University of Kansas Publications MUSEUM OF NATURAL HISTORY The University of Kansas Publications, Museum of Natural History, beginning with volume 1 in 1946, was discontinued with volume 20 in 1971. Shorter research papers formerly pub- lished in the above series are now published as Occasional Papers, Museum of Natural History. The Miscellaneous Publica- tions, Museum of Natural History, began with number 1 in 1946. Longer research papers are pubhshed in that series. Monographs of the Museum of Natural History were initiated in 1970. All manuscripts are subject to critical review by intra- and extra-mural specialists; final acceptance is at the discretion of the publications committee. Institutional libraries interested in exchanging publications may obtain the Occasional Papers and Miscellaneous Publica- tions, by addressing the Exchange Librarian, The University of Kansas Library, Lawrence, Kansas 66045. Individuals may pur- chase separate numbers of all series. Prices may be obtained upon request addressed to Publications Secretary, Museum of Natural History, The University of Kansas, Lawrence, Kansas 66045. Editor: Linda Trueb PRINTED BY UNIVERSITY OF KANSAS PRINTING SERVICE LAWRENCE, KANSAS C M A I , ^ MUS. COM p. ZOOL. b " N/\ ' ux\AJrence. library OCCASIONAL PAPERS ^^^ 1 5 1977 HARVARD of the UNIVERSITY MUSEUM OF NATURAL HISTORY The University of Kansas Lawrence, Kansas NUMBER 64, PAGES 1-46 APRIL 1, 1977 PHYLOGENY OF PLEURODEMA (ANURA: LEPTO- DACTYLIDAE ) : A BIOGEOGRAPHIC MODEL By William E. Duellman^ and Alberto Veloso M." Modern biogeographic thinking is undergoing drastic modifi- cations with the reassessment of Croizat's (1964) conceptual con- tributions by Brundin (1966), Croizat, Nelson, and Rosen (1974), Ball (1975), and Rosen (1975). Recent studies on frogs within the Neotropical Realm (Duellman, 1972; Duellman and Cmmp, 1974; Heyer, 1973) have provided additional support for the importance of climatic fluctuation and concomitant ecological modifications in the patterns of distribution and differentiation of organisms in the American tropics as postulated by Haffer (1969), Vanzolini and WilHams (1970) and Miiller (1973). With the exception of Miiller (1973), who fomiulated a series of dispersal centers throughout South America, the biogeographic concerns have been primarily with the tropical regions, to the neglect of temperate South America. Southern South America has an endemic and evolutionarily important frog fauna, consisting primarily of telmatobiine lepto- dactylids. The major exception is the leptodactyline genus Pleuro- dema containing 14 species, only two of which occur in the tropics. Pleurodema bufonina is the southernmost frog, reaching the Straits of Magellan at 53° S Lat.; two species range northward in the Andes to Peru, and one extends to Panama and the Carribbean coast of South America (13° N Lat.). Pleurodema are small (< 60mm), 1 Curator, Division of Herpetology, Museum of Natural History, and Pro- fessor, Department of Systematics and Ecology, The University of Kansas, Lawrence, Kansas 66045, U.S.A. - Associate in Herpetology, Museum of Natural History, The University of Kansas, and Associate Professor, Departmento de Biologfa Celular y Genetica, Universidad de Chile, Casilla 6556, Santiago, Chile. 2 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY toad-like anuraiis; many species are fossorial. These small frogs exhibit a diversity of reproductive behavior and combinations of moiphological traits that provide ideal material for a phylogenetic study. The data for the present paper have been derived from our studies on the morphology of adults and tadpoles, mating calls, breeding behavior, and karyotypes, and from the extensive liter- ature on the genus. These data fonn the basis for a phylogenetic construct that, in combination with infomiation on past and present climatic and vegetation patterns in southern South America, provide the basis for the biogeographic model. Acknowledgments Our field studies on FJeurodema were accomplished independ- ently. Veloso is indebted to Mario Penna for assistance in recording calls and to Nelson Diaz for measuring frogs. Duellman's field work in South America was supported by a grant (No. 1304) from the National Geographic Society. In the field he was ably assisted by Dana K. Duellman, John E. Simmons, and Linda Trueb. Stephan Halloy also assisted in collecting specimens in Argentina and Chile, and Raymond F. Laurent kindly guided us to a locality for P. tucumana. Through the courtesy of Nelly Carillo de Espinoza, Duellman was able to examine specimens in the Museo de Historia Natrn-al "Javier Prado" in Lima. The research was completed at The University of Kansas and supported by a grant (No. 3272) from the General Research Fund of the Graduate School, The University of Kansas. Veloso's travel to Kansas was made possible by the Programa Multinacional de Genetica, OEA. We are grateful to John E. Simmons for his laboratory assistance; Jose M. Cei for data on certain Argentinian specimens; W. Ronald Heyer, John D. Lynch, Jaime Pefaur, and Edward O. Wiley for helpful suggestions; and Linda Trueb for executing the illustrations. '& Materials and Methods We have relied on the literatiu-e for data on P. hihroni, guatjopae, and kriegi; we have had no field experience with these species or with P. diplolistris and nelmlosa. Of the other 10 species, we have examined the large collections in the Museum of Natural History at The University of Kansas (KU) and in the Museo Javier Prado in Lima, Peru (MJP). Tape recordings were made on Uher 4000-L recorders at a speed of 19 cm /sec and were analyzed by means of a Sonagraph Model 7029 ( Kay Electric Co. ) . Chromosomes were obtained from squashes of corneal epithelium in the manner described by Bogart (1970). Measurements of chromosomes were made from projected PHYLOGENY OF PLEURODEMA 3 negatives by means of a map measurer. Each arm of both homol- ogous chromosomes was measured in order to construct the idio- grams. Nomenclature of centromeric position is that proposed by Levan, Fredga, and Sandberg (1964). HISTORICAL resume; Although frogs of the genus Pleurodema have been recognized as a distinct genus since 1838, when Tschudi proposed the generic name for the Uruguayan species, P. hihroni, there has been great taxonomic confusion at the species level. Throughout most of the literature on the Chilean and Uruguayan species, they have been known as P. hibroni and P. dartvini, respectively. Donoso-Barros and Cei (1962) and Donoso-Barros (1969b) showed that the cor- rect name for the Chilean species is P. thaiil and for the Uruguayan species, P. hibroni. Parker (1927) provided the first systematic re- view of the genus, and \^ellard ( 1960 ) treated the Andean species in Perii. Gorham (1966) listed 11 species in the genus, one of which (verrucosa) was placed in Ischnochema by Lynch (1971). The most recent review of the genus was by Gallardo (1968), who resurrected P. borellii from the synonymy of P. cinerea; Barrio and Rinaldi de Chieri (1970) did not recognize P. horcUii as distinct from P. cinerea. Lynch ( 1971 ) reviewed the generic relationships of the leptodactylid frogs and placed Pleurodema in the Lepto- dactylinae; he treated P. sagittifer ( Schmidt, 1857 ) as a species in- quierenda. Dunn (1944) questionably assigned P. sagittifer named from "Neu-Grenada" to the synonymy of P. marmorata, not an un- reasonable assertion in that several species named by Schmidt and supposedly originating from Panama and Colombia have subse- quently been found only in Peru and Bolivia.-' All trivial names associated with the genus Pleurodema and their current status are listed in Table 1. Cei (1962) summarized the existing knowledge of P. thaul (as P. bibroni). Veloso, Galleguillos, and Diaz (1973) analyzed the karyotypes of allopatric populations of P. thaul. Cei and Codoceo (1957) and Cei (1961) reported on the reproductive cycle of P. bufonina. Cei and Roig (1961) discussed the distribution of P, bufonina; this species was reported from additional Patagonian lo- calities in Chile by Donoso-Barros ( 1969a), Webb and Greer (1969), and Veloso, Diaz, and Galleguillos (1973). Cei (1972a) presented an analysis of serological variation in P. bufonina. Barrio (1964) utilized differences in mating calls for distinguishing three cryptic species of Pleurodema in the subandean desert region in Argentina. 3 Cochran and Coin (1970:486) referred one specimen (USNM 144910) from Barranquilla, Colombia, to P. sagiUifer. Examination of the specimen reveals that it unquestionably is P. brachyops. 