OF THE AMAZON VALLEY. 511 



prived of them. It is not unreasonable to suppose that some species are taken by in- 

 sectivorous animals, whilst others flying in company with them are avoided. I could 

 not, from their excessive scarcity, ascertain on the spot that the Leptalides were thus 

 picked out. I noticed, however, that other genera of their family (Fieridce) were much 

 persecuted. We have proof, in the case of Sand- Wasps, which provision their nests 

 with insects, that a single species is very generally selected out of numbers, even of the 

 same genus, existing in the same locality. I was quite convinced in the case of Cerceris 

 binodis of South America, which destroys numbers of a Megalostomis (family Clythridce), 

 that the great rarity of the Beetle was owing to its serving as prey to the Cerceris. We 

 cannot point out all the conditions of life of each species concerned in these mimetic 

 analogies. All that we can say is, that some species show, by their great abundance in 

 the adult state, that during this period, before they propagate their kind, they enjoy by 

 some means immunity from effective persecution, and that it is therefore an advantage 

 to others not so fortunate, and otherwise unprovided for, if they are so like as to be 

 mistaken for them. 



The process by which a mimetic analogy is brought about in nature is a problem 

 which involves that of the origin of all species and all adaptations. What I have pre- 

 viously said regarding the variation of species, and the segregation of local races from 

 variations, the change of species of SeliconidcB from one locality to another, and the 

 probable vital necessity of their counterfeits which accompany them keeping to the 

 exact imitation in each locality, has prepared the way to the explanation I have to give. 

 In the cases of local variation of the Seliconidce, there was nothing, as before remarked, 

 very apparent in the conditions of the localities to show why one or more of the varieties 

 should prevail in each over their kindred varieties. There was nothing to show plainly 

 that any cause of the formation of local varieties existed, other than the direct action of 

 physical conditions on the individuals, although this might be seen to be clearly incom- 

 petent to explain the occurrence of several varieties of the same species in one locality. 

 We could only conclude, from the way in which the varieties occur in nature, as described 

 in the case of Mechanitis Polijmnia, that the local conditions favoured the increase of one 

 or more varieties in a district at the expense of the others — the selected ones being 

 different in different districts. What these conditions were, or have been, was not revealed 

 by the facts. With the mimetic species Leptalis Theono'e the case is different. We see 

 here a similar segregation of local forms to that of Mechanitis Polymnia ; but we believe 

 we know the conditions of life of the species, and find that they vary from one locality to 

 another. The existence of the species, in each locality, is seen to depend on its form and 

 colours, or dress, being assimilated to those of the ItJiomice of the same district, which 

 IthomicB are changed from place to place, such assimilation being apparently its only 

 means of escaping extermination by insectivorous animals. Thus we have here the 

 reason why local races are formed out of the natural variations of a species : the question 

 then remains, how is this brought about ? 



The explanation of this seems to be quite clear on the theory of natural selection, as 

 recently expounded by Mr. DarMdn in the ' Origin of Species.' The local varieties or 

 races cannot be supposed to have been formed by the du-ect action of physical conditions 



VOL. XXIII. 3 z 



