The Pennsylvania State University

The Graduate School

College of Earth and Mineral Sciences

AUSTRALIAN ANALOGS FOR AN EOCENE PATAGONIAN PALEORAINFOREST

A Thesis in

Geosciences

by

Lisa Merkhofer

Submitted in Partial Fulfillment

of the Requirements

for the Degree of

Master of Science

August 2014

The thesis of Lisa M. Merkhofer was reviewed and approved* by the following:

Peter Wilf Professor of Geosciences Thesis Advisor

Mark Patzkowsky Professor of Geosciences

Timothy Bralower Professor of Geosciences

Lee Kump Professor of Geosciences Head of the Department of Geosciences

*Signatures are on file in the Graduate School.

ii

ABSTRACT

The diverse Laguna del Hunco (LH) paleoflora from early Eocene Patagonia, Argentina, has striking similarities to subtropical and tropical Australian rainforests. Previous research recognized the Simple Notophyll Vine Forests (SNVFs) of montane New South Wales, Australia, as potential analogs for the paleoflora in terms of diversity, floristic composition, leaf size, and environment. In this study, I test this hypothesis by: (1) inferring the Laguna del Hunco rainforest type; (2) quantitatively comparing the paleoflora to 596 Australian rainforest plots; and (3) comparing these results to non-Australian regions in Australasia and Southeast Asia where Laguna del Hunco’s nearest living relatives also occur. First, I inferred rainforest type using paleofloristics and fossil dicot leaf area. Fossil leaf area was measured directly or estimated with the Cain and Castro formula, the Raunkiaer-Webb size classes, or vein scaling, a new method that has not yet been applied to fossils and uses a scaling relationship between leaf area and secondary vein density. By testing all three methods on 159 fossil leaves with intact areas, I found that vein scaling was as accurate at predicting leaf area as the Raunkiaer-Webb size classes, but more applicable to fragmentary leaves. When I used vein scaling to reconstruct the areas of 94 fragmented specimens from LH, the paleoflora’s grand mean leaf area increased by ~360 mm2, recovering large leaf areas that were previously undetected. Across 1152 fossil leaves representing 154 dicot species, Laguna del Hunco’s mean leaf size was 1755 mm2, or large microphyll. The paleoflora’s leaf size index and floristic composition were found to support its similarity to an SNVF. Secondly, I found that subtropical rainforests with moderate, but not montane elevations, were the closest Australian analogs for LH both in terms of leaf size, family composition, and the generic occurrences of Laguna del Hunco’s nearest living relatives (NLRs). Lastly, I found that Laguna del Hunco’s NLRs occurred in three non-overlapping climate regions: cool-dry areas in subtropical Australia, hot-wet areas in tropical Australia, and cool-wet areas not found in Australia, but in montane Australasia and Southeast Asia. These results suggest that Australia no longer has the cool, wet montane environment needed to support some of the lineages from the paleoflora. This study uses a novel, quantitative method of fossil-modern comparison that can be applied to other paleofloras, allowing paleoecological interpretations to be more precisely based on both taxon-free and taxon-informed data.

iii

TABLE OF CONTENTS LIST OF TABLES ...... V! LIST OF FIGURES ...... VI! ACKNOWLEDGMENTS ...... VII! INTRODUCTION ...... 1! Leaf area in paleorainforest reconstructions ...... 1! The Laguna del Hunco paleoflora ...... 4! Australian rainforests as modern analogs ...... 5! MATERIALS AND METHODS ...... 8! Laguna del Hunco setting ...... 8! Paleofloristics and nearest living relatives ...... 9! Fossil specimens and repository ...... 13! Dicot leaf area analysis ...... 13! Australian rainforest dataset ...... 18! Analog rainforest analysis ...... 20! Climate space analysis ...... 22! RESULTS ...... 25! Accuracy of the vein scaling method ...... 25! Laguna del Hunco dicot leaf size ...... 25! Leaf size comparisons ...... 29! Floristic comparisons ...... 33! Climate space comparisons ...... 36! DISCUSSION ...... 39! Including fragmented leaves in fossil leaf area measurements ...... 39! Assessing taphonomic bias in leaf area ...... 40! Laguna del Hunco as a Simple Notophyll Vine Forest ...... 42! Closest analog rainforests ...... 43! Australian climate spaces ...... 45! CONCLUSIONS ...... 47! LITERATURE CITED ...... 49! APPENDIX A. SUPPLEMENTAL LEAF AREA METHODS AND RESULTS ...... 60! APPENDIX B. ADDITIONAL FOSSIL-MODERN COMPARISON RESULTS ...... 95!

iv

LIST OF TABLES

Table 1. Taxa from Laguna del Hunco paleoflora used in floristic analyses ...... 10! Table 2. Breakdown of fossil specimens used for Laguna del Hunco leaf area ...... 17! Table 3. Geography, environment, and species diversity of living Australian rainforests ...... 19! Table 4. Climate ranges of selected nearest living relatives of fossil species ...... 22! Table 5. Taxon-specific leaf area for Laguna del Hunco and Australian nearest living relatives 31! Table 6. Living Australian rainforests with the highest number of Laguna del Hunco’s nearest living relative genera ...... 37!

v

LIST OF FIGURES

F igure 1. Measurement technique for estimating leaf area vein scaling ...... 15! Figure 2. Evaluating methods for measuring fossil leaf area ...... 26! Figure 3. Accuracy of the vein scaling method ...... 27! Figure 4. Laguna del Hunco fossil dicot leaf areas...... 28! Figure 5. Leaf areas of Laguna del Hunco fossils and Australian nearest living relatives ...... 30! Figure 6. Grand mean leaf areas of Laguna del Hunco and living Australian rainforests ...... 32! Figure 7. Family compositions of Laguna del Hunco and living Australian rainforests ...... 34! Figure 8. Occurrences of Laguna del Hunco’s Australian nearest living relative genera ...... 35! Figure 9. Climate spaces of Laguna del Hunco’s nearest living relatives ...... 38!

vi

ACKNOWLEDGMENTS

I owe many thanks to my advisor Dr. Peter Wilf for his unflagging guidance throughout the course of this project. The many hours he devoted to commenting on drafts, ensuring that I had all necessary resources, and interpreting troublesome results was invaluable, as was his ever- present jovial encouragement. I would also like to thank to my committee members, Dr. Mark

Patzkowsky and Dr. Timothy Bralower, for their multiple rounds of helpful feedback and assistance with data analysis. Thanks also to Dr. Robert Kooyman (Macquarie University,

Australia), who provided me access to top-quality data. I am also very grateful to Dr. Lawren

Sack and Christine Scoffoni (UCLA) for patiently coaching me on how to measure leaf veins.

Thanks to Dr. Dana Royer (Wesleyan University) for his constructive advice and help finding essential data. I would especially like to thank Dr. Rubén Cúneo, Pablo Puerta, Eduardo

Ruigómez, Dr. Ignacio Escapa, Laura Reiner, and staff from the Museo Paleontológico Egidio

Feruglio, in Trelew, Argentina, for allowing me to visit the fossil collections, lending photography equipment, and locating and prepping specimens. Funding for this project came from the Penn State Geosciences P.D. Krynine Memorial Fund and NSF Award DEB-0919071 to Peter Wilf.

vii

INTRODUCTION

Strong similarities exist between Gondwanan rainforest paleofloras and living subtropical and tropical rainforests. Molecular phylogenetics suggest that the majority of plants in southern- hemisphere rainforests have maintained ancestral biomes over tens of millions of years (Crisp et al., 2009). Cenozoic paleofloras from South America and Australia indicate that some ancient rainforests contained genera that are extant in living rainforests, frequently with the same floristic associations (Hill, 2004; Carpenter, 2012; Wilf et al., 2009; Wilf, 2012; Knight and

Wilf, 2013; Carvalho et al., 2013). However, little is known about how ancient and living rainforests compare ecologically. Because rainforests have high species diversity, it is difficult to compare them in terms of dominant species. Instead, the different types of rainforest habitat are mainly differentiated by structural features, such as the spacing of trees, the height of the canopy, and the quality of epiphytes, vines, and lianas (Webb, 1959, 1968, 1984; Richards, 1996).

Therefore, in order to make comparisons to living rainforests, both the structural and floristic characteristics of Gondwanan paleorainforests need to be taken into account.

Leaf area in paleorainforest reconstructions

Leaf area is a useful tool for inferring rainforest structure and ecology. In living rainforests, woody angiosperm leaf area correlates in predictable ways with climate and environmental conditions. Leaves become smaller with increasing latitude (Bailey and Sinnott, 1916), decreasing mean annual temperature (Webb, 1968; Greenwood, 1992), declining concentrations of soil nutrients (Hovenden and Schoor, 2004), increasing elevation (McDonald et al., 2003;

Christophel and Gordon, 2004; Hovenden and Schoor, 2004), and decreasing rainfall (Givnish,

1984; Richards, 1996; Wilf et al., 1998). The global occurrence of these trends suggests that leaf

1

size reflects the consistent trade-offs of particular environmental conditions (Wolfe, 1993; Peppe et al., 2011).

In a series of classic studies, Webb (1959, 1968, 1984) demonstrated that Australia’s different rainforest habitats, or rainforest types, could primarily be distinguished by the dominant leaf size of the canopy. Webb grouped leaf size into the classes microphyll (225 to 2025 mm2), notophyll (2026 to 4500 mm2), or mesophyll (4501 to 18225 mm2), creating the “notophyll” class for the numerous small mesophyll leaves he found consistently associated with particular environments in Australia. Webb determined four main rainforest types in Australia: Complex

Mesophyll Vine Forest, Complex Notophyll Vine Forest, Simple Notophyll Vine Forest, and

Microphyll Mossy Forest. Webb’s rainforest types refer to specific structural and leaf size features. For example, a Simple Notophyll Vine Forest characteristically has a majority of notophyll and microphyll leaves, occasional vines and epiphytes, widely spaced trunks, sparse tree crowns, and an even canopy level with emergent Araucaria.

Using a paleoecological technique that is important to the present work, Greenwood (1994) applied the correlation between dominant canopy leaf size and rainforest type to classify several

Australian Paleocene paleofloras. By measuring leaf litter, Greenwood determined the signature range of leaf sizes of each Australian rainforest type. Then, he used leaf size and floristics to match paleofloras to the most similar living rainforest types and inferred their associated ecological characteristics.

Greenwood’s method requires accurate measurements of fossil leaf size. However, measuring fossil leaf area is complicated by the fact that most leaf specimens are fragmented.

The intact leaf areas of fragmented leaves can be estimated using two standard methods, the

Raunkiaer-Webb leaf size classes or the Cain and Castro formula (1959). The first

2

accommodates some uncertainties in leaf area by categorizing leaves into discrete groups

(Raunkiaer, 1934; Webb, 1959). The Cain and Castro (1959) formula calculates intact size following the simple equation: leaf area = 2/3 × leaf length × leaf width. However, both methods require leaves that are mostly intact and cannot be applied to even moderately fragmented leaves.

This problem may be resolved by a new technique that utilizes a scaling relationship between leaf area and leaf secondary vein density that was observed globally across living dicot leaves (Sack et al., 2012). The scaling relationship corresponds to a model of leaf development based on living Arabidopsis. In this model, secondary (2º) veins form during the transition from growth through cell division to growth through cell expansion. Any increase in leaf size beyond this stage pushes 2º veins apart, decreasing 2º vein density and, therefore, allowing 2º vein density to be used to predict leaf area in a nearly consistent manner across taxa. Unlike the other methods of estimating leaf area, vein scaling has a physiological basis and is valid across angiosperm phylogeny, making it more likely to be applicable to Cenozoic leaves (Sack et al.,

2012). If leaf area has the same scaling relationship with 2º vein density in fossil leaves, then the otherwise unmeasurable areas of fragmented leaves can be estimated. Because fragmented leaves are so common in paleofloras, including them in paleofloral leaf area measurements greatly increases the sample size and may potentially recover the areas of large leaves that are more prone to fragmentation.

Studies on modern leaf litter suggest that some additional taphonomic considerations are needed for comparing fossil leaf sizes to those of living rainforests. In Australian rainforests, the most common size of leaves collected from forest floor leaf litter can be up to one leaf size class smaller than that of the surrounding canopy (Greenwood, 1992). Leaf litter that is collected from

3

lakes tends to over-represent the smaller, tougher sun leaves of the outer forest canopy

(Burnham, 1989, 1992; Greenwood, 1992). The bias against large leaves is further exaggerated with leaf collection distance from the lakeshore and river transport (Roth and Dilcher, 1978;

Spicer, 1981; Hill and Gibson, 1986; Greenwood, 1992), mainly due to leaf entrapment and fragmentation (Steart et al., 2002). For these reasons, paleofloral leaf areas are usually considered underestimates of those of the source rainforest.

The Laguna del Hunco paleoflora

The early Eocene Laguna del Hunco (LH) paleoflora, from Chubut Province, Argentina

(Berry, 1925), has an increasing number of identified nearest living relatives (Wilf et al., 2003;

2005; 2009; 2014, González et al., 2007; Hermsen et al., 2012; Wilf, 2012; Knight and Wilf,

2013), making robust paleoecological reconstruction and detailed comparisons to living rainforests that have the same lineages possible.

The Laguna del Hunco paleoflora (52.2 Ma) records southern mid-latitude vegetation during the Early Eocene Climatic Optimum, when global temperatures were significantly higher than today (Zachos et al., 2001). During the early Eocene, Antarctica had not yet separated from

South America by the opening of the Drake Passage, nor from Australia by the opening of the

Tasman Strait (Scher and Martin, 2006; Lawver et al., 2011). In this period, Antarctica had large areas of rainforest that were floristically similar to living tropical and subtropical Australasian rainforests (Pross et al., 2012). Evidence of the biotic interchange of both plants and animals existed between South America, Antarctica, and Australia until at least the early middle Eocene

(Wilf et al., 2013). As a result of this interchange and extinction in Patagonia, Laguna del Hunco has many nearest living relatives that are only present in the tropical and subtropical rainforests of Southeast Asia and Australasia (Berry, 1938; Romero and Hickey, 1976; Gandolfo et al.,

4

1988; Wilf et al., 2003, 2005, 2013, 2014; Zamaloa et al., 2006; Wilf, 2012). For example, the tree genera Akania and Agathis are known as fossils from LH (Romero and Hickey, 1976; Wilf et al., 2014), but today Akania is restricted to Australia, and Agathis is found from Sumatra to

New Zealand, including Australia.

Many of the fossil species identified at LH are also found in the middle Eocene Río

Pichileufú paleoflora (47.7 Ma), from Río Negro, Argentina, suggesting that LH was a part of an extensive, long-lasting rainforest biome (Berry, 1938; Wilf et al., 2005; Wilf, 2012).

Reconstructing the Laguna del Hunco paleorainforest will improve our understanding of a highly diverse, mid-latitude rainforest ecosystem that is extinct today.

Australian rainforests as modern analogs

Australia has a particularly high number of the nearest living relatives of the Laguna del

Hunco paleoflora, including several genera that are extant only in Australia or greater Australasia and Southeast Asia. These include the conifers Agathis (Wilf et al., 2014), Papuacedrus (Wilf et al., 2009), Dacrycarpus (Wilf, 2012), and Acmopyle (Florin, 1940; Wilf et al., 2009; Wilf, 2012).

Additionally, the fern Todea (Carvalho et al., 2013) and the angiosperms Akania (Romero and

Hickey, 1976), Eucalytpus (Gandolfo et al., 2011; Hermsen et al., 2012), and Ceratopetalum

(Gandolfo and Hermsen, 2012) follow the same pattern. Like Laguna del Hunco, Australian rainforest floras also have high species diversity and commonly include the families Proteaceae,

Atherospermataceae, Lauraceae, Monimiaceae, and Cunoniaceae that are known from the fossil site (Table 3; Wilf et al., 2009). For these reasons, living Australian rainforests were chosen for comparisons with Laguna del Hunco because they are better studied (Webb, 1959, 1968; Royer et al., 2009; Kooyman et al., 2011, 2012) than the most comparable rainforests in other regions of Australasia and Southeast Asia.

5

Several conifers known from Laguna del Hunco have a fossil record in Australia, but living ranges in Australasia and Southeast Asia. Decreasing rainfall in Australia is known to have played a significant role in the extinction of the ex-Australian genera Papuacedrus,

Dacrycarpus, Retrophyllum, and Acmopyle (Hill and Brodribb, 2003). Rainforest covered most of Australia in the Eocene and early Oligocene (Greenwood and Christophel, 2005). However, by the mid-Miocene Australia experienced drying evidenced by increasing fossil macro- and microfloras with affinities to extant dry, open vegetation (Macphail and Truswell, 1989) and oxygen isotopes characteristic of decreasing rainfall within offshore ocean cores (Shackleton and

Kennett, 1975). Therefore, rainforests in Southeast Asia and Australasia, outside of Australia, are also discussed in this study as analogs for Laguna del Hunco.

The Simple Notophyll Vine Forests of New South Wales, Australia, have been recognized as potential analogs for the paleoflora based on similar high species diversity, family composition, and leaf area (Wilf et al., 2009; Carvalho et al., 2013). Some montane representatives of these rainforests grow on steep slopes with volcanic soils that are somewhat equivalent to Laguna del Hunco’s volcanic caldera paleoenvironment and, therefore, have been noted as the closest Australian analogs for the Laguna del Hunco paleorainforest (Wilf et al.

2003; 2005; Carvalho et al., 2013). However, the differences between LH and living Australian rainforests have not been quantitatively assessed.

In this study, I test the hypothesis that the closest analogs for Laguna del Hunco within

Australia are subtropical montane SNVFs. To this purpose, I first infer Laguna del Hunco’s rainforest type by utilizing floristics and leaf size. Then, I quantitatively compare the paleoflora to Australian rainforest plots using leaf size, family composition, and the occurrences of eleven genera of Laguna del Hunco’s nearest living relatives. Lastly, I compare these results to non-

6

Australian regions in Australasia and Southeast Asia where Laguna del Hunco’s nearest living relatives also occur.

7

MATERIALS AND METHODS

Laguna del Hunco setting

The Laguna del Hunco paleoflora was collected within the Tufolitas Laguna del Hunco of the Middle Chubut River volcanic-pyroclastic complex in northwest Chubut, Argentina,

Patagonia (Aragón and Romero, 1984; Aragón and Mazzoni, 1997). Three primary ash-fall tuffs stratigraphically associated with the fossil beds have 40Ar-39Ar dates of 51.8 ± 0.2, 52.8 ± 0.16, and 52.1 ± 0.50 Ma (Wilf et al., 2003, 2005). One of these was reanalyzed and yielded a revised sanidine date of 52.2 ± 0.22 Ma, in the early Eocene (Ypresian; Wilf et al., 2005; M. Smith in

Wilf, 2012). Radiometric dates are supported by six magnetic reversals within the Tufolitas that place the fossil-bearing beds near the base of magnetic polarity Chron 23 (Wilf et al., 2003,

2005). The inferred paleolatitude of Laguna del Hunco is ~47ºS, nearly similar to the present day latitude (Wilf et al., 2003).

The tuffaceous mudstones and sandstones of the Tufolitas contain impression-compression fossils of angiosperms and conifers as well as smaller amounts of cycads, palms, ferns, and ginkgophytes (Wilf et al., 2005). Most fossils were within 50 m of stratigraphic section (Wilf,

2003). Fossil caddisfly cases (Fidalgo and Smith, 1987; Genise and Petrulevičius, 2001;

Petrulevičius and Nel, 2003), frogs (Casamiquela, 1961; Báez and Trueb, 1997), and fish

(Azpelicueta and Cione, 2011) have also been found in the same beds as fossil plants.

The depositional environment of the Laguna del Hunco paleoflora is interpreted to be a volcanically active caldera lake with a maximum length of 25 km (Aragón and Romero, 1984;

Aragón and Mazzoni, 1997). The paleoflora includes exceptionally well-preserved specimens of delicate vegetation, like Eucalyptus flower buds (Gandolfo et al., 2011; Hermsen et al., 2012 ), ferns with in situ sori (Carvalho et al., 2010), large intact leaves (up to 12500 mm2 in area),

8

articulated compound cycad leaves (up to 75 cm long), and possible Papuacedrus seedlings

(Wilf et al., 2012). The presence of fragile plant organs, as well as fragmented leaves, suggests that local and regional vegetation were mixed within the caldera lake prior to burial.

Laguna del Hunco’s paleoclimate is similar to that of living montane tropical and subtropical rainforests in Australasia (Wilf, 2012). Laguna del Hunco is inferred to have had a mean annual temperature of 16.9°C based on regional digital leaf physiognomy (Peppe et al.,

2011). This estimate is within the 16-17°C range of coeval sea surface temperatures inferred from multiple Atlantic cores at Laguna del Hunco’s bounding paleolatitudes (Zachos et al.,

1994). Further, several extant taxa indicative of a warm subtropical climate are known from LH, including cycads and palms (Wilf et al., 2003, 2005). The mean annual precipitation of Laguna del Hunco is thought to have been wet (>2500 mm) due to the presence of the genera Acmopyle,

Podocarpus, Dacrycarpus, and Gymnostoma that today have demonstrated physiological limitations to drought (Wilf et al., 2003, 2005, 2009; Zamaloa et al., 2006; Wilf, 2012).

Paleofloristics and nearest living relatives

The Laguna del Hunco paleoflora is unusually diverse, with over 215 morphotypes of leaves and/or reproductive organs (Wilf et al. 2005), including six newly discovered leaf morphotypes

(Appendix A). This analysis uses all dicot leaf species known from the paleoflora for leaf area measurements, and fifteen species that have identified nearest living relatives in Australasia and

Southeast Asia (Table 1). As used here, “Southeast Asia” includes Myanmar (Burma), Thailand,

Laos, Vietnam, Cambodia, Singapore, Philippines, Brunei, Malaysia, Borneo, Indonesia, and

East Timor. “Australasia” includes Australia, New Guinea, New Zealand, and neighboring

Pacific islands, including Fiji and New Caledonia.

9

Table 1. Taxa from Laguna del Hunco paleoflora used in floristic analyses

Fossil species Organ(s) Source NLR(s) Acmopyle engelhardti (Berry) L Florin, 1940; Wilf et al., 2009; Wilf, A. pancheri (Brongn & Florin 2012 Gris) Pilg.† Agathis zamunerae Wilf L, S, SC, Wilf et al., 2014 A. sahniana Buchholz & PC N.E. Gray†

Akania patagonica Gandolfo, L Romero and Hickey, 1976; A. atropurpurea Hyland Dibbern and Romero Gandolfo et al., 1988 Araucaria pichileufensis Berry L, S Berry, 1938 A. lenticula de Laub† Atherospermophyllum guinazui L Knight and Wilf, 2013 A. microstachya J.F. C.L. Knight Bailey & C.T. White

Ceratopetalum sp. S Gandolfo and Hermsen, 2012 A. robusta (C. Moore ex F. Muell.) F.M. Bailey cf. Caldcluvia L Wilf et al., in prep A. bidwillii (Hend. ex Hogg) Mabb. Cochlospermum previtifolium S Berry, 1935, 1938; Wilf et al., 2005; A. cunninghamii Mudie Berry Gandolfo et al., 2007; González, 2009 Dacrycarpus puertae Wilf L, SC PC Wilf, 2012 Daphnandra apatela Schodde* Eucalyptus frenguelliana L, S, F Gandolfo et al., 2011; Hermsen et D. repandula F. Muell. Gandolfo and Zamaloa al., 2012

Monimiophyllum callidentatum L Knight and Wilf, 2013 D. tenuipes G. Perkins C.L. Knight

Orites bivascularis Romero, S Romero et al., 1988; González et C. apetalum D. Don Dibbern and Gandolfo al., 2007 Papuacedrus prechilensis Wilf L, SC Wilf et al., 2009 C. corymbosum C.T. White Podocarpus andiniformis Berry L Berry, 1938; Wilf, 2012 C. succirubrum C.T.White

Retrophyllum sp. (undescribed) L Wilf, 2012 C. virchowii F.Muell. Wilf

Notes: The following fossil species are members of the Laguna del Hunco paleoflora with the same or closely related genera extant in Australasia or Southeast Asia. Organs refer to the fossil leaves (L), fruits and/or seeds (S), seed cones (SC), pollen cones (PC), and/or flowers (F) used to describe each fossil species. The selected nearest living relatives (NLRs) are of either the same or the most morphologically similar living genus as that identified for fossil species. All NLRs are species of genera that occur in Australia, or if the genus is extinct in Australia, Australasia and Southeast Asia. *identified as the most similar living species in the listed source(s); †Genus extinct in Australia, but extant in other regions of Australasia

10

Most of the dicot species from Laguna del Hunco with nearest living relatives in Australia are extant only in Australia or greater Australasia (Table 1, Table 4). Akania patagonica

Gandolfo, Dibbern, and Romero (Akaniaceae) has a genus with only one living species that is limited to subtropical Australia (Romero and Hickey, 1976; Gandolfo et al., 1988).

