NOVEMUER 2 I, I 907 J 



NA JURE 



55 



" now views," nor limi.ed myself to matters which came 

 within the range of the " new science." The criterion of 

 merit struclv me as highly remarl<able and certainly very 

 new, and indicative of a degree of toleration which, if not 

 altogether new. Is at least unusual amongst men of 

 science. The passage quoted by Mr. Cunningham demon- 

 strates that I expressed myself very badly. Soon, however, 

 I shall have an opportunity of trying to do better, and 

 I suggest that until iln.'ii .Mr. Cunningham shall suspend 

 judgment. 



In previous communications to N..\ture I have admitted 

 that Mendelism may conceivably shed a light on the func- 

 tion of se.\, but I challenged its exponents to mention a 

 single other problem on which it has the remotest bearing. 

 \o one has as yet mentioned another problem, and Mr. 

 Cunningham denies, apparently, that it has a bearing on 

 that of sex. Must we assume, then, that it sheds no light 

 on anything at all? 



If, instead of dwelling on the dangers incurred by those 

 who venture to differ from Darwin, Mr. Lock 

 (November 14) demonstrated my errors, he would be more 

 convincing ; and, since he is probably the only human 

 being who doubts the blending of the black and white 

 races in Mulattoes and their descendants, it would be well 

 if he, rather than I, undertook the collection of pedigrees. 

 He would feel himself on the track of a great discovery 

 which would enlighten even Mulattoes, whereas I should 

 feel 1 was wasting time. I do not know what I can gain 

 from the renewed study which he is good enough to 

 suggest. I am well aware of the three principal Mendelian 

 doctrines — segregation of units, independent inheritance of 

 characters, Mendel's law is the greatest of biological dis- 

 coveries — and the more I see and read the more thoroughly 

 I am convinced that they are all demonstrably erroneous, 

 and that nothing but the restricted area covered by 

 Mendelian studies has prevented a recognition of that fact 

 by Mendelians themselves. 



.According to selectionists blending is the function, or at 

 least the effect of conjugation. .According to Mendelians, 

 not blending, but segregation, occurs. Taken by itself, 

 this doctrine assigns no function to conjugation ; it merely 

 controverts the theory of blending. Taken in conjunction 

 with the Mendelian hypothesis of the independent inherit- 

 ance of characters, it assigns to conjugation the function 

 of effecting an exchange of units between the paternal and 

 maternal sets of allelomorphs. That much the Mendelian 

 doctrines imply — that much and no more. Clearly, then, 

 Mendelism is concerned solely with the function of sex. 

 At any rate, I can conceive of nothing else, and, judging 

 from their spoken and written communications, Mendelians 

 have been so rapt in contemplation of the grandeur of 

 ihe discovery that they have given no further thought to 

 the matter. 



" But everybody said," ciuoth he, 

 "That 'twas a famous victory." 



Mr. Lock declares that, since conjugation is nearly 

 universal, all, or nearly all, the questions of heredity are 

 problems of the function of sex. He might as reasonably 

 ru'gue that, since assimilation and death are universal, all 

 problems of heredity are problems of assimilation and 

 rlf-ath. The looseness of thought which pervades 

 Mendelism is well illustrated by his remarks on partheno- 

 genesis and bud-variation. Mendelian segregation implies 

 ihe separation of allelomorphs which, through the union 

 of two gametes, have previously met in a zygote. There 

 ran be no meeting, and therefore no separation when re- 

 production is parthenogenetic. Mr. Lock, in fact, 

 enunciates the surprising doctrine that all variation is 

 Mendelian segregation. I hope he will forgive the blunt- 

 ness of the expression, but he trifles with established terms. 



Souths°a, November 17. G. .Archdall Reid. 



The Winding of Rivers in Plains. 

 Sir Oliver I.odce's letter in N.\ture of November 7 is 

 itself an illustration of his comment on the way in which 

 misunderstanding of cause may lead to misrepresentation 

 of fact. The statements quoted by him are, as is often 

 the case in text-books, inaccurate in so far as they, are 

 incomplete, but, nevertheless, in closer accordance with the 



NO. 1986, VOL. 77] 



facts of nature, viewed from the gcxjlogical point of view, 

 than he supposes. 



