34 MOSTLY MAMMALS 
The occurrence of bands on the legs and sometimes on 
the shoulders of mules and dun-coloured horses, and like- 
wise the presence of dark bars on the limbs of otherwise 
uniformly coloured species of cats, like the Egyptian cat 
and the bay cat, are further proofs of the same law. 
Moreover, the fact that in the young of pigs—and, to a 
certain extent, those of tapirs—the markings take the form 
of longitudinal stripes, whereas in the more specialised 
deer, whether young or old, they are in the shape of spots 
arranged in more or less well-defined lines, is, so far as it 
goes, a confirmation of the theory that spots are newer 
than stripes. And the presence of transverse stripes in 
the still more highly specialised antelopes tends to support 
the derivation of this type of marking from spots, es- 
pecially if it be remembered that the harnessed antelopes 
are partly spotted. Still, it must be borne in mind that 
these instances apply only to light markings, which, as 
already stated, may have a totally different origin from 
dark ones. 
There are, however, apparently insuperable difficulties as 
regards longitudinal and transverse striping in mammals. 
In the first place, instead of finding a number of the 
polyprotodont, or more primitive marsupials, showing longi- 
tudinal stripes, we have in this group only the three- 
striped and single-striped opossums thus marked, and in 
these the stripes are respectively reduced to the numbers 
indicated by their names. This, however, is not all, for 
the banded ant-eater takes its name from the narrow trans- 
verse white stripes with which the back is marked; while 
the thylacine, which cannot in any sense be regarded as a 
specialised type, is similarly marked with broader dark 
stripes, neither of these animals having any trace of a 
longitudinal stripe down the back. The water-opossum, 
