Evolutionary trends and the origin of the mammalian lower jaw

Sidor, Christian A

Abstract.-The single bony element forming the lower jaw of living mammals, the dentary, has been interpreted as representing the culmination of a long and gradual evolutionary trend. Numerous fossil nonmammalian synapsids ("mammal-like reptiles") show varying degrees of enlargement of the dentary and concomitant reduction in the postdentary bones. To quantitatively reexamine patterns of morphological change in the evolution of the mammalian lower jaw, measurement and discrete character data were collected from 322 fossil synapsid mandibles spanning Late Carboniferous through Jurassic time. Measurements confirm that the relative contribution of the dentary increased in theriodont (advanced therapsid) evolution with regard to both stratigraphic and phylogenetic position. However, dentary enlargement and postdentary reduction failed to typify all therapsid subclades. Qualitative characters of the mandible were used to quantify morphological similarity with regard to the early mammal Morganucodon. Analyses contrasting stratigraphic and phylogenetic position with mammalian similarity indicate that mandibular evolution was primarily conservative, with only anomodont therapsids evolving substantial morphological novelty. Scaling analyses comparing the area of the dentary and postdentary regions to jaw length uniformly show isometry or slight positive allometry, although cynodont therapsids have a smaller postdentary region than any other therapsid subgroup. These results suggest that body size decreases cannot fully explain the reduction of the postdentary bones. Finally, step size bias was tested as a mechanism for explaining long-term trends. Qualitative data reveal no significant difference in the magnitude of character changes occurring in mammalian and nonmammalian directions.

Introduction

Mammals are unique among extant vertebrates in possessing a lower jaw (mandible) formed by a single bony element, the dentary. By contrast, the lower jaws of other vertebrates retain a host of postdentary bones (e.g., four to six in most lizards, five in crocodiles and many birds, and typically even greater numbers in fishes). Recorded from rocks dating from over 300 Ma, the mandibles of the earliest nonmammalian synapsids possessed up to seven postdentary bones (Fig. 1), whereas stratigraphically more recent taxa show various stages in the reduction and eventual loss of these bones (Fig. 2) (Romer and Price 1940; Crompton 1963; Allin 1975). The evolutionary fate of the mammalian postdentary bones has been well established; Reichert (1837) used embryological evidence to homologize the incus and malleus of the mammalian middle ear with the quadrate and articular, respectively, of nonmammalian vertebrates. The transformation of several postdentary jaw bones into sound-conducting middle ear bones within synapsids is one of the best-documented examples of a major evolutionary transformation in the vertebrate fossil record (Hopson 1966; Allin 1975; Allin and Hopson 1992; Luo and Crompton 1994). Indeed, synapsid mandibular evolution has come to be regarded as recording a gradual trend whereby enlargement of the dentary occurs at the expense of the postdentary bones (Crompton and Jenkins 1973; Kemp 1982; Hopson 1987). In this study, I use measurement and discrete character data to: (1) quantify the morphological changes that occurred in the evolution of the lower jaw between pelycosaur-grade synapsids and their mammalian descendants, and (2) address several previously proposed hypotheses concerning the nature and magnitude of morphological trends during the first ~100 Myr of synapsid history.

Background to Study Taxa.-Theories of synapsid evolution have traditionally been couched in terms of several adaptive radiations or grades of organization representing successive steps in the mammalian direction. However, a recent proliferation of numerical cladistic analyses has contributed greatly to our understanding of synapsid phylogeny (Fig. 3), and regions of broad consensus are gradually emerging (Rubidge and Sidor 2001).

The earliest occurring and phylogenetically most primitive synapsids are the "pelycosaurs" of traditional terminology. These taxa form a paraphyletic series and are primarily known from Upper Carboniferous to Lower Permian rocks in Europe and North America (Reisz 1986) although several taxa persisted into the Middle Permian in Russia and South Africa (Reisz et al. 1998; Modesto et al. 2001). Sphenacodontids, such as the familiar sailback Dimetrodon, are among the most advanced pelycosaur subgroups (Reisz et al. 1992). All more derived synapsids form the clade Therapsida.

All of the major therapsid clades first appear in the fossil record during the Middle and Late Permian (e.g., Biarmosuchia, Dinocephalia, Anomodontia, Gorgonopsia, Therocephalia, and Cynodontia) and-except for dicynodont anomodonts, some advanced therocephalians, and cynodonts-went extinct in this time interval as well. Therapsids taxonomically and ecologically dominated the end-Paleozoic Pangaean landscape and established the first herbivore-based food chains among vertebrates in the terrestrial realm (Olson 1962; King et al. 1989; Reisz and Sues 2000). The presence of several derived features recently led Laurin and Reisz (1990, 1996) to suggest that Tetraceratops insignis, from the Early Permian of Texas, is phylogenetically the most primitive therapsid (but see Conrad and Sidor 2001).

Cynodonts are first recorded from Upper Permian strata in southern Africa and Russia and represent the therapsid subclade that includes mammals. Numerous derived features associated with obtaining food and its mastication characterize Cynodontia, including the presence of a fossa for the neomorphic masseter muscle on the lateral surface of the dentary, postcanine teeth with accessory cusps and lingual cingula, and a complete sagittal crest for the origin of temporalis musculature. According to the phylogenetic hypothesis proposed by Hopson (1991b, 1994; Hopson and Kitching 2001), a key dichotomy in cynodont phylogeny occurred with the Triassic divergence of the cynognathian and probainognathian lineages. Terminal cynognathians (tritylodontids) range stratigraphically upwards into the Lower Cretaceous (Tatarinov and Matchenko 1999), whereas terminal probainognathians (mammals) first appear in Upper Triassic or Lower Jurassic rocks and survive until the Recent (Lucas and Luo 1993; Luo 1994).

Vertebrate paleontologists have traditionally defined mammals as possessing a well-formed dentary-squamosal jaw joint (Simpson 1960). Taxa included under this (apomorphy-based) definition include Morganucodon and Kuehneotherium, although these and other early Mesozoic forms (e.g., Sinoconodon) probably possessed a functional quadrate-articular jaw joint as well (Hopson 1991b; Luo and Crompton 1994). More recently, Rowe (1988) and Rowe and Gauthier (1992) have advocated using a crown-group definition for Mammalia, and they have termed the larger clade-including traditional mammals that lie phylogenetically outside the clade bounded by extant forms-Mammaliaformes. My use of Mammalia and of the terms "mammal" and "mammalian" correspond to traditional usage (see also Luo et al. 2002).

