November 15, 1894] 



NATURE 



55 



We must all agree with Sir Edward Fry's desire to obtain a 

 clear and exact definition of an "acquired character," as this term 

 has been used in the discussions upon hereditary transmissibility. 

 I do not think, however, that those who have taken part in the 

 various controversies and discussions which have raged inter- 

 mittently during the last seven years, have been misled by the 

 lack of asuthciently exact definition orthemultiplicity of inexact 

 ones. I believe that both sides have known well enough the kind 

 of character which was called acquired, even though no suffi- 

 ciently clear definition was forthcoming. And it may be that this 

 mutual understanding has tended to obscure the demand for a 

 definition. 



An acquired character has generally been brielly defined as 

 "the result of the operation of some external force upon an 

 organism," and I still think that this is as satisfactory as any 

 definition of equal brevity can be. But some want of clearness 

 follows from the elasticity of the word "result." Everything 

 that follows the operation of some external force may be called 

 its "result"; but the definition interpreted in this way would 

 include much that is not within the meaning of the word 

 "acquired." Some increased precision may be added by using 

 the words "direct result"; but a perfectly satisfactory defini- 

 tion should, I think, imply the admission that the result (in its 

 wide sense) of an external force on an organism must always 

 contain elements which are not due to the force — which are not 

 acquired — as well a- those which are due to the force and which 

 are acquired. I think that the following definition will meet 

 the case; "Whenever an organism reacts under an external 

 force, that part of the reaction which is directly due to the force 

 is an acquired character. ' 



In many cases the external force acts only as a shock, with 

 the starling of reaction as its only direct result. In such a case 

 the occurrence of the reaction, as contrasted with the sequence 

 of events which make up the reaction itself, is the acquired 

 character. In examples such as these, those who maintain the 

 transmission of acquired characters would be required to prove 

 that the reaction which could only be started by an external 

 force in the parent, started without this stimulus in the off- 

 spring. 



I believe the definition suggested above meets all Sir 

 Edward Fry's conditions — viz., that it includes all "acquired" 

 characters, and excludes all that are not acquired ; that it is 

 physical and not metaphysical ; that it is not "stated in terms 

 derived from hereditability or the contrary, or in terms of any 

 hypothesis or theory " ; and that it admits of ascertainment 

 and verification. 



That a reaction under an external force is compounded of 

 two parts, due respectively to the body which reacts, and 

 to the force which causes the reaction, is a fact and 

 not a theory or hypothesis. It may be urged, however, 

 that the separation of the two constituents does not admit of 

 ** ascertainment and verification." This may be true, in the 

 present state of our knowledge, for certain cases ; and if so, these 

 cases would be unsuitable for the purposes of an inquiry 

 into the transmissibility of acquired characters. But I do not 

 admit that it is proved that the two constituents of the reaction 

 cannot be separated in every case by a sufficiently careful 

 investigation. For the purposes of this inquiry it is sufficient, 

 however, if we can prove beyond doubt that some part of a 

 reaction is the direct result of an external force, even if we 

 have not thereby exhausted the whole of the direct results 

 contained in the reaction. For if this can be done in a vast 

 number of cases, an immense body of evidence will be provided, 

 and we may expect that, if acquired characters are traus- 

 missible, some proof will be forthcoming. 



I propose to test the elficiency of the definition given above, 

 by showing how it can be applied to some of the examples given 

 in Sir Edward Fry's letter. 



In the case of the "exercierknochen " it is clear that 

 the occurrence of the reaction — the existence of the bony 

 growth — is the direct result of the external force. Here then 

 is an acquired character which will be admitted by everyone, 

 which can be witnessed in a vast number of examples, and which 

 can be conveniently applied to test the transmissibility of such 

 characters. There may, or may not, be other direct results 

 contained in the reaction: some of I he processes of osseous 

 growth may have followed directly from continuous or inter- 

 mittent pressure. But in the first place the verification is much 

 more difficult, although not, I believe, beyond the reach of 

 scientific inciuiry, and, in the second place, such proof, if 



NO. 1307, VOL. 51] 



obtained, would yield evidence which would be far more 

 difficult to obtain. in very large quantity. 



