Cretaceous angiosperm leaves Fossil evidence for low gas ...Fossil evidence for low gas exchange capacities for Early Cretaceous angiosperm leaves Taylor S. Feild, Garland R. Upchurch

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofitpublishers, academic institutions, research libraries, and research funders in the common goal of maximizing access tocritical research.

Fossil evidence for low gas exchange capacities for EarlyCretaceous angiosperm leavesAuthor(s): Taylor S. Feild, Garland R. Upchurch Jr., David S. Chatelet, TimothyJ. Brodribb, Kunsiri C. Grubbs, Marie-Stéphanie Samain, and Stefan WankeSource: Paleobiology, 37(2):195-213. 2011.Published By: The Paleontological SocietyDOI: 10.1666/10015.1URL: http://www.bioone.org/doi/full/10.1666/10015.1

BioOne (www.bioone.org) is an electronic aggregator of bioscience research content,and the online home to over 160 journals and books published by not-for-profit societies,associations, museums, institutions, and presses.

Your use of this PDF, the BioOne Web site, and all posted and associatedcontent indicates your acceptance of BioOne’s Terms of Use, available atwww.bioone.org/page/terms_of_use.

Usage of BioOne content is strictly limited to personal, educational, and non-commercialuse. Commercial inquiries or rights and permissions requests should be directed to theindividual publisher as copyright holder.

http://www.bioone.org/doi/full/10.1666/10015.1

http://www.bioone.org

http://www.bioone.org/page/terms_of_use

Fossil evidence for low gas exchange capacities for EarlyCretaceous angiosperm leaves

Taylor S. Feild, Garland R. Upchurch Jr., David S. Chatelet, Timothy J. Brodribb,Kunsiri C. Grubbs, Marie-Stephanie Samain, and Stefan Wanke

Abstract.—The photosynthetic gas exchange capacities of early angiosperms remain enigmatic.Nevertheless, many hypotheses about the causes of early angiosperm success and how angiospermsinfluenced Mesozoic ecosystem function hinge on understanding the maximum capacity for earlyangiosperm metabolism. We applied structure-functional analyses of leaf veins and stomatal poregeometry to determine the hydraulic and diffusive gas exchange capacities of Early Cretaceous fossilleaves. All of the late Aptian–early Albian angiosperms measured possessed low vein density and lowmaximal stomatal pore area, indicating low leaf gas exchange capacities in comparison to modernecologically dominant angiosperms. Gas exchange capacities for Early Cretaceous angiosperms wereequivalent or lower than ferns and gymnosperms. Fossil leaf taxa from Aptian to Paleocene sedimentspreviously identified as putative stem-lineages to Austrobaileyales and Chloranthales had the samegas exchange capacities and possibly leaf water relations of their living relatives. Our results providefossil evidence for the hypothesis that high leaf gas exchange capacity is a derived feature of laterangiosperm evolution. In addition, the leaf gas exchange functions of austrobaileyoid andchloranthoid fossils support the hypothesis that comparative research on the biology of living basalangiosperm lineages reveals genuine signals of Early Cretaceous angiosperm ecophysiology.

Taylor S. Feild.* School of Biological Sciences, Monash University, Clayton Campus, Australia. E-mail:[email protected]

David S. Chatelet. Department of Ecology and Evolutionary Biology, University of Tennessee, Knoxville,Tennessee 37996

Timothy J. Brodribb. Department of Plant Science, University of Tasmania, Hobart, Tasmania, AustraliaGarland R. Upchurch. Department of Biology, Texas State University, San Marcos, TexasKunsiri C. Grubbs. Department of Biology, Winthrop University, Rock Hill, South CarolinaMarie-Stephanie Samain. Ghent University, Department of Biology, Research Group Spermatophytes, B-9000

Ghent, BelgiumStefan Wanke. Technische Universitat Dresden, Institut fur Botanik, 01062 Dresden, Germany*Corresponding author

Accepted: 29 September 2010

Introduction

Evolutionary transitions in the ways orga-nisms process energy and resources representpivotal turning points in the history of life(Vermeij 1999). In particular, evolutionarychanges in plant metabolic function cascadeacross a network of planet-wide biogeochem-ical and hydrological processes. Of the majortransitions in plant gas exchange function,diverse lines of evidence point to the emer-gence of highly photosynthetically activeangiosperm leaves as a transformative shiftunderpinning the assembly of modern eco-system processes (Boyce et al. 2009; Brodribband Feild 2010). For example, maximal leafgas exchange capacities that define today’secologically dominant angiosperms far out-strip those of all known non-angiosperms

(Korner 1995; Lusk et al. 2003; Brodribb et al.2007; Boyce et al. 2009; Brodribb and Feild2010). The evolution of high leaf gas exchangecapacity may have stabilized or increasedgross primary productivity of the vegetationunder conditions of falling CO2 during theCretaceous and/or later Cenozoic (Volk 1989;Robinson 1994; McElwain et al. 2005; Bondand Scott 2010). In addition, the functionalprerequisites for constructing leaves capableof high rates of photosynthesis and functionalby-products of such leaves as they transpireand eventually decompose may have elicitedenhanced positive feedbacks on vegetation–hydrological cycle interactions, fire frequen-cy, nutrient cycles, and weathering andbiomineralization processes that changedclimates and/or formed new selective inter-

Paleobiology, 37(2), 2011, pp. 195–213

’ 2011 The Paleontological Society. All rights reserved. 0094-8373/11/3702–0002/$1.00

faces for diverse organisms during the Creta-ceous (Volk 1989; Robinson 1994; Martin 1995;Boyce et al. 2009; Taylor et al. 2009; Bond andScott 2010; Brodribb and Feild 2010). Howev-er, an unresolved issue is when high leaf gasexchange of angiosperm leaves first evolved,and how this functional transition influencedearly angiosperm evolution (Wing and Bou-cher 1998; Feild et al. 2009; Brodribb and Feild2010).

An influential hypothesis is that the earliestangiosperms functioned with high photosyn-thetic gas exchange capacities (Stebbins 1974;Doyle and Hickey 1976; Hickey and Doyle1977; Retallack and Dilcher 1981, 1986; Taylorand Hickey 1996). Paleoecological interpreta-tions of the ecomorphology and depositionalenvironments of Early Cretaceous fossil an-giosperm leaves suggested that high photo-synthetic capacity and high rates of weedygrowth favored initial angiosperm success insun-exposed point-bar zones along fast mov-ing rivers (Stebbins 1974; Doyle and Hickey1976; Hickey and Doyle 1977; Retallack andDilcher 1981, 1986; Taylor and Hickey 1996).Thus, the evolution of high leaf gas exchangecapacity by angiosperms did not spark theglobal ecological sweep of the angiosperms.Such is the case because the earliest angio-sperms, while capable of high productivity,remained rare and ecologically confined toearly successional zones for nearly 20 Myrafter their first fossil appearance (Hickey andDoyle 1977; Wing and Boucher 1998; Heim-hofer et al. 2005). Later evolution of otherfunctions in the reproductive system and/ormutualisms with animals gave considerablelift to angiosperm ecological success andallowed the export of high leaf gas exchangecapacity to diverse environments.

By contrast, comparative ecophysiologicalevidence from extant early diverging angio-sperm lineages motivated a hypothesis thatthe first angiosperms functioned with lowmetabolic capacity (Feild et al. 2004, 2009).Low leaf gas exchange capacity was associat-ed with the early diversification of angio-sperms in low-evaporative-demand, shadyhabitats underneath forest canopies formedby gymnosperms and ferns. Later, angio-sperms evolved greater capacities for photo-

synthesis and transpiration, as they dominat-ed forest canopies and open disturbed zones.Under this hypothesis, the rise in ecologicalabundance of angiosperms evolved synchro-nously with their ability to expand into arange of high photosynthetic capacities pre-viously unexplored by other vascular plants(Brodribb and Feild 2010). However, there isno known fossil evidence supporting thehypothesis that the earliest angiospermsfunctioned with low leaf gas exchange poten-tial. Indeed, any literal reading of earlyangiosperm leaf gas exchange evolutionbased on extant basal angiosperm taxa isfraught with uncertainty. Uncertainty existsbecause these lineages experienced consider-able range contraction over the last 100 Myr,they may have become ecologically modifiedby later angiosperm evolution or shifts inglobal environment, and they may fail tosample extinct early angiosperm functionaldiversity (Feild et al. 2009; Royer et al. 2010).

Recent discoveries on how leaf vein andstomatal pore anatomy determine leaf gasexchange capacity offer new potential fortesting hypotheses on early angiosperm leafgas exchange function in the fossil record(Brodribb et al. 2007; Boyce et al. 2009;Brodribb and Feild 2010; McKown et al.2010). These new approaches are potentiallyinformative because previous inferences werebased on traits, including leaf size, leaf shape,major vein architecture patterns, and estimat-ed leaf mass per area of fossil leaves, whichcannot accurately specify where a species fallsalong a spectrum of low to high leaf gasexchange capacity nor specify the hydrauliccosts of leaf photosynthesis (Doyle andHickey 1976; Hickey and Doyle 1977; Retal-lack and Dilcher 1981, 1986; Taylor andHickey 1996; Ackerly and Donoghue 1998;Wing and Boucher 1998; Royer et al. 2010).

The first goal of our investigation was toexplore how venation and stomatal porestructure are linked to leaf gas exchangecapacity and to use these mechanistic linkag-es to test previous hypotheses on how earlyangiosperm fossil leaves functioned. Thefocus is on fossil Zone I leaves from thePotomac Group of North America becausethese fossils represent one of the oldest

196 TAYLOR S. FEILD ET AL.

known records of early angiosperm leaves,and these fossils formed the conceptualcornerstone of the ancestral weed hypothesis(Doyle and Hickey 1976; Hickey and Doyle1977; Taylor and Hickey 1996). The secondgoal was to determine whether gas exchangecapacities of fossil leaves related to extantterrestrial basal angiosperm lineages, specifi-cally Austrobaileyales and Chloranthales, felloutside the range of extant relatives. Thepurpose of these comparisons was to evaluatethe hypothesis that similar leaf gas exchangecapacities have been conserved in extant basalangiosperm leaves since the Cretaceous. Todo so, we determined the relations amongvein and stomatal pore anatomy with leaf gasexchange capacity across a broad sample ofextant basal angiosperm leaves and usedthem to interpret fossil leaf function.

