510 



NA TURE 



[March aS, 1895 



LETTERS TO THE EDITOR. 



\_7kt Editor does not hold himself responsible for opinions ex- 

 pressed ty his correspondents. Neither can he undertake 

 to return, or to correspond with the writers oj, rejected 

 manuscripts intended for this or any other part of NaturK. 

 No notice is taken of anonymous communications.} 



The Statistical Investigation of Evolution. 



Mr. Thiselton-Dver states ihat Prof. Weldon his shown 

 that "selective destruction " takes place in early life amangst 

 individuals which deviate from the "mean specific form." lie 

 further says thai the actual statistical demonstration of the fact 

 that "minimum destruction is in position coincident with the 

 mean of the whole system," deserves to rank amongst the most 

 remarkable achievements in connection with the theory of 

 evolution. But, to judge from the paper by Prof. Weldon, 

 printed in Nature of March 7, he does not claim to have made 

 this remarkable achievement. He says that, according to the 

 results of the statistical investigation, in two dimensions of the 

 shore crab, the frequency of deviations increased during an 

 early period of growth, and that in one case the increase was 

 followed by a decrease ; in the other case it was not. Prof. 

 Weldon stales that if a certain law of growth can be shown 

 to be true by experimental tests, then the result implies 

 a selective destruction in the one case and not in the other. So 

 that all we have is the possibility in the future of a statistical 

 demonstration of selective destruction in the case of one par- 

 ticular dimensional character, and the rigid proof in the present 

 that in the case of the other dimensional character selective de- 

 struction does not take place. Surely every man of science 

 must admit that Prof. Weldon's results, on his own showing, 

 have done more against selective destruction than for it. 



Prof. Weldon says that if we know that a given deviation 

 from the mean is associated with a greater or less percentage of 

 death-rate, wc do not require to know how tiie increase or de- 

 crease of death-rate is brought about, and all ideas of functional 

 adaptation become unnecessary. This may be his own state of 

 mind on the subject, but I venture to state that it is not Dar- 

 winism, and that he cannot shut out others fro:n the most 

 interesting and most important fields of biology in this way. 

 Darwinism states that selective destruction is caused by the 

 struggle for existence, and that a selected character confers an 

 advantage in the competition to get food and beget offspring. 

 If a certain deviation is shown to be associated with an increased 

 or decreased chance of life, we want to know how it acts, and 

 no statistical Gallio can prevent us trying to find out. 



It does not require much search to find deviations which are 

 associated with an increased death-rate. In the human subject 

 cyanosis, due to the retention of the fecial communication be- 

 tween the two sides of the heart, is a well-known abnormality 

 or deviation in the infant ; but I believe few, if any, children 

 born in that condition reach the age of 20. Here we have no 

 difhculty in understanding the reason : the deviation necessarily 

 leads to death. But now, in comparison, lake the case of a 

 child born blind, or deaf and dumb. Here there is no intrinsic 

 reason why life should be shortened ; but in a severe com- 

 petition, it the individual dejiended entirely on his own exer- 

 tions, he might be, probably would be, starved or trampled to 

 death before he had lived very long. I think it is of some 

 interest and importance to know of any given character or 

 deviation, whether it is intrinsically harmful or beneficial, 

 extrinsically so {i.e. in the struggle for existence and repro- 

 duction), or quite indifferent. 



Prof. Weldon is silent, to some extent, about the cases which 

 tell against the idea of -elective destruction. He found that 

 deviations in Aurelia were as numerous in the adult as in the 

 Ephyra:. He told me in conversation, and did not say it was 

 in confidence, that he abandoned some experiments on the selec- 

 tion of Daphnia, because he found that the mere fact of keei>ing 

 a large number in the same water caused a progressive dis- 

 appearance of a certain conspicuous spine. His invcs.igations 

 also entirely ignore the diagnostic value of the characters he 

 deals with. It seems to me that a more valuable result woulil 

 be gained if a parallel investigation were made of two charac- 

 ters — one obviously di-ignostic, the other obviously adaptive. 

 Such char.-ictcrs could be found in a swimming crab. 



