The evolution and functional significance of leaf shapein the angiosperms

Adrienne B NicotraAH Andrea LeighB C Kevin BoyceC Cynthia S JonesD Karl J Niklas EDana L RoyerF and Hirokazu TsukayaG

AResearch School of Biology The Australian National University Canberra ACT 0200 AustraliaBSchool of the Environment University of Technology Sydney PO Box 123 Broadway NSW 2007 AustraliaCDepartment of the Geophysical Sciences 5734 S Ellis Avenue Chicago IL 60637 USADDepartment of Ecology and Evolutionary Biology University of Connecticut 75 N Eagleville RoadUnit-3043 Storrs CT 06269 USA

EDepartment of Plant Biology Cornell University 412 Mann Library Building Cornell University IthacaNY 14853 USA

FDepartment of Earth and Environmental Sciences Wesleyan University 265 Church Street MiddletownCT 06459 USA

GGraduate School of Science University of Tokyo Science Build 2 7-3-1 Hongo Tokyo 113-0033 JapanHCorresponding author Email adriennenicotraanueduau

This paper is part of an ongoing series lsquoThe Evolution of Plant Functionsrsquo

Abstract Angiosperm leavesmanifest a remarkable diversity of shapes that range fromdevelopmental sequenceswithin ashoot and within crown response to microenvironment to variation among species within and between communities andamong orders or families It is generally assumed that because photosynthetic leaves are critical to plant growth and survivalvariation in their shape reflects natural selection operating on function Several non-mutually exclusive theories have beenproposed to explain leaf shape diversity These include thermoregulation of leaves especially in arid and hot environmentshydraulic constraints patterns of leaf expansion in deciduous species biomechanical constraints adaptations to avoidherbivory adaptations to optimise light interception and even that leaf shape variation is a response to selection on flowerform However the relative importance or likelihood of each of these factors is unclear Here we review the evolutionarycontext of leaf shape diversification discuss the proximal mechanisms that generate the diversity in extant systems andconsider the evidence for each the above hypotheses in the context of the functional significance of leaf shape The synthesisof these broad ranging areas helps to identify points of conceptual convergence for ongoing discussion and integrateddirections for future research

Additional keywords compound leaf leaf dissection leaf margin leaf size leaves lobbing

Introduction

A casual look at any local flora reveals great diversity inangiosperm leaf shape lsquoShapersquo describes a dimensionlessdescriptor for example lengthwidth which quantifies form interms of the natural dimensions or size of any object Thefunctional significance of shape variation among leaves hasbeen the subject of debate for many years and there are arange of different approaches to describing leaf shape(eg Nicotra 2010) The diversity of shape suggests that thereis no one ecological strategy that is dependent exclusively onleaf shape Even within a single genus leaf shape variationcan be tremendous (Fig 1) Here we provide an overview ofthe evolutionary history of leaf shape and then examine geneticdevelopmental physiological and ecological determinants of

leaf shape to better understand the evolutionary drivers ofdiversification in leaf shape and the functional significancetherein

Research over the past decade has produced sound ecologicalexplanations for the significance of some key non-shape leaftraits The leaf economic spectrumdescribes a continuum rangingfrom leaves with low to high mass per unit area (LMA) Leaveswith high LMA represent a high investment in structure andare typically long-lived with lower photosynthetic rates (Wrightet al 2004) Leaf size varies in a similar way small leaves areassociated with harsh conditions such as cold (Gates 1980) hot(Smith and Nobel 1977) dry (Thoday 1931 Raunkiaer 1934Gates et al 1968 Parkhurst and Loucks 1972 Specht and Specht1999 Fonseca et al 2000 McDonald et al 2003) high light

CSIRO PUBLISHING Review

wwwpublishcsiroaujournalsfpb Functional Plant Biology 2011 38 535ndash552

CSIRO 2011 101071FP11057 1445-440811070535

(Smith and Nobel 1977 Bragg and Westoby 2002) exposed(Ackerly et al 2002) nutrient poor (Beadle 1966 Cunninghamet al 1999 Fonseca et al 2000) and saline environments (Ballet al 1988) or many of these factors in combination (egMcDonald et al 2003) Leaf shape in contrast has beenshown to vary less predictably across environments or biomesthan size or LMA (McDonald et al 2003) Thus the functionalsignificance of leaf shape and the evolutionary drivers of itsdiversity remain the subject of discussion

The theories about leaf shape are many and not mutuallyexclusive thermoregulation of leaves especially in arid and hotenvironments hydraulic constraints patterns of leaf expansionin deciduous species mechanical constraints adaptations toavoid herbivory adaptations to optimise light interceptionand given that leaves are hypothesised to be developmentalhomologues of floral organs and it has even been suggestedthat leaf shape reflects the effects of selection on flower formFinally there is the chance that leaf shape variation has littlefunctional or adaptive significance and instead reflects randomvariation within the context of phylogenetic history Howevergiven the importance of the leaf we believe the latter option ratherunlikely To distinguish among the formerwe here synthesise the

varied fields involved so that we may assess the relativeimportance of each and identify the ecological situations inwhich one or the other of these mechanisms is likely to besignificant

In this reviewwefirst examine leaf shape froman evolutionaryperspective and in the process demonstrate that shape is highlylabile and widely explored in evolutionary history We bring thefocus onto the angiosperms in particular and then consider theproximate determinants of angiosperm leaf shape The currentunderstandingof thegenetic signals anddevelopmental processesunderlying shape differences are then reviewed to explore thegenetic controls and constraints on leaf shape evolution Thehypotheses about the function of leaf shape are next reviewedwith an evolutionary and genetic perspective in mind Wehighlight those functional issues that we see as particularlyrelevant to leaf shape evolution and then turn to threeexamples of leaf shape variation across communities withinlineages and within individuals to explore how an evolutionaryperspective alters understanding of the relationship between leafform and function We conclude with suggestions on how thisperspective can be used to direct future research on the functionand evolution of leaf shape

(a)

(b)

(e)

(f ) (g)

0 10 20 30 40 50 mm

(h) (i ) ( j )

(c) (d )

Fig 1 Leaf shape in the genus Pelargonium ranges from single and entire to compound and highly dissected Species are (a) P bowkeri (b) P reniforme(c) P klinghardtense (d) P fulgidum (e) P carnosum ( f ) P cucullatum (g) P abrotanifolium (h) P citronellum (i) P australe and ( j) P alternansPhotographs courtesy of Stuart Hay ANU Photography

536 Functional Plant Biology A B Nicotra et al

A brief history of the angiosperm leaf and shapediversity thereinTo understand the evolution of angiosperm leaf shape someperspective on the evolution of the leaf itself is needed as leavesof many shapes have evolved numerous times independentlyin evolutionary history For the purpose of this review ofangiosperm leaf shape we define a leaf as a vascularasymmetric appendicular structure initiated at the shoot apicalmeristem This definition excludes the functionally analogousstructures of mosses and leafy liverworts Morphologicallysimple leaves microphylls arose in the lycopsids Largerleaves with more complex shapes and venation networkscalled megaphylls arose in the seed plants in the extinctprogymnosperm and sphenophyll lineages and in one ormore lineages of lsquofernsrsquo (Galtier 1981 Kenrick and Crane1997 Niklas 1997 Boyce and Knoll 2002 Tomescu 2009Boyce 2010) Thus the seed plants represent just one ofseveral independent evolutionary origins of large leaves withdiverse shapes among the vascular plants

Despite the many independent origins of leaves several traitsnow associatedwith leaves can be considered homologous acrossall vascular plants through a shared monoplyletic origin from aleafless common ancestor First all vascular plant leaves aredistinct from the leaf-like organs of bryophytes that rely onexternal surfaces for photosynthetic gas exchange (as wouldalso be true of the external appendages independently derivedin many algal lineages Niklas 2000) Vascular plant leaves useinternal airspaces regulated by stomata for gas exchange (Raven1996 Boyce 2008a) Second evidence from both living plantsand the venation patterns of fossil leaves indicates that theancestral form of tissue production in the growing leaves of allvascular plants was limited to discrete marginal zones of growth(Pray 1960 Zurakowski and Gifford 1988 Boyce and Knoll2002 Boyce 2007) Other aspects of leaf organography suchas the evolution of differentiated abaxialadaxial domains (seesection on lsquoProximal mechanisms underlying leaf shapediversityrsquo below) and determinate leaf growth appear to ariseindependently in seed plants and other lineages however severalof these lineages have co-opted the same underlying geneticpathways to regulate these developmental processes (seesection on lsquoProximal mechanisms underlying leaf shapediversityrsquo Bharathan et al 2002 Harrison et al 2005 Sanderset al 2007 Tomescu 2009 Boyce 2010)

Within this broader evolutionary context angiosperm leavesrepresent a rangeof particularly aberrant variants of vascular plantleaves Some of the distinctive characteristics of angiospermleaves have arisen independently in other lineages whereasother traits are unique to the flowering plants Angiospermshave evolved leaf growth that occurs diffusely throughout theleaf without being limited to the margin (Pray 1955 Poethig andSussex 1985a 1985b) thus enabling the enormous diversity andcomplexity of angiosperm leaf venation patterns and shapes Thephylogenetic distribution of complex venation patterns suggestthat the evolutionary departures from marginal growth observedin the angiosperms have also arisen repeatedly in two or threeother seed plant lineages and 10 or more groups of ferns (Boyce2005) In these likewise diffuse growth and complex venationpatterns can be seen to occur with diverse leaf shapes Despitethese similarities to the other seed plants angiosperm leaf

evolution remains distinctive in that it ran counter to thegeneral temporal trend seen in the vasculature of other seedplants The vascular network of other seed plant lineagesevolved towards simplification After initially exhibiting thecomplete range of morphologies consistent with marginalgrowth including many involving two vein orders with amidvein and open or reticulate networks of secondary veins(morphologies now common only among the ferns) the leafarchitecture in the other major vascular plant lineages becameprogressively limited to a narrow subset of forms In cycadsconifers Ginkgo and several other extinct seed plant lineages asingle order of veins lead straight parallel courses to a distalmarginAngiosperms however exhibitmany hierarchical ordersof reticulate internally directed veins (Boyce 2005) Furtherangiosperm leaves are unique in their possession ofextraordinarily high vein densities (Fig 2) whereas all otherplants living or extinct average ~2mm of vein length per squaremm of leaf surface angiosperms average between 8 and10mmmmndash2 and can range to above 20mmmmndash2 (Boyce2008a 2008b 2009 Brodribb and Feild 2010 Brodribb et al2010)

Together these changes that accompanied angiospermevolution represent a substantial revision of the functionalpossibilities of a leaf First the release of venation patterns fromstrictly marginal growth allows the production of novel leafshapes and sizes that would not otherwise be possible Seconddifferences in angiosperm venation allowed a complete revolutionin how water is distributed within a leaf (Boyce 2008a) In anyleaf the tissue furthest from the leaf base has access only to thatwater which has not been lost to transpiration in more proximaltissues (Zwieniecki et al 2004a 2006) Reticulate venation andthe minute size of the final order of veins in flowering plantsleads to equitable distribution of water between successive veinorders (Zwieniecki et al 2002) Finally the high vein densities ofangiosperms shorten the path length along xylem and mesophyllthus enabling assimilation and transpiration capacities higherthan in any other group of plants (Bond 1989 Sack and Frole2006 Brodribb et al 2007 Boyce et al 2009) The uniquely hightranspiration capacities of angiosperms may also have influencedthe evolution of leaf shape by relaxing some of the thermalconstraints on leaf size and shape (see lsquoTemperature and waterrsquosection)

It remains controversial when and in which lineages thesecharacteristics of angiosperm leaves arose The distinctivenessof angiosperms leaves has led to polarised interpretations oftheir ancestry Some researcher propose that similarities inleaf venation tie the early angiosperms to particular fossil seedplants (eg Melville 1969) but the group of plants with themost similar venation are the clearly unrelated dipterid fernsThe various seed plant lineages most often favoured asangiosperm relatives based upon reproductive characteristics(Doyle and Hickey 1976 Hilton and Bateman 2006 Frohlichand Chase 2007) tend to have very dissimilar leaves Otherssuggest that angiosperm leaves are so distinct as to require acomplete reinvention of a leaf after passing through a leaflessintermediate such as one that was aquatic- or desert-adapted(Doyle and Hickey 1976) But the appearance of angiosperm-like leaf traits in a variety of fern lineages suggest thatinvoking a leafless intermediate is unnecessary Although the

The evolution of leaf shape Functional Plant Biology 537

history of early angiosperm leaf evolution is still debated weare certain that there have been numerous independent originsof diverse leaf shapes and that leaf shape diversificationis associated with both high vein densities and reticulatevenation patterns

Proximal mechanisms underlying leaf shape diversity

What are the genetic mechanisms underlying this leaf shapediversification and how similar are they given the many

independent origins of diverse leaf shapes What we know ofthe processes of leaf development (and leaf shape determination)is largely a product of studies on model species systems Inparticular Arabidopsis thaliana L (Heynh) (hereafter referredto as Arabidopsis) has been widely used for the identificationof genes determining leaf form In the following sections weexamine the development of the angiosperm leaf specificallythe establishment of dorsiventral domains and flatteningexpansion of the lamina and formation of leaf margin anddiscuss how these factors interact to determine leaf shape

(a) (b)

(c)

Fig 2 Leaf venation architecture in (a) the cycadZamia furfuracea (b) the fernBlechnumgibbum and (c) theangiosperm Boehmeria nivea overlain with the fern Polypodium formosanum Both the Zamia and theBlechnum have marginally ending veins suggesting marginal growth but the single order of parallel veins inZamiamust distribute water over more than 10 cm of length whereas the finest veins in theBlechnum are a fewmm The finest veins distributing water in angiosperms can be even shorter and the density of veins muchhigher the Boehmeria has ~8mm of vein mmndash2 of leaf versus 1mmmmndash2 in the Polypodium

538 Functional Plant Biology A B Nicotra et al

Establishment of dorsiventral domainsand lamina flattening

Since leaf primordia arise as protrusions from shoot apicalmeristems (SAM) suppression of the regular SAMdevelopmental program is the first step required for normaldevelopment of a leaf primordium Recently Sarojam et al(2010) found that genes in the YABBY family suppress SAMidentity and promote lamina growth in Arabidopsis In mostbifacial leaves such as leaves of Arabidopsis and snapdragon(Antirrhinum majus L) the radial primordium differentiatesinto abaxial (lower) and adaxial (upper) surfaces to establishdorsiventrality (Waites and Hudson 1995) Our currentunderstanding is that the molecular genetic regulation ofidentification of the abaxial and adaxial domains depends onactions of both small RNAs and directindirect targets suchas class III Zip genes KANADI genes and AUXIN RESPONSEFACTORS3 (ARF3)ETT in Arabidopsis (Husbands et al 2009Floyd and Bowman 2010 Kidner 2010) Without establishmentof the adaxial and abaxial domains it is believed that leavescannot expand in any angiosperm species rather they becomestalk like since the so-called plate meristem is never activated toexpand the laminas (Fig 3)

The monocot clade is typified by many unifacial leaves thathave only abaxial identity in the lamina as seen in leek (Alliumampeloprasum var porrum L) and Iris spp Leek leaves arecylindrical since they lack adaxial identity in their leaf bladeIris in contrast has flat leaves that expand in a distinct

perpendicular direction relative to that of normal bifacialleaves Likewise the rush Juncus prismatocarpus RBr hasunifacial flat leaves the primordium of the leaf blade growsinto a flat lamina as a result of extensive thickening growthThis developmental pathway is dependent on the geneDROOPING LEAF (DL Yamaguchi et al 2010 Fig 3) thatis also a member of the YABBY gene family but with a monocot-specific function As seen in this case diversified morphology ina particular taxon sometimes depends on some taxon-specificmolecular mechanisms that are modifications of gene functionfrom ancestral lineages

Mechanisms for expansion of lamina

Once dorsiventral identity is generated the leaf lamina can thendiversify in morphology to form simple or compound leaveswith entire or serrated margins differing in thickness length andwidth and in orientation (Tsukaya 2006) The leaf length widthratio is regulated by polar-dependent cell expansion and cellproliferationdistribution Several key genes for its regulationhave been identified from Arabidopsis ANGUSTIFOLIA (AN)and ROTUNDIFOLIA3 (ROT3) regulate the shape of cellswhereas ANGUSTIFOLIA3 (AN3) and ROTUNDIFOLIA4(ROT4) affect the number of cells in the lamina (Tsukaya2006) Loss-of-function mutations of AN and AN3 result innarrower leaves whereas loss-of-function of ROT3 oroverexpression of ROT4 make leaves shorter In addition

Fig 3 Twodifferentmechanismsof laminagrowth Inmost leaves such as leaves ofArabidopsis (upper left)dorsiventral identity (adaxial upper side (shown in green in colour version online) and abaxial lower side(shown in blue in colour version)) identities are established Flattening lamina growth occurs along the junctionbetween adaxial and abaxial domains as a result of activity of the plate meristem Loss of either abaxial oradaxial identity results in radialised growth (lower right) Stick-like leaf morphology seen in leek and Juncuswallichianus (centre) is the typical example of unifacial leaves that have no adaxial domain In the monocotclade however some species develop flat lamina irrespective of lack of adaxial identity in the lamina (rightSisyrinchium rosulatum) that is supported by enhanced growth in the leaf-thickness direction Meristematicactivity occurs in center region of both leaves (orange in colour version) relative position of the shoot apicalmeristem (SAM) is shown by the cross-hatched dot (green in colour version)

The evolution of leaf shape Functional Plant Biology 539

LONGIFOLIA1 (LON1) and LON2 regulate the length of the leaflamina in Arabidopsis (Lee et al 2006) A vast array of genes isknown to influence leaf area (Horiguchi et al 2006) For exampleheteroblastic change in leaf area is governed by the miR156-SQUAMOSAPROMOTERBINDINGPROTEIN-LIKE (SPL)-miR172 pathway via regulation of cell size and cell numbers inArabidopsis (Usami et al 2009)Alterations of activities of anyofthese factors could contribute to natural variation in leaflength width ratio and area None of these genetic variationsresult in loss of reproductive structures suggesting that thesegenes could be under direct environmentally mediated selectionAs yet we do not know whether the above patterns hold in othernon-model species

Leaves also vary in the extent to which the lamina is flator curved CINCINATA (CIN) of snapdragon was oncethought to function to keep the leaf flat by orchestrating cellproliferation along the medio-lateral axis of the leaf primordia(Nath et al 2003) More recently however analyses of theTCP genes which are homologues of CIN in Arabidopsisindicate that the major role of the TCP is in promotion ofmaturation in laminar tissues not in regulation of cellproliferation (Efroni et al 2008 Sarojam et al 2010) The roleof CIN in snapdragon may also be the same as that of TCPsin Arabidopsis The other candidate genetic mechanism forcontrol of leaf curvature is an auxin-related pathway Severalhyponastic leaf mutants have been isolated from Arabidopsisand many of the responsible genes are related to auxinsignalling (Perez-Perez et al 2002 2009) Since auxinsignalling is also very important for venation patterningspecies-specific modification of auxin signalling in leafprimordia may have caused diversification of both leafcurvature and vasculature patterning

Genetic regulation for patterning of leaf marginserration and leaflet formation

Variation in leaf shape as distinct from length and area orcurvature arises from variation in leaf margin and lobeformation Kawamura et al (2010) showed that spatialpatterning of serrations or leaf teeth (see lsquoLeaf shape variationwithin lineagesrsquo section) on the leaf margin depends onformation of auxin maxima on the leaf margin in ArabidopsisLocation of these maxima is governed by polar auxin transportand the stabilisation of auxinmaxima ismaintained by prolongedexpression of CUP-SHAPED COTYLEDON2 (CUC2) around it(Kawamura et al 2010) The mechanism of rhythmic patternformation of the auxin maximaminima on the leaf margin isthought to involve self-organising pattern formationmechanismsby polar auxin transport similar to that involved in patternformation of phyllotaxy (de Reuille et al 2006 Jonsson et al2006) Either the loss of distinct auxinmaxima or an instability ofthe auxin maxima will result in an entire leaf margin (Kawamuraet al 2010) Changes in functions of these genes howeverseverely influence many biological processes including floralorgan development as seen in the pin-formed 1 mutant (Okadaet al 1991) Thus variation among species in pattern of marginserration might accompany variation in floral organs or morelikely reflect differential regulation for example by downstreamgenetic pathways

Combined control of auxin-maxima formation by PIN andCUC genes as described for serration above is also known tohave an important role in leaflet patterning in bittercress(Cardamine flexuosa With) an Arabidopsis relative withcompound leaves (Barkoulas et al 2008) Again this controlvaries among species in tomato (Solanum lycopersicum L) lossof CUC3 function does not convert compound leaves intosimple leaves (Blein et al 2008) Thus it is clear that theindependent evolution of compound leaves from simple leavesis likely driven by species-specific mechanisms (Efroni et al2010) though these may reflect modifications of similarpathways

Recent studies have identified several important key genes forregulation of compound versus simple leaf forms For exampleabsence or prolonged expression of class I Knotted1-likehomeobox (KNOX) genes in leaf primordia is important tocompound leaf formation in a wide range of angiosperms andgymnosperms (Bharathan et al 2002) In some legume specieshowever the LEAFY (LFY)UNIFOLIATA (UNI) gene is the keyfor compound leaf formation instead of the class I KNOX genes(Hofer et al 1997) Changes in these genes could be linked tothe evolution of compound leaves although loss-of-function ofLFYUNI also causes infertility in Arabidopsis and thus it isunlikely that selection directly on these particular genes couldresult in theevolutionof variation in leaf shapeDiversified spatialpatterning of petiolepetiolule and laminaleaflet in compoundleaves have also been suggested to be linked to the distributionpattern of AS1-ROUGH SHEATH2-PHANTASTICA (ARP)expression domains (Kim et al 2003) However based on thefact that the ARP gene has only a limited role in compound leafformation in some species it is unclear whetherARPwill turn outto be major factor in others (Efroni et al 2010)

As described above the independent evolutionary origins ofseveral angiosperm leaf features should caution against theassumption that a single ancestral genetic system underlies leafdevelopment The developmental genetic approaches outlinedhave revealed several key mechanisms for leaf shape controlbut have also demonstrated variation in the function of thesegenes among lineages In a few studies researchers have gonebeyond one or two species studies and have looked for deeperevolutionary links (Illing et al 2009) Studies of the role of theclass 1 KNOX genes have considered not just angiospermsbut other lineages Among simple-leafed dicots and monocots(eg Arabidopsis and Zea) KNOX genes are strongly expressedduring the indeterminate growth of the apical meristem but arerapidly downregulated in the cells flanking the apical domein regions defined by leaf primordia Although some arginine-rich protein (ARP) genes repress the expression of KNOXtranscription factors during the development of both micro-and megaphyll primordia (Harrison et al 2005) the role ofhomeodomain-leucine zipper (HD-ZIP) III genes which areinvolved in the specification of foliar adaxialabaxial fatediffers in the development of these two leaf types (Floyd andBowman 2006) HD-ZIP III genes have been identified in themoss Physcomitrella (Sakakibara et al 2001) and in the fernCeratopteris (Aso et al 1999) they are also expressed duringnormal vascular tissue development in Arabidopsis (Zhong andYe 1999) suggesting that they are very ancient but alsomultifunctional

540 Functional Plant Biology A B Nicotra et al

Accumulating knowledge on molecular genetic mechanismsof leaf morphogenesis is leading to an increasingly evo-devoapproach to studies of leaf diversity (Tsukaya 2010)this progress is aided by establishing new model species(eg Cardamine hirsuta and the rush Canales et al 2010Yamaguchi and Tsukaya 2010) Over the next decade weanticipate that genomic approaches on multiple speciesanalysed in phylogenetic frameworks (eg Illing et al 2009)will resolvemuchof theenigma regarding the evolutionofdiverseleaf forms

Genetic constraints on flower petal and leaf form

One additional developmental consideration is that as famouslydescribed by von Goethe in 1790 flower petals are derivatives ofthe apical meristem and serial homologues to leaves (von Goethe2009) Thus one might hypothesise that although the evolutionof leaves substantially predates that of flowers selection on theform of either of these organs would also result in changes inthe other such that selection on leaf shape and size subsequent tothe evolution of floral organs may be constrained by pleiotropiccontrol of floral form Recent genetic data identifies several genesthat have roles in both leaf and flower form (Dinneny et al 2004Schmid et al 2005 Street et al 2008) For example geneticchanges in both leaf length to width ratio (eg rot3) and leaf size(eg an3) influence not only leaf proportions but also that of thefloral organs in Arabidopsis (Kim et al 1999 Horiguchi et al2005) The garden pea (Pisum sativum L) is a good case in pointas discussed above compound leaf formation depends on LFYUNI but the same gene is also required for flowering Thus ifcurrent selection favoured simple leaves in garden pea changes inleaf shape could be constrained by pleiotropic control offlowering (Hofer et al 1997) Likewise Shalit et al (2009)recently reported that the action of florigen ndash a plant hormonefor flowering ndash influences complexity of leaves by changingnumbers of leaflets in tomato Such relationships suggest thatsome floral characters as well as other metabolic pathways maynowbeclosely linkedwith regulationof leaf shapeand sizeA fewecological studies have also assessed correlations between petaland leaf form (Berg 1960 Armbruster et al 1999) This questionis of particular interest in lineages where animal pollination isassociated with increased diversification rates either as a result ofdirect selection or as a secondary reinforcing factor (van der Nietet al 2006 Armbruster and Muchhala 2009 Kay and Sargent2009)

In situations where leaf and petal form are under similargenetic control and both under selection one of two thingscould happen If leaf shape does not have substantialfunctional impact then one might expect leaf form to changewith selection on petal form However given the likelyimportance of leaf form to photosynthetic function it is morelikely that a breakdown in genetic correlations between leafand flower form (eg by alterations of downstream regulatorypathways) would evolve Several factors argue against thehypothesis that leaf shape is frequently a product of selectionon flower form First diverse leaf shapes were in evidencelong before the origin of flowers in the angiosperms Secondthe genetic mechanisms underlying leaf shape revealconsiderable evolvability with similar genes being co-opted

for different functions in different species Finally given thepotential importance of both petal and leaf shape we suggestthat genetic constraints on the independent evolution of bothstructures are likely to have broken down in those lineages withhighly diversified leaf shapes over the many millions of yearssince petals first evolved

Ecological correlates and the functional significanceof leaf shape

Evolutionary history plays a part in determining leafmorphology but our consideration of leaf shape evolution andits developmental genetic basis demonstrates that diverselineages have repeatedly gained and lost many leaf featuresThus developmental and genetic constraints on the evolutionof leaf form and shape in particular have been overcomemany times or have led to new evolutionary opportunities Itis of particular interest then to ask what are the likely ecologicaldrivers andor functional significance of leaf shape

Temperature and water

Conventionalwisdomwould say that thermoregulation is perhapsthe main driver of leaf shape evolution The maintenance ofleaf temperatures within certain limits is critical for a plantrsquosgrowth and survival and although photosynthetic tissue of somespecies canwithstand temperatures below6C (Ball et al 2006)and above 60C (Clum1926Nobel 1988) irreversible damage toleaves can occur at temperatures well within this range (Levitt1980 Jones 1992 Ball et al 2004 Groom et al 2004 Sharkey2005) Leaf size and shape potentially have large effects onleaf temperature because the two-dimensional proportions of aflattened leaf determine the rate of heat transfer between acrossthe leafndashair interface by influencing the thickness of its boundarylayer

All else being equal the thickness of a boundary layerincreases with length from the windward edge so that heatconvection from small leaves is more rapid than from largeleaves (Raschke 1960 Gates 1968 Vogel 1970 Parkhurst andLoucks 1972Grace et al 1980Geller andSmith 1982Monteithand Unsworth 1990 Schuepp 1993) Likewise because leaflobing reduces the distance across the lamina the rate of heattransfer is predicted to be greater in a lobed leaf than an unlobedleaf with equivalent area (Parkhurst et al 1968 Vogel 1968Lewis 1972 Givnish 1978 Gurevitch and Schuepp 1990a)

The rate of heat transfer from lobed metal plates or lobedleaves coated in metal is greater than from shallow-lobed or un-lobed ones (Parkhurst et al 1968 Thom 1968 Vogel 1970Gottschlich and Smith 1982 Gurevitch and Schuepp 1990bRoth-Nebelsick 2001) Artificial or coated leaves have beenused because they enable boundary layer resistance to beinvestigated in the absence of transpiration However thisapproach is biologically flawed because it does not enable anassessment of the relative importance of a range of traits (eg leafwater content absorbance) in determining the actual operatingtemperatures of leaves in the field

Actual leaf temperature measurements show that large leavescan operate at below-ambient temperatures even on hot days asa result of high rates of latent heat loss through transpiration(Drake et al 1970 Gates 1980 Hegazy and El Amry 1998)

The evolution of leaf shape Functional Plant Biology 541

Thus conditions of evaporative demand and the availability ofsoil water for transpiration are key to understanding leafshapendashtemperature relationships under biologically realisticconditions Because of their much greater transpirationalcapacities (see above) flowering plants have far greaterleeway than other plants regarding the size and shape of theirleaves (see lsquoA brief history of the angiosperm leaf and shapediversity thereinrsquo section Boyce et al 2009)

In addition to reducing heat loads the morphology of deeplylobed leaves may also reflect direct selection for increasedhydraulic efficiency Water moving through a plant encountersnumerous resistances on its path from soil to roots to stemsthrough leaves and out to the external atmosphere Within theliquid phase of water flow through a plant leaf hydraulicresistance Rleaf accounts for more than 30 of total plantresistance (Sack and Holbrook 2006) The high resistance inleaves is due first to a system of veins of decreasing size theresistance of which is inversely proportional to the fourth powerof the radius of the component conduits in these veins Thereforeas a proportion of the considerable resistance in the leaf(Zwieniecki et al 2002) the minor veins provide the greatestresistance to water flow (Sack et al 2004)Water also encountershigh resistance as it exits veins laterally and travels to othercells mainly via apoplastic pathways through the mesophyllAccordingly on the order of half of total Rleaf can occur outsidethe major veins (Sack and Holbrook 2006) As resistanceincreases along the pathway towards sites of evaporationlocalised water potential (Y) becomes progressively morenegative (Brodribb et al 2010) The result of how thisphenomenon might affect a leaf has long been observedregions of leaves lying furthest from the main supply channelsare the first to lsquowitherrsquo when exposed to strong winds (Fig 4Yapp 1912) Leaf lobing represents an effective removal of thispotentially stress-prone tissue and has been suggested as anadaptation to dry conditions (Thoday 1931 Givnish 1979) Ifdeeply lobed leaves have a lower ratio of mesophyll tissue tolarge highly conductive veins they should then have reducedresistance relative to less- or un-lobed leaves (Sack and Tyree2005)

Although leaf lobing might confer adaptive benefits withrespect to maintaining stable hydraulic supply it may also bethat the extent of lobing or margin dissection is itself determinedby hydraulic limitation (Sack et al 2003 Zwieniecki et al2004b Boyce 2009) For species with lsquosunrsquo and lsquoshadersquoleaves the morphology of small deeply lobed sun leaves atthe outer canopy is proposed to result from water pressuredrop across the leaf lamina caused by reduced water deliveryto expanding cells during growth (Fig 5 Zwieniecki et al 2004bBoyce 2009 Leigh et al 2011)Alternatively recent studies havesuggested that many of the characteristics of lsquosunrsquo and lsquoshadersquoleaves are the result of branch autonomy within the plant canopy(Niklas and Cobb 2010)

Thus although the effects of leaf shape on leaf thermalregulation hold when tested on model leaves predictions areless straightforward when hydraulic function is accounted forAdditionally consideration of the relative contribution of a rangeof leaf and branch properties to leaf temperature is critical It islikely that leaf shape is only one amongmany factors influencingleaf thermal regulation other factors could include water

content leaf thickness spectral reflectance orientation andplant architecture Further given the observation of high leafshape diversity withinmany plant communities from hot and aridenvironments it seems likely that regulating leaf temperature isnot the singlemost important evolutionary influence on leaf shapediversity The association between leaf shape and hydraulicproperties which is affected by leaf temperature in contrast islikely to be of greater importance

Leaf structure and functional trade-offs

Many foliar leaf traits including shape reflect functional trade-offs that have been resolved in various ways by differentspecies depending on their ecological settings (eg hydro- vsmesophytes) physiological attributes (eg C3 vsC4metabolism)developmental capabilities (eg heterobaric vs non-heterobaricleaves) or evolutionary histories (eg micro- vs megaphylls)Life history and optimisation theory show that the number ofphenotypic solutions that allow for different equally successfultrait combinations increases as the number of trade-offsincreases ndash a conclusion that applies to traits within the leaf(eg for shape) as well as to leafndashbranch relationships (Niklas1988)

It is evident that leaves perform several functionssimultaneously that is they intercept light transport liquidsand solutes exchange gases with the atmosphere dissipateheat cope with externally applied mechanical forces and

(a)

(e)(f )

( i )

(g)

(h)

(b)

(c)

(d )

Fig 4 Localised lsquowitheringrsquo of leaves exposed to strong wind Reprintedfrom Yapp (1912) with permission from Oxford University Press

542 Functional Plant Biology A B Nicotra et al

defend themselves from herbivores and pathogens Importantlymany of these functions have conflicting physiologicalanatomical or morphological requirements For exampleorientations of lamina that maximise light interception alsominimise the ability to dissipate heat Further all of thesefunctions are intrinsically size dependent For example thequantity of mechanical tissues required to support laminaedepends on the mass of laminae whereas the ability tointercept light depends on laminae area

When seen in this light the trade-offs required for thesuccessful functionality of foliar leaves generate rather thanconstrain ecological and evolutionary opportunities This canbe illustrated mathematically and empirically in the context ofphyllotaxy and its consequences on light interception Computersimulations of mathematically generated shoots to assess theinfluence of leaf shape size and phyllotactic patterns on theability to intercept direct solar radiation show that differences inphyllotaxy significantly influence light interception (particularlyfor shoots with a rosette growth habit) They also show thatcomparatively small differences in leaf shape can compensatefor the negative effects of leaf overlap resulting from virtuallyany phyllotactic pattern For example lobed leaves or pinnifid

compound leaves facilitate light penetration through shootsbearing densely packed leaves Thus phyllotaxy can beviewed as a developmental limiting factor that can drivecompensatory changes in morphological features such asshape that are not directly controlled by patterns of leafinitiation In this way the lsquoconstraintsrsquo imposed by phyllotaxycan foster evolutionary modifications at the level of the entireshoot (Niklas 1988)

One particularly important trade-off operating at the level ofthe architecture of a shoot is that between leaf size and numberFor any given leaf biomass allocation an individual shoot caneither have a few large leaves or many smaller ones Acrossdifferent herbaceous species and juvenile specimens of treespecies drawn from diverse lineages and ecological settingsthe relationship between total leaf biomass per plant (ML) andtotal stemmass per plant (MS) fallswithin a comparatively narrowcorridor of variation that obeys a positive isometric scalingrelationship (ie MLMS

~10 Niklas 2004) Accordinglyacross these species any stem mass investment yields a moreor less proportional increase in leaf mass Therefore it followsmathematically that across these species the relationshipamong the average mass of a typical leaf mL total leaf numberNL and total stem mass complies with the proportionality mL

NLMS~10 When individual shoots are sampled across

diverse species the relationship between mL and NL per shootis negative and isometric (ie mLNL

ndash1 Kleiman and Aarssen2007) a relationship that appears to be insensitive to howspecies are grouped according to other functional traits oraccording to their habitat preference (Yang et al 2008) Thatthis phenomenology reflects a biomass allocation trade-off isclear What remains uncertain is whether it is a simple trade-offbetween leaf mass investment and leaf size or a more complexset of relationships imposed by twig mass investments andhydraulics or biomechanics

Another trade-off that appears to hold across many but not allspecies is the relationship between leaf surface areaA and averageleaf mass mL For more than a few species-groupings thisrelationship is allometric with a slope that is less than onewhich indicates that increases in leaf mass investment do notresult in proportional increases in leaf area (ieAmL

lt10Niklaset al 2007) This trend of lsquodiminishing returnsrsquo has also beenreported when total leaf area per plant is plotted against total leafmass for plants differing in size (Niklas and Cobb 2008 2010)Several factorsmay contribute to these trends eg biomechanicalanalyses show that disproportionately larger investments must bemade to mechanically support photosynthetic tissues as leavesincrease in their surface area (Niklas et al 2009) In both the leafmass to stem mass and leaf mass to area relationships describedabove leaf shapevariation canprovide anextra degreeof freedomfor selection to act on at both the leaf and branch level

Case studies of leaf shape variation at different scales

Above we have discussed evolutionary and proximal drivers ofleaf shape as well as ecological correlates and functionalsignificance thereof Here we present case studies at threescales where we can examine the ecological and evolutionaryimplications of leaf shape These are the cross community scalecaseof leaf teeth as a subset of shapemodificationswithin lineage

200 bottomtop

150

100

50

0

0

08

07

06

05

04

03

5 10

Days after bud break

Leaf

siz

e (c

m2 )

Leaf

lobi

ng in

dex

(cm

2 cm

ndash2)

Days after bud break

15 20 25

0 5 10 15 20 25

Fig 5 Variation in (a) leaf size and (b) shape complexity inQuercus rubraleaves growing at the top and bottom of the tree crown Reprinted fromZwieniecki et al (2004b) with permission from John Wiley amp Sons

The evolution of leaf shape Functional Plant Biology 543

considerations of the genus Pelargonium and the familyProteaceae and finally the case of leaf shape variation within asingle genetic individual heteroblasty and heterophylly

Leaf teeth leaf shape variation at the community scale

Leaf teeth are conspicuous features inmanyplants and are notableas a leaf-shape trait that varies in predictable ways across plantcommunities Nearly 100 years ago Bailey and Sinnott (19151916) observed that plant communities from regions with lowmean annual temperature (MAT) have a higher site meanproportion of non-monocotyledonous species with toothedleaves Paleobotanists quickly seized upon this relationship asan index for inferring paleotemperature Little et al (2010)recently documented the rich interest in this topic compiling351 literature references related to leaf physiognomy andtemperature of which the primary focus is leaf teeth The linkbetween teeth and temperature is observed on all vegetatedcontinents (Greenwood 2005a Peppe Royer et al 2011) andis also seen in related metrics such as tooth size and tooth number(Huff et al 2003 Royer et al 2005 Peppe et al 2011) Furtherthese tooth variables can display phenotypic plasticity totemperature change (Fig 6 Royer et al 2009b)

What is knownabout the functional bases of these leafndashclimaterelationships Leaf teeth commonly expand and mature fasterthan the bulk leaf (eg Billings 1905 Feild et al 2005) and youngteeth are disproportionately vigorous in their gas exchange(Baker-Brosh and Peet 1997 Royer and Wilf 2006) Fromthese observations Royer and Wilf (2006) hypothesised thatone role of teeth is to increase sap flow thereby deliveringnutrients and other solutes to young emerging leaves Inhabitats with low temperatures teeth could play an importantrole in maximising the potential for carbon gain and growth earlyin the season In warmer climates with longer growing seasonsthe carbon benefit conferred by teeth would be outweighed bytheir unavoidable water cost Alternatively Feild et al (2005)hypothesised that teeth help remove freezendashthaw embolismsthrough positive root pressure and guttation Importantly thetransport of fluids is common to both hypotheses From this itmay be surmised that water-limiting conditions select againstteeth (Bailey andSinnott 1916) Indeed this pattern is clearly seenacross local water availability gradients (Burnham et al 2001

Kowalski and Dilcher 2003 Greenwood 2005b Royer et al2009a Peppe et al 2011) but not globally against mean annualprecipitation (eg Peppe et al 2011)

If the hypothesis by Royer andWilf (2006) is correct toothedspecies should have a higher photosynthetic rate at least at thebeginning of the growing season As such teethmay be related tothe leaf economics spectrum (Wright et al 2004) Some studiesfind support for this idea leaves with a short leaf lifespan(deciduous) (Peppe et al 2011) and low leaf mass per area(Royer et al 2005) two leaf economic variables are morelikely to be toothed It also follows that teeth should be morefunctionally related to growing season climatic variables likeseasonality growing season length and growing degree days notmean annual temperature however few studies find support forthis idea (Royer et al 2005 Peppe et al 2011) Ultimately itwould be valuable to have a better understanding of the geneticprocesses underlying tooth development but presently little isknown outside model species (see lsquoGenetic regulation forpatterning of leaf margin serration and leaflet formationrsquosection) Further phylogenetic history has typically beenassumed to have only a minor role in these leaf-climatepatterns but rigorous tests demonstrate that this assumption isfalse (Little et al 2010) This outcome is of particular concern forpaleobotanists because systematic placement of fossils can bedifficult especially for the Cretaceous and Paleogene Tools toaccommodate for this phylogenetic dependency are needed

Leaf shape variation within lineages

Proteaceae

Proteaceae is an ancient angiosperm family originating inGondwana over 100million year ago and possessing 1700 extantspecies (Weston 2007) The familyrsquos greatest representation is inAustralia followed by SouthAfrica with smaller numbers acrossother Gondwanan segments (Cowling and Lamont 1998 Hootand Douglas 1998 Weston 2007) Within the Proteaceae thereexists an enormous range of leaf sizes and shapes ranging froma few mm2 to 500 cm2 and including entire toothed broadlylobed deeply dissected needle-like simple or compound leavesIntriguingly most of this shape diversity is represented inAustralia The majority of taxa outside Australia possessentire leaves the exception being four South African genera

(a) (b) (c) (d )

Fig 6 Representative leaves of Acer rubrum showing genetic variation and plasticity in leaf teeth Leaf derived from coolclimate (Ontario) seed stock grown in (a) Rhode Island (cool climate) and (b) Florida (warm climate) leaf derived fromwarm climate (Florida) seed stock grown in (c) Rhode Island and (d) Florida Bar (for all leaves) = 1 cm Reprinted from Royeret al (2009) under Creative Commons Attribution Licence

544 Functional Plant Biology A B Nicotra et al

that possess dissected (Serruria and Paranomis) or toothed(Mimites and Leucospermum) leafed species all of whichoccur in a single clade (Sauquet et al 2009)

The reason for the extensive radiation in leaf shape diversityin Australian Proteaceae is unclear Among South AfricanProteaceae leaf size decreases with mean annual temperatureand narrow leaves have been shown to overheat less than theirwider counterparts (Yates et al 2010) Given that leaf dissectionand small size are functionally equivalent with respect to heatloss (see lsquoTemperature and waterrsquo section) it is tempting tosuggest that leaf shape variation in Australia represents anadaptive response to increasing hot dry conditions Yet thereappears to be no consistent distributional pattern of Proteaceaeleaf dimensions inAustraliaManymediumor large un-dissectedleaves exist in the hottest habitats Grevilleas with small simpleleaves can be found across most regions including high rainfallareas and some of the species found in the shaded rainforestunderstorey have among the most deeply dissected leaves of allLeaf shape also varies within a single individual For examplefern-leaf stenocarpus Stenocarpus davallioides Foreman ampBHyland has tripinnatisect juvenile leaves but adult leavescan be pinnate or bipinnate (Foreman and Hyland 1995)Other rainforest species such as Atherton Oak (Athertoniadiversifolia (CTWhit) LAS Johnson amp BGBriggs) havevery large lobed intermediate leaves but adult and seedlingleaves are small and entire (Weston 1995) Alternatively leaflobing within a plant can be random having no obvious spatialor developmental association or homogenous across a speciesExamining a phylogeny of extant taxa one finds lobed ordissected leaves alongside entire-leafed species ndash across mostlineages in the group (Weston 2007) Therefore althoughsuggestions of microclimatic adaptation may be offered atthe species level it seems unlikely that an environmentalassociation with leaf shape could be generalised acrossAustralian Proteaceae Rather leaf shape variation in theProteaceae is likely to reflect the range of ecological solutionsto leaf structurefunction trade-offs discussed above

Pelargonium

Similar to the Proteaceae the genus Pelargonium isdistributed predominantly in the southern hemisphere withnearly 80 of its almost 300 species found in South AfricaAlthough the age of the genus is estimated to be around30million years (Bakker et al 2005) the highest rate ofspecies accumulation has occurred in the last 10million years(Martinez-Cabrera 2010) The combination of climatic andedaphic heterogeneity over short distances and limited geneflow (Latimer et al 2005 Linder 2005) is frequently offeredas an explanation for the extraordinary plant species diversity thatcharacterises southern Africa Pelargonium the third largestgenus in the region is characterised by remarkable variation ingrowth form that ranges from annuals and geophytes to shrubsVariation in leaf shape is no less dramatic ranging from entireto highly dissected (Fig 1) often within growth forms andtaxonomic sections Parsing of leaf form categorically revealedthat the extent of dissection of the blade is highly evolvablewithinanoverall ovate leaf shapeoutline that is evolutionarily conserved(Jones et al 2009)

In addition to high evolutionary lability of leaf dissectionthere is variation among major clades in the evolutionary patternof major veins One major clade (Clade C) is dominated bypalmate venation whereas another (Clade A) is dominated bypinnate venation although both types are found in both cladesWe note that pinnate venation is significantly associated withincreased dissection and reductions in functional leaf sizecharacteristic of the highly diverse lsquoxerophyticrsquo clade (CladeA2) of the winter rainfall region

To examine whether leaf shape translates into functionalvariation carbon gain and water use efficiency were measuredin pairs of species contrasted for leaf shape and replicated acrossthree subclades of Clade A and one subclade of the third majorclade CladeB (Nicotra et al 2008) In each case the species withthemore dissected leaves showed higher rates of carbon gain butalso higher rates of water loss in all treatment conditions Thatpaper argued that the differences in carbon gain were not directresults of leaf shape per se but rather that the same conditionsthat result in selection for more dissected leaves also favourthe evolution of high photosynthetic rates high leaf nitrogencontent and opportunistic use of water when available Evenin this study however there was no correlation between leafdissection and the climate present in native ranges suggestingeither that climate is selective at a much smaller scale (ie themicrohabitat) or that for any given set of climate variablesmultiple leaf shapes can be adaptive in association withvariation in other leaf traits

Heteroblasty and heterophylly variationwithin a single genetic individual

We have so far focussed primarily on cross species variation inleaf shape As discussed for the Proteaceae genera above (lsquoTheProteaceaersquo section) however a single genome can produce aremarkable range of leaf shapes during the lifetime of a plantThere are two broad categories of leaf shape variation that can beexpressed by a single individual This variation can arise either inresponse to different external environments or to ontogeneticsignals (some of which may be influenced by environment aswell)

Environmentally induced heterophylly

Environmentally induced heterophylly results when leafshape responds to environmental cues The classic case is thedramatically different emergent and submergent leaves in aquaticplants Although many aquatic plants are also heteroblastic (seebelow Sculthorpe 1967) environmentally induced heterophyllyis not specific to position or plant age and is reversible dependinglargely on the environmental signals (mediated by hormonalsignals) perceived by cells of the developing leaf primordiumThe developmental age of the leaf at the time of signal perceptionis critical Very young primordia (L1ndashL3) generally respondcompletely to the novel environment whereas primordiathat were slightly older at the time of environmental changeshow intermediate features as mature leaves Characteristics ofspecific cells are related to the position of that cell within thedifferentiation gradient characteristic of that species at the timeof environmental change (eg Goliber and Feldman 1990 Bruniet al 1996 Kuwabara and Nagata 2006)

The evolution of leaf shape Functional Plant Biology 545

Among terrestrial plants the lsquosunndashshadersquo leaf response isfrequently cited as the classic example of environmentallyinduced heterophylly Historically differences in leaf formwithin a canopy have been assumed to be responses to alteredlight environments (eg larger and less-lobed leaves are producedin shade)However these different leaf shapes also confer distincthydraulic advantages (see lsquoTemperature and waterrsquo section)Whether in response to light or to hydraulic limitation lsquosunshadersquo heterophylly is accompanied by similar and well-knownanatomical differences across a wide range of species (ie sunleaves are generally smaller and thicker with smaller cells thanshade leaves Taiz and Zeiger 2002) However when theseanatomical differences become determined developmentally(Dengler 1980 Yano and Terashima 2004) has been lesscommonly studied The presence of intermediate anatomicalcharacteristics in leaves transferred among light environmentsshortly after expansion ceased suggests that leaves of somespecies have the capacity to respond anatomically to alteredlight levels until very late in development (Oguchi et al2005) In contrast recent studies suggest that systemic lightand humidity signals sent from older to younger leavesallow younger leaves to develop phenotypes suited for theiranticipated environments (Lake et al 2001 Yano andTerashima 2001 Thomas et al 2004) Exactly when final leafshape is determined in lsquosunshadersquo leaves remains an openquestion for most species

Heteroblasty describes ontogenetically dependentchanges in leaf features most notably shape

In 1900 Goebel originally distinguished differences in lsquojuvenileformrsquo from lsquoadult formrsquo on the basis of dramatic differences inleaf shape and associated traits (Balfour 1969) As an example he

described the shift from compound leaves to phyllodes in Acaciaas heteroblastic and reserved the term homoblastic for the lessdramatic variation exhibited by Casurina It is now generallyaccepted that heteroblasty refers to the full suite of features thatchange along the shoot during development (Jones 1999)

Heteroblasty results from the interaction of developmentalprograms expressed at both the leaf and the shoot- or whole-plant level (Kerstetter and Poethig 1998 Tsukaya et al 2000Usami et al 2009 Wu et al 2009 Chuck and OrsquoConnor 2010)Heteroblasty is generally not reversible unless reiterative growthoccurs (ie following damage) Morphological differencescharacteristic of heteroblasty such as differences in lobing areusually expressed relatively early in development shortly afterleaf initiation (Jones 1995)

The sequence of heteroblastic leaf shapes produced alongthe shoot can be accelerated or delayed in response to differentgrowth environments or architectural manipulations (Ashby1948 Forster and Bonser 2009 Jaya et al 2010)Furthermore both heteroblasty and plasticity in heteroblastyhave been shown to be under strong genetic control (Anderson1989 Climent et al 2006) Forster et al (2011) suggested thatboth heteroblasty and plasticity in heteroblasty are likely to occurin response to lsquowhich leaf type confers the greatest benefit fora given light levelrsquo Several studies over the last decade havepointed towards functional consequences of heteroblastygenerally with reference to single environmental variables (egHansen 1986 Gamage and Jesson 2007 Kubien et al 2007)It is likely that heteroblasty influences multiple leaf functionsbecause suites of leaf traits change with heteroblasty (Jones2001) Indeed this conclusion was supported in a recent studythat examined differences inmore than a hundred leaf traits in koa(Acacia koa AGray Pasquet-Kok et al 2010) Heteroblasty hasalso been proposed as an adaptation to herbivory (Karban and

Table 1 Promising directions for cross-disciplinary investigation of leaf shape

No Direction

1 A multi-species lsquoevo-devorsquo approach similar to that by Illing et al (2009) is likely to be informative with regard to many enigmatic features regardingthe evolution of diverse leaf forms and more generally provides an exciting system for examining the convergence and divergence in the evolutionof function

2 A phylogenetic analysis of leaf shape examining correlated evolution with other traits such as venation size leaf surface characteristics and branch andarchitectural measures would enable a thorough investigation of the situations in which leaf shape diversification arises

3 Further research into both the genetic and developmental processes underlying lobes and teeth and the ecological correlates of each would provide anexcellent opportunity to link across wide environmental scales

4 Cross species studies ideally in the field of the relative importance of leaf shape among other factors in determining leaf temperature and water status arenecessary to test the conclusion put forward here that the functional significance of leaf shape lies more often in water relations than directly in leafthermoregulation

5 Studies of the scale of environmental change that selects for ontogenetic response in leaf shape in heteroblastic species fromdiverse environments will shedlight on links between genes development and ecology

6 Detailed quantifications of the extent towhich leaf shape differswithin canopies particularly of dioecious specieswhere selection on reproductive functioncan lead to dimorphic conditions will help link leaf shape to whole plant processes

7 Studies at the scale of transects across the leaf surface that simultaneously consider hydraulic resistance temperature and light will reveal the extent oflocalised hydraulic and thermal stresses within leaf surfaces that might act as selective forces on leaf shape per se Imaging approaches incorporatingmeasures of 3-dimensional shape temperature and fluorescence provide exciting opportunities for examining spatial heterogeneity in temperature andphysiological processes

8 Since patterns of leaf venation affect both hydraulic supply and capacity for photosynthate transport studies that simultaneously examine both willreveal the extent to which interactions among these fundamental processes either co-determine leaf shape or establish trade-offs ultimately reflectedin leaf shape

9 Finally studies examining the fitness values of different leaf shapes or plasticity in leaf shapes under a range of environmental conditions are needed tobetter identify which aspects of leaf shape are most likely to be adaptive

546 Functional Plant Biology A B Nicotra et al

Thaler 1999 Brennan et al 2001 Fadzly et al 2009) and it hasbeen associated with differences in frost resistance (Darrow et al2001) It is worth noting that studies of the functional significanceof heteroblasty have focussed on relatively few plant lineagesGiven that heteroblasty occurs widely among the angiospermsas well as in some ferns and gymosperms (eg Wagner 1952Mueller 1982 Pryer and Hearn 2009) the next questions mayfocus on the cost of heteroblasty and its fitness consequenceswhat scale of environmental change selects for ontogeneticresponse in leaf shape

Synthesis and next steps

Whether considered at community lineage or single genomescales leaf shape is a trait that is highly labile and responsive to arange of biotic and abiotic factors Although some leaf traitscan be explained in terms of particularly strong trade-offsbetween two or more factors that result in clear scalingrelationships (eg leaf mass-to-area traits or leaf mass-to-stemmass) leaf shape in contrast does not emerge from the aboveconsideration as a factor capable of explaining ecologicaldifferentiation between species Rather leaf shape emerges asa trait for which there are many quite varied functional trade-offs

As stated above these trade-offs for successful functionalityof leaves may generate ecological opportunities leaf shape isperhaps best viewed not as a single major axis but rather asan option that fine tunes the leaf to its conditions over bothshort and evolutionary time spans Numerous elements of thisreview support this conclusion Over evolutionary history theextant range of variation in leaf shape has been explored bymanylineages although only the angiosperms have explored thisrange with the added element of massive increases in veindensities Likewise our growing understanding of geneticcontrols on leaf shape reveal that it is a complex traitdetermined by many different genes at several differentdevelopmental stages Although some genes have conservedaffects on leaf morphology including shape the exact natureof their affects can vary widely Likewise the function to whichagivenelement of leaf shape lsquois putrsquo canvary For example closerinvestigation of the internal architecture and developmentalpathways associated with different leaf shapes suggests thatleaf teeth and leaf lobes often have distinctly differentfunctional foundations and may play different functions indifferent environments

Despite amyriadof functional roles the one recurrent theme inour review is that leaf shape is affected by and strongly influencesleaf water relations From a functional perspective the role of leafshape in leaf thermoregulation has perhaps received the mostattention in previous considerations of the functional significanceof leaf shape The relationship between leaf size and temperaturebeing physically determined by basic energy balance principlesthis was a logical starting point of enquiry But as we haveshown the connection between leaf shape (and size) and leaftemperatures under field conditions is not well supportedempirically Rather leaf shape variation as distinct from sizemore often will reflect water supply trade-offs High veindensities support high photosynthetic rates (Brodribb et al2010) and dissected leaves maintain a greater proportion ofleaf tissue closer to main veins Leaf temperature plays a role

in this story as higher leaf temperatures lead to greaterevaporative water loss but many factors other than leaf shapeinfluence leaf temperature (eg leaf thickness orientation andreflectivity and branch and plant architecture) Finally it is alsoworth noting that although there appears to be an obviouslink between shape and water supply there is a concomitantinteraction between water supply and phloem function thatremains relatively unexplored

Thus our review suggests several angles for further research(Table 1) In particular we advocate those approaches thatenable investigation across the scales examined here (genedevelopment function and evolutionary ecology) and thatfocus in particular on the links among leaf shape developmentand water relations

Acknowledgements

The authors thank the Botanical Society of America for their support of thesymposiumfromwhich this reviewaroseWealso thankLawrenSackand twoanonymous reviewers for useful comments on an earlier version of thismanuscript

References

Ackerly DD Knight CA Weiss SB Barton K Starmer KP (2002) Leafsize specific leaf area and microhabitat distribution of chaparral woodyplants contrasting patterns in species level and community level analysesOecologia 130 449ndash457 doi101007s004420100805

Anderson S (1989) Variation in heteroblastic succession among populationsof Crepis tectorum Nordic Journal of Botany 8 565ndash573 doi101111j1756-10511989tb01730x

Armbruster WS Muchhala N (2009) Associations between floralspecialization and species diversity cause effect or correlationEvolutionary Ecology 23 159ndash179 doi101007s10682-008-9259-z

Armbruster WS Di Stilio VS Tuxill JD Flores TC Runk JLV (1999)Covariance and decoupling of floral and vegetative traits in nineneotropical plants a re-evaluation of Bergrsquos correlation-pleiadesconcept American Journal of Botany 86 39ndash55 doi1023072656953

Ashby E (1948) Studies in the morphogenesis of leaves I An essay on leafshape New Phytologist 47 153ndash176 doi101111j1469-81371948tb05098x

Aso K Kato M Banks JA Hasebe M (1999) Characterization ofhomeodomain-leucine zipper genes in the fern Ceratopteris richardiiand the evolution of the homeodomain-leucine zipper gene family invascular plants Molecular Biology and Evolution 16 544ndash552

Bailey IW Sinnott EW (1915) A botanical index of Cretaceous and Tertiaryclimates Science 41 831ndash834 doi101126science411066831

Bailey IW Sinnott EW (1916) The climatic distribution of certain types ofangiosperm leaves American Journal of Botany 3 24ndash39 doi1023072435109

Baker-Brosh KF Peet RK (1997) The ecological significance of lobed andtoothed leaves in temperate forest trees Ecology 78 1250ndash1255

Bakker FT Culham A Marais EM Gibby M (2005) Nested radiation inCape Pelargonium In lsquoPlant species-level systematics new perspectiveson pattern and process Vol 143rsquo (Eds FT Bakker LW Chartrou BGravendeel PB Pielser) pp 75ndash100 (ARG Ganter Verlag KG RuggellLichtstein)

Balfour IB (Transl) (1969) lsquoOrganography of plants I General organographyrsquoby K Goebel 1900 (Hafner Publishing Co Inc New York)

Ball MC Cowan IR Farquhar GD (1988) Maintenance of leaf temperatureand the optimization of carbon gain in relation to water loss in a tropicalmangrove forest Australian Journal of Plant Physiology 15 263ndash276doi101071PP9880263

The evolution of leaf shape Functional Plant Biology 547

(Smith and Nobel 1977 Bragg and Westoby 2002) exposed(Ackerly et al 2002) nutrient poor (Beadle 1966 Cunninghamet al 1999 Fonseca et al 2000) and saline environments (Ballet al 1988) or many of these factors in combination (egMcDonald et al 2003) Leaf shape in contrast has beenshown to vary less predictably across environments or biomesthan size or LMA (McDonald et al 2003) Thus the functionalsignificance of leaf shape and the evolutionary drivers of itsdiversity remain the subject of discussion

The theories about leaf shape are many and not mutuallyexclusive thermoregulation of leaves especially in arid and hotenvironments hydraulic constraints patterns of leaf expansionin deciduous species mechanical constraints adaptations toavoid herbivory adaptations to optimise light interceptionand given that leaves are hypothesised to be developmentalhomologues of floral organs and it has even been suggestedthat leaf shape reflects the effects of selection on flower formFinally there is the chance that leaf shape variation has littlefunctional or adaptive significance and instead reflects randomvariation within the context of phylogenetic history Howevergiven the importance of the leaf we believe the latter option ratherunlikely To distinguish among the formerwe here synthesise the

varied fields involved so that we may assess the relativeimportance of each and identify the ecological situations inwhich one or the other of these mechanisms is likely to besignificant

