Molecular Phylogenetics and Evolution Vol. 18, No. 3, March, pp. 327 334, 2001 doi:10.1006/mpev.2000.0881, available online at http://www.idealibrary.com on The Coachella Valley Fringe-Toed Lizard (Uma inornata): Genetic Diversity and Phylogenetic Relationships of an Endangered Species Tanya L. Trépanier and Robert W. Murphy 1 Centre for Biodiversity and Conservation Biology, Royal Ontario Museum, 100 Queen s Park, Toronto, Ontario, Canada M5S 2C6, and Department of Zoology, University of Toronto, 25 Harbord Street, Toronto, Ontario, Canada M5S 3G5 Received November 2, 1999; revised October 2, 2000; published online January 23, 2001 A phylogeny was reconstructed for 23 populations of fringe-toed lizards (genus Uma) from the three most northern species of the genus, including the Mojave fringe-toed lizard U. scoparia, the Colorado Desert fringe-toed lizard U. notata, and the endangered Coachella Valley fringe-toed lizard U. inornata. The outgroup taxa were the zebra-tailed lizard, Callisaurus draconoides; the lesser earless lizard, Holbrookia maculata; and the greater earless lizard, Cophosaurus texanus. Evaluation of 1630 combined nucleotide sequence from the mitochondrial genes ATPase 6 and cytochrome b yielded 10 most parsimonious trees. Reweighting the characters using the rescaled consistency index eliminated eight of these trees. The remaining two trees differ only in the placement of two individuals from the Superstition Mountains which either formed a monophlyetic unit or grouped with one individual from the Anza-Borrego population. The preferred phylogeny, one more consistent with geography, had two primary clades: one consisting of U. scoparia and the other placing U. inornata inside the clade containing U. notata. Uma inornata was most closely related to nearby U. notata notata, as opposed to more distant U. notata rufopunctata. 2001 Academic Press INTRODUCTION 1 To whom reprint requests should be addressed. Fax: (416) 586-5553. E-mail: drbob@rom.on.ca Fringe-toed lizards, genus Uma, occur on North American desert sand dunes and ramps in the southwestern United States and central and northwestern Mexico. The genus contains five species: Uma inornata, U. notata, U. scoparia, U. exsul, and U. paraphygas, as well as the subspecies U. notata rufopunctata. Uma inornata is an endangered species, and U. scoparia and U. notata are listed as species of special concern in the State of California (Jennings and Hayes, 1994) due to their declining numbers and dwindling distributions. In Mexico, the fringe-toed lizard U. paraphygas is considered endemic and endangered and U. exsul is listed as being endemic and rare (Norma Oficial Mexicana, 1994). The systematic relationships within the genus have been investigated using anatomy (Schmidt, 1922; Heifetz, 1941; Smith, 1946; Schmidt and Bogert, 1947; Stebbins, 1954; Norris, 1958), allozymes (Adest, 1977; de Queiroz, 1992; Murphy and Doyle, 1998), behavior (Carpenter, 1963), and reproductive data (Mayhew, 1964). However, many questions remain unresolved. Although the various data sets are generally congruent in diagnosing at least three species, a considerable difference of opinion remains among biologists as to whether or not some forms are species or subspecies. The ambiguity arises within the three northernmost species, which are diagnosed on the basis of minor differences in pattern and color. Whereas some biologists consider the northernmost forms to be separate species (Heifetz, 1941; Smith, 1946; Schmidt and Bogert, 1947; Stebbins, 1954; Mayhew, 1964), Schmidt (1922) and Adest (1977) considered the three forms to be subspecies of U. notata, and yet others consider them to be two species and one subspecies (Norris, 1958; Carpenter, 1963). Although de Queiroz (1992) pursued an allozyme study of Uma, only one or two locations were sampled, which were assumed to be representative of each species. In addition, because U. inornata was listed as an endangered species, no tissue samples were available for his allozyme analysis. To date, the genetic relationships among populations of a single species have not been addressed. Because fringetoed lizards are restricted to isolated sand dunes, genetic data might differentiate among populations due to the absence of gene flow. Therefore, each population has the potential to be diagnosable, at least using moderately variable markers such as mitochondrial DNA sequences. Fringe-toed lizards occupy sand dunes and ramps in the Mojave Desert, the Colorado Desert extension of the Sonoran Desert, and the Chihuahuan Desert. Streams were largely responsible for the deposition of the sand and thus the historical connections among populations. As the streams dried at the end of the 327 1055-7903/01 $35.00 Copyright 2001 by Academic Press All rights of reproduction in any form reserved.
