Search tips
Search criteria 


Logo of annbotAboutAuthor GuidelinesEditorial BoardAnnals of Botany
Ann Bot. 2016 April; 117(5): 859–879.
Published online 2015 December 9. doi:  10.1093/aob/mcv135
PMCID: PMC4845799

Why should we investigate the morphological disparity of plant clades?


Background Disparity refers to the morphological variation in a sample of taxa, and is distinct from diversity or taxonomic richness. Diversity and disparity are fundamentally decoupled; many groups attain high levels of disparity early in their evolution, while diversity is still comparatively low. Diversity may subsequently increase even in the face of static or declining disparity by increasingly fine sub-division of morphological ‘design’ space (morphospace). Many animal clades reached high levels of disparity early in their evolution, but there have been few comparable studies of plant clades, despite their profound ecological and evolutionary importance. This study offers a prospective and some preliminary macroevolutionary analyses.

Methods Classical morphometric methods are most suitable when there is reasonable conservation of form, but lose traction where morphological differences become greater (e.g. in comparisons across higher taxa). Discrete character matrices offer one means to compare a greater diversity of forms. This study explores morphospaces derived from eight discrete data sets for major plant clades, and discusses their macroevolutionary implications.

Key Results Most of the plant clades in this study show initial, high levels of disparity that approach or attain the maximum levels reached subsequently. These plant clades are characterized by an initial phase of evolution during which most regions of their empirical morphospaces are colonized. Angiosperms, palms, pines and ferns show remarkably little variation in disparity through time. Conifers furnish the most marked exception, appearing at relatively low disparity in the latest Carboniferous, before expanding incrementally with the radiation of successive, tightly clustered constituent sub-clades.

Conclusions Many cladistic data sets can be repurposed for investigating the morphological disparity of plant clades through time, and offer insights that are complementary to more focused morphometric studies. The unique structural and ecological features of plants make them ideally suited to investigating intrinsic and extrinsic constraints on disparity.

Keywords: Disparity, Embryophyta, morphological diversity, morphospace, angiosperms, conifers, ferns, macroevolution, clade shapes, developmental robustness, species diversity


The number of species within higher taxa, or within clades of a similar age (Magallón and Sanderson 2001), is hugely variable, even for sister groups diverging (by definition) at the same time. While rates and patterns of extinction are clearly influential, some clades appear much more adept at sub-dividing nîche space and speciating than others, even in comparison with their closest relatives. Some groups foster enormous radiations in diversity despite maintaining conservative bodyplans and displaying only modest morphological variety relative to that in their parent clades. Insects, as the best example, have a highly constrained body organization (a fixed number of appendages and tagmata) relative to other groups of arthropods (cf. crustaceans and branchiopods in particular), yet constitute over half of all described arthropod species (Mayhew, 2007). Similarly, beetles display remarkably conservative organization within insects, despite their notoriously high contribution to global species richness (Erwin, 1997). There is no necessary relationship, therefore, between the number of species within a group (species richness or diversity) and its morphological diversity. Indeed, there are suggestions that a constrained and entrenched bodyplan might actually be conducive to higher diversity (Rabosky et al., 2012).

In order to study the relationship between species richness and bodyplan conservation, we need to quantify both diversity and morphological variety or disparity for large groups. Methods for studying diversity are well established (Peet, 1974; Gotelli and Colwell, 2001; Benton, 2009; Ezard et al., 2011; Mayhew et al., 2012), but approaches for quantifying disparity are less familiar, particularly in the botanical literature (Chartier et al., 2014). While it is possible and informative to study diversity and disparity across clades within the extant biota (or, indeed, in any time slice), insights into the dynamics of their interaction are most fruitfully gained by investigating the trajectories of clades throughout their evolution. Most studies to date have focused on animals (Foote, 1994, 1997; Moyne and Neige, 2007; Hughes et al., 2013), but the long evolutionary history (Wellman, 2014) and rich fossil record of land plants (embryophytes) make them ideally suited for comparison. Diversity patterns through time within vascular plants have been studied for many years, typically deriving from species-level compilations of originations and extinctions (Knoll et al., 1979; Niklas et al., 1980; Lidgard and Crane, 1990; Kovach and Batten, 1993; Cascales-Miñana and Cleal, 2012, 2014). Results have differed in some details (Niklas and Tiffney 1994), but are broadly consistent in showing (a) a radiation of pteridophytes and gymnosperms in the Late Devonian-Carboniferous; (b) a gymnosperm-dominated flora in the early-mid Mesozoic of comparatively constant diversity; and (c) a mid-late Cretaceous to Tertiary diversity increase, due primarily to the radiation of the angiosperms. The presence of novel morphological features within this group raised the question of whether phases of embryophyte diversification could be explained by the acquisition of ‘key innovations’ within angiosperms (Endress, 2001), seed plants (Rudall and Bateman, 2007) and early land plants (Bateman et al., 1998; Renzaglia et al., 2000). Advances in plant phylogenetics have revealed that the timing of many plant radiations does not match the first appearance of hypothesized innovations (Sanderson and Donoghue, 1994; Davies et al., 2004; Vamosi and Vamosi, 2010), implying instead that the evolution of suites of characters over an extended period of time may enable diversification (Donoghue, 2009). The hunt for specific drivers has shifted to focus either on competitive interactions, for example between plants and herbivores (Agrawal, 2007; Futuyma and Agrawal, 2009), or on environmental factors such as climatic change (McElwain et al., 1999; Beerling et al., 2001; Willis and Niklas, 2004; Beerling and Berner, 2005; Feild and Arens, 2007; Boyce et al., 2009; Willis and McElwain, 2013).

In marked constrast to diversity, for which temporal patterns have been investigated for many years (Crane et al., 1994; Kenrick and Crane, 1997; Lupia, 1999; Schneider et al., 2004; Soltis and Soltis, 2004; Crisp and Cook, 2011; Donoghue, 2008), there have been only a handful of studies on the morphological disparity of plants (Lupia, 1999; Niklas, 1999; Chartier et al., 2014). Often these studies have focused on a specific aspect of plant morphology such as leaves (Boyce and Knoll, 2002; Boyce, 2005) or vascular systems (Wilson and Knoll, 2010; Feild et al., 2011). Disparity analyses have furnished an important means of assessing macroevolutionary patterns in animals for some years, and we believe that their application to plants would be equally insightful.


This paper has two primary aims. The first is to provide an overview of the methods used to quantify morphological disparity, with particular emphasis on their application to plant evolution. We contrast concepts of disparity with those of diversity or species richness, and explain how exploring both trajectories through time can shed light on the evolutionary dynamics of clades. Morphological disparity is usually quantified with reference to the axes of some form of morphospace; an n-dimensional space in which the distances between species or other operational taxonomic units are proportional to some measure of the morphological distances between them. We therefore distinguish between theoretical and empirical morphospaces and discuss their relative advantages and disadvantages for the study of plants. We also explore a variety of potential data sources and consider their relative merits. Particular emphasis is given to character-based empirical methods, which have proved broadly applicable to animal clades at a wide range of taxonomic levels (Hughes et al., 2013), but have yet to be utilized in plants. The second objective is to demonstrate the application of these methods to a select number of published character matrices for major plant groups. We compare and contrast the observed patterns of disparity through time with those seen in animals, and offer a prospectus for future studies of plant disparity.

What is disparity and why should we study it?

The macroevolution of any major clade through deep time can be characterized in a number of ways. There is perennial interest in how diversity changes (Sepkoski et al., 1981; Sepkoski, 1997; Sepkoski and Miller, 1998), particularly with regards to how species and higher taxonomic richness responds to major physical or biotic changes such as mass extinctions, the opening up of new habitats or the origination of other major groups. Equally fundamentally, we may wish to know how the constituent taxa of a clade are related, and may use phylogeny to better inform the patterns above. Increasingly, however, palaeobiologists are also focusing on the manner in which groups diversified morphologically to give rise to new bodyplans or architectures (Fortey et al., 1996). The range or variance of morphological form across species or other taxa is usually referred to as ‘morphological variety’, ‘morphological disparity’ or simply ‘disparity’ in context. Disparity is therefore a property of a sample of taxa rather than of individual species, and is also measured relative to some set of quantifiable variables. Trajectories of disparity through time are often different from patterns of species and higher taxonomic diversity, and are also difficult to predict from phylogeny.

Although all morphological variety is generated within the context of a phylogeny, diversity and disparity are fundamentally decoupled (Foote, 1991; Fortey et al., 1996, 1997; Moyne and Neige, 2007). Large samples of morphologically very similar species typically have much lower disparity than small groups of morphologically highly dissimilar species. Specifically, numerous basal groups of animals show levels of disparity greater than or equal to their more diverse, derived counterparts (Fig. 1) (Foote, 1992, 1994, 1997; Wills et al., 1994; Wills, 1998) although exceptions exist (Benson et al., 2012). At a coarse level, higher taxonomic diversity (e.g. numbers of orders or classes) tends to be a better proxy for disparity than numbers of species or genera (Foote 1990). Plots of relative disparity through time are therefore often used alongside plots of diversity in order to understand the dynamics of clade evolution more fully.

Fig. 1.
Diversity and disparity are often decoupled, particularly when sampling at lower taxonomic levels. Data for crinoids from Foote (1999) have been adapted and are reproduced with permission from Cambridge University Press. When crinoids first appear in ...

Much of the initial impetus for quantifying levels of disparity came from claims about the evolutionary significance of the fossils from the Middle Cambrian Burgess Shale (Whittington, 1985; Conway Morris, 1989). In particular, it was claimed (Gould, 1989) that the range of morphological variety amongst Cambrian arthropods was far greater than that realized at any time subsequently; an argument couched (at least initially) in the perceived higher taxonomic status (i.e. sub-phylum or class) of many Burgess Shale genera. Gould (1991) subsequently propounded an ‘inverted iconography’ model for the evolution of life. An initial phase of experimentation and looser constraint on bodyplan evolution was posited to yield early maximal disparity, followed by a phase of winnowing in which most bodyplans were lost and the survivors consolidated and canalized. Subsequent evolution would typically yield few new bodyplans, but would see increases in diversity, i.e. increasing numbers of variations (species) upon a more limited number of constrained themes. However, empirical studies of marine invertebrates found that the disparity of Cambrian and Recent faunas was essentially equivalent (Briggs et al., 1992; Wills et al., 1994, 2012; Fortey et al., 1996) (Fig. 2). Subsequent studies have examined the disparity of clades at numerous successive time intervals, often demonstrating relatively high early disparity even while diversity is low (Foote, 1992, 1994; Wills, 1998). Recently, this approach has been applied to a larger data set of exclusively fossil animal clades (Hughes et al., 2013). The shape of the disparity profile of a clade through time can be summarized as a centre of gravity index (CG). Clades with precisely symmetrical patterns through time have indices of 0·50, those with higher levels of disparity early in their history have values <0·5 (bottom heavy), while those peaking late tend to >0·50 (top heavy). In a sample of 98 extinct clades that did not go extinct coincident with one of the ‘Big Five’ (Hallam and Wignall, 1997; Bambach, 2006) mass events, there was a significant bias towards bottom heaviness and early high disparity. Groups persisting to the present tend to have top-heavy profiles; not least because they are artificially truncated by the Recent. Those disappearing coincident with one of the ‘Big Five’ mass events also tend to be top heavy, and for similar reasons.

