<%BANNER%>

Taxonomic Distribution of Modern Fin-Winged Fruits and the Fossil History of the Combretaceae in the United States Based...

Permanent Link: http://ufdc.ufl.edu/UFE0020721/00001

Material Information

Title: Taxonomic Distribution of Modern Fin-Winged Fruits and the Fossil History of the Combretaceae in the United States Based on Fin-Winged Fruits
Physical Description: 1 online resource (81 p.)
Language: english
Creator: O'Leary, Elizabeth L
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: Botany -- Dissertations, Academic -- UF
Genre: Botany thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Angiosperms have evolved wind dispersed fruits and seeds many times. The wings of fruits may be formed from different structures including outgrowths of the ovary or hypanthium, perianth parts, and bracts or bracteoles. The wings may also form various morphological and biomechanical shapes. One such of these fruits is 'fin-winged' that is, having a longitudinal central axis from which several longitudinally oriented wings radiate. This form can be found in capsules, samaras, as well as in other fruit types where the wings may be formed from accessory structures. This thesis examines the diversity of these fin-winged fruits at the generic level within modern Angiosperms and gives brief descriptions of these fruits to aid in identification of dispersed fin-winged fruits. These descriptions are applied to determine if three fruit fossils, described as the genus Terminalia (Combretaceae), have sufficient characters to be differentiated from other fin-winged genera, and if they truly belong in the Combretaceae. The descriptions are also be used to make inferences about the evolution of these fruits. This survey indicated that fin-winged fruits are present among 23 families of angiosperm and are particularly well represented in the rosids.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Elizabeth L O'Leary.
Thesis: Thesis (M.S.)--University of Florida, 2007.
Local: Adviser: Manchester, Steven R.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0020721:00001

Permanent Link: http://ufdc.ufl.edu/UFE0020721/00001

Material Information

Title: Taxonomic Distribution of Modern Fin-Winged Fruits and the Fossil History of the Combretaceae in the United States Based on Fin-Winged Fruits
Physical Description: 1 online resource (81 p.)
Language: english
Creator: O'Leary, Elizabeth L
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: Botany -- Dissertations, Academic -- UF
Genre: Botany thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Angiosperms have evolved wind dispersed fruits and seeds many times. The wings of fruits may be formed from different structures including outgrowths of the ovary or hypanthium, perianth parts, and bracts or bracteoles. The wings may also form various morphological and biomechanical shapes. One such of these fruits is 'fin-winged' that is, having a longitudinal central axis from which several longitudinally oriented wings radiate. This form can be found in capsules, samaras, as well as in other fruit types where the wings may be formed from accessory structures. This thesis examines the diversity of these fin-winged fruits at the generic level within modern Angiosperms and gives brief descriptions of these fruits to aid in identification of dispersed fin-winged fruits. These descriptions are applied to determine if three fruit fossils, described as the genus Terminalia (Combretaceae), have sufficient characters to be differentiated from other fin-winged genera, and if they truly belong in the Combretaceae. The descriptions are also be used to make inferences about the evolution of these fruits. This survey indicated that fin-winged fruits are present among 23 families of angiosperm and are particularly well represented in the rosids.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Elizabeth L O'Leary.
Thesis: Thesis (M.S.)--University of Florida, 2007.
Local: Adviser: Manchester, Steven R.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0020721:00001


This item has the following downloads:


Full Text





TAXONOMIC DISTRIBUTION OF MODERN FIN-WINGED FRUITS AND THE FOSSIL
HISTORY OF THE COMBRETACEAE INT THE UNITED STATES BASED ON FIN-
WINGED FRUITS




















By

ELIZABETH LEO O'LEARY


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2007

































02007 Elizabeth Leo O'Leary


































To Craig Greene










ACKNOWLEDGMENTS

I wish to thank my advisor, Steven R. Manchester and my committee members Walter S.

Judd and David Dilcher for their useful comments and valuable time. I also wish to thank the

museums and herbaria that allowed me to use their collections for my research, particularly those

at Harvard, Berkeley, and the Smithsonian. I thank the Evolving Earth Foundation for funding

this proj ect. The support of my family and friends during this time was extensive, and much

appreciated. Special thanks go to Kristine Hoffmann for the use of her apartment as a second

office.












TABLE OF CONTENTS



page

ACKNOWLEDGMENTS .............. ...............4.....


LIST OF TABLES .........__.. ..... .__. ...............6....


LI ST OF FIGURE S .............. ...............7.....


AB S TRAC T ......_ ................. ............_........8


CHAPTER


1 INTRODUCTION .............. ...............9.....


M odern Frits .............. ...............9.....
Combretaceae ................. ...............11.......... .....


2 MATERIALS AND METHODS .............. ...............14....


Modern Fruit Descriptions ................. ...............14........... ....
Fossil and Modern Combretaceae .............. ...............15....


3 MODERN FRUIT SURVEY .............. ...............34....


4 RE SULT S .............. ...............62....


Fossil Fin-Winged Fruits of North America With Possible Affinities to Combretaceae .......62
Carpolithes prangosoides Berry ................. ...............62........... ....
Terminalia evatllinina"~ MacGinitie .............. .... ..... .. ..............6
Terminalia oregona (Lakhanpal) Meyer and Manchester ................. ....._._ ...............67


5 CONCLUSIONS .............. ...............70....


Modern Winged Fruits................ ...............70
Fossil W inged Fruits ................. ...............72...............


LIST OF REFERENCES .....__ ................. ........._._.........7


BIOGRAPHICAL SKETCH .............. ...............8 1....










LIST OF TABLES


Table


page


2-1 Modern fin-winged fruits............... ...............17.











LIST OF FIGURES


Figure page


2-1 Hernandiaceae-Onagraceae ........._.. .......__. ...............20.....

2-2 Combretaceae (Combretum). .............. ...............21....

2-3 Combretaceae (Terminalia). ............. ...............22.....

2-4 Combretaceae (other) ................. ...............23........... ....

2-5 Malvales............... ...............24


2-6 Zygophyll aceae--Begoni aceae ........._._ ...... .... ...............25.

2-7 Rhamnaceae-Tropaeolumaceae ........._._. ...._. ........ ............2

2-8 Rutaceae- Apiaceae ................. ...............27....... ......

2-9 Distribution of fin winged fruits on Angiosperm phylogeny.............__ .........___.......28

2-10 Terminalia vera............... ...............29..


2-11 Cai po~lithe1~ prangosoides ........._._ ...... .... ...............30.

2-12 Terminalia vera and Ma'~cropteranthus fitzalanin ................ ....___ ........___.........

2-13 Term inalia esttttttttttttttttttta i n .............. ...............32....

2-14 Term inalia esttttttttttttttttttta i n .............. ...............33....




Authors note: All figures are original photographs by the author. The designations listed in the
figures is the location of where the samples were obtained from.









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

TAXONOMIC DISTRIBUTION OF MODERN FIN-WINGED FRUITS AND THE FOSSIL
HISTORY OF THE COMBRETACEAE INT THE UNITED STATES BASED ON FIN-
WINGED FRUITS

By

Elizabeth Leo O'Leary

December 2007

Chair: Steven Manchester
Major: Botany

Angiosperms have evolved wind dispersed fruits and seeds many times. The wings of

fruits may be formed from different structures including outgrowths of the ovary or hypanthium,

perianth parts, and bracts or bracteoles. The wings may also form various morphological and

biomechanical shapes. One such of these fruits is "fin-winged" that is, having a longitudinal

central axis from which several longitudinally oriented wings radiate. This form can be found in

capsules, samaras, as well as in other fruit types where the wings may be formed from accessory

structures. This thesis examines the diversity of these fin-winged fruits at the generic level

within modern Angiosperms and gives brief descriptions of these fruits to aid in identification of

dispersed fin-winged fruits. These descriptions are applied to determine if three fruit fossils,

described as the genus Terminalia (Combretaceae), have sufficient characters to be differentiated

from other fin-winged genera, and if they truly belong in the Combretaceae. The descriptions

are also be used to make inferences about the evolution of these fruits. This survey indicated that

fin-winged fruits are present among 23 families of angiosperm and are particularly well

represented in the rosids.









CHAPTER 1
INTTRODUCTION

Modern Fruits

Winged fruits are found in many Tertiary fossil localities and are commonly preserved

together with fossil leaves and other debris in lake and pond deposits (e.g. Tidwell 1975, Wilde

and Frankenhaiuser 1997, Manchester 2001). The emphasis in many paleobotanical studies has

been on the leaves, and the co-occurring winged fruits and seeds are often mentioned only in

passing. Nevertheless, fruits and seeds are especially useful in taxonomy because they are part

of the reproductive structures, which often have diagnostic features in angiosperms (Tiffney

1990). Winged fruits also have many characters potentially useful for identification including

size, number of wings, patterns of wing venation, wing shape and position, persistence of styles)

and pedicel, placentation type, seed number and orientation, position of micropyle and raphe and

epidermal characters (Spjut 1994).

Despite this, it is only in the past few decades that fossilized winged fruits have been

treated in more detail. Some examples of recent treatments are Acer (Wolfe and Tanai 1987),

Fraxinus (Call and Dilcher 1992), Eucommia (Call and Dilcher 1997), Ceratopetalum (Barnes

and Hill 1999), Dipteronia (McClain and Manchester 2001) and Ailanthus (Corbett and

Manchester 2004). Recent treatments of extinct genera include Chaneya (Wang and Manchester

2000), Cedrelospermum (Manchester 1989), and Diplodipelta (Manchester and Donoghue 1995).

In these instances, the fruits were more informative than leaves for the taxonomic

determinations.

Many additional kinds of winged fruits are known from the Tertiary of North America that

have not been identified. The difficulty in identifying their familial and generic affinities is

largely due to the inaccessibility of comparative data on modern genera with winged fruits










(Tiffney 1990). An important recent contribution was published on asymmetrically winged fruits

of the Americas (Mirle and Burnham 1999). It was accompanied by some line diagrams to aid in

keying out unknown specimens, but did not include photographic images of the many different

genera surveyed.

There are at least 30 extant genera of winged fruits that are popularly known and readily

accessible in western North America, including Acer, Fraxinus, Ulmus, Carpinus, and Halesia

(Schopmeyer 1977). If a fossil fruit does not match one of these familiar types, the comparative

work for identification becomes a more daunting challenge because winged fruits occur in

numerous genera scattered through at least 93 families of flowering plants (Dallwitz et al. 2000

onwards). There is no single source for comparative data on modern winged fruit genera and not

even a comprehensive worldwide list. If a fossil winged fruit does not match a known living

fruit type, it may be extinct, or it may correspond to an extant genus with which the investigator

is unfamiliar. This lack of familiarity may occur more often when a genus has a fossil record

across several continents, but has a limited distribution in modern times, as in the case of the

extant Asian genus Craigia, whose fossil fruits in Europe and North America were formerly

thought to represent an extinct genus Pteleaecarpum Weyland (Buzek et al. 1989), until their

common identity was discovered (Kvadek et al. 1991).

Various morphological categories of winged fruits can be recognized, such as those with

an apical or basal whorl of propeller-like wings, those with a single surrounding wing, and they

vary in aerodynamic efficiency and mode of falling (Augsburger 1988). In this treatment, I

focused on fruits which have two or more wings radiating from the vertical central axis as in the

case of the anthocarps of the Nyctaginaceae (Galloway 1975). These fruits were referred to as

"tumblers" (Augsburber 1988) in relation to flight pattern, as "finned" (Matlack 1987) and










cyclicallyy winged" (Wurdak et al. 2004). I use the term "fin-winged" because "tumblers" refers

to flight pattern and some of these fruits do not fly; "finned" could refer to ribs and does not fully

convey that they are winged; and cyclicallyy winged" could refer to fruits with wings radiating

laterally. In some cases the entire fruit is the unit of dispersal and in others, the wings are on

mericarps which fall off to disperse the seed. Some are capsular and dehisce entirely and others

remain attached to the twig and dehisce their seeds as the fruit is shaken by the wind. To aid in

systematic work, descriptions of many genera with the fin-winged morphology will be presented.

These descriptions will allow comparisons to be made and will help determine if it is possible to

distinguish between different families and genera based solely on the morphology of their fruit.

The family Combretaceae is one in which fin-winged fruits occur in several genera with some

genera more similar than others. This makes it an ideal family to study in detail.

Combretaceae

Some fossils of interest are those that have been assigned to the white mangrove family

(Combretaceae) from the Eocene of the United States. Today this family is distributed

pantropically, and can be found in the state of Florida. Many of the reported fossils are leaves,

but because the entire-margined, pinnately veined leaves of this family are of a common

morphology encountered in many angiosperm families, additional features are needed to confirm

their identity. The family has diagnostically useful features such as combretaceous hairs and

extrafloral nectaries on the leaves, but these have not been found on these impression fossils.

The Combretaceae were revised in 1965 by the two main workers in this group, A.W

Excell and C.A. Stace (Excell and Stace 1965). They recognized two subfamilies, the

Strephonematoideae and the Combretoideae. Strephonematoideae contains the genus

Strephonema which is distinct in its semi-inferior ovary, two-armed hairs, paracytic stomata, and

revolute domatia. The Combretoideae in turn have a fully inferior ovary and convolute or plicate










cotyledons and the stomata are cyclocytic or anomocytic. The fruits of Stephonema are

drupaceous and do not have any specialized dispersal mechanism (Jongkind 1995), and are not

considered here.

The Combretoideae have been studied phylogenetically by Tan et al. (2002). The study

included 18 taxa and was based on ITS, rbcL, PY-IGS but included no morphological data.

Strephonema was used as the outgroup and the results were similar to what would be expected

from the traditional taxonomy. The subfamily split into the tribe containing Laguncularia and

Lumnizera (Laguncularieae) and the tribe Combretaeae which further divided into two clades;

Combretinae and Terminaliinae. Traditionally, there is a third subtribe the Pteleopsidinae, but

this was not included in Tan's analysis. All of these groups contain species with winged fruits.

The genera of Combretaceae that contain species with winged fruits are Calycopteris ,

Combretum, Pteleopsis, M~eiostemon, Terminalia, and Thiloa. Wings may be derived from ovary

tissue, bracteoles or perianth (Excel and Stace 1965).

There are three unique synapomorphies for the Combretaceae; anauxotelic inflorescences

(inflorescence not ending in flower and growth not extending beyond inflorescence),

"combretaceous" hairs (long, straight, sharply pointed, unicellular, very thick walled, with

conical internal component at base), and a "combretaceous" ovary (ovary inferior and unilocular;

ovules few, pendulous on elongate funiculi from the tip of locule) (Johnson and Briggs 1984,

Conti et al. 1997). Any of these features could be preserved in the right conditions, but none of

the fossil fruit species considered here are common and all are preserved as either compression

or impression fossils unattached to branches or leaves.

Fossil fruits have also been attributed to this family. Three North American fossil species

have been identified as belonging to this family. Fossil fruits assigned to Terminalia vera are









found in Eocene claypits in Tennessee and Kentucky and were first described by Edward Berry

in 1930. This fruit was described as similar to 7 brownii and 7 darlingii.~~ddd~~~dd~~~dd Ternzinalia estanttttt~~~~~~tttttttin

was described by MacGinitie (1941) based on leaves and associated winged fruits from the

Eocene Chalk Bluffs flora of the Central Sierra Nevada, California, and was compared to extant

I. hainensis and 7 triflora. Ternzinalia oregona was described from the Oligocene of Oregon by

Meyer and Manchester (1997). It was compared to extant 7 paniculata, and to MacGinitie' s

fossil fruits of Ternzinalia estanttttt~~~~~~tttttttina

Characters for identifying combretaceous fruits have not been clearly presented before, nor

have rigorous morphological comparisons been made between the winged fruits borne by

different genera of the family. Most of the fruits in this family are dispersed abiotically either by

water or wind, and fin-winged fruits are also present in several genera. Without detailed

morphological information on the fruits of extant taxa, it is difficult to have confidence in the

diagnosis of these fossiliferous fruits. MacGinitie (1941) reported that the seed cavity of

Conabretunt is typically longer relative to the wing than those of Ternzinalia, but he did not make

a full comparison or mention the status of the other genera.

Study of the morphology of these fossils and comparison with the modern genera of this

and other families with fin-winged fruits will provide more characters to evaluate whether these

fossils belong to the family, and if so, will give a strong record for the family in the United States

during the Eocene.








CHAPTER 2
MATERIALS AND METHODS

Modern Fruit Descriptions

A list was made of fruits known to have the fin-winged fruit type from internet and

literature searches that I made. The fruit morphology of genera in this list was verified when

possible by examination of specimens in herbaria. The taxonomy was updated to the APG

system (Angiosperm Phylogeny Group 2003).

Extant fruits were examined and photographed with a Nikon Coolpix 5400 digital camera

at the Gray (GH) and Arnold Arboretum (A) Herbaria at Harvard University, Cambridge, MA,

at the US National Museum at the Smithsonian Institution, Washington DC (US), and the United

States National Seed Herbarium, Beltsville, MD (USNSH). Photographs taken of specimens at

the Harvard Herbaria for which the source was unclear (A or GH) are provisionally marked

HARVARD or HARVARD SEED COLLECTION (if they came from the seed collection). The

lighting, both transmitted, and reflected, was adjusted to best show characters such as venation

and trichomes.

The photographs were then used to determine important characters for these genera.

Taxonomic literature (e.g. Goldberg 1986, Hutchinson 1964, and Judd 2006) was consulted to

help interpret the photographs. The fruits were characterized by number of wings, number of

carpels, dehiscence, ovary position, overall shape, wing texture, venation, vein spacing,

accessory parts (such as disk scar), and presence of a marginal vein. In some cases, the state was

not observable in the photographs nor available in the literature. Parentheses are included

around characters for which information is given from familial descriptions, when the details for

the particular genus were not available. Fruit type is given when clear, but indehiscent fruits are

often simply marked "indehiscent" to avoid confusing terms such as samara/samarium. Length








and width are recorded for individual specimens with slashes between different measured

specimens or, in the case of some of the genera in Combretaceae are given as ranges.

Fossil and Modern Combretaceae

Comparisons were made with extant genera of all angiosperm families known to possess

plants with the fin-winged fruit type. Specimens were examined and photographed at the

herbaria mentioned above. Fossil fruits were studied from collections of the Florida Museum of

Natural History, Gainesville (UF), the Smithsonian Institution [United States National Museum],

Washington, D.C., (USNM), the University of California, Museum of Paleontology, Berkeley

(UCMP), and the Field Museum of Natural History, Chicago (FMNH).

The Eocene fruits from western Tennessee and Kentucky were collected along with fossil

leaves, fruits, flowers and pollen from clay pits in the middle Eocene Claiborne Formation of

Tennessee and Kentucky. Large portions of the flora, including the species considered here,

were monographed by Berry (1916, 1930), but many of the determinations are still undergoing

revision (Dilcher and Jones 1980, Jones and Dilcher 1980, Herendeen et al., 1990, Herendeen

and Dilcher 1990). The age estimate is based on palynological correlations.

A total of 25 specimens were examined, including ten from Lamkin clay pit, one from

Warman, five from Grable, ten from Puryear, two from Wilbank, and one each from Powers

localities. A dissecting microscope was used to examine details of the fruits, and sediment was

chipped with needles to expose details of the wings. Each specimen was photographed with a

Nikon Coolpix 955 digital camera.

The fossils were initially exposed by splitting the clay with chisels, revealing the fruits in

longitudinal section. Because the plane of fracture usually exposes only two of the wings, more

detailed preparation was necessary to demonstrate the full number of wings. To reveal the

complete complement of wings, both counterpart halves of the same specimen were reassembled








and then fractured transversely. To control the orientation of the transverse fracture, we used a

dry circular diamond saw to pre-cut the surrounding clay in the desired plane.

Wing venation was studied under magnification and a camera lucida was used to record

details of venation. Line drawings were used to determine if each wing was a composite of two

fused laminae or if it was a single structure. Venation patterns of part and counterpart were

traced and overlaid to determine if the venation was the same on each side.

The Eocene specimens from California, from the Chalk Bluffs flora of the Sierra Nevada

Mountains were previously described by MacGinitie (1941) under the name Terminalia

estamina. Specimens were studied from the UCMP and USNM collections

The Oligocene specimens from Oregon were previously treated by Lakhanpal (1958),

Brown (1959) and Meyer and Manchester (1997). We studied the original specimens deposited

at UF, USNM, and UCMP.

The fossils were then compared to the modern genera of fruit with fin wings and genera

with similar features were compared to determine whether the fossils could be confidently

identified to a particular taxonomic group.










Table 2-1. Modern fin-winged fruits


Longer/
Wider/
ovary marginal Persistance (ruhy
position vein of parts ruhy
equal
(Q)


wing carpel dispersal
num. num. mode


Family


Genus


4-8 fruit Sup.


finned
2 5 Inf
menlcarp


3 3/5 anemoballisty Inf

3 3 ? Inf

2 2 fruit Sup.

3 3--5 mericarp Sup.

3 3-loc fruit Sup.

2-3-
2-2-3-loc fuit Sup.
(4)

2-5 4--5 fruit Inf.

4 2 fruit Inf.

2 fruit Inf.

2 2 fruit Inf.

4 2 fruit Inf.

2,4,5 4--5 fruit Inf.

3 3 ? Sup.

3-4
1--2 ocls fruit Sup.

3 3 anemoballisty Inf.

2 1 fruit Sup.

2 1 fruit Inf.

5 5 fruit Sup.

3 3--4 fruit I

4 5 fruit S

5 5 capsule S

5 5 capsule S

5 2--10 ? S

5 5 ? Sup.


Q

L

L

L/W/Q


?

L

L

L/

L/W/

L/WQ

LW

L

L

LWQ


L

Q

L/W

L

W

L

L


W

W

L

L


Achariaceae Carpotroche

Apiaceae Prangos

Apiaceae Steganotaenia

Begoniaceae Begonia

Burmanniaceae Burmannia

Cardiopteridaceae Cardiopteris

Celastraceae Stackhousia

Celastraceae Tripterigium

Celastraceae Wimmeria


disk

Perianth

Perianth

Stamenal
disk


? yl

Style

prath

Perli nth




(style)

Perianth

apical
protrusion
No

No

No

No


Combretaceae

Combretaceae

Combretaceae

Combretaceae

Combretaceae

Combretaceae

Cunoniaceae


Cyrillaceae

Dioscoreaceae

Fabaceae

Hernandiaceae

Lophopixidaceae

Lauranthaceae

Malvaceae

Malvaceae

Malvaceae

Malvaceae

Malvaceae


Combretum

Macropteranthus

Meiostemon

Pteleopsis

Thiloa

Terminalia

Gillbeea

Cliftonia

Dioscorea

Fissicalyx

Illigera

Lophopyxis

Nuyitsia

Cavanillesia

Abroma

Kleinhovia

Pentace

Pterostyrax


? Perianth

suture Perianth

No No

No No

No Sth e

No No

No apical
protrusion
suture Perianth

suture Perianth

No Perianth

No P










Table 2-1. Continued


Longer/
Wider/
ovary marginal Persistance(ruhy
position vein of parts eqa
(Q)

Sup. No ? W

Sup. sues/e S L

Sup. No S W

Sup. Yes No L

? ?? ?

Sup. No No L

Sup. No No L

Inf. suture No Q

Inf. suture No L/Q

Sup. Yes No Q

Sup. No Perianth Q

Sup. Yes No L

Sup. No apical L
protrusion

Sup. No Perianth W

Sup. No No L

Sup. No No L/W

Sup. No No L

Sup. No No Q

Sup. apical disk W
marginal

Sup. No L
marginal

Sup. No W
marginal

Sup. No No W

Sup. No No L/Q

Sup. ? ? ?

Sup. Yes basal disk L

Sup. Yes basal disk ?

Sup. Yes basal disk ?

Sun. Yes basal disk L


wing carpel dispersal
num. num. mode


Family


Genus


Malvaceae

Malvaceae

Malvaceae

Malvaceae

Malpighiaceae

Malpighiaceae

Nyctaginaceae

Onagraceae

Onagraceae

Pedaliaceae

Pedaliaceae

Phyllanthaceae

Polygalaceae

Polygonaceae

Polygonaceae

Polygonaceae

Polygonaceae

Polygonaceae

Rhamnaceae

Rhamnaceae

Rhamnaceae

Rutaceae

Rutaceae

Sapindaceae

Sapindaceae

Sapindaceae

Sapindaceae

Sapindaceae


Berrya

Burretiodendron

Colona

Craigia

Tetrapterys

Aspidopterys

Phaeoptilum

Megapterium

Oenothera

Holubia

Pterodiscus

Hymenocardia

Polygala

Atraphaxis

Rheum

Rumex

Polygonum

Antigonon

Gouania

Chaydia

Reissikia

Balfourodendron

Ptelea

Dodonaea

Urvillea

Majidea

Stocksia

Thouindium


5

5

3/4/5

5



3

4

4

4

4

4

2



3

3

3

3

3

3

3

3

4

2--3

3

2

2?



3


5 capsule

5 capsule

? capsule

3/4 capsule

? ?

3 fruit

1 fruit

4 anemoballisty

4 anemoballisty

2 fruit

(2-4) fruit

3 fruit

(2-3) fruit

(2-3) fruit

(2-3) fruit

(2-3) fruit

(2-3) fruit

(2-3) fruit

3 capsule

3 capsule

3 mericarp

4--5 fruit

2-3 loc No

?3 ?

(2-3) capsule

(2-3) capsule

(2-3) capsule

(2-3) capsule










Table 2-1. Continued


Longer/
Wider/
ovary marginal Persistance (ruhy
position vein of parts eqa
(Q)
Sup. Yes basal disk ?

Inf. Yes Perianth L

Sup. Yes No Q

Sun. Yes ? L


wing carpel dispersal
num. num. mode


Family


Genus


Sapindaceae

Styraceae

Trigoniaceae

Tropeaeoceae


Toulicia 3 (2-3) capsule
2 or
Halesia 42 fruit

Humbertiodendron 3 3 fruit

Tropaelum 3 3 (1
Tropaelum 3 locule) mericarp


Zygophyllaceae Bulnesia

Zygophyllaceae Guiacum

Zygophyllaceae Morkillia

Zygophyllaceae Porleria

Zygophyllaceae Zygophyllum


5 5 mericarp

? 5 Yes

4 (2-5) capsule

3 (2-5) mericarp

(2-5) mericarp


Yes ?

? ?

No?

? ?

? ?





















































Figure 2-1. Hernandiaceae-Onagraceae. A) Illigera grandiflora (Heng 11446, A). B)
Dioscorea (McPherson 43, FLAS). C) Dioscorea (Mcheson43 FLAS). D) Phaeoptilunt.
spirosunt (R.Sryel 19" USNSH). E) Runrex venosus (R USNSH). F) Oenothera
naissouriensis ( ISteen 51 GH). G) M~egapteriunt nissouriense (L.EWha, USNSH).
















D


H I-










Figure 2-2. Combretaceae (Combretum). A) Conabretunt apiculatunt (SI832 USNSH). B)
Conabretunt gazense (PG136, USNSH). C) Conabretunt fruticosunt (PI 104111,USNSH). D)
Conabretunt rhodesicunt (SPI 48244, Harvard). E) Conabretunt grandiflorunt (SPI 72993,
USNSH). F) Conabretuntprinsigenunt (SPI 61642, USNSH). G) Conabretuntpanicula~tun (P
sgnUSNSH). H) Conabretunt snzeathnzanii (PI 73930, USNSH). I) Conabretunt inaberbe (SPI
48243, USNSH).
































F





















Figure 2-3. Combretaceae (Terminalia). A) Terminalia argentea (PI 300640, USNSH). B)
Terminalia superba (Hris and ay 781 Harvard). C) Terminalia macroptera (221-, Harvard).
D) Terminalia tomentosa (Hara et al. 630075, A). E) Terminalia tomentosa (Butteric 2 A).
F) Terminalia virens (Maguire 30764, USNSH). G) Terminalia sericera (Parker 1428,
Harvard). H) Terminalia ama~zonia (Stinbch 617 A). I) Terminalia browneii (Harvard from
Flora Uganda). J) Terminalia pedicella~ta (Maxwell 01-461, A)
























A BC


Figure 2-4. Combretaceae (other). A) M~eiostemon tetrardra (Grosuenor snl966, USNSH). B)
Pteleopsis hylodendron (Claesaers 401, USNSH). C) Pteleopsis sp. (oselr150
USNSH). D) Guiera sp (Bamps ~247, USNSH). E) Ma'~cropteranthus fitzala~ni (FanisSN
1920, USNSH). F) Ma'~croptera~nthus kelovickii (NByrns 59, USNSH). G) Thiloa sp. (photo
of a photograph from "Types from Munich Herbarium").


