<%BANNER%>

Relationship of Flower Thrips to Hardlock of Cotton

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 E20101208_AAAAJR INGEST_TIME 2010-12-08T20:43:56Z PACKAGE UFE0018161_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 7052 DFID F20101208_AABVAN ORIGIN DEPOSITOR PATH mailhot_d_Page_005thm.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
5ad6eb1dbe41d8554d826d7e6b3e84aa
SHA-1
b9afb3584bd2a8b1ecccdb56873e292d4ea1dac6
21707 F20101208_AABUVH mailhot_d_Page_097.pro
28781e5fe833c592c8e138d6b21d06b1
c6b03f7a40bb022df03bf2bdc0fa1f917a3c0e0c
27002 F20101208_AABUUS mailhot_d_Page_077.pro
d9811a25dc6b6c25c59569348cc48ee0
3f127de31ae60064ce2935d6939e6cfeac9e2f84
23513 F20101208_AABVBB mailhot_d_Page_009.QC.jpg
75c26212109443636d2aa373f633e3ff
0527ec957f4d36eec13b31b2425dea8b4c6aebb1
37131 F20101208_AABVAO mailhot_d_Page_023.QC.jpg
1d4af1947825f0f32d2cbb91a7d9aaa0
1a55e967a2eb603a42148eb4d5b8b8ef8fc33b25
27774 F20101208_AABUVI mailhot_d_Page_098.pro
ed0665e9bb02dda883d090bf2e649e04
4d798a51c13bbcd19791067d3534915a24ab53cf
51037 F20101208_AABUUT mailhot_d_Page_078.pro
95e1c0b33de011e400adc22d2ed18bb2
5802ab40a1106e5ba27b4a476dc1d3c078727ebf
9590 F20101208_AABVBC mailhot_d_Page_108thm.jpg
20c1023840c67266186ab3cf2e30ee0f
0378ce2bec533447c987b90a57d91e87f9ea536d
4634 F20101208_AABVAP mailhot_d_Page_105.QC.jpg
0bf209903d44e13786b27369a2627214
816fac45f719c00cdbd6461a0bbbc449c66a0a0d
22639 F20101208_AABUVJ mailhot_d_Page_099.pro
6d22ebe6287c074c684360b28522dc69
5c0a8599d82cf3e616181bc1731feb85657cfa55
51382 F20101208_AABUUU mailhot_d_Page_079.pro
484c02d6bd4755098bb020817dac39f0
86641a693366f45ac46c52a616546a35142e71e2
8970 F20101208_AABVBD mailhot_d_Page_032thm.jpg
74d6979c049ce7962368fc173f8fd55e
7f505197dbac9efd441da09ff3c35466fb7f1fb6
8777 F20101208_AABVAQ mailhot_d_Page_012thm.jpg
941abf7104c4cde36abc511157523f77
726bf699a81105de77a23dd6bc65f90be98a91af
70102 F20101208_AABUVK mailhot_d_Page_100.pro
a1d4f8b977d32e079e745577a7206a89
2b564799e9fbeb119eb04a60993cdf1b9ba45045
55001 F20101208_AABUUV mailhot_d_Page_080.pro
8a2387677388ee320caae7b243163169
e02fe0db5af7fafa96a1ead18b1a6a7c0185dc9b
36040 F20101208_AABVBE mailhot_d_Page_041.QC.jpg
bcf8d3ae04baf6392dd17e7dd3516f0d
3eaf3de64e42ec17c0ac01ea18f156bcbea488cd
8886 F20101208_AABVAR mailhot_d_Page_034thm.jpg
69f471838bfe900565da6b0c87176f90
48a354433e0681a7a86f360059de482201b06008
54163 F20101208_AABUVL mailhot_d_Page_101.pro
42a7f5a0672a91207cc11db86338e9aa
7cc2d3559f4e3281af65e61f1a531c97aa637653
56079 F20101208_AABUUW mailhot_d_Page_082.pro
b460a4c99e5d293b2c225afcc131e1e4
3ceaf568360bbfa1a2605caaebe6af8476b98af5
9187 F20101208_AABVBF mailhot_d_Page_016thm.jpg
48d79ad182405c7a6866f66bb2f5a239
e082fdea3add63ce89fff25f33e433ddacc3f2e9
9457 F20101208_AABVAS mailhot_d_Page_083thm.jpg
ac006e7ed047c38870413d4d99af1713
3714c690c3f05a2d1b0d856c4f4ad1a1f637bfa0
51206 F20101208_AABUVM mailhot_d_Page_102.pro
77536f9476a61ef3fb1d5f74ec8f660c
435575a5897097280370a2512d121341bf9ecc6b
54728 F20101208_AABUUX mailhot_d_Page_084.pro
8e2774e7433b6e8bab8fe070cd8620bf
ae77a728c7f5274205f6d92865110427f4a4cde4
18464 F20101208_AABVBG mailhot_d_Page_097.QC.jpg
c11b8442c558c74fe33e41999116125e
eb9a5346a8f68b066c28f14ea5ec315da39c5546
123 F20101208_AABUWA mailhot_d_Page_003.txt
45eebaed9c3a93a2d3f99ae6a5aefffd
fc4f4f19159892101846a6d059afec820b6b94ab
8803 F20101208_AABVAT mailhot_d_Page_102thm.jpg
c19014ce2322614bc4a39f1c17743b51
a51fc36cd0fe306222ab0cb0da7d1b66fc4e4688
52562 F20101208_AABUVN mailhot_d_Page_103.pro
547507afcfbd5acd11ff70c85a17334b
9937560fa9bf0c855f5487e4757f6bcf05f11dce
55490 F20101208_AABUUY mailhot_d_Page_085.pro
a904c7b041e241ac26b7491545c08561
abcd729b6ee54a9f4a6397cb2dd9d3e0cda30370
37201 F20101208_AABVBH mailhot_d_Page_060.QC.jpg
35373f8fc89d3124e9342eb64cce0d68
5d306e3bf64e3d02037849bbe83960e5d5c298f6
1356 F20101208_AABUWB mailhot_d_Page_004.txt
d3cfa3cdc160165eda0d37382c5fc7a0
8017e4c423a6cc5237476c8b1077c7a8252981c2
37042 F20101208_AABVAU mailhot_d_Page_028.QC.jpg
ffe2a21a0b260a3700dd044960267de5
83d2ef1327bf589e5da3219c6a5d5a17603e2b01
54437 F20101208_AABUVO mailhot_d_Page_104.pro
67eee09dbb0c598cbac186dea46c704a
3e0d27cafb93b12587bab2603fe21d912394733c
55043 F20101208_AABUUZ mailhot_d_Page_086.pro
a289f7ddf8c9285d82e1a5b1439c05fb
4902470675692ce601d34f55a82ee94c8b332477
9234 F20101208_AABVBI mailhot_d_Page_031thm.jpg
2620d67681fdaca2ebff11431d17adf1
811891a453c2df8d38456ea68bcec8ba8bb68c27
4029 F20101208_AABUWC mailhot_d_Page_005.txt
9e0d2d0f7578846b8b4e48faa8b98336
f325a4ef8fcb86cd20684ea89599ef0d4cdfc98f
26321 F20101208_AABVAV mailhot_d_Page_008.QC.jpg
b71f023053ac2e36b01392d649c89a40
b4a102e0cbf733566fcef33ef1112dae034312d2
8381 F20101208_AABVBJ mailhot_d_Page_015thm.jpg
f12c1b67832c9cee7dcf77e969b59359
fa85aa45fe1b89aa43a4a5b3b2cf0beb26ce0fa8
1499 F20101208_AABUWD mailhot_d_Page_006.txt
3d11e082dba82785fcd60786bc27e44d
8ba01704a30e786e896598a51f2392c844d08334
37092 F20101208_AABVAW mailhot_d_Page_036.QC.jpg
843c99392475f29934fcebb09019860a
53e4dc37af80be6091cadcd4b8055beaa560bc34
4567 F20101208_AABUVP mailhot_d_Page_105.pro
c9be388158134e1c303c063cc1242ecd
5e5ff9f06c4b4ffc92573f98ae14565feb0d9a7b
39487 F20101208_AABVBK mailhot_d_Page_063.QC.jpg
4309ae5801e71234851e03e809f88e31
d12db2e44f1cdbfd9966925065c3c40f0eef51e5
2725 F20101208_AABUWE mailhot_d_Page_007.txt
c560a834aeb7ef129fe813418781208c
566f8bcc63ca71c7739bf6df95bda220defab819
35858 F20101208_AABVAX mailhot_d_Page_104.QC.jpg
61c46d730f2c02bbf044f44309eb0cef
337061703f566071f0e62f69ae1efa6f8cdf3077
63908 F20101208_AABUVQ mailhot_d_Page_106.pro
909ffe6f9bb7271fbc6540bc575a8599
e5ac1982c5807a21f8dbd158df3401c930a8133f
11754 F20101208_AABVBL mailhot_d_Page_113.QC.jpg
73c8e119f236c84da1de997d2d7fdbdb
1d75d7cfc5b4fd7837840148d9477374f134e061
1911 F20101208_AABUWF mailhot_d_Page_008.txt
e52ad4ba40b1b7a6d166dbf972486e0a
ce209671e9edd189c5573d24f60222b2e75f3511
9384 F20101208_AABVAY mailhot_d_Page_112.QC.jpg
6c54dab5b92b3b79cc8ef94a26c34e91
8b23fab75a9944fec53830c1acdbf6767462dc06
70833 F20101208_AABUVR mailhot_d_Page_107.pro
24807967e8fe6d93e4b6ca0590a8a208
9287f8242c3b893a869b69b6a86862c0d22f5c96
36613 F20101208_AABVCA mailhot_d_Page_069.QC.jpg
e84f70a1d6d01818248051eec6d43917
4e34902117da6fc3812b2037cef9b6a47208ba33
8888 F20101208_AABVBM mailhot_d_Page_081thm.jpg
15848cb0ef8dc7f7b7005d3775bbbd14
aa67147f4d2f326a3a44bf017dbb569bbd2c7198
1667 F20101208_AABUWG mailhot_d_Page_009.txt
4ee68a21ff44c3c4b109e14d599bb4b7
e7eb442c2035d7f34ebfcd92d7606adddb242d14
68199 F20101208_AABUVS mailhot_d_Page_108.pro
fcf00d0e93640a18dd6a4f61fbd02f40
8177f42103f8604d70d2b69397acb2088a214484
35162 F20101208_AABVCB mailhot_d_Page_029.QC.jpg
bc0eaba12ca14aca2f183c181980e4ab
220b612d88e4287dcfb770b7aa397c124155ed13
21908 F20101208_AABVBN mailhot_d_Page_049.QC.jpg
5103bc0aed573bdc07248a3063c2b164
18ffebbea773a95fd589a3212fee99a09a5b2cc8
2029 F20101208_AABUWH mailhot_d_Page_010.txt
620c3dcf8095c8c18e78c3fb07c5aeec
a8b1722feff03c364c98467302cf998fe44addb5
64469 F20101208_AABUVT mailhot_d_Page_109.pro
8cbcb72bf7ae93ce691ae8cea24995b6
ba5ab39b4dbe1aeeafa440034c9bd6759ed72a68
10972 F20101208_AABVCC mailhot_d_Page_099.QC.jpg
7ef65e0a083c14e236de328ac77ff1af
65c9d5e2d6992efbeed6ee23cf6322197cf9f5b0
39399 F20101208_AABVBO mailhot_d_Page_108.QC.jpg
85bc1aeeea237dae0f4e0acf60b3f34d
d9b46f527dae8c08fd069fd3248150dd73b2c149
716 F20101208_AABUWI mailhot_d_Page_011.txt
92b47750e9f20e1396534de0c817e087
93bffc00065532ba536b414974eed6ed101040c0
8919 F20101208_AABVAZ mailhot_d_Page_069thm.jpg
49243861b2db27883c8f953d920e1f12
663a81624aa2f9359f557493204c2196d575e746
65776 F20101208_AABUVU mailhot_d_Page_110.pro
366ab77f5943e14b753931439df29150
bce03745fc9b0f172ad2d1660f7fccace4e8b22c
36205 F20101208_AABVCD mailhot_d_Page_084.QC.jpg
cea789df9905e93634d9d39f2dd061cb
76691ac2edde0e56dda7555c34a9cd98a420c7aa
9269 F20101208_AABVBP mailhot_d_Page_082thm.jpg
ec426d55601d9d6eb34c1be774d0608b
c8b5541374d4ed17e7987ee845727d5d3c3c9777
2210 F20101208_AABUWJ mailhot_d_Page_012.txt
9287df6ecfed2e168dc26b7e84fdc7c9
39dec62388b5c691ffd644e1b648c6850d22b2a6
65323 F20101208_AABUVV mailhot_d_Page_111.pro
5e40efb52576030f896fb87c616b74ae
ca8a2cd09380601c327c6c602e9265174fd12498
8565 F20101208_AABVCE mailhot_d_Page_030thm.jpg
09f1a1946fd7b1254300f0c50bf7ea66
ce829dacf9bcf04c55d964ab82a55e96707ab7cc
8568 F20101208_AABVBQ mailhot_d_Page_103thm.jpg
7ecab74e47646bd68e59ef5f955d0616
c06d31261bed6b32597b001bd611781c37b39ff6
2141 F20101208_AABUWK mailhot_d_Page_014.txt
021b66ca3800c146d631f614c97ce656
d8931aa95ff92a7e571c8e3ce5e5e180bc564a67
13118 F20101208_AABUVW mailhot_d_Page_112.pro
f8fbc4c45def4a24b398cabb062d4a80
8ab7d07eb87a5057828ada0050d9543a4ada21a1
8782 F20101208_AABVCF mailhot_d_Page_029thm.jpg
0e4b8716876949f412a071250ba2362e
3512d99199657d6e4b410cb21ac9a85c4995615a
37387 F20101208_AABVBR mailhot_d_Page_085.QC.jpg
af4207b65698541079ebc8a19b3f2b32
64f939d498a9856ab845a10b1e297e3ca6d63e00
2133 F20101208_AABUWL mailhot_d_Page_015.txt
73be586810c697bef0d6d58fb18bfe38
ee3bf2bb5b4d9edb4c91ef2ebb2b8e2d882e73e3
15484 F20101208_AABUVX mailhot_d_Page_113.pro
add6a55a13f50634875a5c3d2610fbd8
e09038e3c7073047baa7acc4c5769a3b98647bfd
35705 F20101208_AABVCG mailhot_d_Page_061.QC.jpg
f56451e19c3521f699e28fba254aba80
a738a51f7abd7da384443d079dce48e19bbbf6c4
2142 F20101208_AABUXA mailhot_d_Page_032.txt
594ec9a412a419c76032936140982fd9
96a032d375ee4f659c8adc928731f8b04b93b386
9177 F20101208_AABVBS mailhot_d_Page_106thm.jpg
9bec045b03ae365312cfc882c7c6b009
948085531f92a637d0de0cd2f789d986a8e545e5
2056 F20101208_AABUWM mailhot_d_Page_017.txt
04dff2ee259c794495c2e5f7173e637e
1166edfb577529baeaa269528a7e449d34aac44f
412 F20101208_AABUVY mailhot_d_Page_001.txt
cf65603dc2c496e41fb71903940cd121
d80c72d9a9f1aa10a31357030c7d0f108e65fd25
37162 F20101208_AABVCH mailhot_d_Page_082.QC.jpg
6641152e16a06830934879815718ef17
3110abfb5e20932bef8679c6e8a221666a18145b
2184 F20101208_AABUXB mailhot_d_Page_034.txt
d112590cb38c5757cde22a6340029b09
406e8755bc0fe16d9f412436000af5d4552b2ec8
7972 F20101208_AABVBT mailhot_d_Page_013thm.jpg
0693444c8d98fe40ddad327d6507a443
57b3dccd65ec2f615daae40de9178957d12e61d3
2070 F20101208_AABUWN mailhot_d_Page_018.txt
142bcd52c3185db37dcc4ee9bc2728e7
f08e45dc5ad9961b8afaf0c532323d69beeebf4c
121 F20101208_AABUVZ mailhot_d_Page_002.txt
1296dd4ceaca93f81f5651be87cd1a33
3dc2eb8489555642eb3878017743f7121fbd980f
34326 F20101208_AABVCI mailhot_d_Page_047.QC.jpg
891852f1fd70847f98d75d6ed0eda3c2
59136f32af0d22169884a7d314ece8f4001b7399
2156 F20101208_AABUXC mailhot_d_Page_035.txt
86c4371a3f87917509b87f4a20815fb3
1380e64d9bc281592ee22a63e48d8a6f0117e0cf
36936 F20101208_AABVBU mailhot_d_Page_111.QC.jpg
a9a93de9f8b7b2685abd385d7a6e53b5
7027a0cbef474fbde47a6700a79cab508edfa6c3
2182 F20101208_AABUWO mailhot_d_Page_019.txt
a6244820c0767e0a5201238090f8c859
075e7ce96ac7360a4631fd4d3e09ef0f11eece5e
36748 F20101208_AABVCJ mailhot_d_Page_071.QC.jpg
c6537351cf0739bf9dfcbdf505460ad6
7a43cfcc25aff49c6651f157fbf14f8e6fd68fc4
2127 F20101208_AABUXD mailhot_d_Page_036.txt
5ce9fc6c00d7c349c0e97a21d6ab02aa
717177b30d9823bdccca9b16c3bb844ff776042e
8230 F20101208_AABVBV mailhot_d_Page_033thm.jpg
599e1790361ae49564ef39681e9278de
ba81a73f68ee0906477eebc1fc36f355e904c012
2085 F20101208_AABUWP mailhot_d_Page_020.txt
01197ed340a9c7589a983caeff77348a
c3bef44687cafbdd92bdb976fd7ad52372938ddd
21289 F20101208_AABVCK mailhot_d_Page_088.QC.jpg
3c42493a5cb41df7a3ffc780102de412
d0703722e9fb64811d9ead278fcc392af76c059f
2246 F20101208_AABUXE mailhot_d_Page_037.txt
2971e45e6bc2341f84472b9ea6b62427
f9dcb00186d02b9869ed6b08049179919c3cc2fa
9097 F20101208_AABVBW mailhot_d_Page_036thm.jpg
94eb43a720cbb1bb9f84ff213381f6a7
5d52bd7d80ea7d40127b4c38aa635b59ad3e9d48
6455 F20101208_AABVCL mailhot_d_Page_008thm.jpg
1354ef813f20630f93cd54eb0c9fc044
7d4a89d5bb49c3d92c42d6754bff5a9ae54de062
2300 F20101208_AABUXF mailhot_d_Page_038.txt
178781256e5b0f932d1fc5bd8b13b0ae
f6ba6999067ab19ee23b07f1877c926b99865bf3
8739 F20101208_AABVBX mailhot_d_Page_101thm.jpg
72a75d20c7d37c300180b23e489b6319
3c2d4b5980e75f33c1478ab9f9be05beb83c7308
2150 F20101208_AABUWQ mailhot_d_Page_021.txt
dc02c912c3d82906a40813b1f7534d03
eef2aa6906a9a409b8f2b18a2cad8a3c61346a27
8462 F20101208_AABVCM mailhot_d_Page_007thm.jpg
b914362c477182bd6dee3fd88e966cd8
003ff807084e0b156672e52bdb257f3949e9dbf5
2191 F20101208_AABUXG mailhot_d_Page_039.txt
5771a275763e144b41219942deb9e053
57b266cf39fb5088d8a9b63d6ddb5cbef81ad19d
8478 F20101208_AABVBY mailhot_d_Page_092thm.jpg
7d797c404ed56272fe33b2ff0a62039f
f36c726f36e676f19e92d1ab6ebdc167bc138954
2155 F20101208_AABUWR mailhot_d_Page_022.txt
995c54234a780e7d8a74900f309bcdd7
a445ab2f85fcbd9a89b5ce00813aa901489c0b13
8552 F20101208_AABVDA mailhot_d_Page_062thm.jpg
10bf0605007cb045769af6d978a28d0e
3dc5b87124e2670242be8ab3a48d2ce3415a7f2a
8237 F20101208_AABVCN mailhot_d_Page_047thm.jpg
56cc9fb13257533b4e58270a5c939a53
f99930b90c35e3a3e08e350e496d55c91ae2ecd4
2157 F20101208_AABUXH mailhot_d_Page_040.txt
d1df30dd9e7b845e5b3304efe483613c
5a5f0de1123712a3bb1a5c20b4f320435f563a42
11671 F20101208_AABVBZ mailhot_d_Page_048.QC.jpg
095492a65fc9cabdf322d6e51e5238cd
94a41c17cf742816c2cfdfcdb7c1ae67c98a29e2
2222 F20101208_AABUWS mailhot_d_Page_023.txt
af40a7cba5874c2da789488a62aedc34
54cd6e076f5ab2c4278f646c6930243e3ab2aa33
35886 F20101208_AABVDB mailhot_d_Page_021.QC.jpg
a2f96e929c36789c2cdde21e32b07450
80bee610f75ff5709b445e0b2d251e14c8a9c649
9062 F20101208_AABVCO mailhot_d_Page_017thm.jpg
b3af3ee4a3e4f097e088f4a378d4cf26
9fe6e2df836819bb82d275cd6e635e994558f75a
2159 F20101208_AABUXI mailhot_d_Page_041.txt
6f69440334403161fddc9619caa75a99
aba552ba49088af6a2d95360c6556e715bcdabdc
1985 F20101208_AABUWT mailhot_d_Page_025.txt
ff7c349ca1391ee67fe273152d9b770d
74f044d2b224d251992cf0587b3f0171e4a08e7f
37086 F20101208_AABVDC mailhot_d_Page_016.QC.jpg
00b5bd45e7d1119d28fe924ded840765
9348a6a70c221fd9fe655ea18e4af083c0099ed7
38292 F20101208_AABVCP mailhot_d_Page_031.QC.jpg
0b017e083937492d7bf92d8dc343e2dc
f29e3ac70ef0690c6add44e7f0174dc157a9bcfd
1714 F20101208_AABUXJ mailhot_d_Page_043.txt
bde47de602160f19806d1078768457d1
5ad9a5553b1cf9c9addca9662b7fee83971e6bba
2223 F20101208_AABUWU mailhot_d_Page_026.txt
7186ee82acf9644e82d4e48daecb08a1
bd41d7cc4d9592edb4b0a6a415e73a1cea0ab37b
34704 F20101208_AABVDD mailhot_d_Page_079.QC.jpg
ffa4205188fe2d08c9ce4d9172366000
61e7f532bde21dff37c645caab29726c4c5659bd
4402 F20101208_AABVCQ mailhot_d_Page_046thm.jpg
1900a712f10bfa05dae0c9ffe1fb7b6b
28f928f8fe1f0928b3be042763c74c24f9253e23
2760 F20101208_AABUXK mailhot_d_Page_045.txt
265da321a3b8cec2248dc7b98b9f5c5a
4792db736cecfea3006c2a8412ec759caadd420b
F20101208_AABUWV mailhot_d_Page_027.txt
512f46bf5854bed026c167084160461c
96e1f8a95ab67c36bd8d86cd27ed15563e4e5d13
1787 F20101208_AABVDE mailhot_d_Page_002.QC.jpg
f584ec09c52436f71ce5592a3a4d7bd6
3af673ad3ef7284ff9ca7e8f810aca2d9891af18
9043 F20101208_AABVCR mailhot_d_Page_111thm.jpg
b99580a122bfe38d45af6cd239f22978
1fb4722498c068fce8c9f78e7a7bb8ef8ae2bc2b
2791 F20101208_AABUXL mailhot_d_Page_046.txt
b52482f239ec10468de48cd89e5ac06c
ee16b73b3d1edf2f7a777cd2397432e44fe0a5b3
F20101208_AABUWW mailhot_d_Page_028.txt
871203e64aad0183197e7722268bd247
ed6f11b6814f4982f4c5173876c27606d14c6315
34530 F20101208_AABVDF mailhot_d_Page_017.QC.jpg
ec945704552be39b42e5bbfac90fb731
f352b25b0dd735fc9c4e8ec04b79d7c1e00b1325
1980 F20101208_AABUYA mailhot_d_Page_062.txt
3fb8b61c2af57f40bc57cf9ebb417459
2b4130a970c9923a95befe86ce818a074ec989e4
3046 F20101208_AABVCS mailhot_d_Page_048thm.jpg
8e5b1e0ea34426e7ffbcc4102279c7df
c375161e0110809a72bf044cd52f00e0a089294f
1042 F20101208_AABUXM mailhot_d_Page_048.txt
171cb6025ff2db8ee3f656b016f964bf
8bc5579652ead5d1e9c693613c3b5fb652c7172e
2093 F20101208_AABUWX mailhot_d_Page_029.txt
dfb5f5d077e5b239ae37c36e2ae1daf8
75858ae0f566bc6df6272de838d52747de91d6a9
34624 F20101208_AABVDG mailhot_d_Page_078.QC.jpg
f7d1e49889684d753e41344d3a396d55
d836ee55f9a8fd20848fbdc25630166306488882
2319 F20101208_AABUYB mailhot_d_Page_063.txt
659163e3c285d8236f8be22605c35837
ee7d56c37ab3171f387e78a1fdc22295e00d707c
2476 F20101208_AABVCT mailhot_d_Page_112thm.jpg
9cffa910d7539d93276e80abf57a5710
194849464d8887bbc4f4ec8b5772cb5e12d7893b
1448 F20101208_AABUXN mailhot_d_Page_049.txt
cf58ab2e2d8d5a7eddebc1a55be23fe2
8014657a5c67133ab5e8867db1e50d4de9c587ff
2094 F20101208_AABUWY mailhot_d_Page_030.txt
96f19fc6490e5cc7dc6b7a8be53dfc09
ea76ff191a1fd29d9d9bd65c8017a8debbbf2fb8
34887 F20101208_AABVDH mailhot_d_Page_018.QC.jpg
17da85df1eb821a5047ddc44a7051251
1e55329c88e0e284958123de047f64fdac9a662a
2198 F20101208_AABUYC mailhot_d_Page_064.txt
eab770e842ec63d1e76aa5372815d65a
6bceddceedf37dc90a5fee757c05128bc2b3c9a0
22260 F20101208_AABVCU mailhot_d_Page_095.QC.jpg
9d3598bccd945d61c4cab222d60b8228
4f41f8758bcc2fd49feaa4d5501be299b990a713
984 F20101208_AABUXO mailhot_d_Page_050.txt
c5382ff67d80ceaf43a5802a1a7d86f0
a29e1712ccc228a2cf041ee011d8b0134fb1cd6e
2177 F20101208_AABUWZ mailhot_d_Page_031.txt
1bd4f85de5708f946413e1e1622b94f9
ce1ab75550c4b5b6dba939c999fe6d6794c133c0
8960 F20101208_AABVDI mailhot_d_Page_087thm.jpg
cd9788886188717112589b9b4f8b7387
6fcf91cc498546633894d79e5d8e284f1f05963d
2227 F20101208_AABUYD mailhot_d_Page_065.txt
ce2b8cd555cde672c483c917cad63016
2a4a12b4694799b8b8e2553d5ef8140488b7696a
8505 F20101208_AABVCV mailhot_d_Page_079thm.jpg
fca61c70d9aa0c1e0b9d41f667a539a6
4312bd3dc53ad39e47f7dd3120a850c90845b7d1
823 F20101208_AABUXP mailhot_d_Page_051.txt
e15ff7e7cff3ca41807c5ad840fd41cf
66e41512d8958cceab0d4f738f2cb8df54da6e9a
37641 F20101208_AABVDJ mailhot_d_Page_037.QC.jpg
dc2cd15da8ec2dc31f42c7d3bc39266f
025329fe9a3abee50cc1962d9dedf307f7a254cd
2078 F20101208_AABUYE mailhot_d_Page_066.txt
a14db39b2f27b8ee4dcf14713d6bfb91
fb2a6df24606c05d6a5591f6030c7f3564cb1ddf
5944 F20101208_AABVCW mailhot_d_Page_050thm.jpg
9521852d7d0b2527088f45adecae3e3d
e7323572dfc847aeb94f5e020a1b016272d65420
118 F20101208_AABUXQ mailhot_d_Page_052.txt
158d5a7334a82e9d6194241410e9e637
52b7c2f941fdc82ec739975f20eb1e8a3a2761cc
35555 F20101208_AABVDK mailhot_d_Page_059.QC.jpg
7d3b15fb64d74e1f5b914928862ef31f
cb4bfb0f9b0327cb73ba394e3456676254e6bdd7
2083 F20101208_AABUYF mailhot_d_Page_067.txt
93ff2ac7937eb5f12c44f2e99b70fb27
b75d743ea10e3595021cef3e7cc1f58bb39b490c
7079 F20101208_AABVCX mailhot_d_Page_043thm.jpg
1f832e89fbc3f2047a36e942281ea218
ea71213bedcc1cbdc554e11d676d0467de54a47c
8848 F20101208_AABVDL mailhot_d_Page_041thm.jpg
f9bd13c688fdf0fceb8dc802d1f44f09
fa220e24b5899f67ff3df1b7448d201081637738
2129 F20101208_AABUYG mailhot_d_Page_068.txt
9f270e542aad20cfe228d0ba6b283b8e
10d2b051c5a158966ab1a45e3c4ae6a82bb8973e
3476 F20101208_AABVCY mailhot_d_Page_006thm.jpg
8aad2dbf4e664b822379e6cd3bb720d8
97d3019a107013daf01c7bedfbd108f1bb4344b7
616 F20101208_AABUXR mailhot_d_Page_053.txt
3a3ac9cc12e7b5cede76391368eec7e6
e764550604d893647df3c38e11f0f10c150b5be8
9265 F20101208_AABVEA mailhot_d_Page_085thm.jpg
49ef684d7f63eed1a0a65ead5faa2539
b61d2d3f4699f64f3dc706e74e78e1cc38a23132
37986 F20101208_AABVDM mailhot_d_Page_083.QC.jpg
ac450ad0935484fe46eacda619b49e2a
9020315a66130267f92cdefc6f71489ab77b65af
2185 F20101208_AABUYH mailhot_d_Page_069.txt
cda042eb06160975f6056761adc6e519
cbdf086ed8683a125e2a8d6ea67cc2027b636a99
8549 F20101208_AABVCZ mailhot_d_Page_091thm.jpg
dc796e91dbf53a7c4b792210ba8c5367
80413d3e342bebf690341b2da62ab1dadd19e945
456 F20101208_AABUXS mailhot_d_Page_054.txt
bf690037031bcb31d97f56fd41f9ccfe
6629e9b62000d26f1ae0f113ac4299eafcec7e39
9006 F20101208_AABVEB mailhot_d_Page_080thm.jpg
f45c56cfe30ed955a2d2cd97e062de39
e1294d8f38ad10d3d5d764fa38fe3c49149bc570
3962 F20101208_AABVDN mailhot_d_Page_100thm.jpg
24cf79b1d3befe7bfd1f938e0830159c
5abf1318a4846080b5a8aa855cd4f299736f6e6d
2225 F20101208_AABUYI mailhot_d_Page_071.txt
a5e1369a50a5f0c0f2d75b5027a03042
b4c892ca4c5f422bc8ac907107baff48d1c5b0ef
1822 F20101208_AABUXT mailhot_d_Page_055.txt
8fe372c4bc910313d4aa1567905ae396
232786af014b005198cf7c9e0b83f9e0bc35b55b
13691 F20101208_AABVEC mailhot_d_Page_011.QC.jpg
8d5a20f18ff4d5d899658d447363a5f8
d46b35d41b26a8763b12b03280b9a383c81e58bf
9130 F20101208_AABVDO mailhot_d_Page_064thm.jpg
38f4e6da11f5b77b1e1b9f4c07d3ae20
bb26271e98e94dfca5a4af557204e8ff4d35a3a4
3750 F20101208_AABUYJ mailhot_d_Page_073.txt
173adbb604c3f2c24fdb9eb7dbe46272
1d91a5c1dc8a30774df3c76e973dcbee50fc576e
791 F20101208_AABUXU mailhot_d_Page_056.txt
27c8f6c237012e5a6987c7c8f8e66376
cae48ce203c4fe415e5fc8726bdba4db13133583
170566 F20101208_AABVED UFE0018161_00001.xml FULL
f5d3a6db7a52a28c6ad3d63a9b0ddf09
2cd80d5d2066c8b0d6e4ce67543d569e2968df2b
8779 F20101208_AABVDP mailhot_d_Page_040thm.jpg
d79c79aef253e0676c3075f97693cd9b
4812ecc6ed5e684502d0bfc90628d52656e06766
1210 F20101208_AABUYK mailhot_d_Page_075.txt
e22117444f86efa211f9fca2edd64af4
efd8162816d91314993baea46958dcde2e8c6b6f
1112 F20101208_AABUXV mailhot_d_Page_057.txt
7f818ffc4f6ad471a2eb115fbd188506
53969b30ca1a22783740cadd3742a27d485a7543
7578 F20101208_AABVEE mailhot_d_Page_001.QC.jpg
93f28b6221c90789c27fe199044bc80c
feef804de433359c69ae0ad9fe0736b052b25d2b
37337 F20101208_AABVDQ mailhot_d_Page_080.QC.jpg
649b61646586bcbf4351175ad3109eb0
623a1b5725d8d4d2113f079d7f2fb8ff7603220a
423 F20101208_AABUYL mailhot_d_Page_076.txt
51ffee0a3e609368b0907b09af58a53b
5802cefce862e570f164e61a24cc4eb49ab4051c
1837 F20101208_AABUXW mailhot_d_Page_058.txt
f2f525d0a7f73fc8c1551a80907e437e
71a86915e8e6b2c2bc14833f693e6eab1f78866f
657 F20101208_AABVEF mailhot_d_Page_002thm.jpg
326200a794d5dc8fdb533e2af9fbf6a9
c2bc839b3b4702de85be5d8d0ecb0c26c3f3e90d
6088 F20101208_AABVDR mailhot_d_Page_049thm.jpg
55130df2814ad50a7aa81aa06b438b0c
ca5f6d6904edc0930a85de44c3e5b1f04804e093
1416 F20101208_AABUYM mailhot_d_Page_077.txt
4497bb5d9429e469afdc606162c8f2d6
8b2419ea405020e78b6480ca9fcfb2a700bc2fd5
2134 F20101208_AABUXX mailhot_d_Page_059.txt
43c36dce72f9e2a0e463e734040ff825
ca7fff27fc8a510334a52ac9e757cdc8a826c7ec
5847 F20101208_AABVEG mailhot_d_Page_004thm.jpg
ed4050eb40992769c22d34d1b295bc04
8a7f91d9fec90c6fd803ee9292e764a6489f3931
4213 F20101208_AABUZA mailhot_d_Page_091.txt
3a3d29d24e87846be49c99cfff7bca0c
5a1c922511494714c37f6ae7a6f50e676ab5f8e4
8944 F20101208_AABVDS mailhot_d_Page_104thm.jpg
e355d2cd1063ff90871478ff083071d1
44a597d9b4d2ce5399fd4636ebc2f9ca79f4d9e0
2108 F20101208_AABUYN mailhot_d_Page_078.txt
3f391b2e99c9c9383d5d143eb7e10fbe
e5bdb9a62f16b0a958ebd2f136bdbce9748f8391
2193 F20101208_AABUXY mailhot_d_Page_060.txt
214bcdad45bbb0743bcbe2b51ee3ccb5
521f954296a8fcfde7ee65137435610f9441dbd5
23947 F20101208_AABVEH mailhot_d_Page_004.QC.jpg
213ac16511d9c1cb4fb99f5b34103653
2f71433af11bbf3bbbf22b1901a790b70deabdf5
2335 F20101208_AABUZB mailhot_d_Page_093.txt
40af093f095bbe067e2b44bee7ab555b
793d3a42bb321d6a6040a8daf420472a0546e639
29994 F20101208_AABVDT mailhot_d_Page_010.QC.jpg
9caedf82951a3aae8cd83e3f366c8ee9
539eb3287d7e269cf5c3d35877571a8cb7089156
2022 F20101208_AABUYO mailhot_d_Page_079.txt
bd9b27085a3ea06ee30639d35c125a14
eb7b4cae72a59dba8cc09f9a2d601cb6b3e9d220
2153 F20101208_AABUXZ mailhot_d_Page_061.txt
18b80bdf17c0ad6888a68e1059adbc93
23e39482c86c7f7f7e223720a09d5349a1b5f01b
27627 F20101208_AABVEI mailhot_d_Page_005.QC.jpg
81bd525a4b0e15cdb90ffec1d9563141
54f4e66fb0845111bad93ebfc6d3d3a7a63e0300
2068 F20101208_AABUZC mailhot_d_Page_094.txt
6a29c5a505e3254cb603ded59c9d4654
59b95966ef5b2264c6c4b188799fb0d7b202a612
37134 F20101208_AABVDU mailhot_d_Page_090.QC.jpg
b79e1f326431ab28257bac5bffed7d31
c7343eca173440ff9e451b9c3d272fc81ca0cde0
2208 F20101208_AABUYP mailhot_d_Page_080.txt
13d37fd5fc0a953cde3ee1ce83094e09
4d99262f97ef00429f7ce52a68e466618cf7c0a7
12901 F20101208_AABVEJ mailhot_d_Page_006.QC.jpg
f95d8e58b3fbf20e5e844f3a9489475a
93bee470706139d05cf4da187ae2ad7e666768cb
1286 F20101208_AABUZD mailhot_d_Page_095.txt
8681d6ed47a5898dde77edf0db5e894f
8ea6d986729cfd6fca9f060bb28e863d99c9866e
22007 F20101208_AABVDV mailhot_d_Page_074.QC.jpg
2a5fc6e19c1f96150a44f2af8760f9e6
c60d47db821881342a3c7dfa935976ed72020743
2151 F20101208_AABUYQ mailhot_d_Page_081.txt
3b9e972391160861a33e355d310cd29d
c3acca16a4e0d12011ea577a82a81517a10311d6
35435 F20101208_AABVEK mailhot_d_Page_007.QC.jpg
998291500d83eef61a0c845ec0013163
acf6f6fe2f2d274d0cac89e95bdd1e423ab58ee9
892 F20101208_AABUZE mailhot_d_Page_096.txt
c7f57cb2fdcb4194cabb4093f13e375d
3186a5a06b1d2991c3cbe8b4634cf7dbf7d8350d
17533 F20101208_AABVDW mailhot_d_Page_075.QC.jpg
9e1405f2d36abf64974ef0fc07c057a8
03c5664e1480195b6bdf50fd94b91998d180e613
F20101208_AABUYR mailhot_d_Page_082.txt
c87dba064dc8cefc2e65f142f055d61b
61afcc2f9cdcfdd78e9359468251231f995cd033
7729 F20101208_AABVEL mailhot_d_Page_010thm.jpg
89841ca0d7ca1f6b12a77837f24704ed
3a3bee16ae3c7d6e49e045c834e3a318e55b16f5
1019 F20101208_AABUZF mailhot_d_Page_097.txt
4f12bb5a65217f1b58aa3cc3d048a49e
e419b66746d09e576e0365c26bad861082e96d34
17412 F20101208_AABVDX mailhot_d_Page_051.QC.jpg
f01acdb9635810139981aa3dbdc538ff
c1a15a50b704f047730dc942285686f46331de69
37582 F20101208_AABVFA mailhot_d_Page_039.QC.jpg
68d9fdf3491e86d4f176ac827b2b9bd6
88b2887809f8dd5b2f89baa626dfc2bd863cd31d
3343 F20101208_AABVEM mailhot_d_Page_011thm.jpg
ac118404619f1e169231b2fd0dc9dc28
5ccbff350aa113c8b6277a39519e6fc2e0c2bbc9
F20101208_AABUZG mailhot_d_Page_098.txt
438fec65ab01b9b1b8c6e53a079591ff
800201ce6344b811bb90f57db38622b26de96b17
14876 F20101208_AABVDY mailhot_d_Page_045.QC.jpg
cbbc44e2f9563aa1b136cf2033f3b6bf
9a58d265c115d7477eb60f74a1ada2d05635379e
2230 F20101208_AABUYS mailhot_d_Page_083.txt
84b7ece15d78493e583f620148dc2516
34c58c78f6e50d0994a21baa3fb568443efd17fe
35586 F20101208_AABVFB mailhot_d_Page_040.QC.jpg
5027fb6f69b68525a340719c96cfa775
268561812a024a47be890ba837380fcbcbbe2649
35839 F20101208_AABVEN mailhot_d_Page_015.QC.jpg
4ddde9d08928e99074d23769f922f0fc
f35d3076d9d40beca64cc87607d9e6eec652c31c
917 F20101208_AABUZH mailhot_d_Page_099.txt
d7f3ffe512fef92df22a5488ad0145f2
49d2e931c44d926b2b7af3b656f3ed88aaad005a
33775 F20101208_AABVDZ mailhot_d_Page_033.QC.jpg
081008f75bc738309a20081ca6ba0d49
edc658d3bd40305a994c45abf8eb3f7e24e2c9dd
2147 F20101208_AABUYT mailhot_d_Page_084.txt
cf73229e8713c714acf6054edd9987df
73d7f2172901c3626ea4dba5cef0270b493e907c
36934 F20101208_AABVFC mailhot_d_Page_042.QC.jpg
5a3918ba4df0a772984c49f3c44f656a
138aff00ec5fbf4197ae9c0f856439fcae5f27e4
36506 F20101208_AABVEO mailhot_d_Page_019.QC.jpg
82d0ad77e0ca5d672fe77b6f11d612c4
2a126874824ca3442e17420c532e5c98e9eb6e18
3144 F20101208_AABUZI mailhot_d_Page_100.txt
9304c46c8433ebe360e564843eea5169
9254f0e9937989e49a6d6642998c5ed469758c43
2172 F20101208_AABUYU mailhot_d_Page_085.txt
15c43c66170de1b172d1f95fb92ce434
c995a2aaef44474e7995bb770056e68c0deb40c5
29253 F20101208_AABVFD mailhot_d_Page_043.QC.jpg
429a381da1628fce16383a65afb11cfc
2182a3036d116f3d947c40808f61f07f2eb9ecfa
9209 F20101208_AABVEP mailhot_d_Page_023thm.jpg
0fc6accf907d0ec15d04b9141a1da8fc
8190885a7a9747a9771c7151572be35e320341f1
F20101208_AABUZJ mailhot_d_Page_101.txt
18b3b12cfc1c52329bafe4cd039dabe9
616c1e325c6573f10b50bed36ad770d613512841
2197 F20101208_AABUYV mailhot_d_Page_086.txt
8853ddfba552e216eb459c9f285e2e80
cd2b88524b1662f53f8e84585eb6f6e57a07cbd9
2994 F20101208_AABVFE mailhot_d_Page_044thm.jpg
ba882b9890f079d7c5c9718ff81b3d18
227393f92a07adc128f8740a8b190f5955766f1a
34844 F20101208_AABVEQ mailhot_d_Page_024.QC.jpg
d0f1856f5cbf25883ad7f0049d408e5c
91ebc1acf9caed967630357951965f38b049a150
2019 F20101208_AABUZK mailhot_d_Page_102.txt
c4bf0a8499a451eb426a5f46a647f599
7363f267f37ab0b7e47bf02dd3c2d4400e3958ec
F20101208_AABUYW mailhot_d_Page_087.txt
466016c78355ade5826fcded9e713f3b
2f1525a070416433c60982154acbcff066b0efb6
10589 F20101208_AABVFF mailhot_d_Page_044.QC.jpg
09f470c369556bcf23a099ebc9734a57
0b435df3723474486a7c2d0109875112537744da
8162 F20101208_AABVER mailhot_d_Page_025thm.jpg
877d88e8027b4fe8b313000c32fddd73
7b9d134ba0797723a6c9b27daa99e8ff89869312
2071 F20101208_AABUZL mailhot_d_Page_103.txt
e1b91bae796cdc2483b4bd2208e9e9d9
7b6db62078b60abf9419a8c1581b9192a1e4c24b
1194 F20101208_AABUYX mailhot_d_Page_088.txt
1124ee27c5c23463386d77d53e31e9f9
77184c78f0a0b222fa0d50aad65a76cfe0c92138
14085 F20101208_AABVFG mailhot_d_Page_046.QC.jpg
4c411637c58382f3f1fe203f6c10b35a
e87a782f82ce0dd02c894b9e77a72f6101a0be05
34087 F20101208_AABVES mailhot_d_Page_025.QC.jpg
0862f5b8f066e9eea2abd9a8ca7da085
ada6567b2f4d008a29b038a21e42e230218bb55c
2140 F20101208_AABUZM mailhot_d_Page_104.txt
83dcdba67b565d342a8008abad949cf6
1b1771a018176953e58842e38b0baf8a5d4b35f0
19356 F20101208_AABVFH mailhot_d_Page_050.QC.jpg
49f6785477198796f5b8e7c5112a5de3
2297ba5214e69ec6d71399e85a08b13be9dfca6e
35799 F20101208_AABVET mailhot_d_Page_026.QC.jpg
9a1bdf098f7e86547ab414c5ad44e30d
c8b803987e426d12f4d820235dbaead722fbbe3f
184 F20101208_AABUZN mailhot_d_Page_105.txt
3cb386f7acad54391cc4932705325173
944cabc5e389b87a613bed092942d8a12c652551
99 F20101208_AABUYY mailhot_d_Page_089.txt
7e2ac43a0df9695e0e2109ff077e9151
0631d753d049557864cdb808faa07b384245900f
5138 F20101208_AABVFI mailhot_d_Page_051thm.jpg
d104dea8b34984671dbf552ac20bc344
7867c0c89a5e4e16ae46e22fce286c3eeb3140ae
9067 F20101208_AABVEU mailhot_d_Page_027thm.jpg
37ad17033b8d074c25be1272eab52dab
ecce2abb0cfa0d4c8da565c9e1f71ccab4139197
2591 F20101208_AABUZO mailhot_d_Page_106.txt
4fe14ad317a6bded20a7e7b4be845820
041338eef58444acc810afb727efb0a0be665814
4112 F20101208_AABUYZ mailhot_d_Page_090.txt
87a0156e6cc855922d254b30e9dc10f6
b123f0835ad2ea170f28485003e82c84bdf83653
15940 F20101208_AABVFJ mailhot_d_Page_052.QC.jpg
904234d308876c361077efc21199f56d
3ad5544b9cf0ac59768ee8d4216496e809712b03
37083 F20101208_AABVEV mailhot_d_Page_032.QC.jpg
b03dcced2a0971694b06551ef437de5a
c9e97194bfc5b4d655310e0dc692b9f7e6f49cfd
2738 F20101208_AABUZP mailhot_d_Page_108.txt
54994c4e56282cc3cd603eff68213346
ef612140a39ad0bf437efcd094c33718336deaf7
4550 F20101208_AABVFK mailhot_d_Page_053thm.jpg
0bbc3666abb461268ff9f52c9ef742b6
bb105a74e6e55bc9d68f3da71c0fbbb28a1b781c
37138 F20101208_AABVEW mailhot_d_Page_035.QC.jpg
ea381ad76e3d4aa2c9706b4418129bc6
01f4b7a8248e5b4ffcfe46418eb1bd6c3fc0b15e
2600 F20101208_AABUZQ mailhot_d_Page_109.txt
accb889d694f615411961cb19ca69717
cef619402e761cd4975e3a7fed973e12e10a6e64
15712 F20101208_AABVFL mailhot_d_Page_053.QC.jpg
15d6b478bda01c490f6009036824f614
bb9839b6befdfb3d54d83bc5de8b189445937972
9219 F20101208_AABVEX mailhot_d_Page_038thm.jpg
6f42c0bfc748e2cfc239f6b9e679033d
8ea57ebbe042120f69193406c8ea8210f6f86d14
2650 F20101208_AABUZR mailhot_d_Page_110.txt
638a4f8194432894e169778914173e5e
791fe312d4b718bd5c1b350e3911a1afcbad106a
4024 F20101208_AABVFM mailhot_d_Page_054thm.jpg
d18b560145f7c9c08ce57c61b3da73ac
f85d76a9041ad4ce948271a9ac47383f84a9164d
38628 F20101208_AABVEY mailhot_d_Page_038.QC.jpg
a83e7e18a7d4134814e86da7293d8e39
70faf768510b35559c88e82bef875d170068d127
2654 F20101208_AABUZS mailhot_d_Page_111.txt
85b80f9622f48c02d1d26af58e0e0243
acfaaaa0f3cd4f5dc1ae4ed7e25c7d3dd8d93ed9
35451 F20101208_AABVGA mailhot_d_Page_068.QC.jpg
5ae1df1070f6ef4514fb3568072e6fe8
32bbcf6f7c0694a00300ea2b608a72b5ef37771e
12099 F20101208_AABVFN mailhot_d_Page_054.QC.jpg
c7f6b198d5448b1e7c0cb4554b59a70e
53773ffd7b1b8987d8e1bb913b5c0f60c5279aed
8872 F20101208_AABVEZ mailhot_d_Page_039thm.jpg
3f78363d8c85f98a92c77149427ad1a4
9b18e95b49b1118e4196ab10c3fff84b76a9ad04
36206 F20101208_AABVGB mailhot_d_Page_070.QC.jpg
18055d18bde054f1ef584cbdcf4015ab
538b6d09120affe689f0ff61b8c419e69d39dbe3
28556 F20101208_AABVFO mailhot_d_Page_055.QC.jpg
7367d1371f3b4142c41acf5791171a40
462fe3ee50240f6d27e06dbb63730ccf382afb95
537 F20101208_AABUZT mailhot_d_Page_112.txt
fee390c5c9f1b0355a6257d2ff8fcbb7
d242c71f6c815565de8d0b69a7dfb171a9d714e5
1053954 F20101208_AABUDA mailhot_d_Page_107.tif
721105e6293f236b4b24bab6bf36bb4f
a46caecf2ca4034542e4545ca9c19373f85c4a22
7018 F20101208_AABVGC mailhot_d_Page_072thm.jpg
cddc073cc000c1d02d4a464bc542ed2e
2d18d36c3ae01002fa22fb485db94ef23f5d9986
16483 F20101208_AABVFP mailhot_d_Page_057.QC.jpg
b6f6ce8c31f7ce9a6c339836d2f9ba7f
a641c8f7e40ecc0b24292a731c38072469fb1ff6
652 F20101208_AABUZU mailhot_d_Page_113.txt
b34f019f5849415810194aaafd68e3c2
5780eacfb9c6ff9f9d59f5a880a95b080c8e14f3
918754 F20101208_AABUDB mailhot_d_Page_006.jp2
ccd8728471348151ff5b0cb40d8959f4
e49666e0a52ae61f885ce7dace1d8cdbcf1ade47
27989 F20101208_AABVGD mailhot_d_Page_072.QC.jpg
47d923f78260678e015b242887dd2113
9510bbccbcab80875ce40fc4613955fc16ffd303
F20101208_AABVFQ mailhot_d_Page_059thm.jpg
f83203b09ed76c5c26ff9f5e2e8cb85c
a7eea528c7f7211fa230126a0f350f07474f37f7
2044 F20101208_AABUZV mailhot_d_Page_001thm.jpg
1397b230853103e650c4b0b6fc4ded18
2bbacd8855a34e61d0e25200821c7824a93430fd
114084 F20101208_AABUDC mailhot_d_Page_060.jpg
e72b744f4e6bf8b3564e4227edf1ab08
8f29c74000ee4d6897b3d142a21e8fa5fa7cc424
8119 F20101208_AABVGE mailhot_d_Page_073thm.jpg
0518c45b116b45b34aba4896e9941447
ee4887e8caf90871f7857dc591feb0ad1c95e921
9070 F20101208_AABVFR mailhot_d_Page_060thm.jpg
ddd0fc134c0c025e1ed6556f97afdd5e
9dec780ed27da75224fd505cc06739678738c9fe
438696 F20101208_AABUZW mailhot_d.pdf
9742b8d76d5ae22de1a79da7b36b1229
eb323c59a973aeb55870be50d59248fd52789e8c
9285 F20101208_AABUDD mailhot_d_Page_110thm.jpg
828ff17090f3d2b67bb38f66f9ec0c57
b09b87ada10f97bc421f3e165f53defc366b425d
6546 F20101208_AABVGF mailhot_d_Page_074thm.jpg
a9baf77296ab8b5a8ce493e4ce65b201
16ef327ef1b4f441db1e11eadbb70affbadc628f
8883 F20101208_AABVFS mailhot_d_Page_061thm.jpg
7e5eb35ff0da6a133e1ebdd6ac8c279e
4e47b970b46278a8901c9d78bdaf36e1faf913aa
8851 F20101208_AABUZX mailhot_d_Page_021thm.jpg
8ce520c75f33faf8c435b4ec495ad29f
7ba946aa1bfa08a1ad3422f351977768c6ccddad
F20101208_AABUDE mailhot_d_Page_101.tif
883acfacf1346c7e70fdd03a4fad3133
9c8b883e38a42ca073fdb331ee5bdf58eaffd2c2
5413 F20101208_AABVGG mailhot_d_Page_075thm.jpg
bc721011eb20b33b24cb0f7d19ab6543
8911e5fb0176eb53d254f1f4bae4b542d11d5412
9326 F20101208_AABVFT mailhot_d_Page_063thm.jpg
691789a6b106cb53ada576762ab75c03
b4b4c7d12179baf0021309c241128e7626c355fd
19604 F20101208_AABUZY mailhot_d_Page_058.QC.jpg
93c3d36df87848312be4db101b8381de
1e85b04092564ba3eefa6ed0e2b3802aa0a0153c
95628 F20101208_AABUDF mailhot_d_Page_043.jp2
ac1fe58fd6ca9d22bf95644080547779
4570f4902006d22e8ed861cafc6e90f1f4944fec
2966 F20101208_AABVGH mailhot_d_Page_076thm.jpg
3d4bbbd013449912c48a3a658767e45f
48f465c1fcc73d36c7b67ff61bdb320fcfe2200f
9141 F20101208_AABVFU mailhot_d_Page_065thm.jpg
6ed55456a8e62c9534d58a00a1642ae3
dc6b562ded94a6178f7aea660d030efc5450f42f
4796 F20101208_AABUZZ mailhot_d_Page_058thm.jpg
d00d8744055451a50712d2b22d684f44
f755548c9999cc204f59a2f72846c765dc5fb64a
67850 F20101208_AABUDG mailhot_d_Page_088.jp2
0386d47fe00f9f6c7ca7d4427824dcdb
1b64f7deddd5841deb601922673bc956f9feefe4
9152 F20101208_AABVGI mailhot_d_Page_076.QC.jpg
084422fdf799b2887cfd69bede7cd0aa
7e346c1d230e2d4cd24b45a41a186df40b14d8cf
56097 F20101208_AABUCS mailhot_d_Page_064.pro
19a7a309875ab594569d91390804ff2b
2e92eed859e34c4f8a6d630327ac7e04232fa991
38605 F20101208_AABVFV mailhot_d_Page_065.QC.jpg
3c63f192c5ad7d8b8a9bd54d33ee3589
9e339fc070ff107d66276915639e401f516931b4
35747 F20101208_AABUDH mailhot_d_Page_091.QC.jpg
0b2d6c69245c59b8dab83809037ad576
b630abce7765c9b919ea9caf17371b869eb27a47
22735 F20101208_AABVGJ mailhot_d_Page_077.QC.jpg
b40bbcbb459e57d9344e1bee31d698bf
1c8df290a038f95d6f53e4cc89385122a303fa4f
8450 F20101208_AABUCT mailhot_d_Page_090thm.jpg
e98ca2f565e5cb9c766a43fab01a611c
71003d098dd3ae1c4d1dc1ca82a6be890a496a86
14792 F20101208_AABUDI mailhot_d_Page_095.pro
fd0622a8ad9460f0e80a5c1c1efe1e13
9c675faa1faa3d2a17d7cd800da68d8f8f325fe0
8570 F20101208_AABVGK mailhot_d_Page_078thm.jpg
db3ae58a10fddb48361c496f44fde1b4
4ae0dcb3f69e3039f6c86383e8deaa4a2236ca4a
88065 F20101208_AABUCU mailhot_d_Page_093.jpg
8c06ed3a487f8b80f0cf897773fdf4a6
b98c41fbbe5830b8be3cbaa81a538e3525e9325f
8635 F20101208_AABVFW mailhot_d_Page_066thm.jpg
e5d3d86672cd906ac9c77ae8108db247
3d2320e2d1ec0e44504d8c54f9517e40f490c108
F20101208_AABUDJ mailhot_d_Page_004.tif
f5db059b40aa765eced3fb5d98fc32a4
950053db9da1befaa8551c8c9ececdf032f438b2
37228 F20101208_AABVGL mailhot_d_Page_081.QC.jpg
4ca6bfea0020477c744e52336ee1bf62
74745147874199873b560d8649bb0875f5d6c685
2162 F20101208_AABUCV mailhot_d_Page_070.txt
33c22c1499d733b47c196a7d47c4f738
c8f843b9421e7553eeef7d489fe6be12c237bba9
34721 F20101208_AABVFX mailhot_d_Page_066.QC.jpg
32c6dc4d57cdae2114da61f509aa2dbe
69eedc727fb4ced333db951e0399a61da6aca53f
33321 F20101208_AABVHA mailhot_d_Page_102.QC.jpg
e1fcbe979dea2cfffc0ae142694d8d6d
44995ed245b62248c6d471f188c6412bcf5d28c8
51887 F20101208_AABUDK mailhot_d_Page_020.pro
4d7c260b09c5086ae8354f16f58f91ab
c0d5d0038c3df04601bee74dd46a7abb5bcfd516
8832 F20101208_AABVGM mailhot_d_Page_086thm.jpg
61da2828e842adbe482c74f685d10321
fca589a6ee3c19f7efe02eadfd19f3fdb10ce296
27787 F20101208_AABUCW mailhot_d_Page_076.jpg
2e8ccec10c9383acf03af9cae152c8ab
01908d9fe437bb62d0b7a8d42fab347ffa130cb4
F20101208_AABVFY mailhot_d_Page_067thm.jpg
ebead2475594999c52967b6dc4772a2d
ab3cafd6105c79c4271ba1d0f4811a9da2353842
35102 F20101208_AABVHB mailhot_d_Page_103.QC.jpg
916653dfcbd9677880ca0e1a206c75b8
5e484d7cb8087c933889991368fefd137c54d6c4
54621 F20101208_AABUDL mailhot_d_Page_061.pro
e1cc2cb61bcbc98c531e78c209eae175
b8d20e0487f6aedb1400e1ea5c69be51f8e409bf
37328 F20101208_AABVGN mailhot_d_Page_086.QC.jpg
7457f060e48b7949a1e6ec4697c605ef
e5788cb2e4ea8b655dc45256bc4c902aac56e2cc
18702 F20101208_AABUCX mailhot_d_Page_056.QC.jpg
e70d38455302e5fb0dce21305a4717e1
b07bc3605802ba6fe263af2bb5325ddcd623ce58
8761 F20101208_AABVFZ mailhot_d_Page_068thm.jpg
01a8ffd87302787f3b040f8436ba7d38
d993e9040fb5f3c1f3c9a1dd5467fc6394c5ba84
1238 F20101208_AABVHC mailhot_d_Page_105thm.jpg
d2d4a23202803ad8f45fe6ca674ac759
905bd4728583f0f148d7792b5936462f6b463076
109868 F20101208_AABUEA mailhot_d_Page_040.jpg
a9d677243eef3dffa600d372228ebdf4
bd88c629caa279bb533456f2b47d08ede138a576
119154 F20101208_AABUDM mailhot_d_Page_042.jp2
a51d92ab7c65b03ff56d9611b0e43493
b1fd0781a8a06245ee8e844237b737f4d3123111
4962 F20101208_AABVGO mailhot_d_Page_088thm.jpg
096821d0a264a330787e6b61a9958b50
32127b6f8ff0010ac12e401ce3a4926067282454
56616 F20101208_AABUCY mailhot_d_Page_023.pro
8289961ca794dec5772f83d145fbfda5
00b9d861edcb608602eb19a4cfc3c2eeae240ce6
37423 F20101208_AABVHD mailhot_d_Page_106.QC.jpg
ea2a914f57eb86f2821da831e3521b29
2969ca92ed622131e44fff575880d00746600783
2220 F20101208_AABUEB mailhot_d_Page_042.txt
e4e805627ebb06f3baf58f34c8a776d6
f5bf454dc584ca6af0bbe60fad28afea86f0aed1
110586 F20101208_AABUDN mailhot_d_Page_068.jpg
134949badb9be4460ac20d92f229bcc7
90af3b2ad089587ff64915022d2c58d516b29d41
2587 F20101208_AABVGP mailhot_d_Page_089thm.jpg
e8951ea0233a61fea6b5670af1a21479
91e085f6b5a5faabb08c4be00a671eedb7e0f9e7
F20101208_AABUCZ mailhot_d_Page_103.tif
e6d29e39a37605da833d7e7ab37701ab
fdfc43d472b9af6927c77b00d0ea7f60befe6dc4
9519 F20101208_AABVHE mailhot_d_Page_107thm.jpg
7f6b279d1112ac5c6555bd88dba03077
94a56cbe4990f72b52be2265d0ad5d3a60a23c19
34057 F20101208_AABUEC mailhot_d_Page_062.QC.jpg
0e671e56b357a203e7f954ced5bbd019
1023186f67ee883997f7189c3c634d94af2bdf2c
119668 F20101208_AABUDO mailhot_d_Page_035.jp2
292f62a8054581a026ed024c92989ce9
bea57b37fcd65a7e9f3d4d3fdc473f12ba3d7c23
37238 F20101208_AABVGQ mailhot_d_Page_092.QC.jpg
d7b40ad9257b9003ccca1d0a6353f344
67d83f4e4da86a5545a05142ecb54e8890023983
9149 F20101208_AABVHF mailhot_d_Page_109thm.jpg
4799784c7f0999cf86cecda561a260a6
e49936c10b26f5add0388eccd8d28a1040860ea3
116940 F20101208_AABUED mailhot_d_Page_024.jp2
adb82a7d4ff382191be004c6398bbf09
39a9d285eac2684b7964f0bbd0dc43bc43b3f1d2
78575 F20101208_AABUDP mailhot_d_Page_009.jpg
d60aca26089a2050d4fc33420d67ccfb
fcc7a59079c4642a76cbb19052edb591cd80ddc8
6673 F20101208_AABVGR mailhot_d_Page_093thm.jpg
db71fd8d101cba214f4aa27bc54044c6
b793ce99b01193470a8d36e741749b0274308a9f
36649 F20101208_AABVHG mailhot_d_Page_109.QC.jpg
2d98a9baf6a9ba670fd73c514a638f21
1497cfa22fbeb09ba3066a8e3ec15bf71d8b9bab
9065 F20101208_AABUEE mailhot_d_Page_024thm.jpg
d2c5daa54952c96b48cf2faa33a90ca6
577cbb378e7656315679237ac4bb46aa1f248c31
1605 F20101208_AABUDQ mailhot_d_Page_072.txt
e9db893a391a8e35ded0686a344993f1
7627cff9ac7eb78326d745243d61467201006235
4544 F20101208_AABVGS mailhot_d_Page_094thm.jpg
63e0d3d96dfd3ef62ad98a0e0df55609
8059b74b42b34fc8a70d02e78dd3a0000306e4c3
37733 F20101208_AABVHH mailhot_d_Page_110.QC.jpg
b22a54b1dc2ea207db1af15b9952bdd1
afd38e27b4518fbc8807d9718a1304ae51ca1fbb
F20101208_AABUEF mailhot_d_Page_020.tif
c3c6471c023be2e675e285e2b4a33a34
680eb03a248c8e16309c11c9f4acace62d83f825
50265 F20101208_AABUDR mailhot_d_Page_062.pro
df81b5c011350437ac4b1469871e1768
1c642db3500a0962c2be0849291e9678d6b70c79
6780 F20101208_AABVGT mailhot_d_Page_095thm.jpg
2f218c683d3ea6484a69d45ff1c7b2b6
3f881205c7b36fd47c5a4a8a69cf676a31bb35ea
2224 F20101208_AABUEG mailhot_d_Page_016.txt
94711c819eeea847c5852bdcbf0ae1f6
685a3ee0074f6d71673fbb395ea1723c5340455f
F20101208_AABUDS mailhot_d_Page_038.tif
8e01dbe1cd15f3f2972b63befa01ac78
d440c8ad25449563c6551b800b6bb3acee520708
22812 F20101208_AABVGU mailhot_d_Page_096.QC.jpg
801c65090e6b4486de8b4ebedfa99f05
7ec4a75d8d49aa8ea82e518059d8f993c08915ae
24373 F20101208_AABUEH mailhot_d_Page_075.pro
00ad6b0f2953176718d78341c70f9569
58aa655a1e78ae343a9c8dcae0888cdf1ea3c038
115590 F20101208_AABUDT mailhot_d_Page_012.jp2
ccabe105fec1d5790349016837efc9ef
60f4dd8df03e43a49ff42d231aab5bf05877e2c5
4963 F20101208_AABVGV mailhot_d_Page_097thm.jpg
6d4d0336a73ea09163e7bc803fd2eece
b8df6cb3d4c9f1b6cd163208ccb30cfb60d5d383
116787 F20101208_AABUEI mailhot_d_Page_061.jp2
70a10e7c7a8230cc066e68733695db8d
0d3417c9049fa78550c22449eecbc35f8f76dd1a
F20101208_AABUDU mailhot_d_Page_015.tif
893273456a3e8ef22e9f92f32b3dee42
3b6da02aab67d4982b9b73cb56cf3b7558692cbd
5728 F20101208_AABVGW mailhot_d_Page_098thm.jpg
16756d8e35f4e1d2136704f73bef552b
a77a275e8b0449579db365e72c18b8743c4ecafb
114451 F20101208_AABUEJ mailhot_d_Page_028.jpg
62d015f421367a79750de3c628fd90e1
e3fde1aecc20ff568abfa9ddcc93d96211124d1a
F20101208_AABUDV mailhot_d_Page_033.txt
d9b44ffbbc382e4fec4d839467b88fc2
760488c247ad9b70376db833309162fe940c8c97
19407 F20101208_AABVGX mailhot_d_Page_098.QC.jpg
1bebadb2f49d7175f7a399850dad79bb
b24bb224655efc07e255b1ac0de20fe2d0e361d6
2786 F20101208_AABVGY mailhot_d_Page_099thm.jpg
9f1135c0df7b8750ebef9d07512aca76
7b3d57bce033adcb313317b6c1c2d9b25302b407
122805 F20101208_AABUEK mailhot_d_Page_083.jp2
c4ce5973439be5512db92e854898424e
3058bce4247d61c203a5d1e7f61dd840c7d89f49
F20101208_AABUDW mailhot_d_Page_061.tif
2abf506960bb4b884a70696c2cbff376
2c3d3d1412489e95ef20c2105234bf52d9a8e606
17307 F20101208_AABVGZ mailhot_d_Page_100.QC.jpg
6a901d5a7a04032d786b62f02b24d9de
b3bb4b7f11592d39d93c793a5d74796156211777
4436 F20101208_AABUEL mailhot_d_Page_045thm.jpg
5fadfb475504639681b9ee7a84ffdc05
b3251207a1b63b3c6e3647145d75c3670181291d
F20101208_AABUDX mailhot_d_Page_016.tif
56e655cd445a60077cc39c0a4eae66d5
37a09940cce0d0569b6b6ba1dc563b6949114eee
4760 F20101208_AABUEM mailhot_d_Page_057thm.jpg
451cf6bd1c56e77e928fa2d113d900d7
5262d7cb4b42d61347a96c28682a69e9c95409da
36359 F20101208_AABUDY mailhot_d_Page_022.QC.jpg
c104df1cd9ed235990c268bfe3a2e2f1
ecc8bf370fef4c9e9c2ac904c9043fdd2dad5c5a
121650 F20101208_AABUFA mailhot_d_Page_060.jp2
dadec25a7ee173575a81762b29526815
2587c2b669c7540aa616c84351ed64a0674e6f9a
36611 F20101208_AABUEN mailhot_d_Page_012.QC.jpg
10fcd15d1e78956c9a9cdb3e76ce5fc5
eeec1cad22fee58902c8fa082cea2fb31c2d1538
33285 F20101208_AABUDZ mailhot_d_Page_004.pro
eb736d7877413b8f4dd9a15bef59f046
122e91626cf144ba2ad8eebfdaeb7815f8372a36
34618 F20101208_AABUFB mailhot_d_Page_006.pro
8450335017ee6c291cbaae869a6420e1
c84c66f87fecb3259a755c5daac47f1a91558d6e
40483 F20101208_AABUEO mailhot_d_Page_107.QC.jpg
ba3b6a18f186d2a66ddbf70109c3a373
d7895219a7e2d4f54cb30717d4f4e8d92eede874
F20101208_AABUFC mailhot_d_Page_027.tif
ed0d92ed0e0ec833deb73d67b68fe546
de06e6c910980c3478ee3619f0f501bbaec9e098
6916 F20101208_AABUEP mailhot_d_Page_096thm.jpg
5f3ceb406161543536e877d6031bed7b
cbadff4d571dcc0ab6f14322c3796bd97372f43d
35124 F20101208_AABUFD mailhot_d_Page_067.QC.jpg
2b7c066fb7b333f03f5865456fdc8f4b
054575e3f5a9ad1ee0007b11c980e2ab01b032a3
1425 F20101208_AABUEQ mailhot_d_Page_074.txt
8a418d0cd9d46c0d1058a687ca502a76
09db8b8a5ecca8a081769c3255b8453b2e63782c
51727 F20101208_AABUFE mailhot_d_Page_066.pro
562ace2bfc5498d35c9e8fd775d2d947
9967c1a5d9e386939fa8658e3900a48ccccf70d5
54127 F20101208_AABUER mailhot_d_Page_036.pro
6492e948a02c51cf5e4049639281f5e9
aa922cc918bab58a194f2297a9f01803e6f6b9f2
115447 F20101208_AABUFF mailhot_d_Page_029.jp2
16b1560750bbe330963e22cae88a5469
ad6995cfafab3a5af98cb7b3f318b34800bd5ef9
120606 F20101208_AABUES mailhot_d_Page_022.jp2
b08e52105c7705987360c71171c66168
4fa0163f80d4bc1eef132f6b219af39e93c886f6
111519 F20101208_AABUFG mailhot_d_Page_069.jpg
ce91be791512282e00b246876a4d0c63
6e551d707883cc508c58bf2f2eb7a09a49290a86
35944 F20101208_AABUET mailhot_d_Page_087.QC.jpg
2e78f0097af5a230b2f22184d21459d3
8194bf8970f5b6fb5205924866581db2ca2239c9
1948 F20101208_AABUFH mailhot_d_Page_013.txt
1919bb0bce636ebc645aaa517c214e52
508028ad668112a6b841915d598de0de79b6433c
690 F20101208_AABUEU mailhot_d_Page_003thm.jpg
b31b8c77375cc66ea830fdf3791d6bbd
1abce56c866d93df89612280ffd5e52c87cf56e6
54712 F20101208_AABUFI mailhot_d_Page_069.pro
55e1ac8b0dad43926ddbf3573545993a
b5ce2339a24d0b57938adc8b52db909da59625c9
34374 F20101208_AABUEV mailhot_d_Page_020.QC.jpg
46bcbae42ecd739d9a343b7d11ce1383
a981c7d6d597dd67e13e3a38b3cc0622f3f081e8
F20101208_AABUFJ mailhot_d_Page_003.tif
aec1c4d26ebcb2e276a5494b88b1fa31
73aa2192c88eeea57bbf069eb45286df0942f2ae
9129 F20101208_AABUEW mailhot_d_Page_014thm.jpg
6364a196d0ed8c2289982c019a9737e4
39f71899952d05742a58ecd6053326769eebb811
111264 F20101208_AABUFK mailhot_d_Page_086.jpg
163e92ea8408def34c2c5f8601d3e923
02f117544aafefb6893921d0444299d3cafbc34a
66242 F20101208_AABUEX mailhot_d_Page_046.jp2
f7a5a346bc48ad9c721d60585127444d
69e4a4405aae31a86ba89e6afbe772813a4c54f3
117712 F20101208_AABUFL mailhot_d_Page_041.jp2
129b27d640e6bc87edd11b88526d857d
7271e77964406873e9959fbdcca0f07053afa203
47607 F20101208_AABUEY mailhot_d_Page_094.pro
62357bc21f81e2252bf900c865959389
9584bdd48fa05dcfe5e288a0575d37b5502fa3ad
447503 F20101208_AABUGA mailhot_d_Page_053.jp2
436ee31e5d66529a2ea5662474af4d8e
60d03b8b7d4df3cd55a69338b84894f80f09534c
F20101208_AABUFM mailhot_d_Page_022.tif
070253dd3c8b9f1a925a203a25f06162
38d4bb2c1efe0e7ce76cdcceb43fa4876c03177e
3446 F20101208_AABUEZ mailhot_d_Page_047.txt
1a01eaeafd4663f4ceafa21cd4a30149
3ab21cc3d8dc58e04456ebf333c5fa158a4f30d5
2941 F20101208_AABUGB mailhot_d_Page_113thm.jpg
68ebfe15f3f1738caeeb822f1ccad59f
64a4de10035e4eba175bb69ab74793741e36daec
33371 F20101208_AABUFN mailhot_d_Page_013.QC.jpg
5d43111eadab03db5fea6d91315a2da4
062171c553cc5cd36bd64a6a855558d6c111bb8b
333428 F20101208_AABUGC mailhot_d_Page_089.jp2
a9b08b11a49d542a521e91859f6ce38c
bc9fcd4688f0609ee2646b31b8064c518bcb3007
627953 F20101208_AABUFO mailhot_d_Page_049.jp2
8c0497a7d14447a749c3bc73c248a370
cd55bb7ea4810c05a52a6defd742bcf158b27379
9052 F20101208_AABUGD mailhot_d_Page_019thm.jpg
5fc401aeb5c1505281b425a8a6845784
587f09075524b705c9f81fb12aa4d6044735ee3d
3582 F20101208_AABUFP mailhot_d_Page_092.txt
da3c4b75b630eb6287597f386c6891b2
2cbd3f739119f01cfac1168c214635d8ae1b2c3b
94750 F20101208_AABUGE mailhot_d_Page_090.pro
14cb7e4e85c4659f37f53aec84aa52c5
9bd7a0a290f50c72b385c2ce3dabd77e3d5438e5
30048 F20101208_AABUFQ mailhot_d_Page_073.QC.jpg
4b8e514d42e7e19070512079992dcc24
97096b4da282c6449f39ab31df7a2135c17c8660
6907 F20101208_AABUGF mailhot_d_Page_077thm.jpg
2828441696c36ce84d3bc6687526bb91
12f62c1ab0e74851438cd5df95a2a9c86e5f35f1
9248 F20101208_AABUFR mailhot_d_Page_071thm.jpg
e71fc88d01a8360e732707f38f64d175
18b45c922f493f694bc333683a5ca1b96f7dffa2
9087 F20101208_AABUGG mailhot_d_Page_028thm.jpg
c100d976419953e9a928c238eb881c57
9b75e2cd6fa0ea247b1f5e8244521310fbc9c8e9
1636 F20101208_AABUFS mailhot_d_Page_003.QC.jpg
f259cbf9cc4d17adbe893a0a5b06c8fc