4 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY Table 1. — Alphabetical Synonymy of the Species of Pleiirodenia. Trivial Name, Original Generic Name, Author, Date Current Name alpina (Paliidicola) Andersson, 1906 ^ cinerea andicola (Pleurodema) Philippi, 1902 thaid andina (Pleurodema) Philippi, 1902 thaul aspera (Pleurodema) Philippi, 1902 thaid belli (Pleurodema) Philippi, 1902 thaul bibroni (Pleurodema) Tschudi, 1838 bibroni borellii (Paludicola) Peracca, 1895 borellii brachijops (Lijstris) Cope, 1868 brachyops hufonina (Pleurodema) Bell, 1843 bufonina carbonaria (Pleurodema) Philippi, 1902 thaxd cinereum (Pleurodema) Cope, 1877 cinerea coquimbensis (Pleurodema) Philippi, 1902 thaul darwini (Pleurodema) Bell, 1843 bibroni diplolistris (Cijstignathus) Peters, 1870 diplolistris elegans (Pleurodema) Bell, 1843 thaul frenata (Paludieola) Cope, 1890 thaul fusca (Pleurodema) Philippi, 1902 thaul glandulosa (Pleurodema) Philippi, 1902 thatd gracilis (Pleurodema bibroni var.) Philippi, 1902 thaul guaijapae (Pleurodema) Barrio, 1964 guaijapae infuscata (Pleurodema marmorata) Vellard, 1960 marmorata kriegi (Paludicola) Miiller, 1926 kriegi longipes (Pleurodema) Philippi, 1902 thaul marmoratus (Leiuperus) Dumeril and Bibron, 1841 marmorata montevidense (Pleurodema) Philippi, 1902 bibroni nebulosus (Leiuperus) Burmeister, 1861 nebulosa phrtjniscoides (Pleurodema) Philippi, 1902 thaul plebeija (Pleurodema) Philippi, 1902 thaul pseudophnjne (Pleurodema) Philippi, 1902 thaul saehsi (Pleurodema) Peters, 1877 brachyops sagittifer (Leiuperus) Schmidt, 1857 marmorata ? salarius (Leiuperus) Bell, 1843 hufonina thaul (Bufo) Lesson, 1826 thaul tucumana (Pleurodema) Parker, 1927 tucumana verrucosa (Pleurodema) Philippi, 1902 thaul viridis (Leiuperus) Tschudi, 1845 marmorata Barrio and Rinaldi de Chieri ( 1970 ) reviewed the karyological data on Pleurodema and showed that P. hihroni and P. kriegi are polyploids, and Laurent (1975) demonstrated morphometric differ- ences between those species. ANALYSIS OF CHARACTERS Morphological, karyological, acoustic, and reproductive charac- ters used in the study are discussed below. Some characters have been used only in a phenetic sense, whereas in others it has been possible to ascertain evolutionaiy direction from primitive to de- rived states. The detennination of primitive and derived states PHYLOGENY OF PLEURODEMA 5 depended upon the utilization of the criteria of Wagner ( 1961 ) as discussed by Kluge and Farris (1969) and Trueb (1973). Lynch (1971) and Heyer (1975) discussed evolutionary direction in char- acters in leptodactylid frogs. Morphological Structure Morphological features, such as sizes and proportions, demial modifications, osteological structures, and coloration exhibit various combinations of differences among the species of Plcurodema. The more important of these are treated below. Size and sexual dimorphism. — The species of PJeurodema are moderately small frogs, ranging in snout-vent length from 33 mm in males of P. kriegi to 56 mm in females of P. hufonina. The largest males are in P. hracliijops ( 49 mm ) and the smallest females in P. kriegi (35 mm). Males are smaller than females; the ratio of snout- vent length between the sexes ranges from 1:104 in P. hrachyops to 1:122 in P. hufonina, fJiauI (southern Chile), and tucumana. Dermal Structure. — The skin on the venter is smooth, except for granules on the proximal posteroventral surfaces of the thighs. The skin on the dorsum is smooth (P. hrachyops, diplolistris), shagreened (P. hufonina, tucumana), smooth with scattered pustules (P. gua- yapae, nehidosa, thaul from Argentina and central Chile), smooth with scattered tubercles and folds in scapular region (P. horellii, cinerea), rugose with folds laterally (P. hihroni, kriegi), or bearing longitudinal rows of large glandular tubercles ( P. thaul from south- ern Chile). Evolutionaiy trends in dermal conditions are not evi- dent. Certainly scapular and lateral folds must be derived charac- ters. Most likely the dermal conditions are reflections of micro- environmental conditions in which the frogs live. Liunhar glands. — Many species of Pleurodema (hihroni, horellii, hrachyops, hufonina, cinerea, kriegi, and thaul) have a pair of prominent elevated glands dorsolaterally just posterior to the sac- rum. In P. hufonina the glands are large and hemi-ovoid (Fig. 1). In the other species the glands are smaller and more globular, pro- truding posteriorly in some. The glands of P. hufonina are usually distinctly darker than the surrounding skin. In the other species part of each gland is black; the rest of the gland is a contrasting color-yellow, red, or pale blue. The physiological function of these glands is unknown. Cei and Espina (1957) reported on the apo- sematic behavior of P. thaul, which lowers the head and elevates the sacral region, thereby displaying the prominent, colored glands, which in this position appear as eyes. Thus, the simulated head gives the impression of a much larger animal. The effect supposedly discourages potential predators. Although this behaxior is common in P. tliaul, we have not observed similar behavior in other species 6 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY Fig. 1. — Left Pleurodema biifonina, $ , 45 mm snout-\'ent length, 161375. Right P. tiicumana, 2 , 35 mm snout-vent length, KU 160844. KU having lumbar glands. In the leptodactyline frogs lumbar glands are derived characters; such glands are present only in Pleurodema. Tarsal fold. — Most species of Pleurodema lack an inner tarsal fold; P. borellii, cinerea, and thaul have a fold on the distal one- fourth of the tarsus, whereas the fold extends more than half the length of the tarsus in P. hufonina. Apparently the primitive con- dition is a short fold; the absence of a fold or presence of a long fold are independently derived characters. Tarsal tubercle. — Tubercles are absent on the tarsus of all spe- cies, except P. diploUstris, which has a large conical tubercle prox- imally on the tarsus. This unique character state obviously is derived. Inner metatarsal tubercle. — The tubercle is ovoid in most spe- cies; enlarged but not compressed in P. bibroni, kriegi, and tucu- mana; enlarged and compressed in P. brachyops, diploUstris, gua- ijapae, and nebulosa. The ovoid condition is considered to be prim- itive; the enlarged tubercle represents an independently derived state, with the compression of the enlarged tubercle being a pro- gressively derived state. Tympanic annuhis. — All species of Pleurodema have an ear; the upper edge of the tympanum is covered by a suprat\'mpanic dennal fold. The tympanic annulus is distinct in most species but partially or completely concealed in P. bibroni, guayapae, kriegei and thaul. The concealment of the tympanum is considered to be the derived character state. Prevomerine odontophores. — These structures are present in all species; in some species the odontophores are large and bear numer- ous teeth, where as in others they are small and have only two or three teeth. Among the species having small odontophores, the structures are buried in the dermal mucosa in P. bibroni, guayapae, kriegi, marmorata, nebulosa, tucumana. Prominent odontophores are considered to be primitive. PHYLOGENY OF PLEURODEMA Fig. 2. — Sternal structure in Pleiirodema: A. P. thaul (southern Chile), KU 162287. B. P. thaul (southern Argentina), KU 160810. C. P. brachyops, KU 167360. D. P. guatjapae (from Barrio, 1964:483). E. P. kriegi (from Gal- lardo, 1968:183). Sternum. — As in all leptodactylines, the sternal plate is ossified; the cartilaginous xiphisternum is partly calcified and variously shaped (Fig. 2). The following character states are identified by letters on figure 2: A. Primitive — P. J)oreIUi, hiifonina, cinerea, thaul (except in southern Argentina), tucumana; B. P. marmorata, thaul (southern Argentina); C. P. brachyops, diplolistris; D. P. guayapae, nehulosa; E. P. hihroni, kriegi. Each of character states B-E prob- ably were derived independently from A. The significance of the differences in sternal structure is not known. Since the sternum is the point of insertion and origin of various pectoral and abdominal muscles, differences in locomotion, burrowing, and /or amplexus might be reflected in its design. Vocal sacs and slits. — Large median, subgular vocal sacs with lateral expansions are present in all species except P. hufonina, which has small vocal slits. All species having vocal sacs have a mating call; P. hufonina has a release call. The presence of vocal sacs is primitive; the absence of vocal sacs is derived, correlated with the loss of voice. Testes. — The testes of breeding males are elongate, con\'oluted, and unpigmented, except in some populations of P. thaul, in which they are pigmented, and in P. marmorata, in which they are enor- mously enlarged ( zb one-third length of body), renifomi, and heavily pigmented. The functional significance of this apparently derived condition is unknown. Coloration. — Most species of Pleurodema are dull brown or gray dorsally, usually with some irregular darker markings; the venter is white, except for dark pigment on the throat in males and yellow, pink, or orange in the groin and on the hidden surfaces of the thighs of P. horcllii, brachyops, and cinerea. In P. thaul (except in south- ern Chile) and marmorata the dorsum is green, tan, orange, or 8 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY reddish brown with darker spots. The dorsum is dull gray or brown, and the venter is gray in P. timid from southern Chile. No evolu- tionary trends in coloration are evident. Karyology With the exception of P. hihroni and kriegi, the chromosomes of the species of Pleiirodema are essentially alike. Barrio and Rinaldi de Chieri (1970) noted that P. boreUii (their P. cinerea from Tucu- man, Argentina), Imfonina, cinerea, guayapae, nehulosa, than], and tucumana all have 2N:r=22 with a secondary constriction on the eighth pair of chromosomes. Veloso, Galleguillos, and Diaz (1973) reported on interpopulational variation in P. thaid; in specimens from Provincia Llanquihue, Chile, secondary constrictions are pres- ent on the short amis of pair 8 and on the long arms of pair 9. We present here a description of the karyotype of P. marmorata. Examination of preparations of seven males and two females from Putre, Provincia Tarapaca, Chile, revealed consistent chromosome structure in both sexes. Of the 11 pairs of chromosomes, pairs 1, 5, 9 and 10 are metacentric; pairs 2, 3, 6, 7, 8, and 11 are submeta- centric, and pair 4 is subtelocentric (Figs. 3-4). A