Atherospermophyllum guinazui C.L. Knight (Atherospermataceae) shares numerous leaf characteristics with the living Australasian genus Daphnandra, and it is most similar to the

Australian species D. apatela (Knight and Wilf, 2013). Similarly, Monimiophyllum callidentatum C.L. Knight (Monimiaceae) has the most shared leaf characters with the living

Australian genus Wilkiea, particularly the species W. hugeliana (Knight and Wilf, 2013). The iconic Australian and Australasian genus Eucalyptus (Myrtaceae) is known from LH from over one hundred leaves of the fossil species E. frenguelliana Gandolfo and Zamaloa (Gandolfo et al.,

2011; Hermsen et al., 2012). A species of Caldcluvia (Cunoniaceae) was identified at LH from a combination of several fossil leaf features, including compound leaf arrangement, leaf shape, and the locations of domatia (hair tufts in secondary vein axils) that are distinctive of this genus (Wilf et al., in prep.). Caldcluvia has a larger modern range than the genera so far listed; it occurs in

Australasia, South America, and Southeast Asia.

Three dicot fossil species used in this analysis are known from LH as fruits but not as leaves. Ceratopetalum (Cunoniaceae) was identified from fossil fruits with sepal venation that is distinctive of this genus, only found in modern Australasia (Gandolfo and Hermsen, 2012). The species Cochlospermum previtifolium Berry (Cochlospermaceae) has living relatives of the same genus in tropical Australia, Africa, and the Americas (Wilf et al., 2005; Gandolfo et al. 2007;

González, 2009; Berry 1935, 1938). Orites bivascularis Romero, Dibbern and Gandolfo

11

(Proteaceae), also from LH, has a genus that is extant in temperate Australia as well as southern

South America (Romero et al., 1988; González et al., 2007).

The quantitative comparisons in this study do not include the genus Gymnostoma

(Casuarinaceae) that is also known from LH as leaf and inflorescence species (Zamaloa et al.,

2006). Gymnostoma is unusual among dicot angiosperms; it has photosynthetic branchlets and greatly reduced leaves and consequently was not included in dicot leaf size measurements.

Today Gymnostoma is found in regions with other Laguna del Hunco lineages, such as tropical

Australia, Fiji, New Caledonia, and Malesia (i.e., Papua New Guinea, Indonesia, Malaysia,

Brunei, the Moluccas, and the Philippines). However, detailed data on Australian Gymnostoma occurrences were also unavailable.

In addition to the fossil dicots described above, this analysis used three fossil conifer species with genera extant in Australia. Araucaria pichileufensis Berry (Araucariaceae) is section

Eutacta, which today is restricted to Australia and New Caledonia (Berry, 1938). Agathis zamunerae Wilf (Araucariaceae) has a genus that is extant in Australia and Malaysia. Fossil

Agathis is most similar to the living Malaysian species A. lenticula, in terms of leaf, pollen cone, and seed-cone scale morphology (Wilf et al., 2014). The broad-leafed fossil conifer Podocarpus andiniformis Berry (Podocarpaceae) is of a living genus found widely across Africa, Asia, and the Americas, with three species extant in tropical Australia (Berry, 1938).

Fossil species of ex-Australian genera that are known from the Australian fossil record but have living regions outside of Australia, in Australasia or Southeast Asia, were included in this study for comparing climate spaces. Acmopyle engelhardti (Berry) Florin and Dacrycarpus puertae Wilf are both podocarp genera (Florin, 1940; Wilf et al., 2009; Wilf, 2012). Acmopyle is extant only in Fiji and New Caledonia, and Dacrycarpus is extant in Southeast Asian and

12

Australasian rainforests. Papuacedrus prechilensis Wilf et al. (Cupressaceae) has a genus with only one living species extant in montane Papua New Guinea and the Moluccas (Wilf et al.,

2009).

Fossil specimens and repository

Laguna del Hunco fossil material was studied at the Museo Paleontológico Egidio

Feruglio (MEF) in Trelew, Chubut, Argentina. Specimens were collected during several field seasons from 1999 to 2009 and are curated at the MEF under field collection numbers or the repository prefix MPEF-Pb (Wilf et al., 2003, 2005; Wilf 2012).

Dicot leaf area analysis

Dicot fossil leaves were measured digitally from high-resolution photographs, using a light microscope as needed for reference. Photographs were taken with a Nikon D90 camera or chosen from an image library of specimens digitally extracted from the matrix by B. Cariglino

(2007) and later used by Peppe et al. (2011). If necessary, Adobe Camera Raw was used to adjust image contrast and brightness.

I analyzed 159 dicot specimens with intact leaf areas to test whether fossil leaf area scales with 2º vein density as it does in living dicot leaves (Sack et al., 2012). Intact leaf specimens were leaves or leaflets chosen for 2º vein preservation, representing 76 fossil species. Leaves were considered intact if no more than 1 cm of margin was missing. I measured leaf areas digitally (Image J; National Institute of Health, Bethesda, MD, USA) by tracing the margin of each leaf to the point of petiole attachment. If a specimen had both a part and counterpart, only one was measured. The studied intact leaves covered a wide range of areas, 67 to >8,000 mm2

(nanophyll to mesophyll).

13

Primary (1º) vein width, 1º vein density, and 2º vein density were recorded to assess their scaling with leaf area (see Appendix A for additional methods). Secondary veins were identified with reference to the Manual of Leaf Architecture (Ellis et al., 2009). I measured 2º vein density as the total vein length per area for all 2º veins including intersecondaries, minor secondaries, and interior secondaries. Secondary vein density was also subsampled in four rectangular areas along one half of the leaf lamina (Fig. 1). These regions were in the center of the basal, middle, and apical thirds of the leaf, and adjacent to the midvein in the middle third of the leaf. If a specimen was poorly preserved in one of these regions, that region was not measured.

Subsampled areas were sized to include at least two 2º veins and had a mean area of 1.4 cm2 with a standard deviation of 0.11 cm2. Subsampled 2º vein density was averaged across all subsampled areas for each specimen. I used subsampled 2º vein density to estimate intact leaf area with the following equation derived from living leaves (Sack et al., 2012):

2 2 Log10 (leaf area cm )=1.96-2.04×Log10 (subsampled 2º vein density cm/cm ) [1]

A fossil-specific vein scaling equation was also determined using ordinary linear regression fitted to the measured intact fossil leaf areas and 2º vein density:

2 2 Log10 (leaf area cm ) = 1.51-1.31×Log10 (subsampled 2º vein density cm/cm ) [2]

I tested the modern and fossil scaling equations for variation in slope and found no significant difference (F (1, 314)=0.410, P =0.52). Because the Sack et al. (2012) regression

(equation 1) describes a broader dataset of living leaves, it was chosen for application to fragmented fossil leaves from LH.

The Cain and Castro (1959) formula and Raunkiaer-Webb leaf size classes (Raunkiaer

1934; Webb 1959) were also used to estimate leaf areas for all the intact leaves studied for comparison to vein scaling estimates. Leaf size classes were used to estimate leaf area by taking

14

Figure 1. Measurement technique for estimating leaf area vein scaling Secondary vein density was measured across 4 rectangular regions (small rectangles) of the leaf lamina divided into thirds (large rectangles). Three of these regions were chosen from the center of each leaf third on one half of the leaf, and the forth region was measured in the center of the middle third adjacent to the midvein. The fossil shown is a Cunoniaceae leaflet, Caldcluvia sp., morphotype TY116, and specimen number LH02-1086. Leaf is 5.7 cm in length.

15

the average of the natural logarithm transformed upper and lower bounds of the size class of each specimen, following the methods of Wilf et al. (1998). For the Cain and Castro (1959) formula, leaf length and width were quantified using the rectangular method, in which length is measured as the distance between leaf apex and base at the midvein, and width is measured at the widest point perpendicular to leaf length. Areas were converted to ln mm2 units in all analyses.

In addition to the leaf area measurements completed for testing the vein scaling method, 224 new, high-resolution leaf area measurements were made of Laguna del Hunco fossil leaves of known species. Each specimen was measured using the most accurate method appropriate for fossil completeness, either by tracing leaf margin, using leaf size classes, or applying vein scaling (Table 2). Fossil specimens were classified as intact, largely intact, or fragmented.

Largely intact leaves were defined as having identifiable length and width, but with more than 1 cm of the leaf margin missing. The areas of largely intact leaf areas were estimated digitally, by tracing the leaf perimeter with reference to other specimens of the same species. Fragmented leaves were classified by having unidentifiable dimensions of length and width. I measured 120 fragmented leaves that varied in size from 90 to 31571 mm2 (4.50 to 10.36 ln mm2), larger on average than intact fossil leaves. Fragmented leaf areas were reconstructed using the vein scaling method. In 26 fragmented leaves, vein-scaling significantly underestimated leaf area, yielding estimates that were smaller than the original, unreconstructed fragment areas. In these cases, area was instead recorded using estimates from leaf size classes.

16

Table 2. Breakdown of fossil specimens used for Laguna del Hunco leaf area

Leaf completeness (n) Source Method of measurement (n) Complete (758) Wilf et al. 2003, 2005 Cain and Castro formula (549) This study Direct (164) Largely complete (319) This study Direct (41) Wilf et al. 2003, 2005 Size classes (278) Fragment (120) This study Vein scaling (94) This study Size classes (26) Total: 1152 Notes: All 1152 fossil leaf specimens used for leaf area measurements are described by leaf completeness, the level of fragmentation, and method of measurement. See Materials and Methods for method and leaf completeness definitions; See Appendix A for all specimen leaf area measurements; n is the number of specimens in each category.

All measurements were compiled with previously published leaf area data for Laguna del

Hunco dicots (Wilf et al., 2003, 2005). Wilf and colleagues obtained these areas by using the

Cain and Castro (1959) formula on complete leaves and the size class method on largely complete leaves. The resulting dicot leaf area database described 1152 Laguna del Hunco leaf specimens, representing 154 leaf species, with sizes ranging from 53 to 18398 mm2 (3.98 to 9.82 ln mm2), leptophyll to macrophyll.

Species mean leaf areas were calculated as the average of the natural log of the largest and smallest leaf of each species. In order to make comparisons to living rainforests, Laguna del

Hunco’s leaf size index (LSI) was calculated using the following formula:

LSI = (m + 2n + 3M – 100)/2 [3]

Where m = % microphyll, n = % notophyll, and M = % mesophyll, by dicot species (Wolfe and Upchurch, 1978; Burnham, 1989), and where microphylls are defined as less than 2025 mm2; notophylls are between 2025 and 4500 mm2; and mesophylls are larger than 4500 mm2.

17

Australian rainforest dataset

Laguna del Hunco leaf areas and floristics were compared to those of living Australian rainforests to identify the closest analog plots. A detailed, ecological dataset of mature woody plants compiled by Kooyman et al. (2012) was used to characterize living Australian rainforests.

This dataset describes 1137 species from 95 families, excluding all plants less than 1 m in height and vines, palms, and ferns. In addition, the dataset describes 596 plots with areas that varied from 0.1 to 0.5 ha (1000 – 5000 m2), defined by the space of one tree and its ~30 nearest neighbor trees from the canopy or subcanopy. This area is reported to be sufficiently detailed to record forest structure (Kooyman et al., 2012).

Australian plots are from tropical latitudes in Queensland and subtropical latitudes in New

South Wales. Sampled plots formed five geographic regions: Cape York and the Wet Tropics from the tropics, and the Nightcap-Border Ranges, Washpool, and Dorrigo from the subtropics

(Table 3). These regions span more than twenty degrees of latitude and cover a wide range of environments. Tropical regions had mean annual temperatures that are on average five to ten degrees Celsius higher than subtropical regions. Mean annual rainfall was highest in tropical regions; the Wet Tropics had the highest average rainfall (2432 mm), and Washpool had the lowest average (1159 mm). The Nightcap-Border Ranges had the highest species diversity when averaged across plots, with up to 117 species recorded from a single plot (Table 3). When species diversity was averaged by region, the two tropical regions were the most diverse.

18

Table 3. Geography, environment, and species diversity of living Australian rainforests

Region n Latitude (ºS) MAT (ºC) MAP (mm) Elevation (m) Species diversity CY 140 15.9 to 10.7 25.3 (22-26) 1603 (1022-2041) 87.2 (1-500) 35.12 (9-60); 650 WT 146 19.5 to 15.5 22.0 (18-24) 2432 (913-4170) 452.6 (3-1500) 39.23 (4-80); 436 NB 140 28.2 to 29.0 17.1 (14-19) 1562 (966-2197) 427.2 (65-1036) 45.25 (24-117); 288 W 43 29.1 to 29.6 15.3 (13-17) 1159 (1080-1236) 786.3 (285-1125) 29.21 (12-55); 113 D 127 30.0 to 30.8 16.2 (12-18) 1616 (1024-1914) 434.1 (9-1044) 32.56 (19-99); 200 Notes: The latitude, mean annual temperature (MAT), mean annual precipitation (MAT), elevation, and species diversity of Australian rainforest plots are given as means (ranges); totals for each major region. Regions are Cape York (CY) and the Wet Tropics (WT) from tropical latitudes and the Nightcap-Border Ranges (NB), Washpool (W), and Dorrigo (D) from subtropical latitudes; n is the number of 0.1 – 0.5 ha plots in each region. All data are calculated from Kooyman et al. (2012).

The Webb (1959) rainforest types of Australian plots were complied from several sources.

R. Kooyman classified tropical plots (personal communication), and Royer et al. (2009) classified subtropical plots from northern New South Wales. Rainforest types were simplified to the four principle categories: Complex Mesophyll Vine Forest (CMVF), Complex Notophyll

Vine Forest (CNVF), Simple Notophyll Vine Forest (SNVF), and Microphyll Mossy Forest

(MMF; Webb, 1959). The forest types of the remaining 291 plots were inferred using LSI

(Equation 3) calculated from the mean leaf sizes of the dicot species in each plot. Forest type was then inferred using the range of LSI values of canopy leaves of Australian rainforests, where

MMFs are below 20; SNVFs are between 20 and 50; mixed SNVFs/CNVFs are between 51 and

57; CNVFs are between 58 and 65; and CMVFs are above 65 (Greenwood, 1992, 1994).

Laguna del Hunco was compared to Australian rainforests on both a fine scale (within the

596 individual plots) and coarse scale (within the 5 geographic regions). Both scales are included to account for uncertainties in the amount of space and time represented by the paleoflora. For example, the paleoflora’s species diversity is most similar to that of Australian rainforests measured at the regional level (Table 3). Australian regions have total species counts between

113 and 650, similar to the number of species identified from LH. However, Laguna del Hunco’s family composition is most similar to that of Australian rainforests measured at the plot level

19

(Appendix B). I also tested an intermediate scale that grouped geographically close plots, but found that it produced a range of family diversities that was too high for comparison with LH.

Additionally, plot-level comparisons allow for the assessment of detailed ecological variations and rainforest types that are not apparent at coarser scales. By making comparisons at both the plot and regional levels, I can use all available fossil data to find the closest living analog rainforests.

Analog rainforest analysis

Australian species leaf areas were reported by Kooyman et al. (2012) and represent mature leaves or the lateral leaflets of compound leaves. Kooyman and colleagues used the formula: leaf area = leaf length x leaf width x 0.70, so I multiplied their areas with a conversion factor to fit results from the Cain and Castro (1959) formula. The grand mean leaf area of each rainforest plot was calculated as the mean of the natural log transformed species mean areas of all species occurring in that plot.

Taxon-specific leaf areas were compared between eight fossil species and their nearest living relatives in Australia (Table 1). Fossil Araucaria was compared to the one living

Australian species of the same section Eutacta, A. cunninghamii.

Fossil Agathis, Araucaria, and Podocarpus were not included in Laguna del Hunco’s grand mean leaf area measurement because they are not dicots. Fossil areas for these genera were estimated for comparisons to modern species of the same genus using the Cain and Castro (1959) formula. For Agathis, leaf length and width data are from Wilf et al. (2014). For Araucaria, leaf lengths and widths were occasionally averaged across several separate leaves of the same branch because Araucaria’s small needle leaves tended to be occluded by other leaf bases. Extant

Australian conifer leaf areas were from Kooyman et al. (2012).

20

I completed two floristic comparisons to living rainforests, one using family composition and the other using the occurrences of the eleven genera of the paleoflora’s nearest living relatives (NLRs). Similarity in family composition was evaluated using the non-metric multidimensional scaling (NMS) method of ordination. I chose this method because it is known to find environmental gradients in complex ecological data (McCune and Grace, 2002). This method also makes no assumptions about the underlying relationships between variables, which is advantageous because two spatially separate geographic areas were analyzed.

Ordination of Laguna del Hunco and each Australian rainforest was based on the presences and absences of 95 angiosperm and gymnosperm families, 22 of which are known from LH

(Appendix B). Four Australian rainforest plots were excluded because they had identical composition with other plots, incompatible with NMS analysis. Identical plots were geographically adjacent to each other, and their removal did not alter the overall variety of the

Australian rainforest climates and environments analyzed. Because compositional data were presence/absence, no standardizations were necessary. Ordination was completed using the R package ‘vegan’ with Bray-Curtis distance. A three-dimensional solution was chosen with a stress of 18.26 (Appendix B). Ordination dimensions one, two, and three were found to represent

71% of the variance between variables. Dimension one represented the large majority of the total variance, with dimensions two and three each contributing approximately 12% (Appendix B).

Floristic comparisons at the genus-level were made by evaluating the occurrences of eleven genera of Laguna del Hunco’s nearest living relatives in Australian rainforest plots (Table 1).

The eleven genera include three genera that are known from LH as fossil fruits and seeds, but not as leaves.

21

Climate space analysis

Some of Laguna del Hunco’s NLRs are of ex-Australian genera, genera that are extinct in

Australia, but extant in other areas of Australasia and Southeast Asia (Table 1). I inferred the climate spaces of Laguna del Hunco’s NLRs, including ex-Australian genera, to determine the kinds of climates occupied by the surviving Gondwanan lineages (Table 4).

Table 4. Climate ranges of selected nearest living relatives of fossil species

Nearest living relatives n MAT (ºC) MAP (mm) Elevation (m) Extant regions Acmopyle pancheri† 72 20±1 1792±292 566±190 New Caledonia Acmopyle sahniana† 4 22±1 2820±348 530±204 Fiji Agathis atropurpurea 2 19±0 2408±576 1085±21 Tropical Australia Agathis lenticula‡ n/a 19±2 2900±900 1375±325 Malaysia Agathis microstachya 1 20±0 1912±0 790±0 Tropical Australia Agathis robusta 16 22±1 1813±473 609±239 Tropical Australia Akania bidwillii 41 17±1 1659±192 394±218 Subtropical Australia Araucaria bidwillii 1 21±0 1792±0 1030±0 Tropical Australia Caldcluvia australiensis 11 22±2 2156±545 778±250 Tropical Australia Caldcluvia paniculata 127 16±2 1541±318 583±283 Subtropical Australia Ceratopetalum apetalum 72 16±2 1714±304 591±234 Subtropical Australia Ceratopetalum corymbosum 3 23±0 2356±0 627±87 Tropical Australia Ceratopetalum succirubrum 11 20±1 2008±638 925±189 Tropical Australia Ceratopetalum virchowii 3 19±1 3177±628 1067±58 Tropical Australia Cochlospermum gillivraei 10 25±1 1329±270 49±26 Tropical Australia Dacrycarpus cinctus† 55 15±3 2736±511 2391±734 New Guinea Dacrycarpus dacrydioides† 490 11±2 2913±1373 318±248 NZ Dacrycarpus imbricatus† 170 18±4 2569±630 1729±873 SE Asia/Australasia Dacrycarpus kinabuluensis† 3 12±2 2092±5 2934±406 Borneo Dacrycarpus vieillardii† 33 21±1 1655±357 360±254 New Caledonia Daphnandra apatela 96 17±1 1451±308 507±235 Subtropical Australia Daphnandra repandula 27 21±1 2399±809 625±358 Subtropical Australia Daphnandra tenuipes 10 18±1 1861±60 351±202 Subtropical Australia Dacrycarpus compactus† 49 13±3 2940±766 2860±510 New Guinea Eucalyptus acmenoides 43 17±1 1537±279 334±192 Subtropical Australia Eucalyptus campanulata 20 16±1 1410±299 706±122 Australia Eucalyptus grandis 58 17±1 1630±270 285±201 Subtropical Australia Eucalyptus microcorys 107 17±1 1572±281 419±233 Subtropical Australia

22

Nearest living relatives n MAT (ºC) MAP (mm) Elevation (m) Extant regions Eucalyptus pellita 2 25±1 1679±115 253±350 Tropical Australia Eucalyptus pilularis 42 17±2 1701±174 356±224 Subtropical Australia Orites excelsus 92 16±2 1622±512 723±263 Subtropical Australia Orites megacarpa 2 23±0 2356±0 691±86 Tropical Australia Papuacedrus papuana§ 352 15±5 2950±1050 2050±1450 New Guinea Podocarpus dispermus 1 22±0 2645±0 680±0 Tropical Australia Podocarpus elatus 21 18±2 1561±323 422±215 Australia Podocarpus grayae 45 24±2 2135±558 275±291 Tropical Australia Retrophyllum comptonii† 50 20±1 1825±319 603±250 New Caledonia Retrophyllum vitiense† 21 23±3 3285±912 767±679 Malesia Wilkiea angustifolia 15 21±2 2507±535 828±326 Tropical Australia Wilkiea austroqueenslandica 33 18±1 1772±218 355±212 Subtropical Australia Wilkiea Barong 24 22±1 2279±721 485±382 Tropical Australia Wilkiea huegeliana 156 17±1 1596±284 464±281 Subtropical Australia Wilkiea macrophylla 3 17±1 1832±21 324±230 Subtropical Australia Wilkiea McIlwraith 2 24±0 1781±258 360±28 Tropical Australia Wilkiea MtHemmant 4 23±1 2201±712 473±283 Tropical Australia Wilkiea MtMolloy 5 22±0 1813±571 611±105 Tropical Australia Notes: Climate values are means ± one standard deviation for mean annual temperature (MAT), mean annual precipitation (MAP), and elevation, and are based on the occurrences of living species in Australia, or Australasia and Southeast Asia. Genera extant in Australia are only represented by their Australian species even if the genus exists outside of Australia, except for Agathis. Climate range data are calculated from Kooyman et al. (2012) unless otherwise noted; n is the number of occurrences of each species. *identified as the most similar living species (See Table 1 for sources); †data from Biffin et al. (2012); ‡data from Farjon, 2010; Wilf et al. (2014) and this study; §data from Wilf et al. (2012) and this study.

Species climate spaces were determined for fifteen genera from LH, eleven from previous comparative analyses, plus the four ex-Australian genera Acmopyle, Dacrycarpus, Retrophyllum, and Papuacedrus. Climate spaces were characterized in terms of the mean annual temperature

(MAT), mean annual precipitation (MAP), and elevation values calculated from the living occurrences of each species. These values were estimated from species occurrences within

Australia, or within Australasia and Southeast Asia for species of ex-Australian genera. For

Australian species, climate ranges were calculated from the occurrences in individual rainforest plots (Kooyman et al., 2012). Occurrences and climate values for the ex-Australian genera

23

Acmopyle, Dacrycarpus, and Retrophyllum are those reported by Biffin et al. (2012) from regions including Borneo, Malaysia, New Guinea, New Caledonia, Fiji, and New Zealand.

Climate values for Agathis lenticula and Papuacedrus papuana were estimated using georeferenced occurrence data from the Global Biodiversity Information Facility (GBIF; accessed through GBIF Data Portal; data.gbif.org, 2014-2). R statistical software and the packages ‘sp’ and ‘raster’ were used to derive temperature limits from the reported elevation and precipitation limits (Farjon, 2010; Wilf et al., 2012, 2014) using WorldClim (Hijmans et al.,

2005). Mean climate variables were calculated as the average of range endpoints, with one standard deviation estimated as the difference between range endpoints and range means. This method overestimates the range size of Agathis lenticula and Papuacedrus papuana but uses the most detailed climate data available.

24

RESULTS

Accuracy of the vein scaling method

When tested on 159 intact fossil leaves, the vein scaling method estimated fossil leaf areas as accurately as using leaf size classes (R2 = 0. 68 and 0.63, respectively, P < 0.001; Fig. 2A).

Both methods were found to be less accurate than the Cain and Castro formula (R2 = 0. 94, P <

0.001).

Vein scaling tended to overestimate leaf areas, predicting areas with a wider range and higher mean than the other two methods (Fig. 2B). The standard error of individual vein-scaling predictions varied from 2.0 to 2.1 mm2 (0.72 to 0.74 ln mm2) across all intact leaves studied (Fig.

3). No correlation was found between the size of intact leaves and the accuracy of vein scaling method (Fig. 3; Appendix A). For the 135 leaves where 2º vein density could be sampled across all four regions, the apical, middle, basal, and central portions of the lamina, vein scaling estimates were more accurate than those estimated from specimens with fewer sampled regions

(R2 = 0.75, P < 0.001).

Other vein traits were found to scale with fossil leaf area in the same direction as those in living leaves: primary vein diameter scaled positively with leaf area, and primary vein density scaled negatively with leaf area (Appendix A; Sack et al., 2012). In fossil leaves, the strongest area-scaling relationship observed was with total 2º vein density (i.e., when vein density could be scored over the entire leaf lamina); this correlation was as significant as it is for living leaves

(Appendix A; R2 = 0.75, P < 0.001).

