<%BANNER%>

The Role of Frugivores in Shaping the Evolution of Fruits in Ficus (Moraceae)

Permanent Link: http://ufdc.ufl.edu/UFE0022103/00001

Material Information

Title: The Role of Frugivores in Shaping the Evolution of Fruits in Ficus (Moraceae)
Physical Description: 1 online resource (152 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: analysis, birds, comparative, components, dispersal, evolution, evolutionary, ficus, figs, fruit, guinea, megachiroptera, new, papua, phylogenetic, principal, seed
Zoology -- Dissertations, Academic -- UF
Genre: Zoology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: We attempt to resolve a central controversy of plant-animal interactions: whether seed dispersers are an important factor in shaping the evolution of fruit traits. By focusing on a genus (Ficus) with extremely variable fruit traits, we could study fruit variation using a phylogeny to control for the evolutionary non-independence of fig species. We studied six characters that were hypothesized to most likely influence a frugivore?s ability to find and access fruits: color, odor, size, softness, exposure away from foliage, and location on the tree. This is the first study on the role of seed dispersers in fruit evolution to combine quantitative measures of color and odor in a detailed phylogenetic comparative analysis. We took a closer look at fruit color because it is probably the first trait mediating fruit-frugivore interactions and because it has historically been mismeasured or ignored in studies of fruit trait evolution. We asked: (1) Can frugivores discriminate between different colors? (2) Have fruit colors of plants that share dispersers converged to a common signal? (3) Do fruit characters co-occur in non-random patterns among species of Ficus? (4) If they do, are the patterns most likely the result of phylogeny or selective pressure by different types of frugivores? Results show that color can be used as a signal by frugivores because it can be discriminated by birds and primates, two of the most common frugivores worldwide. Fruit color has converged in distantly related species, which can be predicted under selective pressure by frugivores. However, we cannot discard the role of microbes as a selective force that determines the amount of the pigment anthocyanin, which has antifungal properties. Color, especially brightness and contrast against the background, have evolved in correlation with fruit odor, size, softness, and exposure away from leaves. Results show that the correlation among these traits cannot be explained solely by phylogeny: a large portion of the variation in fruit traits is adaptive. The most likely explanation for the correlated evolution of six of the traits studied here is selection by frugivores. This pattern holds for figs in one geographic region and worldwide, and that it holds both within and between fig species. Taken together, this evidence argues for the prevalence of seed dispersers over phylogeny as the main driver of the evolution of fruit traits in figs.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Levey, Douglas J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022103:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022103/00001

Material Information

Title: The Role of Frugivores in Shaping the Evolution of Fruits in Ficus (Moraceae)
Physical Description: 1 online resource (152 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: analysis, birds, comparative, components, dispersal, evolution, evolutionary, ficus, figs, fruit, guinea, megachiroptera, new, papua, phylogenetic, principal, seed
Zoology -- Dissertations, Academic -- UF
Genre: Zoology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: We attempt to resolve a central controversy of plant-animal interactions: whether seed dispersers are an important factor in shaping the evolution of fruit traits. By focusing on a genus (Ficus) with extremely variable fruit traits, we could study fruit variation using a phylogeny to control for the evolutionary non-independence of fig species. We studied six characters that were hypothesized to most likely influence a frugivore?s ability to find and access fruits: color, odor, size, softness, exposure away from foliage, and location on the tree. This is the first study on the role of seed dispersers in fruit evolution to combine quantitative measures of color and odor in a detailed phylogenetic comparative analysis. We took a closer look at fruit color because it is probably the first trait mediating fruit-frugivore interactions and because it has historically been mismeasured or ignored in studies of fruit trait evolution. We asked: (1) Can frugivores discriminate between different colors? (2) Have fruit colors of plants that share dispersers converged to a common signal? (3) Do fruit characters co-occur in non-random patterns among species of Ficus? (4) If they do, are the patterns most likely the result of phylogeny or selective pressure by different types of frugivores? Results show that color can be used as a signal by frugivores because it can be discriminated by birds and primates, two of the most common frugivores worldwide. Fruit color has converged in distantly related species, which can be predicted under selective pressure by frugivores. However, we cannot discard the role of microbes as a selective force that determines the amount of the pigment anthocyanin, which has antifungal properties. Color, especially brightness and contrast against the background, have evolved in correlation with fruit odor, size, softness, and exposure away from leaves. Results show that the correlation among these traits cannot be explained solely by phylogeny: a large portion of the variation in fruit traits is adaptive. The most likely explanation for the correlated evolution of six of the traits studied here is selection by frugivores. This pattern holds for figs in one geographic region and worldwide, and that it holds both within and between fig species. Taken together, this evidence argues for the prevalence of seed dispersers over phylogeny as the main driver of the evolution of fruit traits in figs.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Levey, Douglas J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022103:00001


