Gr. H. F. Nuttall 
343 
(there were several more besides) is explained by the tendency infected 
corpuscles have to become aggregated into groups. In parasite A the 
commencement of the “budding” process is seen at 12 minutes, and, 
after the expiration of 85 minutes the parasites are almost completely 
formed but remain joined. In parasite B the first drawing was made 
at 14 minutes (at a later stage than the first stages of A) and, after 69 
minutes, we see two clearly separated pyriform parasites lying with their 
points crossed within the corpuscle. Whereas, in the early stages of 
division, the “ buds ” protruded laterally in parasite A, they protruded 
at one end of the oblong parasites C and D. The latter was observed at 
36° C. in a sample of blood taken from Control Cow I (this vol. p. 244) 
on the 8th day after inoculation. The blood was examined on two 
occasions on the same day at times when the number of infected cor¬ 
puscles amounted to 32 °/ 0 and 1’2 °/ 0 respectively, and the corresponding 
proportions of dividing forms were 2 °/ 0 and 3%. Parasite D rotated 
upon its long axis as division proceeded, consequently the two parasites 
which were being formed appear slightly superposed. But for the fact 
that the process of multiplication is slower, it appears in all respects 
identical with what we have observed in P. canis. 
Figures E to G represent the character of the amoeboid motion, 
which is usually not at all marked. Parasite E was observed in the 
blood of Cow X (p. 242) at a time when •8°/ 0 of the corpuscles were 
infected, that being the maximum percentage of infection attained. 
Parasite F was observed on the succeeding day in the blood of the same 
animal when the number of infected corpuscles had fallen to 01 °/o- 
Parasite G was sketched at the same time as A and B; it showed more 
marked pseudopodia than usual, but the movements were not to be 
compared in point of rapidity with those of P. canis. 
22—2 
