  OCCASIONAL PAPERS of the MUSEUM OF NATURAL HISTORY The University of Kansas Lawrence, Kansas
A SYNOPSIS OF NEOTROPICAL HYLID FROGS, GENUS OSTEOCEPHALUS By [1] Research Associate, Division of Herpetology, Museum of Natural History, University of Kansas. [2] Curator, Division of Herpetology, Museum of Natural History, University of Kansas. When we initiated a study of the herpetofauna at Santa Cecilia in Amazonian Ecuador in 1966, we were immediately confronted with many kinds of animals that we could not identify with the existing literature. Comparisons of our specimens with those preserved in other museums resolved some of the problems, but many identifications could be made only after study of type specimens; even then some determinations remained questionable. We now find that in order to prepare a meaningful account of the herpetofauna of Santa Cecilia, we must complete several taxonomic studies, the limits of which extend far beyond eastern Ecuador. Because of our interests in hylids we have begun our studies on these frogs. One of us (Trueb, 1970a) studied the cranial osteology of casque-headed hylid frogs and redefined the genus Osteocephalus but did not determine the number of species in the genus. Our work in Amazonian Ecuador resulted in the discovery of the sympatric occurrence of three species at each of two localities; one of these species was found with a fourth species at another locality. Study of museum specimens confirmed the recognition of these four species in the Amazon Basin and lower Amazonian slopes of the Andes. A fifth species from Bolivia and PerÚ also is included in the genus. Examination of museum specimens has provided data on the geographic variation in, and distribution of, the five species. However, our conclusions pertaining to some populations need substantiation, because we have been hampered by inadequate material from areas beyond Ecuador. More than half of the 905 specimens of Osteocephalus are from Ecuador, a relatively small part of the total range of the genus. In this paper we are presenting a taxonomic review of the genus Osteocephalus; of necessity our study has been at the alpha level. We have utilized all of the usual external characters, as well as osteological features in our definitions of the species. Tadpoles and mating calls are available for only one species, O. verrucigerus (Trueb and Duellman, 1970); these and other important systematic characters, such as karyotypes, are not available for the group at this time. Our tendency has been to take a conservative view of species; thus it is doubtful that any subsequent worker will recognize fewer species in the genus. Our observations on these frogs in Amazonian Ecuador are presented in a final section of this paper. ACKNOWLEDGMENTS For the loan of specimens or for the provision of working space in their respective institutions, we are indebted to James E. BÖhlke, Werner C. A. Bokermann, F. W. Braestrup, Nelly Carrillo de Espinoza, Osvaldo R. da Cunha, Josef Eiselt, M. J. Fouquette, Jr., Alice G. C. Grandison, Jean GuibÉ, Birgitta Hansson, Walter Hellmich, M. J. Hoogmoed, Robert F. Inger, Konrad Klemmer, Jean Lescure, Alan E. Leviton, Clarence J. McCoy, Robert H. Mount, Charles W. Myers, Umberto Parenti, GÜnther Peters, James A. Peters, William F. Pyburn, Juan A. Rivero, Dorothy M. Smith, Paulo E. Vanzolini, Greta Vestergren, David B. Wake, Charles F. Walker, Ernest E. Williams, and Richard G. Zweifel. Study of specimens in European museums was made possible by a grant (No. 5063) from the Penrose Fund of the American Philosophical Society. Field work in Ecuador was partially supported by grants from the Watkins Fund of the Museum of Natural History, University of Kansas. At our base camp at Santa Cecilia, Ecuador, we enjoyed the hospitality of Ing. Ildefonso MuÑoz B. Transportation in Ecuador was generously provided by the Texaco Petroleum Company. During the course of our field work Stephen R. Edwards and Thomas H. Fritts contributed directly to our study of Osteocephalus. Michael J. Tyler of the South Australian Museum provided information on the vocal sac structure. We extend our sincere thanks to all of these persons for their contributions to our endeavors. MATERIALS AND METHODS We have examined 893 preserved frogs, including the type specimens of all included nominal taxa, 8 skeletons, 1 lot of eggs, and 