This item has the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101109_AAAAAX INGEST_TIME 2010-11-09T14:08:12Z PACKAGE UFE0022103_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 2121 DFID F20101109_AAASOV ORIGIN DEPOSITOR PATH lomascolo_s_Page_075.txt GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
6af82cb61a70b296cdc673148cef0a46
SHA-1
06cefeec1d6d7fd83686333dd39af09bf783fad7
2046 F20101109_AAASOW lomascolo_s_Page_076.txt
146c93712b9c9b29c29c8936acdd4884
b13933c99e0360b8b70ebb23863826da4eedf7c0
2077 F20101109_AAASPK lomascolo_s_Page_090.txt
af4e427c61a79d6fc6660c0a7f93672f
fbdf627fa5d7ea5aede00ab07dbd7b6c1082f0dd
2129 F20101109_AAASOX lomascolo_s_Page_077.txt
d5fbe9f808402bac884ba01cb13bbb37
6d69d33684a7c994bb7b710accb7f6acd03bb5fe
2204 F20101109_AAASQA lomascolo_s_Page_107.txt
405eb375d804aefee31ed5452c1537be
dcca8fdbb66de4e88706ddc1234460c25e10197f
2267 F20101109_AAASPL lomascolo_s_Page_091.txt
405fc50eaba0cc3a4dd1ef93470d03fe
1069d4c5dd4becc32339852332017d8e4a283208
2277 F20101109_AAASOY lomascolo_s_Page_078.txt
4ec5a8b04cc1387621c50933acb7600f
8937c9617721ad386629bb42d4a18f612ae9acf8
2183 F20101109_AAASQB lomascolo_s_Page_108.txt
dfcd7c8c842ca5678c2f02736dfaefc6
c6fd78851323e9363b670ba05ce9b0730047b57d
2265 F20101109_AAASPM lomascolo_s_Page_092.txt
c00d52a73f89d7efee4e6e873b34d456
24a3d3e02289cef0c6fe3b6b8d3b4ed265054cd2
2132 F20101109_AAASOZ lomascolo_s_Page_079.txt
e42323bb5ab6b3c4f765f18b35e22a4c
835ec82846c26e340c3c4bb5c6802b003b637490
2220 F20101109_AAASQC lomascolo_s_Page_109.txt
83e904fab580f991e2133f2326876196
63d0246965ae63a3dcc72c3f5cdd788aeb907e9b
2162 F20101109_AAASPN lomascolo_s_Page_093.txt
cda148382291a211418ed8988fa10a90
ecaae8050cabef8f89652c8b5b2e20204ebc6450
2118 F20101109_AAASQD lomascolo_s_Page_110.txt
4baf734e631baf47854aa6910030ebf7
0826c0d1325427ff73d1eda5cb69b89fc706cccd
1084 F20101109_AAASPO lomascolo_s_Page_094.txt
1935651e9295d16814c7def75f7ac336
7ea208a933058acbc53929938292dc697a8975e6
1932 F20101109_AAASQE lomascolo_s_Page_111.txt
bd475c793304a4611d0b4f97b5c4fa99
c7e29fb5699cedda56f065ca425c371c232e28f5
959 F20101109_AAASPP lomascolo_s_Page_096.txt
6f6fb9b216341f7748dd5967a8acd16a
1c56439bec5f90b268def92350d211aed440a882
2252 F20101109_AAASQF lomascolo_s_Page_112.txt
093856e758c9edcb2170df82ac4040b1
f65d2d5b683e14ee37450607bb6451b02e86e26e
295 F20101109_AAASPQ lomascolo_s_Page_097.txt
f0f4a23a9ce19d5bd6377326d6fb27fc
4ff56e3d4b876b08c55888a849442d59ff51d8b7
2067 F20101109_AAASQG lomascolo_s_Page_113.txt
f993f2f1eabc0c453174da5d96dcc2c0
2aa3f73bf95a2a8fcf34e636ebc8ad62f8116082
796 F20101109_AAASPR lomascolo_s_Page_098.txt
66a956e30aae306c05a16e136bd86ff1
1032186431b243bbb8303360320be017588b8fe5
2025 F20101109_AAASQH lomascolo_s_Page_114.txt
37268956b8c7e5a4dca1a70d144837f9
8783cc14570efcb65f1be0a3263da88bce3e29b6
236 F20101109_AAASPS lomascolo_s_Page_099.txt
117798257ef97e5b337d260eb1b5bc44
5f613d19fe3f327ad6874b989dce64c8bdf4776a
2259 F20101109_AAASQI lomascolo_s_Page_115.txt
82e12424442e74dc48f1c967a9661e17
e48752e0e366321ef587bca0e3ca43bd8be0eea9
1644 F20101109_AAASPT lomascolo_s_Page_100.txt
9cfc1dad23bdd3f0a1bca199dc9a8a2f
4c2f5a98d345c0751df364a7a371f65b17970b70
2111 F20101109_AAASQJ lomascolo_s_Page_116.txt
8ca6f8576648b5b2659689cc2f07b202
da5e44e355a724fac23c10c4fd201618304cf689
1120 F20101109_AAASPU lomascolo_s_Page_101.txt
c498416191e21e54cc1a316e58521cee
29b02ba73e84effbaf1d92e0efa7ef6d32b5669a
2172 F20101109_AAASQK lomascolo_s_Page_117.txt
ffa249781de141c21f619796a1da7410
946dc82103b484759f6b3376baf4262fa43c317c
1533 F20101109_AAASPV lomascolo_s_Page_102.txt
e126260c0122d7b5c34a6724f7574860
de116d0ed5c0ef2f5363123ff71379a90120fcd3
789 F20101109_AAASPW lomascolo_s_Page_103.txt
9f89d9a802d0c97bb426cf38c70091f1
b795218eb79f0abfdf436ce6850e75f6785b47ac
2238 F20101109_AAASRA lomascolo_s_Page_134.txt
b3e079b3116d21a47275a35e0aaef358
00c25f1907a586cc0d02db6c5235748afe39cbfb
2187 F20101109_AAASQL lomascolo_s_Page_118.txt
1cd2612400bab487218509435dc04946
f6ec32564ad7329b658738d6d1bea280449855d3
1562 F20101109_AAASPX lomascolo_s_Page_104.txt
b83d7825fff8610e8f0eedda0909ba37
c1e1f21a4527236cd23f1370e763e879eea4b3af
1576 F20101109_AAASRB lomascolo_s_Page_135.txt
1b5f31a7f665ccfc12432ec7fd5ac2d4
e41174a7fe5f2cd786f54fe4202dc3b3408ad6e0
2309 F20101109_AAASQM lomascolo_s_Page_119.txt
14a658aac83c883d9584c82cfe22fdbb
a54cbe48fb2a02a72e9a9604a26c8e3a32d4c5ce
2149 F20101109_AAASPY lomascolo_s_Page_105.txt
adc1c8b1215544e1603f4d6308b5a95f
36ff19ed92d19b477270b25375f96b28a16d99e8
2175 F20101109_AAASRC lomascolo_s_Page_136.txt
2c5df3ea82b267886a8488637438f691
6f910aa2a8056210d8b20abc60890bb528a57559
2255 F20101109_AAASQN lomascolo_s_Page_121.txt
c45d59e09410a6e5e42d45dde84e558d
b9b46932a45a60290d60e6d24e2391c91ec3a85c
1782 F20101109_AAASPZ lomascolo_s_Page_106.txt
17226c9a43ce5652f8f060782f622ed5
22ca4fb980bf15a4b9a478286e6304332ba4cae6
2328 F20101109_AAASRD lomascolo_s_Page_137.txt
d078a608c8cf176080bb56a297d37bf3
54f853974190666fbfd59f70328068f7e3e38fce
2070 F20101109_AAASQO lomascolo_s_Page_122.txt
0a6b1059a8f4ce4e980015b975f8be53
9e663f5ee6bbbeae9c14df0bb4f95c83ecfe6f34
2282 F20101109_AAASRE lomascolo_s_Page_138.txt
6a9f258ce71661e5b67753b0ab55fece
ed1537e06ea13b35d7a74ecd35118ffb59c0699a
953 F20101109_AAASQP lomascolo_s_Page_123.txt
643118eeff45b80ceb8aa8ddbebd88ad
aa042f5beaf8d2bae66a49ff51770e416dade582
319 F20101109_AAASQQ lomascolo_s_Page_124.txt
888d47fb45cd1d09c04cea7270c6f499
7b90e4c084fe92e093575806d01366fe01b729a6
2461 F20101109_AAASRF lomascolo_s_Page_139.txt
b0a7c0516a0c5eb4480531fd56e9a0a8
1d9050c3839851f79c27c0a7b53cb1f05ac555b7
650 F20101109_AAASQR lomascolo_s_Page_125.txt
dd5821f0bd14793858eea16ee36a850d
719975f113a976d6fd61d40c33aca290e815bb87
2270 F20101109_AAASRG lomascolo_s_Page_140.txt
b11c8b8c0860b8683761d62364d9d917
a54adf84c94685d84722f9eb2115935d16eaf80b
418 F20101109_AAASQS lomascolo_s_Page_126.txt
3e734d57ac92467d01e8cfef52f65d5b
544a7ec009f7b3a7d063c6c1f91404f9f3d58205
2308 F20101109_AAASRH lomascolo_s_Page_141.txt
e822f92825801a23e670f26543b0a324
1d073f16c3c1b43d2ac1bc2aa8dee5bf4763618a
806 F20101109_AAASQT lomascolo_s_Page_127.txt
f048b5c43093e3e99cf61c5d5aa6432e
c58a6a8e1eb4cfe520e27f482cc8023040f00b30
2257 F20101109_AAASRI lomascolo_s_Page_142.txt
ad13b8bd749da190b50f9093d319a79e
4ab95a4ddac81d333f399751071e672dc2e36b98
1523 F20101109_AAASQU lomascolo_s_Page_128.txt
b7519645e67ad63b094db44426c14681
bb4050eeea5cf88aa7eaf1a143d55828094f7824
32164 F20101109_AAAROH lomascolo_s_Page_100.jpg
09331f0ca64ff5d431228e4642bf944f
36b46fa3b870ba982fe46499bc9a0c44f303c212
2272 F20101109_AAASRJ lomascolo_s_Page_143.txt
56e579509f548e779a6730d7404801f2
1c58bd06c48f4ab46a6f0f2191e0a65a899e0058
1919 F20101109_AAASQV lomascolo_s_Page_129.txt
fa712dcd8b2c82e969cb7cf37b82588d
ec13ab74d693079e2247c98dce6f9dc3890815c8
25271604 F20101109_AAAROI lomascolo_s_Page_074.tif
920cde0bdb2e09c4dc5d96160b215693
bc056113825b63f1c7ead984d119b2c94fb2fc34
F20101109_AAASRK lomascolo_s_Page_144.txt
f06eed796a6d625ace74676397f9965b
3b23bbcdc223cad18eb9d285849a9865c4715b15
1911 F20101109_AAASQW lomascolo_s_Page_130.txt
485bbd9a98225f70f8ed70d89791ba29
085b34a6d08020bfe1147f79ab76df57028cd410
7331 F20101109_AAAROJ lomascolo_s_Page_075thm.jpg
c3df0e0977b953db09e6ca434b7bc498
ecf1ea2557208c66de60fe47ba212b9dd2b51925
2526 F20101109_AAASRL lomascolo_s_Page_145.txt
909976d5ae2fbd653c96ca0f49440e7b
490417f829e57bb3ea942d48ec21d0405c5d3e13
2182 F20101109_AAASQX lomascolo_s_Page_131.txt
745de170c2d8d05454f9507ff2302a49
b7c7b42c672b8cff118ff3de6680cf7c2c91aa62
27463 F20101109_AAASSA lomascolo_s_Page_004.QC.jpg
75a30af157c78d403e2a25232024f46e
743932eb16231cff2ae6b954fbadabae12e734a3
2231 F20101109_AAASQY lomascolo_s_Page_132.txt
e8ba394af0b65ef1e48bba6546b99598
ef012dc03aeca511f2a9c1c7425bea49f8c11d03
7388 F20101109_AAASSB lomascolo_s_Page_004thm.jpg
0e9aff7c3fb96dca84a7c91b17317cb6
654d6214b00edf13d6002e3c5f20d2b5fc79fa76
26206 F20101109_AAAROK lomascolo_s_Page_090.QC.jpg
a7c0a5a74997e88a397ef502dc321e9d
e20a36ce041b10d9007cb486bf54af10af003a28
2192 F20101109_AAASRM lomascolo_s_Page_146.txt
b32e502555a1847e9291c8a483fcf3b6
b41b8354d5106a2b11cf0bda1300d7cb6fb43103
2246 F20101109_AAASQZ lomascolo_s_Page_133.txt
c379fe90350b3b8b1263ca808d1dc709
a7482906734d9d67571d20efea867d4cb519ddb8
24513 F20101109_AAARPA lomascolo_s_Page_001.jpg
f52563cbca9157ddc9a17a7a7de54693
577b04f6fa8f4c3841fe458fb9bb5329b099c381
26957 F20101109_AAASSC lomascolo_s_Page_005.QC.jpg
48ea6eddc9a5b63df7be7c0994402a0e
1c04487a4cee780040a043a0412117fdca9a2ca8
2119 F20101109_AAAROL lomascolo_s_Page_120.txt
e8b6d5c7191f3c12b22e3a60552563e9
402e1cd0f513cbe962c1276b55d9007855cc34bb
2410 F20101109_AAASRN lomascolo_s_Page_147.txt
b971f2684750a00c8d356be5d1990a36
8d8d38799637026ddef8c2ad281fda01bdb0e628
15593 F20101109_AAARPB lomascolo_s_Page_003.jpg
45bad90fced68196fa419566084d058d
928b50b22c3dd24eeaa2d5da5c6d93f02a3f1d98
7069 F20101109_AAASSD lomascolo_s_Page_005thm.jpg
f886303b8ec6cd632f715ca57504874a
a184c838cb4321efcccc1c3f9348b48393780ba2
1051956 F20101109_AAAROM lomascolo_s_Page_119.jp2
54dc3175dbc686c5cb1d6e56c5df88a0
0b809c418b96e026216f1903a7ecf25efedff951
2494 F20101109_AAASRO lomascolo_s_Page_148.txt
949770505e2ba0ae1e70c340156d96d6
72bece8aafa63a3b19c42c9cf12f2161f571409c
85766 F20101109_AAARPC lomascolo_s_Page_004.jpg
7bb27b1defb15b501101ca50aefe4f07
8b17351f986929a18d429409dae32a4ed7d2c76e
26641 F20101109_AAASSE lomascolo_s_Page_006.QC.jpg
cde52a186b26cdb42b552dc94172b6bb
a5cc864626a1dd7e81750f209b461478b8178018
903423 F20101109_AAARON lomascolo_s_Page_010.jp2
19721316011d0868c81701e87061b2f3
fdd1e682c35dc15dfd0e566c7b21474f48926bf4
2567 F20101109_AAASRP lomascolo_s_Page_149.txt
4f98981044652f2db0d09a3a49be431b
9d0fe4aeb803d9abb8c168858f3ea1b33ab7d38f
84756 F20101109_AAARPD lomascolo_s_Page_005.jpg
9e3f5b031f17a80e8d4285c26b5d2ddf
60f67a389f4cb976dceb644fc5a47fb3bcd7d174
7164 F20101109_AAASSF lomascolo_s_Page_006thm.jpg
f997d7185e15bcc0fb13f47ba50ca9ea
26e03e15726d76fb230d8a9c139a12464aabaf0e
F20101109_AAAROO lomascolo_s_Page_034.tif
1f3ee3671c01e279a999eaefadde1b89
be63156bad0be6da3c5540c879a8cc743e2f6b6f
2217 F20101109_AAASRQ lomascolo_s_Page_150.txt
a1fa36bf373ba43b206ab2dbb4ee50c7
d68f6eb2050ae060c3689a5283ef77e41f8e4e26
83018 F20101109_AAARPE lomascolo_s_Page_006.jpg
cebd889953dd39261d82f049e49cc140
f2a01cc55e12cc0d83b4e4d6d1c8495b0e8c0c7f
20157 F20101109_AAASSG lomascolo_s_Page_007.QC.jpg
237012fe834973bbe31db0f52809d0b7
0c80fc1961f7d58bb775ba695f8f99eb180f1c2f
1519 F20101109_AAAROP lomascolo_s_Page_007.txt
e0f7c3583a29847e4389bc7ebea48ffd
958b1bdb790370e12afcf468258e2e6d40907136
466 F20101109_AAASRR lomascolo_s_Page_151.txt
6d589b3a89724422467b96ed588d31dd
57cbd1d4f39e7cebb51a85c8485dc3ab4eae43ef
63149 F20101109_AAARPF lomascolo_s_Page_007.jpg
c46dfab2ef70f167723a4dce1341b164
d37d263b1ede8c1d0f2bcfa989b34dfc29177573
5833 F20101109_AAASSH lomascolo_s_Page_007thm.jpg
760d406e8800c299ee5d903a8f08ad0b
3ab46ca1c32bb8704969b19d79b6beeec8871b0d
F20101109_AAAROQ lomascolo_s_Page_095.txt
04d1aa2ba7b969d0149aaf7eaac937c4
7559994663fc1c91e232769ceeebb866585fe842
1978 F20101109_AAASRS lomascolo_s_Page_152.txt
bcddc24cc081c58123de391aff3449db
cee9cc207e8102b49ddbadf837de52b9430fab8f
81503 F20101109_AAARPG lomascolo_s_Page_008.jpg
03d5d77c0da038984998b879ed8a1606
7e6cd488a67a1eac578605e3151513027b347772
21245 F20101109_AAASSI lomascolo_s_Page_008.QC.jpg
e2b991115996460e8413cdb3f2c4264c
9116ba931ffbfcafef479dd2df5daa6fc85194c5
72914 F20101109_AAAROR lomascolo_s_Page_019.jpg
99a13df5c3874da477656b6e3ef8e239
bd4a8d6cd10f75d81b8a45bbe5d039bce8258b64
2446 F20101109_AAASRT lomascolo_s_Page_001thm.jpg
eb03b272142608725f1cf1efb3abf153
edbbfcfd97634da8543eea6b99b442b4d385a444
103461 F20101109_AAARPH lomascolo_s_Page_009.jpg
44eedd01f8042bf538f8e6665fd4ee94
c9f51534bbbdb4c49fa655474921c26ac32fcddb
5429 F20101109_AAASSJ lomascolo_s_Page_008thm.jpg
d577bf14461eeb3c18dcd93625c4f37c
71bba262388f5a48ee80c805460b74a1bed0610f
10307 F20101109_AAAROS lomascolo_s_Page_002.jpg
27f9671f0790d63d372e5ebf069ad0ae
407869c1c226bca1900bf16b158c5f44b3da5f4c
1147804 F20101109_AAASRU lomascolo_s.pdf
243e4aa4ebce140670ab65cf4b2fa773
4fc54973ad51c02da303bf19b9d62bbce32ee991
38403 F20101109_AAARPI lomascolo_s_Page_010.jpg
653585c8dcf0e21eee51c209e571e816
02c5f7c146e8fc4030ff3492f0dd4cd1a5ccf889
25970 F20101109_AAASSK lomascolo_s_Page_009.QC.jpg
af8de6036f35b328e7c76c08b5cb1cb0
dccf31a521fd524e2a91288b7537d6592daee142
2164 F20101109_AAAROT lomascolo_s_Page_021.txt
ee3d4683dcdb41b31cd8ed2900275132
608ac2b50e3b1e002f91f1ddf73755b0cfad1d26
7677 F20101109_AAASRV lomascolo_s_Page_001.QC.jpg
1d567d8ef358a9734ea5ee6e3eb596fb
4c043c99bbbecd86bf5a752164d0412cfc3da95d
49104 F20101109_AAARPJ lomascolo_s_Page_011.jpg
b6f26b97d61808ea3868aede72e15c98
fd2c7704c9ac4313f1423689daa2fb50a738d9c8
6156 F20101109_AAASSL lomascolo_s_Page_009thm.jpg
3c0ec18d09d88842cc3bf8b1ab2defe9
729704d256fccb0d23a6a95fc9b6ec0bfbf219ba
F20101109_AAAROU lomascolo_s_Page_027.txt
5ce89ebdcfde7ebaa41d13d8d353ed9a
3073e3c2b739f8c196d062fe4e10d61ab2166725
3181 F20101109_AAASRW lomascolo_s_Page_002.QC.jpg
0029ed3f8fd052ba15a5c736ad699b52
b02d7e1c7c1e5aac6fd087941f54b46b57dc4f45
27818 F20101109_AAASTA lomascolo_s_Page_017.QC.jpg
c243c762edac6bfb0c1bc0b1e26319c9
298a95eaa234c47ffc6a6ca40202e3194939585c
91129 F20101109_AAARPK lomascolo_s_Page_012.jpg
73bd9a63e7a56f1ba3de1ba0bf9e7f47
daf25844480903424635033b8753e6ba86365d70
10259 F20101109_AAASSM lomascolo_s_Page_010.QC.jpg
4c80856eb9ce47d4c32c0320a2fe0271
3ae749b5c7814c19b3acea4a7081d0442c055e72
F20101109_AAAROV lomascolo_s_Page_057.tif
eb758d78f79ba096cd6f5f87145f4b8d
c0c320b01d3f62b324e91e45add83c69f284d957
1450 F20101109_AAASRX lomascolo_s_Page_002thm.jpg
06ed66aad39016e0cac5136d85228f32
c8679e0a0575f0680d9fccf51ea757eb328ab1d2
7334 F20101109_AAASTB lomascolo_s_Page_017thm.jpg
389777c38377cda1f25bd53eda592c81
15ea46ffef0b076b6273128dba67f5d45335d0e7
7624 F20101109_AAAROW lomascolo_s_Page_139thm.jpg
edc4fa4a04cc476b5ec7176981919f63
33208f3c45a2c7e210629b60214d523b136e4f8d
4374 F20101109_AAASRY lomascolo_s_Page_003.QC.jpg
8331a9300be150f2b6fde306c60f260d
88bc3851e82863ca7a7f919f5e8506f3a85baa75
28186 F20101109_AAASTC lomascolo_s_Page_018.QC.jpg
7432713f1eb5b2ff9d857f09effc1f59
1c51ee63f785e6c14a7f3179de0c7109e20533a1
2936 F20101109_AAASSN lomascolo_s_Page_010thm.jpg
60a3e17e66a6544a8d1b84fac2446ae3
e72bd260711d8814a52036b390400776881ec3fa
89667 F20101109_AAARQA lomascolo_s_Page_029.jpg
867c14db801427510bdc7a9264040bc7
42e8af33dc403e2f72bebdae82ba2b0cfd15be3c
12245 F20101109_AAARPL lomascolo_s_Page_013.jpg
9fb46f751e7f1e6c5afc270c9d54dc4d
1846738bae5907abfb047ecd81124a14d90fd31f
230535 F20101109_AAAROX UFE0022103_00001.xml FULL
174a75d3f221d3b79486ef10749d6a49
a3d2c024d61eafbcb683cd8ab7133e473a6e68c0
1806 F20101109_AAASRZ lomascolo_s_Page_003thm.jpg
89bd57e558b027c9a4c5e18f9430254a
f5c81234a3c0b53eca52e6d3dd6207ba5e3fde8a
7647 F20101109_AAASTD lomascolo_s_Page_018thm.jpg
647c4106d703ede18177761c8d962e96
d3b337ca21f0e45b183ed364e0f11e92c8386d36
14168 F20101109_AAASSO lomascolo_s_Page_011.QC.jpg
ad9f42610ddce683781b0533304f5bbf
3850c566944999c9d51b676efab1c867bbd89f2e
83826 F20101109_AAARQB lomascolo_s_Page_030.jpg
9799333694f2d822113eb0a8c931cbc4
933b370f6298b20aac3e252d673163b1e18a4e41
77824 F20101109_AAARPM lomascolo_s_Page_014.jpg
20bae51bffa682bee83e373407e03850
1364e50301fd8b60ab36de3f105faff4d6f85a88
23684 F20101109_AAASTE lomascolo_s_Page_019.QC.jpg
79669494cd5f999c980ee9c19db30283
d54ff054e29524d7567624dc54f5eb8c78f48175
4041 F20101109_AAASSP lomascolo_s_Page_011thm.jpg
8688087632b13a981050f0bb5afd7f40
58300826658a5d8b142bf2f2a2452b5078e3944a
74885 F20101109_AAARQC lomascolo_s_Page_031.jpg
3b10ff9be1739c4743270c00100fee12
cb287e9e9c795d53e510109bb7cd4dfbce71690b
44466 F20101109_AAARPN lomascolo_s_Page_015.jpg
939cf61ac47b49e11bef0a129110d27a
6f25a9ed64066d652958cceecb87fbfcb0bf589f
6440 F20101109_AAASTF lomascolo_s_Page_019thm.jpg
c241547d218b6a8bd1046af8e0a77b93
489fac83183c8d81ddfcf497f96c8fc44bc8ccec
25769 F20101109_AAASSQ lomascolo_s_Page_012.QC.jpg
154fc4a1c0cd249708165d3adb1aadd7
60402a6b2492672cd047fe2cf21785b119b2f866
85033 F20101109_AAARQD lomascolo_s_Page_032.jpg
ba36d07aefa0b30ea37dd1a949423bd5
df966221ac89141cccf5679269884e01faa715f2
80505 F20101109_AAARPO lomascolo_s_Page_016.jpg
06ba7f20cb379b76de4892237da35672
02ca0d8bbe62f9781dceb4bc9787bd0b2ece3c51
28000 F20101109_AAASTG lomascolo_s_Page_020.QC.jpg
850dc2c90793b6fb50045dcaf916d8d9
bfe41c571d22c4345a07eb91fd68bb01d4c8a15b
6712 F20101109_AAASSR lomascolo_s_Page_012thm.jpg
da0d46a1361b582fc586231fddb5ab3f
90d82aa8aa394d576ad89842d2356f0f1bd7b203
82426 F20101109_AAARQE lomascolo_s_Page_033.jpg
ec9d98cc529f1de08ebc00e5c9d5e0ca
d7147ebae920dd3761c5e9896869a0a819e009bc
88561 F20101109_AAARPP lomascolo_s_Page_017.jpg
766c5071a4664bde09b54809cf7b6110
545862448cf6402186f772ce149e4f494c07defc
7434 F20101109_AAASTH lomascolo_s_Page_020thm.jpg
566a97f5c5398e09480767c26f90c2d2
c2dd0a01ade5518f627389091e253c991409fa0f
3907 F20101109_AAASSS lomascolo_s_Page_013.QC.jpg
1599a34233732802f1a3fb47aee8d517
2e2f37cdb6c1bcb3c51f5a19c10056ff39858edc
87504 F20101109_AAARQF lomascolo_s_Page_034.jpg
1e3a745465afa133ee954193c984be83
1c4ecf66d3bbcbfff3df291c0d9fb3d35a6ac107
88338 F20101109_AAARPQ lomascolo_s_Page_018.jpg
85cc33f3bf6a1be8306a2a65fb27a59d
92a8a96c4a0ad8b82aafd66055ac9794d0503173
27656 F20101109_AAASTI lomascolo_s_Page_021.QC.jpg
b48b2c77295e3a5786cfb366893d8815
3eb6e331cd4ade93e07049ae2b9f0345e9d61ff5
1553 F20101109_AAASST lomascolo_s_Page_013thm.jpg
d35deb302c10f2531daa6dd33505ce02
e109db67ab57266fd61cb6accbffd3847f6c4a83
85835 F20101109_AAARQG lomascolo_s_Page_035.jpg
910f50d97e5168f24ad7b6f0caf31c89
ca7f86a281cd115c04b733a5ac6d36b09289c624
87345 F20101109_AAARPR lomascolo_s_Page_020.jpg
b53a45f2d5d947d66f92deb5ca36da85
d65a915034950471fd55deeaa0d9c80d8e609e99
7430 F20101109_AAASTJ lomascolo_s_Page_021thm.jpg
e1eaf0837d74ab837b805bd13b078ebe
12c4786f93b7f8eadb7d87b9d677f2f665bef714
23490 F20101109_AAASSU lomascolo_s_Page_014.QC.jpg
ea21af137442450dc3392e5f17f209f0
5c7d494c5f4cf0b9577ce467cae5bee6128fcfbc
88360 F20101109_AAARQH lomascolo_s_Page_036.jpg
5bd488a46f21b53fb90fd0f0bbbd5105
a40cc552af1448382c3e81617a7aeb16cc2ec8e5
87122 F20101109_AAARPS lomascolo_s_Page_021.jpg
8255a5a4a6ab9a1e2868ba6c241197f8
9c8a1e6de40cf877c3c2570cf17ada40c2c27bf0
28124 F20101109_AAASTK lomascolo_s_Page_022.QC.jpg
439481d78aa7ae73a09da449114442ac
a9b5a5c861db8633898196c1963025e3e1aeba7e
6610 F20101109_AAASSV lomascolo_s_Page_014thm.jpg
b279ad36684520a876a5bcb65d7b0df1
302ea2a5dce7ab98f2dc8b26ee3cc05c425518a2
88230 F20101109_AAARQI lomascolo_s_Page_037.jpg
3e086886d815bfc8ce0504a9cdb43c57
3fa3c0af55c4f73e7bf1f0ea10aacbb1700d38b1
88894 F20101109_AAARPT lomascolo_s_Page_022.jpg
c18bb869f0a862a4c7db10a2a66770b6
73603db7a6c227d2d2152e276718af9f736dadcb
7252 F20101109_AAASTL lomascolo_s_Page_022thm.jpg
bfb25aac84caa7db75e91284a7cae400
e1f9edf2d4a7c60eb3ab872d07c514505f164e15
14025 F20101109_AAASSW lomascolo_s_Page_015.QC.jpg
cceecc98fa2029b893dbf69aed824e6d
54faa831f0757bb200030238dd51c5ab40c71733
29979 F20101109_AAARQJ lomascolo_s_Page_038.jpg
b018779bfa19497a16d0e5fc3fdefdaa
c73ac80e24ad22c10000329fe1b06d59461958e4
73735 F20101109_AAARPU lomascolo_s_Page_023.jpg
3140c8ac40ed33282dd34717f8c0bd5e
4f75279b6dd59580949f0cad8bebadd0d59fb78a
26222 F20101109_AAASUA lomascolo_s_Page_030.QC.jpg
d40ac3383d55ec1c73f2bf443261d78e
8ea6f7e0d426fb36f9ab218437ab99462b3e419f
23641 F20101109_AAASTM lomascolo_s_Page_023.QC.jpg
2f866b732cb72d40079cdba19567b51c
dbfb2782d5bd2c6a6ef6f270806288f9d524f7c0
4152 F20101109_AAASSX lomascolo_s_Page_015thm.jpg
bc9275c4981243fdfeebec372818b86e
3107284ba808914d6f79594f642209396c65f4cb
63196 F20101109_AAARQK lomascolo_s_Page_039.jpg
b78316b67ca87ecc7dd2e22f3c416f41
590e32d595494edd3516e76e931bef943b5621da
85599 F20101109_AAARPV lomascolo_s_Page_024.jpg
689fd5b1c818a6c03cfda59851735826
359a66c90deaf29dc6c86b69a3f9b1d3f6363548
7296 F20101109_AAASUB lomascolo_s_Page_030thm.jpg
a01441e5bff6eccb6451088c77dad790
e5df7916c908d9388f6d7b32ec7b8e7862423ca6
6317 F20101109_AAASTN lomascolo_s_Page_023thm.jpg
0bb7a89568190ab6d4fedd0e706cfbec
8e9cbf0bd9d20e346499a659ad25d40a13fdbe6c
25681 F20101109_AAASSY lomascolo_s_Page_016.QC.jpg
998e19b1640ad1619baf2939a2079ec8
1f7c90f8775c81380123427f74d924ff92c54765
33438 F20101109_AAARQL lomascolo_s_Page_040.jpg
259a0bb8ee3e7a61f8e5b239ca072296
a489ee193fc6dd6e81ed04d617ae3b1ec250e280
83618 F20101109_AAARPW lomascolo_s_Page_025.jpg
e9803e669c0470b1d21f479184d1f62a
f6339450607a1f213bd128346560b27e0dfacf15
24038 F20101109_AAASUC lomascolo_s_Page_031.QC.jpg
a22771d1e3b5d070f307778614bce020
b753eebf92488e53a61e5954ed94c24744adedd1
6878 F20101109_AAASSZ lomascolo_s_Page_016thm.jpg
d16d53204fddc00e08992b225ca47064
e8c168f7371d91645ceec6b3de26e5530e7fd991
81192 F20101109_AAARPX lomascolo_s_Page_026.jpg
f4ccd1a617dfdb8069435996ee45a7e0
34b7aa4c8757f3eab92cb6f01bff90bcec233206
86208 F20101109_AAARRA lomascolo_s_Page_055.jpg
2a0b7aedfc6b5e2e871d66e8da525c39
013668cec019b2fc6fb209c96cf8759a82fa2b34
6815 F20101109_AAASUD lomascolo_s_Page_031thm.jpg
bfcd54b39c745e07c8963892339bd4a1
f045d00dcd0d1f2ffbdf863b382d4c6890018092
27514 F20101109_AAASTO lomascolo_s_Page_024.QC.jpg
a24d07942abdd858915a86fe6381f4c4
127baa9fbd457c3d3face39276ec0a0a5d378759
57431 F20101109_AAARQM lomascolo_s_Page_041.jpg
68c4a6b1bb7b4e1c64e9d959c0966e79
bde75f94b6a39f70daf3e784bd0e4858ba013e0a
87440 F20101109_AAARPY lomascolo_s_Page_027.jpg
b578503561cf5ade8465f035ea2fecf7
6a37ffb1e451b411f9afc2497423697f199d6c3c
87267 F20101109_AAARRB lomascolo_s_Page_056.jpg
6e4ccbcaa818dc3a7e7cf8a23cfce968
84d6e675e232cdaf0814db5281819671b794d9f7
26959 F20101109_AAASUE lomascolo_s_Page_032.QC.jpg
54e9f80c3b5d6988fd0d78d73aa8db0b
e8f33e00a8ab3aad896e46d4429c1d176471a251
7407 F20101109_AAASTP lomascolo_s_Page_024thm.jpg
ec75d936f4c76b6a5adf0abfa8b98e4d
a1f970de9b8607464a9cc8e274012e8510702298
53131 F20101109_AAARQN lomascolo_s_Page_042.jpg
9ba84b3b53015339c1ca1b7910a1a7f6
d81695766a7682df5079bd5de546379c21726326
83404 F20101109_AAARPZ lomascolo_s_Page_028.jpg
9f2fa0d7329a4fd4067079f241119548
fe4336ef5d0fdc45e95d8eea08516f7c63f0ba5a
87026 F20101109_AAARRC lomascolo_s_Page_057.jpg
45102143646915013356178710551007
6e6ab803cb87c638b4a6eb4376cceef9491e9dc9
7226 F20101109_AAASUF lomascolo_s_Page_032thm.jpg
6ce904a85bb26bb93401e5a0456fc4c3
e74a6b1d5dfd2b4efc89e381af84bfdb1356b6dc
26303 F20101109_AAASTQ lomascolo_s_Page_025.QC.jpg
8ef9986a57b29516c9c41a5a6ad065cf
69b4cc7cafd489e1a4a1632cf0082f3f92140f9a
49162 F20101109_AAARQO lomascolo_s_Page_043.jpg
aa319486c3418688c2fecfdc5d8d1b94
e253da2159146f9a54497dd5f37afa0967286d64
80105 F20101109_AAARRD lomascolo_s_Page_058.jpg
d9bf73e9f360c2e022ed864a915c5809
23b03532b98825cc51ac3fed2a2e5b8688fb6fe6
28447 F20101109_AAATAA lomascolo_s_Page_109.QC.jpg
821dbb7c2c29932ba1b4b549a7cb6dfb
4d442ea925b4e626e638317722e55f91366203a6
25890 F20101109_AAASUG lomascolo_s_Page_033.QC.jpg
63aeb29094a1453df9d8541d5d4934a6
9dde95daff76f8256f021e7c32884e515dde7560
F20101109_AAASTR lomascolo_s_Page_025thm.jpg
7a63de61e982ea4d27f00a4e2fb4904b
f5b0ea2a45f12761067fa2446f0eb891d6fc68de
42137 F20101109_AAARQP lomascolo_s_Page_044.jpg
1208990baf69d9608df6ec3d49d331a2
56fe08aef8c989f5e6f47a3565071cfff74b6a2e
80159 F20101109_AAARRE lomascolo_s_Page_059.jpg
a369c0befc49494b823e8b697bdd6c38
ee76a09b85c69022982932195021e3c7dfc8840e
7706 F20101109_AAATAB lomascolo_s_Page_109thm.jpg
e470e35eb6202ab841c2bf3ef91f82b0
51777460bed53b09d96e2b096a3cb198e819fa55
7300 F20101109_AAASUH lomascolo_s_Page_033thm.jpg
db12b1fac870d56fa0312974c7d37bbd
e96552d1d77659919ccdeeee642b4e8111fd622e
26291 F20101109_AAASTS lomascolo_s_Page_026.QC.jpg
8f2013da891055761b6cae5b7dc28938
397504a98b56aa877b0d46bd332dfd85c80a1b63
33672 F20101109_AAARQQ lomascolo_s_Page_045.jpg
a4bbbb37faaa1b1ee3f52de6b2235baa
57470801c7b76ce328ebc34ca09dee29d33b0903
84377 F20101109_AAARRF lomascolo_s_Page_060.jpg
6726876160c2587b67af4ac27d95b09c
ee7d4b60c95ec57f9e58ee1ff925a59e1208a837
27664 F20101109_AAATAC lomascolo_s_Page_110.QC.jpg
48ad156266e88bd24e894939247527ff
88a649f80b24539d1e4dd80c89a7c97c5751e459
27858 F20101109_AAASUI lomascolo_s_Page_034.QC.jpg
5dfe87085c1cee3e64b310d64242213b
f074c3e69d827806ce1d822e1a684a13397fa99f
7215 F20101109_AAASTT lomascolo_s_Page_026thm.jpg
8cfb6acdf57ee2507d19c286ab2ed3ed
3c3ff6004b24fad5582fdea8c6bbee4c178e0ac3
30995 F20101109_AAARQR lomascolo_s_Page_046.jpg
286be6ba8d1cb5c89b7b933a9b74a0e4
b168d712a5cd7f4bd14c7f583c636b0fc17eaba2
87704 F20101109_AAARRG lomascolo_s_Page_061.jpg
51311d2aa4df0a07e4bdf0c1fc96d306
87a9d2f5509e468914983cff7d5f52ace66979ab
7206 F20101109_AAATAD lomascolo_s_Page_110thm.jpg
64fa5d12dc4cff3c9bd2e15e70b62d93
09d0e0d8c3cb2ac95c401c6f1f15831aec258a96
7462 F20101109_AAASUJ lomascolo_s_Page_034thm.jpg
d5f19972eca8b293430d0789c90c092d
f1acfa4761397f8e765279a8495cc4b054f3578e
28099 F20101109_AAASTU lomascolo_s_Page_027.QC.jpg
ece43efb1eadb5b7feb5869fbb9eac36
5db423d915db641cd3ac66ed3e930f4c62d9eb8d
88961 F20101109_AAARRH lomascolo_s_Page_062.jpg
9d9fcc81356d20e8e9e53b248d0f0d96
0412e8276064789f1b2bdd221a1f417d5ab43c81
24162 F20101109_AAARQS lomascolo_s_Page_047.jpg
cb01bb77cedc383dbb14e47a991cafd2
5312c4016d05871f809052dca3294c4990f13fed
25033 F20101109_AAATAE lomascolo_s_Page_111.QC.jpg
275ab040015a851b933cbbf63aa9b65d
cbff83793fec51b9ef63086ba07ffec1a1fa7871
27238 F20101109_AAASUK lomascolo_s_Page_035.QC.jpg
63062cef5f61437b871c35b757f9c734
809260ee5a0d86b22d705dc1ea9c0436704af3c5
7507 F20101109_AAASTV lomascolo_s_Page_027thm.jpg
2ab85dd70a7eca94f049b12c10a1f0a2
d5b4e3e4c468d7dfab4a80fe7e593e2bb3803a89
84835 F20101109_AAARRI lomascolo_s_Page_063.jpg
716658932d63025306c65b2d31f85e3b
b250df97151e1445880a65c855aaa736b9a2bcca
27356 F20101109_AAARQT lomascolo_s_Page_048.jpg
6fab0cd8ba353b051d7c325455004d04
b96796deb8a73a2c093e8e9ee62e7c35c473ac23
6827 F20101109_AAATAF lomascolo_s_Page_111thm.jpg
89b0401981867917ab21b4ddcbaf5bfe
5c7b4010c3a877bd8758805fef48e5be02e01800
7238 F20101109_AAASUL lomascolo_s_Page_035thm.jpg
3d724afe7d1b84444e31d6f62042d09d
9af1e9a0e810091990d81d36746b387df10367e0
26238 F20101109_AAASTW lomascolo_s_Page_028.QC.jpg
2b835e3b0dd08ee26ccbb9da4d920d1f
f750dea15bedaf122ee0bb88d4c74c245c1b53dd
88486 F20101109_AAARRJ lomascolo_s_Page_064.jpg
bf825e3e878830d7219e15ece7bee033
200fd7201820909e2e5bc318a6e26dbfe68fa762
29228 F20101109_AAARQU lomascolo_s_Page_049.jpg
171ea1b9fb506f0b76f2833033e1a6ad
b241759f0d8dd00c48aa107e488952b4ae11198e
28465 F20101109_AAATAG lomascolo_s_Page_112.QC.jpg
90c4d84aa0a1100503a7d7ea93da2738
2915b07e3d361210e67fbdf1dfc1f77c8a97e5aa
13880 F20101109_AAASVA lomascolo_s_Page_043.QC.jpg
61265bb80ca64a630e3fac4a5383218a
9656eb3f8028574b30afc18f0048aadefd018888
27955 F20101109_AAASUM lomascolo_s_Page_036.QC.jpg
2c49da637b4d5e1b86a6486c50b7a944
d97ca256a6b1d843b9fb27e7c99a7b222b01ccc1
7328 F20101109_AAASTX lomascolo_s_Page_028thm.jpg
3495f2cc03919dbe899e4b05f7dc082a
e5b69d8b09e4b2e3f7059c9c60f9bc04bb605596
82613 F20101109_AAARRK lomascolo_s_Page_065.jpg
bfefc68e5c2f68d77c303d94df14892b
2e853b2c5498a3eed574e451e6c7efa2360797e4
22742 F20101109_AAARQV lomascolo_s_Page_050.jpg
9e409666771bee2aa8c759de850602a2
eb550ba6a630c3ccc89bb324c7ab29cce1538084
7616 F20101109_AAATAH lomascolo_s_Page_112thm.jpg
cdc88b3ada0f2bcb0e64fdf217caf8a7
ff6d7d23e406b901de3bcc6102c33914eb6357dd
4080 F20101109_AAASVB lomascolo_s_Page_043thm.jpg
504f7fe356503afeabafe9af966f0932
a12131d615535329bcecc4b8c7274deaf4b59a3f
7365 F20101109_AAASUN lomascolo_s_Page_036thm.jpg
33205eedd44679c31401f5ec196ae686
f34456b472253af30ee0a16ecd7419a0f38eda1c
28348 F20101109_AAASTY lomascolo_s_Page_029.QC.jpg
55432525f30e3732f28a3507ac612463
f4100cc95621398fd8fb5ad838c305277c93c006
46124 F20101109_AAARRL lomascolo_s_Page_066.jpg
20706a58c369a0feb57b2af9a0c3c2c3
4752f88ca585af27cb50b3f52f5312c13c867c25
87906 F20101109_AAARQW lomascolo_s_Page_051.jpg
b66ed3ef71bd0c20a4686caf9dae7744
d9bc7203279c73ecf78dc88c30a53112e1785fbd
27149 F20101109_AAATAI lomascolo_s_Page_113.QC.jpg
231d082078f13fcebffc2e8335b0742e
1d2b02f84b5e28bfa92feaf835f01bcf927bf4f2
12293 F20101109_AAASVC lomascolo_s_Page_044.QC.jpg
a50a678dbf5c83b0475beeb284e146d7
310489d9640cbdc074b95f1071b0039352cbdf12
F20101109_AAASUO lomascolo_s_Page_037.QC.jpg
0ce719c4662e42d7cdea1d3b63f419db
be8b9095ffc111943c312ce493a6b0e0d4f08bce
7655 F20101109_AAASTZ lomascolo_s_Page_029thm.jpg
b70b34095b35faeb486c698eb60536bb
a4291b565a5677647a72138b99aef169276fa863
83806 F20101109_AAARSA lomascolo_s_Page_081.jpg
20ee1fc6b235afc78b731cf3fd34fb98
b9b1450e4a44269009822f1b8bbf7f60e55ab31a
22599 F20101109_AAARRM lomascolo_s_Page_067.jpg
4c606562d52b78a3dfefb589f46906ed
a3e0d7f252ef49f55f7348f4ab63ad3e77f75e1e
81673 F20101109_AAARQX lomascolo_s_Page_052.jpg
ee0cfe20e0bf709e5346582a6ed7295c
28b85e1b53dac42a1baa6bdfc97c36f6ee1cc494
7355 F20101109_AAATAJ lomascolo_s_Page_113thm.jpg
253cd8062129833b51f1c128ab6a95c7
856b4a2f0c575c8807480ca761af2c381318726d
3840 F20101109_AAASVD lomascolo_s_Page_044thm.jpg
3168f45cb4160a8b125fbed6e0f244f7
42d6b38e4eb5c42d7398d4d7b75f93d8a9cdc4ff
86295 F20101109_AAARSB lomascolo_s_Page_082.jpg
ae6959419646e61bd140bf965fc07487
062f39149f2522d860eb67f05524c5c0a78106e5
74634 F20101109_AAARQY lomascolo_s_Page_053.jpg
f44c485ade72a6a755a4b6490409c207
8830c2232e97e1fa1b11675a2eba5ac601c16140
25454 F20101109_AAATAK lomascolo_s_Page_114.QC.jpg
255a0b441c38de585a7975601322acaa
4928bc685432d7e495d169418d8de11465fa6b2f
11066 F20101109_AAASVE lomascolo_s_Page_045.QC.jpg
9436fdbae1019bb19a52072d84cbd879
a98f06f373fe5a0739a7e0ee5c844963696c3c4e
7637 F20101109_AAASUP lomascolo_s_Page_037thm.jpg
9af36093ec8754a31f1ec0c8cde1c908
983c0c69ca15855ab831f3cd6b1718411be6b535
85543 F20101109_AAARSC lomascolo_s_Page_083.jpg
96688552cf7bb5df212e81dfb52ad30d
100a2457d6d6ba91c3163c4454e9caaac1300404
41220 F20101109_AAARRN lomascolo_s_Page_068.jpg
29f979e7fd8ca2ff49202753a2e54406
436be37ee4a3295f81ab26ba592ef1b30b974bac
87864 F20101109_AAARQZ lomascolo_s_Page_054.jpg
af2a0e77203253e63e8924e13b8ec8b0
686ab085b2f59c2c8707c5e9d14e83c9bdf0dd41
7211 F20101109_AAATAL lomascolo_s_Page_114thm.jpg
aaef980abea5faed938cdbe6deeb4b08
03103988b52976d9c3273a03a989490b8b37b6a6
3555 F20101109_AAASVF lomascolo_s_Page_045thm.jpg
479e22da332495c21da4eb2b440ada06
d2db44a968438a2a67176883c45ab252b37cfc2c
9998 F20101109_AAASUQ lomascolo_s_Page_038.QC.jpg
38fbab45dee1fdefde204303cb535a69
54cf16f7d9219815648a48f79e471e0b974739e9
88954 F20101109_AAARSD lomascolo_s_Page_084.jpg
dfd46a6eb483a2293bf7ecd9a80fe71c
7b71fb35ac9caad7c37115034b186c7da599fa32
21540 F20101109_AAARRO lomascolo_s_Page_069.jpg
62267a24f7ee69943be4156814867837
53a327ccbbe5aa11b7d8cdf4a42b81c75c417bb0
26486 F20101109_AAATBA lomascolo_s_Page_122.QC.jpg
1dab963a174b243991402d06fa0d3c5b
984bb032b17c73eda4affcfee076e0ad0bd86142
28343 F20101109_AAATAM lomascolo_s_Page_115.QC.jpg
8b977e068b16487187c574c8cc67f1af
ac787d38dfab35ee2955b40f73c7aa0fbf6f0566
10194 F20101109_AAASVG lomascolo_s_Page_046.QC.jpg
055c7c65967f0e3e81c8297da95925c9
67b5bd8d6034df5933279ea184023249e0f670b3
3119 F20101109_AAASUR lomascolo_s_Page_038thm.jpg
a5a9b7840a5525ef2e6c812022c2561e
bf81778090c6303d79c253fc0c2abb3c950f3b83
83249 F20101109_AAARSE lomascolo_s_Page_085.jpg
3b19631c410d94a8e94fa1dc7d4814c7
e45880a1034f814f276ef6cc6d9cb7be6cd3f672
89778 F20101109_AAARRP lomascolo_s_Page_070.jpg
7dde0a7846beafd79ee546182d990069
67721a902fa1b3972bfa2b1c75ff8042acfb0515
7087 F20101109_AAATBB lomascolo_s_Page_122thm.jpg
ba13a75a6a5696748ca2ae8c9b13a94b
8304856aa6373076067a8f8f7636046875000ffe
7547 F20101109_AAATAN lomascolo_s_Page_115thm.jpg
17d3536e9965a7ff4701fa982c18a76c
682c807f98cf00097f446335f88b7025ae75c038
3327 F20101109_AAASVH lomascolo_s_Page_046thm.jpg
c564696b1b0301040134ca4e9474aa6a
87c68c519468639e7fe29bb9196ce11bee4ff842
19245 F20101109_AAASUS lomascolo_s_Page_039.QC.jpg
8faa2b7729d505e513474b393238861d
ce1f2f50fda7a6f69d08cb55efbbbe7ca5dbd490
90451 F20101109_AAARSF lomascolo_s_Page_086.jpg
2a9d4db6ec4e8db587a39b95c12e02f8
a101331124d875e6ce9814510d2db2c094170d27
83518 F20101109_AAARRQ lomascolo_s_Page_071.jpg
6a1d867b2b43595980d98dfb6c69f3e4
8300229ffb124687345c9cc133732a3dcefd9a1b
11777 F20101109_AAATBC lomascolo_s_Page_123.QC.jpg
9aa96f4e60b6cce8f5a1ee87726a3d16
6f40c238a975f9196a81e76d2a29d603606ef8c4
25706 F20101109_AAATAO lomascolo_s_Page_116.QC.jpg
0fef1ccff3b8cf2659ecfa47d711fa71
99af6c8b6199d02514c5d90e2c10a36df601766f
7813 F20101109_AAASVI lomascolo_s_Page_047.QC.jpg
618ba2a687cf51244dd1b524e3cdb9b3
0e23bc2ca5c4e105434c6453b0e2a2fa3d0e907a
5623 F20101109_AAASUT lomascolo_s_Page_039thm.jpg
7e9b989bf1efbd9f373d43ef566504fa
57c3ad35a2929664c4913c5b8bb2ac72d9b6683b
89244 F20101109_AAARSG lomascolo_s_Page_087.jpg
5a170748b01349f53eff632e3feb87de
892888f6c858825a7b3b5438e789244f29d2ec98
89878 F20101109_AAARRR lomascolo_s_Page_072.jpg
c6e873b64595397bfa7dadfa76eac05c
19ec180e6d5ac75db3d1e3a140dd86669242643e
3611 F20101109_AAATBD lomascolo_s_Page_123thm.jpg
acf2e7539df8c41f4e1d35371fd1f0a3
44c5fc503aa65b34f426a180c180ecbc6799434b
7047 F20101109_AAATAP lomascolo_s_Page_116thm.jpg
addc8b1baeb833cc822fdb920c26c9ec
b7a1df4d3fb132520e75671e0b424b08e3a27e84
2958 F20101109_AAASVJ lomascolo_s_Page_047thm.jpg
0bf487af148643dd1303f5a6bc782181
afe42ad485a011b3d3044f53daf45a96a8ce1bec
9626 F20101109_AAASUU lomascolo_s_Page_040.QC.jpg
af4e73ec588587b4177f2e8d42916fe2
bc2c5f2526937cbb8cf671425076087691de3373
87363 F20101109_AAARSH lomascolo_s_Page_088.jpg
21b2fd30119e881efbc040fdd9ed5975
7c0ec8f348d14f9d684a445ecaee3ac06d083e5a
83336 F20101109_AAARRS lomascolo_s_Page_073.jpg
e771cf954a18bda04bce9bdd37738e0f
73003cdfb321eedbe58953d08794a9b5ea974d4c
6918 F20101109_AAATBE lomascolo_s_Page_124.QC.jpg
add38ed4e13bd10b6f5f5d7a4e715b0a
49aedb219e9796ea36f1011aa791890854386b9e
27393 F20101109_AAATAQ lomascolo_s_Page_117.QC.jpg
c620f0e8cd1d62921bf7edf2962c5839
04807c4e8a1f02fa7205d064a1e2d3a25a4c2717
8979 F20101109_AAASVK lomascolo_s_Page_048.QC.jpg
97cb3a6c6a72c41a387287a53ff17b30
4fbc4d9861cbdce2146540dd3e322c9cdd6161d3
2976 F20101109_AAASUV lomascolo_s_Page_040thm.jpg
9d5ba1f33f0afa2ac7bfe620b782056c
ee2f9cb7b63508578cc2c2841a65169b64b92bfc
81459 F20101109_AAARSI lomascolo_s_Page_089.jpg
e5eb2ed4f9cc47343fc0c3a62b68e39f
141634e25306c235a4bd2053dab5e18a3b0be67e
89168 F20101109_AAARRT lomascolo_s_Page_074.jpg
7d318b8fdb2985e3c5a74c73a9f8f9fe
904ab074a27f34920d543cd430b5aa63f68ca815
2549 F20101109_AAATBF lomascolo_s_Page_124thm.jpg
986c621009433151bcef174f926f284a
80664ba40d04d26686c644590213a8ed26279625
7363 F20101109_AAATAR lomascolo_s_Page_117thm.jpg
08ab716ed2a73f5be5f288aadcc6a1bb
d6c15c3de6b5a6a98638df21728da8ca9b8c30eb
3192 F20101109_AAASVL lomascolo_s_Page_048thm.jpg
a7f2af731de79d1fb4ff2776a43f119c
97c1e34146607f06d5872ad4ce0548df88f70cdb
14142 F20101109_AAASUW lomascolo_s_Page_041.QC.jpg
7fa6022ed768ea688e6e78221482af2a
5634a547db2032a1fcfd6fca0cc32c37a808e5b6
82779 F20101109_AAARSJ lomascolo_s_Page_090.jpg
452c439b3b27075212e704298f754b0d
7047aa68b3001887b8b725ce3d74172089eeed60
84924 F20101109_AAARRU lomascolo_s_Page_075.jpg
6e5b002dcb87fbfd880780f51fb5d070
35e5ef0699c50c67988e3c52503d5bba78193d16
10888 F20101109_AAATBG lomascolo_s_Page_125.QC.jpg
5a7cc6db967b8fc74b0288cf44696844
d655592f816e154f7f0c42fada80236143a925ef
27262 F20101109_AAATAS lomascolo_s_Page_118.QC.jpg
fd6162854196aa5014055db9150aef2e
ed0b06038c4e707ae393f8018b946446d991fc7b
27682 F20101109_AAASWA lomascolo_s_Page_056.QC.jpg
77797bf7b87a59712a5579eb6b2cda4c
3f93595bd0bb67ba00858f4a74cd514a7816fed2
9687 F20101109_AAASVM lomascolo_s_Page_049.QC.jpg
7a621dc8064759ece5ac07eed3d5b66e
66e0357439b316f9d21c6c2f53566344206a8200
3655 F20101109_AAASUX lomascolo_s_Page_041thm.jpg
eba7ebc21cf4f8de1fbe0e57a8e4f4a3
9abdbb12096e13ddde3a153d9a426a05ea915c7a
89948 F20101109_AAARSK lomascolo_s_Page_091.jpg
d58f177a74fc630125c506b759dfdbf3
df78e84a15ccd3746a3b9feccf9327020454d5ad
81550 F20101109_AAARRV lomascolo_s_Page_076.jpg
f351193c98744d97c29008c635c09147
a35a0be3e7b0fda1dc9622e87f9bfc0cc782ff2d
3482 F20101109_AAATBH lomascolo_s_Page_125thm.jpg
6eefbd615175cf046eb0fbab730087f6
5cd6cc89e085df9fe7d5e8c44d362c78114a182f
7245 F20101109_AAATAT lomascolo_s_Page_118thm.jpg
ca80f45efdb5aa310957121d71a3548a
006b946937de42112c1b278e62fe309b2627e7df
7305 F20101109_AAASWB lomascolo_s_Page_056thm.jpg
9150e811834748d8333e4f8f2d555add
a5105f4482a390c3f216e9c3ae46df1f23248ccf
3323 F20101109_AAASVN lomascolo_s_Page_049thm.jpg
50fe8fee6f300edcb1f0f9a4deee4b65
432af57da89471a2ac62c4ced8c846022dd30a53
16999 F20101109_AAASUY lomascolo_s_Page_042.QC.jpg
04ac0b0dbfa3066c98c1fb78dcc52bbe
2aab03e4b3420e4a6ad445dea35914dec0e1569c
89647 F20101109_AAARSL lomascolo_s_Page_092.jpg
2316d7d2b3be0ebd685515cc27903956
83813e8ee8d9107ad9f5996d414a7f5368190aee
85643 F20101109_AAARRW lomascolo_s_Page_077.jpg
098468023cc7363c590d99d4a54a3a8f
c1cfe7d8eb4739b8cd56e067a7f290d15fe1d00d
6019 F20101109_AAATBI lomascolo_s_Page_126.QC.jpg
cb45f500e89432d4b463db81f30a6a7b
a90df226a2f09bf23f696efc58b48ace4de91c46
28676 F20101109_AAATAU lomascolo_s_Page_119.QC.jpg
ce30144372a3e2b574a70def52c2d80f
0a5938ecd854ac38bbf720fe0124cb48620a5cdb
27533 F20101109_AAASWC lomascolo_s_Page_057.QC.jpg
26a7ca26ce43621abab81fb3fa616586
bf8e0cefe926a26fcadb7d77a72915f64cf45215
6995 F20101109_AAASVO lomascolo_s_Page_050.QC.jpg
55057e1daa284edc1d29a806d028680f
6acea3c932d684c918709edf062a9716b54d518f
5396 F20101109_AAASUZ lomascolo_s_Page_042thm.jpg
c88adbd190ba73023d7c39dd11f7117f
3023920f7287e23c9286226156f26516012864fb
86408 F20101109_AAARSM lomascolo_s_Page_093.jpg
ab852df572a33bdcfebcdba22921173b
7692c05ea97b8f4ba767483f457257a5ad40fcaa
90051 F20101109_AAARRX lomascolo_s_Page_078.jpg
722ac2e1f2e55d58c647d3bd8eb72e85
578f063787a96c7856a71a0e0f4a403929883791
87110 F20101109_AAARTA lomascolo_s_Page_108.jpg
17faf99a154592d60ae5007d4f5bfa8e
7f855d2d53ce558637e035c0ba473e9a96ffc691
2001 F20101109_AAATBJ lomascolo_s_Page_126thm.jpg
bbf8ef78e80e68f5543a25c8fa378062
b6f41fb006cabf841b9829d90da80f6666c7d3f6
7695 F20101109_AAATAV lomascolo_s_Page_119thm.jpg
6f075a47dbfed254ddec43cd09394990
81eda9021cf16524144c154d7c7bc7bdbd4eb8a0
7442 F20101109_AAASWD lomascolo_s_Page_057thm.jpg
9c66a9b1e6c3eee0835ecb483be4ada9
722f4ebd3ef18a194e8d8a5ffb1b7b597b7ab513
2498 F20101109_AAASVP lomascolo_s_Page_050thm.jpg
7a72cbf1f192273bcbbe2ae98d7bed23
f2c9949b3c546b7671cc9a2bb4691c9936d675ab
46865 F20101109_AAARSN lomascolo_s_Page_094.jpg
67e1791ad579e8afbde32d17ae233609
5c044c9dafee904f9fada7f6bdf29dbf8ef76576
86113 F20101109_AAARRY lomascolo_s_Page_079.jpg
40d9a252f0d33bb9dabd827db28f2bfa
9199b7047993d545c3dff9f9518014e6ae11f43d
89190 F20101109_AAARTB lomascolo_s_Page_109.jpg
b9ff3f3b222297171a2a6023d529295b
d6bdd059b519b5675d40fe537ee052af278e3f4a
16404 F20101109_AAATBK lomascolo_s_Page_127.QC.jpg
9feff26d19cab49771468dc1f5cff7fb
6bb1c5a9e1549eaa9839265dc72a4f7583409e96
26608 F20101109_AAATAW lomascolo_s_Page_120.QC.jpg
4fb80ddf6bbbf62fa39ebd43357b3c1c
6e883b150e7b29adf2148260db5a6afbf59a80d3
26152 F20101109_AAASWE lomascolo_s_Page_058.QC.jpg
02846faeb808a1801e406b87b354ce0a
3d9e014afc9e5b62f67b10023e69b623e591de9e
82909 F20101109_AAARRZ lomascolo_s_Page_080.jpg
8a663744824d76e6d9eba8b6a8a3d8c5
199c99de73a1c1251fedc161770a29f85d3cfb24
85804 F20101109_AAARTC lomascolo_s_Page_110.jpg
656bc8bee693ac99bab5b20fdf386257
7b1fa4eb29135870640bb49d97512b815e4355ba
5282 F20101109_AAATBL lomascolo_s_Page_127thm.jpg
7dd9c199ab6f9388b4a0ca08427daf6e
e3902bd2def9cd86a041506d93f293738d48cf86
7390 F20101109_AAATAX lomascolo_s_Page_120thm.jpg
2a872261187e04e8f79b9b866177f3c1
b4b49927e60635eb83af4c7fb9b5d6781d6e4d08
6775 F20101109_AAASWF lomascolo_s_Page_058thm.jpg
8655c6460826fa8d7a917a81b835ab54
cd34103ea9f8bf4a7f82088a4fcd3116c3004c87
27971 F20101109_AAASVQ lomascolo_s_Page_051.QC.jpg
7b198a86b50c0d29109dabf1ff3d7f07
009a3b5c2919a02c805e9a23f4bb1f760f9ea878
49210 F20101109_AAARSO lomascolo_s_Page_095.jpg
fb6c478490eaf0972159cdbab296a189
3befa967cbc7ff0ff81141812aed082837c626cd
76171 F20101109_AAARTD lomascolo_s_Page_111.jpg
6d7eb67146d6e7ba6b099bbdd7b9d4a6
94dc6b6242e292f878c1b74afd4102d4ecc74ff8
20577 F20101109_AAATCA lomascolo_s_Page_135.QC.jpg
c1fc82e933bebb96064a57fa2dc5c7da
bfefabf2668d3433ca47e141546e846b7af6c0bb
10062 F20101109_AAATBM lomascolo_s_Page_128.QC.jpg
3854caaf3e6a9fd93f4828bdca6da409
3ce06a2306bca01693bb4f24ca854efccd8fec23
28689 F20101109_AAATAY lomascolo_s_Page_121.QC.jpg
8b491b63287df606b18ba171c35812b3
684c594dfed65918fe88642b555e3588f9d35a40
25269 F20101109_AAASWG lomascolo_s_Page_059.QC.jpg
653b8b42b84cc8c4a194ebaee5ee25af
63986a6ce7c578f83a144c716b0c872f945bc596
7464 F20101109_AAASVR lomascolo_s_Page_051thm.jpg
03eeb6329ee1125514112f0a6157f1a1
efd80d6b716e4fc20ddbf8be6765a785605c7e3d
38429 F20101109_AAARSP lomascolo_s_Page_096.jpg
9a65135fb97e29d4de554ee483b1d543
647dcba9af3cb260fda050314d3eccea8a7ceaa9
90470 F20101109_AAARTE lomascolo_s_Page_112.jpg
c03476e34870d8d79322d17e89ec7d21
38f805a32e5deac8a405e6288f7bf1496b86a8d1
5643 F20101109_AAATCB lomascolo_s_Page_135thm.jpg
68268219654d58a2e5ba54029729b755
9a861313a2c67172ea2a1965d8aae0d1376ea378
3637 F20101109_AAATBN lomascolo_s_Page_128thm.jpg
fdaf8a1d8f37d7ab0a494f350e7b8c1c
b8244e4f81fbefdbc73a6f2127532c2db2442078
7605 F20101109_AAATAZ lomascolo_s_Page_121thm.jpg
0bdee2e69711050e29a01487dd717f1e
db4ec0b7f0cedb8baed5771a13ea2cf9c1164840
6948 F20101109_AAASWH lomascolo_s_Page_059thm.jpg
838efd756d358fad29566bdebb7f9d28
037f7f8c5d275a7d97b7a77a492e55ac242c44be
26177 F20101109_AAASVS lomascolo_s_Page_052.QC.jpg
fd93ea28be6413d50f63cb6580746828
4680179ce321716caf0bfd9ffabec8e4cae4d01b
19153 F20101109_AAARSQ lomascolo_s_Page_097.jpg
49a5b02107303f1ec1a33dd53f96c98f
4babea5a341915ad634f105dab304dc52e4e585c
83371 F20101109_AAARTF lomascolo_s_Page_113.jpg
30e2d8a5ac0cee852ba4881a03271346
119388a1d8f629c174ecdadc835d5b0a166109fe
25247 F20101109_AAATCC lomascolo_s_Page_136.QC.jpg
de97e4ac099a536d11cd0229e69cfa31
bf7604a6c91cb6fa40e07f5f70f5db938cabe61d
13903 F20101109_AAATBO lomascolo_s_Page_129.QC.jpg
8b15c505b2f0e8b20e686c12fefdc58d
bcd1b5403452632d54047f1959950d324a269247
27048 F20101109_AAASWI lomascolo_s_Page_060.QC.jpg
d6ee2f4b79c3009e09f405fb7e4dfa5d
aca83cc6008d4b2bbcc22d14bc514969b6254ae5
7102 F20101109_AAASVT lomascolo_s_Page_052thm.jpg
95762cc281cee13d3e289f6e5a9d2c0f
7933e14735dfe5e8ce9bee1449baa23759b084a4
45552 F20101109_AAARSR lomascolo_s_Page_098.jpg
af738b9ddc81a1fca5eb6dfadf9022d6
ad0090ec9069a214065a1f8509eef4b7d0a7d707
80603 F20101109_AAARTG lomascolo_s_Page_114.jpg
0a9802ff869c8610b9c5dad05d4c04fa
2bc36329b7e103fd1c335d41f1a4d391db0e6f57
7046 F20101109_AAATCD lomascolo_s_Page_136thm.jpg
316ae7ce84879a56f0c03c463818eed4
4b0240370279153ef3a28f8e248dd54becebc6db
4021 F20101109_AAATBP lomascolo_s_Page_129thm.jpg
307855ea5987124791882a5dcd2a94d4
2f88f6530fdf0d3a1875bdd15fb5e0bc620f112c
7350 F20101109_AAASWJ lomascolo_s_Page_060thm.jpg
6cc65c85bfda837eaa57b15d4e78690d
d6daf02755185205c28e025dc61dbcdfa6d71c13
23470 F20101109_AAASVU lomascolo_s_Page_053.QC.jpg
7e232917dcaf670a34e7f816db172e8c
6fbd2d5a5d93b8a1dbc4f10cdd58e675165c704f
16518 F20101109_AAARSS lomascolo_s_Page_099.jpg
24db581e82b89dcb4eeec7a475df48eb
772b8117e12747e13fe33cfdf8aa4ac13ed065bd
89671 F20101109_AAARTH lomascolo_s_Page_115.jpg
dbde222e7fb50eb3f3ff4d29dc5d0708
ec4ce27392104f6bc260ad02be60fcf48bfffe8e
26479 F20101109_AAATCE lomascolo_s_Page_137.QC.jpg
15dce81b85896b4b9f4e7384a678e7c0
25d03e5a2e465e0ae3dc757bc85a75568dbb21c4
14917 F20101109_AAATBQ lomascolo_s_Page_130.QC.jpg
af8c7269d320dce086820bce2d544696
2bba47fe5017a9dca20107c0115827898f76ccc6
27766 F20101109_AAASWK lomascolo_s_Page_061.QC.jpg
838cefbcb21752c9a6631e236d541aa4
39d88804753ec442f7f51192fbb7d03a78705654
6826 F20101109_AAASVV lomascolo_s_Page_053thm.jpg
653e7291031d106cc1cc072eca05f671
6d8ba068ee5d21fe2350dc1a524a5fa6fb3bbba5
14315 F20101109_AAARST lomascolo_s_Page_101.jpg
1e8eda1e000dab2a22338497ababc160
20baae8ab9a2ea822a4a4e1cc56c88421783c1a9
83437 F20101109_AAARTI lomascolo_s_Page_116.jpg
471b21740c53ba96da9aaeccd4d92e99
b50082bc6ebb77447bfb9474c17011553b212425
F20101109_AAATCF lomascolo_s_Page_137thm.jpg
f1a77e804b990e950735be98b9a68078
a18ce51df48c23f5c26a06df4811f66248c42465
4595 F20101109_AAATBR lomascolo_s_Page_130thm.jpg
1840f7c41993520508b93b3ddc44445e
f23efd32dd73e746849939feefaea6081a7a06d6
F20101109_AAASWL lomascolo_s_Page_061thm.jpg
90907b4b957c97a5df5e834fedd36df5
dc2401a27776c6d0a4f8ffa33fe7404cc4e09142
28254 F20101109_AAASVW lomascolo_s_Page_054.QC.jpg
22ba104574be7d52dbb0c1f564a24c6b
5eedbe1e9cfb9535e5d79fa45cfeab0982962ac3
26063 F20101109_AAARSU lomascolo_s_Page_102.jpg
b47bf21b1bb5c63a24656a132b79e311
6dd870300b053f229687efb2e4f286a4be349775
87249 F20101109_AAARTJ lomascolo_s_Page_117.jpg
f6c69b8725226b1201412dcdf3dea703
214687a68edb5ad4dad1b3ed88d93f8c043d748b
25866 F20101109_AAATCG lomascolo_s_Page_138.QC.jpg
809d1a5f9ab8fbe70bae3d8943299b09
2b24388da65d096c13c6ea51c6b73f942baf7a3d
26819 F20101109_AAATBS lomascolo_s_Page_131.QC.jpg
6124ddb2b2ffbd1f0e0c115100fa083c
85c4e833a4115b372553f85fe715fd534a0c8d6e
6462 F20101109_AAASXA lomascolo_s_Page_069.QC.jpg
943c0af618168702fef79facef48fab6
c2c89566aa26278edc513f358272c6162c736bee
27980 F20101109_AAASWM lomascolo_s_Page_062.QC.jpg
dd46c6d1d52c42daacb1985594c0b6c2
0d2ca9c1f56dbe193fba2dc14381d1cade365f8f
7704 F20101109_AAASVX lomascolo_s_Page_054thm.jpg
ee63fe4735fa2c5acf1698286e93814a
bc14c953cf51dd935e4cca950943b6b93833fb6c
48853 F20101109_AAARSV lomascolo_s_Page_103.jpg
0d4365751a8d69657abeee67a0b4ca40
9c3273154309beaa75057d4465c3c5396dc19e34
85464 F20101109_AAARTK lomascolo_s_Page_118.jpg
677a89eedfa1c791919053252c258356
3157503195267e6a2fd154a0b80f990feb6c3310
7121 F20101109_AAATCH lomascolo_s_Page_138thm.jpg
493c586995b1665bf691cac91918803e
2213f35281a7e9c03bad30cddb662c3a1b4d76b5
7138 F20101109_AAATBT lomascolo_s_Page_131thm.jpg
9e75478b4c0217de68cc5e08e781b810
690325d44ee87c40b54b12313462b5442ef5524d
2222 F20101109_AAASXB lomascolo_s_Page_069thm.jpg
b4aa8af7f8f049204895b72f438fada5
1f8fbf521688c3b2a4b6195cd133d92b84bc18d4
7373 F20101109_AAASWN lomascolo_s_Page_062thm.jpg
12047eedc01cd5291f541b38b42cc3b2
c1b3be590b323da89e2eb6a66fd642ec73b00160
26740 F20101109_AAASVY lomascolo_s_Page_055.QC.jpg
9a47c26150a784c854203034036e6c1b
b4bb0a6c45f0fe7cd847b23ba0672638cc21d7e1
20712 F20101109_AAARSW lomascolo_s_Page_104.jpg
4ffe95be6e072011ad70f798b7638d35
8d1bf6385aec43c2594a7621a0a5d6ee45deb95b
90389 F20101109_AAARTL lomascolo_s_Page_119.jpg
6e5e7e9b1a62f97d843ae9d050e18711
cecf3e8b43da478317daaede3cf3c2b90be7b8af
28338 F20101109_AAATCI lomascolo_s_Page_139.QC.jpg
98940df27ab965d8f47e8829595f2d23
dd32cef071941c4a21142ac830f172c529e92297
27929 F20101109_AAATBU lomascolo_s_Page_132.QC.jpg
d0a886e26c258c1c5255967e4bd4bdb2
fba6ac63cee7b35cceb5e572ed106e359fc6e9f0
28243 F20101109_AAASXC lomascolo_s_Page_070.QC.jpg
e92351135aca7b19d3d31064a2f16cb4
db014ac1bd552fd5de292f40dd44c4a5a01c1bd5
27089 F20101109_AAASWO lomascolo_s_Page_063.QC.jpg
f3f3ad8fca19353e7ad5a00402f8f162
43875d457665350f448f7168d196a201af2af590
7498 F20101109_AAASVZ lomascolo_s_Page_055thm.jpg
3321c70bdd6e5a9c12109a9df8925091
b7a972eecc795e69fc96889ecd0195f2279884da
54813 F20101109_AAARSX lomascolo_s_Page_105.jpg
3cc0519d45784801e56bfb22a2d690ff
6a298ce829653c294f4a26e768ce66bf3c6c33d7
90118 F20101109_AAARUA lomascolo_s_Page_134.jpg
7b2c83678c1451bbb64e91f6028d4675
7ceead8d34fa91705d420e9ef5dcb5df9bdc9714
84006 F20101109_AAARTM lomascolo_s_Page_120.jpg
c6e0a1efc65ff9fc6aae408a6f46b067
7abcc295f66ef4e1e4eb87d4bb2ec3e97f5143dc
25948 F20101109_AAATCJ lomascolo_s_Page_140.QC.jpg
18f5adedd2a70335045abacc8d5bfed5
fc905b89cd8b31bcac9de28134c41ec516904985
7522 F20101109_AAATBV lomascolo_s_Page_132thm.jpg
e92b15bc0e916fb9ba38c4b8e7c695d6
8e13a02e7def232b4e2d5974c6213824cfadfb54
7643 F20101109_AAASXD lomascolo_s_Page_070thm.jpg
e161f0fe15ca354af7dc0e3bbd988111
ad42e528a768d15ffa8866281fcd49ab1b421f97
7190 F20101109_AAASWP lomascolo_s_Page_063thm.jpg
344fce8e3111bb0cdf404b6d5f07308c
96b132bdfa0adc53c497e684cc7bd9b51ca948f4
66798 F20101109_AAARSY lomascolo_s_Page_106.jpg
8feacf0d701ece2ea4dbd7293f4ac099
a9d6894317e1ba2f0e24edd4e9f5199b1a62d797
64257 F20101109_AAARUB lomascolo_s_Page_135.jpg
527cfe8aa8ce46a113519e9cd4f12122
e6f00e9f94e499c908c39209e248f1d84d9d08fc
89903 F20101109_AAARTN lomascolo_s_Page_121.jpg
6af0d1ea9e58816d65489d21a1ba3bdf
24a9399a0ac6f2b53d5bb5f4c8fc831a91d53a6f
7126 F20101109_AAATCK lomascolo_s_Page_140thm.jpg
f13ac1fa775a554f909ef42e064efd2b
f68ede586339965ee9f6a483e7e11455d524a7e7
28496 F20101109_AAATBW lomascolo_s_Page_133.QC.jpg
f7415516b14b2bd4beca3574b52b8812
8f7aa9bcb3cfc62cb027f6bbe88598bdae43c601
26799 F20101109_AAASXE lomascolo_s_Page_071.QC.jpg
0ac491e201c3f9d09aaeb784cdcd6906
a03f376b4ebc6f664ef89da40b69f69d33f5bc7d
28358 F20101109_AAASWQ lomascolo_s_Page_064.QC.jpg
a3f27004e2cc34cc48788b066c51f729
0c5351e0b18cb2d1b7bc0e21a5b115fe634feb50
86484 F20101109_AAARSZ lomascolo_s_Page_107.jpg
35ac0297c9a47625e7ffa6684e9f1d95
dd540ffcf0fffa937d1d91eb1ffbac9d73b3adae
F20101109_AAARUC lomascolo_s_Page_136.jpg
588a63c217ac252b077dbad069feefa7
e33637ed6725630bf3ee3e934d9f23fe113722c0
82602 F20101109_AAARTO lomascolo_s_Page_122.jpg
702012a2f14a98ead87fa4daaa3833a6
fc7f0092a61ece291b78484ffabc2dcf48d84af9
25994 F20101109_AAATCL lomascolo_s_Page_141.QC.jpg
b452a5a978ffdc5d55adcf63fa3385c2
1a10bb770fd68fdec519cb4c5ed592c9bfac807a
7538 F20101109_AAATBX lomascolo_s_Page_133thm.jpg
b72fd2c1bbc3e983c4623da7048e8c0f
f59e2bb83eb995763c2b3348f5e172d6614a7d0e
F20101109_AAASXF lomascolo_s_Page_071thm.jpg
a4c8576dc88e065117501d55f948e86e
5a02e51ac2dd7f0d549a561480d4b3aa72603b79
90818 F20101109_AAARUD lomascolo_s_Page_137.jpg
3d136715617a4bdb903eacfad7c8c51d
818705aceb799f18d230430b2bbbb2bb9d0df0d4
7123 F20101109_AAATDA lomascolo_s_Page_148thm.jpg
805ade39e621005882546c7a9f902fb6
a70b46e4d9b80c451644d511d3be517cdbb12987
7524 F20101109_AAATCM lomascolo_s_Page_141thm.jpg
180d06e842e87eca3d5ca486a3a599d8
2c6993044a97d28ce2aed0406c31b225f23a23d4
28639 F20101109_AAATBY lomascolo_s_Page_134.QC.jpg
a0241e87863f82996531c1b0e791efce
b8c086cfae211e3d93e8341e671bf4d1ae498d0c
28668 F20101109_AAASXG lomascolo_s_Page_072.QC.jpg
0b08b1a78895f4fa633a8e73dd2a1418
493ebdd4e2461fb5b65d12c155504216354698d7
7421 F20101109_AAASWR lomascolo_s_Page_064thm.jpg
57d231e5ca3b0edf3a3e0a69cc14d354
4d470648d0d93131ba0642f12832045b97c0f231
87018 F20101109_AAARUE lomascolo_s_Page_138.jpg
727d609c002ffdab8dd55ad1aaabad19
90e449d03a81253ba126028a5a77a445652e37b8
38881 F20101109_AAARTP lomascolo_s_Page_123.jpg
a4b52dadb7533010cae424ba1731ae7e
1d65e9d1f53c6505a8c274fd24767f90c1c3edcc
28173 F20101109_AAATDB lomascolo_s_Page_149.QC.jpg
05b81b27c713c6bded09fdcf1a8c1d83
151ef6d1a04eb5817ea11c62aee2ffa063586df7
25119 F20101109_AAATCN lomascolo_s_Page_142.QC.jpg
e845153e2b061be36add071b4470484b
ae3570bfc87ce168e5428cb11f9e9e6b263a35b3
7675 F20101109_AAATBZ lomascolo_s_Page_134thm.jpg
1cc39a6249ef542d0b9ee6f2f5351a93
4296ddc9079727ff66c157c2ba371cbeab74692b
7532 F20101109_AAASXH lomascolo_s_Page_072thm.jpg
8ce8f83ca2c0a71e241d28dd723b744d
89ec1e2d2899980fc74750612fb2c648bbf8308f
25640 F20101109_AAASWS lomascolo_s_Page_065.QC.jpg
14122f2a886f68a16fd3552f01819413
be063339bab5d87e3c3ea8899dc71923246bdb36
98108 F20101109_AAARUF lomascolo_s_Page_139.jpg
2bcab3a24ce2761bc5b89811f82ae0ba
b1b6517531cd93bd6ae3c6128488dfe1e50028bb
19564 F20101109_AAARTQ lomascolo_s_Page_124.jpg
51cef71d094b140e4c9079675279477d
f3033c68f3baa27b45a0215353ae96133876eeee
7468 F20101109_AAATDC lomascolo_s_Page_149thm.jpg
50b20c9a5c84494f2c6a94d78a983316
998f476f4e6841b085f861d05323946aaeab4b05
7236 F20101109_AAATCO lomascolo_s_Page_142thm.jpg
9849e6734709747e5c3adf9aca5abbfc
e8282460bb12a995197ca6b690ca1d4619aff0b2
26729 F20101109_AAASXI lomascolo_s_Page_073.QC.jpg
c264509e380796dac1384e9d78b4a021
c69f84feed9e4b2863ccda4311bbafa0cf02b7b0
7021 F20101109_AAASWT lomascolo_s_Page_065thm.jpg
c380647695cfb1cb370f4aa9e605a8b9
cfbc4a3672857455295921880c5116757ba53d4a
89507 F20101109_AAARUG lomascolo_s_Page_140.jpg
716ef1a5fb83884c74f51a3d203e26b1
1e718e577fbb04b8f1b7e13d94cb7354a1ca39e2
37138 F20101109_AAARTR lomascolo_s_Page_125.jpg
46d02f867fafcad5e0631bcaa561129a
4785ac0adfea28b5ca76766e7d08fbeec0a2b03d
1051971 F20101109_AAASAA lomascolo_s_Page_140.jp2
2d5a9ca3ca026bedbe321db1c40932db
5a1c785f8c8ae7fb7020f3646a305da284b1c362
25518 F20101109_AAATDD lomascolo_s_Page_150.QC.jpg
b17e5558f48492796e683fae7c692995
e4ffa666902578827cfb44d69c9122a3b73595f3
25952 F20101109_AAATCP lomascolo_s_Page_143.QC.jpg
855a5d6ef111e590febb61f2051c472a
4642ae40017219b2baa4bb0d813bceab06542fdf
7262 F20101109_AAASXJ lomascolo_s_Page_073thm.jpg
398a9bbbd3bedaebd234528df9c1b29d
a91b52b9fdec3f76acf774bfd6c33548ab0430e3
14707 F20101109_AAASWU lomascolo_s_Page_066.QC.jpg
bd03fced0aa75601804c31868679130b
8e9e5e9ccafe844fb92a121453120d6bd34477fc
93047 F20101109_AAARUH lomascolo_s_Page_141.jpg
16718d55dbf95242d4efd7fb1708d04e
7dfc4c324070b29483b4eb52ac4f2f01576a4d13
22072 F20101109_AAARTS lomascolo_s_Page_126.jpg
d48ce66d69e9cfbc48cf447b74be27c7
1ec5049cdc3b0b5e2b35d3ddb5f4079ebd73da5e
1051959 F20101109_AAASAB lomascolo_s_Page_141.jp2
fed609b3529708edb7a450b250990efb
48ade84d12ba91c854cc108338ace5ba0a301f32
7254 F20101109_AAATDE lomascolo_s_Page_150thm.jpg
364289c371e6cb3fc31ce76eb0a82bfd
9d52cb215c80c8f1d3300a4612d012ed01acd74d
7139 F20101109_AAATCQ lomascolo_s_Page_143thm.jpg
cab57582cd6920e5af7a5a151458983d
511a27c1ac4a28817af12de005d0f89c23f36a57
28020 F20101109_AAASXK lomascolo_s_Page_074.QC.jpg
ff8c1b3f8593bd2f94a35c186d606b20
6fd53a70b3c7ed5960b75f4596b93489e750f21d
4124 F20101109_AAASWV lomascolo_s_Page_066thm.jpg
971f907c61fdb1147dd32ac1415fb6eb
9f75b23d2384fda7f1098e534607f96794595df5
87475 F20101109_AAARUI lomascolo_s_Page_142.jpg
668e05f735c46ae60c3c844e926f3f23
563fef6b8021a2d2fb44c56b39420e9b49ed7ed1
47304 F20101109_AAARTT lomascolo_s_Page_127.jpg
87aa2c25055aa614693b0f407ed6477c
a17d1aa3fb800df52421f33f176242701f27ca36
1051939 F20101109_AAASAC lomascolo_s_Page_142.jp2
2c4aed0616daa1a9c0603f3793b899bc
48bf1649689388db26b7396bf34b8b8dcb17eaca
7250 F20101109_AAATDF lomascolo_s_Page_151.QC.jpg
0dd368ab1aa1f88cc726ae87ebcdb933
6e8161a281653f724f054dd3ac262d383bb10ed2
26022 F20101109_AAATCR lomascolo_s_Page_144.QC.jpg
bddf2d88c08d4edd6f768bc71b993e23
df1ec7a90d3f4cd785df22cee5489c4e8b49713a
7520 F20101109_AAASXL lomascolo_s_Page_074thm.jpg
5485b39ae7b5d99b8c5ecf56150ad21a
7b1bb4415ca82e99ff5df4ff619992c8d7c9943e
7261 F20101109_AAASWW lomascolo_s_Page_067.QC.jpg
a193a38dcb57070388ad18f04b11acd8
9503a30261ab2a5fefa2cf593ad1aab9653ae0ad
89563 F20101109_AAARUJ lomascolo_s_Page_143.jpg
38d5f37e26e158d557ee8a4740750518
7bd676b43a56e44adab2aa8462b4c77bc9710721
30347 F20101109_AAARTU lomascolo_s_Page_128.jpg
1bf4bdfa3254eaaa725a0d3367346212
242fe4eac6a17b0d8dd6d5bd05d227b4ffb11899
1051916 F20101109_AAASAD lomascolo_s_Page_143.jp2
c67290fde3e80ead6d61840c8d052bce
a7516c499a1364308599710709c00b716f065134
F20101109_AAATDG lomascolo_s_Page_151thm.jpg
741dbe8d71e780cd25944f477b7465f8
3d50c0803a6fa96e5e24d08e9f8061809b148e17
7370 F20101109_AAATCS lomascolo_s_Page_144thm.jpg
71ba64ead46034f45ba529cd8736f355
012ecadf5e52b01744178be4019dd0d9ef1a7f05
7391 F20101109_AAASYA lomascolo_s_Page_082thm.jpg
0630b09417958e80bb9105bed0aaf888
441481ab2b24bedadccfe622f98324f417e88138
26356 F20101109_AAASXM lomascolo_s_Page_075.QC.jpg
49018499bfbbb219211879d25095c475
12e092a2ef8b6c5e20c87064753d8e67868967d9
2550 F20101109_AAASWX lomascolo_s_Page_067thm.jpg
8edcaf8173eb2893f5cfcbb27bed537e
185fa66d7cda4244c805af94c1d5f4ff025fc5f7
91218 F20101109_AAARUK lomascolo_s_Page_144.jpg
0d465886399abaaa6b603ff30431bf2f
b80b2a4d3c8bb2e3c5001c9eace734ecdf34cac6
43075 F20101109_AAARTV lomascolo_s_Page_129.jpg
6b2c94c3e1dbc9925be601ae4a109ad5
6d70ec608044420c2257261036de67e362c99810
1051985 F20101109_AAASAE lomascolo_s_Page_144.jp2
85bebd2f8506932766b5922a0cecf11c
688dc20169357cd8846d834cc24c6f2a1ae8d69c
25007 F20101109_AAATDH lomascolo_s_Page_152.QC.jpg
ed522ea455baf1619f6139032579ec21
8fe30a16f6eb7a9aa72b0e5b876cf393e6dc10bd
27877 F20101109_AAATCT lomascolo_s_Page_145.QC.jpg
8bdf32ae41bc7f74b24c342759aed388
2bc2ac9b26801056dc8339cc5274612599fd62fe
26853 F20101109_AAASYB lomascolo_s_Page_083.QC.jpg
007ebe554fba9f0fff19c909d1406774
dfd964c1ba65bbf2dd13a667a797ee3ec689f318
26166 F20101109_AAASXN lomascolo_s_Page_076.QC.jpg
14dec37e5f31de7c8e0387fd3bf059e6
fd5f9f56e55d2397443333b16a10506a5f776ef8
12784 F20101109_AAASWY lomascolo_s_Page_068.QC.jpg
8a21a494e446ca1391830e80dd017bdd
2ef71b77804f6a092415950081bcaa8ec16b5a15
98925 F20101109_AAARUL lomascolo_s_Page_145.jpg
4638296e1df4955b6def319a9e759299
82cd271bb691b0fe194601e3857038d1940637c1
51735 F20101109_AAARTW lomascolo_s_Page_130.jpg
bf839bef3d73496379047cc2fed2286d
7a2c8ffc97547348fd5eeed5dee9abd11590ab0b
1051984 F20101109_AAASAF lomascolo_s_Page_145.jp2
a7bd1deb8aa47dec880fb6be3b22f3d4
df2f9f33e7daf5b5091d41bec06fe9e5b3bee645
6911 F20101109_AAATDI lomascolo_s_Page_152thm.jpg
170fe0e5f4ce8e97d5469ff7443078bf
8b910310d14ea3caddded0d7730cb7cf6e0cd63e
7595 F20101109_AAATCU lomascolo_s_Page_145thm.jpg
300b0ebdd691371566d3249c0dd02615
ad19fdc674d80d056c7cdefdf455b0edb6053295
7264 F20101109_AAASYC lomascolo_s_Page_083thm.jpg
323452ebdd7a8f9f37208866559ea5e3
13317b8ae7cf3d103371c6ed9e984070c4f053ae
7174 F20101109_AAASXO lomascolo_s_Page_076thm.jpg
ef3d6bad0e358622f397c89f259b18f8
2c89e303118dc0e35513b9716b733da00938b22d
3818 F20101109_AAASWZ lomascolo_s_Page_068thm.jpg
57d0852eca3ddcc591b442babb62ca69
925946bac3efb8a984ae493aed9ce331a4ab69d0
1051899 F20101109_AAARVA lomascolo_s_Page_008.jp2
0cdf120eb6e5c9cf9a3924c289da304a
a934de5300e830e46bf8322fb75b2b8f0f776d58
88610 F20101109_AAARUM lomascolo_s_Page_146.jpg
051afa8f08e576b78e644ed0cd9c55d7
b615c8c77c30880a26515645e5a16b496b23e7c7
85096 F20101109_AAARTX lomascolo_s_Page_131.jpg
6b4bd43b950ce9a24fb588b9fba4d61b
0b06f5eb4375939f366df0be0d0a79e1f4886d8c
1051949 F20101109_AAASAG lomascolo_s_Page_146.jp2
c813fc6cc6a1088b535c2111e11fe563
6ef9e49cb494cdbcdc949d99e541569d021f37e3
177889 F20101109_AAATDJ UFE0022103_00001.mets
e046635b82a34e0307dbf5a060b65a95
b6bc3361b9442883c0694a7b326d37580238b49a
25151 F20101109_AAATCV lomascolo_s_Page_146.QC.jpg
7dd8031e1699345767f8230a31bfd462
eb2742a554cd5a253e992dc660a34b7ddcb7fe8e
28341 F20101109_AAASYD lomascolo_s_Page_084.QC.jpg
8c09065ef42091bb39a8ce3b293313d3
23b69e7fb20cb8504218f143f5cc7b414b7f3c76
26996 F20101109_AAASXP lomascolo_s_Page_077.QC.jpg
8d768ca313b6a14ddb38692d5f677903
4ba4bce67658e0109c66f64fe5bb96d0e1683a22
1051970 F20101109_AAARVB lomascolo_s_Page_009.jp2
1ccd2257073f00797b0b466776fcab5e
55d4097fd9479aed3e6d7d60c9faa75b7ca3c74d
95216 F20101109_AAARUN lomascolo_s_Page_147.jpg
baeb91afde46e55243c7e5392a0b0642
36bd6f3ecf5e9aa9f54aa1e783bd774dcec7442c
88606 F20101109_AAARTY lomascolo_s_Page_132.jpg
6ec48ac0e8ebe831ca278317ea34c850
2f70e5eb6412f3ece08a9336ab4aabf32e39d6ed
F20101109_AAASAH lomascolo_s_Page_147.jp2
fda1192c43b5e93bbb817033882e81d6
40ab6694d1505f992c0e880ac6e262eac2229139
7120 F20101109_AAATCW lomascolo_s_Page_146thm.jpg
6a872846a41c6406acaa2ad4c337fc58
4fd68e50afd7c50e48aee9354748db581e59ced6
7488 F20101109_AAASYE lomascolo_s_Page_084thm.jpg
5c0aacd0bb1c4a4f3bc83c10c6208058
4405f76ff2ba5a38a6f3c4064c508afcaddeed61
7448 F20101109_AAASXQ lomascolo_s_Page_077thm.jpg
225a18ab3062f73bc3f99882a70689a8
1ed2ee847f210b1c1520264c0aedefb8bb165e91
F20101109_AAARVC lomascolo_s_Page_011.jp2
56878cbd5245dc0dffdb27d57d4c292c
230b98e5669c21cfb7e3512da483425b5a230596
93774 F20101109_AAARUO lomascolo_s_Page_148.jpg
fff59c0e804823b9f3c366355d3205db
6d4ae69b85afa5c4dd76bbc9bb9deb6590cbf49d
89637 F20101109_AAARTZ lomascolo_s_Page_133.jpg
dfc6511c445a7d47f40f7e898216d33d
bb832b95e73dc2f0989f0fb82fdc28f981b22dee
1051980 F20101109_AAASAI lomascolo_s_Page_148.jp2
ba6b6298bd79e90532c063001a3225b5
8c9d49821fdd5db3608a09002864f97619f181c8
27109 F20101109_AAATCX lomascolo_s_Page_147.QC.jpg
4d308260e5073dd2b236fe7749f0aea3
8b90e8b8697eda351db33821b8aa57456bd5dd9c
25791 F20101109_AAASYF lomascolo_s_Page_085.QC.jpg
130c3f1aa5fd20b2089e1a975f50ae68
87fa58e592731991ceff5699f157b8527e9ee517
28106 F20101109_AAASXR lomascolo_s_Page_078.QC.jpg
bb67c2c42a5dd0d7eeec05f1130908c8
21c8ace06735c1e9c9235299f4de390841904dff
1051978 F20101109_AAARVD lomascolo_s_Page_012.jp2
276811df455e7bc513b9066bccc97896
5c363fd9fa7cb4963320d9a5d2d7a655139ff548
101091 F20101109_AAARUP lomascolo_s_Page_149.jpg
b52d3159bf7410af06a7e4dfcce53c05
341850d02154f2272762d89c88f6696e17af9011
F20101109_AAASAJ lomascolo_s_Page_149.jp2
7e7489d27268f1860dfe52148e8ab9a0
d3f47621a8b347a94f559fac3332a30f99cba4dc
F20101109_AAATCY lomascolo_s_Page_147thm.jpg
6d2acd3530f347c611e8ba27eba0e914
4f7c333b23bb6d4a940d9b5b5b735ee71e4892c0
7382 F20101109_AAASYG lomascolo_s_Page_085thm.jpg
73e9d81708298e317cfeae3580e2e00a
5bc05af9a3ce121a8f75a2f40c92f6fc3e0dc379
138361 F20101109_AAARVE lomascolo_s_Page_013.jp2
f8429d96ac1192faad94ec6f1fd5d035
68cd4c39ca68cf8aafd0bc40047690fa61a919ef
1051981 F20101109_AAASAK lomascolo_s_Page_150.jp2
4ce069e8045354e0df1f78231bab289f
a7bc7566af4209975b062c18faeb9672ceeb8492
26517 F20101109_AAATCZ lomascolo_s_Page_148.QC.jpg
aad9510263c7d09f0a282035c13f4b7a
2967d6151812d4bbacc2d8781eb120e409853cc9
29156 F20101109_AAASYH lomascolo_s_Page_086.QC.jpg
2f2a7795cce3d4f886391ba72026a700
5a2ef6c0820ba35d802f861eb54884642d1043e4
7575 F20101109_AAASXS lomascolo_s_Page_078thm.jpg
77d67b68201ffe88f55365e05b869606
a01eb6c3fe6dfcca23a070009c7ff70567229f2e
1051961 F20101109_AAARVF lomascolo_s_Page_014.jp2
807533806df3f971eaa7424f95a11412
6d1fb4f4d09e684f1c619ffea9d1cf7cb4a2a440
87608 F20101109_AAARUQ lomascolo_s_Page_150.jpg
07233361e9d55a4252e7ea20bab8b35b
b859e44ef79e2706f364ca8d64143317a8dddf13
309226 F20101109_AAASAL lomascolo_s_Page_151.jp2
7ed09315dbc8a172ee50bebd66a84ec7
b7cbdf2a52be47913dc39d80103e910e1645c8b6
7659 F20101109_AAASYI lomascolo_s_Page_086thm.jpg
cb178ab47580a856aca6de7469295b20
715f9e5e6517a4093b660161c6c4e2416d402e43
27012 F20101109_AAASXT lomascolo_s_Page_079.QC.jpg
1601f757f9d76458c1de72b59ed576db
614014c031ce77665ae70973283db73b2d87f3ae
574486 F20101109_AAARVG lomascolo_s_Page_015.jp2
13f5ec986c7979bf130eeded0443ea56
0c11b89f73bda11c362faf2f9db8bd27a8992a80
25966 F20101109_AAARUR lomascolo_s_Page_151.jpg
b8eaf868d6abeba663e0992a88aeea75
74c5d57f00fc4bd0c08f69094f895a6c885630fe
F20101109_AAASBA lomascolo_s_Page_014.tif
f6df6b5b0f3965bce584dccbf99fd2f2
cd2bb48ae3b53f474d993a3c0316b1db39a7ae73
1051982 F20101109_AAASAM lomascolo_s_Page_152.jp2
17db29dcae51a4dc7104c9e639c40b16
0328986b57c39ae1cb73046964bf8eb2b543399d
27965 F20101109_AAASYJ lomascolo_s_Page_087.QC.jpg
150234f41940598c8521c6b7a0c2dd87
bef5fcb163c86ead628765af4c441e8f4b5a0cf0
F20101109_AAASXU lomascolo_s_Page_079thm.jpg
08391529686202dd1bfb7bd0fec5da68
092fe2b8329806137501b95334b844a45e23a2f4
1051963 F20101109_AAARVH lomascolo_s_Page_016.jp2
86998e9b6a598edf7f6fb2e86fc29fa6
c6e3ac664e820606893653074fa1b74092513d9f
80137 F20101109_AAARUS lomascolo_s_Page_152.jpg
ce81ef45f1338613896143753c0824b8
b279d4d717ec75a6c0ad709ea641bd0b8d893125
F20101109_AAASBB lomascolo_s_Page_015.tif
018075e310199e5b4132208c6e414e59
01b6c37adc4b099d7477d6379d396fc1cfe8a24d
F20101109_AAASAN lomascolo_s_Page_001.tif
90e7e0e9668fed768bce31dfc98fd290
1fdc2ef502d20ae414a7f6a520f6085fc9150cb7
7542 F20101109_AAASYK lomascolo_s_Page_087thm.jpg
a0ad2991382bb04a3c971cee905fc57c
45d705140970a001eb4c4886110d2e5c653d4905
26469 F20101109_AAASXV lomascolo_s_Page_080.QC.jpg
260406eaf6889686eb692f8881344737
379b91def9c161a121e85fd36a3aa693d9b3cbbe
F20101109_AAARVI lomascolo_s_Page_017.jp2
09b5afa32d0f04054fac7ae10f7befa3
5dfffd2160927b063b334672744311961adca7d9
245015 F20101109_AAARUT lomascolo_s_Page_001.jp2
e874f247d31efb1d16f5fcd3da5cb11e
167d80a294881ffebfa242aa745bf5bc6effbb0c
F20101109_AAASBC lomascolo_s_Page_016.tif
249efdb9f488865272d3c125a24cee3c
fe15052a708e0d59431435941a7b66ebed2e2e88
F20101109_AAASAO lomascolo_s_Page_002.tif
8d484931a4e705ebf4b2da8782f0cdfa
e46d6e7fd50ccef0c8c0d397e24f4e15c8f534fc
27893 F20101109_AAASYL lomascolo_s_Page_088.QC.jpg
c8fef01068678b3ca62a31045253c64b
6a1a16bd4c1d6bf411802bb19b54e7380ab97bc4
7396 F20101109_AAASXW lomascolo_s_Page_080thm.jpg
ba6f4c3fdece53b68fbb635bc0009f76
3b717105f3f076089868d0e55cc0945d2df4de2d
1051935 F20101109_AAARVJ lomascolo_s_Page_018.jp2
0151cc90f86cf9e76f49c6dc12a80aed
dc95f0cf13adc773ab6679ea53f5c1264b07c61a
31320 F20101109_AAARUU lomascolo_s_Page_002.jp2
2c3f33d5bd27d4e8491ef40dba575744
1ee79b71442b52bf8d65ee25ab504c26d8c27d41
F20101109_AAASBD lomascolo_s_Page_017.tif
88c1c1470b2563be6891eda516fa6342
897eb63184e79f7bfd8fde1fe6e3d90ef8f6e078
F20101109_AAASAP lomascolo_s_Page_003.tif
7d955c82b669829762a92eb951321bdc
fcb441cdee112dd21b164e9dc91fce9efd7d1193
12073 F20101109_AAASZA lomascolo_s_Page_096.QC.jpg
2ad83b2cc19d573eefb3041c2c54e8a9
10bc38cc4d1c6da6d948cb10e79445facd3778a1
7202 F20101109_AAASYM lomascolo_s_Page_088thm.jpg
a90ea6fbc919b0df7d867675e2d91996
8fab776ab35b6c026a07936bb1c8ffb403b675e1
26510 F20101109_AAASXX lomascolo_s_Page_081.QC.jpg
9daa773f4398f7afea3b3f36f0dca49f
e929146cb19d67e6bf44aed1ce585a56cadddc35
1013605 F20101109_AAARVK lomascolo_s_Page_019.jp2
a3cf86322167cb4c89f13bcb5dbea533
2c7ebaeab0344d8ada46cff646e6fe23a07497e8
100733 F20101109_AAARUV lomascolo_s_Page_003.jp2
18d98e944f6b23e2d7e9814bc5b2266e
3afded7bbd116aee4682d7ff51e3f0c14d8cbb1c
F20101109_AAASBE lomascolo_s_Page_018.tif
eba22c50f9e68f321b409ffe2358ad67
9523f1ce7551e7336064f11d3fa9bfd657f580a9
F20101109_AAASAQ lomascolo_s_Page_004.tif
7a223b0fa87ba9c2bdf3b45f618e7c38
4e646e62a243be709d1d066c0ad5a39460a85b4e
3964 F20101109_AAASZB lomascolo_s_Page_096thm.jpg
4c8adc8cddd66dd935aade364acc3eb5
0dd51d2ab93f5d6bf457386aded9472ef5f6b201
25829 F20101109_AAASYN lomascolo_s_Page_089.QC.jpg
0b4f86cc34d8e21f1e063a4ffc879d4d
8623eb8a63d578ce16599d02e8eb810c5dacceb0
7527 F20101109_AAASXY lomascolo_s_Page_081thm.jpg
c82b5b259049fed60eb2e66cd287ae0d
9dd890f16be045be1f83464d729e6b41531d039a
1051974 F20101109_AAARVL lomascolo_s_Page_020.jp2
22cb2cc95b2eb5fe837b95b47f7a29c6
cf838918e2e7d02f6229849842b14a15b7fe27ca
1051968 F20101109_AAARUW lomascolo_s_Page_004.jp2
e762160f730028e3b2b5059ff2f22633
42657323b5405dd23d2aa78bab8fe92b3eafaaeb
F20101109_AAASBF lomascolo_s_Page_019.tif
77af2405910bb0729a1916568b96694e
a6f8f157a70528257bc0211e64c4664466de403f
F20101109_AAASAR lomascolo_s_Page_005.tif
f642482a82343d89e57c44a56cbe637b
ef7fdad8b0bec61756c3a7c729152415542659cb
5541 F20101109_AAASZC lomascolo_s_Page_097.QC.jpg
72f83cbed5d54b7f5fd4b3a086f29386
14e06ac258337793a76c13f310a1829d01c441b6
7050 F20101109_AAASYO lomascolo_s_Page_089thm.jpg
2970c2f54abf5545e792cd77363d35d7
810257f52454d17a8cdbf6d99761039b39e32f0f
27082 F20101109_AAASXZ lomascolo_s_Page_082.QC.jpg
8dbb6a2fcd03a75531de95707caf4f08
e0e2bd26856f079f2967b55d6b1f50c1588be14b
1051965 F20101109_AAARVM lomascolo_s_Page_021.jp2
993c7adb1e04eb70fa4f1d0fec4a07ba
335971a122feb55587372919471e3b785bee33c8
1051952 F20101109_AAARUX lomascolo_s_Page_005.jp2
205d76e07d3eafc3eaa8dbcc01858461
595db88b378f99ac0d416f552d1e55a29823ce6a
F20101109_AAASBG lomascolo_s_Page_020.tif
6060dbd0b079a3b70ec805b3657f3895
147eeddcfa9ec12e83c25535a1b6bd5c12e27945
1051962 F20101109_AAARWA lomascolo_s_Page_035.jp2
ef384e1041d9c7d027f33aa1a286d8b6
47c5046363a8669a317f674cfbdc587ef0512ab7
F20101109_AAASAS lomascolo_s_Page_006.tif
b714463c7c09b617bf4078f4b305809d
d3d4fa6a7e5feebdc468fded4e99665daeadc8b8
1931 F20101109_AAASZD lomascolo_s_Page_097thm.jpg
7543d0309706f3ced8a31636e6cb6cfa
3e1e715309881a61bb5d56871570ce339c51d0a8
7151 F20101109_AAASYP lomascolo_s_Page_090thm.jpg
c792c34291fb0dc65cece5b116b4fa15
1217d53229c913d8634106aed0d55d8382ed2ffc
1051933 F20101109_AAARVN lomascolo_s_Page_022.jp2
0ab697a631f5d246340daf39fae16f21
921365d102b827a108fc729c6f569655d1661921
1051960 F20101109_AAARUY lomascolo_s_Page_006.jp2
194cfff86efed00c90df99e24d5f7041
a848a461778c5398bed1b993ebfa23c37ced6697
F20101109_AAASBH lomascolo_s_Page_021.tif
38fb811d361e56545173eb0bd13b923d
d9d0823c33c16fc245ef0146fded1ffa34f87c13
1051953 F20101109_AAARWB lomascolo_s_Page_036.jp2
87a57a55d22838acd457ba27153c0a48
211729397c7fec6c578565b47b14f7d0e8c34ee0
F20101109_AAASAT lomascolo_s_Page_007.tif
3bef6474ab430863761bab3f6c351c41
616ebff02683d7c70e097eb09ac1422a9b101108
14472 F20101109_AAASZE lomascolo_s_Page_098.QC.jpg
5942c5d5e528e6366d9c2effdb51b897
f6ba32b60b3e422330ea4d151a6f01985d3b0346
28615 F20101109_AAASYQ lomascolo_s_Page_091.QC.jpg
427b079b2ef4ecfe6ec7fbcd76704a56
b60a7f20c87a4ab190728e967fc0d26f32969c82
1025748 F20101109_AAARVO lomascolo_s_Page_023.jp2
fd5e357fac2980e931b81a9669f6f9a9
4d2fcbfd05a4f2326508f369bd40740b068ba700
848672 F20101109_AAARUZ lomascolo_s_Page_007.jp2
3a1586c26040db664f4621cb67532e08
debc53acb8c23595091a18a4a59e7099d4965a30
F20101109_AAASBI lomascolo_s_Page_022.tif
55b860cf0562e25e1b274ac2fa207ec0
b5d32f93fbb2fb42504ea6c1020eaba310488772
1051937 F20101109_AAARWC lomascolo_s_Page_037.jp2
0985e106f2c1ab9aec0e0463f602e206
08ef4f600d472932a84c3850daca83035d379652
F20101109_AAASAU lomascolo_s_Page_008.tif
5f679fa94a3fe28ed7e3be75709c9b05
c5d10806bf60dc70e567dac0e5ec35293f728d0d
4317 F20101109_AAASZF lomascolo_s_Page_098thm.jpg
579326313cbdb027fd80c7ea54e35c91
d8f7f38269571c0dfd7ba98a1108f6b2e47135b0
7686 F20101109_AAASYR lomascolo_s_Page_091thm.jpg
a51613370c2c1592934a52d2e1bc78f7
bbf07cb9fec24e41ff8b7992a1b2c26ef97f1628
F20101109_AAARVP lomascolo_s_Page_024.jp2
ec7b0a643aaf97c4c8b55e4ce271d54e
e4d4c12739acae47f29f0b5c0469f4a0c707a5c1
F20101109_AAASBJ lomascolo_s_Page_023.tif
4bda61c73c2b08fa1e7f0523ba5e8340
b3009039a431516d81ca7140c2e26969819a930c
359576 F20101109_AAARWD lomascolo_s_Page_038.jp2
ebb224620aec06e170aa1d9333732a20
521f702564028dd7aa18c2113323917ed669ae0f
F20101109_AAASAV lomascolo_s_Page_009.tif
4d77aebe27d9f721810a0810d0a21769
d5485239a1f3bbb637535df51e37037635394575
4959 F20101109_AAASZG lomascolo_s_Page_099.QC.jpg
968af545b5348c023771b0bc6184b52e
12d0f85a515e4210ac3a1b92f6537fafda402e40
28272 F20101109_AAASYS lomascolo_s_Page_092.QC.jpg
5b115b6a54c63d973cb89f74d5b9c567
928f47957d85ce329f60c21b139ba75f950ed4a4
F20101109_AAASBK lomascolo_s_Page_024.tif
c2f3a0856ca8a019fce23baee05a45a4
57e6c5cc1182eb44daec17811e9152f9cc08a186
860239 F20101109_AAARWE lomascolo_s_Page_039.jp2
5ef5062b8cf3019b5b4dadb43400618a
55c03f2131638e8302b02d0a246167a3e3af8488
F20101109_AAASAW lomascolo_s_Page_010.tif
a89a5516a6d0a160d6e605fe4e6f6cb9
86642f616f7ff977c48f86271b6faadaf6dc826e
F20101109_AAARVQ lomascolo_s_Page_025.jp2
ebb91dd02265614f9c0639554859ff61
b267b5b64f87fb4cffbd117c82e5788794ee9ba2
1768 F20101109_AAASZH lomascolo_s_Page_099thm.jpg
3e658937ec29e8ef5fef685f7646e00f
8bde8b170b08ae8acd93b76b2d2e72eec9fe37cc
F20101109_AAASBL lomascolo_s_Page_025.tif
5e34b60deb8ecb2dee4e121aa5c15b40
770d549cfe913e99dd6527d5f832f7cc984d48e5
389441 F20101109_AAARWF lomascolo_s_Page_040.jp2
7ac4dfc1179115344593e2e2a8eba366
c932528b0379b64691fdff12f5d4bf256cc24778
F20101109_AAASAX lomascolo_s_Page_011.tif
24501379acb428f0715be3d9e4da5bdb
4092b5e0dfdf63a6cda5dbf86f58d3d6b681af5b
11308 F20101109_AAASZI lomascolo_s_Page_100.QC.jpg
8fbadb19eef5bf738cd6dab504ff7260
5f14266db969fe27a87ed42910e0c1481c9146b9
7450 F20101109_AAASYT lomascolo_s_Page_092thm.jpg
b3f50e1b306718a44c4ea0b2b58b3692
8c48aafa459cea0619570239261464f0a0eafb8a
F20101109_AAASCA lomascolo_s_Page_041.tif
80c9d5784011537487c9f5c053871c7f
4a9f7ef16da6dfbf8b097354c47bd1e1d71a96b7
F20101109_AAASBM lomascolo_s_Page_026.tif
b80eb7fbea289307c91224377ba85c0e
2a27b851360fe472d72bdbba77cc0b88ece91436
760560 F20101109_AAARWG lomascolo_s_Page_041.jp2
0196384e4d9f0919dbcad6d69646adc5
b7a049689af74f52231aac04b124185aaf6838fa
F20101109_AAASAY lomascolo_s_Page_012.tif
1da1037a66b1905cbb32219b0e5ab044
f8e5945f60713e97936347f35a78ab8c6878a999
1051955 F20101109_AAARVR lomascolo_s_Page_026.jp2
941d94a35291d7e693e8ac6d4ba73af6
0dca737dfb3beeaae2011ca0eac9129ffd74e4ca
3909 F20101109_AAASZJ lomascolo_s_Page_100thm.jpg
5e75868b00ceb0c0eb8e05f6a94bb55e
edfef5dde937b8d90bb5edbccb68081acc606062
26948 F20101109_AAASYU lomascolo_s_Page_093.QC.jpg
474f7c6063c36d919e937847fdc6d477
e51978f8775c9f2bdb7a6ea2188a4034edaa1651
F20101109_AAASCB lomascolo_s_Page_042.tif
9049eab76ee03ff3ee9907fef4b45cbf
8a7f66a94df6be666531cd365b65ef662fdac57b
F20101109_AAASBN lomascolo_s_Page_027.tif
b8db5b3b2eeadf533d8e188376445134
42bbc5ec57bfed94f99aa01781d95e0c8b3bf534
693166 F20101109_AAARWH lomascolo_s_Page_042.jp2
c17ccbb8bf05b83d1afa0cc4e284b91d
1594094071b805e49641bc476ddc4dc47464ace6
F20101109_AAASAZ lomascolo_s_Page_013.tif
e7fe2c4d8fca67ffa87e40e394fcb6d6
415baa8833ecb0314fb72c28b86680588720f36d
F20101109_AAARVS lomascolo_s_Page_027.jp2
1d3e79f09993a33b499bed51e5c9c51f
2f1679d5177298b8df6d300e21c7af0779025b82
5186 F20101109_AAASZK lomascolo_s_Page_101.QC.jpg
b9269563b26a7be1ff996fe42a0eb3e9
1d022c9b64d01881ab4bc22d7723a451a68e9837
7379 F20101109_AAASYV lomascolo_s_Page_093thm.jpg
fe734257ddfdc550781c35e4f78e56a1
5eeac01c004ce174a7f84b9bc5f82a6592084140
1054428 F20101109_AAASCC lomascolo_s_Page_043.tif
2335125e0d2dc64bbbe0e3a20ebf0ec5
5b7720c26e3b72d64e3d7b63cc1fd660eb00d433
F20101109_AAASBO lomascolo_s_Page_028.tif
f9c035cf17c661d25fe956a1fb00610c
fee7b9e4b90d82ff0fb9e761e7b99ed1ddec67f7
109508 F20101109_AAARWI lomascolo_s_Page_043.jp2
d8db949c8c03f787fe12d8488b9a3711
bcb1abb6f5ed641f3dce6bd86bc011af3797476e
F20101109_AAARVT lomascolo_s_Page_028.jp2
45bed905a9736b9882101c351fc13c5f
cce311ef66df79f13baf869af85c8428dd6638fa
2095 F20101109_AAASZL lomascolo_s_Page_101thm.jpg
5b16f55bcbf389d5bf279e002fad68d2
637bdf41e3e0abba0f17fdc119feb9fe0bd82523
15074 F20101109_AAASYW lomascolo_s_Page_094.QC.jpg
a8296401020d4e167463256889fa61ef
3b6b59623e27b2e783de3d423e2e0066a68ea275
F20101109_AAASCD lomascolo_s_Page_044.tif
35c775b6908485b84924884fe36a4308
198b9c8f0a2d1ded8a27388559ce06f896dc43be
F20101109_AAASBP lomascolo_s_Page_029.tif
cce7ce004926feaefab9de723f11f76d
06a20285db29a78cc31bd8bd56337876c9acb3e6
97272 F20101109_AAARWJ lomascolo_s_Page_044.jp2
163ec6621707244ddf69dad4d6572b68
f82beee52dbe194318f0e93c94195f5a2ac1fdec
F20101109_AAARVU lomascolo_s_Page_029.jp2
597d6b548ce0f876b774a3d8aa99502b
219be688419cb2413f6d60a263fd34b678313ae1
7802 F20101109_AAASZM lomascolo_s_Page_102.QC.jpg
d164d5e0c28abfce8e493fa8c20ad88e
5b443317bba43c65b352764ec4ef02c1cc0cd736
F20101109_AAASYX lomascolo_s_Page_094thm.jpg
bcf382b98f342b8613ee20a339eee0aa
b53a20fe8f1e76b6df9e8aa632bafca21de6343f
F20101109_AAASCE lomascolo_s_Page_045.tif
1878bbe7e6f408764d5461e288acc726
08f1699a2aee4712e05e1aeb9d32a560d0780e49
F20101109_AAASBQ lomascolo_s_Page_030.tif
59027cd69ef1f6909bd825271ff86798
f53f2b0cb979b575acf4a9cc75bf95dedf94fa10
77892 F20101109_AAARWK lomascolo_s_Page_045.jp2
cb92180e0fdc7bcbce50e6b8dbfe9a47
c2f64d00059803e931f2f6541ff1537387ebb70c
F20101109_AAARVV lomascolo_s_Page_030.jp2
bfdb6e430a0be4d407b87af49cbe82f3
1e3cac98fd75687b374576b274991806bce03fb5
2912 F20101109_AAASZN lomascolo_s_Page_102thm.jpg
cfaac8122c4e870c6e9f30074cddb503
28953e96403cea55296f77e337dee5bf6fd16869
15482 F20101109_AAASYY lomascolo_s_Page_095.QC.jpg
2798db7a3ba0a619508b7d2ec90fbcf1
566cc40d6b7c4e9177ab114930383dda354ae7fc
F20101109_AAASCF lomascolo_s_Page_046.tif
edc0fba8b6a53e7dfa6b797d984e8823
78b3ea33b29cf62deec2324eab817613fe911f68
F20101109_AAASBR lomascolo_s_Page_031.tif
d1fdde7969fcbc36606ce7c36bb460e7
53e9f3430bf76bbf80e3f83dca38bf4d126d02f9
68582 F20101109_AAARWL lomascolo_s_Page_046.jp2
88a41a00978531a6b79d2c5bb2324125
aeabe9e9b9d6a8449d9408a4fb8c95d25d20686f
1043810 F20101109_AAARVW lomascolo_s_Page_031.jp2
49890ce68a23902f29fa29ca9a4cd4fc
52f7f7de702afd04f0bd3ea356b673f20a6521e2
14494 F20101109_AAASZO lomascolo_s_Page_103.QC.jpg
add6c1513ec8986eb1425e335494fc62
46c89726777b26628abf2100d10e8d029b3323ef
4799 F20101109_AAASYZ lomascolo_s_Page_095thm.jpg
b800b651886e2c534ded68d6662d8c31
7300d09acfb137ec628425855a88e24e83297f5a
F20101109_AAASCG lomascolo_s_Page_047.tif
1ae7c013c29320062336a8a66df8e458
386ae90fc5575a3c2708384f697ce446bcf50630
F20101109_AAARXA lomascolo_s_Page_061.jp2
a73b0b9930c02cc92e07f94b9d9363e7
8d8950071b56470d37281343a29327679fc793b5
F20101109_AAASBS lomascolo_s_Page_032.tif
4ae78a4573405bdd74f873150f5658ab
72eca40d817623092a4dd67fcfd78b426f5bebbf
54059 F20101109_AAARWM lomascolo_s_Page_047.jp2
7240a0da04d7ab45ee63a4047323ae9d
9595043d5861e3ff77c851f89b3f4c817d497d93
F20101109_AAARVX lomascolo_s_Page_032.jp2
5110f7f5cad3b59c18286b494367328c
c93acf71bf1e4f1709bd0255868b55439498bdf2
4307 F20101109_AAASZP lomascolo_s_Page_103thm.jpg
1ec1c03f82f29b18cc46bd03aa882b09
c6a0800f23a451e06b6265637391db5e2ae61056
F20101109_AAASCH lomascolo_s_Page_048.tif
f450d03a5c591be7d02452627e1a473d
b35c8b945faf32dfbf734e09d106b7687490f5ae
1051957 F20101109_AAARXB lomascolo_s_Page_062.jp2
39a293568fa5c85c9be6a75ccf2097b4
9eb27364a7d70c3791ec4a861d270023104c70ff
F20101109_AAASBT lomascolo_s_Page_033.tif
824ed92a6d39c8924af6edaf82eb1ea9
0732cf0ddfba2a627fd8c568120e4bcbea7f0cfa
62274 F20101109_AAARWN lomascolo_s_Page_048.jp2
636be51c4d301352ed2e593c1ff5d825
ae4002adf9f9ce5300127d660c93f03c16602e12
1051972 F20101109_AAARVY lomascolo_s_Page_033.jp2
3984c859203b5cbb14f638d4e510e899
07a6d39931b0d84d3fd4f2b22dcc9ee1eceada88
6854 F20101109_AAASZQ lomascolo_s_Page_104.QC.jpg
dab16b6d5afb8c33c5c9f9931397c1b8
223834b34e95abeecf12306b563999e2fcf2a1fb
F20101109_AAASCI lomascolo_s_Page_049.tif
b5bdc4d2a8e91151bd6c626a251e5690
1459d9082896896727bbdf7a08de57ef493137f4
F20101109_AAARXC lomascolo_s_Page_063.jp2
1d1eb563539ec1fb2594a38764160117
89c8ed78c56c4c0ceb5a854322a0df0c19b19637
F20101109_AAASBU lomascolo_s_Page_035.tif
4329600454de1ccaa9f12df8df514c54
97894322ec85f5082aa4da9440ad0d24ff2066fa
68121 F20101109_AAARWO lomascolo_s_Page_049.jp2
bdb04be8519cb9c6e6e612d2ad591262
79f6f233bc346e1f86c6ba7287fedc75008116c3
1051944 F20101109_AAARVZ lomascolo_s_Page_034.jp2
e6e1d80aa5243e2d9618f882a964b16c
72b96d815f315133671f3cf371c6642507b73618
2437 F20101109_AAASZR lomascolo_s_Page_104thm.jpg
6ef9ba7d6c09952f61ed024c41cb1d17
266c580f6477470cc37673de07a41b24e0a773fa
F20101109_AAASCJ lomascolo_s_Page_050.tif
7d260e01952cc49290877a35081a78e3
f2547251ac0b5f6d3609e318244b71349be9970d
1051975 F20101109_AAARXD lomascolo_s_Page_064.jp2
9b96b7cd0387456867a3c0d4d5cd18cf
2948d39d928c0b7b52aae1425a5b82a155f40f84
F20101109_AAASBV lomascolo_s_Page_036.tif
5c104ccfc48b16bfd4a5b226c040660b
694ec8d94d92f8a85e4df991292465171facfce2
49214 F20101109_AAARWP lomascolo_s_Page_050.jp2
fc471c9ec0a7785341398901209a42a9
a8a61fae671e52b518d99216e216c3d7c8591b2f
16513 F20101109_AAASZS lomascolo_s_Page_105.QC.jpg
886b3974f72d4f1129e5f54ea4791f2e
359ce629b5d1eefd79240b63b9e011395b70d684
F20101109_AAASCK lomascolo_s_Page_051.tif
a9caf16d8bad6f110988124ef5ba722d
f03271a7abce68e59dab8a0a9ad8e2c298e03174
1051954 F20101109_AAARXE lomascolo_s_Page_065.jp2
7f8dc3cc5d134f34f8592f9ffc909e21
a2bd7ab47788a41f929ab485336120aa3c52f552
F20101109_AAASBW lomascolo_s_Page_037.tif
3fbecec3aa689395b484aa4966589e18
014aea3d387252a56b7135749c87c3b823b33257
1051929 F20101109_AAARWQ lomascolo_s_Page_051.jp2
39e290ea86070dd2a6743b1380b70554
455e3d5d6fec4ad7c30ac1cbf7a352dbe46f19df
4931 F20101109_AAASZT lomascolo_s_Page_105thm.jpg
7f04e281506a41a45cba533ffbcb16e9
48a5737409100e0f8c5d1e48c5ae24e79e5bb968
F20101109_AAASCL lomascolo_s_Page_052.tif
fd262199e899f6fb8cad10ccbdb90b1e
63303231b9587c9e0e9e997aedac66ec4721440a
592582 F20101109_AAARXF lomascolo_s_Page_066.jp2
a4f2722dd99fbe8e5dd1b1ca2bad3abe
8250c0578e028171d030b9de9ca11c433311f7ac
F20101109_AAASBX lomascolo_s_Page_038.tif
1430f369fb3e104ab38c7ecc7b093796
7cbe050ba893acc4d5278b7d6ef335e2d5988ec6
F20101109_AAARWR lomascolo_s_Page_052.jp2
c34c252e71fb1f8d0a8d63a4f7209a9f
e2f40a24c9444caf46bf0c8f326df906b240fdde
F20101109_AAASCM lomascolo_s_Page_053.tif
20ba083453dcb30d394a76f27bff3639
1bf129c9f5aa3c6cd13c39c458e4eaebe443808f
458857 F20101109_AAARXG lomascolo_s_Page_067.jp2
bd7a3db6785bf357617e770de61dfa20
a9f3e27edf386e2db49632bfa833fe31e2fed157
F20101109_AAASBY lomascolo_s_Page_039.tif
9cf5ac94ea0ce1d16aaf1dec66994264
18bb10232d17530dcb48a9339659dd40cc1c678a
F20101109_AAASDA lomascolo_s_Page_068.tif
bd000585d06d8764e3881fb21c8afb53
b5ea69201e00e6e87f8e8263560dc3f8a7ccb3cb
19885 F20101109_AAASZU lomascolo_s_Page_106.QC.jpg
942b68bcb4f60579d7ec08a80ce900f6
470bf22588cc1d9389dde4c99ceadbf9c42bb82f
477272 F20101109_AAARXH lomascolo_s_Page_068.jp2
34a4de41fae5efa6f7f8d09aaca88229
426d786d56400dee9946f64020a8dd6cd050ddb8
F20101109_AAASBZ lomascolo_s_Page_040.tif
8263a593347e1a6ccc3eb96f10295381
43cc78314eff3fcb0fefe1aaaf61047328bc839a
1050723 F20101109_AAARWS lomascolo_s_Page_053.jp2
d529bb896f62c404a27cba8d5ef54bbc
c31161affb0817cdaca5fcb6b850fd29e448806f
F20101109_AAASDB lomascolo_s_Page_069.tif
2c3bb3d1fecfe0d2b30a4e570da97aad
6dbd017d02629b93dd94326c0568eb7f7efd8df7
F20101109_AAASCN lomascolo_s_Page_054.tif
fabeeddd6fb580208620c98afcf5d19a
adf01867afe9abba73feeee20011f37a2168f594
5625 F20101109_AAASZV lomascolo_s_Page_106thm.jpg
64fbd21c9527d1287edbf8b642540458
43a69ffa7f5282b3c87465bd8955ce0486d1f501
194924 F20101109_AAARXI lomascolo_s_Page_069.jp2
5c775623261af67aba8ace8a3f6104e3
365b51681430be121918041ae64d3e8632eccb2c
F20101109_AAARWT lomascolo_s_Page_054.jp2
6ade8d98dd2183ee36c38efcd0299d3c
03f9737dcf107437a5f0b84f93dd21f58b2ddd82
F20101109_AAASDC lomascolo_s_Page_070.tif
d3e4764d14e84414e6c4d524e97b2187
7e546b6fec7d278898c8c72f26affca2ede5c2b2
F20101109_AAASCO lomascolo_s_Page_055.tif
95d86a151c9257702598819642ab4d42
0dd12f0811fc123ea1ac2e140ec2fd88e17394f6
27279 F20101109_AAASZW lomascolo_s_Page_107.QC.jpg
95bc2340394901b99f4d284de368eae6
503c103ac2dd913ae01a6dd79594ff741b0440d8
1051964 F20101109_AAARXJ lomascolo_s_Page_070.jp2
6a4b9ec84dc18f902c8212440abe7654
2ed2d62656db337962e2f523e37e8dd0bada6326
F20101109_AAARWU lomascolo_s_Page_055.jp2
c02a3f215313eba8e4aa8afa7f9c7de2
77fbf030f928c2211d3782b5b2a768defc112fd8
F20101109_AAASDD lomascolo_s_Page_071.tif
5d6e83a3a1f9cb3c7e0f7350f1f3445c
6ed9f91d3cc9249db1bbb1457ce7ebb518a73e68
F20101109_AAASCP lomascolo_s_Page_056.tif
4f28f68e2e0645c89a2a01ca44689f37
8ecbd272c2466dd5efb20b7f87390c820266ef37
7248 F20101109_AAASZX lomascolo_s_Page_107thm.jpg
22d3bdd086c079f492da651ef4ad150b
07a1fcad86db56f9f369cf7fb394ab0cbdf64dbf
F20101109_AAARXK lomascolo_s_Page_071.jp2
533fbb10681d66ef969b98e2badabc21
bf9cffe2e11f70aef537988d125f0feb53128e9b
F20101109_AAARWV lomascolo_s_Page_056.jp2
dee14aff7f2d9996956abe7f9524e950
cd22dba74154942559514a03ad77315a5a88c33c
F20101109_AAASDE lomascolo_s_Page_072.tif
4805419280d33b1e48c43e7c9145e5fb
b72b3a57767b9ff7bec5f614ca1eef39f7fe55a7
F20101109_AAASCQ lomascolo_s_Page_058.tif
73ea243a896d8d5bdbfa38e2648ff103
dc01ec7e9845a05cce377a37f30809b50f652449
27813 F20101109_AAASZY lomascolo_s_Page_108.QC.jpg
4729b5617174578882190d2e4a2be925
5d3434e7a3609b5b965c6d09d7939683acb8bfef
1051983 F20101109_AAARXL lomascolo_s_Page_072.jp2
87c2ecbc95aaeb4ee7e9918be236dd4f
be6dc4348e8da99fb1aba7ab3b1450012748af49
1051973 F20101109_AAARWW lomascolo_s_Page_057.jp2
5663b8b62711adc4bb116b4cfda350de
7225aa2264e55e0f51fcbe819fb7f59225036daf
F20101109_AAASDF lomascolo_s_Page_073.tif
0a02ea760c13c34c522cee6cc189ebaa
b2456d53267477b55c65fafac656147040edd162
F20101109_AAASCR lomascolo_s_Page_059.tif
5e8489058a037369d5397a152bbdf128
a8b126115cb5dd72680624a2eca8f9a4c44b936d
7167 F20101109_AAASZZ lomascolo_s_Page_108thm.jpg
e65cffd063ae77857180582e818713a9
c84d0a63d45618d56c1ed61a81ff2f0b63c831bd
1051986 F20101109_AAARXM lomascolo_s_Page_073.jp2
61001f94b9f87151023305455ca428e7
6620a7a92282dcd6821e6f5d0cd421a30bb26db3
1051918 F20101109_AAARWX lomascolo_s_Page_058.jp2
ead1a8150d879023901fca122034183e
102a935830b3364668e68971befb9a18d026e1dd
F20101109_AAASDG lomascolo_s_Page_075.tif
17e9de51452373fdd7c91807ee23d9a2
a6551ceff4eca10c580b2747a788f0dd4d5bfd05
F20101109_AAARYA lomascolo_s_Page_087.jp2
1efce03d78304604904e90be901a5f90
76bcd737837015f0c6b97321a70cedce710f9f19
F20101109_AAASCS lomascolo_s_Page_060.tif
5fbbc6cf06ee373a9971f527637fc151
9f298824ab6463da387a80d6892d61a7c78dfeb6
F20101109_AAARXN lomascolo_s_Page_074.jp2
f25edbe5f2ba2b71dc01077d25cf99da
7ee9da3825352920708b079d2bdfe86516aa9e4e
F20101109_AAARWY lomascolo_s_Page_059.jp2
87942e6d7d32cabaafda26b57b8fe56d
33facf0e63f8c2248e749d1e1ff27a1a860b4cce
F20101109_AAASDH lomascolo_s_Page_076.tif
e6c5322e6a551cf84fb19f3a5d893281
b2af5abb6e94a531aa5f314d3751756abfc69d2d
1051934 F20101109_AAARYB lomascolo_s_Page_088.jp2
3c503ac23e001cb783d5332e19c66120
2c3f5906e9199538802a1ec37ddaab4080c2c2d0
F20101109_AAASCT lomascolo_s_Page_061.tif
f114fdb37e8fb959306e080ea95f902b
cb9137694eef1b8fae02299e5d72a9d779f44e9c
1051958 F20101109_AAARXO lomascolo_s_Page_075.jp2
b03597dfe4b1c2aecc4686fc3cbe8aa2
24af59b3872c490719ce03fc1015ce4a1559ab2e
1051977 F20101109_AAARWZ lomascolo_s_Page_060.jp2
a9b61fc73e179c46adaf247c2949ea73
ccad448c4073cce20df55ccb5da1b81bcd779413
F20101109_AAASDI lomascolo_s_Page_077.tif
ccb767a04edfebc4c97738b720ca3832
059d36f0d2ab3d2e35d938ef9a1af8de295677f9
F20101109_AAARYC lomascolo_s_Page_089.jp2
fccb15b9f7b976fb6d3f47e11b22a93d
3db66182a63085c5b5364a2299a7cbfc7a68a8dc
F20101109_AAASCU lomascolo_s_Page_062.tif
3463d533b533e2e3b39bbacabd7de872
c532d812d66b2afbd99d592073f21caa441f311d
F20101109_AAARXP lomascolo_s_Page_076.jp2
69b42f7925cc8108dc9e9f4985105458
d48bd177675f2d638be39fd394160715f6070f89
F20101109_AAASDJ lomascolo_s_Page_078.tif
813abdef2c91e6fc3ca741f67dd45555
49be4b201fc3a2c13e336af91fedafe6cb897e33
1051941 F20101109_AAARYD lomascolo_s_Page_090.jp2
30de92b220e02c4a37df370dfda45736
5432c85d52186039779780b090655853d76987d0
F20101109_AAASCV lomascolo_s_Page_063.tif
68b028d6fb111549bbd7b7db6fdf7d66
1378e7b9fc47806bd962691db82e0f752b7771a3
F20101109_AAARXQ lomascolo_s_Page_077.jp2
ae9f22dd457b9d3e74f4e5292ba0afc9
6a977f843956dff9c97d237126dcfbd42a2d78cb
F20101109_AAASDK lomascolo_s_Page_079.tif
b2e31462487764bc40d580df6f5b7fd8
232123bdd173e445bd31df5c26f148142b5e0285
1051942 F20101109_AAARYE lomascolo_s_Page_091.jp2
bb7d439a5858d8a94f2b3dc3e93bb974
40942d9ac8f8006fb254a110f948f80e2993d3c8
F20101109_AAASCW lomascolo_s_Page_064.tif
b47d5d37594c73d19999266c61494bf3
a9c6f03d072f958a8732803d68c3b0a31456d5f7
F20101109_AAARXR lomascolo_s_Page_078.jp2
a8bee24442833675f14a2720d7052185
5f6264a43fdfb5059f97452df2e9acf91e3ae9fa
F20101109_AAASDL lomascolo_s_Page_080.tif
9ef25ffd8cb6d1b63092247b26e56af0
05a15b7ba3da93ae9d6272ab9196c6954cf6b8f3
F20101109_AAARYF lomascolo_s_Page_092.jp2
4ab3dddf687358d66f0f6438162cdd83
b0ee564d82c65c5b554605aae1bdb505fbac5f99
F20101109_AAASCX lomascolo_s_Page_065.tif
19bd3efe952ac322c562161b752fe6ac
480c5d9d03ce6eee49d3f5d8b96c77941d158eff
1051976 F20101109_AAARXS lomascolo_s_Page_079.jp2
12bf94618a3609910ed5eb8696b55508
72f0075fded5dba58556d284f2a6cc94a55d0418
F20101109_AAASEA lomascolo_s_Page_095.tif
504b30a88431f3419d13b42a1e863124
83fa60365921955f630336e0716afc8bf6daff74
F20101109_AAASDM lomascolo_s_Page_081.tif
2ecc104f4d57b58b2bd8d1d9d35fd117
3f2d42b6494a85a76cbe2abf766d1a204c9946c9
1051938 F20101109_AAARYG lomascolo_s_Page_093.jp2
9f040532069bd5cc0c689b30e2418664
8921afe81d4f1093de521ffa4d542b016af72b1e
F20101109_AAASCY lomascolo_s_Page_066.tif
c3497a9ccfb76b086d007ea93159d61a
1ea3c5c11ab2b805ac46cbb9bd7ae17a8be76407
F20101109_AAASEB lomascolo_s_Page_096.tif
28daae165f3724c4d06ae4a087b8320d
4b2f2ea19140d46d58c88d95058c30427e878bb7
F20101109_AAASDN lomascolo_s_Page_082.tif
9e94207fad6a1c6a9b381a49c3e34aa9
b9ba46dbecd7cbad8144b3f3656cbd472f8a4b1c
604623 F20101109_AAARYH lomascolo_s_Page_094.jp2
51cb30c32500f85b9c9f0e989f949d07
9cbcc84a1f3a8a9696dca550ecf617cbffb3fcf8
25265604 F20101109_AAASCZ lomascolo_s_Page_067.tif
e5b466062588b0783cd8fa46805e3dcc
0559cbf10d07231b683b3eb6e5885b0d17ddc105
F20101109_AAARXT lomascolo_s_Page_080.jp2
a3dc78a8fc60ee32a93c7624ef0066a1
0214488d6f48608e3f2e372d1416f97e71f49b14
F20101109_AAASEC lomascolo_s_Page_097.tif
0518c102f0331f258c74d2adfa8f71cd
67ed1e6ae856d8725f5328ba15097896f10cd338
F20101109_AAASDO lomascolo_s_Page_083.tif
e8fd250cac7d2ed2606b1d884093afec
762643c2a78d11d9fcb7590f480af65446fffe1f
696936 F20101109_AAARYI lomascolo_s_Page_095.jp2
60d669544f7c25b5e42846f69a6a54e3
d5a83f4543fd1e2278945cb338811ee96cc6c4c9
F20101109_AAARXU lomascolo_s_Page_081.jp2
c4f57f81d2ec03f9b2b99cd09d0b05e4
a957c3d6aba9a50b5d772cf68b9945650240d94e
F20101109_AAASED lomascolo_s_Page_098.tif
f380884da149888e1826d05139f2b92a
24900e02c8735682004ecfb7c593e098c275c244
F20101109_AAASDP lomascolo_s_Page_084.tif
a8d64fc4d578aaa49160a24183ac11b5
21b30f1153644ea86aca451ba858c17d36b6038f
422096 F20101109_AAARYJ lomascolo_s_Page_096.jp2
ec393463b12fc95a7ef614ef99552b03
ae42ba9e1ead62b36389bb53e7c5118bb085c8a7
F20101109_AAARXV lomascolo_s_Page_082.jp2
0cb407f753fcb086042182bbc5844079
2eb74852968f69de20e2edd35ed3c1fd75c6c725
F20101109_AAASEE lomascolo_s_Page_099.tif
8e79af1315d7b6b50beb8b77d0d319df
be5698aa39eb17563bc78d79629a7031536c2915
F20101109_AAASDQ lomascolo_s_Page_085.tif
201bef829b65d2faa15643fc34da7b7b
fe622da13a2df82075f3d9dd0644d6342b5a6b5f
158812 F20101109_AAARYK lomascolo_s_Page_097.jp2
1f2a2f2c34a9c1ed1cc53357b112c24e
5e17fa8acf6b5c51e10497de8ded724250cd62f3
F20101109_AAARXW lomascolo_s_Page_083.jp2
570e9dfd7c7f76ac04ab17e81ce28949
53de53e867d15ab5a2d197efc2f8470c134dc15f
F20101109_AAASEF lomascolo_s_Page_100.tif
51d953776d913098d0956296ebcff104
4cd6b1443face20d6610052886f36b64e3012830
F20101109_AAASDR lomascolo_s_Page_086.tif
11b785bb078703d5de308402293eff16
e6fc589272a6976e022bca13982cf11b4891e94d
564188 F20101109_AAARYL lomascolo_s_Page_098.jp2
d7e0c2a2fa5aed1772c9af76caf7af17
1ded3b4432714a628322bb960508b66db51b8a7b
1051909 F20101109_AAARXX lomascolo_s_Page_084.jp2
e1f4dcb763ec581f8f5217b481a53ae7
21195d1284993736e30522f19ef7012aa71c1639
F20101109_AAASEG lomascolo_s_Page_101.tif
a1a24e93a15ee39d8f338885ab2cff58
aef5eb1eb59ea773ec02c943a73b3bfb36232601
1051925 F20101109_AAARZA lomascolo_s_Page_113.jp2
b8c0d9628fbfd9cc8ebbb0b20a50c16e
ffd935e27f72b44ea227d509e65f09913a1dba30
F20101109_AAASDS lomascolo_s_Page_087.tif
b2f1be82717ee447f21dcccca832a4a8
31d901d7d2f0aa43caf624ba34a64eef03efdbe3
127280 F20101109_AAARYM lomascolo_s_Page_099.jp2
aaf4fbbceeee72a922aa354198c2c950
ebf61d7d2a8a25923eed21cbdb32f1168f1fb6aa
F20101109_AAARXY lomascolo_s_Page_085.jp2
725c1a18a7f72c8d9feb32dd0f6cab86
b45ad5a21df24518cc3ab54627e47fade6a0d3ab
F20101109_AAASEH lomascolo_s_Page_102.tif
384419b61d7241f9590c31c1d1b6a958
b696b0ba903599ee6a4ad5c7fa352c54fa89d5cb
F20101109_AAARZB lomascolo_s_Page_114.jp2
fd3bcdb1936dfca476d297dba08f3194
1baa624adf0976072efbe54a55565408a8ab8cd7
F20101109_AAASDT lomascolo_s_Page_088.tif
f98fe2c39ed58ad30945e448f36bb381
ad7727eeca4c675c71731137831b814403c9efb2
414272 F20101109_AAARYN lomascolo_s_Page_100.jp2
9c16fb838084539189ca35d0fd043dda
cc34e63b066413b88eaeb33fec489f87bbdd38d0
F20101109_AAARXZ lomascolo_s_Page_086.jp2
c6b06e1d61f9de8c236411c64744e72e
d4388dc7001e3a754d9a3308a0232523d4ab3bbc
F20101109_AAASEI lomascolo_s_Page_103.tif
c22e44438c0a387e03446a75e763527f
e06ae1b050b508e326a8dcfaaff281e65913f445
F20101109_AAARZC lomascolo_s_Page_115.jp2
4ef4d7381e6718566a455ee2bbfc5dde
b7cb78b1521eb421d5b6f7aa99a8f1578d02fcd8
F20101109_AAASDU lomascolo_s_Page_089.tif
967c8b006fef1c28477933a9a4079bd2
a5bf31e89f861c91cc8356959fe8fb71896b0b54
210964 F20101109_AAARYO lomascolo_s_Page_101.jp2
8b5f0c8130fb29846ac768e9729ec128
bc29640bead4ac6b8bb8c56713be92e4e4e8a77a
F20101109_AAASEJ lomascolo_s_Page_104.tif
fb7db70f451991d83459b4b7132d60ce
b6e8185c8f7a20db9ed98aaea3f77eaf67948b0c
1051859 F20101109_AAARZD lomascolo_s_Page_116.jp2
428f322804cb45031fea76917e05edb4
d38e91e0923469ad935d930df264667c0f27e6d8
F20101109_AAASDV lomascolo_s_Page_090.tif
a8c33e29a58696dd88fbc85471d4cd87
e8e500ed1ed1d83c80a3f7ae0710a37a7d6ec214
494900 F20101109_AAARYP lomascolo_s_Page_102.jp2
4a1569ab4599640ee1edd6804491ac45
2c2e18c23e08e3556c4d1bc9ea7e9c86587951a9
F20101109_AAASEK lomascolo_s_Page_105.tif
9df4732cbce919720fc71016b3e3940a
e63c11a4a3ff045965dbb98eea8c33710397e7cc
F20101109_AAARZE lomascolo_s_Page_117.jp2
6c9cfc0a76974f4281de1bb83f7102ca
3e6cedbcb34fb54eaf109e40d15d0cd4dfff1e14
F20101109_AAASDW lomascolo_s_Page_091.tif
8ce1356d2801fdffb21287e9c4b38066
b0b378f95fad554eed158270e9b02b02d5839359
660719 F20101109_AAARYQ lomascolo_s_Page_103.jp2
ba4a67487e4a728d6e9e8b2af9506469
08c1e7ceee4554f188ca596c246ffe91aede17d8
F20101109_AAASEL lomascolo_s_Page_106.tif
b19f45472f33aed9c695fcf701e6526c
4eed581fb70a73d8e1354416320a646cb2dfca84
F20101109_AAARZF lomascolo_s_Page_118.jp2
1bf0d1a3848df491394e0006fbee91f1
085c18280c0b9a68a6289a4eeeb34c4bc8c725f1
F20101109_AAASDX lomascolo_s_Page_092.tif
56949cee9f62b1c84241532e74db9139
98bbd6337eb0782aa927b801592afbd292b3da0c
308266 F20101109_AAARYR lomascolo_s_Page_104.jp2
1b9601ee7cae73b67cdfd3cab9bf3afc
c4747a40e5259c3347e8c848e91dbb13442f6f37
F20101109_AAASEM lomascolo_s_Page_107.tif
9947ac883e059a627a6b62d62c235e4d
d00e0fa4f1b72c5a8a544d441b8868a22caa5d5b
1051967 F20101109_AAARZG lomascolo_s_Page_120.jp2
3b782a15daa515539a0bbc815944d4fb
ddd9527cdd3425d87f3db2b0a88dd8825024931a
F20101109_AAASDY lomascolo_s_Page_093.tif
1bce34623672859b9c9abf5622a2e28d
e72df9659be1d772b0eaef622d97bd4308f91185
690662 F20101109_AAARYS lomascolo_s_Page_105.jp2
32a70a212c09e138c928723dca4c768b
5be473930215246de50a30b818676b6d4216e10d
F20101109_AAASFA lomascolo_s_Page_121.tif
246011a0f80f890f4883b20ee5d25142
7ec05f9814d117b0edd4e43493f07657bdd4d58b
F20101109_AAASEN lomascolo_s_Page_108.tif
9608e0d7cce9e797be75cbbc235c9d31
e92a3df07d0d1a89c230507abc15505e3f416870
F20101109_AAARZH lomascolo_s_Page_121.jp2
60ebebf26ee1f260acd5028f5d6d3781
98a52b8db10f6dabaad9a0824b2d4900c37cb826
F20101109_AAASDZ lomascolo_s_Page_094.tif
7f1564c53bf2008c58963de447f49b8f
bde378a55923f2e8d6bc97f43570f0c2822bf47f
944926 F20101109_AAARYT lomascolo_s_Page_106.jp2
cb63a9b5f51507ab75ed092c9d29e362
e7f5b259150e5a20397098e46aee9f5bcd6d6905
F20101109_AAASFB lomascolo_s_Page_122.tif
a722a43fc353ac0a737da51650e03df1
82245d6572badf9599b977bf0530cd9841e21b89
F20101109_AAASEO lomascolo_s_Page_109.tif
d8343c4cef70ba5df23e64de4074201c
79b3ccfbd8f21cfb810d751f7d03355b8915d978
1051946 F20101109_AAARZI lomascolo_s_Page_122.jp2
163fd2ec648e5731c0dac719b4978f16
5342e9d4c0b6b42e69d3284c85956af79e329e22
F20101109_AAASFC lomascolo_s_Page_123.tif
fbd9594a482dbc92d805bf81bea22b2b
b6516560f6e2a158759255ae96cffc12daa3b790
F20101109_AAASEP lomascolo_s_Page_110.tif
01eea8500bd5fa2e925f083d08fa5428
b5edaebed94648bd97b213c38aeb402c97bb0410
417415 F20101109_AAARZJ lomascolo_s_Page_123.jp2
642b80a756c19cf72df6be6affd0d269
3a17703cb1425bb1ff24703e10765570ee397f36
F20101109_AAARYU lomascolo_s_Page_107.jp2
a551ede9f2ead92e6a8cdf26acd106c4
f901677e60e13abebc0011b9883c4168aedfe6ed
F20101109_AAASFD lomascolo_s_Page_124.tif
4797dcf43ba99d1613d70d39a4a466c5
9f8009c0058c912be3e4b66b5aca66eafba447c4
F20101109_AAASEQ lomascolo_s_Page_111.tif
2eb4014523028ad59bcb5f3f349cc4c0
849b0bf27008460587d0fe771f64785594bc8c9a
148618 F20101109_AAARZK lomascolo_s_Page_124.jp2
ed4ac137b711babf2d0098e2807300ea
64edf127f95caa09e3a0df8961cea28d6c83feb0
1051911 F20101109_AAARYV lomascolo_s_Page_108.jp2
7491d214b16318d95bf462ad9f314fce
0ffb1fe0f60135a4ae57a11bdf8ebec7b0788fce
F20101109_AAASFE lomascolo_s_Page_125.tif
5ea209d573d5404d4d23a78fbf20d55b
5d4f89d48386a17f2829e82c8e10fc023a923a1d
F20101109_AAASER lomascolo_s_Page_112.tif
8c3a71e72200f7613eedb3f766788917
9e4f314a7d89d02d32c0c1188ef0b32810efc3e4
428347 F20101109_AAARZL lomascolo_s_Page_125.jp2
76c69677b0e3e633d421169188875388
01365814a1dd76bb45f7c60cc8144dabd6edc089
F20101109_AAARYW lomascolo_s_Page_109.jp2
edd4eea65db795ba86a87be96ee6d122
8045575220a7743d018683286c4a5925e49d9185
F20101109_AAASFF lomascolo_s_Page_126.tif
3d24ee147e3510cdc4f091c11dc2b227
9d3ee1f85ae3936ad2ae942b1f3618e21ac01a09
F20101109_AAASES lomascolo_s_Page_113.tif
2e906680281b1a0d5750576607418b15
0075daf4b93a65f58af99a374aac516ae7c437c6
217713 F20101109_AAARZM lomascolo_s_Page_126.jp2
060b8f3e7066292c0a84325489cd0ded
794f54ea476e425256a88af9a77a7ffd2adda832
F20101109_AAARYX lomascolo_s_Page_110.jp2
7c0afb58bf90528b9d442c762e0febb8
c2250b758623994f48070aa1b276b23d86c90807
F20101109_AAASFG lomascolo_s_Page_127.tif
741639e16cfc1726f1c0684dd29977c1
d890726d90bcb07a002c97212698d54dbf973a3f
F20101109_AAASET lomascolo_s_Page_114.tif
384eb5911f8abd11b6599fa79242d12b
8130ffdfd375f64b1a15944e53db6ed0041538ae
480564 F20101109_AAARZN lomascolo_s_Page_127.jp2
47fd448451b8d4a24316c27c5e48d81e
35c9faef78b90b75f1a06c655d87f1b0821710e0
F20101109_AAARYY lomascolo_s_Page_111.jp2
66e12b2f0e130fa5626122bc4960d305
2b57f94d25f3e459585d594739227876b95c38ff
F20101109_AAASFH lomascolo_s_Page_128.tif
85bdef584df488fa5f0411b480017bfc
7494efb00614eac19cab058cb4dc024d684eaaed
F20101109_AAASEU lomascolo_s_Page_115.tif
df5358c68312acc41df1e529ebb23e24
ceea93189670a9f1e294e42ff9a04c23261f124a
290938 F20101109_AAARZO lomascolo_s_Page_128.jp2
6c29fcb4742aa13556fce2a5bdf585dc
4db148106387e11641b748053ac1006537994492
F20101109_AAARYZ lomascolo_s_Page_112.jp2
d8271ff867bc843cd78c99ae88da5b71
93f293be86b1f163ef958cb99c3a0f396d0d81ee
F20101109_AAASFI lomascolo_s_Page_129.tif
2f6a0a1f15fac4abc3320912ddf4b85d
d178b058d54584be27836ddf90f562e855e6af2a
F20101109_AAASEV lomascolo_s_Page_116.tif
f18075ab217957ba431aecdd886f6b06
d38423507e4497f82e1cde350037c2f111a49cfe
451865 F20101109_AAARZP lomascolo_s_Page_129.jp2
3e0a8653af7f9fa6400815b0ab04e02f
6379770743c0f5b3f05b0a1338aa82bfdbbd9d0d
F20101109_AAASFJ lomascolo_s_Page_130.tif
11fb2ec3bfe73c0ca6c8031fa4feea6a
53b33f4f2b6ec698155f26d7ab31315539aeb506
F20101109_AAASEW lomascolo_s_Page_117.tif
bd2069aced47e2ad231170ac7cd267b0
794c0e731db7ebb6a5864985ac0b5cf1599bf2ec
634377 F20101109_AAARZQ lomascolo_s_Page_130.jp2
dde236d05110e1972c72c13f1713f72c
7096cc5b0c90e3d3e454b794da9a1c71b8de62f9
F20101109_AAASFK lomascolo_s_Page_131.tif
276044b074a268a4cff7b1b68ef998ed
00cf5bf3a925a702a1a1f6b80b17da0dd1315c75
F20101109_AAASEX lomascolo_s_Page_118.tif
7a13bcb026dba61f847a0ed509848654
4cf250fb56d20b296ecacfc67d4792f9c2ff0cfb
F20101109_AAARZR lomascolo_s_Page_131.jp2
406841a60d8899f3bc2a8e61485c2df0
d5c6f0db283ae726d384a585aaae0b818440bf60
F20101109_AAASGA lomascolo_s_Page_147.tif
fd99442110fa55f8d5c1bb57149ce39b
731b57993eb048401da82b6f373ba1da351e67a5
F20101109_AAASFL lomascolo_s_Page_132.tif
a153f8664a5fcecd566a4f079020a1ea
ef9dfbefbef015304b23de709e0568baccf87d4c
F20101109_AAASEY lomascolo_s_Page_119.tif
04934b1248b18804367a8c6849547f6a
2779e0015ef59f811a5c354e6fce6e54f1ed8fca
F20101109_AAARZS lomascolo_s_Page_132.jp2
a1c65e6289b9a6b113bf804378f55260
1c6f562cbfdb51b65e1bf985993e1dfed36e4040
F20101109_AAASFM lomascolo_s_Page_133.tif
73260fcc9524986be485195df89f51e3
769928b7b8774aa92657995fea811badf7a1bc4d
F20101109_AAASEZ lomascolo_s_Page_120.tif
bfa8ac6a051c42115db5d30b82cd165b
4a12c3e96bb12df9f02c3371aaf2fb86e1495da2
F20101109_AAARZT lomascolo_s_Page_133.jp2
2990fc680021ecee4678a20685957879
fcf152470fe3741bff06b09f71fdabd2ec95facf
F20101109_AAASGB lomascolo_s_Page_148.tif
3713df8627b7253ef27a082eedf5d1f0
38ddf3759db347373c7af4f29ade72dd7084733d
F20101109_AAASFN lomascolo_s_Page_134.tif
2826d1a7ccd92389b1707cb02d1e9b07
0f32e76ac6d62823774ce0c8823ca543d58fd1d1
F20101109_AAARZU lomascolo_s_Page_134.jp2
e8a16f366637f196eeb1fabe00c2e9ab
b83655a965b22abd4c72b6ddb00f7c689564a22f
F20101109_AAASGC lomascolo_s_Page_149.tif
3b3a7a8fad05a4588f144ffd2457bcd4
befe379a0f37410ce08f0dadb0a38f77ffac25ff
F20101109_AAASFO lomascolo_s_Page_135.tif
8ed035b88b1673a678f348e8b76c7bfd
ac1617622e7d8838fb4671b565777a16289f638a
F20101109_AAASGD lomascolo_s_Page_150.tif
3d1adca99c7aa8addbcd7aca6260386c
8428e335bdeeaaa76f70e03f6aa563a03b53cf8e
F20101109_AAASFP lomascolo_s_Page_136.tif
559485a983ee772944ea72268d27080f
a11170701ba4884fd2f412fafb7c9d4f660d05f8
871946 F20101109_AAARZV lomascolo_s_Page_135.jp2
f0902d0dbca9b7ee62898b60a69ef94d
729894fbc00d721a8723775f06b19c4b802c2b5c
F20101109_AAASGE lomascolo_s_Page_151.tif
aab01cec618b4d85532712257c55a49c
6a4a7449077d6d4b41679b7af24c354dba8cc066
F20101109_AAASFQ lomascolo_s_Page_137.tif
7864824cc8fe9379b26a62679c168c86
7ebff6a77607adf1d6b4f23afa4b72533896ef12
1051930 F20101109_AAARZW lomascolo_s_Page_136.jp2
beeda63661d82b35eb243f1177340fb7
a583490c09781f6cc796e771089e4421fc06b870
F20101109_AAASGF lomascolo_s_Page_152.tif
b7fa522d42a7b53bc49251c705c8944a
65c311797af2b0998767122f12894411a2e58a36
F20101109_AAASFR lomascolo_s_Page_138.tif
ac403ba9e11243d2be2040672fb001ff
1fc73fbf6342b43dbaece70bc7a97135559e4050
F20101109_AAARZX lomascolo_s_Page_137.jp2
d1b3c2ea2e2a4bc56a506bd11edeb27b
c21bf0d339f8db8bd1de41eb0a8966fd71d556d3
8030 F20101109_AAASGG lomascolo_s_Page_001.pro
c0a341f2ba5923cec4767d48ff1b8ef3
e4c8bdd024d4546123441eaea5f2e40830da7d47
F20101109_AAASFS lomascolo_s_Page_139.tif
56c9979fbd93ace03f6e97a008a442cd
447ceeaf60901f977fb0fa3cc37fec2ca61a5879
1051921 F20101109_AAARZY lomascolo_s_Page_138.jp2
8c6191c1bcbee30f210495015014339e
fa271a61b158beffc27d78f99260e8bad55fc9c5
1097 F20101109_AAASGH lomascolo_s_Page_002.pro
40db8a621612197f48a48a8a9a727fc6
e170416a976aaae0b68b683963a734ed6df647af
F20101109_AAASFT lomascolo_s_Page_140.tif
4ed40ff4b0090da1d1ce8f21adaa819c
d8e8fecc9ed3f4e0d1b1ea83a2c158cee3b4442f
F20101109_AAARZZ lomascolo_s_Page_139.jp2
873c14ef6e58df3eabbd2e31d4f6cd6e
ba8cc908c92e8a52a097d9eecd42461ffc2553d9
4245 F20101109_AAASGI lomascolo_s_Page_003.pro
70b59ed424ccfc548787ab3871ccd4b3
30acc4c2b2092b302c9c8dd8fb84bb8bbb1bbe99
F20101109_AAASFU lomascolo_s_Page_141.tif
1edac85832c8c7231598714899a4915e
ef288cc83fdb0d10968e4c88842220b56a2c296a
54111 F20101109_AAASGJ lomascolo_s_Page_004.pro
4d9b5c16f42ab65174f01325bd2bfccf
60795b1b09afe43ba6121dc8c632495656a9032f
F20101109_AAASFV lomascolo_s_Page_142.tif
d6d10aa0bf3cd4ebb15877236e0b0f11
b99cb4c7fc02b8ea9b87589614fef040a35141d3
51931 F20101109_AAASGK lomascolo_s_Page_005.pro
6830d90d4a798487f7d2ca79dc33a0dc
2463c7a180a598ae59f9d8851fc8aecc291fff61
F20101109_AAASFW lomascolo_s_Page_143.tif
594bc5c35d0dd08d095da8c220d1568c
7c10d169fea1f8e04d68bce41febd7c65e3f44a0
55108 F20101109_AAASHA lomascolo_s_Page_021.pro
42a6d478fcd9c8d00b0dd29e7ed53999
fe3d36de7055c9d457ed07d64ae21898d171f715
51832 F20101109_AAASGL lomascolo_s_Page_006.pro
1b42edb40f9fc2d5a1b823e5759dfc68
d54f36947b02059a419c8dfeb86984e96bd002c1
F20101109_AAASFX lomascolo_s_Page_144.tif
54890030ef1b6bbcd81b574f6199382a
480ea6b4a5e2bdd748ee4f45d43d8ba196c1e989
56190 F20101109_AAASHB lomascolo_s_Page_022.pro
1b8cbbbc72a854e0a39095a96e09b1b6
e762b5f32cf6133d665b194bd28b1327661fbf84
37866 F20101109_AAASGM lomascolo_s_Page_007.pro
1d4320cb4770619308036d69f277f9fa
b4383185585c551bb7a123fce8c35861c5abc13a
F20101109_AAASFY lomascolo_s_Page_145.tif
3e3695a0324b0faa7d8cd6b6ad0a3bd7
08c292894ca61e719d43faad83d56e00e0b0024f
90201 F20101109_AAASGN lomascolo_s_Page_008.pro
2ea1b323e6f4f7b2a62ac47a845cddad
50f37896049511ba558adad752dfd463fb50e01a
F20101109_AAASFZ lomascolo_s_Page_146.tif
577669ff6ce975d4f3af7933b02a1e59
f08a86bd7db4a22dcca4d013a9add37e151201aa
45654 F20101109_AAASHC lomascolo_s_Page_023.pro
1786c34540aa8aca9a53cec9152e5a66
6424cae29962ee5ad27cd5bb2d590c3e6815db93
119875 F20101109_AAASGO lomascolo_s_Page_009.pro
21d9f4e0e49f862637948367c9aeb199
838e3b8ffcfc1acb67aa51f56b5b8174c71d7901
54145 F20101109_AAASHD lomascolo_s_Page_024.pro
0b41c91a93e6ca9a6937e45d57f8eb18
07046dafadf34f9cf2363b4d7cf39583b0622933
35223 F20101109_AAASGP lomascolo_s_Page_010.pro
3bd1d0eeb7b692af24dc51d140848001
da471ac748a50eecdfc8a004627334cf04ee57dc
52678 F20101109_AAASHE lomascolo_s_Page_025.pro
29cfab3ed2eb930b3d733006730baded
abcf45dd2f080c8b007ae57b67aa81a7b2a5f660
34973 F20101109_AAASGQ lomascolo_s_Page_011.pro
0c54b0f840a874708d47463072d3a539
22a41524fe62abd12d1900c77de7f305f0d4c55b
50490 F20101109_AAASHF lomascolo_s_Page_026.pro
83ed1d992263a2535afa397bbc41a87c
3c3be1411fe2c255d2e8df9ad838566ee6df9660
73136 F20101109_AAASGR lomascolo_s_Page_012.pro
a3c6de703025dd6226e40cd53910c6b1
24e4575f8d9a650866bb433e809e4a14375ae888
55081 F20101109_AAASHG lomascolo_s_Page_027.pro
2ef8a3be89128f1b1e9656f007f03798
127463fb74741cb8cd53f1d33547492bfe8ff001
2787 F20101109_AAASGS lomascolo_s_Page_013.pro
ab17cfe180217df2bc85d69d391c43c2
dce2bed028d293d5b353bfa417b7648ba6be1ba5
51862 F20101109_AAASHH lomascolo_s_Page_028.pro
276d2990d177d0b97e58ad1828208b6e
46af2e4b4275e8d9efbb92f1f6f9397658c71467
47669 F20101109_AAASGT lomascolo_s_Page_014.pro
f043286e85a94a30169f70b66816ff42
8e9cb9044d95c52a47d0d57f5eaf189a10bdfb90
55665 F20101109_AAASHI lomascolo_s_Page_029.pro
950676c6971b6f1d371adeeb4b66ac55
f768704303feb7b726ae362fb78ab847a6f9a56b
25660 F20101109_AAASGU lomascolo_s_Page_015.pro
dc00b7ed6bd78358a81eb756d29f3147
2f238e841a184c09bcdb666b5f3205aae2f9cefc
52105 F20101109_AAASHJ lomascolo_s_Page_030.pro
2f65f92a0fad726b2dd1983a63b66f58
8d10f68c4408525f4bf5de7ca5ec372cd8be56a0
49881 F20101109_AAASGV lomascolo_s_Page_016.pro
ef378ab8c0d5be64ac58309c24c09993
466ab5c7da798180e549d46f62edff7e8eb19292
56982 F20101109_AAASGW lomascolo_s_Page_017.pro
519aa2442e87e8fbb0ba05f8c3c3c5b3
f975d6fb98d9f5ffb134e215d4fee2bf4339d1f7
46609 F20101109_AAASHK lomascolo_s_Page_031.pro
9fb537b735238a1cf89e3ec6ecacd476
dc6363d2f8a3a8c8cea5a65be29de103101a7c2b
56691 F20101109_AAASGX lomascolo_s_Page_018.pro
338d6f3cd0529d6e0b6063c6b5538b35
c84c346c125d21aba1d9bddfb9a4cad5b2eb43cb
24075 F20101109_AAASIA lomascolo_s_Page_047.pro
e6d9bb47e5254fbc924a7ba41b4588d4
03449e2c49b8ab7dc04b83d6530086304bfc74fa
54642 F20101109_AAASHL lomascolo_s_Page_032.pro
e6d3d5285cc4d69e317c0b49fd7efdf2
7bf1e9e326806765df2796131d88d097f570ea27
45678 F20101109_AAASGY lomascolo_s_Page_019.pro
9ad211863e453d884a74b789255a0947
57780d936bc18b40447d002b8aed16fabc426922
27778 F20101109_AAASIB lomascolo_s_Page_048.pro
b7d97d79ce8f270d163dedf5d2232e00
aecbf7c7982512e8198459758a3a61a3f6ddbcfc
52527 F20101109_AAASHM lomascolo_s_Page_033.pro
4894562660494e48c2cf527eaa609d2c
32c452e54b9fdc7dd92ea2a5b14bf18237b5077b
53699 F20101109_AAASGZ lomascolo_s_Page_020.pro
1a108ed134fa1b916b976fdc493332c6
39475092f196175dc00fe96226e8b7167673cfcc
30316 F20101109_AAASIC lomascolo_s_Page_049.pro
b77c438eb304705c812e414591723f4d
728d21a627f8931dcb31fd7c47d96960fe4f6850
56294 F20101109_AAASHN lomascolo_s_Page_034.pro
6e706c214508dc899b14c4d6c3c6f95c
7f9462d40f26c52b0c142f8b35a4d9f5e1481734
54394 F20101109_AAASHO lomascolo_s_Page_035.pro
f1bd2bed486a1e9394d95bda083fb37f
350788f62cfb4140ad7f41d6d2f5bc67f0e8beb4
21957 F20101109_AAASID lomascolo_s_Page_050.pro
7e62501569435803524263a60cf00e48
f7962a7bee692204d2e9deb8016cd516b4a1ff49
57030 F20101109_AAASHP lomascolo_s_Page_036.pro
e911f1eabf2d7e3bfe47334f5afc8087
7b8f58003a3bf2bf2326a48227b76bfd7a1f5994
55150 F20101109_AAASIE lomascolo_s_Page_051.pro
fe156e30fde6d5707decacbdc4e43dea
3bd418adc55ece7b3394fcc40136ba9932441701
55832 F20101109_AAASHQ lomascolo_s_Page_037.pro
93129ccc6074b683a1dfac3ce78db60e
6a0a049e00d7cc64e9f789c234bb2c499db00a9b
52130 F20101109_AAASIF lomascolo_s_Page_052.pro
0585806f0cd7a53b1590bee0f4de5bfa
cabdb49c1fbf7279f2dee55663aa568097340cd3
15771 F20101109_AAASHR lomascolo_s_Page_038.pro
c90c2ec9124c720633435bcc8e38b95b
3401602065fb482e0b71e899b0697c912c42dbb8
48220 F20101109_AAASIG lomascolo_s_Page_053.pro
4495204cf19990c332486a7227a1bd32
68e3384173b87412feb26aade8d3f562ecca0588
41690 F20101109_AAASHS lomascolo_s_Page_039.pro
5ba9cfbbf32608490376fcccbf402a33
089966b868d831a00906ae5658de230dc6bde46f
56328 F20101109_AAASIH lomascolo_s_Page_054.pro
da8f25fd6b8a66895612d87c9f936e4e
d406c63b80f87c1b700dc75f2010bf98696bb225
16544 F20101109_AAASHT lomascolo_s_Page_040.pro
0417cfabf037e2e02917c79ab0806b82
02c2fbd110238bc62c2bcc1b2fb6af0d47e1e519
55030 F20101109_AAASII lomascolo_s_Page_055.pro
8c274e676e2fd0da24bd7e1fbeccf73c
6911112fe542be8d709640001990c9b52940d2a6
35162 F20101109_AAASHU lomascolo_s_Page_041.pro
61c8853ccd0bde7b951d5e3cbfa6e4f8
f40967ddd52e712f2b7f47072294aac71a83b24a
55615 F20101109_AAASIJ lomascolo_s_Page_056.pro
d1d0a77029abbb20a3cac5dc066d3582
c7ecc1c36657bde570eb3661696e981e1b61de11
29676 F20101109_AAASHV lomascolo_s_Page_042.pro
763dbd6f01cad61f948f055045bbced5
52b92094424de71b8abf8ac58147a39734739fe5
55548 F20101109_AAASIK lomascolo_s_Page_057.pro
e4233c48351f3b1363ecffa79a42523b
c3ce3057ec47c868c9a63dbc4909e654fc970950
52230 F20101109_AAASHW lomascolo_s_Page_043.pro
d2156e1fe93a999bbfe53464500bb2b1
43149c1d0f68936d3e6b49893bbcd945abf8b461
53470 F20101109_AAASJA lomascolo_s_Page_073.pro
cb0c4e710579d4bb3b7fef29c5b82162
5fffaba3e49318992f6a4b2bee0113bb17e14dac
50129 F20101109_AAASIL lomascolo_s_Page_058.pro
0ad1eb49cf9687454d0ab4d0d284b590
b84ae9820b5b483b66955a66d87d22df8ba3fd1e
46834 F20101109_AAASHX lomascolo_s_Page_044.pro
8507500404874136b96515d018ee0ac4
0868d85dc6e686a15610e8e37664d1e06c0ccd0c
57375 F20101109_AAASJB lomascolo_s_Page_074.pro
8f98364eb4ecfdf3ef1e7b891f301f33
46837e6e5a3f3db6c664a10d667af347fc821068
50701 F20101109_AAASIM lomascolo_s_Page_059.pro
8931eff66b8a36b0f3edc13f113ec07e
3bd413d1abb0018a19e5b18e8445d30c7077e5c5
34974 F20101109_AAASHY lomascolo_s_Page_045.pro
05a34bebbdeb2cf47e548c55da0efabe
2a0557bbc32b58e01f62fd4015fd5071ef3df08b
53017 F20101109_AAASJC lomascolo_s_Page_075.pro
8e7148b4808011d1b6db12b12a181a11
ee20ccf4880d3565894c100c0e664148aeff0b59
54763 F20101109_AAASIN lomascolo_s_Page_060.pro
d0911689702148d2918f5e50d336f02f
8bd8ded2f64c0fd89a196204c0ff57945333cbbb
30288 F20101109_AAASHZ lomascolo_s_Page_046.pro
0f84ea7233e50533589e570e249f7f97
2321b110192e395f81822e503725291e2c7ef136
51901 F20101109_AAASJD lomascolo_s_Page_076.pro
0dd4d98b987f815df4a95fc0d607d240
249dec9b5507af4b302bbcb7f3275d6717209bb6
56171 F20101109_AAASIO lomascolo_s_Page_061.pro
6a4cc21538786573484f48ada0ec56b3
616a66638a1a883422c4cae1b16aad31c3d6d95f
56841 F20101109_AAASIP lomascolo_s_Page_062.pro
d886e7702b3d8bfc3510b69685660cc8
864ed4f143558e87a8574a81ec774ffeb9f16e8e
54245 F20101109_AAASJE lomascolo_s_Page_077.pro
40bdf203952f343f6413d7747b7a2ec1
92739affd2ae65d310bd68e3d0b054f7c820518c
54702 F20101109_AAASIQ lomascolo_s_Page_063.pro
eb27c5afc7b210403c86b5efc75988c4
f9908f4657c5d38a745d2d3be6fdbba57c55b36f
58227 F20101109_AAASJF lomascolo_s_Page_078.pro
acf79c72f143da45bc4668a4fb809089
7f4f3284d58ab4bfa2761d6ff2c455a9795152e0
56741 F20101109_AAASIR lomascolo_s_Page_064.pro
cee40632555e08b5aa5fa853be8850a8
bf5ff050162dba003df366de515ac8167078b152
54079 F20101109_AAASJG lomascolo_s_Page_079.pro
f0b1fe6d3d481c3303589ecd5a9bd06b
51e37e9258847db7e2b23a95a99e5e9008fdec00
52102 F20101109_AAASIS lomascolo_s_Page_065.pro
1d790a7c9ed01b062f1d6b6810a719e8
f68759faf8153dbcfeece4b99a52806d87ff7163
52927 F20101109_AAASJH lomascolo_s_Page_080.pro
d43c4e0e4eb6211b97f9f1a350037e71
022d309097680838744e2e65ef79991de3a05696
26180 F20101109_AAASIT lomascolo_s_Page_066.pro
f8008e575661d4dac2c819c03c756ab5
518a665e24532447ad5406e9664bc898bc32019d
53430 F20101109_AAASJI lomascolo_s_Page_081.pro
6c6049b108184dbd06f4bf0ec6b48e8f
8d49e7dadb55b2b2416a3f5ba47555d738887a46
13361 F20101109_AAASIU lomascolo_s_Page_067.pro
b73cdece68593c6f3fb330f2d3e27165
39ca4bdea16cd789728c10365d576a3ff0392f6b
53900 F20101109_AAASJJ lomascolo_s_Page_082.pro
0a70c06723827afc32f44cf3901c6991
963a4d186e789d00e8f5f862bf6cbb2da5a0c0c9
19440 F20101109_AAASIV lomascolo_s_Page_068.pro
4f26cf6b2c04c9f605a575a931acff65
e0d097d5511f3d8b7a9338ead910f188255d0d29
54683 F20101109_AAASJK lomascolo_s_Page_083.pro
6f1e39201c3eae286ab9324f28c2035c
18efc326d2d4048cbf39a3a11d61f9cbf3526123
8555 F20101109_AAASIW lomascolo_s_Page_069.pro
d8d11ac1a0dcdc7a53a3c7977d0fc9ee
21e30ef9a4402ac60e264368508aed098a2288d2
55521 F20101109_AAASJL lomascolo_s_Page_084.pro
27a3351b6fb2c71994a4b5d088f9363f
94700bde44e9a81e9ded9c2de8762d942ef2d12d
54680 F20101109_AAASIX lomascolo_s_Page_070.pro
7544f1f5ae5bcfe535ceed9616089c77
5afc1f3a6ea41ce3be6efe9248518842e77a31fc
5441 F20101109_AAASKA lomascolo_s_Page_099.pro
8b29d9b7131addd16a3f603eba2fd595
2d90de3e6cd76bbeb2ee4e4dc95a3fa29e2b4a06
50825 F20101109_AAASJM lomascolo_s_Page_085.pro
f7e632d0801803858b1f478349e64df4
bcd1736b60d7fe5105cf69cc4b40e03aabd26621
53673 F20101109_AAASIY lomascolo_s_Page_071.pro
86f40741bf7c3e1d1ad077982aee81c6
f76e82a7983f088d82aea35a51af95afe40ea0b9
21903 F20101109_AAASKB lomascolo_s_Page_100.pro
8ba388f1fc131b500996eb1862766f36
59e250bad1e28af2c70fdbeac9532088658bb37a
57950 F20101109_AAASJN lomascolo_s_Page_086.pro
ca2ec6611a00890c9911b72ac7de83be
8e802fb4e8bdf2a665e9c2126ceb666c6ca28ec1
58045 F20101109_AAASIZ lomascolo_s_Page_072.pro
896ff0232c4fe254e66d15a6595bb8f9
16f1d9ca0d9d0290573020155982bc64b0f450c8
10965 F20101109_AAASKC lomascolo_s_Page_101.pro
bc2d80ff6ae6ffd8aa1a8bf10c47aa41
e9bfa9c5d0c65ea5d35110ff5cdcaf2ac34cc0f0
56608 F20101109_AAASJO lomascolo_s_Page_087.pro
24d38aba3fba4a79f68b4dd17d7b8d22
eb02177fdcad82342da1229452cb59bec09cd8d5
25296 F20101109_AAASKD lomascolo_s_Page_102.pro
5dd5bdb538ed229581195bf7b117adc0
f64b4cac5b827584352a05d609bc0c96d8d87e67
54714 F20101109_AAASJP lomascolo_s_Page_088.pro
6aa4e626dbea5debf72d4eddb315b5e0
5213e8a3c41ef45e93cff59aea85daa775760b4c
12734 F20101109_AAASKE lomascolo_s_Page_103.pro
6a18b3a4f1d0f9cbbe6c8dfd12c51887
a9086f8eea8a915c7d6ba04fb157dd5689e55d2b
51700 F20101109_AAASJQ lomascolo_s_Page_089.pro
ceff07ded6189b436d83b9b28840f528
bd36dae811e8d8340a28bbb594cf3f94b73ea60c
52738 F20101109_AAASJR lomascolo_s_Page_090.pro
d2e817a2e65ca27ff8a85660a61d3948
fc5485a2322bf064b3bbc9a10907ff1acfd85e39
17292 F20101109_AAASKF lomascolo_s_Page_104.pro
fad9d660d9f0a33ce61297b521f1fcad
c0b9547503a0f41cc9f526b45d0030a6c3a5bfe8
57987 F20101109_AAASJS lomascolo_s_Page_091.pro
8f6e6f8842d176d0ec4f73429616dfad
d1f8c0019613e5606433b17cf98378e891d8611b
41126 F20101109_AAASKG lomascolo_s_Page_105.pro
50e7e04f43ba1066d85607b7d169b4ab
2817778f3ea3e13136926e7baa91360237358706
57888 F20101109_AAASJT lomascolo_s_Page_092.pro
deb190fb3c142eec0550d41da1e4b1cf
56d72a2deb19f9653ad9bb7ddfea743f1ee16e9a
43507 F20101109_AAASKH lomascolo_s_Page_106.pro
02a85c6da32d004018dc4374df62ffeb
c0baffa58f74125fcac8695555be17a03f25b446
54963 F20101109_AAASJU lomascolo_s_Page_093.pro
eb6ca36e7676878f2b2bfd2e4226bc08
480411007baa263444f561fe639f302941a401f3
53144 F20101109_AAASKI lomascolo_s_Page_107.pro
a248c342583aa922297adbc674979737
18ab7d6329b7712a669196d0974ad620df9808a3
27246 F20101109_AAASJV lomascolo_s_Page_094.pro
df742ee45172f6fcafd179b45adc6258
b016ce8b81be110bc814f51549cfb2ebf3632ccd
55662 F20101109_AAASKJ lomascolo_s_Page_108.pro
2dfa00ea5333c6ccc35d4dc022ba3095
0c1e6eacabb34e0a7c5a599bb87c575f7d2e69f6
27558 F20101109_AAASJW lomascolo_s_Page_095.pro
897337b3c268f78bc5621f5388f4ac54
7f142d3bb2be45e75d327f2f887030ebe9cf6394
56534 F20101109_AAASKK lomascolo_s_Page_109.pro
3f6810b26a2682434c1740bfec0970ba
561d2083127bec95971c2a1ead3c9055d92723c0
16268 F20101109_AAASJX lomascolo_s_Page_096.pro
40078fe3e146895ac29ba28543d0ed57
fd84e58a8b8e868bb88f2cb66074e4d4521ad7a6
13319 F20101109_AAASLA lomascolo_s_Page_125.pro
0587d866896cacb4c43adb9b4b4db658
16881330c2afbf1781f3f80b35df8067f75cd0ae
53933 F20101109_AAASKL lomascolo_s_Page_110.pro
f5ef4fa7f8746e0259929440c810c3c8
85e9bb2275b64b13c7a5a74549a790c5c8dfea21
6975 F20101109_AAASJY lomascolo_s_Page_097.pro
efddf0d9013588ae3e1d74d9c7cdc52f
ec112ca08cd6f1592a6ff186df2586fcd349da72
9631 F20101109_AAASLB lomascolo_s_Page_126.pro
11e5df7fbe162433bf6b240c99e71435
8be89560fe3f6a5d2b0e0b9e447e4797ee775f6a
47881 F20101109_AAASKM lomascolo_s_Page_111.pro
f03e07fbb05eaaf9d950416ccd394d21
91cf8aa0f7fd5f1e70a879f092b9d2eb92d06dc5
16417 F20101109_AAASJZ lomascolo_s_Page_098.pro
38b242fa7dd36390b07009f9242a056a
deffa3275bf151a25fb3436917dc2c9632277aad
15108 F20101109_AAASLC lomascolo_s_Page_127.pro
250fb938c02561fcb79143a4a4edea98
6c4368878eb3c3e635ca981077a85b606a10398b
57684 F20101109_AAASKN lomascolo_s_Page_112.pro
b2c96791a8731deb7c13c4dc8fc22ca5
b3f3ad275541e71650f17e5722f24aaf32aa192b
21000 F20101109_AAASLD lomascolo_s_Page_128.pro
6ac147ff819e7f512f45a2aa7b76af89
50a0dcdc1329c9f15c4aaa789f4b2dc7e9e87105
52265 F20101109_AAASKO lomascolo_s_Page_113.pro
4e8e7194038e7206e43bf2f63556c9a8
0532ec4378564ac24cf692c8b72b9ddc6322c787
32970 F20101109_AAASLE lomascolo_s_Page_129.pro
e1e8a059e2cc7a9658968a9a44cf0578
c8cee3d2a375cafd71319cddf439b3a320d8b67d
51276 F20101109_AAASKP lomascolo_s_Page_114.pro
2ddaf1ecfc12de3cf23c64ed3e738b78
6f505150831ea7f49573594360dda0f8f7059784
39319 F20101109_AAASLF lomascolo_s_Page_130.pro
c935b7b3b2ef0595fdb1f6016499f490
71a75a7a1c4968add9fea9bf2897846c7147fa72
57717 F20101109_AAASKQ lomascolo_s_Page_115.pro
b65e3219adf12302d2a087c93fb7c3dd
e4f3b889de478f366059668bda7de55d2dc47e11
52496 F20101109_AAASKR lomascolo_s_Page_116.pro
efa7ceb043d639f7074cf3a6fb73c720
0ccbbff0bb5105df8dd7b3038d3cc80776e9e389
53657 F20101109_AAASLG lomascolo_s_Page_131.pro
0fe3c490c7b5c2cdc09ef729bc348e45
d89e7f752805d85c614cc46b63049103a7a26dd3
55369 F20101109_AAASKS lomascolo_s_Page_117.pro
a89a3cecaeab292b5369684093c2a1fe
d58e0dfd12a5430e8e42d6724a182f6557eff73e
57036 F20101109_AAASLH lomascolo_s_Page_132.pro
f72a04adf61d5a5c6b6746fe4846ebbe
f3f4ee76d09999751cad329febecaf7b9e4d8d67
54793 F20101109_AAASKT lomascolo_s_Page_118.pro
6bcee81302bc1d10ad41126be6c24caa
b8feb964586f7909137ad0c322be88fd12b9c2b6
57355 F20101109_AAASLI lomascolo_s_Page_133.pro
10f6d04de4842901f1eb229fe3c8142e
3910ac953c9723683ef26489f5e3305eb1ca7d63
59101 F20101109_AAASKU lomascolo_s_Page_119.pro
e89416ba3ae8bc6fc0416c385e32b262
d200ce411cc919f16554fe068961dd67051cc211
57243 F20101109_AAASLJ lomascolo_s_Page_134.pro
3058bb8332de688a26c00837410fe79a
6c023c630ad5324896d589c85f009266b6935f5b
53774 F20101109_AAASKV lomascolo_s_Page_120.pro
947acadb58ba8d2bd5ccf772229ec8b5
42796b559864f8042844ea3a7d2236678daf8077
39556 F20101109_AAASLK lomascolo_s_Page_135.pro
0bc2b454f0111a1a57aec83cd65d6937
15169cfc6521dbfb910cf71c6877a08310c1f5c8
57658 F20101109_AAASKW lomascolo_s_Page_121.pro
eb8dad00e7cd4170012af07efefc767d
5b1f11303d89528fe631747857e2d51642939e39
11128 F20101109_AAASMA lomascolo_s_Page_151.pro
3e29cd4ffeca01efb2fad95a6441356b
f74b0b2f9e2e945eecd19ec5acef3b2691aa9893
53090 F20101109_AAASLL lomascolo_s_Page_136.pro
baea4640c70ecd5da538e9025d39ee6d
4d4c41bbfa948297f0c263d278a5c367f1733556
52603 F20101109_AAASKX lomascolo_s_Page_122.pro
15418293f97a171f0812eacabed09752
a23fe6be5cd2defd4a601e963b9bf10943825ed5
49121 F20101109_AAASMB lomascolo_s_Page_152.pro
7fa425520ad2425741c4196c7dc131dd
685e9cfa3bb9c2961766e51877a3b9064a8e965f
57260 F20101109_AAASLM lomascolo_s_Page_137.pro
f2aeef11c45d318b97a39bce5690bdaa
e5d8179d4ce55019ac80d2351db5eed3d5b2650c
15449 F20101109_AAASKY lomascolo_s_Page_123.pro
dca55bb3643de86ded4d437a14cd75c9
9c0e9ddbd250a99892497885204788becbbca9ae
439 F20101109_AAASMC lomascolo_s_Page_001.txt
5e091b56cee56191f45baf17c44aee6a
4c368617a235e915d95c75cf2e50c00a02e4e7f7
55514 F20101109_AAASLN lomascolo_s_Page_138.pro
c1995bce1ee7ac70e643e11ac31e64ea
9ed777c6aeeab991341e20435126e45713c601ba
5181 F20101109_AAASKZ lomascolo_s_Page_124.pro
793913d0cd5a026655a75537fbd1642b
2812891462cc51199adeb9658541caefb9861619
100 F20101109_AAASMD lomascolo_s_Page_002.txt
c437e3b6c66a5ab7df3bc5a8e17d7d4d
c7c6b1803d395004ffb5c132c44f640737ad0d45
60504 F20101109_AAASLO lomascolo_s_Page_139.pro
a6bef4a2f46087ada95504bbed2ee883
797551068d177f65adaf2c959b3dc9a0da59c0b7
283 F20101109_AAASME lomascolo_s_Page_003.txt
69eef2763d2accec4e2d75889f8e6a90
c798f1bb3148df3dac6ddd29e466bb5e0dae7946
55106 F20101109_AAASLP lomascolo_s_Page_140.pro
ae434de569621a0b02122b9a5b30c692
dd395097e3c997a17c7ecdaeb4fe1e09874e9214
2161 F20101109_AAASMF lomascolo_s_Page_004.txt
ac0ad6588b44ddc2d707eecd302c34ae
31b1eac1fa135ead657039accd9137840ace154a
56264 F20101109_AAASLQ lomascolo_s_Page_141.pro
bb94b73e96210e656cbddfbf40cc5a20
4c6e666fce10449f724cf9c4b58a75f75a26e72a
2038 F20101109_AAASMG lomascolo_s_Page_005.txt
a50ed21433c588346661c86e68342919
793deefbed39fff73d98a4061c819511e2d3905c
55017 F20101109_AAASLR lomascolo_s_Page_142.pro
25e0707ffbd9a756eb9c3d2674db52d3
15cfe9ca9f560f97d40f009dec0c137812f62397
55856 F20101109_AAASLS lomascolo_s_Page_143.pro
6db090b5818a1d2845c740c8fecac5e6
802e504f9beb994fe67e5eb6ac32647f69047f9e
55704 F20101109_AAASLT lomascolo_s_Page_144.pro
293a26cf5a02166fafdadad561058160
0ad44e117af169f6f8a31260493b8f03cde25c2d
2043 F20101109_AAASMH lomascolo_s_Page_006.txt
76cd9420249dd7a7f36ab487554fd27f
5fcc02f009b644720336c12df3efda6bc8274e9f
61602 F20101109_AAASLU lomascolo_s_Page_145.pro
0ec1f1b974c8625dd76fe88a07a9a37d
656b66802b73bc26f099f60d8b8f168fe10f4ffc
3823 F20101109_AAASMI lomascolo_s_Page_008.txt
eda222584fb0174f594ad5703f2d1c06
0f5aee76e3ae89840f5555c5e7e12a083bdce39e
53613 F20101109_AAASLV lomascolo_s_Page_146.pro
6a043a7e5188ebdeab454aa4b31eaeb3
31fe86c606ea72ba8f7f3eb92ffa152e45dec3cf
4951 F20101109_AAASMJ lomascolo_s_Page_009.txt
faf981d04b2b77ef30303674980b1300
3f79ba8caf6b922922dd5aacacbe9f51fc842a71
58764 F20101109_AAASLW lomascolo_s_Page_147.pro
501638eb36c3a8279b62db4635ec9657
4f884c49eca547f34c8c29fb082d6c45377c4d5d
1570 F20101109_AAASMK lomascolo_s_Page_010.txt
d010a8faf79efb1d52dab4f5014a165e
b779aad3d25b52aa39f288e05531b928b7174674
60651 F20101109_AAASLX lomascolo_s_Page_148.pro
fb4e69b8e9fdb4158d19b96dc81ef5ac
0e469e670ccabbc2afe9c086ac9bd22bf8fb4324
2051 F20101109_AAASNA lomascolo_s_Page_028.txt
f422b6c685d8a6bb9c1a77ba88b3dafa
163534d9fda41142e0899a37487feada0785a453
1453 F20101109_AAASML lomascolo_s_Page_011.txt
af574decb17c46513b3f3a1b35b55ea2
532696e6a262fdc20e7d785196b694b0cf7b9645
63158 F20101109_AAASLY lomascolo_s_Page_149.pro
24cc70f19a09c81ca82169da25b00220
760e314bd901d5b90344887256e618b3f06bce19
2181 F20101109_AAASNB lomascolo_s_Page_029.txt
4f944d05f7317ea9475eef22a6219402
901e3bb90ce70c46e4e548e899ca6d44fcfb0d84
2949 F20101109_AAASMM lomascolo_s_Page_012.txt
d3749d0cc8125c99d6718a797befe044
3f13700e633cd55d06cb50bfea5d091161c2064d
54094 F20101109_AAASLZ lomascolo_s_Page_150.pro
25963bfbb67b03618846e3664ccec448
d4703f8c763000035fe99d052fc9aa6202dcb95c
2056 F20101109_AAASNC lomascolo_s_Page_030.txt
242fbab2f191ddbe84f18196714255f6
9cae9e7fc553135f88e78eb1b8a09705be4fb620
115 F20101109_AAASMN lomascolo_s_Page_013.txt
4944859d53729e85cfaaf8cbe13f902d
6594d45825233e8ccde3b741d39c51a7d48176ec
1892 F20101109_AAASND lomascolo_s_Page_031.txt
8080a21d8da48a232e5df4a9fe4bedbb
3f363c9b649a2edb24faa0fd015ba1db0211c405
2052 F20101109_AAASMO lomascolo_s_Page_014.txt
0b35f5a42d5177142f181f83cfad7eb5
a2d674c11c74c9058552b516ff45c7a05ba7462a
2148 F20101109_AAASNE lomascolo_s_Page_032.txt
959e7ae86d35607a47b788031f035854
7f48dd97601d33f5daa91f7fc2a3de9c3bfa2a1d
1017 F20101109_AAASMP lomascolo_s_Page_015.txt
c974a26d32b9a122ad68d9cca04aaf95
14fb9084b5ecf2b4cb4024d52e33104d53554f40
2108 F20101109_AAASNF lomascolo_s_Page_033.txt
06116b9a26eb803c17cb800c0d88093d
32bada406945941a38763aa5ecd8651a6a9a6dde
2080 F20101109_AAASMQ lomascolo_s_Page_016.txt
1ff96f39206f7d5a61a9041e4231bcf2
34471db50b27899caa4e90d8edfa7b4bafa4c223
2209 F20101109_AAASNG lomascolo_s_Page_034.txt
39001db4020110c06100f837b0c6f6f5
e7fc36fd02f1c52fcdc1340f226f1b5e31f93168
2233 F20101109_AAASMR lomascolo_s_Page_017.txt
04c7443bb84223b510acc00a8707bee3
6b11275cca6d410330de3ab4b42a114355c11b6e
2138 F20101109_AAASNH lomascolo_s_Page_035.txt
b2c1db562e9aee054606048f67f522db
8c047c8abd33e6457cffdb562352d28ee8be3e5c
F20101109_AAASMS lomascolo_s_Page_018.txt
ee3142cb4ff90faa2945880815ccc218
8a9ad5b3742c69b1bf2ace6a0c24156383c18323
1818 F20101109_AAASMT lomascolo_s_Page_019.txt
1db741c898427e8a4fc4fa37dbd7aa2d
9ac27d68cccd2b92ec5f339671fa510ba2d5d488
2230 F20101109_AAASNI lomascolo_s_Page_036.txt
4090c068535777f71b284e011dd19a37
44e510558aaa1b3ce63900a1a980d6553577a8ed
2215 F20101109_AAASMU lomascolo_s_Page_020.txt
1cfa9a38abf938770188298841832131
50f3a4d01c7221833302393397550d6e7d3f4c89
2184 F20101109_AAASNJ lomascolo_s_Page_037.txt
cdaf18e36f04202b7ba2e1db333fc337
3d78b2bc9e1351a8bfe15ad312da4c1a7220ef8f
2199 F20101109_AAASMV lomascolo_s_Page_022.txt
5daa52e90fb8d977ebb8598674a877d5
c1dfe65dd60f9501bf70427e2739541a43346bf6
633 F20101109_AAASNK lomascolo_s_Page_038.txt
5995563ca6b8c75fdd7c339a7e00382d
6c3476b6c62a5a6d5db76d836c5c899d78bd7eb6
1849 F20101109_AAASMW lomascolo_s_Page_023.txt
5f4b65cdfec2ca57ce34a1e2fb47efec
53f2183c272c5f688d1aeaf3cb1aaa4c4b5cacc5
F20101109_AAASOA lomascolo_s_Page_054.txt
673cbb893e4c57faafc355bf4b2c7a02
73e6ec4e7cdcfda7aa6070c019a5434cc9851e70
5348 F20101109_AAASNL lomascolo_s_Page_039.txt
df591e8b0c1bcc741ef072ec6f2da53f
00d45a95b5797b5e48c338745e362fc7d70cea8e
2133 F20101109_AAASMX lomascolo_s_Page_024.txt
ac2ae13e6bf6f52092f0c8adcabef6d1
ab8f397eb0c8a3a9a305988f9f019bb082624e8c
2155 F20101109_AAASOB lomascolo_s_Page_055.txt
ace3bf5282d75b74e41c652ea57a9c0c
984fa4e5e88bc3ef66dc28279fc397cea18c7a2a
800 F20101109_AAASNM lomascolo_s_Page_040.txt
8890a45cb5e9b4ac48d1e4d851a4d76e
c7bbabe536ca020cb719a2446551a83221b6cc68
2068 F20101109_AAASMY lomascolo_s_Page_025.txt
5dd2aefc058fdf0aafc7f07dcb1c7550
ca126a9d617c00c0e9a773902c2108dabf297b20
2177 F20101109_AAASOC lomascolo_s_Page_056.txt
c889ec6ca4da1012af00ae8f4d63a205
5979ab620692117826c0fc6dbc397087669827ef
1541 F20101109_AAASNN lomascolo_s_Page_041.txt
9928cf65e9901b8f0cd25a628eac74cc
ee9c17849345e517b17eb963b261085f98d99fb7
2000 F20101109_AAASMZ lomascolo_s_Page_026.txt
224a06bd07e78bc9d2a44a289db58b73
6db9b2e235379fea3086655d31ba0addf3c38d4d
2179 F20101109_AAASOD lomascolo_s_Page_057.txt
8195abc161f9353b771168c8850ba273
a858d7e244779c18207f10b424ead46f8ac074fd
2124 F20101109_AAASNO lomascolo_s_Page_042.txt
5ae6bc530f5b8f6acf33fa0b6a700b7c
9193ca4bfc04e353031c3068f825f38ee2a51b3c
2016 F20101109_AAASOE lomascolo_s_Page_058.txt
985ceee797b213a56392730395d45529
c80c0bbbd5265992d03533e5d175c9bfcd3396e8
2340 F20101109_AAASNP lomascolo_s_Page_043.txt
af0995cee169abcbd36cfc83833c6830
520790ccedc8a2e15caaf7c9014863298bb09b27
2045 F20101109_AAASOF lomascolo_s_Page_059.txt
8d003aa203d0b5c2f2653e1dc0f9bdf8
5f2df455531f4ceb2a32a7bc333ae932389f7466
2206 F20101109_AAASNQ lomascolo_s_Page_044.txt
1808c0ac6fa836548c6b759e3ff8af99
3d1f0a9a9a8ea850ae066271c1763214a63e6017
2150 F20101109_AAASOG lomascolo_s_Page_060.txt
517bff35a134139954bf6cf5edfb342a
255eb2dda05622b5a0db860425cd2622810366f9
1672 F20101109_AAASNR lomascolo_s_Page_045.txt
0e8cdf3e9bea2e31ff5b35fe85448d6e
20547311dacc318331fbb0150424dea21cac4da1
F20101109_AAASOH lomascolo_s_Page_061.txt
69b5e23d17d7aec2980de8fdaa99424b
312ce4d3c26982dec8f4801481c35ddee8bcbb68
1458 F20101109_AAASNS lomascolo_s_Page_046.txt
5465d778a18b42a8367e78d972fc3de9
1c04e7232b6870ed9e46e858864059ccb3e4f269
2224 F20101109_AAASOI lomascolo_s_Page_062.txt
640570347718388124b2482352ee5f0c
1e8be77676b31afacd697ce6c0bb3b6297699c4f
1350 F20101109_AAASNT lomascolo_s_Page_047.txt
b3dc4a3f24769bfd44fc34ccb553269a
9f89673ae8c14fdae8bf2a7ba83abf8375c6db89
1521 F20101109_AAASNU lomascolo_s_Page_048.txt
dc64e38cfa8561a1e69c89a6889b9e5f
d04abfa48aa7477f38176c857a7f5a0a36e40001
F20101109_AAASOJ lomascolo_s_Page_063.txt
a314720f19036952f4ec612e376bf9b0
7d7de9c96b13410b595ac52162ffc4d78beefd0c
1933 F20101109_AAASNV lomascolo_s_Page_049.txt
e1a4d89d6c52a967463aa4a89c4d6ba1
392d3e8bf6827fb209d6bd5b438c495bc31033d4
2226 F20101109_AAASOK lomascolo_s_Page_064.txt
69a3a715e8f7da116b63d1d4df9b8050
6a56cd34357dd269d3b588208e75744e3782c912
1218 F20101109_AAASNW lomascolo_s_Page_050.txt
a9f119713eafe5af7fa31d5bb87ac48f
9d99c502b88570f95fa280952f334f2a29538fd1
2049 F20101109_AAASOL lomascolo_s_Page_065.txt
e372d2ec4874080a35915ed8483861d4
44bb3b0504a3e52dbbfbcb014822a154f40b8399
2245 F20101109_AAASNX lomascolo_s_Page_051.txt
d0c54ea4127f916cf104f13eb9bc759e
63c22f771fdf5c7cc732fb5ff40ed7ac47dc345b
2079 F20101109_AAASPA lomascolo_s_Page_080.txt
59c5ac578770d49bd1450cd70c8f33ea
f93825036538e70b1fa83efcfac13ce942943453
1049 F20101109_AAASOM lomascolo_s_Page_066.txt
b99ec7eac8616644737aa477b13d8ed2
d066b744dd7b70e6914d098864b52b527ebbd40c
2055 F20101109_AAASNY lomascolo_s_Page_052.txt
49976f9b7e35645dfdbeb391db12e0dd
254c19369448af93b408802323bab0e6acb344dd
2110 F20101109_AAASPB lomascolo_s_Page_081.txt
159eb5f2223ec903069c0722c6ae6bcd
4225d78b42a87031734e303906ffba3ad3672fa5
747 F20101109_AAASON lomascolo_s_Page_067.txt
0e54fcc525f98d360c15f0a1b091fd73
c6efb1ed0b09933a4ac5145ce2bffe2dd3f64c25
2011 F20101109_AAASNZ lomascolo_s_Page_053.txt
58124e2c2cfa1328a72b5709d659e155
a318810c1d9135f246f929b903b937b0c52fcafd
2116 F20101109_AAASPC lomascolo_s_Page_082.txt
b1e668276d1053b9f4d2773acbf39a1e
9cbb5a1794b0787d549c6da3af659746f3888097
1245 F20101109_AAASOO lomascolo_s_Page_068.txt
2de2241ddefa332f691d11475607232b
7e2b5b8a4ccdcb006bc417c935a892c5a05e4e3f
2145 F20101109_AAASPD lomascolo_s_Page_083.txt
eb6fb12f7b2c7b9511fa763721082d2c
d47b0a24578aaf03b2bbce4e35a080efacc1508d
408 F20101109_AAASOP lomascolo_s_Page_069.txt
6263a9b37fd210e2eb327fcb7c84cbf5
2848bd7966ff7e97eedc93570536092bfcaf022c
2176 F20101109_AAASPE lomascolo_s_Page_084.txt
4c1dd144e6ad0b68253641e682be79b6
9aacc88a438ddcf59778cfcfe2b36d4e63d9db24
F20101109_AAASOQ lomascolo_s_Page_070.txt
18c5098df75d3f1a39029e3fa43b0ea7
9c5852315e66918cad8d7dabee3894d1d8b6d681
F20101109_AAASPF lomascolo_s_Page_085.txt
c0407be59493cf0d03bb4cb588d685b8
2dd1cfe7063d9d52701ed227f2ab891f3dcc1a39
F20101109_AAASOR lomascolo_s_Page_071.txt
7aa95bf3121d73dd4fb94396232179d9
faecbd94009a46de6e691153ad6f825e89619892
2266 F20101109_AAASPG lomascolo_s_Page_086.txt
bcfd52d0dc678ba1435ac038d6717a66
6954c3d48b9e9bdb96c2de15bda3a50ad57c3194
2269 F20101109_AAASOS lomascolo_s_Page_072.txt
0c633e1f14679b0447086a9acfb57255
d600fc30faf0e8620148f9c1d49d6819002c3faa
2223 F20101109_AAASPH lomascolo_s_Page_087.txt
5183123b67ca1c9f188f951292c2e915
6be64906201bab6a475d4995f8bb61ef66d3d295
2099 F20101109_AAASOT lomascolo_s_Page_073.txt
63e523f0c60ee9709fb73a726a97fa07
32f7ac688d04c8fbfa62d5605c61315b0c8d4102
F20101109_AAASPI lomascolo_s_Page_088.txt
bce279e6c656816bd7943817f7948ec9
88d0ae41232885734c73a136e679a0cabc4e1c32
2243 F20101109_AAASOU lomascolo_s_Page_074.txt
826f20a607177a30fc41319834d9cd40
b4b83041f822b8d580e4d76dcc753d279f801ce7
2039 F20101109_AAASPJ lomascolo_s_Page_089.txt
1f10b5671a875be8907a4f6d37dc19af
880f6f2fbbdcd9baf6343ca27ef75489dd450984







