THE PURE LINE HYPOTHESIS AND THE 

 INHERITANCE OF SMALL VARIATIONS. 



By Prof. Ernest Warren. D.Sc. 



{Read, July 9, 1918.) 



{Plate 16.) 



The question whether small, casual variations can be in- 

 herited is of the greatest importance for any clear understanding 

 of the methods of evolution, and the experiments described in 

 this paper have been carried out in the hope of affording evidence 

 on the subject. 



Certain aspects of Mendelism and the factorial hypothesis 

 are first briefly discussed in order that the object of the enquiry 

 may be more readily understood. 



I. Theoretical. 



1. Dissimilarity of Organisms. 



From practical experience it may be affirmed that two ferti- 

 lised egg-cells never develop into two organisms which are 

 identically the same in bodily structure. This diversity among 

 the individuals of a species is called variation, and the existenc; 

 of such is, of course, universally recognised. 



The nature of the adult body of an individual depends on the 

 potentialities of the fertilised tgg, and on the environment to 

 which the developing embryo and the fully-formed organism are 

 subjected. 



Neither the environment nor the potentialities of any two 

 fertilised egg-cells can be imagined to be identical in every 

 respect, and the variations exhibited by any two individuals of a 

 species result from a variable environment and a variable poten- 

 tiality. These somewhat trite statements will be admitted as 

 self-evident. 



2. Unit-Characters. 



In recent years it has become increasingly the practice to 

 regard an organism as being composed of a number of semi- 

 independent characters, and each of such characters is supposed 

 to have been represented in the fertilised egg-cell in some definite 

 manner. As a non-committal term, the name " factor " was 

 applied by Bateson to the potentiality of the fertilised o.^^ with 

 respect to any one of these characters. 



It is, however, impossible to conceive of almost any 

 character of an organism which does not involve several other 

 characters, e.g., stature is a highly complicated character. It will 

 thus be seen that in this regard the expression " character " has 

 really a technical or conventional meaning; and it implies 

 a group of characters which shows a tendency to be inherited 

 independently of other characters and as a whole. Such a 

 character is frequently spoken of as a " unit-character." 



