( Ixxxi ) 



these products and the mechanism which produces them as 

 the result of natural selection acting upon physiological 

 variability as a basis. Whether these defensive secretions 

 have been directly developed for that particular purpose, or 

 Avhether, as appears more probable, a waste-product of 

 metabolism has been utilised, is at present an open question, 

 but the main contention that physiological processes can be 

 brought under the influence of natural selection is not 

 thereby influenced. 



Incases such as have now been considered there are outward, 

 visible, and therefore " specific " characters in the systematic 

 sense, such as colour, the formic acid gland, etc., which may 

 be regarded as the correlates of special physiological pro- 

 cesses. But if external characters such as these, which are 

 significant, can be legitimately correlated with special physio- 

 logical activities, is it not reasonable to suspect that there 

 may exist in any species special physiological activities which 

 are equally correlated with external non-significant characters 

 which have been allowed to persist simply because they are 

 indifferent ? It seems to me that such a suspicion would be 

 well founded. It is not going too far to say that the pigment 

 colours of organisms generally are due to their special phy- 

 siological activities, and any modification of these activities 

 would result in a modification of colour. Now, those external 

 conditions classed under the action of food and climate, etc., 

 are just those conditions which might ba expected to affect 

 physiological processes. If these processes are thus affected, 

 a local race, or "geographical species," or, as far as we 

 know, a permanent species, might be produced by a true pro- 

 cess of physiological selection — not in the sense used by the 

 late Dr. Eomanes— but by the ordinary operation of natural 

 selection. The suggestion is, I venture to think, worthy of 

 serious consideration. The mode of action may be rendered 

 more explicit by an example. 



Supposing a butterfly physiologically susceptible to climate, 

 i.e., a species whose special vital chemical processes are upset 

 in the pupal state by change of temperature, degree of 

 humidity, etc., ranges into new districts, or experiences a 

 change of climate in its own region. Since no two indi- 



