Dear Mr. Henslow,--You will correctly infer from this address that I shall not be able to attend the Linnean Society meeting on the 16th prox. For two or three years past my health has been breaking up, and several months ago I had a stroke of paralysis. So I have had to knock off all work, and have just arrived here to spend the winter--finding your letter, forwarded from Oxford, awaiting me. It has interested me very much, and some time I should like to see the paper to which it refers, whether in MS. or print. As far as I can gather, you are spontaneously following in the footsteps of Asa Gray, Nägeli, and some other botanists. But, it seems to me, this self-adaptation doctrine is equivalent to an _a priori_ abandoning of all hope to obtain any naturalistic explanation of the phenomena in question. It simply refers the facts of adaptation immediately to some theory of design, and so brings us back again to Paley, Bell, and Chalmers. As when a child asks why a flower closes at night, and we answer him: Because God has made it so, my dear. _C'est magnifique, mais ce n'est pas la science._ But do not mistake me. My quarrel is with the term self-adaptation, which seems to imply causes of a non-naturalistic kind. Which, of course, is quite a different thing from doubting whether the naturalistic explanation given by Darwin is adequate to meet all the facts. I am myself more and more given to question 'the all-sufficiency of natural selection,' and this, whether or not use-inheritance is one of the supplementary factors. But that there are some hitherto undiscovered factors of this kind where many of the phenomena of adaptation are concerned, I am more and more disposed to suspect. Nevertheless I believe, in the light of analogy, that they will all prove to be natural causes, and therefore not correctly definable as due to 'self-adaptation.' My hemiplegia has given me a terrible shake, so I cannot write much. Indeed, this is the longest of the few letters which I have written since my attack. So please excuse seeming bluntness, and believe me to remain, Ever yours, very truly and most interestedly, GEO. J. ROMANES. P.S.--Of course you would not in any case expect to find so much variability of the _conspicuously indefinite_ kind in nature as in cultivation. For, by hypothesis, natural selection is present in the one case (to destroy useless variations) while absent in the other. But I allow this does not apply to the examples you give me. Only remember the point in publishing your paper. Hôtel Costebelle, Hyères: February 10, 1894.