3 lots of tadpoles that we refer to the genus Osteocephalus; in addition skulls were removed from five preserved specimens, and radiographs were made of 12 other preserved specimens. We have been fortunate in seeing living individuals of all species, except O. pearsoni, but we have colored photographs of a living specimen of that species. Figures 1 and 2 were drawn from projected colored transparencies of living frogs. Terminology follows that of Duellman (1970b). On the distribution maps solid symbols indicate localities from which we have examined specimens; open symbols represent additional locality records based on the literature. Throughout the text specimens are listed by their catalogue numbers preceded by the appropriate museum abbreviation, as follows:
HISTORICAL RESUMÉ Because of the taxonomic confusion that has surrounded the generic name Osteocephalus and two of the species (and their synonyms), we present a brief resumÉ of the taxonomic history of the group. Among the amphibians sent to the MusÉum National d'Histoire Naturelle in Paris by a Monsieur Leprieur in French Guiana was a single female specimen of a moderately large hylid having a well-ossified skull and smooth dorsal skin. This specimen escaped from the covetous eyes of Johann Tschudi, who prematurely named several species on the basis of specimens in Paris, and survived without an epithet until DumÉril and Bibron (1841) proposed for it the name Hyla leprieurii. The description of the species is fairly detailed, but the specimen was not illustrated. This is the earliest trivial name now associated with Osteocephalus. Fitzinger (1843) in his generic synopsis of amphibians and reptiles proposed the generic name Osteocephalus but did not associate a specific name with the genus. Consequently, Osteocephalus Fitzinger, 1843, is a nomen nudum. Franz Steindachner followed Leopoldo Fitzinger at the Naturhistorisches Museum in Vienna, where he had access to Fitzinger's notes and, of course, the important collections housed in that museum. Steindachner (1862) named two species of Osteocephalus on the basis of Brasilian specimens collected by Johann Natterer. Both species were named in the same publication; O. taurinus appeared on page 77, and O. favolineatus, on p. 80. This is the earliest association of the generic name Osteocephalus with a specific name and a description, both of which satisfy the Code of Zoological Nomenclature for generic availability. Therefore, Steindachner is the authority for the generic name Osteocephalus, which has O. taurinus as the type species by original designation. It is not possible to determine whether or not Steindachner's usage of Osteocephalus was the same as that intended by Fitzinger 19 years earlier. Steindachner (1862) gave reasonably good descriptions of his two new species and provided excellent illustrations of the two specimens, both large females. Apparently impressed by the similarities between Trachycephalus nigromaculatus Tschudi, 1838, and Osteocephalus taurinus, Steindachner (1867) used the combination Trachycephalus (Osteocephalus) taurinus. This ambiguous usage for the 1860's precludes our determining if Steindachner was in effect synonymizing Osteocephalus with Trachycephalus or whether he was placing Osteocephalus in a subgeneric status. Steindachner (1867) did not mention O. flavolineatus; perhaps by that time he had concluded that flavolineatus was only a color morph of taurinus. Cope (1867) placed Hyla leprieurii in the genus Hypsiboas Wagler, 1830. Cope (1874) named Osteocephalus planiceps from Nauta, PerÚ. The single specimen was among the collections made by the Orton Expedition to the upper Amazon Basin and was deposited in the Academy of Natural Sciences in Philadelphia. Boulenger (1882) placed both Osteocephalus and Trachycephalus in the synonymy of Hyla; he recognized Hyla taurina (with O. flavolineatus as a synonym), H. leprieurii, and H. planiceps. In the same publication Boulenger named Hyla buckleyi on the basis of 10 specimens in the British Museum from Ecuador; in the description he stated that buckleyi was like leprieurii and taurinus in having paired lateral vocal sacs. Boulenger held a lasting influence on taxonomic herpetology, and his generic synonymy of Osteocephalus was unchallenged until only a decade ago. Goin (1961) presented a generic synopsis of the genera of hylid