The Development of the Chondrocranium of Typhlonectes compressicaudus (Gymnophiona), with Comparison to Other Species

The Development of the Chondrocranium of Typhlonectes compressicaudus (Gymnophiona), with Comparison to Other Species Marvalee H. Wake; Jean-Marie Exbrayat; Michel Delsol Journal of Herpetology, Vol. 19, No. 1. (Mar., 1985), pp. 68-77. Stable URL: http://links.jstor.org/sici?sici=0022-1511%28198503%2919%3a1%3c68%3atdotco%3e2.0.co%3b2-5 Journal of Herpetology is currently published by Society for the Study of Amphibians and Reptiles. Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/ssar.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org. http://www.jstor.org Sun Sep 9 14:54:40 2007

1ournal of Herpetology, Voi. 19, No. 1, pp. 68-77, 1985 Copyright 1985 Society for the Study of Amphibians and Reptrles The Development of the Chondrocranium of Typhlonectes compressicaudus (Gymnophiona), with Comparison to Other Species MARVALEE H. WAKE,' JEAN-MARIE EX BRAY AT,^ AND MICHELDELSOL~ 'Department of Zoology and Museum of Vertebrate Zoology, University of California, Berkeley, California 94720, USA and 2Laboratoire de Biologie Generale de la Faculte Catholique des Sciences de Lyon et Laboratoire d'etude du Developpement des Vertebres lnferieurs de 1'Ecole Pratique des Haut Etudes, 25, rue du Plat, 69288 Lyon Cedex 02, France Ass~~~cr.-The development and degeneration of the chondrocranium of Typhlonectes compressicaudus (Typhlonectidae: Gymnophiona: Amphibia) is described and compared with data for other gymnophione species. The typhlonectid pattern exemplifies many of the constraints typical of gymnophiones (emphasis on otic and nasal capsule development, late development of stapes, retention of cartilaginous nasal capsule rims), but has a number of unique features as well, especially extensive lateral vaulting of the braincase, the absence of a cartilaginous mesethmoid component and its precursors, and concomitant modification~ of nasal capsule development. Other aspects of development are apparently correlated with both the live-bearing mode (early mineralization of tooth crowns and jaw articulation surfaces) and with the aquatic habits of the species (reduced overlap of dermal elements). Information is slowly accruing about formed the basis of assessment of the the developmental morphology of the relationships of caecilians to other amskull of caecilians. Much of our current phibians and to other vertebrates (see knowledge is based on only a few stages De Beer, 1937). Recently, however, for diverse taxa. One hundred years ago, Wake and Hanken (1982) described Sarasin and Sarasin (1887-90) and Peter skull development in the Central (1898) examined the Sri Lankan Ich- American caeciliid Dermophis mexicanus thyophis glutinosus, and provided much from a virtually complete series, allowinformation about the structure of the ing more extensive comparison among skull. Ramaswami (1941)also examined caecilians and among amphibians in I. glutinosus, and in addition provided general. crucial information about another Recent work in French Guyana has member of the family Ichthyophiidae, provided extensive material represent- Uraeotyphlus narayani, as well as an ad- ing an additional family of caecilians, ditional Indian form, the caeciliid Ge- the Typhlonectidae. Typhlonectids are geneophis carnosus. Marcus and his stu- aquatic, restricted to northern and cendents, especially Marcus et al. (1935) tral South America, and are live-beardescribed development in the Sey- ers, so as with Dermophis mexicanus, colchelles caeciliids Hypogeophis rostratus lection of pregnant females at different and Grandisonia alternans, including the times of the year has yielded embryos ossification sequence. A major problem and fetuses at a number of stages of dewith these analyses is that they are velopment. Further, due to the work of based on very incomplete series, rarely Delsol et al. (1981), Exbrayat (1983),and more than two or three stages of devel- Exbrayat et al. (1981, 1982) on the reopment from which information about productive biology of Typhlonectes comchondrocranial or bone developmental pressicaudus and that of Wake (1977a, b, sequences can be gathered. Yet, the data 1980, 1982) on reproductive biology and of the Sarasins, Peter, and Marcus et al. development in diverse live-bearing

