458 



SCIENCE 



[N. S. Vol. KXXVIII. No. 979 



have arrived at an accomplished fact. 

 These things, in spite of all their variety in 

 structure and function, being homogenes, 

 tell us nothing, because according to our 

 mode of procedure we can not compare 

 that monophyletie Anlage with anything 

 else, since we have reduced all the homo- 

 genous modifications to one. Logically it 

 is true that there can have been only one, 

 but in the living world of nature there are 

 no such ironbound categories and absolute 

 distinctions. For instance, if we compare 

 the organs of one and the same individual, 

 we at once observe repetition, e. g., that of 

 serial homology, which implies many diffi- 

 culties, with very different interpretations. 

 Even in such an apparently simple case as 

 the relation between shoulder girdle and 

 pelvis we are at a loss, since the decision 

 depends upon our view as to the origin of 

 the paired limbs, whether both are modified 

 visceral arches, and in this case serially re- 

 peated homogenes, or whether they are the 

 derivatives from one lateral fin, which is 

 itself a serial compound, from which, how- 

 ever, the proximal elements, the girdles, are 

 supposed to have arisen independently. 

 "What is metamerism? Is it the outcome 

 of a process of successive repetitions so 

 that the units are homogenes, or did the 

 division take place at one time all along 

 the line, or is it due to a combination of 

 the two procedures? 



The same vagueness finds its parallel 

 when dealing with the corresponding or- 

 gans of different animals, since these afford 

 the absolute chance that organs of the same 

 structure and function may not be re- 

 ducible to one germ, but may be shown to 

 have arisen independently in time as well as 

 with reference to the space they occupy in 

 their owners. As heterogenes they can be 

 compared as to their causes. In the study 

 of the evolution of homogenes the problem 

 is to account for their divergencies, whilst 



the likeness, the agreements, so to speak 

 their greatest common measure, is eo ipso 

 taken to be due to inheritance. When, on 

 the contrary, dealing with heterogenes we 

 are attracted by their resemblances, which 

 since they can not be due to inheritance 

 must have a common cause outside them- 

 selves. Now, since a leading feature of the 

 evolution of homogenes is divergence, 

 whilst that of heterogenes implies converg- 

 ence from different starting-points, it fol- 

 lows that the more distant are these re- 

 spective starting-points (either in time or 

 in the material) the better is our chance of 

 extracting the greatest common measure 

 out of the unknown number of causes 

 which combine in the production of even 

 the apparently simplest organ. 



These resemblances are a very promising 

 field and the balance of importance will 

 more and more incline towards the investi- 

 gation of function, a study which, however, 

 does not mean mere physiology with its 

 present-day aims in the now tacitly ac- 

 cepted sense, but that broad study of life 

 and death which is to yield the answer to 

 the question Why? 



Meantime, comparative anatomy will not 

 be shelved; it will always retain the cast- 

 ing-vote as to the degree of affinity among 

 resemblances, but emphatically its whole 

 work is not to be restricted to this occupa- 

 tion. It will increasingly have to reckon 

 with the functions, indeed never without 

 them. The animal refuses to yield its 

 secrets unless it be considered as a living 

 individual. It is true that Gegenbaur 

 himself was most emphatic in asserting 

 that an organ is the result of its function. 

 Often he held up to scorn the embryog- 

 rapher's method of muddling cause and 

 effect, or he mercilessly showed that in the 

 reconstruction of the evolution of an organ 

 certain features can not have been phases 

 unless they imply physiological continuity. 