THE ROLE OF FRUGIVORES IN THE EVOLUTION OF FRUIT IN FICUS (MORACEAE)


By

SILVIA BEATRIZ LOMASCOLO

















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2008


































2008 Silvia Beatriz Lomascolo

































To my husband and son, who are the compass that keeps me focused on the true North of my
life.





To my parents, for just giving, from the very beginning.









ACKNOWLEDGMENTS

I first want to thank Doug Levey, my advisor, for his mentoring, patience, respect for my

choices in life and, especially, for believing in me even at times when I doubted of my own

capacity to get to the end. Doug was the best adviser I could have asked for my Ph.D. He

became a role model as a scientist, teacher, and parent. Every time a piece of equipment broke

while I was in the field, Doug rushed to send a replacement piece half a world away. Moreover,

the equipment always came accompanied by many chocolate bars, an important "pat in the back"

for a woman in despair in the field.

The other members of my committee, Rebecca Kimball, Ben Bolker, Mike Miyamoto, and

Emilio Bruna, offered great insight from different perspectives. Drs. David Dilcher and Colin

Chapman were once part of my committee. They had to be replaced for regulatory and logistic

reasons, but were very helpful at the beginning stages of my dissertation. Special thanks go to

Rebecca and Ben, who were extremely helpful with data analysis and always willing to offer

their time to brainstorm for new ideas on how to tackle my (uncooperative) data. Without

Rebecca, the phylogenetic analysis would have been extremely difficult to tame. Without Ben, I

would still be wandering aimlessly through hallways, books, and internet forums. Rebecca and

Ben are true assets to the Department of Zoology.

My time in Papua New Guinea (PNG) was one of the richest, hardest, and most intense

periods of my life. Things would have been much harder and a lot less enjoyable without the

help of Andy Mack, Ross Sinclair, Debra Wright, and the rest of the Wildlife Conservation

Society PNG crew. They unselfishly helped in every aspect of my field work, offering resources

and (at times much needed) emotional support. Moreover, they became long-lasting friends.

M6nica Alberto and Laurent Bertin hosted me for weeks at a time while I anxiously waited

for equipment to arrive, or for rides to the field. I am immensely thankful for their endless









generosity, and for offering me a place to stay, when all else failed. Madang wouldn't have felt

like such a familiar place without them. The management at Jais Aben Resort was also generous

to offer an extremely good deal to a low-budgeted graduate student. The staff were always

extremely nice and made me feel "at home". Big thanks go to Gisele Maisonneuve too, who

shared her house with us a few times, facilitating my work with the coastal fig species around

Madang. Unknowingly, she provided a safer environment for the first month of development of

our child.

Volatile compound work would not have been possible without the help of Peter Teal,

Hans Alborn, and Barbara Dueben, who is no longer with us. They are great scientists and

amazing people who "adopted" me and helped me without asking anything in return. We need

more people like them in science (and everywhere, really). Mark Whitten of the Florida

Museum of Natural History also helped me in the fledgling stages of the volatile work. He

helped me get started and he led me to Peter Teal, two events that made the volatile compound

work a reality.

Many people in the Department of Zoology helped in many aspects of my dissertation.

Brian McNabb with his contagious curiosity and love for PNG helped me financially and

logistically with my first trip to explore that amazing land. He was right: PNG is one of those

places on Earth immensely worth seeing. Thanks to many professors, including Ed Braun, Craig

Osenberg, Marta Wayne, Larry McEdward, Carmine Lanciani, Richard Kiltie, Colette St. Mary,

and others for interesting discussions in their offices or during courses that in one way or another

helped to shape this dissertation as it is today. Karen Pallone was always vigilant in helping me

over administrative hurdles; I don't know what Zoology graduate students would do without

such an extremely efficient person in our Department. Pete Ryschkewitsch helped me prepare









for my fieldwork in many ways and always made my trips to the basement a fun break from my

computer. Ken Albergotti and Jared Wright are the most efficient and knowledgeable computer

people we've had since I came to the Department. I thank them for their many trips to the 6th

floor to save me from my computer illiteracy. Thanks also go to Kent Vliet, Kenetha Johnson,

Tangelyn Mitchell, Lelia Long, and Jimie Norton from Biological Sciences, who helped support

me as a Teaching Assistant.

The many hours spent on the 6th floor led me to meet great people who have come and

gone through the Department, but who will remain an important part of my life. Special thanks

go to Manjula Tiwari, Andreas Heyland, Cristina Dockx, Cedric Worman, Toshi Okuyama,

Martin and Veronika Schaefer, and Charlotte Skov, who are long-lasting friends, and who I'll

take home with me in my fond memories.

The collaborations I developed throughout my Ph.D. ended up being an important part of

this dissertation. I thank George Weiblen for introducing me to Ohu and for his constant help

with PNG figs. I want to especially acknowledge Martin Schaefer and Pablo Speranza for fun

and productive collaborations. I hope there are many more. Thanks also go to Eric Dyreson for

walking me through EPCA, a statistical analysis that seemed written to fill a gap in my

dissertation, and for patiently answering all my questions (some of them, more than once).

Tamaki Yuri pointed me to EPCA and helped me weave through Mesquite and get the analysis

to work. I also want to thank Nina Ronsted for generously providing sequences for all the fig

species I needed, before she had even published her work.

Thanks go to my undergraduate volunteers, who spent long hours entering and juggling

data. I really hope they gained as much as I did from the experience. I thank Eliza Chiang and

Derek Cox for sticking with me until the very end. Jamie Morris, Sara Ouly, and Sabrina Simms









also helped at different stages. Kore Tau and Francesca Dem, from PNG, helped me by taking

full charge of parts of my project, even after I left PNG. They also helped me get to know the

mysterious Melanesian culture a bit better.

Thanks go to the people in Ohu, especially to my field assistants Amos, Frank, and Ismael,

without whom I couldn't have gathered half of my data. To Brus Isua, for sharing his immense

knowledge on the figs of Madang, and to the whole Isua family for hosting me for so long. To

Lucy and the many women of Ohu who have showed me how strong a woman can really be.

My parents and my sister are an inherent part of this work because they have always been

role models for me. They constantly show me with their example how to be a good professional,

a good daughter, a good sibling and, now, a good parent. I still have a lot to learn from them,

now from a bit closer.

Above all, I want to thank my husband for making enjoyable the (at times arid and rough)

journey towards finishing this dissertation. He helped with everything that I have thanked other

people for: fieldwork, equipment, ideas, interesting discussions about my work, and, especially,

emotional support. He taught me the value of persistence and of self-teaching. He was a pillar

that helped me stay on my feet in the toughest moments, and a true partner in this enterprise. He

also gave me my child, which is what I'm most thankful for in life.










TABLE OF CONTENTS

page

A CK N O W LED G M EN T S ................................................................. ........... ............. .....

L IS T O F T A B L E S ..................................................................................................... ........... 1 1

LIST OF FIGURES .................................. .. .... ..... ................. 12

ABSTRAC T ................................................... ............... 14

CHAPTER

1 G EN ER A L IN TR O D U C TIO N ................................................................... .....................16

2 CORRELATED EVOLUTION OF FIG SIZE AND COLOR SUPPORTS THE
DISPERSAL SYNDROME HYPOTHESIS ................. ................................ 20

In tro d u ctio n ................... ...................2...................0..........
M eth o d s ...........................................................................2 3
Study System ................................................................ ..... ..... ......... 23
G general A pproach................................................... 23
Fruit Trait Data ............... ................. ............................... ........ 24
Phylogenetic Data and Analyses ................. .................................26
Sequences .............................................................................................................26
A lig n m e n t ........................................................................................................... 2 7
P hylogenetic analyses ...........................................................28
Fruit Trait Analyses ...................................... ....................29
Phylogenetic comparative analysis .............................. ............... 29
S en sitiv ity an aly sis ............................................................................................ 3 0
N on-phylogenetic com parative analysis ...................................... ........... ....31
Results ........................ ...................... ................. ...............31
P hylogenetic R econstru action ..................................................................................... 3 1
N on-Phylogenetic Com parative Analysis ....................................................... 31
Phylogenetic Com parative Analysis................................ ................... 32
D iscu ssio n ................... ...................3...................3..........

3 SIGNAL CONVERGENCE IN FRUITS: A RESULT OF SELECTION BY
F R U G IV O R E S ? ................................................................................ 5 1

Introdu action ......................... .........................................5 1
M materials an d M eth o d s ..................................................................................................... 54
Fruit Color and Disperser Data ............................ ......................... 54
V isio n M o d el .............................................................................5 5
Data Analysis....................................................... 56
Prediction 1: Color discrimination by birds and primates..................56
Prediction 2: Conspicuousness of the fruits to frugivores ................... ................. 57


8









Prediction 3: Phylogenetic control ........................................ ....................... 57
R e su lts ............................. ........................................................................... .............................5 8
Prediction 1: Color Discrimination by Birds and Primates...........................................58
Prediction 2: Conspicuousness of the Fruits to Frugivores............................................59
Prediction 3: Phylogenetic C control ........................................... ........... ............... 59
D iscu ssio n ................... ...................5...................9..........

4 REVISITING THE DISPERSAL SYNDROME HYPOTHESIS: A PHYLOGENETIC
COMPARATIVE STUDY OF FRUIT TRAITS IN FiCUS (MORACEAE) ........................70

In tro d u ctio n ................... ...................7...................0..........
M e th o d s ..................................................................................7 5
Study Site and G general Specifications ........................................ ........................ 75
Study system: the natural history of figs ................................................75
Q uantification of V isits to Figs ............................................... ................. .......... .77
Fruit Traits ............................. .....................................................................78
Fruit color............................................... 78
F ru it o d o r ............................................................................................................ 8 0
F ru it size, an d so ftn ess ....................................................................................... 8 0
Fruit exposure traits .................................................... .... ................ 81
D ata A nalysis....... ........ .......................... .................. .... ..... ........ .... 8 1
Identification of correlation in fruit traits...................................... .................. 81
Phylogenetic control ........................ .............................................. ............. 82
Volatile compounds in figs with different dispersers........................................84
Phylogenetic reconstruction .................................................. .............................. 84
R e su lts ................... ...................8...................5..........
V isits to F ig T rees ..................................................... .. 85
D isperser PCA and EPCA ................. ......... ......................................... ............... 85
Fruit Trait PCA and EPCA ................. ......... ........................................ ............... 86
Volatile Com pounds ........ .. .................................. .. .. .. .... .... ................. 87
D iscu ssion ............. .. ............. ........................................................ 88

5 WITHIN SPECIES TEST OF THE DISPERSAL SYNDROME HYPOTHESIS: A
STU D Y O F D IO E C IO U S FIG S..................................................................................107

Introdu action ...................................... ............................................................... ............... 107
G general B background and G oals.................... ........................ ............... ...107
Natural History of Fig Pollination and Seed Dispersal ...............................................109
Methods ........................................................................... 111
Study Site and General Specifications ............................... ......................111
Quantification of Visits to Figs .............. .......... ........... ......................... 111
F ru it T raits ......... .... .. ............... .....................................................................1 12
F ru it co lo r ............. ................................................................................... 1 12
Fruit odor ................. ................................114
Fruit size, softness, and peduncle length ......................................................114
D ata Analysis ............................................................................ .............. . ................. 115
Frugivore visits to female versus male trees ............. .....................................115


9









Difference in fruit traits of seed and gall figs.................................... ............... 115
Changes from unripe to ripe stage in monoecious and dioecious figs ...................115
U V reflectan ce ............... ..................................................... ...... .... .... 16
R e su lts ............................ .. .. ........ .......... ............................ ................ 1 16
Frugivore Visits to Female Versus Male Trees.................. ........... ............... 116
Difference in Fruit Traits of Seed and Gall Figs..................................... ..............116
Changes From Unripe to Ripe Stage in Monoecious and Dioecious Figs ................117
U V R efl ectance ............... ..... .. ................................................ .. ........ 118
D iscu ssion ................ ......................................... ...............................118

6 GENERAL CONCLUSION S ................................................................. ............... 131

L IST O F R E F E R E N C E S ......... ..................... ......................................... ...................................136

B IO G R A PH IC A L SK E T C H ......... ................. ............................................. .......................... 152







































10









LIST OF TABLES


Table page

2-1 Results from M addison's Concentrated Changes test ....................................... .......... 43

2-2 Data on fruit diameter, color, animal consumers, and distribution for the 64 species
included in our phylogeny, as found in the literature ................................ ............... 44

3-1 Loadings of the quantum catch variables for primate and bird cone sensitivities for
the "training" DFA.................... ........................ ............ 69

4-1 Loadings of variables for the Disperser PCA and EPCA. ............................................... 105

4-2 Loadings of variables for the Fruit Trait PCA and EPCA......................... ...........105

4-3 Information on abbreviations used in figures, disperser recorded, and source of
information for dispersal mode.................... .................................. 106

5-1 Discriminant Function Analysis of female versus male figs using eight fruit variables .130

5-2 Discriminant Function Analysis of female, monoecious and male figs at different
m maturity stages .......................................................................... 130

5-3 Mean UV reflectance for fig species of different sex and maturity stages..................130









LIST OF FIGURES


Figure page

2-1 Bootstrap support and Bayesian posterior probabilities on the maximum likelihood
phylogeny of 64 F icus species................................................ ............................... 39

2-2 Box and whiskers plot comparing the mean diameter of Type A (red, purple) and
Type B (green, yellow, brown, and orange) fruits .........................................................40

2-3 Maximum likelihood phylogeny of 64 Ficus species.................................. ............... 41

3-1 Calculation of quantum catch values. ........................................ .......................... 67

3-2 Results of the Discriminant Function analysis (DFA) based on primate cone
excitation and bird cone excitation .............................................................................. 68

4-1 Map of Papua New Guinea in reference with a world map...................... ................95

.4-2 Loadings of the first three principal components (PC) on fruit reflectance spectra
from 300 to 750nm, measured at 5nm intervals............................ .................... 96

4-3 M aximum Likelihood tree of 42 fig species. .......................... ......... ................97

4-4 Parsimony reconstruction of dispersal mode .......... ................................ ...............99

4-5 Results of the Disperser PCA and EPCA .......................... ...... ...............101

4-6 Plot of PC 1 vs 2 for the Fruit Trait PCA and EPCA for 42 species of figs................ 102

4-7 Correlation between genetic distance and difference in the score on PC 1...................103

4-8 Discriminant Functions on original and randomized matrix of 73 volatile
com pounds. ...............................................................................104

.5-1 Loadings of the first three principal components (PC) on fruit reflectance spectra
from 300 to 750nm, measured at 5nm intervals................... ........... ............... ......... 123

5-2 Plot of Discriminant Function analysis of female and male figs................................... 124

5-3 Scores on the linear discriminant function (LDF 1) resulting from the discriminant
analysis of female and male figs............... ...... .. ................................. 125

5-4 Maximum Likelihood phylogeny of the species included in the seed-gall fig
co m p ariso n ......... ......................................................................................... 12 6

5-5 Discrimination among figs of different sex and maturity stages using a Discriminant
Function analysis ........................................... ........................... 128









5-6 Pair-wise comparisons among fig groups of different sex and maturity stages ............129









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

THE ROLE OF FRUGIVORES IN THE EVOLUTION OF FRUITS IN FICUS (MORACEAE)

By

Silvia Beatriz Lomascolo

May 2008

Chair: Douglas J. Levey
Major: Zoology

We attempt to resolve a central controversy of plant-animal interactions: whether seed

dispersers are an important factor in shaping the evolution of fruit traits. By focusing on a genus

(Ficus) with extremely variable fruit traits, we could study fruit variation using a phylogeny to

control for the evolutionary non-independence of fig species. We studied six characters that

were hypothesized to most likely influence a frugivore's ability to find and access fruits: color,

odor, size, softness, exposure away from foliage, and location on the tree. This is the first study

on the role of seed dispersers in fruit evolution to combine quantitative measures of color and

odor in a detailed phylogenetic comparative analysis. We took a closer look at fruit color because

it is probably the first trait mediating fruit-frugivore interactions and because it has historically

been mismeasured or ignored in studies of fruit trait evolution. We asked: (1) Can frugivores

discriminate between different colors? (2) Have fruit colors of plants that share dispersers

converged to a common signal? (3) Do fruit characters co-occur in non-random patterns among

species of Ficus? (4) If they do, are the patterns most likely the result of phylogeny or selective

pressure by different types of frugivores?

Results show that color can be used as a signal by frugivores because it can be

discriminated by birds and primates, two of the most common frugivores worldwide. Fruit color









has converged in distantly related species, which can be predicted under selective pressure by

frugivores. However, we cannot discard the role of microbes as a selective force that determines

the amount of the pigment anthocyanin, which has antifungal properties. Color, especially

brightness and contrast against the background, have evolved in correlation with fruit odor, size,

softness, and exposure away from leaves. Results show that the correlation among these traits

cannot be explained solely by phylogeny: a large portion of the variation in fruit traits is

adaptive. The most likely explanation for the correlated evolution of six of the traits studied here

is selection by frugivores. This pattern holds for figs in one geographic region and worldwide,

and that it holds both within and between fig species. Taken together, this evidence argues for

the prevalence of seed dispersers over phylogeny as the main driver of the evolution of fruit traits

in figs.









CHAPTER 1
GENERAL INTRODUCTION

"It is interesting to contemplate an entangled bank, clothed i i jh many plants of many kinds, i ith/

birds singing on the bushes, i/ ith various insectsflitting about, and\ i/th worms crawling through

the damp earth, and to reflect that these elaborately constructed forms, so different from each

other, and dependent on each other in so complex a manner, have all been produced by laws

acting around us."

Darwin, 1859, On The Origin of Species



Species interactions are ubiquitous and important in shaping many ecological and

evolutionary processes, as depicted by the paragraph by Darwin cited above (1859). All

biologists will probably agree on the fact that every system we study is affected in one way or

another by the interaction among and within species. What we might not all agree on, however,

is what types of species interactions are most important (Bertness et al. 1999; Bruno et al. 2003).

Negative interactions have received the most attention by ecologists, which is illustrated by two

facts. First, more studies have been published on negative interactions, including competition,

predation and herbivory, than on positive interactions, such as pollination, seed dispersal and

facilitation (-146,000 and -27,000, respectively, as resulting from a Web of Science search

using the name of the interactions mentioned above together with "ecology" as keywords). And

second, research on negative interactions have had stronger incidence in ecological theory, as

illustrated by some of the most used models in ecology, namely, Lotka and Volterra's

interspecific competition model, Gause's competitive exclusion principle, and Hutchinson's and

MacArthur's work on the niche and species packing (Bruno et al. 2003). Positive interactions

have been largely overlooked, although they have an important role in several ecological









processes, such as facilitation through habitat modification (Bertness and Leonard 1997; Halpem

et al. 2007), and plant reproduction through pollination and seed dispersal. This leads to the

focus of this study, which is the positive interaction between fleshy-fruited plants and the

animals that eat the fruits and disperse the seed, potentially aiding in plant reproduction.

The field of seed dispersal is wrestling with a major contradiction: although ecological

evidence demonstrates that frugivores are important for the reproduction of fruiting plants, any

evolutionary role of seed dispersers in shaping fruit characteristics is difficult to detect. Those

who believe in the evolutionary importance of seed dispersers in shaping fruit evolution point to

the existence of apparently non-random combinations of fruit traits that match the sensory

abilities of the frugivores that feed on them (Janson 1983; Kalko et al. 1996; Korine et al. 2000;

Lord et al. 2002). These studies argue that frugivores must have selected for these suites of

traits, which are referred to as dispersal syndromes. On the opposite side of the issue are those

who argue that other factors, such as phylogeny or physiological constrains, play a more

prevalent role than seed dispersers in shaping fruit traits (Fischer and Chapman 1993; Herrera

1985; 1987; 1992). They explain the match between fruit traits and frugivores' sensory abilities

as preexisting preferences of frugivores for certain types of fruits: the frugivores will feed on the

fruits that are most easy to find and accessible to them. This process is called "ecological fitting"

(Janzen 1980; 1985).

We suggest that much of this controversy stems from the disparate approaches of different

studies, the qualitative measurements of fruit traits (especially color and odor), and the lack of

consistency among studies in the fruits traits that are argued to be important for frugivores to find

and access fruit. Studies that have claimed an association between fruit traits and seed dispersers

have generally not controlled for phylogenetic effects and hence attribute any similarity in fruit









traits among species dispersed by similar frugivores to common selective forces by those

frugivores; they tend to disregard the possibility that a group of species might bear similar fruits

simply because they have shared a common ancestor (Felsenstein 1985; Harvey and Pagel 1991).

Studies that have claimed a lack of association between fruit traits and seed dispersers have

either focused at high taxonomic levels and consequently have ignored great variation at the

species level (Fischer and Chapman 1993); have taken place in temperate regions, and therefore

not incorporated the diversity of dispersers and fruiting plants in tropical regions (Herrera 1987;

1992); or have neglected traits that may be very important in mediating plant-animal interactions,

such as color and odor (Jordano 1995).

We test the Dispersal Syndrome hypothesis, which states that frugivores have selected for

the non-random combination of fruit traits found in nature. We study fruit characteristics at the

species level in the genus Ficus, based on a molecular phylogeny to control for the evolutionary

non-independence of species. We focus on morphological characters of fruit that most likely

affect frugivore fruit choice, and trace their evolution in the genus Ficus (Moraceae) to infer the

evolutionary trajectories and underlying causes of the appearance of modem-day figs. We show

that fruit color and size seem to have evolved in correlation, as predicted by the Dispersal

Syndrome hypothesis (Chapter 2). We then test an important assumption of the Dispersal

Syndrome hypothesis, namely that the fruit traits used to define dispersal syndromes can actually

be perceived and differentiated by the frugivores that the traits are signaling (Chapter 3). We

focus on color, which has been argued to be the most important trait determining fruit choice by

frugivores (Voigt et al. 2004) but has been largely mismeasured or ignored in studies of fruit

evolution, probably due to the difficulties of quantifying it in an objective way. We specifically

test the Signal Convergence hypothesis, which states that if color is a reliable signal to









dispersers, then colors should converge among species dispersed by similar frugivores. We

show that frugivores can distinguish the color of fruits they disperse from the color of fruits

dispersed by other frugivores. We conclude that color can be used as a signal to dispersers. We

also show that colors that attract different frugivores have evolved independently in different

genera and families of fruiting plants.

The most important finding of our work is that when seven of the traits hypothesized to be

most important for frugivores to find and access available fruits are analyzed together, they

suggest that dispersers are an important selective force in determining the diversity of fruiting

plants, even when phylogenetic effects are accounted for (Chapter 4). Lastly, a within-species

study of dioecious figs comparing individuals with viable seeds and often consumed by

frugivores (female plants, called seed figs), with individuals without viable seeds and not

consumed by frugivores (male plants, called gall figs), shows the same pattern just described for

interspecific comparisons (Chapter 5).

Overall, the results of my study show that seed dispersers have had an important role in

shaping the evolution of fruit traits. Other alternative hypotheses that were used to explain the

diversity of specific traits such as color, or the correlated evolution of pairs of traits, such as

color and size, were not tested, but are discussed. Although some alternative hypotheses cannot

be discarded to explain the diversity of specific traits, such as the antimicrobial role of pigments

in ripe fruits, only the Dispersal Syndrome hypothesis can account for the correlated evolution of

the six fruit traits that we examined.









CHAPTER 2
CORRELATED EVOLUTION OF FIG SIZE AND COLOR SUPPORTS THE DISPERSAL
SYNDROME HYPOTHESIS

Introduction

Although seed dispersers clearly influence the fitness and spatial distribution of fruiting

plants (Harms et al. 2000; Howe and Miriti 2000; Marquez et al. 2004; Terborgh et al. 2002),

their effect on the evolution of fruit traits remains controversial (Bolmgren and Eriksson 2005;

Fischer and Chapman 1993; Herrera 1992; Jordano 1995; Lord et al. 2002; Mack 1993).

Evolutionary ecologists have proposed that differences in the behavior and ecology of seed

dispersers have led to corresponding differences in fruit species favored by those dispersers

(Gautier-Hion et al. 1985; Janson 1983; Kalko et al. 1996, Schaeffer et al. 2006 ). For example,

fruit-eating birds are diurnal, have excellent color vision, a poor sense of smell, and are limited

by gape width in the size of fruit they can swallow (Jordano 2000; Levey 1987; Rey et al. 1997;

Wheelwright and Janson 1985). Therefore, fruiting plants dispersed by birds should tend to

stand out against the green foliage, and be odorless and small. Colors that stand out against

leaves include red, black, and purple, all of which have either high chromatic or achromatic

contrast against foliage (Burns and Dalen 2002; Janson 1983; Schmidt et al. 2004b). Fruit-eating

mammals, on the other hand, are often nocturnal, have teeth (i.e., are not gape limited), and

exhibit a keen sense of smell. Thus, mammal-dispersed fruits do not need to stand out against

green foliage (e.g., green, yellow, brown, or pale orange, Janson 1983), and are more likely to be

relatively large and odoriferous.

These sets of fruit traits define what have been termed "bird syndrome" and "mammal

syndrome" fruits, respectively (Herrera 2002a; McKey 1975; van der Pijl 1969). At least

superficially, fruit traits such as size and color seem to be associated non-randomly, as expected

(Gautier-Hion et al. 1985; Janson 1983; Kalko et al. 1996; Korine et al. 2000; Shanahan and









Compton 2001; Wheelwright and Janson 1985). However, two opposing hypotheses explain co-

occurrence of fruit size and color. One we termed the Phylogenetic Inertia hypothesis, which

states that co-occurrence of fruit traits is phylogenetically determined; the size and color of a

given fruit species is most strongly determined by the size and color of ancestral species. Under

this hypothesis, frugivores have pre-existing preferences for particular types of fruits and

disperse seeds in fruits that match those preferences -- an apparently common process called

"ecological fitting" (Agosta 2006; Janzen 1980; 1985; Strauss et al. 2006) that does not drive the

evolution of fruit traits. Alternatively, the Dispersal Syndrome hypothesis states that the co-

occurrence of fruit traits is determined by selective pressure from sets of seed dispersers that

differ in behavior and ecology, as described above.

Tests of these hypotheses with phylogenetic comparative methods have commonly rejected

the Dispersal Syndrome hypothesis (Fischer and Chapman 1993; Herrera 1987; 1989; 1992; but

see Jordano 1995). However, this result may reflect methodological challenges more than

biology. In particular, these studies have pooled data from species of the same genus (Fischer

and Chapman 1993) or they have focused on temperate species (Herrera 1987; 1989; 1992),

thereby overlooking immense intrageneric variation, especially in tropical taxa. This variation

remains largely unexplored in phylogenetic comparative analyses.

Several lines of evidence from studies with a broad geographic scope support the Dispersal

Syndrome hypothesis. These studies controlled for phylogenetic inertia by comparing fruit traits

among plants of the same genus (Lord 2004; Lord et al. 2002), family (Mack 1993) or species

(Hampe and Bairlein 2000). For example, fruit traits in New Zealand are consistent with

hypothesized selective pressures by avian and reptilian frugivores, the only frugivores

historically present in New Zealand (Lord 2004; Lord et al. 2002). Also, tropical fruits in the









New World tend to be smaller than those of related plant species in the Old World (Mack 1993),

which is consistent with the observation that New World mammalian frugivores are smaller than

in the Old World (Fleming et al. 1987; Mack 1993). Lastly, within-species differences in fruit

traits match the characteristics of the different disperser guilds found in two different regions of

Europe (Hampe and Bairlein 2000). Despite these seemingly non-random patterns, such studies

are limited by low replication (typically, one region is compared to another) and lack of an

explicit phylogenetic framework.

We provide a phylogenetic test of the Dispersal Syndrome hypothesis at the species level

in the genus Ficus (Moraceae) by testing two predictions that arise under this hypothesis: (1)

Ficus species with ripe fruit that are red, purple, or black will produce smaller fruit than species

with green, yellow, brown, and orange fruits, and these combinations of color and size will occur

independently in different clades; and (2) interspecific variation in fruit size of red, purple, or

black figs will be less than interspecific variation in fruit size of green, yellow, brown, or orange

figs. The rationale for the second prediction is that birds are more restricted than mammals in

the size of fruits they can ingest because mammals can manipulate fruits with their forelimbs and

can eat them piecemeal using their teeth (Alcantara and Rey 2003). Although disperser guilds

such as birds and mammals include animals that vary considerably within each group (e.g.,

tanagers to cassowaries, and bats to elephants) (Lord et al. 2002) we are testing traits that should

be affected in a similar way by most members of each guild. We believe it is a fair

generalization to say that most birds have more acute color vision than most mammals

(especially nocturnal mammals) (Bowmaker et al. 1997; Bowmaker and Hunt 2006; Das et al.

1999) and that, overall, mammals are larger than birds and can therefore feed on larger fruits

than birds. Even smaller mammals, such as bats, rodents, and many marsupials may be able to









eat larger fruits than birds of comparable size because they have teeth and can eat the fruits

piecemeal, and they have forelimbs and can handle the fruits in more ways than birds can with

their beak. Through this conservative test of the Dispersal Syndrome hypothesis we found that

color and size of figs seem to evolve in correlation. The patterns of correlated evolution were

identified despite the conservative nature of our test given such broad disperser guilds.

Methods

Study System

Ficus is an ideal group for testing the Dispersal Syndrome hypothesis because its fruits

vary immensely among its ca. 750 species. Figs can be yellow, brown, red, purple, or orange,

and may be as small as 4mm in diameter to as big as 80mm (Dominy et al. 2003; Laman and

Weiblen 1998, S. Lomascolo pers. obs.). Ficus has an unusually wide distribution, as they are

found in almost all tropical forests worldwide, and their fruits are consumed by a taxonomically

diverse array of vertebrates, including 1274 bird and mammal species in 523 genera, in addition

to reptiles and fishes (Shanahan et al. 2001b). Furthermore, molecular data are available to build

a phylogeny and fruit size and color are described in the literature for many species. We used 64

Ficus species for which genetic and fruit size and color information were available in the

literature.

General Approach

We first built a molecular phylogeny of the genus Ficus, which is necessary to tease apart

the Dispersal Syndrome hypothesis from the Phylogenetic Inertia hypothesis. To ensure that the

reconstructed phylogeny was not affected by the method chosen to build it, we constructed

phylogenies using maximum likelihood (ML), Bayesian, and parsimony searches.









Correlated evolution of fruit color and size was tested using Maddison's (1990)

Concentrated Changes test and Pagel's (1994) Likelihood Ratio test on the parsimony and ML

trees.

Because phylogenetic comparative methods give priority to phylogeny over ecology when

explaining shared characters (Westoby et al. 1995) and we do not know the degree of

phylogenetic niche conservatism (Harvey and Pagel 1991) in figs, we also tested for correlation

of fruit color and size using raw data (hereafter non-phylogenetic comparative analyses). Non-

phylogenetic comparative analyses were performed on the set of species that is included in our

phylogeny and on all Ficus species for which fruit data were available in the literature. This

allowed us to test whether the limited set of species in the phylogeny was representative of a

much larger set of species. The second prediction was tested without incorporating phylogeny

because all phylogenetic comparative tests that we are aware of require data for each species in

the phylogeny to reconstruct ancestral character state. Because we are testing variation in fruit

size between the two types of fruits, to do a phylogenetic analysis we would need data on

intraspecific fruit size variation for each species in our phylogeny, which was not available in the

literature. In most cases, only a mean diameter was reported without any measure of within

species variation. Thus, we calculated interspecific variation for mammal-colored and bird-

colored fruits, and compared them using an F-test. This approach tests the results of selective

pressure on fruit size by birds and mammals at the community level.

Fruit Trait Data

Fruit trait data were obtained primarily from published sources (Croat 1978; Dominy et al.

2003; Laman and Weiblen 1998), and secondarily from a Ficus expert (Weiblen, pers. comm.)

and field data collected by S. Lomascolo. All data are from ripe fruits; unripe fruits were

excluded. Because the phylogenetic comparative tests we used are designed for dichotomous









variables, we transformed color data into two categories, following Janson's (1983) classification

of Type A (red and purple) and Type B (green, yellow, brown, and orange) fruits. Likewise, size

data were transformed into two categories, based on the mode diameter (10mm) of species

included in our study: "small" (< 10 mm) and "large" (> 10 mm). The 10 mm cutoff point also

coincides with the mode for all Ficus species for which diameter data were available in the

literature (Croat 1978; Dominy et al. 2003, Weiblen pers. comm.; Laman and Weiblen 1998),

with the mean fruit diameter and bird gape width reported by Wheelright (1985), and with the

largest diameter of fruits typically consumed by some of the most common fruit-eating passerine

birds in Costa Rica (Levey 1987).

Occasionally, different sources provided contradictory information about fruit color of a

given species. For example, Ficus asperifolia is reported to produce yellow or red fruits

(Dominy et al. 2003), which means it could be classified either as Type A or B. Because we

cannot discern whether these situations result from a polymorphism, phenotypic plasticity, or

simply a reporting error, we performed separate analyses for each possible assignment of color

type. We reconstructed character change in two ways. For the most conservative reconstruction

we chose the option that would minimize support for the Dispersal Syndrome hypothesis. For

example, if a species had either yellow or red figs and was classified in the "small" category, we

recorded it as Type B, so that the combination of color and size would not match the prediction

of Type A color and small size. The second reconstruction was the non-conservative

reconstruction where, for each ambiguous classification, we chose the color that matched the

hypothesis of correlated evolution between the two traits. By comparing results of these two

approaches, we are able to assess the sensitivity of studies like ours to sometimes arbitrary









character state reconstruction a commonly encountered limitation of data gleaned from the

literature.

Phylogenetic Data and Analyses

Although three phylogenies have been published for Ficus, the comparative tests that we

used require a phylogeny that has branch lengths and includes only the species for which fruit

data were available (Likelihood Ratio test, Pagel 1994), or else a phylogeny that is completely

resolved (Concentrated Changes test, Maddison 1990). Thus, we constructed our own

phylogeny.

Sequences: Internal transcribed spacer (ITS) sequences for 58 Ficus species were

obtained from GenBank. For six additional species (F. congensis, F. exasperata, F. mucuso, F.

natalensis, F. asperifolia2, and F. vallis-choudae), we sequenced the ITS region ourselves.

Voucher specimens for these species were deposited at the Florida Museum of Natural History

Herbarium under numbers FLAS 218158 FLAS 218164, and the resulting ITS sequences were

submitted to Genbank (Accession nos. EF520706- EF520713). Ficus ITS1 through ITS2

sequences were amplified using primers ITS1 through 5 (White et al. 1990). All PCR

amplifications were carried out in 25 pL reactions containing 0.4 units ofNEB Taq polymerase

(New England Biolabs, Beverly, MA) 1.5mM MgC12, 0.4 pM of each primer and 0. ImM of

each dNTP in the manufacturer's buffer. Amplification was carried out in a Biometra T3

Thermoblock with the same program for all primer combinations. The program consisted of an

initial denaturing step of 95C for 2 min; five cycles of 95C for 1 min, 53C for 1 min, and 72C

for 2 min with a decrease of 1C per cycle in the annealing temperature; 35 cycles with an

annealing temperature of 480C and a final extension step of 72C for 12min. PCR products were

cleaned with Wizard SV Gel and PCR Clean-up System (Promega, Maddison, WI) and diluted

to approximately Ing/pL for sequencing.









