104 



NATURE 



[January 26, iq22 



Letters to the Editor. 



\The Editor does not hold himself responsible for opinions 

 expressed by his correspondents. Neither can he undertake 

 to return, or to correspond with the writers of, rejected 

 manuscripts intended for this or any other part of Nature. 

 No notice is taken of anonymous communications.^ 



Some Problems in Evolution. 



Since I am not, in the ordinary meaning, a bio- 

 logist, I have sometimes difficulty in understanding 

 biological language. • Doubtless, also, I am often 

 ignorant of recent developments in knowledge and 

 thought. But certain problems of disease and educa- 

 tion interest me, and I cannot get on with them 

 unless some points, essentially biological, are cleared 

 up. In the hope of enlightenment I wrote to Nature. 

 Immediately the discussion became acrimonious : at 

 least, I became acrimonious, I was told, in effect, 

 that I had no business in the august deliberations of 

 biologists. It is not in human nature, or my variety 

 of it, to accept that pontifical attitude. However, 

 there seems now some prospect of the desired lucidity, 

 and I shall be very ready to accept it with an humble 

 and a contrite heart. 



I fear, however, that Dr. Cunningham's letter in 

 Nature of January 12 does not greatly help. He 

 writes :— " Sir Archdall Reid argues, as though it 

 were a remarkable discovery, that characters are not 

 present as such in the fertilised ovum from which an 

 organism develops." But is that quite fair? I 

 argued only that if, as all biologists are aware, no 

 characters as such are present in the germ, then it 

 must follow that, in the case of any and every 

 character, nothing but germinal potentiality (pre- 

 disposition, diathesis, capacity, ability) to produce (in 

 response to fitting nurture) can 'be transmitted ; 

 whence it follows further that all characters are alike 

 as regards innateness, acquiredness, and inheritability ; 

 whence, again, it follows that if we classify characters 

 with respect to these qualities, there is, as Prof. 

 Goodrich says, only one kind of character. On the 

 other hand, as all biologists know (I protest I do 

 not claim this as a new discovery), there are two 

 kinds of variations : (a) those which result from 

 germinal, and (6) those which result from nurtural, 

 differences. 



Of course, we can classify characters in all sorts 

 of ways, useful and useless— according to colour, 

 weight, size, shape, obviousness of recapitulation, 

 frequency of reproduction, and so on. In a classifica- 

 tion which physiologists have found useful, characters 

 are ranked according to the influences which cause 

 them to develop. This tabulation has the merit of 

 forcing the inquirer to bear in mind the plain truth 

 that frequency of reproduction depends (except when 

 germinal^ variations occur) altogether on the frequency 

 with which fitting nurtures are experienced, and not 

 at all on^ the frequency of inheritance. For example, 

 under this scheme of classification the inquirer bears 

 in mind that rose comb and single comb in poultry 

 are not more inheritable than corns on oarsmen's 

 hands, but that they are more frequently reproduced 

 only because the proper nurture is more frequently 

 experienced. With respect to inheritance, his mind is 

 fixed on the nature of the individual (the germ- 

 plasm); with respect to reproduction, on the nurture 

 received. Moreover, the student is compelled to 

 realise that when he transfers the distinguishing 

 terms "innate," "acquired," and "inheritable" from 

 likenesses and differences between individuals to the 

 characters in which those likenesses and differences 

 are revealed, he has shifted his ground. It is one 

 thing to compare separate individuals, and quite 

 NO. 2726, VOL. 109] 



another thing to compare characters which may 

 occur in the same individual. The old terms may 

 still be applicable; but that is the question which 

 has been raised. It will be gathered that they do 

 not seem applicable to me, and that their constant 

 and (to me) inexplicable transference is one of the 

 causes of my puzzlement. It may be noted also that 

 Darwin, in all that remains oermanent of his work, 

 used these terms in relation to variations, while 

 Lamarck and Weismann applied them especially to 

 characters. I may be mistaken, but I believe that I 

 am right when I say that no one (including Darwin) 

 has ever doubted the all-suflficiency of natural selection 

 unless he has, in his thinking, transferred the terms 

 "innate," "acquired," and "inheritable" from 

 variations to characters, or has confused inheritance 

 with reproduction. 



Again, we may employ our words with unusual 

 meanings and reason on that basis. Thus "inherit " 

 may be used in the sense of "reproduce," when, of 

 course, the "intensity of inheritance" of combs is 

 infinitely greater than that of corns. But now we 

 are asking for trouble and in sight of confusion. We 

 are in danger of using as counters in thought and 

 discussion, not realities in nature, but mere words. 

 Our inquiries,, notwithstanding our language, relate 

 not to the natures of individuals, but to their nurtures. 

 Does or does not the impure dominant inherit the 

 recessive trait which it does not reproduce? Does 

 the pure extracted recessive which is unlike its 

 parent inherit nothing? When a pigeon or a fowl 

 belonging to a fancy breed reproduces the wild ances- 

 tral coloration, from whom does it inherit? From 

 an exceedingly remote ancestor? It passes my non- 

 biological comprehension to understand how an indi- 

 vidual can inherit except through, and therefore 

 from, his parent. In practice the difficulty is sur- 

 mounted by using "inherit" with the usual, or with 

 the unusual, meaning as exigencies of argument dic- 

 tate. For example. Dr. Cunningham employs the 

 word with the ordinary meaning when he declares "a 

 character may be inherited when ii is apparent only 

 in one parent or in neither," and with the unusual 

 meaning when he insists that combs are more inherit- 

 able than corns. 



Consider the Lamarckian dictum: "Acquired as 

 well as innate characters are inheritable"; and the 

 neo-Darwinian : " Innate, but not acquired, characters 

 are inheritable. " What do " innate " and " acquired " 

 mean here? No one can tell. Definitions are im- 

 possible, for none can be framed which cover the 

 whole of common and accepted usage. What does 

 "inherit" mean? When applied to "innate" 

 characters it may have, as already indicated, its 

 ordinary meaning, or it may mean "reproduce." If 

 a cock reproduced a comb under the same conditions 

 as those in which its parent produced it (in response 

 to similar nurture) all biologists would regard the 

 comb as inherited — and rightly, for reproduction 

 under the same conditions implies inheritance, 

 though inheritance does not necessarily imply repro- 

 duction. The case is different with resp>ect to 

 "acquired" characters. If a dhild reproduced an 

 oarsman's corn under the same conditions as the 

 parent produced it, few biologists would regard the 

 corn as inherited. It would be regarded as inherited 

 only if the child developed it under conditions in 

 which the parent did not and could not have developed 

 it. ,The word now means "vary,'' i.e. non-inherit, 

 for non-inheritance is variation. It seems, then, that 

 an acquired character is not inherited when it is in- 

 herited, and is inherited when it Is not inherited — 

 i.e. a single word In a single sentence has two con- 