CHONDROCRANIUM OF GYMNOPHIONA 69 species, we are able to assess the development of T. compressicaudus with greater understanding of caecilian biology. The comparison of development in livebearers from two different families should give information about the conservativeness and/or the diversity of developmental pattern in caecilians. Sixteen specimens of Typhlonectes compressicaudus collected in French Guyana in 1979 and 1980 were cleared and differentially stained for bone and cartilage (Hanken and Wassersug, 1981). The series includes 10 embryos and fetuses (16, 22, 32, 42, 45, 55, 65 [2], 73, and 140 mm total length [TL]), two newborns (95 and 125 mm TL), two juveniles (194 and 208 mm TL), and two adults (307 and 362 mm TL). Unfortunately, all specimens except the newborns were fixed in Bouin's solution. Bone was therefore demineralized, so we cannot assess the sequence of ossification in this material. However, the chondrocrania and the hyoid apparatus are well stained, and some indications of bone development are present. We therefore are able to examine chondrocranium and hyoid apparatus development and the degeneration of the chondrocranium as it is replaced by bone, and to compare this species with data for others. Further, the newborns provide information about the state of 0ssification at birth, and one of us (Wake) has fetuses and adults of this species from Amazonian Colombia and Brazil which provides additional information about ossification. DESCRIPTION 16 mm TL (Stage 111-2 ldelsol et al., 19811).-The chondrocranium is 2.7 mm long (Fig. 1A). Trabeculae and parachordals are thin rods of cartilage, fused at their juncture. Parachordals are united posteriorly by a thin parachorda1 plate, which has an anterior projection under the brain that encloses the anterior end of the notochord. Preoptic, postoptic, and antotic pilae join the trabeculae and parachordals to the orbital cartilage, which is united posteriorly with the otic capsule. The orbital cartilage has an anterior projection, and the trabeculae a bifurcated anterior projection, but the nasal capsules have not yet formed. The otic capsules are very thin, the lateral rims only of thickened cartilage. A pair of occipital arches is well developed, and the otic capsules are weakly attached to them. The palatoquadrate is an irregularly shaped vertical rectangle, with indications of ascending and posterior processes. The stapes is a small cartilaginous rod, slightly free of the quadrate. Meckel's cartilage is stout, with a retroarticular process that is approximately one-fourth the length of the cartilage, and anteromedially the cartilage forms a large boss. Five to seven rows of tiny teeth are aggregated on the boss. Elements of the hyoid apparatus are present but unfused. A v-shaped basihyal is present; a pair of ceratohyal bars are developing beside, but not attached to, it. Posteriorly two pairs of ceratobranchial bars are initiated. They are widely separated medially. 22 mm TL (Stage 111-4).-The chondrocranium is much more fully developed. All of the structures described above are stouter, of heavier cartilage (Fig. lb, C). The chondrocranium is 3.7 mm long. The otic capsules are better developed, and the medial walls are forming. Most noteworthy is the state of development of the nasal capsules (Fig. 2A). The anteromedial union of the trabeculae has formed the ethmoid plate, now a flat structure ventral to the forebrain. From it project the nasal capsules. The cartilagi obliquae form thin lateral rims, and the medial rim is irregular. The cupola anterior is incomplete dorsally and lies somewhat behind the strut forming the ventromedial rim of the cavsule. The solum nasi is stout, and from it the ventral floor of the capsule, the lamina transversalis, extends. The capsules are separate, so there is no septim nasi or

M. H. WAKE ET AL. FIG. 1. Camera lucida drawings of developing chondrocrania of T. compressicaudus.a. Lateral view of chondrocranium of 16 mm total length (TL) (stage 111-2, Delsol et al., 1981) embryo. B. Lateral view of chondrocranium of 22 mm TL (stage 111-4) embryo. C. Dorsal view of chondrocranium of 22 mm TL embryo. D. Lateral view of chondrocranium of 42 mm TL (stage 111-5) embryo. E. Dorsal view of chondrocranium of 42 mm TL embryo. Abbreviations for all Figs.: a = articulating facet of the palatoquadrate; ap = antotic pila; apr = ascending process; ca = cupola anterior; cob = cartilago obliqua; ep = ethmoid plate; f = frontal; ft = fetal teeth; If = lateral flange; It = lamina transversalis; m = Meckel's cartilage; mr = Meckel's cartilage remnant; mw = medial wall; mx =maxilla; nc = nasal capsule; npm = naso-premaxilla; nr = notochordal remnant; o = otic process; oa = occipital arch; oar = occipital arch rudiment; oc = otic capsule; or = orbital cartilage; orb = orbital cartilage remnant; p = parachordal; pn = pares nasi; pnp = prenasal process; pp = postoptic pila; pq = palatoquadrate; pr = preoptic pila; prt = parietal; ptp = pterygoid process; q = squamosal; s = stapes; ser = sphenethmoid remnant; sm = septomaxilla; t = trabecula; tt = trabecular plate; vmc = ventro-medial wall cartilage. Bar in all Figs. = 1 mm.