Data Collection

Taxon Sampling.-Fossil synapsids included in this study range from the earliest-appearing (Late Carboniferous) pelycosaur-grade taxa through some of the most primitive mammals, such as the Early Jurassic genera Morganucodon and Sinoconodon. In total, 19 "pelycosaurs," six basal therapsids, 13 dinocephalians (including five anteosaurians and eight tapinocephalians), 25 anomodonts, ten gorgonopsians, ten therocephalians, and 25 cynodonts were included. The cynodonts include six non-eucynodonts, 11 cynognathians (including six tritylodontids), and eight probainognathians (including two Mesozoic mammals). All taxa were at the genus level or, in several instances, below.

The stratigraphic range of each taxon was collected from original museum locality information or the literature (e.g., Kitching 1977; Rubidge 1995; Ivachnenko et al. 1997) and then binned into one or more of 18 age ranks (AR) for the purpose of analysis. ARs are non-overlapping stratigraphic bins in an ordered sequence (Gauthier et al. 1988). Importantly, ARs are not necessarily of equal duration; some ARs are equivalent to a single geological formation, whereas others encompass several formations or groups. The goal of this type of binning is a single, resolved sequence of the synapsid fossil record despite its derivation from a variety of widely separated continental deposits (see also Sidor 2001). One major drawback to the AR approach is that gaps in the synapsid record are effectively ignored; time periods lacking synapsid fossils are not represented in the analysis. For example, a major hiatus in the synapsid record occurs between the youngest continental deposits in North America (e.g., the San Angelo and Flowerpot Formations) and the oldest in Russia and South Africa (e.g., Mezen and the Eodicynodon Assemblage Zone, respectively) (Lucas and Heckert 2001). This approximately 2-Myr hiatus encompasses much of Roadian time, but is not evident between ARs 6 and 7. Appendix 4 reports the geological formations and vertebrate biozones making up each AR.

From a recent compilation of synapsid cladistic analyses, I also collected phylogenetic inference data, which consisted of each taxon's clade rank (CR) (Gauthier et al. 1988) and the number of branch points from the root of the cladogram (i.e., patristic distance, PD) (Figs. 3, 4). A rationale for this specific arrangement of synapsid relationships is provided in Appendix 1. CR equals the number of branching points a taxon is positioned up the phylogenetic trajectory from Synapsida to Mammalia (Fig. 3). Branching within a terminal taxon on this pectinate tree is not considered. For example, every species within Gorgonopsia has a CR of 11. In contrast, PD measures the total number of nodes passed from the root of the cladogram to the taxon in question because branching within side-branches is taken into account. Only when a singleton attaches directly to the primary spine of the cladogram (e.g., Tetraceratops or Dvinia) are CR and PD equal.

Data were collected from study of fossil specimens at the following institutions: Albany Museum, Grahamstown, South Africa; American Museum of Natural History, New York; Field Museum of Natural History, Chicago; Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts; National Museum of Natural History, Washington, D.C.; University of California Museum of Paleontology, Berkeley; Bernard Price Institute for Palaeontological Research, University of the Witwatersrand, Johannesburg; National Museum, Bloemfontein; South African Museum, Cape Town; Transvaal Museum, Pretoria; The Natural History Museum, London; Museum of Zoology, University of Cambridge, Cambridge, United Kingdom; Oxford University Museum, Oxford; Paleontological Institute, Moscow; Bayerische Staatssammlung fur Palaontologic und Historische Geologie, Munich; Humboldt Museum fur Naturkunde, Berlin; and Institut und Museum fur Geologie und Palaontologie, Tubingen. Only four taxa were coded exclusively from the literature: Bienotheroides wanhsienensis, Ecteninion lunensis, Probelesodon sanjuanensis, and Sinoconodon sp. (Sun 1984; Crompton and Luo 1993; Martinez and Forster 1996; Martinez et al. 1996).

Qualitative Data.-I used 82 qualitative characters to describe morphological variation within the synapsid mandible. These characters included 22 pertaining to the dentary and 41 to the postdentary bones, six general shape features, and 13 dentition-related features. The last set specifically did not include characters describing the morphology of the dentition itself (such as cusp patterns). Rather, these characters focused on dental features manifesting themselves on the form of the lower jaw itself (e.g., whether or not the teeth were set in sockets). Many of the characters and character states were taken from previous cladistic analyses of synapsid relationships. Characters, character state descriptions, and literature references are presented in Appendix 2. The corresponding taxon/character data matrix is in Appendix 3.

The data matrix in Appendix 3 was converted to a taxon/taxon similarity matrix using the Simple Matching Coefficient (S^sub SM^) (Sokal and Sneath 1963), which is equal to the number of characters for which two taxa match divided by the number of characters for which they could possibly match (in order to accommodate missing or inapplicable characters). Appendix 4 contains the final line in the similarity matrix, which compares the early mammal Morganucodon with every other taxon.

Quantitative Data.-Two areas and four linear measurements constitute the quantitative data set (Fig. 5). The area of the dentary and postdentary bones was calculated by digitizing their respective outlines in NIH Image. Synapsid mandibles were oriented norma lateralis, and then the following measurements were taken parallel to the long axis of the jaw: (1) jaw length, measured from the anterior-most point on the dentary to the midpoint of the craniomandibular joint; (2) dorsal length of the dentary, measured from the anterior tip of the dentary to the sutural contact between the dentary and surangular along the dorsal margin of the mandible; (3) ventral length of the dentary, measured from the anterior tip of the dentary to the sutural contact between the dentary and angular along the ventral margin of the mandible; and (4) perpendicular to the previous measurements, height of the coronoid region, equal to the distance from the mandibular joint to the dorsalmost point on the lower jaw. Measurements under 200 mm were taken with digital calipers and recorded to the nearest one-tenth millimeter. Measurements over 200 mm were taken with a measuring tape and recorded to the nearest half-millimeter. Raw data are presented in Appendix 5.

