Sei'tember 3, 1896] 



.VA TURE 



413 



LETTERS TO THE EDITOR. 

 ( The Eiiilor 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.^ 



Utility of Specific Characters. 



Proi\ Lankester's lucid statement in Nature for August zo, 

 shows that a part of his objection to my position is due to my 

 own wain of skill in stating clearly what I mean. 



I am far from wishing to reject the method of imaginative 

 hypothesis and subsequent experiment or observation. I respect 

 that method as sincerely as Prof. Lankester himself, and although 

 I cannot pretend to his measure of skill in using it, yet, so far as 

 I can sei', I have, in my work on the frontal breadth of crabs, 

 c-mployed this very method to the best of my ability. The 

 hypothesis with which I started was, that if natural selection 

 acted upon the frontal breadth of crabs at all, there ought to be 

 a demonstrable difference between the percentage of abnormal 

 frontal Ijrcadth in young crabs, and the percentage of the same 

 abnormalities in older crabs ; and I proceeded to test this 

 hypothesis by measurement of crabs of different sizes. The 

 result showed that a change in the frequency of abnormal 

 frontal breadth could, in fact, be observed. The effort of imag- 

 ination was here small enough, but, such as it was, it served to 

 guide my first step ; and, having made this first step, I had to 

 formulate a second hypothesis. A diminution in the frequency 

 of abnormal frontal breadth, with increasing size of crabs, might 

 be due either to a selective destruction of abnormal crjibs during 

 growth, or to a modification of these crabs, by which abnormal 

 individuals lose their abnormality as they grow. In order to 

 decide which of these imaginative hypotheses should be adopted, 

 I have spent a great part of the last two years in ascertaining the 

 law of growth of crabs, so far as their frontal breadth is con- 

 cerned. Setting the question of skill on one side, the only 

 difference I can perceive between the method of this whole 

 investigation and that of any research conducted by Prof. 

 Lankester, is a difference in the tools employed in verification of 

 hypotheses. The only tool which I have used has been some 

 kind of measuring scale ; and, although this kind of tool is more 

 unpleasant to work with than those used Ijy more fortunate 

 persons, it docs not imply any dift'erence in the method of work. 



Further, assuming the law of growth to yield evidence of 

 selective destruction, so that change in frontal breadth is cor- 

 related with change in death-rate, I heartily agree with Prof. 

 Lankester that a further hypothesis ought to be formulated as to 

 the whole process connecting change in frontal breadth (and the 

 whole group of characters correlated with it) with change in 

 death-rate. The only step taken by Prof. Lankester, which I 

 cannot follow, is the admission of hypotheses in which some of 

 thel.artorsof the problem are neglected. I should like to explain 

 what I mean by this. 



In Citrcinus miTnas I have shown that change of frontal 

 breadili is correlated with change in several other dimensions of 

 the exoskeleton ; and I have no doubt that it is correlated also 

 with change in the size and shape of several internal organs, .such 

 as the brain, liver, kidneys, and others. I have not measured 

 such an oxyrhynchous crab as Stenorhynchus ; but it is probable 

 that the changes among internal organs correlated with change 

 in frontal breadth, will prove to be very difi'erenl in such a crab 

 as S/iiiorhynchus from the corresponding changes in Carciniis. 



Let us su|>])ose, therefore, that the liver is shown to vary when 

 the frontal breadth of Carciiius varies, but not when the frontal 

 breadth of Stenorhynchus varies ; and suppose, further, that an 

 hypothesis is submitted as to the process by which change in 

 the liver oi Carcinus leads to change in the death-rate. It seems 

 to me that, unless one of the steps in this process involves a 

 change in frontal breadth, the hypothesis must be rejected, be- 

 cause one of the properties of the liver of Carcinus is not 

 accounted for. The hypothesis submitted may be true of 

 Stenorhynchus ; but, since it neglects one of the differences 

 between that animal and Carcinus, it cannot be true of both. 



