WCP4220

[1]

Parkstone, Dorset

Feb[ruar]y 18^th 1891

My dear Mr Cockerell

As you are a student of variation I thought you would see my point without explanation. Now I will explain. The following three points I consider to be proved by overwhelming evidence a summary of which is given in "Darwinism" Chap[ter]. III.

(1). All increasing or dominant species’ (& it is from there that new species arise) vary considerably, in all their parts organs & faculties, in every generation.

(2). The amount of this variation is so large that when only 20 to 50 adults are compared it reaches from 10 to 20 per cent of the mean value of such characters as can be accurately measured.

(3). The proportion of individuals which vary considerably is large, reaching to one [2] fourth, or one third of the whole number compared. In other words, the curve of variation is low —

Thus

not thus.

[two diagrams, one illustrating the curve of variation Wallace believes is correct and one illustrating a curve of variation Wallace believes is incorrect.]

Hence it follows that whatever character is increased or diminished in individuals by the effect of the environment, a similar increase or diminution will occur by genetic variation, in each generation, and in certainly 5 or 10 per cent of the individuals dealt with. Hence your supposition that in the check lots no such modification would occur as in those exposed to special conditions [3] is almost an impossible one, and an effect produced on one or even on five or 10 per cent by special conditions would be imperceptible, because similar effects would occur through normal variation & often to a much greater amount. Hence I said, that to be clear and decisive the effect produced by the conditions should be inherited by a large proportion of the offspring. You may say that the effects of conditions would be additional to the normal effects of variation. True. And if largely inherited they would soon show it, but if as you first supposed only one per cent, that would be entirely swamped by the irregularities of normal variation & inheritance. You must remember too that [4] experiments on a very large scale, & with check experiments on an equally large scale, & all carried on for many years would require a very large establishment & ample funds not likely to be obtained. Again the whole raison d’etre of this enquiry is to decide whether inheritance of acquired characters is of any importance in the origin of species. To be of importance it must rank in generality with variation, otherwise it is entirely superfluous, even if it exists, and variation could do perfectly well without it. Yet again ^either there is a fundamental cause of such inheritance of there is not. If there is, — if such inheritance is a law of nature why should it not rank with the inheritance of genetic variations? — which are I presume to the extent of about one half? If it were only one per cent it might be a fluke! It would require innumerable experiments to prove¹ it was anything else.

Yours faithfully | Alfred R Wallace [signature]

I² return your note which is good so far as it goes, but I am not sure that it is any good theorising on the subject till we have some facts.