C: bs +) 
In both of these cases, and in many others that might be 
adduced, the dependence of resemblance on locality extends 
further than to the general aspect of an assemblage of species 
common to the whole region. It is found that where modifica- 
tions of the pattern occur in different parts of the same ex- 
tended region, the whole number of insects associated by 
participation in the common pattern undergoes the same 
kind of change; they all become similarly modified together. 
One of the most striking instances of this phenomenon is 
afforded by a well-known assemblage of butterflies from 
Central and South America which has been fully dealt with by 
Prof. Poulton. These butterflies are characterised by a 
peculiar arrangement of the colours red, yellow and black; 
and the assemblage contains representatives of many different 
subfamilies, including Ithomiines, Heliconiines, Danaines, 
Nymphalines and Pierines, to say nothing of certain moths. 
The general facies, while retaining its common elements, shows 
certain modifications according to the special locality; one 
such modification being peculiar to Central America, another 
to Venezuela, and others to Trinidad, Guiana, East Brazil, 
the Upper Amazon and Ecuador. These modifications, be it 
understood, are not confined to one or two of the constituent 
members of the combination, but are shared in by all or 
most of the associated species. A somewhat similar phe- 
nomenon may be observed in the parallel African genera 
Mylothris and Phrissura which were mentioned just now. 
Both of these are Pierine genera, though the affinity between 
them is not very close. 
In considering such cases as these, the number of which 
might be almost indefinitely extended, we are naturally 
inclined to ask whether the geographical conditions may 
not have something to do with the resemblances noted. No 
doubt they have; but the question remains whether such 
influence is a direct or an indirect one. That the influence of 
a climatic condition may sometimes be direct is suggested by 
the temperature experiments of many investigators, including 
Merrifield, Standfuss and Fischer. But there are great, and, 
as it would seem, insurmountable difficulties in the way of 
adopting the hypothesis of a direct climatic or geographical 