secondary con- striction is present on the short arms of both homologues of pair 8. Most species of Pleiirodema are consei^vative karyologically; the presence of a secondary constriction on the ninth pair in P. fhatd in southern Chile is the only noticeable divergence. Barrio and Rinaldi de Chieri (1970) reported tetraploidy in P. hihroni and kriegi; this is a striking departiu-e from the otherwise conservative karyology of the group. As is evident from Barrio and Chieri's fig- Pleurodema marmorata %\lllhl%%\% iSU SI «« ■• Pleurodema thaul tf\S ii ^1 IS SI U *« II % Pleurodema bufonina .. i \\ W \\ S'i V% as •* » ^' *• 1234 5678 9 10 11 Fig. 3. — Chromosomes of Pleurodema. PHYLOGENY OF PLEURODEMA 9 ures 9 and 11, it is not possible to arrange the chromosomes in groups of fours. There are differences in size and structure of the supposed homologous pairs 1-2 and 15-16 in P. bihroni and in pairs 1-2, 11-12, and 15-16 in P. kriegi. Thus, Barrio and Rinaldi de Chieri suggested that these species are allopolyploids. However, a different mechanism possibly is responsible for the differences in the homologs. If a species has a long evolutionary histoiy as a polyploid, new arrangements of genetic material might explain the absence of complete homology ( Bogart and Wasserman, 1972). According to Barrio and Rinaldi de Chieri (1970), the Pleurodema marmorafa \\\\\ i 1 1 1 Pleurodema thaul (CO MiMiiiii Pleurodema thaul (SO ilil i i s I Pleurodema bufonina iMiSiiii. 123456789 10 11 Fig. 4. — Idiograms of Pleurodetua. Note secondary constriction on long anns of nintli pair in P. thaul (soutliern Chile). 10 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY polyploid species are the result of the combination of chromosomes of two parental species. Due to the morphological similarities of these disjunct polyploid species, we assume that the polyploid con- dition evolved only once, prior to the vicariation of the modern species. The identification of parental species cannot be detemiined with certainty. Karyologically, any of the diploid species of Pleuro- dema could have been parental species. Structurally and behavior- ally, the ideal parental stocks would have had lumbar glands, axil- lary amplexus, and foam nests. Furthermore, the ancestral stocks should have occurred in western or central Argentina. These cri- teria are met in their entirety only by P. horellii. Assuming P. boreUii to be one of the parental species involved in the develop- ment of the allopolyploid condition, the following alternatives for the second parental species can be explored: 1) P. tucumana. — Lumbar glands absent. Presently occurs in sympatry with P. boreUii in the vicinity of Tucuman, Argentina; no evidence of hybridization in zone of sympatry. 2) P. twbulosa and P. guayapoe. — Lumbar glands absent. Ranges of both species overlap in western Argentina; species hy- bridize in zone of sjanpatry (Barrio, 1964). 3) P. thaiil. — Lumbar glands present; differences in sternal structure from other species; no foam nest. Presently occurs in Nothofagiis forest. 4) A species having lumbar glands, axillary amplexus, and foam nest, but now extinct. We view the alternatives in priority of their numerical sequence, except the fourth, which exists as an untestable possibility. Pletiro- dema horellii and tucumana occur sympatrically at present and do not interbreed; furthermore, P. tucumana lacks lumbar glands. Pleurodema nebidosa and guayapae fonn hybrids; thus, because their genetic divergence is not great, they may be relatively recent in origin, more recent than the fomiation of the polyploid stock that gave rise to P. bibroni and kriegi. It seems unlikely that P. thaul was a parental species. Pleurodema boreUii and the polyploid spe- cies inhabit semi-arid environments, whereas P. thaul inhabits Nothofagus forest. Furthermore, the differences in reproductive biology seem to eliminate the likelihood of its interbreeding with a species having a foam nest. On the basis of this evidence it is difficult to explain the polyploid condition of P. bibroni and kriegi as having arisen by allopolyploidy. The polyploid condition could have arisen in P. horellii or a P. borellii-\ike ancestor by autopolyploidy. The latter apparently is a generalized way of polyploid formation in other polyploid anurans (Becak, et al, 196S; Becak, 1970; Becak and Becak,' 1974; Bogart, 1976). PHYLOGENY OF PLEURODEMA U Reproductive Biology Different aspects of reproductive biology in Pleurodema can be utilized in speculating on phylogenetic relationships within the genus. Amplectic position. — Frogs exhibit two basic kinds of amplexus- inguinal and axillary. Inguinal amplexus is considered to be the primitive condition (Lynch, 1973; Rabb, 1973); this amplectic position is characteristic of the archaeobatrachian families and the Australian myobatrachines. Inguinal amplexus was reported in the telmatobiine leptodactylid Batracliyla leptopus by Fonnas (1976), and illustrated in Tebnafobius somuncurensis by Cei (1972b). Cei (1962) reported inguinal amplexus in P. thaul from southern Chile. We have observed inguinal amplexus in P. thaul in southern Chile and in P. bufonina. These are the only records of inguinal amplexus in New World leptodactylids. All other PJeurodemo have, or are presumed to have, axillary amplexus. The presence of inguinal amplexus in two species of Pleurodema is paradoxical. Although in frogs as a group, inguinal amplexus is primitive, we consider the condition to be derived in Pleurodema. Eggs. — Three different kinds of egg clutch structure are found in Pleurodema. In P. thaul (except in southern Chile) and P. mar- morata eggs are deposited in clumps in water, whereas eggs are deposited in short, irregular strings in P. bufonina and thaul (in southern Chile). This type of egg deposition seems to be associated with inguinal amplexus in Pleurodema. All other species have, or are presumed to have, an aquatic foam nest-an advanced reproduc- tive mode ( Lynch, 1971 ) . We consider the aquatic clumps of eggs to be primitive (shared with telmatobiines) and eggs in strings (among leptodactylids unique to some Pleurodema) and foam nests (characteristic of most leptodactylines) to be derived independently. Tadpoles. — Tadpoles of P. thaul were described by Cei ( 1962 ) . We have examined tadpoles of P. brachyops, bufonina, cinerea, marmorata and thaul (southern Argentina, southern and central Chile) and have found that all are alike structurally in having: 1) an ovoid body about as wide as deep; 2) small eyes directed dorso- laterally; 3 ) sinistral spiracle directed posterodorsally at about mid- length of body; 4) moderately robust caudal musculature about equal to depth of dorsal fin at midlength of tail; 5) a dorsal fin not extending onto body; 6) median vent; 7) small anteroventral mouth with the upper part of the lip bare and elsewhere bearing a single row of small papillae; (S) lips barely folded laterally; 9) two upper and three lower rows of denticles with the second upper row inter- rupted medially; 10) beaks broadly arched and finely serrate. All of the tadpoles have a dull grayish brown body with a dark gray belly. The tail is tan or brown with fine dark brown reticu- 12 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY lations, least developed in P. hrachyops and cinerea and heaviest in P. thaul. In life, gold or greenish flecks or streaks are present ventrolaterally in some tadpoles. The lack of morphological diversitv' among these generalized leptodactylid tadpoles limits the use of larval characters in a phylogenetic analysis. Mating Call Barrio (1964) analyzed the mating calls of P. gumjapae, nebu- losa, and tucumana. Here we present analyses of the mating calls of P. hrachyops, cinerea, mormorata, and thaul and of the release call of P. hufonina (Figs. 5-6). Pleurodema hrachyops. — One individual was recorded at Nueva Gorgona and another at 3.4 km north of Montijo, Panama. Both were calling while floating in shallow ponds; water temperatures were 25° and 29.5° C, respectively. The call consists of long series of loud snore-like notes, 36-40 (x r= 38) per minute. The notes have a duration of 0.5-1.5 (x = 1.0) second. The notes are well modu- lated; the third harmonic at 650-750 (x = 700) Hz (Hertz) is domi- nant, and the pulse rate is 60 per second. Pleurodema cinerea. — Two individuals from the Rio Tupiza, Departamento Potosi, Bolivia, were calling from water having a temperatiu-e of 16° C; the air temperature was 18.5° C. Nine call groups of two individuals were analyzed. The call is a series of pulsed notes separated by short intervals. The notes are moderately well modulated with 11 hamionics evident between 80 and 8000 Hz. The third harmonic at about 1600 Hz is dominant. The duration of the notes is 0.14-0.20 (x = 0.17) of a second, and the pulse rate is 50-60 (x ■= 59.9) pulses per second. Pleurodema marmorafa. — Two individuals were recorded at Tambo Machay, Departamento Cuzco, Peru, at 11.5° C and at 5 km W of Tincopalca, Departamento Puno, Peru, at 30° C. Fourteen call groups of the four individuals were analyzed. The call consists of a series of moderately long notes having durations of 0.24-0.48 (x =: 0.34) of a second. The notes are moderately well modulated; the harmonics rise during the length of the note. The dominant frequency is at 1600-1900 (x= 1716) Hz at the beginning of the notes, rising to a mean of 1958 Hz at the end of the notes. A second emphasized hamionic is at 5258 Hz at the beginning, rising to 5861 Hz at the end of the notes. The individuals from Tambo Machay produced 60-80 (x = 70) pulses per second; those from Tincopalca, 50 pulses per second. Pleurodema thaul. — Analyses were made of 53 calls of 25 indi- viduals from 13 different localities in Chile, ranging from Pajonales (29°21'S) to Valdivia (39°50'S). Air temperatures were 13.0-19.0 PHYLOGENY OF PLEURODEMA 13 2- 0- 3- 2- 1- M I- cr UJ X o 0- 3- 1- 0- 3- 2- ■<«aMa»*«aWHM>.,,ida»> «■■»> -jm^ . »■ » ,' ^^„ , »-&-^-^-*'3P^Tp;3?i^ Jet*- «E-il^.;.