December 15, 1910] 



NATURE 



203 



LETTERS TO THE EDITOR. 



[The Editor does not hold himself responsible for opinions 

 expressed 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 o/ Nature. 

 .\o notice is taken of anonymous communications.] 



Morphological Method and the Ancestry of Vettebrates. 



There has just reached me the Proceedings of the 

 Linnean Society for October, containing the report of an 

 interesting discussion upon the " Origin of Vertebrates " | 

 which took place at meetings of the society in January 

 and February of this year. Apart altogether from state- 

 ments as to matters of fact which seem to be open to 

 challenge, this discussion appears to me to raise points j 

 in regard to the methods employed by the morphologisl | 

 which are of practical importance to those interested in j 

 the science of morphology. j 



The remarks made by Dr. Gaskell and his supporters | 

 make it apparent that there exist wide differences between i 

 what they accept as the correct principles of morpho- | 

 logical research and those which are accepted by other 

 working morphologists. Personally, I have been de- 

 voting myself for some time past to researches dealing 

 with the evolution of vertebrate structure, and my 

 impression is that there can be no wider gulf than that 

 • esasting between the working principles apparently adopted 

 by Dr. Gaskell and those adopted by myself and many i 

 other morphologists. If the principles of morphological i 

 speculation employed by Dr. Gaskell are sound, it seems 

 to follow that those heid by the majority of morphologists 

 are absurd, and that the work based upon them represents 

 in great part wasted labour. I will endeavour to throw 

 Into relief some of these radical differences of opinion 

 as to general principles which appear to separate many 

 of us from Dr. Gaskell and his friends. I feel all the 

 more impelled to do so when I read the words of a dis- 

 tinguished physiologist who took part in the discussion : — 

 " I am convinced that the principles on which he [Dr. 

 Gaskell] has proceeded are the only ones which will lead 

 to a solution of the problem." I fully realise that there 

 are physiologists who would feel it rash to express them- 

 selves so decidedly as to what are, or what are not, the 

 correct principles upon which to work in a science other 

 than their own. Nevertheless, it seems quite clear that 

 Dr. Gaskell's views — though they may not appeal to 

 many who are specialists in vertebrate morphology — do 

 produce a strong impression on many w^orkers in other 

 departments of biological science. 



What are the principles which must be followed in 

 speculations regarding phylogeny if these speculations are 

 to be trustworthy? It is clear, in the first place, that 

 whenever possible such speculations should rest upon a 

 tripod basis of comparative anatomy, embryology, and 

 palaeontology. It is clear, further, that in fashioning 

 each foot of the tripod certain definite rules must be 

 j adhered to if the foot is to be sound and sufficient to 

 I support the weight which is to rest upon it. For example. 

 In making use of the data of palaeontology, we have to 

 bear in mind, firstly, that the geological record, and still 

 I more our knowledge of that record, is and must always 

 I be of the most fragmentary character, and, secondly, that 

 "1 all phylogenetic speculation it is unsafe to trust to 

 ita dealing only with one single set of organs, whether 

 rse be the skeletal organs — alone, as a rule, available 

 the palaeontologist — or any other system of organs. In 

 ii case of comparative anatomy and embryology we must 

 j bear in mind that a feature has a prima facie greater 

 importance or not according as to whether or not it occurs 

 j in a more or less " primitive " group. Then, again, in 

 both comparative anatomy and embryology it is necessary 

 I to devote great care and attention to the sifting out of 

 features which are mere adaptations to modern environ- 

 mental conditions from those which are ancestral. Finally, 

 has to be borne in mind that in comparative anatomy 

 are beset by the same kind of difficulty as that of the 

 jjiotozoologist when he tries to piece together isolated 

 ] observations on dead material into a connected life- 

 history, while in embryology, on the other hand, the facts, 

 ;ch as they are, appear to be presented to us all ready 



NO. 2146, VOL. 85] 



arranged in their phylogenetic sequence. These general 

 principles apply to all phylogenetic speculations, and 

 morphologists are probably all in agreement regarding 

 them. 



