87 
enough, however, the types, from Brava, South Somaliland, are fairly inter- 
mediate in their colour between the two, although, if anything, rather more 
like the one geographically most distant, that from Kilimanjaro. 
In the skulls again, while, as is usual in these Antelopes, considerable 
differences are to be observed between any two skulls compared together, 
these differences do not appear to be correlated either with locality or colour- 
characters. In fact, with regard to the extension backwards of the pre- 
maxille towards the nasals, one specimen in the British Museum Collection 
has the two extremes on the two sides of its skull, showing conclusively that 
this character cannot be relied upon. 
We have therefore come to the conclusion that, so far as the colour and 
skull-characters here mentioned are concerned, the Kilimanjaro, Lamu, and 
Brava Dik-diks cannot be separated from one another, even as subspecies or 
local races. 
December, 1895. 
helped materially to bring us to the conclusion we have come to above, as their wide range of 
variation shows that certain differences in the hoof-lengths that we had previously noted 
in the different forms cannot be regarded as of any value for distinguishing the species, and 
must merely be due to individual variation. 