.1






,b
i, G











AI


BC





Q)G
i ,II11111111


D2 F




I4







Figure 2-5. Malvales. A) Cavanellisia platanifolia (Fster224, GH). B) Abromaa~ngusta
(Wag 4164 A). C) Kleinhovia hospital (Huang et al 16132, A). D) M~anvellia lepidota
(Hriman 3129 A) ). E) Pterostyrax corymbosa (Mie Min-Xiann 92250, A). F) Halesia
diptera (Chester 1124, A). G) Berrya javanica (KK and SS 11, A). H) Colona sp. (Suejarto 59,
A). I) Burretiodendron esquirolli (Malelen et al. 1160, A). J) Cragia yunnanensis (1124 UF).
K) Pentace laxiflora (Sn 78602, A).














C


)E


jG


J


Figure 2-6. Zygophyllaceae-Begoniaceae. A) Bulnesia chilensis (Wedemam 29 GH). B)
Bulnesia arborea (1069, UF). C) Zygophyllum morgan2sa (Marloth 12275, A). D) Porleria
angustifolia (Moore and Stevermark 3288, Harvard). E) Wimmeria mexicana (Ceinger Jul
1999, GH). F) Wimmeria concolor (Goldmn 227 GH). G) Tripterigium regelii (USNSH
Togashi 531). H) Gillbeea papuana (Brass 31815, A). I) Gillbeeapapuana (Takuchi608, A).
J) Begonia sp. (Ksterman 6022 A). K) Begonia nelubifolia (Poropus 8543, GH).


ID


1~B
1


1111151\11

















A B



































Figure 2-7. Rhamnaceae-Tropaeolumaceae. A) Gouania dontingensis (Rdrgue266 A).
B) Chaydaia (Huder 601, USNSH). C) Reisseika (L.O. Williams 6834, USNSH). D) Gouania
hipuloides (Brumbach 9537, GH). E) Carpotroche anzazonica (No.9001, USNSH). F)
Hunabertodendron (Capman 180535F, USNSH). G) Lophopyris (Buderus NGF 23932,
USNSH). Tropaeobia (Donat51, UNSH).












1\


11111111"


B


Figure 2-8. Rutaceae- Apiaceae. A) Balfourodendron riedelan2um (Fer, 14 GH). B)
Ptelea trifolata (Harvard Seed Collection). C) Ptelea crenulata (Harvard Seed Collection). D)
Urvilleaa~ndersonii (Anderson 9150, NYBG). E) Chiftonia monophylla (Rhade Sa03 GH).
F) Holubia saccata (AOD Mogg 24419, USNSH) G) Pterodiscus sp. (Dinker 4935, A). H)
Cardiopteris moluccanum (Wang 8 1070, A). I) Steganotaenia araliocea (Meye 7962
USNSH). J) Prangos odotoptera (Rekigef 97 USNSH).


~xi;d














.;I ~t


~Il




















































Oxalidales





Cuourbltales




Huerleale






Slpindale.



Cornales

Ericales

Ouarryales

Gentiarsales

Larniales~alr

colanales

Aqulfollale~

Dipsicales



Apideslf


Figure 2-9. Distribution of fin winged fruits on Angiosperm phylogeny. Each star represents a

family that contains at least one species with fin winged fruits. Topology taken from (Stevens,

2001 onwards).


E ~soereales

Poales

Alberatalec

Liliales





Figure 2-10. Terminalia vera. A) (FLMNH 15815-48329). B) (FLMNH 15815-48329).
C) (FLMNH 15815-48326). D) (UF 15875-48329). E) (UF 15875-48329). F) (UF 18884-
32723) G. (UF 18884-32723).






















































Figure 2-11. Call~~irdithe pranzgosoides. A) (USNM 36005). B) (FLMNH 15815-48328). C.)
(FLMNH 15820-48318). D) (FLMNH 18884-32781). E) (FLMNH 18884-32781). F)
(UFl5820-48332) G) (FLMNH 18952-46057). H) (FLMNH15826-48331). I) (FLMNH). J)
(FLMNH). K) (FLMNH 15815-48325).



















































B



Figure 2-12. Terminalia vera and Macropteranthus fitzalani. A) (USNM). B) (USNSH Francis
SN 1920).











.i
--r



;Y


B) (USNM 313946). C) (UCMP 2321).


Figure 2-13. Terminalia est amina.tt~~ttt~~tt A) (UCMP 2321).
D) (UCMP 2321).


111 z l"'~l"''I""r









`f


r


II i l0 I H I III : l ii


Figure 2-14.
15842-7032).
250-10497).


Terminalia estamtttttttt~~~~~~~~~ina A) Willemette flora (USNM42334). B) Fossil Oregon (UF
C) Lyons flora (USNM 42353). D) (UF 250-10497). E) (UF 250-10497). F) (UF


B
111"''!1'"''~" "!~!I"''"I['!?I'"'


a IIII,, IP,,l,,l II I 1ll Ill~ lllll ll









CHAPTER 3
MODERN FRUIT SURVEY

As a result of this survey, I became aware of at least 87 genera in which radially finned

fruits are present. Each genus is described below (arranged alphabetically by family). These

data serve as the basis for the comparative analyses presented in Chapter 5.

Achariaceae
Carpotroche (Figure 2-7E)
Species examined: Carpotroche ama~zonica Mart. (N.9001, USNSH)
Type of fruit: Indehiscent
Length: 7.8 cm
Width: 7.1 cm
Shape in outline: Globose
Ovary condition: Superior
Persistence of styles/perianth: Disk on basal surface
Persistence of pedicel/stipe: No
Carpels: 4-8
Number of wings: 6 to many (~20); often dissected
Margins: Entire/undulate
Wing derivation: Ovary
Wing thickness: Coriaceous
Trichomes: None observed
Venation: Wing was too opaque to observe venation
Distribution: Tropical America

Apiaceae

Anatomy and morphology of fin-winged fruit genera of Apiaceae and Araliacaeae

(Annesorhiza, Asteriscium, Astrotricha, Choritaenia, Da~sispermum, Elaeoselinum, Heptaptera,

Hermas, Heteromorpha, Laretia, Molopospermum, M~yodocarpus, Pachypleurum, Peucedanum,~~dddd~~~ddd~~~

Polemanniopsis, Polylophium, Rouya, and Tordylium) has been recently studied in detail (Liu et

al. 2006). They found that fruit characters including developmental origin of the wings, the

shape of the carpel, presence of vittae, and other anatomical features to correspond with

molecular cladograms, which suggests that in this group at least, fruits have great taxonomic

value. Described below are two genera which were not included. Pranzgos is obviously winged








but the species of Steganotaenia I observed appeared only slightly finned although they can be

winged (Hyde and Wursten 2002-7).

Pranzgos (Figure 2-8J)
Species: Pranzgos odontotlera Boiss. (K.H Beckinguif 997, USNSH)
Type of fruit: Schizocarp
Length: 2 cm. (+0.5 pedicel)
Width: 1 cm
Shape in outline: Elliptical
Ovary condition: Inferior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: Pedicel
Carpels: 5
Number of wings: 5
Margins: Entire
Wing derivation: From ribs on ovary
Wing thickness: Papery
Trichomes: None observed
Venation: None apparent

Steganotaenia (Figure 2-8I)
Species examined: Steganotaenia araliacea Hochst. (F.G. Meters 7962, USNSH)
Type of fruit: Schizocarp
Length: 1.6 cm. (+.4 pedicel)
Width: 1 cm
Shape in outline: Elliptical
Ovary condition: Inferior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: Pedicel
Carpels: 5
Number of wings: This species is more ridged than winged, but other species prominently
2 winged
Margins: Entire
Wing derivation: Ovary
Wing thickness: Coriaceaous
Trichomes: No
Venation: N/A

Begoniaceae

Begoniaceae have 2 genera with the prominent winged ovaries Begonia and

Hillebrandia. Begonias have many stamens, and there is a disk at the apex of the fruit where

these stamens, along with petals and stigma were attached. Begonias may be wind dispersed,

animal dispersed, or rain dispersed. Different fruit morphologies are associated with each of








these different dispersal syndromes. Wind dispersed begonias tend to have three locules with

three equal or close to equal sized wings while rain dispersed begonias have one large wing and

two smaller wings. Ecologically, wind dispersed begonias are more often found in open forest

habitat while rain or animal dispersed begonias are found in humid closed forest habitats (Tebbitt

et al. 2006).

Begonia (Figure 2-6JK)
Species examined: Begonia sp. (Hove e a. 88 A), Begonia nehembiifolia Cham. &
Schltdl. (Popus 8543), G Begonia sp. (Kosterman 6022, A)
Type of fruit: Dehi scent/Indehi scent. capsule splitting along sutures and releasing seeds
via anemoballisty
Length: Variable
Width: Variable longer than wide or wider than long
Shape in outline: Variable
Ovary condition: Inferior
Persistence of styles/perianth: Yes- staminal disk
Persistence of pedicel/stipe: Yes stipe
Carpels: 3-5
Symmetry: Often with one locule and wing larger than the others
Number of wings: 3
Margins: Entire
Wing derivation: Ovary outgrowth
Wing thickness: Membranous
Trichomes: No
Venation : Transverse; no marginal vein; some looping; areoles transverse elongate

Burmaniaceae are closely related to Dioscorea and some have similar fruits. Three

winged with minute seeds. The three carpels are sometimes basally affixed.

Cardiopteridacerae

Cardiopteris (Figure 2-8H)
Species examined: Cardiopteris mohiccana Blume (Wann 81070, A)
Type of fruit: Indehiscent
Length: 3.5 cm
Width: 2.8 cm
Shape in outline: Transversely Elliptical
Ovary condition: Superior
Persistence of styles/perianth: One large stylar protrusion and calyx present at base
Persistence of pedicel/stipe: maybe very small stipe
Carpels/Locules: 2 carpels, 1 locule
Number of wings: 2
Margins: Entire, smooth








Wing derivation: Ovary
Wing thickness: Chartaceous
Trichomes: No
Venation: Marginal vein only

Celastraceae

Stackhousia. These herbs are found in Australia/Malesia (Mabberly 1989). The plants

accumulate nickel in various organs including the fruits (Bhatia et al. 2003). Their fruits are

Schizocarpic and readily identifiable by their wings with radiating veins completely free from

interconnections and dichotomizations and with a marginal vein. Interestingly, the germination

of the seeds is increased in the presence of smoke (Vigilante et al. 1998).

Species examined: Stackhousia brunonis Benth. (A. Travers 19, A)
Type of fruit: Schizocarp
Length: 2.4 cm
Width: 1.7 cm
Shape in outline: Ovate
Apex: Acuminate
Base: Cordate
Ovary condition: Superior
Persistence of styles/perianth: Calyx at base and apical protrusions
Persistence of pedicel/stipe: Very small stipe
Carpels: 3-5
Number of wings: 3
Margins: Entire with marginal vein
Wing derivation: Pericarp
Wing thickness: Chartaceous
Trichomes: Yes, unbranched
Venation: Radiating out and arcing slightly upward; thicker at central body and thinning
towards margin; joining marginal vein.

Tripterygium (Figure 2-6G). Triptelygium is a liana or scandent shrub which is used in

herbal medicine. The ovules are epitropous.

Species examined: Tripterigium regelii Sprague & Takeda (Tonashi 531, A);
Tripterigium wilfordii Hook.f. (Cvadrc 36 A)
Type of fruit: Indehiscent
Length: 1.2 cm
Width: 1.1 cm
Shape in outline: Longitudinally Elliptical
Apex: Rounded/cordate
Base: Lobate








Ovary condition: Superior
Persistence of styles/perianth: Apical protrusion and floral remnants at base
Persistence of pedicel/stipe: Pedicel
Carpels: 3 locules
Number of wings: 3
Margins: Entire (marginal vein)
Wing derivation: Ovary?
Wing thickness: Chartaceous
Trichomes: No
Venation: Arching upwards subparrallel and only occasionally dichotomizing; central
body with longitudinal veins and some transverse crossveins (marginal vein)
Distribution: East China to Taiwan

Wimmeria (Figure 2-6 E F)
Species examined: Wimmeria concolor Cham. & Schltdl. (Goldman 227, GH),
Wimmeria confuse Hemsl. (Lundell 1938, GH), Wimmeria mexicana (DC.) Lundell

Type of fruit: Indehiscent samara
Length: 1.4/ 2 cm
Length: 1.8/ 1.6 cm
Shape in outline: Elliptical
Ovary condition: Superior
Persistence of styles/perianth: Styles at apex, perianth at base
Persistence of pedicel/stipe: Pedicel
Carpels: 2-3 locules
Number of wings: Usually 3, but may be 2 or 4 (Lundel, 193 8)
Margins: Entire
Wing derivation: Ovary
Wing thickness: Chartaceous
Trichomes: No
Venation: Radiating subparallel dichotomizing and anastomosing
Fruit may have silvery cottonlike filaments when breaking open

Combretaceae


Combretaceae are a family with fruits that are water dispersed, animal dispersed, and

wind dispersed. Wind dispersal occurs in many clades within the family. This family is

discussed in greater detail in Chapter 3.

Combretum (Figure 2-2)
Species examined: Combretum apiculatum Sond. (SP 2842 USNSH); C.
erythrophyllum Sond. (PIl l2639, USNSH); C. farinosum H.B. & K. (FPI 81266,
USNSH); C. fr~uticosum Stuntz (PI 104111, USNSH); C. gazense Swynn. & Baker f
(PIG13469, USNSH); C. grandiflorum G. Don (SPI 72993, USNSH); C. imberbe








Wawra(SPI 48243, USNSH); C. paniculatunt Vent. (PI sn, USNSH): C. prinsigenunt
Marloth. (SPI 48244, USNSH); C. snzeathnzannii G.Don (PI 73930, USNSH)
Type of fruit: Indehiscent. diclesium (Spjut, 1994)
Length: 1-5 cm
Width: 1-4 cm
Shape in outline: Variable from widely elliptical, ovate, to narrowly elliptical
Ovary condition: Inferior
Persistence of styles/perianth: Sometimes stylar protusion at apex
Persistence of pedicel/stipe: Usually a pedicel
Carpels: 4-5 (one locule)
Number of wings: 4-5
Margins: Usually tattered
Wing derivation: Perianth
Wing thickness: Chartaceous
Trichomes: Not observed
Venation: Generally radiating subparallel

Macropteranthes (Figure 2-4 E F)
S ecies examined: M~ Jtzalan2i F. Muell (SN 1960, USNSH); M~kelorickii Benth. (E
Byrnes 1592, USNSH)
Type of fruit: Indehiscent
Length: 1 cm/ 2 cm+
Width: 1.5 cm/ 1.5 cm
Shape in outline: Wide elliptical. to elongate obovate
Ovary condition: Inferior
Persistence of styles/perianth: Calyx lobes at apex
Persistence of pedicel/stipe: Pedicel on fruits ofM. fitzalan2i
Carpels: 2 but pseudomonomerous
Number of wings: 4 each (composed of 2 lamina)
Margins: Entire
Wing derivation: Bracteoles
Wing thickness: Chartaceous
Trichomes: Not observed but presumably there may be combretaceous hairs.
Venation: Primaries looping or not with secondaries interconnecting and looping at
margins (or not), and tertiaries when present interconnecting or forming blind ended
branched veinlets

Pteleopsis (Figure 2-4 B C)
Species examined: P. hylodendron Mildber. (J. lssens 401J USNSH) P. sp
(Gsswele 350 USNSH).
Type of fruit: Indehiscent
Length: 1.5-2.5 cm
Width: 0.5-1.5 cm
Shape in outline: Elliptical., sometimes with v-shaped notch at apex or stylar protrusion,
base truncate
Ovary condition: Inferior
Persistence of styles/perianth: Sometimes stylar protrusion
Persistence of pedicel/stipe: Pedicel








Carpels: 2, but appearing as one
Number of wings: 2
Margins: Entire
Wing derivation: Perianth
Wing thickness: Chartaceous
Trichomes: Not observed but presumably there may be combretaceous hairs
Venation: Radiating subparallel

Thiloa (Figure 2-4 G)
Species examined: Thiloa glaucocarpa (Mart.) Eich. Image was observed at Harvard
from "Types of the Munich Herbarium"
Type of fruit: Indehiscent
Length: ?
Width: ?
Shape in outline: Rounded
Ovary condition: Inferior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: Pedicel
Carpels: 2, but appearing as one
Number of wings: 4
Margins: Entire
Wing derivation: Perianth
Wing thickness: Chartaceous
Trichomes: Not observed but presumably there may be combretaceous hairs
Venation: Not observed

Terminalia (Figure 2-3)
Species examined: T. argentea Mart. (P 3060 USNSH); T. virens (Spruce ex
Eicheler) Alwen & Stace (B. Manuire et al. 30764, USNSH); T. alata Herb.Madr. ex
Wall. (Maxwell 92-19, A); T. ama~zonica Excell (Steinbach 6617, A); 7: argentea
Mart. (Mello et al. 608, GH); T. glaucescens Planch. Ex Benth. (Bretleler 248, A); 7:
macroptera Mart. (221-9 Harvard); T. pedicellata W.Nanakorn (Mxwel 141
A); T. pyrifolia Kurz. (Burma Forest School Herbarium 61, A); T. tomentosa Mart.
Ex Eichl. (Hra1004 A).
Type of fruit: Indehiscent diclesium (Spjut, 1994)
Length: 0.5- 7 cm
Width: 1- 6cm
Shape in outline: Variable from elongate to elliptical to very wide elliptical
Ovary condition: Inferior
Persistence of styles/perianth: Stylar protrusion sometimes; diclesium.
Persistence of pedicel/stipe: Sometimes stipe or pedicel
Carpels: 2 but appearing as one
Number of wings: 2, 4, 5
Margins: Entire, may be tattered or reinforced by increased dichotomizations near margin
Wing derivation: Perianth
Wing thickness: Chartaceous to coriaceous
Trichomes: Combretaceous hairs present or absent








Venation: Usually radiating subparrallel but may dichotomize and anastomose either
occatinally or as a reinforcement to the margin

Cunonaceae


Gillbeea (Figure 2-6 H I)
Gillbeea is found in New Guinea and Queensland and has a three winged one seeded
fruit. Anatomical information for the flowers of this genus can be found in Matthews
et al. (2001). The broken pieces had two distinctive veins from the central body to
the apex.
Species examined: Gillbeea papuana Schltr. (Takuchi608, A) (Brass 31815, A)
Type of fruit: Described as Indehiscent., but I found pieces broken apart
Length: 1.6 cm
Width: 1 cm
Shape in outline: Longitudinally Elliptical
Ovary condition: Superior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: Pedicel
Carpels: 3
Number of wings: 3
Margins: Wavy/entire with marginal vein
Wing derivation: Ovary wall outgrowths
Wing texture: Chartaceous
Trichomes: Yes
Venation: Radiating reticulum with midveins running apically from central axis and a
marginal vein present

Cyrillaceae


Cliftonia (Figure 2-8E)
Species examined: Cliftonia monophylla Britton (Rhoads S203 GH).
Type of fruit: Indehiscent
Length: 0.5 cm
Width: 0.5 cm
Shape in outline: Longitudinally Elliptical
Apex: Emarginate
Base: Rounded
Ovary condition: Superior
Persistence of styles/perianth: Persistent sepals at base of fruit
Persistence of pedicel/stipe: Pedicel?
Carpels: 3-4
Number of wings: 2-3
Margins: Entire
Wing derivation: Ovary
Wing thickness: Coriaceous
Trichomes: No








Venation: Unknown

Dioscorenceae


Dioscorea (Figure 2-1 B C)
Species examined: Dioscorea alata L. (McPherson 43, FLAS)
Type of fruit: Apically opening capsule (dispersing via anemoballisty)
Length: Variable 4.3 cm
Width: Variable 4.3 cm
Shape in outline: Variable from longer than wide to round to wider than long
Ovary condition: Inferior
Persistence of styles/perianth: Sometimes perianth persists at apex
Persistence of pedicel/stipe: Small pedicel
Carpels: 3
Number of wings: 3
Margins: Thick marginal suture
Wing derivation: Ovary
Wing thickness: Chartaceous
Trichomes: No
Venation: Striations radiating outward to true venation
Seeds: Often with winged seeds

Fabac eae


There are several genera in the Fabaceae that have winged fruits. Of these, at least three

are known to have the Ein-winged morphology. Much has been published about the fruits of this

family (Kirkbride et al. 2003; Stirton and Zarucchi 1989, Gunn 1984) including the morphology

and aerodynamics of wind dispersed legumes. At least four genera have Ein-winged fruit

(Fissicalyx, Piscidia, Sesbania, and Tetracarpon). Piscidia is found from Florida to Venezuala

and was figured in Flora of Panama (Dwyer 1965). The fruits are pedicellate and stipitate, four

winged, both apex and base are truncate, with pinnate venation. Fissicalyx is described below.

Fissicalyx. Fissicalyx is a member of the Dalbergioid legumes, a group which has at

least 16 genera with wings (Lavin et al. 2001). The wings of Fissicalyx are derived from

expansion of the ovary wall, but in other winged genera the wings are formed from expansion of

ovary sutures or from the stipe. This genus is figured in (Dwyer 1965).








Species examined: Image of Fissicalyx taken from USNSH website
Type of fruit: Indehiscent pod
Length: 6 cm
Width: 5.3 cm
Shape in outline: Elliptical with an acuminate apex
Ovary condition: Superior
Persistence of styles/perianth: Style at apex
Persistence of pedicel/stipe: No
Carpels: 1
Number of wings: 2
Margins: Entire
Wing derivation: Expansion of ovary wall
Wing thickness: Chartaceous?
Trichomes: No
Venation: Radiating

Hernandiaceae


Illigera (Figure 2-1 A)
Species examined:1Illigera celibica Miq. (How 73378, A), Illigera cordata Dunn
(Delaway 4 Sept. 1883, A), Illigera grandiflora W.W.Sm. & Jeffrey (Henn 11446, A)
(He1 1346 A)
Fruit type: Indehiscent., coded as a drupe (Renner, 1999)
Length: 3.6/3.6 cm
Width: 6.9/3.7 cm
Shape in outline: Laterally elliptical or rounded
Apex: Cordate
Base: Cordate/ cuneate
Ovary condition: Inferior
Persistence of styles/perianth: Perhaps sometimes style
Persistence of pedicel/stipe: No
Carpels: One
Symmetry: 2 wings larger than third
Number of wings: 2 or 3
Margins: Entire slightly wavy
Wing derivation: Ovary
Wing thickness: Chartaceous
Trichomes: No
Venation: Veins radiating to margin very sinuous all of a single order

Lophopyxidaceae


Lophopyxis (Figure 2-7 G). Lophopyxis is the only genus in the Lophopyxidaceae, it is
sometimes placed in Celastraceae but has characters such as tomentose ovaries which
are different from the Celastraceae (Simmons et al. 1999)








Species examined: Lophopyxis combretocarpa Engl. Ex Pax (Abraia345 USNSH)
Lophopyxis maingayi Hook.f. Scode503, USNSH)
Type of fruit: Indehiscent samara
Length: 3 cm
Width: 1.5 cm
Shape in outline: Elongate obovate
Ovary condition: Superior.
Persistence of styles/perianth: Stylar remnants at apex and perianth remnants at base
Persistence of pedicel/stipe: Small stipe
Carpels: 5
Number of wings: 5
Margins: Entire
Wing derivation: Ovary
Wing thickness: Chartaceous
Trichomes: Densely hairy, hairs unbranched
Venation: Downward arching subparrallel

Loranthaceae


Nuyitsia. Loranthaceae are a family known for being parasitic on tree branches. Fruits

of this genus are usually fleshy. Nuytsia is a tree or shrub and has three-winged wind dispersed

fruits. Lamont (1985) studied the dispersal of these fruits and found a maximum dispersal

capability of 50 meters. Nevertheless, he concluded that this supports evidence that this genus

reproduces mainly by suckering.

Species examined: Nuytsia sp. (USNSH)
Type of fruit: Indehiscent
Length: 1 cm
Width: 0.8cm
Shape in outline: Elliptical
Ovary condition: Inferior.
Persistence of styles/perianth: No
Persistence of pedicel/stipe: Pedicel
Carpels: 3-4
Number of wings: 3
Margins: Entire
Wing derivation: Ovary
Wing thickness: Chartaceous
Trichomes: No
Venation: Transverse veins occasionally dichotomizing, arising from midline of fruit








Malvacene


Cavanillesia (Figure 2-5A). The fruits of Cavanillesia are the largest fin-winged fruits at

over 12 cm. long. Garwood (1985) studied the role of mucilage in seedling development of

Cavanillesia. This genus produces large amounts of mucilage in the fruits, and the mucilage is

present during dispersal.

Species examined: Cavanillesia platanifolia H.B. & K (R. Foster 2247, GH)
Type of fruit: Indehiscent capsule
Length: 12 cm
Width: 14 cm
Shape in outline: Widely elliptical
Apex: Cordate
Base: Cordate
Ovary condition: Superior
Persistence of styles/perianth: Apical protuberance
Persistence of pedicel/stipe: No
Carpels: 5 (one or two seeded fruits)
Number of wings: 4
Margins: Tattered
Wing derivation: Ovary
Wing thickness: Membranous
Trichomes: No
Venation: At first glance, veins appear to be subparrallel and of a single order, but closer
examination reveals a reticulum of lower order veins between the primaries which
form areoles of 1mm or less, the veins are very fibrous and remain as a skeleton after
the wing surface is lost.

Abroma (Figure 2-5 B)
Species examined: Abroma augusta L.f. (CWag 4164 A) (A.D.E Elmer 21446, A)
Type of fruit: Loculicidal capsule
Length: 4-5cm.
Width 5-6 cm
Shape in outline: Obovate
Apex: Truncate
Base: Rounded
Ovary condition: Superior
Persistence of styles/perianth: Calyx
Persistence of pedicel/stipe: No
Locules: 5
Number of wings: 5
Margins: Thick suture
Wing derivation: Ovary
Wing thickness: Chartaceous








Trichomes: Tri-branched hairs.
Venation: Veins thick, ascending to tips, dichotomising and anastomosing occasionally
forming transversely long areoles.

Kleinzhovia (Figure 2-5C). Kleinhovia is a vine or a tree and grows in Asia and Australia.

Hj erpe and Hedenas (2001) found that Kleinhovia hospital is a species that thrives in the

disturbed habitat after a cyclone.

Species examined: Kleinhovia hospital L. (Huanget al.16132 A)
Type of fruit: Inflated capsule
Length: 1.4 cm
Width: 2.3 cm
Shape in outline: Obcordate
Ovary condition: Superior
Persistence of styles/perianth: Remnants at base
Persistence of pedicel/stipe: Pedicel and stipe
Carpels: 5
Number of wings: 5
Margins: Sutured
Wing derivation: Perianth
Wing thickness: Chartaceous
Trichomes: No
Venation: Veins coursing apically forming large transversely elongate areoles. Veinlets
ending within areoules with a small hook.

Maxwellia (Figure 2-5D). M~amellia is a New Caledonia endemic with the unusual

condition of having many small seeds in an indehiscent fruit (Wilkins and Chappell 2002,

Robyns et al. 1977).

Species examined: Mamwellia lepidota Baill. (Hiirliman 13129, A)
Type of fruit: Indehiscent
Length: 3 cm
Width: 2.3 cm.
Shape in outline: Elliptical
Apex: Obcordate
Base: Cordate
Ovary condition: Superior
Persistence of styles/perianth: Perianth at base
Persistence of pedicel/stipe: Pedicel on specimen examined
Carpels: 3-4
Number of wings: 3-4(-5)
Margins: Entire and wavy
Wing derivation: Ovary
Wing thickness: Coriaceous








Trichomes: No
Venation: Wing too thick to observe

Craigia (Figure 2-5J). Modern Craigia is found in Eastern China and Taiwan; but there

are also fruit fossils from the Tertiary of North America, Europe and Asia (Kva~ek et al. 1991).

Species examined: Craigia yunnanensis (UF 1124)
Type of fruit: Capsule
Length: 3.1 cm
Length: 2.3 cm
Shape in outline: Elliptical
Ovary condition: Superior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: No
Carpels: 3-4
Number of wings: 5
Margins: Entire; marginal vein
Wing derivation: Ovary?
Wing thickness: Papery/membranous
Trichomes: No
Venation: Primaries radiating dichotomizing and anastomosing to form elongate areoles,
secondaries interconnecting and also forming freely ending veinlets which are
sometimes branched.