From the six original variables, four ratios were calculated: (A) the relative position of the dentary/surangular suture, calculated as the ratio of measurement (2) to measurement (1), (B) the relative position of the dentary/angular suture, calculated as the ratio of measurement (3) to measurement (1); (C) the relative height of the coronoid process, calculated as the ratio of measurement (4) to measurement (1); and (D) the relative area of the dentary, calculated as the area of the dentary divided by the total area of the lower jaw. Measurements were originally collected from 764 mandibles (Sidor 2000: App. 6.1). Of these, 322 lower jaws preserved at least two measurements and were used to calculate mean values for each taxon for each of the four dentary ratios (Appendix 5). For the purpose of summarizing changes in all four ratios, each taxon's dentary index (DI) was calculated as the average of the four ratios when each was standardized to have a mean of zero and unit variance. Appendix 4 contains the four original ratios and the summary DI, in addition to each taxon's first and last appearances (in ARs), CR, and PD. Because of fossil incompleteness, not every taxon has a complete set of measurements and so could not be included in all analyses.

Measurement-based Analyses

Stratigraphic Results.-Figure 6A summarizes the results of the measurement-based analyses. The dentary index (DI) is plotted on the abscissa, with increasingly positive values indicating an overall larger contribution of the dentary to the composition of the lower jaw. Although only the DI is depicted, the four individual dentary ratios show similar patterns (Table 1). The ordinate represents the stratigraphic distribution of each taxon in ARs. Significant, positive correlation between the two axes corresponds to a preferential positioning of dentary relative sizes through time.

The pattern depicted in Figure 6A confirms that the earliest-occurring, pelycosaur-grade taxa had the relatively smallest dentaries and largest complement of postdentary bones, and that the latest-occurring synapsids (e.g., tritylodontids and mammals) had mandibles that were almost exclusively formed by the dentary. Importantly, both the maximum and average dentary size increased over time in this study sample. However, it is interesting to note that several late-occurring synapsids retained relatively small dentaries.

Table 1 displays the results of a series of analyses that examine evolutionary patterns within several synapsid subclades. These analyses show that a significant positive correlation between DI and AR is nearly uniformly present in those clades encompassing mammals (e.g., Synapsida, Therapsida, Theriodontia, Cynodontia). Probainognathia is the exception to this pattern, but this may be due to the relatively few intervals that this clade spans. In contrast, clades not encompassing mammals (i.e., side branches such as anomodonts) generally have nonsignificant correlations. This crucial disagreement suggests that only the ancestral lineage leading to mammals (i.e., along the backbone of the cladogram) shows a consistent dentary enlargement (see below), and that clades budded off from this line retained their ancestral proportions but did not systematically continue the trend. It is worth noting that pelycosaur-grade synapsids show little indication of directionality, even though they span seven long intervals (ARs 1-7; Late Carboniferous to early Middle Permian, or approximately 35 Myr) and represent the primitive morphotype from which all subsequent change was derived.

Phylogenetic Results.-Figure 6B plots the relationship between DI and each taxon's cladogram position, as measured by CR (see Table 2 for complete results). It is clear from this graph that the degree to which a synapsid clade shares ancestry with mammals has a strong, positive relationship with that clade's average dentary size. However, when the inferred primitive condition for each consecutive clade is highlighted (filled circles), this point does not consistently reside in the left tail of that clade's range of DI values. This position suggests that diversification within each synapsid subclade expanded the range of DI values but did not uniformly increase the relative contribution of the dentary.

Directionality within subclades is more fully considered in Figure 6C (and Table 3), which contrasts the number of branch points separating each taxon from the root of the cladogram (patristic distance; PD) with its DI. Because stratigraphic and phylogenetic position show a strong relationship in synapsids (Sidor and Hopson 1998), Figure 6C is very similar to 6A. Taxa diverging relatively early (i.e., with low PDs) tend to have small dentaries, whereas phylogenetically more derived taxa show a wider range of values. The expanding range of values observed at high PDs can be attributed to the persistence of small-dentaried anomodonts (plus signs) with theriodonts (filled circles) that consistently enlarge the dentary. However, just as with the stratigraphic analyses, the individual theriodont subclades that lack mammals as a subgroup lack a corresponding trend (Table 3). Again, this suggests that increasing the dentary size was not a universal feature of synapsid evolution.

Character-based Analyses

The six measurements used above can provide only a limited view of morphological changes occurring within synapsid mandibular evolution. Potentially more informative is quantifying morphological similarity with reference to an exemplar primitive mammal (Morganucodon, in this case) using discrete characters (Appendix 4). Phenetic similarity is an appropriate metric to use in this case because the convergent acquisition of a certain phenotype pertains to net, rather than total, morphological change (Foote 1996). As with the measurement-based analyses, significantly positive correlations between the degree of similarity to mammals and stratigraphic or phylogenetic position would support the hypothesis of a morphological trend toward gaining mammal-like features, whereas non-significant correlations would refute it. Furthermore, negative correlations correspond to increasing dissimilarity; i.e., the morphological modifications experienced by a clade's lower jaw consistently distance it from the mammalian position in morphospace.

Stratigraphic and Phylogenetic Results.-Figure 7 and Table 4 contain the principal results of the discrete character-based analyses, which are remarkably similar to those based on measurements (compare with Fig. 6). This similarity implies that both data sets are capturing a common signal from synapsid evolution. When compared with stratigraphic position (Fig. 7A), Late Carboniferous and Early Permian pelycosaur-grade synapsids begin with approximately 60% of their (comparable) lower-jaw characters matching the condition in Morganucodon (AR 1-6). By the middle of Late Permian times (AR 9), however, therapsid diversification expanded this range of values, with anomodonts becoming increasingly dissimilar to mammals, and theriodonts becoming increasingly similar (presumably through synapomorphy). The "increase in variance" pattern continued until the demise of anomodonts in the Late Triassic (AR 17), whereby only the advanced cynodonts (i.e., the right tail of the distribution) remained. As with the measurement-based results, subclades encompassing mammals typically show significant correlations between AR and the degree of similarity to mammals, whereas side-branches often do not (Table 4).

Figure 7B compares the degree of mammal mandibular similarity against each taxon's CR, with the inferred primitive condition at each point highlighted. An increasingly mammal-like lower jaw is expected to correlate with higher CRs, given that some of the features used in this analysis have been proposed as synapomorphies diagnosing higher-level synapsid clades. An interesting result is the relatively low degree of divergence (i.e., range of values) from the presumed ancestral condition at each CR (filled circles). Only anomodonts, and in particular their derived dicynodont subclade (e.g., Fig. 2D), show substantial morphological divergence. Although I attempted to be as exhaustive as possible in my character selection, doubtless additional characters could be discovered and affect this low degree of subclade morphological divergence.

