July 30, 1896] 



NA TURE 



293 



LETTERS TO THE EDITOR. 

 ( The Editor does not hold himself responsible for opinions ex- 

 pressed 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.'\ 



The Utility of Specific Characters. 

 I HOI'E that my friend Prof. Lankester will forgive me if I 

 find myself unable to accept the version which he has published 

 of a portion of the remarks which I made at the recent discussion 

 at the Linnean Society. 



I entirely agree with Mr. Mivart, that the " Problem of 

 Utility" really involves the validity of the Darwinian theory. 



I staled in my remarks what I have often stated before, that 

 I was more and more confirmed in the belief that specific 

 characters in flowering plants are utilitarian. I showed that this 

 was obviously so in familiar cases. If time had allowed, I 

 might have strengthened my position by reference to the large 

 amount of important and convincing work which has been done 

 in this direction in other countries. At home, for reasons which 

 are not far to seek, this kind of research is now almost entirely 

 neglected. The result is that the Darwinian theory of organic 

 evolution seems hardly to have a convinced supporter left except 

 Mr. Wallace. In its place we have the "Physiological Selec- 

 tion" of Dr. Romanes, the " Discontinuous Variation " of Mr. 

 Bateson, and, last of all, the extended " Correlation Principle" 

 of Prof. Lankester. A common feature of each is their more or 

 less definite rejection of the principle of utility as accounting for 

 specific characters. 



I was examining with Mr. Darwin, at Kew, a collection of 

 Pitcher-plants (Nepenthes). The specific differences lie mainly 

 in the apjiendages of the pitchers. I hazarded the remark that 

 il seemed hopeless to attempt to explain these on teleological 

 grounds, and that it was difficult to believe that the differences 

 were not due to merely fortuitous variation. Mr. Darwin 

 replied, that he was not prepared to admit it. He gave two 

 reasons: (l) that many plant-structures which at first sight it 

 was scarcely conceivable could be adaptive, had been proved to 

 be so in the most unexpected manner ; (2) that to assume that 

 phenomena were not susceptible of explanation, was to .shut the 

 door to discovery. 



Now I stated, perhaps unnecessarily, that the two reasons were 

 not on the same plane. The first implicitly asserts an inductive 

 principle, the probability of which, it seems to me, subsequent 

 research confirms every day ; the second, I described as of a 

 moral kind. It would have been better to have spoken of it as 

 t'thical, or as a counsel of scientific prudence. In point of fact, 

 I never used the word "immorality"; that was imported into 

 the discussion, and I think in a somewhat sarcastic spirit, by 

 Prof. Lankester himself. 



A more serious point, however, is this. Prof. Lankester quotes 

 me as the authority for the statement that Mr. Darwin " appears 

 to have dejirecated . . . the invocation of this theory of cor- 

 relation as an explanation of cases of apparently useless parts in 

 animals and plants." Now I made no reference whatever to 

 correlation, which I do not think enters into the particular case I 

 referred to. The question I put to Mr. Darwin amounted 

 simply to this : — "Is it probable that these specific differences 

 will turn out to be adaptive ? " And his reply was, in effect : "I 

 think it is." 



I confess that the use to which Prof. Lankester has turned the 

 correlation principle fills me with some surprise. As with every 

 difiicully which is from time to lime brought up against the 

 Darwinian theory, it will be found that Mr. Darwin has 

 thoroughly considered the matter himself, and has said pretty 

 much all that is to be said about il. The whole subject is ex- 

 haustively discus.sed in the twenty-fifth chapter of "Animals and 

 Plants under Domestication." He treats at considerable length 

 " the cases in which we can partly understand the bond of con- 

 nection," and then gives more briefly " the cases in which we 

 <;annot even conjecture, or can only very obscurely see, what is 

 the nature of the bond." This is characteristic of Mr. Darwin's 

 fairness ; he adopted precisely the same method in regard to 

 cases which seemed to make for Lamarckism, and were not 

 readily explicable by the principle of natural selection. Most of 

 these have been since cleared up, and I do not doubt that the 

 same thing will happen in regard to correlation. The nexus, 

 which is now obscure, will sooner or later be revealed. 