Fig. 2.
Simplified models of the pattern of morphological disparity through the Phanerozoic. The ‘traditional’ model assumes that patterns of disparity loosely track diversity, which increases (albeit irregularly) through time. Gould (1989) espoused ...

Other research agendas have become increasingly important within particular clades. One is the extent to which bodyplans are modular, and comprise units within which changes are relatively tightly correlated, but between which there is greater flexibility (Klingenberg et al., 2004; Monteiro and Nogueira, 2010; Cooper et al., 2010; Drake and Klingenberg, 2010) Another is the extent to which developmental vs. environmental factors constrain bodyplans over evolutionary time (Allen et al., 2008; Anderson et al., 2011). Increasingly, there is also interest in quantifying functional disparity, notably in fish and basal tetrapods (Friedman, 2010; Anderson et al., 2013).

Why study the disparity of plants?

In contrast to animals, there have been few studies investigating the morphological disparity of plant clades. It is possible that the patterns in plants may differ from those in animals; both the trends observed in statistical samples of clades, and the overall pattern of disparity through time for the group as a whole. In this latter context, it may be informative to compare plots of ordinal diversity through time (compiled from Benton, 1993), insofar as counts of higher taxa afford a very rough approximation to disparity (Fig. 3). Animals reach relatively high levels of ordinal diversity relatively early in their history; commensurate with the patterns revealed in explicit studies of disparity. The pattern observed in vascular plants differs markedly. Even accounting for the much later origin of vascular plants compared with animals, plants show a much more gradual increase in ordinal diversity, reaching 50 % of their maximum relatively late in their evolutionary history. Plants show ordinal diversity increases in three discrete phases: (1) the Late Devonian, corresponding to the initial radiation of pteridophytes and gymnosperms; (2) a smaller increase at the start of the Cretaceous, coincident with the appearance of the angiosperms; and (3) a Late Cretaceous increase, corresponding to the appearance of many modern angiosperm groups (Niklas and Tiffney, 1994).

Fig. 3.
Ordinal diversity of animals (Eumetazoa; blue) and plants (Embryophta: green) through the Phanerozoic. Numbers of orders per geological stage have been tallied from Benton (1993) for animals and from Cascales-Minãna and Cleal (2014) for plants. ...

Ordinal diversity profiles (Fig. 3) suggest that vascular plants have fewer fundamentally different modes of morphological organization than animals, and acquired novel bodyplans more gradually. Strikingly, plants appear to be relatively unperturbed by the mass extinction events that were catastrophic for animals; or at least the recovery of plants was rapid enough to mask any significant diversity decreases in the fossil record (Rees, 2002; McElwain and Punyasena, 2007; Cascales-Miñana and Cleal, 2014). Plants therefore appear to have greater resilience to certain types of ecological disturbance than animals (Cascales-Miñana and Cleal, 2012); a surprising inference given that many aspects of plant morphology are thought to be tightly mechanically and physiologically constrained to optimize photosynthetic efficiency and structural support (Niklas and Kerchner, 1984). Even relatively simple optimization models with a small number of variables can produce the diverse spectrum of habits and gross phenotypes seen across plant groups (Farnsworth and Niklas, 1995; Niklas, 1999) (Fig. 4); ecological disturbance may actually serve as a driver for increasing phenotypic diversity. Therefore, although basic structural components (e.g. phytomers in the case of branches) may be relatively morphologically conserved across taxonomic groups, they can nevertheless produce markedly different gross morphologies, even between closely related species or within species. This scale-dependent disparity is one of the defining characteristics of vascular plants and probably facilitates the unparalleled level of phenotypic plasticity seen within many plant species (Schlichting, 1986, 2002; Bradshaw, 2006). The hierarchical modularity in many aspects of plant form (Barthélémy and Caraglio, 2007; Klingenberg et al., 2012) may also have profound implications for plant evolution (Friedman and Williams, 2003).

Fig. 4.
Simulation of bifurcate branching structures capturing aspects of vascular plant morphology (after Niklas, 1999). (A) Illustration of the three parameters used: the bifurcation angle ϕ, the rotation angle Y and the probability of apical bifurcation ...

Studies of plant disparity to date have mostly focused on specific structures in which shape variation is believed to be of particular functional importance, rather than on holistic analyses of form. Leaf and shoot disparity, in particular, has been extensively studied. Boyce and Knoll (2002) investigated trends in leaf shape in fossil plant lineages, revealing a rapid expansion of leaf morphospace in the Early/Middle Carboniferous. The genetic controls on leaf shape (Langlade et al., 2005; Chitwood et al., 2014) and compound leaf structures are gradually being better understood (Klingenberg et al., 2012). Leaf shape appears to be correlated with shoot morphology (Lacroix et al., 2003; Jeune et al., 2006), although the importance of selective, functional and historical constraints in the evolution of these hierarchical systems is poorly understood (Burns et al., 2008). Floral morphology, despite having long been recognized as a critical component of angiosperm disparity (Stebbins, 1951), has received relatively little attention until recently (Whibley et al., 2006; Stournaras et al., 2013; Chartier et al., 2014). Similar considerations apply to the architecture of inflorescences (Prusinkiewicz et al., 2007). Other work has investigated the evolution and possible adaptive value of different types of pollen (Lupia, 1999; Ressayre and Godelle, 2002) as well as physiological properties in the conductive vessels of major seed plant groups (Wilson and Knoll, 2010). Rather than attempt to assess disparity from large collations of morphological data, more holistic approaches tend to consider habit and gross architecture (Niklas and Kerchner, 1984; Niklas, 1999; Silva and Batalha, 2011).

The decoupling of diversity and disparity within higher plant clades appears every bit as great as that within animal groups (Yu et al., 2014). For example, the true grasses (Poaceae) and the bromeliads (Bromelilaceae) are both families of angiosperms in the order Poales. However, the true grasses are represented by about 10 000 species (The Plant List, of varying size but relatively limited floral disparity, while the bromeliads contain just over 3000 species but show huge variation in inflorescence morphology (Benzing, 2000; Sajo et al., 2004). It is clear that a complete picture of plant disparity cannot be captured by focusing exclusively on the disparity of specific structures (as there is strong scale dependence) or by using diversity as a proxy. Holistic approaches that use a broad suite of characters sampled over large numbers of taxa will probably constitute the best way of quantifying plant disparity at macroevolutionary scales. Here, we take some preliminary steps in this direction for a sample of higher plant clades.

Types of data

There are many approaches to quantifying morphology (Moore and Moser, 1995; Chapman and Rasskin-Gutman, 2001; Lockwood et al., 2002), and the most suitable usually depends upon the application and the question being addressed. Where the forms being compared are broadly similar (e.g. typically species within genera or families), a variety of morphometric approaches can be used to derive sets of continuous variables describing shape and shape change, usually with some implicit standardization for size and orientation (Rohlf and Marcus, 1993; Adams et al., 2004) (Fig. 5). Three-dimensional, landmark-based approaches operate by identifying biologically (or functionally) homologous points (e.g. intersections between homologous structures) across all of the species or higher taxa (hereafter ‘operational taxonomic units’ or OTUs) being compared (Marcus, 2000; von Cramon-Taubadel et al., 2007; Mitteroecker and Gunz, 2009). Outline-based methods describe shapes in more detail. This can be either using a more limited number of discrete points (homologous landmarks), possibly interspersed with semi-landmarks to specify the form further (Bookstein 1997; Perez et al., 2006), or using continuous functions (e.g. Fourier analysis) describing shape (Rohlf and Archie, 1984; Crampton, 1995). Where the forms being compared are more divergent (e.g. across higher taxa), it often becomes difficult to identify a sufficient number of homologous or functional landmarks to capture all but the most limited and conservative aspects of form variation (Bocxlaer and Schultheiß, 2010). Here, it is possible to use an array of discretely coded characters, each recognizing two or more alternative states, as descriptors of morphological variation (Wills et al., 1994; Wills, 1998). Such data are more flexible, but entail more assumptions and potential subjectivity concerning the selection and discretization of characters and states. The morphospaces that they define also have properties that differ from those derived from continuous character data (Gavrilets, 1999).

Fig. 5.
Types of data underpinning disparity analyses. (A) Landmarks (in red) from Webster and Zelditch (2008) situated on homologous points of a trilobite cephalon. (B) Equidistant semi-landmark points (in red) from MacLeod (2011), defining the outline of a ...

The first studies quantifying disparity explicitly with empirical data sets were published about 25 years ago (Foote, 1990, 1994; Briggs et al., 1992; Wills et al., 1994) (Fig. 6). The disparity profiles of numerous major animal clades were investigated over the next decade, before a wane in apparent interest. The last few years, however, have seen the resurgence of empirical studies, with a particular emphasis on the use of discrete character data sets (see Supplementary Data Table S1). As a general rule, metazoan clades tend to show an initial rapid increase in disparity, with early levels of disparity being at or close to the maximum levels observed throughout the group’s history.

Fig. 6.
The resurgence of disparity analyses for animal clades and the paucity of plant studies. (A) Bar chart of the number and taxonomic distribution of focal clades in disparity analyses from 1990 to 2014. (B) The decline in the use of outline data and the ...

Biological homology and functional analogy

With all types of data, a distinction can be drawn between those approaches that attempt to capture variation in biologically homologous aspects of morphology (Rohlf, 2002; Klingenberg et al., 2004), and those that are more concerned with the functional parameters of shape (Nogueira et al., 2009; Figueirido et al., 2011; O’Higgins et al., 2011; Anderson et al., 2011, 2013). Morphological disparity can be used to refer to both aspects of variation in form, although the intention is sometimes unspecified (Love, 2007). The distinction can be illustrated with reference to the tails of derived sharks and ichthyosaurs, both of which have convergently evolved dorsal and ventral lobes with a relatively high aspect ratio for high-speed aquatic locomotion (Motani, 2002; Lingham-Soliar, 2005; Lingham-Soliar and Plodowski, 2007). In functional terms, the dorsal lobes of both groups are comparable, as are the ventral lobes. However, the vertebral column of sharks extends into the dorsal tail lobe, while that of ichthyosaurs deviates into the ventral lobe. The tip of both dorsal lobes might therefore constitute a valid functional landmark, but the tip of the dorsal lobe of sharks is biologically homologous to the tip of the ventral lobe in ichthyosaurs. Similar considerations apply to discrete, character data; much depends upon the manner in which characters and states are defined.

