Contributions to Zoology, 70 (I) 1-22 (2001)

SPB Academic Publishing bv, The Hague

Molecules, morphology, and the monophyly of diapsid reptiles

Michael+S.Y. Lee

Department of Zoology, University of Queensland Brisbane, QLD 4072, Australia, e-mail: [email protected]

Keywords: Reptilia, testudines, cladistics, osteology, fossils, molecular phylogenetics

Abstract Contents

The morphological and molecular evidence forhigher-level reptile Introduction 1 relationships is reassessed, A combined analysis of 176 osteo- Morphological and molecular data sets 4

logical, 40 soft anatomical, and 2903 (1783 aligned) molecular Data analyses 6 characters in 28 amniote taxa yields the traditional reptile tree. Phylogenetic results 7

Synapsids the sister taxon to all other 7 (including mammals) are Separate analyses

' amniotes, all Turtles with 8 including extant reptiles. group anapsid Homogeneity tests parareptiles and fall outside a monophyletic Diapsida. Within Combined analyses 9 diapsids, squamates and Sphenodon form a monophyletic Molecules, morphology and the importance of fossils 11

Lepidosauria, and crocodiles plus birds form a monophyletic Acknowledgements 16 Archosauria. This tree is identical to the tree strongly supported References 16 by the data included. 18 osteological alone when fossils are In a Appendix combined analysis the strong osteological signal linking turtles

with is anapsids sufficient to override a soft anatomical signal

placing turtles next to a heterodox archosaur-mammal clade, Introduction and a weaker molecular signal linking turtles with archosaurs.

However, the turtle-archosaur clade cannot be statistically

rejected. When fossils of the are ignored, the signal in the osteological Despite over a century research, relationships

data set disappears and, in a combined analysis of morphology of amniotes between the major groups living (mam- and molecules, the molecular (turtle-archosaur) signal prevails. mals, turtles, squamates, Sphenodon and archosaurs) These results highlight the importance of fossils, not just in remain contentious. There is now a consensus that osteological studies, but even in “combined” analyses where all form i.e. that turtles they cannot be living reptiles a clade, are scored for the majority ofcharacters (soft anatomy and molecules). Although the total number of molecular traits more closely related to other living reptiles than

is much (2903) than the total number of greater morphological to mammals. This arrangement has now been sup- taits (216), when only characters informative at the relevant ported compelling evidence from both morphol- levels by are considered, the two data sets are approximatelyequal (Gauthier et ah, 1988;Laurin and Reisz, 1995; size. The ogy in partition homogeneity test yields unreliable results de and unless uninformative Lee, 1997; Braga Rieppel, 1997) and mol- (invariant and autapomorphic) characters are excluded. ecules Eernisse and Analyses ofthe mitochondrial data suggest that (Marshall, 1992; Kluge, 1993;

recent evidence from nuclear for heterodox turtle- genes a Van de Peeret ah, 1993; Caspers et ah, 1996; Zar- crocodile clade be artefact of inade- (excluding birds) might an doya and Meyer, 1998). However, relationships quate of sampling a diverse outgroup (mammals) and thus, within the clade remain controversial, due problems with reptile rooting the reptile tree. mainly to disagreement over the affinities of turtles

with the remaining reptiles, which are all conven-

termed the basis of their tionally “diapsids” on pos-

session oftwo temporal fenestrae(Williston, 1917;

Romer, 1966; Carroll, 1988; Benton, 1996). There

Downloaded from Brill.com10/08/2021 07:17:21AM via free access 2 M.S. Y. Lee - Phylogenetics of diapsid reptiles

archosaurs are two major clades of living diapsids,

(crocodiles and birds) and lepidosaurs (Sphenodon

and and squamates). The monophyly of archosaurs, of lepidosaurs, is extremely strongly corroborated

(e.g. Gauthier, 1984; Benton, 1984,, 1985; Evans,

1984, 1988; Gauthier et ah, 1988; Kemp, 1988;

but Rieppel and de Braga, 1996; Dilkes, 1998; see

Hedges and Poling, 1999 and comments below).

If these two diapsid clades are provisionally ac-

in which turtles cepted, there are three ways might outside be related to diapsids: they might fall a

monophyletic Diapsida, or might render diapsids

(as currently construed) paraphyletic by being the Fig. I. The three proposed positions for turtles with respect to

sister to either or archosaurs. group lepidosaurs living diapsids. 1 is the “anapsid”hypothesis, 2 is the “lepidosaur”

These three possibilities are here called the “anap- hypothesis, and 3 is the “archosaur” hypothesis.

sid”, “lepidosaur” and “archosaur” hypotheses re-

of diapsid mono- spectively (Fig. I). Each has some recent charac- The widespread assumption and or was challenged by Rieppel ter support from either morphology molecules. phyly recently and 1997), who The anapsid hypothesis is the traditional arrange- deBraga (1996; deBraga Rieppel,

the of detailed of ment. It implies that turtles retain primitively on basis a analysis osteological

the second - that turtles unfenestrated (anapsid) skulls, and that the presence traits suggested arrangement

nested within as the sister to of two temporal openings is one of the diagnostic are diapsids group

and con- derived characters uniting archosaurs and lepido- extant lepidosaurs. DeBraga Rieppel’s

clusions were based on the first saurs as a monophyletic Diapsida (e.g. Williston, comprehensive of basal amniotes (including 1917; Gaffney, 1980; Reisz, 1981; Gauthier, 1984; phylogenetic analysis

assumed to be Benton, 1985, 1996; Gaffney and Meylan, 1988; turtles) where diapsids were not

The results when Carroll, 1988). Under the most recent interpreta- monophyletic. were unexpected: into all its lin- tions of this hypothesis, the nearest relatives to the Diapsida was split component

these lineages did not form a clade. Rather, turtles are primitive anapsid reptiles, in particular, eages,

the cluster of procolophonoids (Laurin and Reisz, 1995) or turtles nested within diapsid lineages, marine as the sister to sauropterygians, a pareiasaurs (Gregory, 1946; Ivachnenko, 1987; group of and Lee, 1997). The hypothesis of diapsid monophyly radiation consisting placodonts, plesiosaurs,

relatives. were the nearest has been supported largely by morphological, and their Lepidosaurs extant

the clade. This in particular skeletal, traits. However, even the relatives of sauropterygian-turtle

also that turtles have skeletal traits are relatively few, and some are implies secondarily

correlated with temporal fenestration (e.g. Reisz, closed their temporal fenestra. However, a reanalysis

1981; Gauthier, 1984; Benton, 1985; Evans, 1988; of deBraga and RieppeTs data showed that, even

ifall their character Laurin and Reisz, 1995). Furthermore, soft ana- codings were accepted as cor-

tomical for their over traits support a radically different arrange- rect, support preferred phylogeny

not ment of amniotes (Gauthier et ah, 1988; Gardiner, the traditional phylogeny was significant

the Further where 1982, 1993; see below). Thus, morphological (Wilkinson et ah, 1997). reanalysis,

evidence for diapsid monophyly is not strong. Nev- some miscoded characters in the data set were with turtles ertheless, (his arrangement was widely accepted, rescored, yielded the traditional tree, and outside and analyses of basal amniote relationships con- grouping with anapsid reptiles falling

However, as tinued to assume diapsids were monophyletic, and a monophyletic Diapsida. relatively

thus were to cause this represented them using only one or two pre- few characters changed topo-

tree not sumably basal forms (e.g. Laurin and Reisz, 1995; logical shift, the revised (traditional) was

Lee, 1995). strongly supported over the heterodox topology

Downloaded from Brill.com10/08/2021 07:17:21AM via free access Contributions to Zoology, 70 (1) - 2001 3

(Lee, 1997). Thus, although deBraga and Rieppel’s sizeable portions of the mitochondrialand nuclear analysis did not demonstrate that turtles were molecular data support the archosaur hypothesis, modified diapsids, it certainly highlighted that the the results are not yet conclusive. traditional anapsid hypothesis is far weaker than Given the disagreement over reptile relationships generally assumed, and that the lepidosaur hypo- and the observation that both the morphological thesis deserves serious consideration.A more recent and molecular evidence is ambiguous, it was de-

combined study (Rieppel and Reisz, 1999) has again argued cided to integrate these datainto a phylo- for a turtle-sauropterygian clade, and discussed a genetic analysis. This study thus revisits an impor- scenario for an aquatic origin of turtles. tant earlier study of amniote relationships (Eernisse

It was suggested that molecular sequences would and Kluge, 1993) that included an earlier version provide independent data to resolve the issue (Wil- of the osteological data (Gauthier et ah, 1988) and kinson et the molecular data the short and ah, 1997). However, only relatively poorly sampled gene have then surprisingly supported neither of the two sequences that were available. That analysis

included arrangements proposed on the basis ofmorphology. also only two “anapsid” groups (turtles

molecular results seemed and thus could the Instead, early most consis- captorhinids), and not test mono- tent - with the third possibility that turtles are most phyly of diapsids with respect to other anapsids, closely related to extant archosaurs (see Eernisse or the possible relationships of turtles with other and Kluge, 1993 for review). A recent study (Zar- anapsids (e.g. pareiasaurs and procolophonoids) doya and included from within combined and molecular Meyer, 1998) sequences a morphological two complete mitochondrial genes(12S rRNA and framework-. Also, that analysis did not include 16S rRNA, totalling 2903 aligned base pairs) from euryapsids, which is a potentially important omis- the two that major lineages of turtles (pleurodires and sion since they are the lineage apparently “pulls” cryptodires), archosaurs (crocodiles and birds), and turtles into diapsids in one study (deBraga and

lepidosaurs and The and The (Sphenodon squamates). op- Rieppel, 1997; Rieppel Reisz, 1999). timal trees underparsimony, neighbour-joining, and present study includes all major anapsid lineages, likelihood the sister and thus address analyses grouped turtles as and euryapsids, can these ques-

to group archosaurs, but the traditional morpho- tions with a combined morphological and molecular

logical of arrangement a monophyletic Diapsida analysis. was almost equally consistent with the data. The Combining data sets can allow secondary phylo- turtle-lepidosaur hypothesis (deBraga and Rieppel, genetic signals to emerge, which are not apparent

1997) was the least supported, but even this ar- when the data sets are analysed individually (Barrett

rangementcould not be statistically rejected (Zar- et al., 1991; Olmstead and Sweere, 1994; Nixon doya and Meyer, 1998). Analysis of complete mito- and Carpenter, 1996). Thus, there is the possibility

chondrial DNA genomes also grouped turtles and that morphological and molecular data sets, which diapsids (represented only by squamates and archo- each give equivocal results, when combined might saurs) to the exclusion of mammals (Kumazawa interact to yield a strong and/or unexpected signal. and Nishida, 1999). Another recent study (Hedges Also, larger data sets (e.g. molecular sequences) and Poling, additional molecular be to smaller data 1999) proposed might expected swamp sets (e.g. support for an even more extreme version of the morphology), since if the strength of the signal

archosaur turtles in hypothesis, in which are placed (per character) is the same both data sets, the within archosaurs, as the would sister group ofcrocodilians larger data set have a larger number ofchar- alone (to the exclusion of its birds). While both nuclear acters supporting arrangement. However, em- and mitochondrial genes were evaluated in this pirical results suggest that this pattern often does

study, the heterodox crocodile-turtle clade is with coherent sup- not occur, a signal in a smaller mor- ported by the in signal found only most (but not phological data set outweighing an ambiguous sig- all) of the nuclear contradicted nal in data; it was by size- the larger molecular set (e.g. see Nixon and

able ' portions of their nuclear data, and all of their Carpenter, 1996; Baker et al., 1998). Thus, there mitochondrial data (see discussion). Thus, while is no reason to expect that the emergent signal in

Downloaded from Brill.com10/08/2021 07:17:21AM via free access 4 M.S. Y. Lee - Phylogenetics of diapsid reptiles

included and coded the combined analysis will be that of 2000 aligned performed with ichthyosaurs

their nucleotides, rather than that of 200 morphological according to Motani et ai. (1998): inclusion - did not tree characters; it could even be some other signal change topology they always emerged

the sister to without entirely (e.g. turtles as sister group to Sphenodon as group euryapsids changing

- had alone). As the results show, the traditional anapsid existing relationships and only a minor effect

values. because hypothesis (and thus diapsid monophyly), despite on bootstrap Thus, their exclusion

have effect of the above does a repeated recent challenges, remains the most strong- problems not major results. ly supported by the combined data sets. The turtle- on the

but anatomical data set was derived from archosaur clade remains a possibility, the turtle- The soft

the review in Gauthier et al. To lepidosaur clade can be statistically rejected. detailed (1988).

