Apkil 1, 1904.] 



SCIENCE. 



549 



I believe, therefore, that the right plan is 

 to put back the meeting of the American Asso- 

 ciation to its old place near the close of the 

 summer vacation, and to leave convocation 

 week for the smaller, more homogeneous, and 

 less popular associations of vforking scien- 

 tists. William North Rice. 



NATURAL SELECTION IN KINETIC EVOLUTION.* 



That there are species, varieties, mutations 

 or hybrids v^hich differ in one, two, or three 

 characters, as commonly assumed in discus- 

 sions of Mendel's laws, is a misleading as- 

 sumption. To speak of a species as having 

 developed in one direction or as having a 

 single peculiar character may be permissible 

 for taxonomic purposes, but in evolutionary 

 studies it is careless to forget that the di- 

 versity is general, if not complete. The di- 

 versity of varieties and species is like that of 

 individuals, but greater. Evolution, which 

 is a continuous summary or integTation of 

 this individual diversity, is not a simple 

 process, but highly multiplex; as much so, 

 indeed, as the lines of descent in which the 

 life of the species goes forward. A composite 

 general direction is maintained by the species 

 because the multitudinous strands of individ- 

 ual descent are bound together by interbreed- 

 ing. The variations take place in particular 

 threads, but evolution signifies rather the 

 progressive change of the whole organic net- 

 work, -f ' 



The evolution of a new type means changes 

 in many directions and of many kinds, in the 

 germ cells and in the various tissues and 

 organs, as well as in the external form of the 

 complex cell-colony which we are accustomed 

 to look upon as a single individual. Each 

 cell, tissue, organ and feature is undergoing 

 evolution, and for normal and permanent 

 progress these manifold developments must 

 keep together. When single lines or slender 

 strands of descent are separated from the 

 main network the congruence of type is lost. 

 The normal variation and individual diversity 



* Read before the Biological Society of Wash- 

 ington, March 19, 1904. 



I The Popular Science Monthly, March, 1904, p. 

 451. 



of the species having been eliminated, the 

 evolutionary coordination of cells, organs and 

 functions breaks down, and abrupt changes or 

 aberrations of heredity appear. These degen- 

 erative mutations may not differ in their es- 

 sential nature from normal variations, but the 

 conditions of their appearance are abnormal, 

 and the results often disastrotis.* 



A domestic variety may be ' improved ' by 

 the further increase of the one or two char- 

 acters or qualities which render it valuable, 

 but a new specific or generic type is the com- 

 pound or resultant of many variations in 

 many characters. By close selection which 

 restricts evolutionary progress to a narrow line 

 of descent a ' single character ' may push out 

 farther in a decade than the natural multiplex 

 evolution would carry it in a century or a 

 millennium, but such a specialization weakens 

 and unbalances the organism, and is a process 

 of degeneration rather than a constructive 

 evolution. Selective inbreeding and other 

 forms of isolation accentuate single charac- 

 ters, but the interbreeding of normally diverse 

 individuals (symbasis) weaves new types out 

 of the variations of many lines of descent. 



The neglect of this distinction vitiates much 

 evolutionary literature, both that which treats 

 selection as an actuating ' force,' and that 

 which rejects selection for ' discontinuous 

 variation' or 'the mutation theory.' f It is 



* Mutations, like hybrids, are sometimes com- 

 pletely sterile, and they may have at the same 

 time an increased vegetative vigor. The vegeta- 

 tive vigor of many mutative varieties of domesti- 

 cated plants has doubtless delayed the recognition 

 of their abnormal evolutionary status, though 

 the abnormality of infertile hybrids has long been 

 appreciated. It is paradoxical, indeed, that the 

 increased vigor which accompanies normal varia- 

 tions and crosses should also attend degenerative 

 changes, but there is room for this apparent 

 contradiction in so complex and many-sided a 

 process as evolution. 



f Very recent examples of the latter tendency 

 are found in Professor Morgan's ' Evolution and 

 Adaptation ' and also in Dr. D. T. MacDougal's 

 review of this work {Torreya, 3: 185, December, 

 1903). Professor Morgan refers (p. 368) with 

 approval to an admission bj' Darwin that selec- 

 tion can not explain dimorphism in plants be- 