e30561912b5b80904d354dde8301c8554e781864
244019 F20101208_AABUGH mailhot_d_Page_076.jp2
9cc2e5ba3fdfbe2c3333a84d6d760f02
99ce74789308350844e957eedff55dc597ec5e7f
1346 F20101208_AABUFT mailhot_d_Page_044.txt
32e48cb83fd24003cc492b08f7f48152
8c27f968faee2bb460645c2a7cdb4a233bd36988
25271604 F20101208_AABUGI mailhot_d_Page_049.tif
189931d822b59b3daf4b4f51ee9979a2
e0df9a132364a460b525dfa01ddf9b0c6d5b9207
37458 F20101208_AABUFU mailhot_d_Page_064.QC.jpg
b8fbb3c989c56b3e43f54326fb3fba89
f37fcd22668d41c86bb4df92564926dfe0f34f49
8820 F20101208_AABUGJ mailhot_d_Page_020thm.jpg
4af52077556f90343fc1eaaa00213120
3bf7d0c5473b050779583fabac51c8377b4c2c93
53399 F20101208_AABUFV mailhot_d_Page_081.pro
5c78d4c4675f2a6bc32d73d1f0386370
b73c00d008804d54a0144e724511242d31aa8f64
57001 F20101208_AABUGK mailhot_d_Page_083.pro
856fb1d1796315d1c3594622e80c4c1e
888dee9e338d5cb224fa76a46cba1dc8a777189d
5705 F20101208_AABUFW mailhot_d_Page_056thm.jpg
822f680823ea2cfa6cf62a5b2ddcf142
4dc129be0b99c31976b467c1ec66be6a611c115e
72035 F20101208_AABUHA mailhot_d_Page_077.jpg
831e2f3b163d4a53dcecc28b9584fa90
f582d248d4dba7e3eaa8cef9b02b57375014325a
9195 F20101208_AABUGL mailhot_d_Page_037thm.jpg
ef543d8f8606e80481b49d64e29a3fac
b9be2b9a25667407871e07db2c22c93444ecc034
F20101208_AABUFX mailhot_d_Page_090.tif
ad45197dc8ec4fd42178559ad7b80579
1989aa9a7a483fc4b5ac828c76f58521fff7d54e
F20101208_AABUGM mailhot_d_Page_011.tif
7a0df747b3310162c2ef8fb1c1a97dea
c088419b698c4b5cd854d5ce9fe785a967d04288
36835 F20101208_AABUFY mailhot_d_Page_034.QC.jpg
711d3224306cbba954ed76fd58c7d4f7
870d85b3efb4a86d7c6a1c812155486eeae6b788
F20101208_AABUHB mailhot_d_Page_060.tif
c4ba3ee3b7884bb2a34e8b4d167eda3a
02f0fc2920de6af3931e4e89dd4c82eeda75898f
2109 F20101208_AABUGN mailhot_d_Page_024.txt
31ef1faf09c32072c1e7384402fc0796
ef131c73a192ecc66c757ca7b6b1938090682d7e
118086 F20101208_AABUFZ mailhot_d_Page_032.jp2
17bf7a7a199ffec5039dee446b000ccd
603ce89d05461adc0b9e096e4c87db9e44ba4ed3
131735 F20101208_AABUHC UFE0018161_00001.mets
88ddbff0d9bbdc18af9f34dbeb725a8f
578b16e445f61a5dd4b5fadd124fae297cb5a9dc
8758 F20101208_AABUGO mailhot_d_Page_026thm.jpg
5a51ba76c364f3d3d5e31d261aa02e35
cbfe0f4fdec51631f98be0f7c32586c1a9ef60a3
108335 F20101208_AABUGP mailhot_d_Page_015.jpg
38caf0a48b3261e333c326787e709a58
b838a1291c5aa3eac7c9b8af4e9ba59be56f691a
103662 F20101208_AABUGQ mailhot_d_Page_102.jpg
cceca142b03253c219630eb310f4cdb8
0c3b4d35e6f30a20eb2baf0629eb5a27ddec65ff
23749 F20101208_AABUHF mailhot_d_Page_001.jpg
b6be72e20e40c96a2d6c4de0865d0a49
59d8bcc07aa4907520fe1d1e47e3b5297cace3e1
58965 F20101208_AABUGR mailhot_d_Page_038.pro
5c78404261cbfa14de31318c0ae0abc6
7c52993242639cc5e1a4ef9142b27d2eadb525d0
5145 F20101208_AABUHG mailhot_d_Page_002.jpg
801afa6bb9e180f3ead90e88d05e53e4
fc135c344fc7142035373d0c0f476a873b5b3b08
55463 F20101208_AABUGS mailhot_d_Page_019.pro
f1c11fa437662c99ba6a6ae6a8f6a184
07538cce12faecbb13d315b65aa72c8329b7500d
6060 F20101208_AABUHH mailhot_d_Page_003.jpg
efdc19fbc30846f8fe03ac62b1fc33a2
69f2b6235280708a3f430d6aab2b11fe1de37b5c
40593 F20101208_AABUGT mailhot_d_Page_058.pro
0d4d4ad138ac139ea40feb9a65d231f4
b797e5304b03c83b59d83126074bcd68f8b5c78f
8383 F20101208_AABUGU mailhot_d_Page_018thm.jpg
998a58fd2ce7320d01e3e4813a7004ba
3da237617f2eef6e4af6fc05d7b05b5f7661a7a0
72992 F20101208_AABUHI mailhot_d_Page_004.jpg
d5d6929ad9e41536b2bcdc5a07ce0133
457954a3ebdb9efd134e0d5fb33833aa424afe29
115309 F20101208_AABUGV mailhot_d_Page_037.jpg
3d12755260d9674234cbd9d82f004ab3
ab3eb01bcfa645b3523e586617c509bcb3735f80
124550 F20101208_AABUHJ mailhot_d_Page_005.jpg
59022fa773334947188734bfc9d79147
021f048f3b30f4b914dd8feae3adab2c9e9a3790
99044 F20101208_AABUGW mailhot_d_Page_073.jpg
a5378620b1572387a0f24566488a4707
32dfad4e8bbddcb31bdc09e29266f50bdecfedd9
51826 F20101208_AABUHK mailhot_d_Page_006.jpg
2ecf2b3c6ddd15b28f408e7f6e716cfc
1d9d9432529c2bd714e2fde78243c285f76bef18
2848 F20101208_AABUGX mailhot_d_Page_107.txt
f0f4b9cfdfe555e7f15987f971342fa8
0f480f601e0391806582176bfa7b128f74859f2e
109068 F20101208_AABUIA mailhot_d_Page_024.jpg
c18609641dd40454d347452c4aeef30b
c2a9055a1c34c86eb0edde78a7a247e7024869a4
122493 F20101208_AABUHL mailhot_d_Page_007.jpg
4fb3afaff11ff0bdcff1c14071d8864b
1c99de05baedf73ed2a8722370f15868e9533324
408714 F20101208_AABUGY mailhot_d_Page_050.jp2
1994557e6403bff3867abbf1ddefb345
16afbf0ae07b8b644d2d68d30ef2247fc1ac9c31
103996 F20101208_AABUIB mailhot_d_Page_025.jpg
0ba6ef04ca23c380605695ae9531469a
01a53f7974f9c7a763584d97ae0d667240c1b499
93553 F20101208_AABUHM mailhot_d_Page_008.jpg
23c85b7ed9089866fe13dde467f54064
3ec7b02be0bb0ce3754f9296ebdf2a1d68f97d00
7230 F20101208_AABUGZ mailhot_d_Page_055thm.jpg
124ec774b1e23f8e60c894ed61968bcc
a1ec7916c8904ab6107bb24ad418bce0517bf326
96839 F20101208_AABUHN mailhot_d_Page_010.jpg
5d6322f767d5edf6d94e686df753097f
3833799c014385dadb947924c75417bd92e982ea
109039 F20101208_AABUIC mailhot_d_Page_026.jpg
af7ecc4dabeb161522bf72d63fa83d64
da603de3afa686107a291c14573ba6be93776214
39798 F20101208_AABUHO mailhot_d_Page_011.jpg
4b8c31b082bffd2d569229296f8908d1
7d82212c58f2c7d583c9669ae379d76d5f476ae8
107015 F20101208_AABUID mailhot_d_Page_027.jpg
c87af3e47993765697a8f27876a1bdc5
fde03b55ee60b32955ba313707ee591174c60f92
111037 F20101208_AABUHP mailhot_d_Page_012.jpg
1d3e5f8ea94541fe239c9c3f13507168
3f52e74b60a2f53bb3e441b0e326a0281c365a88
108812 F20101208_AABUIE mailhot_d_Page_029.jpg
2d9115266f3cd48d608bd559a8443af9
38b7203498043a3a54480b1104cc3da27a97ffc0
100750 F20101208_AABUHQ mailhot_d_Page_013.jpg
f140bf8cb84ad209de9b6937158742a2
a0bdcd40db36a9ce89d4e218323a47f8f60a594d
105299 F20101208_AABUIF mailhot_d_Page_030.jpg
70df34f8c34178a8698bdfbaf7787e8a
19a5b29a03fe0090a4e99f72bb91522190d2fadc
112324 F20101208_AABUHR mailhot_d_Page_014.jpg
a7a1101259544466a2f5c4c43332d0f3
617ba8531ca2e1714915ad8bfa52c494feeffb1f
112306 F20101208_AABUIG mailhot_d_Page_031.jpg
c1bad47d0c53712918878ee467020dbf
530a8d43545dac58de6c5dbe7920646a5c5bfb76
114383 F20101208_AABUHS mailhot_d_Page_016.jpg
169c9612c95fc14bd1c0d069b26f0c5b
b36317bd8655fed14297ff9673019558468aeda0
110299 F20101208_AABUIH mailhot_d_Page_032.jpg
758e09c1042b779c0b4dc5c126c4e199
48e9a8e877c55fc1b4631cfac844c7eafd24775c
106917 F20101208_AABUHT mailhot_d_Page_017.jpg
a2fcddd88406db06ea2b8a0071c9a964
0047bffc1b58079749a488ee3485788f584a4695
101700 F20101208_AABUII mailhot_d_Page_033.jpg
118120e4917f3365e2dbe894846c2ee2
2109b7d90f16f49c79bcf7839f0ec5de4e6c1b19
106910 F20101208_AABUHU mailhot_d_Page_018.jpg
ad9966067b24656000259aa431f67781
b57efedea329414d6c30fe37bfb2bc1067f7f82d
111568 F20101208_AABUIJ mailhot_d_Page_034.jpg
00f3881618066772cc709c43e2b9a1e5
aae50501cd5092512653a22c5e5028b6b3ff4241
111474 F20101208_AABUHV mailhot_d_Page_019.jpg
977967e6795c91ea7e836877b8a8bc36
e35275ef0271a6b4fdc353ca3a2747a653c739d7
113017 F20101208_AABUIK mailhot_d_Page_035.jpg
e7df0822433053b5a3174903cb1a81f0
00e37a0af77bcf04a195a13566f068ca07b3edda
103995 F20101208_AABUHW mailhot_d_Page_020.jpg
08272b26ecd714b0c23a86112fdf4e40
e2f93920dfcab27fd5e7b5323e701e11e1608dd5
45564 F20101208_AABUJA mailhot_d_Page_053.jpg
802ce6605c7ef5fdfc43fafbd6793b4d
6b7f774124e44ee4b4d339ab812df0816113ad60
111553 F20101208_AABUIL mailhot_d_Page_036.jpg
5c51a2138c424694df86b9f2332f4e9d
a6866485725149b1edd3f9d30539857ce99cd616
110848 F20101208_AABUHX mailhot_d_Page_021.jpg
57eb770603dc261ec8b9d2eda02a87a4
c654be445b3dc6075bdc6bef84cc14f1b8270a7e
34307 F20101208_AABUJB mailhot_d_Page_054.jpg
4e66d1d45a6d8bad973b77b97eb18637
f77d38d7da55831ba8f8dab79ceb5da5bedf7535
119201 F20101208_AABUIM mailhot_d_Page_038.jpg
e47e4a883f97632dfe4280a99a202ada
7a9844af6dbe8ff84952f5c5f534d401d6d6f2df
112066 F20101208_AABUHY mailhot_d_Page_022.jpg
54b21f55c6f3f1f934a522d05b12bd04
05b5a382a0b21aeafa7a0989a2dc7263ea99e17f
93268 F20101208_AABUJC mailhot_d_Page_055.jpg
389d2fbd386fd4fe0ebb8fd447ec3a18
f2e155364e86a45602722ecd7b1d4895692a4456
111110 F20101208_AABUIN mailhot_d_Page_039.jpg
429d6331b489d3a379283a8363722027
8bfbd7c15721d05f1b8e45455b46b6e312036204
115061 F20101208_AABUHZ mailhot_d_Page_023.jpg
ba9730ddb679bcda2039c39d24f48411
33b6b462eeea6d9a0af8d3bdcb21679064c5d877
112097 F20101208_AABUIO mailhot_d_Page_041.jpg
14ad9412189a3676336d9b886ab11977
e1018df80b4b69dbbd716bf58a8ee51300526c5e
54124 F20101208_AABUJD mailhot_d_Page_056.jpg
b675170a967e396ddc6393c746653cf4
1cabd89bfa6956162f4d4aca8c52094ee7b3d52c
113532 F20101208_AABUIP mailhot_d_Page_042.jpg
d75c580af198ab1de414afcb90c803df
6e16063c19244132d656405f13b98816ffd4c16e
50019 F20101208_AABUJE mailhot_d_Page_057.jpg
a30a6696168d797103ed0c0bbc23c5db
6b63566821f97ff0b59a89921335ee9d2dd89cba
89789 F20101208_AABUIQ mailhot_d_Page_043.jpg
5132d97cf467d3974a270c4cbf1206a0
2368b1d9932389dd1b891b01e39e2c086e7984d9
68167 F20101208_AABUJF mailhot_d_Page_058.jpg
ec5bc0d806e1c039096cbcd3dbb0320e
bb7cfbc70f2b9e73dae36c0034034e4d0859a409
35034 F20101208_AABUIR mailhot_d_Page_044.jpg
7afdfd3f4af0183218a3022c4a43d44e
3d99b321a2b1c2ca8c0769cefc4fc3a2001f7aaf
105669 F20101208_AABUJG mailhot_d_Page_059.jpg
57f04c34241d018659ca0f33a940af18
e50ae1c2e249a03f5bea8bc5fa84a4451a4f6b8f
50332 F20101208_AABUIS mailhot_d_Page_045.jpg
18e3095f8b8ea0f57eda012c68c912ef
17c24c7b01d27516630eea68210a48402aab592e
109510 F20101208_AABUJH mailhot_d_Page_061.jpg
644f1c130a62fa2e333bfe04a391937e
d21d3c92f0555584d06305248db8ffea0a137132
46002 F20101208_AABUIT mailhot_d_Page_046.jpg
58d6f36f2f261d0a16df5f2a853db1f7
7fa81fbc9fa52527083b3067ce9c5f4148549719
104612 F20101208_AABUJI mailhot_d_Page_062.jpg
41d249be2ac511aeeccfe96da768ff09
3cbfd19539cc4e51539a1bd568f5bddab5f64981
122623 F20101208_AABUIU mailhot_d_Page_047.jpg
9fba8246d04be2fac3fbf6201d17d1f5
a91c30fe0db8c7d47793bc9b9f4c64cfcc1906c3
120790 F20101208_AABUJJ mailhot_d_Page_063.jpg
3bd846bd50b0e1cd03a0a456d7264258
ca904c4951a0a929fb8c2e96b492f50daf27e7c4
38965 F20101208_AABUIV mailhot_d_Page_048.jpg
2e7e8060bb0d976112c4140afd536a09
38059fed9a5c483bc825b225661dddf48715a992
114823 F20101208_AABUJK mailhot_d_Page_064.jpg
1cbbb63918fbd825aad11b8065cf8360
4f24dc39e38b7f232219da53c8f91ad30f20742a
69206 F20101208_AABUIW mailhot_d_Page_049.jpg
e6a11a790861536cae94d1e96e08ac6e
92155a7a093c1446cc5f84f54476e0e190a8a324
116969 F20101208_AABUJL mailhot_d_Page_065.jpg
744ce3d628f86f32d8d6fe6226f592cd
41ef6d5add4c657f4c5f16c2ad40627af2c5ac76
52394 F20101208_AABUIX mailhot_d_Page_050.jpg
ece9567d3dfe5e411b0a78d7816a9302
81eb328fba744489dcc2a38bc5f9b3123fc19bd3
113608 F20101208_AABUKA mailhot_d_Page_085.jpg
d3057be84f545ed66cc6c1a95fb89bfb
5d908f2de88f6c929723a1711189337f291f9f2b
106479 F20101208_AABUJM mailhot_d_Page_066.jpg
02fba211e17990276efd8d0e1cbd4674
9b2922257183f43e705b1e42c40c0770ab90f9ed
51922 F20101208_AABUIY mailhot_d_Page_051.jpg
5a01592e98798eebeacfe5c8e6cdb794
908322f1a7b2dd9d51c819513e5bca038b2e5ce3
111840 F20101208_AABUKB mailhot_d_Page_087.jpg
193ab62c915ed0f26bbccf7558640e22
1d7552b9eeebe1651b2e52061d2a8e1833d8fd97
107487 F20101208_AABUJN mailhot_d_Page_067.jpg
b4d25261308a8edf22d08a09842e04e0
bf2a6451262ae13f24a53cb5f7dc3667e7dbf744
42371 F20101208_AABUIZ mailhot_d_Page_052.jpg
8e656482a6a305a41f7acaf5c7cfe84a
c9ab9768e7e242312f8259bb5366c8cdeec6d653
63827 F20101208_AABUKC mailhot_d_Page_088.jpg
346994b74b1ee2461a85e80a0cdf074d
5821aa71da936bd60905a1e7cdf8e8237dd662e3
111441 F20101208_AABUJO mailhot_d_Page_070.jpg
0a59fbf7a8e1af8aaa65acc5694174f7
3a4393792e475bb1dbb31ced827201f77a2b6a66
23279 F20101208_AABUKD mailhot_d_Page_089.jpg
5c4b0fdc4cca23831d94fa9777b6f7c7
6f94d5b808a71024ad188be749e199db2cce298f
113665 F20101208_AABUJP mailhot_d_Page_071.jpg
3fbc11b84e3f2f38eae09f32153b8341
131845ea85209cfd02813679253c085a3e7bc616
83179 F20101208_AABUJQ mailhot_d_Page_072.jpg
8fc2148da4ef5e241bf99c9fa6fb632f
3b557641e21c606e5e5c71187e544827a31c26e6
138263 F20101208_AABUKE mailhot_d_Page_090.jpg
5e40ec69b0a576f7b68c004a9fd35d27
a87a64fd93167369165d358fbc825a6000ad98a7
67643 F20101208_AABUJR mailhot_d_Page_074.jpg
db55cb3a211effa44ce189a6e5be5b32
88339b0d2b766e0913718d436000094ae6c63f8b
133736 F20101208_AABUKF mailhot_d_Page_091.jpg
d1b515b05517c487920dd61c71297512
f49c35ddc6a30241d6ee79f79e287b457fb42acf
56890 F20101208_AABUJS mailhot_d_Page_075.jpg
0660f1b449c09f169764b465029d3de6
a2433a0f5f6d89ed52dc5e0f9787894d0091c7d5
138110 F20101208_AABUKG mailhot_d_Page_092.jpg
5c149e6fba58ff47b4318310beaf3835
1d733cbce01ff1cfa4fa0099496f548c2ea1bbc1
104827 F20101208_AABUJT mailhot_d_Page_078.jpg
b015c89392803bee02b1af51ce5fff16
8fc927aacf0836064005895b669afb2744d669d7
70517 F20101208_AABUKH mailhot_d_Page_094.jpg
40e985621bceca5ce19debe9e35717b9
9c0fc49f224738372c425c5a4cf2defea25ac786
107083 F20101208_AABUJU mailhot_d_Page_079.jpg
9dc4cf8f05260ee0365478cf2a3e0079
2fe3adff6878e61aeffe7e5861585ac27881d23d
58634 F20101208_AABUKI mailhot_d_Page_095.jpg
800c2397d79c811b2ed2be96e77167cb
119c12388d74ee0b87b9c8409ed515f0c7459766
113553 F20101208_AABUJV mailhot_d_Page_080.jpg
483da4f7d9e31f64f13000512a776260
dd47dea18aa3875d4084f31c32cbb1e9c9a33545
58903 F20101208_AABUKJ mailhot_d_Page_096.jpg
6496cbdd6792afc5e2866406e1020411
a7e9f399e27c5b022a0850d366ace5a4f541d790
107726 F20101208_AABUJW mailhot_d_Page_081.jpg
60310379d3d46ee65c2a847dc4f3b15c
dae48bb125db134e48b02787258a92771f3fdd22
57178 F20101208_AABUKK mailhot_d_Page_097.jpg
ff42e61d6e805acddee022fef3647561
daea0ee5ba03aa9f266752e35db194419373bbef
114720 F20101208_AABUJX mailhot_d_Page_082.jpg
f8d0b169a8e91b80df459ccca8b0955d
27780408bfb74f92ab401f201ebb72a6115c36cc
23478 F20101208_AABULA mailhot_d_Page_001.jp2
7ee0375f6dadd6c785cfa3ccf95e8c13
0403da2c9854ad16fb4f590b60d200b7032c7756
62448 F20101208_AABUKL mailhot_d_Page_098.jpg
36d1ff31c01e2d16cfdf5edccc0c0a53
dd4a11c50d1bbad779160dc253f9784c59601078
117252 F20101208_AABUJY mailhot_d_Page_083.jpg
78b60987a0575828f484855ebd4371a3
8497567720ed0e9357f6b8cb75b42c86a8fdc2b5
6156 F20101208_AABULB mailhot_d_Page_002.jp2
3e9b0fec8ccf39b60171a5862beac6c3
de1f39486bf2183fafee454eef42e6d24e88c768
37549 F20101208_AABUKM mailhot_d_Page_099.jpg
a800dd156a6dc63a9e3886c2b680587b
b5388899302c15f45487cc889ab1ba28fea4b397
110826 F20101208_AABUJZ mailhot_d_Page_084.jpg
c220a5f24d1539eed57edb139ad12dcf
0ff8abe8a78cd0c7de6f8806b0107b250a317845
7137 F20101208_AABULC mailhot_d_Page_003.jp2
f516174444b51bbffd7a7832b5a5395d
0a7bccce6285143f1b71e29a72ea29c29e8a2bcb
68178 F20101208_AABUKN mailhot_d_Page_100.jpg
9557780b316f65710ac2a566ee54351d
868095b0fa17aa543c66d894636bb06a2a7015fc
74585 F20101208_AABULD mailhot_d_Page_004.jp2
3159d682849c3b4c598d76cc650d24bc
92575ec5e82bd9c0e624c144ff5ffdbcbfb6df1b
110348 F20101208_AABUKO mailhot_d_Page_101.jpg
f383960b1438e9b766cf8048a7a4dcdb
d3f54c5c310788f85b6aa176c417033737defc22
1051971 F20101208_AABULE mailhot_d_Page_005.jp2
cbaff33a137014f87573363ceb1daa45
32fbf1443d69b52a55188e6c08c90033eaabb3b2
107087 F20101208_AABUKP mailhot_d_Page_103.jpg
a07ae54eec961e88b684712618bdc41c
ae8386d4b1918ea1c6122178f0608b4903db096e
109838 F20101208_AABUKQ mailhot_d_Page_104.jpg
62fb38b089a219a42fa8d47cf233f224
408ff665237549eb378c579a2314c9ecd18df3c6
1051984 F20101208_AABULF mailhot_d_Page_007.jp2
af07bb33786e477db8b26a64f8f5aed1
21fc585abbbebd4c20585f2379eb728aea780b80
12577 F20101208_AABUKR mailhot_d_Page_105.jpg
2e677e53bc9fcc08bcfcf4330a7aa563
e3a985e558e4e749d0c508e29bcb45934c4359ba
1051976 F20101208_AABULG mailhot_d_Page_008.jp2
2dbfa6700dbddd5bc2a8e65e2b62c087
333b00770600ba02be1e1f24920193ce0ea4788d
132572 F20101208_AABUKS mailhot_d_Page_106.jpg
7e95d86cf089daa63b5ef26d9323a0ea
3c276bdbb41c50b0a6be93906d9cbc5fdf3240cc
1051962 F20101208_AABULH mailhot_d_Page_009.jp2
286331a2b5d79e29c958982c60b150c3
4d7bad6567ff383ff59dd6dc5e2c85fd4ad6f215
145634 F20101208_AABUKT mailhot_d_Page_107.jpg
bcd6364952df3fe1980a78124ed0ccca
9cdd50ae66cbfac6185a09d546fe4ab599c98b6b
101208 F20101208_AABULI mailhot_d_Page_010.jp2
f1fe16ab8cd7311f01f24997def8635f
df71c1127b299270c7d089ccc5052708d8802f33
137951 F20101208_AABUKU mailhot_d_Page_108.jpg
e407dd63ca25967fbf2d4c80ce64bf07
d03edd6085fa7b9cd9b6ab7f3db69bbfc79df751
43224 F20101208_AABULJ mailhot_d_Page_011.jp2
27b23de1275a2e3fbe2ae40917ea0415
80af074c14f55026fe55755210efbc2674e13a2c
122309 F20101208_AABUKV mailhot_d_Page_109.jpg
1feb25c1c6a96418a0ba50157d12f92a
f9676a07d368d681c7c938a39d935ee337563080
106633 F20101208_AABULK mailhot_d_Page_013.jp2
57cab13e62154c6b94b36131efe07a34
2a6f0e8df5c257bac6b5731ffa095b1fc70bff19
134088 F20101208_AABUKW mailhot_d_Page_110.jpg
b29db5af806497b65cb1189b080183bf
4607ba082abf136cd8029122447a99b5db82cc58
F20101208_AABUMA mailhot_d_Page_033.jp2
5608ab437b433cd3801ad9b33ffd6a2a
f346ba71bda5f1a91321ef3a3bd3a4c0475e39f5
117716 F20101208_AABULL mailhot_d_Page_014.jp2
8f1cab24936340d7c709329747f7cba7
48044828523c5b1d05634bb834f05732936a7bb7
130858 F20101208_AABUKX mailhot_d_Page_111.jpg
4d528e777e0b82ac163999ebf44f38c1
bee5ae8534bf7acc74a849732a7c021ea2ccdb07
118503 F20101208_AABUMB mailhot_d_Page_034.jp2
29a9691ec895ef68f7f8b9a5d1760903
61ca6161e2280b8807f98d2d8bdec14d0f83a72e
114255 F20101208_AABULM mailhot_d_Page_015.jp2
3bfdbbe4ed684d125be61e836556075a
9fb45a05671347fcc6937cc148b882b460d202b0
30409 F20101208_AABUKY mailhot_d_Page_112.jpg
d6dfed33fa20ba8df2d0100b4bd348b9
b819382f7ae045a6495da328d7e843d53ff3304c
117289 F20101208_AABUMC mailhot_d_Page_036.jp2
868f1a7c195ca6927ae18a0f6ba83143
db19d88d3d7f7789f37b1c8fb32eeb423c199ed0
121477 F20101208_AABULN mailhot_d_Page_016.jp2
7bbd6208efed9b36347abeb738f434d9
e5dbbe636b2ea59a9ca65606cc38be23c95eeed7
36342 F20101208_AABUKZ mailhot_d_Page_113.jpg
42519c827da4bdad20159fee8900099c
22772a975f83fdeabd36a527ecf7a570418eeef7
122377 F20101208_AABUMD mailhot_d_Page_037.jp2
eb1eeb7df38d603ab9aace58e4eb2e48
4ffc60bc2709c76fc664a5d8b8aa075c23484bae
113138 F20101208_AABULO mailhot_d_Page_017.jp2
d3cdc2f2e6b8850b4e2cc45152b0d0cc
74bc9e4b54c6f5cb72fbc67f6e3d47a4186aedb8
124358 F20101208_AABUME mailhot_d_Page_038.jp2
efc187400d9722698318d7356f7f9b52
2eeb74dd2405f5a687cfbd4bff68d44b50dc8309
112987 F20101208_AABULP mailhot_d_Page_018.jp2
26eed78aaab8d9dc41c70579153b3a43
d2f8e9af706f300558d49f3d64ea376c3841d282
116986 F20101208_AABUMF mailhot_d_Page_039.jp2
b979d30ae14334d998ecc70e8212a252
72f71051739b7cddf1464372c66b0ee67f8b7421
118793 F20101208_AABULQ mailhot_d_Page_019.jp2
34f7e8088e29324eb4bd8d33cd7a6aad
31934f23166dcf0451d11f33359519b434edb621
111796 F20101208_AABULR mailhot_d_Page_020.jp2
6c4ed67355eabb562955b9a284a0355d
4dc5658a6ad48ef50d4795f835ac86e6c375bcc7
118961 F20101208_AABULS mailhot_d_Page_021.jp2
77a391882c9fdf79927f6858e96e6174
2339d0d5709d51335d4c2e658e59f511c6ae7b7b
116866 F20101208_AABUMG mailhot_d_Page_040.jp2
5c0521887a97d71921915235dfd9d1df
1a47d5100703e03414786d5bd5cd32a6c41a83f9
122715 F20101208_AABULT mailhot_d_Page_023.jp2
051994edf9d1f05067e0c4d1d1760f44
d493adafc1684c063554f229b6a2ed0c4beec1ef
33046 F20101208_AABUMH mailhot_d_Page_044.jp2
eb6419ff09edb4236f9d80d917f109eb
6f0f2ae2c62eff2876a142c393f3edf81b1f49a5
108920 F20101208_AABULU mailhot_d_Page_025.jp2
12ab68149370961f89649a76eedede81
443e87678f734afa33ee1b1b8037ac966ccb2c21
72049 F20101208_AABUMI mailhot_d_Page_045.jp2
8b1205f07729920f6b963afcda6b973e
c03f025ebd137cc93dc25b4eaadc7ac35cd4a292
115288 F20101208_AABULV mailhot_d_Page_026.jp2
40c1def4272f057841d74009e0ec1e21
3209bb07d8fee32dccd92068c64da84aa2f0ea90
118810 F20101208_AABUMJ mailhot_d_Page_047.jp2
dbc94ae255f76b49d6ddb0b71c5ecbb0
3082c4861c712237160d55db9dd42febf3a62b5f
113968 F20101208_AABULW mailhot_d_Page_027.jp2
b139ff3d7a58042c36a0461e6242499f
8a0c9dd8d8af6fdac7f7a45fef8e2d556406d476
39116 F20101208_AABUMK mailhot_d_Page_048.jp2
e03dd9a807506c3a3ea5fc39e42b5d7f
ac4f6ab206efb687207e76c538ab14015308e5a5
121577 F20101208_AABULX mailhot_d_Page_028.jp2
dc537b7db24d378b48eb08e5c7d39e64
cddbe1389d189d2facbbb8c4d75eb586d301d9fa
118400 F20101208_AABUNA mailhot_d_Page_069.jp2
bcfe123e08cce407312e779d8ef321f6
41736fa8097fa978eaa99b84c0f9686ce3daf2fc
493500 F20101208_AABUML mailhot_d_Page_051.jp2
3f3e5f63bd2aa3f6554c01c7e2e46833
c252fb835f14e5afb65d77e69cc4a4143148dbfb
114458 F20101208_AABULY mailhot_d_Page_030.jp2
d1d02a83be0343e61feea1def568eac3
77b8668d4d473d381926948a9ca6b01ea34a9b8f
119104 F20101208_AABUNB mailhot_d_Page_070.jp2
eef990049dddfe7c9f077992e4dc2e74
01e1d856163f65185a137127e2b1a812c029b0e8
390905 F20101208_AABUMM mailhot_d_Page_052.jp2
97a203aaa37dc6be349b8a5f727aad80
e837e6f3da34fa3d93d923bb7e164b2fa4cacce8
118803 F20101208_AABULZ mailhot_d_Page_031.jp2
c02e1e5b91f2f0c2f00c46a6f1929113
983b656d1c9e5dbfdc517458c733601de857b014
119438 F20101208_AABUNC mailhot_d_Page_071.jp2
118d2fba8649ef6856066f4b70f16ae0
13e120ae779b8315efe6703985972170186d5188
313588 F20101208_AABUMN mailhot_d_Page_054.jp2
e5d67ee62c73752c8e1dc3598193acb2
6d65c6c1e2a94b173be600b3b0faba910c322cbf
89463 F20101208_AABUND mailhot_d_Page_072.jp2
e20cd2baa7fdeb91930ffe24d98d82ba
3e36a43c014c1f708abf8d256179fb98755d8929
85394 F20101208_AABUMO mailhot_d_Page_055.jp2
85301bed691a230adcac3391778b9b78
aa47fd66a8cb814d542b4c8794abb920d3fe5124
98463 F20101208_AABUNE mailhot_d_Page_073.jp2
03cc6d294f6614982526e25ad20b399f
cc9fd1a737703a811211a63ace2c3c8f91fabc50
575353 F20101208_AABUMP mailhot_d_Page_056.jp2
38ec5fe75e8543ef519b424fabab884d
b52e0357887d043d892fe9ed10fb2a639b2c223c
660589 F20101208_AABUNF mailhot_d_Page_074.jp2
a4d4a0a10e6dad1a6520fdefbf18c9a1
35451f35f3f57bd0f35811bae327af15f9b184ad
500106 F20101208_AABUMQ mailhot_d_Page_057.jp2
ab899dcebca822c2f93301f975002074
1a4b0fda1ab3b8f7218f1cdd9e131c9eb3c61c6c
539780 F20101208_AABUNG mailhot_d_Page_075.jp2
a5af0c17ac552d8cf4b934d45ea2f5e3
29fdbcd8da499a6c62e1ed6cb63942e2873ba6a7
67759 F20101208_AABUMR mailhot_d_Page_058.jp2
236f94399f1bfbb428d3e2d4e3e59dbf
d38cc1b052693a76f902288f313ca1bf6d36602e
111656 F20101208_AABUMS mailhot_d_Page_059.jp2
a569b53af5b11967950a3d38525a7c64
f873de825c5a608a5100fc084f525227cfc563ad
677681 F20101208_AABUNH mailhot_d_Page_077.jp2
56ab924f7779ec86437d623842f6601a
6f7411cd0eac2cd8bf326f791bef06ee103bcb2d
110951 F20101208_AABUMT mailhot_d_Page_062.jp2
7210d73a85b419717022c5eb43289cdc
c3309b8e7c65e3738a3c6a52b5d222f4479baccd
110420 F20101208_AABUNI mailhot_d_Page_078.jp2
3dbb62230e5b80c80216db9de8ef8ff7
7dae416c8521d9be6b36df1094f4c3691ca71b39
126558 F20101208_AABUMU mailhot_d_Page_063.jp2
d740caf57612f04931aa3b25593e8efd
427e8b00bb99b6d606a7b4164b872d253ec2219d
112853 F20101208_AABUNJ mailhot_d_Page_079.jp2
e8081f56c570c9b9d89575433735d9bf
709e218f380c9ecaff68039744617f5ec975e93d
121624 F20101208_AABUMV mailhot_d_Page_064.jp2
2408d35072104ed3c495a6943e1d3977
1d97c2a9605ce49ada052fe111f0f0b6323dd8d1
119434 F20101208_AABUNK mailhot_d_Page_080.jp2
4ff94e4f8258528916664d8e9142bcfd
68e1d016fd428896304f2103ef8beb45f1209b36
123914 F20101208_AABUMW mailhot_d_Page_065.jp2
0839866e2cf655086799476705da8d3e
f8c02bca171343cb62a640072d78ecd485a7503d
40246 F20101208_AABUOA mailhot_d_Page_099.jp2
9b954feb07c457ad73bc2ab260d03368
22717a0adf3185a74bf6ce4ac02442f46c99dfe3
115956 F20101208_AABUNL mailhot_d_Page_081.jp2
3f3b3d70fe94e06103becca3aceb2491
45c3de8f68403e6d39f37d9c72b478bb1cad1a2f
111809 F20101208_AABUMX mailhot_d_Page_066.jp2
07607bacbea45bea3ae7aa1ab65345c8
cd9be30f22ca73cf8bd1c7e7f83e9560c7f3f700
123840 F20101208_AABUOB mailhot_d_Page_100.jp2
33e946ae8da6f16f4e3b60e529ab95c1
58a07f4d956a64c10ebdf7a57e7476f89f2913d7
122075 F20101208_AABUNM mailhot_d_Page_082.jp2
2205f3eafd688ac73f6f98d727e589a3
f3dff90d9c62dc4f1f8790f979aca1f03d71e18a
112907 F20101208_AABUMY mailhot_d_Page_067.jp2
934a80ca0517b861a849d638ea1b239d
ea33adf6235c0864f863abc2dbbd0073adf24259
116230 F20101208_AABUOC mailhot_d_Page_101.jp2
71714d8ec43ff5fc9450449771330bb1
14bba741a0cfd6b6bae80d32677a97de1b07cd7c
117571 F20101208_AABUNN mailhot_d_Page_084.jp2
14eb94c7c32e4578aac0380120abbeb1
5f5dc3b29e7c6675d9c696851b8e47b7652d7048
115728 F20101208_AABUMZ mailhot_d_Page_068.jp2
5c1d95f6549966f9341192bcde159e39
dbab30b653cef28be74604ff39714379d882f024
111305 F20101208_AABUOD mailhot_d_Page_102.jp2
60688090f6f621c903c9f4277cca4243
40e1a24bf5103fbb44f2879b107e87c6d7bbc8d2
119789 F20101208_AABUNO mailhot_d_Page_085.jp2
a69825995a2ca6edd26df2da3f645df7
97dd45c12159e71b41cc83eff1ec6a7b210d6b41
113687 F20101208_AABUOE mailhot_d_Page_103.jp2
fe17e1aaacd971268a4c2ba0fc995c46
b506dcf1cfdd5764b45bb1d5dc71dc98cfa493c1
118274 F20101208_AABUNP mailhot_d_Page_086.jp2
5c23e7545bd53ce5fdd6615940f301f2
b00fe5d8b94d392007faf150a72aaafc0aeca462
117395 F20101208_AABUOF mailhot_d_Page_104.jp2
b84013005f161226c912ab40577c2837
151273e6e0f5f1790fcbcee8868a83b17471946b
118132 F20101208_AABUNQ mailhot_d_Page_087.jp2
a6081dc32db7d9a1d9efd2ce2f1f4027
d391f684082194ef740c1747001e58457a0efad5
13290 F20101208_AABUOG mailhot_d_Page_105.jp2
c43cc86d74ef17043b86b4d784d75706
0623d01d30d878570ad2ecb1c2b25db3df64fb9a
158686 F20101208_AABUNR mailhot_d_Page_090.jp2
930c176a7fdeb3a45280944cdef6afaf
1f11a5c8b618d7000bf06380b4d3fe0167c93cfd
1051953 F20101208_AABUOH mailhot_d_Page_106.jp2
4e8fd3dab6cd8768f29f922b2888e04c
4273c6f6d5f273b048cc23bc81116d19fa3304b1
153385 F20101208_AABUNS mailhot_d_Page_091.jp2
f6e59fc4fe6a768ed43d1b344c0e786b
4c6bf5634e6575cdce5159b3a1a98d1088c2246c
159209 F20101208_AABUNT mailhot_d_Page_092.jp2
ca9581e5f752bd70d803b1649e60fbfa
163e3d54601b442c0ffde69e0824aa86d818b80b
146236 F20101208_AABUOI mailhot_d_Page_107.jp2
8caa01e30a0a15273974ddf9b083282d
e597b1b1c0f40ba3db811a610fe79548ed6c3ed4
73641 F20101208_AABUNU mailhot_d_Page_093.jp2
96f25b5d1448ee080c2b8f1bedeb2bbe
5d13b8a87a12055bb2890227f17505db62e6d9ad
139282 F20101208_AABUOJ mailhot_d_Page_108.jp2
7bdb7011f0ef39488ab511dbd638b1fe
8f59e73f74cc615007c0567bbc961fe4c170132e
76831 F20101208_AABUNV mailhot_d_Page_094.jp2
646398a9ded555937c76bba043c00f89
7e2a553fa4186c2649cf96265609d7eeb440d615
1051965 F20101208_AABUOK mailhot_d_Page_109.jp2
1439037b0aeac52caea11fdee03db156
dc73cc763cb9f056aefdebb48c23f2d478efaff9
535361 F20101208_AABUNW mailhot_d_Page_095.jp2
ba9781686b101c2248ab5102774c773e
b1f770ce36cd02893f865cdfbd36f7731673997c
138786 F20101208_AABUOL mailhot_d_Page_110.jp2
65014e54c3afceb64862a4a3e2747bb1
e8f4652d3976c0dc7cf216963a1b65a4ea709c78
546060 F20101208_AABUNX mailhot_d_Page_096.jp2
eb3992de2183f9ed16baa70193bc7624
150a831639fa29c09752e7bb5520281f5887a72d
F20101208_AABUPA mailhot_d_Page_017.tif
46645cd993b3c7294af34aa3f746791f
392433ba6b296e6d43a2dae82c7ac72a99841aec
136950 F20101208_AABUOM mailhot_d_Page_111.jp2
50ff97dcc0aea40ad5a72e47cb7ac4bc
9c7d27644e268d6edf2084a7ee16feb71958ab7e
515795 F20101208_AABUNY mailhot_d_Page_097.jp2
887c65fffd3f0cdf0e9ae8badf174a3f
a0eb2eba32dbf4113de5232898188f541375b5a9
F20101208_AABUPB mailhot_d_Page_018.tif
b061f3afc3ca7d00353c3c9abe9a3604
2e99ed50a8c39f8979a3bfe0ab122beffb36f0d0
310401 F20101208_AABUON mailhot_d_Page_112.jp2
2d47fa3df163822612b3ce7290217675
e92939e2f44fd6c586d2d49b28dd0c69492e345e
590953 F20101208_AABUNZ mailhot_d_Page_098.jp2
f5476f9b64a0710ea15ae4f69bc49188
727ccb4cec8eef5afb5f577c029c57ad81a9b838
F20101208_AABUPC mailhot_d_Page_019.tif
35a2a54ef2ad3ecb09b136241dd1bd2d
ed32767ec2e248da9982a328af790d717cd1d1b8
37573 F20101208_AABUOO mailhot_d_Page_113.jp2
ac36be9beb2b7b68c6b4e1f9c1ce4aaf
57e2d69ba9647eb8bb441f490f939b351dcdce5e
F20101208_AABUPD mailhot_d_Page_021.tif
edd87e87ea7ff1e1b2044a9b59317e5a
375c7b7b7ef2fa2954b2f802e9357ace91118899
F20101208_AABUOP mailhot_d_Page_001.tif
9d0d2388e2727cd57e7179e5712a6026
fa611980655ac0663c70f59d19eaef088c648067
F20101208_AABUPE mailhot_d_Page_023.tif
8a1b12d9d3d80fea55c72fd5bcf3e2a9
55956216d61b2c1709e7ea038cb3e82e139fe1db
F20101208_AABUOQ mailhot_d_Page_002.tif
f983c795f2423e870607ce8139884091
98a0348d82e87ee30dc4e695d62c1c6008ba4f89
F20101208_AABUPF mailhot_d_Page_024.tif
f65c076ef273a81a91369fd34878beca
8735af0eb4b7f3c1cf349965fc4b57d6fdf8eeea
F20101208_AABUOR mailhot_d_Page_005.tif
6f30a7b6c726e5ddf3728d9f71f1bba4
26e52ee6609401de4e4e52aa815814b52c7ada7e
F20101208_AABUPG mailhot_d_Page_025.tif
774cef19074744f46313b0a11b0ac385
fe74f7cd5ef08d9e8112a20d5129128c50e7696d
F20101208_AABUOS mailhot_d_Page_006.tif
c901c442a04ae93f7b92d28e4f6ea63b
7ce364184b0cf1e05d853a8c302d2abc6278a7bc
F20101208_AABUPH mailhot_d_Page_026.tif
4e5e618e5c8396b76c39bc8d4b439b82
fb18ba9defba1d3b1a9c31b3e8d610a2537c71f5
F20101208_AABUOT mailhot_d_Page_007.tif
88619b2e3da677d49f4d99c924b39a00
8659e1392372faf9f7609e9dd0c7089666e76d65
F20101208_AABUPI mailhot_d_Page_028.tif
fce9c0d579d4c4d65f65e27c3ed61a47
d02c68c7e14241ff562b7d8c0389237a0fc8ad08
F20101208_AABUOU mailhot_d_Page_008.tif
27c62c42957f85fd1431e0667f651068
d36ed7a3efdfb28bb454390c5ac5c90db8f209d1
F20101208_AABUOV mailhot_d_Page_009.tif
b8c60974e07cdc13f9f295e95964e8b5
52e30343e5cbee772a6df6d97e5a1379d039906e
F20101208_AABUPJ mailhot_d_Page_029.tif
eda664ae4e93800a55a946c802b6daf3
e0fbe98c3ad14490f4d31ab1cd1c4e7e63708096
F20101208_AABUOW mailhot_d_Page_010.tif
e91ef0b9b66cfa94f37d555508ba7b3e
bce309ebb1df884faec6418d93834142b31f0721
F20101208_AABUPK mailhot_d_Page_030.tif
676c9a8aa3efce20dc61fcbd41ffce73
67b6cc0900d0886f646ed35595345b1f63899d7c
F20101208_AABUOX mailhot_d_Page_012.tif
a501d8061cdd37c1eb38ebc2e78e8304
8b897878e1716e8f1fd0f41702b5f379a6848040
F20101208_AABUQA mailhot_d_Page_047.tif
2664c68895475dba70c4281e964767a4
e485fe350fdaf3be243ae282ead5d6bf5f2ce896
F20101208_AABUPL mailhot_d_Page_031.tif
3676a0aeed603ff77513a95c18a79cb7
05c10f8e9349f25f64c4d448dbc4a313df376b9a
F20101208_AABUOY mailhot_d_Page_013.tif
ea57f97ce10703b970d74310aef7a820
5ae727ae0c8b85f7aabe7f5c49d863a3a5546561
F20101208_AABUQB mailhot_d_Page_048.tif
7afa76cbe5f4666313b42ee06b2e5a9e
35a05945af55e2f3fdea48c7f9dc0510eef2f55c
F20101208_AABUPM mailhot_d_Page_032.tif
2b2844c4ee90ff0bcd0ffdfb83a56d9f
e5ab2335113c3b041c4cb6b34ee7ceeb5eb31795
F20101208_AABUOZ mailhot_d_Page_014.tif
dffb8e47a16c60f1655d8819c33aafae
48b85e3dc068f4c1bc22b2e3e56277063c63aaab
F20101208_AABUQC mailhot_d_Page_050.tif
1a952ff7922e0858834465b33f6147a0
fcbbf1ade849303e3a17a94d59ef02be56a98252
F20101208_AABUPN mailhot_d_Page_033.tif
bc733f71792d1497f98265da0fa2e370
8418ff16f9144bcb907686e516772061e7f35c6e
F20101208_AABUQD mailhot_d_Page_051.tif
9008677434eb504abeb3b68d4a33f373
332570e080ee2dc1fee4ca84750e7e0eecda35b9
F20101208_AABUPO mailhot_d_Page_034.tif
ac6fb49f21c4392537e4ad2242450484
d8607245ba43ee1d2b538aa1e2f772fe0410a347
F20101208_AABUQE mailhot_d_Page_052.tif
fc8d4b663d37175bbcf73af14683e1c2
8d0559a090ec3188e50d8d76acbd38c9892be620
F20101208_AABUPP mailhot_d_Page_035.tif
6af756ebd587fe642362522bc61d9fb6
3530a9ed642da754056aeb239e5dd1dfed3d0642
F20101208_AABUQF mailhot_d_Page_053.tif
5de5778dbee0672347f4ef46de97da95
4bd2e6a02f1d38f463141010d03dfd238d4ad7aa
F20101208_AABUPQ mailhot_d_Page_036.tif
7a3fb7da96d13f605048c58bd64ec7f5
ec31b00b27044e912657afb01da6097df2387cb9
F20101208_AABUQG mailhot_d_Page_054.tif
e0dfee8c69ebc66e7ae7e243a7c03e85
da5ef471b27443faa73a9d63f80ea5b8f0199af9
F20101208_AABUPR mailhot_d_Page_037.tif
7e9c123de7ac3023ebd8ae77ddbc651d
f94a717dfadfc5defb39b564ae99cfee786734b4
F20101208_AABUQH mailhot_d_Page_055.tif
1af78770b02424d1fe1921ee14127ea6
69221851979cebcf60f631f9f113eff3ae7b8eee
F20101208_AABUPS mailhot_d_Page_039.tif
c476467ea9f9152bed71eb9f423205c6
f1b022f7a7e4a4356199a5c6d8a013b32ebd36bc
F20101208_AABUQI mailhot_d_Page_056.tif
240f6c5ca85291f870246ef1f4c0446e
ec7ea741b04a367248e36d62d3bab1549489b079
F20101208_AABUPT mailhot_d_Page_040.tif
0aaea0b5a6d979d2658853e61de9aed2
ae58d924fea03630881109f755cf71d2f6b7d49d
F20101208_AABUQJ mailhot_d_Page_057.tif
635016c567a7498c28507e5cb08ab3ff
ede2d14aa1f0303f675467ba28f6306949db7b89
F20101208_AABUPU mailhot_d_Page_041.tif
5c09aa1321de242c5b902215b6fea086
f4be48062f54642dbfda568052cd6de4673a68ef
F20101208_AABUPV mailhot_d_Page_042.tif
c1bc6972644c9fb38ff80dea14279361
11d5b3faf13333b7ff4b69723a1533890fffe466
F20101208_AABUQK mailhot_d_Page_058.tif
342ee4e9fcea529072939330160f85e6
f80f970206c6e015b9f392b020c90c532d77ffb5
F20101208_AABUPW mailhot_d_Page_043.tif
7225db83251dc17f1329dd76551a93cf
49d3526a9db3785c4b3b0a9e7d1f6a8329f1325a
F20101208_AABURA mailhot_d_Page_076.tif
d030f1096833ad6d4bbb49545734f521
0254ca6bd6b8b3123c307d74527172feea8eda8f
F20101208_AABUQL mailhot_d_Page_059.tif
2c6d2219bfdc2b4b95355f54be0a5989
50a5e3b6605f07d2f74a6c68e44e06a51b19986c
F20101208_AABUPX mailhot_d_Page_044.tif
f7e5a0e553ad6289d45c8792197fc177
6081877a3247b9d096ca56477ba9372309ef56df
F20101208_AABURB mailhot_d_Page_077.tif
515a0555a11df7bfe3f57bdfa835486f
1034ae37facada7a9a4ebb2d12b156ed635c2928
F20101208_AABUQM mailhot_d_Page_062.tif
74e49ab634483fbf67459a9f77a3a107
454a64a167677a846b1b25ea47e42844efe9fa92
1054428 F20101208_AABUPY mailhot_d_Page_045.tif
c93b201d4c6381029a3d3f9b0cef7819
4e1f2e78369a3ae9ca876083922c22c57fb2a4b9
F20101208_AABURC mailhot_d_Page_078.tif
ed5ee34f619ba77c0f123a1641e0de9c
a7edd8ffc1f705360d9361507fe8ba7b480b6c95
F20101208_AABUQN mailhot_d_Page_063.tif
5be86d10b4f7662a3bef269a95c59239
276155ed1d5349118c2b0ecfdafc1141aeba53bb
F20101208_AABUPZ mailhot_d_Page_046.tif
e6d3c051b54e7950247e74c18a159710
1f346c27666019111f3637037d89ac539e17702b
F20101208_AABUQO mailhot_d_Page_064.tif
d4b1f0cab381b1b80bd32001634849fb
4d9f82feb5098def89b5e02969094acd23bbe455
F20101208_AABURD mailhot_d_Page_079.tif
97d327e36a2097b6ae7e7d4c7cf280ef
4677a3f997ce7122eb98faf91d71506633457734
F20101208_AABUQP mailhot_d_Page_065.tif
35f76f9e645edb81af7fe0e7f5d3e7c4
be35757ff5934407d18f060b3a9d130fd82cb87c
F20101208_AABURE mailhot_d_Page_080.tif
bbda1c8b8f4f894a69fa8a0e4e483809
c0d14b1bdec347483cb714d8f1d75183c5ec369a
F20101208_AABUQQ mailhot_d_Page_066.tif
6bd4093286b2cc4a7b837dd575c439d3
65b3e0da3cf7e55f289fcfc997133ab5fd7c6790
F20101208_AABURF mailhot_d_Page_081.tif
0fc8daf86ae59e3efa6f1399f6d3c448
7c6a1defc26ba691d035a77e8c767fac8d8530ff
F20101208_AABUQR mailhot_d_Page_067.tif
67fc9a89012c6dcf0f343371589aa6c9
357c735b9582022a5b27203007c0b57d83c7b2ac
F20101208_AABURG mailhot_d_Page_082.tif
3f80975af89f48233f32ba1c4f970094
2130387a7606e5de6a195d793f74356521dfa1b6
F20101208_AABUQS mailhot_d_Page_068.tif
3194607a067106080469e4129495bbb0
11970dcd3fcee4a41362c0335dc056b2be0f9dbd
F20101208_AABURH mailhot_d_Page_083.tif
52ec8aed64c89c0834f25ff9a0b4ecff
56c3c3bbdb5dc8668d0442c45dab0c804077a37f
F20101208_AABUQT mailhot_d_Page_069.tif
5815dbb375f789883f5affe389642e33
03ceefa25cee1063c95ad8a6b8853968134b022f
F20101208_AABURI mailhot_d_Page_084.tif
0c37e7c99e2a668a70d158d383557c15
f46b5687ea01f2393f8376a2aa9f5a3d1b5f1427
F20101208_AABUQU mailhot_d_Page_070.tif
3ee98bcc17631d7b91388df46094a96d
9b35a07c1bbba009b045dea3a1a735f71e991073
F20101208_AABURJ mailhot_d_Page_085.tif
7fac9522bde90af9a65852e2e99f42a0
4a4cb27f0559164b7f8242bced9a2175b364e3b7
F20101208_AABUQV mailhot_d_Page_071.tif
08414b52f52ebd690f02fe3cacc7b195
6c763a1cacfbaffc79e3327162d0e20e2f309e7f
F20101208_AABURK mailhot_d_Page_086.tif
5482d2f43fa6e60c66c3f35e7b28f468
6fd941d1a0ab0d1e67b94b405f47d5a8765e27fe
F20101208_AABUQW mailhot_d_Page_072.tif
7f17e3d9304441bf8442e50efc5ee5bf
c0203a0896d04c13d3e68568d464e04f7613e969
F20101208_AABUQX mailhot_d_Page_073.tif
19885fe08f8f05d7d60c60fad1a4b8ef
ba9b97f9880436270386fd32d328422a20ae8d71
F20101208_AABUSA mailhot_d_Page_105.tif
a89b35492c30f3abbcbf1fa4f52bd695
890c21948e2bf7d90de999d4b9f716b8648b3744
F20101208_AABURL mailhot_d_Page_087.tif
8fa8a908bc58a44b7b89aa13dc42081f
d1d7c436c4e2c36923094f5929e821d7b8ed6b72
F20101208_AABUQY mailhot_d_Page_074.tif
72cf70b13994fcf83884d70cc350ce0d
4c0a40ca44ed17d47de73cefba4212eba31304a4
F20101208_AABUSB mailhot_d_Page_106.tif
49d33ba9ba8e5c066b4e337470f88f0b
afe7eb29981c3c4d5cd35ad4592382fded09aa38
F20101208_AABURM mailhot_d_Page_088.tif
d6c745cb330ad7e0dd1ce0fc3a96ceb4
433fd8c9f127da116b0f2fc95a1c3ef52c4b20d1
F20101208_AABUQZ mailhot_d_Page_075.tif
bf8e6dcbc3477138e4c08f8261c1f405
3e656aaab267d140bcd26f22e543e6453b22e619
F20101208_AABUSC mailhot_d_Page_108.tif
7990500682c14973b7b1f04b6d37e6ee
bcd5d95c936b0a25d06817d477b426d71d400e01
F20101208_AABURN mailhot_d_Page_089.tif
9e66987297b8e6c7af17548dae7c0589
24e49b418afc235ce6ed3f0c1ab8229ea5b2ee1e
F20101208_AABUSD mailhot_d_Page_109.tif
d9e19e76665a18d8377b5990815fa857
2bd5522f849554e63b22ec226b4aae2f70f54136
F20101208_AABURO mailhot_d_Page_091.tif
483b1a7029097fb74a9aa91fc19d2697
6cd2bc56a8440640b1df7370c496fd9c79e5f364
F20101208_AABUSE mailhot_d_Page_110.tif
da3f4afe32ec4b4a34ec605602e721dc
17876772d0425a62c675b225a16ec38b6c8146ab
F20101208_AABURP mailhot_d_Page_092.tif
0080eafaf861b6d0c998f6e243c9d3b9
b81f4d4785830431c053f5ec67ee107ec0e68452
F20101208_AABUSF mailhot_d_Page_111.tif
563b2fc43f64222ef094c990c48d4802
1d55031f3356316f69446acd1dc9cc3fd57e4e55
F20101208_AABURQ mailhot_d_Page_093.tif
e3fcb57eee4f2290abe2b3a7186e50be
57388c12d82e8c5632fb9df30fe3723962c2053b
F20101208_AABUSG mailhot_d_Page_112.tif
c2ee4f90ea5f948d073f3b7cc0a341ca
f5f066feb87c36fda5134055acac7b2b35e0b504
F20101208_AABURR mailhot_d_Page_094.tif
fdbd74adf7767e3f8c485e6f26c009b1
b679da4aba4cd83a14d23d4ff258c884fb1c2097
F20101208_AABUSH mailhot_d_Page_113.tif
b741004cfe36900bcb73e650665cfcd2
347186cd25abd04ff1aedb1eb6ea9a339220740f
F20101208_AABURS mailhot_d_Page_095.tif
b29b5661c8eb50f7dc2cb247f3174a72
21a9653b632f8a5bde0d394cb57f03350eaa7529
7346 F20101208_AABUSI mailhot_d_Page_001.pro
fd507690e5c69b46146fd2c4b5249023
0b5a6c515af5d84110f45c5c220b0ad53be4659e
F20101208_AABURT mailhot_d_Page_096.tif
379492360f9e27fd38edd0e0d63e03ab
1769c3b81323042017177aa028b54711c854ba8b
1332 F20101208_AABUSJ mailhot_d_Page_002.pro
c680d494025820b1dbab6ac6c4f68acd
6c78859428d169edca4e0ee54745933fba09902f
F20101208_AABURU mailhot_d_Page_097.tif
c50c390f7849f4cfc015fd7f5cb11f24
cecdd9f1816d3ad5d7edfd72d35e2126ea962c3c
1796 F20101208_AABUSK mailhot_d_Page_003.pro
7b956b64f300457d4590cc05a4cb2bb3
53d48f33f344ce5877fac6e5568c5615993b2eee
F20101208_AABURV mailhot_d_Page_098.tif
c7fa5f1ffc812139654803e201672cbf
fc4466d8db0c3c433031e8b3b647b4070e09daab
98503 F20101208_AABUSL mailhot_d_Page_005.pro
8a1f058b530d74a9fbe7e94b5ce5a825
9b3492b25efb78b6d7e17de200542531b41bff2c
F20101208_AABURW mailhot_d_Page_099.tif
c50e1f97e8a9612dc55742c1eb6c37eb
977ebfcbff3f5ef98de53e29ea9985b258e05d17
53672 F20101208_AABUTA mailhot_d_Page_024.pro
663696072d7572d9a47a9bbaf8ee1d44
89ac6dbff1f88bc4d1201a5a089dc181ffce3722
F20101208_AABURX mailhot_d_Page_100.tif
5329d0a7fcc9bb1b0808691b5fa85dc4
b232546b558d19032aa71434e5387f34a5bb7bc5
50248 F20101208_AABUTB mailhot_d_Page_025.pro
ae56f138c417a69affc923e2e92c7530
3f4ad6a48ce56c509a69de6e675043860750ad51
68005 F20101208_AABUSM mailhot_d_Page_007.pro
fe54ad331459d6373d75f8a615c24060
413c6169a9dc6066ac4545566401ab4ae4ad460e
F20101208_AABURY mailhot_d_Page_102.tif
efc2edb39029319314e509f93f0cb609
86cc0c49c1bf793b4ec0e5c3e0a9feaa3fa373f5
53757 F20101208_AABUTC mailhot_d_Page_026.pro
4dc501dd3f76037f864f878b05a95a32
4005b1edd982492d401ee50670c3c506380768e3
47237 F20101208_AABUSN mailhot_d_Page_008.pro
654ab1783e309f56004e76288a097ff9
d2554b6a046478fbdcb5adebb692bf3f6bbce6d2
F20101208_AABURZ mailhot_d_Page_104.tif
65e7cf759721b76d956607ec09accf33
e787bd663a65e5c0ba49d4912373662ac51405a7
52533 F20101208_AABUTD mailhot_d_Page_027.pro
db7fd2beb43e3f7f99cb0ab7ad62a9ea
57be8d79d4c56212befd763274e12658f7c985e1
40813 F20101208_AABUSO mailhot_d_Page_009.pro
cd9bc99cc1d5dbb009ce5e0b46ea16a7
e1b0b366303fe29673ce1556f4d8def94239de10
56643 F20101208_AABUTE mailhot_d_Page_028.pro
4770235dc364f56e39bb311b2bc4fd5c
77e8b882410bc552f632c8362be0d81383cd2e57
46616 F20101208_AABUSP mailhot_d_Page_010.pro
fc3f33cc1a963cd8900c0eb7d5ea4b33
e2997f3c0d6fd34804a563b7309d227ae5d0ff05
53180 F20101208_AABUTF mailhot_d_Page_029.pro
87395fb962ba2683ada72846b0d58635
4302d66faa766f86d3157deb50bab7180daa012c
18021 F20101208_AABUSQ mailhot_d_Page_011.pro
cfc37d3ff283508c864017a8c2024327
aea53bf93dfe47ab31adcfbf633ff8121b79ef52
52315 F20101208_AABUTG mailhot_d_Page_030.pro
691d1650b80fb27af1b35157f9b95294
8999b8816d0445df00101cfd73a62ce15e04c969
53641 F20101208_AABUSR mailhot_d_Page_012.pro
cca88681ecb871196a66828a2c789018
559f6175baf10e0b37637def33f7b976232e64b8
55540 F20101208_AABUTH mailhot_d_Page_031.pro
0df52afa33f4d9ab745b691c91be3b95
047819befeac08c240204f9dbd29623cdc85ec6e
49100 F20101208_AABUSS mailhot_d_Page_013.pro
be43b96fc5d857812be0d92dc48e6f91
2043c62c13def951c25dd774afb3a9de6309808f
54523 F20101208_AABUTI mailhot_d_Page_032.pro
56abd7f6b504cd760fd78bb34eb60a51
ca913ae440eb8f99151fddb1f1dbce3329c1c234
54380 F20101208_AABUST mailhot_d_Page_014.pro
6a79235fea84552c9975fa84492f34fb
f91b22bf056ea3193e20df02be8e5459e70eb39d
50326 F20101208_AABUTJ mailhot_d_Page_033.pro
129a6bd2ec34a6c5719ebffb22f2d04b
59113b594958b5755dd397be1929417b101eb413
53291 F20101208_AABUSU mailhot_d_Page_015.pro
8737c82bc2dd631013cbb918e687c9df
387f24ca079fb2c2fec385e6c0d3359163e949b2
55646 F20101208_AABUTK mailhot_d_Page_034.pro
3e5bb251aa768a6cda4b2d27fa7f4a86
b60fa9a86003519ca78ef1e6a8a6ee3d2ef1f073
56477 F20101208_AABUSV mailhot_d_Page_016.pro
c84b357bea7452c01d7109f998dd7ef3
9fa5e043ca7f1e9bddf5da77a373415a8f571db0
55076 F20101208_AABUTL mailhot_d_Page_035.pro
a56a47c865f808ffa5976b426273ef61
21438790f00002f1cc90c702a046a2adc8cf0e79
52301 F20101208_AABUSW mailhot_d_Page_017.pro
0e014cd6855b59d241e94e0059aa7baf
4b7cff12a0630472cced9a18dc49ffacad7b9bc6
57414 F20101208_AABUTM mailhot_d_Page_037.pro
bd8210658c83c9ac9c7104e64e34bb42
ae46f203d2f8189b5d7cbcfd16bc0c72eade1c22
52413 F20101208_AABUSX mailhot_d_Page_018.pro
927024a9843af6b36620845640099590
f84e3ddf79cf09ce9bdb75bd872dc193d75429d2
834 F20101208_AABUUA mailhot_d_Page_052.pro
25282ffccee10dfd426b9769c73fa400
52fb8b7608563bedb0f9b6979330d4afd46a25be
54722 F20101208_AABUSY mailhot_d_Page_021.pro
ceaf3277b81ee9077d30887bd56c2be3
47cd61e270c1ddba31663a5b9982f21fc651fd25
10999 F20101208_AABUUB mailhot_d_Page_053.pro
5ee4e872e67e321dc9bbd727cf707471
e9975d9633e3c46d89b84567fef74e9534079dfe
54536 F20101208_AABUTN mailhot_d_Page_039.pro
2f228649732fa53b2e7eb9a792f706c1
47954cd55b2961e91a904f7a8331c16a247bd4a7
54766 F20101208_AABUSZ mailhot_d_Page_022.pro
3804b6ef6ab7b6767a2035ecaa6a1717
ad7d2e76d820bce07af00d1d637b574ff0088283
6649 F20101208_AABUUC mailhot_d_Page_054.pro
593642bb76f757ed333d69d52fe1f996
5ba5d14cc9877062cd4907e6548a9b6af4a821ea
55027 F20101208_AABUTO mailhot_d_Page_040.pro
3d7610a7e6a1bbfb0335c1e356cec8a5
a65534071ecdfd944e55c2079846b3fd62ffff46
39679 F20101208_AABUUD mailhot_d_Page_055.pro
f816208c1c1b8f90af7293ec4f65f555
1a4d8d3a269ca27bbb5a1cbe022dd483e282db16
55035 F20101208_AABUTP mailhot_d_Page_041.pro
f3f259566d3a48760fdb174d5f99e534
d5f0149cb966888adce308e46914cffba5bbaf06
12974 F20101208_AABUUE mailhot_d_Page_056.pro
5c105b5da55f13eeda98c0bb59ada488
1e3f9a4698e292773b6b31b458521e81739dbb55
56572 F20101208_AABUTQ mailhot_d_Page_042.pro
a31925ca2eeb1ec5d2d8febdc53da28b
46f2e500a89057495e0f7e80958fc3a71dc5e3fe
18635 F20101208_AABUUF mailhot_d_Page_057.pro
8e3ed24c1e131c59e219c459e9862403
730505a4e3aff73adf35fed49ec63a088e7c772e
43121 F20101208_AABUTR mailhot_d_Page_043.pro
426fbcb2d6d65406ca795e073d851079
2ff37b20da86b0d7c61135df7878e4a730751045
19140 F20101208_AABVAA mailhot_d_Page_094.QC.jpg
cd349e6f6fbc88ed0c6c29f005c8839f
461f2b0f5590a9670276f3a2a039a42b8aee485e
51721 F20101208_AABUUG mailhot_d_Page_059.pro
273a28ef10222d34425563f9dc1a487c
993692463b41e11d6904324147a7fbd8f477aa59
22600 F20101208_AABUTS mailhot_d_Page_044.pro
331460f9adb7a16586510fb56e49de31
2be20298b73159e3cb2f62fade0a79500b057158
8962 F20101208_AABVAB mailhot_d_Page_022thm.jpg
9a448b3ace642a58aa8fd8456262abf9
cdb4f53d32fcadf04c1d1498b61fae2d96bd61f2
56109 F20101208_AABUUH mailhot_d_Page_060.pro
a303953fbbc741b85d003652c04bb2fd
6d7a8d92828824589e72d80bd6a9b6572e9cc2af
48257 F20101208_AABUTT mailhot_d_Page_045.pro
38389a30f217192b2b2a10a962209999
071acfdb1d114affcc3546134249b27f3d5d1e59
36983 F20101208_AABVAC mailhot_d_Page_014.QC.jpg
ac02a6ff618e4b3c2d22d240f1029b08
1650ee9c2cdb9b980ee82db402377a1f0738e588
59428 F20101208_AABUUI mailhot_d_Page_063.pro
5f69a343d3a7a54a12cc74715708c89a
f11550164a2244859afe8c3a72f97a1ec701fa93
45219 F20101208_AABUTU mailhot_d_Page_046.pro
8c68391bc317fb4368ed62011efd4627
b23dd4d4f53a6386a41e6319b1bee25900d5f868
5735 F20101208_AABVAD mailhot_d_Page_009thm.jpg
911d8808335e440cd985eb3a3480046b
9fdc3bdee6439c5de158a529c5ac3dc7d01571bd
56609 F20101208_AABUUJ mailhot_d_Page_065.pro
246697c7545e67b7449739e58a0eda89
a379c940a28584ec0cebfe71f10b570427d6444d
75576 F20101208_AABUTV mailhot_d_Page_047.pro
1bfe5efc3a399b35c223da522873e1c7
2deafb988c126886403b6203e8501bdd16fb9af5
9075 F20101208_AABVAE mailhot_d_Page_042thm.jpg
de91b2fe39e2edd05587b22a1b264ab1
9708e1252ec0f4dced119b251a8d34dd35ac444e
52695 F20101208_AABUUK mailhot_d_Page_067.pro
1323838b185c4ca3d954c0871153b940
72b454379e924653d063a324dd92f4788b7674d3
22810 F20101208_AABUTW mailhot_d_Page_048.pro
57476902bee286bf7cba442df52cf220
1887fe91bce388cc23a37210474789e6b0a05563
5128 F20101208_AABVAF mailhot_d_Page_052thm.jpg
b91de17adef950e895825227a15f8a0b
97d8667e1ddbe0cf904159f3b7f277f7e763d570
53804 F20101208_AABUUL mailhot_d_Page_068.pro
4c5f9f878a441a601e639cb2095626d2
eba018bd4ae1b084c5a70aadb350f48743f7054a
24514 F20101208_AABUTX mailhot_d_Page_049.pro
0c4dd90189a0380cd1922ddd9876536e
9b259c1be59d0ae788469dfdbc9fcb613bf03847
8987 F20101208_AABVAG mailhot_d_Page_084thm.jpg
7047c656000be355fea1c6956f2f198f
8a5d996942dd74b283dbd1731b5f3632728c81b6
54693 F20101208_AABUVA mailhot_d_Page_087.pro
7d507a6a8e29dfc84bc22886695bc8a3
17998f3beb17ca3d9101bf9011e7660f8bb353c2
54903 F20101208_AABUUM mailhot_d_Page_070.pro
820628c1d097f1670051f272b76ad221
f3ee1ccd1a53ebd2275714176d33209dced13e06
11105 F20101208_AABUTY mailhot_d_Page_050.pro
e1104a073925fb6890ce417d94545b43
61d8c8a8e3484f971707a6c0f21e181fdc754743
9092 F20101208_AABVAH mailhot_d_Page_035thm.jpg
3a666f9223b9ee2c963e79b43155803b
8150b96aeffce0513910b325fb2dd68d02937e04
29892 F20101208_AABUVB mailhot_d_Page_088.pro
38098bb075fa1c738e4207325be67a90
4abf1c0106e2180c57607328c130e6b55f114967
56882 F20101208_AABUUN mailhot_d_Page_071.pro
4ac8c61059dc312f66284a321caa7209
5b9271679bbb87a6eb7c1da709a2606052b9314d
12929 F20101208_AABUTZ mailhot_d_Page_051.pro
bbac27bad22b32b7125118743d73b8a5
b67dea6dfbc7dbc199e00b287519e9392d8d9df9
7563 F20101208_AABVAI mailhot_d_Page_089.QC.jpg
1d8edee30968735a1fa2074ab29be2d1
f866e133e92b67bc7d7667f77af363b9843e349b
1744 F20101208_AABUVC mailhot_d_Page_089.pro
3e590d7c76a4e107cbe39278cd1de67f
68ab9342cb893b38a0173133c40f418d92d9702c
35815 F20101208_AABVAJ mailhot_d_Page_101.QC.jpg
99aab9c58494fac9c4c0af8957b93f53
baf6788cd354877d3c9afc465043922591d45894
93576 F20101208_AABUVD mailhot_d_Page_091.pro
a13a0915c8834d7a17e0445c5a9ca9fa
23293a856d1ed34b203230401b23c2258b1722a6
40324 F20101208_AABUUO mailhot_d_Page_072.pro
61c2ed548ad58018c22f9516bd491964
b458b5fc46c26748257b7ad6f9b465fda454b494
28398 F20101208_AABVAK mailhot_d_Page_093.QC.jpg
361877c1532ae195941c0da66be4bc47
8e4cd8909b65424b83f51e9cd645f1779008119b
90079 F20101208_AABUVE mailhot_d_Page_092.pro
bb64bea04dcb07a7ba722ec5cffd59a3
6fbc5e58f960a1652f78089c695b6d3d38e62ea5
70402 F20101208_AABUUP mailhot_d_Page_073.pro
8822d23d7d2c49a6771f37a7622ade03
7de2e55df9f59956227a78c60bd3a2a644cc86a9
34264 F20101208_AABVAL mailhot_d_Page_027.QC.jpg
e139031fc5e124cff88f3b3dfa5419b9
37ae702ccab523dd5a3edf444129aebb69aad267
52328 F20101208_AABUVF mailhot_d_Page_093.pro
194daae3737d3a8ba901cf014954cc7d
e0d1f85d642a3624ec8baba845d60f5aad4c6704
28609 F20101208_AABUUQ mailhot_d_Page_074.pro
4927758dff7db6bdd0a90a7095db4e4e
6bd060eedab52a20f0241af1e5ca7aeeb9e82bda
8850 F20101208_AABVAM mailhot_d_Page_070thm.jpg
d4b70ffcb03370b09be494fa4015fede
2bd42b2390ed97eefee90477632e51d185015a79
15403 F20101208_AABUVG mailhot_d_Page_096.pro
4efe4239fc87d7bae62fc6346b03f574
40cd0687a8c4d1e54440a1e3fe261c392fef8c9c
7845 F20101208_AABUUR mailhot_d_Page_076.pro
9e650d261b482f4f352998b47339e0b3
f2133282a411bc8e87a83a395590efc20409c77b
35089 F20101208_AABVBA mailhot_d_Page_030.QC.jpg
a0c7767fa7f0d728d904269bfd2f2731
36593795335b32cf1955acd82464b7f2d2d32355