We sequenced all regions in both directions on a CEQ 8000 capillary sequencer

(Beckman-Coulter, Fullerton, CA) using 14 reaction volumes with the addition of 80mM Tris and

2 mM MgC12 (pH 9) to complete the volume of a full reaction. To edit the sequences manually

we used Sequencher TM (V4.1.4, Genecodes, AnnArbor, MI).

Alignment: We prealigned the sequences with the Clustal-W (Thompson et al. 1994)

algorithm included in BioEdit (V 5.0.6, Hall 1997) and manually adjusted the resulting

alignment. The nrDNA region includes ITS1 and ITS2, separated by the 5.8S ribosomal RNA

gene. Because the 5.8S region differed from ITS1 and 2 in that it contained few variable sites

and appeared to evolve under a different model of evolution (see below), the 5.8S region was

excluded from analyses. Base positions that could not be unambiguously aligned, particularly

due to insertions or deletions (indels) of nucleotides, were recorded as missing data. There were

15 indels two or more base pairs in length in regions where alignment was unambiguous. For

parsimony analyses only, these were considered single characters, coded as present/absent, and

included in analyses. Otherwise, all single base pairs deletions and multiple base pairs deletions

that could not be unambiguously aligned were treated as missing data. A low complexity region

ranging from 20 to 40 base pairs in ITS 2 containing a poly-G was not readily alignable;

however, preliminary parsimony searches conducted without this region did not result in

sufficiently resolved trees. To make use of the information contained in this region, clades that

had appeared consistently in our preliminary analyses and were congruent with the phylogenies

shown in Weiblen (2000), Jousellin et al. (2003), and Ronsted et al (2005) were aligned

separately for this region. This approach is similar to the stepwise alignment strategy described

by Hsiao et al. (1999). Although this strategy resulted in unambiguous within-clade alignments,

alignment across clades remained difficult for this region. Consequently, in order









to avoid spuriously reinforcing support for the clades that had been aligned separately, all

informative positions within this region were manually de-aligned between pre-aligned groups

by introducing gaps when the individual alignments were pooled in a single matrix.

Phylogenetic analyses: We used the clade formed by F. maxima, F. insipida and F.

yoponensis as a functional outgroup based on the fact that it appeared as sister to the rest of the

genus in all our preliminary analyses conducted using Morus alba as an outgroup. This topology

was fully consistent with Weiblen's (2000), Jousselin et al.'s (2003), and Ronsted et al's (2005)

analyses, which were performed with different data sets.

The maximum parsimony criterion searches for the phylogenetic tree that minimizes the

number of evolutionary changes that must have occurred. Using PAUP* 4.0b10 (Swofford

2003), a preliminary heuristic search was performed with tree bisection-reconnection (TBR)

branch swapping and the minimum tree length recorded. The final searches were conducted for

5000 sequence addition replicates (SAR) by saving only trees that were shorter or equal to the

minimum tree found in the preliminary search. A bootstrap analysis was also conducted for

1000 bootstrap replicates with 100 SAR within each replicate using TBR.

The maximum likelihood criterion searches for the phylogenetic tree that best fits the

data, given a particular model of evolution. The model of evolution that best describes our data

was determined by the hierarchical likelihood ratio test in MODELTEST 3.06 (Posada and

Crandall 1998). We initially examined the best model for ITS1, ITS2, and the 5.8S region

independently. The best models selected for ITS1 and ITS2 were similar and more complex than

the model selected for the 5.8S region. Thus, the 5.8S region was excluded from analyses and

the best model for the combined ITS regions was used for the likelihood analysis. We conducted









a heuristic search in PAUP* with 10 random sequence additions using the model (TN93) and

parameters recommended by MODELTEST 3.06 (Posada and Crandall 1998).

We assessed support on the phylogeny using two different approaches. Bayesian analyses

were conducted using MrBayes 3.1.2 (Huelsenbeck and Ronquist 2001; Ronquist and

Huelsenbeck 2003). We ran our chains for 1 x 107 generations, sampling every 1000

generations. The first 2 x 106 generations were discarded as "burn-in", and a consensus of the

remaining trees was used to determine the posterior probabilities at each node. We also ran 200

bootstrap replicates using maximum likelihood using GARLI 0.951 (Zwickl 2006). We

considered nodes strongly supported when posterior probabilities were greater than 95% or when

bootstrap values were greater than 70% (Alfaro et al. 2003; Hillis and Bull 1993).

Fruit Trait Analyses

Phylogenetic comparative analysis: To test whether the pattern of association between

fruit color and size constitutes evidence of correlated evolutionary change in both traits

(prediction #1) we performed Pagel's Likelihood Ratio test (1994), as implemented in Discrete

v. 4.0 (Pagel 1994; 1999), which compares the goodness of fit of a model of independent

evolution to one of correlated evolution. This test assumes that changes in character state are

more probable on long branches than on short branches and takes into account branch length for

character state reconstruction. We also performed Maddison's (1990) Concentrated Changes test

using MacClade 4.0 (Maddison and Maddison 2000), which tests for non-random association of

traits by determining if an evolutionary change in one trait is correlated with change in a second

trait. The test reconstructs character changes using a parsimony criterion and assumes a

dependent and an independent variable. We determined whether color or size was more likely to

be independent using Discrete (Pagel 1994; 1997). For example, if changes in color precede

changes in size, rates of change in size should differ depending upon color type while rates of









color change should not differ by size category. By constraining rates of change among joint

character states (e.g., small-Type A to small-Type B set equal to large-Type A to large-Type B),

the likelihood of color preceding size could be compared to that of size preceding color. Since

this approach does not test nested models, the different models were compared using the AIC

(Akaike's Information Criterion) (Burnham and Anderson 2003). Since results of this test were

not conclusive, and since mode of evolution of fruit traits is not well-understood, we performed

the Concentrated Changes test (Maddison 1990) in two different ways: first using color as the

independent variable, and then using size as the independent variable.

The Concentrated Changes test cannot deal with unresolved nodes, so we resolved the

polytomies in our phylogeny by using the resolution of our majority rule tree, Weiblen's (2000),

Jousellin et al's (2003), or Ronsted et al's (2005) reconstructions.

Sensitivity analysis: The Concentrated Changes test requires specification of several

different algorithms, although defining the algorithms may be arbitrary if the mode of trait

evolution is not well known. To address this, we performed a sensitivity analysis using all

available algorithm combinations to explore if algorithm choice had a major influence on the test

of prediction #1 and, if so, which algorithm seemed to most influence the results. The

algorithms that we varied were (1) ACCTRAN or DELTRAN, which are options for resolving

ambiguous branches and that produce character changes earlier (at the basal branches) or later (at

the terminal branches) in phylogenetic history, respectively. ACCTRAN increases the number

of losses, whereas DELTRAN increases the number of gains; and (2) MINSTATE or

MAXSTATE, which are options for reconstructing character states at ambiguous nodes.

MINSTATE uses the lowest trait value, whereas MAXSTATE uses the largest trait value.









As part of the sensitivity analysis, we ran the analyses using the conservative and non-

conservative trait reconstruction, using color and size as the independent variables.



Non-phylogenetic comparative analysis: To test if type A fruits had smaller diameter

than type B fruits (prediction #1), we used a Kolmogorov-Smirnov non-parametric test because

our data were not normal. To test whether type B fruits have greater variation in fruit size than

type A fruits (prediction #2), we performed an F-test (Zar 1999). Both tests were performed for

all species ofFicus found in the literature and for the subset of species included in the

phylogeny.

Results

Phylogenetic Reconstruction

The phylogenies generated using maximum likelihood, Bayesian, and parsimony

analyses were similar to each other, as well as to published phylogenies (Jousselin et al. 2003;

Ronsted et al. 2005; Weiblen 2000). The majority of nodes were well-supported in our analyses

(Fig. 2-1). With taxa shared among our phylogenies and published phylogenies, all the major

clades with high bootstrap values were essentially identical. Conflicts occurred only in poorly

supported groups and those conflicts were minor. Slight differences in topology between the ML

and parsimony tree did not affect the results of our phylogenetic comparative analyses (see

below).

Non-Phylogenetic Comparative Analysis

When we analyzed data only for those species ofFicus included in our phylogeny, we

omitted a species (F. ruginerva) whose fruits were reported as larger than 10mm, but no exact

diameter was given. For this limited set of species, type A fruits were smaller than type B fruits,

in agreement with prediction #1 ( x A= 14.86 mm, nA = 36; x B= 27.41, nB = 27; Kolmogorov-









Smirnov Z = 1.818; p = 0.003). For the entire set of species, the difference was in the same

direction but not as large ( x A= 15.46 mm, nA = 56; x B = 18.13 mm, nB = 72; Kolmogorov-

Smirnov Z = 1.359; p = 0.050). The variance in diameter of type B fruits was significantly larger

than the variance of type A fruits for the subset of species in our phylogeny, in agreement with

prediction #2 (o2A = 79.63; a2B = 281.69; F35,26 = 3.54; p < 0.001). However, this difference in

variance was not as large when all species were included in the analysis (C2A = 106.68; C2B =

116.36; F71,55 = 1.09; One tailed test p > 0.25) (Fig. 2-2).

Phylogenetic Comparative Analysis

For two reasons, we emphasize the analyses performed on the ML tree (Fig. 2-3). First,

Pagel's Likelihood Ratio test (1994) requires branch lengths, which were only available on the

ML trees. Second, the parsimony searches came up with over 5000 most parsimonious trees and

performing the comparative analyses in all 5000 of them was unwieldy. A preliminary

comparative analysis performed on a randomly chosen parsimony tree showed no difference with

the comparative analyses performed on the ML tree.

With Pagel's Likelihood Ratio test (Pagel 1994), the model of correlated evolution of fruit

size and color fits the data significantly better than the model of independent evolution, which

prediction #1, regardless of the coding of ambiguous fruit colors (Conservative reconstruction:

likelihood ratio = 4.889; p = 0.027; 1000 simulations. Non-conservative reconstruction:

likelihood ratio = 4.990; p = 0.027; 1026 simulations).

To determine the independent variable for Maddison's Concentrated Changes test

(Maddison 1990), assuming color precedes size is more likely (In likelihood = -80.31) than

assuming size precedes color (In likelihood = -81.68). However, while color may be more likely

to precede size, examination of the Akaike weights cannot exclude size preceding color.

Assuming color is independent is more consistent with the observation that color tends to change









at basal branches while size tends to change at more terminal branches (Fig. 2-3). When color is

considered as the independent variable, the results for Maddison's Concentrated Changes test

(Maddison 1990) indicate that fruit size changes from large to small on branches with brightly

colored fruits more often than expected by chance. The p-values from all the tests performed

with the different algorithm combinations and trait reconstructions ranged from 0.002 to 0.03

(Table 2-1). When size is considered the independent variable, however, results vary greatly

according to the algorithm combination and character state reconstruction used and support for

prediction #1 is mixed (0.04 < p < 0.54; Table 2-1).



Discussion

Overall, our results support the Dispersal Syndrome hypothesis. Mean diameter is smaller

for Type A fruits than for Type B fruits, supporting prediction #1 of the Dispersal Syndrome

hypothesis. This pattern holds when using raw data for the subset of species included in our

phylogeny and when using raw data for all the species of figs found in the literature. It also

holds when examined with phylogenetic comparative methods: fruit size tends to change from

large to small on species with type A colors (e.g., red and purple) more often than expected by

chance. This result was consistent across all combinations of algorithms for the Concentrated

Changes test (Maddison 1990) when color was considered the independent variable. The

tendency of color and size to evolve in correlation, at least in the subset of 64 species of Ficus

included in our analyses, supports the Dispersal Syndrome hypothesis rather than the

Phylogenetic Inertia hypothesis, suggesting that seed dispersers have been important for the

evolution of at least some fruit traits in Ficus.

The Dispersal Syndrome hypothesis was supported to a lesser extent by our test of

prediction #2. In the subset of species used in our phylogeny, Type A fruits are significantly less









variable among species in diameter than Type B fruits, as predicted. However, in analyzing

variation of this type we could not control for phylogeny and the significance of the difference in

variation between Type A and B fruits disappeared when all species were included.

Returning to prediction #1, an alternative hypothesis for the observed non-random

combinations of fig color and size is physiological. Because the production of reproductive

parts, including fruits, can be costly to plants (Jurik 1983; Reekie and Bazzaz 1987), and because

green fruits can photosynthesize even after they are ripe, green fruits may have lower costs of

production than fruits of other colors (Cipollini and Levey 1991). Given that large fruits must be

more costly to produce than small fruits, green photosynthetic fruits should be favored over fruits

of different colors. However, a positive carbon balance in green fruits has been found only at

high light levels and the balance becomes negative at low levels (Cipollini and Levey 1991).

Because light levels may change considerably during a plant's lifetime, it seems unlikely that

extra photosynthetic tissue of large fruits would be selectively advantageous, unless the plant

grew exclusively in brightly lit habitats and ripe fruits were fully exposed to the sun. Also,

considering the large surface area provided by leaves, it seems unlikely that photosynthetic ripe

fruits would benefit the parent plant sufficiently to increase its fitness relative to other plants

with fruits of different color and similar size.

The sensitivity analysis for Maddison's Concentrated Changes test (Maddison 1990)

helped us identify which source of variation has the stronger impact on our results, given

uncertainty in both the phylogenetic reconstruction and the mode of fruit trait evolution. The

largest impact in results was generated by switching color and size as independent variables in

the Concentrated Changes test (Maddison 1990, Table 1). When using size as the independent

variable, the strength of the correlation varied widely, especially between the conservative and









non-conservative reconstructions. We think that the choice of the independent variable affects

the results in our study because when size is the independent variable, few changes of color

occur on branches with small-sized fruits, primarily since small-sized fruits appear in more

terminal branches. When so few changes occur (a small sample size for the purpose of the

Concentrated Changes test), calculating the probability of the observed number of gains and

losses seems to be very sensitive to small changes in the number of gains and losses. This

sensitivity highlights the importance of understanding the mode of fruit trait evolution (i.e.,

which trait evolves first) before one can draw conclusions about the importance of seed

dispersers for the evolution of fruit traits. Even though the test used to decide which variable

should be used as independent was fairly inconclusive, observing the pattern of gains and losses

as reconstructed using a parsimony criterion gives us more confidence that color can be

considered the independent variable.

An important caveat of our methodology is that our categorization of colors was based on

the perception of colors by humans. Birds, however, see the world differently than humans; they

can detect UV (Bennett et al. 1994; Bennett and Thery 2007; Bowmaker et al. 1997; Das et al.

1999; Hart 2001; Schaefer et al. 2006). Likewise, cone type and color perception can vary

widely in mammals (Bowmaker and Hunt 2006; Peichl 2005). Unfortunately, data that

incorporate the entire range of wavelengths reflected by fruits, and color perception by different

frugivores (Schaefer et al. 2006) are not available in the literature for the fig species included in

our study. The technology necessary to quantify reflectance in the field has only recently

become available (Altshuler 2001; Altshuler 2003).

Another limitation of this study is the oversimplification of fruit traits and disperser guilds

used to define the syndromes. Fruit size and color were the only fruit traits readily available in









the literature and easily comparable among studies and, hence, we were unable to include traits

such as odor, location on the tree, texture, crop size, among other traits that may be potentially

important for frugivores to find fruits (Dominy 2004; Dominy et al. 2001). Because we included

only color and size, a broad definition of disperser guilds was needed, as it was not feasible to

define disperser guilds more narrowly on the basis of their potential selective pressure on just

fruit color and size. Despite potentially important differences among different taxa of

mammalian frugivores such as bats, rodents, or primates, we consider it safe to assume that,

because most frugivorous mammals can manipulate fruits using their forelimbs and teeth, and

because they are generally larger than birds, mammals should be able to consume larger fruits

than can birds. As for color, despite minor differences in avian color receptors (Bowmaker et al.

1997; Das et al. 1999), birds have a much more complex visual system than mammals and, in

general, a higher variety of cone cells, which are responsible for color vision (Bennett and Thery

2007; Bowmaker et al. 1997; Bowmaker and Hunt 2006). Hence, we consider it a safe

assumption that birds can exert a stronger selective pressure on fruit color than mammals can,

which are often nocturnal and have less color receptors in their retina. Considering only fruit

color and size and broadly defined disperser guilds makes our analyses a conservative test of the

Dispersal Syndrome hypothesis. In short, the patterns of correlated evolution that we found were

identified despite this generalization, not because of it.

Another drawback in our definition of disperser guilds is that we did not incorporate the

potential characteristics of the paleofauna that figs have interacted with historically. Modern day

figs appear to date back at least 60 million years (Ronsted et al. 2005) and they have almost

certainly interacted with frugivores that differ from today's frugivores, at least in body size (e.g.,

extinct New World megafauna, Janzen and Martin 1982). However, we do not have any reason









to believe that historic birds and mammals exerted selective pressures that were significantly

different from those of current birds and mammals, and any discussion on this would be highly

speculative. Even with this limitation, we believe our conclusions remain solid because we

predicted how current birds and mammals should affect the evolution of fruits, and our results

generally supported our predictions.

Our results can be used to generate readily testable hypotheses about which frugivores

consume a given type of fig. For example, it can be predicted that the figs of Ficusphaeocyse

and F. benjamin, bearing type A colors and classified in the "small" size category, are dispersed

most commonly and efficiently by birds, whereas F. botryocarpa, F. copiosa, and F.

dammaropsis, all of which bear type B colors and are in the "large" size category, are dispersed

most commonly and efficiently by mammals. Qualitative data from a published source

(Shanahan et al. 2001b) and from fieldwork (S. Lomascolo, unpublished data) on fruit removal

by animals for most fig species in our study show that most figs are consumed by a wide

diversity of frugivores, including birds and mammals (Table 2-2). At first glance, this may seem

to contradict the importance of any particular group of frugivore in shaping the evolution of fruit

traits. However, the data in Table 2, although highly informative, should be taken with caution

because they do not provide information that would allow one to assess the potential of a given

frugivore as a selective force. Such information would include fruit removal rate, gut treatment,

deposition of seeds, and overall frugivore effectiveness (sensu Schupp 1993). That said, some

general patterns arise from Table 2-2. For example, large fruits often have contrasting colors, a

combination that seemingly falsifies the Dispersal Syndrome hypothesis. These species tend to

occur in the Old World tropics, where larger birds such as hombills and cassowaries occur.

These may be "exceptions that prove the rule." We predict that contrasting colored fruits should









be smaller in places where such large frugivorous birds are less common, i.e., in the New World.

However, more data from the New World tropics is needed to test this idea. We encourage the

collection and application of quantitative data that address disperser effectiveness to determine

whether our findings match what is found in nature.

Taken together, our results suggest that differences among taxa of seed dispersers have

influenced the evolution of fruit traits. In particular, we conclude that the correlated evolution of

fruit size and color support the Dispersal Syndrome hypothesis.











t 70% ML bootstrap F. mucuso
F sycomorus
A 0.95 Bayesian F. vallis-choudae
posterior probability F bolryoides
F sur
F. racemosa
S F. robusta
F. nodosa
F. auriculata
F uncinata
F. pungens
a F bemaysii
FR hispidioides
SA F. septica
F bo ryocarpa
A F. condensa
F. variegata
F, mollior
F dammaropsis
F, semiveslla
F adenosperma
F. itoana
F microdictya
conocephalifolia
F. phaeosyce
F. copiosa
F. wassa
11A F. tinctoria
F. virgata
F. parietalis
F. asperifolia 1
A F. asperifolia 2
A exasperata
SF trachypison
*A F. aurata
A F. grossularoides
a *a F. punctata
F ruginerva
F deltoidea
A F. prasinicarpa
F. virens
A F. superba
F- religiosa
r A F. consociata
F. xylophylla
F. subgelderi
F. microcarpa
S Fbenjamina
A F. binnendykii
F. elastica
SF. pellucidopunctata
F. macrophylla
F rubiginosa
F. xylosycia
F. glumosa
F natalensis
"A I IFF. congensis
F. lutea
F pertusa
IA F schumacheri
F nymphaeifolia
F. insipida
BA F. maxima
F. yoponensis


Figure 2-1. Bootstrap support and Bayesian posterior probabilities on the maximum likelihood
phylogeny of 64 Ficus species used for analyses. Nodes with strong bootstrap support
in maximum likelihood analyses (greater than or equal to 70%) are indicated with a
circle; nodes with strong support using Bayesian posterior probabilities (greater than
or equal to 0.95) are indicated with a triangle



























A specseE AlI qRe2 FRIlqeny Ptop"W
ViWA Iype EB peeS q*eM
TFppeA pe B



Figure 2-2. Box and whiskers plot comparing the mean diameter of Type A (red, purple) and
Type B (green, yellow, brown, and orange) fruits (Janson 1983) for the set of Ficus
species in our phylogeny (nA = 36, nB = 27) and for all Ficus species found in the
literature (nA = 56, nB = 72). The boxes' upper and lower sides indicate the first and
third quartiles, and the horizontal lines inside the boxes marks the median. The
vertical lines show the largest and smallest observations that fall within 1.5 times the
box size from the nearest quartile









Figure 2-3. Maximum likelihood phylogeny of 64 Ficus species. Color is mapped onto the
phylogeny and was done using a parsimony criterion in MacClade (the only criterion
available in this software), and is one of the reconstructions tested with the
Concentrated Changes test. Black branches represent type A fruits, grey branches
represent type B fruits, and white branches represent ambiguous reconstruction. At
the tip of the branches, character state for diameter is mapped for each fig species.
Forty two species show the combinations predicted by the Dispersal Syndrome
hypothesis (type A colors and small size, or type B colors and large size), whereas 22
species show a mismatching combination. The ancestral state for fruit diameter was
reconstructed as large. Changes of diameter from large to small (called gains, and
represented by white dashes) and small to large (called losses, and represented by
grey dashes) are marked on the branches where they are hypothesized to have
occurred according to the parsimony criterion used for the reconstruction. For this
particular reconstruction, out of a total of 12 gains and 5 losses of small size, eight
and four, respectively, occurred on branches with type A fruits. For branch support
value using maximum likelihood bootstrap and Bayesian posterior probabilities see
Fig. 1.












Color reconstruction
Non-contrasting
- llContrasting
1 Equivocal
Diameter
Large
- Small


Diameter
F. mucuso
3 F sycomorus
F. vallis-choudae
F. botryoides
F. sur
F. racemose
F. robusta
F. nodosa
F auriculata
S F. uncinata
SF. pungens
F. bernaysi,
F. hispidioides
F. septica
F. botryocarpa
F condensa
m m F. variegata
F mollior
F. dammaropsis
F. semivestita
I F. adenosperma
F. itoana
S F. microdictya
SF conocephalitolia
SF. phaeosyce
F. copiosa
SF. wassa
S F. tinctoria
^ F. virgata
F. parietalis
M F. asperifolia 1
F. asperifolia 2
F. exasperata
F. trachypison
S F. aurata
SF. grossularoides
F. punctata
F. ruginerva
F deltoidea
m F. prasinicarpa
F. virens
F. superba
F. religiosa
F. consociata
F. xylophylla
F. subgelderi
F. E microcarpa
SF. benjamin
F. binnendykii
F. elastic
F. pellucidopunctata
S F. macrophylla
F rubiginosa
F. xvlosycia
F g umosa
F natalensis
SF. congensis
F lutea
F pertusa
I l IF. schumacheri
F nymphaeifola
F. insipida
F. maxima
F. yoponensis










Table 2-1. Results from Maddison's Concentrated Changes tests showing all combinations of algorithms and character state
reconstructions performed on the maximum likelihood tree using color and size as the independent variable (I and II,
respectively).
Independent Non-conservative reconstruction Conservative reconstruction
variable


Resolving
option to
count gains
and losses




p-value
Resolving
option to
count gains
and losses


deltran


Color


Diameter


p-value


Total: 17 gains, 0
losses
Branches: 12 gains, 0
losses
minstate maxstate
0.012 0.005



Total:12 gains, 2
losses
Branches: 5 gains, 0
losses
minstate maxstate
0.131 0.085


acctran

Total:12 gains, 5
losses
Branches: 9 gains, 4
losses
minstate maxstate
0.011 0.007



Total:7 gains, 7 losses

Branches:3 gains, 1
loss
minstate maxstate
0.049 0.067


deltran

Total:17 gains, 0
losses
Branches:13 gains, 0
losses
minstate maxstate
0.002 0.002



Total:9 gains, 7 losses

Branches:2 gains, 3
losses
minstate maxstate
0.539 0.544


acctran

Total:12 gains, 5
losses
Branches: 9 gains, 4
losses
minstate maxstate
0.034 0.015



Total:9 gains, 7 losses

Branches:3 gains, 2
losses
minstate maxstate
0.197 0.204


Total: total number of gains and losses using each resolving option (DELTRAN, which delays changes, or ACCTRAN, which
accelerates changes) Branches: number of gains and losses on the branches with the character state of interest. When considering color
as the independent variable, Branches is the total gains and losses of "small size" that occur on branches with Type A (brightly colored)
fruits. When considering size as the independent variable Branches is the total gains and losses of Type A colors on branches with
small-sized fruits. MINSTATE and MAXSTATE are options for reconstructing character states at ambiguous nodes. MINSTATE uses
the lowest trait value, whereas MAXSTATE uses the largest trait value.









Table 2-2. Data on fruit diameter, color, animal consumers, and distribution for the 64 species included in our phylogeny, as found in
the literature. Superscripts next to the Ficus species indicate the source from which data on fruit size and color were
obtained. The column "diameter code" indicates whether the species was classified as small or large based on the 10mm
cutoff point described in the text. The column "Fruit type" indicates whether the species fell in the Type A or B color class
based on Janson's (1983) classification, as explained in the text. Data on animal consumers were obtained mainly from
Shanahan et al. (2001b), and secondarily from fieldwork by S. Lomascolo. The superscript next to the distribution
indicates the source from which this information was obtained.
Ficus species Diameter Color Size Fruit Avian consumers Mammalian Other Distribution
(mm) code Type Consumersh consumers


F. adenosperma
F. asperifolia

F. aurata
F. auriculata


F. benjamin


Green a
Yellow
or reda
Orange
Greeng


8a Purplea


Small
Small

Large
Large


Small


B No data found
A or B Pigeons


Passeriforms


Galliforms,
Pisciforms,
A hornbills,
Cuculiforms,
parrots, pigeons,
Passeriforms


No data found
OW primates


Rodents
OW bats


NW and OW bats, OW
primates, civet, rodents


Cassowary,
Redf Large B hornbills parrots,
pigeon,
Passeriforms
Pisciforms,
Reda Small B hornbills, pigeons,
passeriforms


Marsupials, OW bats


OW primates, rodents


New
Guineab


Green large B Marsupials


Fish


New
Guineab
Africaa

Borneoa
Chinab


Borneoa


F. bernaysii



F. binnendykii


New
Guineab


Peninsular
Malaysiaa


F. botryocarpa 35a









Table 2-2. Continued
F. botryoides
F. condensa
F. congensis
F.
conocephalifolia


F. consociata


29e
20a
25g

28.5g


12 a


Rede
Green a
Redg


large
Large
Large


Redg Large


Reda Large


No data found

No data found


No data found
OW bats
No data found


Madagascarb
Borneoa
Africaa
New
Guineab


Peninsular
Malaysiaa


OW bats


Pisciforms,
hornbills,
A Cuculiforms,
pigeons,
Passeriforms


Tree shrews, Asiatic
Black bear, OW
primates, rodents


F. copiosa


F. dammaropsis
F. deltoidea

F. elastica

F. exasperata


F. glumosa

F. grossularoides

F. hispidioides
F. insipida


29a


60.5g
9a


>10g

9a


9a


Yellow Large B Cassowary,
hornbills, Fig parrot


Greeng
Reda


Large
Small


Honey eater


jreeng Large B Pigeons,
Passeriforms
Reda Small A Hornbills, Turaco

Pisciforms,


Reda Small A pigeons,
Passeriforms
Redd Small A Passerifroms


Green
Green


Large
Large


New
Guineab


Marsupials, OW bats


Marsupials, OW bats
OW primates

OW primates

OW bats, OW primates,
Artiodactyla

OW primates


OW bats
NW primates, NW bats,
Procyonids, tapirs


Turtle,
fish


New
Guineab
Borneoa
New
Guineaf

Africaa


Africaa

Borneob
New
Guineab
Venezuelab


New
Guineab


Green Large B OW bats


(


F. itoana 34a









Table 2-2. Continued
F. lutea

F. macrophylla


15a Yellow
or reda
17.5g Purpleg


F. maxima


Large A or B Hornbills, pigeons,
Passeriforms
Large A Cuculiforms,
parrots, pigeons,
Passeriforms


Green Large B


F. microcarpa


<10g Purpleg


Small


Pisciforms,
B hornbills,
Cuculiforms,
parrots, pigeons,
Passeriforms


OW bats, OW primates,
elephants
OW bats


NW bats, NW primates

Marsupials, OW bats,
OW primates


F. microdictya
F. mollior
F. mucuso


F. natalensis



F. nodosa

F. nymphaeifolia


F. parietalis


Reda
Green
Orangeg


Small
Large
Large


Redg Large


No data found
No data found
Pigeons


Pisciforms,
hornbills, Turacos,
pigeons,
A Passeriforms


Green Large B Hornbills


22a Green or
reda


Large


A orB


Pisciforms,
Reda Large A hornbills,
Passeriforms


No data found
No data found
OW bats, OW primates,
elephants

OW bats, OW primates,
Artiodactyla, elephants


Marsupials, OW bats

NW bats


OW primates, rodents


Africaa


Australiac


Brazilb
New
Guineab


New
Guineab
Australiac
Africaa


Africaa


New
Guineab
South
Americaa


Bruneic









Table 2-2. Continued

F. pellucidopunctata


9a Purplea Small


Pisciforms,
A hornbills,
Trogoniforms,
pigeons,
Passeriforms


Yellow
or reda


F. pertusa





F. phaeosyce


F. prasinicarpa
F. punctata

F. pungens



F. racemosa


Large A or B


Guans, Pisciforms,
Coraciforms,
Trogoniforms,
parrots,
Passeriforms


9g Redg Small A Cassowary,
hornbills, pigeons,
Passeriforms


Reda
Reda


Small
Large


Marsupials, NW bats,
NW primates,
Procyonids



Marsupials, OW bats


Parrots


7g Redg Small A Hornbills, parrots,
pigeons,
Passeriforms
Galliforms,
25e Green or Large A or B Pisciforms,
rede hornbills,
Cuculiforms,
parrots, pigeons,
Passeriforms


OW primates, rodents

Marsupials, OW bats



OW bats, OW primates,
bears, civet, Artiodactyla,
rodents


Venezuelab


New
Guineab


Solomon
Islandsb
Borneob
New
Guineab





Australiab


Borneob









Table 2-2. Continued


F. religiosa


F. robusta
F. rubiginosa


F. ruginerva

F. schumacheri

F. semivestita

F. septica


F. subgelderi


Purpleg Small


75g Greeng Large
12.59 Greeng Large


37.5d Oranged


Large


Pisciforms,
hornbills,
A Cuculiforms,
pigeons,
Passeriforms


B No data found
B Cuculiforms,
parrots, pigeons,
Passeriforms
B Hornbills


5a Greena Small B No data found


Greenf


Large


Coraciiform,
30a Greena Large B pigeons,
Passeriform
Pisciforms,
12d Oranged Large A hornbills, parrots,
pigeons,
Passeriforms


OW bats, OW primates,
Artiodactyla


No data found
OW bats


OW primates, rodents

No data found

OW bats

Marsupials, OW bats,
OW primates, rodents

Tree shrew, OW
primates, rodents


Galliforms,
Pisciforms,
Purpleg Large A hornbills,
Cuculiforms,
parrots, pigeons,
Passeriforms


OW bats, OW primates,
civet, rodents


Lebanon'


New
Guineab
Australiac


Borneob
French
Guyanac
New
Guineab
New
Guineab


F. superba


Bruneic


Borneob










Table 2-2. Continued


Pisciforms,
19a Reda Small B hornbills, Turacos,
pigeons,
Passeriforms
Pisciforms,
26a Yellowa Large B hornbills, Turacos,
pidgeons,
Passeriforms

14a Reda Large A Hornbills, parrots,
pigeons,
Passeriforms

7a Reda Small A Hornbills, parrots,
pigeons,
Passeriforms


Reda
Yellow
or reda


Large
Large


Pigeons


Cassowary,
27a Green or Large A or B hornbill, parrots,
reda Passeriforms
Cassowary,
8a Yellowa Small A Pisciforms,
hornbills,
Cuculiforms,
parrots, pigeons,
Passeriforms,
Cuculiform,
8a Orangea Small A pigeons,
Passeriforms


F. sur


Africaa


OW bats, OW primates,
Artiodactyla


OW bats, OW primates,
civet, hyrax

Marsupials, OW bats,
OW
primates, rodents


Marsupials, OW bats


Rodents
OW bats, OW primates


Marsupials, OW bast,
OW primates, rodents

OW bats, OW primates,
carnivores, rodents




OW primates


F. sycomorus


F. tinctoria


F. trachypison


F. uncinata
F. vallis-choudae


F. variegata


F. virens


F. virgata


Africaa


New
Guineab


New
Guineab


Bruneic
Africaa


Australiab

New
Guineab


New
Guineab









Table 2-2. Continued


Cassowary,
10a Reda Small A hornbill, parrots,
pigeons,
Passeriforms
25a Orangea Large A Pigeons,
Passeriforms


Reda Large


Marsupials, OW bats


A Pigeons,
Passeriforms


F. yoponensis 18a Greena Large B Passeriforms NW bats, NW primates, Panamaa
Procyonids, Artiodactyla,
rodents
Superscripts denote the source from which the information was obtained: a. Dominy et al. (2003); b. Weiblen (2000); c. Jousellin et al.
(2003); d. Laman and Weiblen (1998); e. Missouri Botanical Garden website, at http://mobot.mobot.org/W3T/Search/vast.html; f. George
S Weiblen, pers. comm.; g. Pers. obs.; h. Shanahan (2000)
0 NW stands of New World, and OW stands for Old World.


F. wassa


F. xylophylla


F. xylosycia


New
Guineab


Bruneic


New
Guineab









CHAPTER 3
SIGNAL CONVERGENCE IN FRUITS: A RESULT OF SELECTION BY FRUGIVORES?

Introduction

To overcome the constraints imposed by immobility, many plants depend on animals for

dispersal of pollen and seeds. Animal pollinators and dispersers cue on different plant signals to

find flowers and ripe fruits. Yet the great diversity of plant signals is poorly understood -- it has

been variously attributed to chance, to environmental factors, and to the sensory ecology of

mutualistic animals (Ben-Tal and King 1997; Gumbert et al. 1999; Johnson and Steiner 2000).

The latter, in particular, constitutes a long-debated central question in the field of plant-animal

interactions: Do animal mutualists shape the evolution of plants, causing plant signals to

converge into pollination and seed dispersal syndromes (Janson 1983; Johnson and Steiner 2000;

Ridley 1930; van der Pijl 1961; 1969)? Syndromes are sets of flower or fruit traits hypothesized

to evolve under the selective pressure of pollinators and seed dispersers, respectively. Because

many unrelated plants are visited by similar mutualists, and because visitation rate is commonly

density-dependent, an adaptive explanation for signal convergence is that plants sharing the same

suites of visitors might benefit from converging on a similar phenotype to attract mutualists.

This "Signal Convergence hypothesis" is the main focus of this study.

Nevertheless, the general convergence of fruit or flower traits to syndromes that are

reliably associated with (i.e., presumably signal to) specific animal vectors has been questioned.

Detractors of the syndrome hypothesis argue that plants tend to benefit from different types of

animal mutualists and, conversely, plant-visiting animals do not tend to specialize on only one

type of plant (Herrera 1985; Johnson and Steiner 2000; Waser et al. 1996). This argument may

be particularly relevant to the evolution of seed dispersal syndromes because plants show a lower

degree of specialization towards seed dispersers than they do towards pollinators (Bluthgen et al.









2007). From an adaptive viewpoint, the low degree of specialization in seed dispersal will result

in disparate selective pressures on fruit appearance and will thus reduce the likelihood of signal

convergence. Some authors claim that apparently non-random combinations of fruit traits are

evidence of syndromes (Gautier-Hion et al. 1985; Janson 1983; Kalko et al. 1996; Knight and

Siegfried 1983; Korine et al. 2000; Willson et al. 1989). However, when those combinations of

traits have been scrutinized in phylogenetically controlled studies, the syndromes are generally

not supported (Fischer and Chapman 1993; Herrera 1987). This difference in approach and

opinion has generated much controversy.

We argue that one reason studies addressing the role of frugivores in shaping fruit

evolution have reached disparate conclusions is that fruit color, one of the most important fruit

traits reflecting differential food selection by distinct frugivore assemblages (Voigt et al. 2004),

has been measured inappropriately. Color as a visual signal might be particularly important in

defining syndromes because signals represent the first step in the interactions between plants and

animals. Natural selection should favor signals (fruit colors) that elicit a response in the receiver

(the frugivore) that increases the fitness of the sender (the plant) (Endler 1992). From this, we

can derive that the efficiency of a signal not only depends on the signal's form (i.e., the spectral

reflectance of the fruit), but also on the receiver's perception (i.e., the frugivore's neural circuitry

to interpret the color signal) (Endler 1992). Nonetheless, evolutionary studies on fruits and

frugivores have typically reported color qualitatively and according to human visual perception

(Fischer and Chapman 1993; Gautier-Hion et al. 1985; Janson 1983; Lord et al. 2002), or they

have ignored color completely (Jordano 1995), probably because of the difficulty of quantifying

color.









Another important property of signals is their conspicuousness, which is quantified as the

signal-to-noise ratio, and in the case of visual signals is directly correlated with the contrast

between the color of the object and that of the surrounding environment. Natural selection

should maximize conspicuousness to the receiver. In the case of fruits, the contrast between the

color of the fruit and the foliage background should be greatest to the disperser to which the

signal is targeted. Although fruit conspicuousness has been shown to be important for fruit

detection (Burs and Dalen 2002; Regan et al. 2001; Schmidt et al. 2004a; Sumner and Mollon

2000a), the contrast between the color of the fruit and that of its background has never been

incorporated in a study of dispersal syndromes.

Our study focuses on the role of fruit color as a signal to two of the most common

frugivorous taxa, primates and birds. Under the Signal Convergence hypothesis, we test whether

fruit color can be used as a signal to seed dispersers. We predict that:

1. Primates and birds should be able to distinguish visually between fruits that are

primarily dispersed by one of them and those that are primarily dispersed by

other group;

2. Fruit colors should be more conspicuous to the disperser (primates or birds) to

which they are signaling.

3. The color of fruits dispersed mainly by primates should converge to colors that

are different from those dispersed mainly by birds.

One assumption of our study is that primates and birds can have differential effects on

plant fitness according to their differences in seed treatment and recruitment requirements of

different plants. Although support is scant, some published studies suggest that this is a fair

assumption (Clark et al. 2001; Poulsen et al. 2002).









To test prediction 1 of the Signal Convergence hypothesis we use vision models to analyze

fruit signals according to the color vision capabilities of the main types of frugivores, primates

and birds (Osorio and Vorobyev 1996; Regan et al. 2001; Vorobyev and Osorio 1998). These

vision models approximate visual capability of specific frugivores based on their cone sensitivity

(see Materials and Methods). To test prediction 2, we define conspicuousness as the contrast in

color between the fruit and the foliage background (Endler 1992) and use vision models to

approximate conspicuousness of fruits to each frugivore type. To test prediction 3, we compare

the fruit colors of bird and primate dispersed plants belonging to the same family and genus.

To test these predictions, we use the largest spectral data set of fruit colors analyzed to this

date. For 406 plant species with known fruit consumers, we demonstrate that primates are

physiologically capable of discriminating between fruits dispersed by them and those dispersed

by birds based solely on color. Likewise, we show that birds can do the same and that this

pattern transcends geographic areas, and persists even when phylogenetic effects are controlled.

We also discuss why fruit colors were not always more conspicuous to the targeted disperser.

Materials and Methods

Fruit Color and Disperser Data

Data on reflectance spectra of fruits and fruit consumers for each plant species were

obtained mainly from the literature and partly from our fieldwork. A total of 406 species were

included, of which 177 were primate-dispersed species from Guyana (111), Uganda (46), and

Venezuela (20), and 229 were bird-dispersed species from Florida (29), Europe (99), Venezuela

(31), and other unspecified regions (70) (Regan et al. 2001; Schaefer et al. 2007; Schmidt et al.

2004a; Sumner and Mollon 2000b). In all cases, color was quantified using a portable

spectrometer and a special light source that reflects equally in all parts of the spectrum. Fruit

reflectance was obtained as the proportion of reflectance by a white standard. The spectrometer,









light, and white standard varied among studies but were consistent within each study. Although

the use of different equipment adds an uncontrolled source of variation in the data, we

considered that, if the predicted patterns were found in reflectance of primate versus bird

dispersed fruits, this would be despite the lack of consistence in color measuring devices, and not

because of it.

Spectra were averaged in 5 nm intervals from 400-700 nm. Unfortunately, data on fruit

reflectance under 400 nm (the UV range) were not available for primate-dispersed fruits, so we

left this part of the visual range of birds out of our analyses. Our results are therefore

conservative because differences between primate and bird perception of fruit color are likely

more pronounced if UV vision is accounted for.

Vision Model

Because color is interpreted by the brain based on the input of light to different types of

cones cells with different sensitivity to specific wavelengths (Bowmaker and Hunt 2006; Regan

et al. 2001), we calculated the light input to the different cone types for each frugivore group.

Our vision model combines the reflectance spectrum of each fruit with the color vision ability for

each frugivore (represented by the peak sensitivity of the cone cells for each frugivore) and

yields three variables per species of fruit, corresponding to the input to the brain by the short,

medium, and long wavelength cones of the frugivore (Fig. 3-1). For primates, we used cone peak

sensitivities corresponding to humans, (440, 540, and 570 nm), which are very similar to those of

some other trichromatic primates (e.g., Alouatta spp.) (Bowmaker and Hunt 2006; Regan et al.

2001). For birds, we used cone peak sensitivities for the Blue Tit (Cyanistes caeruleus, 372,

456, 544, and 609 nm). We calculated the relative cone excitation values, called quantum

catches (Fig. 3-1), which represent the basic sensory input for color vision (Kelber et al. 2003).

Although cone peak sensitivity varies within each of these groups of frugivores (Bowmaker et al.









1997; Bowmaker and Hunt 2006), these differences are small relative to differences between

birds and primates.

All cone excitation values were transformed to the natural logarithm to achieve normality

and were analyzed with Discriminant Function analysis (DFA) to test whether bird-dispersed

fruits and primate-dispersed fruits were significantly discriminated using the quantum catch of

bird and primate cones. Discriminant Function Analysis finds the combination of the variables

that best discriminates between the groups, and allows us to identify each variable's contribution

to the discrimination. These variable combinations are called Linear Discriminant Functions

(LDF). If birds can visually discriminate between fruits commonly dispersed by them and those

commonly dispersed by primates, a significant discrimination should be obtained using the

variables yielded by our vision model for birds and, likewise, for primates.

Data Analysis

Prediction 1: Color discrimination by birds and primates: To evaluate the accuracy of

the DFA and its predictive power, we split our data set of 406 species into two parts and derived

the LDFs with one part (species from Guyana, Florida, and unspecified geographic regions),

called the "training" DFA, and tested the predictive value of LDFs on the other part of our data, a

completely independent data set (species from Uganda, Europe, and Venezuela), called the

validation DFA. This is a powerful test of the validity of the discrimination between primate and

bird fruits by primate and bird vision, because the pattern of discrimination will remain

significant only if it is consistent across very different biogeographical regions, such as the

Paleotropics, Paleo-temperate areas, and the Neotropics. We also tested the predictive value of

the linear discriminant functions by classifying the cases as "bird" or "primate" fruits based on

their score on the LDF derived with the "training" DFA. The higher the percentage of species

that were well classified (known bird-dispersed species correctly classified as a bird-dispersed









species), the higher the predictive value is of the linear discriminant functions. This was done

for the "training" and the validation data sets.

Prediction 2: Conspicuousness of the fruits to frugivores: To test whether primate-

dispersed fruits were more contrasting against the foliage than bird-dispersed fruits from a

primate's visual perspective, we first calculated the chromatic contrasts of 86 primate fruits and

188 bird fruits against the foliage background as the Euclidean distance between the color of a

fruit and the color of the structure against which it is viewed by the frugivore. Spectra were

normalized to the same brightness by dividing the reflectance at each wavelength by the total

reflectance for each species. This provides contrasts due to the color of the fruits and not to their

brightness (Schmidt et al. 2004a), which can vary greatly with light availability. The resulting

contrasts for primate and bird fruits were entered into the primate-vision model and the results

were compared using a Mann-Whitney U test. To ensure that the differences between fruit types

are actually due to differences in reflectance by the fruit itself and not by the background, which

is beyond the scope of the Signal Convergence hypothesis, we also calculated fruit contrasts

against the mean of the background for the 256 species included in the analysis and compared

the results with another Mann-Whitney U test.

Prediction 3: Phylogenetic control: Closely related species are not independent data

points because they share a common ancestor. We controlled for such phylogenetic effects by

doing a pair wise comparison of the mean score on the linear discriminant function for bird and

primate dispersed species within each family of plants. To do such a comparison, we ran another

DFA with all the species together and then only for the 22 families that had both bird and

primate dispersed species. We calculated the mean score for bird and primate dispersed species

per family. Based on the results of the training and validation data sets, plants dispersed by birds









scored consistently higher on the LDF separating the groups, and therefore we tested whether the

scores of bird-dispersed species were significantly higher than primate-dispersed species of the

same families using a one-tailed Wilcoxon Signed Rank test. We performed the same test on the

mean scores within 12 genera for which we had bird and primate dispersed species. If the

difference between bird and primate-dispersed species within each family or genus was

significant, we concluded that the pattern of discrimination observed was not driven by

phylogeny.

Results

Prediction 1: Color Discrimination by Birds and Primates

The "training" DFA resulted in significant discrimination between bird- and primate-

dispersed fruits when using bird cone sensitivity peaks (Wilk's X = 0.680, numerator d.f. = 3,

denominator d.f. = 205, p < 0.00001) and primate cone sensitivity peaks (Wilk's X= 0.649,

numerator d.f. = 3, denominator d.f. = 205, p < 0.00001) (Fig. 3-2). Classification accuracy was

fairly high, with 82% of 111 primate-dispersed species classified correctly and 76% of 98 bird-

dispersed fruits classified correctly. Given the high negative loading of the medium wavelength

variable (representing green light that primarily stimulates the medium-wavelength cone) and the

high positive loading of the short wavelength variable (representing blue light) (Table 3-1), our

results suggest that primate-dispersed fruits tend to be greener than bird-dispersed fruits, which

tend to be more blue or purple.

The validation DFA also showed significant discrimination between bird and primate-

dispersed fruits for bird (Wilk's X = 0.638, numerator d.f. = 3, denominator d.f. = 192, p <

0.00001) and primate cone sensitivities (Wilk's X = 0.616, numerator d.f. = 3, denominator d.f. =

192, p < 0.00001) (Fig. 3-2). Classification accuracy was also high, with 80% of 66 primate-









dispersed species being classified correctly and 82% of 130 bird-dispersed fruits classified

correctly.

Prediction 2: Conspicuousness of the Fruits to Frugivores

Contrary to the predictions under the Signal Convergence hypothesis, primate-dispersed

fruits are less conspicuous (i.e., contrast less against the foliage) to primates than are bird-

dispersed fruits (Mann-Whitney U = 4,273, p < 0.001, n = 256). Unfortunately, we could not

test conspicuousness of bird and primate-dispersed fruits to birds because UV reflectance is

required to obtain an accurate measure of conspicuousness to birds. We are confident that the

difference in contrasts is due to differences in fruit color and not background color because the

contrasts calculated with the mean reflectance of the background for all 256 species included was

also significantly different from a primate's perspective (Primates: Mann-Whitney U = 8,584, p

< 0.001, n = 256).

Prediction 3: Phylogenetic Control

Our phylogenetic test comparing the mean scores on the linear discriminant functions of

bird and primate-dispersed fruits within each family using a Wilcoxon Signed Rank test showed

that bird-dispersed fruits tended to score higher on the LDF than did primate-dispersed fruits,

although the difference was not are large as that found at the genus level (Family level: one-

tailed test, V = 77, p = 0.056 for both bird cone peak sensitivity, and V = 77, p = 0.056 for both

primate cone peak sensitivity; Genus level: one-tailed test, V= 10, p = 0.021 for bird cone peak

sensitivity, and V = 7, p = 0.005 for primate cone peak sensitivity).

Discussion

Our results generally support the first prediction of the Signal Convergence hypothesis,

which states that frugivores should be able to discriminate between the color of fruits dispersed

by them and those of fruits dispersed by other frugivores. Based on the model representing the









visual capabilities of primates and birds, we conclude that these frugivores can differentiate

between fruits typically dispersed by primates and those by birds. This is important because

color could not be considered a signal to frugivores if frugivores were not capable of

distinguishing colors.

Even though the discrimination is significant, a considerable amount of overlap exists

between bird- and primate-dispersed fruits. This result is expected because, although a frugivore

may more often consume fruits that are more conspicuous to them than those that are more

cryptic, they will not feed exclusively on fruits with conspicuous colors. Frugivores can also find

fruits based on senses other than vision, and they can feed on fruits found by chance from a

nearby fruiting tree. Plants with exclusive seed dispersers are exceptional, and neither the Signal

Convergence, nor the Dispersal Syndrome hypotheses imply exclusiveness of frugivores. If

syndromes are an evolutionary response to seed dispersers, they should result from the response

of a plant species to selection by those seed dispersers that improve its fitness, even though many

other not-so-efficient frugivores may still feed on the same plant.

If we could include other cues used by these frugivores to find fruits in our model, we

would likely obtain sharper discrimination. The fact that, as shown by our study, the color of

some bird-dispersed fruits fall well within the range of a primate's visual perception, does not

mean that bird fruits are necessarily conspicuous to primates and that they can easily find them.

Primates may need a combination of cues to make the fruits conspicuous and fully available to

them, such as color, odor, shape, texture, and/or location on the tree (Dominy 2004; Dominy et

al. 2001). Our study is limited to testing only one trait (color), which we use to define dispersal

syndromes and, although we believe color is one of the most important cue used by diurnal

frugivores to find fruits, we acknowledge that it is certainly not the only one.









It is surprising that the quantum catch on the long wavelength cone, representing red light

vision, was unimportant in discriminating between bird- and primate-dispersed fruits. Red fruits

are commonly associated with bird dispersal more than with primate dispersal (Janson 1983;

Wheelwright and Janson 1985; Willson and Whelan 1990), but so are black fruits. Red and

black reflectance spectra differ dramatically in the amount of long wavelength light reflected and

hence, bird dispersed fruits result in highly variable quantum catch for the cone responsible of

long wavelength absorbance. Hence, reflectance at long wavelengths is not a consistent

discriminant because the within group variance is too high. What red and black fruits do have in

common is their high contrast against the foliage, which may be why birds are often associated

with fruits of those colors (Schmidt et al. 2004a).

Our second prediction, that the fruits dispersed by a frugivore should be more conspicuous

to that frugivore than fruits not dispersed by them, was not supported by our data. Primate fruits

seem to be less conspicuous to primates than bird fruits. We propose several alternative

explanations for the lack of congruence of the results for primates and birds, and the lack of

complete support to our hypothesis, and we divide them into "adaptive" and "non-adaptive"

explanation from the perspective of plant-frugivore communication. The adaptive explanations

are consistent with the Signal Convergence hypothesis. First, it is possible that, while birds have

such well developed color vision and may be primarily visual foragers, primates use other

signals besides color, such as odor, size, and texture of fruits to assess their edibility (Dominy

2004; Laska et al. 2007). This may therefore derive in a stronger selective pressure on

conspicuousness by birds than by primates. If this is the case, we predict that a similar test on

conspicuousness to birds, bird-dispersed fruits should be more conspicuous to birds than

primate-dispersed fruits. Also, by including fruits from varied geographical regions from the









New and Old World, we included a wide range of primates that have very different color vision

capabilities (Bowmaker and Hunt 2006; Regan et al. 2001; Surridge et al. 2003). In order to

answer the specific question of whether the fruit/background color combination is optimized for

dispersal by primates, we should differentiate between dichromatic and trichromatic primates,

and we should also take into account the polymorphisms found among platyrrhine (i.e., New

World) monkeys. If primates exert a strong selective pressure to maximize the signal-to-noise

ratio in the fruits they eat, the signal will most likely be seen when studying specific groups of

primates with well defined and similar color vision capabilities. This may not be as crucial when

testing conspicuousness to birds because birds have a much higher within-guild similarity in

cone sensitivities among very different taxa (Hart 2001), and therefore their potential selective

pressure on fruit color may be similar among different bird species. We mentioned in the

Introduction that variation in color vision among disperser types is greater than that within

disperser type. However, within group variability, even if small, may affect the results of the test

for fruit contrasts more than that of color discrimination.

Second, it is also possible that our measure of conspicuousness is simplistic. Sumner and

Mollon (2000b) argue that, at equal contrast between background and fruit color, larger fruits are

more conspicuous to primates simply because they are larger. That is, large-fruited plants may

not be under as strong selective pressure as small-fruited plants to maximize the difference

between background and fruit color. Under this hypothesis we predict that, among the primate-

dispersed fruits, species with lower score on LDF 1 will be larger than species with higher LDF 1

score, because LDF 1 is negatively associated with reflectance of green light, which is less

conspicuous against the foliage. This is similar to what Sumner and Mollen (2000b) showed

when they correlated fruit size and reflectance at longer wavelength. However, his negative









correlation can also be explained by the effect of smaller, more visual frugivores, such as birds

feeding on smaller fruits, as predicted by the Dispersal Syndrome hypothesis, rather than the sole

effect of selection on conspicuousness by one type of frugivore (Lomascolo et al. 2008; Sumner

and Mollon 2000b). Whatever the ultimate explanation for this correlation, we suggest that

conspicuousness to frugivores should be measured as a function of fruit size, in addition to the

contrast against the background.

The non-adaptive explanations consider roles of color that are independent of plant-

frugivore communication. Willson and Whelan (1990) identify 13 non-exclusive hypotheses that

may explain fruit color patterns, of which six are completely unrelated to seed dispersers and

include defense against pathogens or seed predators, as well as physiological, or phylogenetic

constraints. Although we will not attempt a thorough discussion of all Willson and Whelan's

(1990) hypotheses, we want to generally address some that seem particularly relevant to our

study. First, the anthocyanins responsible for red and black colors in fruits have been shown to

reduce fungal growth (Schaefer and Schaefer 2007) and, thus, attraction of seed dispersers by

anthocyanines may be a by-product of the antimicrobial properties of these pigments. Red is

also inconspicuous to certain insects that eat fruit but do not disperse seeds, so this pigment may

have evolved to hide fruits from predators -- not to make them more conspicuous to seed

dispersers (Willson and Whelan 1990). In this case, color should be considered an exaptation

(Gould and Vrba 1982) instead of an adaptation to endozoochory.

Second, fruit pigmentation has been hypothesized to decrease some physiological

constraints, such as thermoregulation. Because dark colors absorb more heat than light colors, it

has been hypothesized that dark fruits have higher metabolic rates and therefore can reach

maturity faster, reducing exposure to fruit predators. However, dark colors would also cool









faster as the radiation emitted by a body must equal the radiation absorbed (Kirchhoff s Law of

Thermal Radiation), so dark fruits should be at a disadvantage in many tropical rain forests

where clouds cover the sun light long periods of time. Yet dark-colored fruits are extremely

common in these cloudy environments (Wheelwright and Janson 1985; Willson et al. 1989;

Willson and Whelan 1990). The thermoregulation hypothesis does not explain the advantage of

dark-colored fruits in the shaded understory either, where they are also very common (pers. obs.)

Lastly, as Wheelwright and Janson (1985) point out, dark pigments are only deposited at the end

of the ripening period and, hence, may contribute very little to the development of the fruit.

Third, fruit colors have also been hypothesized to be related to the plant's metabolic

constraints. Because the production of reproductive parts, including fruits, can be costly to

plants (Jurik 1983; Reekie and Bazzaz 1987), and because green fruits can photosynthesize even

after they are ripe, green fruits may have lower costs of production than fruits of other colors

(Cipollini and Levey 1991). This is especially relevant to large-fruited species: given that large

fruits must be more costly to produce than small fruits, green photosynthetic fruits should be

favored over fruits of different colors. However, a positive carbon balance in green fruits has

been found only at high light levels and the balance becomes negative at low levels (Cipollini

and Levey 1991), which does not explain the importance of green fruits in environments of

varying light levels, such as a rain forest or the forest understory.

Last but not least are phylogenetic constrains. Fruit traits in general have been shown to be

highly conserved (Jordano 1995), probably due to the long generation time of plants in

comparison with seed dispersers (Herrera 1985). Our comparison between species in the same

genus suggests that similar colors among species dispersed by primates and those dispersed by

birds are the result of convergence rather than shared common ancestry, at least in the 12 genera









included. Although not statistically significant, the trend found in 22 families also suggests the

independent evolution of fruit color.

We emphasize the importance of measuring color objectively, using newly available

portable spectrometers (Altshuler 2001; Altshuler 2003) and to interpret these colors based on

the visual capabilities of the frugivores under study (Chittka and Raine 2006). Although

primates have sometimes been merged with birds in a bird-primate syndrome because no

differences in color were found between fruits dispersed by these two groups of frugivores

(Gautier-Hion et al. 1985), we show that, while primates and birds overlap in the color of the

fruits they consume, primates disperse more green fruits than do birds, and that birds disperse

more blue fruits than do primates. These results match those of earlier studies in fruit choice by

frugivores (Janson 1983; Wheelwright and Janson 1985), even when those studies classified

color from a human perspective. We believe that this match, despite such different methods of

measuring color, is because we are testing frugivores whose color vision capability is similar to

that of humans. This is, of course, with the non-trivial exception of UV perception in birds,

which we were unable to consider due to lack of UV reflectance information in the literature.

We emphasize that using vision models may be especially important if the visual system of the

frugivores under study (e.g., a dichromatic primate) differs significantly from our visual system.

In the case of this study, if fruit reflectance in the UV range were included in our analysis, the

discrimination between primate and bird-dispersed fruits using all four of a bird's cone peak

sensitivity would almost certainly be sharper than that shown here. The results from the

primate's perspective should not change because we would be incorporating reflectance

information on a part of the spectrum that primates cannot see.









Although we cannot be sure that our model exactly represents how birds and primates

perceive color, as color vision entails much more than cone excitation, our model provides an

objective approximation of color vision by these frugivores. This is the first study that addresses

the Signal Convergence hypothesis in a fruit-frugivore system, using such an objective measure

of color and such a large data set of reflectance spectra (the largest analyzed to date).

Color is an important mediator in plant-animal communication (Chittka and Raine 2006;

Johnson and Steiner 2000; Schaefer et al. 2004, this study) and we believe that learning the

signaling role of color will shed light on understanding how animals and plants affect each

other's fitness and evolution. This can only be done by using vision models that incorporate the

receptive and cognitive processes of the receivers, which adds much needed objectivity in the

assessment of how plant signal to animals (Schaefer et al. 2004).











Red fruit


c


8

X


400 450 500 550 600 650 70
Wavelength (nm)


E

(0
-L










400 450 500 550 6 650 700
Wavelength (nm)


0440 0540 Q570


Figure 3-1. Reflectance spectra of two different fruits (left) combined with the peak cone sensitivities of a frugivore (middle) yields
the quantum catch for the three cone types for each fruit (right). The quantum catch values approximate the frugivore's
perception of color.


Q440 Q540 Q570





Green fruit











SGreen -- Blue A
40


30



i s
LD

10 f



C -55 -45 -35 -25 -15 -05 05 15 25 35


I B
L0


368


25
a. %

15 \







LDF 1



Figure 3-2. Results of the Discriminant Function analysis (DFA) based on (A) primate cone
excitation and (B) bird cone excitation. Black lines correspond to the "training" DFA
and grey lines to the validation DFA. Dashed lines represent primate-dispersed fruits
and solid lines represent bird-dispersed fruits. Primate-dispersed fruits tend to have
negative scores on the linear discriminant function, because they are greener than
bird-dispersed fruits, which have more of a blue hue than primate-dispersed fruits
















68











Table 3-1. Loadings of the quantum catch variables for primate and bird cone sensitivities for
the "training" DFA


Variable
Short (Blue light)
Medium (Green light)
Long (Red light)
Proportion of variation


Primate vision
1.396
-2.758
0.771


1.00


Loading on LDF 1
Bird vision
1.234
-1.797
-0.071


1.00









CHAPTER 4
REVISITING THE DISPERSAL SYNDROME HYPOTHESIS: A PHYLOGENETIC
COMPARATIVE STUDY OF FRUIT TRAITS IN FICUS (MORACEAE)

Introduction

Seed dispersal by vertebrates occurs in tropical, temperate, xeric, and moist environments

alike, both on land and in aquatic habitats (Adams et al. 2007; Godinez-Alvarez et al. 2002;

Herrera 1987; Korine et al. 2000; Shanahan et al. 2001b; Spiegel and Nathan 2007; Traveset et

al. 2004). Its selective advantage in various geological scenarios and historical periods is

evidenced by its independent evolution in many plant lineages, from seed ferns, to gymnosperms

and angiosperms (Herrera 1989; 2002b). It is particularly ubiquitous in moist tropical

environments, where over 90% of woody species are dispersed by animals (Herrera 2002b), and

where it has been proposed as on of the most important proximate mechanisms explaining high

tree diversity (Janzen 1970; Terborgh et al. 2002). Yet, the question "How important are seed

dispersers to fruiting plants?" may elicit fundamentally different answers from the ecologists and

the evolutionary biologists in the field. While ecologists argue convincingly that seed dispersal is

important to the fitness of fruiting plants, many evolutionary biologists argue just as strongly that

there is little evidence seed dispersal "matters" to plants (i.e., that it has influenced the evolution

of fruit traits).

Among the most cited lines of evidence for the contemporary role of seed dispersal in plant

and animal communities are that it links the last phase of a plant's reproductive cycle to the first

phase of population recruitment (Herrera et al. 1994; Nathan and Muller-Landau 2000; Rey and

Alcantara 2000; Schupp and Fuentes 1995), contributes disproportionately to forest recovery

after natural disasters (Hjerpe et al. 2001; Thornton et al. 1996; Whittaker and Jones 1994), and

dramatically reduces density-dependent mortality near the parent plant (Fragoso et al. 2003;

Howe 1986; Howe and Miriti 2000; 2004; Howe et al. 1985; Schupp 1988). There is also









growing evidence that dispersal limitation (i.e., the inability for seeds to colonize sites

appropriate for germination) may be a major process limiting plant recruitment, affecting plant

spatial distributions and contributing to the maintenance of biodiversity (but see Clark et al.

2007; Clark et al. 1999; Hurtt and Pacala 1995; Muller-Landau et al. 2002; Murray and Garcia-C

2002; Turnbull et al. 2000). Thus, plants species without dispersers may face extinction

(Cardoso da Silva and Tabarelli 2000; Chapman and Onderdonk 1998; Cordeiro and Howe 2001;

2003). Conversely, reductions in fruit abundance may threaten frugivore populations

(Moegenburg and Levey 2002; Redford 1992).

But evolutionary evidence is less clear-cut. Some argue that seed dispersal is

consequential to the evolution of fruits and point to the apparently non-random combination of

fruit traits (Cipollini et al. 2002; Gautier-Hion et al. 1985; Janson 1983; Korine et al. 2000),

called "dispersal syndromes", that match presumed preferences and fruit-handling abilities of

different sets of frugivores. However, most evidence to date suggests that seed dispersal is

inconsequential in shaping fruit traits, as factors other than seed dispersers can better explain

fruit traits these factors include phylogeny (Fischer and Chapman 1993; Herrera 1987; 1989;

1992; Jordano 1995), architectural constraints (Herrera 1987; 2002a), and growth form (Herrera

1987). Moreover, fruit traits presumably selected for by frugivore taxa such as birds or

mammals are not always appealing to their putative dispersers (Tamboia et al. 1996), casting

doubt on the validity of adaptive explanations for the association of those traits with particular

sets of dispersers.

It is puzzling that researchers have failed to find support for the potential role of seed

dispersers in the evolution of fruits because many assumptions that are central to the adaptive

role of fruit traits hold true, namely: 1. seed dispersal is often important for plant reproduction









(Fragoso et al. 2003; Harms et al. 2000; Howe and Miriti 2000; 2004); 2. seed dispersers can

exert disparate selective pressures on fruiting plants because disperser guilds treat seeds

differently (Clark et al. 2001; Godinez-Alvarez et al. 2002; Poulsen et al. 2002), and may

therefore affect plant fitness differentially; 3. disperser guilds perceive fruit traits differently and

can differentiate fruit traits well (Hodgkison et al. 2007; Korine and Kalko 2005, Chapter 3, this

dissertation), making it feasible for plants to use fruit traits to attract beneficial consumers and/or

to deter detrimental consumers, and 4. some traits such as fruit size and color are not strongly

phylogenetically conserved, suggesting potential for selection by seed dispersers (Jordano 1995;

Lomascolo et al. 2008 (Chapter 2, this Dissertation); Voigt et al. 2004).

We argue that one of the reasons for the contradictory conclusions of evolutionary studies

is the analytical approach used by different studies. First, some studies have failed to consider

the role of phylogeny in determining the similarity in fruit-trait combination among plant species

(Janson 1983; Kalko et al. 1996; Korine et al. 2000), which fails to account for the fact that

species may not be independent data points. Second, those who controlled for the non-

independence of species have either focused at higher taxonomic levels or have used estimates of

the phylogenetic relatedness of species (instead of explicit phylogenies), methods that often

decrease the power to find a correlation between disperser type and variation in fruit traits, when

it exists (Bremer and Eriksson 1992; Fischer and Chapman 1993; Jordano 1995). Third, some of

the most influential studies come from temperate regions (Herrera 1987; 2002a) where fruit

diversity is limited compared to tropical regions; therefore great variation in tropical fruit traits

remains unexplained. Fourth, thorough studies that investigated the variation of fruit traits with

dispersal mode have analyzed fruit traits separately (Herrera 2002a; Jordano 1995), whereas

frugivores most likely assess whole fruits based on many variables simultaneously. Finally,









dispersal guilds have often been defined very broadly (i.e., birds vs. mammals, Janson 1983;

Jordano 1995), which hinders the definition of the potential selective pressure of frugivores on

the fruit traits of interest.

Perhaps the most glaring reason why evolutionary studies are not conclusive on whether

frugivores have selected for fruit traits is that some of the most important fruit traits used by

frugivores to find and select fruits have been excluded or mismeasured. The most striking

examples of this are fruit color and odor. These traits are commonly used by plants to signal

fruit ripeness (Sumner and Mollon 2000b) or nutritional composition (Catoni et al. 2008; Goff

and Klee 2006; Schaefer et al. 2008; Schaefer and Schmidt 2004), and are two of the most

important cues used by frugivores to find and select fruits (Kalko et al. 1996; Laska et al. 2007;

Thies et al. 1998; Voigt et al. 2004; Wheelwright and Janson 1985; Willson and Whelan 1990).

In particular, they are quantified from the perspective of human perception. This can generate

significant and unintended biases, given that other animals sense the world differently than

humans and that most plants use color and odor to signal to animals other than humans (Bennett

et al. 1994; Schaefer et al. 2004). For example, most passerine birds can see UV light, while

humans and most other mammals cannot (Bennett and Cuthill 1994; Bennett and Thery 2007;

Hart 2001; Peichl 2005). Likewise, most mammals have a more developed sense of olfaction

than humans and, especially, than birds (Clark 1991; Evans 2006; Heymann 2006; Wysocki and

Preti 2004).

Our work improves on previous studies by 1. focusing at the species level and controlling

for the non-independence of species using a molecular phylogeny; 2. studying tropical plant

genus with highly variable fruit traits; 3. using a multivariate phylogenetic comparative approach

to simultaneously test for the correlated evolution of seven fruit characteristics that are predicted









to be important for frugivores to find fruits, namely fruit size, color, odor, softness, contrast

against the background, location on the tree, and exposure away from foliage; 4. focusing on two

well defined disperser guilds with clearly different potential for selection on fruit traits; and 5.

reducing human biases by using quantitative measures of color using a spectrometer, and of odor

through gas chromatography/mass spectrometry.

We test two hypotheses for co-occurring sets of fruit traits. First, the Phylogenetic

Hypothesis: fruit traits are inherited from ancestors they are the result of shared phylogenetic

history. Second, the Dispersal Syndrome Hypothesis: fruit syndromes are the result of selective

pressure by sets of seed dispersing frugivores that differ in behavior and ecology. Our objective

is to tease apart these hypotheses by determining how much variation in fruit traits is attributable

to phylogeny and how much is attributable to selective pressure by frugivores. We focus on two

well defined groups of frugivores, birds and bats, which are predicted to select for the following

dispersal syndromes: 1. Passerine bird syndrome: Because passerine birds have a relatively

narrow gape, lack teeth, have acute color vision (tetrachromatic, including a peak in the UV

range, Bowmaker et al. 1997; Das et al. 1999; Hart 2001), a poor sense of smell, and usually

perch to pick fruits, fruits preferred by passerines should be small, visually conspicuous (hues of

red, purple or black, contrasting against leaf or bark background), less odorous than bat-

dispersed fruits, and presented among foliage along branches. 2. Bat syndrome: Because bats

have teeth (and therefore can manipulate fruits with their mouth and eat them piecemeal), are

mostly nocturnal (which limits their perception of color), use chemical cues to find fruit

(Bianconi et al. 2007; Hodgkison et al. 2007; Mikich et al. 2003), and grab fruits in flight or land

on trees and crawl to reach fruits (Elangovan et al. 2001; Kalko et al. 1996), fruits preferred by

them should be larger than bird-dispersed fruits, odorous, and less visually conspicuous (hues of









green, yellow, pale orange, or brown, non-contrasting against leaf or bark background). Also,

they will grow along the trunk (caulicarpic) and, in general, they will be more exposed away

from the foliage than fruits eaten by birds (Ridley 1930; van der Pijl 1969). We demonstrate that

these two groups of frugivores have had an evolutionary impact on fruit traits within Ficus.

Methods

Study Site and General Specifications

Data were collected between September 2004 and December 2005 in a lowland forest near

the village of Ohu (1450 E, 50 S) in Madang Province, Papua New Guinea, (Fig. 4-1). Data on

fruit color, color contrast against the background, odor, size, softness, and exposure to frugivores

(location of fruits on branches or trunk, distance from closest leaves, and peduncle length) for

each of 42 fig species were obtained by averaging records from between 3 and 8 individuals per

species. For each individual tree, between one and 10 fruits were measured and averaged for all

traits.

Study system: the natural history of figs

Ficus (Moraceae) provide a unique opportunity to study the evolution of fruit traits in

relation to seed dispersers because its species are unusually variable in fruit color, size,

presentation, and type of disperser. Fig fruits are technically closed infructescences, that are

hollow and shaped like an urn, most probably selected to allow the characteristic internal

pollination provided by wasps (Janzen 1979; Weiblen 2002). (For more information on fig

pollination and its relation with the seed dispersal mutualism, see Chapter 5, this dissertation.)

Fig infructescences are called syconiaa", a term we use interchangeably with "fruits" because

they are functionally fruits; i.e., they are comprised of seeds and nutritious pulp, and they attract

and reward seed dispersing animals. They can be red, purple, black, pink, orange, yellow, green,

or brown. Syconia range in diameter from 3-80 mm and grow from the tips of branches, at the









base of leaves, on the trunk from ground level to the canopy, and even from the roots (Laman

and Weiblen 1998, G. Weiblen pers. comm., S. Lomasclo pers. obs.). In terms of life form, figs

can be free standing, epiphytic, hemi-epiphytic, or "parasitic" stranglers of other trees. They are

also geographically widespread (Shanahan et al. 2001b). More than 700 species are found

worldwide in tropical forests, thereby regularly co-occurring with different sets of dispersers.

Dispersers include bats, arboreal and terrestrial mammals, volant passeriness, hornbills, fruit

doves, parrots, etc.) and flightless birds (e.g, Cassowaries), fishes, and reptiles (Shanahan et al.

2001b).

Figs are best known for their tight mutualism with pollinating wasps, in which one species

of fig is typically pollinated by one species of wasp. The mechanism of pollination is quite

similar in all species ofFicus: a gravid female wasp carrying with pollen enters the closed

inflorescence, pollinates the flowers and lays eggs in some of the ovaria. Therefore, viable seeds

develop from some flowers and wasp larvae develop from others (Janzen 1979). Because

pollinators may influence flower anatomy (Temeles and Kress 2003) the fact that all fig species

have virtually the same pollinator contributes to making Ficus a good study system: floral

morphology is fairly constant and, therefore, most differences found in fruit characteristics

among figs should be largely independent of floral traits.

In many forests and across continents, figs are considered a "keystone" resource because

they produce large fruit crops that ripen during seasons when few other fruits are available

(Lambert and Marshall 1991; Terborgh 1986). Figs may also be important to frugivore

communities due to their high calcium content (O'Brien et al. 1998). On the flip side, frugivores

appear important for figs because figs depend heavily on seed dispersal for regeneration (Titus et

al. 1990).









Figs have often been deemed generalists because they produce large amounts of fruits with

low nutritional quality, fruit year-round, and attract large numbers of dispersers from different

guilds. These attributes have likely discouraged researchers from considering figs as a study

system for detecting dispersal syndromes. However, when figs are studied with attention to

species-specific differences, it becomes quickly apparent that many species are almost

exclusively visited by specific guilds of frugivores, such as bats, arboreal mammals, and birds

(Kalko et al. 1996; Korine et al. 2000; Shanahan 2000; Shanahan and Compton 2001; Shanahan

et al. 2001a), and only some of them fruit continuously (S. Lomascolo, pers. obs.).

Quantification of Visits to Figs

All data on fig consumption come from figs that contain viable seeds; these figs are

commonly called "seed" figs (see Chapter 5). We recorded consumption of figs using two video

cameras (Sony DCR-HC40), which were set up 3-5m away from the fruiting Ficus tree and

visited only to change tapes or batteries. To record both diurnal and nocturnal frugivores, we

started videotaping at 06:00 and ended at 10:30, and continued again between 18:30 until 23:00.

We recorded nocturnal visits using an infrared light (Sony HVL-IRH2). We aimed one camera

at the ground to record terrestrial frugivores and the other either at a branch or at the trunk,

depending on where most of the ripe figs were located. We recorded between one and four

individuals of 29 species of figs for a total of 1019 hours. Unless a frugivore was seen not to eat

any fruit, all frugivores that entered the tree were considered seed dispersers. For some species

for which fruit data were collected but no dispersers were recorded, we obtained frugivore

information from Shanahan et al. (2001b). One species for which no frugivore was recorded, F.

botryocarpa had been repeatedly pointed out by villagers as a commonly bat-dispersed species,

so we also included it in the analysis as a bat-dispersed species.









Fruit Traits

Fruit color: We quantified fruit color using a USB2000 portable spectrometer (Ocean

Optics, Inc) and a PX-2 pulsed xenon light source, which took reliable readings between 300 and

740nm. This spectrum includes the range of wavelengths visible by humans, plus the ultraviolet

range. We scanned fruits using a sensor with six optical fibers illuminating the target fruit and a

seventh fiber that returned the reflected light to the spectrometer. The scanning angle was fixed

at 450 by using a black metal stand with a hole positioned at that angle. The metal stand also

blocked external light. To keep the distance between the end of the sensor and the fruits constant

we fastened a non-UV filtering microscope slide to the opening of the hole where the sensor was

introduced. To obtain an average reflectance spectrum per fruit, we scanned each fruit three

times in three different spots on the fruit. We calculated reflectance as the proportion of a

Spectralon white reflectance standard (Labsphere, North Sutton, NH) and, hence, reflectance

measures do not have units. The three spectra for each species were averaged every 5nm.

We obtained a quantitative measure of fruit color for each species using a Principal

Components Analysis (PCA), which summarized the information for the complete spectrum by

finding linear combinations of the original variables -- in this case, reflectance intensity every 5

nm (Clotfelter et al. 2007; Cuervo and Shine 2007; de Ayala et al. 2007; Endler 1990). This

technique reduces multivariate data to several axes that summarize the variance in, and

covariance among, the original variables. A separate PCA was done to obtain color variables for

the group of 42 species of figs for which fruit data were collected, and for the group of 19

species of figs for which exclusive dispersal by birds or bats was recorded. These two groups

were used in different statistical analyses to answer separate questions (see below). In both cases

the first principal component (PC) had high positive loadings for all wavelengths throughout the

spectrum, and hence represents fruit brightness (Fig. 4-2), i.e., the total amount of light reflected









by the fruit. Therefore, light fruits had high scores on PC 1 and dark fruits had low scores.

Principal components 2 and 3 represent different hues. Hue refers to the "everyday" meaning of

color, summarizing differences in reflectance of different wavelengths. The second PC

distinguishes reflectance in the red versus the UV-blue wavelengths because it had positive

loadings for wavelengths from 300 to -600nm, and negative loadings from 600 to 740nm.

Therefore, UV-blue fruits had a high positive score on PC 2, whereas fruits reflecting in the red

part of the spectrum had high negative scores. The third principal component distinguishes

reflectance in the green versus red and UV-blue range because it had high positive loadings of

wavelengths between 450nm and 640nm. Thus, green fruits had a high score on PC 3 and UV-

blue and red (and, therefore, purple) fruits had a low score. These three PCs explained 91% of

the variance in reflectance and all had eigenvalues greater than one, which means that each

explained more variation than any single original variable alone. Although PCs 4 and 5 also had

eigenvalues higher than one, they only explained 4 and 3% of the variation, respectively.

Because reflectance in all parts of the spectrum of interest was summarized by PCs 1 to 3, we

considered that additional PCs did not contribute sufficiently to the color description of each fig

species. Hence, we use only scores on PCs 1 to 3 in subsequent analyses.

Color contrasts were calculated as the Euclidean distance between the color of a fruit and

the color of the structure against which it was exposed to the frugivore, which consisted of either

leaves or bark, depending on the species. Spectra were normalized to the same brightness by

dividing the reflectance at each wavelength by the total reflectance for each species. This

generates contrasts due to the color of the fruits not to their brightness. These contrasts are

called "chromatic" contrasts (Schmidt et al. 2004a) and are calculated as

D = [Qf(A) Qb(A)]2










were Qfis the color spectrum of the fruit and Qb is the color spectrum of the background

structure; X is the wavelength in nm and the sum corresponds to the complete spectrum between

300-740nm.

Fruit odor: To capture volatile compounds, we collected fruits in the field, brought them

to the lab, and placed them inside bags made of inert material (Reynolds oven baking bags). A

vacuum pump (Welch, model #2522B-01) sucked air into the bag through a carbon filter on one

end and forced it out of the bag through a filter at the opposite end filled with Super Q (Alltech),

which adsorbed volatiles emitted by the fruits. For a given collection of fruits in a bag, we ran

the pump for four continuous hours, which began no more than three hours after the fruits had

been collected from the tree. To extract fruit volatiles from the Super Q filters, we flushed

methylene chloride through the filters. We then used gas chromatography to separate and

quantify the amount of different (unknown) volatiles. We also used mass spectrometry to

identify some of the compounds. Because there were too many compounds (n -75) to identify or

analyze separately and because we were most interested in determining whether bird and

mammal fruits differ in overall production of volatile compounds, we calculated the total amount

of volatile compounds by summing the areas under each compound's peak for each species. We

standardized for number and size of fruits in a bag by dividing the amount of each volatile

contained in the sample by the total surface area of the fruits in the bag that produced the

volatile.

Fruit size, and softness: We measured fruit diameter with calipers to the nearest 0.5 mm

and we used this as an estimate of overall fruit size. To measure fruit softness, we determined an

arbitrary scale between 1 (hardest; fruit surface did not indent when pressed against fingers) to 4

(softer; fruit surface indented easily when pressure was applied to fruit between the fingers). To









achieve consistency in this arbitrary measure the same researcher recorded this measurement

throughout the study. Fruit size and softness, together with fruit color and odor will be referred

to as "appearance traits", to differentiate them from traits reflecting exposure to the frugivores

("exposure traits").

Fruit exposure traits: To quantify fruit position away from the foliage, we estimated the

distance between the fruit and the nearest leaf to 1.0 cm for the highest and the lowest fruit in the

tree, and averaged these values to incorporate the variation within each tree. Another exposure

variable was dichotomous: fruits growing from branches versus those growing from the trunk. A

third measure of exposure was peduncle length, the distance ( 0.5 mm) between the base of the

fruit and the point where the peduncle attached to the branch or trunk.