In this reviewwefirst examine leaf shape froman evolutionaryperspective and in the process demonstrate that shape is highlylabile and widely explored in evolutionary history We bring thefocus onto the angiosperms in particular and then consider theproximate determinants of angiosperm leaf shape The currentunderstandingof thegenetic signals anddevelopmental processesunderlying shape differences are then reviewed to explore thegenetic controls and constraints on leaf shape evolution Thehypotheses about the function of leaf shape are next reviewedwith an evolutionary and genetic perspective in mind Wehighlight those functional issues that we see as particularlyrelevant to leaf shape evolution and then turn to threeexamples of leaf shape variation across communities withinlineages and within individuals to explore how an evolutionaryperspective alters understanding of the relationship between leafform and function We conclude with suggestions on how thisperspective can be used to direct future research on the functionand evolution of leaf shape

(a)

(b)

(e)

(f ) (g)

0 10 20 30 40 50 mm

(h) (i ) ( j )

(c) (d )

Fig 1 Leaf shape in the genus Pelargonium ranges from single and entire to compound and highly dissected Species are (a) P bowkeri (b) P reniforme(c) P klinghardtense (d) P fulgidum (e) P carnosum ( f ) P cucullatum (g) P abrotanifolium (h) P citronellum (i) P australe and ( j) P alternansPhotographs courtesy of Stuart Hay ANU Photography

536 Functional Plant Biology A B Nicotra et al

A brief history of the angiosperm leaf and shapediversity thereinTo understand the evolution of angiosperm leaf shape someperspective on the evolution of the leaf itself is needed as leavesof many shapes have evolved numerous times independentlyin evolutionary history For the purpose of this review ofangiosperm leaf shape we define a leaf as a vascularasymmetric appendicular structure initiated at the shoot apicalmeristem This definition excludes the functionally analogousstructures of mosses and leafy liverworts Morphologicallysimple leaves microphylls arose in the lycopsids Largerleaves with more complex shapes and venation networkscalled megaphylls arose in the seed plants in the extinctprogymnosperm and sphenophyll lineages and in one ormore lineages of lsquofernsrsquo (Galtier 1981 Kenrick and Crane1997 Niklas 1997 Boyce and Knoll 2002 Tomescu 2009Boyce 2010) Thus the seed plants represent just one ofseveral independent evolutionary origins of large leaves withdiverse shapes among the vascular plants

Despite the many independent origins of leaves several traitsnow associatedwith leaves can be considered homologous acrossall vascular plants through a shared monoplyletic origin from aleafless common ancestor First all vascular plant leaves aredistinct from the leaf-like organs of bryophytes that rely onexternal surfaces for photosynthetic gas exchange (as wouldalso be true of the external appendages independently derivedin many algal lineages Niklas 2000) Vascular plant leaves useinternal airspaces regulated by stomata for gas exchange (Raven1996 Boyce 2008a) Second evidence from both living plantsand the venation patterns of fossil leaves indicates that theancestral form of tissue production in the growing leaves of allvascular plants was limited to discrete marginal zones of growth(Pray 1960 Zurakowski and Gifford 1988 Boyce and Knoll2002 Boyce 2007) Other aspects of leaf organography suchas the evolution of differentiated abaxialadaxial domains (seesection on lsquoProximal mechanisms underlying leaf shapediversityrsquo below) and determinate leaf growth appear to ariseindependently in seed plants and other lineages however severalof these lineages have co-opted the same underlying geneticpathways to regulate these developmental processes (seesection on lsquoProximal mechanisms underlying leaf shapediversityrsquo Bharathan et al 2002 Harrison et al 2005 Sanderset al 2007 Tomescu 2009 Boyce 2010)

Within this broader evolutionary context angiosperm leavesrepresent a rangeof particularly aberrant variants of vascular plantleaves Some of the distinctive characteristics of angiospermleaves have arisen independently in other lineages whereasother traits are unique to the flowering plants Angiospermshave evolved leaf growth that occurs diffusely throughout theleaf without being limited to the margin (Pray 1955 Poethig andSussex 1985a 1985b) thus enabling the enormous diversity andcomplexity of angiosperm leaf venation patterns and shapes Thephylogenetic distribution of complex venation patterns suggestthat the evolutionary departures from marginal growth observedin the angiosperms have also arisen repeatedly in two or threeother seed plant lineages and 10 or more groups of ferns (Boyce2005) In these likewise diffuse growth and complex venationpatterns can be seen to occur with diverse leaf shapes Despitethese similarities to the other seed plants angiosperm leaf

evolution remains distinctive in that it ran counter to thegeneral temporal trend seen in the vasculature of other seedplants The vascular network of other seed plant lineagesevolved towards simplification After initially exhibiting thecomplete range of morphologies consistent with marginalgrowth including many involving two vein orders with amidvein and open or reticulate networks of secondary veins(morphologies now common only among the ferns) the leafarchitecture in the other major vascular plant lineages becameprogressively limited to a narrow subset of forms In cycadsconifers Ginkgo and several other extinct seed plant lineages asingle order of veins lead straight parallel courses to a distalmarginAngiosperms however exhibitmany hierarchical ordersof reticulate internally directed veins (Boyce 2005) Furtherangiosperm leaves are unique in their possession ofextraordinarily high vein densities (Fig 2) whereas all otherplants living or extinct average ~2mm of vein length per squaremm of leaf surface angiosperms average between 8 and10mmmmndash2 and can range to above 20mmmmndash2 (Boyce2008a 2008b 2009 Brodribb and Feild 2010 Brodribb et al2010)

Together these changes that accompanied angiospermevolution represent a substantial revision of the functionalpossibilities of a leaf First the release of venation patterns fromstrictly marginal growth allows the production of novel leafshapes and sizes that would not otherwise be possible Seconddifferences in angiosperm venation allowed a complete revolutionin how water is distributed within a leaf (Boyce 2008a) In anyleaf the tissue furthest from the leaf base has access only to thatwater which has not been lost to transpiration in more proximaltissues (Zwieniecki et al 2004a 2006) Reticulate venation andthe minute size of the final order of veins in flowering plantsleads to equitable distribution of water between successive veinorders (Zwieniecki et al 2002) Finally the high vein densities ofangiosperms shorten the path length along xylem and mesophyllthus enabling assimilation and transpiration capacities higherthan in any other group of plants (Bond 1989 Sack and Frole2006 Brodribb et al 2007 Boyce et al 2009) The uniquely hightranspiration capacities of angiosperms may also have influencedthe evolution of leaf shape by relaxing some of the thermalconstraints on leaf size and shape (see lsquoTemperature and waterrsquosection)

It remains controversial when and in which lineages thesecharacteristics of angiosperm leaves arose The distinctivenessof angiosperms leaves has led to polarised interpretations oftheir ancestry Some researcher propose that similarities inleaf venation tie the early angiosperms to particular fossil seedplants (eg Melville 1969) but the group of plants with themost similar venation are the clearly unrelated dipterid fernsThe various seed plant lineages most often favoured asangiosperm relatives based upon reproductive characteristics(Doyle and Hickey 1976 Hilton and Bateman 2006 Frohlichand Chase 2007) tend to have very dissimilar leaves Otherssuggest that angiosperm leaves are so distinct as to require acomplete reinvention of a leaf after passing through a leaflessintermediate such as one that was aquatic- or desert-adapted(Doyle and Hickey 1976) But the appearance of angiosperm-like leaf traits in a variety of fern lineages suggest thatinvoking a leafless intermediate is unnecessary Although the

The evolution of leaf shape Functional Plant Biology 537

history of early angiosperm leaf evolution is still debated weare certain that there have been numerous independent originsof diverse leaf shapes and that leaf shape diversificationis associated with both high vein densities and reticulatevenation patterns

Proximal mechanisms underlying leaf shape diversity

What are the genetic mechanisms underlying this leaf shapediversification and how similar are they given the many

independent origins of diverse leaf shapes What we know ofthe processes of leaf development (and leaf shape determination)is largely a product of studies on model species systems Inparticular Arabidopsis thaliana L (Heynh) (hereafter referredto as Arabidopsis) has been widely used for the identificationof genes determining leaf form In the following sections weexamine the development of the angiosperm leaf specificallythe establishment of dorsiventral domains and flatteningexpansion of the lamina and formation of leaf margin anddiscuss how these factors interact to determine leaf shape

(a) (b)

(c)

Fig 2 Leaf venation architecture in (a) the cycadZamia furfuracea (b) the fernBlechnumgibbum and (c) theangiosperm Boehmeria nivea overlain with the fern Polypodium formosanum Both the Zamia and theBlechnum have marginally ending veins suggesting marginal growth but the single order of parallel veins inZamiamust distribute water over more than 10 cm of length whereas the finest veins in theBlechnum are a fewmm The finest veins distributing water in angiosperms can be even shorter and the density of veins muchhigher the Boehmeria has ~8mm of vein mmndash2 of leaf versus 1mmmmndash2 in the Polypodium

538 Functional Plant Biology A B Nicotra et al

Establishment of dorsiventral domainsand lamina flattening

Since leaf primordia arise as protrusions from shoot apicalmeristems (SAM) suppression of the regular SAMdevelopmental program is the first step required for normaldevelopment of a leaf primordium Recently Sarojam et al(2010) found that genes in the YABBY family suppress SAMidentity and promote lamina growth in Arabidopsis In mostbifacial leaves such as leaves of Arabidopsis and snapdragon(Antirrhinum majus L) the radial primordium differentiatesinto abaxial (lower) and adaxial (upper) surfaces to establishdorsiventrality (Waites and Hudson 1995) Our currentunderstanding is that the molecular genetic regulation ofidentification of the abaxial and adaxial domains depends onactions of both small RNAs and directindirect targets suchas class III Zip genes KANADI genes and AUXIN RESPONSEFACTORS3 (ARF3)ETT in Arabidopsis (Husbands et al 2009Floyd and Bowman 2010 Kidner 2010) Without establishmentof the adaxial and abaxial domains it is believed that leavescannot expand in any angiosperm species rather they becomestalk like since the so-called plate meristem is never activated toexpand the laminas (Fig 3)

The monocot clade is typified by many unifacial leaves thathave only abaxial identity in the lamina as seen in leek (Alliumampeloprasum var porrum L) and Iris spp Leek leaves arecylindrical since they lack adaxial identity in their leaf bladeIris in contrast has flat leaves that expand in a distinct

perpendicular direction relative to that of normal bifacialleaves Likewise the rush Juncus prismatocarpus RBr hasunifacial flat leaves the primordium of the leaf blade growsinto a flat lamina as a result of extensive thickening growthThis developmental pathway is dependent on the geneDROOPING LEAF (DL Yamaguchi et al 2010 Fig 3) thatis also a member of the YABBY gene family but with a monocot-specific function As seen in this case diversified morphology ina particular taxon sometimes depends on some taxon-specificmolecular mechanisms that are modifications of gene functionfrom ancestral lineages

Mechanisms for expansion of lamina

Once dorsiventral identity is generated the leaf lamina can thendiversify in morphology to form simple or compound leaveswith entire or serrated margins differing in thickness length andwidth and in orientation (Tsukaya 2006) The leaf length widthratio is regulated by polar-dependent cell expansion and cellproliferationdistribution Several key genes for its regulationhave been identified from Arabidopsis ANGUSTIFOLIA (AN)and ROTUNDIFOLIA3 (ROT3) regulate the shape of cellswhereas ANGUSTIFOLIA3 (AN3) and ROTUNDIFOLIA4(ROT4) affect the number of cells in the lamina (Tsukaya2006) Loss-of-function mutations of AN and AN3 result innarrower leaves whereas loss-of-function of ROT3 oroverexpression of ROT4 make leaves shorter In addition

Fig 3 Twodifferentmechanismsof laminagrowth Inmost leaves such as leaves ofArabidopsis (upper left)dorsiventral identity (adaxial upper side (shown in green in colour version online) and abaxial lower side(shown in blue in colour version)) identities are established Flattening lamina growth occurs along the junctionbetween adaxial and abaxial domains as a result of activity of the plate meristem Loss of either abaxial oradaxial identity results in radialised growth (lower right) Stick-like leaf morphology seen in leek and Juncuswallichianus (centre) is the typical example of unifacial leaves that have no adaxial domain In the monocotclade however some species develop flat lamina irrespective of lack of adaxial identity in the lamina (rightSisyrinchium rosulatum) that is supported by enhanced growth in the leaf-thickness direction Meristematicactivity occurs in center region of both leaves (orange in colour version) relative position of the shoot apicalmeristem (SAM) is shown by the cross-hatched dot (green in colour version)

The evolution of leaf shape Functional Plant Biology 539

LONGIFOLIA1 (LON1) and LON2 regulate the length of the leaflamina in Arabidopsis (Lee et al 2006) A vast array of genes isknown to influence leaf area (Horiguchi et al 2006) For exampleheteroblastic change in leaf area is governed by the miR156-SQUAMOSAPROMOTERBINDINGPROTEIN-LIKE (SPL)-miR172 pathway via regulation of cell size and cell numbers inArabidopsis (Usami et al 2009)Alterations of activities of anyofthese factors could contribute to natural variation in leaflength width ratio and area None of these genetic variationsresult in loss of reproductive structures suggesting that thesegenes could be under direct environmentally mediated selectionAs yet we do not know whether the above patterns hold in othernon-model species

Leaves also vary in the extent to which the lamina is flator curved CINCINATA (CIN) of snapdragon was oncethought to function to keep the leaf flat by orchestrating cellproliferation along the medio-lateral axis of the leaf primordia(Nath et al 2003) More recently however analyses of theTCP genes which are homologues of CIN in Arabidopsisindicate that the major role of the TCP is in promotion ofmaturation in laminar tissues not in regulation of cellproliferation (Efroni et al 2008 Sarojam et al 2010) The roleof CIN in snapdragon may also be the same as that of TCPsin Arabidopsis The other candidate genetic mechanism forcontrol of leaf curvature is an auxin-related pathway Severalhyponastic leaf mutants have been isolated from Arabidopsisand many of the responsible genes are related to auxinsignalling (Perez-Perez et al 2002 2009) Since auxinsignalling is also very important for venation patterningspecies-specific modification of auxin signalling in leafprimordia may have caused diversification of both leafcurvature and vasculature patterning

Genetic regulation for patterning of leaf marginserration and leaflet formation

Variation in leaf shape as distinct from length and area orcurvature arises from variation in leaf margin and lobeformation Kawamura et al (2010) showed that spatialpatterning of serrations or leaf teeth (see lsquoLeaf shape variationwithin lineagesrsquo section) on the leaf margin depends onformation of auxin maxima on the leaf margin in ArabidopsisLocation of these maxima is governed by polar auxin transportand the stabilisation of auxinmaxima ismaintained by prolongedexpression of CUP-SHAPED COTYLEDON2 (CUC2) around it(Kawamura et al 2010) The mechanism of rhythmic patternformation of the auxin maximaminima on the leaf margin isthought to involve self-organising pattern formationmechanismsby polar auxin transport similar to that involved in patternformation of phyllotaxy (de Reuille et al 2006 Jonsson et al2006) Either the loss of distinct auxinmaxima or an instability ofthe auxin maxima will result in an entire leaf margin (Kawamuraet al 2010) Changes in functions of these genes howeverseverely influence many biological processes including floralorgan development as seen in the pin-formed 1 mutant (Okadaet al 1991) Thus variation among species in pattern of marginserration might accompany variation in floral organs or morelikely reflect differential regulation for example by downstreamgenetic pathways

Combined control of auxin-maxima formation by PIN andCUC genes as described for serration above is also known tohave an important role in leaflet patterning in bittercress(Cardamine flexuosa With) an Arabidopsis relative withcompound leaves (Barkoulas et al 2008) Again this controlvaries among species in tomato (Solanum lycopersicum L) lossof CUC3 function does not convert compound leaves intosimple leaves (Blein et al 2008) Thus it is clear that theindependent evolution of compound leaves from simple leavesis likely driven by species-specific mechanisms (Efroni et al2010) though these may reflect modifications of similarpathways

Recent studies have identified several important key genes forregulation of compound versus simple leaf forms For exampleabsence or prolonged expression of class I Knotted1-likehomeobox (KNOX) genes in leaf primordia is important tocompound leaf formation in a wide range of angiosperms andgymnosperms (Bharathan et al 2002) In some legume specieshowever the LEAFY (LFY)UNIFOLIATA (UNI) gene is the keyfor compound leaf formation instead of the class I KNOX genes(Hofer et al 1997) Changes in these genes could be linked tothe evolution of compound leaves although loss-of-function ofLFYUNI also causes infertility in Arabidopsis and thus it isunlikely that selection directly on these particular genes couldresult in theevolutionof variation in leaf shapeDiversified spatialpatterning of petiolepetiolule and laminaleaflet in compoundleaves have also been suggested to be linked to the distributionpattern of AS1-ROUGH SHEATH2-PHANTASTICA (ARP)expression domains (Kim et al 2003) However based on thefact that the ARP gene has only a limited role in compound leafformation in some species it is unclear whetherARPwill turn outto be major factor in others (Efroni et al 2010)

As described above the independent evolutionary origins ofseveral angiosperm leaf features should caution against theassumption that a single ancestral genetic system underlies leafdevelopment The developmental genetic approaches outlinedhave revealed several key mechanisms for leaf shape controlbut have also demonstrated variation in the function of thesegenes among lineages In a few studies researchers have gonebeyond one or two species studies and have looked for deeperevolutionary links (Illing et al 2009) Studies of the role of theclass 1 KNOX genes have considered not just angiospermsbut other lineages Among simple-leafed dicots and monocots(eg Arabidopsis and Zea) KNOX genes are strongly expressedduring the indeterminate growth of the apical meristem but arerapidly downregulated in the cells flanking the apical domein regions defined by leaf primordia Although some arginine-rich protein (ARP) genes repress the expression of KNOXtranscription factors during the development of both micro-and megaphyll primordia (Harrison et al 2005) the role ofhomeodomain-leucine zipper (HD-ZIP) III genes which areinvolved in the specification of foliar adaxialabaxial fatediffers in the development of these two leaf types (Floyd andBowman 2006) HD-ZIP III genes have been identified in themoss Physcomitrella (Sakakibara et al 2001) and in the fernCeratopteris (Aso et al 1999) they are also expressed duringnormal vascular tissue development in Arabidopsis (Zhong andYe 1999) suggesting that they are very ancient but alsomultifunctional

540 Functional Plant Biology A B Nicotra et al

Accumulating knowledge on molecular genetic mechanismsof leaf morphogenesis is leading to an increasingly evo-devoapproach to studies of leaf diversity (Tsukaya 2010)this progress is aided by establishing new model species(eg Cardamine hirsuta and the rush Canales et al 2010Yamaguchi and Tsukaya 2010) Over the next decade weanticipate that genomic approaches on multiple speciesanalysed in phylogenetic frameworks (eg Illing et al 2009)will resolvemuchof theenigma regarding the evolutionofdiverseleaf forms

Genetic constraints on flower petal and leaf form

One additional developmental consideration is that as famouslydescribed by von Goethe in 1790 flower petals are derivatives ofthe apical meristem and serial homologues to leaves (von Goethe2009) Thus one might hypothesise that although the evolutionof leaves substantially predates that of flowers selection on theform of either of these organs would also result in changes inthe other such that selection on leaf shape and size subsequent tothe evolution of floral organs may be constrained by pleiotropiccontrol of floral form Recent genetic data identifies several genesthat have roles in both leaf and flower form (Dinneny et al 2004Schmid et al 2005 Street et al 2008) For example geneticchanges in both leaf length to width ratio (eg rot3) and leaf size(eg an3) influence not only leaf proportions but also that of thefloral organs in Arabidopsis (Kim et al 1999 Horiguchi et al2005) The garden pea (Pisum sativum L) is a good case in pointas discussed above compound leaf formation depends on LFYUNI but the same gene is also required for flowering Thus ifcurrent selection favoured simple leaves in garden pea changes inleaf shape could be constrained by pleiotropic control offlowering (Hofer et al 1997) Likewise Shalit et al (2009)recently reported that the action of florigen ndash a plant hormonefor flowering ndash influences complexity of leaves by changingnumbers of leaflets in tomato Such relationships suggest thatsome floral characters as well as other metabolic pathways maynowbeclosely linkedwith regulationof leaf shapeand sizeA fewecological studies have also assessed correlations between petaland leaf form (Berg 1960 Armbruster et al 1999) This questionis of particular interest in lineages where animal pollination isassociated with increased diversification rates either as a result ofdirect selection or as a secondary reinforcing factor (van der Nietet al 2006 Armbruster and Muchhala 2009 Kay and Sargent2009)

In situations where leaf and petal form are under similargenetic control and both under selection one of two thingscould happen If leaf shape does not have substantialfunctional impact then one might expect leaf form to changewith selection on petal form However given the likelyimportance of leaf form to photosynthetic function it is morelikely that a breakdown in genetic correlations between leafand flower form (eg by alterations of downstream regulatorypathways) would evolve Several factors argue against thehypothesis that leaf shape is frequently a product of selectionon flower form First diverse leaf shapes were in evidencelong before the origin of flowers in the angiosperms Secondthe genetic mechanisms underlying leaf shape revealconsiderable evolvability with similar genes being co-opted

for different functions in different species Finally given thepotential importance of both petal and leaf shape we suggestthat genetic constraints on the independent evolution of bothstructures are likely to have broken down in those lineages withhighly diversified leaf shapes over the many millions of yearssince petals first evolved

Ecological correlates and the functional significanceof leaf shape

Evolutionary history plays a part in determining leafmorphology but our consideration of leaf shape evolution andits developmental genetic basis demonstrates that diverselineages have repeatedly gained and lost many leaf featuresThus developmental and genetic constraints on the evolutionof leaf form and shape in particular have been overcomemany times or have led to new evolutionary opportunities Itis of particular interest then to ask what are the likely ecologicaldrivers andor functional significance of leaf shape

Temperature and water

Conventionalwisdomwould say that thermoregulation is perhapsthe main driver of leaf shape evolution The maintenance ofleaf temperatures within certain limits is critical for a plantrsquosgrowth and survival and although photosynthetic tissue of somespecies canwithstand temperatures below6C (Ball et al 2006)and above 60C (Clum1926Nobel 1988) irreversible damage toleaves can occur at temperatures well within this range (Levitt1980 Jones 1992 Ball et al 2004 Groom et al 2004 Sharkey2005) Leaf size and shape potentially have large effects onleaf temperature because the two-dimensional proportions of aflattened leaf determine the rate of heat transfer between acrossthe leafndashair interface by influencing the thickness of its boundarylayer

All else being equal the thickness of a boundary layerincreases with length from the windward edge so that heatconvection from small leaves is more rapid than from largeleaves (Raschke 1960 Gates 1968 Vogel 1970 Parkhurst andLoucks 1972Grace et al 1980Geller andSmith 1982Monteithand Unsworth 1990 Schuepp 1993) Likewise because leaflobing reduces the distance across the lamina the rate of heattransfer is predicted to be greater in a lobed leaf than an unlobedleaf with equivalent area (Parkhurst et al 1968 Vogel 1968Lewis 1972 Givnish 1978 Gurevitch and Schuepp 1990a)

The rate of heat transfer from lobed metal plates or lobedleaves coated in metal is greater than from shallow-lobed or un-lobed ones (Parkhurst et al 1968 Thom 1968 Vogel 1970Gottschlich and Smith 1982 Gurevitch and Schuepp 1990bRoth-Nebelsick 2001) Artificial or coated leaves have beenused because they enable boundary layer resistance to beinvestigated in the absence of transpiration However thisapproach is biologically flawed because it does not enable anassessment of the relative importance of a range of traits (eg leafwater content absorbance) in determining the actual operatingtemperatures of leaves in the field

Actual leaf temperature measurements show that large leavescan operate at below-ambient temperatures even on hot days asa result of high rates of latent heat loss through transpiration(Drake et al 1970 Gates 1980 Hegazy and El Amry 1998)

The evolution of leaf shape Functional Plant Biology 541

Thus conditions of evaporative demand and the availability ofsoil water for transpiration are key to understanding leafshapendashtemperature relationships under biologically realisticconditions Because of their much greater transpirationalcapacities (see above) flowering plants have far greaterleeway than other plants regarding the size and shape of theirleaves (see lsquoA brief history of the angiosperm leaf and shapediversity thereinrsquo section Boyce et al 2009)

In addition to reducing heat loads the morphology of deeplylobed leaves may also reflect direct selection for increasedhydraulic efficiency Water moving through a plant encountersnumerous resistances on its path from soil to roots to stemsthrough leaves and out to the external atmosphere Within theliquid phase of water flow through a plant leaf hydraulicresistance Rleaf accounts for more than 30 of total plantresistance (Sack and Holbrook 2006) The high resistance inleaves is due first to a system of veins of decreasing size theresistance of which is inversely proportional to the fourth powerof the radius of the component conduits in these veins Thereforeas a proportion of the considerable resistance in the leaf(Zwieniecki et al 2002) the minor veins provide the greatestresistance to water flow (Sack et al 2004)Water also encountershigh resistance as it exits veins laterally and travels to othercells mainly via apoplastic pathways through the mesophyllAccordingly on the order of half of total Rleaf can occur outsidethe major veins (Sack and Holbrook 2006) As resistanceincreases along the pathway towards sites of evaporationlocalised water potential (Y) becomes progressively morenegative (Brodribb et al 2010) The result of how thisphenomenon might affect a leaf has long been observedregions of leaves lying furthest from the main supply channelsare the first to lsquowitherrsquo when exposed to strong winds (Fig 4Yapp 1912) Leaf lobing represents an effective removal of thispotentially stress-prone tissue and has been suggested as anadaptation to dry conditions (Thoday 1931 Givnish 1979) Ifdeeply lobed leaves have a lower ratio of mesophyll tissue tolarge highly conductive veins they should then have reducedresistance relative to less- or un-lobed leaves (Sack and Tyree2005)

Although leaf lobing might confer adaptive benefits withrespect to maintaining stable hydraulic supply it may also bethat the extent of lobing or margin dissection is itself determinedby hydraulic limitation (Sack et al 2003 Zwieniecki et al2004b Boyce 2009) For species with lsquosunrsquo and lsquoshadersquoleaves the morphology of small deeply lobed sun leaves atthe outer canopy is proposed to result from water pressuredrop across the leaf lamina caused by reduced water deliveryto expanding cells during growth (Fig 5 Zwieniecki et al 2004bBoyce 2009 Leigh et al 2011)Alternatively recent studies havesuggested that many of the characteristics of lsquosunrsquo and lsquoshadersquoleaves are the result of branch autonomy within the plant canopy(Niklas and Cobb 2010)

Thus although the effects of leaf shape on leaf thermalregulation hold when tested on model leaves predictions areless straightforward when hydraulic function is accounted forAdditionally consideration of the relative contribution of a rangeof leaf and branch properties to leaf temperature is critical It islikely that leaf shape is only one amongmany factors influencingleaf thermal regulation other factors could include water

content leaf thickness spectral reflectance orientation andplant architecture Further given the observation of high leafshape diversity withinmany plant communities from hot and aridenvironments it seems likely that regulating leaf temperature isnot the singlemost important evolutionary influence on leaf shapediversity The association between leaf shape and hydraulicproperties which is affected by leaf temperature in contrast islikely to be of greater importance

Leaf structure and functional trade-offs

Many foliar leaf traits including shape reflect functional trade-offs that have been resolved in various ways by differentspecies depending on their ecological settings (eg hydro- vsmesophytes) physiological attributes (eg C3 vsC4metabolism)developmental capabilities (eg heterobaric vs non-heterobaricleaves) or evolutionary histories (eg micro- vs megaphylls)Life history and optimisation theory show that the number ofphenotypic solutions that allow for different equally successfultrait combinations increases as the number of trade-offsincreases ndash a conclusion that applies to traits within the leaf(eg for shape) as well as to leafndashbranch relationships (Niklas1988)

It is evident that leaves perform several functionssimultaneously that is they intercept light transport liquidsand solutes exchange gases with the atmosphere dissipateheat cope with externally applied mechanical forces and

(a)

(e)(f )

( i )

(g)

(h)

(b)

(c)

(d )

Fig 4 Localised lsquowitheringrsquo of leaves exposed to strong wind Reprintedfrom Yapp (1912) with permission from Oxford University Press

542 Functional Plant Biology A B Nicotra et al

defend themselves from herbivores and pathogens Importantlymany of these functions have conflicting physiologicalanatomical or morphological requirements For exampleorientations of lamina that maximise light interception alsominimise the ability to dissipate heat Further all of thesefunctions are intrinsically size dependent For example thequantity of mechanical tissues required to support laminaedepends on the mass of laminae whereas the ability tointercept light depends on laminae area

When seen in this light the trade-offs required for thesuccessful functionality of foliar leaves generate rather thanconstrain ecological and evolutionary opportunities This canbe illustrated mathematically and empirically in the context ofphyllotaxy and its consequences on light interception Computersimulations of mathematically generated shoots to assess theinfluence of leaf shape size and phyllotactic patterns on theability to intercept direct solar radiation show that differences inphyllotaxy significantly influence light interception (particularlyfor shoots with a rosette growth habit) They also show thatcomparatively small differences in leaf shape can compensatefor the negative effects of leaf overlap resulting from virtuallyany phyllotactic pattern For example lobed leaves or pinnifid

compound leaves facilitate light penetration through shootsbearing densely packed leaves Thus phyllotaxy can beviewed as a developmental limiting factor that can drivecompensatory changes in morphological features such asshape that are not directly controlled by patterns of leafinitiation In this way the lsquoconstraintsrsquo imposed by phyllotaxycan foster evolutionary modifications at the level of the entireshoot (Niklas 1988)

One particularly important trade-off operating at the level ofthe architecture of a shoot is that between leaf size and numberFor any given leaf biomass allocation an individual shoot caneither have a few large leaves or many smaller ones Acrossdifferent herbaceous species and juvenile specimens of treespecies drawn from diverse lineages and ecological settingsthe relationship between total leaf biomass per plant (ML) andtotal stemmass per plant (MS) fallswithin a comparatively narrowcorridor of variation that obeys a positive isometric scalingrelationship (ie MLMS

~10 Niklas 2004) Accordinglyacross these species any stem mass investment yields a moreor less proportional increase in leaf mass Therefore it followsmathematically that across these species the relationshipamong the average mass of a typical leaf mL total leaf numberNL and total stem mass complies with the proportionality mL

NLMS~10 When individual shoots are sampled across

diverse species the relationship between mL and NL per shootis negative and isometric (ie mLNL

ndash1 Kleiman and Aarssen2007) a relationship that appears to be insensitive to howspecies are grouped according to other functional traits oraccording to their habitat preference (Yang et al 2008) Thatthis phenomenology reflects a biomass allocation trade-off isclear What remains uncertain is whether it is a simple trade-offbetween leaf mass investment and leaf size or a more complexset of relationships imposed by twig mass investments andhydraulics or biomechanics

Another trade-off that appears to hold across many but not allspecies is the relationship between leaf surface areaA and averageleaf mass mL For more than a few species-groupings thisrelationship is allometric with a slope that is less than onewhich indicates that increases in leaf mass investment do notresult in proportional increases in leaf area (ieAmL

lt10Niklaset al 2007) This trend of lsquodiminishing returnsrsquo has also beenreported when total leaf area per plant is plotted against total leafmass for plants differing in size (Niklas and Cobb 2008 2010)Several factorsmay contribute to these trends eg biomechanicalanalyses show that disproportionately larger investments must bemade to mechanically support photosynthetic tissues as leavesincrease in their surface area (Niklas et al 2009) In both the leafmass to stem mass and leaf mass to area relationships describedabove leaf shapevariation canprovide anextra degreeof freedomfor selection to act on at both the leaf and branch level

Case studies of leaf shape variation at different scales

Above we have discussed evolutionary and proximal drivers ofleaf shape as well as ecological correlates and functionalsignificance thereof Here we present case studies at threescales where we can examine the ecological and evolutionaryimplications of leaf shape These are the cross community scalecaseof leaf teeth as a subset of shapemodificationswithin lineage

200 bottomtop

150

100

50

0

0

08

07

06

05

04

03

5 10

Days after bud break

Leaf

siz

e (c

m2 )

Leaf

lobi

ng in

dex

(cm

2 cm

ndash2)

Days after bud break

15 20 25

0 5 10 15 20 25

Fig 5 Variation in (a) leaf size and (b) shape complexity inQuercus rubraleaves growing at the top and bottom of the tree crown Reprinted fromZwieniecki et al (2004b) with permission from John Wiley amp Sons

The evolution of leaf shape Functional Plant Biology 543

considerations of the genus Pelargonium and the familyProteaceae and finally the case of leaf shape variation within asingle genetic individual heteroblasty and heterophylly

Leaf teeth leaf shape variation at the community scale

Leaf teeth are conspicuous features inmanyplants and are notableas a leaf-shape trait that varies in predictable ways across plantcommunities Nearly 100 years ago Bailey and Sinnott (19151916) observed that plant communities from regions with lowmean annual temperature (MAT) have a higher site meanproportion of non-monocotyledonous species with toothedleaves Paleobotanists quickly seized upon this relationship asan index for inferring paleotemperature Little et al (2010)recently documented the rich interest in this topic compiling351 literature references related to leaf physiognomy andtemperature of which the primary focus is leaf teeth The linkbetween teeth and temperature is observed on all vegetatedcontinents (Greenwood 2005a Peppe Royer et al 2011) andis also seen in related metrics such as tooth size and tooth number(Huff et al 2003 Royer et al 2005 Peppe et al 2011) Furtherthese tooth variables can display phenotypic plasticity totemperature change (Fig 6 Royer et al 2009b)

What is knownabout the functional bases of these leafndashclimaterelationships Leaf teeth commonly expand and mature fasterthan the bulk leaf (eg Billings 1905 Feild et al 2005) and youngteeth are disproportionately vigorous in their gas exchange(Baker-Brosh and Peet 1997 Royer and Wilf 2006) Fromthese observations Royer and Wilf (2006) hypothesised thatone role of teeth is to increase sap flow thereby deliveringnutrients and other solutes to young emerging leaves Inhabitats with low temperatures teeth could play an importantrole in maximising the potential for carbon gain and growth earlyin the season In warmer climates with longer growing seasonsthe carbon benefit conferred by teeth would be outweighed bytheir unavoidable water cost Alternatively Feild et al (2005)hypothesised that teeth help remove freezendashthaw embolismsthrough positive root pressure and guttation Importantly thetransport of fluids is common to both hypotheses From this itmay be surmised that water-limiting conditions select againstteeth (Bailey andSinnott 1916) Indeed this pattern is clearly seenacross local water availability gradients (Burnham et al 2001

Kowalski and Dilcher 2003 Greenwood 2005b Royer et al2009a Peppe et al 2011) but not globally against mean annualprecipitation (eg Peppe et al 2011)

If the hypothesis by Royer andWilf (2006) is correct toothedspecies should have a higher photosynthetic rate at least at thebeginning of the growing season As such teethmay be related tothe leaf economics spectrum (Wright et al 2004) Some studiesfind support for this idea leaves with a short leaf lifespan(deciduous) (Peppe et al 2011) and low leaf mass per area(Royer et al 2005) two leaf economic variables are morelikely to be toothed It also follows that teeth should be morefunctionally related to growing season climatic variables likeseasonality growing season length and growing degree days notmean annual temperature however few studies find support forthis idea (Royer et al 2005 Peppe et al 2011) Ultimately itwould be valuable to have a better understanding of the geneticprocesses underlying tooth development but presently little isknown outside model species (see lsquoGenetic regulation forpatterning of leaf margin serration and leaflet formationrsquosection) Further phylogenetic history has typically beenassumed to have only a minor role in these leaf-climatepatterns but rigorous tests demonstrate that this assumption isfalse (Little et al 2010) This outcome is of particular concern forpaleobotanists because systematic placement of fossils can bedifficult especially for the Cretaceous and Paleogene Tools toaccommodate for this phylogenetic dependency are needed

Leaf shape variation within lineages

Proteaceae

Proteaceae is an ancient angiosperm family originating inGondwana over 100million year ago and possessing 1700 extantspecies (Weston 2007) The familyrsquos greatest representation is inAustralia followed by SouthAfrica with smaller numbers acrossother Gondwanan segments (Cowling and Lamont 1998 Hootand Douglas 1998 Weston 2007) Within the Proteaceae thereexists an enormous range of leaf sizes and shapes ranging froma few mm2 to 500 cm2 and including entire toothed broadlylobed deeply dissected needle-like simple or compound leavesIntriguingly most of this shape diversity is represented inAustralia The majority of taxa outside Australia possessentire leaves the exception being four South African genera

(a) (b) (c) (d )

Fig 6 Representative leaves of Acer rubrum showing genetic variation and plasticity in leaf teeth Leaf derived from coolclimate (Ontario) seed stock grown in (a) Rhode Island (cool climate) and (b) Florida (warm climate) leaf derived fromwarm climate (Florida) seed stock grown in (c) Rhode Island and (d) Florida Bar (for all leaves) = 1 cm Reprinted from Royeret al (2009) under Creative Commons Attribution Licence

544 Functional Plant Biology A B Nicotra et al

that possess dissected (Serruria and Paranomis) or toothed(Mimites and Leucospermum) leafed species all of whichoccur in a single clade (Sauquet et al 2009)

The reason for the extensive radiation in leaf shape diversityin Australian Proteaceae is unclear Among South AfricanProteaceae leaf size decreases with mean annual temperatureand narrow leaves have been shown to overheat less than theirwider counterparts (Yates et al 2010) Given that leaf dissectionand small size are functionally equivalent with respect to heatloss (see lsquoTemperature and waterrsquo section) it is tempting tosuggest that leaf shape variation in Australia represents anadaptive response to increasing hot dry conditions Yet thereappears to be no consistent distributional pattern of Proteaceaeleaf dimensions inAustraliaManymediumor large un-dissectedleaves exist in the hottest habitats Grevilleas with small simpleleaves can be found across most regions including high rainfallareas and some of the species found in the shaded rainforestunderstorey have among the most deeply dissected leaves of allLeaf shape also varies within a single individual For examplefern-leaf stenocarpus Stenocarpus davallioides Foreman ampBHyland has tripinnatisect juvenile leaves but adult leavescan be pinnate or bipinnate (Foreman and Hyland 1995)Other rainforest species such as Atherton Oak (Athertoniadiversifolia (CTWhit) LAS Johnson amp BGBriggs) havevery large lobed intermediate leaves but adult and seedlingleaves are small and entire (Weston 1995) Alternatively leaflobing within a plant can be random having no obvious spatialor developmental association or homogenous across a speciesExamining a phylogeny of extant taxa one finds lobed ordissected leaves alongside entire-leafed species ndash across mostlineages in the group (Weston 2007) Therefore althoughsuggestions of microclimatic adaptation may be offered atthe species level it seems unlikely that an environmentalassociation with leaf shape could be generalised acrossAustralian Proteaceae Rather leaf shape variation in theProteaceae is likely to reflect the range of ecological solutionsto leaf structurefunction trade-offs discussed above

Pelargonium

Similar to the Proteaceae the genus Pelargonium isdistributed predominantly in the southern hemisphere withnearly 80 of its almost 300 species found in South AfricaAlthough the age of the genus is estimated to be around30million years (Bakker et al 2005) the highest rate ofspecies accumulation has occurred in the last 10million years(Martinez-Cabrera 2010) The combination of climatic andedaphic heterogeneity over short distances and limited geneflow (Latimer et al 2005 Linder 2005) is frequently offeredas an explanation for the extraordinary plant species diversity thatcharacterises southern Africa Pelargonium the third largestgenus in the region is characterised by remarkable variation ingrowth form that ranges from annuals and geophytes to shrubsVariation in leaf shape is no less dramatic ranging from entireto highly dissected (Fig 1) often within growth forms andtaxonomic sections Parsing of leaf form categorically revealedthat the extent of dissection of the blade is highly evolvablewithinanoverall ovate leaf shapeoutline that is evolutionarily conserved(Jones et al 2009)

In addition to high evolutionary lability of leaf dissectionthere is variation among major clades in the evolutionary patternof major veins One major clade (Clade C) is dominated bypalmate venation whereas another (Clade A) is dominated bypinnate venation although both types are found in both cladesWe note that pinnate venation is significantly associated withincreased dissection and reductions in functional leaf sizecharacteristic of the highly diverse lsquoxerophyticrsquo clade (CladeA2) of the winter rainfall region

To examine whether leaf shape translates into functionalvariation carbon gain and water use efficiency were measuredin pairs of species contrasted for leaf shape and replicated acrossthree subclades of Clade A and one subclade of the third majorclade CladeB (Nicotra et al 2008) In each case the species withthemore dissected leaves showed higher rates of carbon gain butalso higher rates of water loss in all treatment conditions Thatpaper argued that the differences in carbon gain were not directresults of leaf shape per se but rather that the same conditionsthat result in selection for more dissected leaves also favourthe evolution of high photosynthetic rates high leaf nitrogencontent and opportunistic use of water when available Evenin this study however there was no correlation between leafdissection and the climate present in native ranges suggestingeither that climate is selective at a much smaller scale (ie themicrohabitat) or that for any given set of climate variablesmultiple leaf shapes can be adaptive in association withvariation in other leaf traits

Heteroblasty and heterophylly variationwithin a single genetic individual

We have so far focussed primarily on cross species variation inleaf shape As discussed for the Proteaceae genera above (lsquoTheProteaceaersquo section) however a single genome can produce aremarkable range of leaf shapes during the lifetime of a plantThere are two broad categories of leaf shape variation that can beexpressed by a single individual This variation can arise either inresponse to different external environments or to ontogeneticsignals (some of which may be influenced by environment aswell)

Environmentally induced heterophylly

Environmentally induced heterophylly results when leafshape responds to environmental cues The classic case is thedramatically different emergent and submergent leaves in aquaticplants Although many aquatic plants are also heteroblastic (seebelow Sculthorpe 1967) environmentally induced heterophyllyis not specific to position or plant age and is reversible dependinglargely on the environmental signals (mediated by hormonalsignals) perceived by cells of the developing leaf primordiumThe developmental age of the leaf at the time of signal perceptionis critical Very young primordia (L1ndashL3) generally respondcompletely to the novel environment whereas primordiathat were slightly older at the time of environmental changeshow intermediate features as mature leaves Characteristics ofspecific cells are related to the position of that cell within thedifferentiation gradient characteristic of that species at the timeof environmental change (eg Goliber and Feldman 1990 Bruniet al 1996 Kuwabara and Nagata 2006)

The evolution of leaf shape Functional Plant Biology 545

Among terrestrial plants the lsquosunndashshadersquo leaf response isfrequently cited as the classic example of environmentallyinduced heterophylly Historically differences in leaf formwithin a canopy have been assumed to be responses to alteredlight environments (eg larger and less-lobed leaves are producedin shade)However these different leaf shapes also confer distincthydraulic advantages (see lsquoTemperature and waterrsquo section)Whether in response to light or to hydraulic limitation lsquosunshadersquo heterophylly is accompanied by similar and well-knownanatomical differences across a wide range of species (ie sunleaves are generally smaller and thicker with smaller cells thanshade leaves Taiz and Zeiger 2002) However when theseanatomical differences become determined developmentally(Dengler 1980 Yano and Terashima 2004) has been lesscommonly studied The presence of intermediate anatomicalcharacteristics in leaves transferred among light environmentsshortly after expansion ceased suggests that leaves of somespecies have the capacity to respond anatomically to alteredlight levels until very late in development (Oguchi et al2005) In contrast recent studies suggest that systemic lightand humidity signals sent from older to younger leavesallow younger leaves to develop phenotypes suited for theiranticipated environments (Lake et al 2001 Yano andTerashima 2001 Thomas et al 2004) Exactly when final leafshape is determined in lsquosunshadersquo leaves remains an openquestion for most species

Heteroblasty describes ontogenetically dependentchanges in leaf features most notably shape

In 1900 Goebel originally distinguished differences in lsquojuvenileformrsquo from lsquoadult formrsquo on the basis of dramatic differences inleaf shape and associated traits (Balfour 1969) As an example he

described the shift from compound leaves to phyllodes in Acaciaas heteroblastic and reserved the term homoblastic for the lessdramatic variation exhibited by Casurina It is now generallyaccepted that heteroblasty refers to the full suite of features thatchange along the shoot during development (Jones 1999)

Heteroblasty results from the interaction of developmentalprograms expressed at both the leaf and the shoot- or whole-plant level (Kerstetter and Poethig 1998 Tsukaya et al 2000Usami et al 2009 Wu et al 2009 Chuck and OrsquoConnor 2010)Heteroblasty is generally not reversible unless reiterative growthoccurs (ie following damage) Morphological differencescharacteristic of heteroblasty such as differences in lobing areusually expressed relatively early in development shortly afterleaf initiation (Jones 1995)

The sequence of heteroblastic leaf shapes produced alongthe shoot can be accelerated or delayed in response to differentgrowth environments or architectural manipulations (Ashby1948 Forster and Bonser 2009 Jaya et al 2010)Furthermore both heteroblasty and plasticity in heteroblastyhave been shown to be under strong genetic control (Anderson1989 Climent et al 2006) Forster et al (2011) suggested thatboth heteroblasty and plasticity in heteroblasty are likely to occurin response to lsquowhich leaf type confers the greatest benefit fora given light levelrsquo Several studies over the last decade havepointed towards functional consequences of heteroblastygenerally with reference to single environmental variables (egHansen 1986 Gamage and Jesson 2007 Kubien et al 2007)It is likely that heteroblasty influences multiple leaf functionsbecause suites of leaf traits change with heteroblasty (Jones2001) Indeed this conclusion was supported in a recent studythat examined differences inmore than a hundred leaf traits in koa(Acacia koa AGray Pasquet-Kok et al 2010) Heteroblasty hasalso been proposed as an adaptation to herbivory (Karban and

Table 1 Promising directions for cross-disciplinary investigation of leaf shape

No Direction

1 A multi-species lsquoevo-devorsquo approach similar to that by Illing et al (2009) is likely to be informative with regard to many enigmatic features regardingthe evolution of diverse leaf forms and more generally provides an exciting system for examining the convergence and divergence in the evolutionof function

2 A phylogenetic analysis of leaf shape examining correlated evolution with other traits such as venation size leaf surface characteristics and branch andarchitectural measures would enable a thorough investigation of the situations in which leaf shape diversification arises

3 Further research into both the genetic and developmental processes underlying lobes and teeth and the ecological correlates of each would provide anexcellent opportunity to link across wide environmental scales

4 Cross species studies ideally in the field of the relative importance of leaf shape among other factors in determining leaf temperature and water status arenecessary to test the conclusion put forward here that the functional significance of leaf shape lies more often in water relations than directly in leafthermoregulation

5 Studies of the scale of environmental change that selects for ontogenetic response in leaf shape in heteroblastic species fromdiverse environments will shedlight on links between genes development and ecology

6 Detailed quantifications of the extent towhich leaf shape differswithin canopies particularly of dioecious specieswhere selection on reproductive functioncan lead to dimorphic conditions will help link leaf shape to whole plant processes

7 Studies at the scale of transects across the leaf surface that simultaneously consider hydraulic resistance temperature and light will reveal the extent oflocalised hydraulic and thermal stresses within leaf surfaces that might act as selective forces on leaf shape per se Imaging approaches incorporatingmeasures of 3-dimensional shape temperature and fluorescence provide exciting opportunities for examining spatial heterogeneity in temperature andphysiological processes

8 Since patterns of leaf venation affect both hydraulic supply and capacity for photosynthate transport studies that simultaneously examine both willreveal the extent to which interactions among these fundamental processes either co-determine leaf shape or establish trade-offs ultimately reflectedin leaf shape

9 Finally studies examining the fitness values of different leaf shapes or plasticity in leaf shapes under a range of environmental conditions are needed tobetter identify which aspects of leaf shape are most likely to be adaptive

546 Functional Plant Biology A B Nicotra et al

Thaler 1999 Brennan et al 2001 Fadzly et al 2009) and it hasbeen associated with differences in frost resistance (Darrow et al2001) It is worth noting that studies of the functional significanceof heteroblasty have focussed on relatively few plant lineagesGiven that heteroblasty occurs widely among the angiospermsas well as in some ferns and gymosperms (eg Wagner 1952Mueller 1982 Pryer and Hearn 2009) the next questions mayfocus on the cost of heteroblasty and its fitness consequenceswhat scale of environmental change selects for ontogeneticresponse in leaf shape

Synthesis and next steps

Whether considered at community lineage or single genomescales leaf shape is a trait that is highly labile and responsive to arange of biotic and abiotic factors Although some leaf traitscan be explained in terms of particularly strong trade-offsbetween two or more factors that result in clear scalingrelationships (eg leaf mass-to-area traits or leaf mass-to-stemmass) leaf shape in contrast does not emerge from the aboveconsideration as a factor capable of explaining ecologicaldifferentiation between species Rather leaf shape emerges asa trait for which there are many quite varied functional trade-offs

As stated above these trade-offs for successful functionalityof leaves may generate ecological opportunities leaf shape isperhaps best viewed not as a single major axis but rather asan option that fine tunes the leaf to its conditions over bothshort and evolutionary time spans Numerous elements of thisreview support this conclusion Over evolutionary history theextant range of variation in leaf shape has been explored bymanylineages although only the angiosperms have explored thisrange with the added element of massive increases in veindensities Likewise our growing understanding of geneticcontrols on leaf shape reveal that it is a complex traitdetermined by many different genes at several differentdevelopmental stages Although some genes have conservedaffects on leaf morphology including shape the exact natureof their affects can vary widely Likewise the function to whichagivenelement of leaf shape lsquois putrsquo canvary For example closerinvestigation of the internal architecture and developmentalpathways associated with different leaf shapes suggests thatleaf teeth and leaf lobes often have distinctly differentfunctional foundations and may play different functions indifferent environments

Despite amyriadof functional roles the one recurrent theme inour review is that leaf shape is affected by and strongly influencesleaf water relations From a functional perspective the role of leafshape in leaf thermoregulation has perhaps received the mostattention in previous considerations of the functional significanceof leaf shape The relationship between leaf size and temperaturebeing physically determined by basic energy balance principlesthis was a logical starting point of enquiry But as we haveshown the connection between leaf shape (and size) and leaftemperatures under field conditions is not well supportedempirically Rather leaf shape variation as distinct from sizemore often will reflect water supply trade-offs High veindensities support high photosynthetic rates (Brodribb et al2010) and dissected leaves maintain a greater proportion ofleaf tissue closer to main veins Leaf temperature plays a role

in this story as higher leaf temperatures lead to greaterevaporative water loss but many factors other than leaf shapeinfluence leaf temperature (eg leaf thickness orientation andreflectivity and branch and plant architecture) Finally it is alsoworth noting that although there appears to be an obviouslink between shape and water supply there is a concomitantinteraction between water supply and phloem function thatremains relatively unexplored

Thus our review suggests several angles for further research(Table 1) In particular we advocate those approaches thatenable investigation across the scales examined here (genedevelopment function and evolutionary ecology) and thatfocus in particular on the links among leaf shape developmentand water relations

Acknowledgements

The authors thank the Botanical Society of America for their support of thesymposiumfromwhich this reviewaroseWealso thankLawrenSackand twoanonymous reviewers for useful comments on an earlier version of thismanuscript

References

Ackerly DD Knight CA Weiss SB Barton K Starmer KP (2002) Leafsize specific leaf area and microhabitat distribution of chaparral woodyplants contrasting patterns in species level and community level analysesOecologia 130 449ndash457 doi101007s004420100805

Anderson S (1989) Variation in heteroblastic succession among populationsof Crepis tectorum Nordic Journal of Botany 8 565ndash573 doi101111j1756-10511989tb01730x

Armbruster WS Muchhala N (2009) Associations between floralspecialization and species diversity cause effect or correlationEvolutionary Ecology 23 159ndash179 doi101007s10682-008-9259-z

Armbruster WS Di Stilio VS Tuxill JD Flores TC Runk JLV (1999)Covariance and decoupling of floral and vegetative traits in nineneotropical plants a re-evaluation of Bergrsquos correlation-pleiadesconcept American Journal of Botany 86 39ndash55 doi1023072656953

Ashby E (1948) Studies in the morphogenesis of leaves I An essay on leafshape New Phytologist 47 153ndash176 doi101111j1469-81371948tb05098x

Aso K Kato M Banks JA Hasebe M (1999) Characterization ofhomeodomain-leucine zipper genes in the fern Ceratopteris richardiiand the evolution of the homeodomain-leucine zipper gene family invascular plants Molecular Biology and Evolution 16 544ndash552

Bailey IW Sinnott EW (1915) A botanical index of Cretaceous and Tertiaryclimates Science 41 831ndash834 doi101126science411066831

Bailey IW Sinnott EW (1916) The climatic distribution of certain types ofangiosperm leaves American Journal of Botany 3 24ndash39 doi1023072435109

Baker-Brosh KF Peet RK (1997) The ecological significance of lobed andtoothed leaves in temperate forest trees Ecology 78 1250ndash1255

Bakker FT Culham A Marais EM Gibby M (2005) Nested radiation inCape Pelargonium In lsquoPlant species-level systematics new perspectiveson pattern and process Vol 143rsquo (Eds FT Bakker LW Chartrou BGravendeel PB Pielser) pp 75ndash100 (ARG Ganter Verlag KG RuggellLichtstein)

Balfour IB (Transl) (1969) lsquoOrganography of plants I General organographyrsquoby K Goebel 1900 (Hafner Publishing Co Inc New York)

Ball MC Cowan IR Farquhar GD (1988) Maintenance of leaf temperatureand the optimization of carbon gain in relation to water loss in a tropicalmangrove forest Australian Journal of Plant Physiology 15 263ndash276doi101071PP9880263

The evolution of leaf shape Functional Plant Biology 547

A brief history of the angiosperm leaf and shapediversity thereinTo understand the evolution of angiosperm leaf shape someperspective on the evolution of the leaf itself is needed as leavesof many shapes have evolved numerous times independentlyin evolutionary history For the purpose of this review ofangiosperm leaf shape we define a leaf as a vascularasymmetric appendicular structure initiated at the shoot apicalmeristem This definition excludes the functionally analogousstructures of mosses and leafy liverworts Morphologicallysimple leaves microphylls arose in the lycopsids Largerleaves with more complex shapes and venation networkscalled megaphylls arose in the seed plants in the extinctprogymnosperm and sphenophyll lineages and in one ormore lineages of lsquofernsrsquo (Galtier 1981 Kenrick and Crane1997 Niklas 1997 Boyce and Knoll 2002 Tomescu 2009Boyce 2010) Thus the seed plants represent just one ofseveral independent evolutionary origins of large leaves withdiverse shapes among the vascular plants

Despite the many independent origins of leaves several traitsnow associatedwith leaves can be considered homologous acrossall vascular plants through a shared monoplyletic origin from aleafless common ancestor First all vascular plant leaves aredistinct from the leaf-like organs of bryophytes that rely onexternal surfaces for photosynthetic gas exchange (as wouldalso be true of the external appendages independently derivedin many algal lineages Niklas 2000) Vascular plant leaves useinternal airspaces regulated by stomata for gas exchange (Raven1996 Boyce 2008a) Second evidence from both living plantsand the venation patterns of fossil leaves indicates that theancestral form of tissue production in the growing leaves of allvascular plants was limited to discrete marginal zones of growth(Pray 1960 Zurakowski and Gifford 1988 Boyce and Knoll2002 Boyce 2007) Other aspects of leaf organography suchas the evolution of differentiated abaxialadaxial domains (seesection on lsquoProximal mechanisms underlying leaf shapediversityrsquo below) and determinate leaf growth appear to ariseindependently in seed plants and other lineages however severalof these lineages have co-opted the same underlying geneticpathways to regulate these developmental processes (seesection on lsquoProximal mechanisms underlying leaf shapediversityrsquo Bharathan et al 2002 Harrison et al 2005 Sanderset al 2007 Tomescu 2009 Boyce 2010)

Within this broader evolutionary context angiosperm leavesrepresent a rangeof particularly aberrant variants of vascular plantleaves Some of the distinctive characteristics of angiospermleaves have arisen independently in other lineages whereasother traits are unique to the flowering plants Angiospermshave evolved leaf growth that occurs diffusely throughout theleaf without being limited to the margin (Pray 1955 Poethig andSussex 1985a 1985b) thus enabling the enormous diversity andcomplexity of angiosperm leaf venation patterns and shapes Thephylogenetic distribution of complex venation patterns suggestthat the evolutionary departures from marginal growth observedin the angiosperms have also arisen repeatedly in two or threeother seed plant lineages and 10 or more groups of ferns (Boyce2005) In these likewise diffuse growth and complex venationpatterns can be seen to occur with diverse leaf shapes Despitethese similarities to the other seed plants angiosperm leaf

evolution remains distinctive in that it ran counter to thegeneral temporal trend seen in the vasculature of other seedplants The vascular network of other seed plant lineagesevolved towards simplification After initially exhibiting thecomplete range of morphologies consistent with marginalgrowth including many involving two vein orders with amidvein and open or reticulate networks of secondary veins(morphologies now common only among the ferns) the leafarchitecture in the other major vascular plant lineages becameprogressively limited to a narrow subset of forms In cycadsconifers Ginkgo and several other extinct seed plant lineages asingle order of veins lead straight parallel courses to a distalmarginAngiosperms however exhibitmany hierarchical ordersof reticulate internally directed veins (Boyce 2005) Furtherangiosperm leaves are unique in their possession ofextraordinarily high vein densities (Fig 2) whereas all otherplants living or extinct average ~2mm of vein length per squaremm of leaf surface angiosperms average between 8 and10mmmmndash2 and can range to above 20mmmmndash2 (Boyce2008a 2008b 2009 Brodribb and Feild 2010 Brodribb et al2010)

Together these changes that accompanied angiospermevolution represent a substantial revision of the functionalpossibilities of a leaf First the release of venation patterns fromstrictly marginal growth allows the production of novel leafshapes and sizes that would not otherwise be possible Seconddifferences in angiosperm venation allowed a complete revolutionin how water is distributed within a leaf (Boyce 2008a) In anyleaf the tissue furthest from the leaf base has access only to thatwater which has not been lost to transpiration in more proximaltissues (Zwieniecki et al 2004a 2006) Reticulate venation andthe minute size of the final order of veins in flowering plantsleads to equitable distribution of water between successive veinorders (Zwieniecki et al 2002) Finally the high vein densities ofangiosperms shorten the path length along xylem and mesophyllthus enabling assimilation and transpiration capacities higherthan in any other group of plants (Bond 1989 Sack and Frole2006 Brodribb et al 2007 Boyce et al 2009) The uniquely hightranspiration capacities of angiosperms may also have influencedthe evolution of leaf shape by relaxing some of the thermalconstraints on leaf size and shape (see lsquoTemperature and waterrsquosection)

It remains controversial when and in which lineages thesecharacteristics of angiosperm leaves arose The distinctivenessof angiosperms leaves has led to polarised interpretations oftheir ancestry Some researcher propose that similarities inleaf venation tie the early angiosperms to particular fossil seedplants (eg Melville 1969) but the group of plants with themost similar venation are the clearly unrelated dipterid fernsThe various seed plant lineages most often favoured asangiosperm relatives based upon reproductive characteristics(Doyle and Hickey 1976 Hilton and Bateman 2006 Frohlichand Chase 2007) tend to have very dissimilar leaves Otherssuggest that angiosperm leaves are so distinct as to require acomplete reinvention of a leaf after passing through a leaflessintermediate such as one that was aquatic- or desert-adapted(Doyle and Hickey 1976) But the appearance of angiosperm-like leaf traits in a variety of fern lineages suggest thatinvoking a leafless intermediate is unnecessary Although the

The evolution of leaf shape Functional Plant Biology 537

history of early angiosperm leaf evolution is still debated weare certain that there have been numerous independent originsof diverse leaf shapes and that leaf shape diversificationis associated with both high vein densities and reticulatevenation patterns

Proximal mechanisms underlying leaf shape diversity

What are the genetic mechanisms underlying this leaf shapediversification and how similar are they given the many

independent origins of diverse leaf shapes What we know ofthe processes of leaf development (and leaf shape determination)is largely a product of studies on model species systems Inparticular Arabidopsis thaliana L (Heynh) (hereafter referredto as Arabidopsis) has been widely used for the identificationof genes determining leaf form In the following sections weexamine the development of the angiosperm leaf specificallythe establishment of dorsiventral domains and flatteningexpansion of the lamina and formation of leaf margin anddiscuss how these factors interact to determine leaf shape

(a) (b)

(c)

