8o Asa Arthur Schaeffer 



METHODS or INVESTIGATION 



There are two methods by means ot which choice of food can be 

 investigated in such an organism as Stentor. 



One method is specific, consisting in observing and recording the 

 path and fate of each particle that is fed to the Stentor. This is 

 accomphshed as follows: A capillary pipette is made by drawing 

 out an ordinary pipette to a very fine hair having an internal dia- 

 meter of about 75/i or less. Food particles, such as Phacus, 

 Euglen^e, etc., or indigestible particles, as the experiment ma\' 

 demand, are then sucked up into the pipette with some water. 

 Several Stentors are transferred from the original culture dish into 

 a watch glass with a few cubic centimeters of the culture solution 

 and placed on the stage of a binocular microscope of the Braus- 

 Driienr type. A magnification of about 65 diameters is used. 

 After the Stentors have become attached to bits of detritus in the 

 watch glass, the particles are fed from the pipette, the end of which 

 is held very carefully about the diameter of the disk away from 

 and above the Sten tor's disk. The particles are for the most part 

 fed successively, in each case waiting until the foregoing particle 

 is swallowed before the succeeding one is set free from the pipette. 

 Much time is of course taken up in the recording of results and m 

 getting the pipette into position again for feeding. Some idea ot 

 the time required in making such feeding experiments may be 

 obtained from the fact that a successful experiment in which 120 

 particles are fed extends over about an hour and three-quarters. 

 When it was desired to feed two or more kinds of food, the parti- 

 cles were first mixed in the desired proportion and then sucked into 

 the pipette. If the various sorts did not come in the desired order 

 some of the particles were merely dropped to the bottom of the 

 dish and not allowed to touch the disk of the Stentor at all. In 

 this way the order of substances in a mixed stream was under con- 

 trol. Nothing was fed when it was not intended. With a method 

 of this degree of exactness there seems to be no reason why the 

 results should not be thoroughly reliable. The only considerable 

 variant which seems possible is the physiologic state of Stentor — 

 which it is the purpose of this paper to investigate. 



