534 H. S. Jennings and George T. Hargitt. 



which had merely been warmed to 50° Cent., in order to destroy 

 any foreign Paramecia it might contain. This infusion was used 

 quite fresh, and was made anew each day. The animals do not 

 flourish so well if the fluid is allowed to become twenty-four 

 hours old before use. From March 29 on the infusion was em- 

 ployed at but one-half the concentration mentioned; that is, 

 200 cc. of water was added to the gram of hay, in place of 100 cc. 

 This ''i^-standard" infusion was found to be on the whole better. 

 Every day the number of fissions during the previous twenty- 

 four hours was recorded, and a single specimen was transferred 

 with a capillary pipette to three drops of the fresh infusion on a 

 clean slide. In order that there might be little danger of losing 

 the series by death of a single individual, from each of the original 

 specimens two lines of propagation were carried on, and in a part 

 of the experiment four. Thus from every race there were at all 

 times at least four parallel lines of propagation in progress, and 

 during a part of the experiment there were at least eight lines for 

 each race. In the case of several of the races the number of parallel 

 experiments carried through was considerably greater than this, 

 and the experiments were continued much longer. The slides 

 with the different lines were kept side by side in the same moist 

 chambers so that the conditions were all identical for the differ- 

 ent races. 



The experiment showed that there are certain characteristic 

 differences in the rate of fission among the different races. ^ The 

 characteristic rate of fission for the different races, in the case of 

 healthy lines that have not recently conjugated (though by no 

 means in a state of "depression") are shown for the entire period 

 of the experiment and for the successive weeks in table 5. The 

 table is based, save in one or two cases specified, on lines that lived 

 throughout the experiment. In each case lines from two indi- 

 viduals at the beginning are employed, and so far as possible two 

 lines from each of these individuals are followed, so that data for 

 four lines for each race appear in this record. In cases where 



9 The results as to other influences affecting the rate of fission, and particularly 

 as to differentiation in this matter within the limits of a single race, will be dealt 

 with in another paper. 