=^- \ -.0^ -2i-. " ---^'- ■■ *- ^^^ ~ 1,1 .1, ■ tt -till- M-i-i.n 'rlntiT-.- I m, l^^ , , >_— . \ A,. V ^ 1 1 =?^f5fe !►!: ^ y^^jp^y^y^i^^, ' "<4i, ^ ^ .►M^MM^^^^^^^^ O-H- 0.0 .Z.Z.i: 0.2 0.4 SECONDS 0.6 Fig. 5. — Audiospectrogiams of mating calls of Pleuwdema. A. P. brachyops, KU Tape 867, 25° C. B. P. cinerea, KU tape 1206, 16° C. C. P. marmorala, KU tape 1223, 6.5° C. D. P. thaul, KU Tape 1213, 16° C. 14 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY (x = 16.75 ± 2.25) ° C. The call consists of a modulated trill having a duration of 0.2-9.0 (x =z 2.54) seconds. The second (of five) har- monic at 1500-2300 (x = 2047) Hz is dominant, and the fourth hamionic at about 5900 Hz also is emphasized. The call durations given above are for single individuals. Males calling in large cho- ruses sometimes call continuously for nearly one minute. At Cuesta de los Condores, Provincia Talca, Chile, a chorus of four males was observed to have an initially organized chorus struc- ture ( as defined by Duellman, 1967 ) . One male emitted short, loud, poorly modulated notes followed by the usual mating call. These two kinds of calls had different harmonic structures. As soon as the trill began, the other individuals began trilling. Pleiiroclema hufonina. — Although we have collected males in breeding condition at Laguna del Maule, Provincia Talca, Chile, and Laguna Blanca, Provincia Neuquen, Argentina, we have not heard the species produce a mating call. Jose M. Cei (pers. comm.) has confinned these observations on the basis of his extensive ex- perience with this species. However, P. hufonina does have a re- lease call and warning vibration; Cei and Espina (1957) reported the presence of warning vibrations in P. thaul, an activity also observed by us, but without a release call. Analyses were made of six release calls by two individuals from Laguna del Maule; the recordings were made in the laboratoiy at 25° C. The release call consists of two short notes each having a duration of about 0.02 of a second and separated from one another by an interval of about 0.25 of a second and then a longer note about 0.1 of a second in duration. The long note is moderately well modulated, having three or four hamionics. The first harmonic is dominant at about 1000 Hz initially and then declining to about 700 Hz; the pulses are produced at a rate of 140 per second. ACCOUNTS OF SPECIES In the following accounts of the species of PleiirocJema, a brief diagnosis is followed bv an account of the ecologv and distribution of each species. Additional information is presented in Remarks. Pleurodema bibroni Tschudi Pleurodeiua bibroni Tschudi, 1838:85. — Montevideo, Pio\incia Monte\ideo, Uruguay. Diagnosis. — Small species ( S 38 mm, 9 40 mm) showing little sexual dimorphism in size; skin on dorsum weakly rugose with folds laterally; lumbar gland large, prominent; inner tarsal fold absent; inner metatarsal tubercle ovoid; tympanic annulus concealed; pre- vomerine odontophores covered; sternum shallo\\'ly notched pos- teriorly; dorsum gray with large dark spots; belly white; 2N=44. PHYLOGENY OF PLEURODEMA 15 N I- cc UJ X o 0.0 0.4 0.8 SECONDS Fig. 6. — Audiospectrogram of release call of Pleuwdema bufonina, KU Tape 1357, 25° C. Ecology and Distribution. — Presumably terrestrial and having axillary amplexus and depositing eggs in foam nest. Known only from seasonally wet pampas of southeastern Uruguay from Treinta y Tres to Montevideo. Remarks. — We have no experience with this species, which was known as P. darivini until Donoso-Barros (1969b) demonstrated that the name P. hihroni (long associated with a Chilean species) was the correct name for the Uruguayan species. Pleurodema borellii (Peracca) Pahidicola boreUii Peracca, 1895:26. — San Migiiel de Tucuman, Provincia Tucu- man, Argentina. Diagnosis. — Large species ( S 44 mm, 9 49 mm ) showing only moderate sexual dimorphism in size; skin on dorsum smooth with scattered pustules dorsally and folds laterally; lumbar gland prom- inent; inner tarsal fold extending about one-fourth length of tarsus; inner metatarsal tubercle ovoid; tympanic annulus distinct; pre- vomerine odontophores prominent; sternum deeply incised pos- teriorly; dorsum orange-tan to grayish tan with brown or dark gray spots; belly white with tan Hecks laterally; ventral surfaces of thighs pinkish lavender; 2N=22 with secondary constriction on eighth pair. Ecology and Distribution. — Inhabiting sub-humid subtropical environments including broad-leafed deciduous forest and xeroph- ilous scrub forest and breeding in temporary ponds after seasonal rains. Amplexus axillary; eggs deposited in foam nest. Northwest- ern Argentina from northern Salta and eastern Jujuy through Tucu- man to Catamarca and La Rioja, occurring at elevations of 400 to nearly 3000 m. 16 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY Remarks. — Parker ( 1927 ) considered P. borellii to be a junior synonym of P. cinerea, but Gallardo (1968) recognized P. borellii as a distinct species. Comparison of specimens of P. cinerea from the Altiplano of southern Perii and Bohvia with specimens of P. borellii from the vicinity of Tucuman, Argentina ( ± 500 m ) reveals few differences. In comparison with P. cinerea, P. borellii is shghtly larger and has proportionately longer legs and feet. The snout in P. borellii is slightly narrower and more acuminate than the broader, blunter snout in P. borellii. The yellow coloration in the groin of P. cinerea was not observed in P. borellii. Although we have re- cordings of P. cinerea, we do not know the call of P. borellii. Until such time as decisive evidence is available, we prefer to recognize P. borellii and P. cinerea, species that live in strikingly different habits. Gallardo (1968) reported P. borellii from La Quiaca on the Altiplano in northern Provincia Jujuy, Argentina. We have speci- mens from the Altiplano collected at Abra Pampa south of La Quiaca. Our specimens are referable to P. cinerea. Other speci- mens from Rio Grande, Provincia Jujuy, also are P. cinerea. Con- sequently, the identifications of all specimens from Provincia Jujuy reported by Gallardo (1968) are suspect. In the southern part of its range P. borellii occurs in sympatiy with P. tiiciimana. The two species were found at the same pond at El Cadillal, Provincia Tucuman, Argentina, on the night of 2 December 1974. Pleurodema brachyops (Cope) Ltjstris brachyops Cope, 1868:312. — Rio Magdalena, Colombia. Diagnosis. — Large species ( S 49 mm, 9 51 mm) showing little sexual dimorphism in size; skin on dorsum smooth; lumbar gland prominent; inner trasal absent; inner metatarsal tubercle enlarged, compressed; tympanic annulus distinct; prevomerine odontophores prominent; sternum broadly dilated posteriorly with shallow in- dentation; dorsum pale green or tan with dark brown blotches; belly white; groin and hidden surfaces of legs orange; 2N=22. Ecology and Distribution. — Inhabiting savannas, llanos, and xeric scrub forests and breeding in temporary ponds after seasonal rains. Amplexus axillary; eggs deposited in foam nest. Guyana and extreme northern Brasil westward into Venezuelan llanos and along dry north coast of Venezuela and Colombia; savannas of Panama; coastal islands of Venezuela and Curacao. Remarks. — This species is distinctive in the genus by having deep orange "flash colors" in the groin and hidden surfaces of the limbs. Also, it is distinctive in having pale blue spots on the other- wise black lumbar gland. PHYLOGENY OF PLEURODEMA 17 One of us (Duellinan) has observed this speeies in Panama, Colombia, and Venezuela. Males eall while sitting in shallow water or while floating on the surface of the water. Although the lumbar gland is large and distinctively colored, no aposematic behavior displaying the gland was observed. Pleurodema bufonina Bell Pleurodema bufonina Bell, 1943:39. — Puerto Deseado, Provincia Santa Cruz, Argentina. Diagnosis. — Large species ( c5 45 mm, 9 56 mm) showing ex- treme sexual dimorphism in size; skin on dorsum shagreened, thick, glandular; lumbar gland enormous, approximately one-third length of body; inner tarsal fold extending at least to midlength of tarsus; inner metatarsal tubercle ovoid; tympanic annulus distinct; pre- vomerine odontophores prominent; sternum deeply incised pos- teriorly; dorsum dull brown with or without darker brown spots and/or pale middorsal stripe; venter creamy tan; 2N=22 with secondary constriction on eighth pair. Ecology and Distribution. — Terrestrial, frequenting arroyos and margins of lakes in the cold sub-humid or semi-arid Patagonian steppe and timdra. Amplexus inguinal; eggs deposited in short strings in water. Patagonian Argentina and southern Chile from the Straits of Magellan northward to 36° S principally east of the Andes but following tongues of Patagonian habitat through Andean passes at Laguna del Maule (36° S), Laguna La Laja (37°30'S), and Paso Pino Hachado (36°40'S), Chile. Elevational range from sea level to 2300 m. Remarks. — This is the southernmost frog; its distribution extends to 53° S. Cei and Roig ( 1961) reported on the distribution and sur- vival in harsh Patagonian environments. Cei and Codoceo ( 1957 ) and Cei ( 1961 ) noted the seasonally arrested spermatogenetic cycle in the species. Capurro (1957) reported P. bufonina from Isla Chiloe, Chile; examination of the illustration of the specimen reveals that it is P. thaul. Cei (1962) reported that M. Codoceo had observed P. bufo- nina and P. bibroni {= P. thaul) together in Aisen, Chile. Our field work in Chile and Argentina has not revealed the sympatric occur- rence of the two species. They are parapatric at Laguna El Maule and at Laguna La Laja. At these Chilean localities there are in- trusions of Patagonian vegetation into more mesic Andean scrub and forest. Pleurodema bufonina occurs only in the Patagonian vegetation, where P. thatd does not occur. The two species also were found in close proximity in Provincia Neuquen, Argentina; there, too, P. bufonina was found in the Patagonian steppe, whereas P. thaul occurred only in the Nothofagus forest. 