make the terminal taxa in that analysis comparable

to those used in the osteological analysis, crocodiles

combined into the Archosauri- Morphological and molecular data sets and birds were taxon

formes, and lepidosaurs were split into Sphenodon

and The data matrix includes all 40 The osteological data set of deBraga and Rieppcl squamates. that characters identified in that were infor- (1997) was revised and expanded, yielding a total study

between mam- of 176 characters for 28 ingroup taxa. New char- mative with respect to relationships and acters, and modifications to existing character mals, turtles, archosauriforms, Sphenodon

The used to root definitions and codings, are listed in the appendix. squamates (appendix). outgroup with the condition found in Most of these modifications have been previously the tree was coded

the relatives of am- published by Lee (1997a) and subsequently accepted lissamphibians, nearest extant

niotes. The soft anatomical characters discussed (Ricppel and Rcisz, 1999), and are thus not con-

in Gardiner and are reanal- tentious. However, there remains disagreement over (1982) Lovtrup (1985)

in Gauthier et al. and as such have some characters, and the reasons for the codings ysed (1988),

in the been included in the present study. adopted in this study are presented appendix, data consists of the 12S along with references that discuss and/or illustrate The molecular complete and I6S rRNA mitochondrial DNA for these problematic characters. In order to reduce sequences

and ar- computing time, the two procolophonoids, and the mammals, turtles, Sphenodon, squamates, 1120 three pareiasaurs, each coded separately in that chosaurs. Of the total of 2903 sites, ambigu-

here into terminal sites were excluded, a total analysis, are combined single ously aligned leaving

of 1783 sites for and Meyer, taxa (Procolophonoidea and Pareiasauridae respec- analysis (Zardoya if 1998). were to evaluate there tively). This is not a problematic simplification since Analyses performed in the excluded sites: in the analyses of the original (deBraga and Rieppel, was any phylogenetic signal

was none below). in order to make 1997), and recoded data sets, the procolophonoid there (see Again,

lin- 'the terminal taxa used in the molecular lineages formed a clade, as did the pareiasaur analysis

the with those in the other various eages. For the osteological data set, outgroup comparable analyses, in the molecular matrix had to be used to root the tree was coded with the condition species grouped

The and found in diadectomorphs, the nearest relatives of into higher taxa. monotreme, marsupial,

were combined into a amniotes (Gauthier et ah, 1988). Ichthyosaurs, an two placental sequences single

“mammal” of the os- important taxon omitted from the original analysis, sequence (representing part

taxon the and were added to the data set by Motani et al. (1998): teological Cynodontia), plcurodire

into a “turtle” however, their codings were questionable as they cryptodire sequences single sequence

of the taxon and the were based on a single (albeit primitive) ichthyo- (part osteological Testudines),

rather than of basal bird and crocodile into a “Archo- saur, on a range ichthyosaurs. sequences single

for these saur” of the taxon Furthermore, they are being recoded traits sequence (part osteological

by other workers (M. Caldwell and P. Spencer, in Archosauriformes). Construction of such “consen-

sus” done follows. When the two prep). Hence, they are not included in the formal sequences was as

turtles and both had the analysis here. However, exploratory analyses were (a pleurodire cryptodire)

Downloaded from Brill.com10/08/2021 07:17:21AM via free access Contributions to Zoology, 70 (I) - 2001 5

same nucleotide at a given site (e.g. “A”), the structures of doubtful homology (e.g. hair and

“testudine” scored with that nucleo- the sequence was feathers), or are cladistically uninformative, tide at that site. However, at sites where they differed derived state being uniformly present in only one

(e.g. the pleurodire with a “T” and the cryptodire terminal taxon (e.g. subclavian artery origin, uri- with a the “testudine” scored Gardiner’s doubt includes “C”), sequence was nary bladder). study no as polymorphic (“TC”). This approach of recon- some novel and phylogenetically useful traits that structing consensus sequences for higher taxa, have been overlooked by previous workers, but

used in the and commonly morphological analyses, is now because state definitions, distributions, need being increasingly used in molecular studies as well to be more rigorously assessed, these characters

and (e.g. Zravy et ah, 1998), yields more accurate are not included in this study. results than discussed nuclear arbitrarily using a single species to The recently sequences (Hedg-

in also been represent a large clade an analysis (Wiens, 1998; es and Poling, 1999) have not considered

Bininda-Emonds et ah, 1998). Analyses of the here, because the full details of that study could molecular data be due and alone, with the terminal species com- not published to space considerations, bined into and consensus sequences in this manner, the presumed alignments phylogenetic analyses yielded almost identical results (see below) to the are currently being reevaluated by several workers. previous of the analysis Zardoya and Meyer (1998), Though the crocodile-turtle clade proposed on where the terminal of the nuclear data contradict species were coded individually. basis appears to the

This that the of of the terminal suggests consensus sequence approach monophyly one taxa used in this

the the faithfully retained phylogenetic signal that was analysis (Archosauriformes), support for this present in the weaker original data set. heterodox clade appears than suggested. The combines what to be The turtle-crocodile-bird was present study appear trichotomy interpreted recent, resolved comprehensive data sets for each category by Hedges and Poling (1999) as in favour of traits. Of course, not all characters have been of a crocodile-turtle clade. However, four of the

included. There doubtless other nuclear a crocodile-bird are many osteo- analyses supported clade, logical, soft anatomical and molecular characters while two other nuclear analyses supported a turtle- that have been discussed in the literature: no study bird clade. These latter clades were also supported

can hope include of their mitochondrial which to every trait in every category. by some data, were In the in particular, soft anatomical traits of Gardiner not included the “combined” analyses. Thus, a

(1993) and the DNA data of and ofthese would nuclear Hedges more conservative interpretation data Poling (1999) were not used, for reasons discussed be to treat the turtle-crocodile-bird trichotomy as below. Also, the available complete mitochondrial unresolved, which is consistent with the widespread

genomes (Kumaza and Nishida, 1999) do not yet assumption of monophyly ofArchosauriformes. For include been the of of Sphenodon, and thus also have not purposes this study, subdivision Archo-

used in this because analysis. sauriformes is not necessary all three data

Gardiner considered here (1993) listed additional traits supporting sets individually strongly support the Haematothermia hypothesis (a bird-mammal the monophyly of Archosauriformes: the osteology clade); however, these newer data require rigorous (e.g. Gauthier et ah, 1988; de Braga and Rieppel, reassessment. Only 25 of the 97 characters used 1997) the soft anatomy (Gauthier et ah, 1988), in Gardiner’s (1993) the the DNA and analysis are discussed; and mitochondrial (Zardoya Meyer, remainder are only briefly tabulated. Many of the 1998) Even if the Archosauriformes was split and

inadequately defined characters for appear to have been coded as several taxa, the data considered here miscoded but a these proper assessment of the correct taxa would have emerged as a monophyletic character states for various taxa can only be done group in all the individual and in the combined after these states are defined. The there is no to properly major- analyses. Thus, necessity split up of ity the 25 characters that are discussed have Archosauriformes for this analysis. However, in eithei been the inadequately surveyed across amniotes order to rigorously evaluate crocodile-turtle

(e.g. true sleep”), lump together vaguely similar hypothesis in a future analysis involving nuclear

Downloaded from Brill.com10/08/2021 07:17:21AM via free access 6 M.S. Y. Lee - Phylogenetics of diapsid reptiles

genes (which might not support monophyly ofAr- cal and molecular analyses each includedonly extant

it will be subdivide chosauriformes), necessary to taxa (since fossils could not be scored for either

the taxon into its component lineages: Crurotarsi class of trait).

(crocodiles and stem-group relatives), Ornithodira The partition homogeneity (= incongruence

and and all (birds stem-group relatives), more basal length difference) test (Farris et al., 1994) as imple-

archosauriform outgroups to these two clades, such mented in PAUP* was applied to see ifany of the

as Proterochampsidae, Euparkeria, Erythrosuchidae, data sets were significantly incongruent. This test

and Proterosuchidae (e.g. Benton and Clark, 1988; randomly assigns characters to partitions equal in

Gower then Sereno, 1991; Juul, 1994; Witmer, 1997; size to the original partitions, and compares

and Weber, 1998). the between-data set incongruence of the rando-

mised partitions to that in the original partitions.

These tests will be inaccurate if invariant characters

Data analyses are unequally distributed in the original data sets

(e.g. if they are only present in molecular partitions

but in The three data sets here were analysed according not morphological partitions), as the randomi-

to maximum parsimony using PAUP* (Swofford, sation procedure will assign them to both resampled

1999), the only analytic method generally applicable data partitions. It has been recommended that such

of characters be the to all three types traits. (Other analyses applicable deleted before employing test

to the molecular data, i.e. maximum likelihood and (Cunningham, 1997). The same arguments apply

neighbour joining, have already been performed to other uninformative (e.g. autapomorphic) char-

and which should therefore be deleted well by Zardoya Meyer (1998), and yielded very acters, as

similar results to the parsimony analysis). In all (Lee, 2000). Accordingly, two sets of partition

analyses, characters were equally weighted and all homogeneity tests were performed, one including

multistate characters unordered. The all characters, and the other with uninformative were permu-

tation tail probability (PTP) test (Archie, 1989; Faith (invariant and autapomorphic) characters deleted. and Cranston, 1991) was employed to identify phy- Partition homogeneity tests might also be adversely

if logenetic structure in the data, but the results were affected characters with large amounts of missing

data interpreted conservatively (Slowinski and Crother, data are unequally distributed in the original

1998). Bootstrapping frequencies (Felsenstein, sets (e.g. DNA and soft anatomical characters can

1985) were based on 500 heuristic searches in be scored only for living taxa, while osteological

be PAUP: all cladistically uninformative characters characters can potentially scored for all taxa).

were excluded In the randomised data traits scorable in liv- during resampling since large num- sets, bers of these characters in molecu- forms and in all will be (e.g. as many ing only, forms, again

lar data sets) have been demonstrated to lower distributed evenly throughout the data partitions,

bootstrap values (Carpenter, 1996). In order to which will therefore again not be representative

ofthe standardise results, therefore, it seems best to de- original partitions. However, unlike the case

lete all uninformative characters from with uninformative there is no bootstrap characters, easy way

analysis. The alternative, of including such char- to correct for this bias. Deleting fossil taxa will

is difficult make traits in each scorable in all acters, as there are potentially an infi- partition taxa,

nite numberof“uninformative” characters for each but might also significantly change the phyloge-

class of netic in the data. Bremer support (Bremer, 1988) for signal osteological data set (Gauthier each elade was calculated using PAUP commands et ah, 1988; Donoghue et al., 1989; Lee, 1998; see

TreeRot generated in (Sorenson, 1996), which were results below).

modified is whether so that each search employed 100 rather It arguable significant incongruence

than 10 random addition replicates. Initially, the justifies keeping data sets separate during analyses

three data sets Bull et and were analysed separately. The os- (e.g. see al., 1993; Nixon Carpenter,

data was all Ballard of teological analysed two ways: taxa 1996; et al., 1998). Thus, regardless the

and ofthe included, only extant taxa. The soft anatomi- results partition homogeneity tests, combined

Downloaded from Brill.com10/08/2021 07:17:21AM via free access Contributions to Zoology, 70 (1) - 2001 7

analyses were performed, incorporating all 176 os- teological, 40 soft anatomical, and 1783 aligned molecular characters. Two such combined analyses were undertaken: one including all taxa (including fossil taxa, which could not be scored for soft anato- mical or molecular traits), and one including only taxa with living representatives (i.e. terminals which could be scored for all traits). Comparison of the results of these two analyses would demonstrate whether fossils could have any effect on tree topo-

logy in a combined analysis where they could only be scored for 176 characters (at most) out of nearly 2000.