Methods

Abbreviations.—Dv, vein density (mmmm22); gc

STOMA, maximum stomatal conduc-tance to water vapor calculated from stomatalpore geometry (mmol H2O m22 s21); gc

VEIN,maximal stomatal conductance to water va-por calculated from vein density (mmol H2Om22 s21); gm, measured maximum stomatalconductance to water vapor (mmol H2Om22 s21); PC, measured maximum photosyn-thetic capacity on leaf area basis (mmol CO2

m22 s21); Y, water potential (MPa); Yleaf, leafwater potential (MPa); Ysoil, soil water poten-tial (MPa); SD, stomatal density (numbermm22); SL, stomatal guard cell length (mm);SW stomatal guard cell width (mm); PoreL,stomatal pore length (mm); PoreD, stomatalpore depth (mm); SPA, stomatal pore area atmaximal aperture (m2); VPD, vapor pressuredeficit (kPa).

Extant Species and Fossil Leaf Collections.—The comparative investigations of leaf struc-ture–function focused on 87 species of extantbasal angiosperms, including Amborella, Nym-phaeales, Austrobaileyales, and Chloranthales(Appendix I in the supplementary materialonline at http://dx.doi.org/10.1666/10015.s1). Most species studied were in naturalpopulations in Australia, Costa Rica, China,Dominican Republic, French Polynesia, Fiji,Jamaica, New Caledonia, New Zealand, Peru,

Papua New Guinea, Thailand, United States,and Vietnam. Ten individuals per species weresampled. In addition, we studied 18 species,represented by three to ten individuals, in anoutdoor garden collection in Knoxville,Tennessee. Plants were watered using atimed drip irrigation system, and all plantsreceived fertilization every two to three weeksto ensure healthy leaves for physiologicalmeasurements. The species sampled encom-passed the modern diversity of life form, lifezone, and regeneration ecology found in extantterrestrial basal angiosperms (Appendix I,online) (Feild et al. 2004, 2009). Diverse linesof phylogenetic and paleobotanical datasupport these lineages as having divergednear the base of extant angiosperm phylogeny(Jansen et al. 2007; Moore et al. 2007; Saarela etal. 2007; Feild et al. 2009; Endress and Doyle2009).

To examine intraspecific variation of leafstructure in relation to light, we sampled sunand shade leaves of each species whenpossible. Sun and shade environments weredesignated by field observations (Keeling andPhillips 2007). Sun leaves were taken as thosethat fully expanded under greater than 70%exposure to open sky. Shade leaves weretaken as those that fully expanded in theforest understory, defined as less than 5%exposure to open sky. For shade-demandingand short-lived pioneer species, only shadeand sun leaves were available for sampling,respectively.

Details on the localities, ages, and proposedsystematics of the sampled fossils are sum-marized in Appendix II (online). The fossilsrepresented seven localities spanning approx-imately 60 Myr (early Aptian to earliestPaleocene). Two groups of fossils werefocused on (1) leaf fossils from Zone I of thePotomac Group (Doyle and Hickey 1976;Hickey and Doyle 1977) and (2) leaf fossilsthat previous systematic investigations iden-tified as possible stem-lineage relatives toextant Austrobaileyales and Chloranthales,referred hereafter as austrobaileyoids andchloranthoids, respectively (Appendix II, on-line). Appendix III (online) presents evidencefor stratigraphic ages and systematic place-ments for the fossils measured.

EARLY ANGIOSPERM LEAF GAS EXCHANGE 197

Maximum Photosynthetic Rates and StomatalConductances.—To test how leaf venation andstomatal pore structure related to photosyn-thetic gas exchange, water vapor and CO2

exchange fluxes were measured with aphotosynthesis infrared gas analyzer (LiCOR6400XT, Li-COR Biosciences, Lincoln, NB,United States of America). We chose 30 basalangiosperm species to obtain a broad sam-pling of phylogenetic and ecological diversity(Appendix I, online) (Feild et al. 2004, 2009).

Stomatal water vapor conductance (gm,mmol H2O m22 s21) and photosyntheticcapacity (PC, mmol CO2 m22 s21) were mea-sured on clear mornings (0900–1130 h) to: (1)avoid heterogeneities on leaf gas exchangedue to passing clouds, (2) ensure leaves weredry, and (3) ensure that maximal stomatalopening and light-induction of photosynthe-sis occurred before midday stomatal closure.Microclimate around leaves during measure-ments was controlled at 25 6 1.5uC, 1300 mmolquanta m22 s21 photosynthetic photon fluxdensity (PPFD), vapor pressure deficit (VPD)between 0.9 and 1.1 kPa, and 380 6 5 mL L21

CO2. The light intensities provided saturatedphotosynthesis but did not induce photoinhi-bition (Feild et al. 2004). Measurements underthese optimal conditions served as the pho-tosynthetic and stomatal conductance maxi-ma attainable for a species. Photosyntheticgas exchange capacity for each species wasbased on a sample of five undamaged andfully expanded leaves from each of fiveindividuals.

Vein Density (Dv) of Extant and Fossil Leavesto Calculate Leaf Gas Exchange Capacity.—Veindensity is the length of veins ramifying in agiven amount of leaf area (mm mm22). Datafrom living plants demonstrated that photo-synthetic capacity and stomatal conductanceto water vapor can be predicted from Dv

(Brodribb et al. 2007; Boyce et al. 2009;Brodribb and Feild 2010; McKown et al.2010). Dv is a major determinant of leafCO2/H2O exchange because water transportand photosynthetic gas exchange are coupled(Sperry 2003). Leaf hydraulic capacity is setby vein structure because increased veinbranching brings xylem tissues that arespecialized for water transport closer to the

sites of water evaporation in the leaf (Brod-ribb et al. 2007; McKown et al. 2010). Hence,Dv defines the hydraulic supply limit of watervapor exchange that secondarily dictatesmaximum CO2 assimilation by leaves (Brod-ribb et al. 2007; Brodribb and Feild 2010).

Dv of extant taxa was measured on leavescleared using sodium hydroxide, sodiumhypochlorite, and heat (Hudson et al. 2010).Veins were stained in safranin and measuredusing an upright microscope over approxi-mately 6 mm2 of area (Axio-Imager, Carl-Zeiss, Germany). Digital images were cap-tured with an AxioCam camera (Carl-Zeiss,Germany) and processed using ImageJ(http://rsb.info.nih.gov/ij/; NIH, Bethesda,Maryland) to measure Dv. Dv was measuredon compression fossils by tracing vein lengthsusing ImageJ. Images were obtained with adigital camera and macro lens (Nikon D300Swith Nikkor 60 mm lens, Nikon, Japan) or bydigitizing 10.1 3 12.7 cm photographicnegatives and prints of fossils at 600 dpi.Three Dv measurements on each sampledfossil for each species were taken. Only fossilswith well-preserved venation were sampled(Fig. 1). Details on how fossils were selectedfor measurements are provided in AppendixIII (online).

We used a published model to reconstructphotosynthetic capacity (PC) and maximumstomatal conductance to water vapor of fossilleaves from Dv (see Brodribb et al. 2007;Brodribb and Feild 2010). The model assumesthat under non-limiting conditions of soilwater availability, maximum leaf hydraulicconductance (Kleaf) can be calculated from thedistance water must flow from the veinterminals to the sites of evaporation (dm;Brodribb et al. 2007). By knowing Kleaf onecan calculate maximum stomatal conductanceto water vapor from vein density (gc

VEIN),which provides a basis for calculating maxi-mum leaf PC. To emphasize the impact ofvein evolution on gas exchange, we usedfixed concentrations of CO2 and O2 (currentambient concentrations) to reconstruct photo-synthetic capacity. Our emphasis was ondiffusive-hydrodynamic constraints to leafgas exchange rather than biochemical con-straints upon photosynthesis such as maxi-


mum electron transport and carboxylationrates. These enzymatic processes are notpreserved in fossils. Instead of attempting ahighly uncertain reconstruction leaf gas ex-change capacities across an uncertain range ofcross-varying Cretaceous ambient CO2 andO2 concentrations (McElwain et al. 2005;Fletcher et al. 2008; Barclay et al. 2010;Glasspool and Scott 2010; Royer 2010), atmo-spheric gas concentrations were constrainedto allow a comparison of gas exchangecapacities under standard conditions. Param-eters for VPD and water potential gradientacross the leaf were set at 1 kPa and 20.5 MPa,respectively. These physiological conditionsare associated with maximal gas exchangecapacity for terrestrial basal angiosperms(Feild et al. 2009). Leaf thickness affectsestimates of maximum CO2/H2O gas ex-change by varying dm (Brodribb et al. 2007).Thus, leaf gas exchange capacities werecalculated at lower (70 mm) and upper

(140 mm) ends for the range of vein-epidermalthicknesses for mesophytic angiosperm leaves(Brodribb et al. 2007; D. S. Chatelet and T. S.Feild unpublished data 2009). This procedurewas applied because leaf thickness cannot bemeasured in compression fossils.

Measurements of Stomatal Apparatus Geome-try in Living and Fossil Leaves and Calculation ofMaximum Stomatal Conductance to Water Va-por.—An accepted equation to calculate themaximum diffusive conductance of the sto-mata (gstoma) was used (Parlange and Wag-goner 1970; Van Gardenigen et al. 1989;Kaiser 2009):

gstoma~1= d=p � a � bð Þ½ �f

zln 4a=bð Þ=p � a�=D �Ng ð1Þwhere a 5 guard cell pore length/2, b 5

guard cell pore width/2, d 5 guard cell poredepth; D 5 diffusivity of water vapor in air, N5 guard cell density. Measurements of d weremade on cross-sections of FAA fixed leavesembedded in plastic resin (JB-4, PolysciencesInc., Warrington, Pennsylvania), and sec-tioned at 5-mm thickness on a rotary micro-tome (RM2245, Leica Microsystems, Ger-many). From each species, 20 guard cellpores were imaged at 4003 and measuredwith ImageJ. a and N were determined onmacerated cuticles stained in safranin. Cuti-cles were macerated using acetic acid andhydrogen peroxide. We measured five 4-mm2

sheets of cuticle for N at 2003. Geometriclandmarks for assessing guard cell porelength and depth followed previous criteria(Lawson et al. 1998). We assumed thatmaximum guard cell width (b) was approx-imated by one-third of the guard cell porelength (Osborne et al. 2004). Using the lengthand width of the stomatal pore, and approx-imating the pore as an ellipse, we calculatedstomatal pore area at maximal aperture (SPA,m2).