18 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY Although breeding activit)^ has been observed, no mating calls have been heard. Breeding males do not have the laterally dis- tended, pigmented vocal sacs characteristic of most of the other species; small vocal slits are present. Males do have a release call and a warning vibration. In the northern part of its range P. hufonino commonly is found in association with Bufo spinulostis Wiegmann. Tadpoles of both species were found in pools in arroyos in Provincia Neuquen, Argen- tina, in December. At Laguna Blanca, Provincia Neuquen, P. hufo- nina was found beneath rocks at the edge of the lake; terrestrial stages of the primarily aquatic Telmatohius patagonicus Cei were under the same rocks. Pleurodema cinerea Cope Pleurodema cinereiim Cope, 1877:40. — Juliaca, Depaitmento Piino, Pern. Diagnosis. — Large species ( $ 42 mm, 9 45mm), showing little sexual dimorphism in size; skin on dorsum smooth with scattered pustules; lumbar gland prominent; inner tarsal fold extending about one-fourth length of tarsus; inner metatarsal tubercle ovoid; t>ni- panic annulus distinct; prevomerine odontophores prominent; ster- num deeply incised posteriorly; dorsum olive-tan to grayish brown with dark brown to black irregular spots; belly white; groin and hidden surfaces of thighs yellow; 2N=22 with secondary constric- tion on eighth pair. Ecology and Distribution. — Inhabiting puna (bunch-grass asso- ciations) and semi-arid montane environments and breeding in temporaiy ponds fomied by seasonal precipitation. Amplexus axil- lary; eggs deposited in foam nest. The Altiplano and some dissected valleys from the Titicaca Basin in southern Peru to northern Argen- tina, occurring at elevations of 2900-4100 m. Remarks. — Males call while sitting at the bases of bunch grass in water or floating on the surface of the water. In the Titicaca Basin P. cinerea occurs in sympatry with the smaller P. marmorata. Comparisons between P. cinerea and P. horellii are given in the account of the latter species. Pleurodema diplolistris (Peters) Cystignaihus diplolistris Peters, 1870:648. — Ceara, Brasil. Diagnosis. — Small species ( $ 40 mm, $ 43 mm ) showing little sexual dimorphism in size; skin on dorsum smooth; lumbar gland absent; inner tarsal fold absent; inner metatarsal tubercle enlarged, compressed; tympanic annulus distinct; prevomerine odontophores prominent; sternum broadly expanded and shallowly notched pos- teriorly; dorsum tan with irregular brown spots; venter white. PHYLOGENY OF PLEURODEMA 19 Ecology and Di.strihulion. — Occurring in the arid caatinga of northeastern Brasil (states of Ceara and Bahia). Remarks. — We have not observed this species in the field; there are no reports in the hterature on its Hfe history and karyology. Presumably the species has 22 chromosomes and exhibits axillary amplexus with eggs deposited in foam nests. Pleurodema diplolistris is unique among species in the genus by having a prominent tarsal tubercle. Pleurodema guayapae Barrio Pleurodema guayapae Barrio, 1964:478. — Guayapa, near Patquia, Provincia La Rioja, Argentina. Diagnosis. — Small species ( S 38 mm, 9 42 mm ) showing little sexual dimoiphism in size; skin on dorsum pustulate; lumbar gland absent; inner tarsal fold absent; inner metatarsal tubercle enlarged, compressed; tympanic annulus concealed; prevomerine odonto- phores covered; sternum with posterior osseous bifurcation; dorsum brownish pink with small dark spots; venter white; 2Nzrr22 with secondary constriction on eighth pair. Ecology and Distribution. — A fossorial species living in arid and semi-arid habitats and breeding in temporaiy pools after brief sea- sonal rains. Amplexus axillary; eggs in foam nest. Central Argen- tina from southwestern Santiago del Estero to central La Rioja, including extreme southern Catamarca and northwestern Cordoba. Remarks. — Barrio (1964) distinguished this species from P. nehidosa on the basis of the mating call. He discovered hybrids between P. guayapae and P. nehidosa at Guayapa. The mating call of the hybrid is intermediate between those of the parental species. The ranges of the two species overlap only in the central part of Provincia La Rioja. The range of P. guayapae lies completely within that of P. tucumana. Pleurodema kriegi (Miiller) Paludicola kriegi Miiller, 1926:194. — Sierra Grande, Provincia Cordoba, Ar- gentina. Diagnosis. — Small species ( S 33 mm; $ 35 mm) showing little sexual dimorphism in size; skin on dorsum slightly rugose with folds laterally; lumbar gland large, prominent; inner tarsal fold absent; inner metatarsal tubercle ovoid; tympanic annulus concealed, pre- vomerine odontophores covered; sternum shallowly notched pos- teriorly; dorsum dull reddish gray with large black spots; belly white becoming ochre on chest and reddish yellow on undersides of legs; 2N=44. Ecology and Distribution. — Probably terrestrial, inhabiting semi- arid habitats; presumably having axillary amplexus and depositing 20 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY eggs in foam nests. Known only from the Sierra Grande and Pampa de Achala, western Cordoba, Argentina. Remarks. — We have no experience with this species, which Parker (1927) synonymized with P. cinerea. Gallardo (1968) re- described essentially topotypic specimens and placed P. kriegi in the synonymy of P. danvini ( := P. ]}iJ)roni). Barrio and Rinaldi de Chieri ( 1970 ) noted the tetraploid condition of P. kriegi and recog- nized the species as being distinct from P. hihroni, which also is tetraploid. They noted secondary constrictions on the 15th pair of chromosomes of both species and also on the 12th pair of P. kriegi, although the latter constriction is not evident in their figure. Lau- rent (1975) noted slight diflerences in proportions and in coloration. The ranges of the two species are separated by a distance of nearly 1000 km and includes the pampas of eastern Argentina, a region well known herpetologically. Pleurodema marmorata (Dumeril and Bibron) Leiupertis maiinoratus Dumeril and Bibvon, 1841:421. — Potosi, Departamento Potosi, Bolivia. Diagnosis. — Small species ( <5 38 mm; ? 46 mm) showing moder- ate sexual dimorphism in size; skin on dorsum smooth with scattered pustules; lumbar gland absent; inner tarsal fold absent; inner meta- tarsal tubercle ovoid; tympanic annulus usually distinct; prevomer- ine odontophores covered; sternum rounded posteriorly; dorsum gray, tan, green, red, or brown with dark green, brown, or black markings; venter white; 2N=:22 with secondary constriction on eighth pair. Ecology and Distribution. — Inhabiting wet and dry puna (bunch-grass associations) between 3000 and 5000 m. Amplexus axillary; eggs deposited in masses in shallow temporary or perma- nent ponds. Ranging from central Peru to central Bolivia and ex- treme northeastern Chile, occurring on the western slopes of the Cordillera Oriental, the Altiplano and the Cordillera Occidental. Remarks. — \'ellard (1960) recognized three subspecies of this highly variable species. Examination of \^ellard"s specimens plus extensive series from throughout the range in the collections at the Museum of Natural Historv, The Universitv of Kansas, reveals that P. marmorata is a highly variable species and that within-population variation in proportions and coloration frequently is greater than between-population variation. Consequently, we prefer not to rec- ognize taxonomically any populations within P. marmorata. In the Titicaca Basin in southern Peru P. marmorata occurs sympatrically with P. cinerea, a larger species depositing eggs in foam nests. Comparison of series of P. marmorata from the sym- patric zone (Santa Rosa, Departamento Puno, Perii) with specimens PHYLOGENY OF PLEURODEMA 21 from north of the range of P. cinerea ( Departamento Ayacucho, Peru) revealed tliat southern specimens are larger and have pro- portionately longer feet than the northern ones. No detailed mor- phometric analysis of samples from throughout the range was undertaken. Calling males were observed at many localities in Perii. Males sit on the ground at the edge of shallow pools or sit in shallow water. None was observed calling while floating in the water. This species reaches higher elevations in the Andes than any other amphibian. We have observed it to elevations of 4675 m, and Vellard (1960) stated that it reaches 5000 m. At these high ele- vations, the frogs usually are found beneath rocks -frequently on wet ground with small pools of water that freeze nightly. In the Cordillera Occidental in central Peru and northern Chile, and in the Cordillera in Bolivia we have found numerous individuals be- neath rocks covered