328 TRÉPANIER AND MURPHY TABLE 1 Sample Sizes and Localities of Sequenced Fringe-Toed Lizard Samples Species Locality Sample size Voucher a U. scoparia Dumont Dunes, Inyo County, CA 1 ROM 19840 Rice Dunes, San Bernadino County, CA 1 ROM 19848 Dale Dry Lake, San Bernadino County, CA 1 ROM 19851 Red Pass, Fort Irwin, San Bernadino County, CA 1 ROM 3406 U. notata Algodones, Imperial County, CA 5 ROM 19875-9 Anza-Borrego, San Diego County, CA 2 ROM 3412, 3414 Superstition Mountains, Imperial County, CA 2 ROM 3416-7 Salton Sea, Imperial County, CA 1 ROM 3415 North of San Felipe, Baja California (Norte), Mexico 2 ROM 34082-3 U. n. rufopunctata Mohawk, Yuma County, AZ 3 ROM 19893-5 Yuma, Yuma County, AZ 4 ROM 19861, 19863-5 Pinta Sands, Yuma County, AZ 5 ROM 19896-900 San Pedro, Sonora, Mexico 3 ROM 4276-8 ROM 4331, 4333, 4338-9, 4341 Puerto Peñasco, Sonora, Mexico 5 U. inornata SSW of Thousand Palms, Riverside County, CA 1 CAP 1702 Whitewater River Reserve, Riverside County, CA 1 CAP 1718 Coachella Valley Preserve, Riverside County, CA 1 CAP 1723 Willow Hole Preserve, Riverside County, CA 1 CAP 1735 North of Thousand Palms, Riverside County, CA 1 CAP 1745 Eastern Indio Hills, Riverside County, CA 1 CAP 1748 Windy Point, Riverside County, CA 1 CAP 1755 Coachella Valley Preserve N, Riverside County, CA 1 CAP 1768 Western Indio Hills, Riverside County, CA 1 CAP 1771 C. draconoides Ocotillo, Imperial County, CA 1 ROM 4146 C. texanus 15 km North of Lab Mapimi, Durango, Mexico 1 ROM 32048 H. maculata 15 km North of Lab Mapimi, Durango, Mexico 1 ROM 32049 a ROM, Royal Ontario Museum; CAP, Christopher A. Phillips. Pleistocene, the dune populations became isolated. In most cases, fringe-toed lizard distributions seem to be related to paths of pluvial drainage and, consequently, fringe-toed lizards occupy most aeolian sand concentrations in the Mojave Desert. The same process was probably responsible for the formation of Chihuahuan and Sonoran dunes (Norris, 1958), although isolation of the Mexican fringe-toed lizards may have occurred as early as the Late Tertiary (Morafka et al., 1992). Unsuitable habitat prevents dispersal among dunes. Therefore, fringe-toed lizards are essentially island inhabitants and, as a result, they are highly susceptible to factors that directly and indirectly affect their sand dune habitat. These factors include habitat destruction caused by the removal of sand, development and an increase in landfill sites, misuse of off-road vehicles, alterations in wind patterns for sand deposition due to urban development, and habitat loss from the conversion of land for agriculture and golf courses (see Turner et al., 1984; U.S. Fish and Wildlife Service, 1984; Jennings and Hayes, 1994; Barrows, 1996). Mitochondrial DNA gene sequence data from multiple populations of Mojave, Colorado Desert, Sonora, and Coachella Valley fringe-toed lizards were used to evaluate the genealogical history among these populations and taxa. The phylogeny was used to evaluate the validity of the current taxonomy of the three northernmost species of fringe-toed lizards. Specifically we investigated the validity of U. inornata as a species, which has been considered by many to be a subspecies of U. notata. This paper is not intended to provide a phylogeny of the genus. Rather it evaluates the validity of the Coachella Valley fringe-toed lizard as a species as defined by the evolutionary species concept (Wiley, 1978; Wiley and Mayden, 2000) using the phylogenetic species concept (Nixon and Wheeler, 1990) as the operational criterion. MATERIALS AND METHODS Specimen Collection Forty-five specimens of fringe-toed lizards (genus Uma) were sampled from 23 localities (Table 1). The outgroup included the zebra-tailed lizard, Callisaurus draconoides; the lesser earless lizard, Holbrookia maculata; and the greater earless lizard, Cophosaurus texanus, with U. scoparia as the primary outgroup species. Studies by Norris (1958), de Querioz (1989), and Carpenter (1963) revealed that the most appropriate outgroup for U. notata, U. notata rufopunctata, and U. inornata is U. scoparia. The monophyly of the three
