ON THE VARIATION OF SrECIES. 91 



characters under such circumstances, the proof of the natural 

 origination of a species is complete : it could not be much 

 more so were we able to watch the process step by step. 

 It mioht be objected that the difference between our two 

 species is but slight, and that by classing them as varieties, 

 nothing further Avould be proved by them. But the dif- 

 ferences between them are such as obtain between allied 

 species generally. Large genera are composed, in great part, 

 of such species; and it is interesting to show how the great 

 and beautiful diversity within a large genus is brought about 

 by the working of laws within our comprehension. 



A few remarks on the way races are produced will be 

 here in place. Naturalists have been generally inclined to 

 attribute the formation of local varieties or races of a species 

 to the direct action of physical conditions on individuals 

 belonging to it, which have migrated into new localities. 

 It might be said, therefore, that our Heliconius Thelxiope 

 of the moist forests has resulted from such operation of the 

 local conditions on H. Melpomene, especially as intermediate 

 varieties are fojind in districts of intermediate character and 

 position. It is true that external agencies — such as food 

 and climate, causing delayed or quickened growth — have 

 great effect on insects, acting on their adolescent states, and 

 so by correlation of growth on the shape and colours of 

 the adult forms. But there is no proof that a complete local 

 variety or race has been produced wholly by this means, 

 modifications acquired by individuals not being generally 

 transmissible to offspring. The examination of these races 

 or closely-allied species of Heliconii, with reference to their 

 geographical distribution, throws hght also on this subject. 

 Thus, Heliconius Thelxiope is disseminated over a district 

 2,000 miles in length from east to west, from the mouth of 

 the Amazons to the eastern slopes of the Andes, but shows 