Data Analysis

All statistical tests were done using R version 2.6.0 (R Development Core Team 2008),

except for the Evolutionary Principal Components Analysis (Dyreson and Maddison 2006), for

which we used Mesquite 2.01 (Maddison and Maddison 2007).

Identification of correlation in fruit traits: Principal Components Analysis (PCA) was

used to identify the main axes of variation in the fruit data. In this case, PCA was used to

identify the main differences between bird and bat-dispersed figs, and to define which variables

covaried along the axis separating figs dispersed by these two guilds. Because variables were

measured in different units and differed by up to three orders of magnitude, they were

standardized to zero mean and unit standard deviation.

We performed two different PCAs, one on the 19 species of figs that were dispersed by

either birds or bats (hereafter "Disperser PCA"), and one on the 42 species of figs for which we

collected fruit data (hereafter "Fruit Trait PCA"). The Disperser PCA was done to objectively

define fruit syndromes by identifying the suites of traits that differed between bird-dispersed and









bat-dispersed figs. Species with mixed dispersal were not included in this analysis because they

were predicted to have intermediate characteristics between species exclusively dispersed by

birds or by bats, and our goal was to identify the differences between bird and bat fruits. We

tested whether the difference between bird and bat dispersed species on the PC axis explaining

the most variation between the two groups was significant using a Mann-Whitney U test. The

Fruit Trait PCA included all the fig species for which fruit trait data were collected, and it

included species dispersed by birds, by bats, by both, and those for which disperser data was not

available. The goal of the Fruit Trait PCA was to explore the overall pattern of variation in the

fruit variables and to identify whether the trait correlation observed in the bird and bat-dispersed

species could be generalized to a wider set of fig species. Also, we explored whether fig species

with mixed dispersal had intermediate trait values between bird and bat fruits.

Phylogenetic control: To test whether the pattern of fruit trait variation was most likely

explained by phylogeny or could be considered adaptive, we performed an Evolutionary

Principal Components Analysis (EPCA) (Dyreson and Maddison 2006; Schlick-Steiner et al.

2006). This EPCA is based on evolutionary transitions (the difference in trait value between

each species and the most recent ancestor, as reconstructed by a parsimony criterion, minimizing

the sum of squared changes) instead of the raw variables used in a standard PCA. While

standard PCA maximizes the variation in the traits among species, EPCA maximizes

evolutionary change incurred by each species after they split from their ancestor, thus avoiding

the problem of non-independence of species (Dyreson and Maddison 2006). Just as explained

above for PCA, EPCA was completed on the 19 species set of bat and bird-dispersed species

(Disperser EPCA) and on the 42 species set (Fruit Trait EPCA). To test whether there was a









significant difference between bird and bat dispersed species on the EPC 1 resulting from the

Disperser EPCA we used a Mann-Whitney U test.

A second method of testing the role of phylogeny in explaining variation in fruit traits is to

correlate that variation with genetic relatedness among species (Voigt et al. 2004). If what

determines covariation of fruit traits (i.e., dispersal syndromes) is phylogenetic history, then

closely related plant species should share combinations of fruit traits. If, instead, dispersal

syndromes result from divergent selective pressures of different types of seed dispersers,

similarity of fruits between any given pair of species should not be related to the genetic

relatedness of those species. For each possible pair wise comparison of the 42 fig species for

which we collected fruit data, we calculated genetic relatedness between all possible pairs of

taxa, quantified in PAUP* (Swofford 2003) as the number of character changes along the tree

between each pair of species corrected by the model of evolution chosen to build the

phylogenetic tree. This genetic difference between two taxa is called "patristic distance" and is

proportional to the sum of branch lengths between them. We also calculated the difference in

fruit appearance between all possible pairs of taxa as the absolute value of the difference in the

score on the PC 1 resulting from the Fruit Trait PCA, which was the most important in separating

bird and bat-dispersed species. We then tested for a correlation between these estimates of

genetic relatedness and difference in fruit appearance (n = 861 pairs) using a Mantel test (Mantel

1967). The Mantel test avoids the problem of non-independence of species by first calculating

two distance matrices, one based on genetic distance and the other based on difference in trait

appearance, and then randomly permuting rows and columns within each matrix. The

probability of the correlation observed, as quantified by r2, is calculated based on the null

distribution created by the randomization.









Volatile compounds in figs with different dispersers: A Discriminant Function Analysis

based on presence/absence of 73 volatile compounds was used to identify the most important

compounds distinguishing fig species dispersed primarily by birds (5 fig species), bat (13

species), bandicoot (2 species), and by more than one of these groups (10 species). To test the

significance of the discrimination, we calculated a Kruskal-Wallace (K-W) statistic for the

difference in the score on Linear Discriminant Function (LDF) 1, which mainly separated

between the groups in the original data matrix, then randomized the matrix and re-calculated a

K-W statistic for the random matrix. Column totals were fixed to control for common volatile

compounds, while row totals were allowed to vary. This process was repeated 10,000 times to

obtain a random distribution of the K-W statistic and then compared our observed K-W statistic

to the random distribution.

Phylogenetic reconstruction

Although four phylogenies of the genus Ficus have been published (Jousselin et al. 2003;

Lomascolo et al. 2008; Ronsted et al. 2005; Weiblen 2000), none have all the species for which

we collected fruit data. We therefore reconstructed our own phylogeny, using DNA sequences

from GenBank and some from N. Ronsted (Department of Medicinal Chemistry, University of

Copenhagen). We constructed one tree for the 19 species exclusively dispersed by birds and bats

(necessary for the Disperser EPCA), and one for the 42 species for which we had data on fruit

traits (necessary for the Fruit Trait EPCA) (Fig 4-3). A third tree that included 29 species with

known dispersers (the 19 species exclusively dispersed by birds and bats, and 10 species with

mixed dispersal) was constructed in order to map dispersal mode (Fig. 4-4). We used a

maximum likelihood criterion, which searches for the phylogenetic tree that best fits the data,

given a particular model of evolution. The model of evolution that best described our data was

determined using the Akiake Information Criterion from the model set examined by









MODELTEST 3.6 (Posada and Crandall 1998). We conducted a heuristic search in PAUP* with

10 random sequence additions using the appropriate model (GTR+G for all three data sets) and

parameters recommended by MODELTEST 3.6 (Posada and Crandall 1998). We assessed

support on the phylogeny using 500 maximum likelihood bootstrap replicates using GARLI

0.951 (Zwickl 2006). Unresolved nodes were resolved for further comparative analyses based

on published phylogenies. In the few cases where the species were not included in any published

phylogeny, we resolved it randomly. Several random combinations were tested and all yielded

similar results, so all results reported are based on one resolution, as seen in Figure 4-3.

Results

Visits to Fig Trees

Out of the 29 fig species that were videotaped, we obtained disperser data for 15. Data on

frugivore consumption of 13 additional species were obtained from Shanahan et al (2001b), and

one from the knowledge of local villagers. Most species were dispersed by bats (14 species),

birds (5 species), or both (10 species). Two species were recorded as exclusively bandicoot

dispersed and they were included in the exploratory analysis of volatile compounds in bat, bird,

bandicoot and bird dispersed fruits. Two observations were too few to justify the addition of a

Bandicoot Syndrome, so we did not include it in other analyses.

Based on a parsimony reconstruction of dispersal mode we identify at least three switches

from bat to bird dispersal and from bat to mixed dispersal (Figure 4-4). Although many sister

species share the same dispersal mode, this reconstruction suggest dispersal mode has changed

independently many times in this group of figs.

Disperser PCA and EPCA

The first principal component (hereafter PC) explained 38% of the variation in the data,

and is associated, in decreasing order of magnitude, with contrast against the background (-),









softness (-), distance from leaves (+), growth from branches (-), peduncule length (+), size (+),

and brightness (+) (Fig. 4-5A, Table 4-1). Bird-dispersed species tend to have lower scores on

PC 1 than bat-dispersed species, suggesting that dispersed species are more contrasting against

the background, softer, more often growing among the foliage and on branches, have shorter

peduncules and are smaller than bat-dispersed species (Fig. 4-5A). All of these characteristics

match those predicted for bird and bat syndrome fruits. The second PC explained 17% of the

variation and it seemed to mainly separate two fig species (F. semivestita and F. edelfeltii) from

the rest. We saw no identifiable biological relevance of this separation with respect to disperser

type, and therefore we will not report this PC further. Controlling for phylogeny, the EPCA

reveals a similar pattern (Fig. 4-5B), which is confirmed by a highly significant correlation

between the scores of each fig species on PC land EPC 1 (r2 = 0.77, d.f. = 17, p-value < 0.0001).

The loadings for contrast against the background, softness, distance to closest leaf, peduncule

length, size, and did not change substantially between PC 1 and EPC 1, although their order of

importance changed slightly (Table 4-1). The fruit traits that showed most difference between

PC 1 and EPC 1 were fruit brightness and odor, which became more important in distinguishing

between bird and bat figs when phylogenetic effects were controlled and location, and location

of fruits on branches or trunk, which became less important (Table 4-1). Although the loading of

the color variable defining UV-blue versus yellow (PC3) is not high, it is worth noting that its

sign changed from positive in PC 1 to negative in EPC 1.

Fruit Trait PCA and EPCA

The Fruit Trait PCA with 42 fig species shows a pattern of ordination similar to the

Disperser PCA. The first PC explained 38% of the variation in fruit data and was associated, in

decreasing order of magnitude, with contrast against the background (-), growth from branches

(-), peduncule length (+), distance from leaves (+), softness (-), size (+), and odor (+) (Fig. 4-6A,









Table 4-2). Except for contrast against the background, color variables were not as important to

explain the variation among fig species. The EPCA reveals a similar pattern to that of PCA, but

the ordination is not as clearly matched to the PC axes, as it is in the Disperser EPCA. Fig

species seem to be separated on EPC 2 in a similar fashion to their spread on PC 1 (Fig. 4-6).

Because EPC 1 and EPC 2 explain a very similar amount of variation (21.4% vs 20.3%.

respectively) and defining which is the main axis of variation is unimportant, and because bird

and bat dispersed species are separated in both dimensions, we report both axes in detail. A

strong, positive correlation (r2 = 0.79, d.f. = 40, p-value < 0.0001) between the score of each

species on PC 1 and on EPC 2 suggests that the ordination of species does not change

substantially after controlling for phylogeny. EPC 2 shows higher loadings of exposure

variables, as it is associated, in order of magnitude with distance from leaves (-), growth from the

trunk (-), and peduncle length (-), and also with contrast against the background (+) and odor (-),

whereas EPC 1 is largely based on differences in fruit appearance, as it is associated, in order of

magnitude, with brightness (+), size (+), and softness (-), and also with peduncule length

(+)(Table 4-2). Most exposure variables have similar loadings on EPC 2 to those of PC 1, but

the relative importance is slightly different. In particular, fruit size and two of the color variables

defining fruit hue became less important in EPC 2. On the other hand, the loadings of most

appearance variables on EPC 1 are similar to those on PC 1.

The correlation between genetic relatedness, as measured by branch length between pairs

of species, and the difference in the score on the PC1 axis is significant but explains less than 3%

of the variation in the data (r2 = 0.023, d.f. = 859, p-value < 0.02) (Fig. 4-7).

Volatile Compounds

The DFA on volatile compounds resulted in significant discrimination between bird, bat,

bandicoot, and mixed dispersal modes (Fig. 4-8A). The twelve most important compounds









associated with bat-dispersed species are shown in Table 3. The scores on LDF 1, which

separated bird from bandicoot and bat dispersed figs, were compared with a Kruskal-Wallis test

and are significantly different from each other (K-W x = 20.38, d.f. = 3, p-value = 0.0001). The

randomization of the data matrix and re-calculation of the Kruskal-Wallis statistic 10,000 times

showed that the probability of obtaining by chance a discrimination as significant as that found

with the original matrix was 0.0002. Figure 4-8B shows an example of the discrimination

achieved with a random matrix.

Volatile compounds more strongly associated with bat dispersal were mostly esters

(Pentanol acetate, Acetamide, Ethyl octanoate, and Hexenoic acid, and a small, unidentified

compound), with some sesquiterpenes (Muurolene, Panasimsen, and Germacrene D) and an

unsaturated alcohol (unidentified), which generally coincides with previous studies of volatiles in

bat-dispersed figs (Hodgkison et al. 2007).

Discussion

Our phylogenetic comparative study of 10 variables describing fruit appearance and

accessibility to frugivores suggests that fruit traits covary in a way predicted only under selective

pressure by frugivores. We believe our results are particularly robust for two reasons. First, we

controlled for phylogenetic effects using a species-level phylogeny. Although phylogenetic

effects can be controlled in other ways (Fischer and Chapman 1993; Herrera 1987; Jordano

1995; Voigt et al. 2004), using a species-level phylogeny is currently the most accurate manner

for estimating the true phylogenetic history of a given taxon. Second, we used a multivariate

phylogenetic comparative analysis, which allowed us to search for patterns in many variables

simultaneously without violating assumptions of independence. This approach provided an

holistic analysis of the multiple traits that frugivores perceive and use to find and assess fruits. It

is simplistic to assume that a frugivore assesses, for example, the size of a fruit independently of









its color, odor, or position in the tree. Just as the conspicuousness of any given fruit color is

higher in large than small fruits (Sumner and Mollon 2000b), conspicuousness almost certainly

depends on a combination of many traits that together make the fruit relatively easy or hard to

find by a frugivore, based on its overall perception of the world.

Our results strongly suggest that bird and bat syndromes indeed exist in figs. In Papua

New Guinea, figs in the Bird Syndrome are darker, more contrasting against the background, less

odorous, less exposed away from the foliage, smaller and softer than figs in the Bat Syndrome.

This list generally matches the set of traits previously suggested for Bird Syndrome and Bat

Syndrome fruits (Knight and Siegfried 1983; Shanahan et al. 2001b). Our approach, however,

allows us to conclude that this particular combination of traits reflects evolutionary transitions --

i.e., changes undergone by each fig species after splitting from an ancestral species. Therefore,

these sets of traits are likely adaptive.

Birds and bats also showed an association with figs growing from branches or directly

from the trunk, respectively, but the importance of these variables was dramatically reduced in

the EPCA, which may suggest that the location of figs on a tree is most strongly determined

phylogenetically. Although several alternative hypotheses have been proposed to explain the

colors of fruits (Willson and Whelan 1990), or the correlated evolution between, for example,

fruit color and size (Cipollini and Levey 1991; Lomascolo et al. 2008 (Chapter 2, this

Dissertation)), we are not aware of any alternative hypothesis that can explain the correlated

evolution of fruit size, brightness, contrast against the background, odor, softness, and exposure

away from the foliage. The most parsimonious explanation for the correlated evolution of these

six traits is natural selection by seed dispersers.









The same general pattern is apparent in a much larger data set of fig species, although data

on consumers are not available for all species. This suggests that Dispersal Syndromes exist and

are adaptive, at least for the set of 42 fig species included in this study. Our results generate

testable predictions about potential consumers and seed dispersers of fig species, especially those

falling in the extremes of the ordination shown in Figure 4-6. Our results also show that

frugivores overlap in the fruits they eat. Many fig species that fall close to the "bird" extreme of

the dispersal syndrome axis (EPC 2 in the Fruit Trait EPCA) were recorded to be also (e.g., F.

microcarpa, F. phaeocyse, F. pungens, F. wassa,) or even exclusively (e.g., F. subulata, F.

virgata) dispersed by bats. These "exceptions" do not undermine our argument of the

importance of frugivore selective pressure because the Dispersal Syndrome hypothesis does not

assume exclusiveness of frugivores. The assumption is, instead, that the strongest selective

pressure will come from the overall most efficient (sensu Schupp 1993) disperser. Because our

disperser data came from tree visits recorded in the field or reported in the literature, we are

likely considering only the most common disperser(s) of each fig species. Therefore, an

important assumption behind our analysis is that the most common seed disperser is the most

efficient seed disperser. Admittedly, this is an oversimplified view many processes important

to plant fitness occur after a frugivore consumes a fruit. We argue that the pattern we observed

holds despite this generalization, not because of it. A better test of the Dispersal Syndrome

hypothesis would consider dispersers that directly increase the fitness of the dispersed plant (i.e.,

increase seedling survival). We believe that if those data were used in a test similar to the one

we have presented here, the Dispersal Syndromes we described here would be even more

pronounced.









Another test of phylogeny's influence on co-variation in fruit traits was the correlation

between genetic relatedness and fruit appearance (i.e., scores on PC 1 derived from the Fruit

Trait PCA). Although closely related species tended to be more similar in appearance than non-

closely related species, less than 3% of the overall variation in fruit trait combination is

explained by genetic relatedness. Closely related species will always be more similar to each

other than to random species because, by definition, they shared a much longer history. What is

striking to us is that 97% of the similarity between figs must be explained by factors other than

phylogenetic relatedness. This lack of correlation can also be visually appreciated in Figure 4-3,

where the PC 1 score is mapped onto the phylogeny. Our results support the idea that fruit

appearance is more labile than it had been previously argued (Bremer and Eriksson 1992;

Herrera 1987; Jordano 1995) and, at least in figs, an important part of this variation appears to be

explained by differences among dispersers. We want to note that our estimates of the amount of

variation in fig traits that can be accounted for by frugivores may be conservative because EPCA

and the correlation between genetic relatedness with difference in trait characteristics they take

into account what is not shared by closely related species. As it has been previously noted by

Westoby and colleagues (1995) and also by Harvey and Pagel (1991), trait variation that is

inherited by common ancestry and is, thus, shared by closely related species, may also be

adaptive. This is because closely related species might tend to converge on similar adaptations

because they respond similarly to similar selective pressures because of their shared genes, and

because they may tend to colonize similar habitats and, hence, be subject to similar selective

pressures. Moreover, a characteristic that is shared by all members of the same clade might have

been fixed because of its high adaptive value. Many phylogenetic comparative methods give

priority to phylogeny over ecology to explain shared characteristics, but because we show results









from non-phylogenetic and phylogenetic comparisons, we allow the reader to draw their own

conclusions on the extent to which shared characteristics can be explained by either process

We suggest that Dispersal Syndromes should be thought of in terms of traits that make a

particular species of fruit more easy tofind by frugivores, and not more attractive to them.

Although a seemingly subtle semantic distinction, we argue that distinguishing among traits that

make a given fruit more or less attractive to a frugivore may lead to confusion and to false

expectations under the Dispersal Syndrome hypothesis. Imagine, for example, a frugivore given

a choice between fruits with different characteristics. Based on previous experience with other

fruits and the post-ingestional benefits particular to those fruits (e.g., nutritional and health-

related benefits, (Catoni et al. 2008; Goff and Klee 2006; Schaefer et al. 2008)), it may choose

the one that is more attractive in those terms. However, a more natural scenario is one in which

frugivores are not presented with choices but are instead searching for food to the best of their

ability, given the limitations imposed by their physiology, anatomy, and foraging behavior. If a

frugivore, while searching for, or feeding on a preferred fruit with high nutritional quality,

encounters by chance fruits that are easily accessible, it will most probably consume those fruits

too, even if those fruits are otherwise not attractive, just because other fruits may be scarce. In

this way, birds may consume bat-syndrome fruits, and vice versa. The more attractive fruit may

be preferred over the less attractive by the frugivore in a feeding trial during which the frugivore

must choose between the fruits. But whenever fruits are relatively uncommon, such preferences

likely disappear iffound both species will be equally consumed in nature. Instead, if a plant

highly benefits from the dispersal of its seeds by a specific type of frugivore, the selective

pressure on the plant will be to be found more readily by such a frugivore. With this distinction

in mind, the expectation of frugivore exclusiveness may be unrealistic.









It is worth noting that fruit color per se, as described by hue, was relatively unimportant in

discriminating between bird and bat fruits. However, two components of fruit color fruit

brightness and contrast against the background were important in distinguishing figs dispersed

by birds and bats. This result echoes results of previous studies that emphasize the role of

contrast for bird-dispersed fruits (Schaefer et al. 2006; Schmidt et al. 2004a). Note, however,

that in our dataset hue co-varied with brightness and may therefore still play a role (albeit a weak

one) in defining Disperser Syndromes. Fig species with green and yellow fruits tended to have

higher scores on PC 1 and EPC 1, and red and purple fruits tended to have low scores on these

axies, which were highly positively associated with brightness. Chromatic contrasts also tended

to be associated with color, as red and black fruits have higher contrast against green foliage

(Schmidt et al. 2004a). This suggests that these other dimensions of color are more directly

associated with bat or bird dispersal than reflected in our analysis. It would be interesting to

repeat this analysis using other quantitative measures of color, such as hue and chroma (defined

as the purity of the color, Endler 1990), to determine whether brightness and contrast still prevail

over fruit hue in discriminating between bird and bat-dispersed fruits.

Although bat-dispersal seemed to be associated with specific volatile compounds, we did

not test the evolutionary role of bats in selecting for those particular compounds. Because some

of these compounds are also present in some fig species when fig inflorescences are receptive to

pollinating wasps (Grison-Pige et al. 2002a; Grison-Pige et al. 2002b), it is possible that those

compounds may have been selected for by fig pollinators, not dispersers. Hence, until more

detailed studies on the evolution of these compounds are available, we suggest viewing these

compounds as an exaptation (Gould and Vrba 1982), not an adaptation, to frugivore attraction.









Many fruit traits may be subject to selective pressure from multiple factors (Cipollini and

Levey 1997; Mack 2000; Willson and Whelan 1990), but this does not undermine the importance

of seed dispersers as a selective force that may generate and/or maintain fleshy fruit diversity.

Any fruit trait may have several functions and therefore be subject to disparate selective

pressures, generating complex patterns of trait covariation. Even so, traits with a strong

influence on frugivore attraction will be subject to selection by frugivores. Although our data do

not allow us to conclude that any particular trait of bird or bat-dispersed figs originated from the

selective pressure of these seed dispersers (i.e., is an adaptation to seed dispersal), we believe

that the correlated evolution of so many traits that were predicted apriori to play a role in

frugivore attraction, strongly argues for an important role of seed dispersers on the evolution of

fruits as a whole and, hence, in explaining the diversity of fleshy fruits.




















EW


S20o Nartl.-r Ca-j 1:-


PAPUA NEW GUINEA
0 150 mI
0 150 km


Pacific Ocean


SMussou
. -N/rngo Manus
Cop Iln ^C: t ire n u ^,. t9
mnik, A a/ty s rBnsmarck m -r ay,'
Archipelago '-. L/hi Croup
,---..k / .,. ,,..,.o-% .
B ismarck Sea ar.
INDONESIA ,, ismal r Se

Lo e d.. A rang -- r

,i '" ,' l lr rn j Ar a
':"? '/. "^- l h ., ,.%. .. ,."
,, ", F .z, ,,."'tV,


Arafura Sea


AUSTR
@ s National Geographic Soc


...C -ca.r-j- 'n btlond Lur.kLa i" ,
S, n ..s ISLAND
P, D ---I .ldl.. I "rk
SMo re ibv 'y -0r,
ila .-te sour Is.
Ju 10, 5
A L I. ,. a r 1 A .. Louisiode
AL"IA t. am.arr A rch/~peago


:ietyj


145' E


Corai Sea


':. a 155'E


Figure 4-1. Map of Papua New Guinea in reference with a world map. Data were collected from
areas surrounding the city of Madang, mainly from Ohu village (not shown).


Equator 0


.- r


,-P F;;:
R
I~clrl
















I L

X A



x
Z x
A )


o o
000


A
A
A
A


x< A


x A
x ,A
X A


A A

x LA
x ,


xx
XX
Xx


PC 2

^%w^


x x
*'Vi


300 350 400 450 500 550 600 650 700 750

Wavelength (nm)





.igure 4-2. Loadings of the first three principal components (PC) on fruit reflectance spectra
from 300 to 750nm, measured at 5nm intervals. PC 1 represents variation in mean
brightness, PC 2 in green versus red coloration, and PC 3 in UV and blue coloration.


PC 3


i' 0
0
0
0


PC 1


i)
P _
0


>< A
AX
A x


A
A

A

A


A
z









Figure 4-3. Maximum Likelihood tree of 42 fig species. Nodes with bootstrap support value >
70 are marked with a star. The trait mapped in the phylogeny is the score of each
species on PC 1, the dispersal syndrome axis, using a maximum parsimony criterion.












F. adenosperma

oF adelpha

oF congesta

o Fbernaysti

SF ihspdloides

SF tematana

oF pacchyrachis


PC 1 score
M1-2.485 to -1.7002
M -1 7682 to -1.0474
M-1 0474 to -0.3286
W -0 32886 to 0 39020
S0 39020 to 1.1090
1.1090 to 1.82780
S1 82780 to 2.54680
W 2 54860 to 3.26540
E 3 28540 to 3.98420
W 3 98420 to 4.7030
S4.7030 to 5.42180


F septica


] F pungens



Iva



SF- variegata


m F benjamnina

i F. microcarpa


F conocephalifolia


iF tnctoria


macrofiyncha


odoardi


oF. itoana

oiF dammaropsis
08 I I I I I I I I
008 0.07 0_06 0_05 0_04 0_03 0_02 0_01 0.00









Figure 4-4. Parsimony reconstruction of dispersal mode. The number by the species names is the
score on PCA axis 1, which separates figs according to dispersal mode. Disperser
mode is shown in key.













I F. benjamin -2.338

n F. microcarpa -1.768

SF. hesperidiformis 0.192

I F. xylosycia -1.337

F. conocephalifolia -0.395

SF. phaeocyse -1.088

mF.copiosa 1.318

o F. wassa -0.783


m F subulata -1.421

m F. virgata -2.401


F. gul -1.709

a -:. macroryncha -2.483

L-o F trachypison -1.320

m F edelfeltil 0.301

hombroniana -1.045


i.862 F. dammaropsis 0.199

SF. itoana 2.862

a F. nodosa 3.603

m F semivestita 0.276

F. variegata 2.593

o F pungens 0.683

m F. botryocarpa 2.965

o F. septica -0.727

w* F. arfakensis 1.587

m F. bemaysii 2.633
Disperser
a F. hispidoides 2.150 Bat
Mixed
F. congesta 2.562 Bird

F. morobensis 4.703

SF'. pachyrrachis 2.633


o o 0 0 0 0
IllIIOOllll
oooooooooo
















A
A









A

A
AA
A
A


A
A
A





A A


-3 -2 -1 0 1 2 3

PC 1


(N -


CD -


-3 -2
-3 -2


-1 0

EPC1


Figure 4-5. Results of the Disperser PCA (A) and EPCA (B). Bird-dispersed figs are represented by a solid triangle and bat-dispersed

figs are represented by a white triangle.






















101


o-


B


A






A









A A
A
A
Aa
A
A A A
A
A
A


A A A
AA


A


1 2 3


















tin phawa
o o
prari ,1;
tra
amrdru


suL
ben
ela


ber
ter *
var
bot

0 *
Ito


hosla
mel
sep
0


mor
ade


OdO
Ioo -
pol dam
I I I I I I
-2 -1 0 1 2 3 4

PC 1


B
l pol
ri amlu ada mel WaI hes
tino horn *
mc pragul
o u ede dam
o sub mol
pha tra
0 con

oo

his
e
Ln pun
ter

I l:,'3r, ade
aade
Ln arf

ito

Smor
o
nod

-2 -1 0 1 2 3

EPC1


Figure 4-6. Plot of PC 1 vs 2 for the Fruit Trait PCA (A) and EPCA (B) for 42 species of figs. The importance of the original
variables in defining the PCs can be seen in Table 4-1. Species names are abbreviated as explained in Table 4-3. Red dots:
bird-dispersed species. Blue dots: bat-dispersed species. Yellow dots: mixed dispersal. Fig species without dots had no data
on disperser.













00 g0
0 0 0
0
0O

0 0 0 .
0 0 0
a hY~8~
0t 0 0
00 00
00 OAD AC
9 Qo c?I<
c^ a. ifr


o 0
0
0 0


0 00
0


O 0o


0.05


I I I I
0.10


I I 015
0.a5


Genetic distance




Figure 4-7. Correlation between genetic distance and difference in the score on the PCA axis 1
for all possible pairs among the 42 fig species for which fruit trait data were obtained.
They ellipse defines a 95% confidence region.


0 4
CL


0-
c
0




a














A B










A
& .


A

J1







X
Ax


X X
ax
& x


x Ax
x
x x



-5 -4 3 -2 .1 0 1 2 -3 -2 -1 0 1 2

LDF 1 LDF 1




Figure 4-8. Discriminant Functions plot based on (A) original data matrix for 30 Ficus species on presence/absence of 73 volatiles
compounds, and on (B) the same matrix randomized, keeping the column totals fixed. Symbols represent bird-dispersed
(A), bat-dispersed (A), bandicoot-dispersed (*), and mixed-dispersed (x).









Table 4-1. Loadings of variables for the Disperser PCA and EPCA.
PC1 PC2
Diameter 0.336 0.305
Softness -0.383 -0.331
PC1 0.205 0.236
PC2 -0.050 -0.416
PC3 0.153 -0.036
Odor 0.154 0.355
Peduncule length 0.346 0.061
Trunk 0.261 -0.529
Branch -0.366 0.314
Contrast against background -0.437 0.067
Distance from closest leaf 0.368 -0.235


Proportion Variance


0.378


0.172


Corresponds to Figure 4-5.
EPC1 EPC2
0.275 0.208
-0.266 -0.204
0.555 0.204
-0.171 0.452
-0.197 0.102
0.423 0.193
0.341 -0.107
0.082 -0.574
0.010 0.363
-0.316 0.083
0.273 -0.376


0.260


0.244


Table 4-2. Loadings of variables for the Fruit Trait PCA and EPCA. Corresponds to Figure 4-6.
PC1 PC2 EPC1 EPC2
Diameter 0.271 -0.461 0.515 0.013
Softness -0.308 0.402 -0.445 -0.122
PC1 0.170 -0.559 0.551 0.133
PC2 0.116 0.099 -0.292 0.020
PC3 0.171 0.174 -0.089 -0.012
Odor 0.218 -0.108 0.043 -0.191
Peduncule length 0.370 -0.077 0.301 -0.264
Trunk 0.339 0.312 0.012 -0.476
Branch -0.402 -0.300 0.090 0.451
Contrast against background -0.432 -0.116 -0.197 0.294
Distance from closest leaf 0.335 0.234 0.032 -0.587


Proportion of Variance


0.378


0.144


0.214


0.203










Table 4-3. Information on abbreviations used in figures, disperser recorded, and source of
information for dispersal mode.
Ficus species Abbreviation Disperser Source


F. adenosperma
F. adelpha
F. congesta
F. bernaysii
F. hispidioides
F. ternatana
F. pacchyrachis
F. morobensis
F. arfakensis
F. botryocarpa
F. septica
F. pungens
F. nodosa
F. semivestita
F. mollior
F. variegata
F. benjamin
F. microcarpa
F. elastica
F. drupacea
F. hesperidiiformis
F. xylosycia
F. conocephalifolia
F. phaeosyce
F. copiosa
F. wassa
F. subulata
F. virgata
F. tinctoria
F. gul
F. macrorhyncha
F. trachypison
F. ampelas
F. melinocarpa
F. odoardi
F. prasinicarpa
F. edelfeltii
F. hombroniana
F. polyantha
F. subtrinervia
F. itoana
F. dammaropsis


ada
ade
con
ber
his
ter
pac
mor
arf
bot
sep
pun
nod
sem
moll
var
ben
mic
ela
dru
hes
xyl
cof
pha
cop
was
sub
vir
tin
gul
mac
tra
amp
mel
odo
pra
ede
hom
pol
suv
ito
dam


no data
no data
bat
bat
bat
no data
bird
bat
bat
bat
mixed
mixed
mixed
bat
no data
bat
mixed
mixed
no data
no data
mixed
bird
bat
mixed
bat
mixed
bat
bat
no data
bird
bird
bird
no data
no data
no data
no data
bat
bat
no data
no data
bat
mixed


field work
field work
field work

field work
field work
field work
field work
Shanahan et al 2001
field work
Shanahan et al 2001
field work

field work
Shanahan et al 2001
Shanahan et al 2001


field work
Shanahan et al 2001
Shanahan et al 2001
Shanahan et al 2001
field work
Shanahan et al 2001
field work
field work

field work
field work
field work





field work
field work


field work
Shanahan et al 2001









CHAPTER 5
WITHIN SPECIES TEST OF THE DISPERSAL SYNDROME HYPOTHESIS: A STUDY OF
DIOECIOUS FIGS

Introduction

General Background and Goals

Ecological evidence strongly suggests that seed dispersal is important to fruiting plants, to

the point that some authors believe the question is no longer heither seed dispersal matters, but

when and where it matters most (Howe and Miriti 2000; 2004). If dispersed seeds are indeed

more likely to establish than undispersed seeds (Fragoso et al. 2003; Harms et al. 2000), then

individual trees that attract more seed dispersers should leave more offspring. Hence, seed

dispersers should be a strong selective force on fruit characteristics.

Evidence for the evolutionary role of seed dispersers on fruit traits, on the other hand, is

mixed. Many evolutionary ecologists have described patterns in co-occurring traits that they

attribute to different selective pressures of seed-dispersing vertebrates (Gautier-Hion et al. 1985;

Janson 1983; Lord 2004; Lord et al. 2002; Mack 1993). In particular, they champion the

existence of "dispersal syndromes" non-random combinations of fruit traits that seem to match

the foraging capabilities and overall characteristics of seed disperser taxa that forage in disparate

ways (McKey 1975; van der Pijl 1969). Other evolutionary ecologists have concluded that

dispersal syndromes do not exist or that differences among seed disperser taxa cannot account

for them (Fischer and Chapman 1993; Herrera 1987; 1989; 1992; 2002a; Jordano 1995).

The Dispersal Syndrome hypothesis posits that seed dispersers are important for plant

reproduction and that fruit traits that attract more seed dispersers are selectively favored. Hence,

one should find that dispersal syndromes have evolved independently in unrelated taxa that share

similar seed dispersers. Many preceding studies testing this hypothesis have been criticized

because they did not control for phylogenetic effects, instead considering each species as an









independent data point (e.g., Gautier-Hion et al. 1985; Janson 1983; Kalko et al. 1996; Korine et

al. 2000). Patterns of association between suites of fruit traits and taxa of seed dispersers were

interpreted as the result of co-evolution when, in fact, they could be more easily explained by the

process of "ecological fitting" -frugivores with preexisting preferences encounter fruits with

different traits, but will feed on the fruits that are most easy to find and accessible to them

(Felsenstein 1985; Harvey and Pagel 1991; Herrera 1987; 1989; Janzen 1980; 1985; Jordano

1995).

The genus Ficus offers an unusual opportunity to test the Dispersal Syndrome hypothesis

in a way that tightly controls for phylogenetic effects because some species produce

fundamentally different types of "fruits". (The fruiting structures of Ficus are termed syconia,

but for simplicity we refer to them as "fruits".) Functionally female trees produce "seed figs,"

which contain viable seeds. These individual trees therefore benefit from attraction of seed

dispersers. Functionally male trees produce "gall figs" whose seeds are all consumed by their

pollinators' larvae; they therefore do not benefit from attraction of seed dispersers.

Dumont et al (2004) showed that in one dioecious species, F. pungens, fruit bats preferred

seed figs over gall figs in feeding trials. Seed figs were larger and had more soluble

carbohydrates and fat, and less fiber than gall figs, characteristics that are all energetically

important for bats (Dumont et al. 2004). In this study, we expand on Dumont et al's (2004) work

by (1) quantifying frugivore preference for seed figs in natural conditions by videotaping 26 seed

figs and 20 gall fig trees of seven species (2) comparing the difference between gall and seed figs

of 27 dioecious fig species with respect to five traits that are known to be important in attraction

of seed dispersers: fruit size, color, odor, softness, and exposure on the tree (Gautier-Hion et al.

1985; Janson 1983; Kalko et al. 1996; Korine et al. 2000; Lomascolo et al. 2008; Lord 2004;









Lord et al. 2002; Mack 1993; Schaefer et al. 2006; van der Pijl 1969; Willson and Whelan 1990,

Chapter 4, this dissertation); and (3) quantifying at changes in those five traits during the

ripening process for gall and seed figs and for figs of monoecious species, to determine whether

male trees retain characters from unripe fruits when their fruits are fully mature. We also take a

closer look at UV reflectance because it has been shown to attract some seed dispersers

(Altshuler 2001). Because most hymenoptera also perceive UV reflectance, and because of the

tight mutualism of figs with pollinating wasps, we looked at UV reflectance at the flower, unripe,

and ripe stage of seed and gall figs. If UV has a role in frugivore attraction, ripe female figs

should show the highest reflectance. If, however, UV is mainly a signal to pollinators, syconia at

the flower stage should show the highest reflectance.

Assuming traits that attract seed dispersers are costly to the plant, we predict that only figs

with viable seeds (seed figs and figs of monoecious species) should invest in nutrient-rich

exocarp because only they would benefit from attracting seed-dispersing frugivores.

Specifically, seed figs should be larger, have more pigment and volatile compounds that make

them odorous, be softer (higher water to fiber ratio), and have longer peduncles to expose the

fruits away from the foliage and trunk. They should also reflect more in the UV range of the

spectrum. It has been hypothesized that fruit traits that attract frugivores, especially color and

odor, signal ripeness (Sumner and Mollon 2000b) i.e., that the appearance of fruit changes

when seeds are ready to be dispersed. We predict that seed and monoecious figs should show

large changes between the unripe and ripe stages, whereas male figs should show relatively

smaller changes between unripe and ripe stages.

Natural History of Fig Pollination and Seed Dispersal

Functional dioecy in figs is the result of one of the most tightly coevolved plant-pollinator

systems in nature (Janzen 1979; Weiblen 2002), in which typically one species of wasp









pollinates one species of fig. Fertilized female wasps carrying pollen enter the fig inflorescence

through a small hole ostiolele"), often losing their wings and part of their antennae (Wiebes

1979). They then pollinate the female flowers and lay an egg in each flower's ovary. The wasp

larvae develop in the fig ovaries inside the seed coat and consume the endosperm (Weiblen

2002). Once the wasps have completely matured, they exit the seed coat through a hole and mate

inside the fig. Females collect pollen from the now mature male flowers and exit the fig through

a hole made by the male wasps, which stay and die inside the fig where they emerged. Fertilized

female wasps carrying pollen then search for other figs at the correct stage of development to

continue their life cycle. In functionally dioecious fig species, female figs produce only female

flowers, and they all have long styles that prevent oviposition. Hence, a female fig acts as a

wasp trap because female wasps cannot lay eggs in them and cannot exit the fig to search for

another. Since female figs produce viable seeds and no wasps, they are called seed figs (Dumont

et al. 2004).

Functionally male figs, on the other hand, produce male and female flowers, but all female

flowers are short-styled and wasps lay eggs in every one of them, preventing development of any

viable seeds; they are called gall figs and produce wasps that disperse their pollen. In

monoecious figs, wasps oviposit in some flowers and not others, so that all syconia produce

some wasps and some viable seeds.

Fruits of gall and seed figs differ in many traits, including time of development, duration of

receptivity of inflorescences to pollinators and duration of fruit development (Patel and McKey

1998), season of ripening (Hill 1967; Patel and McKey 1998), shape and size (Hill 1967;

Lambert 1992), color (Dumont et al. 2004; Hill 1967; Lambert 1992), nutrient and fiber content

(Dumont et al. 2004, Weiblen, Lomascolo, and Dumont, unpublished data), and texture (Dumont









et al. 2004; Lambert 1992). These differences are presumably the result of optimizing the male

and female functions -- i.e., pollination success in gall figs and frugivore attraction in seed figs

(Dumont et al. 2004; Lambert 1992; Patel and McKey 1998).

In this study, we focus on the optimization of the female function of seed figs and explore

whether selection pressure by seed dispersers can account for differentiation of gall and seed figs

when each is at the ripe stage.

Methods

Study Site and General Specifications

We collected data from September 2004 until December 2005 in a lowland forest in

Madang Province, Papua New Guinea, at Ohu Village (1450 E, 50 S) (Fig. 4-1). To obtain data

on fruit color, color contrast against the background, odor, size, softness, and peduncle length for

male and female figs of each species we averaged records from between 3 and 8 individuals per

species. For each individual tree, between one and 20 fruits were measured and averaged for all

traits.

Quantification of Visits to Figs

We recorded consumption of female and male figs using two video cameras (Sony DCR-

HC40), which were set up 3-5m away from the tree and to minimize disturbance, visited only to

change tapes or batteries. Videotaping started at 6am and ended at 10:30am, and the again at

6:30pm until 11:00pm, to record both diurnal and nocturnal frugivores. Nocturnal recording was

done using an infrared light (Sony HVL-IRH2). We aimed one camera at the ground to record

terrestrial frugivores and one either at a branch or at the trunk, depending on where most of the

ripe figs were. We recorded 26 female trees (347 hours) and 20 male trees (343 hours) of seven

species (Ficus congesta, F. hispidoides, F. morobensis, F. pachyrrachis, F. pungens, F. septica,

and F. virgata).









Fruit Traits

Fruit color: We quantified fruit color using a USB2000 portable spectrometer (Ocean

Optics, Inc) and a PX-2 Pulsed Xenon light source, which took reliable readings between 300

and 740nm. This spectrum includes the range of wavelengths visible by humans, plus the

ultraviolet range. We scanned fruits using a sensor with six optical fibers illuminating the target

fruit and a seventh fiber that returned the reflected light to the spectrometer. The scanning angle

was fixed at 450 by using a black metal stand with a hole at that angle. The metal stand also

blocked external light. We fastened a non-UV filtering microscope slide to the opening of the

hole where the sensor was introduced to keep the distance between the end of the sensor and the

fruits constant. To obtain an average reflectance spectrum per fruit, we scanned each fruit three

times in three different spots on the fruit. We calculated reflectance as the proportion of a

certified white reflectance standard (Labsphere). The three spectra for each species were

averaged every 5nm. We obtained a quantitative measure of fruit color for each species by doing

a Principal Components Analysis (PCA), which summarizes the information for the complete

spectrum by finding linear combinations of the original variables (in this case, reflectance

intensity every 5 nm) (Clotfelter et al. 2007; Cuervo and Shine 2007; de Ayala et al. 2007;

Endler 1990). We did two separate PCAs: one to obtain color variables for the comparison

between ripe fruits of male vs. female figs, and a second one to obtain color variables for the

comparison between ripe and unripe male, female, and monoecious figs. In both cases, the first

principal component (PC) had high positive loadings for all wavelengths throughout the

spectrum, and hence it represents fruit brightness, the total amount of light reflected by a fruit

(Fig. 5-1). Therefore, light fruits had high scores on PC 1 and dark fruits had low scores. The

second PC represents reflectance in the yellow versus the UV-blue wavelengths because it had

negative loadings for wavelengths from 300 to around 500nm, and evenly positive loadings from









500 to 740nm. Therefore, yellow fruits had a high positive score on PC 2, whereas fruits

reflecting in the UV or blue part of the spectrum had high negative scores. The third principal

component represents reflectance in the green versus red and UV-blue range, because it had high

negative loadings of wavelengths between 420nm and 610nm. Thus, green fruits have a low

score on PC 3 and UV-blue and red (and, therefore, purple) fruits have a high score. These three

PCs explained 94% of the variance, and all had eigenvalues greater than one, which means that

each explains more variation than any one original variable alone. Although PC 4 and 5 also had

eigenvalues higher than one, they only explained 3 and 2% of the variation, respectively.

Because reflectance in all parts of the spectrum of interest was summarized by PC 1 through 3,

we considered that additional PCs did not contribute sufficiently to the color description of each

fig species. Hence, we use only scores on PC 1 through 3 in subsequent analyses.

We calculated color contrasts as the Euclidean distance between the color of a fruit and the

color of the structure against which it is exposed to the frugivore. We normalized spectra to the

same brightness by dividing the reflectance at each wavelength by the total reflectance for each

species. This represents contrasts due to the color of the fruits and not to their brightness, and

are called chromatic contrasts (Schmidt et al. 2004a).

D = [Qf(A) -Qb(A)]A2

Qf is the color spectrum of the fruit and Qb is the color spectrum of the background structure; X

is the wavelength in nm and the sum corresponds to the complete spectrum between 300-740nm.

We analyzed reflectance in the UV range of the spectrum separately because we were

interested in including syconia at the flower stage in the comparison. Mean reflectance between

350 and 400nm was calculated and compared between seed and gall figs at the flower, unripe,

and ripe stage.









Fruit odor: To collect odor, we collected fruits in the field, brought them to the lab, and

placed them inside plastic bags (Reynolds oven baking bags) that had a carbon filter at one end

to clean the air coming into the bag, and a filter with Super Q (Alltech) in it to adsorb the

volatiles emitted by the fruits. The filter was in turn connected to a vacuum pump (Welch,

model no. 2522B-01), which sucked the air through a carbon filter into the bag containing the

figs, and through the Super Q filter. All odor collections were run for four continuous hours, and

began no more than three hours after the fruits had been collected from the tree. To extract the

fruit volatiles from the Super Q filters, we injected methylene chloride through the filters, which

eluted the volatile compounds trapped in the Super Q. We used gas chromatography to separate

volatile compounds and mass spectrometry to identify some of them. For the purpose of this

study, however, only the total amount of volatile compounds is used as a measure of how

odorous the fruits were. Because the number of fruits we were able to collect and the size of the

fruits varied among trees, we had to standardize the total amount of volatiles by dividing the

amount of each volatile contained in the sample by the total surface area of the fruits used to

collect odor.

Fruit size, softness, and peduncle length: We measured fruit diameter with calipers to

the nearest 0.5 mm and used as a measure of overall fruit size. To measure fruit softness, an

arbitrary scale between 1 (hardest, fruit surface did not indent when pressed against fingers) to 4

(softer, fruit surface indented easily when pressure was applied to fruit between the fingers) was

determined by the authors and measured by the same person throughout the study. Peduncle

length (+ 0.5 mm) was measured between the base of the fruit and the point where the peduncle

attached to the branch or trunk.









Data Analysis

Frugivore visits to female versus male trees: To test whether female fruits were visited

more often than male fruits, we compared the number of visited (12) and not-visited (14) female

trees to the number of visited (1) and not-visited (19) male trees using a Chi-square test. Unless

a frugivore was specifically seen not to eat any fruits the entire time it was in a tree, any

frugivore that entered the tree was assumed to have been attracted to the tree and its presence

was recorded as a visit. A tree was recorded as visited if at least one frugivore entered the tree.

Difference in fruit traits of seed and gall figs: To identify the variables that distinguish gall

and seed figs, we performed a Discriminant Function analysis (DFA) on seed and gall figs of 25

species. Discriminant Function analysis finds the combination of traits that best discriminates

between the groups of interest, in this case, seed and gall figs (i.e., two groups).

Changes from unripe to ripe stage in monoecious and dioecious figs: To identify how

traits changed from the unripe to the ripe stages, and to determine whether these changes are

similar between figs with viable seeds (female and monoecious figs) and those without viable

seeds (male figs), we ran another DFA on the data including unripe monoecious, unripe seed,

unripe gall, ripe monoecious, ripe seed, and ripe gall figs (i.e., six groups). For both DFAs, all

variables were standardized to have a mean of zero and standard deviation of one because fruit

variables were measured in different scales and had differences of up to three orders of

magnitude. We tested the significance of the discrimination by calculating Wilk's Lambda,

which represents the proportion of the variance not explained by the predictor variables. This

test, however, does not inform which groups differ from which. To analyze which groups in

particular could be considered different, we performed an ANOVA on the Linear Discriminant

Function (LDF) that summarized most of the variation among the groups and performed a Tukey

test. We were particularly interested in comparing: 1. male fruits (not consumed by frugivores)









and female figs (consumed by frugivores); 2. female and monoecious fruits, which are both

reportedly consumed by frugivores; 3. ripe male and unripe male fruits, to see whether males

retain the characteristics of unripe fruits; 4. unripe male, female and monoecious figs, to see

whether the differentiation between fruits consumed by frugivores and those not consumed by

frugivores starts at ripening or whether these fruits are different from earlier on.

Another way we tested the predictive value of the linear discriminant functions is the

predictive classification of cases, which is done using Baye's Rule to calculate the posterior

probabilities of each case of being in each group, based on each case's discriminant score. The

higher the percentage of species that were well classified (e.g., known females correctly

classified as a female) the higher the predictive value of the LDF.

UV reflectance: We used Mann-Whitney U tests for pair-wise comparisons of mean UV

reflectance by figs of different sex and maturity stage. A sequential Bonferroni correction (Rice

1989) was used to calculate the actual critical p-value in multiple comparisons.

Results

Frugivore Visits to Female Versus Male Trees

Of the 26 female trees videotaped, 12 were visited by frugivores. Of the 20 male trees

videotaped, one was visited by a frugivore. Visits to female trees were significantly higher than

visits to male trees (x2 = 7.52, d.f. = 1, p = 0.006).

Difference in Fruit Traits of Seed and Gall Figs

One linear discriminant function (LDF) was sufficient to separate female from male figs as

two groups can only be separated in one dimension (Fig. 5-2). In order of importance, as

measured by the loadings of variables on the LDF (Table 5-1), female fruits were softer, had

more volatile compounds, were larger, darker, and contrary to predictions, reflected more in the

green part of the spectrum than did male figs. Female figs were also more contrasting against the









foliage and had shorter peduncles, but the loadings of these variables were quite low and, hence,

less important in discriminating between male and female figs. Discrimination of female and

male figs based on the fruit variables mentioned was statistically significant (Wilk's X = 0.339,

numerator d.f. = 8, denominator d.f = 41, p < 0.0001). When the scores of female and male

trees of the same species are compared, all except one species (F. dammaropsis) show a higher

score for females than for males (Fig. 5-3). Cross classification resulted in 94% of the cases

correctly classified.

The difference in fruit traits for gall and seed figs is present in two different clades that

seem to have evolved dioecy independently (Fig. 5-4)

Changes From Unripe to Ripe Stage in Monoecious and Dioecious Figs

The fruit variables included in this study can significantly discriminate between figs of

different sex and ripening stage (Wilk's X = 0.043, numerator d.f. = 40, denominator d.f. =

473.56, p < 0.0001) (Fig. 5-5). In general, monoecious and female ripe fruits scored highest on

LDF 1, followed by male ripe fruits, and then by all unripe fruits. According to the variable

loadings on LDF 1 (Table 5-2), this means that monoecious and female fruits are softer, darker,

and had higher reflectance in the red and UV-blue part of the spectrum than did male ripe fruits

and all unripe fruits. Monoecious and female ripe fruits were also slightly more odorous and had

longer peduncles, but these variables had low loadings and are therefore less important in

discriminating between the groups than are softness, brightness and green reflectance. The rest

of the variables had even lower loadings on LDF 1. A post-hoc Tukey test showed that unripe

figs from different sexes were not significantly different from each other, but were significantly

different from all ripe figs. Ripe gall figs differed significantly from all unripe figs and from ripe

seed and monoecious figs. Monoecious and seed figs did not differ from each other but did

differ from all other figs (Fig. 5-6).









UV Reflectance

The number of figs included in each comparison and mean reflectance for each sex and

maturity stage are summarized in Table 5-3. Against our prediction under the hypothesis that

UV signals frugivores, mean UV reflectance by ripe gall figs was significantly higher than UV

reflectance by ripe seed figs (W = 951, p = 0.00007). Reflectance of UV light does not seem to

have an exclusive role in signaling to wasps either, as reflectance did not differ between the

flower and the ripe stage for both sexes pooled together (W = 482, p = 0.87). No difference was

detected between gall and seed syconia at the flower stage, although this may be due to the small

sample size. UV reflectance was higher for unripe seed figs than for ripe seed figs (W = 1321, p

= 0.003), but this difference was not present between unripe gall figs (n = 26, x = 2.35) and ripe

gall figs (W = 292, p = 0.22).

Discussion

Seed dispersers were attracted to seed figs, which showed many of the characteristics that

were predicted to attract frugivores. As predicted, seed figs are softer than gall figs, which

probably facilitates frugivore consumption (Dumont 1999; Dumont and O'Neal 2004); more

odorous, which is especially important for mammalian frugivores (Hodgkison et al. 2007; Korine

and Kalko 2005); larger; and darker. Contrary to expectation, seed figs reflected more in the

green and less in the red and UV-blue range of the spectrum than did gall figs. This is puzzling,

at least at first, because only seed figs showed intense red, purple, or black coloration. A

potential explanation is that many red seed figs showed a dark red coloration, whereas the gall

figs for the same species showed a lighter red, almost pink, or an orange-red (e.g., F. itoana, F.

conocephalifolia, F. dammaropsis, F. macrorrhyncha, F. odoardi, F. phaeosyce, and F. wassa).

This results in the overall reflectance in the red spectrum being higher in the lighter gall figs than

in the darker seed figs. In other words, the main difference between these gall and seed figs is









not so much which part of the spectrum reflects most light (which is what the PC variables

represent), but instead the saturation of the red coloration. Saturation, also called chroma,

represents the "purity" of a color, and is measured by the steepness of the reflectance spectrum,

in this case, in the red part of the spectrum (Endler 1990). Saturation of a pink or orange fruit is

lower than that of a red fruit, but overall reflectance in the red part of the spectrum can still be

higher if the pink or orange fruit is brighter. Unfortunately, we did not quantify saturation.

An alternative explanation, non-exclusive of the previous one, is that the high positive

loading of PC 3 ("UV-blue-redness" in Table 1) on LDF 1 is more representative of reflectance

in the UV range than in the red range. When reflectance in the UV range is investigated in more

detail for all species in this study, gall figs reflect more in this part of the spectrum than female

figs. Moreover, unripe seed figs reflect more in the UV than ripe seed figs. These two results in

conjunction may result because some fruits have a waxy layer at the unripe stage that protects

them from desiccation, which often wears off as time passes and the fruits ripen (H. M. Schaefer,

pers. comm.). Because gall figs often detach from the tree as soon as the wasps emerge whereas

seed figs stay attached until a frugivore eats them or they start rotting, the gall figs we collected

may have been "younger" than the seed figs we collected and hence have retained more of the

waxy layer than did the seed figs. UV reflectance may also be selected for at the receptive stage

to attract pollinating wasps and be retained as the fig matures. This is supported by our results

showing that UV reflectance for gall and seed figs was similar at the flower stage, as both are

under strong selective pressure to attract pollinating wasps. However, this would not explain the

higher UV reflectance in gall figs compared to seed figs (Dumont et al. 2004). What is most

puzzling, however, is the lack of difference in UV reflectance between figs at the flower and ripe

stage. The trend in UV reflectance as the fig goes from the flower, to the unripe, and to the ripe









stage seems to differ between seed and gall figs (Table 5-3). The number of figs at the flower

stage is very low compared to the number of ripe figs and therefore the results for this

comparison are best considered inconclusive.

Although contrast against the background showed the predicted trend, i.e., seed figs

contrasting more against the foliage than gall figs, it did not appear as important as the

previously discussed variables in discriminating between the groups. This may be because, even

though some of the seed figs were red and purple (which have the highest contrast against the

foliage, Schmidt et al. 2004a), more than a third of the seed figs were green when ripe and had a

very low chromatic contrast against the foliage. Although peduncles varied highly among fig

species (Chapter 5, this dissertation), it was less important in differentiating between seed and

gall figs.

Figs with viable seeds were predicted to be under selective pressure by seed dispersers and

were indeed significantly different from all types of figs without viable seeds (i.e., gall and

unripe figs of both sexes). Figs with viable seeds were softer, darker, and reflected more in the

red and UV-blue part of the spectrum than all other figs, as represented from their higher scores

on LDF 1. To a lesser degree, they were also more odorous and larger, and had longer peduncles

than all other figs. Contrary to my predictions, gall figs were significantly discriminated from

unripe figs, which means that they do not necessarily retain all the characteristics of unripe figs.

Even though they do not have viable seeds and do not need to signal to seed dispersers, their

traits do change during the ripening process, although to a lesser extend than do figs with viable

seeds, as seen from their intermediate mean score on LDF 1 (Fig. 5-5).

A limitation of this study, and possibly the reason why some of the predictions did not

hold, is that we established predictions for the differences between seed and gall figs based on









the potential selective pressure of seed dispersers in general even though, for example, birds and

mammals may exert very different, and potentially even conflicting selective pressures on

different fig species (Chapter 4, this dissertation). It is especially revealing that the seed figs that

were exclusively consumed by birds (F. gul, F. macrorrhyncha, F. pachyrrachys, and F.

xylocisia) (Chapter 4, this dissertation) have a higher average score on LDF 1 than the mean for

all seed figs (5.21 vs. 3.61, respectively), whereas the mammal dispersed seed figs (F.

adenosperma, F. arfakensis, F. bernaysii, and F. botryocarpa, F. conocephalifolia, F. congesta,

F. copiosa, and F. hispidioides, F. itoana, F. morobensis, F. variegata, and F. virgata) (Chapter

4, this dissertation) average lower than the mean for all seed figs (3.47 vs. 3.61, respectively).

This apparent separation within the group of seed figs suggests that bird- and mammal-dispersed

figs have different characteristics. The within-group differences are insignificant compared to

the between group differences, as shown by the highly significant results for the DFAs in this

study. Comparing gall and seed figs while distinguishing between bird- and mammal-consumed

figs should be particularly enlightening. Birds are diurnal frugivores and have acute color vision

(Bowmaker et al. 1997; Bowmaker and Hunt 2006; Hart 2001) that they use to find fruit. On the

other hand, all frugivorous mammals in Papua New Guinea are either nocturnal or

crepuscular/nocturnal, and must therefore rely more heavily on senses other than vision to find

fruit. Odor, for example, has been reported as important for nocturnal mammals of other regions

when searching for fruit (Hodgkison et al. 2007; Korine and Kalko 2005; Thies et al. 1998).

A particularly interesting comparison to elucidate the difference in gall and seed figs for

mammal- and bird-dispersed figs would be to examine the difference in traits between gall and

seed figs for bird- and mammal-dispersed figs. Trait values can be subtracted between gall and

seed figs of each species, and the differences used as variables on a DFA for bird-dispersed









versus mammal-dispersed figs. we predict that in bat-dispersed figs, the major differences

between gall and seed figs will be the amount of volatile compounds produced, given that bats

are mostly crepuscular/nocturnal and cannot rely so much on visual cues, and in structures such

as the peduncle that expose the fruits away from the foliage and trunk, given that most of the

smaller flying foxes common in Papua New Guinea either grab the fruit in flight or land for a

few seconds on the fruit itself and then fly away with the fruit in their mouth (pers. obs). In bird-

dispersed figs, on the other hand, major differences in gall and seed figs should be in the

production of pigments that make the fruits visually conspicuous by contrasting against the

foliage, given that birds have such acute color vision (Bowmaker et al. 1997; Bowmaker and

Hunt 2006; Hart 2001; Schaefer et al. 2006; Schaefer et al. 2007; Schmidt et al. 2004b), and in

fruit softness, given that birds do not have teeth and can only use their bills to handle the fruits.

Although we do not know the proximate mechanism that determines the differences

between seed and gall figs, our results suggest that the ultimate mechanism is selective pressure

by seed dispersers. Seed figs show combinations of fruit traits that were predicted (Chapter 4,

this dissertation) and shown (this Chapter) to attract seed dispersers, while these attractive trait

combinations are not present in individuals that do not benefit from attracting seed dispersers.

This is true for both groups of figs that have independently evolved dioecy (Figure 5-4),

rendering further support to the adaptive significance of the difference between seed and gall

figs. Despite not distinguishing between frugivore guilds, the differences found among gall figs

and figs with viable seeds monoeciouss and seed figs) support the hypothesis that, in general,

frugivores are an important selective force in shaping the evolution of fruit traits in figs.
















xx
X


PC3


PC 1


x
x
x
x
X
x
X,-


Zp

ZI

PC2 A
AAAA4C


x
x
X

x
x
x


x A
>WX A

x
x x


x x
x x
x x
x
Xx

x x

XX'XO


I I I I I I I I I I
300 350 400 450 500 550 600 650 700 750

Wavelength(nm)


Figure 5-1. Loadings of the first three principal components (PC) on fruit reflectance spectra
from 300 to 750nm, measured at 5nm intervals. PC 1 represents variation in mean
brightness, PC 2 in green versus red coloration, and PC 3 in UV and blue coloration.


""I



















-3 -2 -1 0 1 2 3



B


-3 -2 -1 0
LDF 1


1 2 3


Figure 5-2. Plot of Discriminant Function analysis. (A) Female figs (B) Male figs


I -_...__1



























A A


III I I II II liii BI



IL 8 ~ u iL
IfL


Figure 5-3. Scores on the linear discriminant function (LDF 1) resulting from the discriminant
analysis of female and male figs. For each species, the paired scores on the LDF 1
for female (o) and male (A) individuals are plotted. Note that for all species, except
one (Ficus dammaropsis) female trees score higher than male trees. The horizontal
lines points to a value of zero on the LDF score, and is drawn simply as a point of
reference.









Figure 5-4. Maximum Likelihood phylogeny of the species included in the seed-gall fig
comparison. From the mapping of the breeding system, it is apparent that dioecy
evolved twice in this group of Ficus. This phylogeny was obtained as explained in
Chapter 4, and species that were not included in the analyses done in this chapter
were trimmed off the tree.












" F adenosperrnma

" F adelpha

SF congesta

" F bemaysii

SF hispidokles

SF morobensis

" F pacchyrachis












F variegata
" F. arfakensis

F botryocarpa

F_ septica








| F. espendiamis










SF. nocephalfolia

* F phaeosyce
F molliosr










* F wassa
F arsubulata
mF benjamin













SF macrocarpa
F. F elastica


F. drupaceae
F hespeFrdiiformis

F xylosycia






.F_ wassa









* F subulata
SF inctoa a

F dmmgul
l F macrortyncha

F. trachypison

F. odoardi

F prasmicarpa

. F edelreft

IF hombroniana

F. polyantha

F. subtimervia

F itoara

F dammaropsis


Breeding system

M Dioecious
I Monoecious












c(N -


UX X X 0
m x x

tY X








LDF










Figure 5-5. Discrimination among figs of different sex and maturity stages using a Discriminant
Function analysis. Symbols represent: monoecious ripe (o), female ripe (*), male ripe
(x), monoecious unripe (A), female unripe (m), and male unripe (*) figs.


















128
C
A 0 o

A A









-4 -2 0 2 4 6

LDF i


Figure 5-5. Discrimination among figs of different sex and maturity stages using a Discriminant
Function analysis. Symbols represent: monoecious ripe (o), female ripe (), male ripe
(x), monoecious unripe (A), female unripe (i), and male unripe (*) figs.


















128


X X
xx
* X

U x
*X

4

- x










p-value

BR-BU i I I < 0.0001
FR- BR 1.00

FU BR < 0.0001

MR BR 0.0009

MU BR < 0.0001
FR BU < 0.0001

FU BU 0.6932

MR- BU < 0.0001

MU BU 0.5441

FU FR < 0.0001

MR FR < 0.0001
MU FR < 0.0001

MR-FU i < 0.0001

MU FU -- 0.9996
MU-MR- < 0.0001

-5 0 5

Difference in mean score on LDF1

Figure 5-6. Pair-wise comparisons among fig groups of different sex and maturity stages using
Tukey tests. BR: monoecious ripe, BU: monoecious unripe, FR: female ripe, FU:
female unripe, MR: male ripe, MU: male unripe. The only groups whose difference is
not significantly different from zero were female and monoecious ripe figs, and all
pairwise comparisons of unripe figs.










Table 5-1. Discriminant Function Analysis of female versus male figs using eight fruit variables.
Variables with higher loadings (above 0.3) are marked with an asterisk
Variable Loading on LD 1
softness *-1.239
diameter *-0.551
Odor *-1.113
Contrast -0.171
PC 1 (brightness) *0.763
PC 2 yellownessss") *0.338
PC 3 ("UV-blue-redness") *0.366
Peduncle length 0.182



Table 5-2. Discriminant Function Analysis of female, monoecious and male figs at different
maturity stages using eight fruit variables. Variables with higher loadings (above 0.3)
are marked with an asterisk
Variable Loading on LD 1 Loading on LD 2
softness *3.026 0.102
diameter 0.199 0.083
Odor 0.202 -0.248
Contrast -0.120 *-1.017
PC 1 (brightness) *-0.321 -0.027
PC 2 yellownessss") -0.145 *-0.337
PC 3 ("UV-blue-redness") *0.360 0.056
Peduncle length 0.261 -0.044

Proportion of variation 0.92 0.05



Table 5-3. Mean UV reflectance for fig species of different sex and maturity stages
Sex Maturity stage Observations Mean UV reflectance
Female flower 7 1.09
Female unripe 44 1.66
Female ripe 44 1.06
Male flower 6 2.39
Male unripe 26 2.35
Male ripe 28 2.04









CHAPTER 6

GENERAL CONCLUSIONS


This study contributes to the advancement of the field of plant-frugivore interactions

because we help resolve a long standing paradox: although animals facilitate plant reproduction

through seed dispersal, they do not seem to shape plant evolution. We show that a large portion

of the variation in fruit characteristics is unexplained by phylogeny, and that the correlated

evolution of seven traits point to seed dispersers as the main selective force explaining this

variation. Our study's success in detecting the evolutionary influence of frugivores on fruit traits

likely results from our ability to avoid these limitations of earlier work by: 1. focusing at the

species level, where most evolutionary meaningful variation occurs; 2. tightly controlling for

phylogenetic effects with a phylogeny, the closest approximation to the true evolutionary history

of the taxa under study, and by doing a within-species study; 3. studying groups of species on

both local and geographic scales; 4. using quantitative measures of fruit color and odor,

important signals to frugivores that had previously been ignored or mismeasured, and taking into

account the frugivore's perspective on color; 5. incorporating a "holistic" view of Dispersal

Syndromes by analyzing all fruit variables together, which we believe better reflects how

dispersers search for and assess fruit; and 6. defining Dispersal Syndromes more narrowly and

grouping dispersers according to their similar potential to select for fruit characteristics.

One method to test for natural selection in the field is to search for correlated evolution

among traits that, in combination, offer higher fitness to the individual bearing them than any

other combination. Convergence of multiple traits of distantly related species form strong

evidence that the traits do not have a neutral adaptive effect (Endler 1986). We have shown that

dispersal syndromes are suites of traits that have converged in distantly related species.









Moreover, we show that these suites of traits can be predicted based on the behavioral and

anatomical characteristics of the animals that feed on the fruits. We generate testable predictions

of the potential frugivores of different fig species that can be used to test our results. More

importantly, these predictions can be generalized to other plant taxa, and to other plant-animal

interactions, such as pollination, for which fruit and floral traits can be quantified.

Although arguably to a lesser extent, students of pollinator-plant interactions also ponder

the question of how specialized this interaction really is, and whether pollinators affect the

evolution of flowering plants (Bluthgen et al. 2007; Johnson and Steiner 2000; Vazquez and

Aizen 2004; Waser et al. 1996). Fruit and flower traits have been hypothesized to act as signals

in plant-animal communication (Chittka and Raine 2006; Goff and Klee 2006; Schaefer et al.

2008; Schaefer and Schaefer 2007; Schaefer et al. 2004; Sumner and Mollon 2000b) and

therefore their importance must be assessed from the perspective of the animal to which the

signal is targeted. That an animal differentially perceives signals directed to it versus directed to

other animals is an important assumption of plant-animal interactions studies. Pollinators, for

example, seem to respond to color cues to locate flowers (Bradshaw and Schemske 2003; Chittka

and Raine 2006; Gumbert et al. 1999). In our test of this assumption in fruit to frugivore

signaling, we show that the color of bird and primate-dispersed fruits can be perceived as

different from the bird's and primate's perspective. This result confirms that fruit color can be

used as a signal by plants that may differentially benefit from bird or primate consumption of

their fruits. A similar assessment of other fruit traits, such as odor signals, would be needed, but

might be immensely more challenging, as assessing perception of odor by animals would likely

require many more "dimensions" than for color (Mamlouk and Martinetz 2004), and these

dimensions are far from being completely identified (Chittka and Brockmann 2005).









With the exception of a few thoroughly studied cases of tight coevolution between plants

and their mutualistic pollinators (Janzen 1979; Smith et al. 2008; Thompson and Pellmyr 1992)

or seed dispersers (Amico and Aizen 2000; Reid 1991) mutualistic plant-animal interactions

seem to be quite generalized (but see Fenster et al. 2004; Fuentes 1995; Herrera 1985; 1998;

Johnson and Steiner 2000; Waser et al. 1996). Moreover, it has been argued that the distribution

in plant-animal mutalistic networks can be reasonably well explained by stochastic processes

(Vazquez 2005). However, the general observation that pollinator-plant systems are more

specialized relative to frugivore-plant systems seems to hold (Bluthgen et al. 2007). The

potential for tight coevolution between one species of seed disperser and one species of plant is

low because, among other things, most plants benefit from the dispersal services of many

different frugivores and, any one frugivore consumes a wide range of fruits. Also, generation

times and historical species turnover rates are very different between frugivores and plants

(Herrera 1985). Moreover, populations of the same species interact with different species along

their geographic range (Gomulkiewicz et al. 2000; Rey et al. 2006; Thompson 1994; 1999). For

all these reasons, it has been argued that frugivores and plants undergo, at best, diffuse

coevolution which occurs when groups of species interact with each other (Janzen 1980).

We believe diffuse coevolution should not be viewed as a hopelessly messy evolutionary

relationship. If one can identify groups of species that interact in a consistent and distinctive

way (e.g., a guild of seed dispersers that exerts similar selective pressure on the plant), and if one

objectively defines what traits these interactants are most likely to exert selection upon, then one

can make specific predictions on the potential for coevolution. We have focused only on one

side of the interaction between frugivores and plants -- the role of seed dispersers in the

evolution of fruit traits. The other side -- the role of plants in the evolution of frugivore anatomy









and physiology of visual, and olfactory systems, or on its foraging behavior -- remains to be

tested (Dominy et al. 2003; Osorio and Vorobyev 1996; Peichl 2005; Surridge et al. 2003).

It can be argued that modern day frugivores may not represent the fauna that historically

interacted and exerted selection on figs. Figs have almost certainly interacted with frugivores that

differ from today's frugivores, at least in body size (e.g., extinct New World megafauna, Janzen

and Martin 1982). However, figs as we know them today are likely to have interacted with

species of bats and birds that are similar to those seen today. Both the origin of modern-day

Ficus and Megachiroptera seem to trace back to the same historic epoch and region. The origin

and diversification of the fig family, Moraceae, is most parsimoniously assumed to be in Eurasia,

from where it would have dispersed to the Southern hemisphere through land connections

(Zerega et al. 2005). Modem day figs diversified some time during the Eocene (Zerega et al.

2005) and this is likely to have occurred in the Asian-Australasian region as this is the center of

distribution for many fig sections, and where the highest fig diversity is found, worldwide (Berg

1989). Megachiroptera are also thought to have originated in the South East Asian-Melanesian

region (Giannini and Simmons 2003). Fully flighted bats are also known from the Eocene and,

although no fossils remains of Megachiroptera have been found in the region, recent ancestor of

their closely related clade, the Microchiroptera, are known from the Australia-New Guinea

region (Bonaccorso 1998). Also, the oldest Megachiroptera fossil found was in Thailand

(Ducrocq et al. 1993). The fig species found in New Guinea likely interacted with modern day

birds since their origin in Eurasia, as birds evolved much earlier than figs, at least 150 mya and

were widespread. However, they likely only came into contact with the birds with which they

interact today only in the Pleistocene through a land bridge that connected the island with

Australia during the last Ice Age. The island of New Guinea is comprised of a section of the









northern part of the Australian continent to the South, and a series of islands that collided with

the Australian continent in the past 40 m.y. to the North (Audley-Charles 1991). Although most

of the flora of New Guinea have closest affinities with Southeast Asia, most New Guinean birds,

are of Australian origin (Beehler et al. 1986). Because the characteristics on which we based the

potential for selection by birds on figs were already present in earlier birds, we consider it

unlikely that selection by birds changed significantly in the evolutionary history of figs.

Returning to the first point made in the Introduction, just as it has been argued that

incorporating positive interactions can drastically change core concepts in ecological theory

(Bruno et al. 2003), we show that, at least one positive interaction that between frugivores and

fruiting plants can help explain an important evolutionary process the diversification of

fleshy fruits. We argue that studies on positive interactions can help us understand processes that

are important for the origin and maintenance of biodiversity.

In our humble quest to identify the role of phylogeny versus frugivores in shaping the

evolution of fruit characteristics in figs, we hope to have contributed, even if a little, to the

understanding of the evolutionary relationships in the "entangled bank, covered with plants of

many kinds" interacting with different animals "dependent on each other in so complex a

manner" (Darwin 1859).









