2124. [5-6-03] has convinced me of the importance of mutations in the productions of specific differences. To say that you are not clear as to what I mean by direct action of Environment, and add "I see no evidence at present that its influence is very great, is permanent in a stock, unless selections that place" & further, "Precisely similar difficulties arise in the code of the inheritance of acquired characters, "which I fancy some of our readers might, I believe seriously, consider to be carried by your paragraph a 'direct action' of Env't", I use the phrase as a half century of spectacular writers has used it in contrast to an indirect (e.g., selective) effect of Environment. That the developing organism is altered with changing environmental conditions has, of course, been shown hundreds of times; and the modification persists so long as the particular conditions do. And this modification may take place in one generation without deaths, as I have shown experimentally in protozoa. Of course, [crossed out 'the'] when the Environment changes again the organisms vary about a new centre, but often retain for some generations the effect of the former conditions. This has been beautifully shown by the work of [?]Stacedfuss of Zurich and of Fischer on Lipidoptera. And as for the inheritance of the acquired, I am quite complaisant at being charged with thinking that possible and even proved for certain cases. The illustrated case you would find so hard to accept has, I believe, been so nearly realized experimentally[?] that I think the burden of experimental proofs rests on those who deny the validity of the results. It seems to me that the experimental evidence that is lacking is not for mutation or for direct action of Environment as factors of specific change, but for selection of [end]