494 SPONTANEOUS AGGREGATION OF FLAGELLATES 



We must now ask, what is the cause of the gradual slow retiral 

 of a Bodo band from the edges of a preparation when left over night 

 at room temperature, of its more rapid retiral in the cold, and of its 

 spreading out again when replaced at a higher temperature? These 

 phenomena also can be imitated with an indigocarmine preparation 

 such as has just been described. When the slide with its yellow 

 central region separated from the blue border by a square purple 

 band, the sides of which are parallel to the edges of the cover-glass, 

 is placed in the ice chest the purple band contracts. It retires towards 

 the center of the solution, the corners of the square becoming rounded 

 until it assumes the form of a ring. When the preparation is now 

 replaced at a higher temperature the purple ring expands again, 

 becoming square as it nears the edges of the cover-glass. When 

 left over night in a moist chamber at room temperature the purple 

 band recedes slowly from the edges of the liquid. Now this behavior 

 of the purple band of partially reduced indigocarmine is due to the 

 different solubilities of oxygen at different temperatures. On the 

 ice more oxygen goes into solution and drives the purple band inwards. 

 When the preparation is warmed some of the oxygen goes out of 

 solution and the purple band approaches the edges again. Left at 

 room temperature the concentration of dissolved oxygen at the edges 

 of the preparation gradually rises because the water was not originally 

 saturated with oxygen, so that the band retires slowly inwards. The 

 cause of the similar behavior of a Bodo band under the same conditions 

 must be precisely the same. The band of flagellates moves to a 

 position nearer to or further from the edge according to the lesser 

 or greater amount of oxygen going into solution at the different 

 temperatures. An alternative explanation is that the Bodo band 

 retires inwards at a low temperature or when left for some time 

 because under these circumstances the flagellates change their oxygen 

 optimum. The suggestion, however, becomes very improbable in 

 view of the parallel behavior of the indigocarmine band, the cause 

 of which is known. 



Confirmatory Facts. 



There are several facts and experiments which strikingly confirm 

 the conclusion that the concentration of dissolved oxygen is the con- 