frogs, in which he recognized Osteocephalus and stated: "There are perhaps eight or ten species of this genus in South America. Certainly taurinus, britti, leprieuri, buckleyi and pearsoni belong here. O. planiceps is surely a synonym of leprieuri and I believe that garbei is as well. The status of such forms as macrotis, riopastazae, and depressa has not yet been settled." Goin defined Osteocephalus as follows: "Males with paired vocal pouches, one at each angle of the jaw; derm of head not co-ossified with skull but roof of skull exostosed." Trueb (1970a) elaborated on Goin's definition and assuredly included only O. taurinus and O. leprieurii in the genus. Goin's inclusion of buckleyi, britti, and pearsoni in Osteocephalus was the first association of any of these names with that genus. Duellman (1970a) demonstrated that Garbeana garbei Miranda-Ribeiro, 1926, was a member of the Hyla rubra group. Hyla macrotis Andersson, 1945, is a Phrynohyas. Trueb and Duellman (1970) determined that Hyla verrucigera Werner, 1901, is the earliest name for an Osteocephalus displaying striking sexual dimorphism in coloration and texture of the dorsal skin; Hyla riopastazae Andersson, 1945 (female holotype), and Hyla orcesi Funkhouser, 1956 (male holotype), were placed in the synonymy of Osteocephalus verrucigerus. Hyla pearsoni Gaige, 1929, is a small species of Osteocephalus. Our findings substantiate Goin's suggestions relative to two other taxa. Hyla leprieurii britti Melin, 1941, from the Rio UaupÉs, Brasil, and Hyla depressa Andersson, 1945, from the RÍo Pastaza watershed, Ecuador, are members of the genus Osteocephalus, but both are synonyms of earlier names—leprieurii and taurinus, respectively. Another name proposed by Melin (1941), Hyla (Trachycephalus) vilarsi from TaracuÁ, Brasil, also is placed in the synonymy of O. taurinus. Cochran and Goin (1970) were unaware of the identities of Hyla verrucigera and riopastazae; they used the later name Osteocephalus orcesi for Colombian frogs that are correctly referred to O. verrucigerus. Although Goin (1961) placed Hyla buckleyi and H. pearsoni in Osteocephalus, Cochran and Goin (1970) recognized a "buckleyi group" in Hyla that included these two species plus a new species, Hyla cabrerai from Amazonian Colombia and Brasil (total of three specimens). Also, these authors named Hyla carri from a single Colombian specimen. Study of the types of Hyla cabrerai, H. carri, and H. festae Peracca, 1904, from Ecuador, reveal that all of these names are synonyms of Osteocephalus buckleyi. Much of the taxonomic confusion and multiplicity of trivial names is due to the great amount of color variation in taurinus and to the sexual dimorphism in the texture of the dorsal skin in all of the species. The details of variation in these and other characters and our justifications for the synonymies are given in the accounts of the species. All of the trivial names that apply to species herein recognized as members of the genus Osteocephalus are listed in table 1. Osteocephalus Steindachner, 1862 Osteocephalus Steindachner, 1862:77 [Type species.—Osteocephalus taurinus Steindachner, 1862, by original designation]. Not Osteocephalus Fitzinger, 1843:50 (nomen nudum). Diagnostic Definition.—1) Skull broader than long; 2) dermal roofing bones of skull well ossified, exostosed, and/or co-ossified in some species; 3) prenasal and internasal bones absent; 4) parasphenoid alae posterolaterally oriented; 5) dentigerous processes of prevomers angular (/— —\); 6) vocal sacs paired, posterior, and when inflated protruding posteroventral or posterolateral to angles of jaws; 7) submentalis muscle moderate in size and araphic; 8) intermandibularis muscle undifferentiated and bearing an elongate median aponeurosis; 9) parotoid glands absent or poorly developed, skin not producing viscous secretion characteristic of Phrynohyas; 10) skin on dorsum tuberculate in males, smooth in females; 11) tympanum large, 60 percent or more of diameter of eye; 12) fingers about one-third, toes more than three-fourths webbed; 13) discs large, round; 14) nuptial excrescences present in breeding males; 15) inner metatarsal tubercle not modified for digging; 16) outer metatarsal tubercle absent; 17) tarsal fold weak or absent; 18) pupil horizontal; 19) palpebrum clear; 20) known tadpoles having two upper and five lower rows of teeth.