PAGE 1

1 RELATIONSHIP OF FLOWER THRI PS TO HARDLOCK OF COTTON By DANIEL JOSEPH MAILHOT 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 2007

PAGE 2

2 Copyright 2007 by Daniel Joseph Mailhot

PAGE 3

3 To those who establish the f oundation for future advances.

PAGE 4

4 ACKNOWLEDGMENTS My heartfelt thanks go to my advisor, Dr. James Marois, for his s upport of my research and his advice on writing this di ssertation. I also thank the ot her members of my supervisory committee, Dr. Daniel Chellemi, Dr. Raymond Galaherr, and Dr. James Kimbrough, for their advice and patience. I thank Dr. David Wright for sharing his knowledge of cotton. I thank Dr. Joseph Funderburk, Dr. Julie Stavisky and Dr. Dean Paini for sharing their expertise regarding flower thrips. Dr. Kathy La wrence at Auburn University, Dr. John Mueller at Clemson University and Dr. Boyd Padgett at Louisiana State University generously provided cotton flowers to confirm my findings in Florida. I thank my lab colleagues Dr. Tawainga Katsvairo, Dr. Breno Leite, Enoch Osekre and Francis Tsig bey for helpful insights and good conversations. I thank Wayne Branch, Brian Kidd and Pawel Wiat rak for maintaining my research plots. I thank Doug Hatfield, John L. Smith and Henry Mc Crone for their assistance with the weather model. This research was suppor ted by Cotton Incorporated and Ce rexagri. Their willingness to provide funding is sincerely ap preciated. Finally, I thank my family for their unwavering support and encouragement during my years in graduate school.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES................................................................................................................ .........9 ABSTRACT....................................................................................................................... ............10 CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW..............................................................12 Cotton......................................................................................................................... ............12 Flower thrips.................................................................................................................. .........15 Hardlock....................................................................................................................... ..........20 2 THRIPS ON FLOWERS OF COTTON.................................................................................26 Introduction................................................................................................................... ..........26 Materials and Methods.......................................................................................................... .30 Field Plots.................................................................................................................... ....30 Sampling of Thrips..........................................................................................................31 Identification of Thrips....................................................................................................32 Multi-State Species Survey.............................................................................................32 Thrips Accumulation in Flowers.....................................................................................33 Results........................................................................................................................ .............33 Prevalent Species.............................................................................................................33 Similarities Across the Southeast....................................................................................34 Impact of Insecticide Treatment s on Flower-Inhabiting Insects.....................................35 Changes in Thrips Numbers During the Growing Season..............................................36 Association of Thrips and Orius ......................................................................................36 Thrips Accumulation in Flowers.....................................................................................38 Discussion..................................................................................................................... ..........39 3 RELATIONSHIP OF FLOWER THRIPS TO FUSARIUM HARDLOCK..........................59 Introduction................................................................................................................... ..........59 Materials and Methods.......................................................................................................... .62 Isolation of Fusarium from Thrips..................................................................................62 Influence of Thrips and Fusarium on Hardlock Incidence in Greenhouse......................63 Effect of Fungicides and Insecticides on Hardlock.........................................................65 Results........................................................................................................................ .............66 Isolation of Fusarium from Thrips..................................................................................66 Influence of Thrips and Fusarium on Hardlock Incidence in Greenhouse......................66

PAGE 6

6 Effect of Fungicides and Insecticid es on Thrips, Hardlock, and Yield...........................67 Discussion..................................................................................................................... ..........69 4 RELATIONSHIP OF HARDLOCK TO WEAT HER, THRIPS, AND YIELD....................78 Introduction................................................................................................................... ..........78 Materials and Methods.......................................................................................................... .80 Results........................................................................................................................ .............81 Weather Models of Hardlock Severity............................................................................81 Connections between Thrips Hardlock, and Yield.........................................................85 Discussion..................................................................................................................... ..........86 5 SUMMARY AND CONCLUSIONS...................................................................................101 LIST OF REFERENCES.............................................................................................................106 BIOGRAPHICAL SKETCH.......................................................................................................113

PAGE 7

7 LIST OF TABLES Table page 2.1 Mean number of inhabitants per flower across all treatments in Quincy, FL....................44 2.2 Mean number of inhabitants per flower across all treatments in Marianna, FL................45 2.3 Mean number of inhabitants per flower in Winnsboro, LA...............................................45 2.4 Mean number of inhabitants per flower in Fairhope, AL..................................................46 2.5 Mean number of inhabitants per flower in Blackville, SC................................................46 2.6 Mean numbers of flower inhabi tants by treatment in Quincy, FL.....................................47 2.7 Mean number of inhabitants per fl ower by treatment in Marianna, FL.............................47 2.8 Number of males per female by tr eatment for each year of sampling...............................48 2.9 Occurance of adult thrips and Orius in flowers sampled...................................................55 2.10 Interspecific associ ation of thrips and Orius .....................................................................55 2.11 Interspecific covariance of thrips and Orius ......................................................................55 2.12 Correlation of mean weather condit ions to thrips numbers at 1000..................................57 2.13 Relationship of mean temperature fro m 0800 to 1000 and relative humidity from 1900 to 1000 to thrips numbers at 1000.............................................................................58 2.14 Rainfall during the sampling times shown in Figure 2.4...................................................58 3.1 Proportion of flowers containi ng thrips-carried Fusarium.................................................73 3.2 Results of greenhouse hardlock experiments.....................................................................73 3.3 Effect of fungicides and insecticides on thrips, hardlock, and yield in Quincy, FL..........73 3.4 Effect of fungicide and insecticide treatments on thrips, hardlock, and yield in Marianna, FL................................................................................................................... ...73 4.1 Relationship between average temperatur e and relative humidity from 0700 to 1900 on the day of bloom and hardlock severity for that day....................................................90 4.2 Relationship between average temperat ure from 0700 to 1900 on the day of bloom and hardlock severity for that day......................................................................................90

PAGE 8

8 4.3 Relationship between average relativ e humidity from 0700 to 1900 on the day of bloom and hardlock severity for that day..........................................................................91 4.4 Relationship between average temperat ure from 0800 to 1000 on the day of bloom and hardlock severity for that day......................................................................................91 4.5 Relationship between average relativ e humidity from 0800 to 1000 on the day of bloom and hardlock severity for that day..........................................................................92 4.6 Relationship between average temperatur e and relative humidity from 0800 to 1000 on the day of bloom to hardlo ck severity for that day.......................................................92 4.7 Association of temperature and hardlo ck severity by hour for each location....................93 4.8 Relationship between average temperatur e from 0000 to 0600 on the day of bloom to hardlock severity for all days.............................................................................................94 4.9 Relationship between average temperatur e from 0000 to 0600 on the day of bloom to hardlock severity for temperatures between 21 and 25C.................................................97 4.10 Mean and standard deviation of models when predictions are applied to data.................98 4.11 Validation of the model by random samp ling and applying predictions to data...............98 4.12 Relationship of temperature from 0000 to 0600 and estimated thrips numbers to hardlock in Marianna, FL..................................................................................................99 4.13 Relationships between thrips, hardlock and yield on a per plot seasonal basis.............100

PAGE 9

9 LIST OF FIGURES Figure page 2.1 Number of males per female by treatment on each day of sampling.................................49 2.2 Variation in adult Orius (minute pirate bug) numbers over time......................................51 2.3 Variation in thrips numbers over time...............................................................................53 2.4 Increase in thrips numbers during first day of bloom in Quincy, FL................................56 3.1 Thrips numbers and hardlock in cidence by year in Marianna, FL....................................74 3.2 Thrips numbers and hardlock incidence by treatment in Marianna, FL............................75 3.3 Thrips numbers and hardlock for all treatments and years in Quincy, FL........................76 3.4 Thrips numbers and hardlock in Quincy, FL.....................................................................77 4.1 Cotton boll exhibiting sy mptoms of hardlock...................................................................89 4.2 Mean temperature from 0000 to 0600 and hardlock severity............................................95 4.3 Distribution of mean temper atures from 0000 to 06:00 for days in the climate model.....96 4.4 Incidence of hardlock during the gr owing season for days evaluated...............................96 4.5 Days in model with >10 bolls recovered...........................................................................97 4.6 Prediction residuals for days w ith >10 bolls, 21 to 25C (N=49).....................................98

PAGE 10

10 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 RELATIONSHIP OF FLOWER THRI PS TO HARDLOCK OF COTTON By Daniel J. Mailhot May 2007 Chair: James J. Marois Major Department: Plant Pathology Hardlock is a limiting factor for cotton yields in Florida, and appears as a failure of the fiber to expand outward following boll opening. A nnual losses are typica lly 30 to 70% of the crop. Prior research has identified Fusarium verticillioides as a causal agent. It is believed to infect the flowers on the day of bloom, and sust ain itself within the developing boll. It was hypothesized flower thrips may increase the chan ce of infection by carrying the spores into flowers or allowing better access by the pa thogen due to their feeding damage. To determine the extent to which thrips in the field transport F. verticillioides thrips were captured and placed on media to isolate Fusarium Approximately 10% of cotton flowers contained thrips that were carrying Fusarium Frankliniella tritici was the most commonly found thrips species, representing mo re than 99% of individuals. Similar results were noted in samples from Louisiana, Alabama, and South Carolin a. It was also determined that the average relative humidity from 1900 on the day prior to 1000 on the day of sampling was negatively associated with the number of thrips in flowers at 1000. Field trials were performed to test the ability of insecticide and fungi cide applications to reduce thrips numbers and hardlock. Insecticides reduced thrips number and hardlock severity, and sometimes improved yields. The number of th rips in a plot was positively associated with

PAGE 11

11 hardlock severity. Greenhouse experiments were performed to test the ability of Fusarium inoculation, thrips, or thrips exposed to Fusarium to cause hardlock. Thrips exposed to Fusarium resulted in the most severe symptoms. Th e effect of weather on hardlock was also explored. The average temperature between 0000 and 0600 on the day of bloom was negatively associated with hardlock severity in the resul ting bolls. These findings demonstrate hardlock is associated with flower thrips and can be re duced by managing thrips numbers. Hardlock incidence is also influenced by temperature, and this information may allow control measures to be used when they would be most effective.

PAGE 12

12 CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW Cotton Cotton, Gossypium hirsutum L., is an important crop in th e southeastern US. Diseases and insects affecting its production are well documen ted, and it has been a topic of research since the 1800s. In 2003 cotton was produced on 38,000 hectares in Florida. The harvest of 117,000 bales each weighing 218 kg. yielded a market value of $33.49 million. Since 1995, yields have usually ranged from 450-670 kg/ha These lower yields have b een attributed to hardlock, a situation in which cotton fiber fa ils to fluff out of the boll af ter opening. From the early 1980s through the early 1990s, Florida yields were close to 785 kg/ha, and peaked in 1984 at 1105 kg/ha (Anonymous, 2005). Yields in Florida tend to be lower than other areas of the Southeast, resulting in smaller profit margins, and the pot ential for greater gains from new research. Lower yields have occurred despite the re lease of improved varieties. Many of these varieties have been genetical ly modified to produce the Bacillus thuringiensis (Bt) endotoxin and to be resistant to glyphosate (Round Up Ready or RR). The most commonly used Bt toxin in cotton is the Cry1Ac protein, and it is highly effective against lepi dopterans. Bt toxin expression is influenced slightly by parent variety b ackground, but is sufficien t for field conditions (Adamczyk and Sumerford, 2001). In addition to re ducing insect damage to cotton, it has also allowed a substantial reduction in the number of insecticide applica tions during the season (Cattaneo, 2006). Resistance to gl yphosate permits the he rbicide to be used as a broad-spectrum herbicide while cotton seedlings are in the field. This is advantageous, si nce it allows control of broad-leaf weeds which would otherwise be diffi cult and require several other materials in a conventional control program. Absorption of the compound by humans is low, it is not metabolized, and does not accumulate in tissues. It is non-carcinogenic and has no impact on

PAGE 13

13 fertility or reproductive parame ters. Breakdown of the compound in the environment is also fairly rapid, with a half-life of 14 days under typical conditions (Accinelli, 2004). The type of cotton produced in Florida and the southeastern US is known as upland cotton ( G. hirsutum ). This is in contrast to the lo ng-staple Pima and Egyptian cottons ( G. barbadense ) that are sometimes produced in the southwestern US. Prior to 1920, Sea Island cotton ( G. barbadense ) was produced on the Florida peninsul a and coastal islands of Georgia and the Carolinas. It was limited to these locatio ns due to the long growing-season required for its production, and is still produ ced in the Caribbean. The even tual arrival of the boll weevil made its production unprofitable, and the industr y collapsed. However, Sea Island production in the US preceded that of upland cotton, and its high profitability formed the basis for expansion of upland cotton into the interior of the southeast (May and Lege, 1999). The profitability of cotton farm ing varies substantially from year to year. In 2003, the estimated return after all expenses was $236.62 per hectare, in the Southeast. From 1997 to 2004, only 3 of 8 years showed a profit, with to tal losses exceeding tota l profits. Production expenses, yields and prices received vary a nnually. In 2003, total per hectare operating costs were $793.14 and total allocated costs were $449.36. A lint yield of 932 kg/ha at $1.41/kg and a seed yield of 1507 kg/ha. at $0.09/ kg yielded the $236.62/ha quoted a bove. The estimated farm size for the estimates listed above was 535 acres. Fl orida yields are typically lower than the rest of the cotton belt, and reducing the previous quo ted yield to the actual planted yield for 2003 of 597 lbs/ac. results in a loss of approximately $123/ha (National Cott on Council of America, 2005, 2006).

PAGE 14

14 Cotton is typically planted betw een April 20 and June 1. On average, emergence occurs at 7 days after planting (DAP), first square at 39 DAP, and first bloom at 62 DAP. Development is determined in large part by growing degree days, so warmer weather leads to faster maturation. The normal developmental sequence of cott on bolls has been thoroughly documented. Cotton flowers require approximately 35 days to de velop from carpels to anthesis, the stage of opening. The square appears 10-14 days into the process, and pollen mother cells are undergoing meiosis by days 13 and 14. Floral nectaries begin devel oping around day 10. Ovule number, which determines the number of locule s per boll, is thought to be influenced by environmental factors, although varieties also differ (Gipson, 1982). On the day of bloom, indicated by a white flower, polle n germinates on the stigma and the pollen tube begins growing toward the ovary. This growth requires 12 to 30 hours to reach the ovary and accomplish fertilization. By the second day of bloom, the flower has ch anged from white to a red-purple color. The flower usually detaches from the plant on the 3rd or 4th day after bloom. The zygote begins cell division 4 to 5 days after bloom (Pollock and Jensen, 1964). The initiation of flowering is primarily determined by the accumulated growing de gree-days since planting. This is expressed by plants in warmer conditions initiating the first bloom on a lo wer branch (Roussopoulos et al, 1998). Some cotton varieties are photoperiod-sensitiv e, but they are not gr own commercially in the US. Water stress prior to flowering was observed to increase th e subsequent rate of flowering (Guinn, 1979). High night temperatures ( 25C) have been reported to delay the onset of flowering (Mauney, 1966). Epidermal cells on the surface of the developing embryo undergo a period of extreme elongation for approximately 20 days. These elongated cells attain a length of 25 to 35 mm (Quisenberry and Kohel, 1975). Th ey later enter a phase of secondary cell wall

PAGE 15

15 thickening. At maturity the cells dry, leaving a hollow tube of cellulose which is utilized as fiber. The time required for this process is strongly influenced by temperature (Gipson and Joham, 1968). When about 60% of bolls have op ened, a chemical defoliant is applied to the plants to speed development of unopened bolls and prepare the crop for harvest. Plants are defoliated to reduce contaminants in the harveste d fiber, and this is performed 10 to 14 days before harvest. Approximately 155 days are re quired from planting till harvest in Florida (Wright and Brecke, 2002). Cotton flowers provide many food resources for insects. Pollen provides a source of protein for egg development, resulting in increased fecundity (Trichilo an d Leigh, 1988). Nectar is also an important resource fo r attracting pollinator s and predatory insects to cotton plants (Wackers and Bonifay, 2004). In cotton, new flowers are rapidly colonized by thrips. Their role in cotton production and biology has not been investigated. Flower thrips Flower thrips, mainly Frankliniella tritici (eastern flower thrips), but also F. occidentalis (western flower thrips), and F. bispinosa (Florida flower thrips) ra pidly colonize cotton flowers immediately after opening. More thrips accumula te as the day progresses, by the end of which flowers may contain 10-40 thrips. This occu rs in many production ar eas, although the specific species involved may vary. Thrips are a frequent problem on cotton seedli ngs in the US, resulting in distortion of expanding leaves, and small discolored spots. The most common species are Frankliniella occidentalis (western flower thrips), Frankliniella fusca (tobacco thrips) and Thrips tabaci (onion thrips). Thrips control is usually achieved using a granular insecticide at planting. Foliar insecticide applications are used if more than 2 to 3 thrips are present per plant and damage is observed (Sprenkel, 2005). Applications of jasm onic acid to cotton seed lings can reduce thrips

PAGE 16

16 feeding by 80%, although leaf area is also reduced by 28% (Omer et al., 2001). However, this is not used as a management technique. Thrips feed ing is generally not a problem when the plants are more mature. However, severe damage was reported in Turkey (Atakan and Ozgur, 2001a) by Frankliniella intonsa on mature cotton plants. Although mo re than 350 thrips were observed per flower, pollination was not adversely affect ed. Feeding by thrips larvae resulted in boll shedding, although ovipositioning by females in flower parts had a larger impact. Several studies have examined the tendency of fl ower thrips to disperse in search of food resources. In British Columbia, F. occidentalis dispersal was shown to occur when windspeed was less than 15 km/h, although dispersal was mo st likely to occur in the absence of wind (Pearsall, 2002). In Turkey, F. intonsa was shown to follow a similar pattern. Red flowers were found to contain their highest numbers of thrips at 05:30, after which their numbers fell steadily until 11:30. This dispersal coincides with the opening of new white flowers, which were shown to accumulate between 20 and 200 individuals. The authors also suggested red flowers could serve as a refuge from insecticide applications (Atakan and Ozgur, 2001b). Several flower-inhabiting thrips species are present in north Florida. In an 18-month study of 37 wild plant species, 78% of thrips were a dults, and 87% of adults were from the genus Frankliniella The most common species were F. tritici F. bispinosa F. occidentalis and F. fusca The relative contributions of each specie s fluctuated substantially during the study. Cotton flowers were not examined, but during its bloom period of June, July and August, F. tritici and F. bispinosa were the most common species on w ild hosts (Chellemi et al., 1994). The primary influence on Frankliniella populations appeared to be the availability of suitable flowers. F. occidentalis populations increased mo re rapidly than other species, probably due to its wider plant host range. It is non-native, and was not reported in the area until 1981 (Beshear,

PAGE 17

17 1983). This is prior to the peri od of high yields (and presumably low hardlock) from the early 1980s to 1990s, so its arrival wa s probably not involved in the in creased hardlock experienced since that time. Thrips numbers are affected by pr edatory insects and parasites. Orius insidiosus the minute pirate bug, is a predator of flower thrips that can consume 12.5 thrips per day (Tommasini and Nicoli, 1993). If a sufficient number of Orius are present in an area, localized extinction of flower thrips may occur. Orius will remain in the area, feeding on extrafloral nectaries or pollen and preventing the thrips population from reboundi ng. This diet is sufficient to allow development and oviposition by Orius Small hair tufts called domatia are present on the underside of leaves of some plants, and are associated with larger numbers of predatory insects. However, cotton does not produce these structures. A parasitic nematode Thripinema fuscum was shown to infect F. fusca in north Florida. Among F. fusca found on peanut, as many as 51 and 67% of females on certain sampling date s were found to be infected, resulting in sterility. T. fuscum infection of F. tritici and F. occidentalis was far less common, occurring in only 2% of individuals (Funder burk et al., 2002; Stavisky et al ., 2001; Ramachandran et al., 2001). Species and sex differences have been observe d in thrips behavior. On greenhouse pepper plants, movement of F. occidentalis was limited, while F. tritici and F. bispinosa were found to disperse relatively rapidly. Ma les of all three species were shown to be more mobile among plants (Ramachandran et al., 2001). Females of F. occidentalis have been shown to spend more time feeding and produce more feeding-associated scars on petunias than do males (van de Wetering et al., 1998). Th ese differences in behavior resulted in F. tritici and F. bispinosa being less susceptible to predation than F. occidentalis (Ramachandran et al., 2001.

PAGE 18

18 Thrips species also differ in th eir response to insecticides. F. occidentalis populations have been shown to increase following applic ations of acephate and esfenvalerate, while spinosad causes a reduction. However, acepha te and esfenvalerate were highly toxic to F. tritici and F. bispinosa while spinosad was less effective (Rei tz et al., 2003). Ot her studies have shown spinosad to be equally effective agai nst all three species (E ger et al., 1998). The effectiveness of insecticides on thrips numbers can be complicated by its effect on predators. Several insecticides have been evaluated for both lethal and sub-lethal effects on Orius which could affect its reproduction. Sp inosad was found to have no e ffect, lethal or sub-lethal, on Orius populations (Studebaker and Kring, 2000). C yhalothrin resulted in high mortality, but no sub-lethal effect was observed in survivors (Studebaker and Kring, 2000) Acephate was not tested, but is a broad-spectrum organophosphate which affects a larger number of arthropods. The development rate of thrips is dependant on temperature (Lewis, 1973) It can best be modeled by the use of growing degree days ( GDD), which is also used for predicting plant maturity (Toapanta, et al., 1996). This method was later confirmed and refined by Toapanta et al. (2001), demonstrating it is essential to use te mperature measurements from within the plant canopy (where developing thrips are located) rather than above the plants The life history of flower thrips differs slightly by species. Development time of F. tritici is approximately one day shorter than that of F. occidentalis (approximately 8.5 vs. 9.5 days). Oviposition rates and the number of offspring produced during the female lif etime are not statistically different. However, F. tritici does produce a larger portio n of offspring early in its adult life. Despite these advantages, F. tritici does not show a greater rate of populatio n increase in the field (Reitz et al., 2001).

PAGE 19

19 Thrips possess only one mandible which is used for cutting into plant tissue, and a stylet through which food is drawn. This results in open wounds to the plant, which could allow easier penetration of the tissue by pathogens. Females of F. occidentalis have been shown to feed more frequently and intensely than males, resulting in more tissue damage (van de Wetering et al., 1999). Pickett et al. (19 88) reported 68% of adult F. occidentalis found on cotton plants occurred on fruiting structures, wi th most of these occurring in the flower itself. This is consistent with the concept that cotton pollen ma y be preferable to leaves as a food source for flower thrips (Agrawal et al., 1999). In tomat o, amino acid analysis shows phenylalanine content to be associated with thrips num bers (Brodbeck et al, 2001). Farrar and Davis (1991) investig ated the relationship between F. occidentalis and fusarium ear rot of corn. Fusarium ear rot of corn is caused by Fusarium verticillioides and can result in large yield losses in some years in addition to contamination of the crop with mycotoxins. Disease incidence had been connected previous ly to husk tightness and insect damage. Insecticide applications reduced the numbers of F. occidentalis observed and the incidence of disease. They concluded that th rips may be acting as vectors of F. verticillioides or promoting infection as wounding agents by fe eding on the plant tissue. Several observations regarding hardlock suggest flower thrips might be involved. First, there has been an increase in hardlock in re cent years. Although hardlock rates have not historically been measured, this view is held by some cotton observers and is confirmed by the relatively lower yields experien ced from the late 1990s (Nati onal Cotton Council of America, 2005). This coincides with reduced spraying of in secticides during the flowering stage, due to the use of genetically modi fied varieties producing the Bacillus thuringiensis endotoxin. Large numbers of flower thrips often o ccur within the flowers. Relativ ely higher rates of hardlock are

PAGE 20

20 observed along field margins, and thrips numbers sometimes follow this same pattern. However, hardlock was reported to have been one of the major factors taking cott on out of the Southeast along with boll weevils in the last half of the 20th century (Wright et al., 2004). Hardlock Hardlock is characterized as a failure of the cotton fiber to expand outward after boll opening. Instead, it remains in compressed locules, similar to an orange slice, which may have a grey or occasionally faint pink color. These compressed locules are frequently missed or knocked to the ground by mechanical harvesters. The resulting yield lo ss often ranges from 20 to 60%, depending on the year. Until this di ssertation, hardlock had not been formally recognized as a disease, although it could be c onsidered a subset of the boll rot complex. Hardlock is most severe along the gulf coast, possibly due to the region s high temperatures and humidity. It is also associated with rainfall, hi gh nitrogen, plant size and density. Attempts have been made to avoid harvest problems by the us e of ultra-narrow row plantings and harvesting using a stripper, but this has not proven feasible due to higher production costs and lower fiber quality (Wright et al., 2004). Previous research has shown ha rdlock to be associated with F. verticillioides (Marois et al., 2002). Prior to Michai lieds and Morgan (1998), F. verticillioides was referred to as F. moniliforme F. verticillioides commonly occurs in cotton fi elds, and can be found growing saprophytically on crop residue. It has been isolat ed from both seeds and peduncles. It can also sometimes be found on mature cotton fiber in th e field, yielding the pi nk coloration described previously. It is hypothesized that F. verticillioides infects via the flowers, and colonizes the developing boll. Inoculation of flowers with a spore suspension of Fusarium has been shown to result in more hardlock (Marois et al., 2005).

PAGE 21

21 F. verticillioides is capable of surviving for extend ed periods of time on crop debris. Cotten and Munkvold (1998) soaked maize stalk pieces in spore suspensions of F. moniliforme F. proliferatum and F. subglutinans and left pieces in an Iowa field at the surface and several depths in the soil profile under se veral crop rotations. Th e recovery rate for all three species was over 50% during the first 300 days, and was arou nd 10 to 20% after 600 days. Rohrbach and Taniguchi (1984) observed th e rate of infection by F. moniliforme on pineapple during the flowering stage was best predicted by the number of hours per week the temperature was between 21 and 27C. They also noted a significan t negative correlation at 27 to 32C and 32 to 38C. Infections were also associated with rainfall, although a clea r correlation was not demonstrated. Subbarao and Michailides (1995 ) determined the optimal temperature for F. moniliforme infection on pollinator figs is 30C. It also resulted in the shortest incubation and latent periods (approximately 15 and 40 hours, respectively). At 35C, these were increased to 70 to 90 and 90 to 120 hours. These periods were also increased at temp eratures below 30C, but less drastically. Palmateer et al. (2004) found F. moniliforme to be relatively uncommon among Fusarium species isolated from living cotton plant tissue. F. proliferatum was also fairly uncommon. In contrast, Baird and Carling (1998) found Fusarium species to be present on 22 to 38% of dead cotton roots. Although F. verticillioides and F. proliferatum were not listed in the 6 most frequently isolated Fusarium species, 5 other unlisted species were isolated from 1 to 6% of samples. Bolkan et al. (1979) demonstrated F. moniliforme conidia to be short-lived (6 to 13 weeks) in the soil in the absen ce of host tissues. However, in corporating stem and leaf tissue from pineapple into the soil incr eased survival to at least 12 m onths. It should be noted cotton stems, roots, and fiber from the previous year are commonly found in cotton fields.

PAGE 22

22 Other studies have examined F. moniliforme spore release under field conditions. Sanders and Snow (1978) found the amount of airborne sp ores increased for several weeks after first bloom, possibly due to saprophytic growth on shed flowers and other tissue, before finally declining approximately 6 week s later. That trend was not observed in this study, although taking more samples might have showed it to be the case. It was later confirmed (Snow and Sanders, 1979) that shed flowers, bolls and squares were a suitable substrate to produce Fusarium spores. This was observed on 46 to 96% of flowers and 54 to 88% of bolls and squares. Ooka and Kommedahl (1977) observed a similar situation with F. moniliforme spores in corn fields. Spore numbers were lowest while the plant wa s actively growing, and increased as it reached maturity. They also observed wind-blown soil containing F. moniliforme spores which was likely to have traveled 300-400 km. However, Fernando et al. (2000) found more airborne spores of Fusarium graminearum within 1.5 m of an infected wheat field than at 5 m, suggesting local production of inoculum may be most importa nt. Sanders and Snow (1978) found the release of boll-rotti ng pathogen spores (including Fusarium ) to be highest between 18:00 and 06:00. Fernando et al (2000) found spor e release of F. moniliforme to be highest in wheat fields between 16:00 and 08:00. The availabi lity of moisture is important for growth. Torres et al. (2003) examined the role of wa ter activity (relative humidity expressed as a decimal, abbreviated aw) and temperature on germination of F. verticillioides on maize kernels. Of the two temperatures (20 and 30C) and three levels of aw (0.92, 0.95 and 0.98) evaluated, the combination of 30C and 0.98 resulted in th e most rapid hyphal growth. Growth was substantially slower at lower temperature and levels of aw. Fusarium is associated with boll rots in cotton, and the symptoms of hardlock are often lumped with boll rots. Arndt (1950) reported an especially severe year of boll rots in South

PAGE 23

23 Carolina, and a symptom he described as the ti ght-lock condition. The prevalence of tightlock ranged from 1 to 90%, with areas closer to the coast having a higher incidence. Bagga (1968) reported F. monliforme as the second most isolated boll rot species, occurring on 9 to 13% of samples. Fungal pathogens, including Fusarium can be found within cotton bolls by the first several weeks of developm ent (Roncadori, 1969). When i noculated directly into the pericarp of a developing boll, F. moniliforme causes disease on both adjacent carpels as well as the locule (Sparnicht and Roncadori, 1972). The po ssibility of inoculating flowers with spores of a pathogen to cause boll rot has been demonstrat ed (Edgerton, 1912). The incidence of boll rot is higher with a closed canopy, due to increased humidity, although this is sometimes alleviated by a lack of rainfall. Reduced nitrogen ferti lizer rates sometimes resulted in less boll rot, although the canopy characteristics were not affected (Roncadori et al., 1975). Boll rot is also influenced by the quantity of ai rborne inoculum available. At two locations in Louisiana, spore numbers were found to peak approximately 50 days after first bloom, and the first infected bolls were observe d at 60 days. The spore concentr ation was considerably less 10100 m from the cotton, so it is likely spores were originating in the cr op. Spore concentration also varied considerably duri ng the day, with the highest le vels between 1800 and 0600 hr. Among boll-rotting fungi, Fusarium spores were second only to Diplodia gossypina in numbers. No relationship was found linking te mperature, humidity or rainfa ll to spore numbers. It was suggested the increasing levels of spores c ould have been generated by fungal growth on naturally shed flowers, bolls, squares, and leaves (Sanders and Snow, 1978). In the case of boll rot caused by Colletotrichum capsici damage to lint can vary from a tight-locked condition to seve re degradation of fiber (Robert s and Snow, 1984). Susceptibility to boll rot is also influenced by gossypol (a polyphenol in the Malvaceae), production of which

PAGE 24

24 can be increased by mechanical damage of the plan t (Bell, 1967). Alterna tive leaf shape has also been explored as an option for reducing boll rot in cidence. Okra-leaf vari eties have deeply lobed leaves, similar to that of okra. This results in less boll rot, probably due to a less humid plant canopy. These varieties also produce more flower s during a season, in th e range of 150 to 210 per meter of row compared to 100 to 140 in conv entional types. A more open plant canopy also improves the efficiency with which pesticides can be applied to the plants. Okra-leaf varieties produce better yields under a dverse conditions, but lower yi elds under optimal conditions compared to conventional varieties. They have not been commercially successful in the US, although they account for approximately 50% of th e cotton acreage in Au stralia (Heitholt and Meredith, 1998). Marois and Wright (2004) reported a climate model for predicting hardlock incidence. The mean temperature and humidity from 0700 to 1900 on the day of bloom were recorded during 2002, and correlated to hard lock incidence in subsequent bo lls originating from the white flowers of that particular day. The model was based on eight separate dates, and yielded a R2 value of 0.935 and p < 0.005. Since F. verticillioides was shown to be associated with hardlock (Marois et al., 2002), fungicide applications have been evaluated as a control measure. Marois and Wright (2004) showed significant reduc tions in hardlock and increases in yield during the 2002 growing season with applications of thiophanate -methyl. Benomyl applications we re made to cotton under three nitrogen fertilizer regimes, but only reduced hardlock at the highe st nitrogen rate (201 kg/ha). Applications to blooms instead of bolls were shown to be most effective. During the 2004 season, fungicide applications did not significantly affect hardlock or yield (Maroi s et al., 2005). However, significant increases in leaf area index (LAI) and decrease s in leaf disease were noted.

PAGE 25

25 Although significant yield differences were not noted, a positive correlation with LAI and a negative correlation to leaf disease were demonstrated. Other studies have failed to show reductions in hardlock due to fungicide applicati ons. Seebold et al. (2004) examined the use of fungicide applications in five states in the Southeast duri ng the 2003 season. Fungicides did not reduce hardlock, although it improved seed cotton yi eld in Georgia. The nu mber of applications was shown to be more important than the rate or timing of those applications. The regional project was expanded to ten states for the 2004 season (Seebold and Kemerait, 2005). Fungicide applications had little impact on hardlock, regardless of the number of applications. No impact on yield was noted, except in Virg inia. Yields in Louisiana, Fl orida and Georgia may have been lower due to inclement weather, which resulted in a complete loss of the Alabama study. In Florida, Georgia and Tennessee, fungicides increa sed LAI, and in Florida and Georgia this was determined by the number of applications. Maro is et al. (2006) did not find improvements to LAI, hardlock, or yield from fungicide a pplications during the 2005 season in Florida. Hardlock is a major limiting factor for cotton yields in Florida, and control strategies are not available. It is hypothesi zed that flower thrips may be involved in the problem, and that reducing their numbers could reduce the severity of hardlock. Studies were performed from 2003 to 2005 to investigate this possibility. The first objective was to quantify the density and diversity of arthropod species in cotton flowers and the effects of insecticides on their populations. The second objective was to determin e if there was a relationship between thrips species and hardlock. The third objective was to assess if thrips numbers could be managed to reduced hardlock and improve yield.