The stomata of extant basal angiospermsand the fossil stomata investigated possessedprominent peristomatal rims over the guardcell pores (Upchurch 1984a,b, 1995). Vesti-bules could lengthen the diffusional pathlength from the stomatal pore to the bulkphase. As a result, the accuracy of equation

FIGURE 1. Comparison of fossil preservation of minorveins in Ficophyllum crassinerve (USNM192353), a putativeaustrobaileyoid stem-lineage with a vein density of3.52 mm mm22 (A) and a sun leaf of extant Amborellatrichopoda with vein density of 3.77 mm mm22 (B). Scalebars, 1.5 mm.


(1) will be affected. The calculated maximumconductance of the stomatal pore complex(gc

STOMA mmol m22 s21) to water vapor is:

gcSTOMA~1= rstomazrvestibuleð Þ ð2Þ

where rstoma is the resistance of the stoma andrvestibule is the resistance of the vestibule atopa stoma. The resistances of each term werecalculated by substituting the length, width,and depth of the vestibule or the stoma intoequation (1). Vestibule depth was measuredat 4003 from the same cross-sections asdescribed above, and vestibule length andwidth determined from cuticle macerations at4003. To improve the accuracy of vestibulegeometry measurements, we observed stoma-tal vestibules of some species with scanningelectron microscopy on critically dried leaves(observations not shown). An approximationof the cuticular vestibule and stomatal poreresistors as additive in series makes threeassumptions: (1) the architectures can beapproximated as connected pores of simplecylindrical geometry; (2) diffusion within thetwo components does not involve exchangethrough the walls of each; and (3) otherresistors to water vapor diffusion, such asinternal cuticle and intercellular conductance,are not significant. Assumptions one and twoare reasonable because vestibules consist of athick cuticle that is likely to be impermeableto water vapor. Assumption three is validbecause maximum conductance was calculat-ed (Kaiser 2009).

On fossil leaf cuticles, measurements weremade on previous cuticular preparationsfrom four localities (Appendix III, online)(Upchurch 1984a,b, 1995; Upchurch andDilcher 1990). Guard cell pore length as wellas the length and width of the vestibuleaperture were measured at 4003. Guard cellwidth was approximated as one-third of thepore length. The taxon means are from tenstomata. Stomata pore or vestibule depthscould not be measured on the prepared fossilcuticles. However, stomatal pore depth cor-related with stomatal pore length (n 5 113; r2

5 0.87) across a broad range of stomatal size(i.e., 18 mm to ,100 mm in maximum length[D. S. Chatelet unpublished data 2010]). Thus,stomatal length measurements were used to

calculate stomatal pore depth. Stomatal ves-tibule depth of fossil leaf cuticles was ap-proximated in the calculations as the averagedepth (5.48 6 2.6 SD mm, n 5 113) foundacross all of the extant angiosperms sampled.

Phylogenetic Analyses.—To compare leaffossils to reconstructed nodes of trait evolu-tion across extant angiosperm phylogeny, weassembled a composite phylogenetic tree bygrafting species-level trees onto two well-supported backbone topologies using Mes-quite version 2.6 (Maddison and Maddison2008). The species-level phylogenies includedthe following clades: Chloranthales, Illicia-ceae, Schisandraceae, Trimenia, and Nym-phaeales (Jansen et al. 2007; Moore et al.2007; Saarela et al. 2007; Endress and Doyle2009; Feild et al. 2009). Published consensustrees were used, and the polytomies treated assoft. The composite tree was pruned to leavethe species sampled.

Ancestral state values for functional traitswere reconstructed using weighted squared-change parsimony over species terminalsusing Mesquite. This method minimizes thesum of squared change along all branches ofthe tree to reconstruct the values of internalnodes from the trait values of the species’terminals (Maddison and Maddison 2008).For species with sun and shade leaves,ancestral state trait reconstructions wereperformed using the maximum trait values.Then, a separate analysis was run using theaverage of sun and shade values to define aspecies trait value. In cases where the phylo-genetic relations were uncertain, Mesquitewas used to randomly resolve phylogeneticrelations within clades (100 times), and thetrait values for nodes within the tree wereaveraged (Appendix IV, online).

Results

Leaf Venation Density and Gas ExchangeCapacity in Extant and Extinct Leaves.—Leafphotosynthetic capacity (PC) and maximumstomatal conductance to water vapor (gm)increased linearly with increasing vein density(Dv; Fig. 2A,B). Nymphaeales, however, rep-resented exceptions with high PC and gm butwith comparatively low Dv (Fig. 2A). Acrossall species, PC increased with gm (Fig. 2C).


Vein densities of extant Chloranthalesranged from 1.62 mm mm22 for shade leavesof Hedyosmum orientale to 5.06 mm mm22 forsun leaves of Ascarina maheswari. ExtantAustrobaileyales Dv ranged from 1.94 mmmm22 for shade leaves of Austrobaileya scan-dens to 6.35 mm mm22 for sun leaves ofKadsura heteroclita. Mean Dv of all Zone I fossilspecies measured from the Potomac Groupextended over approximately 50% of theextant terrestrial basal angiosperm range,varying from 2.8 mm mm22 6 0.36 SD inProteaephyllum reniforme to 4.1 mm mm22 6

0.18 SD in ‘‘Sapindopsis’’ elliptica. The mean forZone I leaves (3.38 mm mm22 6 0.35 SD; n 5

12 species) did not differ significantly fromthe extant terrestrial basal angiosperm mean(results not shown). Zone I fossil leaves alsoexhibited Dv values similar to the reconstruct-ed Dv trait values over the base of extantangiosperm phylogeny (Fig. 2; SupplementalTable 1 and Fig. S1).

Vein densities for chloranthoids averaged3.58 mm mm22 6 0.62 SD (n 5 11 species;Fig. 3) and ranged from 2.47 mm mm22 6 0.08SD in Reynoldsiophyllum nebrascense (RoseCreek, Late Albian) to 4.61 mm mm22 6 0.46in Crassidenticulum cracendentis (Courtland,Cenomanian). Dv of austrobaileyoid leavesaveraged 3.61 mm mm22 6 0.46 SD (n 5 5)ranged from 3.12 mm mm22 6 0.21 SD in theFredericksburg sample of Eucalyptophyllumoblongifolium (Upper Zone I, Potomac Group)to 4.36 mm mm22 6 0.32 SD in Longstrethiavaridentata (latest Albian, Rose Creek; Fig. 3).In addition, mean vein densities for chlor-anthoids and austrobaileyoids did not differsignificantly (results not shown). No trend inDv across the sampled fossil species wasobserved through time (Fig. 3). Finally, max-imum CO2 exchange and stomatal conduc-tances to water vapor calculated from veindensity (gc

VEIN) for chloranthoid and austro-baileyoid species through time fell within the10th and 90th percentiles’ extant species andreconstructed ancestral node trait values overthe base of extant phylogeny (Fig. 3).

Stomatal Pore Anatomy and Gas ExchangeCapacity in Leaves of Extant and ExtinctSpecies.—Stomatal size measured on fossilcuticles encompassed approximately 60% of

FIGURE 2. Coordination of structure-function relations ofthe venation system and leaf gas exchange capacity across35 extant basal angiosperm species. A, Relation betweenleaf photosynthetic capacity (PC, mmol CO2 m22 s21) andleaf vein density (Dv, mm mm22; excluding Nym-phaeales, y 5 2.444x, r2 5 0.69). B, Relation betweenmeasured maximum stomatal conductance to watervapor (gm mmol H2O m22 s21) and Dv (y 5 36.1x, r2 50.62). C. Relation of PC to gm (with Nymphaeales, y 50.059x, r2 5 0.79; excluding Nymphaeales, y 5 0.065x, r2 5

0.42). Standardized measurement environmental condi-tions are described in ‘‘Methods.’’ Symbols denotegrowth forms for A–C: #, shrubs and trees; e lianas;%, sub-shrubs; n, terrestrial herbs; ,, floating leaves ofaquatic Nymphaeales herbs. Points are the averages ofthree observations made on five to ten leaves per species,and error bars denote standard deviation about the mean.The dotted lines in A and B refer to modeled photosyn-thetic capacities and maximum stomatal conductancesbased on vein density at 140 and 70 mm leaf thicknesses.Species information and data are provided in AppendixI (online).


the stomatal size and density range for extantAustrobaileyales and Chloranthales (Fig. 4A).Across extant terrestrial basal angiosperms(excluding the sun leaves of Kadsura coccineathat possessed unusually large stomata),maximum calculated stomatal conductancefrom pore geometry (gc

STOMA) increased withmaximal stomatal pore area (SPA) for diffu-sive exchange (Fig. 4B; gc

STOMA 5 30.15 SPA 2

53.57; n 5 104; r2 5 0.8). Similarly, gcSTOMA

increased with greater SPA in fossil cuticles[(gc

STOMA) 5 34.36 (SPA) 2 75.73, n 5 10, r2 5

0.82; Fig. 4B].Maximum calculated stomatal conductance

from pore geometry ranged from 105 mmolH2O m22 s21 in the austrobaileyoid Ficophyl-lum crassinerve to 327 mmol H2O m22 s21 inCelastrophyllum latifolium (Figs. 4B, 5). Over

FIGURE 3. Leaf vein density (Dv, mm mm22) and leaf CO2/H2O gas exchange capacities (photosynthetic capacity, PC,mmol CO2 m22 s21 and stomatal conductance to water vapor, gc

