To Asa Gray 24 August 
Down Farnborough Kent
My dear Dr. Gray
I received your letter1 about a fortnight ago & immediately forwarded the note to Hooker, who since then has started for a tour in Germany with Dr. Lindley’s eldest son.—
I am very sorry you shd. have had such trouble about the Dytiscus,2 but it was not my causing.
And now I really hardly know how to thank you enough for the very great trouble which the list of “close species” must have caused you.—3 What knowledge & labour & judgment is condensed in that little sheet of note-paper! I fear that you will think the object not at all worth the labour; but I can only say that if I could have done it myself, I would have done it, had it caused me ten times the labour which it must have caused you. I had met with a remark by Fries that the species of large genera are more closely related to each other, than are the species of small genera.4 I consulted a very good entomologist,5 & Hooker & Bentham, & they did not at all believe in this. But several facts & considerations, nevertheless, made me think that there might be some truth in it; and all general statements of such kind it is my object in my eclectic, peddling sort of work to test. It occurred to me that if I could get some good systematists, not species-splitters, to mark (without the object being known) the close species in a list; then if I counted the average number of the species in such genera, & compared it with the general average (for this end all the genera with single species have to be omitted; & I have omitted Salix & Carex also) of the species to the genera in the same country; it would, to a certain extent, tell whether on average the close species occurred in the larger genera.
Now in your M.S list (Salix & Carex being omitted) there are 115 genera & these have 6.37 species to genus: whereas in your Manual (omitting Salix & Carex & all genera with single species) the average is 4.67. So that it seems that when many organic forms are allied, making what is called a genus, some of them are apt to be more closely allied than are the species in the smaller genera.6 Mr. H. C. Watson has marked for me the British Flora, & the same result is given. I know how vague all such results must be, & there may be some fallacy (should the fallacy be apparent to you, I shd be most grateful to be informed.) in the result, but I cannot detect it; & am inclined to believe that the above proposition may be trusted; but I shall of course try to test it by other means.—
Pray accept my true & cordial thanks for all your very great kindness, & believe me, Your’s very truly | Ch. Darwin
Have you seen Decandolles new great work on Geograph. Distrib.7 Hooker thinks very well of it.—8
P.S. | As it is ten to one you will not be able to give me any facts on the following head, I venture to put it before you for the chance.—
I have met several (chiefly amongst animals) cases, so many that I can hardly think it purely accidental, in which, when the species of a genus differed in some organ or part, which is usually constant in the species of the same genus, then that one or more of the species individually varied in some degree in this same organ or character.9 Thus to give a very bad example, I remember seeing it somewhere asserted that the position of the embryo differed in a very unusual manner in the species of Helianthemum; & hence, if there be any truth in the above rule, some of the species might be expected to vary in this same respect.—
To give good example in animal kingdom in Pyrgoma (a cirripede) the opercular valves (usually very constant) differ wonderfully in the different species, & in some of the species, these valves, (usually so constant) vary in the most perplexing manner, so that they differ more in the very same species, than often in different species of different genera.—10
"Close" species in large and small genera.
Alphonse de Candolle on geographical distribution [Géographie botanique raisonnée (1855)].