114 University of California Publications in Zoology [Vol.19 



From table 2 we see that the greater number of these cysts were 

 found in the small intestine, and in one series they were found in 

 the rectum. This was interpreted to mean that encystment com- 

 menced in the small intestine (duodenum) and the cysts were in the 

 rectum because they had been carried there by the peristaltic move- 

 ment of the bowels. 



Encystment, it appears, begins with the formation of a wall around 

 a single individual, and the immediately subsequent stages are sim- 

 ilar to those in binary fission preceding plasmotomy of the two-zooid 

 somatella. Many of the stages following the first division of the nuclei 

 were not found. Some of the cysts showed twelve nuclei (pi. 1, fig. 9) 

 due to the fact that some of the nuclei have not divided the third time. 

 In some cysts (pi. 1, fig. 10) two of the nuclei were seen to be dividing 

 for their last time (third division). No plasmotomy was seen to occur 

 within the eyst, but from the evidence given by Kofoid and Christian- 

 sen (1915) it is more probable that plasmotomy takes place outside the 

 cysts after the digestion of the'wall of the cysts in the small intestine. 



This method of reproduction within the cysts does not seem to be the 

 common method employed by this flagellate, for the cysts were detected 

 in only two cases in my preparations of G. microti from the meadow 

 mouse and were never detected in some two hundred examinations 

 of cysts of a species of Giardia found in the rat. It is a method 

 which results in a greater proliferation of the individuals when com- 

 pared with binary fission. In the rat the binary cysts were found in 

 the faeces at regular intervals, showing that encystment was cyclic 

 and not continuous or sporadic in nature. But since the multiple 

 cysts were not found with such regularity it appears that this method 

 of reproduction does not follow at regular intervals unless its interval 

 of recurrence is much longer than that found for binary fission. It 

 may be, then, that disturbance of the normal environment causes an 

 increased fecundity on the part of the flagellates which results in the 

 method of multiple fission for its expression. The possibility of 

 multiple fission following the formation of a zygote by syngamy is not 

 excluded. 



These cysts when ingested would have their cyst walls digested 

 away as in the case of the binary cysts. Since each one of the multi- 

 nucleate cysts contained a somatella of sixteen nuclei an eight-zooid 

 somatella would be liberated to continue its plasmotomy in a free 

 state. Cysts showing part of the cyst wall absent were figured by 

 Kofoid and Christiansen (1915, pi. 7, figs. 42, 53) and these cysts 



