1 877.] Animal Geography. 65 
It has been maintained that, in the NearCtic region at 
least, the Mammalia increase in size with the latitude and 
altitude of their birthplace. This view is disputed by Mr. 
Allen, who declares that it does not agree with the develop- 
ment of the American Carnivora. Indeed as regards alti- 
tude, and consequently rarefaction of the ambient medium, 
the very opposite law has been proposed. The largest ani- 
mals are now found in the denser medium, water ; many of 
the next largest species— such as the elephants, hippopota- 
mus, &c. — inhabit river-marshes and deltas almost on a level 
with the sea. There is also good reason for supposing that 
the atmosphere in the epochs which produced the gigantic 
extindt animals must have been denser than it is at present. 
Mr. Allen considers that the largest individuals of every 
species, and the largest, best-developed, and most typical 
species of every group, will be found near its centre of 
distribution. 
It has also been suggested, at least as regards inseCts, 
that in all regions dark-coloured species are characteristic 
of woods, whilst white or light-coloured forms occur in the 
open plains. This law holds good in some well-known 
genera of butterflies, such as the “ whites” as compared 
with the forest-loving Hipparchias , Erebias, &c. ; but the 
deeply-coloured Vanessas inhabit the open country. Among 
Coleoptera we find the dung-beetles, mainly black in colour, 
abundant in the open country, like the ruminants on whose 
excrement they prey. The common ground-beetles, also 
(Harpalidae, &c.), are chiefly sombre in colour, and certainly 
show no exclusive preference for woods. On the other hand, 
the chafers (Melolonthidae) and the Buprestids frequent the 
woods, and yet display some of the most striking instances 
of pale colouration to be found among the entire tribe of 
Coleoptera. 
For further details on the influence of locality we must 
refer the reader to Mr. Wallace’s most instructive Address, 
but we cannot help expressing our regret that he has not 
introduced the consideration of this subject into his magnum 
opus. 
We may rest assured that the peculiarities to which we 
have so briefly referred point to causes of variation at work 
other than protective mimicry or than sexual selection. 
This is evidently the opinion of Mr. Wallace, who insists 
that “ one of these causes is an influence depending strictly 
on locality, whose nature we cannot yet understand, but 
whose effects are everywhere to be seen when carefully 
VOL. vii. (n.s.) f 
