PETERS. — METABOLISM AND DIVISION IN PROTOZOA. 453 



became very old, so that it was reduced to the condition of a decaying 

 cellulose medium, it frequently produced Stentors. Likewise when the 

 amount of hay put into the above salt solution was very sparing a tran- 

 sient culture might result. A series of chemical determinations was 

 made upon hay infusion and upon otlier media for comparison. The 

 results showed that hay furnished so large an amount of fermentable matter 

 as to produce too much acidity in a medium intended for Stentor. This 

 fact also explains tlie observations upon the age of a hay infusion referred 

 to above, and the amount of hay that may permit a growth of Stentors. 



Some operations will now be described that, are useful for improving 

 poor cultures or for reviving old ones. Some account will also be given 

 of chemical estimations by the volumetric method that were made in 

 order to compare cultures in different stages. The latter observations 

 were found to furnish a good index of the conditions prevailing in the 

 culture liquid. Only the estimation of acidity, the most useful of these 

 determinations, will be described. Operating upon 5 cc. of the culture 

 liquid I determined the amount of 0.01 normal acid (hydrochloric) or 

 alkali (sodic hydrate or calcic hydrate) required to produce reaction with 

 methyl orange and phenol thalein respectively. The results were com- 

 pared and other estimations also were made. By making a series of 

 observations upon a single culture from the time of setting it, some idea 

 was gained of its progressive chemical changes. Experience showed that 

 the titration with phenolthalein was the most instructive and the only one 

 necessary. This indicator has the advantage for our purpose of being 

 useful in the titration of carbonic acid and organic acids. These come 

 especially into play in fermenting cultures. lu these estimations it was 

 imjiortant not to deprive the culture of much liquid, as refilling it dis- 

 turbed the naturally prevailing conditions. Yet repeated tests were re- 

 quired, and so for this reason the small quantity of 5 cc. was invariably 

 adopted for a test. In the titration of carbonic acid with 0.01 normal 

 alkali (usually sodic hydrate) much difficulty was at first experienced 

 owing to the absorption of this gas from the atmosphere. Hence con- 

 stant results could not be obtained in successive estimations of the same 

 liquid. The following procedure was devised to overcome the difficulty. 

 The liquid (5 cc.) to be titrated was transferred with a pipette from the 

 culture jar to a short thick-walled test-tube about one cm, in diameter. 

 A few cubic centimeters of kerosene were then poured in to cover the 

 liquid and so prevent absorption of CO^ by the alkali introduced for titra- 

 tion. Phenolthalein was added. A glass rod was also kept in the tube 

 for stirring. The titration was performed over a white tile. If neces- 