We are not dependent on the calculations of mathe- 

 maticians for our knowledge' of the behaviour of rivers ; 

 at one time I had a goodly collection of notes of observ- 

 ations made and published by others, but having un- 

 fortunately lost this, I shall write only of what I have 

 seen myself, having had many opportunities of observing 

 the behaviour of rivers which are subject to large 

 variations of flow. In the dry season, when the water 

 is low and the stream flows quietly in a channel of 

 more than sufficient size to carry the discharge, I have 

 noticed that the current is often perceptibly more rapid 

 over the shallow water on the inner side of the bend 

 than in the deep channel on the outer. The line of 

 maximum velocity of flow would, consequently, lie nearer 

 the inner than the outer side of the bend, and be less 

 sinuous than the general course of the river, in accord- 

 ance with the investigation quoted by Sir Oliver Lodge ; 

 but at these times the geological activity of the stream 

 is so small as to be almost negligible. When, on the 

 other hand, the same stream is in flood, and a day's work 

 is more than that of a decade or a century of the placid 

 flow, another state of things prevails ; then the current 

 sweeps rapidly round the outer side of the curve, and 

 on the inner side is comparatively slack water. The line 

 of maximum velocity must, therefore, become more 

 sinuous than the river itself, and not only is this result 

 to be reached by deduction, but in some cases it has been 

 visibly perceptible. Further, the undertow on the outer 

 edge of the curve has not been visibly distinguishable in 

 anv case that I have seen ; if existing, its effect must 

 have been insignificant in comparison with that of the 

 horizontal movement of the stream, and often it certainly 

 does not exist. At times, and in certain circumstances, 

 sand and even pebbles may be thrown up to the surface 

 of the water near the outer bank of the stream, and 

 where the waters have overflowed the banks pebbles may 

 be found Iving on the dry ground after the flood has passed 

 away; these facts show that there must, in some cases' at 

 least, be an upward, not a downward, current along the 

 bed on the outer side of the bend of a stream. 



Do not let me be misunderstood. So far as Sir Oliver 

 I^odge is pleading for accuracy in text-books we are all 

 with him, but when he states, as a positive fact, that the 

 line of maximum velocity of current in a river is less 

 sinuous than the river itself, and that the current along 

 the outer bank of a curve is more rapid than along the 

 inner, then I must join issue with him and maintain that 

 these statements are only sometimes true. This is no 

 question of the accuracy of Prof. James Thomson's 

 calculations, but they evidently cover only that part of 

 the problem which is least important from a geological 

 point of view. R. D. Oldham. 



"Magic Mirror" Effects. 



I HAD occasion recently to coat with collodion a silver 

 surface mirror on patent plate 2 millimetres thick. During 

 the operation the mirror was held with one of the rubber 

 pneumatic holders frequently used by photographers when 

 coating or varnishing plates. As the film of collodion set, 

 a series of interference colours disposed in concentric 

 circles appeared immediatelv over the region of the suction 

 disc of the holder. I could scarcely bring myself to 

 encourage the idea which at once occurred to me, viz. 

 that the slight suction of the pneumatic holder was actually 

 deforming so thick a plate of glass and producing an 

 appreciable concavity in its vicinity. 



But this seems really to have been the case. For when 

 the beam of light from a lantern (placed with its back to 

 the screen) was reflected back on to the screen by the 

 mirror held w-ith the pneumatic holder, there appeared in 

 the rectangular patch of light determined by the size and 

 shape of the mirror a much brighter internal circular 

 patch which changed its position conformably with any 

 alteration of the position of the pneumatic holder. 



I have never seen any reference made to deformations 

 produced in this way ; yet such deformations might be 

 found to have a practical significance in critical coating 

 operations where absolute uniformity in the thickness of 

 the coating is desired. Douglas Carnegie. 



