492 SPONTANEOUS AGGREGATION OF FLAGELLATES 



It is not carbonic acid, then, that drives out the Bodo from the center 

 and we must conclude that, while the organisms collect into a region 

 where the oxygen concentration is lower than the saturation con- 

 centration for the atmospheric partial pressure, yet when the oxygen 

 in this region falls below a certain limiting value the organisms are 

 forced to move away again. They remain in a band surrounding 

 the central area of lowest oxygen concentration, this band representing 

 the optimum concentration of oxygen for them. It lies between the 

 central exhausted region and the outside liquid into which oxygen 

 continually dissolves from the air to replace that used up. Further, 

 since in the crowded band the Bodo consume the available oxygen 

 more rapidly than it can be replaced from outside, the size of the 

 central area of lowest oxygen concentration continually increases, 

 forcing the band to approach nearer and nearer to the edges of the 

 cover-slip. 



The fact that the Bodo band is a zone of optimum oxygen con- 

 centration at once gives the clue to the third question originally 

 asked — why does the advancing band cease to advance when it has 

 reached a certain distance from the edges of the cover-glass? As 

 the band advances towards the water-air surface at the edges of the 

 preparation it must eventually reach a position where there is a state 

 of equilibrium between the oxygen used up by the flagellates and 

 that diffusing inwards from the edges. At this point the band will 

 remain stationary. 



That this inward diffusion of oxygen is really the factor which 

 controls the distance of the band of flagellates from the edge can be 

 shown very simply as follows. A preparation with a Bodo band which 

 has become stationary is placed in a gas chamber on the stage of a 

 microscope and oxygen is passed through the chamber. Almost 

 immediately the flagellate band commences to retire inwards towards 

 the center of the liquid under the cover-slip. It gets gradually less 

 square and more circular until it forms a small ring and finally becomes 

 one mass of flagellates at the center of the preparation. When now a 

 stream of hydrogen is passed through the gas chamber in place of 

 the oxygen, the band reforms and slowly but continuously increases 

 in circumference. In the first case the oxygen dissolving in the 

 edges of the liquid and diffusing inwards causes the position of opti- 