Content.—As defined here, the genus contains five known species: O. buckleyi (Boulenger), O. leprieurii (DumÉril and Bibron), O. pearsoni (Gaige), O. taurinus Steindachner, and O. verrucigerus (Werner). Distribution.—The Guianas and Amazon Basin; also in the upper Orinoco and Magdalena drainages. Most localities are at elevations below 500 m, but the genus ascends the Amazonian slopes of the Andes to elevations of about 1800 m. Analysis of Characters Size and Proportions.—Frogs of the genus Osteocephalus are moderate to large hylids. The largest species is taurinus, attaining a snout-vent length of 103.1 mm; the smallest is pearsoni, which attains a length of 54.7 mm. Considerable intraspecific geographic variation occurs in adult size, especially in taurinus. Females of all species attain a noticeably larger size than males, but no significant differences are apparent in proportions (Table 2). Table 2.—Comparison of Size and Proportions in the Species of Osteocephalus. (Means are given in parentheses below observed ranges)
Coloration.—All Osteocephalus are predominantly brown frogs usually with some darker dorsal markings (Figs. 1 and 2). Osteocephalus verrucigerus has a nearly uniform dark brown dorsum and no distinct transverse bars on the limbs, whereas all of the other species have distinct bars on the limbs. The dorsal markings on the body consist of irregular blotches in buckleyi, pearsoni, and taurinus but are narrow transverse marks in leprieurii. A narrow middorsal cream or yellow stripe is present in some individuals of buckleyi and taurinus but absent in all individuals of the other species. The flanks are uniform pale tan in leprieurii and uniform reddish brown in verrucigerus; in the other species the flanks are cream to brown with dark brown or black spots (also dark diagonal marks in some buckleyi). A creamy white anal stripe is present in some specimens of leprieurii but absent in all individuals of other species. The postocular region, encompassing the tympanum, is dark brown in most specimens. In adults of pearsoni and taurinus the upper lips are dark brown. A pale cream or tan suborbital spot is present in pearsoni and in some taurinus; in some specimens of taurinus the suborbital spot is expanded posteriorly forming a labial stripe on the posterior part of the lip. The labial markings of verrucigerus are similar to the latter pattern, except that in females a distinct, light labial stripe extends the length of the lip. Osteocephalus leprieurii has a distinct, broad, pale labial stripe. The lips are barred cream and dark brown in buckleyi. The venter is uniform creamy white or pale tan in leprieurii, uniform white in some buckleyi (most males), and uniform tan in some taurinus. The other species and some individuals of taurinus and buckleyi (most females) have dark ventral markings. These markings are most distinctive in verrucigerus, in which the venter is white with bold black mottling and spots (Fig. 3c). Those individuals of taurinus having ventral markings usually have indistinct, diffuse brown spots on the throat and chest (Fig. 3b). Osteocephalus pearsoni is characterized by a fine brown reticulation on the venter and on the anterior and posterior surfaces of the thighs in adults (Fig. 3a). Individuals of buckleyi that have ventral markings vary between the patterns illustrated for pearsoni and taurinus (Figs. 3b and c). Ontogenetic change in coloration is slight or non-existent in buckleyi, pearsoni, and taurinus, except that juveniles lack ventral markings. A dark blotch on the back and distinct transverse bars on the limbs are evident in juveniles of verrucigerus; these markings are obscured in the adults. Juveniles of leprieurii are olive-brown with yellow dorsolateral stripes; the transverse dark marks, characteristic of the adults, appear before the stripes are lost.