Phylogenetic results

Separate analyses

The analysis of the 176 osteological characters,

including extant and fossil taxa, yielded a single

most = parsimonious tree (length 771, consistency

index = 0.54, retention index = 0.67) consistent with the traditional, anapsid hypothesis (Fig. 2). The PTP test found significant phylogenetic sig- nal in this data set (P= 0.0057). The relationships

discovered the among living forms were mammals Fig. 2. Thephylogeny of amniotes supported by the osteological

(turtles(archosaurs+lepidosaurs)). Turtles fell out- data alone,by the morphological data (osteology + soft anatomy),

side and by the entiredata set (osteology + soft anatomy+ molecules). diapsid reptiles and grouped strongly with anap- Fossils included in all taxa in bold sid are analyses. Living text, “parareptiles”; in particular, with pareiasaurs. fossils in plain text. Turtles are related to anapsids, and diapsids All diapsid reptiles, including fossil forms, formed arc monophyletic. Bremer and bootstrapping support for each a robust clade (Bremer and bootstrap support for of these analyses in this order is listed at each node. all groups shown in Fig. 2). When the osteological analysis was performed with only living taxa, three most parsimonious trees sidered. When fossils are ignored, the osteology icsulted (L=323, CI=0.84, RI=0.40): turtles either is equally consistent with all three hypotheses. fell outside diapsids, or with grouped lepidosaurs, The characters soft anatomical supported a very °i with grouped archosaurs (i.e. the anapsid, heterodox tree (L=63, CI=0.70, Rl=0.61) which lepidosaur, and archosaur all hypotheses were turtles placed as the sister group of an archosaur- equally supported). The strict consensus of the three mammal clade This is (Fig. 4). arrangement very trees is shown in Figure 3. The PTP test found the similar to that proposed by Gardiner (1982, 1993) data set contained a significant phylogenetic signal and Lovtrup (1985), who both considered many (P-0.007), presumably concentrated at the two of these but does characters, not appear to have icsolved clades in the strict consensus (reptiles and other modern advocates. be any It can concluded it lepidosaurs). Thus, appears that the osteological that the soft anatomical data by itself does not data strongly supports diapsid (i.e. the monophyly strongly support any of the three currently accepted anapsid when hypothesis) only fossil taxa are con- for positions turtles. The PTP test again suggested

Downloaded from Brill.com10/08/2021 07:17:21AM via free access 8 M.S. Y. Lee - Phylogenetics of diapsid reptiles

Fig. J, The phylogeny ofamniotes supported by the osteological Fig. 5. The phylogeny of amniotes supported by the molecular

the entire set + soft data alone when fossil taxa are excluded. The tree is a strict data alone, and by data (osteology anatomy

which + when fossils are Numbers nodes refer consensus of three most parsimonious trees, represent molecules) ignored. at

all three proposed positions for turtles(Fig. 1). Numbers at nodes to Bremer supportand bootstrap frequency for the two respective

Bremer refer to support and bootstrap frequency. analyses.

The PTP test indicated significant phylogenetic

structure (P=0.008).

Analyses including all (well and poorly-aligned)

sites yielded an identical tree, and almost identi-

cal branch supports, to those including only well-

aligned sites. Analyses including only the poorly

sites tree but with aligned gave a single extremely

all clades had a of poor support on (all bootstrap

less than50% and a Bremer index of 1). Relaxation

of parsimony by a single step produces a totally

unresolved tree. The poorly-aligned sites also failed

the PTP test (P=0.21). These results indicate that

sites had Fig. 4. The phylogeny of amniotes supported by the soft anatomy exclusion of the poorly aligned almost alone. Numbers at nodes refer to Bremer and that support bootstrap no effect on the phylogeny, and there is almost frequency. in It no signal the poorly aligned sites. is recom-

mendedthat these tests should be routinely perform-

ed sites assumed to be significant phylogenetic signal in this data set on excluded ambiguously

few do (P=0.001). aligned; at the moment studies this, and

influence of excluded sites is not The analysis of the 1783 unambiguously-aligned the explicitly

nucleotide sites yielded a fully resolved tree (L= investigated.

2785, Cl =0.93, RI=0.35) supporting the Archo-

sauria hypothesis: mammals(lepidosaurs(turtles+

archosaurs)). The high Cl is an artefact of the inclu- Homogeneity tests

sion of large numbersofautapomorphic characters;

the index 0.80 each falls to when such characters are The partition homogeneity test was applied to excluded. of data All clades received moderate to strong possible pair sets, and to a comparison

the entire support (bootstrap around 70% or more, Bremer between morphological (osteological plus

>7), including the turtle-archosaur clade (Fig. 5). soft anatomical) and molecular data (Table 1). For

Downloaded from Brill.com10/08/2021 07:17:21AM via free access Contributions to Zoology, 70 (I) - 2001 9

Table L The results of partition homogeneity (incongruence when uninformative characters are excluded, as

length difference) tests for various comparisons. recommendedhere, this inexplicable positive result

with ail Partitions All Characters Informative Chars disappears, comparisons having insignifi-

P Only cant values (P>0.13).

Osteology vs Soft

Anatomy P = 0.95 P = 0.95 Combined = analyses Osteology vs Molecules P = 0.77 P 0.33 Soft Anatomy vs

Molecules = = P 0.037* P 0.13 The analysis of the combined 216 morphological All Morphology vs (osteological and soft anatomical) characters, with

Molecules P = 0.68 P = 0.21 all taxa included, yielded a tree (L=837, CI=0.55,

* Denotes = identical in to the significant at P 0.05. RI=0.66) topology osteological

tree with all taxa, and with very similar Bremer

and values When all 1999 char- the soft bootstrap (Fig. 2). anatomy versus molecules comparison, only acters (osteology, soft anatomy, molecules) were living forms were included since fossil forms could

included, the same tree results (L=3630, CI=0.84, not be scored for either data set. For all other com- RI=0.59), again with very similar bootstrap and parisons, all taxa (living and fossil) were included, Bremer 2). This tree is consistent since the supports (Fig. latter could be scored for one of the two with the anapsid hypothesis: turtles(lepidosaurs+ data sets. Two analyses were performed, with all archosaurs). of extant diapsids, in re- characters Monophyly included, and with uninformative char- lation other was to extant amniotes, strongly sup- acters excluded. No significant heterogeneity was ported in all these analyses. Moreover, the positions detected in seven of the eight comparisons: the ex- of the fossil taxa in these combined analyses were ception was the comparison involving the soft identical to that found in the osteological analysis. anatomy and the molecules with all characters that fossils and characters It appears the osteological included. However, when only informative char- largely determinedthe results of the combined anal- acters were included the soft anatomy and molecular yses, even the latter a small data though comprised only sets arc not significantly incongruent. proportion of the total data set (176 out of 1999 The arguments that uninformative characters characters). should be excluded before applying this test (see In order to test the lepidosaur and archosaur Lee 2000) are supported by these results (Table hypotheses against the osteological, morphological 1 )• Firstly, excluding uninformative characters sub- and entire data sets, which all yielded the same stantially changed the resultant P values in all phylogeny, the “backbone constraints” function in comparisons for the soft vs oste- except anatomy PAUP was employed to find the best trees consistent ology. I his is because all comparisons the except with the following relationships between extant taxa: latter contained a molecular partition which had uninformative many characters. The soft anatomy A. mammals((lepidosaurs)(turtlcs+archosaurs)), i.e. vs osteology comparison involved two morpho- the “archosaur” hypothesis. data logical sets, both with few uninforma- very B. mammals((lepidosaurs+turtles)(archosaurs)), i.e. tive characters, so exclusion of the latter did not the “lepidosaur” hypothesis. change the test results. Secondly, employing the test with all characters included rather produces In each of these analyses, all fossil taxa were allowed counterintuitive results. In particular, the osteology to “float”, i.e. insert anywhere in the backbone is very congruent with both soft anatomical (P=0.95) phylogeny ofextant forms. The most parsimonious and molecular (P=0.77) data, that the consistent with suggesting trees each backbone constraint were attei two data sets should also be quite congruent found using three data sets: the osteological data with one another. Yet, the test that the suggests alone, the morphological data (osteology + soft molecular and soft anatomical data in- are highly anatomy), and for the entire data (osteology + soft

congruent with one another (P=0.037). However, anatomy + molecules).

Downloaded from Brill.com10/08/2021 07:17:21AM via free access 10 M.S. Y. Lee - Phylogenetics of diapsid reptiles

significantly worse than the most parsimonious

(traditional) tree for the osteological, and for the

combined morphological (osteology + soft anatomy)

data (Table 2). However, for the entire (morpho-

logical + molecular) data, these trees are not signifi-

than cantly worse the most parsimonious tree (Table

2).

For backbone constraint B (the lepidosaur hypo-

thesis), the same two most parsimonious trees were

found for all three data combinations(Fig. 7). These

trees both grouped turtles with sauropterygians in

a lepidosauromorph clade, and are very similar to

the arrangement proposed by deBraga and Rieppel

(1997). Templeton tests (Table 2) indicate that both

trees are significantly worse than the most parsi-

monious (traditional) tree for the osteological, and

the combined morphological data. For the entire

data the difference under set, verges on significant

the two-tailed test (P=0.059 and P= 0.064) and is

significant under the more liberal one-tailed test

(P=0.029 and P=0.032). As the Templeton test is

conservative and often underes- rather appears to Fig. 6. The two most parsimonious trees consistent with the timatedifferences(Larson, 1994), and the one-tailed “backbone” constraint ofa turtle-archosaur clade to the exclusion

test is here the alternative tree of other living reptiles. These same two trees are found in anal- more applicable as

the yses employing osteological, morphological (osteology + is known to be shorter a priori, these results thus

soft and entire + soft +molecules) anatomy) (osteology anatomy constitute evidence against the lepidosaur hypo-

data. Living taxa in bold text, fossils in plain text. thesis.

Thus, the osteological and combined morpho-

For backbone constraint A (the archosaur hypo- logical data sets both support the traditional, anapsid

and thesis), the same two most parsimonious trees were hypothesis and statistically reject the archosaur

found for all three data combinations (Fig. 6). Both lepidosaur hypotheses. The entire (morphological

trees grouped turtles with rhynchosaurs in the archo- and molecular) data also supports the anapsid hypo-

sauromorph clade. Templeton tests (Templeton, thesis over the other two hypotheses, and can sta-

1983; Larson, 1994) indicate that these trees are tistically reject the lepidosaur, but not the archo-

Table 2. Support for the most parsimonious trees consistent with the turtle-archosaur and turtle-lepidosaur hypotheses compared to

the most trees to entire data parsimonious (turtle-anapsid) tree. The trees are illustrated in Figs 2, 6 and 7. The same apply the set, the

morphology, or just the osteology. The extra length, and two- and one-tailed Templeton test results, for each tree and each data set are

shown.

MPT’s consistent with Entire data Morphology Osteology alternative hypotheses (Ost+Soft+Mol) (Ost+Soft)

= = turtle-archosanr 1 (Fig. 6) + 10, P =0.34, 0.17 + 18, P 0.029**, 0.015** + 16, P 0.045**, 0,023**

5 turtle-archosaur 2 (Fig. 6) + 10, Z = 0.36, 0.17 + 18, P = 0.029**, 0.015** + 16, P = 0.045**, 0.023**

= turtle-lepidosaur 1 (Fig. 7) +19, P =0.059*, 0.029** +20, P 0.015**, 0.008** + 17, P = 0,035**, 0.018**

turtle-lepidosaur 2 (Fig. 7) + 19, P =0.064*, 0.032** +20, P =0.018**, 0.009** +17, P = 0.041**, 0.020**

* ** = significant at P<0.1, =significant at P<0.05

Downloaded from Brill.com10/08/2021 07:17:21AM via free access - 11 Contributions to Zoology, 70 (1) 2001

of amniotes the combined Fig. 8. The phylogeny supported by

morphological (osteological and soft anatomical) data, when

Numbers at nodes refer to Bremer fossils are ignored. support

and bootstrap frequency.

consistent with the archosaur hypothesis.

