CHOICE OF FOOD IN AMEBA 245 



be called to the fact that man does not acquire a knowledge of 

 the chemical composition of any single substance directly by 

 a sense organ. The nearest approach to obtaining such knowl- 

 edge is in the smell sensations of the elements chlorine, bromine 

 and iodine; but even with these substances it has been pointed 

 out that the mucous membrane, aside from the olfactory nerve 

 endings, may be stimulated, and that nascent hydrogen may 

 combine with these elements before the sense organ is stimulated. 

 We cannot conclude, therefore, that iodine, bromine and chlorine 

 are sensed in the elemental condition. Now since we do not 

 find any conclusive evidence in man (or elsewhere) that the 

 chemical composition of substances is sensed, not even in a 

 single case, one must, to say the least, feel suspicious of the 

 correctness of the chemical hypothesis. Differences in surface 

 tension, adsorption, rate and direction of diffusion of dissolved 

 molecules, ionization, reflection of light from suspended particles, 

 etc., are probably among the properties which play a part in 

 determining the feeding behavior of ameba. The disturbance 

 due to moving animals in water, which is transmitted as waves 

 of changes of pressure to the ameba, and which, as we have 

 seen, is the most potent stimulus for setting off the feeding 

 reaction, is purely a physical property, and it is not unlikely 

 that the effect of the pressure waves is a change in the distri- 

 bution of surface energy on the part of the ameba affected. 

 While we are far from explaining all the acts of selection in 

 ameba on a physical basis, a respectable beginning has never- 

 theless been made, as we have seen in the section devoted to 

 the effect of mechanical agitation. 



What has been said about the basis of selection in this section 

 has had reference only to individual acts of selection; the eating 

 of a grain of carmine or of globulin, or the avoidance of a grain 

 of carbon, each taken by itself. A very important element 

 affecting selection in a broader sense, that is, selection as the 

 sum of the feeding behavior of an ameba, has not yet been 

 mentioned. This element is the effect of previous experience. 



The importance of this factor in selection is equalled by the 

 variety of ways in which it may express itself. The previous 

 stimulus and response characteristically affects the succeeding 

 response. In some cases the effect is slight compared with the 

 stimulus ; in others the effect of the previous stimulus and response 