Skin.—The dorsal skin of all males of Osteocephalus is tuberculate to varying degrees, whereas the dorsal skin of females is smooth, or nearly so (Fig. 4). Osteocephalus verrucigerus differs from other members of the genus by the presence of numerous, large tubercles bearing keratinized tips. The tubercles of leprieurii are numerous and spinous but much smaller than those of verrucigerus; those of taurinus are spinous but less numerous than in leprieurii. Osteocephalus buckleyi has a mixture of large and small, non-spinous tubercles, and pearsoni has only a few, small, scattered, non-spinous tubercles. Fleshy tubercles occur on the eyelids and supratympanic fold in females of buckleyi; a few small tubercles are present on the back of females of pearsoni, whereas the dorsal skin in females of the other species is smooth. The skin on the flanks of both sexes of buckleyi is weakly areolate; in the other species the flanks are smooth. The skin on the top of the head in taurinus is rugose as a consequence of co-ossification. In all species the anal opening is directed posteriorly at the upper level of the thighs. Hands and Feet.—The feet of Osteocephalus are fully webbed or nearly so. Webbing between fingers one and two is basal in all species. Webbing between fingers two, three, and four is most extensive in taurinus, in which the three fingers are about one-half webbed (Fig. 5). Osteocephalus buckleyi, pearsoni, and verrucigerus have reduced webbing between fingers two and three, and leprieurii has reduced webbing between fingers two, three, and four. All members of the genus have well-developed subconical subarticular tubercles on the fingers and toes; there is a tendency for the distal tubercle on the fourth finger to be weakly bifid. Palmar and plantar supernumerary tubercles are well developed in taurinus, moderately developed in buckleyi, leprieurii, and pearsoni, and barely evident in verrucigerus. All of the species have a noticeable fold on the wrist and enlarged prepollices, bearing horny nuptial excrescences in breeding males. The prepollex is least enlarged in buckleyi. Outer metatarsal tubercles are absent. The inner metatarsal tubercle is moderately well developed and ovoid in leprieurii and pearsoni; it is elliptical and flat in the other species. Tarsal folds are absent in all species except verrucigerus, in which the folds are barely evident.
Cranium.—As a genus, the cranial structure is remarkably uniform and quite generalized when viewed in the context of the family Hylidae. The skulls are broad and relatively flat, each being only slightly more broad than long and about one-third as high as long. In dorsal aspect the snouts are broadly rounded; the snout of buckleyi is somewhat less rounded and appears to be slightly longer than the snouts of other species. This subtle difference relates to the relative narrowness of the premaxillaries in buckleyi.
The genus is characterized by well-developed dermal roofing bones and an unusually large exposure of the sphenethmoid dorsally (Fig. 6). The conformation of the sphenethmoid exposed dorsally is determined by the marginal positions of the adjacent, overlapping elements—the nasals and frontoparietals. Medially the nasals overlap the lateral margins of the sphenethmoid. Anteromedially, the nasals converge in leprieurii and taurinus, are narrowly separated in buckleyi and pearsoni, or are more widely separated in verrucigerus. In all species the nasals terminate at the anterodorsal corner of the orbit. The frontoparietals of buckleyi, leprieurii, and taurinus have an anterolateral extension, which marginally overlaps the dorsolateral edge of the sphenethmoid and articulates with the posterodorsal corner of the nasal in buckleyi and taurinus; the bones are narrowly separated in leprieurii. The frontoparietals of pearsoni and verrucigerus have more extensive median ossification and less extensive anterolateral ossification. Thus, in those species the posteromedian portion of the sphenethmoid is obscured, and the lateral margins are partly exposed. The frontoparietal fontanelle is completely covered in all species, except buckleyi and leprieurii, in which an irregular, median separation of the frontoparietals exposes a small portion of the fontanelle. The posterolateral margins of the frontoparietals lie medial to the epiotic eminences. Dermal ornamentation, involving the nasals, frontoparietals, and sphenethmoid, occurs in taurinus and, to a limited extent, in pearsoni. In the latter species marginal portions of the frontoparietals, the dorsal surfaces of the nasals, and the posteromedial part of the exposed sphenethmoid are slightly exostosed, resulting in an open, reticulate pattern of dermal sculpturing of exceedingly low relief (Fig. 6c). Osteocephalus taurinus is characterized by casquing, co-ossification, and a rather intricate pattern of dermal sculpturing, which was described in detail and illustrated by Trueb (1970a). The squamosals of all species are moderately large structures having otic plates that overlie the lateral portion of the cristae paroticae. The posterior rami are short; the zygomatic rami of all species, except taurinus, extend slightly more than one-half of the distance to the maxillary. In taurinus the zygomatic ramus extends nearly to, or articulates with, the maxillary. The maxillary arches are complete and relatively robust. The alary processes of the premaxillaries are vertically oriented in leprieurii, pearsoni, and taurinus and very slightly inclined posteriorly in buckleyi and verrucigerus. The maxillaries are uniformly characterized by the absence of postorbital processes and by the presence of preorbital processes that articulate with the maxillary processes of the nasals. The partes facialae of the maxillaries are moderately developed in all species, except taurinus, in which the pars fascialis tends to articulate with the lateral margin of the nasal in well-ossified individuals. The partes palatinae are poorly developed in all species, except buckleyi, in which the pars palatina of the premaxillary is moderately robust. The pterygoids are uniformly tri-radiate structures. The anterior rami terminate at about the mid-level of the orbit, and the medial rami articulate firmly with the anterolateral corner of the otic capsule. The palatines are well-developed elements bearing ventral ridges; the ridges are somewhat irregular in buckleyi, taurinus, and verrucigerus but smooth in leprieurii and pearsoni. The parasphenoids are large elements characterized by acuminate cultriform processes and posterolaterally inclined alary processes. The basal areas of the cultriform processes bear small odontoid protuberances in buckleyi, taurinus, and verrucigerus, whereas they are smooth in leprieurii and pearsoni. The prevomers are remarkably uniform, moderately well-developed structures. In each species the anterior ramus lies adjacent to the premaxillary, and the lateral wings form the anterior, medial, and posteromedial margins of the internal nares. The dentigerous processes are characteristically large and angular and bear numerous teeth. The sphenethmoid and otoccipitals are well ossified; a dermal sphenethmoid is present only in taurinus.