Pentace (Figure 2-8 K)
Species examined: Pentace laxiflora Merr. (Leopold 78602, A)
Type of fruit: Indehiscent samara
Length: 1.7 cm
Width: 1 cm
Shape in outline: Elliptical
Ovary condition: Superior
Persistence of styles/perianth: Perianth at base
Persistence of pedicel/stipe: Pedicel
Carpels: 2-10
Number of wings: 5
Margins: Slightly tattered
Wing derivation: Ovary?
Wing thickness: Papery
Trichomes: No
Venation: Sinuous, radiating, some dichotomizing and anastomosing near the margin

Malpighiaceae


Aspidopterys. Aspidopterys is a liana which grows in tropical Asia.
Species examined: Aspidopterys roxburghiana A.Juss. (Griffith 924, A) (Naravanaswami
18, A)








Type of fruit: Samara
Length: 3 cm/ 4.3 cm
Width: 2.3 cm/ 1.5 cm
Shape in outline: Ovate/ oblong
Ovary condition: Superior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: No
Carpels: 3
Number of wings: 3
Margins: Entire with veins looping along margin
Wing derivation: Ovary
Wing thickness: Chartaceous
Trichomes: No
Venation: Primaries forming loops with the secondaries interconnecting and looping at
the margin, tertiaries interconnect

Nyetaginaceae

This family has many species and genera with fin-winged fruits formed from accrescent

tepals (Spjut 1994). Seed have a curved embryo. Only one genus is detailed here, a survey

within the family needs to undertaken to document the diversity within the family.

Abronia. Abronia has been thoroughly studied for anatomy (Wilson 1974, Wilson,

1975), dispersability (Wilson 1976) and compared to the similar genus Tripterocalyx (Galloway

1975).

Phaeoptilum (Figure 2-1C)
Species examined: P. spinosum Radlk. (R. rydl "E 99" USNSH)
Type of fruit: Indehiscent anthocarp
Length: 2.1 cm
Width: 1.8 cm
Shape in outline: Elliptical with lobate base
Ovary condition: Superior
Persistence of styles/perianth: Yes wings are derived from accrescent tepals
Persistence of pedicel/stipe: No
Carpels: 1
Number of wings: 4
Margins: Entire
Wing derivation: Accrescent tepals
Wing thickness: Chartaceous
Trichomes: No
Venation: Radiating slightly sinuous








Onagraceae


Megapterium (Figure 2-1G)
Species examined: Megapterium naissouriense Spach. (L.E Wham s.n., USNSH)
Type of fruit: Apically opening capsule (dispersing via anemoballisty)
Length: 6.8 cm
Width: 6.8 cm
Shape in outline: Transversly elliptical/ round
Apex: Rounded
Base: Rounded
Ovary condition: Inferior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: No
Number of wings: 4
Margins: Entire and sutured?
Wing derivation: Ovary.
Wing thickness: Chartaceous.
Trichomes: No.
Venation: Striations radiating.

Oenothera (Figure 2-1F)
Species examined: Oenothera naissouriensis Sims (Stevens 561, Harvard Seed
Collection), Oenothera nzacrocarpa (Hitchcock 165, Harvard Seed Collection)
(Wagner s1n. Seed Herbarium)
Type of fruit: Loculicidal capsule dispersing via anemoballisty
Length: 3.5/ 4 cm
Width: 1.8/ 4 cm
Shape in outline: Slightly ovate/ slightly obovate/ longitudinally elliptical/ round
Ovary condition: Inferior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: No/ sometimes small pedicel?
Locules: 4
Number of wings: 4
Margins: Entire and sutured
Wing derivation: Ovary?
Wing thickness: Chartaceous
Trichomes: Sometimes small hairs
Venation: Radiating striations

Pedaliaceae


Holubia (Figure 2-8F). Holubia is an herb native to Madagascar
Species examined: Hohibia saccata Oliver ( IAnonmos 01 USNSH)
Type of fruit: Indehiscent
Length: 5.4 cm
Width: 5.3 cm








Shape in outline: Round
Apex: Rounded slightly cordate
Base: Rounded slightly cordate.
Ovary condition: Superior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: Pedicel
Carpels: 2 locules
Number of wings: 4
Margins: Entire sinuous with marginal vein
Wing derivation:
Wing thickness: Chartaceous
Trichomes: No
Venation: One order, radiating outward; dichotomizing and anastomosing to form
transverse longitudinal to polygonal areoles; marginal vein present

Pterodiscus (Figure 2-8G)
Species examined: Pterodiscus aurantiacus Welw. (Werdrmann2290 GH) (Dse
4635, GH)
Type of fruit: Indehiscent
Length: 3.1 cm/ 3.3 cm
Width: 3.1cm/ 3.4 cm
Shape in outline: Round
Apex: Rounded
Base: Lobate
Ovary condition: Superior
Persistence of styles/perianth: Perianth at base, small apical protrusion
Persistence of pedicel/stipe: Pedicle
Carpels: 2-4 reported for the family
Number of wings: 4
Margins: Entire with marginal vein
Wing derivation:
Wing thickness: Chartaceous
Trichomes: No
Venation: Radiating outward with some dichotomizing and anastomosing particularly
towards the margin. Marginal vein is present

Phyllanthaceae


Hymenocardia. Hymenocardia is the only wind dispersed member or the

Phyllaanthaceae and has previously been put into its own family (Hymenocardiaceae) (Wurdak

et al. 2004).

Species examined: Hymenocardia ulmoides Oliver (Robyns 799, USNSH)
Type of fruit: Schizocarp
Length: 2.3 cm








Width: 1.9 cm
Shape in outline: Round-Elliptical
Apex: Obsaggitate
Base: Rounded
Ovary condition: Superior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: Both
Carpels: Bilocular
Number of wings: 2
Margins: Entire with marginal vein
Wing derivation: Ovary
Wing thickness: Chartaceous
Trichomes: No
Venation: Central vein running apically through the basal portion of the wing and over
the apically placed seed cavity. Secondary veins radiating transversely and
dichotomizing and anastomosing irregularly, marginal vein present.

Polygalaceae


Poly gala
Species examined: Polygala sp. (Britton1029 USNSH)
Type of fruit: Dehiscent or indehiscent
Length: 4 cm
Width: 3 cm
Shape in outline: Obovate-round or asymmetrical elliptical
Ovary condition: Superior
Persistence of styles/perianth: Apical protrusion may be present and small disk at base
Persistence of pedicel/stipe: Pedicel
Carpels: 2
Number of wings: 2-3
Margins: Entire
Wing derivation: Ovary
Wing thickness: Papery
Trichomes: Yes, unbranched
Venation: Radiating, dichotomizing and anastomosing infrequently

Polygonaceae

Polygonaceae are a family with many genera with fin-winged fruit. The fruit type in this

family is an achene, and the wings, in these fruits are formed from perianth parts. The fruits may

be biwinged or triwinged and the shape is quite variable. Laubengayer (1937) detailed the

anatomy of the flower. Genera with fin-winged fruits include Antigonon, Aturaphaxis








Calligonium, Polygonum, Rheum, Rumex, and Triplaris. Three of these genera are described

below.

Antigonon
Species: Antigonon guatemalense Meisn. (PI 146197, USNSH).
Type of fruit: Achene with perianth forming wings
Length: 3 cm
Width: 3 cm
Shape in outline: Round with lobate base
Ovary condition: Superior
Persistence of styles/perianth: Perianth forming wings
Persistence of pedicel/stipe: Pedicel
Carpels: 2-3
Number of wings: 3
Margins: Entire
Wing derivation: Perianth
Wing thickness: Papery
Trichomes: No
Venation: Primaries form reticulum with polygonal areoles, a few secondaries
interconnect or end freely.

Atraphaxis
Species examined: Atraphaxis spinosa Eichw. (Viedmnu 127-2 A)
Type of fruit: Indehiscent
Length: 0.8 cm
Width: 0.85 cm
Shape in outline: Ovate, apex rounded, base cordate
Ovary condition: Superior
Persistence of styles/perianth: Wings and perianth at base
Persistence of pedicel/stipe: Pedicel
Carpels: 2-3
Number of wings: 3
Margins: Entire
Wing derivation: Perianth
Wing thickness: Papery
Trichomes: No
Venation: Single primary running from base to apex over central body, secondaries
dichotomizing and anastomosing occasionally across the wing.

Rum ex
Species examined: Rumex obtusifolius Herb.Buch.-Ham. Ex Wall. (HARVARD SEED
COLLECTION); Rumex venosus Pursh (77, USNSH).
Type of fruit: Indehiscent
Length: .8 cm/ 2.3 cm
Width: .45 cm/ 3 cm
Shape in outline: Obovate/ wide elliptical
Ovary condition: Superior








Persistence of styles/perianth: Perianth forming wings
Persistence of pedicel/stipe: Stipe on obtusifolius
Carpels: 3
Number of wings: 3
Margins: Entire
Wing derivation: Tepals
Wing thickness: Papery
Trichomes: No
Venation: Single order forming reticulum with polygonal areoles or primaries forming
reticulum with secondaries running laterally forming interconnections.

Rhamnaceae

Anatomy and dispersability of the winged fruits of one subfamily, the Gouaniaeae, has

been studied in detail (Medan 1988).

Gouania (Figure 2-7A)
Species examined: Gouania domingensis Aubl. (Rdrgez256 GH) Gouania
hipuloides Urb. (Brumac 937 GH) (Brass 2113 8, GH)
Type of fruit: Capsule
Length: 1 cm/ 1.1 cm
Width: 1.4 cm/ 0.9 cm
Shape in outline: Wide elliptical
Ovary condition: Superior
Persistence of styles/perianth: Apical disk
Persistence of pedicel/stipe: Stipe
Carpels: 3
Number of wings: 3
Margins: Entire often with an intramarginal vein
Wing derivation: Ovary
Wing thickness: Coriaceous
Trichomes: No
Venation: Irregular meshwork of irregularly spaced asymmetrical veins forming large
irregular areoles marginal vein sometimes present.

Chaydaia (Figure 2-7B)
Species examined: Chaydaia berchemiaefolia Koidz. (Hrdr 01 USNSH)
Type of fruit: Septicidal capsule
Length: 1.4 cm
Width: 1.1 cm
Shape in outline: Circular
Ovary condition: Superior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: Stipe
Carpels: 3
Number of wings: 3
Margins: Undulating








Wing derivation: Ovary/hypanthium
Wing thickness: Papery
Trichomes: No
Venation: Single order of irregularly looping veins forming large polyhedral areoles
sometimes looping to form an intramarginal vein.

Reissekia (Figure 2-7C)
Species examined: Reissekia smilacina Endl. (L. Wlliams 6834, USNSH)
Type of fruit: Inflated schizocarp
Length: 1 cm
Width: 1.6 cm
Shape in outline: Butterfly shaped
Ovary condition: Superior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: No
Carpels: 3
Number of wings: 3
Margins: Entire with marginal vein
Wing derivation:
Wing thickness: Papery
Trichomes: No
Venation: Marginal vein and one order of veins radiating outwards dichotomizing and
anastomosing rarely in wing but frequently over the central body.

Rutaceae

The subfamily Pteleinae has three genera with winged fruits (Ptelea, Helietta, and

Balfourodendron), and one genus (Taravalia)~TTT~~TT~~~TT~~ with a nut like capsule. Of these, Ptelea and

Balfouodendron are fin winged while Helietta is similar to Acer in having two asymmetrical

mericarps (Pirani 1998). Spathelia is another genus in the family with fin-winged fruits, but is

not described here.

Balfourodendron (Figure 2-8A). Balfourodendron is found in both Central and South

America. Pirani (1998) describes the samara as typically having four one-seeded locules in

which most of the seeds are aborted. The epicarp, which forms the wings, is described as thin

and coriaceous, the mesocarp thicker, soft and dotted glandular dotted, the endocarp forms a hard

lignified stone (Pirani 1998).

Species examined: Balfourodendron riedelianum Engl. (Vanni et al. 385, GH)
Type of fruit: Indehiscent samara








Length: 1.7 cm
Width: 2.3 cm
Shape in outline: Widely elliptical
Ovary condition: Superior
Persistence of styles/perianth: Remnants only
Persistence of pedicel/stipe: Stipe present
Carpels: 4-5
Number of wings: 4
Margins: Entire
Wing derivation: Ovary
Wing thickness: Papery
Trichomes: No
Venation: Three orders of venation, the first order arcing downwards, the second
interconnecting, and third forming a reticulum. Venation is sinuous. Areoles
polygonal and small (2/10 mm).

Ptelea (Figure 2-8BC). Fossilized Ptelea samaras from North America were critically

examined by Call and Dilcher (1995). They found three features to be particularly important in

identifying the fruits of this genus 1) Superior. ovary demonstrated by floral disk at the base of

the fruit where it j oins the pedicel, 2) wings fused above and below the seed cavity, and 3) three

slightly diverging veins extending medially through the lower half of the samara forming a

coarse transversely oriented reticulum over the surface of the fruit body and a radiating looping

reticulate pattern on the wings. I agree with these characters, although P. crenulata venation

does not loop, but rather dichotomizes.

Species examined: Ptelea crenulata Greene (HARVARD SEED COLLECTION), Ptelea
polyadenia Greene (HARVARD SEED COLLECTION), Ptelea trifolata L.
(HARVARD SEED COLLECTION)
Type of fruit: Samara
Length: 2.3 cm/ 1.5 cm
Width: 2.5 cm/ 1.2 cm
Shape in outline: Rounded
Ovary condition: Superior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: Pedicel and stipe
Carpels: 2-3 locules
Number of wings: 2 or 3
Margins: Entire somewhat wavy; tattered in P. crenulata
Wing derivation: Ovary
Wing thickness: Papery
Trichomes: No








Venation: In P. crenulata veins dichotomizing and anastomosing irregularly forming
elongate areoles; in P. trifolata forming loops to create a reticulum of polygonal
areoles.

Sapindaceae

The Sapindaceae have many genera with fruits that are dispersed by air. Some of these

fruits are fin-winged and others are not. All of the fin-winged fruits have a marginal vein.

Detailed comparisons of the fruits of the Paullineae have been made (Weckerle and Rutishauser,

2005). The fruits in this section may have internal hairs that are either unicellular or

multicellular and all are three carpellate. Genera with fin-winged fruits include Dodonaea,

Guioa, Majidea, Serjania 7houindium Toulicia Stocksia and Urvillea. Five of these genera are

described below.

Guioa
Species examined: Guioa comesperma Radlk. (Taucuhi 673 GH)
Type of fruit: Capsule
Length: 2 cm
Width: 2.3cm
Shape in outline: Obcordate
Ovary condition: Superior
Persistence of styles/perianth: Disk at base
Persistence of pedicel/stipe: No
Carpels: 1 (3 locular)
Number of wings: 3
Margins: Entire
Wing derivation: Ovary
Wing thickness: Woody
Trichomes: No
Venation: Unavailable because the wing was too woody

Stocksia
Species examined: Stocksia brauhica image from Steven Manchester only
Shape in outline: Ovate
Ovary condition: Superior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: Pedicel
Margins: Entire with marginal vein
Wing derivation: Ovary
Venation: Reticulum reaching to margin with secondaries interconnecting or feeling
ending in branched or unbranched veinlets.








Thouindium
Species examined: Thouindium image from Steven Manchester only.
Type of fruit: Capsule.
Shape in outline: Elongate, base slightly cordate, apex opening.
Ovary condition: Superior.
Persistence of styles/perianth: Basal disk.
Persistence of pedicel/stipe: Small pedicel.
Number of wings: 3.
Margins: Entire with marginal vein.
Wing derivation: Ovary.
Venation: Single order coursing downwards dichotomizing and anastomosing to form
polygonal to elongate areoles; veins entering the marginal vein.

Toulicia
Species examined: Toulicia image from Steven Manchester only.
Shape in outline: Rounded obovate.
Ovary condition: Superior.
Persistence of styles/perianth: No.
Persistence of pedicel/stipe: No.
Number of wings: 3.
Margins: Entire with marginal vein.
Wing derivation: Ovary.
Trichomes: No.
Venation: Primaries radiating, dichotomizing and anastomosing entering marginal vein;
secondaries interconnecting and forming reticulum of polygonal areoles or freely
ending.

Urvillea (Figure 2-8D)
Species examined: Urvillea andersonii Ferrucci (Anderson 9150, NY)
Type of fruit: Septifragal capsule
Length: 3.6 cm
Width: 1.7 cm
Shape in outline: Elongate
Ovary condition: Superior
Persistence of styles/perianth: Styles at apex
Persistence of pedicel/stipe: Pedicel present, stipe present
Carpels: 3
Number of wings: 2
Margins: Entire with marginal vein
Wing derivation: Ovary
Wing thickness: Chartaceous
Trichomes: No
Venation: Marginal vein and radiating sub parallel with some dichotomizing








Styracaceae


Halesia (Figure 2-5F). Halesia fruit fossils are found in Europe and reports have been

made of fruits in North America. Fritsch et al. (2001) used these reports, and molecular

phylogenies to suggest a North American origin for the genus. They admit that the fossils do not

show the intramarginal vein found in Halesia species, but they still consider it to be this genus.

Species examined: Halesia carolina L. (Hill 17001, GH), Halesia diptera L. (Habiso
1129, GH)
Type of fruit: Indehiscent drupe
Length: 3.7cm
Width: 1.7 cm
Shape in outline: Spatulate
Apex: Cordate-sagittate
Base: Obtuse
Ovary condition: Inferior
Persistence of styles/perianth: Calyx at apex
Persistence of pedicel/stipe: Pedicel
Carpels: 2
Number of wings: 2, 4
Margins: Entire wavy
Wing derivation: Ovary
Wing thickness: Thick semi-opaque
Trichomes: No
Venation: Intramarginal veins present
Other: Fruits may persist on branches even after leaves seasonally fall, prominent apical
beak present.

Trigonaceae

Humbertiodendron (Figure 2-7F)
Species examined: Humbertiodendron soboundanidddd~~~~~~ddddd Leandri (Anonymous, 244, USNSH)
Type of fruit: Indehiscent samara
Length: 2.5 cm
Width: 2.3 cm
Shape in outline: Round
Ovary condition: Superior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: None dispersed, pedicle present on branch
Carpels: 3
Number of wings: 3
Margins: Entire.
Wing derivation: Ovary
Wing thickness: Chartaceous








Trichomes: Yes (long-pilose)
Venation: Marginal vein; primary veins transverse to radiating upwards at approximately
45 degrees; occasionally dichotomizing in wing but dichotomizing more frequently
near margmn.

Tropaeolaceae

Tropaeolum (Figure 2-7H)
Species examined: Tropaeolum porifolium (Cay.) L.Andersson & S.Andersson (A Dna
51, USNSH)
Type of fruit: Schizocarp
Length: 1.6 cm
Width: 1.3 cm.
Shape in outline: Longitudinally elliptical
Ovary condition: Superior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: No
Carpels: 3
Number of wings: 3
Margins: Entire not smooth with a marginal vein
Wing derivation:
Wing thickness: Papery
Trichomes: No
Venation: Marginal vein; radiating outwards occasionally dichotomizing near the margin

Zygophyllaceae

Fruits of Zygophyllaceae may be capsular or Schizocarp.ic (Beier 2003). Some of these

fruits develop wings in the fin-winged fashion. Four genera are described below.

Bulnzesia (Figure 2-6AB)
Species examined: Bublesia chilensis Gay. (Werdermann 429, GH); Bublesia foliose
Griseb. (Schreiter 2849, GH); Bublesia retamtttttttt~~~~~~~~~o Griseb. (Hutchinson & Wrinht 7115,
GH)
Type of fruit: Schizocarp
Length: 1.6cm/ 4cm
Width: 1.3/ 4 cm
Shape in outline: Elliptical to round/elliptical
Ovary condition: Superior
Persistence of styles/perianth: Perianth at base in B. foliosa and B. chiliensis; stylar
protrusion in B. foliose
Persistence of pedicel/stipe: Stipe, sometimes pedicle
Carpels: 5.
Number of wings: 5.
Margins: Entire.
Wing derivation: Ovary.
Wing thickness: Chartaceous.








Trichomes: Yes unbranched on B. foliose and B. riva~s-martinezii (see Navarro, 1994).
Venation: Radiating, dichotomizing, looping at margin with some interconnections
particularly towards the margin.

Morkillia
Species examined: Morkillia mexicana Rose & Painter (Sous 460, GH) (Chase 7332,
GH)
Type of fruit: Capsule
Length: 3.5 cm/ 3.75 cm
Width: 2.5 cm/ 2 cm
Shape in outline: Elliptical-ovate with truncate apex or apex with rounded invagination
creating a saggitate appearance
Ovary condition: Superior
Persistence of styles/perianth: No
Persistence of pedicel/stipe: Pedicel
Carpels: 4
Number of wings: 4
Margins: Entire
Wing derivation: Ovary
Wing thickness: Coriaceous
Trichomes: Yes, densely hairy
Venation: Unknown

Porleria (Figure 2-6D)
Species examined: Porlieria angustifolia A.Gray (Moore 3288, Harvard)
Type of fruit: Schizocarp
Length: 1.7 cm
Width: 1.7 cm
Shape in outline: Widely ovate
Ovary condition: Superior
Persistence of styles/perianth: Stylar protrusion and basal disk
Persistence of pedicel/stipe: Yes, both
Carpels: 4
Number of wings: 3
Margins: Entire
Wing derivation: Ovary
Wing thickness: Coriaceous
Trichomes: Yes unbranched
Venation: Not visible by transmitted or reflected light








Zygophyllum (Figure 2-6C)
Species examined: (Zygophyllunt norgsana L. (Marloth 12275, A); Zygophyllunt
retrofr~actunt Thunb. (Maroth650 A)
Type of fruit: Schizocarp
Length: 2 cm (+1cm. pedicel)/ 3 cm (+several centimeter pedicel)
Width: 2.5cm/ 3cm
Shape in outline: Elliptical
Persistence of styles/perianth: Style at apex
Persistence of pedicel/stipe: Pedicel
Carpels: 4
Number of wings: 4
Margins: Entire
Wing derivation: Ovary
Wing thickness: Chataceous
Trichomes: No
Venation: Dichotomizing anastomosing and looping
Ovary condition: Superior. to form a reticulum of areoles, that are transversely elongate
towards the central body and increasingly polygonal toward the margin.









CHAPTER 4
RESULTS

Fossil Fin-Winged Fruits of North America With Possible Affinities to Combretaceae

The fossil history of the Combretaceae has been reported to include fossilized leaves and

fruits (ex. Berry 1916), wood (ex. Sakala 2002), a possible flower (Takahashi et al. 1999) and

pollen (ex. Salzman 2000). Three winged-fruit fossils from North America will be treated here

(Terminalia vera, Terminalia estamtttttttt~~~~~~~~~in and Terminalia oregona). These fruits were not assigned

to the genus Terminalia because it was the best fit based on careful comparisons, but simply

because it is a genus in the Combretaceae. Additionally, these fruits were not compared to all of

the genera that have fin-winged fruits, and it may be that there is a better match among the

modern fin-winged fruits than any of the genera of Combretaceae.

Carpolithes prangosoides Berry

Basionym: Call~~ipalithe pranzgosoides Berry 1916, USGS Prof. Paper 91, p. 351, pl. 104,

Figure 2-9.

Terminalia vera. Berry 1926. Torrey Botanical Club Bulletin. Vol. 53, p. 61, figs. 1-5.

Carpolithus henryensis. Berry 1916, USGS Prof. Paper 91, p. 352, pl. 112, fig. 16.

Emended Description

Fruit fin-winged, usually radially symmetrical, appearing bisymmetric in face view,

elliptical, 1.2 -3.3 cm long, 1-1.5cm wide; composed of a central body with 5 fin-like,

radiating wings, pedicel not dispersed with fruit. Style not persistent.

Fruit body elongate, appearing smooth or longitudinally striated. Each wing lies in a

radial plane intersecting with the longitudinal axis of the fruit, auriform to half-elliptic in shape.

Near the distal end of the fruit, apical flaps persist in some specimens and it is unclear from what

tissue these flaps arise. The wing venation is uniformly oriented towards the apex at about 30










degrees from the fruit's central axis. Venation consists of numerous fine veins, frequently

dichotomizing, forming loops along the way toward the margin, with secondary veins

interconnecting and forming areoles of the reticulum 0.7-1.5 mm. Marginal or intramarginal vein

absent.

Comments. This species is common in claypits of the Middle Eocene Claiborne Fm in

Tennessee and Kentucky. Although the original specimen figured as Calrpolithus pranzgosoides

(Berry 1916) does not clearly show the wing venation, it does show a pair of elongate lateral

wings that can be matched with specimens found later and described as Terminalia vera Berry

(1930). An elongate central bulge in this specimen, representing the fruit body shows a median

longitudinal groove that we interpret as a third wing, while the counterpart of this specimen

(broken) shows at least one additional groove, which we interpret as another wing. Caiptlithes~

henryensis is represented by only a single specimen, which lacks a counterpart. This specimen

also shows a central body with elongate lateral wings with a groove, which is interpreted as a

third wing. Based on the observed features, these specimens can be matched to those Berry

(1930) illustrated later under the name Terminalia vera, and we place all of these in synonymy

under the earlier name Cai po~lithe1~ prangosoides.

Most of the C. prangosoides specimens are exposed in longitudinal fractures that show

only two visible wings, positioned laterally on either side of the fusiform central body and lying

in a common plane (Figures 2-10 & 2-11). This configuration is an artifact of the horizontal

orientation of the fruit within the sediment in which it was buried, and of the lengthwise plane

through the sediment that intercepted and exposed the fossil. In describing Terminalia vera,

Berry considered that the fruits possessed only two wings. (Berry 1930). Careful microscopy of

some specimens shows a stairstep fracture pattern in which multiple levels on which different










wings are exposed. In order to determine the full number of wings, two specimens were

deliberately fractured transversely in the laboratory and show unequivocally that there are five

wings (Figure 2-10).

None of the specimens show the pedicel, styles or the locule condition. Because pedicels

have not been found attached to the fruits, it may be inferred that the fruits were sessile or that

they were shed by abscising from the top of the pedicel. Because the fruit is always found with

a full complement of wings it is inferred that it was not schizocarpic. But the available

characters, including the number and position of wings and their venation allows for detailed

comparison with extant fruits of similar form.

Although this species was assigned to the Combretaceae based on its lateral wings,

detailed comparison of wing venation distinguishes the fossil from most of the genera within this

family. The complex system with primary veins forming loops and secondaries interconnecting

is similar to the venation in the wings of Calycopteris and ]\ Art i pIi anth )rs~lrr Calycopteris does

not have a Ein-winged fruit; its wings are arranged like a propeller around the apex, and are

derived from accrescent perianth. 3\~I A llrtiopcitheslr', which is a member of a separate tribe,

Lagunculariaeae (Excell and Stace 1965) is Ein-winged with four wings derived from two

bracteoles, such that a pair of wings extends from each bracteole. The venation is strikingly

similar to that of the fossil. The geometry of a Hyve-winged fruit would be difficult to reconcile

with formation of two wings per bracteole.

Other families with Hyve winged fruits include Apiaceae, Lophopyxidaceae, Malvaceae

and Zygophyllaceae (Table 2-1). The fruits of Zygophyllaceae are mericarps and those of

Malvaceae are usually capsular so those two families can be eliminated. Apiaceae do not have

prominent venation and can be eliminated and Lophopyxidaceae lack the looping pattern of









venation on the fruits. Although no genus fits closely; the family, Combretaceae seems to be

morphologically most similar to these fossils. The fruit lacks the distinguishing characteristics of

the family such as combretaceous hairs, and because these fruits may be those of an in an extinct

genus, there is not enough evidence to place this fossil in the family.

"Terminalia stamina MacGinitie

MacGinitie (1941) described Terminalia estamtttttttt~~~~~~~~~in based on three unattached leaf

specimens and a winged fruit from the Middle Eocene Chalk bluffs flora of California. He

called the cited specimens "cotypes," but did not designate a holotype. Subsequently

(MacGinitie 1969, p. 129), he made a formal new combination, transferring the species to

M~astixia estamtttttttt~~~~~~~~~in (MacGinitie) MacGinitie, based on two of the leaf specimens that had been

Eigured earlier. In this transfer, he cited the two of the leaf specimens but specifically excluded

the remaining leaf and fruit specimens, stating that "only Figure 2-2, plate 42 represents a leaf of

Terminalia" and "The fruit, Figure 2-4, plate 43, is correctly assigned to Terminalia." This

created the awkward situation that the two specimens that MacGinitie (1969) still considered to

represent Terminalia were left without an epithet. To resolve this situation, a new epithet will

have to be given to this fossil.