Thus, when all relevant taxa are considered, the

traditional anapsid hypothesis is supported in the

morphological and combined (morphological +

molecular) analyses', mainly dueto a strong signal

fig- 7. the The two most parsimonious trees consistent with from the osteology that is sufficient to override “backbone” to the exclusion constraint ofa turtle-lepidosaur clade mole- the weaker signals from the soft anatomy and ot other living reptiles. These same two trees are found in analyses cules. However, if fossils are ignored, the osteo- employing the osteological, morphological (osteology + soft and the heterodox and + logical signal disappears arrange- anatomy) entire (osteology + soft anatomy molecules)

data. in bold the molecular remains Living taxa text, fossils in plain text. ment supported by sequences

uncontradicted, even in the combined analyses that

saur, hypothesis. In terms of rank order, the anapsid include osteology. hypothesis is the most supported in all three analy-

ses, followed by the archosaur, with the lepidosaur hypothesis and the being the least supported (Table 2). Molecules, morphology importance of

Interestingly, when fossil taxa are excluded, the fossils

osteological signal for the anapsid hypothesis dis-

and that the appears, the osteological data are equally con- The above results suggest anapsid hypo-

sistent - thus with all three hypotheses (see above results thesis of turtle affinities and diapsid mono- lor osteological data, and Fig. 3). As would be ex- phyly - remains not only tenable, but is the most

pected, when therefore, when fossils were excluded, the supported arrangement the morphological,

combined is considered simul- analyses were not greatly influenced by molecular and fossil evidence the osteological data. The combined 216 morpho- taneously. The results are thus consistent with those

logical amniote based (osteological and soft anatomical) characters of an earlier study of relationships

yielded a tree (L=389, CI=0.81, RI=0.46) consistent on morphology and molecules (Eernisse and Kluge, with the but with also anapsid hypothesis diapsid mo- 1993), which found evidence that the tradi- nophyly very poorly supported (bootstrap=58%, tional, osteological tree was preserved ina combined Bremer=2) (Fig, 8). The entire morphological and analysis. The broad structure of the current tree molecular data set yielded a tree (L=3176, CI=0.91, (Fig. 2) is highly congruent with traditional views,

RI-0.38) with the mod- same topology as that implied and most widely-recognised groupings are by the molecular data alone, and with bootstrap erately to strongly corroborated. The “pelycosaur” hequencies hardly changed (Fig. 5). This tree is and “therapsid” lineages cluster as successively

Downloaded from Brill.com10/08/2021 07:17:21AM via free access 12 M.S. Y. Lee - Phylogenetics of diapsid reptiles

while the molecular workers, sequences weakly

support the archosaur hypothesis. The combined

data supports the anapsid hypothesis but cannot

reject the archosaur hypothesis. Both the anapsid

and archosaur hypotheses therefore merit further

consideration, especially in view ofthe new nuclear

data proposed for a crocodile-turtle clade (Hedges

and Poling, 1999). These results should therefore

and encourage further investigation of anapsid

archosauromorph taxa that might be related to

turtles. If turtles the results here are anapsids, sug-

for gest pareiasaurs are the strongest candidates

The their nearest relatives. If turtles are archosauro- Fig. “conservative” tree based on the results published

in and (1999). Hedges Poling morphs, this study suggests the rhynchosaurs are

worth investigating. Both pareiasaurs and rhyncho-

closer outgroups to mammals as part of a mono- saurs are stout, slow-moving, herbivorous reptiles

phyletic Synapsida (e.g. Broom, 1932; Romer, 1966; with shearing feeding mechanisms, a plausible

All other fossil with for a turtle In Kemp, 1982). reptiles group ecotype ancestor. addition, pareiasaurs

dermal while living reptiles, comprising the Sauropsida. Saurop- possess armour, rhynchosaurs have

sids consist of two major lineages, parareptiles and beaks - two further, highly distinctive turtle traits.

The The and eureptiles. parareptile clade includes, among lepidosaur hypothesis (deßraga Rieppel,

other does not receive from individual things, millerettids, procolophonoids, pareia- 1997) support any

data saurs and turtles (e.g. Laurin and Reisz, 1995; Lee, set (osteology, soft anatomy, mitochondrial

1995). Within parareptiles, the nearest relatives of DNA, or nuclear DNA), is statistically rejected by

turtles the combined and to be the least are pareiasaurs, slow-moving, armoured analysis, appears

herbivores (e.g. Gregory, 1946; Lee, 1995). These likely of the three. Ifturtles are forced to be lepido-

results indeed thus differ from those ofEernisse and Kluge sauromorphs, they group with the marine

(1993), which grouped turtles with captorhinids. sauropterygians, as suggested by Rieppel and col-

Here, captorhinids, protorothyridids, and a mono- leagues (deßraga and Rieppel, 1997; Rieppel and

to form clade Reisz, this is phyletic Diapsida group together, a 1999). However, tree significantly

ofEureptiles (Laurin and Reisz, 1995; Lee, 1997). worse than the most parsimonious tree. There is

Within diapsids, araeoscelids, younginiforms and thus no phylogenetic support for an aquatic origin

certain noted other early forms are basal lineages out- of turtles. Also, it should be that all basal side the extant crown-clade (Laurin, 1991; deBraga turtles described to date (e.g. Proganochelys, and Rieppel, 1997; Motani et al., 1998). Finally, Australochelys, Palaeochersis and Proterochersis)

be advanced, crown-clade diapsids (Sauria) consist appearto terrestrial, being found in fully orpartly of diverse sister terrestrial and tortoise-like two groups, lepidosauromorphs deposits exhibiting and archosauromorphs (Gauthier, 1984; Benton, features such as extremely short, robust digits and

1985; Evans, 1988; Gauthier et al., 1988; deBraga domed shells (Rougier et al. 1995). and Previous have demonstrated Rieppel, 1997). The marine sauropterygians - surveys how mor-

placodonts, - can plesiosaurs and relatives are part of phological traits drive the results of combined

the and Lepidosauromorpha (deBraga and Rieppel, morphological molecular analyses, even when

1997; Motani et al., 1998; Rieppel and Reisz, 1999). they are in the minority (Nixon and Wheeler, 1996;

Baker When the data sets are analysed separately, the et ah, 1998). In this analysis, however, it is results are more equivocal. The osteology strongly notable that the traditional tree is not retrieved when

the supports anapsid hypothesis when all taxa are morphological traits are added to the molecular

the soft considered, anatomy gives a heterodox ar- data set without increasing taxonomic sampling.

rangement considered untenable by almost all Rather, the additionof morphological traits together

Downloaded from Brill.com10/08/2021 07:17:21AM via free access Contributions to Zoology, 70 (1) - 2001 13

with fossil taxa is required to change tree topology. consistent with the “anapsid”, “lepidosaur” and

While of fossil while the the phylogenetic importance taxa “diapsid” hypotheses (Fig. 1), soft ana-

has been disparaged in the past (e.g. Patterson, 1981; tomical data produces a highly heterodox phylogeny

Gardiner, 1982), it is now well establishedthat fossil inconsistent with all three osteological trees (Fig.

taxa the anatomical can overturn even well-corroborated clades 4). However, when soft traits are

found in of with the of morphological analyses only extant taxa combined osteology, one the three os-

Gauthier - the - (e.g. et al., 1988; Donoghue et al., 1989; teological trees “anapsid” tree emerges as Lee, 1998). However, most of these examples in- best supported. This occurs because the soft ana- volve morphological analyses which have focused tomical data is more consistent (or less inconsis-

on the with osteological or hard-part characters determin- tent) with “anapsid” osteological tree than able in fossils. The importance of fossils in such the other possible osteological trees. Thus, even

“osteocentric” analyses is not surprising, since if though at first glance the soft anatomical datamight

for little it sufficiently well-preserved they can be scored seem to contain information since implies

all bizarre it still characters. The results here demonstrate that a rather tree, serves to arbitrate be-

fossils can three also play a decisive role in combined tween the equally-parsimonious osteologi-

of the analyses hard parts, soft anatomy, and molecules. cal hypotheses. Thus, affinities of turtles are A similar and unresolved when the is pattern occurs in a morphological osteology analysed sepa-

molecular resolved when soft analysis of squamate phylogeny (Lee rately, strangely the anatomy and - but resolved Reeder, unpubl. data), where tree topology is is analysed separately (albeit weakly) determined by three critical fossil taxa. The ef- in favour ofiihc traditional view when the data sets fects of fossils This is a were not explicitly addressed in the are combined. good example of a secondary

study of in a amniote relationships by Eernisse and phylogenetic signal emerging combined analy- Kluge when (1993), but as theircombined (morphologi- sis, which is not apparent the data sets are

cal + molecular) tree was identical to their mor- partitioned and analysed separately (Barrett et al.,

phological Nixon and tree, and their morphological tree was 1991; and Wheeler, 1996; Wiens Reeder, shown elsewhere to be structured by fossil taxa 1997).

(Gauthier et be also underscores the al., 1988), the same effects can This study observation that,

inferred. these molecular data often contain Thus, studies challenge the tempt- while sets many more ing assumption that fossils might be unimportant characters than morphological ones, when only

m combined informative the molecular and morphological analyses, characters at relevant levels are because they can be scored for few characters considered, the difference can be much less very pro-

(only a nounced. This is because molecular studies varying proportion of the “hard part” mor- sequence phology). A certain recent analysis ofrelationships within all sites in a region. This increased objec- turtles based on morphology and molecules (Shaffer tivity, however, means that most characters in et al., data will 1997) showed a much weaker effect of fossils aligned sequence sets not be cladistically consistent with this assumption: in that combined informative. There will often be alignment ambi- analysis, addition of fossils did not change topology guities which will exclude certain segments from but reduced support for certain clades, because they analysis. Of the aligned sites, many will either be contained large amounts of missing data and/or invariant or cladistically uninformative (e.g. each inserted along branches leading to (previously) well- derived condition characterising only a single ter- supported clades. The results here 'demonstrate that minal taxon). Finally, of the remaining sites which ossils are not in always so indifferent, but can contri- are cladistically informative molecular analyses,

ute and strong unique signals in phylogenetic some will not be informative at the relevant levels analyses. but serve to unite species which are already “known” Another For the interesting result concerns the separate to be closely related. instance, analysis of and combined analyses of the osteology and soft Zardoya and Meyer (1998) was aimed at testing anatomy, when fossils are omitted. The osteological 'relationships between mammals, turtles, archosaurs

analysis produces three their most-parsimonious trees, and lepidosaurs. However, terminal taxa were

Downloaded from Brill.com10/08/2021 07:17:21AM via free access 14 M.S. Y. Lee - Phylogenetics of diapsid reptiles

species, and thus not all the aligned, cladistically molecular traits. Hence, it is not surprising that

informative characters helped illuminate the rela- the strong morphological signal regarding relation-

tionships between these four groups, i.e. they are ships between turtles, lepidosaurs and archosaurs

not informative the For the at relevant level. some managed to overwhelm more ambiguous mo-

informative sites, the derived state united only the lecular signal addressing these relationships. Not

the two turtles, or two archosaurs, or the two lepido- only that, but the addition of the 2903 (1783 aligned)

saurs, or various mammal clades, and were thus base pairs of molecular data to the morphological

effectively “autapomorphies” of widely-accepted data set had only minor effects on the Bremer and

terminal taxa (turtles, archosaurs, lepidosaurs, bootstrap support for all clades. There is certainly

mammals). In other sites the derived state cuts across a strong phylogenetic signal in both the morphologi- the four putatively monophyletic groups of inter- cal and molecular data sets, as shown by the PTP

for est: uniting, instance, a turtle(e.g. the pleurodire) test results. However, while a large portion of the

and a mammal(e.g. the monotreme); such charac- signal in the morphological data contributed to the

ters were also effectively uninformative with respect resolution of relationships between turtles, archo-

to the question being investigated (relationships saurs, and lepidosaurs, much of the molecular signal

four between the groups), since they are equally consisted of support for parts of the tree (e.g. mono-

consistent with of the of any phylogenetic hypotheses phyly mammals, therians, lepidosaurs, archosaurs

tested. it and being These comments, should be em- turtles) not directly relevant to the problem

phasised, only apply to parsimony analyses. Like- under investigation (turtle-diapsid relationships).

lihood and neighbour-joining analyses make use Finally, although an attempt to integrate the

of more all of the data. likelihood nuclear or se-quence In data (Hedges and Poling, 1999) into these

and analyses, for instance, invariant sites, changes analyses is currently premature since these data within putatively monophyletic groups, will con- are currently being reevaluated and supplemented

tribute towards calculations of several it be worthwhile gamma parameters, by groups, might specu-

and invariant sites, transition-transversion ratios, lating how the traditional arrangement supported etc. in this study might be reconciled with the hetero-

The same problems of course also affect mor- dox arrangement proposed in that study. As stated

data: phological many morphological characters in the introduction, a conservative interpretation

will the be uninformative at relevant levels, being of their datasuggests a turtle-bird-crocodylian tri- of dubious homology (= “alignment”), or invariant chotomy (Fig. 9). If this interpretation is adopted, in all taxa, or autapomorphies of a single terminal there is a single unrooted network for reptiles that

taxon of accepted monophyly, or convergently is consistent with both the traditional arrangement, present only in derived members of accepted mono- and the heterodox phylogeny proposed by Hedges

phyletic terminal taxa. However such characters and Poling (Fig. 10, II). If the tree is rooted at

arc usually deliberately culled frommorphological turtles, the traditional topology retrieved in the

data sets before analysis - whether or not this is current study results. If, however, this tree is rooted advisable (e.g. Yeates, 1995). Thus, unlike in mo- at squamates, a tree consistent with Hedges and lecular analyses, almost all the characters in the Poling’s (conservative) tree results, with the

relevant morphological data sets are directly to squamates and Sphenodon being basal to a turtle- resolving relationships among the groups of in- crocodile-bird clade.

terest. The methodology ofmorphological system- There remains the problem of deciding which atics is therefore eclectic and but more subjective, rooting position is more likely to be correct. A

it means that the information of such data density rooting problem is not as likely to affect the mito-

sets is In this higher. analysis, although initially chondrial or morphological data sets considered

the data consisted of 216 morphological and 2903 here since the outgroups to reptiles in both these molecular traits, when only traits cladistically in- data sets were well-sampled: basal mammals (mar-

formative at the relevant there levels are considered, supials and monotremes) as well as eutherians were

204 are actually morphological traits and 288 included in the mitochondrial and soft anatomical

Downloaded from Brill.com10/08/2021 07:17:21AM via free access Contributions to Zoology, 70 (I) - 2001 15

data sets, along with the next extant outgroup (am-

The phibians). osteological data set used all mam-

fossil and mals as well as synapsids (pelycosaurs

therapsids), and again employed the next nearest

scorable outgroup (diadectomorphs). Thus, the root-

ing near turtles implied by the combined signal

from these data seems reasonably robust, given the

exhaustive characterisation of the outgroups. How-

is to affect ever, a rooting problem more likely the

nuclear data, since only two derived eutherian mam-

mals (rodent and primate) were used and more basal

eutherians (e.g. edentates), marsupials and mono-

not This observation is tremes were sampled. par-

F‘S- 10. The unrooted network consistent with all three hypo- ticularly interesting since analyses of the 12S and theses of turtleaffinities, and the recent proposed different arrangement 16S sequences using samples of mammals by Hedges and Poling Rooting at position 1 yields the (1999). variable results. to root a “reptile” tree gave highly traditional “anapsid” hypothesis, position 2 gives the “archosaur” Under parsimony, for instance, four different rooting hypothesis, position 3 gives the “lepidosaur” hypothesis, and

were three ofthese position 5 a tree that of and positions found, corresponding yields very similar to Hedges Poling (see Fig. 9). to the “anapsid”, “archosaur”, and “lepidosaur” hy-

1 F Ve t*'^erent d network in and the of mammalian roote trees consistent with the unrooted Fig. 10, combinations outgroups that ' iJt yield each particular The results of likelihood are listed. Note how rooting. a and parsimony analyses outgroup sampling can c range the - - rooting position - and thus, the preferred tree. P - Platypus, O Opossum, W Whale, H Human. OP means the tree is 06 W1'b only the and with the etc. Note that two opossum platypus, W means the tree is rooted only whale, equally-parsimonious rees were found when the tree was rooted with the human and the whale (*).

