Post on 06-Feb-2018
The Trimorphodon biscutatus (Squamata: Colubridae)
Species Complex Revisited: A Multivariate Statistical
Analysis of Geographic Variation
Thomas J. Devitt1,3, Travis J. LaDuc2, and Jimmy A. McGuire3
The Western Lyresnake (Trimorphodon biscutatus) inhabits arid regions from the desert southwestern United
States southward along the Pacific lowland versant to northwestern Costa Rica and exhibits substantial
geographic variation in size, squamation, and color pattern across its range. We examined patterns of
geographic variation within T. biscutatus using multivariate statistical analyses of 33 morphological characters
scored from 429 specimens. Principal components and discriminant analysis revealed six morphologically
distinct groups that are generally concordant with lineages recovered in a phylogeographic analysis of
mitochondrial DNA and with taxa traditionally recognized as species or subspecies. We conclude that
Trimorphodon biscutatus (sensu lato) comprises six evolutionary species (including the recently elevated T.
vilkinsonii) and recommend elevating T. biscutatus (sensu stricto), T. lambda, T. lyrophanes, T. paucimaculatus, and T.
quadruplex to the species level. A key to the species of Trimorphodon is provided.
Trimorphodon biscutatus habita regiones aridas desde los desiertos del suroeste de Estados Unidos hacia el sur, a lo
largo de las tierras bajas del pacıfico de Mexico hasta el noroeste de Costa Rica. Esta especie presenta
considerable variacion geografica en tamano, escamacion, y patrones de coloracion a lo largo de su rango, y por
lo tanto ha tenido una inestable historia taxonomica. Investigamos poblaciones de T. biscutatus usando analisis
estadısticos multivariados de datos morfologicos para delimitar especies dentro de este taxon politıpico y de
amplia distribucion. Analisis de componentes principales y analisis discriminante de 33 caracteres de 429
especımenes resultaron en seis grupos morfologicamente distintos, consistentes con taxa reconocidos
tradicionalmente como subspecies. Concluimos que Trimorphodon biscutatus (sensu lato) comprende seis linajes
evolutivos (incluyendo a T. vilkinsonii, recientemente designada como especie) y recomendamos elevar a T.
biscutatus (sensu stricto), T. lambda, T. lyrophanes, T. paucimaculatus, y T. quadruplex al rango de especie. Una clave
al nivel de especie de Trimorphodon es incluida.
MEMBERS of the genus Trimorphodon (commonlyknown as ‘‘lyresnakes’’ owing to the lyre-shapedor chevron-shaped marking present on the top of
the head of most forms) are long, slender, rear-fanged,nocturnal snakes inhabiting arid regions throughout thesouthwestern United States, Mexico, and Central America.At least two species comprise the genus: the widelydistributed, polytypic Western Lyresnake, T. biscutatus(Fig. 1A–G), ranging from the southwestern United Statessouthward along the Pacific lowland versant of Mexicoto northwestern Costa Rica (Fig. 2), and the MexicanLyresnake, Trimorphodon tau (Fig. 1H), found over much ofmainland Mexico north of the Isthmus of Tehuantepec(Scott and McDiarmid, 1984a, 1984b). Trimorphodon biscu-tatus and T. tau have long been recognized as distinctspecies, differing in several diagnostic features and occurringin sympatry at lower elevations in Michoacan, Morelos,Sinaloa and Sonora (McDiarmid and Scott, 1970; Scott andMcDiarmid, 1984a, 1984b).
Both species have been further subdivided into multiplesubspecies, but the taxonomy of T. biscutatus in particularhas been unstable due to varying interpretations of
geographic variation in size, squamation, and color pattern.In total, six species have been described from specimens thatare now considered to be T. biscutatus (Cope, 1870, 1886,1887; Klauber, 1924; Taylor, 1938; Smith, 1941, 1942, 1943).A number of authors subsequently suggested that recog-nized forms were subspecies of one widespread, polytypictaxon (Bogert and Oliver, 1945; Duellman, 1957; Jones andFindley, 1963; Loomis and Stephens, 1967; Hardy andMcDiarmid, 1969). Gehlbach (1971) provided the mostrecent review of the entire Trimorphodon biscutatus complex,uniting all species under one name with six subspecies (T. b.biscutatus, T. b. lambda, T. b. lyrophanes, T. b. quadruplex, T. b.vandenburghi, and T. b. vilkinsonii) citing clinal variationacross both latitude and longitude in meristic characters.However, several of these characters show clear demarcationbetween samples from different geographic regions, failingto exhibit the rank order in character values that would beexpected for a species that varies clinally. FollowingGehlbach (1971), Scott and McDiarmid (1984a) defined T.biscutatus as one species subdivided into six subspecies,despite earlier observations (McDiarmid and Scott, 1970)that two distinct species (T. biscutatus and T. lambda) were
1 Museum of Natural Science and Department of Biological Sciences, 119 Foster Hall, Louisiana State University, Baton Rouge, Louisiana70803; E-mail: (TJD) tdevitt@berkeley.edu. Send reprint requests to TJD.
2 Section of Integrative Biology and Texas Natural Science Center, The University of Texas at Austin, Austin, Texas 78712-1064.3 Present address: Museum of Vertebrate Zoology and Department of Integrative Biology, 3101 Valley Life Sciences Building, University of
California, Berkeley, California 94720-3160.Submitted: 16 February 2007. Accepted: 3 October 2007. Associate Editor: D. Kizirian.F 2008 by the American Society of Ichthyologists and Herpetologists DOI: 10.1643/CH-07-045
Copeia 2008, No. 2, 370–387
Fig. 1. Geographic variation within members of the Trimorphodon biscutatus complex (A–G) and Trimorphodon tau (H). (A) T. b. lyrophanes: SanDiego Co., California (TJD 088); (B) T. b. lambda: Cochise Co., Arizona (LSUMZ 83684); (C) T. vilkinsonii: Presidio Co., Texas (TNHC 66511); (D) T.paucimaculatus: Jalisco, Mexico (UTA R 52919); (E) T. b. biscutatus: Jalisco, Mexico (UTA R 52932); (F) T. b. quadruplex: Guanacaste Province, CostaRica (specimen not collected); (G) presumed hybrid between T. b. lambda and T. vilkinsonii illustrating reduced head pattern characteristic of T.vilkinsonii: Cochise Co., Arizona (LSUMZ 83685); (H) T. tau: Sinaloa, Mexico (MZFC 19476). All photos by T. J. Devitt.
Devitt et al.—Variation in Trimorphodon biscutatus 371
likely sympatric in southwestern Jalisco. LaDuc and Johnson(2003) elevated T. b. vilkinsonii back to species level based onmorphological differences, resurrecting the name T. vilk-insonii to represent populations inhabiting the ChihuahuanDesert.
Devitt (2006) investigated the phylogeography of theTrimorphodon biscutatus complex using mitochondrial DNA(mtDNA) sequence data and Bayesian relaxed molecularclock divergence time estimation, recovering five well-supported clades whose boundaries are concordant withexisting geographic or environmental barriers. This mtDNAphylogeny provides a working hypothesis about the numberof presumed species within this polytypic taxon. Here, weexamine morphological variation within T. biscutatus toprovide a nuclear background against which the mtDNAgene genealogy may be evaluated, and test whethersufficient evidence exists for recognizing separate evolu-tionary lineages (sensu Wiley, 1978).
MATERIALS AND METHODS
Sampling.—We scored a suite of 38 meristic, mensural, andcolor-pattern characters (Appendix 1) from 219 specimensassigned to Trimorphodon biscutatus, with an emphasis onpopulations in California, Mexico, and Central America.These data were combined with LaDuc and Johnson’s (2003)dataset consisting of the same characters obtained from 210specimens primarily from Arizona, Texas, New Mexico, andSonora, for a total of 429 specimens. A subsample ofspecimens from LaDuc and Johnson’s data were re-exam-ined in order to verify inter-measurer consistency (Lee,1990). Sex was determined by dissection. We analyzed malesand females separately to remove the effects of sexualdimorphism. Museum acronyms follow Leviton et al. (1985)except MZFC (Museo de Zoologıa, Facultad de Ciencias) andTJD (personal field series of T. J. Devitt).
Statistical analyses.—Each sample was classified a priori intoone of six groups traditionally recognized as species orsubspecies (T. vilkinsonii, T. b. lambda, T. b. lyrophanes, T.paucimaculatus, T. b. biscutatus, and T. b. quadruplex; Fig. 2)
based on the distribution of Trimorphodon vilkinsonii sensuLaDuc and Johnson (2003) and the five major mitochon-drial lineages recovered by Devitt (2006). Standard univar-iate descriptive statistics (mean 6 1 standard deviation,range, sample size) were calculated for each group. Equalityof variances among groups was examined using an F test.We performed pairwise comparison of means (ANOVA)using Scheffe’s F procedure for post hoc comparisons(Scheffe, 1953) because it is robust to violations ofassumptions inherent in multiple comparison procedures(e.g., the assumption of homogeneity of variances). Wetested continuous size measurements (other than snout–vent length [SVL]) for differences among groups usinganalysis of covariance (ANCOVA) where SVL was thecovariate. We regressed morphometric characters againstsnout–vent length as an indicator of overall size in order toseparate size and shape variation, resulting in residuals usedin subsequent multivariate analyses (Reist, 1986). Allunivariate statistical tests were implemented in StatView5.0 (SAS Institute, 1998).
In addition to univariate methods, we also used multivar-iate statistical techniques for both description and infer-ence. Principal components analysis (PCA) was implement-ed to explore and describe patterns of variation in the data,by organizing specimens along a meaningful gradientemphasizing variation among, rather than similarity within,samples (McGarigal et al., 2000). Scatterplots for each pair ofprincipal components were constructed to check for outliers(McGarigal et al., 2000). We used discriminant functionanalysis (DFA) to investigate differences among prespecifiedgroups and to determine which variables contribute themost variation to group dispersion (McGarigal et al., 2000).Statistical significance of the canonical functions wasassessed through jackknifing and cross-validation.
For both PCA and DFA to be strictly applicable, the datamust meet a number of fundamental assumptions, includ-ing random sampling, independence among samples, andmultivariate normality (McGarigal et al., 2000). However, asthere is no objective way to fully evaluate the assumption ofmultivariate normality, we transformed only those charac-ters that were significantly different (P , 0.05) from theideal normal distribution, identified using a Kolmogorov–Smirnov (K–S) test. We performed analyses with both log-transformed and raw data to assess robustness of the data toviolations of the assumption of multivariate normality,although even moderate violations of the assumptionsunderlying PCA and DFA do not drastically alter resultsprovided the sample size is sufficient (Harris, 1975; McGar-igal et al., 2000). In multivariate analyses, we included onlythose characters that were significantly different (P , 0.05)in at least one pairwise comparison. Specimens missing datawere excluded from multivariate analyses, which wereconducted with SYSTATH v.8.0 Statistics (SPSS Inc., 1998).
Species delimitation.—In evaluating whether multiple speciesexist within the Trimorphodon biscutatus complex, we followthe Evolutionary Species Concept (ESC) proposed bySimpson (1961:153) and later modified by Wiley (1978).Under the ESC, ‘‘A species is a single lineage of ancestraldescendant populations of organisms which maintains itsidentity from other such lineages and which has its ownevolutionary tendencies and historical fate.’’ (Wiley,1978:18). We consider morphological differences corrobo-rated by genetic divergence between populations in combi-
Fig. 2. Distribution of the Western Lyresnake (Trimorphodon biscuta-tus) complex after Gehlbach (1971) and Scott and McDiarmid (1984a).
372 Copeia 2008, No. 2
nation with geographic distribution as sufficient evidencefor recognizing separate evolutionary species (Wiley, 1978).We maintain that some populations remain distinct despitelimited interbreeding with other populations, and that theeffect of this admixture does not diminish the independenceof these evolutionary lineages (de Queiroz, 1998).
RESULTS
Re-examination of a subsample of LaDuc and Johnson’s(2003) dataset indicated that nearly identical measurementswere taken from the same specimens by both observers (TJDand TJL). Six geographically contiguous sampling entitiesassorted in ordination space for both males (Fig. 3A) andfemales (Fig. 3B) in the PCA. Univariate descriptive statisticsand results from ANOVA are provided in Appendix 2. Posthoc tests revealed that most characters were significantlydifferent (P , 0.05) in at least one pairwise groupcomparison in both sexes. Characters that were notsignificantly different among populations varied slightlybetween sexes and were excluded from subsequent multi-variate analyses (Appendix 2). Regressions of all headmeasurements against SVL were not significantly differentamong populations in either sex. Additional non-significantcharacters included the number of preocular scales amongboth males and females, as well as the numbers of postocularand loreal scales among females only. Equality of variances Ftests revealed that most characters were significantlyheteroscedastic in at least one pairwise group comparisonin both sexes. Snout–vent length in males was the onlycontinuous character that departed significantly from anideal normal distribution. Log-transformation of this char-acter did not change results from principal components anddiscriminant analyses, and thus raw values were retained infinal analyses. All continuous characters measured from
females were normally distributed and thus were nottransformed.