VEIN) calculated from a coupled venation hydraulic-photosynthetic model with CO2 concentration set at 380 mL L21 CO2 for early angiosperm leaf fossils. Gray regions onthe timescale refer to the stratigraphic age range for the fossil localities (Zone I Potomac Group [lower upper Aptian],Braun Ranch and Rose Creek [upper Albian], Courtland [Cenomanian]). PC and gc

VEIN were parameterized using twovein-to-epidermal hydraulic distances from 70 mm to 140 mm to produce a span of predicted values encompassing thelikely morphological variability in the thicknesses of angiosperm leaves. The two points for PC and gc

VEIN for each fossilleaf taxon reflect these values. Symbols refer to previously hypothesized systematic affinities of fossil leaves: stem-lineage austrobaileyoids (&) and chloranthoids (N), and fossil leaves of uncertain systematic placement (e). Numbersrefer to the species: 1, Proteaephyllum reniforme; 2, Celastrophyllum latifolium; 3, Eucalyptophyllum oblongifoliumFredericksburg sample; 4, Celastrophyllum sp. Drewry’s Bluff; 5, Drewry’s Bluff Leaf Type #1 (Moutonia); 6,Celastrophyllum sp. (C. obovatum); 7, Vitiphyllum multifidum; 8, Quercophyllum tenunerve; 9, Eucalyptophyllum oblongifoliumDrewry’s Bluff; 10, Ficophyllum crassinerve; 11, Rogersia angustifolia; 12, cf. Ficophyllum; 13, Sapindopsis elliptica; 14,Reynoldsiophyllum nebrascense; 15, Crassidenticulum landiase; 16, Crassidenticulum decurrens Braun Ranch samples; 17,Longstrethia aspera; 18, Dennsinervum kaulii Rose Creek samples; 19, Crassidenticulum decurrens Rose Creek samples; 20,Longstrethia varidentata; 21, Densinervum kaulii Courtland sample; 22, Crassidenticulum cracendtis. The dashed lines acrossthe three fossil data panels indicate the 10th and 90th percentiles of gc

STOMA across the extant species. The box plotsdepict the variation within extant clades with the bottom and top of the box indicating the 25th and 75th percentiles,respectively, the two whiskers the 10th and 90th percentiles, respectively, and the horizontal line within the box, themedian value. Symbols beyond the whiskers are outliers.


the fossil species sampled, no time-dependentpattern in gc

STOMA was found (Fig. 5). Maxi-mum calculated stomatal conductance frompore geometry of fossil cuticles nested withinthe range of extant species (Figs. 4B, 5).Consistent with the result that SPA drivesmuch of gc

STOMA, water vapor conductances ofstomatal vestibules were three to six timesgreater than calculated stomatal conductancesof just the stomata pore across all of the extantand extinct leaves sampled (Fig. 4C).

gcSTOMA was related positively to maximum

stomatal conductance (gm) as measured byleaf gas exchange analysis across terrestrialbasal angiosperm species (gc

STOMA 5 1.632 (gm)+ 9.04, n 5 30; r2 5 0.64; Fig. 6A). gc

STOMA

values, however, were greater than gm interrestrial basal angiosperm species (Fig. 6A).By comparison, maximal stomatal conduc-tance to water vapor calculated from veindensity (gc

VEIN) yielded estimates closer tomeasured stomatal conductance maxima onextant leaves (gc

VEIN 5 0.514 (gm) 2 57.56, n 5

30; r2 5 0.60; Fig. 6A). Nymphaeales leaves,however, exhibited much greater gm incomparison to gc

STOMA and gcVEIN (Fig. 5B). Dv

and SPA were unrelated, as were gcSTOMA and

gcVEIN (Fig. 6C).

Discussion

Fossil Evidence for Low Gas Exchange Capacityin Early Angiosperm Leaves.—All Zone IPotomac Group fossil angiosperm leavesmeasured had structural traits related tolower gas exchange capacities, in contrast tomodern ecologically dominant angiosperms.Dv values of late Aptian–early Albian fossilangiosperm leaves occurred in the narrowrange of low densities found across ferns andgymnosperm clades that dominated plantcommunities during the Early Cretaceous(Lupia et al. 1999; Boyce et al. 2009). Theuniform pattern of low vein densities acrossZone I angiosperm fossil leaves contrasts withthe widespread dominance of high vein

FIGURE 4. Relations among stomatal morphology andmaximum stomatal conductance to water vapor (mmolH2O m22 s21) for extant basal angiosperms and fossilcuticles. A, Stomatal size (mm2) versus stomata density(number mm22 of leaf area) for extant basal angiospermsand fossil cuticles. B, Maximal stomatal conductance towater vapor calculated from stomatal pore dimensions(gc

STOMA, mmol H2O m22 s21) versus stomatal pore surfacearea (m2 3 106) across extant and extinct fossil cuticles. C,Calculated maximal water vapor conductance of thestomatal vestibule (gc

VESTIBULE, mmol H2O m22 s21) versusgc

STOMA for fossil and extant cuticles. The dashed line showsthe one-to-one relation. #, extant basal angiosperm leaves

r

(including Amborella, Austrobaileyales, Chloranthales, andNymphaeales); and N, fossil cuticles (see Appendix II,online, for the taxa sampled).


densities characterizing the vast majority ofecologically abundant angiosperms today(Boyce et al. 2009; Brodribb and Feild 2010).

The functional contrast between Zone Iangiosperms and extant angiosperms be-comes especially striking when regenerationhabitat is controlled. Extant floodplain colo-nizing angiosperms from warm temperateand tropical regions, which represent ecolo-gies thought to approximate the depositionalsettings for many Zone I angiosperms, exhibitvein densities ranging from 10 to 20 mm mm22

(Doyle and Hickey 1976; Hickey and Doyle1977; Brodribb et al. 2007; Boyce et al. 2009;Brodribb and Feild 2010). Photosyntheticcapacities and stomatal conductances of thesehigh Dv angiosperms delimit the extant limitsof C3 photosynthesis in woody plants, oper-ating close to 30 mmol CO2 m22 s21 and700 mmol H2O m22 s21 (Feild and Balun2008; T. S. Feild, unpublished data fromPapua New Guinea 2009–2010). By contrast,the scaling of vein density with leaf CO2 and

H2O exchange capacity across vascular plantsindicates that maximum leaf gas exchangecapacities of low-Dv late Aptian–early Albianangiosperms would be on average three timeslower than in modern weedy riparian angio-sperms photosynthesizing under physiologi-cally optimal conditions (Fig. 3) (Brodribb etal. 2007; Boyce et al. 2009). Instead, photosyn-thetic and transpirational capacities modeledfrom veins and the stomatal pore geometry ofZone I fossil leaves as well as extant terrestrialbasal angiosperm leaves equaled the lowcapacities of mesic shade-tolerant woodyangiosperms, ferns, and some extant conifersand cycads (Korner 1995; Brodribb and Hill1997; Brodribb and Feild 2000, 2010; Lusk etal. 2003; Brodribb et al. 2007; Feild and Balun2008; Franks and Beerling 2009; Kaiser 2009).Other ferns (Dicranopteris, Dipteris, Gleichenia),lycopods (Lycopodium), conifers (several Pinusspecies), and Gnetum from sun-exposed dis-turbed habitats possessed higher leaf gasexchange capacities than those inferred for

FIGURE 5. Maximum stomatal conductance to water vapor calculated from stomatal pore dimensions (gcSTOMA, mmol

H2O m22 s21) of fossil austrobaileyoid (&), chloranthoid (N), illicioid cuticles (% filled gray), and of uncertainsystematic placement (e) as compared to extant Chloranthales, Amborella-Austrobaileyales, and Illiciaceae plusSchisandraceae. Numbers refer to taxa: 1, cf. Ficophyllum; 2, Eucalyptophyllum; 3, Drewry’s Bluff Leaf Type #1; 4,Celastrophyllum sp.; 5, Celastrophyllum latifolium; 6, Moutonia; 7, Longstrethia varidentata; 8, Illiciales Type I; 9, IllicialesType 2; 10, Protoilliciales. Gray regions on the timescale refer to the stratigraphic ages for the fossil localities ([Zone IPotomac Group [lower upper Aptian], Rose Creek and Kiowa Formation [upper Albian], and Sugarite Coal [upperMaastrichtian and lowest Paleocene]). The dashed lines across the three fossil data panels indicate the over extant rangefrom the 10th and 90th percentiles of gc

STOMA. The box plots depict the variation within extant clades with the bottomand top of the box indicating the 25th and 75th percentiles, respectively, the two whiskers the 10th and 90th percentiles,respectively, and the horizontal line within the box, the median value. Symbols beyond the whiskers are outliers.


late Aptian–early Albian angiosperms (Brod-ribb et al. 2007; Brodribb and Feild 2008; Feildand Balun 2008; T. S. Feild, unpublished datafrom New Caledonia, 2009).

Evidence for Conservatism of Low Leaf GasExchange Capacities in Extant Terrestrial BasalAngiosperm Leaves.—Vein densities, maximalstomatal pore areas, and calculated gasexchange capacities for all species of austro-baileyoid and chloranthoid fossil leaves sam-pled fell within the narrow range of lowvalues characterizing modern diversity ofextant terrestrial basal angiosperm leaves(Appendix I, online). In addition, the func-tional trait values of fossils nested within therange of ancestral trait values reconstructedfrom extant angiosperms across major early-diverging nodes up to the common ancestorof eudicots, monocots, and magnoliids(Figs. 3, 5). Thus, our results support thehypothesis that extant Austrobaileyales andChloranthales retain genuine functional sig-nals of the photosynthetic capabilities foundin some Early Cretaceous angiosperms.

A caveat to the view that metabolic capac-ities of extant basal angiosperm leaves equalthose of some Early Cretaceous angiosperms isthat phylogenetic relations of the hypothesizedfossil austrobaileyoid and chloranthoid leaffossils are difficult to know. A robust phylo-genetic analysis of fossil leaf taxa sampled andextant angiosperms remains elusive becausetoo few characters are preserved (Doyle 2007).Another issue is that some characters suggest-ing that fossils are stem-lineages to Amborella,Austrobaileyales, and/or Chloranthales, suchas chloranthoid leaf teeth and low regularity ofvenation patterns (low leaf rank), also occur inextant magnoliids and basal eudicots (Up-church 1984a,b; Upchurch and Dilcher 1990;Carpenter 2005). Thus, such traits are plesio-morphic for angiosperms as a whole and notdiagnostic of extant basal lineages (Doyle2007). This phylogenetic pattern is importantbecause most magnoliids and basal eudicotspossess leaves with greater CO2/H2O ex-change capacities than Amborella, Austrobai-leyales, and Chloranthales species (Brodribband Feild 2010).