COACHELLA VALLEY FRINGE-TOED LIZARD 329 northernmost forms is well supported. Our paper does not address the phylogeny of the fringe-toed lizards, but rather investigates the genetic diversity and phylogenetic placement of the endangered Coachella Valley fringe-toed lizard. Thus, the closely related Mexican fringe-toed lizards, U. paraphygas and U. exsul, were not used as additional outgroup taxa. Lizards were captured by hand or by using nooses. For most specimens, tail tips were removed and preserved in 95% ethanol, and the animals were released on site. Some specimens were euthanized and subsequently preserved in 95% ethanol, following accepted Animal Welfare Protocols. All blood samples from U. inornata were collected from an infraorbital sinus, and all animals were released at the point of capture. DNA Extraction, Amplification, Sequencing, and Sequence Alignment Standard phenol-extraction methods were used to extract DNA from tail muscle, following procedures from Palumbi (1996) and Hillis et al. (1996). The mitochondrial gene ATPase 6 was amplified and sequenced using the primers 5 atg aac cta agc ttc ttc gac caa tt 3 (Haddrath, personal communication) and 5 acg aat acg tag gct tgg att a 3 (Fu, 1999). Amplification and sequencing of cytochrome b was performed using three primers flanking this region. The primers 5 cca tcc aac atc tca gca tga tga aa 3 (Kocher et al., 1989), 5 tga gga caa ata tcc ttc tga gg 3 (Fu, 1999), and 5 gtc ttc agt ttt tgg ttt aca aga c 3 (Kocher et al., 1989) were used. Double-stranded DNA was prepared for sequencing using the polymerase chain reaction (PCR, Saiki et al., 1988). Reactions of 25 l were set up using 1 l of genomic DNA (0.05 g/ l), 1 l of each primer (10 mm), 0.08 l of a dntp s mix (containing 10 mm of each datp, dgtp, dttp, dctp) (Perkin Elmer Cetus), 0.2 l of Amplitaq polymerase (Perkin Elmer Cetus), and 2.5 l of a reaction buffer (containing 67 mm Tris, ph 8.8, 2 mm MgCl) (Sigma). The 0.5-ml reaction tubes were placed in a thermal cycler. The parameters varied as follows: Initial cycle 94 C for 2 min; 92 C for 30 s, 42 50 C for 45 s, 72 C for 45 s (36 38 cycles); and final extension file at 72 C for 5 min. Double-stranded amplified product was electrophoresed in a 2% agarose (BDH) gel containing ethidium bromide (1 g/ml; BDH) for 15 30 min at 100 120 V in a Tris acetate buffer. Successful amplifications were excised from the gel and purified by GeneClean (Bio/Can Scientific). Protocols followed directly the manufacturer s instructions. The PCR product from the ATPase 6 mtdna gene was a 600-bp fragment. The other PCR product was a 1100-bp fragment from the cytochrome b mtdna gene, which included a portion of the Thr t- RNA. Purified double-stranded DNA was labeled using 33 P-labeled terminator cycle sequencing kit (Amersham). The sequenced DNA was loaded into a 40 cm, 5% long ranger (J. T. Baker) polyacrylamide gel in a 0.6% TBE buffer, and electrophoresed at 65 V for 2.5 5.5 h. The gels were dried on filter paper and exposed for 20 30 h on Kodak XAR autoradiograph film. Sequences were aligned by eye using BioEdit and then exported as a PAUP/Nexus file for phylogenetic analyses. Transition and transversion ratios were calculated for each gene separately, using the average number of events across all most parsimonious reconstructions in MacClade 3.04 (Maddison and Maddison, 1992). Phylogenetic Analysis The two genes were analyzed both separately and combined. Unweighted maximum parsimony analyses were performed as implemented in PAUP* 4.02b (Swofford, 1998). Most parsimonious trees (MPTs) were obtained by employing the heuristic tree search algorithm. All PAUP* 4.02b analyses used random addition sequence, 50 200 replicates while retaining minimal trees only, tree bisection reconnection branch swapping with steepest descent, and collapsed zero length branches. All multistate characters were evaluated as nonadditive (unordered). All characters were initially evaluated as unweighted. Nodal consistency was assessed using (1) functional ingroup outgroup evaluations (FIG/FOG; Watrous and Wheeler, 1981; Murphy et al., 1983; Fu and Murphy, 1997), (2) nodal-specific permutation tail probabilities (n-ptps) for character covariation (Fu and Murphy, 1999), and (3) bootstrap proportions (BSP; Felsenstein, 1985). FIG/FOG, BSP, and