It is clear that when Prof. Weismann admits that " the 

 periodical change of leaf in temperate climates has been 

 produced in relation to the recurring alternation of summer and 

 winter," he is referring to the selection of inherent characters, 

 and not the production of acquired characters. The sentences 

 which follow the one quoted (p. 406), leave no doubt upon this 

 point. Sir Edward Fry may feel assured that when any direct 

 results of heat, cold, air, food, moisture, gravity, or light upon 

 the organism are proved to be heritable, the controversy is at an 

 end. 



The case of geotropism logically resembles that of the 

 "exercierknochen." iThe occurrence of the reaction is 

 certainly a direct result of the external force — an acquired 

 character ; and here too we have an immense body of evidence 

 to which to appeal, and which points only in one direction. In 

 spite of the innumerable generations during which plants have 

 assumed certain relative proportions under the influence of 

 gravity, this influence is just as necessary to-day as it has ever 

 been, and the youngest generation starts unbiassed by the direct 

 result of external forces upon its ancestors. 



As regards the "extra fingers or toes, patches of grey hair, 

 moles, cStc," the question is raised as to whether external forces 

 are not involved as direct causes. If this can be proved the 

 question at issue is settled, for such characters are known to be 

 transmissible. If not, the observation merely shows us that 

 certain characters, not proved to be acquired, are transmissible. 

 But if the non-transmi.ssibility of those proved to be acquired 

 has been established on a sufficiently large scale, then the 

 observation in question, accompanied by the continued absence 

 of proof that the characters in question are acquired, may be 

 fairly held to indicate the existence of two contrasted classes of 

 characters, which we may call spontaneous or inherent, and 

 acquired. 



We are asked if we have any scientific knowledge of the 

 organic world independently of any external influence. This 

 method of eliciting an answer must not be allowed to disguise, 

 as it appears to do, the very positive knowledge we possess of 

 the separate effects of the several external influences. This is a 

 legitimate province of scientific investigation, and a large 

 amount of research at the present day is devoted to such 

 questions. 



In handwriting the two constituents of the reaction are some- 

 what difficult, but by no means impossible, to distinguish. The 

 external influence of training operates upon the most complex 

 part of the organism, the nervous system, which again directs 

 the muscular system. Is the style of handwriting due to the 

 external force, or the organism which reacts? We can eliminate 

 pen, ink, and paper as influences by only considering the cases 

 in which these have been identical. There remains the influence 

 of the teacher, and in order to prove that this has been the 

 direct cause of style, it must be shown that the teacher had 

 produced the same style in many pupils. If a style so produced 

 became hereditary, evidence of transmissibility of an acquired 

 character would be provided. Conversely, variety of style 

 under the same conditions of teaching, &c., would favour the 

 view that we are not dealing with an acquired character in this 

 part of the reaction. 



It is unnecessary to consider further the cases of mutilation 

 and wounds, for I imagine that Prof. Weismann, and all who 

 agree with him on this subject, will be willing to accept the 

 clear statements of Sir Edward Fry's letter. " What the 

 organism transmits is the capacity or predisposition, and not the 

 actual result ol the reaction." The latter in these cases is an 

 acquired chan^cter, while no one has ever shown that there is 

 any probability that the former is acquired. 



I have, in this letter, avoided reference to many points raised 

 by Sir Edward Fry, not from want of interest or inclination, but 

 in order to keep to the main issue — the attempt to furnish a 

 clear definition of the class of characters in question. 



If acquired characters are transmissible, we must expect that 

 sooner or later among the vast body of characters which are 

 or will be admitted on all hands to be acquired, some valid 

 instances of hereditary transmission will be furthcoming. 



Such cases as that mentioned by Dr. Ilill in Nature of 

 October 25, when on a sufficient scale andadequately sifted, would 

 supply the requisite evidence. But up to the present such satis- 

 factory evidence has not been forthcoming, although it has been 

 sought for by many observers. Edward B. PoultoN. 



Oxford, November 4, 



