752 



NA TURE 



[December 2, 1922 



spoke of " The inadequacy of the theory of natural 

 selection as an explanation of the facts of geographical 

 distribution and evolution." Dr. Willis pointed out 

 that Darwin's immortal service to science consisted 

 in the firm establishment of the doctrine of evolution. 

 Tins was effected by devising the mechanism of the 

 natural selection of infinitesimal variations, the 

 principle usually known under the name of Darwin- 

 ism. This theory involves many assumptions : 

 ig others, that such variations are (i) continuous, 

 (2) hereditary, (3) differentiating, (4) selected, and 

 (5) that the necessary differentiating variations for 

 the associated characters appear together. For all 

 of these the proof is as yet insufficient. 



I >r. Willis proceeded to consider the extent to 

 which natural selection of small variations could be 

 held to explain the facts of geographical distribution, 

 morphology, and evolution, special reference being 

 made to the grasses and to the Chrysomelid beetles. 

 It was then pointed out that natural selection was 

 helpless to explain the differences in distribution of 

 closely related species, which, on the other hand, could 

 be explained on the hypothesis of " Age and Area "— 

 i.e. that the area occupied by any group of allied 

 spei ies (at least ten) depends chiefly upon the ages 

 of the species. On this hypothesis predictions could 

 be made which were found to be justified by facts. 

 Dispersal of species is held to be mainly mechanical : 

 so much dispersal in so much time. This suggested 

 the further hypothesis of " Size and Space " ; that, 

 in groups of ten allied genera, the total space occupied 

 goes with the total number of species. If this be 

 true, whatever phenomena are shown by " Area " 

 should also be shown by " Size." This in fact is 

 shown to be the case when the number of allied 

 species occupying areas of increasing size and the 

 number of species in allied genera are plotted in the 

 form of curves. The shape of the curves is invariably 

 uniform. But sizes of genera are clearly the result 

 of evolution. According to the theory of natural 

 selection, the sizes of genera must depend upon their 

 success, and it is, therefore, inconceivable that they 

 should show such uniformity of expression. Such 

 facts, however, are easily explained by the hypothesis 

 that geographical distribution and evolution extend 

 with age — i.e. that the factors causing them act at a 

 more or less uniform rate. Natural selection, which 

 is essentially differentiating, cannot explain these 

 facts. 



In consequence, however useful it may be to 

 explain details of certain adaptations, and although 

 everything at birth must pass through the sieve of 

 natural selection, it seems that the latter principle 

 must be abandoned as an important factor in geo- 

 graphical distribution and evolution. Finally, Dr. 

 Willis considered it necessary to accept large muta- 

 tions as being of greatest importance in evolution. 

 In his opinion Guppy's theory of differentiation 

 should replace the Darwinian position that evolution 

 has proceeded from individual through variety to 

 species, genus, etc., for the theorv of " Age and Area " 

 showed clearly that the family is older than the 

 genus, and that the genus is older than the species. 



Mr. G. Udny Yule spoke upon " A mathematical 

 conception of evolution based on the theory of Age, 

 Size, and Space." He suggested that if the size of 

 (In genus be considered an index of its age, species 

 might be regarded as thrown by the genus much 

 offspring are thrown by a stock, and that the 

 number of species originating from a given initial 

 ies will • increase in geometric ratio with the 

 time. The forms of frequency distribution for 

 numbers of genera with numbers of species were 

 shown to be in accordance with the facts, and the 

 possibility was suggested of determining from such 



distributions the ratio between the rates of increase 

 of genera and species and the age of the family in 

 terms of the doubling period for species. 



Mr. C. Tate Regan stated that in his special study 

 of fishes he had formed conclusions as to the origin 

 and relationships of species and genera which were 

 quite different from those of Dr. Willis. He pointed 

 out that the hollow curves of the previous speakers 

 were extreme types of asymmetrical curves which 

 could also be obtained from many sorts of data — e.g. 

 I\ plotting graphically the wealth of the community, 

 grading from many poor to few very rich, or from 

 the numbers of occurrences of surnames in the 

 London Telephone Directory. All these curves were 

 simply graphic representations of certain facts the 

 meaning of which could be ascertained only by 

 detailed analysis. According to his own view, the 

 first step in the origin of a species had been not a 

 change of structure but some form of isolation. 

 The extreme mutationists, who thought that adapta- 

 tions originated as large transformations without 

 relation to use or environment, seemed to have returned 

 to the special creation theory. Darwin's theory of 

 evolution was that species had been modified by the 

 natural selection of slight variations, aided by the 

 inherited effects of use and disuse, and, in an un- 

 important manner (so far as adaptations were con- 

 cerned) by the direct action of the environment. 

 That theory was put forward by a man who knew 

 the facts to be explained. Mr. Tate Regan claimed 

 that Darwin's theory explained them and that no 

 other theory stood the test. 



Prof. W. Johannsen spoke from the point of view 

 of a geneticist. He pointed out that selection could 

 not produce anything, but it should be borne in mind 

 that I 'arwin's belief in a productive power of selection 

 was fully logical from the naive view of his time. 

 The mutations which we knew did not explain the 

 nature of evolution or the origin of large differences 

 such as the differences between families. Modern 

 genetics could scarcely contribute to a solution of the 

 main problems of evolution, but it seemed to have 

 cleared the ground from the erroneous Lamarckian 

 and Darwinian views. He himself and, he thought, 

 most geneticists were agnostics as to the mechanismr 

 of evolution. 



Air. J. T. Cunningham thought that natural 

 selection was " as extinct as the dodo," and that the 

 origin of species was due to mutations. Specific 

 characters were for the most part useless, but other 

 groups might be distinguished by adaptive and non- 

 adaptive characters. He discussed adaptation, which 

 he considered to have arisen in a Lamarckian manner. 

 Modern discoveries concerning internal secretions 

 showed how many adaptations exhibiting recapitula- 

 tion might have been produced by stimuli and 

 functional exercise. 



Dr. H. Wager urged that there was more in the 

 theory of natural selection than was implied by 

 Dr. Willis and Mr. Cunningham. Fluctuating varia- 

 tions were dismissed from having evolutionary 

 significance, but mutations were not necessarily 

 large. He reminded the audience that an alternative 

 title given by Darwin to " The Origin of Species by 

 means of Natural Selection " was " The preservation 

 of favoured races in the struggle for life," which 

 might be interpreted in modern terms as " The 

 preservation of favourable mutations." 



Prof. E. B. Poulton discussed the theory of " Age 

 and Area" in relation to mimicry, and pointed out 

 that in certain African butterflies the younger form 

 is distributed over a much wider area than the 

 ancestral type. 



Dr. Chalmers Mitchell supported Mr. Tate Regan, 

 and considered that Dr. Willis had presented merely 



NO. 2770, VOL. I io] 



