104 THE OCEANIC HYDROZOA. 



The polyjntes (fig. A) attain a length of three quarters of an inch or more. Like the 

 tentacles they exhibit great irritability and contractility, and are in constant motion, at 

 the same time assuming every variety of form from that of a cylinder to that of a pear, 

 or that of a wide-mouthed and narrow-necked flask. The latter is the shape most commonly 

 assumed. 



The lip is covered with many thread-cells, and is usually colourless or yellowish white ; 

 the dilated body of the organ is blue, and appears, even to the naked eye, to be dotted 

 over with numerous dark spots i}?^). The ectoderm of the polypite is thin and dense. The 

 endoderm, thicker, occasionally exhibits an appearance as if it were longitudinally fibrillated, 

 and its inner surface is covered with long and beautiful cilia, which are especially obvious 

 about the oral aperture. The ectoderm and endoderm are in close contact about the mouth 

 of the polypite, but elsewhere a narrov;, clear space exists between them ; and in the 

 enlarged, globular portion of the polypite this space was increased at intervals by the 

 endoderm becoming folded (fig. 6). The folds were not at regular intervals, nor, so far 

 as I could observe, did they ever extend completely round the circumference of the sac. 

 They were certainly not produced by any mere contraction of the endoderm, inasmuch 

 as the state of extension of the polypite had no influence whatever upon their existence. 

 The space left between the endoderm and ectoderm appeared (as might be expected) 

 greenish, and as if filled by a liquid, but no trace of globules or granules could be detected 

 in it. 



The dark spots which have been referred to above, arose from the presence of delicate 

 villous processes of the endoderm (figs. 5 — 7). These villi were found wherever the fold^ 

 existed ; and, whenever I succeeded in tracing a villus completely to its attachment, I found 

 its base to pass into one of these folds. The villi were especially numerous towards the 

 proximal end of the polypite. They were cylindrical, or slightly conical, but became almost 

 dendritic when compressed, in consequence of the thrusting asunder of the large globular 

 masses, of which their free ends appeared to be composed. These masses were but 

 indistinctly visible towards the base of the villus, but, towards its apex, they projected, 

 so as to give the extremity of the organ a slightly lobed appearance, and were filled 

 with densely packed, coloured granules. 



The surface of the villi was richly ciliated, and thread-cells were imbedded here and 

 there in their substance. The axis of the villus was traversed by a clear canal, continuous at 

 its basal end with the interspace between the endoderm and ectoderm. The apical 

 termination of this canal was obscured by the pigment of the villi. It would seem as 

 if the endoderm had a distinct basement membrane, for, in two instances, where the soft 

 investment of the villi had been torn asunder and compressed, they remained connected with 

 the endoderm of the polypite by a delicate membranous tube of the same diameter as 

 the central canal. 



The greater part of the surface of the cocnosarc is covered by a bluish-green, granular 

 mass, mixed with small polypites and numerous tentacles of the smaller sort (figs. 13, 13ff, 13i). 

 On microscopic examination, it is seen that the granular mass is formed by an infinity 

 of small appendages, attached to the extremity of ramified processes of the hydrosoma, whose 

 internal cavities are, of course, continuous with that of the hydrosoma. 



Most of these appendages are imperfect, and are to be found in every stage of 