However, extant basal eudicot and magno-liid leaves of low leaf rank (defined as 1r0–

FIGURE 6. Comparisons of maximum stomatal conduc-tances to water vapor (mmol H2O m22 s21) as calculatedfrom stomatal pore geometry (gc

STOMA) and vein density(gc

VEIN) across extant basal angiosperm leaves withmeasured stomatal conductances gm (mmol H2Om22 s21) and leaf structural traits. A, gc

STOMA and gcVEIN

versus gm. Dashed line represents the one-to-one relation.Symbols are as in Figure 2, with open symbols denotinggc

STOMA and filled symbols denoting gcVEIN. B, Stomatal

pore area (m2 3 106) versus vein density (mm mm22) overall of the extant angiosperm species sampled for gasexchange change measurements. C, gc

STOMA versus gcVEIN

for all of the extant angiosperm species sampled for gasexchange change measurements. Error bars have beenomitted for clarity.


2r3; Todzia and Keating 1991) and withchloranthoid teeth exhibited low Dv

(Fig. 7A,B). Also, only a few basal eudicotspecies with chloranthoid teeth developedcomparatively higher Dv (.6 mm mm22).These taxa included most Gunnera species, thepoikilohydric shrub Myrothamnus flabellifolius,and a few xeric-adapted Papaveraceae spe-cies. These species, however, exhibited higherleaf rank than most basal lineages and Zone Ifossil leaves (Doyle and Hickey 1976; Hickeyand Doyle 1977; Todzia and Keating 1991;Fuller and Hickey 2005). Thus irrespective ofphylogenetic affinity, fossil angiospermleaves of low rank, and especially if they alsopossess chloranthoid teeth, mark relativelylow leaf gas exchange capacity. These resultssupport a hypothesis that Dv and leaf rankcorrelate across angiosperms (Brodribb andFeild 2010). If increasing regularity of majorvein development is associated with theability of a higher-order venation system tomore efficiently fill in the finest areoles that

dictate hydraulic flux at the vein terminal(Brodribb et al. 2007; McKown et al. 2010),then leaf rank and vein density may bedevelopmentally linked. However, more re-search is needed to test such a hypothesis. Wedo not expect that a mechanistic relationshipholds between low vein density and chlor-anthoid teeth, because chloranthoid leaf teethare involved in releasing sap pushed to theleaves by root pressure (Feild et al. 2005). At aminimum, our results support the hypothesisthat Early Cretaceous Zone I leaves, as well asspecies of austrobaileyoid and chloranthoidfossils, exhibited ranges of photosyntheticand transpiration capacities analogous toextant terrestrial basal angiosperm lineages.

Evidence for Wet Habitat Adaptation inAncient Austrobaileyoid/Chloranthoid FossilLeaves.—Although reconstructed leaf gas ex-change capacities of austrobaileyoid andchloranthoid leaf fossils compared well withtheir hypothesized extant descendants, simi-larity in low gas capacities, by itself, cannot

FIGURE 7. Vein densities (Dv, mm mm22) of extant basal angiosperms, 65 species of magnoliids, and basal eudicotleaves. A, Basal eudicot leaves with chloranthoid leaf teeth. B, Leaf venation architecture of low rank organization. Boxplots depict the variation within extant clades with the bottom and top of the box indicating the 25th and 75thpercentiles, respectively, the two whiskers the 10th and 90th percentiles, respectively, and the horizontal line within thebox, the median value. Symbols beyond the whiskers are outliers. The statistical differences between group means(Student’s independent t-test, at least p , 0.05) are indicated by different letters. Basal lineage data comprise Amborella,Austrobaileyales, and Chloranthales. Basal eudicot lineages comprise Ranunculales (Ranuculaceae, Papaveraceae,Lardizabalaceae, Berberidaceae), Trochodendrales, Buxales, Gunnerales, and Proteales. Magnoliid lineages compriseMagnoliales, Canellales, Laurales, and Piperales. Dv values are taken from Brodribb and Feild (2010), and leaf rankscores from Todzia and Keating (1991).


diagnose whether these fossil leaves camefrom plants adapted to wet (.2000 mm yr21

rainfall) and forest understory habitats likethose characterizing Amborella as well as mostAustrobaileyales and Chloranthales (Feild etal. 2004, 2009). Specifying the habitat contextof austrobaileyoid/chloranthoids is criticalfor determining how functional trait diversityin extant basal lineages bears on how the firstflowering plants functioned and the selectivecontexts responsible for diverse hypothesizedkey innovations of early angiosperm success(Hickey and Doyle 1977; Retallack and Dil-cher 1981; Taylor and Hickey 1996; Feild andArens 2007; Williams 2008; Feild et al. 2009).Paleoenvironmental proxy records indicatehigh annual rainfalls (up to 4500 mm yr21)and tropical to paratropical temperatures forall of the fossils sampled (Upchurch andWolfe 1987; Upchurch and Dilcher 1990;Upchurch 1995; White et al. 2001; Ufnar etal. 2008). However, there is tremendousvariation in evaporative demand within near-ly all plant communities, and thereforediscerning the evaporative niches of fossilleaves demands a finer specification of cano-py position (Feild et al. 2004).

Across extant basal angiosperms, low veindensity and stomatal pore area signal largehydraulic constraints on photosynthesis andwhole plant function, and, therefore, occur-rence in wet low-evaporative-demand habi-tats. Strong coupling between veins andhabitat occurs in basal angiosperms becausethese taxa cannot overcome the constraints ofhigh xylem hydraulic resistance by operatingat large drops in water potential (i.e., toleratehigh drought stress) from the soil to leaf todrive high transpiration (Sperry et al. 2007;Feild et al. 2009). Instead, leaves of extantbasal angiosperms wilt and the stem xylemvasculatures become embolized at modestdrought stress (Sperry et al. 2007; Feild et al.2009). Thus, to determine if austrobaileyoidand chloranthoid leaves managed hydraulicfunction differently than extant relatives, aknowledge of their water stress physiology isnecessary. In addition, how such processesinteracted with environmental conditions thathave no modern analog—particularly highCO2 and O2 during the Cretaceous, which

vary the amount of leverage that vein densityand water-use efficiency exert on whole plantfunction and ecological distribution—must beconsidered (Sperry 2003; McElwain et al.2005; Brodribb and Feild 2010).

A key piece of structural evidence pointingto low drought-stress tolerance in austrobai-leyoids and chloranthoids is that several ofthese fossils possess glandular chloranthoidleaf teeth (Hickey and Doyle 1977; Upchurch1984a,b, 1995; Upchurch and Dilcher 1990;Wang and Dilcher 2006). In extant basalangiosperms, chloranthoid leaf teeth releaseguttation sap during root pressure, whichpressurizes the xylem (Feild et al. 2005). Rootpressure occurs when the soil is at fullhydration and the atmosphere saturated withwater vapor—conditions that are fleeting inall but the wettest terrestrial habitats (Sperry2003). Nonetheless, root pressure is an essen-tial mechanism for restoring lost xylemtransport capacity following drought byrefilling embolized conduits across terrestrialbasal angiosperms (Sperry et al. 2007). There-fore, we hypothesize that Early Cretaceousangiosperms with chloranthoid teeth likelyguttated and had low drought tolerance.Because of their low vein density andstomatal pore area, such leaves are predictedto have occurred in wet zones of lowtranspirational pull.

Austrobaileyoid fossils, some of which lackchloranthoid teeth, were also likely adapted todamp, low-evaporation habitats because theypossessed large stomatal vestibules (Upchurch1984a,b, 1995). Stomatal vestibules do not actas anti-transpirants (Fig. 4C), but insteaddecrease leaf surface wettability (Schonherrand Bukovac 1972; Feild et al. 2005, 2009).Vestibules in other plants have been demon-strated to keep guard cell pores from beingdrowned by continuous water films formedover the cuticle during high-humidity conden-sation and heavy rainfall, which otherwiseinhibit CO2 uptake, favor fungal invasion, andleach nutrients by excessive cuticle hydration(Schonherr and Bukovac 1972; Feild et al.2005). The widespread pattern of large, low-density stomata found across extant terrestrialbasal angiosperms and austrobaileyoid/chlor-anthoid fossils suggests that these leaves share


slow stomatal response kinetics and lowoptimization of long-term carbon gain withrespect to water loss—functional traits ofleaves adapted to wet humid zones (AppendixI, online) (Franks and Beerling 2009). In thefuture, measurements of tracheary elementthickness-to-span ratio in leaf veins to quantifytension-induced implosion resistance mayoffer a direct window on the hydraulic anddrought-stress limits of early angiospermleaves (Blackman et al. 2010). At present,interpretations of the function of fossil leafstructure are consistent with the hypothesisthat austrobaileyoid and chloranthoid fossilleaves occurred in wet, very humid habitatssuch as underneath a canopy of non-angio-sperms.

A remaining question is how several of thelow-Dv and -SPA Aptian–earliest Albianangiosperm leaves we measured, includingsome of the austrobaileyoids and chlor-anthoids, came to be deposited in coarse-grained, low-carbon sediments interpreted assampling the front of regenerating vegetationon riparian point bars (Doyle and Hickey1976; Hickey and Doyle 1977; Taylor andHickey 1996). By contrast, no extant Austro-baileyales and Chloranthales occur in thepioneer thickets of open sandy point bars inlowland tropical and temperate riparianvegetation (Feild 2009; Feild et al. 2009). Thisextant exclusion has a physiological basis,because living terrestrial basal angiospermscannot tolerate the high evaporative demandand wide fluctuations in water availabilityfound in riparian point bar zones (Puhakka etal. 1992; Rood et al. 2003; Merigliano 2005;Robertson and Augspurger 1999). Whenextant basal angiosperm taxa with low-Dv

leaves are forced into high evaporativedemand, chronic photoinhibition of leaf pho-tosynthesis, dysfunction of flowering, shootdieback, and/or whole plant mortality occureven when soils are prevented from dryingdown (Feild et al. 2009).