The exclusive use of putatively biologically homologous discrete variables restricts consideration to the same pool of characters used by cladists. In practice, and especially when dealing with fossil taxa, cladistic homology is established on operational grounds of detailed similarity and relationships to other structures (Pinna, 1991; Butler and Saidel, 2000). Such characters may also be functionally analogous, but are not necessarily so (Ruvinsky and Gibson-Brown, 2000; Shubin et al., 2009). Cladistic matrices therefore offer a rich resource for quantifying morphological variety across more conservative suites of putatively biologically homologous characters. Moreover, in the absence of homoplasy, we would expect the inter-OTU morphological distances assessed from such data to correlate closely with the evolutionary or patristic distances inferred on most parsimonious or otherwise optimal phylogenetic trees. With the progressive introduction of more character conflict and homoplasy (Sanderson and Donoghue, 1989), this correlation will increasingly break down (Kelly et al., 2014), as will the inferred validity of many of the homology statements underpinning the data. Cladograms must account for the distribution of states across taxa by introducing hypotheses of convergence and reversal along branches. The metrics of morphological differences underpinning analyses of morphological disparity do not account for similarity due to the convergent acquisition or loss of traits, and are therefore intrinsically more phenetic in approach. Indeed, as levels of homoplasy increase (and more putative homologies are revealed to be analogies), patterns of morphological variety inferred from homologies and those inferred from statements of functional similarity become progressively more similar.

Morphospaces: theoretical and empirical

Once a set of morphological descriptors or variables has been established for a given group, it is possible to assess the morphological variety of constituent subgroups (e.g. clades) or of chronological sub-samples (e.g. taxa from successive geological periods). This can be done directly from the data, but it is more typical to visualize patterns of taxonomic distributions in some form of morphospace; an abstract, multidimensional space in which distances correlate with morphological differences. A distinction (although one not universally embraced; Mitterocker and Huttegger, 2009) can be drawn between theoretical and empirical morphospaces. Theoretical morphospaces typically have dimensions that each capture a single quantifiable aspect of form, and (despite being parameterized with reference to real organisms) are defined a priori without the need for an empirical data set. The most frequently cited examples are those describing mollusc shells, which variously quantify form and growth using a very modest number of variables (Raup and Michelson, 1965; Skalak et al., 1997; Hammer and Bucher, 2005; Urdy et al., 2010). Real specimens can be located within theoretical morphospaces, but empirical data are not necessary in order to define them. Empirical morphospaces, in contrast, are constructed from a particular set of empirical morphological data. Their dimensionality tends to be high (Raup and Michelson, 1965; Foote, 1997; McGhee, 1999; Mitteroecker and Huttegger, 2009); much higher than that of their theoretical counterparts. For this reason, a number of data reduction techniques (usually multivariate ordination such as principal components or co-ordinates analyses) are used to condense the dimensionality of the space. This makes it possible to summarize morphological variation using a smaller number of abstracted variables, whilst minimizing distortion. These abstracted axes often cannot be described verbally, but may allow the relative disparity of groups to be visualized and quantified more readily. Many of these approaches necessitate a distillation of the multivariate differences between taxa into a single measure of difference or distance for all possible taxon pairs (often realized as a triangular distance matrix analogous to that used to tabulate distances in a road atlas). The precise distance metric used depends upon the nature of the data and the desired properties of the resultant space and/or disparity indices. These complexities are discussed elsewhere at length (Wills, 1998, 2001; Hughes et al., 2013).

Two issues deserve emphasis. First, all morphospaces are abstractions, and necessarily based upon a sub-set of morphological variables. Variable choice inevitably determines the nature of the space. Many practitioners seek to sample variables as widely as possible from all aspects of morphology, thereby deriving spaces that reflect overall form. This is not always possible, however, as in many cases where only variation in particular organs or aspects of form can be codified across taxa (Pretorius and Scholtz, 2001; Lindbladh, 2002; Miller and Venable, 2003; Neige, 2006; Jones et al., 2009). Morphospaces derived from particular aspects of form or using data from particular organ systems or modules may be well suited to addressing particular evolutionary questions. However, ‘morphological disparity’ is usually conceived as referring holistically to overall form. Secondly, indices of disparity are necessarily relative, and comparisons are only possible within the parameters of a given morphospace or underlying data set. Hence, while it is possible to make inferences regarding the relative disparity of a group at different times in its evolutionary history, or to compare the disparity of constituent subgroups within an analysis, it is not possible to make comparisons between groups from independently constructed morphospaces or data sets. This is also the reason why supermatrices uncritically assembled from multiple published data sets (and containing large blocks of inapplicable codes for large groups of taxa) may lose traction on some of the largest and deepest comparative questions.

A variety of disparity indices have been discussed in the literature (Foote, 1991, 1994, 1997; Wills et al., 1994; Wills, 2001; Hughes et al., 2013), but it is not our intention to rehearse the relative merits of these here. Among the most widely used approaches are those that distil the dispersion of taxa on multiple axes of the morphospace into a single value. The dispersion on a single axis can be quantified as either the range (defined by the outliers) or the variance of scores; the latter has the advantage of a relative insensitivity to sample size differences. Measures on multiple axes can be combined either as their product – effectively calculating the hypervolume of a hypercube – or as their sum. While hypervolumes are superficially more intuitive, they effectively give disproportionate weighting to smaller differences on later axes. Most ordination methods sequester progressively smaller fractions of total variance in later axes, but multiplying the univariate indices of dispersion means that halving the spread on any axis (whether the first or last) will halve the resultant hypervolume. Products also collapse to zero whenever the dispersion of taxa on a given axis is zero. Summing the univariate indices of dispersion (rather than multiplying them) avoids these problems. The sum of variances has particularly desirable properties, therefore, and has been used throughout the present study.


Data collection

In general, we followed the protocols set out in Hughes et al. (2013). Morphological matrices for six major tracheophyte groups [Angiospermae (Doyle et al., 1994; Nandi et al., 1998; Doyle and Endress, 2014), Arecaceae (Baker et al., 2009), Nymphaeales (Borsch et al., 2008), Pinophyta (Hart, 1987), Pinaceae (Klymiuk and Stockey, 2012) and Polypodiidae (Pryer et al., 1995)] were selected from the literature. These represent the most diverse extant higher taxa of vascular plants, in addition to well-sampled sub-clades within both the angiosperms and conifers. A breakdown of the number and type (broad anatomical focus) of characters present in each data set is given in Supplementary Data Table S2. All of the matrices – with the exception of the cone-only pine data set – included a wide range of characters representing most aspects of morphology. The morphospaces derived from them might therefore reasonably be expected to capture overall form. Outgroup taxa were removed from these source matrices in several cases; either because there were missing taxa between the ingroup and outgroup, or because the outgroup OTUs were sampled at a higher taxonomic level than the ingroup. In some data sets, we also had to overcome inhomogeneity of sampling within the ingroup, which we achieved by amalgamating OTUs in such a way as to render them homogeneous at a higher taxonomic level. Character amalgamation utilized modal states. Some characters were rendered uninformative as a result of these condensations, and were therefore removed (specifically Pinaceae – 46, 47, 51; Arecales – 6, 10, 15, 21, 22, 48, 78, 91, 92, 10; Nymphaeales – 4, 5, 6, 12, 14, 22, 28, 39, 57). Stratigraphic ranges were assigned to stages using the International Stratigraphic Chart (Gradstein et al., 2004; Ogg et al., 2008). Stratigraphic range data were sourced from the Paleobiology Database (, Sepkoski Online (Sepkoski, 2002) and The Fossil record 2 (Benton, 1993). Ranges were treated as continuous between first and last occurrences, with data being grouped into stage-level time bins. In cases where first and last occurrences were resolved only to intervals above the stage level, we coded for the stage corresponding to the midpoint of the interval. There were very few fossils within the Nymphaeales and we therefore estimated ranges using the time-calibrated molecular phylogeny of Yoo et al. (2005). Temporal bins with sample sizes of one were also amalgamated so that disparity could be calculated for these intervals. Comparisons between our sampled diversity curves and those compiled from more comprehensive sources are given in Supplementary Data Fig. S1.


For each exemplary clade, intertaxon distance matrices were calculated using the Generalised Euclidean Distance (GED) metric of Wills (1998), and as implemented in Hughes et al. (2013). Distance matrices were ordinated in R (R Core Team, 2013) using principal co-ordinates analysis (Wills et al., 1994), and incorporating Caillez’s correction for negative eigenvectors (Cailliez 1983). Disparity for each time bin was calculated as the sum of variances on all axes of the morphospace, yielding a trajectory of disparity through time. The centre of gravity of each trajectory was used to distinguish between those clades whose temporal mean disparity was located early (bottom heavy), late (top heavy) or in the middle of their evolutionary history (symmetrical). A centre of gravity metric (Gould et al., 1987; Uhen, 1996) in absolute time (CG) was calculated for each clade as:


where di is the disparity at the ith stratigraphic interval and ti the temporal midpoint in absolute time [million years (My)] of the ith stratigraphic interval. This was then scaled between the ages of the oldest (toldest) and youngest (tyoungest) intervals to yield an index of observed CG (CGscaled) between 0 and 1.


The expected CGscaled for a clade of constant disparity through time is unlikely to be 0·50, but is determined by the durations of the time bins over which the profile is measured. The observed CGscaled was therefore compared with the inherent CG (CGi) for a clade of uniform disparity spanning the same intervals. A bootstrapping procedure was used to generate a distribution of 1000 resampled differences between CGscaled and CGi, and clades for which >97·5 % of bootstrapped replicates lay either above or below the centre of gravity inherent in the time scale (P < 0·05) were deemed to be significantly top or bottom heavy, respectively (Foote, 1991). Observed CGscaled was then expressed relative to CGi as a baseline; hereafter simply CG.

An ancillary test from Hughes et al. (2013) was used to determine whether the taxa observed in the first two stages had significantly less disparity than the maximum observed in any time bin. The disparity profile of the clade was bootstrapped 1000 times. For each replicate curve, the difference in disparity between the first two stages and the disparity maximum was calculated, yielding a distribution. If a difference of zero was within the 95 % limits of this distribution, we were unable to reject the null hypothesis: namely that there was no difference between the initial disparity and the maximum (early high disparity). In such cases, maximal disparity was achieved in the earliest stages of the clade’s evolution. A similar test was applied to the end of each group’s history (late high disparity).


Patterns of plant disparity through time

Our results are presented as preliminary explorations of the manner in which our selected clades have explored one form of morphospace through time. While more detailed work will certainly follow, our findings highlight several general patterns and permit certain conclusions.

For extinct clades with homogeneous birth/death dynamics and characters evolving under a Brownian model, the null expectation is that clade disparity profiles should be somewhat top heavy on average (a mean clade CG >0·5) (Foote, 1991). This is because the morphology of new lineages is contingent upon the morphology of those from which they have evolved; clades would therefore be expected to explore morphospaces in a progressive manner. The extinction of lineages, in contrast, can occur in any pattern with respect to the morphospace. Random extinction, in particular, will tend to maintain a relatively wide morphospatial distribution, introducing a fundamental asymmetry into clade evolution. This is an oversimplistic model for the clades studied here, because all are extant; the Recent effectively truncates their evolution. As demonstrated by Hughes et al. (2013), extant clades (as well as those becoming extinct coincident with one of the ‘Big Five’ mass events) have a much greater tendency towards top-heaviness merely by virtue of their persistence to the Recent. It is therefore unsurprising that most of our exemplar clades (with the exception of two of the three angiosperm data sets: Doyle et al., 1994; Nandi et al., 1998) show significantly top-heavy (CG >0·5) profiles (Table 1). More remarkably, several of these clades show initial disparity levels close to their maxima, or reach this level early in their history. Arecaceae (palms) (Baker et al., 2009) first appeared at their maximum disparity, while all three of our angiosperm data sets (Nandi et al., 1998; Doyle et al., 1994; Doyle and Endress, 2014) showed initial disparity levels >90 % of the maxima in each case. Polypodiales (ferns) (Pryer et al., 1995) and Pinaceae (pines) (Klymiuk and Stockey, 2012) both reached their maxima within three time bins. The Nymphaeales (water lilies) (Borsch et al., 2008) are represented by a small data set (just 22 taxa) partitioned into larger time bins. Despite their apparently slow start, early disparity levels were not significantly different from the maximum (Table 1).