PAGE 26

26 CHAPTER 2 THRIPS ON FLOWERS OF COTTON Introduction Thrips are small insects, approximately 1mm in length (family Thripidae). They range in color from light tan to dark brown, depending on the species. They possess primitive wings that allow limited flight. Small hairs (setae) o ccur on some areas of the body, and along with antennae are characteristics useful for identifying the species. Mo st thrips of concern in cotton production belong to the genus Frankliniella (subfamily Thripinae). Frankliniella fusca is a pest of cotton seedlings and peanuts. Other species such as Frankliniella occidentalis Frankliniella tritici and Frankliniella bispinosa occur more often on mature vegetation or flowers. Thrips are known vector s of plant viruses. F. occidentais (Sakimura, 1962), F. fusca (Pappu et al., 1998) and F. bispinosa (Avila et al., 2006) are capable of transmitting the tospovirus tomato spotted wilt virus. F. fusca has also been shown to transmit Pantoea ananatis the bacterium responsible for cente r rot of onion (Gitiatis et al., 2003). Thrips can increase the incidence of corn ear rot, caused by Fusarium verticillioides (Farrar and Davis, 1991). Thrips can vector the pathogen and, by feeding, establish in fection sites. Thrips can overwinter in the soil as larvae. F. occidentalis has been shown to survive whol e-body freezing at temperatures as low as -16C if gradually acclimated (McDonald et al., 1997). In milder climates, adults are active during the entire year although the maturation rate declines (Toapanta, 2001). Several flower-inhabiting thrips species are present in north Florida. In an 18-month study of 37 wild plant species, 78% of thrips found were adults, and 87% of these adults were from the genus Frankliniella (Chellemi et al., 1994). The most common species were F. tritici F. bispinosa F. occidentalis and F. fusca The relative contributions of each species fluctuated substantially during the study. Co tton flowers were not examined, but during its bloom period of

PAGE 27

27 June, July and August, F. tritici and F. bispinosa were the most common species on wild hosts. The primary influence on Frankliniella populations appeared to be the availability of suitable flowers. F. occidentalis populations increased more rapidly during the spring than other species, probably due to its wider plant host range. It is non-native, and wa s not reported in the area until 1981 (Beshear, 1983). Thrips numbers are affected by pr edatory insects and parasites. Orius insidiosus (minute pirate bug) is a common predator that can cons ume 12.5 thrips per day (Tommasini and Nicoli, 1993). If a sufficient number of Orius are present in an area, locali zed extinction of flower thrips may occur. Orius will remain in the area, feeding on extrafloral nectaries or pollen and preventing the thrips population from rebounding. This diet is sufficient to allow development and oviposition by Orius A parasitic nematode Thripinema fuscum was shown to infect F. fusca in north Florida. Among F. fusca found on peanut, as many as 51 and 67% of females on certain sampling dates were found to be infected, resulting in sterility. T. fuscum infection of F. tritici and F. occidentalis was far less common, occurring in only 2% of individuals (Funderburk et al., 2002; Stavisky et al., 2001; Ramachandran et al., 2001). Species and sex differences have been observe d in thrips behavior. On greenhouse pepper plants, movement of F. occidentalis was limited, while F. tritici and F. bispinosa were found to disperse relatively rapidly. Male s of all three species were also shown to be more mobile among plants (Ramachandran et al., 2001). Females of F. occidentalis have been shown to spend more time feeding and produce more feeding-associated scars on petunias than do males (van de Wetering et al., 1998). Th ese differences in behavior resulted in F. tritici and F. bispinosa being less susceptible to predation by Orius than is F. occidentalis (Ramachandran et al., 2001.

PAGE 28

28 Thrips species also differ in th eir response to insecticides. F. occidentalis populations have been shown to increase following applic ations of acephate and esfenvalerate, while spinosad causes a reduction. However, acephate and esfenvalerate are highly toxic to F. tritici and F. bispinosa while spinosad is less effective (Reitz et al., 2003). Other studies have shown spinosad to be equally effectiv e against all three sp ecies (Eger et al., 1998). Spinosad is a macrolide that contains two active ingredients, spinosyn A and spinosyn D. They are derived from Saccharopolyspora spinosa and result in hyperexcitation of neurons in the central nervous system resulting in eventual paralysis (Salga do, 1998). The effectiven ess of insecticides on thrips numbers can be complicated by its effect on predators, which probably explains increases in F. occidentalis Studebaker and Kring (2000) evaluate d several insecticides for both lethal and sub-lethal effects on Orius which could affect its reproducti on. Spinosad was found to have no effect, lethal or sub-lethal, on Orius populations. Cyhalothrin resu lted in high mortality, but no sub-lethal effect was observed in survivors. Acephate was not tested, but is a broad-spectrum organophosphate which affects a large number of arthropods. Thrips possess only one mandible which is used for cutting into plant tissue, and a stylet through which food is drawn. This results in open wounds to the plant, which could allow easier penetration of the tissue by pathogens. Females of F. occidentalis have been shown to feed more frequently and intensely than males, resulting in more tissue damage (van de Wetering et al., 1999). Pickett et al. (19 88) reported 68% of adult F. occidentalis found on cotton plants occurred on fruiting structures, wi th most of these occurring in the flower itself. This is consistent with the concept that cotton pollen ma y be preferable to leaves as a food source for some thrips species (Agrawal et al., 1999). Pollen provides a source of protein for egg development, resulting in increased fecundity (Trichilo and Leigh, 1988). In tomato, amino acid

PAGE 29

29 analysis showed phenylalanine content to be asso ciated with thrips numbers (Brodbeck et al, 2001). Nectar is also an important resource for attracting pollinators a nd predatory insects to cotton plants (Wackers and Bonifay, 2004). Thrips are a frequent problem on cotton seedli ngs in the US, resulting in distortion of expanding leaves, and small discolored spots. The most common species are Frankliniella occidentalis (western flower thrips), Frankliniella fusca (tobacco thrips) and Thrips tabaci (onion thrips). Thrips control is usually achieved using a granular insecticide at planting. Foliar insecticide applications are used if more than 2 to 3 thrips are present per plant and damage is observed (Sprenkel, 2005). Applications of jasm onic acid to cotton seed lings can reduce thrips feeding by 80%, although leaf area is also reduced by 28% (Omer et al., 2001). However, this is not used as a management technique. Thrips feed ing is generally not a problem when the plants are more mature. However, severe damage was reported in Turkey (Atakan and Ozgur, 2001a) by Frankliniella intonsa on mature cotton plants. Although mo re than 350 thrips were observed per flower, pollination was not adversely affect ed. Feeding by thrips larvae resulted in boll shedding, although ovipositioning by females in flower parts had a larger impact. In recent years there has been increasing inte rest in hardlock of cotton. Hardlock is a failure of the cotton fiber to expand outward fr om the boll after opening, and it instead remains in compact wedges. Affected locules will re main on the plant or be knocked to the ground during harvest. Fiber quality is not usually affected, but yields can be reduced considerably. Hardlock is associated with the fungus Fusarium verticillioides and is believed to infect through the flower on the day of bloom (Marois et al., 20 02). Most control strategies have focused on the application of fungicides to fl owers and maturing bolls (Marois and Wright, 2004). It has also

PAGE 30

30 been suggested flower thrips coul d be involved in hardlock. If that is the case, reducing their numbers may limit the severity of hardlock. Previous studies have examined the thrips sp ecies associated with cotton plants. However these have been conducted outside the southeastern US, in drier climates, and the thrips species found differ from those in this area. Frankliniella intonsa has been reported as a pest of cotton in Greece (Deligeorgidis et al., 2002), Turkey (Atakan and Ozgur, 2001a). Frankliniella schultzei and F. occidentalis have been reported as pests of co tton flowers and leaves in Brazil (Monteiro, 2001). Other studies within the Sout heast have focused on damage caused to cotton seedlings by thrips feeding. Th e objectives of this study are to describe the insect species found in cotton flowers and determine how they are affected by insecticide applications. Materials and Methods Field Plots Two field studies were performed, approximately 40 miles apart, at branches of the North Florida Research and Education Center in Qu incy and Marianna, Florida. Cultivar DPL 555 Bt/RR was used, and plots were maintained acco rding to the recommendati ons of the University of Florida extension service unless otherwise note d. Acephate (Orthene) and lambda cyhalothrin (Karate) were used when needed to control Nezara viridula (southern green stink bug) and Euschistus servus (brown stink bug). In Quincy, a large fungicide-insecticide study was utilized to evaluate thrips, hardlock, and yield for 2 years. It was a randomized complete block design, with 4 blocks. There were 28 treatments in 2004 and 10 treatments in 2005. Pl ots were 4 rows (0.9 m between rows) by 9 m long. Control and insecticide-treated plots (w ith or without fungicide depending on the year) were sampled for thrips. Other treatments we re varied rates and timings of fungicide applications, and were not sampled. In 2004, th e insecticide treatment consisted of weekly

PAGE 31

31 applications of 0.10 kg a.i. (active ingredient)/ ha of spinosad (Tracer) on Mondays and 0.56kg/ha acephate + 0.04 kg a.i./ha lambda cyhalothrin (Warrior) on Thursdays. In 2005, 0.02 kg a.i./ha Karate (lambda cyhalothrin) was substituted fo r Warrior, and 0.9 kg/ha of thiophanate-methyl (Topsin M) was applied every 2 weeks. The Marianna study examined the effects of insecticides and fungicides on thrips, hardlock and yield for 3 years. The site was part of a Paspalum notatum (bahiagrass) rotation, and the cotton was planted after peanuts each year. The plots were eight rows in width, with 0.9 m between rows, and 18 m in length. Rows were or iented north to south, and at each end a 6 m wide section of peanuts was planted. Peanuts support large numbers of F. fusca and are often planted in proximity to cotton. It was suspected they could influence the species ratio found in cotton flowers. A randomized complete block desi gn was used with 4 blocks and 4 treatments. The experiment included unsprayed control plots and three other treatments which were applied during the bloom period. The in secticide treatment consisted of spinosad at 0.07 kg a.i./ha alternated weekly with acepha te at 0.9 kg a.i./ha. The f ungicide treatment consisted of thiophanate-methyl at 1.1 kg a.i./ha applied week ly. A fourth treatment included a weekly application of both the insecticide and fungicide sprays listed above. Sampling of Thrips Cotton flowers are white on the first day they open, but by even ing the fringes of the petals are often pink. On the second day, the petals have changed to a solid dark pink or red color. In this study, white, first day, flowers were collect ed and placed into individual 60 ml vials containing 70% ethanol, 30% deioni zed water. The flower was placed into the vial with the peduncle located at the opening. This allowed the flower contents to fall into the bottom of the vial. Flowers were sampled on a weekly basis from the two outer rows of each plot, between 11:00 and 13:00. In the Quincy study, 12 flower s were sampled from each of 4 control and 4

PAGE 32

32 insecticide plots. During 2004, the interval between insecticide appl ications and sampling varied. In 2005, insectic ide-only plots were not available, so an insecticide+f ungicide treatment was substituted. The interval between treatmen t sprayings and sampling time was held constant at 2 days. In the Marianna study, 16 flowers we re sampled from each of the 16 plots. The interval between insecticide applications and sampling varied between 2 and 6 days. In both locations, treatments were applied on the first week of bloom, and sampling began later that week. Sampling was discontinued in late August, since blooms after th at time would not result in harvestable bolls. In 2004, a hurricane prevented sampling on the 5th week, and by the following week the plants had stopped flowering. Identification of Thrips Vials containing thrips were ke pt at room temperature until the sample was evaluated. During 2003 and 2004, the liquid from the sample wa s poured into a Petri dish and the flower was then immersed in the dish to dislodge any th rips present. The sample was then examined, and the insects recorded. In 2005, the flower wa s removed and insects were counted while still in the vial. Flowers were period ically dissected to ensure thrips were not remaining within the flowers. Thrips were recorded to species ba sed on overall color, and antennae pigmentation and ornamentation, while sex was determined by pres ence or absence of the ovipositor, and abdomen width and curvature. Other commonly found or easil y identified insects were classified to order or genus. Insects that were rare ly observed were not recorded. The data were analyzed using the SAS GLM procedure, and means were separa ted using Tukeys Studentized Range Test. Multi-State Species Survey To determine if the species found in Quincy a nd Marianna were typical of the Southeast in general, or specific to north Florida, samples we re taken in other states Collaborators collected 50-80 flowers from Louisiana, Alabama and Sout h Carolina and sent them to Quincy, FL for

PAGE 33

33 identification. The Louisiana samples were coll ected at the Macon Ridge Research Station in Winnsboro, LA. The Alabama samples were coll ected at the Auburn University Gulf Coast Research Station located in Fairhope, AL. The South Carolina samples were collected at the Edisto Research and Education Center in Blackv ille, SC. Flowers were sampled for 3 years at each location. Thrips Accumulation in Flowers In 2004, white flowers were sampled at 2-hour intervals from 10:00 to 16:00 or 10:00 to 18:00 in Quincy, Florida. In 2005, this was repeat ed at two other sites within approximately 300 meters of the 2004 site. At each time interval 15 flowers were collected from each site. Samples were processed as described previous ly. Sampling was discontinued on several days due to rainwater from thunderstorms remaining in flowers. Results Prevalent Species Flower thrips species identified from Quincy and Marianna were consistent at both sites and across all years of the study (Tables 2.1 and 2.2). F. tritici (eastern flower thrips) was the most common species, and flower s contained an average of 1 .4 to 4.2 individuals. Females outnumbered males, typically by a 2:1 to 5:1 ratio. F. occidentalis (western flower thrips) was rarely found in 2003 and 2004 (2 to 3 thrips per 1000 cotton flowers), and was not present in 2005. Although slightly more common (3 to 10 thri ps per 1000 cotton flower s), the same pattern of diminishing prevalence was observed in F. bispinosa (Florida flower thrips). These three species are all flower feeders, and likely competitors for resources. F. fusca was observed at low levels (3 to 16 per 1000 flowers) in all years. It is a foliar-feed ing species, and could have been found in the flowers as a result of dispersion in search of suitable host plants. Immature thrips were also present, and their numbe rs ranged from 0.05 to 0.10 per flower.

PAGE 34

34 Orius species are predators of thrips, but can s ubsist on nectar when prey is unavailable. Observed ratios of Orius to thrips ranged from 1:35 to 1:220 depending on location and year (Tables 2.1 and 2.2). Previous research in field pepper (Funderburk et al. 2000) has shown thrips suppression to occur at 1:200, and several days after reaching 1:40 loca lized near-extinction occurred. Although low ratios were often observed, in cotton they were not sufficient to cause further reductions in thrips numbers. Aphis sp (aphids) were observed at levels of 1.2 to 5.3 per flower. They are known for secreting sticky honeydew, and if bolls are open th is can result in sticky cotton. This problem was not observed in the plots, a nd is distinct from hardlock. Members of the order Formicidae (ants) were also observed, with as many as 2.9 per 10 flowers. They were highly aggregated, and flowers containing ants typi cally contained 5 or more. A lthough they were not classified further, at least two species a ppeared to be present. In 2005, members of order Forficulidae (earwigs) were first recorded. Th ey had been observed but not r ecorded previously due to their extremely low numbers. Beetles from order Staphilinidae were also first observed in high numbers (4 to 56 per hundred flowers) in 2005. Similarities Across the Southeast Samples from Louisiana, Alabama, and South Ca rolina were also examined to see if they were similar to those in Florida (T ables 2.3, 2.4, and 2.5). In all locations, F. tritici was most common. Their numbers ranged from 0.42 to 14.8 pe r flower. Louisiana in 2003 contained the highest number of F. occidentalis recorded (0.31 per flower), but they still constituted a minority of flower thrips present. One individual each of both F. bispinosa and T. palmi were found. The presence of F. bispinosa was unexpected since they are most common to peninsular Florida. The sex ratios varied considerably, with the highest being 1 male per 48 females. In a third of the location/years, males were more common than fema les. The large range of thrips numbers and

PAGE 35

35 sex ratios observed may have been influenced by crop management practices. The number of immature thrips also varied consid erably, from 0.01 to 0.96 per flower. Orius was generally uncommon, except in two instances wher e 0.15 and 0.21 per flower were found. Impact of Insecticide Treatments on Flower-Inhabiting Insects The influence of insecticide treatments on fl ower-inhabiting insects was examined (Tables 2.6 and 2.7). Insecticide applications, whether al one or in combination with fungicide, reduced thrips numbers. The reduction in thrips numbers varied depending on location and year. Thrips were reduced by 84 and 92% in Quincy, and 32, 20 and 36% in Marianna. Both males and females were affected, but males experienced larger percentage declines. Fungicide applications did not influence thrips numbers. Immature th rips were reduced by insecticides by 94 and 100% in Quincy. The number of immature thrips wa s not significantly reduced in Marianna. The number of male thrips per female was usually between 0.1 and 0.7 (Table 2.8). In Quincy the proportion of males was reduced in both years, by about 50%. In Marianna, insecticide applications significantly reduced the propor tion of males in 2003, but not in 2004 or 2005. However, combining the data from all three year s showed an overall si gnificant reduction from 0.5 per female in the control to 0.4 in the insecticide-treated pl ots. The proportion of males varied considerably during the season (Fig. 2.1). Orius was adversely affected by insecticides in Quincy (Table 2.6). Adults were reduced by 83% during one year, and immatures by 80 and 100% in both years. The number of Orius observed was very low in all treatments, and considering the sample size, th is result should be viewed with caution. Marianna, insecticides were less harmful to Orius (Table 2.7). Adults were not affected, while the number of immatures in the insecticide plot was reduced by 66% compared to the co ntrol during one of the three years. Aphis sp were not consistently affected by the treatments. There was also little to

PAGE 36

36 no impact on representatives from the order Fo rmicidae (ants). The insecticide treatments appeared to reduce thrips numbers while having a small impact on other flower inhabitants. Changes in Thrips Numbers During the Growing Season The number of thrips observed varied considerably during th e growing season (Fig. 2.3). At later sampling dates in Marianna in 2003 and 2005, thrips numbers were approximately 55 and 170% higher than earlier in the seasons and were associated with a declining number of white flowers available in the field. This may have caused crowding in the remaining flowers. The overall seasonal trend was sim ilar in both of these years, a lthough 2005 had fewer thrips. In 2004, flowering stopped unusually ea rly, preventing continued sampling. This was preceded by a drastic decline in thrips numbers, from approxi mately 4.5 to 0.1 per flower. In Quincy in 2004, the season began with an unusually rapid increase in thrips numbers, followed by an abrupt decline. The control and insec ticide plots declined by 70 and 80% from their mid-season peaks. In 2005, treatment differences were apparent. On the first sa mpling date, the control plots contained 13 thrips per flower, but declined to 4 by the next sampling date and declined for the remainder of the season. The insecticide-treate d plots showed less vari ation between days, but thrips numbers were generally lower during the second half of the season. Association of Thrips and Orius Orius populations also fluctuated during the seasons (Fig. 2.2). In Marianna, control and insecticide plots generally showed similar fluctuations during the growing season for each year. In 2005, there was a rapid increase in the number of Orius present during the last two sampling dates. On the last sampling date of that year, Orius numbers were approximately 400% higher than before their increase (0.40 vs. 0.05 per flowe r). When the fluctuations in thrips thrips numbers (Fig. 2.3) are compared to Orius (Fig. 2.2), it is difficult to discern a clear relationship.

PAGE 37

37 To determine if thrips and Orius were associated, the number of flowers containing either thrips or Orius both species, or neither was calcul ated. The predicted distribution was determined (Table 2.9), and the significance of their association was calculated using a Chisquare test (Table 2.10). The number of flowers containing both thrips and Orius was higher than predicted, and this was signi ficant (p<0.05) in 3 of 5 loca tion-years. Three indexes of species association were used: Ochiai, Dice, and Jaccard. Values can range from 0 (not associated) to 1 (always associ ated), and those computed for this study were between 0.06 and 0.37. The covariation of thrips and Orius was evaluated by calculating their correlation on a per flower and a per plot*day basis for each location year (Table 2.11). On a per flower basis, the correlations were fairly low ( 0.20). By comparing the mean number of thrips and Orius for each plot and day, correlations as high as 0.50 were obtained. It appe ars that thrips and Orius are associated, but not strongly in cotton flowers. In a predatory relationship, a positive association suggests Orius fluctuates in response to thrips numbers. In cont rast, a negative association would result if Orius resulted in a localized de pression in thrips numbers. If both these scenarios occur, then the observed relations hip would appear weak. There we re several instances that may show the Orius population being influenced by thrips num bers. In the 2004 Marianna control plots (section A of Fig. 2.2 and 2.3), the decline in Orius numbers was less dras tic than in thrips at the end of the season, suggesting they were depressing thrips numbers. In the 2003 Marianna insecticide plots (section B of Fig. 2.2 and 2.3) thrips reache d their lowest numbers on the second week, then rebounded for the remainder of th e season. This same pattern occurred in the Orius population, except it was delayed by one week. In the 2004 Quincy insecticide plots, a seasonal peak in thrips numbers occurred one week before the seasonal peak for Orius It appears each species is capable of influenci ng the numbers of the other at certain times.

PAGE 38

38 Thrips Accumulation in Flowers Cotton flower buds are very tightly closed, a nd occasional dissections did not reveal any thrips within them prior to flower opening. The flower opens at approxim ately 09:00, and thrips begin arriving almost immediately (Fig. 2.4). On most days in 2005, there were 2 or fewer thrips found in flowers at the first sa mpling time of 10:00. At site 2 (2006), flowers at the 10:00 sampling time typically had 4 thrips per flower. Site 3 (2006) had slightly more, with 2 to 8 per flower. Their numbers usually increased until 14:00. At 14:00, flowers typically contained 4 to 14 (Site 1, 2005), 10 to 15 (Site 2, 2006), and 14 to 25 (Site 3, 2006) thrips. After 14:00, thrips numbers were generally as likely to decline as to increase further, but tended to remain in a similar range to the numbers at 14:00. When sampling days are compared to each other, there are obvious differences in the initi al number of thrips, rate of increase, and time of maximum numbers. At sites 2 and 3 on 7/28/2006, thrips numbers at 10: 00 were far higher than on any other day sampled that year. This was associated with lower humidity than on other days. There is considerable variation in thrips numbers betw een days, and it was hypothesized this was due to weather conditions. Temperature, relative humid ity, solar radiation, wind speed and rainfall were compared to the number of thrips presen t. Correlations were pe rformed between thrips numbers at 10:00 and temperatur e and relative humidity for each hour from 16:00 of the day prior to the time of sampling. Means were comp uted for temperature and relative humidity for the most highly correlated time periods. These means were then correlated to thrips numbers (Table 12.2). At sites 2 and 3, relative humid ity from 19:00 on the day before sampling to 10:00 of the next day was strongly correlated to the number of thrips present in flowers at 10:00 (-0.87 and -0.85, respectively, p<0.08). Combining data from sites 2 and 3 resulted in a slightly lower, but highly significant correlati on (-0.81, p=0.0046). Temperature from 8:00 to 10:00 was also correlated to thrips numbers in sites 2 a nd 3 (0.54 and 0.69, respectively, p<0.35). Combining

PAGE 39

39 both days improved the relationshi p (0.60, p=0.07). Relative humid ity and temperature were not correlated to thrips numbers at site 1. Sampling on those days was distributed over a 2-week period, while sites 2 and 3 were sampled on consecu tive days. It is possi ble variations in the total thrips population duri ng the course of sampling at site 1 ob scured the role of weather. The sampling dates also showed narrower ranges of temperature (27.6 to 29.4C) and relative humidity (76.0 to 83.3%) in 2005 than in 2006 (26.6 to 30.5C and 66.3 to 84.7%). The narrower weather variable ranges in 2005 may ha ve been insufficient to influence thrips numbers. Regressions were performed co mparing the weather variable s described previously to thrips numbers (Table 2.13). Adjusted R2 values were high for relative humidity and low for temperature. Using both variab les to predict thrips numbers was less useful than relative humidity alone. Relative humidity appears to be the best predictor of thrips numbers at 10:00. Solar radiation was not predictive of thrips numbers, possibly due to its limited ability to penetrate the plant cano py in early morning. Wind speeds were relatively low (0 to 11 km/h) on most days, and did not predict th rips numbers. Thrips numbers after 10:00 were not influenced by the weather variables examined. After 10:00, th e best predictor of thrips numbers was the number observed at the previous sampling time. Rainfall also occurred during the dates sampled (Table 2.14). On one occasion (7/29/2006) rainfall slightly reduced thrips numbers, although this was only temporary. It is likely some thrips left the flowers, while others were sheltered between overlapping cotton petals. Overlapping co tton petals are also a favored position during normal weather conditions, and may harbor 40 to 80% of the thrips present in a flower. Discussion Cotton flowers contain valuable resources for insects, and regardless of what control measures are attempted, some organism is likely to utilize them. Flower inhabitants interact

PAGE 40

40 through competition and predation, and their pop ulations can fluctuate for unknown reasons. Several observations from these st udies warrant further comment. Based on its prevalence on vegetable crops and wild hosts prior to 2003, F. occidentalis was predicted to be a common sp ecies in cotton flowers. Inst ead, it was rare in 2003 and not found in 2005. This decline in F. occidentalis numbers was also observed on other crops. It may be that the ratio of flower thrips speci es found in cotton flowers is closer to an approximation of the species found in the surroundi ng area rather than being a unique ecological niche. If that is the case, fl uctuations may occur in the ratio of thrips species. This could influence predators which feed on thrips. Some studies have suggested Orius feeds preferentially on F. bispinosa and F. occidentalis relative to F. tritici. This is probably due to the increased activity levels noted in F. tritici which requires increased effort for predators (Reitz et al., 2001). Under certain conditions, pr edation may influence the species ratio. As time progresses, the most common species may face gr eater pressures as predator and parasite populations adjust themselves to better exploit it. It is possible this may prevent any one species from comprising a majority indefinitely. Having mostly one thrips species in a field does simplify management. Reitz et al. (2003) determined effects of insecticides varied according to species in pepper production. They f ound spinosad was effective against F. occidentalis but not F. tritici In contrast, esfenvalerate a nd acephate reduced populations of F. tritici and F. bispinosa but resulted in hi gher populations of F. occidentalis This suggests the spinosad component of the insecticide applications in this study may not have been as important. However, the high proportion of F. tritici in thrips populations occurr ed in multiple locations in the Southeast, and was present in plots not spra yed with spinosad. It is unlikely the use of spinosad biased the species ratio in this study.

PAGE 41

41 Insecticide applications generally resulted in fe wer males compared to females in the plots, and this could be relevant from a management pe rspective. Thrips are haplodiploid, with males being haploid and females being diploid. Sexua l reproduction between male and female thrips results in exclusively female of fspring, since sex is determined by ploidy level. Females also reproduce parthenogenically, which results in exclusively male offspring. Decreasing the number of males available for breeding mi ght increase the likeli hood of parthenogenic reproduction, resulting in fewer female offspri ng being produced. The reason for a reduction in males in the insecticide plots is unclear. Male s tend to disperse more readily than females, suggesting they would re-colonize the insecticide plots faster than females. This is the opposite of what was observed. Research on tomato has shown males are more lik ely to occur in the upper portion of the plant canopy (Reitz et al ., 2002). This increases their exposure to insecticide applications compared to females, which are more evenly distributed throughout the canopy. Greater exposure to spraying probably explains the lower proportion of males in insecticide-treated plots. Alternating applications of sp inosad and acephate in Marianna did not appear to harm Orius populations. However, Orius is highly mobile and may ha ve re-colonized plots after treatment. It should be noted that spinosad is among the least harm ful insecticides to Orius (Ramachandran, 2001; Reitz et al., 2001). Insecticide applications did reduce Orius numbers in Quincy, and this may be a result of using lam bda cyhalothrin instead of acephate. In pepper production, Orius has been demonstrated to cause localized near-extinctions of thrips (Reitz et al., 2003). That did not occur in th is study, possibly due to either di fferent thrips species or host plant characteristics. F. tritici is more mobile, and more likely to evade Orius than is F. occidentalis (Reitz et al., 2001). Als o, the near-extinctions observed in pepper do not occur in

PAGE 42

42 tomato production and it is believed characteris tics of the plant itself are involved. However, a positive association between thrips and Orius was observed. In this study, the mean number of thrips dec lined each year in all treatments. In 2004, this was primarily due to a sharp decline in thrips num bers in the fourth week of bloom. The reason for that decline is unclear, but it preceded th e unusually early end of flowering. Nitrogen concentration in plant tissue influences both thrips feeding and flowering. It is possible declining nitrogen availability within the plan t led to reduced phenylal anine content, causing thrips to emigrate in search of better hosts (B rodbeck et al., 2001). Lo w nitrogen availability might also have caused premature termination of flowering. This seems unlikely considering the abruptness with which this sequence occurred. Another possibility is planting date. Due to weather and technical problems, planting dates were slightly later each ye ar, with the onset of flowering ranging from June 20 in 2003 to about Ju ly 5 in 2005. Thrips p opulations are at their peak in the spring, but decline rapidly in May (Chellemi et. al, 1994). It is possible delayed planting of cotton resulted in more of a gap in food availability between the wild spring-hosts and the cotton flowers, reducing th rips numbers. It is also possi ble that various parasites and predators more suited to F. tritici have increased since the time of its presumed replacement of F. occidentalis in early 2003. Parasitic nematodes from the genus Thipinema are common worldwide. T. fuscum infects 40-80% of F. fusca females in north Florida, but fewer than 2% of F. occidentalis and F. tritici (Funderburk et al., 2002; Stavisky et al., 2001; Ramachandran et al., 2001). It likely there are other such parasites which more commonly affect F. tritici The observed increase in thrips numbers per flower during the day is consistant with previous studies. In Turkey, F. intonsa was shown to follow a similar pattern. Red flowers (day after bloom) were found to contain their highest numbers of thrips at 05:30, after which their

PAGE 43

43 numbers fell steadily until 11:30. This disper sal coincides with the opening of new white flowers, which were shown to accumulate between 20 and 200 individuals. The authors also suggested red flowers could serve as a refuge from insecticide applications (Atakan and Ozgur, 2001b). In British Columbia, it was determined F. occidentalis dispersal could occur if windspeed was less than 15 km/h, although it was most common in the absence of wind (Pearsall, 2002). Wind suppressing thrips move ment was not observed in this study, probably because wind speed in this study did not exceed 16 km/h, and was typically below 8 km/h. The association of lower temperature and relative humi dity prior to 10:00 with higher thrips numbers was unexpected. A prior study (Toapanta, 2001) has shown thrips generation lengths to be determined by growing degree days, with warmer temperatures resulting in faster maturity. Higher morning temperatures in this study decr eased thrips movement, perhaps because the temperatures encountered were already above th e optimum. The lower humidity might result in less dew on the plants, allowing thrips to m ove more easily within the plant canopies. Thrips control in cotton flowers had not b een attempted previously, and its potential effectiveness was unknown. Insecticide applic ations reduced total thrips numbers by approximately 20 to 90% depending on the year and location. This study demonstrates insecticide applications are an effective strategy for reducing thrips numbers in cotton flowers. This reduction in thrips numbers was associated with reduction in hard lock severity, and is described in Ch.3.

PAGE 44

44 Table 2.1. Mean number of inhabitants per flow er across all treatments in Quincy, FL. Thrips 2004 2005 Juvenile 0.10 0.03 0.06 0.02 Adult MaleFemaleMale Female Frankliniella tritici 1.10 0.12 2.41 0.18 0.39 0.06 1.88 0.17 F. bispinosa <0.01 0.00 <0.01 0.00 0 0 F. fusca 0 <0.01 0.00 0 0.01 0.00 F. occidentalis <0.01 0.00 0 0 0 Thrips palmi 0 0 0 0 Other insects observed Juvenile Orius sp. 0.03 0.01 0.01 0.01 Adult Orius sp. 0.05 0.01 0.03 0.01 Aphis sp. 3.84 0.36 5.32 0.32 Formicidae 0.25 0.06 0.08 0.03 Forficulidae --0 Staphilinidae --0.04 0.01

PAGE 45

45Table 2.2. Mean number of inhabitants per flow er across all treatments in Marianna, FL. Thrips 2003 2004 2005 Juvenile 0.13 0.02 0.05 0.01 0.02 0.00 Adult Male Female Male Female Male Female Frankliniella tritici 1.16 0.08 3.21 0.08 0.68 0.04 2.10 0.080.40 0.031.04 0.04 F. bispinosa 0.01 0.00 0.07 0.00 <0. 01 0.000.01 0.000 0 F. fusca 0 0.02 0.00 0 0.01 0.000 0.01 0.00 F. occidentalis <0.01 0.00<0.01 0.00<0.01 0.000 0 0 Thrips palmi <0.01 0.00<0.01 0.000 0 0 0 Other insects observed Juvenile Orius sp. 0.06 0.00 0.04 0.00 0.04 0.00 Adult Orius sp. 0.10 0.00 0.05 0.00 0.15 0.00 Aphis sp. 1.20 0.09 1.83 0.16 5.48 0.37 Formicidae 0.10 0.02 0.29 0.03 0.25 0.03 Forficulidae ----<0.01 0.00 Staphilinidae ----0.56 0.03 Table 2.3. Mean number of inhabitant s per flower in Winnsboro, LA. Thrips 2003 2004 2005 Juvenile 0.08 0.03 0.13 0.07 0.01 0.00 Adult Male Female Male Female Male Female Frankliniella tritici 1.30 0.27 1.51 0.24 0.83 0.15 4.30 0.53 0.01 0.01 0.41 0.15 F. bispinosa 0 0 0 0 0 0 F. fusca 0 0 0 0 0 0 F. occidentalis 0.12 0.06 0.19 0.05 0 0 0 0 Thrips palmi 0 0.01 0.01 0 0 0 0 Other insects observed Juvenile Orius sp. 0 0.02 0.02 0 Adult Orius sp. 0.01 0.01 0 0 Aphis sp. 3.02 0.65 18.06 2.61 0.04 0.02 Formicidae 0 0 2.34 0.45 Forficulidae ----0.13 0.05 Staphilinidae ----0.51 0.22

PAGE 46

46Table 2.4. Mean number of inhabitant s per flower in Fairhope, AL. Thrips 2003 2004 2005 Juvenile 0.96 0.26 0.29 0.11 0.62 0.12 Adult Male Female Male Female Male Female Frankliniella tritici 2.96 0.36 1.52 0.21 6.32 0.78 8.50 1.01 1.88 0.25 1.31 0.14 F. bispinosa 0 0.02 0.02 0 0 0 0 F. fusca 0 0 0 0 0 0 F. occidentalis 0.02 0.02 0 0 0 0 0 Thrips palmi 0 0 0 0 0 0 Other insects observed Juvenile Orius sp. 0.14 0.05 0.07 0.03 0.09 0.03 Adult Orius sp. 0 0.21 0.07 0.03 0.02 Aphis sp. 0.40 0.23 0.13 0.07 0.43 0.12 Formicidae 0.20 0.07 0.09 0.05 0.01 0.01 Forficulidae ----0 Staphilinidae ----0 Table 2.5. Mean number of inhabitant s per flower in Blackville, SC. Thrips 2003 2004 2005 Juvenile 0.14 0.06 0.02 0.02 0.18 0.06 Adult Male Female Male Female Male Female Frankliniella tritici 3.00 0.45 1.35 0.20 0.18 0.06 0.98 0.16 1.28 0.23 2.73 0.28 F. bispinosa 0 0 0 0 0 0 F. fusca 0 0.01 0.01 0 0 0 0 F. occidentalis 0 0 0 0 0 0 Thrips palmi 0 0 0 0 0 0 Other insects observed Juvenile Orius sp. 0 0 0.07 0.04 Adult Orius sp. 0 0.07 0.04 0.15 0.04 Aphis sp. 0.83 0.29 0.54 0.16 14.66 2.62 Formicidae 0.18 0.01 0.20 0.09 0.03 0.02 Forficulidae ----0 Staphilinidae ----0.01 0.01

PAGE 47

47 Table 2.6. Mean numbers of flower inha bitants by treatment in Quincy, FL. F. tritici Orius sp. year treatment N male female immature thrips immature adult Aphis sp. Formicidae 2004 C 188 2.05 a 4.03 a 0.19 a 0.05 a 0.06 a 3.43 a 0.51 a 2005 C 288 0.73 b 3.43 a 0.11 a 0.03 a 0.06 a 3.78 a 0.13 b 2004 C 188 2.05 a 4.03 a 0.19 a 0.05 a 0.06 a 3.43 a 0.51 a 2004 I 190 0.15 b 0.81 b 0.01 b 0.01 b 0.04 a 4.25 a 0.00 b 2005 C 288 0.73 a 3.43 a 0.11 a 0.03 a 0.06 a 3.78 b 0.13 a 2005 IF 280 0.04 b 0.28 b 0.00 b 0.00 b 0.01 b 6.91 a 0.03 a Numbers within a column between horizontal lines followed by the same letter are not significantly different according to Tukeys Studentized Range Test Range Test (p 0.05). Treatment C is a control, I is a weekly application of lambda cyhalothrin and spinosad insecticid es (separate days), and IF is a combination of the fungicide and insecticide treatments. Table 2.7. Mean number of inhabitants per flower by treatment in Marianna, FL. F. tritici Orius sp. year treatment N male female Immature thrips immature adult Aphis sp. Formicidae 2003 all 1052 1.16 a 2.72 a 0.13 a 0.04 a 0.11 b 1.41 b 0.09 b 2004 all 1023 0.68 b 2.10 b 0.05 b 0.04 a 0.05 c 1.83 b 0.29 a 2005 all 1190 0.40 c 1.04 c 0.02 b 0.04 a 0.15 a 5.48 a 0.25 a all C 838 0.92 a 2.21 a 0.06 a 0.05 a 0.10 a 2.73 a 0.25 a all F 803 0.86 a 2.05 a 0.09 a 0.04 a 0.13 a 3.21 a 0.25 a all IF 792 0.57 b 1.72 b 0.04 a 0.03 a 0.10 a 3.32 a 0.16 a all I 832 0.56 b 1.68 b 0.06 a 0.03 a 0.09 a 2.85 a 0.20 a 2003 C 278 1.54 a 3.06 a 0.10 a 0.04 a 0.10 a 0.85 c 0.15 a F 247 1.35 a 2.90 a b 0.17 a 0.04 a 0.15 a 1.57 a b 0.08 a IF 244 0.94 b 2.62 a b 0.08 a 0.03 a 0.09 a 2.03 a 0.05 a I 283 0.82 b 2.33 b 0.15 a 0.03 a 0.08 a 1.28 c b 0.10 a 2004 C 256 0.77 a 2.47 a 0.06 a b 0.03 a 0.04 a 1.68 a 0.34 a F 255 0.72 a 2.20 a b 0.09 a 0.03 a 0.08 a 1.59 a 0.20 a IF 256 0.56 a 1.81 b 0.02 b 0.04 a 0.04 a 1.79 a 0.28 a I 256 0.66 a 1.91 b 0.02 b 0.05 a 0.04 a 2.25 a 0.34 a 2005 C 304 0.48 a 1.20 a 0.03 a 0.06 a 0.14 a 5.34 a 0.25 a F 301 0.58 a 1.21 a 0.01 a 0.04 a b 0.15 a 5.93 a 0.44 a IF 292 0.28 b 0.88 b 0.02 a 0.02 b 0.15 a 5.74 a 0.15 a I 293 0.24 b 0.84 b 0.01 a 0.02 b 0.15 a 4.88 a 0.17 a Numbers within a column between horizontal lines followed by the same letter are not significantly different according to Tukeys Studentized Range Test (p 0.05). Treatment C is control, F is thiophanate-methyl fungicide, I is a weekly application of acephate or sp inosad insecticides, and IF is a combination of the fungicide and insecticide treatments.

PAGE 48

48 Table 2.8. Number of males per female by treatment for each year of sampling. Quincy, FL Treatment 2004 2005 C 0.52 a 0.21 a IF 0.10 b I 0.25 b P-value 0.0178 0.1574 Marianna, FL Treatment 2003 2004 2005 all years C 0.63 a 0.29 a 0.45 a b 0.48 a F 0.57 ab 0.31 a 0.50 a 0.47 a IF 0.40 b 0.35 a 0.33 a b 0.36 b I 0.40 b 0.39 a 0.27 b 0.36 b P-value 0.0295 0.4766 0.0402 0.0155 Numbers within a column between horizontal lines followed by the same letter are not significantly different according to Tukeys Studentized Range Test (p 0.05). Treatment C is control, F is thiophanate-methyl fungicide, I is a weekly application of acephate or sp inosad insecticides, and IF is a combination of the fungicide and insecticide treatments.

PAGE 49

49 A0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 7/15/037/24/03**7/31/03**8/7/038/23/03 DateMales/female C F I IF B0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 7/17/047/25/048/1/04**8/9/04DateMales/female C F I IF Figure 2.1. Number of males per female by treatment on each day of sampling. A Marianna, 2003. B Marianna, 2004. C Marianna, 2005. D Quincy, 2004. E Quincy, 2005. Treatment C is control, F is thiopha nate-methyl fungicide, I is a weekly application of acephate or spinosad insect icides, and IF is a combination of the fungicide and insecticide treatments. A doubl e asterisk (**) indi cates the control and insecticide treatments differed significantl y (p< 0.05) according to Duncans multiple range test for that specific date.

PAGE 50

50 C0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 7/20/057/28/058/11/058/25/059/1/05DateMales/female C F I IF D0.0 0.2 0.4 0.6 0.8 1.0 1.2 7/16/047/23/047/30/04**8/6/04 DateMales/female C I E0.0 0.2 0.4 0.6 0.8 1.0 1.2 7/23/057/27/058/10/058/17/05**8/20/058/26/05 DateMales/female C FI Figure 2.1. Continued

PAGE 51

51 A0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 12-Jul22-Jul1-Aug11-Aug21-Aug31-Aug10-Sep DateOrius 2003 2004 2005 B0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 12-Jul22-Jul1-Aug11-Aug21-Aug31-Aug10-Sep DateOrius 2003 2004 2005 Figure 2.2. Variation in adult Orius (minute pirate bug) numbers over time. A Marianna, FL unsprayed control. B Marianna FL insecticide treatment. C Quincy, FL unsprayed control. D insecticide treatment.

PAGE 52

52 C0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 12-Jul22-Jul1-Aug11-Aug21-Aug31-Aug DateOrius 2004 2005 D0 0.02 0.04 0.06 0.08 0.1 0.12 12-Jul22-Jul1-Aug11-Aug21-Aug31-Aug DateOrius 2004 2005 Figure 2.2. Continued

PAGE 53

53 A0 1 2 3 4 5 6 7 8 12-Jul22-Jul1-Aug11-Aug21-Aug31-Aug10-Sep DateAdult Thrips 2003 2004 2005 B0 1 2 3 4 5 6 7 12-Jul22-Jul1-Aug11-Aug21-Aug31-Aug10-SepDateAdult Thrips 2003 2004 2005 Figure 2.3. Variation in thrips numbers over time. A Marianna, unsprayed control. B Marianna, insecticide treatment. C Quincy, unsprayed control. D Quincy, insecticide treatment.

PAGE 54

54 C0 2 4 6 8 10 12 14 16 12-Jul22-Jul1-Aug11-Aug21-Aug31-AugDateAdult Thrips 2004 2005 D0 0.5 1 1.5 2 2.5 3 12-Jul22-Jul1-Aug11-Aug21-Aug31-AugDateAdult thrips 2004 2005 Figure 2.3. Continued

PAGE 55

55 Table 2.9. Occurance of adult thrips and Orius in flowers sampled. thrips + Orius thrips alone Orius alone neither present N Quincy 2004 actual 14 225 4 135 378 predicted 11 228 7 132 378 Quincy 2005 actual 15 247 4 302 568 predicted 9 253 10 296 568 Marianna 2003 actual 89 831 7 125 1052 predicted 84 836 12 120 1052 Marianna 2004 actual 41 670 8 304 1023 predicted 34 677 15 297 1023 Marianna 2005 actual 118 561 31 480 1190 predicted 85 594 64 447 1190 Table 2.10. Interspecific asso ciation of thrips and Orius association p Ochiai Dice Jaccard Quincy 2004 positive <0.25 0.21 0.11 0.06 Quincy 2005 positive <0.005 0.21 0.11 0.06 Marianna 2003 positive <0.25 0.30 0.18 0.10 Marianna 2004 positive <0.05 0.22 0.11 0.06 Marianna 2005 positive <0.005 0.37 0.29 0.17 Table 2.11. Interspecific covariance of thrips and Orius correlation p N grouping Quincy 2004 0.0446 0.3876 378 Quincy 2005 0.2011 <0.0001568 Quincy 0.1341 <0.0001946 Marianna 2003 0.0651 0.0347 1052 per flower Marianna 2004 0.0932 0.0029 1023 Marianna 2005 0.1855 <0.00011190 Marianna 0.0754 <0.00013265 Quincy 2004 0.1905 0.2965 32 Quincy 2005 0.4568 0.0011 48 Quincy 0.3512 0.0014 80 Marianna 2003 0.1785 0.0263 155 per day*plot Marianna 2004 0.1936 0.0285 128 Marianna 2005 0.5067 <0.0001160 Marianna 0.1291 0.0065 443

PAGE 56

56 A0 2 4 6 8 10 12 14 16 18 20 8101214161820time (h)Adult thrips 7/7/2005 7/9/2005 7/12/2005 7/17/2005 7/18/2005 B0 5 10 15 20 25 30 8101214161820time (h)Adult thrips 7/28/2006 7/29/2006 7/30/2006 7/31/2006 8/1/2006 Figure 2.4. Increase in thrips numbers duri ng first day of bloom in Quincy, FL. A Site 1, 2005. B Site 2, 2006. C Site 3, 2006.

PAGE 57

57 C0 5 10 15 20 25 30 35 40 45 8101214161820time (h)Adult thrips 7/28/2006 7/29/2006 7/30/2006 7/31/2006 8/1/2006 Figure 2.4. Continued Table 2.12. Correlation of mean weather c onditions to thrips numbers at 1000. Temperature 0800 to 1000 RH 1900 to 1000 Site Year N Corr p Corr p 1 2005 5 -0.4146 0.4877 0.1221 0.8449 2 2006 5 0.5437 0.3436 -0.8727 0.0535 3 2006 5 0.6906 0.1967 -0.8523 0.0666 2+3 10 0.6001 0.0667 -0.8089 0.0046 1+2+3 15 0.4211 0.1180 -0.6049 0.0169

PAGE 58

58 Table 2.13. Relationship of mean temperatur e from 0800 to 1000 and relative humidity from 1900 to 1000 to thrips numbers at 1000 Site N Equation R2 adj.-R2 p 1 5 thrips = 82.93826 -T*2.88495 0.1719-0.1041 0.4877 2 5 thrips = -26.52871 +T*1.12867 0.29560.0608 0.3436 3 5 thrips = -64.85198 +T*2.54632 0.47690.3026 0.1967 2+3 10 thrips = -45.69035 +T*1.83749 0.36010.2801 0.0667 1+2+3 15 thrips = -39.65781 +T*1.59549 0.17730.114 0.118 1 5 thrips = -8.83217 +RH*0.12930 0.0149-0.3134 0.8449 2 5 thrips = 35.52886 -RH*0.34261 0.76160.6821 0.0535 3 5 thrips = 59.40033 -RH*0.59429 0.72640.6351 0.0666 2+3 10 thrips = 47.46459 -RH*0.46845 0.65430.6111 0.0046 1+2+3 15 thrips = 42.51782 -RH*0.41942 0.366 0.3172 0.0169 1 5 thrips = 74.90710 -T*3.711 56 +RH*0.33889 0.2602-0.4795 0.7398 2 5 thrips = 69.29711 -T*0. 81540 -RH*0.46463 0.81920.6385 0.1808 3 5 thrips = 52.78204 +T*0.15981 -RH*0.57037 0.72710.4541 0.2729 2+3 10 thrips = 61.03958 -T*0. 32780 -RH*0.51750 0.65860.5611 0.0232 1+2+3 15 thrips = 38.22897 +T*0.11002 -RH*0.40594 0.36640.2608 0.0647 Table 2.14. Rainfall during the sampling times shown in Figure 2.4. date hour rain (cm) 7/9/2005 13 1.27 7/9/2005 18 0.71 7/29/2006 13 0.20 7/29/2006 18 1.40 7/30/2006 14 0.15 7/30/2006 15 0.74 7/30/2006 16 0.15

PAGE 59

59 CHAPTER 3 RELATIONSHIP OF FLOWER THRIPS TO FUSARIUM HARDLOCK Introduction Hardlock is expressed as a failure of the locule of fiber to ex pand outward after boll opening. Instead, it remains in compressed locules, which may have a faint pink or orange color. These compressed locules are frequently mi ssed or knocked to the ground by mechanical harvesters. The resulting yield loss often ranges from 20 to 60%, depending on the year. It has not been formally recognized as a disease, althou gh it could be considered a subset of the boll rot complex. Hardlock is most severe along the gulf coast, possibly due to the regions high temperatures and humidity. It is also associated with rainfall high nitrogen, plant size and density. Attempts have been made to avoid harvest problems by the use of ultra-narrow row plantings and harvesting using a stripper, but this has not proven feasible due to higher production costs and lower fiber qua lity (Wright et al., 2004). Previous research has shown ha rdlock to be associated with F. verticillioides (Marois et al., 2002). Prior to Michai lieds and Morgan (1998), F. verticillioides was referred to as F. moniliforme F. verticillioides commonly occurs in cotton fi elds, and can be found growing saprophytically on crop residue. It has been isolat ed from both seeds and peduncles. It can also sometimes be found on mature cotton fiber in the field, yielding the coloration described previously. It is hypothesized that F. verticillioides infects via the flowers, and colonizes the developing boll. Inoculation of fl owers with a spore suspension of Fusarium has been shown to result in more hardlock (Marois et al., 2005). Fusarium is strongly associated with boll rots in cotton, and the symptoms of hardlock are often lumped with boll rots. Arndt (1950) reported an especially severe year of boll rots in South Carolina, and a symptom he described as the tight-lock condition. The prevalence of

PAGE 60

60 tight-lock ranged from 1 to 90% with areas closer to the co ast having a higher incidence. Bagga (1968) reported F. monliforme as the second most isolated boll rot species, occurring on 9 to 13% of samples. F ungal pathogens, including Fusarium can be found within cotton bolls by the first several weeks of development (Roncador i, 1969). When inoculated directly into the pericarp of a developing boll, F. moniliforme causes disease on both adjacent carpels as well as the locule (Sparnicht and Roncadori, 1972). The po ssibility of inoculating flowers with spores of a pathogen to cause boll rot has been demonstrat ed (Edgerton, 1912). The incidence of boll rot is higher with a closed canopy, due to increased humidity, although this is sometimes alleviated by a lack of rainfall. Reduced nitrogen ferti lizer rates sometimes resulted in less boll rot, although the canopy characteristics were not affected (Roncadori et al., 1975). Boll rot is also influenced by the quantity of airborne inoculum av ailable. At two locations in Louisiana, spore numbers were found to peak approximately 50 days after first bloom, and the first infected bolls were observed at 60 days. The spore concentr ation was considerably less 10-100 m from the cotton, so it is likely spores were originating in the crop. Spore concentration also varied considerably during the day, w ith the highest levels between 1800 and 0600 hr. Among bollrotting fungi, Fusarium spp. spores were second only to Diplodia gossypina in numbers. No relationship was found linking temperature, humidity or rainfall to spor e numbers. It was suggested the increasing levels of spores c ould have been generated by fungal growth on naturally shed flowers, bolls, squa res, and leaves (Sanders and S now, 1978). In the case of boll rot caused by Colletotrichum capsici damage to lint can vary from a tight-locked condition to severe degradation of fiber (Roberts and Snow 1984). Susceptibility to boll rot is also influenced by gossypol (a polyphenol in the Ma lvaceae), production of which can be increased by mechanical damage of the plant (Bell, 1967).

PAGE 61

61 Cotton flowers usually open between 09:00 and 09:30, and are rapidly colonized by flower thrips. In north Florida, Frankliniella tritici (eastern flower thrips ) are most common, but F. occidentalis (western flower thrips) and F. bispinosa (Florida flower thrips) are also present (Mailhot, 2006). Their numbers increase as the day progresses, and flowers eventually contain 10 to 40 thrips. Predatory insects such as Orius insidiosus (minute pirate bug) can reduce thrips numbers (Ramachandran et al., 2001), but can some times be harmed by insecticide applications used to control thrips (St udebaker and Kring, 2000). Acephate is highly toxic to F. tritici (Reitz et al., 2003). Spinosad is less toxic to F. tritici but non-toxic to O. insidiosus (Studebaker and Kring, 2000). Thrips possess only one mandible which is used for cutting into plant tissue, and a stylet through which food is drawn. This results in open wounds to the plant, which could allow easier penetration of the tissue by Fusarium spores. Females of F. occidentalis have been shown to feed more frequently and intensel y than males, resulting in more tissue damage (van de Wetering et al., 1999). Farrar and Davis (1991) investigated the relationship between F. occidentalis and Fusarium ear rot of corn. They concluded that thrips may be acting as vectors of F. verticillioides or as wounding agents by feeding on the pl ant tissue. Pickett et al. (1988) reported 68% of adult F. occidentalis found on cotton plants occurred on fruiting structures, with most of these occurring in the flower itself. This is c onsistent with the concept that cotton pollen may be preferable to leaves as a food sour ce for thrips (Agrawal et al., 1999). In order to demonstrate the association of thrips with Fusarium hardlock, a series of studies were performed. The objectives were to determine the prevalence of thrips exposed to Fusarium in the field, whether artificially-exposed th rips would cause hardlock in a controlled setting, and if field applications of insecticide would reduce hardlock.

PAGE 62

62 Materials and Methods Isolation of Fusarium from Thrips White cotton flowers were collected from a crop rotation study in Quincy, FL. In 2004, cultivar DPL 458 BG/RR was used, producing the Bacillus thuringiensis endotoxin and providing resistance to glyphosate herbicide. In 2005, DPL 555 BG/RR was used, providing the same benefits. The plants were maintained acco rding to recommendations of the University of Florida extension service, and tr eated with insecticides when necessary. Fungicides were not applied during the growing season. Flowers were collected from border rows of plots between 10:00 and 11:00. Each flower was placed into a se parate plastic bag, and sealed shut to contain any thrips which were present. The sample bags were then refrigerated for approximately three hours to reduce thrips mobility. All thrips from the particular bag were then placed onto Petri dishes of one-quarter strength ac idified potato dextrose agar. The dish was then covered and sealed with parafilm to prevent any thrips fr om escaping. The thrips were permitted to move around the dish, allowing any spores on them to be spread across the plate. In 2005 the procedure was modified. Thrips were collected from flowers in the field using an aspirator, placed directly onto APDA Petri dishes, and sealed with parafilm. The thrips survived for approximately three days, after which time f ungal colonies were counted, and suspected Fusarium colonies were marked. Cladosporium and Trichoderma were also recorded due to their frequency and distinctive appearances After seven to ten days, suspected Fusarium colonies were re-examined to verify the orig inal count. Between 20 and 100 flowers were sampled on each date, depending on the number of thrips present and weather conditions. The data were analyzed using the SAS GLM procedur e, and means were separated using Duncans multiple range test.

PAGE 63

63 Influence of Thrips and Fusarium on Hardlock Incidence in Greenhouse Cotton was planted and maintained in a gr eenhouse at the North Florida Research and Education Center in Quincy, FL. The temperat ure varied between 25 and 45C and the plants were watered as necessary. A potting mix including peat moss and composted bark was used. Each cotton seedling was treated with 0.17 ml of Admire (imidacloprid) to prevent insect infestations. Approximately 5 g of water-solu ble 15-30-15 fertilizer wa s applied to each plant early in the experiment. Pix (mepiquat chloride ) was applied at a rate of 0.01 ml/plant 1 or 2 times, as needed to reduce plant height. Excessi ve height and boll weight required some plants to be attached to bamboo stakes to remain upri ght. Shortly before flower ing, plants were divided into four groups: a control, thrips-only, thrips exposed to Fusarium and Fusarium -inoculated. The thrips-only treatment consisted of 10 thrips being placed into each open flower. Fusarium verticillioides cultures (2, 5, and 6) isolated from infect ed bolls were transferred to strength acidified potato dextrose agar. These isolates have been shown pr eviously to result in hardlock (Marois and Wright, 2004) if inoc ulated into flowers. One transfer from each was used to produce a mixed quantity of inoculum to better a pproximate that found in the field. Later work analyzing an internal transcribe d spacer (ITS) region of the geno me established that isolates 5 and 6 are actually F. proliferatum (Leite, et al., 2006). After 10 to 14 days, spores were rinsed from the surface using deionized water. A suspension of 90,000-330,000 spores per ml was created, and with refriger ation remained viable for 3 days. Thrips were exposed to inoculum by placing them on a plate of F. verticillioides for approximately one hour, and then 10 of these thrips were transferred into each available flower Unlike flowers produced in a field setting, the petals in the greenhouse expanded further apart, yielding a more open flower. Treatments were spatially separated to prevent thrips from movi ng between treatments. Flowers were tagged with ribbons to indicate the treatment and date. Af ter all bolls on a plant were open, they were

PAGE 64

64 evaluated for hardlock. Locules displaying the characteristic failu re of fiber to expand were deemed to be affected. The numbers of affected and total locules for each boll were recorded. The first study was conducted from Janua ry to May of 2003, using variety DPL 555 BG/RR. Flowering occurred from March to Apri l, and 15 plants were used in each of 4 treatments. Frankliniella occidentalis was raised in cages, using green beans and pollen as a food source. Frankliniella occidentalis frequently occurred prior to 2003, and was suspected to be a common species in cotton flowers. In the Fusarium inoculation treatment, approximately 5 ml of spore suspension was sprayed into each open flower. In the second study, flowering occurred from April-May 2005. The Fusarium inoculation procedure was modified by using a syringe to place approximately 1 ml of inoculum (same concentration as before) onto the stigma. Exposure to water results in lysing of cotton po llen, reducing the chance of successful pollination (Burke, 2002). This modification reduced the fl ower abortion rate and was maintained in subsequent studies. However, insufficient thri ps were produced, resulting in too few treated bolls to evaluate the thrips treatments. The third study, flow ering from July to August, substituted wild-captured F. tritici to more effectively model fi eld conditions. Thrips were captured from the same plot as the Fusarium -isolation study, and sometimes kept in captivity for several days using tomatillo ( Physalis ixocarpa ) fruit as a food source. A second cotton variety, DPL 444 BG/RR, was added and 8 plants per variety were used in each of the 4 treatments. The DPL 444 plants required 1-2 extra applications of mepiquat chlori de to keep them similar in height to DPL 555. DPL 444 also continued flow ering for 4-5 weeks longer than DPL 555. The third study was replicated from October to No vember. The first and second studies were analyzed individually using the SAS GLM procedur e, and means were separated using Duncans multiple range test. The third study consiste d of two replications separated by time.

