CORRESPONDENCE 641 



itself is a mere distributor. The variations thus distributed are, as Darwin 

 long ago urged, neutral — of every description, bad, good, and indifferent ; 

 and it is natural selection which eliminates the bad, and so the good 

 are left. 



Prof. MacBride also leaves out of account all reference to modifying 

 factors, of small and yet Mendelian differences, and other points which 

 are of importance for the presentation of the Mendelian case. 



In one specific instance he also, I venture to say, makes a difficulty 

 where none exists. On p. 397 he refers to the well-known experiments whereby 

 a lens is formed from a portion of epidermis from which lenses do not usually 

 develop ; and he raises an argument about the existence of special lens- 

 determiners which " recalls the delicious blends of thirty years ago," when 

 Weismann was to the fore. Surely, if we simply postulate (a) optic cup, 

 (b) epidermis, (c) a substance formed by the optic cup which causes epidermis 

 to develop into lens, we have a formal explanation which is no harder to 

 swallow than many definite facts of physiology — e.g. the specific action of 

 adrenalin. It is, at any rate, infinitely harder to explain the facts from any 

 Lamarckian standpoint ! Similarly with the regeneration of an amphibian 

 leg (p. 396), Prof. MacBride makes his adversaries think in terms of 

 determiners, whereas they would much prefer to do so in terms of organic 

 equilibrium. 



I might also refer to his statement on p. 404 : " It is necessary [to carry 

 out convincing experiments on the inheritance of acquired characters] to find 

 an animal that will respond to a change of environment by a change of habit 

 or structure, and this is by no means easy." To the uninitiated this would 

 appear to detract from the importance of Lamarckism as an evolutionary 

 factor. 



Finally, he makes an assumption as to Guyer's work (p. 405) which appears 

 to me to be wholly unjustified. Guyer states that he has caused hereditary 

 changes in the lens of the eye of rabbits by injecting anti-lens serum. 

 Prof. MacBride's comment is as follows : " If this conclusion be accepted, 

 the cardinal principle of the theory of the inheritability of acquired characters, 

 viz. the influence of the soma on the germ, is conceded" (italics mine). 

 Personally, I had always supposed that the " cardinal principle of the 

 inheritability of acquired characters " was that adaptive somatic changes, 

 especially those concerned with vise and disuse, were transmitted to the 

 offspring in some degree. I am not acquainted with any leading biologist 

 since Weismann's time who would deny that changes in the body might 

 affect the germ-plasm, although the process does not appear to be common. 

 Such a phenomenon, known as somatic induction, is familiar in all the 

 textbooks — but it has very little in common with Lamarckism. If Guyer's 

 results are confirmed, we are brought face to face with new possibilities in 

 biology — the identification of factors in the germ-plasm through the methods 

 of immunology. The reaction would appear to depend on protein specificity, 

 and to be very unhke (to use Prof. MacBride's own argument) anything 

 that might occur in nature. 



In conclusion, I should like to point out that I have not concerned myself 

 to attack Prof. MacBride for upholding this or that doctrine. I have 

 only wished to protest against his belittling the positive achievements of 

 branches of science which do not happen to square with his beliefs, and 

 against his using, as support for the theories which he upholds, results which 

 do not support those theories as they are ordinarily stated. 



With apologies for trespassing upon your space at such a length. 



Yours faithfully, 



Julian S. Huxley. 

 New College, Oxford, 

 January 28, 1921. 