Table 1.
Expected (or inherent) and observed centres of gravity (CGscaled) for clade disparity profiles, along with the results of bootstrapping tests (CG P-value) to determine if these differ

Conifers (Hart, 1987) have the most dynamic disparity trajectory, with initial Carboniferous and Permian levels significantly lower than at any subsequent times (Fig. 7). These modest levels persisted until after the end of the Permian, whereupon there were significant increases into the early Mesozoic. Although disparity appears to decline between the Middle and Late Triassic, it increases subsequently to reach maximum levels at the end of the Jurassic. Levels then decline gradually until the Recent, with extant disparity being significantly lower than the maximum levels observed at the end of the Jurassic. Conifers also show more intensive clustering of taxa in the morphospace at a variety of spatial scales than do the other clades in our study (Fig. 8). Disparity within the pine family (Fig. 9) shows broad similarities with conifers as a whole from their origins in the Jurassic; a reassuring finding given that pines represent a significant proportion of conifer diversity from this time. The initial increase in disparity for pines occurs slightly later than the corresponding increase in conifers as a whole, and is maintained until the present day.

Fig. 7.
(A) Disparity and diversity profiles for conifers using data from Hart (1987). Disparity (black circles) calculated as the sum of variances on all principal co-ordinate axes within several time bins. Values are the mean of 1000 bootstrap replicates ± ...
Fig. 8.
To what extent are taxa clustered within their respective morphospaces at different levels of granularity? Highly clustered, spatially heterogeneous distributions can be approximated with smaller numbers of principal points than can diffuse, spatially ...
Fig. 9.
(A) Disparity and diversity profiles for the pine family using data from Klymiuk and Stockey (2012). Disparity (black circles) calculated as the sum of variances on all principal co-ordinate axes within several time bins. Values are the mean of 1000 bootstrap ...

Both angiosperms as a whole (Doyle et al., 1994) (Fig. 10) and the palm sub-clade (Baker et al., 2009) (Fig. 11) show approximately constant disparity through time. Palm disparity undergoes a slight decrease through the end of the Mesozoic and the early Palaeogene, such that the disparity of living taxa is lower than the realized maximum of the past. In contrast, our results suggest that the water lilies (Borsch et al., 2008) did not reach present levels of disparity until the Neogene (Fig. 12), with markedly lower levels for the first 10 My of their history. We note that this is our smallest data set (22 taxa), resulting in large estimates of error relative to observed fluctuations in disparity.

Fig. 10.
(A) Disparity and diversity profiles for angiosperms using data from Doyle and Endress (2014). Disparity (black circles) calculated as the sum of variances on all principal co-ordinate axes within several time bins. Values are the mean of 1000 bootstrap ...
Fig. 11.
(A) Disparity and diversity profiles for palms using data from Baker et al. (2009). Disparity (black circles) calculated as the sum of variances on all principal co-ordinate axes within several time bins. Values are the mean of 1000 bootstrap ...
Fig. 12.
(A) Disparity and diversity profiles for water lilies using data from Borsch et al. (2008). Disparity (black circles) calculated as the sum of variances on all principal co-ordinate axes within several time bins. The sampled number of species ...

In polypod ferns, disparity increases through the Permian and Triassic, reaching or slightly exceeding modern levels by the Early Jurassic (Fig. 13). Disparity increased slightly thereafter to peak levels around the K/Pg but subsequently declined significantly in the last few million years.

Fig. 13.
(A) Disparity and diversity profiles for extant leptosporangiate ferns using data from Pryer et al. (1995). Disparity (black circles) calculated as the sum of variances on all principal co-ordinate axes within several time bins. Values are the ...

An unexpected observation is that high levels of disparity were maintained for the past 80+ My in our largest clades (conifers, pines, ferns and angiosperms), despite successive radiations of sub-groups and catastrophic environmental and faunal upheavals over this time, including the K/Pg event (Ehleringer and Sage, 1991; Cerling et al., 1997; Zachos et al., 2001). Indeed, while there is evidence of significant local faunal turnover in plants (McElwain and Punyasena, 2007), recent work suggests that only two major extinction pulses are supported in the plant fossil record: one at the Carboniferous–Permian transition and another during the middle-late Permian (Cascales-Miñana and Cleal, 2014). Of the groups analysed, only conifers spanned this second event and actually show a significant increase in disparity during this time. It is therefore possible that conifers were evolving into areas of ecospace formerly occupied by other plant groups that declined at the end of the Permian (Retallack, 1995).

The high initial disparity of many of the plant groups investigated here results from the appearance of a small number of morphologically highly distinct taxa close to the base of each clade. In most of our groups, fossils quickly define the extremes of the empirical envelope as soon as they appear, with subsequent lineages gradually filling the intervening morphospace rather than colonizing more eccentric regions of it. Conifers exhibit a rather different pattern (Fig. 7), with the gradual appearance of sub-clades that each occupy distinct regions of the space (Fig. 8). Rather than rapid morphospace occupation followed by subsequent saturation, conifers appear to show several phases of morphospace colonization and subsequent diversity increase in tightly defined regions centred around pioneers with novel character combinations. This suggests that the evolution of conifers may have been characterized by the intermittent acquisition of novel morphologies or ‘key’ innovations, followed by subsequent diversification. Such events may include the radiation of the pines in the Jurassic and the cypresses in the Cretaceous and early Palaeogene. The high degree of morphospace clustering may result from competition with other groups (such as angiosperms), constraining the available morphospace. However, it is more likely to be a function of greater structural or developmental constraints acting upon suites of characters within the conifer data set (moreover, conifers appear to show relatively tight clustering in the Triassic and Jurassic, prior to the inferred appearance of basal angiosperms). Pines show much weaker clustering than conifers as a whole. Characters within the pine data set (Klymiuk and Stockey, 2012) were derived from cone morphology, strongly implying that Pinacae were able to explore the majority of possible cone forms rapidly and early in their evolution in a relatively unconstrained manner.

Because most of the discrete character matrices analysed here included a broad sample of characters from many different anatomical regions, it is reasonable to assume that the gross morphology of the taxa in the sample was reasonably represented. Our three angiosperm matrices had marked differences in character and taxon composition (Fig. 14), but showed similar overall patterns of disparity through time.

Fig. 14.
Disparity profiles for three cladistic data sets of angiosperms. Disparity (black circles) calculated as the sum of variances on all principal co-ordinate axes within several time bins. Values are the mean of 1000 bootstrap replicates ± s.e. Despite ...

Why are there so few studies of plant disparity?

There are a number of possible reasons why empirical morphospace approaches have been underutilized within the plant sciences, aside from the usual methodological considerations underpinning the choice of data and indices (Rohlf, 1998). Many morphometric approaches entail time-consuming data collection, which may limit tractable sample sizes. There are also difficulties in establishing variable or character sets that can be measured or coded across higher taxa. Most studies therefore focus upon smaller plant clades or else derive data from particular structures (Chartier et al., 2014) rather than investigating overall morphological disparity. Moreover, the often fragmentary nature of fossil material may mean that holistic treatments are impractical, or that many types of morphometric data cannot be obtained (Adams et al., 2004).

Utilizing existing discrete morphological data matrices

New morphological character matrices for plants are becoming increasingly rare (Gottlieb, 1988; Sytsma et al., 1991); mounting evidence from molecular phylogenetics implies that morphological convergence is obfuscating our understanding of plant relationships (Bowe and Coat, 2000; Donoghue and Doyle, 2000; Schneider et al., 2009). However, we believe that morphological character data have important uses beyond that of inferring phylogeny (Thorne et al., 2011); not least for quantifying patterns of disparity change throughout morphologically and taxonomically diverse clades with long evolutionary histories. In this context, the problems of homoplasy and convergence that bedevil phylogenetic inference are less marked, since morphospaces are conceived for a variety of purposes and can be intended to reflect a variety of aspects of evolution. Discrete character morphospaces offer a framework for quantifying patterns of morphological disparity within large clades, but also highlight questions that can be addressed in a more focused manner using other morphometric techniques (Goodman, 2002). More comprehensive analyses of existing plant character matrices would represent an efficient use of legacy data, allow some of the commonalities suggested in this paper to be properly tested and would powerfully complement existing and future morphometric studies.

Despite the abundance of discrete, morphological data in the literature, there are a number of considerations when using explicitly cladistic matrices for disparity studies. Morphological cladists usually seek to resolve phylogeny (Forey et al., 1998), but are not always concerned with representing accurate branch lengths and evolutionary distances. Even in the extreme approach adopted by pattern cladistics, which views the cladistic method as being divorced from evolutionary assumptions of descent through modification (Brady, 1982; Brower, 2000), there is still an imperative to recognize hierarchical groupings within sets of taxa (Hennig, 1966; Estabrook et al., 1975). There may therefore be a tendency to sub-divide morphological variety more finely within taxa that are morphologically conservative overall in order to resolve their relationships or structure. Conversely, taxa supported by long evolutionary branches may be morphologically very distinct from their nearest sampled relatives, but there may be no imperative to quantify all of these differences to the same degree of resolution as in highly diverse and morphologically similar groups. More generally, it is reasonable to expect character matrices to be biased towards distinctive features and/or those which have been demonstrated to be good at distinguishing groups in previous studies. An allied issue is the assumption that all characters should be treated equally. This may not always be desirable, particularly in cases where some groups are characterized by a limited number of highly distinctive and variable characters while others are defined by broader suites of gross morphological features that are nevertheless coded as a single character. For example, it is probably simplistic to treat the presence or absence of sclereids in the leaves on an equal footing with scandent vs. arborescent growth habits (Foster, 1956; Rury and Dickison, 1984). While there are a variety of objective approaches for the differential weighting of characters in phylogenetic studies, these are derived from predictions or empirical estimates of levels of homoplasy or the phylogenetic information content of characters (Farris, 1969; Sharkey, 1989; Goloboff, 1993, 2014). In disparity analyses, what may be required is rather some weighting derived from the ontogenetic priority, developmental (Riedl, 1977; Arthur, 1984, 1988; Wimsatt, 1986) or structural depth (Stebbins, 1969; Pettersson, 2009) of characters, although such weights are notoriously difficult to assign.