We considered factor loadings with an absolute valuegreater than 0.50 highly significant in interpreting principalcomponent structure (McGarigal et al., 2000; Tabachnickand Fidell, 2001). Characters that loaded significantly on thefirst principal component for both sexes included SVL, scalecounts (ventrals, subcaudals, dorsal scale rows, and infra-labials), and differences in color pattern (blotch length,division, and number; Table 1). In both sexes, the firstcomponent provided a division between populations in theNearctic deserts of northern Mexico and the southwesternU.S. (T. b. lyrophanes, T. b. lambda, T. vilkinsonii) andpopulations found in tropical deciduous forest from south-ern Sonora southward (T. paucimaculatus, T. b. biscutatus,and T. b. quadruplex). Characters with large loadings on thesecond principal component for both sexes includedadditional color pattern characters (blotch length, shape,and distance between blotches), which provided furtherseparation among northern (T. b. lyrophanes, T. b. lambda,and T. vilkinsonii) and southern (T. paucimaculatus, T.biscutatus, and T. b. quadruplex) populations (Fig. 3). Canon-ical analysis of discriminance revealed highly significantdifferences among the sampling entities specified by thePCA in both males (Wilks’ l 5 0.0004, P , 0.0001; Fig. 4A,C) and females (Wilks’ l 5 0.0001, P , 0.0001; Fig. 4B, D).Canonical variate scores indicated that ventral and sub-caudal scale counts, differences in blotch size, condition ofthe anal plate, and completeness of the lyre-shaped markingbest discriminated among these prespecified groups (Ap-pendix 1). The classification function correctly classified97% of the specimens in both sexes (Table 2).
Post hoc tests identified characters that were significantlydifferent (P , 0.05) among populations (Appendix 1).Northern populations (T. b. lyrophanes, T. b. lambda, and
Fig. 3. Principal components scatterplots for males (A) and females (B).
Devitt et al.—Variation in Trimorphodon biscutatus 373
T. vilkinsonii) did not differ in size, but were significantlysmaller than southern populations (T. paucimaculatus, T. b.biscutatus, and T. b. quadruplex). Populations of T. paucima-culatus were significantly larger than adjacent T. b. lambdapopulations to the north, but significantly smaller than T. b.biscutatus populations to the south. Populations of T. b.biscutatus did not differ significantly in size from disjunct T.b. quadruplex populations from Central America. Similar tothe pattern of size variation, mean numbers of ventral,subcaudal, and dorsal scales were significantly higher insouthern populations (Appendix 1). Geographically proxi-mate samples of males were significantly different in countsof ventral scales except for the following pairwise compar-isons: T. b. lambda–T. vilkinsonii and T. b. biscutatus–T. b.quadruplex. Among females, non-significant pairwise com-parisons of ventral scales included T. b. lambda–T. vilkinsoniiand T. b. lyrophanes–T. b. lambda. For subcaudal scale countsamong males, T. b. lambda was not significantly differentfrom adjacent T. vilkinsonii populations. Among females,mean number of subcaudals did not differ between T. b.
lambda and T. vilkinsonii or between T. b. biscutatus and T. b.quadruplex. Numbers of dorsal scale rows differed primarilybetween, but not within, northern and southern popula-tions. For example, number of dorsal scale rows at midbodywere significantly different among geographically proxi-mate samples for both sexes except for the followingpairwise comparisons: T. b. lyrophanes–T. b. lambda, T. b.lambda–T. vilkinsonii, and T. b. biscutatus–T. b. quadruplex.Number of dorsal scale rows counted at two additionalpoints on the body generally followed this same trend.Means of several head scale counts were significantlydifferent among a small subset of pairwise comparisons,although only the number of infralabial scales contributedto group dispersion in multivariate analyses (Table 1).Variation in this character approximated the trend innorth–south differences observed in other meristic charac-ters, with southern populations characterized by highercounts (Appendix 2). The only group in which a significantnumber of individuals (68%) possessed an entire anal platewas T. b. lyrophanes (Appendix 2).
Table 1. Factor Loadings for the First and Second Principal Components and Canonical Discriminant Functions for the First, Second, and Third
Canonical Variate Scores, Standardized by Within-Group Variances. Values greater than 0.50 in absolute value are considered significant and are shown in
bold for emphasis. N. S. indicates a given character was not significantly different (P , 0.05) in any pairwise comparison of populations using Scheffe’s F
procedure for post hoc comparisons, and thus was excluded from multivariate analyses. See Appendix 1 for character abbreviations.
PC 1 PC 2 CV 1 CV 2 CV 3
males females males females males females males females males females
SVL 0.631 0.532 20.060 0.132 0.095 0.084 0.153 0.138 0.004 0.156VNT 0.848 0.811 0.193 0.224 0.655 0.564 0.153 0.059 0.157 20.069SBCDL 0.779 0.699 0.345 0.392 0.657 0.614 0.108 20.136 0.071 20.017ANAL 0.320 0.361 0.308 0.360 0.006 20.241 0.177 20.031 20.545 20.632SRW1 0.860 0.816 0.127 0.069 0.305 0.364 20.242 0.124 0.284 0.310SRW2 0.838 0.832 0.155 0.188 0.242 0.283 0.155 20.026 0.041 0.126SCRW3 0.824 0.727 0.037 0.068 0.128 0.336 0.014 0.255 0.106 0.125SWBTH1 0.511 0.610 20.528 20.574 20.046 20.065 20.388 20.714 20.466 20.257SWBTH2 0.831 0.797 20.335 20.370 0.203 0.351 20.207 20.166 0.065 20.015SWBTH3 0.856 0.848 20.116 20.232 0.181 0.064 20.048 20.129 0.129 0.137LHTSP1 0.234 0.314 20.533 20.509 0.016 0.043 20.014 0.142 0.054 20.069LHTSP2 0.492 0.475 20.449 20.335 20.049 0.026 20.300 20.269 20.062 0.126LHTSP3 0.416 0.398 20.396 20.397 20.099 20.204 20.143 20.385 0.039 0.020SUPRAL 0.198 N. S. 0.257 N. S. 0.026 N. S. 0.045 N. S. 20.017 N. S.SUPRAR 0.181 N. S. 0.230 N. S. 0.054 N. S. 0.159 N. S. 20.085 N. S.INFRAL 0.475 0.564 0.169 0.106 0.007 0.023 0.084 20.046 20.071 20.250INFRAR 0.535 0.462 0.060 0.135 20.076 20.098 20.046 0.025 20.152 0.044POSTL 20.325 N. S. 0.107 N. S. 20.012 N. S. 20.041 N. S. 0.020 N. S.POSTR 20.266 N. S. 20.053 N. S. 20.026 N. S. 0.011 N. S. 0.048 N. S.LORL 0.229 N. S. 20.127 N. S. 20.159 N. S. 20.110 N. S. 0.015 N. S.LORR 0.269 N. S. 20.156 N. S. 0.142 N. S. 20.035 N. S. 20.033 N. S.LYRE 0.405 0.363 20.331 20.457 0.025 20.146 20.219 20.587 0.004 0.069BBTH 20.527 20.553 20.483 20.489 20.377 20.241 20.203 20.384 0.417 0.483TBTH 20.072 20.262 20.401 20.437 0.134 0.300 20.068 20.078 0.191 0.342PBDIV 0.123 0.087 20.468 20.479 20.112 20.153 0.006 0.220 20.024 20.203SBDIV 0.581 0.495 20.333 20.278 0.165 0.097 20.295 20.193 20.437 20.492HSBTH 0.217 0.255 0.694 0.726 0.108 0.061 0.078 0.120 20.107 20.019BSQ12 0.222 0.114 0.825 0.878 0.080 0.051 0.048 0.132 0.046 20.195BSQ78 0.098 0.137 0.851 0.850 0.142 0.071 0.227 0.252 20.019 0.307BSQ1314 0.072 20.033 0.841 0.776 20.043 0.062 0.265 0.290 20.136 20.258BSQAN 20.121 20.088 0.788 0.746 20.169 20.031 0.113 20.261 0.156 0.123Eigenvalue 7.909 7.05 5.598 5.58 20.673 19.406 5.432 8.238 3.132 4.575% total variance 25.5 28.2 18.1 22.3 — — — — — —% total dispersion — — — — 65.1 54.7 82.3 77.9 92.1 90.8
374 Copeia 2008, No. 2
Table 2. Classification Matrix from the Discriminant Analysis of Males/Females Using Randomized Cross-Validation. Off-diagonal numbers
represent misclassifications.
Taxon biscutatus lambda lyrophanes quadruplex vilkinsonii paucimaculatus % Correct
biscutatus 39/11 0/0 0/0 0/0 0/0 0/0 100/100lambda 0/0 117/46 0/0 0/0 1/0 2/0 98/100lyrophanes 0/0 6/3 39/15 0/0 0/0 0/0 87/83quadruplex 0/0 0/0 0/0 34/10 0/0 0/0 100/100vilkinsonii 0/0 0/0 0/0 0/0 42/18 0/0 100/100paucimaculatus 0/0 0/0 0/0 0/0 0/0 16/6 100/100Total 39/11 123/49 39/15 34/10 43/18 18/6 97/97
Fig. 4. Canonical analysis of discriminance scatterplots for males (A, C) and females (B, D).
Devitt et al.—Variation in Trimorphodon biscutatus 375
Several color pattern characters were also identified bypost hoc tests as significantly different (P , 0.05) amongpopulations (Appendix 1). Within northern populations, T.vilkinsonii has significantly fewer, narrower, and morewidely separated blotches than adjacent T. b. lambda tothe west (Klauber, 1940; LaDuc and Johnson, 2003;Appendix 2). In turn, T. b. lambda possesses fewer, broader,more widely separated blotches compared to neighboring T.b. lyrophanes, which has relatively many, narrowly separatedblotches (Klauber, 1928, 1940). Populations of T. paucima-culatus have significantly fewer, more narrowly separated,elongate blotches compared to adjacent T. b. lambdapopulations to the north (Taylor, 1938). The nominate T.b. biscutatus is further differentiated from the narrowlysympatric T. paucimaculatus by the presence of blotches withmarginal extensions, giving the markings the appearance ofthe letter ‘H’, and usually with smaller spots between theprimary blotches (Smith, 1941). Specimens of T. b. quad-ruplex are distinguished by blotches that are divided by thecentral light spots into secondary blotches, which arefurther subdivided by an additional light space (Smith,1941).
The head markings of Trimorphodon are highly variable(Klauber, 1940; Jones and Findley, 1963; LaDuc andJohnson, 2003), but may be useful in distinguishingbetween these taxa when used in combination with othercharacters. Consistent with the overall reduction in patternexhibited by T. vilkinsonii, this species usually lacks acomplete lyre-shaped marking. Instead, the head pattern ischaracterized by dark markings consisting only of spots or avery foreshortened chevron (Taylor, 1939; Jones andFindley, 1963; LaDuc and Johnson, 2003). Most T. b. lambdapopulations possess a complete, chevron-shaped lyre, exceptfor some specimens from extreme southeastern Arizona(Cochise County) and southwestern New Mexico (Grant andHidalgo counties) that may exhibit a reduced lyre-shapedmarking. Additionally, T. vilkinsonii usually lack the darkinterocular bar across the frontal and prefrontal scalespresent in other taxa (Gehlbach, 1971). Populations of T.b. lyrophanes are characterized by a dark marking that mostclosely resembles a lyre (Klauber, 1940), becoming incom-plete near the apex in 72% of specimens examined(Appendix 2). Occasional specimens of T. paucimaculatuspossess head markings that resemble the very foreshortenedchevron exhibited by T. vilkinsonii, while in others themarking becomes nearly incomplete near the apex, ap-proaching the condition seen in T. b. lyrophanes. Theremainder of populations in southern Mexico (south ofthe Trans-Mexican Volcanic Belt [TMVB]) and CentralAmerica usually possess complete V-shaped head markings(Taylor, 1939).
DISCUSSION
The six morphologically distinct groups identified here aregenerally concordant with the major clades recovered byDevitt (2006) in his phylogeographic analysis of mtDNA. Anoverlay of current taxonomy (Gehlbach, 1971; Scott andMcDiarmid, 1984a) on these groups, however, revealsconflict between taxa traditionally recognized as subspeciesand the geographic boundaries of the groups identified here.Inaccurate taxonomy, interspecific hybridization, and/orincomplete lineage sorting may be responsible for thisdiscordance (reviewed in Funk and Omland, 2003). Previousworkers hypothesized that gene flow between these lineages
was the result of primary intergradation and dispersalaround pre-existing geographic barriers (Gehlbach,1971:200). However, the concordance between speciesborders and geographical barriers is more consistent with apattern of vicariant allopatric divergence and speciation,followed by secondary contact (Devitt, 2006). Below, weevaluate the geographical boundaries of the morphologicalgroups identified here against current taxonomic designa-tions.