Laguna del Hunco dicot leaf size

Although the vein scaling method was applied only to a small fraction of fossil specimens, including their reconstructed leaf areas increased both species mean leaf areas and the grand

25

Figure 2. Evaluating methods for measuring fossil leaf area Three methods of estimating fossil leaf area were tested on 159 complete fossil leaves from Laguna del Hunco, vein scaling (VS), the Cain and Castro formula (CC), and leaf size classes (SC). All estimated areas were compared to the directly measured area of each leaf. A. The correlation between estimated and measured fossil leaf areas; all R2 values are significant at P < 0.001. B. The range of estimated leaf areas relative to measured leaf areas (Direct). Box plot tails go to data extremes; boxes show 1st and 3rd quartiles, bold lines show medians, and bold points show means. No box is shown for the size class boxplot because the 3rd, and 1st quartiles were equal to the median.

26

9 )

2 2 8 R = 0.68 m m 7 6 Leaf area (ln 5

−2.5 −2.0 −1.5 −1.0 −0.5 0.0

Secondary vein density (ln mm/mm2)

Figure 3. Accuracy of the vein scaling method The relationship between the measured leaf area and subsampled 2º vein density is shown for 159 fossils with intact leaf areas. The black line indicates the leaf areas predicted from the Sack et al. (2012) scaling equation based on subsampled 2º vein density. Dashed lines show the 95% confidence (back) and 95% prediction (grey) limits, P > 0.001.

mean leaf area for the paleoflora (Fig. 4). Thirty species were moved from smaller leaf size classes to the larger notophyll and mesophyll groups. The grand mean leaf area of the paleoflora was found to be 7.47 ln mm2 (1755 mm2), or large microphyll, approximately 360 mm2 higher than when no fragmented leaves were analyzed (1394 mm2, 7.24 ln mm2; Fig. 4). Including fragmented leaves added 94 additional specimens representing 76 species, 30 of which previously had no area measurements. Many of these previously unmeasured species had large notophyll or mesophyll leaf areas (Appendix A). An average of 2.8 of the four sampling regions for vein density could be measured in fragmented specimens.

27

The Laguna del Hunco paleoflora was dominated by large microphyll-sized species with much smaller numbers of notophylls and mesophylls (Fig. 4). Laguna del Hunco’s leaf size index (LSI) was 27, squarely within the 13 to 38 range defined for Simple Notophyll Vine

Forests from the analysis of Australian rainforest floor leaf litter (Greenwood, 1994). Slightly fewer than half of the Laguna del Hunco species were represented by a single-leaf measurement.

Species with more than one measurement had an average 14-fold variation in unlogged intraspecific leaf areas.

Including vein scaling results

Excluding vein

100 Micro scaling results 80 60 Noto 40 Meso Species counts 20 Lepto Nano Macro 0

2 4 6 8 10 12

Species mean area (ln mm2)

Figure 4. Laguna del Hunco fossil dicot leaf areas. Species mean leaf areas are shown for 154 dicot species and binned by leaf size class (abbreviated). The grand mean leaf area for Laguna del Hunco is 7.47 ln mm2 (large microphyll) shown as a bold line. Species mean areas that were modified by including fragmented specimens with intact areas estimated by vein scaling are striped. The number of species in each leaf size class is as follows: 1 leptophyll (0.6%), 0 nanophyll (0%), 92 microphyll (59.3%), 40 notophyll (25.8%), 20 mesophyll (13.5%), and 1 macrophyll (0.6%). See Table 2 for the methods used for leaf area measurements and additional data sources.

28

Leaf size comparisons

When Laguna del Hunco leaf areas were compared to those of Australian species of the same or most closely related genus, mean fossil leaf areas were consistently smaller (Fig. 5;

Table 5). Of the eight fossil species studied, five were, at most, one leaf size class smaller than at least one of their NLRs. The fossil species with the greatest differences in leaf area from their nearest living relatives were Podocarpus andiniformis and Monimiophyllum callidentatum. M. callidentatum is represented only by one fossil specimen. The mean areas of the fossil species

Agathis zamunerae, cf. Caldcluvia, Atherospermophyllum guinazui, and Eucalyptus frenguelliana overlapped or nearly overlapped with their living equivalents. There was no correlation between the sample size of a fossil species and its similarity to its Australian nearest living relatives (Fig. 5, Table 5).

Laguna del Hunco’s grand mean leaf area of 1755 mm2 (7.47 ln mm2) was also smaller than those of all living Australian rainforest plots, which had grand mean leaf areas varying from

2038 – 15063 mm2 (7.62 to 9.62 ln mm2; Fig. 6). The rainforest plots with the most similar grand mean leaf areas to LH were from the subtropics: the Nightcap-Border Ranges, Dorrigo, and

Washpool. The ten Australian rainforest plots with the most similar grand mean leaf areas to LH were on average cool (16.7 ºC MAT), moderately wet (1655.5 mm MAP), and at foothill elevations (311.2 m; Table B5), and most commonly had metamorphic bedrock. The forest types of the ten plots were either SNVF or mixed SNVF-CNVF, and on average had lower diversity

(24.1 species/plot) than the typical subtropical Australian rainforest (Table B5; Table 3).

29

M

*

* e

i * s a s s i i * s

- a u

m a u

n s s

a r u

a i i o z

m c i u p a t

m

a i l n

i

e i r r r

l h v l a l p

u

t n m

e

n n o a y a t u f n r f l y e

i

a o i l

c

a u c h a u e c u t

u

g g n

a k u o p i e p d

g m n l

g a l

c A a i d

o A d

t s i

n a r e a u h o

n a

o z e d m A m c i r E C r a i

P p i l

l f

u

e l . n p l a f h o y t c c h A M p

Figure 5. Leaf areas of Laguna del Hunco fossils and Australian nearest living relatives The leaf areas of eight Laguna del Hunco species (grey boxplots) are compared to the average leaf areas of their Australian nearest living relatives (diamonds). Some Australian species areas are not shown if they overlapped with those of other species of the same genus (see Table 5 for all Australian species). Box plot tails go to data extremes; boxes show 1st and 3rd quartiles, bold lines show medians, and points show means. Leaf size classes are labeled (abbreviated); *Most similar living species; **areas are of leaflets

30

Table 5. Taxon-specific leaf area for Laguna del Hunco and Australian nearest living relatives

Fossil species Mean LA Australian NLRs Mean LA n mm2 ln mm2 mm2 ln mm2 Agathis zamunerae 44 323.7 5.78 A. atropurpurea 1224.1 7.11 A. microstachya 1571.8 7.36 A. robusta 8.20 Akania patagonica 6 1702.7 7.44 A. bidwillii 3640.9 9.08

Araucaria pichileufensis 8 13.8 2.63 A. cunninghamii 42.1 3.74 Caldcluvia sp. 55 943.8 6.85 Caldcluvia 2951.3 7.99 australiensis 6310.7 8.75 C. paniculata Atherospermophyllum 8 2018.2 7.61 Daphnandra apatela* 4188.1 8.34 guinazui D. repandula 12708.1 9.45 D. tenuipes 2724.4 7.91 Eucalyptus frenguelliana 102 336.9 5.82 E. acmenoides 2100.6 7.65 E. campanulata 3165.3 8.06 E. grandis 3361.0 8.12 E. microcorys 2100.8 7.65 E. pellita 4491.8 8.41 E. pilularis 3361.0 8.12 Podocarpus 9 88.2 4.48 P. dispermus 11968.1 9.39 andiniformis P. elatus 2275.6 7.73 P. grayae 3498.2 8.16 Monimiophyllum 1 757.4 6.63 Wilkiea hugeliana* 5271.1 8.57 callidentatum W. angustifolia 12456.5 9.43 W. austro- queenslandica 9798.6 9.19 W. Barong 19930.4 9.90 W. macrophylla 15677.8 9.66 W. McIlwraith 7331.9 8.90 W. MtHemmant 5653.3 8.64 W. MtMolloy 6063.2 8.71 W. wardellii 5486.2 8.61 Notes: Australian nearest living relative (NLR; Table 1) leaf areas are calculated from Kooyman et al. (2012); LA is leaf area; n is number of fossil leaves measured; *identified as the most similar living species (see Table 1 for sources).

31

) 2 10 m m 9 Meso 8 Noto LH 7 Micro Grand mean leaf area (ln Grand CY WT NB W D

Figure 6. Grand mean leaf areas of Laguna del Hunco and living Australian rainforests The distribution of grand mean leaf areas, the mean of all dicot species mean areas, of living rainforest plots within each region of Australia are shown as boxplots. The Australian regions are Cape York (CY), Wet Tropics (WT), Nightcap-Border Ranges (BD), Washpool (W), and Dorrigo (D). Boxplots for tropical regions are light grey and subtropical regions are dark grey. The grand mean leaf area for LH is shown as a bold line. Box plot tails go to data extremes, boxes show 1st and 3rd quartiles, bold lines show medians, and points show means. Leaf size classes are labeled (abbreviated).

32

Floristic comparisons

In the ordination based on the presences and absences of 95 woody plant families, Laguna del Hunco’s family composition was found to be within the range of variation of those of living

Australian rainforests (Fig. 7). Of the ten rainforest plots closest to LH in ordination, eight were from the subtropics (Table B4). These plots shared 73-80% family-level compositional similarity with LH (Table B2); however, none had the families Cupressaceae, Bixaceae, Escalloniaceae,

Euphorbiaceae, Salicaceae, Urticaceae, or Akaniaceae, which are all known from LH and extant in Australia (Table 3B). The mean climate of the ten most similar Australian rainforest plots was cool (16.6 ºC MAT), moderately wet (1547.3 mm MAP), and at upland elevations (659.7 m;

Table B4). Of the ten closest plots, nine were CMVFs and had moderate species diversity (mean

36.8 species/plot; Table B5; Table 3).

Genera of Laguna del Hunco’s NLRs were found to occur widely throughout Australia; all rainforest plots had at least one of the eleven genera studied (Fig. 8). These genera had varying abundances; in terms of the percentage of plots where each occurred, they were Cochlospermum

(2%), Agathis (3%), Akania (7%), Podocarpus (11%), Araucaria (13%), Ceratopetalum (15%),

Orites (16%), Daphnandra (22%), Caldcluvia (23%), and Eucalyptus (26%).

Overall, subtropical plots were found to have more NLR genera than tropical plots (Table 6,

Fig. 8), but lacked the genera Agathis, Podocarpus, Cochlospermum, and Gymnostoma that were only found in tropical Australia in this dataset. Up to seven NLR genera were found to co-occur in two individual plots, one in the Moonpar State Forest in Dorrigo and the other in the Whian

33

Figure 7. Family compositions of Laguna del Hunco and living Australian rainforests A. The 3-dimensional nonmetric multidimensional scaling (NMS) ordination of LH and living Australian rainforest plots based on the presences and absences of 95 woody plant families. Ordinated points are samples: LH (yellow) and living Australian rainforest plots from tropical (red) or subtropical (blue) latitudes. See Appendix B for additional NMS methods and support. B. Similarity between LH and living Australian rainforest plots. Similarity was calculated as the three dimensional relative Euclidean distance between the ordination score of LH and each Australian rainforest plot. Smaller distances reflect higher similarity. Australian regions are Cape York (CY) and the Wet Tropics (WT) from tropical latitudes (light grey) and the Nightcap- Border Ranges (BD), Washpool (W), and Dorrigo (D) from subtropical latitudes (dark grey). Boxplot tails go to data extremes; boxes show 1st and 3rd quartiles, and bold lines show medians.

34

10

Wilkiea 10 8 Podocarpus Orites 8

6 Eucalyptus Daphnadra 6

4 Cochlospermum 4 Ceratopetalum 2 2 Caldcluvia Araucaria

Akania 0 0 Agathis Number of occurrences within plots Number of occurrences with regions D W CY WT NB W D CY NB WT

Figure 8. Occurrences of Laguna del Hunco’s Australian nearest living relative genera The distributions of eleven genera of Laguna del Hunco’s Australian nearest living relatives are shown within rainforest plots from each Australian region; plots from tropical regions are light grey, plots from subtropical regions are dark grey. Australian regions are Cape York (CY), Wet Tropics (WT), Nightcap-Border Ranges (BD), Washpool (W), and Dorrigo (D). Results include 3 genera known from LH as fossil fruits and seeds, Orites, Cochlospermum, and Ceratopetalum. Boxplot tails go to data extremes; boxes show 1st and 3rd quartiles, bold lines show medians, and points show means.

35

Whain State Forest in the Nightcap-Border Ranges (Table 6). Both of these plots have the rare genus Akania found only in 7% of the plots studied.

Of the ten rainforests plots with the highest number of NLR genera, five were from the

Nightcap-Border Ranges. The mean climate of the ten most similar plots was cool (16.9 ºC

MAT), moderately wet (1596.5 mm MAP), and at low elevations (548.5 m). All of the closest ten plots had metamorphic or igneous bedrock. Of these ten plots, four were CMVFs, four were

CNVFs, and two were SNVFs. On average, the ten plots had higher species diversity compared to other rainforests in Australia (71.9 species/plot; Table B5; Table 3).

Climate space comparisons

The studied nearest living relatives of the Laguna del Hunco paleoflora formed three distinct regions when their climate spaces were defined by the MAT, MAP, and elevation of their occurrences (Table 5; ). These regions corresponded to whether NLRs were subtropical

Australian species, tropical Australian species, or ex-Australian species. There was very little overlap between regions; only two species, Orites excelsus and Podocarpus elatus, occurred in more than one region. Because the majority of the occurrences of these two species were in subtropical Australia, they were grouped with subtropical Australian species (Fig. 9).

All tropical Australian species, even those with the highest elevation occurrences, had warmer climate spaces than subtropical Australian species. All NLR genera extant in Australia had species in both tropical Australia and subtropical Australia, except for Agathis,

Cochlospermum, and Akania that were exclusive to one part of Australian in this dataset. Ex-

Australian species nearly always had higher elevation and wetter climate spaces than Australian species.

36 Table 6. Living Australian rainforests with the highest number of Laguna del Hunco’s nearest living relative genera

n Forest Compositional Species Leaf Area Elevation MAT MAP Plot ID Location Bedrock Region NLRs Type similarity diversity (ln mm2) (m) (ºC) (mm) D144 7 Moonpar S.F. CNVF* m D 0.62 97 8.21 646 16 1622 N61 7 Whian Whian S.F. SNVF r NB 0.75 92 8.28 189 19 1846 N10 6 Whian Whian S.F. SNVF b NB 0.61 103 8.49 160 18 1808 D141 6 Dorrigo N.P. CNVF* m D 0.69 75 8.24 611 14 1876 N32 5 Border Ranges N.P. CMVF* b NB 0.57 49 8.61 744 16 1795 N99 5 Mt Warning N.P. CMVF* g NB 0.66 84 8.48 225 18 1849 N19 5 Beaury S.F. CMVF* b NB 0.35 67 8.47 717 16 1120 D159 5 Bellinger River N.P. CMVF* b D 0.50 54 8.38 799 16 1566 P45 5 Bakers Blue Mt. CNVF g WT 0.52 61 8.76 790 21 1317 W273 5 Washpool N.P. CNVF* m W 0.61 37 8.26 604 15 1166 Mean 71.9 8.41 548.5 16.9 1596 Range 37-103 8.21-8.75 160-799 14-21 1120-1876 Notes: The ten Australian rainforests with the highest number of Laguna del Hunco’s nearest living relatives (NLRs) genera are shown with their associated climatic, geographic, and ecologic data. See Methods for sources for forest types; *Forest type inferred from canopy leaf size index. Rainforest plot bedrock, region, floristic, and climate data are sourced from Kooyman et al. (2012). Compositional similarity at the family level determined from ordination, where smaller values are more similar. Leaf area is grand mean leaf area. Abbreviations: n, number; N.P., national park; S.F., state forest; bedrock values: m, metamorphic; r, rhyolite; b, basalt; g, granite; climate values: MAT; mean annual temperature; MAP, mean annual precipitation; geographic regions: WT, Wet Tropics; NB, Nightcap-Border Ranges; D, Dorrigo; W, Washpool.

37

Figure 9. Climate spaces of Laguna del Hunco’s nearest living relatives Each species climate space is defined by one ellipsoid, centered at the average value of the mean annual temperature (MAT), mean annual precipitation (MAP), and elevation of the localities where the species has been recorded, and with ellipsoid radii equal to one standard deviation away from average values. Ellipsoids for species that are extant in the Australian subtropics (blue) and tropics (red) represent only recorded occurrences within Australia, even if the species is known from other regions. Ellipsoids for species that are extinct in Australia (green) represent all recorded occurrences in Australasia and Southeast Asia. See Table 4 for a list of all species used, climate range values, and the regions from which living occurrence data were analyzed.

38

DISCUSSION

Including fragmented leaves in fossil leaf area measurements

The hypothesis that leaf size was controlled by similar developmental constraints in the

Cenozoic as it is today (Fig. 2, Fig. 3; Sack et al., 2012) is supported by the finding that fossil leaves have nearly the same area and vein scaling relationships as those observed in living dicot leaves. Because vein scaling was found to predict intact leaf area as accurately as leaf size classes, it can be used to reconstruct the areas of fragmented fossil leaves without compromising measurement accuracy.

The accuracy of vein scaling predictions was correlated to the number of regions of fossil leaf lamina that were measured for secondary vein density. For leaves where all four sampling regions could be measured (Fig. 1), the correlation between predicted and measured areas was much higher than for leaves where fewer regions were measured (R2 = 0.75, 0. 68, respectively;

P < 0.001). In this study, an average of 2.8 of the four sampling regions could be measured, either because other regions were poorly preserved or not intact. Because of this, the vein scaling method may be improved in future studies by choosing sampling regions to maximize sample area, rather than to standardize the portions of leaf lamina that are sampled across specimens.

By including the reconstructed areas of fragmented leaves in Laguna del Hunco’s leaf size measurements, I added 94 specimens and 30 species that previously had no area estimates to the dataset. Moreover, applying vein scaling helped recover some large fossil leaf areas that would have otherwise gone unrepresented (Fig. 4). Although fragmented leaves made up only a small percentage of the specimens used in leaf area measurements, including their reconstructed areas had a significant effect on leaf area results, increasing the grand mean leaf area of the paleoflora by ~360 mm2. Vein scaling tended to overestimate the areas of intact leaves, but this tendency

39

cannot fully explain the observed increase in Laguna del Hunco grand mean leaf area. When the unreconstructed areas of fragmented leaves were included in leaf area measurements, Laguna del

Hunco’s grand mean leaf area still increased by a smaller, but substantial, ~101 mm2. Therefore, part of the observed increase in Laguna del Hunco grand mean leaf area is independent of any overestimation of the vein scaling method. This new technique of using vein scaling to estimate intact leaf area can be widely applied and may be especially useful for other paleofloras that have more fragmentary leaf specimens than Laguna del Hunco.

Assessing taphonomic bias in leaf area

While this study does more to mitigate the effects of taphonomic bias in leaf area than any prior analysis, Laguna del Hunco’s species mean leaf areas and grand mean leaf area must still be considered minimum estimates. The unreconstructed areas of fragmented leaves were larger than those of intact leaves by an average of ~708 mm2. This finding supports the hypothesis that larger fossil leaves are more frequently fragmented than smaller fossil leaves (Roth and Dilcher,

1978; Spicer, 1981; Hill and Gibson 1986; Greenwood, 1992). Using vein scaling was found to help recover some large leaf areas (Fig. 4), but vein-scaling measurements were restricted to fragments of known species. Therefore, only fragmented specimens that were large enough for species identification were included in leaf area measurements. If larger leaves are more likely to be torn into small fragments before burial, then omitting the small, unidentified fragments in this analysis may have left a portion of measurable large leaf area unrepresented.

There is some evidence that suggests that the taphonomic bias in Laguna del Hunco’s leaf size data is limited. Laguna del Hunco species with more than one area measurement had an average intraspecific variation of 14-fold, larger than that observed across 157 species of living woody dicots from a variety of environments (6-fold; Milla and Reich, 2007). Because fossil leaf

40

areas have a greater intraspecific variation than living leaf areas, Laguna del Hunco’s area results are likely to have sampled the larger leaves of most fossil species. Therefore, paleoecological interpretations of Laguna del Hunco’s leaf area results are considered appropriate as long as modern taphonomic samples, like leaf litter accumulations, are used for reference (Roth and

Dilcher, 1978; Spicer, 1981; Hill and Gibson, 1986; Burnham, 1989, 1992; Greenwood, 1992).

The mean leaf areas of all eight Laguna del Hunco fossil species were smaller than those of

Australian species of the same or most similar genus (Fig. 5). Because this trend was shared between distantly related taxonomic groups, including dicots and conifers, it most likely reflects a common taphonomic bias against large leaves, rather than a common evolutionary increase in leaf area. This is supported by the finding that the leaf size difference between Laguna del Hunco fossil and living Australian leaves is similar to that seen in Australian rainforests, where species mean leaf areas measured from leaf litter were found to be up to one size class smaller than those measured from the local canopy (Greenwood, 1992). Five fossil species had mean areas that were less than one size class smaller than at least one of their NLRs (Fig. 5). If the maximum leaf areas of fossil species are considered to be less biased estimates of the paleorainforest canopy species mean areas, then these results indicate little difference in leaf area between most fossil species and their NLRs in Australia.

The fossil species Araucaria pichileufensis may be an exception to the taphonomic explanation for the differences between fossil and living leaf areas. This conifer species has scale-like leaves with the smallest mean areas of all of the fossil species studied. As fossils,

Araucaria leaves were always found attached to branches and not fragmented, suggesting that the leaves of this species may be the least subject to taphonomic bias. However, the mean leaf area of Araucaria cunninghamii is more than three times larger than that of fossil A.

41

pichileufensis (42 mm2 and 14 mm2, respectively) indicating that the leaf size difference between fossil and living Australian Araucaria species may be independent of the preservation of large leaves. Further, outside of Australia, species of Araucaria Section Eutacta have leaf sizes that are more similar to fossil Araucaria. For example, the New Caledonian endemic A. nemorosa has a species mean leaf area of 13 mm2 (2.39 ln mm2) as calculated using the Cain and Castro

(1955) formula on leaf lengths and widths (Silba, 1986). This size is much closer to that measured for fossil Araucaria. Therefore, fossil Araucaria leaf sizes are most likely characteristically smaller than Australian species of the same genus without the influence of taphonomic bias.

Laguna del Hunco as a Simple Notophyll Vine Forest

The Laguna del Hunco paleoflora had a majority of large-microphyll leaves with fewer notophylls and mesophylls, resulting in a leaf size index of 27. This leaf size index was within the range of values characteristic of leaf litter from Australian Simple Notophyll Vine Forests

(SNVFs), but no other forest type (Fig. 4; Greenwood, 1991, 1994). Laguna del Hunco’s floristics also support its affinity with SNVFs. In Australian rainforests, SNVFs are defined by the presence of vines (Webb, 1959) and Laguna del Hunco includes several plant families, like

Menispermaceae, that commonly have vine life forms. Additionally, Australian SNVFs characteristically have sclerophyllous emergents like Araucaria, Agathis, and Eucalyptus, all genera known from Laguna del Hunco. The Laguna del Hunco paleorainforest is also likely to have included emergents because of the presence of large silicified tree stumps (Petersen, 1946) and other genera like Agathis, Gymnostoma, and Dacrycarpus that reach heights of 60 m in living rainforests (Paijmans, 1970; Farjon, 2010). Therefore, this study suggests that the Laguna del Hunco paleoflora has the greatest similarity to living Australian SNVFs.

42

Laguna del Hunco also has floristic similarity with the Eocene Anglesea and Golden Grove paleofloras, from Victoria, Australia, that have been interpreted as subtropical SNVFs

(Greenwood, 1994). These paleofloras include the families Proteaceae, Myrtaceae, and

Lauraceae (Christophel and Greenwood, 1988) and Podocarpus (Greenwood, 1987) and

Gymnostoma (Christophel, 1980), all taxa known from Laguna del Hunco.

Closest analog rainforests

Given the vast geographic and temporal distances involved, living subtropical Australian rainforests were found to be remarkably similar to Laguna del Hunco. In comparisons to LH, subtropical Australian rainforests made up all ten of the plots with the most similar leaf sizes, the majority of the ten plots with the closest family compositions, and the majority of the ten plots with the highest numbers of NLR genera (Table B5, B4, 6). Further, the closest subtropical rainforests in terms of each of these three criteria were strikingly analogous to LH. One subtropical rainforest had a grand mean leaf area of 7.62 ln mm2, just above the 7.47 ln mm2 value for LH (Table B5). Subtropical Australian rainforests were likewise found to be close floristic analogs; one Australian plot shared 80% of its family composition with LH, and another two had seven of the eleven genera of Laguna del Hunco’s NLRs, including the rare genus

Akania (Table B5; Table 6).

The rainforest plots found to be the most similar in each of these criteria did not overlap.

Australian rainforest plots with the highest number of NLR genera did not also have the closest family composition or the closest leaf areas to LH. However, the closest plots did share a similar climate when climate values were averaged across the ten analogs found in each of these categories. This climate was cool (~16.7 - 16.9 ºC MAT), moderately wet (1547 - 1655 mm

MAP; Table B5; Table B4; Table 6), and at moderate elevation (311-658 m), corresponding to

43

mostly foothill and upland environments. These results support the hypothesis that LH has the most similar leaf size and floristics to subtropical rainforests within Australia, but not specifically to montane subtropical rainforests as expected.