Description. Fruit elliptical in outline, symmetrical, rounded over the apex, length is 3-

3.5 cm, width 2.5 cm. The fossil fruit has a fusiform central body, 13 mm long, 4 mm wide,

from which at least Hyve longitudinal wings radiate. Fruit is indehiscent. An axial vein extends

about 7 mm from the base of the fruit to the base of the central body. Venation of the wings

radiates from the central body toward the margin. Veins of the wing Eine, dichotomizing and

anastomosing, spaced from 0.5-1 mm apart without prominent looping or cross-veins. No

marginal vein present. Features of style and perianth are unknown.









Comments. I have seen only two fruit specimens of this species, both of which are

broken and incomplete. One of the specimens has been fractured obliquely across the fruit, such

that impression surfaces of five wings can be counted (Figure 2-11). Both specimens seem to

have the wings intact and because no dispersed mericarps of this type have been found in the

fossil flora, we believe it to be non-schizocarpic, indehiscent fruit.

In his treatment, MacGinitie compared the Eocene fossil fruits that he called Terminalia

estamtttttttt~~~~~~~~~in to the modern species Terminalia hainanensis and Terminalia triflora, which I have not

observed in fruit. The other species of the genus that I examined also showed some similarity of

form and venation.

Important features of this fossil fruit include apparent indehiscence, a fusiform central

body, and presence of five wings with fine venation which dichotomizes. Other genera of

angiosperms with a similar pattern of venation include Ptelea, Wimmeria, Tripterigium,

Bulnesia, Tropaelum, and Pterodiscus. None of these genera have five wings as does the fossil

and as do members of the Combretaceae (Table 2-1). Pentace is a genus with similar venation

and five wings, but the wing edges in this genus are reinforced by many dichotomizations near

the margin, which is not consistent with the fossil because the dichotomizations lie within the

wing rather than along the margin. If we exclude those modern genera with fewer than five

wings and those with dehiscent fruits, the only remaining candidates for comparison with this

fossil are in the Lophopyxidaceae and Combretaceae (Table 2-1). The vein density of

Lophopyxis is much greater with at least five veins per millimeter rather than 1-2 veins per

millimeter of the fossil. Fruits of Combretum and Terminalia (Combretaceae) can show similar

form to the fossil, including the presence of a fusiform central body that is positioned above the









base of the fruit. They also tend to have higher vein density than the fossil and their margins

tend to be ragged due to lack of reinforcing venation.

Of the modern fin-winged fruits, fruits of genera within the Combretaceae seem to be

most similar to these fossils. Some diagnostic features of the family could not be observed in

these fossils including the inferior ovary, the number and positioning of seeds, presence of

combretaceous hairs. Additionally, the venation is quite different in .\ iL i i inh esllr'

M~eiostemon and Thiloa have only four wings and Pteleopsis has only two. Within the

Combretaceae, the fruit fits best as either the genus Terminalia or the genus Combretum. This

does not mean that this necessarily belongs to this family because the similarity lies partly in the

absence of any distinguishing characters, and the fossils may represent an extinct species in

another family.

Indeed, the fossil seems to differ from both Combretum and Terminalia in some aspects.

The genus Combretum tends to have edges which fray because the veins so rarely dichotomize

and the venation is very fine so that there is nothing to stop a small tear from becoming a large

one. The fossil in contrast has smooth edges and there are clearly spaces between the veins.

Terminalia may also have these very fine veins and tattered edges, but it can also have more

widely spaced veins and smoother edges (e.g. Terminalia). The assignment of this fossil to the

genus Combretaceae is not certain without confirming characters such as combretaceous hairs or

apical placentation, and the assignment to the genus Terminalia is certainly not justified.

Terminalia oregona (Lakhanpal) Meyer and Manchester

Basionym: Halesia oregona Lakhanpal, 1958, pro parte, University of California

Publications in Geological Sciences 35, p. 36, pl. 9, fig. 1, Pl. 10, Figure 2-4 only.

Terminalia sp. Brown 1959, pro pate, Journal of Paleontology 33, p. 128; Pl. 24, Figure

2-16 only.









Terminalia oregona (Lakhanpal) Meyer and Manchester, 1997, University of California

Publications in Geological Sciences 141, p. 128, pl. 51, Figures 2-1 through 2-4.

Description. This fruit is elliptical to obovate, pedicellate, symmetrical, base ranging

from cuneate and acute to rounded and cordate, apex obcordate and rounded, length (20 mm, 25

mm, 45 mm, 33 mm, 40 mm, 55 mm, 33 mm), width (18 mm, 20 mm, 31 mm, 30 mm, 20 mm,

27 mm, 25 mm). Pedicel is 1 mm wide and up to 14 mm long. The fruit is nonschizocarpic. The

fossil fruit has a fusiform central body mainly confined to the upper 1/3 of the fruit, 17-21 mm

long, and 5-6.5mm wide with several prominent longitudinal ribs. A single vein leads from the

pedicel to the central body. Fruits have up to at least three wings. Fruits with at least two wings,

and in one specimen (Figure 2-13 5,6) at least three were observed. Veins of moderate

thickness, radiating outward from the central axis and occasionally dichotomizing, spaced 0.5

mm apart, without prominent looping or cross veins, no marginal vein present. Stylar protrusion

with a thickened disk like structure at its junction with the central body.

Comments. Important features include apparently three wings, inferior ovary with

persistent stylar protrusion and apical disk, central body ridged and situated in the upper 1/3 of

the fruit. Modern fin-winged fruits with an inferior ovary include Apiaceae, Begoniaceae,

Burmanniaceae, Combretaceae, Dioscoreaceae, Hernandiaceae, Loranthaceae, Onagraceae, and

Styracaceae. We are uncertain if they were variable in wing number from 2-3 or whether those

specimens showing two wings are incompletely preserved.

Dioscoreaceae and Begoniaceae are eliminated because they have a marginal vein which

is not found on the fossil. Lakhanpal (1958) believed it might be Halesia (Styracaceae). Halesia

is distinguished by prominent intramarginal veins on the wings distinctive of this genus and not

seen in the fossil. Halesia has perianth remaining at the apex where the fossil does not (Figure










2-7F). Apiaceae fruits are typically smaller without obvious venation. Hernandiaceae venation

was sinuous without dichotomizations. Onagraceae venation tends to be obscure or arching

upward. Loranthaceae venation tends to be more closely spaced. Each character in this fossil

can be found in some species of the Combretaceae, but not in combination.

Terminalia oregona is a fossil with characters not inconsistent with the Combretaceae.

This fruit has three wings with radiating venation that only rarely dichotomizes. This venation

pattern is consistent with some species in both Terminalia and Combretum. The apical

protrusion is similar to that found in M~eiostemon and the apical position of the seed cavity can be

found in Pteleopsis.

None of these fossils shows characters that would clearly place it into the Combretaceae,

but none are wholly inconsistent. Terminalia vera has venation strikingly similar to

Ma'~cropteranthus, but the difference in wing number is difficult to reconcile. Terminalia

estamtttttttt~~~~~~~~~in and Terminalia oregona are both most similar to either Combretum or Terminalia.

Terminalia estamtttttttt~~~~~~~~~in has five wings and radiating venation, but the veins are not closely spaced

as is the common condition. Terminalia oregona, also has venation which is more widely

spaced than usual for the family.

Overall, the lack of any diagnostic character placing these fruits in this family, the lack of

critical examination of the leaves formerly placed in Combretaceae leaves me unwilling to state

that there was a record of the family in the United States during the Eocene-Oligocene.









CHAPTER 5
CONCLUSIONS

Based on the previous chapters I have been able to compare the various modern fin-

winged fruits and determine the status of the fossils. These results are presented below.

Modern Winged Fruits

Based on the survey presented in chapter three, fin-winged fruits occur in at least 19

orders of Angiosperms, occurring in magnoliids, monocots, and eudicots. Within eudicots, such

fruits are particularly well represented among the rosids, with 18 of the 30O families that include

genera with fin-winged fruits. These fruits are also represented in basal eudicots and asterids

(Figure 2-8). Anemoballistic fruits were found convergently in three families.

Of the 87 genera known, only 34 had an inferior ovary, less than half had a marginal

vein, suture, or intermarginal vein. Many of those that lack the marginal vein did have

reinforcement towards the edge, either dichotomizing more frequently or forming loops distally.

Presumably this helps prevent the wings from becoming tattered. Fruits of Conabretuns, which

lack marginal veins and looping, are often frayed along the margin (Figure 2-2).

Wings can be formed from ovary tissue, from perianth parts and/or bracts. Those formed

from perianth or bracts may take the form of lamina appressed to the ovary. In these cases, the

wings may be formed from the fusion or partial fusion of two adj acent lamina (Reisseika) or each

lamina may form two wings freely (Ma'~cropteranthus).

The fin-winged fruit type is found on every continent except Antarctica, and in habitats

ranging from desert (Ma'~cropteranthus) to rainforest (Cavanellesia). These fruits occur on trees

(Cavanellisia, CraigiaBalfourodendron), shrubs (Phaeoptihtna, M~ancellia, Winaneria) vines

(Dioscorea, Triptergium, Kleinhovia) and herbs (Polygonunt, Stackhousia, Abronia).










Fin-winged fruits have enough distinguishing characters to separate them by family. A

dichotomous key to the Ein-winged fruits (by family) is given below.

la. Fruit from an inferior ovary
2a. Fruit dispersing via anemoballisty with many seeds
3a. Fruit with two or four wings ................. ...............Onagraceae...... ...
3b. Fruit with three wings
4a. Fruit opening apically, wings equal in size, seeds
may be winged ................. ............. ..... Dioscoreaceae/Burmanniaceae
4b. Fruit splitting along sutures, wings equal or unequal
in size, seeds not winged ................. .......... ............... Begoniaceae.
2b. Fruit dispersing otherwise, capsule, mericarp, samara etc.
Sa. Fruits wing with intramarginal vein ........._..._.._ ...._._. ............... Styracaceae
5b. Fruits without intramarginal vein
6a. Fruit a schizocarp ............... ...............Apiaceae._ ......
6b. Fruit non-schizocarpic
7a. Unicarpellate fruit with sinuous venation ................... ......... Hernandiaceae
7b. Three-Hyve carpellate, venation otherwise
8a. Fruit small (1cm), wings 3, carpels 3-4 ................... .......Loranthaceae
8b. Fruits usually greater than 1cm, carpels 4-5,
wings 2, (3), 4, 5 ................. ............... ......... ...Combretaceae
lb. Fruit from a superior ovary
9a. Number of wings Hyve or greater
10a. Venation arching downwards ................. ........... ................ Lophopyxidaceae
10b. Venation otherwise (radiating, transverse, ascending)
11a. Wings six-many, may be incomplete, fruit indehiscent. .................. Achariaceae
11Ib. Wings Hyve (or fewer in M~amellia), wings complete,
membranous-chartaceous coriaceouss in M~amellia),
fruit a capsule .................. .. .......... .............. M alvaceae...
9b. Number of wings less than five (2, 3, or 4)
12a. Marginal vein present
13a. Four wings, fruit nearly round in outline ................. .........._._ ....Pedaliaceae
13b. Two-three wings, fruit not round in outline
14a. Fruits indehiscent
15a. Marginal vein only, style persistent, long (1cm)
and thick (2mm) ................. .......... ................ Cardiopteridaceae
15b. Venation more than just marginal vein, style if
persistent less than 1cm and less than 2mm thick.
16a. Calyx persisting at the base of fruit ................... ....... Celastraceae
16b. Calyx not persisting ................. ...................... Cunonaceae
14b. Fruits dehiscent, capsular or schizocarpic
17a. Two winged
18a. Fruit schizocarpic, bilocular ovary ................... .... Phyllanthaceae
18b.Fruit capsular, ovary with three carpels ...Sapindaceae (Urvillea)
17b. Three winged










19a. Fruit schizocarpic
20a.Calyx persistent at base, venation radiating
from central body with no interconnections,
joining the marginal vein ................... ......... Stackhousiaceae
20b. Calyx not persistent, venation otherwise
21a. Fruit elliptical, longer than wide............. Tropaelaceae
21b. Fruit butterfly shaped, wider
than long............... ............... Rhamnaceae (Reisseika)
19b.Fruit capsular
22a. Venation irregular, asymmetrical, forming
variably sized polygonal areoles ................... .... Rhamnaceae
22b. Venation radiating, dichotomizing and
anastomosing............... ........... Sapid a ea
12b.Marginal vein absent
23a. Fruit schizocarpic, four wings ................. ................. .... Zygophyllaceae
23b. Fruit non-schizocarpic, wings 2,3, or 5
24a. Fruit coriaceous with many small seeds/fruit........ Malvaceae (Maxwellia)
24b. Fruit not coriaceous, fewer seeds/fruit
25a. Fruit formed from persistent perianth
26a. Fruit an achene .............. ................... Poly gonaceae
26b. Frit otherwise ................. .....___..................Nytaginaceae
25b. Fruit formed from ovary
27a. Sepals or disk persistent at base
28a. Sepals persistent, no persistent disk.................... Cyrillaceae
28b. Sepals not persistent, disk persistent................ Polygalaceae
27b. Sepals/disk not persistent
29a. Three-winged
30a. Veins forming loops, trichomes absent.. Malpighiaceae
30b. Veins dichotomizing but not looping,
trichomes present ................. ................ ..Trigonaceae
29b. Two or four winged
31a. Fruit longer than wide ............. ...............Fabaceae
31Ib. Fruit wider than long or nearly equal.............. Rutaceae

Fossil Winged Fruits

As presented in Chapter 4, three different occurrences of fossil fin-winged fruits have

been attributed previously to Combretaceae in North America. Two are from the Eocene, one

from the Oligocene, and all have been assigned to the genus Terminalia. Each of these was

reexamined with attention to the original fossil specimens and closely compared with extant

fruits of Combretaceae and other extant families. The features preserved in these fossils are

insufficient to confirm identification to that family. Although similar in many respects to










Conabretunt and Ternzinalia, these fossils may represent convergence in fruit form and could

represent another family.

In order to confidently assign a fossil fin-winged fruit to the Combretaceae, characters

such as cross sectional anatomy, seed condition, or presence of "combretaceous hairs"(long,

straight, sharply pointed, unicellular, very thick walled, with conical internal component at base),

should be present. Tattered wing margins are common in Conabretunt and could be a

contributing character when analyzing a fossil.

It is hoped that the survey of extant fin-winged fruits presented here will provide an

improved basis for evaluating similar types of fruits from the fossil record. Such winged fruits

are known as early as the late Albian of Kansas (Dilcher, pers. comm. 2006) and are well

represented in Tertiary floras. Special attention to details of venation as well as features of

persistent perianth, bracts, disks, when present can facilitate confident identification of extant

genera, and provides a basis for recognizing extinct taxa.









LIST OF REFERENCES


Anderberg, AA & Zhang X, 2002. Phylogenetic relationships of Cyrillaceae and Clethraceae
(Ericales) with special emphasis on the genus Purdiaeae planch. Organisms Diversity and
Evolution 2:127-137.

Angiosperm Phylogeny Group 2003. An update of the angiosperm phylogeny group
classification for the orders and families of flowering plants: APG II Botanical Journal of
the Linnean Society 141:399-436.

Augsperger, CK, 1988. Mass allocation, moisture content, and dispersal capacity of wind-
dispersed tropical diaspores. New Phytologist 108:357-368.

Bailey, VL, 1962. Revision of the genus Ptelea (Rutaceae). Brittonia 14:1-45.

Barneby, RC & Grimes, JW, 1984. Two leguminous forest trees new to the flora of French
Guiana. Brittonia 36:45-48.

Barnes, RW & Hill, RS, 1999. Ceratopetalum fruits from Australian Cainozoic sediments and
their significance for petal evolution in the genus. Australian Systematic Botany 12:635-
645.

Beier, B-A, Chase, MW & Thulin, M, 2003. Phylogenetic relationships and taxonomy of
subfamily Zygophylloideae (Zygophyllaceae) based on molecular and morphological data.
Plant Systematics and Evolution 240: 11-39.

Berry, EW, 1916. The Lower Eocene floras of Southeastern North America. US Geological
Survey Professional Paper 91:1-481.

Berry, EW, 1926. Terminalia in the Lower Eocene of Southeastern North America. Bulletin of
the Torrey Botanical Club 53:59-65.

Berry, EW, 1930. Revision of the Lower Eocene Wilcox flora of the Southeastern States with
descriptions of new species, chiefly from Tennessee and Kentucky. U.S. Geological
Survey Professional Paper 156:1-196.

Bradford, JC & Barnes, RW, 2001. Phylogenetics and classification of Cunoniaceae
(Oxalidales) using chloroplast DNA sequences and morphology. Systematic Botany
26:354-385.

Brown, RW, 1959. A bat and some plants from the Upper Oligocene of Oregon. Journal of
Paleontology 33:125-129.

Burnham, RJ, 1995. A new species of winged fruit from the Miocene of Ecuador: Tipuana
Ecuatoriana (Legumniosae). American Journal of Botany 82:1599-1607.

Bikiek, C, Kvadek, Z, & Manchester, SR, 1989. Sapindaceous affinities of the Pteleaecarpum
fruits from the Tertiary of Eurasia and North America. Botanical Gazette 150:477-489.










Burrows, CJ, 1995. Germination behavior of the seeds of six New Zealand woody plant species.
New Zealand Joumnal of Botany 33:365-377.

Caddick, LR, Wilkin, P, Rudall, PJ, Hedderson, TAJ, & Chase, MW, 2002. Yams reclassified:
A recircumscription of Dioscoreaceae and Dioscoreales. Taxon 51:103-114.

Call, VB & Dilcher, DL, 1992. Investigations of Angiosperms from the Eocene of Southeastemn
North America: Samaras of Fraxinus wiilcoxiana berry. Review of Palaeobotany and
Palynology 74:249-266.

Call, VB & Dilcher, DL, 1995. Fossil Ptelea Samaras (Rutaceae) in North America. American
Journal of Botany 82: 1069-1073.

Call, VB & Dilcher, DL, 1997. The fossil record ofEucommia (Eucommiaceae) in North
America. American Journal of Botany 84:789-814.

Cars, P, Decraene, LPR, Smets, E & Clinckemaillie, D, 2002. The uncertain systematic position
of Symplocos (Symplocaceae) evidence from a floral ontogenetic study. International
Journal of Plant Sciences 163:67-74.

Conti, E., Litt, A, Wilson, PG, Grahm, SA, Briggs, BG, Johnson, LAS & Stoma, KJ, 1997.
Interfamilial relationships in Myrtles: Molecular phylogeny and patterns of morphological
evolution. Systematic Botany 22:629-647.

Corbett, SL & Manchester, SR, 2004. Phytogeography and fossil history ofAilan2thus
(Simroubaceae). International Joumnal of Plant Sciences 165:671-690.

Dallwitz, MJ, Paine, TA and Zurcher, EJ, 2000 onwards. Principles of interactive keys.
http ://delta-intkeyccom.

Duke, JA, 1961. Flora of Panama. Part IV. Fascicle IV. Annals of the Missouri Botanical
Garden 48:1-106.

Dwyer, JD & Hermann, FJ, 1965. Flora of Panama. Part V. Fascicle 4. Annals of the Missouri
Botanical Garden 52:1-54.

Excell AW and Stace, CA, 1965. Revision of the Combretaceae. Boletim da Sociedade
Broteriana 40:5-26.

Ferrucci, MS, 2006. A new species of th villea (Sapindaceae) from Northwestern Venezuela.
Brittonia 58:83-87.

Fritsch, P, Morton, CM, Chen, T, & Meldrum, C, 2001. Phylogeny and biogeography of the
Styracaceae. International Journal of Plant Sciences 16:S95-S116.

Galloway, L, 1975. Systematics of the North American desert species of Abronia and
Tripterocalyx (Nyctaginaceae). Brittonia 27:328-347.










Garwood, NC, 1985. The role of mucilage in the germination of Cuipo, Cavanillesia
platanifolia (H. & B.) H.B.K. (Bombacaceae), a Tropical Tree. American Journal of
Botany 72:1095-1105.

Gianoli, E, 2004. Evolution of a climbing habit promotes diversification in flowering plants.
The Royal Society 271:2011-2015.

Goldberg, A, 1986. Classification, evolution, and phylogeny of the families of Dicotyledons.
Smithsonian Institution Press. Washington D.C.

Herendeen PS & Dilcher, DL, 1990. Fossil mimisoid legumes from the Eocene and Oligocene of
Southeastern North America. Review of Palaeobotany and Palynology 62:339-361.

Herendeen, PS, Les, DH, & Dilcher, DL, 1990. Fossil Ceratophyllunt (Ceratophyllaceae) from
the Tertiary of North America. American Joumnal of Botany 77:7-16.

Hj erpe, J & Hedenas, H, 2001. Tropical rain forest recovery from cyclone damage and fire in
Samoa. Biotropica 33:249-259.

Hutchinson, J, 1964. The genera of flowering plants. Clarendon Press. Oxford.

Johnson, LAS & Briggs, BG, 1984. Myrtales and Myrtaceae a phylogenetic analysis. Annals
of the Missouri Botanical Garden 71:700-756.

Jones, JH & Dilcher, DL, 1980. Investigations of angiosperms from the Eocene of North
America: Rha~nnus nzarginatus (Rhamnaceae) reexamined. American Journal of Botany
67:959-967.

Jongkind, CCH, 1995. Review of the genus Strephonenta (Combretaceae). Annals of the
Missouri Botanical Garden 82:535-541.

Judd, WS, 2006. Advanced plant taxonomy. Target Copy. Gainesville, FL.

Karehed, J, 2001. Multiple origin of the tropical forest tree family Icacinaceae. American
Journal of Botany 88:2259-2274.

Kva~ek, Z, Buzek, C, & Manchester, SR, 1991. Fossil fruits of~teleaecarpunt weylan2d-
Tiliaceous, not Sapindaceous. Botanical Gazette 152:522-523.

Lakhampal, RN, 1958. The Rujada flora of West Central Oregon. University of Califomnia
Publications in Geological Sciences 35:1-66.

Lamont, B, 1985. Dispersal of the winged fruits ofl~uytsia floribunda (Loranthaceae). Austral
Ecology 10:187-193.

Laubengayer, RA, 1937. Studies in the anatomy and morphology of the polygonaceous flower.
American Joumnal of Botany 24:329-343.










Lavin, M, Pennington, RT, Klitgaard, BB, Sprent, JI, De Lima, HC, & Gasson, PE, 2001. The
Dalberioid legumes (Fabaceae): Delimitation of a pantropical monophyletic clade.
American Joumnal of Botany 88:503-533.

Levin, RA, 2000. Phylogenetic relationships within Nytaginaceae nyctagineae: Evidence from
nuclear and chloroplast genomes. Systematic Botany 25:738-750.

Liu, M (R), Plunkett, GM, Lowry II, PP, Van Wyk, BE, & Tilney, PM, 2006. The taxonomic
value of fruit wing types in the order Apiales. American Joumnal of Botany 93:1357-1368.

Mabberly, DJ, 1989. The plant-book; A portable dictionary of the higher plants. Cambridge
University Press. New York.

MacGinitie, HD, 1941. A middle Eocene flora from the central Sierra Nevada. Camegie
Institute Washington Publication 584:1-78.

Manchester, SR, 2001. Update on the megafossil flora of Florissant, Colorado. Proceedings of
the Denver Museum of Nature & Science 4: 137-161.

Manchester, SR & Donohue, M, 1995. Winged fruits of Linnaeeae (Caprifoliaceae) in the
Tertiary of Westemn North America: Diplodipelta Gen. Nov. International Journal of Plant
Sciences 156:709-722.

Matlack, GR, 1987. Diaspore size, shape, and fall behavior in wind-dispersed plant species.
American Joumnal of Botany 74: 115 0-1 160.

Matthews, ML, Endress, PK, Schonenberger, J, & Friis, EM, 2001. A comparison of floral
structures of Anisophylleaceae and Cunonaceae and the problem of their systematic
position. Annals of Botany 88:439-455.

Matthews, ML & Endress, PK, 2005. Comparative floral structure and systematics in Celastrales
(Celastraceae, Parnassiaceae, Lepidobotryaceae). Botanical Joumnal of the Linnean Society
149:129-194.

McClain, AM & Manchester, SR, 2001. Dipteronia (Sapindaceae) from the Tertiary of North
America and implications for the phytogeographic history of the Aceroideae. American
Journal of Botany 88:1316-1325.

Medan, D, 1989. Diaspore diversity in the anemochorous Gouanicae (Rhamanceae). Plant
Systematics and Evolution 168:149-158.

Meyer, HW & Manchester, SR, 1997. The Oligocene Bridge Creek flora of the John Day
Formation, Oregon. University of California Publications in Geological Science 141:1-
195, 75 pl.










Mirle, C & Burnham, R, 1999. Identification of asymmetrically winged samaras from the
Western Hemisphere. Brittonia 51:1-14.

Navarro, G, 1994. A new species ofBulnesia (Zygophyllaceae) from the Xerothermic Southern
Puna of Bolivia. Novon 4:280-284.

Pirani, JR, 1998. A revi sion of Helietta and Balfourodendron (Rutaceae-Pteleinae) Brittonia
50: 348-380.

Razi, BA, 1950. A contribution towards the study of the dispersal mechanisms in flowering
plants of Mysore (South India). Ecology 31:282-286.

Renner, SS, 1986. Circumscription and phylogeny of the Laurales: Evidence from molecular
and morphological data. American Journal of Botany 86:1301-1315.

Schopmeyer, CS, 1974. Seeds of woody plants in the United States. Forest Service, U.S.
Department of Agriculture. Washington D.C.

Sakala, J, 2000. Silicified angiosperm wood from the Dangu locality (Ypresian of the Gisors
region, Eure, France)- final part: the problem of palaeoclimate reconstruction based on
fossil wood. Geodiversitas 22:493-497.

Stace, CA, 1965. The use of epidermal characters in phylogenetic considerations. The New
Phytologist 65:304-319.

Stace, CA, 1968. A revision of the genus Thiloa (Combretaceae). Bulletin of the Torrey
Botanical Club 95:156-165.

Stevens, PF, (2001 onwards). Angiosperm phylogeny website. Version 7, May 2006 [and more
or less continuously updated since].http://www. mobot. org/MOBOT/research/APweb/.

Spjut, RW, 1994. A systematic treatment of fruit types. Memoirs of the New York Botanical
Garden 70:1-182.

Takahashi M, Crane, PR & Ando, H, 1999. Fossil flowers and associated plant fossils from the
Kamikitaba locality (Ashizawa Formation, Futuba Group, lower Coniacian, Upper
Cretaceous) of northeast Japan. Journal of Plant Research 1 12: 187-206.

Tan, F, Shuhua, S, Zhong, Y, Gong, X, & Wang, Y, 2002. Phylogenetic relationships of the
Combretoideae (Combretaceae) inferred from plastid, nuclear gene and spacer sequences.
Journal of Plant Research 115:475-481.

Tebbitt, MC, Lowe-Forrest, L, Santoriello, A, Clement, WL, & Swensen, SM, 2006.
Phylogenetic Relationships of Asian Begonia, with and emphasis on the evolution of rain-
ballist and animal dispersal mechanisms in sections Platycentrum, Sphenanthera and
Leptosae.










Tidwell, D, 1975. Common fossils plants of Westemn North America; Brigham Young University
Press.

Tiffney, B, 1990. The collection and study of dispersed angiosperm fruits and seeds. Palaios
5:499-519.

Tobe, H, Wagner, WL & Chin, HC, 1987. A systematic and evolutionary study of Oenothera
(Onagraceae): Seed Coat Anatomy. Botanical Gazette 148:235-257.

Vanvinckenroye, P, Cresens, E, Ronse Decraene, L-P & Smets, E, 1993. A comparative floral
development study in Pisonia, Bougainvillea and M~irabilis (Nyctaginaceae) with special
emphasis on the Gynoecium and floral nectaries. Bulletin du Jardin botanique de
Belgique/ Bulletin van de National Plantentuin van Belgie 62:69-96.

Vigilante, T, Dixon, K, Sieler, I, Roche, S & Tieu, A, 1998. Smoke germination in Australian
plants. Rural Industries Research and Development Corporation.