Presuming that the characters used herein are an unbiased sample from the total pool of possible lower-jaw characters, an interesting pattern emerges: except for caseasaurs (CR 1), the inferred primitive condition at each CR lies at the mammal-like (right-hand) tail for non-theriodonts (CRs 2-10), shifts to an intermediate value within gorgonopsians and therocephalians (CRs 11 and 12), and then lies on the non-mammal-like (left-hand) end for cynodonts onward (CRs 13-22). This implies that morphological change within each subclade went from being primarily divergent, within non-cynodonts, to convergent, within cynodonts (but see below). It is also worth noting that discontinuities between the inferred ancestral condition between adjacent CRs could indicate gaps in the fossil record, if synapsid evolution was predominantly monotonic (Sidor and Hopson 1998), or variation in the rate of character acquisition, if taxon sampling probabilities were relatively constant (Sereno et al. 1999).

Figure 7C plots the number of inferred branch points from the root of the cladogram to each terminal taxon (PD) against the degree to which each taxon's lower jaw is similar to that of Morganucodon. Taxa positioned near the base of the tree (with low PDs) hover around a mammal mandibular similarity of 60%, whereas anomodonts and nonmammalian theriodonts expand this range by roughly 20% in negative and positive directions, respectively. Thus, prior to the early mammal Sinoconodon, the lower jaws of advanced nonmammalian cynodonts such as Probainognathus maintained substantial differences from those of early mammals. This plot most clearly depicts the Y-shaped pattern hinted at in several other graphs (compare Figs. 6A,C and 7A,C), where anomodonts and theriodonts morphologically diverge from one another after an early phase of nondirectionality within more basal synapsids. The gap between the branches of the Y is due to the relatively low diversity and short temporal duration of the clades with intermediate similarity values (viz. gorgonopsians and therocephalians).

Scaling Patterns

Synapsids diversified into an impressive array of body sizes and presumed ecologies during the Permian and Triassic. However, the possibility that changes in body size were important factors in the reduction of the postdentary bones has received scant attention in the literature, instead, most analyses have focused on the detailed morphology of several exemplar taxa assumed to be phylogenetically close to the line leading toward mammals (Allin 1975).

Figure 8A compares dentary area and total jaw length for the 160 synapsids with both measurements (Appendix 5). A line with a slope of two indicates isometry in this case because an area is being plotted against a linear measurement. A reduced major axis regression (RMA) indicates that synapsids as a whole conform to this expectation (slope = 2.031 + or - 0.101). Furthermore, various synapsid subgroups show either near isometry or slight positive allometry (Table 5).

Figure 8B plots postdentary area against total jaw length for 154 fossil synapsids. In contrast to the dentary area results, the RMA regression for Synapsida is significantly greater than isometry (slope = 2.617 + or - 0.180). However, this apparent allometry is due to the mixing of two regressions. When, synapsids are subdivided into cynodonts and non-cynodont components, both of these groups show a relationship between postdentary area and jaw length that is indistinguishable from isometry (slopes of 2.174 + or - 0.184 and 2.108 + or - 0.180, respectively). However, the cynodont regression is offset below that of all other synapsids, indicating that this group had a proportionally more gracile postdentary region. See Table 6 for complete results.

Step-Size Analysis

The analyses presented above show a complicated pattern of results but do not address the underlying mechanisms by which trends could develop. One such mechanism could be a bias in step size (Fisher 1986; McShea 1994; Wagner 2000b). For example, even if dentary increases and decreases were equally likely to occur during the course of synapsid evolution, if increases were twice the magnitude of decreases, then the dentary would be predicted to enlarge over time. The possibility that unequal degrees of mammalian versus nonmammalian morphological change brought about a trend for an increasingly mammalian lower jaw is examined below.

Methods.-To address the hypothesis of step-size bias with the discrete character data, I used MacClade (Maddison and Maddison 1992) to contrast the number of internodal character state changes leading to each pair of sister taxa at every CR along the spine of the cladogram in Figure 3. For example, at CR 7 (Therapsida) between three and ten character-state changes (depending on optimization) occur along the branch to Tetraceratops, whereas six to nine occur along the branch to CR 8. If evolution along the mammalian line typically produced larger than average changes, then we might expect the number of character-state changes between CRs to be consistently larger than between CR nodes and side branches.

These comparisons are based on the premise that morphological changes occurring between consecutive nodes on this cladogram produce increased similarity to mammals (because they are synapomorphic), whereas changes accumulated on the side-branches (i.e., toward the individual terminal taxa) should produce increased dissimilarity. One shortcoming of this type of analysis is that it only uses the first possible comparison at each CR (i.e., changes occurring on the first inter-node in either direction) and thereby disregards subsequent (i.e., more deeply nested) changes within each sister clade.

Results.-Wilcoxon sign-rank tests found no significant difference between the number of character changes in mammalian and nonmammalian directions (Table 7, Fig. 9). This result was the same regardless of whether minimum or maximum numbers of character-state changes were used.

Discussion

The hypothesis that disparate groups of synapsids independently acquired mammallike characteristics has a long pedigree (Olson 1944, 1959, 1962; Romer 1965; Simpson 1959). However, some examples of "convergence" probably arose from the taxonomic framework accepted at that time-one that recognized paraphyletic and polyphyletic grades of organization (Hopson 1994). The application of cladistic methods to synapsid systematics has dispelled some cases of morphologic homoplasy as unnecessary when viewed from the standpoint of total character congruence (Hopson and Barghusen 1986; Rowe 1986; Kemp 1988b). Here, I have readdressed the oft-noted observation that the size of the dentary increased during the course of synapsid evolution. Both quantitative and discrete data indicate that a lower jaw of increasingly mammalian cast was a prevalent feature of pre-mammalian synapsid evolution (Fig. 10), but finer scales of phylogenetic resolution yield more complex patterns.

In Theriodontia and its subordinate clades that encompass mammals, the pattern of both measurements and similarity values is suggestive of a driven trend (in that both the minimum and maximum values steadily increase). Thus, the measurement results accord well with Allin's (1975) hypothesis that reduction of the postdentary bones improved high-frequency hearing in these taxa and was therefore selectively advantageous. However, corresponding directionality is not apparent within the "side-branch" clades (Tables 1-4), which suggests that a common driving force is doubtful. In the most extreme case, anomodonts show the exact opposite trend: decreasing dentary size and increasing their lower jaw's distinctiveness from that of mammals. This suggests either that high-frequency hearing was not important to anomodonts or that selection for this feature was not exclusively molding mandibular evolution in this group.