Tropical Australian rainforests were also found to share similarities with Laguna del Hunco, although close analogs from this region were less common. All tropical rainforests had much larger leaf areas than Laguna del Hunco, but some had similar floristics. One tropical rainforest shared 78.9% of its family composition with the paleoflora (Table B4). Tropical rainforests contained some Laguna del Hunco taxa that were not found in any other parts of Australia in this dataset, including Podocarpus, Gymnostoma, Agathis, and Cochlospermum. Extant Agathis and

Gymnostoma are known to be highly susceptible to drought (Brodribb and Holbrook, 2005;

Zamaloa et al., 2006) and only tropical rainforests in Australia have mean annual temperatures over 2200 mm (Table 3). Therefore, tropical Australian rainforests can contain Laguna del

Hunco taxa and a high-rainfall climate that are found nowhere else in Australia.

The rainforest types of the closest Australian analogs varied. Laguna del Hunco was interpreted to be most similar to living Australian SNVFs on the similarity of its floristics and leaf size distribution to those Australian rainforest leaf litter (Greenwood, 1994). In comparisons to living rainforests, the ten plots with the most similar grand mean leaf areas to LH were all either SNVF, or mixed SNVF and Complex Notophyll Vine Forest (CNVF; Table B5), supporting the association between leaf size and rainforest type. In floristic comparisons, the

Australian rainforest plots with the most similar family composition to LH were nearly all

Complex Mesophyll Vine Forests (Table B4). However, the rainforest plots with six or seven of

Laguna del Hunco’s NLR genera were either SNVF or CNVF (Table 6). These results suggest

44

that subtropical Australia includes some rainforests with similar leaf size, genera, and rainforest type as Laguna del Hunco.

Australian climate spaces

Although many lineages from the paleoflora survive in Australia, the three non- overlapping climate regions of Laguna del Hunco’s NLRs indicate that they do not co-occur in the same climates in Australia. In tropical Australia, Laguna del Hunco’s NLRs have climate spaces that are mostly wetter and always warmer (by at least two degrees Celsius in MAT) than those of the subtropical Australian NLRs (Fig. 9). Further, species of the ex-Australian genera

Dacrycarpus, Acmopyle, Papuacedrus, and Retrophyllum were found to have wetter and higher- elevation climate spaces than any of Laguna del Hunco’s Australian NLRs. Therefore, Australia lacks an important cool, wet montane climate needed to support some of the lineages from

Laguna del Hunco.

Outside of Australia, Australasian and Southeast Asian rainforests with a wet montane climate may have both small leaf areas, like some subtropical Australian rainforests, and high numbers of Laguna del Hunco taxa that are extinct in Australia. For example, the lower and upper-montane rainforests of Mt. Kerigomna and Mt. Wilhelm, in Papua New Guinea, have a

MAT of 7.8 to 14.3 ºC, a MAP of at least 3985 mm, and elevations between 2500-3550 m

(Grubb and Stevens, 1985). Both rainforest types have a majority of microphyll or notophyll leaves, similar to the leaf areas found for LH. Common genera in these rainforests include

Dacrycarpus and Papuacedrus, which are extinct in Australia, and Podocarpus and Caldcluvia.

These results suggest that Gondwanan lineages from Laguna del Hunco survived to the present through different actions. In Australia, some lineages lived in drier subtropical climates and others in hotter tropical climates. Outside of Australia, lineages tracked the cool, wet climate

45

to higher-elevation equatorial areas like Borneo and New Guinea. Although Australian rainforests are very close analogs for the paleoflora, surviving Laguna del Hunco lineages did not have a coherent floristic response over the past tens of millions of years. This study supports the idea that Australia provided refugia for some Gondwanan rainforest taxa to move from southern latitudes into warmer, wetter areas in equatorial latitudes and the northern hemisphere

(Wilf et al., 2009; Hill, 2004; Zamaloa et al., 2006).

46

CONCLUSIONS

Among Australian rainforests, the Laguna del Hunco paleoflora was most similar to Simple

Notophyll Vine Forests, based on floristic and leaf area comparisons.

Leaf area measurements for Laguna del Hunco included fragmented leaves with areas reconstructed using a new vein scaling technique. When tested on fossil leaves with intact leaf areas, vein scaling predicted leaf areas as accurately as using leaf size classes, but was applicable to more fragmentary leaves that otherwise could not be measured. Applying vein scaling to a small number of fragmented leaf specimens from LH increased the paleoflora’s grand mean leaf area, recovering large leaf areas that were previously unmeasured. The mean leaf areas of

Laguna del Hunco species were consistently smaller than those of living Australian species of the same or most morphologically similar genus, suggesting that Laguna del Hunco leaf areas remain taphonomically biased against large leaves. For most fossil species, the difference between fossil and living leaf areas was comparable to that observed in living Australian rainforests between leaves collected from leaf litter and those collected from the forest canopy.

Laguna del Hunco was dominated by large microphyll-sized leaves and had an LSI that was within the range characteristic of leaf litter from living Australian SNVFs.

In quantitative comparisons to Laguna del Hunco, subtropical Australian rainforests had the closest leaf areas, made up the majority of plots with the most similar family compositions, and contained the highest numbers of the genera of Laguna del Hunco’s NLRs. A few tropical

Australian rainforests also had high floristic similarity with LH and contained Laguna del Hunco taxa that were not found in any other part of Australia. Laguna del Hunco’s NLRs were found to form three distinct climate regions. Ex-Australian species had climate spaces with higher elevation and higher rainfall than the climate spaces of any Australian species. These results

47

suggest that surviving Gondwanan lineages known from LH either occupied cool-dry climates in subtropical Australia or hot-wet climates in tropical Australia, or tracked a cool-wet climate to montane environments outside of Australia, in Australasia and Southeast Asia.

48

LITERATURE CITED

ARAGÓN, E., AND E. J. ROMERO. 1984. Geología, paleoambientes y paleobotánica de yacimientos

Terciarios del occidente de Río Negro, Neuquén y Chubut. Actas del IX Congreso

Geológico Argentino, San Carlos de Bariloche 4: 475-507.

ARAGÓN, E., AND M. M. MAZZONI. 1997. Geología y estratigrafía del complejo volcánico

piroclástico del Río Chubut Medio (Eoceno), Chubut, Argentina. Revista de la Asociación

Geológica Argentina 52: 243-256.

AZPELICUETA, M. D. L. M., AND A. L. CIONE. 2011. Redescription of the Eocene catfish

Bachmannia chubutensis (Teleostei: Bachmanniidae) of southern South America. Journal of

Vertebrate Paleontology 31: 258-269.

BÁEZ, A. M., AND L. TRUEB, L. 1997. Redescription of the Paleogene Shelania pascuali from

Patagonia and its bearing on the relationships of fossil and Recent pipoid frogs. Scientific

Papers of the Natural History Museum, University of Kansas 4: 1-41.

BAILEY, I. W., AND E. W. SINNOTT. 1916. The climatic distribution of certain types of

angiosperm leaves. American Journal of Botany 3: 24-39.

BERRY, E. W. 1925. Fossil plants from the Tertiary of Patagonia and their significance.

Proceedings of the National Academy of Sciences 11: 404-405.

BERRY, E. W. 1935. A fossil Cochlospermum from northern Patagonia. Bulletin of the Torrey

Botanical Club 62: 65-67.

BERRY, E. W. 1938. Tertiary flora from the Rio Pichifleufu. Geological Society of America

Special Papers 12: 1-149.

49

BIFFIN, E., T. J. BRODRIBB, R. S. HILL, P. THOMAS, AND A. J. LOWE. 2012. Leaf evolution in

southern hemisphere conifers tracks the angiosperm ecological radiation. Proceedings of the

Royal Society B: Biological Sciences 279: 341-348.

BRODRIBB, T. J., AND N. M. HOLBROOK. 2005. Water stress deforms tracheids peripheral to the

leaf vein of a tropical conifer. Plant Physiology 137: 1139-1146.

BURNHAM, R. J. 1989. Relationships between standing vegetation and leaf litter in a paratropical

forest: Implications for paleobotany. Review of Palaeobotany and Palynology, 58: 5-32.

CAIN, S. A., AND G. M. D. O. CASTRO. 1959. Manual of Vegetation Analysis. Harper and

Brothers, New York, USA.

CARPENTER, R. J. 2012. Proteaceae Leaf Fossils: Phylogeny, Diversity, Ecology and Austral

Distributions. The Botanical Review, 78:261-287.

CARVALHO M. R., P. WILF, M. A. GANDOLFO, N. R. CÚNEO, AND K. R. JOHNSON. 2011. Fossil

ferns from the Eocene of Argentina and the deep-time links between southern hemisphere

rainforests. Proceedings of the XVIII International Botanical Congress 250-51. International

Botanical Congress, Melbourne, Australia.

CARVALHO, M. R., P. WILF, E. J. HERMSEN, M. A. GANDOLFO, N. R. CÚNEO, AND K. R. JOHNSON.

2013. First record of Todea (Osmundaceae) in South America, from the early Eocene

paleorainforests of Laguna del Hunco (Patagonia, Argentina). American Journal of Botany

100: 1831-1848.

CASAMIQUELA, R. M. 1961. Un pipoideo fósil de Patagonia. Revista del Museo de La Plata,

Sección Paleontología 4: 71-123.

50

CHRISTOPHEL, D. C., AND D.R. GREENWOOD. 1988. A comparison of Australian tropical

rainforest and Tertiary fossil leaf-beds. Proceedings of the Ecological Society of Australia

15: 139-148.

CHRISTOPHEL, D., AND P. GORDON. 2004. Genotypic control and environmental plasticity: Foliar

physiognomy and paleoecology. New Phytologist 161: 327-329.

CHRISTOPHEL, D.C. 1980. Occurrence of Casuarina megafossils in the Tertiary of South-Eastern

Australia. Australian Journal of Botany 28: 249–259.

CRISP, M. D., M. T. ARROYO, L. G. COOK, M. A. GANDOLFO, G. L. JORDAN, M. S. MCGLONE,

AND H. P. LINDER. 2009. Phylogenetic biome conservatism on a global scale. Nature, 458:

754-756.

ELLIS, B., D. C. DALY, L. J. HICKEY, K. R. JOHNSON, J. D. MITCHELL, P. WILF, AND S. L. WING.

2009. Manual of Leaf Architecture. Cornell University Press, New York, USA.

FARJON, A. 2010. A Handbook of the World's Conifers. Brill, Leiden, Netherlands.

FIDALGO, P., AND D. SMITH. 1987. A fossil Siricidae (Hymenoptera) from Argentina.

Entomological News 89: 63-66.

FLORIN, R. 1940. Die heutige und frühere Verbreitung der Koniferengattung Acmopyle Pilger.

Svensk Botanisk Tidskrift 34: 117-140.

GANDOLFO, M. A., AND E. J. HERMSEN. 2012. The emerging Patagonian fossil record of

Cunoniaceae and its biogeographical significance. Japanese Journal of Palynology 58: 66-

67.

GANDOLFO, M. A., E. J. HERMSEN, M. C. ZAMALOA, K. C. NIXON, C. C. GONZÁLEZ, P. WILF, N.

R. CÚNEO, AND K. R. JOHNSON. 2011. Oldest known Eucalyptus macrofossils are from South

America. PLoS One 6: e21084.

51

GANDOLFO, M. A., M. C. DIBBERN, AND E. J. ROMERO. 1988. Akania patagonica n. sp. and

additional material on Akania americana Romero & Hickey (Akaniaceae), from Paleocene

sediments of Patagonia. Bulletin of the Torrey Botanical Club 115: 83-88.

GANDOLFO, M. A., M. C. ZAMALOA, C. C. GONZÁLEZ, N. R. CÚNEO, P. WILF, AND K. R.

JOHNSON. 2007. Bixaceae: A tropical component of the early Eocene Laguna del Hunco

paleoflora, Chubut, Patagonia, Argentina. Geological Society of America Annual Meeting,

Denver, Abstracts with Programs 39: 589.

GENISE, J., AND J. PETRULEVICIUS. 2001. Caddisfly cases from the early Eocene of Chubut,

Patagonia, Argentina. Second International Congress on Paleoentomology, Kraków, Poland,

Abstracts 12-13.

GIVNISH, T. 1984. Leaf and canopy adaptations in tropical forests. In E. Medina, H. A. Mooney,

C. Vázquez-Yanes [eds.], Physiological Ecology of Plants of the Wet Tropics. 51-84.

Springer, The Hague, Netherlands.

GONZÁLEZ, C. C., M. A. GANDOLFO, M. C. ZAMALOA, N.R. CÚNEO, P. WILF, AND K. R. JOHNSON.

2007. Revision of the Proteaceae macrofossil record from Patagonia, Argentina. The

Botanical Review 72: 235-266.

GONZÁLEZ, C.C. 2009. Revisión taxonómica y biogeográfica de las familias de angiospermas

dominantes de la “Flora del Hunco” (Eoceno Temprano), Chubut, Argentina. PhD

dissertation, Universidad de Buenos Aires, Buenos Aires, Argentina.

GREENWOOD, D. R. 1987. Early Tertiary Podocarpaceae-megafossils From the Eocene Anglesea

locality, Victoria, Australia. Australian Journal of Botany 35: 111-133.

GREENWOOD, D. R. 1991. The taphonomy of plant macrofossils. In S.K. Donovan [ed.],

Fossilization: The Processes of Taphonomy. 145-169. Belhaven Press, London, England.

52

GREENWOOD, D. R. 1992. Taphonomic constraints on foliar physiognomic interpretations of Late

Cretaceous and Tertiary paleoclimates. Review of Palaeobotany and Palynology 71: 149-

190.

GREENWOOD, D. R., D. CHRISTOPHEL, E. BERMINGHAM, C. DICK, AND C. MORITZ. 2005. The

origins and Tertiary history of Australian "tropical" rainforests. In E. Bermingham and C.W.

Dick and C. Moritz [eds.], Tropical Rainforests: Past, Present and Future. 336–373.

University of Chicago Press, Chicago, USA.

GREENWOOD, D.R. 1994. Palaeobotanical evidence for Tertiary climates. In D. Greenwood [ed.],

History of the Australian Vegetation: Cretaceous to Recent, 44-59. Cambridge University

Press, Cambridge, England.

GRUBB, P. J., AND P. F. STEVENS. 1985. The Forests of the Fatima Basin and Mt. Kerigomna,

Papua New Guinea, with a Review of Montane and Subalpine Rainforests in Papuasia.

Australian National University, Canberra, Australia.

HERMSEN, E. J., M. A. GANDOLFO, AND M. C. ZAMALOA. 2012. The fossil record of Eucalyptus in

Patagonia. American Journal of Botany 99: 1356-1374.

HIJMANS, R. J., S. E. CAMERON, J. L. PARRA, P. G. JONES, AND A. JARVIS. 2005. Very high

resolution interpolated climate surfaces for global land areas. International Journal of

Climatology 25: 1965-1978.

HILL, R. S. 2004. Origins of the southeastern Australian vegetation. Philosophical Transactions

of the Royal Society of London B: Biological Sciences 359: 1537-1549.

HILL, R. S., AND N. GIBSON. 1986. Distribution of potential macrofossils in Lake Dobson,

Tasmania. The Journal of Ecology 47: 373-384.

53

HILL, R. S., AND T. J. BRODRIBB. 2003. Evolution of conifer foliage in the southern hemisphere.

Proceedings of the 4th International Conifer Conference 615: 53-58.

HOVENDEN, M. J., AND J. K. V. SCHOOR. 2004. Nature vs. nurture in the leaf morphology of

Southern Beech, Nothofagus cunninghamii (Nothofagaceae). New Phytologist 161: 585-594.

IUCN. 2013. IUCN Red List of Threatened Species. Version 2013.2. .

Downloaded on 23 May 2014.

JACOBS, B.F. 2002. Estimation of low-latitude paleoclimates using fossil angiosperm leaves:

examples from the Miocene Tugen Hills, Kenya. Paleobiology, 28: 399-421.

KNIGHT, C. L., AND P. WILF. 2013. Rare leaf fossils of Monimiaceae and Atherospermataceae

(Laurales) from Eocene Patagonian rainforests and their biogeographic significance.

Palaeontologia Electronica 16: 26A; 1-39.

KOOYMAN, R., M. ROSSETTO, W. CORNWELL, AND M. WESTOBY. 2011. Phylogenetic tests of

community assembly across regional to continental scales in tropical and subtropical rain

forests. Global Ecology and Biogeography 20:707-716.

KOOYMAN, R., ROSSETTO, M., ALLEN, C., & CORNWELL, W. (2012). Australian tropical and

subtropical rain forest community assembly: Phylogeny, functional biogeography, and

environmental gradients. Biotropica 44: 668-679.

LAWVER, L. A., L. M. GAHAGAN, AND I. W. D. DALZIEL. 2011. A different look at gateways:

Drake Passage and Australia/Antarctica. In J. Anderson & J. Wellner [Eds.], American

Geophysical Union Special Publication 063: Tectonic, Climatic, and Cryospheric Evolution

of the Antarctic Peninsula, 5-33. AGU, Washington, DC, USA.

MACPHAIL, M., AND E. TRUSWELL. 1989. Palynostratigraphy of the central west Murray Basin.

BMR Journal of Australian Geology and Geophysics 11: 301-331.

54

MCCUNE, B., J. B. GRACE, AND D. L. URBAN. 2002. Analysis of Ecological Communities. MjM

Software Design, Oregon, USA.

MCDONALD, P. G., C. R. FONSECA, J. M. OVERTON, AND M. WESTOBY. 2003. Leaf-size

divergence along rainfall and soil-nutrient gradients: Is the method of size reduction

common among clades? Functional Ecology 17: 50-57.

MILLA, R., AND P. B. REICH. 2007. The scaling of leaf area and mass: The cost of light

interception increases with leaf size. Proceedings of the Royal Society B: Biological

Sciences 274: 2109-2115.

PAIJMANS, K. 1970. An analysis of four tropical rain forest sites in New Guinea. Journal of

Ecology 58: 77-101.

PEPPE, D. J., ROYER, D. L., CARIGLINO, B., OLIVER, S. Y., NEWMAN, S., LEIGHT, E., ENIKOLOPOV,

et al. 2011. Sensitivity of leaf size and shape to climate: global patterns and paleoclimatic

applications. New Phytologist 190: 724-739.

PETERSEN, C. S. 1946. Estudios geológicos en la región del Río Chubut medio. Dirección de

Minas y Geología Boletín 59: 1-137.

PETRULEVIČIUS, J. F., AND A. NEL. 2003. Frenguelliidae, a new family of dragonflies from the

earliest Eocene of Argentina (Insecta: Odonata): Phylogenetic relationships within Odonata.

Journal of Natural History 37: 2909-2917.

PROSS, J., L. CONTRERAS, P. K. BIJL, D. R. GREENWOOD, S. M. BOHATY, S. SCHOUTEN, J. A.

BENDLE, et al. 2012. Persistent near-tropical warmth on the Antarctic continent during the

early Eocene epoch. Nature 488: 73-77.

RAUNKIAER, C. 1934. The Life Forms of Plants and Statistical Plant Geography; Being the

Collected Papers of C. Raunkiaer. Clarendon Press, Oxford, England.

55

RICHARDS, P. W. 1996. The Tropical Rain Forest: An Ecological Study, 2d ed. Cambridge

University Press, Cambridge, England.

ROMERO, E. J. AND L. J. HICKEY. 1976. A fossil leaf of Akaniaceae from Paleocene beds in

Argentina. Bulletin Torrey Botanical Club 103:126–131.

ROMERO, E. J., M. DIBBERN, AND M. GANDOLFO. 1988. Revisión de Lomatia bivascularis (Berry)

Frenguelli (Proteaceae) del yacimiento de la Laguna del Hunco (Paleoceno), Pcia. del

Chubut. Actas del IV Congreso Argentino de Paleontología y Bioestratigrafía 3: 125-130.

ROTH, J. AND D. DILCHER. 1978. Some considerations in leaf size and leaf margin analysis of

fossil leaves. Courier Forschungsinst Senckenberg 30: 165-171.

ROYER, D. L., R. M. KOOYMAN, S. A. LITTLE, AND P. WILF. 2009. Ecology of leaf teeth: A multi-

site analysis from an Australian subtropical rainforest. American Journal of Botany 96: 738-

750.

SACK, L., C. SCOFFONI, A. D. MCKOWN, K. FROLE, M. RAWLS, J. C. HAVRAN, H. TRAN, AND T.

TRAN. 2012. Developmentally based scaling of leaf venation architecture explains global

ecological patterns. Nature Communications 3: 837-847.

SCHER, H. D. AND MARTIN, E. E. (2006). Timing and climatic consequences of the opening of

Drake Passage. Science, 312: 428-430.

SHACKLETON, N. AND J. KENNETT. 1975. Paleotemperature history of the Cenozoic and the

initiation of Antarctic glaciation: Oxygen and carbon isotope analyses in DSDP Sites 277,

279, and 281. Initial Reports of the Deep Sea Drilling Project 29: 743-755.

SILBA, J. (1986). Phytologia memoirs VIII. Encyclopedia coniferae. Corvallis, Oregon, 205-207.

56

SPICER, R. A. 1981. Sorting and deposition of allochthonous plant material in a modern

environment at Silwood Lake, Silwood Park, Berkshire, England. U.S. Geological Survey

Professional Paper 1143: 1-77.

STEART, D. C., P. I. BOON, D. R. GREENWOOD, AND N. T. DIAMOND. 2002. Transport of leaf litter

in upland streams of Eucalyptus and Nothofagus forests in south-eastern Australia. Archiv

für Hydrobiologie 156: 43-61.

WEBB, L. J. 1959. A physiognomic classification of Australian rain forests. Journal of Ecology

47: 551-570.

WEBB, L. J. 1968. Environmental relationships of the structural types of Australian rain forest

vegetation. Ecology 49: 296-311.

WEBB, L. J., J. G. TRACEY, AND W. T. WILLIAMS. 1984. A floristic framework of Australasian

rainforests. Australian Journal of Ecology 9: 169-198.

WILF, P. 2012. Rainforest conifers of Eocene Patagonia: Attached cones and foliage of the extant

southeast-Asian and Australasian genus Dacrycarpus (Podocarpaceae). American Journal of

Botany 99: 562-584.

WILF, P., I. H. ESCAPA, N. R. CÚNEO, R. M. KOOYMAN, K. R. JOHNSON, AND A. IGLESIAS. 2014.

First South American Agathis (Araucariaceae), Eocene of Patagonia. American Journal of

Botany 101: 156-179.

WILF, P., K. R. JOHNSON, N. R. CÚNEO, M. E. SMITH, B. S. SINGER, AND M. A. GANDOLFO. 2005.

Eocene plant diversity at Laguna del Hunco and Río Pichileufú, Patagonia, Argentina.

American Naturalist 165: 634-650.

57

WILF, P., N. R. CÚNEO, I. H. ESCAPA, D. POL, AND M. O. WOODBURNE. 2013. Splendid and

seldom isolated: The paleobiogeography of Patagonia. The Annual Review of Earth and

Planetary Sciences 41: 561-603.

WILF, P., N. R. CÚNEO, K. R. JOHNSON, J. F. HICKS, S. L. WING, AND J. D. OBRADOVICH. 2003.

High plant diversity in Eocene South America: Evidence from Patagonia. Science 300: 122-

125.

WILF, P., S. A. LITTLE, A. IGLESIAS, M. C. ZAMALOA, M. A. GANDOLFO, N. R. CÚNEO, AND K. R.

JOHNSON. 2009. Papuacedrus (Cupressaceae) in Eocene Patagonia, a new fossil link to

Australasian rainforests. American Journal of Botany 96: 2031-2047.

WILF, P., S. L. WING, D. R. GREENWOOD, AND C. L. GREENWOOD. 1998. Using fossil leaves as

paleoprecipitation indicators: An Eocene example. Geology 26: 203-206.

WOLFE, J. A. 1971. Tertiary climatic fluctuations and methods of analysis of Tertiary floras.

Palaeogeography Palaeoclimatology Palaeoecology 9: 27-57.

WOLFE, J. A. 1993. A method of obtaining climatic parameters from leaf assemblages. US

Geological Survey Bulletin 2040: 1-71.

WOLFE, J. A., AND G. R. UPCHURCH JR. 1987. North American nonmarine climates and

vegetation during the Late Cretaceous. Palaeogeography, Palaeoclimatology,

Palaeoecology 61: 33-77.

ZACHOS, J. C., L. D. STOTT, AND K. C. LOHMANN. 1994. Evolution of early Cenozoic marine

temperatures. Paleoceanography 9:353–387.

ZACHOS, J., M. PAGANI, L. SLOAN, E. THOMAS, AND K. BILLUPS, K. 2001. Trends, rhythms, and

aberrations in global climate 65 Ma to Present. Science 292: 686-693.

58

ZAMALOA, M. C., M. A. GANDOLFO, C. C. GONZÁLEZ, E. J. ROMERO, N. R. CÚNEO, AND P. WILF.

2006. Casuarinaceae from the Eocene of Patagonia, Argentina. International Journal of

Plant Sciences 167: 1279-1289.