Vijayaraghavan, MR, 1969. Studies in the Family Cyrillaceae. I. Development of male and
female gametophytes in Cliftonia monophylla (Lam.) Britton ex Sarg. Bulletin of the
Torrey Botanical Club 4:484-489.

Wang, Y & Manchester, SR, 2000. Chaneya, a new genus of winged fruit from the Tertiary of
North America and Eastemn Asia. International Journal of Plant Sciences. 2000. 161:167-
178.

Weckerle, CS & Rutishauser, R, 2005. Gynoecium, fruit and seed structure of Paullinicae
(Sapindaceae). Botanical Joumnal of the Linnean Society 147:159-189.

Whitlock, BA, Bayer, C, & Baum, DA, 2001. Phylogenetic relationships and floral evolution of
the Byttnerioideae ("Sterculiaceae" or Malvaceae s.1.) Based on sequences of the
chloroplast gene ndhF. Systematic Botany 26:420-437.

Wilde, V & Frankenhauser, H, 1998. The Middle Eocene plant taphocoenosis from Eckfeld
(Eifel, Germany). Review of Palaeobotany and Palynology 101:7-24.

Wilkins, CF & Chappill, JA, 2002. Seed and seedling morphology and seed anatomy of
Lasiopetaleae (Malvaceae s.1. or Sterculiaceae). Australian Systematic Botany 15:545-
563.

Wilson, RC, 1974. Abronia: II. Anthocarp polymorphism and anatomy for nine species of
Abronia found in California. Aliso 8:113-128.

Wilson, RC, 1975. Abronia III. Pericarp and seed coat anatomy and its ecological implications
for Nine Species ofAbronia. Aliso 8:289-299.










Wilson, RC, 1976. Abronia: IV. Anthocarp dispersibility and its ecological implications for
nine species of Abronia. Aliso 8:493-506.

Wolfe, JA and Tanai, T, 1987. Systematics, phylogeny, and distribution ofAcer (Maples) in the
Cenozoic of Westemn North America. Journal of the Faculty of Science, Hokkaido
University, (Japan) series IV, 22: 1-246.

Wurdak. KJ, Hoffmann, P, Samuel, R, de Bruijn, A, van der Bank, M, & Chase, MW, 2004.
Molecular phylogenetic analysis of Phyllanthaceae (Phyllanthoideae pro pate,
Euphorbiaceae sensu lato) using plastid RBCL DNA sequences. American Journal of
Botany 91:1882-1900.

Zhuge, R, 1990. On the genus Burretiodendron Sensu Lato (Tiliaceae). Journal of the Amold
Arboretum 71:371-38.









BIOGRAPHICAL SKETCH

Elizabeth O'Leary was born in Pittsfield, Massachusetts, in 1981. She attended elementary

school through high school in the public school system of Natick, Massachusetts and graduated

at age 17 in the year 1999. She entered College of the Atlantic in the same year, maj oring in

human ecology. She graduated from College of the Atlantic in 2003; her final proj ect was an

easy to read book about plant reproduction. In 2004 and 2005, she took several courses through

Harvard Extension School and then applied for graduate school at the University of Florida. She

began graduate school in the spring of 2005 with a focus in paleobotany, working under the

supervision of Steven Manchester. She graduated with a Master of Science degree in 2007.





PAGE 1

1 TAXONOMIC DISTRIBUTION OF MODERN FI N-WINGED FRUITS AND THE FOSSIL HISTORY OF THE COMBRETACEAE IN THE UNITED STATES BASED ON FINWINGED FRUITS By ELIZABETH LEO OLEARY A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007

PAGE 2

2 Elizabeth Leo OLeary

PAGE 3

3 To Craig Greene

PAGE 4

4 ACKNOWLEDGMENTS I wish to thank my advisor, Steven R. Ma nchester and my committee members Walter S. Judd and David Dilcher for their useful comments and valuable time. I also wish to thank the museums and herbaria that allowed me to use thei r collections for my res earch, particularly those at Harvard, Berkeley, and the Smithsonian. I thank the Evolving Earth Foundation for funding this project. The support of my family and fr iends during this time wa s extensive, and much appreciated. Special thanks go to Kristine Hoffmann for the use of her apartment as a second office.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........6 LIST OF FIGURES................................................................................................................ .........7 ABSTRACT....................................................................................................................... ..............8 CHAPTER 1 INTRODUCTION................................................................................................................... ..9 Modern Fruits.................................................................................................................. .........9 Combretaceae................................................................................................................... ......11 2 MATERIALS AND METHODS...........................................................................................14 Modern Fruit Descriptions......................................................................................................14 Fossil and Modern Combretaceae..........................................................................................15 3 MODERN FRUIT SURVEY.................................................................................................34 4 RESULTS........................................................................................................................ .......62 Fossil Fin-Winged Fruits of North America With Possible Affinities to Combretaceae.......62 Carpolithes prangosoides Berry.............................................................................................62 Terminalia estamina MacGinitie........................................................................................65 Terminalia oregona (Lakhanpal) Meyer and Manchester......................................................67 5 CONCLUSIONS....................................................................................................................70 Modern Winged Fruits........................................................................................................... .70 Fossil Winged Fruits........................................................................................................... ....72 LIST OF REFERENCES............................................................................................................. ..74 BIOGRAPHICAL SKETCH.........................................................................................................81

PAGE 6

6 LIST OF TABLES Table page 2-1 Modern fin-winged fruits.................................................................................................. ......17

PAGE 7

7 LIST OF FIGURES Figure page 2-1 HernandiaceaeOnagraceae.............................................................................................20 2-2 Combretaceae (Combretum)..............................................................................................21 2-3 Combretaceae (Terminalia)...............................................................................................22 2-4 Combretace ae (other)....................................................................................................... ..23 2-5 Malvales.................................................................................................................. ...........24 2-6 ZygophyllaceaeBegoniaceae..........................................................................................25 2-7 RhamnaceaeTropaeolumaceae.......................................................................................26 2-8 Rutaceae Apiaceae........................................................................................................ .27 2-9 Distribution of fin winged fruits on Angiosperm phylogeny.............................................28 2-10 Terminalia vera ..................................................................................................................29 2-11 Carpolithes prangosoides ..................................................................................................30 2-12 Terminalia vera and Macropteranthus fitzalani ................................................................31 2-13 Terminalia estamina ..........................................................................................................32 2-14 Terminalia estamina ..........................................................................................................33 Authors note: All figures are or iginal photographs by th e author. The designations listed in the figures is the location of where the samples were obtained from.

PAGE 8

8 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science TAXONOMIC DISTRIBUTION OF MODERN FI N-WINGED FRUITS AND THE FOSSIL HISTORY OF THE COMBRETACEAE IN THE UNITED STATES BASED ON FINWINGED FRUITS By Elizabeth Leo OLeary December 2007 Chair: Steven Manchester Major: Botany Angiosperms have evolved wind dispersed fr uits and seeds many times. The wings of fruits may be formed from different structures including outgrowths of the ovary or hypanthium, perianth parts, and bracts or bracteoles. The wings may also form various morphological and biomechanical shapes. One such of these fruits is fin-winged that is, having a longitudinal central axis from which several l ongitudinally oriented wings radiat e. This form can be found in capsules, samaras, as well as in other fruit t ypes where the wings may be formed from accessory structures. This thesis examines the diversity of these fin-winged fru its at the generic level within modern Angiosperms and gives brief descripti ons of these fruits to aid in identification of dispersed fin-winged fruits. Thes e descriptions are applied to de termine if three fruit fossils, described as the genus Terminalia (Combretaceae), have sufficient ch aracters to be differentiated from other fin-winged genera, and if they truly belong in the Combretaceae. The descriptions are also be used to make inferences about the evol ution of these fruits. This survey indicated that fin-winged fruits are present among 23 families of angiosperm and are particularly well represented in the rosids.

PAGE 9

9 CHAPTER 1 INTRODUCTION Modern Fruits Winged fruits are found in many Tertiary fo ssil localities and are commonly preserved together with fossil leaves and other debris in lake and pond deposits (e.g. Tidwell 1975, Wilde and Frankenhuser 1997, Manchester 2001). The emphasis in many paleobotanical studies has been on the leaves, and the co-occurring winged fruits and seeds are often mentioned only in passing. Nevertheless, fruits and seeds are especi ally useful in taxonomy because they are part of the reproductive structures, which often have diagnostic feat ures in angiosperms (Tiffney 1990). Winged fruits also have many characters potentially useful fo r identification including size, number of wings, patterns of wing venation, wi ng shape and position, persistence of style(s) and pedicel, placentation type, seed number a nd orientation, position of micropyle and raphe and epidermal characters (Spjut 1994). Despite this, it is only in the past few decad es that fossilized winged fruits have been treated in more detail. Some examples of r ecent treatments are Acer (Wolfe and Tanai 1987), Fraxinus (Call and Dilcher 1992), Eucommia (C all and Dilcher 1997), Ceratopetalum (Barnes and Hill 1999), Dipteronia (McClain and Ma nchester 2001) and Ailanthus (Corbett and Manchester 2004). Recent treatments of extinct genera include Chaneya (Wang and Manchester 2000), Cedrelospermum (Manchester 1989), and Diplodipelta (Manchester and Donoghue 1995). In these instances, the fruits were more informative than leaves for the taxonomic determinations. Many additional kinds of winged fruits are known from the Tertiary of North America that have not been identified. The difficulty in id entifying their familial and generic affinities is largely due to the inaccessibility of compara tive data on modern genera with winged fruits

PAGE 10

10 (Tiffney 1990). An important rece nt contribution was published on asymmetrically winged fruits of the Americas (Mirle and Burnham 1999). It was accompanied by some line diagrams to aid in keying out unknown specimens, but did not includ e photographic images of the many different genera surveyed. There are at least 30 extant genera of winged fruits that are popular ly known and readily accessible in western North America, including Acer, Fraxinus, Ulmus, Carpinus, and Halesia (Schopmeyer 1977). If a fossil fruit does not matc h one of these familiar types, the comparative work for identification becomes a more dauntin g challenge because winged fruits occur in numerous genera scattered through at least 93 fa milies of flowering plants (Dallwitz et al. 2000 onwards). There is no single sour ce for comparative data on mode rn winged fruit genera and not even a comprehensive worldwid e list. If a fossil winged fr uit does not match a known living fruit type, it may be extinct, or it may correspond to an extant genus with which the investigator is unfamiliar. This lack of familiarity may occur more often when a genus has a fossil record across several continents, but has a limited distribu tion in modern times, as in the case of the extant Asian genus Craigia, whose fossil fruits in Europe and North America were formerly thought to represent an extinct genus Ptelea ecarpum Weyland (Buzek et al. 1989), until their common identity was discovered (Kva ek et al. 1991). Various morphological categorie s of winged fruits can be re cognized, such as those with an apical or basal whorl of propeller-like wings, those with a single surrounding wing, and they vary in aerodynamic efficiency and mode of fa lling (Augsburger 1988). In this treatment, I focused on fruits which have two or more wings radi ating from the vertical central axis as in the case of the anthocarps of the Nyctaginaceae (Gallo way 1975). These fruits were referred to as tumblers (Augsburber 1988) in relation to fli ght pattern, as finne d (Matlack 1987) and

PAGE 11

11 cyclically winged (Wurdak et al 2004). I use the term fin-winge d because tumblers refers to flight pattern and some of these fruits do not fly; finned could refer to ribs and does not fully convey that they are winged; and cyclically wing ed could refer to fruits with wings radiating laterally. In some cases the entire fruit is th e unit of dispersal and in others, the wings are on mericarps which fall off to disperse the seed. So me are capsular and dehisce entirely and others remain attached to the twig and dehisce their seed s as the fruit is shaken by the wind. To aid in systematic work, descriptions of many genera with the fin-winged morphology will be presented. These descriptions will allow comparisons to be made and will help determine if it is possible to distinguish between different families and genera based solely on the morphology of their fruit. The family Combretaceae is one in which fin-winged fruits occur in several genera with some genera more similar than others. This ma kes it an ideal family to study in detail. Combretaceae Some fossils of interest are those that have been assigned to the white mangrove family (Combretaceae) from the Eocene of the United St ates. Today this family is distributed pantropically, and can be found in the state of Florid a. Many of the reported fossils are leaves, but because the entire-margined, pinnately vein ed leaves of this family are of a common morphology encountered in many angiosperm families, additional features are needed to confirm their identity. The family has diagnostically us eful features such as combretaceous hairs and extrafloral nectaries on the leav es, but these have not been found on these impression fossils. The Combretaceae were revised in 1965 by the two main workers in this group, A.W Excell and C.A. Stace (Excell and Stace 1965). They recognized two subfamilies, the Strephonematoideae and the Combretoideae. Strephonematoideae contains the genus Strephonema which is distinct in its semi-inferior ova ry, two-armed hairs, paracytic stomata, and revolute domatia. The Combretoideae in turn have a fully inferior ovary and convolute or plicate

PAGE 12

12 cotyledons and the stomata are cyclocy tic or anomocytic. The fruits of Stephonema are drupaceous and do not have any specialized dispersal mechanism (Jongkind 1995), and are not considered here. The Combretoideae have been studied phylogene tically by Tan et al. (2002). The study included 18 taxa and was based on ITS, rbcL PY-IGS but included no morphological data. Strephonema was used as the outgroup and the results were similar to what would be expected from the traditional taxonomy. The s ubfamily split into the tribe containing Laguncularia and Lumnizera (Laguncularieae) and the tribe Combretaea e which further divided into two clades; Combretinae and Terminaliinae. Traditionally, th ere is a third subtribe the Pteleopsidinae, but this was not included in Tans an alysis. All of these groups cont ain species with winged fruits. The genera of Combretaceae that cont ain species with winged fruits are Calycopteris Combretum Pteleopsis Meiostemon Terminalia, and Thiloa Wings may be derived from ovary tissue, bracteoles or perianth (Excel and Stace 1965). There are three unique synapomor phies for the Combretaceae; anauxotelic inflorescences (inflorescence not ending in flower and growth not extending beyond inflorescence), combretaceous hairs (long, straight, sharply po inted, unicellular, very thick walled, with conical internal component at base), and a comb retaceous ovary (ovary in ferior and unilocular; ovules few, pendulous on elongate funiculi fr om the tip of locule ) (Johnson and Briggs 1984, Conti et al. 1997). Any of these features could be preserved in the right conditions, but none of the fossil fruit species considered here are co mmon and all are preserve d as either compression or impression fossils unattach ed to branches or leaves. Fossil fruits have also been attributed to this family. Three North American fossil species have been identified as belonging to th is family. Fossil fruits assigned to Terminalia vera are

PAGE 13

13 found in Eocene claypits in Tennessee and Kentuc ky and were first described by Edward Berry in 1930. This fruit was described as similar to T. brownii and T. darlingii Terminalia estamina was described by MacGinitie (1941) based on leav es and associated winged fruits from the Eocene Chalk Bluffs flora of the Central Sierra Ne vada, California, and was compared to extant T. hainensis and T. triflora Terminalia oregona was described from the Oligocene of Oregon by Meyer and Manchester (1997). It was compared to extant T. paniculata and to MacGinities fossil fruits of Terminalia estamina Characters for identifying combretaceous fruits have not been clearly presented before, nor have rigorous morphological comparisons been made between the winged fruits borne by different genera of the family. Most of the fruits in this family are dispersed abiotically either by water or wind, and fin-winged fr uits are also present in seve ral genera. Without detailed morphological information on the fruits of extant taxa, it is difficult to have confidence in the diagnosis of these fossiliferous fruits. MacGin itie (1941) reported th at the seed cavity of Combretum is typically longer relative to the wing than those of Terminalia but he did not make a full comparison or mention the status of the other genera. Study of the morphology of these fossils and co mparison with the modern genera of this and other families with fin-winged fruits will prov ide more characters to evaluate whether these fossils belong to the family, and if so, will give a strong record for the family in the United States during the Eocene.

PAGE 14

14 CHAPTER 2 MATERIALS AND METHODS Modern Fruit Descriptions A list was made of fruits known to have th e fin-winged fruit type from internet and literature searches that I made. The fruit morphology of genera in this list was verified when possible by examination of specimens in herb aria. The taxonomy was updated to the APG system (Angiosperm Phylogeny Group 2003). Extant fruits were examined and photographe d with a Nikon Coolpix 5400 digital camera at the Gray (GH) and Arnold Arboretum (A) He rbaria at Harvard University, Cambridge, MA, at the US National Museum at the Smithsonian Institution, Washington DC (US), and the United States National Seed Herbarium, Beltsville, MD (USNSH). Photographs taken of specimens at the Harvard Herbaria for which the source was unclear (A or GH) are provisionally marked HARVARD or HARVARD SEED COLLECTION (if they came from the seed collection). The lighting, both transmitted, and reflected, was adjusted to best show characters such as venation and trichomes. The photographs were then used to determin e important characters for these genera. Taxonomic literature (e.g. Goldberg 1986, Hutchinson 1964, and Judd 2006) was consulted to help interpret the photogr aphs. The fruits were characteri zed by number of wings, number of carpels, dehiscence, ovary position, overall sh ape, wing texture, venation, vein spacing, accessory parts (such as disk scar), and presence of a marginal vein. In some cases, the state was not observable in the photographs nor available in the literatur e. Parentheses are included around characters for which information is given from familial descriptions, when the details for the particular genus were not available. Fruit type is given when clear, but indehiscent fruits are often simply marked indehiscent to avoid conf using terms such as samara/samarium. Length

PAGE 15

15 and width are recorded for individual specimen s with slashes between different measured specimens or, in the case of some of the ge nera in Combretaceae are given as ranges. Fossil and Modern Combretaceae Comparisons were made with extant genera of all angiosperm families known to possess plants with the fin-winged fruit type. Speci mens were examined and photographed at the herbaria mentioned above. Fossil fruits were stud ied from collections of the Florida Museum of Natural History, Gainesville (UF), the Smithsonian Institution [United States National Museum], Washington, D.C., (USNM), the University of California, Museum of Paleontology, Berkeley (UCMP), and the Field Museum of Natural History, Chicago (FMNH). The Eocene fruits from western Tennessee a nd Kentucky were collected along with fossil leaves, fruits, flowers and pollen from clay p its in the middle Eocene Claiborne Formation of Tennessee and Kentucky. Large por tions of the flora, including the species considered here, were monographed by Berry (1916, 1930), but many of the determinations are still undergoing revision (Dilcher and Jones 1980, Jones and D ilcher 1980, Herendeen et al., 1990, Herendeen and Dilcher 1990). The age estimate is based on palynological correlations. A total of 25 specimens were examined, incl uding ten from Lamkin clay pit, one from Warman, five from Grable, ten from Puryear, tw o from Wilbank, and one each from Powers localities. A dissecting microscope was used to examine details of the fruits, and sediment was chipped with needles to expose details of the wings. Each specimen was photographed with a Nikon Coolpix 955 digital camera. The fossils were initially exposed by splitting the clay with chis els, revealing the fruits in longitudinal section. Because the plane of fract ure usually exposes only two of the wings, more detailed preparation was necessary to demonstrate the full number of wings. To reveal the complete complement of wings, both counterpart halves of the same specimen were reassembled

PAGE 16

16 and then fractured transversely. To control the orientation of the transverse fracture, we used a dry circular diamond saw to pre-cut the surrounding clay in the desired plane. Wing venation was studied under magnification and a camera lucida was used to record details of venation. Line drawings were used to determine if each wing was a composite of two fused laminae or if it was a single structure. Venation patterns of part and counterpart were traced and overlaid to determine if the venation was the same on each side. The Eocene specimens from California, from th e Chalk Bluffs flora of the Sierra Nevada Mountains were previously described by M acGinitie (1941) under the name Terminalia estamina. Specimens were studied from the UCMP and USNM collections The Oligocene specimens from Oregon were previously treated by Lakhanpal (1958), Brown (1959) and Meyer and Manche ster (1997). We studied the original specimens deposited at UF, USNM, and UCMP. The fossils were then compared to the modern genera of fruit with fin wings and genera with similar features were compared to dete rmine whether the fossils could be confidently identified to a particular taxonomic group.

PAGE 17

17 Table 2-1. Modern fin-winged fruits Family Genus wing num. carpel num. dispersal mode ovary position marginal vein Persistance of parts Longer/ Wider/ (roughly) equal (Q) Achariaceae Carpotroche 620+ 4-8 fruit Sup. No disk Q Apiaceae Prangos 5 5 finned mericarp Inf. No Perianth L Apiaceae Steganotaenia 2 5 finned mericarp Inf No Perianth L Begoniaceae Begonia 3 3/5 anemoballisty Inf Yes Stamenal disk L/W/Q Burmanniaceae Burmannia 3 3 ? Inf ? ? ? Cardiopteridaceae Cardiopteris 2 2 fruit Sup. Yes Style L Celastraceae Stackhousia 3 3--5 mericarp Sup. Yes Style, perianth L Celastraceae Tripterigium 3 3-loc fruit Sup. Yes Style, Perianth L Celastraceae Wimmeria 2-3(4) 2-3-loc fruit Sup. Yes L/W Combretaceae Combretum 2-5 4--5 fruit Inf. No (style) L/W/Q Combretaceae Macropteranthus 4 2 fruit Inf. No Perianth L/W Combretaceae Meiostemon 2 fruit Inf. No apical protrusion L Combretaceae Pteleopsis 2 2 fruit Inf. No No L Combretaceae Thiloa 4 2 fruit Inf. No No Q Combretaceae Terminalia 2,4,5 4--5 fruit Inf. No No L/W/Q Cunoniaceae Gillbeea 3 3 ? Sup. Yes No L Cyrillaceae Cliftonia 1--2 3-4 locules fruit Sup. ? Perianth Q Dioscoreaceae Dioscorea 3 3 anemoballisty Inf. suture Perianth L/W Fabaceae Fissicalyx 2 1 fruit Sup. No No L Hernandiaceae Illigera 2 1 fruit Inf. No No W Lophopixidaceae Lophopyxis 5 5 fruit Sup. No Style, perianth L Lauranthaceae Nuyitsia 3 3--4 fruit I No No L Malvaceae Cavanillesia 4 5 fruit S No apical protrusion W Malvaceae Abroma 5 5 capsule S suture Perianth W Malvaceae Kleinhovia 5 5 caps ule S suture Perianth W Malvaceae Pentace 5 2--10 ? S No Perianth L Malvaceae Pterostyrax 5 5 ? Sup. No P L

PAGE 18

18 Table 2-1. Continued Family Genus wing num. carpel num. dispersal mode ovary position marginal vein Persistance of parts Longer/ Wider/ (roughly) equal (Q) Malvaceae Berrya 5 5 capsule Sup. No ? W Malvaceae Burretiodendron 5 5 capsule Sup. Yes, suture S L Malvaceae Colona 3/4/5 ? capsule Sup. No S W Malvaceae Craigia 5 3/4 capsule Sup. Yes No L Malpighiaceae Tetrapterys ? ? ? ? ? ? ? Malpighiaceae Aspidopterys 3 3 fruit Sup. No No L Nyctaginaceae Phaeoptilum 4 1 fruit Sup. No No L Onagraceae Megapterium 4 4 anemoballisty Inf. suture No Q Onagraceae Oenothera 4 4 anemoballisty Inf. suture No L/Q Pedaliaceae Holubia 4 2 fruit Sup. Yes No Q Pedaliaceae Pterodiscus 4 (2-4) fruit Sup. No Perianth Q Phyllanthaceae Hymenocardia 2 3 fruit Sup. Yes No L Polygalaceae Polygala (2-3) fruit Sup. No apical protrusion L Polygonaceae Atraphaxis 3 (2-3) fruit Sup. No Perianth W Polygonaceae Rheum 3 (2-3) fruit Sup. No No L Polygonaceae Rumex 3 (2-3) fruit Sup. No No L/W Polygonaceae Polygonum 3 (2-3) fruit Sup. No No L Polygonaceae Antigonon 3 (2-3) fruit Sup. No No Q Rhamnaceae Gouania 3 3 capsule Sup. I. marginal apical disk W Rhamnaceae Chaydia 3 3 capsule Sup. I. marginal No L Rhamnaceae Reissikia 3 3 mericarp Sup. I. marginal No W Rutaceae Balfourodendron 4 4--5 fruit Sup. No No W Rutaceae Ptelea 2--3 2-3 loc No Sup. No No L/Q Sapindaceae Dodonaea 3 ?3 ? Sup. ? ? ? Sapindaceae Urvillea 2 (2-3) caps ule Sup. Yes basal disk L Sapindaceae Majidea 2? (2-3) caps ule Sup. Yes basal disk ? Sapindaceae Stocksia ? (2-3) caps ule Sup. Yes basal disk ? Sapindaceae Thouindium 3 (2-3) c apsule Sup. Yes basal disk L

PAGE 19

19 Table 2-1. Continued Family Genus wing num. carpel num. dispersal mode ovary position marginal vein Persistance of parts Longer/ Wider/ (roughly) equal (Q) Sapindaceae Toulicia 3 (2-3) c apsule Sup. Yes basal disk ? Styraceae Halesia 2 or 4 2 fruit Inf. Yes Perianth L Trigoniaceae Humbertiodendron 3 3 fruit Sup. Yes No Q Tropeaeoceae Tropaeolum 3 3 (1 locule) mericarp Sup. Yes ? L Zygophyllaceae Bulnesia 5 5 mericarp Sup. Yes ? ? Zygophyllaceae Guiacum ? 5 Yes Sup. ? ? ? Zygophyllaceae Morkillia 4 (2-5) capsule Sup. No? L Zygophyllaceae Porleria 3 (2-5) mericarp Sup. ? ? ? Zygophyllaceae Zygophyllum (2-5) mericarp Sup. ? ? ?

PAGE 20

20 Figure 2-1. HernandiaceaeOnagraceae. A) Illigera grandiflora (Heng 11446 A). B) Dioscorea (McPherson 43 FLAS). C) Dioscorea (McPherson 43 FLAS). D) Phaeoptilum. spirosum (R. Srydel 199 USNSH). E) Rumex venosus (77, USNSH). F) Oenothera missouriensis (Stevens 561 GH). G) Megapterium missouriense (L.E Wham USNSH).

PAGE 21

21 Figure 2-2. Combretaceae (Combretum). A) Combretum apiculatum (SPI 28342 USNSH). B) Combretum gazense (PIG 13469 USNSH). C) Combretum fruticosum (PI 104111 ,USNSH). D) Combretum rhodesicum (SPI 48244 Harvard). E) Combretum grandiflorum (SPI 72993 USNSH). F) Combretum primigenum (SPI 61642 USNSH). G ) Combretum paniculatum (PI s.n., USNSH). H) Combretum smeathmanii (PI 73930 USNSH). I) Combretum imberbe (SPI 48243, USNSH).

PAGE 22

22 Figure 2-3. Combretaceae (Terminalia). A) Terminalia argentea (PI 300640 USNSH). B) Terminalia superba (Harris and Fay 781 Harvard). C) Terminalia macroptera (221-9 Harvard). D) Terminalia tomentosa (Hara et al. 630075 A). E) Terminalia tomentosa (Butterwick 2 A). F) Terminalia virens (Maguire 30764 USNSH). G) Terminalia sericera (Parker 1428 Harvard). H) Terminalia amazonia (Steinbach 6617 A). I) Terminalia browneii (Harvard from Flora Uganda). J) Terminalia pedicellata (Maxwell 01-461 A)

PAGE 23

23 Figure 2-4. Combret aceae (other). A) Meiostemon tetrardra (Grosuenor sn1966 USNSH). B) Pteleopsis hylodendron (Claesaers 401 USNSH). C) Pteleopsis sp. (Gossweiler 13550 USNSH). D) Guiera sp (Bamps 2477 USNSH). E) Macropteranthus fitzalani (Francis SN 1920 USNSH). F) Macropteranthus kekwickii (N.Byrnes 1592 USNSH). G) Thiloa sp. (photo of a photograph from Types from Munich Herbarium).

PAGE 24

24 Figure 2-5. Malvales. A) Cavanellisia platanifolia (Foster 2247 GH). B) Abroma angusta (Wang 41264 A). C) Kleinhovia hospita (Huang et al 16132 A). D) Maxwellia lepidota (Hurliman 13129 A) ). E) Pterostyrax corymbosa (Mie Min-Xiang 92250 A). F) Halesia diptera (Chester 1124 A). G) Berrya javanica (KK and SS 11 A). H) Colona sp (Suejarto 59 A). I) Burretiodendron esquirolli (Malelen et al. 1160 A). J) Cragia yunnanensis (1124, UF). K) Pentace laxiflora (San 78602 A).

