248 A. A. SCHAEFFER 



and paramecia, but there is no ease where it has been definitely 

 shown that a characteristic response is due to some chemical 

 effect. The reactions of Paramecium, therefore, can be con- 

 sistently and adequately explained by assuming that physical 

 qualities act as the stimuli. 



It remains to mention that selection of food is accomplished 

 roughly by the avoiding reaction. If a Paramecium comes to 

 a locality where an injurious chemical is dissolved in the water, 

 the Paramecium avoids the locality and all solid particles in it. 

 If, however, an attractive chemical such as carbon dioxide is 

 present, the Paramecium enters the water where the diffused 

 chemical is and eats all the particles of proper size there without 

 expressing any choice among them, for carbon dioxide is not 

 injurious in dilute solutions. Inside of the funnel or gullet of 

 Paramecium there is practically no selection, though Metalnikow 

 has shown that the power to discriminate may be developed 

 very appreciably by proper experimentation (Metalnikow, '12). 



Stentor caeruleus. The blue stentor feeds on small particles 

 such as bacteria, and also on larger forms such as ciliates, 

 flagellates, rotifers, zoospores, unicellular algae, etc. It is 

 difficult to say what the normal diet of stentor is, but it may 

 be mentioned that laboratory cultures can be raised very success- 

 fully on bacteria and flagellates such as chilomonas. ' The 

 number of particles eaten per day depends of course upon the 

 size of the particles. Ten paramecia may be quite sufficient, 

 but it would require thousands of flagellates and millions of 

 bacteria, if either of these kinds of organisms were eaten 

 exclusively. 



Stentor has a food sorting mechanism by means of which 

 any particle can be ingested or rejected ' at will.' The selective 

 mechanism acts with great precision, especially when the particles 

 are of appreciable size. In such case relatively few particles 

 need to be eaten to maintain life, and there is consequently 

 sufficient time for careful selection of individual particles. 



It is reasonable to suppose that it is advantageous to stentor 

 to sort out the food particles from all that are brought to it, 

 whenever these particles are of large size. 



Stentor still possesses the power to choose its food especially 

 where larger particles are concerned. But indigestible material 

 may be eaten when very hungry. Very small grains of indi- 