Trimorphodon b. biscutatus and T. paucimaculatus:sympatric ‘‘subspecies’’.—McDiarmid and Scott (1970:36)first noted that two distinct species of the T. biscutatusgroup (T. biscutatus and T. lambda) occur in ‘‘sympatry ornear sympatry’’ in the southwestern corner of Jalisco, aresult confirmed by Devitt (2006). Morphological interme-diates have not been reported, suggesting complete repro-ductive isolation. The larger southern form (hereafter, T.biscutatus) extends southward along the coast to westernGuatemala (Devitt, 2006). The smaller, northern form(hereafter, T. paucimaculatus) is found along the coast fromsouthwestern Jalisco through Nayarit to northern Sinaloa(Hardy and McDiarmid, 1969; Devitt, 2006). These lineagesare nearly as divergent in mtDNA from each other as anypopulation of the T. biscutatus complex is from T. tau(approx. 9–10% based on uncorrected pairwise sequencedivergence; Devitt, 2006).
Taylor (1938) originally described populations fromSinaloa as a distinct species, T. paucimaculatus, based onpreocular scales separated from the frontal scales, and‘‘dorsal spots greatly elongated and fewer in number’’(1938:527). Despite these morphological differences, Fuglerand Dixon (1961) relegated this taxon to subspecific status.Subsequently, Gehlbach (1971) subsumed it under T. b.biscutatus based on inferred gene flow with Sonoran DesertT. b. lambda populations in southern Sonora, though henoted ‘‘The Sierra Madre, which converges on coastallowlands in the Navojoa-Alamos region of Sonora, appar-ently limits gene flow along the Pacific versant of Mexico.’’(Gehlbach, 1971:207). Regardless of any putative contem-porary gene flow, morphological evidence presented here, incombination with genetic data (Devitt, 2006), indicatessubstantial historical isolation between populations fromsouthern Sonora and those from northern Sinaloa. Thesedata support Taylor’s conclusion that ‘‘bi-scutatus’’ as usedby previous workers (Gunther, 1895:174; Boulenger,1896:54) represents a ‘‘composite’’ of more than one species(Taylor, 1938:529).
Trimorphodon b. biscutatus and T. b. quadruplex: imperfect‘‘subspecies’’ boundaries.—Gehlbach (1971:207) suggestedthat inundation of the Isthmus of Tehuantepec separatedpopulations in southern Mexico (T. b. biscutatus) from thosein Central America (T. b. quadruplex) long enough for colorpattern differences to evolve between these taxa. Resultspresented here, however, together with mtDNA (Devitt,2006) indicate that the break occurs further south across theGuatemalan highlands. Populations from western Guate-mala (representing T. b. quadruplex sensu Gehlbach, 1971)group with populations further west from Mexico south ofthe TMVB in molecular and morphological analyses, ratherthan with populations from eastern Guatemala and CentralAmerica (Devitt, 2006). An area of unsuitable habitat alongthe coast of Chiapas and Guatemala separates T. b. biscutatus
376 Copeia 2008, No. 2
of southern Mexico and interior valleys of western Guate-mala from its sister group (T. b. quadruplex) in easternGuatemala, Honduras, Nicaragua, and Costa Rica (Devitt,2006). An abrupt shift from xeric, subhumid vegetation tohumid forest occurs in the vicinity of Tonala, Chiapas,Mexico, and this humid forest extends southeastward alongthe coast to near Escuintla in Guatemala (Campbell, 1999).A number of arid-adapted lineages, including Trimorphodon,are absent from this cooler, wetter region along the coast,but are found inland in the isolated, arid interior valleys ofthe Rıos Grijalva and Motagua that drain to the Atlantic(Campbell and Vannini, 1988). These arid valleys exist inthe rain shadow of coastal volcanoes and show a high degreeof similarity to tropical deciduous forests of the Pacificfoothills in western Mexico (Campbell and Vannini, 1988).
Trimorphodon b. lyrophanes and T. b. lambda: dispersalor vicariance? ––While Gehlbach (1971) concluded thathistorical fragmentation of Trimorphodon populations acrossthe Isthmus of Tehuantepec was responsible for themorphological differences he observed between populationsin that region, he suggested that recent dispersal around thepre-existing Gulf of California from mainland Mexico bestexplained the distribution of Trimorphodon on the BajaCalifornia Peninsula (Gehlbach, 1971:200). Devitt (2006),however, found an abrupt phylogeographic break betweenTrimorphodon populations from the Baja California Penin-sula region and those from adjacent mainland Mexico, apattern consistent with a model of vicariant allopatricdivergence. He concluded that the formation of the Gulfof California was probably responsible for dividing theselineages during the late Miocene to early Pliocene (approx.8–4 Ma; Devitt, 2006).
Trimorphodon b. lambda and T. vilkinsonii: introgression orincomplete lineage sorting? ––As with his dispersal scenariofor explaining the presence of Trimorphodon on the BajaCalifornia Peninsula, Gehlbach (1971:200) postulated thatlyresnakes inhabiting the Chihuahuan Desert reached theirpresent range by dispersing eastward around the northernend of the Sierra Madre Occidental from the Pacific lowlandsof Mexico. LaDuc and Johnson (2003) demonstrated clearmorphological differences between T. b. lambda and T.vilkinsonii of the Sonoran and Chihuahuan deserts, respec-tively, elevating T. vilkinsonii back to the species level, butleaving T. b. lambda as a subspecies of T. biscutatus.Interestingly, these morphologically distinct taxa (T. b.lambda and T. vilkinsonii) exhibit relatively little geneticdifferentiation in mtDNA (Devitt, 2006). Devitt (2006)concluded that either introgression or incomplete lineagesorting could explain this pattern equally well, becausedistinguishing between their effects is difficult from genegenealogies alone (Nielsen and Wakeley, 2001). Morpho-logical data (Gehlbach, 1971; LaDuc and Johnson, 2003)indicate that recent gene flow has occurred between theselineages across the desert grassland ecotone between theChihuahuan and Sonoran deserts in southeastern Arizonaand southwestern New Mexico (the Cochise filter barrier;Morafka, 1977). Although a pattern of geographicallylocalized, interspecifically-shared haplotypes is not expectedunder incomplete lineage sorting as it is under localintrogression (Funk and Omland, 2003), incomplete lineagesorting cannot be ruled out. Taking into consideration itsrestricted range, T. vilkinsonii may be a recently derived
peripheral isolate that budded off from ancestral popula-tions in the Sonoran Desert and colonized the northwesternmargin of the Chihuahuan Desert during Quaternaryclimatic oscillations and has had insufficient time to sortto monophyly (Devitt, 2006). In cases of peripatric specia-tion, recognizing the nested and parental lineages asdistinct, or, conversely, recognizing them as the same, bothmask the distinct process of peripheral isolation (de Queirozand Donoghue, 1988; Wiens and Penkrot, 2002; Funk andOmland, 2003). Regardless of the underlying evolutionaryprocess, these lineages are maintaining their own separateidentities despite some recurrent gene flow.
Revised taxonomy.—Clearly, current taxonomy underesti-mates species-level diversity with this group; sympatry of‘‘subspecies’’ is an untenable condition for most biologists,regardless of species delimitation criteria. Taxonomic deci-sions involving geographically differentiated taxa that areexperiencing limited gene flow, however, are more difficultto resolve (Wiens, 2004). Morphologically intermediatespecimens from near putative contact zones suggest thatlimited gene flow may occur between some northernpopulations (e.g., T. lambda–T. vilkinsonii, T. lambda–T.lyrophanes, and T. lambda–T. paucimaculatus). However, theeffect of this hybridization does not diminish the indepen-dence of these evolutionary lineages. Although all speciesconcepts require some subjective decisions, a taxonomicclassification constructed under the ESC (Wiley, 1978)represents a significant improvement over the traditionaltaxonomy because it reflects evolutionary history. Thus, wepropose a taxonomic reclassification of this group recogniz-ing six evolutionary species, including T. biscutatus (sensustricto), T. lambda, T. lyrophanes, T. paucimaculatus, T.quadruplex, and the recently elevated T. vilkinsonii (LaDucand Johnson, 2003).
Trimorphodon biscutatus (Dumeril, Bibron, and Dumeril,1854)Figure 1E
Dipsas bi-scutata Dumeril, Bibron, and Dumeril, 1854:1153.Dispsadomorphus biscutatus Gunther, 1858:176.Trimorphodon biscutatus Cope, 1861a:297.
Eteirodipsas biscutata Jan, 1863:105 (in part).Trimorphodon major Cope, 1870:153.Trimorphodon (Dipsas) biscutata Duges, 1882:145.Dipsas bisculata Duges, 1884:337.Sibon biscutatum Garman, 1884a:16 (in part).Sibon biscutata Garman, 1884b:22 (in part).Dipsas bicutata Velasco, 1890:54.
Trimorphodon biscutatum Duges, 1893:295.Trimorphodon biscutatus biscutatus Smith, 1941:159.Trimorphodon biscutatus semirutus Smith, 1943:492.
Holotype.—MNHN 5900, sex unknown, Mexique, collectiondate and collectors unknown (not examined by authors).
Diagnosis.—This species can be distinguished from all otherspecies of Trimorphodon by the presence of primary darkbody blotches shaped like the letter ‘H’ (especially anterior-ly), and usually with additional smaller dorsal markingsbetween primary blotches.
A species of Trimorphodon defined by the followingcombination of characteristics: primary dark body blotches
Devitt et al.—Variation in Trimorphodon biscutatus 377
fewer than 30 (mean of 19), 4–12 scales wide alongmiddorsal line at midbody (mean of seven), shaped likethe letter ‘H’ (especially anteriorly), central pale spotsusually dividing primary blotches, resulting secondaryblotches undivided; usually with additional smaller dorsalmarkings between primary blotches that may be paired andelongate, rounded, or H-shaped; seventh and eighthprimary blotches separated by seven scale rows on average;ventrals 232–279 in males, 254–286 in females; midbodyscale rows 22–28 (mean of 25); marking on top of head lyre-shaped, with a dark bar between the eyes or across snout;anal divided; size large (maximum observed SVL of males 5
1255 mm, females 5 1590 mm).
Distribution.—Found from the vicinity of the Rıo Purifica-cion, Jalisco south along the Pacific Coast throughoutColima, Michoacan, Guerrero, extreme southern Puebla,Oaxaca, southwestern Chiapas to the vicinity of Tonala, andalso in arid, interior valleys of southwestern Guatemala suchas the upper Rıo Grijalva valley in the state of Huehuete-nango.
Remarks.—Cope (1875a) synonymized Trimorphodon majorwith T. biscutatus. Duellman (1954) synonymized T. b.semirutus with T. b. biscutatus. Trimorphodon biscutatus issympatric with T. paucimaculatus in southwestern Jalisco, aswas first suggested by McDiarmid and Scott (1970).Although previous workers concluded that the breakbetween T. b. biscutatus and T. b. quadruplex occurred atthe Isthmus of Tehuantepec (Gehlbach, 1971), phylogeneticanalyses (Devitt, 2006) together with color pattern differ-ences (this study) indicate genetic and morphologicaldiscontinuities further south across the Guatemalan high-lands. Smith and Taylor (1950) restricted the type locality toTehuantepec, Oaxaca, Mexico.
Suggested common name.—Western Lyresnake.
Trimorphodon lambda Cope, 1886Figure 1B
Trimorphodon lyrophanes Cope, 1875b:38 (in part). Arizonaspecimens.
Trimorphodon lambda Cope, 1886:286.
Trimorphodon lambda lambda Dixon et al., 1962:98.
Trimorphodon biscutatus lambda Gehlbach, 1971:208.
Holotype.—USNM 13487, juvenile, Mexico, Sonora, Guay-mas, 1883, H. F. Emerich (not examined by authors).
Diagnosis.—This species is distinguished from T. tau by theabsence of a broad, pale band across the nape, from T.lyrophanes by having fewer body blotches (mean of 24 in T.lambda, 34 in T. lyrophanes) that are more widely separated(seventh and eighth blotches separated by four scale rows onaverage in T. lambda, two in T. lyrophanes), from T.vilkinsonii by having a greater number of body blotches(mean of 24 in T. lambda, 21 in T. vilkinsonii) that are morenarrowly separated (seventh and eighth blotches separatedby four scale rows on average in T. lambda, eight in T.vilkinsonii), from T. paucimaculatus by having a greaternumber of body blotches (mean of 24 in T. lambda, 21 inT. paucimaculatus) that are narrower (mean of five scaleswide along middorsal line at midbody in T. lambda, eight in
T. paucimaculatus), from T. biscutatus by the absence of H-shaped blotches and additional markings between theprimary blotches, and from T. quadruplex by havingundivided secondary blotches.