PAGE 25

25 Figure 2-6. ZygophyllaceaeBegoniaceae. A) Bulnesia chilensis (Weidermaum 429 GH). B) Bulnesia arborea (1069 UF). C) Zygophyllum morgansa (Marloth 12275 A). D) Porleria angustifolia (Moore and Steyermark 3288 Harvard). E) Wimmeria mexicana (Clevinger July 1999 GH). F) Wimmeria concolor (Goldman 227 GH). G) Tripterigium regelii (USNSH Togashi 531 ). H) Gillbeea papuana (Brass 31815 A). I) Gillbeea papuana (Takeuchi 6084 A). J) Begonia sp. (Kosterman 6022 A). K) Begonia nelubifolia (Poropus 8543 GH).

PAGE 26

26 Figure 2-7. RhamnaceaeTropaeolumaceae. A) Gouania domingensis (Rodriguez 2566 A). B) Chaydaia (Huder 601 USNSH). C) Reisseika (L.O. Williams 6834 USNSH). D) Gouania lupuloides (Brumbach 9537 GH). E) Carpotroche amazonica (No. 9001 USNSH). F) Humbertodendron (Capman 180535F USNSH). G) Lophopyxis (Buderus NGF 23932 USNSH). Tropaeolum (Donat 51 USNSH).

PAGE 27

27 Figure 2-8. Rutaceae Apiaceae. A) Balfourodendron riedelanum (Fiebrig 6104 GH). B) Ptelea trifolata (Harvard Seed Collection). C) Ptelea crenulata (Harvard Seed Collection). D) Urvillea andersonii (Anderson 9150 NYBG). E) Cliftonia monophylla (Rhoades 5a203 GH). F) Holubia saccata (AOD Mogg 24419 USNSH) G) Pterodiscus sp. (Dinker 4935 A). H) Cardiopteris moluccanum (Wang 81070 A). I) Steganotaenia araliocea (Meyer 7962 USNSH). J) Prangos odotoptera (Reckingerf 997 USNSH).

PAGE 28

28 Figure 2-9. Distribution of fin winged fruits on Angiosperm phylogeny. Each star represents a family that contains at least one species with fin winged fruits. Topol ogy taken from (Stevens, 2001 onwards).

PAGE 29

29 Figure 2-10. Terminalia vera A) (FLMNH 15815-48329). B) (FLMNH 15815-48329). C) (FLMNH 15815-48326). D) (UF 15875-48329) E) (UF 15875-48329). F) (UF 1888432723) G. (UF 18884-32723).

PAGE 30

30 Figure 2-11. Carpolithes prangosoides A) (USNM 36005). B) (FLMNH 15815-48328). C.) (FLMNH 15820-48318). D) (FLMNH 18884-32781) E) (FLMNH 18884-32781). F) (UF15820-48332) G) (FLMNH 18952-46057). H) (FLMNH15826-48331). I) (FLMNH). J) (FLMNH). K) (FLMNH 15815-48325).

PAGE 31

31 Figure 2-12. Terminalia vera and Macropteranthus fitzalani A) (USNM). B) (USNSH Francis SN 1920).

PAGE 32

32 Figure 2-13. Terminalia estamina A) (UCMP 2321). B) (USN M 313946). C) (UCMP 2321). D) (UCMP 2321).

PAGE 33

33 Figure 2-14. Terminalia estamina A) Willemette flora (USNM42334). B) Fossil Oregon (UF 15842-7032). C) Lyons flora (USNM 42353). D) (UF 250-10497). E) (UF 250-10497). F) (UF 250-10497).

PAGE 34

34 CHAPTER 3 MODERN FRUIT SURVEY As a result of this survey, I became aware of at least 87 genera in which radially finned fruits are present. Each genus is described be low (arranged alphabetically by family). These data serve as the basis for the compar ative analyses presented in Chapter 5. Achariaceae Carpotroche (Figure 2-7E) Species examined: C arpotroche amazonica Mart. (No. 9001 USNSH) Type of fruit: Indehiscent Length: 7.8 cm Width: 7.1 cm Shape in outline: Globose Ovary condition: Superior Persistence of styles/perianth: Disk on basal surface Persistence of pedicel/stipe: No Carpels: 4-8 Number of wings: 6 to many (~20); often dissected Margins: Entire/undulate Wing derivation: Ovary Wing thickness: Coriaceous Trichomes: None observed Venation: Wing was too opa que to observe venation Distribution: Tropical America Apiaceae Anatomy and morphology of fin-winged fruit genera of Apiaceae and Araliacaeae ( Annesorhiza, Asteriscium, Astrotricha, Chorita enia, Dasispermum, Elaeoselinum, Heptaptera, Hermas, Heteromorpha, Laretia, Molopospermum, Myodocarpus, Pachypleurum, Peucedanum, Polemanniopsis, Polylophium, Rouya, and Tordylium) has been recently studied in detail (Liu et al. 2006). They found that fruit characters in cluding developmental origin of the wings, the shape of the carpel, presence of vittae, and other anatomical featur es to correspond with molecular cladograms, which suggest s that in this group at least, fruits have great taxonomic value. Described below are two ge nera which were not included. Prangos is obviously winged

PAGE 35

35 but the species of Steganotaenia I observed app eared only slightly finne d although they can be winged (Hyde and Wursten 2002-7). Prango s (Figure 2-8J) Species : Prangos odontotlera Boiss. (K.H Beckinguif 997 USNSH) Type of fruit: Schizocarp Length: 2 cm. (+0.5 pedicel) Width: 1 cm Shape in outline: Elliptical Ovary condition: Inferior Persistence of styles/perianth: No Persistence of pedicel/stipe: Pedicel Carpels: 5 Number of wings: 5 Margins: Entire Wing derivation: From ribs on ovary Wing thickness: Papery Trichomes: None observed Venation: None apparent Steganotaenia (Figure 2-8I) Species examined: S teganotaenia araliacea Hochst. (F.G. Meyers 7962 USNSH) Type of fruit: Schizocarp Length: 1.6 cm. (+.4 pedicel) Width: 1 cm Shape in outline: Elliptical Ovary condition: Inferior Persistence of styles/perianth: No Persistence of pedicel/stipe: Pedicel Carpels: 5 Number of wings: This species is more ridged than winged, but other species prominently 2 winged Margins: Entire Wing derivation: Ovary Wing thickness: Coriaceaous Trichomes: No Venation: N/A Begoniaceae Begoniaceae have 2 genera with the prominent winged ovaries Begonia and Hillebrandia Begonias have many stamens, and there is a disk at the apex of the fruit where these stamens, along with petals and stigma we re attached. Begonias may be wind dispersed, animal dispersed, or rain disper sed. Different fruit morphologies are associated with each of

PAGE 36

36 these different dispersal syndromes. Wind disper sed begonias tend to have three locules with three equal or close to equal sized wings while rain dispersed begonias have one large wing and two smaller wings. Ecologically, wind disperse d begonias are more often found in open forest habitat while rain or animal dispersed begonias ar e found in humid closed forest habitats (Tebbitt et al. 2006). Begonia (Figure 2-6JK) Species examined: B egonia sp (Hoover et al. 988 A), Begonia nelumbiifolia Cham. & Schltdl. (Porpus 8543, GH ), Begonia sp. (Kosterman 6022 A) Type of fruit: Dehiscent/Indehiscent. capsule splitting along sutures and releasing seeds via anemoballisty Length: Variable Width: Variable longer than wide or wider than long Shape in outline: Variable Ovary condition: Inferior Persistence of styles/peria nth: Yesstaminal disk Persistence of pedi cel/stipe: Yes stipe Carpels: 3-5 Symmetry: Often with one locule and wing larger than the others Number of wings: 3 Margins: Entire Wing derivation: Ovary outgrowth Wing thickness: Membranous Trichomes: No Venation : Transverse; no marginal vein; some looping; areoles transverse elongate Burmaniaceae are closely related to Diosco rea and some have similar fruits. Three winged with minute seeds. The three carpels are sometimes basally affixed. Cardiopteridacerae Cardiopteris (Figure 2-8H) Species examined: C ardiopteris moluccana Blume (Wang 81070 A) Type of fruit: Indehiscent Length: 3.5 cm Width: 2.8 cm Shape in outline: Transversely Elliptical Ovary condition: Superior Persistence of styles/perianth: One large stylar protrusion and calyx present at base Persistence of pedicel/stipe: maybe very small stipe Carpels/Locules: 2 carpels, 1 locule Number of wings: 2 Margins: Entire, smooth

PAGE 37

37 Wing derivation: Ovary Wing thickness: Chartaceous Trichomes: No Venation: Marginal vein only Celastraceae Stackhousia. These herbs are found in Australia/Mal esia (Mabberly 1989). The plants accumulate nickel in various orga ns including the fruits (Bhatia et al. 2003). Their fruits are Schizocarpic and readily identifiable by their wi ngs with radiating veins completely free from interconnections and dichotomizatio ns and with a marginal vein. Interestingly, the germination of the seeds is increased in the pres ence of smoke (Vigilante et al. 1998). Species examined: S tackhousia brunonis Benth. (A. Travers 19, A) Type of fruit: Schizocarp Length: 2.4 cm Width: 1.7 cm Shape in outline: Ovate Apex: Acuminate Base: Cordate Ovary condition: Superior Persistence of styles/perianth: Caly x at base and ap ical protrusions Persistence of pedicel/stipe: Very small stipe Carpels: 3-5 Number of wings: 3 Margins: Entire with marginal vein Wing derivation: Pericarp Wing thickness: Chartaceous Trichomes: Yes, unbranched Venation: Radiating out and ar cing slightly upward; thicker at central body and thinning towards margin; joining marginal vein. Tripterygium (Figure 2-6G). Tripterygium is a liana or scandent shrub which is used in herbal medicine. The ovules are epitropous. Species examined: T ripterigium regelii Sprague & Takeda (Togashi 531 A); Tripterigium wilfordii Hook.f. (Cavaderic 3316 A) Type of fruit: Indehiscent Length: 1.2 cm Width: 1.1 cm Shape in outline: Longitudinally Elliptical Apex: Rounded/cordate Base: Lobate

PAGE 38

38 Ovary condition: Superior Persistence of styles/perianth: Apical protrusion and floral remnants at base Persistence of pedicel/stipe: Pedicel Carpels: 3 locules Number of wings: 3 Margins: Entire (marginal vein) Wing derivation: Ovary? Wing thickness: Chartaceous Trichomes: No Venation: Arching upwards subparrallel and only occasionally dichotomizing; central body with longitudinal veins and some tr ansverse crossveins (marginal vein) Distribution: East China to Taiwan Wimmeria (Figure 2-6 E F) Species examined: W immeria concolor Cham. & Schltdl. (Goldman 227 GH), Wimmeria confusa Hemsl. (Lundell 1938 GH), Wimmeria mexicana (DC.) Lundell (Clevinger July 1999 GH) Type of fruit: Indehiscent samara Length: 1.4/ 2 cm Length: 1.8/ 1.6 cm Shape in outline: Elliptical Ovary condition: Superior Persistence of styles/perianth: St yles at apex, perianth at base Persistence of pedicel/stipe: Pedicel Carpels: 2-3 locules Number of wings: Usually 3, but may be 2 or 4 (Lundel, 1938) Margins: Entire Wing derivation: Ovary Wing thickness: Chartaceous Trichomes: No Venation: Radiating subparallel dichotomizing and anastomosing Fruit may have silvery cottonlik e filaments when breaking open Combretaceae Combretaceae are a family with fruits that are water dispersed, animal dispersed, and wind dispersed. Wind dispersal occurs in many cl ades within the family. This family is discussed in greater detail in Chapter 3. Combretum (Figure 2-2) Species examined: C ombretum apiculatum Sond. (SPI 28342 USNSH); C. erythrophyllum Sond. (PI112639 USNSH); C. farinosum H.B. & K. (FPI 81266 USNSH); C. fruticosum Stuntz (PI 104111 USNSH); C. gazense Swynn. & Baker f. (PIG 13469 USNSH); C. grandiflorum G. Don (SPI 72993 USNSH); C. imberbe

PAGE 39

39 Wawra(SPI 48243 USNSH); C. paniculatum Vent. (PI s.n. USNSH): C primigenum Marloth. (SPI 48244, USNSH); C. smeathmannii G.Don (PI 73930 USNSH) Type of fruit: Indehiscent. diclesium (Spjut, 1994) Length: 1-5 cm Width: 1-4 cm Shape in outline: Variable from widely elliptical, ovate, to narrowly elliptical Ovary condition: Inferior Persistence of styles/perianth: So metimes stylar protusion at apex Persistence of pedicel/s tipe: Usually a pedicel Carpels: 4-5 (one locule) Number of wings: 4-5 Margins: Usually tattered Wing derivation: Perianth Wing thickness: Chartaceous Trichomes: Not observed Venation: Generally ra diating subparallel Macropteranthes (Figure 2-4 E F) Species examined: M. fitzalani F. Muell (SN 1960 USNSH); M kekwickii Benth. (W. Byrnes 1592 USNSH) Type of fruit: Indehiscent Length: 1 cm/ 2 cm+ Width: 1.5 cm/ 1.5 cm Shape in outline: Wide elliptical. to elongate obovate Ovary condition: Inferior Persistence of styles/peria nth: Calyx lobes at apex Persistence of pedicel/sti pe: Pedicel on fruits of M. fitzalani Carpels: 2 but pseudomonomerous Number of wings: 4 each (composed of 2 lamina) Margins: Entire Wing derivation: Bracteoles Wing thickness: Chartaceous Trichomes: Not observed but presumably there may be combretaceous hairs. Venation: Primaries looping or not with secondaries interconn ecting and looping at margins (or not), and tertiaries when pres ent interconnecting or forming blind ended branched veinlets Pteleopsis (Figure 2-4 B C) Species examined: P. hylodendron Mildber. (J. Classens 401 J, USNSH) P. sp (Gossweiler 13530 USNSH). Type of fruit: Indehiscent Length: 1.5-2.5 cm Width: 0.5-1.5 cm Shape in outline: Elliptical., sometimes with vshaped notch at apex or stylar protrusion, base truncate Ovary condition: Inferior Persistence of styles/perianth: Sometimes stylar protrusion Persistence of pedicel/stipe: Pedicel

PAGE 40

40 Carpels: 2, but appearing as one Number of wings: 2 Margins: Entire Wing derivation: Perianth Wing thickness: Chartaceous Trichomes: Not observed but presumably there may be combretaceous hairs Venation: Radiating subparallel Thiloa (Figure 2-4 G) Species examined: T hiloa glaucocarpa (Mart.) Eich. Image was observed at Harvard from Types of the Munich Herbarium Type of fruit: Indehiscent Length: ? Width: ? Shape in outline: Rounded Ovary condition: Inferior Persistence of styles/perianth: No Persistence of pedicel/stipe: Pedicel Carpels: 2, but appearing as one Number of wings: 4 Margins: Entire Wing derivation: Perianth Wing thickness: Chartaceous Trichomes: Not observed but presumably there may be combretaceous hairs Venation: Not observed Terminalia (Figure 2-3) Species examined: T argentea Mart. (PI 300640 USNSH); T. virens (Spruce ex Eicheler) Alwen & Stace (B. Maguire et al. 30764 USNSH); T. alata Herb.Madr. ex Wall. (Maxwell 92-19, A); T. amazonica Excell (Steinbach 6617 A); T. argentea Mart. (Mello et al. 608 GH); T. glaucescens Planch. Ex Benth. (Bretleler 248, A); T. macroptera Mart. (221-9 Harvard); T. pedicellata W.Nanakorn (Maxwell, 01-461, A); T. pyrifolia Kurz. (Burma Forest School Herbarium 61 A); T. tomentosa Mart. Ex Eichl. (Hara 100014 A). Type of fruit: Indehiscent diclesium (Spjut, 1994) Length: 0.57 cm Width: 16cm Shape in outline: Variable from elongate to elliptical to very wide elliptical Ovary condition: Inferior Persistence of styles/per ianth: Stylar protrusi on sometimes; diclesium. Persistence of pedicel/stipe: Sometimes stipe or pedicel Carpels: 2 but appearing as one Number of wings: 2, 4, 5 Margins: Entire, may be tattered or reinfor ced by increased dichotomizations near margin Wing derivation: Perianth Wing thickness: Chartaceous to coriaceous Trichomes: Combretaceous hairs present or absent

PAGE 41

41 Venation: Usually radiating subparrallel but may dichotomize and anastomose either occatinally or as a reinforcement to the margin Cunonaceae Gillbeea (Figure 2-6 H I) Gillbeea is found in New Guinea and Queensland and has a three winged one seeded fruit. Anatomical information for the flow ers of this genus can be found in Matthews et al. (2001). The broken pieces had two di stinctive veins from the central body to the apex. Species examined: G illbeea papuana Schltr. (Takeuchi 6084 A) (Brass 31815 A) Type of fruit: Described as Indehiscent., but I found pieces broken apart Length: 1.6 cm Width: 1 cm Shape in outline: Longitudinally Elliptical Ovary condition: Superior Persistence of styles/perianth: No Persistence of pedicel/stipe: Pedicel Carpels: 3 Number of wings: 3 Margins: Wavy/entire with marginal vein Wing derivation: Ovary wall outgrowths Wing texture: Chartaceous Trichomes: Yes Venation: Radiating reticulum with midveins running apically from central axis and a marginal vein present Cyrillaceae Cliftonia (Figure 2-8E) Species examined: C liftonia monophylla Britton (Rhoades 5a203 GH). Type of fruit: Indehiscent Length: 0.5 cm Width: 0.5 cm Shape in outline: Longitudinally Elliptical Apex: Emarginate Base: Rounded Ovary condition: Superior Persistence of styles/per ianth: Persistent sepa ls at base of fruit Persistence of pedicel/stipe: Pedicel? Carpels: 3-4 Number of wings: 2-3 Margins: Entire Wing derivation: Ovary Wing thickness: Coriaceous Trichomes: No

PAGE 42

42 Venation: Unknown Dioscoreaceae Dioscorea (Figure 2-1 B C) Species examined: Dioscorea alata L. (McPherson 43 FLAS) Type of fruit: Apically opening capsule (dispersing via anemoballisty) Length: Variable 4.3 cm Width: Variable 4.3 cm Shape in outline: Variable from longer th an wide to round to wider than long Ovary condition: Inferior Persistence of styles/per ianth: Sometimes perianth persists at apex Persistence of pedicel/stipe: Small pedicel Carpels: 3 Number of wings: 3 Margins: Thick marginal suture Wing derivation: Ovary Wing thickness: Chartaceous Trichomes: No Venation: Striations radiati ng outward to true venation Seeds: Often with winged seeds Fabaceae There are several genera in the Fabaceae that ha ve winged fruits. Of these, at least three are known to have the fin-winged morphology. Much has been published about the fruits of this family (Kirkbride et al. 2003; Stirton and Zarucchi 1989, Gunn 1984) including the morphology and aerodynamics of wind dispersed legumes. At least four genera have fin-winged fruit ( Fissicalyx, Piscidia, Sesbania and Tetracarpon ). Piscidia is found from Florida to Venezuala and was figured in Flora of Pana ma (Dwyer 1965). The fruits are pedicellate and stipitate, four winged, both apex and base are tr uncate, with pinnate venation. Fissicalyx is described below. Fissicalyx. Fissicalyx is a member of the Dalbergi oid legumes, a group which has at least 16 genera with wings (Lav in et al. 2001). The wings of Fissicalyx are derived from expansion of the ovary wall, but in other winged genera the wings are formed from expansion of ovary sutures or from the stipe. Th is genus is figured in (Dwyer 1965).

PAGE 43

43 Species examined: Image of Fissicalyx taken from USNSH website Type of fruit: Indehiscent pod Length: 6 cm Width: 5.3 cm Shape in outline: Elliptical with an acuminate apex Ovary condition: Superior Persistence of styles/perianth: Style at apex Persistence of pedicel/stipe: No Carpels: 1 Number of wings: 2 Margins: Entire Wing derivation: Expansion of ovary wall Wing thickness: Chartaceous? Trichomes: No Venation: Radiating Hernandiaceae Illigera (Figure 2-1 A) Species examined : Illigera celibica Miq. (How 73378 A), Illigera cordata Dunn (Delaway 4 Sept. 1883 A), Illigera grandiflora W.W.Sm. & Jeffrey (Heng 11446 A) (Heng 10346 A) Fruit type: Indehiscent., code d as a drupe (Renner, 1999) Length: 3.6/3.6 cm Width: 6.9/3.7 cm Shape in outline: Latera lly elliptical or rounded Apex: Cordate Base: Cordate/ cuneate Ovary condition: Inferior Persistence of styles/periant h: Perhaps sometimes style Persistence of pedicel/stipe: No Carpels: One Symmetry: 2 wings larger than third Number of wings: 2 or 3 Margins: Entire slightly wavy Wing derivation: Ovary Wing thickness: Chartaceous Trichomes: No Venation: Veins radiating to margin very sinuous all of a single order Lophopyxidaceae Lophopyxis (Figure 2-7 G). Lophopyxi s is the only genus in the Lophopyxidaceae, it is sometimes placed in Celastraceae but has characters such as tomentose ovaries which are different from the Celastraceae (Simmons et al. 1999)

PAGE 44

44 Species examined: L ophopyxis combretocarpa Engl. Ex Pax (Abrahia 3405 USNSH) Lophopyxis maingayi Hook.f. (Schode 5063 USNSH) Type of fruit: Indehiscent samara Length: 3 cm Width: 1.5 cm Shape in outline: Elongate obovate Ovary condition: Superior. Persistence of styles/perianth: Stylar remnants at apex and perianth remnants at base Persistence of pedicel/stipe: Small stipe Carpels: 5 Number of wings: 5 Margins: Entire Wing derivation: Ovary Wing thickness: Chartaceous Trichomes: Densely hairy, hairs unbranched Venation: Downward arching subparrallel Loranthaceae Nuyitsia. Loranthaceae are a family known for being parasitic on tree branches. Fruits of this genus are usually fleshy. Nuytsia is a tree or shrub and ha s three-winged wind dispersed fruits. Lamont (1985) studied the dispersal of these fruits and found a maximum dispersal capability of 50 meters. Nevertheless, he conclu ded that this supports evidence that this genus reproduces mainly by suckering. Species examined: N uytsia sp. (USNSH) Type of fruit: Indehiscent Length: 1 cm Width: 0.8cm Shape in outline: Elliptical Ovary condition: Inferior. Persistence of styles/perianth: No Persistence of pedicel/stipe: Pedicel Carpels: 3-4 Number of wings: 3 Margins: Entire Wing derivation: Ovary Wing thickness: Chartaceous Trichomes: No Venation: Transverse veins occasionally dic hotomizing, arising from midline of fruit

PAGE 45

45 Malvaceae Cavanillesia (Figure 2-5A). The fruits of Cavanillesia are the largest fin-winged fruits at over 12 cm. long. Garwood (1985) studied the role of mucilage in s eedling development of Cavanillesia This genus produces large amounts of muc ilage in the fruits, and the mucilage is present during dispersal. Species examined: C avanillesia platanifolia H.B. & K (R. Foster 2247 GH) Type of fruit: Indehiscent capsule Length: 12 cm Width: 14 cm Shape in outline: Widely elliptical Apex: Cordate Base: Cordate Ovary condition: Superior Persistence of styles/peria nth: Apical protuberance Persistence of pedicel/stipe: No Carpels: 5 (one or two seeded fruits) Number of wings: 4 Margins: Tattered Wing derivation: Ovary Wing thickness: Membranous Trichomes: No Venation: At first glance, veins appear to be subparrallel and of a si ngle order, but closer examination reveals a reticulum of lower order veins between the primaries which form areoles of 1mm or less, the veins are very fibrous and remain as a skeleton after the wing surface is lost. Abroma (Figure 2-5 B) Species examined: A broma augusta L.f. (C.Wang 41264 A) (A.D.E Elmer 21446 A) Type of fruit: Loculicidal capsule Length: 4-5cm. Width 5-6 cm Shape in outline: Obovate Apex: Truncate Base: Rounded Ovary condition: Superior Persistence of styles/perianth: Calyx Persistence of pedicel/stipe: No Locules: 5 Number of wings: 5 Margins: Thick suture Wing derivation: Ovary Wing thickness: Chartaceous

PAGE 46

46 Trichomes: Tri-branched hairs. Venation: Veins thick, ascending to tips, di chotomising and anastomosing occasionally forming transversely long areoles. Kleinhovia (Figure 2-5C). Kleinhovia is a vine or a tree and grows in Asia and Australia. Hjerpe and Hedenas (2001) found that Kleinhovia hospita is a species that thrives in the disturbed habitat after a cyclone. Species examined: K leinhovia hospita L. (Huang et al. 16132 A) Type of fruit: Inflated capsule Length: 1.4 cm Width: 2.3 cm Shape in outline: Obcordate Ovary condition: Superior Persistence of styles/peria nth: Remnants at base Persistence of pedicel/s tipe: Pedicel and stipe Carpels: 5 Number of wings: 5 Margins: Sutured Wing derivation: Perianth Wing thickness: Chartaceous Trichomes: No Venation: Veins coursing apically forming la rge transversely elongate areoles. Veinlets ending within areoules with a small hook. Maxwellia (Figure 2-5D). Maxwellia is a New Caledonia endemic with the unusual condition of having many small seeds in an indehiscent fruit (Wilkins and Chappell 2002, Robyns et al. 1977). Species examined: M axwellia lepidota Baill. (Hrliman 13129 A) Type of fruit: Indehiscent Length: 3 cm Width: 2.3 cm. Shape in outline: Elliptical Apex: Obcordate Base: Cordate Ovary condition: Superior Persistence of styles/per ianth: Perianth at base Persistence of pedicel/stipe: Pedicel on specimen examined Carpels: 3-4 Number of wings: 3-4(-5) Margins: Entire and wavy Wing derivation: Ovary Wing thickness: Coriaceous

PAGE 47

47 Trichomes: No Venation: Wing too thick to observe Craigia (Figure 2-5J). Modern Craigia is found in Eastern China and Taiwan; but there are also fruit fossils from the Tertiary of North America, Europe and Asia (Kva ek et al. 1991). Species examined: C raigia yunnanensis (UF 1124) Type of fruit: Capsule Length: 3.1 cm Length: 2.3 cm Shape in outline: Elliptical Ovary condition: Superior Persistence of styles/perianth: No Persistence of pedicel/stipe: No Carpels: 3-4 Number of wings: 5 Margins: Entire; marginal vein Wing derivation: Ovary? Wing thickness: Papery/membranous Trichomes: No Venation: Primaries radiating dichotomizing and anastomosing to form elongate areoles, secondaries interconnecting and also fo rming freely ending veinlets which are sometimes branched. Pentace (Figure 2-8 K) Species examined: P entace laxiflora Merr. (Leopold 78602 A) Type of fruit: Indehiscent samara Length: 1.7 cm Width: 1 cm Shape in outline: Elliptical Ovary condition: Superior Persistence of styles/peria nth: Perianth at base Persistence of pedicel/stipe: Pedicel Carpels: 2-10 Number of wings: 5 Margins: Slightly tattered Wing derivation: Ovary? Wing thickness: Papery Trichomes: No Venation: Sinuous, radiating, some dichotom izing and anastomosing near the margin Malpighiaceae Aspidopterys. Aspidopterys is a liana which grows in tropical Asia. Species examined: A spidopterys roxburghiana A.Juss. (Griffith 924 A) (Narayanaswami 18 A)

PAGE 48

48 Type of fruit: Samara Length: 3 cm/ 4.3 cm Width: 2.3 cm/ 1.5 cm Shape in outline: Ovate/ oblong Ovary condition: Superior Persistence of styles/perianth: No Persistence of pedicel/stipe: No Carpels: 3 Number of wings: 3 Margins: Entire with veins looping along margin Wing derivation: Ovary Wing thickness: Chartaceous Trichomes: No Venation: Primaries forming loops with the secondaries interconnec ting and looping at the margin, tertiaries interconnect Nyctaginaceae This family has many species and genera with fin-winged fruits formed from accrescent tepals (Spjut 1994). Seed have a curved embryo. Only one genus is detailed here, a survey within the family needs to undertaken to doc ument the diversity within the family. Abronia. Abronia has been thoroughly studied for anatomy (Wilson 1974, Wilson, 1975), dispersability (Wilson 1976) and compared to the similar genus Tripterocalyx (Galloway 1975). Phaeoptilum (Figure 2-1C) Species examined : P. spinosum Radlk. (R. Srydel 199 USNSH) Type of fruit: Indehiscent anthocarp Length: 2.1 cm Width: 1.8 cm Shape in outline: Elliptical with lobate base Ovary condition: Superior Persistence of styles/perianth: Yes wings are derived from accrescent tepals Persistence of pedicel/stipe: No Carpels: 1 Number of wings: 4 Margins: Entire Wing derivation: Accrescent tepals Wing thickness: Chartaceous Trichomes: No Venation: Radiating slightly sinuous