Now as regards the phylogeny of the Vertebrata. 

 When a student under Sedgwick at Cambridge I began to 

 realise that perhaps the most clamant need in vertebrate 

 morphology was for a broadening of its observational 

 basis, particularly in regard to the embryology of the 

 more primitive forms of vertebrates. There seemed 

 general agreement that the elasmobranchs, crossoptery- 

 gians, lung-fishes, and urodele amphibians were all 

 " primitive " groups as compared, e.g., with teleostean 

 fishes, reptiles, birds, and mammals. It seemed clearly 

 indicated that if we were to have the general ideas of 

 vertebrate embryology upon a trustworthy basis it was 

 essential that the development of the crossopterygian 

 ganoids and lung-fishes should be worked out to a similar 

 degree of detail to what had been accompxlished in the 

 case of the other groups. My own research work during 

 the last fourteen years has been directed according to 

 this ideal, and has been devoted to the study of the 

 morpholt^y of the four relatively primitive groups men- 

 tioned above, together with, of course, the cyclostomes. 

 As I fail to see how there can be any other scientific 

 method of approaching the problem of vertebrate phylogeny 

 than that which starts from a comprehensive study of 

 such surviving forms as are admitted to be the more 

 pi imitive, I may be excused for venturing to offer these 

 remarks regarding the discussion at the Linnean Society. 



One of the points rais'Kl in the discussion is the very 

 important one, upon which I have elsewhere expressed 

 opinions — the relations of physiology and morphology. 

 No one has, I dare say, a greater horror than I have of 

 that type of zoologist sometimes referred to as the 

 " mere " morphologist — the zoologist who fails to keep 

 before him at every moment of his work that he is deal- 

 ing with living functional organisms, who fails to realise 

 that at every stage of its evolution an organ must, in the 

 first place, be able to function. I have at various times 

 criticised evolutionary speculations, e.g. on the evolution 

 of renal organs, or of the vertebrate limbs, or on the 

 inadequacy of natural selection, because they seemed to 

 ignore important physiological or functional considerations. 

 While taking the greatest care to keep physiological con- 

 siderations before us, we must, however, not forget that 

 the facts upon which the evolution theory is based are 

 morphological facts. We know nothing as yet of " re- 

 capitulation " in physiology ; we have in the rocks no 

 record of physiological change ; even the science of com- 

 parative physiology is still in its infancy. The whole 

 record of evolution is and must necessarily be mainly, if 

 not entirely, a morphological record. 



Another principle which is of much importance in 

 morphological work is this : the greatest care must be 

 taken to make the observational basis of speculation as 

 broad as possible. The fewer the organs or the organisms 

 that are made use of the less trustworthy are the con- 

 clusions drawn. E.g. a detailed study of the radiate eye 

 of an insect and of a stalk-eyed crustacean brings to light 

 the most striking resemblances, so much so that if we 

 only knew one genus of each group and only the eye 

 structure in that genus we should be inclined to suppose 

 the two genera to be closely allied. I need not say that 

 a study of the general morphology of the Antennata on 

 the one hand and the Crustacea on the other indicates 

 that they have been evolved separately from a simple 

 ancestral form which existed probably long before the 

 radiate eye had become evolved. The same kind of prin- 

 ciple holds in regard to organisms. The detailed study 

 of the structure of two animals may bring to light the 

 most astounding resemblances in details of structure, and 

 it may, nevertheless, be the height of rashness to attribute 

 the resemblance to genetic affinity unless there exists 

 collateral evidence which supports the view. Yet Dr. 

 Gaskell says, " my object throughout has been by the 

 study of Ammocoetes to find out a clue to the past history 

 of these extraordinary early forms of fish." Is the 

 ordinary zoologist" like myself completely in error when 

 he thinks that for " .AmmoccEtes " there should have been 

 written the words " C3clostomes, crossopterygians, elasmo- 

 branchs, lung-fishes, and amphibians " in formulating 



