Georychus and tts Allies, 443 
shape, and old ones of the latter, but individual exceptions to 
this rule are very numerous. No doubt the shape of the 
foramen is a good enough character, if care be taken that it 
is regarded as an average and not an invariable one, but 
species should certainly not be described on it alone without 
quite considerable series. 
Of course, in species isolated geographically, such as Cryp- 
tomys foxt or zecht, the characters are more constant, but 
where, as in the southern half of Africa, the animals are 
found everywhere and no barrier to distribution exists, it 
may almost be said that any character of skull may be found 
in any species, thus absolutely nullifying geographical splitting 
of the usual type. 
Even size, when the giant northern species are put aside, 
is of but little use, for in any series of skulls a few individuals 
will be found much larger than the great majority, these 
animals appearing to add to the bulk of their skulls long after 
they have attained adult life. For example, the type-skulls 
of both C. bocagei, de Wint., and C. whytet, Thos., are so 
much larger than others. from the same areas that they might 
readily be thought specifically different from them. 
Of course, I do not mean that the skull-characters are 
negligible, but only that a much more liberal view of their 
inconstancy and want of importance should be taken than is 
the case in other mammals. Let anyone look at Mr. Austin 
Roberts’s figures of skulls, and he will see how essentially 
similar they all are to one another. 
The mammary formula of Cryptomys appears to be always 
2—1=6. At least, this is the number I find in every spirit- 
female we have, whatever its locality. Mr. Roberts’s state- 
ment that some of the species have no inguinal pair is probably 
due to his having attempted to count mamme on fresh speci- 
mens, which is almost as difficult and untrustworthy as doing 
so on skins. Spirit-specimens are alone to be trusted for 
this particular purpose, as every mammalogist of any expe- 
rience 1S aware. 
It may be of interest to record that Cryptomys—or, at 
least, some of the species of it—possesses that most unusual 
structure, an os clitoridis. In a specimen of C. mechow? it is 
an oblong flattened bone 5 mm. in length, 1°5 in breadth, and 
0-5 in thickness, not very much smaller than the corresponding 
both from the same farm at Willbrook, Estcourt (Wroughton § Turner), 
which equally have the extremes as regards the posterior nasal regions, 
other specimens from the same series filling up the gap between them. 
