ON THE FEEDING HABITS OF AMEBA 569 



shortly after 3.15, with the carmme still inside the ameba, having 

 been ingested about four hours before. 



These experiments indicate that the endoplasm of at least some 

 strains of raptorial amebas differ strikingly from the endoplasm 

 of granular amebas in their sensitiveness to carmine. 



Summary of experiments with carmine. Carmine, whether 

 pure or mixed with a slight amount of egg white, is readily eaten 

 by amebas, especially by those of the granular type. Raptorial 

 amebas ingest carmine much less readily than the granular. • 



The granular amebas do not keep carmine in their bodies for 

 anything like the length of time food substances are kept, but it 

 is excreted in from three to fifteen minutes after eating it, the 

 average length of time being about seven minutes. The raptorial 

 amebas on the other hand may retain the ingested carmine for 

 four hours or more. The reason for this difference is not clear 

 but it is possible that the character of their usual food may have 

 something to do with it. The food of the granular type con- 

 sists mostly of soft bodied organisms, like flagellates and cili- 

 ates, which break up quickly after ingestion. The food of the 

 raptorial amebas, which kept carmine in their bodies for several 

 hours, consists of organisms with hard shells — diatoms and 

 desmids — which resist for a considerable time (over forty-four 

 minutes in the case of one diatom) the action of the digestive 

 juices. It is not impossible, therefore, that some rhythm or 

 habit may have become established in the protoplasm by many 

 and frequent repetitions, so that the ingested particles of different 

 composition tend to be treated as if they had the qualities (in 

 this case, slow of digestion) of the usual food substances, and con- 

 sequently kept for a similar length of time in the body. On a 

 purely chemical basis no explanation of this observation suggests 

 itself. 



The excretion of a particle of carmine may be delayed by the 

 ingestion of a food particle. 



A piece of carmine is eaten only once if the ameba is in a state 

 of mild hunger ; but the same piece may be eaten several times in 

 succession if the ameba is very hungry. The ameba reacts less 

 and less strongly to the same carmine grain if presented a num- 



THE JOURNAL OP EXPERIMENTAL ZOOLOGY, VOL. 20, NO. 4 