LIST OF REFERENCES


Adams SB, Hamel PB, Connor K, Burke B, Gardiner ES, Wise D (2007) Potential roles of fish,
birds, and water in swamp privet (Forestiera acuminata) seed dispersal. Southeastern
Naturalist 6:669-682

Agosta SJ (2006) On ecological fitting, plant-insect associations, herbivore host shifts, and host
plant selection. Oikos 114:556-565

Alcantara JM, Rey PJ (2003) Conflicting selection pressures on seed size: evolutionary ecology
of fruit size in a bird-dispersed tree, Olea europaea. Journal of Evolutionary Biology
16:1168-1176

Alfaro ME, Zoller S, Lutzoni F (2003) Bayes or bootstrap? A simulation study comparing the
performance of Bayesian Markov chain Monte Carlo sampling and bootstrapping in
assessing phylogenetic confidence. Molecular Biology and Evolution 20:255-266

Altshuler DL (2001) Ultraviolet reflectance in fruits, ambient light composition and fruit
removal in a tropical forest. Evolutionary Ecology Research 3:767-778

Altshuler DL (2003) Flower color, hummingbird pollination, and habitat irradiance in four
neotropical forests. Biotropica 35:344-355

Amico G, Aizen MA (2000) Ecology Mistletoe seed dispersal by a marsupial. Nature 408:929-
930

Audley-Charles MG (1991) Tectonics of the New-Guinea Area. Annual Review of Earth and
Planetary Sciences 19:17-41

Beehler BM, Pratt TK, Zimmerman DA (1986) Birds of New Guinea. Princeton University
Press, Princeton, New Jersey

Bennett ATD, Cuthill IC (1994) Ultraviolet vision in birds What is its function? Vision
Research 34:1471-1478

Bennett ATD, Cuthill IC, Norris KJ (1994) Sexual selection and the mismeasure of color.
American Naturalist 144:848-860

Bennett ATD, Thery M (2007) Avian color vision and coloration: Multidisciplinary evolutionary
biology. American Naturalist 169:S1-S6

Ben-Tal Y, King RW (1997) Environmental factors involved in colouration of flowers of
Kangaroo Paw. Scientia Horticulturae 72:35-48

Berg CC (1989) Classification and Distribution of Ficus. Experientia 45:605-611









Bertness MD, Leonard GH (1997) The role of positive interactions in communities: Lessons
from intertidal habitats. Ecology 78:1976-1989

Bertness MD, Leonard GH, Levine JM, Schmidt PR, Ingraham AO (1999) Testing the relative
contribution of positive and negative interactions in rocky intertidal communities.
Ecology 80:2711-2726

Bianconi GV, Mikich SB, Teixeira SD, Maia B (2007) Attraction of fruit-eating bats with
essential oils of fruits: A potential tool for forest restoration. Biotropica 39:136-140

Bluthgen N, Menzel F, Hovestadt T, Fiala B (2007) Specialization, constraints, and conflicting
interests in mutualistic networks. Current Biology 17:341-346

Bolmgren K, Eriksson O (2005) Fleshy fruits origins, niche shifts, and diversification. Oikos
109:255-272

Bonaccorso FJ (1998) Bats of Papua New Guinea. Conservation International, Washington, DC

Bowmaker JK, Heath LA, Wilkie SE, Hunt DM (1997) Visual pigments and oil droplets from six
classes of photoreceptor in the retinas of birds. Vision Research 37:2183-2194

Bowmaker JK, Hunt DM (2006) Evolution of vertebrate visual pigments. Current Biology
16:R484-R489

Bradshaw HD, Schemske DW (2003) Allele substitution at a flower colour locus produces a
pollinator shift in monkeyflowers. Nature 426:176-178

Bremer B, Eriksson O (1992) Evolution of Fruit Characteristics and Dispersal Modes in the
Tropical Family Rubiaceae. Biological Journal of the Linnean Society 47:79-95

Bruno JF, Stachowicz JJ, Bertness MD (2003) Inclusion of facilitation into ecological theory.
Trends in Ecology & Evolution 18:119-125

Burnham KP, Anderson D (2003) Model selection and multi-model inference., 2nd edn.
Springer, New York, NY

Burns KC, Dalen JL (2002) Foliage color contrasts and adaptive fruit color variation in a bird-
dispersed plant community. Oikos 96:463-469

Cardoso da Silva JM, Tabarelli M (2000) Tree species impoverishment and the future flora of the
Atlantic forest of Northeast Brazil. Nature 404:72-74

Catoni C, Schaefer HM, Peters A (2008) Fruit for health: the effect of flavonoids on humoral
immune response and food selection in a frugivorous bird. Functional Ecology. In press









Chapman CA, Onderdonk DA (1998) Forests without primates: Primate/plant codependency.
American Journal of Primatology 45:127-141

Chittka L, Brockmann A (2005) Perception space The final frontier. Plos Biology 3:564-568

Chittka L, Raine NE (2006) Recognition of flowers by pollinators. Current Opinion in Plant
Biology 9:428-435

Cipollini ML, Bohs LA, Mink K, Paulk E, Bohning-Gaese K (2002) Secondary metabolites of
ripe fleshy fruits: ecology and phylogeny in the genus Solanum. In: Levey DJ, Silva WR,
Galetti M (eds) Seed dispersal and Frugivory: Ecology, Evolution and Conservation.
CABI Publishing, Wallingford, pp 111-128

Cipollini ML, Levey DJ (1991) Why some fruits are green when they are ripe carbon balance in
fleshy fruits. Oecologia 88:371-377

Cipollini ML, Levey DJ (1997) Secondary metabolites of fleshy vertebrate-dispersed fruits:
Adaptive hypotheses and implications for seed dispersal. American Naturalist 150:346-
372

Clark CJ, Poulsen JR, Levey DJ, Osenberg CW (2007) Are plant populations seed limited? A
critique and meta-analysis of seed addition experiments. American Naturalist 170:128-
142

Clark CJ, Poulsen JR, Parker VT (2001) The role of arboreal seed dispersal groups on the seed
rain of a lowland tropical forest. Biotropica 33:606-620

Clark JS et al. (1999) Interpreting recruitment limitation in forests. American Journal of Botany
86:1-16

Clark L (1991) Odor detection thresholds in tree swallows and Cedar Waxwings. Auk 108:177-
180

Clotfelter ED, Ardia DR, McGraw KJ (2007) Red fish, blue fish: trade-offs between
pigmentation and immunity in Betta splendens. Behavioral Ecology 18:1139-1145

Cordeiro NJ, Howe HF (2001) Low recruitment of trees dispersed by animals in African forest
fragments. Conservation Biology 15:1733-1741

Cordeiro NJ, Howe HF (2003) Forest fragmentation severs mutualism between seed dispersers
and an endemic African tree. Proceedings of the National Academy of Sciences of the
United States of America 100:14052-14056

Croat TB (1978) Flora of Barro Colorado Island. Standford University Press, Standford,
California









Cuervo JJ, Shine R (2007) Hues of a dragon's belly: morphological correlates of ventral
coloration in water dragons. Journal of Zoology 273:298-304

Darwin C (1859) On the origin of species by means of natural selection or the preservation of
favoured races in the struggle of life, London

Das D, Wilkie SE, Hunt DM, Bowmaker JK (1999) Visual pigments and oil droplets in the retina
of a passerine bird, the canary Serinus canaria: microspectrophotometry and opsin
sequences. Vision Research 39:2801-2815

de Ayala RM, Saino N, Moller AP, Anselmi C (2007) Mouth coloration of nestlings covaries
with offspring quality and influences parental feeding behavior. Behavioral Ecology
18:526-534

Dominy NJ (2004) Fruits, fingers, and fermentation: The sensory cues available to foraging
primates. Integrative and Comparative Biology 44:295-303

Dominy NJ, Lucas PW, Osorio D, Yamashita N (2001) The sensory ecology of primate food
perception. Evolutionary Anthropology 10:171-186

Dominy NJ, Svenning JC, Li WH (2003) Historical contingency in the evolution of primate color
vision. Journal of Human Evolution 44:25-45

Ducrocq S, Jaeger JJ, Sige B (1993) Un megachiroptere dans l'Eocene superieur de Thaiilande:
incidence dans la discussion phylogenique du group. N. Jahrb. Geol. Palaotol. Mh.
9:561-575

Dumont ER (1999) The effect of food hardness on feeding behaviour in frugivorous bats
(Phyllostomidae): an experimental study. Journal of Zoology 248:219-229

Dumont ER, O'Neal R (2004) Food hardness and feeding behavior in old world fruit bats
(Pteropodidae). Journal of Mammalogy 85:8-14

Dumont ER, Weiblen GD, Winkelmann JR (2004) Preferences of fig wasps and fruit bats for
figs of functionally dioecious Ficus pungens. Journal of Tropical Ecology 20:233-238

Dyreson E, Maddison WP (2006) Rhetenor package for morphometrics In Mesquite 1.11

Elangovan V, Marimuthu G, Kunz TH (2001) Temporal patterns of resource use by the short-
nosed fruit bat, Cynopterus sphinx (Megachiroptera : Pteropodidae). Journal of
Mammalogy 82:161-165

Endler JA (1986) Natural selection in the wild. Princeton University Press, Princeton

Endler JA (1990) On the Measurement and Classification of Color in Studies of Animal Color
Patterns. Biological Journal of the Linnean Society 41:315-352










Endler JA (1992) Signals, Signal Conditions, and the Direction of Evolution. American
Naturalist 139:S125-S153

Evans CS (2006) Accessory chemosignaling mechanisms in primates. American Journal of
Primatology 68:525-544

Felsenstein J (1985) Phylogenies and the Comparative Method. American Naturalist 125:1-15

Fenster CB, Armbruster WS, Wilson P, Dudash MR, Thomson JD (2004) Pollination syndromes
and floral specialization. Annual Review of Ecology Evolution and Systematics 35:375-
403

Fischer KE, Chapman CA (1993) Frugivores and fruit syndromes Differences in patterns at the
genus and species level. Oikos 66:472-482

Fleming TH, Breitwisch R, Whitesides GH (1987) Patterns of tropical vertebrate frugivore
diversity. Annual Review of Ecology and Systematics 18:91-109

Fragoso JMV, Silvius KM, Correa JA (2003) Long-distance seed dispersal by tapirs increases
seed survival and aggregates tropical trees. Ecology 84:1998-2006

Fuentes M (1995) How specialized are fruit-bird interactions? Overlap of frugivore assemblages
within and between plant species. Oikos 74:324-330

Gautier-Hion A et al. (1985) Fruit characters as a basis of fruit choice and seed dispersal in a
tropical forest vertebrate community. Oecologia 65:324-337

Giannini NP, Simmons NB (2003) A phylogeny of megachiropteran bats (Mammalia :
Chiroptera : Pteropodidae) based on direct optimization analysis of one nuclear and four
mitochondrial genes. Cladistics-the International Journal of the Willi Hennig Society
19:496-511

Godinez-Alvarez H, Valiente-Banuet A, Rojas-Martinez A (2002) The role of seed dispersers in
the population dynamics of the columnar cactus Neobuxbaumia tetetzo. Ecology
83:2617-2629

Goff SA, Klee HJ (2006) Plant volatile compounds: Sensory cues for health and nutritional
value? Science 311:815-819

Gomulkiewicz R, Thompson JN, Holt RD, Nuismer SL, Hochberg ME (2000) Hot spots, cold
spots, and the geographic mosaic theory of coevolution. American Naturalist 156:156-
174

Gould SJ, Vrba ES (1982) Exaptation a Missing Term in the Science of Form. Paleobiology
8:4-15










Grison-Pige L, Bessiere JM, Hossaert-McKey M (2002a) Specific attraction of fig-pollinating
wasps: Role of volatile compounds released by tropical figs. Journal of Chemical
Ecology 28:283-295

Grison-Pige L, Hossaert-McKey M, Greeff JM, Bessiere JM (2002b) Fig volatile compounds a
first comparative study. Phytochemistry 61:61-71

Gumbert A, Kunze J, Chittka L (1999) Floral colour diversity in plant communities, bee colour
space and a null model. Proceedings of the Royal Society of London Series B-Biological
Sciences 266:1711-1716

Hall E (1997) Biological sequence alignment editor for Windows 95/98/NT. In
http://www.mbio.ncsu.edu/BioEdit/bioedit.html

Halpern BS, Silliman BR, Olden JD, Bruno JP, Bertness MD (2007) Incorporating positive
interactions in aquatic restoration and conservation. Frontiers in Ecology and the
Environment 5:153-160

Hampe A, Bairlein F (2000) Modified dispersal-related traits in disjunct populations of bird-
dispersed Frangula alnus (Rhamnaceae): a result of its Quaternary distribution shifts?
Ecography 23:603-613

Harms KE, Wright SJ, Calderon O, Hernandez A, Herre EA (2000) Pervasive density-dependent
recruitment enhances seedling diversity in a tropical forest. Nature 404:493-495

Hart NS (2001) The visual ecology of avian photoreceptors. Progress in Retinal and Eye
Research 20:675-703

Harvey PH, Pagel MD (1991) The comparative method in evolutionary biology. Oxford
University Press, Oxford

Herrera CM (1985) Determinants of Plant-Animal Coevolution the Case of Mutualistic
Dispersal of Seeds by Vertebrates. Oikos 44:132-141

Herrera CM (1987) Vertebrate-dispersed plants of the Iberian Peninsula A study of fruit
characteristics. Ecological Monographs 57:305-331

Herrera CM (1989) Seed dispersal by animals: a role in angiosperm diversification? American
Naturalist 133:309-322

Herrera CM (1992) Interspecific variation in fruit shape: allometry, phylogeny, and adaptation
to dispersal agents. Ecology 73:1832-1841

Herrera CM (1998) Long-term dynamics of Mediterranean frugivorous birds and fleshy fruits: A
12-year study. Ecological Monographs 68:511-538










Herrera CM (2002a) Correlated evolution of fruit and leaf size in bird-dispersed plants: species-
level variance in fruit traits explained a bit further? Oikos 97:426-432

Herrera CM (2002b) Seed dispersal by vertebrates. In: Herrera CM, Pellmyr O (eds) Plant-
Animal Interactions. An Evolutionary Approach. Blackwell Science Ltd., Oxford, pp
185-208

Herrera CM, Jordano P, Lopezsoria L, Amat JA (1994) Recruitment of a Mast-Fruiting, Bird-
Dispersed Tree Bridging Frugivore Activity and Seedling Establishment. Ecological
Monographs 64:315-344

Heymann EW (2006) The neglected sense-olfaction in primate behavior, ecology, and evolution.
American Journal of Primatology 68:519-524

Hill DS (1967) Figs (Ficus Spp) and Fig-Wasps (Chalcidoidea). Journal of Natural History
1:413-&

Hillis DM, Bull JJ (1993) An empirical test of bootstrapping as a method for assessing
confidence in phylogenetic analysis. Systematic Biology 42:182-192

Hjerpe J, Hedenas H, Elmqvist T (2001) Tropical rain forest recovery from cyclone damage and
fire in Samoa. Biotropica 33:249-259

Hodgkison R et al. (2007) Chemical ecology of fruit bat foraging behavior in relation to the fruit
odors of two species of paleotropical bat-dispersed figs (Ficus hispida and Ficus
scortechinii). Journal of Chemical Ecology 33:2097-2110

Howe HF (1986) Seed dispersal by fruit-eating birds and mammals. In: Murray D (ed) Seed
dispersal. Academic Press, New York, pp 123-189

Howe HF, Miriti MN (2000) No question: seed dispersal matters. Trends in Ecology &
Evolution 15:434-436

Howe HF, Miriti MN (2004) When seed dispersal matters. Bioscience 54:651-660
Howe HF, Schupp EW, Westley LC (1985) Early Consequences of Seed Dispersal for a
Neotropical Tree (Virola-Surinamensis). Ecology 66:781-791

Hsiao C, Jacobs SWL, Chatterton NJ, Asay KH (1999) A molecular phylogeny of the grass
family (Poaceae) based on the sequences of nuclear ribosomal DNA (ITS). Australian
Systematic Botany 11:667-688

Huelsenbeck JP, Ronquist F (2001) MRBAYES: Bayesian inference of phylogenetic trees.
Bioinformatics 17:754-755









Hurtt GC, Pacala SW (1995) The Consequences of Recruitment Limitation Reconciling
Chance, History and Competitive Differences between Plants. Journal of Theoretical
Biology 176:1-12

Janson CH (1983) Adaptation of fruit morphology to dispersal agents in a Neotropical forest.
Science 219:187-189

Janzen DH (1970) Herbivores and Number of Tree Species in Tropical Forests. American
Naturalist 104:501-&

Janzen DH (1979) How to be a fig. Annual Review of Ecology and Systematics 10:13-51

Janzen DH (1980) When is it coevolution? Evolution 34:611-612

Janzen DH (1985) On ecological fitting. Oikos 45:308-310

Janzen DH, Martin PS (1982) Neotropical Anachronisms the Fruits the Gomphotheres Ate.
Science 215:19-27

Johnson SD, Steiner KE (2000) Generalization versus specialization in plant pollination systems.
Trends in Ecology & Evolution 15:140-143

Jordano P (1995) Angiosperm fleshy fruits and seed dispersers A comparative analysis of
adaptation and constraints in plant-animal interactions. American Naturalist 145:163-191

Jordano P (2000) Fruits and frugivory. In: Fenner M (ed) Seeds. The Ecology of Regeneration in
Plant Communities. CABI Publishing, New York, pp 125-165

Jousselin E, Rasplus JY, Kjellberg F (2003) Convergence and coevolution in a mutualism:
Evidence from a molecular phylogeny of Ficus. Evolution 57:1255-1269

Jurik TW (1983) Reproductive Effort and CO2 Dynamics of Wild Strawberry Populations.
Ecology 64:1329-1342

Kalko EKV, Herre EA, Handley CO (1996) Relation of fig fruit characteristics to fruit-eating
bats in the New and Old World tropics. Journal of Biogeography 23:565-576

Kelber A, Vorobyev M, Osorio D (2003) Animal colour vision behavioral tests and
physiological concepts. Biological Reviews 78:81-118

Knight RS, Siegfried WR (1983) Inter-relationships between type, size and colour of fruits and
dispersal in Southern African trees. Oecologia 56:405-412

Korine C, Kalko EKV (2005) Fruit detection and discrimination by small fruit-eating bats
(Phyllostomidae): echolocation call design and olfaction. Behavioral Ecology and
Sociobiology 59:12-23










Korine C, Kalko EKV, Herre EA (2000) Fruit characteristics and factors affecting fruit removal
in a Panamanian community of strangler figs. Oecologia 123:560-568

Laman TG, Weiblen GD (1998) Figs of Gunung Palung National Park (West Kalimantan,
Indonesia). Tropical Biodiversity 5:245-297

Lambert FR (1992) Fig Dimorphism in Bird-Dispersed Gynodioecious Ficus. Biotropica 24:214-
216

Lambert FR, Marshall AG (1991) Keystone Characteristics of Bird-Dispersed Ficus in a
Malaysian Lowland Rain-Forest. Journal of Ecology 79:793-809

Laska M, Freist P, Krause S (2007) Which senses play a role in nonhuman primate food
selection? A comparison between squirrel monkeys and spider monkeys. American
Journal of Primatology 69:282-294

Levey DJ (1987) Seed size and fruit-handling techniques of avian frugivores. American
Naturalist 129:471-485

Lomascolo S, Speranza P, Kimball R (2008) Correlated evolution of fig size and color supports
the Dispersal Syndromes hypothesis. Oecologia Published online DOI 10.1007/s00442-008-
1023-0

Lord JM (2004) Frugivore gape size and the evolution of fruit size and shape in southern
hemisphere floras. Austral Ecology 29:430-436

Lord JM, Markey AS, Marshall J (2002) Have frugivores influenced the evolution of fruit traits
in New Zealand? In: Levey DJ, Silva WR, Galetti M (eds) Seed dispersal and frugivory:
Ecology, Evolution and Conservation. CABI Publishing, Wallingford, pp 55-68

Mack AL (1993) The sizes of vertebrate-dispersed fruits: a Neotropical-Paleotropical
comparison. American Naturalist 142:840-856

Mack AL (2000) Did fleshy fruit pulp evolve as a defence against seed loss rather than as a
dispersal mechanism? Journal of Biosciences 25:93-97

Maddison DR, Maddison WP (2000) MacClade 4: Analysis of phylogeny and character
evolution In Sinauer Associates, Inc.

Maddison DR, Maddison WP (2007) Mesquite: A modular system for evolutionary analysis In
2.01

Maddison WP (1990) A method for testing the correlated evolution of two binary characters -
Are gains or losses concentrated on certain branches of a phylogenetic tree? Evolution
44:539-557










Mamlouk AM, Martinetz T (2004) On the dimensions of the olfactory perception space.
Neurocomputing 58-60:1019-1025

Mantel N (1967) Detection of Disease Clustering and a Generalized Regression Approach.
Cancer Research 27:209

Marquez AL, Real R, Vargas JM (2004) Dependence of broad-scale geographical variation in
fleshy-fruited plant species richness on disperser bird species richness. Global Ecology
and Biogeography 13:295-304

McKey DS (1975) The ecology of coevolved seed dispersal systems. In: Gilbert LE, Raven PH
(eds) Co-evolution of animals and plants. University of Texas Press, Austin, pp 159-191

Mikich SB, Bianconi GV, Helena B, Maia NS, Teixeira SD (2003) Attraction of the fruit-eating
bat Carollia perspicillata to Piper gaudichaudianum essential oil. Journal of Chemical
Ecology 29:2379-2383

Moegenburg SM, Levey DJ (2002) Prospects for conserving biodiversity in Amazonian
extractive reserves. Ecology Letters 5:320-324

Muller-Landau HC, Wright SJ, Calder6n O, Hubbell SP, Foster RB (2002) Assessing
recruitment limitation: concepts, methods and case studies from a tropical forest. In:
Levey DJ, Silva WR, Galetti M (eds) Seed dispersal and frugivory: ecology, evolution
and conservation. CABI Publishing, Wallingford

Murray KG, Garcia-C JM (2002) Contributions of seed dispersal and demography to recruitment
limitation in a Costa Rican cloud forest. In: Levey DJ, Silva WR, Galetti M (eds) Seed
dispersal and frugivory: ecology, evolution and conservation. CABI Publishing,
Wallingford

Nathan R, Muller-Landau HC (2000) Spatial patterns of seed dispersal, their determinants and
consequences for recruitment. Trends in Ecology & Evolution 15:278-285

Osorio D, Vorobyev M (1996) Colour vision as an adaptation to frugivory in primates.
Proceedings of the Royal Society of London Series B-Biological Sciences 263:593-599

Pagel M (1994) Detecting correlated evolution on phylogenies: a general method for the
comparative analysis of discrete characters. Proceedings of the Royal Society of London
Series B-Biological Sciences 255:37-45

Pagel M (1997) Inferring evolutionary processes from phylogenies. Zoologica Scripta 26:331-
348

Pagel M (1999) The maximum likelihood approach to reconstructing ancestral character states of
discrete characters on phylogenies. Systematic Biology 48:612-622










Patel A, McKey D (1998) Sexual specialization in two tropical dioecious figs. Oecologia
115:391-400

Peichl L (2005) Diversity of mammalian photoreceptor properties: Adaptations to habitat and
lifestyle? Anatomical Record Part a-Discoveries in Molecular Cellular and Evolutionary
Biology 287A: 1001-1012

Posada D, Crandall KA (1998) MODELTEST: testing the model of DNA substitution.
Bioinformatics 14:817-818

Poulsen JR, Clark CJ, Connor EF, Smith TB (2002) Differential resource use by primates and
hornbills: Implications for seed dispersal. Ecology 83:228-240

R Development Core Team (2008) R: A language and environment for statistical computing In R
Foundation for Statistical Computing

Redford KH (1992) The Empty Forest. Bioscience 42:412-422

Reekie EG, Bazzaz FA (1987) Reproductive Effort in Plants .1. Carbon Allocation to
Reproduction. American Naturalist 129:876-896

Regan BC, Julliot C, Simmen B, Vienot F, Charles-Dominique P, Mollon JD (2001) Fruits,
foliage and the evolution of primate colour vision. Philosophical Transactions of the
Royal Society of London Series B-Biological Sciences 356:229-283

Reid N (1991) Coevolution of Mistletoes and Frugivorous Birds. Australian Journal of Ecology
16:457-469

Rey PJ, Alcantara JM (2000) Recruitment dynamics of a fleshy-fruited plant (Olea europaea):
connecting patterns of seed dispersal to seedling establishment. Journal of Ecology
88:622-633

Rey PJ, Gutierrez JE, Alcantara J, Valera F (1997) Fruit size in wild olives: implications for
avian seed dispersal. Functional Ecology 11:611-618

Rey PJ et al. (2006) The geographic mosaic in predispersal interactions and selection on
Helleborus foetidus (Ranunculaceae). Journal of Evolutionary Biology 19:21-34
Rice WR (1989) Analyzing Tables of Statistical Tests. Evolution 43:223-225

Ridley HN (1930) The dispersal of plants throughout the world. L. Reeve & Co., Ltd., Ashford

Ronquist F, Huelsenbeck JP (2003) MrBayes 3: Bayesian phylogenetic inference under mixed
models. Bioinformatics 19:1572-1574









Ronsted N, Weiblen GD, Cook JM, Salamin N, Machado CA, Savolainen V (2005) 60 million
years of co-divergence in the fig-wasp symbiosis. Proceedings of the Royal Society B-
Biological Sciences 272:2593-2599

Schaefer HM, Levey DJ, Schaefer V, Avery ML (2006) The role of chromatic and achromatic
signals for fruit detection by birds. Behavioral Ecology 17:784-789

Schaefer HM, McGraw KJ, Catoni C (2008) Birds use fruit colour as honest signal of dietary
antioxidant rewards. Functional Ecology In press

Schaefer HM, Schaefer V (2007) The evolution of visual fruit signals: concepts and constraints.
In: Dennis AJ, Schupp EW, Green RJ, Wescott DA (eds) Seed dispersal: Theory and its
application in a changing world. CABI, Wallingford, UK, pp 59-77

Schaefer HM, Schaefer V, Levey DJ (2004) How plant-animal interactions signal new insights in
communication. Trends in Ecology & Evolution 19:577-584

Schaefer HM, Schaefer V, Vorobyev M (2007) Are fruit colors adapted to consumer vision and
birds equally efficient in detecting colorful signals? American Naturalist 169:S159-S169

Schaefer HM, Schmidt V (2004) Detectability and content as opposing signal characteristics in
fruits. Proceedings of the Royal Society of London Series B-Biological Sciences
271:S370-S373

Schlick-Steiner BC et al. (2006) A multidisciplinary approach reveals cryptic diversity in
Western Palearctic Tetramorium ants (Hymenoptera : Formicidae). Molecular
Phylogenetics and Evolution 40:259-273

Schmidt V, Schaefer HM, Winkler H (2004a) Conspicuousness, not colour as foraging cue in
plant-animal signaling. Oikos 106:551-557

Schmidt V, Schaefer HM, Winkler H (2004b) Conspicuousness, not colour as foraging cue in
plant-animal signalling. Oikos 106:551-557

Schupp EW (1988) Seed and Early Seedling Predation in the Forest Understory and in Treefall
Gaps. Oikos 51:71-78

Schupp EW (1993) Quantity, Quality and the Effectiveness of Seed Dispersal by Animals.
Vegetatio 108:15-29

Schupp EW, Fuentes M (1995) Spatial Patterns of Seed Dispersal and the Unification of Plant-
Population Ecology. Ecoscience 2:267-275

Shanahan M (2000) Ficus seed dispersal guilds: ecology, evolution and conservation
implications. In. University of Leeds, UK









Shanahan M, Compton SG (2001) Vertical stratification of figs and fig-eaters in a Bornean
lowland rain forest: how is the canopy different? Plant Ecology 153:121-132

Shanahan M, Harrison RD, Yamuna R, Boen W, Thornton IWB (2001a) Colonization of an
island volcano, Long Island, Papua New Guinea, and an emergent island, Motmot, in its
caldera lake. V. Colonization by figs (Ficus spp.), their dispersers and pollinators. Journal
of Biogeography 28:1365-1377

Shanahan M, So S, Compton SG, Corlett R (2001b) Fig-eating by vertebrate frugivores: a global
review. Biological Reviews 76:529-572

Smith CI, Pellmyr O, Althoff DM, Balcazar-Lara M, Leebens-Mack J, Segraves KA (2008)
Pattern and timing of diversification in Yucca (Agavaceae): specialized pollination does
not escalate rates of diversification. Proceedings of the Royal Society B-Biological
Sciences 275:249-258

Spiegel O, Nathan R (2007) Incorporating dispersal distance into the disperser effectiveness
framework: frugivorous birds provide complementary dispersal to plants in a patchy
environment. Ecology Letters 10:718-728

Strauss SY, Lau JA, Carroll SP (2006) Evolutionary responses of natives to introduced species:
what do introductions tell us about natural communities? Ecology Letters 9:354-371

Sumner P, Mollon JD (2000a) Catarrhine photopigments are optimized for detecting targets
against a foliage background. Journal of Experimental Biology 203:1963-1986

Sumner P, Mollon JD (2000b) Chromaticity as a signal of ripeness in fruits taken by primates.
Journal of Experimental Biology 203:1987-2000

Surridge AK, Osorio D, Mundy NI (2003) Evolution and selection of trichromatic vision in
primates. Trends in Ecology & Evolution 18:198-205

Swofford DL (2003) PAUP*: Phylogenetics analysis using parsimony (*and other methods),
version 4.0 In Sinauer

Tamboia T, Cipollini ML, Levey DJ (1996) An evaluation of vertebrate seed dispersal
syndromes in four species of black nightshade (Solanum sect Solanum). Oecologia
107:522-532

Temeles EJ, Kress WJ (2003) Adaptation in a plant-hummingbird association. Science 300:630-
633

Terborgh J (1986) Keystone plant resources in the tropical forest. In: Soule ME (ed)
Conservation Biology: the science of scarcity and diversity. Sinauer Associates,
Massachusets, pp 330-344









Terborgh J, Pitman N, Silman M, Schichter H, Nfifez P (2002) Maintenance of tree diversity in
tropical forests. In: D. J. Levey WRS, and M. Galetti (ed) Seed dispersal and frugivory:
Ecology, Evolution and Conservation. CABI Publishing, Wallingford, pp 1-17

Thies W, Kalko EKV, Schnitzler HU (1998) The roles of echolocation and olfaction in two
Neotropical fruit-eating bats, Carollia perspicillata and C-castanea, feeding on Piper.
Behavioral Ecology and Sociobiology 42:397-409

Thompson JD, Higgins DG, Gibson TJ (1994) CLUSTAL W: Improving the sensitivity of
progressive multiple sequence alignment through sequence weighting, position-specific
gap penalties, and weight matrix choice. Nucleic acids research 22:4673-4680

Thompson JN (1994) The coevolutionary process. The University of Chicago Press

Thompson JN (1999) Specific hypotheses on the geographic mosaic of coevolution. American
Naturalist 153:S1-S14

Thompson JN, Pellmyr O (1992) Mutualism with Pollinating Seed Parasites Amid Co-
Pollinators Constraints on Specialization. Ecology 73:1780-1791

Thornton IWB, Compton SG, Wilson CN (1996) The role of animals in the colonization of the
Krakatau Islands by fig trees (Ficus species). Journal of Biogeography 23:577-592

Titus JH, Holbrook NM, Putz FE (1990) Seed-Germination and Seedling Distribution of Ficus-
Pertusa and F-Tuerckheimii Are Strangler Figs Autotoxic. Biotropica 22:425-428

Traveset A, Willson MF, Verdu M (2004) Characteristics of fleshy fruits in southeast Alaska:
phylogenetic comparison with fruits from Illinois. Ecography 27:41-48

Tumbull LA, Crawley MJ, Rees M (2000) Are plant populations seed-limited? A review of seed
sowing experiments. Oikos 88:225-238

van der Pijl L (1961) Ecological Aspects of Flower Evolution .2. Zoophilous Flower Classes.
Evolution 15:44-&

van der Pijl L (1969) Principles of dispersal of higher plants. Academic Press, Orlando, Fla.
Vazquez DP (2005) Degree distribution in plant-animal mutualistic networks: forbidden links or
random interactions? Oikos 108:421-426

Vazquez DP, Aizen MA (2004) Asymmetric specialization: A pervasive feature of plant-
pollinator interactions. Ecology 85:1251-1257

Voigt FA, Bleher B, Fietz J, Ganzhorn JU, Schwab D, Bohning-Gaese K (2004) A comparison
of morphological and chemical fruit traits between two sites with different frugivore
assemblages. Oecologia 141:94-104









Vorobyev M, Osorio D (1998) Receptor noise as a determinant of colour thresholds. Proceedings
of the Royal Society of London Series B-Biological Sciences 265:351-358

Waser NM, Chittka L, Price MV, Williams NM, Ollerton J (1996) Generalization in pollination
systems, and why it matters. Ecology 77:1043-1060

Weiblen GD (2000) Phylogenetic relationships of functionally dioecious Ficus (Moraceae) based
on ribosomal DNA sequences and morphology. American Journal of Botany 87:1342-
1357

Weiblen GD (2002) How to be a fig wasp. Annual Review of Entomology 47:299-330

Westoby M, Leishman MR, Lord JM (1995) On misinterpreting the phylogeneticc correction'.
Journal of Ecology 83:531-534

Wheelwright NT (1985) Fruit size, gape width, and the diets of fruit-eating birds. Ecology
66:808-818

Wheelwright NT, Janson CH (1985) Colors of fruit displays of bird-dispersed plants in two
tropical forests. American Naturalist 126:777-799

White TJ, Bruns T, Lee S, Taylor J (1990) Amplification and direct sequencing of fungal
ribosomal RNA genes for phylogenetics. In: Gelfand D, Sminsky J, White T (eds) PCR
protocols: a quide to methods and applications. Academic Press, San Diego, pp 315-322

Whittaker RJ, Jones SH (1994) The role of frugivorous bats and birds in the rebuilding of a
tropical forest ecosystem, Krakatau, Indonesia. Journal of Biogeography 21:245-258

Wiebes JT (1979) Co-Evolution of Figs and Their Insect Pollinators. Annual Review of Ecology
and Systematics 10:1-12

Willson MF, Irvine AK, Walsh NG (1989) Vertebrate Dispersal Syndromes in Some Australian
and New-Zealand Plant-Communities, with Geographic Comparisons. Biotropica 21:133-
147

Willson MF, Whelan CJ (1990) The evolution of fruit color in fleshy-fruited plants. American
Naturalist 136:790-809

Wysocki CJ, Preti G (2004) Facts, fallacies, fears, and frustrations with human pheromones.
Anatomical Record Part a-Discoveries in Molecular Cellular and Evolutionary Biology
281A:1201-1211


Zar JH (1999) Biostatistical Analysis, 4th edn. Prentice-Hall, New Jersey









Zerega NJC, Clement WL, Datwyler SL, Weiblen GD (2005) Biogeography and divergence
times in the mulberry family (Moraceae). Molecular Phylogenetics and Evolution
37:402-416

Zwickl DJ (2006) Genetic algorithm approaches for the phylogenetic analysis of large biological
sequence datasets under the maximum likelihood criterion. In:
www.bio.utexas.edu/faculty/antisense/garli/Garli.html. University of Texas, Austin.









BIOGRAPHICAL SKETCH

I completed my undergraduate studies at the Universidad Nacional de Tucuman (UNT), in

Argentina, and obtained the title of Licenciate in Biological Sciences, with an orientation in

Zoology in 1998. I obtained various honors for graduating with the highest average in the class

of 1998 for all students in the College of Natural Sciences, namely, being the carrier of the

National flag in official events, three medals given by the Department of Zoology, the College of

Natural Sciences, and the Graduate School, and honorary diplomas by the Biological Sciences

Graduate Association and the Argentinean Federation of University Women.

I obtained a graduate fellowship from the Secretariat of Science and Technology at the

UNT to complete my graduate studies there. I started a project on the importance of birds and

bats for the dispersal of a local species, Solanum riparium (Solanaceae) but, shortly after I also

obtained a Fulbright Fellowship to go to graduate school in the United States. I pursued that

opportunity and came to the University of Florida, where I enrolled in the MS and then Ph.D.

programs in the Department of Zoology in August of 2000.

I obtained several grants totaling approximately $35,000 to complete my field work in

Papua New Guinea. Funding was provided by NSF, Rufford Foundation, Garden Club of

America, Lubee Foundation, Sigma Xi (two Grants in Aid for Research), American Women in

Science, Sophie Danforth Foundation, among others. I also obtained several fellowships to

support myself while taking courses and writing my dissertation, such as the Fulbright

Fellowship mentioned above, an International Women Fellowship form P.E.O., and a

Dissertation Fellowship from CLAS. I was also credited for "outstanding academic

achievement" by an International Student Award by the University of Florida.





PAGE 1

1 THE ROLE OF FRUGIVORES IN TH E EVOLUTION OF FRUIT IN FICUS (MORACEAE) By SILVIA BEATRIZ LOMASCOLO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Silvia Beatriz Lomscolo

PAGE 3

3 To my husband and son, who are the compass that keeps me focused on the true North of my life. To my parents, for just giving, from the very beginning.

PAGE 4

4 ACKNOWLEDGMENTS I first want to thank Doug Levey, m y advisor, for his mentoring, patie nce, respect for my choices in life and, especially, for believing in me even at times when I doubted of my own capacity to get to the end. Doug was the best adviser I could have asked for my Ph.D. He became a role model as a scientist, teacher, and parent. Every time a piece of equipment broke while I was in the field, Doug rushed to send a replacement piece half a world away. Moreover, the equipment always came accompanied by many choc olate bars, an important pat in the back for a woman in despair in the field. The other members of my committee, Rebecca Kimball, Ben Bolker, Mike Miyamoto, and Emilio Bruna, offered great insight from different perspectives. Drs. David Dilcher and Colin Chapman were once part of my committee. They had to be replaced for regulatory and logistic reasons, but were very helpful at the beginning stages of my disse rtation. Special thanks go to Rebecca and Ben, who were extremely helpful with data analysis and always willing to offer their time to brainstorm for new ideas on how to tackle my (uncooperative) data. Without Rebecca, the phylogenetic analysis would have been extremely difficult to tame. Without Ben, I would still be wandering aimlessly through hall ways, books, and internet forums. Rebecca and Ben are true assets to th e Department of Zoology. My time in Papua New Guinea (PNG) was one of the richest, hardest, and most intense periods of my life. Things would have been much harder and a lot le ss enjoyable without the help of Andy Mack, Ross Sinclair, Debra Wright and the rest of the Wildlife Conservation Society PNG crew. They unselfishly helped in ev ery aspect of my field work, offering resources and (at times much needed) emotional support. Moreover, they became long-lasting friends. Mnica Alberto and Laurent Bertin hosted me for weeks at a time while I anxiously waited for equipment to arrive, or for rides to the fi eld. I am immensely thankful for their endless

PAGE 5

5 generosity, and for offering me a place to stay, when all else failed. Madang wouldnt have felt like such a familiar place without them. The management at Jais Aben Resort was also generous to offer an extremely good deal to a low-budgete d graduate student. The staff were always extremely nice and made me feel at home. Big thanks go to Gisele Maisonneuve too, who shared her house with us a fe w times, facilitating my work with the coastal fig species around Madang. Unknowingly, she provided a safer environm ent for the first month of development of our child. Volatile compound work would not have been possible without the help of Peter Teal, Hans Alborn, and Barbara Dueben, who is no longe r with us. They are great scientists and amazing people who adopted me and helped me w ithout asking anything in return. We need more people like them in science (and everywhere, really). Mark Whitten of the Florida Museum of Natural History also helped me in the fledgling stages of the volatile work. He helped me get started and he led me to Peter Teal, two events that made the volatile compound work a reality. Many people in the Department of Zoology he lped in many aspects of my dissertation. Brian McNabb with his contagious curiosity a nd love for PNG helped me financially and logistically with my fi rst trip to explore that amazing land. He was right: PNG is one of those places on Earth immensely worth seeing. Thanks to many professors, including Ed Braun, Craig Osenberg, Marta Wayne, Larry McEdward, Carmine Lanciani, Richard Kiltie, Colette St. Mary, and others for interesting discussions in their offi ces or during courses that in one way or another helped to shape this dissertation as it is today. Karen Pallone wa s always vigilant in helping me over administrative hurdles; I dont know what Zoology gradua te students would do without such an extremely efficient person in our Depart ment. Pete Ryschkewitsch helped me prepare

PAGE 6

6 for my fieldwork in many ways and always made my trips to the basement a fun break from my computer. Ken Albergotti and Jared Wright are the most efficient and knowledgeable computer people weve had since I came to the Department. I thank them for their many trips to the 6th floor to save me from my com puter illiteracy. Thanks also go to Kent Vliet, Kenetha Johnson, Tangelyn Mitchell, Lelia Long, and Jimie Norton from Biological Sciences, who helped support me as a Teaching Assistant. The many hours spent on the 6th floor led me to meet great people who have come and gone through the Department, but who will remain an important part of my life. Special thanks go to Manjula Tiwari, Andreas Heyland, Cris tina Dockx, Cedric Worman, Toshi Okuyama, Martin and Veronika Schaefer, and Charlotte Skov, who are long-lasti ng friends, and who Ill take home with me in my fond memories. The collaborations I developed throughout my Ph.D. ended up be ing an important part of this dissertation. I thank George Weiblen for in troducing me to Ohu and for his constant help with PNG figs. I want to especially acknowledg e Martin Schaefer and Pablo Speranza for fun and productive collaborations. I hope there are many more. Thanks also go to Eric Dyreson for walking me through EPCA, a statistical analysis that seemed written to fill a gap in my dissertation, and for patiently an swering all my questions (some of them, more than once). Tamaki Yuri pointed me to EPCA and helped me weave through Mesquite and get the analysis to work. I also want to thank Nina Rnsted for generously providing sequences for all the fig species I needed, before she had even published her work. Thanks go to my undergraduate volunteers, who spent long hours entering and juggling data. I really hope they gained as much as I did from the experience. I thank Eliza Chiang and Derek Cox for sticking with me until the very end. Jamie Morris, Sara Ouly, and Sabrina Simms

PAGE 7

7 also helped at different stag es. Kore Tau and Francesca Dem, from PNG, helped me by taking full charge of parts of my projec t, even after I left PNG. They also helped me get to know the mysterious Melanesian culture a bit better. Thanks go to the people in Ohu, especially to my field assistants Amos, Frank, and Ismael, without whom I couldnt have gathered half of my data. To Brus Isua, for sharing his immense knowledge on the figs of Madang, and to the whole Isua family for hosting me for so long. To Lucy and the many women of O hu who have showed me how strong a woman can really be. My parents and my sister are an inherent part of this work because they have always been role models for me. They constantly show me with their example how to be a good professional, a good daughter, a good sibling and, now a good parent. I still have a lot to learn from them, now from a bit closer. Above all, I want to thank my husband for ma king enjoyable the (at ti mes arid and rough) journey towards finishing this di ssertation. He helped with everything that I have thanked other people for: fieldwork, equipment, ideas, interes ting discussions about my work, and, especially, emotional support. He taught me the value of pe rsistence and of self-te aching. He was a pillar that helped me stay on my feet in the toughest moments, and a true partner in this enterprise. He also gave me my child, which is what Im most thankful for in life.

PAGE 8

8 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ........11 LIST OF FIGURES.......................................................................................................................12 ABSTRACT...................................................................................................................................14 CHAP TER 1 GENERAL INTRODUCTION.............................................................................................. 16 2 CORRELATED EVOLUTION OF FIG SIZE AND COLOR SUPPORTS THE DISPERSAL SYNDROME HYPOTHESIS ..........................................................................20 Introduction................................................................................................................... ..........20 Methods..................................................................................................................................23 Study System...................................................................................................................23 General Approach............................................................................................................ 23 Fruit Trait Data............................................................................................................... .24 Phylogenetic Data and Analyses..................................................................................... 26 Sequences.................................................................................................................26 Alignment.................................................................................................................27 Phylogenetic analyses..............................................................................................28 Fruit Trait Analyses......................................................................................................... 29 Phylogenetic comparative analysis.......................................................................... 29 Sensitivity analysis................................................................................................... 30 Non-phylogenetic comparative analysis.................................................................. 31 Results.....................................................................................................................................31 Phylogenetic Reconstruction...........................................................................................31 Non-Phylogenetic Comparative Analysis....................................................................... 31 Phylogenetic Comparative Analysis................................................................................ 32 Discussion...............................................................................................................................33 3 SIGNAL CONVERGENCE IN FRUITS: A RESULT OF SELECTION BY FRUGI VORES?.................................................................................................................... .51 Introduction................................................................................................................... ..........51 Materials and Methods...........................................................................................................54 Fruit Color and Disperser Data........................................................................................ 54 Vision Model...................................................................................................................55 Data Analysis...................................................................................................................56 Prediction 1: Color discrimination by birds and primates........................................ 56 Prediction 2: Conspicuousness of the fruits to frugivores ........................................ 57

PAGE 9

9 Prediction 3: Phylogenetic control...........................................................................57 Results.....................................................................................................................................58 Prediction 1: Color Discrimina tion by Birds and Prim ates............................................. 58 Prediction 2: Conspicuousness of the Fruits to F rugivores............................................. 59 Prediction 3: Phylogenetic Control.................................................................................59 Discussion...............................................................................................................................59 4 REVISITING THE DISPERSAL SYNDRO ME HYPOTHESIS: A PHYLOGENETIC COMPARATIVE STUDY OF FRUIT TRAITS IN FiCUS (MORACEAE)........................ 70 Introduction................................................................................................................... ..........70 Methods..................................................................................................................................75 Study Site and General Specifications............................................................................ 75 Study system: the natural history of figs......................................................................... 75 Quantification of Visits to Figs....................................................................................... 77 Fruit Traits.......................................................................................................................78 Fruit color.................................................................................................................78 Fruit odor.................................................................................................................. 80 Fruit size, and softness............................................................................................. 80 Fruit exposure traits.................................................................................................. 81 Data Analysis...................................................................................................................81 Identification of correla tion in fruit traits .................................................................81 Phylogenetic control.................................................................................................82 Volatile compounds in figs with different dispersers ............................................... 84 Phylogenetic reconstruction............................................................................................84 Results.....................................................................................................................................85 Visits to Fig Trees........................................................................................................... 85 Disperser PCA and EPCA...............................................................................................85 Fruit Trait PCA and EPCA.............................................................................................. 86 Volatile Compounds........................................................................................................87 Discussion...............................................................................................................................88 5 WITHIN SPECIES TEST OF THE DISPE RS AL SYNDROME HYPOTHESIS: A STUDY OF DIOECIOUS FIGS........................................................................................... 107 Introduction................................................................................................................... ........107 General Background and Goals.....................................................................................107 Natural History of Fig Pollination and Seed Dispersal................................................. 109 Methods................................................................................................................................111 Study Site and General Specifications.......................................................................... 111 Quantification of Visits to Figs..................................................................................... 111 Fruit Traits.....................................................................................................................112 Fruit color...............................................................................................................112 Fruit odor................................................................................................................ 114 Fruit size, softness, and peduncle length................................................................ 114 Data Analysis.................................................................................................................115 Frugivore visits to female versus male trees.......................................................... 115

PAGE 10

10 Difference in fruit traits of seed and gall figs......................................................... 115 Changes from unripe to ripe stage in m onoecious and dioecious figs................... 115 UV reflectance........................................................................................................116 Results...................................................................................................................................116 Frugivore Visits to Female Versus Male Trees............................................................. 116 Difference in Fruit Traits of Seed and Gall Figs........................................................... 116 Changes From Unripe to Ripe Stag e in Monoecious and Dioecious Figs .................... 117 UV Reflectance............................................................................................................. 118 Discussion.............................................................................................................................118 6 GENERAL CONCLUSIONS............................................................................................... 131 LIST OF REFERENCES.............................................................................................................136 BIOGRAPHICAL SKETCH.......................................................................................................152

PAGE 11

11 LIST OF TABLES Table page 2-1 Results from Maddisons Concentrated Changes test.......................................................43 2-2 Data on fruit diameter, color, animal consum ers, and distribution for the 64 species included in our phylogeny, as found in the literature........................................................ 44 3-1 Loadings of the quantum catch variables for prim ate and bird cone sensitivities for the training DFA.............................................................................................................69 4-1 Loadings of variables for the Disperser PCA and EPCA................................................ 105 4-2 Loadings of variables for the Fruit Trait PCA and EPCA...............................................105 4-3 Information on abbreviations used in fi gures, disperser record ed, and source of inform ation for dispersal mode........................................................................................ 106 5-1 Discriminant Function Analysis of female versus m ale figs using eight fruit variables. 130 5-2 Discriminant Function Analysis of fema le, m onoecious and male figs at different maturity stages................................................................................................................ .130 5-3 Mean UV reflectance for fig species of different sex and m aturity stages...................... 130

PAGE 12

12 LIST OF FIGURES Figure page 2-1 Bootstrap support and Bayesian posteri or probabilities on the m aximum likelihood phylogeny of 64 Ficus species........................................................................................... 39 2-2 Box and whiskers plot comparing the m ean dia meter of Type A (red, purple) and Type B (green, yellow, brown, and orange) fruits............................................................ 40 2-3 Maximum likelihood phylogeny of 64 Ficus species.. ......................................................41 3-1 Calculation of quantum catch values................................................................................. 67 3-2 Results of the Discriminant Function analysis (D FA) based on primate cone excitation and bird cone excitation.................................................................................... 68 4-1 Map of Papua New Guinea in reference with a world map............................................... 95 4-2 Loadings of the first three principal components (PC) on fruit reflectance spectra from 300 to 750nm, measured at 5nm intervals................................................................ 96 4-3 Maximum Likelihood tree of 42 fig species...................................................................... 97 4-4 Parsimony reconstruction of dispersal mode..................................................................... 99 4-5 Results of the Disperser PCA and EPCA........................................................................ 101 4-6 Plot of PC 1 vs 2 for the Fruit Tr ait P CA and EPCA for 42 species of figs.................... 102 4-7 Correlation between genetic distance and difference in the score on PC 1 ..................... 103 4-8 Discriminant Functions on original and random ized matrix of 73 volatile compounds.......................................................................................................................104 5-1 Loadings of the first three principal components (PC) on fruit reflectance spectra from 300 to 750nm, measured at 5nm intervals.............................................................. 123 5-2 Plot of Discriminant Function an alysis of fe male and male figs..................................... 124 5-3 Scores on the linear discriminant function (LDF 1) resulting from the discriminant analysis of female and male figs...................................................................................... 125 5-4 Maximum Likelihood phylogeny of the sp ecies included in the seed-gall f ig comparison..................................................................................................................... ..126 5-5 Discrimination among figs of different sex and m aturity stages using a Discriminant Function analysis............................................................................................................. 128

PAGE 13

13 5-6 Pair-wise comparisons among fig groups of diffe rent sex and maturity stages..............129

PAGE 14

14 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE ROLE OF FRUGIVORES IN TH E EVOLUTION OF FRUITS IN FICUS (MORACEAE) By Silvia Beatriz Lomscolo May 2008 Chair: Douglas J. Levey Major: Zoology We attempt to resolve a central controversy of plant-animal interactions: whether seed dispersers are an important factor in shaping the evolution of fruit traits. By focusing on a genus ( Ficus ) with extremely variable frui t traits, we could study fruit variation using a phylogeny to control for the evolutionary non-independence of fig species. We studied six characters that were hypothesized to most likely influence a frugi vores ability to find and access fruits: color, odor, size, softness, exposure away from foliage, and location on the tree. This is the first study on the role of seed dispersers in fruit evolution to combine quan titative measures of color and odor in a detailed phylogenetic comparative analysis We took a closer look at fruit color because it is probably the first trait mediating fruit-frugivor e interactions and because it has historically been mismeasured or ignored in studies of fru it trait evolution. We asked: (1) Can frugivores discriminate between different colors? (2) Have fruit colors of plants that share dispersers converged to a common signal? (3) Do fruit characters co-occur in non-random patterns among species of Ficus ? (4) If they do, are the patterns most likely the result of phylogeny or selective pressure by different types of frugivores? Results show that color can be used as a signal by frugivores because it can be discriminated by birds and primates, two of the most common frugivores worldwide. Fruit color

PAGE 15

15 has converged in distantly related species, whic h can be predicted under selective pressure by frugivores. However, we cannot discard the role of microbes as a selective force that determines the amount of the pigment anthocyanin, which ha s antifungal properties. Color, especially brightness and contrast against the background, have evolved in co rrelation with fruit odor, size, softness, and exposure away from leaves. Results show that the correlation among these traits cannot be explained solely by phylogeny: a large portion of the variation in fruit traits is adaptive. The most likely explanation for the correla ted evolution of six of the traits studied here is selection by frugivores. This pattern holds for figs in one geographic region and worldwide, and that it holds both within and between fig speci es. Taken together, this evidence argues for the prevalence of seed di spersers over phylogeny as the main driver of the evolution of fruit traits in figs.

PAGE 16

16 CHAPTER 1 GENERAL INTRODUCTION It is interesting to contemplat e an entangled bank, clothed with many plants of many kinds, with birds singing on the bushes, with various insects flitting about, and with worms crawling through the damp earth, and to reflect that these elaboratel y constructed forms, so different from each other, and dependent on each other in so comp lex a manner, have all been produced by laws acting around us. Darwin, 1859, On The Origin of Species Species interactions are ubiquitous and im portant in shaping many ecological and evolutionary processes, as depicted by the paragraph by Darwin c ited above (1859). All biologists will probably agree on the fact that ever y system we study is affected in one way or another by the interaction among an d within species. What we might not all agree on, however, is what types of species interactions are most important (Bertness et al. 1999; Bruno et al. 2003). Negative interactions have received the most a ttention by ecologists, whic h is illustrated by two facts. First, more studies ha ve been published on negative in teractions, including competition, predation and herbivory, than on positive interact ions, such as pollination, seed dispersal and facilitation (~146,000 and ~27,000, respectively, as re sulting from a Web of Science search using the name of the interactions mentioned ab ove together with ecology as keywords). And second, research on negative interactions have ha d stronger incidence in ecological theory, as illustrated by some of the most used models in ecology, namely, Lotka and Volterras interspecific competition model, Gauses competitive exclusion principle, and Hutchinsons and MacArthurs work on the niche and species packi ng (Bruno et al. 2003). Positive interactions have been largely overlooked, although they ha ve an important role in several ecological

PAGE 17

17 processes, such as facilitation through habitat modification (Ber tness and Leonard 1997; Halpern et al. 2007), and plant reproducti on through pollination and seed dispersal. This leads to the focus of this study, which is th e positive interaction between fleshy-fruited plants and the animals that eat the fruits and disperse the seed, potentially aiding in plant reproduction. The field of seed dispersal is wrestling w ith a major contradiction: although ecological evidence demonstrates that frugivores are importa nt for the reproduction of fruiting plants, any evolutionary role of seed dispersers in shaping fruit characteristics is difficult to detect. Those who believe in the evolutionary im portance of seed dispersers in shaping fruit evolution point to the existence of apparently nonrandom combinations of fruit traits that match the sensory abilities of the frugivores that feed on them (Janson 1983; Kalko et al. 1996; Korine et al. 2000; Lord et al. 2002). These studies argue that frugivores must ha ve selected for these suites of traits, which are referred to as dispersal syndromes On the opposite side of the issue are those who argue that other factors, such as phyloge ny or physiological constrains, play a more prevalent role than seed disper sers in shaping fruit traits (Fischer and Chapman 1993; Herrera 1985; 1987; 1992). They explain the match between fruit traits and frugivor es sensory abilities as preexisting preferences of frugivores for certain types of fruits: the frugivores will feed on the fruits that are most easy to find and accessible to them. This process is called ecological fitting (Janzen 1980; 1985). We suggest that much of this controversy stem s from the disparate approaches of different studies, the qualitative measuremen ts of fruit traits (especially color and odor), and the lack of consistency among studies in the fruits traits that are argued to be important for frugivores to find and access fruit. Studies that have claimed an asso ciation between fruit traits and seed dispersers have generally not controlled for phylogenetic effect s and hence attribute any similarity in fruit

PAGE 18

18 traits among species dispersed by similar frugivores to common selective forces by those frugivores; they tend to disregard the possibility that a group of species might bear similar fruits simply because they have shared a common ance stor (Felsenstein 1985; Harvey and Pagel 1991). Studies that have claimed a lack of association between fruit tr aits and seed dispersers have either focused at high taxonomic levels and consequently have ignored great variation at the species level (Fischer and Chapman 1993); have taken place in temperate regions, and therefore not incorporated the diversity of dispersers and fruiting plants in tropical regions (Herrera 1987; 1992); or have neglected traits that may be very important in medi ating plant-animal interactions, such as color and odor (Jordano 1995). We test the Dispersal Syndrome hypothesis, which states that frugivores have selected for the non-random combination of fru it traits found in nature. We st udy fruit characteristics at the species level in the genus Ficus based on a molecular phylogeny to control for the evolutionary non-independence of species. We focus on morphological charac ters of fruit that most likely affect frugivore fruit choice, and tr ace their evolution in the genus Ficus (Moraceae) to infer the evolutionary trajectories and unde rlying causes of the appearance of modern-day figs. We show that fruit color and size seem to have evolve d in correlation, as predicted by the Dispersal Syndrome hypothesis (Chapter 2). We then test an important assumption of the Dispersal Syndrome hypothesis, namely that the fruit traits used to define dispersal syndromes can actually be perceived and differentiated by the frugivores that the traits ar e signaling (Chapter 3). We focus on color, which has been argued to be the most important trait determining fruit choice by frugivores (Voigt et al. 2004) but has been largely mismeasured or ignored in studies of fruit evolution, probably due to the difficulties of quan tifying it in an objective way. We specifically test the Signal Convergence hypothesis, which stat es that if color is a reliable signal to

PAGE 19

19 dispersers, then colors should converge among sp ecies dispersed by similar frugivores. We show that frugivores can distinguis h the color of fruits they disperse from the color of fruits dispersed by other frugivores. We conclude that color can be used as a signal to dispersers. We also show that colors that attract different fr ugivores have evolved i ndependently in different genera and families of fruiting plants. The most important finding of our work is that when seven of the traits hypothesized to be most important for frugivores to find and access available fruits are analyzed together, they suggest that dispersers are an important selectiv e force in determining the diversity of fruiting plants, even when phylogenetic effects are accounted for (Chapter 4). Lastly, a within-species study of dioecious figs compar ing individuals with viable seeds and often consumed by frugivores (female plants, called seed figs), w ith individuals without viable seeds and not consumed by frugivores (male plan ts, called gall figs), shows the same pattern just described for interspecific comparisons (Chapter 5). Overall, the results of my study show that seed dispersers have had an important role in shaping the evolution of fruit trai ts. Other alternative hypotheses that were used to explain the diversity of specific traits such as color, or the correlated evolu tion of pairs of traits, such as color and size, were not tested, but are disc ussed. Although some alternative hypotheses cannot be discarded to explain the divers ity of specific traits, such as th e antimicrobial role of pigments in ripe fruits, only the Dispersal Syndrome hypothe sis can account for the co rrelated evolution of the six fruit traits that we examined.

PAGE 20

20 CHAPTER 2 CORRELATED EVOLUTION OF FIG SIZE AND COLOR SU PPORTS THE DISPERSAL SYNDROM E HYPOTHESIS Introduction Although seed dispersers clearl y influence the fitness and sp atial dis tribution of fruiting plants (Harms et al. 2000; Howe and Miriti 2000; Mrquez et al. 2004; Terborgh et al. 2002), their effect on the evolution of fruit traits remains controversial (Bolmgren and Eriksson 2005; Fischer and Chapman 1993; Herre ra 1992; Jordano 1995; Lord et al. 2002; Mack 1993). Evolutionary ecologists have pr oposed that differences in the behavior and ecology of seed dispersers have led to correspond ing differences in fruit specie s favored by those dispersers (Gautier-Hion et al. 1985; Janson 1983; Kalko et al. 1996, Schaeffer et al. 2006 ). For example, fruit-eating birds are diurnal, ha ve excellent color vision, a poor se nse of smell, and are limited by gape width in the size of fruit they can swallow (Jordano 2000; Leve y 1987; Rey et al. 1997; Wheelwright and Janson 1985). Therefore, fruiting plants dispersed by birds should tend to stand out against the green folia ge, and be odorless and small. Colors that stand out against leaves include red, black, and purple, all of which have either high ch romatic or achromatic contrast against foliage (Burns and Dalen 2002; Janson 1983; Schmidt et al. 2004b). Fruit-eating mammals, on the other hand, are often nocturnal, have teeth (i.e., are not gape limited), and exhibit a keen sense of smell. Thus, mammal-di spersed fruits do not need to stand out against green foliage (e.g., green, yellow, brown, or pale orange, Janson 1983), and are more likely to be relatively large and odoriferous. These sets of fruit traits de fine what have been termed "bird syndrome" and "mammal syndrome" fruits, respectively (Herrera 2002a ; McKey 1975; van der Pijl 1969). At least superficially, fruit traits such as size and color seem to be associated non-randomly, as expected (Gautier-Hion et al. 1985; Janson 1983; Kalko et al. 1996; Korine et al. 2000; Shanahan and

PAGE 21

21 Compton 2001; Wheelwright and Janson 1985). However, two opposing hypotheses explain cooccurrence of fruit size and colo r. One we termed the Phyloge netic Inertia hypothesis, which states that co-occurrence of fruit traits is phylogenetically determined; the size and color of a given fruit species is most str ongly determined by the size and colo r of ancestral species. Under this hypothesis, frugivores have pre-existing pref erences for particular types of fruits and disperse seeds in fruits that match those preferences -an apparently common process called ecological fitting (Agosta 2006; Janzen 1980; 1985; Strauss et al. 2006) that does not drive the evolution of fruit traits. Alternatively, the Dispersal Syndrom e hypothesis states that the cooccurrence of fruit traits is determined by selectiv e pressure from sets of seed dispersers that differ in behavior and ecol ogy, as described above. Tests of these hypotheses with phylogenetic co mparative methods have commonly rejected the Dispersal Syndrome hypothesis (Fischer a nd Chapman 1993; Herrera 1987; 1989; 1992; but see Jordano 1995). However, this result may reflect methodological challenges more than biology. In particular, these stud ies have pooled data from species of the same genus (Fischer and Chapman 1993) or they have focused on temperate species (H errera 1987; 1989; 1992), thereby overlooking immense intrageneric variation, especially in tropical taxa. This variation remains largely unexplored in phylogenetic comparative analyses. Several lines of evidence from studies with a broad geographic scope support the Dispersal Syndrome hypothesis. These studies controlled for phylogenetic iner tia by comparing fruit traits among plants of the same genus (Lord 2004; Lord et al. 2002), family (Mack 1993) or species (Hampe and Bairlein 2000). For example, fruit traits in New Zealand are consistent with hypothesized selective pressures by avian and reptilian frugivores, the only frugivores historically present in New Zeal and (Lord 2004; Lord et al. 2002). Also, tropical fruits in the

PAGE 22

22 New World tend to be smaller than those of re lated plant species in the Old World (Mack 1993), which is consistent with the observation that New World mammalian frugivores are smaller than in the Old World (Fleming et al. 1987; Mack 1993). Lastly, within-species differences in fruit traits match the characteristics of the different disperser guilds found in two different regions of Europe (Hampe and Bairlein 2000). Despite th ese seemingly non-random patterns, such studies are limited by low replication (t ypically, one region is compared to another) and lack of an explicit phylogenetic framework. We provide a phylogenetic test of the Dispersal Syndrome hypot hesis at the species level in the genus Ficus (Moraceae) by testing two predictions that arise under this hypothesis: (1) Ficus species with ripe fruit that are red, purple, or black will pr oduce smaller fruit than species with green, yellow, brown, and orange fruits, and these combinations of co lor and size will occur independently in different clades ; and (2) interspecific variation in fruit size of red, purple, or black figs will be less than interspecific variati on in fruit size of green, yellow, brown, or orange figs. The rationale for the sec ond prediction is that birds are mo re restricted than mammals in the size of fruits they can ingest because mammals can manipulate fruits with their forelimbs and can eat them piecemeal using thei r teeth (Alcantara and Rey 2 003). Although disperser guilds such as birds and mammals include animals th at vary considerably within each group (e.g., tanagers to cassowaries, and bats to elephants) (Lord et al. 2002) we are testing traits that should be affected in a similar way by most member s of each guild. We believe it is a fair generalization to say that most birds have more acute color vision than most mammals (especially nocturnal mammals) (Bowmaker et al. 1997; Bowmaker and Hunt 2006; Das et al. 1999) and that, overall, mammals ar e larger than birds and can th erefore feed on larger fruits than birds. Even smaller mammals, such as bats, rodents, and many mars upials may be able to

PAGE 23

23 eat larger fruits than birds of comparable size because they have teeth and can eat the fruits piecemeal, and they have forelimbs and can handle th e fruits in more ways than birds can with their beak. Through this conservative test of the Dispersal Syndrome hypothesis we found that color and size of figs seem to evolve in corr elation. The patterns of correlated evolution were identified despite the conservative nature of our test given such broad disperser guilds. Methods Study System Ficus is an ideal group f or testing the Disper sal Syndrome hypothesis because its fruits vary immensely among its ca. 750 species. Figs can be yellow, brown, red, purple, or orange, and may be as small as 4mm in diameter to as big as 80mm (Dominy et al. 2003; Laman and Weiblen 1998, S. Lomscolo pers. obs.). Ficus has an unusually wide distribution, as they are found in almost all tropical forests worldwide, an d their fruits are consumed by a taxonomically diverse array of vertebrates, including 1274 bird and mammal spec ies in 523 genera, in addition to reptiles and fishes (Shanahan et al. 2001b). Fu rthermore, molecular data are available to build a phylogeny and fruit size and co lor are described in the literature for many species. We used 64 Ficus species for which genetic a nd fruit size and color information were available in the literature. General Approach We first built a m olecular phylogeny of the genus Ficus which is necessary to tease apart the Dispersal Syndrome hypothesis from the Phylogene tic Inertia hypothesis. To ensure that the reconstructed phylogeny was not affected by the method chosen to build it, we constructed phylogenies using maximum likelihood (ML), Bayesian, and parsimony searches.

PAGE 24

24 Correlated evolution of fr uit color and size was test ed using Maddisons (1990) Concentrated Changes test and Pagels (1994 ) Likelihood Ratio test on the parsimony and ML trees. Because phylogenetic comparative methods give priority to phylogeny over ecology when explaining shared characters (Westoby et al. 1995) and we do not know the degree of phylogenetic niche conservatism (Harvey and Pagel 1991) in figs, we also tested for correlation of fruit color and size using raw data (hereaft er non-phylogenetic comparative analyses). Nonphylogenetic comparative analyses were performed on the set of species that is included in our phylogeny and on all Ficus species for which fruit data were available in the literature. This allowed us to test whether the limited set of sp ecies in the phylogeny wa s representative of a much larger set of species. The second predic tion was tested without incorporating phylogeny because all phylogenetic comparative tests that we are aware of require data for each species in the phylogeny to reconstruct ancestral character state. Because we are tes ting variation in fruit size between the two types of fruits, to do a phylogenetic analysis we would need data on intraspecific fruit size variation for each species in our phylogeny, which was not available in the literature. In most cases, only a mean diamet er was reported without any measure of within species variation. Thus, we calculated interspe cific variation for ma mmal-colored and birdcolored fruits, and compared them using an F-test This approach tests the results of selective pressure on fruit size by birds and mammals at the community level. Fruit Trait Data Fruit trait data were obtained prim arily from published sources (Croat 1978; Dominy et al. 2003; Laman and Weiblen 1998), and secondarily from a Ficus expert (Weiblen, pers. comm.) and field data collected by S. Lomscolo. All data are from ripe fruits; unripe fruits were excluded. Because the phylogenetic comparative tests we used are designed for dichotomous

PAGE 25

25 variables, we transformed color data into two categories, followi ng Jansons (1983) classification of Type A (red and purple) and Type B (green, ye llow, brown, and orange) fruits. Likewise, size data were transformed into tw o categories, based on the mode diameter (10mm) of species included in our study: small ( 10 mm) and large (> 10 mm). The 10 mm cutoff point also coincides with the mode for all Ficus species for which diameter data were available in the literature (Croat 1978; Dominy et al. 2003, Weiblen pers. comm.; Laman and Weiblen 1998), with the mean fruit diameter and bird gape width reported by Wheelri ght (1985), and with the largest diameter of fruits typically consumed by some of the most common fruit-eating passerine birds in Costa Rica (Levey 1987). Occasionally, different sources provided contra dictory information a bout fruit color of a given species. For example, Ficus asperifolia is reported to produce yellow or red fruits (Dominy et al. 2003), which means it could be classi fied either as Type A or B. Because we cannot discern whether these situations result from a polymorphism, phenotypic plasticity, or simply a reporting error, we performed separate analyses for each possible assignment of color type. We reconstructed characte r change in two ways. For the most conservative reconstruction we chose the option that would minimize support for the Dispersal Syndrome hypothesis. For example, if a species had either yellow or red fi gs and was classified in the small category, we recorded it as Type B, so that the combinati on of color and size would not match the prediction of Type A color and small size. The s econd reconstruction was the non-conservative reconstruction where, for each ambiguous classifi cation, we chose the color that matched the hypothesis of correlated evolution between the two traits. By co mparing results of these two approaches, we are able to assess the sensitivity of studies like ours to sometimes arbitrary

PAGE 26

26 character state reconstruction a commonly encountered limitati on of data gleaned from the literature. Phylogenetic Data and Analyses Although three phylogenies have been published for Ficus the com parative tests that we used require a phylogeny that has branch length s and includes only the species for which fruit data were available (Likelihood Ra tio test, Pagel 1994), or else a phylogeny that is completely resolved (Concentrated Changes test, Ma ddison 1990). Thus, we constructed our own phylogeny. Sequences: Internal transcribed sp acer (ITS) sequences for 58 Ficus species were obtained from GenBank. For six additional species (F. congensis, F. exasperata, F. mucuso, F. natalensis, F. asperifolia2 and F. vallis-choudae), we sequenced the ITS region ourselves. Voucher specimens for these species were deposited at the Florida Museum of Natural History Herbarium under numbers FLAS 218158 FLAS 218164, and the resulting ITS sequences were submitted to Genbank (Accession nos. EF520706EF520713). Ficus ITS1 through ITS2 sequences were amplified using primers IT S1 through 5 (White et al. 1990). All PCR amplifications were carried out in 25 L reactions containing 0.4 units of NEB Taq polymerase (New England Biolabs, Be verly, MA) 1.5mM MgCl2, 0.4 M of each primer and 0.1mM of each dNTP in the manufacturers buffer. Amplif ication was carried out in a Biometra T3 Thermoblock with the same program for all prim er combinations. The program consisted of an initial denaturing step of 95C for 2 min; five cycles of 95C for 1 mi n, 53C for 1 min, and 72C for 2 min with a decrease of 1C per cycle in the annealing temp erature; 35 cycles with an annealing temperature of 48C and a final extensi on step of 72C for 12min. PCR products were cleaned with Wizard SV Gel and PCR Cleanup System (Promega, Maddison, WI) and diluted to approximately 1ng/ L for sequencing.

PAGE 27

27 We sequenced all regions in both dir ections on a CEQ 8000 capillary sequencer (Beckman-Coulter, Fullerton, CA) using reaction volumes with the addition of 80mM Tris and 2 mM MgCl2 (pH 9) to complete the volume of a full reaction. To edit the sequences manually we used Sequencher TM (V4.1.4, Genecodes, AnnArbor, MI). Alignment : We prealigned the sequences with the Clustal-W (Thom pson et al. 1994) algorithm included in BioEdit (V 5.0.6, Hall 199 7) and manually adju sted the resulting alignment. The nrDNA region includes ITS1 and ITS2, separated by the 5.8S ribosomal RNA gene. Because the 5.8S region differed from ITS1 and 2 in that it contained few variable sites and appeared to evolve under a different mode l of evolution (see below), the 5.8S region was excluded from analyses. Base positions that could not be unambiguously aligned, particularly due to insertions or deletions (indels) of nucle otides, were recoded as missing data. There were 15 indels two or more base pairs in length in regions where alignment was unambiguous. For parsimony analyses only, these were considered single characters, coded as present/absent, and included in analyses. Otherwise, all single base pairs deletions and multiple base pairs deletions that could not be unambiguously aligned were tr eated as missing data. A low complexity region ranging from 20 to 40 base pairs in ITS 2 c ontaining a poly-G was not readily alignable; however, preliminary parsimony searches conduct ed without this region did not result in sufficiently resolved trees. To make use of the information contained in this region, clades that had appeared consistently in our preliminary an alyses and were congruent with the phylogenies shown in Weiblen (2000), Jous ellin et al. (2003), and Rnste d et al (2005) were aligned separately for this region. This approach is si milar to the stepwise alignment strategy described by Hsiao et al. (1999). Although th is strategy resulted in unambi guous within-clade alignments, alignment across clades remained difficult for this region. Consequently, in order

PAGE 28

28 to avoid spuriously reinforcing support for the clades that had been aligned separately, all informative positions within this region were manually de-aligned betw een pre-aligned groups by introducing gaps when the individual ali gnments were pooled in a single matrix. Phylogenetic analyses : We used the clade formed by F. maxima, F. insipida and F. yoponensis as a functional outgroup based on the fact that it appeared as sister to the rest of the genus in all our preliminary analyses conducted using Morus alba as an outgroup. This topology was fully consistent with Weiblens (2000), Jousselin et al.s (2003), and Rnsted et als (2005) analyses, which were performed with different data sets. The maximum parsimony criterion searches for the phylogenetic tree that minimizes the number of evolutionary changes that must have occurred. Using PAUP* 4.0b10 (Swofford 2003), a preliminary heuristic search was perf ormed with tree bisec tion-reconnection (TBR) branch swapping and the minimum tree length reco rded. The final searches were conducted for 5000 sequence addition replicates (SAR) by saving onl y trees that were shor ter or equal to the minimum tree found in the preliminary search. A bootstrap analysis was also conducted for 1000 bootstrap replicates with 100 SAR within each replicate using TBR. The maximum likelihood criterion searches for the phylogenetic tree that best fits the data, given a particular model of evolution. The model of evoluti on that best describes our data was determined by the hierarchical likeli hood ratio test in MODELTEST 3.06 (Posada and Crandall 1998). We initially examined the best model for ITS1, ITS2, and the 5.8S region independently. The best models selected for ITS1 and ITS2 were similar and more complex than the model selected for the 5.8S region. Thus, the 5.8S region was excluded from analyses and the best model for the combined ITS regions wa s used for the likelihood analysis. We conducted

PAGE 29

29 a heuristic search in PAUP* with 10 random sequence additions using the model (TN93) and parameters recommended by MODELTEST 3.06 (Posada and Crandall 1998). We assessed support on the phylogeny using two different approaches. Bayesian analyses were conducted using MrBayes 3.1.2 (Huels enbeck and Ronquist 2001; Ronquist and Huelsenbeck 2003). We ran our chains for 1 x 107 generations, sampling every 1000 generations. The first 2 x 106 generations were di scarded as burn-in, and a consensus of the remaining trees was used to determine the poste rior probabilities at ea ch node. We also ran 200 bootstrap replicates using maximum lik elihood using GARLI 0.951 (Zwickl 2006). We considered nodes strongly supported when posterior probabilities were greater than 95% or when bootstrap values were greater than 70% (A lfaro et al. 2003; Hillis and Bull 1993). Fruit Trait Analyses Phylogenetic comparative analysis : To test whether the patte rn of association between f ruit color and size constitutes evidence of correlated evolutionary change in both traits (prediction #1) we performed Pa gels Likelihood Ratio test (1994), as implemented in Discrete v. 4.0 (Pagel 1994; 1999), which compares the goodness of fit of a model of independent evolution to one of correlated evolution. This te st assumes that changes in character state are more probable on long branches than on short bran ches and takes into account branch length for character state reconstruction. We also perfor med Maddisons (1990) Concentrated Changes test using MacClade 4.0 (Maddison and Maddison 2000), which tests for non-random association of traits by determining if an evolutionary change in one trait is correlated with change in a second trait. The test reconstructs character changes using a pa rsimony criterion and assumes a dependent and an independent variable. We determ ined whether color or size was more likely to be independent using Discrete (Pagel 1994; 1997). For example, if changes in color precede changes in size, rates of change in size shoul d differ depending upon color type while rates of

PAGE 30

30 color change should not differ by size category. By constraining rates of change among joint character states (e.g., small-Type A to small-Type B set equal to large-Type A to large-Type B), the likelihood of color preceding size could be compared to that of size preceding color. Since this approach does not test nested models, the different models were compared using the AIC (Akaikes Information Criterion) (Burnham and A nderson 2003). Since results of this test were not conclusive, and since mode of evolution of fruit traits is not well -understood, we performed the Concentrated Changes test (Maddison 1990) in two different ways: first using color as the independent variable, and then using size as the independent variable. The Concentrated Changes test cannot deal with unresolved nodes, so we resolved the polytomies in our phylogeny by using the resolution of our majority rule tree, Weiblens (2000), Jousellin et als (2003) or Rnsted et als (2005) reconstructions. Sensitivity analysis : The Concentrated Changes test requires specification of several different algorithms, although defining the algor ithms may be arbitrary if the mode of trait evolution is not well known. To address this, we performed a sensitivity analysis using all available algorithm combinations to explore if algorithm choice ha d a major influence on the test of prediction #1 and, if so, which algorithm seemed to most influence the results. The algorithms that we varied were (1) ACCTRAN or DELTRAN, which are options for resolving ambiguous branches and that produce character changes earlier (at the basal branches ) or later (at the terminal branches) in phylogenetic histor y, respectively. ACCTRA N increases the number of losses, whereas DELTRAN increases the number of gains; and (2) MINSTATE or MAXSTATE, which are options for reconstruc ting character states at ambiguous nodes. MINSTATE uses the lowest trait value, wherea s MAXSTATE uses the largest trait value.

PAGE 31

31 As part of the sensi tivity analysis, we ran the analyses using the conservative and nonconservative trait reconstruction, using colo r and size as the independent variables. Non-phylogenetic comparative analysis : To test if type A fruits had smaller diameter than type B fruits (prediction #1), we used a Kolmogorov-Smirnov non-parametric test because our data were not normal. To test whether type B fruits have great er variation in fruit size than type A fruits (prediction #2), we performed an Ftest (Zar 1999). Both tests were performed for all species of Ficus found in the literature and for the subset of species included in the phylogeny. Results Phylogenetic Reconstruction The phylogenies generated using maxi mum likelihood, Bayesian, and parsimony analyses were similar to each other, as well as to published phylogenies (Jousselin et al. 2003; Rnsted et al. 2005; Weiblen 2000). The majority of nodes were well-suppor ted in our analyses (Fig. 2-1). With taxa shared among our phyloge nies and published phylo genies, all the major clades with high bootstrap values were essentially identical. C onflicts occurred only in poorly supported groups and those conflicts were minor. Slight differences in topology between the ML and parsimony tree did not affect the results of our phylogenetic comparative analyses (see below). Non-Phylogenetic Comparative Analysis When we analyzed data only for those species of Ficus included in our phylogeny, we om itted a species ( F. ruginerva ) whose fruits were reported as larger than 10mm, but no exact diameter was given. For this limited set of species type A fruits were smaller than type B fruits, in agreement with prediction #1 (x A= 14.86 mm, nA = 36; x B= 27.41, nB = 27; Kolmogorov-

PAGE 32

32 Smirnov Z = 1.818; p = 0.003). For the entire set of species, the difference was in the same direction but not as large (x A = 15.46 mm, nA = 56;x B = 18.13 mm, nB = 72; KolmogorovSmirnov Z = 1.359; p = 0.050). The variance in diameter of type B fruits wa s significantly larger than the variance of type A fruits for the subset of species in our phyloge ny, in agreement with prediction #2 ( 2 A = 79.63; 2 B = 281.69; F35,26 = 3.54; p < 0.001). However, this difference in variance was not as large when all speci es were included in the analysis ( 2 A = 106.68; 2 B = 116.36; F71,55 = 1.09; One tailed test p > 0.25) (Fig. 2-2). Phylogenetic Comparative Analysis For two reas ons, we emphasize the analyses pe rformed on the ML tree (Fig. 2-3). First, Pagels Likelihood Ratio test ( 1994) requires branch lengths, which were only available on the ML trees. Second, the parsimony searches came up with over 5000 most parsimonious trees and performing the comparative analyses in al l 5000 of them was unwieldy. A preliminary comparative analysis performed on a randomly chosen parsimony tree showed no difference with the comparative analyses performed on the ML tree. With Pagels Likelihood Ratio test (Pagel 1994), the model of correlated evolution of fruit size and color fits the data significantly better than the model of independent evolution, which prediction #1, regardless of the coding of ambiguous fruit colors (Conservative reconstruction: likelihood ratio = 4.889; p = 0.027; 1000 simulations. Non-conservative reconstruction: likelihood ratio = 4.990; p = 0.027; 1026 simulations). To determine the independent variable fo r Maddisons Concentrated Changes test (Maddison 1990), assuming color precedes size is more likely (ln likelihood = -80.31) than assuming size precedes color (ln likelihood = -81.68) However, while color may be more likely to precede size, examination of the Akaike weights cannot exclude size preceding color. Assuming color is independent is mo re consistent with the observat ion that color tends to change

PAGE 33

33 at basal branches while size tends to change at mo re terminal branches (Fig. 2-3). When color is considered as the independent variable, the re sults for Maddisons Con centrated Changes test (Maddison 1990) indicate that fru it size changes from large to sm all on branches with brightly colored fruits more often than expected by chance. The p-values from all the tests performed with the different algorithm combinations a nd trait reconstructions ranged from 0.002 to 0.03 (Table 2-1). When size is considered the inde pendent variable, however results vary greatly according to the algorithm combination and char acter state reconstruction used and support for prediction #1 is mixed (0.04 < p < 0.54; Table 2-1). Discussion Overall, our results support th e Dispersal Syndrom e hypothesis. Mean diameter is smaller for Type A fruits than for Type B fruits, s upporting prediction #1 of the Dispersal Syndrome hypothesis. This pattern holds wh en using raw data for the subset of species included in our phylogeny and when using raw data for all the species of figs found in the literature. It also holds when examined with phylogenetic compara tive methods: fruit size tends to change from large to small on species with type A colors (e .g., red and purple) more often than expected by chance. This result was consistent across all combinations of algorithms for the Concentrated Changes test (Maddison 1990) when color was considered the independent variable. The tendency of color and size to evolve in correla tion, at least in the s ubset of 64 species of Ficus included in our analyses, s upports the Dispersal Syndrome hypothesis rather than the Phylogenetic Inertia hypothesis, suggesting that seed dispersers have been important for the evolution of at least some fruit traits in Ficus The Dispersal Syndrome hypothesis was supported to a lesser extent by our test of prediction #2. In the subset of species used in our phylogeny, Type A fruits are significantly less

PAGE 34

34 variable among species in diameter than Type B fruits, as predicted. However, in analyzing variation of this type we coul d not control for phylogeny and the significance of the difference in variation between Type A and B fruits disa ppeared when all species were included. Returning to prediction #1, an alterna tive hypothesis for the observed non-random combinations of fig color and size is physiolo gical. Because the production of reproductive parts, including fruits, can be costly to plan ts (Jurik 1983; Reekie and Bazzaz 1987), and because green fruits can photosynthesize even after they are ripe, green fruits may have lower costs of production than fruits of other colors (Cipollini and Levey 1991). Given that large fruits must be more costly to produce than small fruits, green p hotosynthetic fruits should be favored over fruits of different colors. However, a positive carbon balance in green fruits has been found only at high light levels and the balance becomes negati ve at low levels (Cipollini and Levey 1991). Because light levels may change considerably during a plants lifetime, it seems unlikely that extra photosynthetic tissue of la rge fruits would be selectivel y advantageous, unless the plant grew exclusively in brightly lit habitats and ri pe fruits were fully exposed to the sun. Also, considering the large surface area provided by leaves, it seems unlikely that photosynthetic ripe fruits would benefit the parent pl ant sufficiently to increase its fitness relative to other plants with fruits of different color and similar size. The sensitivity analysis for Maddisons Concentrated Changes test (Maddison 1990) helped us identify which source of variation has the stronger impact on our results, given uncertainty in both the phylogene tic reconstruction and the mode of fruit trait evolution. The largest impact in results was generated by switching color and si ze as independent variables in the Concentrated Changes test (Maddison 1990, Tabl e 1). When using si ze as the independent variable, the strength of the correlation varied widely, especial ly between the conservative and

PAGE 35

35 non-conservative reconstructions. We think that the choice of th e independent variable affects the results in our study because when size is the independent variable, few changes of color occur on branches with small-sized fruits, primar ily since small-sized fruits appear in more terminal branches. When so few changes occur (a small sample size for the purpose of the Concentrated Changes test), calculating the pr obability of the observed number of gains and losses seems to be very sensitive to small cha nges in the number of gains and losses. This sensitivity highlights the importance of understand ing the mode of fruit trait evolution (i.e., which trait evolves first) before one can dr aw conclusions about the importance of seed dispersers for the evolution of fruit traits. Ev en though the test used to decide which variable should be used as independent wa s fairly inconclusive, observing the pattern of gains and losses as reconstructed using a parsimony criterion gi ves us more confidence that color can be considered the independent variable. An important caveat of our methodology is that our categorization of colors was based on the perception of colors by humans. Birds, howev er, see the world differently than humans; they can detect UV (Bennett et al. 1994; Bennett and Thery 2007; Bowmaker et al. 1997; Das et al. 1999; Hart 2001; Schaefer et al 2006). Likewise, cone type and color perception can vary widely in mammals (Bowmaker and Hunt 2006; Peichl 2005). Unfortunately, data that incorporate the entire range of wavelengths refl ected by fruits, and colo r perception by different frugivores (Schaefer et al. 2006) ar e not available in the literature for the fig species included in our study. The technology necessary to quantif y reflectance in the field has only recently become available (Altshu ler 2001; Altshuler 2003). Another limitation of this study is the oversimplification of fr uit traits and disperser guilds used to define the syndromes. Fruit size and colo r were the only fruit tra its readily available in

PAGE 36

36 the literature and easily comparable among studies and, hence, we were unable to include traits such as odor, location on the tree, texture, crop size, among other traits th at may be potentially important for frugivores to find fruits (Dominy 2004; Dominy et al. 2001). Because we included only color and size, a broad definition of disperse r guilds was needed, as it was not feasible to define disperser guilds more narrowly on the basis of their potential sel ective pressure on just fruit color and size. Despite potentially important differe nces among different taxa of mammalian frugivores such as bats, rodents, or primates, we cons ider it safe to assume that, because most frugivorous mammals can manipulate fruits using their forelimbs and teeth, and because they are generally larger than birds, ma mmals should be able to consume larger fruits than can birds. As for color, despite minor diffe rences in avian color receptors (Bowmaker et al. 1997; Das et al. 1999), birds have a much more complex visual system than mammals and, in general, a higher variety of cone cells, which ar e responsible for color vision (Bennett and Thery 2007; Bowmaker et al. 1997; Bowmaker and H unt 2006). Hence, we consider it a safe assumption that birds can exert a stronger select ive pressure on fruit color than mammals can, which are often nocturnal and have less color re ceptors in their retina. Considering only fruit color and size and broadly defined disperser guild s makes our analyses a conservative test of the Dispersal Syndrome hypothesis. In short, the patterns of correlate d evolution that we found were identified despite this generalization, not because of it. Another drawback in our defini tion of disperser guilds is that we did not incorporate the potential characteristics of the pale ofauna that figs have interacted with historically. Modern day figs appear to date back at least 60 million years (Rnsted et al. 2005) and they have almost certainly interacted with frugivor es that differ from todays frugi vores, at least in body size (e.g., extinct New World megafauna, Janzen and Martin 1982). However, we do not have any reason

PAGE 37

37 to believe that historic birds and mammals exer ted selective pressures that were significantly different from those of current birds and mammals, and any disc ussion on this would be highly speculative. Even with this limitation, we be lieve our conclusions re main solid because we predicted how current birds and mammals should a ffect the evolution of fruits, and our results generally supported our predictions. Our results can be used to generate read ily testable hypotheses about which frugivores consume a given type of fig. For example, it can be predicted that the figs of Ficus phaeocyse and F. benjamina bearing type A colors and classified in the small size category, are dispersed most commonly and efficiently by birds, whereas F. botryocarpa, F. copiosa and F. dammaropsis, all of which bear type B colors and ar e in the large size category, are dispersed most commonly and efficiently by mammals. Qualitative data from a published source (Shanahan et al. 2001b) and from fieldwork (S. Lomscolo, unpublished data) on fruit removal by animals for most fig species in our study s how that most figs are consumed by a wide diversity of frugivores, including bi rds and mammals (Table 2-2). At first glance, this may seem to contradict the importance of any particular gr oup of frugivore in shapi ng the evolution of fruit traits. However, the data in Table 2, although highly informative, should be taken with caution because they do not provide information that w ould allow one to assess the potential of a given frugivore as a selective force. Such informati on would include fruit removal rate, gut treatment, deposition of seeds, and overall frugivore effectiveness (sensu Schupp 1993) That said, some general patterns arise from Table 2-2. For example, large fruits often have contrasting colors, a combination that seemingly falsifies the Disper sal Syndrome hypothesis. These species tend to occur in the Old World tropics, where larger bird s such as hornbills and cassowaries occur. These may be exceptions that prove the rule. We predict that contrasting colored fruits should

PAGE 38

38 be smaller in places where such large frugivorous birds are less common, i.e., in the New World. However, more data from the New World tropics is needed to test this idea. We encourage the collection and application of quantitative data that address disperser effectiveness to determine whether our findings match what is found in nature. Taken together, our results sugge st that differences among taxa of seed dispersers have influenced the evolution of fruit traits. In partic ular, we conclude that the correlated evolution of fruit size and color support the Di spersal Syndrome hypothesis.