Downloaded from Brill.com10/08/2021 07:17:21AM via free access 16 M.S. Y. Lee - Phylogenetics of diapsid reptiles

and the potheses, other being very similar to the any analyses, morphological or molecular, that fail

and tree of Hedges Poling, with lepidosaurs (Sphe- to sample outgroups adequately.

nodon and squamates) being paraphyletic with

respect to a turtle-archosaur clade (Fig. 10, 1 IE).

Inclusion of all mammal outgroups, and most Acknowledgements

of the subsets outgroups that included the mono-

treme (platypus), resulted in a rooting near turtles I thank Olivier Rieppel for discussion and providing the original

data of and which has (position I) consistent with the “anapsid” hypo- set deBraga Rieppel (1997) been modified and extended here. 1 am especially grateful to Axel Meyer for thesis and diapsid monophyly (Fig. 11 A). How- ofthe 12S and data which have been providing a copy 16S sets if the ever, monotreme sequence was excluded, the modified here, and for helpful discussion, Nick Fraser, Chris tree tended to be rooted within lepidosaurs, resulting Lambkin, Alan de Queiroz, Andrew Hugall, David Yeates, Jeff in a and a turtle-archosaur Patrice Bouchet and referees all made paraphyletic Lepidosauria Skevington, anomymous

clade 11 This is similar the helpful comments on various versions of the This (Fig. B,D). very to tree manuscript.

work is Australian Research Council in and is supported by an QEI1 proposed Hedges Poling (1999). It per- Research Fellowship and Large Research grant. haps significant that Hedges and Poling’s study

used only two eutherians (rodent and primate) to

root their and did tree not considerany monotreme References or marsupial sequences. This might predispose it

to root within lepidosaurs, especially since lepido- Archie JW. 1989. A randomization test for phylogenetic in- saurs appear to be at the end of long branches for formation in systematic data. Syst. Zool. 38: 219-252. molecular (A. com.) many sequences Meyer, pers. Baker RH, Yu X, DeSalle R. 1998. Assessing the relative

and long branch attraction can affect rooting posi- contribution of molecular and morphological characters

in tion (Milinkovitch and Lyons-Weiler, 1998; Graham simultaneous analysis trees. Mol. Phyl. Evol. 9: 427- 436. et al., 1998). If the results for the mitochondrial Ballard JW, Thayer MK, Newton AF Jr., Grismer ER. 12S I6S and rRNA also hold for the nuclear se- 1998. Data sets, partitions and characters: philosophies wider quences, inclusion of a sample of mammal and procedures for analysing multiple data sets. Syst. Biol.

sequences (especially those from monotremes) 47: 367-396.

break the Barrett Sober E. 1991, consensus. might up long branch between the ingroup M, DonoghueMJ, Against Syst. Biol. 40: 486-493. and outgroup, and thus shift the root from within Benton MJ. 1984. The relationships and early evolution of lepidosaurs to near turtles, or at least weaken the the Diapsida. Symp. Zool. Soc. Land. 52: 575-596, evidence for placing the root within lepidosaurs. Benton MJ. 1985. Classification and phylogeny of the diapsid This wouldmake the results heterodox from nuclear reptiles. Zool. J. Linn. Soc. 84: 97-164.

much genes more consistent with previous mor- Benton MJ. 1996. Vertebrate Palaeontology: Biology and

Evolution. Second Edition. London: Unwin Hyman. phological (and some previous molecular) results. Benton MJ, Clarke JM. 1988. Archosaur and A phylogeny recent example of this phenomenon concerns 1 the relationships of the Crocodylia. In: Benton MJ, ed. where whales, two (apparently) very different phylo- The Phylogeny and Classification of the Tetrapods. Vol. genies the and molecular implied by morphological I. Oxford: Clarendon Press, 295-338. data are actually consistent with the same unrooted Bininda-Emonds OR, Bryant HN, Russell AP. 1998.

terminals cladistic network, and the morphological rooting is robust Supraspecific taxa as in analysis: implicit of assumptions monophyly and a comparison of methods. while the molecular rooting is sensitive to sampling Biol. J. Linn. Soc. 64: 101-133. within the artiodactyl outgroup, and ambiguous Boonstra LD. 1932. Pareiasaurian studies. Part VIII. The and alignments (Messenger McGuire, 1998; Milin- and of osteology myology the locomotor apparatus. Part kovitch and Lyons-Weiler, 1998). Such problems B. Fore limb. Ann. S. Afr. Mus. 28: 436-503.

be Bremer K. 1988. The Limits of Amino Acid Data might prevalent when only a few members are Sequence in Angiosperm Reconstruction. Evolution used diverse Phylogenetic 42: to represent a very outgroup (mammals, 795-803. artiodactyls). This is not to imply that rooting prob- Broom R. 1932. The Mammal-like Reptiles of South Africa lems affect only molecular analyses, and do not and the Origin of Reptiles. London: F. and G. Witherby. affect ones. will affect morphological Rather, they Bull JJ, Huelsenbeck JP, Cunningham CW, Swofford DL,

Downloaded from Brill.com10/08/2021 07:17:21AM via free access Contributions to Zoology, 70 (I) - 2001 17

Waddel PJ. 1993. Partitioning and combining data in tematic cate.g.ories of the Diapsida. Ph.D dissertation,

phylogenetic analysis. Syst. Biol. 42: 384-397. University of California, Berkeley.

Carpenter JM. 1996. Uninformative bootstrapping. Cladistics Gauthier J, Kluge AG, Rowe T. 1988. Amniote phylogeny

12: 177-181. and the importance of fossils. Cladistics 4: 105 - 209. Carroll RL. 1988. Vertebrate, and Evolution. Paleontology Gow CE. 1972. The osteology and relationships of the Mil- New York: W, H. Freeman. lerettidae (Reptilia: Cotylosauria). J. Zool. 167: 219-264, Caspers GJ, Reindeers GJ, Leunisscn JA, Watte J, de Gow CE. 1975. The morphology and relationships of Youngina Jong WW. 1996. Protein indicate that turtles sequences capensis Broom and Prolacerta broomi Farrington. Pa-

branched off from the amniote tree after mammals. J. laeont. Afr. 18: 89-131.

Mol. Evol. 42: 580-586. Gower DJ, Weber E. 1998. The braincase of Euparkeria,

Cunningham CW. 1997. Can three tests incongruence pre- and the evolutionary relationships ofbirds and crocodilians. dict when data should be combined? Mol. Biol. Evol. 14: Biol. Rev. 73: 367-412. 733-740. Graham SW, Kohn JR, Morton BR, Eckenwaldcr JE, DeUraga M, Reisz RR. 1996. The Permian early reptile Barrett SCH. 1999. and dis- Phylogenetic congruence Acleisthorhinus pteroticus and its phylogenetic position. cordance and three molecular among one morphological J. Vert. Pa leant. 16: 384-395. data sets from Pontederiaceae. Syst. Biol. 47: 545-567. DeBraga M, RicppcI O. 1997. Reptile phylogeny and the 1946. Pareiasaurs versus Gregory WK. placodonts as near affinities of turtles. Zool. J. Linn. Soc. 120: 281-354. ancestors to the turtles. Bull. Amer. Mus. Nat. Hist. 86: Dilkcs DW. 1998. The early Triassic rhynchosaur Mesosuchus 276-323. hrowni and the interrelationships ofbasal archosauromorph Hedges SB, Poling LL. 1999. A molecular phylogeny of reptiles. Phil. Trans. R. Soc. Land. 13 353: 501-541. reptiles. Science 283: 998-1001. Donoghue MJ, Doyle JA, Gauthier J, Kluge AG, Rowe Ivachnenko ME. 1987. [Permian parareptiles of the USSR]. f. 1989.The importance offossils in phylogeny reconstruc- Trudy Pal. Inst. Acad. Nauk SSSR 223: 3-160. In Russian. tion. Ann. Rev. Ecol. Syst. 20: 431-460. Juul L. 1994. The phylogeny of basal archosaurs. Palaeont. Eernisse DJ, AG. 1993. Kluge Taxonomic congruence versus Afr. 31: 1-39. total evidence, and amniote phylogeny inferred from fossils, Kemp TS. 1982. Mammal-like reptiles and the origin of molecules and morphology. Mol. Biol. Evol. 10: 1170- mammals. London: Academic Press. 1195.

Kemp TS. 1988. Haemothermia or Archosauria?: The inter- Evans SE. 1984. The classification ofthe Lcpidosauria. Zool. of birds and Zool. J. J. relationships mammals, crocodiles. Linn. Soc. 82: 87-100.

Linn. Soc. 92; 67-104. Evans SE. 1987. The braincase of Youngina capensis (Rep- Kumazawa Nishida M. 1999. mitochondrial tilia; Y, Complete Diapsida; Permian). N. Jahrb. Geol. Paldont. Monals. DNA of the turtle and mole 1984: 193-203. sequences green blue-tailed

skink: statistical evidence for archosaurian ofturtles. Evans SE. 1988. affinity The early history and relationships of the Mol. Biol. Evol. 16: 784-792. Diapsida. in: Benton MJ, ed. The Phylogeny and Classifica- Larson AR. 1994. The of tion comparison morphological and of the Telrapods. Oxford: Clarendon Press, 221-260.

l ailh molecular data in In: Schierwater DP, Cranston PS. 1991. Could this short phylogenetic systematics. a cladogram have arisen B, Steit B, Wagner GP, deSatle R, eds. Molecular ecology by chance alone? On permutation tests for cladistic and evolution: approaches and applications. Basel, Switzer- structure. Cladistics 7: 1-28.

Earris .IS, land; Birkhauser Verlag, 371-390. Kallcrsjii M, Kluge AG, Bult C. 1994. Testing

significance Laurin M. 1991. The osteology of a Lower Permian eosuchian of incongruence. Cladistics 10: 315-319,

Fclsenstein J. 1985. from Texas and a review of diapsid phylogeny. Zool. J. Confidence limits on phylogenies: An

Linn. Soc. 101: 59-95. approach utilizing the bootstrap. Evolution 39: 783-791.

ES. attney 1980. Phylogenetic relationships of the major Laurin M, Reisz RR. 1995. A reevaluation of early amniote

ol gioups amniotes. In: Panchen AL, ed. The Terrestrial phylogeny. Zool. J. Linn. Soc. 113: 165-223. Environment and the Origin of Land Vertebrates. New Lee MSY. 1995. Historical burden in systematics and inter- York: Academic Press, 593-610. relationships of “parareptiles”. Biol. Rev. 70: 459-547.