n-ptp trials were performed in PAUP* 4.02b. BSP and n-ptp evaluations involved 1000 random replicates of the data. RESULTS The combined data consisted of 1630 nucleotide sites from the mitochondrial genes ATPase 6 (599 bp) and cytochrome b (1031 bp). Of these, 232 sites (14.2%) were variable, with 176 (10.8%) potentially phylogenetically informative. For the ATPase 6 gene, 73 (12.2%) of the 599 sites were variable, with 52 (8.7%) potentially phylogenetically informative. The transition to transversion ratio was averaged across all MPTs and found to be 1.7:1. For the cytochrome b gene, 159 (15.4%) of the 1031 sites were variable, with 124 (12.0%) potentially phylogenetically informative. The transition to transversion ratio was averaged across all MPTs and found to be 3.5:1. Among the 232 variable sites, most changes were silent in that they did not code for alternative amino acids, whereas the nucleotide substitutions at 38 positions resulted in the coding of alternative amino acids. Of the alternative amino acids, 19 were found in the ATPase 6 gene, and 19 were found in the cytochrome b gene. All sequences were deposited in GenBank (ATPase 6 AF301913 AF301960; cytochrome b AF301961 AF302008).
330 TRÉPANIER AND MURPHY FIG. 1. Strict consensus trees for the ATPase 6 (a) and cytochrome b genes (b). Intraspecific and Interspecific Variation Sequence site variation within U. scoparia and U. notata ranged from 2.37 0.43% and 3.57 0.00%, respectively. Within U. inornata, sequence variation among the seven populations was 0.25 0.00%. U. scoparia differed from U. notata at 7.48 8.41% of the sites sequenced. Similarly, U. scoparia differed from U. inornata at 7.90 8.40% of sites. In contrast, U. inornata differed from U. notata at only 0.62 3.32% of sequenced sites. Phylogeny Analysis of the ATPase 6 data resulted in 75 MPTs with a length of 200 steps (CI 0.68, RI 0.83). A strict consensus tree for the ATPase 6 gene is shown in Fig. 1a. The cytochrome b data resulted in 4 MPTs with a tree length of 416 steps (CI 0.61, RI 0.81). A strict consensus tree for the cytochrome b gene is shown in Fig. 1b. The two gene trees are completely congruent in resolved relationships among the populations, with the exception of two nodes (see Fig. 1). In both data sets, U. scoparia and U. inornata are phylogenetically diagnosable as species, yet U. notata is paraphyletic and nonexclusive with respect to U. inornata. In addition, the subspecies U. notata rufopuntata is not monophyletic. An unweighted analysis of the combined data resulted in 10 MPTs with a length of 626 steps (CI 0.62, RI 0.81). The characters were reweighted using the rescaled consistency index and reanalyzed using the heuristic search routine. The second repetition obtained tree length stability and resulted in two trees, both of which were among the 10 original MPTs. These two trees differed only in the placement of two individuals from the Superstition Mountains, which either formed a monophyletic unit or grouped with one individual from the Anza-Borrego population. The phylogeny that places the two individuals from the Supersti-
COACHELLA VALLEY FRINGE-TOED LIZARD 331 FIG. 2. The preferred phylogenetic hypothesis for Uma using the ATPase 6 and cytochrome b combined data. Numbers on the terminal branches of the cladogram correspond to map locations. Vertical bars indicate membership in any of the three recognized species or subspecies. Bootstrap proportions greater than 70% are indicated on branches. Thick horizontal bars indicate nodes supported by highly significant character covariance (P 0.001), whereas nodes by dubious character covariance (0.050 P 0.001) are depicted by open horizontal bars. tion Mountains as a monophyletic unit forms our preferred tree (Fig. 2). As with the separate gene analyses, U. scoparia and U. inornata are phylogenetically diagnosable as species (monophyletic and exclusive), yet U. notata is paraphyletic with respect to U. inornata. In addition, U. notata rufopunctata from the Mohawk, Pinta Sands, Yuma, San Pedro, and Puerto Peñasco localities is paraphyletic with respect to U. notata (Fig. 2). Nodal Stability Nodal stability evaluations were performed on the combined data set. Nodes having