No. 494] ZOOLOGICAL PROGRESS 



1 L9 



time, it is perfectly legitimate to subdivide previously 

 described single species into smaller aggregates to be 

 denominated new species. 



Where this movement will lead to is not easy to fore- 

 tell. It was hoped that the statistical methods of bio- 

 metric work would make possible a solution of some of 

 the difficulties in defining species, but, while these methods 

 enable the investigator to discover and express differences 

 between large groups of individuals vastly more accu- 

 rately than the old methods of simple inspection and 

 rough mental estimate did, they do not settle what char- 

 acters or how many such shall be used in denning a spe- 

 cies or what degree of difference must be arrived at 

 before a given aggregate may be divided into two or more 

 species. Nor does the recent proposition of de Vries come 

 nearer to solving the problem. According to this investi- 

 gator real species can be denned by a certain combination 

 of characters which vary not in a continuous way hut by 

 leaps. These discontinuous variations constitute differ- 

 ences that may be as distinct as the differences between 

 chemical elements ; and any real or elementary species, as 

 de Vries designates it. is distinguished by one or more of 

 these elementary characters. Such elementary species, 

 though open to variation, are cut off absolutely in their 

 specific characters from other such species by a gulf 

 that is never bridged by intermediate forms. Hence 

 they ought to be as easily and distinctly describable as 

 chemical compounds or even chemical elements. These 

 elementary species, which have been for the most part 

 heretofore ignored or passed over simply as races, etc., 

 are the real units of systematic zoology and are much 

 more numerous than the ordinary or Linnapan species. 

 In fact, each Linnaean species probably consists of many 

 elementary species and consequently the acceptance of 

 this proposition would multiply the number of described 

 species many times. But the difficulty with this proposal 

 is to be found in the fact that an inspection of the so- 

 called elementary characters shows them to be not so 