CHONDROCRANIUM OF GYMNOPHIONA FIG.2. Comparison of caecilian nasal capsule development. A. 22 mm TL T. compressicnudus. 8. 25 mm TL Dermophis mexicanus. Note particularly the incomplete ventral floor of the capsules and the absence of the prenasal process in T. compressicaudus. The condition in D. mexicnnus is typical of embryos of most species examined. See text, and Fig. 1 for abbreviations. prenasal process. The palatoquadrate has assumed definitive form (Fig. 3A). It is a three-dimensional structure rather than a flat plate of cartilage. The ascending process curves dorsally and laterally, and has a lateral blade. The pterygoid process is pronounced. A descending process provides a more complex articulation with Meckel's cartilage than in the earlier stage. The stapes is continuous with the stapedial process of the palatoquadrate. The palatoquadrate is fused to the anterior end of the otic capsule near the top of the ascending process. Meckel's cartilage is stouter and the articulating facet and the anterior boss are enlarged. Numerous rows of fetal teeth are arrayed in a plate on the boss, and the plate is beginning to extend posteriorly on the cartilage. The hyoid apparatus is also much larger and more FIG. 3. Palatoquadrate shape change in early development. A. 22 mm TL embryo of T. compressicaudus; 8.32 mm TL (stage 111-5) embryo of T. compressicnudus. Maximal development has occurred at 22 mm; however, the palatoquadrate is presumed functional in intra-oviducal feeding until and after the quadrate has mineralized. See text, and Fig. 1 for abbreviations. nearly fused compared to the condition in the 16 mm specimen. The ceratohyal and ceratobranchial rods are longer, and ceratobranchial I (CB I) is stouter and of more densely staining cartilage than the other elements. The ceratohyals are not yet fused to the basihyal, and CB I and I1 are not fused medially. Ceratobranchial I11 (the fourth element of the apparatus) is fused medially, and is a flat, thick plate of cartilage. At its posteromedial end, it has incorporated CB IV, which extends medially. 32 mm TL (Stage 111-5).-This stage is notable for the more extensive development of the nasal capsules, and the inception of some degeneration of cartilaginous components of the chondrocranium, indicating the inception of both endochondral and dermal bone. The nasal capsules have added cartilage, especially to the medial wall and to the ventral and lateral components. The increase in size and forward projection of the nasal capsules, coupled

72 M. H. WAKE ET AL. with the enlargement of the otic capsules and the general growth of the head, results in a more elongate chondrocranium, in contrast to the rather round crania of the preceding stages. The parachordal plate is changed in shape, forming a rim around the notochordal extension, but reducing otherwise, except for the completion of the ventral floor between the otic capsules and the bases of the occipital arches. There is some degeneration of the orbital cartilage and trabecula at the level of the orbit. The palatoquadrate has begun to degenerate, and is free of the otic capsule (Fig. 3B). The stapes in this specimen is free on the left side, fused to the palatoquadrate on the right (see Discussion). Meckel's cartilage is degenerated at its articulation with the palatoquadrate and in the retroarticular process. This probably indicates replacement by bone at the articulation, similar to the condition reported by Wake and Hanken (1982) in Dermophis. The hyoid apparatus is little changed from that of the 22 mm specimen. The chondrocranium is 4.5 mm long. 42 mm TL (Also Stage 111-5).-The anterior end of the chondrocranium continues to grow, whereas elements elsewhere begin to degenerate (Fig. ID, E). The nasal capsules are virtually enclosed, open only dorsally and at the narial apertures. There is also a slight opening at the juncture of the lamina transversalis and the cartilago obliqua. The anterior rims of the capsules are of heavy cartilage. A flange of cartilage extends laterally from the orbital cartilage and the nasal capsule, roofing the orbit. The lateral wall of the braincase from the trabecula to the orbital cartilage and the nasal capsule is also formed of thin cartilage, so the anterior part of the chondrocranium is extensively cartilaginous-more so than any other caecilian reported (see Discussion below). The palatoquadrate has continued to degenerate, but in this specimen the posterior edge of the right palatoquadrate is fused to the otic capsule. The palatoquadrate cartilage is heaviest where the stapes abuts it, but the stapes is free. Meckel's cartilage is more degenerate, but the tooth plate bears several rows of teeth. The elements of the hyoid apparatus are more elongate, but not fused. The chondrocranium is 4.8 mm long. 45 mm TL (Also Stage 111-5).