PAGE 49

49 Onagraceae Megapterium (Figure 2-1G) Species examined: M egapterium missouriense Spach. (L.E Wham s.n. USNSH) Type of fruit: Apically opening capsule (dispersing via anemoballisty) Length: 6.8 cm Width: 6.8 cm Shape in outline: Transv ersly elliptical/ round Apex: Rounded Base: Rounded Ovary condition: Inferior Persistence of styles/perianth: No Persistence of pedicel/stipe: No Number of wings: 4 Margins: Entire and sutured? Wing derivation: Ovary. Wing thickness: Chartaceous. Trichomes: No. Venation: Striations radiating. Oenothera (Figure 2-1F) Species examined: O enothera missouriensis Sims (Stevens 561 Harvard Seed Collection), Oenothera macrocarpa (Hitchcock 165 Harvard Seed Collection) (Wagner s.n ., Seed Herbarium) Type of fruit: Loculicidal capsule dispersing via anemoballisty Length: 3.5/ 4 cm Width: 1.8/ 4 cm Shape in outline: Slightly ovate/ sligh tly obovate/ longitudina lly elliptical/ round Ovary condition: Inferior Persistence of styles/perianth: No Persistence of pedicel/stipe: No/ sometimes small pedicel? Locules: 4 Number of wings: 4 Margins: Entire and sutured Wing derivation: Ovary? Wing thickness: Chartaceous Trichomes: Sometimes small hairs Venation: Radiating striations Pedaliaceae Holubia (Figure 2-8F). Holubia is an herb native to Madagascar Species examined: H olubia saccata Oliver (Anonymous 301, USNSH) Type of fruit: Indehiscent Length: 5.4 cm Width: 5.3 cm

PAGE 50

50 Shape in outline: Round Apex: Rounded slightly cordate Base: Rounded slightly cordate. Ovary condition: Superior Persistence of styles/perianth: No Persistence of pedicel/stipe: Pedicel Carpels: 2 locules Number of wings: 4 Margins: Entire sinuous with marginal vein Wing derivation: Wing thickness: Chartaceous Trichomes: No Venation: One order, radiating outward; dichotomizing and anastomosing to form transverse longitudinal to polygonal areoles; marginal vein present Pterodiscus (Figure 2-8G) Species examined: P terodiscus aurantiacus Welw. (Werdermann 2290 GH) (Dister 4635 GH) Type of fruit: Indehiscent Length: 3.1 cm/ 3.3 cm Width: 3.1cm/ 3.4 cm Shape in outline: Round Apex: Rounded Base: Lobate Ovary condition: Superior Persistence of styles/per ianth: Perianth at base small apical protrusion Persistence of pedicel/stipe: Pedicle Carpels: 2-4 reported for the family Number of wings: 4 Margins: Entire with marginal vein Wing derivation: Wing thickness: Chartaceous Trichomes: No Venation: Radiating outward with some di chotomizing and anastomosing particularly towards the margin. Marginal vein is present Phyllanthaceae Hymenocardia. Hymenocardia is the only wind dispersed member or the Phyllaanthaceae and has previously been put in to its own family (Hymenocardiaceae) (Wurdak et al. 2004). Species examined : Hymenocardia ulmoides Oliver (Robyns 799 USNSH) Type of fruit: Schizocarp Length: 2.3 cm

PAGE 51

51 Width: 1.9 cm Shape in outline: Round-Elliptical Apex: Obsaggitate Base: Rounded Ovary condition: Superior Persistence of styles/perianth: No Persistence of pedicel/stipe: Both Carpels: Bilocular Number of wings: 2 Margins: Entire with marginal vein Wing derivation: Ovary Wing thickness: Chartaceous Trichomes: No Venation: Central vein running apically th rough the basal portion of the wing and over the apically placed seed cavity. Secondary veins ra diating transversely and dichotomizing and anastomosing irre gularly, marginal vein present. Polygalaceae Polygala Species examined: P olygala sp. (Britton 10129 USNSH) Type of fruit: Dehiscent or indehiscent Length: 4 cm Width: 3 cm Shape in outline: Obovate-round or asymmetrical elliptical Ovary condition: Superior Persistence of styles/perianth: Apical protrusion may be present and small disk at base Persistence of pedicel/stipe: Pedicel Carpels: 2 Number of wings: 2-3 Margins: Entire Wing derivation: Ovary Wing thickness: Papery Trichomes: Yes, unbranched Venation: Radiating, dichotomizing and anastomosing infrequently Polygonaceae Polygonaceae are a family with many genera with fin-winged fruit. The fruit type in this family is an achene, and the wings, in these fruits are formed from perianth parts. The fruits may be biwinged or triwinged and the shape is qu ite variable. Laubengayer (1937) detailed the anatomy of the flower. Genera with fin-winged fruits include Antigonon, Atraphaxis

PAGE 52

52 Calligonium, Polygonum, Rheum, Rumex and Triplaris Three of these genera are described below. Antigonon Species: A ntigonon guatemalense Meisn. (PI 146197 USNSH). Type of fruit: Achene with perianth forming wings Length: 3 cm Width: 3 cm Shape in outline: Round with lobate base Ovary condition: Superior Persistence of styles/periant h: Perianth forming wings Persistence of pedicel/stipe: Pedicel Carpels: 2-3 Number of wings: 3 Margins: Entire Wing derivation: Perianth Wing thickness: Papery Trichomes: No Venation: Primaries form reticulum with polygonal areoles, a few secondaries interconnect or end freely. Atraphaxis Species examined: Atraphaxis spinosa Eichw. (Viedumnu 1927-2 A) Type of fruit: Indehiscent Length: 0.8 cm Width: 0.85 cm Shape in outline: Ovate, apex rounded, base cordate Ovary condition: Superior Persistence of styles/perianth: Wings and perianth at base Persistence of pedicel/stipe: Pedicel Carpels: 2-3 Number of wings: 3 Margins: Entire Wing derivation: Perianth Wing thickness: Papery Trichomes: No Venation: Single primary running from base to apex over central body, secondaries dichotomizing and anastomosing occasionally across the wing. Rumex Species examined: Rumex obtusifolius Herb.Buch.-Ham. Ex Wall. (HARVARD SEED COLLECTION); Rumex venosus Pursh (77, USNSH). Type of fruit: Indehiscent Length: .8 cm/ 2.3 cm Width: .45 cm/ 3 cm Shape in outline: Obovate/ wide elliptical Ovary condition: Superior

PAGE 53

53 Persistence of styles/periant h: Perianth forming wings Persistence of pedicel/sti pe: Stipe on obtusifolius Carpels: 3 Number of wings: 3 Margins: Entire Wing derivation: Tepals Wing thickness: Papery Trichomes: No Venation: Single order forming reticulum w ith polygonal areoles or primaries forming reticulum with secondaries running la terally forming interconnections. Rhamnaceae Anatomy and dispersability of the winged fr uits of one subfamily, the Gouaniaeae, has been studied in detail (Medan 1988). Gouania (Figure 2-7A) Species examined: Gouania domingensis Aubl. (Rodriguez 2566, GH) Gouania lupuloides Urb. (Brumbach 9537 GH) (Brass 21138 GH) Type of fruit: Capsule Length: 1 cm/ 1.1 cm Width: 1.4 cm/ 0.9 cm Shape in outline: Wide elliptical Ovary condition: Superior Persistence of styles/p erianth: Apical disk Persistence of pedicel/stipe: Stipe Carpels: 3 Number of wings: 3 Margins: Entire often with an intramarginal vein Wing derivation: Ovary Wing thickness: Coriaceous Trichomes: No Venation: Irregular meshwork of irregularly spaced asymmetrical veins forming large irregular areoles marginal vein sometimes present. Chaydaia (Figure 2-7B) Species examined: Chaydaia berchemiaefolia Koidz. (Horder 601 USNSH) Type of fruit: Septicidal capsule Length: 1.4 cm Width: 1.1 cm Shape in outline: Circular Ovary condition: Superior Persistence of styles/perianth: No Persistence of pedicel/stipe: Stipe Carpels: 3 Number of wings: 3 Margins: Undulating

PAGE 54

54 Wing derivation: Ovary/hypanthium Wing thickness: Papery Trichomes: No Venation: Single order of irregularly loopi ng veins forming large polyhedral areoles sometimes looping to form an intramarginal vein. Reissekia (Figure 2-7C) Species examined: Reissekia smilacina Endl. (L.O Williams 6834 USNSH) Type of fruit: Inflated schizocarp Length: 1 cm Width: 1.6 cm Shape in outline: Butterfly shaped Ovary condition: Superior Persistence of styles/perianth: No Persistence of pedicel/stipe: No Carpels: 3 Number of wings: 3 Margins: Entire with marginal vein Wing derivation: Wing thickness: Papery Trichomes: No Venation: Marginal vein and one order of veins radiating outwards dichotomizing and anastomosing rarely in wing but frequently over the central body. Rutaceae The subfamily Pteleinae has thre e genera with winged fruits ( Ptelea Helietta and Balfourodendron) and one genus ( Taravalia ) with a nut like capsule. Of these Ptelea and Balfouodendron are fin winged while Helietta is similar to Acer in having two asymmetrical mericarps (Pirani 1998). Spathelia is another genu s in the family with fin-winged fruits, but is not described here. Balfourodendron (Figure 2-8A). Balfourodendron is found in both Central and South America. Pirani (1998) describes the samara as typically having four one-seeded locules in which most of the seeds are aborted. The epicarp, which forms the wings, is described as thin and coriaceous, the mesocarp thicker, soft and dot ted glandular dotted, the endocarp forms a hard lignified stone (Pirani 1998). Species examined: B alfourodendron riedelianum Engl. (Vanni et al. 385 GH) Type of fruit: Indehiscent samara

PAGE 55

55 Length: 1.7 cm Width: 2.3 cm Shape in outline: Widely elliptical Ovary condition: Superior Persistence of styles/p erianth: Remnants only Persistence of pedicel /stipe: Stipe present Carpels: 4-5 Number of wings: 4 Margins: Entire Wing derivation: Ovary Wing thickness: Papery Trichomes: No Venation: Three orders of venation, the first order arcing downwards, the second interconnecting, and third forming a reti culum. Venation is sinuous. Areoles polygonal and small (2/10 mm). Ptelea (Figure 2-8BC). Fossilized Ptelea samaras from North America were critically examined by Call and Dilcher (1995). They found th ree features to be particularly important in identifying the fruits of this genus 1) Superior. ovary demonstrated by floral disk at the base of the fruit where it joins the pedicel, 2) wings fu sed above and below the seed cavity, and 3) three slightly diverging veins extendi ng medially through the lower half of the samara forming a coarse transversely oriented reticulum over th e surface of the fruit body and a radiating looping reticulate pattern on the wings. I agree with these characters, although P. crenulata venation does not loop, but rather dichotomizes. Species examined: P telea crenulata Greene (HARVARD SEED COLLECTION), Ptelea polyadenia Greene (HARVARD SEED COLLECTION), Ptelea trifolata L. (HARVARD SEED COLLECTION) Type of fruit: Samara Length: 2.3 cm/ 1.5 cm Width: 2.5 cm/ 1.2 cm Shape in outline: Rounded Ovary condition: Superior Persistence of styles/perianth: No Persistence of pedicel/s tipe: Pedicel and stipe Carpels: 2-3 locules Number of wings: 2 or 3 Margins: Entire somewhat wavy; tattered in P. crenulata Wing derivation: Ovary Wing thickness: Papery Trichomes: No

PAGE 56

56 Venation: In P. crenulata veins dichotomizing and anas tomosing irregularly forming elongate areoles; in P. trifolata forming loops to create a reticulum of polygonal areoles. Sapindaceae The Sapindaceae have many genera with fruits th at are dispersed by air. Some of these fruits are fin-winged and others are not. All of the fin-winged fruits ha ve a marginal vein. Detailed comparisons of the fruits of the Paullineae have been made (Weckerle and Rutishauser, 2005). The fruits in this section may have in ternal hairs that are either unicellular or multicellular and all are three carpellate. Genera with fin-winged fruits include Dodonaea, Guioa, Majidea, Serjania Thouindium Toulicia Stocksia and Urvillea Five of these genera are described below. Guioa Species examined: G uioa comesperma Radlk. (Talucuchi 8673, GH) Type of fruit: Capsule Length: 2 cm Width: 2.3cm Shape in outline: Obcordate Ovary condition: Superior Persistence of styles/per ianth: Disk at base Persistence of pedicel/stipe: No Carpels: 1 (3 locular) Number of wings: 3 Margins: Entire Wing derivation: Ovary Wing thickness: Woody Trichomes: No Venation: Unavailable becau se the wing was too woody Stocksia Species examined: Stocksia brauhica image from Steven Manchester only Shape in outline: Ovate Ovary condition: Superior Persistence of styles/perianth: No Persistence of pedicel/stipe: Pedicel Margins: Entire with marginal vein Wing derivation: Ovary Venation: Reticulum reaching to margin with secondaries interconnecting or freeling ending in branched or unbranched veinlets.

PAGE 57

57 Thouindium Species examined: T houindium image from Steven Manchester only. Type of fruit: Capsule. Shape in outline: Elongate, base s lightly cordate, apex opening. Ovary condition: Superior. Persistence of styles/p erianth: Basal disk. Persistence of pedicel/stipe: Small pedicel. Number of wings: 3. Margins: Entire with marginal vein. Wing derivation: Ovary. Venation: Single order coursing downwards dichotomizing and anastomosing to form polygonal to elongate areoles; vein s entering the marginal vein. Toulicia Species examined: Toulicia image from Steven Manchester only. Shape in outline: Rounded obovate. Ovary condition: Superior. Persistence of styl es/perianth: No. Persistence of pedicel/stipe: No. Number of wings: 3. Margins: Entire with marginal vein. Wing derivation: Ovary. Trichomes: No. Venation: Primaries radiating, dichotomizi ng and anastomosing entering marginal vein; secondaries interconnecting and forming re ticulum of polygonal areoles or freely ending. Urvillea (Figure 2-8D) Species examined: U rvillea andersonii Ferrucci (Anderson 9150 NY) Type of fruit: Septifragal capsule Length: 3.6 cm Width: 1.7 cm Shape in outline: Elongate Ovary condition: Superior Persistence of styles/per ianth: Styles at apex Persistence of pedicel/stipe: Pedicel present, stipe present Carpels: 3 Number of wings: 2 Margins: Entire with marginal vein Wing derivation: Ovary Wing thickness: Chartaceous Trichomes: No Venation: Marginal vein and radiating sub parallel with some dichotomizing

PAGE 58

58 Styracaceae Halesia (Figure 2-5F). Halesia fruit fossils are found in Europe and reports have been made of fruits in North America. Fritsch et al. (2001) used these reports, and molecular phylogenies to suggest a North Amer ican origin for the genus. They admit that the fossils do not show the intramarginal vein found in Halesia species, but they still cons ider it to be this genus. Species examined: H alesia carolina L. (Hill 17001 GH), Halesia diptera L. (Harbison 1129 GH) Type of fruit: Indehiscent drupe Length: 3.7cm Width: 1.7 cm Shape in outline: Spatulate Apex: Cordate-sagittate Base: Obtuse Ovary condition: Inferior Persistence of styles/perianth: Calyx at apex Persistence of pedicel/stipe: Pedicel Carpels: 2 Number of wings: 2, 4 Margins: Entire wavy Wing derivation: Ovary Wing thickness: Thick semi-opaque Trichomes: No Venation: Intramarginal veins present Other: Fruits may persist on branches even after leaves seasonally fall, prominent apical beak present. Trigonaceae Humbertiodendron (Figure 2-7F) Species examined: H umbertiodendron soboundani Leandri (Anonymous, 244 USNSH) Type of fruit: Indehiscent samara Length: 2.5 cm Width: 2.3 cm Shape in outline: Round Ovary condition: Superior Persistence of styles/perianth: No Persistence of pedicel/stipe: None dispersed, pedicle present on branch Carpels: 3 Number of wings: 3 Margins: Entire. Wing derivation: Ovary Wing thickness: Chartaceous

PAGE 59

59 Trichomes: Yes (long-pilose) Venation: Marginal vein; primar y veins transverse to radiating upwards at approximately 45 degrees; occasionally dichotomizing in wing but dichotomizing more frequently near margin. Tropaeolaceae Tropaeolum (Figure 2-7H) Species examined: Tropaeolum porifolium (Cav.) L.Andersson & S.Andersson (A. Donat 51 USNSH) Type of fruit: Schizocarp Length: 1.6 cm Width: 1.3 cm. Shape in outline: Longitudinally elliptical Ovary condition: Superior Persistence of styles/perianth: No Persistence of pedicel/stipe: No Carpels: 3 Number of wings: 3 Margins: Entire not smooth with a marginal vein Wing derivation: Wing thickness: Papery Trichomes: No Venation: Marginal vein; radi ating outwards occasionally di chotomizing near the margin Zygophyllaceae Fruits of Zygophyllaceae may be capsular or Schizocarp.ic (Beier 2003). Some of these fruits develop wings in the fin-winged fash ion. Four genera are described below. Bulnesia (Figure 2-6AB) Species examined: B ulnesia chilensis Gay. (Werdermann 429, GH ); Bulnesia foliosa Griseb. (Schreiter 2849 GH); Bulnesia retamo Griseb. (Hutchinson & Wright 7115 GH) Type of fruit: Schizocarp Length: 1.6cm/ 4cm Width: 1.3/ 4 cm Shape in outline: Ellipt ical to round/elliptical Ovary condition: Superior Persistence of styles/periant h: Perianth at base in B. foliosa and B. chiliensis ; stylar protrusion in B. foliosa Persistence of pedicel/stipe: Stipe, sometimes pedicle Carpels: 5. Number of wings: 5. Margins: Entire. Wing derivation: Ovary. Wing thickness: Chartaceous.

PAGE 60

60 Trichomes: Yes unbranched on B. foliosa and B. rivas-martinezii (see Navarro, 1994). Venation: Radiating, dichotomizing, looping at margin with some interconnections particularly towards the margin. Morkillia Species examined: M orkillia mexicana Rose & Painter (Sousa 4601 GH) (Chase 7332 GH) Type of fruit: Capsule Length: 3.5 cm/ 3.75 cm Width: 2.5 cm/ 2 cm Shape in outline: Elliptical-ovate with trunc ate apex or apex with rounded invagination creating a saggitate appearance Ovary condition: Superior Persistence of styles/perianth: No Persistence of pedicel/stipe: Pedicel Carpels: 4 Number of wings: 4 Margins: Entire Wing derivation: Ovary Wing thickness: Coriaceous Trichomes: Yes, densely hairy Venation: Unknown Porleria (Figure 2-6D) Species examined: Porlieria angustifolia A.Gray (Moore 3288 Harvard) Type of fruit: Schizocarp Length: 1.7 cm Width: 1.7 cm Shape in outline: Widely ovate Ovary condition: Superior Persistence of styles/perianth: Stylar protrusion and basal disk Persistence of pedicel/stipe: Yes, both Carpels: 4 Number of wings: 3 Margins: Entire Wing derivation: Ovary Wing thickness: Coriaceous Trichomes: Yes unbranched Venation: Not visible by transmitted or reflected light

PAGE 61

61 Zygophyllum (Figure 2-6C) Species examined: ( Zygophyllum morgsana L. (Marloth 12275, A ); Zygophyllum retrofractum Thunb. (Marloth 650 A) Type of fruit: Schizocarp Length: 2 cm (+1cm. pedicel)/ 3 cm (+several centimeter pedicel) Width: 2.5cm/ 3cm Shape in outline: Elliptical Persistence of styles/perianth: Style at apex Persistence of pedicel/stipe: Pedicel Carpels: 4 Number of wings: 4 Margins: Entire Wing derivation: Ovary Wing thickness: Chataceous Trichomes: No Venation: Dichotomizing anastomosing and looping Ovary condition: Superior. to form a reticulum of areoles, that are transversely elongate towards the central body and increas ingly polygonal toward the margin.

PAGE 62

62 CHAPTER 4 RESULTS Fossil Fin-Winged Fruits of North America With Possible Affinities to Combretaceae The fossil history of the Combretaceae has been reported to include fossilized leaves and fruits (ex. Berry 1916), wood (ex. Sakala 2002), a possible flower (Takahashi et al. 1999) and pollen (ex. Salzman 2000). Three winged-fruit fossils from North America will be treated here ( Terminalia vera, Terminalia estamina and Terminalia oregona ). These fruits were not assigned to the genus Terminalia because it was the best fit based on careful comparisons, but simply because it is a genus in the Combretaceae. Additiona lly, these fruits were not compared to all of the genera that have fin-winged fruits, and it may be that there is a better match among the modern fin-winged fruits than any of the genera of Combretaceae. Carpolithes prangosoides Berry Basionym: Carpolithes prangosoides Berry 1916, USGS Prof. Paper 91, p. 351, pl. 104, Figure 2-9. Terminalia vera Berry 1926. Torrey Botanical Club Bu lletin. Vol. 53, p. 61, figs. 1-5. Carpolithus henryensis. Berry 1916, USGS Prof. Pa per 91, p. 352, pl. 112, fig. 16. Emended Description Fruit fin-winged, usually radially symmetri cal, appearing bisymmet ric in face view, elliptical, 1.2 .3 cm long, 1.5c m wide; composed of a centr al body with 5 fin-like, radiating wings, pedicel not dispersed w ith fruit. Style not persistent. Fruit body elongate, appearing smooth or long itudinally striated. Each wing lies in a radial plane intersecting with the longitudinal axis of the fruit, auriform to half-elliptic in shape. Near the distal end of the fruit, apical flaps pers ist in some specimens and it is unclear from what tissue these flaps arise. The wi ng venation is uniformly oriented towards the apex at about 30

PAGE 63

63 degrees from the fruit's central axis. Venation co nsists of numerous fine veins, frequently dichotomizing, forming loops along the way to ward the margin, with secondary veins interconnecting and forming areoles of the reticu lum 0.7-1.5 mm. Marginal or intramarginal vein absent. Comments. This species is common in claypits of the Middle Eocene Claiborne Fm in Tennessee and Kentucky. Although the original specimen figured as Carpolithus prangosoides (Berry 1916) does not clearly show the wing vena tion, it does show a pair of elongate lateral wings that can be matched with spec imens found later and described as Terminalia vera Berry (1930). An elongate central bulge in this specimen, representi ng the fruit body shows a median longitudinal groove that we inte rpret as a third wing, while the counterpart of this specimen (broken) shows at least one additional gr oove, which we interpret as another wing. Carpolithes henryensis is represented by only a single specimen, wh ich lacks a counterpart. This specimen also shows a central body with elongate lateral wings with a gr oove, which is interpreted as a third wing. Based on the observed features, thes e specimens can be matched to those Berry (1930) illustrated later under the name Terminalia vera and we place all of these in synonymy under the earlier name Carpolithes prangosoides Most of the C. prangosoides specimens are exposed in long itudinal fractures that show only two visible wings, positioned laterally on either side of the fusiform central body and lying in a common plane (Figures 2-10 & 2-11). This configuration is an ar tifact of the horizontal orientation of the fruit within the sediment in which it was buried, and of the lengthwise plane through the sediment that intercepted a nd exposed the fossil. In describing Terminalia vera Berry considered that the fruits possessed only two wings. (Berry 1930). Careful microscopy of some specimens shows a stairstep fracture patter n in which multiple levels on which different

PAGE 64

64 wings are exposed. In order to determine the full number of wings, two specimens were deliberately fractured transversely in the labora tory and show unequivocally that there are five wings (Figure 2-10). None of the specimens show the pedicel, styl es or the locule condi tion. Because pedicels have not been found attached to the fruits, it may be inferred that the fruits were sessile or that they were shed by abscising from the top of the pe dicel. Because the fruit is always found with a full complement of wings it is inferred that it was not schizocarpic. But the available characters, including the number and position of wings and their venation allows for detailed comparison with extant fruits of similar form. Although this species was assigned to the Combretaceae based on its lateral wings, detailed comparison of wing venation distinguishes th e fossil from most of the genera within this family. The complex system with primary vein s forming loops and secondaries interconnecting is similar to the venation in the wings of Calycopteris and Macropteranthes Calycopteris does not have a fin-winged fruit; its wings are arra nged like a propeller around the apex, and are derived from accrescent perianth. Macropteranthes which is a member of a separate tribe, Lagunculariaeae (Excell and Stace 1965) is finwinged with four wings derived from two bracteoles, such that a pair of wings extends from each bracteole. The venation is strikingly similar to that of the fossil. The geometry of a five-winged fruit would be difficult to reconcile with formation of two wings per bracteole. Other families with five winged fruits include Apiaceae, Lophopyxidaceae, Malvaceae and Zygophyllaceae (Table 2-1). The fruits of Zygophyllaceae are meri carps and those of Malvaceae are usually capsular so those two families can be eliminated. Apiaceae do not have prominent venation and can be eliminated and Lophopyxidaceae lack th e looping pattern of

PAGE 65

65 venation on the fruits. Although no genus fits cl osely; the family, Comb retaceae seems to be morphologically most similar to these fossils. The fruit lacks the distinguis hing characteristics of the family such as combretaceous hairs, and becaus e these fruits may be those of an in an extinct genus, there is not enough evidence to place this fossil in the family. Terminalia stamina MacGinitie MacGinitie (1941) described Terminalia e stamina based on three unattached leaf specimens and a winged fruit from the Middle Eo cene Chalk bluffs flora of California. He called the cited specimens cotypes, but did not designate a holotype. Subsequently (MacGinitie 1969, p. 129), he made a formal ne w combination, transferring the species to Mastixia estamina (MacGinitie) MacGinitie, based on two of the leaf specimens that had been figured earlier. In this transfer he cited the two of the leaf sp ecimens but specifically excluded the remaining leaf and fruit specimens, stating that only Figure 2-2, plate 42 represents a leaf of Terminalia and The fruit, Figure 2-4, pl ate 43, is correctly assigned to Terminalia This created the awkward situa tion that the two specimens that MacG initie (1969) sti ll considered to represent Terminalia were left without an epithet. To resolve this situation, a new epithet will have to be given to this fossil. Description Fruit elliptical in outline, symmet rical, rounded over th e apex, length is 33.5 cm, width 2.5 cm. The fossil fruit has a fusiform central body, 13 mm long, 4 mm wide, from which at least five longitudi nal wings radiate. Fruit is inde hiscent. An axial vein extends about 7 mm from the base of the fruit to th e base of the central body. Venation of the wings radiates from the central body toward the margi n. Veins of the wing fine, dichotomizing and anastomosing, spaced from 0.5-1 mm apart without prominent looping or cross-veins. No marginal vein present. Features of style and perianth are unknown.

