326 GEORG DUNCKER [NOVEMBER 
That elementary complex of individuals, which is usually the 
starting-point for zoological and botanical investigations, is the species. 
More or less exclusively, all biological, systematic, and anatomical 
results are referred to this. But the species is by no means an 
elementary group; even if we omit its systematic sub-groups, the 
variety and the race, we find it empirically to be composed of in- 
dividuals which are separated by space and time, and are allied to 
each other in different degrees of kinship. In those individuals of 
the same species there regularly occur morphological differentiations 
of their common characters, caused by constitutional factors (conditions 
of sex, stage of development) as well as by the sum total of recognis- 
able external conditions of life (locality, geological formation, climate, 
food, ete.). Really elementary complexes of individuals, eo ipso 
coherent, are only those of which the morphological qualities have 
not been differentiated by any of the factors just enumerated. But 
even in such a “FORM-UNIT,” as I have called it [7], we find on 
investigating the distinctive characters that there are individual 
differences. 
Therefore the species is not elemental, a conclusion strengthened 
by the difficulty (bordering upon impossibility) of definition. It splits 
up in numerous variable form-units, produced by different factors, 
which frequently may be united into races or varieties. Each form- 
unit is a sum of more or less different individuals, the characters of 
which change in the course of development, that is, in time, but 
appear constant at a given moment, so that it is not justifiable to 
speak of varying or non-varying individuals. On the other hand, 
groups of individuals are variable in every moment of their existence 
and in each of their characters. Therefore the fact of variation is to 
be seen only in the characters of groups of individuals, and to be 
investigated only in these. 
The exact study of variation affords a better understanding of the 
systematic relations between groups of individuals; it is a means of 
distinguishing pathological from normal states ; but it owes its highest 
importance to its bearing on the theoretical explanation of the re- 
lations between organic individuals, 7c. in regard to heredity and 
evolution. 
The objects of an investigation on variability are the characters of 
a complex of individuals and, according to the laws of induction, at 
first of the most primitive complex, the form-unit. The aim of this 
investigation is twofold; on the one hand qualitative, to discover the 
real individual differences in these characters, which we may call the 
variants ; on the other, quantitative, to discover the relative frequencies 
of the single variants of each character determined. 
The principal difference between a character of a single individual 
and one of a complex of individuals, therefore, consists in the possibility 
of expressing the former by a single variant, while the latter requires 
ea 