Gaffney ES. 1990. The comparative osteology ofthe Triassic Lee MSY. 1997. Reptile relationships turn turtle. Nature turtle Bull. 389: 245-246. Proganochelys. Amer. Mu's. Nat. Hist. 194: 1- 263. Lee MSY. 1998. Convergent evolution and character corre- Gaftney ES, Meylan PA. 1988. A of turtles. In: lation in towards resolution of phylogeny burrowing reptiles: a squa- Benton MJ, ed. The Phytogeny and Classification ofTetra- mate phylogeny. Biol. J. Linn. Soc. 65; 369-453. P°ds. Oxford: Clarendon Lee MSY. 2000. Uninformative characters and Press, 157-219. apparent Gardiner BG. 1982. Tetrapod classification. Zool. J. Linn. conflict between morphology and molecules. Mol. Biol. Soc. 74: 207-232. Evol. in press. Gardiner BC.. 1993. Haematothermia: warm-blooded am- Lovtrup S. 1985. On the classification ofthe taxon Tetrapoda. niotes. Cladistics 9: 369-395. Syst. Zool. 34: 463-470. Gauthier, J. 1984. A cladistic of the analysis higher sys- Marshall CR. 1992. Substitution bias, weighted parsimony,

Downloaded from Brill.com10/08/2021 07:17:21AM via free access 18 M. .S'. Y. Lee - Phylogenetics of diapsid reptiles

and amniotc phylogeny as inferred from 18S rRNA se- Wiens JJ, Reeder TW. 1997. Phylogeny of the spiny lizards

Mol. Evol. based molecular and quences. Biol. 9: 370-373. ( Sceloporus) on morphological evi-

Messenger SL, McGuire JA. 1998. Morphology, molecules dence. Herp. Monog. 11: 1-101.

and the phylogenetics of cetaceans. Sysl. Biol. 47: 90- Wilkinson M, Thorley J, Benton MJ. 1997. Uncertain turtle

124. relationships. Nature 387: 466.

Milinkovitch MC, Lyons-VVcilcr J. 1998. Finding optimal Williston SW. 1917. The phylogeny and classification of

ingroup topologies and convexities when the choice of reptiles. J. Geol. 25: 411-421.

outgroups is not obvious. Mot. Phyl. Evol. 9: 348-357. Witmcr LM. 1997. The evolution of the antorbital cavity of

Motani R, Minoura N, Ando T. 1998. Ichthyosaurian re- archosaurs: a study in soft-tissue reconstruction in the

illuminated skeletons from fossil record with of the function of lationships by new primitive an analysis pneuma-

Japan. Nature 393: 255-257. ticity. Mem. Soc. Vert. Paleont. 3: 1-73.

Nixon KC, Carpenter JM. 1996. On simultaneous analysis. Yeates D. 1995. Why remove autapomorphies? Cladistics

Cladistics 12: 221-241. 8: 387-389.

Olmstcad Swccre JA. 1999. data in A. 1998. mitochondrial RG, Combining phylo- Zardoya R, Meyer Complete genome

genetic systcmatics: an empirical approach using three indicates diapsid affinities of turtles. Proc. Natl. Acad.

molecular data sets in the Solanaceae. Syst. Biol. 43: 467- Sci. 95: 14226-14231.

481, Zrzav'y J, Mihuika S, Kcpka P, Bczdck A, Tietz D. 1998.

Patterson C. 1981. Significance of fossils in determining Phylogeny of the Metazoa based on morphological and

evolutionary relationships. Ann. Rev. Ecol. Syst. 12: 195- I8S ribosomal DNA evidence. Cladistics 14: 249-285.

223.

Reisz RR. 1981. A diapsid reptile from the Pennsylvanian First draft received: 14 November 1999

of Kansas. Spec. Publ. Mus. Nat. Hist., Univ. Kansas 1:

1-74.

Rieppel O, dcBraga M. 1996. Turtles as diapsid reptiles. Appendix Nature 384: 453-455.

Rieppcl O, Reisz RR. 1999. The origin and early evolution

of turtles. Ann. Rev. Ecol. Syst. 20: 1022. Osteological Characters (1-176)

Romer AS. 1966. Vertebrate Paleontology. Third Edition.

Characters 1-168 described in and Chicago: University of Chicago Press. are as deBraga Rieppel (1997)

modifications discussed Lee and Motani Romer AS, Price LI. 1940. Review of the Pelycosauria. with as by (1997) et

al. These have been Geol. Soc. Amer. Spec. Pap. 28: 1-538. (1998) incorporated. changes mostly accepted

and thus notcontentious and Reisz, 1999, Rougier GW, de la Fuente MS, Arcucci AB. 1995. Late (Rieppel p.3). However,

characters Triassic turtles from South America, Science 268: 855- changes regardingeight (characters 65, 82,103,120,

858. 121,127,140& 152) are questionedby Rieppel and Reisz (1999)

and discussed below. Some additional made in Sercno PC. 1991. Basal archosaurs: phylogenetic relation- are changes are

the and also discussed below. Reasons for the ships and functional implications. Soc. Vert. Pal. Memoir present analysis

2: 1-53. codings adopted in this analysis discussed, and the references

cited illustrate the contentious characters and/or describe Shaffer IIB, Meylan P, McKnight ML. 1997. Tests ofturtle them

Other characters are not discussed in order to phylogeny; molecular, morphological, and paleontological fully. save space.

Characters 169-176 are new and thus all listed. approaches. Syst. Biol. 46: 235-268. and Reisz Slowinski JB, Crother Bl. 1998. Is the PTP test useful? Rieppel (1999 p.3, paragraph 2) listed, without

additional also made Cladistics 14: 297-302. \ discussion, changes they to the matrix of

and these not here until Sorenson MR. 1996. Tree Rot. Computer Program for Macin- deBraga Rieppel (1997); are adopted

tosh, Distributed by the author, University of Michigan, their supporting reasons are presented.

Ann Arbor.

19. Well-known lanthanosuchids have state Swofford 1)L. 1999. PA UP*. Phylogenetic Analysis Using 0 (Ivachnenko,

Parsimony (*and Other Methods). Version 4. Sunderland 1987)

MA: Sinauer Associates. 32. Both Acleistorhinus and lanthanosuchids have state 0

1987; and Templeton AR. 1983. Phylogenetic inference from restric- (Ivachnenko, Lee, 1995; deBraga Reisz, 1996).

41. is coded unknown tion site endonuclease site with Acleistorhinus as following deBraga cleavage maps particular

and Reisz where the is noted be reference to the humans and Evolution 37: 221-244. (1996), area to poorly apes.

Van and the relevant de Peer Y, Neels JM, dc Rijk P, De Wachter R. preserved suture not discussed orillustrated in the reconstructions. 1993. Reconstructing evolution from eukaryotic small-

ribosomal-subunit Lower fenestra absent (0); The com- RNA sequences: calibration of the mo- 51. temporal present (1).

lecular clock. J. Mol. Evol. 37: 221-232. plex multistate character 51 of deBraga and Rieppel

included three binaries - the Wiens JJ. 1998. The of methods for and of the lower accuracy coding presence temporal

for fenestra (this character), the ventral ofthe fenestra sampling higher-level taxa phylogenetic analysis; a opening

simulation study. Syst. Biol. 47; 397-413. (character 169 below), and the contributionofthe quadrato-

Downloaded from Brill.com10/08/2021 07:17:21AM via free access Contributions to Zoology, 70 (1) - 2001 19

to the fenestra with the absence 128. The condition in lanthanosuchids is have jugal (correlated presence or known; they state

of the quadratojugal). Contrary to de Braga and Rieppel 0 (Lee, 1995).

(1997), nycteroleterids lack a lower fenestra (Ivachnenko, 140. Pareiasaurs lack a discrete fourth trochanter, i.e. a discrete

1987). flange on the ventral surface of the femur. Rieppel and

63. millerettids Younginids (Gow, 1975; Evans, 1987), (Gow, Reisz (1999) interpretthem as having a ‘weak 4th trochanter

and all and retain which has shifted to the of the 1972) pelycosaurs (Romer Price, 1940) ... edge femur’, and thus

the ventral otic fissure. recognise a weaker structure in a different position as

65. Pareiasaurs have basal tubera which formed both This are by potentially homologous. again is a subjective disagree-

the basisphenoid and basioccipital, in contrast to the basal ment that is difficult to resolve. Regardless ofthe potential

tubera ofturtles and other taxa which are formed entirely homology, pareiasaurs lack a well-developed fourth

by the basioccipital. All basal tubera are here considered trochanter, hence the coding adopted here.

sufficiently similar to be potentially homologous in this 152. The first distal tarsal is present in most turtles, including

study; however, Rieppel and Reisz (1999) suggest that all adequately known basal forms (e.g. Gaffney, 1990);

are different fail the of and Reisz code either absence they enough to similarity test Rieppel (1999) presence or

as for turtles without homology. potentially primitive citing reasons.

72. This is here redefined basicranial articulation 169. Lower fenestra closed character as temporal ventrally (0); open ventrally

unfused (0) or fused (1), rather than the inferredfunctional (1). A subdivision ofDR51; not applicable in taxa without

of states palate“kinetic” or “akinetic”. Thepalate in Acleis- a fenestra.

torhinus and lanthanosuchids is not fused; rather, it is a 170. Suborbital foramen. Absent (0), present (1). LR49

tight but non-sutural joint. Under the redefined character 171. Quadrate flange ofpterygoid. Long, almostreaches quadrate

they now have state 0. condyle (0); short, does not approach quadrate condyle

74. Millerettids, captorhinids, nycteroleterids and procolo- (1). From Lee (1995).

phonids do not have a suborbital fenestra (only a small 172. Exoccipitals. Without lateral flanges(0); with lateral flanges

foramen, which has been coded as a separate character, (1). From Lee (1995).

170). See Laurin and Reisz (1995). 173. Anterior end of spl'enial. Tapered (0); forked (1). From

82. The mandibularjoint is here considered to be primitively Lee (1995).

in front of the occiput in turtles, since this is the condition 174. Foramen jugular anterius. Small (0); large (I). From Lee

found in all basal turtles such as Proganochelys, Kayen- (1995).

tachelys and Auslralochelys and most pleurodires and 175. Epipophysis of atlas vertebra. Present (0); absent (1). From

cryptodires (e.g. Gaffney, 1990; Rougier et ah, 1995). In Lee (1995).

contrast, Rieppel and Reisz (1999) code this character as 176. Paired anterior and posterior concavities in neural arches.

polymorphic in turtles. Absent (0); present (1), From Lee (1995).

97, The condition in lanthanosuchids is known; they have state 0 (Lee 1995).

103. The narrow ventral margin on the cervicals ofpareiasaurs Soft anatomical characters (177-216) is considered to the potentially homologous ‘keels’ on

diapsids; Rieppel and Reisz (1999) consider them ‘ridged codes refer to the but not Alphanumeric following descriptions numbering keeled’, thusarguing that they are not similar enough

to of Gauthier et al. warrant system (1988). conjecture ofhomology. This againis a subjective

disagreement that is difficultto resolve. However, it should

177. External meatus. Small G23. be noted that basal turtles exhibit auditory (0); large (1). a very similar morphology 178. membrane and enclosed metotic to that of Secondary tympanic pareiasaurs, which Rieppel and Reisz accept as fissure. Absent (0); present (1). A21. potentially homologous to that of diapsids. 20, 179. Extracolumella. Without Huxley’s foramen (0); with 1 Pareiasaurs are here coded with two coracoid ossifications (e.g. Boonstra, Huxley’s foramen (1). A25 modified. 1932, Gregory 1946); Rieppel and Reisz 180. Nasal conchae. Absent G24a. (1999) code them with (0); present (1). one but did not cite the source. (21. Pareiasaurs 181. External nasal gland. Within nasal capsule (0); outside are here coded with the coracoid foramen within the nasal capsule (1). A6. coracoid, at least when the lateral surface of the 182. Cochlea. Short (0); elongate (1). G27. scapulocoracoid is considered (e.g. Boonstra, 1932; Gregory, 183. Cerebellum. Small or moderately developed (0); enlarged 1946); Rieppel and Reisz (1999) code them with G7 modified. the foramen between (1). the coracoid and scapula without 184. Olfactory bulbs. Without with citing a source. peduncles (0); peduncles 7. An (1). AI4. cctepicondylar foramen is found in most turtles, 185. Dorsoventricular of including all basal ridge telencephalon. Absent (0); forms (e.g. Gaffney, 1990), and thus pre- sent LIS. coded as (1). primitively present. It is absent mainly in highly ' 186. Neurofilament derived forms. proteins. Two types (0); three types (1). aquatic Rieppel and Riesz (1999) code either L9. state as potentially primitive for turtles without stating reasons. 187. Pineal gland. Sensory (0); secretory (1). G13 modified.

Downloaded from Brill.com10/08/2021 07:17:21AM via free access 20 M.S. Y. Lee - Phylogenetics of diapsid reptiles

Sinus 188. cavernosos.Absent (0); present (I). LI3. 205. Lumen ofstomach. Entirely anterior to pyloris (0); partially

189. Cells ofiris and ciliary muscles. Smooth (0); striated (1). posterior to pyloris (1). A7.

A17. 206. Blood in low in plasma. High urea (0); urea (1). A9.