-The chondrocranium continues to degenerate medially and posteriorly. The otic capsules remain intact, as do the nasal capsules and the adjacent cartilaginous structure. The palatoquadrate has degenerated, except for its dorsal rim. The stapes is a cartilaginous rod with a footplate, lying free in the window of the otic capsule. The parachordal plate is much reduced, and the occipital arches are reduced. The ceratohyals are tenuously fused to the basihyal, and the ceratobranchials are thicker cartilaginous plates, but only CB 111-IV is fused medially. The posterior margin of the latter element is filled in with cartilage, obscuring the medial arm that is CB IV. The chondrocranium is 4.6 mm long. 55 mm TL (Stage IV-I).-The specimen is poorly fixed, and staining is affected. However, the chondrocranium is much degenerated, only the nasal and otic capsules, bits of other anterior elements, and of Meckel's cartilage remaining. The stapes and the hyobranchial apparatus are also cartilaginous. 65 mm TL (Stage IV-2).-Two specimens show considerable variation. In one, the cartilage is degenerate, much like the specimen described above. In the other, chondrocranial structure is much more extensive, though degeneration is clearly advancing. The postoptic pila and the adjacent trabecularparachordal base and the orbital cartilage above are absent; the lateral flanges behind the nasal capsules and the medial walls of the braincase are reduced. Trabeculae and parachordals are reduced in thickness, as are the otic capsules and the posterior parts of the nasal capsules. The occipital arches and posterior ends of Meckel's cartilages are degenerated, and the parachordal plate

CHONDROCRANIUM OF GYMNOPHIONA 73 is represented only by a cartilaginous pad in the region of the notochordal extension. Random bits of cartilage occur in the otic capsules, and bits of the palatoquadrates are cartilaginous. Even in this demineralized specimen, alizarinstained bone remains in the articular component of the lower jaw and in the articular end of the quadrate, suggesting early and extensive mineralization (see Discussion). Ceratobranchials I are fused medially to the basihyal; the left ceratohyal is not yet fused. The chondrocranium is 5.2 mm long. 73 mm TL (Stage IV-2).-This specimen is similar to the more degenerate 65 mm specimen described above, and adds no additional information. 95 mm TL (Stage IV-3).-This newborn specimen was well fixed in neutral buffered formalin, so yields information about the state of ossification at birth, but herein we emphasize the state of the remaining chondrocranial components. The nasal capsules retain cartilaginous anterior rims that project beyond the naso-premaxillae-a condition that persists throughout life. Bits of cartilage remain at the anterior end of the lower jaw, suggesting a Meckel's process or mento-meckelian cartilage as in frogs. There is no cartilaginous prenasal component of the mesethmoid, as is present in many adult caecilians; this is simply a correlate of the fact that the nasal capsules are widely separated throughout their development, and a septum nasi and prenasal process never develop in this species (see Discussion below). Most of the dermal and endochondral bones are well formed, but some information about ossification stages can be determined. Neither the dorsal shelf of the maxilla nor the squamosal are complete. Large independent septomaxillae are present. The vomer also is not fully formed, particularly the posterior struts. There is not a free ectopterygoid. None of the dermal investing bones of the skull overlap at their borders, save a bit of the vomer and palatine with the complex os ba- sale. Fusions of naso-premaxillae and maxillo-palatine are extensive. The os basale, incorporating both endochondral and dermal components (see Wake and Hanken, 1982), is complete. A large temporal fossa between squamosal and parietal is present; this will close little more during adulthood. A massive plate of fetal teeth is present on the lower jaw of this newborn; a single row of germs of adult teeth is found on the maxillarynaso-premaxillary rim. The hyoid apparatus is cartilaginous, and the ceratal elements are broader. Ceratobranchial I1 is still not fused medially; the ceratohyal-basihyal fusion is still weak. The skull is 7.0 mm long. 125 mm TL (Stage IV-3).