BSPs of 70% or higher are shown in Fig. 2. Low BSPs may result from a small number of unambiguous synapomorphies and therefore do not indicate a lack of confidence in the node (Felsenstein, 1985). N-PTPs for locating significant character covariation (Fu and Murphy, 1999) were calculated. Nodes supported by highly significant character covariance (P 0.001) are depicted in Fig. 2 by thick horizontal bars, whereas nodes with dubious character covariance (0.050 P 0.001) are depicted by open horizontal bars. DISCUSSION Intraspecific and Interspecific Variation Genetic variation within U. scoparia and U. notata was much greater than within U. inornata. Uma inornata has a much smaller distribution than the other species, and today its range is much smaller than in the recent past (England and Nelson, 1976; U.S. Department of the Interior, 1980). Indeed, the extent of variation within U. inornata is equivalent to that observed within a single population of U. scoparia or U. notata. The homogeneity within U. inornata likely reflects bottlenecking and continued loss of variability is expected given ongoing destruction and degradation of the sand dune habitats.
332 TRÉPANIER AND MURPHY Levels of variation within U. notata are not substantially greater than the level of variation within U. scoparia. Phylogeny At a species level, the relationships resolved among our samples (Fig. 2) are consistent with most previous hypotheses of relationships for the northernmost Uma. Based on anatomical characters, Norris (1958) and de Querioz (1989) concluded that U. notata and U. inornata were more closely related to each other than to U. scoparia. Carpenter (1963) reached the same conclusion in his analyses of display behavior. These results differ from those of Mayhew (1964) who evaluated body and cloacal temperatures, testis volume, and male breeding season. He concluded that U. inornata and U. scoparia were more similar. Similarly, Zalusky et al. (1980) studied 21 anatomical characters and concluded that U. scoparia and U. inornata are phenetically more similar to each other than either is to U. notata. In both cases greater similarity is likely attributed to symplesiomorphies and not synapomorphies. Recognition of U. inornata has been contentious because the species has been defined primarily on the basis of color pattern (Cope, 1895; Heifetz, 1941). Adest (1977) evaluated allozyme genetic distances and concluded that only three species groups could be defined within the genus, one of which combined U. scoparia, U. notata, and U. inornata. Although dorsal color and presence of a ventrolateral blotch unite U. notata and U. scoparia, Adest (1978) rejected these characters; ventrolateral pigmentation, including the number, size, and color of bars or spots, varies within and among populations of U. notata and U. scoparia, and bars or spots are present in some individuals of U. inornata. Such variation seems to be common among Uma. Morafka et al. (1992) reported that the width and number of ventrolateral blotches vary within and among populations and species of Mexican fringe-toed lizards, as well as among the left and right sides of a single lizard. Norris (1958) concluded that dorsal color was a response to sand color. The discrepancies among the evaluations derive from similarity being equated with relatedness, which is an unjustified assumption. Our cladistic evaluation not only revealed unambiguous synapomorphies uniting U. notata and U. inornata, but the latter species branches off within the former resulting in a taxonomy that does not reflect genealogical history. Taxonomic Implications We used the evolutionary species concept (Wiley, 1978; Wiley and Mayden, 2000) to assess the taxonomic status of populations and clades using the phylogenetic species concept (Nixon and Wheeler, 1990) as the operational criterion. The taxonomy based on our genetic data required that (1) monophyletic groups were unambiguously diagnosed based on a number of synapomorphies and (2) no mixing of haplotypes (histories) occurred among taxa. Within this context of phylogenetic taxonomy, the current arrangement of three species and one subspecies from the northernmost clade must be redefined. Either a