with snow. One individual was observed hop- ping across a snow field by day. Males were heard calling when the air temperature was 1° C. Pleurodema nebulosa (Burmeister) Leitiperus nehulosus Burmeister, 1861:532. — Mendoza, Provincia Mendoza, Argentina. Diagnosis. — Small species ( i 36 mm; 9 40 mm) showing little sexual dimorphism in size; skin on dorsum weakly pustulate; lumbar gland absent; inner tarsal fold absent; inner metatarsal tubercle enlarged, compressed; tympanic annulus distinct ventrally; pre- vomerine odontophores covered; sternum with posterior osseous bifurcation; dorsum pinkish gray with black flecks; venter white; 2N^22 with secondary constriction on eighth pair. Ecology and Distribution. — A fossorial species living in arid habitats and breeding in temporary rain pools. Amplexus axillary; eggs in foam nest. Western Argentina from southwestern Cata- marca to central Mendoza, including western San Luis, with dis- junct populations in southwestern Cordoba and northern La Pampa. Remarks. — Barrio (1964) discovered the sympatric occurrence of and hybridization of cryptic species at Guayapa ( see account of P. guayapae ) . The two species differ structurally in the concealment of the tympanic annulus in P. guayapae and in mating calls. Pleurodema thaul (Lesson) Bufo thaul Lesson, 1826:64. — Concepcion, Provincia Concepcion, Chile. Diagnosis. — Small to large species depending upon popidation, showing moderate sexual dimorphism in size; skin on dorsum smooth, pustulate, or with rows of large pustules; lumbar gland prominent; inner tarsal fold extending about one-fourth length of 22 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY tarsus; inner metatarsal tubercle ovoid; tympanic annulus partly concealed; sternum rounded or deeply incised posteriorly; dorsum green, gray, or tan with darker markings, or dark grayish brown with or without faintly darker markings; venter white or gray; 2Nr:r 22 with secondary constriction on eighth pair ( also on ninth pair in some populations ) . Ecology and Distribution. — Occurring in humid and subhumid austral forests and subhumid chaparral (matorral) vegetation from Aisen northward to La Serena, Provincia Coquimbo, and thence with discontinuous populations northward to Copiapo, Provincia Atacama, Chile; in Argentina occurring in austral forests along the eastern base of the Andes. Elevational range from sea level to 1500 m. Amplexus axillary or inguinal; eggs deposited in strings or masses in water. Remarks. — We regard P. thaul as a composite species of at least three forms differing in size, proportions, coloration, karyology, amplectic position, and eggs. These three forms we designate solely by their geographic positions-Southern Chile ( SC ) , Centi'al Chile (CC), and southern Argentina (SA). The three forms are diag- nosed, as follows- Southern Chile (SC): Large frogs ( $ 40 mm, 9 53 mm) hav- ing extreme sexual dimorphism in size; snout narrowly rounded in dorsal profile; skin on dorsum bearing longitudinal rows of large tubercles; sternum deeply incised posteriorly; nuptial pad well de- veloped; dorsum predominately dull dark brown or gray, some in- dividuals having paler brown with large dark spots; venter gray; secondary constrictions on eighth and ninth pair of chromosomes; amplexus inguinal; eggs deposited in short strings in water. Central Chile (CC): Moderate-sized frogs ( $ 44 mm, 9 48 mm) exhibiting minor sexual dimorphism in size; snout broadly rounded in dorsal profile; skin on dorsum smooth with or without small pustules; sternum deeply incised posteriorly; nuptial pad poorly developed; dorsum gray with small dark green spots or tan or dark green with small dark brown spots; venter white; secondary constriction on eighth pair of chromosomes; amplexus axillary; eggs deposited in masses in water. Southern Argentina (SA): Moderate-sized frogs ( i 41 mm; 9 48 mm) exhibiting considerable sexual dimorphism in size; snout broadly rounded in dorsal profile; skin on dorsum smooth with or without small pustules; sternum round posteriorly; nuptial pad well developed; dorsum gray, green, or tan with dark brown or green irregular spots; venter white; secondary constriction on eighth pair of chromosomes; amplexus axillary; eggs deposited in masses in water. In addition to these three forms we also note the existence of a PHYLOGENY OF PLEURODEMA 23 moiph presently known only from an isolated saline pond in the desert at Pajonales in northern Provincia Coquimbo, Chile. Frogs from this locality have completely smooth skin dorsally and a pale coloration with yellow in the inguinal region. The southern Chilean fonn occurs throughout Provincia Llan- quihue and on Isla Chiloe; presumably this is the form in the ex- treme south in Aisen. The central Chilean form ranges from the Rio Bio-Bio northward to La Serena. The southern Argentinian form occurs in the western parts of Provincia Neuquen and Pro- vincia Rio Negro. In south-central Chile between Concepcion and Valdivia there is a confusing array of Pleurodema with respect to polymorphism in dorsal pustules and coloration. Cei and Capurro (1937) and Cei (1958) analyzed the variation in coloration and morphological features in P. thoiil, and Cei ( 1958) concluded that additional studies on this polymoiphic species should be able to define distinct taxa. Our major contribution to the problem has been the discovery of differences in sternal condi- tions, amplectic positions, egg deposition, and karyotypes-all char- acters used at the specific level in Pleurodema. However, we re- frain from taxonomic recognition of the fomis at this time due to our incomplete knowledge of the distribution and variation within the fomis, especially in the region from Concepcion to Valdivia. Furthermore, once the variation within the taxa is better knowai it is necessary to determine the accurate assignment of Philippi's (1902) many specific names, all of which now reside in the syn- onymy of P. thaiil. The presently confusing array of variation in populations between Concepcion and Valdivia perhaps is indicative of the existence there of representatives of both the northern and southern forms. There are no apparent barriers to gene flow in Pleurodema from Provincia Llanquihue to central Chile ( vicinity of Santiago), although the Rio Bio-Bio is the northern boundary for some species restricted to the austral forests. Possibly an additional taxon exists in this region. In southern Chile and Argentina this species occurs in Notho- fagus and Araucaria forests; farther north in the lowlands it occurs in chaparral (matorral) associations and in relict stands of Notho- fagus. The species also occurs in the deeply incised Andean Valleys in central Chile where it attains elevations of 1500 m. In two of these valleys (Laguna del Maule and Laguna La Laja), P. thaul is parapatric with P. hufonina, which occurs only in the Patagonian vegetation in the upper parts of the valleys. Although the species has been found on the western slopes of passes in southern Chile to elevations of 1010 m, it has not been found on the tops of Paso de Petrohue ( 980 m ) and Paso de Puyehue ( 1479 m ) , where potential gene flow might occur between populations in southern Chile and southern Argentina. 24 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY Pleurodema tucumana Parker Pleurodema tucumana Parker, 1927:468. — Tapia, Provincia Tucuman, Ar- gentina. Diagnosis. — Large species ( S 40 mm, 9 53 mm) showing ex- treme sexual dimorphism in size; skin on dorsum weakly granular; lumbar gland absent; inner tarsal fold absent; inner metatarsal tu- bercle enlarged, not compressed; tympanic annulus distinct; pre- vomerine odontophores usually covered; sternum deeply incised posteriorly; dorsum brown with irregular dark brown spots; venter white; 2N=r22, probably with secondary constriction of eighth pair. Ecology and Distribution. — A fossorial species inhabiting sub- humid to arid habitats and breeding in temporary pools after sea- sonal rains. Amplexus axillary; eggs in foam nest. Central Argentina from northern Tucuman and western Santiago del Estero through southeastern Catamarca and northwestern Cordoba to central La Rioja. Remarks. — The range of P. guayapae is completely encompassed within the range of P. tuctiniana. In central La Rioja both species occur in sympatry with P. nebiilosa. In Tucuman P. tucumana oc- curs with P. ])oreUii; at El Cadillal, Provincia Tucuman, both species were observed at the same pond on the night of 2 December 1974 (Duellman, per. observ. ). INTRAGENERIC RELATIONSHIPS Using the cladistic approaches of Hennig ( 1966 ) , we have attempted to detennine branching sequences and sister groups of Pleurodema. In this method the phylogeny of a group is recon- structed by defining vicariance events; all species on a polytypic branch share derived characters and differ from species on the other branch ( sister group ) in one or more derived characters. In the following list of characters, the primitive state is desig- nated 0; derived states are 1, 2, etc., as noted. The characters are described in the foregoing section on character analysis, and the character states are tabulated in Table 2. A. Amplectic position. 0. Axillary 1. Inguinal B. Egg deposition (1 and 2 independently derived) 0. Masses 1. Strings 2. Foam nest C. Lumbar gland 0. Absent 1. Present PHYLOGENY OF PLEURODEMA 25 D. Chromosomes 0. 22 1. 