64 THE VARIATION OF ANIMALS IN NATURE 



difference in terminology conceals a real difference in the 

 type of variation (i.e. in distribution). 



Below the level of varieties and subspecies the ordinary 

 task of the systematist is not pursued. All that we have 

 said concerning the validity of the species-concept applies 

 with equal truth to the subdivisions of the species itself, viz. 

 the uncertainty as to their genetic status and the difficulty of 

 standardising the concepts. 



It remains for us to notice the various attempts that have 

 been made to incorporate the results of population-analysis 

 into taxonomy. A good account of this is given by du Rietz 

 (I.e.), who reviewed and attempted to harmonise all the various 

 terms proposed. It is enough to state that intensive popula- 

 tion-analysis (dating from Alexis Jordan's pioneer work) has 

 revealed the presence within systematic species of various 

 subordinate elements which are imperfectly represented by 

 the old terms variety and subspecies. It is clear that there is a 

 basic distinction, now generally recognised and described in 

 detail by du Rietz (I.e., pp. 349-354), between a population 

 forming a local (variety) as opposed to a regional (subspecies) 

 element in a species. The extent to which the Jordanon 

 (Lotsy), microspecies and elementary species (Jordan), natio 

 (Semenov-Tian-Shansky), etc., are merely synonymous with 

 one or the other of these is an academic point, and it is similarly 

 obvious that the line between ' local ' and ' geographical ' 

 race is quite arbitrary. The differentiation of populations 

 into a large number of intercrossing ' biotypes ' and the way 

 in which such subordinate elements are distinguished by 

 isolation lead to a very finely graded hierarchy of local 

 groupings (cf. Crampton, 1 916-1932 ; Gulick, 1905 ; Heincke, 

 1898), and it would be undesirable to attempt to define 

 these by a rigid terminology. Some taxonomists have recog- 

 nised a finer distinction under the name ' forma ' to designate 

 a purely fluctuational type (— 'modification') or, with a 

 more non-committal connotation, to designate a type ' occur- 

 ring sporadically in a species-population and not forming a 

 distinct local or regional facies in it ' (du Rietz) . 



Finally, we would draw attention to the attempt which has 

 been made by Fenton (1931, p. 30) to remodel the traditional 

 Linnean system so as to suit the findings of palaeontology. 

 His definitions of ' subspecies ' and ' form ' are not to be 