The specialized structure of the anomodont mandible is an interesting exception to another result of these analyses-the relative scarcity of divergent lower-jaw morphologies among synapsid side-branches. Although there are certainly some features that are autapomorphic for the clades that do not encompass mammals (e.g., the extremely slender dentary of varanopseids, the near-vertical ridge on the reflected lamina of gorgonopsians, or the elongate angular process of the dentary in some advanced cynognathians), no one synapsid subgroup amasses more than a few such specializations, except for the anomodonts. Importantly, this lack of mandibular autapomorphy indicates that the acquisition of only a few mammalian characters would be sufficient to drive an apparent trend toward a mammal-like jaw.

Disruptive Patterns.-The Y-shaped pattern of dentary size and mammalian similarity (Figs. 6, 7) that emerged from several analyses is strikingly similar to that of disruptive selection within modern populations (i.e., when selection acts against intermediates and favors morphological extremes). Foote (1993) showed that blastoids exhibit a similar disruptive pattern, but he suggested that if a bias against intermediates were present, then its explanation would require investigation at finer scales. In the case of synapsids, the lack of intermediates is due to the early extinction of gorgonopsians and therocephalians, compared with the relatively long-lived anomodont and cynodont clades.

Combining Methods.-Both stratigraphy- and phylogeny-based methods have been used to examine patterns of morphological change in fossil lineages (Gingerich 1976; Benton 1990; McShea 1994; Wagner 1996). Importantly, the potential weaknesses of either approach might be overcome by using both methods in a study. For example, if cladistic estimates of synapsid phylogeny have been led astray by rampant homoplasy, then the stratigraphic distribution of the taxa may yield a more informative measure of relatively primitive and derived taxa. Conversely, if the fossil record does not accurately portray the first appearances of synapsids because preservation rates vary widely, then phylogenetic measures might yield a more reliable sequence of branching events. The concordant results found in this study suggest that the synapsid fossil record is relatively well sampled and that the cladistic hypothesis of synapsid relationships presented here is in line with the distribution of fossil finds (Sidor and Hopson 1998).

Conclusions

The prevalence of homoplasy in synapsid evolution has been a hotly contested topic (Kemp 1988a; Rowe 1988; Hopson 1991a). Hopson (1994: p. 212) suggested that although "[t]he polyphyletic origin of mammals is no longer a tenable hypothesis. . . this is not to say that parallelism and convergence have not been significant aspects of pre-mammalian synapsid evolution."

The present study supports the following main conclusions:

1. The lack of a well-supported phylogeny has exaggerated previous estimates of morphological convergence or parallelism in the synapsid fossil record. The hypothesis of multiple therapsid groups arising independently from pelycosaur-grade ancestors (e.g., Olson 1962; Boonstra 1972) necessitated rampant homoplasy and are now considered untenable (Rubidge and Sidor 2001). Certain lower jaw characteristics and proportions are better viewed as broadly distributed synapomorphies indicative of common ancestry.

2. Despite the striking differences between the lower jaws of basal synapsids (i.e., "pelycosaur") and mammals, mandibular evolution within synapsids was predominantly conservative. Except for dicynodont anomodonts, most therapsid subclades do not acquire substantial morphological novelty in their lower jaw structure.

3. The area of the dentary and postdentary regions scales either isometrically or with slight positive allometry when compared with jaw length. This suggests that body-size trends are not sufficient to drive the reduction of the postdentary bones in synapsid evolution. Importantly, when compared with other synapsid subgroups, cynodonts are characterized by smaller-than-predicted postdentary areas.

4. Selection acting to decrease the size of the postdentary bones, and thereby improving high-frequency hearing, is still the most tenable mechanism for the evolution of the mammalian lower jaw (Allin 1975; Allin and Hopson 1992). However, this mechanism by itself has difficulty explaining the converse pattern in anomodont therapsids (i.e., decreasing the size of the dentary and increasing the size of the postdentary bones).

These conclusions, in combination with those of recent studies on long-term patterns of epipodial (Hopson 1995) and cranial (Sidor 2001) evolution, suggest that morphological trends within synapsids should be reinvestigated within a quantitative and phylogenetic framework.

Acknowledgments

This project was part of my dissertation research at the University of Chicago. I thank my committee, M. Foote, R. Reisz, P. Sereno, P. Wagner, and especially my advisor, J. Hopson, for comments on previous versions of the manuscript. M. Carrano's review of the first draft helped considerably. I also acknowledge the support and assistance provided by my fellow graduate students, including H. Larsson, J. Wilson, R. Blob, J. Socha, R. O'Keefe, D. Croft, P. Magwene, F. Lando, A. Beck, J. Conrad, E. Love, and J. Tsao. Data for this project were gathered on trips to several domestic and foreign research collections. For their help in this critical aspect of my dissertation research, I sincerely thank the following curators and museum personnel: N. Hotton, M. Brett-Surman, G. Wilson, K. Padian, M. Norell, G. Gaffney, F. Jenkins, A. Crompton, C. Schaff, M.-A. Turner, H.-D. Sues, R. Reisz, S. Modesto, D. Scott, J. Bolt, O. Rieppel, B. Rubidge, M. Raath, C. Gow, F. Thackeray, H. Fourie, J. Neveling, J. Welman, R. Smith, S. Kaal, A. Milner, S. Chapman, A. Friday, R. Symonds, T. Kemp, M. Wills, D. Sigogneau-Russell, D. Dutheil, M. Maisch, P. Wellnhofer, D. Unwin, M. Ivachnenko, N. Kalandadze, A. Kurkin, V. Bulanov, and V. Golubev. I also acknowledge the following granting organizations and institutions for supporting my research: National Science Foundation Doctoral Dissertation Improvement Grant (NSF DEB-9801342), Hinds Fund (University of Chicago), the American Museum of Natural History, and the Richard Estes Memorial Award from the Society of Vertebrate Paleontology. This manuscript profited from reviews by T. Kemp, S. Wing, J. Lillegraven, and D. McShea.

Literature Cited

Allin, E. F. 1975. Evolution of the mammalian middle ear. Journal of Morphology 147:403-438.

Allin, E. F., and J. A. Hopson. 1992. Evolution of the auditory system in Synapsida ("mammal-like reptiles" and primitive mammals) as seen in the fossil record. Pp. 587-614 in D. B. Webster, R. R. Fay, and A. N. Popper, eds. The evolutionary biology of hearing. Springer, New York.

Angielczyk, K. D. 2001. Preliminary phylogenetic analysis and stratigraphic congruence of the dicynodont anomodonts (Synapsida: Therapsida). Palaeontologia Africana 37:53-79.

