LI^"NEAN SOCIETY OF LOKEON". 85 



Megahi-nhriis, which was said to be opisthocceloLis, the prococlous 

 coiuhtion actually occurred, and in the following year lioulenger, 

 referring to this point, concluded ('oS, p. 4US) : " it is therefore 

 clear that this character, liowever important it may appear 

 at first, is worthless even as a specilic character in these 

 Batrachians." 



The Lifrequency of Variation in the Mode of Vertehral 

 Articulation. 



A fact, however, which lends weiglit to the value of the character 

 of the vertebral centra for systematic purposes is the extraordinary 

 rarity of the occuiTence of individual variation. In not one of 

 at least a hundred specimens of the procoelous families Biifonidae, 

 Hylidae, and Cystignathidtc (represented by over seventy species) 

 did I find a case of indi\idual variation in the character of the 

 vertebral centra, although variation affecting the vertebrae in 

 other ways was not infrequent ! Indeed, so far as I can discover, 

 oidy once has an abnormality affecting the centra been recorded 

 in a member of these families, this being described by Cope ('66) 

 in Borhorocaites. In the vertebral column of the Pelobatidse 

 individual variation is peculiarly frequent, but even in this family, 

 very few cases of variation in the character of the centra have 

 been met with outside of the genus Mcijalophrys. 



Among the Diphisiocffilous families, also, the occurrence of 

 individual variation in the condition of the centra is extra- 

 ordinarily rare — indeed, it is almost certainly the rarest of all 

 abnormalities. AVhen such variation does occur it appears 

 invariably to take the form of a reversion to the more primitive 

 (strictly proccBlous) condition. Thus, Lloyd Morgan has described 

 one such case in the common frog, while I have myself met with 

 Iavo other instances of this reversion. The first of these was 

 found in a specimen of B. escahnta, and the second may be seen 

 in a specimen which is preserved in the British Museum Collec- 

 tiou. It is labelled liana sp., but is, almost certainly, 2i. tigrina. 

 Apart from these three examples * there are, I believe, no known 

 cases of variation in the condition of the centra in diplasioca3lous 

 forms. Moreover, I have recently examined the vertebral column 

 of more than one hundred and fifty specimens of the common 

 frog, and although I find that the vertebral column is variable in 

 8 per cent, of the specimens examined, yet there was not one 

 whicli departed from the typical diplasiocoelous condition of the 

 centra. 



T7ie Character of the Vertehral Centra not " Adaptive." 



The precise manner in which the vertebral centra articulate 

 does not appear to be related to the habits of the various forms, 

 tliat is, it cannot be described as an adaptive character. There 



* An almormal vertebral cohiniii {R. tempomria) described by Howes ('86) 

 liad ten vertobrie all of vvliicli were procoelous. 



