i 3 8 



UNIVERSITY OF MISSOURI STUDIES 



The theory thus developed does not pretend to be the 

 ultimate solution of the problems attacked. We do not pos- 

 sess the data upon which to found a final 

 theory. But we shall scarcely obtain these 

 data without the guidance of a the- 

 ory. Experimental research must be 

 systematic, must start from a theory, how- 

 ever imperfect this may be, in order to lead to 

 scientific advancement. If the theory here offered succeeds 

 in stimulating' experimental research in a field somewhat neg- 

 lected for many years, the author's hope will be realized. 



The necessity 

 of a 

 theory 



