356 THE AMERICAN NATURALIST [Vol.XLVII 



veniently expresed, the absence of something present 

 in the germ cell that produced the brown eyes. The ap- 

 parent absurdity of saying that something is deter- 

 mined by nothing disappears when we understand that 

 this merely means that the difference between the given 

 case (blue eyes) and some other (brown eyes) is due to 

 the lack in the former of something present in the latter. 

 This sort of analysis is necessary for all statements re- 

 garding determiners in Mendelian inheritance, and when 

 properly carried out it reveals their true meaning and 

 rids them of offense. 



9. The question may be raised whether this way of 

 looking at causation is a mere practical device for clear- 

 ing thought in particular cases, or whether it has a wider 

 significance. Is all causation only of differences? Is it 

 only of differences that a causal explanation can prop- 

 erly be given? Is causal formulation inapplicable to 

 things taken by themselves, without differentiation or 

 comparison? Does all causal formulation necessarily 

 imply comparison? It appears that all this might be 

 affirmed; here the matter is raised merely as a question. 1 



3. Illustrative Questions for Radically Experimental 

 Analysis 



A. Some assert that a certain chromosome is a deter- 

 miner of sex ; others dissent. 



What experiment or experiments would decide? Or 

 has the word determiner here no experimental meaning? 

 The positive assertion is evidently absurd if it is taken 

 to mean that the chromosome contains all the conditions 

 necessary for the production of the sex characteristics 

 (male or female). Interpreted in accordance with our 

 two rules, it means merely that if two similar eggs side 

 by side produce animals of the same sex, and if from 

 one of these a certain chromosome could be removed (or 



1 Mills 's "method of differences" set forth in his "Logic" is not the 

 search for the causes of differences between cases, recommended above, but 