PAGE 66

66 Comments. I have seen only two fruit specimens of this species, both of which are broken and incomplete. One of the specimens has been fractured obliquely across the fruit, such that impression surfaces of five wings can be counted (Figure 2-11). Both specimens seem to have the wings intact and because no dispersed me ricarps of this type have been found in the fossil flora, we believe it to be non-schizocarpic, indehiscent fruit. In his treatment, MacGinitie compared the Eocene fossil fruits that he called Terminalia estamina to the modern species Terminalia hainanensis and Terminalia triflora, which I have not observed in fruit. The other species of the genus that I examined also showed some similarity of form and venation. Important features of this fossil fruit include apparent indehiscence, a fusiform central body, and presence of five wings with fine vena tion which dichotomizes. Other genera of angiosperms with a similar pattern of venation include Ptelea Wimmeria Tripterigium Bulnesia Tropaelum and Pterodiscus None of these genera have five wings as does the fossil and as do members of the Combretaceae (Table 2-1). Pentace is a genus with similar venation and five wings, but the wing edges in this genus are reinforced by many dichotomizations near the margin, which is not consistent with the fo ssil because the dichotomizations lie within the wing rather than along the margin. If we exclude those modern genera with fewer than five wings and those with dehiscent fruits, the only remaining candidates for comparison with this fossil are in the Lophopyxidaceae and Combret aceae (Table 2-1). The vein density of Lophopyxis is much greater with at least five vein s per millimeter rather than 1-2 veins per millimeter of the fossil. Fruits of Combretum and Terminalia (Combretaceae) can show similar form to the fossil, including the presence of a fusiform central body that is positioned above the

PAGE 67

67 base of the fruit. They also tend to have hi gher vein density than the fossil and their margins tend to be ragged due to lack of reinforcing venation. Of the modern fin-winged fruits, fruits of genera within the Combretaceae seem to be most similar to these fossils. Some diagnostic fe atures of the family could not be observed in these fossils including the inferior ovary, the number and positioning of seeds, presence of combretaceous hairs. Additionally, the venation is quite different in Macropteranthes Meiostemon and Thiloa have only four wings and Pteleopsis has only two. Within the Combretaceae, the fruit fits best as either the genus Terminalia or the genus Combretum This does not mean that this necessarily belongs to this family because th e similarity lies partly in the absence of any distinguishing characters, and th e fossils may represent an extinct species in another family. Indeed, the fossil seems to differ from both Combretum and Terminalia in some aspects. The genus Combretum tends to have edges which fray becau se the veins so rarely dichotomize and the venation is very fine so that there is nothing to stop a small t ear from becoming a large one. The fossil in contrast has smooth edges and there are clearly spaces between the veins. Terminalia may also have these very fine veins and tattered edges, but it can also have more widely spaced veins and smoother edges (e.g. Terminalia) The assignment of this fossil to the genus Combretaceae is not certain without confirming characters such as combretaceous hairs or apical placentation, and th e assignment to the genus Terminalia is certainly not justified. Terminalia oregona (Lakhanpal) Meyer and Manchester Basionym: Halesia oregona Lakhanpal, 1958, pro parte, Un iversity of California Publications in Geological Sciences 35, p. 36, pl. 9, fig. 1, Pl. 10, Figure 2-4 only. Terminalia sp. Brown 1959, pro parte, Journal of Paleontology 33, p. 128; Pl. 24, Figure 2-16 only.

PAGE 68

68 Terminalia oregona (Lakhanpal) Meyer and Manchester 1997, University of California Publications in Geological Sciences 141, p. 128, pl. 51, Figures 2-1 through 2-4. Description This fruit is elliptical to obovate pedicellate, symmet rical, base ranging from cuneate and acute to rounded and cordate, apex obcordate and r ounded, length (20 mm, 25 mm, 45 mm, 33 mm, 40 mm, 55 mm, 33 mm), width (18 mm, 20 mm, 31 mm, 30 mm, 20 mm, 27 mm, 25 mm). Pedicel is 1 mm wi de and up to 14 mm long. The fruit is nonschizocarpic. The fossil fruit has a fusiform central body mainly confined to the upper 1/ 3 of the fruit, 17-21 mm long, and 5-6.5mm wide with severa l prominent longitudinal ribs. A single vein leads from the pedicel to the central body. Fruits have up to at least three wings. Fruits with at least two wings, and in one specimen (Figure 2-13 5,6) at leas t three were observed. Veins of moderate thickness, radiating outward from the central axis and occasionally dichotomizing, spaced 0.5 mm apart, without prominent looping or cross vein s, no marginal vein present. Stylar protrusion with a thickened disk like structure at its junction with the central body. Comments. Important features include apparently three wings, inferior ovary with persistent stylar protrusion and apical disk, central body ridged a nd situated in the upper 1/3 of the fruit. Modern fin-winged fruits with an inferior ovary include Apiaceae, Begoniaceae, Burmanniaceae, Combretaceae, Dioscoreaceae, He rnandiaceae, Loranthaceae, Onagraceae, and Styracaceae. We are uncerta in if they were variable in wing number from 2-3 or whether those specimens showing two wings are incompletely preserved. Dioscoreaceae and Begoniaceae are eliminated because they have a marginal vein which is not found on the fossil. Lakhanpal (1958) believed it might be Halesia (Styracaceae). Halesia is distinguished by promin ent intramarginal veins on the wings distinctive of this genus and not seen in the fossil. Halesia has perianth remaining at the apex where the fossil does not (Figure

PAGE 69

69 2-7F). Apiaceae fruits are typically smaller without obvious venation. Hernandiaceae venation was sinuous without dichotomizations. Onagrace ae venation tends to be obscure or arching upward. Loranthaceae venation tends to be more cl osely spaced. Each character in this fossil can be found in some species of the Co mbretaceae, but not in combination. Terminalia oregona is a fossil with characters not inconsistent with the Combretaceae. This fruit has three wings with radiating venation that only rarely dichotomizes. This venation pattern is consistent w ith some species in both Terminalia and Combretum The apical protrusion is similar to that found in Meiostemon and the apical position of the seed cavity can be found in Pteleopsis None of these fossils shows characters that would clearly place it into the Combretaceae, but none are wholly inconsistent. Terminalia vera has venation strikingly similar to Macropteranthus but the difference in wing numbe r is difficult to reconcile. Terminalia estamina and Terminalia oregona are both most similar to either Combretum or Terminalia Terminalia estamina has five wings and radiating venation, but the veins are not closely spaced as is the common condition. Terminalia oregona also has venation which is more widely spaced than usual for the family. Overall, the lack of any diagnos tic character placing these fruits in this family, the lack of critical examination of the leaves formerly pl aced in Combretaceae leaves me unwilling to state that there was a record of the family in th e United States during the Eocene-Oligocene.

PAGE 70

70 CHAPTER 5 CONCLUSIONS Based on the previous chapters I have been able to compare the various modern finwinged fruits and determine the status of th e fossils. These results are presented below Modern Winged Fruits Based on the survey presented in chapter three, fin-winged fruits occur in at least 19 orders of Angiosperms, occurring in magnoliids, monocots, and eudicots. Within eudicots, such fruits are particularly well represented among th e rosids, with 18 of the 30 families that include genera with fin-winged fruits. Th ese fruits are also represented in basal eudicots and asterids (Figure 2-8). Anemoballistic fruits were found convergently in three families. Of the 87 genera known, only 34 had an inferior ovary, less than half had a marginal vein, suture, or intermarginal vein. Many of those that lack the marginal vein did have reinforcement towards the edge, either dichotomizi ng more frequently or forming loops distally. Presumably this helps prevent the wings from becoming tattered. Fruits of Combretum which lack marginal veins and looping, are ofte n frayed along the margin (Figure 2-2). Wings can be formed from ovary tissue, from perianth parts and/or bracts. Those formed from perianth or bracts may take the form of la mina appressed to the ovar y. In these cases, the wings may be formed from the fusion or partial fusion of two adjacent lamina ( Reisseika ) or each lamina may form two wings freely ( Macropteranthus ). The fin-winged fruit type is found on every co ntinent except Antarcti ca, and in habitats ranging from desert ( Macropteranthus ) to rainforest ( Cavanellesia ). These fruits occur on trees ( Cavanellisia, Craigia,Balfourodendron ), shrubs ( Phaeoptilum, Maxwellia, Wimmeria ) vines ( Dioscorea, Triptergium, Kleinhovia ) and herbs ( Polygonum, Stackhousia, Abronia ).

PAGE 71

71 Fin-winged fruits have enough di stinguishing characters to se parate them by family. A dichotomous key to the fin-winged fr uits (by family) is given below. 1a. Fruit from an inferior ovary 2a. Fruit dispersing via anemoballisty with many seeds 3a. Fruit with two or four wings.....................................................................Onagraceae 3b. Fruit with three wings 4a. Fruit opening apically, wings equal in size, seeds may be winged............................................Dioscoreaceae/Burmanniaceae 4b. Fruit splitting along sutures, wings equal or unequal in size, seeds not winged.................................................................Begoniaceae 2b. Fruit dispersing otherwise, capsule, mericarp, samara etc. 5a. Fruits wing with intr amarginal vein........................................................Styracaceae 5b. Fruits without intramarginal vein 6a. Fruit a schizocarp..................................................................................Apiaceae 6b. Fruit non-schizocarpic 7a. Unicarpellate fruit with sinuous venation............................Hernandiaceae 7b. Three-five carpellate, venation otherwise 8a. Fruit small (1cm), wings 3, carpels 3-4..........................Loranthaceae 8b. Fruits usually greater than 1cm, carpels 4-5, wings 2, (3), 4, 5............................................................Combretaceae 1b. Fruit from a superior ovary 9a. Number of wings five or greater 10a. Venation arching downwards.........................................................Lophopyxidaceae 10b. Venation otherwise (radi ating, transverse, ascending) 11a. Wings six-many, may be incomple te, fruit indehiscent...................Achariaceae 11b. Wings five (or fewer in Maxwellia ), wings complete, membranous-chartaceous (coriaceous in Maxwellia ), fruit a capsule.....................................................................................Malvaceae 9b. Number of wings less than five (2, 3, or 4) 12a. Marginal vein present 13a. Four wings, fruit nearly round in outline..........................................Pedaliaceae 13b. Two-three wings, fruit not round in outline 14a. Fruits indehiscent 15a. Marginal vein only, style persistent, long (1cm) and thick (2mm)......................................................Cardiopteridaceae 15b. Venation more than ju st marginal vein, style if persistent less than 1cm and less than 2mm thick. 16a. Calyx persisting at the base of fruit..........................Celastraceae 16b. Calyx not persisting...................................................Cunonaceae 14b. Fruits dehiscent, capsular or schizocarpic 17a. Two winged 18a. Fruit schizocarpic, bilocular ovary.......................Phyllanthaceae 18b. Fruit capsular, ovary with three carpels...Sapindaceae (Urvillea) 17b. Three winged

PAGE 72

72 19a. Fruit schizocarpic 20a.Calyx persistent at base, venation radiating from central body with no interconnections, joining the marginal vein............................Stackhousiaceae 20b. Calyx not persistent, venation otherwise 21a. Fruit elliptic al, longer than wide.............Tropaelaceae 21b. Fruit butterfly shaped, wider than long................................Rhamnaceae ( Reisseika ) 19b. Fruit capsular 22a. Venation irregular, asymmetrical, forming variably sized polygonal areoles.......................Rhamnaceae 22b. Venation radiating, dichotomizing and anastomosing......................................................Sapindaceae 12b. Marginal vein absent 23a. Fruit schizocarpic, four wings....................................................Zygophyllaceae 23b. Fruit non-schizocarpic, wings 2,3, or 5 24a. Fruit coriaceous with many small seeds/fruit........Malvaceae (Maxwellia) 24b. Fruit not coriaceous, fewer seeds/fruit 25a. Fruit formed from persistent perianth 26a. Fruit an achene.......................................................Polygonaceae 26b. Fruit otherwise.........................................................Nytaginaceae 25b. Fruit formed from ovary 27a. Sepals or disk persistent at base 28a. Sepals persistent no persistent disk....................Cyrillaceae 28b. Sepals not persistent disk persistent................Polygalaceae 27b. Sepals/disk not persistent 29a. Three-winged 30a. Veins forming loops, trichomes absent..Malpighiaceae 30b. Veins dichotomizing but not looping, trichomes present........................................Trigonaceae 29b. Two or four winged 31a. Fruit longer than wide.....................................Fabaceae 31b. Fruit wider than long or nearly equal..............Rutaceae Fossil Winged Fruits As presented in Chapter 4, three different o ccurrences of fossil fin-winged fruits have been attributed previously to Combretaceae in North America. Two are from the Eocene, one from the Oligocene, and all have been assigned to the genus Terminalia Each of these was reexamined with attention to the original fossi l specimens and closely compared with extant fruits of Combretaceae and other extant families. The features preserved in these fossils are insufficient to confirm identific ation to that family. Although similar in many respects to

PAGE 73

73 Combretum and Terminalia these fossils may represent conve rgence in fruit form and could represent another family. In order to confidently assi gn a fossil fin-winged fruit to the Combretaceae, characters such as cross sectional anatomy, seed condi tion, or presence of co mbretaceous hairs(long, straight, sharply pointed, unicellula r, very thick walled, with conical internal component at base), should be present. Tattered wing margins are common in Combretum and could be a contributing character when analyzing a fossil. It is hoped that the survey of extant finwinged fruits presented here will provide an improved basis for evaluating similar types of fru its from the fossil record. Such winged fruits are known as early as the late Albian of Ka nsas (Dilcher, pers. comm. 2006) and are well represented in Tertiary floras. Special attention to details of venation as well as features of persistent perianth, bracts, disks, when present can facilitate confident identification of extant genera, and provides a basis for recognizing extinct taxa.

PAGE 74

74 LIST OF REFERENCES Anderberg, AA & Zhang X, 2002. Phylogenetic relationships of Cyrillaceae and Clethraceae (Ericales) with special emphasis on the genus Purdiaeae planch. Organisms Diversity and Evolution 2:127-137. Angiosperm Phylogeny Group 2003. An update of the angiosperm phylogeny group classification for the orders and families of flowering plants: APG II Botanical Journal of the Linnean Society 141:399. Augsperger, CK, 1988. Mass allocation, moisture content, and dispersal capacity of winddispersed tropical diaspores. New Phytologist 108:357-368. Bailey, VL, 1962. Revision of the genus Ptelea (Rutaceae). Brittonia 14:1-45. Barneby, RC & Grimes, JW, 1984. Two leguminous fo rest trees new to the flora of French Guiana. Brittonia 36:45-48. Barnes, RW & Hill, RS, 1999. Ceratopetalum fruits from Australian Cainozoic sediments and their significance for petal evolution in the genus. Australian Systematic Botany 12:635645. Beier, B-A, Chase, MW & Thulin, M, 2003. Phylogenetic relationships and taxonomy of subfamily Zygophylloideae (Zygophyllaceae) base d on molecular and morphological data. Plant Systematics and Evolution 240:11-39. Berry, EW, 1916. The Lower Eocene floras of S outheastern North America. US Geological Survey Professional Paper 91:1-481. Berry, EW, 1926. Terminalia in the Lower Eocene of Southeastern North America. Bulletin of the Torrey Botanical Club 53:59-65. Berry, EW, 1930. Revision of the Lower Eocene Wilc ox flora of the Southeastern States with descriptions of new species chiefly from Tennessee and Kentucky. U.S. Geological Survey Professional Paper 156:1-196. Bradford, JC & Barnes, RW, 2001. Phyloge netics and classification of Cunoniaceae (Oxalidales) using chloroplast DNA sequen ces and morphology. Systematic Botany 26:354-385. Brown, RW, 1959. A bat and some plants from the Upper Oligocene of Oregon. Journal of Paleontology 33:125-129. Burnham, RJ, 1995. A new species of winged fruit from the Miocene of Ecuador: Tipuana Ecuatoriana (Legumniosae). American Journal of Botany 82:1599-1607. B ek, Kva ek, Z, & Manchester, SR, 1989. Sapindaceous affinities of the Pteleaecarpum fruits from the Tertiary of Eurasia and North America. Botanical Gazette 150:477-489.

PAGE 75

75 Burrows, CJ, 1995. Germination behavior of the s eeds of six New Zealand woody plant species. New Zealand Journal of Botany 33:365-377. Caddick, LR, Wilkin, P, Rudall, PJ, Hedderson, TA J, & Chase, MW, 2002. Yams reclassified: A recircumscription of Dioscoreac eae and Dioscoreales. Taxon 51:103-114. Call, VB & Dilcher, DL, 1992. Investigations of Angiosperms from the Eocene of Southeastern North America: Samaras of Fraxinus wiilcoxiana berry. Review of Palaeobotany and Palynology 74:249-266. Call, VB & Dilcher, DL, 1995. Fossil Ptelea Samaras (Rutaceae) in North America. American Journal of Botany 82:1069-1073. Call, VB & Dilcher, DL, 1997. The fossil record of Eucommia (Eucommiaceae) in North America. American Journal of Botany 84:789 814. Caris, P, Decraene, LPR, Smets, E & Clinckema illie, D, 2002. The uncertain systematic position of Symplocos (Symplocaceae) evidence from a floral ontogenetic study. International Journal of Plant Sciences 163:67-74. Conti, E., Litt, A, Wilson, PG, Grahm, SA Briggs, BG, Johnson, LAS & Stoma, KJ, 1997. Interfamilial relationships in Myrtles: Mo lecular phylogeny and patte rns of morphological evolution. Systematic Botany 22:629-647. Corbett, SL & Manchester, SR, 2004. Phytogeography and fossil history of Ailanthus (Simroubaceae). International Jour nal of Plant Sciences 165:671-690. Dallwitz, MJ, Paine, TA and Zurcher, EJ, 2000 onwards. Principles of interactive keys. http://delta-intkey.com Duke, JA, 1961. Flora of Panama. Part IV. Fa scicle IV. Annals of the Missouri Botanical Garden 48:1-106. Dwyer, JD & Hermann, FJ, 1965. Flora of Panama. Pa rt V. Fascicle 4. Annals of the Missouri Botanical Garden 52:1-54. Excell AW and Stace, CA, 1965. Revision of th e Combretaceae. Boletim da Sociedade Broteriana 40:5-26. Ferrucci, MS, 2006. A new species of Urvillea (Sapindaceae) from Northwestern Venezuela. Brittonia 58:83-87. Fritsch, P, Morton, CM, Chen, T, & Meldrum, C, 2001. Phylogeny and biogeography of the Styracaceae. International Journa l of Plant Sciences 16:S95-S116. Galloway, L, 1975. Systematics of the North American desert species of Abronia and Tripterocalyx (Nyctaginaceae). Brittonia 27:328-347.

PAGE 76

76 Garwood, NC, 1985. The role of mucilage in the germination of Cuipo, Cavanillesia platanifolia (H. & B.) H.B.K. (Bombacaceae), a Tr opical Tree. American Journal of Botany 72:1095-1105. Gianoli, E, 2004. Evolution of a climbing habit promotes diversification in flowering plants. The Royal Society 271:2011-2015. Goldberg, A, 1986. Classificati on, evolution, and phylogeny of the families of Dicotyledons. Smithsonian Institution Press. Washington D.C. Herendeen PS & Dilcher, DL, 1990. Fossil mimisoid legumes from the Eocene and Oligocene of Southeastern North America. Review of Palaeobotany a nd Palynology 62:339-361. Herendeen, PS, Les, DH, & Dilcher, DL, 1990. Fossil Ceratophyllum (Ceratophyllaceae) from the Tertiary of North America. American Journal of Botany 77:7-16. Hjerpe, J & Hedenas, H, 2001. Tropical rain forest recovery from cyclone damage and fire in Samoa. Biotropica 33:249-259. Hutchinson, J, 1964. The genera of flower ing plants. Clarendon Press. Oxford. Johnson, LAS & Briggs, BG, 1984. Myrtales and My rtaceae a phylogenetic analysis. Annals of the Missouri Botanical Garden 71:700-756. Jones, JH & Dilcher, DL, 1980. Investigations of angiosperms from the Eocene of North America: Rhamnus marginatus (Rhamnaceae) reexamined. American Journal of Botany 67:959-967. Jongkind, CCH, 1995. Review of the genus Strephonema (Combretaceae). Annals of the Missouri Botanical Garden 82:535-541. Judd, WS, 2006. Advanced plant taxonom y. Target Copy. Gainesville, FL. Karehed, J, 2001. Multiple origin of the tropi cal forest tree family Icacinaceae. American Journal of Botany 88:2259-2274. Kva ek, Z, Buzek, C, & Manchester, SR, 1991. Fossil fruits of Pteleaecarpum weyland Tiliaceous, not Sapindaceous Botanical Gazette 152:522-523. Lakhampal, RN, 1958. The Rujada flora of West Central Oregon. University of California Publications in Geological Sciences 35:1-66. Lamont, B, 1985. Dispersal of the winged fruits of Nuytsia floribunda (Loranthaceae). Austral Ecology 10:187-193. Laubengayer, RA, 1937. Studies in the anatom y and morphology of the polygonaceous flower. American Journal of Botany 24:329-343.

PAGE 77

77 Lavin, M, Pennington, RT, Klitgaard, BB, Sprent JI, De Lima, HC, & Gasson, PE, 2001. The Dalberioid legumes (Fabaceae): Delimitation of a pantropical monophyletic clade. American Journal of Botany 88:503-533. Levin, RA, 2000. Phylogenetic relationships within Nytaginaceae nyctagineae : Evidence from nuclear and chloroplast genomes Systematic Botany 25:738-750. Liu, M (R), Plunkett, GM, Lowry II, PP, Van Wyk, BE, & Tilney, PM, 2006. The taxonomic value of fruit wing types in the order Apia les. American Journal of Botany 93:1357-1368. Mabberly, DJ, 1989. The plant-book; A portable dict ionary of the higher plants. Cambridge University Press. New York. MacGinitie, HD, 1941. A middle Eocene flora fro m the central Sierra Nevada. Carnegie Institute Washington Publication 584:1-78. Manchester, SR, 2001. Update on the megafossil fl ora of Florissant, Colorado. Proceedings of the Denver Museum of Nature & Science 4:137-161. Manchester, SR & Donohue, M, 1995. Winged fr uits of Linnaeeae (Cap rifoliaceae) in the Tertiary of Western North America: Diplodipelta Gen. Nov. International Journal of Plant Sciences 156:709-722. Matlack, GR, 1987. Diaspore size, shape, and fall behavior in wind-disp ersed plant species. American Journal of Botany 74:1150-1160. Matthews, ML, Endress, PK, Schonenberger, J, & Friis, EM, 2001. A comparison of floral structures of Anisophylleaceae and Cunonaceae and the problem of their systematic position. Annals of Botany 88:439-455. Matthews, ML & Endress, PK, 2005. Comparative fl oral structure and syst ematics in Celastrales (Celastraceae, Parnassiaceae, Lepidobotryaceae). Botanical Journal of the Linnean Society 149:129-194. McClain, AM & Manchester, SR, 2001. Dipteronia (Sapindaceae) from the Tertiary of North America and implications for the phytogeographic history of the Aceroideae. American Journal of Botany 88:1316-1325. Medan, D, 1989. Diaspore diversity in the an emochorous Gouanieae (Rhamanceae). Plant Systematics and Evolution 168:149-158. Meyer, HW & Manchester, SR, 1997. The Oligoc ene Bridge Creek flora of the John Day Formation, Oregon. University of California Publications in Geol ogical Science 141:1195, 75 pl.

PAGE 78

78 Mirle, C & Burnham, R, 1999. Identification of asymmetrica lly winged samaras from the Western Hemisphere. Brittonia 51:1-14. Navarro, G, 1994. A new species of Bulnesia (Zygophyllaceae) from the Xerothermic Southern Puna of Bolivia. Novon 4:280-284. Pirani, JR, 1998. A revision of Helietta and Balfourodendron (Rutaceae-Pteleinae). Brittonia 50: 348-380. Razi, BA, 1950. A contribution towards the study of the dispersal mechanisms in flowering plants of Mysore (Sout h India). Ecology 31:282-286. Renner, SS, 1986. Circumscription and phylogeny of the Laurales: Evidence from molecular and morphological data. Ameri can Journal of Botany 86:1301-1315. Schopmeyer, CS, 1974. Seeds of woody plants in the United States. Forest Service, U.S. Department of Agriculture. Washington D.C. Sakala, J, 2000. Silicified angi osperm wood from the Dangu local ity (Ypresian of the Gisors region, Eure, France)final par t: the problem of palaeoclim ate reconstruction based on fossil wood. Geodiversitas 22:493-497. Stace, CA, 1965. The use of epidermal character s in phylogenetic considerations. The New Phytologist 65:304-319. Stace, CA, 1968. A revision of the genus Thiloa (Combretaceae). Bulletin of the Torrey Botanical Club 95:156-165. Stevens, PF, (2001 onwards). Angiosperm phylogeny website. Version 7, May 2006 [and more or less continuously updated since].http ://www.mobot.org/MOBOT/research/APweb/. Spjut, RW, 1994. A systematic treatment of fruit types. Memoirs of the New York Botanical Garden 70:1-182. Takahashi M, Crane, PR & Ando, H, 1999. Fossil fl owers and associated plant fossils from the Kamikitaba locality (Ashizawa Formati on, Futuba Group, lower Coniacian, Upper Cretaceous) of northeast Japan. Jour nal of Plant Research 112: 187-206. Tan, F, Shuhua, S, Zhong, Y, Gong, X, & Wang, Y, 2002. Phylogenetic relationships of the Combretoideae (Combretaceae) inferred from pl astid, nuclear gene and spacer sequences. Journal of Plant Research 115:475-481. Tebbitt, MC, Lowe-Forrest, L, Santoriell o, A, Clement, WL, & Swensen, SM, 2006. Phylogenetic Relationships of Asian Begonia, with and emphasis on the evolution of rainballist and animal dispersal mechanisms in sections Platycentrum, Sphenanthera and Leptosae.

PAGE 79

79 Tidwell, D, 1975. Common fossils plants of West ern North America; Bri gham Young University Press. Tiffney, B, 1990. The collection and study of disper sed angiosperm fruits and seeds. Palaios 5:499-519. Tobe, H, Wagner, WL & Chin, HC, 1987. A syst ematic and evolutionary study of Oenothera (Onagraceae): Seed Coat Anat omy. Botanical Gazette 148:235-257. Vanvinckenroye, P, Cresens, E, Ronse Decraene, L-P & Smets, E, 1993. A comparative floral development study in Pisonia Bougainvillea and Mirabilis (Nyctaginaceae) with special emphasis on the Gynoecium and floral nectar ies. Bulletin du Jardin botanique de Belgique/ Bulletin van de National Plantentuin van Belgie 62:69-96. Vigilante, T, Dixon, K, Sieler, I, Roche, S & Tieu, A, 1998. Smoke germination in Australian plants. Rural Industries Research and Development Corporation. Vijayaraghavan, MR, 1969. Studies in the Family Cyrillaceae. I. Development of male and female gametophytes in Cliftonia monophylla (Lam.) Britton ex Sarg. Bulletin of the Torrey Botanical Club 4:484-489. Wang, Y & Manchester, SR, 2000. Chaneya a new genus of winged fruit from the Tertiary of North America and Eastern Asia. Internat ional Journal of Pl ant Sciences. 2000. 161:167178. Weckerle, CS & Rutishauser, R, 2005. Gynoecium fruit and seed structure of Paullinieae (Sapindaceae). Botanical Journa l of the Linnean Society 147:159-189. Whitlock, BA, Bayer, C, & Baum, DA, 2001. Phyloge netic relationships and floral evolution of the Byttnerioideae (Sterculiaceae or Malva ceae s.l.) Based on sequences of the chloroplast gene ndhF. Systematic Botany 26:420-437. Wilde, V & Frankenhauser, H, 1998. The Mi ddle Eocene plant taphocoenosis from Eckfeld (Eifel, Germany). Review of Palaeobotany and Palynology 101:7-24. Wilkins, CF & Chappill, JA, 2002. Seed and seedling morphology and seed anatomy of Lasiopetaleae (Malvaceae s.l. or Sterculiaceae ). Australian Systematic Botany 15:545563. Wilson, RC, 1974. Abronia : II. Anthocarp polymorphism a nd anatomy for nine species of Abronia found in California. Aliso 8:113-128. Wilson, RC, 1975. Abronia III. Pericarp and seed coat anatomy and its ecological implications for Nine Species of Abronia Aliso 8:289-299.

PAGE 80

80 Wilson, RC, 1976. Abronia : IV. Anthocarp dispersibility and its ecological implications for nine species of Abronia Aliso 8:493-506. Wolfe, JA and Tanai, T, 1987. System atics, phylogeny, and distribution of Acer (Maples) in the Cenozoic of Western North America. Journa l of the Faculty of Science, Hokkaido University, (Japan) series IV, 22: 1-246. Wurdak. KJ, Hoffmann, P, Samuel, R, de Bruijn, A, van der Bank, M, & Chase, MW, 2004. Molecular phylogenetic analysis of P hyllanthaceae (Phyllanthoideae pro parte, Euphorbiaceae sensu lato) using plastid RBCL DNA sequences. American Journal of Botany 91:1882-1900. Zhuge, R, 1990. On the genus Burretiodendron Se nsu Lato (Tiliaceae). Journal of the Arnold Arboretum 71:371-38.

PAGE 81

81 BIOGRAPHICAL SKETCH Elizabeth OLeary was born in Pittsfield, Ma ssachusetts, in 1981. She attended elementary school through high school in th e public school system of Natic k, Massachusetts and graduated at age 17 in the year 1999. She entered College of the Atlantic in the same year, majoring in human ecology. She graduated from College of th e Atlantic in 2003; her final project was an easy to read book about plant re production. In 2004 and 2005, sh e took several courses through Harvard Extension School and then applied for graduate school at th e University of Florida. She began graduate school in the spring of 2005 w ith a focus in paleobotany, working under the supervision of Steven Manchester She graduated with a M aster of Science degree in 2007.