190. Ciliary processes. Poorly developed (0); well developed 207. Phi (= beta) keratins in epidermis. Absent (0); present {1).

(1). G32. G5c.

191. muscle of Absent G34. 208. erectile Absent G30. Pyramidalis eye. (0); present (1). Single, penis. (0); present (1).

192. Absent small Nictitating membrane. or (0); large (I). G34b 209. True epiphyses. Absent (0); present (1). G36a modified.

modified. 210. canals in Cartilage epiphysis. Absent (0); present (1). L10.

193. Tendon of nictitans. Absent (0); present (I). G34c. 211. Caruncle. Absent (0); present (1). LI9.

194. Tendon ofnictitans. Attaches interorbital 212. albumen. Small to eye (0); septum Egg amount (0); large amount (1). LI8.

(1). G34a. 213. Eggshell. Calcareous (0); parchment (1). L15.

195. Colour vision. Absent All. 214. With without (0); present (1). Eggshell. large pores (0); large pores (1),

196. Masticatory muscle plate. Undivided (0); divided (1), A19. LI5a.

197. Postmandibular branchial arches. Three or more (0); 215. Paired membranes. Present absent two tertiary egg (0); (1),

or fewer (1). A29. LI 5b.

198. Two or more tracheal rings. Absent (0); present (1). G35a. 216. Nest building utilising plant material. Absent (0); present

199. Heart. Incompletely divided (0); completely divided (I). (1). A12.

G20.

200. Septum sinu-venosi. Absent (0); present (1). 021.

201. Semilunar valves ofpulmonary artery. Two (0); three (1). Mitochondrial DNA (characters 217-3119).

G3.

202. Subclavian arteries. Near third fourth aortic Consensus ofthe sites in or arches, or sequences 2903 the complete aligned

anterior G37, 12S and I6S ribosomal RNA and more posterior (0); near carotids, or more (1). sequences (Zardoya Meyer,

203. and adrenal Kidney gland. Adjacent (0); separated (1). 1998) were constructed for Mammalia, Archosauria, and

A20. Testudines as discussed in the text.

204. Adrenals. Adjacent to body wall (0); suspendedby gonadal

mesenteries (1). L7 modified.

Data Matrix

Appendix Table I. The morphological data matrix. Characters Appendix Table I. Continued.

1-176 are from osteology, and 177-216 are from soft anatomy. - N no data = (i.e. missing information), - inapplicable (i.e,

taxon loo derived to be objectively scored). Within-taxon

variability is represented as follows: A=0&1, B=0&2, C=l&2,

D=0&l&2, E=0&2&3, F=2&3, G=0&3,

Tr Pr Pr o o lacer SQUAMATA Araeoscel TESTUDINES CYNODONTIA Caseidae OUTGROUP Prolacerti SQUAMATA Araeoscel tor TESTUDINES CYNODONTIA Caseidae OUTGROUP t Placodontia Gorgonopsia Sphenacodon Ophiacodont Placodontia Gorgonopsia Ophiacodont Rhynchosauria i Choristodera Choristodera Sphenacodon f Younginiformes Captorhinidae Pareiasauridae Millerettidae Edaphosauridae Rhynchosauria Younginiformes Captorhinidae Millerettidae RHYNCHOCEPHALIA Kuehneosauridae Pareiasauridae t Edaphosauridae ilophosauridae ARCHOSAURIFORMES Eosauropterygia Claudiosauridae ididaeProtorothyrididae Procolophonoidea Nycteroleteridae Lanthanosuchidae Acleistorhinidae Trilophosauridae ARCHOSAURIFORMES RHYNCHOCEPHALIA Kuehneosauridae Eosauropterygia Claudiosauridae Procolophonoidea Nycteroleteridae Lanthanosuchidae Acleistorhinidae idae ididae idae ormes tidae formes othyrididae idae

1 1 1 1 1 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N N N N N N 1 0 0 1 1 1 0 0 0 0 1 1 1 N N N 1 0 N 0 N 0 0 0 0 2 N 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0

1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 A A 0 0 0 1 0 0 0 0 0 1

1 1 1 1 A A 0 N A 0 0 0 0 0 0 A 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 A A 0 0 0 0 N 0 0 0 0 0 0 0 1 1 0 0 1 1 0 0 0 A

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 N 0 0 0 0 0 1 1 0 0 1 A 1 A 1 0

0 1 1 1 1 0 0 1 0 0 0 0 0 0 0 1 1 1 1 0 0 A 0 0 1 1 1 1 1 0 0 0 0 1 1 1 0 1 0 0 0 0 0 0 A A A 1 0 0 A 0 1 0 0 1 0 0

0 1 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 A

N 1 0 0 N 0 0 0 A 1 0 1 0 0 0 0 1 1 0 0 0 0 0 2 1 0 0 0 0 0 A 1 1 0 1 1 1 A 0 0 0 0 0 0 A 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 1 D 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0

0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 '0 0 0 0 0 0 1 1 1 1 1 0 0 N 0 1 1 1 0 1 0 0 0 1 N N N N N N N N N N N N N N N 0

N 1 1 A 0 A 1 1 1 1 0 N 0 0 0 1 1 1 1 1 1 0 1 1 1 0 0 1 0 1 1 1 N 1 1 N 1 1 1 1 1 1 1 1 0 0 2 2 N 0 0 0 1 1 1 0 0 0 0 3

0 0 0 0 0 0 0 0 0 0 0 0 0 0 A A 0* 0 0 0 1 0 0 0 0 0 00 0 0 1 1 1 1 0 1 0 1 1 1 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 N 0 0 0 0 0 0 0 1 1 1 1 1 1 0 0 0 0 0 0 0 N 1 1 0 1 1 1 1 1 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 1 A 1 1 A

N 0 A A 1 0 0 1 0 1 0 1 1 0 1 0 1 1 0 0 0 0 1 1 N 1 1 0 0 1 0 0 0 0 0 1 1 0 0 0 A 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0

A A N A 0 A 1 1 A 0 0 1 0 0 0 A 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 0 0

0 0 0 N 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 A 1 0 0 0 N 1 1 2 0 2 0 2 0 1 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0

1 1 1 1 2 2 2 N 1 2 1 1 0 0 0 1 N N 0 1 1 1 1 0 2 1 1 0 0 0 0 N 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 A 0 0 0

N N A N 1 1 1 N 1 1 0 N N 0 A 0 0 N N N N N 1 N 0 0 0 0 0 N 0 0 0 N 0 1 N 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0

1 0 A 1 2 D 0 0 A 2 1 1 1 0 0 B 0 0 0 0 1 1 1 1 1 1 1 0 0 0 1 1 A 1 0 0 1 A 1 1 0 0 0 0 1 0 1 1 0 0 0 2 2 0 0 0 0 0

Downloaded from Brill.com10/08/2021 07:17:21AM via free access Contributions to Zoology, 70 (I) - 2001 21

Appendix Table I. Continued. Appendix Table I. Continued,

Tr Pr Mi Ophi Tr Young 11 Young Procol olacer SQUAMATA in Araeoscel sir© CYNODONT CdS6ldd6 OUTGROUP Prolacert SQUAMATA in Araeoscel Caseidae OUTGROUP ARCHOSAURI i TESTUDINES Pareiasaur i TESTODINES CYNODOrTTIA Edaphosaur t Placodontia f 11 Gorgonopsia acodon Kuehneosaur Placodontia f Protorothyr Gorgonopsia Sphenacodon Qphiacodont i IA ilophosaurRhynchosauria Choristodera Captorhinidae i ilophosaur Rhynchosauria Choristodera Captorhinidae Millerettidae Edaphosauridae t Pareiasauridae ophono RHYNCHOCEPHALIA Kuehneosauridae Claudiosauridae t Eosauropterygia das RHYNCHOCEPHALIA Eosauropterygia Claudiosauridae ormes idae Procolophonoidea Nycteroleteridae ormes Lanthanosuchidae Acleistorhinidae Sphenacodontidae ARCHOSAURIFORMES Nycteroleteridae Lanthanosuchidae Acleistorhinidae idae FORMES formes ididae Protorothyrididae idae idae ididae idae idae idae iformes ididae idea idae

1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 A 0 0 0 0 0 0 0 1 1 1 0 1 2 2 N A 1 A 1 0 0 0 1 1 1 1 1 N 0 1 1 0 0 0 0 0

0 0 0 0 0 1 1 A 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 N N 1 1 N 1 1 N 0 0 N 1 0 0 0 0 0 0 N N N 1 1 0 0 0 0 0

N 1 1 1 N 1 1 1 N 1 1 1 1 1 0 0 0 0 0 N 1 0 A 0 0 0 0 2 0 0 0 0 0 2 c 1 1 A N 0 1 1 0 0 0 0 0 N N 0 N 0 1 1 1 1 1 0

0 2 2 F F 3 2 3 F 3 0 2 0 0 0 1 0 1 0 .1 2 2 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 1 0 0 N N 0 2 1 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 A 0 1 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 N N N N N 0 0 0 N N 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N N 0 0 N 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 N 1 1 1 N N 0 1 1 1 1 1 1 0 0 2 1

N 1 1 1 1 1 1 0 A A 0 0 0 0 0 1 1 0 0 0 0 0 N N N N N N 1 N 0 0 0 1 0 1 N N 0 0 0 0 0 0 0 0 0 0 0 0 0 N 0 0 0 0 0 0

1 1 1 1 1 1 1 N 1 N 0 0 0 1 1 1 N N 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 N N A A A N N N 0 0 0 0 0 0 0

2 2 A 0 1 A A A 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 1 0 0 1 1 0 0 0 0 0 1 A A 0 0 0 1 0 0 0 0 0 0 0 0 N 0 0

1 1 1 1 1 1 1 -0 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 1 1 1 1 1 0 1 1 1 0 0 0 0 0 N N 1 0 0 0 0 0 0 0

3 3 F 3 G E B 2 C 2 1 1 A 1 1 1 A 0 1 1 1 1 0 0 0 1 N 0 0 N N N 0 0 0 0 B 3 2 2 2 B 2 0 2 1 1 1 1 2 0 0 0 0 0 0 0 0

1 1 1 1 1 C 1 1 1 1 1 1 1 1 0 0 0 0 0 0 2 1 1 2 1 1 1 1 1 1 1 2 1 2 2 2 C 1 1 2 1 0 0 0 0 0 0 0 0 0 0 0 0 0 5 N N 0 0 0

i 0 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 B B 2 0 0 B 1 2 2 N 0 0 0 0 0 0 1 1 A 1 1 1 1 1 1 0 1 A 0 0 0 A 0 N N 2 2 0 0 0 0 0

2 2 2 2 2 2 2 2 2 2 0 0 0 1 2 1 B 0 1 1 1 1 1 1 1 0 1 1 0 0 0 0 0 1 0 A 0 0 N 0 0 0 N 1 0 1 1 1 1 1 0 1 A 0 0 0 0 0 2 2 C C 2 c C 2 2 C 2 2 C 2 1 1 1 1 1 B 0 0 0 0 0 0 2 2 1 1 1 0 0 0 0 0 0 0 2 A 0 0 0 0 0 0 0 0 0 0 N N 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 A N 1 1 0 0 0 1 1 0 N N N 1 0 0 0 0 0 0 0 3 1 2 2 2 2 2 2 2 2 2 2 B 2 2 2 2 0 1 1 1 1 1 1 0 N N 0 0 2 2 0 0 0 1 1 1 1 0 0 A 1 1 N 0 0 0 0 1 0 0 0 0 1 1 0 N 0 0 0 1 1 0 0 N A 1 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 N N 1 1 0 0 0 A 0 0 0 N N N 1 0 0 N 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 A 0 1 1 1 1 1 0 0 0 0 0 1 0 0 0 0 1 1 N N 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 0 0 0 0 0 N 0 0 N N 1 0 0 0 0 1 1 1 1 N N 1 0 0 0 0 0 0 N' 0 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 A 0 1 N 1 1 1 1 1 1 0 0 N N 0 1 0 0 0 0 0 0 1 1 A 1 1 0 A 0 A 0 0 0 0 0 0 0 0 0 0 2 2 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 1 0 a 0 0 0 0 0 N N 0 0 0 0 0 0 0 0

i 1 1 0 0 0 0 N 0 0 0 0 0 0 0 1 1 0 0 0 0 1 1 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 1 1 0 N N 0 1 1 N 0 0 0 0 0

