Selection of Food in Sientor Cceruleus {Ehr.) 93 



which will be taken up a little later. If this is the case the S ten tor 

 discriminated very well indeed what was Phacus and what was 

 glass or starch. An interesting feature of this experiment is the 

 number of loops that are recorded for the Phacus which were 

 rejected and the absence of loops in the paths of the particles of 

 glass. 



So far, these experiments show that Stentor can select Phacus 

 from a stream of mixed particles in which there are one or two 

 kinds of indigestible particles mixed with Phacus. There are 

 two possibilities as to the way this selection is accomplished. 

 First it may be that Stentor ingests from a mixed stream only one 

 kind of food particles (such as Phacus triqueter), and rejects all 

 other kinds of food and indigestible substances. The other pos- 

 sibility is that Stentor ingests all sorts of food particles and re- 

 jects all sorts of particles that are not food. To determine which 

 of these alternatives is the one which actually obtains, the follow- 

 ing experiment was performed in which two kinds of food par- 

 ticles, Phacus triqueter and Euglena viridis, were fed in a mixed 

 stream with two kinds of indigestible substances, powdered sul- 

 phur and powdered glass. The following are the results: 



Experiment ^. Selection of Phacus and Euglena from Sulphur ajid 



Glass 



The Stentor upon which this experiment was tried was the same 

 one which had submitted to the second experiment, on the pre- 

 vious day (p. 90). It still contained considerable amounts of 

 partly digested food which probably represented the Phacus that 

 were eaten the day previous. The starch grain which was eaten 

 in Experiment 2 was not to be seen, and it is probable that it was 

 voided some hours after it was ingested. That it was digested 

 is made highly improbable in view of the work of Meissner ('88) 

 who found that nearly all of the potato starch which the S ten tors 

 ate was not digested, and that some of the starch which remained 

 in Stentors for over 48 hours was practically in the same con- 

 dition as when fed. There is therefore no good ground for sup- 

 posing that any part of the behavior of the Stentor in this experi- 

 ment was due to the fact that a starch grain was ingested on the 



