Annals of the Transvaal Museum. 
•) ♦> 
earth’s surface: this is not a circular argument, as the evidence for these 
latter hypotheses is based on sound geological facts. The very fact of the 
occurrence of the two main branches of the Cystignathidae in Australia 
and South America respectively is sufficient to imply the existence of this 
family in Cretaceous times, for according to all accounts the land connec- 
tions between those two countries must have terminated by early tertiary 
times at the latest. At the same time we must add that the Cystig- 
nathidae is perhaps the oldest family of the Anura. 
Again, it is obvious that the immigration of Australian types into 
southern Africa can be dated less remotely the more northward the place 
of entry. Now, even the present-day distribution of this genus is very 
imperfectly known : it was first taken in the Stellenbosch neighbourhood 
and recently at Knysna, but as apparently it is of purely arboreal habit 
it must easily escape notice, and we may reasonably expect that it will 
prove to have a much wider range : very likely the genus will be found 
to occur in the forest region of the whole coastal strip of Cape Province 
at least. Nevertheless, it should be mentioned that from analogy with 
the plants, the western portion of Cape Province is just the locality where 
we might have expected to meet with Australian types, for Dr. Schonland 
says : “ It is certainly most extraordinary that wherever we find the closest 
agreement between the floras of the two countries (South Africa and 
Australasia), it is between the boxed-up flora of the Cape Province (of 
botanists, i.e. south-west Cape Colony) and the flora of Australia, chiefly 
west and south-west Australia, and when we consider that this connection 
exists in widely separated orders these facts become all the more wonder- 
ful.” But even from a botanical point of view it seems necessary to 
postulate a former wider distribution in South Africa for this particular 
flora, and in the same paper we read : "A former eastward extension (of 
the typical Cape flora) is certainly demanded by our theory . . . ; we 
know that even now many outliers of the typical Cape flora reach far 
east, and their number would probably be much greater if the tropical 
African flora did not appear as a formidable competitor in the coast 
regions.” In this connection I may remark that the various botanical 
areas of South Africa have little significance from the point of view of 
vertebrate zoology : it is quite true indeed that the peculiar South African 
element is specially concentrated in south-western Cape Province, but 
thence it is diffused all over southern Africa, the Congo basin excepted, 
and only two areas, an eastern and a western separated by the Drakensberg 
Kange, have any claim even to sub-regional rank, though within a 
sub-region the several species of a genus may range themselves 
according to the various environmental conditions. Amongst the South 
African reptiles perhaps the most striking case of Australian relationship 
is that of the genus Oedura, which is confined to the South African and 
Australian regions. The distribution of our species is as follows : 
0. ajricana* Damaraland, eastern Cape Colony, Natal : 0. nivaria , first 
taken in Natal at the highest point of the Drakensberg Range and more 
recently in the Pirie Bush (F. A. 0. Pym). The case of Oedura will 
* The Damaraland record is open to suspicion : nivaria is probably a synonym of africana. 
3 