In view of fossil ecomorphic traits linked tolow drought tolerance (see above), austrobai-leyoid and chloranthoids also would beexpected to be excluded from sunny riparianpoint bars. Thus, low water stress tolerance inthese early angiosperms rejects one hypothe-

sis for why extant basal angiosperms do notequal early angiosperms. Unlike Ginkgo,which was likely displaced from riparianzones by angiosperms (Royer et al. 2003),the living descendants of low Dv chlor-anthoids and austrobaileyoids—Chlor-anthales and Austrobaileyales—cannot repre-sent the sun ruderals of the Cretaceous thatlost out and secondarily retreated to the forestunderstory following competitive displace-ment by later waves of more metabolicallyescalated angiosperms in the ancestral sunny,disturbed riparian zone.

The highly sensitive turgor relations (i.e.,wilting at a loss of ,8% relative water contentand leaf water potentials from 20.8 to21.2 MPa) of extant basal angiosperm leaves,and by inference those of austrobaileyoidsand chloranthoids, means that even thehighest CO2 concentrations predicted for theCretaceous (,2000 mL L21) will not saveenough water to empower such plants intosites with high evaporative load (Jones 1993;Sperry 2003; Sperry et al. 2007; Feild et al.2009; Royer 2010). More broadly, there is noevidence that elevated CO2 can shift thecanopy positioning or evaporative niche ofany extant species. Studies reporting theeffects of increased CO2 on leaf gas exchangeand water relations demonstrated small or nochange in water use, leaf water potentialunder maximum evaporative load, leaf andstem xylem hydraulic conductance, anddrought tolerance under high CO2 (Kerstiens1998; Wullschleger et al. 2002; Korner 2009). Itis notable that trait values were invariantfrom the Aptian–earliest Albian to earlyPaleocene (a ,40-Myr-long period) despiteconsiderable atmospheric and ecologicalchange through this interval (Wing andBoucher 1998; Lupia et al. 1999; Royer 2010).However, future fossil tests are needed to testwhether water use physiologies of extinctaustrobaileyoids or chloranthoids changed inways different from living relatives duringfrom the Paleocene to Quaternary. We sug-gest that the indirect effects of CO2 on varyingregional evaporative demand by increasedglobal temperature, feedbacks with cloudi-ness, and variations in the intensity of theglobal hydrological cycle were more impor-


tant in influencing ecological distribution ofEarly Cretaceous angiosperms (White et al.2001; Ufnar et al. 2008).

Instead, the depositional patterns of Ap-tian–early Albian angiosperms investigatedare probably explained by taphonomic com-plexities. Hickey and Doyle (1977) suggestedthat the small size, fragmentary nature, andrarity of Zone I leaves were consistent withshort-distance transport. Such small transportpossibilities from understory sites are highlyplausible because the depositional environ-ments responsible for leaf compression fossilassemblages formed by meandering riparianzones and floodplains are complex in waysthat obscure the specification of evaporativepreference based on sediments alone (Gas-taldo et al. 1987; Robertson and Augspurger1999; Feild et al. 2004; Merigliano 2005;Richardi-Branco et al. 2009). Importantly, theregeneration niches of Austrobaileyales andChloranthales provide living proof that mi-crosites combining fluvial disturbance withlow evaporative demand, low carbon, andsandy sedimentology exist (Feild et al. 2004;Ito et al. 2006; Feild 2009). Such dark anddisturbed sites are often near higher flow-energy channels in tropical floodplain zones,semi-shaded stream margins (not necessarilywith tall closed forest canopies) along up-stream cutbanks of floodplains (Ito et al. 2006;Feild and Arens 2007; Feild et al. 2009).

Veins Versus Stomata and the RetrodictingFossil Leaf Gas Exchange Capacity from LeafStructure.—Our data allowed us to test theaccuracy of different models for reconstruct-ing the gas exchange performance of fossilleaves. Diffusional maxima calculated fromstomatal pore anatomy were much lessaccurate than analyses based on vein densityto predict the maximum capacity of leaves tolose water and take up CO2 (Fig. 5). Theseresults emphasize the advantages of measur-ing veins over stomata to infer gas exchangecapacity of fossil leaves.

Why are stomata-based estimates of maxi-mal stomatal conductance less accurate thanthose derived from veins? A key problem isthat stomata function through movement.Animate functioning means that experimen-tally difficult-to-test assumptions must invari-

ably be made about what length-to-widthratios (PL:PW) of the pore define the maximaldiffusive potential (Van Gardingen et al. 1989;Kaiser 2009). The standard approach has beento assume that maximal width is a constantfraction of the pore length (Osborne et al.2004; Franks and Beerling 2009). However,PL:PW has a phylogenetic component acrossvascular plants (Franks and Farquhar 2007).Such an approach is inaccurate for cladessuch as grasses that open maximally at a one-to-one ratio of PL:PW (Franks and Farquhar2007). Nymphaeales may represent a similarcase. If PL:PW is an assumed one-to-one ratiofor Nymphaeales’ stomata, then calculatedstomatal conductance maxima for most spe-cies fall on the one-to-one line with measuredstomatal conductance. Unknown maximalpore dimensions are important because smallerrors in pore area have large effects oncalculated conductances for one-dimensionaldiffusion (Parlange and Waggoner 1970).Such errors are most problematic when thestomatal area is large, resulting from large-sized stomata or high stomatal density(Fig. 6A). Another significant unknown isthe effects of stomatal ornamentations, suchas crypts, plugs, and peristomatal rims, whichpare back effective pore area (Brodribb andHill 1997). For stomatal vestibules of Austro-baileyales and Chloranthales, at least, theeffects on calculated conductance are small.

By contrast, veins are fixed in space bymesophyll tissue, which means that hydraulicdistribution is approximated by a staticgeometry. Geometric spacing of veins is oftenwell preserved in fossils because veins arelignified. Veins, however, are not withoutsignificant problems. Nymphaeales, for ex-ample, illustrate that aquatic plants representa functional type with venation-dependent,liquid-phase transport uncoupled to the ca-pacity for diffusive gas exchange. Futurestudies could elucidate how vein constructioncosts and how major versus minor veinhydraulics influence leaf gas exchange capac-ity (McKown et al. 2010).

Conclusions

The low vein densities and low maximalstomatal pore areas found in late Aptian–


early Albian fossil terrestrial angiospermleaves indicate that high leaf gas exchangecapacities evolved later than previously as-sumed during angiosperm evolution (Feild etal. 2004; Brodribb and Feild 2010). Althoughpreviously at least one taxon from Zone Isediments was proposed as adapted to theforest understory (Doyle and Hickey 1976),our results expose a diversity of earlyangiosperms that functioned with low gasexchange capacities and low drought toler-ance. During the Aptian–earliest Albian,angiosperms were minor ecological players,making up less than 5% of the globalabundance and species diversity (Lupia etal. 1999; Heimhofer et al. 2005; McElwain etal. 2005). Consistent with this pattern, angio-sperm leaves from the Zone I of the PotomacGroup explored only 10% of the functionalmorphospace of leaf vein densities exhibitedby modern angiosperms (Boyce et al. 2009;Brodribb and Feild 2010). Thus, many of theZone I taxa of the Potomac Group fossilleaves that originally motivated the ancestralweed hypothesis did not function with highgas exchange capacities and opportunisticenergy use as found in modern weedyangiosperms. The vast majority of livingterrestrial basal angiosperms were similarlylimited in their exploration of this functionalperformance space.

Although to our knowledge the fossilleaves measured represent the oldest assessedfor ecophysiological performance so far, ourwork is limited in two important ways: thefossils sampled represent a single region inspace and time, and they are not the earliestknown angiosperms. Therefore, other gasexchange capacities could be ancestral. Fossilleaves from other localities that are coevalwith Zone I of the Potomac Group or 5–10 Myrolder are known (Sun and Dilcher 2002;Cuneo and Gandolfo 2005; Coiffard et al.2007; Archangelsky et al. 2009). In addition,an older radiation of the angiosperm clade,nearly 20 Myr older than Zone I of thePotomac Group, is evidenced by fossil pollen(Brenner 1996; Doyle 1999). Nevertheless, theavailable structural observations for otherAptian and older fossils are consistent withthe hypothesis that low leaf gas exchange

capacities and drought intolerance are ances-tral. All of the coeval and older leaves withpreserved venation appear to be of low leafrank, and some fossils possess chloranthoidleaf teeth—traits associated with low Dv andwet habitats (Sun and Dilcher 2002; Cuneoand Gandolfo 2005; Archangelsky et al. 2009).For the older pollen fossils, it is significantthat chloranthoids represent a major compo-nent (Brenner 1996; Doyle 1999). Analogizingthese fossils to extant Chloranthales andyounger chloranthoid leaf fossils suggeststhat the plants that produced these grainsfunctioned with low gas exchange anddrought intolerance.

Finally, our results on Early Cretaceousangiosperm leaf form and function provide afirst test of the extent to which extant basalangiosperms are ecophysiologically equiva-lent to some early angiosperms. We foundthat fossil leaves hypothesized as stem-line-age relatives to extant basal angiospermspossessed gas exchange capacities, and likelyleaf water relations, similar to those of theirclosest extant descendants. Leaf carbon–wateruse represents a fundamental determinant ofplant distribution in space and time as well asof whole plant life history (Stebbins 1974;Bond 1989; Sack and Holbrook 2006). Thus,our results provide evidence for the ofteninherently assumed hypothesis that compar-ative research on extant basal angiospermbiology reveals genuine functional signalsfrom the Early Cretaceous (Williams 2008;Endress and Doyle 2009; Feild et al. 2009).