-The structure of the skull of this formalin-fixed larger newborn is very similar to that described above. There are some modifications of the dentition. The large fetal tooth plate is still present on the lower jaw, but composed of a reduced number of rows (8-10, rather than 11-14), and a row of small teeth of adult shape and arrangement is present lingually. Also, there has been more growth of some of the dermal bones, particularly the maxillary and the squamosal. The septomaxillary is large and not yet fused to the maxilla. The cartilages of the anterior ends of the nasal capsules are very large, especially laterally. The skull is 8.1 mm long. 140 mm TL (Stage IV-2).-This very large fetus is particularly instructive with regard to the state of the chondrocranium near birth. It is preserved in Bouin's, with resultant demineralization of bone. However, excellent staining of vestiges of cartilage allows a determination of the chondrocranial components that are last to break down, for they are not shielded by alizarinstained bone (Fig. 4). It is therefore clear that all of the posterior part of the chondrocranium, including the otic capsules and the stapes, is replaced by bone. Vestiges of the orbital cartilages remain anterior and posterior to the orbits, and there are bits of the trabecular-

M. H. WAKE ET AL. FIG. 4. Cartilaginous elements remaining in the skull of a near-term fetus of T. cornpressicaudus (140 mm TL; stage IV-2). The posterior part of the chondrocranium has been replaced by bone, and only vestiges of cartilage remain medially. A medial remnant of the ethmoid plate, the anterior rims of the nasal cartilages (which will remain cartilaginous throughout life), and the anterior component of Meckel's cartilage are cartilaginous in the anterior part of the skull. Investing bones are outlined. See text, and Fig. 1for abbreviations. parachordal base. There is a medial remnant of the cartilaainous " ethmoid plate. It does not have a vertical extension, however. The nasal capsules are much eroded posteriorly, and have irregular posterior borders. The anterior end of Meckel's cartilage remains cartilaginous as well. It is noteworthy that thislarge but unborn fetus has a fully adult dentition, with dentary and splenial rows on the lower jaw and maxillary-naso-premaxillary and vomeropalatine rows in the upper jaw. The hyoid cartilages are well formed, and CB I1 is finally fused medially in this 8.0 mm skull. Chondrocranial morphology in Typhlonectes compressicaudus is largely similar to that of other species for which there are data. This rather complete ontogenetic series allows comparison with Dermophis mexicanus (Wake and Han- rr ken, 1982), the only other species for which there are similarly complete data. As noted above, this also allows comparison of two live-bearing species that have fetal teeth but are in different families. Information on the chondrocranium, or elements of it, is available for a few other species. As noted above, development in lchthyophis glutinosus was commented upon by Sarasin and Sarasin (1887-90), Peter (1898), and by Ramaswami (1941). Further information about that species is provided by de Jager (1939) on the quadrate, and by Visser (1963) who examined skull morphology in I. glutinosus and I. monochrous. Visser had juvenile and adult specimens, and provided description of numerous elements of the skull, but did not State the sizes of his carefully compare cartilage reduction. Still. his work. varticularlv on the nasal I.I capsule, is useful. Ramaswami's work on Uraeotyphlus (1941) and on Gegeneophis (1948) has important information on chondrocrania, and that of Marcus et al. (1935) on Hypogeophis and Grandisonia remains basic. Els (1963) provided information on the nasal capsule of Schistornetopum thomensis as part of his description of the adult skull, and Brand (1956) did the same for Scolecomorphus uluguruensis. This work focusing on nasal capsules and quadrates provides for comparison among diverse species. Since much significant variation is seen in these structures among caecilians, we are pleased to add information about Typhlonectes compressicaudus. The chondrocranium of T. compressicaudus differs from that of other species in having rather more extensive nasal capsules and cartilaginous walls of the anterior part of the braincase, and in completely lacking a septum nasi and prenasal process throughout development (Fig. 2A). The ontogeny of the palatoquadrate and the stapes also differs in some respects from that reported for other species. We consider this analysis very tentative, however, for most of the data with which we are compar-

