WOHOMN \< KAK 



13 



gena. According to Fischer and Miss Waterhouse 

 the aoospores come to real on the host hyphae and 

 develop germ tubes of variable lengths (tigs. S 7, 

 16 which penetrate the host wall. The contents flow 

 into the fungus hyphae through this tube and soon 

 become obscured by and almost completely lost to 



sight in the host protoplasm. Although it is not visi- 

 ble as a clearly-defined body its presence is never- 

 theless evident by the increased density and opacity 

 of the host protoplasm in the region of infection. 

 Fischer reported that the young thallus of A'. septi- 

 gena loses all individuality as it mixes with the host 

 protoplasm and develops into a plasmodium. hut his 

 account is not based on observations of fixed and 

 stained material. It is not improbable that the para- 

 site remains naked until very late in development, 

 hut it is apparently immiscible with the host proto- 

 plasm. If it is amoeboid in shape with numerous fine 

 pseudopods it may well infiltrate the interstices of 

 the host protoplasm and appear to be fused or mixed 

 with the latter. This, however, remains to be deter- 

 mined by intensive cytologic a] study of fixed and 

 stained material. 



As the parasite increases in size and attains vege- 

 tative maturity, numerous small vacuoles usually 

 appear in the cytoplasm of the hypertrophied por- 

 tions of the host hyphae (figs. 9, 16), but it is not 

 always certain whether these vacuoles relate to the 

 cytoplasm of the parasite or the host. However. 

 since they seem to fuse later and form the large 

 central vacuole of the parasite's sporangium (fig. 

 10) they probably relate to the parasite. Such vacu- 

 oles may move about and undergo marked changes 

 in shape from time to time (fig. <)). According to 

 Butler, the vacuolate stage is not very marked in R. 

 inflate. At this stage one or more dome-shaped exit 

 papillae are usually present which project through 

 the host wall (figs. 9, 10). Their presence is prob- 

 ably an indication that the parasite has reached 

 vegetative maturity and been transformed into an 

 incipient sporangium. The formation of the spo- 

 rangium wall has never been observed, but it has 

 been described in the literature as indistinguishable 

 from and seemingly fused with that of the host. 

 However, by plasmolytic experiments Miss Water- 

 house demonstrated that R. waterhouseii has a dis- 

 tinct membrane of its own which may be readily 

 separated from the host wall. Whether or not it is 

 composed largely of cellulose had not been deter- 

 mined. 



Cytokinesis is probably accomplished by centri- 

 fugal cleavage furrows which progress from the 

 border of the central vacuole to the periphery, al- 

 though it has not yet been clearly demonstrated. Ac- 

 cording to Miss Waterhouse, the risible changes in 

 the protoplasm preparatory to sporogenesis .ire quite 



characteristic. Tin- central vacuole may disappear, 



leaving the protoplasm quite clear and homogeneous 



except for a few dark granules in tin- center. Shortly 



thereafter the protoplasm takes on the appearance 



as if it had undergone cleavage into zoospore initials, 

 but this phase persists only for a short time. The 



protoplasm becomes optically clear again, and after 

 a period of about half an hour tin- exit papillae <h 



liquesce and disappear completely, having a cyto- 

 plasmic membrane across the orifice, following this 

 stage the protoplasm becomes finely reticulate in :i|p 

 pearance, and shortly thereafter the definitive spore 

 initials are formed (fig. 11). As these become more 

 clearly defined the whole mass of segments begins 

 to glide and nunc around. This movement increases 

 in intensity until the zoospores are swarming and 

 swirling around in the sporangium. Within a few 

 minutes the membrane across the orifice bursts and 

 the zoospores arc discharged in a dense cloud-like 

 mass. 



All species of liozellopsh are reported to have 

 heterocont zoospores with the shorter flagellum ex- 

 tending forward in swimming (figs. f. 18, 15). The 

 exact point of insertion of the flagclla is not certain 

 in R. waterkouseii, but in R. in flata and R. septigena 

 Fischer and Butler reported that the short flagellum 

 arises from the anterior end (fig. 15) while tin- 

 longer one is laterally inserted. In R. simulant, on 

 the other hand, both flagclla arise from the anterior 

 end (fig. 23) as in the Plasmodiophorales, according 

 to Tokunaga. In swimming the zoospores move along 

 more evenly and smoothly in a straight or curved 

 path in marked contrast to the jerky darting motion 

 of the zoospores of Rozella species. They may also 

 round up and encyst (fig. 14), but so far no evidence 

 of diplanetism has been observed. 



The development of septigenous species is similar 

 to that described above for aseptigenous members, 

 with the exception that the thallus produces more 

 than one sporangium or resting spore. Fischer de- 

 scribed the thallus in R. septigena and R. simulans 

 as a plasmodium which fragments or undergoes 

 schizogony into several segments. Although he fre- 

 quently noted instances of multiple infection ( fig. 

 10) Fischer nonetheless believed that each plasmo- 

 dium is the product of a single infection. As evidence 

 that the thallus fragments, he reported cases of sin- 

 gle spores giving rise to 5, 2, t. and 7 sporangia. In 

 cases of multiple infection the resultant thai li or 

 Plasmodia remain separate and do not fuse, accord- 

 ing to Fischer. As noted elsewhere his studies do not 



relate to experimentally controlled monospore in- 

 fections, and until such studies have been made the 

 problem of whether or not a single infection eventu- 

 ally gives rise to several sporangia or resting spores 

 remains to be conclusively settled. However formed, 

 the thallus or plasmodium is reported to fragment 

 and the portions become separated by transverse 

 host walls and mature in basipetal succession (figs. 

 17.22). 



The process of resting spore formation in the 

 septigenous species appears to be the same as that 

 of Rosella, although it must be noted in this connec- 

 tion that Fischer's figures (figs. 18-21 ) may possi 

 bly relate to Eozella instead of Rozelloptit. Assum- 

 ing that tin- stages shown in figures IK to 21 belong 

 to the parasite with billagel late heterocont zoospores 

 (fig. 15), it is evident that the process is strikingly 