: N 0 1 1 1 A 1 1 1 1 1 N 1 1 0 0 0 0 0 1 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 A 1 0 1 1 0 0 , A N N 0 0 0 1 1 1 1 1 1 1 N 0 N 0 1 1 1 1 1 1 1 1 1 1 1 1 N N N N 1 0 0 0 N 0 1 1 0 0 0 0 0 1 0 0 0 0 0 0 0 N N 0 1 0 0 0 0 0 0

i 0 0 1 A 0 0 N A 0 1 0 1 0 0 N N N . 0 1 1 0 0 0 0 0 1 1 0 0 0 0 0 1 1 0 N 0 A A N A 0 N 0 0 0 1 1 0 0 0 1 1 0 0 0 0 0 0 4 1 1 0 N 1 0 1 1 1 N 1 0 0 0 0 0 1 C 1 0 0 0 0 2 2 2 2 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 1 A 0 0 N N 0 2 2 0 0 0 0 0 1 1 1 0 1 1 1 N 1 c A 0 0 0 0 0 0 1 1 A N 0 0 0 1 1 1 1 0 0 0 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 N N 0 1 1 0 0 0 0 0

i o 1 N 1 1 1 N 1 1 N N N N 0 1 0 0 0 1 0 0 0 1 1 0 1 0 0 0 1 0 0 N N N N 1 N 1 0 N 0 1 A 0 0 0 N N N 1 0 0 0 0 0 0

, : 1 N 1 N N 1 A N 1 1 N N N 0 1 1 0 0 1 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 1 1 1 1 N N 1 1 0 i N N N N N N N 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 A 0 N A 1 1 0 1 0 0 1 0 0 0 0 N N N 0 0 0 0 0 0 . 1 1 1 1 1 1 1 N 1 1 0 N N N 0 A 0 0 N N N N 1 1 0 0 0 0 0 0 7 0 1 0 0 N 1 1 N 1 N 0 0 0 0 0 1 0 0 0 N N 0 1 1 0 0 0 0 0

N N 1 N N A 0 N N N N N I . N N N N 0 A 0 0 N N N N 2 2 0 0 0 0 0 0 1 1 0 1 1 0 0 0 0 0 0 0 1 0 0 0 N N N 1 1 0 0 0 0 0 3 0 A 0 N 0 0 0 1 1 0 0 0 1 1 1 1 0 0 A 1 0 0 0 0 0 1 1 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 N N 1 1 1 1 1 1 1 0

i o o 0 0 0 0 0 0 2 2 0 0 0 0 0 0 D 0 0 0 0 0 0 0 2 2 2 2 0 1 1 0 0 0 0 0 0 0 0 1 1 1 0 2 2 N : A 0 1 1 0 N N 0 0 0 0 0 : 2 1 l- 1 1 1 1 1 0 0 1 1 1 N 0 2 2 0 0 0 0 0 0 0 0 N 1 1 N 0 0 0 0 0 0 0 1 N A 0 N 0 1 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 1

; . L 1 1 0 1 1 1 NN N 1 1 0 0 1 1 1 A 1 0 0 N 0 0 0 0 0 0 0 0 0 0 0 N 0 1 1 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 1 1 0 N 0 0 0

i o 0 1 0 1 A 1 0 0 A A 0 1 N 1 1 0 0 0 0 0 0 0 0 0 N 0 0 0 N N A 2 0 0 0 A N 0 0 0 0 1 0 1 1 1 N 1 0 0 2 2 1 1 0 0 0

: - i a o 0 0 1 0 1 1 1 1 1 1 1 1 1 0 0 0 0 0 1 1 1 1 1 1 1 0 N N 0 0 1 1 0 0 1 N N N N 0 0 0 : . 1 1 0 0 0 0 0 0 0 0 0 0 0 : 2 2 c C 1 2 N 2 0 2 2 1 1 1 1 1 1 1 0 0 1 0 0 0 0 1 0 0 0 N N 0 0 0 0 0 0 0 1 1 1 1 1 2 1 2 1 1 1 1 2 2 1 1 1 1 0 N

: . l 1 1 1 0 0 1 N 1 0 1 1 1 1 1 A 1 1 1 N 0 0 0 0 0 0 0 1 0 0 0 N N 0 1 0 0 N 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 1 0 0 0

: 1 1 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 N 0 0 0 N 0 0 0 0 0 0 0 8 0 0 1 1 0 1 N 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 A N A N N N N 0 0 2 0 1 1 1 1 1 2 2 1 2 N 1 1 1 1 1 2 2 2 1 N N 1 1 1 0 0 0 0 0 0 1

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 N N 0 1 0 0 N 0 0 0 1 1 1 1 1 0 0 1 1 0 A 1 1 1 1 N 1 1 1 1 1 N N N 0 0 0 N 0 0 0 0 6 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 2 1 0 N N 0 2 2 0 N 0 0 0 0 0 101010N0 10D11AA1 00011010 10AA10N0 1011111000011NN0 00001N10 00001000 00001N00 02001000 13201111 10201111 00211111 00211N10 00A11NN0 002N1N10 00201000 00221010 00021000 00000000 01000000 00000000 00000000 1 1 1 1 1 1 1 1 0 0 1 0 1 1 0 0 0 1 0 1 0 0 0 0 0 A1N10 A N N 300111111 00DA1A10. 0 0 0 1 1 1 1 1 1 1 N 0 0 1 1 1 1 0 0 0 0 0 N N 1 1 0 0 0 0 0 0 N

i 1 1 N 1 0 N 1 1 0 N N 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 N 0 0 0 0 N 0 0 0 10 0 0 1 0 2 0 0 0 0 0 0 o N 1 1 10 0 0 1 0 0 0 0 0 0 2 0 0 0 0 0 2 c A 0 0 0 0 0 0 110 0 0 0 0 0 0 0 N 0 0 2 0 0 0 0 0 0 0 0 0 0 0 o 0 0 0 10 0 0 0 0 N N N N N N 0 N N N N N 0 D 0 N N 0 0 0 0 0 0 0 0 0 1 0 N N N N N N N N N N N 0 9 n 1 A N N N 0 0 0 10 10 10 10 0 0 0 0 0 0 0 0 N N 0 0 0 0 0 1 A 0 0 0 0 0 0 0 0 0 0 0 0 1 1 N 1 0 0 0 i 1 0N001N0NN10N110 1 1 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 1 0 0 1 N 0 1 0 0 0 0 N N N N 0 0 0 0 1 1 1 0 1 0 0 i 1 1 1 0 0 1 0 1 A 1 1 1 0 1 0 N N N N 0 0 0 0 NN01112: 1110 1 1 1 1 1 1 A N N N A 0 0 0 0 N 1111 1110 1110 1 1 1 1 11110 1 o 0 1 110 110 110 110 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 N 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 7 1

; 1 0 1 A 1 1 0 1 0 0 0 0 0 0 0 0 1 1 1 1 1 N 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 N 0 0 0 1 1 0 1 0 0 A N 0 0 0 N 0 1 1 1 0 0 0 0 0

0 0 0 1 0 1 0 0 0 0 0 0 0 0 1 1 1 1 1 0 N N N 0 0 0 0 0 0 0

0 0*0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 0 0 0 0 0 0 N 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N 1 1 1 1 N 0 0 0 0 0 0 0 0 0 0 1

N N 1 N N 0 0 N N N 0 N N N N N N N 0 N N N N N N 1 N N N 0

N N N 1 N N 1 0 N N N N N N N N 0 N N N N N N A N N N N N 0

0 1 A 1 N N N N N N N N 0 N 0 N N N N N N N N N N N N N N 0

0 N 1 N N 1 1 N N N N N N N N 0 N N N N N N 1 N 0 N N N N N 0 8 1

0 N N 1 N N 1 1 N N N N N N N N 1 N N N N N N 0 N N N N N 0

0 0 0 N 1 N N 1 N N N N N N N N N N N N N N N 1 N N N N N 0

0 N N 1 N N 0 0 N N N N N N N N 0 N N N N N N 1 N N N N N 0

N N N 0 N N A N N 1 1 N N N N N 0 N N N N N N 0 N N N N N 0

N N N 310111210001001100100011011 0 N N 1 N N 1 1 N N N N N 1 N N N N N N 0 N N N N N 0

1101A1NNN11AB011A0N10011110 N N N N N ■00N1C0110110O0N1011 10001D0N1A11A1011011 10001B0N1A11A0011011IDOOlBAllOOllOOllAlA10001101101110011011 A0N01A1N11111010101A1000100N11N1110N10NN 1000000N1111111N1N11 1000000110010001N0A1 000000001001000000001000N001101100001011 00000000100100000000 N 0 1 1 N 1 N N 0 0 N N N N N N N N 0 N N N 1 N N N N 0 OOOOOOOOOOOOOOOOOOAO A00A0000A00200110010101A000N1002001111A0OOOOOOOOOOO2OOINONOO 0 N 1 N N 0 0 N N N N N N N N 1 N N 1 A01N1001A0011011 ONNNNNNNNNNNNNNNNNN100N00DNNNN200NNNNN0 010A1010110110111011 01001000100110111000 01000000100100100000 0100000000010000000010000000100100100000 00000000000100000000 00000000000000000000 N N N N N N N N N N 0

N N 1 N N 0 0 N N N N N N N N 1 N N 1 N N N N N N N N N N 0

N 0 N 1 N N 1 1 N N N N N N N N 1 N N N N N N 0 N N N N N 0

N N 1 N N 0 0 N N N N N N N 0 N N N 1 N N N N 1 N N 0 N N N 0 9 1

Downloaded from Brill.com10/08/2021 07:17:21AM via free access 22 M.S. Y. Lee - Phylogenetics of diapsid reptiles

Appendix Table I. Continued.

Tr i Chor Prolacertif is SQUAMATA RHYNCHOC Placodontia TESTUDINES CYNODONTIA Gorgonopsia Sphenacodon Caseidae OUTGROUP 1ophosaurRhynchosauria todera Younginiformes Captorhinidae Millerettidae ARCHOSAURIFORMES EPHAL Kuehneosauridae Eosauropterygia Claudiosauridae AraeoscelididaeProtorothyrididae Pareiasauridae Procolophonoidea Nycteroleteridae Lanthanosuchidae Acleistorhinidae Edaphosauridae Ophiacodontidae idae ormes IA tidae

N 1 N N N 0 0 N N N N N N N N 1 N N N N N N 0 N N N N N 0

1 1 1 N N N N 1 N N N N N N N N N N N N N N N 0 N N N N 0

N N N 1 N 1 1 N N N N N N N N 0 N N N N N N 1 N N N N N 0

N N 0 1 1 N N N N N N N N N N 0 N N N N N N - N N N N N _

N N A N 1 1 1 N N N N 0 N N N N N N N N N N N N N N N N 0

1 N N N N 1 1 N N N N N N N N 1 N N N N 0 N N N N N N N 0

1 N N N N 1 1 N N N N N N N N 1 N N N N N N 0 N N N N N 0

N N 1 N 1 1 N N N N N N N N N 1 N N N N N N 0 N N N N N 0

N N 1 0 0 N N N N 0 N N N N N N N N N N N N 1 N N N N N 0

N N 1 N N 1 1 N N N N N N N N 0 N N N N N N 1 N N N N N 0 0 0 2

1 0 0 N N N N N N N N N N N N 0 N N N N N N 1 N N N N N 0

N N 1 N N 0 0 N N N N N N N N N N N N 0 N N N N 1 N N N 0

N N 1 N N 1 1 N N 0 0 N N N N N N N N N N N N N N N N N 0

N 0 N 1 1 N N N N N N N N N N 0 N N N N N N 0 N N N N N 0

N N 1 0 0 N N N N N N N N N N 0 N N N N N N 1 N N N N N 0

N N 1 N N 1 1 N 0 N N N N N N N N N N N N N 0 N N N N N 0

N N 1 N N 1 1 N N N N N N N 1 N N N N N N N 0 N N N N N 0

1 N N N N 0 0 N N N N N N N N 1 N N N N N N 1 N N N N N 0

N N 0 1 1 N N N N N N N N N N 0 N N N N N N 1 N N N N N 0

N N 1 N 0 0 N N N N N N N N N 1 N N N N N N 1 N N N N N 0 0 1 2

N N 1 N N 0 0 N N N N N N N N 1 N N N N N N 1 N N N N N 0

1 N N N N 0 1 N N N N N N N N 1 N N N 1 N N N N N N N N 0

N N 0 N 1 1 N N N N N N N N N 0 N N N N N N 0 N N N N N _

N N 0 N N 1 1 N N N N N N N N 0 N N N N N N 0 N N N N N -

N N 0 N N 1 1 N N N N 0 N N N 0 N N N N N N N N N N N N -

1 N 0 N N N N 0 N N N N N N N 0 N N N N N N 1 N N N N N 0 6 1 2

Downloaded from Brill.com10/08/2021 07:17:21AM via free access