NO. 1396, VOL. 54] 



The animal organism is a " complex " which, in the vast 

 majority of cases, we are far from understanding. It has under- 

 gone in a high degree what Mr. Herbert Spencer calls " in- 

 tegration. " It is not surprising, therefore, that organic correlation 

 has been obscured. Prof Lankester lays it down, that "pre- 

 sumably also plants " are in the .same predicament. But I do 

 not admit this. In plants integration has not been carried to 

 anything like the same extent. For many purposes of biological 

 research I therefore hold with Mr. Darw'in, that plants are better 

 subjects of investigation than animals, because the phenomena 

 are less complex. 



The result is that in plants most cases of correlated variation are 

 at once explicable. All the appendicular organs are homologous. 

 A variation which affects one runs through the whole. Amongst 

 many thousand Snowdroj^is from Asia Minor, grown at Kew this 

 spring, a few had exceptionally broad leaves ; this was accom- 

 panied by a corresponding dilatation of the perianth-segments. 

 There is a variety of the common oak with marbled foliage. A 

 tree at Tortworth has borne acorns, and these are striped. At 

 first sight it might seem odd that a variation in foliage and fruit 

 should be correlated. But it is not so : the marbling is due to 

 the partial suppression of chlorophyll in those portions of the 

 ground-tissue which are exposed to light ; and this tract of tissue 

 is continuous in the leaves and the carpels. 



I cannot but think that even in animals, of which I know 

 little. Prof Lankester is building on a rash foundation in 

 attempting to generalise widely from cases in which, in the 

 light of present knowledge, the obvious but, as he thinks, useless 

 distinctive character may (or may not) be linked with the un- 

 obvious adaptive variation. 



It seems to me that when an explicable correlation persists in 

 a species, we are not justified in assuming any part of the chain 

 to be useless. The whole is, in fact, part of the specific 

 character ; and this was what I took to be Prof Weldon's point. 

 I do not see that our ignorance of the nature of the "bond" 

 makes any difference ; nor do I see how Prof Lankester ex- 

 tricates himself from the effect of his admission that the parts of 

 the chain are always subject to selection. It seems to me that 

 we are justified in inferring that what survive as specific differ- 

 ences do so because they are useful. 



I doubt if the case which has so impressed him is a very satis- 

 factory one. He thinks that in " tropical regions" the colour of 

 the skin is linked with the chemical activity of the leucocytes in the 

 blood. Assuming that immunity from fever is due to the latter, 

 he infers that the former is not a "useful character." This is, 

 in the present state of our knowledge, taking a good deal for 

 granted. But I frankly admit that such a case, if completely 

 established, would give the utility of specific characters, and 

 with it the Darwinian theory, a serious blow ; and Prof. 

 Lankester would have the satisfaction of arriving at the same 

 result as Dr. Romanes, but by a different path. 



But is he sure of his ground ? Mr. Darwin touches on the 

 connection between "complexion and constitution," but does 

 not appear to think the evidence points to any definite conclu- 

 sion. Nor, I confess, do I, from such facts as are within my 

 knowledge. I have sent a good many men from Kew to Africa, 

 and the belief of my staff is that fair men enjoy better health 

 than dark. But I do not consider that the data are sufficient. 

 On the other hand, men of African descent, transferred from the 

 West Indies to Africa, are said to be more susceptible to febrile 

 maladies than the natives. Certainly the natives of India do 

 not appear to enjoy any immunity from fever. " It is," says Sir 

 Clements Markham, " by far the most prolific cause of death, 

 carrying off . . . very many more than all other disease and 

 accidents put together " (except cholera). I cannot but be 

 impressed with the fact, because it was to combat this state of 

 things that the Government of India introduced Cinchona culti- 

 vation, by far the most important enterprise in which Kew ever 

 took part. 



It appears to me that the relation of a stationary population 

 to local febrile diseases is governed by natural selection, and 

 has possibly nothing to do with epidermal pigment. The more 

 susceptible die oft', the more immune survive. Variation in the 

 phagocytes would do the whole business. In this way disease 

 and population reach an equilibrium. In some cases a disease 

 actually attenuates, to recover its virulence when, as in the case 

 of measles in Fiji, it reaches new ground. At any rate, I think 

 Wells's theory can hardly be accepted as a scientific fact. But 

 it does not follow that epidermal pigment is useless because one 

 explanation of it seems to fail. 