PAGE 65

65 Effect of Fungicides and Insecticides on Hardlock Two field studies were performed, approximately 40 miles apart, at branches of the North Florida Research and Education Center in Quincy and Marianna, Florida. These were described in chapter 2. Variety DPL 555 Bt/RR was used, and plots were maintained according to the recommendations of the University of Florida ex tension service unless otherwise noted. Orthene (acephate) and Karate (lambda cyha lothrin) were used when needed to control the southern green stink bug ( Nezara viridula ) and the brown stink bug ( Euschistus servus ). In Quincy, a large fungicide-insecticide study was utilized to evaluate thrips, hardlock, and yield for 2 years. It was a randomized complete block design, with 4 blocks. There were 28 treatments in 2004 and 10 treatments in 2005. Pl ots were 4 rows (0.9 m between rows) by 9 m long. Control and insecticide-treated plots (w ith or without fungicide depending on the year) were sampled for thrips. Other treatments we re varied rates and timings of fungicide applications, and were not sampled. In 2004, th e insecticide treatment consisted of weekly applications of 0.10 kg a.i. (active ingredient)/ ha of spinosad (Tracer) on Mondays and 0.56kg/ha acephate + 0.04 kg a.i./ha lambda cyhalothrin (Warrior) on Thursdays. In 2005, 0.02 kg a.i./ha Karate (lambda cyhalothrin) was substituted fo r Warrior, and 0.9 kg/ha of thiophanate-methyl (Topsin M) was applied every 2 weeks. In each of the 8 plots, 12 flowers were collected weekly, and stored for identification. Hardlock severity was assesse d approximately 2 weeks after defoliation. Five plants were se lected at random, and the number of hardlocked a nd total locules per boll was recorded for each plot. The two cente r rows of each plot were harvested with a spindle plot picker. The data were analyzed using the SAS GLM procedure, and means were separated using Duncans multiple range test (SAS, 2003). The Marianna study examined the effects of inse cticides and fungicides on thrips, hardlock and yield for 3 years. The site was part of a Paspalum notatum (bahiagrass) rotation, and the

PAGE 66

66 cotton was planted after peanuts each year. Th e plots were eight rows in width, with 0.9 m between rows, and 18 m in length. Rows were or iented north to south, and at each end a 6 m wide section of peanuts was planted. Peanuts support large numbers of F. fusca and are often planted in proximity to cotton. It was suspected they could influence the species ratio found in cotton flowers. A randomized complete block desi gn was used with 4 blocks and 4 treatments. The experiment included unsprayed control plots and three other treatments which were applied during the bloom period. The in secticide treatment consisted of spinosad at 0.07 kg a.i./ha alternated weekly with acepha te at 0.9 kg a.i./ha. The f ungicide treatment consisted of thiophanate-methyl at 1.1 kg a.i./ha applied week ly. A fourth treatment included a weekly application of both the insecticid e and fungicide sprays listed a bove. Thrips were sampled from the two outer rows of each plot, while yield da ta was obtained from tw o of the inner rows. Flowers were sampled weekly, 8 from each plot, and stored until thrips could be counted. The data were analyzed using the SAS GLM procedur e, and means were separated using Duncans multiple range test. Results Isolation of Fusarium from Thrips Between 7 and 13 percent of flowers contained thrips that were carrying Fusarium (Table 3.1). No significant differences were observed by year. The number of Fusarium isolations also varied by day, with several days contributing most of the isol ations for the season. High inoculum days were randomly scattered thr oughout the season. W eather conditions were compared, but no similarities between hi gh inoculum days were identified. Influence of Thrips and Fusarium on Hardlock Incidence in Greenhouse In 2003, the addition of thrips ( F. occidentalis ) previously exposed to F. verticillioides resulted in the most hardlock (Table 3.2). The Fusarium -inoculation treatment was also

PAGE 67

67 significantly higher than the control. Thrips in the absence of Fusarium did not differ significantly from the control group. In the second study, the Fusarium -inoculated and control groups only differed at the p=0.10 level of significance. In the third study, the results differed slight ly by variety, although similar results were obtained in both replications. In DPL 555, the same pattern was observed as in the first study, despite using F. tritici instead of F. occidentalis Fusarium -exposed thrips resulted in the highest levels of hardlock. Fusarium -inoculation and thrips each increased hardlock compared to the control treatment. In DPL 444, Fusarium -inoculation and thrips each resulted in significantly more ha rdlock than the control group, although Fusarium -exposed thrips did not differ significantly from the control. In many cases, fewer bolls successfully reached maturity when they had been treated with Fusarium -carrying thrips. This was primarily due to more bolls aborting in their first 2 weeks of development, although flowers also aborted slightly more often. On several days during the experiments, flowers were not treated and the resulting bolls were not included in the experiment. However, the bolls which formed fr om those untreated flower s rarely aborted. If hardlock symptoms were a resu lt of shortages in photosynthates, the reduction in boll numbers might have alleviated some of the problem. A lternatively, the aborted bolls could represent a more severe or advanced category of infection, although this is unlikely since F. verticillioides could not be re-isolated from aborted bolls. Effect of Fungicides and Insecticides on Thrips, Hardlock, and Yield In Quincy, the use of insecticide, both al one and with a fungicide significantly reduced thrips numbers (Table 3.3) by 84% in 2005 and 92% in 2005. In 2005, spraying reduced hardlock from 32% to 18%, but was not significa nt in 2004. Yield was not significantly affected in either year.

PAGE 68

68 In Marianna, insecticide provi ded significant reductions in thrips numbers for all years (Table 3.4), while fungicide pr edictably had no impact. The reduction was more modest than experienced in Quincy, ranging from 21 to 36%. In secticide significantly reduced hardlock in all years, ranging from approximately 30 to 50%. F ungicide applications were significant in some years, but not all. Combining applications of insecticide and fungicide did not provide additional reductions in hardlock in any year It should also be noted that the greater percentage reduction in thrips due to spraying in 2005 compared to pr evious years coincided with a greater reduction in hardlock. Yields were low in 2003, and sprayi ng was not beneficial. Yields were higher in 2004, but only the combined sprays resulted in a significant improvement. In 2005, yields were higher than previous years, and insecticide appl ications provided significa nt increases in yield. The relationship between thrips numbers a nd hardlock incidence in Marianna, FL is illustrated in Fig. 3.1. The mean number of thri ps found within flowers of a plot during the entire season is compared to the hardlock incide nce for that same plot. Clear patterns can be distinguished for 2003 and 2005, with R2 values of 0.19 and 0.34. In 2004, removing the outlier with the highest hardlock incidence only increases the R2 value to 0.06, so other factors may have proved more important in e xplaining hardlock incidence. When treatments are examined individually with all three years included (F ig. 3.2), the relationshi ps are more clear. R2 values range from 0.29 to 0.73. Removing one outlier each from the control and insecticide-only plots increased R2 values to 0.54 and 0.77, respectively. When data is combined from all treatments and years (Fig. 3.3), the rela tionship is also apparent. Although Duncans multiple range test did not show consistant differences in hardlock severity in the Quincy study (Table 3.3), a regressi on showed a similar pattern to that observed in

PAGE 69

69 Marianna. Combining all treatmen ts and year resulted in an R2 value of 0.25. The relationship was actually stronger in 2004 than 2005, with R2 values of 0.63 and 0.39. Discussion The rate at which Fusarium was isolated from thrips in flowers was lower than the observed rate of hardlock in the field. Because thrips numbers within flowers increase during the day, sampling in the afternoon coul d have resulted in is olation rates more similar to observed rates of hardlock. Although hardlock incidence declined between 2003 and 2005, there was no change in the frequency with which Fusarium was isolated from thrips in the field. This suggests yearly fluctuations in inoculum quantity may not be an important determinant of the severity of hardlock. Sanders and Snow (1978) found the amount of airborne spores increased for several weeks after first bloom possibly due to saprophytic growth on shed flowers and other tissue, before finally declining approximately 6 w eeks later. That trend was not observed in this study, although taking more samples might have s hown it to be the case. Ooka and Kommedahl (1977) observed a similar situation with F. moniliforme spores in corn fields. Spores numbers were lowest while the plant was actively growing and increased as it reached maturity. They also observed wind-blown soil containing F. moniliforme spores which was likely to have traveled 300-400 km. However, Fernando et al. (2000) found more airborne spores within 1.5 m of an infected wheat field than at 5 m, sugge sting local production of inoculum may be most important. The thripsFusarium isolation study was undertaken with the assumption thrips were either transporting inoculum into the cotton flowers or their feed ing damage to plant tissue made infection more likely to occur. The results show thrips are capable of transporting inoculum, and this appears realistic based on previous studies. Atakan (2001) observe d thrips populations in red flowers were highest at 05:30, after which th ey began dispersing. Sanders and Snow (1978)

PAGE 70

70 found the release of boll-rotti ng pathogen spores (including Fusarium ) to be highest between 18:00 and 06:00. Fernando et al (2000) found spor e release of F. moniliforme to be highest in wheat fields between 16:00 and 08:00. This allows a time period between 05:30 and around 09:30 before thrips enter new fl owers and during which they could come into contact with recently deposited spores from the previous night. This does not preclude the possibility that thrips damage allows more infections, and perhaps both pathways work synergistically. The 3 isolates used in the greenhouse st udy had been identified in fall of 2002 as F. verticillioides based on spore types and shapes, as well as pigmentation in culture. It was not until summer of 2006 that further work revealed isolates 5 and 6 to be F. proliferatum Previous inoculation studies showed no signi ficant differences between the isolates in the ability to cause hardlock (Marois and Wright, 2004; Marois et al., 2005). This suggests at least two Fusarium species are capable of causing symp toms. Palmateer et al. (2004) found F. moniliforme to be relatively uncommon among Fusarium species isolated from living cotton plant tissue. F. proliferatum was also fairly uncommon. In c ontrast, Baird and Carling (1998) found Fusarium species to be present on 22 to 38% of dead cotton roots. Although F. verticillioides and F. proliferatum were not listed in the 6 most frequently isolated Fusarium species, 5 other unlisted species were isolated from 1 to 6% of sa mples. Bolkan et al. (1979) demonstrated F. moniliforme conidia to be short-lived (6 to 13 weeks) in the soil in the absence of host tissues. However, incorporating stem and leaf tissue from pineapple into the soil increased survival to at least 12 months. It should be noted cotton stems, roots, and fiber from the previous year are commonly found in cotton fields. In the greenhouse inoculation studies, hardlo ck symptoms occurred in the control group, although at lower rates than the experimental tr eatments. Flowers were observed closely for

PAGE 71

71 thrips to ensure Fusarium was not being transported to non-inoc ulated flowers. In addition, the air entering this greenhouse passed through an ev aporative cooler, reducing the opportunity for airborne inoculum from outdoors to reach the flow ers. It appears a certain amount of observed hardlock may result from unknown factors. The first and third greenhouse studies also suggest the particular species of thrips may not be im portant in hardlock, since similar results were obtained with both F. occidentalis and F. tritici It also demonstrated cotton varieties do not respond identically to hardlock. However, the tw o varieties also showed large differences in response to growth regulators, flower production, and boll rete ntion, so comparisons between them should not be extrapolated to field performance. The observ ation of more bolls aborting in their first 2 to 3 weeks of development may be related to the higher levels of gossypol observed at that stage in deve lopment (Bell, 1967). Several observations from Figs. 3.1 to 3.4 are worth noting. In Fig. 3.1, the only treatment which is consistently above the trendline is the un sprayed control. Within a season, it is assumed factors other than thrips that might affect hardlock influence all plots equally. This suggests something other than the quantity of thrips is influencing hardlock. It may be that in addition to reducing their numbers, the insectic ide applications reduce activitie s of thrips which contribute to hardlock. In Fig. 3.2, plots we re grouped by treatment across years. This is useful in that it would reduce the possible complicating factor just described. However, because it is unknown why thrips numbers decreased each year of the study, or if that cause also reduced hardlock independently of the thrips numbe rs, the results should be viewed with caution. The same could be said of Fig. 3.3, although it includes more overlap betwee n years. In Fig. 3.4, the 2004 control plots show the largest ra nge of values along the trendl ine, making it most useful for confirming the thrips-hardlock connection. Each of the other 3 groups is tightly clustered, and

PAGE 72

72 these clusters are considerable distances apart. Because of this, any comparison between these groups will yield a good R2 value. While this makes them le ss useful for showing the connection between thrips and hardlock, they are very useful for demonstrating the benefits obtained from a spray program. Applications of insecticide to field plots we re effective at reduci ng thrips numbers, and this effect was observed in both Quincy a nd Marianna. This was expected based on F. tritici being the dominant species and the particular in secticides chosen to control thrips without severely affecting Orius By preserving the natural predator, the mixed results sometimes associated with spraying (Funder burk et al., 2000) were avoided. The connection between thrips numbers and hardlock was clear in Marianna, but only obvious in one of two years in Quincy. This suggests that while thrips appear to be the most signific ant factor, they are not the only determinant of hardlock incidence. Although it can be concluded that reducing thrips numbers within a season reduced the incidence of hardlo ck, it is not clear whet her declining thrips populations led to reduced hardlock or if both were being influen ced by a third factor. Yield can be adversely affected by many factors, and in some cases hardlock may not be a significant contributor to low yields. At the same time, reduc ing thrips numbers, regardless of their starting point, was usually beneficial, suggesting it may be a useful management strategy for controlling hardlock.

PAGE 73

73 Table 3.1. Proportion of flowers cont aining thrips-carried Fusarium. Year Percent Fusarium N 2003 7.0 a 329 2004 13 a 23 2005 6.6 a 273 Numbers followed by the same letter are not significantly different according to Duncans Multiple Range Test (p 0.05). Table 3.2. Results of greenhouse hardlock experiments. 2003 2005 March-April AprilMay Jul-Aug Oct-Nov Percent Hardlock N Percent Hardlock N Percent Hardlock N DP 444 Fusarium+thrips ------37 a b 37 Fusarium ------44 a 51 Thrips ------45 a 58 Control ------27 b 41 DP 555 Fusarium+thrips 66 a 25 ---57 a 26 Fusarium 56 a b 58 20 a 30 37 b 80 Thrips 39 c b 36 ---40 b 85 Control 32 c 50 12 a 67 23 c 76 Numbers in a column followed by the same letter between horizontal lines are not si gnificantly different according to Duncans Multiple Range Test (p 0.05). Table 3.3. Effect of fungicides and insecticides on thrips, hardlock, and yield in Quincy, FL. Treatment thrips hardlock yield 2004 2005 2004 2005 2004 2005 Control 6.10 a 4.18 a 0.48 a 0.32 a 1369 a 1547 a Insecticide +fungicide --0.32 b 0.42 a 0.18 b 1530 a 1590 a Insecticide 0.95 b --0.42 a --1280 a --Numbers followed by the same letter are not significantly different according to Duncans Multiple Range Test (p 0.05). Table 3.4. Effect of fungicide and insecticide treatments on th rips, hardlock, and yield in Marianna, FL. Treatment thrips 2003 2004 2005 Control 4.05 a 3.28 a 1.68 a Fungicide 3.84 a b 2.96 a b 1.79 a Insecticide +fungicide 3.14 c b 2.38 b 1.16 b Insecticide 2.83 c 2.59 b 1.08 b hardlock yield 2003 2004 2005 2003 2004 2005 Control 0.72 a 0.41 a 0.41 a 528 a b 911 b 1276 b Fungicide 0.69 a 0.31 b 0.33 b 533 a 1003 a b 1237 b Insecticide +fungicide 0.52 b 0.28 b 0.23 c 417 b 1178 a 1688 a Insecticide 0.47 b 0.29 b 0.19 c 499 a b 1110 a b 1602 a Numbers followed by the same letter in a column are not significantly different accor ding to Duncans Multiple Range Test (p 0.05).

PAGE 74

74 Figure 3.1. Regression of thrips numbers and hardlock inciden ce by year in Marianna, FL. C = control, unsprayed F = fungicide treatment, weekly application of Topsin M at 1.25 lb/acre I = insecticide treatment, weekly alternating applica tion of Tracer at 2 oz/acre or Orthene at 1 lb/acre IF= insecticide with fungicide treatment simultaneous application of insecticide and fungicide regimes listed above 20030.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 0.001.002.003.004.005.006.00 thripshardlock C F I IF 20040.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.001.002.003.004.005.00 thripshardlock C F I IF 20050.00 0.10 0.20 0.30 0.40 0.50 0.000.501.001.502.002.503.00 thripshardlock C F I IF 2003y = 0.0673x + 0.3466 R2 = 0.1901 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 0.001.002.003.004.005.006.00 thripshardlock 2004y = 0.0212x + 0.2668 R2 = 0.0188 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.001.002.003.004.005.00 thripshardlock 2005y = 0.1258x + 0.1109 R2 = 0.3361 0.00 0.10 0.20 0.30 0.40 0.50 0.000.501.001.502.002.503.00 thripshardlock

PAGE 75

75 Figure 3.2. Regression of thrips nu mbers and hardlock incidence by treatment in Marianna, FL. C = control, unsprayed F = fungicide treatment, weekly application of Topsin M at 1.25 lb/acre I = insecticide treatment, weekly alternating applica tion of Tracer at 2 oz/acre or Orthene at 1 lb/acre IF= insecticide with fungicide treatment simultaneous application of insecticide and fungicide regimes listed above Control 2003-2005y = 0.0688x + 0.2953 R2 = 0.294 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 0.001.002.003.004.005.006.00 thripshardlock Fungicide 2003-2005y = 0.1175x + 0.0969 R2 = 0.5277 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.001.002.003.004.005.006.00 thripshardlock Insecticide 2003-2005y = 0.0841x + 0.1019 R2 = 0.7719 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.001.002.003.004.005.00 thripshardlock Insecticide + Fungicide 2003-2005y = 0.1252x + 0.0518 R2 = 0.7295 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.001.002.003.004.005.00 thripshardlock

PAGE 76

76 All treatments 2003-2005y = 0.1064x + 0.1202 R2 = 0.5125 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 0.001.002.003.004.005.006.00 thripshardlock Figure 3.3. Regression of thrips nu mbers and hardlock for all trea tments and years in Quincy, FL.

PAGE 77

77 Figure 3.4. Regression of thrips numb ers and hardlock in Quincy, FL. 2004 C = control, unsprayed 2004 I = insecticide treatment, 2.9 oz/acre of Tracer on Mondays and 8 oz/acre Orthene + 5 oz/acre Warrior on Thursdays 2005 C = control, unsprayed 2005 IF = insecticide with fungicide treatment, 2.9 oz/acre of Tracer on Mondays, 8 oz/acre Orthene + 2 oz/acre Karate on Thursdays and 1.0 lb/acre of Topsin M every 2 weeks All treatments 2004-2005y = 0.0259x + 0.2557 R2 = 0.2549 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.002.004.006.008.0010.00 thripshardlock 2004y = 0.016x + 0.3971 R2 = 0.6306 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.002.004.006.008.0010.00 thripshardlock 2005y = 0.0136x + 0.1768 R2 = 0.3927 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.001.002.003.004.005.006.00 thripshardlock All treatments 2004-20050.00 0.10 0.20 0.30 0.40 0.50 0.60 0.002.004.006.008.0010.00 thripshardlock 2004 C 2004 I 2005 C 2005 IF Control 2004-2005y = 0.0501x + 0.106 R2 = 0.4655 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.002.004.006.008.0010.00 thripshardlock Insecticide/insecticide-fungicide 2004-2005y = 0.2894x + 0.1126 R2 = 0.6906 0.00 0.10 0.20 0.30 0.40 0.50 0.000.501.001.50 thripshardlock

PAGE 78

78 CHAPTER 4 RELATIONSHIP OF HARDLOCK TO WEATHER, THRIPS, AND YIELD Introduction Hardlock of cotton ( Gossypium hirsutum L.) occurs when the fiber does not fluff out as the boll opens at maturity. Mature locules look lik e wedges of an orange when broken apart (Fig. 4.1). Although the quality of the cotton fiber may not be severely affected, conventional spindle harvesting equipment is not able to capture th e fiber and bring it into the harvester as the hardlocked cotton is knocked from the plant and falls to the ground or is strung out of the boll giving the appearance of poor harvesting procedures Attempts to scrap the field by running the picker a second time to get the hardlocked cotton of ten results in little lint increase, more trash and lower lint quality. The severity of hardlock in cotton has been associated with high nitrogen, high plant density, high temperature and humidity, insect damage, and seed rot. Because bolls affected by hardlock are not harvested by conventional picker s, yield losses of over 50% have occurred in states in the Southeast. In 2002, hardlock was shown to be associated with Fusarium verticillioides (Marois et al., 2002). Before a publication by Mi chailieds and Morgan (1998), F. verticillioides was referred to as Fusarium moniliforme. F. verticillioides occurs frequently in cotton fields, growing saprophytically on crop residue. It has been isolated from seeds, peduncles, and mature fiber. It is hypothesized that F. verticillioides infects the flowers on the da y of bloom and enters the immature boll before flower dehiscence. Flower inoculation us ing a suspension of F. verticillioides micro and macroconida has been shown to increase hardlo ck severity (Marois et al., 2005). Hardlock can be differentiated from the tradit ional boll rots. Boll rots result from pathogen damage where the carpel turns brown or black an d never opens or after the bolls have opened,

PAGE 79

79 microorganisms destroy the cotton fibers. Bats on (2001) identified 15 species of fungi and bacteria associated with boll rots but he does not describe hardlo ck independently of boll rot. Boll rots occur during wet weather when the cott on boll or fiber is colonized by a number of microbes, although only a few fungi are responsible for the majority of infections (Pinkard and Chilton, 1966). These include Alternaria gossypina (Thuem.) Hopkins, Curvularia spp., Diplodia gossypina Cke., Helminthosporium gossypii Tucker, Fusarium spp., and Phomopsis spp. (Palmateer et al., 2003: Pinkard and Chilton, 1966). Sanders and Snow (1978) found a correlation between the numbers of airborne spores of these fungi and th e incidence of boll rot caused by them. They also proposed that the like ly source of the fungi was not from infected bolls but from spores being produced on shed pl ant parts, as there was a correlation between airborne spore numbers and the normal seasonal sh edding of flowers, bolls, squares, and leaves (Sanders and Snow, 1977). Flower thrips rapidly colonize cotton flower s on the morning they open, typically between 0900 and 0930. The most common species in north Florida is Frankliniella tritici (eastern flower thrips). F. occidentalis (western flower thrips) and F. bispinosa (Florida flower thrips) are also present (Chapter 2). Thrips continue entering th e flowers for much of the day, typically reaching a peak of 10 to 40 per flower between 1400 and 1800. Thrips numbers can be reduced by predatory insects such as Orius insidiosus the minute pirate bug (Ramachandran et al., 2001), but these predators can sometimes be harmed by ins ecticide applications used to control thrips (Studebaker and Kring, 2000). A connection between thrips numb ers and temperature has been demonstrated. The average temperature betwee n 0800 and 1000 is negatively correlated with the number of thrips in the flowers at 1000 (Ch.2).

PAGE 80

80 In 2004, a model was proposed linking temperature and relative humidity on the day of bloom to hardlock incidence in those same bolls after opening (Marois et al., 2004). Temperature and relative humidity between 0700 and 1900 were shown to be positively associated with hardlock severity. This model was consistent with the geographic hypothesis of coastal areas having more hardlock due to their clim ate. It was also intuitive since F. verticillioides germinates and grows more readily in high temperature, high moisture conditions. Subbarao and Michailides (1995) determined the optimal temperature for F. verticillioides (moniliforme) infection on pollinator figs is 30C. The temperatures observed in the model were usually below 30C. The objectives of this stu dy were to explore the connec tion between weather conditions and hardlock, and to clarify the relations hips between thrips, hardlock and yield. Materials and Methods Flowers were tagged using ribbons on a weekly basis. This occurred at 4 separate locations in Florida (Altha, Jay, Marianna, and Quin cy) for 2 to 4 years. Temperature and relative humidity were recorded at 15-minute interval s using a CR10X data logger and 2 HMP 45 AC temperature-relative humidity probes. This allowed comparisons between each of the 2 temperature and 2 relative humidity measurem ents to improve accuracy. The weather measurement system was located within 100 m of the study flowers. After defoliation, all bolls from tagged flowers were removed from the plants and evaluated for hardlock severity. Hardlock severity was calculated as the number of hardlocked locules of the total present. Data were recorded separately for each boll. Bolls were categorized by date of bloom, and the average hardlock severity was calculated for each date. The number of bolls recovered for each date va ried considerably. Typically 20 to 40 bolls were recovered, and dates with fewer than 5 bo lls were excluded from the analysis. The 2 temperature and 2 relative humidity values for e ach time interval were averaged to produce 1

PAGE 81

81 temperature and 1 relative humidity value. After choosing the time interval of interest, the mean for each weather variable was determined using SAS for each day for which flowers had been tagged. The PROC REG command was used to perf orm regressions for each location and year. The relationship between thrips numbers, hard lock, and yield was evaluated in fungicideinsecticide studies at branch st ations of the North Florida Res earch and Education Center in Marianna and Quincy, FL. These studies were re ported in Ch.2 and Ch.3, but all 3 variables were not compared. In this analysis it is the thrips numbers, and not the treatments, that are of interest. Results Weather Models of Hardlock Severity The model involving the temperature and rela tive humidity from 0700 to 1900, as reported in (Marois et al., 2004), was tested for each locati on and year of this study (Table 4.1). It was significant for the data set from which it wa s developed (Quincy, 2002), showing a connection between higher temperature and humidity to increa sed hardlock serverity. At that same location in 2003 it was also significant. However, the re lationship between hardlock and the temperature component of the model was reversed from the pr evious year. In other location years the model was frequently insignificant, and the relationships betw een model components varied regularly. Grouping the data sets by year or location did not improve si gnificance. Grouping all 63 days from the data sets showed little or no rela tionship. The temperature or relative humidity components were also analyzed separatel y, but their significa nce and adjusted R2 values were generally low, and their relationships to hardlo ck varied from positive to negative (Tables 4.2 and 4.3). In Chapter 2, the temperature between 0800 and 1000 was shown to be negatively associated with thrips numbers at 1000. To determine if weather conditions may influence hardlock indirectly, through its impact on thrips numbers in fl owers, the mean temperature

PAGE 82

82 between 0800 and 1000 was compared to hardlock incidence for that day (Table 4.4). When viewed by location-year, the relationships were mo stly negative, and more significant than the first model. When examined in groupings by loca tion or year, the models were typically better or equal to the individual-locati on year models. Grouping all days produced a significant (p=0.07), negative relationship with an extremely low R2 value (0.05). The average relative humidity during this period was also examined (Table 4.5). The relationships between humidity and hardlock fluctuated between positiv e and negative, and the adjusted R2 values were usually low. Constructing a model using both temperature and relative humidity from 0800 to 1000 was no better than using temperature alone, and the re lative humidity component varied between a positive and negative association (Table 4.6). Although the temperature from 0800 to 1000 predic ted hardlock incidence better than the first model, it was unclear if othe r time periods might be a better predictor. To determine this, regression analyses were performed for each lo cation-year comparing hardlock incidence to temperature (Table 4.7) for each hour from 1700 on the day prior to bloom to 2400 on the day of bloom. Hours showing a significant (p < 0.10) rela tionship were compared for consistency across each of the location years. Significant relati onships were observed from 1700 (prior to bloom day) to 1000 and 1300 to 2400 on the day of bl oom. These ranges vari ed considerably by location-year, and when analyzed together on ly the period from 0000 to 0600 was shown to be important, based on significance and adjusted R2 values (>0.20). The peak significance time within this period varied by modelyear, so an average for the period was calculated for each. The resulting model tested a linear relationship between temperatur e from 0000 to 0600 and hardlock incidence for that same day (Table 4.8). In all location-years, locations, y ears, and combined data, night time temperature was negatively associated with hardlock incidence. In 4 of 11 location-

PAGE 83

83 years, the model was significant at p < 0.20 and adjusted R2 values were >0.50. Grouped by location, p < 0.05, and adjusted R2 values varied from 0.15 to 0.79. Examined by year the results were similar, with p < 0.10, and adjusted R2 values varied from 0.12 to 0.28. The combined data set showed a significance of p = <0.0001 and an adjusted R2 value of 0.27. To determine if this result was complicated by plant maturity or varying hardlock intensity within the growing seasons, the temperatures (Fig. 4.3) and hardlock se verity (Fig. 4.4) were plotted against the date. These did not appear to influence the model. A dding the date (in the form of Julian day) as a model component did not increase the adjusted R2 value (data not shown). A problem with this model was that the residuals be tween the predicted and actual ha rdlock did not fit a normal distribution, and were larger toward the extremes of the temperature range. To further refine the model, days on whic h fewer than 10 bolls were recovered were removed from the analysis (Table 4.9 ). This improved the adjusted R2 value from 0.26 to 0.30. Then days with temperatures below 21 or above 25C were also excluded, leaving 49 of the 63 (77%) original days (Fig. 4.5). This improved the adjusted R2 value to 0.40 (Table 4.9). When the resulting equation was used to predict hardlock in that same data, the residuals were normally distributed, with a mean of 0.00 and a standard de viation of 0.20 (Table 4.10). To further explore the data, days with 10 or more bolls were a ssigned to two temperature groups for separate regression analyses: 17.1 to 22.9C or 23.0 to 26.0 C. Days between 17.1 and 22.9C showed a gradual slope, significant (p=0.06) relationship, and low adjusted R2 value (0.07). The second group (23.0 to 26.0C) showed a more gradual slope an insignificant re lationship (p=0.66), and an adjusted R2 value of .04. The mean hardlock severity for low and high temperature groups were 62 and 30%, respectively. The change in hardlo ck severity at 23C is fairly abrupt. As an alternative model, days above 23C were categor ized as having 30% hardlock severity, while

PAGE 84

84 those below 23C were categorized as having 62% ha rdlock. The resulting residuals were similar to those achieved using the linear regression model (Table 4.10). For days within the 21 to 25C range, the linear regression model had a slightly better residual m ean and standard deviation than did the categorical model. When applying thes e models to the full temperature range (17 to 26C), the categorical model performed better. Th is suggests the specific temperature of the day should determine which model to use. To validate the model (linear regression, >10 bolls, 21 to 25C), 25 individual days were randomly sampled from the data set. A regressi on was performed, and the resulting equation was used to predict hardlock in the other 24 days. The residuals between the predicted and actual hardlock severity for those days were calculate d (Table 4.11). This procedure was performed a total of four times. The mean residual ranged from .007 to .090, and the standard deviations were between 0.16 and 0.23. The association of flower thrips with hardlo ck for individual days was also tested. The flowers tagged in Marianna were from rows ad jacent to the fungicide-i nsecticide hardlock study described in Ch.2 and Ch.3. The 2 adjacent plots were a control and an insecticide+fungicide treatment. The mean number of thrips from the 2 adjacent plots was used as an estimate of thrips numbers within the tagged rows, since they were not sampled. Thrips were not a good predictor of hardlock (Table 4.12). When both thrips numbers and temperature from 0000 to 0600 were included in the model, the results were better than using thrips alone but worse than using temperature alone. However, these m odels using thrips have several sh ortfalls. First, if thrips are important, they should be significant on their own. Thrips numbers vary throughout an area, and their numbers appear to fluctuate independently of the overall trend. It is possible the estimates derived from adjacent plots were not representati ve of the tagged flowers. A second problem is

PAGE 85

85 the temperature model did not perform as well in Marianna as other locations, so combining it with the thrips numbers which themselves did not predict well does not seem likely to improve the accuracy. The models were based on a relativel y small number of days, and adding additional variables to the models makes them less reliable. Connections between Thrips, Hardlock, and Yield The relationship between thrips numbers, hard lock, and yield on a pe r plot basis at the season-level was examined (Table 4.13). In Ma rianna, higher thrips numbers were associated with more hardlock. The model was significant in 2 of 3 years, and also when the data for all the years was merged. Both R2 and adjusted R2 values remained below 0.40. Hardlock was inversely associated with yield, and this was significant in all years except 2003. In 2003, the yields were very low in all treatments (Ch.3) and it is likely that factors other than hardlock were more important. When the data was merged across ye ars, the result was hi ghly significant (p < 0.0001), and R2 and adjusted R2 values were both 0.57. The association between thrips numbers and yield was less consistent, but negative ove rall. In only 1 of 3 years was the relationship significant (p = 0.0129), and in the non-significant years higher thrips numbers we re associated with higher yielding plots. Combining the data from all 3 years improved the signifi cance of the model (p = <0.0001), and boosted the R2 and adjusted R2 values to 0.48. Using thrips numbers and hardlock to predict yield proved highly significant in 2005 (p = 0.0009) and when data was combined across years (p = <0.0001). The highest adjusted R2 value by year was 0.34, but reached 0.64 when the data was combined across years. Simila r results were obtained in Quincy. Higher thrips numbers were associated with higher hardlo ck in 2004 (p = 0.02) and 2005 (p = 0.08). R2 and adjusted R2 values were higher than in Marianna for both years. Higher-yielding plots were generally associated with lower hardlock, although the relationship was weaker than in Marianna.

PAGE 86

86 Thrips numbers were not directly related to yield. Using thrips numbers and hardlock as a model to predict yield did not pr ove significant in Quincy. Discussion In this study, the association of weather conditions and hardlo ck incidence was explored. The first reported hardlock-weather model (Maroi s et al., 2004) was test ed, but did not prove adequate in all locations. A s econd model, previously used for predicting thrips numbers, was also evaluated. Although it was an improvement fr om the first model, its predictive ability was very limited. Examining the temperatures by hour revealed the time be tween 0000 and 0600 to be most important for predicting hardlock. A mode l was constructed using the average temperature for that time period. This model out-performed t hose tested previously. It was further refined by limiting it to days within the temp erature range of 21 to 25C. It was also determined that days can be assigned to a high or low hardlock severi ty group with some success based on whether they are below or above 23C. The first two models tested had straightforward biological explanations. In the first model, temperature and relative humidity were positively associated with hardlock incidence. Such conditions would be favorable for the germ ination and growth of the causal agent, F. verticillioides In the second model, temperature fr om 0800 to 1000 were negatively associated with hardlock. This temperature range had prev iously been shown to be negatively associated with thrips numbers at 1000. Since warm temperatur es slow the movement of thrips into flowers, it is assumed their harmful effects would be dimi nished, resulting in less hardlock. However, neither model fit the data as effectively as di d the third. It is possi ble that cooler night temperatures have a subtle effect on the deve loping flower making it more susceptible to infection. Oosterhuis and Jernst edt (1999) noted cool temperatur es can delay anthesis (period when pollen is shed) by two to three hours. Th is would delay pollination, possibly increasing the

PAGE 87

87 amount of time the stigma is receptive and susceptible to infection by Fusarium Although the low temperatures from 0800 to 1000 that are as sociated with higher th rips numbers did not consistently predict hardlock, in some instances they might be important. It may be possible to develop a model to predict hardlock based on te mperature and the number of thrips known to be inhabiting cotton flowers. At the Marianna site (Chapter 3), larger incr eases in the proportion of non-hardlocked cotton due to insecticide applications were obtained in the years with the highest hardlock severity. If temperatures from 0000 to 0600 are used to estimate hardlock severity, this could allow a more optimal use of insecticide applications. Models have been developed previously to optimize management of cotton. The GOSSYM/COMAX model has been used for estimating irrigation needs (Staggenborg et al, 1996) and exploring the result of various fertilizer and defoliation strategies on yield (McKinion et al., 1989) The relationships between thrips numbers, hardlo ck severity, and yield were also explored. The models comparing thrips to hardlock were mostly significant. Among the significant ones, the adjusted R2 values ranged from 0.09 to 0.54, sugges ting thrips are an important, but not exclusive, factor in hardlock severity. This lends further support to the fi ndings of Chapter 3, in which significant treatment effects were observed in the severity of hardlock. There are probably other factors influencing the proba bility of infection, progression of the disease, and expression of symptoms after boll opening. The relationship be tween hardlock and yield was significant in many cases, and those location-years in which it wasnt were characterized by below average yield. This suggests hardlock is one of many factors that can influence yields. It is also interesting to note that when viewed by year, there was no relationship between thrips numbers and yield. This suggests they have no direct imp act on the cotton plants, a nd any effect is due to hardlock.

PAGE 88

88 In conclusion, the hardlock se verity for a given day can be predicted with an adjusted R2 of 0.40 using the mean temperature between 0000 and 0600. Cooler temperatur es are associated with higher hardlock severity, a nd these predictions are most va lid for temperatures between 21 and 25C. Below 21C severity is typically above 50%, while above 25C severity is below 50%. This model could be used to optimize spraying, but needs to be tested further. The impact of thrips control at different temp eratures is unknown, but this must be determined for the model to optimize spraying. This could be done by tagging fl owers in control and insecticide-treated plots, and quantifying the number of thrips in each plot on that day. Night temperatures can vary considerably from day to day, and tagging flow ers on consecutive days may reduce interference from any other sources. Each fl ower is at risk of infection fo r one day, and this occurs over an eight week period. Targeting protec tive strategies to those days w ith the highest risk allows the most judicious use of pesticides, thereby ma ximizing economic and environmental benefits.

PAGE 89

89 Figure 4.1. Cotton boll exhibiti ng symptoms of hardlock.

PAGE 90

90 Table 4.1. Relationship between average temperat ure and relative humidity from 0700 to 1900 on the day of bloom and hardlock severity for that day. location year N days equation p R2 adj.-R2 Altha 2003 6 H = 3.38485 -T719*0.03388 RH719*0.02245 0.6703 0.2341 -0.2765 Altha 2004 4 H = 1.49718 +T719*0.10730 RH719*0.06072 0.3958 0.8433 0.5300 Jay 2003 8 H = 1.61997 -T719*0.03756 -RH719*0.00376 0.5768 0.1976 -0.1234 Jay 2004 8 H = 8.06744 -T719*0.23627 -RH719*0.01480 0.1048 0.5943 0.4320 Jay 2005 4 H = 0.86125 +T719*0.01482 RH719*0.01169 0.2698 0.9272 0.7816 Marianna 2003 4 H = -12.52706 +T719*0.28892 +RH719*0.06857 0.6429 0.5867 -0.2400 Marianna 2004 4 H = 0.12447 +T719*0.02388 RH719*0.00769 0.7222 0.4784 -0.5648 Marianna 2005 5 H = 31.72419 -T719*0.70494 -RH719*0.14080 0.5165 0.4835 -0.0330 Quincy 2002 8 H = -5.53090 +T719*0.13061 +RH719*0.03539 0.0232 0.7780 0.6892 Quincy 2003 8 H = 1.62170 -T719*0.05959 +RH719*0.00662 0.0240 0.7749 0.6849 Quincy 2004 4 H = 5.07364 -T719*0.10155 -RH719*0.02803 0.1916 0.9633 0.8899 Altha 10 H = 6.32756 -T719*0.11106 RH719*0.03409 0.5888 0.1404 -0.1052 Jay 20 H = 3.64325 -T719*0.08808 -RH719*0.01061 0.0316 0.3341 0.2557 Marianna 13 H = 2.78049 -T719*0.08754 +RH719*0.00429 0.1003 0.3687 0.2424 Quincy 20 H = -2.14070 +T719*0.05701 +RH719*0.01397 0.7066 0.0400 -0.0729 2003 26 H = -1.15743 +T719*0.02203 + RH719*0.01389 0.4931 0.0596 -0.0221 2004 20 H = 4.36586 -T719*0.10977 -RH719*0.01140 0.0602 0.2816 0.1970 2005 9 H = 1.83714 -T719*0.03499 -RH719*0.00510 0.6282 0.1435 -0.1419 all all 63 H = 1.46886 -T719*0.02710 -RH719*0.00259 0.5649 0.0189 -0.0138 Table 4.2. Relationship between average temperat ure from 0700 to 1900 on the day of bloom and hardlock severity for that day. location year N days Temp. C Percent hardlock equation p R2 adj.-R2 Altha 2003 6 25.5 30.6 54 97 H = -0 .52745 +T719*0.0446 0. 3620 0.2090 0.0113 Altha 2004 4 29.2 30.6 6 80 H = -7. 53823 +T719*0.27067 0.5045 0.2455 -0.1317 Jay 2003 8 26.2 31.1 4 44 H = 1. 00990 -T719*0.02586 0.2811 0.1894 0.0543 Jay 2004 8 25.7 28.9 12 94 H = 5. 88195 -T719*0.19692 0.0669 0.4541 0.3631 Jay 2005 4 26.6 29.2 37 59 H = 2. 30545 -T719*0.06512 0.3206 0.4616 0.1925 Marianna 2003 4 26.1 29.9 72 91 H = 1. 52993 -T719*0.02607 0.4803 0.2701 -0.0949 Marianna 2004 4 28.5 30.5 26 39 H = 1. 29241 -T719*0.03191 0.5291 0.2217 -0.1674 Marianna 2005 5 28.1 31.4 19 80 H = 1. 82540 -T719*0.04629 0.6963 0.0580 -0.2560 Quincy 2002 8 27.6 33.0 40 96 H = 0. 05871 +T719*0.02353 0.6928 0.0279 -0.1342 Quincy 2003 8 26.1 30.5 29 70 H = 2.88915 -T719*0.08660 0.0043 0.7684 0.7298 Quincy 2004 4 29.1 31.0 2 34 H = 3. 06464 -T719*0.09699 0.4876 0.2626 -0.1061 Altha 10 H = 0.21026 +T 719*0.01483 0.7937 0.0091 -0.1148 Jay 20 H = 2.18815 -T719*0.06390 0.0360 0.2219 0.1787 Marianna 13 H = 3.48033 -T719*0.10098 0.0316 0.3553 0.2967 Quincy 20 H = 0.47849 +T 719*0.00140 0.9729 0.0001 -0.0555 2003 26 H = 1.28771 -T719*0.02744 0.3221 0.0408 0.0009 2004 20 H = 2.77654 -T719*0.08238 0.0459 0.2035 0.1593 2005 9 H = 1.59363 -T719*0.03912 0.4173 0.0959 -0.0332 all all 63 H = 1.05481 -T 719*0.01921 0.3283 0.0157 -0.0005

PAGE 91

91 Table 4.3. Relationship between average relative humidity from 0700 to 1900 on the day of bloom and hardlock severity for that day. location year N days RH % Percent hardlock equation p R2 adj.-R2 Altha 2003 6 68 86 54 97 H = 1.72035 -R H719*0.01310 0.3363 0.2296 0.0370 Altha 2004 4 65 73 6 80 H = 5.04400 -R H719*0.06591 0.1001 0.8098 0.7146 Jay 2003 8 62 87 4 44 H = -0.18292 +RH719*0.00621 0.3837 0.1282 -0.0170 Jay 2004 8 62 85 12 94 H = 0.84503 -RH719*0.00462 0.7695 0.0154 -0.1487 Jay 2005 4 56 79 37 59 H = 1.18929 -RH719*0.01039 0.0421 0.9176 0.8764 Marianna 2003 4 68 86 72 91 H = 0.32869 +RH719*0.00617 0.4322 0.3224 -0.0164 Marianna 2004 4 54 71 26 39 H = 0.68295 -RH719*0.00530 0.3335 0.4443 0.1664 Marianna 2005 5 66 81 19 80 H = 0.10734 +RH719*0.00463 0.8544 0.0131 -0.3158 Quincy 2002 8 60 79 40 96 H = -0.33560 +RH719*0.01663 0.1660 0.2929 0.1750 Quincy 2003 8 67 85 29 70 H = -1.08503 +RH719*0.02004 0.0057 0.7457 0.7033 Quincy 2004 4 60 70 2 34 H = 1.98995 -RH719*0.02752 0.1780 0.6757 0.5135 Altha 10 H = 1.29985 -RH719*0.00906 0.5130 0.0553 -0.0628 Jay 20 H = 0.55598 -RH719*0.00208 0.7558 0.0055 -0.0497 Marianna 13 H = -0.42226 +RH719*0.01323 0.1398 0.1872 0.1133 Quincy 20 H = 0.26393 +RH719*0.00362 0.6800 0.0097 -0.0453 2003 26 H = -0.12656 +RH719*0.00847 0.2429 0.0564 0.0170 2004 20 H = 0.48189 -RH719*0.00118 0.8911 0.0011 -0.0544 2005 9 H = 0.88559 -RH719*0.00605 0.4965 0.0684 -0.0646 all all 63 H = 0.39754 +RH719*0.00144 0.7409 0.0018 -0.0146 Table 4.4. Relationship between average temperat ure from 0800 to 1000 on the day of bloom and hardlock severity for that day. location year N days Temp. C Percent hardlock equation p R2 adj.-R2 Altha 2003 6 24.2 28.4 54 97 H = -0 .86394 +T810*0.06303 0. 2939 0.2670 0.0838 Altha 2004 4 25.5 27.6 6 80 H = 9. 68415 -T810*0.34434 0.0738 0.8579 0.7868 Jay 2003 8 23.8 27.4 4 44 H = 1.47193 -T810*0.04634 0.3596 0.1409 -0.0023 Jay 2004 8 22.2 27.1 12 94 H = 2. 94612 -T810*0.09854 0.1263 0.3442 0.2349 Jay 2005 4 21.7 26.8 37 59 H = 1. 56850 -T810*0.04426 0.0266 0.9475 0.9212 Marianna 2003 4 25.5 30.3 72 91 H = 1.87963 -T810*0.03817 0.0061 0.9878 0.9817 Marianna 2004 4 24.8 27.3 26 39 H = 1.43672 -T810*0.04154 0.2332 0.5879 0.3819 Marianna 2005 5 25.1 28.6 19 80 H = 1. 97583 -T810*0.05644 0.5042 0.1603 -0.1196 Quincy 2002 8 23.4 27.8 40 96 H = 0. 18795 +T810*0.02307 0.7143 0.0240 -0.1387 Quincy 2003 8 23.5 27.2 29 70 H = 2.45497 -T810*0.07822 0.0967 0.3921 0.2908 Quincy 2004 4 26.0 27.4 2 34 H = 2. 37327 -T810*0.08271 0.6656 0.1118 -0.3323 Altha 10 H = 1.70746 -T810*0.04154 0.5209 0.0533 -0.0650 Jay 20 H = 2.47235 -T810*0.08200 0.0061 0.3488 0.3127 Marianna 13 H = 0.06460 +T 810*0.01672 0.6983 0.0142 -0.0754 Quincy 20 H = 3.00672 -T810*0.09678 0.0536 0.1916 0.1467 2003 26 H = 0.08964 +T810* 0.01632 0.5871 0.0125 -0.0287 2004 20 H = 3.18640 -T810*0.10754 0.0048 0.3651 0.3292 2005 9 H = 1.43866 -T810*0.03766 0.1939 0.2278 0.1174 all all 63 H = 1.44105 -T810*0.03629 0.0692 0.0531 0.0376

PAGE 92

92 Table 4.5. Relationship between average relative humidity from 0800 to 1000 on the day of bloom and hardlock severity for that day. location year N days RH % Percent hardlock equation p R2 adj.-R2 Altha 2003 6 77 94 54 97 H = 2.50525 -R H810*0.02010 0.1750 0.4040 0.2550 Altha 2004 4 77 81 6 80 H = 4.96424 -R H810*0.05535 0.6117 0.1508 -0.2739 Jay 2003 8 76 92 4 44 H = 0.33838 -RH810*0.00072459 0.9559 0.0006 -0.1660 Jay 2004 8 70 99 12 94 H = 0.59850 -RH810*0.00121 0.9366 0.0011 -0.1653 Jay 2005 4 71 89 37 59 H = 1.12832 -RH810*0.00798 0.4064 0.3524 0.0285 Marianna 2003 4 70 87 72 91 H = 0.00280 +RH810*0.01027 0.0266 0.9476 0.9214 Marianna 2004 4 74 83 26 39 H = 1.34000 -RH810*0.01254 0.2326 0.5889 0.3833 Marianna 2005 5 77 94 19 80 H = -0.61222 +RH810*0.01245 0.5012 0.1624 -0.1168 Quincy 2002 8 76 94 40 96 H = -1.16870 +RH810*0.02271 0.0214 0.6140 0.5496 Quincy 2003 8 79 96 29 70 H = -0.94665 -RH810*0.01597 0.1952 0.2615 0.1384 Quincy 2004 4 76 85 2 34 H = 2.17161 -RH810*0.02457 0.2680 0.5358 0.3037 Altha 10 H = 0.78812 -RH810*0.00177 0.9095 0.0017 -0.1231 Jay 20 H = 0.92170 -RH810*0.00618 0.4426 0.0331 -0.0206 Marianna 13 H = 0.60392 -RH810*0.00103 0.9243 0.0009 -0.0900 Quincy 20 H = -1.13329 +RH810*0.01930 0.0838 0.1569 0.1101 2003 26 H = 1.00986 -RH810*0.00579 0.4669 0.0223 -0.0185 2004 20 H = 0.83577 -RH810*0.00535 0.6368 0.0127 -0.0422 2005 9 H = 0.25670 +RH810*0.00236 0.8013 0.0097 -0.1318 all all 63 H = 0.24149 +RH810*0.00309 0.5502 0.0059 -0.0104 Table 4.6. Relationship between average temperat ure and relative humidity from 0800 to 1000 on the day of bloom to hardlock severity for that day. location year N days equation p R2 adj.-R2 Altha 2003 6 H = 5.03406 -T810*0.05457 RH810*0.03313 0.4252 0.4345 0.0576 Altha 2004 4 H = 7.38611 -T810*0.47509 + RH810*0.07181 0.1097 0.9880 0.9639 Jay 2003 8 H = 2.00074 -T810*0.05174 -RH810*0.00454 0.6452 0.1608 -0.1749 Jay 2004 8 H = 3.38390 -T810*0.10157 -RH810*0.00439 0.3290 0.3590 0.1026 Jay 2005 4 H = 1.47966 -T810*0.05652 +RH810*0.00476 0.0091 0.9999 0.9998 Marianna 2003 4 H = 3.25879 -T810*0.06587 -RH810*0.00768 0.0584 0.9966 0.9898 Marianna 2004 4 H = 1.41568 -T810*0.02101 -RH810*0.00654 0.6292 0.6042 -0.1875 Marianna 2005 5 H = -0.23261 -T810*0.00833 +RH810*0.01064 0.8375 0.1625 -0.6751 Quincy 2002 8 H = -3.70374 +T810*0.08013 +RH810*0.02873 0.0073 0.8599 0.8038 Quincy 2003 8 H = 2.44961 -T810*0.07812 -RH810*0.00003128 0.2881 0.3921 0.1490 Quincy 2004 4 H = 6.73137 -T810*0.15094 -RH810*0.03074 0.3545 0.8743 0.6230 Altha 10 H = 6.22996 -T810*0.12984 RH810*0.02631 0.4737 0.1922 -0.0386 Jay 20 H = 2.69654 -T810*0.08001 -RH810*0.00327 0.0231 0.3579 0.2824 Marianna 13 H = -0.45543 -T810*0.02583 -RH810*0.00335 0.9079 0.0191 -0.1770 Quincy 20 H = 1.37903 -T810*0.06973 -RH810*0.01089 0.1126 0.2266 0.1356 2003 26 H = 1.42445 -T810*0.00911 -RH810*0.00786 0.7626 0.0233 -0.6160 2004 20 H = 3.85551 -T810*0.10964 -RH810*0.00754 0.0150 0.3900 0.3183 2005 9 H = 1.19601 -T810*0.03831 -RH810*0.00309 0.4315 0.2443 -0.0075 all all 63 H = 1.53183 -T810*0.03742 -RH810*0.00072999 0.1928 0.0534 0.0218

PAGE 93

93 Table 4.7. Association of temperature and hardlock severity by hour for each location. Altha Jay Marianna Quincy all locations hour p relat. p relat. p relat. p relat.p relat. R2 adj-R2 17b 0.63 + 0.28 0.00 0.99 + 0.18 0.02 0.01 18b 0.66 0.29 0.00 0.32 + 0.35 0.01 0.00 19b 0.72 0.34 0.00 0.50 + 0.25 0.02 0.00 20b 0.55 0.22 0.00 0.59 0.02 0.08 0.06 21b 0.10 0.13 0.00 0.02 0.00 0.22 0.21 22b 0.04 0.10 0.00 0.13 0.00 0.17 0.16 23b 0.02 0.13 0.00 0.27 0.00 0.17 0.15 24b 0.01 0.16 0.01 0.18 0.00 0.16 0.14 1 0.00 0.11 0.02 0.00 0.00 0.23 0.22 2 0.00 0.07 0.01 0.00 0.00 0.27 0.26 3 0.00 0.05 0.02 0.00 0.00 0.30 0.29 4 0.00 0.03 0.03 0.00 0.00 0.29 0.27 5 0.00 0.04 0.13 0.00 0.00 0.26 0.24 6 0.00 0.03 0.38 0.01 0.00 0.22 0.21 7 0.00 0.04 0.43 + 0.04 0.00 0.12 0.10 8 0.00 0.04 0.52 + 0.03 0.01 0.08 0.07 9 0.23 0.01 0.76 + 0.00 0.02 0.07 0.06 10 0.92 + 0.00 0.92 + 0.07 0.06 0.05 0.03 11 0.99 0.06 0.65 0. 94 0.61 0.00 -0.01 12 0.89 + 0.23 0.15 0.97 0.49 0.00 0.00 13 0.92 0.34 0.04 0.77 + 0.41 0.01 0.00 14 0.90 + 0.23 0.03 0.59 0.16 0.03 0.01 15 0.64 + 0.29 0.03 0.51 0.24 0.02 0.00 16 0.04 + 0.47 0.03 0. 83 0.82 0.00 -0.01 17 0.07 + 0.88 0.06 0. 83 + 0.77 + 0.00 -0.01 18 0.81 + 0.99 + 0.04 0. 85 + 0.91 0.00 -0.01 19 0.58 0.42 0.01 0.66 0.25 0.02 0.00 20 0.29 0.42 0.01 0.08 0.03 0.07 0.05 21 0.20 + 0.07 0.12 0.00 0.05 0.05 0.04 22 0.04 + 0.00 0.13 0.00 0.00 0.11 0.09 23 0.05 + 0.00 0.17 0.00 0.00 0.15 0.14 24 0.09 + 0.00 0.11 0.00 0.00 0.14 0.13

PAGE 94

94 Table 4.8. Relationship between average temperat ure from 0000 to 0600 on the day of bloom to hardlock severity for all days. location year N days Temp. C Percent hardlock equation p R2 adj.-R2 Altha 2003 6 21.8 22.9 54 97 H = 7. 90058 -T0106*0.32403 0.0253 0.7520 0.6901 Altha 2004 4 20.6 24.8 6 80 H = 4. 61249 -T0106*0.18074 0.0385 0.9244 0.8866 Jay 2003 8 21.9 24.6 4 44 H = 1.38649 -T0106*0.04749 0.4719 0.0894 -0.0624 Jay 2004 8 18.7 25.2 12 94 H = 1.45101 -T0106*0.04274 0.4591 0.0944 -0.0565 Jay 2005 4 19.7 24.5 37 59 H = 1. 57428 -T0106*0.04930 0.0259 0.9489 0.9233 Marianna 2003 4 22.2 -23.5 72 91 H = 1.08427 -T0106*0.01260 0.9087 0.0083 -0.4875 Marianna 2004 4 22.6 26.0 26 39 H = 0.42002 -T0106*0.00327 0.9182 0.0067 -0.4900 Marianna 2005 5 21.8 24.3 19 80 H = 3.10125 -T0106*0.11400 0.2816 0.3637 0.1517 Quincy 2002 8 17.1 22.3 40 96 H = 1.65947 -T0106*0.04368 0.2845 0.1870 0.0515 Quincy 2003 8 21.7 23.3 29 70 H = 1. 09049 -T0106*0.02837 0.8331 0.0080 -0.1573 Quincy 2004 4 19.7 23.9 2 34 H = 1.64555 -T0106*0.06688 0.1646 0.6979 0.5468 Altha 10 H = 5.30923 -T 0106*0.20881 0.0004 0.8136 0.7903 Jay 20 H = 1.75501 -T0106*0.05940 0.0495 0.1976 0.1531 Marianna 13 H = 2.99630 -T0106*0.10534 0.0426 0.3233 0.2617 Quincy 20 H = 2.60580 -T0106*0.09657 0.0051 0.3614 0.3259 2003 26 H = 4.27822 -T0106*0.16566 0.0066 0.2689 0.2385 2004 20 H = 1.72184 -T0106*0.05796 0.0682 0.1729 0.1270 2005 9 H = 1.94613 -T0106*0.06506 0.0809 0.3726 0.2829 all all 63 H = 2.33353 -T0106*0.08148 <0.0001 0.2770 0.2652

PAGE 95

95 A0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 17.018.019.020.021.022.023.024.025.026.0 Temperature (C)Proportion hardlocked Altha Jay Marianna Quincy By = -0.0812x + 2.3261 R2 = 0.2749 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 17.018.019.020.021.022.023.024.025.026.0 Temperature (C)Proportion hardlocked Figure 4.2. Mean temperature from 0000 to 0600 and hardlock severity. A Days by location. B Regression line.

PAGE 96

96 y = -0.0205x + 821.66 R2 = 0.0503 15.0 17.0 19.0 21.0 23.0 25.0 27.0 2-Jul12-Jul22-Jul1-Aug11-Aug21-Aug31-Aug10-Sep20-Sep DateTemperature (C) Figure 4.3. Distribution of mean te mperatures from 0000 to 0600 for days in the climate model. y = 0.0021x 81.031 R2 = 0.0193 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 2-Jul12-Jul22-Jul1-Aug11-Aug21-Aug31-Aug10-Sep20-Sep DateProportion hardlocked Figure 4.4. Incidence of hardlock during the growing season for days evaluated.

PAGE 97

97 Table 4.9. Relationship between average temperat ure from 0000 to 0600 on the day of bloom to hardlock severity for temperatures between 21 and 25C. N days Temp. hardlock bolls recovered equation p R2 adj.-R2 63 17.1 26.0 0.02 0.97 >5 H = 2. 32467 -T*0.08106 <0.0001 0.2731 0.2611 58 17.1 26.0 0.02 0.97 >10 H = 2. 42539 -T*0.08543 <0.0001 0.3145 0.3022 49 21.4 24.8 0.02 0.97 >10 H = 4. 57773 -T*0.17898 <0.0001 0.4173 0.4049 37 17.1 22.9 0.29 0.97 >10 H = 1. 66745 -T*0.04858 0.0602 0.0973 0.0715 21 23.1 26.0 0.02 0.80 >10 H = 0.97826 -T*0.02803 0.6627 0.0102 -0.0419 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 17.019.021.023.025.027.0 Temperature (C)Proportion hardlocked Figure 4.5. Days in model with >10 bolls recovered.

