88 SECTIONAL ADDRESSES. 



phylogeny is as a subject of study, it is not necessarily tlie most smtable 

 basis of classification.' I am not sure that I quite understand what is 

 implied by the second of these statements, but I do not suppose that even 

 Dr. Bather would be prepared to suggest a system of classification entirely 

 divorced from phylogenetic considerations. 



Forty years ago the reconstruction of the evolutionary history of the 

 major divisions of the animal kingdom was almost universally regarded as 

 the chief end of zoological research. To-day, except among palaeontolo- 

 gists, one might almost say that the phylogenetic period in the history of 

 zoology has come to an end. When one recalls the extravagances of its 

 later developments, the derivation of Vertebrates from Arachnids and of 

 Echinoderms from Cirripedes, one cannot be surprised that zoologists of 

 the modern school take little interest in it. If we accept this attitude, it 

 follows that problems of affinity and relationship are not worth worrying 

 about. We are told, in so many words, that our business as systematists 

 is identification, not classification ; that what we have to do is merely to 

 devise some kind of key or card-index that will enable animals to be quickly 

 and easily sorted into species. As far as the really scientific branches of 

 zoology are concerned an artificial system of classification is as good as, 

 and may even be better than, any other. An illustration of this attitude of 

 mind is seen in a paper recently issued from Cambridge in which Liihodes 

 is replaced, without explanation or discussion, among the Brachyura ; 

 which, on the card-index system, is doubtless its appropriate place. 



It is quite true that the categories of the physiologist, the ecologist, the 

 geneticist, and so on, often cut across the dividing lines of the most 

 natural classification we can devise, but both the divergences and the 

 coincidences are worthy of closer consideration than they sometimes 

 receive. If there is any truth in the theory of Evolution it is obvious that 

 functions and habits have an evolutionary history behind them, but it is no 

 less obvious that this history has not been independent of the history of the 

 organisms that display them. The details of this history we shall never 

 fully know and even its broad outlines may perhaps always remain misty. 

 A Natural system of classification expressing even these broad outlines 

 may prove to be an unattainable ideal, but each step towards it holds out 

 the promise of usefulness in other and possibly remote fields of research. 



A great deal of current work and still more of current speculation in 

 zoology seems to me to suffer from this neglect of the taxonomic outlook. 

 In the zoology of the later nineteenth century the comparative method was 

 still the chief tool of morphology. The relative importance of structural 

 characters was measured by the extent of their persistence through larger 

 or smaller divisions of the animal kingdom. This point of view tends to 

 be lost sight of with the increasing emphasis on the experimental method. 

 The systematic zoologist, in listening to the exponents of the modern lines 

 of research is apt to be impressed by the little account that is taken of the 

 vast variety of animal life. To say this is not to underrate in any way the 

 advances that have been made in these lines within the present century or 

 the revolutionary changes they have made in our views on many funda- 

 mental questions. Physiology, for example, is to-day a vastly different 

 science from what it was thirty years ago, partly because the physiological 

 laboratory has a more varied fauna than it had then. Nevertheless, the 