59

APPENDIX A. SUPPLEMENTAL LEAF AREA METHODS AND RESULTS

CONTENTS Additional fossil leaf vein measurement methods ...... 61 Table A1. Scaling of fossil leaf vein traits with leaf area ...... 61 Figure A1. Relationship between species sample size and mean area...... 62 Table A2. All Laguna del Hunco dicot fossil specimen leaf areas...... 63 Table A3. All Laguna del Hunco dicot fossil species mean leaf areas ...... 91

60

Additional fossil leaf vein measurement methods Leaf venation traits were measured in 159 fossil leaves with intact areas from Laguna del Hunco to test if fossil vein traits correlate with leaf area in a similar way as in living leaves. Fossil measurement follows the same protocol used in living leaves (Sack et al., 2012). Primary vein width was measured perpendicularly to 1º vein length at 4 regions along the midvein: the leaf base, and the centers of the basal, middle and apical thirds of the leaf. The basal width measurement was made just above the petiole insertion point, or at the basal-most portion of the midvein preserved if the petiole was absent. Primary vein width measurements were averaged across each leaf specimen. Primary vein density was calculated as the sum of the lengths of primary veins divided by total leaf area. Secondary vein density was calculated as the sum of lengths of secondary veins divided by total leaf area. If the leaf was symmetrical, 2º vein density measurements were conducted on one medial side of the midvein and doubled. Subsampled 2º vein density was measured as described in Materials and Methods.

Table A1. Scaling of fossil leaf vein traits with leaf area. The scaling relationships between four fossil vein traits and intact leaf area are shown for natural log transformed data. R2 = Pearson’s coefficient of determination; n = number of fossil specimens analyzed; modern scaling relationships observed across 485 living species are shown for comparison (Sack et al., 2012). All modern and fossil values are significant at P < 0.001.

Fossil leaf vein trait n R2 Living leaf R2 Mean 1º vein width 78 0.39 0.58-0.70 1º vein density 145 0.74 0.90 2º vein density 101 0.75 0.66-0.78 Subsampled 2º vein density 159 0.68 0.74

61

10 9 ) 2 m 8 m 7 6 5 Leaf area (ln 4 3

0 20 40 60 80 100

Total specimens measured

Figure A1. Relationship between species sample size and mean area. For 154 dicot fossil species analyzed, the total specimens measured for each species is plotted relative to the species mean leaf area. There is no correlation between species mean leaf area and the number of specimens representing a fossil species, and many large leaves are contributed by species with only one area measurement.

62

Table A2. All Laguna del Hunco dicot fossil specimen leaf areas. Leaf areas for 1152 specimens are given in ln mm2. See Materials and Methods for definitions of leaf completeness and area measurement methods. All leaf areas were either directly measured or estimated using vein scaling (VS), the Cain and Castro formula (CC), or leaf size classes (SC) by taking the mean of the logarithm transformed upper and lower bounds of the size class of each specimen.

Specimen field # Morphotype Leaf completeness ln Area Method Source LH02-0117 TY017 Complete 6.27 Direct this study LH02-0217 TY017 Complete 6.11 Direct this study LH02-1246 TY017 Complete 6.63 Direct this study LH02-0315 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0552 TY018 Complete 6.24 CC Wilf et al. 2003, 2005 LH02-0553 TY018 Complete 6.58 CC Wilf et al. 2003, 2005 LH02-0554 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0557 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0009 TY018 Complete 7.17 CC Wilf et al. 2003, 2005 LH04-0032 TY018 Complete 7.80 CC Wilf et al. 2003, 2005 LH04-0092 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0129 TY018 Complete 7.56 CC Wilf et al. 2003, 2005 LH04-0134 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0135 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0137 TY018 Complete 6.68 CC Wilf et al. 2003, 2005 LH04-0140 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0160 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0163 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0184 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-1056 TY018 Complete 6.33 CC Wilf et al. 2003, 2005 LH04-1187 TY018 Complete 7.20 CC Wilf et al. 2003, 2005 LH06-0046 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0135 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0142 TY018 Complete 7.49 Direct this study LH06-0152 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0212 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0218 TY018 Complete 4.54 CC Wilf et al. 2003, 2005 LH06-0258 TY018 Complete 7.04 CC Wilf et al. 2003, 2005 LH06-0302 TY018 Complete 6.68 CC Wilf et al. 2003, 2005 LH06-1041 TY018 Complete 6.06 CC Wilf et al. 2003, 2005 LH10-0005 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH11-0004 TY018 Complete 7.13 CC Wilf et al. 2003, 2005 LH13-0015 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0162 TY018 Complete 6.62 CC Wilf et al. 2003, 2005 LH13-0289 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0500 TY018 Complete 6.51 SC Wilf et al. 2003, 2005

63

Specimen field # Morphotype Leaf completeness ln Area Method Source LH13-0501 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0502 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0503 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-1026 TY018 Complete 5.59 CC Wilf et al. 2003, 2005 LH13-1029 TY018 Complete 5.65 Direct this study LH13-1094 TY018 Complete 6.18 CC Wilf et al. 2003, 2005 LH13-1512 TY018 Complete 6.48 CC Wilf et al. 2003, 2005 LH16-0008 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH17-0040 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH17-0041 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH17-0042 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH17-0043 TY018 Complete 6.51 SC Wilf et al. 2003, 2005 LH18-0013 TY018 Complete 6.19 CC Wilf et al. 2003, 2005 LH23-0037 TY018 Complete 6.04 CC Wilf et al. 2003, 2005 LH27(2009)-0256 TY018 Complete 6.40 Direct this study LH02-0043 TY019 Complete 8.01 SC Wilf et al. 2003, 2005 LH02-0073 TY019 Complete 8.01 SC Wilf et al. 2003, 2005 LH02-0105 TY019 Complete 7.79 CC Wilf et al. 2003, 2005 LH02-0158 TY019 Complete 8.01 SC Wilf et al. 2003, 2005 LH02-0291 TY019 Complete 7.35 CC Wilf et al. 2003, 2005 LH04-0003 TY019 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-0004 TY019 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0024 TY019 Complete 9.11 SC Wilf et al. 2003, 2005 LH04-0052 TY019 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-0055 TY019 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0059 TY019 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0064 TY019 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-0069 TY019 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0081 TY019 Complete 9.11 SC Wilf et al. 2003, 2005 LH04-0083 TY019 Complete 8.41 CC Wilf et al. 2003, 2005 LH04-0089 TY019 Complete 8.34 CC Wilf et al. 2003, 2005 LH04-0103 TY019 Complete 8.81 CC Wilf et al. 2003, 2005 LH04-0118 TY019 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-0153 TY019 Fragment 7.06 VS this study LH04-0154 TY019 Fragment 6.74 VS this study LH04-0155 TY019 Fragment 6.70 VS this study LH04-0156 TY019 Fragment 6.78 VS this study LH04-0157 TY019 Fragment 6.90 VS this study LH04-0170 TY019 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-0191 TY019 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-1031 TY019 Complete 8.78 CC Wilf et al. 2003, 2005 LH04-1139 TY019 Complete 8.25 CC Wilf et al. 2003, 2005

64

Specimen field # Morphotype Leaf completeness ln Area Method Source LH04-1157 TY019 Complete 7.55 CC Wilf et al. 2003, 2005 LH04(2006)-0032 TY019 Complete 8.97 Direct this study LH06-0114 TY019 Complete 7.24 CC Wilf et al. 2003, 2005 LH06-0300 TY019 Complete 9.11 SC Wilf et al. 2003, 2005 LH13-1109 TY019 Complete 9.09 CC Wilf et al. 2003, 2005 LH13-1140 TY019 Complete 8.12 CC Wilf et al. 2003, 2005 LH13-7117 TY019 Complete 6.50 CC Wilf et al. 2003, 2005 LH15-1000 TY019 Complete 8.89 CC Wilf et al. 2003, 2005 LH16-0001 TY019 Complete 7.69 CC Wilf et al. 2003, 2005 LH16-0032 TY019 Complete 8.01 SC Wilf et al. 2003, 2005 LH20-0006 TY019 Complete 8.01 SC Wilf et al. 2003, 2005 LH23-0038 TY019 Complete 7.71 CC Wilf et al. 2003, 2005 LH01-0013 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0008 TY020 Complete 7.09 CC Wilf et al. 2003, 2005 LH02-0034 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0044 TY020 Complete 9.11 SC Wilf et al. 2003, 2005 LH02-0062 TY020 Complete 7.67 CC Wilf et al. 2003, 2005 LH02-0066 TY020 Complete 6.97 CC Wilf et al. 2003, 2005 LH02-0074 TY020 Complete 4.32 SC Wilf et al. 2003, 2005 LH02-0102 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0122 TY020 Complete 7.59 CC Wilf et al. 2003, 2005 LH02-0197 TY020 Complete 7.12 CC Wilf et al. 2003, 2005 LH02-0200 TY020 Complete 7.76 CC Wilf et al. 2003, 2005 LH02-0225 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0240 TY020 Complete 7.67 CC Wilf et al. 2003, 2005 LH02-0280 TY020 Complete 6.36 CC Wilf et al. 2003, 2005 LH02-0301 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0318 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-1055 TY020 Complete 7.80 CC Wilf et al. 2003, 2005 LH02-1083 TY020 Complete 7.21 Direct this study LH02-1084 TY020 Complete 7.76 CC Wilf et al. 2003, 2005 LH02-1139 TY020 Complete 7.43 CC Wilf et al. 2003, 2005 LH02-1258 TY020 Complete 6.96 CC Wilf et al. 2003, 2005 LH02-1282 TY020 Complete 6.68 CC Wilf et al. 2003, 2005 LH03-0004 TY020 Complete 6.78 CC Wilf et al. 2003, 2005 LH03-0016 TY020 Complete 7.35 CC Wilf et al. 2003, 2005 LH04-0012 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0015 TY020 Complete 7.65 CC Wilf et al. 2003, 2005 LH04-0020 TY020 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-0022 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0023 TY020 Complete 7.03 CC Wilf et al. 2003, 2005 LH04-0031 TY020 Complete 7.73 CC Wilf et al. 2003, 2005

65

Specimen field # Morphotype Leaf completeness ln Area Method Source LH04-0037 TY020 Complete 7.99 CC Wilf et al. 2003, 2005 LH04-0038 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0039 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0040 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0048 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0050 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0058 TY020 Complete 7.86 CC Wilf et al. 2003, 2005 LH04-0060 TY020 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-0063 TY020 Complete 7.58 CC Wilf et al. 2003, 2005 LH04-0065 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0066 TY020 Complete 7.64 CC Wilf et al. 2003, 2005 LH04-0068 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0075 TY020 Complete 6.78 CC Wilf et al. 2003, 2005 LH04-0080 TY020 Complete 7.23 CC Wilf et al. 2003, 2005 LH04-0088 TY020 Complete 6.24 CC Wilf et al. 2003, 2005 LH04-0105 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0111 TY020 Complete 7.17 CC Wilf et al. 2003, 2005 LH04-0113 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0123 TY020 Complete 7.93 CC Wilf et al. 2003, 2005 LH04-0124 TY020 Complete 7.70 CC Wilf et al. 2003, 2005 LH04-0130 TY020 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-0171 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0189 TY020 Complete 6.33 CC Wilf et al. 2003, 2005 LH04-0192 TY020 Complete 5.95 CC Wilf et al. 2003, 2005 LH04-0197 TY020 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-0219 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-1013 TY020 Complete 7.70 CC Wilf et al. 2003, 2005 LH04-1045 TY020 Complete 6.91 CC Wilf et al. 2003, 2005 LH04-1119 TY020 Complete 7.36 Direct this study LH04-1123 TY020 Complete 8.21 CC Wilf et al. 2003, 2005 LH04-1155 TY020 Complete 6.84 Direct this study LH04-1182 TY020 Complete 6.97 CC Wilf et al. 2003, 2005 LH04-1201 TY020 Complete 6.63 CC Wilf et al. 2003, 2005 LH04-1233 TY020 Complete 6.73 CC Wilf et al. 2003, 2005 LH04-1239 TY020 Complete 6.82 CC Wilf et al. 2003, 2005 LH04-1246 TY020 Complete 6.93 CC Wilf et al. 2003, 2005 LH04-1299 TY020 Complete 5.38 CC Wilf et al. 2003, 2005 LH04-1319 TY020 Complete 8.91 Direct this study LH06-0033 TY020 Complete 7.73 CC Wilf et al. 2003, 2005 LH06-0155 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0156 TY020 Complete 7.46 CC Wilf et al. 2003, 2005 LH06-0237 TY020 Complete 6.58 CC Wilf et al. 2003, 2005

66

Specimen field # Morphotype Leaf completeness ln Area Method Source LH06-0245 TY020 Complete 7.46 CC Wilf et al. 2003, 2005 LH06-1034 TY020 Complete 6.65 CC Wilf et al. 2003, 2005 LH06-1082 TY020 Complete 6.29 CC Wilf et al. 2003, 2005 LH06-1131 TY020 Complete 6.38 CC Wilf et al. 2003, 2005 LH06-1167 TY020 Complete 7.33 CC Wilf et al. 2003, 2005 LH06-1246 TY020 Complete 6.84 CC Wilf et al. 2003, 2005 LH08-0004 TY020 Complete 7.03 CC Wilf et al. 2003, 2005 LH13-0004 TY020 Complete 6.61 CC Wilf et al. 2003, 2005 LH13-0050 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0179 TY020 Complete 7.17 CC Wilf et al. 2003, 2005 LH13-0188 TY020 Complete 8.01 SC Wilf et al. 2003, 2005 LH13-0241 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0287 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-1039 TY020 Complete 5.26 Direct this study LH13-1410 TY020 Complete 7.29 CC Wilf et al. 2003, 2005 LH15-0082 TY020 Complete 8.04 CC Wilf et al. 2003, 2005 LH16-0044 TY020 Complete 7.56 CC Wilf et al. 2003, 2005 LH17-0002 TY020 Complete 7.00 CC Wilf et al. 2003, 2005 LH17-0003 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH17-0009 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH17-0011 TY020 Complete 6.51 SC Wilf et al. 2003, 2005 LH17-0039 TY020 Complete 7.25 CC Wilf et al. 2003, 2005 LH18-0014 TY020 Complete 7.87 CC Wilf et al. 2003, 2005 LH20-0046 TY020 Fragment 9.62 SC this study LH22-0102 TY020 Complete 8.60 CC Wilf et al. 2003, 2005 LH27(2009)-0221 TY020 Complete 6.10 Direct this study LH4(2009)-0435 TY020 Complete 4.79 Direct this study LH4(2009)-0437 TY020 Largely complete 7.83 Direct this study MPEF-Pb 1450 TY020 Complete 6.41 Direct this study LH01-0021 TY021 Complete 4.32 SC Wilf et al. 2003, 2005 LH01-0022 TY021 Complete 6.33 CC Wilf et al. 2003, 2005 LH02-0023 TY021 Fragment 6.51 SC this study LH02-0166 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0209 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-1037 TY021 Complete 6.31 CC Wilf et al. 2003, 2005 LH02-1050 TY021 Complete 6.41 CC Wilf et al. 2003, 2005 LH02-1111 TY021 Complete 4.93 CC Wilf et al. 2003, 2005 LH02-1240 TY021 Complete 4.20 CC Wilf et al. 2003, 2005 LH02-1383 TY021 Complete 5.45 CC Wilf et al. 2003, 2005 LH03-0011 TY021 Complete 6.04 CC Wilf et al. 2003, 2005 LH04-0010 TY021 Complete 6.37 CC Wilf et al. 2003, 2005 LH04-0035 TY021 Complete 6.51 SC Wilf et al. 2003, 2005

67

Specimen field # Morphotype Leaf completeness ln Area Method Source LH04-0084 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0177 TY021 Complete 4.32 SC Wilf et al. 2003, 2005 LH04-1043 TY021 Complete 6.04 CC Wilf et al. 2003, 2005 LH04-1138 TY021 Complete 5.08 CC Wilf et al. 2003, 2005 LH05-0001 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0056 TY021 Complete 4.32 SC Wilf et al. 2003, 2005 LH06-0068 TY021 Complete 6.29 CC Wilf et al. 2003, 2005 LH06-0081 TY021 Complete 4.32 SC Wilf et al. 2003, 2005 LH06-0113 TY021 Complete 4.32 SC Wilf et al. 2003, 2005 LH06-0127 TY021 Complete 4.32 SC Wilf et al. 2003, 2005 LH06-0131 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0141 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0153 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0165 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0199 TY021 Complete 6.49 CC Wilf et al. 2003, 2005 LH06-0204 TY021 Complete 5.80 SC Wilf et al. 2003, 2005 LH06-0229 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0230 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0247 TY021 Complete 5.98 CC Wilf et al. 2003, 2005 LH06-0262 TY021 Complete 5.59 CC Wilf et al. 2003, 2005 LH06-0270 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0275 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0280 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0303 TY021 Complete 6.66 CC Wilf et al. 2003, 2005 LH06-0306 TY021 Complete 6.28 CC Wilf et al. 2003, 2005 LH06-0309 TY021 Complete 6.15 CC Wilf et al. 2003, 2005 LH06-1089 TY021 Complete 6.33 CC Wilf et al. 2003, 2005 LH06-1108 TY021 Complete 6.09 CC Wilf et al. 2003, 2005 LH06-1123 TY021 Complete 6.23 CC Wilf et al. 2003, 2005 LH06-1203 TY021 Complete 5.77 CC Wilf et al. 2003, 2005 LH06-1210 TY021 Complete 5.28 CC Wilf et al. 2003, 2005 LH06-1216 TY021 Complete 5.85 Direct this study LH06-1275 TY021 Complete 5.68 CC Wilf et al. 2003, 2005 LH06-1282 TY021 Complete 4.81 CC Wilf et al. 2003, 2005 LH10-0010 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0003 TY021 Complete 5.17 Direct this study LH13-0013 TY021 Complete 4.32 SC Wilf et al. 2003, 2005 LH13-0019 TY021 Complete 6.69 Direct this study LH13-0022 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0025 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0026 TY021 Complete 6.04 CC Wilf et al. 2003, 2005 LH13-0053 TY021 Complete 6.51 SC Wilf et al. 2003, 2005

68

Specimen field # Morphotype Leaf completeness ln Area Method Source LH13-0085 TY021 Complete 8.01 SC Wilf et al. 2003, 2005 LH13-0088 TY021 Complete 6.40 CC Wilf et al. 2003, 2005 LH13-0089 TY021 Complete 5.92 Direct this study LH13-0111 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0112 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0141 TY021 Complete 6.25 CC Wilf et al. 2003, 2005 LH13-0146 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0147 TY021 Complete 5.83 CC Wilf et al. 2003, 2005 LH13-0149 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0160 TY021 Complete 6.07 CC Wilf et al. 2003, 2005 LH13-0170 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0171 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0189 TY021 Complete 4.32 SC Wilf et al. 2003, 2005 LH13-0194 TY021 Complete 6.24 CC Wilf et al. 2003, 2005 LH13-0197 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0199 TY021 Complete 6.06 CC Wilf et al. 2003, 2005 LH13-0205 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0207 TY021 Complete 6.62 CC Wilf et al. 2003, 2005 LH13-0208 TY021 Complete 6.00 CC Wilf et al. 2003, 2005 LH13-0219 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0220 TY021 Complete 6.35 CC Wilf et al. 2003, 2005 LH13-0222 TY021 Complete 6.15 CC Wilf et al. 2003, 2005 LH13-0223 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0226 TY021 Complete 5.62 CC Wilf et al. 2003, 2005 LH13-0240 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0249 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0252 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0291 TY021 Complete 3.62 CC Wilf et al. 2003, 2005 LH13-0293 TY021 Complete 5.77 CC Wilf et al. 2003, 2005 LH13-0300 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0307 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0506 TY021 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-1020 TY021 Complete 6.12 CC Wilf et al. 2003, 2005 LH13-1051 TY021 Complete 6.25 CC Wilf et al. 2003, 2005 LH13-1186 TY021 Complete 6.09 Direct this study LH13-1259 TY021 Largely complete 6.07 Direct this study LH13-1283 TY021 Complete 6.66 CC Wilf et al. 2003, 2005 LH13-1401 TY021 Complete 5.68 CC Wilf et al. 2003, 2005 LH13-1468 TY021 Complete 7.04 CC Wilf et al. 2003, 2005 LH13-1472 TY021 Complete 7.14 CC Wilf et al. 2003, 2005 LH13-1497 TY021 Complete 5.61 CC Wilf et al. 2003, 2005 LH15-0029 TY021 Complete 6.63 CC Wilf et al. 2003, 2005

69

Specimen field # Morphotype Leaf completeness ln Area Method Source LH16-0500 TY021 Complete 4.32 SC Wilf et al. 2003, 2005 LH17-1005 TY021 Complete- poor vein preservation 5.22 Direct this study LH22-0003 TY021 Complete 5.74 CC Wilf et al. 2003, 2005 LH23-0002 TY021 Complete 5.85 CC Wilf et al. 2003, 2005 LH02-1034 TY022 Complete 6.77 Direct this study LH04-0087 TY022 Complete 7.87 Direct this study LH04-0166 TY022 Complete 6.66 CC Wilf et al. 2003, 2005 LH04-1033 TY022 Complete 6.78 CC Wilf et al. 2003, 2005 LH02-0150 TY023 Complete 8.14 CC Wilf et al. 2003, 2005 LH02-0156 TY023 Complete 9.11 SC Wilf et al. 2003, 2005 LH02-1071 TY023 Complete 8.72 CC Wilf et al. 2003, 2005 LH02-1212 TY023 Complete 6.09 CC Wilf et al. 2003, 2005 LH04-0091 TY023 Complete 7.89 CC Wilf et al. 2003, 2005 LH04-0106 TY023 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-0122 TY023 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-1143 TY023 Complete 8.21 CC Wilf et al. 2003, 2005 LH06-0019 TY023 Complete 7.61 CC Wilf et al. 2003, 2005 LH06-0167 TY023 Complete 7.90 Direct this study LH06-0257 TY023 Complete 8.10 CC Wilf et al. 2003, 2005 LH06-0294 TY023 Complete 8.56 CC Wilf et al. 2003, 2005 LH06-0295 TY023 Complete 7.83 CC Wilf et al. 2003, 2005 LH06-0310 TY023 Complete 7.88 CC Wilf et al. 2003, 2005 LH06-0317 TY023 Complete 6.07 CC Wilf et al. 2003, 2005 LH06-1161 TY023 Complete 8.49 CC Wilf et al. 2003, 2005 LH06-1186 TY023 Complete 8.41 CC Wilf et al. 2003, 2005 LH06-1241 TY023 Fragment 8.65 SC this study LH11-0001 TY023 Complete 9.15 CC Wilf et al. 2003, 2005 LH13-0169 TY023 Complete 7.78 CC Wilf et al. 2003, 2005 LH13-1178 TY023 Complete 7.55 CC Wilf et al. 2003, 2005 LH13-1324 TY023 Fragment 7.21 SC this study LH15-0017 TY023 Complete 9.13 CC Wilf et al. 2003, 2005 LH15-0032 TY023 Complete 8.53 CC Wilf et al. 2003, 2005 LH15-0045 TY023 Complete 7.02 CC Wilf et al. 2003, 2005 LH22-0015 TY023 Complete 9.19 CC Wilf et al. 2003, 2005 LH22-0526 TY023 Complete 9.06 CC Wilf et al. 2003, 2005 LH4(2009)-0441 TY023 Complete 7.06 Direct this study LHF-0016 TY023 Complete 6.99 Direct this study LH17-0001 TY024 Fragment 8.50 SC this study LH17-0033 TY024 Fragment 7.24 VS this study LH02-0116 TY025 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-1000 TY025 Complete 6.64 CC Wilf et al. 2003, 2005 LH03-0020 TY025 Complete 7.07 CC Wilf et al. 2003, 2005

70

Specimen field # Morphotype Leaf completeness ln Area Method Source LH04-0026 TY025 Complete 6.72 CC Wilf et al. 2003, 2005 LH04-0042 TY025 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0056 TY025 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0071 TY025 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0095 TY025 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0119 TY025 Complete 6.63 CC Wilf et al. 2003, 2005 LH04-0144 TY025 Complete 7.03 CC Wilf et al. 2003, 2005 LH04-0149 TY025 Complete 6.51 CC Wilf et al. 2003, 2005 LH04-0150 TY025 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0207 TY025 Complete 6.85 CC Wilf et al. 2003, 2005 LH04-1036 TY025 Complete 6.22 Direct this study LH04-1289 TY025 Complete 7.98 CC Wilf et al. 2003, 2005 LH04-1306 TY025 Complete 7.10 CC Wilf et al. 2003, 2005 LH04(2006)-0003a TY025 Largely complete 7.36 Direct this study LH05-50' TY025 Complete 4.32 SC Wilf et al. 2003, 2005 LH06-0001 TY025 Largely complete 7.81 Direct this study LH06-1152 TY025 Complete 7.40 CC Wilf et al. 2003, 2005 LH13-0232 TY025 Complete 4.32 SC Wilf et al. 2003, 2005 LH17-0054 TY025 Complete 6.51 SC Wilf et al. 2003, 2005 LH17-0055 TY025 Complete 6.51 SC Wilf et al. 2003, 2005 LH17-0056 TY025 Complete 6.51 SC Wilf et al. 2003, 2005 LH22-0100 TY025 Complete 8.07 CC Wilf et al. 2003, 2005 LH22-1009 TY025 Complete 6.82 CC Wilf et al. 2003, 2005 LH13-0073 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-1240-1 TY026 Complete 5.56 CC Wilf et al. 2003, 2005 LH15-0024 TY026 Complete 6.94 Direct this study LH15-0063 TY026 Complete 7.24 CC Wilf et al. 2003, 2005 LH15-0064 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0065 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0066 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0067 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0068 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0069 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0070 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0071 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0072 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0073 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0074 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0086 TY026 Complete 8.01 SC Wilf et al. 2003, 2005 LH15-0087 TY026 Complete 8.01 SC Wilf et al. 2003, 2005 LH15-0091 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0092 TY026 Complete 6.51 SC Wilf et al. 2003, 2005

