248 c. ISHIKAWA : 



is developed from the rest of the spindle-fibres of the last divi- 

 sion of the S2:)erm- cells, but the fate of the centrosphere after 

 division of the body is complete, had not been ascertained in 

 my previous researches. In order to determine this point, I 

 have made new observations upon numerous individuals in 

 various stages of division, which have led me to the conclusion 

 that a part of the centrosphere goes to form the tentacle. 



Fig. 2 represents a fresh Nociiluca that has just linislied 

 its division, but in which the two individuals are still connected 

 together by a rather broad bridge of cytoplasm, which in the 

 bodies ramifies in all directions into the usual pseudopodia-like 

 processes. The nucleus appears to l)e quite transparent, but 

 strong magnification shows in it numerous elongated chromo- 

 somes all situated nearer to the side of the nucleus on which 

 the centrosphere lies. The latter structure has, when in a fresh 

 condition, its usual granular appearance, and is easily recognis- 

 able. A part of the cell-wall lying above or very close to the 

 centrosphere is to be seen, slightly raised up in the form of a 

 small kno])-like protuberance, the contents of which have the 

 same granular apj)earance as the centrosphere, and this proves 

 to be the commencement of the tentacle. In specimens killed 

 with Flemming's solution and stained with iron-haematoxyliu 

 (Fig. 1), we observe under a high power, that the substance 

 filling the tentacle is exactly similar to that of the centrosphere 

 and is easily distinguishable from the surrounding cytoplasm, 

 strong evidence in itself that the formation of the tentacle oc- 

 curs at the cost of the centrosphere. 



A somewhat later stage in the formation of the tentacle is 

 represented in Fig. 3, in which will again be seen the same 

 relation existing between the tentacle and the centrosphere. 



