Vol. 70.] A'ERTEBRATES FROM BRITISH EAST AFRICA. 1S1 



Testudo partialis, the length of the complete plastron must have 

 been about 84 centimetres. 



The epiplastra (Epi.), which form a slight projection at the 

 anterior end of the plastron, are greatly thickened, and, as in 

 T. partialis, bear on their upper surface a strong prominence, 

 which is deeply hollowed out posteriorly to form a high lip. 

 The entoplastron (Eat.) is lozenge-shaped, its transverse diameter 

 being the longest. In the middle line on its upper surface is a 

 strong ridge, produced posteriorly into a pointed process, which 

 seems to have projected behind the rest of the bone. On the 

 ventral face of the plastron, the postero-lateral borders of the ento- 

 plastron are marked by strong grooves, as if a line of division in 

 the epidermal shields occurred there. The hyoplastra (Hrjo.) seem 

 to have been relatively longer than in T. partialis. 



The paired gular shields (y.) are very short; they cover the 

 projecting portions of the epiplastra, and their posterior angle is 

 considerably in front of the anterior angle of the entoplastron. 

 This last character distinguishes this species from all with which 

 it has been compared, except T. perpiniana Deperet & Donnezan 1 

 from the Pliocene of Roussillon. From this, however, the 

 species here described is distinguished by the form of the 

 epiplastra. In consequence of the small size of the gulars, the 

 humeral shields (hum.) appear very large; the line of union be- 

 tween them and the pectorals is straight, and passes immediately 

 behind the posterior angle of the entoplastral bone. The peculiar 

 grooves running along the hinder border of this bone, and perhaps 

 indicating a partial division of the humeral shields, have been 

 mentioned above. 



The pectoral shields (pect.) are very short from before back- 

 wards : they are widest at their median ends, and narrow outwards 

 almost to a point. The boundary between the abdominal and the 

 femoral shields is not preserved. 



This Chelonian may best be referred to the genus Testutio, 

 although it is possible that, when the whole shell is known, it will 

 be found necessary to separate it. It may be, for the present, 

 called Test a a ] o crass a, sp. nov. 



Some dimensions (in centimetres) of this specimen are : — 



Estimated total length 84 - 



Width immediately in front of the axillary buttress ... 39 '5 



Length of the entoplastral bone 16*2 



Width of the same 19'4 



An imperfect left scapula of a Chelonian of gigantic size may 

 here be mentioned. The specimen is from Karungu, probably 

 from the Lower Miocene deposits, and was sent to the British 

 Museum by Mr. C. W. Hobley, C.M.Gr. It is rather larger than 

 the corresponding bone of a skeleton of T. gigantea 8chweiu r u'- 

 from Aldabra, the shell of which is about 123 centimetres long in 



1 Mem. Soc. Geol. France : Paleontologie, vol. iv, Mem. 3 (1893) pi. [ii] xv, 