PAGE 39

39 Figure 2-1. Bootstrap support and Bayesian pos terior probabilities on the maximum likelihood phylogeny of 64 Ficus species used for analyses. Node s with strong bootstrap support in maximum likelihood analyses (greater than or equal to 70%) are indicated with a circle; nodes with st rong support using Bayesian posterior probabilities (greater than or equal to 0.95) are i ndicated with a triangle

PAGE 40

40 Figure 2-2. Box and whiskers plot comparing th e mean diameter of Type A (red, purple) and Type B (green, yellow, brown, and orange ) fruits (Janson 1983) for the set of Ficus species in our phylogeny (nA = 36, nB = 27) and for all Ficus species found in the literature (nA = 56, nB = 72). The boxes upper and lowe r sides indicate the first and third quartiles, and the horizontal lines inside the boxes marks the median. The vertical lines show the largest and smallest observations that fall within 1.5 times the box size from the nearest quartile

PAGE 41

41 Figure 2-3. Maximum likelihood phylogeny of 64 Ficus species. Color is mapped onto the phylogeny and was done using a parsimony crite rion in MacClade (the only criterion available in this software), and is one of the reconstructions tested with the Concentrated Changes test. Black branches represent type A fruits, grey branches represent type B fruits, a nd white branches represent ambiguous reconstruction. At the tip of the branches, character state for diameter is mapped for each fig species. Forty two species show the combinati ons predicted by the Dispersal Syndrome hypothesis (type A colors and small size, or type B colors and large size), whereas 22 species show a mismatching combination. The ancestral state for fruit diameter was reconstructed as large. Cha nges of diameter from large to small (called gains, and represented by white dashes) and small to large (called losses, and represented by grey dashes) are marked on the branches where they are hypothesized to have occurred according to the parsimony criteri on used for the reconstruction. For this particular reconstruction, out of a total of 12 gains and 5 losses of small size, eight and four, respectively, occurred on branches with type A fruits. For branch support value using maximum likelihood bootstrap and Bayesian posterio r probabilities see Fig. 1.

PAGE 42

42

PAGE 43

43 Table 2-1. Results from Maddisons Concentr ated Changes tests showing all combinati ons of algorithms and character state reconstructions performed on the maximum likelihood tree using color and size as the independent variable (I and II, respectively). Independent variable Non-conservative reconstruction Conservative reconstruction deltran acctran deltran acctran Color Total: 17 gains, 0 losses Total:12 gains, 5 losses Total:17 gains, 0 losses Total:12 gains, 5 losses Resolving option to count gains and losses Branches: 12 gains, 0 losses Branches: 9 gains, 4 losses Branches:13 gains, 0 losses Branches: 9 gains, 4 losses minstate maxstate minstate maxstate minstate maxstate minstate maxstate p-value 0.012 0.005 0.011 0.007 0.002 0.002 0.034 0.015 Diameter Total:12 gains, 2 losses Total:7 gains, 7 losses Total:9 gains, 7 losses Total:9 gains, 7 losses Resolving option to count gains and losses Branches: 5 gains, 0 losses Branches:3 gains, 1 loss Branches:2 gains, 3 losses Branches:3 gains, 2 losses minstate maxstate minstate maxstate minstate maxstate minstate maxstate p-value 0.131 0.085 0.049 0.067 0.539 0.544 0.197 0.204 Total: total number of gains and losses using each resolvi ng option (DELTRAN, which delays changes, or ACCTRAN, which accelerates changes) Branches : number of gains and losses on the br anches with the character state of interest. When considering color as the independent variable, Branches is the total gains and losses of small size that occur on bran ches with Type A (brightl y colored) fruits. When considering size as the indepe ndent variable Branches is the total gain s and losses of Type A colors on branches with small-sized fruits. MINSTATE and MAXSTATE are options for reconstructing character st ates at ambiguous nodes. MINSTATE uses the lowest trait value, whereas MAXSTATE uses the largest trait value.

PAGE 44

44 Table 2-2. Data on fruit diameter, color, animal consumers, and distribution for th e 64 species included in our phylogeny, as f ound in the literature. Superscripts next to the Ficus species indicate the source from which data on fruit size and color were obtained. The column diameter code indicates whether the sp ecies was classified as small or large based on the 10mm cutoff point described in the text. The co lumn Fruit type indicates whether the speci es fell in the Type A or B color class based on Jansons (1983) classifica tion, as explained in the te xt. Data on animal consumers were obtained mainly from Shanahan et al. (2001b), and secondarily from fieldwork by S. Lomscolo. The superscript next to the distribution indicates the source from which this information was obtained. Ficus species Diameter (mm) Color Size code Fruit Type Avian consumersh Mammalian Consumersh Other consumersh Distribution F. adenosperma 10a Greena Small B No data found No data found New Guineab F. asperifolia 10a Yellow or reda Small A or B Pigeons OW primates Fish Africaa F. aurata 11a Orangea Large A Passeriforms Rodents Borneoa F. auriculata 60g Greeng LargeB OW bats Chinab F. benjamina 8a Purplea Small A Galliforms, Pisciforms, hornbills, Cuculiforms, parrots, pigeons, Passeriforms NW and OW bats, OW primates, civet, rodents Borneoa F. bernaysii 12f Redf Large B Cassowary, hornbills parrots, pigeon, Passeriforms Marsupials, OW bats New Guineab F. binnendykii 8a Reda Small B Pisciforms, hornbills, pigeons, passeriforms OW primates, rodents Peninsular Malaysiaa F. botryocarpa 35a Greena large B Marsupials New Guineab

PAGE 45

45 Table 2-2. Continued F. botryoides 29e Rede large A No data found No data found Madagascarb F. condensa 20a Greena LargeB OW bats Borneoa F. congensis 25g Redg LargeA No data found No data found Africaa F. conocephalifolia 28.5g Redg LargeA OW bats New Guineab F. consociata 12a Reda Large A Pisciforms, hornbills, Cuculiforms, pigeons, Passeriforms Tree shrews, Asiatic Black bear, OW primates, rodents Peninsular Malaysiaa F. copiosa 29a Yellowa Large B Cassowary, hornbills, Fig parrot Marsupials, OW bats New Guineab F. dammaropsis 60.5g Greeng LargeB Honey eater Marsupials, OW bats New Guineab F. deltoidea 9a Reda SmallA OW primates Borneoa F. elastica 10g Greeng Large B Pigeons, Passeriforms OW primates New Guineaf F. exasperata 9a Reda Small A Hornbills, Turaco OW bats, OW primates, Artiodactyla Africaa F. glumosa 9a Reda Small A Pisciforms, pigeons, Passeriforms OW primates Africaa F. grossularoides 10d Redd SmallA Passerifroms Borneob F. hispidioides 35a Greena LargeB OW bats New Guineab F. insipida 40a Greena Large B NW primates, NW bats, Procyonids, tapirs Turtle, fish Venezuelab F. itoana 34a Greena LargeB OW bats New Guineab

PAGE 46

46 Table 2-2. Continued F. lutea 15a Yellow or reda Large A or B Hornbills, pigeons, Passeriforms OW bats, OW primates, elephants Africaa F. macrophylla 17.5g Purpleg Large A Cuculiforms, parrots, pigeons, Passeriforms OW bats Australiac F. maxima 25a Greena LargeB NW bats, NW primates Brazilb F. microcarpa <10g Purpleg Small B Pisciforms, hornbills, Cuculiforms, parrots, pigeons, Passeriforms Marsupials, OW bats, OW primates New Guineab F. microdictya 10a Reda SmallA No data found No data found New Guineab F. mollior 16a Greena LargeB No data found No data found Australiac F. mucuso 40g Orangeg Large B Pigeons OW bats, OW primates, elephants Africaa F. natalensis 13g Redg Large A Pisciforms, hornbills, Turacos, pigeons, Passeriforms OW bats, OW primates, Artiodactyla, elephants Africaa F. nodosa 28a Greena LargeB Hornbills Marsupials, OW bats New Guineab F. nymphaeifolia 22a Green or reda Large A or B NW bats South Americaa F. parietalis 15a Reda Large A Pisciforms, hornbills, Passeriforms OW primates, rodents Bruneic

PAGE 47

47 Table 2-2. Continued F. pellucidopunctata 9a Purplea Small A Pisciforms, hornbills, Trogoniforms, pigeons, Passeriforms Borneob F. pertusa 14a Yellow or reda Large A or B Guans, Pisciforms, Coraciforms, Trogoniforms, parrots, Passeriforms Marsupials, NW bats, NW primates, Procyonids Venezuelab F. phaeosyce 9g Redg Small A Cassowary, hornbills, pigeons, Passeriforms Marsupials, OW bats New Guineab F. prasinicarpa 7a Reda SmallA Parrots Solomon Islandsb F. punctata 49a Reda LargeA OW primates, rodents Borneob F. pungens 7g Redg Small A Hornbills, parrots, pigeons, Passeriforms Marsupials, OW bats New Guineab F. racemosa 25e Green or rede Large A or B Galliforms, Pisciforms, hornbills, Cuculiforms, parrots, pigeons, Passeriforms OW bats, OW primates, bears, civet, Artiodactyla, rodents Australiab

PAGE 48

48 Table 2-2. Continued F. religiosa 10g Purpleg Small A Pisciforms, hornbills, Cuculiforms, pigeons, Passeriforms OW bats, OW primates, Artiodactyla Lebanonc F. robusta 75g Greeng LargeB No data found No data found New Guineab F. rubiginosa 12.5g Greeng Large B Cuculiforms, parrots, pigeons, Passeriforms OW bats Australiac F. ruginerva 37.5d Oranged LargeB Hornbills OW primates, rodents Borneob F. schumacheri 5a Greena SmallB No data found No data found French Guyanac F. semivestita 33f Greenf LargeB OW bats New Guineab F. septica 30a Greena Large B Coraciiform, pigeons, Passeriform Marsupials, OW bats, OW primates, rodents New Guineab F. subgelderi 12d Oranged Large A Pisciforms, hornbills, parrots, pigeons, Passeriforms Tree shrew, OW primates, rodents Bruneic F. superba 20g Purpleg Large A Galliforms, Pisciforms, hornbills, Cuculiforms, parrots, pigeons, Passeriforms OW bats, OW primates, civet, rodents Borneob

PAGE 49

49 Table 2-2. Continued F. sur 19a Reda Small B Pisciforms, hornbills, Turacos, pigeons, Passeriforms OW bats, OW primates, Artiodactyla Africaa F. sycomorus 26a Yellowa Large B Pisciforms, hornbills, Turacos, pidgeons, Passeriforms OW bats, OW primates, civet, hyrax Africaa F. tinctoria 14a Reda Large A Hornbills, parrots, pigeons, Passeriforms Marsupials, OW bats, OW primates, rodents New Guineab F. trachypison 7a Reda Small A Hornbills, parrots, pigeons, Passeriforms Marsupials, OW bats New Guineab F. uncinata 22a Reda LargeA Rodents Bruneic F. vallis-choudae 33a Yellow or reda Large B Pigeons OW bats, OW primates Africaa F. variegata 27a Green or reda Large A or B Cassowary, hornbill, parrots, Passeriforms Marsupials, OW bast, OW primates, rodents Australiab F. virens 8a Yellowa Small A Cassowary, Pisciforms, hornbills, Cuculiforms, parrots, pigeons, Passeriforms, OW bats, OW primates, carnivores, rodents New Guineab F. virgata 8a Orangea Small A Cuculiform, pigeons, Passeriforms OW primates New Guineab

PAGE 50

50 Table 2-2. Continued F. wassa 10a Reda Small A Cassowary, hornbill, parrots, pigeons, Passeriforms Marsupials, OW bats New Guineab F. xylophylla 25a Orangea Large A Pigeons, Passeriforms Bruneic F. xylosycia 13a Reda Large A Pigeons, Passeriforms New Guineab F. yoponensis 18a Greena Large B Passeriforms NW bats, NW primates, Procyonids, Artiodactyla, rodents Panamaa Superscripts denote the source fr om which the information was obtained: a. Do miny et al. (2003); b. Weiblen (2000); c. Jouselli n et al. (2003) ; d. Laman and Weiblen (1998) ; e. Mi ssouri Botanical Garden website, at http ://mobot.mobot.org/W3T/Search/vast.html; f. George Weiblen, pers. comm.; g. Pers. obs.; h. Shanahan (2000) NW stands of New World, and OW stands for Old World.

PAGE 51

51 CHAPTER 3 SIGNAL CONVERGENCE IN FRUITS: A RESULT OF SELECTION BY FRUGIVORES? Introduction To overcome the cons traints imposed by immobility, many plants depend on animals for dispersal of pollen and seeds. Animal pollinator s and dispersers cue on different plant signals to find flowers and ripe fruits. Yet the great diversity of plant signa ls is poorly understood -it has been variously attributed to chance, to envir onmental factors, and to the sensory ecology of mutualistic animals (Ben-Tal and King 1997; Gumbert et al. 1999; J ohnson and Steiner 2000). The latter, in particular, consti tutes a long-debated central questi on in the field of plant-animal interactions: Do animal mutualists shape th e evolution of plants, causing plant signals to converge into pollination and seed dispersal syndromes (Janson 1983; Johnson and Steiner 2000; Ridley 1930; van der Pijl 1961; 1969)? Syndromes are sets of flower or fruit traits hypothesized to evolve under the selective pressure of pollin ators and seed dispersers, respectively. Because many unrelated plants are visited by similar mutualists, and because visitation rate is commonly density-dependent, an adaptive expl anation for signal convergence is that plants sharing the same suites of visitors might benefit from convergi ng on a similar phenotype to attract mutualists. This Signal Convergence hypothesis is the main focus of this study. Nevertheless, the general conve rgence of fruit or flower tr aits to syndromes that are reliably associated with (i.e., presumably signa l to) specific animal vectors has been questioned. Detractors of the syndrome hypothesis argue that plants tend to benefit from different types of animal mutualists and, conversely, plant-visiti ng animals do not tend to specialize on only one type of plant (Herrera 1985; Johnson and Stei ner 2000; Waser et al. 1996 ). This argument may be particularly relevant to the evolution of seed dispersal syndromes because plants show a lower degree of specializati on towards seed dispersers than they do towards pollinators (Bluthgen et al.

PAGE 52

52 2007). From an adaptive viewpoint, the low degree of specialization in s eed dispersal will result in disparate selective pressure s on fruit appearance and will th us reduce the likelihood of signal convergence. Some authors claim that apparen tly non-random combinations of fruit traits are evidence of syndromes (Gautier-Hion et al. 1985 ; Janson 1983; Kalko et al. 1996; Knight and Siegfried 1983; Korine et al. 2000; Willson et al 1989). However, when those combinations of traits have been scrutin ized in phylogenetically controlled studies, the syndromes are generally not supported (Fischer and Chapman 1993; Herrera 1987). This difference in approach and opinion has generated much controversy. We argue that one reason studies addressi ng the role of frugivores in shaping fruit evolution have reached disparate conclusions is th at fruit color, one of the most important fruit traits reflecting differential f ood selection by distinct frugivore assemblages (Voigt et al. 2004), has been measured inappropriately. Color as a vi sual signal might be pa rticularly important in defining syndromes because signals represent the firs t step in the interactio ns between plants and animals. Natural selection should favor signals (fr uit colors) that elicit a response in the receiver (the frugivore) that increases the fitness of the sender (the plant) (Endle r 1992). From this, we can derive that the efficiency of a signal not only depends on the signals form (i.e., the spectral reflectance of the fruit), but also on the receivers perception (i.e., the frugivores neural circuitry to interpret the color signal) (Endler 1992). N onetheless, evolutionary studies on fruits and frugivores have typically reported color qualitativ ely and according to hu man visual perception (Fischer and Chapman 1993; Gautier-Hion et al. 1985; Janson 1983; Lord et al. 2002), or they have ignored color completely (Jordano 1995), pr obably because of the difficulty of quantifying color.

PAGE 53

53 Another important property of signals is thei r conspicuousness, which is quantified as the signal-to-noise ratio, and in the ca se of visual signals is direc tly correlated with the contrast between the color of the object and that of the surrounding environment. Natural selection should maximize conspicuousness to the receiver. In the case of fruits, the contrast between the color of the fruit and the foliage background should be greatest to the disperser to which the signal is targeted. Although fru it conspicuousness has been shown to be important for fruit detection (Burns and Dalen 2002; Regan et al. 2001; Schmidt et al. 2004a; Sumner and Mollon 2000a), the contrast between the color of the fr uit and that of its background has never been incorporated in a study of dispersal syndromes. Our study focuses on the role of fruit colo r as a signal to two of the most common frugivorous taxa, primates and bi rds. Under the Signal Convergen ce hypothesis, we test whether fruit color can be used as a signal to seed dispersers. We predict that: 1. Primates and birds should be able to dis tinguish visually between fruits that are primarily dispersed by one of them and those that are primarily dispersed by other group; 2. Fruit colors should be more conspicuous to the disperser (primates or birds) to which they are signaling. 3. The color of fruits dispersed mainly by primates should converg e to colors that are different from those dispersed mainly by birds. One assumption of our study is that primates and birds can have differential effects on plant fitness according to their differences in se ed treatment and recruitment requirements of different plants. Although support is scant, some published studies suggest that this is a fair assumption (Clark et al. 2001; Poulsen et al. 2002).

PAGE 54

54 To test prediction 1 of the Signal Convergence hypothesis we use vision models to analyze fruit signals according to the color vision capabili ties of the main types of frugivores, primates and birds (Osorio and Vorobyev 1996; Regan et al. 2001; Vorobyev and Osorio 1998). These vision models approximate visual capability of specific frugivores based on their cone sensitivity (see Materials and Methods). To te st prediction 2, we define cons picuousness as the contrast in color between the fruit and the foliage backgr ound (Endler 1992) and use vision models to approximate conspicuousness of fruits to each frugivore type. To test prediction 3, we compare the fruit colors of bird and primate dispersed pl ants belonging to the same family and genus. To test these predictions, we use the largest spectr al data set of fruit colors analyzed to this date. For 406 plant species w ith known fruit consumers, we demonstrate that primates are physiologically capable of discriminating between fruits dispersed by them and those dispersed by birds based solely on color. Likewise, we s how that birds can do th e same and that this pattern transcends geographic areas, and persists even when phylogenetic effects are controlled. We also discuss why fruit colors were not always more conspicuous to the targeted disperser. Materials and Methods Fruit Color and Disperser Data Data on reflectance spectra of fruits and fr uit con sumers for each plant species were obtained mainly from the literatu re and partly from our fieldw ork. A total of 406 species were included, of which 177 were prim ate-dispersed species from Guyana (111), Uganda (46), and Venezuela (20), and 229 were bird -dispersed species from Florid a (29), Europe (99), Venezuela (31), and other unspecified regions (70) (Regan et al. 2001; Schaefer et al. 2007; Schmidt et al. 2004a; Sumner and Mollon 2000b). In all cases color was quantifie d using a portable spectrometer and a special light source that reflec ts equally in all parts of the spectrum. Fruit reflectance was obtained as the pr oportion of reflectance by a white standard. The spectrometer,

PAGE 55

55 light, and white standard varied among studies but were consistent within each study. Although the use of different equipment adds an uncontro lled source of variation in the data, we considered that, if the predicted patterns were found in reflectance of primate versus bird dispersed fruits, this would be despite the lack of consistence in color measuring devices, and not because of it. Spectra were averaged in 5 nm intervals from 400-700 nm. Unfortunately, data on fruit reflectance under 400 nm (the UV range) were not available for primate-dispersed fruits, so we left this part of the visual range of birds out of our analyses. Our results are therefore conservative because differences between primate and bird perception of fruit color are likely more pronounced if UV vision is accounted for. Vision Model Because color is in terpreted by the brain based on the input of light to different types of cones cells with different sensi tivity to specific wavelengths (Bowmaker and Hunt 2006; Regan et al. 2001), we calculated the light input to the different cone type s for each frugivore group. Our vision model combines the reflectance spectrum of each fruit with the co lor vision ability for each frugivore (represented by the peak sensitiv ity of the cone cells for each frugivore) and yields three variables pe r species of fruit, corresponding to th e input to the brain by the short, medium, and long wavelength cones of the frugivore (Fig. 3-1). For primates, we used cone peak sensitivities correspondi ng to humans, (440, 540, and 570 nm), which are very similar to those of some other trichromatic primates (e.g., Alouatta spp.) (Bowmaker and Hunt 2006; Regan et al. 2001). For birds, we used cone peak sensitivities for the Blue Tit ( Cyanistes caeruleus, 372, 456, 544, and 609 nm). We calculated the relati ve cone excitation values, called quantum catches (Fig. 3-1), which represen t the basic sensory input for colo r vision (Kelber et al. 2003). Although cone peak sensitivity varies within each of these groups of frugivores (Bowmaker et al.

PAGE 56

56 1997; Bowmaker and Hunt 2006), these differences are small relative to differences between birds and primates. All cone excitation values were transformed to the natural logarithm to achieve normality and were analyzed with Discrimi nant Function analysis (DFA) to test whether bird-dispersed fruits and primate-dispersed fruits were significantly discriminated using the quantum catch of bird and primate cones. Discriminant Function Analysis finds the combination of the variables that best discriminates between the groups, and a llows us to identify each variables contribution to the discrimination. These variable combinations are called Linear Discriminant Functions (LDF). If birds can visually discriminate between fruits commonly dispersed by them and those commonly dispersed by primates, a significant discrimination should be obtained using the variables yielded by our vi sion model for birds and, likewise, for primates. Data Analysis Prediction 1: Color discrimination by birds and primates : To evaluate the accu racy of the DFA and its predictive power, we split our da ta set of 406 species into two parts and derived the LDFs with one part (species from Guyana, Florida, and unspecified geographic regions), called the training DFA, and tested the predictive value of LDFs on the other part of our data, a completely independent data set (species from Uganda, Europe, and Venezuela), called the validation DFA. This is a powerful test of the validity of the discrimina tion between primate and bird fruits by primate and bird vision, becau se the pattern of discrimination will remain significant only if it is consistent across very different biogeographica l regions, such as the Paleotropics, Paleo-temperate area s, and the Neotropics. We also tested the predictive value of the linear discriminant functions by classifying the cases as bir d or primate fruits based on their score on the LDF derived with the trainin g DFA. The higher the percentage of species that were well classified (known bird-dispersed species correctly classified as a bird-dispersed

PAGE 57

57 species), the higher the predictive value is of the linear discrimi nant functions. This was done for the training and th e validation data sets. Prediction 2: Conspicuousness of the fruits to frugivores : To test whether primatedispersed fruits were more contrasting against the foliage than bird-dis persed fruits from a primates visual perspective, we first calculated the chromatic contrasts of 86 primate fruits and 188 bird fruits against the foliage background as the Euclidean distance between the color of a fruit and the color of the structure against whic h it is viewed by the frugivore. Spectra were normalized to the same brightness by dividing the reflectance at each wavelength by the total reflectance for each species. This provides contrasts due to the color of the fruits and not to their brightness (Schmidt et al. 2004a), which can vary greatly with light availa bility. The resulting contrasts for primate and bird fr uits were entered into the prim ate-vision model and the results were compared using a Mann-Whitney U test. To ensure that the differences between fruit types are actually due to differences in reflectance by the fruit itself and not by the background, which is beyond the scope of the Signal Convergence hypothesis, we also calculated fruit contrasts against the mean of the background for the 256 speci es included in the analysis and compared the results with another Mann-Whitney U test. Prediction 3: Phylogenetic control : Closely related species are not independent data points because they share a common ancestor. We controlled for such phylogenetic effects by doing a pair wise comparison of the mean score on the linear discriminant function for bird and primate dispersed species within each family of plants. To do such a comparison, we ran another DFA with all the species together and then only for the 22 families that had both bird and primate dispersed species. We calculated the m ean score for bird and primate dispersed species per family. Based on the results of the training a nd validation data sets, pl ants dispersed by birds

PAGE 58

58 scored consistently higher on the LDF separating the groups, and therefore we tested whether the scores of bird-dispersed species were significan tly higher than primate-di spersed species of the same families using a one-tailed Wilcoxon Signed Rank test. We performed the same test on the mean scores within 12 genera for which we had bird and primate dispersed species. If the difference between bird and primate-disperse d species within each family or genus was significant, we concluded that the pattern of discriminati on observed was not driven by phylogeny. Results Prediction 1: Color Discrimination by Birds and Primates The tra ining DFA resulted in significan t discrimination between birdand primatedispersed fruits when using bird cone sensitivity peaks (Wilks = 0.680, numerator d.f. = 3, denominator d.f. = 205, p < 0.00001) and prim ate cone sensitivity peaks (Wilks = 0.649, numerator d.f. = 3, denominator d.f. = 205, p < 0.00001) (Fig. 3-2). Classification accuracy was fairly high, with 82% of 111 prim ate-dispersed species classified correctly and 76% of 98 birddispersed fruits classified corre ctly. Given the high negative load ing of the medium wavelength variable (representing green light that primarily stimulates the medium-wavelength cone) and the high positive loading of the short wavelength vari able (representing blue light) (Table 3-1), our results suggest that primate-disper sed fruits tend to be greener th an bird-dispersed fruits, which tend to be more blue or purple. The validation DFA also showed significant discrimination between bird and primatedispersed fruits for bird (Wilks = 0.638, numerator d.f. = 3, denominator d.f. = 192, p < 0.00001) and primate cone sensitivities (Wilks = 0.616, numerator d.f. = 3, denominator d.f. = 192, p < 0.00001) (Fig. 3-2). Classi fication accuracy was also high, with 80% of 66 primate-

PAGE 59

59 dispersed species being classified correctly and 82% of 130 bird-dispersed fruits classified correctly. Prediction 2: Conspicuousness of the Fruits to Frugivores Contrary to the predictions under the Signal Convergence hypothesis, primate-dispersed fruits are less conspicuous (i.e., contrast less ag ainst the foliage) to primates than are birddispersed fruits (Mann-Whitney U = 4,273, p < 0.001, n = 256). Unfortunately, we could not test conspicuousness of bird and primate-disper sed fruits to birds be cause UV reflectance is required to obtain an accurate measure of conspi cuousness to birds. We are confident that the difference in contrasts is due to differences in fruit color and not background color because the contrasts calculated with the mean reflectance of the background for all 256 species included was also significantly different from a primate s perspective (Primates: Mann-Whitney U = 8,584, p < 0.001, n = 256). Prediction 3: Phylogenetic Control Our phylogenetic test comparing the mean scores on the linear discri minant functions of bird and primate-dispersed fruits within each family using a Wilcoxon Signed Rank test showed that bird-dispersed fruits tende d to score higher on the LDF than did primate-dispersed fruits, although the difference was not are large as that found at the ge nus level (Family level: onetailed test, V = 77, p = 0.056 for both bird cone peak sensitivity, and V = 77, p = 0.056 for both primate cone peak sensitivity; Genus level: one -tailed test, V= 10, p = 0.021 for bird cone peak sensitivity, and V = 7, p = 0.005 for prim ate cone peak sensitivity). Discussion Our results generally support the first prediction of th e Signal Convergence hypothesis, which states that frugivores should be able to discrim inate between the color of fruits dispersed by them and those of fruits dispersed by other frugivores. Based on the model representing the

PAGE 60

60 visual capabilities of primates a nd birds, we conclude that th ese frugivores can differentiate between fruits typically disperse d by primates and those by bird s. This is important because color could not be considered a signal to frugivores if frugivores were not capable of distinguishing colors. Even though the discrimination is significant, a considerable amount of overlap exists between birdand primate-dispersed fruits. This result is expected because, although a frugivore may more often consume fruits that are more c onspicuous to them than those that are more cryptic, they will not feed exclusively on fruits with conspicuous colors. Frugivores can also find fruits based on senses other than vision, and they can feed on fruits found by chance from a nearby fruiting tree. Plants with exclusive seed dispersers are exceptional, and neither the Signal Convergence, nor the Dispersal Syndrome hypothese s imply exclusiveness of frugivores. If syndromes are an evolutionary response to seed dispersers, they should result from the response of a plant species to se lection by those seed dispersers that improve its fitness, even though many other not-so-efficient frugivores may still feed on the same plant. If we could include other cues used by these frugivores to find fruits in our model, we would likely obtain sharper discrimination. The fact that, as shown by our study, the color of some bird-dispersed fruits fall well within the range of a primates vi sual perception, does not mean that bird fruits are necessa rily conspicuous to primates and that they can easily find them. Primates may need a combination of cues to make the fruits conspicuous and fully available to them, such as color, odor, shape, texture, a nd/or location on the tree (Dominy 2004; Dominy et al. 2001). Our study is limited to testing only one tr ait (color), which we use to define dispersal syndromes and, although we believe color is one of the most important cue used by diurnal frugivores to find fruits, we acknowledge th at it is certainly not the only one.

PAGE 61

61 It is surprising that the quantum catch on th e long wavelength cone, representing red light vision, was unimportant in discriminating between birdand primatedispersed fruits. Red fruits are commonly associated with bird dispersal more than with primat e dispersal (Janson 1983; Wheelwright and Janson 1985; Willson and Whelan 1990), but so are black fruits. Red and black reflectance spectra differ dramatically in the amount of long wavele ngth light reflected and hence, bird dispersed fruits resu lt in highly variable quantum catch for the cone responsible of long wavelength absorbance. Hence, reflectanc e at long wavelengths is not a consistent discriminant because the within group variance is too high. What red and black fruits do have in common is their high contrast against the foliage, which may be why birds are often associated with fruits of those colors (Schmidt et al. 2004a). Our second prediction, that the fruits dispersed by a frugivor e should be more conspicuous to that frugivore than fruits not dispersed by them was not supported by our data. Primate fruits seem to be less conspicuous to primates than bird fruits. We propose several alternative explanations for the lack of c ongruence of the results for primat es and birds, and the lack of complete support to our hypothesis, and we divi de them into adaptive and non-adaptive explanation from the perspective of plant-frugivore communication. The adaptive explanations are consistent with the Signal Convergence hypothesis. Fi rst, it is possible that, while birds have such well developed color vision and may be pr imarily visual foragers, primates use other signals besides color, such as odor, size, and text ure of fruits to assess their edibility (Dominy 2004; Laska et al. 2007). This may therefore derive in a stronger se lective pressure on conspicuousness by birds than by primates. If this is the case, we predict that a similar test on conspicuousness to birds, bird-dispersed fruits should be more conspicuous to birds than primate-dispersed fruits. Als o, by including fruits from varied geographical regions from the

PAGE 62

62 New and Old World, we included a wi de range of primates that ha ve very different color vision capabilities (Bowmaker and Hunt 2006; Regan et al. 2001; Surridge et al 2003). In order to answer the specific question of whether the fr uit/background color combination is optimized for dispersal by primates, we should differentiate between dichromatic and trichromatic primates, and we should also take into account the pol ymorphisms found among platyrrhine (i.e., New World) monkeys. If primates exert a strong sele ctive pressure to maximize the signal-to-noise ratio in the fruits they eat, the signal will most likely be seen when studying specific groups of primates with well defined and similar color vision capabilities. This may not be as crucial when testing conspicuousness to birds because birds have a much highe r within-guild similarity in cone sensitivities among very different taxa (H art 2001), and therefore th eir potential selective pressure on fruit color may be similar among di fferent bird species. We mentioned in the Introduction that variation in co lor vision among disperser types is greater than that within disperser type. However, within gr oup variability, even if small, may affect the results of the test for fruit contrasts more than that of color discrimination. Second, it is also possible that our measure of conspicuousne ss is simplistic. Sumner and Mollon (2000b) argue that, at equal contrast between background and fruit color, larger fruits are more conspicuous to primates simply because they are larger. That is, large-fruited plants may not be under as strong selective pressure as sm all-fruited plants to maximize the difference between background and fruit color. Under this hypothesis we predict that, among the primatedispersed fruits, species with lower score on LDF 1 will be larger than species with higher LDF 1 score, because LDF 1 is negatively associated with reflectanc e of green light, which is less conspicuous against the foliage. This is sim ilar to what Sumner and Mollen (2000b) showed when they correlated fruit size and reflectance at longer wavelength. However, his negative

PAGE 63

63 correlation can also be explained by the effect of smaller, more visual frugivores, such as birds feeding on smaller fruits, as predicted by the Dispersal Syndrome hypothesis, rather than the sole effect of selection on conspic uousness by one type of frugivore (Lomscolo et al. 2008; Sumner and Mollon 2000b). Whatever the ultimate explana tion for this correlation, we suggest that conspicuousness to frugivores should be measured as a function of fruit si ze, in addition to the contrast against the background. The non-adaptive explanations consider roles of color that are independent of plantfrugivore communication. Willson and Whelan (1990) identify 13 non-exclusive hypotheses that may explain fruit color patterns, of which six ar e completely unrelated to seed dispersers and include defense against pathogens or seed predat ors, as well as physiological, or phylogenetic constraints. Although we will not attempt a thorough discussion of all Willson and Whelans (1990) hypotheses, we want to generally address so me that seem particularly relevant to our study. First, the anthocyanins responsible for red and black colors in fruits have been shown to reduce fungal growth (Schaefer and Schaefer 2007) and, thus, at traction of seed dispersers by anthocyanines may be a by-product of the antimicr obial properties of these pigments. Red is also inconspicuous to certain inse cts that eat fruit but do not disper se seeds, so this pigment may have evolved to hide fruits from predators -not to make them more conspicuous to seed dispersers (Willson and Whelan 1990). In this case color should be considered an exaptation (Gould and Vrba 1982) instead of an adaptation to endozoochory. Second, fruit pigmentation has been hypothe sized to decrease some physiological constraints, such as thermoregulation. Because dark colors absorb more heat than light colors, it has been hypothesized that dark fruits have higher metabolic rates and therefore can reach maturity faster, reducing exposure to fruit predators. However, dark colors would also cool

PAGE 64

64 faster as the radiation emitted by a body must equal the radiati on absorbed (Kirchhoffs Law of Thermal Radiation), so dark fr uits should be at a disadvantage in many tropical rain forests where clouds cover the sun light long periods of time. Yet da rk-colored fruits are extremely common in these cloudy environments (Wheelwr ight and Janson 1985; Willson et al. 1989; Willson and Whelan 1990). The thermoregulation hypothesis does not explain the advantage of dark-colored fruits in the shaded understory either, where they are also very common (pers. obs.) Lastly, as Wheelwright and Janson (1985) point out, dark pigments are onl y deposited at the end of the ripening period and, hence, may contribute ve ry little to the development of the fruit. Third, fruit colors have also been hypothesi zed to be related to the plants metabolic constraints. Because the production of reproductive parts, includi ng fruits, can be costly to plants (Jurik 1983; Reekie and Bazzaz 1987), and because green fruits can photosynthesize even after they are ripe, green fruits may have lower costs of production than fruits of other colors (Cipollini and Levey 1991). This is especially relevant to large-fr uited species: given that large fruits must be more costly to produce than sm all fruits, green photosynthetic fruits should be favored over fruits of different colors. However, a positive carbon balance in green fruits has been found only at high light levels and the balance becomes nega tive at low levels (Cipollini and Levey 1991), which does not explain the impor tance of green fruits in environments of varying light levels, such as a rain forest or the forest understory. Last but not least are phylogeneti c constrains. Fruit traits in general have been shown to be highly conserved (Jordano 1995), probably due to the long generation time of plants in comparison with seed dispersers (Herrera 1985). Our comparison between species in the same genus suggests that similar colors among species dispersed by primates and those dispersed by birds are the result of convergence rather than sh ared common ancestry, at least in the 12 genera

PAGE 65

65 included. Although not statistically significant, the trend found in 22 families also suggests the independent evolution of fruit color. We emphasize the importance of measuring color objectively, using newly available portable spectrometers (Altshuler 2001; Altshule r 2003) and to interpret these colors based on the visual capabilities of the frugivores under study (Chittka and Raine 2006). Although primates have sometimes been merged with birds in a bird-primate syndrome because no differences in color were found between fruits dispersed by these two groups of frugivores (Gautier-Hion et al. 1985), we show that, while pr imates and birds overlap in the color of the fruits they consume, primates disperse more green fruits than do birds, and that birds disperse more blue fruits than do primates. These results match those of earlier st udies in fruit choice by frugivores (Janson 1983; Wheelwright and Janson 1985), even when those studies classified color from a human perspective. We believe that this match, despite such different methods of measuring color, is because we are testing frugiv ores whose color vision cap ability is similar to that of humans. This is, of course, with th e non-trivial exception of UV perception in birds, which we were unable to consider due to lack of UV reflectance information in the literature. We emphasize that using vision models may be esp ecially important if the visual system of the frugivores under study (e.g., a dichro matic primate) differs significan tly from our visual system. In the case of this study, if fr uit reflectance in the UV range were included in our analysis, the discrimination between primate and bird-dispersed fruits using all four of a birds cone peak sensitivity would almost certainly be sharper than that shown here. The results from the primates perspective should not change becau se we would be incorporating reflectance information on a part of the spectru m that primates cannot see.

PAGE 66

66 Although we cannot be sure that our model ex actly represents how birds and primates perceive color, as color vision entails much mo re than cone excitation, our model provides an objective approximation of color vi sion by these frugivores. This is the first study that addresses the Signal Convergence hypothesis in a fruit-frugi vore system, using such an objective measure of color and such a large data set of reflectance spectra (the largest analyzed to date). Color is an important mediator in plantanimal communication (Chittka and Raine 2006; Johnson and Steiner 2000; Schaefer et al. 2004, this study) and we believe that learning the signaling role of color will shed light on unde rstanding how animals and plants affect each others fitness and evolution. This can only be done by using vision models that incorporate the receptive and cognitive processes of the receivers, which adds much needed objectivity in the assessment of how plant signal to animals (Schaefer et al. 2004).

PAGE 67

67 Figure 3-1. Reflectance spectra of tw o different fruits (left) combined with the peak cone sensitivities of a frugivore (middle ) yields the quantum catch for the three cone type s for each fruit (right). The quantum catch values approximate the frugivores perception of color.

PAGE 68

68 Figure 3-2. Results of the Discriminant Functi on analysis (DFA) based on (A) primate cone excitation and (B) bird cone excitation. Bl ack lines correspond to the training DFA and grey lines to the validation DFA. Dash ed lines represent primate-dispersed fruits and solid lines represent bird-d ispersed fruits. Primate-disp ersed fruits tend to have negative scores on the linear discriminant function, because they are greener than bird-dispersed fruits, which have more of a blue hue than primate-dispersed fruits

PAGE 69

69 Table 3-1. Loadings of the quantum catch variab les for primate and bird cone sensitivities for the training DFA Loading on LDF 1 Variable Primate vision Bird vision Short (Blue light) 1.396 1.234 Medium (Green light) -2.758 -1.797 Long (Red light) 0.771 -0.071 Proportion of variation 1.00 1.00

PAGE 70

70 CHAPTER 4 REVISITING THE DISPERSAL SYNDRO ME HYPOTHESIS: A PHYLOGENETIC COMPARATIVE STUDY OF FRUIT TRAITS IN FICUS (MORACEAE) Introduction Seed dispersal by vertebrates occurs in tropical temperate, xeric, and m oist environments alike, both on land and in aquatic habitats (Adams et al. 2007; Godinez-Alvarez et al. 2002; Herrera 1987; Korine et al. 2000; Shanahan et al. 2001b; Spiegel and Nathan 2007; Traveset et al. 2004). Its selective advantag e in various geological scenario s and historical periods is evidenced by its independent evolution in many pl ant lineages, from seed ferns, to gymnosperms and angiosperms (Herrera 1989; 2002b). It is particularly ubiquitous in moist tropical environments, where over 90% of woody species are dispersed by animal s (Herrera 2002b), and where it has been proposed as on of the most important proximate mechanisms explaining high tree diversity (Janzen 1970; Terborgh et al. 2002). Yet, the question How important are seed dispersers to fruiting plants? may elicit fundamentally different answers from the ecologists and the evolutionary biologists in the field. While ecol ogists argue convincingly th at seed dispersal is important to the fitness of fruiting plants, many e volutionary biologists argu e just as strongly that there is little evidence se ed dispersal matters to plants (i.e., that it has influenced the evolution of fruit traits). Among the most cited lines of evidence for the c ontemporary role of seed dispersal in plant and animal communities are that it links the last phase of a plants reproductive cycle to the first phase of population recruitment (Herrera et al 1994; Nathan and Muller-Landau 2000; Rey and Alcantara 2000; Schupp and Fuentes 1995), contributes disproportionately to forest recovery after natural disasters (Hjerpe et al. 2001; T hornton et al. 1996; Whittaker and Jones 1994), and dramatically reduces density-dependent mortality near the parent plant (Fragoso et al. 2003; Howe 1986; Howe and Miriti 2000; 2004; Howe et al. 1985; Schupp 1988). There is also

PAGE 71

71 growing evidence that dispersal limitation (i.e ., the inability for seeds to colonize sites appropriate for germination) may be a major process limiting plant recruitment, affecting plant spatial distributions and contributing to the main tenance of biodiversity (but see Clark et al. 2007; Clark et al. 1999; Hurtt a nd Pacala 1995; Muller-Landau et al 2002; Murray and Garca-C 2002; Turnbull et al. 2000). Thus, plants spec ies without dispersers may face extinction (Cardoso da Silva and Tabarelli 2000; Chapma n and Onderdonk 1998; Cordeiro and Howe 2001; 2003). Conversely, reductions in fruit a bundance may threaten frugivore populations (Moegenburg and Levey 2002; Redford 1992). But evolutionary evidence is less clear-c ut. Some argue th at seed dispersal is consequential to the evolution of fruits and point to the apparently non-random combination of fruit traits (Cipollini et al 2002; Gautier-Hion et al. 1985; Janson 1983; Korine et al. 2000), called dispersal syndromes, th at match presumed preferences and fruit-handling abilities of different sets of frugivores. However, most evidence to date suggests that seed dispersal is inconsequential in shaping fruit traits, as factors other than seed dispersers can better explain fruit traits these factors include phylogeny (Fischer and Chapman 1993; Herrera 1987; 1989; 1992; Jordano 1995), architectural c onstraints (Herrera 1987; 2002a), and growth form (Herrera 1987). Moreover, fruit traits presumably sel ected for by frugivore taxa such as birds or mammals are not always appealing to their put ative dispersers (Tamboia et al. 1996), casting doubt on the validity of adaptive explanations for th e association of those traits with particular sets of dispersers. It is puzzling that re searchers have failed to find support for the potential role of seed dispersers in the evolution of fruits because many assumptions that are central to the adaptive role of fruit traits ho ld true, namely: 1. seed dispersal is often important for plant reproduction

PAGE 72

72 (Fragoso et al. 2003; Harms et al. 2000; Howe and Miriti 2000; 2004); 2. seed dispersers can exert disparate selective pressures on fruiting plants because disperser guilds treat seeds differently (Clark et al. 2001; Godinez-Alvarez et al. 2002; Poulsen et al. 2002), and may therefore affect plant fitness diffe rentially; 3. disperser guilds percei ve fruit traits differently and can differentiate fruit traits well (Hodgkison et al. 2007; Korine and Ka lko 2005, Chapter 3, this dissertation), making it feasible for plants to use fruit traits to at tract beneficial consumers and/or to deter detrimental consumers, and 4. some trai ts such as fruit size and color are not strongly phylogenetically conserved, suggest ing potential for selection by seed dispersers (Jordano 1995; Lomscolo et al. 2008 (Chapter 2, this Dissertation); Voigt et al. 2004). We argue that one of the reasons for the cont radictory conclusions of evolutionary studies is the analytical approach used by different studies. First, some studies have failed to consider the role of phylogeny in determining the similarity in fruit-trait combination among plant species (Janson 1983; Kalko et al. 1996; Korine et al. 2000), which fails to account for the fact that species may not be independent data points. Second, those who controlled for the nonindependence of species have either focused at hi gher taxonomic levels or have used estimates of the phylogenetic relatedness of sp ecies (instead of explicit ph ylogenies), methods that often decrease the power to find a correlation between di sperser type and variation in fruit traits, when it exists (Bremer and Eriksson 1992; Fischer a nd Chapman 1993; Jordano 1995). Third, some of the most influential studies come from temp erate regions (Herrera 1987; 2002a) where fruit diversity is limited compared to tropical regions; therefore great variation in tropical fruit traits remains unexplained. Fourth, thorough studies that investigated the variation of fruit traits with dispersal mode have analyzed fruit traits separately (Herrera 2002a ; Jordano 1995), whereas frugivores most likely assess w hole fruits based on many variab les simultaneously. Finally,

PAGE 73

73 dispersal guilds have often been defined very broadly (i.e., birds vs. mammals, Janson 1983; Jordano 1995), which hinders the definition of th e potential selective pr essure of frugivores on the fruit traits of interest. Perhaps the most glaring reason why evoluti onary studies are not conclusive on whether frugivores have selected for fruit traits is that some of the most important fruit traits used by frugivores to find and select fruits have been excluded or mismeasured. The most striking examples of this are fruit color and odor. Th ese traits are commonly used by plants to signal fruit ripeness (Sumner and Mollon 2000b) or nut ritional composition (Cat oni et al. 2008; Goff and Klee 2006; Schaefer et al 2008; Schaefer and Schmidt 2004) and are two of the most important cues used by frugivores to find and sele ct fruits (Kalko et al. 1996; Laska et al. 2007; Thies et al. 1998; Voigt et al. 2004; Wheelwright and Janson 1985; Willson and Whelan 1990). In particular, they are quantifie d from the perspective of human perception. This can generate significant and unintended biases, given that ot her animals sense the world differently than humans and that most plants use color and odor to signal to animals other than humans (Bennett et al. 1994; Schaefer et al. 2004). For example, most passerine birds can see UV light, while humans and most other mammals cannot (Be nnett and Cuthill 1994; Bennett and Thery 2007; Hart 2001; Peichl 2005). Likewi se, most mammals have a more developed sense of olfaction than humans and, especially, than birds (Clark 1991; Evans 2006; Heymann 2006; Wysocki and Preti 2004). Our work improves on previous studies by 1. fo cusing at the species level and controlling for the non-independence of species using a mo lecular phylogeny; 2. studying tropical plant genus with highly variable fruit traits; 3. using a multivariate phylogenetic comparative approach to simultaneously test for the correlated evolution of seven fruit characteristics that are predicted

PAGE 74

74 to be important for frugivores to find fruits, na mely fruit size, color, odor, softness, contrast against the background, location on the tree, and exposure away from foliage; 4. focusing on two well defined disperser guilds with clearly different potential for se lection on fruit traits; and 5. reducing human biases by using quantitative measur es of color using a spectrometer, and of odor through gas chromatography/mass spectrometry. We test two hypotheses for co-occurring sets of fruit traits. Firs t, the Phylogenetic Hypothesis: fruit traits are inherited from ances tors they are the result of shared phylogenetic history. Second, the Dispersal Syndrome Hypothesi s: fruit syndromes are the result of selective pressure by sets of seed disper sing frugivores that differ in beha vior and ecology. Our objective is to tease apart these hypotheses by determining how much variation in fruit traits is attributable to phylogeny and how much is attributable to selective pressure by frugivores. We focus on two well defined groups of frugivores, birds and bats, which are predicted to select for the following dispersal syndromes: 1. Passerine bird syndrome : Because passerine birds have a relatively narrow gape, lack teeth, have acute color vision (tetrachromatic, including a peak in the UV range, Bowmaker et al. 1997; Das et al. 1999; Ha rt 2001), a poor sense of smell, and usually perch to pick fruits, fruits preferred by passerine s should be small, visually conspicuous (hues of red, purple or black, contrasti ng against leaf or bark back ground), less odorous than batdispersed fruits, and presented among foliage along branches. 2. Bat syndrome : Because bats have teeth (and therefore can manipulate fruits with their mouth and eat them piecemeal), are mostly nocturnal (which limits their perception of color), use chemical cues to find fruit (Bianconi et al. 2007; Hodgkison et al 2007; Mikich et al. 2003), and gr ab fruits in flight or land on trees and crawl to reach fru its (Elangovan et al. 2001; Kalko et al. 1996), fruits preferred by them should be larger than bird-dispersed frui ts, odorous, and less visually conspicuous (hues of

PAGE 75

75 green, yellow, pale orange, or brown, non-contra sting against leaf or bark background). Also, they will grow along the trunk (caulicarpic) and, in general, they will be more exposed away from the foliage than fruits eaten by birds (Ridley 1930; van der Pijl 1969). We demonstrate that these two groups of frugivores have had an evolutionary im pact on fruit traits within Ficus. Methods Study Site and General Specifications Data were collected between September 2004 and Decem ber 2005 in a lowland forest near the village of Ohu (145 E, 5 S) in Madang Pr ovince, Papua New Guinea, (Fig. 4-1). Data on fruit color, color contrast against the background, odor, size, softness, a nd exposure to frugivores (location of fruits on branches or trunk, distance from closest leaves, and peduncle length) for each of 42 fig species were obtained by averaging records from between 3 and 8 individuals per species. For each individual tree, between one and 10 fruits were measured and averaged for all traits. Study system: the natural history of figs Ficus (Mo raceae) provide a unique opportunity to study the evolution of fruit traits in relation to seed dispersers because its species are unusually variable in fruit color, size, presentation, and type of disperse r. Fig fruits are technically closed infructescences, that are hollow and shaped like an urn, most probably se lected to allow the characteristic internal pollination provided by wasps (Janzen 1979; Weib len 2002). (For more information on fig pollination and its relation with the seed dispersal mutualism, see Chapter 5, this dissertation.) Fig infructescences are called syconia, a term we use interchangeably with fruits because they are functionally fruits; i.e., they are comprised of seeds and nutritious pulp, and they attract and reward seed dispersing animals. They can be red, purple, black, pi nk, orange, yellow, green, or brown. Syconia range in diameter from 3-80 mm and grow from the tips of branches, at the

PAGE 76

76 base of leaves, on the trunk fr om ground level to the canopy, and even from the roots (Laman and Weiblen 1998, G. Weiblen pers. comm., S. Lomsclo pers. obs.). In terms of life form, figs can be free standing, epiphytic, hemi-epiphytic, or par asitic stranglers of ot her trees. They are also geographically widespread (Shanahan et al. 2001b). Mo re than 700 species are found worldwide in tropical forests, thereby regularly co-occurring with different sets of dispersers. Dispersers include bats, arborea l and terrestrial mammals, volant (passerines, hornbills, fruit doves, parrots, etc.) and flightle ss birds (e.g, Cassowaries), fishes and reptiles (Shanahan et al. 2001b). Figs are best known for their tight mutualism with pollinating wasps, in which one species of fig is typically pollinated by one species of wasp. The mechanism of pollination is quite similar in all species of Ficus : a gravid female wasp carrying with pollen enters the closed inflorescence, pollinates the flower s and lays eggs in some of the ovaria. Therefore, viable seeds develop from some flowers and wasp larvae develop from others (Janzen 1979). Because pollinators may influence flower anatomy (Temeles and Kress 2003) the fact that all fig species have virtually the same pollinator contributes to making Ficus a good study system: floral morphology is fairly constant and, therefore, mo st differences found in fruit characteristics among figs should be largely i ndependent of floral traits. In many forests and across continents, figs are considered a keystone resource because they produce large fruit crops that ripen during seasons when few other fruits are available (Lambert and Marshall 1991; Terborgh 1986). Figs may also be important to frugivore communities due to their high calciu m content (OBrien et al. 1998). On the flip side, frugivores appear important for figs because figs depend he avily on seed dispersal for regeneration (Titus et al. 1990).

PAGE 77

77 Figs have often been deemed generalists becau se they produce large am ounts of fruits with low nutritional quality, fruit year-round, and attract large numbers of dispersers from different guilds. These attributes have likely discouraged researchers fr om considering figs as a study system for detecting dispersal syndromes. Howeve r, when figs are studied with attention to species-specific differences, it becomes quick ly apparent that many species are almost exclusively visited by specific gui lds of frugivores, such as bats arboreal mammals, and birds (Kalko et al. 1996; Korine et al. 2000; Shanahan 2000; Shanahan and Compton 2001; Shanahan et al. 2001a), and only some of them fruit continuously (S Lomascolo, pers. obs.). Quantification of Visits to Figs All data on f ig consumption come from figs that contain viable seeds; these figs are commonly called seed figs (see Chapter 5). We recorded consumption of figs using two video cameras (Sony DCR-HC40), which were set up 3-5m away from the fruiting Ficus tree and visited only to change tapes or batteries. To record both diur nal and nocturnal frugivores, we started videotaping at 06:00 and ended at 10:30 and continued again between 18:30 until 23:00. We recorded nocturnal visits using an infrared light (Sony HVL -IRH2). We aimed one camera at the ground to record terrestrial frugivores a nd the other either at a branch or at the trunk, depending on where most of the ripe figs were located. We recorded between one and four individuals of 29 species of figs for a total of 1019 hours. Unless a frugivore was seen not to eat any fruit, all frugivores that ente red the tree were considered s eed dispersers. For some species for which fruit data were collected but no disp ersers were recorded, we obtained frugivore information from Shanahan et al. (2001b). On e species for which no frugivore was recorded, F. botryocarpa had been repeatedly pointed out by villagers as a comm only bat-dispersed species, so we also included it in the anal ysis as a bat-dispersed species.

PAGE 78

78 Fruit Traits Fruit color : W e quantified fruit color using a USB2000 portable spectrometer (Ocean Optics, Inc) and a PX-2 pulsed xenon light source which took reliable readings between 300 and 740nm. This spectrum includes the range of wavelengths visible by humans, plus the ultraviolet range. We scanned fruits using a sensor with six optical fibers illuminating the target fruit and a seventh fiber that returned the reflected light to the spectrometer. The scanning angle was fixed at 45 by using a black metal stand with a hole pos itioned at that angle. The metal stand also blocked external light. To keep the distance between the end of the sensor and the fruits constant we fastened a non-UV filtering microscope slide to the opening of the hole where the sensor was introduced. To obtain an average reflectance sp ectrum per fruit, we scanned each fruit three times in three different spots on the fruit. We calculated reflectance as the proportion of a Spectralon white reflectance standard (Labs phere, North Sutton, NH) and, hence, reflectance measures do not have units. The three spectra for each species were averaged every 5nm. We obtained a quantitative measure of fru it color for each species using a Principal Components Analysis (PCA), which summarized th e information for the complete spectrum by finding linear combinations of the original variab les -in this case, reflectance intensity every 5 nm (Clotfelter et al. 2007; Cuervo and Shine 2007; de Ayala et al. 2007; Endler 1990). This technique reduces multivariate data to several axes that summarize the variance in, and covariance among, the original variables. A separate PCA was done to obta in color variables for the group of 42 species of figs for which fru it data were collected, and for the group of 19 species of figs for which exclusive dispersal by birds or bats was recorded. These two groups were used in different statistical analyses to answ er separate questions (see below). In both cases the first principal component (P C) had high positive loadings fo r all wavelengths throughout the spectrum, and hence represents fruit brightness (Fi g. 4-2), i.e., the total am ount of light reflected

PAGE 79

79 by the fruit. Therefore, light fruits had high scores on PC 1 and dark fruits had low scores. Principal components 2 and 3 represent different hue s. Hue refers to the everyday meaning of color, summarizing differences in reflectance of different wavelengths. The second PC distinguishes reflectance in the red versus the UV-blue wavelengths because it had positive loadings for wavelengths from 300 to ~600nm, and negative loadings from 600 to 740nm. Therefore, UV-blue fruits had a high positive scor e on PC 2, whereas fruits reflecting in the red part of the spectrum had high negative scores. The third principal component distinguishes reflectance in the green versus red and UV-blue range because it had high positive loadings of wavelengths between 450nm and 640nm. Thus, green fruits had a high score on PC 3 and UVblue and red (and, therefore, purple) fruits had a low score. Th ese three PCs explained 91% of the variance in reflectance and all had eigenvalues greater than one, which means that each explained more variation than any single original variable alone. Although PCs 4 and 5 also had eigenvalues higher than one, they only explai ned 4 and 3% of the variation, respectively. Because reflectance in all parts of the spectru m of interest was summarized by PCs 1 to 3, we considered that additional PCs did not contribute su fficiently to the color description of each fig species. Hence, we use only scores on PCs 1 to 3 in subsequent analyses. Color contrasts were calculated as the Euclid ean distance between the color of a fruit and the color of the structure against which it was expos ed to the frugivore, which consisted of either leaves or bark, depending on the species. Spectra were normalized to the same brightness by dividing the reflectance at each wavelength by the total reflectance for each species. This generates contrasts due to the color of the fruits not to their brightness. These contrasts are called chromatic contrasts (Schmidt et al. 2004a) and are calculated as 2)]^(([QbQfD

PAGE 80

80 were Qf is the color spectrum of the fruit and Qb is the color spect rum of the background structure; is the wavelength in nm and the sum corresponds to the complete spectrum between 300-740nm. Fruit odor : To capture volatile compounds, we colle cted fruits in the field, brought them to the lab, and placed them inside bags made of inert material (Reynolds oven baking bags). A vacuum pump (Welch, model #2522B-01) sucked ai r into the bag throug h a carbon filter on one end and forced it out of the bag through a filter at the opposite end filled with Super Q (Alltech), which adsorbed volatiles emitted by the fruits. Fo r a given collection of fruits in a bag, we ran the pump for four continuous hours, which began no more than three hours after the fruits had been collected from the tree. To extract fruit volatiles from the Super Q filters, we flushed methylene chloride through the filters. We then used gas chromatography to separate and quantify the amount of different (unknown) vola tiles. We also used mass spectrometry to identify some of the compounds. Because there we re too many compounds (n ~75) to identify or analyze separately and because we were most interested in determining whether bird and mammal fruits differ in overall production of volatile compounds, we calculated the total amount of volatile compounds by summing the areas under each compounds peak for each species. We standardized for number and size of fruits in a bag by dividing the amount of each volatile contained in the sample by the total surface area of the fruits in the bag that produced the volatile. Fruit size, and softness : We measured fruit diameter with calipers to the nearest 0.5 mm and we used this as an estimate of overall fruit si ze. To measure fruit softness, we determined an arbitrary scale between 1 (hardest; fruit surface di d not indent when pressed against fingers) to 4 (softer; fruit surface indented easily when pressure was applied to fruit between the fingers). To

PAGE 81

81 achieve consistency in this arbitrary measure th e same researcher recorded this measurement throughout the study. Fruit size and softness, toge ther with fruit color and odor will be referred to as appearance traits, to differentiate them fr om traits reflecting expo sure to the frugivores (exposure traits). Fruit exposure traits : To quantify fruit position away from the foliage, we estimated the distance between the fruit and the ne arest leaf to 1.0 cm for the highe st and the lowest fruit in the tree, and averaged these values to incorporate the variation within each tree. Another exposure variable was dichotomous: fruits growing from branches versus those growing from the trunk. A third measure of exposure was peduncle length, the distance (+ 0.5 mm) between the base of the fruit and the point where the peduncle attached to the branch or trunk. Data Analysis All statistical tests were done using R version 2.6.0 (R De velopm ent Core Team 2008), except for the Evolutionary Principal Compone nts Analysis (Dyreson and Maddison 2006), for which we used Mesquite 2.01 (Maddison and Maddison 2007). Identification of correlation in fruit traits : Principal Components Analysis (PCA) was used to identify the main axes of variation in the fruit data. In this case, PCA was used to identify the main differences between bird and ba t-dispersed figs, and to define which variables covaried along the axis separating figs disperse d by these two guilds. Because variables were measured in different units and differed by up to three orders of magnitude, they were standardized to zero mean and unit standard deviation. We performed two different PCAs, one on the 19 species of figs that were dispersed by either birds or bats (hereafter Disperser PCA), and one on the 42 species of figs for which we collected fruit data (hereafter Fruit Trait PCA). The Disperser PCA was done to objectively define fruit syndromes by identifying the suites of traits that differed be tween bird-dispersed and

PAGE 82

82 bat-dispersed figs. Species with mixed dispersal were not included in this analysis because they were predicted to have intermediate characte ristics between species exclusively dispersed by birds or by bats, and our goal was to identify the differences between bird and bat fruits. We tested whether the difference between bird and bat dispersed species on the PC axis explaining the most variation between the two groups was significant using a Mann-Whitney U test. The Fruit Trait PCA included all the fig species for which fruit trait data were collected, and it included species dispersed by birds, by bats, by bo th, and those for which disperser data was not available. The goal of the Fruit Trait PCA was to explore the overall pattern of variation in the fruit variables and to identify whether the trait correlation observed in the bird and bat-dispersed species could be generalized to a wider set of fig species. Also, we explored whether fig species with mixed dispersal had intermediate trait values between bird and bat fruits. Phylogenetic control : To test whether the pattern of fruit trait variation was most likely explained by phylogeny or could be considered adaptive, we performed an Evolutionary Principal Components Analysis (EPCA) (Dyres on and Maddison 2006; Sch lick-Steiner et al. 2006). This EPCA is based on e volutionary transitions (the di fference in trait value between each species and the most recent ancestor, as reconstructed by a parsimony criterion, minimizing the sum of squared changes) instead of the raw variables used in a standard PCA. While standard PCA maximizes the variation in the traits among species, EPCA maximizes evolutionary change incurred by e ach species after they split from their ancestor, thus avoiding the problem of non-independence of species (Dyreson and Maddis on 2006). Just as explained above for PCA, EPCA was completed on the 19 sp ecies set of bat and bird-dispersed species (Disperser EPCA) and on the 42 sp ecies set (Fruit Trait EPCA). To test whether there was a

PAGE 83

83 significant difference between bird and bat disp ersed species on the EPC 1 resulting from the Disperser EPCA we used a Mann-Whitney U test. A second method of testing the role of phylogeny in explaining variation in fruit traits is to correlate that variation with genetic relatedne ss among species (Voigt et al. 2004). If what determines covariation of fruit traits (i.e., di spersal syndromes) is phylogenetic history, then closely related plant species should share combin ations of fruit traits. If, instead, dispersal syndromes result from divergent selective pressu res of different types of seed dispersers, similarity of fruits between any given pair of species should not be related to the genetic relatedness of those species. For each possible pair wise comparison of the 42 fig species for which we collected fruit data, we calculated genetic relatedness between all possible pairs of taxa, quantified in PAUP* (Swofford 2003) as th e number of character changes along the tree between each pair of species corrected by th e model of evolution chosen to build the phylogenetic tree. This genetic difference between two taxa is ca lled patristic distance and is proportional to the sum of branch lengths between them. We also calculated the difference in fruit appearance between all possibl e pairs of taxa as the absolute value of the difference in the score on the PC 1 resulting from the Fruit Trait PCA, which was the most important in separating bird and bat-dispersed species. We then tested for a correlation between these estimates of genetic relatedness and difference in fruit appearance (n = 861 pairs) using a Mantel test (Mantel 1967). The Mantel test avoids the problem of non-independence of species by first calculating two distance matrices, one based on genetic dist ance and the other based on difference in trait appearance, and then randomly permuting ro ws and columns within each matrix. The probability of the correlation observed, as quantified by r2, is calculated based on the null distribution created by the randomization.

PAGE 84

84 Volatile compounds in figs with different dispersers : A Discriminant Function Analysis based on presence/absence of 73 volatile compounds was used to identify the most important compounds distinguishing fig species dispersed pr imarily by birds (5 fig species), bat (13 species), bandicoot (2 species), and by more than one of these groups (10 species). To test the significance of the discrimination, we calculate d a Kruskal-Wallace (K-W) statistic for the difference in the score on Linear Discrimina nt Function (LDF) 1, which mainly separated between the groups in the original data matrix, then randomized the matrix and re-calculated a K-W statistic for the random matrix. Column to tals were fixed to cont rol for common volatile compounds, while row totals were allowed to va ry. This process was repeated 10,000 times to obtain a random distribution of the K-W statisti c and then compared our observed K-W statistic to the random distribution. Phylogenetic reconstruction Although four phylogenies of the genus Ficus have been published (Jousselin et al. 2003; Lom scolo et al. 2008; Rnsted et al. 2005; Weib len 2000), none have all the species for which we collected fruit data. We therefore r econstructed our own phylogeny, using DNA sequences from GenBank and some from N. Rnsted (Depar tment of Medicinal Chemistry, University of Copenhagen). We constructed one tree for the 19 species exclusively dispersed by birds and bats (necessary for the Disperser EPCA), and one for the 42 species for which we had data on fruit traits (necessary for the Fruit Tr ait EPCA) (Fig 4-3). A third tr ee that included 29 species with known dispersers (the 19 species exclusively dispersed by birds and bats, and 10 species with mixed dispersal) was constructed in order to map dispersal mode (Fig. 4-4). We used a maximum likelihood criterion, which searches for th e phylogenetic tree that best fits the data, given a particular model of evolution. The model of evolution that best described our data was determined using the Akiake Information Criterion from the model set examined by

PAGE 85

85 MODELTEST 3.6 (Posada and Crandall 1998). We conducted a heuristic search in PAUP* with 10 random sequence additions using the appropriat e model (GTR+G for all three data sets) and parameters recommended by MODELTEST 3.6 (Posada and Crandall 1998). We assessed support on the phylogeny using 500 maximum like lihood bootstrap repli cates using GARLI 0.951 (Zwickl 2006). Unresolved nodes were resolv ed for further comparative analyses based on published phylogenies. In the few cases where the species were not included in any published phylogeny, we resolved it randomly. Several random combinations were te sted and all yielded similar results, so all results reported are ba sed on one resolution, as seen in Figure 4-3. Results Visits to Fig Trees Out of the 29 fig species that were videotap ed, w e obtained disperser data for 15. Data on frugivore consumption of 13 additional species were obtained from Shanahan et al (2001b), and one from the knowledge of local villagers. Most species were dispersed by bats (14 species), birds (5 species), or both (10 sp ecies). Two species were recorded as exclusively bandicoot dispersed and they were included in the explorat ory analysis of volatile compounds in bat, bird, bandicoot and bird dispersed frui ts. Two observations were too fe w to justify the addition of a Bandicoot Syndrome, so we did not in clude it in other analyses. Based on a parsimony reconstruction of dispersal mode we identify at least three switches from bat to bird dispersal and from bat to mi xed dispersal (Figure 44). Although many sister species share the same dispersal mode, this r econstruction suggest dispersal mode has changed independently many times in this group of figs. Disperser PCA and EPCA The first principal com ponent (hereafter PC) explained 38% of the va riation in the data, and is associated, in decreasing order of magn itude, with contrast against the background (-),

PAGE 86

86 softness (-), distance from leaves (+), growth fr om branches (-), peduncule length (+), size (+), and brightness (+) (Fig. 4-5A, Ta ble 4-1). Bird-dispersed specie s tend to have lower scores on PC 1 than bat-dispersed species, suggesting that dispersed species are more contrasting against the background, softer, more often growing amo ng the foliage and on branches, have shorter peduncules and are smaller than bat-dispersed speci es (Fig. 4-5A). All of these characteristics match those predicted for bird and bat syndrom e fruits. The second PC explained 17% of the variation and it seemed to main ly separate two fig species ( F. semivestita and F. edelfeltii ) from the rest. We saw no identifiable biological relevan ce of this separation with respect to disperser type, and therefore we will not report this PC further. Controlling for phylogeny, the EPCA reveals a similar pattern (Fig. 4-5B), which is confirmed by a highly significant correlation between the scores of each fig species on PC 1and EPC 1 (r2 = 0.77, d.f. = 17, p-value < 0.0001). The loadings for contrast against the background, softness, distance to closest leaf, peduncule length, size, and did not change substantially between PC 1 a nd EPC 1, although their order of importance changed slightly (Table 4-1). The fru it traits that showed most difference between PC 1 and EPC 1 were fruit brightness and odor, which became more important in distinguishing between bird and bat figs when phylogenetic eff ects were controlled a nd location, and location of fruits on branches or trunk, which became less important (Table 4-1). Although the loading of the color variable defining UV-blue versus yello w (PC3) is not high, it is worth noting that its sign changed from positive in PC 1 to negative in EPC 1. Fruit Trait PCA and EPCA The Fruit Trait PCA with 42 fig species shows a pattern of ordination sim ilar to the Disperser PCA. The first PC explained 38% of th e variation in fruit data and was associated, in decreasing order of magnitude, with contrast against the background (-), growth from branches (-), peduncule length (+), distance from leaves (+ ), softness (-), size (+), and odor (+) (Fig. 4-6A,

PAGE 87

87 Table 4-2). Except for contrast against the back ground, color variables were not as important to explain the variation among fig species. The EPCA reveals a similar pattern to that of PCA, but the ordination is not as clearly matched to the PC axes, as it is in the Disperser EPCA. Fig species seem to be separated on EPC 2 in a simila r fashion to their spread on PC 1 (Fig. 4-6). Because EPC 1 and EPC 2 explain a very si milar amount of variation (21.4% vs 20.3%. respectively) and defining which is the main axis of variation is unimportant, and because bird and bat dispersed species are separated in both dimensions, we report both axes in detail. A strong, positive correlation (r2 = 0.79, d.f. = 40, p-value < 0.0001) between the score of each species on PC 1 and on EPC 2 suggests that the ordination of species does not change substantially after c ontrolling for phylogeny. EPC 2 show s higher loadings of exposure variables, as it is associated, in order of magnitude with distance fr om leaves (-), growth from the trunk (-), and peduncle length (-), and also with contrast against the background (+) and odor (-), whereas EPC 1 is largely based on differences in fr uit appearance, as it is associated, in order of magnitude, with brightness (+), size (+), and softness (-), and also with peduncule length (+)(Table 4-2). Most exposure variables have si milar loadings on EPC 2 to those of PC 1, but the relative importance is slightly different. In particular, fruit size and tw o of the color variables defining fruit hue became less important in EPC 2. On the other hand, the loadings of most appearance variables on EPC 1 are similar to those on PC 1. The correlation between genetic relatedness, as measured by branch length between pairs of species, and the difference in the score on the PC1 axis is signi ficant but explains less than 3% of the variation in the data (r2 = 0.023, d.f. = 859, p-valu e < 0.02) (Fig. 4-7). Volatile Compounds The DFA on volatile compounds resulted in sign ificant discrim ination between bird, bat, bandicoot, and mixed dispersal modes (Fig. 4-8A). The twelve most important compounds

PAGE 88

88 associated with bat-dispersed species are shown in Table 3. The scores on LDF 1, which separated bird from bandicoot and bat dispersed figs, were compared with a Kruskal-Wallis test and are significantly different from each other (K-W 2 = 20.38, d.f. = 3, p-value = 0.0001). The randomization of the data matrix and re-calcula tion of the Kruskal-Wallis statistic 10,000 times showed that the probability of obtaining by chance a discrimination as significant as that found with the original matrix was 0.0002. Figure 48B shows an example of the discrimination achieved with a random matrix. Volatile compounds more strongl y associated with bat dispersal were mostly esters (Pentanol acetate, Acetamide, Ethyl octanoate and Hexenoic acid, and a small, unidentified compound), with some sesquiterpenes (Muurolen e, Panasimsen, and Germacrene D) and an unsaturated alcohol (unidentified) which generally coincides with previous studies of volatiles in bat-dispersed figs (Hodgki son et al. 2007). Discussion Our phylogenetic com parative study of 10 vari ables describing fruit appearance and accessibility to frugivores suggests that fruit traits covary in a way predicted only under selective pressure by frugivores. We believe our results are particularly r obust for two reasons. First, we controlled for phylogenetic effects using a species-level phylogeny. Although phylogenetic effects can be controlled in other ways (F ischer and Chapman 1993; Herrera 1987; Jordano 1995; Voigt et al. 2004), using a species-level phylogeny is curre ntly the most accurate manner for estimating the true phylogenetic history of a given taxon. Second, we used a multivariate phylogenetic comparative analysis, which allowed us to search for patterns in many variables simultaneously without violating assumptions of independence. This approach provided an holistic analysis of the multiple traits that frugivo res perceive and use to find and assess fruits. It is simplistic to assume that a frugivore assesses, for example, the size of a fruit independently of

PAGE 89

89 its color, odor, or position in the tree. Just as the conspicuousness of any given fruit color is higher in large than small fruits (Sumner a nd Mollon 2000b), conspicuousness almost certainly depends on a combination of many traits that togeth er make the fruit relatively easy or hard to find by a frugivore, based on its overall perception of the world. Our results strongly suggest that bird and bat syndromes indeed exist in figs. In Papua New Guinea, figs in the Bird Syndrome are dark er, more contrasting ag ainst the background, less odorous, less exposed away from the foliage, smalle r and softer than figs in the Bat Syndrome. This list generally matches the set of traits previously suggested for Bird Syndrome and Bat Syndrome fruits (Knight and Siegfried 1983; Sh anahan et al. 2001b). Our approach, however, allows us to conclude that this particular combin ation of traits reflects e volutionary transitions -i.e., changes undergone by each fig species after split ting from an ancestral species. Therefore, these sets of traits are likely adaptive. Birds and bats also showed an association wi th figs growing from branches or directly from the trunk, respectively, but the importance of these variables was dramatically reduced in the EPCA, which may suggest that the location of figs on a tree is most strongly determined phylogenetically. Although severa l alternative hypotheses have b een proposed to explain the colors of fruits (Willson and Whelan 1990), or th e correlated evolution between, for example, fruit color and size (Cipollini and Levey 1991; Lomscolo et al. 2008 (Chapter 2, this Dissertation)), we are not awar e of any alternative hypothesis th at can explain the correlated evolution of fruit size, bright ness, contrast against the bac kground, odor, softness, and exposure away from the foliage. The most parsimonious e xplanation for the correlated evolution of these six traits is natural selection by seed dispersers.

PAGE 90

90 The same general pattern is apparent in a much larger data set of fig species, although data on consumers are not available for all species. This suggests that Dispersal Syndromes exist and are adaptive, at least for the se t of 42 fig species included in th is study. Our results generate testable predictions about potentia l consumers and seed dispersers of fig species, especially those falling in the extremes of the ordination shown in Figure 4-6. Our results also show that frugivores overlap in the fruits they eat. Many fi g species that fall close to the bird extreme of the dispersal syndrome axis (EPC 2 in the Frui t Trait EPCA) were reco rded to be also (e.g., F. microcarpa, F. phaeocyse, F. pungens, F. wassa, ) or even exclusively (e.g., F. subulata, F. virgata ) dispersed by bats. These exceptions do not undermine our argument of the importance of frugivore selective pressure be cause the Dispersal Syndrome hypothesis does not assume exclusiveness of frugivores. The assump tion is, instead, that th e strongest selective pressure will come from the overall most effi cient (sensu Schupp 1993) disperser. Because our disperser data came from tree visi ts recorded in the field or re ported in the literature, we are likely considering only the most common disperser(s) of each fig species. Therefore, an important assumption behind our analysis is that the most common seed disperser is the most efficient seed disperser. Admittedly, this is an oversimplified view many processes important to plant fitness occur after a fr ugivore consumes a fruit. We ar gue that the pattern we observed holds despite this generalization, not because of it. A better test of the Dispersal Syndrome hypothesis would consider dispersers that directly increase the fitn ess of the dispersed plant (i.e., increase seedling survival). We believe that if th ose data were used in a test similar to the one we have presented here, the Dispersal Syndrom es we described here would be even more pronounced.

PAGE 91

91 Another test of phylogenys influence on co-var iation in fruit traits was the correlation between genetic relatedness and fruit appearance (i.e., scores on PC 1 derived from the Fruit Trait PCA). Although closely rela ted species tended to be more similar in appearance than nonclosely related species, less than 3% of the ove rall variation in fruit trait combination is explained by genetic relatedness. Closely relate d species will always be more similar to each other than to random species becau se, by definition, they shared a much longer history. What is striking to us is that 97% of the similarity between figs must be explai ned by factors other than phylogenetic relatedness. This lack of correlation can also be visu ally appreciated in Figure 4-3, where the PC 1 score is mapped onto the phyloge ny. Our results support the idea that fruit appearance is more labile than it had been previously argued (Bre mer and Eriksson 1992; Herrera 1987; Jordano 1995) and, at least in figs, an important part of this variation appears to be explained by differences among dispersers. We want to note that our estim ates of the amount of variation in fig traits that can be accounted fo r by frugivores may be cons ervative because EPCA and the correlation between genetic relatedness with difference in trait characteristics they take into account what is not shared by closely related species. As it has been previously noted by Westoby and colleagues (1995) and also by Harvey and Pagel (1991), tra it variation that is inherited by common ancestry and is, thus, shared by closely related species, may also be adaptive. This is because closely related spec ies might tend to converge on similar adaptations because they respond similarly to similar selectiv e pressures because of their shared genes, and because they may tend to colonize similar habita ts and, hence, be subject to similar selective pressures. Moreover, a characteri stic that is shared by all member s of the same clade might have been fixed because of its high adaptive value. Many phylogenetic comparative methods give priority to phylogeny over ecology to explain shared characteristics, but because we show results

PAGE 92

92 from non-phylogenetic and phylogenetic comparisons, we allow the reader to draw their own conclusions on the extent to which shared charac teristics can be explai ned by either process We suggest that Dispersal Syndromes should be thought of in terms of traits that make a particular species of fruit more easy to find by frugivores, and not more attractive to them. Although a seemingly subtle semantic distinction, we argue that distinguis hing among traits that make a given fruit more or less attractive to a frugivore may lead to confusion and to false expectations under the Dispersal Syndrome hypothesis. Imagine, for example, a frugivore given a choice between fruits with different characteris tics. Based on previous experience with other fruits and the post-ingestional be nefits particular to those fruits (e.g., nutritional and healthrelated benefits, (Caton i et al. 2008; Goff and Klee 2006; Schaefer et al. 2008)), it may choose the one that is more attractive in those terms. However, a more natural scenario is one in which frugivores are not presented with choices but are instead searching for food to the best of their ability, given the limitations imposed by their physiology, anatomy, and fora ging behavior. If a frugivore, while searching for, or feeding on a preferred fruit with high nutritional quality, encounters by chance fruits that are easily accessi ble, it will most probably consume those fruits too, even if those fruits are otherwise not attractive, just because other fruits may be scarce. In this way, birds may consume bat-syndrome fruits, and vice versa. The more attractive fruit may be preferred over the less attractive by the frugi vore in a feeding trial during which the frugivore must choose between the fruits. But whenever fr uits are relatively uncommon, such preferences likely disappear if found both species will be equally consumed in nature. Instead, if a plant highly benefits from the dispersal of its seed s by a specific type of frugivore, the selective pressure on the plant will be to be found more readily by such a frugivor e. With this distinction in mind, the expectation of frugivore exclusiveness may be unrealistic.

PAGE 93

93 It is worth noting that fruit color per se, as described by hue, was relatively unimportant in discriminating between bird and bat fruits. Ho wever, two components of fruit color fruit brightness and contrast against the background were important in distinguishing figs dispersed by birds and bats. This result echoes results of previous studies that emphasize the role of contrast for bird-dispersed fru its (Schaefer et al. 2006; Schmidt et al. 2004a). Note, however, that in our dataset hue co-varied with brightness and may therefore still play a role (albeit a weak one) in defining Disperser Syndromes. Fig species with green and yellow fruits tended to have higher scores on PC 1 and EPC 1, and red and purp le fruits tended to have low scores on these axies, which were highly positively associated with brightness. Chromatic contrasts also tended to be associated with color, as red and black fruits have higher contra st against green foliage (Schmidt et al. 2004a). This suggests that thes e other dimensions of color are more directly associated with bat or bird dispersal than reflec ted in our analysis. It would be interesting to repeat this analysis using other quantitative measures of color, such as hue and chroma (defined as the purity of the color, Endler 1990), to determine whether bright ness and contrast still prevail over fruit hue in discriminating between bird and bat-dispersed fruits. Although bat-dispersal s eemed to be associated with specific volatile compounds, we did not test the evolutionary role of bats in selecting for those pa rticular compounds. Because some of these compounds are also present in some fig species when fig inflorescences are receptive to pollinating wasps (Grison-Pige et al. 2002a; GrisonPige et al. 2002b), it is possible that those compounds may have been selected for by fig pol linators, not dispersers Hence, until more detailed studies on the evoluti on of these compounds are available, we suggest viewing these compounds as an exaptation (Gould and Vrba 1982) not an adaptation, to frugivore attraction.

PAGE 94

94 Many fruit traits may be subject to selective pressure from multiple factors (Cipollini and Levey 1997; Mack 2000; Willson and Whelan 1990), but this does not undermine the importance of seed dispersers as a selective force that may generate and/or maintain fleshy fruit diversity. Any fruit trait may have several functions and therefore be subject to disparate selective pressures, generating complex patterns of trait covariation. Even so, traits with a strong influence on frugivore attraction will be subject to se lection by frugivores. Although our data do not allow us to conclude that any particul ar trait of bird or bat-dispersed figs originated from the selective pressure of these s eed dispersers (i.e., is an adaptation to seed dispersal), we believe that the correlated evolution of so many traits that were predicted a priori to play a role in frugivore attraction, strongly argues for an important role of s eed dispersers on the evolution of fruits as a whole and, hence, in expl aining the diversity of fleshy fruits.

