86 AMOEBAE LIVING IN MAN 



often occur in abundance, whilst the intermediate 4-nucleate stage may 

 be rare or even absent. This is probably due to the fact that the cyst 

 rests for a long time in the binucleate stage, and then rapidly completes 

 the two further nuclear divisions. There is probably a very brief resting 

 period at the 4-nucleate stage. This is also indicated by the fact that 

 the majority of 4-nucleate cysts of E. colt, if carefully examined, show 

 one or more of their nuclei preparing to divide or actually in the form of 

 a spindle (cf. fig. 61, PI. IV). This is often another very useful character 

 for discriminating between the 4-nucleate stages of the two species. 



It should be noted that the size of the nucleus in the uninucleate 

 cyst of E. coli is very variable. At the beginning of development 

 (fig. 56, PI. IV) it is comparatively small, resembling the nucleus in the 

 precystic amoeba (fig. 13, PI. I). For the first nuclear division it then 

 increases in size enormously (fig. 57, PI. IV), finally forming a typical 

 spindle (fig. 58, PI. IV). The subsequent nuclear divisions are all 

 similar, and resemble those of E. histolytica. Sometimes definite 

 chromosomes appear to be present, however, as in the nuclear spindle 

 shown in fig. 58 (PI. IV) : but I have not yet been able to satisfy myself 

 of their invariable occurrence at all stages. I hope to return to this 

 question later, when I have been able to examine more preparations. 

 I would merely note here that, as in E. histolytica, the dividing nuclei in 

 cysts discharged in the stools very soon become abnormal; and unless 

 the cysts are fixed immediately, they show most misleading and abnormal 

 nuclear figures. All the divisions appear to be accomplished in the 

 same manner — the nuclear spindles, etc., in succeeding stages differing 

 only in their size. Division of the nuclei is not always synchronous, 

 and this occasionally leads to the production of cysts containing — 

 probably only temporarily, in ordinary circumstances — odd numbers of 

 nuclei, such as 3, 5, or 7. Cysts with 6 nuclei may also be found some- 

 times. Supernucleate cysts also occur from time to time. Instead of 

 the nuclei entering into a permanent resting stage in the 8-nucleate cyst, 

 one or more may again divide. In this manner cysts containing any 

 number of nuclei from 9 to 16 may be formed. Fig. 67 (PI. IV) shows 

 a i6-nucleate cyst which is, as is generally the case, of unusually large 

 size. Such cysts are not common, and do not represent normal stages 

 in development, in my opinion.* They are, I think, supernucleate 

 " freaks," like the 8-nucleate cysts of E. nana or the binucleate cysts of 

 /. biltschlii, which are both equally rare. I have very rarely found cysts 

 of E. coli containing more than 16 nuclei. I have seen cysts — examined 

 in iodine solution — containing 18 nuclei, on several occasions; and 

 I have once found one containing more — at least 20. The nuclei, which 

 lie at different levels, are very difficult to count with certainty in such 

 cysts, and the only safe way is to draw them very carefully with the 

 camera lucida.f 



Nuclei of difterent sizes are sometimes seen in the cysts of E. coli. 

 I do not attach any importance to these, but regard them as abnormali- 

 ties. They are, on the whole, uncommon : but occasionally an infected 



* Cf. Mathis and Mercier (1917 d), who regard them as " schizogonic " cysts. 



t As an instance of the ease with which mistakes can be made, I may mention that 

 I began to draw the cyst shown in fig. 67, PI. IV in the belief that it contained 12 nuclei. 

 It was only after every nucleus had been outlined with great care with the camera lucida 

 that I realized my error. 