two-species classification or a five-species classification is required. In the two species classification, U. inornata would not be recognized as species. It forms a relatively recent group clustered among older distinct clades of Uma notata (Fig. 2); younger clades cannot be recognized unless older clades are also given species status. Uma notata Baird, 1859, has priority by date of publication over U. inornata Cope, 1895, and U. n. rufopunctata Cope, 1895. The Sonora fringe-toed lizard, U. n. rufopunctata (Cope, 1895) varies slightly from U. notata in blotch width and breeding color (Norris, 1958). It occupies dunes in the Yuma Desert from southwestern Arizona into extreme northwestern Mexico. We evaluated representatives of U. notata rufopunctata from the Mohawk Dunes, Pinta Sands and Yuma Dunes in Arizona, and in Mexico from San Pedro and Puerto Peñasco. Because samples from these dune systems do not cluster together (Figs. 1 and 2), there is no genealogical support for recognizing U. n. rufopunctata. Consequently, U. n. rufopunctata Cope, 1895, would be considered a junior synonym of U. notata Baird, 1859, if only one species is recognized. The second monophyletic species is U. scoparia. Alternatively, because the Coachella Valley fringetoed lizard U. inornata is genetically and geographically isolated and therefore on its own evolutionary trajectory, one could argue that the species should be retained. We believe this arrangement is preferable. Breeding coloration among sexes within the Coachella Valley fringe-toed lizard is distinct from the other northernmost forms of Uma. Females display a vivid lateral color when gravid, and males display only a faint wash of lateral color, which is the opposite of that observed in the other nearby fringe-toed lizard populations (Fisher, personal communication). The markings and color changes on the Coachella Valley fringe-toed lizards are far more distinct than in U. notata at Algodones Dunes (Barrows, personal communication). This morphological distinctiveness, and the fact that the Coachella Valley fringe-toed lizard is isolated and on its own evolutionary trajectory, provides justification for the recognition of this species. Thus, in order to maintain monophyly in species names, the Coachella Valley fringe-toed lizard maintains species status, and U. notata from California and Baja California, Mexico, should be retained as a species. Uma n. rufopunctata from all dunes except Mohawk Dunes also needs to be considered a full species, U. rufopunctata. Finally, the Mohawk Dunes population is a cryptic species that needs to be described, as no prior name is avail-
COACHELLA VALLEY FRINGE-TOED LIZARD 333 able. This description is in progress (T. L. Trépanier and R. W. Murphy, manuscript in preparation). The fifth monophyletic species in this phylogenetic taxonomy is U. scoparia. Conservation Measures Formerly, the smallest taxonomic unit of conservation was the species level. More recently, in the United States it has become possible to conserve single or multiple populations of species (U.S. Fish and Wildlife Service, 1994). In order to qualify for protection, populations must meet certain criteria. Specifically, they must form evolutionarily significant units. The term evolutionarily significant unit, first coined by Ryder (1986), is adopted in the Endangered Species Act. It identifies distinct population segments for conservation and management purposes that are unique as demonstrated by quantitative measures of genetic or morphological discontinuity (U.S. Department of the Interior, 1996). The protection of populations is crucial, in addition to species, because populations are responsible for high levels of diversity within species, and the potential exists for genetic, morphological, physiological, or behavioral diversification among populations. Because conservation efforts should be directed toward protecting genetic diversity, and not merely taxonomic names, protection of all the remaining populations of fringe-toed lizards should be enhanced, not withdrawn. The Coachella Valley fringe-toed lizard is morphologically and genetically distinct, whether considered a species or population segment, and therefore deserves