Poulton, Mimicry and Natural Selection. - 
posterior part of the visible dorsal surface black. It contains 
many species of the Lycid models, and also Coleoptera belonging 
to the Telephoridae, Melyridae, Phytophaga, Cantharidae, 
and Longicorns, several species of aculeate Hymenoptera, a few 
Hemipterous insects, two species of Lepidoptera Heterocera and 
one of Diptera. We have here all kinds of habits and all kinds 
of life-histories, larvae living in the open, larvae burrowing in 
plant-stems, carnivorous larvae, leaf-eating larvae, larvae with special 
food stored in cells. It is simply childish to appeal vaguely to 
the direct action of like forces as the explanation of the remarkable 
likeness which runs through the group: for the environing forces 
are not like but extremely unlike, because of the very diverse 
conditions under which various members of the group live and 
grow. 
All the butterfly sub-families which furnish the chief models 
for Mimicry are remarkable for a uniformity of colour and pattern 
among groups of species in each of the countries they inhabit. These 
sub-families are the Danainae found all over the tropics, and 
the allied Ithomiinae (Neotropinae) of tropical America, the 
Acraeinae almost confined to Africa and tropical America, and the 
allied Heliconinae practically restricted to the latter. A very strong 
family likeness runs through long series of species, as anyone may 
see by a glance at the successive drawers of a collection of African 
Acraeinae or Oriental E uploeina and comparing them with an 
equal number of species in any sub-family which does not provide 
models for Mimicry. Compare for instance our European Van- 
essidae with sets of local species of any of the four above- 
named sub-families. The species of Vanessa do indeed possess 
homologous markings!) and many of the gaps between them can 
be filled up, but we have to hunt the world in order to do it, and 
even then we only obtain a partial continuity between extreme 
differences, whereas in the specially protected sub-families there is 
not only continuity but uniformity in large groups of species. Mr. A. 
G. Mayer?) has found that among 450 species of Neotropical 
Ithomiinae and Heliconinae there are only 15 shades of colour, 
whereas among 200 species of Neotropical Papilioninae there 
are 36 shades. And this is not by any means due to the scarcity 
of variation in the former; for individual differences in each 
locality, aud geographical differences, as we pass from one district 
to another, are very prevalent. Combined with the uniformity 
within these sub-families is a marked tendency to resemble other 
protected sub- families within the same region, a tendency which 
is so pronounced in the case of the Ithomiinae and Helico- 
ninae that they were long regarded as a single group although 
1) See F. A. Dixey in Trans. Entom,. Soc. London, 1890, p. 89. 
2) Bulletin of the Mus. of Comp. Zool. at Harvard Coll., Feb. 1897, p. 169. 
