F. KiDD AND C. West 119 



size; this is a common practice in seed selection, water-culture experi- 

 ments, etc., and in scientific work generally. The main conclusion to be 

 drawn from the available evidence, to which we shall allude immediately, 

 concerning the selection and use of seed of different size, both in the 

 case of forest-trees and in the case of herbaceous plants of economic 

 importance, is distinctly in favour of the general conclusion that large 

 seed produces a more vigorous crop and larger yield, although im- 

 portant contradictions to this statement have been published. 



In view of its importance in practice it is essential in considering this 

 question to analyse the conditions which affect the size of the seed. The 

 latter may depend either upon the environmental conditions of the 

 parent plant, or upon the position of the flower on the inflorescence and 

 of the inflorescence on the plant, or upon a hereditary factor. 



In experimental work planned to determine differences in crop pro- 

 duction arising from the selection of seeds of different size there has 

 usually been no distinction drawn between the various causes influencing 

 the size of the seed. This failure has introduced some confusion into the 

 interpretation of the results which have hitherto been obtained in re- 

 search work on this subject. If we are dealing with a hereditary factor 

 for large seed, the average weight of seed in the crop, but not neces- 

 sarily the total weight of the crop (cf. Waldron(75)), would probably be 

 pre-determined by heredity, and the question with which w^e are 

 dealing, namely, a question of physiological pre-determination, would 

 not come in. 



(a) Determined by Climatic and Edaykic Factors acting wpon 

 the Parent Plant. 



When the size of the seed is altered by climatic or edaphic factors 

 acting through the parent plant, any resulting increase or decrease in 

 yield from the seed thus altered can only be a matter of physiological 

 pre-determination . 



The literature bearing on this question is very scanty. The effect 

 of various light-intensities upon seed-formation and the size of seeds 

 has been subjected to investigation by Combes (ii). He tested the ger- 

 mination capacity of the seeds of different sizes which he obtained by 

 differential treatments of the parent plants. His plants were grown for 

 the whole period of their existence under five intensities of light obtained 

 by glass screens of various thicknesses which gave ^, I, |, |, and 1 of 

 the total insolation as determined bv Wiesner's method. 