Fig 2 Leaf venation architecture in (a) the cycadZamia furfuracea (b) the fernBlechnumgibbum and (c) theangiosperm Boehmeria nivea overlain with the fern Polypodium formosanum Both the Zamia and theBlechnum have marginally ending veins suggesting marginal growth but the single order of parallel veins inZamiamust distribute water over more than 10 cm of length whereas the finest veins in theBlechnum are a fewmm The finest veins distributing water in angiosperms can be even shorter and the density of veins muchhigher the Boehmeria has ~8mm of vein mmndash2 of leaf versus 1mmmmndash2 in the Polypodium

538 Functional Plant Biology A B Nicotra et al

Establishment of dorsiventral domainsand lamina flattening

Since leaf primordia arise as protrusions from shoot apicalmeristems (SAM) suppression of the regular SAMdevelopmental program is the first step required for normaldevelopment of a leaf primordium Recently Sarojam et al(2010) found that genes in the YABBY family suppress SAMidentity and promote lamina growth in Arabidopsis In mostbifacial leaves such as leaves of Arabidopsis and snapdragon(Antirrhinum majus L) the radial primordium differentiatesinto abaxial (lower) and adaxial (upper) surfaces to establishdorsiventrality (Waites and Hudson 1995) Our currentunderstanding is that the molecular genetic regulation ofidentification of the abaxial and adaxial domains depends onactions of both small RNAs and directindirect targets suchas class III Zip genes KANADI genes and AUXIN RESPONSEFACTORS3 (ARF3)ETT in Arabidopsis (Husbands et al 2009Floyd and Bowman 2010 Kidner 2010) Without establishmentof the adaxial and abaxial domains it is believed that leavescannot expand in any angiosperm species rather they becomestalk like since the so-called plate meristem is never activated toexpand the laminas (Fig 3)

The monocot clade is typified by many unifacial leaves thathave only abaxial identity in the lamina as seen in leek (Alliumampeloprasum var porrum L) and Iris spp Leek leaves arecylindrical since they lack adaxial identity in their leaf bladeIris in contrast has flat leaves that expand in a distinct

perpendicular direction relative to that of normal bifacialleaves Likewise the rush Juncus prismatocarpus RBr hasunifacial flat leaves the primordium of the leaf blade growsinto a flat lamina as a result of extensive thickening growthThis developmental pathway is dependent on the geneDROOPING LEAF (DL Yamaguchi et al 2010 Fig 3) thatis also a member of the YABBY gene family but with a monocot-specific function As seen in this case diversified morphology ina particular taxon sometimes depends on some taxon-specificmolecular mechanisms that are modifications of gene functionfrom ancestral lineages

Mechanisms for expansion of lamina

Once dorsiventral identity is generated the leaf lamina can thendiversify in morphology to form simple or compound leaveswith entire or serrated margins differing in thickness length andwidth and in orientation (Tsukaya 2006) The leaf length widthratio is regulated by polar-dependent cell expansion and cellproliferationdistribution Several key genes for its regulationhave been identified from Arabidopsis ANGUSTIFOLIA (AN)and ROTUNDIFOLIA3 (ROT3) regulate the shape of cellswhereas ANGUSTIFOLIA3 (AN3) and ROTUNDIFOLIA4(ROT4) affect the number of cells in the lamina (Tsukaya2006) Loss-of-function mutations of AN and AN3 result innarrower leaves whereas loss-of-function of ROT3 oroverexpression of ROT4 make leaves shorter In addition

Fig 3 Twodifferentmechanismsof laminagrowth Inmost leaves such as leaves ofArabidopsis (upper left)dorsiventral identity (adaxial upper side (shown in green in colour version online) and abaxial lower side(shown in blue in colour version)) identities are established Flattening lamina growth occurs along the junctionbetween adaxial and abaxial domains as a result of activity of the plate meristem Loss of either abaxial oradaxial identity results in radialised growth (lower right) Stick-like leaf morphology seen in leek and Juncuswallichianus (centre) is the typical example of unifacial leaves that have no adaxial domain In the monocotclade however some species develop flat lamina irrespective of lack of adaxial identity in the lamina (rightSisyrinchium rosulatum) that is supported by enhanced growth in the leaf-thickness direction Meristematicactivity occurs in center region of both leaves (orange in colour version) relative position of the shoot apicalmeristem (SAM) is shown by the cross-hatched dot (green in colour version)

The evolution of leaf shape Functional Plant Biology 539

LONGIFOLIA1 (LON1) and LON2 regulate the length of the leaflamina in Arabidopsis (Lee et al 2006) A vast array of genes isknown to influence leaf area (Horiguchi et al 2006) For exampleheteroblastic change in leaf area is governed by the miR156-SQUAMOSAPROMOTERBINDINGPROTEIN-LIKE (SPL)-miR172 pathway via regulation of cell size and cell numbers inArabidopsis (Usami et al 2009)Alterations of activities of anyofthese factors could contribute to natural variation in leaflength width ratio and area None of these genetic variationsresult in loss of reproductive structures suggesting that thesegenes could be under direct environmentally mediated selectionAs yet we do not know whether the above patterns hold in othernon-model species

Leaves also vary in the extent to which the lamina is flator curved CINCINATA (CIN) of snapdragon was oncethought to function to keep the leaf flat by orchestrating cellproliferation along the medio-lateral axis of the leaf primordia(Nath et al 2003) More recently however analyses of theTCP genes which are homologues of CIN in Arabidopsisindicate that the major role of the TCP is in promotion ofmaturation in laminar tissues not in regulation of cellproliferation (Efroni et al 2008 Sarojam et al 2010) The roleof CIN in snapdragon may also be the same as that of TCPsin Arabidopsis The other candidate genetic mechanism forcontrol of leaf curvature is an auxin-related pathway Severalhyponastic leaf mutants have been isolated from Arabidopsisand many of the responsible genes are related to auxinsignalling (Perez-Perez et al 2002 2009) Since auxinsignalling is also very important for venation patterningspecies-specific modification of auxin signalling in leafprimordia may have caused diversification of both leafcurvature and vasculature patterning

Genetic regulation for patterning of leaf marginserration and leaflet formation

Variation in leaf shape as distinct from length and area orcurvature arises from variation in leaf margin and lobeformation Kawamura et al (2010) showed that spatialpatterning of serrations or leaf teeth (see lsquoLeaf shape variationwithin lineagesrsquo section) on the leaf margin depends onformation of auxin maxima on the leaf margin in ArabidopsisLocation of these maxima is governed by polar auxin transportand the stabilisation of auxinmaxima ismaintained by prolongedexpression of CUP-SHAPED COTYLEDON2 (CUC2) around it(Kawamura et al 2010) The mechanism of rhythmic patternformation of the auxin maximaminima on the leaf margin isthought to involve self-organising pattern formationmechanismsby polar auxin transport similar to that involved in patternformation of phyllotaxy (de Reuille et al 2006 Jonsson et al2006) Either the loss of distinct auxinmaxima or an instability ofthe auxin maxima will result in an entire leaf margin (Kawamuraet al 2010) Changes in functions of these genes howeverseverely influence many biological processes including floralorgan development as seen in the pin-formed 1 mutant (Okadaet al 1991) Thus variation among species in pattern of marginserration might accompany variation in floral organs or morelikely reflect differential regulation for example by downstreamgenetic pathways

Combined control of auxin-maxima formation by PIN andCUC genes as described for serration above is also known tohave an important role in leaflet patterning in bittercress(Cardamine flexuosa With) an Arabidopsis relative withcompound leaves (Barkoulas et al 2008) Again this controlvaries among species in tomato (Solanum lycopersicum L) lossof CUC3 function does not convert compound leaves intosimple leaves (Blein et al 2008) Thus it is clear that theindependent evolution of compound leaves from simple leavesis likely driven by species-specific mechanisms (Efroni et al2010) though these may reflect modifications of similarpathways

Recent studies have identified several important key genes forregulation of compound versus simple leaf forms For exampleabsence or prolonged expression of class I Knotted1-likehomeobox (KNOX) genes in leaf primordia is important tocompound leaf formation in a wide range of angiosperms andgymnosperms (Bharathan et al 2002) In some legume specieshowever the LEAFY (LFY)UNIFOLIATA (UNI) gene is the keyfor compound leaf formation instead of the class I KNOX genes(Hofer et al 1997) Changes in these genes could be linked tothe evolution of compound leaves although loss-of-function ofLFYUNI also causes infertility in Arabidopsis and thus it isunlikely that selection directly on these particular genes couldresult in theevolutionof variation in leaf shapeDiversified spatialpatterning of petiolepetiolule and laminaleaflet in compoundleaves have also been suggested to be linked to the distributionpattern of AS1-ROUGH SHEATH2-PHANTASTICA (ARP)expression domains (Kim et al 2003) However based on thefact that the ARP gene has only a limited role in compound leafformation in some species it is unclear whetherARPwill turn outto be major factor in others (Efroni et al 2010)

As described above the independent evolutionary origins ofseveral angiosperm leaf features should caution against theassumption that a single ancestral genetic system underlies leafdevelopment The developmental genetic approaches outlinedhave revealed several key mechanisms for leaf shape controlbut have also demonstrated variation in the function of thesegenes among lineages In a few studies researchers have gonebeyond one or two species studies and have looked for deeperevolutionary links (Illing et al 2009) Studies of the role of theclass 1 KNOX genes have considered not just angiospermsbut other lineages Among simple-leafed dicots and monocots(eg Arabidopsis and Zea) KNOX genes are strongly expressedduring the indeterminate growth of the apical meristem but arerapidly downregulated in the cells flanking the apical domein regions defined by leaf primordia Although some arginine-rich protein (ARP) genes repress the expression of KNOXtranscription factors during the development of both micro-and megaphyll primordia (Harrison et al 2005) the role ofhomeodomain-leucine zipper (HD-ZIP) III genes which areinvolved in the specification of foliar adaxialabaxial fatediffers in the development of these two leaf types (Floyd andBowman 2006) HD-ZIP III genes have been identified in themoss Physcomitrella (Sakakibara et al 2001) and in the fernCeratopteris (Aso et al 1999) they are also expressed duringnormal vascular tissue development in Arabidopsis (Zhong andYe 1999) suggesting that they are very ancient but alsomultifunctional

540 Functional Plant Biology A B Nicotra et al

Accumulating knowledge on molecular genetic mechanismsof leaf morphogenesis is leading to an increasingly evo-devoapproach to studies of leaf diversity (Tsukaya 2010)this progress is aided by establishing new model species(eg Cardamine hirsuta and the rush Canales et al 2010Yamaguchi and Tsukaya 2010) Over the next decade weanticipate that genomic approaches on multiple speciesanalysed in phylogenetic frameworks (eg Illing et al 2009)will resolvemuchof theenigma regarding the evolutionofdiverseleaf forms

Genetic constraints on flower petal and leaf form

One additional developmental consideration is that as famouslydescribed by von Goethe in 1790 flower petals are derivatives ofthe apical meristem and serial homologues to leaves (von Goethe2009) Thus one might hypothesise that although the evolutionof leaves substantially predates that of flowers selection on theform of either of these organs would also result in changes inthe other such that selection on leaf shape and size subsequent tothe evolution of floral organs may be constrained by pleiotropiccontrol of floral form Recent genetic data identifies several genesthat have roles in both leaf and flower form (Dinneny et al 2004Schmid et al 2005 Street et al 2008) For example geneticchanges in both leaf length to width ratio (eg rot3) and leaf size(eg an3) influence not only leaf proportions but also that of thefloral organs in Arabidopsis (Kim et al 1999 Horiguchi et al2005) The garden pea (Pisum sativum L) is a good case in pointas discussed above compound leaf formation depends on LFYUNI but the same gene is also required for flowering Thus ifcurrent selection favoured simple leaves in garden pea changes inleaf shape could be constrained by pleiotropic control offlowering (Hofer et al 1997) Likewise Shalit et al (2009)recently reported that the action of florigen ndash a plant hormonefor flowering ndash influences complexity of leaves by changingnumbers of leaflets in tomato Such relationships suggest thatsome floral characters as well as other metabolic pathways maynowbeclosely linkedwith regulationof leaf shapeand sizeA fewecological studies have also assessed correlations between petaland leaf form (Berg 1960 Armbruster et al 1999) This questionis of particular interest in lineages where animal pollination isassociated with increased diversification rates either as a result ofdirect selection or as a secondary reinforcing factor (van der Nietet al 2006 Armbruster and Muchhala 2009 Kay and Sargent2009)

In situations where leaf and petal form are under similargenetic control and both under selection one of two thingscould happen If leaf shape does not have substantialfunctional impact then one might expect leaf form to changewith selection on petal form However given the likelyimportance of leaf form to photosynthetic function it is morelikely that a breakdown in genetic correlations between leafand flower form (eg by alterations of downstream regulatorypathways) would evolve Several factors argue against thehypothesis that leaf shape is frequently a product of selectionon flower form First diverse leaf shapes were in evidencelong before the origin of flowers in the angiosperms Secondthe genetic mechanisms underlying leaf shape revealconsiderable evolvability with similar genes being co-opted

for different functions in different species Finally given thepotential importance of both petal and leaf shape we suggestthat genetic constraints on the independent evolution of bothstructures are likely to have broken down in those lineages withhighly diversified leaf shapes over the many millions of yearssince petals first evolved

Ecological correlates and the functional significanceof leaf shape

Evolutionary history plays a part in determining leafmorphology but our consideration of leaf shape evolution andits developmental genetic basis demonstrates that diverselineages have repeatedly gained and lost many leaf featuresThus developmental and genetic constraints on the evolutionof leaf form and shape in particular have been overcomemany times or have led to new evolutionary opportunities Itis of particular interest then to ask what are the likely ecologicaldrivers andor functional significance of leaf shape

Temperature and water

Conventionalwisdomwould say that thermoregulation is perhapsthe main driver of leaf shape evolution The maintenance ofleaf temperatures within certain limits is critical for a plantrsquosgrowth and survival and although photosynthetic tissue of somespecies canwithstand temperatures below6C (Ball et al 2006)and above 60C (Clum1926Nobel 1988) irreversible damage toleaves can occur at temperatures well within this range (Levitt1980 Jones 1992 Ball et al 2004 Groom et al 2004 Sharkey2005) Leaf size and shape potentially have large effects onleaf temperature because the two-dimensional proportions of aflattened leaf determine the rate of heat transfer between acrossthe leafndashair interface by influencing the thickness of its boundarylayer

All else being equal the thickness of a boundary layerincreases with length from the windward edge so that heatconvection from small leaves is more rapid than from largeleaves (Raschke 1960 Gates 1968 Vogel 1970 Parkhurst andLoucks 1972Grace et al 1980Geller andSmith 1982Monteithand Unsworth 1990 Schuepp 1993) Likewise because leaflobing reduces the distance across the lamina the rate of heattransfer is predicted to be greater in a lobed leaf than an unlobedleaf with equivalent area (Parkhurst et al 1968 Vogel 1968Lewis 1972 Givnish 1978 Gurevitch and Schuepp 1990a)

The rate of heat transfer from lobed metal plates or lobedleaves coated in metal is greater than from shallow-lobed or un-lobed ones (Parkhurst et al 1968 Thom 1968 Vogel 1970Gottschlich and Smith 1982 Gurevitch and Schuepp 1990bRoth-Nebelsick 2001) Artificial or coated leaves have beenused because they enable boundary layer resistance to beinvestigated in the absence of transpiration However thisapproach is biologically flawed because it does not enable anassessment of the relative importance of a range of traits (eg leafwater content absorbance) in determining the actual operatingtemperatures of leaves in the field

Actual leaf temperature measurements show that large leavescan operate at below-ambient temperatures even on hot days asa result of high rates of latent heat loss through transpiration(Drake et al 1970 Gates 1980 Hegazy and El Amry 1998)

The evolution of leaf shape Functional Plant Biology 541

Thus conditions of evaporative demand and the availability ofsoil water for transpiration are key to understanding leafshapendashtemperature relationships under biologically realisticconditions Because of their much greater transpirationalcapacities (see above) flowering plants have far greaterleeway than other plants regarding the size and shape of theirleaves (see lsquoA brief history of the angiosperm leaf and shapediversity thereinrsquo section Boyce et al 2009)

In addition to reducing heat loads the morphology of deeplylobed leaves may also reflect direct selection for increasedhydraulic efficiency Water moving through a plant encountersnumerous resistances on its path from soil to roots to stemsthrough leaves and out to the external atmosphere Within theliquid phase of water flow through a plant leaf hydraulicresistance Rleaf accounts for more than 30 of total plantresistance (Sack and Holbrook 2006) The high resistance inleaves is due first to a system of veins of decreasing size theresistance of which is inversely proportional to the fourth powerof the radius of the component conduits in these veins Thereforeas a proportion of the considerable resistance in the leaf(Zwieniecki et al 2002) the minor veins provide the greatestresistance to water flow (Sack et al 2004)Water also encountershigh resistance as it exits veins laterally and travels to othercells mainly via apoplastic pathways through the mesophyllAccordingly on the order of half of total Rleaf can occur outsidethe major veins (Sack and Holbrook 2006) As resistanceincreases along the pathway towards sites of evaporationlocalised water potential (Y) becomes progressively morenegative (Brodribb et al 2010) The result of how thisphenomenon might affect a leaf has long been observedregions of leaves lying furthest from the main supply channelsare the first to lsquowitherrsquo when exposed to strong winds (Fig 4Yapp 1912) Leaf lobing represents an effective removal of thispotentially stress-prone tissue and has been suggested as anadaptation to dry conditions (Thoday 1931 Givnish 1979) Ifdeeply lobed leaves have a lower ratio of mesophyll tissue tolarge highly conductive veins they should then have reducedresistance relative to less- or un-lobed leaves (Sack and Tyree2005)

Although leaf lobing might confer adaptive benefits withrespect to maintaining stable hydraulic supply it may also bethat the extent of lobing or margin dissection is itself determinedby hydraulic limitation (Sack et al 2003 Zwieniecki et al2004b Boyce 2009) For species with lsquosunrsquo and lsquoshadersquoleaves the morphology of small deeply lobed sun leaves atthe outer canopy is proposed to result from water pressuredrop across the leaf lamina caused by reduced water deliveryto expanding cells during growth (Fig 5 Zwieniecki et al 2004bBoyce 2009 Leigh et al 2011)Alternatively recent studies havesuggested that many of the characteristics of lsquosunrsquo and lsquoshadersquoleaves are the result of branch autonomy within the plant canopy(Niklas and Cobb 2010)

Thus although the effects of leaf shape on leaf thermalregulation hold when tested on model leaves predictions areless straightforward when hydraulic function is accounted forAdditionally consideration of the relative contribution of a rangeof leaf and branch properties to leaf temperature is critical It islikely that leaf shape is only one amongmany factors influencingleaf thermal regulation other factors could include water

content leaf thickness spectral reflectance orientation andplant architecture Further given the observation of high leafshape diversity withinmany plant communities from hot and aridenvironments it seems likely that regulating leaf temperature isnot the singlemost important evolutionary influence on leaf shapediversity The association between leaf shape and hydraulicproperties which is affected by leaf temperature in contrast islikely to be of greater importance

Leaf structure and functional trade-offs

Many foliar leaf traits including shape reflect functional trade-offs that have been resolved in various ways by differentspecies depending on their ecological settings (eg hydro- vsmesophytes) physiological attributes (eg C3 vsC4metabolism)developmental capabilities (eg heterobaric vs non-heterobaricleaves) or evolutionary histories (eg micro- vs megaphylls)Life history and optimisation theory show that the number ofphenotypic solutions that allow for different equally successfultrait combinations increases as the number of trade-offsincreases ndash a conclusion that applies to traits within the leaf(eg for shape) as well as to leafndashbranch relationships (Niklas1988)

It is evident that leaves perform several functionssimultaneously that is they intercept light transport liquidsand solutes exchange gases with the atmosphere dissipateheat cope with externally applied mechanical forces and

(a)

(e)(f )

( i )

(g)

(h)

(b)

(c)

(d )

Fig 4 Localised lsquowitheringrsquo of leaves exposed to strong wind Reprintedfrom Yapp (1912) with permission from Oxford University Press

542 Functional Plant Biology A B Nicotra et al

defend themselves from herbivores and pathogens Importantlymany of these functions have conflicting physiologicalanatomical or morphological requirements For exampleorientations of lamina that maximise light interception alsominimise the ability to dissipate heat Further all of thesefunctions are intrinsically size dependent For example thequantity of mechanical tissues required to support laminaedepends on the mass of laminae whereas the ability tointercept light depends on laminae area

When seen in this light the trade-offs required for thesuccessful functionality of foliar leaves generate rather thanconstrain ecological and evolutionary opportunities This canbe illustrated mathematically and empirically in the context ofphyllotaxy and its consequences on light interception Computersimulations of mathematically generated shoots to assess theinfluence of leaf shape size and phyllotactic patterns on theability to intercept direct solar radiation show that differences inphyllotaxy significantly influence light interception (particularlyfor shoots with a rosette growth habit) They also show thatcomparatively small differences in leaf shape can compensatefor the negative effects of leaf overlap resulting from virtuallyany phyllotactic pattern For example lobed leaves or pinnifid

compound leaves facilitate light penetration through shootsbearing densely packed leaves Thus phyllotaxy can beviewed as a developmental limiting factor that can drivecompensatory changes in morphological features such asshape that are not directly controlled by patterns of leafinitiation In this way the lsquoconstraintsrsquo imposed by phyllotaxycan foster evolutionary modifications at the level of the entireshoot (Niklas 1988)

One particularly important trade-off operating at the level ofthe architecture of a shoot is that between leaf size and numberFor any given leaf biomass allocation an individual shoot caneither have a few large leaves or many smaller ones Acrossdifferent herbaceous species and juvenile specimens of treespecies drawn from diverse lineages and ecological settingsthe relationship between total leaf biomass per plant (ML) andtotal stemmass per plant (MS) fallswithin a comparatively narrowcorridor of variation that obeys a positive isometric scalingrelationship (ie MLMS

~10 Niklas 2004) Accordinglyacross these species any stem mass investment yields a moreor less proportional increase in leaf mass Therefore it followsmathematically that across these species the relationshipamong the average mass of a typical leaf mL total leaf numberNL and total stem mass complies with the proportionality mL

NLMS~10 When individual shoots are sampled across

diverse species the relationship between mL and NL per shootis negative and isometric (ie mLNL

ndash1 Kleiman and Aarssen2007) a relationship that appears to be insensitive to howspecies are grouped according to other functional traits oraccording to their habitat preference (Yang et al 2008) Thatthis phenomenology reflects a biomass allocation trade-off isclear What remains uncertain is whether it is a simple trade-offbetween leaf mass investment and leaf size or a more complexset of relationships imposed by twig mass investments andhydraulics or biomechanics

Another trade-off that appears to hold across many but not allspecies is the relationship between leaf surface areaA and averageleaf mass mL For more than a few species-groupings thisrelationship is allometric with a slope that is less than onewhich indicates that increases in leaf mass investment do notresult in proportional increases in leaf area (ieAmL

lt10Niklaset al 2007) This trend of lsquodiminishing returnsrsquo has also beenreported when total leaf area per plant is plotted against total leafmass for plants differing in size (Niklas and Cobb 2008 2010)Several factorsmay contribute to these trends eg biomechanicalanalyses show that disproportionately larger investments must bemade to mechanically support photosynthetic tissues as leavesincrease in their surface area (Niklas et al 2009) In both the leafmass to stem mass and leaf mass to area relationships describedabove leaf shapevariation canprovide anextra degreeof freedomfor selection to act on at both the leaf and branch level

Case studies of leaf shape variation at different scales

Above we have discussed evolutionary and proximal drivers ofleaf shape as well as ecological correlates and functionalsignificance thereof Here we present case studies at threescales where we can examine the ecological and evolutionaryimplications of leaf shape These are the cross community scalecaseof leaf teeth as a subset of shapemodificationswithin lineage

200 bottomtop

150

100

50

0

0

08

07

06

05

04

03

5 10

Days after bud break

Leaf

siz

e (c

m2 )

Leaf

lobi

ng in

dex

(cm

2 cm

ndash2)

Days after bud break

15 20 25

0 5 10 15 20 25

Fig 5 Variation in (a) leaf size and (b) shape complexity inQuercus rubraleaves growing at the top and bottom of the tree crown Reprinted fromZwieniecki et al (2004b) with permission from John Wiley amp Sons

The evolution of leaf shape Functional Plant Biology 543

considerations of the genus Pelargonium and the familyProteaceae and finally the case of leaf shape variation within asingle genetic individual heteroblasty and heterophylly

Leaf teeth leaf shape variation at the community scale

Leaf teeth are conspicuous features inmanyplants and are notableas a leaf-shape trait that varies in predictable ways across plantcommunities Nearly 100 years ago Bailey and Sinnott (19151916) observed that plant communities from regions with lowmean annual temperature (MAT) have a higher site meanproportion of non-monocotyledonous species with toothedleaves Paleobotanists quickly seized upon this relationship asan index for inferring paleotemperature Little et al (2010)recently documented the rich interest in this topic compiling351 literature references related to leaf physiognomy andtemperature of which the primary focus is leaf teeth The linkbetween teeth and temperature is observed on all vegetatedcontinents (Greenwood 2005a Peppe Royer et al 2011) andis also seen in related metrics such as tooth size and tooth number(Huff et al 2003 Royer et al 2005 Peppe et al 2011) Furtherthese tooth variables can display phenotypic plasticity totemperature change (Fig 6 Royer et al 2009b)

What is knownabout the functional bases of these leafndashclimaterelationships Leaf teeth commonly expand and mature fasterthan the bulk leaf (eg Billings 1905 Feild et al 2005) and youngteeth are disproportionately vigorous in their gas exchange(Baker-Brosh and Peet 1997 Royer and Wilf 2006) Fromthese observations Royer and Wilf (2006) hypothesised thatone role of teeth is to increase sap flow thereby deliveringnutrients and other solutes to young emerging leaves Inhabitats with low temperatures teeth could play an importantrole in maximising the potential for carbon gain and growth earlyin the season In warmer climates with longer growing seasonsthe carbon benefit conferred by teeth would be outweighed bytheir unavoidable water cost Alternatively Feild et al (2005)hypothesised that teeth help remove freezendashthaw embolismsthrough positive root pressure and guttation Importantly thetransport of fluids is common to both hypotheses From this itmay be surmised that water-limiting conditions select againstteeth (Bailey andSinnott 1916) Indeed this pattern is clearly seenacross local water availability gradients (Burnham et al 2001

Kowalski and Dilcher 2003 Greenwood 2005b Royer et al2009a Peppe et al 2011) but not globally against mean annualprecipitation (eg Peppe et al 2011)

If the hypothesis by Royer andWilf (2006) is correct toothedspecies should have a higher photosynthetic rate at least at thebeginning of the growing season As such teethmay be related tothe leaf economics spectrum (Wright et al 2004) Some studiesfind support for this idea leaves with a short leaf lifespan(deciduous) (Peppe et al 2011) and low leaf mass per area(Royer et al 2005) two leaf economic variables are morelikely to be toothed It also follows that teeth should be morefunctionally related to growing season climatic variables likeseasonality growing season length and growing degree days notmean annual temperature however few studies find support forthis idea (Royer et al 2005 Peppe et al 2011) Ultimately itwould be valuable to have a better understanding of the geneticprocesses underlying tooth development but presently little isknown outside model species (see lsquoGenetic regulation forpatterning of leaf margin serration and leaflet formationrsquosection) Further phylogenetic history has typically beenassumed to have only a minor role in these leaf-climatepatterns but rigorous tests demonstrate that this assumption isfalse (Little et al 2010) This outcome is of particular concern forpaleobotanists because systematic placement of fossils can bedifficult especially for the Cretaceous and Paleogene Tools toaccommodate for this phylogenetic dependency are needed

Leaf shape variation within lineages

Proteaceae

Proteaceae is an ancient angiosperm family originating inGondwana over 100million year ago and possessing 1700 extantspecies (Weston 2007) The familyrsquos greatest representation is inAustralia followed by SouthAfrica with smaller numbers acrossother Gondwanan segments (Cowling and Lamont 1998 Hootand Douglas 1998 Weston 2007) Within the Proteaceae thereexists an enormous range of leaf sizes and shapes ranging froma few mm2 to 500 cm2 and including entire toothed broadlylobed deeply dissected needle-like simple or compound leavesIntriguingly most of this shape diversity is represented inAustralia The majority of taxa outside Australia possessentire leaves the exception being four South African genera

(a) (b) (c) (d )

Fig 6 Representative leaves of Acer rubrum showing genetic variation and plasticity in leaf teeth Leaf derived from coolclimate (Ontario) seed stock grown in (a) Rhode Island (cool climate) and (b) Florida (warm climate) leaf derived fromwarm climate (Florida) seed stock grown in (c) Rhode Island and (d) Florida Bar (for all leaves) = 1 cm Reprinted from Royeret al (2009) under Creative Commons Attribution Licence

544 Functional Plant Biology A B Nicotra et al

that possess dissected (Serruria and Paranomis) or toothed(Mimites and Leucospermum) leafed species all of whichoccur in a single clade (Sauquet et al 2009)

The reason for the extensive radiation in leaf shape diversityin Australian Proteaceae is unclear Among South AfricanProteaceae leaf size decreases with mean annual temperatureand narrow leaves have been shown to overheat less than theirwider counterparts (Yates et al 2010) Given that leaf dissectionand small size are functionally equivalent with respect to heatloss (see lsquoTemperature and waterrsquo section) it is tempting tosuggest that leaf shape variation in Australia represents anadaptive response to increasing hot dry conditions Yet thereappears to be no consistent distributional pattern of Proteaceaeleaf dimensions inAustraliaManymediumor large un-dissectedleaves exist in the hottest habitats Grevilleas with small simpleleaves can be found across most regions including high rainfallareas and some of the species found in the shaded rainforestunderstorey have among the most deeply dissected leaves of allLeaf shape also varies within a single individual For examplefern-leaf stenocarpus Stenocarpus davallioides Foreman ampBHyland has tripinnatisect juvenile leaves but adult leavescan be pinnate or bipinnate (Foreman and Hyland 1995)Other rainforest species such as Atherton Oak (Athertoniadiversifolia (CTWhit) LAS Johnson amp BGBriggs) havevery large lobed intermediate leaves but adult and seedlingleaves are small and entire (Weston 1995) Alternatively leaflobing within a plant can be random having no obvious spatialor developmental association or homogenous across a speciesExamining a phylogeny of extant taxa one finds lobed ordissected leaves alongside entire-leafed species ndash across mostlineages in the group (Weston 2007) Therefore althoughsuggestions of microclimatic adaptation may be offered atthe species level it seems unlikely that an environmentalassociation with leaf shape could be generalised acrossAustralian Proteaceae Rather leaf shape variation in theProteaceae is likely to reflect the range of ecological solutionsto leaf structurefunction trade-offs discussed above

Pelargonium

Similar to the Proteaceae the genus Pelargonium isdistributed predominantly in the southern hemisphere withnearly 80 of its almost 300 species found in South AfricaAlthough the age of the genus is estimated to be around30million years (Bakker et al 2005) the highest rate ofspecies accumulation has occurred in the last 10million years(Martinez-Cabrera 2010) The combination of climatic andedaphic heterogeneity over short distances and limited geneflow (Latimer et al 2005 Linder 2005) is frequently offeredas an explanation for the extraordinary plant species diversity thatcharacterises southern Africa Pelargonium the third largestgenus in the region is characterised by remarkable variation ingrowth form that ranges from annuals and geophytes to shrubsVariation in leaf shape is no less dramatic ranging from entireto highly dissected (Fig 1) often within growth forms andtaxonomic sections Parsing of leaf form categorically revealedthat the extent of dissection of the blade is highly evolvablewithinanoverall ovate leaf shapeoutline that is evolutionarily conserved(Jones et al 2009)

In addition to high evolutionary lability of leaf dissectionthere is variation among major clades in the evolutionary patternof major veins One major clade (Clade C) is dominated bypalmate venation whereas another (Clade A) is dominated bypinnate venation although both types are found in both cladesWe note that pinnate venation is significantly associated withincreased dissection and reductions in functional leaf sizecharacteristic of the highly diverse lsquoxerophyticrsquo clade (CladeA2) of the winter rainfall region

To examine whether leaf shape translates into functionalvariation carbon gain and water use efficiency were measuredin pairs of species contrasted for leaf shape and replicated acrossthree subclades of Clade A and one subclade of the third majorclade CladeB (Nicotra et al 2008) In each case the species withthemore dissected leaves showed higher rates of carbon gain butalso higher rates of water loss in all treatment conditions Thatpaper argued that the differences in carbon gain were not directresults of leaf shape per se but rather that the same conditionsthat result in selection for more dissected leaves also favourthe evolution of high photosynthetic rates high leaf nitrogencontent and opportunistic use of water when available Evenin this study however there was no correlation between leafdissection and the climate present in native ranges suggestingeither that climate is selective at a much smaller scale (ie themicrohabitat) or that for any given set of climate variablesmultiple leaf shapes can be adaptive in association withvariation in other leaf traits

Heteroblasty and heterophylly variationwithin a single genetic individual

We have so far focussed primarily on cross species variation inleaf shape As discussed for the Proteaceae genera above (lsquoTheProteaceaersquo section) however a single genome can produce aremarkable range of leaf shapes during the lifetime of a plantThere are two broad categories of leaf shape variation that can beexpressed by a single individual This variation can arise either inresponse to different external environments or to ontogeneticsignals (some of which may be influenced by environment aswell)

Environmentally induced heterophylly

Environmentally induced heterophylly results when leafshape responds to environmental cues The classic case is thedramatically different emergent and submergent leaves in aquaticplants Although many aquatic plants are also heteroblastic (seebelow Sculthorpe 1967) environmentally induced heterophyllyis not specific to position or plant age and is reversible dependinglargely on the environmental signals (mediated by hormonalsignals) perceived by cells of the developing leaf primordiumThe developmental age of the leaf at the time of signal perceptionis critical Very young primordia (L1ndashL3) generally respondcompletely to the novel environment whereas primordiathat were slightly older at the time of environmental changeshow intermediate features as mature leaves Characteristics ofspecific cells are related to the position of that cell within thedifferentiation gradient characteristic of that species at the timeof environmental change (eg Goliber and Feldman 1990 Bruniet al 1996 Kuwabara and Nagata 2006)

The evolution of leaf shape Functional Plant Biology 545

Among terrestrial plants the lsquosunndashshadersquo leaf response isfrequently cited as the classic example of environmentallyinduced heterophylly Historically differences in leaf formwithin a canopy have been assumed to be responses to alteredlight environments (eg larger and less-lobed leaves are producedin shade)However these different leaf shapes also confer distincthydraulic advantages (see lsquoTemperature and waterrsquo section)Whether in response to light or to hydraulic limitation lsquosunshadersquo heterophylly is accompanied by similar and well-knownanatomical differences across a wide range of species (ie sunleaves are generally smaller and thicker with smaller cells thanshade leaves Taiz and Zeiger 2002) However when theseanatomical differences become determined developmentally(Dengler 1980 Yano and Terashima 2004) has been lesscommonly studied The presence of intermediate anatomicalcharacteristics in leaves transferred among light environmentsshortly after expansion ceased suggests that leaves of somespecies have the capacity to respond anatomically to alteredlight levels until very late in development (Oguchi et al2005) In contrast recent studies suggest that systemic lightand humidity signals sent from older to younger leavesallow younger leaves to develop phenotypes suited for theiranticipated environments (Lake et al 2001 Yano andTerashima 2001 Thomas et al 2004) Exactly when final leafshape is determined in lsquosunshadersquo leaves remains an openquestion for most species

Heteroblasty describes ontogenetically dependentchanges in leaf features most notably shape

In 1900 Goebel originally distinguished differences in lsquojuvenileformrsquo from lsquoadult formrsquo on the basis of dramatic differences inleaf shape and associated traits (Balfour 1969) As an example he

described the shift from compound leaves to phyllodes in Acaciaas heteroblastic and reserved the term homoblastic for the lessdramatic variation exhibited by Casurina It is now generallyaccepted that heteroblasty refers to the full suite of features thatchange along the shoot during development (Jones 1999)

Heteroblasty results from the interaction of developmentalprograms expressed at both the leaf and the shoot- or whole-plant level (Kerstetter and Poethig 1998 Tsukaya et al 2000Usami et al 2009 Wu et al 2009 Chuck and OrsquoConnor 2010)Heteroblasty is generally not reversible unless reiterative growthoccurs (ie following damage) Morphological differencescharacteristic of heteroblasty such as differences in lobing areusually expressed relatively early in development shortly afterleaf initiation (Jones 1995)

The sequence of heteroblastic leaf shapes produced alongthe shoot can be accelerated or delayed in response to differentgrowth environments or architectural manipulations (Ashby1948 Forster and Bonser 2009 Jaya et al 2010)Furthermore both heteroblasty and plasticity in heteroblastyhave been shown to be under strong genetic control (Anderson1989 Climent et al 2006) Forster et al (2011) suggested thatboth heteroblasty and plasticity in heteroblasty are likely to occurin response to lsquowhich leaf type confers the greatest benefit fora given light levelrsquo Several studies over the last decade havepointed towards functional consequences of heteroblastygenerally with reference to single environmental variables (egHansen 1986 Gamage and Jesson 2007 Kubien et al 2007)It is likely that heteroblasty influences multiple leaf functionsbecause suites of leaf traits change with heteroblasty (Jones2001) Indeed this conclusion was supported in a recent studythat examined differences inmore than a hundred leaf traits in koa(Acacia koa AGray Pasquet-Kok et al 2010) Heteroblasty hasalso been proposed as an adaptation to herbivory (Karban and

Table 1 Promising directions for cross-disciplinary investigation of leaf shape

No Direction

1 A multi-species lsquoevo-devorsquo approach similar to that by Illing et al (2009) is likely to be informative with regard to many enigmatic features regardingthe evolution of diverse leaf forms and more generally provides an exciting system for examining the convergence and divergence in the evolutionof function

2 A phylogenetic analysis of leaf shape examining correlated evolution with other traits such as venation size leaf surface characteristics and branch andarchitectural measures would enable a thorough investigation of the situations in which leaf shape diversification arises

3 Further research into both the genetic and developmental processes underlying lobes and teeth and the ecological correlates of each would provide anexcellent opportunity to link across wide environmental scales

4 Cross species studies ideally in the field of the relative importance of leaf shape among other factors in determining leaf temperature and water status arenecessary to test the conclusion put forward here that the functional significance of leaf shape lies more often in water relations than directly in leafthermoregulation

5 Studies of the scale of environmental change that selects for ontogenetic response in leaf shape in heteroblastic species fromdiverse environments will shedlight on links between genes development and ecology

6 Detailed quantifications of the extent towhich leaf shape differswithin canopies particularly of dioecious specieswhere selection on reproductive functioncan lead to dimorphic conditions will help link leaf shape to whole plant processes

7 Studies at the scale of transects across the leaf surface that simultaneously consider hydraulic resistance temperature and light will reveal the extent oflocalised hydraulic and thermal stresses within leaf surfaces that might act as selective forces on leaf shape per se Imaging approaches incorporatingmeasures of 3-dimensional shape temperature and fluorescence provide exciting opportunities for examining spatial heterogeneity in temperature andphysiological processes

8 Since patterns of leaf venation affect both hydraulic supply and capacity for photosynthate transport studies that simultaneously examine both willreveal the extent to which interactions among these fundamental processes either co-determine leaf shape or establish trade-offs ultimately reflectedin leaf shape

9 Finally studies examining the fitness values of different leaf shapes or plasticity in leaf shapes under a range of environmental conditions are needed tobetter identify which aspects of leaf shape are most likely to be adaptive

546 Functional Plant Biology A B Nicotra et al

Thaler 1999 Brennan et al 2001 Fadzly et al 2009) and it hasbeen associated with differences in frost resistance (Darrow et al2001) It is worth noting that studies of the functional significanceof heteroblasty have focussed on relatively few plant lineagesGiven that heteroblasty occurs widely among the angiospermsas well as in some ferns and gymosperms (eg Wagner 1952Mueller 1982 Pryer and Hearn 2009) the next questions mayfocus on the cost of heteroblasty and its fitness consequenceswhat scale of environmental change selects for ontogeneticresponse in leaf shape

Synthesis and next steps

Whether considered at community lineage or single genomescales leaf shape is a trait that is highly labile and responsive to arange of biotic and abiotic factors Although some leaf traitscan be explained in terms of particularly strong trade-offsbetween two or more factors that result in clear scalingrelationships (eg leaf mass-to-area traits or leaf mass-to-stemmass) leaf shape in contrast does not emerge from the aboveconsideration as a factor capable of explaining ecologicaldifferentiation between species Rather leaf shape emerges asa trait for which there are many quite varied functional trade-offs

As stated above these trade-offs for successful functionalityof leaves may generate ecological opportunities leaf shape isperhaps best viewed not as a single major axis but rather asan option that fine tunes the leaf to its conditions over bothshort and evolutionary time spans Numerous elements of thisreview support this conclusion Over evolutionary history theextant range of variation in leaf shape has been explored bymanylineages although only the angiosperms have explored thisrange with the added element of massive increases in veindensities Likewise our growing understanding of geneticcontrols on leaf shape reveal that it is a complex traitdetermined by many different genes at several differentdevelopmental stages Although some genes have conservedaffects on leaf morphology including shape the exact natureof their affects can vary widely Likewise the function to whichagivenelement of leaf shape lsquois putrsquo canvary For example closerinvestigation of the internal architecture and developmentalpathways associated with different leaf shapes suggests thatleaf teeth and leaf lobes often have distinctly differentfunctional foundations and may play different functions indifferent environments

Despite amyriadof functional roles the one recurrent theme inour review is that leaf shape is affected by and strongly influencesleaf water relations From a functional perspective the role of leafshape in leaf thermoregulation has perhaps received the mostattention in previous considerations of the functional significanceof leaf shape The relationship between leaf size and temperaturebeing physically determined by basic energy balance principlesthis was a logical starting point of enquiry But as we haveshown the connection between leaf shape (and size) and leaftemperatures under field conditions is not well supportedempirically Rather leaf shape variation as distinct from sizemore often will reflect water supply trade-offs High veindensities support high photosynthetic rates (Brodribb et al2010) and dissected leaves maintain a greater proportion ofleaf tissue closer to main veins Leaf temperature plays a role

in this story as higher leaf temperatures lead to greaterevaporative water loss but many factors other than leaf shapeinfluence leaf temperature (eg leaf thickness orientation andreflectivity and branch and plant architecture) Finally it is alsoworth noting that although there appears to be an obviouslink between shape and water supply there is a concomitantinteraction between water supply and phloem function thatremains relatively unexplored

Thus our review suggests several angles for further research(Table 1) In particular we advocate those approaches thatenable investigation across the scales examined here (genedevelopment function and evolutionary ecology) and thatfocus in particular on the links among leaf shape developmentand water relations

Acknowledgements

The authors thank the Botanical Society of America for their support of thesymposiumfromwhich this reviewaroseWealso thankLawrenSackand twoanonymous reviewers for useful comments on an earlier version of thismanuscript

References

Ackerly DD Knight CA Weiss SB Barton K Starmer KP (2002) Leafsize specific leaf area and microhabitat distribution of chaparral woodyplants contrasting patterns in species level and community level analysesOecologia 130 449ndash457 doi101007s004420100805

Anderson S (1989) Variation in heteroblastic succession among populationsof Crepis tectorum Nordic Journal of Botany 8 565ndash573 doi101111j1756-10511989tb01730x

Armbruster WS Muchhala N (2009) Associations between floralspecialization and species diversity cause effect or correlationEvolutionary Ecology 23 159ndash179 doi101007s10682-008-9259-z

Armbruster WS Di Stilio VS Tuxill JD Flores TC Runk JLV (1999)Covariance and decoupling of floral and vegetative traits in nineneotropical plants a re-evaluation of Bergrsquos correlation-pleiadesconcept American Journal of Botany 86 39ndash55 doi1023072656953

Ashby E (1948) Studies in the morphogenesis of leaves I An essay on leafshape New Phytologist 47 153ndash176 doi101111j1469-81371948tb05098x

Aso K Kato M Banks JA Hasebe M (1999) Characterization ofhomeodomain-leucine zipper genes in the fern Ceratopteris richardiiand the evolution of the homeodomain-leucine zipper gene family invascular plants Molecular Biology and Evolution 16 544ndash552

Bailey IW Sinnott EW (1915) A botanical index of Cretaceous and Tertiaryclimates Science 41 831ndash834 doi101126science411066831

Bailey IW Sinnott EW (1916) The climatic distribution of certain types ofangiosperm leaves American Journal of Botany 3 24ndash39 doi1023072435109

Baker-Brosh KF Peet RK (1997) The ecological significance of lobed andtoothed leaves in temperate forest trees Ecology 78 1250ndash1255

Bakker FT Culham A Marais EM Gibby M (2005) Nested radiation inCape Pelargonium In lsquoPlant species-level systematics new perspectiveson pattern and process Vol 143rsquo (Eds FT Bakker LW Chartrou BGravendeel PB Pielser) pp 75ndash100 (ARG Ganter Verlag KG RuggellLichtstein)

Balfour IB (Transl) (1969) lsquoOrganography of plants I General organographyrsquoby K Goebel 1900 (Hafner Publishing Co Inc New York)

Ball MC Cowan IR Farquhar GD (1988) Maintenance of leaf temperatureand the optimization of carbon gain in relation to water loss in a tropicalmangrove forest Australian Journal of Plant Physiology 15 263ndash276doi101071PP9880263

The evolution of leaf shape Functional Plant Biology 547

history of early angiosperm leaf evolution is still debated weare certain that there have been numerous independent originsof diverse leaf shapes and that leaf shape diversificationis associated with both high vein densities and reticulatevenation patterns

Proximal mechanisms underlying leaf shape diversity

What are the genetic mechanisms underlying this leaf shapediversification and how similar are they given the many

independent origins of diverse leaf shapes What we know ofthe processes of leaf development (and leaf shape determination)is largely a product of studies on model species systems Inparticular Arabidopsis thaliana L (Heynh) (hereafter referredto as Arabidopsis) has been widely used for the identificationof genes determining leaf form In the following sections weexamine the development of the angiosperm leaf specificallythe establishment of dorsiventral domains and flatteningexpansion of the lamina and formation of leaf margin anddiscuss how these factors interact to determine leaf shape

(a) (b)

(c)

Fig 2 Leaf venation architecture in (a) the cycadZamia furfuracea (b) the fernBlechnumgibbum and (c) theangiosperm Boehmeria nivea overlain with the fern Polypodium formosanum Both the Zamia and theBlechnum have marginally ending veins suggesting marginal growth but the single order of parallel veins inZamiamust distribute water over more than 10 cm of length whereas the finest veins in theBlechnum are a fewmm The finest veins distributing water in angiosperms can be even shorter and the density of veins muchhigher the Boehmeria has ~8mm of vein mmndash2 of leaf versus 1mmmmndash2 in the Polypodium

538 Functional Plant Biology A B Nicotra et al

Establishment of dorsiventral domainsand lamina flattening

Since leaf primordia arise as protrusions from shoot apicalmeristems (SAM) suppression of the regular SAMdevelopmental program is the first step required for normaldevelopment of a leaf primordium Recently Sarojam et al(2010) found that genes in the YABBY family suppress SAMidentity and promote lamina growth in Arabidopsis In mostbifacial leaves such as leaves of Arabidopsis and snapdragon(Antirrhinum majus L) the radial primordium differentiatesinto abaxial (lower) and adaxial (upper) surfaces to establishdorsiventrality (Waites and Hudson 1995) Our currentunderstanding is that the molecular genetic regulation ofidentification of the abaxial and adaxial domains depends onactions of both small RNAs and directindirect targets suchas class III Zip genes KANADI genes and AUXIN RESPONSEFACTORS3 (ARF3)ETT in Arabidopsis (Husbands et al 2009Floyd and Bowman 2010 Kidner 2010) Without establishmentof the adaxial and abaxial domains it is believed that leavescannot expand in any angiosperm species rather they becomestalk like since the so-called plate meristem is never activated toexpand the laminas (Fig 3)

The monocot clade is typified by many unifacial leaves thathave only abaxial identity in the lamina as seen in leek (Alliumampeloprasum var porrum L) and Iris spp Leek leaves arecylindrical since they lack adaxial identity in their leaf bladeIris in contrast has flat leaves that expand in a distinct

perpendicular direction relative to that of normal bifacialleaves Likewise the rush Juncus prismatocarpus RBr hasunifacial flat leaves the primordium of the leaf blade growsinto a flat lamina as a result of extensive thickening growthThis developmental pathway is dependent on the geneDROOPING LEAF (DL Yamaguchi et al 2010 Fig 3) thatis also a member of the YABBY gene family but with a monocot-specific function As seen in this case diversified morphology ina particular taxon sometimes depends on some taxon-specificmolecular mechanisms that are modifications of gene functionfrom ancestral lineages

Mechanisms for expansion of lamina

Once dorsiventral identity is generated the leaf lamina can thendiversify in morphology to form simple or compound leaveswith entire or serrated margins differing in thickness length andwidth and in orientation (Tsukaya 2006) The leaf length widthratio is regulated by polar-dependent cell expansion and cellproliferationdistribution Several key genes for its regulationhave been identified from Arabidopsis ANGUSTIFOLIA (AN)and ROTUNDIFOLIA3 (ROT3) regulate the shape of cellswhereas ANGUSTIFOLIA3 (AN3) and ROTUNDIFOLIA4(ROT4) affect the number of cells in the lamina (Tsukaya2006) Loss-of-function mutations of AN and AN3 result innarrower leaves whereas loss-of-function of ROT3 oroverexpression of ROT4 make leaves shorter In addition

Fig 3 Twodifferentmechanismsof laminagrowth Inmost leaves such as leaves ofArabidopsis (upper left)dorsiventral identity (adaxial upper side (shown in green in colour version online) and abaxial lower side(shown in blue in colour version)) identities are established Flattening lamina growth occurs along the junctionbetween adaxial and abaxial domains as a result of activity of the plate meristem Loss of either abaxial oradaxial identity results in radialised growth (lower right) Stick-like leaf morphology seen in leek and Juncuswallichianus (centre) is the typical example of unifacial leaves that have no adaxial domain In the monocotclade however some species develop flat lamina irrespective of lack of adaxial identity in the lamina (rightSisyrinchium rosulatum) that is supported by enhanced growth in the leaf-thickness direction Meristematicactivity occurs in center region of both leaves (orange in colour version) relative position of the shoot apicalmeristem (SAM) is shown by the cross-hatched dot (green in colour version)

The evolution of leaf shape Functional Plant Biology 539

LONGIFOLIA1 (LON1) and LON2 regulate the length of the leaflamina in Arabidopsis (Lee et al 2006) A vast array of genes isknown to influence leaf area (Horiguchi et al 2006) For exampleheteroblastic change in leaf area is governed by the miR156-SQUAMOSAPROMOTERBINDINGPROTEIN-LIKE (SPL)-miR172 pathway via regulation of cell size and cell numbers inArabidopsis (Usami et al 2009)Alterations of activities of anyofthese factors could contribute to natural variation in leaflength width ratio and area None of these genetic variationsresult in loss of reproductive structures suggesting that thesegenes could be under direct environmentally mediated selectionAs yet we do not know whether the above patterns hold in othernon-model species

Leaves also vary in the extent to which the lamina is flator curved CINCINATA (CIN) of snapdragon was oncethought to function to keep the leaf flat by orchestrating cellproliferation along the medio-lateral axis of the leaf primordia(Nath et al 2003) More recently however analyses of theTCP genes which are homologues of CIN in Arabidopsisindicate that the major role of the TCP is in promotion ofmaturation in laminar tissues not in regulation of cellproliferation (Efroni et al 2008 Sarojam et al 2010) The roleof CIN in snapdragon may also be the same as that of TCPsin Arabidopsis The other candidate genetic mechanism forcontrol of leaf curvature is an auxin-related pathway Severalhyponastic leaf mutants have been isolated from Arabidopsisand many of the responsible genes are related to auxinsignalling (Perez-Perez et al 2002 2009) Since auxinsignalling is also very important for venation patterningspecies-specific modification of auxin signalling in leafprimordia may have caused diversification of both leafcurvature and vasculature patterning

Genetic regulation for patterning of leaf marginserration and leaflet formation

Variation in leaf shape as distinct from length and area orcurvature arises from variation in leaf margin and lobeformation Kawamura et al (2010) showed that spatialpatterning of serrations or leaf teeth (see lsquoLeaf shape variationwithin lineagesrsquo section) on the leaf margin depends onformation of auxin maxima on the leaf margin in ArabidopsisLocation of these maxima is governed by polar auxin transportand the stabilisation of auxinmaxima ismaintained by prolongedexpression of CUP-SHAPED COTYLEDON2 (CUC2) around it(Kawamura et al 2010) The mechanism of rhythmic patternformation of the auxin maximaminima on the leaf margin isthought to involve self-organising pattern formationmechanismsby polar auxin transport similar to that involved in patternformation of phyllotaxy (de Reuille et al 2006 Jonsson et al2006) Either the loss of distinct auxinmaxima or an instability ofthe auxin maxima will result in an entire leaf margin (Kawamuraet al 2010) Changes in functions of these genes howeverseverely influence many biological processes including floralorgan development as seen in the pin-formed 1 mutant (Okadaet al 1991) Thus variation among species in pattern of marginserration might accompany variation in floral organs or morelikely reflect differential regulation for example by downstreamgenetic pathways

Combined control of auxin-maxima formation by PIN andCUC genes as described for serration above is also known tohave an important role in leaflet patterning in bittercress(Cardamine flexuosa With) an Arabidopsis relative withcompound leaves (Barkoulas et al 2008) Again this controlvaries among species in tomato (Solanum lycopersicum L) lossof CUC3 function does not convert compound leaves intosimple leaves (Blein et al 2008) Thus it is clear that theindependent evolution of compound leaves from simple leavesis likely driven by species-specific mechanisms (Efroni et al2010) though these may reflect modifications of similarpathways

Recent studies have identified several important key genes forregulation of compound versus simple leaf forms For exampleabsence or prolonged expression of class I Knotted1-likehomeobox (KNOX) genes in leaf primordia is important tocompound leaf formation in a wide range of angiosperms andgymnosperms (Bharathan et al 2002) In some legume specieshowever the LEAFY (LFY)UNIFOLIATA (UNI) gene is the keyfor compound leaf formation instead of the class I KNOX genes(Hofer et al 1997) Changes in these genes could be linked tothe evolution of compound leaves although loss-of-function ofLFYUNI also causes infertility in Arabidopsis and thus it isunlikely that selection directly on these particular genes couldresult in theevolutionof variation in leaf shapeDiversified spatialpatterning of petiolepetiolule and laminaleaflet in compoundleaves have also been suggested to be linked to the distributionpattern of AS1-ROUGH SHEATH2-PHANTASTICA (ARP)expression domains (Kim et al 2003) However based on thefact that the ARP gene has only a limited role in compound leafformation in some species it is unclear whetherARPwill turn outto be major factor in others (Efroni et al 2010)

As described above the independent evolutionary origins ofseveral angiosperm leaf features should caution against theassumption that a single ancestral genetic system underlies leafdevelopment The developmental genetic approaches outlinedhave revealed several key mechanisms for leaf shape controlbut have also demonstrated variation in the function of thesegenes among lineages In a few studies researchers have gonebeyond one or two species studies and have looked for deeperevolutionary links (Illing et al 2009) Studies of the role of theclass 1 KNOX genes have considered not just angiospermsbut other lineages Among simple-leafed dicots and monocots(eg Arabidopsis and Zea) KNOX genes are strongly expressedduring the indeterminate growth of the apical meristem but arerapidly downregulated in the cells flanking the apical domein regions defined by leaf primordia Although some arginine-rich protein (ARP) genes repress the expression of KNOXtranscription factors during the development of both micro-and megaphyll primordia (Harrison et al 2005) the role ofhomeodomain-leucine zipper (HD-ZIP) III genes which areinvolved in the specification of foliar adaxialabaxial fatediffers in the development of these two leaf types (Floyd andBowman 2006) HD-ZIP III genes have been identified in themoss Physcomitrella (Sakakibara et al 2001) and in the fernCeratopteris (Aso et al 1999) they are also expressed duringnormal vascular tissue development in Arabidopsis (Zhong andYe 1999) suggesting that they are very ancient but alsomultifunctional

540 Functional Plant Biology A B Nicotra et al

Accumulating knowledge on molecular genetic mechanismsof leaf morphogenesis is leading to an increasingly evo-devoapproach to studies of leaf diversity (Tsukaya 2010)this progress is aided by establishing new model species(eg Cardamine hirsuta and the rush Canales et al 2010Yamaguchi and Tsukaya 2010) Over the next decade weanticipate that genomic approaches on multiple speciesanalysed in phylogenetic frameworks (eg Illing et al 2009)will resolvemuchof theenigma regarding the evolutionofdiverseleaf forms

Genetic constraints on flower petal and leaf form

One additional developmental consideration is that as famouslydescribed by von Goethe in 1790 flower petals are derivatives ofthe apical meristem and serial homologues to leaves (von Goethe2009) Thus one might hypothesise that although the evolutionof leaves substantially predates that of flowers selection on theform of either of these organs would also result in changes inthe other such that selection on leaf shape and size subsequent tothe evolution of floral organs may be constrained by pleiotropiccontrol of floral form Recent genetic data identifies several genesthat have roles in both leaf and flower form (Dinneny et al 2004Schmid et al 2005 Street et al 2008) For example geneticchanges in both leaf length to width ratio (eg rot3) and leaf size(eg an3) influence not only leaf proportions but also that of thefloral organs in Arabidopsis (Kim et al 1999 Horiguchi et al2005) The garden pea (Pisum sativum L) is a good case in pointas discussed above compound leaf formation depends on LFYUNI but the same gene is also required for flowering Thus ifcurrent selection favoured simple leaves in garden pea changes inleaf shape could be constrained by pleiotropic control offlowering (Hofer et al 1997) Likewise Shalit et al (2009)recently reported that the action of florigen ndash a plant hormonefor flowering ndash influences complexity of leaves by changingnumbers of leaflets in tomato Such relationships suggest thatsome floral characters as well as other metabolic pathways maynowbeclosely linkedwith regulationof leaf shapeand sizeA fewecological studies have also assessed correlations between petaland leaf form (Berg 1960 Armbruster et al 1999) This questionis of particular interest in lineages where animal pollination isassociated with increased diversification rates either as a result ofdirect selection or as a secondary reinforcing factor (van der Nietet al 2006 Armbruster and Muchhala 2009 Kay and Sargent2009)

In situations where leaf and petal form are under similargenetic control and both under selection one of two thingscould happen If leaf shape does not have substantialfunctional impact then one might expect leaf form to changewith selection on petal form However given the likelyimportance of leaf form to photosynthetic function it is morelikely that a breakdown in genetic correlations between leafand flower form (eg by alterations of downstream regulatorypathways) would evolve Several factors argue against thehypothesis that leaf shape is frequently a product of selectionon flower form First diverse leaf shapes were in evidencelong before the origin of flowers in the angiosperms Secondthe genetic mechanisms underlying leaf shape revealconsiderable evolvability with similar genes being co-opted

for different functions in different species Finally given thepotential importance of both petal and leaf shape we suggestthat genetic constraints on the independent evolution of bothstructures are likely to have broken down in those lineages withhighly diversified leaf shapes over the many millions of yearssince petals first evolved

Ecological correlates and the functional significanceof leaf shape

Evolutionary history plays a part in determining leafmorphology but our consideration of leaf shape evolution andits developmental genetic basis demonstrates that diverselineages have repeatedly gained and lost many leaf featuresThus developmental and genetic constraints on the evolutionof leaf form and shape in particular have been overcomemany times or have led to new evolutionary opportunities Itis of particular interest then to ask what are the likely ecologicaldrivers andor functional significance of leaf shape