PAGE 95

95 Figure 4-1. Map of Papua New Guinea in reference with a world map. Data were collected from areas surrounding the city of Madang, main ly from Ohu village (not shown).

PAGE 96

96 Figure 4-2. Loadings of the first three princi pal components (PC) on fruit reflectance spectra from 300 to 750nm, measured at 5nm interval s. PC 1 represents variation in mean brightness, PC 2 in green versus red colo ration, and PC 3 in UV and blue coloration.

PAGE 97

97 Figure 4-3. Maximum Likelihood tree of 42 fig sp ecies. Nodes with bootstrap support value 70 are marked with a star. The trait mappe d in the phylogeny is the score of each species on PC 1, the dispersal syndrome axis using a maximum parsimony criterion.

PAGE 98

98

PAGE 99

99 Figure 4-4. Parsimony reconstruction of dispersal mode. The number by the species names is the score on PCA axis 1, which separates figs according to dispersal mode. Disperser mode is shown in key.

PAGE 100

100

PAGE 101

101 101 Figure 4-5. Results of the Disper ser PCA (A) and EPCA (B). Bird-dispersed figs are represented by a solid triangle and bat-disp ersed figs are represented by a white triangle.

PAGE 102

102 102 Figure 4-6. Plot of PC 1 vs 2 for the Fru it Trait PCA (A) and EPCA (B) for 42 species of figs. The importance of the original variables in defining the PCs can be seen in Table 4-1. Species names are abbrevia ted as explained in Table 4-3. Red dots: bird-dispersed species. Blue dots: bat-disp ersed species. Yellow dots: mixed dispersa l. Fig species without dots had no data on disperser.

PAGE 103

103 Figure 4-7. Correlation between gene tic distance and difference in the score on the PCA axis 1 for all possible pairs among the 42 fig species for which fruit trait data were obtained. They ellipse defines a 95% confidence region.

PAGE 104

104 Figure 4-8. Discriminant Functions plot based on (A) or iginal data matrix for 30 Ficus species on presence/absence of 73 volatiles compounds, and on (B) the same matrix randomized, keeping the column totals fixed. Symbol s represent bird-dispersed ( ), bat-dispersed ( ), bandicoot-dispersed ( ), and mixed-dispersed ().

PAGE 105

105 Table 4-1. Loadings of variables for the Disp erser PCA and EPCA. Corresponds to Figure 4-5. PC1 PC2 EPC1 EPC2 Diameter 0.336 0.305 0.275 0.208 Softness -0.383 -0.331 -0.266 -0.204 PC1 0.205 0.236 0.555 0.204 PC2 -0.050 -0.416 -0.171 0.452 PC3 0.153 -0.036 -0.197 0.102 Odor 0.154 0.355 0.423 0.193 Peduncule length 0.346 0.061 0.341 -0.107 Trunk 0.261 -0.529 0.082 -0.574 Branch -0.366 0.314 0.010 0.363 Contrast against background -0.437 0.067 -0.316 0.083 Distance from closest leaf 0.368 -0.235 0.273 -0.376 Proportion Variance 0.378 0.172 0.260 0.244 Table 4-2. Loadings of variables for the Fruit Trait PCA and EPCA. Corresponds to Figure 4-6. PC1 PC2 EPC1 EPC2 Diameter 0.271 -0.461 0.515 0.013 Softness -0.308 0.402 -0.445 -0.122 PC1 0.170 -0.559 0.551 0.133 PC2 0.116 0.099 -0.292 0.020 PC3 0.171 0.174 -0.089 -0.012 Odor 0.218 -0.108 0.043 -0.191 Peduncule length 0.370 -0.077 0.301 -0.264 Trunk 0.339 0.312 0.012 -0.476 Branch -0.402 -0.300 0.090 0.451 Contrast against background -0.432 -0.116 -0.197 0.294 Distance from closest leaf 0.335 0.234 0.032 -0.587 Proportion of Variance 0.378 0.144 0.214 0.203

PAGE 106

106 Table 4-3. Information on abbreviations used in figures, disperser reco rded, and source of information for dispersal mode. Ficus species Abbreviation Disperser Source F. adenosperma ada no data F. adelpha ade no data F. congesta con bat field work F. bernaysii ber bat field work F. hispidioides his bat field work F. ternatana ter no data F. pacchyrachis pac bird field work F. morobensis mor bat field work F. arfakensis arf bat field work F. botryocarpa bot bat field work F. septica sep mixed Shanahan et al 2001 F. pungens pun mixed field work F. nodosa nod mixed Shanahan et al 2001 F. semivestita sem bat field work F. mollior moll no data F. variegata var bat field work F. benjamina ben mixed Shanahan et al 2001 F. microcarpa mic mixed Shanahan et al 2001 F. elastica ela no data F. drupacea dru no data F. hesperidiiformis hes mixed field work F. xylosycia xyl bird Shanahan et al 2001 F. conocephalifolia cof bat Shanahan et al 2001 F. phaeosyce pha mixed Shanahan et al 2001 F. copiosa cop bat field work F. wassa was mixed Shanahan et al 2001 F. subulata sub bat field work F. virgata vir bat field work F. tinctoria tin no data F. gul gul bird field work F. macrorhyncha mac bird field work F. trachypison tra bird field work F. ampelas amp no data F. melinocarpa mel no data F. odoardi odo no data F. prasinicarpa pra no data F. edelfeltii ede bat field work F. hombroniana hom bat field work F. polyantha pol no data F. subtrinervia suv no data F. itoana ito bat field work F. dammaropsis dam mixed Shanahan et al 2001

PAGE 107

107 CHAPTER 5 WITHIN SPECIES TEST OF THE DISPERSAL SYNDROME HYPOT HESIS: A STUDY OF DIOECIOUS FIGS Introduction General Background and Goals Ecological evidence s trongly suggests that seed di spersal is important to fruiting plants, to the point that some authors believe the question is no longer whether seed dispersal matters, but when and where it matters most (Howe and Miriti 2000; 2004). If dispersed seeds are indeed more likely to establish than undispersed seed s (Fragoso et al. 2003; Harms et al. 2000), then individual trees that attract more seed disper sers should leave more offspring. Hence, seed dispersers should be a strong selectiv e force on fruit characteristics. Evidence for the evolutionary role of seed di spersers on fruit traits on the other hand, is mixed. Many evolutionary ecologists have describe d patterns in co-occurring traits that they attribute to different selective pressures of s eed-dispersing vertebrates (Gautier-Hion et al. 1985; Janson 1983; Lord 2004; Lord et al. 2002; Mack 1993). In particular, they champion the existence of dispersal syndromes non-random comb inations of fruit traits that seem to match the foraging capabilities and overall ch aracteristics of seed disperser taxa that forage in disparate ways (McKey 1975; van der Pijl 1969). Other evolutionary ecologists have concluded that dispersal syndromes do not exist or that differences among seed disperser taxa cannot account for them (Fischer and Chapman 1993; Herrera 1987; 1989; 1992; 2002a; Jordano 1995). The Dispersal Syndrome hypothesis posits that seed dispersers are important for plant reproduction and that fruit traits th at attract more seed dispersers are selectively favored. Hence, one should find that dispersal syndr omes have evolved independently in unrelated taxa that share similar seed dispersers. Many pr eceding studies testing this h ypothesis have been criticized because they did not control for phylogenetic eff ects, instead considering each species as an

PAGE 108

108 independent data point (e.g., Gau tier-Hion et al. 1985; Janson 1983; Kalko et al. 1996; Korine et al. 2000). Patterns of association be tween suites of fruit traits and taxa of seed dispersers were interpreted as the result of co-evolution when, in fact, they could be more easily explained by the process of ecological fitting frugivores with preexisting preferences encounter fruits with different traits, but will feed on the fruits that are most easy to find and accessible to them (Felsenstein 1985; Harvey and Pagel 1991; He rrera 1987; 1989; Janzen 1980; 1985; Jordano 1995). The genus Ficus offers an unusual opportunity to te st the Dispersal Syndrome hypothesis in a way that tightly controls for phylogenetic effects because some species produce fundamentally different types of fru its. (The fruiting structures of Ficus are termed syconia, but for simplicity we refer to them as fruits. ) Functionally female trees produce seed figs, which contain viable seeds. These individual tr ees therefore benefit from attraction of seed dispersers. Functionally male trees produce gal l figs whose seeds are all consumed by their pollinators larvae; they therefore do not be nefit from attraction of seed dispersers. Dumont et al (2004) showed that in one dioecious species, F. pungens fruit bats preferred seed figs over gall figs in feeding trials. Seed figs were larger and had more soluble carbohydrates and fat, and less fiber than gall fi gs, characteristics that are all energetically important for bats (Dumont et al. 2004). In th is study, we expand on Dumont et als (2004) work by (1) quantifying frugivore preference for seed fi gs in natural conditions by videotaping 26 seed figs and 20 gall fig trees of seven species (2) comp aring the difference between gall and seed figs of 27 dioecious fig species with re spect to five traits that are know n to be important in attraction of seed dispersers: fruit size, co lor, odor, softness, and exposure on the tree (Gautier-Hion et al. 1985; Janson 1983; Kalko et al. 1996; Korine et al. 2000; Lomscolo et al. 2008; Lord 2004;

PAGE 109

109 Lord et al. 2002; Mack 1993; Schaefer et al 2006; van der Pijl 1969; Willson and Whelan 1990, Chapter 4, this dissertation); a nd (3) quantifying at changes in those five traits during the ripening process for gall and seed figs and for figs of monoecious species, to determine whether male trees retain characters from unripe fruits when their fruits are fully mature. We also take a closer look at UV reflectance because it has been shown to attract some seed dispersers (Altshuler 2001). Because most hymenoptera also perceive UV reflectance, and because of the tight mutualism of figs with pollinating wasps, we looked at UV reflectance at the flower, unripe, and ripe stage of seed and gall figs. If UV has a role in frugivore attraction, ripe female figs should show the highest reflectance. If, however, UV is mainly a signal to pollinators, syconia at the flower stage should show the highest reflectance. Assuming traits that attract seed dispersers are costly to the plant, we predict that only figs with viable seeds (seed figs and figs of monoecious species) should invest in nutrient-rich exocarp because only they would benefit fr om attracting seed-dispersing frugivores. Specifically, seed figs should be larger, have more pigment and volatile compounds that make them odorous, be softer (higher water to fiber ratio), and have longer peduncles to expose the fruits away from the foliage and trunk. They should also reflect more in the UV range of the spectrum. It has been hypothesized that fruit tr aits that attract frugivor es, especially color and odor, signal ripeness (Sumner and Mollon 2000b) i.e., that the appearance of fruit changes when seeds are ready to be dispersed. We pr edict that seed and monoecious figs should show large changes between the unripe and ripe stages, whereas male figs should show relatively smaller changes between unrip e and ripe stages. Natural History of Fig Pollination and Seed Dispersal Functional dioecy in figs is th e result of one of th e m ost tightly coevolved plant-pollinator systems in nature (Janzen 1979; Weiblen 2002), in which typically one species of wasp

PAGE 110

110 pollinates one species of fig. Fertilized female wasps carrying pollen enter the fig inflorescence through a small hole (ostiole), often losing thei r wings and part of their antennae (Wiebes 1979). They then pollinate the female flowers and lay an egg in each flowers ovary. The wasp larvae develop in the fig ovaries inside the seed coat and consume the endosperm (Weiblen 2002). Once the wasps have completely matured, they exit the seed coat through a hole and mate inside the fig. Females collect pollen from the now mature male flowers and exit the fig through a hole made by the male wasps, which stay and die inside the fig where they emerged. Fertilized female wasps carrying pollen then search for othe r figs at the correct stage of development to continue their life cycle. In functionally dioecious fig species female figs produce only female flowers, and they all have long styles that prevent oviposition. Hence, a female fig acts as a wasp trap because female wasps cannot lay eggs in them and cannot exit the fig to search for another. Since female figs produce viable seeds and no wasps, they are called seed figs (Dumont et al. 2004). Functionally male figs, on the other hand, produce male and female flowers, but all female flowers are short-styled and wasps lay eggs in every one of them, preventing development of any viable seeds; they are called gall figs and produce wasps that disperse their pollen. In monoecious figs, wasps oviposit in some flowers and not others, so that all syconia produce some wasps and some viable seeds. Fruits of gall and seed figs differ in many tra its, including time of de velopment, duration of receptivity of inflorescences to pollinators and duration of fruit development (Patel and McKey 1998), season of ripening (Hill 1967; Patel and McKey 1998), shape and size (Hill 1967; Lambert 1992), color (Dumont et al. 2004; Hill 1967; Lambert 1992), nutrient and fiber content (Dumont et al. 2004, Weiblen, Lomscolo, and Du mont, unpublished data), and texture (Dumont

PAGE 111

111 et al. 2004; Lambert 1992). These differences are presumably the result of optimizing the male and female functions -i.e., polli nation success in gall figs and fr ugivore attraction in seed figs (Dumont et al. 2004; Lambert 1992; Patel and McKey 1998). In this study, we focus on the optimization of the female function of seed figs and explore whether selection pressure by seed dispersers can account for differe ntiation of gall and seed figs when each is at the ripe stage. Methods Study Site and General Specifications We collected data from September 2004 until December 2005 in a lowland forest in Madang Province, Papua New Guinea, at Ohu Village (145 E, 5 S) (Fig. 4-1). To obtain data on fruit color, color contrast against the background, odor, size, softness, and peduncle length for male and female figs of each species we averag ed records from between 3 and 8 individuals per species. For each individual tree, between one and 20 fruits were measured and averaged for all traits. Quantification of Visits to Figs We recorded consum ption of female and male figs using two video cameras (Sony DCRHC40), which were set up 3-5m away from the tree and to minimize disturba nce, visited only to change tapes or batteries. Videotaping starte d at 6am and ended at 10:30am, and the again at 6:30pm until 11:00pm, to record both diurnal and nocturnal frugivores. No cturnal recording was done using an infrared light (Sony HVL-IRH2). We aimed one camera at the ground to record terrestrial frugivores and one eith er at a branch or at the tru nk, depending on where most of the ripe figs were. We recorded 26 female trees (347 hours) and 20 male trees (343 hours) of seven species (Ficus congesta F. hispidoides, F. morobensis F. pachyrrachis F. pungens, F. septica and F. virgata ).

PAGE 112

112 Fruit Traits Fruit color : W e quantified fruit color using a USB2000 portable spectrometer (Ocean Optics, Inc) and a PX-2 Pulsed Xenon light sour ce, which took reliable readings between 300 and 740nm. This spectrum incl udes the range of wavelengths visible by humans, plus the ultraviolet range. We scanned fruits using a sens or with six optical fibers illuminating the target fruit and a seventh fiber that retu rned the reflected light to the spectrometer. The scanning angle was fixed at 45 by using a black metal stand with a hole at that angle. The metal stand also blocked external light. We fa stened a non-UV filtering microscope slide to the opening of the hole where the sensor was introduced to keep the distance between the end of the sensor and the fruits constant. To obtain an average reflectance spectrum per fruit, we scanned each fruit three times in three different spots on the fruit. We calculated reflectance as the proportion of a certified white reflectance standard (Labsphere ). The three spectra for each species were averaged every 5nm. We obtained a quantitative measure of fruit color for each species by doing a Principal Components Analysis (PCA), which summarizes the information for the complete spectrum by finding linear combinations of the or iginal variables (in this case, reflectance intensity every 5 nm) (Clotfelter et al. 2007; Cuervo and Shine 2007; de Ayala et al. 2007; Endler 1990). We did two separate PCAs: one to obtain color variables for the comparison between ripe fruits of male vs. female figs, and a second one to obtain color variables for the comparison between ripe and unripe male, female, and monoecious figs. In both cases, the first principal component (PC) had high positive lo adings for all wavelengths throughout the spectrum, and hence it represents fruit brightness, the total amount of light reflected by a fruit (Fig. 5-1). Therefore, light fru its had high scores on PC 1 and dark fruits had low scores. The second PC represents reflectance in the yellow versus the UV-blue wavelengths because it had negative loadings for wavelengths from 300 to around 500nm, and evenly positive loadings from

PAGE 113

113 500 to 740nm. Therefore, yellow fruits had a high positive score on PC 2, whereas fruits reflecting in the UV or blue part of the spectrum had high negativ e scores. The third principal component represents reflectance in the green ve rsus red and UV-blue range, because it had high negative loadings of wavelengths between 420nm and 610nm. Thus, green fruits have a low score on PC 3 and UV-blue and red (and, therefore, purple) fruits have a high score. These three PCs explained 94% of the variance, and all had ei genvalues greater than one, which means that each explains more variation than any one origin al variable alone. Although PC 4 and 5 also had eigenvalues higher than one, they only explai ned 3 and 2% of the variation, respectively. Because reflectance in all parts of the spectru m of interest was summarized by PC 1 through 3, we considered that additional PCs did not contribute sufficiently to the color description of each fig species. Hence, we use only scores on PC 1 through 3 in subsequent analyses. We calculated color contrasts as the Euclidean distance between the color of a fruit and the color of the structure against whic h it is exposed to the frugivore We normalized spectra to the same brightness by dividing the re flectance at each wavelength by the total reflectance for each species. This represents contrasts due to the colo r of the fruits and not to their brightness, and are called chromatic contrast s (Schmidt et al. 2004a). 2)]^(([QbQfD Qf is the color spectrum of the fruit and Qb is the color spectrum of the background structure; is the wavelength in nm and the sum correspond s to the complete spectrum between 300-740nm. We analyzed reflectance in the UV range of the spectrum separately because we were interested in including syconia at the flower stage in the comparison. Me an reflectance between 350 and 400nm was calculated and co mpared between seed and gall figs at the flower, unripe, and ripe stage.

PAGE 114

114 Fruit odor : To collect odor, we collected fruits in the field, brought them to the lab, and placed them inside plastic bags (Reynolds oven baking bags) that had a carbon filter at one end to clean the air coming into the bag, and a filte r with Super Q (Alltech) in it to adsorb the volatiles emitted by the fruits. The filter was in turn connected to a vacuum pump (Welch, model no. 2522B-01), which sucked the air throu gh a carbon filter into the bag containing the figs, and through the Super Q filter. All odor collections were run for four continuous hours, and began no more than three hours after the fruits had been collected from the tree. To extract the fruit volatiles from the Super Q filters, we inject ed methylene chloride through the filters, which eluted the volatile compounds trapped in the Super Q. We used gas chromatography to separate volatile compounds and mass spectrometry to identif y some of them. For the purpose of this study, however, only the total amount of volatil e compounds is used as a measure of how odorous the fruits were. Because th e number of fruits we were able to collect and the size of the fruits varied among trees, we had to standardize the total amount of volatiles by dividing the amount of each volatile contained in the sample by the total surface area of the fruits used to collect odor. Fruit size, softness, and peduncle length : We measured fruit di ameter with calipers to the nearest 0.5 mm and used as a measure of ove rall fruit size. To meas ure fruit softness, an arbitrary scale between 1 (hardest, fruit surface di d not indent when pressed against fingers) to 4 (softer, fruit surface indented eas ily when pressure was applied to fruit between the fingers) was determined by the authors and measured by the same person throughout the study. Peduncle length (+ 0.5 mm) was measured between the base of the fruit and the point where the peduncle attached to the branch or trunk.

PAGE 115

115 Data Analysis Frugivore v isits to female versus male trees : To test whether female fruits were visited more often than male fruits, we compared the numb er of visited (12) and not-visited (14) female trees to the number of visited (1 ) and not-visited (19) male trees using a Chi-square test. Unless a frugivore was specifically seen not to eat any fruits the entire time it was in a tree, any frugivore that entered the tree was assumed to ha ve been attracted to the tree and its presence was recorded as a visit. A tree was recorded as visited if at least one frugivore entered the tree. Difference in fruit traits of seed and gall figs : To identify the variables that distinguish gall and seed figs, we performed a Discriminant Function analysis (DFA) on seed and gall figs of 25 species. Discriminant Function analysis finds the combination of traits that best discriminates between the groups of interest, in this case, seed and gall figs (i.e., two groups). Changes from unripe to ripe stage in monoecious and dioecious figs : To identify how traits changed from the unripe to the ripe stages, and to determine whether these changes are similar between figs with viable seeds (femal e and monoecious figs) an d those without viable seeds (male figs), we ran another DFA on the data including unripe monoecious, unripe seed, unripe gall, ripe monoecious, ripe seed, and ripe gall figs (i.e., six groups ). For both DFAs, all variables were standardized to have a mean of zero and standard deviation of one because fruit variables were measured in different scales and had differences of up to three orders of magnitude. We tested the significance of the discrimination by calcula ting Wilks Lambda, which represents the proportion of the variance not explained by th e predictor variables. This test, however, does not inform which groups diffe r from which. To analyze which groups in particular could be considered different, we performed an ANOVA on the Linear Discriminant Function (LDF) that summarized most of the va riation among the groups and performed a Tukey test. We were particularly interested in comparing: 1. male fruits (n ot consumed by frugivores)

PAGE 116

116 and female figs (consumed by frugivores); 2. female and monoecious fruits, which are both reportedly consumed by frugivores; 3. ripe male and unripe male fruits, to see whether males retain the characte ristics of unripe fruits; 4. unripe male, female and monoecious figs, to see whether the differentiation between fruits consumed by frugivor es and those not consumed by frugivores starts at ripening or whether th ese fruits are different from earlier on. Another way we tested the predictive value of the linear discrimina nt functions is the predictive classification of cases which is done using Bayes Ru le to calculate the posterior probabilities of each case of being in each group, based on each cases discriminant score. The higher the percentage of species that were well classified (e.g., known females correctly classified as a female) the higher th e predictive value of the LDF. UV reflectance : We used Mann-Whitney U tests for pair-wise comparisons of mean UV reflectance by figs of different sex and maturity stage. A sequential Bonferroni correction (Rice 1989) was used to calculate the actual cr itical p-value in multiple comparisons. Results Frugivore Visits to Fema le Versu s Male Trees Of the 26 female trees videotaped, 12 were visited by frugivores. Of the 20 male trees videotaped, one was visited by a frug ivore. Visits to female trees were significantly higher than visits to male trees ( 2 = 7.52, d.f. = 1, p = 0.006). Difference in Fruit Traits of Seed and Gall Figs One linear discriminant function (LDF) was suffici ent to separate female from male figs as two groups can only be separated in one dimension (Fig. 5-2). In order of importance, as measured by the loadings of variables on the LDF (Table 5-1), female fr uits were softer, had more volatile compounds, were larger, darker, and contrary to predictions, reflected more in the green part of the spectrum than did male figs. Female figs were also more contrasting against the

PAGE 117

117 foliage and had shorter peduncles, but the loadings of these variab les were quite low and, hence, less important in discriminating between male a nd female figs. Discrimination of female and male figs based on the fruit variables menti oned was statistically significant (Wilks = 0.339, numerator d.f. = 8, denominator d.f. = 41, p < 0.0001) When the scores of female and male trees of the same species are compared, all except one species ( F. dammaropsis ) show a higher score for females than for males (Fig. 5-3). Cr oss classification resulte d in 94% of the cases correctly classified. The difference in fruit traits for gall and seed figs is present in two different clades that seem to have evolved dioecy independently (Fig. 5-4) Changes From Unripe to Ripe Stage in Monoecious and Dioecious Figs The fruit variables included in this study can significantly discriminate between figs of different sex and ripening stage (Wilks = 0.043, numerator d.f. = 40, denominator d.f. = 473.56, p < 0.0001) (Fig. 5-5). In general, monoecious and female ripe fruits scored highest on LDF 1, followed by male ripe fruits, and then by all unripe fruits. Acco rding to the variable loadings on LDF 1 (Table 5-2), this means that monoecious and female fru its are softer, darker, and had higher reflectance in the red and UV-blue part of the spect rum than did male ripe fruits and all unripe fruits. Monoecious and female ripe fruits were also slightly more odorous and had longer peduncles, but these variables had low loadings and are therefore less important in discriminating between the groups than are softness, brightness and green reflectance. The rest of the variables had even lowe r loadings on LDF 1. A post-hoc Tukey test showed that unripe figs from different sexes were not significantly different from each other, but were significantly different from all ripe figs. Ripe gall figs differed significantly from all unripe figs and from ripe seed and monoecious figs. Monoecious and seed figs did not differ from each other but did differ from all other figs (Fig. 5-6).

PAGE 118

118 UV Reflectance The num ber of figs included in each comparison and mean reflectance for each sex and maturity stage are summarized in Table 5-3. Against our prediction unde r the hypothesis that UV signals frugivores, mean UV reflectance by ripe gall figs was signifi cantly higher than UV reflectance by ripe seed figs (W = 951, p = 0.00007). Reflectance of UV light does not seem to have an exclusive role in signaling to wasps either, as reflectance did not differ between the flower and the ripe stage for both sexes pooled together (W = 482, p = 0.87). No difference was detected between gall and seed syconia at the flow er stage, although this may be due to the small sample size. UV reflectance was higher for unripe s eed figs than for ripe seed figs (W = 1321, p = 0.003), but this difference was not pres ent between unripe gall figs (n = 26, x = 2.35) and ripe gall figs (W = 292, p = 0.22). Discussion Seed dispersers were attracted to seed figs, which showed many of th e characteristics that were predicted to attract frugivor es. As predicted, seed figs are softer than gall figs, w hich probably facilitates frugivore consumption (D umont 1999; Dumont and O'Neal 2004); more odorous, which is especially important for mamma lian frugivores (Hodgkison et al. 2007; Korine and Kalko 2005); larger; and darker. Contrary to expectation, seed figs reflected more in the green and less in the red and UV-bl ue range of the spectrum than di d gall figs. This is puzzling, at least at first, because only seed figs s howed intense red, purple, or black coloration. A potential explanation is that many red seed figs showed a dark red coloration, whereas the gall figs for the same species showed a light er red, almost pink, or an orange-red (e.g., F. itoana, F. conocephalifolia F. dammaropsis F. macrorrhyncha, F. odoardi F. phaeosyce, and F. wassa). This results in the overall reflectance in the red spectrum being higher in th e lighter gall figs than in the darker seed figs. In other words, the ma in difference between these gall and seed figs is

PAGE 119

119 not so much which part of the spectrum reflects most light (whi ch is what the PC variables represent), but instead the satu ration of the red coloration. Saturation, also called chroma, represents the purity of a color, and is meas ured by the steepness of the reflectance spectrum, in this case, in the red part of the spectrum (Endler 1990). Saturation of a pink or orange fruit is lower than that of a red fruit, but overall reflec tance in the red part of the spectrum can still be higher if the pink or orange fru it is brighter. Unfortunately, we did not quantify saturation. An alternative explanation, non-exclusive of th e previous one, is that the high positive loading of PC 3 (UV-blue-redness in Table 1) on LDF 1 is more representative of reflectance in the UV range than in the red range. When refl ectance in the UV range is investigated in more detail for all species in this study, gall figs reflect more in this part of the spectrum than female figs. Moreover, unripe seed figs reflect more in the UV than ripe seed figs. These two results in conjunction may result because some fruits have a waxy layer at the unripe stage that protects them from desiccation, which often wears off as time passes and the fruits ripen (H. M. Schaefer, pers. comm.). Because gall figs often detach fr om the tree as soon as the wasps emerge whereas seed figs stay attached until a frugivore eats them or they start rotting, the gall figs we collected may have been younger than the seed figs we collected and hence have retained more of the waxy layer than did the seed figs. UV reflectance may also be selected for at the receptive stage to attract pollinating wasps and be retained as th e fig matures. This is supported by our results showing that UV reflectance for gall and seed figs was similar at the flower stage, as both are under strong selective pressure to attract pollinating wasps. Howe ver, this would not explain the higher UV reflectance in gall figs compared to se ed figs (Dumont et al. 2004). What is most puzzling, however, is the lack of difference in UV reflectance between figs at the flower and ripe stage. The trend in UV reflectance as the fig goes from the flower, to the unripe, and to the ripe

PAGE 120

120 stage seems to differ between seed and gall figs (T able 5-3). The number of figs at the flower stage is very low compared to the number of ripe figs and therefor e the results for this comparison are best considered inconclusive. Although contrast against the background showed the predicte d trend, i.e., seed figs contrasting more against the foliage than gall figs, it did not appear as important as the previously discussed variables in discriminating between the groups. This may be because, even though some of the seed figs were red and purple (which have the highest contrast against the foliage, Schmidt et al. 2004a), more than a third of the seed figs were green when ripe and had a very low chromatic contrast against the foliage. Although peduncles varied highly among fig species (Chapter 5, this dissert ation), it was less important in differentiating between seed and gall figs. Figs with viable seeds were pr edicted to be under selective pr essure by seed dispersers and were indeed significantly different from all types of figs wit hout viable seeds (i.e., gall and unripe figs of both sexes). Figs with viable seeds were softer, da rker, and reflected more in the red and UV-blue part of the spectrum than all ot her figs, as represented from their higher scores on LDF 1. To a lesser degree, they were also more odorous and larger, and had longer peduncles than all other figs. Contrary to my predictions gall figs were significan tly discriminated from unripe figs, which means that they do not necessarily retain all the characte ristics of unripe figs. Even though they do not have viable seeds and do not need to signal to seed dispersers, their traits do change during the ripeni ng process, although to a lesser ex tend than do figs with viable seeds, as seen from their intermedia te mean score on LDF 1 (Fig. 5-5). A limitation of this study, and possibly the re ason why some of the predictions did not hold, is that we established predictions for the differences between seed and gall figs based on

PAGE 121

121 the potential selective pressure of seed disperse rs in general even though, for example, birds and mammals may exert very different, and potenti ally even conflicting selective pressures on different fig species (Chapter 4, this dissertation). It is especially revealing that the seed figs that were exclusively consumed by birds ( F. gul, F. macrorrhyncha, F. pachyrrachys, and F. xylocisia ) (Chapter 4, this dissertation) have a highe r average score on LDF 1 than the mean for all seed figs (5.21 vs. 3.61, respectively), whereas the mammal dispersed seed figs ( F. adenosperma, F. arfakensis, F. bernaysii, and F. botryocarpa, F. conocephalifolia, F. congesta, F. copiosa, and F. hispidioides, F. itoana, F. morobensis, F. variegata, and F. virgata ) (Chapter 4, this dissertation) average lower than the mean for all seed figs (3.47 vs. 3.61, respectively). This apparent separation within the group of seed figs suggests that birdand mammal-dispersed figs have different characteristics. The within-group differences are insignificant compared to the between group differences, as shown by the highl y significant results for the DFAs in this study. Comparing gall and seed figs while dis tinguishing between bird and mammal-consumed figs should be particularly enli ghtening. Birds are diurnal frugi vores and have acute color vision (Bowmaker et al. 1997; Bowmaker and Hunt 2006; Hart 2001) that they use to find fruit. On the other hand, all frugivorous mammals in Pa pua New Guinea are either nocturnal or crepuscular/nocturnal, and must th erefore rely more heavily on se nses other than vision to find fruit. Odor, for example, has been reported as important for nocturnal mammals of other regions when searching for fruit (Hodgkis on et al. 2007; Korine and Kal ko 2005; Thies et al. 1998). A particularly interesting comparison to elucidate the difference in gall and seed figs for mammaland bird-dispersed figs would be to ex amine the difference in traits between gall and seed figs for birdand mammal-dispersed figs. Trait values can be subt racted between gall and seed figs of each species, and the differences used as variables on a DFA for bird-dispersed

PAGE 122

122 versus mammal-dispersed figs. we predict that in bat-dispersed figs, the major differences between gall and seed figs will be the amount of volatile compounds produced, given that bats are mostly crepuscular/nocturnal and cannot rely so much on visual cues, and in structures such as the peduncle that expose the fruits away from the foliage a nd trunk, given that most of the smaller flying foxes common in Papua New Guinea ei ther grab the fruit in flight or land for a few seconds on the fruit itself and then fly away with the fruit in their mouth (pers. obs). In birddispersed figs, on the other hand, major differen ces in gall and seed figs should be in the production of pigments that make the fruits vi sually conspicuous by c ontrasting against the foliage, given that birds have such acute colo r vision (Bowmaker et al. 1997; Bowmaker and Hunt 2006; Hart 2001; Schaefer et al. 2006; Schaef er et al. 2007; Schmidt et al. 2004b), and in fruit softness, given that birds do not have teeth and can only use thei r bills to handle the fruits. Although we do not know the proximate mechan ism that determines the differences between seed and gall figs, our results suggest that the ultimate mechanism is selective pressure by seed dispersers. Seed figs show combinations of fruit traits that we re predicted (Chapter 4, this dissertation) and shown (this Chapter) to attract seed dispersers, while these attractive trait combinations are not present in i ndividuals that do not benefit from attracting seed dispersers. This is true for both groups of figs that have independently evolve d dioecy (Figure 5-4), rendering further support to the adaptive significa nce of the difference between seed and gall figs. Despite not distinguishing between frugi vore guilds, the differences found among gall figs and figs with viable seeds (monoecious and seed figs) support the hypothes is that, in general, frugivores are an important selective force in shaping the evolution of fruit traits in figs.

PAGE 123

123 Figure 5-1. Loadings of the first three princi pal components (PC) on fruit reflectance spectra from 300 to 750nm, measured at 5nm interval s. PC 1 represents variation in mean brightness, PC 2 in green versus red colo ration, and PC 3 in UV and blue coloration.

PAGE 124

124 Figure 5-2. Plot of Discriminant Function analysis. (A) Female figs (B) Male figs

PAGE 125

125 Figure 5-3. Scores on the linear di scriminant function (LDF 1) re sulting from the discriminant analysis of female and male figs. For each species, the paired scores on the LDF 1 for female ( ) and male ( ) individuals are plotted. Note that for all species, except one ( Ficus dammaropsis ) female trees score higher th an male trees. The horizontal lines points to a value of zero on the LDF score, and is drawn simply as a point of reference.

PAGE 126

126 Figure 5-4. Maximum Likelihood ph ylogeny of the species incl uded in the seed-gall fig comparison. From the mapping of the breedi ng system, it is apparent that dioecy evolved twice in this group of Ficus. This phylogeny was obtai ned as explained in Chapter 4, and species that were not include d in the analyses done in this chapter were trimmed off the tree.

PAGE 127

127

PAGE 128

128 Figure 5-5. Discrimination among figs of different sex and maturity stages using a Discriminant Function analysis. Symbols re present: monoecious ripe ( ), female ripe (), male ripe (), monoecious unripe ( ), female unripe ( ), and male unripe (*) figs.

PAGE 129

129 Figure 5-6. Pair-wise comparisons among fig groups of different sex and maturity stages using Tukey tests. BR: monoecious ripe, BU: monoecious unripe, FR: female ripe, FU: female unripe, MR: male ripe, MU: male unripe. The only groups whose difference is not significantly different from zero were female and monoecious ripe figs, and all pairwise comparisons of unripe figs.

PAGE 130

130 Table 5-1. Discriminant Function Analysis of female versus male figs using eight fruit variables. Variables with higher loadings (above 0.3) are marked with an asterisk Variable Loading on LD 1 softness *-1.239 diameter *-0.551 Odor *-1.113 Contrast -0.171 PC 1 (brightness) *0.763 PC 2 (yellowness) *0.338 PC 3 ("UV-blue-redness") *0.366 Peduncle length 0.182 Table 5-2. Discriminant Function Analysis of fe male, monoecious and male figs at different maturity stages using eight fruit variables. Variables with higher loadings (above 0.3) are marked with an asterisk Variable Loading on LD 1 Loading on LD 2 softness *3.026 0.102 diameter 0.199 0.083 Odor 0.202 -0.248 Contrast -0.120 *-1.017 PC 1 (brightness) *-0.321 -0.027 PC 2 (yellowness) -0.145 *-0.337 PC 3 ("UV-blue-redness") *0.360 0.056 Peduncle length 0.261 -0.044 Proportion of variation 0.92 0.05 Table 5-3. Mean UV reflectance for fig species of different sex and maturity stages Sex Maturity stage Observations Mean UV reflectance Female flower 7 1.09 Female unripe 44 1.66 Female ripe 44 1.06 Male flower 6 2.39 Male unripe 26 2.35 Male ripe 28 2.04

PAGE 131

131 CHAPTER 6 GENERAL CONCLUSIONS This study contributes to the advancem ent of the field of plantfrugivore interactions because we help resolve a long standing paradox: although animals facilitate plant reproduction through seed dispersal, they do not seem to shap e plant evolution. We show that a large portion of the variation in fruit characteristics is unexplained by phylogeny, and that the correlated evolution of seven traits point to seed dispersers as the main selective force explaining this variation. Our studys success in detecting the evolu tionary influence of frugivores on fruit traits likely results from our ability to avoid these lim itations of earlier work by: 1. focusing at the species level, where most evolutionary meani ngful variation occurs; 2. tightly controlling for phylogenetic effects with a phylogeny, the closest appr oximation to the true evolutionary history of the taxa under study, and by doing a within-species study; 3. studying groups of species on both local and geographic scales; 4. using quan titative measures of fruit color and odor, important signals to frugivores that had previously been ignored or mismeasured, and taking into account the frugivores perspectiv e on color; 5. incorporating a holistic view of Dispersal Syndromes by analyzing all frui t variables together, which we believe better reflects how dispersers search for and assess fruit; and 6. defining Dispersal Syndromes more narrowly and grouping dispersers according to their similar pot ential to select for fr uit characteristics. One method to test for natural selection in th e field is to search for correlated evolution among traits that, in combinati on, offer higher fitness to the i ndividual bearing them than any other combination. Convergence of multiple trai ts of distantly related species form strong evidence that the traits do not ha ve a neutral adaptive effect (Endl er 1986). We have shown that dispersal syndromes are suites of traits that ha ve converged in distan tly related species.

PAGE 132

132 Moreover, we show that these suites of traits can be predicted based on the behavioral and anatomical characteristics of the animals that feed on the fruits. We genera te testable predictions of the potential frugivores of different fig species that can be used to test our results. More importantly, these predictions can be generalized to other plant taxa, and to other plant-animal interactions, such as pollination, for which fr uit and floral traits can be quantified. Although arguably to a lesser extent, students of pollinator-plant interactions also ponder the question of how specialized th is interaction really is, and whether pollinators affect the evolution of flowering plants (Bluthgen et al. 2007; Johnson and Steiner 2000; Vzquez and Aizen 2004; Waser et al. 1996). Frui t and flower traits have been hypothesized to act as signals in plant-animal communication (Chittka and Raine 2006; Goff and Klee 2006; Schaefer et al. 2008; Schaefer and Schaefer 2007; Schaefer et al. 2004; Sumner and Mollon 2000b) and therefore their importance must be assessed from the perspective of the animal to which the signal is targeted. That an animal differentially perceives signals di rected to it versus directed to other animals is an important assumption of plantanimal interactions studies. Pollinators, for example, seem to respond to color cues to lo cate flowers (Bradshaw a nd Schemske 2003; Chittka and Raine 2006; Gumbert et al. 1999). In our test of this assumption in fruit to frugivore signaling, we show that the color of bird and primate-dispersed fruits can be perceived as different from the birds and prim ates perspective. This result c onfirms that fruit color can be used as a signal by plants that may differentially benefit from bird or primate consumption of their fruits. A similar assessment of other fruit traits, such as odor signals, would be needed, but might be immensely more challenging, as asse ssing perception of odor by animals would likely require many more dimensions than for color (Mamlouk and Martinetz 2004), and these dimensions are far from being completely identified (Chittka and Brockmann 2005).

PAGE 133

133 With the exception of a few thoroughly studied cases of tight coevol ution between plants and their mutualistic pollinators (Janzen 1979; Smith et al. 2008; Thompson and Pellmyr 1992) or seed dispersers (Amico and Aizen 2000; Reid 1991) mutualistic plantanimal interactions seem to be quite generalized ( but see Fens ter et al. 2004; Fuentes 1995; Herrera 1985; 1998; Johnson and Steiner 2000; Waser et al. 1996). More over, it has been argue d that the distribution in plant-animal mutalistic networks can be re asonably well explained by stochastic processes (Vzquez 2005). However, the general observa tion that pollinator-plant systems are more specialized relative to frugivore-plant systems seems to hold (Bluthgen et al. 2007). The potential for tight coevolution be tween one species of seed disper ser and one species of plant is low because, among other things, most plants be nefit from the dispersal services of many different frugivores and, any one frugivore consumes a wide range of fruits. Also, generation times and historical species tu rnover rates are very different between frugivores and plants (Herrera 1985). Moreover, populations of the same species intera ct with different species along their geographic range (Gomulkiewicz et al. 2000; Rey et al. 2006; Thompson 1994; 1999). For all these reasons, it has been argued that frugivores and plants unde rgo, at best, diffuse coevolution which occurs when groups of speci es interact with each other (Janzen 1980). We believe diffuse coevolution should not be viewed as a hopelessly messy evolutionary relationship. If one can identify groups of species that interact in a consistent and distinctive way (e.g., a guild of seed dispersers that exerts similar selective pr essure on the plan t), and if one objectively defines what traits th ese interactants are most likely to exert selection upon, then one can make specific predictions on the potential fo r coevolution. We have focused only on one side of the interaction between frugivores and pl ants -the role of s eed dispersers in the evolution of fruit traits. The othe r side -the role of plants in the evolution of frugivore anatomy

PAGE 134

134 and physiology of visual, and olfactory systems, or on its foraging behavior -remains to be tested (Dominy et al. 2003; Osorio and Vorobye v 1996; Peichl 2005; Surridge et al. 2003). It can be argued that modern day frugivores may not represent the fauna that historically interacted and exerted selection on figs. Figs have almost certainly interacted with frugivores that differ from todays frugivores, at least in body size (e.g., extinct New World megafauna, Janzen and Martin 1982). However, figs as we know th em today are likely to have interacted with species of bats and birds that are similar to those seen today. Both the origin of modern-day Ficus and Megachiroptera seem to trace back to th e same historic epoch and region. The origin and diversification of the fig family, Moraceae, is most parsimoniously assumed to be in Eurasia, from where it would have dispersed to the Southern hemisphere through land connections (Zerega et al. 2005). Modern day figs diversif ied some time during the Eocene (Zerega et al. 2005) and this is likely to have occurred in the As ian-Australasian region as this is the center of distribution for many fig sections, and where the highest fig diversity is found, worldwide (Berg 1989). Megachiroptera are also t hought to have origin ated in the South East Asian-Melanesian region (Giannini and Simmons 2003). Fully flighted bats are also known from the Eocene and, although no fossils remains of Megachiroptera have been found in the region, recent ancestor of their closely related clade, the Microchiropt era, are known from the Australia-New Guinea region (Bonaccorso 1998). Also, the oldest Megachiroptera fossil found was in Thailand (Ducrocq et al. 1993). The fig species found in New Guinea likely interacted with modern day birds since their origin in Eurasia, as birds ev olved much earlier than fi gs, at least 150 mya and were widespread. However, they likely only came into contact with the birds with which they interact today only in the Plei stocene through a land bridge th at connected the island with Australia during the last Ice Age. The island of New Guinea is comprised of a section of the

PAGE 135

135 northern part of the Australian continent to the South, and a series of islands that collided with the Australian continent in th e past 40 m.y. to the North (A udley-Charles 1991). Although most of the flora of New Guinea have closest affinities with Southeast Asia, most New Guinean birds, are of Australian origin (Beehler et al. 1986). Be cause the characteristic s on which we based the potential for selection by birds on figs were alre ady present in earlier birds, we consider it unlikely that selection by birds changed signifi cantly in the evolutionary history of figs. Returning to the first point made in the Introduction, just as it has been argued that incorporating positive interactions can drastically change core concepts in ecological theory (Bruno et al. 2003), we show that, at least one po sitive interaction that between frugivores and fruiting plants can help explai n an important evolutionary pr ocess the diversification of fleshy fruits. We argue that studies on positive interactions can help us understand processes that are important for the origin a nd maintenance of biodiversity. In our humble quest to identify the role of phylogeny versus frugivores in shaping the evolution of fruit characteristics in figs, we hope to have contributed, even if a little, to the understanding of the evolutionary relationships in the entangled bank, covered with plants of many kinds interacting with different animals dependent on each other in so complex a manner (Darwin 1859).

PAGE 136

LIST OF REFERENCES Ada ms SB, Hamel PB, Connor K, Burke B, Gardiner ES, Wise D (2007) Po tential roles of fish, birds, and water in swamp privet (Forestiera acuminata) seed dispersal. Southeastern Naturalist 6:669-682 Agosta SJ (2006) On ecological f itting, plant-insect a ssociations, herbivore host shifts, and host plant selection. Oikos 114:556-565 Alcantara JM, Rey PJ (2003) Conflicting selecti on pressures on seed size: evolutionary ecology of fruit size in a bird-dispersed tree, Olea europaea. Journal of Evolutionary Biology 16:1168-1176 Alfaro ME, Zoller S, Lutzoni F (2003) Bayes or bootstrap? A simula tion study comparing the performance of Bayesian Markov chain Monte Carlo sampling and bootstrapping in assessing phylogenetic confidence. Mo lecular Biology and Evolution 20:255-266 Altshuler DL (2001) Ultraviolet reflectance in fruits, ambient light composition and fruit removal in a tropical forest. Evol utionary Ecology Research 3:767-778 Altshuler DL (2003) Flower color, hummingbird pollination, and habitat irradiance in four neotropical forests. Biotropica 35:344-355 Amico G, Aizen MA (2000) Ecology Mistletoe seed disper sal by a marsupial. Nature 408:929930 Audley-Charles MG (1991) Tectonics of the New-Guinea Area Annual Review of Earth and Planetary Sciences 19:17-41 Beehler BM, Pratt TK, Zimmerman DA (1986) Bi rds of New Guinea. Princeton University Press, Princeton, New Jersey Bennett ATD, Cuthill IC (1994) Ultraviolet visi on in birds What is its function? Vision Research 34:1471-1478 Bennett ATD, Cuthill IC, Norris KJ (1994) Sexual selection and the mismeasure of color. American Naturalist 144:848-860 Bennett ATD, Thery M (2007) Avian color vision a nd coloration: Multidisciplinary evolutionary biology. American Naturalist 169:S1-S6 Ben-Tal Y, King RW (1997) Environmental fact ors involved in colouration of flowers of Kangaroo Paw. Scientia Horticulturae 72:35-48 Berg CC (1989) Classification and Distribution of Ficus. Experientia 45:605-611

PAGE 137

Bertness MD, Leonard GH (1997) The role of pos itive interactions in communities: Lessons from intertidal habitats. Ecology 78:1976-1989 Bertness MD, Leonard GH, Levine JM, Schmidt PR, Ingraham AO (1999) Testing the relative contribution of positive and negative interactions in rocky intertidal communities. Ecology 80:2711-2726 Bianconi GV, Mikich SB, Teixeira SD, Maia B (2007) Attraction of fr uit-eating bats with essential oils of fruits: A potential tool for forest restorat ion. Biotropica 39:136-140 Bluthgen N, Menzel F, Hovestadt T, Fiala B (2 007) Specialization, cons traints, and conflicting interests in mutualistic netw orks. Current Biology 17:341-346 Bolmgren K, Eriksson O (2005) Fleshy fruits origins, niche shifts, a nd diversification. Oikos 109:255-272 Bonaccorso FJ (1998) Bats of Papua New Guinea. Conservation International, Washington, DC Bowmaker JK, Heath LA, Wilkie SE, Hunt DM ( 1997) Visual pigments and oil droplets from six classes of photoreceptor in the retinas of birds. Vision Research 37:2183-2194 Bowmaker JK, Hunt DM (2006) Evolution of vertebrate visual pigments. Current Biology 16:R484-R489 Bradshaw HD, Schemske DW (2003) Allele subs titution at a flower colour locus produces a pollinator shift in monkeyflowers. Nature 426:176-178 Bremer B, Eriksson O (1992) Evolution of Fru it Characteristics and Dispersal Modes in the Tropical Family Rubiaceae. Biological J ournal of the Linnean Society 47:79-95 Bruno JF, Stachowicz JJ, Bertness MD (2003) Inclus ion of facilitation in to ecological theory. Trends in Ecology & Evolution 18:119-125 Burnham KP, Anderson D (2003) Model selection and multi-model inference., 2nd edn. Springer, New York, NY Burns KC, Dalen JL (2002) Foliage color contrast s and adaptive fruit colo r variation in a birddispersed plant community. Oikos 96:463-469 Cardoso da Silva JM, Tabarelli M (2000) Tree species impoverishment and the future flora of the Atlantic forest of Northe ast Brazil. Nature 404:72-74 Catoni C, Schaefer HM, Peters A (2008) Fruit for health: the effect of flavonoids on humoral immune response and food selection in a fr ugivorous bird. Functional Ecology. In press

PAGE 138

Chapman CA, Onderdonk DA (1998) Forests wit hout primates: Primate/plant codependency. American Journal of Primatology 45:127-141 Chittka L, Brockmann A (2005) Perception space The final frontier. Plos Biology 3:564-568 Chittka L, Raine NE (2006) Recognition of flowers by pollinators. Current Opinion in Plant Biology 9:428-435 Cipollini ML, Bohs LA, Mink K, Paulk E, Bhning-Gaese K (2002) Secondary metabolites of ripe fleshy fruits: ecology and phylogeny in the genus Solanum In: Levey DJ, Silva WR, Galetti M (eds) Seed dispersal and Frugi vory: Ecology, Evolution and Conservation. CABI Publishing, Wallingford, pp 111-128 Cipollini ML, Levey DJ (1991) Why some fruits are green when they are ripe carbon balance in fleshy fruits. Oecologia 88:371-377 Cipollini ML, Levey DJ (1997) Secondary metabolit es of fleshy vertebrate-dispersed fruits: Adaptive hypotheses and implications for s eed dispersal. American Naturalist 150:346372 Clark CJ, Poulsen JR, Levey DJ, Osenberg CW (2007) Are plant populations seed limited? A critique and meta-analysis of seed addition experiments. Ameri can Naturalist 170:128142 Clark CJ, Poulsen JR, Parker VT (2001) The role of arboreal seed dispersal groups on the seed rain of a lowland tropical forest. Biotropica 33:606-620 Clark JS et al. (1999) Interpreti ng recruitment limitation in forests. American Journal of Botany 86:1-16 Clark L (1991) Odor detection thresholds in tree swallows and Cedar Waxwings. Auk 108:177180 Clotfelter ED, Ardia DR, McGraw KJ (2007) Red fish, blue fish: trade-offs between pigmentation and immunity in Betta splendens. Behavioral Ecology 18:1139-1145 Cordeiro NJ, Howe HF (2001) Low recruitment of trees dispersed by animals in African forest fragments. Conservati on Biology 15:1733-1741 Cordeiro NJ, Howe HF (2003) Forest fragmentat ion severs mutualism between seed dispersers and an endemic African tree. Proceedings of the National Academy of Sciences of the United States of America 100:14052-14056 Croat TB (1978) Flora of Barro Colorado Isla nd. Standford University Press, Standford, California

PAGE 139

Cuervo JJ, Shine R (2007) Hues of a dragon' s belly: morphological correlates of ventral coloration in water dragons. Journal of Zoology 273:298-304 Darwin C (1859) On the origin of species by mean s of natural selection or the preservation of favoured races in the struggle of life, London Das D, Wilkie SE, Hunt DM, Bowmaker JK (1999) Visual pigments and oil droplets in the retina of a passerine bird, the canary Serinus canaria : microspectrophotometry and opsin sequences. Vision Research 39:2801-2815 de Ayala RM, Saino N, Moller AP, Anselmi C ( 2007) Mouth coloration of nestlings covaries with offspring quality and influences pare ntal feeding behavior Behavioral Ecology 18:526-534 Dominy NJ (2004) Fruits, fingers and fermentation: The sensory cues available to foraging primates. Integrative and Comparative Biology 44:295-303 Dominy NJ, Lucas PW, Osorio D, Yamashita N (2001) The sensory ecology of primate food perception. Evolutionary Anthropology 10:171-186 Dominy NJ, Svenning JC, Li WH (2 003) Historical contingency in the evolution of primate color vision. Journal of Hu man Evolution 44:25-45 Ducrocq S, Jaeger JJ, Sig B (1993) Un mgach iroptre dans l'Eocne suprieur de Thalande: incidence dans la discussion phylognique du groupe. N. Jahrb. Geol. Palotol. Mh. 9:561-575 Dumont ER (1999) The effect of food hardne ss on feeding behaviour in frugivorous bats (Phyllostomidae): an experimental study. Journal of Zoology 248:219-229 Dumont ER, O'Neal R (2004) Food hardness and feeding behavior in old world fruit bats (Pteropodidae). Journal of Mammalogy 85:8-14 Dumont ER, Weiblen GD, Winkelmann JR (2004) Pr eferences of fig wasps and fruit bats for figs of functionally dioecious Ficus pungens Journal of Tropical Ecology 20:233-238 Dyreson E, Maddison WP (2006) Rhetenor pa ckage for morphometrics In Mesquite 1.11 Elangovan V, Marimuthu G, Kunz TH (2001) Tem poral patterns of resource use by the shortnosed fruit bat, Cynopterus sphinx (Megachiroptera : Pteropodidae). Journal of Mammalogy 82:161-165 Endler JA (1986) Natural selection in the wild. Princeton Un iversity Press, Princeton Endler JA (1990) On the Measurement and Classi fication of Color in Studies of Animal Color Patterns. Biological Journal of the Linnean Society 41:315-352

PAGE 140

Endler JA (1992) Signals, Signal Conditions, and the Direction of Evolution. American Naturalist 139:S125-S153 Evans CS (2006) Accessory chemosignaling mechanisms in primates. American Journal of Primatology 68:525-544 Felsenstein J (1985) Phylogenies and the Compar ative Method. American Naturalist 125:1-15 Fenster CB, Armbruster WS, Wilson P, Dudash MR, Thomson JD (2004) Pollination syndromes and floral specialization. A nnual Review of Ecology Evolution and Systematics 35:375403 Fischer KE, Chapman CA (1993) Fr ugivores and fruit syndromes Di fferences in patterns at the genus and species level. Oikos 66:472-482 Fleming TH, Breitwisch R, Wh itesides GH (1987) Patterns of tropical vertebrate frugivore diversity. Annual Review of Ec ology and Systematics 18:91-109 Fragoso JMV, Silvius KM, Correa JA (2003) Longdistance seed dispersal by tapirs increases seed survival and aggregates tropical trees. Ecology 84:1998-2006 Fuentes M (1995) How specialized are fruit-bird interactions? Overlap of frugivore assemblages within and between plant species. Oikos 74:324-330 Gautier-Hion A et al. (1985) Fru it characters as a basis of fru it choice and seed dispersal in a tropical forest vertebrate community. Oecologia 65:324-337 Giannini NP, Simmons NB (2003) A phylogeny of megachiropteran bats (Mammalia : Chiroptera : Pteropodidae) based on direct optimization analysis of one nuclear and four mitochondrial genes. Cladistics-the Intern ational Journal of the Willi Hennig Society 19:496-511 Godinez-Alvarez H, Valiente-Banuet A, Rojas-Mart inez A (2002) The role of seed dispersers in the population dynamics of the column ar cactus Neobuxbaumia tetetzo. Ecology 83:2617-2629 Goff SA, Klee HJ (2006) Plant volatile compounds: Sensory cues for health and nutritional value? Science 311:815-819 Gomulkiewicz R, Thompson JN, Holt RD, Nuismer SL, Hochberg ME (2000) Hot spots, cold spots, and the geographic mosaic theory of coevolution. American Naturalist 156:156174 Gould SJ, Vrba ES (1982) Exaptation a Miss ing Term in the Science of Form. Paleobiology 8:4-15

PAGE 141

Grison-Pige L, Bessiere JM, Hossaert-McKey M (2002a) Specific attraction of fig-pollinating wasps: Role of volatile compounds released by tropical figs. Journal of Chemical Ecology 28:283-295 Grison-Pige L, Hossaert-McKey M, Greeff JM, Bessiere JM ( 2002b) Fig volatile compounds a first comparative study. Phytochemistry 61:61-71 Gumbert A, Kunze J, Chittka L (1999) Floral col our diversity in plant communities, bee colour space and a null model. Proceedings of the Royal Society of London Series B-Biological Sciences 266:1711-1716 Hall E (1997) Biological se quence alignment editor for Windows 95/98/NT. In http://www.mbio.ncsu.edu/BioEdit/bioedit.html Halpern BS, Sillim an BR, Olden JD, Bruno JP Bertness MD (2007) Incorporating positive interactions in aquatic re storation and conservation. Fr ontiers in Ecology and the Environment 5:153-160 Hampe A, Bairlein F (2000) Modified dispersal-re lated traits in disjunct populations of birddispersed Frangula alnus (Rhamnaceae): a result of its Quaternary distribution shifts? Ecography 23:603-613 Harms KE, Wright SJ, Calderon O, Hernandez A, Herre EA (2000) Pervasive density-dependent recruitment enhances seedling diversity in a tropical forest. Nature 404:493-495 Hart NS (2001) The visual ecology of avian photoreceptors. Progress in Retinal and Eye Research 20:675-703 Harvey PH, Pagel MD (1991) The comparativ e method in evolutionary biology. Oxford University Press, Oxford Herrera CM (1985) Determinants of Plant-Anim al Coevolution the Case of Mutualistic Dispersal of Seeds by Vertebrates. Oikos 44:132-141 Herrera CM (1987) Vertebrate-dispersed plants of the Iberian Peninsula A study of fruit characteristics. Ecological Monographs 57:305-331 Herrera CM (1989) Seed dispersal by animals: a role in angiosperm diversification? American Naturalist 133:309-322 Herrera CM (1992) Interspecific variation in fruit shape: a llometry, phylogeny, and adaptation to dispersal agents. Ecology 73:1832-1841 Herrera CM (1998) Long-term dynamics of Medite rranean frugivorous birds and fleshy fruits: A 12-year study. Ecological Monographs 68:511-538

PAGE 142

Herrera CM (2002a) Correlated evol ution of fruit and leaf size in bird-dispersed plants: specieslevel variance in fruit traits expl ained a bit further? Oikos 97:426-432 Herrera CM (2002b) Seed dispersal by vertebra tes. In: Herrera CM, Pellmyr O (eds) PlantAnimal Interactions. An Evolutionary Approach. Blackwell Science Ltd., Oxford, pp 185-208 Herrera CM, Jordano P, Lopezsoria L, Amat JA (1994) Recruitment of a Mast-Fruiting, BirdDispersed Tree Bridging Frugivore Activit y and Seedling Establishment. Ecological Monographs 64:315-344 Heymann EW (2006) The neglected sense-olfacti on in primate behavior, ecology, and evolution. American Journal of Primatology 68:519-524 Hill DS (1967) Figs (Ficus Spp) and Fig-Wasp s (Chalcidoidea). Journal of Natural History 1:413-& Hillis DM, Bull JJ (1993) An empirical test of bootstrapping as a method for assessing confidence in phylogenetic analys is. Systematic Biology 42:182-192 Hjerpe J, Hedenas H, Elmqvist T (2001) Tropical rain forest recovery from cyclone damage and fire in Samoa. Biotropica 33:249-259 Hodgkison R et al. (2007) Chemical ecology of fru it bat foraging behavior in relation to the fruit odors of two species of paleotropical batdispersed figs (Ficus hispida and Ficus scortechinii). Journal of Chemical Ecology 33:2097-2110 Howe HF (1986) Seed dispersal by fruit-eati ng birds and mammals. In: Murray D (ed) Seed dispersal. Academic Press, New York, pp 123-189 Howe HF, Miriti MN (2000) No question: s eed dispersal matters. Trends in Ecology & Evolution 15:434-436 Howe HF, Miriti MN (2004) When seed dispersal matters. Bioscience 54:651-660 Howe HF, Schupp EW, Westley LC (1985) Early Consequences of Seed Dispersal for a Neotropical Tree (Virola-Su rinamensis). Ecology 66:781-791 Hsiao C, Jacobs SWL, Chatterton NJ, Asay KH (1999) A molecular phylogeny of the grass family (Poaceae) based on the sequences of nuclear ribosomal DNA (ITS). Australian Systematic Botany 11:667-688 Huelsenbeck JP, Ronquist F (2001) MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17:754-755

PAGE 143

Hurtt GC, Pacala SW (1995) The Consequences of Recruitment Lim itation Reconciling Chance, History and Competitive Differences between Plants. Journal of Theoretical Biology 176:1-12 Janson CH (1983) Adaptation of fr uit morphology to dispersal agen ts in a Neotropical forest. Science 219:187-189 Janzen DH (1970) Herbivores and Number of Tree Species in Tropical Forests. American Naturalist 104:501-& Janzen DH (1979) How to be a fig. Annual Re view of Ecology and Systematics 10:13-51 Janzen DH (1980) When is it co evolution? Evolution 34:611-612 Janzen DH (1985) On ecological fitting. Oikos 45:308-310 Janzen DH, Martin PS (1982) Neotropical Anac hronisms the Fruits the Gomphotheres Ate. Science 215:19-27 Johnson SD, Steiner KE (2000) Generalization versus specialization in plant pollination systems. Trends in Ecology & Evolution 15:140-143 Jordano P (1995) Angiosperm fleshy fruits and se ed dispersers A comparative analysis of adaptation and constraints in plant-animal interactions. American Naturalist 145:163-191 Jordano P (2000) Fruits and frugivory. In: Fenner M (ed) Seeds. The Ecology of Regeneration in Plant Communities. CABI Publishing, New York, pp 125-165 Jousselin E, Rasplus JY, Kjellberg F (2003) Convergence and coevolution in a mutualism: Evidence from a molecular phylogeny of Ficus. Evolution 57:1255-1269 Jurik TW (1983) Reproductive Effort and CO2 Dynamics of Wild Strawberry Populations. Ecology 64:1329-1342 Kalko EKV, Herre EA, Handley CO (1996) Relation of fig fruit ch aracteristics to fruit-eating bats in the New and Old World tr opics. Journal of Biogeography 23:565-576 Kelber A, Vorobyev M, Osorio D (2003) Anim al colour vision behavioural tests and physiological concepts. Bi ological Reviews 78:81-118 Knight RS, Siegfried WR (1983) In ter-relationships between type, size and colour of fruits and dispersal in Southern African trees. Oecologia 56:405-412 Korine C, Kalko EKV (2005) Fruit detection and discrimination by small fruit-eating bats (Phyllostomidae): echolocation call design and olfaction. Behavioral Ecology and Sociobiology 59:12-23

PAGE 144

Korine C, Kalko EKV, Herre EA (2000) Fruit char acteristics and factors affecting fruit removal in a Panamanian community of strangler figs. Oecologia 123:560-568 Laman TG, Weiblen GD (1998) Figs of G unung Palung National Park (West Kalimantan, Indonesia). Tropical Bi odiversity 5:245-297 Lambert FR (1992) Fig Dimorphism in Bird-Dis persed Gynodioecious Ficus. Biotropica 24:214216 Lambert FR, Marshall AG (1991) Keystone Char acteristics of Bird-Dispersed Ficus in a Malaysian Lowland Rain-Forest. Journal of Ecology 79:793-809 Laska M, Freist P, Krause S (2007) Which senses play a role in nonhuman primate food selection? A comparison between squirrel monkeys and spider monkeys. American Journal of Primatology 69:282-294 Levey DJ (1987) Seed size and fruit-handling techniques of avian frugivores. American Naturalist 129:471-485 Lomscolo S, Speranza P, Kimball R (2008) Correlated evolution of fig size and color supports the Dispersal Syndromes hypothesis. Oeco logia Published online DOI 10.1007/s00442-0081023-0 Lord JM (2004) Frugivore gape size and the evol ution of fruit size a nd shape in southern hemisphere floras. Austral Ecology 29:430-436 Lord JM, Markey AS, Marshall J (2002) Have frugivor es influenced the evolution of fruit traits in New Zealand? In: Levey DJ, Silva WR, Ga letti M (eds) Seed dispersal and frugivory: Ecology, Evolution and Conservation. CABI Publishing, Wallingford, pp 55-68 Mack AL (1993) The sizes of vertebrate-dis persed fruits: a Neotropical-Paleotropical comparison. American Naturalist 142:840-856 Mack AL (2000) Did fleshy fruit pulp evolve as a defence against seed loss rather than as a dispersal mechanism? Journal of Biosciences 25:93-97 Maddison DR, Maddison WP (2000) MacClade 4: Analysis of phylogeny and character evolution In Sinauer Associates, Inc. Maddison DR, Maddison WP (2007) Mesquite: A modular system fo r evolutionary analysis In 2.01 Maddison WP (1990) A method for testing the correlated evolution of two binary characters Are gains or losses concentrated on certain branches of a phylogene tic tree? Evolution 44:539-557

PAGE 145

Mamlouk AM, Martinetz T (2004) On the dimens ions of the olfactory perception space. Neurocomputing 58-60:1019-1025 Mantel N (1967) Detection of Disease Cluste ring and a Generalized Regression Approach. Cancer Research 27:209 Mrquez AL, Real R, Vargas JM (2004) Dependence of broad-scale geographical variation in fleshy-fruited plant species richness on disp erser bird species richness. Global Ecology and Biogeography 13:295-304 McKey DS (1975) The ecology of coevolved seed dispersal systems. In: Gilbert LE, Raven PH (eds) Co-evolution of animals and plants. Un iversity of Texas Press, Austin, pp 159-191 Mikich SB, Bianconi GV, Helena B, Maia NS, Te ixeira SD (2003) Attrac tion of the fruit-eating bat Carollia perspicillata to Piper gaudichaudianum essentia l oil. Journal of Chemical Ecology 29:2379-2383 Moegenburg SM, Levey DJ (2002) Prospects for conserving biodiversity in Amazonian extractive reserves. Ecology Letters 5:320-324 Muller-Landau HC, Wright SJ, Caldern O, Hubbell SP, Foster RB (2002) Assessing recruitment limitation: concepts, methods a nd case studies from a tropical forest. In: Levey DJ, Silva WR, Galetti M (eds) Seed dispersal and frugivory: ecology, evolution and conservation. CABI Publishing, Wallingford Murray KG, Garca-C JM (2002) Contributions of seed dispersal and dem ography to recruitment limitation in a Costa Rican cloud forest. In: Levey DJ, Silva WR, Ga letti M (eds) Seed dispersal and frugivory: ecology, evoluti on and conservation. CABI Publishing, Wallingford Nathan R, Muller-Landau HC (2000) Spatial pattern s of seed dispersal, their determinants and consequences for recruitment. Trends in Ecology & Evolution 15:278-285 Osorio D, Vorobyev M (1996) Colour vision as an adaptation to frugivory in primates. Proceedings of the Royal Society of London Series B-Biological Sciences 263:593-599 Pagel M (1994) Detecting co rrelated evolution on phylogenies: a general method for the comparative analysis of disc rete characters. Proceedings of the Royal Society of London Series B-Biological Sciences 255:37-45 Pagel M (1997) Inferring evol utionary processes from phylogenies. Zoologica Scripta 26:331348 Pagel M (1999) The maximum likelihood approach to reconstructing ancestral character states of discrete characters on phylogenies. Systematic Biology 48:612-622

PAGE 146

Patel A, McKey D (1998) Sexual specialization in two tropical dioe cious figs. Oecologia 115:391-400 Peichl L (2005) Diversity of mammalian photoreceptor properties: Adaptations to habitat and lifestyle? Anatomical Record Part a-Discoveri es in Molecular Cellular and Evolutionary Biology 287A:1001-1012 Posada D, Crandall KA (1998) MODELTEST: testing the model of DNA substitution. Bioinformatics 14:817-818 Poulsen JR, Clark CJ, Connor EF, Smith TB (20 02) Differential resource use by primates and hornbills: Implications for seed dispersal. Ecology 83:228-240 R Development Core Team (2008) R: A language a nd environment for statistical computing In R Foundation for Statistical Computing Redford KH (1992) The Empty Forest. Bioscience 42:412-422 Reekie EG, Bazzaz FA (1987) Reproductive Effo rt in Plants .1. Carbon Allocation to Reproduction. American Naturalist 129:876-896 Regan BC, Julliot C, Simmen B, Vienot F, Ch arles-Dominique P, Mollon JD (2001) Fruits, foliage and the evolution of primate colour vision. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences 356:229-283 Reid N (1991) Coevolution of Mistletoes and Fr ugivorous Birds. Australian Journal of Ecology 16:457-469 Rey PJ, Alcantara JM (2000) Recruitment dynamics of a fleshy-fruited plant (Olea europaea): connecting patterns of seed dispersal to seedling estab lishment. Journal of Ecology 88:622-633 Rey PJ, Gutierrez JE, Alcantara J, Valera F (1997) Fruit size in wild ol ives: implications for avian seed dispersal. Functional Ecology 11:611-618 Rey PJ et al. (2006) The geogra phic mosaic in predispersal in teractions and selection on Helleborus foetidus (Ranunculaceae). Jour nal of Evolutionary Biology 19:21-34 Rice WR (1989) Analyzing Tables of St atistical Tests. Evolution 43:223-225 Ridley HN (1930) The dispersal of plants thr oughout the world. L. Reeve & Co., Ltd., Ashford Ronquist F, Huelsenbeck JP ( 2003) MrBayes 3: Bayesian phyl ogenetic inference under mixed models. Bioinformatics 19:1572-1574

PAGE 147

Rnsted N, Weiblen GD, Cook JM, Salamin N, Machado CA, Savolainen V (2005) 60 million years of co-divergence in the fig-wasp symb iosis. Proceedings of the Royal Society BBiological Sciences 272:2593-2599 Schaefer HM, Levey DJ, Schaefer V, Avery ML (2006) The role of chromatic and achromatic signals for fruit detection by bird s. Behavioral Ecology 17:784-789 Schaefer HM, McGraw KJ, Catoni C (2008) Birds us e fruit colour as hone st signal of dietary antioxidant rewards. Functional Ecology In press Schaefer HM, Schaefer V (2007) The evolution of vi sual fruit signals: conc epts and constraints. In: Dennis AJ, Schupp EW, Green RJ, Wescott DA (eds) Seed dispersal: Theory and its application in a changing wo rld. CABI, Wallingford, UK, pp 59-77 Schaefer HM, Schaefer V, Levey DJ (2004) How pl ant-animal interactions signal new insights in communication. Trends in Ec ology & Evolution 19:577-584 Schaefer HM, Schaefer V, Vorobyev M (2007) Are fruit colors adapted to consumer vision and birds equally efficient in dete cting colorful signals? Amer ican Naturalist 169:S159-S169 Schaefer HM, Schmidt V (2004) Detectability a nd content as opposing signal characteristics in fruits. Proceedings of the Royal Societ y of London Series B-Biological Sciences 271:S370-S373 Schlick-Steiner BC et al. (2006) A multidisciplinary approach reveals cryptic diversity in Western Palearctic Tetramorium ants (Hymenoptera : Formicidae). Molecular Phylogenetics and Evolution 40:259-273 Schmidt V, Schaefer HM, Winkler H (2004a) Conspicuousness, not colour as foraging cue in plant-animal signaling. Oikos 106:551-557 Schmidt V, Schaefer HM, Winkler H (2004b) Cons picuousness, not colour as foraging cue in plant-animal signalling. Oikos 106:551-557 Schupp EW (1988) Seed and Early Seedling Predati on in the Forest Understory and in Treefall Gaps. Oikos 51:71-78 Schupp EW (1993) Quantity, Quality and the Eff ectiveness of Seed Dispersal by Animals. Vegetatio 108:15-29 Schupp EW, Fuentes M (1995) Spatial Patterns of Seed Dispersal and the Unification of PlantPopulation Ecology. Ecoscience 2:267-275 Shanahan M (2000) Ficus seed dispersal guilds: ecol ogy, evolution and conservation implications. In. University of Leeds, UK

PAGE 148

Shanahan M, Compton SG (2001) Vertical stratif ication of figs and fig-eaters in a Bornean lowland rain forest: how is the canopy different? Plant Ecology 153:121-132 Shanahan M, Harrison RD, Yamuna R, Boen W, Thornton IWB (2001a) Colonization of an island volcano, Long Island, Papua New Guinea, and an emergent island, Motmot, in its caldera lake. V. Colonization by fi gs (Ficus spp.), their disper sers and pollinators. Journal of Biogeography 28:1365-1377 Shanahan M, So S, Compton SG, Corlett R (2001 b) Fig-eating by vertebra te frugivores: a global review. Biological Reviews 76:529-572 Smith CI, Pellmyr O, Althoff DM, Balcazar-L ara M, Leebens-Mack J, Segraves KA (2008) Pattern and timing of diversification in Yu cca (Agavaceae): speciali zed pollination does not escalate rates of diversification. Pro ceedings of the Royal Society B-Biological Sciences 275:249-258 Spiegel O, Nathan R (2007) Incorporating disper sal distance into the disperser effectiveness framework: frugivorous birds provide comple mentary dispersal to plants in a patchy environment. Ecology Letters 10:718-728 Strauss SY, Lau JA, Carroll SP (2006) Evolutionary responses of natives to introduced species: what do introductions tell us about natu ral communities? Ec ology Letters 9:354-371 Sumner P, Mollon JD (2000a) Ca tarrhine photopigments are optim ized for detecting targets against a foliage background. Journal of Experimental Biology 203:1963-1986 Sumner P, Mollon JD (2000b) Chromaticity as a si gnal of ripeness in fruits taken by primates. Journal of Experime ntal Biology 203:1987-2000 Surridge AK, Osorio D, Mundy NI (2003) Evoluti on and selection of trichromatic vision in primates. Trends in Ecology & Evolution 18:198-205 Swofford DL (2003) PAUP*: Phylogenetics an alysis using parsimony (*and other methods), version 4.0 In Sinauer Tamboia T, Cipollini ML, Levey DJ (1996) An evaluation of vertebrate seed dispersal syndromes in four species of black night shade (Solanum sect Solanum). Oecologia 107:522-532 Temeles EJ, Kress WJ (2003) Adaptation in a plant-hummingbird association. Science 300:630633 Terborgh J (1986) Keystone plant resources in the tropical forest. In: Soule ME (ed) Conservation Biology: the science of scar city and diversity. Sinauer Associates, Massachusets, pp 330-344

PAGE 149

Terborgh J, Pitman N, Silman M, Schichter H, N ez P (2002) Maintenance of tree diversity in tropical forests. In: D. J. Levey WRS, and M. Galetti (ed) Seed di spersal and frugivory: Ecology, Evolution and Conservation. CABI Publishing, Wallingford, pp 1-17 Thies W, Kalko EKV, Schnitzler HU (1998) The roles of echolocation and olfaction in two Neotropical fruit-eating bats, Carollia perspi cillata and C-castanea, feeding on Piper. Behavioral Ecology and Sociobiology 42:397-409 Thompson JD, Higgins DG, Gibson TJ (1994) CLUSTAL W: Improving the sensitivity of progressive multiple sequence alignment th rough sequence weighting, position-specific gap penalties, and weight matrix c hoice. Nucleic acids research 22:4673-4680 Thompson JN (1994) The coevolutionary proc ess. The University of Chicago Press Thompson JN (1999) Specific hypotheses on the ge ographic mosaic of coevolution. American Naturalist 153:S1-S14 Thompson JN, Pellmyr O (1992) Mutualism w ith Pollinating Seed Parasites Amid CoPollinators Constraints on Sp ecialization. Ecology 73:1780-1791 Thornton IWB, Compton SG, Wilson CN (1996) The role of animals in the colonization of the Krakatau Islands by fig trees (Ficus species). Journal of Biogeography 23:577-592 Titus JH, Holbrook NM, Putz FE (1990) Seed-G ermination and Seedling Distribution of FicusPertusa and F-Tuerckheimii Are Strangler Figs Autotoxic. Biotropica 22:425-428 Traveset A, Willson MF, Verdu M (2004) Characteri stics of fleshy fruits in southeast Alaska: phylogenetic comparison with fruits from Illinois. Ecography 27:41-48 Turnbull LA, Crawley MJ, Rees M (2000) Are plan t populations seed-limited? A review of seed sowing experiments. Oikos 88:225-238 van der Pijl L (1961) Ecological Aspects of Flower Evolution .2. Zoophilous Flower Classes. Evolution 15:44-& van der Pijl L (1969) Principles of dispersal of higher plants. Academic Press, Orlando, Fla. Vzquez DP (2005) Degree distribu tion in plant-animal mutualisti c networks: forbidden links or random interactions? Oikos 108:421-426 Vzquez DP, Aizen MA (2004) Asymmetric spec ialization: A pervasiv e feature of plantpollinator interactions. Ecology 85:1251-1257 Voigt FA, Bleher B, Fietz J, Ganzhorn JU, Schwab D, Bohning-Gaese K (2004) A comparison of morphological and chemical fruit traits betwee n two sites with different frugivore assemblages. Oecologia 141:94-104

PAGE 150

Vorobyev M, Osorio D (1998) Receptor noise as a determinant of colour thresholds. Proceedings of the Royal Society of London Seri es B-Biological Sciences 265:351-358 Waser NM, Chittka L, Price MV, Williams NM, O llerton J (1996) Generalization in pollination systems, and why it matters. Ecology 77:1043-1060 Weiblen GD (2000) Phylogenetic relationships of functionally dioecious Ficus (Moraceae) based on ribosomal DNA sequences and morphology. American Journal of Botany 87:13421357 Weiblen GD (2002) How to be a fig wasp. Annual Review of Entomology 47:299-330 Westoby M, Leishman MR, Lord JM (1995) On mi sinterpreting the 'phylogenetic correction'. Journal of Ecology 83:531-534 Wheelwright NT (1985) Fruit size, gape width, and the diets of fruit-eating birds. Ecology 66:808-818 Wheelwright NT, Janson CH (1985) Colors of fru it displays of bird-dis persed plants in two tropical forests. Ameri can Naturalist 126:777-799 White TJ, Bruns T, Lee S, Taylor J (1990) Am plification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Ge lfand D, Sminsky J, White T (eds) PCR protocols: a quide to methods and applic ations. Academic Press, San Diego, pp 315-322 Whittaker RJ, Jones SH (1994) The role of frugi vorous bats and birds in the rebuilding of a tropical forest ecosystem, Krakatau, Indonesia. Journal of Biogeography 21:245-258 Wiebes JT (1979) Co-Evolution of Figs and Thei r Insect Pollinators. Annual Review of Ecology and Systematics 10:1-12 Willson MF, Irvine AK, Walsh NG (1989) Vertebrate Dispersal Syndromes in Some Australian and New-Zealand Plant-Communities, with Geographic Comparisons. Biotropica 21:133147 Willson MF, Whelan CJ (1990) The evolution of fr uit color in fleshy-fruited plants. American Naturalist 136:790-809 Wysocki CJ, Preti G (2004) Facts, fallacies, f ears, and frustrations with human pheromones. Anatomical Record Part a-Discoveries in Molecular Cellular and Evolutionary Biology 281A:1201-1211 Zar JH (1999) Biostatistical Analysis, 4th edn. Prentice-Hall, New Jersey

PAGE 151

Zerega NJC, Clement WL, Datwyler SL, Weiblen GD (2005) Biogeography and divergence times in the mulberry family (Moraceae). Molecular Phylogenetics and Evolution 37:402-416 Zwickl DJ (2006) Genetic algorith m approaches for the phylogeneti c analysis of large biological sequence datasets under the maxi mum likelihood criterion. In: www.bio.utexas.edu/faculty/antisense/garli/Garli.html University of Texas, Austin.

PAGE 152

BIOGRAPHICAL SKETCH I com pleted my undergraduate studies at the Universidad Nacional de Tucumn (UNT), in Argentina, and obtained the title of Licenciate in Biological Scie nces, with an orientation in Zoology in 1998. I obtained various honors for graduating with the highest average in the class of 1998 for all students in the College of Natura l Sciences, namely, being the carrier of the National flag in official events three medals given by the Depart ment of Zoology, the College of Natural Sciences, and the Graduate School, a nd honorary diplomas by the Biological Sciences Graduate Association and the Argentinean Federation of University Women. I obtained a graduate fellowship from the S ecretariat of Science and Technology at the UNT to complete my graduate studies there. I started a project on the importance of birds and bats for the dispersal of a local species, Solanum riparium (Solanaceae) but, shortly after I also obtained a Fulbright Fellowship to go to graduate school in the United States. I pursued that opportunity and came to the University of Florid a, where I enrolled in the MS and then Ph.D. programs in the Department of Zoology in August of 2000. I obtained several grants totaling approxima tely $35,000 to complete my field work in Papua New Guinea. Funding was provided by NSF, Rufford Foundation, Garden Club of America, Lubee Foundation, Sigma Xi (two Grants in Aid for Research), American Women in Science, Sophie Danforth Foundation, among others I also obtained several fellowships to support myself while taking courses and writi ng my dissertation, such as the Fulbright Fellowship mentioned above, an Internati onal Women Fellowship form P.E.O., and a Dissertation Fellowship from CLAS. I was also credited for outstanding academic achievement by an International Student Award by the University of Florida.