LAW OF GEMINATE SPECIES JORDAN 119 



hypothesis into that of fact. But the application of this law or rule to plants 

 and to one-celled animals has been questioned. So far as rhizopods are con- 

 cerned, Dr. Kofoid finds that the species are in general sharply defined and 

 of the widest distribution in the sea, so that we can hardly state laws as de- 

 fining their geographical distribution. To these minute floating animals, the 

 sea scarcely offers barriers at all, and the recognized species do not seem to be 

 products of geographical isolation. Doubtless these species in duration and 

 in nature correspond more nearly to genera or families of higher animals 

 than to actual species. Perhaps minor specific difEerences such as we note 

 among arthopods or vertebrates are intangible or non-existent. The effects 

 of isolation may be tangible only among forms which possess more varied 

 relations with their environment. 



The application of this law to plants has also been denied. But gem- 

 inate species are just as common in botany as in zoology, and the effects 

 of isolation in species-forming are just as distinct. The law is just as patent 

 in the one case as in the other. It is merely obscured by other laws or con- 

 ditions which obtain among plants. 



In the nature of things, most physical barriers are more easily crossed 

 by plants than by animals. The possibilities of reinvasion are thus doubtless 

 much increased. The plant is limited by climate, rainfall, nature of soil, and 

 the same species is likely to occupy all suitable locations within a large area. 

 Animals are more mobile than plants within their range, a fact which tends 

 to keep the interbreeding masses more uniform. In the struggle for exist- 

 ence, the plant is pitted against its environment. Whether the plant survives 

 or not depends not much on the nature of the seed, but mainly on its relation 

 to the spot on which it falls. There is little selection within the species due 

 to the choice of one individual as against another. Selection can only happen 

 where plants are overcrowded, and there the survival is mainly that of the 

 seed whose roots run deepest. There is little room for struggle between 

 closely related species. Each individual grows — if it can — on the spot 

 where it falls. The variations among plants are great, but these variations 

 are mostly lost unless reinforced by segregation. There is no likelihood of 

 the survival of DeVries' mutants (or hybrids) of the evening primrose if 

 these forms are left free to mix in the same field. 



Among plants we often notice the fact — rare, though not unknown 

 among animals — of numerous species of the same genus occupying the same 

 area. In some cases these species are closely related, suggesting mutants, and 

 in other cases the relation indicates the existence of hybrids. In California, 

 for example, there are in the same general region many species of Lupinus, 

 of Calochortus, of Ceanothus, of Arctostaphylos, of Eschscholtzia, of 