Temperature and water

Conventionalwisdomwould say that thermoregulation is perhapsthe main driver of leaf shape evolution The maintenance ofleaf temperatures within certain limits is critical for a plantrsquosgrowth and survival and although photosynthetic tissue of somespecies canwithstand temperatures below6C (Ball et al 2006)and above 60C (Clum1926Nobel 1988) irreversible damage toleaves can occur at temperatures well within this range (Levitt1980 Jones 1992 Ball et al 2004 Groom et al 2004 Sharkey2005) Leaf size and shape potentially have large effects onleaf temperature because the two-dimensional proportions of aflattened leaf determine the rate of heat transfer between acrossthe leafndashair interface by influencing the thickness of its boundarylayer

All else being equal the thickness of a boundary layerincreases with length from the windward edge so that heatconvection from small leaves is more rapid than from largeleaves (Raschke 1960 Gates 1968 Vogel 1970 Parkhurst andLoucks 1972Grace et al 1980Geller andSmith 1982Monteithand Unsworth 1990 Schuepp 1993) Likewise because leaflobing reduces the distance across the lamina the rate of heattransfer is predicted to be greater in a lobed leaf than an unlobedleaf with equivalent area (Parkhurst et al 1968 Vogel 1968Lewis 1972 Givnish 1978 Gurevitch and Schuepp 1990a)

The rate of heat transfer from lobed metal plates or lobedleaves coated in metal is greater than from shallow-lobed or un-lobed ones (Parkhurst et al 1968 Thom 1968 Vogel 1970Gottschlich and Smith 1982 Gurevitch and Schuepp 1990bRoth-Nebelsick 2001) Artificial or coated leaves have beenused because they enable boundary layer resistance to beinvestigated in the absence of transpiration However thisapproach is biologically flawed because it does not enable anassessment of the relative importance of a range of traits (eg leafwater content absorbance) in determining the actual operatingtemperatures of leaves in the field

Actual leaf temperature measurements show that large leavescan operate at below-ambient temperatures even on hot days asa result of high rates of latent heat loss through transpiration(Drake et al 1970 Gates 1980 Hegazy and El Amry 1998)

The evolution of leaf shape Functional Plant Biology 541

Thus conditions of evaporative demand and the availability ofsoil water for transpiration are key to understanding leafshapendashtemperature relationships under biologically realisticconditions Because of their much greater transpirationalcapacities (see above) flowering plants have far greaterleeway than other plants regarding the size and shape of theirleaves (see lsquoA brief history of the angiosperm leaf and shapediversity thereinrsquo section Boyce et al 2009)

In addition to reducing heat loads the morphology of deeplylobed leaves may also reflect direct selection for increasedhydraulic efficiency Water moving through a plant encountersnumerous resistances on its path from soil to roots to stemsthrough leaves and out to the external atmosphere Within theliquid phase of water flow through a plant leaf hydraulicresistance Rleaf accounts for more than 30 of total plantresistance (Sack and Holbrook 2006) The high resistance inleaves is due first to a system of veins of decreasing size theresistance of which is inversely proportional to the fourth powerof the radius of the component conduits in these veins Thereforeas a proportion of the considerable resistance in the leaf(Zwieniecki et al 2002) the minor veins provide the greatestresistance to water flow (Sack et al 2004)Water also encountershigh resistance as it exits veins laterally and travels to othercells mainly via apoplastic pathways through the mesophyllAccordingly on the order of half of total Rleaf can occur outsidethe major veins (Sack and Holbrook 2006) As resistanceincreases along the pathway towards sites of evaporationlocalised water potential (Y) becomes progressively morenegative (Brodribb et al 2010) The result of how thisphenomenon might affect a leaf has long been observedregions of leaves lying furthest from the main supply channelsare the first to lsquowitherrsquo when exposed to strong winds (Fig 4Yapp 1912) Leaf lobing represents an effective removal of thispotentially stress-prone tissue and has been suggested as anadaptation to dry conditions (Thoday 1931 Givnish 1979) Ifdeeply lobed leaves have a lower ratio of mesophyll tissue tolarge highly conductive veins they should then have reducedresistance relative to less- or un-lobed leaves (Sack and Tyree2005)

Although leaf lobing might confer adaptive benefits withrespect to maintaining stable hydraulic supply it may also bethat the extent of lobing or margin dissection is itself determinedby hydraulic limitation (Sack et al 2003 Zwieniecki et al2004b Boyce 2009) For species with lsquosunrsquo and lsquoshadersquoleaves the morphology of small deeply lobed sun leaves atthe outer canopy is proposed to result from water pressuredrop across the leaf lamina caused by reduced water deliveryto expanding cells during growth (Fig 5 Zwieniecki et al 2004bBoyce 2009 Leigh et al 2011)Alternatively recent studies havesuggested that many of the characteristics of lsquosunrsquo and lsquoshadersquoleaves are the result of branch autonomy within the plant canopy(Niklas and Cobb 2010)

Thus although the effects of leaf shape on leaf thermalregulation hold when tested on model leaves predictions areless straightforward when hydraulic function is accounted forAdditionally consideration of the relative contribution of a rangeof leaf and branch properties to leaf temperature is critical It islikely that leaf shape is only one amongmany factors influencingleaf thermal regulation other factors could include water

content leaf thickness spectral reflectance orientation andplant architecture Further given the observation of high leafshape diversity withinmany plant communities from hot and aridenvironments it seems likely that regulating leaf temperature isnot the singlemost important evolutionary influence on leaf shapediversity The association between leaf shape and hydraulicproperties which is affected by leaf temperature in contrast islikely to be of greater importance

Leaf structure and functional trade-offs

Many foliar leaf traits including shape reflect functional trade-offs that have been resolved in various ways by differentspecies depending on their ecological settings (eg hydro- vsmesophytes) physiological attributes (eg C3 vsC4metabolism)developmental capabilities (eg heterobaric vs non-heterobaricleaves) or evolutionary histories (eg micro- vs megaphylls)Life history and optimisation theory show that the number ofphenotypic solutions that allow for different equally successfultrait combinations increases as the number of trade-offsincreases ndash a conclusion that applies to traits within the leaf(eg for shape) as well as to leafndashbranch relationships (Niklas1988)

It is evident that leaves perform several functionssimultaneously that is they intercept light transport liquidsand solutes exchange gases with the atmosphere dissipateheat cope with externally applied mechanical forces and

(a)

(e)(f )

( i )

(g)

(h)

(b)

(c)

(d )

Fig 4 Localised lsquowitheringrsquo of leaves exposed to strong wind Reprintedfrom Yapp (1912) with permission from Oxford University Press

542 Functional Plant Biology A B Nicotra et al

defend themselves from herbivores and pathogens Importantlymany of these functions have conflicting physiologicalanatomical or morphological requirements For exampleorientations of lamina that maximise light interception alsominimise the ability to dissipate heat Further all of thesefunctions are intrinsically size dependent For example thequantity of mechanical tissues required to support laminaedepends on the mass of laminae whereas the ability tointercept light depends on laminae area

When seen in this light the trade-offs required for thesuccessful functionality of foliar leaves generate rather thanconstrain ecological and evolutionary opportunities This canbe illustrated mathematically and empirically in the context ofphyllotaxy and its consequences on light interception Computersimulations of mathematically generated shoots to assess theinfluence of leaf shape size and phyllotactic patterns on theability to intercept direct solar radiation show that differences inphyllotaxy significantly influence light interception (particularlyfor shoots with a rosette growth habit) They also show thatcomparatively small differences in leaf shape can compensatefor the negative effects of leaf overlap resulting from virtuallyany phyllotactic pattern For example lobed leaves or pinnifid

compound leaves facilitate light penetration through shootsbearing densely packed leaves Thus phyllotaxy can beviewed as a developmental limiting factor that can drivecompensatory changes in morphological features such asshape that are not directly controlled by patterns of leafinitiation In this way the lsquoconstraintsrsquo imposed by phyllotaxycan foster evolutionary modifications at the level of the entireshoot (Niklas 1988)

One particularly important trade-off operating at the level ofthe architecture of a shoot is that between leaf size and numberFor any given leaf biomass allocation an individual shoot caneither have a few large leaves or many smaller ones Acrossdifferent herbaceous species and juvenile specimens of treespecies drawn from diverse lineages and ecological settingsthe relationship between total leaf biomass per plant (ML) andtotal stemmass per plant (MS) fallswithin a comparatively narrowcorridor of variation that obeys a positive isometric scalingrelationship (ie MLMS

~10 Niklas 2004) Accordinglyacross these species any stem mass investment yields a moreor less proportional increase in leaf mass Therefore it followsmathematically that across these species the relationshipamong the average mass of a typical leaf mL total leaf numberNL and total stem mass complies with the proportionality mL

NLMS~10 When individual shoots are sampled across

diverse species the relationship between mL and NL per shootis negative and isometric (ie mLNL

ndash1 Kleiman and Aarssen2007) a relationship that appears to be insensitive to howspecies are grouped according to other functional traits oraccording to their habitat preference (Yang et al 2008) Thatthis phenomenology reflects a biomass allocation trade-off isclear What remains uncertain is whether it is a simple trade-offbetween leaf mass investment and leaf size or a more complexset of relationships imposed by twig mass investments andhydraulics or biomechanics

Another trade-off that appears to hold across many but not allspecies is the relationship between leaf surface areaA and averageleaf mass mL For more than a few species-groupings thisrelationship is allometric with a slope that is less than onewhich indicates that increases in leaf mass investment do notresult in proportional increases in leaf area (ieAmL

lt10Niklaset al 2007) This trend of lsquodiminishing returnsrsquo has also beenreported when total leaf area per plant is plotted against total leafmass for plants differing in size (Niklas and Cobb 2008 2010)Several factorsmay contribute to these trends eg biomechanicalanalyses show that disproportionately larger investments must bemade to mechanically support photosynthetic tissues as leavesincrease in their surface area (Niklas et al 2009) In both the leafmass to stem mass and leaf mass to area relationships describedabove leaf shapevariation canprovide anextra degreeof freedomfor selection to act on at both the leaf and branch level

Case studies of leaf shape variation at different scales

Above we have discussed evolutionary and proximal drivers ofleaf shape as well as ecological correlates and functionalsignificance thereof Here we present case studies at threescales where we can examine the ecological and evolutionaryimplications of leaf shape These are the cross community scalecaseof leaf teeth as a subset of shapemodificationswithin lineage

200 bottomtop

150

100

50

0

0

08

07

06

05

04

03

5 10

Days after bud break

Leaf

siz

e (c

m2 )

Leaf

lobi

ng in

dex

(cm

2 cm

ndash2)

Days after bud break

15 20 25

0 5 10 15 20 25

Fig 5 Variation in (a) leaf size and (b) shape complexity inQuercus rubraleaves growing at the top and bottom of the tree crown Reprinted fromZwieniecki et al (2004b) with permission from John Wiley amp Sons

The evolution of leaf shape Functional Plant Biology 543

considerations of the genus Pelargonium and the familyProteaceae and finally the case of leaf shape variation within asingle genetic individual heteroblasty and heterophylly

Leaf teeth leaf shape variation at the community scale

Leaf teeth are conspicuous features inmanyplants and are notableas a leaf-shape trait that varies in predictable ways across plantcommunities Nearly 100 years ago Bailey and Sinnott (19151916) observed that plant communities from regions with lowmean annual temperature (MAT) have a higher site meanproportion of non-monocotyledonous species with toothedleaves Paleobotanists quickly seized upon this relationship asan index for inferring paleotemperature Little et al (2010)recently documented the rich interest in this topic compiling351 literature references related to leaf physiognomy andtemperature of which the primary focus is leaf teeth The linkbetween teeth and temperature is observed on all vegetatedcontinents (Greenwood 2005a Peppe Royer et al 2011) andis also seen in related metrics such as tooth size and tooth number(Huff et al 2003 Royer et al 2005 Peppe et al 2011) Furtherthese tooth variables can display phenotypic plasticity totemperature change (Fig 6 Royer et al 2009b)

What is knownabout the functional bases of these leafndashclimaterelationships Leaf teeth commonly expand and mature fasterthan the bulk leaf (eg Billings 1905 Feild et al 2005) and youngteeth are disproportionately vigorous in their gas exchange(Baker-Brosh and Peet 1997 Royer and Wilf 2006) Fromthese observations Royer and Wilf (2006) hypothesised thatone role of teeth is to increase sap flow thereby deliveringnutrients and other solutes to young emerging leaves Inhabitats with low temperatures teeth could play an importantrole in maximising the potential for carbon gain and growth earlyin the season In warmer climates with longer growing seasonsthe carbon benefit conferred by teeth would be outweighed bytheir unavoidable water cost Alternatively Feild et al (2005)hypothesised that teeth help remove freezendashthaw embolismsthrough positive root pressure and guttation Importantly thetransport of fluids is common to both hypotheses From this itmay be surmised that water-limiting conditions select againstteeth (Bailey andSinnott 1916) Indeed this pattern is clearly seenacross local water availability gradients (Burnham et al 2001

Kowalski and Dilcher 2003 Greenwood 2005b Royer et al2009a Peppe et al 2011) but not globally against mean annualprecipitation (eg Peppe et al 2011)

If the hypothesis by Royer andWilf (2006) is correct toothedspecies should have a higher photosynthetic rate at least at thebeginning of the growing season As such teethmay be related tothe leaf economics spectrum (Wright et al 2004) Some studiesfind support for this idea leaves with a short leaf lifespan(deciduous) (Peppe et al 2011) and low leaf mass per area(Royer et al 2005) two leaf economic variables are morelikely to be toothed It also follows that teeth should be morefunctionally related to growing season climatic variables likeseasonality growing season length and growing degree days notmean annual temperature however few studies find support forthis idea (Royer et al 2005 Peppe et al 2011) Ultimately itwould be valuable to have a better understanding of the geneticprocesses underlying tooth development but presently little isknown outside model species (see lsquoGenetic regulation forpatterning of leaf margin serration and leaflet formationrsquosection) Further phylogenetic history has typically beenassumed to have only a minor role in these leaf-climatepatterns but rigorous tests demonstrate that this assumption isfalse (Little et al 2010) This outcome is of particular concern forpaleobotanists because systematic placement of fossils can bedifficult especially for the Cretaceous and Paleogene Tools toaccommodate for this phylogenetic dependency are needed

Leaf shape variation within lineages

Proteaceae

Proteaceae is an ancient angiosperm family originating inGondwana over 100million year ago and possessing 1700 extantspecies (Weston 2007) The familyrsquos greatest representation is inAustralia followed by SouthAfrica with smaller numbers acrossother Gondwanan segments (Cowling and Lamont 1998 Hootand Douglas 1998 Weston 2007) Within the Proteaceae thereexists an enormous range of leaf sizes and shapes ranging froma few mm2 to 500 cm2 and including entire toothed broadlylobed deeply dissected needle-like simple or compound leavesIntriguingly most of this shape diversity is represented inAustralia The majority of taxa outside Australia possessentire leaves the exception being four South African genera

(a) (b) (c) (d )

Fig 6 Representative leaves of Acer rubrum showing genetic variation and plasticity in leaf teeth Leaf derived from coolclimate (Ontario) seed stock grown in (a) Rhode Island (cool climate) and (b) Florida (warm climate) leaf derived fromwarm climate (Florida) seed stock grown in (c) Rhode Island and (d) Florida Bar (for all leaves) = 1 cm Reprinted from Royeret al (2009) under Creative Commons Attribution Licence

544 Functional Plant Biology A B Nicotra et al

that possess dissected (Serruria and Paranomis) or toothed(Mimites and Leucospermum) leafed species all of whichoccur in a single clade (Sauquet et al 2009)

The reason for the extensive radiation in leaf shape diversityin Australian Proteaceae is unclear Among South AfricanProteaceae leaf size decreases with mean annual temperatureand narrow leaves have been shown to overheat less than theirwider counterparts (Yates et al 2010) Given that leaf dissectionand small size are functionally equivalent with respect to heatloss (see lsquoTemperature and waterrsquo section) it is tempting tosuggest that leaf shape variation in Australia represents anadaptive response to increasing hot dry conditions Yet thereappears to be no consistent distributional pattern of Proteaceaeleaf dimensions inAustraliaManymediumor large un-dissectedleaves exist in the hottest habitats Grevilleas with small simpleleaves can be found across most regions including high rainfallareas and some of the species found in the shaded rainforestunderstorey have among the most deeply dissected leaves of allLeaf shape also varies within a single individual For examplefern-leaf stenocarpus Stenocarpus davallioides Foreman ampBHyland has tripinnatisect juvenile leaves but adult leavescan be pinnate or bipinnate (Foreman and Hyland 1995)Other rainforest species such as Atherton Oak (Athertoniadiversifolia (CTWhit) LAS Johnson amp BGBriggs) havevery large lobed intermediate leaves but adult and seedlingleaves are small and entire (Weston 1995) Alternatively leaflobing within a plant can be random having no obvious spatialor developmental association or homogenous across a speciesExamining a phylogeny of extant taxa one finds lobed ordissected leaves alongside entire-leafed species ndash across mostlineages in the group (Weston 2007) Therefore althoughsuggestions of microclimatic adaptation may be offered atthe species level it seems unlikely that an environmentalassociation with leaf shape could be generalised acrossAustralian Proteaceae Rather leaf shape variation in theProteaceae is likely to reflect the range of ecological solutionsto leaf structurefunction trade-offs discussed above

Pelargonium

Similar to the Proteaceae the genus Pelargonium isdistributed predominantly in the southern hemisphere withnearly 80 of its almost 300 species found in South AfricaAlthough the age of the genus is estimated to be around30million years (Bakker et al 2005) the highest rate ofspecies accumulation has occurred in the last 10million years(Martinez-Cabrera 2010) The combination of climatic andedaphic heterogeneity over short distances and limited geneflow (Latimer et al 2005 Linder 2005) is frequently offeredas an explanation for the extraordinary plant species diversity thatcharacterises southern Africa Pelargonium the third largestgenus in the region is characterised by remarkable variation ingrowth form that ranges from annuals and geophytes to shrubsVariation in leaf shape is no less dramatic ranging from entireto highly dissected (Fig 1) often within growth forms andtaxonomic sections Parsing of leaf form categorically revealedthat the extent of dissection of the blade is highly evolvablewithinanoverall ovate leaf shapeoutline that is evolutionarily conserved(Jones et al 2009)

In addition to high evolutionary lability of leaf dissectionthere is variation among major clades in the evolutionary patternof major veins One major clade (Clade C) is dominated bypalmate venation whereas another (Clade A) is dominated bypinnate venation although both types are found in both cladesWe note that pinnate venation is significantly associated withincreased dissection and reductions in functional leaf sizecharacteristic of the highly diverse lsquoxerophyticrsquo clade (CladeA2) of the winter rainfall region

To examine whether leaf shape translates into functionalvariation carbon gain and water use efficiency were measuredin pairs of species contrasted for leaf shape and replicated acrossthree subclades of Clade A and one subclade of the third majorclade CladeB (Nicotra et al 2008) In each case the species withthemore dissected leaves showed higher rates of carbon gain butalso higher rates of water loss in all treatment conditions Thatpaper argued that the differences in carbon gain were not directresults of leaf shape per se but rather that the same conditionsthat result in selection for more dissected leaves also favourthe evolution of high photosynthetic rates high leaf nitrogencontent and opportunistic use of water when available Evenin this study however there was no correlation between leafdissection and the climate present in native ranges suggestingeither that climate is selective at a much smaller scale (ie themicrohabitat) or that for any given set of climate variablesmultiple leaf shapes can be adaptive in association withvariation in other leaf traits

Heteroblasty and heterophylly variationwithin a single genetic individual

We have so far focussed primarily on cross species variation inleaf shape As discussed for the Proteaceae genera above (lsquoTheProteaceaersquo section) however a single genome can produce aremarkable range of leaf shapes during the lifetime of a plantThere are two broad categories of leaf shape variation that can beexpressed by a single individual This variation can arise either inresponse to different external environments or to ontogeneticsignals (some of which may be influenced by environment aswell)

Environmentally induced heterophylly

Environmentally induced heterophylly results when leafshape responds to environmental cues The classic case is thedramatically different emergent and submergent leaves in aquaticplants Although many aquatic plants are also heteroblastic (seebelow Sculthorpe 1967) environmentally induced heterophyllyis not specific to position or plant age and is reversible dependinglargely on the environmental signals (mediated by hormonalsignals) perceived by cells of the developing leaf primordiumThe developmental age of the leaf at the time of signal perceptionis critical Very young primordia (L1ndashL3) generally respondcompletely to the novel environment whereas primordiathat were slightly older at the time of environmental changeshow intermediate features as mature leaves Characteristics ofspecific cells are related to the position of that cell within thedifferentiation gradient characteristic of that species at the timeof environmental change (eg Goliber and Feldman 1990 Bruniet al 1996 Kuwabara and Nagata 2006)

The evolution of leaf shape Functional Plant Biology 545

Among terrestrial plants the lsquosunndashshadersquo leaf response isfrequently cited as the classic example of environmentallyinduced heterophylly Historically differences in leaf formwithin a canopy have been assumed to be responses to alteredlight environments (eg larger and less-lobed leaves are producedin shade)However these different leaf shapes also confer distincthydraulic advantages (see lsquoTemperature and waterrsquo section)Whether in response to light or to hydraulic limitation lsquosunshadersquo heterophylly is accompanied by similar and well-knownanatomical differences across a wide range of species (ie sunleaves are generally smaller and thicker with smaller cells thanshade leaves Taiz and Zeiger 2002) However when theseanatomical differences become determined developmentally(Dengler 1980 Yano and Terashima 2004) has been lesscommonly studied The presence of intermediate anatomicalcharacteristics in leaves transferred among light environmentsshortly after expansion ceased suggests that leaves of somespecies have the capacity to respond anatomically to alteredlight levels until very late in development (Oguchi et al2005) In contrast recent studies suggest that systemic lightand humidity signals sent from older to younger leavesallow younger leaves to develop phenotypes suited for theiranticipated environments (Lake et al 2001 Yano andTerashima 2001 Thomas et al 2004) Exactly when final leafshape is determined in lsquosunshadersquo leaves remains an openquestion for most species

Heteroblasty describes ontogenetically dependentchanges in leaf features most notably shape

In 1900 Goebel originally distinguished differences in lsquojuvenileformrsquo from lsquoadult formrsquo on the basis of dramatic differences inleaf shape and associated traits (Balfour 1969) As an example he

described the shift from compound leaves to phyllodes in Acaciaas heteroblastic and reserved the term homoblastic for the lessdramatic variation exhibited by Casurina It is now generallyaccepted that heteroblasty refers to the full suite of features thatchange along the shoot during development (Jones 1999)

Heteroblasty results from the interaction of developmentalprograms expressed at both the leaf and the shoot- or whole-plant level (Kerstetter and Poethig 1998 Tsukaya et al 2000Usami et al 2009 Wu et al 2009 Chuck and OrsquoConnor 2010)Heteroblasty is generally not reversible unless reiterative growthoccurs (ie following damage) Morphological differencescharacteristic of heteroblasty such as differences in lobing areusually expressed relatively early in development shortly afterleaf initiation (Jones 1995)

The sequence of heteroblastic leaf shapes produced alongthe shoot can be accelerated or delayed in response to differentgrowth environments or architectural manipulations (Ashby1948 Forster and Bonser 2009 Jaya et al 2010)Furthermore both heteroblasty and plasticity in heteroblastyhave been shown to be under strong genetic control (Anderson1989 Climent et al 2006) Forster et al (2011) suggested thatboth heteroblasty and plasticity in heteroblasty are likely to occurin response to lsquowhich leaf type confers the greatest benefit fora given light levelrsquo Several studies over the last decade havepointed towards functional consequences of heteroblastygenerally with reference to single environmental variables (egHansen 1986 Gamage and Jesson 2007 Kubien et al 2007)It is likely that heteroblasty influences multiple leaf functionsbecause suites of leaf traits change with heteroblasty (Jones2001) Indeed this conclusion was supported in a recent studythat examined differences inmore than a hundred leaf traits in koa(Acacia koa AGray Pasquet-Kok et al 2010) Heteroblasty hasalso been proposed as an adaptation to herbivory (Karban and

Table 1 Promising directions for cross-disciplinary investigation of leaf shape

No Direction

1 A multi-species lsquoevo-devorsquo approach similar to that by Illing et al (2009) is likely to be informative with regard to many enigmatic features regardingthe evolution of diverse leaf forms and more generally provides an exciting system for examining the convergence and divergence in the evolutionof function

2 A phylogenetic analysis of leaf shape examining correlated evolution with other traits such as venation size leaf surface characteristics and branch andarchitectural measures would enable a thorough investigation of the situations in which leaf shape diversification arises

3 Further research into both the genetic and developmental processes underlying lobes and teeth and the ecological correlates of each would provide anexcellent opportunity to link across wide environmental scales

4 Cross species studies ideally in the field of the relative importance of leaf shape among other factors in determining leaf temperature and water status arenecessary to test the conclusion put forward here that the functional significance of leaf shape lies more often in water relations than directly in leafthermoregulation

5 Studies of the scale of environmental change that selects for ontogenetic response in leaf shape in heteroblastic species fromdiverse environments will shedlight on links between genes development and ecology

6 Detailed quantifications of the extent towhich leaf shape differswithin canopies particularly of dioecious specieswhere selection on reproductive functioncan lead to dimorphic conditions will help link leaf shape to whole plant processes

7 Studies at the scale of transects across the leaf surface that simultaneously consider hydraulic resistance temperature and light will reveal the extent oflocalised hydraulic and thermal stresses within leaf surfaces that might act as selective forces on leaf shape per se Imaging approaches incorporatingmeasures of 3-dimensional shape temperature and fluorescence provide exciting opportunities for examining spatial heterogeneity in temperature andphysiological processes

8 Since patterns of leaf venation affect both hydraulic supply and capacity for photosynthate transport studies that simultaneously examine both willreveal the extent to which interactions among these fundamental processes either co-determine leaf shape or establish trade-offs ultimately reflectedin leaf shape

9 Finally studies examining the fitness values of different leaf shapes or plasticity in leaf shapes under a range of environmental conditions are needed tobetter identify which aspects of leaf shape are most likely to be adaptive

546 Functional Plant Biology A B Nicotra et al

Thaler 1999 Brennan et al 2001 Fadzly et al 2009) and it hasbeen associated with differences in frost resistance (Darrow et al2001) It is worth noting that studies of the functional significanceof heteroblasty have focussed on relatively few plant lineagesGiven that heteroblasty occurs widely among the angiospermsas well as in some ferns and gymosperms (eg Wagner 1952Mueller 1982 Pryer and Hearn 2009) the next questions mayfocus on the cost of heteroblasty and its fitness consequenceswhat scale of environmental change selects for ontogeneticresponse in leaf shape

Synthesis and next steps

Whether considered at community lineage or single genomescales leaf shape is a trait that is highly labile and responsive to arange of biotic and abiotic factors Although some leaf traitscan be explained in terms of particularly strong trade-offsbetween two or more factors that result in clear scalingrelationships (eg leaf mass-to-area traits or leaf mass-to-stemmass) leaf shape in contrast does not emerge from the aboveconsideration as a factor capable of explaining ecologicaldifferentiation between species Rather leaf shape emerges asa trait for which there are many quite varied functional trade-offs

As stated above these trade-offs for successful functionalityof leaves may generate ecological opportunities leaf shape isperhaps best viewed not as a single major axis but rather asan option that fine tunes the leaf to its conditions over bothshort and evolutionary time spans Numerous elements of thisreview support this conclusion Over evolutionary history theextant range of variation in leaf shape has been explored bymanylineages although only the angiosperms have explored thisrange with the added element of massive increases in veindensities Likewise our growing understanding of geneticcontrols on leaf shape reveal that it is a complex traitdetermined by many different genes at several differentdevelopmental stages Although some genes have conservedaffects on leaf morphology including shape the exact natureof their affects can vary widely Likewise the function to whichagivenelement of leaf shape lsquois putrsquo canvary For example closerinvestigation of the internal architecture and developmentalpathways associated with different leaf shapes suggests thatleaf teeth and leaf lobes often have distinctly differentfunctional foundations and may play different functions indifferent environments

Despite amyriadof functional roles the one recurrent theme inour review is that leaf shape is affected by and strongly influencesleaf water relations From a functional perspective the role of leafshape in leaf thermoregulation has perhaps received the mostattention in previous considerations of the functional significanceof leaf shape The relationship between leaf size and temperaturebeing physically determined by basic energy balance principlesthis was a logical starting point of enquiry But as we haveshown the connection between leaf shape (and size) and leaftemperatures under field conditions is not well supportedempirically Rather leaf shape variation as distinct from sizemore often will reflect water supply trade-offs High veindensities support high photosynthetic rates (Brodribb et al2010) and dissected leaves maintain a greater proportion ofleaf tissue closer to main veins Leaf temperature plays a role

in this story as higher leaf temperatures lead to greaterevaporative water loss but many factors other than leaf shapeinfluence leaf temperature (eg leaf thickness orientation andreflectivity and branch and plant architecture) Finally it is alsoworth noting that although there appears to be an obviouslink between shape and water supply there is a concomitantinteraction between water supply and phloem function thatremains relatively unexplored

Thus our review suggests several angles for further research(Table 1) In particular we advocate those approaches thatenable investigation across the scales examined here (genedevelopment function and evolutionary ecology) and thatfocus in particular on the links among leaf shape developmentand water relations

Acknowledgements

The authors thank the Botanical Society of America for their support of thesymposiumfromwhich this reviewaroseWealso thankLawrenSackand twoanonymous reviewers for useful comments on an earlier version of thismanuscript

References

Ackerly DD Knight CA Weiss SB Barton K Starmer KP (2002) Leafsize specific leaf area and microhabitat distribution of chaparral woodyplants contrasting patterns in species level and community level analysesOecologia 130 449ndash457 doi101007s004420100805

Anderson S (1989) Variation in heteroblastic succession among populationsof Crepis tectorum Nordic Journal of Botany 8 565ndash573 doi101111j1756-10511989tb01730x

Armbruster WS Muchhala N (2009) Associations between floralspecialization and species diversity cause effect or correlationEvolutionary Ecology 23 159ndash179 doi101007s10682-008-9259-z

Armbruster WS Di Stilio VS Tuxill JD Flores TC Runk JLV (1999)Covariance and decoupling of floral and vegetative traits in nineneotropical plants a re-evaluation of Bergrsquos correlation-pleiadesconcept American Journal of Botany 86 39ndash55 doi1023072656953

Ashby E (1948) Studies in the morphogenesis of leaves I An essay on leafshape New Phytologist 47 153ndash176 doi101111j1469-81371948tb05098x

Aso K Kato M Banks JA Hasebe M (1999) Characterization ofhomeodomain-leucine zipper genes in the fern Ceratopteris richardiiand the evolution of the homeodomain-leucine zipper gene family invascular plants Molecular Biology and Evolution 16 544ndash552

Bailey IW Sinnott EW (1915) A botanical index of Cretaceous and Tertiaryclimates Science 41 831ndash834 doi101126science411066831

Bailey IW Sinnott EW (1916) The climatic distribution of certain types ofangiosperm leaves American Journal of Botany 3 24ndash39 doi1023072435109

Baker-Brosh KF Peet RK (1997) The ecological significance of lobed andtoothed leaves in temperate forest trees Ecology 78 1250ndash1255

Bakker FT Culham A Marais EM Gibby M (2005) Nested radiation inCape Pelargonium In lsquoPlant species-level systematics new perspectiveson pattern and process Vol 143rsquo (Eds FT Bakker LW Chartrou BGravendeel PB Pielser) pp 75ndash100 (ARG Ganter Verlag KG RuggellLichtstein)

Balfour IB (Transl) (1969) lsquoOrganography of plants I General organographyrsquoby K Goebel 1900 (Hafner Publishing Co Inc New York)

Ball MC Cowan IR Farquhar GD (1988) Maintenance of leaf temperatureand the optimization of carbon gain in relation to water loss in a tropicalmangrove forest Australian Journal of Plant Physiology 15 263ndash276doi101071PP9880263

The evolution of leaf shape Functional Plant Biology 547

Establishment of dorsiventral domainsand lamina flattening

Since leaf primordia arise as protrusions from shoot apicalmeristems (SAM) suppression of the regular SAMdevelopmental program is the first step required for normaldevelopment of a leaf primordium Recently Sarojam et al(2010) found that genes in the YABBY family suppress SAMidentity and promote lamina growth in Arabidopsis In mostbifacial leaves such as leaves of Arabidopsis and snapdragon(Antirrhinum majus L) the radial primordium differentiatesinto abaxial (lower) and adaxial (upper) surfaces to establishdorsiventrality (Waites and Hudson 1995) Our currentunderstanding is that the molecular genetic regulation ofidentification of the abaxial and adaxial domains depends onactions of both small RNAs and directindirect targets suchas class III Zip genes KANADI genes and AUXIN RESPONSEFACTORS3 (ARF3)ETT in Arabidopsis (Husbands et al 2009Floyd and Bowman 2010 Kidner 2010) Without establishmentof the adaxial and abaxial domains it is believed that leavescannot expand in any angiosperm species rather they becomestalk like since the so-called plate meristem is never activated toexpand the laminas (Fig 3)

The monocot clade is typified by many unifacial leaves thathave only abaxial identity in the lamina as seen in leek (Alliumampeloprasum var porrum L) and Iris spp Leek leaves arecylindrical since they lack adaxial identity in their leaf bladeIris in contrast has flat leaves that expand in a distinct

perpendicular direction relative to that of normal bifacialleaves Likewise the rush Juncus prismatocarpus RBr hasunifacial flat leaves the primordium of the leaf blade growsinto a flat lamina as a result of extensive thickening growthThis developmental pathway is dependent on the geneDROOPING LEAF (DL Yamaguchi et al 2010 Fig 3) thatis also a member of the YABBY gene family but with a monocot-specific function As seen in this case diversified morphology ina particular taxon sometimes depends on some taxon-specificmolecular mechanisms that are modifications of gene functionfrom ancestral lineages

Mechanisms for expansion of lamina

Once dorsiventral identity is generated the leaf lamina can thendiversify in morphology to form simple or compound leaveswith entire or serrated margins differing in thickness length andwidth and in orientation (Tsukaya 2006) The leaf length widthratio is regulated by polar-dependent cell expansion and cellproliferationdistribution Several key genes for its regulationhave been identified from Arabidopsis ANGUSTIFOLIA (AN)and ROTUNDIFOLIA3 (ROT3) regulate the shape of cellswhereas ANGUSTIFOLIA3 (AN3) and ROTUNDIFOLIA4(ROT4) affect the number of cells in the lamina (Tsukaya2006) Loss-of-function mutations of AN and AN3 result innarrower leaves whereas loss-of-function of ROT3 oroverexpression of ROT4 make leaves shorter In addition

Fig 3 Twodifferentmechanismsof laminagrowth Inmost leaves such as leaves ofArabidopsis (upper left)dorsiventral identity (adaxial upper side (shown in green in colour version online) and abaxial lower side(shown in blue in colour version)) identities are established Flattening lamina growth occurs along the junctionbetween adaxial and abaxial domains as a result of activity of the plate meristem Loss of either abaxial oradaxial identity results in radialised growth (lower right) Stick-like leaf morphology seen in leek and Juncuswallichianus (centre) is the typical example of unifacial leaves that have no adaxial domain In the monocotclade however some species develop flat lamina irrespective of lack of adaxial identity in the lamina (rightSisyrinchium rosulatum) that is supported by enhanced growth in the leaf-thickness direction Meristematicactivity occurs in center region of both leaves (orange in colour version) relative position of the shoot apicalmeristem (SAM) is shown by the cross-hatched dot (green in colour version)

The evolution of leaf shape Functional Plant Biology 539

LONGIFOLIA1 (LON1) and LON2 regulate the length of the leaflamina in Arabidopsis (Lee et al 2006) A vast array of genes isknown to influence leaf area (Horiguchi et al 2006) For exampleheteroblastic change in leaf area is governed by the miR156-SQUAMOSAPROMOTERBINDINGPROTEIN-LIKE (SPL)-miR172 pathway via regulation of cell size and cell numbers inArabidopsis (Usami et al 2009)Alterations of activities of anyofthese factors could contribute to natural variation in leaflength width ratio and area None of these genetic variationsresult in loss of reproductive structures suggesting that thesegenes could be under direct environmentally mediated selectionAs yet we do not know whether the above patterns hold in othernon-model species

Leaves also vary in the extent to which the lamina is flator curved CINCINATA (CIN) of snapdragon was oncethought to function to keep the leaf flat by orchestrating cellproliferation along the medio-lateral axis of the leaf primordia(Nath et al 2003) More recently however analyses of theTCP genes which are homologues of CIN in Arabidopsisindicate that the major role of the TCP is in promotion ofmaturation in laminar tissues not in regulation of cellproliferation (Efroni et al 2008 Sarojam et al 2010) The roleof CIN in snapdragon may also be the same as that of TCPsin Arabidopsis The other candidate genetic mechanism forcontrol of leaf curvature is an auxin-related pathway Severalhyponastic leaf mutants have been isolated from Arabidopsisand many of the responsible genes are related to auxinsignalling (Perez-Perez et al 2002 2009) Since auxinsignalling is also very important for venation patterningspecies-specific modification of auxin signalling in leafprimordia may have caused diversification of both leafcurvature and vasculature patterning

Genetic regulation for patterning of leaf marginserration and leaflet formation

Variation in leaf shape as distinct from length and area orcurvature arises from variation in leaf margin and lobeformation Kawamura et al (2010) showed that spatialpatterning of serrations or leaf teeth (see lsquoLeaf shape variationwithin lineagesrsquo section) on the leaf margin depends onformation of auxin maxima on the leaf margin in ArabidopsisLocation of these maxima is governed by polar auxin transportand the stabilisation of auxinmaxima ismaintained by prolongedexpression of CUP-SHAPED COTYLEDON2 (CUC2) around it(Kawamura et al 2010) The mechanism of rhythmic patternformation of the auxin maximaminima on the leaf margin isthought to involve self-organising pattern formationmechanismsby polar auxin transport similar to that involved in patternformation of phyllotaxy (de Reuille et al 2006 Jonsson et al2006) Either the loss of distinct auxinmaxima or an instability ofthe auxin maxima will result in an entire leaf margin (Kawamuraet al 2010) Changes in functions of these genes howeverseverely influence many biological processes including floralorgan development as seen in the pin-formed 1 mutant (Okadaet al 1991) Thus variation among species in pattern of marginserration might accompany variation in floral organs or morelikely reflect differential regulation for example by downstreamgenetic pathways

Combined control of auxin-maxima formation by PIN andCUC genes as described for serration above is also known tohave an important role in leaflet patterning in bittercress(Cardamine flexuosa With) an Arabidopsis relative withcompound leaves (Barkoulas et al 2008) Again this controlvaries among species in tomato (Solanum lycopersicum L) lossof CUC3 function does not convert compound leaves intosimple leaves (Blein et al 2008) Thus it is clear that theindependent evolution of compound leaves from simple leavesis likely driven by species-specific mechanisms (Efroni et al2010) though these may reflect modifications of similarpathways

Recent studies have identified several important key genes forregulation of compound versus simple leaf forms For exampleabsence or prolonged expression of class I Knotted1-likehomeobox (KNOX) genes in leaf primordia is important tocompound leaf formation in a wide range of angiosperms andgymnosperms (Bharathan et al 2002) In some legume specieshowever the LEAFY (LFY)UNIFOLIATA (UNI) gene is the keyfor compound leaf formation instead of the class I KNOX genes(Hofer et al 1997) Changes in these genes could be linked tothe evolution of compound leaves although loss-of-function ofLFYUNI also causes infertility in Arabidopsis and thus it isunlikely that selection directly on these particular genes couldresult in theevolutionof variation in leaf shapeDiversified spatialpatterning of petiolepetiolule and laminaleaflet in compoundleaves have also been suggested to be linked to the distributionpattern of AS1-ROUGH SHEATH2-PHANTASTICA (ARP)expression domains (Kim et al 2003) However based on thefact that the ARP gene has only a limited role in compound leafformation in some species it is unclear whetherARPwill turn outto be major factor in others (Efroni et al 2010)

As described above the independent evolutionary origins ofseveral angiosperm leaf features should caution against theassumption that a single ancestral genetic system underlies leafdevelopment The developmental genetic approaches outlinedhave revealed several key mechanisms for leaf shape controlbut have also demonstrated variation in the function of thesegenes among lineages In a few studies researchers have gonebeyond one or two species studies and have looked for deeperevolutionary links (Illing et al 2009) Studies of the role of theclass 1 KNOX genes have considered not just angiospermsbut other lineages Among simple-leafed dicots and monocots(eg Arabidopsis and Zea) KNOX genes are strongly expressedduring the indeterminate growth of the apical meristem but arerapidly downregulated in the cells flanking the apical domein regions defined by leaf primordia Although some arginine-rich protein (ARP) genes repress the expression of KNOXtranscription factors during the development of both micro-and megaphyll primordia (Harrison et al 2005) the role ofhomeodomain-leucine zipper (HD-ZIP) III genes which areinvolved in the specification of foliar adaxialabaxial fatediffers in the development of these two leaf types (Floyd andBowman 2006) HD-ZIP III genes have been identified in themoss Physcomitrella (Sakakibara et al 2001) and in the fernCeratopteris (Aso et al 1999) they are also expressed duringnormal vascular tissue development in Arabidopsis (Zhong andYe 1999) suggesting that they are very ancient but alsomultifunctional

540 Functional Plant Biology A B Nicotra et al

Accumulating knowledge on molecular genetic mechanismsof leaf morphogenesis is leading to an increasingly evo-devoapproach to studies of leaf diversity (Tsukaya 2010)this progress is aided by establishing new model species(eg Cardamine hirsuta and the rush Canales et al 2010Yamaguchi and Tsukaya 2010) Over the next decade weanticipate that genomic approaches on multiple speciesanalysed in phylogenetic frameworks (eg Illing et al 2009)will resolvemuchof theenigma regarding the evolutionofdiverseleaf forms

Genetic constraints on flower petal and leaf form

One additional developmental consideration is that as famouslydescribed by von Goethe in 1790 flower petals are derivatives ofthe apical meristem and serial homologues to leaves (von Goethe2009) Thus one might hypothesise that although the evolutionof leaves substantially predates that of flowers selection on theform of either of these organs would also result in changes inthe other such that selection on leaf shape and size subsequent tothe evolution of floral organs may be constrained by pleiotropiccontrol of floral form Recent genetic data identifies several genesthat have roles in both leaf and flower form (Dinneny et al 2004Schmid et al 2005 Street et al 2008) For example geneticchanges in both leaf length to width ratio (eg rot3) and leaf size(eg an3) influence not only leaf proportions but also that of thefloral organs in Arabidopsis (Kim et al 1999 Horiguchi et al2005) The garden pea (Pisum sativum L) is a good case in pointas discussed above compound leaf formation depends on LFYUNI but the same gene is also required for flowering Thus ifcurrent selection favoured simple leaves in garden pea changes inleaf shape could be constrained by pleiotropic control offlowering (Hofer et al 1997) Likewise Shalit et al (2009)recently reported that the action of florigen ndash a plant hormonefor flowering ndash influences complexity of leaves by changingnumbers of leaflets in tomato Such relationships suggest thatsome floral characters as well as other metabolic pathways maynowbeclosely linkedwith regulationof leaf shapeand sizeA fewecological studies have also assessed correlations between petaland leaf form (Berg 1960 Armbruster et al 1999) This questionis of particular interest in lineages where animal pollination isassociated with increased diversification rates either as a result ofdirect selection or as a secondary reinforcing factor (van der Nietet al 2006 Armbruster and Muchhala 2009 Kay and Sargent2009)

In situations where leaf and petal form are under similargenetic control and both under selection one of two thingscould happen If leaf shape does not have substantialfunctional impact then one might expect leaf form to changewith selection on petal form However given the likelyimportance of leaf form to photosynthetic function it is morelikely that a breakdown in genetic correlations between leafand flower form (eg by alterations of downstream regulatorypathways) would evolve Several factors argue against thehypothesis that leaf shape is frequently a product of selectionon flower form First diverse leaf shapes were in evidencelong before the origin of flowers in the angiosperms Secondthe genetic mechanisms underlying leaf shape revealconsiderable evolvability with similar genes being co-opted

for different functions in different species Finally given thepotential importance of both petal and leaf shape we suggestthat genetic constraints on the independent evolution of bothstructures are likely to have broken down in those lineages withhighly diversified leaf shapes over the many millions of yearssince petals first evolved

Ecological correlates and the functional significanceof leaf shape

Evolutionary history plays a part in determining leafmorphology but our consideration of leaf shape evolution andits developmental genetic basis demonstrates that diverselineages have repeatedly gained and lost many leaf featuresThus developmental and genetic constraints on the evolutionof leaf form and shape in particular have been overcomemany times or have led to new evolutionary opportunities Itis of particular interest then to ask what are the likely ecologicaldrivers andor functional significance of leaf shape

Temperature and water

Conventionalwisdomwould say that thermoregulation is perhapsthe main driver of leaf shape evolution The maintenance ofleaf temperatures within certain limits is critical for a plantrsquosgrowth and survival and although photosynthetic tissue of somespecies canwithstand temperatures below6C (Ball et al 2006)and above 60C (Clum1926Nobel 1988) irreversible damage toleaves can occur at temperatures well within this range (Levitt1980 Jones 1992 Ball et al 2004 Groom et al 2004 Sharkey2005) Leaf size and shape potentially have large effects onleaf temperature because the two-dimensional proportions of aflattened leaf determine the rate of heat transfer between acrossthe leafndashair interface by influencing the thickness of its boundarylayer

All else being equal the thickness of a boundary layerincreases with length from the windward edge so that heatconvection from small leaves is more rapid than from largeleaves (Raschke 1960 Gates 1968 Vogel 1970 Parkhurst andLoucks 1972Grace et al 1980Geller andSmith 1982Monteithand Unsworth 1990 Schuepp 1993) Likewise because leaflobing reduces the distance across the lamina the rate of heattransfer is predicted to be greater in a lobed leaf than an unlobedleaf with equivalent area (Parkhurst et al 1968 Vogel 1968Lewis 1972 Givnish 1978 Gurevitch and Schuepp 1990a)

The rate of heat transfer from lobed metal plates or lobedleaves coated in metal is greater than from shallow-lobed or un-lobed ones (Parkhurst et al 1968 Thom 1968 Vogel 1970Gottschlich and Smith 1982 Gurevitch and Schuepp 1990bRoth-Nebelsick 2001) Artificial or coated leaves have beenused because they enable boundary layer resistance to beinvestigated in the absence of transpiration However thisapproach is biologically flawed because it does not enable anassessment of the relative importance of a range of traits (eg leafwater content absorbance) in determining the actual operatingtemperatures of leaves in the field

Actual leaf temperature measurements show that large leavescan operate at below-ambient temperatures even on hot days asa result of high rates of latent heat loss through transpiration(Drake et al 1970 Gates 1980 Hegazy and El Amry 1998)

The evolution of leaf shape Functional Plant Biology 541

Thus conditions of evaporative demand and the availability ofsoil water for transpiration are key to understanding leafshapendashtemperature relationships under biologically realisticconditions Because of their much greater transpirationalcapacities (see above) flowering plants have far greaterleeway than other plants regarding the size and shape of theirleaves (see lsquoA brief history of the angiosperm leaf and shapediversity thereinrsquo section Boyce et al 2009)

In addition to reducing heat loads the morphology of deeplylobed leaves may also reflect direct selection for increasedhydraulic efficiency Water moving through a plant encountersnumerous resistances on its path from soil to roots to stemsthrough leaves and out to the external atmosphere Within theliquid phase of water flow through a plant leaf hydraulicresistance Rleaf accounts for more than 30 of total plantresistance (Sack and Holbrook 2006) The high resistance inleaves is due first to a system of veins of decreasing size theresistance of which is inversely proportional to the fourth powerof the radius of the component conduits in these veins Thereforeas a proportion of the considerable resistance in the leaf(Zwieniecki et al 2002) the minor veins provide the greatestresistance to water flow (Sack et al 2004)Water also encountershigh resistance as it exits veins laterally and travels to othercells mainly via apoplastic pathways through the mesophyllAccordingly on the order of half of total Rleaf can occur outsidethe major veins (Sack and Holbrook 2006) As resistanceincreases along the pathway towards sites of evaporationlocalised water potential (Y) becomes progressively morenegative (Brodribb et al 2010) The result of how thisphenomenon might affect a leaf has long been observedregions of leaves lying furthest from the main supply channelsare the first to lsquowitherrsquo when exposed to strong winds (Fig 4Yapp 1912) Leaf lobing represents an effective removal of thispotentially stress-prone tissue and has been suggested as anadaptation to dry conditions (Thoday 1931 Givnish 1979) Ifdeeply lobed leaves have a lower ratio of mesophyll tissue tolarge highly conductive veins they should then have reducedresistance relative to less- or un-lobed leaves (Sack and Tyree2005)

Although leaf lobing might confer adaptive benefits withrespect to maintaining stable hydraulic supply it may also bethat the extent of lobing or margin dissection is itself determinedby hydraulic limitation (Sack et al 2003 Zwieniecki et al2004b Boyce 2009) For species with lsquosunrsquo and lsquoshadersquoleaves the morphology of small deeply lobed sun leaves atthe outer canopy is proposed to result from water pressuredrop across the leaf lamina caused by reduced water deliveryto expanding cells during growth (Fig 5 Zwieniecki et al 2004bBoyce 2009 Leigh et al 2011)Alternatively recent studies havesuggested that many of the characteristics of lsquosunrsquo and lsquoshadersquoleaves are the result of branch autonomy within the plant canopy(Niklas and Cobb 2010)

Thus although the effects of leaf shape on leaf thermalregulation hold when tested on model leaves predictions areless straightforward when hydraulic function is accounted forAdditionally consideration of the relative contribution of a rangeof leaf and branch properties to leaf temperature is critical It islikely that leaf shape is only one amongmany factors influencingleaf thermal regulation other factors could include water

content leaf thickness spectral reflectance orientation andplant architecture Further given the observation of high leafshape diversity withinmany plant communities from hot and aridenvironments it seems likely that regulating leaf temperature isnot the singlemost important evolutionary influence on leaf shapediversity The association between leaf shape and hydraulicproperties which is affected by leaf temperature in contrast islikely to be of greater importance

Leaf structure and functional trade-offs

Many foliar leaf traits including shape reflect functional trade-offs that have been resolved in various ways by differentspecies depending on their ecological settings (eg hydro- vsmesophytes) physiological attributes (eg C3 vsC4metabolism)developmental capabilities (eg heterobaric vs non-heterobaricleaves) or evolutionary histories (eg micro- vs megaphylls)Life history and optimisation theory show that the number ofphenotypic solutions that allow for different equally successfultrait combinations increases as the number of trade-offsincreases ndash a conclusion that applies to traits within the leaf(eg for shape) as well as to leafndashbranch relationships (Niklas1988)

It is evident that leaves perform several functionssimultaneously that is they intercept light transport liquidsand solutes exchange gases with the atmosphere dissipateheat cope with externally applied mechanical forces and

(a)

(e)(f )

( i )

(g)

(h)

(b)

(c)

(d )

Fig 4 Localised lsquowitheringrsquo of leaves exposed to strong wind Reprintedfrom Yapp (1912) with permission from Oxford University Press

542 Functional Plant Biology A B Nicotra et al

defend themselves from herbivores and pathogens Importantlymany of these functions have conflicting physiologicalanatomical or morphological requirements For exampleorientations of lamina that maximise light interception alsominimise the ability to dissipate heat Further all of thesefunctions are intrinsically size dependent For example thequantity of mechanical tissues required to support laminaedepends on the mass of laminae whereas the ability tointercept light depends on laminae area

When seen in this light the trade-offs required for thesuccessful functionality of foliar leaves generate rather thanconstrain ecological and evolutionary opportunities This canbe illustrated mathematically and empirically in the context ofphyllotaxy and its consequences on light interception Computersimulations of mathematically generated shoots to assess theinfluence of leaf shape size and phyllotactic patterns on theability to intercept direct solar radiation show that differences inphyllotaxy significantly influence light interception (particularlyfor shoots with a rosette growth habit) They also show thatcomparatively small differences in leaf shape can compensatefor the negative effects of leaf overlap resulting from virtuallyany phyllotactic pattern For example lobed leaves or pinnifid

compound leaves facilitate light penetration through shootsbearing densely packed leaves Thus phyllotaxy can beviewed as a developmental limiting factor that can drivecompensatory changes in morphological features such asshape that are not directly controlled by patterns of leafinitiation In this way the lsquoconstraintsrsquo imposed by phyllotaxycan foster evolutionary modifications at the level of the entireshoot (Niklas 1988)

One particularly important trade-off operating at the level ofthe architecture of a shoot is that between leaf size and numberFor any given leaf biomass allocation an individual shoot caneither have a few large leaves or many smaller ones Acrossdifferent herbaceous species and juvenile specimens of treespecies drawn from diverse lineages and ecological settingsthe relationship between total leaf biomass per plant (ML) andtotal stemmass per plant (MS) fallswithin a comparatively narrowcorridor of variation that obeys a positive isometric scalingrelationship (ie MLMS

~10 Niklas 2004) Accordinglyacross these species any stem mass investment yields a moreor less proportional increase in leaf mass Therefore it followsmathematically that across these species the relationshipamong the average mass of a typical leaf mL total leaf numberNL and total stem mass complies with the proportionality mL

NLMS~10 When individual shoots are sampled across

diverse species the relationship between mL and NL per shootis negative and isometric (ie mLNL

ndash1 Kleiman and Aarssen2007) a relationship that appears to be insensitive to howspecies are grouped according to other functional traits oraccording to their habitat preference (Yang et al 2008) Thatthis phenomenology reflects a biomass allocation trade-off isclear What remains uncertain is whether it is a simple trade-offbetween leaf mass investment and leaf size or a more complexset of relationships imposed by twig mass investments andhydraulics or biomechanics

Another trade-off that appears to hold across many but not allspecies is the relationship between leaf surface areaA and averageleaf mass mL For more than a few species-groupings thisrelationship is allometric with a slope that is less than onewhich indicates that increases in leaf mass investment do notresult in proportional increases in leaf area (ieAmL

lt10Niklaset al 2007) This trend of lsquodiminishing returnsrsquo has also beenreported when total leaf area per plant is plotted against total leafmass for plants differing in size (Niklas and Cobb 2008 2010)Several factorsmay contribute to these trends eg biomechanicalanalyses show that disproportionately larger investments must bemade to mechanically support photosynthetic tissues as leavesincrease in their surface area (Niklas et al 2009) In both the leafmass to stem mass and leaf mass to area relationships describedabove leaf shapevariation canprovide anextra degreeof freedomfor selection to act on at both the leaf and branch level

Case studies of leaf shape variation at different scales

Above we have discussed evolutionary and proximal drivers ofleaf shape as well as ecological correlates and functionalsignificance thereof Here we present case studies at threescales where we can examine the ecological and evolutionaryimplications of leaf shape These are the cross community scalecaseof leaf teeth as a subset of shapemodificationswithin lineage

200 bottomtop

150

100

50

0

0

08

07

06

05

04

03

5 10

Days after bud break

Leaf

siz

e (c

m2 )

Leaf

lobi

ng in

dex

(cm

2 cm

ndash2)

Days after bud break

15 20 25

0 5 10 15 20 25

Fig 5 Variation in (a) leaf size and (b) shape complexity inQuercus rubraleaves growing at the top and bottom of the tree crown Reprinted fromZwieniecki et al (2004b) with permission from John Wiley amp Sons

The evolution of leaf shape Functional Plant Biology 543

considerations of the genus Pelargonium and the familyProteaceae and finally the case of leaf shape variation within asingle genetic individual heteroblasty and heterophylly

Leaf teeth leaf shape variation at the community scale

Leaf teeth are conspicuous features inmanyplants and are notableas a leaf-shape trait that varies in predictable ways across plantcommunities Nearly 100 years ago Bailey and Sinnott (19151916) observed that plant communities from regions with lowmean annual temperature (MAT) have a higher site meanproportion of non-monocotyledonous species with toothedleaves Paleobotanists quickly seized upon this relationship asan index for inferring paleotemperature Little et al (2010)recently documented the rich interest in this topic compiling351 literature references related to leaf physiognomy andtemperature of which the primary focus is leaf teeth The linkbetween teeth and temperature is observed on all vegetatedcontinents (Greenwood 2005a Peppe Royer et al 2011) andis also seen in related metrics such as tooth size and tooth number(Huff et al 2003 Royer et al 2005 Peppe et al 2011) Furtherthese tooth variables can display phenotypic plasticity totemperature change (Fig 6 Royer et al 2009b)

What is knownabout the functional bases of these leafndashclimaterelationships Leaf teeth commonly expand and mature fasterthan the bulk leaf (eg Billings 1905 Feild et al 2005) and youngteeth are disproportionately vigorous in their gas exchange(Baker-Brosh and Peet 1997 Royer and Wilf 2006) Fromthese observations Royer and Wilf (2006) hypothesised thatone role of teeth is to increase sap flow thereby deliveringnutrients and other solutes to young emerging leaves Inhabitats with low temperatures teeth could play an importantrole in maximising the potential for carbon gain and growth earlyin the season In warmer climates with longer growing seasonsthe carbon benefit conferred by teeth would be outweighed bytheir unavoidable water cost Alternatively Feild et al (2005)hypothesised that teeth help remove freezendashthaw embolismsthrough positive root pressure and guttation Importantly thetransport of fluids is common to both hypotheses From this itmay be surmised that water-limiting conditions select againstteeth (Bailey andSinnott 1916) Indeed this pattern is clearly seenacross local water availability gradients (Burnham et al 2001

Kowalski and Dilcher 2003 Greenwood 2005b Royer et al2009a Peppe et al 2011) but not globally against mean annualprecipitation (eg Peppe et al 2011)

If the hypothesis by Royer andWilf (2006) is correct toothedspecies should have a higher photosynthetic rate at least at thebeginning of the growing season As such teethmay be related tothe leaf economics spectrum (Wright et al 2004) Some studiesfind support for this idea leaves with a short leaf lifespan(deciduous) (Peppe et al 2011) and low leaf mass per area(Royer et al 2005) two leaf economic variables are morelikely to be toothed It also follows that teeth should be morefunctionally related to growing season climatic variables likeseasonality growing season length and growing degree days notmean annual temperature however few studies find support forthis idea (Royer et al 2005 Peppe et al 2011) Ultimately itwould be valuable to have a better understanding of the geneticprocesses underlying tooth development but presently little isknown outside model species (see lsquoGenetic regulation forpatterning of leaf margin serration and leaflet formationrsquosection) Further phylogenetic history has typically beenassumed to have only a minor role in these leaf-climatepatterns but rigorous tests demonstrate that this assumption isfalse (Little et al 2010) This outcome is of particular concern forpaleobotanists because systematic placement of fossils can bedifficult especially for the Cretaceous and Paleogene Tools toaccommodate for this phylogenetic dependency are needed

Leaf shape variation within lineages

Proteaceae

Proteaceae is an ancient angiosperm family originating inGondwana over 100million year ago and possessing 1700 extantspecies (Weston 2007) The familyrsquos greatest representation is inAustralia followed by SouthAfrica with smaller numbers acrossother Gondwanan segments (Cowling and Lamont 1998 Hootand Douglas 1998 Weston 2007) Within the Proteaceae thereexists an enormous range of leaf sizes and shapes ranging froma few mm2 to 500 cm2 and including entire toothed broadlylobed deeply dissected needle-like simple or compound leavesIntriguingly most of this shape diversity is represented inAustralia The majority of taxa outside Australia possessentire leaves the exception being four South African genera

(a) (b) (c) (d )

Fig 6 Representative leaves of Acer rubrum showing genetic variation and plasticity in leaf teeth Leaf derived from coolclimate (Ontario) seed stock grown in (a) Rhode Island (cool climate) and (b) Florida (warm climate) leaf derived fromwarm climate (Florida) seed stock grown in (c) Rhode Island and (d) Florida Bar (for all leaves) = 1 cm Reprinted from Royeret al (2009) under Creative Commons Attribution Licence

544 Functional Plant Biology A B Nicotra et al

that possess dissected (Serruria and Paranomis) or toothed(Mimites and Leucospermum) leafed species all of whichoccur in a single clade (Sauquet et al 2009)

