934 



SCIENCE. 



[N. S. Vol. VIII. No. 209. 



arrangements based on such characters was 

 an expression of deep-seated affinities. It 

 was not till 1865 that a key to the definitive 

 arrangement of the group was discovered 

 by Cope in the osteological characters and 

 especially modifications of the sternal ap- 

 paratus. Cope's arrangement has been 

 called by Mr. Boulenger an ' epoch-making- 

 classification' and, in the form to which it 

 has been brought by the later labors of 

 Cope and Boulenger, will doubtless be near 

 that which will gain ultimate general ac- 

 ceptance. Nevertheless, it may be thought 

 too much value has been attached to the 

 arciferons type of sternum. An arrange- 

 ment of the phaneroglossate forms into three 

 superfamilies, of which the arciferous fam- 

 ily of Discoglossidoe is the most generalized, 

 may be more acceptable to some ; this has 

 been named Discoglossoidea. The other 

 Arcifera are the Bufonoidea, and the Fie- 

 MiSTERNiA are equivalent to another super- 

 family of nearly equal systematic impor- 

 tance — the Ranoidea. 



The Discoglossoidea differ from both the 

 others in the possession of ribs in the adult, 

 and by the median position of the spirac- 

 ulum or outlet from the branchial cham- 

 ber in the tadpole. Less important pe- 

 cularities are the arrangement of the male 

 urogenital apparatus (so that ' the sper- 

 matozoa are conveyed through the vasa 

 efferentia direct to the seminal duct ; the 

 latter produced forward beyond the kidney') 

 and the disposition of the labial teeth 

 series ' in two or three rows.' The single 

 family — Biscoglossidw — contains three Euro- 

 pean genera and five species ; the genera 

 were widely separated by the old systema- 

 tists, and it was to Cope that the earliest 

 appreciation of their relationship was due. 



The Bufonoidea, or typical Arcifera, are 

 represented in Europe by three families — 

 Pelobaiidce, BiifonidcB and Hijlidce. The 

 Pelobatidce of Boulenger were distributed by 

 Cope among two families, one of the Euro- 



pean genera (Pelodytes) having been desig- 

 nated as the type of Pelodytidte, and the 

 other (Pelobates) associated with the Amer- 

 ican Scapliiopus in the family Scaphiopidte. 

 There can scarcely be any question that 

 Boulenger is correct in reducing the two 

 Copian families to one. The only characters 



sp.- 



Alytes 



Eyla 



used to differentiate them were the ' sacrum 

 united with the coccyx by condyle ' in the 

 Pelodytidte, and the ' urostyle without condy- 

 loid articulation, its axial portion restrict- 

 ing that of the sacrum and connate with it, ' 

 in the Scaphiopidce.^- 



The inconstancy of this character in some 

 groups has been shown by Boulenger. "Du- 

 ges, basing his observations [on the Pelohates 

 cidtri2}es] , has denied the fusion of the sacral 

 vertebra with the coccyx described by Mer- 

 tens in Pelohates fuscus, with which P. cidtri- 

 pes was then confounded ; he observes, how- 

 ever, that the articulation, by means of 

 one condyle, is an almost immovable one. 



"Cope Batr. N. Am., 296. 