But above all, what wc want is a comparative investigation 

 of the results of (election without change of conditions, and of 

 change of conditions without selection. I began, not long ago, 



NO. 1326, VOL. 51] 



to try to inaugurate a society for carrying on a thorough inves- 

 tigation of this kind, but have not at present received enough 

 support to carry out the scheme. The method of the inves- 

 tigation is fairly obvious and not difficult, but the difficulty is to 

 get the money and the time to carry it out. I ciili'er from Prof. 

 Weldon in thinking that the questions raised by the Darwinian 

 hypothesis are not purely statistical, but experimental, and I 

 agree with Mr. Thiselton-Dyer— that to talk of experimentally 

 checking the hypothesis by the statistical method is a con- 

 tradiction in teims. J. T. Cunningham. 

 Cleethorpes, March 15. 



A True Spectrum Top and a Complementary One. 



To make a true specirum top — whicli is not copyright, so far 

 as I know — take a disc of white paper, and one of black, of 

 equal size. Spin the white one on a disc of cardboard mounted 

 on a nail, an 1 while it Is spinning draw a small brusli well 

 charged with lamp-black water-colour paint, steadily and not too 

 slowly from centre to circumference of the disc, thus describing 

 a spiral line. Make a radial cut in each of the discs, and after 

 interlocking them as in the well-known colour discs, place them 

 on the top. We thus obtain a top in which the lines are spiral, 

 and the relative sizes of black and of white areas are easily 

 regula'cd by turning one disc to right or left, while the other 

 is held still. If the lines be not too thin or too thick, and not 

 loo near togeth r, and if the relative areas of black and white 

 be adjusted suitably to the light, the top exhibits, when spun, 

 broad bands of colour, each band containing all the colours of 

 the spectrum in their natur.1l order. The spaces between the 

 lines should be not less than five times as broad as the lines 

 themselves. The brightest effects are produced in my own case, 

 by lamplight, with the areas of light and dark almost equal ; by 

 daylight, with the dark area about three times as great as the 

 light. Other proportions, however, seem to give better results 

 with other people. 



A " complementary lop," yielding colours complementary to 

 those of the spectrum {i.e. the colours of mother-of-pearl) in a 

 continuous band ranging from lemon-yellow, through puce to 

 electric-blue green, is made in the same way, except that the 

 .spiral line is to be drawn in wliite on the black disc. 



In both cases the colours are somewhat dilute, but the proper 

 regulation of the relative areas of black and white reduces this 

 defect very considerably, and I have obtained bands on my 

 spectrum top brighter and purer than any which I could get by 

 painting a specirum with colours on paper. 



I communicate this description before my experiments are 

 complete, in order to prevent anyone who may make the same 

 discovery, from obtaining a copyright for the design of either 

 these tops or of earlier ones which I made, in which one half 

 the disc was black and the other white, with a white spiral on 

 the black, or a black spiral on the white, or with both at dilTerent 

 dist.ances from the centre on the same top. .Vnyone who wishes 

 to do so may make as many tops or lantern-discs as he chooses 

 from the above description, jirovided he does not attempt to 

 hinder anybody else from making or selling similar ones. 



C. Hekuert Hurst. 



Owens College, Manchester, March 24. 



A Foucault Pendulum at Dublin. 



It may perhaps interest some of your readers to loam that 

 Koucault's pendulum experiment has recently been performed 

 in Trinity College, Dublin, with complete success. 



Immediately under one of the glass domes, by means of which 

 the hall of the New Building is lit, a cast-iron bar was securely 

 bolted, which terminatis in a cylindrlcallyshaped piece of metal 

 the axis of which is vertical. Into this cylinder a steel plug 

 was inserted, which was drilled to receive the upper end of 

 the wire supporting the Imb, which was fitted with a screw, 

 liy placing the upper end of the wire in this position, I'rnf. 

 FilzGcraldand I secured a kngth of 45 feet for our pendulum ; 

 but, under the circumstances, we were unable to use the same 

 weight as that adopted by Sir R. Hall when making the 

 experiment, viz. 300 lbs., and were obliged to content ourselves 

 with a liub weighing 16 lbs., which, however, answered 

 admirably. 



The experiment is made in the following manner :— About 

 two feet behind llic position of equilibrium of the bob, wc place 

 the electric lamp, and at a suitable distance in front a lens, so 