The reason for the extensive radiation in leaf shape diversityin Australian Proteaceae is unclear Among South AfricanProteaceae leaf size decreases with mean annual temperatureand narrow leaves have been shown to overheat less than theirwider counterparts (Yates et al 2010) Given that leaf dissectionand small size are functionally equivalent with respect to heatloss (see lsquoTemperature and waterrsquo section) it is tempting tosuggest that leaf shape variation in Australia represents anadaptive response to increasing hot dry conditions Yet thereappears to be no consistent distributional pattern of Proteaceaeleaf dimensions inAustraliaManymediumor large un-dissectedleaves exist in the hottest habitats Grevilleas with small simpleleaves can be found across most regions including high rainfallareas and some of the species found in the shaded rainforestunderstorey have among the most deeply dissected leaves of allLeaf shape also varies within a single individual For examplefern-leaf stenocarpus Stenocarpus davallioides Foreman ampBHyland has tripinnatisect juvenile leaves but adult leavescan be pinnate or bipinnate (Foreman and Hyland 1995)Other rainforest species such as Atherton Oak (Athertoniadiversifolia (CTWhit) LAS Johnson amp BGBriggs) havevery large lobed intermediate leaves but adult and seedlingleaves are small and entire (Weston 1995) Alternatively leaflobing within a plant can be random having no obvious spatialor developmental association or homogenous across a speciesExamining a phylogeny of extant taxa one finds lobed ordissected leaves alongside entire-leafed species ndash across mostlineages in the group (Weston 2007) Therefore althoughsuggestions of microclimatic adaptation may be offered atthe species level it seems unlikely that an environmentalassociation with leaf shape could be generalised acrossAustralian Proteaceae Rather leaf shape variation in theProteaceae is likely to reflect the range of ecological solutionsto leaf structurefunction trade-offs discussed above

Pelargonium

Similar to the Proteaceae the genus Pelargonium isdistributed predominantly in the southern hemisphere withnearly 80 of its almost 300 species found in South AfricaAlthough the age of the genus is estimated to be around30million years (Bakker et al 2005) the highest rate ofspecies accumulation has occurred in the last 10million years(Martinez-Cabrera 2010) The combination of climatic andedaphic heterogeneity over short distances and limited geneflow (Latimer et al 2005 Linder 2005) is frequently offeredas an explanation for the extraordinary plant species diversity thatcharacterises southern Africa Pelargonium the third largestgenus in the region is characterised by remarkable variation ingrowth form that ranges from annuals and geophytes to shrubsVariation in leaf shape is no less dramatic ranging from entireto highly dissected (Fig 1) often within growth forms andtaxonomic sections Parsing of leaf form categorically revealedthat the extent of dissection of the blade is highly evolvablewithinanoverall ovate leaf shapeoutline that is evolutionarily conserved(Jones et al 2009)

In addition to high evolutionary lability of leaf dissectionthere is variation among major clades in the evolutionary patternof major veins One major clade (Clade C) is dominated bypalmate venation whereas another (Clade A) is dominated bypinnate venation although both types are found in both cladesWe note that pinnate venation is significantly associated withincreased dissection and reductions in functional leaf sizecharacteristic of the highly diverse lsquoxerophyticrsquo clade (CladeA2) of the winter rainfall region

To examine whether leaf shape translates into functionalvariation carbon gain and water use efficiency were measuredin pairs of species contrasted for leaf shape and replicated acrossthree subclades of Clade A and one subclade of the third majorclade CladeB (Nicotra et al 2008) In each case the species withthemore dissected leaves showed higher rates of carbon gain butalso higher rates of water loss in all treatment conditions Thatpaper argued that the differences in carbon gain were not directresults of leaf shape per se but rather that the same conditionsthat result in selection for more dissected leaves also favourthe evolution of high photosynthetic rates high leaf nitrogencontent and opportunistic use of water when available Evenin this study however there was no correlation between leafdissection and the climate present in native ranges suggestingeither that climate is selective at a much smaller scale (ie themicrohabitat) or that for any given set of climate variablesmultiple leaf shapes can be adaptive in association withvariation in other leaf traits

Heteroblasty and heterophylly variationwithin a single genetic individual

We have so far focussed primarily on cross species variation inleaf shape As discussed for the Proteaceae genera above (lsquoTheProteaceaersquo section) however a single genome can produce aremarkable range of leaf shapes during the lifetime of a plantThere are two broad categories of leaf shape variation that can beexpressed by a single individual This variation can arise either inresponse to different external environments or to ontogeneticsignals (some of which may be influenced by environment aswell)

Environmentally induced heterophylly

Environmentally induced heterophylly results when leafshape responds to environmental cues The classic case is thedramatically different emergent and submergent leaves in aquaticplants Although many aquatic plants are also heteroblastic (seebelow Sculthorpe 1967) environmentally induced heterophyllyis not specific to position or plant age and is reversible dependinglargely on the environmental signals (mediated by hormonalsignals) perceived by cells of the developing leaf primordiumThe developmental age of the leaf at the time of signal perceptionis critical Very young primordia (L1ndashL3) generally respondcompletely to the novel environment whereas primordiathat were slightly older at the time of environmental changeshow intermediate features as mature leaves Characteristics ofspecific cells are related to the position of that cell within thedifferentiation gradient characteristic of that species at the timeof environmental change (eg Goliber and Feldman 1990 Bruniet al 1996 Kuwabara and Nagata 2006)

The evolution of leaf shape Functional Plant Biology 545

Among terrestrial plants the lsquosunndashshadersquo leaf response isfrequently cited as the classic example of environmentallyinduced heterophylly Historically differences in leaf formwithin a canopy have been assumed to be responses to alteredlight environments (eg larger and less-lobed leaves are producedin shade)However these different leaf shapes also confer distincthydraulic advantages (see lsquoTemperature and waterrsquo section)Whether in response to light or to hydraulic limitation lsquosunshadersquo heterophylly is accompanied by similar and well-knownanatomical differences across a wide range of species (ie sunleaves are generally smaller and thicker with smaller cells thanshade leaves Taiz and Zeiger 2002) However when theseanatomical differences become determined developmentally(Dengler 1980 Yano and Terashima 2004) has been lesscommonly studied The presence of intermediate anatomicalcharacteristics in leaves transferred among light environmentsshortly after expansion ceased suggests that leaves of somespecies have the capacity to respond anatomically to alteredlight levels until very late in development (Oguchi et al2005) In contrast recent studies suggest that systemic lightand humidity signals sent from older to younger leavesallow younger leaves to develop phenotypes suited for theiranticipated environments (Lake et al 2001 Yano andTerashima 2001 Thomas et al 2004) Exactly when final leafshape is determined in lsquosunshadersquo leaves remains an openquestion for most species

Heteroblasty describes ontogenetically dependentchanges in leaf features most notably shape

In 1900 Goebel originally distinguished differences in lsquojuvenileformrsquo from lsquoadult formrsquo on the basis of dramatic differences inleaf shape and associated traits (Balfour 1969) As an example he

described the shift from compound leaves to phyllodes in Acaciaas heteroblastic and reserved the term homoblastic for the lessdramatic variation exhibited by Casurina It is now generallyaccepted that heteroblasty refers to the full suite of features thatchange along the shoot during development (Jones 1999)

Heteroblasty results from the interaction of developmentalprograms expressed at both the leaf and the shoot- or whole-plant level (Kerstetter and Poethig 1998 Tsukaya et al 2000Usami et al 2009 Wu et al 2009 Chuck and OrsquoConnor 2010)Heteroblasty is generally not reversible unless reiterative growthoccurs (ie following damage) Morphological differencescharacteristic of heteroblasty such as differences in lobing areusually expressed relatively early in development shortly afterleaf initiation (Jones 1995)

The sequence of heteroblastic leaf shapes produced alongthe shoot can be accelerated or delayed in response to differentgrowth environments or architectural manipulations (Ashby1948 Forster and Bonser 2009 Jaya et al 2010)Furthermore both heteroblasty and plasticity in heteroblastyhave been shown to be under strong genetic control (Anderson1989 Climent et al 2006) Forster et al (2011) suggested thatboth heteroblasty and plasticity in heteroblasty are likely to occurin response to lsquowhich leaf type confers the greatest benefit fora given light levelrsquo Several studies over the last decade havepointed towards functional consequences of heteroblastygenerally with reference to single environmental variables (egHansen 1986 Gamage and Jesson 2007 Kubien et al 2007)It is likely that heteroblasty influences multiple leaf functionsbecause suites of leaf traits change with heteroblasty (Jones2001) Indeed this conclusion was supported in a recent studythat examined differences inmore than a hundred leaf traits in koa(Acacia koa AGray Pasquet-Kok et al 2010) Heteroblasty hasalso been proposed as an adaptation to herbivory (Karban and

Table 1 Promising directions for cross-disciplinary investigation of leaf shape

No Direction

1 A multi-species lsquoevo-devorsquo approach similar to that by Illing et al (2009) is likely to be informative with regard to many enigmatic features regardingthe evolution of diverse leaf forms and more generally provides an exciting system for examining the convergence and divergence in the evolutionof function

2 A phylogenetic analysis of leaf shape examining correlated evolution with other traits such as venation size leaf surface characteristics and branch andarchitectural measures would enable a thorough investigation of the situations in which leaf shape diversification arises

3 Further research into both the genetic and developmental processes underlying lobes and teeth and the ecological correlates of each would provide anexcellent opportunity to link across wide environmental scales

4 Cross species studies ideally in the field of the relative importance of leaf shape among other factors in determining leaf temperature and water status arenecessary to test the conclusion put forward here that the functional significance of leaf shape lies more often in water relations than directly in leafthermoregulation

5 Studies of the scale of environmental change that selects for ontogenetic response in leaf shape in heteroblastic species fromdiverse environments will shedlight on links between genes development and ecology

6 Detailed quantifications of the extent towhich leaf shape differswithin canopies particularly of dioecious specieswhere selection on reproductive functioncan lead to dimorphic conditions will help link leaf shape to whole plant processes

7 Studies at the scale of transects across the leaf surface that simultaneously consider hydraulic resistance temperature and light will reveal the extent oflocalised hydraulic and thermal stresses within leaf surfaces that might act as selective forces on leaf shape per se Imaging approaches incorporatingmeasures of 3-dimensional shape temperature and fluorescence provide exciting opportunities for examining spatial heterogeneity in temperature andphysiological processes

8 Since patterns of leaf venation affect both hydraulic supply and capacity for photosynthate transport studies that simultaneously examine both willreveal the extent to which interactions among these fundamental processes either co-determine leaf shape or establish trade-offs ultimately reflectedin leaf shape

9 Finally studies examining the fitness values of different leaf shapes or plasticity in leaf shapes under a range of environmental conditions are needed tobetter identify which aspects of leaf shape are most likely to be adaptive

546 Functional Plant Biology A B Nicotra et al

Thaler 1999 Brennan et al 2001 Fadzly et al 2009) and it hasbeen associated with differences in frost resistance (Darrow et al2001) It is worth noting that studies of the functional significanceof heteroblasty have focussed on relatively few plant lineagesGiven that heteroblasty occurs widely among the angiospermsas well as in some ferns and gymosperms (eg Wagner 1952Mueller 1982 Pryer and Hearn 2009) the next questions mayfocus on the cost of heteroblasty and its fitness consequenceswhat scale of environmental change selects for ontogeneticresponse in leaf shape

Synthesis and next steps

Whether considered at community lineage or single genomescales leaf shape is a trait that is highly labile and responsive to arange of biotic and abiotic factors Although some leaf traitscan be explained in terms of particularly strong trade-offsbetween two or more factors that result in clear scalingrelationships (eg leaf mass-to-area traits or leaf mass-to-stemmass) leaf shape in contrast does not emerge from the aboveconsideration as a factor capable of explaining ecologicaldifferentiation between species Rather leaf shape emerges asa trait for which there are many quite varied functional trade-offs

As stated above these trade-offs for successful functionalityof leaves may generate ecological opportunities leaf shape isperhaps best viewed not as a single major axis but rather asan option that fine tunes the leaf to its conditions over bothshort and evolutionary time spans Numerous elements of thisreview support this conclusion Over evolutionary history theextant range of variation in leaf shape has been explored bymanylineages although only the angiosperms have explored thisrange with the added element of massive increases in veindensities Likewise our growing understanding of geneticcontrols on leaf shape reveal that it is a complex traitdetermined by many different genes at several differentdevelopmental stages Although some genes have conservedaffects on leaf morphology including shape the exact natureof their affects can vary widely Likewise the function to whichagivenelement of leaf shape lsquois putrsquo canvary For example closerinvestigation of the internal architecture and developmentalpathways associated with different leaf shapes suggests thatleaf teeth and leaf lobes often have distinctly differentfunctional foundations and may play different functions indifferent environments

Despite amyriadof functional roles the one recurrent theme inour review is that leaf shape is affected by and strongly influencesleaf water relations From a functional perspective the role of leafshape in leaf thermoregulation has perhaps received the mostattention in previous considerations of the functional significanceof leaf shape The relationship between leaf size and temperaturebeing physically determined by basic energy balance principlesthis was a logical starting point of enquiry But as we haveshown the connection between leaf shape (and size) and leaftemperatures under field conditions is not well supportedempirically Rather leaf shape variation as distinct from sizemore often will reflect water supply trade-offs High veindensities support high photosynthetic rates (Brodribb et al2010) and dissected leaves maintain a greater proportion ofleaf tissue closer to main veins Leaf temperature plays a role

in this story as higher leaf temperatures lead to greaterevaporative water loss but many factors other than leaf shapeinfluence leaf temperature (eg leaf thickness orientation andreflectivity and branch and plant architecture) Finally it is alsoworth noting that although there appears to be an obviouslink between shape and water supply there is a concomitantinteraction between water supply and phloem function thatremains relatively unexplored

Thus our review suggests several angles for further research(Table 1) In particular we advocate those approaches thatenable investigation across the scales examined here (genedevelopment function and evolutionary ecology) and thatfocus in particular on the links among leaf shape developmentand water relations

Acknowledgements

The authors thank the Botanical Society of America for their support of thesymposiumfromwhich this reviewaroseWealso thankLawrenSackand twoanonymous reviewers for useful comments on an earlier version of thismanuscript

References

Ackerly DD Knight CA Weiss SB Barton K Starmer KP (2002) Leafsize specific leaf area and microhabitat distribution of chaparral woodyplants contrasting patterns in species level and community level analysesOecologia 130 449ndash457 doi101007s004420100805

Anderson S (1989) Variation in heteroblastic succession among populationsof Crepis tectorum Nordic Journal of Botany 8 565ndash573 doi101111j1756-10511989tb01730x

Armbruster WS Muchhala N (2009) Associations between floralspecialization and species diversity cause effect or correlationEvolutionary Ecology 23 159ndash179 doi101007s10682-008-9259-z

Armbruster WS Di Stilio VS Tuxill JD Flores TC Runk JLV (1999)Covariance and decoupling of floral and vegetative traits in nineneotropical plants a re-evaluation of Bergrsquos correlation-pleiadesconcept American Journal of Botany 86 39ndash55 doi1023072656953

Ashby E (1948) Studies in the morphogenesis of leaves I An essay on leafshape New Phytologist 47 153ndash176 doi101111j1469-81371948tb05098x

Aso K Kato M Banks JA Hasebe M (1999) Characterization ofhomeodomain-leucine zipper genes in the fern Ceratopteris richardiiand the evolution of the homeodomain-leucine zipper gene family invascular plants Molecular Biology and Evolution 16 544ndash552

Bailey IW Sinnott EW (1915) A botanical index of Cretaceous and Tertiaryclimates Science 41 831ndash834 doi101126science411066831

Bailey IW Sinnott EW (1916) The climatic distribution of certain types ofangiosperm leaves American Journal of Botany 3 24ndash39 doi1023072435109

Baker-Brosh KF Peet RK (1997) The ecological significance of lobed andtoothed leaves in temperate forest trees Ecology 78 1250ndash1255

Bakker FT Culham A Marais EM Gibby M (2005) Nested radiation inCape Pelargonium In lsquoPlant species-level systematics new perspectiveson pattern and process Vol 143rsquo (Eds FT Bakker LW Chartrou BGravendeel PB Pielser) pp 75ndash100 (ARG Ganter Verlag KG RuggellLichtstein)

Balfour IB (Transl) (1969) lsquoOrganography of plants I General organographyrsquoby K Goebel 1900 (Hafner Publishing Co Inc New York)

Ball MC Cowan IR Farquhar GD (1988) Maintenance of leaf temperatureand the optimization of carbon gain in relation to water loss in a tropicalmangrove forest Australian Journal of Plant Physiology 15 263ndash276doi101071PP9880263

The evolution of leaf shape Functional Plant Biology 547

LONGIFOLIA1 (LON1) and LON2 regulate the length of the leaflamina in Arabidopsis (Lee et al 2006) A vast array of genes isknown to influence leaf area (Horiguchi et al 2006) For exampleheteroblastic change in leaf area is governed by the miR156-SQUAMOSAPROMOTERBINDINGPROTEIN-LIKE (SPL)-miR172 pathway via regulation of cell size and cell numbers inArabidopsis (Usami et al 2009)Alterations of activities of anyofthese factors could contribute to natural variation in leaflength width ratio and area None of these genetic variationsresult in loss of reproductive structures suggesting that thesegenes could be under direct environmentally mediated selectionAs yet we do not know whether the above patterns hold in othernon-model species

Leaves also vary in the extent to which the lamina is flator curved CINCINATA (CIN) of snapdragon was oncethought to function to keep the leaf flat by orchestrating cellproliferation along the medio-lateral axis of the leaf primordia(Nath et al 2003) More recently however analyses of theTCP genes which are homologues of CIN in Arabidopsisindicate that the major role of the TCP is in promotion ofmaturation in laminar tissues not in regulation of cellproliferation (Efroni et al 2008 Sarojam et al 2010) The roleof CIN in snapdragon may also be the same as that of TCPsin Arabidopsis The other candidate genetic mechanism forcontrol of leaf curvature is an auxin-related pathway Severalhyponastic leaf mutants have been isolated from Arabidopsisand many of the responsible genes are related to auxinsignalling (Perez-Perez et al 2002 2009) Since auxinsignalling is also very important for venation patterningspecies-specific modification of auxin signalling in leafprimordia may have caused diversification of both leafcurvature and vasculature patterning

Genetic regulation for patterning of leaf marginserration and leaflet formation

Variation in leaf shape as distinct from length and area orcurvature arises from variation in leaf margin and lobeformation Kawamura et al (2010) showed that spatialpatterning of serrations or leaf teeth (see lsquoLeaf shape variationwithin lineagesrsquo section) on the leaf margin depends onformation of auxin maxima on the leaf margin in ArabidopsisLocation of these maxima is governed by polar auxin transportand the stabilisation of auxinmaxima ismaintained by prolongedexpression of CUP-SHAPED COTYLEDON2 (CUC2) around it(Kawamura et al 2010) The mechanism of rhythmic patternformation of the auxin maximaminima on the leaf margin isthought to involve self-organising pattern formationmechanismsby polar auxin transport similar to that involved in patternformation of phyllotaxy (de Reuille et al 2006 Jonsson et al2006) Either the loss of distinct auxinmaxima or an instability ofthe auxin maxima will result in an entire leaf margin (Kawamuraet al 2010) Changes in functions of these genes howeverseverely influence many biological processes including floralorgan development as seen in the pin-formed 1 mutant (Okadaet al 1991) Thus variation among species in pattern of marginserration might accompany variation in floral organs or morelikely reflect differential regulation for example by downstreamgenetic pathways

Combined control of auxin-maxima formation by PIN andCUC genes as described for serration above is also known tohave an important role in leaflet patterning in bittercress(Cardamine flexuosa With) an Arabidopsis relative withcompound leaves (Barkoulas et al 2008) Again this controlvaries among species in tomato (Solanum lycopersicum L) lossof CUC3 function does not convert compound leaves intosimple leaves (Blein et al 2008) Thus it is clear that theindependent evolution of compound leaves from simple leavesis likely driven by species-specific mechanisms (Efroni et al2010) though these may reflect modifications of similarpathways

Recent studies have identified several important key genes forregulation of compound versus simple leaf forms For exampleabsence or prolonged expression of class I Knotted1-likehomeobox (KNOX) genes in leaf primordia is important tocompound leaf formation in a wide range of angiosperms andgymnosperms (Bharathan et al 2002) In some legume specieshowever the LEAFY (LFY)UNIFOLIATA (UNI) gene is the keyfor compound leaf formation instead of the class I KNOX genes(Hofer et al 1997) Changes in these genes could be linked tothe evolution of compound leaves although loss-of-function ofLFYUNI also causes infertility in Arabidopsis and thus it isunlikely that selection directly on these particular genes couldresult in theevolutionof variation in leaf shapeDiversified spatialpatterning of petiolepetiolule and laminaleaflet in compoundleaves have also been suggested to be linked to the distributionpattern of AS1-ROUGH SHEATH2-PHANTASTICA (ARP)expression domains (Kim et al 2003) However based on thefact that the ARP gene has only a limited role in compound leafformation in some species it is unclear whetherARPwill turn outto be major factor in others (Efroni et al 2010)

As described above the independent evolutionary origins ofseveral angiosperm leaf features should caution against theassumption that a single ancestral genetic system underlies leafdevelopment The developmental genetic approaches outlinedhave revealed several key mechanisms for leaf shape controlbut have also demonstrated variation in the function of thesegenes among lineages In a few studies researchers have gonebeyond one or two species studies and have looked for deeperevolutionary links (Illing et al 2009) Studies of the role of theclass 1 KNOX genes have considered not just angiospermsbut other lineages Among simple-leafed dicots and monocots(eg Arabidopsis and Zea) KNOX genes are strongly expressedduring the indeterminate growth of the apical meristem but arerapidly downregulated in the cells flanking the apical domein regions defined by leaf primordia Although some arginine-rich protein (ARP) genes repress the expression of KNOXtranscription factors during the development of both micro-and megaphyll primordia (Harrison et al 2005) the role ofhomeodomain-leucine zipper (HD-ZIP) III genes which areinvolved in the specification of foliar adaxialabaxial fatediffers in the development of these two leaf types (Floyd andBowman 2006) HD-ZIP III genes have been identified in themoss Physcomitrella (Sakakibara et al 2001) and in the fernCeratopteris (Aso et al 1999) they are also expressed duringnormal vascular tissue development in Arabidopsis (Zhong andYe 1999) suggesting that they are very ancient but alsomultifunctional

540 Functional Plant Biology A B Nicotra et al

Accumulating knowledge on molecular genetic mechanismsof leaf morphogenesis is leading to an increasingly evo-devoapproach to studies of leaf diversity (Tsukaya 2010)this progress is aided by establishing new model species(eg Cardamine hirsuta and the rush Canales et al 2010Yamaguchi and Tsukaya 2010) Over the next decade weanticipate that genomic approaches on multiple speciesanalysed in phylogenetic frameworks (eg Illing et al 2009)will resolvemuchof theenigma regarding the evolutionofdiverseleaf forms

Genetic constraints on flower petal and leaf form

One additional developmental consideration is that as famouslydescribed by von Goethe in 1790 flower petals are derivatives ofthe apical meristem and serial homologues to leaves (von Goethe2009) Thus one might hypothesise that although the evolutionof leaves substantially predates that of flowers selection on theform of either of these organs would also result in changes inthe other such that selection on leaf shape and size subsequent tothe evolution of floral organs may be constrained by pleiotropiccontrol of floral form Recent genetic data identifies several genesthat have roles in both leaf and flower form (Dinneny et al 2004Schmid et al 2005 Street et al 2008) For example geneticchanges in both leaf length to width ratio (eg rot3) and leaf size(eg an3) influence not only leaf proportions but also that of thefloral organs in Arabidopsis (Kim et al 1999 Horiguchi et al2005) The garden pea (Pisum sativum L) is a good case in pointas discussed above compound leaf formation depends on LFYUNI but the same gene is also required for flowering Thus ifcurrent selection favoured simple leaves in garden pea changes inleaf shape could be constrained by pleiotropic control offlowering (Hofer et al 1997) Likewise Shalit et al (2009)recently reported that the action of florigen ndash a plant hormonefor flowering ndash influences complexity of leaves by changingnumbers of leaflets in tomato Such relationships suggest thatsome floral characters as well as other metabolic pathways maynowbeclosely linkedwith regulationof leaf shapeand sizeA fewecological studies have also assessed correlations between petaland leaf form (Berg 1960 Armbruster et al 1999) This questionis of particular interest in lineages where animal pollination isassociated with increased diversification rates either as a result ofdirect selection or as a secondary reinforcing factor (van der Nietet al 2006 Armbruster and Muchhala 2009 Kay and Sargent2009)

In situations where leaf and petal form are under similargenetic control and both under selection one of two thingscould happen If leaf shape does not have substantialfunctional impact then one might expect leaf form to changewith selection on petal form However given the likelyimportance of leaf form to photosynthetic function it is morelikely that a breakdown in genetic correlations between leafand flower form (eg by alterations of downstream regulatorypathways) would evolve Several factors argue against thehypothesis that leaf shape is frequently a product of selectionon flower form First diverse leaf shapes were in evidencelong before the origin of flowers in the angiosperms Secondthe genetic mechanisms underlying leaf shape revealconsiderable evolvability with similar genes being co-opted

for different functions in different species Finally given thepotential importance of both petal and leaf shape we suggestthat genetic constraints on the independent evolution of bothstructures are likely to have broken down in those lineages withhighly diversified leaf shapes over the many millions of yearssince petals first evolved

Ecological correlates and the functional significanceof leaf shape

Evolutionary history plays a part in determining leafmorphology but our consideration of leaf shape evolution andits developmental genetic basis demonstrates that diverselineages have repeatedly gained and lost many leaf featuresThus developmental and genetic constraints on the evolutionof leaf form and shape in particular have been overcomemany times or have led to new evolutionary opportunities Itis of particular interest then to ask what are the likely ecologicaldrivers andor functional significance of leaf shape

Temperature and water

Conventionalwisdomwould say that thermoregulation is perhapsthe main driver of leaf shape evolution The maintenance ofleaf temperatures within certain limits is critical for a plantrsquosgrowth and survival and although photosynthetic tissue of somespecies canwithstand temperatures below6C (Ball et al 2006)and above 60C (Clum1926Nobel 1988) irreversible damage toleaves can occur at temperatures well within this range (Levitt1980 Jones 1992 Ball et al 2004 Groom et al 2004 Sharkey2005) Leaf size and shape potentially have large effects onleaf temperature because the two-dimensional proportions of aflattened leaf determine the rate of heat transfer between acrossthe leafndashair interface by influencing the thickness of its boundarylayer

All else being equal the thickness of a boundary layerincreases with length from the windward edge so that heatconvection from small leaves is more rapid than from largeleaves (Raschke 1960 Gates 1968 Vogel 1970 Parkhurst andLoucks 1972Grace et al 1980Geller andSmith 1982Monteithand Unsworth 1990 Schuepp 1993) Likewise because leaflobing reduces the distance across the lamina the rate of heattransfer is predicted to be greater in a lobed leaf than an unlobedleaf with equivalent area (Parkhurst et al 1968 Vogel 1968Lewis 1972 Givnish 1978 Gurevitch and Schuepp 1990a)

The rate of heat transfer from lobed metal plates or lobedleaves coated in metal is greater than from shallow-lobed or un-lobed ones (Parkhurst et al 1968 Thom 1968 Vogel 1970Gottschlich and Smith 1982 Gurevitch and Schuepp 1990bRoth-Nebelsick 2001) Artificial or coated leaves have beenused because they enable boundary layer resistance to beinvestigated in the absence of transpiration However thisapproach is biologically flawed because it does not enable anassessment of the relative importance of a range of traits (eg leafwater content absorbance) in determining the actual operatingtemperatures of leaves in the field

Actual leaf temperature measurements show that large leavescan operate at below-ambient temperatures even on hot days asa result of high rates of latent heat loss through transpiration(Drake et al 1970 Gates 1980 Hegazy and El Amry 1998)

The evolution of leaf shape Functional Plant Biology 541

Thus conditions of evaporative demand and the availability ofsoil water for transpiration are key to understanding leafshapendashtemperature relationships under biologically realisticconditions Because of their much greater transpirationalcapacities (see above) flowering plants have far greaterleeway than other plants regarding the size and shape of theirleaves (see lsquoA brief history of the angiosperm leaf and shapediversity thereinrsquo section Boyce et al 2009)

In addition to reducing heat loads the morphology of deeplylobed leaves may also reflect direct selection for increasedhydraulic efficiency Water moving through a plant encountersnumerous resistances on its path from soil to roots to stemsthrough leaves and out to the external atmosphere Within theliquid phase of water flow through a plant leaf hydraulicresistance Rleaf accounts for more than 30 of total plantresistance (Sack and Holbrook 2006) The high resistance inleaves is due first to a system of veins of decreasing size theresistance of which is inversely proportional to the fourth powerof the radius of the component conduits in these veins Thereforeas a proportion of the considerable resistance in the leaf(Zwieniecki et al 2002) the minor veins provide the greatestresistance to water flow (Sack et al 2004)Water also encountershigh resistance as it exits veins laterally and travels to othercells mainly via apoplastic pathways through the mesophyllAccordingly on the order of half of total Rleaf can occur outsidethe major veins (Sack and Holbrook 2006) As resistanceincreases along the pathway towards sites of evaporationlocalised water potential (Y) becomes progressively morenegative (Brodribb et al 2010) The result of how thisphenomenon might affect a leaf has long been observedregions of leaves lying furthest from the main supply channelsare the first to lsquowitherrsquo when exposed to strong winds (Fig 4Yapp 1912) Leaf lobing represents an effective removal of thispotentially stress-prone tissue and has been suggested as anadaptation to dry conditions (Thoday 1931 Givnish 1979) Ifdeeply lobed leaves have a lower ratio of mesophyll tissue tolarge highly conductive veins they should then have reducedresistance relative to less- or un-lobed leaves (Sack and Tyree2005)

Although leaf lobing might confer adaptive benefits withrespect to maintaining stable hydraulic supply it may also bethat the extent of lobing or margin dissection is itself determinedby hydraulic limitation (Sack et al 2003 Zwieniecki et al2004b Boyce 2009) For species with lsquosunrsquo and lsquoshadersquoleaves the morphology of small deeply lobed sun leaves atthe outer canopy is proposed to result from water pressuredrop across the leaf lamina caused by reduced water deliveryto expanding cells during growth (Fig 5 Zwieniecki et al 2004bBoyce 2009 Leigh et al 2011)Alternatively recent studies havesuggested that many of the characteristics of lsquosunrsquo and lsquoshadersquoleaves are the result of branch autonomy within the plant canopy(Niklas and Cobb 2010)

Thus although the effects of leaf shape on leaf thermalregulation hold when tested on model leaves predictions areless straightforward when hydraulic function is accounted forAdditionally consideration of the relative contribution of a rangeof leaf and branch properties to leaf temperature is critical It islikely that leaf shape is only one amongmany factors influencingleaf thermal regulation other factors could include water

content leaf thickness spectral reflectance orientation andplant architecture Further given the observation of high leafshape diversity withinmany plant communities from hot and aridenvironments it seems likely that regulating leaf temperature isnot the singlemost important evolutionary influence on leaf shapediversity The association between leaf shape and hydraulicproperties which is affected by leaf temperature in contrast islikely to be of greater importance

Leaf structure and functional trade-offs

Many foliar leaf traits including shape reflect functional trade-offs that have been resolved in various ways by differentspecies depending on their ecological settings (eg hydro- vsmesophytes) physiological attributes (eg C3 vsC4metabolism)developmental capabilities (eg heterobaric vs non-heterobaricleaves) or evolutionary histories (eg micro- vs megaphylls)Life history and optimisation theory show that the number ofphenotypic solutions that allow for different equally successfultrait combinations increases as the number of trade-offsincreases ndash a conclusion that applies to traits within the leaf(eg for shape) as well as to leafndashbranch relationships (Niklas1988)

It is evident that leaves perform several functionssimultaneously that is they intercept light transport liquidsand solutes exchange gases with the atmosphere dissipateheat cope with externally applied mechanical forces and

(a)

(e)(f )

( i )

(g)

(h)

(b)

(c)

(d )

Fig 4 Localised lsquowitheringrsquo of leaves exposed to strong wind Reprintedfrom Yapp (1912) with permission from Oxford University Press

542 Functional Plant Biology A B Nicotra et al

defend themselves from herbivores and pathogens Importantlymany of these functions have conflicting physiologicalanatomical or morphological requirements For exampleorientations of lamina that maximise light interception alsominimise the ability to dissipate heat Further all of thesefunctions are intrinsically size dependent For example thequantity of mechanical tissues required to support laminaedepends on the mass of laminae whereas the ability tointercept light depends on laminae area

When seen in this light the trade-offs required for thesuccessful functionality of foliar leaves generate rather thanconstrain ecological and evolutionary opportunities This canbe illustrated mathematically and empirically in the context ofphyllotaxy and its consequences on light interception Computersimulations of mathematically generated shoots to assess theinfluence of leaf shape size and phyllotactic patterns on theability to intercept direct solar radiation show that differences inphyllotaxy significantly influence light interception (particularlyfor shoots with a rosette growth habit) They also show thatcomparatively small differences in leaf shape can compensatefor the negative effects of leaf overlap resulting from virtuallyany phyllotactic pattern For example lobed leaves or pinnifid

compound leaves facilitate light penetration through shootsbearing densely packed leaves Thus phyllotaxy can beviewed as a developmental limiting factor that can drivecompensatory changes in morphological features such asshape that are not directly controlled by patterns of leafinitiation In this way the lsquoconstraintsrsquo imposed by phyllotaxycan foster evolutionary modifications at the level of the entireshoot (Niklas 1988)

One particularly important trade-off operating at the level ofthe architecture of a shoot is that between leaf size and numberFor any given leaf biomass allocation an individual shoot caneither have a few large leaves or many smaller ones Acrossdifferent herbaceous species and juvenile specimens of treespecies drawn from diverse lineages and ecological settingsthe relationship between total leaf biomass per plant (ML) andtotal stemmass per plant (MS) fallswithin a comparatively narrowcorridor of variation that obeys a positive isometric scalingrelationship (ie MLMS

~10 Niklas 2004) Accordinglyacross these species any stem mass investment yields a moreor less proportional increase in leaf mass Therefore it followsmathematically that across these species the relationshipamong the average mass of a typical leaf mL total leaf numberNL and total stem mass complies with the proportionality mL

NLMS~10 When individual shoots are sampled across

diverse species the relationship between mL and NL per shootis negative and isometric (ie mLNL

ndash1 Kleiman and Aarssen2007) a relationship that appears to be insensitive to howspecies are grouped according to other functional traits oraccording to their habitat preference (Yang et al 2008) Thatthis phenomenology reflects a biomass allocation trade-off isclear What remains uncertain is whether it is a simple trade-offbetween leaf mass investment and leaf size or a more complexset of relationships imposed by twig mass investments andhydraulics or biomechanics

Another trade-off that appears to hold across many but not allspecies is the relationship between leaf surface areaA and averageleaf mass mL For more than a few species-groupings thisrelationship is allometric with a slope that is less than onewhich indicates that increases in leaf mass investment do notresult in proportional increases in leaf area (ieAmL

lt10Niklaset al 2007) This trend of lsquodiminishing returnsrsquo has also beenreported when total leaf area per plant is plotted against total leafmass for plants differing in size (Niklas and Cobb 2008 2010)Several factorsmay contribute to these trends eg biomechanicalanalyses show that disproportionately larger investments must bemade to mechanically support photosynthetic tissues as leavesincrease in their surface area (Niklas et al 2009) In both the leafmass to stem mass and leaf mass to area relationships describedabove leaf shapevariation canprovide anextra degreeof freedomfor selection to act on at both the leaf and branch level

Case studies of leaf shape variation at different scales

Above we have discussed evolutionary and proximal drivers ofleaf shape as well as ecological correlates and functionalsignificance thereof Here we present case studies at threescales where we can examine the ecological and evolutionaryimplications of leaf shape These are the cross community scalecaseof leaf teeth as a subset of shapemodificationswithin lineage

200 bottomtop

150

100

50

0

0

08

07

06

05

04

03

5 10

Days after bud break

Leaf

siz

e (c

m2 )

Leaf

lobi

ng in

dex

(cm

2 cm

ndash2)

Days after bud break

15 20 25

0 5 10 15 20 25

Fig 5 Variation in (a) leaf size and (b) shape complexity inQuercus rubraleaves growing at the top and bottom of the tree crown Reprinted fromZwieniecki et al (2004b) with permission from John Wiley amp Sons

The evolution of leaf shape Functional Plant Biology 543

considerations of the genus Pelargonium and the familyProteaceae and finally the case of leaf shape variation within asingle genetic individual heteroblasty and heterophylly

Leaf teeth leaf shape variation at the community scale

Leaf teeth are conspicuous features inmanyplants and are notableas a leaf-shape trait that varies in predictable ways across plantcommunities Nearly 100 years ago Bailey and Sinnott (19151916) observed that plant communities from regions with lowmean annual temperature (MAT) have a higher site meanproportion of non-monocotyledonous species with toothedleaves Paleobotanists quickly seized upon this relationship asan index for inferring paleotemperature Little et al (2010)recently documented the rich interest in this topic compiling351 literature references related to leaf physiognomy andtemperature of which the primary focus is leaf teeth The linkbetween teeth and temperature is observed on all vegetatedcontinents (Greenwood 2005a Peppe Royer et al 2011) andis also seen in related metrics such as tooth size and tooth number(Huff et al 2003 Royer et al 2005 Peppe et al 2011) Furtherthese tooth variables can display phenotypic plasticity totemperature change (Fig 6 Royer et al 2009b)

What is knownabout the functional bases of these leafndashclimaterelationships Leaf teeth commonly expand and mature fasterthan the bulk leaf (eg Billings 1905 Feild et al 2005) and youngteeth are disproportionately vigorous in their gas exchange(Baker-Brosh and Peet 1997 Royer and Wilf 2006) Fromthese observations Royer and Wilf (2006) hypothesised thatone role of teeth is to increase sap flow thereby deliveringnutrients and other solutes to young emerging leaves Inhabitats with low temperatures teeth could play an importantrole in maximising the potential for carbon gain and growth earlyin the season In warmer climates with longer growing seasonsthe carbon benefit conferred by teeth would be outweighed bytheir unavoidable water cost Alternatively Feild et al (2005)hypothesised that teeth help remove freezendashthaw embolismsthrough positive root pressure and guttation Importantly thetransport of fluids is common to both hypotheses From this itmay be surmised that water-limiting conditions select againstteeth (Bailey andSinnott 1916) Indeed this pattern is clearly seenacross local water availability gradients (Burnham et al 2001

Kowalski and Dilcher 2003 Greenwood 2005b Royer et al2009a Peppe et al 2011) but not globally against mean annualprecipitation (eg Peppe et al 2011)

If the hypothesis by Royer andWilf (2006) is correct toothedspecies should have a higher photosynthetic rate at least at thebeginning of the growing season As such teethmay be related tothe leaf economics spectrum (Wright et al 2004) Some studiesfind support for this idea leaves with a short leaf lifespan(deciduous) (Peppe et al 2011) and low leaf mass per area(Royer et al 2005) two leaf economic variables are morelikely to be toothed It also follows that teeth should be morefunctionally related to growing season climatic variables likeseasonality growing season length and growing degree days notmean annual temperature however few studies find support forthis idea (Royer et al 2005 Peppe et al 2011) Ultimately itwould be valuable to have a better understanding of the geneticprocesses underlying tooth development but presently little isknown outside model species (see lsquoGenetic regulation forpatterning of leaf margin serration and leaflet formationrsquosection) Further phylogenetic history has typically beenassumed to have only a minor role in these leaf-climatepatterns but rigorous tests demonstrate that this assumption isfalse (Little et al 2010) This outcome is of particular concern forpaleobotanists because systematic placement of fossils can bedifficult especially for the Cretaceous and Paleogene Tools toaccommodate for this phylogenetic dependency are needed

Leaf shape variation within lineages

Proteaceae

Proteaceae is an ancient angiosperm family originating inGondwana over 100million year ago and possessing 1700 extantspecies (Weston 2007) The familyrsquos greatest representation is inAustralia followed by SouthAfrica with smaller numbers acrossother Gondwanan segments (Cowling and Lamont 1998 Hootand Douglas 1998 Weston 2007) Within the Proteaceae thereexists an enormous range of leaf sizes and shapes ranging froma few mm2 to 500 cm2 and including entire toothed broadlylobed deeply dissected needle-like simple or compound leavesIntriguingly most of this shape diversity is represented inAustralia The majority of taxa outside Australia possessentire leaves the exception being four South African genera

(a) (b) (c) (d )

Fig 6 Representative leaves of Acer rubrum showing genetic variation and plasticity in leaf teeth Leaf derived from coolclimate (Ontario) seed stock grown in (a) Rhode Island (cool climate) and (b) Florida (warm climate) leaf derived fromwarm climate (Florida) seed stock grown in (c) Rhode Island and (d) Florida Bar (for all leaves) = 1 cm Reprinted from Royeret al (2009) under Creative Commons Attribution Licence

544 Functional Plant Biology A B Nicotra et al

that possess dissected (Serruria and Paranomis) or toothed(Mimites and Leucospermum) leafed species all of whichoccur in a single clade (Sauquet et al 2009)

The reason for the extensive radiation in leaf shape diversityin Australian Proteaceae is unclear Among South AfricanProteaceae leaf size decreases with mean annual temperatureand narrow leaves have been shown to overheat less than theirwider counterparts (Yates et al 2010) Given that leaf dissectionand small size are functionally equivalent with respect to heatloss (see lsquoTemperature and waterrsquo section) it is tempting tosuggest that leaf shape variation in Australia represents anadaptive response to increasing hot dry conditions Yet thereappears to be no consistent distributional pattern of Proteaceaeleaf dimensions inAustraliaManymediumor large un-dissectedleaves exist in the hottest habitats Grevilleas with small simpleleaves can be found across most regions including high rainfallareas and some of the species found in the shaded rainforestunderstorey have among the most deeply dissected leaves of allLeaf shape also varies within a single individual For examplefern-leaf stenocarpus Stenocarpus davallioides Foreman ampBHyland has tripinnatisect juvenile leaves but adult leavescan be pinnate or bipinnate (Foreman and Hyland 1995)Other rainforest species such as Atherton Oak (Athertoniadiversifolia (CTWhit) LAS Johnson amp BGBriggs) havevery large lobed intermediate leaves but adult and seedlingleaves are small and entire (Weston 1995) Alternatively leaflobing within a plant can be random having no obvious spatialor developmental association or homogenous across a speciesExamining a phylogeny of extant taxa one finds lobed ordissected leaves alongside entire-leafed species ndash across mostlineages in the group (Weston 2007) Therefore althoughsuggestions of microclimatic adaptation may be offered atthe species level it seems unlikely that an environmentalassociation with leaf shape could be generalised acrossAustralian Proteaceae Rather leaf shape variation in theProteaceae is likely to reflect the range of ecological solutionsto leaf structurefunction trade-offs discussed above

Pelargonium

Similar to the Proteaceae the genus Pelargonium isdistributed predominantly in the southern hemisphere withnearly 80 of its almost 300 species found in South AfricaAlthough the age of the genus is estimated to be around30million years (Bakker et al 2005) the highest rate ofspecies accumulation has occurred in the last 10million years(Martinez-Cabrera 2010) The combination of climatic andedaphic heterogeneity over short distances and limited geneflow (Latimer et al 2005 Linder 2005) is frequently offeredas an explanation for the extraordinary plant species diversity thatcharacterises southern Africa Pelargonium the third largestgenus in the region is characterised by remarkable variation ingrowth form that ranges from annuals and geophytes to shrubsVariation in leaf shape is no less dramatic ranging from entireto highly dissected (Fig 1) often within growth forms andtaxonomic sections Parsing of leaf form categorically revealedthat the extent of dissection of the blade is highly evolvablewithinanoverall ovate leaf shapeoutline that is evolutionarily conserved(Jones et al 2009)

In addition to high evolutionary lability of leaf dissectionthere is variation among major clades in the evolutionary patternof major veins One major clade (Clade C) is dominated bypalmate venation whereas another (Clade A) is dominated bypinnate venation although both types are found in both cladesWe note that pinnate venation is significantly associated withincreased dissection and reductions in functional leaf sizecharacteristic of the highly diverse lsquoxerophyticrsquo clade (CladeA2) of the winter rainfall region

To examine whether leaf shape translates into functionalvariation carbon gain and water use efficiency were measuredin pairs of species contrasted for leaf shape and replicated acrossthree subclades of Clade A and one subclade of the third majorclade CladeB (Nicotra et al 2008) In each case the species withthemore dissected leaves showed higher rates of carbon gain butalso higher rates of water loss in all treatment conditions Thatpaper argued that the differences in carbon gain were not directresults of leaf shape per se but rather that the same conditionsthat result in selection for more dissected leaves also favourthe evolution of high photosynthetic rates high leaf nitrogencontent and opportunistic use of water when available Evenin this study however there was no correlation between leafdissection and the climate present in native ranges suggestingeither that climate is selective at a much smaller scale (ie themicrohabitat) or that for any given set of climate variablesmultiple leaf shapes can be adaptive in association withvariation in other leaf traits

Heteroblasty and heterophylly variationwithin a single genetic individual

We have so far focussed primarily on cross species variation inleaf shape As discussed for the Proteaceae genera above (lsquoTheProteaceaersquo section) however a single genome can produce aremarkable range of leaf shapes during the lifetime of a plantThere are two broad categories of leaf shape variation that can beexpressed by a single individual This variation can arise either inresponse to different external environments or to ontogeneticsignals (some of which may be influenced by environment aswell)

Environmentally induced heterophylly

Environmentally induced heterophylly results when leafshape responds to environmental cues The classic case is thedramatically different emergent and submergent leaves in aquaticplants Although many aquatic plants are also heteroblastic (seebelow Sculthorpe 1967) environmentally induced heterophyllyis not specific to position or plant age and is reversible dependinglargely on the environmental signals (mediated by hormonalsignals) perceived by cells of the developing leaf primordiumThe developmental age of the leaf at the time of signal perceptionis critical Very young primordia (L1ndashL3) generally respondcompletely to the novel environment whereas primordiathat were slightly older at the time of environmental changeshow intermediate features as mature leaves Characteristics ofspecific cells are related to the position of that cell within thedifferentiation gradient characteristic of that species at the timeof environmental change (eg Goliber and Feldman 1990 Bruniet al 1996 Kuwabara and Nagata 2006)

The evolution of leaf shape Functional Plant Biology 545

Among terrestrial plants the lsquosunndashshadersquo leaf response isfrequently cited as the classic example of environmentallyinduced heterophylly Historically differences in leaf formwithin a canopy have been assumed to be responses to alteredlight environments (eg larger and less-lobed leaves are producedin shade)However these different leaf shapes also confer distincthydraulic advantages (see lsquoTemperature and waterrsquo section)Whether in response to light or to hydraulic limitation lsquosunshadersquo heterophylly is accompanied by similar and well-knownanatomical differences across a wide range of species (ie sunleaves are generally smaller and thicker with smaller cells thanshade leaves Taiz and Zeiger 2002) However when theseanatomical differences become determined developmentally(Dengler 1980 Yano and Terashima 2004) has been lesscommonly studied The presence of intermediate anatomicalcharacteristics in leaves transferred among light environmentsshortly after expansion ceased suggests that leaves of somespecies have the capacity to respond anatomically to alteredlight levels until very late in development (Oguchi et al2005) In contrast recent studies suggest that systemic lightand humidity signals sent from older to younger leavesallow younger leaves to develop phenotypes suited for theiranticipated environments (Lake et al 2001 Yano andTerashima 2001 Thomas et al 2004) Exactly when final leafshape is determined in lsquosunshadersquo leaves remains an openquestion for most species

Heteroblasty describes ontogenetically dependentchanges in leaf features most notably shape

In 1900 Goebel originally distinguished differences in lsquojuvenileformrsquo from lsquoadult formrsquo on the basis of dramatic differences inleaf shape and associated traits (Balfour 1969) As an example he

described the shift from compound leaves to phyllodes in Acaciaas heteroblastic and reserved the term homoblastic for the lessdramatic variation exhibited by Casurina It is now generallyaccepted that heteroblasty refers to the full suite of features thatchange along the shoot during development (Jones 1999)

Heteroblasty results from the interaction of developmentalprograms expressed at both the leaf and the shoot- or whole-plant level (Kerstetter and Poethig 1998 Tsukaya et al 2000Usami et al 2009 Wu et al 2009 Chuck and OrsquoConnor 2010)Heteroblasty is generally not reversible unless reiterative growthoccurs (ie following damage) Morphological differencescharacteristic of heteroblasty such as differences in lobing areusually expressed relatively early in development shortly afterleaf initiation (Jones 1995)

The sequence of heteroblastic leaf shapes produced alongthe shoot can be accelerated or delayed in response to differentgrowth environments or architectural manipulations (Ashby1948 Forster and Bonser 2009 Jaya et al 2010)Furthermore both heteroblasty and plasticity in heteroblastyhave been shown to be under strong genetic control (Anderson1989 Climent et al 2006) Forster et al (2011) suggested thatboth heteroblasty and plasticity in heteroblasty are likely to occurin response to lsquowhich leaf type confers the greatest benefit fora given light levelrsquo Several studies over the last decade havepointed towards functional consequences of heteroblastygenerally with reference to single environmental variables (egHansen 1986 Gamage and Jesson 2007 Kubien et al 2007)It is likely that heteroblasty influences multiple leaf functionsbecause suites of leaf traits change with heteroblasty (Jones2001) Indeed this conclusion was supported in a recent studythat examined differences inmore than a hundred leaf traits in koa(Acacia koa AGray Pasquet-Kok et al 2010) Heteroblasty hasalso been proposed as an adaptation to herbivory (Karban and

Table 1 Promising directions for cross-disciplinary investigation of leaf shape

No Direction

1 A multi-species lsquoevo-devorsquo approach similar to that by Illing et al (2009) is likely to be informative with regard to many enigmatic features regardingthe evolution of diverse leaf forms and more generally provides an exciting system for examining the convergence and divergence in the evolutionof function

2 A phylogenetic analysis of leaf shape examining correlated evolution with other traits such as venation size leaf surface characteristics and branch andarchitectural measures would enable a thorough investigation of the situations in which leaf shape diversification arises

3 Further research into both the genetic and developmental processes underlying lobes and teeth and the ecological correlates of each would provide anexcellent opportunity to link across wide environmental scales

4 Cross species studies ideally in the field of the relative importance of leaf shape among other factors in determining leaf temperature and water status arenecessary to test the conclusion put forward here that the functional significance of leaf shape lies more often in water relations than directly in leafthermoregulation

5 Studies of the scale of environmental change that selects for ontogenetic response in leaf shape in heteroblastic species fromdiverse environments will shedlight on links between genes development and ecology

6 Detailed quantifications of the extent towhich leaf shape differswithin canopies particularly of dioecious specieswhere selection on reproductive functioncan lead to dimorphic conditions will help link leaf shape to whole plant processes

7 Studies at the scale of transects across the leaf surface that simultaneously consider hydraulic resistance temperature and light will reveal the extent oflocalised hydraulic and thermal stresses within leaf surfaces that might act as selective forces on leaf shape per se Imaging approaches incorporatingmeasures of 3-dimensional shape temperature and fluorescence provide exciting opportunities for examining spatial heterogeneity in temperature andphysiological processes

8 Since patterns of leaf venation affect both hydraulic supply and capacity for photosynthate transport studies that simultaneously examine both willreveal the extent to which interactions among these fundamental processes either co-determine leaf shape or establish trade-offs ultimately reflectedin leaf shape

9 Finally studies examining the fitness values of different leaf shapes or plasticity in leaf shapes under a range of environmental conditions are needed tobetter identify which aspects of leaf shape are most likely to be adaptive

546 Functional Plant Biology A B Nicotra et al

Thaler 1999 Brennan et al 2001 Fadzly et al 2009) and it hasbeen associated with differences in frost resistance (Darrow et al2001) It is worth noting that studies of the functional significanceof heteroblasty have focussed on relatively few plant lineagesGiven that heteroblasty occurs widely among the angiospermsas well as in some ferns and gymosperms (eg Wagner 1952Mueller 1982 Pryer and Hearn 2009) the next questions mayfocus on the cost of heteroblasty and its fitness consequenceswhat scale of environmental change selects for ontogeneticresponse in leaf shape

Synthesis and next steps

Whether considered at community lineage or single genomescales leaf shape is a trait that is highly labile and responsive to arange of biotic and abiotic factors Although some leaf traitscan be explained in terms of particularly strong trade-offsbetween two or more factors that result in clear scalingrelationships (eg leaf mass-to-area traits or leaf mass-to-stemmass) leaf shape in contrast does not emerge from the aboveconsideration as a factor capable of explaining ecologicaldifferentiation between species Rather leaf shape emerges asa trait for which there are many quite varied functional trade-offs

As stated above these trade-offs for successful functionalityof leaves may generate ecological opportunities leaf shape isperhaps best viewed not as a single major axis but rather asan option that fine tunes the leaf to its conditions over bothshort and evolutionary time spans Numerous elements of thisreview support this conclusion Over evolutionary history theextant range of variation in leaf shape has been explored bymanylineages although only the angiosperms have explored thisrange with the added element of massive increases in veindensities Likewise our growing understanding of geneticcontrols on leaf shape reveal that it is a complex traitdetermined by many different genes at several differentdevelopmental stages Although some genes have conservedaffects on leaf morphology including shape the exact natureof their affects can vary widely Likewise the function to whichagivenelement of leaf shape lsquois putrsquo canvary For example closerinvestigation of the internal architecture and developmentalpathways associated with different leaf shapes suggests thatleaf teeth and leaf lobes often have distinctly differentfunctional foundations and may play different functions indifferent environments

Despite amyriadof functional roles the one recurrent theme inour review is that leaf shape is affected by and strongly influencesleaf water relations From a functional perspective the role of leafshape in leaf thermoregulation has perhaps received the mostattention in previous considerations of the functional significanceof leaf shape The relationship between leaf size and temperaturebeing physically determined by basic energy balance principlesthis was a logical starting point of enquiry But as we haveshown the connection between leaf shape (and size) and leaftemperatures under field conditions is not well supportedempirically Rather leaf shape variation as distinct from sizemore often will reflect water supply trade-offs High veindensities support high photosynthetic rates (Brodribb et al2010) and dissected leaves maintain a greater proportion ofleaf tissue closer to main veins Leaf temperature plays a role

in this story as higher leaf temperatures lead to greaterevaporative water loss but many factors other than leaf shapeinfluence leaf temperature (eg leaf thickness orientation andreflectivity and branch and plant architecture) Finally it is alsoworth noting that although there appears to be an obviouslink between shape and water supply there is a concomitantinteraction between water supply and phloem function thatremains relatively unexplored

Thus our review suggests several angles for further research(Table 1) In particular we advocate those approaches thatenable investigation across the scales examined here (genedevelopment function and evolutionary ecology) and thatfocus in particular on the links among leaf shape developmentand water relations

Acknowledgements

The authors thank the Botanical Society of America for their support of thesymposiumfromwhich this reviewaroseWealso thankLawrenSackand twoanonymous reviewers for useful comments on an earlier version of thismanuscript

References

Ackerly DD Knight CA Weiss SB Barton K Starmer KP (2002) Leafsize specific leaf area and microhabitat distribution of chaparral woodyplants contrasting patterns in species level and community level analysesOecologia 130 449ndash457 doi101007s004420100805

Anderson S (1989) Variation in heteroblastic succession among populationsof Crepis tectorum Nordic Journal of Botany 8 565ndash573 doi101111j1756-10511989tb01730x

Armbruster WS Muchhala N (2009) Associations between floralspecialization and species diversity cause effect or correlationEvolutionary Ecology 23 159ndash179 doi101007s10682-008-9259-z

Armbruster WS Di Stilio VS Tuxill JD Flores TC Runk JLV (1999)Covariance and decoupling of floral and vegetative traits in nineneotropical plants a re-evaluation of Bergrsquos correlation-pleiadesconcept American Journal of Botany 86 39ndash55 doi1023072656953

Ashby E (1948) Studies in the morphogenesis of leaves I An essay on leafshape New Phytologist 47 153ndash176 doi101111j1469-81371948tb05098x

Aso K Kato M Banks JA Hasebe M (1999) Characterization ofhomeodomain-leucine zipper genes in the fern Ceratopteris richardiiand the evolution of the homeodomain-leucine zipper gene family invascular plants Molecular Biology and Evolution 16 544ndash552

Bailey IW Sinnott EW (1915) A botanical index of Cretaceous and Tertiaryclimates Science 41 831ndash834 doi101126science411066831

Bailey IW Sinnott EW (1916) The climatic distribution of certain types ofangiosperm leaves American Journal of Botany 3 24ndash39 doi1023072435109

Baker-Brosh KF Peet RK (1997) The ecological significance of lobed andtoothed leaves in temperate forest trees Ecology 78 1250ndash1255

Bakker FT Culham A Marais EM Gibby M (2005) Nested radiation inCape Pelargonium In lsquoPlant species-level systematics new perspectiveson pattern and process Vol 143rsquo (Eds FT Bakker LW Chartrou BGravendeel PB Pielser) pp 75ndash100 (ARG Ganter Verlag KG RuggellLichtstein)

Balfour IB (Transl) (1969) lsquoOrganography of plants I General organographyrsquoby K Goebel 1900 (Hafner Publishing Co Inc New York)

Ball MC Cowan IR Farquhar GD (1988) Maintenance of leaf temperatureand the optimization of carbon gain in relation to water loss in a tropicalmangrove forest Australian Journal of Plant Physiology 15 263ndash276doi101071PP9880263

The evolution of leaf shape Functional Plant Biology 547

Accumulating knowledge on molecular genetic mechanismsof leaf morphogenesis is leading to an increasingly evo-devoapproach to studies of leaf diversity (Tsukaya 2010)this progress is aided by establishing new model species(eg Cardamine hirsuta and the rush Canales et al 2010Yamaguchi and Tsukaya 2010) Over the next decade weanticipate that genomic approaches on multiple speciesanalysed in phylogenetic frameworks (eg Illing et al 2009)will resolvemuchof theenigma regarding the evolutionofdiverseleaf forms

Genetic constraints on flower petal and leaf form

One additional developmental consideration is that as famouslydescribed by von Goethe in 1790 flower petals are derivatives ofthe apical meristem and serial homologues to leaves (von Goethe2009) Thus one might hypothesise that although the evolutionof leaves substantially predates that of flowers selection on theform of either of these organs would also result in changes inthe other such that selection on leaf shape and size subsequent tothe evolution of floral organs may be constrained by pleiotropiccontrol of floral form Recent genetic data identifies several genesthat have roles in both leaf and flower form (Dinneny et al 2004Schmid et al 2005 Street et al 2008) For example geneticchanges in both leaf length to width ratio (eg rot3) and leaf size(eg an3) influence not only leaf proportions but also that of thefloral organs in Arabidopsis (Kim et al 1999 Horiguchi et al2005) The garden pea (Pisum sativum L) is a good case in pointas discussed above compound leaf formation depends on LFYUNI but the same gene is also required for flowering Thus ifcurrent selection favoured simple leaves in garden pea changes inleaf shape could be constrained by pleiotropic control offlowering (Hofer et al 1997) Likewise Shalit et al (2009)recently reported that the action of florigen ndash a plant hormonefor flowering ndash influences complexity of leaves by changingnumbers of leaflets in tomato Such relationships suggest thatsome floral characters as well as other metabolic pathways maynowbeclosely linkedwith regulationof leaf shapeand sizeA fewecological studies have also assessed correlations between petaland leaf form (Berg 1960 Armbruster et al 1999) This questionis of particular interest in lineages where animal pollination isassociated with increased diversification rates either as a result ofdirect selection or as a secondary reinforcing factor (van der Nietet al 2006 Armbruster and Muchhala 2009 Kay and Sargent2009)

