170 K. KUDO 



fifteen to twenty turns in the capsule. Figure 71 shows the 

 extruded filaments which suggests that it was coiled seventeen 

 times, and at the same time demonstrates the fact that the polar 

 filament is coiled in the way as was suggested by the writer ('16) 

 in Nosema bombycis and also in Thelohania magna (text figure) , 

 but not in the way figured by Stempell ('09), who thought the 

 filament was coiled around a central axial portion, which con- 

 ception was supported by Zander ('11) and Strickland ('13). 

 One will notice the wavy course throughout the entire filament, 

 the height of the wave being smallest at the extremities. This 

 condition of the extruded filament agrees well with the oval 

 shape of the polar capsule and the suggested arrangement of the 

 filament in the capsule. 



Of all the known species of the genus Thelohania, twenty-three 

 in number, including T. magna, T. legeri Hesse is related most 

 closely to the present form. As was stated above, this species is 

 parasitic in the adipose cells of Anopheles maculipennis, and has 

 spores of similar structure. One may be inclined to think that 

 this form and the species recorded here are one and the same, 

 although the former was observed in France. Indeed, except 

 for the irregularity and shght difference in the dimensions of the 

 spores and the difference of the host species, the two species 

 would be distinguished from each other only with difficulty. 



The difference in the length of extruded filament has not the 

 importance in the identification of species of Microsporidia 

 which some authors have given it (for example, Strickland,' 13) 

 because there is very often a conspicuous variation even in one 

 and the same species. Besides, the different methods employed 

 for the study of the filament frequently bring out entirely dif- 

 ferent results. This is best demonstrated by the filament of the 

 spore of Nosema bombycis which have been studied by three 

 investigators, each using a different method. Thelohan ('94), 

 for the first time, proved the presence of the filament in a micro- 

 sporidian spore by treating spores of Nosema bombycis with 

 nitric acid, and recorded it as 12 to 14ai long. Stempell ('09), 

 on the other hand, used iodine alcohol, and found that the length 

 of the filament was 32 to 34^. Kudo ('13, '16, '18), by using 