CHONDROCRANIUM OF GYMNOPHIONA 75 ing ours are based on structure in adults and/or juveniles, with the describers inferring ontogeny, or on data from only a very few developmental stages, rather than complete ontogenetic sequences. For example, the extensively cartilaginous anterior part of the chondrocranium is a transitory feature, found in fetuses between 40 and 60 mm TL. However, the only comparable ontogenetic data that exist are for Dermophis mexicanus, and comparable stages are lacking for other species for which there is any information about the anterior end of the chondrocranium. Therefore we really cannot say whether the extensive structure of T. compressicaudus is the more typical condition, or whether the lack of such structure as in D, mexicanus is the standard mode of development (if there is a standard mode). A number of features of the chondrocranium do characterize T. compressicaudus. The pronounced anteromedial boss of Meckel's cartilage and the fetal teeth arrayed as a plate composed of the fused pedicels of the teeth are typical of the genus (Wake, 1976, 1977a, 1978). The absence of a septum nasi, a prenasal process, and a lamina perpendicularis of the mesethmoid that extends between the nasal capsules all appear to be unique to the taxon. Brand (1956) makes a point of the lack of a prenasal process that extends anteriorly between the nasal capsules in Scolecomorphus uluguruensis, but that species has an elongate lamina perpendicularis, which never appears in T. compressicaudus. A prenasal process of the mesethmoid that remains cartilaginous throughout life characterizes D. mexicanus (Wake and Hanken, 1982) (Fig. 28) and many other species (Wake, unpubl.). Despite the extensive development of the nasal capsules during ontogeny, their degeneration, but for the anterior rims, is complete. This is in apparent contrast to the condition in adults of lchthyophis glutinosus (Visser, 1963) and Scolecomorphus uluguruensis (Brand, 1956), in which cartilago infranaria or cartilago obliqua and cartilago ectochoanali remain to connect the nasal capsule to a lamina orbitonasalis. Neither Brand nor Visser say whether the latter is ossified nor to what else it attaches (if anything), but Visser mentions that the lamina is absent in the region of the tentacle in the adult. This is the region represented in fig. 38 of his paper, as well as in fig. 4, so we cannot interpret his conception of this element. The cartilaginous rims of the nasal capsules illustrated by Els (1963) for Schistometopum thomensis are clearly not attached to any part of the rest of the skull, but have a slender posterior-trailing cartilago obliqua. we infer that the nasal sac is otherwise contained in bone, as in D. mexicanus (Wake and Hanken, 1982) and several other species examined with the technique used in this study by the senior author. It is noteworthy that the nasal capsule conformation in T. compressicaudus superficially resembles that of salamanders and of some frogs during their development, particularly in the absence of the prenasal process (see figs. in De Beer, 1937, for examples). However, the components giving rise to the overall shape and the pattern of their associations are different in Typhlonectes from those in the other amvhibian orders. We therefore consider this a very superficial sort of convergence, and cannot ascribe any functional properties to it. Presence of a prenasal process is the primitive condition in caecilians; Typhlonectes is a highly derived genus based on a diversity of characters, so we presume a secondary loss of the prenasal process. The early degeneration of the palatoquadrate and the articular surface of Meckel's cartilage and the early mineralization of those components and of the fetal teeth is similar to the condition in D. mexicanus reported by Wake and Hanken (1982), and may be characteristic of viviparous species with intra-oviducal feeding, as is the very presence of a species-specific fetal dentition as suggested by Wake (1977a, b, 1980,