In situations where leaf and petal form are under similargenetic control and both under selection one of two thingscould happen If leaf shape does not have substantialfunctional impact then one might expect leaf form to changewith selection on petal form However given the likelyimportance of leaf form to photosynthetic function it is morelikely that a breakdown in genetic correlations between leafand flower form (eg by alterations of downstream regulatorypathways) would evolve Several factors argue against thehypothesis that leaf shape is frequently a product of selectionon flower form First diverse leaf shapes were in evidencelong before the origin of flowers in the angiosperms Secondthe genetic mechanisms underlying leaf shape revealconsiderable evolvability with similar genes being co-opted

for different functions in different species Finally given thepotential importance of both petal and leaf shape we suggestthat genetic constraints on the independent evolution of bothstructures are likely to have broken down in those lineages withhighly diversified leaf shapes over the many millions of yearssince petals first evolved

Ecological correlates and the functional significanceof leaf shape

Evolutionary history plays a part in determining leafmorphology but our consideration of leaf shape evolution andits developmental genetic basis demonstrates that diverselineages have repeatedly gained and lost many leaf featuresThus developmental and genetic constraints on the evolutionof leaf form and shape in particular have been overcomemany times or have led to new evolutionary opportunities Itis of particular interest then to ask what are the likely ecologicaldrivers andor functional significance of leaf shape

Temperature and water

Conventionalwisdomwould say that thermoregulation is perhapsthe main driver of leaf shape evolution The maintenance ofleaf temperatures within certain limits is critical for a plantrsquosgrowth and survival and although photosynthetic tissue of somespecies canwithstand temperatures below6C (Ball et al 2006)and above 60C (Clum1926Nobel 1988) irreversible damage toleaves can occur at temperatures well within this range (Levitt1980 Jones 1992 Ball et al 2004 Groom et al 2004 Sharkey2005) Leaf size and shape potentially have large effects onleaf temperature because the two-dimensional proportions of aflattened leaf determine the rate of heat transfer between acrossthe leafndashair interface by influencing the thickness of its boundarylayer

All else being equal the thickness of a boundary layerincreases with length from the windward edge so that heatconvection from small leaves is more rapid than from largeleaves (Raschke 1960 Gates 1968 Vogel 1970 Parkhurst andLoucks 1972Grace et al 1980Geller andSmith 1982Monteithand Unsworth 1990 Schuepp 1993) Likewise because leaflobing reduces the distance across the lamina the rate of heattransfer is predicted to be greater in a lobed leaf than an unlobedleaf with equivalent area (Parkhurst et al 1968 Vogel 1968Lewis 1972 Givnish 1978 Gurevitch and Schuepp 1990a)

The rate of heat transfer from lobed metal plates or lobedleaves coated in metal is greater than from shallow-lobed or un-lobed ones (Parkhurst et al 1968 Thom 1968 Vogel 1970Gottschlich and Smith 1982 Gurevitch and Schuepp 1990bRoth-Nebelsick 2001) Artificial or coated leaves have beenused because they enable boundary layer resistance to beinvestigated in the absence of transpiration However thisapproach is biologically flawed because it does not enable anassessment of the relative importance of a range of traits (eg leafwater content absorbance) in determining the actual operatingtemperatures of leaves in the field

Actual leaf temperature measurements show that large leavescan operate at below-ambient temperatures even on hot days asa result of high rates of latent heat loss through transpiration(Drake et al 1970 Gates 1980 Hegazy and El Amry 1998)

The evolution of leaf shape Functional Plant Biology 541

Thus conditions of evaporative demand and the availability ofsoil water for transpiration are key to understanding leafshapendashtemperature relationships under biologically realisticconditions Because of their much greater transpirationalcapacities (see above) flowering plants have far greaterleeway than other plants regarding the size and shape of theirleaves (see lsquoA brief history of the angiosperm leaf and shapediversity thereinrsquo section Boyce et al 2009)

In addition to reducing heat loads the morphology of deeplylobed leaves may also reflect direct selection for increasedhydraulic efficiency Water moving through a plant encountersnumerous resistances on its path from soil to roots to stemsthrough leaves and out to the external atmosphere Within theliquid phase of water flow through a plant leaf hydraulicresistance Rleaf accounts for more than 30 of total plantresistance (Sack and Holbrook 2006) The high resistance inleaves is due first to a system of veins of decreasing size theresistance of which is inversely proportional to the fourth powerof the radius of the component conduits in these veins Thereforeas a proportion of the considerable resistance in the leaf(Zwieniecki et al 2002) the minor veins provide the greatestresistance to water flow (Sack et al 2004)Water also encountershigh resistance as it exits veins laterally and travels to othercells mainly via apoplastic pathways through the mesophyllAccordingly on the order of half of total Rleaf can occur outsidethe major veins (Sack and Holbrook 2006) As resistanceincreases along the pathway towards sites of evaporationlocalised water potential (Y) becomes progressively morenegative (Brodribb et al 2010) The result of how thisphenomenon might affect a leaf has long been observedregions of leaves lying furthest from the main supply channelsare the first to lsquowitherrsquo when exposed to strong winds (Fig 4Yapp 1912) Leaf lobing represents an effective removal of thispotentially stress-prone tissue and has been suggested as anadaptation to dry conditions (Thoday 1931 Givnish 1979) Ifdeeply lobed leaves have a lower ratio of mesophyll tissue tolarge highly conductive veins they should then have reducedresistance relative to less- or un-lobed leaves (Sack and Tyree2005)

Although leaf lobing might confer adaptive benefits withrespect to maintaining stable hydraulic supply it may also bethat the extent of lobing or margin dissection is itself determinedby hydraulic limitation (Sack et al 2003 Zwieniecki et al2004b Boyce 2009) For species with lsquosunrsquo and lsquoshadersquoleaves the morphology of small deeply lobed sun leaves atthe outer canopy is proposed to result from water pressuredrop across the leaf lamina caused by reduced water deliveryto expanding cells during growth (Fig 5 Zwieniecki et al 2004bBoyce 2009 Leigh et al 2011)Alternatively recent studies havesuggested that many of the characteristics of lsquosunrsquo and lsquoshadersquoleaves are the result of branch autonomy within the plant canopy(Niklas and Cobb 2010)

Thus although the effects of leaf shape on leaf thermalregulation hold when tested on model leaves predictions areless straightforward when hydraulic function is accounted forAdditionally consideration of the relative contribution of a rangeof leaf and branch properties to leaf temperature is critical It islikely that leaf shape is only one amongmany factors influencingleaf thermal regulation other factors could include water

content leaf thickness spectral reflectance orientation andplant architecture Further given the observation of high leafshape diversity withinmany plant communities from hot and aridenvironments it seems likely that regulating leaf temperature isnot the singlemost important evolutionary influence on leaf shapediversity The association between leaf shape and hydraulicproperties which is affected by leaf temperature in contrast islikely to be of greater importance

Leaf structure and functional trade-offs

Many foliar leaf traits including shape reflect functional trade-offs that have been resolved in various ways by differentspecies depending on their ecological settings (eg hydro- vsmesophytes) physiological attributes (eg C3 vsC4metabolism)developmental capabilities (eg heterobaric vs non-heterobaricleaves) or evolutionary histories (eg micro- vs megaphylls)Life history and optimisation theory show that the number ofphenotypic solutions that allow for different equally successfultrait combinations increases as the number of trade-offsincreases ndash a conclusion that applies to traits within the leaf(eg for shape) as well as to leafndashbranch relationships (Niklas1988)

It is evident that leaves perform several functionssimultaneously that is they intercept light transport liquidsand solutes exchange gases with the atmosphere dissipateheat cope with externally applied mechanical forces and

(a)

(e)(f )

( i )

(g)

(h)

(b)

(c)

(d )

Fig 4 Localised lsquowitheringrsquo of leaves exposed to strong wind Reprintedfrom Yapp (1912) with permission from Oxford University Press

542 Functional Plant Biology A B Nicotra et al

defend themselves from herbivores and pathogens Importantlymany of these functions have conflicting physiologicalanatomical or morphological requirements For exampleorientations of lamina that maximise light interception alsominimise the ability to dissipate heat Further all of thesefunctions are intrinsically size dependent For example thequantity of mechanical tissues required to support laminaedepends on the mass of laminae whereas the ability tointercept light depends on laminae area

When seen in this light the trade-offs required for thesuccessful functionality of foliar leaves generate rather thanconstrain ecological and evolutionary opportunities This canbe illustrated mathematically and empirically in the context ofphyllotaxy and its consequences on light interception Computersimulations of mathematically generated shoots to assess theinfluence of leaf shape size and phyllotactic patterns on theability to intercept direct solar radiation show that differences inphyllotaxy significantly influence light interception (particularlyfor shoots with a rosette growth habit) They also show thatcomparatively small differences in leaf shape can compensatefor the negative effects of leaf overlap resulting from virtuallyany phyllotactic pattern For example lobed leaves or pinnifid

compound leaves facilitate light penetration through shootsbearing densely packed leaves Thus phyllotaxy can beviewed as a developmental limiting factor that can drivecompensatory changes in morphological features such asshape that are not directly controlled by patterns of leafinitiation In this way the lsquoconstraintsrsquo imposed by phyllotaxycan foster evolutionary modifications at the level of the entireshoot (Niklas 1988)

One particularly important trade-off operating at the level ofthe architecture of a shoot is that between leaf size and numberFor any given leaf biomass allocation an individual shoot caneither have a few large leaves or many smaller ones Acrossdifferent herbaceous species and juvenile specimens of treespecies drawn from diverse lineages and ecological settingsthe relationship between total leaf biomass per plant (ML) andtotal stemmass per plant (MS) fallswithin a comparatively narrowcorridor of variation that obeys a positive isometric scalingrelationship (ie MLMS

~10 Niklas 2004) Accordinglyacross these species any stem mass investment yields a moreor less proportional increase in leaf mass Therefore it followsmathematically that across these species the relationshipamong the average mass of a typical leaf mL total leaf numberNL and total stem mass complies with the proportionality mL

NLMS~10 When individual shoots are sampled across

diverse species the relationship between mL and NL per shootis negative and isometric (ie mLNL

ndash1 Kleiman and Aarssen2007) a relationship that appears to be insensitive to howspecies are grouped according to other functional traits oraccording to their habitat preference (Yang et al 2008) Thatthis phenomenology reflects a biomass allocation trade-off isclear What remains uncertain is whether it is a simple trade-offbetween leaf mass investment and leaf size or a more complexset of relationships imposed by twig mass investments andhydraulics or biomechanics

Another trade-off that appears to hold across many but not allspecies is the relationship between leaf surface areaA and averageleaf mass mL For more than a few species-groupings thisrelationship is allometric with a slope that is less than onewhich indicates that increases in leaf mass investment do notresult in proportional increases in leaf area (ieAmL

lt10Niklaset al 2007) This trend of lsquodiminishing returnsrsquo has also beenreported when total leaf area per plant is plotted against total leafmass for plants differing in size (Niklas and Cobb 2008 2010)Several factorsmay contribute to these trends eg biomechanicalanalyses show that disproportionately larger investments must bemade to mechanically support photosynthetic tissues as leavesincrease in their surface area (Niklas et al 2009) In both the leafmass to stem mass and leaf mass to area relationships describedabove leaf shapevariation canprovide anextra degreeof freedomfor selection to act on at both the leaf and branch level

Case studies of leaf shape variation at different scales

Above we have discussed evolutionary and proximal drivers ofleaf shape as well as ecological correlates and functionalsignificance thereof Here we present case studies at threescales where we can examine the ecological and evolutionaryimplications of leaf shape These are the cross community scalecaseof leaf teeth as a subset of shapemodificationswithin lineage

200 bottomtop

150

100

50

0

0

08

07

06

05

04

03

5 10

Days after bud break

Leaf

siz

e (c

m2 )

Leaf

lobi

ng in

dex

(cm

2 cm

ndash2)

Days after bud break

15 20 25

0 5 10 15 20 25

Fig 5 Variation in (a) leaf size and (b) shape complexity inQuercus rubraleaves growing at the top and bottom of the tree crown Reprinted fromZwieniecki et al (2004b) with permission from John Wiley amp Sons

The evolution of leaf shape Functional Plant Biology 543

considerations of the genus Pelargonium and the familyProteaceae and finally the case of leaf shape variation within asingle genetic individual heteroblasty and heterophylly

Leaf teeth leaf shape variation at the community scale

Leaf teeth are conspicuous features inmanyplants and are notableas a leaf-shape trait that varies in predictable ways across plantcommunities Nearly 100 years ago Bailey and Sinnott (19151916) observed that plant communities from regions with lowmean annual temperature (MAT) have a higher site meanproportion of non-monocotyledonous species with toothedleaves Paleobotanists quickly seized upon this relationship asan index for inferring paleotemperature Little et al (2010)recently documented the rich interest in this topic compiling351 literature references related to leaf physiognomy andtemperature of which the primary focus is leaf teeth The linkbetween teeth and temperature is observed on all vegetatedcontinents (Greenwood 2005a Peppe Royer et al 2011) andis also seen in related metrics such as tooth size and tooth number(Huff et al 2003 Royer et al 2005 Peppe et al 2011) Furtherthese tooth variables can display phenotypic plasticity totemperature change (Fig 6 Royer et al 2009b)

What is knownabout the functional bases of these leafndashclimaterelationships Leaf teeth commonly expand and mature fasterthan the bulk leaf (eg Billings 1905 Feild et al 2005) and youngteeth are disproportionately vigorous in their gas exchange(Baker-Brosh and Peet 1997 Royer and Wilf 2006) Fromthese observations Royer and Wilf (2006) hypothesised thatone role of teeth is to increase sap flow thereby deliveringnutrients and other solutes to young emerging leaves Inhabitats with low temperatures teeth could play an importantrole in maximising the potential for carbon gain and growth earlyin the season In warmer climates with longer growing seasonsthe carbon benefit conferred by teeth would be outweighed bytheir unavoidable water cost Alternatively Feild et al (2005)hypothesised that teeth help remove freezendashthaw embolismsthrough positive root pressure and guttation Importantly thetransport of fluids is common to both hypotheses From this itmay be surmised that water-limiting conditions select againstteeth (Bailey andSinnott 1916) Indeed this pattern is clearly seenacross local water availability gradients (Burnham et al 2001

Kowalski and Dilcher 2003 Greenwood 2005b Royer et al2009a Peppe et al 2011) but not globally against mean annualprecipitation (eg Peppe et al 2011)

If the hypothesis by Royer andWilf (2006) is correct toothedspecies should have a higher photosynthetic rate at least at thebeginning of the growing season As such teethmay be related tothe leaf economics spectrum (Wright et al 2004) Some studiesfind support for this idea leaves with a short leaf lifespan(deciduous) (Peppe et al 2011) and low leaf mass per area(Royer et al 2005) two leaf economic variables are morelikely to be toothed It also follows that teeth should be morefunctionally related to growing season climatic variables likeseasonality growing season length and growing degree days notmean annual temperature however few studies find support forthis idea (Royer et al 2005 Peppe et al 2011) Ultimately itwould be valuable to have a better understanding of the geneticprocesses underlying tooth development but presently little isknown outside model species (see lsquoGenetic regulation forpatterning of leaf margin serration and leaflet formationrsquosection) Further phylogenetic history has typically beenassumed to have only a minor role in these leaf-climatepatterns but rigorous tests demonstrate that this assumption isfalse (Little et al 2010) This outcome is of particular concern forpaleobotanists because systematic placement of fossils can bedifficult especially for the Cretaceous and Paleogene Tools toaccommodate for this phylogenetic dependency are needed

Leaf shape variation within lineages

Proteaceae

Proteaceae is an ancient angiosperm family originating inGondwana over 100million year ago and possessing 1700 extantspecies (Weston 2007) The familyrsquos greatest representation is inAustralia followed by SouthAfrica with smaller numbers acrossother Gondwanan segments (Cowling and Lamont 1998 Hootand Douglas 1998 Weston 2007) Within the Proteaceae thereexists an enormous range of leaf sizes and shapes ranging froma few mm2 to 500 cm2 and including entire toothed broadlylobed deeply dissected needle-like simple or compound leavesIntriguingly most of this shape diversity is represented inAustralia The majority of taxa outside Australia possessentire leaves the exception being four South African genera

(a) (b) (c) (d )

Fig 6 Representative leaves of Acer rubrum showing genetic variation and plasticity in leaf teeth Leaf derived from coolclimate (Ontario) seed stock grown in (a) Rhode Island (cool climate) and (b) Florida (warm climate) leaf derived fromwarm climate (Florida) seed stock grown in (c) Rhode Island and (d) Florida Bar (for all leaves) = 1 cm Reprinted from Royeret al (2009) under Creative Commons Attribution Licence

544 Functional Plant Biology A B Nicotra et al

that possess dissected (Serruria and Paranomis) or toothed(Mimites and Leucospermum) leafed species all of whichoccur in a single clade (Sauquet et al 2009)

The reason for the extensive radiation in leaf shape diversityin Australian Proteaceae is unclear Among South AfricanProteaceae leaf size decreases with mean annual temperatureand narrow leaves have been shown to overheat less than theirwider counterparts (Yates et al 2010) Given that leaf dissectionand small size are functionally equivalent with respect to heatloss (see lsquoTemperature and waterrsquo section) it is tempting tosuggest that leaf shape variation in Australia represents anadaptive response to increasing hot dry conditions Yet thereappears to be no consistent distributional pattern of Proteaceaeleaf dimensions inAustraliaManymediumor large un-dissectedleaves exist in the hottest habitats Grevilleas with small simpleleaves can be found across most regions including high rainfallareas and some of the species found in the shaded rainforestunderstorey have among the most deeply dissected leaves of allLeaf shape also varies within a single individual For examplefern-leaf stenocarpus Stenocarpus davallioides Foreman ampBHyland has tripinnatisect juvenile leaves but adult leavescan be pinnate or bipinnate (Foreman and Hyland 1995)Other rainforest species such as Atherton Oak (Athertoniadiversifolia (CTWhit) LAS Johnson amp BGBriggs) havevery large lobed intermediate leaves but adult and seedlingleaves are small and entire (Weston 1995) Alternatively leaflobing within a plant can be random having no obvious spatialor developmental association or homogenous across a speciesExamining a phylogeny of extant taxa one finds lobed ordissected leaves alongside entire-leafed species ndash across mostlineages in the group (Weston 2007) Therefore althoughsuggestions of microclimatic adaptation may be offered atthe species level it seems unlikely that an environmentalassociation with leaf shape could be generalised acrossAustralian Proteaceae Rather leaf shape variation in theProteaceae is likely to reflect the range of ecological solutionsto leaf structurefunction trade-offs discussed above

Pelargonium

Similar to the Proteaceae the genus Pelargonium isdistributed predominantly in the southern hemisphere withnearly 80 of its almost 300 species found in South AfricaAlthough the age of the genus is estimated to be around30million years (Bakker et al 2005) the highest rate ofspecies accumulation has occurred in the last 10million years(Martinez-Cabrera 2010) The combination of climatic andedaphic heterogeneity over short distances and limited geneflow (Latimer et al 2005 Linder 2005) is frequently offeredas an explanation for the extraordinary plant species diversity thatcharacterises southern Africa Pelargonium the third largestgenus in the region is characterised by remarkable variation ingrowth form that ranges from annuals and geophytes to shrubsVariation in leaf shape is no less dramatic ranging from entireto highly dissected (Fig 1) often within growth forms andtaxonomic sections Parsing of leaf form categorically revealedthat the extent of dissection of the blade is highly evolvablewithinanoverall ovate leaf shapeoutline that is evolutionarily conserved(Jones et al 2009)

In addition to high evolutionary lability of leaf dissectionthere is variation among major clades in the evolutionary patternof major veins One major clade (Clade C) is dominated bypalmate venation whereas another (Clade A) is dominated bypinnate venation although both types are found in both cladesWe note that pinnate venation is significantly associated withincreased dissection and reductions in functional leaf sizecharacteristic of the highly diverse lsquoxerophyticrsquo clade (CladeA2) of the winter rainfall region

To examine whether leaf shape translates into functionalvariation carbon gain and water use efficiency were measuredin pairs of species contrasted for leaf shape and replicated acrossthree subclades of Clade A and one subclade of the third majorclade CladeB (Nicotra et al 2008) In each case the species withthemore dissected leaves showed higher rates of carbon gain butalso higher rates of water loss in all treatment conditions Thatpaper argued that the differences in carbon gain were not directresults of leaf shape per se but rather that the same conditionsthat result in selection for more dissected leaves also favourthe evolution of high photosynthetic rates high leaf nitrogencontent and opportunistic use of water when available Evenin this study however there was no correlation between leafdissection and the climate present in native ranges suggestingeither that climate is selective at a much smaller scale (ie themicrohabitat) or that for any given set of climate variablesmultiple leaf shapes can be adaptive in association withvariation in other leaf traits

Heteroblasty and heterophylly variationwithin a single genetic individual

We have so far focussed primarily on cross species variation inleaf shape As discussed for the Proteaceae genera above (lsquoTheProteaceaersquo section) however a single genome can produce aremarkable range of leaf shapes during the lifetime of a plantThere are two broad categories of leaf shape variation that can beexpressed by a single individual This variation can arise either inresponse to different external environments or to ontogeneticsignals (some of which may be influenced by environment aswell)

Environmentally induced heterophylly

Environmentally induced heterophylly results when leafshape responds to environmental cues The classic case is thedramatically different emergent and submergent leaves in aquaticplants Although many aquatic plants are also heteroblastic (seebelow Sculthorpe 1967) environmentally induced heterophyllyis not specific to position or plant age and is reversible dependinglargely on the environmental signals (mediated by hormonalsignals) perceived by cells of the developing leaf primordiumThe developmental age of the leaf at the time of signal perceptionis critical Very young primordia (L1ndashL3) generally respondcompletely to the novel environment whereas primordiathat were slightly older at the time of environmental changeshow intermediate features as mature leaves Characteristics ofspecific cells are related to the position of that cell within thedifferentiation gradient characteristic of that species at the timeof environmental change (eg Goliber and Feldman 1990 Bruniet al 1996 Kuwabara and Nagata 2006)

The evolution of leaf shape Functional Plant Biology 545

Among terrestrial plants the lsquosunndashshadersquo leaf response isfrequently cited as the classic example of environmentallyinduced heterophylly Historically differences in leaf formwithin a canopy have been assumed to be responses to alteredlight environments (eg larger and less-lobed leaves are producedin shade)However these different leaf shapes also confer distincthydraulic advantages (see lsquoTemperature and waterrsquo section)Whether in response to light or to hydraulic limitation lsquosunshadersquo heterophylly is accompanied by similar and well-knownanatomical differences across a wide range of species (ie sunleaves are generally smaller and thicker with smaller cells thanshade leaves Taiz and Zeiger 2002) However when theseanatomical differences become determined developmentally(Dengler 1980 Yano and Terashima 2004) has been lesscommonly studied The presence of intermediate anatomicalcharacteristics in leaves transferred among light environmentsshortly after expansion ceased suggests that leaves of somespecies have the capacity to respond anatomically to alteredlight levels until very late in development (Oguchi et al2005) In contrast recent studies suggest that systemic lightand humidity signals sent from older to younger leavesallow younger leaves to develop phenotypes suited for theiranticipated environments (Lake et al 2001 Yano andTerashima 2001 Thomas et al 2004) Exactly when final leafshape is determined in lsquosunshadersquo leaves remains an openquestion for most species

Heteroblasty describes ontogenetically dependentchanges in leaf features most notably shape

In 1900 Goebel originally distinguished differences in lsquojuvenileformrsquo from lsquoadult formrsquo on the basis of dramatic differences inleaf shape and associated traits (Balfour 1969) As an example he

described the shift from compound leaves to phyllodes in Acaciaas heteroblastic and reserved the term homoblastic for the lessdramatic variation exhibited by Casurina It is now generallyaccepted that heteroblasty refers to the full suite of features thatchange along the shoot during development (Jones 1999)

Heteroblasty results from the interaction of developmentalprograms expressed at both the leaf and the shoot- or whole-plant level (Kerstetter and Poethig 1998 Tsukaya et al 2000Usami et al 2009 Wu et al 2009 Chuck and OrsquoConnor 2010)Heteroblasty is generally not reversible unless reiterative growthoccurs (ie following damage) Morphological differencescharacteristic of heteroblasty such as differences in lobing areusually expressed relatively early in development shortly afterleaf initiation (Jones 1995)

The sequence of heteroblastic leaf shapes produced alongthe shoot can be accelerated or delayed in response to differentgrowth environments or architectural manipulations (Ashby1948 Forster and Bonser 2009 Jaya et al 2010)Furthermore both heteroblasty and plasticity in heteroblastyhave been shown to be under strong genetic control (Anderson1989 Climent et al 2006) Forster et al (2011) suggested thatboth heteroblasty and plasticity in heteroblasty are likely to occurin response to lsquowhich leaf type confers the greatest benefit fora given light levelrsquo Several studies over the last decade havepointed towards functional consequences of heteroblastygenerally with reference to single environmental variables (egHansen 1986 Gamage and Jesson 2007 Kubien et al 2007)It is likely that heteroblasty influences multiple leaf functionsbecause suites of leaf traits change with heteroblasty (Jones2001) Indeed this conclusion was supported in a recent studythat examined differences inmore than a hundred leaf traits in koa(Acacia koa AGray Pasquet-Kok et al 2010) Heteroblasty hasalso been proposed as an adaptation to herbivory (Karban and

Table 1 Promising directions for cross-disciplinary investigation of leaf shape

No Direction

1 A multi-species lsquoevo-devorsquo approach similar to that by Illing et al (2009) is likely to be informative with regard to many enigmatic features regardingthe evolution of diverse leaf forms and more generally provides an exciting system for examining the convergence and divergence in the evolutionof function

2 A phylogenetic analysis of leaf shape examining correlated evolution with other traits such as venation size leaf surface characteristics and branch andarchitectural measures would enable a thorough investigation of the situations in which leaf shape diversification arises

3 Further research into both the genetic and developmental processes underlying lobes and teeth and the ecological correlates of each would provide anexcellent opportunity to link across wide environmental scales

4 Cross species studies ideally in the field of the relative importance of leaf shape among other factors in determining leaf temperature and water status arenecessary to test the conclusion put forward here that the functional significance of leaf shape lies more often in water relations than directly in leafthermoregulation

5 Studies of the scale of environmental change that selects for ontogenetic response in leaf shape in heteroblastic species fromdiverse environments will shedlight on links between genes development and ecology

6 Detailed quantifications of the extent towhich leaf shape differswithin canopies particularly of dioecious specieswhere selection on reproductive functioncan lead to dimorphic conditions will help link leaf shape to whole plant processes

7 Studies at the scale of transects across the leaf surface that simultaneously consider hydraulic resistance temperature and light will reveal the extent oflocalised hydraulic and thermal stresses within leaf surfaces that might act as selective forces on leaf shape per se Imaging approaches incorporatingmeasures of 3-dimensional shape temperature and fluorescence provide exciting opportunities for examining spatial heterogeneity in temperature andphysiological processes

8 Since patterns of leaf venation affect both hydraulic supply and capacity for photosynthate transport studies that simultaneously examine both willreveal the extent to which interactions among these fundamental processes either co-determine leaf shape or establish trade-offs ultimately reflectedin leaf shape

9 Finally studies examining the fitness values of different leaf shapes or plasticity in leaf shapes under a range of environmental conditions are needed tobetter identify which aspects of leaf shape are most likely to be adaptive

546 Functional Plant Biology A B Nicotra et al

Thaler 1999 Brennan et al 2001 Fadzly et al 2009) and it hasbeen associated with differences in frost resistance (Darrow et al2001) It is worth noting that studies of the functional significanceof heteroblasty have focussed on relatively few plant lineagesGiven that heteroblasty occurs widely among the angiospermsas well as in some ferns and gymosperms (eg Wagner 1952Mueller 1982 Pryer and Hearn 2009) the next questions mayfocus on the cost of heteroblasty and its fitness consequenceswhat scale of environmental change selects for ontogeneticresponse in leaf shape

Synthesis and next steps

Whether considered at community lineage or single genomescales leaf shape is a trait that is highly labile and responsive to arange of biotic and abiotic factors Although some leaf traitscan be explained in terms of particularly strong trade-offsbetween two or more factors that result in clear scalingrelationships (eg leaf mass-to-area traits or leaf mass-to-stemmass) leaf shape in contrast does not emerge from the aboveconsideration as a factor capable of explaining ecologicaldifferentiation between species Rather leaf shape emerges asa trait for which there are many quite varied functional trade-offs

As stated above these trade-offs for successful functionalityof leaves may generate ecological opportunities leaf shape isperhaps best viewed not as a single major axis but rather asan option that fine tunes the leaf to its conditions over bothshort and evolutionary time spans Numerous elements of thisreview support this conclusion Over evolutionary history theextant range of variation in leaf shape has been explored bymanylineages although only the angiosperms have explored thisrange with the added element of massive increases in veindensities Likewise our growing understanding of geneticcontrols on leaf shape reveal that it is a complex traitdetermined by many different genes at several differentdevelopmental stages Although some genes have conservedaffects on leaf morphology including shape the exact natureof their affects can vary widely Likewise the function to whichagivenelement of leaf shape lsquois putrsquo canvary For example closerinvestigation of the internal architecture and developmentalpathways associated with different leaf shapes suggests thatleaf teeth and leaf lobes often have distinctly differentfunctional foundations and may play different functions indifferent environments

Despite amyriadof functional roles the one recurrent theme inour review is that leaf shape is affected by and strongly influencesleaf water relations From a functional perspective the role of leafshape in leaf thermoregulation has perhaps received the mostattention in previous considerations of the functional significanceof leaf shape The relationship between leaf size and temperaturebeing physically determined by basic energy balance principlesthis was a logical starting point of enquiry But as we haveshown the connection between leaf shape (and size) and leaftemperatures under field conditions is not well supportedempirically Rather leaf shape variation as distinct from sizemore often will reflect water supply trade-offs High veindensities support high photosynthetic rates (Brodribb et al2010) and dissected leaves maintain a greater proportion ofleaf tissue closer to main veins Leaf temperature plays a role

in this story as higher leaf temperatures lead to greaterevaporative water loss but many factors other than leaf shapeinfluence leaf temperature (eg leaf thickness orientation andreflectivity and branch and plant architecture) Finally it is alsoworth noting that although there appears to be an obviouslink between shape and water supply there is a concomitantinteraction between water supply and phloem function thatremains relatively unexplored

Thus our review suggests several angles for further research(Table 1) In particular we advocate those approaches thatenable investigation across the scales examined here (genedevelopment function and evolutionary ecology) and thatfocus in particular on the links among leaf shape developmentand water relations

Acknowledgements

The authors thank the Botanical Society of America for their support of thesymposiumfromwhich this reviewaroseWealso thankLawrenSackand twoanonymous reviewers for useful comments on an earlier version of thismanuscript

References

Ackerly DD Knight CA Weiss SB Barton K Starmer KP (2002) Leafsize specific leaf area and microhabitat distribution of chaparral woodyplants contrasting patterns in species level and community level analysesOecologia 130 449ndash457 doi101007s004420100805

Anderson S (1989) Variation in heteroblastic succession among populationsof Crepis tectorum Nordic Journal of Botany 8 565ndash573 doi101111j1756-10511989tb01730x

Armbruster WS Muchhala N (2009) Associations between floralspecialization and species diversity cause effect or correlationEvolutionary Ecology 23 159ndash179 doi101007s10682-008-9259-z

Armbruster WS Di Stilio VS Tuxill JD Flores TC Runk JLV (1999)Covariance and decoupling of floral and vegetative traits in nineneotropical plants a re-evaluation of Bergrsquos correlation-pleiadesconcept American Journal of Botany 86 39ndash55 doi1023072656953

Ashby E (1948) Studies in the morphogenesis of leaves I An essay on leafshape New Phytologist 47 153ndash176 doi101111j1469-81371948tb05098x

Aso K Kato M Banks JA Hasebe M (1999) Characterization ofhomeodomain-leucine zipper genes in the fern Ceratopteris richardiiand the evolution of the homeodomain-leucine zipper gene family invascular plants Molecular Biology and Evolution 16 544ndash552

Bailey IW Sinnott EW (1915) A botanical index of Cretaceous and Tertiaryclimates Science 41 831ndash834 doi101126science411066831

Bailey IW Sinnott EW (1916) The climatic distribution of certain types ofangiosperm leaves American Journal of Botany 3 24ndash39 doi1023072435109

Baker-Brosh KF Peet RK (1997) The ecological significance of lobed andtoothed leaves in temperate forest trees Ecology 78 1250ndash1255

Bakker FT Culham A Marais EM Gibby M (2005) Nested radiation inCape Pelargonium In lsquoPlant species-level systematics new perspectiveson pattern and process Vol 143rsquo (Eds FT Bakker LW Chartrou BGravendeel PB Pielser) pp 75ndash100 (ARG Ganter Verlag KG RuggellLichtstein)

Balfour IB (Transl) (1969) lsquoOrganography of plants I General organographyrsquoby K Goebel 1900 (Hafner Publishing Co Inc New York)

Ball MC Cowan IR Farquhar GD (1988) Maintenance of leaf temperatureand the optimization of carbon gain in relation to water loss in a tropicalmangrove forest Australian Journal of Plant Physiology 15 263ndash276doi101071PP9880263

The evolution of leaf shape Functional Plant Biology 547

Thus conditions of evaporative demand and the availability ofsoil water for transpiration are key to understanding leafshapendashtemperature relationships under biologically realisticconditions Because of their much greater transpirationalcapacities (see above) flowering plants have far greaterleeway than other plants regarding the size and shape of theirleaves (see lsquoA brief history of the angiosperm leaf and shapediversity thereinrsquo section Boyce et al 2009)

In addition to reducing heat loads the morphology of deeplylobed leaves may also reflect direct selection for increasedhydraulic efficiency Water moving through a plant encountersnumerous resistances on its path from soil to roots to stemsthrough leaves and out to the external atmosphere Within theliquid phase of water flow through a plant leaf hydraulicresistance Rleaf accounts for more than 30 of total plantresistance (Sack and Holbrook 2006) The high resistance inleaves is due first to a system of veins of decreasing size theresistance of which is inversely proportional to the fourth powerof the radius of the component conduits in these veins Thereforeas a proportion of the considerable resistance in the leaf(Zwieniecki et al 2002) the minor veins provide the greatestresistance to water flow (Sack et al 2004)Water also encountershigh resistance as it exits veins laterally and travels to othercells mainly via apoplastic pathways through the mesophyllAccordingly on the order of half of total Rleaf can occur outsidethe major veins (Sack and Holbrook 2006) As resistanceincreases along the pathway towards sites of evaporationlocalised water potential (Y) becomes progressively morenegative (Brodribb et al 2010) The result of how thisphenomenon might affect a leaf has long been observedregions of leaves lying furthest from the main supply channelsare the first to lsquowitherrsquo when exposed to strong winds (Fig 4Yapp 1912) Leaf lobing represents an effective removal of thispotentially stress-prone tissue and has been suggested as anadaptation to dry conditions (Thoday 1931 Givnish 1979) Ifdeeply lobed leaves have a lower ratio of mesophyll tissue tolarge highly conductive veins they should then have reducedresistance relative to less- or un-lobed leaves (Sack and Tyree2005)

Although leaf lobing might confer adaptive benefits withrespect to maintaining stable hydraulic supply it may also bethat the extent of lobing or margin dissection is itself determinedby hydraulic limitation (Sack et al 2003 Zwieniecki et al2004b Boyce 2009) For species with lsquosunrsquo and lsquoshadersquoleaves the morphology of small deeply lobed sun leaves atthe outer canopy is proposed to result from water pressuredrop across the leaf lamina caused by reduced water deliveryto expanding cells during growth (Fig 5 Zwieniecki et al 2004bBoyce 2009 Leigh et al 2011)Alternatively recent studies havesuggested that many of the characteristics of lsquosunrsquo and lsquoshadersquoleaves are the result of branch autonomy within the plant canopy(Niklas and Cobb 2010)

Thus although the effects of leaf shape on leaf thermalregulation hold when tested on model leaves predictions areless straightforward when hydraulic function is accounted forAdditionally consideration of the relative contribution of a rangeof leaf and branch properties to leaf temperature is critical It islikely that leaf shape is only one amongmany factors influencingleaf thermal regulation other factors could include water

content leaf thickness spectral reflectance orientation andplant architecture Further given the observation of high leafshape diversity withinmany plant communities from hot and aridenvironments it seems likely that regulating leaf temperature isnot the singlemost important evolutionary influence on leaf shapediversity The association between leaf shape and hydraulicproperties which is affected by leaf temperature in contrast islikely to be of greater importance

Leaf structure and functional trade-offs

Many foliar leaf traits including shape reflect functional trade-offs that have been resolved in various ways by differentspecies depending on their ecological settings (eg hydro- vsmesophytes) physiological attributes (eg C3 vsC4metabolism)developmental capabilities (eg heterobaric vs non-heterobaricleaves) or evolutionary histories (eg micro- vs megaphylls)Life history and optimisation theory show that the number ofphenotypic solutions that allow for different equally successfultrait combinations increases as the number of trade-offsincreases ndash a conclusion that applies to traits within the leaf(eg for shape) as well as to leafndashbranch relationships (Niklas1988)

It is evident that leaves perform several functionssimultaneously that is they intercept light transport liquidsand solutes exchange gases with the atmosphere dissipateheat cope with externally applied mechanical forces and

(a)

(e)(f )

( i )

(g)

(h)

(b)

(c)

(d )

Fig 4 Localised lsquowitheringrsquo of leaves exposed to strong wind Reprintedfrom Yapp (1912) with permission from Oxford University Press

542 Functional Plant Biology A B Nicotra et al

defend themselves from herbivores and pathogens Importantlymany of these functions have conflicting physiologicalanatomical or morphological requirements For exampleorientations of lamina that maximise light interception alsominimise the ability to dissipate heat Further all of thesefunctions are intrinsically size dependent For example thequantity of mechanical tissues required to support laminaedepends on the mass of laminae whereas the ability tointercept light depends on laminae area

When seen in this light the trade-offs required for thesuccessful functionality of foliar leaves generate rather thanconstrain ecological and evolutionary opportunities This canbe illustrated mathematically and empirically in the context ofphyllotaxy and its consequences on light interception Computersimulations of mathematically generated shoots to assess theinfluence of leaf shape size and phyllotactic patterns on theability to intercept direct solar radiation show that differences inphyllotaxy significantly influence light interception (particularlyfor shoots with a rosette growth habit) They also show thatcomparatively small differences in leaf shape can compensatefor the negative effects of leaf overlap resulting from virtuallyany phyllotactic pattern For example lobed leaves or pinnifid

compound leaves facilitate light penetration through shootsbearing densely packed leaves Thus phyllotaxy can beviewed as a developmental limiting factor that can drivecompensatory changes in morphological features such asshape that are not directly controlled by patterns of leafinitiation In this way the lsquoconstraintsrsquo imposed by phyllotaxycan foster evolutionary modifications at the level of the entireshoot (Niklas 1988)

One particularly important trade-off operating at the level ofthe architecture of a shoot is that between leaf size and numberFor any given leaf biomass allocation an individual shoot caneither have a few large leaves or many smaller ones Acrossdifferent herbaceous species and juvenile specimens of treespecies drawn from diverse lineages and ecological settingsthe relationship between total leaf biomass per plant (ML) andtotal stemmass per plant (MS) fallswithin a comparatively narrowcorridor of variation that obeys a positive isometric scalingrelationship (ie MLMS

~10 Niklas 2004) Accordinglyacross these species any stem mass investment yields a moreor less proportional increase in leaf mass Therefore it followsmathematically that across these species the relationshipamong the average mass of a typical leaf mL total leaf numberNL and total stem mass complies with the proportionality mL

NLMS~10 When individual shoots are sampled across

diverse species the relationship between mL and NL per shootis negative and isometric (ie mLNL

ndash1 Kleiman and Aarssen2007) a relationship that appears to be insensitive to howspecies are grouped according to other functional traits oraccording to their habitat preference (Yang et al 2008) Thatthis phenomenology reflects a biomass allocation trade-off isclear What remains uncertain is whether it is a simple trade-offbetween leaf mass investment and leaf size or a more complexset of relationships imposed by twig mass investments andhydraulics or biomechanics

Another trade-off that appears to hold across many but not allspecies is the relationship between leaf surface areaA and averageleaf mass mL For more than a few species-groupings thisrelationship is allometric with a slope that is less than onewhich indicates that increases in leaf mass investment do notresult in proportional increases in leaf area (ieAmL

lt10Niklaset al 2007) This trend of lsquodiminishing returnsrsquo has also beenreported when total leaf area per plant is plotted against total leafmass for plants differing in size (Niklas and Cobb 2008 2010)Several factorsmay contribute to these trends eg biomechanicalanalyses show that disproportionately larger investments must bemade to mechanically support photosynthetic tissues as leavesincrease in their surface area (Niklas et al 2009) In both the leafmass to stem mass and leaf mass to area relationships describedabove leaf shapevariation canprovide anextra degreeof freedomfor selection to act on at both the leaf and branch level

Case studies of leaf shape variation at different scales

Above we have discussed evolutionary and proximal drivers ofleaf shape as well as ecological correlates and functionalsignificance thereof Here we present case studies at threescales where we can examine the ecological and evolutionaryimplications of leaf shape These are the cross community scalecaseof leaf teeth as a subset of shapemodificationswithin lineage

200 bottomtop

150

100

50

0

0

08

07

06

05

04

03

5 10

Days after bud break

Leaf

siz

e (c

m2 )

Leaf

lobi

ng in

dex

(cm

2 cm

ndash2)

Days after bud break

15 20 25

0 5 10 15 20 25

Fig 5 Variation in (a) leaf size and (b) shape complexity inQuercus rubraleaves growing at the top and bottom of the tree crown Reprinted fromZwieniecki et al (2004b) with permission from John Wiley amp Sons

The evolution of leaf shape Functional Plant Biology 543

considerations of the genus Pelargonium and the familyProteaceae and finally the case of leaf shape variation within asingle genetic individual heteroblasty and heterophylly

Leaf teeth leaf shape variation at the community scale

Leaf teeth are conspicuous features inmanyplants and are notableas a leaf-shape trait that varies in predictable ways across plantcommunities Nearly 100 years ago Bailey and Sinnott (19151916) observed that plant communities from regions with lowmean annual temperature (MAT) have a higher site meanproportion of non-monocotyledonous species with toothedleaves Paleobotanists quickly seized upon this relationship asan index for inferring paleotemperature Little et al (2010)recently documented the rich interest in this topic compiling351 literature references related to leaf physiognomy andtemperature of which the primary focus is leaf teeth The linkbetween teeth and temperature is observed on all vegetatedcontinents (Greenwood 2005a Peppe Royer et al 2011) andis also seen in related metrics such as tooth size and tooth number(Huff et al 2003 Royer et al 2005 Peppe et al 2011) Furtherthese tooth variables can display phenotypic plasticity totemperature change (Fig 6 Royer et al 2009b)

What is knownabout the functional bases of these leafndashclimaterelationships Leaf teeth commonly expand and mature fasterthan the bulk leaf (eg Billings 1905 Feild et al 2005) and youngteeth are disproportionately vigorous in their gas exchange(Baker-Brosh and Peet 1997 Royer and Wilf 2006) Fromthese observations Royer and Wilf (2006) hypothesised thatone role of teeth is to increase sap flow thereby deliveringnutrients and other solutes to young emerging leaves Inhabitats with low temperatures teeth could play an importantrole in maximising the potential for carbon gain and growth earlyin the season In warmer climates with longer growing seasonsthe carbon benefit conferred by teeth would be outweighed bytheir unavoidable water cost Alternatively Feild et al (2005)hypothesised that teeth help remove freezendashthaw embolismsthrough positive root pressure and guttation Importantly thetransport of fluids is common to both hypotheses From this itmay be surmised that water-limiting conditions select againstteeth (Bailey andSinnott 1916) Indeed this pattern is clearly seenacross local water availability gradients (Burnham et al 2001

Kowalski and Dilcher 2003 Greenwood 2005b Royer et al2009a Peppe et al 2011) but not globally against mean annualprecipitation (eg Peppe et al 2011)

If the hypothesis by Royer andWilf (2006) is correct toothedspecies should have a higher photosynthetic rate at least at thebeginning of the growing season As such teethmay be related tothe leaf economics spectrum (Wright et al 2004) Some studiesfind support for this idea leaves with a short leaf lifespan(deciduous) (Peppe et al 2011) and low leaf mass per area(Royer et al 2005) two leaf economic variables are morelikely to be toothed It also follows that teeth should be morefunctionally related to growing season climatic variables likeseasonality growing season length and growing degree days notmean annual temperature however few studies find support forthis idea (Royer et al 2005 Peppe et al 2011) Ultimately itwould be valuable to have a better understanding of the geneticprocesses underlying tooth development but presently little isknown outside model species (see lsquoGenetic regulation forpatterning of leaf margin serration and leaflet formationrsquosection) Further phylogenetic history has typically beenassumed to have only a minor role in these leaf-climatepatterns but rigorous tests demonstrate that this assumption isfalse (Little et al 2010) This outcome is of particular concern forpaleobotanists because systematic placement of fossils can bedifficult especially for the Cretaceous and Paleogene Tools toaccommodate for this phylogenetic dependency are needed

Leaf shape variation within lineages

Proteaceae

Proteaceae is an ancient angiosperm family originating inGondwana over 100million year ago and possessing 1700 extantspecies (Weston 2007) The familyrsquos greatest representation is inAustralia followed by SouthAfrica with smaller numbers acrossother Gondwanan segments (Cowling and Lamont 1998 Hootand Douglas 1998 Weston 2007) Within the Proteaceae thereexists an enormous range of leaf sizes and shapes ranging froma few mm2 to 500 cm2 and including entire toothed broadlylobed deeply dissected needle-like simple or compound leavesIntriguingly most of this shape diversity is represented inAustralia The majority of taxa outside Australia possessentire leaves the exception being four South African genera

(a) (b) (c) (d )

Fig 6 Representative leaves of Acer rubrum showing genetic variation and plasticity in leaf teeth Leaf derived from coolclimate (Ontario) seed stock grown in (a) Rhode Island (cool climate) and (b) Florida (warm climate) leaf derived fromwarm climate (Florida) seed stock grown in (c) Rhode Island and (d) Florida Bar (for all leaves) = 1 cm Reprinted from Royeret al (2009) under Creative Commons Attribution Licence

544 Functional Plant Biology A B Nicotra et al

that possess dissected (Serruria and Paranomis) or toothed(Mimites and Leucospermum) leafed species all of whichoccur in a single clade (Sauquet et al 2009)

The reason for the extensive radiation in leaf shape diversityin Australian Proteaceae is unclear Among South AfricanProteaceae leaf size decreases with mean annual temperatureand narrow leaves have been shown to overheat less than theirwider counterparts (Yates et al 2010) Given that leaf dissectionand small size are functionally equivalent with respect to heatloss (see lsquoTemperature and waterrsquo section) it is tempting tosuggest that leaf shape variation in Australia represents anadaptive response to increasing hot dry conditions Yet thereappears to be no consistent distributional pattern of Proteaceaeleaf dimensions inAustraliaManymediumor large un-dissectedleaves exist in the hottest habitats Grevilleas with small simpleleaves can be found across most regions including high rainfallareas and some of the species found in the shaded rainforestunderstorey have among the most deeply dissected leaves of allLeaf shape also varies within a single individual For examplefern-leaf stenocarpus Stenocarpus davallioides Foreman ampBHyland has tripinnatisect juvenile leaves but adult leavescan be pinnate or bipinnate (Foreman and Hyland 1995)Other rainforest species such as Atherton Oak (Athertoniadiversifolia (CTWhit) LAS Johnson amp BGBriggs) havevery large lobed intermediate leaves but adult and seedlingleaves are small and entire (Weston 1995) Alternatively leaflobing within a plant can be random having no obvious spatialor developmental association or homogenous across a speciesExamining a phylogeny of extant taxa one finds lobed ordissected leaves alongside entire-leafed species ndash across mostlineages in the group (Weston 2007) Therefore althoughsuggestions of microclimatic adaptation may be offered atthe species level it seems unlikely that an environmentalassociation with leaf shape could be generalised acrossAustralian Proteaceae Rather leaf shape variation in theProteaceae is likely to reflect the range of ecological solutionsto leaf structurefunction trade-offs discussed above

Pelargonium

Similar to the Proteaceae the genus Pelargonium isdistributed predominantly in the southern hemisphere withnearly 80 of its almost 300 species found in South AfricaAlthough the age of the genus is estimated to be around30million years (Bakker et al 2005) the highest rate ofspecies accumulation has occurred in the last 10million years(Martinez-Cabrera 2010) The combination of climatic andedaphic heterogeneity over short distances and limited geneflow (Latimer et al 2005 Linder 2005) is frequently offeredas an explanation for the extraordinary plant species diversity thatcharacterises southern Africa Pelargonium the third largestgenus in the region is characterised by remarkable variation ingrowth form that ranges from annuals and geophytes to shrubsVariation in leaf shape is no less dramatic ranging from entireto highly dissected (Fig 1) often within growth forms andtaxonomic sections Parsing of leaf form categorically revealedthat the extent of dissection of the blade is highly evolvablewithinanoverall ovate leaf shapeoutline that is evolutionarily conserved(Jones et al 2009)

In addition to high evolutionary lability of leaf dissectionthere is variation among major clades in the evolutionary patternof major veins One major clade (Clade C) is dominated bypalmate venation whereas another (Clade A) is dominated bypinnate venation although both types are found in both cladesWe note that pinnate venation is significantly associated withincreased dissection and reductions in functional leaf sizecharacteristic of the highly diverse lsquoxerophyticrsquo clade (CladeA2) of the winter rainfall region

To examine whether leaf shape translates into functionalvariation carbon gain and water use efficiency were measuredin pairs of species contrasted for leaf shape and replicated acrossthree subclades of Clade A and one subclade of the third majorclade CladeB (Nicotra et al 2008) In each case the species withthemore dissected leaves showed higher rates of carbon gain butalso higher rates of water loss in all treatment conditions Thatpaper argued that the differences in carbon gain were not directresults of leaf shape per se but rather that the same conditionsthat result in selection for more dissected leaves also favourthe evolution of high photosynthetic rates high leaf nitrogencontent and opportunistic use of water when available Evenin this study however there was no correlation between leafdissection and the climate present in native ranges suggestingeither that climate is selective at a much smaller scale (ie themicrohabitat) or that for any given set of climate variablesmultiple leaf shapes can be adaptive in association withvariation in other leaf traits

Heteroblasty and heterophylly variationwithin a single genetic individual

We have so far focussed primarily on cross species variation inleaf shape As discussed for the Proteaceae genera above (lsquoTheProteaceaersquo section) however a single genome can produce aremarkable range of leaf shapes during the lifetime of a plantThere are two broad categories of leaf shape variation that can beexpressed by a single individual This variation can arise either inresponse to different external environments or to ontogeneticsignals (some of which may be influenced by environment aswell)

Environmentally induced heterophylly

Environmentally induced heterophylly results when leafshape responds to environmental cues The classic case is thedramatically different emergent and submergent leaves in aquaticplants Although many aquatic plants are also heteroblastic (seebelow Sculthorpe 1967) environmentally induced heterophyllyis not specific to position or plant age and is reversible dependinglargely on the environmental signals (mediated by hormonalsignals) perceived by cells of the developing leaf primordiumThe developmental age of the leaf at the time of signal perceptionis critical Very young primordia (L1ndashL3) generally respondcompletely to the novel environment whereas primordiathat were slightly older at the time of environmental changeshow intermediate features as mature leaves Characteristics ofspecific cells are related to the position of that cell within thedifferentiation gradient characteristic of that species at the timeof environmental change (eg Goliber and Feldman 1990 Bruniet al 1996 Kuwabara and Nagata 2006)

The evolution of leaf shape Functional Plant Biology 545

Among terrestrial plants the lsquosunndashshadersquo leaf response isfrequently cited as the classic example of environmentallyinduced heterophylly Historically differences in leaf formwithin a canopy have been assumed to be responses to alteredlight environments (eg larger and less-lobed leaves are producedin shade)However these different leaf shapes also confer distincthydraulic advantages (see lsquoTemperature and waterrsquo section)Whether in response to light or to hydraulic limitation lsquosunshadersquo heterophylly is accompanied by similar and well-knownanatomical differences across a wide range of species (ie sunleaves are generally smaller and thicker with smaller cells thanshade leaves Taiz and Zeiger 2002) However when theseanatomical differences become determined developmentally(Dengler 1980 Yano and Terashima 2004) has been lesscommonly studied The presence of intermediate anatomicalcharacteristics in leaves transferred among light environmentsshortly after expansion ceased suggests that leaves of somespecies have the capacity to respond anatomically to alteredlight levels until very late in development (Oguchi et al2005) In contrast recent studies suggest that systemic lightand humidity signals sent from older to younger leavesallow younger leaves to develop phenotypes suited for theiranticipated environments (Lake et al 2001 Yano andTerashima 2001 Thomas et al 2004) Exactly when final leafshape is determined in lsquosunshadersquo leaves remains an openquestion for most species

Heteroblasty describes ontogenetically dependentchanges in leaf features most notably shape

In 1900 Goebel originally distinguished differences in lsquojuvenileformrsquo from lsquoadult formrsquo on the basis of dramatic differences inleaf shape and associated traits (Balfour 1969) As an example he

described the shift from compound leaves to phyllodes in Acaciaas heteroblastic and reserved the term homoblastic for the lessdramatic variation exhibited by Casurina It is now generallyaccepted that heteroblasty refers to the full suite of features thatchange along the shoot during development (Jones 1999)

Heteroblasty results from the interaction of developmentalprograms expressed at both the leaf and the shoot- or whole-plant level (Kerstetter and Poethig 1998 Tsukaya et al 2000Usami et al 2009 Wu et al 2009 Chuck and OrsquoConnor 2010)Heteroblasty is generally not reversible unless reiterative growthoccurs (ie following damage) Morphological differencescharacteristic of heteroblasty such as differences in lobing areusually expressed relatively early in development shortly afterleaf initiation (Jones 1995)

The sequence of heteroblastic leaf shapes produced alongthe shoot can be accelerated or delayed in response to differentgrowth environments or architectural manipulations (Ashby1948 Forster and Bonser 2009 Jaya et al 2010)Furthermore both heteroblasty and plasticity in heteroblastyhave been shown to be under strong genetic control (Anderson1989 Climent et al 2006) Forster et al (2011) suggested thatboth heteroblasty and plasticity in heteroblasty are likely to occurin response to lsquowhich leaf type confers the greatest benefit fora given light levelrsquo Several studies over the last decade havepointed towards functional consequences of heteroblastygenerally with reference to single environmental variables (egHansen 1986 Gamage and Jesson 2007 Kubien et al 2007)It is likely that heteroblasty influences multiple leaf functionsbecause suites of leaf traits change with heteroblasty (Jones2001) Indeed this conclusion was supported in a recent studythat examined differences inmore than a hundred leaf traits in koa(Acacia koa AGray Pasquet-Kok et al 2010) Heteroblasty hasalso been proposed as an adaptation to herbivory (Karban and

Table 1 Promising directions for cross-disciplinary investigation of leaf shape

No Direction

1 A multi-species lsquoevo-devorsquo approach similar to that by Illing et al (2009) is likely to be informative with regard to many enigmatic features regardingthe evolution of diverse leaf forms and more generally provides an exciting system for examining the convergence and divergence in the evolutionof function

2 A phylogenetic analysis of leaf shape examining correlated evolution with other traits such as venation size leaf surface characteristics and branch andarchitectural measures would enable a thorough investigation of the situations in which leaf shape diversification arises

3 Further research into both the genetic and developmental processes underlying lobes and teeth and the ecological correlates of each would provide anexcellent opportunity to link across wide environmental scales

4 Cross species studies ideally in the field of the relative importance of leaf shape among other factors in determining leaf temperature and water status arenecessary to test the conclusion put forward here that the functional significance of leaf shape lies more often in water relations than directly in leafthermoregulation

5 Studies of the scale of environmental change that selects for ontogenetic response in leaf shape in heteroblastic species fromdiverse environments will shedlight on links between genes development and ecology

6 Detailed quantifications of the extent towhich leaf shape differswithin canopies particularly of dioecious specieswhere selection on reproductive functioncan lead to dimorphic conditions will help link leaf shape to whole plant processes

7 Studies at the scale of transects across the leaf surface that simultaneously consider hydraulic resistance temperature and light will reveal the extent oflocalised hydraulic and thermal stresses within leaf surfaces that might act as selective forces on leaf shape per se Imaging approaches incorporatingmeasures of 3-dimensional shape temperature and fluorescence provide exciting opportunities for examining spatial heterogeneity in temperature andphysiological processes

8 Since patterns of leaf venation affect both hydraulic supply and capacity for photosynthate transport studies that simultaneously examine both willreveal the extent to which interactions among these fundamental processes either co-determine leaf shape or establish trade-offs ultimately reflectedin leaf shape

9 Finally studies examining the fitness values of different leaf shapes or plasticity in leaf shapes under a range of environmental conditions are needed tobetter identify which aspects of leaf shape are most likely to be adaptive

546 Functional Plant Biology A B Nicotra et al

Thaler 1999 Brennan et al 2001 Fadzly et al 2009) and it hasbeen associated with differences in frost resistance (Darrow et al2001) It is worth noting that studies of the functional significanceof heteroblasty have focussed on relatively few plant lineagesGiven that heteroblasty occurs widely among the angiospermsas well as in some ferns and gymosperms (eg Wagner 1952Mueller 1982 Pryer and Hearn 2009) the next questions mayfocus on the cost of heteroblasty and its fitness consequenceswhat scale of environmental change selects for ontogeneticresponse in leaf shape

Synthesis and next steps

Whether considered at community lineage or single genomescales leaf shape is a trait that is highly labile and responsive to arange of biotic and abiotic factors Although some leaf traitscan be explained in terms of particularly strong trade-offsbetween two or more factors that result in clear scalingrelationships (eg leaf mass-to-area traits or leaf mass-to-stemmass) leaf shape in contrast does not emerge from the aboveconsideration as a factor capable of explaining ecologicaldifferentiation between species Rather leaf shape emerges asa trait for which there are many quite varied functional trade-offs

As stated above these trade-offs for successful functionalityof leaves may generate ecological opportunities leaf shape isperhaps best viewed not as a single major axis but rather asan option that fine tunes the leaf to its conditions over bothshort and evolutionary time spans Numerous elements of thisreview support this conclusion Over evolutionary history theextant range of variation in leaf shape has been explored bymanylineages although only the angiosperms have explored thisrange with the added element of massive increases in veindensities Likewise our growing understanding of geneticcontrols on leaf shape reveal that it is a complex traitdetermined by many different genes at several differentdevelopmental stages Although some genes have conservedaffects on leaf morphology including shape the exact natureof their affects can vary widely Likewise the function to whichagivenelement of leaf shape lsquois putrsquo canvary For example closerinvestigation of the internal architecture and developmentalpathways associated with different leaf shapes suggests thatleaf teeth and leaf lobes often have distinctly differentfunctional foundations and may play different functions indifferent environments

Despite amyriadof functional roles the one recurrent theme inour review is that leaf shape is affected by and strongly influencesleaf water relations From a functional perspective the role of leafshape in leaf thermoregulation has perhaps received the mostattention in previous considerations of the functional significanceof leaf shape The relationship between leaf size and temperaturebeing physically determined by basic energy balance principlesthis was a logical starting point of enquiry But as we haveshown the connection between leaf shape (and size) and leaftemperatures under field conditions is not well supportedempirically Rather leaf shape variation as distinct from sizemore often will reflect water supply trade-offs High veindensities support high photosynthetic rates (Brodribb et al2010) and dissected leaves maintain a greater proportion ofleaf tissue closer to main veins Leaf temperature plays a role

in this story as higher leaf temperatures lead to greaterevaporative water loss but many factors other than leaf shapeinfluence leaf temperature (eg leaf thickness orientation andreflectivity and branch and plant architecture) Finally it is alsoworth noting that although there appears to be an obviouslink between shape and water supply there is a concomitantinteraction between water supply and phloem function thatremains relatively unexplored

Thus our review suggests several angles for further research(Table 1) In particular we advocate those approaches thatenable investigation across the scales examined here (genedevelopment function and evolutionary ecology) and thatfocus in particular on the links among leaf shape developmentand water relations