special consideration. This species is isolated geographically from other populations, has genetic and morphological traits that are specific to it, and is on a diverging evolutionary trajectory. It is likely that other genetically distinctive population segments of fringetoed lizards occur within the genus and these deserve equal protection. ACKNOWLEDGMENTS D. J. Morafka, C. Phillips, C. Barrows, M. Fisher, R. MacCulloch, and M. Green provided comments and assistance. We give special thanks to A. Muth, C. Barrows, M. Fisher, C. Phillips, J. Losos, and P. Brylski for collection and donation of Coachella Valley fringe-toed lizard tissue samples (U. S. Fish and Wildlife Permit No. UCRBDC-4 issued to A. Muth). D. J. Morafka, S. Hillard, I. Girard, and M. Marolda assisted in collecting Mojave fringe-toed lizards (CDFG Permit No. 801068-03 issued to D. J. Morafka). D. Turner assisted in collecting and donated Colorado Desert and Sonora fringe-toed lizard tissues (Arizona Fish and Game Commission Permit No. HK081491). Collections in the Chihuahuan Desert were assisted by H. E. Gadsden, H. López-Corrujedo, J. Estrada-Rodríguez, U. Romero-Méndez, A. Orona E., and A. Herrera (Collecting Permit No. DOO.750-1268 issued to RWM). Collections in Baja California were assisted by D. J. Morafka, S. Papadakos-Morafka, and C. Huang (Collecting Permit No. DOO.02-2802, Export Permit No. 12140 issued to R.W.M.). We give special thanks to Oswaldo Suntillán L., Alfredo Zavala González, and Rogelio del Hoyo Ramírez of PROFEPA in Ensenada, Mexico. Specimens from Sonora were collected by D. J. Morafka, G. A. Adest, and G. Aguirre L. Comments that were provided by three anonymous reviews are sincerely appreciated. This research was supported by the Natural Sciences and Engineering Research Council (NSERC) of Canada Grant A3148 to R.W.M. and contracts from Fort Irwin NTC to R.W.M. and D. J. Morafka; we give special thanks to M. Quillman of Fort Irwin. This is contribution 230 from the Centre for Biodiversity and Conservation Biology of the Royal Ontario Museum. REFERENCES Adest, G. A. (1977). Genetic relationships in the genus Uma (Iguanidae). Copeia 1977: 47 52. Adest, G. A. (1978). The Relations of the Sand Lizards Uma, Callisaurus and Holbrookia (Sauria: Iguanidae): An Electrophoretic Study, Ph.D. dissertation. University of California, Los Angeles, CA. Barrows, C. (1996). An ecological model for the protection of a dune ecosystem. Conserv. Biol. 10(3): 888 891. Carpenter, C. C. (1963). Patterns of behavior in three forms of the fringe-toed lizards (Uma Iguanidae). Copeia 1963: 406 412. Cope, E. D. (1895). On the species of Uma and Xantusia. Am. Natur. 29: 938 939. de Queiroz, K. (1989). Morphological and Biochemical Evolution in the Sand Lizards, Ph.D. dissertation. University of California, Berkeley, CA. de Queiroz, K. (1992). Phylogenetic relationships and rates of allozyme evolution among the lineages of sceloporine sand lizards. Biol. J. Linnean Soc. 45: 333 362. England, A. S., and Nelson, S. G. (1976). Status of the Coachella Valley Fringe-Toed Lizard (Uma inornata), Inland Fisheries Admin. Report 77 1. Department of Game and Fish, State of California. Felsenstein, J. (1985). Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39: 783 791. Fu, J. (1999). Phylogeny of Lacertid Lizards (Squamata: Lacertidae) and the Evolution of Unisexuality, Ph.D. dissertation. University of Toronto, Toronto. Fu, J., and Murphy, R. W. (1997). Phylogeny of Chinese Oreolalax and the use of functional outgroups to select among multiple equally parsimonious trees. Asiatic Herpetol. Res. 7: 38 43. Fu, J., and Murphy, R. W. (1999). Discriminating and locating character covariance: An application of permutation tail probability (PTP) analyses. Syst. Biol. 48(2): 380 395. Heifetz, W. (1941). A review of the lizards of the genus Uma. Copeia 1941: 99 111. Hillis, D. M., Mable, B. K., and Moritz, C. (1996). Applications of molecular systematics. In Molecular Systematics (D. M. Hillis, C. Moritz, and B. K. Mable, Eds.), 2 nd ed., pp. 515 543. Sinauer, Sunderland, MA. Jennings, M. R., and Hayes, M. P. (1994). Amphibian and Reptile Species of Special Concern in California, Final Report. California Department of Fish and Game, Rancho Cordova, CA. Kocher, T. D., Thomas, W. K., Meyer, A., Edwards, S. V., Paabo, S., Villablanca, F. X., and Wilson., A. C. (1989). Dynamics of mitochondrial DNA evolution in animals: Amplification and sequencing with conserved primers. Proc. Natl. Acad. Sci. USA 86: 6196 6200. Maddison, W., and Maddison, D. (1992). MacClade: Analysis of Phylogeny and Character Evolution, Version 3.05. Sinauer, Sunderland, MA. Mayhew, W. W. (1964). Taxonomic status of California populations of the lizard genus Uma. Herptologica 20: 170 183. Morafka, D. J., Adest, G. A., Reyes, L. M., Aguirre, L. G., and