Some cladistic matrices are constructed in order to address particular questions, most commonly sequences of character acquisition across important evolutionary transitions; for example, tetrapods from fishes (Wagner and Chiu, 2001; Long and Gordon, 2004; Ruta et al., 2006; Wagner et al., 2006) and birds from dinosaurs (Garner et al., 1999; Xu, 2006; Heers et al., 2014; Brusatte et al., 2014). Such data intentionally focus on the taxa and characters bracketing these changes, with deliberately much sparser sampling outside of this. More generally, outgroup taxa – often included for rooting purposes – are more sparsely sampled than those of the ingroups (Graybeal, 1998; Heath et al., 2008). Morphological cladistic characters may therefore sample morphological variation unevenly across taxa and through time. Not all data sets are suitable for investigating temporal and taxonomic patterns of morphological variation therefore, and many require some form of moderation. Hughes et al. (2013), for example, standardized sampling according to higher taxonomy, and removed outgroups.

One final issue is the inclusion or otherwise of autapomorphic character states; those present in just a single taxon (Yeates, 1992; Bryant, 1995). Such states cannot influence inferred cladistic branching structure, but they do affect branch lengths (without introducing homoplasy) and indices of morphological difference. In two state characters, an autapomophic state renders the entire character cladistically (but not phenetically) uninformative. This property is flagged by most phylogenetic software, which usually results in their removal from cladistic matrices. Autapomorphic states are more likely to be retained in multistate characters (those with three or more states), since the character remains informative overall. More generally, cladists do not actively seek to include autapomorphic states, such that cladistic matrices usually omit this aspect of morphological variation. Empirically, however, the inclusion/exclusion of autapomorphies makes relatively little difference to assessments of morphological variety (Wills, 2001). The precise effect of autapomorphic states will depend upon the overall properties of the data set and the mode of analysis, but in general they merely cause the taxa possessing them to appear marginally more divergent from the overall mean morphology than they would otherwise be.

There is an increasing desire for large, complete phylogenies to underpin various forms of evolutionary and ecological analyses (Guyer and Slowinski, 1993; Phillimore and Freckleton, 2006; Tamura et al., 2012). Large matrices of molecular characters (supermatrices) are frequently assembled de novo using open data resources and automated algorithms (Liu et al., 2001; Davies et al., 2004; Bininda-Emonds, 2004; Davis and Page, 2008). There are no similar repositories or tools for morphological matrices. Assembling large matrices comprising hundreds or thousands of OTUs and characters from first principles would ensure greater consistency, but is hugely time-consuming. Hence, morphological supermatrices are often assembled by amalgamating the largest data sets or synthetic treatments available for constituent groups. However, this approach may entail its own set of problems. The first is alluded to above; the differential sampling of taxa and characters. Taxon sampling can be standardized more readily, but uniform character sampling requires more detailed knowledge and entails greater subjectivity. More problematically, it is often difficult or impossible to code many of the characters in the constituent matrices for the ‘outgroup’ taxa (those represented in the other matrices), thereby resulting in large blocks of inferred plesiomorphies (typically ‘0’ or absent) and inapplicable codings (‘?’). Depending upon the manner in which such inapplicabilities are treated, this phenomenon can result in artificially distinct clusters of taxa, strongly but spuriously demarcated by these discontinuities in knowledge and character sampling (Wilkinson et al., 2005; Cotton et al., 2006). For these reasons, large published cladistic matrices compiled from first-hand observations of specimens (or from careful treatments of the primary literature) have many potential advantages over those assembled by conjoining data from disparate published sources (de Dequeiroz and Gatesy, 2007).


1. The concept of morphological disparity is distinct from those of diversity and species richness (Wills, 2001). Indices of disparity attempt to codify the morphological variety of a sample of taxa, are calculated relative to some set of morphological variables or characters, and often utilize a plot of taxa in a multidimensional morphospace. Morphospaces are abstract spaces in which the geometric distances between taxa are proportional to some measure of the morphological differences between them. The nature of a morphospace is entirely contingent upon the underlying data, the manner in which differences between taxa are summarized as distances, and the methods used to project these distances into an n dimensional space. The precise approach will depend upon the purpose for which the morphospace is intended. It follows that there is no objective morphospace (in the sense that there is an objective phylogeny), and that the dispersion of taxa in different spaces cannot be compared directly (comparisons between sub-groups within the space are possible, but these are necessarily only relative). Morphospaces derived from large samples of characters or variables encompassing most aspects of form are most likely to offer insights into overall morphological variety. Indices of disparity variously assess the relative dispersion of samples of taxa within a morphospace, or provide some distillation of the morphological differences between them.

2. Diversity and disparity appear to be fundamentally decoupled. A significant majority of the animal clades investigated show relatively high disparity early in their evolution (Hughes et al., 2013) at times when diversity is still comparatively low (i.e. there are modest numbers of taxa but these are morphologically highly distinct from each other). The subsequent evolution of such groups often sees an increase in diversity with little or no concomitant increase in disparity; there are increasing numbers of taxa within a restricted number of morphological ‘themes’. Disparity may even decline as diversity is rising, since some of the most speciose clades have particularly constrained bodyplans but are able to partition ecospace and morphospace particularly finely. A substantial minority of animal clades show other patterns, including high initial disparity at low diversity (Foote, 1990).

3. There have been relatively few studies of morphological disparity in plants, and no studies have attempted to assess patterns of overall disparity in major clades through time. Temporal patterns of diversity in plants and animals show significantly different patterns (Knoll et al., 1979), with plants counterintuitively being less affected at times of global mass extinction (Cascales-Miñana and Cleal, 2014). An assessment of patterns of disparity in major plant clades is therefore overdue, and may provide insights into plant macroevolution to complement those being obtained for animals.

4. There are numerous morphometric methods that allow shape and shape change to be quantified across taxa. However, as the morphogical variety of the forms being compared increases (usually in tandem with the taxonomic scope of the study), the ability of such approaches to compare increasingly disparate forms becomes more limited. Discrete character data sets have certain advantages in this context. There are rich resources of discrete character matrices already available for numerous plant clades, and, although initially intended for inferring phylogeny, these data sets can be repurposed for disparity studies within certain strictures.

5. Our preliminary disparity analyses for six exemplary plant clades demonstrate that initial levels of disparity are usually high, if not indistinguishable from (or at) the maximum ultimately achieved by the group. Most regions of the morphospace are colonized early in the history of each plant clade, with subsequent evolution serving merely to increase diversity within these regions. The notable exception are the conifers, in which sub-clades appear intermittently, and progressively colonize distinct regions of the space. This results in conifer disparity increasing incrementally over the first half of the group’s history. All of our exemplary plant clades have disparity profile shapes with a centre of gravity higher than the intrinsic null (significantly so in all except for two angiosperm data sets). This is unsurprising, however, since all are extant groups, with profiles truncated by the Recent (Hughes et al., 2013). Combining detailed empirical morphometric studies of specific anatomical regions with the more holistic approach illustrated here will probably be reciprocally illuminating, and offer insights into plant macroevolution.


Supplementary data are available online at and consist of the following. Table S1: list of published disparity studies. Table S2: distribution of characters in sampled plant data sets. Figure S1: comparison between actual (‘total’) and sampled diversity curves.

Supplementary Data:


This paper evolved from a presentation by M.A.W. at Euro Evo Devo 2014 in Vienna, and we thank Rainer Melzer and Günter Theißen for their invitation to speak at their symposium. We are grateful to the authors who generated the cladistic matrices repurposed and reanalysed herein. This work was generously supported by grant no. 43915 from the John Templeton Foundation (which funds J.W.O.’s PhD work), and by the Leverhulme Trust grant F/00 351/Z to M.A.W. (which underpinned our work on disparity). J.W.O. wrote the manuscript, collated plant morphological and stratigraphic data sets, drafted figures and analysed data. M.H. ran disparity analyses and statistics, summarized the disparity literature and drafted figures. S.G. produced clustering plots. M.A.W. conceived and designed the study, wrote the manuscript and drafted one figure.


