REGULATION IN BEHAVIOR 341 



exhausted. The animal is again swimming about in the light, and the 

 green chlorophyll bodies which it contains are producing a little oxygen 

 which the infusorian uses in its metabolic processes. Now it comes 

 again to a dark region. In the darkness the production of oxygen by 

 the green bodies ceases ; they no longer supply the metabolic processes 

 with this necessary factor. Now we find that the infusorian rejects the 

 darkness and turns in another direction. The white Paramecium can- 

 datum does not do this, nor do the colorless bacteria. Possessing no 

 chlorophyll, they receive no more oxygen in the light than in the 

 darkness, and they pass into darkness as readily as into light. But 

 many colored bacteria do reject the darkness. They require light in 

 certain other metabolic processes, in their assimilation of inorganic 

 compounds, and when they come to the boundary between light and 

 darkness, they return into the light. Most bacteria reject regions con- 

 taining no oxygen, as we have seen. But in certain bacteria, oxygen is 

 not required for the metabolic processes; on the contrary, it impedes 

 them. These bacteria reject regions containing oxygen, swimming back 

 into the light. In some cases among unicellular organisms the relation 

 of behavior to the metabolic processes is exceedingly precise. Thus, 

 Engelmann (1882 a) proved that in Bacterium (or Chromatium) photo- 

 metricum the ultra-red and the yellow-orange rays are those most favor- 

 able to the metabolic processes (assimilation of carbon dioxide, etc.). 

 When a microspectrum is thrown on these bacteria, they are found to 

 react in such a way as to collect in precisely the ultra-red and the yellow- 

 orange. The reaction consists in a change of behavior, a reversal of 

 movement, at the moment of passing from the ultra-red or the yel- 

 low-orange to any other part of the spectrum. At that same instant 

 the metabolic processes of course suffer interference. Bacteria are not 

 in nature subjected to pure spectral colors in bands, so that there has 

 been no opportunity for the production of this correspondence between 

 behavior and favorable conditions, through the natural selection of vary- 

 ing individuals. 



In all these cases the behavior depends upon the metabolic processes, 

 and is of such a character as to favor them. Throughout the present 

 volume we have found similar relations to hold for all sorts of organisms. 

 We find even that when the metabolic processes of a given individual 

 change, the behavior changes in a corresponding way. 



Why does the bacterium or infusorian change its behavior and shrink 

 back from the darkness or the region containing no oxgyen ? As a mat- 

 ter of fact, it needs the light or the oxygen in its metabolic processes, 

 and it does not shrink back from their absence unless it does need them. 

 But we have no reason to attribute to the bacterium anything like a 