71

Specimen field # Morphotype Leaf completeness ln Area Method Source LH15-0093 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0094 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0095 TY026 Complete 6.28 CC Wilf et al. 2003, 2005 LH15-0096 TY026 Complete 6.32 CC Wilf et al. 2003, 2005 LH15-0502 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0504 TY026 Complete 6.44 CC Wilf et al. 2003, 2005 LH15-0506 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-24(1) TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-24(2) TY026 Complete 7.09 CC Wilf et al. 2003, 2005 LH15-24(3) TY026 Complete 7.09 CC Wilf et al. 2003, 2005 LH15-24(4) TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-24(5) TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-24(6) TY026 Complete 7.08 CC Wilf et al. 2003, 2005 LH15-24(7) TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-91-96 TY026 Complete- poor vein preservation 6.23 Direct this study LH15(2009)-0376b TY026 Complete 5.74 Direct this study LH25-0500 TY026 Complete 7.00 CC Wilf et al. 2003, 2005 LH25-0501 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH25-0502 TY026 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0001 TY027 Complete 6.58 Direct this study LH02-0196 TY028 Complete 7.41 Direct this study LH02-0198 TY028 Complete 7.76 CC Wilf et al. 2003, 2005 LH02-1162 TY028 Fragment 8.41 VS this study LH13-0145 TY029 Complete 7.96 CC Wilf et al. 2003, 2005 LH13-1130 TY029 Complete 8.33 CC Wilf et al. 2003, 2005 LH13-1323 TY029 Complete 7.48 Direct this study LH13-1539 TY029 Complete 6.78 CC Wilf et al. 2003, 2005 LH22-0021 TY029 Complete 8.43 CC Wilf et al. 2003, 2005 LH22-0022 TY029 Complete 7.34 CC Wilf et al. 2003, 2005 LH23-0045' TY029 Fragment 8.52 VS this study LH02-1108 TY030 Complete 7.44 CC Wilf et al. 2003, 2005 LH04-1162 TY030 Complete 7.93 CC Wilf et al. 2003, 2005 LH06-0162 TY030 Complete 8.01 SC Wilf et al. 2003, 2005 LH06-0162b TY030 Fragment 9.37 VS this study LH06-1132 TY030 Complete 7.78 CC Wilf et al. 2003, 2005 LH06-1180b TY030 Complete 7.44 Direct this study LH06-1268 TY030 Complete 7.74 CC Wilf et al. 2003, 2005 LH13-0235 TY030 Complete 8.01 SC Wilf et al. 2003, 2005 LH13-1007 TY030 Complete 7.95 CC Wilf et al. 2003, 2005 LH13-1424 TY030 Complete 8.21 Direct this study LH13-1486 TY030 Complete 8.39 CC Wilf et al. 2003, 2005 LH20-0022 TY030 Complete 6.51 SC Wilf et al. 2003, 2005

72

Specimen field # Morphotype Leaf completeness ln Area Method Source LH22-1006 TY030 Fragment 9.46 VS this study LH25-0011 TY030 Complete 7.63 Direct this study LH25-0012 TY030 Fragment 8.88 VS this study LH13-0001 TY032 Complete 8.96 Direct this study LH13-0033 TY032 Complete 8.01 SC Wilf et al. 2003, 2005 LH13-0055 TY032 Complete 9.11 SC Wilf et al. 2003, 2005 LH13-0128 TY032 Complete 8.24 CC Wilf et al. 2003, 2005 LH13-0212 TY032 Complete 8.19 CC Wilf et al. 2003, 2005 LH13-1171 TY032 Fragment 8.82 VS this study LH13-1463 TY032 Fragment 9.18 VS this study LH23-0055 TY032 Complete 9.11 SC Wilf et al. 2003, 2005 LH01-0017 TY035 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0002 TY035 Complete 7.81 CC Wilf et al. 2003, 2005 LH02-0202 TY035 Complete 7.31 CC Wilf et al. 2003, 2005 LH02-0259 TY035 Complete 7.65 Direct this study LH02-1051 TY035 Fragment 7.33 VS this study LH04-0125 TY035 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0050 TY035 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0248 TY035 Complete 8.01 SC Wilf et al. 2003, 2005 LH06-1107 TY035 Complete 7.03 CC Wilf et al. 2003, 2005 LH06-1185 TY035 Complete 7.44 CC Wilf et al. 2003, 2005 LH06-1217 TY035 Complete 7.55 CC Wilf et al. 2003, 2005 LH06-1273 TY035 Complete 7.19 CC Wilf et al. 2003, 2005 LH13-0086 TY035 Complete 6.95 CC Wilf et al. 2003, 2005 LH25-0002 TY035 Complete 7.73 CC Wilf et al. 2003, 2005 LH4(2009)-0429 TY035 Complete 6.86 Direct this study LH02-0096 TY036 Largely complete 7.37 Direct this study LH02-0169 TY037 Fragment 8.54 VS this study LH20-0021 TY037 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0026 TY038 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0051 TY038 Complete 6.46 CC Wilf et al. 2003, 2005 LH02-0085 TY038 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0152 TY038 Complete 6.48 Direct this study LH02-0219 TY038 Complete 7.67 CC Wilf et al. 2003, 2005 LH13-0118 TY038 Fragment 8.13 VS this study LH13-0271 TY038 Complete 7.18 Direct this study LH06-0013 TY039 Fragment 8.01 SC this study LHF-0039 TY040 Largely complete 9.82 Direct this study LH02-0186 TY041 Complete 5.89 CC Wilf et al. 2003, 2005 LH02-0212 TY041 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0338 TY041 Complete 6.68 CC Wilf et al. 2003, 2005 LH02-1030 TY041 Complete 7.53 CC Wilf et al. 2003, 2005

73

Specimen field # Morphotype Leaf completeness ln Area Method Source LH02-1189 TY041 Complete 6.17 CC Wilf et al. 2003, 2005 LH02-1208 TY041 Complete 6.06 Direct this study LH02-1267 TY041 Complete 6.95 Direct this study LH04-0110 TY041 Complete 6.00 CC Wilf et al. 2003, 2005 LH06-0012 TY041 Complete 7.13 CC Wilf et al. 2003, 2005 LH06-0061 TY041 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0304 TY041 Complete 7.31 CC Wilf et al. 2003, 2005 LH06-0325 TY041 Complete 6.76 CC Wilf et al. 2003, 2005 LH06-1074 TY041 Complete 5.89 CC Wilf et al. 2003, 2005 LH13-0043 TY041 Complete 7.38 CC Wilf et al. 2003, 2005 LH13-0272 TY041 Fragment 6.51 SC this study LH22-0028 TY041 Complete 6.81 Direct this study LH02-0176 TY042 Complete 6.39 CC Wilf et al. 2003, 2005 LH02-1185 TY042 Complete 6.47 CC Wilf et al. 2003, 2005 LH04-0043 TY042 Complete 7.02 CC Wilf et al. 2003, 2005 LH04-0093 TY042 Complete 7.22 CC Wilf et al. 2003, 2005 LH04-0128 TY042 Fragment 8.33 VS this study LH04-0190 TY042 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0194 TY042 Complete 6.91 Direct this study LH04-0202 TY042 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0222 TY042 Complete 7.38 CC Wilf et al. 2003, 2005 LH04-1063 TY042 Complete 5.79 CC Wilf et al. 2003, 2005 LH04-1085 TY042 Complete 6.66 CC Wilf et al. 2003, 2005 LH04-1104 TY042 Complete 6.90 CC Wilf et al. 2003, 2005 LH04-1130 TY042 Complete 7.50 CC Wilf et al. 2003, 2005 LH04-1158 TY042 Complete 7.15 CC Wilf et al. 2003, 2005 LH04-1172 TY042 Complete 7.17 CC Wilf et al. 2003, 2005 LH04-1256 TY042 Complete 7.21 CC Wilf et al. 2003, 2005 LH06-0160 TY042 Largely complete 8.08 Direct this study LH13-0102 TY042 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-1501 TY042 Complete 7.06 CC Wilf et al. 2003, 2005 LH16-0024 TY042 Complete 6.88 CC Wilf et al. 2003, 2005 LH16-0025 TY042 Complete 8.01 SC Wilf et al. 2003, 2005 LH02-0006 TY043 Complete 7.39 CC Wilf et al. 2003, 2005 LH02-0006b TY043 Fragment 7.32 VS this study LH02-0031 TY043 Complete 8.01 SC Wilf et al. 2003, 2005 LH02-0075 TY043 Complete 6.73 CC Wilf et al. 2003, 2005 LH02-0153 TY043 Complete 7.58 CC Wilf et al. 2003, 2005 LH02-0555 TY043 Fragment 7.34 VS this study LH02-1249 TY043 Complete 5.33 CC Wilf et al. 2003, 2005 LH02-1350 TY043 Complete 5.93 CC Wilf et al. 2003, 2005 LH02-0014 TY044 Complete 9.11 SC Wilf et al. 2003, 2005

74

Specimen field # Morphotype Leaf completeness ln Area Method Source LH02-0040 TY044 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0013 TY044 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-0021 TY044 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0027 TY044 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-0112 TY044 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-1047 TY044 Fragment 8.01 SC this study LH04-1106 TY044 Largely complete 8.03 Direct this study LH13-1123 TY044 Fragment 8.27 VS this study LH16-0011 TY044 Complete 6.51 SC Wilf et al. 2003, 2005 LH16-0059 TY044 Complete 6.51 SC Wilf et al. 2003, 2005 LH17-0061 TY044 Complete 8.01 SC Wilf et al. 2003, 2005 LH20-1036 TY044 Fragment 6.51 SC this study LH21-0001 TY044 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0010 TY045 Fragment 9.38 VS this study LH02-17' TY045 Complete 9.11 SC Wilf et al. 2003, 2005 LH04-1222 TY045 Fragment 8.80 VS this study LH04-1244 TY045 Fragment 8.69 VS this study LH02-1012 TY046 Complete 8.62 CC Wilf et al. 2003, 2005 LH04-1035 TY046 Complete 8.19 CC Wilf et al. 2003, 2005 LH06-1190 TY046 Fragment 9.25 VS this study MPEF-0195 TY046 Fragment 9.88 VS this study LHF-0038 TY047 Complete 8.35 Direct this study LH12(1999)-0001 TY048 Fragment 7.95 VS this study LH02-0254 TY050 Complete 8.63 CC Wilf et al. 2003, 2005 LH02-1331 TY050 Fragment 8.91 VS this study LH02-254a TY050 Fragment 9.05 VS this study LH06-0215 TY053 Complete 7.81 CC Wilf et al. 2003, 2005 LH23-0045 TY053 Complete 7.67 Direct this study LH02-0110 TY054 Complete 8.03 Direct this study LH02-1080 TY054 Largely complete 6.60 Direct this study LH02-1271 TY054 Complete 4.94 CC Wilf et al. 2003, 2005 LH04-0098 TY054 Complete 6.96 CC Wilf et al. 2003, 2005 LH04-1253 TY054 Complete 6.81 CC Wilf et al. 2003, 2005 LH06-0027 TY054 Largely complete 6.90 Direct this study LH15-0047 TY054 Complete 5.49 CC Wilf et al. 2003, 2005 LH06-0276 TY055 Fragment 7.93 VS this study LH02-0037 TY056 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0267 TY056 Complete 9.11 SC Wilf et al. 2003, 2005 LH02-1007 TY056 Fragment 8.98 VS this study LH02-1009 TY056 Complete 7.14 CC Wilf et al. 2003, 2005 LH02-1063 TY056 Complete 6.97 CC Wilf et al. 2003, 2005 LH02-1065 TY056 Complete 7.07 CC Wilf et al. 2003, 2005

75

Specimen field # Morphotype Leaf completeness ln Area Method Source LH03-0017 TY056 Complete 7.55 CC Wilf et al. 2003, 2005 LH03-0028 TY056 Complete 6.87 CC Wilf et al. 2003, 2005 LH04-0172 TY056 Complete 6.44 CC Wilf et al. 2003, 2005 LH04-0178 TY056 Complete 7.21 CC Wilf et al. 2003, 2005 LH04-1133 TY056 Complete 7.11 CC Wilf et al. 2003, 2005 LH04-1146 TY056 Complete 7.34 CC Wilf et al. 2003, 2005 LH06-0017 TY056 Complete 7.55 CC Wilf et al. 2003, 2005 LH06-0119 TY056 Complete 7.22 CC Wilf et al. 2003, 2005 LH06-0292b TY056 Complete 7.66 Direct this study LH06-1195 TY056 Complete 6.75 CC Wilf et al. 2003, 2005 LH06-1304 TY056 Complete 6.91 CC Wilf et al. 2003, 2005 LH11-0002 TY056 Complete 7.78 CC Wilf et al. 2003, 2005 LH13-1072 TY056 Complete 7.96 Direct this study LH17-12' TY056 Complete 6.51 SC Wilf et al. 2003, 2005 LH24-0002 TY056 Complete 7.94 CC Wilf et al. 2003, 2005 LH27(2009)-0243a TY056 Complete 7.41 Direct this study LH27(2009)-0244b TY056 Complete 6.61 Direct this study LH13-1061 TY057 Complete 7.92 CC Wilf et al. 2003, 2005 LH13-1145 TY057 Fragment 8.43 VS this study LH13-1389 TY057 Complete 7.13 CC Wilf et al. 2003, 2005 LH13-1419 TY057 Complete 7.94 CC Wilf et al. 2003, 2005 LH19-0001 TY057 Fragment 8.58 VS this study LH15-0036 TY058 Complete 6.19 Direct this study LH04-0161 TY059 Fragment 7.66 VS this study LH24-0015 TY060 Complete 5.50 Direct this study LH25-0015 TY060 Complete 6.96 Direct this study LH04-0102 TY061 Complete 6.41 CC Wilf et al. 2003, 2005 LH04-0215 TY061 Complete 6.51 SC Wilf et al. 2003, 2005 LH05-0002 TY061 Complete 5.84 CC Wilf et al. 2003, 2005 LH06-0031 TY061 Complete 6.76 CC Wilf et al. 2003, 2005 LH06-0087 TY061 Complete 6.91 CC Wilf et al. 2003, 2005 LH06-0090 TY061 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0099 TY061 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0128 TY061 Complete 7.43 CC Wilf et al. 2003, 2005 LH06-0133 TY061 Complete 6.55 CC Wilf et al. 2003, 2005 LH06-0139 TY061 Complete 7.14 CC Wilf et al. 2003, 2005 LH06-0166 TY061 Complete 7.00 CC Wilf et al. 2003, 2005 LH06-0185 TY061 Complete 7.17 CC Wilf et al. 2003, 2005 LH06-0193 TY061 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0232 TY061 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0250 TY061 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0261 TY061 Complete 7.56 CC Wilf et al. 2003, 2005

76

Specimen field # Morphotype Leaf completeness ln Area Method Source LH06-0266 TY061 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0312 TY061 Complete 5.77 CC Wilf et al. 2003, 2005 LH06-1001 TY061 Complete 7.56 CC Wilf et al. 2003, 2005 LH06-1002 TY061 Complete 6.83 Direct this study LH06-1065 TY061 Complete 7.53 CC Wilf et al. 2003, 2005 LH06-1077 TY061 Complete 7.38 CC Wilf et al. 2003, 2005 LH06-1176 TY061 Complete 7.09 CC Wilf et al. 2003, 2005 LH06-1191 TY061 Complete 7.39 CC Wilf et al. 2003, 2005 LH06-1224 TY061 Complete 5.30 CC Wilf et al. 2003, 2005 LH06-1260 TY061 Complete 7.94 CC Wilf et al. 2003, 2005 LH06-1269 TY061 Complete 6.12 CC Wilf et al. 2003, 2005 LH13-0009 TY061 Complete 6.72 CC Wilf et al. 2003, 2005 LH13-0012 TY061 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0020 TY061 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0048 TY061 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0076 TY061 Complete 6.21 CC Wilf et al. 2003, 2005 LH13-0080 TY061 Complete 6.79 CC Wilf et al. 2003, 2005 LH13-0098 TY061 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0116 TY061 Complete 6.28 CC Wilf et al. 2003, 2005 LH13-0165 TY061 Complete 6.42 CC Wilf et al. 2003, 2005 LH13-0193 TY061 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0202 TY061 Complete 6.58 CC Wilf et al. 2003, 2005 LH13-0210 TY061 Complete 5.85 Direct this study LH13-0214 TY061 Complete 6.38 CC Wilf et al. 2003, 2005 LH13-0215 TY061 Complete 7.52 CC Wilf et al. 2003, 2005 LH13-0267 TY061 Complete 6.90 CC Wilf et al. 2003, 2005 LH13-1009 TY061 Complete 5.70 Direct this study LH13-1137 TY061 Complete 6.36 Direct this study LH13-1253 TY061 Complete 6.67 Direct this study LH13-1325 TY061 Complete 6.14 CC Wilf et al. 2003, 2005 LH13-1335 TY061 Complete 5.77 CC Wilf et al. 2003, 2005 LH13-1442 TY061 Complete 5.96 Direct this study LH13-1458 TY061 Complete 7.56 CC Wilf et al. 2003, 2005 LH13-1573 TY061 Complete 7.03 Direct this study LH20-0035 TY061 Complete 6.90 CC Wilf et al. 2003, 2005 LH22-0005 TY061 Complete 6.89 CC Wilf et al. 2003, 2005 LH22-0008 TY061 Complete 6.97 CC Wilf et al. 2003, 2005 LH22-0010 TY061 Complete 4.97 CC Wilf et al. 2003, 2005 LH22-0106 TY061 Complete 6.91 CC Wilf et al. 2003, 2005 LH25-0008 TY061 Complete 6.62 CC Wilf et al. 2003, 2005 LH25-0029 TY061 Complete 6.32 CC Wilf et al. 2003, 2005 LH25-0031 TY061 Complete 6.40 CC Wilf et al. 2003, 2005

77

Specimen field # Morphotype Leaf completeness ln Area Method Source LH25-0033 TY061 Complete 7.57 CC Wilf et al. 2003, 2005 LH27(2009)-0249 TY061 Fragment 4.32 SC this study LH15-0025 TY062 Complete 8.43 CC Wilf et al. 2003, 2005 LH15-0025a TY062 Complete 8.56 Direct this study LH15-0038 TY064 Complete 6.03 Direct this study LH01-0020 TY066 Fragment 7.87 VS this study LH13-1139 TY066 Complete 6.27 Direct this study LH13-1228 TY066 Complete 8.27 CC Wilf et al. 2003, 2005 LH13-1480 TY066 Complete 6.80 CC Wilf et al. 2003, 2005 LH02-0324 TY067 Fragment 7.79 VS this study LH25-0010 TY068 Complete 7.38 Direct this study LH06-0073 TY069 Complete 6.87 CC Wilf et al. 2003, 2005 LH02-0149 TY070 Complete 7.74 CC Wilf et al. 2003, 2005 LH03-0010 TY070 Complete 7.48 CC Wilf et al. 2003, 2005 LH18-0007 TY070 Fragment 8.55 VS this study LH18-0008 TY070 Complete 8.45 CC Wilf et al. 2003, 2005 LH18-0009 TY070 Complete 8.56 CC Wilf et al. 2003, 2005 LH18-0010 TY070 Complete 9.11 SC Wilf et al. 2003, 2005 LH02-0069 TY072 Fragment 6.51 SC this study LH13-0094 TY073 Complete 7.39 CC Wilf et al. 2003, 2005 LH20-0033 TY074 Fragment 9.44 VS this study LH02-0021 TY075 Fragment 6.51 SC this study LH02-1024 TY075 Fragment 4.32 SC this study LH02-1140 TY075 Fragment 6.51 SC this study LH13-1331 TY076 Complete 7.48 Direct this study LH13-1447 TY076 Complete 7.46 Direct this study LH23-0048 TY076 Fragment 8.83 VS this study LH02-0091 TY077 Largely complete 6.43 Direct this study LH06-0146 TY079 Complete 7.01 Direct this study LH13-1531 TY079 Fragment 8.17 VS this study LH17-0028 TY080 Complete 6.26 Direct this study LH01-0008 TY081 Complete 7.85 CC Wilf et al. 2003, 2005 LH20-0047 TY081 Complete 6.64 CC Wilf et al. 2003, 2005 LH21-0006 TY081 Fragment 8.88 VS this study LH15-0052 TY082 Fragment 7.51 VS this study LH02-0127 TY083 Fragment 9.19 VS this study LH01-0011 TY084 Largely complete 7.07 Direct this study LH01-0024 TY084 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0123 TY084 Fragment 8.66 VS this study LH02-0201 TY084 Complete 6.17 CC Wilf et al. 2003, 2005 LH02-0218 TY084 Largely complete 8.24 Direct this study LH02-0271 TY084 Complete 5.49 Direct this study

78

Specimen field # Morphotype Leaf completeness ln Area Method Source LH02-0298 TY084 Complete 5.81 CC Wilf et al. 2003, 2005 LH02-0550 TY084 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-1005 TY084 Complete 7.51 CC Wilf et al. 2003, 2005 LH02-1017 TY084 Largely complete 6.98 Direct this study LH02-1045 TY084 Fragment 8.19 VS this study LH02-1048 TY084 Complete 6.83 Direct this study LH02-1054 TY084 Complete 5.89 CC Wilf et al. 2003, 2005 LH02-1163 TY084 Complete 6.25 CC Wilf et al. 2003, 2005 LH02-1197 TY084 Complete 6.63 CC Wilf et al. 2003, 2005 LH02-1333 TY084 Complete 6.86 CC Wilf et al. 2003, 2005 LH03-0015 TY084 Complete 6.51 SC Wilf et al. 2003, 2005 LH03-0021 TY084 Complete 6.68 CC Wilf et al. 2003, 2005 LH04-0117 TY084 Complete 6.76 CC Wilf et al. 2003, 2005 LH04-0136 TY084 Complete 6.25 CC Wilf et al. 2003, 2005 LH04-0175 TY084 Complete 7.68 CC Wilf et al. 2003, 2005 LH04-0204 TY084 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-0205 TY084 Complete 8.01 SC Wilf et al. 2003, 2005 LH04-0206 TY084 Complete 7.58 CC Wilf et al. 2003, 2005 LH04-1171 TY084 Complete 8.01 CC Wilf et al. 2003, 2005 LH04-1220 TY084 Complete 7.18 CC Wilf et al. 2003, 2005 LH04-1227 TY084 Complete 7.05 CC Wilf et al. 2003, 2005 LH04-1230 TY084 Complete 7.56 CC Wilf et al. 2003, 2005 LH04-1234 TY084 Complete 6.87 CC Wilf et al. 2003, 2005 LH04-1238 TY084 Complete 6.93 CC Wilf et al. 2003, 2005 LH04-1257 TY084 Complete 6.60 CC Wilf et al. 2003, 2005 LH04-1270 TY084 Complete 5.73 CC Wilf et al. 2003, 2005 LH04-1272 TY084 Complete 5.41 CC Wilf et al. 2003, 2005 LH04-1283 TY084 Complete 6.37 CC Wilf et al. 2003, 2005 LH04-1336 TY084 Complete 7.02 CC Wilf et al. 2003, 2005 LH06-0059 TY084 Complete 8.01 SC Wilf et al. 2003, 2005 LH09-0001 TY084 Complete 7.13 CC Wilf et al. 2003, 2005 LH13-0056 TY084 Complete 7.53 Direct this study LH13-0297 TY084 Complete 6.87 CC Wilf et al. 2003, 2005 LH13-1034 TY084 Complete 7.70 CC Wilf et al. 2003, 2005 LH13-1053 TY084 Complete 7.41 Direct this study LH13-1120 TY084 Complete 6.87 CC Wilf et al. 2003, 2005 LH13-1219 TY084 Complete 7.10 CC Wilf et al. 2003, 2005 LH13-1295 TY084 Complete 7.60 CC Wilf et al. 2003, 2005 LH13-1418 TY084 Complete 7.47 CC Wilf et al. 2003, 2005 LH13-1422 TY084 Complete 7.26 CC Wilf et al. 2003, 2005 LH13-1428 TY084 Complete 7.13 Direct this study LH13-1519 TY084 Complete 6.93 CC Wilf et al. 2003, 2005