Acknowledgments

This research benefited from commentsprovided by J. Doyle, A. Iglesias, G. Jordan,C. Jaramillo, J. Williams, B. Gomez, P. Wilf, D.Royer, C. Feild, H. Feild, and K. Boyce. Forhelpful discussion on Dakota Formationangiosperms, we thank D. Dilcher, H. Wang,and S. Manchester (Florida Museum ofNatural History). We extend our thanks tothose involved in the field component of theresearch. In Papua New Guinea, L. Balun, R.Banka, B. Bau, K. Tuck, S. Saulei, and R.Kipranis are thanked for support. In Vietnam,we thank Ton That Minh for help in organiz-ing fieldwork in Lam Dong Province at


BiDoup Nui-Ba National Park. In Thailand,we thank Queen Sirikit Botanical Gardens inChiang Mai for logistical support and fieldadvice. For work in Peru, we thank R. Valega,J. Janovec, and M. N. Raurau Quisiyupanqui.We thank J.-Y. Meyer for help in coordinatingfieldwork in French Polynesia. Field samplingwas collected under permission granted byThe National Institute of Natural Resources(Peru), Province Nord (New Caledonia),National Environment and Planning Agency(Jamaica), Delegation a la recherche—Polynesiefrancaise (French Polynesia), and NationalResearch Insitute (Papua New Guinea). Thisresearch was supported by a National ScienceFoundation grant (IOB-0714156) to T.S.F.

Literature Cited

Ackerly, D. D., and M. J. Donoghue. 1998. Leaf size, sapling

allometry and Corner’s rules: a phylogenetic study of correlat-

ed evolution in maples (Acer). American Naturalist 152:767–

791.

Archangelsky, S., V. Barreda, M. G. Passalia, M. A. Gandolfo, M.

Pramparo, E. Romero, R. Cuneo, A. Zamuner, A. Iglesias, M.

Llorens, G. G. Puebla, M. Quattrocchio, and W. Volkheimer.

2009. Early angiosperm diversification: evidence from southern

South America. Cretaceous Research 30:1073–1082.

Barclay, R. S., J. C. McElwain, and B. B. Sageman. 2010. Volcanic

CO2 pulse activates carbon sequestration during Oceanic

Anoxic Event 2. Nature Geoscience 3:205–208.

Blackman, C. J., T. J. Brodribb, and G. J. Jordan. 2010. Leaf

hydraulic vulnerability is related to conduit dimensions and

drought resistance across a diverse range of woody angio-

sperms. New Phytologist 188: doi: 10.1111/j.1469-8137.

2010.03439.x

Bond, W. J. 1989. The tortoise and the hare—ecology of

angiosperm dominance and gymnosperm persistence. Biolog-

ical Journal of the Linnean Society 36:227–249.

Bond, W. J., and A. C. Scott. 2010. Fire and the spread of flowering

plants in the Cretaceous. New Phytologist 188: doi: 10.1111/

j.1469-8137.2010.03418.x

Boyce, C. K., T. J. Brodribb, T. S. Feild, and M. A. Zwieniecki.

2009. Angiosperm leaf vein evolution was physiologically and

environmentally transformative. Proceedings of the Royal

Society of London B 276:1771–1776.

Brenner, G. J. 1996. Evidence for the earliest stage of angiosperm

pollen evolution: a paleoequatorial section from Israel. Pp. 91–

115 in D. W. Taylor and L. J. Hickey, eds. Flowering plant origin,

evolution, and phylogeny. Chapman and Hall, New York.

Brodribb, T., and T. S. Feild. 2000. Stem hydraulic supply is linked

to leaf photosynthetic capacity: evidence from New Caledonian

and Tasmanian rainforests. Plant, Cell and Environment

23:1381–1388.

———. 2008. The evolutionary significance of flat-leaves in the

conifer, Pinus krempfii in Vietnamese rainforest. New Phytolo-

gist 178:201–209.

———. 2010. Leaf hydraulic evolution led to a surge in leaf

photosynthetic capacity during early angiosperm diversifica-

tion. Ecology Letters 13:175–183.

Brodribb, T. J., and R. S. Hill. 1997. Imbricacy and stomatal wax

plugs reduce maximum leaf conductance in Southern Hemi-

sphere conifers. Australian Journal of Botany 45:657–668.

Brodribb, T. J., T. S. Feild, and G. J. Jordan. 2007. Leaf maximum

photosynthetic rate and venation are linked by hydraulics.

Plant Physiology 144:1890–1898.

Carpenter, K. J. 2005. Stomatal architecture and evolution in basal

angiosperms. American Journal of Botany 92:1595–1615.

Coiffard, C., B. Gomez, and F. Thevenard. 2007. Early Cretaceous

angiosperm invasion of western Europe and major environ-

mental changes. Annals of Botany 100:545–553.

Cuneo, R., and M. A. Gandolfo. 2005. Angiosperm leaves from the

Kachaike Formation, Lower Cretaceous of Patagonia, Argen-

tina. Review of Palaeobotany and Palynology 136:29–47.

Doyle, J. A. 1999. The rise of angiosperms as seen in the African

Cretaceous pollen record. Pp. 3–29 in L. Scott, A. Cadman and

R. Verhoeven, eds. Palaeoecology of Africa and the surround-

ing islands, Vol. 26. A. A. Balkema, Rotterdam.

———. 2007. Systematic value of leaf architecture across the

angiosperms in light of molecular phylogenetic analyses.

Courier Forschungsinstitut Senckenberg 258:21–37.

Doyle, J. A., and L. J. Hickey. 1976. Pollen and leaves from the

mid-Cretaceous Potomac Group and their bearing on early

angiosperm evolution. Pp. 139–206 in C. B. Beck, ed. Origin and

early evolution of angiosperms. Columbia University Press,

New York.

Endress, P. K., and J. A. Doyle. 2009. Reconstruction the ancestral

angiosperm flower and its initial specializations. American

Journal of Botany 96:22–66.

Feild, T. S. 2009. Regeneration ecology of early angiosperm seeds

and seedlings: integrating inferences from extant basal lineages

and fossils. Pp. 130–149 in M. A. Leck, V. T. Parker, R. L.

Simpson, eds. Seedling ecology and evolution. Cambridge

University Press.

Feild, T. S., and N. C. Arens. 2007. The ecophysiology of early

angiosperms. Plant, Cell and Environment 30:291–309.

Feild, T. S., and L. Balun. 2008. Xylem hydraulic and photosyn-

thetic function of Gnetum (Gnetales) species from Papua New

Guinea. New Phytologist 177:665–675.

Feild, T. S., N. C. Arens, J. A. Doyle, T. E. Dawson, and M. J.

Donoghue. 2004. Dark and disturbed: a new image of early

angiosperm ecology. Paleobiology 30:82–107.

Feild, T. S., T. L. Sage, C. Czerniak, and W. J. D. Illes. 2005.

Hydathodal leaf teeth of Chloranthus japonicus prevent gutta-

tion-induced flooding of the mesophyll. Plant, Cell, and

Environment 28:1179–1190.

Feild, T. S., D. S. Chatelet, and T. J. Brodribb. 2009. Ancestral

xerophobia: a hypothesis on the whole plant ecophysiology of

early angiosperms. Geobiology 7:237–264.

Fletcher, B. J., S. J. Brentnall, C. W. Anderson, R. A. Berner, and D.

J. Beerling. 2008. Atmospheric carbon dioxide linked with

Mesozoic and early Cenozoic climate change. Nature Geosci-

ence 1:43–48.

Franks, P. J., and D. J. Beerling. 2009. Maximum leaf conductance

driven by CO2 effects on stomatal size and density over

geologic time.Proceedings of the National Academy of Sciences

USA 106:10343–10347.

Franks, P. J., and G. D. Farquhar. 2007. The mechanical diversity

of stomata and its significance in gas-exchange control. Plant

Physiology 143:78–87.

Fuller, D. Q., and L. J. Hickey. 2005. Systematics and leaf

architecture of the Gunneraceae. Botanical Review 71:295–353.

Gastaldo, R. A., D. P. Douglass, and S. M. McCarroll. 1987. Origin,

characteristics, and provenance of plant macrodetritus in a

Holocene crevasse splay, Mobile Delta, Alabama. Palaios 2:229–

240.

Glasspool, I. J., and A. C. Scott. 2010. Phanerozoic concentrations

of atmospheric oxygen reconstructed from sedimentary char-

coal. Nature Geoscience (in press).

Heimhofer, U., P. Hochuli, S. Burla, J. Dinis, and H. Weissert.

2005. Timing of Early Cretaceous angiosperm diversification


and possible links to major paleoenvironmental change.

Geology 33:141–144.

Hickey, L. J., and J. A. Doyle 1977. Early Cretaceous fossil

evidence for angiosperm evolution. Botanical Review 43:3–104.

Hudson, P. J., J. Razanatsoa, and T. S. Feild. 2010. Early vessel

evolution and the diversification of wood function: insights

from Malagasy Canellales. American Journal of Botany 97:80–

93.

Ito, H., S. Ito, T. Matsui, and T. Marutani. 2006. Effect of fluvial

and geomorphic disturbances on habitat segregation of tree

species in a sedimentation-dominated riparian forest in warm-

temperate mountainous region in southern Japan. Journal of

Forest Research 11:405–417.

Jansen, R. K., Z. Cai, L. A. Raubeson, H. Daniell, C. W.

dePamphilis, J. Leebens-Mack, K. F. Muller, M. Guisinger-

Bellian, R. C. Haberle, A. K. Hansen, T. W. Chumley, S. B. Lee,

R. Peery, J. R. McNeal, J. V. Kuehl, and J. L. Boore. 2007.

Analysis of 81 genes from 64 plastid genomes resolves

relationships in angiosperms and identifies genome-scale

evolutionary patterns. Proceedings of the National Academy

of Sciences USA 104:19369–19374.

Jones, H. J. 1993. Plants and microclimates. Cambridge University

Press, Cambridge.

Kaiser, H. 2009. The relation between stomatal aperture and gas

exchange under consideration of pore geometry and diffusional

resistance in the mesophyll. Plant, Cell and Environment

32:1091–1098.

Keeling, H. C., O. L. Phillips. 2007. A calibration method for the

crown illumination index for assessing forest light environ-

ments. Forest Ecology and Management 242:431–437.

Kerstiens, G. 1998. Shade-tolerance as a predictor of responses to

elevated CO2 in trees. Physiologia Plantarum 102:472–480.

