84 rnOCEEDlXGS OF T1[E 



for what will probiibly prove to be inerely individual variations 

 (vide infra), of the uniformly proccelous type. They differ, how- 

 ever, from the remaining procoelous families, in that the convexity 

 upon the sacral vertebral centrum (for articulation with the 

 coccyx) is single instead of double as in all other Anura. 



It will be seen, therefore, that the Anura may be divided 

 sharply into five groups, according to the structure of the 

 vertebral column (and attached ribs). 



(i.) Sacral vertebrae always fused with coccyx ; pre-sacral 

 vertebrae seven (or fewer), opisthoccelous : with ribs in develop- 

 ment. (Aglossa.) 



(ii.) Sacral vertebra free, with biconvex centrum ; pre-sacral 

 vertebriB not less than eight, opisthoccelous : with ribs. 

 (Discoglossidaj.) 



(iii.) Sacral vertebra proccelous, ankylosed to coccyx, or if free, 

 with but a single condjde for articulation with coccyx ; pre-sacral 

 vertebra} eight, proccelous: without ribs. (Pelobatida;.) 



(iv.) Sacral vertebra free, proccelous, witb double condyle for 

 the coccyx; pre-sacral vertebrae eight, proccelous: without ribs. 

 (Remaining Arcifera.) 



(v.) Sacral vertebra free, biconvex ; eighth vertebra biconcave ; 

 first seven vertebrae proccelous : without ribs. (Firmisternia.) 



Discussion. 



Earlier Vicivs as to the importance of the condition of the 

 Vertebral Centra as a systematic character. 



As is well known, a varying degree of importance has been 

 attached, by different authors, to the condition of the vertebral 

 centra in the Anura. Thus both Cope ('66) and Lataste ('79) laid 

 considerable stress upon the opisthoccelous or proccelous character. 

 Eoulenger ('82), in his ' Catalogue of the Eatrachia Salientia 

 s. Ecaudata,' made use, principally, of the condition of the 

 pectoral g rdle and of the dentition, and pointed out that 

 the mode of vertebral articulation appeared to be of unequal 

 importance throughout the group. 



Some three years later, Blanchard ('S5, p. ^»^;S) went so far as 

 to propose that the tailless Batracliia should be separated into 

 two principal groups, the " Opisthocceliennes " and the " Pro- 

 cceliennes." Jlis suggestion did not meet with acceptance, and 

 subsequent writers seem to have attached less and less value to 

 the condition of the centra as a systematic character. Thus, in 

 1901, Gadow wrote ('01, p. 19): "The systematic value of this 

 pro- or opistho-ccxjlous character has been much exaggerated." 

 He continued (p. 20), " it is not difficult to imagine that in the 

 Anura the production of pro- or opistho-ccelous vertebrae depends 

 6imi)ly upon the centra or articulating knobs happening to fuse 

 either with the hind or the front end of the vertebra\" 



In 19u7 Beddard pointed out that in some species of the genus 