PAGE 98

98 y = 0.0007x 0.0155 R2 = 1E-05 -0.60 -0.40 -0.20 0.00 0.20 0.40 0.60 0.80 1.00 21.021.522.022.523.023.524.024.525.0 Temperature (C)Proportion hardlocked Figure 4.6. Prediction residuals for days with >10 bolls, 21 to 25C (N=49). Table 4.10. Mean and standard deviation of mode ls when predictions are applied to data. linear categorical 21-25C mean 0.00 0.01 st. dev. 0.20 0.22 17-26C mean 0.05 0.00 st. dev. 0.27 0.22 Table 4.11. Validation of the model by random sa mpling and applying predictions to data. models established from random sampling of days prediction residuals N equation p R2 adj.-R2 N mean SD 25 H = 3.55426 -T*0.13543 0.0071 0.2752 0.2437 24 -0.0581 0.2218 25 H = 4.77974 -T*0.18862 0.0034 0.316 0.2863 24 -0.0372 0.1644 25 H = 3.33086 -T*0.12455 0.0053 0.2921 0.2613 24 -0.0068 0.2324 25 H = 4.37491 -T*0.17204 0.0003 0.4392 0.4149 24 -0.0904 0.1928

PAGE 99

99 Table 4.12. Relationship of temperature from 00 00 to 0600 and estimated thrips numbers to hardlock in Marianna, FL. location year N days thrips equation p R2 adj.-R2 Marianna 2003 4 2.3 4.4 H = 0. 62834 +TH*0.05242 0.3332 0.4446 0.1669 Marianna 2004 4 0.4 4.7 H = 0.32735 +TH*0.00524 0.8473 0.0233 -0.4650 Marianna 2005 5 1.4 2.5 H = 0.71081 -TH*0.13613 0.6866 0.0619 -0.2509 Marianna all 13 H = 0.35030 +T H*0.06850 0.2534 0.1166 0.0363 Marianna 2003 4 2.3 4.4 H = 3.17168 -T*0 .11778 +TH*0.09484 0.3430 0.8823 0.6470 Marianna 2004 4 0.4 4.7 H = 0.30124 +T*0 .00101 +TH*0.00580 0. 9881 0.0237 -1.9289 Marianna 2005 5 1.4 2.5 H = 3.09767 -T*0.1 0907 -TH*0.05674 0.6262 0.3738 -0.2524 Marianna all 13 H = 2.65757 -T*0.094 86 +TH*0.03723 0.1121 0.3545 0.2254

PAGE 100

100 Table 4.13. Relationships between thrips, hardlock and yield on a per plot seasonal basis. location year N thrips hardlock yield equation p R2 adj.-R2 Marianna 2003 32 x x hardlock = 0.4116 5 +thrips*0.05216 0.0460 0.1263 0.0971 Marianna 2004 32 x x hardlock = 0.3498 0 -thrips*0.00825 0.7560 0.0033 -0.0300 Marianna 2005 32 x x hardlock = 0.1191 8 +thrips*0.11944 0.0007 0.3219 0.2993 Marianna all 96 x x hardlock = 0.1583 1 +thrips*0.09310 0.0001 0.3902 0.3837 Marianna 2003 32 x x yield = 516.60053 -hardlock*36.06410 0. 7810 0.0026 -0.0306 Marianna 2004 32 x x yield = 1366.33802 -hardlock*967.06381 0.0026 0.2648 0.2403 Marianna 2005 32 x x yield = 1902.35967 -hardlock*1560.96214 0.0002 0.3735 0.3526 Marianna all 96 x x yield = 1711.36408 -hardlock*1740.30903 <0.0001 0.5664 0.5618 Marianna 2003 32 x x yield = 474.833 74 +thrips*5.09984 0.7888 0.0024 -0.0308 Marianna 2004 32 x x yield = 955.192 83 +thrips*33.86048 0.4962 0.0156 -0.0172 Marianna 2005 32 x x yield = 1783.89 191 -thrips*233.84959 0.0129 0.1892 0.1621 Marianna all 96 x x yield = 1645.94 504 -thrips*239.86385 <0.0001 0.4844 0.4790 Marianna 2003 32 x x x yield = 497.64089 +thrips* 7.98992 -hardlock*55.4035 8 0.8924 0.0078 -0.0606 Marianna 2004 32 x x x yield = 1289.87769 +thrips* 25.96324 -hardlock*956. 78225 0.0096 0.2739 0.2238 Marianna 2005 32 x x x yield = 1947.46192 -thrips*69 .92373 -hardlock*1372.49023 0.0009 0.3849 0.3425 Marianna all 96 x x x yield = 1836.74673 -thrips*12 7.65422 -hardlock*1205.25120 <0.0001 0.6501 0.6426 Quincy 2004 8 x x hardlock = 0.3990 7 +thrips*0.01581 0.0230 0.6052 0.5394 Quincy 2005 8 x x hardlock = 0.1745 8 +thrips*0.03057 0.0751 0.4352 0.3411 Quincy all 16 x x hardlock = 0.2642 0 +thrips*0.02937 0.0189 0.3347 0.2872 Quincy 2004 8 x x yield = 468.27660 +hardlock*1860.10638 0.2863 0.1858 0.0501 Quincy 2005 8 x x yield = 1817.62244 -hardlock*1021.01514 0.2288 0.1841 0.0481 Quincy all 16 x x yield = 1731.20696 -hardlock*814.36697 0.1271 0.1582 0.0980 Quincy 2004 8 x x yield = 1219.43516 +thrips*29.73564 0.4113 0.1150 -0.0325 Quincy 2005 8 x x yield = 1599.82076 -thrips*13.73293 0.7689 0.0155 -0.1486 Quincy all 16 x x yield = 1437.60645 +thrips*3.13140 0.9118 0.0009 -0.0705 Quincy 2004 8 x x x yield = 489.77366 +thrips* 0.82832 +hardlock*1828.39 815 0.5981 0.1859 -0.1398 Quincy 2005 8 x x x yield = 1854.99692 +thrips*30 .95703 -hardlock*1461.6911 2 0.5227 0.2286 -0.0800 Quincy all 16 x x x yield = 1775.20536 +thrips*40 .65899 -hardlock*1277.7 9295 0.1413 0.2600 0.1461

PAGE 101

101 CHAPTER 5 SUMMARY AND CONCLUSIONS The most common (>99%) flower thrips species observed was F. tritici (eastern flower thrips), and flowers contained an average of 1.4 to 4.2 individuals. Females outnumbered males, typically by a 2:1 to 5:1 ratio. F. occidentalis (western flower thrips) was rarely found in 2003 and 2004 (2 to 3 thrips per 1000 cotton flowers), and was not present in 2005. Although slightly more common (3 to 10 thrips per 1000 cotton flow ers), the same pattern of diminishing numbers was observed in F. bispinosa (Florida flower thrips). These th ree species are all flower feeders, and likely competitors for resources. F. fusca was observed at low levels (3 to 16 per 1000 flowers) in all years. It is a foliar-feeding spec ies, and could have been found in the flowers as a result of dispersion in search of suitable host plan ts. Immature thrips were also present, and their numbers ranged from 0.05 to 0.10 per flower. Orius species, predators of thrips, were also present at a rate of 0.03 to 0.14 pe r flower. Observed ratios of Orius to thrips ranged from 1:35 to 1:220 depending on location and year. Orius probably reduced thrips numbers, but this did not lead to localized extinctions as observed on other crops. Aphis sp (aphids) were observed at levels of 1.2 to 5.3 per flower. Samples from Louisiana, Alabama, and South Ca rolina were also examined to see if they were similar to those in Fl orida. In all locations, F. tritici was most common. Their numbers ranged from 0.42 to 14.8 per flower. Louisian a in 2003 contained the highest number of F. occidentalis recorded (0.31 per flow er). One individual of F. bispinosa was found in Alabama, although they are most common to peninsular Florid a. The sex ratios varied considerably, with the highest being 1 male per 48 females. In a third of the location/years, males were more common than females. The large range of thri ps numbers and sex ratios observed may have been influenced by crop management practices. The number of immature thrips also varied

PAGE 102

102 considerably, from 0.01 to 0.96 per flower. Orius was generally uncommon, except in two instances where 0.15 and 0.21 per flower were found. The number of thrips observed weekly fluctuat ed during the growing season. In 2 years at Marianna, a rapid rise of 55 to 170% in thrips numbers was observe d for the last sampling dates. This was associated with a decline in the num ber of white flowers, and possible crowding of thrips. The rate at which thrips accumulate in flowers was also examined. Cotton flowers open between 0900 and 0930. By 1000, there were typically 4 or fewer thrips per flower. Thrips numbers increased until 1400, at wh ich point 4 to 35 thrips were present. The rate at which thrips arrived in flowers varied substantially by day and it was hypothesized this was due to weather conditions. Cool temperatures and low humidity from 0800 to 1000 was associated with more thrips at 1000. The ability of insecticide treatment to redu ce thrips numbers was also evaluated. In Marianna, FL, acephate was alternated weekly w ith spinosad reduced thrips numbers by 20 to 35%. In Quincy, FL, lambda cyhalothrin was alte rnated weekly with sp inosad, and reduced by 84 to 92%. To demonstrate the association of flower thrips with hardlock, a series of experiments was performed. Live thrips were captured from co tton flowers in research plots and placed on acidified potato dextro se agar to isolate Fusarium It was determined that between 7 and 13% of flowers contained thrips that were carrying Fusarium In another experiment, cotton flowers inside a greenhouse were inoculated with Fusarium spores, thrips, or thrips that had been exposed to Fusarium cultures. The result varied by cultiv ar, but the combination of thrips and Fusarium typically resulted in the most severe ha rdlock symptoms, while each of them alone increased severity compared to the contro l group. The combination of thrips and Fusarium also

PAGE 103

103 resulted in bolls aborting at a higher rate. In field plots, fungicide a nd/or insecticides were applied to determine the impact on thrips, hardlock and yield. In Quincy, insecticides reduced thrips numbers and sometimes reduced hardlock but did not improve yield. In Marianna, fungicides did not affect thrips and usually reduced hardloc k, but did not improve yield. Insecticides reduced thrips numbers, reduced ha rdlock more consistently and further than fungicides, and improved yield in 1 of 3 years. Combining fungicide and in secticide applications had the same impact on thrips and hardlock as di d insecticide alone, but increased yield in 2 of 3 years. Models comparing weather conditions on the day of bloom to hardlock severity in resulting bolls were evaluated. The first reported hardlock-wea ther model (Marois et al., 2004) was based on a positive association between the average temperature and relative humidity from 0700 to 1900 and hardlock severity. Although it f it the early data set well, it did not prove adequate in all locations. A s econd model, previously used for predicting thrips numbers, was also evaluated. Although it was an improvement fr om the first model, its predictive ability was still very limited. Examining the temperatures by hour revealed the ti me between midnight and 0600 to be most important for predicting hardloc k. A model was constructed using the average temperature for that time period. This model out -performed those tested previously. It was further refined by limiting it to days within the temperature range of 21 to 24C. Below that range, it was more accurate to assume hardlo ck would be above 50%, and above that range assume it would be less than 50%. The first 2 models tested had straightforward biological explanations. In the first model, temperature and relative humidity were positively associated with hardlock incidence. Such conditions would be favorable for the germ ination and growth of the causal agent, F.

PAGE 104

104 verticillioides In the second model, temperature from 0800 to 1000 was negatively associated with hardlock. This temperature range had prev iously been shown to be negatively associated with thrips numbers at 1000. By slowing the moveme nt of thrips into flow ers, it is assumed their harmful effects would be diminished, resulting in less hardlock. However, neither model fit the data as effectively as did the third. It is possi ble that cooler night temp eratures have a subtle effect on the developing flower making it less su sceptible to infection. Although the impact of temperature on thrips numbers does not appear to be the primary determinant of hardlock severity, it is possible the cool night temperatures have a re sidual effect on thrips that supplements any direct effect on the plant. The relationships between thrips numbers, hardlock severity, and yield were also explored. Regression analyses comparing thrips to hardlock were mostly significant. Among the significant ones, the adjusted R2 values ranged from 0.09 to 0.54, suggesting thrips are an important, but not exclusive, factor in hardlo ck severity. There are probably other factors influencing the probability of in fection, progression of the diseas e, and expression of symptoms after boll opening. The relationship between hard lock and yield was significant in many cases, and those location-years in whic h it wasnt were characterize d by below average yield. This suggests hardlock is one of th e many factors that can influenc e yields. Comparing thrips numbers to yield on a yearly basis showed no re lationship. This suggest s they have no direct impact on the cotton plants, and any effect is due to hardlock. These experiments have shown thrips numbers can be reduced in a field setting with insecticide applications. This reduced hardlock severity, and sometimes improved yield. Thrips in flowers were shown to be carrying Fusarium and the greenhouse study showed this to increase hardlock severity compared to flowers that were untreated or contained thrips without

PAGE 105

105 Fusarium Temperature was shown to influence both thrips numbers and hardlock severity. Flower thrips and weather conditions ar e important contributors to hardlock.

PAGE 106

106 LIST OF REFERENCES Accinelli, C., C. Screpanti, A. Vicari, P. Ca tizone. 2004. Influence of insecticidal toxins from Bacillus thuringiensis subsp. kurstaki on the degradation of glyphosate and glufosinateammonium in soil samples. Agriculture, Ecosystems, & Environment 103:497-507. Adamczyk Jr., J.J. and D.V. Sumerford. 2001. Potential factors impacting season-long expression of Cry1Ac in 13 commercial variet ies of Bollgard cotton. Journal of Insect Science 1:13. Agrawal, A.A., C. Kobayashi, and J.S. Thaler. 1999. Influence of prey availability and induced host-plant resistance on omni vory by western flower thri ps. Ecology. 80(2):518-523. Anonymous. 2006. State cotton area, yield, and production. National Cotton Council of America [Online]. Available at: http://www.cotton.org/econ/crop info/cropdata/state_data.cfm Accessed: 12/13/2006 Anonymous. 2005. Cotton production costs and re turns. National Cotton Council of America [Online]. Available at: http://www.cotton.org/econ/cropin fo/costsreturns/seaboard.cfm Accessed: 12/13/2006 Arndt, C. H. 1950. Boll rots of cotton in S outh Carolina in 1949. Pl ant Disease Reporter 34(6):176. Atakan, E. and A.F. Ozgur. 2001a. Preliminary investigation on damage by Frankliniella intonsa to cotton in the Cukurova region of Turkey. Th rips and Tospoviruses: Proceedings of the 7th International Symposium on Thysanoptera 133-140. Atakan, E. and A.F. Ozgur. 2001b. Determining the favorable sampling time for Frankliniella intonsa on cotton. Thrips and Tospoviruses: Proceedings of the 7th International Symposium on Thysanoptera 225-227. Avila, Y., J. Stavisky, S. Hague, J. Funderbur k, S. Reitz, and T. Momol. 2006. Evaluation of Frankliniella bispinosa (Thysanoptera: Thripidae) as a v ector of the tomato spotted wilt virus in pepper. Florida Entomologist 89:204-207. Bagga, H. S. 1968. Fungi associated with cott on boll rot and their fr equency. Plant Disease Reporter 52(7):582-584. Baird, R. and D. Carling. 1998. Survival of para sitic and saprophytic fung i on intact senescent cotton roots. Journal of Cotton Science 2:27-34. Batson, W.E. 2001. Boll rots. Pages 36-38 in: Compendium of Cotton Diseases, 2nd ed. Kirkpatrick, T. L., and C. S. Rothrock, eds. American Phytopatholog ical Society, St. Paul, MN. Bell, A.A. 1967. Formation of gossypol in infe cted or chemically irritated tissues of Gossypium species. Phytopathology 57:759-764.

PAGE 107

107 Beshear, R.J. 1983. New records of thrips in Georgia (Thysanoptera, Terebrantia). J. GA Entomol. Soc. 18:342-344. Bolkan, H.A., J.C. Dianese, and F.P. Cuperti no. 1979. Survival and colo nization potential of Fusarium moniliforme var. subglutinans in soil. Phytopathology 69:1298-1301. Brodbeck, B.V., J. Stavisky, J.E. Funderburk, P.C. Andersen, and S.M. Olson. 2001. Flower nitrogen status and populations of Frankliniella occidentalis feeding on Lycopersicon esculentum Entomologia Experimentalis et Applicata 99:165. Burke, J.J. 2002. Moisure sens itivity of cotton pollen; an emasculation tool for hybrid production. Agron. J. 94:883-888. Cattaneo, M.G., C. Yafuso, C. Schmidt, C.Y. Hu ang, M. Rahman, C. Olson, C. Ellers-Kirk, B.J. Orr, S.E. Marsh, L. Antilla, P. Dutilleu, Y. Carriere. 2006. Farm-scale evaluation of the impacts of transgenic cotton on biodiversity, pe sticide use, and yield. Proceedings of the National Academy of Sciences of th e United States of America 103:7571-7576. Chellemi, D.O., J.E. Funderburk, and D.W. Ha ll. 1994. Seasonal abundance of flower-inhabiting Frankliniella species (Thysanoptera: Thripidae) on wild plant species. Environmental Entomology 23:337-342. Cotten, T.K. and G.P. Munkvold. 1998. Survival of Fusarium moniliforme F. proliferatum and F. subglutinans in maize stalk residu e. Phytopathology 88:550-555. Deligeorgidis, P.N., C.G. Athanassiou and N.G. Kavallieratos. 2002. Seas onal abundance, spatial distribution and sampling indices of thrips populations on cotton; a 4-year survey from central Greece. J. Appl. Ent. 126:343-348. Edgerton, C.W. 1912. Flower infection with cotton boll rots. Phyt opathology 2: 23-27. Eger Jr., J.E., J. Stavisky, J.E. Funderbur k. 1998. Comparative toxicity of spinosad to Frankliniella spp. (Thysanoptera: Thripidae), with notes on a bioassay technique. Florida Entomologist 81(4) 547-551. Farrar, J.J. and R.M. Davis. 1991. Relationshi ps among ear morphology, we stern flower thrips, and Fusarium ear rot of corn. Phytopathology 81:661-666. Fernando, W.G.D., J.D. Miller, W.L. Seaman, K. Seifert and T.C. Paulitz. 2000. Daily and seasonal dynamics of airborne spores of Fusarium graminearum and other Fusarium species sampled over wheat plots. Can. J. Bot./Rev. Can. Bot. 78(4): 497-505. Funderburk, J., J. Stavisky, C. Tipping, D. Gorb et, T. Momol and R. Berger. 2002. Infection of Frankliniella fusca (Thysanoptera: Thripidae) in P eanut by the Parasitic Nematode Thripinema fuscum (Tylenchidae: Allantonematidae) Environ. Entomo l. 31(3): 558-563. Gipson, J.R. 1982. Integration of events in boll development: temperature effects. In Cotton Physiology: a treatise. USDA.

PAGE 108

108 Gipson, J.R. and H.E. Joham. 1968. Influence of night temperature on growth and development of cotton ( Gossypium hirsutim L.) fiber properties. Agron. J. 60:296-298. Gitaitis, R. D., R. R. Walcott M. L. Wells, J. C. D. Perez, and F. H. Sanders. 2003. Transmission of Pantoea ananatis causal agent of center rot of onion, by tobacco thrips, Frankliniella fusca Plant Dis. 87:675-678. Guinn, G. 1979. Hormonal relations in flowerin g, fruiting and cut-out. P.265-276. Proceedings Beltwide Cotton Prod. Res. Conf. Heitholt, J.J. and W.R. Meredi th, Jr. 1998. Yield, flowering, and l eaf area index of okra-leaf and normal-leaf cotton isolin es. Crop Sci. 38:643-648. Leite, B., J.J. Marois, D.L. Wright, E.A. Osekre, T.W. Katsvairo, and P.J. Wiatrak. 2006. Symptom Development and Characterization of Cotton Hardlock: Focus on Fumonisin Production and Stored Lipids. In Annual mee tings abstracts [CD-RO M]. ASA, CSSA, and SSSA, Madison, WI. Marin, S., V. Sanchis, A. Teixido, R. Saenz, A.J. Ramos, I. Vinas and N. Magan. 1996. Water and temperature relations and microconidial germination of Fusarium moniliforme and Fusarium proliferatum from maize. Can. J. Microbiol. 42(10):1045-1050. Marois, J.J., D.J. Mailhot, D.L. Wright and P. J. Wiatrak. 2006. Application of pesticides to control fusarium hardlock of cotton in Fl orida. P.121-125 in: Proceedings of Beltwide Cotton Production Research Conferences. National Cotton Council of America. Marois, J.J., D.L. Wright and P.J. Wiatrak. 2005. Update of 2004 Fusarium hardlock research in Florida. P.152-156 in: Proc. Be ltwide Cotton Prod. Res. Conf. Marois, J.J. and D.L. Wright. 2004. Etiology, epidem iology, and control of Fusarium hardlock of cotton in the southeast: the possibilities. P.331-336 in Proc. Beltwide Cotton Prod. Res. Conf. Marois, J.J., D.L. Wright and P.J. Wiatrak. 2002. Association of Fusarium sp. with hardlock of cotton in the southeastern U.S. Pr oc. Beltwide Cotton Prod. Res. Conf. Mauney, J.R. 1966. Floral initiation of upland cotton Gossypium hirsutum L. in response to temperature. J. Exp. Bot. 17:452-459. May, O.L. and K.E. Lege. 1999. Developmen t of the world cotton industry. p. 65-98. In C. Smith and J. Cothren (ed.) Cotton: orig in, history, techno logy, and production. McDonald, J.R., J.S. Bale and K.F.A. Walters. 1997. Rapid cold hardening in the western flower thrips Frankliniella occidentalis J. Insect Physiol. 43:759-766. McKinion, J.M., D.N. Baker, F.D. Whisler and J.R. Lambert. 1989. Application of GOSSYM/COMAX system to cotton crop management. Agr. Syst. 31:55-65.

PAGE 109

109 Michailieds, T. J., and Morgan D. P. 1998. Spread of endosepsis in Calimyearna fig orchards. Phytopathology 88:637-647. Omer, A.D., J. Granett, R. Karban and E.M. Villa. 2001. Chemically-indu ced resistance against multiple pests of cotton. International j ournal of pest management 47(1):49-54. Ooka, J.J. and T. Kommedahl. 1977. Wind and rain dispersal of Fusarium moniliforme in corn fields. Phytopathology 67:1023-1026. Oosterhuis, D.M. and J. Jernstedt. 1999. Mo rphology and anatomy of the cotton plant. p. 175206. In Cotton: origin, history, t echnology, and production. C. Smith and J. Cothren (ed.). Palmateer, A.J., K.S. McLean, G. Morgan-J ones and E. van Santen. 2004. Frequency and diversity of fungi colonizing tissues of upland cotton. Mycopathologia 157:303-316. Pappu, H.R., J.W. Todd, A.K. Culbreath, M.D. Bandla and J.L. Sherwood. 1998. First report on the multiplication of tomato spotted wilt tospovirus in tobacco thrips, Frankliniella fusca Plant Disease 82:1282. Pearsall, I.A. 2002. Daily flight activity of the western flower thrips (Thysan., Thripidae) in nectarine orchards in British Columbia, Canada. J. Appl. Ent. 126:293. Pickett, C.H., L.T. Wilson, and D. Gonzalez. 1988. Population dynamics and withinplant distribution of the western fl ower thrips (Thysanoptera: Thripidae), and early-season predator of spider mites infesting co tton. Environ. Entomol. 17(3):551-559. Pinckard, J. A., and S. J. P. Chilton. 1966. The economic importance and classification of cotton boll rots in Louisiana. Proc. La. Acad. Sci. 29:12-22. Pollock, E.G. and W.A. Jensen. 1964. Cell development during ear ly embryogenesis in Capsella and Gossypium Am. J. Bot. 51:915-921. Quisenberry, J.E., and R.J. Kohel. 1975. Growth and development of fiber and seed in upland cotton. Crop Sci. 15:463. Ramachandran, S., J. Funderburk, J. Stavis ky and S. Olson. 2001. Population abundance and movement of Frankliniella species and Orius insidiosus in field pepper. Agricultural and Forest Entomology 3:129-137. Reitz, S.R. 2002. Seasonal and within plant distribution of Frankliniella thrips (Thysanoptera: Thripidae) in north Florida tomato es. Florida Entomologist 85:431-439. Reitz S.R., E.L. Yearby, J.E. Funderburk, J. Stavisky, M.T. Momol and S.M. Olson. 2003. Integrated Management Tactics for Frankliniella Thrips (Thysanoptera: Thripidae) in Field-Grown Pepper. Journal of Economic Entomology 96:1201-1214(14).

PAGE 110

110 Reitz, S.R., J.E. Funderburk, E.A. Hansen, I. Baez, S. Waring, S. Ramachandran. 2001. Interspecific variation in behavi or and its role in thrips eco logy. Thrips and Tospoviruses: Proceedings of the 7th Internationa l Symposium on Thysanoptera 133-140. Roberts, R.G. and J.P. Snow. 1984. Hi stopathology of cotton boll rot caused by Colletotrichum capsici Phytopathology 74:390-397. Rohrbach, K.G. and G. Taniguchi. 1984. Effect s of temperature, moisture, and stage of inflorescence development on infection of pineapple by Penicillium funiculosum and Fusarium moniliforme var. subglutinans Phytopathology 74:995-1000. Roncadori, R.W., S.M. McCarter and J.L. Crawford. 1975. Evaluation of various control measures for cotton boll rot. Phytopathology 65:567-570. Roncadori, R.W. 1969. Fungal invasion of developing cotton bolls. Phytopathology 59:13561359. Roussopoulos, D., A. Liakatas and W.J. Wh ittington. 1998. Controlled-temperature effects on cotton growth and development. The J ournal of Agricultura l Science 130:451-462. Salgado, V.L. 1998. Studies on the mode of ac tion of spinosad: insect symptoms and physiological correlates. Pesticide Bi ochemistry and Physiology 60:91-102. Sakimura, K. 1963. Frankliniella occidentalis (Thysanoptera:Thripidae), a vector of the tomato spotted wilt virus, with special reference to the colorforms. Ann. Entomol. Soc. Am. 55:387-389. Sanders, D.E. and J.P. Snow. 1978. Dispersal of airborne spores of boll-rotting fungi and the incidence of cotton boll ro t. Phytopathology 68:1438-1441. Sanders, D. E., and J. P. Snow. 1977. A possible source of inoculum of bo ll rotting fungi. Proc. Beltwide Cotton Prod. Res. Conf. Seebold, K. and R. Kemerait. 2005. Results from the 2004 regional hardlock project effects on disease and yield. Page 159 in: Proceedings of Beltwide Cotton Production Research Conferences. National Cott on Council of America. Seebold, K., R.C. Kemerait, J. Marois, D. Wright B. Padgett, J. Mueller, K. McLean, and T. Estes. 2004. Results from the 2003 regiona l hardlock project. Pages 337-340 in: Proceedings of Beltwide Cotton Production Research Conferences. National Cotton Council of America. Snow, J.P. and D.E. Sanders. 1979. Role of ab scised cotton flowers, bolls and squares in production of inoculum by boll-rotting Fusarium spp Plant Disease Reporter 63:288-289. Sparnicht, R.H. and R.W. Roncadori. 1972. Fusa rium boll rot of cott on: pathogenicity and histopathology. Phytopathology 62:1381-1386.

PAGE 111

111 Sprenkel, R.K. 2005. Cotton Plant and Pest Mon itoring Manual for Florida. University of Florida, Extension Publication ENY-830. Staggenborg, S.A., R.J. Lascano and D.R. Krieg. 1996. Determining cotton water use in a semiarid climate with the GOSSYM cotton simulation model. Agronomy J. 88:689-716. Stavisky, J., J. Funderburk, T. Momol, D. Gorbet. 2001. Influence of parasitism by Thripinema fuscum on dynamics of local populations of Frankliniella fusca Thrips and Tospoviruses: Proceedings of the 7th Internationa l Symposium on Thysanoptera 141-143. Stewart, J.McD. 1975. Fiber initiation on the cotton ovule ( G. hirsutum ). Am. J. Bot. 62:723730. Studebaker, G. E. and T.J. Kri ng. 2000. Lethal and sublethal effect s of early season insecticides on insidious flower bug ( Orius insidiosus ): an important predator in cotton. P.221-225 in Proceedings of the 2000 Cotton Research Meeting. Subbarao, K.V. and T.J. Michailides. 1995. E ffects of temperature on isolates of Fusarium moniliforme causing fig endosepsis and Aspergillus niger causing smut. Phytopathology 85:662-668. Toapanta, M.A., J.E. Funderburk and D. Chellemi. 2001. Development of Frankliniella species (Thysanoptera:Thripidae) in re lation to microclimatic temper atures in vetch. Journal of Entomological Science 36:426-437. Toapanta, M., J. Funderburk, S. Webb, D. Chellemi and J. Tsai. 1996. Abundance of Frankliniella spp. (Thysanoptera: Thripidae) on winter and spring host plants. Environmental Entomology 25:793-800. Tommasini, M.G. and G. Nicoli. 1993. Adult activity of four Orius species reared on two preys. Bulletin of the International Organization of Biological Control/Western Pacific Region 16:181-184. Torres, M.R., A.J. Ramos, J. Soler, V. Sanc his and S. Marin. 2003. SEM study of water activity and temperature effects on the initial growth of Aspergillus ochraceus Alternaria alternata and Fusarium verticillioides on maize grain. Interna tional Journal of Food Microbiology 81:185-193. Trichilo, P.J. and T.F. Leigh. 1988. Influence of resource quality on the re productive fitness of flower thrips (Thysanoptera: Thripidae). Anna ls of the Entomological Society of America 81:64-70. van de Wetering, F., J. Julshof, K. Posthuma, P. Harrewijn, R. Goldbach and D. Peters. 1998. Distinct feeding behavior between sexes of Frankliniella occidentalis results in higher scar production and lower tospovirus transmission by females. Entomologia Experimentalis et Applicata 88:9-15.

PAGE 112

112 Wright, D.L. and B.J. Brecke. 2002. 2002 Cotton defo liation and harvest aid guide. University of Florida, Extension. Publication SS-AGR-181. Wackers, F.L. and C. Bonifay. 2004. How to be sweet? Extrafloral nectar allocation by Gossypium hirsutum fits optimal defense theo ry predictions. Ecology 85:1512-1518. Wright, D.L., J.J. Marois, P.J. Wiatrak and T. Katsvairo. 2004. History and overview of the hardlock problem in humid areas of the d eep south. Proc. Beltwide Cotton Prod. Res. Conf.

PAGE 113

113 BIOGRAPHICAL SKETCH Daniel Joseph Mailhot was bor n in Tallahassee, FL in 1981. After graduating from Leon High School in 1999, he attended the University of Fl orida. Daniel graduated with a Bachelor of Science degree in plant science in the summer of 2002. He accepted an assistantship at the University of Florida under the supervision of James Marois in the Plant Pathology Department, and started graduate school in fall of 2002. His research has focused on hardlock of cotton and its relationship to flower thrips and weather co nditions. Daniel will gr aduate in spring of 2007.


Permanent Link: http://ufdc.ufl.edu/UFE0018161/00001

Material Information

Title: Relationship of Flower Thrips to Hardlock of Cotton
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0018161:00001

Permanent Link: http://ufdc.ufl.edu/UFE0018161/00001

Material Information

Title: Relationship of Flower Thrips to Hardlock of Cotton
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0018161:00001


This item has the following downloads:


Full Text





RELATIONSHIP OF FLOWER THRIPS TO HARDLOCK OF COTTON


By

DANIEL JOSEPH MAILHOT













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

2007

































Copyright 2007

by

Daniel Joseph Mailhot

































To those who establish the foundation for future advances.









ACKNOWLEDGMENTS

My heartfelt thanks go to my advisor, Dr. James Marois, for his support of my research

and his advice on writing this dissertation. I also thank the other members of my supervisory

committee, Dr. Daniel Chellemi, Dr. Raymond Galaherr, and Dr. James Kimbrough, for their

advice and patience. I thank Dr. David Wright for sharing his knowledge of cotton. I thank Dr.

Joseph Funderburk, Dr. Julie Stavisky and Dr. Dean Paini for sharing their expertise regarding

flower thrips. Dr. Kathy Lawrence at Auburn University, Dr. John Mueller at Clemson

University and Dr. Boyd Padgett at Louisiana State University generously provided cotton

flowers to confirm my findings in Florida. I thank my lab colleagues Dr. Tawainga Katsvairo,

Dr. Breno Leite, Enoch Osekre and Francis Tsigbey for helpful insights and good conversations.

I thank Wayne Branch, Brian Kidd and Pawel Wiatrak for maintaining my research plots. I

thank Doug Hatfield, John L. Smith and Henry McCrone for their assistance with the weather

model. This research was supported by Cotton Incorporated and Cerexagri. Their willingness to

provide funding is sincerely appreciated. Finally, I thank my family for their unwavering

support and encouragement during my years in graduate school.









TABLE OF CONTENTS

page

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

LIST OF TABLES ................... .... ...................................................................................................

LIST OF FIGURES .................................. .. ..... ..... ................. .9

A B S T R A C T ......... ....................... ............................................................ 10

CHAPTER

1 INTRODUCTION AND LITERATURE REVIEW ............................................................12

C o tto n ..................................................................................................................................... 1 2
F lo w e r th rip s ..................................................................................................................... 1 5
H ard lo ck ...........................................................................2 0

2 THRIPS ON FLOWERS OF COTTON ............ ..... .............................. 26

Introduction ................. .................................... ............................26
M materials and M methods ...................................... .. .......... ....... ...... 30
Field Plots ............. ............ .. .................... ........ ....... ................. .. 30
Sam pling of Thrips .............. ................. ........... .......... ........... ..3... 31
Identification of Thrips ......................................... .............. ........ .... .. 32
M ulti-State Species Survey ..................................................................... ..................32
Thrips A ccum ulation in Flow ers ......................................................................... ..... 33
R e su lts ................... ...................3...................3..........
Prevalent Species................................................ 33
Sim ilarities A cross the Southeast ................................. ........................... ............. 34
Impact of Insecticide Treatments on Flower-Inhabiting Insects...............................35
Changes in Thrips Numbers During the Growing Season ...........................................36
A association of Thrips and Orius ..................................................... ...................36
T hrips A ccum ulation in Flow ers ......................................................................... ... 38
D isc u ssio n ................... ........................................................... ................ 3 9

3 RELATIONSHIP OF FLOWER THRIPS TO FUSARIUM HARDLOCK ..........................59

Introdu action ................... .......................................................... ................. 59
M materials and M ethods ............... .... ...................................................................................... 62
Isolation of Fusarium from Thrips ............. ................................................................62
Influence of Thrips and Fusarium on Hardlock Incidence in Greenhouse...................63
Effect of Fungicides and Insecticides on Hardlock ............. ........................................65
R e su lts................. ... .. ................. ............................................................. 6 6
Isolation of Fusarium from Thrips .................................. ............................................66
Influence of Thrips and Fusarium on Hardlock Incidence in Greenhouse....................66









Effect of Fungicides and Insecticides on Thrips, Hardlock, and Yield.........................67
D iscu ssion ............ ......... .................................... ..................................69

4 RELATIONSHIP OF HARDLOCK TO WEATHER, THRIPS, AND YIELD ....................78

In tro d u c tio n ....................................................................................................................... 7 8
M materials and M methods ...................................... .. ......... ......... .....80
Results ................... ...... ................ .......... ... 81
W weather M models of H ardlock Severity ........................................ ...................... ........ 81
Connections between Thrips, Hardlock, and Yield............................................. 85
D iscu ssio n ................... ...................8...................6..........

5 SUMMARY AND CONCLUSIONS........................................................ ............. 101

L IST O F R E F E R E N C E S .................................................................................... ...................106

B IO G R A PH IC A L SK E T C H ......................................................................... ........................ 113





































6









LIST OF TABLES


Table page

2.1 Mean number of inhabitants per flower across all treatments in Quincy, FL....................44

2.2 Mean number of inhabitants per flower across all treatments in Marianna, FL................45

2.3 Mean number of inhabitants per flower in Winnsboro, LA.......................................45

2.4 Mean number of inhabitants per flower in Fairhope, AL .........................................46

2.5 Mean number of inhabitants per flower in Blackville, SC ........................................46

2.6 Mean numbers of flower inhabitants by treatment in Quincy, FL..................................47

2.7 Mean number of inhabitants per flower by treatment in Marianna, FL.............................47

2.8 Number of males per female by treatment for each year of sampling ..............................48

2.9 Occurance of adult thrips and Orius in flowers sampled........................ .....................55

2.10 Interspecific association of thrips and Orius ............................ .....................55

2.11 Interspecific covariance of thrips and Orius ............................... ................................. 55

2.12 Correlation of mean weather conditions to thrips numbers at 1000 .............................57

2.13 Relationship of mean temperature from 0800 to 1000 and relative humidity from
1900 to 1000 to thrips num bers at 1000........................................................................ ....58

2.14 Rainfall during the sampling times shown in Figure 2.4.......................................... 58

3.1 Proportion of flowers containing thrips-carried Fusarium..................... .............. 73

3.2 Results of greenhouse hardlock experiments .............................. ................................. 73

3.3 Effect of fungicides and insecticides on thrips, hardlock, and yield in Quincy, FL..........73

3.4 Effect of fungicide and insecticide treatments on thrips, hardlock, and yield in
M arian n a F L .......................................................................... 7 3

4.1 Relationship between average temperature and relative humidity from 0700 to 1900
on the day of bloom and hardlock severity for that day ............................................90

4.2 Relationship between average temperature from 0700 to 1900 on the day of bloom
and hardlock severity for that day ......................................................................... ... ... 90









4.3 Relationship between average relative humidity from 0700 to 1900 on the day of
bloom and hardlock severity for that day ............................................... ............... 91

4.4 Relationship between average temperature from 0800 to 1000 on the day of bloom
and hardlock severity for that day ......................................................................... ...91

4.5 Relationship between average relative humidity from 0800 to 1000 on the day of
bloom and hardlock severity for that day ............................................... ............... 92

4.6 Relationship between average temperature and relative humidity from 0800 to 1000
on the day of bloom to hardlock severity for that day ....................................................92

4.7 Association of temperature and hardlock severity by hour for each location ..................93

4.8 Relationship between average temperature from 0000 to 0600 on the day of bloom to
hardlock severity for all days.................................................. ............................... 94

4.9 Relationship between average temperature from 0000 to 0600 on the day of bloom to
hardlock severity for temperatures between 21 and 250C ............................................97

4.10 Mean and standard deviation of models when predictions are applied to data .................98

4.11 Validation of the model by random sampling and applying predictions to data ...............98

4.12 Relationship of temperature from 0000 to 0600 and estimated thrips numbers to
hardlock in M arianna, F L ......................................................................... ................... 99

4.13 Relationships between thrips, hardlock and yield on a per plot, seasonal basis.............100









LIST OF FIGURES


Figure page

2.1 Number of males per female by treatment on each day of sampling.............................49

2.2 Variation in adult Orius (minute pirate bug) numbers over time ....................................51

2.3 V ariation in thrips num bers over tim e ........................................ ......................... 53

2.4 Increase in thrips numbers during first day of bloom in Quincy, FL ..............................56

3.1 Thrips numbers and hardlock incidence by year in Marianna, FL ....................................74

3.2 Thrips numbers and hardlock incidence by treatment in Marianna, FL..........................75

3.3 Thrips numbers and hardlock for all treatments and years in Quincy, FL ......................76

3.4 Thrips numbers and hardlock in Quincy, FL........................ ............................ 77

4.1 Cotton boll exhibiting symptoms of hardlock ......... ........................................89

4.2 Mean temperature from 0000 to 0600 and hardlock severity ................. ............ .......95

4.3 Distribution of mean temperatures from 0000 to 06:00 for days in the climate model.....96

4.4 Incidence of hardlock during the growing season for days evaluated............. ...............96

4.5 D ays in m odel w ith >10 bolls recovered ........................................ ....................... 97

4.6 Prediction residuals for days with >10 bolls, 21 to 250C (N=49) ...................................98









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

RELATIONSHIP OF FLOWER THRIPS TO HARDLOCK OF COTTON

By

Daniel J. Mailhot

May 2007

Chair: James J. Marois
Major Department: Plant Pathology

Hardlock is a limiting factor for cotton yields in Florida, and appears as a failure of the

fiber to expand outward following boll opening. Annual losses are typically 30 to 70% of the

crop. Prior research has identified Fusarium verticillioides as a causal agent. It is believed to

infect the flowers on the day of bloom, and sustain itself within the developing boll. It was

hypothesized flower thrips may increase the chance of infection by carrying the spores into

flowers or allowing better access by the pathogen due to their feeding damage.

To determine the extent to which thrips in the field transport F. verticillioides, thrips were

captured and placed on media to isolate Fusarium. Approximately 10% of cotton flowers

contained thrips that were carrying Fusarium. Frankliniella tritici was the most commonly

found thrips species, representing more than 99% of individuals. Similar results were noted in

samples from Louisiana, Alabama, and South Carolina. It was also determined that the average

relative humidity from 1900 on the day prior to 1000 on the day of sampling was negatively

associated with the number of thrips in flowers at 1000.

Field trials were performed to test the ability of insecticide and fungicide applications to

reduce thrips numbers and hardlock. Insecticides reduced thrips number and hardlock severity,

and sometimes improved yields. The number of thrips in a plot was positively associated with









hardlock severity. Greenhouse experiments were performed to test the ability ofFusarium

inoculation, thrips, or thrips exposed to Fusarium to cause hardlock. Thrips exposed to

Fusarium resulted in the most severe symptoms. The effect of weather on hardlock was also

explored. The average temperature between 0000 and 0600 on the day of bloom was negatively

associated with hardlock severity in the resulting bolls. These findings demonstrate hardlock is

associated with flower thrips and can be reduced by managing thrips numbers. Hardlock

incidence is also influenced by temperature, and this information may allow control measures to

be used when they would be most effective.









CHAPTER 1
INTRODUCTION AND LITERATURE REVIEW

Cotton

Cotton, Gossypium hirsutum L., is an important crop in the southeastern US. Diseases

and insects affecting its production are well documented, and it has been a topic of research since

the 1800s. In 2003 cotton was produced on 38,000 hectares in Florida. The harvest of 117,000

bales each weighing 218 kg. yielded a market value of $33.49 million. Since 1995, yields have

usually ranged from 450-670 kg/ha. These lower yields have been attributed to hardlock, a

situation in which cotton fiber fails to fluff out of the boll after opening. From the early 1980s

through the early 1990s, Florida yields were close to 785 kg/ha, and peaked in 1984 at 1105

kg/ha (Anonymous, 2005). Yields in Florida tend to be lower than other areas of the Southeast,

resulting in smaller profit margins, and the potential for greater gains from new research.

Lower yields have occurred despite the release of improved varieties. Many of these

varieties have been genetically modified to produce the Bacillus i//Il illgie'lli\ (Bt) endotoxin and

to be resistant to glyphosate (Round Up Ready or RR). The most commonly used Bt toxin in

cotton is the CrylAc protein, and it is highly effective against lepidopterans. Bt toxin expression

is influenced slightly by parent variety background, but is sufficient for field conditions

(Adamczyk and Sumerford, 2001). In addition to reducing insect damage to cotton, it has also

allowed a substantial reduction in the number of insecticide applications during the season

(Cattaneo, 2006). Resistance to glyphosate permits the herbicide to be used as a broad-spectrum

herbicide while cotton seedlings are in the field. This is advantageous, since it allows control of

broad-leaf weeds which would otherwise be difficult and require several other materials in a

conventional control program. Absorption of the compound by humans is low, it is not

metabolized, and does not accumulate in tissues. It is non-carcinogenic and has no impact on









fertility or reproductive parameters. Breakdown of the compound in the environment is also

fairly rapid, with a half-life of 14 days under typical conditions (Accinelli, 2004).

The type of cotton produced in Florida and the southeastern US is known as "upland"

cotton (G. hirsutum). This is in contrast to the long-staple Pima and Egyptian cottons (G.

barbadense) that are sometimes produced in the southwestern US. Prior to 1920, Sea Island

cotton (G. barbadense) was produced on the Florida peninsula and coastal islands of Georgia

and the Carolinas. It was limited to these locations due to the long growing-season required for

its production, and is still produced in the Caribbean. The eventual arrival of the boll weevil

made its production unprofitable, and the industry collapsed. However, Sea Island production in

the US preceded that of upland cotton, and its high profitability formed the basis for expansion

of upland cotton into the interior of the southeast (May and Lege, 1999).

The profitability of cotton farming varies substantially from year to year. In 2003, the

estimated return after all expenses was $236.62 per hectare, in the Southeast. From 1997 to

2004, only 3 of 8 years showed a profit, with total losses exceeding total profits. Production

expenses, yields and prices received vary annually. In 2003, total per hectare operating costs

were $793.14 and total allocated costs were $449.36. A lint yield of 932 kg/ha at $1.41/kg and a

seed yield of 1507 kg/ha. at $0.09/kg yielded the $236.62/ha quoted above. The estimated farm

size for the estimates listed above was 535 acres. Florida yields are typically lower than the rest

of the cotton belt, and reducing the previous quoted yield to the actual planted yield for 2003 of

597 lbs/ac. results in a loss of approximately $123/ha (National Cotton Council of America,

2005, 2006).









Cotton is typically planted between April 20 and June 1. On average, emergence occurs at

7 days after planting (DAP), first square at 39 DAP, and first bloom at 62 DAP. Development is

determined in large part by growing degree days, so warmer weather leads to faster maturation.

The normal developmental sequence of cotton bolls has been thoroughly documented.

Cotton flowers require approximately 35 days to develop from carpels to anthesis, the stage of

opening. The square appears 10-14 days into the process, and pollen mother cells are

undergoing meiosis by days 13 and 14. Floral nectaries begin developing around day 10. Ovule

number, which determines the number of locules per boll, is thought to be influenced by

environmental factors, although varieties also differ (Gipson, 1982). On the day of bloom,

indicated by a white flower, pollen germinates on the stigma and the pollen tube begins growing

toward the ovary. This growth requires 12 to 30 hours to reach the ovary and accomplish

fertilization.

By the second day of bloom, the flower has changed from white to a red-purple color. The

flower usually detaches from the plant on the 3rd or 4th day after bloom. The zygote begins cell

division 4 to 5 days after bloom (Pollock and Jensen, 1964). The initiation of flowering is

primarily determined by the accumulated growing degree-days since planting. This is expressed

by plants in warmer conditions initiating the first bloom on a lower branch (Roussopoulos et al,

1998). Some cotton varieties are photoperiod-sensitive, but they are not grown commercially in

the US. Water stress prior to flowering was observed to increase the subsequent rate of

flowering (Guinn, 1979). High night temperatures (250C) have been reported to delay the onset

of flowering (Mauney, 1966). Epidermal cells on the surface of the developing embryo undergo

a period of extreme elongation for approximately 20 days. These elongated cells attain a length

of 25 to 35 mm (Quisenberry and Kohel, 1975). They later enter a phase of secondary cell wall









thickening. At maturity the cells dry, leaving a hollow tube of cellulose which is utilized as

fiber. The time required for this process is strongly influenced by temperature (Gipson and

Joham, 1968). When about 60% of bolls have opened, a chemical defoliant is applied to the

plants to speed development of unopened bolls and prepare the crop for harvest. Plants are

defoliated to reduce contaminants in the harvested fiber, and this is performed 10 to 14 days

before harvest. Approximately 155 days are required from planting till harvest in Florida

(Wright and Brecke, 2002).

Cotton flowers provide many food resources for insects. Pollen provides a source of

protein for egg development, resulting in increased fecundity (Trichilo and Leigh, 1988). Nectar

is also an important resource for attracting pollinators and predatory insects to cotton plants

(Wackers and Bonifay, 2004). In cotton, new flowers are rapidly colonized by thrips. Their role

in cotton production and biology has not been investigated.

Flower thrips

Flower thrips, mainly Frankliniella tritici (eastern flower thrips), but also F. occidentalis

(western flower thrips), and F. bispinosa (Florida flower thrips) rapidly colonize cotton flowers

immediately after opening. More thrips accumulate as the day progresses, by the end of which

flowers may contain 10-40 thrips. This occurs in many production areas, although the specific

species involved may vary.

Thrips are a frequent problem on cotton seedlings in the US, resulting in distortion of

expanding leaves, and small discolored spots. The most common species are Frankliniella

occidentalis (western flower thrips), Frankliniellafusca (tobacco thrips) and Thrips tabaci

(onion thrips). Thrips control is usually achieved using a granular insecticide at planting. Foliar

insecticide applications are used if more than 2 to 3 thrips are present per plant and damage is

observed (Sprenkel, 2005). Applications ofjasmonic acid to cotton seedlings can reduce thrips









feeding by 80%, although leaf area is also reduced by 28% (Omer et al., 2001). However, this is

not used as a management technique. Thrips feeding is generally not a problem when the plants

are more mature. However, severe damage was reported in Turkey (Atakan and Ozgur, 200 la)

by Frankliniella intonsa on mature cotton plants. Although more than 350 thrips were observed

per flower, pollination was not adversely affected. Feeding by thrips larvae resulted in boll

shedding, although ovipositioning by females in flower parts had a larger impact.

Several studies have examined the tendency of flower thrips to disperse in search of food

resources. In British Columbia, F. occidentalis dispersal was shown to occur when windspeed

was less than 15 km/h, although dispersal was most likely to occur in the absence of wind

(Pearsall, 2002). In Turkey, F. intonsa was shown to follow a similar pattern. Red flowers were

found to contain their highest numbers of thrips at 05:30, after which their numbers fell steadily

until 11:30. This dispersal coincides with the opening of new white flowers, which were shown

to accumulate between 20 and 200 individuals. The authors also suggested red flowers could

serve as a refuge from insecticide applications (Atakan and Ozgur, 2001b).

Several flower-inhabiting thrips species are present in north Florida. In an 18-month study

of 37 wild plant species, 78% of thrips were adults, and 87% of adults were from the genus

Frankliniella. The most common species were F. tritici, F. bispinosa, F. occidentalis and F.

fusca. The relative contributions of each species fluctuated substantially during the study.

Cotton flowers were not examined, but during its bloom period of June, July and August, F.

tritici and F. bispinosa were the most common species on wild hosts (Chellemi et al., 1994).

The primary influence on Frankliniella populations appeared to be the availability of suitable

flowers. F. occidentalis populations increased more rapidly than other species, probably due to

its wider plant host range. It is non-native, and was not reported in the area until 1981 (Beshear,









1983). This is prior to the period of high yields (and presumably low hardlock) from the early

1980s to 1990s, so its arrival was probably not involved in the increased hardlock experienced

since that time.

Thrips numbers are affected by predatory insects and parasites. Orius insidiosus, the

minute pirate bug, is a predator of flower thrips that can consume 12.5 thrips per day

(Tommasini and Nicoli, 1993). If a sufficient number of Orius are present in an area, localized

extinction of flower thrips may occur. Orius will remain in the area, feeding on extrafloral

nectaries or pollen and preventing the thrips population from rebounding. This diet is sufficient

to allow development and oviposition by Orius. Small hair tufts called domatia are present on

the underside of leaves of some plants, and are associated with larger numbers of predatory

insects. However, cotton does not produce these structures. A parasitic nematode Thripinema

fuscum was shown to infect F. fusca in north Florida. Among F. fusca found on peanut, as many

as 51 and 67% of females on certain sampling dates were found to be infected, resulting in

sterility. T fuscum infection ofF. tritici and F. occidentalis was far less common, occurring in

only 2% of individuals (Funderburk et al., 2002; Stavisky et al., 2001; Ramachandran et al.,

2001).

Species and sex differences have been observed in thrips behavior. On greenhouse pepper

plants, movement ofF. occidentalis was limited, while F. tritici and F. bispinosa were found to

disperse relatively rapidly. Males of all three species were shown to be more mobile among

plants (Ramachandran et al., 2001). Females ofF. occidentalis have been shown to spend more

time feeding and produce more feeding-associated scars on petunias than do males (van de

Wetering et al., 1998). These differences in behavior resulted in F. tritici and F. bispinosa being

less susceptible to predation than F. occidentalis (Ramachandran et al., 2001.









Thrips species also differ in their response to insecticides. F. occidentalis populations

have been shown to increase following applications of acephate and esfenvalerate, while

spinosad causes a reduction. However, acephate and esfenvalerate were highly toxic to F. tritici

and F. bispinosa, while spinosad was less effective (Reitz et al., 2003). Other studies have

shown spinosad to be equally effective against all three species (Eger et al., 1998). The

effectiveness of insecticides on thrips numbers can be complicated by its effect on predators.

Several insecticides have been evaluated for both lethal and sub-lethal effects on Orius which

could affect its reproduction. Spinosad was found to have no effect, lethal or sub-lethal, on

Orius populations (Studebaker and Kring, 2000). Cyhalothrin resulted in high mortality, but no

sub-lethal effect was observed in survivors (Studebaker and Kring, 2000). Acephate was not

tested, but is a broad-spectrum organophosphate which affects a larger number of arthropods.

The development rate of thrips is dependant on temperature (Lewis, 1973). It can best be

modeled by the use of growing degree days (GDD), which is also used for predicting plant

maturity (Toapanta, et al., 1996). This method was later confirmed and refined by Toapanta et

al. (2001), demonstrating it is essential to use temperature measurements from within the plant

canopy (where developing thrips are located) rather than above the plants. The life history of

flower thrips differs slightly by species. Development time ofF. tritici is approximately one day

shorter than that ofF. occidentalis (approximately 8.5 vs. 9.5 days). Oviposition rates and the

number of offspring produced during the female lifetime are not statistically different. However,

F. tritici does produce a larger portion of offspring early in its adult life. Despite these

advantages, F. tritici does not show a greater rate of population increase in the field (Reitz et al.,

2001).









Thrips possess only one mandible which is used for cutting into plant tissue, and a stylet

through which food is drawn. This results in open wounds to the plant, which could allow easier

penetration of the tissue by pathogens. Females ofF. occidentalis have been shown to feed more

frequently and intensely than males, resulting in more tissue damage (van de Wetering et al.,

1999). Pickett et al. (1988) reported 68% of adult F. occidentalis found on cotton plants

occurred on fruiting structures, with most of these occurring in the flower itself. This is

consistent with the concept that cotton pollen may be preferable to leaves as a food source for

flower thrips (Agrawal et al., 1999). In tomato, amino acid analysis shows phenylalanine content

to be associated with thrips numbers (Brodbeck et al, 2001).

Farrar and Davis (1991) investigated the relationship between F. occidentalis and fusarium

ear rot of corn. Fusarium ear rot of corn is caused by Fusarium verticillioides, and can result in

large yield losses in some years in addition to contamination of the crop with mycotoxins.

Disease incidence had been connected previously to husk tightness and insect damage.

Insecticide applications reduced the numbers ofF. occidentalis observed and the incidence of

disease. They concluded that thrips may be acting as vectors ofF. verticillioides or promoting

infection as wounding agents by feeding on the plant tissue.

Several observations regarding hardlock suggest flower thrips might be involved. First,

there has been an increase in hardlock in recent years. Although hardlock rates have not

historically been measured, this view is held by some cotton observers and is confirmed by the

relatively lower yields experienced from the late 1990s (National Cotton Council of America,

2005). This coincides with reduced spraying of insecticides during the flowering stage, due to

the use of genetically modified varieties producing the Bacillus iilnl iigien\i\ endotoxin. Large

numbers of flower thrips often occur within the flowers. Relatively higher rates of hardlock are









observed along field margins, and thrips numbers sometimes follow this same pattern. However,

hardlock was reported to have been one of the major factors taking cotton out of the Southeast

along with boll weevils in the last half of the 20th century (Wright et al., 2004).

Hardlock

Hardlock is characterized as a failure of the cotton fiber to expand outward after boll

opening. Instead, it remains in compressed locules, similar to an orange slice, which may have a

grey or occasionally faint pink color. These compressed locules are frequently missed or

knocked to the ground by mechanical harvesters. The resulting yield loss often ranges from 20

to 60%, depending on the year. Until this dissertation, hardlock had not been formally

recognized as a disease, although it could be considered a subset of the boll rot complex.

Hardlock is most severe along the gulf coast, possibly due to the region's high temperatures and

humidity. It is also associated with rainfall, high nitrogen, plant size and density. Attempts have

been made to avoid harvest problems by the use of ultra-narrow row plantings and harvesting

using a stripper, but this has not proven feasible due to higher production costs and lower fiber

quality (Wright et al., 2004).

Previous research has shown hardlock to be associated with F. verticillioides (Marois et

al., 2002). Prior to Michailieds and Morgan (1998), F. verticillioides was referred to as F.

moniliforme. F. verticillioides commonly occurs in cotton fields, and can be found growing

saprophytically on crop residue. It has been isolated from both seeds and peduncles. It can also

sometimes be found on mature cotton fiber in the field, yielding the pink coloration described

previously. It is hypothesized that F. verticillioides infects via the flowers, and colonizes the

developing boll. Inoculation of flowers with a spore suspension ofFusarium has been shown to

result in more hardlock (Marois et al., 2005).









F. verticillioides is capable of surviving for extended periods of time on crop debris.

Cotten and Munkvold (1998) soaked maize stalk pieces in spore suspensions ofF. moniliforme,

F. proliferatum, and F. subglutinans, and left pieces in an Iowa field at the surface and several

depths in the soil profile under several crop rotations. The recovery rate for all three species was

over 50% during the first 300 days, and was around 10 to 20% after 600 days. Rohrbach and

Taniguchi (1984) observed the rate of infection by F. moniliforme on pineapple during the

flowering stage was best predicted by the number of hours per week the temperature was

between 21 and 270C. They also noted a significant negative correlation at 27 to 320C and 32 to

380C. Infections were also associated with rainfall, although a clear correlation was not

demonstrated. Subbarao and Michailides (1995) determined the optimal temperature for F.

moniliforme infection on pollinator figs is 300C. It also resulted in the shortest incubation and

latent periods (approximately 15 and 40 hours, respectively). At 350C, these were increased to

70 to 90 and 90 to 120 hours. These periods were also increased at temperatures below 300C,

but less drastically.

Palmateer et al. (2004) found F. moniliforme to be relatively uncommon among Fusarium

species isolated from living cotton plant tissue. F. proliferatum was also fairly uncommon. In

contrast, Baird and Carling (1998) found Fusarium species to be present on 22 to 38% of dead

cotton roots. Although F. verticillioides and F. proliferatum were not listed in the 6 most

frequently isolated Fusarium species, 5 other unlisted species were isolated from 1 to 6% of

samples. Bolkan et al. (1979) demonstrated F. moniliforme conidia to be short-lived (6 to 13

weeks) in the soil in the absence of host tissues. However, incorporating stem and leaf tissue

from pineapple into the soil increased survival to at least 12 months. It should be noted cotton

stems, roots, and fiber from the previous year are commonly found in cotton fields.









Other studies have examined F. moniliforme spore release under field conditions. Sanders

and Snow (1978) found the amount of airborne spores increased for several weeks after first

bloom, possibly due to saprophytic growth on shed flowers and other tissue, before finally

declining approximately 6 weeks later. That trend was not observed in this study, although

taking more samples might have showed it to be the case. It was later confirmed (Snow and

Sanders, 1979) that shed flowers, bolls and squares were a suitable substrate to produce

Fusarium spores. This was observed on 46 to 96% of flowers and 54 to 88% of bolls and

squares. Ooka and Kommedahl (1977) observed a similar situation with F. moniliforme spores

in corn fields. Spore numbers were lowest while the plant was actively growing, and increased

as it reached maturity. They also observed wind-blown soil containing F. moniliforme spores

which was likely to have traveled 300-400 km. However, Fernando et al. (2000) found more

airborne spores of Fusarium graminearum within 1.5 m of an infected wheat field than at 5 m,

suggesting local production of inoculum may be most important. Sanders and Snow (1978)

found the release of boll-rotting pathogen spores (including Fusarium) to be highest between

18:00 and 06:00. Fernando et al. (2000) found spore release ofF. moniliforme to be highest in

wheat fields between 16:00 and 08:00. The availability of moisture is important for growth.

Torres et al. (2003) examined the role of water activity (relative humidity expressed as a

decimal, abbreviated aw) and temperature on germination ofF. verticillioides on maize kernels.

Of the two temperatures (20 and 300C) and three levels of aw (0.92, 0.95 and 0.98) evaluated, the

combination of 300C and 0.98 resulted in the most rapid hyphal growth. Growth was

substantially slower at lower temperature and levels of aw.

Fusarium is associated with boll rots in cotton, and the symptoms of hardlock are often

lumped with boll rots. Arndt (1950) reported an especially severe year of boll rots in South









Carolina, and a symptom he described as the "tight-lock condition." The prevalence of "tight-

lock" ranged from 1 to 90%, with areas closer to the coast having a higher incidence. Bagga

(1968) reported F. monliforme as the second most isolated boll rot species, occurring on 9 to

13% of samples. Fungal pathogens, including Fusarium, can be found within cotton bolls by the

first several weeks of development (Roncadori, 1969). When inoculated directly into the

pericarp of a developing boll, F. moniliforme causes disease on both adjacent carpels as well as

the locule (Spamicht and Roncadori, 1972). The possibility of inoculating flowers with spores of

a pathogen to cause boll rot has been demonstrated (Edgerton, 1912). The incidence of boll rot

is higher with a closed canopy, due to increased humidity, although this is sometimes alleviated

by a lack of rainfall. Reduced nitrogen fertilizer rates sometimes resulted in less boll rot,

although the canopy characteristics were not affected (Roncadori et al., 1975).

Boll rot is also influenced by the quantity of airborne inoculum available. At two locations

in Louisiana, spore numbers were found to peak approximately 50 days after first bloom, and the

first infected bolls were observed at 60 days. The spore concentration was considerably less 10-

100 m from the cotton, so it is likely spores were originating in the crop. Spore concentration

also varied considerably during the day, with the highest levels between 1800 and 0600 hr.

Among boll-rotting fungi, Fusarium spores were second only to Diplodia gossypina in numbers.

No relationship was found linking temperature, humidity or rainfall to spore numbers. It was

suggested the increasing levels of spores could have been generated by fungal growth on

naturally shed flowers, bolls, squares, and leaves (Sanders and Snow, 1978).

In the case of boll rot caused by Colletotrichum capsici, damage to lint can vary from a

"tight-locked" condition to severe degradation of fiber (Roberts and Snow, 1984). Susceptibility

to boll rot is also influenced by gossypol (a polyphenol in the Malvaceae), production of which









can be increased by mechanical damage of the plant (Bell, 1967). Alternative leaf shape has also

been explored as an option for reducing boll rot incidence. Okra-leaf varieties have deeply lobed

leaves, similar to that of okra. This results in less boll rot, probably due to a less humid plant

canopy. These varieties also produce more flowers during a season, in the range of 150 to 210

per meter of row compared to 100 to 140 in conventional types. A more open plant canopy also

improves the efficiency with which pesticides can be applied to the plants. Okra-leaf varieties

produce better yields under adverse conditions, but lower yields under optimal conditions

compared to conventional varieties. They have not been commercially successful in the US,

although they account for approximately 50% of the cotton acreage in Australia (Heitholt and

Meredith, 1998).

Marois and Wright (2004) reported a climate model for predicting hardlock incidence.

The mean temperature and humidity from 0700 to 1900 on the day of bloom were recorded

during 2002, and correlated to hardlock incidence in subsequent bolls originating from the white

flowers of that particular day. The model was based on eight separate dates, and yielded a R2

value of 0.935 and p < 0.005.

Since F. verticillioides was shown to be associated with hardlock (Marois et al., 2002),

fungicide applications have been evaluated as a control measure. Marois and Wright (2004)

showed significant reductions in hardlock and increases in yield during the 2002 growing season

with applications of thiophanate-methyl. Benomyl applications were made to cotton under three

nitrogen fertilizer regimes, but only reduced hardlock at the highest nitrogen rate (201 kg/ha).

Applications to blooms instead of bolls were shown to be most effective. During the 2004

season, fungicide applications did not significantly affect hardlock or yield (Marois et al., 2005).

However, significant increases in leaf area index (LAI) and decreases in leaf disease were noted.









Although significant yield differences were not noted, a positive correlation with LAI and a

negative correlation to leaf disease were demonstrated. Other studies have failed to show

reductions in hardlock due to fungicide applications. Seebold et al. (2004) examined the use of

fungicide applications in five states in the Southeast during the 2003 season. Fungicides did not

reduce hardlock, although it improved seed cotton yield in Georgia. The number of applications

was shown to be more important than the rate or timing of those applications. The regional

project was expanded to ten states for the 2004 season (Seebold and Kemerait, 2005). Fungicide

applications had little impact on hardlock, regardless of the number of applications. No impact

on yield was noted, except in Virginia. Yields in Louisiana, Florida and Georgia may have been

lower due to inclement weather, which resulted in a complete loss of the Alabama study. In

Florida, Georgia and Tennessee, fungicides increased LAI, and in Florida and Georgia this was

determined by the number of applications. Marois et al. (2006) did not find improvements to

LAI, hardlock, or yield from fungicide applications during the 2005 season in Florida.

Hardlock is a major limiting factor for cotton yields in Florida, and control strategies are

not available. It is hypothesized that flower thrips may be involved in the problem, and that

reducing their numbers could reduce the severity of hardlock. Studies were performed from

2003 to 2005 to investigate this possibility. The first objective was to quantify the density and

diversity of arthropod species in cotton flowers and the effects of insecticides on their

populations. The second objective was to determine if there was a relationship between thrips

species and hardlock. The third objective was to assess if thrips numbers could be managed to

reduced hardlock and improve yield.









CHAPTER 2
THRIPS ON FLOWERS OF COTTON

Introduction

Thrips are small insects, approximately 1mm in length (family Thripidae). They range in

color from light tan to dark brown, depending on the species. They possess primitive wings that

allow limited flight. Small hairs setaee) occur on some areas of the body, and along with

antennae are characteristics useful for identifying the species. Most thrips of concern in cotton

production belong to the genus Frankliniella (subfamily Thripinae). Frankliniellafusca is a pest

of cotton seedlings and peanuts. Other species such as Frankliniella occidentalis, Frankliniella

tritici, and Frankliniella bispinosa occur more often on mature vegetation or flowers.

