142 PROF. G. B. HOWES AND W. RIDEWOOD ON [Mat, 6, 



Wiedersheim and his pupil Kehrer, and of Baur (1), have added 

 considerably to our knowledge of the lirr.b-skeleton of the Urodela, 

 that of the Aniira has received much less attention. The most 

 important investigations recorded since the classic of Gegenbaur ( 1 8), 

 are those of Born (3, 5). Briihl has published (11) figures of a 

 somewhat extraordinary character, pertaining to some seven or eight 

 genera ; but concerning these we have little to say, except that we 

 fully endorse Bom's criticisms and corrections of them (6, pp. 48 

 et seq.). 



Bom's latest communication purports to be somewhat revolu- 

 tionary, so far at least as the carpus is concerned (6, p. 62). It will 

 be seen, however, in the sequel, that we are unable to accept his 

 determinations ; and as the discovery of new elements and relation- 

 ships has led us to differ from most of our predecessors, in our 

 estimate of the morphological value of leading elements of both 

 carpus and tarsus, it is binding upon us to preface this paper with 

 an account of those observations upon which our dissensions are 

 based. In accordance, then, with the exigencies of the case, we shall 

 divide the paper into two sections : viz. : — firstly, a ffeyieral part, in 

 which will be given a review of the morphology of the carpal and 

 tarsal elements, together with a detailed account of those hitlierto 

 unrecognized structures, whose discovery has necessitated a redeter- 

 mination of the value of any one or more constituents ; and, secondly, 

 a special part, in which the leading families will betaken successively, 

 the characters of the carpus and tarsus of each being noted in the 

 light of the preceding section, and special peculiarities dealt with. 



b. Nomenclature. — Our investigations do not profess to be 

 exhaustive ; they leave much to be settled, and we have accordingly 

 refrained from introducing a nomenclature which might embody a 

 premature expression of homological relationships with the limbs of 

 other animals. In dealing with elements whose morphological 

 value remains in the least degree doubtful, we have, for these 

 reasons, adopted the nomenclature of Ecker (17) as being the more 

 empirical and better suited to our purpose than the alternative one 

 of Duges (16). 



In describing the pre-hallux (calcar), we have, in order to 

 facilitate comparison with the works of our predecessors, adopted 

 the older system of nomenclature, calling the outermost digit the 

 fifth, the innermost one the pre-hallux or calcar, and that next to it 

 the hallux or first digit. 



New terms have been introduced only where unavoidable ; and 

 some of these are, for reasons already given, purely empirical. 



c. Material. — It will be seen, from the second part of the paper, 

 that we have examined the limbs of 37 genera and 60 species — 

 adequate representatives of all the leading families as defined by 

 Boulenger (7), with the exception of the Dyscophidce, Dendrophry- 

 niscidce, Amphignatkodontid<e, and Hemiphracfidce. For these 

 specimens our grateful acknowledgments are due mainly to Prof. 

 Huxley, and to Mr. G. A. Boulenger of the Natural History 

 Museum. To the last-named gentleman we are indebted for further 



