Geometrica-Group of S.A. Tortoises. 197 



some characters have retained their ascendency in both sub-groups, 

 while at the same time other and divergent determinants have become 

 uppermost, maybe as a result of varying external conditions. 



The two contrasted species are widely separated geographically. 

 Geomctrica is limited to the low districts around the south-western 

 extremity of Cape Colony, whereas ociilifera is found over a wide 

 area, along the high, northern borders of Cape Colony, extending 

 into Damaraland and the tropical parts of Bechuanaland. 



There can be no question that oculifcra is very closely related 

 to the verreauxii sub-group, but the specimens at my command can 

 scarcely be regarded as directly transitional where the combination 

 of all the characters is taken into account. Until the intermediate 

 region between those from which verreauxii and ociilifera have been 

 obtained is thoroughly explored, the form must be accorded specific 

 rank. As in the case of geometrica, it is probably our present 

 ignorance of transitional forms which forces one to this conclusion, 

 and strong efforts are being made to remedy this. 



c. Testudo tentoria, Bell, 1828. 



The forms embraced under this species present great variations 

 in all the characters under consideration. In many ways they come 

 nearest to what we may suppose the common ancestor of the 

 geometrica-gvoup to have been ; the characters are very fluctuating, 

 and in almost every direction suggest conditions which are more 

 elaborated elsewhere. When discussing the separate characters, the 

 tentoria sub-group was in nearly every instance taken as the starting 

 point, as representing the most generalized or incipient condition. 

 Better than in any other sub-group, its different members suggest the 

 lines of elaboration which the various characters will pursue. 



Within this sub-group we find the transitions from the uniformly 

 arranged radiating yellow rays on the carapace shields, all of equal 

 value, to the distinction into primary, secondary, and tertiary rays ; 

 the transitions from the uniformly black, central patch and yellow 

 sides of the plastron, to the rayed pattern ; the transitions from the 

 rows of uniformly enlarged antebrachial shields to the excessive 

 growth of a more central one, and from the absence of an enlarged 

 femoral spine to a strongly developed one ; transitions from four or 

 five inframarginal shields to only one ; from an extremely knobbed 

 carapace to one with almost flat shields ; and from a rudimentary, 

 incipient nuchal shield to one well developed. 



The species has a wide geographical range, and, placing 

 together suites of specimens from different localities, marked differ- 

 ences are recognizable ; yet, when compared with the many transi- 

 tional varieties, no one would dream of giving them specific rank. 

 It is unquestionably a very plastic form in every respect, but 

 with local assemblages showing more or less constancy. 



The representatives occurring around the southern and eastern 

 coasts are least specialized or furthest removed structurally from the 

 other sub-groups, but as we approach the more inland and northern 



