BEHAVIOR 15 



pouch was such that pieces of stentor prey could be bitten off. 



As in many other cihates, the food vacuoles after they are 

 pinched off from the inner end of the gullet may be guided into 

 the interior by long fibrils dangling therefrom, first described in 

 Stentor by Schuberg (1890). But these could also serve a different 

 purpose. Andrews (1946) observed them in gullets everted by 

 pressure. When released the gullet can reinvert in only 10 

 minutes and this may be accomplished by traction of the fibers 

 in question. 



Perhaps a nice point of morphology is that the rim of the oral 

 pouch over which rejected particles are dumped is definitely 

 below the level of the frontal field and membranelles so that 

 rejects probably do not return to the oral stream. 



Gullet ciha and membranelles can work independently, for 

 Dierks noticed that ingestion may occur while the membranelles 

 circling the anterior end have for some reason stopped. 



Cannibalism has been observed in the three species of Stentor 

 most commonly cultured and may also occur in others. Ingestion 

 of its fellows by coeruleus was first reported by Johnson (1893) 

 and was the subject of a special study by Gelei (1925) who also 

 noticed cannibalism in roeseli; and Ivanic (1927) claimed that 

 cannibalism occurs in polymorphus. I recorded indubitable evidence 

 of cannibalism in all of the 9 stocks of coeruleus which I have 

 growing in my laboratory. To paraphrase Gelei, at least three 

 problems come to mind in regard to this peculiar food choice: 

 Why stentors come to eat each other, how they are able to ingest 

 such large objects, and what the consequences are for the can- 

 nibalizer, particularly whether it is able to digest its own species 

 of protoplasm. These topics will be considered in that order. 



Gelei noticed that hunger or the absence of other food organisms 

 is not the cause of cannibalism, as may be inferred from the fact 

 that stentors will ingest more of their fellows when they already 

 have one or more of these huge '* meals " in the process of digestion. 

 I found one coeruleus with five others inside. I also noticed can- 

 nibalism to be most frequent in cultures only a day or two fol- 

 lowing their nutrification. But neither is satiety the cause; for 

 cannibalism is found in starving samples, but not as frequently as 

 one might expect. These observations disprove Ivanic's contention 

 that cannibalizing stentors and other protozoa have a need and 