76 M. H. WAKE ET AL. 1982). It should be noted that T. compressicaudus, and probably other typhlonectids, have an additional mode of intra-uterine nutrition during development, that of placental analogues of the ventral epidermis acting as an ectotrophoblast (Delsol et al., 1981), and of the sac-like embryonic gills. The shape of the palatoquadrate and its processes and its relation to the stapes are of much concern in the earlier literature on caecilian skull structure. Since most of this discussion relates to the shape of the ossified quadrate, which is not within the scope of this paper, we add little to the debate. However, we do note several aspects of the development of this structure. As had been reported for Ichthyophis glutinosus (de Jager, 1939, 1947) and for Hypogeophis and Grandisonia (Marcus, et al., 1935), the palatoquadrate is transitorily attached to the neurocranium (in Typhlonectes, to the anterior rim of the otic capsule, rather than to the taenia marginalis proper), and the stapes is transitorily attached to the palatoquadrate as the above authors report in their material. We do not consider this evidence for or against the question of hyoid or capsular origin of the stapes. We lack the critical earlier stages of development of the stapes, and infer that the other authors who have been concerned about this have not had the data either. The shape of the palatoquadrate in T. compressicaudus is much more like that of other caecilians for which there is information than it is like that of D. mexicanus. In D, mexicanus, the palatoquadrate.lacks ascending and basal processes, and the pterygoid process is very short. The element is a flat vertical plate. In T. cornpressicaudus, all of the processes are well formed but transitory, and the palatoquadrate has a three-dimensional shape at 22 mm TL (Fig. 3). We can add only a little information about ossification other than that already presented. It is clear that the stapes mineralizes late in ontogeny, typical of many species (see table 1 in Wake and Hanken, 1982). A free septomaxilla is present at birth, fused later to the maxilla. As noted above, most of the other fusions typical of the Typhlonectes skull as noted by Taylor (1969) have occurred by birth. However, we have noted that much development of certain of the dermal investing bones remains to take place (maxillary shelf, squamosal blade, posterior processes of the vomer, etc.). The typhlonectid skull does not exhibit the overlap of dermal elements typical of adults of most of the terrestrial species (see particularly Wiedersheim, 1879; Taylor, 1969; Wake and Hanken, 1982). Further, a large temporal fenestra between squamosal and parietal remains open. Whether this contributes to an actively kinetic skull, or whether it is a consequence of the aquatic mode of life, with reduced emphasis on burrowing in a dense substrate and concomitant release of binding of the skull elements to resist the shear and compression forces generated by burrowing, cannot yet be determined. We conclude that the chondrocranium and hyoid apparatus of Typhlonectes compressicaudus develop according to a pattern typical of most caecilians for which we have any information. However, the completeness of the structure of the anterior part of the chondrocranium at the time of maximal chondrocranial development may be unique. The wide separation of the nasal capsules and the absence of a septum nasi and prenasal process may also be unique to Typhlonectes. Some aspects of the pattern of development, especially the fetal dentition and its early mineralization, and early mineralization of the jaw articulation components, may be related to the viviparous mode of development, with maternal nutrition and a prolonged gestation period. Comparison of the ontogeny of the chondrocranium and hyoid apparatus of Typhlonectes compressicaudus with that of Dermophis mexicanus and with the developmental data for other species sug-

CHONDROCRANIUM OF GYMNOPHIONA 77 gests a common, rather constrained pattern of development, but some features that are correlated with differences in reproductive mode and adult habits. Acknowledgments.-We thank the Foundation Singer-Polignac for support for collection of the material. Laboratory work and analysis were supported by National Science Foundation grant DEB 80-09505 and discussion was facilitated by travel funds from the American Philosophical Society to M. H. Wake, to which agencies - we express our gratitude. BRAND, D. J. 1956. On the cranial morphology of Scolecomorphus uluguruensis (Barbour and Loveridge). Ann. Univ. Stellenbosch 32:l-25. DE BEER, G. R. 1937. The Development of the Vertebrate Skull. London: Oxford University Press. Pp. 1-552; 143 plates. DE JAGER, E. F. J. 1939. The gymnophione quadrate and its processes, with special reference to the processus ascendens in a juvenile lch- MARCUS, H., E. STIMMELMAYR, AND G. PORSCH. 