Dr. C. W. Andrews — Gigantic Eocene Land Tortoise. 529 



In the type and in several other specimens the epiplastral region 

 is produced forwards so as to project considerably beyond the 

 anterior border of the carapace ; this epiplasti'al prominence, like 

 the deep concavity of the middle portion of the plastron, is probably 

 a male chai'acter. In the Museum at Cairo there is a beautifully 

 preserved shell of which, the carapace differs very slightly from 

 that of the type (e.g. in the smaller degree of convexity of the 

 vertebral shields), while the plastron is very dissimilai". In the 

 first place it is only slightly concave, then the epiplastral region, 

 though prominent, is not produced forward in the same way, and 

 lastly the portion of the xiphiplastrals covered by the anal shields 

 is much narrower, so tliat the posterior emargination of the plastron 

 is narrower and its angle more acute ; this specimen may be regarded 

 as probably the female, form of the present species. 



The dimensions of the type-specimen (Plate XVII) are : — 



Length of the carapace in a straight line 

 Length of carapace along curve 

 Width of carapace in straight line . . . 

 Width of carapace along curve 

 Length of plastron (greatest) 

 Width of anterior lobe of plastron . . . 

 Width of posterior lobe , , 



AVidth of episternal prolongation . . . 



The dimensions of the female specimen are much smaller; some 

 of these are : — 



Length of carapace 



Width of carapace ... 



Length of plastron ... 



Width of anterior lobe of plastron 



Width of posterior lobe 



In the former account, owing to the fact that the presence of 

 a nuchal shield was not recognised, it was stated that these species 

 approached the Galapagos tortoises rather than those of the African 

 islands. Now, however, it seems possible that there may be some 

 relationship with the species inhabiting Madagascar and the Aldabra 

 Islands, a much more probable suggestion. 



The presence of the nuchal shield, on the other hand, separates this 

 form somewhat more widely from some of the Tertiary species with 

 which it has been compared. Nevertheless, in many respects it is 

 similar to Testudo perjnniana, described by Deperet and Donnezan 

 from the Lower Pliocene of Eoussillon, and it may also be allied to 

 T. Uheronis, Deperet, from the Miocene of Mount Leberon, and 

 T. gigas, Bravard, from the Upper Oligocene of the Allier. 



As was suggested in the former note, " it seems probable that in 

 the present species we have an early representative of a large group 

 of tortoises, members of which occur at several horizons in the 

 Tertiary beds of Europe, and of which T. pardalis and T. calcarata 

 may be modern forms. T. Atlas and T. Cautleiji, of the Siwalik 

 Hills, and the still existing T. Sumeirei, of Mauritius, may also fall 

 into this group." To this it may be added that probably the giant 

 species of Madagascar and the Aldabra Islands are also I'elated. 