Thrips are known vectors of plant viruses. F. occidentais (Sakimura, 1962), F. fusca

(Pappu et al., 1998) and F. bispinosa (Avila et al., 2006) are capable of transmitting the

tospovirus tomato spotted wilt virus. F. fusca has also been shown to transmit Pantoea ananatis,

the bacterium responsible for center rot of onion (Gitiatis et al., 2003). Thrips can increase the

incidence of corn ear rot, caused by Fusarium verticillioides (Farrar and Davis, 1991). Thrips

can vector the pathogen and, by feeding, establish infection sites. Thrips can overwinter in the

soil as larvae. F. occidentalis has been shown to survive whole-body freezing at temperatures as

low as -16C if gradually acclimated (McDonald et al., 1997). In milder climates, adults are

active during the entire year, although the maturation rate declines (Toapanta, 2001).

Several flower-inhabiting thrips species are present in north Florida. In an 18-month study

of 37 wild plant species, 78% of thrips found were adults, and 87% of these adults were from the

genus Frankliniella (Chellemi et al., 1994). The most common species were F. tritici, F.

bispinosa, F. occidentalis and F. fusca. The relative contributions of each species fluctuated

substantially during the study. Cotton flowers were not examined, but during its bloom period of









June, July and August, F. tritici and F. bispinosa were the most common species on wild hosts.

The primary influence on Frankliniella populations appeared to be the availability of suitable

flowers. F. occidentalis populations increased more rapidly during the spring than other species,

probably due to its wider plant host range. It is non-native, and was not reported in the area until

1981 (Beshear, 1983).

Thrips numbers are affected by predatory insects and parasites. Orius insidiosus (minute

pirate bug) is a common predator that can consume 12.5 thrips per day (Tommasini and Nicoli,

1993). If a sufficient number of Orius are present in an area, localized extinction of flower thrips

may occur. Orius will remain in the area, feeding on extrafloral nectaries or pollen and

preventing the thrips population from rebounding. This diet is sufficient to allow development

and oviposition by Orius.

A parasitic nematode Thripinemafuscum was shown to infect F. fusca in north Florida.

Among F. fusca found on peanut, as many as 51 and 67% of females on certain sampling dates

were found to be infected, resulting in sterility. T. fuscum infection ofF. tritici and F.

occidentalis was far less common, occurring in only 2% of individuals (Funderburk et al., 2002;

Stavisky et al., 2001; Ramachandran et al., 2001).

Species and sex differences have been observed in thrips behavior. On greenhouse pepper

plants, movement ofF. occidentalis was limited, while F. tritici and F. bispinosa were found to

disperse relatively rapidly. Males of all three species were also shown to be more mobile among

plants (Ramachandran et al., 2001). Females ofF. occidentalis have been shown to spend more

time feeding and produce more feeding-associated scars on petunias than do males (van de

Wetering et al., 1998). These differences in behavior resulted in F. tritici and F. bispinosa being

less susceptible to predation by Orius than is F. occidentalis (Ramachandran et al., 2001.









Thrips species also differ in their response to insecticides. F. occidentalis populations

have been shown to increase following applications of acephate and esfenvalerate, while

spinosad causes a reduction. However, acephate and esfenvalerate are highly toxic to F. tritici

and F. bispinosa, while spinosad is less effective (Reitz et al., 2003). Other studies have shown

spinosad to be equally effective against all three species (Eger et al., 1998). Spinosad is a

macrolide that contains two active ingredients, spinosyn A and spinosyn D. They are derived

from Saccharopolyspora spinosa, and result in hyperexcitation of neurons in the central nervous

system resulting in eventual paralysis (Salgado, 1998). The effectiveness of insecticides on

thrips numbers can be complicated by its effect on predators, which probably explains increases

in F. occidentalis. Studebaker and Kring (2000) evaluated several insecticides for both lethal

and sub-lethal effects on Orius which could affect its reproduction. Spinosad was found to have

no effect, lethal or sub-lethal, on Orius populations. Cyhalothrin resulted in high mortality, but

no sub-lethal effect was observed in survivors. Acephate was not tested, but is a broad-spectrum

organophosphate which affects a large number of arthropods.

Thrips possess only one mandible which is used for cutting into plant tissue, and a stylet

through which food is drawn. This results in open wounds to the plant, which could allow easier

penetration of the tissue by pathogens. Females ofF. occidentalis have been shown to feed more

frequently and intensely than males, resulting in more tissue damage (van de Wetering et al.,

1999). Pickett et al. (1988) reported 68% of adult F. occidentalis found on cotton plants

occurred on fruiting structures, with most of these occurring in the flower itself. This is

consistent with the concept that cotton pollen may be preferable to leaves as a food source for

some thrips species (Agrawal et al., 1999). Pollen provides a source of protein for egg

development, resulting in increased fecundity (Trichilo and Leigh, 1988). In tomato, amino acid









analysis showed phenylalanine content to be associated with thrips numbers (Brodbeck et al,

2001). Nectar is also an important resource for attracting pollinators and predatory insects to

cotton plants (Wackers and Bonifay, 2004).

Thrips are a frequent problem on cotton seedlings in the US, resulting in distortion of

expanding leaves, and small discolored spots. The most common species are Frankliniella

occidentalis (western flower thrips), Frankliniellafusca (tobacco thrips) and Thrips tabaci

(onion thrips). Thrips control is usually achieved using a granular insecticide at planting. Foliar

insecticide applications are used if more than 2 to 3 thrips are present per plant and damage is

observed (Sprenkel, 2005). Applications ofjasmonic acid to cotton seedlings can reduce thrips

feeding by 80%, although leaf area is also reduced by 28% (Omer et al., 2001). However, this is

not used as a management technique. Thrips feeding is generally not a problem when the plants

are more mature. However, severe damage was reported in Turkey (Atakan and Ozgur, 200 la)

by Frankliniella intonsa on mature cotton plants. Although more than 350 thrips were observed

per flower, pollination was not adversely affected. Feeding by thrips larvae resulted in boll

shedding, although ovipositioning by females in flower parts had a larger impact.

In recent years there has been increasing interest in hardlock of cotton. Hardlock is a

failure of the cotton fiber to expand outward from the boll after opening, and it instead remains

in compact wedges. Affected locules will remain on the plant or be knocked to the ground

during harvest. Fiber quality is not usually affected, but yields can be reduced considerably.

Hardlock is associated with the fungus Fusarium verticillioides, and is believed to infect through

the flower on the day of bloom (Marois et al., 2002). Most control strategies have focused on the

application of fungicides to flowers and maturing bolls (Marois and Wright, 2004). It has also









been suggested flower thrips could be involved in hardlock. If that is the case, reducing their

numbers may limit the severity of hardlock.

Previous studies have examined the thrips species associated with cotton plants. However

these have been conducted outside the southeastern US, in drier climates, and the thrips species

found differ from those in this area. Frankliniella intonsa has been reported as a pest of cotton

in Greece (Deligeorgidis et al., 2002), Turkey (Atakan and Ozgur, 2001a). Frankliniella

schultzei and F. occidentalis have been reported as pests of cotton flowers and leaves in Brazil

(Monteiro, 2001). Other studies within the Southeast have focused on damage caused to cotton

seedlings by thrips feeding. The objectives of this study are to describe the insect species found

in cotton flowers and determine how they are affected by insecticide applications.

Materials and Methods

Field Plots

Two field studies were performed, approximately 40 miles apart, at branches of the North

Florida Research and Education Center in Quincy and Marianna, Florida. Cultivar DPL 555

Bt/RR was used, and plots were maintained according to the recommendations of the University

of Florida extension service unless otherwise noted. Acephate (Orthene) and lambda cyhalothrin

(Karate) were used when needed to control Nezara viridula (southern green stink bug) and

Euschistus servus (brown stink bug).

In Quincy, a large fungicide-insecticide study was utilized to evaluate thrips, hardlock, and

yield for 2 years. It was a randomized complete block design, with 4 blocks. There were 28

treatments in 2004 and 10 treatments in 2005. Plots were 4 rows (0.9 m between rows) by 9 m

long. Control and insecticide-treated plots (with or without fungicide, depending on the year)

were sampled for thrips. Other treatments were varied rates and timings of fungicide

applications, and were not sampled. In 2004, the insecticide treatment consisted of weekly









applications of 0.10 kg a.i. (active ingredient)/ha of spinosad (Tracer) on Mondays and 0.56kg/ha

acephate + 0.04 kg a.i./ha lambda cyhalothrin (Warrior) on Thursdays. In 2005, 0.02 kg a.i./ha

Karate (lambda cyhalothrin) was substituted for Warrior, and 0.9 kg/ha of thiophanate-methyl

(Topsin M) was applied every 2 weeks.

The Marianna study examined the effects of insecticides and fungicides on thrips,

hardlock and yield for 3 years. The site was part of a Paspalum notatum (bahiagrass) rotation,

and the cotton was planted after peanuts each year. The plots were eight rows in width, with 0.9

m between rows, and 18 m in length. Rows were oriented north to south, and at each end a 6 m

wide section of peanuts was planted. Peanuts support large numbers ofF. fusca, and are often

planted in proximity to cotton. It was suspected they could influence the species ratio found in

cotton flowers. A randomized complete block design was used with 4 blocks and 4 treatments.

The experiment included unsprayed control plots and three other treatments which were applied

during the bloom period. The insecticide treatment consisted of spinosad at 0.07 kg a.i./ha

alternated weekly with acephate at 0.9 kg a.i./ha. The fungicide treatment consisted of

thiophanate-methyl at 1.1 kg a.i./ha applied weekly. A fourth treatment included a weekly

application of both the insecticide and fungicide sprays listed above.

Sampling of Thrips

Cotton flowers are white on the first day they open, but by evening the fringes of the petals

are often pink. On the second day, the petals have changed to a solid dark pink or red color. In

this study, white, first day, flowers were collected and placed into individual 60 ml vials

containing 70% ethanol, 30% deionized water. The flower was placed into the vial with the

peduncle located at the opening. This allowed the flower contents to fall into the bottom of the

vial. Flowers were sampled on a weekly basis from the two outer rows of each plot, between

11:00 and 13:00. In the Quincy study, 12 flowers were sampled from each of 4 control and 4









insecticide plots. During 2004, the interval between insecticide applications and sampling

varied. In 2005, insecticide-only plots were not available, so an insecticide+fungicide treatment

was substituted. The interval between treatment sprayings and sampling time was held constant

at 2 days. In the Marianna study, 16 flowers were sampled from each of the 16 plots. The

interval between insecticide applications and sampling varied between 2 and 6 days. In both

locations, treatments were applied on the first week of bloom, and sampling began later that

week. Sampling was discontinued in late August, since blooms after that time would not result

in harvestable bolls. In 2004, a hurricane prevented sampling on the 5th week, and by the

following week the plants had stopped flowering.

Identification of Thrips

Vials containing thrips were kept at room temperature until the sample was evaluated.

During 2003 and 2004, the liquid from the sample was poured into a Petri dish and the flower

was then immersed in the dish to dislodge any thrips present. The sample was then examined,

and the insects recorded. In 2005, the flower was removed and insects were counted while still

in the vial. Flowers were periodically dissected to ensure thrips were not remaining within the

flowers. Thrips were recorded to species based on overall color, and antennae pigmentation and

ornamentation, while sex was determined by presence or absence of the ovipositor, and abdomen

width and curvature. Other commonly found or easily identified insects were classified to order

or genus. Insects that were rarely observed were not recorded. The data were analyzed using the

SAS GLM procedure, and means were separated using Tukey's Studentized Range Test.

Multi-State Species Survey

To determine if the species found in Quincy and Marianna were typical of the Southeast in

general, or specific to north Florida, samples were taken in other states. Collaborators collected

50-80 flowers from Louisiana, Alabama and South Carolina and sent them to Quincy, FL for









identification. The Louisiana samples were collected at the Macon Ridge Research Station in

Winnsboro, LA. The Alabama samples were collected at the Auburn University Gulf Coast

Research Station located in Fairhope, AL. The South Carolina samples were collected at the

Edisto Research and Education Center in Blackville, SC. Flowers were sampled for 3 years at

each location.

Thrips Accumulation in Flowers

In 2004, white flowers were sampled at 2-hour intervals from 10:00 to 16:00 or 10:00 to

18:00 in Quincy, Florida. In 2005, this was repeated at two other sites within approximately 300

meters of the 2004 site. At each time interval, 15 flowers were collected from each site.

Samples were processed as described previously. Sampling was discontinued on several days

due to rainwater from thunderstorms remaining in flowers.

Results

Prevalent Species

Flower thrips species identified from Quincy and Marianna were consistent at both sites

and across all years of the study (Tables 2.1 and 2.2). F. tritici (eastern flower thrips) was the

most common species, and flowers contained an average of 1.4 to 4.2 individuals. Females

outnumbered males, typically by a 2:1 to 5:1 ratio. F. occidentalis (western flower thrips) was

rarely found in 2003 and 2004 (2 to 3 thrips per 1000 cotton flowers), and was not present in

2005. Although slightly more common (3 to 10 thrips per 1000 cotton flowers), the same pattern

of diminishing prevalence was observed in F. bispinosa (Florida flower thrips). These three

species are all flower feeders, and likely competitors for resources. F. fusca was observed at low

levels (3 to 16 per 1000 flowers) in all years. It is a foliar-feeding species, and could have been

found in the flowers as a result of dispersion in search of suitable host plants. Immature thrips

were also present, and their numbers ranged from 0.05 to 0.10 per flower.









Orius species are predators of thrips, but can subsist on nectar when prey is unavailable.

Observed ratios of Orius to thrips ranged from 1:35 to 1:220 depending on location and year

(Tables 2.1 and 2.2). Previous research in field pepper (Funderburk et al. 2000) has shown thrips

suppression to occur at 1:200, and several days after reaching 1:40 localized near-extinction

occurred. Although low ratios were often observed, in cotton they were not sufficient to cause

further reductions in thrips numbers.

Aphis sp. (aphids) were observed at levels of 1.2 to 5.3 per flower. They are known for

secreting sticky honeydew, and if bolls are open this can result in "sticky cotton". This problem

was not observed in the plots, and is distinct from hardlock. Members of the order Formicidae

(ants) were also observed, with as many as 2.9 per 10 flowers. They were highly aggregated,

and flowers containing ants typically contained 5 or more. Although they were not classified

further, at least two species appeared to be present. In 2005, members of order Forficulidae

(earwigs) were first recorded. They had been observed but not recorded previously due to their

extremely low numbers. Beetles from order Staphilinidae were also first observed in high

numbers (4 to 56 per hundred flowers) in 2005.

Similarities Across the Southeast

Samples from Louisiana, Alabama, and South Carolina were also examined to see if they

were similar to those in Florida (Tables 2.3, 2.4, and 2.5). In all locations, F. tritici was most

common. Their numbers ranged from 0.42 to 14.8 per flower. Louisiana in 2003 contained the

highest number ofF. occidentalis recorded (0.31 per flower), but they still constituted a minority

of flower thrips present. One individual each of both F. bispinosa and T. palmi were found. The

presence ofF. bispinosa was unexpected since they are most common to peninsular Florida. The

sex ratios varied considerably, with the highest being 1 male per 48 females. In a third of the

location/years, males were more common than females. The large range of thrips numbers and









sex ratios observed may have been influenced by crop management practices. The number of

immature thrips also varied considerably, from 0.01 to 0.96 per flower. Orius was generally

uncommon, except in two instances where 0.15 and 0.21 per flower were found.

Impact of Insecticide Treatments on Flower-Inhabiting Insects

The influence of insecticide treatments on flower-inhabiting insects was examined (Tables

2.6 and 2.7). Insecticide applications, whether alone or in combination with fungicide, reduced

thrips numbers. The reduction in thrips numbers varied depending on location and year. Thrips

were reduced by 84 and 92% in Quincy, and 32, 20 and 36% in Marianna. Both males and

females were affected, but males experienced larger percentage declines. Fungicide applications

did not influence thrips numbers. Immature thrips were reduced by insecticides by 94 and 100%

in Quincy. The number of immature thrips was not significantly reduced in Marianna. The

number of male thrips per female was usually between 0.1 and 0.7 (Table 2.8). In Quincy the

proportion of males was reduced in both years, by about 50%. In Marianna, insecticide

applications significantly reduced the proportion of males in 2003, but not in 2004 or 2005.

However, combining the data from all three years showed an overall significant reduction from

0.5 per female in the control to 0.4 in the insecticide-treated plots. The proportion of males

varied considerably during the season (Fig. 2.1). Orius was adversely affected by insecticides in

Quincy (Table 2.6). Adults were reduced by 83% during one year, and immatures by 80 and

100% in both years. The number of Orius observed was very low in all treatments, and

considering the sample size, this result should be viewed with caution. Marianna, insecticides

were less harmful to Orius (Table 2.7). Adults were not affected, while the number of

immatures in the insecticide plot was reduced by 66% compared to the control during one of the

three years. Aphis sp. were not consistently affected by the treatments. There was also little to









no impact on representatives from the order Formicidae (ants). The insecticide treatments

appeared to reduce thrips numbers while having a small impact on other flower inhabitants.

Changes in Thrips Numbers During the Growing Season

The number of thrips observed varied considerably during the growing season (Fig. 2.3).

At later sampling dates in Marianna in 2003 and 2005, thrips numbers were approximately 55

and 170% higher than earlier in the seasons and were associated with a declining number of

white flowers available in the field. This may have caused crowding in the remaining flowers.

The overall seasonal trend was similar in both of these years, although 2005 had fewer thrips. In

2004, flowering stopped unusually early, preventing continued sampling. This was preceded by

a drastic decline in thrips numbers, from approximately 4.5 to 0.1 per flower. In Quincy in 2004,

the season began with an unusually rapid increase in thrips numbers, followed by an abrupt

decline. The control and insecticide plots declined by 70 and 80% from their mid-season peaks.

In 2005, treatment differences were apparent. On the first sampling date, the control plots

contained 13 thrips per flower, but declined to 4 by the next sampling date and declined for the

remainder of the season. The insecticide-treated plots showed less variation between days, but

thrips numbers were generally lower during the second half of the season.

Association of Thrips and Orius

Orius populations also fluctuated during the seasons (Fig. 2.2). In Marianna, control and

insecticide plots generally showed similar fluctuations during the growing season for each year.

In 2005, there was a rapid increase in the number of Orius present during the last two sampling

dates. On the last sampling date of that year, Orius numbers were approximately 400% higher

than before their increase (0.40 vs. 0.05 per flower). When the fluctuations in thrips thrips

numbers (Fig. 2.3) are compared to Orius (Fig. 2.2), it is difficult to discern a clear relationship.









To determine if thrips and Orius were associated, the number of flowers containing either

thrips or Orius, both species, or neither was calculated. The predicted distribution was

determined (Table 2.9), and the significance of their association was calculated using a Chi-

square test (Table 2.10). The number of flowers containing both thrips and Orius was higher

than predicted, and this was significant (p<0.05) in 3 of 5 location-years. Three indexes of

species association were used: Ochiai, Dice, and Jaccard. Values can range from 0 (not

associated) to 1 (always associated), and those computed for this study were between 0.06 and

0.37. The covariation of thrips and Orius was evaluated by calculating their correlation on a per

flower and a per plot*day basis for each location year (Table 2.11). On a per flower basis, the

correlations were fairly low (<0.20). By comparing the mean number of thrips and Orius for

each plot and day, correlations as high as 0.50 were obtained. It appears that thrips and Orius are

associated, but not strongly in cotton flowers. In a predatory relationship, a positive association

suggests Orius fluctuates in response to thrips numbers. In contrast, a negative association

would result if Orius resulted in a localized depression in thrips numbers. If both these scenarios

occur, then the observed relationship would appear weak. There were several instances that may

show the Orius population being influenced by thrips numbers. In the 2004 Marianna control

plots (section A of Fig. 2.2 and 2.3), the decline in Orius numbers was less drastic than in thrips

at the end of the season, suggesting they were depressing thrips numbers. In the 2003 Marianna

insecticide plots (section B of Fig. 2.2 and 2.3) thrips reached their lowest numbers on the

second week, then rebounded for the remainder of the season. This same pattern occurred in the

Orius population, except it was delayed by one week. In the 2004 Quincy insecticide plots, a

seasonal peak in thrips numbers occurred one week before the seasonal peak for Orius. It

appears each species is capable of influencing the numbers of the other at certain times.









Thrips Accumulation in Flowers

Cotton flower buds are very tightly closed, and occasional dissections did not reveal any

thrips within them prior to flower opening. The flower opens at approximately 09:00, and thrips

begin arriving almost immediately (Fig. 2.4). On most days in 2005, there were 2 or fewer thrips

found in flowers at the first sampling time of 10:00. At site 2 (2006), flowers at the 10:00

sampling time typically had 4 thrips per flower. Site 3 (2006) had slightly more, with 2 to 8 per

flower. Their numbers usually increased until 14:00. At 14:00, flowers typically contained 4 to

14 (Site 1, 2005), 10 to 15 (Site 2, 2006), and 14 to 25 (Site 3, 2006) thrips. After 14:00, thrips

numbers were generally as likely to decline as to increase further, but tended to remain in a

similar range to the numbers at 14:00. When sampling days are compared to each other, there

are obvious differences in the initial number of thrips, rate of increase, and time of maximum

numbers. At sites 2 and 3 on 7/28/2006, thrips numbers at 10:00 were far higher than on any

other day sampled that year. This was associated with lower humidity than on other days. There

is considerable variation in thrips numbers between days, and it was hypothesized this was due to

weather conditions. Temperature, relative humidity, solar radiation, wind speed and rainfall

were compared to the number of thrips present. Correlations were performed between thrips

numbers at 10:00 and temperature and relative humidity for each hour from 16:00 of the day

prior to the time of sampling. Means were computed for temperature and relative humidity for

the most highly correlated time periods. These means were then correlated to thrips numbers

(Table 12.2). At sites 2 and 3, relative humidity from 19:00 on the day before sampling to 10:00

of the next day was strongly correlated to the number of thrips present in flowers at 10:00 (-0.87

and -0.85, respectively, p<0.08). Combining data from sites 2 and 3 resulted in a slightly lower,

but highly significant correlation (-0.81, p=0.0046). Temperature from 8:00 to 10:00 was also

correlated to thrips numbers in sites 2 and 3 (0.54 and 0.69, respectively, p<0.35). Combining









both days improved the relationship (0.60, p=0.07). Relative humidity and temperature were not

correlated to thrips numbers at site 1. Sampling on those days was distributed over a 2-week

period, while sites 2 and 3 were sampled on consecutive days. It is possible variations in the

total thrips population during the course of sampling at site 1 obscured the role of weather. The

sampling dates also showed narrower ranges of temperature (27.60 to 29.40C) and relative

humidity (76.0 to 83.3%) in 2005 than in 2006 (26.6 to 30.50C and 66.3 to 84.7%). The

narrower weather variable ranges in 2005 may have been insufficient to influence thrips

numbers.

Regressions were performed comparing the weather variables described previously to

thrips numbers (Table 2.13). Adjusted R2 values were high for relative humidity and low for

temperature. Using both variables to predict thrips numbers was less useful than relative

humidity alone. Relative humidity appears to be the best predictor of thrips numbers at 10:00.

Solar radiation was not predictive of thrips numbers, possibly due to its limited ability to

penetrate the plant canopy in early morning. Wind speeds were relatively low (0 to 11 km/h) on

most days, and did not predict thrips numbers. Thrips numbers after 10:00 were not influenced

by the weather variables examined. After 10:00, the best predictor of thrips numbers was the

number observed at the previous sampling time. Rainfall also occurred during the dates sampled

(Table 2.14). On one occasion (7/29/2006) rainfall slightly reduced thrips numbers, although

this was only temporary. It is likely some thrips left the flowers, while others were sheltered

between overlapping cotton petals. Overlapping cotton petals are also a favored position during

normal weather conditions, and may harbor 40 to 80% of the thrips present in a flower.

Discussion

Cotton flowers contain valuable resources for insects, and regardless of what control

measures are attempted, some organism is likely to utilize them. Flower inhabitants interact









through competition and predation, and their populations can fluctuate for unknown reasons.

Several observations from these studies warrant further comment.

Based on its prevalence on vegetable crops and wild hosts prior to 2003, F. occidentalis

was predicted to be a common species in cotton flowers. Instead, it was rare in 2003 and not

found in 2005. This decline in F. occidentalis numbers was also observed on other crops. It

may be that the ratio of flower thrips species found in cotton flowers is closer to an

approximation of the species found in the surrounding area rather than being a unique ecological

niche. If that is the case, fluctuations may occur in the ratio of thrips species. This could

influence predators which feed on thrips. Some studies have suggested Orius feeds

preferentially on F. bispinosa and F. occidentalis relative to F. tritici. This is probably due to

the increased activity levels noted in F. tritici, which requires increased effort for predators

(Reitz et al., 2001). Under certain conditions, predation may influence the species ratio. As time

progresses, the most common species may face greater pressures as predator and parasite

populations adjust themselves to better exploit it. It is possible this may prevent any one species

from comprising a majority indefinitely. Having mostly one thrips species in a field does

simplify management. Reitz et al. (2003) determined effects of insecticides varied according to

species in pepper production. They found spinosad was effective against F. occidentalis, but not

F. tritici. In contrast, esfenvalerate and acephate reduced populations ofF. tritici and F.

bispinosa, but resulted in higher populations ofF. occidentalis. This suggests the spinosad

component of the insecticide applications in this study may not have been as important.

However, the high proportion ofF. tritici in thrips populations occurred in multiple locations in

the Southeast, and was present in plots not sprayed with spinosad. It is unlikely the use of

spinosad biased the species ratio in this study.









Insecticide applications generally resulted in fewer males compared to females in the plots,

and this could be relevant from a management perspective. Thrips are haplodiploid, with males

being haploid and females being diploid. Sexual reproduction between male and female thrips

results in exclusively female offspring, since sex is determined by ploidy level. Females also

reproduce parthenogenically, which results in exclusively male offspring. Decreasing the

number of males available for breeding might increase the likelihood of parthenogenic

reproduction, resulting in fewer female offspring being produced. The reason for a reduction in

males in the insecticide plots is unclear. Males tend to disperse more readily than females,

suggesting they would re-colonize the insecticide plots faster than females. This is the opposite

of what was observed. Research on tomato has shown males are more likely to occur in the

upper portion of the plant canopy (Reitz et al., 2002). This increases their exposure to

insecticide applications compared to females, which are more evenly distributed throughout the

canopy. Greater exposure to spraying probably explains the lower proportion of males in

insecticide-treated plots.

Alternating applications of spinosad and acephate in Marianna did not appear to harm

Orius populations. However, Orius is highly mobile and may have re-colonized plots after

treatment. It should be noted that spinosad is among the least harmful insecticides to Orius

(Ramachandran, 2001; Reitz et al., 2001). Insecticide applications did reduce Orius numbers in

Quincy, and this may be a result of using lambda cyhalothrin instead of acephate. In pepper

production, Orius has been demonstrated to cause localized near-extinctions of thrips (Reitz et

al., 2003). That did not occur in this study, possibly due to either different thrips species or host

plant characteristics. F. tritici is more mobile, and more likely to evade Orius than is F.

occidentalis (Reitz et al., 2001). Also, the near-extinctions observed in pepper do not occur in









tomato production and it is believed characteristics of the plant itself are involved. However, a

positive association between thrips and Orius was observed.

In this study, the mean number of thrips declined each year in all treatments. In 2004, this

was primarily due to a sharp decline in thrips numbers in the fourth week of bloom. The reason

for that decline is unclear, but it preceded the unusually early end of flowering. Nitrogen

concentration in plant tissue influences both thrips feeding and flowering. It is possible

declining nitrogen availability within the plant led to reduced phenylalanine content, causing

thrips to emigrate in search of better hosts (Brodbeck et al., 2001). Low nitrogen availability

might also have caused premature termination of flowering. This seems unlikely considering the

abruptness with which this sequence occurred. Another possibility is planting date. Due to

weather and technical problems, planting dates were slightly later each year, with the onset of

flowering ranging from June 20 in 2003 to about July 5 in 2005. Thrips populations are at their

peak in the spring, but decline rapidly in May (Chellemi et. al, 1994). It is possible delayed

planting of cotton resulted in more of a gap in food availability between the wild spring-hosts

and the cotton flowers, reducing thrips numbers. It is also possible that various parasites and

predators more suited to F. tritici have increased since the time of its presumed replacement ofF.

occidentalis in early 2003. Parasitic nematodes from the genus Thipinema are common

worldwide. T. fuscum infects 40-80% ofF. fusca females in north Florida, but fewer than 2% of

F. occidentalis and F. tritici (Funderburk et al., 2002; Stavisky et al., 2001; Ramachandran et al.,

2001). It likely there are other such parasites which more commonly affect F. tritici.

The observed increase in thrips numbers per flower during the day is consistent with

previous studies. In Turkey, F. intonsa was shown to follow a similar pattern. Red flowers (day

after bloom) were found to contain their highest numbers of thrips at 05:30, after which their









numbers fell steadily until 11:30. This dispersal coincides with the opening of new white

flowers, which were shown to accumulate between 20 and 200 individuals. The authors also

suggested red flowers could serve as a refuge from insecticide applications (Atakan and Ozgur,

2001b). In British Columbia, it was determined F. occidentalis dispersal could occur if

windspeed was less than 15 km/h, although it was most common in the absence of wind

(Pearsall, 2002). Wind suppressing thrips movement was not observed in this study, probably

because wind speed in this study did not exceed 16 km/h, and was typically below 8 km/h. The

association of lower temperature and relative humidity prior to 10:00 with higher thrips numbers

was unexpected. A prior study (Toapanta, 2001) has shown thrips generation lengths to be

determined by growing degree days, with warmer temperatures resulting in faster maturity.

Higher morning temperatures in this study decreased thrips movement, perhaps because the

temperatures encountered were already above the optimum. The lower humidity might result in

less dew on the plants, allowing thrips to move more easily within the plant canopies.

Thrips control in cotton flowers had not been attempted previously, and its potential

effectiveness was unknown. Insecticide applications reduced total thrips numbers by

approximately 20 to 90% depending on the year and location. This study demonstrates

insecticide applications are an effective strategy for reducing thrips numbers in cotton flowers.

This reduction in thrips numbers was associated with reduction in hardlock severity, and is

described in Ch.3.










Table 2.1. Mean number of inhabitants per flower across all treatments in Quincy, FL.
Thrips 2004 2005
Juvenile 0.10 0.03 0.06 0.02
Adult Male Female Male Female
Frankliniella tritici 1.10 0.12 2.41 0.18 0.39 0.06 1.88 0.17
F. bispinosa <0.01 0.00 <0.01 0.00 0 0
F. fusca 0 <0.01 0.00 0 0.01 0.00
F. occidentalis <0.01 0.00 0 0 0
Thrips palmi 0 0 0 0
Other insects observed
Juvenile Orius sp. 0.03 0.01 0.01 0.01
Adult Orius sp. 0.05 0.01 0.03 0.01
Aphis sp. 3.84 0.36 5.32 0.32
Formicidae 0.25 0.06 0.08 0.03
Forficulidae --- 0
Staphilinidae --- 0.04 0.01










Table 2.2. Mean number of inhabitants per flower across all treatments in Marianna, FL.
Thrips 2003 2004 2005
Juvenile 0.13 0.02 0.05 0.01 0.02 0.00
Adult Male Female Male Female Male Female
Frankliniella tritici 1.16 0.08 3.21 0.08 0.68 0.04 2.10 0.08 0.40 0.03 1.04 0.04
F. bispinosa 0.01 0.00 0.07 0.00 <0.01 0.00 0.01 0.00 0 0
F. fusca 0 0.02 0.00 0 0.01 0.00 0 0.01 0.00
F. occidentalis <0.01 0.00 <0.01 0.00 <0.01 0.00 0 0 0
Thrips palmi <0.01 0.00 <0.01 0.00 0 0 0 0


Other insects observed
Juvenile Orius sp.
Adult Orius sp.
Aphis sp.
Formicidae
Forficulidae
Staphilinidae


0.06 0.00
0.10 0.00
1.20 0.09
0.10 0.02


0.04 0.00
0.05 0.00
1.83 0.16
0.29 0.03


0.04 0.00
0.15 0.00
5.48 0.37
0.25 0.03
<0.01 0.00
0.56 0.03


Table 2.3. Mean number of inhabitants per flower in Winnsboro, LA.
SThrips 2003
Juvenile 0.08 0.03
Adult Male Female
Frankliniella tritici 1.30 0.27 1.51 0.24 0.8
F. bispinosa 0 0
F. fusca 0 0
F. occidentalis 0.12 0.06 0.19 0.05
Thrips palmi 0 0.01 0.01


Other insects observed
Juvenile Orius sp.
Adult Orius sp.
Aphis sp.
Formicidae
Forficulidae
Staphilinidae


0
0.01 0.01
3.02 0.65
0


2004
0.13 0.07
Male Female
3 0.15 4.30 0.53
0 0
0 0
0 0
0 0

0.02 0.02
0
18.06 2.61
0


2005
0.01 0.00
Male Female
0.01 0.01 0.41 0.15
0 0
0 0
0 0
0 0

0
0
0.04 0.02
2.34 0.45
0.13 0.05
0.51 0.22










Table 2.4. Mean number of inhabitants per flower in Fairhope, AL.
Thrips 2003
Juvenile 0.96 0.26
Adult Male Female
Frankliniella tritici 2.96 0.36 1.52 0.21 6
F. bispinosa 0 0.02 0.02
F. fusca 0 0
F. occidentalis 0.02 0.02 0
Thrips palmi 0 0


Other insects observed
Juvenile Orius sp. 0.14 0.05
Adult Orius sp. 0
Aphis sp. 0.40 0.23
Formicidae 0.20 0.07
Forficulidae ---
Staphilinidae ---

Table 2.5. Mean number of inhabitants per flower in Blackville, SC.
SThrips 2003
Juvenile 0.14 0.06
Adult Male Female
Frankliniella tritici 3.00 0.45 1.35 0.20 0.
F. bispinosa 0 0
F. fusca 0 0.01 0.01
F. occidentalis 0 0
Thrips palmi 0 0


Other insects observed
Juvenile Orius sp.
Adult Orius sp.
Aphis sp.
Formicidae
Forficulidae
Staphilinidae


0
0
0.83 0.29
0.18 0.01


2004
0.29 0.11
Male Female
32 0.78 8.50 1.01
0 0
0 0
0 0
0 0

0.07 0.03
0.21 0.07
0.13 0.07
0.09 0.05


2004
0.02 0.02
Male Female
18 0.06 0.98 0.16
0 0
0 0
0 0
0 0

0
0.07 0.04
0.54 0.16
0.20 0.09


2005
0.62 0.12
Male Female
1.88 0.25 1.31 0.14
0 0
0 0
0 0
0 0

0.09 0.03
0.03 0.02
0.43 0.12
0.01 0.01


2005
0.18 0.06
Male Female
1.28 0.23 2.73 0.28
0 0
0 0
0 0
0 0

0.07 0.04
0.15 0.04
14.66 2.62
0.03 0.02
0
0.01 0.01


.










Table 2.6. Mean numbers of flower inhabitants by treatment in Quincy, FL.
F. tritici Orius sp.
immature
year treatment N male female thrips immature adult Aphis sp. Formicidae
2004 C 188 2.05 a 4.03 a 0.19 a 0.05 a 0.06 a 3.43 a 0.51 a
2005 C 288 0.73 b 3.43 a 0.11 a 0.03 a 0.06 a 3.78 a 0.13 b
2004 C 188 2.05 a 4.03 a 0.19 a 0.05 a 0.06 a 3.43 a 0.51 a
2004 I 190 0.15 b 0.81 b 0.01 b 0.01 b 0.04 a 4.25 a 0.00 b
2005 C 288 0.73 a 3.43 a 0.11 a 0.03 a 0.06 a 3.78 b 0.13 a
2005 IF 280 0.04 b 0.28 b 0.00 b 0.00 b 0.01 b 6.91 a 0.03 a
Numbers within a column between horizontal lines followed by the same letter are not significantly different
according to Tukey's Studentized Range Test Range Test (p<0.05). Treatment "C" is a control, "I" is a weekly
application of lambda cyhalothrin and spinosad insecticides (separate days), and "IF" is a combination of the
fungicide and insecticide treatments.

Table 2.7. Mean number of inhabitants per flower by treatment in Marianna, FL.


F. tritici


year treatment
2003 all
2004 all
2005 all
all C
all F
all IF
all I
2003 C
F
IF
I
2004 C
F
IF
I
2005 C
F
IF
I


N
1052
1023
1190
838
803
792
832
278
247
244
283
256
255
256
256
304
301
292
293


male
1.16a
0.68 b
0.40 c
0.92 a
0.86 a
0.57 b
0.56 b
1.54 a
1.35 a
0.94 b
0.82 b
0.77 a
0.72 a
0.56 a
0.66 a
0.48 a
0.58 a
0.28 b
0.24 b


female
2.72 a
2.10 b
1.04 c
2.21 a
2.05 a
1.72 b
1.68 b
3.06 a
2.90 a
2.62 a
2.33
2.47 a
2.20 a
1.81
1.91
1.20 a
1.21 a
0.88 b
0.84 b


Numbers within a column between horizontal
according to Tukey's Studentized Range Test


Immature
thrips
0.13 a
0.05 b
0.02 b
0.06 a
0.09 a
0.04 a
0.06 a
0.10a
b 0.17a
b 0.08 a
b 0.15a
0.06 a b
b 0.09 a
b 0.02 b
b 0.02 b
0.03 a
0.01 a
0.02 a
0.01 a
lines followed


Orius sp.


immature
0.04 a
0.04 a
0.04 a
0.05 a
0.04 a
0.03 a
0.03 a
0.04 a
0.04 a
0.03 a
0.03 a
0.03 a
0.03 a
0.04 a
0.05 a
0.06 a
0.04 a b
0.02 b
0.02 b
by the same


adult
0.11 b
0.05 c
0.15a
0.10a
0.13a
0.10a
0.09 a
0.10a
0.15a
0.09 a
0.08 a
0.04 a
0.08 a
0.04 a
0.04 a
0.14a
0.15a
0.15a
0.15a


Aphis sp.
1.41 b
1.83 b
5.48 a
2.73 a
3.21 a
3.32 a
2.85 a
0.85 c
1.57 a b
2.03 a
1.28 c b
1.68 a
1.59 a
1.79 a
2.25 a
5.34 a
5.93 a
5.74 a
4.88 a


Formicidae
0.09 b
0.29 a
0.25 a
0.25 a
0.25 a
0.16a
0.20 a
0.15a
0.08 a
0.05 a
0.10a
0.34 a
0.20 a
0.28 a
0.34 a
0.25 a
0.44 a
0.15a
0.17a


letter are not significantly different


(p<0.05). Treatment "C" is control, "F" is thiophanate-methyl


fungicide, "I" is a weekly application of acephate or spinosad insecticides, and "IF" is a combination of the
fungicide and insecticide treatments.










Table 2.8. Number of males per female by treatment for each year of sampling.
Quincy, FL
Treatment 2004 2005
C 0.52 a 0.21 a
IF 0.10 b
I 0.25 b
P-value 0.0178 0.1574

Marianna, FL
Treatment 2003 2004 2005 all years
C 0.63 a 0.29 a 0.45 a b 0.48 a
F 0.57 ab 0.31 a 0.50 a 0.47 a
IF 0.40 b 0.35 a 0.33 a b 0.36 b
I 0.40 b 0.39 a 0.27 b 0.36 b
P-value 0.0295 0.4766 0.0402 0.0155
Numbers within a column between horizontal lines followed by the same letter are not significantly different
according to Tukey's Studentized Range Test (p<0.05). Treatment "C" is control, "F" is thiophanate-methyl
fungicide, "I" is a weekly application of acephate or spinosad insecticides, and "IF" is a combination of the
fungicide and insecticide treatments.












1.8
1.6 A
1.4
1.2
E 1.0
S0.8 -F
0.6 E I
0.4 E IF
0.2
0.0
7/15/03 7/24/03** 7/31/03** 8/7/03 8/23/03
Date


0.8
B
0.7
0.6
0.5 -
E C
0.4 I C
W 0.3
El I
0.2 I IF
0.1
0.0
7/17/04 7/25/04 8/1/04** 8/9/04
Date

Figure 2.1. Number of males per female by treatment on each day of sampling. A, Marianna,
2003. B, Marianna, 2004. C, Marianna, 2005. D, Quincy, 2004. E, Quincy, 2005.
Treatment "C" is control, "F" is thiophanate-methyl fungicide, "I" is a weekly
application of acephate or spinosad insecticides, and "IF" is a combination of the
fungicide and insecticide treatments. A double asterisk (**) indicates the control and
insecticide treatments differed significantly (p< 0.05) according to Duncan's multiple
range test for that specific date.














1.2

1.0

0.8

g 0.6 mF

0.4 I
o IF
0.2

0.0 )
7/20/05 7/28/05 8/11/05 8/25/05 9/1/05
Date


1.2
D

1.0


0.8

mC
0.6 I-


0.4


0.2


0.0
7/16/04 7/23/04 7/30/04** 8/6/04
Date

1.2
E
1.0


0.8


0.6 FI


0.4


0.2


0.0
7/23/05 7/27/05 8/10/05 8/17/05** 8/20/05 8/26/05
Date

Figure 2.1. Continued











































































Figure 2.2. Variation in adult Orius (minute pirate bug) numbers over time. A, Marianna, FL

unsprayed control. B, Marianna FL insecticide treatment. C, Quincy, FL unsprayed
control. D, insecticide treatment.


0.5

0.45 A

0.4

0.35 -

0.3 -

T
0.25 -

0.2 -

0.15

0.1 -

0.05 7

0
12-Jul 22-Jul 1-Aug 11-Aug 21-Aug 31-Aug 10-Sep
Date


0.5 -

0.45

0.4

0.35

0.3

0.25

0.2

0.15 -

0.1

0.05

0--
12-Jul


22-Jul 1-Aug 11-Aug 21-Aug 31-Aug 10-Ser
Date

















































Figure 2.2. Continued















7

6

t 5
9-

4 -
a

4 3

2

1

0 -
12-Jul



7--


6


5


4 4-

3I -


2


1 -


0-
12-Jul


21-Aug 31-Aug 10-Sep


22-Jul 1-Aug 11-Aug 21-Aug 31-Aug 10-Sep
Date


Figure 2.3. Variation in thrips numbers over time. A, Marianna, unsprayed control. B,
Marianna, insecticide treatment. C, Quincy, unsprayed control. D, Quincy,
insecticide treatment.


T


I L


22-Jul 1-Aug 11-Aug
Date


T













16


14-

12I

10


._-- :004

6
\T T
4 L1 1





12-Jul 22-Jul 1-Aug 11-Aug 1 -Aug


an

1.5
3


12-Jul 22-Jul


1-Aug


-'04
105








- 9- .I ,, i 1 -A u
S--Aug


Date


Figure 2.3. Continued









Table 2.9. Occurance of adult thrips and Orius in flowers sampled.


Quincy 2004

Quincy 2005

Marianna 2003

Marianna 2004

Marianna 2005


actual
predicted
actual
predicted
actual
predicted
actual
predicted
actual
predicted


Table 2.10. Interspecific association of thrips


Quincy 2004
Quincy 2005
Marianna 2003
Marianna 2004
Marianna 2005


association
positive
positive
positive
positive
positive


Table 2.11. Interspecific covariance of thrips


Quincy 2004
Quincy 2005
Quincy
Marianna 2003
Marianna 2004
Marianna 2005
Marianna
Quincy 2004
Quincy 2005
Quincy
Marianna 2003
Marianna 2004
Marianna 2005
Marianna


correlation
0.0446
0.2011
0.1341
0.0651
0.0932
0.1855
0.0754
0.1905
0.4568
0.3512
0.1785
0.1936
0.5067
0.1291


thrips +
Orius
14
11
15
9
89
84
41
34
118
85


neither
present
135
132
302
296
125
120
304
297
480
447


thrips
alone
225
228
247
253
831
836
670
677
561
594


and Orius.
Ochiai
0.21
0.21
0.30
0.22
0.37


and Orius.
N
378
568
946
1052
1023
1190
3265
32
48
80
155
128
160
443


Orius
alone
4
7
4
10
7
12
8
15
31
64



Dice
0.11
0.11
0.18
0.11
0.29


p
<0.25
<0.005
<0.25
<0.05
<0.005


N
378
378
568
568
1052
1052
1023
1023
1190
1190


Jaccard
0.06
0.06
0.10
0.06
0.17


grouping

per
flower






per
day*plot


p
0.3876
<0.0001
<0.0001
0.0347
0.0029
<0.0001
<0.0001
0.2965
0.0011
0.0014
0.0263
0.0285
<0.0001
0.0065












20

18 A

16 -

14

S. 12 -'':-:'
S_ 10 1 'i 'oos

I 8

6

4
2


8 10 12 14 16 18 20
time (h)


30

B
25


20-


1 7 .0/jOy '
cc




o---------1COO-
15 -


10-






0
8 10 12 14 16 18 20
time (h)


Figure 2.4. Increase in thrips numbers during first day of bloom in Quincy, FL. A, Site 1, 2005.
B, Site 2, 2006. C, Site 3, 2006.













40 -

35

30 /

| 25 -.28 '0

20 0 C

15 -T .

10 T i

5-

0 -
8 10 12 14 16 18 20
time (h)

Figure 2.4. Continued






Table 2.12. Correlation of mean weather conditions to thrips numbers at 1000.
Temperature 0800 to 1000 RH 1900 to 1000
Site Year N Corr p Corr p
1 2005 5 -0.4146 0.4877 0.1221 0.8449
2 2006 5 0.5437 0.3436 -0.8727 0.0535
3 2006 5 0.6906 0.1967 -0.8523 0.0666
2+3 10 0.6001 0.0667 -0.8089 0.0046
1+2+3 15 0.4211 0.1180 -0.6049 0.0169










Table 2.13. Relationship of mean temperature from 0800 to 1000 and relative humidity from
1900 to 1000 to thrips numbers at 1000
Site N Equation R2 adj.-R2 p
1 5 thrips = 82.93826 -T*2.88495 0.1719 -0.1041 0.4877
2 5 thrips = -26.52871 +T*1.12867 0.2956 0.0608 0.3436
3 5 thrips = -64.85198 +T*2.54632 0.4769 0.3026 0.1967
2+3 10 thrips = -45.69035 +T*1.83749 0.3601 0.2801 0.0667
1+2+3 15 thrips = -39.65781 +T*1.59549 0.1773 0.114 0.118
1 5 thrips = -8.83217 +RH*0.12930 0.0149 -0.3134 0.8449
2 5 thrips = 35.52886 -RH*0.34261 0.7616 0.6821 0.0535
3 5 thrips = 59.40033 -RH*0.59429 0.7264 0.6351 0.0666
2+3 10 thrips = 47.46459 -RH*0.46845 0.6543 0.6111 0.0046
1+2+3 15 thrips = 42.51782 -RH*0.41942 0.366 0.3172 0.0169
1 5 thrips = 74.90710 -T*3.71156 +RH*0.33889 0.2602 -0.4795 0.7398
2 5 thrips = 69.29711 -T*0.81540 -RH*0.46463 0.8192 0.6385 0.1808
3 5 thrips = 52.78204 +T*0.15981 -RH*0.57037 0.7271 0.4541 0.2729
2+3 10 thrips = 61.03958 -T*0.32780 -RH*0.51750 0.6586 0.5611 0.0232
1+2+3 15 thrips = 38.22897 +T*0.11002 -RH*0.40594 0.3664 0.2608 0.0647





Table 2.14. Rainfall during the sampling times shown in Figure 2.4.
date hour rain (cm)
7/9/2005 13 1.27
7/9/2005 18 0.71
7/29/2006 13 0.20
7/29/2006 18 1.40
7/30/2006 14 0.15
7/30/2006 15 0.74
7/30/2006 16 0.15









CHAPTER 3
RELATIONSHIP OF FLOWER THRIPS TO FUSARIUM HARDLOCK

Introduction

Hardlock is expressed as a failure of the locule of fiber to expand outward after boll

opening. Instead, it remains in compressed locules, which may have a faint pink or orange color.

These compressed locules are frequently missed or knocked to the ground by mechanical

harvesters. The resulting yield loss often ranges from 20 to 60%, depending on the year. It has

not been formally recognized as a disease, although it could be considered a subset of the boll rot

complex. Hardlock is most severe along the gulf coast, possibly due to the region's high

temperatures and humidity. It is also associated with rainfall, high nitrogen, plant size and

density. Attempts have been made to avoid harvest problems by the use of ultra-narrow row

plantings and harvesting using a stripper, but this has not proven feasible due to higher

production costs and lower fiber quality (Wright et al., 2004).

Previous research has shown hardlock to be associated with F. verticillioides (Marois et

al., 2002). Prior to Michailieds and Morgan (1998), F. verticillioides was referred to as F.

moniliforme. F. verticillioides commonly occurs in cotton fields, and can be found growing

saprophytically on crop residue. It has been isolated from both seeds and peduncles. It can also

sometimes be found on mature cotton fiber in the field, yielding the coloration described

previously. It is hypothesized that F. verticillioides infects via the flowers, and colonizes the

developing boll. Inoculation of flowers with a spore suspension ofFusarium has been shown to

result in more hardlock (Marois et al., 2005).

Fusarium is strongly associated with boll rots in cotton, and the symptoms of hardlock

are often lumped with boll rots. Arndt (1950) reported an especially severe year of boll rots in

South Carolina, and a symptom he described as the "tight-lock condition". The prevalence of









"tight-lock" ranged from 1 to 90%, with areas closer to the coast having a higher incidence.

Bagga (1968) reported F. monliforme as the second most isolated boll rot species, occurring on 9

to 13% of samples. Fungal pathogens, including Fusarium, can be found within cotton bolls by

the first several weeks of development (Roncadori, 1969). When inoculated directly into the

pericarp of a developing boll, F. moniliforme causes disease on both adjacent carpels as well as

the locule (Spamicht and Roncadori, 1972). The possibility of inoculating flowers with spores of

a pathogen to cause boll rot has been demonstrated (Edgerton, 1912). The incidence of boll rot

is higher with a closed canopy, due to increased humidity, although this is sometimes alleviated

by a lack of rainfall. Reduced nitrogen fertilizer rates sometimes resulted in less boll rot,

although the canopy characteristics were not affected (Roncadori et al., 1975). Boll rot is also

influenced by the quantity of airborne inoculum available. At two locations in Louisiana, spore

numbers were found to peak approximately 50 days after first bloom, and the first infected bolls

were observed at 60 days. The spore concentration was considerably less 10-100 m from the

cotton, so it is likely spores were originating in the crop. Spore concentration also varied

considerably during the day, with the highest levels between 1800 and 0600 hr. Among boll-

rotting fungi, Fusarium spp. spores were second only to Diplodia gossypina in numbers. No

relationship was found linking temperature, humidity or rainfall to spore numbers. It was

suggested the increasing levels of spores could have been generated by fungal growth on

naturally shed flowers, bolls, squares, and leaves (Sanders and Snow, 1978). In the case of boll

rot caused by Colletotrichum capsici, damage to lint can vary from a "tight-locked" condition to

severe degradation of fiber (Roberts and Snow, 1984). Susceptibility to boll rot is also

influenced by gossypol (a polyphenol in the Malvaceae), production of which can be increased

by mechanical damage of the plant (Bell, 1967).









Cotton flowers usually open between 09:00 and 09:30, and are rapidly colonized by flower

thrips. In north Florida, Frankliniella tritici (eastern flower thrips) are most common, but F.

occidentalis (western flower thrips) and F. bispinosa (Florida flower thrips) are also present

(Mailhot, 2006). Their numbers increase as the day progresses, and flowers eventually contain

10 to 40 thrips. Predatory insects such as Orius insidiosus (minute pirate bug) can reduce thrips

numbers (Ramachandran et al., 2001), but can sometimes be harmed by insecticide applications

used to control thrips (Studebaker and Kring, 2000). Acephate is highly toxic to F. tritici (Reitz

et al., 2003). Spinosad is less toxic to F. tritici but non-toxic to 0. insidiosus (Studebaker and

Kring, 2000).

Thrips possess only one mandible which is used for cutting into plant tissue, and a stylet

through which food is drawn. This results in open wounds to the plant, which could allow easier

penetration of the tissue by Fusarium spores. Females ofF. occidentalis have been shown to

feed more frequently and intensely than males, resulting in more tissue damage (van de Wetering

et al., 1999). Farrar and Davis (1991) investigated the relationship between F. occidentalis, and

Fusarium ear rot of corn. They concluded that thrips may be acting as vectors ofF.

verticillioides or as wounding agents by feeding on the plant tissue. Pickett et al. (1988) reported

68% of adult F. occidentalis found on cotton plants occurred on fruiting structures, with most of

these occurring in the flower itself. This is consistent with the concept that cotton pollen may be

preferable to leaves as a food source for thrips (Agrawal et al., 1999).

In order to demonstrate the association of thrips with Fusarium hardlock, a series of

studies were performed. The objectives were to determine the prevalence of thrips exposed to

Fusarium in the field, whether artificially-exposed thrips would cause hardlock in a controlled

setting, and if field applications of insecticide would reduce hardlock.









Materials and Methods


Isolation of Fusarium from Thrips

White cotton flowers were collected from a crop rotation study in Quincy, FL. In 2004,

cultivar DPL 458 BG/RR was used, producing the Bacillus thuringiensis endotoxin and

providing resistance to glyphosate herbicide. In 2005, DPL 555 BG/RR was used, providing the

same benefits. The plants were maintained according to recommendations of the University of

Florida extension service, and treated with insecticides when necessary. Fungicides were not

applied during the growing season. Flowers were collected from border rows of plots between

10:00 and 11:00. Each flower was placed into a separate plastic bag, and sealed shut to contain

any thrips which were present. The sample bags were then refrigerated for approximately three

hours to reduce thrips mobility. All thrips from the particular bag were then placed onto Petri

dishes of one-quarter strength acidified potato dextrose agar. The dish was then covered and

sealed with parafilm to prevent any thrips from escaping. The thrips were permitted to move

around the dish, allowing any spores on them to be spread across the plate. In 2005 the

procedure was modified. Thrips were collected from flowers in the field using an aspirator,

placed directly onto APDA Petri dishes, and sealed with parafilm. The thrips survived for

approximately three days, after which time fungal colonies were counted, and suspected

Fusarium colonies were marked. Cladosporium and Trichoderma were also recorded due to

their frequency and distinctive appearances. After seven to ten days, suspected Fusarium

colonies were re-examined to verify the original count. Between 20 and 100 flowers were

sampled on each date, depending on the number of thrips present and weather conditions. The

data were analyzed using the SAS GLM procedure, and means were separated using Duncan's

multiple range test.









Influence of Thrips and Fusarium on Hardlock Incidence in Greenhouse

Cotton was planted and maintained in a greenhouse at the North Florida Research and

Education Center in Quincy, FL. The temperature varied between 250 and 450C and the plants

were watered as necessary. A potting mix including peat moss and composted bark was used.

Each cotton seedling was treated with 0.17 ml of Admire (imidacloprid) to prevent insect

infestations. Approximately 5 g of water-soluble 15-30-15 fertilizer was applied to each plant

early in the experiment. Pix (mepiquat chloride) was applied at a rate of 0.01 ml/plant 1 or 2

times, as needed to reduce plant height. Excessive height and boll weight required some plants

to be attached to bamboo stakes to remain upright. Shortly before flowering, plants were divided

into four groups: a control, thrips-only, thrips exposed to Fusarium, and Fusarium-inoculated.

The thrips-only treatment consisted of 10 thrips being placed into each open flower. Fusarium

verticillioides cultures (2, 5, and 6) isolated from infected bolls were transferred to 14 strength

acidified potato dextrose agar. These isolates have been shown previously to result in hardlock

(Marois and Wright, 2004) if inoculated into flowers. One transfer from each was used to

produce a mixed quantity of inoculum to better approximate that found in the field. Later work

analyzing an internal transcribed spacer (ITS) region of the genome established that isolates 5

and 6 are actually F. proliferatum (Leite, et al., 2006). After 10 to 14 days, spores were rinsed

from the surface using deionized water. A suspension of 90,000-330,000 spores per ml was

created, and with refrigeration remained viable for 3 days. Thrips were exposed to inoculum by

placing them on a plate ofF. verticillioides for approximately one hour, and then 10 of these

thrips were transferred into each available flower. Unlike flowers produced in a field setting, the

petals in the greenhouse expanded further apart, yielding a more open flower. Treatments were

spatially separated to prevent thrips from moving between treatments. Flowers were tagged with

ribbons to indicate the treatment and date. After all bolls on a plant were open, they were









evaluated for hardlock. Locules displaying the characteristic failure of fiber to expand were

deemed to be affected. The numbers of affected and total locules for each boll were recorded.

The first study was conducted from January to May of 2003, using variety DPL 555

BG/RR. Flowering occurred from March to April, and 15 plants were used in each of 4

treatments. Frankliniella occidentalis was raised in cages, using green beans and pollen as a

food source. Frankliniella occidentalis frequently occurred prior to 2003, and was suspected to

be a common species in cotton flowers. In the Fusarium inoculation treatment, approximately 5

ml of spore suspension was sprayed into each open flower. In the second study, flowering

occurred from April-May 2005. The Fusarium inoculation procedure was modified by using a

syringe to place approximately 1 ml of inoculum (same concentration as before) onto the stigma.

Exposure to water results in lysing of cotton pollen, reducing the chance of successful pollination

(Burke, 2002). This modification reduced the flower abortion rate and was maintained in

subsequent studies. However, insufficient thrips were produced, resulting in too few treated

bolls to evaluate the thrips treatments. The third study, flowering from July to August,

substituted wild-captured F. tritici to more effectively model field conditions. Thrips were

captured from the same plot as the Fusarium-isolation study, and sometimes kept in captivity for

several days using tomatillo (Physalis ixocarpa) fruit as a food source. A second cotton variety,

DPL 444 BG/RR, was added and 8 plants per variety were used in each of the 4 treatments. The

DPL 444 plants required 1-2 extra applications of mepiquat chloride to keep them similar in

height to DPL 555. DPL 444 also continued flowering for 4-5 weeks longer than DPL 555. The

third study was replicated from October to November. The first and second studies were

analyzed individually using the SAS GLM procedure, and means were separated using Duncan's

multiple range test. The third study consisted of two replications separated by time.









Effect of Fungicides and Insecticides on Hardlock

Two field studies were performed, approximately 40 miles apart, at branches of the North

Florida Research and Education Center in Quincy and Marianna, Florida. These were described

in chapter 2. Variety DPL 555 Bt/RR was used, and plots were maintained according to the

recommendations of the University of Florida extension service unless otherwise noted. Orthene

(acephate) and Karate (lambda cyhalothrin) were used when needed to control the southern green

stink bug (Nezara viridula) and the brown stink bug (Euschistus servus).

In Quincy, a large fungicide-insecticide study was utilized to evaluate thrips, hardlock, and

yield for 2 years. It was a randomized complete block design, with 4 blocks. There were 28

treatments in 2004 and 10 treatments in 2005. Plots were 4 rows (0.9 m between rows) by 9 m

long. Control and insecticide-treated plots (with or without fungicide, depending on the year)

were sampled for thrips. Other treatments were varied rates and timings of fungicide

applications, and were not sampled. In 2004, the insecticide treatment consisted of weekly

applications of 0.10 kg a.i. (active ingredient)/ha of spinosad (Tracer) on Mondays and 0.56kg/ha

acephate + 0.04 kg a.i./ha lambda cyhalothrin (Warrior) on Thursdays. In 2005, 0.02 kg a.i./ha

Karate (lambda cyhalothrin) was substituted for Warrior, and 0.9 kg/ha of thiophanate-methyl

(Topsin M) was applied every 2 weeks. In each of the 8 plots, 12 flowers were collected weekly,

and stored for identification. Hardlock severity was assessed approximately 2 weeks after

defoliation. Five plants were selected at random, and the number of hardlocked and total locules

per boll was recorded for each plot. The two center rows of each plot were harvested with a

spindle plot picker. The data were analyzed using the SAS GLM procedure, and means were

separated using Duncan's multiple range test (SAS, 2003).

The Marianna study examined the effects of insecticides and fungicides on thrips, hardlock

and yield for 3 years. The site was part of a Paspalum notatum (bahiagrass) rotation, and the









cotton was planted after peanuts each year. The plots were eight rows in width, with 0.9 m

between rows, and 18 m in length. Rows were oriented north to south, and at each end a 6 m

wide section of peanuts was planted. Peanuts support large numbers ofF. fusca, and are often

planted in proximity to cotton. It was suspected they could influence the species ratio found in

cotton flowers. A randomized complete block design was used with 4 blocks and 4 treatments.

The experiment included unsprayed control plots and three other treatments which were applied

during the bloom period. The insecticide treatment consisted of spinosad at 0.07 kg a.i./ha

alternated weekly with acephate at 0.9 kg a.i./ha. The fungicide treatment consisted of

thiophanate-methyl at 1.1 kg a.i./ha applied weekly. A fourth treatment included a weekly

application of both the insecticide and fungicide sprays listed above. Thrips were sampled from

the two outer rows of each plot, while yield data was obtained from two of the inner rows.

