140 EDWIN CARLETON MACDOWELL 



It would be difficult to conceive of an organism that could 

 mature free from all forces except those conditioned by the 

 germ plasm. Every stage of an individual's growth is shaped 

 by the reaction of the germinal to the extra-germinal, the 

 hereditary to the environmental. In the case of the flies, the 

 role of the environment in controlling the bristle numbers is 

 obvious; we see changes in one, we see changes in the other. In 

 some other cases of variation it is not possible to see, in this 

 clear and simple way, changes in the environment associated 

 with changes in somatic structures. Failing to find such a 

 relationship, it is tempting to conclude that the variations in 

 the soma do not indicate environmental changes but rather ger- 

 minal variations. Can it rightly be supposed that in the ab- 

 sence of correlation between obvious fluctuations in temperature 

 and food and such like, that there are no other controlling 

 influences than those originating in the germ plasm? The an- 

 ticipated criticism of this work is that the environmental influ- 

 ences are too potent for the germinal changes to appear. But 

 suppose the power to influence bristle numbers, that seems as a 

 matter of fact to lie in the amount of food eaten, etc., rested in 

 some unknowable condition outside the germ plasm, such as 

 the age of the sperm before fertilization, or the temperature of 

 the mother's body during the maturation of the egg, then this 

 objection would probable not be considered. The individual 

 may be considered to be a reaction between the germ plasm and 

 the environment but does it not sometimes seem that students 

 of selection are so intent upon their consideration of the germ 

 plasm that there is a tendency to ignore the scope of the influence 

 of environment? In this way one of the most essential tools in 

 attacking the problem is neglected. The point to be empha- 

 sized is that there is very Httle support for the supposition that 

 the soma mirrors the germ plasm in all cases except when obvious 

 environmental relations are found. It seems hardly possible 

 that one can look forward to ever establishing firmly an exact 

 relationship between soma and germ plasm, at least in bi-sexual 

 multicellular animals. In the present case the difficulties are 

 all to clear too permit the statement that the germ plasm did not 



