70 The Ottawa Naturalist. [July 



two kinds or aggregates of individuals may be spoken of as 

 distinct species, is involved in the title of this paper, "What is 

 a species"? 



The query is not an easy one to answer where very similar 

 forms, and, in many instances, even where totally dissimilar 

 forms are concerned. For it must be understood from the first 

 that dissimilarity of form does not necessarily indicate distinct- 

 ness of species in the broadest sense of the word. The late 

 Charles Darwin wrote volumes dealing in one way or another 

 with the subject. From his point of view, about the best defini- 

 tion that can be given a species is "an aggregate of individuals 

 capable of producing, under natural conditions, progeny of their 

 own form, through successive generations". That is the most 

 exclusive sense in which the term can be used. But it is obvious 

 that the difficulty of discussing how far that ability exists, or 

 whether it exists at all, in a very large proportion of the multitu- 

 dinous forms of organic life, has given rise to much of the past or 

 existing controversy of the relationship of forms or kinds. The 

 power of reproduction exists in very many instances between 

 allied kinds generally admitted to be distinct species; generally 

 speaking the more closely allied two species are, the more fre- 

 quently will crosses between them be found in localities where 

 the two live together. But amongst animals, with few ex- 

 ceptions, the reproductive power in such cases is not transmitted 

 to the offspring. In other words, true hybrids, i.e., the progeny 

 of crosses between different but allied kinds, are themselves 

 infertile, or sterile, or. in the case of the few exceptions, they 

 become sterile in the subsequent generation. This does not 

 apply in the same way to plants, in which the means of perpetu- 

 ation are very different, hybrids much more frequently fertile, 

 and species still harder to define. 



It happens that while some species are confined to very 

 small areas, called "local species", others exist all over a con- 

 tinent, and are called "generally distributed" species. 



Now, supposing it were possible to apply this reproductive 

 test to all the various forms in different groups throughout, say, 

 North America, it would be found that in some cases one species 

 existed in much the same form wherever it was found, that is, 

 that different individuals in the same district showed little or 

 no variation one from the other, and that an individual or 

 specimen from a district, say, on the east coast, differed in no 

 essential characters from one from the west coast. Such is 

 called a constant or non-variable species. In other species, 

 individuals or "specimens" may be found var3nng much from 

 others in the same locality, it may be in colour, size, relative 

 dimensions of different parts, etc. Specimens so differing are 



