1856] GEOGRAPHICAL DISTRIBUTION. 437 



C. Darwin to Asa Gray. 



Down, July 20th [1856]. 



... It is not a little egotistical, but I should like to tell 

 you (and I do not think I have) how I view my work. 

 Nineteen years (!) ago it occurred to me that whilst otherwise 

 employed on Nat. Hist., I might perhaps do good if I noted 

 any sort of facts bearing on the question of the origin of 

 species, and this I have since been doing. Either species 

 have been independently created, or they have descended 

 from other species, Hke varieties from one species. I think it 

 can be shown to be probable that man gets his most distinct 

 varieties by preserving such as arise best worth keeping and 

 destroying the others, but I should fill a quire if I were to go 

 on. To be brief, I assume that species arise like our domestic 

 varieties with much extinction ; and then test this hypothesis 

 by comparison with as many general and pretty well-estab- 

 lished propositions as I can find made out, — in geographical 

 distribution, geological history, affinities, &c., &:c. And it 

 seems to me that, supposing that such hypothesis were to 

 explain such general propositions, we ought, in accordance 

 with the common way of following all sciences, to admit it till 

 some better hypothesis be found out. For to my mind to 

 say that species were created so and so is no scientific expla- 

 nation, only a reverent way of saying it is so and so. But it 

 is nonsensical trying to show how I try to proceed in the 

 compass of a note. But as an honest man, I must tell you 

 that I have come to the heterodox conclusion that there are 

 no such things as independently created species — that species 

 are only strongly defined varieties. I know that this will' 

 make you despise me. I do not much underrate the many 

 huge difficulties on this view, but yet it seems to me to explain 

 too much, otherwise inexplicable, to be false. Just to allude 

 to one point in your last note, viz., about species of the same 

 genus getierally having a common or continuous area ; if they 

 are actual lineal descendants of one species, this of course 

 would be the case ; and the sadly too many exceptions (for 