334 TRÉPANIER AND MURPHY Lieberman, S. S. (1992). Differentiation of North American deserts: A phylogenetic evaluation of a vicariance model. In Biogeography of Mesoamerica (Darwin and Weldon, Eds.), Tulane Studies in Zoology and Botany, Supplementary Publication No. 1, pp. 195 226. Tulane University, New Orleans, LA. Murphy, R. W., Cooper, W. E., and Richardson, C. W. (1983). Phylogenetic relationships of the North American five-lined skinks, genus Eumeces (Sauria: Scincidae). Herpetologica 39: 200 211. Murphy, R. W., and Doyle, K. D. (1998). Phylophenetics: Frequencies and polymorphic characters in genealogical estimation. Syst. Biol. 47: 737 761. Nixon, K. C., and Wheeler, Q. D. (1990). An amplification of the phylogenetic species concept. Cladistics 6: 211 223. Norma Oficial Mexicana (1994). NOM-059-ECOL-1994, que determina las especies y subespecies de floras y fauna silvestres terrestres y acuáticas en peligro de extinción, amenazadas, raras y las sujetas a protección especial, y que establece especificaciones para su protección, Diario Oficial de la Federación, Monday May 16, 1994. Norris, K. S. (1958). The evolution and systematics of the iguanid genus Uma and its relation to the evolution of other North American desert reptiles. Bull. Am. Mus. Nat. Hist. 114: 251 326. Palumbi, S. R. (1996). Nucletic acids II: The polymerase chain reaction. In Molecular Systematics (D. M. Hillis, C. Moritz, and B. K. Mable, Eds.), 2nd ed., pp. 205 247. Sinauer, Sunderland, MA. Ryder, O. A. (1986). Species conservation and systematics: The dilemma of subspecies. Trends Ecol. Evol. 1: 9 10. Saiki, R. K., Gelfand, D. H., Stoeffel, S., Scharf, S. J., Higuchi, R., Horn, G. T., Mullis, K. B., and Erlich, H. A. (1988). Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239: 487 491. Schmidt, K. P. (1922). The amphibians and reptiles of lower California and the neighboring islands. Bull. Am. Mus. Nat. Hist. 46: 607 707. Schmidt, K. P., and Bogert, C. M. (1947). A new fringe-footed sand lizard from Coahuilla, Mexico. Am. Mus. Novitates 1339: 1 9. Smith, H. M. (1946). Handbook of Lizards. Comstock, Ithaca, NY. Stebbins, R. C. (1954). Amphibians and Reptiles of Western North America. McGraw-Hill, New York. Swofford, D. L. (1998). PAUP* Version 4.02b: Phylogenetic Analysis Using Parsimony. Sinauer, Sunderland, MA. Turner, F. B., Weaver, D. C., and Rorabaugh, J. C. (1984). Effects of reduction in windblown sand on the abundance of the fringe-toed lizard (Uma inornata) in the Coachella Valley, California. Copeia 1984: 370 378. U.S. Department of the Interior (1980). Endangered and threatened wildlife and plants: Listing as threatened with critical habitat for the Coachella Valley fringe-toed lizard. Fed. Regist. 45: 63812 63820. U.S. Department of the Interior. (1996). Policy regarding the recognition of distinct vertebrate population. Fed. Regist. 61: 4722 4725. U.S. Fish and Wildlife Service. (1984). Coachella Valley Fringe- Toed Lizard Recovery Plan. U.S. Fish and Wildlife Service, Portland, OR. Watrous, L. E., and Wheeler, Q. D. (1981). The outgroup comparison method of character analysis. Syst. Zool. 30: 1 11. Wiley, E. O. (1978). The evolutionary species concept reconsidered. Syst. Zool. 27: 17 26. Wiley, E. O., and Mayden, R. L. (2000). The evolutionary species concept. In Species Concepts and Phylogenetic Theory: A Debate (Q. D. Wheeler and R. Meier, Eds.), pp. 70 89. Columbia Univ. Press, New York. Zalusky, S. B., Gaudin, A. J., and Swanson, J. R. (1980). A comparative study of cranial osteology in the North American sand lizards, genus Uma (Reptilia: Iguanidae). Copeia 1980: 296 310.