A species of Trimorphodon defined by the followingcombination of characteristics: primary dark body blotchesfewer than 31 (mean of 24), 3–9 scales wide along middorsalline at midbody (mean of five), central pale spots dividingprimary blotches, resulting secondary blotches undivided;seventh and eighth blotches separated by four scale rows onaverage; ventrals 211–240 in males, 223–249 in females;midbody scale rows 19–25 (mean of 22); usually with acomplete lyre-shaped marking on top of head and a dark barbetween the eyes or across snout, but head markings may bereduced, especially in specimens from extreme southeasternArizona (Cochise Co.) and southwestern New Mexico (Grantand Hidalgo counties); anal divided; size small (maximumobserved SVL of males 5 885 mm, females 5 890 mm).
Distribution.—Found from southern Nevada (Nye Co.),extreme southwestern Utah (Washington Co.), souththrough much of Arizona except for the northeasternquarter, southwestern New Mexico (Hidalgo and Grantcounties), southeastern California generally east of theSalton Trough region, south through Sonora and westernChihuahua west of the Sierra Madre Occidental, to northernSinaloa.
Remarks.—Trimorphodon lambda hybridizes with T. vilkinsoniiin extreme southeastern Arizona (Cochise County) andsouthwestern New Mexico (Grant and Hidalgo counties).This species may also come into contact with T. lyrophanesin southeastern California and/or extreme southern Nevada,and with T. paucimaculatus near the Sonora–Sinaloa border.Specimens from contact zones may be difficult to identify.
Suggested common name.—Sonoran Lyresnake.
Trimorphodon lyrophanes (Cope, 1861)Figure 1A
Lycodon lyrophanes Cope, 1861b:343.
Trimorphodon lyrophanes Cope, 1861a:297.
Trimorphodon vandenburghi Klauber, 1924:17.
Lycognathus lyrophanes Cochran, 1961:196.
Trimorphodon biscutatus lyrophanes Gehlbach, 1971:208.
Trimorphodon biscutatus vandenburghi Gehlbach, 1971:209.
Syntypes.—USNM 4680, two specimens (male and female),[Mexico], Lower California, Cape St. Lucas, May 1859, JohnXantus, (not examined by authors).
Diagnosis.—This species is distinguished from T. tau by theabsence of a broad, pale band across the nape, from both T.lambda and T. vilkinsonii by having a greater number of bodyblotches (mean of 34 in T. lyrophanes, 24 in T. lambda, 21 inT. vilkinsonii) that are more narrowly separated (seventh andeighth blotches separated by two scale rows on average in T.lyrophanes, four in T. lambda, eight in T. vilkinsonii), from T.paucimaculatus by having a greater number of body blotches(mean of 34 in T. lyrophanes, 21 in T. paucimaculatus) thatare narrower (mean of five scales wide along middorsal lineat midbody in T. lyrophanes, eight in T. paucimaculatus),from T. biscutatus by the absence of H-shaped blotches and
378 Copeia 2008, No. 2
additional markings between the primary blotches, andfrom T. quadruplex by having undivided secondary blotches.
A species of Trimorphodon defined by the followingcombination of characteristics: primary dark body blotchesfewer than 48 (mean of 34), 2–7 scales wide along middorsalline at midbody (mean of five), central pale spots divideprimary blotches, resulting secondary blotches undivided;seventh and eighth blotches separated by two scale rows onaverage; ventrals 214–237 in males, 221–250 in females;midbody scale rows 19–24 (mean of 22); lyre-shapedmarking on top of head often incomplete at the apex(resembling a lyre), with a dark bar between the eyes oracross snout; anal divided or entire; size small (maximumobserved SVL of males 5 774 mm, females 5 936 mm).
Distribution.—Found throughout most of the Baja CaliforniaPeninsula and on Cerralvo, Danzante, San Jose, San Marcos,and Tiburon islands; in southern California generally westof the Salton Trough region north to the vicinity of LosAngeles on the coast, and inland in the Mohave Desertnorth to the Argus and Amargosa mountains in InyoCounty.
Remarks.—The original catalogue entry for the type ofTrimorphodon lyrophanes contained four specimens: two arein the USNM under number 4860. A third was sent to theAcademy of Natural Sciences, Philadelphia, and a specimenwith the same data as the types was catalogued at theAcademy (ANSP 10146) but could not be located as of 1983(Scott and McDiarmid, 1984a).
Grismer et al. (1994) synonymized Trimorphodon biscutatusvandenburghi with T. b. lyrophanes. Trimorphodon lyrophanesmay come into contact with T. lambda in parts ofsoutheastern California and extreme southern Nevada.Specimens from these contact zones may be difficult toidentify.
Suggested common name.—Peninsular Lyresnake.
Trimorphodon paucimaculatus Taylor, 1939Figure 1D
Trimorphodon paucimaculatus Taylor, 1938:527.Trimorphodon lambda paucimaculata Fugler and Dixon,
1961:17.Trimorphodon lambda paucimaculatus Fouquette and Ross-
man, 1963:198.Trimorphodon biscutatus biscutatus Gehlbach, 1971: 207.
Holotype.—UIMNH 25072, sex unknown, Mexico, Sinaloa,Mazatlan, 24 July 1934, Edward H. Taylor (not examined byauthors).
Diagnosis.—This species is distinguished from T. tau by theabsence of a broad, pale band across the nape, from both T.lambda and T. lyrophanes by having fewer body blotches(mean of 21 in T. paucimaculatus, 24 in T. lambda, 34 in T.lyrophanes) that are wider (mean of eight scales wide alongmiddorsal line at midbody in T. paucimaculatus, five in T.lambda and T. lyrophanes), from T. vilkinsonii by havingwider blotches (mean of eight scales wide along middorsalline at midbody in T. paucimaculatus, three in T. vilkinsonii)that are more narrowly separated (seventh and eighthblotches separated by four scale rows on average in T.
paucimaculatus, eight in T. vilkinsonii), from T. biscutatus bythe absence of H-shaped blotches and additional markingsbetween the primary blotches, and from T. quadruplex byhaving undivided secondary blotches.
A species of Trimorphodon defined by the followingcombination of characteristics: primary dark body blotchesfewer than 31 (mean of 21), 5–13 scales wide alongmiddorsal line at midbody (mean of nine), dark-edged,saddle-shaped; central pale spots variable, but may divideprimary blotches, secondary blotches undivided; seventhand eighth blotches separated by four scale rows on average;ventrals 237–258 in males, 248–265 in females; midbodyscale rows 22–27 (mean 24); complete lyre-shaped markingon top of head, but becoming almost divided at the apex insome specimens, approaching the condition exhibited by T.lyrophanes; a dark bar between the eyes or across snout; analdivided; size medium (maximum observed SVL of males 5
1018 mm, females 5 1074 mm).
Distribution.—Found generally from northern Sinaloa south-ward along the coast to the vicinity of the Rıo Purificacion,Jalisco, Mexico.
Remarks.—Taylor (1938:527) described this taxon based onpreoculars separated from the frontal, and ‘‘dorsal spotsgreatly elongated and fewer in number.’’ Later (1939:60), henoted ‘‘a character not previously mentioned, whichdifferentiates this form from bi-scutatus, is the absence ofkeels or ridges on the scales in the males.’’ We scored onlythe number of preoculars, not whether they were separatedfrom the frontal, and did not observe keeled scales in eithersex.
Gehlbach (1971) synonymized Trimorphodon lambda pau-cimaculatus with T. biscutatus biscutatus. Trimorphodonpaucimaculatus may come into contact with T. lambda innorthern Sinaloa, and is narrowly sympatric with T.biscutatus in southwestern Jalisco.
Suggested common name.—Sinaloan Lyresnake.
Trimorphodon quadruplex (Smith, 1941)Figure 1F
Trimorphodon biscutatus quadruplex Smith, 1941:157.
Holotype.—USNM 89476, female, Nicaragua, Esteli, 1932,James H. Ivy (not examined by authors).
Diagnosis.—This species is distinguished from all otherspecies of Trimorphodon by blotches that are divided bycentral light spots into secondary blotches, which arefurther subdivided by an additional light space.
A species of Trimorphodon defined by the followingcombination of characteristics: primary dark body blotchesfewer than 24 (mean of 19), 5–15 scales wide alongmiddorsal line at midbody (mean of ten), central light spotsdivide body blotches into secondary blotches, which are alsofurther subdivided by central light markings; seventh andeighth blotches separated by four scale rows on average;ventrals 249–266 in males, 250–266 in females; midbodyscale rows 23–27 (mean of 25); lyre-shaped marking on topof head complete, with a dark bar between the eyes or acrosssnout; anal divided; size large (maximum observed SVL ofmales 5 1595 mm, females 1302 mm).
Devitt et al.—Variation in Trimorphodon biscutatus 379
Distribution.—Found from southeastern Guatemala southalong the Pacific Coast to Puntarenas Province, Costa Rica;also present in some arid Atlantic slope valleys in Guate-mala, Honduras, and Nicaragua.
Remarks.—Scott and McDiarmid (1984a) inferred that Smith(1942) elevated Trimorphodon biscutatus quadruplex to specieslevel, but formal taxonomic changes were never actuallymade. Although previous workers concluded that the breakbetween T. b. biscutatus and T. b. quadruplex occurred at theIsthmus of Tehuantepec (Gehlbach, 1971), phylogeneticanalyses (Devitt, 2006) together with color pattern differ-ences (this study) indicate genetic and morphologicaldiscontinuities further south across the Guatemalan high-lands.
Suggested common name.—Central American Lyresnake.
Trimorphodon vilkinsonii Cope, 1886Figure 1C
Trimorphodon vilkinsonii Cope 1886:285.Trimorphodon wilkinsonii Cope, 1887:68.Trimorphodon upsilon Gunther, 1895:175 (in part).Trimorphodon vilkinsoni Werner, 1929:181.Trimorphodon biscutatus vilkinsoni Gehlbach, 1971:209.Trimorphodon lambda vilkinsoni Werler and Dixon, 2000
Holotype.—USNM 14268, juvenile male, Mexico, City ofChihuahua, collection date unknown, Edward Wilkinson(examined by TJL).
Diagnosis.—This species is distinguished from T. tau by theabsence of a broad, pale band across the nape, from both T.lambda and T. lyrophanes by having fewer body blotches(mean of 21 in T. vilkinsonii, 24 in T. lambda, 34 in T.lyrophanes) that are more widely separated (seventh andeighth blotches separated by eight scale rows on average inT. vilkinsonii, four in T. lambda, two in T. lyrophanes), from T.paucimaculatus by having more narrow body blotches (meanof three scales wide along middorsal line at midbody in T.vilkinsonii, eight in T. paucimaculatus) that are more widelyseparated (seventh and eighth blotches separated by eightscale rows on average in T. vilkinsonii, four in T. paucima-culatus), from T. biscutatus by the absence of H-shapedblotches and additional markings between the primaryblotches, and from T. quadruplex by having undividedsecondary blotches.
A species of Trimorphodon defined by the followingcombination of characteristics: primary dark body blotchesfewer than 25 (mean of 21), 2–4 scales wide along middorsalline at midbody (mean of three), central light spots that donot divide blotches; seventh and eighth blotches separatedby eight scale rows on average; ventrals 221–251 in males,225–241 in females; midbody scale rows 21–25 (mean of 22);head pattern variable, but usually consisting of a veryforeshortened chevron or only dark spots on frontals andparietals; without a dark bar between eyes or across snout;anal divided; size small (maximum observed SVL of males 5
668 mm, females 5 837 mm).
Distribution.—Found in southern New Mexico (Dona Ana,Grant, Hidalgo, Luna, and Sierra counties), southwesternTexas (Brewster, El Paso, Hudspeth, Jeff Davis and Presidio
counties), and east of the Sierra Madre Occidental innorthern Chihuahua and probably northwestern Coahuila,Mexico.
Remarks.—Trimorphodon vilkinsonii may hybridize with T.lambda in extreme southeastern Arizona (Cochise County)and southwestern New Mexico (Grant and Hidalgo coun-ties). Werler and Dixon (2000) applied the name Trimorpho-don lambda vilkinsoni to this taxon without explanation.LaDuc and Johnson (2003) elevated Trimorphodon biscutatusvilkinsonii to species level, and resurrected the nameoriginally applied to this taxon, T. vilkinsonii. We recom-mend that the standard English name for this species bechanged to the ‘‘Chihuahuan Lyresnake’’ rather than the‘‘Chihuahuan Desert Lyresnake’’ (sensu LaDuc and Johnson,2003) in the interest of brevity. Gehlbach (1971) restrictedthe type locality to ‘‘vicinity of Ciudad Chihuahua.’’
Suggested common name.—Chihuahuan Lyresnake.