44 E. Tarsal fold (1 and 2 independently derived) 0. Short 1. Long 2. Absent F. Inner metatarsal tubercle (1 and 2 independently derived; 3 derived from 2) 0. Ovoid 1. Small 2. Enlarged 3. Enlarged and compressed • • G. Tympanic annulus 0. Prominent 1. Concealed H. Prevomerine odontophores 0. Prominent 1. Covered 1. Sternum (each state independently derived) 0. Figure 2A 1. Figure 2B 2. Figure 2C 3. Figure 2D 4. Figure 2E Four characters (A-D) were used to generate cladograms de- fining six major branches in Pleurodema (Fig. 7); each branch con- tains one to four species and represents a monophyletic group. Group A.-P. marmorofa ( character states for A-D = 0000 ) Group B.-P. diplolistris, g,iiai/apae, nehiiJosa, tticumana (0200) Group C.-P. than! (SA), thaul (CC) (0010) Group D.-P. bufonina, thaul (SC) (1110) Group E.-P. borellii, cinerea, bracliyops (0210) Group F.-P. bibroni, kriegi (0211) Two equally parsimonious phyletic arrangements of these groups are possible; each requires five evolutionary steps, and each con- tains one pair of convergences. In the first alternative (Fig. 7A) the deposition of eggs in foam nests occurs in two phyletic lines, whereas in the second alternative (Fig. 7B) the development of lumbar glands occurs twice. We favor the first alternative because it is most reasonable biologically. Within the Leptodactylinae large lumbar glands are unique to certain species of Pleurodema, whereas foam nests are common in leptodactylines and also occur in other families. Thus, it seems more reasonable to assume the development of foam nests in two phyletic lines of Pleurodema than to require the evolution of lumbar glands independently to two phyletic lines. 26 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY Table 2. — Character States of 13 Characters in Pleuwdema. ( See text for explanation ) _ u cc ,co ^ S-J f u u e •*-> o « Sw a 53 « C/D u en o c» -12 :=> •^ ^ O o --^ c f-^ f-w O t o 5 rS ^5 5 s s ^ u -JS r^ .« r^ 'G '"^ 6D r^ S s— •s A 0 0 0 1 0 0 0 0 0 0 0 0 1 0 B 2 2 2 1 2 2 2 2 0 2 0 0 1 2 C 1 1 1 1 1 0 0 1 0 0 1 1 1 0 D 1 0 0 0 0 0 0 1 0 0 0 0 0 0 E 2 0 2 1 0 2 2 2 2 2 0 0 0 2 F 2 0 3 0 0 3 3 2 0 3 0 0 0 2 G 1 0 0 0 0 0 1 1 0 0 1 1 1 0 H 1 0 0 0 0 0 1 1 1 1 0 0 0 1 I 3 0 2 0 0 2 4 3 I 4 1 0 0 0 INTERGENERIC RELATIONSHIPS According to Lynch (1974) and Heyer (1975), Pleuroclema is a primitive leptodactyline. Lynch (1971: 207) stated: "The Lepto- dactyhnae are derived from the relative primitive Alsodini (Eup- sophus). The most primitive leptodact\4ine (Pleurodema) is very similar to Eupsophus. The two genera differ in the sternal appa- ratus, breeding biology, and loss of the qnadratojngal. Pleurodema has an osseous sternal style (as do all other leptodactylines ) ; these two characters clearly ally Pleurodema with the Leptodactylinae, although its close relationship with Eupsophus is obvious and could be used to support the argument that the Leptodactylinae are only a tribe of the Telmatobiinae." It should be noted that Lynch's Eupsophus includes frogs now placed in Alsodes, Eupsophus, Phrynopus, and Telmatohius. In Heyer's (1975) cladistic arrange- ments of derived characters of leptodactylid frogs, Pleurodema clustered with the telmatobiine genera Batrachijla, Bafrachophrynus, Eupsophus, and Telmatohius in two cladograms, and with the leptodactyline genus Physalaemus in a third cladogram. It is constructive to compare Pleurodema with Alsodes nodosus (Eupsophus nodosus of Lynch, 1971). Veloso and Itinra (1976) provided karyological evidence for the separation of Alsodes and Eupsophus. The latter is considered to be a primitive telmatobiine having species with 2(S, 30, and 34 chromosomes, many of which are telocentric, thereby giving fundamental numbers of 46, 54, and 58. All Alsodes, except A. nodosus, have 26 metacentric and submeta- centric chromosomes, whereas A. nodosus has 22 metacentric and submetacentric chromosomes; thus, Alsodes have fundamental num- bers of 52 and 44. Most other telmatobiines have 26-34 chromo- PHYLOGENY OF PLEURODEMA 27 Fig. 7. — Alternative phylogenies of six species groups of Pleuwdema based on characters A-D. Character shifts are: 1. Development of lumbar glands. 2. Eggs deposited in foam nest. 3. Inguinal amplexus and eggs laid in strings. 4. Polyploidy. See text for species composition of groups. somes; only Alsodes nodo.ms and Telmatobius somunciirensis (Cei, 1969) have 22 chromosomes. With the exception of the tetraploid species (P. Jjibroni and kriegi), all Pleurodema have 22 metacentric and submetacentric homomorphic chromosomes with a secondary constriction on pair 8. The 22 chromosomes of Alsodes nodosiis are more variable in size than those of Pleurodema; in A. nodosus the fourth pair is hetero- morphic, and a secondary constriction is present on pair 2. Karyo- logically, Pleurodema and Alsodes nodosus could have had a com- mon ancestor, but it is more reasonable to postulate an independent derivation of 22 chromosomes in Pleurodema from an I'Aipsopluis stock. Structurally Alsodes nodosus is distinctly different from Pleuro- 28 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY dema. It is a moderately large, terrestrial species; males have spinous nupital adspersities on the thumb and chest. Males have a mating call, and amplexus is axillary; large unpigmented eggs are deposited in clumps in streams in the Pacific lowlands and lower Andean slopes of central Chile. The large tadpoles have two upper and three lower rows of denticles. Among the species of Eupsophus, vertebralis has 28 chromo- somes (fundamental number 54), roseus 30 (46), and vanzoUnii 34 (58).^ It is easiest to derive the chromosome complement of Pleuro- dema from a stock resembling Eupsophus roseus. This small terres- trial species inhabits humid Nothofagus forest in southern Chile. Males have a mating call but lack nuptial adspersities. Amplexus is axillary, and the eggs are unpigmented; presumably they are deposited in water. Comparison of characters of the primitive telmatobiines (Tribe Telmatobiini, including Alsodes, BatracJwphrynus, Eupsophus, Hylorina, Limnomedusa, and Telmatobkis) with Pleurodema and the subfamily Leptodactylinae (exclusive of Pleurodemo) reveals that Pleurodema is intemiediate between the other groups in some characters, shares some character states with one or the other group, and differs from both groups in some characters (Table 3). The moiphological and karyological evidence supports Lynch's ( 1971 ) contention that Pleurodema can be considered as a primitive leptodactyline and possibly allied with the telmatobiine Eupsophus. The inguinal amplexus exhibited by two species of Pleurodema is unusual among New World leptodactylids. Fomias (1976) reported inguinal amplexus in Batrachyla leptopus, and Cei (1972b) illustrated inguinal amplexus in Tehnatohius somuncurensis. Eggs in short strings, as found in Pleurodema hufonina and P. thaul (SC) are unique in the Leptodact\'lidae. We interpret these repro- ductive differences (inguinal amplexus and eggs in strings) as adap- tive modifications that are divergent from the major evolutionary trends in the Telmatobiinae-Leptodactylinae lineage. With the exception of Pseudopaludicola and some Pleurodema, all leptodactylines deposit eggs in foam nests. Heyer (1969) noted evolutionary adaptations in nests in the Lepfodactylus complex. The trend is from foam nests floating in water to nests in hollows that fill with water; in these instances aquatic tadpoles are present. In Adenomera a foam nest is constructed on land and there are no aquatic larvae. Bogart (1974) discussed the evolutionary significance of karyo- types of the Lepiodactylus complex. Noting the presence of 24 and ■1 Ihina and Vehso (1976) reported 34 chromosomes in Eupsophus cop- pingcri from the Sierra Nahuelbuta, Chile. The frogs in the Sierra Naluielbuta were named as Eupsophus vanzolinii by Donoso-Barros (1974). PHYLOGENY OF PLEURODEMA 29 Table 3. — Comparison of Three Groups of Leptodaetylid Frogs. Character Tehnatobiiiii Piciirodema Leptodactylinae Sternal Style Quadratojugal Cotylar Type Pupil Lar\ al \'ent Amplexus Eggs Chromosomes Cartilaginous Osseous Present Absent II I or II Horizontal or xertical Horizontal Median Median Axillary A\illar\' or inguinal Water Water or foam 22-34 22 Osseous Present or absent II Horizontal Median or de.xtral Axillary Water or foam 22-26 26 chromosomes in Adenomera and 22 chromosomes in Leptodac- tylus, lie suggested that Adenomera was a primitive leptodactyHne and that the subfamily might have evolved through a hneage from Eupsophtis and Bafraehyla through Eleutlierodaetylus to Adeno- mera and thence Lcptodactyhis. Heyer and Diment (1974) criti- cized Bogart's hypothesis, noting that his proposed hneage is con- traiy to all other evidence in the leptodactylines. Bogarfs phylogeny is based solely on karotypes-only one set of characters-to the ex- clusion of data on morphology, life history, and ecology. When all suites of characters are considered, the primitive chromosome num- ber in the Leptodactylinae must be 22. Thus, 24 and 26 chromo- somes in Adenomera must be derived, an hypothesis consistent with the derived morphological and ecological characters of the genus. Pleurodema has 22 chromosomes with a secondary constriction on pair number 8; the same chromosome structure is common in Leptodactylus (Bogart, 1974). The floating foam nests of Pleuro- dema are the same as the floating nests of Leptodactylus and Physalaemus. Pleurodema seems to be an ideal ancestral lepto- dactyline, except that the quadratojugal is absent. This element is absent only in one other genus of leptodactylines, Pseudopaludicola, the only other leptodactyline depositing eggs in water. THE BIOGEOGRAPHIC MODEL Ideally, biogeographic hypotheses should be deductive (Ball, 1975), not