  • Adams DC, Rohlf FJ, Slice DE. 2004. Geometric morphometrics: ten years of progress following the ‘revolution’. Italian Journal of Zoology 71: 5–16.
  • Agrawal AA. 2007. Macroevolution of plant defense strategies. Trends in Ecology and Evolution 22: 103–109. [PubMed]
  • Allen CE, Beldade P, Zwaan BJ, Brakefield PM. 2008. Differences in the selection response of serially repeated color pattern characters: standing variation, development, and evolution. BMC Evolutionary Biology 8: 94. [PMC free article] [PubMed]
  • Anderson PSL, Friedman M, Brazeau MD, Rayfield EJ. 2011. Initial radiation of jaws demonstrated stability despite faunal and environmental change. Nature 476: 206–209. [PubMed]
  • Anderson PSL, Friedman M, Ruta M. 2013. Late to the table: diversification of tetrapod mandibular biomechanics lagged behind the evolution of terrestriality. Integrative and Comparative Biology 53: 197–208. [PubMed]
  • Arthur W. 1984. Mechanisms of morphological evolution. New York: Wiley.
  • Arthur W. 1988. A theory of the evolution of development. New York: Wiley.
  • Baker WJ, Savolainen V, Asmussen-Lange CB, et al. 2009. Complete generic-level phylogenetic analyses of palms (Arecaceae) with comparisons of supertree and supermatrix approaches. Systematic Biology 58: 240–256. [PubMed]
  • Bambach RK. 2006. Phanerozoic biodiversity mass extinctions. Annual Review of Earth and Planetary Sciences 34: 127–155.
  • Barthélémy D, Caraglio Y. 2007. Plant architecture: a dynamic, multilevel and comprehensive approach to plant form, structure and ontogeny. Annals of Botany 99: 375–407. [PMC free article] [PubMed]
  • Bateman RM, Crane PR, DiMichele WA, et al. 1998. Early evolution of land plants: phylogeny, physiology, and ecology of the primary terrestrial radiation. Annual Review of Ecology and Systematics 29: 263–292.
  • Beerling D, Berner R. 2005. Feedbacks and the coevolution of plants and atmospheric CO2. Proceedings of the National Academy of Sciences, USA 102: 1302–1305. [PubMed]
  • Beerling D, Osborne C, Chaloner W. 2001. Evolution of leaf-form in land plants linked to atmospheric CO2 decline in the Late Palaeozoic era. Nature 410: 352–354. [PubMed]
  • Benson RBJ, Evans M, Druckenmiller PS. 2012. High diversity, low disparity and small body size in plesiosaurs (Reptilia, Sauropterygia) from the Triassic–Jurassic boundary. PLoS One 7: e31838. [PMC free article] [PubMed]
  • Benton MJ. 1993. The fossil record 2. London: Chapman & Hall.
  • Benton MJ. 2009. The Red Queen and the Court Jester: species diversity and the role of biotic and abiotic factors through time. Science 323: 728–732. [PubMed]
  • Benzing DH. 2000. Bromeliaceae: profile of an adaptive radiation. Cambridge: Cambridge University Press.
  • Bininda-Emonds ORP. 2004. The evolution of supertrees. Trends in Ecology and Evolution 19: 315–322. [PubMed]
  • Bocxlaer B Van, Schultheiß R. 2010. Comparison of morphometric techniques for shapes with few homologous landmarks based on machine-learning approaches to biological discrimination. Paleobiology 36: 497–515.
  • Bookstein FL. 1997. Landmark methods for forms without landmarks: morphometrics of group differences in outline shape. Medical Image Analysis 1: 225–243. [PubMed]
  • Borsch T, Löhne C, Wiersema J. 2008. Phylogeny and evolutionary patterns in Nymphaeales: integrating genes, genomes and morphology. Taxon 57: 1052–1081.
  • Bowe L, Coat G. 2000. Phylogeny of seed plants based on all three genomic compartments: extant gymnosperms are monophyletic and Gnetales’ closest relatives are conifers. Proceedings of the National Academy of Sciences, USA 97: 4092–4097. [PubMed]
  • Boyce CK. 2005. Patterns of segregation and convergence in the evolution of fern and seed plant leaf morphologies. Paleobiology 31: 117–140.
  • Boyce CK, Knoll AH. 2002. Evolution of developmental potential and the multiple independent origins of leaves in Paleozoic vascular plants. Paleobiology 28: 70–100.
  • Boyce CK, Brodribb TJ, Feild TS, Zwieniecki MA. 2009. Angiosperm leaf vein evolution was physiologically and environmentally transformative. Proceedings of the Royal Society B: Biological Sciences 276: 1771–1776. [PMC free article] [PubMed]
  • Bradshaw AD. 2006. Unravelling phenotypic plasticity – why should we bother? New Phytologist 170: 644–648. [PubMed]
  • Brady RH. 1982. Theoretical issues and ‘pattern cladistics’. Systematic Zoology 31: 286–291.
  • Briggs DEG, Fortey RA, Wills MA. 1992. Morphological disparity in the Cambrian. Science 256: 1670–1673. [PubMed]
  • Brower AVZ. 2000. Evolution is not a necessary assumption of cladistics. Cladistics 16: 143–154.
  • Brusatte SL, Lloyd GT, Wang SC, Norell MA. 2014. Gradual assembly of avian body plan culminated in rapid rates of evolution across the dinosaur–bird transition. Current Biology 24: 2386–2392. [PubMed]
  • Bryant HN. 1995. Why autapomorphies should be removed. Cladistics 11: 381–384.
  • Burns JH, Munguia P, Nomann BE, Braun SJ, Terhorst CP, Miller TE. 2008. Vegetative morphology and trait correlations in 54 species of Commelinaceae. Botanical Journal of the Linnean Society 158: 257–268.
  • Butler AB, Saidel WM. 2000. Defining sameness: historical, biological, and generative homology. BioEssays 22: 846–853. [PubMed]
  • Cailliez F. 1983. The analytical solution of the additive constant problem. Psychometrika 48: 305–308.
  • Cascales-Miñana B, Cleal CJ. 2012. Plant fossil record and survival analyses. Lethaia 45: 71–82.
  • Cascales-Miñana B, Cleal CJ. 2014. The plant fossil record reflects just two great extinction events. Terra Nova 26: 195–200.
  • Cerling TE, Harris JM, MacFadden BJ. 1997. Global vegetation change through the Miocene/Pliocene boundary. Nature 389: 153–158.
  • Chapman R, Rasskin-Gutman D. 2001. Quantifying morphology. In: Briggs E, Crowther P, editors. , eds. Palaeobiology II. Malden: Blackwell Science, 489–492.
  • Chartier M, Jabbour F, Gerber S, et al. 2014. The floral morphospace – a modern comparative approach to study angiosperm evolution. New Phytologist 204: 841–853. [PubMed]
  • Chitwood DH, Ranjan A, Martinez CC, et al. 2014. A modern ampelography: a genetic basis for leaf shape and venation patterning in grape. Plant Physiology 164: 259–272. [PubMed]
  • Conway Morris S. 1989. Burgess Shale faunas and the Cambrian explosion. Science 246: 339–346. [PubMed]
  • Cooper WJ, Parsons K, McIntyre A, Kern B, McGee-Moore A, Albertson RC. 2010. Bentho-pelagic divergence of cichlid feeding architecture was prodigious and consistent during multiple adaptive radiations within African rift-lakes. PLoS One 5: e9551. [PMC free article] [PubMed]
  • Cotton JA, Slater CSC, Wilkinson M. 2006. Discriminating supported and unsupported relationships in supertrees using triplets. Systematic Biology 55: 345–350. [PubMed]
  • von Cramon-Taubadel N Von, Frazier BC, Lahr MM. 2007. The problem of assessing landmark error in geometric morphometrics: theory, methods, and modifications. American Journal of Physical Anthropology 35: 24–35. [PubMed]
  • Crampton JS. 1995. Elliptic Fourier shape analysis of fossil bivalves: some practical considerations. Lethaia 28: 179–186.
  • Crane PR, Friis EM, Pedersen KR. 1994. Paleobotanical evidence on the early radiation of magnoliid angiosperms. Plant Systematics and Evolution 8: 51–72 (suppl.).
  • Crisp M, Cook L. 2011. Cenozoic extinctions account for the low diversity of extant gymnosperms compared with angiosperms. New Phytologist 192: 997–1009. [PubMed]
  • Davies TJ, Barraclough TG, Chase MW, Soltis PS, Soltis DE, Savolainen V. 2004. Darwin’s abominable mystery: insights from a supertree of the angiosperms. Proceedings of the National Academy of Sciences, USA 101: 1904–1909. [PubMed]
  • Davis KE, Page RDM. 2008. Reweaving the tapestry: a supertree of birds. PLoS Currents 6: doi: 10.1371/currents.tol.c1af68dda7c999ed9f1e4b2d2df7a08e. [PMC free article] [PubMed]
  • De Dequeiroz A, Gatesy J. 2007. The supermatrix approach to systematics. Trends in Ecology and Evolution 22: 34–41. [PubMed]
  • Donoghue M. 2008. A phylogenetic perspective on the distribution of plant diversity. Proceedings of the National Academy of Sciences, USA 105: 11549–11555. [PubMed]
  • Donoghue MJ. 2009. Key innovations, convergence, and success: macroevolutionary lessons from plant phylogeny. Paleobiology 31: 77–93.
  • Donoghue MJ, Doyle JA. 2000. Seed plant phylogeny: demise of the anthophyte hypothesis? Current Biology 10: 106–109. [PubMed]
  • Doyle JA, Endress PK. 2000. Morphological phylogenetic analysis of basal angiosperms: comparison and combination with molecular data. International Journal of Plant Sciences 161: S121–S153.
  • Doyle JA, Endress PK. 2014. Integrating Early Cretaceous fossils into the phylogeny of living angiosperms: ANITA lines and relatives of Chloranthaceae. International Journal of Plant Sciences 175: 555–600.
  • Doyle JA, Donoghue MJ, Zimmer EA. 1994. Integration of morphological and ribosomal RNA data on the origin of angiosperms. Annals of the Missouri Botanical Garden 81: 419–450.
  • Drake AG, Klingenberg CP. 2010. Large-scale diversification of skull shape in domestic dogs: disparity and modularity. American Naturalist 175: 289–301. [PubMed]
  • Ehleringer JR, Sage RF. 1991. Climate change and the evolution of C4 photosynthesis. Trends in Ecology and Evolution 6: 95–99. [PubMed]
  • Endress PK. 2001. Origins of flower morphology. Journal of Experimental Zoology 115: 105–115. [PubMed]
  • Erwin TL. 1997. Biodiversity at its utmost: tropical forest beetles. In: Reaka-Kudla ML, Wilson DE, Wilson EO, editors. , eds. Biodiversity II: understanding and protecting our biological resources. Joseph Henry Press, 27–40.
  • Estabrook G, Johnson CS, Morris FM. 1975. An idealized concept of the true cladistic character. Mathematical Biosciences 212: 263–272.
  • Ezard THG, Aze T, Pearson PN, Purvis A. 2011. Interplay between changing climate and species’ ecology drives macroevolutionary dynamics. Science 25: 16–19. [PubMed]
  • Farnsworth KD, Niklas KJ. 1995. Theories of optimization, form and function in branching architecture in plants. Functional Ecology 9: 355–363.
  • Farris J. 1969. A successive approximations approach to character weighting. Systematic Biology 18: 374–385.
  • Feild TS, Arens NC. 2007. The ecophysiology of early angiosperms. Plant, Cell and Environment 30: 291–309. [PubMed]
  • Feild TS, Brodribb TJ, Iglesias A, et al. 2011. Fossil evidence for Cretaceous escalation in angiosperm leaf vein evolution. Proceedings of the National Academy of Sciences, USA 108: 8363–8366. [PubMed]
  • Figueirido B, MacLeod N, Krieger J, De Renzi M, Pérez-Claros JA, Palmqvist P. 2011. Constraint and adaptation in the evolution of carnivoran skull shape. Paleobiology 37: 490–518.
  • Foote M. 1989. Perimeter-based Fourier analysis: a new morphometric method applied to the trilobite cranidium. Journal of Paleontology 63: 880–885.
  • Foote M. 1990. Nearest-neighbor analysis of trilobite morphospace. Systematic Biology 39: 371–382.