KEY TO THE SPECIES OF TRIMORPHODON(after Smith, 1941; Scott and McDiarmid, 1984a)
1a. Head pattern characterized by a broad, pale bandwith a relatively even posterior border across thenape _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ T. tau
1b. Head pattern without a broad, pale band across thenape _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 2
2a. Head pattern consisting of dark chevron-shaped orlyre-shaped marking and with a dark bar betweeneyes or across snout; dark blotches generally wide(.4 scales wide), and narrowly separated (generally,7 scale rows between blotches), often with centrallight spots that divide blotches _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 3
2b. Head pattern consisting of a very foreshortenedchevron-shaped mark or only dark spots on frontalsand parietals, and without a dark bar between eyesor across snout; dark blotches narrow (usually 2–4scales wide), widely separated (7–8 scale rowsbetween blotches), and usually without central lightspots that divide blotches _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ T. vilkinsonii
3a. Primary dark blotches divided by central light bandinto undivided secondary blotches _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 4
3b. Primary dark blotches divided by central light bandinto secondary blotches, most of which are furthersubdivided by central light markings _ _ _ _ T. quadruplex
4a. Primary dark blotches saddle-shaped withoutsmaller, paired dorsal markings between primaryblotches _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 5
4b. Primary dark blotches often H-shaped, especiallyanteriorly, and often with small, elongate paireddorsal markings between primary blotches _ _ _ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ T. biscutatus5a. Fewer than about 25 dark body blotches _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 65b. More than about 25 dark body blotches (average
34), each 4–6 scale rows wide, with about 2–3 scalerows between blotches and with central lightmarkings; often with a lyre-shaped head mar-king _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ T. lyrophanes
6a. Dark blotches elongate (about 10–11 scales wideanteriorly, slightly narrower posteriorly), saddle-shaped, dark-edged, with central light markings,usually numbering about 21 _ _ _ _ _ _ _ T. paucimaculatus
6b. Dark blotches about as broad as long (about 4–8scales wide, 4–5 scale rows between blotches), with
380 Copeia 2008, No. 2
central pale spots that may entirely divide blotch-es, usually numbering about 24 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ T. lambda
MATERIAL EXAMINED
Trimorphodon vilkinsonii (63): Mexico: Chihuahua: AMNH96653, CM 60117–9, UNM 34382–4, USNM 14268. UnitedStates: New Mexico: Dona Ana Co.: NMSU 3615, 5095, 6944;Luna Co.: UTEP 11273; Sierra Co.: NMSU 5489–90, UNM16264–6, UNM 17799, UNM 25668, UTEP 15597. Texas:Brewster Co.: BYU 40886, SRSU 2550, 3713, 4043, 6228,6231, UTEP 10724–5,11200,12146; El Paso Co.: AMNH28862, CAS 139998, UTEP 1752, 1754, 1979, 2328, 2329,2452, 2458, 2482, 2484, 2491, 2492, 2526, 2532, 2533, 2694,4017, 6227, 12104, 12330, 13671–2; Jeff Davis Co.: LSUMZ34726; Presidio Co.: FMNH 201039, SRSU 1689, 4595, 4767,TNHC 3332, 28404, UTEP 10762, 11358.
Trimorphodon lambda (174): Mexico: Sonora: AMNHR151227, BYU 36809, 37202, 37174, 41159, 41196, CAS-SU 24047, 24051, KU 67732, LACM 122903, SRSU 1484,UAZ 9633–4, 9637–41, 27062–3, 28065–6, 30737, 39750,44011–2, 45603, 45902, 46462, 46781. United States:Arizona: Cochise Co.: AMNH 80160, 85043–5, 94858–9,99334–5, 111209, 115624, 126747, BYU 16744, 32086, CAS190689–90, CM 69729, NMSU 2384, SWRS 158–9, UAZ26931, 26958, 26966, 37760, 37762, 39595, 39756–7, 40059,41560–1, 41563, 41576, 43601, 46851, UF 53466, 84195,UMMZ 120313, UNM 25235, 49333, 49742, 50227, 50449,51682, 53654, 53910, 53939, UTEP 5477, 11409; Gila Co.:CAS 80656, CM 53753, FMNH 74970; Graham Co.: UAZ26955; Greenlee Co.: CAS 190691; La Paz Co.: UTEP 15989;Maricopa Co.: CM 48621, 48669, 51418; Pima Co.: CAS33846, 129202, 152522, 158220, 190686, FMNH 74971,LSUMZ 28649, SRSU 1485, 5662, UAZ 26929, 26934, 26947,26951, 26959, 26965, 35788, 37767–8, 37840, 37849, 39593,46270, 46495, 47321, 48894, 49428, 50086, 50197, UTEP2771, 11021, 11199, 15326; Pinal Co.: UAZ 26927, 37764–5;Santa Cruz Co.: LSUMZ 32895, UAZ 26927, 26939-41,26962, UMMZ 75785, 75809, 178023, UTEP 15916; YavapaiCo.: UAZ 37850, 46285; Yuma Co.: UNM 41641, 55161; NewMexico: Catron Co.: NMSU 2959, UNM 48105; Hidalgo Co.:AMNH 80380, MVZ 67228, NMSU 4958, UAZ 35952, UNM4200. Nevada: Clark Co.: CAS-SU 19921; Nye Co.: CAS-SU19922; BYU 17939, 23727; Utah: Washington Co.: BYU 501-2, 653, 1798, 23801.
Trimorphodon lyrophanes (51): Mexico: Baja California: KU185657–8, LACM 126261, 134001; Baja California Sur:LACM 103399, MVZ 11903–4, 57931, 104219–20, 117313–4, 117372, 128495, 140881, 142046, 161562. United States:California: Imperial Co.: LACM 27920, 67280; Inyo Co.: CAS65496, 143731; Kern Co.: KU 62902, LACM 132249; OrangeCo.: LACM 103402; Riverside Co.: LACM 103404–11,103413–6; San Bernadino: LACM 2711; San Diego Co.: KU8497, 74366–8, LACM 2710, 27924, 28700, 103412, 103433,103434–8, 103439.
Trimorphodon paucimaculatus (41): Mexico: Jalisco: KU67734, 187744, LACM 25932; Nayarit: LACM 103374–6,103378–82, MVZ 70278, 71334; Sinaloa: CAS-SU 24052, KU73636, 78931, 80765, LACM 7089–93, 7095, 7097–8, 7100,28716, 103379, 103381, 103383–9, 122377, 122902, ROM14959, UAZ 37770, 37708.
Trimorphodon biscutatus (58): Mexico: Chiapas: CAS140948, 163754, LACM 38210, 122900–1, UMMZ 113770;Colima: KU 67731, LACM 37335, 59135, 130115, MVZ72194, 76716, 161563, UMMZ 80200–1, 114567; Guerrero:
KU 87474, LACM 58136, MVZ 161565–7, UMMZ 85768;Jalisco: KU 95787, 95970, 106288, 37336, 136970–1, LACM37337; Michoacan: KU 29492, MVZ 170791, UMMZ 104606,105154, 112517, 112519–21, 121529, 114565–6, 118953;Oaxaca: CAS 143898, KU 137675, LACM 7103, 8480, 38207–9, 58137–8, 65245, 103370, 103372, 114146, 128455.
Trimorphodon quadruplex (42): Costa Rica: Guanacaste: KU35513, 63894–5, 102532–4, LACM 113916–8, 114144, MVZ170797, 80026, UMMZ 123332, 123588, 131292–3; Puntar-enas: KU 63896, 86586. El Salvador: KU 183972; La Libertad:KU 116968, 183971; Sonsonate: LACM 114145. Guatemala:UMMZ 131294–5; El Progreso: KU 191125; MVZ 146984–5;Santa Rosa: UMMZ 107347–8. Honduras: Choluteca: KU140081, MVZ 78764; Valle: KU 116967. Nicaragua: Chinan-dega: KU 125012; Chontales: KU 63897; Managua: KU86259, 174425–8; Matagalpa: KU 42307–10, 86260; Rivas:KU 101925, 125011.
ACKNOWLEDGMENTS
We thank the McGuire lab for comments on a previous draftof this manuscript. We thank J. Parra for translating theabstract. For participating in a blind evaluation of the key tospecies, we thank P. Nicodemo (AMNH). For assistance in theearly stages of this project, we thank D. Cannatella. We thankthe California Academy of Sciences for a Charles StearnsGrant-in-Aid that allowed the first author to examinespecimens in their collection. For loans of preserved speci-mens, we thank the following institutions and individuals:AMNH (D. Frost, L. Ford); BYU (J. Sites); CAS (B. Drewes, J.Vindum); CM (the late C. J. McCoy, E. Censky); FMNH (A.Resetar); KU (C. Raxworthy, E. Greenbaum); LACM (K.Beaman, R. Bezy, D. Kizirian, J. Wright); LSUMZ (D. Ross-man); MVZ (C. Cicero, H. Greene, E. Taylor); NMSU (C.Benkman, P. Hyder); ROM (R. Murphy); SRSU (J. Scudday);SWRS (W. Sherbrooke); TCWC (J. Dixon, K. Vaughn); TNHC(D. Cannatella, C. Malcolm); UAZ (G. Bradley); UCM (A. deQuieroz); UF (D. Auth); UIMNH (D. Blake, S. Sroka); UMMZ(G. Schneider, A. Kluge); UNM (H. Snell); USNM (R.McDiarmid, S. Gotte, B. Reynolds); UTA (J. Campbell, C.Stewart); UTEP (C. Lieb, R. Webb); and E. A. Liner.
LITERATURE CITED
Bogert, C. M., and J. A. Oliver. 1945. A preliminary analysisof the herpetofauna of Sonora. Bulletin of the AmericanMuseum of Natural History 83:297–426.
Boulenger, G. A. 1896. Catalogue of Snakes in the BritishMuseum (Natural History) Vol. III. The Trustees of theBritish Museum, London.
Campbell, J. A. 1999. Distribution patterns of amphibiansin Middle America, p. 111–210. In: Patterns of Distributionof Amphibians. W. E. Duellman (ed.). The Johns HopkinsUniversity Press, Baltimore, Maryland.
Campbell, J. A., and J. P. Vannini. 1988. A new sub-species of beaded lizard, Heloderma horridum, from theMotagua Valley of Guatemala. Journal of Herpetology 22:457–468.
Cochran, D. M. 1961. Type specimens of reptiles andamphibians in the United States National Museum.United States National Museum Bulletin 220:xv + 291.
Cope, E. D. 1861a. Contributions to the ophiology of LowerCalifornia, Mexico and Central America. Proceedings ofthe Academy of Natural Sciences of Philadelphia 13:292–306.
Devitt et al.—Variation in Trimorphodon biscutatus 381
Cope, E. D. 1861b. Notes and descriptions of new andlittle known species of American reptiles. Proceedingsof the Academy of Natural Sciences of Philadelphia12:339–345.
Cope, E. D. 1870. Seventh contribution to the herpetologyof tropical America. Proceedings of the American Philo-sophical Society 11:147–169.
Cope, E. D. 1875a. On the Batrachia and Reptilia of CostaRica. Journal of the Academy of Natural Sciences ofPhiladelphia, Series 2, 8:93–154.
Cope, E. D. 1875b. Check-list of North American Batrachiaand Reptilia. United States National Museum Bulletin1:1–104.
Cope, E. D. 1886 (‘‘1885’’). Thirteenth contribution to theherpetology of tropical America. Proceedings of theAmerican Philosophical Society 23:271–287.
Cope, E. D. 1887. Catalogue of batrachians and reptiles ofCentral America and Mexico. Bulletin of the United StatesNational Museum 32:1–98.
de Queiroz, K. 1998. The general lineage concept of species,species criteria, and the process of speciation: a conceptualunification and terminological recommendations, p. 57–75.In: Endless Forms: Species and Speciation. D. J. Howard andS. H. Berlocher (eds.). Oxford University Press, Oxford, U.K.
de Queiroz, K., and M. J. Donoghue. 1988. Phylogeneticsystematics and the species problem. Cladistics 4:317–338.
Devitt, T. J. 2006. Phylogeography of the Western Lyresnake(Trimorphodon biscutatus): testing aridland biogeographi-cal hypotheses across the Nearctic–Neotropical transition.Molecular Ecology 15:4387–4407.
Dixon, J. R., M. Sabbath, and R. Worthington. 1962.Comments on snakes from central and western Mexico.Herpetologica 18:91–100.
Duellman, W. E. 1954. The amphibians and reptiles ofJorullo Volcano, Michoacan, Mexico. Occasional Papers ofthe Museum of Zoology, University of Michigan 560:1–24.
Duellman, W. E. 1957. Notes on snakes from the Mexicanstate of Sinaloa. Herpetologica 13:237–240.
Duges, A. 1882. Nota sobre el colcoatl o Trimorphodon(Dipsas) biscutata, D. B. La Naturaleza 6:145–148.
