172 



Transactions of the Society, 



arrived at from these experiments, and that is, that in reckoning 

 up the numher of filaments spuu by Diadema, and going to make 

 up the cable, the count always came below 200. Now this would 

 appear quite insufficient as a product of the spinning-tubes, which 

 in a fine preparation I have of the spinnerets by Bourgogne, 

 certainly exceed 1000 in number ; so it would seem we have to 

 fall back upon the conclusion, that either all the spinnerets are not 

 in action at the same time, or, that a consideraljle proportion of the 

 spinning-tubes, which have apparently some differences in structure, 

 have also a diff'erent function to perform, such as cross lines of the 

 geometric web and " bead globules." This is mere surmise, but 

 the fact of the small number of ultimate filaments in relation to 

 spinning -tubes remains. 



A description of the mechanism of the spinning apparatus 

 would make this paper too long. It would form the basis of a 

 future communication, or the materials are at the service of any 

 microscopist who wishes to work out the subject. 