Korner, C. 1995. Leaf diffusive conductances in the major

vegetation types of the globe. Pp. 463–490 in E.-D. Schulze

and M. M. Caldwell, eds. Ecophysiology of photosynthesis.

Springer, Berlin.

———. 2009. Responses of humid tropical trees to rising CO2.

Annual Review of Ecology and Systematics 40:61–79.

Lawson, T., W. James, and J. Weyers. 1998. A surrogate measure

of stomatal aperture. Journal of Experimental Botany 325:1397–

1403.

Lupia, R., S. Lidgard S., and P. R. Crane. 1999. Comparing

palynological abundance and diversity: implications for biotic

replacement during the Cretaceous angiosperm radiation.

Paleobiology 25:305–340.

Lusk, C. H., I. Wright, and P. B. Reich. 2003. Photosynthetic

differences contribute to competitive advantage of evergreen

angiosperm trees over evergreen conifers in productive

habitats. New Phytologist 160:329–336.

Maddison, W. P., and D. R. V. Maddison. 2008. Mesquite: a

modular system for evolutionary analysis. http://mesquiteproject.

org

Martin, R. E. 1995. Cyclic and secular variation in microfossil

biomineralization: clues to the biogeochemical evolution of

Phanerozoic oceans. Global and Planetary Change 11:1–23.

McElwain, J. C., K. J. Willis, and R. Lupia. 2005. Cretaceous CO2

decline and the radiation and diversification of angiosperms.

Pp. 133–165 in J. R. Ehleringer, T. E. Cerling, and M. D. Dearing,

eds. A history of atmospheric CO2 and its effect on plants,

animals, and ecosystems. Springer, New York.

McKown, A. D., H. Cochard, and L. Sack. 2010. Decoding leaf

hydraulics with a spatially explicit model: principles of

venation architecture and implication for its evolution. Amer-

ican Naturalist 175:447–460.

Merigliano, M. F. 2005. Cottonwood understory zonation and its

relation to flood plain stratigraphy. Wetlands 25:356–374.

Moore, M. J., C. D. Bell, P. S. Soltis, and D. E. Soltis. 2007. Using

plastid genome-scale data to resolve enigmatic relationships

among basal angiosperms. Proceedings of the National

Academy of Sciences USA 49:19363–19368.

Osborne, C. P., D. J. Beerling, D. H. Lomax, and W. G. Chaloner.

2004. Biophysical constraints on the origin of leaves inferred

from the fossil record. Proceedings of the National Academy of

Sciences USA 101:10360–10362.

Parlange, J. Y., and P. E. Waggoner. 1970. Stomatal dimensions

and resistance to diffusion. Plant Physiology 46:337–348.

Puhakka, M., R. Kalliola, M. Rajasilta, and J. Salo. 1992. River

types, site evolution, and successional vegetation patterns in

Peruvian Amazonia. Journal of Biogeography 19:651–665.

Retallack, G. J., and D. L. Dilcher. 1981. A coastal hypothesis for

the dispersal and rise to dominance of flower plants. Pp. 27–77

in K. J. Niklas, ed. Paleobotany, paleoecology, and evolution.

Praeger, New York.

———. 1986. Cretaceous angiosperm invasion of North America.

Cretaceous Research 7:227–252.

Richardi-Branco, F., F. C. Branco, R. J. F. Garcia, R. S. Faria, S. Y.

Pereira, R. Portugal, L. C. Pressenda, and P. R. B. Pereira. 2009.

Plant accumulations along the Itanhaem River Basin, southern

coast of Sao Paulo State, Brazil. Palaios 24:416–424.

Robertson, K. M., and C. K. Augspurger. 1999. Geomorphic

processes and spatial patterns of primary forest succession on

the Bogue Chitto River, United States of America. Journal of

Ecology 87:1052–1063.

Robinson, J. M. 1994. Speculations on carbon dioxide starvation,

Late Tertiary evolution of stomatal regulation and floristic

modernization. Plant, Cell and Environment 17:1–10.

Rood, S. B., J. H. Braatne, and F. M. R. Hughes. 2003.

Ecophysiology of riparian cottonwoods: stream flow depen-

dency, water relations, and restoration. Tree Physiology

23:1113–1124.

Royer, D. L. 2010. Fossils constrain ancient climate sensitivity.

Proceedings of the National Academy of Sciences USA 107:517–

518.

Royer, D. L., L. J. Hickey, and S. Wing. 2003. Ecological

conservatism in the ‘living fossil’ Ginkgo. Paleobiology 29:84–

104.

Royer, D. L., I. M. Miller, D. J. Peppe, and L. J. Hickey. 2010. Leaf

economic traits from fossils support a weedy habit for early

angiosperms. American Journal of Botany 97:1–8.

Saarela, J. M., H. S. Rai, J. A. Doyle, P. K. Endress, S. Mathews, A.

Marchant, B. Briggs, and S. W. Graham. 2007. A new branch

emerges near the root of angiosperm phylogeny. Nature

446:312–315.

Sack, L., and N. M. Holbrook. 2006. Leaf hydraulics. Annual

Review of Plant Biology 57:361–381.

Schonherr, J, and M. J. Bukovac. 1972. Dependence on surface

tension, wettability, and stomatal morphology. Plant Physiol-

ogy 49:813–819.

Sperry, J. S. 2003. Evolution of water transport and xylem

structure. International Journal of Plant Sciences 164:S115–S127.

Sperry, J. S., U. G. Hacke, T. S. Feild, Y. Sano, and E. H. Sikkema.

2007. Hydraulic consequences of vessel evolution. International

Journal of Plant Sciences 168:1127–1139.

Stebbins, G. L. 1974. Flowering plants: evolution above the species

level. Harvard University Press, Cambridge.

Sun, G., and D. L. Dilcher. 2002. Early angiosperms from the

Lower Cretaceous of Jixi, eastern Heilongjiang, China. Review

of Palaeobotany and Palynology 121:91–112.

Taylor, D. W., and L. J. Hickey. 1996. Evidence for and

implications of an herbaceous origin for angiosperms. Pp.

232–266 in D. W. Taylor and L. J. Hickey, eds. Flowering plant

origin, evolution, and phylogeny. Chapman and Hall, New

York.

Taylor, L. L., J. R. Leake, J. Quirk, K. Hardy, S. A. Banwart, and D.

J. Beerling. 2009. Biological weathering and the long-term


carbon cycle: integrating mycorrhizal evolution and function

into the current paradigm. Geobiology 7:171–191.

Todzia, C. A., and R. C. Keating. 1991. Leaf architecture of the

Chloranthaceae. Annals of the Missouri Botanical Garden

78:476–496.

Ufnar, D. F., G. A. Ludvigson, L. A. Gonzalez, and D. R. Grocke.

2008. Precipitation rates and atmospheric heat transport during

the Cenomanian greenhouse warming in North America:

Estimates from a stable isotope mass-balance model. Palaeo-

geography, Palaeoclimatology, Palaeoecology 266:28–38.

Upchurch, G. R. 1984a. Cuticular anatomy of angiosperm leaves

from the Lower Cretaceous Potomac Group. Zone I leaves.

American Journal of Botany 71:192–2002.

———. 1984b. Cuticle evolution in Early Cretaceous angiosperms

from the Potomac Group of Virginia and Maryland. Annals of

the Missouri Botanical Garden 71:522–550.

——— . 1995. Dispersed angiosperm cuticles: their history,

preparation, and application to the rise of angiosperms in

Cretaceous and Paleocene coals, south western interior of

North America. International Journal of Coal Geology 28:161–

227.

Upchurch, G. R., and D. L. Dilcher. 1990. Cenomanian angio-

sperm leaf megafossils, Dakota Formation, Rose Creek Locality,

Jefferson County, Southeastern Nebraska. U.S. Geological

Survey Bulletin 1915.

Upchurch, G. R., and J. A. Wolfe. 1987. Mid-Cretaceous to Early

Tertiary vegetation and climate: evidence from fossil leaves and

woods. Pp. 75–105 in E. M. Friis, W. G. Chaloner, and P. R.

Crane, eds. The origins of angiosperms and their biological

consequences. Cambridge University Press, Cambridge.

Van Gardingen, P. R., C. E. Jeffree, and J. Grace. 1989. Variation in

stomatal aperture in leaves of Avena fatua L. observed by low

temperature scanning electron microscopy. Plant, Cell, and

Environment 12:887–897.

Vermeij, G. J. 1999. Inequality and the directionality of history.

American Naturalist 153:243–253.

Volk, T. J. 1989. Rise of angiosperms as a factor in long-term

climatic cooling. Geology 17:107–110.

Wang, H., and D. L. Dilcher. 2006. Early Cretaceous angiosperm

leaves from the Dakota Formation, Braun Ranch locality,

Kansas, United States of America. Palaeontographica, Abtei-

lung B 273:101–137.

——— . 2009. Late Cretaceous angiosperm leaves from the

Courtland Clay Pit, Minnesota. Palaeontographica, Abteilung

B 281:143–177.

White, T., L. Gonzalez, G. Ludvigson, and C. Poulson. 2001.

Middle Cretaceous greenhouse hydrological cycle of North

America. Geology 29:363–366.

Williams, J. H. 2008. Novelties of the flowering plant pollen tube

underlie diversification of a key life history stage. Proceedings

of the National Academy of Sciences USA 105:11259–11263.

Wing, S. L., and L. D. Boucher. 1998. Ecological aspects of the

Cretaceous flowering plant radiation. Annual Review of Earth

and Planetary Sciences 26:379–421.

Wullschleger, S. D. 1993. Biochemical limitations to carbon

assimilation in C3—a retrospective analysis of the A/Ci curves

from 109 species. Journal of Experimental Botany 44:907–920.

Wullschleger, S. D., T. J. Tschaplinski, and R. J. Norby. 2002. Plant

water relations at elevated CO2 - implications for water-limited

environments. Plant, Cell and Environment 25:319–331.


Documents

Cretaceous angiosperm leaves Fossil evidence for low gas ...Fossil evidence for low gas exchange capacities for Early Cretaceous angiosperm leaves Taylor S. Feild, Garland R. Upchurch