CHAP, in.] VERTEBEyE : BATRACHIA. 125 



for articulation with a tenth vertebra. The right zygapophysis 

 was well formed and articulated with the tenth, but the left was 

 rudimentary. The tenth vertebra itself had an imperfect centrum 

 and the neural arch though complete was markedly asymmetrical. 

 Posteriorly its centrum presented two convexities for articulation 

 with the urostyle. [For details see original figures.] BOURNE, 

 A. G., Quart. J. Micr. Sci., xxiv. 1884, p. 87. 



This is a case of some importance as exhibiting Meristic 

 Variation in a simple form. Of course, as Bourne says, we may 

 say that in this specimen the end of the urostyle has been 

 segmented off and that it is composed of " potential " vertebra?, 

 and as he also remarks, it is interesting in this connexion to 

 notice that some Anura, e.g. Discoylossus, present one or two pairs 

 of transverse processes placed one behind the other at the proximal 

 end of the urostyle. But this description is still some way from 

 expressing all that has happened in this case ; for beyond the 

 separation of a tenth segment from the general mass of the 

 urostyle there is Substantive Variation in the ninth vertebra in 

 correlation with this Meristic Variation. For the ninth has devel- 

 oped a zygapophysis and has two concavities behind, like the 

 vertebra? which in the normal frog are anterior to the ninth. 

 There is therefore a forward Homoeosis, associated with an increase 

 in number of segments, just as there is in such a case as that of 

 Man (No. 26) or in that of Galictis vittata (No. 50). 



It is also interesting in this case to see that the actually last free 

 vertebra here, though it is the 10th, has two convex articular sur- 

 faces behind like the 9th, which is the last in the normal frog, thus 

 shewing a similar forward Homoeosis. Now applying the ordinary 

 conception of Homology to this case, we may, as Bourne says, 

 prove that the 9th in it is homologous with the 9th in a normal 

 frog for its transverse processes are enlarged in the characteristic 

 manner to carry the pelvic girdle. But similarly we may prove 

 also that the tenth in this case is homologous with the ninth of the 

 normal, for its centrum has the peculiar convexities characterizing 

 the last free vertebra, Baur's proof that the first caudal was 

 homologous in the two specimens of Gavialis (see No. 54) rested 

 on the same class of evidence, and for the moment is satisfying, 

 but as here seen this method though so long established leads 

 to a dead-lock. Upon this case it may be well to lay some stress, 

 for the issues raised are here so easily seen. Besides this the 

 imperfect condition of the extra vertebra enables us to see the 

 phenomenon of increase in a transitional state, a condition rarely 

 found. In the instances recorded in Gavialis (No. 54), owing to 

 the perfection and completeness of the variation, the characters of 

 the 1st caudal are definitely present in the 28th though normally 

 proper to the 27th, and therefore it may be argued that the 28th 

 here is the 27th of the type. The frog here described shews that 

 in this conclusion other possibilities are not met. On the analogy 