To put the matter in another form : suppose I wish to obtain 

 hydrogen from sulphuric acid, I can do so by adding to the sul- 

 phuric acid a certain quantity of zinc. From a known quantity 

 of sulphuric acid I can obtain a definite quantity of hydrogen, 

 and I shall, in so doing, dissolve a definite quantity of zinc with 

 the formation of a definite quantity of zinc sulphate. If, instead 



NO. I 40 1, VOL. 54] 



of dissolving zinc, I dissolve iron in my .sulphuric acid, I can 

 still obtain from it the same quantity of hydrogen, but the 

 quantity of iron required will not be the same as the quantity 

 of zinc used in the previous experiment, and the resulting sul- 

 phate will be different. It is, of course, impossible to form an 

 exact hypothesis of what occurs in either of these cases, if I pay 

 attention only to the evolution of hydrogen, and regard the 

 formation of sulphate as an unimportant concomitant. I must 

 in each case form a theory of the behaviour of the metal, the 

 hydrogen, and the acid radicle ; and, so far as it fails to account 

 for any fact concerning any one of these bodies, my theory is 

 imperfect. 



In precisely the same way, it seems to me that we ought not 

 to rest content with any theory of an animal structure which 

 does not account for all the phenomena associated with it ; so 

 that a theory of the function of frontal breadth in a crab should, 

 I think, involve every organ correlated with it. It may be said 

 that such a theory is unattainable because of its complexity ; 

 and this is certainly at present true ; but the habit of regarding 

 one or other of the properties of an organ as unimportant, 

 would for ever prevent the formation of such a theory even if 

 it were otherwise passible. 



It is this sense of the necessary complication of such hypotheses 

 which makes me glad to assert that they are unnecessary to a 

 knowledge of the factors of evolution. It is possible to know 

 that change in frontal breadth in a Carcinus, for example, is 

 associated with change in death-rate under the conditions of 

 Plymouth Sound ; so that those crabs in which the frontal 

 breadth has a particular magnitude, can be known to have a 

 greater chance of living and breeding than those of different 

 frontal breadth. A complete knowledge of the processes asso- 

 ciated with this relation between frontal breadth and death-rate 

 is a thing of very great interest, and I believe, as firmly as Prof. 

 Lankester, that every effort should be made to attain to it ; but, 

 desirable as it is, it is still not necessary in order to know that a 

 crab's chance of living and breeding may be known by measuring 

 its frontal breadth. It is not necessary in order that the change 

 in mean frontal breadth may be measured from generation to 

 generation, and the direction and rate of evolution by this means 

 ascertained. W. F. R. Wei.don. 



Marine Biological Laboratory, Plymouth, August 26. 



The Death of Lilienthal. 



I have received this authentic report of Mr. O. Lilienthal's 

 death. If you think the letter worth publishing in^NATURE, it 

 is at your service. 



Hannover, Technische Hochschule. 



C. RUNGE. 



Vou are right in presuming that I can give you details refer- 

 ring to Otto Lilienthal's death, authentic as far as they can be 

 obtained. 



As early as the beginning of la.st spring, Lilienthal's experi- 

 ments had taken a new departure. He had gradually come 

 to the conclusion that the surfaces employed by him were not 

 sufficient. 



With a surface of twelve to fourteen square metres he could 

 take sufficiently long flights to serve his purpose of observation 

 and practice in strong, gusty wind, but he very rightly considered 

 experimenting in a strong wind to be too dangerous, and with a 

 light breeze about twenty square metres were found necessary. 

 This enormous surface, however, could not be handled with the 

 same certainty and exactness as the older wings, and as his 

 system of steering consisted in shifting his weight within the 

 surface upon which it was suspended, he had hit upon the simple 

 expedient of placing two surfaces one above the other. 



This system promised from the beginning to be a very marked 

 .advance. Informer days Lilienthal had tried, over and over 

 again, to make small paper models that would soar like birds, 

 and had always been disappointed. Now this problem seemed 

 to be solved. These two-story models, which resembled beetles 

 rather than birds, soared in the most astonishing manner. He 

 would let them oft" from the top of the artificial cone which he 

 had erected at Lichterfelde, and they would take long and some- 

 times circuitous flights into the surrounding fields, and never 

 .showed the slightest tendency to take " headers"— a peculiarity 

 very frequently hitherto observed in soaring models. 



These experiments, therefore, seemed to prove that not only 

 would a two-story surface be more easily steered, because a 



