June 15, 1915 Spongospora Subierranea 271 



were previously occupied by the infecting plasmodium, are quite clearly 

 shown in this illustration. 



The cells produced by divisions of the giant cells are all infected. 

 They are approximately the size of the normal potato cells and seem to 

 grow rather slowly. So far as has been observed, these cells rarely 

 divide. Half a dozen, or even more, infected cells may result from one 

 cell originally infected. The writer has never seen any indication that 

 a Plasmodium can pass from a growing infected cell into a haalthy one. 



Within the host cell the plasmodium is closely applied to the host 

 nucleus,, as shown in Plate XXXIX, figure 6. The host nucleus seems 

 to be as thoroughly embedded in the plasmodium as are its owii nuclei. 

 Whether or not the fungus cytoplasm is in direct contact with the 

 membrane of the host nucleus is difficult to determine, but, so far as 

 appearance goes, this seems to be the case. It is interesting to see that 

 the intercellular plasmodia do not kill the host cells, but merely stimulate 

 them to increased growth and division. This relation indicates that 

 S. subierranea is a rather highly specialized parasite. 



Sometimes the host nucleus becomes much lobed and distorted ; it often 

 contains several nucleoli. The chromatin strands become abnormal in 

 appearance or may even entirely disappear. In some instances, however, 

 the host nucleus remains intact even after spore formation. It can then 

 be seen embedded in the spore ball. The relation between the plasmo- 

 dium and the host nucleus is an interesting subject, but the detailed 

 description of the appearance of the diseased host cells will be left for 

 some future time. 



The Plasmodium within the host cell is irregular in shape, as shown 

 in Plate XXXIX, figure 6. The nuclei are rather evenly distributed 

 and stain very readily. Each contains a large nucleolus. The nuclear 

 membrane stands out clearly, and chromatin strands can be seen. The 

 writer is of the opinion that these nuclei do not divide during the early 

 stages of infection, but this point needs further study. Nuclear divisions 

 in the plasmodium seem most common just after the giant host cell has 

 divided into smaller ones. So far as has been observed, all the nuclei in 

 a given plasmodium divide simultaneously. Mitotic divisions are the 

 only kind that have been observed. The equatorial plate stage is shown 

 in Plate XXXIX, figure 7. This is the stage most commonly met with 

 in the preparations, but later stages are also abundant. Spindle fibers 

 are clearly seen, but no astral rays or centrosomes have been observed. 

 The writer agrees with Osborn (14) that nuclear division immediately 

 precedes spore formation, but whether or not there are two successive 

 divisions he is not prepared to say. In Plate XXXIX, figure 8, is shown 

 a small portion of a plasmodium very highly magnified. A dense region 

 can be seen around each nucleus. Appearances like this probably explain 

 how Osborn (14) and others have been led to believe that the cells were 

 infected by uninucleate amoebae. It has generally been supposed that in 