Acknowledgements

The authors thank the Botanical Society of America for their support of thesymposiumfromwhich this reviewaroseWealso thankLawrenSackand twoanonymous reviewers for useful comments on an earlier version of thismanuscript

References

Ackerly DD Knight CA Weiss SB Barton K Starmer KP (2002) Leafsize specific leaf area and microhabitat distribution of chaparral woodyplants contrasting patterns in species level and community level analysesOecologia 130 449ndash457 doi101007s004420100805

Anderson S (1989) Variation in heteroblastic succession among populationsof Crepis tectorum Nordic Journal of Botany 8 565ndash573 doi101111j1756-10511989tb01730x

Armbruster WS Muchhala N (2009) Associations between floralspecialization and species diversity cause effect or correlationEvolutionary Ecology 23 159ndash179 doi101007s10682-008-9259-z

Armbruster WS Di Stilio VS Tuxill JD Flores TC Runk JLV (1999)Covariance and decoupling of floral and vegetative traits in nineneotropical plants a re-evaluation of Bergrsquos correlation-pleiadesconcept American Journal of Botany 86 39ndash55 doi1023072656953

Ashby E (1948) Studies in the morphogenesis of leaves I An essay on leafshape New Phytologist 47 153ndash176 doi101111j1469-81371948tb05098x

Aso K Kato M Banks JA Hasebe M (1999) Characterization ofhomeodomain-leucine zipper genes in the fern Ceratopteris richardiiand the evolution of the homeodomain-leucine zipper gene family invascular plants Molecular Biology and Evolution 16 544ndash552

Bailey IW Sinnott EW (1915) A botanical index of Cretaceous and Tertiaryclimates Science 41 831ndash834 doi101126science411066831

Bailey IW Sinnott EW (1916) The climatic distribution of certain types ofangiosperm leaves American Journal of Botany 3 24ndash39 doi1023072435109

Baker-Brosh KF Peet RK (1997) The ecological significance of lobed andtoothed leaves in temperate forest trees Ecology 78 1250ndash1255

Bakker FT Culham A Marais EM Gibby M (2005) Nested radiation inCape Pelargonium In lsquoPlant species-level systematics new perspectiveson pattern and process Vol 143rsquo (Eds FT Bakker LW Chartrou BGravendeel PB Pielser) pp 75ndash100 (ARG Ganter Verlag KG RuggellLichtstein)

Balfour IB (Transl) (1969) lsquoOrganography of plants I General organographyrsquoby K Goebel 1900 (Hafner Publishing Co Inc New York)

Ball MC Cowan IR Farquhar GD (1988) Maintenance of leaf temperatureand the optimization of carbon gain in relation to water loss in a tropicalmangrove forest Australian Journal of Plant Physiology 15 263ndash276doi101071PP9880263

The evolution of leaf shape Functional Plant Biology 547

defend themselves from herbivores and pathogens Importantlymany of these functions have conflicting physiologicalanatomical or morphological requirements For exampleorientations of lamina that maximise light interception alsominimise the ability to dissipate heat Further all of thesefunctions are intrinsically size dependent For example thequantity of mechanical tissues required to support laminaedepends on the mass of laminae whereas the ability tointercept light depends on laminae area

When seen in this light the trade-offs required for thesuccessful functionality of foliar leaves generate rather thanconstrain ecological and evolutionary opportunities This canbe illustrated mathematically and empirically in the context ofphyllotaxy and its consequences on light interception Computersimulations of mathematically generated shoots to assess theinfluence of leaf shape size and phyllotactic patterns on theability to intercept direct solar radiation show that differences inphyllotaxy significantly influence light interception (particularlyfor shoots with a rosette growth habit) They also show thatcomparatively small differences in leaf shape can compensatefor the negative effects of leaf overlap resulting from virtuallyany phyllotactic pattern For example lobed leaves or pinnifid

compound leaves facilitate light penetration through shootsbearing densely packed leaves Thus phyllotaxy can beviewed as a developmental limiting factor that can drivecompensatory changes in morphological features such asshape that are not directly controlled by patterns of leafinitiation In this way the lsquoconstraintsrsquo imposed by phyllotaxycan foster evolutionary modifications at the level of the entireshoot (Niklas 1988)

One particularly important trade-off operating at the level ofthe architecture of a shoot is that between leaf size and numberFor any given leaf biomass allocation an individual shoot caneither have a few large leaves or many smaller ones Acrossdifferent herbaceous species and juvenile specimens of treespecies drawn from diverse lineages and ecological settingsthe relationship between total leaf biomass per plant (ML) andtotal stemmass per plant (MS) fallswithin a comparatively narrowcorridor of variation that obeys a positive isometric scalingrelationship (ie MLMS

~10 Niklas 2004) Accordinglyacross these species any stem mass investment yields a moreor less proportional increase in leaf mass Therefore it followsmathematically that across these species the relationshipamong the average mass of a typical leaf mL total leaf numberNL and total stem mass complies with the proportionality mL

NLMS~10 When individual shoots are sampled across

diverse species the relationship between mL and NL per shootis negative and isometric (ie mLNL

ndash1 Kleiman and Aarssen2007) a relationship that appears to be insensitive to howspecies are grouped according to other functional traits oraccording to their habitat preference (Yang et al 2008) Thatthis phenomenology reflects a biomass allocation trade-off isclear What remains uncertain is whether it is a simple trade-offbetween leaf mass investment and leaf size or a more complexset of relationships imposed by twig mass investments andhydraulics or biomechanics

Another trade-off that appears to hold across many but not allspecies is the relationship between leaf surface areaA and averageleaf mass mL For more than a few species-groupings thisrelationship is allometric with a slope that is less than onewhich indicates that increases in leaf mass investment do notresult in proportional increases in leaf area (ieAmL

lt10Niklaset al 2007) This trend of lsquodiminishing returnsrsquo has also beenreported when total leaf area per plant is plotted against total leafmass for plants differing in size (Niklas and Cobb 2008 2010)Several factorsmay contribute to these trends eg biomechanicalanalyses show that disproportionately larger investments must bemade to mechanically support photosynthetic tissues as leavesincrease in their surface area (Niklas et al 2009) In both the leafmass to stem mass and leaf mass to area relationships describedabove leaf shapevariation canprovide anextra degreeof freedomfor selection to act on at both the leaf and branch level

Case studies of leaf shape variation at different scales

Above we have discussed evolutionary and proximal drivers ofleaf shape as well as ecological correlates and functionalsignificance thereof Here we present case studies at threescales where we can examine the ecological and evolutionaryimplications of leaf shape These are the cross community scalecaseof leaf teeth as a subset of shapemodificationswithin lineage

200 bottomtop

150

100

50

0

0

08

07

06

05

04

03

5 10

Days after bud break

Leaf

siz

e (c

m2 )

Leaf

lobi

ng in

dex

(cm

2 cm

ndash2)

Days after bud break

15 20 25

0 5 10 15 20 25

Fig 5 Variation in (a) leaf size and (b) shape complexity inQuercus rubraleaves growing at the top and bottom of the tree crown Reprinted fromZwieniecki et al (2004b) with permission from John Wiley amp Sons

The evolution of leaf shape Functional Plant Biology 543

considerations of the genus Pelargonium and the familyProteaceae and finally the case of leaf shape variation within asingle genetic individual heteroblasty and heterophylly

Leaf teeth leaf shape variation at the community scale

Leaf teeth are conspicuous features inmanyplants and are notableas a leaf-shape trait that varies in predictable ways across plantcommunities Nearly 100 years ago Bailey and Sinnott (19151916) observed that plant communities from regions with lowmean annual temperature (MAT) have a higher site meanproportion of non-monocotyledonous species with toothedleaves Paleobotanists quickly seized upon this relationship asan index for inferring paleotemperature Little et al (2010)recently documented the rich interest in this topic compiling351 literature references related to leaf physiognomy andtemperature of which the primary focus is leaf teeth The linkbetween teeth and temperature is observed on all vegetatedcontinents (Greenwood 2005a Peppe Royer et al 2011) andis also seen in related metrics such as tooth size and tooth number(Huff et al 2003 Royer et al 2005 Peppe et al 2011) Furtherthese tooth variables can display phenotypic plasticity totemperature change (Fig 6 Royer et al 2009b)

What is knownabout the functional bases of these leafndashclimaterelationships Leaf teeth commonly expand and mature fasterthan the bulk leaf (eg Billings 1905 Feild et al 2005) and youngteeth are disproportionately vigorous in their gas exchange(Baker-Brosh and Peet 1997 Royer and Wilf 2006) Fromthese observations Royer and Wilf (2006) hypothesised thatone role of teeth is to increase sap flow thereby deliveringnutrients and other solutes to young emerging leaves Inhabitats with low temperatures teeth could play an importantrole in maximising the potential for carbon gain and growth earlyin the season In warmer climates with longer growing seasonsthe carbon benefit conferred by teeth would be outweighed bytheir unavoidable water cost Alternatively Feild et al (2005)hypothesised that teeth help remove freezendashthaw embolismsthrough positive root pressure and guttation Importantly thetransport of fluids is common to both hypotheses From this itmay be surmised that water-limiting conditions select againstteeth (Bailey andSinnott 1916) Indeed this pattern is clearly seenacross local water availability gradients (Burnham et al 2001

Kowalski and Dilcher 2003 Greenwood 2005b Royer et al2009a Peppe et al 2011) but not globally against mean annualprecipitation (eg Peppe et al 2011)

If the hypothesis by Royer andWilf (2006) is correct toothedspecies should have a higher photosynthetic rate at least at thebeginning of the growing season As such teethmay be related tothe leaf economics spectrum (Wright et al 2004) Some studiesfind support for this idea leaves with a short leaf lifespan(deciduous) (Peppe et al 2011) and low leaf mass per area(Royer et al 2005) two leaf economic variables are morelikely to be toothed It also follows that teeth should be morefunctionally related to growing season climatic variables likeseasonality growing season length and growing degree days notmean annual temperature however few studies find support forthis idea (Royer et al 2005 Peppe et al 2011) Ultimately itwould be valuable to have a better understanding of the geneticprocesses underlying tooth development but presently little isknown outside model species (see lsquoGenetic regulation forpatterning of leaf margin serration and leaflet formationrsquosection) Further phylogenetic history has typically beenassumed to have only a minor role in these leaf-climatepatterns but rigorous tests demonstrate that this assumption isfalse (Little et al 2010) This outcome is of particular concern forpaleobotanists because systematic placement of fossils can bedifficult especially for the Cretaceous and Paleogene Tools toaccommodate for this phylogenetic dependency are needed

Leaf shape variation within lineages

Proteaceae

Proteaceae is an ancient angiosperm family originating inGondwana over 100million year ago and possessing 1700 extantspecies (Weston 2007) The familyrsquos greatest representation is inAustralia followed by SouthAfrica with smaller numbers acrossother Gondwanan segments (Cowling and Lamont 1998 Hootand Douglas 1998 Weston 2007) Within the Proteaceae thereexists an enormous range of leaf sizes and shapes ranging froma few mm2 to 500 cm2 and including entire toothed broadlylobed deeply dissected needle-like simple or compound leavesIntriguingly most of this shape diversity is represented inAustralia The majority of taxa outside Australia possessentire leaves the exception being four South African genera

(a) (b) (c) (d )

Fig 6 Representative leaves of Acer rubrum showing genetic variation and plasticity in leaf teeth Leaf derived from coolclimate (Ontario) seed stock grown in (a) Rhode Island (cool climate) and (b) Florida (warm climate) leaf derived fromwarm climate (Florida) seed stock grown in (c) Rhode Island and (d) Florida Bar (for all leaves) = 1 cm Reprinted from Royeret al (2009) under Creative Commons Attribution Licence

544 Functional Plant Biology A B Nicotra et al

that possess dissected (Serruria and Paranomis) or toothed(Mimites and Leucospermum) leafed species all of whichoccur in a single clade (Sauquet et al 2009)

The reason for the extensive radiation in leaf shape diversityin Australian Proteaceae is unclear Among South AfricanProteaceae leaf size decreases with mean annual temperatureand narrow leaves have been shown to overheat less than theirwider counterparts (Yates et al 2010) Given that leaf dissectionand small size are functionally equivalent with respect to heatloss (see lsquoTemperature and waterrsquo section) it is tempting tosuggest that leaf shape variation in Australia represents anadaptive response to increasing hot dry conditions Yet thereappears to be no consistent distributional pattern of Proteaceaeleaf dimensions inAustraliaManymediumor large un-dissectedleaves exist in the hottest habitats Grevilleas with small simpleleaves can be found across most regions including high rainfallareas and some of the species found in the shaded rainforestunderstorey have among the most deeply dissected leaves of allLeaf shape also varies within a single individual For examplefern-leaf stenocarpus Stenocarpus davallioides Foreman ampBHyland has tripinnatisect juvenile leaves but adult leavescan be pinnate or bipinnate (Foreman and Hyland 1995)Other rainforest species such as Atherton Oak (Athertoniadiversifolia (CTWhit) LAS Johnson amp BGBriggs) havevery large lobed intermediate leaves but adult and seedlingleaves are small and entire (Weston 1995) Alternatively leaflobing within a plant can be random having no obvious spatialor developmental association or homogenous across a speciesExamining a phylogeny of extant taxa one finds lobed ordissected leaves alongside entire-leafed species ndash across mostlineages in the group (Weston 2007) Therefore althoughsuggestions of microclimatic adaptation may be offered atthe species level it seems unlikely that an environmentalassociation with leaf shape could be generalised acrossAustralian Proteaceae Rather leaf shape variation in theProteaceae is likely to reflect the range of ecological solutionsto leaf structurefunction trade-offs discussed above

Pelargonium

Similar to the Proteaceae the genus Pelargonium isdistributed predominantly in the southern hemisphere withnearly 80 of its almost 300 species found in South AfricaAlthough the age of the genus is estimated to be around30million years (Bakker et al 2005) the highest rate ofspecies accumulation has occurred in the last 10million years(Martinez-Cabrera 2010) The combination of climatic andedaphic heterogeneity over short distances and limited geneflow (Latimer et al 2005 Linder 2005) is frequently offeredas an explanation for the extraordinary plant species diversity thatcharacterises southern Africa Pelargonium the third largestgenus in the region is characterised by remarkable variation ingrowth form that ranges from annuals and geophytes to shrubsVariation in leaf shape is no less dramatic ranging from entireto highly dissected (Fig 1) often within growth forms andtaxonomic sections Parsing of leaf form categorically revealedthat the extent of dissection of the blade is highly evolvablewithinanoverall ovate leaf shapeoutline that is evolutionarily conserved(Jones et al 2009)

In addition to high evolutionary lability of leaf dissectionthere is variation among major clades in the evolutionary patternof major veins One major clade (Clade C) is dominated bypalmate venation whereas another (Clade A) is dominated bypinnate venation although both types are found in both cladesWe note that pinnate venation is significantly associated withincreased dissection and reductions in functional leaf sizecharacteristic of the highly diverse lsquoxerophyticrsquo clade (CladeA2) of the winter rainfall region

To examine whether leaf shape translates into functionalvariation carbon gain and water use efficiency were measuredin pairs of species contrasted for leaf shape and replicated acrossthree subclades of Clade A and one subclade of the third majorclade CladeB (Nicotra et al 2008) In each case the species withthemore dissected leaves showed higher rates of carbon gain butalso higher rates of water loss in all treatment conditions Thatpaper argued that the differences in carbon gain were not directresults of leaf shape per se but rather that the same conditionsthat result in selection for more dissected leaves also favourthe evolution of high photosynthetic rates high leaf nitrogencontent and opportunistic use of water when available Evenin this study however there was no correlation between leafdissection and the climate present in native ranges suggestingeither that climate is selective at a much smaller scale (ie themicrohabitat) or that for any given set of climate variablesmultiple leaf shapes can be adaptive in association withvariation in other leaf traits

Heteroblasty and heterophylly variationwithin a single genetic individual

We have so far focussed primarily on cross species variation inleaf shape As discussed for the Proteaceae genera above (lsquoTheProteaceaersquo section) however a single genome can produce aremarkable range of leaf shapes during the lifetime of a plantThere are two broad categories of leaf shape variation that can beexpressed by a single individual This variation can arise either inresponse to different external environments or to ontogeneticsignals (some of which may be influenced by environment aswell)

Environmentally induced heterophylly

Environmentally induced heterophylly results when leafshape responds to environmental cues The classic case is thedramatically different emergent and submergent leaves in aquaticplants Although many aquatic plants are also heteroblastic (seebelow Sculthorpe 1967) environmentally induced heterophyllyis not specific to position or plant age and is reversible dependinglargely on the environmental signals (mediated by hormonalsignals) perceived by cells of the developing leaf primordiumThe developmental age of the leaf at the time of signal perceptionis critical Very young primordia (L1ndashL3) generally respondcompletely to the novel environment whereas primordiathat were slightly older at the time of environmental changeshow intermediate features as mature leaves Characteristics ofspecific cells are related to the position of that cell within thedifferentiation gradient characteristic of that species at the timeof environmental change (eg Goliber and Feldman 1990 Bruniet al 1996 Kuwabara and Nagata 2006)

The evolution of leaf shape Functional Plant Biology 545

Among terrestrial plants the lsquosunndashshadersquo leaf response isfrequently cited as the classic example of environmentallyinduced heterophylly Historically differences in leaf formwithin a canopy have been assumed to be responses to alteredlight environments (eg larger and less-lobed leaves are producedin shade)However these different leaf shapes also confer distincthydraulic advantages (see lsquoTemperature and waterrsquo section)Whether in response to light or to hydraulic limitation lsquosunshadersquo heterophylly is accompanied by similar and well-knownanatomical differences across a wide range of species (ie sunleaves are generally smaller and thicker with smaller cells thanshade leaves Taiz and Zeiger 2002) However when theseanatomical differences become determined developmentally(Dengler 1980 Yano and Terashima 2004) has been lesscommonly studied The presence of intermediate anatomicalcharacteristics in leaves transferred among light environmentsshortly after expansion ceased suggests that leaves of somespecies have the capacity to respond anatomically to alteredlight levels until very late in development (Oguchi et al2005) In contrast recent studies suggest that systemic lightand humidity signals sent from older to younger leavesallow younger leaves to develop phenotypes suited for theiranticipated environments (Lake et al 2001 Yano andTerashima 2001 Thomas et al 2004) Exactly when final leafshape is determined in lsquosunshadersquo leaves remains an openquestion for most species

Heteroblasty describes ontogenetically dependentchanges in leaf features most notably shape

In 1900 Goebel originally distinguished differences in lsquojuvenileformrsquo from lsquoadult formrsquo on the basis of dramatic differences inleaf shape and associated traits (Balfour 1969) As an example he

described the shift from compound leaves to phyllodes in Acaciaas heteroblastic and reserved the term homoblastic for the lessdramatic variation exhibited by Casurina It is now generallyaccepted that heteroblasty refers to the full suite of features thatchange along the shoot during development (Jones 1999)

Heteroblasty results from the interaction of developmentalprograms expressed at both the leaf and the shoot- or whole-plant level (Kerstetter and Poethig 1998 Tsukaya et al 2000Usami et al 2009 Wu et al 2009 Chuck and OrsquoConnor 2010)Heteroblasty is generally not reversible unless reiterative growthoccurs (ie following damage) Morphological differencescharacteristic of heteroblasty such as differences in lobing areusually expressed relatively early in development shortly afterleaf initiation (Jones 1995)

The sequence of heteroblastic leaf shapes produced alongthe shoot can be accelerated or delayed in response to differentgrowth environments or architectural manipulations (Ashby1948 Forster and Bonser 2009 Jaya et al 2010)Furthermore both heteroblasty and plasticity in heteroblastyhave been shown to be under strong genetic control (Anderson1989 Climent et al 2006) Forster et al (2011) suggested thatboth heteroblasty and plasticity in heteroblasty are likely to occurin response to lsquowhich leaf type confers the greatest benefit fora given light levelrsquo Several studies over the last decade havepointed towards functional consequences of heteroblastygenerally with reference to single environmental variables (egHansen 1986 Gamage and Jesson 2007 Kubien et al 2007)It is likely that heteroblasty influences multiple leaf functionsbecause suites of leaf traits change with heteroblasty (Jones2001) Indeed this conclusion was supported in a recent studythat examined differences inmore than a hundred leaf traits in koa(Acacia koa AGray Pasquet-Kok et al 2010) Heteroblasty hasalso been proposed as an adaptation to herbivory (Karban and

Table 1 Promising directions for cross-disciplinary investigation of leaf shape

No Direction

1 A multi-species lsquoevo-devorsquo approach similar to that by Illing et al (2009) is likely to be informative with regard to many enigmatic features regardingthe evolution of diverse leaf forms and more generally provides an exciting system for examining the convergence and divergence in the evolutionof function

2 A phylogenetic analysis of leaf shape examining correlated evolution with other traits such as venation size leaf surface characteristics and branch andarchitectural measures would enable a thorough investigation of the situations in which leaf shape diversification arises

3 Further research into both the genetic and developmental processes underlying lobes and teeth and the ecological correlates of each would provide anexcellent opportunity to link across wide environmental scales

4 Cross species studies ideally in the field of the relative importance of leaf shape among other factors in determining leaf temperature and water status arenecessary to test the conclusion put forward here that the functional significance of leaf shape lies more often in water relations than directly in leafthermoregulation

5 Studies of the scale of environmental change that selects for ontogenetic response in leaf shape in heteroblastic species fromdiverse environments will shedlight on links between genes development and ecology

6 Detailed quantifications of the extent towhich leaf shape differswithin canopies particularly of dioecious specieswhere selection on reproductive functioncan lead to dimorphic conditions will help link leaf shape to whole plant processes

7 Studies at the scale of transects across the leaf surface that simultaneously consider hydraulic resistance temperature and light will reveal the extent oflocalised hydraulic and thermal stresses within leaf surfaces that might act as selective forces on leaf shape per se Imaging approaches incorporatingmeasures of 3-dimensional shape temperature and fluorescence provide exciting opportunities for examining spatial heterogeneity in temperature andphysiological processes

8 Since patterns of leaf venation affect both hydraulic supply and capacity for photosynthate transport studies that simultaneously examine both willreveal the extent to which interactions among these fundamental processes either co-determine leaf shape or establish trade-offs ultimately reflectedin leaf shape

9 Finally studies examining the fitness values of different leaf shapes or plasticity in leaf shapes under a range of environmental conditions are needed tobetter identify which aspects of leaf shape are most likely to be adaptive

546 Functional Plant Biology A B Nicotra et al

Thaler 1999 Brennan et al 2001 Fadzly et al 2009) and it hasbeen associated with differences in frost resistance (Darrow et al2001) It is worth noting that studies of the functional significanceof heteroblasty have focussed on relatively few plant lineagesGiven that heteroblasty occurs widely among the angiospermsas well as in some ferns and gymosperms (eg Wagner 1952Mueller 1982 Pryer and Hearn 2009) the next questions mayfocus on the cost of heteroblasty and its fitness consequenceswhat scale of environmental change selects for ontogeneticresponse in leaf shape

Synthesis and next steps

Whether considered at community lineage or single genomescales leaf shape is a trait that is highly labile and responsive to arange of biotic and abiotic factors Although some leaf traitscan be explained in terms of particularly strong trade-offsbetween two or more factors that result in clear scalingrelationships (eg leaf mass-to-area traits or leaf mass-to-stemmass) leaf shape in contrast does not emerge from the aboveconsideration as a factor capable of explaining ecologicaldifferentiation between species Rather leaf shape emerges asa trait for which there are many quite varied functional trade-offs

As stated above these trade-offs for successful functionalityof leaves may generate ecological opportunities leaf shape isperhaps best viewed not as a single major axis but rather asan option that fine tunes the leaf to its conditions over bothshort and evolutionary time spans Numerous elements of thisreview support this conclusion Over evolutionary history theextant range of variation in leaf shape has been explored bymanylineages although only the angiosperms have explored thisrange with the added element of massive increases in veindensities Likewise our growing understanding of geneticcontrols on leaf shape reveal that it is a complex traitdetermined by many different genes at several differentdevelopmental stages Although some genes have conservedaffects on leaf morphology including shape the exact natureof their affects can vary widely Likewise the function to whichagivenelement of leaf shape lsquois putrsquo canvary For example closerinvestigation of the internal architecture and developmentalpathways associated with different leaf shapes suggests thatleaf teeth and leaf lobes often have distinctly differentfunctional foundations and may play different functions indifferent environments

Despite amyriadof functional roles the one recurrent theme inour review is that leaf shape is affected by and strongly influencesleaf water relations From a functional perspective the role of leafshape in leaf thermoregulation has perhaps received the mostattention in previous considerations of the functional significanceof leaf shape The relationship between leaf size and temperaturebeing physically determined by basic energy balance principlesthis was a logical starting point of enquiry But as we haveshown the connection between leaf shape (and size) and leaftemperatures under field conditions is not well supportedempirically Rather leaf shape variation as distinct from sizemore often will reflect water supply trade-offs High veindensities support high photosynthetic rates (Brodribb et al2010) and dissected leaves maintain a greater proportion ofleaf tissue closer to main veins Leaf temperature plays a role

in this story as higher leaf temperatures lead to greaterevaporative water loss but many factors other than leaf shapeinfluence leaf temperature (eg leaf thickness orientation andreflectivity and branch and plant architecture) Finally it is alsoworth noting that although there appears to be an obviouslink between shape and water supply there is a concomitantinteraction between water supply and phloem function thatremains relatively unexplored

Thus our review suggests several angles for further research(Table 1) In particular we advocate those approaches thatenable investigation across the scales examined here (genedevelopment function and evolutionary ecology) and thatfocus in particular on the links among leaf shape developmentand water relations

Acknowledgements

The authors thank the Botanical Society of America for their support of thesymposiumfromwhich this reviewaroseWealso thankLawrenSackand twoanonymous reviewers for useful comments on an earlier version of thismanuscript

References

Ackerly DD Knight CA Weiss SB Barton K Starmer KP (2002) Leafsize specific leaf area and microhabitat distribution of chaparral woodyplants contrasting patterns in species level and community level analysesOecologia 130 449ndash457 doi101007s004420100805

Anderson S (1989) Variation in heteroblastic succession among populationsof Crepis tectorum Nordic Journal of Botany 8 565ndash573 doi101111j1756-10511989tb01730x

Armbruster WS Muchhala N (2009) Associations between floralspecialization and species diversity cause effect or correlationEvolutionary Ecology 23 159ndash179 doi101007s10682-008-9259-z

Armbruster WS Di Stilio VS Tuxill JD Flores TC Runk JLV (1999)Covariance and decoupling of floral and vegetative traits in nineneotropical plants a re-evaluation of Bergrsquos correlation-pleiadesconcept American Journal of Botany 86 39ndash55 doi1023072656953

Ashby E (1948) Studies in the morphogenesis of leaves I An essay on leafshape New Phytologist 47 153ndash176 doi101111j1469-81371948tb05098x

Aso K Kato M Banks JA Hasebe M (1999) Characterization ofhomeodomain-leucine zipper genes in the fern Ceratopteris richardiiand the evolution of the homeodomain-leucine zipper gene family invascular plants Molecular Biology and Evolution 16 544ndash552

Bailey IW Sinnott EW (1915) A botanical index of Cretaceous and Tertiaryclimates Science 41 831ndash834 doi101126science411066831

Bailey IW Sinnott EW (1916) The climatic distribution of certain types ofangiosperm leaves American Journal of Botany 3 24ndash39 doi1023072435109

Baker-Brosh KF Peet RK (1997) The ecological significance of lobed andtoothed leaves in temperate forest trees Ecology 78 1250ndash1255

Bakker FT Culham A Marais EM Gibby M (2005) Nested radiation inCape Pelargonium In lsquoPlant species-level systematics new perspectiveson pattern and process Vol 143rsquo (Eds FT Bakker LW Chartrou BGravendeel PB Pielser) pp 75ndash100 (ARG Ganter Verlag KG RuggellLichtstein)

Balfour IB (Transl) (1969) lsquoOrganography of plants I General organographyrsquoby K Goebel 1900 (Hafner Publishing Co Inc New York)

Ball MC Cowan IR Farquhar GD (1988) Maintenance of leaf temperatureand the optimization of carbon gain in relation to water loss in a tropicalmangrove forest Australian Journal of Plant Physiology 15 263ndash276doi101071PP9880263

The evolution of leaf shape Functional Plant Biology 547

considerations of the genus Pelargonium and the familyProteaceae and finally the case of leaf shape variation within asingle genetic individual heteroblasty and heterophylly

Leaf teeth leaf shape variation at the community scale

Leaf teeth are conspicuous features inmanyplants and are notableas a leaf-shape trait that varies in predictable ways across plantcommunities Nearly 100 years ago Bailey and Sinnott (19151916) observed that plant communities from regions with lowmean annual temperature (MAT) have a higher site meanproportion of non-monocotyledonous species with toothedleaves Paleobotanists quickly seized upon this relationship asan index for inferring paleotemperature Little et al (2010)recently documented the rich interest in this topic compiling351 literature references related to leaf physiognomy andtemperature of which the primary focus is leaf teeth The linkbetween teeth and temperature is observed on all vegetatedcontinents (Greenwood 2005a Peppe Royer et al 2011) andis also seen in related metrics such as tooth size and tooth number(Huff et al 2003 Royer et al 2005 Peppe et al 2011) Furtherthese tooth variables can display phenotypic plasticity totemperature change (Fig 6 Royer et al 2009b)

What is knownabout the functional bases of these leafndashclimaterelationships Leaf teeth commonly expand and mature fasterthan the bulk leaf (eg Billings 1905 Feild et al 2005) and youngteeth are disproportionately vigorous in their gas exchange(Baker-Brosh and Peet 1997 Royer and Wilf 2006) Fromthese observations Royer and Wilf (2006) hypothesised thatone role of teeth is to increase sap flow thereby deliveringnutrients and other solutes to young emerging leaves Inhabitats with low temperatures teeth could play an importantrole in maximising the potential for carbon gain and growth earlyin the season In warmer climates with longer growing seasonsthe carbon benefit conferred by teeth would be outweighed bytheir unavoidable water cost Alternatively Feild et al (2005)hypothesised that teeth help remove freezendashthaw embolismsthrough positive root pressure and guttation Importantly thetransport of fluids is common to both hypotheses From this itmay be surmised that water-limiting conditions select againstteeth (Bailey andSinnott 1916) Indeed this pattern is clearly seenacross local water availability gradients (Burnham et al 2001

Kowalski and Dilcher 2003 Greenwood 2005b Royer et al2009a Peppe et al 2011) but not globally against mean annualprecipitation (eg Peppe et al 2011)

If the hypothesis by Royer andWilf (2006) is correct toothedspecies should have a higher photosynthetic rate at least at thebeginning of the growing season As such teethmay be related tothe leaf economics spectrum (Wright et al 2004) Some studiesfind support for this idea leaves with a short leaf lifespan(deciduous) (Peppe et al 2011) and low leaf mass per area(Royer et al 2005) two leaf economic variables are morelikely to be toothed It also follows that teeth should be morefunctionally related to growing season climatic variables likeseasonality growing season length and growing degree days notmean annual temperature however few studies find support forthis idea (Royer et al 2005 Peppe et al 2011) Ultimately itwould be valuable to have a better understanding of the geneticprocesses underlying tooth development but presently little isknown outside model species (see lsquoGenetic regulation forpatterning of leaf margin serration and leaflet formationrsquosection) Further phylogenetic history has typically beenassumed to have only a minor role in these leaf-climatepatterns but rigorous tests demonstrate that this assumption isfalse (Little et al 2010) This outcome is of particular concern forpaleobotanists because systematic placement of fossils can bedifficult especially for the Cretaceous and Paleogene Tools toaccommodate for this phylogenetic dependency are needed

Leaf shape variation within lineages

Proteaceae

Proteaceae is an ancient angiosperm family originating inGondwana over 100million year ago and possessing 1700 extantspecies (Weston 2007) The familyrsquos greatest representation is inAustralia followed by SouthAfrica with smaller numbers acrossother Gondwanan segments (Cowling and Lamont 1998 Hootand Douglas 1998 Weston 2007) Within the Proteaceae thereexists an enormous range of leaf sizes and shapes ranging froma few mm2 to 500 cm2 and including entire toothed broadlylobed deeply dissected needle-like simple or compound leavesIntriguingly most of this shape diversity is represented inAustralia The majority of taxa outside Australia possessentire leaves the exception being four South African genera

(a) (b) (c) (d )

Fig 6 Representative leaves of Acer rubrum showing genetic variation and plasticity in leaf teeth Leaf derived from coolclimate (Ontario) seed stock grown in (a) Rhode Island (cool climate) and (b) Florida (warm climate) leaf derived fromwarm climate (Florida) seed stock grown in (c) Rhode Island and (d) Florida Bar (for all leaves) = 1 cm Reprinted from Royeret al (2009) under Creative Commons Attribution Licence

544 Functional Plant Biology A B Nicotra et al

that possess dissected (Serruria and Paranomis) or toothed(Mimites and Leucospermum) leafed species all of whichoccur in a single clade (Sauquet et al 2009)

The reason for the extensive radiation in leaf shape diversityin Australian Proteaceae is unclear Among South AfricanProteaceae leaf size decreases with mean annual temperatureand narrow leaves have been shown to overheat less than theirwider counterparts (Yates et al 2010) Given that leaf dissectionand small size are functionally equivalent with respect to heatloss (see lsquoTemperature and waterrsquo section) it is tempting tosuggest that leaf shape variation in Australia represents anadaptive response to increasing hot dry conditions Yet thereappears to be no consistent distributional pattern of Proteaceaeleaf dimensions inAustraliaManymediumor large un-dissectedleaves exist in the hottest habitats Grevilleas with small simpleleaves can be found across most regions including high rainfallareas and some of the species found in the shaded rainforestunderstorey have among the most deeply dissected leaves of allLeaf shape also varies within a single individual For examplefern-leaf stenocarpus Stenocarpus davallioides Foreman ampBHyland has tripinnatisect juvenile leaves but adult leavescan be pinnate or bipinnate (Foreman and Hyland 1995)Other rainforest species such as Atherton Oak (Athertoniadiversifolia (CTWhit) LAS Johnson amp BGBriggs) havevery large lobed intermediate leaves but adult and seedlingleaves are small and entire (Weston 1995) Alternatively leaflobing within a plant can be random having no obvious spatialor developmental association or homogenous across a speciesExamining a phylogeny of extant taxa one finds lobed ordissected leaves alongside entire-leafed species ndash across mostlineages in the group (Weston 2007) Therefore althoughsuggestions of microclimatic adaptation may be offered atthe species level it seems unlikely that an environmentalassociation with leaf shape could be generalised acrossAustralian Proteaceae Rather leaf shape variation in theProteaceae is likely to reflect the range of ecological solutionsto leaf structurefunction trade-offs discussed above

Pelargonium

Similar to the Proteaceae the genus Pelargonium isdistributed predominantly in the southern hemisphere withnearly 80 of its almost 300 species found in South AfricaAlthough the age of the genus is estimated to be around30million years (Bakker et al 2005) the highest rate ofspecies accumulation has occurred in the last 10million years(Martinez-Cabrera 2010) The combination of climatic andedaphic heterogeneity over short distances and limited geneflow (Latimer et al 2005 Linder 2005) is frequently offeredas an explanation for the extraordinary plant species diversity thatcharacterises southern Africa Pelargonium the third largestgenus in the region is characterised by remarkable variation ingrowth form that ranges from annuals and geophytes to shrubsVariation in leaf shape is no less dramatic ranging from entireto highly dissected (Fig 1) often within growth forms andtaxonomic sections Parsing of leaf form categorically revealedthat the extent of dissection of the blade is highly evolvablewithinanoverall ovate leaf shapeoutline that is evolutionarily conserved(Jones et al 2009)

In addition to high evolutionary lability of leaf dissectionthere is variation among major clades in the evolutionary patternof major veins One major clade (Clade C) is dominated bypalmate venation whereas another (Clade A) is dominated bypinnate venation although both types are found in both cladesWe note that pinnate venation is significantly associated withincreased dissection and reductions in functional leaf sizecharacteristic of the highly diverse lsquoxerophyticrsquo clade (CladeA2) of the winter rainfall region

To examine whether leaf shape translates into functionalvariation carbon gain and water use efficiency were measuredin pairs of species contrasted for leaf shape and replicated acrossthree subclades of Clade A and one subclade of the third majorclade CladeB (Nicotra et al 2008) In each case the species withthemore dissected leaves showed higher rates of carbon gain butalso higher rates of water loss in all treatment conditions Thatpaper argued that the differences in carbon gain were not directresults of leaf shape per se but rather that the same conditionsthat result in selection for more dissected leaves also favourthe evolution of high photosynthetic rates high leaf nitrogencontent and opportunistic use of water when available Evenin this study however there was no correlation between leafdissection and the climate present in native ranges suggestingeither that climate is selective at a much smaller scale (ie themicrohabitat) or that for any given set of climate variablesmultiple leaf shapes can be adaptive in association withvariation in other leaf traits

Heteroblasty and heterophylly variationwithin a single genetic individual

We have so far focussed primarily on cross species variation inleaf shape As discussed for the Proteaceae genera above (lsquoTheProteaceaersquo section) however a single genome can produce aremarkable range of leaf shapes during the lifetime of a plantThere are two broad categories of leaf shape variation that can beexpressed by a single individual This variation can arise either inresponse to different external environments or to ontogeneticsignals (some of which may be influenced by environment aswell)

Environmentally induced heterophylly

Environmentally induced heterophylly results when leafshape responds to environmental cues The classic case is thedramatically different emergent and submergent leaves in aquaticplants Although many aquatic plants are also heteroblastic (seebelow Sculthorpe 1967) environmentally induced heterophyllyis not specific to position or plant age and is reversible dependinglargely on the environmental signals (mediated by hormonalsignals) perceived by cells of the developing leaf primordiumThe developmental age of the leaf at the time of signal perceptionis critical Very young primordia (L1ndashL3) generally respondcompletely to the novel environment whereas primordiathat were slightly older at the time of environmental changeshow intermediate features as mature leaves Characteristics ofspecific cells are related to the position of that cell within thedifferentiation gradient characteristic of that species at the timeof environmental change (eg Goliber and Feldman 1990 Bruniet al 1996 Kuwabara and Nagata 2006)

The evolution of leaf shape Functional Plant Biology 545

Among terrestrial plants the lsquosunndashshadersquo leaf response isfrequently cited as the classic example of environmentallyinduced heterophylly Historically differences in leaf formwithin a canopy have been assumed to be responses to alteredlight environments (eg larger and less-lobed leaves are producedin shade)However these different leaf shapes also confer distincthydraulic advantages (see lsquoTemperature and waterrsquo section)Whether in response to light or to hydraulic limitation lsquosunshadersquo heterophylly is accompanied by similar and well-knownanatomical differences across a wide range of species (ie sunleaves are generally smaller and thicker with smaller cells thanshade leaves Taiz and Zeiger 2002) However when theseanatomical differences become determined developmentally(Dengler 1980 Yano and Terashima 2004) has been lesscommonly studied The presence of intermediate anatomicalcharacteristics in leaves transferred among light environmentsshortly after expansion ceased suggests that leaves of somespecies have the capacity to respond anatomically to alteredlight levels until very late in development (Oguchi et al2005) In contrast recent studies suggest that systemic lightand humidity signals sent from older to younger leavesallow younger leaves to develop phenotypes suited for theiranticipated environments (Lake et al 2001 Yano andTerashima 2001 Thomas et al 2004) Exactly when final leafshape is determined in lsquosunshadersquo leaves remains an openquestion for most species

Heteroblasty describes ontogenetically dependentchanges in leaf features most notably shape

In 1900 Goebel originally distinguished differences in lsquojuvenileformrsquo from lsquoadult formrsquo on the basis of dramatic differences inleaf shape and associated traits (Balfour 1969) As an example he

described the shift from compound leaves to phyllodes in Acaciaas heteroblastic and reserved the term homoblastic for the lessdramatic variation exhibited by Casurina It is now generallyaccepted that heteroblasty refers to the full suite of features thatchange along the shoot during development (Jones 1999)

Heteroblasty results from the interaction of developmentalprograms expressed at both the leaf and the shoot- or whole-plant level (Kerstetter and Poethig 1998 Tsukaya et al 2000Usami et al 2009 Wu et al 2009 Chuck and OrsquoConnor 2010)Heteroblasty is generally not reversible unless reiterative growthoccurs (ie following damage) Morphological differencescharacteristic of heteroblasty such as differences in lobing areusually expressed relatively early in development shortly afterleaf initiation (Jones 1995)

The sequence of heteroblastic leaf shapes produced alongthe shoot can be accelerated or delayed in response to differentgrowth environments or architectural manipulations (Ashby1948 Forster and Bonser 2009 Jaya et al 2010)Furthermore both heteroblasty and plasticity in heteroblastyhave been shown to be under strong genetic control (Anderson1989 Climent et al 2006) Forster et al (2011) suggested thatboth heteroblasty and plasticity in heteroblasty are likely to occurin response to lsquowhich leaf type confers the greatest benefit fora given light levelrsquo Several studies over the last decade havepointed towards functional consequences of heteroblastygenerally with reference to single environmental variables (egHansen 1986 Gamage and Jesson 2007 Kubien et al 2007)It is likely that heteroblasty influences multiple leaf functionsbecause suites of leaf traits change with heteroblasty (Jones2001) Indeed this conclusion was supported in a recent studythat examined differences inmore than a hundred leaf traits in koa(Acacia koa AGray Pasquet-Kok et al 2010) Heteroblasty hasalso been proposed as an adaptation to herbivory (Karban and

Table 1 Promising directions for cross-disciplinary investigation of leaf shape

No Direction

1 A multi-species lsquoevo-devorsquo approach similar to that by Illing et al (2009) is likely to be informative with regard to many enigmatic features regardingthe evolution of diverse leaf forms and more generally provides an exciting system for examining the convergence and divergence in the evolutionof function

2 A phylogenetic analysis of leaf shape examining correlated evolution with other traits such as venation size leaf surface characteristics and branch andarchitectural measures would enable a thorough investigation of the situations in which leaf shape diversification arises

3 Further research into both the genetic and developmental processes underlying lobes and teeth and the ecological correlates of each would provide anexcellent opportunity to link across wide environmental scales

4 Cross species studies ideally in the field of the relative importance of leaf shape among other factors in determining leaf temperature and water status arenecessary to test the conclusion put forward here that the functional significance of leaf shape lies more often in water relations than directly in leafthermoregulation

5 Studies of the scale of environmental change that selects for ontogenetic response in leaf shape in heteroblastic species fromdiverse environments will shedlight on links between genes development and ecology

6 Detailed quantifications of the extent towhich leaf shape differswithin canopies particularly of dioecious specieswhere selection on reproductive functioncan lead to dimorphic conditions will help link leaf shape to whole plant processes

7 Studies at the scale of transects across the leaf surface that simultaneously consider hydraulic resistance temperature and light will reveal the extent oflocalised hydraulic and thermal stresses within leaf surfaces that might act as selective forces on leaf shape per se Imaging approaches incorporatingmeasures of 3-dimensional shape temperature and fluorescence provide exciting opportunities for examining spatial heterogeneity in temperature andphysiological processes

8 Since patterns of leaf venation affect both hydraulic supply and capacity for photosynthate transport studies that simultaneously examine both willreveal the extent to which interactions among these fundamental processes either co-determine leaf shape or establish trade-offs ultimately reflectedin leaf shape

9 Finally studies examining the fitness values of different leaf shapes or plasticity in leaf shapes under a range of environmental conditions are needed tobetter identify which aspects of leaf shape are most likely to be adaptive

546 Functional Plant Biology A B Nicotra et al

Thaler 1999 Brennan et al 2001 Fadzly et al 2009) and it hasbeen associated with differences in frost resistance (Darrow et al2001) It is worth noting that studies of the functional significanceof heteroblasty have focussed on relatively few plant lineagesGiven that heteroblasty occurs widely among the angiospermsas well as in some ferns and gymosperms (eg Wagner 1952Mueller 1982 Pryer and Hearn 2009) the next questions mayfocus on the cost of heteroblasty and its fitness consequenceswhat scale of environmental change selects for ontogeneticresponse in leaf shape

Synthesis and next steps

Whether considered at community lineage or single genomescales leaf shape is a trait that is highly labile and responsive to arange of biotic and abiotic factors Although some leaf traitscan be explained in terms of particularly strong trade-offsbetween two or more factors that result in clear scalingrelationships (eg leaf mass-to-area traits or leaf mass-to-stemmass) leaf shape in contrast does not emerge from the aboveconsideration as a factor capable of explaining ecologicaldifferentiation between species Rather leaf shape emerges asa trait for which there are many quite varied functional trade-offs

As stated above these trade-offs for successful functionalityof leaves may generate ecological opportunities leaf shape isperhaps best viewed not as a single major axis but rather asan option that fine tunes the leaf to its conditions over bothshort and evolutionary time spans Numerous elements of thisreview support this conclusion Over evolutionary history theextant range of variation in leaf shape has been explored bymanylineages although only the angiosperms have explored thisrange with the added element of massive increases in veindensities Likewise our growing understanding of geneticcontrols on leaf shape reveal that it is a complex traitdetermined by many different genes at several differentdevelopmental stages Although some genes have conservedaffects on leaf morphology including shape the exact natureof their affects can vary widely Likewise the function to whichagivenelement of leaf shape lsquois putrsquo canvary For example closerinvestigation of the internal architecture and developmentalpathways associated with different leaf shapes suggests thatleaf teeth and leaf lobes often have distinctly differentfunctional foundations and may play different functions indifferent environments

Despite amyriadof functional roles the one recurrent theme inour review is that leaf shape is affected by and strongly influencesleaf water relations From a functional perspective the role of leafshape in leaf thermoregulation has perhaps received the mostattention in previous considerations of the functional significanceof leaf shape The relationship between leaf size and temperaturebeing physically determined by basic energy balance principlesthis was a logical starting point of enquiry But as we haveshown the connection between leaf shape (and size) and leaftemperatures under field conditions is not well supportedempirically Rather leaf shape variation as distinct from sizemore often will reflect water supply trade-offs High veindensities support high photosynthetic rates (Brodribb et al2010) and dissected leaves maintain a greater proportion ofleaf tissue closer to main veins Leaf temperature plays a role

in this story as higher leaf temperatures lead to greaterevaporative water loss but many factors other than leaf shapeinfluence leaf temperature (eg leaf thickness orientation andreflectivity and branch and plant architecture) Finally it is alsoworth noting that although there appears to be an obviouslink between shape and water supply there is a concomitantinteraction between water supply and phloem function thatremains relatively unexplored

Thus our review suggests several angles for further research(Table 1) In particular we advocate those approaches thatenable investigation across the scales examined here (genedevelopment function and evolutionary ecology) and thatfocus in particular on the links among leaf shape developmentand water relations

Acknowledgements

The authors thank the Botanical Society of America for their support of thesymposiumfromwhich this reviewaroseWealso thankLawrenSackand twoanonymous reviewers for useful comments on an earlier version of thismanuscript

References

Ackerly DD Knight CA Weiss SB Barton K Starmer KP (2002) Leafsize specific leaf area and microhabitat distribution of chaparral woodyplants contrasting patterns in species level and community level analysesOecologia 130 449ndash457 doi101007s004420100805

Anderson S (1989) Variation in heteroblastic succession among populationsof Crepis tectorum Nordic Journal of Botany 8 565ndash573 doi101111j1756-10511989tb01730x

Armbruster WS Muchhala N (2009) Associations between floralspecialization and species diversity cause effect or correlationEvolutionary Ecology 23 159ndash179 doi101007s10682-008-9259-z

Armbruster WS Di Stilio VS Tuxill JD Flores TC Runk JLV (1999)Covariance and decoupling of floral and vegetative traits in nineneotropical plants a re-evaluation of Bergrsquos correlation-pleiadesconcept American Journal of Botany 86 39ndash55 doi1023072656953

Ashby E (1948) Studies in the morphogenesis of leaves I An essay on leafshape New Phytologist 47 153ndash176 doi101111j1469-81371948tb05098x

Aso K Kato M Banks JA Hasebe M (1999) Characterization ofhomeodomain-leucine zipper genes in the fern Ceratopteris richardiiand the evolution of the homeodomain-leucine zipper gene family invascular plants Molecular Biology and Evolution 16 544ndash552

Bailey IW Sinnott EW (1915) A botanical index of Cretaceous and Tertiaryclimates Science 41 831ndash834 doi101126science411066831

Bailey IW Sinnott EW (1916) The climatic distribution of certain types ofangiosperm leaves American Journal of Botany 3 24ndash39 doi1023072435109

Baker-Brosh KF Peet RK (1997) The ecological significance of lobed andtoothed leaves in temperate forest trees Ecology 78 1250ndash1255

Bakker FT Culham A Marais EM Gibby M (2005) Nested radiation inCape Pelargonium In lsquoPlant species-level systematics new perspectiveson pattern and process Vol 143rsquo (Eds FT Bakker LW Chartrou BGravendeel PB Pielser) pp 75ndash100 (ARG Ganter Verlag KG RuggellLichtstein)

Balfour IB (Transl) (1969) lsquoOrganography of plants I General organographyrsquoby K Goebel 1900 (Hafner Publishing Co Inc New York)

Ball MC Cowan IR Farquhar GD (1988) Maintenance of leaf temperatureand the optimization of carbon gain in relation to water loss in a tropicalmangrove forest Australian Journal of Plant Physiology 15 263ndash276doi101071PP9880263

The evolution of leaf shape Functional Plant Biology 547

that possess dissected (Serruria and Paranomis) or toothed(Mimites and Leucospermum) leafed species all of whichoccur in a single clade (Sauquet et al 2009)

The reason for the extensive radiation in leaf shape diversityin Australian Proteaceae is unclear Among South AfricanProteaceae leaf size decreases with mean annual temperatureand narrow leaves have been shown to overheat less than theirwider counterparts (Yates et al 2010) Given that leaf dissectionand small size are functionally equivalent with respect to heatloss (see lsquoTemperature and waterrsquo section) it is tempting tosuggest that leaf shape variation in Australia represents anadaptive response to increasing hot dry conditions Yet thereappears to be no consistent distributional pattern of Proteaceaeleaf dimensions inAustraliaManymediumor large un-dissectedleaves exist in the hottest habitats Grevilleas with small simpleleaves can be found across most regions including high rainfallareas and some of the species found in the shaded rainforestunderstorey have among the most deeply dissected leaves of allLeaf shape also varies within a single individual For examplefern-leaf stenocarpus Stenocarpus davallioides Foreman ampBHyland has tripinnatisect juvenile leaves but adult leavescan be pinnate or bipinnate (Foreman and Hyland 1995)Other rainforest species such as Atherton Oak (Athertoniadiversifolia (CTWhit) LAS Johnson amp BGBriggs) havevery large lobed intermediate leaves but adult and seedlingleaves are small and entire (Weston 1995) Alternatively leaflobing within a plant can be random having no obvious spatialor developmental association or homogenous across a speciesExamining a phylogeny of extant taxa one finds lobed ordissected leaves alongside entire-leafed species ndash across mostlineages in the group (Weston 2007) Therefore althoughsuggestions of microclimatic adaptation may be offered atthe species level it seems unlikely that an environmentalassociation with leaf shape could be generalised acrossAustralian Proteaceae Rather leaf shape variation in theProteaceae is likely to reflect the range of ecological solutionsto leaf structurefunction trade-offs discussed above

Pelargonium

Similar to the Proteaceae the genus Pelargonium isdistributed predominantly in the southern hemisphere withnearly 80 of its almost 300 species found in South AfricaAlthough the age of the genus is estimated to be around30million years (Bakker et al 2005) the highest rate ofspecies accumulation has occurred in the last 10million years(Martinez-Cabrera 2010) The combination of climatic andedaphic heterogeneity over short distances and limited geneflow (Latimer et al 2005 Linder 2005) is frequently offeredas an explanation for the extraordinary plant species diversity thatcharacterises southern Africa Pelargonium the third largestgenus in the region is characterised by remarkable variation ingrowth form that ranges from annuals and geophytes to shrubsVariation in leaf shape is no less dramatic ranging from entireto highly dissected (Fig 1) often within growth forms andtaxonomic sections Parsing of leaf form categorically revealedthat the extent of dissection of the blade is highly evolvablewithinanoverall ovate leaf shapeoutline that is evolutionarily conserved(Jones et al 2009)

In addition to high evolutionary lability of leaf dissectionthere is variation among major clades in the evolutionary patternof major veins One major clade (Clade C) is dominated bypalmate venation whereas another (Clade A) is dominated bypinnate venation although both types are found in both cladesWe note that pinnate venation is significantly associated withincreased dissection and reductions in functional leaf sizecharacteristic of the highly diverse lsquoxerophyticrsquo clade (CladeA2) of the winter rainfall region

To examine whether leaf shape translates into functionalvariation carbon gain and water use efficiency were measuredin pairs of species contrasted for leaf shape and replicated acrossthree subclades of Clade A and one subclade of the third majorclade CladeB (Nicotra et al 2008) In each case the species withthemore dissected leaves showed higher rates of carbon gain butalso higher rates of water loss in all treatment conditions Thatpaper argued that the differences in carbon gain were not directresults of leaf shape per se but rather that the same conditionsthat result in selection for more dissected leaves also favourthe evolution of high photosynthetic rates high leaf nitrogencontent and opportunistic use of water when available Evenin this study however there was no correlation between leafdissection and the climate present in native ranges suggestingeither that climate is selective at a much smaller scale (ie themicrohabitat) or that for any given set of climate variablesmultiple leaf shapes can be adaptive in association withvariation in other leaf traits

Heteroblasty and heterophylly variationwithin a single genetic individual

We have so far focussed primarily on cross species variation inleaf shape As discussed for the Proteaceae genera above (lsquoTheProteaceaersquo section) however a single genome can produce aremarkable range of leaf shapes during the lifetime of a plantThere are two broad categories of leaf shape variation that can beexpressed by a single individual This variation can arise either inresponse to different external environments or to ontogeneticsignals (some of which may be influenced by environment aswell)

Environmentally induced heterophylly

Environmentally induced heterophylly results when leafshape responds to environmental cues The classic case is thedramatically different emergent and submergent leaves in aquaticplants Although many aquatic plants are also heteroblastic (seebelow Sculthorpe 1967) environmentally induced heterophyllyis not specific to position or plant age and is reversible dependinglargely on the environmental signals (mediated by hormonalsignals) perceived by cells of the developing leaf primordiumThe developmental age of the leaf at the time of signal perceptionis critical Very young primordia (L1ndashL3) generally respondcompletely to the novel environment whereas primordiathat were slightly older at the time of environmental changeshow intermediate features as mature leaves Characteristics ofspecific cells are related to the position of that cell within thedifferentiation gradient characteristic of that species at the timeof environmental change (eg Goliber and Feldman 1990 Bruniet al 1996 Kuwabara and Nagata 2006)

The evolution of leaf shape Functional Plant Biology 545

Among terrestrial plants the lsquosunndashshadersquo leaf response isfrequently cited as the classic example of environmentallyinduced heterophylly Historically differences in leaf formwithin a canopy have been assumed to be responses to alteredlight environments (eg larger and less-lobed leaves are producedin shade)However these different leaf shapes also confer distincthydraulic advantages (see lsquoTemperature and waterrsquo section)Whether in response to light or to hydraulic limitation lsquosunshadersquo heterophylly is accompanied by similar and well-knownanatomical differences across a wide range of species (ie sunleaves are generally smaller and thicker with smaller cells thanshade leaves Taiz and Zeiger 2002) However when theseanatomical differences become determined developmentally(Dengler 1980 Yano and Terashima 2004) has been lesscommonly studied The presence of intermediate anatomicalcharacteristics in leaves transferred among light environmentsshortly after expansion ceased suggests that leaves of somespecies have the capacity to respond anatomically to alteredlight levels until very late in development (Oguchi et al2005) In contrast recent studies suggest that systemic lightand humidity signals sent from older to younger leavesallow younger leaves to develop phenotypes suited for theiranticipated environments (Lake et al 2001 Yano andTerashima 2001 Thomas et al 2004) Exactly when final leafshape is determined in lsquosunshadersquo leaves remains an openquestion for most species

Heteroblasty describes ontogenetically dependentchanges in leaf features most notably shape

In 1900 Goebel originally distinguished differences in lsquojuvenileformrsquo from lsquoadult formrsquo on the basis of dramatic differences inleaf shape and associated traits (Balfour 1969) As an example he

described the shift from compound leaves to phyllodes in Acaciaas heteroblastic and reserved the term homoblastic for the lessdramatic variation exhibited by Casurina It is now generallyaccepted that heteroblasty refers to the full suite of features thatchange along the shoot during development (Jones 1999)

Heteroblasty results from the interaction of developmentalprograms expressed at both the leaf and the shoot- or whole-plant level (Kerstetter and Poethig 1998 Tsukaya et al 2000Usami et al 2009 Wu et al 2009 Chuck and OrsquoConnor 2010)Heteroblasty is generally not reversible unless reiterative growthoccurs (ie following damage) Morphological differencescharacteristic of heteroblasty such as differences in lobing areusually expressed relatively early in development shortly afterleaf initiation (Jones 1995)

The sequence of heteroblastic leaf shapes produced alongthe shoot can be accelerated or delayed in response to differentgrowth environments or architectural manipulations (Ashby1948 Forster and Bonser 2009 Jaya et al 2010)Furthermore both heteroblasty and plasticity in heteroblastyhave been shown to be under strong genetic control (Anderson1989 Climent et al 2006) Forster et al (2011) suggested thatboth heteroblasty and plasticity in heteroblasty are likely to occurin response to lsquowhich leaf type confers the greatest benefit fora given light levelrsquo Several studies over the last decade havepointed towards functional consequences of heteroblastygenerally with reference to single environmental variables (egHansen 1986 Gamage and Jesson 2007 Kubien et al 2007)It is likely that heteroblasty influences multiple leaf functionsbecause suites of leaf traits change with heteroblasty (Jones2001) Indeed this conclusion was supported in a recent studythat examined differences inmore than a hundred leaf traits in koa(Acacia koa AGray Pasquet-Kok et al 2010) Heteroblasty hasalso been proposed as an adaptation to herbivory (Karban and

Table 1 Promising directions for cross-disciplinary investigation of leaf shape

No Direction

1 A multi-species lsquoevo-devorsquo approach similar to that by Illing et al (2009) is likely to be informative with regard to many enigmatic features regardingthe evolution of diverse leaf forms and more generally provides an exciting system for examining the convergence and divergence in the evolutionof function

2 A phylogenetic analysis of leaf shape examining correlated evolution with other traits such as venation size leaf surface characteristics and branch andarchitectural measures would enable a thorough investigation of the situations in which leaf shape diversification arises

3 Further research into both the genetic and developmental processes underlying lobes and teeth and the ecological correlates of each would provide anexcellent opportunity to link across wide environmental scales

4 Cross species studies ideally in the field of the relative importance of leaf shape among other factors in determining leaf temperature and water status arenecessary to test the conclusion put forward here that the functional significance of leaf shape lies more often in water relations than directly in leafthermoregulation

5 Studies of the scale of environmental change that selects for ontogenetic response in leaf shape in heteroblastic species fromdiverse environments will shedlight on links between genes development and ecology

6 Detailed quantifications of the extent towhich leaf shape differswithin canopies particularly of dioecious specieswhere selection on reproductive functioncan lead to dimorphic conditions will help link leaf shape to whole plant processes

7 Studies at the scale of transects across the leaf surface that simultaneously consider hydraulic resistance temperature and light will reveal the extent oflocalised hydraulic and thermal stresses within leaf surfaces that might act as selective forces on leaf shape per se Imaging approaches incorporatingmeasures of 3-dimensional shape temperature and fluorescence provide exciting opportunities for examining spatial heterogeneity in temperature andphysiological processes

8 Since patterns of leaf venation affect both hydraulic supply and capacity for photosynthate transport studies that simultaneously examine both willreveal the extent to which interactions among these fundamental processes either co-determine leaf shape or establish trade-offs ultimately reflectedin leaf shape

9 Finally studies examining the fitness values of different leaf shapes or plasticity in leaf shapes under a range of environmental conditions are needed tobetter identify which aspects of leaf shape are most likely to be adaptive