1935. Beitrage zur Kenntnis der Gymnophionen XXV. Die Ossifikation des Hypogeophisschadels. Morph. Jahrb. 76:375-420. PETER, K. 1898. Die Entwicklung und funktionele Gestaltung des Schadels ;on lchthyophzs glutinosus. Morph. Jahrb. 25:4-78. RAMASWAMI, L. S. 1941. Some aspects of the cranial morphology of Uraeotyphlus narayani Seshachar (Apoda). Rec. Indian Mus. 43:143-207. -. 1948. The chondrocranium of Gegenophis (Apoda Amphibia). Proc. Zool. Soc. London 118:752-760. SARASIN. P.. AND F. SARASIN. 1887-90. Zur Entwicklungsgeschichte und ~natomie der Ceylonischen Blindwuhle, lchthyophis plutinosa. Ergebn. Naturforsch. Ceylon-1884-6: 4 vols. Wiesbaden: C. W. Kreidel's Verlag. TAYLOR, E. H. 1969. Skulls of Gymnophiona and their significance in the taxonomy of the group. Univ. Kansas Sci. Bull. 48:585-687. VISSER,M. H. C. 1963. The cranial morphology of lchthyophis glutinosus (Linne) and lchthyophis monochrous (Bleeker). Ann. Univ. Stellenbosch 38:67-102. WAKE, M. H. 1976. The development and replacement of teeth in viviparous caecilians. J. Morphol. 148:33-64. -. 1977a. The reproductive biology of cae- thyophis glutinosus. Anat. Anz. 88:223-232. cilians: an evolutionary perspective. In D. H. -. 1947. Some points in the development Taylor and S. I. Guttman (eds.), The Reproof the stapes of Ichthyophisglutinosus. Anat. Anz. ductive Biology of Amphibians. New York: 96:203-210. Plenum Publ. DELSOL,M., J. FLATIN, J.-M. EXBRAYAT, AND J. BONS. -. 1977b. Fetal maintenance and its evolu- 1981. Developpement de Typhlonectes compressicaudus, amphibien apode vivipare. Hy- nophiona. J. Herpetol. 11:379-386. tionary significance in the Amphibia: Gympotheses sur sa nutrition embryonaire et lar- -. 1978. Ontogeny of Typhlonectes obesus, vaire par un ectotrophoblaste. C. R. Acad. Sc. with emphasis on dentition and feeding. Pa- Paris 293:281-285. peis Avulsos, 2001. 12:l-13. ELS, A. J. 1963. Contributions to the cranial mor- -. 1980. Reproduction, growth and popuphology of Schistometopum thomensis. Ann. Univ. Stellenbosch 38:39-64. lation structure of Dermophis mexicanus (Amphibia: Gymnophiona). Herpetologica 36:244- EXBRAYAT, J.-M. 1983. Primieres observations sur 256. le cycle annuel de l'ovaire de Typhlonectes com- -. 1982. Diversity within a framework of pressicaudus (Dumeril et Bibron, 1841), batracien apode vivipare. C. R. Acad. Sc. Paris 296: 493-498. contraints: reproductive modes in Amphibia. In D. Mossakowski and G. Roth (eds.), Environmental Adaptation and Evolution A The- -, M. DELSOL, AND J. FLATIN. 1981. Pre- oretical and Empirical Approach. Stuttgart: mieres remarques sur la gestation chez Typhlonectes compressicaudus (Dumeril et Bibron, 1841), amphibien apode vivipare. C. R. Acad. Sc. Paris 292:417-420. --, AND -. 1982. Observations concernant la gestation de Typhlonectes compressicaudus (Dumeril et Bibron, 1841) amphibien apode vivipare. Bull. 2001. Fr. 107:486. HANKEN, J., AND R. J. WASSERSUG. 1981. The visible skeleton. Funct. Photog. 16:22-26 and 44. Gustav Fischer Verlag. -, AND J. HANKEN. 1982. The development of the skull of Dermophis mexicanus (Amphibia: Gymnophiona), with comments on skull kinesis and amphibian relationships. J. Morphol. 173:203-223. WIEDERSHEIM, R. 1879. Die Anatomie der Gymnophiona. Jena: Gustav Fischer Verlag. Accepted: 2 February 1984.

http://www.jstor.org LINKED CITATIONS - Page 1 of 1 - You have printed the following article: The Development of the Chondrocranium of Typhlonectes compressicaudus (Gymnophiona), with Comparison to Other Species Marvalee H. Wake; Jean-Marie Exbrayat; Michel Delsol Journal of Herpetology, Vol. 19, No. 1. (Mar., 1985), pp. 68-77. Stable URL: http://links.jstor.org/sici?sici=0022-1511%28198503%2919%3a1%3c68%3atdotco%3e2.0.co%3b2-5 This article references the following linked citations. If you are trying to access articles from an off-campus location, you may be required to first logon via your library web site to access JSTOR. Please visit your library's website or contact a librarian to learn about options for remote access to JSTOR. Literature Cited Fetal Maintenance and Its Evolutionary Significance in the Amphibia: Gymnophiona Marvalee H. Wake Journal of Herpetology, Vol. 11, No. 4. (Oct. 31, 1977), pp. 379-386. Stable URL: http://links.jstor.org/sici?sici=0022-1511%2819771031%2911%3a4%3c379%3afmaies%3e2.0.co%3b2-w