Flowers were sampled weekly, 8 from each plot, and stored until thrips could be counted. The

data were analyzed using the SAS GLM procedure, and means were separated using Duncan's

multiple range test.

Results

Isolation of Fusarium from Thrips

Between 7 and 13 percent of flowers contained thrips that were carrying Fusarium (Table

3.1). No significant differences were observed by year. The number of Fusarium isolations also

varied by day, with several days contributing most of the isolations for the season. High

inoculum days were randomly scattered throughout the season. Weather conditions were

compared, but no similarities between high inoculum days were identified.

Influence of Thrips and Fusarium on Hardlock Incidence in Greenhouse

In 2003, the addition of thrips (F. occidentalis) previously exposed to F. verticillioides

resulted in the most hardlock (Table 3.2). The Fusarium-inoculation treatment was also









significantly higher than the control. Thrips in the absence of Fusarium did not differ

significantly from the control group. In the second study, the Fusarium-inoculated and control

groups only differed at the p=0.10 level of significance.

In the third study, the results differed slightly by variety, although similar results were

obtained in both replications. In DPL 555, the same pattern was observed as in the first study,

despite using F. tritici instead ofF. occidentalis. Fusarium-exposed thrips resulted in the

highest levels of hardlock. Fusarium-inoculation and thrips each increased hardlock compared

to the control treatment. In DPL 444, Fusarium-inoculation and thrips each resulted in

significantly more hardlock than the control group, although Fusarium-exposed thrips did not

differ significantly from the control.

In many cases, fewer bolls successfully reached maturity when they had been treated with

Fusarium-carrying thrips. This was primarily due to more bolls aborting in their first 2 weeks of

development, although flowers also aborted slightly more often. On several days during the

experiments, flowers were not treated and the resulting bolls were not included in the

experiment. However, the bolls which formed from those untreated flowers rarely aborted. If

hardlock symptoms were a result of shortages in photosynthates, the reduction in boll numbers

might have alleviated some of the problem. Alternatively, the aborted bolls could represent a

more severe or advanced category of infection, although this is unlikely since F. verticillioides

could not be re-isolated from aborted bolls.

Effect of Fungicides and Insecticides on Thrips, Hardlock, and Yield

In Quincy, the use of insecticide, both alone and with a fungicide, significantly reduced

thrips numbers (Table 3.3) by 84% in 2005 and 92% in 2005. In 2005, spraying reduced

hardlock from 32% to 18%, but was not significant in 2004. Yield was not significantly affected

in either year.









In Marianna, insecticide provided significant reductions in thrips numbers for all years

(Table 3.4), while fungicide predictably had no impact. The reduction was more modest than

experienced in Quincy, ranging from 21 to 36%. Insecticide significantly reduced hardlock in all

years, ranging from approximately 30 to 50%. Fungicide applications were significant in some

years, but not all. Combining applications of insecticide and fungicide did not provide additional

reductions in hardlock in any year. It should also be noted that the greater percentage reduction

in thrips due to spraying in 2005 compared to previous years coincided with a greater reduction

in hardlock. Yields were low in 2003, and spraying was not beneficial. Yields were higher in

2004, but only the combined sprays resulted in a significant improvement. In 2005, yields were

higher than previous years, and insecticide applications provided significant increases in yield.

The relationship between thrips numbers and hardlock incidence in Marianna, FL is

illustrated in Fig. 3.1. The mean number of thrips found within flowers of a plot during the

entire season is compared to the hardlock incidence for that same plot. Clear patterns can be

distinguished for 2003 and 2005, with R2 values of 0.19 and 0.34. In 2004, removing the outlier

with the highest hardlock incidence only increases the R2 value to 0.06, so other factors may

have proved more important in explaining hardlock incidence. When treatments are examined

individually with all three years included (Fig. 3.2), the relationships are more clear. R2 values

range from 0.29 to 0.73. Removing one outlier each from the control and insecticide-only plots

increased R2 values to 0.54 and 0.77, respectively. When data is combined from all treatments

and years (Fig. 3.3), the relationship is also apparent.

Although Duncan's multiple range test did not show consistent differences in hardlock

severity in the Quincy study (Table 3.3), a regression showed a similar pattern to that observed in









Marianna. Combining all treatments and year resulted in an R2 value of 0.25. The relationship

was actually stronger in 2004 than 2005, with R2 values of 0.63 and 0.39.

Discussion

The rate at which Fusarium was isolated from thrips in flowers was lower than the

observed rate of hardlock in the field. Because thrips numbers within flowers increase during the

day, sampling in the afternoon could have resulted in isolation rates more similar to observed

rates of hardlock. Although hardlock incidence declined between 2003 and 2005, there was no

change in the frequency with which Fusarium was isolated from thrips in the field. This

suggests yearly fluctuations in inoculum quantity may not be an important determinant of the

severity of hardlock. Sanders and Snow (1978) found the amount of airborne spores increased

for several weeks after first bloom, possibly due to saprophytic growth on shed flowers and other

tissue, before finally declining approximately 6 weeks later. That trend was not observed in this

study, although taking more samples might have shown it to be the case. Ooka and Kommedahl

(1977) observed a similar situation with F. moniliforme spores in corn fields. Spores numbers

were lowest while the plant was actively growing and increased as it reached maturity. They

also observed wind-blown soil containing F. moniliforme spores which was likely to have

traveled 300-400 km. However, Fernando et al. (2000) found more airborne spores within 1.5 m

of an infected wheat field than at 5 m, suggesting local production of inoculum may be most

important.

The thrips-Fusarium isolation study was undertaken with the assumption thrips were

either transporting inoculum into the cotton flowers or their feeding damage to plant tissue made

infection more likely to occur. The results show thrips are capable of transporting inoculum, and

this appears realistic based on previous studies. Atakan (2001) observed thrips populations in

red flowers were highest at 05:30, after which they began dispersing. Sanders and Snow (1978)









found the release of boll-rotting pathogen spores (including Fusarium) to be highest between

18:00 and 06:00. Fernando et al. (2000) found spore release ofF. moniliforme to be highest in

wheat fields between 16:00 and 08:00. This allows a time period between 05:30 and around

09:30 before thrips enter new flowers and during which they could come into contact with

recently deposited spores from the previous night. This does not preclude the possibility that

thrips damage allows more infections, and perhaps both pathways work synergistically.

The 3 isolates used in the greenhouse study had been identified in fall of 2002 as F.

verticillioides based on spore types and shapes, as well as pigmentation in culture. It was not

until summer of 2006 that further work revealed isolates 5 and 6 to be F. proliferatum. Previous

inoculation studies showed no significant differences between the isolates in the ability to cause

hardlock (Marois and Wright, 2004; Marois et al., 2005). This suggests at least two Fusarium

species are capable of causing symptoms. Palmateer et al. (2004) found F. moniliforme to be

relatively uncommon among Fusarium species isolated from living cotton plant tissue. F.

proliferatum was also fairly uncommon. In contrast, Baird and Carling (1998) found Fusarium

species to be present on 22 to 38% of dead cotton roots. Although F. verticillioides and F.

proliferatum were not listed in the 6 most frequently isolated Fusarium species, 5 other unlisted

species were isolated from 1 to 6% of samples. Bolkan et al. (1979) demonstrated F.

moniliforme conidia to be short-lived (6 to 13 weeks) in the soil in the absence of host tissues.

However, incorporating stem and leaf tissue from pineapple into the soil increased survival to at

least 12 months. It should be noted cotton stems, roots, and fiber from the previous year are

commonly found in cotton fields.

In the greenhouse inoculation studies, hardlock symptoms occurred in the control group,

although at lower rates than the experimental treatments. Flowers were observed closely for









thrips to ensure Fusarium was not being transported to non-inoculated flowers. In addition, the

air entering this greenhouse passed through an evaporative cooler, reducing the opportunity for

airborne inoculum from outdoors to reach the flowers. It appears a certain amount of observed

hardlock may result from unknown factors. The first and third greenhouse studies also suggest

the particular species of thrips may not be important in hardlock, since similar results were

obtained with both F. occidentalis and F. tritici. It also demonstrated cotton varieties do not

respond identically to hardlock. However, the two varieties also showed large differences in

response to growth regulators, flower production, and boll retention, so comparisons between

them should not be extrapolated to field performance. The observation of more bolls aborting in

their first 2 to 3 weeks of development may be related to the higher levels of gossypol observed

at that stage in development (Bell, 1967).

Several observations from Figs. 3.1 to 3.4 are worth noting. In Fig. 3.1, the only treatment

which is consistently above the trendline is the unsprayed control. Within a season, it is assumed

factors other than thrips that might affect hardlock influence all plots equally. This suggests

something other than the quantity of thrips is influencing hardlock. It may be that in addition to

reducing their numbers, the insecticide applications reduce activities of thrips which contribute

to hardlock. In Fig. 3.2, plots were grouped by treatment across years. This is useful in that it

would reduce the possible complicating factor just described. However, because it is unknown

why thrips numbers decreased each year of the study, or if that cause also reduced hardlock

independently of the thrips numbers, the results should be viewed with caution. The same could

be said of Fig. 3.3, although it includes more overlap between years. In Fig. 3.4, the 2004

control plots show the largest range of values along the trendline, making it most useful for

confirming the thrips-hardlock connection. Each of the other 3 groups is tightly clustered, and









these clusters are considerable distances apart. Because of this, any comparison between these

groups will yield a good R2 value. While this makes them less useful for showing the connection

between thrips and hardlock, they are very useful for demonstrating the benefits obtained from a

spray program.

Applications of insecticide to field plots were effective at reducing thrips numbers, and

this effect was observed in both Quincy and Marianna. This was expected based on F. tritici

being the dominant species and the particular insecticides chosen to control thrips without

severely affecting Orius. By preserving the natural predator, the mixed results sometimes

associated with spraying (Funderburk et al., 2000) were avoided. The connection between thrips

numbers and hardlock was clear in Marianna, but only obvious in one of two years in Quincy.

This suggests that while thrips appear to be the most significant factor, they are not the only

determinant of hardlock incidence. Although it can be concluded that reducing thrips numbers

within a season reduced the incidence of hardlock, it is not clear whether declining thrips

populations led to reduced hardlock, or if both were being influenced by a third factor. Yield can

be adversely affected by many factors, and in some cases hardlock may not be a significant

contributor to low yields. At the same time, reducing thrips numbers, regardless of their starting

point, was usually beneficial, suggesting it may be a useful management strategy for controlling

hardlock.










Table 3.1. Proportion of flowers containing thrips-carried Fusarium.
Year Percent Fusarium N
2003 7.0 a 329
2004 13 a 23
2005 6.6 a 273
Numbers followed by the same letter are not significantly different according
to Duncan's Multiple Range Test (p<0.05).


Table 3.2. Results of greenhouse hardlock experiments.
2003 2005
March-April April-May Jul-Aug Oct-Nov
Percent Percent Percent
Hardlock N Hardlock N Hardlock N
DP 444 Fusarium+thrips --- --- -- 37 a b 37
Fusarium --- --- -- 44 a 51
Thrips --- --- -- 45 a 58
Control --- -- --- -- 27 b 41
DP 555 Fusarium+thrips 66 a 25 --- -- 57 a 26
Fusarium 56 a b 58 20 a 30 37 b 80
Thrips 39 cb 36 --- -- 40 b 85
Control 32 c 50 12 a 67 23 c 76
Numbers in a column followed by the same letter between horizontal lines are not significantly different according
to Duncan's Multiple Range Test (p<0.05).

Table 3.3. Effect of fungicides and insecticides on thrips, hardlock, and yield in Quincy, FL.
Treatment thrips hardlock yield
2004 2005 2004 2005 2004 2005
Control 6.10 a 4.18 a 0.48 a 0.32 a 1369 a 1547 a
Insecticide +fungicide --- 0.32 b 0.42 a 0.18 b 1530 a 1590 a
Insecticide 0.95 b --- 0.42 a --- 1280 a ---
Numbers followed by the same letter are not significantly different according to Duncan's Multiple Range Test
(p<0.05).

Table 3.4. Effect of fungicide and insecticide treatments on thrips, hardlock, and yield in
Marianna, FL.
Treatment thrips
2003 2004 2005
Control 4.05 a 3.28 a 1.68 a
Fungicide 3.84 ab 2.96 ab 1.79 a
Insecticide +fungicide 3.14 c b 2.38 b 1.16 b
Insecticide 2.83 c 2.59 b 1.08 b
hardlock yield
2003 2004 2005 2003 2004 2005
Control 0.72 a 0.41 a 0.41 a 528 ab 911 b 1276 b
Fungicide 0.69 a 0.31 b 0.33 b 533 a 1003 a b 1237 b
Insecticide +fungicide 0.52 b 0.28 b 0.23 c 417 b 1178 a 1688 a
Insecticide 0.47 b 0.29 b 0.19 c 499 ab 1110 ab 1602 a
Numbers followed by the same letter in a column are not significantly different according to Duncan's Multiple
Range Test (p<0.05).















2003


090
080
070
060
050
040
030
020
010
000
000 1 00 200


300

thrips




2004


400 500 600


090
080
070 '
060 1l
050
040 *
030
020
010
00
0 00 1 00 200 300 400 500 600


thrips




2004


070
060
050
040
030
020
010


n7n


060
050
040
030
020
010
000
0000


00 1 00 200 300 400 5 0(

thrips




2005


040

030

020

010

000
000


050 1 00 1 50 200 250

thrips


050-

040

030

020

010

000
000


300


1 00 200 300 400

thrips




2005















050 1 00 1 50 200 250

thrips


Figure 3.1. Regression of thrips numbers and hardlock incidence by year in Marianna, FL.


C = control, unsprayed
F = fungicide treatment, weekly application of Topsin M at 1.25 lb/acre
I = insecticide treatment, weekly alternating application of Tracer at 2 oz/acre or Orthene at 1 lb/acre
IF= insecticide with fungicide treatment, simultaneous application of insecticide and fungicide regimes listed above


I X


U

U
U~ 0


uuu
0


050


0
U.
r- U


500


300


2003

















Fungicide 2003-2005


100 200 300 400 500 600

thrips


080
070
060
050
040
030
020
010
000
00


Insecticide 2003-2005


4 1~~~
F- 1'


00 1 00 200 300 400 500

thrips


070
060
050
040
030
020
0 10
000
000


090 -
080
070
S060
8 050
0 040
* 030
020
010
000
000


Figure 3.2. Regression of thrips numbers and hardlock incidence by treatment in Marianna, FL.


C = control, unsprayed
F = fungicide treatment, weekly application of Topsin M at 1.25 lb/acre
I = insecticide treatment, weekly alternating application of Tracer at 2 oz/acre or Orthene at 1 lb/acre
IF= insecticide with fungicide treatment, simultaneous application of insecticide and fungicide regimes listed above


100 200 300 400 500 600

thrips




Insecticide + Fungicide 2003-2005








*





1 00 200 300 400 500

thrips


*

"II~:


F-


050

. 040

S030

= 020

0 10

000


Control 2003-2005













All treatments 2003-2005


0.90
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10


0.00


0.00 1.00 2.00 3.00 4.00 5.00 6.00
thrips


Figure 3.3. Regression of thrips numbers and hardlock for all treatments and years in Quincy,
FL.


0O1CIIF.h C, 1202 *





19*


I















All treatments 2004-2005


50


20 -



10 ,- ,1- ,
30

?0

10

DO
000 200 400 600 800 1000

thrips




2004


60
50

40 v

30

20.

10

00
000 200 400 600 800 1000

thrips


Control 2004-2005


060

050

040

S030

= 020

010

000
000


2 00 400 600

thrips


050

040 *



S20 **

010
n "r,


000 200 400 600 800 1000

thrips




2005


30
*


u25

020

015 *

010

005

000
0 00 1 00 200 300 400 500

thrips




Insecticide/insecticide-fungicide 2004-2005


0 Z


800 1000


050


thrips


Figure 3.4. Regression of thrips numbers and hardlock in Quincy, FL.


2004 C = control, unsprayed
2004 I = insecticide treatment, 2.9 oz/acre of Tracer on Mondays and 8 oz/acre Orthene + 5 oz/acre Warrior
on Thursday
2005 C = control, unsprayed
2005 IF = insecticide with fungicide treatment, 2.9 oz/acre of Tracer on Mondays, 8 oz/acre Orthene + 2
oz/acre Karate on Thursdays and 1.0 lb/acre of Topsin M every 2 weeks


*F ~i 'i-~


All treatments 2004-2005


u


^









CHAPTER 4
RELATIONSHIP OF HARDLOCK TO WEATHER, THRIPS, AND YIELD

Introduction

Hardlock of cotton (Gossypium hirsutum L.) occurs when the fiber does not fluff out as the

boll opens at maturity. Mature locules look like wedges of an orange when broken apart (Fig.

4.1). Although the quality of the cotton fiber may not be severely affected, conventional spindle

harvesting equipment is not able to capture the fiber and bring it into the harvester as the

hardlocked cotton is knocked from the plant and falls to the ground or is strung out of the boll

giving the appearance of poor harvesting procedures. Attempts to "scrap" the field by running the

picker a second time to get the hardlocked cotton often results in little lint increase, more trash and

lower lint quality. The severity of hardlock in cotton has been associated with high nitrogen, high

plant density, high temperature and humidity, insect damage, and seed rot. Because bolls affected

by hardlock are not harvested by conventional pickers, yield losses of over 50% have occurred in

states in the Southeast.

In 2002, hardlock was shown to be associated with Fusarium verticillioides (Marois et al.,

2002). Before a publication by Michailieds and Morgan (1998), F. verticillioides was referred to

as Fusarium moniliforme. F. verticillioides occurs frequently in cotton fields, growing

saprophytically on crop residue. It has been isolated from seeds, peduncles, and mature fiber. It

is hypothesized that F. verticillioides infects the flowers on the day of bloom and enters the

immature boll before flower dehiscence. Flower inoculation using a suspension ofF.

verticillioides micro and macroconida has been shown to increase hardlock severity (Marois et al.,

2005).

Hardlock can be differentiated from the traditional boll rots. Boll rots result from pathogen

damage where the carpel turns brown or black and never opens or after the bolls have opened,









microorganisms destroy the cotton fibers. Batson (2001) identified 15 species of fungi and

bacteria associated with boll rots, but he does not describe hardlock independently of boll rot.

Boll rots occur during wet weather when the cotton boll or fiber is colonized by a number of

microbes, although only a few fungi are responsible for the majority of infections (Pinkard and

Chilton, 1966). These include Alternaria gossypina (Thuem.) Hopkins, Curvularia spp., Diplodia

gossypina Cke., Helminthosporium gossypii Tucker, Fusarium spp., and Phomopsis spp.

(Palmateer et al., 2003: Pinkard and Chilton, 1966). Sanders and Snow (1978) found a

correlation between the numbers of airborne spores of these fungi and the incidence of boll rot

caused by them. They also proposed that the likely source of the fungi was not from infected

bolls but from spores being produced on shed plant parts, as there was a correlation between

airborne spore numbers and the normal seasonal shedding of flowers, bolls, squares, and leaves

(Sanders and Snow, 1977).

Flower thrips rapidly colonize cotton flowers on the morning they open, typically between

0900 and 0930. The most common species in north Florida is Frankliniella tritici (eastern flower

thrips). F. occidentalis (western flower thrips) and F. bispinosa (Florida flower thrips) are also

present (Chapter 2). Thrips continue entering the flowers for much of the day, typically reaching

a peak of 10 to 40 per flower between 1400 and 1800. Thrips numbers can be reduced by

predatory insects such as Orius insidiosus, the minute pirate bug (Ramachandran et al., 2001), but

these predators can sometimes be harmed by insecticide applications used to control thrips

(Studebaker and Kring, 2000). A connection between thrips numbers and temperature has been

demonstrated. The average temperature between 0800 and 1000 is negatively correlated with the

number of thrips in the flowers at 1000 (Ch.2).









In 2004, a model was proposed linking temperature and relative humidity on the day of

bloom to hardlock incidence in those same bolls after opening (Marois et al., 2004). Temperature

and relative humidity between 0700 and 1900 were shown to be positively associated with

hardlock severity. This model was consistent with the geographic hypothesis of coastal areas

having more hardlock due to their climate. It was also intuitive since F. verticillioides germinates

and grows more readily in high temperature, high moisture conditions. Subbarao and Michailides

(1995) determined the optimal temperature for F. verticillioides (moniliforme) infection on

pollinator figs is 300C. The temperatures observed in the model were usually below 300C.

The objectives of this study were to explore the connection between weather conditions

and hardlock, and to clarify the relationships between thrips, hardlock and yield.

Materials and Methods

Flowers were tagged using ribbons on a weekly basis. This occurred at 4 separate

locations in Florida (Altha, Jay, Marianna, and Quincy) for 2 to 4 years. Temperature and relative

humidity were recorded at 15-minute intervals using a CR10X data logger and 2 HMP 45 AC

temperature-relative humidity probes. This allowed comparisons between each of the 2

temperature and 2 relative humidity measurements to improve accuracy. The weather

measurement system was located within 100 m of the study flowers. After defoliation, all bolls

from tagged flowers were removed from the plants and evaluated for hardlock severity. Hardlock

severity was calculated as the number of hardlocked locules of the total present. Data were

recorded separately for each boll. Bolls were categorized by date of bloom, and the average

hardlock severity was calculated for each date.

The number of bolls recovered for each date varied considerably. Typically 20 to 40 bolls

were recovered, and dates with fewer than 5 bolls were excluded from the analysis. The 2

temperature and 2 relative humidity values for each time interval were averaged to produce 1









temperature and 1 relative humidity value. After choosing the time interval of interest, the mean

for each weather variable was determined using SAS for each day for which flowers had been

tagged. The PROC REG command was used to perform regressions for each location and year.

The relationship between thrips numbers, hardlock, and yield was evaluated in fungicide-

insecticide studies at branch stations of the North Florida Research and Education Center in

Marianna and Quincy, FL. These studies were reported in Ch.2 and Ch.3, but all 3 variables were

not compared. In this analysis it is the thrips numbers, and not the treatments, that are of interest.

Results

Weather Models of Hardlock Severity

The model involving the temperature and relative humidity from 0700 to 1900, as reported

in (Marois et al., 2004), was tested for each location and year of this study (Table 4.1). It was

significant for the data set from which it was developed (Quincy, 2002), showing a connection

between higher temperature and humidity to increased hardlock serverity. At that same location

in 2003 it was also significant. However, the relationship between hardlock and the temperature

component of the model was reversed from the previous year. In other location years the model

was frequently insignificant, and the relationships between model components varied regularly.

Grouping the data sets by year or location did not improve significance. Grouping all 63 days

from the data sets showed little or no relationship. The temperature or relative humidity

components were also analyzed separately, but their significance and adjusted R2 values were

generally low, and their relationships to hardlock varied from positive to negative (Tables 4.2 and

4.3).

In Chapter 2, the temperature between 0800 and 1000 was shown to be negatively

associated with thrips numbers at 1000. To determine if weather conditions may influence

hardlock indirectly, through its impact on thrips numbers in flowers, the mean temperature









between 0800 and 1000 was compared to hardlock incidence for that day (Table 4.4). When

viewed by location-year, the relationships were mostly negative, and more significant than the

first model. When examined in groupings by location or year, the models were typically better or

equal to the individual-location year models. Grouping all days produced a significant (p=0.07),

negative relationship with an extremely low R2 value (0.05). The average relative humidity

during this period was also examined (Table 4.5). The relationships between humidity and

hardlock fluctuated between positive and negative, and the adjusted R2 values were usually low.

Constructing a model using both temperature and relative humidity from 0800 to 1000 was no

better than using temperature alone, and the relative humidity component varied between a

positive and negative association (Table 4.6).

Although the temperature from 0800 to 1000 predicted hardlock incidence better than the

first model, it was unclear if other time periods might be a better predictor. To determine this,

regression analyses were performed for each location-year comparing hardlock incidence to

temperature (Table 4.7) for each hour from 1700 on the day prior to bloom to 2400 on the day of

bloom. Hours showing a significant (p < 0.10) relationship were compared for consistency across

each of the location years. Significant relationships were observed from 1700 (prior to bloom

day) to 1000 and 1300 to 2400 on the day of bloom. These ranges varied considerably by

location-year, and when analyzed together only the period from 0000 to 0600 was shown to be

important, based on significance and adjusted R2 values (>0.20). The peak significance time

within this period varied by model-year, so an average for the period was calculated for each. The

resulting model tested a linear relationship between temperature from 0000 to 0600 and hardlock

incidence for that same day (Table 4.8). In all location-years, locations, years, and combined data,

night time temperature was negatively associated with hardlock incidence. In 4 of 11 location-









years, the model was significant at p < 0.20 and adjusted R2 values were >0.50. Grouped by

location, p < 0.05, and adjusted R2 values varied from 0.15 to 0.79. Examined by year the results

were similar, with p < 0.10, and adjusted R2 values varied from 0.12 to 0.28. The combined data

set showed a significance of p = <0.0001 and an adjusted R2 value of 0.27. To determine if this

result was complicated by plant maturity or varying hardlock intensity within the growing

seasons, the temperatures (Fig. 4.3) and hardlock severity (Fig. 4.4) were plotted against the date.

These did not appear to influence the model. Adding the date (in the form of Julian day) as a

model component did not increase the adjusted R2 value (data not shown). A problem with this

model was that the residuals between the predicted and actual hardlock did not fit a normal

distribution, and were larger toward the extremes of the temperature range.

To further refine the model, days on which fewer than 10 bolls were recovered were

removed from the analysis (Table 4.9). This improved the adjusted R2 value from 0.26 to 0.30.

Then days with temperatures below 21 or above 250C were also excluded, leaving 49 of the 63

(77%) original days (Fig. 4.5). This improved the adjusted R2 value to 0.40 (Table 4.9). When

the resulting equation was used to predict hardlock in that same data, the residuals were normally

distributed, with a mean of 0.00 and a standard deviation of 0.20 (Table 4.10). To further explore

the data, days with 10 or more bolls were assigned to two temperature groups for separate

regression analyses: 17.1 to 22.90C or 23.0 to 26.00C. Days between 17.1 and 22.90C showed a

gradual slope, significant (p=0.06) relationship, and low adjusted R2 value (0.07). The second

group (23.0 to 26.00C) showed a more gradual slope, an insignificant relationship (p=0.66), and

an adjusted R2 value of -0.04. The mean hardlock severity for low and high temperature groups

were 62 and 30%, respectively. The change in hardlock severity at 230C is fairly abrupt. As an

alternative model, days above 230C were categorized as having 30% hardlock severity, while









those below 230C were categorized as having 62% hardlock. The resulting residuals were similar

to those achieved using the linear regression model (Table 4.10). For days within the 21 to 250C

range, the linear regression model had a slightly better residual mean and standard deviation than

did the categorical model. When applying these models to the full temperature range (17 to

260C), the categorical model performed better. This suggests the specific temperature of the day

should determine which model to use.

To validate the model (linear regression, >10 bolls, 21 to 250C), 25 individual days were

randomly sampled from the data set. A regression was performed, and the resulting equation was

used to predict hardlock in the other 24 days. The residuals between the predicted and actual

hardlock severity for those days were calculated (Table 4.11). This procedure was performed a

total of four times. The mean residual ranged from -0.007 to -0.090, and the standard deviations

were between 0.16 and 0.23.

The association of flower thrips with hardlock for individual days was also tested. The

flowers tagged in Marianna were from rows adjacent to the fungicide-insecticide hardlock study

described in Ch.2 and Ch.3. The 2 adjacent plots were a control and an insecticide+fungicide

treatment. The mean number of thrips from the 2 adjacent plots was used as an estimate of thrips

numbers within the tagged rows, since they were not sampled. Thrips were not a good predictor

of hardlock (Table 4.12). When both thrips numbers and temperature from 0000 to 0600 were

included in the model, the results were better than using thrips alone but worse than using

temperature alone. However, these models using thrips have several shortfalls. First, if thrips are

important, they should be significant on their own. Thrips numbers vary throughout an area, and

their numbers appear to fluctuate independently of the overall trend. It is possible the estimates

derived from adjacent plots were not representative of the tagged flowers. A second problem is









the temperature model did not perform as well in Marianna as other locations, so combining it

with the thrips numbers which themselves did not predict well does not seem likely to improve the

accuracy. The models were based on a relatively small number of days, and adding additional

variables to the models makes them less reliable.

Connections between Thrips, Hardlock, and Yield

The relationship between thrips numbers, hardlock, and yield on a per plot basis at the

season-level was examined (Table 4.13). In Marianna, higher thrips numbers were associated

with more hardlock. The model was significant in 2 of 3 years, and also when the data for all the

years was merged. Both R2 and adjusted R2 values remained below 0.40. Hardlock was inversely

associated with yield, and this was significant in all years except 2003. In 2003, the yields were

very low in all treatments (Ch.3), and it is likely that factors other than hardlock were more

important. When the data was merged across years, the result was highly significant (p < 0.0001),

and R2 and adjusted R2 values were both 0.57. The association between thrips numbers and yield

was less consistent, but negative overall. In only 1 of 3 years was the relationship significant (p =

0.0129), and in the non-significant years higher thrips numbers were associated with higher

yielding plots. Combining the data from all 3 years improved the significance of the model (p =

<0.0001), and boosted the R2 and adjusted R2 values to 0.48. Using thrips numbers and hardlock

to predict yield proved highly significant in 2005 (p = 0.0009) and when data was combined

across years (p = <0.0001). The highest adjusted R2 value by year was 0.34, but reached 0.64

when the data was combined across years. Similar results were obtained in Quincy. Higher thrips

numbers were associated with higher hardlock in 2004 (p = 0.02) and 2005 (p = 0.08). R2 and

adjusted R2 values were higher than in Marianna for both years. Higher-yielding plots were

generally associated with lower hardlock, although the relationship was weaker than in Marianna.









Thrips numbers were not directly related to yield. Using thrips numbers and hardlock as a model

to predict yield did not prove significant in Quincy.

Discussion

In this study, the association of weather conditions and hardlock incidence was explored.

The first reported hardlock-weather model (Marois et al., 2004) was tested, but did not prove

adequate in all locations. A second model, previously used for predicting thrips numbers, was

also evaluated. Although it was an improvement from the first model, its predictive ability was

very limited. Examining the temperatures by hour revealed the time between 0000 and 0600 to be

most important for predicting hardlock. A model was constructed using the average temperature

for that time period. This model out-performed those tested previously. It was further refined by

limiting it to days within the temperature range of 21 to 250C. It was also determined that days

can be assigned to a high or low hardlock severity group with some success based on whether they

are below or above 23C.

The first two models tested had straightforward biological explanations. In the first model,

temperature and relative humidity were positively associated with hardlock incidence. Such

conditions would be favorable for the germination and growth of the causal agent, F.

verticillioides. In the second model, temperature from 0800 to 1000 were negatively associated

with hardlock. This temperature range had previously been shown to be negatively associated

with thrips numbers at 1000. Since warm temperatures slow the movement of thrips into flowers,

it is assumed their harmful effects would be diminished, resulting in less hardlock. However,

neither model fit the data as effectively as did the third. It is possible that cooler night

temperatures have a subtle effect on the developing flower making it more susceptible to

infection. Oosterhuis and Jernstedt (1999) noted cool temperatures can delay anthesis (period

when pollen is shed) by two to three hours. This would delay pollination, possibly increasing the









amount of time the stigma is receptive and susceptible to infection by Fusarium. Although the

low temperatures from 0800 to 1000 that are associated with higher thrips numbers did not

consistently predict hardlock, in some instances they might be important. It may be possible to

develop a model to predict hardlock based on temperature and the number of thrips known to be

inhabiting cotton flowers. At the Marianna site (Chapter 3), larger increases in the proportion of

non-hardlocked cotton due to insecticide applications were obtained in the years with the highest

hardlock severity. If temperatures from 0000 to 0600 are used to estimate hardlock severity, this

could allow a more optimal use of insecticide applications. Models have been developed

previously to optimize management of cotton. The GOSSYM/COMAX model has been used for

estimating irrigation needs (Staggenborg et al, 1996) and exploring the result of various fertilizer

and defoliation strategies on yield (McKinion et al., 1989)

The relationships between thrips numbers, hardlock severity, and yield were also explored.

The models comparing thrips to hardlock were mostly significant. Among the significant ones,

the adjusted R2 values ranged from 0.09 to 0.54, suggesting thrips are an important, but not

exclusive, factor in hardlock severity. This lends further support to the findings of Chapter 3, in

which significant treatment effects were observed in the severity of hardlock. There are probably

other factors influencing the probability of infection, progression of the disease, and expression of

symptoms after boll opening. The relationship between hardlock and yield was significant in

many cases, and those location-years in which it wasn't were characterized by below average

yield. This suggests hardlock is one of many factors that can influence yields. It is also

interesting to note that when viewed by year, there was no relationship between thrips numbers

and yield. This suggests they have no direct impact on the cotton plants, and any effect is due to

hardlock.









In conclusion, the hardlock severity for a given day can be predicted with an adjusted R2 of

0.40 using the mean temperature between 0000 and 0600. Cooler temperatures are associated

with higher hardlock severity, and these predictions are most valid for temperatures between 21

and 25C. Below 21C severity is typically above 50%, while above 25C severity is below 50%.

This model could be used to optimize spraying, but needs to be tested further. The impact of

thrips control at different temperatures is unknown, but this must be determined for the model to

optimize spraying. This could be done by tagging flowers in control and insecticide-treated plots,

and quantifying the number of thrips in each plot on that day. Night temperatures can vary

considerably from day to day, and tagging flowers on consecutive days may reduce interference

from any other sources. Each flower is at risk of infection for one day, and this occurs over an

eight week period. Targeting protective strategies to those days with the highest risk allows the

most judicious use of pesticides, thereby maximizing economic and environmental benefits.






















Figure 4.1. Cotton boll exhibiting symptoms of hardlock.











Table 4.1. Relationship between average temperature and relative humidity from 0700 to 1900 on
the day of bloom and hardlock severity for that day.
N
location year days equation p R2 adj.-R2
Altha 2003 6 H = 3.38485 -T719*0.03388 -RH719*0.02245 0.6703 0.2341 -0.2765
Altha 2004 4 H = 1.49718 +T719*0.10730 -RH719*0.06072 0.3958 0.8433 0.5300
Jay 2003 8 H = 1.61997 -T719*0.03756 -RH719*0.00376 0.5768 0.1976 -0.1234
Jay 2004 8 H = 8.06744 -T719*0.23627 -RH719*0.01480 0.1048 0.5943 0.4320
Jay 2005 4 H = 0.86125 +T719*0.01482 -RH719*0.01169 0.2698 0.9272 0.7816
Marianna 2003 4 H = -12.52706 +T719*0.28892 +RH719*0.06857 0.6429 0.5867 -0.2400
Marianna 2004 4 H = 0.12447 +T719*0.02388 -RH719*0.00769 0.7222 0.4784 -0.5648
Marianna 2005 5 H = 31.72419 -T719*0.70494 -RH719*0.14080 0.5165 0.4835 -0.0330
Quincy 2002 8 H = -5.53090 +T719*0.13061 +RH719*0.03539 0.0232 0.7780 0.6892
Quincy 2003 8 H = 1.62170 -T719*0.05959 +RH719*0.00662 0.0240 0.7749 0.6849
Quincy 2004 4 H = 5.07364 -T719*0.10155 -RH719*0.02803 0.1916 0.9633 0.8899
Altha 10 H = 6.32756 -T719*0.11106 -RH719*0.03409 0.5888 0.1404 -0.1052
Jay 20 H = 3.64325 -T719*0.08808 -RH719*0.01061 0.0316 0.3341 0.2557
Marianna 13 H = 2.78049 -T719*0.08754 +RH719*0.00429 0.1003 0.3687 0.2424
Quincy 20 H = -2.14070 +T719*0.05701 +RH719*0.01397 0.7066 0.0400 -0.0729
2003 26 H = -1.15743 +T719*0.02203 +RH719*0.01389 0.4931 0.0596 -0.0221
2004 20 H = 4.36586 -T719*0.10977 -RH719*0.01140 0.0602 0.2816 0.1970
2005 9 H = 1.83714 -T719*0.03499 -RH719*0.00510 0.6282 0.1435 -0.1419
all all 63 H = 1.46886 -T719*0.02710 -RH719*0.00259 0.5649 0.0189 -0.0138


Table 4.2. Relationship between average temperature from 0700 to 1900 on the day of bloom and
hardlock severity for that day.
N Percent
location year days Temp. C hardlock equation p R2 adj.-R2
Altha 2003 6 25.5-30.6 54-97 H = -0.52745 +T719*0.0446 0.3620 0.2090 0.0113
Altha 2004 4 29.2-30.6 6-80 H = -7.53823 +T719*0.27067 0.5045 0.2455 -0.1317
Jay 2003 8 26.2-31.1 4-44 H = 1.00990 -T719*0.02586 0.2811 0.1894 0.0543
Jay 2004 8 25.7-28.9 12-94 H = 5.88195 -T719*0.19692 0.0669 0.4541 0.3631
Jay 2005 4 26.6-29.2 37-59 H = 2.30545 -T719*0.06512 0.3206 0.4616 0.1925
Marianna 2003 4 26.1 -29.9 72-91 H = 1.52993 -T719*0.02607 0.4803 0.2701 -0.0949
Marianna 2004 4 28.5-30.5 26-39 H = 1.29241 -T719*0.03191 0.5291 0.2217 -0.1674
Marianna 2005 5 28.1 -31.4 19-80 H = 1.82540-T719*0.04629 0.6963 0.0580 -0.2560
Quincy 2002 8 27.6-33.0 40-96 H = 0.05871 +T719*0.02353 0.6928 0.0279 -0.1342
Quincy 2003 8 26.1 -30.5 29-70 H = 2.88915 -T719*0.08660 0.0043 0.7684 0.7298
Quincy 2004 4 29.1 -31.0 2-34 H = 3.06464 -T719*0.09699 0.4876 0.2626 -0.1061
Altha 10 H= 0.21026 +T719*0.01483 0.7937 0.0091 -0.1148
Jay 20 H= 2.18815-T719*0.06390 0.0360 0.2219 0.1787
Marianna 13 H = 3.48033 -T719*0.10098 0.0316 0.3553 0.2967
Quincy 20 H = 0.47849 +T719*0.00140 0.9729 0.0001 -0.0555
2003 26 H = 1.28771 -T719*0.02744 0.3221 0.0408 0.0009
2004 20 H = 2.77654 -T719*0.08238 0.0459 0.2035 0.1593
2005 9 H = 1.59363-T719*0.03912 0.4173 0.0959 -0.0332
all all 63 H = 1.05481 -T719*0.01921 0.3283 0.0157 -0.0005











Table 4.3. Relationship between average relative humidity from 0700 to 1900 on the day of bloom
and hardlock severity for that day.
N Percent
location year days RH % hardlock equation p R2 adj.-R2
Altha 2003 6 68-86 54-97 H = 1.72035-RH719*0.01310 0.3363 0.2296 0.0370
Altha 2004 4 65-73 6-80 H = 5.04400 -RH719*0.06591 0.1001 0.8098 0.7146
Jay 2003 8 62-87 4-44 H = -0.18292 +RH719*0.00621 0.3837 0.1282 -0.0170
Jay 2004 8 62-85 12-94 H = 0.84503 -RH719*0.00462 0.7695 0.0154 -0.1487
Jay 2005 4 56-79 37-59 H = 1.18929 -RH719*0.01039 0.0421 0.9176 0.8764
Marianna 2003 4 68-86 72-91 H = 0.32869 +RH719*0.00617 0.4322 0.3224 -0.0164
Marianna 2004 4 54-71 26-39 H = 0.68295 -RH719*0.00530 0.3335 0.4443 0.1664
Marianna 2005 5 66-81 19-80 H = 0.10734 +RH719*0.00463 0.8544 0.0131 -0.3158
Quincy 2002 8 60-79 40-96 H = -0.33560 +RH719*0.01663 0.1660 0.2929 0.1750
Quincy 2003 8 67-85 29-70 H = -1.08503 +RH719*0.02004 0.0057 0.7457 0.7033
Quincy 2004 4 60-70 2-34 H = 1.98995 -RH719*0.02752 0.1780 0.6757 0.5135
Altha 10 H = 1.29985-RH719*0.00906 0.5130 0.0553 -0.0628
Jay 20 H = 0.55598-RH719*0.00208 0.7558 0.0055 -0.0497
Marianna 13 H =-0.42226+RH719*0.01323 0.1398 0.1872 0.1133
Quincy 20 H = 0.26393 +RH719*0.00362 0.6800 0.0097 -0.0453
2003 26 H= -0.12656 +RH719*0.00847 0.2429 0.0564 0.0170
2004 20 H= 0.48189-RH719*0.00118 0.8911 0.0011 -0.0544
2005 9 H = 0.88559-RH719*0.00605 0.4965 0.0684 -0.0646
all all 63 H = 0.39754 +RH719*0.00144 0.7409 0.0018 -0.0146

Table 4.4. Relationship between average temperature from 0800 to 1000 on the day of bloom and
hardlock severity for that day.
N Percent
location year days Temp. C hardlock equation p R2 adj.-R2
Altha 2003 6 24.2-28.4 54-97 H = -0.86394 +T810*0.06303 0.2939 0.2670 0.0838
Altha 2004 4 25.5-27.6 6-80 H = 9.68415 -T810*0.34434 0.0738 0.8579 0.7868
Jay 2003 8 23.8-27.4 4-44 H = 1.47193-T810*0.04634 0.3596 0.1409 -0.0023
Jay 2004 8 22.2-27.1 12-94 H = 2.94612 -T810*0.09854 0.1263 0.3442 0.2349
Jay 2005 4 21.7-26.8 37-59 H = 1.56850 -T810*0.04426 0.0266 0.9475 0.9212
Marianna 2003 4 25.5-30.3 72-91 H = 1.87963-T810*0.03817 0.0061 0.9878 0.9817
Marianna 2004 4 24.8-27.3 26-39 H = 1.43672-T810*0.04154 0.2332 0.5879 0.3819
Marianna 2005 5 25.1 -28.6 19 80 H = 1.97583 -T810*0.05644 0.5042 0.1603 -0.1196
Quincy 2002 8 23.4-27.8 40-96 H = 0.18795 +T810*0.02307 0.7143 0.0240 -0.1387
Quincy 2003 8 23.5-27.2 29-70 H = 2.45497 -T810*0.07822 0.0967 0.3921 0.2908
Quincy 2004 4 26.0 27.4 2 34 H = 2.37327 -T810*0.08271 0.6656 0.1118 -0.3323
Altha 10 H= 1.70746-T810*0.04154 0.5209 0.0533 -0.0650
Jay 20 H = 2.47235 -T810*0.08200 0.0061 0.3488 0.3127
Marianna 13 H = 0.06460 +T810*0.01672 0.6983 0.0142 -0.0754
Quincy 20 H = 3.00672 -T810*0.09678 0.0536 0.1916 0.1467
2003 26 H= 0.08964 +T810*0.01632 0.5871 0.0125 -0.0287
2004 20 H = 3.18640-T810*0.10754 0.0048 0.3651 0.3292
2005 9 H = 1.43866-T810*0.03766 0.1939 0.2278 0.1174
all all 63 H = 1.44105-T810*0.03629 0.0692 0.0531 0.0376











Table 4.5. Relationship between average relative humidity from 0800 to
and hardlock severity for that day.


Percent
hardlock
54 97
6 80
4 44
12 94
37 59
72 91
26 39
19-80
40 96
29 70
2 34


RH %
77 94
77-81
76 92
70 99
71 -89
70 87
74 83
77 94
76 94
79 96
76 85


1000 on the day of bloom


location
Altha
Altha
Jay
Jay
Jay
Marianna
Marianna
Marianna
Quincy
Quincy
Quincy
Altha
Jay
Marianna
Quincy




all


year
2003
2004
2003
2004
2005
2003
2004
2005
2002
2003
2004





2003
2004
2005
all


days
6
4
8
8
4
4
4
5
8
8
4
10
20
13
20
26
20
9
63


Table 4.6. Relationship between average temperature and relative humidity from
the day of bloom to hardlock severity for that day.


equation
H = 5.03406 -T810*0.05457 -RH810*0.03313
H = 7.38611 -T810*0.47509 +RH810*0.07181
H = 2.00074 -T810*0.05174 -RH810*0.00454
H = 3.38390 -T810*0.10157 -RH810*0.00439
H = 1.47966 -T810*0.05652 +RH810*0.00476
H = 3.25879 -T810*0.06587 -RH810*0.00768
H = 1.41568 -T810*0.02101 -RH810*0.00654
H = -0.23261 -T810*0.00833 +RH810*0.01064
H = -3.70374 +T810*0.08013 +RH810*0.02873
H = 2.44961 -T810*0.07812 -RH810*0.00003128
H = 6.73137 -T810*0.15094 -RH810*0.03074
H = 6.22996 -T810*0.12984 -RH810*0.02631
H = 2.69654 -T810*0.08001 -RH810*0.00327
H = -0.45543 -T810*0.02583 -RH810*0.00335
H = 1.37903 -T810*0.06973 -RH810*0.01089
H = 1.42445 -T81 0*0.00911 -RH810*0.00786
H = 3.85551 -T810*0.10964 -RH810*0.00754
H = 1.19601 -T810*0.03831 -RH810*0.00309
H = 1.53183 -T810*0.03742 -RH810*0.00072999


P
0.4252
0.1097
0.6452
0.3290
0.0091
0.0584
0.6292
0.8375
0.0073
0.2881
0.3545
0.4737
0.0231
0.9079
0.1126
0.7626
0.0150
0.4315
0.1928


0800 to 1000 on


R2 adi.-R2


0.4345
0.9880
0.1608
0.3590
0.9999
0.9966
0.6042
0.1625
0.8599
0.3921
0.8743
0.1922
0.3579
0.0191
0.2266
0.0233
0.3900
0.2443
0.0534


0.0576
0.9639
-0.1749
0.1026
0.9998
0.9898
-0.1875
-0.6751
0.8038
0.1490
0.6230
-0.0386
0.2824
-0.1770
0.1356
-0.6160
0.3183
-0.0075
0.0218


equation
H = 2.50525 -RH810*0.02010
H = 4.96424 -RH810*0.05535
H = 0.33838 -RH810*0.00072459
H = 0.59850 -RH810*0.00121
H = 1.12832 -RH810*0.00798
H = 0.00280 +RH810*0.01027
H = 1.34000-RH810*0.01254
H = -0.61222 +RH810*0.01245
H = -1.16870 +RH810*0.02271
H = -0.94665 -RH810*0.01597
H = 2.17161 -RH810*0.02457
H = 0.78812 -RH810*0.00177
H = 0.92170 -RH810*0.00618
H = 0.60392 -RH810*0.00103
H = -1.13329 +RH810*0.01930
H = 1.00986-RH810*0.00579
H = 0.83577 -RH810*0.00535
H = 0.25670 +RH810*0.00236
H = 0.24149 +RH810*0.00309


P
0.1750
0.6117
0.9559
0.9366
0.4064
0.0266
0.2326
0.5012
0.0214
0.1952
0.2680
0.9095
0.4426
0.9243
0.0838
0.4669
0.6368
0.8013
0.5502


R2
0.4040
0.1508
0.0006
0.0011
0.3524
0.9476
0.5889
0.1624
0.6140
0.2615
0.5358
0.0017
0.0331
0.0009
0.1569
0.0223
0.0127
0.0097
0.0059


adj.-R2
0.2550
-0.2739
-0.1660
-0.1653
0.0285
0.9214
0.3833
-0.1168
0.5496
0.1384
0.3037
-0.1231
-0.0206
-0.0900
0.1101
-0.0185
-0.0422
-0.1318
-0.0104


location
Altha
Altha
Jay
Jay
Jay
Marianna
Marianna
Marianna
Quincy
Quincy
Quincy
Altha
Jay
Marianna
Quincy




all


year
2003
2004
2003
2004
2005
2003
2004
2005
2002
2003
2004





2003
2004
2005
all


N
days
6
4
8
8
4
4
4
5
8
8
4
10
20
13
20
26
20
9
63










Table 4.7. Association of temperature and hardlock severity by hour for each location.


hour
17b
18b
19b
20b
21b
22b
23b
24b
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24


Altha
p relat.
0.63 +
0.66 -
0.72 -
0.55 -
0.10 -
0.04 -
0.02 -
0.01 -
0.00 -
0.00 -
0.00 -
0.00 -
0.00 -
0.00 -
0.00 -
0.00 -
0.23 -
0.92 +
0.99 -
0.89 +
0.92 -
0.90 +
0.64 +
0.04 +
0.07 +
0.81 +
0.58 -
0.29 -
0.20 +
0.04 +
0.05 +
0.09 +


Jay Marianna
p relat. p relat.
0.28 0.00
0.29 0.00
0.34 0.00
0.22 0.00
0.13 0.00
0.10 0.00
0.13 0.00
0.16 0.01
0.11 0.02
0.07 0.01
0.05 0.02
0.03 0.03
0.04 0.13
0.03 0.38
0.04 0.43 +
0.04 0.52 +
0.01 0.76 +
0.00 0.92 +
0.06 0.65
0.23 0.15
0.34 0.04
0.23 0.03
0.29 0.03
0.47 0.03
0.88 0.06
0.99 + 0.04
0.42 0.01
0.42 0.01
0.07 0.12
0.00 0.13
0.00 0.17
0.00 0.11


Quincy
p relat.
0.99 +
0.32 +
0.50 +
0.59
0.02
0.13
0.27
0.18
0.00
0.00
0.00
0.00
0.00
0.01
0.04
0.03
0.00
0.07
0.94
0.97
0.77 +
0.59
0.51
0.83
0.83 +
0.85 +
0.66
0.08
0.00
0.00
0.00
0.00


p
0.18
0.35
0.25
0.02
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.02
0.06
0.61
0.49
0.41
0.16
0.24
0.82
0.77
0.91
0.25
0.03
0.05
0.00
0.00
0.00


all locations
relat. R2
0.02
0.01
0.02
0.08
0.22
0.17
0.17
0.16
0.23
0.27
0.30
0.29
0.26
0.22
0.12
0.08
0.07
0.05
0.00
0.00
0.01
0.03
0.02
0.00
+ 0.00
0.00
0.02
0.07
0.05
0.11
0.15
0.14


adj-R2
0.01
0.00
0.00
0.06
0.21
0.16
0.15
0.14
0.22
0.26
0.29
0.27
0.24
0.21
0.10
0.07
0.06
0.03
-0.01
0.00
0.00
0.01
0.00
-0.01
-0.01
-0.01
0.00
0.05
0.04
0.09
0.14
0.13











Table 4.8. Relationship between average temperature from 0000 to 0600 on the day of bloom to


hardlock severity for all days.
N Percent
location vear days Temo. OC hardlock eauation


D R2 adi.-R2


Altha 2003 6 21.8-22.9 54-97 H = 7.90058-T0106*0.32403
Altha 2004 4 20.6-24.8 6-80 H = 4.61249-T0106*0.18074
Jay 2003 8 21.9-24.6 4-44 H= 1.38649-T0106*0.04749
Jay 2004 8 18.7-25.2 12-94 H= 1.45101 -T0106*0.04274
Jay 2005 4 19.7-24.5 37-59 H = 1.57428-T0106*0.04930
Marianna 2003 4 22.2-23.5 72-91 H = 1.08427-T0106*0.01260
Marianna 2004 4 22.6-26.0 26-39 H = 0.42002 -T0106*0.00327
Marianna 2005 5 21.8-24.3 19-80 H=3.10125-T0106*0.11400
Quincy 2002 8 17.1-22.3 40-96 H = 1.65947-T0106*0.04368
Quincy 2003 8 21.7-23.3 29-70 H = 1.09049-T0106*0.02837
Quincy 2004 4 19.7-23.9 2-34 H = 1.64555-T0106*0.06688
Altha 10 H= 5.30923-T0106*0.20881
Jay 20 H = 1.75501-T0106*0.05940
Marianna 13 H = 2.99630-T0106*0.10534
Quincy 20 H = 2.60580-T0106*0.09657
2003 26 H= 4.27822-T0106*0.16566
2004 20 H = 1.72184 -T0106*0.05796
2005 9 H = 1.94613 -T0106*0.06506
all all 63 H = 2.33353 -T0106*0.08148


0.0253 0.7520 0.6901
0.0385 0.9244 0.8866
0.4719 0.0894 -0.0624
0.4591 0.0944 -0.0565
0.0259 0.9489 0.9233
0.9087 0.0083 -0.4875
0.9182 0.0067 -0.4900
0.2816 0.3637 0.1517
0.2845 0.1870 0.0515
0.8331 0.0080 -0.1573
0.1646 0.6979 0.5468
0.0004 0.8136 0.7903
0.0495 0.1976 0.1531
0.0426 0.3233 0.2617
0.0051 0.3614 0.3259
0.0066 0.2689 0.2385
0.0682 0.1729 0.1270
0.0809 0.3726 0.2829
<0.0001 0.2770 0.2652


III










































18.0 19.0 20.0 21.0 22.0

Temperature (C)


23.0 24.0


25.0 26.0


18.0 19.0 20.0 21.0 22.0 23.0 24.0 25.0 26.0

Temperature (C)


Figure 4.2. Mean temperature from 0000 to 0600 and hardlock severity. A, Days by location. B,
Regression line.


1.00


0.90


0.80


0.70


o 0.60


.c 0.50


o 0.40
a.
0.30
0.30


0.20


0.10


0.00
17


A 0 *
A A E A


A ,,larianna

A 'JiIInC ,
r
AA

A




A A

*
AA A
U E


.0


1.00


0.90

0.80

v
0.70

0.60


0.50


0.40

0.30


0.20

0.10


0.00 -
17.0













27.0



25.0



23.0



c 21.0

E

19.0



17.0



15.0
2-Jul


12-Jul 22-Jul 1-Aug 11-Aug 21-Aug 31-Aug 10-Sep 20-Sep

Date


Figure 4.3. Distribution of mean temperatures from 0000 to 0600 for days in the climate model.


12-Jul 22-Jul 1-Aug 11-Aug

Date


21-Aug 31-Aug


10-Sep 20-Sep


Figure 4.4. Incidence of hardlock during the growing season for days evaluated.


*






.. .

*



v = -0 02.05- 821 66
R- = 0 050


1.00

0.90

0.80

0.70

0.60

0.50

0.40

0.30

0.20

0.10

0.00
2-Jul


* *


S0 0021 81 01

* *
*











Table 4.9. Relationship between average temperature from 0000 to 0600 on the day of bloom to


hardlock severity for temperatures between 21 and 250C.
N bolls
davs TemD. hardlock recovered eauation


D R2 adi.-R2


17.1 -26.0
17.1 -26.0
21.4-24.8
17.1 -22.9
23.1 -26.0




1.00
0.90 -
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
17.0


0.02- 0.97
0.02- 0.97
0.02- 0.97
0.29- 0.97
0.02- 0.80


19.0


H = 2.32467 -T*0.08106
H = 2.42539 -T*0.08543
H = 4.57773 -T*0.17898
H = 1.66745-T*0.04858
H = 0.97826 -T*0.02803


21.0


23.0


<0.0001
<0.0001
<0.0001
0.0602
0.6627


25.0


0.2731
0.3145
0.4173
0.0973
0.0102


27.0


Temperature (C)


Figure 4.5. Days in model with >10 bolls recovered.


0.2611
0.3022
0.4049
0.0715
-0.0419


**
*




* .
*


*


III























* *
* 21.5 *22.0*


y = 0.0007x 0.0155
R- = 1E-05


q. *


* *


23.0 23.5 24.0 4.5
24 0


22.5


-0.60
Temperature (C)


Figure 4.6. Prediction residuals for days with >10 bolls, 21 to 250C (N=49).







Table 4.10. Mean and standard deviation of models when predictions are applied to data.
linear categorical
21-25C mean 0.00 0.01
st. dev. 0.20 0.22
17-26oC mean 0.05 0.00
st. dev. 0.27 0.22


Table 4.11. Validation of the model by random sampling and applying predictions to data.


models established from random sampling of days


prediction residuals


N equation p R2 adj.-R2 N mean SD
25 H = 3.55426 -T*0.13543 0.0071 0.2752 0.2437 24 -0.0581 0.2218
25 H = 4.77974 -T*0.18862 0.0034 0.316 0.2863 24 -0.0372 0.1644
25 H = 3.33086 -T*0.12455 0.0053 0.2921 0.2613 24 -0.0068 0.2324
25 H = 4.37491 -T*0.17204 0.0003 0.4392 0.4149 24 -0.0904 0.1928


1.00

0.80


0.60

0.40

0.20

0.00

-0.2?0

-0.40


__


25.0











Table 4.12. Relationship of temperature from 0000 to 0600 and estimated thrips numbers to
hardlock in Marianna, FL.


equation
H = 0.62834 +TH*0.05242
H = 0.32735 +TH*0.00524
H = 0.71081 -TH*0.13613
H = 0.35030 +TH*0.06850
H = 3.17168 -T*0.11778 +TH*0.09484
H = 0.30124 +T*0.00101 +TH*0.00580
H = 3.09767 -T*0.10907 -TH*0.05674
H = 2.65757 -T*0.09486 +TH*0.03723


0.:
0.
0.
0
0.:
0.
0.
0.


p R2
3332 0.4446
8473 0.0233
6866 0.0619
.2534 0.1166
3430 0.8823
9881 0.0237
6262 0.3738
1121 0.3545


location
Marianna
Marianna
Marianna
Marianna
Marianna
Marianna
Marianna
Marianna


year
2003
2004
2005
all
2003
2004
2005
all


days
4
4
5
13
4
4
5
13


thrips
2.3-4.4
0.4-4.7
1.4-2.5

2.3-4.4
0.4-4.7
1.4-2.5


adj.-R2
0.1669
-0.4650
-0.2509
0.0363
0.6470
-1.9289
-0.2524
0.2254













Table 4.13. Relationships between thrips, hardlock and yield on a per plot, seasonal basis.


location
Marianna
Marianna
Marianna
Marianna
Marianna
Marianna
Marianna
Marianna
Marianna
Marianna
Marianna
Marianna
Marianna
Marianna
Marianna
S Marianna
o Qu
o Quincy
Quincy
Quincy
Quincy
Quincy
Quincy
Quincy
Quincy
Quincy
Quincy
Quincy
Quincyv


year
2003
2004
2005
all
2003
2004
2005
all
2003
2004
2005
all
2003
2004
2005
all
2004
2005
all
2004
2005
all
2004
2005
all
2004
2005
all


thrips
x
x
x
x





x
x
x
x
x
x
x
x
x
x
x




x
x
x
x
x
x


hardlock
x
x
x
x
x
x
x
x





x
x
x
x
x
x
x
x
x
x




x
x
x


yield equation
hardlock = 0.41165 +thrips*0.05216
hardlock = 0.34980 -thrips*0.00825
hardlock = 0.11918 +thrips*0.11944
hardlock = 0.15831 +thrips*0.09310
x yield = 516.60053 -hardlock*36.06410
x yield = 1366.33802 -hardlock*967.06381
x yield = 1902.35967 -hardlock*1560.96214
x yield = 1711.36408 -hardlock*1740.30903
x yield = 474.83374 +thrips*5.09984
x yield = 955.19283 +thrips*33.86048
x yield = 1783.89191 -thrips*233.84959
x yield = 1645.94504 -thrips*239.86385
x yield = 497.64089 +thrips*7.98992 -hardlock*55.40358
x yield = 1289.87769 +thrips*25.96324 -hardlock*956.78225
x yield = 1947.46192 -thrips*69.92373 -hardlock*1372.49023
x yield = 1836.74673 -thrips*127.65422 -hardlock*1205.25120
hardlock = 0.39907 +thrips*0.01581
hardlock = 0.17458 +thrips*0.03057
hardlock = 0.26420 +thrips*0.02937
x yield = 468.27660 +hardlock*1860.10638
x yield = 1817.62244 -hardlock*1021.01514
x yield = 1731.20696 -hardlock*814.36697
x yield = 1219.43516 +thrips*29.73564
x yield = 1599.82076 -thrips*13.73293
x yield = 1437.60645 +thrips*3.13140
x yield = 489.77366 +thrips*0.82832 +hardlock*1828.39815
x yield = 1854.99692 +thrips*30.95703 -hardlock*1461.69112
x yield = 1775.20536 +thrips*40.65899 -hardlock*1277.79295


P
0.0460
0.7560
0.0007
0.0001
0.7810
0.0026
0.0002
<0.0001
0.7888
0.4962
0.0129
<0.0001
0.8924
0.0096
0.0009
<0.0001
0.0230
0.0751
0.0189
0.2863
0.2288
0.1271
0.4113
0.7689
0.9118
0.5981
0.5227
0.1413


R2
0.1263
0.0033
0.3219
0.3902
0.0026
0.2648
0.3735
0.5664
0.0024
0.0156
0.1892
0.4844
0.0078
0.2739
0.3849
0.6501
0.6052
0.4352
0.3347
0.1858
0.1841
0.1582
0.1150
0.0155
0.0009
0.1859
0.2286
0.2600


adj.-R2
0.0971
-0.0300
0.2993
0.3837
-0.0306
0.2403
0.3526
0.5618
-0.0308
-0.0172
0.1621
0.4790
-0.0606
0.2238
0.3425
0.6426
0.5394
0.3411
0.2872
0.0501
0.0481
0.0980
-0.0325
-0.1486
-0.0705
-0.1398
-0.0800
0.1461