Duges, A. 1884. Elementos de Zoologıa. Oficina Tipograficade la Secretaria de Fomento. Distrito Federal, Mexico.
Duges, A. 1893. Variaciones de coloracion en el Gerrhonotusimbricatus. La Naturaleza (Serie 2) 2:294–295.
Dumeril, A. M. C., G. Bibron, and A. Dumeril. 1854.Erpetologie Generale ou Historie Naturelle Complete desReptiles. Volume 7. Librarie Encyclopedique de Roret,Paris, France.
Fouquette, M. J., and D. A. Rossman. 1963. Noteworthyrecords of Mexican amphibians and reptiles in the FloridaState Museum and the Texas Natural History Collection.Herpetologica 19:185–201.
Fugler, C. M., and J. R. Dixon. 1961. Notes on theherpetofauna of the El Dorado area of Sinaloa, Mexico.Publications of the Museum, Michigan State University(Biological Series) 2:1–24.
Funk, D. J., and K. E. Omland. 2003. Species-levelparaphyly and polyphyly: frequency, causes, and conse-quences, with insights from animal mitochondrial DNA.Annual Review of Ecology, Evolution, and Systematics34:397–423.
Garman, S. 1884a (‘‘1883’’). The reptiles and batrachians ofNorth America, Part I, Ophidia. Memoirs of the Museumof Comparative Zoology 8:xxxi + 185.
Garman, S. 1884b. The North American reptiles andbatrachians. Bulletin of the Essex Institute 16:1–46.
Gehlbach, F. R. 1971. Lyre snakes of the Trimorphodonbiscutatus complex: a taxonomic resume. Herpetologica27:200–211.
Grismer, L. L., J. A. McGuire, and B. D. Hollingsworth.1994. A report on the herpetofauna of the VizcaınoPeninsula, Baja California, Mexico, with a discussion of itsbiogeographic and taxonomic implications. Bulletin ofthe Southern California Academy of Sciences 93:45–80.
Gunther, A. C. L. G. 1858. Catalogue of the ColubrineSnakes in the Collection of the British Museum. Taylorand Francis, London.
Gunther, A. C. L. G. 1895. Reptilia and Batrachia, part 122,p. 169–176. In: Biologia Centrali-Americana; or, Contri-butions to the Knowledge of the Fauna and Flora ofMexico and Central America. O. Salvin and F. D. Godman(eds.). R. H. Porter and Dulan, and Co., London, Reprintedin 1987. Fascimile Reprints in Herpetology, Society for theStudy of Amphibians and Reptiles, Athens, Ohio.
Hardy, L. M., and R. W. McDiarmid. 1969. The amphibiansand reptiles of Sinaloa, Mexico. University of KansasPublications 3:39–252.
Harris, R. J. 1975. A Primer of Multivariate Statistics.Academic Press, New York.
Jan, G. 1863. Elenco Sistematico degli Ofidi Descritti eDisegnati per l’Iconografia Generale. Lombardi, Milan, Italy.
Jones, C., and J. S. Findley. 1963. Second record of the lyresnake, Trimorphodon vilkinsoni, in New Mexico. South-western Naturalist 8:175–177.
Klauber, L. M. 1924. Notes on the distribution of snakes inSan Diego County, California. Bulletin of the ZoologicalSociety of San Diego 1:1–23.
Klauber, L. M. 1928. The Trimorphodon (lyre snakes) ofCalifornia, with notes on the species of the adjacent areas.Transactions of the San Diego Society of Natural History5:183–194.
Klauber, L. M. 1940. The lyre snakes (genus Trimorphodon)of the United States. Transactions of the San Diego Societyof Natural History 9:163–194.
LaDuc, T. J., and J. D. Johnson. 2003. A taxonomic revisionof Trimorphodon biscutatus vilkinsonii (Serpentes: Colubri-dae). Herpetologica 59:365–375.
Lee, J. C. 1990. Sources of extraneous variation in the studyof meristic characters: the effect of size and of inter-observer variability. Systematic Zoology 39:31–39.
Leviton, A. E., R. H. Gibbs, Jr., E. Heal, and C. E. Dawson.1985. Standards in herpetology and ichthyology: part I.Standard symbolic codes for institutional resource collec-tions in herpetology and ichthyology. Copeia 1985:802–832.
Loomis, R. B., and R. C. Stephens. 1967. Additional noteson the snakes taken in and near Joshua Tree NationalMonument, California. Bulletin of the Southern Califor-nia Academy of Sciences 66:1–22.
McDiarmid, R. W., and N. J. Scott, Jr. 1970. Geographicvariation and systematic status of Mexican lyre snakes ofthe Trimorphodon tau group (Colubridae). Los AngelesCounty Museum Contributions in Science 179:1–43.
McGarigal, K., S. Cushman, and S. G. Stafford. 2000.Multivariate Statistics for Wildlife and Ecology Research.Springer-Verlag, New York.
Morafka, D. J. 1977. A Biogeographical Analysis of theChihuahuan Desert through its Herpetofauna. Dr. W.Junk B.V., Publishers, The Hague.
382 Copeia 2008, No. 2
Nielsen, R., and J. Wakeley. 2001. Distinguishing migra-tion from isolation: a Markov Chain Monte Carloapproach. Genetics 158:885–896.
Reist, J. D. 1986. An empirical evaluation of coefficientsused in residual and allometric adjustment of sizecovariation. Canadian Journal of Zoology 64:1363–1368.
Scheffe, H. 1953. A method for judging all contrasts in theanalysis of variance. Biometrika 40:87–104.
Scott, N. J., Jr., and R. W. McDiarmid. 1984a. Trimorphodonbiscutatus. Catalogue of American Amphibians and Rep-tiles 353:1–4.
Scott, N. J., Jr., and R. W. McDiarmid. 1984b. Trimorphodontau. Catalogue of American Amphibians and Reptiles354:1–2.
Simpson, G. G. 1961. Principles of Animal Taxonomy.Columbia University Press, New York.
Smith, H. M. 1941. Notes on the snake genus Trimorphodon.Proceedings of the United States National Museum91:149–168.
Smith, H. M. 1942. Remarks on Mexican king snakes of thetriangulum group. Proceedings of the Rochester Academyof Science 8:196–207.
Smith, H. M. 1943. Summary of the collections of snakesand crocodilians made in Mexico under the WalterRathbone Bacon traveling scholarship. Proceedings ofthe United States National Museum 93:393–504.
Smith, H. M., and E. H. Taylor. 1950. Type localities ofMexican reptiles and amphibians. University of KansasScience Bulletin 33:313–380.
Tabachnick, B. G., and L. S. Fidell. 2001. Using Multivar-iate Statistics, Fourth Edition. Allyn and Bacon, Boston,Massachusetts.
Taylor, E. H. 1938 (‘‘1936’’). Notes on the herpetologicalfauna of the Mexican state of Sinaloa. University of KansasScience Bulletin 24:505–537.
Taylor, E. H. 1939 (‘‘1938’’). On Mexican snakes of thegenera Trimorphodon and Hypsiglena. University of KansasScience Bulletin 25:357–383.
Velasco, A. L. 1890. Geografıa y estadıstica del estado deGuanajuato. Geografıa y estadıstica de la RepublicaMexicana. Tomo V. Oficina Tipografica de la Secretariade Fomento. Distrito Federal, Mexico.
Werler, J. E., and J. R. Dixon. 2000. Texas Snakes:Identification, Distribution, and Natural History. Univer-sity of Texas Press, Austin, Texas.
Werner, F. 1929. Ubersicht der Gattungen und Artender Schlangen aus der Familie Colubridae. III. Teil(Colubrinae). Zoologische Jahrbucher, Abteilung fur
Systematik, Okologie und Geographie der Tiere 57:1–196.
Wiens, J. 2004. What is speciation and how should we studyit? The American Naturalist 163:914–923.
Wiens, J. J., and T. A. Penkrot. 2002. Delimiting speciesusing DNA and morphological variation and discordantspecies limits in spiny lizards (Sceloporus). SystematicBiology 51:69–91.
Wiley, E. O. 1978. The evolutionary species conceptreconsidered. Systematic Zoology 27:17–26.
Appendix 1
Morphological character abbreviations: Snout–vent length(SVL); Tail length (TL); Snout length (SNL); Head length(HDL); Distance between eyes (DBES); Distance between eyeand nostril (DBEN); Number of ventral scales (VNT);Number of subcaudal scales excluding the terminal spine(SBCDL); Number of dorsal scale rows at the 25th ventral(SRW1), midbody (SRW2), and 25th ventral anterior to thevent (SRW3); Number of supralabials on the left (SUPRAL)and right (SUPRAR) sides of the head; Number of infralabialson the left (INFRAL) and right (INFRAR) sides of the head;Number of preoculars on the left (PREOCL) and right(PREOCR) sides of the head; Number of postoculars on theleft (POSTL) and right (POSTR) sides of the head; Number ofloreals on the left (LORL) and right (LORR) sides of the head;Number of body blotches (BBTH); Number of tail blotches(TBTH); Number of scales along the midline encompassedby a blotch (including the light center) at the 25th ventral(SWBTH1), at midbody (SWBTH2), and at the 25th ventralfrom vent (SWBTH3); Number of scales along the midlineencompassed by the light center of a blotch at the 25th
ventral (LHTSP1), at midbody (LHTSP2), and at the 25th
ventral anterior to the vent (LHTSP3); Number of scalesalong the midline in the space between blotches betweenthe first and second blotches (BSQ12), seventh and eighthblotches (BSQ78), thirteenth and fourteenth blotches(BSQ1314), and last set of blotches anterior to vent(BSQAN); Nature (1 5 divided or 0 5 entire) of the primary(PBDIV) and secondary (SBDIV) blotches; Completeness ofthe lyre-shaped head marking (LYRE; 0 5 no markings, 1 5
only spots, 2 5 lyre mostly incomplete, 3 5 lyre mostlycomplete, 4 5 complete lyre present); Number of H-shapedblotches (HSBTH; 0 5 no H-shaped blotches, 1 5 first blotchH- shaped, 2 5 1–5 blotches H-shaped, 3 5 6–10 blotches H-shaped, 4 5 11–15 blotches H-shaped); Nature of the analplate (ANAL; 1 5 divided or 0 5 entire).
Devitt et al.—Variation in Trimorphodon biscutatus 383
Men
sura
l(m
m)
an
dm
eri
stic
(nu
mb
er
ofsc
ales)
vari
atio
nin
Trim
orp
hod
on
bis
cuta
tus
bis
cuta
tus,
T.b
.la
mb
da
,T.b
.lyr
op
ha
nes,
T.b
.qua
dru
ple
x,T.
vilk
inso
nii,
an
dT.
pa
uci
ma
cula
tus.
Table
entrie
sfo
rea
ch
char
acte
rin
clude
mea
n6
1st
andar
ddev
iatio
n,r
ange
,sam
ple
size
,and
resu
ltsfrom
AN
OVA
usi
ng
Schef
fe’s
Fpro
cedure
forpost
hoc
com
par
isons.
TL(t
aill
engt
h),
SNL
(snoutle
ngt
h),
DB
ES(d
ista
nce
bet
wee
ney
es),
and
DB
EN(d
ista
nce
bet
wee
ney
ean
dnost
ril)
wer
ete
sted
fordiff
eren
ces
usi
ng
AN
CO
VA
wher
eSV
Lw
asth
eco
varia
te.N
um
ber
sin
dic
ate
asi
gnifi
cant(
P,
0.0
5)
pai
rwis
ediff
eren
cein
mea
ns
bet
wee
nth
epopula
tion(s
)
liste
dan
dth
epopula
tion
inw
hose
colu
mn
they
appea
r.N
.S.i
ndic
ates
no
sign
ifica
ntd
iffer
ence
sex
ista
mong
any
poss
ible
pai
rwis
eco
mpar
ison.*
Indic
ates
dis
cret
ech
arac
ter.{
Val
ues
indic
ate
freq
uen
cyofa
nen
tire
anal
pla
te(%
),sa
mple
size
,an
dre
sults
from
AN
OVA
usi
ng
Schef
fe’s
Fpro
cedure
for
post
hoc
com
par
isons.
See
Appen
dix
1fo
rch
arac
ter
abbre
viat
ions.
T.vi
lkin
son
ii(1
)T.
b.la
mb
da
(2)
T.b
.ly
rop
ha
nes
(3)
T.p
au
cim
acu
latu
s(4
)T.
b.b
iscu
tatu
s(5
)T.
b.qu
ad
rup
lex
(6)
male
fem
ale
male
fem
ale
male
fem
ale
male
fem
ale
male
fem
ale
male
fem
ale
SVL
49
3.5
68
8.5
56
4.2
61
32
.44
93
.16
12
5.6
59
9.9
61
85
.55
35
.76
94
.85
31
.26
19
9.8
71
2.7
61
65
.08
09
.26
22
2.2
74
3.6
62
42
.68
64
.76
35
4.3
85
4.7
63
71
.47
64
.96
32
8.7
22
2–
66
83
50
–8
37
23
1–
88
52
37
–8
90
36
4–
77
42
40
–9
36
45
7–
10
18
29
3–
10
74
30
3–
12
55
36
5–
15
90
26
8–
15
95
29
5–
13
02
39
21
10
95
82
92
22
41
63
81
93
01
31
,4
,6
11
,4
,6
11
,4
,6
12
,3
,5
N.S.