Vertebral Column.—The cervical cotyles are uniformly widely displaced. The neural arches are low and non-imbricate. The transverse processes of the third presacral vertebrae are approximately equal in width to the sacral diapophyses in all species, except buckleyi, in which the processes of the third presacral are slightly narrower than the diapophyses. Osteocephalus buckleyi is further distinguished by the presence of narrow transverse processes on presacrals five through eight (Fig. 7b); males have narrower processes than do females. The processes are moderately wide but subequal in width in pearsoni, taurinus, and verrucigerus, whereas they are nearly equivalent in width to one another and to the sacral diapophyses in leprieurii (Fig. 7a). The sacral diapophyses of all species are moderately dilated and posterolaterally inclined. The coccyx bears a distinct dorsal ridge and has a bicondylar articulation with the sacrum. Pectoral Girdle.—The pectoral girdles are fully arciferal and bear small, cartilaginous omosterna and moderately large cartilaginous sterna. The coracoids are robust, and the clavicles are strongly arched. Procoracoid cartilage seems to be absent. The scapulae are large, longer than the clavicles, and bicapitate proximally. The suprascapulae are moderately large and well ossified in leprieurii and taurinus. The suprascapula of verrucigerus is poorly ossified, and that of buckleyi is not ossified. Pelvic Girdle.—The ilia of buckleyi, taurinus, and verrucigerus lack any indication of a crest on the shaft, whereas leprieurii has a low crest. The dorsal acetabular expansion of the ilia is moderately low in taurinus and verrucigerus, but distinctly lower in buckleyi and leprieurii. The ilia of all species bear low dorsal protuberances. The ischia of leprieurii, taurinus, and verrucigerus are moderately expanded; that of buckleyi is somewhat less expanded dorsally. The pubis of leprieurii, taurinus, and verrucigerus are calcified, whereas that of buckleyi remains cartilaginous. Throat Musculature and Vocal Sac Structure.—Tyler (1971) described the throat myology of Osteocephalus. The genus is characterized by a moderate-sized araphic submentalis muscle and an undifferentiated intermandibularis having an elongate median aponeurosis. The intermandibularis and submentalis are completely independent in buckleyi, whereas in the other species there is a small attachment between these muscles. According to Tyler (pers. com.), Osteocephalus has three distinctive types of vocal sac structure which result from differences in the development of the interhyoideus muscle and the overlying skin. In leprieurii and verrucigerus the supramandibular portions of the interhyoideus form a simple tubular, posterolateral extension; there is no modification of the associated skin. Osteocephalus buckleyi and pearsoni have more extensive development of the supramandibular portions of the interhyoideus; furthermore, the associated skin forms a broad, loose fold extending from the ventromedial surface of the throat dorsolaterally to the base of the supratympanic fold. Like buckleyi and pearsoni, the supramandibular portion of the interhyoideus is much expanded in taurinus. The vocal sac structure of taurinus differs from that of other members of the genus in that the skin of taurinus forms an everted pouch, which dangles loosely beneath the supratympanic fold. Key to the Species of Osteocephalus | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|