Battail, B. 1991. Les Cynodontes (Reptilia, Therapsida): une phylogenie. Bulletin du Museum National d'Histoire Naturelle 13:17-105.

Benton, M. J. 1990. Reptiles. Pp. 279-300 in K. J. McNamara, ed. Evolutionary trends. Belhaven, London.

Berman, D. S., R. R. Reisz, J. R. Bolt, and D. Scott. 1995. The cranial anatomy and relationships of the synapsid Varanosaurus (Eupelycosauria: Ophlacodontidae) from the Early Permian of Texas and Oklahoma. Annals of the Carnegie Museum 64: 99-133.

Bonaparte, J. F. 1962. Descripcion del craneo y mandibula de Exaeretodon frenguellii, Cabrera, y su comparacion con Diademodontidae, Tritylodontidae y los cinodontes sudamericanos. Publicaciones del Museo Municipal de Ciencias Naturales de Mar del Plata "Lorenzo Scaglia" 1:135-202.

Boonstra, L. D. 1972. Discard the names Theriodontia and Anomodontia: a new classification of the Therapsida. Annals of the South African Museum 59:315-338.

Brinkman, D. 1981. The structure and relationships of the dromasaurs (Reptilia: Therapsida). Breviora 465:1-34.

Clark, J. M., and J. A. Hopson. 1985. Distinctive mammal-like reptile from Mexico and its bearing on the phylogeny of the Tritylodontidae. Nature 315:398-400.

Cluver, M. A., and G. M. King. 1983. A reassessment of the relationships of the Permian Dicynodontia (Reptilia, Therapsida) and a new classification of dicynodonts. Annals of the South African Museum 91:195-273.

Conrad, J., and C. Sidor. 2001. Re-evaluation of Tetraceratops insignis (Synapsida: Sphenacodontia). Journal of Vertebrate Paleontology 21:42A.

Crompton, A. W. 1963. The evolution of the mammalian jaw. Evolution 17:431-439.

_____. 1972. Postcanine occlusion in cynodonts and tritylodontids. Bulletin of the British Museum (Natural History) Geology 21:30-71.

Crompton, A. W., and F. Ellenberger. 1957. On a new cynodont from the Molteno Beds and the origin of the tritylodontids. Annals of the South African Museum 44:1-14.

Crompton, A. W., and N. Hotton. 1967. Functional morphology of the masticatory apparatus of two dicynodonts (Reptilia, Therapsida). Postilla 109:1-51.

Crompton, A. W., and F. A. Jenkins Jr. 1973. Mammals from reptiles: a review of mammalian origins. Annual Review of Earth and Planetary Sciences 1:131-155.

Crompton, A. W., and Z. Luo. 1993. Relationship of the Liassic mammals Sinoconodon, Morganucodon oehleri, and Dinnetherium. Pp. 30-44 in F. S. Szalay, M. J. Novacek, and M. C. McKenna, eds. Mammal phylogeny: Mesozoic differentiation, multituberculates, monotremes, early therians, and marsupials. Springer, New York.

Efremov, I. A. 1940. Ulemosaurus svijagensis Riab.-ein Dincephale aus den Ablagerurgen des Perm der UdSSR. Nova Acta Leopoldina 9:155-205.

Fisher, D. C. 1986. Progress in organismal design. Pp. 99-117 in D. M. Raup and D. Jablonski, eds. Patterns and processes in the history of life. Springer, Berlin.

Foote, M. 1993. Contributions of individual taxa to overall morphological variation. Paleobiology 19:403-419.

_____. 1996. Evolutionary patterns in the fossil record. Evolution 50:1-11.

Gauthier, J., A. Kluge, and T. Rowe. 1988. Amniote phylogeny and the importance of fossils. Cladistics 4:105-209.

Gingerich, P. D. 1976. Paleontology and phylogony: patterns of evolution at the species level in Early Tertiary mammals. American Journal of Science 276:1-28.

Grine, F. E. 1997. Dinocephalians are not anomodonts. Journal of Vertebrate Paleontology 17:177-183.

Hopson, J. A. 1965. Tritylodontid therapsids from Yunnan and the cranial morphology of Bienotherium. Ph.D. dissertation. University of Chicago, Chicago.

_____. 1966. The origin of the mammalian middle car. American Zoologist 6:437-450.

_____. 1987. The mammal-like reptiles: a study of transitional fossils. American Biology Teacher 49:16-26.

_____. 1991a. Convergence in mammals, tritheledonts, and tritylodonts. Journal of Vertebrate Paleontology 11:36A.

_____. 1991b. Systematics of the nonmammalian Synapsida and implications for patterns of evolution in synapsids. Pp. 635-693 in H.-P. Schultze and L. Trueb, eds. Origins of the higher groups of tetrapods: controversy and consensus. Comstock, Ithaca.

_____. 1994. Synapsid evolution and the radiation of non-eutherian mammals. Pp. 190-219 in D. B. Prothero and R. M. Schoch, eds. Major features of vertebrate evolution. Paleontological Society, Knoxville, Tenn.

_____. 1995. Patterns of evolution in the manus and pes of non-mammalian therapsids. Journal of Vertebrate Paleontology 15:615-639.

_____. 1999. Therapsids. Pp. 1256-1266 in R. Singer, ed. Encyclopedia of paleontology. Fitzroy Dearborn, Chicago.

Hopson, J. A., and H. Barghusen. 1986. An analysis of therapsid relationships. Pp. 83-106 in N. Hotton, P. D. MacLean, J. J. Roth, and E. C. Roth, eds. The ecology and biology of the mammal-like reptiles. Smithsonian Institution Press, Washington, D.C.

Hopson, J. A., and J. W. Kitching. 1972. A revised classification of cynodonts (Reptilia; Therapsida). Palaeontologia Africana 14:71-85.

_____. 2001. A probainognathian cynodont from South Africa and the phylogeny of nonmammalian cynodonts. Bulletin of the Museum of Comparative Zoology 156:3-35.

Huelsenbeck, J. P. 1991. When are fossils better than extant taxa in phylogenctic analysis? Systematic Zoology 40:458-469.

Ivachnenko, M. F., V. K. Golubev, Y. M. Gubin, N. N. Kalandadze, I. V. Novikov, A. G. Sennikov, and A. S. Rautian. 1997. Permskie i Triasovye tetrapody Vostochnoi Evropy (Permian and Triassic tetrapods of Eastern Europe). Geos, Moscow.