79

Specimen field # Morphotype Leaf completeness ln Area Method Source LH13-1567 TY084 Complete 7.76 CC Wilf et al. 2003, 2005 LH13-1601 TY084 Complete 6.98 CC Wilf et al. 2003, 2005 LH13-7112 TY084 Complete 7.61 CC Wilf et al. 2003, 2005 LH13-7120 TY084 Complete 6.06 CC Wilf et al. 2003, 2005 LH14-0003 TY084 Complete 6.51 SC Wilf et al. 2003, 2005 LH15-0042 TY084 Complete 6.91 CC Wilf et al. 2003, 2005 LH15-0079 TY084 Complete 4.91 CC Wilf et al. 2003, 2005 LH16-0028 TY084 Complete 7.03 CC Wilf et al. 2003, 2005 LH16-0040 TY084 Complete 6.37 CC Wilf et al. 2003, 2005 LH17-0031 TY084 Complete 6.55 CC Wilf et al. 2003, 2005 LH21-0003 TY084 Complete 7.17 CC Wilf et al. 2003, 2005 LH22-0011 TY084 Complete 6.51 SC Wilf et al. 2003, 2005 LH22-0025 TY084 Complete 7.98 CC Wilf et al. 2003, 2005 LH24-0018 TY084 Complete 6.51 SC Wilf et al. 2003, 2005 LH25-0007 TY084 Complete 7.30 Direct this study LH25-0014 TY084 Complete 6.51 SC Wilf et al. 2003, 2005 LH4(2009)-0430 TY084 Complete 8.19 Direct this study LH02-0133 TY101 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-1037 TY101 Largely complete 8.50 Direct this study LH06-0238 TY101 Complete 6.38 CC Wilf et al. 2003, 2005 LH13-1435 TY101 Complete 7.35 CC Wilf et al. 2003, 2005 LH16-0043 TY101 Complete 7.58 CC Wilf et al. 2003, 2005 LH24-0004 TY101 Complete 7.06 CC Wilf et al. 2003, 2005 LH02-0083 TY102 Complete 6.63 Direct this study LH02-0120 TY102 Largely complete 5.81 Direct this study LH02-0313 TY103 Complete 6.12 CC Wilf et al. 2003, 2005 LH13-0006 TY104 Complete 6.76 Direct this study LH13-1548 TY104 Fragment 6.59 VS this study LH02-0322 TY105 Complete 8.01 SC Wilf et al. 2003, 2005 LH02-1044 TY105 Complete 8.20 CC Wilf et al. 2003, 2005 LH02-1149 TY105 Complete 7.92 CC Wilf et al. 2003, 2005 LH02-1244 TY105 Complete 6.04 Direct this study LH06-0055 TY105 Complete 7.35 CC Wilf et al. 2003, 2005 LH06-0096 TY105 Complete 8.44 Direct this study LH06-0180 TY105 Complete 8.42 CC Wilf et al. 2003, 2005 LH06-0206 TY105 Largely complete 7.83 Direct this study LH06-1154 TY105 Complete 6.96 CC Wilf et al. 2003, 2005 LH06-1158 TY105 Complete 6.74 CC Wilf et al. 2003, 2005 LH13-0063 TY105 Complete 8.14 CC Wilf et al. 2003, 2005 LH13-0133 TY105 Complete 7.65 CC Wilf et al. 2003, 2005 LH13-0203 TY105 Complete 6.36 Direct this study LH13-1165 TY105 Complete 7.06 CC Wilf et al. 2003, 2005

80

Specimen field # Morphotype Leaf completeness ln Area Method Source LH13-1565 TY105 Complete 7.89 CC Wilf et al. 2003, 2005 LH15-0007 TY105 Complete 8.01 SC Wilf et al. 2003, 2005 LH15-0020 TY105 Complete 9.11 SC Wilf et al. 2003, 2005 LH15-0039 TY105 Complete 8.01 SC Wilf et al. 2003, 2005 LH17-0004 TY105 Complete 8.01 CC Wilf et al. 2003, 2005 LH20-0038 TY105 Complete 6.51 SC Wilf et al. 2003, 2005 LH25-0005 TY105 Complete 7.65 CC Wilf et al. 2003, 2005 LH04-0101 TY106 Complete 6.97 CC Wilf et al. 2003, 2005 LH04-1142 TY106 Fragment 10.91 SC this study LH04(2006)-0050 TY106 Fragment 10.91 SC this study LH13-0204 TY106 Complete 9.11 SC Wilf et al. 2003, 2005 LH13-0306 TY106 Complete 9.11 SC Wilf et al. 2003, 2005 LH13-1143 TY106 Fragment 9.62 VS this study LH13-1227 TY106 Fragment 8.25 VS this study LH15-0002 TY106 Complete 9.11 SC Wilf et al. 2003, 2005 LH15-0080 TY106 Complete 9.43 CC Wilf et al. 2003, 2005 LH20-0032 TY106 Complete 10.90 SC Wilf et al. 2003, 2005 LH23-0024 TY106 Complete 10.90 SC Wilf et al. 2003, 2005 LH23-0031 TY106 Complete 9.11 SC Wilf et al. 2003, 2005 LH23-0052 TY106 Fragment 9.80 VS this study LH27(2009)-0296 TY106 Fragment 7.96 VS this study LHF-0002 TY106 Fragment 8.01 SC this study LH02-0245 TY107 Complete 6.73 Direct this study LH02-1101 TY107 Complete 6.48 CC Wilf et al. 2003, 2005 LH02-1386 TY107 Complete 5.45 CC Wilf et al. 2003, 2005 LH02-1413 TY107 Complete 7.09 CC Wilf et al. 2003, 2005 LH03-0026 TY107 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0030 TY107 Complete 7.04 CC Wilf et al. 2003, 2005 LH04-0033 TY107 Fragment 8.12 VS this study LH04-0054 TY107 Complete 6.80 CC Wilf et al. 2003, 2005 LH04-0070 TY107 Complete 5.86 Direct this study LH04-0074 TY107 Complete 7.44 Direct this study LH04-0079 TY107 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0096 TY107 Complete 7.19 CC Wilf et al. 2003, 2005 LH04-0146 TY107 Complete 6.91 CC Wilf et al. 2003, 2005 LH04-0165 TY107 Complete 7.07 CC Wilf et al. 2003, 2005 LH04-0176 TY107 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0209 TY107 Complete 5.92 CC Wilf et al. 2003, 2005 LH04-0214 TY107 Complete 5.97 CC Wilf et al. 2003, 2005 LH04-0220 TY107 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0221 TY107 Complete 6.73 CC Wilf et al. 2003, 2005 LH04-1048 TY107 Complete 6.40 CC Wilf et al. 2003, 2005

81

Specimen field # Morphotype Leaf completeness ln Area Method Source LH04-1329 TY107 Complete 5.30 CC Wilf et al. 2003, 2005 LH04-1413 TY107 Complete 7.36 CC Wilf et al. 2003, 2005 LH15-1011 TY107 Complete 5.80 CC Wilf et al. 2003, 2005 LH16-0017 TY107 Complete 6.51 SC Wilf et al. 2003, 2005 LH16-0035 TY107 Complete 7.34 CC Wilf et al. 2003, 2005 LH16-0037 TY107 Complete 6.51 SC Wilf et al. 2003, 2005 LH16-0045 TY107 Complete 6.52 CC Wilf et al. 2003, 2005 LH16-0046 TY107 Complete 6.86 CC Wilf et al. 2003, 2005 LH16-0052 TY107 Complete 6.25 CC Wilf et al. 2003, 2005 LH17-0005 TY107 Complete 6.55 CC Wilf et al. 2003, 2005 LH17-0007 TY107 Complete 5.97 CC Wilf et al. 2003, 2005 LH17-0019 TY107 Complete 6.51 SC Wilf et al. 2003, 2005 LH17-0024 TY107 Complete 6.51 SC Wilf et al. 2003, 2005 LH17-0026 TY107 Complete 6.16 CC Wilf et al. 2003, 2005 LH17-0029 TY107 Complete 6.73 CC Wilf et al. 2003, 2005 LH20-0007 TY107 Complete 7.05 CC Wilf et al. 2003, 2005 LH20-0027 TY107 Complete 6.89 CC Wilf et al. 2003, 2005 LH20-0041 TY107 Complete 6.51 SC Wilf et al. 2003, 2005 LH21-0005 TY107 Complete 6.05 CC Wilf et al. 2003, 2005 LH27(2009)-0370 TY107 Complete 6.37 Direct this study LH02-0135 TY108 Complete 6.90 CC Wilf et al. 2003, 2005 LH02-0135a TY108 Complete 6.86 Direct this study LH15-0018 TY109 Complete 6.98 CC Wilf et al. 2003, 2005 LH15-0031 TY109 Fragment 8.01 SC this study LH15-0043 TY109 Fragment 7.41 VS this study LH17-IOP09 TY109 Complete 7.48 Direct this study LH02-0082 TY111 Largely complete 8.56 Direct this study LH04-1051 TY111 Largely complete 7.90 Direct this study LH13-1599 TY111 Fragment 7.84 VS this study LH04-1247 TY112 Complete 7.90 CC Wilf et al. 2003, 2005 LH06-0022 TY112 Complete 6.86 CC Wilf et al. 2003, 2005 LH06-0041 TY112 Complete 6.89 CC Wilf et al. 2003, 2005 LH06-0140 TY112 Largely complete 8.01 Direct this study LH06-1019 TY112 Complete 7.83 Direct this study LH06-1175 TY112 Fragment 7.85 VS this study LH13-1393 TY112 Complete 7.62 CC Wilf et al. 2003, 2005 LH13-1404 TY112 Fragment 7.95 VS this study LH02-0063 TY113 Complete 7.11 Direct this study LH02-0167 TY113 Complete 7.03 Direct this study LH02-1049 TY113 Complete 7.39 Direct this study LH02-1096 TY113 Complete 7.43 CC Wilf et al. 2003, 2005 LH06-0040 TY113 Complete 7.75 CC Wilf et al. 2003, 2005

82

Specimen field # Morphotype Leaf completeness ln Area Method Source LH06-1157 TY113 Complete 6.66 CC Wilf et al. 2003, 2005 LH06(2006)-06 TY113 Complete 7.08 Direct this study LH27(2009)-0297 TY113 Largely complete 7.94 Direct this study LH23-0032 TY114 Fragment 8.06 VS this study LH01-0012 TY116 Complete 5.26 CC Wilf et al. 2003, 2005 LH01-0016 TY116 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0084 TY116 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0160 TY116 Complete 4.58 CC Wilf et al. 2003, 2005 LH02-0226 TY116 Complete 5.59 CC Wilf et al. 2003, 2005 LH02-0264 TY116 Complete 5.49 CC Wilf et al. 2003, 2005 LH02-0289 TY116 Complete 5.45 Direct this study LH02-1062 TY116 Complete 6.55 CC Wilf et al. 2003, 2005 LH02-1086 TY116 Complete 6.89 Direct this study LH02-1164 TY116 Complete 6.09 CC Wilf et al. 2003, 2005 LH02-1183 TY116 Complete 7.62 CC Wilf et al. 2003, 2005 LH02-1261 TY116 Complete 6.68 CC Wilf et al. 2003, 2005 LH03-0019 TY116 Complete 5.12 CC Wilf et al. 2003, 2005 LH03-0022 TY116 Complete 6.43 CC Wilf et al. 2003, 2005 LH03-0029 TY116 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-1296 TY116 Complete 6.52 CC Wilf et al. 2003, 2005 LH04(2006)-0030 TY116 Complete 6.96 Direct this study LH06-0085 TY116 Fragment 7.66 VS this study LH06-0101 TY116 Complete 8.01 SC Wilf et al. 2003, 2005 LH06-0157 TY116 Complete 7.07 CC Wilf et al. 2003, 2005 LH06-0190 TY116 Complete 6.81 CC Wilf et al. 2003, 2005 LH06-0203 TY116 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-0301 TY116 Complete 7.57 CC Wilf et al. 2003, 2005 LH13-0095 TY116 Complete 7.02 CC Wilf et al. 2003, 2005 LH13-0126 TY116 Complete 7.03 CC Wilf et al. 2003, 2005 LH13-0186 TY116 Complete 6.49 CC Wilf et al. 2003, 2005 LH13-0224 TY116 Complete 9.11 SC Wilf et al. 2003, 2005 LH13-0231 TY116 Complete 8.01 SC Wilf et al. 2003, 2005 LH13-0285 TY116 Complete 6.58 CC Wilf et al. 2003, 2005 LH13-0295 TY116 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0304 TY116 Complete 6.58 CC Wilf et al. 2003, 2005 LH13-1205 TY116 Complete 6.41 CC Wilf et al. 2003, 2005 LH13-1370:1420 TY116 Fragment 8.60 VS this study LH13-1423 TY116 Complete 5.81 Direct this study LH13-1427 TY116 Complete 7.65 CC Wilf et al. 2003, 2005 LH13-1448 TY116 Complete 6.18 Direct this study LH13-1553 TY116 Complete 5.87 CC Wilf et al. 2003, 2005 LH15-0004 TY116 Fragment 6.51 SC this study

83

Specimen field # Morphotype Leaf completeness ln Area Method Source LH15-0035 TY116 Complete 7.58 CC Wilf et al. 2003, 2005 LH15-0056 TY116 Fragment 7.38 VS this study LH15-0077 TY116 Complete 6.87 CC Wilf et al. 2003, 2005 LH15-56' TY116 Complete 7.02 CC Wilf et al. 2003, 2005 LH15(2009)-0520 TY116 Complete 7.45 Direct this study LH16-0020 TY116 Complete 6.81 CC Wilf et al. 2003, 2005 LH16-0029 TY116 Complete 6.51 SC Wilf et al. 2003, 2005 LH16-0053 TY116 Complete 6.65 CC Wilf et al. 2003, 2005 LH17-0059 TY116 Complete 6.51 SC Wilf et al. 2003, 2005 LH17-1003 TY116 Complete 6.62 CC Wilf et al. 2003, 2005 LH17-1003a TY116 Complete 6.86 Direct this study LH22-0019 TY116 Complete 6.48 CC Wilf et al. 2003, 2005 LH22-0109 TY116 Complete 5.85 CC Wilf et al. 2003, 2005 LH23-0057 TY116 Complete 6.51 SC Wilf et al. 2003, 2005 LH24-0008 TY116 Complete 6.51 SC Wilf et al. 2003, 2005 LH24-0010 TY116 Complete 6.19 Direct this study LH25-0006 TY116 Complete 7.89 CC Wilf et al. 2003, 2005 LH01-0005 TY117 Complete 6.19 CC Wilf et al. 2003, 2005 LH01-0010 TY117 Complete 6.09 CC Wilf et al. 2003, 2005 LH01-0026 TY117 Complete 6.75 CC Wilf et al. 2003, 2005 LH02-0033 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0060 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0070 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0076 TY117 Complete 5.48 CC Wilf et al. 2003, 2005 LH02-0080 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0141 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0151 TY117 Fragment 8.41 VS this study LH02-0180 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0204 TY117 Complete 6.63 CC Wilf et al. 2003, 2005 LH02-0210 TY117 Complete 5.58 CC Wilf et al. 2003, 2005 LH02-0220 TY117 Complete 5.77 CC Wilf et al. 2003, 2005 LH02-0253 TY117 Complete 6.33 CC Wilf et al. 2003, 2005 LH02-0256 TY117 Complete 6.19 CC Wilf et al. 2003, 2005 LH02-0283 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0331 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0332 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH02-0500 TY117 Complete 6.91 CC Wilf et al. 2003, 2005 LH02-1124 TY117 Complete 6.76 CC Wilf et al. 2003, 2005 LH02-1127 TY117 Complete 7.55 CC Wilf et al. 2003, 2005 LH02-1182 TY117 Complete 6.04 CC Wilf et al. 2003, 2005 LH02-1231 TY117 Complete 6.64 CC Wilf et al. 2003, 2005 LH02-1235 TY117 Complete 7.45 CC Wilf et al. 2003, 2005

84

Specimen field # Morphotype Leaf completeness ln Area Method Source LH02-1403 TY117 Complete 6.34 CC Wilf et al. 2003, 2005 LH04-0005 TY117 Complete 5.65 CC Wilf et al. 2003, 2005 LH04-0099 TY117 Complete 6.33 CC Wilf et al. 2003, 2005 LH04-0132 TY117 Complete 7.33 CC Wilf et al. 2003, 2005 LH04-0216 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH04-0223 TY117 Complete 4.32 SC Wilf et al. 2003, 2005 LH04-1149 TY117 Complete 6.89 CC Wilf et al. 2003, 2005 LH04-1161 TY117 Complete 6.58 CC Wilf et al. 2003, 2005 LH04-1194 TY117 Complete 6.22 CC Wilf et al. 2003, 2005 LH04-1298 TY117 Complete 5.66 CC Wilf et al. 2003, 2005 LH06-0236 TY117 Complete 7.88 CC Wilf et al. 2003, 2005 LH06-0308 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-1109 TY117 Complete 6.44 CC Wilf et al. 2003, 2005 LH06-1219 TY117 Complete 6.80 Direct this study LH13-0045 TY117 Complete 6.43 CC Wilf et al. 2003, 2005 LH13-0062 TY117 Complete 6.85 Direct this study LH13-0082 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0123 TY117 Complete 6.20 CC Wilf et al. 2003, 2005 LH13-0148 TY117 Complete 6.58 CC Wilf et al. 2003, 2005 LH13-0150 TY117 Complete 6.17 CC Wilf et al. 2003, 2005 LH13-0156 TY117 Complete 6.89 CC Wilf et al. 2003, 2005 LH13-0167 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0174 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0196 TY117 Complete 5.79 CC Wilf et al. 2003, 2005 LH13-0225 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0230 TY117 Complete 6.17 CC Wilf et al. 2003, 2005 LH13-0264 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0265 TY117 Complete 6.63 CC Wilf et al. 2003, 2005 LH13-0276 TY117 Complete 6.10 CC Wilf et al. 2003, 2005 LH13-1040 TY117 Complete 6.93 Direct this study LH13-1085 TY117 Complete 6.20 Direct this study LH13-1167 TY117 Complete 5.83 Direct this study LH13-1173 TY117 Complete 7.63 CC Wilf et al. 2003, 2005 LH13-1213 TY117 Largely complete 6.12 Direct this study LH15-0030 TY117 Complete 6.97 CC Wilf et al. 2003, 2005 LH15-0058 TY117 Complete 7.09 CC Wilf et al. 2003, 2005 LH15-0076 TY117 Complete 5.73 CC Wilf et al. 2003, 2005 LH15-0085 TY117 Complete 5.77 CC Wilf et al. 2003, 2005 LH15-0097 TY117 Complete 6.95 CC Wilf et al. 2003, 2005 LH15-0508 TY117 Complete 5.30 CC Wilf et al. 2003, 2005 LH15-0509 TY117 Complete 5.55 CC Wilf et al. 2003, 2005 LH15-0510 TY117 Complete 5.60 CC Wilf et al. 2003, 2005

85

Specimen field # Morphotype Leaf completeness ln Area Method Source LH15(2009)-0374 TY117 Complete 6.22 Direct this study LH15(2009)-0385 TY117 Complete 6.03 Direct this study LH15(2009)-0389 TY117 Complete 7.39 Direct this study LH16-0051 TY117 Complete 6.58 CC Wilf et al. 2003, 2005 LH16-0501 TY117 Complete 5.68 CC Wilf et al. 2003, 2005 LH17-0052 TY117 Complete 6.05 CC Wilf et al. 2003, 2005 LH17-IOP5 TY117 Complete 6.58 Direct this study LH18-0001 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH18-0002 TY117 Complete 6.51 SC Wilf et al. 2003, 2005 LH18-0003 TY117 Complete 6.76 CC Wilf et al. 2003, 2005 LH18-0005 TY117 Complete 6.73 CC Wilf et al. 2003, 2005 LH18-1-0005 TY117 Complete 6.92 Direct this study LH23-0040 TY117 Complete 6.87 CC Wilf et al. 2003, 2005 LH24-0014 TY117 Complete 6.66 CC Wilf et al. 2003, 2005 LH24-0019 TY117 Complete 7.00 CC Wilf et al. 2003, 2005 LH04-0036 TY118 Complete 6.97 Direct this study LH04-1024 TY118 Complete 8.50 CC Wilf et al. 2003, 2005 LH04-1083 TY118 Complete 9.11 CC Wilf et al. 2003, 2005 LH04-1144 TY118 Complete 7.78 CC Wilf et al. 2003, 2005 LH04-1189 TY118 Complete 6.92 CC Wilf et al. 2003, 2005 LHF-0022 TY119 Fragment 7.81 VS this study LH01-0015 TY122 Complete 6.60 CC Wilf et al. 2003, 2005 LH01-0027 TY122 Complete 6.11 CC Wilf et al. 2003, 2005 LH02-1060 TY122 Complete 6.70 CC Wilf et al. 2003, 2005 LH02-1256 TY122 Complete 5.78 CC Wilf et al. 2003, 2005 LH04-1213 TY122 Complete 6.09 CC Wilf et al. 2003, 2005 LH04-1285 TY122 Complete 6.54 CC Wilf et al. 2003, 2005 LH04-1293 TY122 Complete 7.29 CC Wilf et al. 2003, 2005 LH04-1314 TY122 Complete 7.17 CC Wilf et al. 2003, 2005 LH06-0062 TY122 Complete 6.37 CC Wilf et al. 2003, 2005 LH06-1011 TY122 Complete 7.10 CC Wilf et al. 2003, 2005 LH06-1110 TY122 Complete 6.31 CC Wilf et al. 2003, 2005 LH06-1134 TY122 Complete 6.60 CC Wilf et al. 2003, 2005 LH06-1184 TY122 Complete 6.58 CC Wilf et al. 2003, 2005 LH06-1239 TY122 Complete 6.72 CC Wilf et al. 2003, 2005 LH06-1280 TY122 Complete 6.09 CC Wilf et al. 2003, 2005 LH06-IOP1 TY122 Complete 6.94 Direct this study LH13-0021 TY122 Complete 6.41 CC Wilf et al. 2003, 2005 LH13-0036 TY122 Complete 7.23 CC Wilf et al. 2003, 2005 LH13-0049 TY122 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0060 TY122 Complete 5.80 CC Wilf et al. 2003, 2005 LH13-0075 TY122 Complete 6.86 CC Wilf et al. 2003, 2005

86

Specimen field # Morphotype Leaf completeness ln Area Method Source LH13-0091 TY122 Complete 6.51 SC Wilf et al. 2003, 2005 LH13-0097 TY122 Complete 6.15 CC Wilf et al. 2003, 2005 LH13-0139 TY122 Complete 6.98 CC Wilf et al. 2003, 2005 LH13-0140 TY122 Complete 6.97 CC Wilf et al. 2003, 2005 LH13-1114 TY122 Largely complete 6.45 Direct this study LH13-1150 TY122 Complete 6.97 CC Wilf et al. 2003, 2005 LH13-1159 TY122 Complete 6.85 Direct this study LH13-1182 TY122 Complete 6.41 CC Wilf et al. 2003, 2005 LH13-1209 TY122 Complete 6.52 Direct this study LH13-1246 TY122 Complete 6.39 CC Wilf et al. 2003, 2005 LH13-1252 TY122 Complete 6.40 CC Wilf et al. 2003, 2005 LH13-1261 TY122 Complete 6.57 Direct this study LH13-1305 TY122 Complete 4.38 CC Wilf et al. 2003, 2005 LH13-1340 TY122 Complete 5.68 CC Wilf et al. 2003, 2005 LH13-1377 TY122 Complete 6.85 CC Wilf et al. 2003, 2005 LH13-1392 TY122 Complete 4.34 Direct this study LH13-1417 TY122 Complete 7.10 CC Wilf et al. 2003, 2005 LH13-1493 TY122 Complete 6.44 CC Wilf et al. 2003, 2005 LH13-1495 TY122 Complete 7.17 Direct this study LH13-1508 TY122 Complete 7.68 CC Wilf et al. 2003, 2005 LH13-1509 TY122 Complete 5.41 CC Wilf et al. 2003, 2005 LH13-1520 TY122 Complete 5.26 Direct this study LH15-0009 TY122 Complete 6.51 CC Wilf et al. 2003, 2005 LH15-0028 TY122 Complete 7.22 Direct this study LH15-1002 TY122 Complete 6.98 CC Wilf et al. 2003, 2005 LH15(2009)-0381 TY122 Complete 6.22 Direct this study LH22-1001 TY122 Complete 7.16 Direct this study LH23-0019 TY122 Complete 6.33 CC Wilf et al. 2003, 2005 LH25-0034 TY122 Complete 7.19 CC Wilf et al. 2003, 2005 LH27(2009)-0242 TY122 Complete- poor vein preservation 5.60 Direct this study LH27(2009)-0246 TY122 Complete 8.58 Direct this study LH04-1030 TY123 Fragment 8.85 VS this study LH15-0053 TY123 Fragment 8.75 VS this study LH06-0032 TY124 Complete 7.18 Direct this study LH13-1222 TY124 Complete 7.70 Direct this study LH04-0001 TY125 Complete 7.08 Direct this study LH02-0003 TY126 Fragment 8.27 VS this study LH04-1214 TY126 Largely complete 7.41 Direct this study LH04-0114 TY132 Largely complete 7.78 Direct this study LHF-0041 TY133 Largely complete 8.90 Direct this study LH06-0138 TY134 Fragment 9.20 VS this study LH02-0090 TY135 Fragment 7.64 VS this study