inductive narratives. In order to make our biogeographic model meaningful to the interpretation of the paleogeography of other groups of organisms, we provide a narrative of: 1) the physi- ographic, climatic, and vegetational history of southern South America, and 2) the distributional history of the diverging groups of Pleurodema. The primitive telmatobiine leptodactylids are restricted to, or are most diverse in, the austral Nothofagus forests, whereas the leptodactylines are primarily non-forest frogs, inhabiting savannas, grasslands, and deserts. Thus, the histories of the austral forests and 30 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY the austral xeric vegetation formations are pertinent to the evokition of the leptodactyhd frogs. The Changing Stage The Tertiary is characterized by gradual cooling and drying subsequent to the Eocene ( Axelrod and Bailey, 1969; Wolfe, 1971 ) . The austral forests with such characteristic floristic elements as Araticarki, Laurelia, and Nothofagus have existed in southern South America at least since the Eocene ( Cooper, 1960; Cerceau-Larrival, 1968). It is now generally conceded that this austral flora had a Gondwanian dispersal (Raven and Axelrod, 1974) -one that also involved leptodactylid frogs (Lynch, 1971; Savage, 1973). The austral flora was more widespread in the early Tertiary than it is now; in the Oligocene Araucaria, Laurelia, and Notlwfagus occurred northward to at least 30° S. Lat. in Chile, and Nothofagus was widely distributed in Patagonia ( Jeannel, 1967 ) . Solbrig (1976) hypothesized the existence in middle latitudes of South America since the early Tertiary of a tropical deciduous or semideciduous forest with some xerophytic adaptations. He temied this the Tertiary-Chaco Paleoflora and hypothesized that this flora persists with modifications to the present. Prior to the Miocene there was no uplift of the Andes; climates in southern South America were more equitable than now and climatic belts extended across the continent. Volcanic activity in Chubut and Santa Cruz in southern Argentina and the uplift of the Bolivian cordillera were initiated in the Miocene (Harrington, 1962; Aubodin, et «/., 1973). With the initiation of the Andean uplift in the Miocene, the climates east of the Andes were modified. According to Solbrig (1976), in the Miocene there was a wide ex- pansion of the Argentinian pampas with Nothofagus forest adjacent to the pampas; the Nothofagus forest was restricted to about its present distribution east of the Andes. Also, semi-desert zones developed between mountains in the western part of the continent and in northern Patagonia. In the Pliocene there was strong volcanic uplift of the Andes (Petersen, 1958; Rudand, et al, 1965) and increased rain shadow effects east of the mountains. In the Pliocene the Chacoan vege- tation expanded northeastward, possibly giving rise to the dry caatinga formation in northeastern Brasil; semidesert associations developed along the eastern face of the Andes and in the south separated the NotJiofagus forests from the pampas (Solbrig, 1976). The semideserts were the precursors of the modern Patagonian and monte vegetations. Also, during the Pliocene the bunch-grass asso- ciation (puna) developed in the Andes. Throughout Miocene and Pliocene times there was a continued PIIYI.OGENY OF PLEURODEMA 31 desiccation of the climates along the Pacific coast from central Chile to northern Peru with the restriction or elimination of elements of the Tertiary-Chaco Paleoflora, leaving many disjunct distributions as far north as southern Ecuador ( Jeannel, 1967). The development of the extreme xeric conditions of the Atacama Desert and the re- striction of the chaparral type of vegetation to central Chile oc- curred in the Pleistocene, when the Andean Orogeny thrust the Cordilleras to their present heights, and the cold Humboldt Current assumed its present course. Montane glaciation and associated climatic depression was wide- spread throughout the Andes; three or four glacial stages are recog- nized (Wiilleumier, 1971). The only major continental glacier ex- tended completely across Patagonia between 41^ and 42° S Lat. (Jeannel, 1967). However, the basaltic Somuncura Plateau was not glaciated (Cei, 1969). The lowlands of southern Chile were dis- sected by tongues of montane glaciers. During the Pleistocene the climatic regimes of southern South America alternated between cold and temperate probably with cor- responding latitudinal shifts in vegetation. In central Argentina the changes involved drier (interglacial) and wetter (glacial) epochs. Pleistocene and Holocene events in tropical South America also in- volved alternating wet and dry epochs resulting in expansion of forests and restriction of non-forest habitats in the humid inter- glacials and expansion of non-forest habitats and restriction of forests in the dry glacial periods. Van der Hammen and Gonzalez (1969), Van der Hammen (1961), Vanzolini and Williams (1970), and Haffer (1974 and numerous sources cited therein) have docu- mented these climatic fluctuations on the basis of palynological evidence, floristics, and patterns of avian and saurian speciation and distribution. If not before the Pleistocene, at least during Pleisto- cene dry periods, there existed a dry corridor connecting the Cha- coan xerophytic vegetation with the dry Caribbean lowlands and llanos of Venezuela, vegetation formations that existed at least since the Pliocene (Solbrig, 1976; Samiiento, 1976). These historical events resulted in the present climatic and vege- tational patterns that exist today (Fig. ' biology of tlie anurans: contemporary re- search on major problems. Uniw Missouri Press, Columbia: 133-182. PHYLOGENY OF PLEURODEMA 45 MuLLEH, L. 1926. Neue Reptilicn unci Batiachier der Zoologischen Samiiilung des Bayeiischen Staats. Zool. Anz., 65:193-200. MiJLLER, P. 1973. The dispersal centres of terrestrial vertebrates in tlie Neo- tropical Realm. Junk, The Hague: 1-244. Parker, H. W. 1927. A revision of the frogs of the genera Psetidopaludicola, Physalaemiis, and Pleuwdcma. Ann. Mag. Nat. Hist., (9)20:450-478. Peracca, M. G. 1895. Viaggio del dott. Alfredo Borelli nella Rep. Argentina e nell Paraguay. Rettili ed AnBbi. Bol. Mus. Zool., 10 (195): 1-32. Peters, W. 1870. Mitteilungen iiber Amphibien des Koniglich Zoologischen Museums. Monats. Akad. Wiss. Berlin, 1870:640-652. Petersen, U. 1958. Structiue and uplift of the Andes of Peru, Bolivia, Chile, and adjacent Argentina. Bol. Soc. Geol. Peru, 33:57-143. Philippi, R. a. 1902. Suplemento a los Batraquios Chilenos descritos en la Historia Fisica i Politica de Chile de don C. Gay. Santiago: 1-161. Rabb, G. B. 1973. Evolutionary aspects of tlie reproductive behavior of frogs. in Vial, J. L. (ed. ). Exolutionary biology of the anurans: contempo- rary research on major problems. Uniw Missouri Press, Columbia: 213- 227. Raven, P. H., Axelrod, D. I. 1974. Angiosperm biogeography and past conti- nental movements. Ann. Missouri Bot. Garden, 61:539-673. Rosen, D. E. 1975. A vicariance model of Caribbean biogeography. Syst. Zool., 24:431-464. Rutland, R. W. R., Guest, J. E., Grasty, R. L. 1965. Isotopic ages and Andean uplift. Nature, 208:677-678. Sarmiento, G. 1976. Evolution of arid vegetation in tropical America, in Goodall, D. W. (ed.). Evoluticm of desert biota. Univ. Texas Press, Austin: 65-99. Savage, J. M. 1973. The geographic distribution of frogs: patterns and pre- dictions, in Vial, J. L. (ed.). E\()lutionary biology of the anurans: contemporary research on major problems. Univ. Missouri Press, Co- lumbia: 351-445. Solbrig, O. T. 1976. The origin and floristic affinities of tlie Soudi American temperate desert and semidesert regions, in Goodall, D. W. (ed.). Evolution of desert biota. Univ. Texas Press, Austin: 7-49. Truer, L. 1973. Bones, frogs, and evolution, in Vial, J. L. (ed.). Evolu- tionary biology of the anurans: contemporary research on major prob- lems. Univ. Missouri Press, Columbia: 65-132. TscHUDi, J. 1838. Classification der Batrachier. Mem. Soc. Sci. Nat. Neuchatel, 5(2):l-99. Van Der Hammen, T. 1961. The Quaternary climatic changes of northern South America. Ann. New York Acad. Sci., 95:673-683. \^\x Der Hammen, T., Gonzalez, E. 1960. Upper Pleistocene and Holocene climate and vegetation of the "Sabana de Bogota." Leiden Geol. Mededel., 25:262-315. Vanzolini, p. E., Williams, E. E. 1970. South American anoles: the geo- graphic differentiation and e\()lution of the Anolis chiysolcpis species group (Sauria, Iguanidae). Arch. Zool. Univ. Sao Paulo, 19:1-298. Vellard, J. 1960. El genero Pleurodema en los Andes Peruanos. Mem. Mus. Hist. Nat. Javier Prado, 10(7):1-12. Veloso, a., Galleguillos, R., Diaz, N. 1973. Karyotypic analysis of allo- patric populations of Pleurodema thanl (Lesson). Amphibia, Leptodac- tylidae. Caryologia, 26:69-76. Veloso, A., Iturra, P. 1976. Chromosome evidence for different phyletic lines in the frog genus £(//»op/2!/.v (Amphibia: Leptodactylidae ) . Herp. Rev., 7:99-100 (abstract). 46 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY VuiLLEUMiEH, B. S. 1971. Pleistoceiie changes in the fauna and flora of South America. Science, 173:771-780. Wagner, W. H., Jr. 1961. Problems in the classification of ferns, in Recent advances in botany. Univ. Toronto Press, Toronto: 841-844. Webb, R. C, Greer, J. K. 1969. Amphibians and reptiles from Malleco Province, Chile. Publ. Mus. Michigan St. Univ., 4:193-226. Wolfe, J. A. 1971. Tertiary climatic fluctuations and metliods of analysis of Tertiary floras. Paleogeog. Paleoclim. Paleoecol., 9:27-57.