34 BULLETIN 31, UNITED STATES NATIONAL MUSEUM. 



Tlic C«cilii(liXi arc generally regarded as represeutiDg a distinct order, 

 which bears the names Apoda or Gymnophiona. The definition given 

 to this order by Mr. Boulenger* is : "No limbs; tail rudimentary. Males 

 with an intromittent copulatory organ. Adapted for burrowing." Of 

 these definitions none is of ordinal value. The tail in some species is 

 distinct. The intromittent copulatory organ in Dermophis mexicanus, 

 Oymnopis proximus, and Herpele ochrocephala is not an especial organ, 

 but is merely the everted cloaca. The hard papilhe observed by Giiu- 

 tbert in the IclithyopMs gluUnosus are wanting in the above species. 

 The protrusion of the cloaca is effected by two especial muscles, which 

 are wanting in Ampbiuraidie. As to limbs, their extremely rudiment- 

 ary character in Amphinma is well known. To regard their condi- 

 tion as indicating ordinal separation from the Cseciliidfe is not in accord- 

 ance with our practice in similar cases in the Eeptilia, as in the order 

 Laceitilia. 



I have endeavored to sustain the order Gymnophiona by the character 

 of the fusion of the nasal and premaxillary bones found in the majority 

 of the genera. | But Stannius § shows that these bones are distinct in 

 Ichthyojihis. Huxley states (Anatomy of Vertebrate Animals, p. 155) 

 that in Ichthyophis gluUnosus a distinct bone nearly encircles the orbit. 

 This he compares to the supra and postorbital bones found in the Stego- 

 cepbali. But in Ohthon^rpeton, Csecilia, Dermophis, and other genera, 

 this bone forms part of the maxillary, so that it is not characteristic of tlie 

 family, and may not be homologous with the bones which occupy the 

 same position in Stegocephali. Wiedersheim calls it maxillary. 



With these fact in view I have united || the Caeciliidai with the Urodela, 

 a proposition which I fully believe to be sustained by the evidence. The 

 Ccvciliidce is a family of Urodela, connected with the typical forms through 

 the Amphiumidce. 



Geographical distrihuUon. — The distribution of the families of the 

 Urodela and their contents is as follows : None exist in the Australian 

 realm and, excepting some Oceciliidte, in the Ethiopian realm. In the 

 iSTeotropical they are restricted to the Central American and Mexican 

 districts, with a few species in the northern Andes, and one reputed to 

 be from the West Indian island, Santo Domingo. 



The CseciliidiB are more widely distributed, belonging especially to 

 trojiical regions. 



* Catalogue of tlio British Maseum, 1882, i>. 88. 

 t Reptiles of British India (Ray Society), p. 441. 

 X American Naturalist, 1884, p. 26. 

 § Zootomie der Amphibien, 1856, p. 44. 



II American Naturalist, 1885, p. 244, note, and Proceeds. Anier. Philos. Soc.^ Feb, 

 1886 p. 442. 



