( 4;)8 ) 



development, whicli allow a fusion of the lespoetive forms, as the nej;a(ive does to 

 the affirmative, and therefore is diflerent from the lower degrees of development not 

 only as to degree, but also as to kind. 



5. From the fourth point it follows that if specific difference means a difference 

 of kind, not only of degree, the chief criterion of specific distinctness of a given form 

 of animal or plant is the impossibility of fusion with other forms. 



Hence we have to accept the following definition of the term " species " : — 



A species is a ijroup of individuals xvldch is differentiated from all other' cooi- 

 teniporary groups by one w more characters, and of ivhich the descendants which 

 are fvlly qualified for propaijation fo')'ni again under all conditions of life cme or 

 more groups of individuals differentiated from the descendants of all other groups 

 by one or more characters. 



The reasons why we do not accept one of the definitions of species as given by 

 other authors are best stated by shortly discussing the definitions of Eimer, 

 Romanes, and Wallace. 



I. Eimer"s definition * is a purely jjhvsiologifal one : — 



" Species are groups of individuals which are so modified that sexual intercourse 

 does no longer take place between them and other groujis, or that successful inter- 

 course is not unlimitedly possible." 



Apart from the criteria of the term •' si)ecies" in tliis di-)inition being negative, 

 we cannot accept the definition simply because it is not a definition in consequence of 

 its allowing us a choice between two different criteria. I..et us discuss these criteria 

 separately : — 



1. " Species are groups of individuals which are so modified that sexual inter- 

 course does no longer take place." 



According to this definition the prevention of intercrossing is a con.secjuence 

 of the modification of the groups of individuals. From several sides attention has 

 been drawn to the occurrence of psychological selection among the individuals of the 

 same species, especially among specimens belonging to varieties of domesticated 

 animals. As in these cases intercrossing does not take place on account of a dislike 

 of the respective varieties, i.e. as the prevention of intercrossing is a consequence 

 of the modification of a species into varieties, these varieties would conform to 

 Elmer's definition, and therefore would have to be considered as distinct species. 



2. " Species are groups of in(li\iduals which are .so modified that successful sexual 

 intercourse is not unlimitedly possible." 



In the second part of his studies on Papiliosf Eimer claims priority over 

 Romanes as to the principal idea ujwn which Romanes based his theory of Physiological 

 Selection. The chief j)remiss of this theory is the occurrence of such a variation 

 among the individuals of the .same species tliat a group of specimens does not only 

 not cross with the rest of the individuals on account of dislike, but is infertile or 

 restrictedly fertile with them, whereas the specimens of that group are inter se com- 

 pletely fertile. If such a physiological variation occurs, and there is no i-eason wliy 

 it should not occur, that group of individuals would constitute a species according to 

 Elmer's definition, in spite of the absence of morphological distinguishing characters, 

 and in spite of the offspring of the phy.siological variety belonging partly to the 



* Kinier. ArthiUhmri mid VcriraiuUgchaft hi-i Schmettrrliinjtn. 1881). j). 16. Konuiiiti-fl, in Darwin ami 

 A/O'r Darwin, 1895. p. 229, says that the *• purely physiological (ictinition is not nowa^tays PTitertainc^l li.v 

 any naturalist." 



t Eimer, Artbilduny unil Verwamltirliaft bri Schmrtterlimjen !I. .lena, IS9.5. p. 14. 



