November io, 192 i] 



NATURE 



335 



Letters to the Editor. 



{The Editor does not hold himself responsible for opinions 

 expressed by his correspondents. Seither 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 Age of the Earth. 



At the discussion on the age of the earth at Edin- 

 burgh we were unfortunately prevented from hearing 

 Dr. JefTreys's contribution owing to lack of time. 1 

 have, however, read it with interest in Nature of 

 October 27. Mv only object in writing now is to 

 demur to an allusion in it: "Lord Rayleigh's sug- 

 gestion that the earth must be becoming hotter." 

 The words I used were carefully qualified, and I do 

 not think that I can do better than quote them : "If 

 [the radio-active materials present in the earth] are 

 generating more [heat than is now leaking out from 

 the earth] (and there is evidence to suggest that 

 they are), the temperature must, according to all 

 received views, be rising." 



I appreciate the difficulty of such a supposition as 

 clearly as Dr. Jeffreys. ' Rayleigh. 



Terling Place, Witham, Essex, October 31. 



Inheritance, Mendelism, and Mutation. 



In Nature (August 18, pp. 780-84) appears an 

 article by Prof. Goldschmidt on "The Determination 

 of Sex." The author supposes he is dealing with 

 inheritance. At any rate, Mendelians suppose they 

 deal with inheritance, and he declares "we may safely 

 say that to-day, in the light of Mendelism and the 

 work accomplished in the realms of cytology, the 

 problem is solved as completely as the methods of 

 biology permit." Very probably tne odd chromosome 

 of which he writes has influence ; but very massive 

 evidence indicates that sex is determined mainly, not 

 by the nature, but by the nurture of the individual — 

 by such things as hormcHies from the - sex-glands, 

 food, temperature, and the like. In other words, it 

 is certain that both sets of sexual characters are 

 inherited by the individual, but, except in true 

 hermaphrodites and in " intersexuality " (abnormal 

 blending), only one set is reproduced. Male is un- 

 developed female, and vice versa. There is alternative 

 patency and latency, alternate reproduction, not alter- 

 nate inheritance. But the alternation is often irre- 

 gular. For example, in aphides, during a long and 

 indeterminate series of generations, the male char- 

 acters are latent. 



So far as it is possible to judge from ordinary ex- 

 perience of life, offspring appear to blend most 

 parental characters, as is best seen when, within 

 limits, the parents are much unlike. For example, 

 when blue-eyed Scandinavian crosses with black-eved 

 negro, the result is a mulatto, who, though he may 

 incline more to one parent than to the other, is 

 consuicuouslv a blend. Mulattos, mated together, 

 continue to breed true, but if succeeding matings be 

 with pure white the touch of the tar-brush grows 

 fainter generation after generation. In sexual char- 

 acters and eve-dolour there is, however, no apparent 

 blending. The child is male or female, with, almost 

 invariably, black eyes — black dominating over blue. 

 But do sexual characters and eve-colour furnish evi- 

 dence that there is here no blending? Is it reasonable 

 to expect the male characters to blend with female 

 characters? Is this what happens in the case of trulv 

 hermaohrodite animals and olants when both sets of 

 sexual organs are patent? For example, do stamens 

 blend with pistils? Plainly, except in cases of "inter- 

 NO. 2715. VOL. 108] 



sexuality," when the male and female characters do 



blend more or less, if there is any blending, it is 

 between the patent male characters of the one parent 

 and the latent male characters of the other. So, also, 

 with the female characters. 



Similarly with eye-colour. With at least rare ex- 

 ceptions, mulattos, no matter for how many genera- 

 tions interbred, have black eyes. The black continues 

 dominant. But if each succeeding generation mates 

 with white, then at length blue eyes suddenly appear. 

 How could this happen if the inheritance, not merely 

 the reproduction, were alternate? The black would 

 then remain as dominant in the octoroon as in the 

 mulatto. But with the inheritance of both characters 

 and the reproduction of only one the thing is com- 

 prehensible. Each reinforcement of blue then weakens 

 the dominance of the black until at last blue becomes 

 dominant. Much the same appears to happen in the 

 case of human monstrosities — idiocy, hare-lip, club- 

 foot, and the like. Here normality is dominant until 

 the conditions of nurture are in some way altered — 

 e.g. bv the reinforcement of the monstrous strain. 

 Apparently, then, the long latency of the blue eye- 

 colour is comparable with the similar latency of the 

 male characters in aphides. 



What is a mutation? As well as I arri able to judge 

 from literature, it is a character the reproduction of 

 which is Mendelian, and which, very commonly, is of 

 •wide amplitude," a •' snort." It has been said that, 

 "unlike fluctuations which are responses to nurture, 

 mutations have representatives in the germ-plasm." 

 But this is merely a misuse of language, for since all 

 characters are products of germinal potentiality and 

 fitting nurture, all must be equally germinal and 

 somatic. As already indicated in a former com- 

 munication (N.ATURE, August 25, pp. 808-10), 

 whenever we are able to observe natural selection 

 actually at work, its choice is obviously among 

 fluctuations, not mutations. Thus every shade of 

 susceptibilitv to tuberculosis may be observed; for 

 ifi the same surroundings some individuals resist 

 disease altogether, others recover more or less quickly 

 from illness, while others die after long or shc«-t 

 illnesses; in different locaHties the stringency of 

 selection is different ; races have been afflicted by 

 the bacillus during diverse durations of time, and 

 every race is resistant in proportion to the length 

 and severitv of its past experience. Moreover, the 

 inheritance ' is blended; thus half-breeds (e.g. Euro- 

 peans crossed with Asiatic, negro, and American 

 Indian) are susceptible in proportion to the blended 

 susceptibilities of their ancestors. For example, 

 Eurasians commonly survive in the cities of the tem- 

 perate zones, where American half-breeds almost in- 

 variablv perish. It is probable that few of the 

 American Indians and Maoris and their half-breeds 

 that came to and sur\'ived the Great War have 

 returned home to live. 



On the other hand, as Darwin noted, man "often 

 begins his selection by some half-monstrous form, 

 or at least bv somp modification prominent enough to 

 catch the eye, or be plainly useful to him." Man is 

 in haste to get results, and usually cannot observe 

 small differences in races other than his own. Thus 

 even the natives of China seem to an Englishpian 

 newlv arrived as like as peas in a pod — as doubtless 

 thev are. In some cases, however, man is obliged to 

 relv on fluctuations — as in breeding for speed in 

 horses, when a thousand co-ordinated structures are 

 involved. .\ thousand co-ordinated mutations occur- 

 ring in one individual are almost inconceivable. It 

 seems, then, that natural selection, which works for 

 the benefit of the individual and his race, chooses 

 fluctuations, while artificial selection, which works 

 for the benefit of man, relies largely on mutations. 