2,3
,5
2,3
,5
2,3
,5
N.S.
TL1
07
.76
19
.11
05
.86
20
.51
07
.36
27
.71
04
.46
32
.09
4.6
61
8.7
81
.66
28
.51
50
62
7.1
14
4.9
63
8.7
15
3.0
65
1.9
14
7.3
65
8.8
16
4.7
66
0.1
14
6.6
65
7.1
46
–1
38
66
–1
46
44
–1
64
41
–1
60
67
–1
51
38
–1
38
99
–1
88
51
–1
95
61
–2
45
61
–2
61
52
–24
96
3–
21
53
61
91
02
53
29
20
24
15
37
17
30
12
1,3
,4
,6
1,3
,4
,6
3,4
,6
1,3
,4
,6
4,5
,6
1,2
,4
,5
,6
2,3
,4
,5
2,3
,5
2,3
,5
2,3
,5
2,3
,5
,6
2,3
,5
SNL
5.4
61
.15
.36
0.9
5.7
61
.45
.76
1.4
5.4
61
.05
.36
1.7
6.5
61
.37
.16
1.6
6.7
61
.97
.86
2.6
7.6
62
.57
.36
2.7
3.2
–8
.43
.7–
7.0
2.7
–9
.43
.2–
8.9
3.9
–8
.33
.0–
9.0
3.6
–8
.63
.36
8.9
3.5
–1
0.9
4.3
–1
3.0
3.7
–1
3.6
3.5
–1
2.1
37
22
10
65
62
92
12
21
33
41
93
01
31
,4
1,4
,6
1,4
1,4
1,4
,6
1,4
,6
N.S.
3,5
2,3
,5
2,3
,5
2,3
,5
2,3
,5
HD
L1
3.1
62
.51
3.1
61
.81
3.9
62
.81
4.2
63
.01
3.4
62
.01
3.0
63
.51
5.7
62
.91
6.8
63
.51
6.7
63
.71
8.5
65
.61
8.4
64
.91
7.3
65
.68
.4–
20
.31
0.7
–1
6.6
8.8
–2
0.6
9.0
–2
1.1
10
.7–
18
.78
.9–
20
9.6
–2
0.4
9.7
–2
1.8
11
.4–
24
.41
1.3
–2
9.9
10
.5–
27
.11
0.1
–2
7.6
37
22
10
95
62
92
12
21
43
51
93
01
31
,4
,6
1,4
,6
1,4
,6
1,4
1,4
,6
1,4
,6
2,3
,5
3,5
2,3
,5
2,3
,5
2,3
,5
,6
2,3
,5
DB
ES5
.96
1.2
5.8
60
.96
.36
1.3
6.2
61
.36
.06
0.8
5.8
61
.47
.26
1.2
7.7
61
.57
.36
1.8
8.1
62
.57
.86
2.3
7.5
62
.33
.7–
9.3
4.2
–7
.24
.0–
9.3
3.6
–9
.24
.5–
8.4
4.0
–8
.64
.2–
8.9
4.2
–9
.74
.5–
10
.94
.2–
13
.04
.2–
13
4.7
–1
1.7
37
21
10
85
62
92
12
11
43
51
92
91
31
,4
,6
1,4
,6
1,4
1,4
,6
1,4
,6
1,4
,6
3,5
2,3
,5
2,3
,5
2,3
,5
2,3
,5
2,3
,5
DB
EN3
.96
0.9
3.9
60
.74
.16
1.0
4.2
61
.13
.86
0.7
3.7
61
.34
.96
1.1
5.2
61
.34
.96
1.5
5.8
62
.05
.56
1.7
5.3
62
.12
.1–
6.6
2.8
–5
.21
.8–
6.6
2.3
–7
.02
.6–
5.5
2.2
–6
.22
.5–
6.4
2.2
–6
.72
.7–
7.7
2.9
–1
0.0
2.8
–9
.62
.6–
8.6
37
21
10
45
62
92
12
11
33
52
53
01
31
,4
1,4
,6
1,4
,6
1,4
,6
1,4
,6
1,4
,6
2,3
,5
2,3
,5
2,3
,5
2,3
,5
,6
2,3
,5
2,3
,5
VN
T2
27
.56
5.2
23
4.4
64
.72
23
.96
4.8
23
1.5
65
.12
27
.76
6.1
23
3.5
66
.92
46
.36
5.0
25
2.9
64
.32
60
.06
7.6
26
7.6
68
.12
58
.06
4.9
26
0.6
64
.22
21
–2
51
22
5–
24
12
11
–2
40
22
3–
24
92
14
–2
37
22
1–
25
02
37
–2
58
24
8–
26
52
32
–2
79
25
4–
28
62
49
–2
66
25
0–2
66
39
21
10
65
72
82
22
41
63
71
93
01
31
,4
,6
1,4
,6
1,3
,4
,6
1,4
,6
1,2
,4
,6
1,4
,6
1,2
,3
,4
,5
1,2
,3
,5
2,3
,5
,6
2,3
,4
,5
,6
2,3
,5
,6
1,2
,3
,5
SBC
DL
78
.56
3.4
71
.16
6.4
77
.76
3.5
69
.16
3.8
69
.86
3.3
63
.26
4.3
85
.66
4.0
74
.76
4.7
93
.66
6.6
85
.96
4.1
88
.66
5.8
79
.86
3.7
69
–8
45
9–
84
69
–8
76
0–
78
66
–7
85
6–
73
75
–9
26
3–
81
77
–1
05
79
–9
37
3–
99
74
–8
43
72
01
03
52
29
20
23
14
36
17
30
12
1,3
,4
,6
1,3
,4
1,3
,4
,6
1,3
,4
1,2
,4
,5
,6
1,2
,4
,5
,6
1,2
,3
,5
1,3
2,3
,4
,5
,6
2,3
,5
,6
1,2
,3
,5
2,3
,5
SRW
12
0.9
60
.42
1.1
60
.52
0.7
60
.72
1.1
60
.72
1.3
60
.62
2.0
61
.02
2.4
60
.82
2.8
61
.02
3.7
60
.92
4.6
60
.82
4.1
60
.82
4.8
60
.72
0–
22
20
–2
21
8–
23
19
–2
32
0–
23
19
–2
32
1–
24
21
–2
52
1–
25
23
–2
62
3–
26
24
–2
64
02
11
11
59
29
21
24
16
37
19
30
12
1,4
,6
1,4
,6
1,3
,4
,6
1,3
,4
,6
1,2
,4
,6
1,2
,4
1,2
,3
,5
1,2
,4
,5
2,3
,5
,6
2,3
,5
,6
2,3
,5
,6
2,3
,5
,6
App
end
ix2
384 Copeia 2008, No. 2
T.vi
lkin
son
ii(1
)T.
b.la
mb
da
(2)
T.b
.ly
rop
ha
nes
(3)
T.p
au
cim
acu
latu
s(4
)T.
b.b
iscu
tatu
s(5
)T.
b.qu
ad
rup
lex
(6)
male
fem
ale
male
fem
ale
male
fem
ale
male
fem
ale
male
fem
ale
male
fem
ale
SRW
22
2.2
60
.52
2.7
60
.72
1.8
60
.72
2.4
60
.72
2.2
61
.02
2.7
60
.72
3.8
60
.82
4.6
60
.92
4.9
60
.82
6.2
61
.02
4.9
61
.12
6.1
60
.92
1–
24
22
–2
51
9–
25
21
–2
51
9–
24
21
–2
42
2–
25
23
–2
72
2–
26
24
–2
82
3–
27
24
–2
73
92
11
12
59
29
22
24
16
38
19
30
13
1,4
,6
1,4
,6
1,4
,6
1,4
,6
1,4
,6
1,4
,6
1,2
,3
,5
1,2
,3
,5
2,3
,5
,6
2,3
,5
,6
2,3
,5
,6
2,3
,5
,6
SRW
31
5.2
60
.51
5.7
60
.61
5.0
60
.51
5.5
60
.51
5.4
60
.61
6.0
60
.71
6.0
60
.61
6.1
60
.41
6.7
60
.41
7.2
60
.81
7.3
60
.71
7.5
60
.81
4–
17
15
–1
71
4–
17
14
–1
61
4–
16
15
–1
71
5–
18
15
–1
71
6–
17
15
–1
81
5–
18
16
–1
94
12
11
12
59
29
22
24
16
38
19
30
13
1,4
,6
1,4
1,3
,4
,6
1,4
1,2
,4
,6
1,4
1,2
,3
,4
,5
1,4
2,3
,4
,5
,6
2,3
,5
,6
1,2
,3
,5
,6
2,3
,5
,6
SUPRAL
9.1
60
.69
.06
0.4
8.9
60
.48
.96
0.5
8.7
60
.58
.66
0.5
8.8
60
.48
.96
0.2
9.0
60
.28
.96
0.2
96
09
.16
0.3
8–
11
8–
10
8–
10
8–
11
8–
98
–9
8–
98
–9
8–
98
–9
99
–1
03
92
11
03
60
29
22
22
16
36
19
30
12
3N
.S.
N.S.
N.S.
5N
.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
SUPRAR
9.1
60
.48
.96
0.4
8.9
60
.59
.06
0.5
8.6
60
.58
.66
0.6
8.9
60
.59
.06
09
.16
0.2
9.0
60
.59
60
.29
60
8–
10
8–
10
8–
10
8–
10
8–
98
–1
08
–1
09
9–
10
8–
10
8–
99
38
21
10
06
02
92
22
21
63
51
93
01
23
N.S.
N.S.
N.S.
1,5
N.S.
N.S.
N.S.
3N
.S.
N.S.
N.S.
INFR
AL
11
.76
0.8
11
.96
0.8
11
.86
0.9
12
.36
0.8
11
.36
0.7
11
.66
0.8
12
.56
0.6
12
.66
0.7
12
.56
0.6
12
.76
0.9
12
.36
0.8
12
.96
0.5
10
–1
31
1–
13
9–
13
11
–1
41
0–
13
10
–1
31
1–
13
12
–1
41
1–
14
11
–1
51
0–
14
12
–1
43
82
11
03
57
29
22
20
14
33
19
30
12
1N
.S.
1N
.S.
1,4
,6
1,4
3N
.S.
2,3
,5
33
3IN
FRAR
11
.66
0.7
11
.96
0.8
11
.96
0.9
12
.16
0.8
11
.46
0.6
11
.66
0.9
12
.36
0.9
12
.66
0.7
12
.66
0.6
12
.76
0.8
12
.46
0.8
12
.46
0.9
10
–1
31
1–
13
9–
14
11
–1
41
0–
12
10
–1
31
0–
14
11
–1
41
2–
14
11
–1
41
1–
14
11
–1
43
62
19
95
62
92
12
11
53
41
83
01
21
,4
,6
N.S.
1N
.S.
1,4
,6
15
N.S.
2,3
,5
33
,5
N.S.
PREO
CL
3.0
60
.23
.06
0.2
3.0
60
.23
.06
0.2
3.0
60
2.9
60
.43
.06
0.2
3.1
60
.23
.06
03
.06
03
60
.23
60
2–
43
–4
2–
42
–4
31
–3
3–
43
–4
33
2–
33
39
21
10
45
82
92
22
21
63
61
93
01
3N
.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
PREO
CR
3.0
60
.23
.16
0.3
3.0
60
.23
.06
0.3
2.9
60
.32
.96
0.3
3.0
60
.23
.06
03
.06
03
.06
03
60
3.1
60
.33
–4
3–
42
–4
2–
42
–3
2–
33
–4
33
33
3–
43
92
11
01
58
29
22
23
16
35
19
30
12
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
N.S.
PO
STL
3.0
60
.22
.86
0.4
3.0
60
.23
.16
0.3
3.0
60
3.0
60
.33
.06
03
.16
0.3
2.9
46
0.2
42
.96
0.2
2.8
60
.42
.96
0.5
2–
32
–3
2–
42
–4
32
–4
33
–4
2–
32
–3
2–
32
–4
39
21
10
55
82
92
22
31
63
51
93
01
3N
.S.
N.S.
4N
.S.
4N
.S.
N.S.
N.S.
N.S.
N.S.
2,3
N.S.
PO
STR
3.0
60
.22
.86
0.4
3.0
60
.23
.16
0.4
3.1
60
.33
.06
0.2
3.1
60
.33
.06
02
.96
0.2
2.9
60
.22
.96
0.3
3.1
60
.33
–4
2–
32
–3
2–
43
–4
3–
43
–4
32
–3
2–
32
–3
3–
43
92
11
02
57
29
22
23
16
36
19
30
12
N.S.
