852 



SCIENCE 



[N. S. Vol. XXX. No. 780 



able variations and such characters as serve to 

 distinguish one kind of poultry or mammal 

 from another. To-day we see more clearly that 

 a new character, such as " angora " hair or an 

 extra toe, belongs to a different category of 

 variations from ordinary fluctuations in the 

 length of the hairs on the body of a cat and 

 variations in the thickness of a toe; for a 

 hair will be more or less long according to the 

 nutrition it receives at the base (and this 

 varies at different times), and the toe will be 

 more or less thick, depending on the use to 

 which it has been put. The variations de- 

 pendent on environment or use are, so far as 

 we know, not inherited, while the new char- 

 acters clearly are. Thus the primary classifi- 

 cation of variations is based on their herita- 

 bility. This much was pre-deVriesian. 



The new vievsrpoint, introduced by deVries, 

 and extended by Johannsen, affects the inter- 

 pretation of those slight variations that seem 

 to be independent of environment and are 

 distributed about a mean value in the form 

 of the familiar " frequency polygon." The 

 biometric " school " laid stress on this sort of 

 variation, and held that by selective breeding 

 from the extreme variants through many 

 generations an indefinitely wide departure 

 from a starting point might be effected. This 

 deVries denied, but. held that, while such se- 

 lection might lead to a certain departure from 

 the mode, the degree of such a departure was 

 restricted through a strong regressive tend- 

 ency. Here Johannsen steps in, analyzing 

 more completely this result of breeding from 

 the extremes of the frequency polygon. 



The fundamental principle of Johannsen is 

 that an ordinary frequency polygon is usually 

 made up of measurements of a characteristic 

 belonging to a non-homogeneous mass of in- 

 dividuals; that it is really analyzable into 

 several elementary masses each of which has 

 a " frequency polygon " of its own. In each 

 elementary polygon the variation is strictly 

 due to non-inheritable somatic modifications, 

 selection of extremes of which has no genetic 

 significance. But the selection for breeding 

 of individuals belonging to different elemen- 

 tary polygons, lying, say, at the extremes of 



the complex, may quickly lead to an isolation 

 of these elementary polygons, the constituent 

 individuals of which reproduce their peculiari- 

 ties as distinct elementary species. Thus 

 Johannsen holds that not only do individuals 

 with qualitatively dissimilar characters be- 

 long to distinct elementary species, but often 

 such as are only quantitatively unlike. The 

 complex variation-groups are called by the 

 author phcenotypes, or false types, the elemen- 

 tary variation groups are genotypes, or genetic 

 types. 



What is the proof of the existence of these 

 two types? It lies in the author's experi- 

 ments in breeding " in the pure line." When- 

 ever, in a self-fertilizing species, a character 

 is measured through successive generations it 

 does not show a regression toward the 

 mediocre position of the entire population, 

 but regression occurs only to a near-by mode 

 of the elementary genotype. In such a species 

 regression to mediocrity occurs only when we 

 consider the offspring of parents which, even 

 though similar, belong to distinct genotypes. 

 Tor, since genotypes overlap, the parents, 

 though quantitatively similar in any organ, 

 may have children that regress in an opposite 

 direction to the modes of their (unlike) geno- 

 types, and thus be quite dissimilar to their 

 parents. In the long run the change from 

 parents to offspring will be in the direction 

 of mediocrity. This is the usual result and it 

 has obscured the facts of genotypes in the 

 midst of, and as constituents of, the pheno- 

 types. Now, although self fertilization is 

 necessary to the proof of the existence of 

 genotypes, such types are believed to be uni- 

 versal and necessary to the interpretation of 

 heredity and evolution. So the author in his 

 500-page book rewrites the science of heredity 

 from the new standpoint. 



The book is in the form of twenty-five lec- 

 tures. The first six are devoted to variability 

 and its statistical analysis; then follow five 

 devoted to selection and regression; three to 

 aberrant and complex frequency polygons; 

 four to correlation; two to types of variation; 

 one to effect of environment ; two to hybridiza- 

 tion; one to nutrition and one to human 