  • Foote M. 1991. Morphological and taxonomic diversity in clade’s history: the blastoid record and stochastic simulations. Contributions from the Museum of Paleontology 28: 101–140.
  • Foote M. 1992. Paleozoic record of morphological diversity in blastozoan echinoderms. Proceedings of the National Academy of Sciences, USA 89: 7325–7329. [PubMed]
  • Foote M. 1994. Morphological disparity in Ordovician–Devonian crinoids and the early saturation of morphological space. Paleobiology 20: 320–344.
  • Foote M. 1997. The evolution of morphological diversity. Annual Review of Ecology and Systematics 28: 129–152.
  • Foote M. 1999. Morphological diversity in the evolutionary radiation of Paleozoic and post-Paleozoic crinoids. Paleobiology 25: 1–115.
  • Forey PL, Humphries CJ, Williams DM. 1998. Cladistics: the theory and practice of parsimony analysis. Oxford: Oxford University Press.
  • Fortey RA, Briggs DEG, Wills MA. 1996. The Cambrian evolutionary ‘explosion’: decoupling cladogenesis from morphological disparity. Biological Journal of the Linnean Society 57: 13–33.
  • Fortey R, Briggs D, Wills M. 1997. The Cambrian evolutionary ‘explosion’ recalibrated. BioEssays 19: 429–434.
  • Foster AS. 1956. Plant idioblasts remarkable examples of cell specialization comparative morphology of idioblasts. Protoplasma 46: 184–193.
  • Friedman M. 2010. Explosive morphological diversification of spiny-finned teleost fishes in the aftermath of the end-Cretaceous extinction. Proceedings of the Royal Society B: Biological Sciences 277: 1675–1683. [PMC free article] [PubMed]
  • Friedman WE, Williams JH. 2003. Modularity of the angiosperm female gametophyte and its bearing on the early evolution of endosperm in flowering plants. Evolution 57: 216–230. [PubMed]
  • Futuyma DJ, Agrawal AA. 2009. Macroevolution and the biological diversity of plants and herbivores. Proceedings of the National Academy of Sciences, USA 106: 18054–18061. [PubMed]
  • Garner JP, Taylor GK, Thomas ALR. 1999. On the origins of birds: the sequence of character acquisition in the evolution of avian flight. Proceedings of the Royal Society B: Biological Sciences 266: 1259–1266.
  • Gavrilets S. 1999. Dynamics of clade diversification on the morphological hypercube. Proceedings of the Royal Society B: Biological Sciences 266: 817–824.
  • Goloboff PA. 1993. Estimating character weights during tree search. Cladistics 9: 83–91.
  • Goloboff PA. 2014. Extended implied weighting. Cladistics 30: 260–272.
  • Goodman N. 2002. Biological data becomes computer literate: new advances in bioinformatics. Current Opinion in Biotechnology 13: 68–71. [PubMed]
  • Gotelli NJ, Colwell RK. 2001. Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecology Letters 4: 379–391.
  • Gottlieb LD. 1988. Towards molecular genetics in Clarkia: gene duplications and molecular characterization of PGI genes. Annals of the Missouri Botanical Garden 75: 1169–1179.
  • Gould SJ. 1989. Wonderful life. New York: WW Norton.
  • Gould SJ. 1991. The disparity of the Burgess Shale arthropod fauna and the limits of cladistic analysis: why we must strive to quantify morphospace. Paleobiology 17: 411–423.
  • Gould S, Gilinsky N, German R. 1987. Asymmetry of lineages and the direction of evolutionary time. Science 236: 1437–1441. [PubMed]
  • Gradstein FM, Ogg JG, Smith AG. 2004. A geologic time scale. Cambridge: Cambridge University Press.
  • Graybeal A. 1998. Is it better to add taxa or characters to a difficult phylogenetic problem? Systematic Biology 47: 9–17. [PubMed]
  • Guyer C, Slowinski JB. 1993. Adaptive radiation and the topology of large phylogenies. Evolution 47: 253–263.
  • Hallam Am Wignall PB. 1997. Mass extinctions and their aftermath. Oxford: Oxford University Press.
  • Hammer Ø, Bucher H. 2005. Models for the morphogenesis of the molluscan shell. Lethaia 38: 111–122.
  • Hart JA. 1987. A cladistic analysis of conifers: preliminary results. Journal of the Arnold Arboretum 68: 269–307.
  • Heath TA, Hedtke SM, Hillis DM. 2008. Taxon sampling and the accuracy of phylogenetic analyses. Journal of Systematics and Evolution 46: 239–257.
  • Heers AM, Dial KP, Tobalske BW. 2014. From baby birds to feathered dinosaurs: incipient wings and the evolution of flight. Paleobiology 40: 459–476.
  • Hennig W. 1966. Phylogenetic systematics. Urbana, IL: University of Illinois Press.
  • Hughes M, Gerber S, Wills MA. 2013. Clades reach highest morphological disparity early in their evolution. Proceedings of the National Academy of Sciences, USA 110: 13875–13879. [PubMed]
  • Jeune B, Barabé D, Lacroix C. 2006. Classical and dynamic morphology: toward a synthesis through the space of forms. Acta Biotheoretica 54: 277–293. [PubMed]
  • Jones D, Purnell MA, von Bitter PH. 2009. Morphological criteria for recognising homology in isolated skeletal elements: comparison of traditional and morphometric approaches in conodonts. Palaeontology 52: 1243–1256.
  • Kelly S, Grenyer R, Scotland RW. 2014. Phylogenetic trees do not reliably predict feature diversity. Diversity and Distributions 20: 600–612.
  • Kenrick P, Crane P. 1997. The origin and early evolution of plants on land. Nature 389: 33–39.
  • Klingenberg CP, Leamy LJ, Cheverud JM. 2004. Integration and modularity of quantitative trait locus effects on geometric shape in the mouse mandible. Genetics 166: 1909–1921. [PubMed]
  • Klingenberg CP, Duttke S, Whelan S, Kim M. 2012. Developmental plasticity, morphological variation and evolvability: a multilevel analysis of morphometric integration in the shape of compound leaves. Journal of Evolutionary Biology 25: 115–129. [PubMed]
  • Klymiuk AA, Stockey RA. 2012. A lower Cretaceous (Valanginian) seed cone provides the earliest fossil record for Picea (Pinaceae). American Journal of Botany 99: 1069–1082. [PubMed]
  • Knoll AH, Niklas KJ, Tiffney BH. 1979. Phanerozoic land-plant diversity in North America. Science 206: 1400–1402. [PubMed]
  • Kovach WL, Batten DJ. 1993. Diversity changes in lycopsid and aquatic fern megaspores through geologic time. Paleobiology 19: 28–42.
  • Lacroix C, Jeune B, Purcell-Macdonald S. 2003. Shoot and compound leaf comparisons in eudicots: dynamic morphology as an alternative approach. Botanical Journal of the Linnean Society 143: 219–230.
  • Langlade NB, Feng X, Dransfield T, et al. 2005. Evolution through genetically controlled allometry space. Proceedings of the National Academy of Sciences, USA 102: 10221–10226. [PubMed]
  • Lidgard S, Crane PR. 1990. Angiosperm diversification and Cretaceous floristic trends: a comparison of palynofloras and leaf macrofloras. Paleobiology 16: 77–93.
  • Lindbladh M. 2002. Morphometric analysis of pollen grains for paleoecological studies: classification of Picea from eastern North America. American Journal of Botany 89: 1459–1467. [PubMed]
  • Lingham-Soliar T. 2005. Caudal fin allometry in the white shark Carcharodon carcharias: implications for locomotory performance and ecology. Naturwissenschaften 92: 231–236. [PubMed]
  • Lingham-Soliar T, Plodowski G. 2007. Taphonomic evidence for high-speed adapted fins in thunniform ichthyosaurs. Naturwissenschaften 94: 65–70. [PubMed]
  • Liu F-GR, Miyamoto MM, Freire NP, et al. 2001. Molecular and morphological supertrees for eutherian (placental) mammals. Science 291: 1786–1789. [PubMed]
  • Lockwood CA, Lynch JM, Kimbel WH. 2002. Quantifying temporal bone morphology of great apes and humans: an approach using geometric morphometrics. Journal of Anatomy 201: 447–464. [PubMed]
  • Long JA, Gordon MS. 2004. The greatest step in vertebrate history: a paleobiological review of the fish tetrapod transition. Physiological and Biochemical Zoology 77: 700–719. [PubMed]
  • Love AC. 2007. Functional homology and homology of function: biological concepts and philosophical consequences. Biology and Philosophy 22: 691–708.
  • Lupia R. 1999. Discordant morphological disparity and taxonomic diversity during the Cretaceous angiosperm radiation: North American pollen record. Paleobiology 25: 1–28.
  • MacLeod N. 2011. Semilandmarks and radial Fourier analysis. Palaeontological Association Newsletter 76: 25–42.
  • Magallón S, Sanderson MJ. 2001. Absolute diversification rates in angiosperm clades. Evolution 55: 1762–1780. [PubMed]
  • Marcus LF. 2000. Application of landmark morphometrics to skulls representing the orders of living mammals. Hystrix 11: 27–47.
  • Mayhew PJ. 2007. Why are there so many insect species? Perspectives from fossils and phylogenies. Biological Reviews of the Cambridge Philosophical Society 82: 425–454. [PubMed]
  • Mayhew PJ, Bell MA, Benton TG, McGowan AJ. 2012. Biodiversity tracks temperature over time. Proceedings of the National Academy of Sciences, USA 109: 15141–15145. [PubMed]
  • McElwain JC, Beerling DJ, Woodward FI. 1999. Fossil plants and global warming at the Triassic–Jurassic boundary. Science 285: 1386–1390. [PubMed]
  • McElwain JC, Punyasena SW. 2007. Mass extinction events and the plant fossil record. Trends in Ecology and Evolution 22: 548–557. [PubMed]
  • McGhee GR. 1999. Theoretical morphology: the concept and its applications. Columbia University Press.
  • Miller JS, Venable DL. 2003. Floral morphometrics and the evolution of sexual dimorphism in Lycium (Solanaceae). Evolution 57: 74–86. [PubMed]
  • Mitteroecker P, Gunz P. 2009. Advances in geometric morphometrics. Evolutionary Biology 36: 235–247.
  • Mitteroecker P, Huttegger SM. 2009. The concept of morphospaces in evolutionary and developmental biology: mathematics and metaphors. Biological Theory 4: 54–67.
  • Monteiro LR, Nogueira MR. 2010. Adaptive radiations, ecological specialization, and the evolutionary integration of complex morphological structures. Evolution 64: 724–744. [PubMed]
  • Moore KJ, Moser LE. 1995. Quantifying developmental morphology of perennial grasses. Crop Science 35: 37–43.
  • Motani R. 2002. Scaling effects in caudal fin propulsion and the speed of ichthyosaurs. Nature 415: 309–312. [PubMed]
  • Moyne S, Neige P. 2007. The space–time relationship of taxonomic diversity and morphological disparity in the Middle Jurassic ammonite radiation. Palaeogeography, Palaeoclimatology, Palaeoecology 248: 82–95.
  • Nandi OI, Chase MW, Endress PK. 1998. A combined cladistic analysis of angiosperms using rbcL and non-molecular data sets. Annals of the Missouri Botanical Garden 85: 137–214.
  • Neige P. 2006. Morphometrics of hard structures in cuttlefish. Vie et Milieu 56: 121–127.
  • Niklas KJ. 1999. Evolutionary walks through a land plant morphospace. Journal of Experimental Botany 50: 39–52.
  • Niklas KJ, Kerchner V. 1984. Mechanical and photosynthetic constraints on the evolution of plant shape. Paleobiology 10: 79–101.
  • Niklas KJ, Tiffney BH. 1994. The quantification of plant biodiversity through time. Philosophical Transactions of the Royal Society B: Biological Sciences 345: 35–44.
  • Niklas KJ, Tiffney BH, Knoll AH. 1980. Apparent changes in the diversity of fossil plants. a preliminary assessment. Evolutionary Biology 12: 1–89.
