NOVITATES ZOOLOGICAE. 



Vol. III. DECEMBER, 1896. No. 4. 



TESTUDO EPHIPPIUM. 



By dr. a. GItNTHER, F.R.S. 



(Plates XX., XXL. and XXII.) 



THE subject of this paper is based upon materials which Dr. G. Baur collected 

 ia Duncau Island during his visit to the Galapagos Archipelago in IsOl, 

 and wliich were afterwards acquired by the Hon. Walter Rothschild. A large 

 stuffed wall- with bones (carapace 291 in. long, measured in a straight line) and the 

 skeleton of a smaller fi'irude (carapace 22i in.) were retained for the Tring 

 Museum, whilst the skeletons of two ni(de.-< (respectively 2Ti and 2.j in. long) 

 were transferred to, and are now in, the Natural History Museum. 



These Duncan Island specimens belong to the species which I described 

 from a stutfed adult male in the Edinburgh Museum of Science and Art under 

 the name of Tesfiido ephippium* The direct comparison with the type, of 

 which carapace, skull, pelvis, and large liml)-bones are preserved, leaves no 

 doubt as to their specific identity. The agreement is so perfect as far as rhe 

 parts mentioned are concerned, that they do not require a new description, but I am 

 enabled to su]iplement my first account by describing the carapacx^ of a voung 

 friivde, and tlic vertebral column which had been lost in the Edinburgh specimen. 



Carapace of the Female, 22i in. long (PI. XX.). — The thickness of the shell 

 is much greater than in Testndo abingdonii, but the individual plates are attenuated 

 on the margins so as to leave now and then a narrow cleft between them. 



As regards general form the differences from the adult male are clearly 

 due either to sex or to less .advanced growth. The front part of the carapace 

 is less compressed, but the upper profile from the middle of the central 

 dorsal plate is almost horizontal ; the sides of the front part are decidedly 

 concave, the concavity being deepened by the reverted  anterior margins. The 

 upper jiart of tlie shell, when viewed from above, is flat and broad ; the hinder 

 part is rounded, its middle portion being steeply declivous, whilst the marginal 

 plates above tlie hind-legs are as strongly arched outwards and upwards as 

 in the adult male. The posterior margin of the sliell is deejdy scalloped, more 

 so than the anterior. Tlie plates are deeply striated, the striated portion lieing 

 very distinct from the areolai'. Although we know the mode antl rate of growtii 

 of the epidermal plates from Aldabra specimens kept in captivity, and although, 

 in a state of nature, tlie growth most probably proceeds intermittently but at 

 ri'gnliir annual jieriods, so that each stria rejiresents the extent of sm annual 

 growth, it would be somewhat hazardous to calculate the age of our individual from 

 tlie number of striae. The oldest .striae gradually disapjiear, and more prominent 

 striae are mixed with less projecting. Thus, if in our sjiecimen each visible 

 stria is taken into consideratiou, its age might have been from fourteen to fifteen 



' I'Idl. Tram. Vol. 10,') (187.".), p. L'7I ; or flhjaitt. Land Turl . (IS??), p. SI, witli plates. 



23 