87

Specimen field # Morphotype Leaf completeness ln Area Method Source LH04-0300 TY136 Fragment 6.51 SC this study LH02-0174 TY138 Complete 6.17 CC Wilf et al. 2003, 2005 LH02-0320 TY138 Largely complete 6.40 Direct this study LH06-0003 TY140 Complete 7.20 CC Wilf et al. 2003, 2005 LH06-0255 TY140 Complete 6.51 SC Wilf et al. 2003, 2005 LH06-1025 TY140 Complete 7.14 Direct this study LH27(2009)-0220 TY140 Fragment 8.00 VS this study LH06-1307 TY141 Complete 8.09 CC Wilf et al. 2003, 2005 LH06-1073 TY142 Complete 7.03 Direct this study LH06-1196 TY144 Complete 8.07 Direct this study LH06-1066 TY147 Fragment 8.57 VS this study LH06-1022 TY148 Fragment 9.27 VS this study LH06-1045 TY148 Largely complete 7.76 Direct this study LH06-1052 TY150 Fragment 7.17 VS this study LH06-1042 TY151 Complete 7.24 Direct this study LH06-1215 TY152 Fragment 8.86 VS this study LH06-1180' TY153 Complete 7.23 CC Wilf et al. 2003, 2005 LH04-1077 TY155 Complete 8.17 Direct this study LH13-1400 TY155 Complete 8.64 CC Wilf et al. 2003, 2005 LH04-1000 TY157 Complete 7.42 Direct this study LH04-1199 TY157 Complete 6.44 Direct this study LH06-0023 TY157 Complete 6.32 Direct this study LH06-1020 TY157 Complete 6.70 CC Wilf et al. 2003, 2005 LH13-1260 TY157 Complete 7.59 Direct this study LH13-1571 TY157 Complete 7.13 CC Wilf et al. 2003, 2005 LH04-1180 TY158 Fragment 8.18 VS this study LH04-1309 TY158 Fragment 7.93 VS this study LH02-1097 TY159 Fragment 8.23 VS this study LH04-1078 TY159 Complete 7.32 Direct this study LH04-1311 TY159 Complete 7.51 CC Wilf et al. 2003, 2005 LH04-1118 TY160 Fragment 7.86 VS this study LH04-1025 TY161 Fragment 8.34 VS this study LH04-1098 TY163 Fragment 6.51 SC this study LH04-1075 TY164 Largely complete 7.51 Direct this study LH04-1113 TY166 Fragment 6.51 SC this study LH02-1057 TY167 Fragment 7.44 VS this study LH02-1178 TY167 Complete 6.93 CC Wilf et al. 2003, 2005 LH04-1195 TY167 Complete 6.86 Direct this study LH04-1334 TY170 Fragment 9.34 VS this study LH13-1027 TY171* Fragment 8.01 SC this study LH13-1532 TY171* Fragment 7.96 VS this study LH13-1028 TY175 Complete- poor vein preservation 3.19 Direct this study

88

Specimen field # Morphotype Leaf completeness ln Area Method Source LH13-1559 TY177 Fragment 8.92 VS this study LH04-0133 TY178 Complete 8.61 CC Wilf et al. 2003, 2005 LH13-1561 TY178 Fragment 8.72 VS this study LH13-1193 TY180 Largely complete 7.08 Direct this study LH13-1080 TY181 Largely complete 6.74 Direct this study LH13-1105 TY181 Complete 6.56 Direct this study LH13-1154 TY183 Largely complete 6.50 Direct this study LH13-1022 TY185 Complete 8.78 Direct this study LH13-0134 TY186 Complete 6.86 CC Wilf et al. 2003, 2005 LH13-1216 TY186 Largely complete 7.27 Direct this study LH13-1494 TY186 Complete 7.06 Direct this study LH13-1372 TY187 Complete 6.12 Direct this study LH15(2009)-0392 TY187 Largely complete 7.44 Direct this study LH13-1276 TY188 Largely complete 8.03 Direct this study LH02-1205 TY191 Fragment 6.51 SC this study LH02-1410 TY191 Fragment 6.51 SC this study LH06-1298 TY191 Complete 5.55 CC Wilf et al. 2003, 2005 LH02-1105 TY192 Largely complete 7.26 Direct this study LH02-1330 TY192 Largely complete 7.81 Direct this study LH02-1359 TY192 Complete 6.69 CC Wilf et al. 2003, 2005 LH02-1040 TY193 Fragment 8.70 VS this study LH13-1487 TY193 Complete 7.34 Direct this study LH02-1122 TY194 Fragment 7.86 VS this study LH02-1279 TY195 Fragment 8.98 VS this study LH02-1160 TY196 Complete 5.70 CC Wilf et al. 2003, 2005 LH02-1210 TY197 Complete 6.71 Direct this study LH06-0189 TY197 Complete 7.94 Direct this study LH17-IOP12 TY197 Largely complete 7.01 Direct this study LH02-1154 TY198 Fragment 8.21 VS this study LH04-1302 TY199 Complete 6.95 Direct this study LH06-1256 TY200 Complete 7.41 Direct this study LH13-1194 TY201 Fragment 6.51 SC this study LH27(2009)-0299 TY201 Complete 8.05 Direct this study LH13-1232 TY202 Fragment 9.15 VS this study LH13-0303 TY203 Complete 6.36 Direct this study LH06-1159 TY204 Complete- poor vein presveration 6.44 Direct this study MPEF-144b TY205 Largely complete 8.96 Direct this study LH04-1192:1193 TY206 Complete 6.35 Direct this study LH24-0025 TY207 Fragment 8.02 VS this study LH13-1311 TY208 Complete 6.13 Direct this study LH13-1315 TY208 Complete 6.41 CC Wilf et al. 2003, 2005 LH06-0284 TY209 Complete 7.46 CC Wilf et al. 2003, 2005

89

Specimen field # Morphotype Leaf completeness ln Area Method Source LH06-1250 TY209 Complete 7.52 Direct this study LH06-0079 TY210 Complete 7.14 CC Wilf et al. 2003, 2005 LH06-1083 TY210 Complete 6.58 CC Wilf et al. 2003, 2005 LH15-0060 TY210 Complete 7.30 Direct this study LH22-0009 TY210 Complete 7.29 CC Wilf et al. 2003, 2005 LH06-0004 TY212 Fragment 8.38 VS this study LH02-1358 TY213 Largely complete 5.52 Direct this study LH06-0292 TY213 Complete 7.51 CC Wilf et al. 2003, 2005 LH02-1058 TY215 Complete 6.45 CC Wilf et al. 2003, 2005 LH06-1183 TY215 Fragment 8.77 VS this study LH06-0074 TY216 Fragment 7.85 VS this study LH26-0001 TY217 Fragment 8.60 VS this study LH25(2006)-0001 TY218 Complete- poor vein presveration 8.76 Direct this study LH04(2006)-0048-2 TY219 Complete 7.93 Direct this study LH17-IOP17 TY220 Largely complete 7.79 Direct this study LH04(2006)-0040 TY222 Complete 6.57 Direct this study LH04(2006)-0018 TY223 Complete 6.10 Direct this study LH15(2009)-0541 TY225** Fragment 8.84 VS this study LH15(2009)-0542 TY226** Fragment 8.18 VS this study LH4(2009)-0453 TY227** Complete 7.85 Direct this study LH25(2009)-0353 TY229** Complete 5.94 Direct this study LH27(2009)-0255 TY230** Fragment 6.86 VS this study LH27(2009)-0255-2 TY230** Complete 5.75 Direct this study LH27(2009)-0298 TY231** Fragment 8.18 VS this study Notes: *Monocot species, but included in this analysis because it has a venation pattern like **Newly discovered fossil species

90

Table A3. All Laguna del Hunco dicot fossil species min, max, and mean leaf areas. Minimum, maximum and mean fossil leaf areas (LA) are shown in mm2 although analyses were conducted with natural log transformed units; n is number of specimens measured. Mean leaf areas were calculated as the average of the minimum and maximum leaf size of each species.

Family Morphotype Species name n Min LA Max LA Mean LA ?Fabaceae TY017 4 451.07 758.58 604.82 Sapindaceae TY018 Schmidelia proedulis Berry 1925 49 93.69 2440.60 1267.15 Fagaceae TY019 “Tetracera” patagonica Berry 1925 39 665.14 9045.29 4855.22 ?Ulmaceae TY020 “Celtis” ameghenoi Berry 1925 102 75.19 15024.24 7549.71 Eucalyptus frenguelliana Gandolfo and Myrtaceae TY021 102 37.34 3010.92 1524.13 Zamaloa Myrtaceae TY022 5 780.55 2697.28 1738.92 Malvaceae TY023 aff. Brachychiton 29 432.68 9798.65 5115.67 ?Anacardaceae TY024 3 1395.29 9045.29 5220.29 Rhamnaceae TY025 “Rhamnidium” sp. 26 75.19 3197.10 1636.15 Proteaceae TY026 Lomatia preferruginea Berry 1938 38 259.82 3010.92 1635.37 Monimiaceae TY027 1 724.12 724.12 724.12 Unknown TY028 3 1660.50 4512.15 3086.32 Atherospermataceae TY029 Atherospermophyllum guinazui C.L knight 7 880.07 5017.83 2948.95 Bixaceae TY030 16 671.83 12798.11 6734.97 Unknown TY032 8 3010.92 9738.07 6374.49 cf. Hovenia (“Banara” prehernandiensis Rhamnaceae TY035 15 671.83 3010.92 1841.37 1938) Unknown TY036 2 1352.89 1582.31 1467.60 ?Leguminosae TY037 Leguminosites patagonicus Berry 1925 2 671.83 5126.95 2899.39 aff. Eucryphia (Schmidelia graciliforma ?Eucryphiaceae TY038 8 639.06 3714.50 2176.78 Berry 1925) ?Leguminosae TY039 2 3018.69 9045.29 6031.99 ?Malvaceae TY040 1 18443.49 18443.49 18443.49 Myrtaceae TY041 “Myrcia” deltoidea Engelhardt 17 361.41 2143.08 1252.24 Salicaceae TY042 “Banara” prehernandiensis Berry 1925 21 327.01 4166.58 2246.80 Escalloniaceae TY043 cf. Escallonia 8 206.44 3010.92 1608.68 Lomatia occidentalis (Berry) Frenguelli Proteaceae TY044 14 671.83 9045.29 4858.56 1943 Malvaceae TY045 aff. Firmiana 5 5825.50 11809.19 8817.34 Euphorbiaceae TY046 4 3604.72 19462.31 11533.52 ?Menispermaceae TY047 1 4209.44 4209.44 4209.44 Unknown TY048 1 2836.99 2836.99 2836.99 Unknown TY050 3 5597.08 8479.83 7038.46 Unknown TY053 2 2153.63 2465.13 2309.38 ?Myrtaceae TY054 8 139.77 3083.75 1611.76 Unknown TY055 2 1211.97 2786.36 1999.16 Unknown TY056 24 626.41 9045.29 4835.85 Unknown TY057 6 1248.88 5825.50 3537.19 Unknown TY058 1 489.84 489.84 489.84

91

Family Morphotype Species name n Min LA Max LA Mean LA Unknown TY059 2 1176.15 2125.79 1650.97 ?Myrtaceae TY060 “Myrcia” inequilateris Berry 1938 2 245.60 1053.69 649.64 Unknown TY061 “Myrica” mira Berry 1938 60 75.00 2807.36 1441.18 Urticaceae TY062 cf. Urtica 2 4582.50 5193.73 4888.12 Unknown TY064 1 417.69 417.69 417.69 Unknown TY066 4 528.97 3904.95 2216.96 Unknown TY067 1 2404.86 2404.86 2404.86 Unknown TY068 1 1598.44 1598.44 1598.44 Unknown TY069 1 962.95 962.95 962.95 Unknown TY070 7 1772.24 9045.29 5408.77 Unknown TY072 “Sterculia” patagonica Berry 1925 2 675.00 1939.14 1307.07 Unknown TY073 1 1619.71 1619.71 1619.71 ?Leguminosae TY074 2 9045.29 12612.19 10828.7 Myrtaceae TY075 4 75.00 1465.57 770.29 Unknown TY076 4 1685.81 6807.87 4246.84 Unknown TY077 2 601.85 617.60 609.72 Unknown TY079 2 1109.52 3520.49 2315.01 Unknown TY080 1 522.56 522.56 522.56 Unknown TY081 4 765.09 7215.67 3990.38 Unknown TY082 2 1465.57 1822.20 1643.88 Unknown TY083 2 8518.54 9838.14 9178.34 Lauraceae TY084 66 135.64 5750.23 2942.93 Akania patagonica Gandolfo, Dibbern and Akaniaceae TY101 6 589.93 4897.87 2743.90 Romero Monimiaceae TY102 Monimiophyllum callidentatum C.L. Knight 1 760.84 760.84 760.84 Unknown TY103 1 454.86 454.86 454.86 Unknown TY104 2 730.16 859.45 794.80 “Cupania” grosse-serrata (Engelhardt) Sapindaceae TY105 22 419.66 9045.29 4732.48 Berry 1925 unknown TY106 17 1064.22 54675.01 27869.62 Unknown TY107 “Coprosma” incerta Berry 1938 40 200.34 3362.54 1781.44 Unknown TY108 2 956.22 992.27 974.25 Sapindales (order) TY109 4 1074.92 3018.69 2046.81 Unknown TY111 3 2548.34 5201.53 3874.94 Unknown TY112 9 953.37 3022.27 1987.82 Unknown TY113 8 780.55 2796.46 1788.51 Unknown TY114 1 3178.85 3178.85 3178.85 Cunoniaceae TY116 Caldcluvia sp. 55 97.51 9045.29 4571.40 Fabaceae TY117 “Cassia” argentinensis Berry 1938 82 75.19 4483.24 2279.22 Menispermaceae TY118 5 1012.32 9045.29 5028.81 Unknown TY119 1 2457.10 2457.10 2457.10 Lauraceae TY122 52 76.44 5318.32 2697.38 Unknown TY123 2 6338.65 6956.67 6647.66

92

Family Morphotype Species name n Min LA Max LA Mean LA Unknown TY124 3 1308.44 2218.69 1763.56 Unknown TY125 1 1186.37 1186.37 1186.37 Unknown TY126 3 1654.99 3918.60 2786.80 Unknown TY132 1 2399.99 2399.99 2399.99 Unknown TY133 1 7314.78 7314.78 7314.78 Unknown TY134 2 9045.29 9906.32 9475.81 Unknown TY135 1 2086.21 2086.21 2086.21 Unknown TY136 1 675.00 675.00 675.00 ?Melastomataceae TY138 2 478.19 603.39 540.79 ?Monimiaceae TY140 4 671.83 2966.95 1819.39 ?Araliaceae TY141 1 3261.69 3261.69 3261.69 Unknown TY142 1 1126.48 1126.48 1126.48 unknown TY144 1 3194.80 3194.80 3194.80 Unknown TY147 1 5288.21 5288.21 5288.21 ?Menispermaceae TY148 2 2347.75 10633.63 6490.69 Unknown TY150 1 1293.42 1293.42 1293.42 Unknown TY151 1 1387.72 1387.72 1387.72 Unknown TY152 1 7020.05 7020.05 7020.05 Bixaceae TY153 1 1380.22 1380.22 1380.22 Araliaceae TY155 3 2703.30 5653.33 4178.31 Myrtaceae TY157 6 555.04 1970.70 1262.87 Unknown TY158 3 2784.61 3789.54 3287.07 Unknown TY159 3 1505.93 3745.33 2625.63 Malvaceae TY160 1 2602.82 2602.82 2602.82 Unknown TY161 1 4189.25 4189.25 4189.25 Unknown TY163 1 675.00 675.00 675.00 Unknown TY164 2 1824.27 1919.85 1872.06 ?Myrtaceae TY166 1 675.00 675.00 675.00 Unknown TY167 ? “Ouratea” firmifolia 3 952.78 1702.23 1327.50 Unknown TY170 . 1 11414.46 11414.46 11414.46 Ripogonaceae TY171* Ripogonum new sp. 3 2344.90 3018.69 2681.80 Fabaceae TY175 2 24.05 24.33 24.19 ?Winteraceae TY177 1 7501.80 7501.80 7501.80 Unknown TY178 3 4491.76 6112.58 5302.17 Unknown TY180 2 1074.92 1189.69 1132.31 Unknown TY181 2 709.44 848.77 779.10 Unknown TY183 1 663.36 663.36 663.36 Unknown TY185 1 6473.63 6473.63 6473.63 Unknown TY186 3 953.37 1438.61 1195.99 Unknown TY187 2 452.71 1703.07 1077.89 Unknown TY188 1 3077.34 3077.34 3077.34 Unknown TY191 4 257.24 962.95 610.09

93

Family Morphotype Species name n Min LA Max LA Mean LA Unknown TY192 3 804.32 2456.32 1630.32 Menispermaceae TY193 3 1533.25 6017.00 3775.12 Malvaceae TY194 2 2597.98 5218.68 3908.33 Unknown TY195 1 7936.48 7936.48 7936.48 Unknown TY196 1 298.87 298.87 298.87 Unknown TY197 3 818.86 2804.78 1811.82 Unknown TY198 1 3684.09 3684.09 3684.09 Unknown TY199 1 1046.40 1046.40 1046.40 Unknown TY200 1 1659.60 1659.60 1659.60 Unknown TY201 2 675.00 3127.70 1901.35 Unknown TY202 1 9461.37 9461.37 9461.37 Anacardiaceae TY203 1 577.88 577.88 577.88 Unknown TY204 1 623.75 623.75 623.75 Menispermaceae TY205 1 7780.60 7780.60 7780.60 ?Burseraceae TY206 1 572.22 572.22 572.22 Unknown TY207 1 3032.19 3032.19 3032.19 Proteaceae TY208 new Proteaceae sp. González et al. 2007 2 460.64 607.89 534.27 Unknown TY209 2 1737.15 1843.25 1790.20 Unknown TY210 4 720.54 1479.34 1099.94 Unknown TY212 1 4368.59 4368.59 4368.59 ?Lauraceae TY213 2 248.89 1826.21 1037.55 Unknown TY215 2 632.70 6409.45 3521.07 Unknown TY216 1 2576.79 2576.79 2576.79 Unknown TY217 1 5448.22 5448.22 5448.22 Myrtaceae TY218 1 6384.09 6384.09 6384.09 Unknown TY219 1 2782.97 2782.97 2782.97 Unknown TY220 1 2417.44 2417.44 2417.44 Unknown TY222 1 710.51 710.51 710.51 Unknown TY223 1 447.55 447.55 447.55 Unknown TY225** 1 6923.07 6923.07 6923.07 Unknown TY226** 1 3576.32 3576.32 3576.32 Unknown TY227** 1 2572.07 2572.07 2572.07 Unknown TY229** 1 380.32 380.32 380.32 Unknown TY230** 2 312.79 950.94 631.86 Unknown TY231** 1 3552.14 3552.14 3552.14 *Monocot species, but included in this analysis because it has a venation pattern like dicot leaves (RJ Carpenter et al., in review). **Newly discovered fossil species

94

APPENDIX B. ADDITIONAL FOSSIL-MODERN COMPARISON RESULTS

CONTENTS Figure B1. Evaluating NMS dimensionality ...... 96 Table B1. Proportion of variance represented by NMS ordination axes ...... 96 Table B2. Family compositional similarity in the ten Australian plots most similar to LH……. 97 Table B3. Laguna del Hunco family occurrences in the ten most compositionally similar Australian plots ...... 98 Table B4. Ten Australian plots with most similar family compositions to LH…………...... 99 Table B5. Living Australian rainforest plots with most similar grand mean leaf areas to LH....100

95

Figure B1. Evaluating NMS dimensionality. A. Stress plot of showing stress decreasing to acceptable levels at three ordination dimensions. B. Shepard diagram of the three dimensional NMS solution. A monotonic fit line shows ordination distances relative to distances in the original data matrix. The non-metric fit R2 statistic is calculated as 1 – stress2, and linear fit R2 is the squared correlation between ordination distances and the monotonic fitted values.

Table B1. Proportion of data variance represented by NMS ordination axes. Values record the R2 correlation between Euclidean distances in ordination space and Euclidean distances in the original data matrix of Australian rainforest plot plant family composition.

Incremental variance explained Cumulative variance explained Dimension 1 0.457 0.457 Dimension 2 0.127 0.584 Dimension 3 0.123 0.707

96

Table B2. Family compositional similarity in the ten most compositionally similar Australian plots The shared number plant family presences and absences are recorded between LH and the 10 nearest Australian plots in NMS ordination; overall percent shared is out of the total of 95 plant families analyzed.

Shared family Shared family Overall % Plot ID presences absences shared N86 9 67 80.0 P73 10 67 78.9 W309 10 65 77.9 D254 13 66 76.8 W269 13 66 76.8 P8 6 66 75.8 D261 9 64 74.7 N42 11 66 74.7 N74 10 62 73.7 D259 8 64 73.7

97

Table B3. Laguna del Hunco family occurrences in the ten Australian plots most similar to LH. Plant families are the 22 families known from LH that are extant in Australia; n is the number of occurrences of each family. *indicates rare families, found in less than 15% of the total samples analyzed across Australia.

Families n Percentage of plots Lauraceae 10 100 Myrtaceae 10 100 Sapindaceae 9 90 Casuarinaceae 7* 70 Cunoniaceae 6 60 Fabaceae 6 60 Monimiaceae 6 60 Proteaceae 6 60 Araliaceae 4 40 Rhamnaceae 3 30 Anacardiaceae 1 10 Araucariaceae 1* 10 Atherospermataceae 1 10 Malvaceae 1 10 Podocarpaceae 1* 10 Bixaceae 0 0 Cupressaceae 0* 0 Escalloniaceae 0* 0 Euphorbiaceae 0 0 Salicaceae 0 0 Urticaceae 0 0 Akaniaceae 0 0

98

Table B4. Living Australian rainforest plots with most similar family compositions to LH LA Elevation MAT MAP Plot Type Dist-ance NLRs n species Bed-rock Region (ln mm2) (m) (ºC) (mm) N86 CMVF* 0.14 2 24 8.80 716 15 1307 s NB P73 SNVF 0.25 2 49 8.94 1050 20 1912 g WT W309 CMVF* 0.25 3 44 8.59 791 15 1159 g W D254 CMVF* 0.29 3 34 8.68 247 16 1397 m D W269 CMVF* 0.29 4 54 8.56 815 15 1182 m W P8 CMVF 0.29 1 19 9.04 40 22 2582 a WT D261 CMVF* 0.30 3 31 8.40 867 15 1509 m D N42 CMVF* 0.32 4 43 8.61 859 14 1222 b NB N74 CMVF* 0.32 3 42 8.60 407 17 1870 b NB D259 CMVF* 0.33 1 23 8.24 805 17 1333 b D Mean 2.6 36.8 8.65 659.7 16.6 1547.3 1159- Range 1-4 19-54 8.23-9.04 40-1050 14-22 2582

Notes: Family compositional similarity was calculated as the three dimensional Euclidian distance between LH and Australian rainforest plots in nonmetric multidimensional scaling ordination space. The mean annual temperature (MAT), mean annual precipitation (MAP), and geographic region are shown for each Australia plot; WT is is the Wet Tropics, NB is the Nightcap-Border Ranges, D is Dorrigo, and W is Washpool. Climate, location, and genera data are calculated from Kooyman et al. (2012). LA is grand mean leaf area. Bedrock values are m= metamorphic, r=rhyolite, b= basalt, and g= granite, s= sandstone. NP is national park, and SF is state forest. See Methods for forest types sources; *Forest type inferred from canopy leaf size index

99

Table B5. Living Australian rainforest plots with most similar grand mean leaf areas to LH LA Dist- n Elevation MAT MAP Bed- Plot Type NLRs (ln Region ance species (m) (ºC) (mm) rock mm2) N77 SNVF* 1.195 1 24 7.62 247 18 1965 r NB D180 SNVF/CNVF* 1.369 3 19 7.65 647 15 1900 m D N133 SNVF* 1.058 1 24 7.65 453 17 1168 s NB D210 SNVF/CNVF* 1.237 2 19 7.67 28 18 1684 m D D233 SNVF/CNVF* 1.013 2 26 7.69 433 16 1602 m D N76 SNVF/CNVF* 1.202 2 25 7.73 226 18 1965 r NB W289 SNVF/CNVF* 0.947 3 25 7.75 377 16 1159 m W D190 SNVF/CNVF* 1.148 2 24 7.75 115 18 1654 m D D251 SNVF/CNVF* 1.104 1 27 7.78 412 15 1569 m D D205 SNVF/CNVF* 1.478 4 28 7.79 174 16 1889 m D Mean 2.1 24.1 7.707 311.2 16.7 1655.5 7.62- 1159- Range 1-4 19-28 28-647 15-18 7.79 1956

Notes: Family compositional similarity was calculated as the three dimensional Euclidian distance between LH and Australian rainforest plots in nonmetric multidimensional scaling ordination space. The mean annual temperature (MAT), mean annual precipitation (MAP), and geographic region are shown for each Australia plot; WT is the Wet Tropics, NB is the Nightcap-Border Ranges, D is Dorrigo, and W is Washpool. Climate, location, and genera data are calculated from Kooyman et al. (2012). LA is grand mean leaf area. Bedrock values are m= metamorphic, r=rhyolite, b= basalt, and g= granite, s= sandstone. NP is national park, and SF is state forest. See Methods for forest type sources; *Forest type inferred from canopy leaf size index

100