Kemp, T. S. 1982. Mammal-like reptiles and the origin of mammals. Academic Press, New York.

_____. 1983. The relationships of mammals. Zoological Journal of the Linnean Society 77:353-384.

_____. 1988a. Haemothermia or Archosauria? The interrelationships of mammals, birds and crocodiles. Zoological Journal of the Linnean Society 92:67-104.

_____. 1988b. Interrelationships of the Synapsida. In M. J. Benton, ed. The phylogeny and classification of the tetrapods, Vol. 2. Mammals. Systematics Association Special Volume 35B:1-22. Clarendon, Oxford.

Kermack, D. M., F. Mussett, and H. W. Rigney. 1973. The lower jaw of Morganucodon. Zoological Journal of the Linnean Society 53:87-175.

King, G. M. 1988. Anomodontia. Pp. 1-174 in P. Wellnhofer, ed. Encyclopedia of paleoherpetology, Part 17C. Gustav Fischer, Stuttgart.

King, G. M., B. W. Oelofsen, and B. S. Rubidge. 1989. The evolution of the dicynodont feeding system. Zoological Journal of the Linnean Society 96:185-211.

Kitching, J. W. 1977. The distribution of the Karroo vertebrate fauna, Memoir 1. Bernard Price Institute for Palaeontological Research, University of the Witwatersrand, Johannesburg.

Laurin, M. 1993. Anatomy and relationships of Haptodus garnettensis, a Pennsylvanian synapsid from Kansas. Journal of Vertebrate Paleontology 13:200-229.

Laurin, M., and R. R. Reisz. 1990. Tetraceratops is the oldest known therapsid. Nature 345:249-250.

_____. 1996. The osteology and relationships of Tetraceratops insignis, the oldest known therapsid. Journal of Vertebrate Paleontology 16:95-102.

Lucas, S. G., and A. B. Heckert. 2001. Olson's gap: a global hiatus in the record of Middle Permian tetrapods. Journal of Vertebrate Paleontology 21:75A.

Lucas, S. G., and Z. Luo. 1993. Adelobasileus from the Upper Triassic of West Texas: the oldest mammal. Journal of Vertebrate Paleontology 13:309-334.

Luo, Z. 1994. Sister-group relationships of mammals and transformations of diagnostic mammalian characters. Pp. 98-128 in N. C. Fraser and H.-D. Sues, eds. In the shadow of the dinosaurs: early Mesozoic tetrapods. Cambridge University Press, New York.

Luo, Z., and A. W. Crompton. 1994. Transformation of the quadrate (incus) through the transition from non-mammalian cynodonts to mammals. Journal of Vertebrate Paleontology 14:341-374.

Luo, Z., and X.-C. Wu. 1995. Correlations of vertebrate assemblage of the Lower Lufeng Formation, Yunnan, China. Pp. 83-88 in A. Sun and Y. Wang, eds. Sixth symposium of Mesozoic terrestrial ecosystems and biota, short papers. China Ocean Press, Beijing.

Luo, Z., Z. Kielan-Jaworoska, and R. L. Cifelli. 2002. In quest for a phylogeny of Mesozoic mammals. Acta Palaeontologica Polonica 47:1-78.

Maddison, W. P., and D. R. Maddison. 1992. MacClade 3.0.8a: interactive analysis of phylogeny and character evolution. Sinauer, Sunderland, Mass.

Martinez, R. N., and C. A. Forster. 1996. The skull of Probelesodon sanjuanensis, Sp. Nov., from the Late Triassic Ischigualasto Formation of Argentina. Journal of Vertebrate Paleontology 16:285-291.

Martinez, R. N., C. L. May, and C. A. Forster. 1996. A new carnivorous cynodont from the Ischigualasto Formation (Late Triassic, Argentina), with comments on eucynodont phylogeny. Journal of Vertebrate Paleontology 16:271-284.

McShea, D. W. 1994. Mechanisms of large-scale evolutionary trends. Evolution 48:1747-1763.

Modesto, S. P. 1995. The skull of the herbivorous synapsid Edaphosaurus boanerges from the Lower Permian of Texas. Palaeontology 38:213-239.

Modesto, S. P., B. S. Rubidge, and J. Welman. 1999. The most basal anomodont therapsid and the primacy of Gondwana in the evolution of the anomodonts. Proceedings of the Royal Society of London B 266:331-337.

Modesto, S., C. A. Sidor, B. S. Rubidge, and J. Welman. 2001. A second varanopseid skull from the Upper Permian of South Africa: implications for Late Permian 'pelycosaur' evolution. Lethaia 34:249-259.

Olson, E. C. 1944. The origin of mammals based on the cranial morphology of the therapsid suborders. Geological Society of America Special Paper 55:1-136.

_____. 1959. The evolution of mammalian characters. Evolution 13:344-353.

_____. 1962. Late Permian terrestrial vertebrates, U.S.A. and U.S.S.R. Transactions of the American Philosophical Society 52:1-224.

Parrington, F. R. 1955. On the cranial anatomy of some gorgonopsids and the synapsid middle ear. Proceedings of the Zoological Society of London 125:1-40.

Reichert, C. 1837. Uber die Visceralbogen der Wirbeltiere im Allgemeinen und deren Metamorphosen bei den Vogeln und Saugetieren. Mullers Archiv fur Anatomie, Physiologie und wissenschaftliche Medizin 1837:120-222.

Reisz, R. R. 1986. Pelycosauria. Pp. 1-102 in P. Wellnhofer, ed. Encyclopedia of paleoherpetology, Part 17A. Gustav Fischer, Stuttgart.

_____. 1988. Two small reptiles from a Late Pennsylvanian quarry near Hamilton, Kansas. Kansas Geological Survey Guidebook 6:189-194.

Reisz, R. R., and D. W. Dilkes. 1995. A small varanopseid (Synapsida: Eupelycosauria) from the uppermost Carboniferous (Upper Pennsylvanian) of Kansas. Journal of Vertebrate Paleontology 15:49A.

_____. 2003. Archaeovenator hamiltoni, a new varanopid (Synapsida: Eupelycosauria) from the Upper Carboniferous of Kansas. Canadian Journal of Earth Sciences 40:667-668.