546 Functional Plant Biology A B Nicotra et al

Thaler 1999 Brennan et al 2001 Fadzly et al 2009) and it hasbeen associated with differences in frost resistance (Darrow et al2001) It is worth noting that studies of the functional significanceof heteroblasty have focussed on relatively few plant lineagesGiven that heteroblasty occurs widely among the angiospermsas well as in some ferns and gymosperms (eg Wagner 1952Mueller 1982 Pryer and Hearn 2009) the next questions mayfocus on the cost of heteroblasty and its fitness consequenceswhat scale of environmental change selects for ontogeneticresponse in leaf shape

Synthesis and next steps

Whether considered at community lineage or single genomescales leaf shape is a trait that is highly labile and responsive to arange of biotic and abiotic factors Although some leaf traitscan be explained in terms of particularly strong trade-offsbetween two or more factors that result in clear scalingrelationships (eg leaf mass-to-area traits or leaf mass-to-stemmass) leaf shape in contrast does not emerge from the aboveconsideration as a factor capable of explaining ecologicaldifferentiation between species Rather leaf shape emerges asa trait for which there are many quite varied functional trade-offs

As stated above these trade-offs for successful functionalityof leaves may generate ecological opportunities leaf shape isperhaps best viewed not as a single major axis but rather asan option that fine tunes the leaf to its conditions over bothshort and evolutionary time spans Numerous elements of thisreview support this conclusion Over evolutionary history theextant range of variation in leaf shape has been explored bymanylineages although only the angiosperms have explored thisrange with the added element of massive increases in veindensities Likewise our growing understanding of geneticcontrols on leaf shape reveal that it is a complex traitdetermined by many different genes at several differentdevelopmental stages Although some genes have conservedaffects on leaf morphology including shape the exact natureof their affects can vary widely Likewise the function to whichagivenelement of leaf shape lsquois putrsquo canvary For example closerinvestigation of the internal architecture and developmentalpathways associated with different leaf shapes suggests thatleaf teeth and leaf lobes often have distinctly differentfunctional foundations and may play different functions indifferent environments

Despite amyriadof functional roles the one recurrent theme inour review is that leaf shape is affected by and strongly influencesleaf water relations From a functional perspective the role of leafshape in leaf thermoregulation has perhaps received the mostattention in previous considerations of the functional significanceof leaf shape The relationship between leaf size and temperaturebeing physically determined by basic energy balance principlesthis was a logical starting point of enquiry But as we haveshown the connection between leaf shape (and size) and leaftemperatures under field conditions is not well supportedempirically Rather leaf shape variation as distinct from sizemore often will reflect water supply trade-offs High veindensities support high photosynthetic rates (Brodribb et al2010) and dissected leaves maintain a greater proportion ofleaf tissue closer to main veins Leaf temperature plays a role

in this story as higher leaf temperatures lead to greaterevaporative water loss but many factors other than leaf shapeinfluence leaf temperature (eg leaf thickness orientation andreflectivity and branch and plant architecture) Finally it is alsoworth noting that although there appears to be an obviouslink between shape and water supply there is a concomitantinteraction between water supply and phloem function thatremains relatively unexplored

Thus our review suggests several angles for further research(Table 1) In particular we advocate those approaches thatenable investigation across the scales examined here (genedevelopment function and evolutionary ecology) and thatfocus in particular on the links among leaf shape developmentand water relations

Acknowledgements

The authors thank the Botanical Society of America for their support of thesymposiumfromwhich this reviewaroseWealso thankLawrenSackand twoanonymous reviewers for useful comments on an earlier version of thismanuscript

References

Ackerly DD Knight CA Weiss SB Barton K Starmer KP (2002) Leafsize specific leaf area and microhabitat distribution of chaparral woodyplants contrasting patterns in species level and community level analysesOecologia 130 449ndash457 doi101007s004420100805

Anderson S (1989) Variation in heteroblastic succession among populationsof Crepis tectorum Nordic Journal of Botany 8 565ndash573 doi101111j1756-10511989tb01730x

Armbruster WS Muchhala N (2009) Associations between floralspecialization and species diversity cause effect or correlationEvolutionary Ecology 23 159ndash179 doi101007s10682-008-9259-z

Armbruster WS Di Stilio VS Tuxill JD Flores TC Runk JLV (1999)Covariance and decoupling of floral and vegetative traits in nineneotropical plants a re-evaluation of Bergrsquos correlation-pleiadesconcept American Journal of Botany 86 39ndash55 doi1023072656953

Ashby E (1948) Studies in the morphogenesis of leaves I An essay on leafshape New Phytologist 47 153ndash176 doi101111j1469-81371948tb05098x

Aso K Kato M Banks JA Hasebe M (1999) Characterization ofhomeodomain-leucine zipper genes in the fern Ceratopteris richardiiand the evolution of the homeodomain-leucine zipper gene family invascular plants Molecular Biology and Evolution 16 544ndash552

Bailey IW Sinnott EW (1915) A botanical index of Cretaceous and Tertiaryclimates Science 41 831ndash834 doi101126science411066831

Bailey IW Sinnott EW (1916) The climatic distribution of certain types ofangiosperm leaves American Journal of Botany 3 24ndash39 doi1023072435109

Baker-Brosh KF Peet RK (1997) The ecological significance of lobed andtoothed leaves in temperate forest trees Ecology 78 1250ndash1255

Bakker FT Culham A Marais EM Gibby M (2005) Nested radiation inCape Pelargonium In lsquoPlant species-level systematics new perspectiveson pattern and process Vol 143rsquo (Eds FT Bakker LW Chartrou BGravendeel PB Pielser) pp 75ndash100 (ARG Ganter Verlag KG RuggellLichtstein)

Balfour IB (Transl) (1969) lsquoOrganography of plants I General organographyrsquoby K Goebel 1900 (Hafner Publishing Co Inc New York)

Ball MC Cowan IR Farquhar GD (1988) Maintenance of leaf temperatureand the optimization of carbon gain in relation to water loss in a tropicalmangrove forest Australian Journal of Plant Physiology 15 263ndash276doi101071PP9880263

The evolution of leaf shape Functional Plant Biology 547

Among terrestrial plants the lsquosunndashshadersquo leaf response isfrequently cited as the classic example of environmentallyinduced heterophylly Historically differences in leaf formwithin a canopy have been assumed to be responses to alteredlight environments (eg larger and less-lobed leaves are producedin shade)However these different leaf shapes also confer distincthydraulic advantages (see lsquoTemperature and waterrsquo section)Whether in response to light or to hydraulic limitation lsquosunshadersquo heterophylly is accompanied by similar and well-knownanatomical differences across a wide range of species (ie sunleaves are generally smaller and thicker with smaller cells thanshade leaves Taiz and Zeiger 2002) However when theseanatomical differences become determined developmentally(Dengler 1980 Yano and Terashima 2004) has been lesscommonly studied The presence of intermediate anatomicalcharacteristics in leaves transferred among light environmentsshortly after expansion ceased suggests that leaves of somespecies have the capacity to respond anatomically to alteredlight levels until very late in development (Oguchi et al2005) In contrast recent studies suggest that systemic lightand humidity signals sent from older to younger leavesallow younger leaves to develop phenotypes suited for theiranticipated environments (Lake et al 2001 Yano andTerashima 2001 Thomas et al 2004) Exactly when final leafshape is determined in lsquosunshadersquo leaves remains an openquestion for most species

Heteroblasty describes ontogenetically dependentchanges in leaf features most notably shape

In 1900 Goebel originally distinguished differences in lsquojuvenileformrsquo from lsquoadult formrsquo on the basis of dramatic differences inleaf shape and associated traits (Balfour 1969) As an example he

described the shift from compound leaves to phyllodes in Acaciaas heteroblastic and reserved the term homoblastic for the lessdramatic variation exhibited by Casurina It is now generallyaccepted that heteroblasty refers to the full suite of features thatchange along the shoot during development (Jones 1999)

Heteroblasty results from the interaction of developmentalprograms expressed at both the leaf and the shoot- or whole-plant level (Kerstetter and Poethig 1998 Tsukaya et al 2000Usami et al 2009 Wu et al 2009 Chuck and OrsquoConnor 2010)Heteroblasty is generally not reversible unless reiterative growthoccurs (ie following damage) Morphological differencescharacteristic of heteroblasty such as differences in lobing areusually expressed relatively early in development shortly afterleaf initiation (Jones 1995)

The sequence of heteroblastic leaf shapes produced alongthe shoot can be accelerated or delayed in response to differentgrowth environments or architectural manipulations (Ashby1948 Forster and Bonser 2009 Jaya et al 2010)Furthermore both heteroblasty and plasticity in heteroblastyhave been shown to be under strong genetic control (Anderson1989 Climent et al 2006) Forster et al (2011) suggested thatboth heteroblasty and plasticity in heteroblasty are likely to occurin response to lsquowhich leaf type confers the greatest benefit fora given light levelrsquo Several studies over the last decade havepointed towards functional consequences of heteroblastygenerally with reference to single environmental variables (egHansen 1986 Gamage and Jesson 2007 Kubien et al 2007)It is likely that heteroblasty influences multiple leaf functionsbecause suites of leaf traits change with heteroblasty (Jones2001) Indeed this conclusion was supported in a recent studythat examined differences inmore than a hundred leaf traits in koa(Acacia koa AGray Pasquet-Kok et al 2010) Heteroblasty hasalso been proposed as an adaptation to herbivory (Karban and

Table 1 Promising directions for cross-disciplinary investigation of leaf shape

No Direction

1 A multi-species lsquoevo-devorsquo approach similar to that by Illing et al (2009) is likely to be informative with regard to many enigmatic features regardingthe evolution of diverse leaf forms and more generally provides an exciting system for examining the convergence and divergence in the evolutionof function

2 A phylogenetic analysis of leaf shape examining correlated evolution with other traits such as venation size leaf surface characteristics and branch andarchitectural measures would enable a thorough investigation of the situations in which leaf shape diversification arises

3 Further research into both the genetic and developmental processes underlying lobes and teeth and the ecological correlates of each would provide anexcellent opportunity to link across wide environmental scales

4 Cross species studies ideally in the field of the relative importance of leaf shape among other factors in determining leaf temperature and water status arenecessary to test the conclusion put forward here that the functional significance of leaf shape lies more often in water relations than directly in leafthermoregulation

5 Studies of the scale of environmental change that selects for ontogenetic response in leaf shape in heteroblastic species fromdiverse environments will shedlight on links between genes development and ecology

6 Detailed quantifications of the extent towhich leaf shape differswithin canopies particularly of dioecious specieswhere selection on reproductive functioncan lead to dimorphic conditions will help link leaf shape to whole plant processes

7 Studies at the scale of transects across the leaf surface that simultaneously consider hydraulic resistance temperature and light will reveal the extent oflocalised hydraulic and thermal stresses within leaf surfaces that might act as selective forces on leaf shape per se Imaging approaches incorporatingmeasures of 3-dimensional shape temperature and fluorescence provide exciting opportunities for examining spatial heterogeneity in temperature andphysiological processes

8 Since patterns of leaf venation affect both hydraulic supply and capacity for photosynthate transport studies that simultaneously examine both willreveal the extent to which interactions among these fundamental processes either co-determine leaf shape or establish trade-offs ultimately reflectedin leaf shape

9 Finally studies examining the fitness values of different leaf shapes or plasticity in leaf shapes under a range of environmental conditions are needed tobetter identify which aspects of leaf shape are most likely to be adaptive

546 Functional Plant Biology A B Nicotra et al

Thaler 1999 Brennan et al 2001 Fadzly et al 2009) and it hasbeen associated with differences in frost resistance (Darrow et al2001) It is worth noting that studies of the functional significanceof heteroblasty have focussed on relatively few plant lineagesGiven that heteroblasty occurs widely among the angiospermsas well as in some ferns and gymosperms (eg Wagner 1952Mueller 1982 Pryer and Hearn 2009) the next questions mayfocus on the cost of heteroblasty and its fitness consequenceswhat scale of environmental change selects for ontogeneticresponse in leaf shape

Synthesis and next steps

Whether considered at community lineage or single genomescales leaf shape is a trait that is highly labile and responsive to arange of biotic and abiotic factors Although some leaf traitscan be explained in terms of particularly strong trade-offsbetween two or more factors that result in clear scalingrelationships (eg leaf mass-to-area traits or leaf mass-to-stemmass) leaf shape in contrast does not emerge from the aboveconsideration as a factor capable of explaining ecologicaldifferentiation between species Rather leaf shape emerges asa trait for which there are many quite varied functional trade-offs

As stated above these trade-offs for successful functionalityof leaves may generate ecological opportunities leaf shape isperhaps best viewed not as a single major axis but rather asan option that fine tunes the leaf to its conditions over bothshort and evolutionary time spans Numerous elements of thisreview support this conclusion Over evolutionary history theextant range of variation in leaf shape has been explored bymanylineages although only the angiosperms have explored thisrange with the added element of massive increases in veindensities Likewise our growing understanding of geneticcontrols on leaf shape reveal that it is a complex traitdetermined by many different genes at several differentdevelopmental stages Although some genes have conservedaffects on leaf morphology including shape the exact natureof their affects can vary widely Likewise the function to whichagivenelement of leaf shape lsquois putrsquo canvary For example closerinvestigation of the internal architecture and developmentalpathways associated with different leaf shapes suggests thatleaf teeth and leaf lobes often have distinctly differentfunctional foundations and may play different functions indifferent environments

Despite amyriadof functional roles the one recurrent theme inour review is that leaf shape is affected by and strongly influencesleaf water relations From a functional perspective the role of leafshape in leaf thermoregulation has perhaps received the mostattention in previous considerations of the functional significanceof leaf shape The relationship between leaf size and temperaturebeing physically determined by basic energy balance principlesthis was a logical starting point of enquiry But as we haveshown the connection between leaf shape (and size) and leaftemperatures under field conditions is not well supportedempirically Rather leaf shape variation as distinct from sizemore often will reflect water supply trade-offs High veindensities support high photosynthetic rates (Brodribb et al2010) and dissected leaves maintain a greater proportion ofleaf tissue closer to main veins Leaf temperature plays a role

in this story as higher leaf temperatures lead to greaterevaporative water loss but many factors other than leaf shapeinfluence leaf temperature (eg leaf thickness orientation andreflectivity and branch and plant architecture) Finally it is alsoworth noting that although there appears to be an obviouslink between shape and water supply there is a concomitantinteraction between water supply and phloem function thatremains relatively unexplored

Thus our review suggests several angles for further research(Table 1) In particular we advocate those approaches thatenable investigation across the scales examined here (genedevelopment function and evolutionary ecology) and thatfocus in particular on the links among leaf shape developmentand water relations

Acknowledgements

The authors thank the Botanical Society of America for their support of thesymposiumfromwhich this reviewaroseWealso thankLawrenSackand twoanonymous reviewers for useful comments on an earlier version of thismanuscript

References

Ackerly DD Knight CA Weiss SB Barton K Starmer KP (2002) Leafsize specific leaf area and microhabitat distribution of chaparral woodyplants contrasting patterns in species level and community level analysesOecologia 130 449ndash457 doi101007s004420100805

Anderson S (1989) Variation in heteroblastic succession among populationsof Crepis tectorum Nordic Journal of Botany 8 565ndash573 doi101111j1756-10511989tb01730x

Armbruster WS Muchhala N (2009) Associations between floralspecialization and species diversity cause effect or correlationEvolutionary Ecology 23 159ndash179 doi101007s10682-008-9259-z

Armbruster WS Di Stilio VS Tuxill JD Flores TC Runk JLV (1999)Covariance and decoupling of floral and vegetative traits in nineneotropical plants a re-evaluation of Bergrsquos correlation-pleiadesconcept American Journal of Botany 86 39ndash55 doi1023072656953

Ashby E (1948) Studies in the morphogenesis of leaves I An essay on leafshape New Phytologist 47 153ndash176 doi101111j1469-81371948tb05098x

Aso K Kato M Banks JA Hasebe M (1999) Characterization ofhomeodomain-leucine zipper genes in the fern Ceratopteris richardiiand the evolution of the homeodomain-leucine zipper gene family invascular plants Molecular Biology and Evolution 16 544ndash552

Bailey IW Sinnott EW (1915) A botanical index of Cretaceous and Tertiaryclimates Science 41 831ndash834 doi101126science411066831

Bailey IW Sinnott EW (1916) The climatic distribution of certain types ofangiosperm leaves American Journal of Botany 3 24ndash39 doi1023072435109

Baker-Brosh KF Peet RK (1997) The ecological significance of lobed andtoothed leaves in temperate forest trees Ecology 78 1250ndash1255

Bakker FT Culham A Marais EM Gibby M (2005) Nested radiation inCape Pelargonium In lsquoPlant species-level systematics new perspectiveson pattern and process Vol 143rsquo (Eds FT Bakker LW Chartrou BGravendeel PB Pielser) pp 75ndash100 (ARG Ganter Verlag KG RuggellLichtstein)

Balfour IB (Transl) (1969) lsquoOrganography of plants I General organographyrsquoby K Goebel 1900 (Hafner Publishing Co Inc New York)

Ball MC Cowan IR Farquhar GD (1988) Maintenance of leaf temperatureand the optimization of carbon gain in relation to water loss in a tropicalmangrove forest Australian Journal of Plant Physiology 15 263ndash276doi101071PP9880263

The evolution of leaf shape Functional Plant Biology 547

Thaler 1999 Brennan et al 2001 Fadzly et al 2009) and it hasbeen associated with differences in frost resistance (Darrow et al2001) It is worth noting that studies of the functional significanceof heteroblasty have focussed on relatively few plant lineagesGiven that heteroblasty occurs widely among the angiospermsas well as in some ferns and gymosperms (eg Wagner 1952Mueller 1982 Pryer and Hearn 2009) the next questions mayfocus on the cost of heteroblasty and its fitness consequenceswhat scale of environmental change selects for ontogeneticresponse in leaf shape

Synthesis and next steps

Whether considered at community lineage or single genomescales leaf shape is a trait that is highly labile and responsive to arange of biotic and abiotic factors Although some leaf traitscan be explained in terms of particularly strong trade-offsbetween two or more factors that result in clear scalingrelationships (eg leaf mass-to-area traits or leaf mass-to-stemmass) leaf shape in contrast does not emerge from the aboveconsideration as a factor capable of explaining ecologicaldifferentiation between species Rather leaf shape emerges asa trait for which there are many quite varied functional trade-offs

As stated above these trade-offs for successful functionalityof leaves may generate ecological opportunities leaf shape isperhaps best viewed not as a single major axis but rather asan option that fine tunes the leaf to its conditions over bothshort and evolutionary time spans Numerous elements of thisreview support this conclusion Over evolutionary history theextant range of variation in leaf shape has been explored bymanylineages although only the angiosperms have explored thisrange with the added element of massive increases in veindensities Likewise our growing understanding of geneticcontrols on leaf shape reveal that it is a complex traitdetermined by many different genes at several differentdevelopmental stages Although some genes have conservedaffects on leaf morphology including shape the exact natureof their affects can vary widely Likewise the function to whichagivenelement of leaf shape lsquois putrsquo canvary For example closerinvestigation of the internal architecture and developmentalpathways associated with different leaf shapes suggests thatleaf teeth and leaf lobes often have distinctly differentfunctional foundations and may play different functions indifferent environments

Despite amyriadof functional roles the one recurrent theme inour review is that leaf shape is affected by and strongly influencesleaf water relations From a functional perspective the role of leafshape in leaf thermoregulation has perhaps received the mostattention in previous considerations of the functional significanceof leaf shape The relationship between leaf size and temperaturebeing physically determined by basic energy balance principlesthis was a logical starting point of enquiry But as we haveshown the connection between leaf shape (and size) and leaftemperatures under field conditions is not well supportedempirically Rather leaf shape variation as distinct from sizemore often will reflect water supply trade-offs High veindensities support high photosynthetic rates (Brodribb et al2010) and dissected leaves maintain a greater proportion ofleaf tissue closer to main veins Leaf temperature plays a role

in this story as higher leaf temperatures lead to greaterevaporative water loss but many factors other than leaf shapeinfluence leaf temperature (eg leaf thickness orientation andreflectivity and branch and plant architecture) Finally it is alsoworth noting that although there appears to be an obviouslink between shape and water supply there is a concomitantinteraction between water supply and phloem function thatremains relatively unexplored

Thus our review suggests several angles for further research(Table 1) In particular we advocate those approaches thatenable investigation across the scales examined here (genedevelopment function and evolutionary ecology) and thatfocus in particular on the links among leaf shape developmentand water relations

Acknowledgements

The authors thank the Botanical Society of America for their support of thesymposiumfromwhich this reviewaroseWealso thankLawrenSackand twoanonymous reviewers for useful comments on an earlier version of thismanuscript

References

Ackerly DD Knight CA Weiss SB Barton K Starmer KP (2002) Leafsize specific leaf area and microhabitat distribution of chaparral woodyplants contrasting patterns in species level and community level analysesOecologia 130 449ndash457 doi101007s004420100805

Anderson S (1989) Variation in heteroblastic succession among populationsof Crepis tectorum Nordic Journal of Botany 8 565ndash573 doi101111j1756-10511989tb01730x

Armbruster WS Muchhala N (2009) Associations between floralspecialization and species diversity cause effect or correlationEvolutionary Ecology 23 159ndash179 doi101007s10682-008-9259-z

Armbruster WS Di Stilio VS Tuxill JD Flores TC Runk JLV (1999)Covariance and decoupling of floral and vegetative traits in nineneotropical plants a re-evaluation of Bergrsquos correlation-pleiadesconcept American Journal of Botany 86 39ndash55 doi1023072656953

Ashby E (1948) Studies in the morphogenesis of leaves I An essay on leafshape New Phytologist 47 153ndash176 doi101111j1469-81371948tb05098x

Aso K Kato M Banks JA Hasebe M (1999) Characterization ofhomeodomain-leucine zipper genes in the fern Ceratopteris richardiiand the evolution of the homeodomain-leucine zipper gene family invascular plants Molecular Biology and Evolution 16 544ndash552

Bailey IW Sinnott EW (1915) A botanical index of Cretaceous and Tertiaryclimates Science 41 831ndash834 doi101126science411066831

Bailey IW Sinnott EW (1916) The climatic distribution of certain types ofangiosperm leaves American Journal of Botany 3 24ndash39 doi1023072435109

Baker-Brosh KF Peet RK (1997) The ecological significance of lobed andtoothed leaves in temperate forest trees Ecology 78 1250ndash1255

Bakker FT Culham A Marais EM Gibby M (2005) Nested radiation inCape Pelargonium In lsquoPlant species-level systematics new perspectiveson pattern and process Vol 143rsquo (Eds FT Bakker LW Chartrou BGravendeel PB Pielser) pp 75ndash100 (ARG Ganter Verlag KG RuggellLichtstein)

Balfour IB (Transl) (1969) lsquoOrganography of plants I General organographyrsquoby K Goebel 1900 (Hafner Publishing Co Inc New York)

Ball MC Cowan IR Farquhar GD (1988) Maintenance of leaf temperatureand the optimization of carbon gain in relation to water loss in a tropicalmangrove forest Australian Journal of Plant Physiology 15 263ndash276doi101071PP9880263

The evolution of leaf shape Functional Plant Biology 547

Ball MC Canny MJ Huang CX Heady RD (2004) Structural changes inacclimated and unacclimated leaves during freezing and thawingFunctional Plant Biology 31 29ndash40 doi101071FP03164

Ball MC Canny MJ Huang CX Egerton JJG Wolfe J (2006) Freezethaw-induced embolism depends on nadir temperature the heterogeneoushydration hypothesis Plant Cell amp Environment 29 729ndash745doi101111j1365-3040200501426x

Barkoulas M Hay A Kougioumoutzi E Tsiantis M (2008) A developmentalframework for dissected leaf formation in the Arabidopsis relativeCardamine hirsuta NatureGenetics 40 1136ndash1141 doi101038ng189

Beadle NCW (1966) Soil phosphate and its role in molding segments of theAustralian flora and vegetation with special reference to xeromorphy andsclerophylly Ecology 47 992ndash1007 doi1023071935647

BergRL (1960)The ecological significance of correlation pleiadesEvolution14 171ndash180 doi1023072405824

Bharathan G Goliber TE Moore C Kessler S Pham T Sinha NR (2002)Homologies in leaf form inferred from KNOXI gene expression duringdevelopment Science 296 1858ndash1860 doi101126science1070343

Billings FH (1905) Precursory leaf serrations of Ulmus Botanical Gazette40 224ndash225 doi101086328669

BleinTPulidoAVialette-GuiraudANikovicsKMorinHHayA JohansenIE Tsiantis M Laufs P (2008) A conserved molecular framework forcompound leaf development Science 322 1835ndash1839 doi101126science1166168

BondWJ (1989) The tortoise and the hare ecology of angiosperm dominanceand gymnosperm persistence Biological Journal of the Linnean SocietyLinnean Society of London 36 227ndash249 doi101111j1095-83121989tb00492x

Boyce CK (2005) Patterns of segregation and convergence in the evolutionof fern and seed plant leaf morphologies Paleobiology 31 117ndash140doi1016660094-8373(2005)031lt0117POSACIgt20CO2

Boyce CK (2007) Mechanisms of laminar growth in morphologicallyconvergent leaves and flower petals International Journal of PlantSciences 168 1151ndash1156 doi101086520730

Boyce CK (2008a) The fossil record of plant physiology and development ndashwhat leaves can tell us Paleontological Society Papers 14 133ndash146

Boyce CK (2008b) How green was Cooksonia The importance of size inunderstanding the early evolution of physiology in the vascular plantlineage Paleobiology 34 179ndash194 doi1016660094-8373(2008)034[0179HGWCTI]20CO2

Boyce CK (2009) Seeing the forest with the leaves ndash clues to canopyplacement from leaf fossil size and venation characteristicsGeobiology 7 192ndash199 doi101111j1472-4669200800176x

Boyce CK (2010) The evolution of plant development in a paleontologicalcontext Current Opinion in Plant Biology 13 102ndash107 doi101016jpbi200910001

Boyce CK Knoll AH (2002) Evolution of developmental potential and themultiple independent origins of leaves in Paleozoic vascular plantsPaleobiology 28 70ndash100 doi1016660094-8373(2002)028lt0070EODPATgt20CO2

Boyce CK Brodribb TJ Feild TS Zwieniecki MA (2009) Angiosperm leafvein evolution was physiologically and environmentally transformativeProceedings of the Royal Society B-Biological Sciences 276 1771ndash1776doi101098rspb20081919

Bragg JGWestobyM (2002) Leaf size and foraging for light in a sclerophyllwoodland Functional Ecology 16 633ndash639 doi101046j1365-2435200200661x

BrennanEBWeinbaumSARosenheimJAKarbanR (2001)Heteroblasty inEucalyptus globulus (Myricales Myricaceae) affects ovipositonaland settling preferences of Ctenarytaina eucalypti and C spatulata(Homoptera Psyllidae) Environmental Entomology 30 1144ndash1149doi1016030046-225X-3061144

Brodribb TJ Feild TS (2010) Leaf hydraulic evolution led a surge in leafphotosynthetic capacity during early angiosperm diversificationEcologyLetters 13 175ndash183 doi101111j1461-0248200901410x

Brodribb TJ Feild TS Jordan GJ (2007) Leaf maximum photosynthetic rateand venation are linked by hydraulicsPlant Physiology 144 1890ndash1898doi101104pp107101352

Brodribb TJ Feild TS Sack L (2010) Viewing leaf structure and evolutionfrom a hydraulic perspective Functional Plant Biology 37 488ndash498doi101071FP10010

Bruni NC Young JP Dengler NG (1996) Leaf developmental plasticity ofRanunculus flabellaris in response to terrestrial and submergedenvironments Canadian Journal of Botany 74 823ndash837 doi101139b96-103

Burnham RJ Pitman NCA Johnson KR Wilf P (2001) Habitat-relatederror in estimating temperatures from leaf margins in a humidtropical forest American Journal of Botany 88 1096ndash1102 doi1023072657093

Canales C Barkoulas M Galinha C Tsiantis M (2010) Weeds of changeCardamine hirsuta as a new model system for studying dissected leafdevelopment Journal of Plant Research 123 25ndash33 doi101007s10265-009-0263-3

Chuck G OrsquoConnor D (2010) Small RNAs going the distance during plantdevelopmentCurrentOpinion inPlantBiology13 40ndash45 doi101016jpbi200908006

Climent J Chambel MR Lopez R Mutke S Alia R Gil L (2006) Populationdivergence for heteroblasty in the Canary Island pine (Pinus canariensisPinaceae) American Journal of Botany 93 840ndash848 doi103732ajb936840

ClumHH (1926) The effect of transpiration and environmental factors on leaftemperatures 1 Transpiration American Journal of Botany 13 194ndash216doi1023072435461

CowlingRMLamontBB (1998)On the nature ofGondwanan speciesflocksdiversity of Proteaceae in Mediterranean south-western Australia andSouth Africa Australian Journal of Botany 46 335ndash355 doi101071BT97040

Cunningham SA Summerhayes B Westoby M (1999) Evolutionarydivergences in leaf structure and chemistry comparing rainfall and soilnutrient gradients Ecological Monographs 69 569ndash588 doi1018900012-9615(1999)069[0569EDILSA]20CO2

Darrow HE Bannister P Burritt DJ Jameson PE (2001) The frost resistanceof juvenile and adult forms of some heteroblastic New Zealand plantsNew Zealand Journal of Botany 39 355ndash363 doi1010800028825X20019512741

de Reuille PB Bohn-Courseau I Ljung K Morin H Carraro N Godin CTraas J (2006) Computer simulations reveal properties of the cellndashcellsignaling network at the shoot apex in Arabidopsis Proceedings of theNational Academy of Sciences of the United States of America 1031627ndash1632 doi101073pnas0510130103

Dengler N (1980) Comparative histological basis of sun and shade leafdimorphism in Helianthus annuus Canadian Journal of Botany 58717ndash730 doi101139b80-092

Dinneny JR Yadegari R Fischer RL Yanofsky MF Weigel D (2004)The role of JAGGED in shaping lateral organs Development 1311101ndash1110 doi101242dev00949

Doyle JHickeyL (1976) Pollen and leaves from themid-Cretaceous potomacgroup and their bearing on early angiosperm evolution In lsquoOrigin andearly evolution of Angiospermsrsquo (Ed C Beck) pp 139ndash206 (ColumbiaUniversity Press New York)

Drake BG Raschke K Salisbury FB (1970) Temperatures andtranspiration resistances of Xanthium leaves as affected by airtemperature humidity and wind speed Plant Physiology 46 324ndash330doi101104pp462324

548 Functional Plant Biology A B Nicotra et al

Efroni I Blum E Goldshmidt A Eshed Y (2008) A protracted and dynamicmaturation schedule underlies Arabidopsis leaf development The PlantCell 20 2293ndash2306 doi101105tpc107057521

Efroni I EshedYLifschitzE (2010)Morphogenesis of simple and compoundleaves a critical review The Plant Cell 22 1019ndash1032 doi101105tpc109073601

FadzlyN JackCSchaeferHMBurnsKC(2009)Ontogenetic colourchangesin an insular tree species signalling to extinct browsing birds NewPhytologist 184 495ndash501 doi101111j1469-8137200902926x

Feild TS Sage TL Czerniak C Iles WJD (2005) Hydathodal leaf teeth ofChloranthus japonicus (Chloranthaceae) prevent guttation-inducedflooding of the mesophyll Plant Cell amp Environment 28 1179ndash1190doi101111j1365-3040200501354x

Floyd SK Bowman JL (2006) Distinct developmental mechanisms reflectthe independent origins of leaves in vascular plants Current Biology 161911ndash1917 doi101016jcub200607067

Floyd SK Bowman JL (2010) Gene expression patterns in seed plantshoot meristems and leaves homoplasy or homology Journal ofPlant Research 123 43ndash55 doi101007s10265-009-0256-2

Fonseca CR Overton JM Collins B Westoby M (2000) Shifts in traitcombinations along rainfall and phosphorus gradients Journal ofEcology 88 964ndash977 doi101046j1365-2745200000506x

Foreman BH Hyland B (1995) Stenocarpus In lsquoFlora of Australia Vol 16rsquo(Ed P McCarthy) pp 364ndash366 (CSIRO Publishing Melbourne)

Forster MA Bonser SP (2009) Heteroblastic development and the optimalpartitioning of traits among contrasting environments in Acacia implexaAnnals of Botany 103 95ndash105 doi101093aobmcn210

Forster MA Ladd B Bonser SP (2011) Optimal allocation of resources inresponse to shading and neighbours in the heteroblastic species Acaciaimplexa Annals of Botany 103 95ndash105 doi101093aobmcn210

FrohlichMWChaseMW(2007)After a dozen years of progress the origin ofangiosperms is still a greatmysteryNature450 1184ndash1189 doi101038nature06393

Galtier J (1981) Structures foliaires de fougegraveres et Pteacuteridospermales duCarbonifegravere Infeacuterieur et leur signification eacutevolutivePalaeontographica Abt B 180 1ndash38

Gamage HK Jesson L (2007) Leaf heteroblasty is not an adaptation to shadeseedling anatomical and physiological responses to light New ZealandJournal of Ecology 31 245ndash254

Gates DM (1968) Transpiration and leaf temperatureAnnual Review of PlantPhysiology 19 211ndash238 doi101146annurevpp19060168001235

Gates DM (1980) lsquoBiophysical ecologyrsquo (Springer-Verlag New York)Gates DM Alderfer R Taylor E (1968) Leaf temperatures of desert plants

Science 159 994ndash995 doi101126science1593818994Geller GN Smith WK (1982) Influence of leaf size orientation and

arrangement on temperature and transpiration in three high-elevationlarge leafed herbs Oecologia 53 227ndash234 doi101007BF00545668

Givnish TJ (1978) Ecological aspects of plant morphology leaf form inrelation to environment Acta Biotheoretica 27 83ndash142

GivnishTJ (1979)On theadaptive significanceof leaf form In lsquoTopics inplantpopulation biologyrsquo (EdsOTSolbrig J SubodhGB Johnson PHRaven)pp 375ndash407 (Columbia University Press New York)

Goliber TE Feldman LJ (1990) Developmental analysis of leaf plasticity inthe heterophyllous aquatic plant Hippuris vulgaris American Journal ofBotany 77 399ndash412 doi1023072444726

Gottschlich DE Smith AP (1982) Convective heat transfercharacteristics of toothed leaves Oecologia 53 418ndash420 doi101007BF00389024

Grace J Fasehun FE Dixon M (1980) Boundary layer conductance of theleaves of some tropical timber trees Plant Cell amp Environment 3443ndash450

Greenwood DR (2005a) Leaf form and the reconstruction of pastclimates New Phytologist 166 355ndash357 doi101111j1469-8137200501380x

Greenwood DR (2005b) Leaf margin analysis taphonomic constraintsPalaios 20 498ndash505 doi102110palo2004P04-58

Groom PK Lamont BB Leighton S Leighton P Burrows C (2004) Heatdamage in sclerophylls is influenced by their leaf properties and plantenvironment Ecoscience 11 94ndash101

Gurevitch J Schuepp PH (1990a) Boundary layer properties of highlydissected leaves ndash an investigation using an electrochemical fluidtunnel Plant Cell amp Environment 13 783ndash792 doi101111j1365-30401990tb01094x

Gurevitch J Schuepp PH (1990b) Boundary layer properties of highlydissected leaves an investigation using an electrochemical fluid tunnelPlant Cell amp Environment 13 783ndash792 doi101111j1365-30401990tb01094x

Hansen DH (1986) Water relations of compound leaves and phyllodes inAcacia koa var latifolia Plant Cell amp Environment 9 439ndash445doi101111j1365-30401986tb01758x

Harrison CJ Corley SBMoylan EC Alexander DL Scotland RW LangdaleJA (2005) Independent recruitment of a conserved developmentalmechanism during leaf evolution Nature 434 509ndash514 doi101038nature03410

Hegazy AK El Amry MI (1998) Leaf temperature of desert sand duneplants perspectives on the adaptability of leaf morphology AfricanJournal of Ecology 36 34ndash43 doi101046j1365-20281998109-89109x

Hilton J Bateman RM (2006) Pteridosperms are the backbone of seedndashplantphylogeny Journal of the Torrey Botanical Society 133 119ndash168doi1031591095-5674(2006)133[119PATBOS]20CO2

Hofer J Turner L Hellens R Ambrose M Matthews P Michael A Ellis N(1997) UNIFOLIATA regulates leaf and flower morphogenesis in peaCurrent Biology 7 581ndash587 doi101016S0960-9822(06)00257-0

Hoot SBDouglasAW(1998) Phylogenyof the Proteaceae based on atpB andatpB-rbcL intergenic spacer region sequences Australian SystematicBotany 11 301ndash320 doi101071SB98027

Horiguchi G Kim GT Tsukaya H (2005) The transcription factor AtGRF5and the transcription coactivator AN3 regulate cell proliferation inleaf primordia of Arabidopsis thaliana The Plant Journal 43 68ndash78doi101111j1365-313X200502429x

Horiguchi G Fujikura U Ferjani A Ishikawa N Tsukaya H (2006)Large-scale histological analysis of leaf mutants using two simple leafobservation methods identification of novel genetic pathways governingthe size and shape of leavesThePlant Journal 48 638ndash644 doi101111j1365-313X200602896x

Huff PM Wilf P Azumah EJ (2003) Digital future for paleoclimateestimation from fossil leaves Preliminary results Palaios 18266ndash274 doi1016690883-1351(2003)018lt0266DFFPEFgt20CO2

Husbands AY Chitwood DH Plavskin Y Timmermans MCP (2009)Signals and prepatterns new insights into organ polarity in plantsGenes amp Development 23 1986ndash1997 doi101101gad1819909

Illing N Klak C Johnson C Brito D Negrao N Baine F van Kets VRamchurnKRSeoigheCRodenL (2009)Duplicationof theAsymmetricLeaves1Rough Sheath 2Phantastica (ARP) gene precedes the explosiveradiation of the Ruschioideae Development Genes and Evolution 219331ndash338 doi101007s00427-009-0293-9

Jaya E Kubien DS Jameson PE Clemens J (2010) Vegetative phasechange and photosynthesis in Eucalyptus occidentalis architecturalsimplification prolongs juvenile traits Tree Physiology 30 393ndash403doi101093treephystpp128

Jones HG (1992) lsquoPlants and microclimate a quantitative approach toenvironmental plant physiologyrsquo 2nd edn (Cambridge UniversityPress Cambridge)

JonesCS (1995)Does shade prolong juvenile developmentAmorphologicalanalysis of leaf shape changes inCucurbita argyrosperma subsp sororia(Cucurbitaceae) American Journal of Botany 82 346ndash359 doi1023072445580

The evolution of leaf shape Functional Plant Biology 549

Jones CS (1999) An essay on juvenility phase change and heteroblasty inseed plants International Journal of Plant Sciences 160 S105ndashS111doi101086314215

Jones CS (2001) The functional correlates of heteroblastic variation in leaveschanges in form and ecophysiology with whole plant ontogeny Bulletinof the Botanical Society of Argentina 36 171ndash184

Jones CS Bakker FT Schlichting CD Nicotra AB (2009) Leaf shapeevolution in the South African genus Pelargonium Lrsquo Her(Geraniaceae) Evolution 63 479ndash497 doi101111j1558-5646200800552x

Jonsson H Heisler MG Shapiro BE Meyerowitz EM Mjolsness E (2006)An auxin-driven polarized transport model for phyllotaxis Proceedingsof the National Academy of Sciences of the United States of America 1031633ndash1638 doi101073pnas0509839103

Karban R Thaler JS (1999) Plant phase change and resistance toherbivory Ecology 80 510ndash517 doi1018900012-9658(1999)080[0510PPCART]20CO2

Kawamura E Horiguchi G Tsukaya H (2010) Mechanisms of leaf toothformation in Arabidopsis The Plant Journal 62 429ndash441 doi101111j1365-313X201004156x

KayKMSargentRD(2009)The roleof animalpollination inplant speciationIntegrating ecology geography and geneticsAnnual Review of EcologyEvolution and Systematics 40 637ndash656 doi101146annurevecolsys110308120310

Kenrick P Crane PR (1997) The origin and early evolution of plants on landNature 389 33ndash39 doi10103837918

Kerstetter RA Poethig RS (1998) The specification of leaf identity duringshoot development Annual Review of Cell and Developmental Biology14 373ndash398 doi101146annurevcellbio141373

Kidner CA (2010) The many roles of small RNAs in leaf developmentJournal of Genetics and Genomics 37 13ndash21 doi101016S1673-8527(09)60021-7

Kim GT Tsukaya H Saito Y Uchimiya H (1999) Changes in theshapes of leaves and flowers upon overexpression of cytochromeP450 in Arabidopsis Proceedings of the National Academy ofSciences of the United States of America 96 9433ndash9437 doi101073pnas96169433

Kim M McCormick S Timmermans M Sinha N (2003) The expressiondomain of PHANTASTICA determines leaflet placement in compoundleaves Nature 424 438ndash443 doi101038nature01820

Kleiman D Aarssen LW (2007) The leaf sizenumber trade-off in treesJournal of Ecology 95 376ndash382 doi101111j1365-2745200601205x

Kowalski EA Dilcher DL (2003) Warmer paleotemperatures forterrestrial ecosystems Proceedings of the National Academy ofSciences of the United States of America 100 167ndash170 doi101073pnas232693599

Kubien DS Jaya E Clemens J (2007) Differences in the structure and gasexchangephysiologyof juvenile and adult leaves inMetrosideros excelsaInternational Journal of Plant Sciences 168 563ndash570 doi101086513485

Kuwabara A Nagata T (2006) Cellular basis of developmental plasticityobserved in heterophyllous leaf formation of Ludwigia arcuata(Onagraceae) Planta 224 761ndash770 doi101007s00425-006-0258-4

Lake JA Quick WP Beerling DJ Woodward FI (2001) Plant developmentsignals from mature to new leaves Nature 411 154 doi10103835075660

Latimer AM Silander JA Cowling RM (2005) Neutral ecological theoryreveals isolation and rapid speciation in a biodiversity hot spot Science309 1722ndash1725 doi101126science1115576

Lee YK Kim GT Kim IJ Park J Kwak SS Choi G Chung WI (2006)LONGIFOLIA1 and LONGIFOLIA2 two homologous genes regulatelongitudinal cell elongation in Arabidopsis Development 1334305ndash4314 doi101242dev02604

Leigh A Zwieniecki MA Rockwell FE Boyce CK Nicotra AB HolbrookNM (2011) Structural and physiological correlates of heterophylly inGinkgo biloba L New Phytologist 189 459ndash470 doi101111j1469-8137201003476x

Levitt J (1980) lsquoResponses of plants to environmental stressesrsquo 2nd edn(Academic Press New York)

Lewis MC (1972) Physiological significance of variation in leaf structureScience Progress 60 25ndash51

Linder HP (2005) Evolution of diversity the Cape flora Trends in PlantScience 10 536ndash541 doi101016jtplants200509006

Little SA Kembel SW Wilf P (2010) Paleotemperature proxies from leaffossils reinterpreted in light of evolutionary historyPLoSONE5 e15161doi101371journalpone0015161

Martinez-Cabrera HI (2010) Influence of climate in functional and speciesdiversification in South African Pelargonium PhD Thesis University ofConnecticut Storrs CT USA

McDonald PG Fonseca CR Overton JM Westoby M (2003) Leaf-sizedivergence along rainfall and soil-nutrient gradients is the method ofsize reduction common among clades Functional Ecology 17 50ndash57doi101046j1365-2435200300698x

Melville R (1969) Leaf venation pattern and origin of angiosperms Nature224 121ndash125 doi101038224121a0

Monteith JL Unsworth MH (1990) lsquoPrinciples of environmental physicsrsquo2nd edn (Edward Arnold London)

Mueller RJ (1982) Shoot ontogeny and the comparative development of theheteroblastic leaf series in Lygodium japonicum (Thunb) Sw BotanicalGazette (Chicago Ill) 143 424ndash438 doi101086337318

Nath U Crawford BCW Carpenter R Coen E (2003) Genetic control ofsurface curvature Science 299 1404ndash1407 doi101126science1079354

Nicotra AB (2010) Leaf size and shape PrometheusWiki Available athttpprometheuswikipublishcsiroautiki-indexphppage=Leaf+size+and+shape [Verified 2 June 2011]

Nicotra AB Cosgrove MJ Cowling A Schlichting CD Jones CS (2008)Leaf shape linked to photosynthetic rates and temperature optima inSouth African Pelargonium species Oecologia 154 625ndash635doi101007s00442-007-0865-1

Niklas KJ (1988) The role of phyllotactic pattern as a developmentalconstraint on the interception of light by leaf surfaces Evolution 421ndash16 doi1023072409111

Niklas KJ (1997) lsquoThe evolutionary biology of plantsrsquo (University ofChicago Press Chicago)

Niklas KJ (2000) The evolution of plant body plans ndash a biomechanicalperspective Annals of Botany 85 411ndash438 doi101006anbo19991100

Niklas KJ (2004) Plant allometry is there a grand unifying theoryBiological Reviews of the Cambridge Philosophical Society 79871ndash889 doi101017S1464793104006499

Niklas KJ Cobb ED (2008) Evidence for lsquodiminishing returnsrsquo from thescaling of stem diameter and specific leaf area American Journal ofBotany 95 549ndash557 doi103732ajb0800034

Niklas KJ Cobb ED (2010) Ontogenetic changes in the numbers of short- vlong-shoots account for decreasing specific leaf area in Acer rubrum(Aceraceae) as trees increase in size American Journal of Botany 9727ndash37 doi103732ajb0900249

Niklas KJ Cobb ED Niinemets U Reich PB Sellin A Shipley B Wright IJ(2007) lsquoDiminishing returnsrsquo in the scaling of functional leaf traitsacross and within species groups Proceedings of the NationalAcademy of Sciences of the United States of America 104 8891ndash8896doi101073pnas0701135104

Niklas KJ Cobb ED Spatz H-C (2009) Predicting the allometry of leafsurface area and dry mass American Journal of Botany 96 531ndash536doi103732ajb0800250

550 Functional Plant Biology A B Nicotra et al

Nobel PS (1988) lsquoEnvironmental biology of Agaves and Cactirsquo (CambridgeUniversity Press Cambridge)

Oguchi R Hikosaka K Hirose T (2005) Leaf anatomy as a constraint forphotosynthetic acclimation differential responses in leaf anatomy toincreasing growth irradiance among three deciduous trees Plant Cellamp Environment 28 916ndash927 doi101111j1365-3040200501344x

Okada K Ueda J Komaki MK Bell CJ Shimura Y (1991) Requirement ofthe auxin polar transport-system in early stages of Arabidopsis floralbud formation The Plant Cell 3 677ndash684

Parkhurst DF LoucksOL (1972)Optimal leaf size in relation to environmentJournal of Ecology 60 505ndash537 doi1023072258359

Parkhurst DF Duncan PR Gates DM Kreith F (1968) Wind-tunnelmodelling of convection of heat between air and broad leaves ofplants Agricultural Meteorology 5 33ndash47 doi1010160002-1571(68)90021-6

Pasquet-Kok J Creese C Sack L (2010) Turning over a new lsquoleafrsquo multiplefunctional significances of leaves versus phyllodes in Hawaiian Acaciakoa Plant Cell amp Environment 33 2084ndash2100 doi101111j1365-3040201002207x

Peppe DJ Royer DL Cariglino B Oliver SY Newman S et al (2011)Sensitivity of leaf size and shape to climate global patterns andpaleoclimatic applications New Phytologist 190 724ndash739 doi101111j1469-8137201003615x

Perez-Perez JM Serrano-Cartagena J Micol JL (2002) Genetic analysis ofnatural variations in the architecture of Arabidopsis thaliana vegetativeleaves Genetics 162 893ndash915

Perez-Perez JM Candela H Robles P Quesada V Ponce MR Micol JL(2009) Lessons from a search for leaf mutants in Arabidopsis thalianaThe International Journal of Developmental Biology 53 1623ndash1634doi101387ijdb072534jp

Poethig RS Sussex IM (1985a) The cellular-parameters of leaf developmentin tobacco ndash a clonal analysis Planta 165 170ndash184 doi101007BF00395039

Poethig RS Sussex IM (1985b) The developmental morphology andgrowth dynamics of the tobacco leaf Planta 165 158ndash169doi101007BF00395038

Pray TR (1955) Foliar venation of angiosperms 2 Histogenesis of thevenation of Liriodendron American Journal of Botany 42 18ndash27doi1023072438589

Pray TR (1960) Ontogeny of the open dichotomous venation in the pinnaof the fern Nephrolepis American Journal of Botany 47 319ndash328doi1023072439217

Pryer KM Hearn DJ (2009) Evolution of leaf form in Marsileaceous fernsevidence for heterochrony Evolution 63 498ndash513 doi101111j1558-5646200800562x

Raschke K (1960) Heat transfer between the plant and the environmentAnnual Review of Plant Physiology and Plant Molecular Biology 11111ndash126

Raunkiaer C (1934) lsquoThe life forms of plants and statistical plant geographyrsquo(Clarendon Press Oxford)

Raven JA (1996) Into the voids the distribution function development andmaintenance of gas spaces in plants Annals of Botany 78 137ndash142doi101006anbo19960105

Roth-Nebelsick A (2001) Computer-based analysis of steady-state andtransient heat transfer of small-sized leaves by free and mixedconvection Plant Cell amp Environment 24 631ndash640 doi101046j1365-3040200100712x

RoyerDLWilf P (2006)Why do toothed leaves correlate with cold climatesGas exchange at leaf margins provides new insights into a classicpaleotemperature proxy International Journal of Plant Sciences 16711ndash18 doi101086497995

RoyerDLWilfP JaneskoDAKowalskiEADilcherDL(2005)Correlationsof climate and plant ecology to leaf size and shape potential proxiesfor the fossil record American Journal of Botany 92 1141ndash1151doi103732ajb9271141

Royer DL Kooyman RM Wilf P (2009a) Ecology of leaf teeth a multi-siteanalysis from an Australian subtropical rainforest American Journal ofBotany 96 738ndash750 doi103732ajb0800282

Royer DL Meyerson LA Robertson KM Adams JM (2009b) Phenotypicplasticity of leaf shape along a temperature gradient inAcer rubrumPLoSONE 4 e7653 doi101371journalpone0007653

Sack L Frole K (2006) Leaf structural diversity is related to hydrauliccapacity in tropical rain forest trees Ecology 87 483ndash491doi10189005-0710

Sack L Holbrook NM (2006) Leaf hydraulics Annual Review ofPlant Biology 57 361ndash381 doi101146annurevarplant56032604144141

Sack L Tyree MT (2005) Leaf hydraulics and its implications in plantstructure and function In lsquoVascular transport in plantsrsquo (Eds NMHolbrook MA Zwieniecki) pp 93ndash114 (Elsevier Academic PressBurlington MA USA)

Sack L Cowan PD Jaikumar N Holbrook NM (2003) The lsquohydrologyrsquo ofleaves co-ordination of structure and function in temperate woodyspecies Plant Cell amp Environment 26 1343ndash1356 doi101046j0016-8025200301058x

Sack L Streeter CM Holbrook NM (2004) Hydraulic analysis of water flowthrough leaves of sugar maple and red oak Plant Physiology 1341824ndash1833 doi101104pp103031203

Sakakibara K Nishiyama T Kato M Hasebe M (2001) Isolation ofhomeodomain-leucine zipper genes from the moss Physcomitrellapatens and the evolution of homeodomain-leucine zipper genes in landplants Molecular Biology and Evolution 18 491ndash502

Sanders H Rothwell GW Wyatt S (2007) Paleontological context for thedevelopmental mechanisms of evolution International Journal of PlantSciences 168 719ndash728 doi101086513519

Sarojam R Sappl PG Goldshmidt A Efroni I Floyd SK Eshed Y BowmanJL (2010) Differentiating Arabidopsis shoots from leaves by combinedYABBY activities The Plant Cell 22 2113ndash2130 doi101105tpc110075853

Sauquet H Weston PH Anderson CL Barker NP Cantrill DJ Mast ARSavolainen V (2009) Contrasted patterns of hyperdiversification inMediterranean hotspots Proceedings of the National Academy ofSciences of the United States of America 106 221ndash225 doi101073pnas0805607106

SchmidMDavisonTSHenz SR PapeUJDemarMVingronM ScholkopfBWeigel D Lohmann JU (2005) A gene expression map of Arabidopsisthaliana development Nature Genetics 37 501ndash506 doi101038ng1543

Schuepp PH (1993) Tansley Review No 59 Leaf boundary layers NewPhytologist 125 477ndash507 doi101111j1469-81371993tb03898x

Sculthorpe CD (1967) Vegetative polymorphism and the problem ofheterophylly In lsquoThe biology of aquatic vascular plantsrsquo pp 218ndash247(Koeltz Scientific Books Konigstein West Germany)

Shalit A Rozman A Goldshmidt A Alvarez JP Bowman JL Eshed YLifschitz E (2009) The flowering hormone florigen functions as a generalsystemic regulator of growthand terminationProceedingsof theNationalAcademy of Sciences of the United States of America 106 8392ndash8397doi101073pnas0810810106

Sharkey TD (2005) Effects of moderate heat stress on photosynthesisimportance of thylakoid reactions rubisco deactivation reactiveoxygen species and thermotolerance provided by isoprene Plant Cellamp Environment 28 269ndash277 doi101111j1365-3040200501324x

Smith WK Nobel PS (1977) Temperature and water relations for sunand shade leaves of a desert broadleaf Hyptis emoryi Journal ofExperimental Botany 28 169ndash183 doi101093jxb281169

Specht RL Specht A (1999) lsquoAustralian plant communities Dynamics ofstructure growth and biodiversityrsquo (Oxford University Press Oxford)

Street NR Sjoumldin A Bylesjouml M Gustafsson P Trygg J Jansson S (2008)A cross-species transcriptomics approach to identify genes involved inleaf development BMCGenomics 9 589 doi1011861471-2164-9-589

The evolution of leaf shape Functional Plant Biology 551

Taiz L Zeiger E (2002) lsquoPlant physiologyrsquo 3rd edn (Sinauer AssociatesSunderland MA USA)

Thoday D (1931) The significance of reduction in the size of leaves Journalof Ecology 19 297ndash303 doi1023072255823

Thom AS (1968) Exchange of momentum mass and heat between anartificial leaf and airflow in a wind tunnel Quarterly Journal of theRoyal Meteorological Society 94 44ndash55 doi101002qj49709439906

Thomas PWWoodward FI QuickWP (2004) Systemic irradiance signallingin tobacco New Phytologist 161 193ndash198 doi101046j1469-8137200300954x

Tomescu AMF (2009) Megaphylls microphylls and the evolution of leafdevelopment Trends in Plant Science 14 5ndash12 doi101016jtplants200810008

Tsukaya H (2006) Mechanism of leaf-shape determination Annual Reviewof Plant Biology 57 477ndash496 doi101146annurevarplant57032905105320

Tsukaya H (2010) Leaf development and evolution Journal of PlantResearch 123 3ndash6 doi101007s10265-009-0285-x

Tsukaya H Shoda K Kim GT Uchimiya H (2000) Heteroblasty inArabidopsis thaliana (L) Heynh Planta 210 536ndash542 doi101007s004250050042

Usami T Horiguchi G Yano S TsukayaH (2009) Themore and smaller cellsmutants of Arabidopsis thaliana identify novel roles for SQUAMOSAPROMOTER BINDING PROTEIN-LIKE genes in the control ofheteroblasty Development 136 955ndash964 doi101242dev028613

van der Niet T Johnson SD Linder HP (2006) Macroevolutionary datasuggest a role for reinforcement in pollination system shiftsEvolution 601596ndash1601 doi10155405-7051

Vogel S (1968) lsquoSun leavesrsquo and lsquoshade leavesrsquo differences in convectiveheatdissipation Ecology 49 1203ndash1204 doi1023071934517

Vogel S (1970) Convective cooling at low airspeeds and the shapes ofbroad leaves Journal of Experimental Botany 21 91ndash101 doi101093jxb21191

von Goethe JW (2009) lsquoThe metamorphosis of plantsrsquo (Reprinted by MITPress Cambridge MA USA) Originally published in 1790

Wagner WH (1952) Types of foliar dichotomy in living ferns AmericanJournal of Botany 39 578ndash592 doi1023072438706

Waites R Hudson A (1995) Phantastica ndash a gene required for dorsoventralityof leaves in Antirrhinum majus Development 121 2143ndash2154

Weston PH (1995) Athertonia In lsquoFlora of Australia Vol 16rsquo Ed PMcCarthy) pp 413ndash415 (CSIRO Publishing Melbourne)

Weston PH (2007) lsquoProteaceae in the families and genera of vascular plantsVol 9FloweringplantsndashEudicotsrsquo (VolumeEdsCJeffrey JWKadereitSeries Ed K Kubitzki) (Springer Berlin)

Wright IJ Reich PB Westoby M Ackerly DD Baruch Z et al (2004)The worldwide leaf economics spectrum Nature 428 821ndash827doi101038nature02403

Wu G Park MY Conway SR Wang JW Weigel D Poethig RS (2009) Thesequential action of miR156 andmiR172 regulates developmental timingin Arabidopsis Cell 138 750ndash759 doi101016jcell200906031

Yamaguchi T Tsukaya H (2010) Evolutionary and developmental studies ofunifacial leaves in monocots Juncus as a model system Journal of PlantResearch 123 35ndash41 doi101007s10265-009-0255-3

Yamaguchi T Yano S Tsukaya H (2010) Genetic framework for flattenedleaf blade formation in unifacial leaves of Juncus prismatocarpusThe Plant Cell 22 2141ndash2155 doi101105tpc110076927

YangDLiGSunS (2008)Thegenerality of leaf size versus number trade-offin temperatewoodyspeciesAnnalsofBotany102 623ndash629doi101093aobmcn135

Yano S Terashima I (2001) Separate localization of light signal perceptionfor sun or shade type chloroplast and palisade tissue differentiationin Chenopodium album Plant amp Cell Physiology 42 1303ndash1310doi101093pcppce183

Yano S Terashima I (2004) Developmental process of sun and shade leavesin Chenopodium album L Plant Cell amp Environment 27 781ndash793doi101111j1365-3040200401182x

Yapp RH (1912) Spirea ulmaria L and its bearing on the problem ofxeromorphy in marsh plants Annals of Botany 26 815ndash870

Yates MJ Verboom GA Rebelo AG Cramer MD (2010) Ecophysiologicalsignificance of leaf size variation in Proteaceae from the Cape FloristicRegion Functional Ecology 24 485ndash492 doi101111j1365-2435200901678x

Zhong RQ Ye ZH (1999) IFL1 a gene regulating interfascicular fiberdifferentiation in Arabidopsis encodes a homeodomain-leucine zipperprotein The Plant Cell 11 2139ndash2152

Zurakowski KA Gifford EM (1988) Quantitative studies of pinnuledevelopment in the ferns Adiantum raddianum and Cheilanthes viridisAmerican Journal of Botany 75 1559ndash1570 doi1023072444706

Zwieniecki MA Melcher PJ Boyce CK Sack L Holbrook NM (2002)Hydraulic architecture of leaf venation in Laurus nobilis L Plant Cell ampEnvironment 25 1445ndash1450 doi101046j1365-3040200200922x

Zwieniecki MA Boyce CK Holbrook NM (2004a) Functional design spaceof single-veined leaves role of tissue hydraulic properties in constrainingleaf size and shape Annals of Botany 94 507ndash513 doi101093aobmch173

Zwieniecki MA Boyce CK Holbrook NM (2004b) Hydraulic limitationsimposed by crown placement determine final size and shape of Quercusrubra L leaves Plant Cell amp Environment 27 357ndash365 doi101111j1365-3040200301153x

Zwieniecki MA Stone HA Leigh A Boyce CK Holbrook NM (2006)Hydraulic design of pine needles one-dimensional optimizationfor single-vein leaves Plant Cell amp Environment 29 803ndash809doi101111j1365-3040200501448x

Manuscript received 28 February 2011 accepted 30 May 2011

552 Functional Plant Biology A B Nicotra et al

httpwwwpublishcsiroaujournalsfpb