380 



THE AMERICAN MUSEUM JOURNAL 



As regards the vertebrate animals this 

 is the origin of species. We can pro\"e it 

 in thousands of cases, as surely as we can 

 prove anything by experiment when 

 nature lays down its conditions. While 

 now and then "mutations" or discon- 

 tinuous variations occur, especially in 

 domestic races of mixed origin, I do not 

 know of a single species of vertebrate 

 animal which could even hypothetically 

 be supposed to have originated from a 

 mutation. 



Among the different races or variants 

 of birds, none is recognized by name, or 

 as having permanence, by competent 

 ornithologists, unless these variants have 

 a geographical basis. Subspecies are 

 therefore geographical races, and these 

 are the stuff that species are made of. 

 The word geography should not be con- 

 strued too literally; one does not need 

 to go far up a mountain side, or into 

 the depths of the sea, to find a condition 

 of things radically different from those 

 at sea level, on land, or in the ocean. 



I have thus far referred only to verte- 

 brate animals because I know these best, 

 and because their intensified life makes 

 their relations to one another and to the 

 environment sharply defined; but I 

 know that exactly the same rules apply 

 to the invertebrates and to plants in 

 their degree. Among plants the in- 

 fluence of some features of the environ- 

 ment are very great, as soil, moisture, 

 and temperature; while barriers are 

 often passed more easily than with most 

 animals. Some plants moreover are 

 self-fertilized, and therefore really iso- 

 lated from their kind, although growing 

 in the same neighborhood. There are 

 more cases of nearly related species 

 growing close together among plants 

 than among animals. Yet even among 

 the higher vertebrates there occur some 

 of these cases — reinvasion perhaps on 

 the part of the separated and modified 

 forms. These however, are not common 



enough to aifect the main proposition, 

 and each case should be studied on its 

 merits. The real point at issue is this;^ 

 Varieties, or incipient species, blend |^" 

 with the main group unless kept separate t 

 in the breeding season. Migratory birds _^ 

 and fishes may mingle outside the breed- 

 ing season M'ithout likelihood of crossing 

 and thus of losing distinctive characters. . 



In the discussion of the " Law of^ 

 Geminate Species," I have approached - c 

 the same problem from another side. 

 Geminate, or twin, species are forms 

 substantially alike in characters and 

 habit, but separated by barriers which 

 hold them permanently apart. These 

 abound in botany — roses, maples, plane 

 trees, brambles, Clintonia, Trientalis, 

 Trillium — they can be found in almost 

 every widespread genus. They are rela- 

 tively equally numerous among birds 

 and fishes. On the two sides of the 

 Isthmus of Panama there are more than 

 a hundred such pairs of species of fishes, 

 separated from each other since the 

 Miocene. It needs no search to find 

 them among Insects or mollusks. The 

 origin of such twin pairs is obvious, and 

 it is plainly not due to mutations after 

 the fashion of those described in a hybrid 

 evening primrose by Hugo de Vries. 



If there were one-thousandth part of 

 the evidence for the origin of species by 

 mutation that is shown for the origin by 

 separation, the theory of de Vries would 

 have standing in science. But to show 

 that species arise in general through 

 separation by barriers is not to exclude 

 any other influence whatever. We know 

 of no species however, which is not the 

 product of internal influences — heredity 

 with variation — modified by the exter- 

 nal limitations arising from selection 

 and isolation. Whatever else may be a 

 factor in the formation of species in 

 nature or in domestication, from these 

 four we can never escape — heredity, 

 variation, selection, and segregation. 



"] 