  • Nogueira MR, Peracchi AL, Monteiro LR. 2009. Morphological correlates of bite force and diet in the skull and mandible of phyllostomid bats. Functional Ecology 23: 715–723.
  • Ogg JG, Ogg G, Gradstein FM. 2008. Concise geologic time scale. Cambridge: Cambridge University Press.
  • O’Higgins P, Cobb SN, Fitton LC, et al. 2011. Combining geometric morphometrics and functional simulation: an emerging toolkit for virtual functional analyses. Journal of Anatomy 218: 3–15. [PubMed]
  • Peet RK. 1974. The measurement of species diversity. Annual Review of Ecology and Systematics 5: 285–307.
  • Perez SI, Bernal V, Gonzalez PN. 2006. Differences between sliding semi-landmark methods in geometric morphometrics, with an application to human craniofacial and dental variation. Journal of Anatomy 208: 769–784. [PubMed]
  • Pettersson M. 2009. Complexity and evolution. Cambridge: Cambridge University Press.
  • Phillimore AB, Freckleton RP. 2006. Ecology predicts large-scale patterns of phylogenetic diversification in birds. American Naturalist 168: 220–229. [PubMed]
  • Pinna MGG. 1991. Concepts and tests of homology in the cladistic paradigm. Cladistics 7: 367–394.
  • Pretorius E, Scholtz CH. 2001. Geometric morphometrics and the analysis of higher taxa: a case study based on the metendosternite of the Scarabaeoidea (Coleoptera). Biological Journal of the Linnean Society 74: 35–50.
  • Prusinkiewicz P, Erasmus Y, Lane B. 2007. Evolution and development of inflorescence architectures. Science 316: 1452–1457. [PubMed]
  • Pryer KM, Smith AR, Skog JE. 1995. Phylogenetic relationships of extant ferns based on evidence from morphology and rbcL sequences. American Fern Journal 85: 205–282.
  • Rabosky DL, Slater GJ, Alfaro ME. 2012. Clade age and species richness are decoupled across the eukaryotic tree of life. PLoS Biology 10: e1001381. [PMC free article] [PubMed]
  • Raup DM, Michelson A. 1965. Theoretical morphology of the coiled shell. Science 147: 1–2. [PubMed]
  • R Core Team. 2013. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL
  • Rees PM. 2002. Land-plant diversity and the end-Permian mass extinction. Geology 30: 827–830.
  • Renzaglia KS, Duff RJ, Nickrent DL, Garbary DJ. 2000. Vegetative and reproductive innovations of early land plants: implications for a unified phylogeny. Philosophical Transactions of the Royal Society B: Biological Ssciences 355: 769–793. [PMC free article] [PubMed]
  • Ressayre A, Godelle B. 2002. Aperture pattern ontogeny in angiosperms. Journal of Experimental Zoology 135: 122–135. [PubMed]
  • Retallack GJ. 1995. Permian–Triassic life crisis on land. Science 267: 77–80. [PubMed]
  • Riedl R. 1977. A systems-analytical approach to macro-evolutionary phenomena. Quarterly Review of Biology 52: 351–370. [PubMed]
  • Rohlf FJ. 1998. On applications of geometric morphometrics to studies of ontogeny and phylogeny. Systematic Biology 47: 147–158. [PubMed]
  • Rohlf FJ. 2002. Geometric morphometrics and phylogeny. In: MacLeod N, Forey PL, editors. , eds. Morphology, shape and phylogeny. CRC Press, 175–193.
  • Rohlf FJ, Archie JW. 1984. A comparison of Fourier methods for the description of wing shape in mosquitoes (Diptera: Culicidae). Systematic Biology 33: 302–317.
  • Rohlf FJ, Marcus LF. 1993. A revolution in morphometrics. Trends in Ecology and Evolution 8: 129–132. [PubMed]
  • Rudall PJ, Bateman RM. 2007. Developmental bases for key innovations in the seed-plant microgametophyte. Trends in Plant Science 12: 317–326. [PubMed]
  • Rury P, Dickison W. 1984. Structural correlations among wood, leaves and plant habit. Contemporary problems in plant anatomy. Orlando, FL: Academic Press, 495–540.
  • Ruta M, Wagner PJ, Coates MI. 2006. Evolutionary patterns in early tetrapods. I. Rapid initial diversification followed by decrease in rates of character change. Proceedings of the Royal Society B: Biological Sciences 273: 2107–2111. [PMC free article] [PubMed]
  • Ruvinsky I, Gibson-Brown JJ. 2000. Genetic and developmental bases of serial homology in vertebrate limb evolution. Development 127: 5233–5244. [PubMed]
  • Sajo MG, Rudall PJ, Prychid CJ. 2004. Floral anatomy of Bromeliaceae, with particular reference to the evolution of epigyny and septal nectaries in commelinid monocots. Plant Systematics and Evolution 247: 215–231.
  • Sanderson MJ, Donoghue MJ. 1989. Patterns of variation in levels of homoplasy. Evolution 43: 1781–1795.
  • Sanderson MJ, Donoghue MJ. 1994. Shifts in diversification rate with the origin of angiosperms. Science 264: 1590–1592. [PubMed]
  • Schlichting CD. 1986. The evolution of phenotypic plasticity in plants. Annual Review of Ecology and Systematics 17: 667–693.
  • Schlichting CD. 2002. Phenotypic plasticity in plants. Plant Species Biology 17: 85–88.
  • Schneider H, Schuettpelz E, Pryer KM, Cranfill R, Magallón S, Lupia R. 2004. Ferns diversified in the shadow of angiosperms. Nature 428: 553–557. [PubMed]
  • Schneider H, Smith AR, Pryer KM. 2009. Is morphology really at odds with molecules in estimating fern phylogeny? Systematic Botany 34: 455–475.
  • Sepkoski JJ. 1997. Biodiversity: past, present, and future. Journal of Paleontology 71: 533–539. [PubMed]
  • Sepkoski JJ, Miller AI. 1998. Analysing diversification through time. Trends in Ecology and Evolution 13: 158–159. [PubMed]
  • Sepkoski JJ, Bambach RK, Raup DM, Valentine JW. 1981. Phanerozoic marine diversity and the fossil record. Nature 293: 435–437.
  • Sharkey MJ. 1989. A hypothesis-independent method of character weighting for cladistic analysis. Cladistics 5: 63–86.
  • Shubin N, Tabin C, Carroll S. 2009. Deep homology and the origins of evolutionary novelty. Nature 457: 818–823. [PubMed]
  • Silva IA, Batalha MA. 2011. Plant functional types in Brazilian savannas: the niche partitioning between herbaceous and woody species. Perspectives in Plant Ecology, Evolution and Systematics 13: 201–206.
  • Skalak R, Farrow D, Hoger A. 1997. Kinematics of surface growth. Journal of Mathematical Biology 35: 869–907. [PubMed]
  • Soltis P, Soltis D. 2004. The origin and diversification of angiosperms. American Journal of Botany 91: 1614–1626. [PubMed]
  • Stebbins GL. 1951. Natural selection and the differentiation of angiosperm families. Evolution 5: 299–324.
  • Stebbins GL. 1969. The basis of progressive evolution. Chapel Hill, NC: University of North Carolina Press.
  • Stournaras KE, Lo E, Bohning-Gaese K, et al. 2013. How colorful are fruits? Limited color diversity in fleshy fruits on local and global scales. New Phytologist 198: 617–629. [PubMed]
  • Sytsma KJ, Smith JF, Berry PE. 1991. The use of chloroplast DNA to assess biogeography and evolution of morphology, breeding systems, and flavonoids in Fuchsia sect, Skinnera (Onagraceae). Systematic Botany 16: 257–269.
  • Tamura K, Battistuzzi FU, Billing-Ross P, Murillo O, Filipski A, Kumar S. 2012. Adaptive radiation and the topology of large phylogenies. Proceedings of the National Academy of Sciences, USA 109: 19333–19338. [PubMed]
  • Thorne PM, Ruta M, Benton MJ. 2011. Resetting the evolution of marine reptiles at the Triassic–Jurassic boundary. Proceedings of the National Academy of Sciences, USA 108: 8339–8344. [PubMed]
  • Uhen M. 1996. An evaluation of clade-shape statistics using simulations and extinct families of mammals. Paleobiology 22: 8–22.
  • Urdy S, Goudemand N, Bucher H, Chirat R. 2010. Allometries and the morphogenesis of the molluscan shell: a quantitative and theoretical model. Journal of Experimental Zoology. Part B, Molecular and Developmental Evolution 314: 280–302. [PubMed]
  • Vamosi JC, Vamosi SM. 2010. Key innovations within a geographical context in flowering plants: towards resolving Darwin’s abominable mystery. Ecology Letters 13: 1270–1279. [PubMed]
  • Wagner GP, Chiu C. 2001. The tetrapod limb: a hypothesis on its origin. Journal of Experimental Zoology 291: 226–240. [PubMed]
  • Wagner PJ, Ruta M, Coates MI. 2006. Evolutionary patterns in early tetrapods. II. Differing constraints on available character space among clades. Proceedings of the Royal Society B: Biological Sciences 273: 2113–2118. [PMC free article] [PubMed]
  • Webster M, Zeldtich ML. 2008. Integration and regulation of developmental systems in trilobites. In Rábano I, Gozalo R, García-Bellido D, editors. , eds. Advances in trilobite research. Cuadernos del Museo Geominero 9. Instituto Geológico y Minero de España, Madrid, 427–433
  • Wellman CH. 2014. The nature and evolutionary relationships of the earliest land plants. New Phytologist 202: 1–3. [PubMed]
  • Whibley AC, Langlade NB, Andalo C, et al. 2006. Evolutionary paths underlying flower color variation in Antirrhinum. Science 313: 963–966. [PubMed]
  • Whittington HB. 1985. The Burgess Shale. New Haven, CT: Yale University Press.
  • Wilkinson M, Pisani D, Cotton JA, Corfe I. 2005. Measuring support and finding unsupported relationships in supertrees. Systematic Biology 54: 823–831. [PubMed]
  • Willis KJ, McElwain JC. 2013. The evolution of plants. Oxford: Oxford University Press.
  • Willis KJ, Niklas KJ. 2004. The role of Quaternary environmental change in plant macroevolution: the exception or the rule? Philosophical Transactions of the Royal Society B: Biological Sciences 359: 159–172. [PMC free article] [PubMed]
  • Wills MA. 1998. Crustacean disparity through the Phanerozoic: comparing morphological and stratigraphic data. Biological Journal of the Linnean Society 65: 455–500.
  • Wills MA. 2001. Disparity vs. diversity. In: Briggs DEG, Crowther PR, editors. , eds. Palaeobiology II. Oxford: Blackwell Science, 495–500.
  • Wills MA, Briggs DEG, Fortey RA. 1994. Disparity as an evolutionary index: a comparison of Cambrian and Recent arthropods. Paleobiology 20: 93–130.
  • Wills MA, Gerber S, Ruta M, Hughes M. 2012. The disparity of priapulid, archaeopriapulid and palaeoscolecid worms in the light of new data. Journal of Evolutionary Biology 25: 2056–2076. [PubMed]
  • Wilson JP, Knoll AH. 2010. A physiologically explicit morphospace for tracheid-based water transport in modern and extinct seed plants. Paleobiology 36: 335–355.
  • Wimsatt W. 1986. Developmental constraints, generative entrenchment, and the innate–acquired distinction. In: Bechtel W, editor. , ed. Integrating scientific disciplines. Dordrecht: Martinus-Nijhof, 185–208.
  • Xu X. 2006. Feathered dinosaurs from China and the evolution of major avian characters. Integrative Zoology 1: 4–11. [PubMed]
  • Yeates D. 1992. Why remove autapomorphies? Cladistics 8: 387–389.
  • Yoo M-J, Bell CD, Soltis PS, Soltis DE. 2005. Divergence times and historical biogeography of Nymphaeales. Systematic Botany 30: 693–704.
  • Yu Y, Heinrichs J, Schäfer-Verwimp A, Zhu R-L, Schneider H. 2014. Inferring the accumulation of morphological disparity in epiphyllous liverworts. Organisms Diversity and Evolution 14: 151–162.
  • Zachos J, Pagani M, Sloan L, Thomas E, Billups K. 2001. Trends, rhythms, and aberrations in global climate 65 Ma to present. Science 292: 686–693. [PubMed]

Articles from Annals of Botany are provided here courtesy of Oxford University Press