N.S.
N.S.
N.S.
4N
.S.
N.S.
N.S.
N.S.
N.S.
3,6
N.S.
Ap
pen
dix
2
Co
ntin
ued.
Devitt et al.—Variation in Trimorphodon biscutatus 385
T.vi
lkin
son
ii(1
)T.
b.la
mb
da
(2)
T.b
.ly
rop
ha
nes
(3)
T.p
au
cim
acu
latu
s(4
)T.
b.b
iscu
tatu
s(5
)T.
b.qu
ad
rup
lex
(6)
male
fem
ale
male
fem
ale
male
fem
ale
male
fem
ale
male
fem
ale
male
fem
ale
LORL
2.2
60
.43
.06
0.2
2.5
60
.62
.96
0.4
2.7
60
.53
.06
0.6
2.9
60
.33
.06
02
.86
0.5
2.8
60
.52
.76
0.5
2.8
60
.42
–3
2–
31
–3
2–
42
–4
2–
52
–3
31
–4
1–
32
–3
2–
33
92
11
05
58
29
22
21
16
36
18
29
13
1,3
,6
N.S.
N.S.
N.S.
5N
.S.
5N
.S.
5N
.S.
N.S.
N.S.
LORR
2.3
60
.42
.86
0.4
2.6
60
.62
.96
0.4
2.8
60
.53
.06
0.6
3.0
60
.33
.06
02
.96
0.4
2.8
60
.52
.86
0.4
2.9
60
.32
–3
2–
31
–4
2–
42
–4
2–
52
–4
31
–4
1–
32
–3
2–
33
92
11
03
58
29
22
21
16
36
18
29
13
1,2
,3
,4
,6
N.S.
1,5
,6
N.S.
5N
.S.
5N
.S.
2,5
N.S.
5N
.S.
BB
TH2
0.5
61
.62
1.9
63
.82
4.0
62
.62
3.8
62
.93
3.0
64
.13
4.9
66
.42
1.1
63
.02
1.3
62
.41
9.3
63
.41
9.7
62
.71
9.3
62
.01
9.2
62
.01
8–
24
17
–3
11
8–
31
19
–3
02
5–
42
23
–4
81
7–
31
18
–2
81
4–
30
16
–24
15
–2
41
7–
23
40
22
11
13
52
92
22
41
53
81
92
91
32
,3
31
,3
,4
,5
,6
31
,2
,4
,5
,6
1,4
,5
,6
2,3
32
,3
32
,3
3TB
TH1
0.2
61
.61
0.7
62
.91
2.1
62
.01
3.2
63
.41
4.9
62
.11
5.6
62
.61
1.8
61
.39
.96
1.4
11
.46
2.5
10
.36
1.8
12
.36
1.9
12
.16
2.0
18
–2
41
–1
68
–1
79
–2
61
1–
19
11
–2
29
–1
58
–1
35
–1
86
–1
38
–1
79
–1
64
02
21
08
58
29
20
23
15
38
16
27
12
2,4
33
,5
31
,2
,4
,5
,6
1,5
,6
33
33
3,5
N.S.
SWB
TH1
5.2
61
.85
.26
1.6
6.6
61
.46
.66
1.3
4.9
60
.95
.36
1.2
10
.46
2.6
11
.46
2.7
3.4
60
.93
.16
0.7
10
.26
2.0
10
.36
1.7
2–
93
–8
4–
10
5–
11
4–
73
–8
6–
16
6–
16
1–
62
–4
7–
15
8–
13
37
18
11
16
02
92
22
81
34
12
12
41
52
,4
,5
,6
4,5
,6
1,3
,4
,5
,6
3,4
,5
,6
2,4
,5
,6
2,4
,5
,6
1,2
,3
,5
1,2
,3
,5
1,2
,3
,4
,6
1,2
,3
,4
,6
1,2
,3
,5
1,2
,3
,5
SWB
TH2
6.9
61
.76
.56
1.6
5.2
61
.15
.36
1.1
4.5
61
.14
.56
1.3
10
.46
2.4
10
.26
2.5
2.9
60
.73
.26
0.7
8.3
61
.38
.76
1.1
4–
12
4–
93
–8
4–
92
–7
3–
75
–1
56
–1
42
–4
2–
46
–1
17
–1
03
81
91
10
60
29
22
29
13
41
21
24
15
2,3
,4
,5
,6
3,4
,5
1,4
,5
,6
4,5
,6
1,4
,5
,6
1,4
,6
1,2
,3
,5
,6
1,2
,3
,5
1,2
,3
,4
,6
1,2
,4
,6
1,2
,3
,4
,5
2,3
,5
SWB
TH3
7.2
62
.16
.86
2.3
4.2
61
.04
.46
0.9
3.6
61
.03
.66
0.9
8.7
62
.59
.26
2.5
2.3
60
.52
.56
0.5
6.0
60
.86
.86
1.5
4–
14
4–
13
0–
73
–7
2–
62
–5
5–
16
5–
15
2–
32
–3
5–
85
–1
13
81
91
12
60
29
22
29
13
41
21
24
15
2,3
,4
,5
,6
2,3
,4
,5
1,4
,5
,6
1,4
,5
,6
1,4
,5
,6
1,4
,6
1,2
,3
,5
,6
1,2
,3
,5
1,2
,3
,4
,6
1,2
,4
,6
1,2
,3
,4
,5
2,3
,5
LHTS
P1
0.6
60
.90
.86
0.9
1.7
60
.61
.66
0.6
1.1
61
.01
.06
0.4
2.5
60
.82
.26
0.7
0.6
60
.90
.46
0.6
1.4
61
.21
.46
1.1
0–
30
–2
0–
31
–3
0–
60
–2
1–
41
–3
0–
30
–2
0–
40
–3
36
18
11
06
02
82
22
81
34
02
11
81
12
,4
,6
41
,3
,4
,5
5,6
2,4
41
,2
,3
,5
,6
1,3
,5
,6
2,4
,6
2,4
1,4
,5
2,4
LHTS
P2
1.1
60
.71
.36
0.6
1.4
60
.51
.26
0.5
1.0
60
.71
.06
0.4
2.1
60
.91
.76
0.6
0.4
60
.60
.76
0.7
1.2
60
.71
.16
0.3
0–
20
–2
1–
20
–2
0–
40
–2
1–
41
–3
0–
20
–2
0–
31
–2
37
19
10
96
02
82
22
91
34
02
11
81
14
,5
N.S.
4,5
54
,5
41
,2
,3
,5
,6
3,5
1,2
,3
,4
,6
2,4
4,5
N.S.
LHTS
P3
1.0
60
.60
.96
0.7
1.1
60
.51
.26
0.4
0.9
60
.70
.96
0.5
1.6
60
.71
.56
0.7
0.2
60
.40
.36
0.5
0.6
60
.50
.76
0.5
0–
20
–3
0–
40
–2
0–
30
–2
1–
31
–3
0–
10
–1
0–
10
–1
37
19
11
06
02
82
22
91
34
02
11
41
14
,5
N.S.
4,5
,6
54
,5
N.S.
1,2
,3
,5
,6
51
,2
,3
,4
2,4
2,4
N.S.
Ap
pen
dix
2
Co
ntin
ued
.
386 Copeia 2008, No. 2
T.vi
lkin
son
ii(1
)T.
b.la
mb
da
(2)
T.b
.ly
rop
ha
nes
(3)
T.p
au
cim
acu
latu
s(4
)T.
b.b
iscu
tatu
s(5
)T.
b.qu
ad
rup
lex
(6)
male
fem
ale
male
fem
ale
male
fem
ale
male
fem
ale
male
fem
ale
male
fem
ale
BSQ
12
8.0
62
.38
.96
2.2
4.7
62
.04
.26
1.7
2.5
60
.52
.56
0.7
4.1
61
.33
.46
0.5
9.2
62
.68
.56
2.2
4.5
61
.23
.76
0.9
3–
14
4–
13
2–
13
2–
10
2–
31
–4
3–
93
–4
3–
14
5–
13
3–
83
–6
41
22
11
16
02
92
22
41
53
81
82
91
32
,3
,4
,6
2,4
,6
1,3
,5
1,5
1,2
,4
,5
1,5
1,5
1,5
2,3
,4
,6
2,3
,4
,6
1,3
,5
1,5
BSQ
78
7.9
62
.58
.36
1.8
4.3
61
.83
.96
1.5
2.3
60
.82
.36
1.0
4.0
61
.23
.66
0.8
7.3
62
.17
.76
2.3
3.8
60
.83
.76
0.9
3–
12
4–
12
2–
11
2–
91
–4
1–
62
–8
2–
53
–1
15
–1
32
–5
2–
54
12
21
11
60
29
22
24
15
38
19
29
13
2,3
,4
,6
2,4
,6
1,3
,5
1,5
1,2
,5
1,5
1,5
1,5
2,3
,4
,6
2,3
,4
,6
1,5
1,5
BSQ
13
14
6.7
62
.07
.66
1.4
4.1
61
.33
.96
1.6
2.5
60
.62
.46
0.7
3.7
60
.94
.06
0.8
5.9
61
.95
.96
1.3
3.4
60
.63
.16
0.8
3–
11
5–
10
2–
92
–9
2–
41
–4
2–
53
–6
3–
12
3–
92
–4
2–
44
12
11
11
60
29
22
24
15
38
19
28
13
2,3
,4
,6
2,3
,4
,6
1,3
,5
1,5
1,2
,5
1,5
1,5
1,5
2,3
,4
,6
2,3
,4
,6
1,5
1,5
BSQ
AN
6.9
61
.36
.16
1.8
3.8
61
.23
.96
1.9
2.8
60
.72
.76
1.0
3.4
61
.03
.56
0.6
5.5
61
.75
.36
2.0
2.4
60
.82
.56
0.7
5–
10
2–
82
–8
2–
12
1–
41
–4
2–
63
–5
2–
10
1–
81
–4
2–
41
72
21
12
60
29
22
24
15
37
18
29
13
1,2
,4
,6
3,4
1,3
,4
,5
N.S.
1,2
,5
1,5
1,5
N.S.
2,3
,4
31
,2
,5
5PB
DIV
*0
.56
0.5
0.7
60
.51
.06
0.2
0.9
60
.40
.96
0.4
1.0
60
1.2
61
.21
.46
0.8
0.6
60
.50
.66
0.7
1.0
60
1.0
60
0–
10
–1
0–
10
–3
0–
11
–1
0–
31
–3
0–
10
–2
1–
11
–1
37
21
10
75
12
91
91
41
03
61
32
61
22
,3
,4
,6
N.S.
1,5
N.S.
5N
.S.
1,5
N.S.
2,4
,6
61
,5
N.S.
SBD
IV*
06
00
.16
0.3
0.1
60
.30
.16
0.3
06
00
60
06
00
60
06
00
.16
0.3
1.0
60
.21
.06
00
–0
0–
10
–1
0–
10
–0
0–
00
–0
0–
00
–0
0–
10
–1
1–
14
12
11
12
53
29
21
15
10
36
16
29
13
44
44
44
44
44
1,2
,3
,5
,6
1,2
,3
,5
,6
LYRE*
1.8
61
.11
.16
0.6
3.6
60
.83
.46
1.0
3.1
60
.52
.96
0.6
3.7
60
.73
.86
0.4
3.9
60
.43
.86
0.4
3.8
60
.43
.46
0.7
1–
40
–3
0–
41
–4
2–
42
–4
1–
43
–4
3–
43
–4
3–
42
–4
34
20
10
24
92
92
02
11
03
51
62
91
31
,2
,3
,4
,6
1,2
,3
,4
,6
3,5
51
,2
,4
,5
55
53
,5
53
,5
5H
SBTH
*2
.76
0.9
2.0
61
.21
.46
1.7
1.2
61
.00
60
06
00
60
06
02
.86
1.3
3.3
60
.71
.46
1.3
0.8
61
.50
–4
0–
30
–4
0–
30
–0
0–
00
–0
0–
00
–4
2–
40
–4
0–
44
02
11
10
53
29
21
17
10
36
16
28
13
1,2
,3
,4
,6
2,3
,4
,6
1,3
,5
,6
1,3
,5
1,2
,4
,5
1,2
,5
1,2
,4
,5
1,5
2,3
,4
,5
,6
2,3
,4
,6
1,3
,5
,6
1,5
AN
AL{
0%
0%
0%
0%
62
%7
9%
4%
0%
0%
7%
0%
0%
37
19
10
24
82
91
92
31
13
51
42
91
23
33
31
,2
,4
,5
,6
1,2
,4
,5
,6
33
33
33
App
end
ix2
Co
ntin
ued
.
Devitt et al.—Variation in Trimorphodon biscutatus 387