Reisz, R. R., and H.-D. Sues. 2000. Herbivory in Late Paleozoic and Triassic terrestrial vertebrates. Pp. 9-41 in H.-D. Sues, ed. Evolution of herbivory in terrestrial vertebrates. Cambridge University Press, New York,

Reisz, R. R., D. S. Berman, and D. Scott. 1992. The cranial anatomy and relationships of Secodontosaurus, an unusual mammal-like reptile (Synapsida: Sphenacodontidae) from the Early Permian of Texas. Zoological Journal of the Linnean Society 104:127-184.

Reisz, R. R., D. W. Dilkes, and D. S. Berman. 1998. Anatomy and relationships of Elliotsmithia longiceps Broom, a small synapsid (Eupelycosauria: Varanopseidae) from the Late Permian of South Africa. Journal of Vertebrate Paleontology 18:602-611.

Renaut, A. J. 2000. A re-evaluation of the cranial morphology and taxonomy of the Triassic dicynodont genus Kannemeyeria. Ph.D. dissertation. University of the Witwatersrand, Johannesburg.

Romer, A. S. 1965. Possible polyphylety of the vertebrate classes. Zoologische Jahrbuch 92:143-156.

_____. 1969. The Chanares (Argentina) Triassic reptile fauna V. A new chiniquodontid cynodont, Probelesodon lewisi. cynodont ancestry. Breviora 333:1-24.

_____. 1970. The Chanares (Argentina) Triassic reptile fauna VI. A chiniquodontid cynodont with an incipient squamosaldentary jaw articulation. Breviora 344:1-18.

Romer, A. S., and L. I. Price. 1940. Review of the Pelycosauria. Geological Society of America Special Paper 28:1-538.

Rowe, T. 1986. Osteological diagnosis of Mammalia, L. 1758, and its relationship to extinct Synapsida. Ph.D. dissertation. University of California, Berkeley.

_____. 1988. Definition, diagnosis, and the origin of Mammalia. Journal of Vertebrate Paleontology 8:241-264.

_____. 1993. Phylogenetic systematics and the early history of mammals. Pp. 129-145 in F. S. Szalay, M. J. Novacek, and M. C. McKenna, eds. Mammal phylogeny: Mesozoic differentiation, multituberculates, monotremes, early therians, and marsupials. Springer, New York.

Rowe, T., and J. Gauthier. 1992. Ancestry, paleontology, and the definition of the name Mammalia. Systematic Biology 41:372-378.

Rubidge, B. S. 1991. A new primitive dinocephalian mammal-like reptile from the Permian of southern Africa. Palaeontology 34:547-559.

_____. 1994. Australosyodon, the first primitive anteosaurid dinocephalian from the Upper Permian of Gondwana. Palaeontology 37:579-594.

_____. 1995. Biostratigraphy of the Beaufort Group (Karoo Series), South Africa. Government Printer, Pretoria.

Rubidge, B. S., and J. A. Hopson. 1996. A primitive anomodont therapsid from the base of the Beaufort Group (Upper Permian) of South Africa. Zoological Journal of the Linnean Society 117:117-139.

Rubidge, B. S., and C. A. Sidor. 2001. Evolutionary patterns among Permo-Triassic therapsids. Annual Review of Ecology and Systematics 32:449-480.

Rubidge, B. S., and J. A. van den Heever. 1997, Morphology and systematic position of the dinocephalian Styracocephalus platyrhinus. Lethaia 30:157-168.

Rybczynski, N. 2000. Cranial anatomy and phylogenetic position of Suminia getmanovi, a basal anomodont (Amniota: Therapsida) from the Late Permian of Eastern Europe. Zoological Journal of the Linnean Society 130:329-373.

Sereno, P. C., A. L. Beck, D. B. Dutheil, H. C. E. Larsson, G. H. Lyon, B. Moussa, R. W. Sadleir, C. A. Sidor, D. J. Varricchio, G. P. Wilson, and J. A. Wilson. 1999. Cretaceous sauropods from the Sahara and the uneven rate of skeletal evolution among dinosaurs. Science 286:1342-1347.

Sidor, C. A. 2000. Evolutionary trends and relationships within the Synapsida. Ph.D. dissertation. University of Chicago, Chicago.

_____. 2001. Simplification as a trend in synapsid cranial evolution. Evolution 55:1419-1442.

Sidor, C. A., and J. A. Hopson. 1998. Ghost lineages and "mammalness": assessing the temporal pattern of character acquisition in the Synapsida. Paleobiology 24:254-273.

Sidor, C. A., and R. M. H. Smith. In press. A new galesaurid (Therapsida: Cynodontia) from the Early Triassic of South Africa. Palaeontology.

Sigogneau, D. 1970. Revision systematique des gorgonopsiens Sud-Africains. Cahiers de Paleontologie, Paris.

Simpson, G. G. 1959. Mesozoic mammals and the polyphyletic origin of mammals. Evolution 13:405-414.

_____. 1960. Diagnosis of the classes Reptilia and Mammalia. Evolution 14:388-392.

Sokal, R. R., and P. H. A. Sneath. 1963. Principles of numerical taxonomy. W. H. Freeman, San Francisco.

Sues, H.-D. 1985. The relationships of the Tritylodontidae (Synapsida). Zoological Journal of the Linnean Society 85:205-217.

Sun, A. 1984. Skull morphology of the tritylodont genus Bienotheroides of Sichuan. Scientia Sinica 27:970-984.

Tatarinov, L. P., and E. N. Matchenko. 1999. A find of an aberrant tritylodont (Reptilia, Cynodontia) in the Lower Cretaceous of the Kemerovo region. Paleontological Journal 33:422-428.

van den Heever, J. A. 1994. The cranial anatomy of the early Therocephalia (Amniota: Therapsida). Universiteit van Stellenbosch Annale 1994/1:1-59.

Wagner, P. J. 1996. Contrasting the underlying patterns of active trends in morphologic evolution. Evolution 50:990-1007.

_____. 2000a. The quality of the fossil record and the accuracy of phylogenetic inferences about sampling and diversity. Systematic Biology 49:65-86.

_____. 2000b. Rate heterogeneity in shell character evolution among lophospiroid gastropods. Paleobiology 27:290-310.

Watson, D. M. S., and A. S. Romer. 1956. A classification of therapsid reptiles. Bulletin of the Museum of Comparative Zoology 114:37-89.

Christian A. Sidor. Department of Anatomy, New York College of Osteopathic Medicine, Old Westbury, New York 11568-8000. E-mail: casidor@iris.nyit.edu

Accepted: 31 January 2003

Copyright Paleontological Society Fall 2003
Provided by ProQuest Information and Learning Company. All rights Reserved