FEEDING PARAMECIUM KNOWN BACTERIA 177 



entirely unprotected against deleterious bacteria if these pre- 

 dominate in infusions in which Paramecium is living. The ani- 

 mals may die from starvation in the midst of an abundance of 

 unavailable food, or from poisoning should these bacteria be 

 toxic. 



The lack of exact knowledge concerning the characteristics of 

 saprophytic bacteria is a great handicap in undertaking the study 

 of any bacterial mixture serving as food for Paramecium. Har- 

 gitt and Fray ('17) found it impossible to identify adequately the 

 organisms with which they worked. Miss Watt and I found the 

 same difficulty in the course of this investigation. I have in- 

 cluded cultural characteristics of the organisms used in these 

 experiments with the idea that they might be of some slight 

 help to anyone desiring to undertake similar work, but there is 

 great need for a thorough investigation of the saprophytic bac- 

 teria. Until such a study shall have been made, it is desirable 

 that any satisfactory cultures which may be discovered shall be 

 carefully kept and made available to any investigators who wish 

 to use them. The cultures A', B', and C are being maintained, 

 and subcultures will be given to those who desire them. 



For ordinary routine work, a satisfactory chance mixture main- 

 tained at a high state of effi.ciency is to be preferred to an artificial 

 mixture such as used in this investigation. The behavior of 

 animals fed with M during the first month amply demonstrates 

 this, for this mixture was superior to the artificial ones during 

 this period. However, if one is investigating the behavior of 

 Paramecium with great care, a known bacterial content is 

 necessary in infusions. Deleterious bacteria certainly lower the 

 rate of metabolism or cause an undue percentage of deaths and 

 so modify normal metabolism as to greatly influence the inter- 

 pretation of results. 



A varied diet is natural for most animals. Although a given 

 environment may be normal for a particular organism, it is not 

 necessarily one in which the optimum of metabolism can be 

 maintained. Variety of food seems to produce an optimum 

 metabolic rate in all cases investigated, largely because no one 

 type of food has been found which combines all the factors neces- 



