THREAD-CAPSULES. 433 



24.) I had noticed that the tentacles of this species 

 were more than usually tenacious : whether this quality 

 may be owing to the barbed armature of the filiferous 

 capsules, I will not certainly say, but I think it not 

 unlikely. 



The small capsules are of a similar shape (fig. 23), 

 but are not more than j^th inch in length ; and they 

 jn'opel a thread, which, with the microscopic power 

 that I use, appears simple, to the length of -^th inch, 

 or six- times that of the capsule. (Fig. 25.) 



The tentacles do not, so far as I have seen, contain 

 any large capsules ; nor are the small ones present in 

 extraordinary number. The larger, j^th in. long, are 

 long- oval, a little curved, and permeated by a clear 

 cylindrical body throughout their centre. (Fig. 26.) 

 The smaller are more linear, f^th inch in length, 

 with a central evanescent line, that appeared zig- 

 zagged, or perhaps spiral (fig. 27) ; but the minute- 

 ness here precluded a satisfactory resolution. The 

 same reason, probably, prevented my seeing the struc- 

 ture of the emitted thread ; but the similarity of 

 appearance between the central column of the larger 

 tentacular capsules, and that of the larger ovarian 

 capsules, suggests a similarity of structure also. 



I cannot help thinking, from facts already recorded 

 in these pages, that the filaments which are so freely 

 shot forth by most Actinioi from pores scattered 

 over their whole surface, are neither seminal nor 

 ovarian ducts, but offensive weapons. In all cases 

 in which I have examined them, they are filled more or 

 less densely with filiferous capsules, and those the 

 most elaborately armed. Why should seminal or 

 P 2 



