222 REP(JKT ON THK PKNNATULIDA. 



iu some points of detail. We are in inucli doubt concerning the 

 function of these canals. Kolliker says they are to be regarded as a 

 modification of the nutrient canals, and possibly subserving some 

 special function. The epithelium lining them has a very glandular 

 appearance, and, bearing in mind their position at the points of com- 

 munication between, on the one hand, the fine canal system which 

 penetrates the mesoderm in all directions, and is iu communication 

 with the polype cavities, and, on the other hand, the main canal 

 system of the rachis and stalk, it has occurred to us that they may 

 very possibly be excretory organs and act as kidneys, separating effete 

 matters from the fluid in the fine nutrient canals, and discharging it 

 into the main canal system. This view derives some slight support 

 from the fact that in more than one case we have seen small collections 

 of debris over the orifices from the radial canals into the main canal, 

 which were apparently being discharged from the foi-mer into the latter. 



The chief difficulty in assigning this or indeed any other important 

 function to this system of canals, lies in the fact that they are found only 

 in certain members of the Pennatulida. They are present in Virgularia 

 and HaUxceptrum ; but Poinatida and Funiculina have no trace of them. 

 They can have nothing to do with the ova, for they are far too small 

 to admit them ; neither, so far as our observations go, do ova ever 

 occur iu the main I'adial canal, though, as we have seen, they do pass 

 into the lateral canals. 



2. — Tlie Stem. — 



The stem or calcareous axis of the rachis and stalk (Figs. 2, 3, 4, 5, 

 and 6, c), is cylindrical, firmly calcified, and brittle. According to 

 Dalyell it contains as much as 85 per cent, of mineral matter, chiefly 

 carbonate and phosphate of lime, and only 15 per cent, of animal 

 matter. 



Not only does the stem of Vir(]ularia differ from that of Pennatula 

 or Funiculina m its greater brittleness, but the proportions at various 

 parts of its length are also very different. Both in Fennntula and 

 Funiculina the stem is thickest at or just above the junction of the stalk 

 and rachis, from which pomt it tapers both upwai'ds and downwards, 

 ending at both ends in fine, imperfectly calcified, and very flexible points 

 {videVl. I., Fig. 2, and PI. III., Fig. 8). In Funiculina the stem extends 

 the whole length of the colony, while in Pennatula the stem reaches the 

 bottom of the stalk, but stops short some distance from the top of the 

 rachis. In Pemiatula it is also bent back on itself at both ends in the 

 form of a hook. 



In Virgularia the stem (Fig. 2) extends the whole length of the 

 colony. In the stalk, according to Dalyell, Kolliker, and Koren and 

 Danielssen, the stem tapers gradually downwards, ending in a fine 

 flexible point, which reaches to the bottom of the bulbous termination 

 of the stalk, and then turns back on itself for a short distance, ending 

 in a small hook, much as in Pennatula. In the rachis, starting 

 from below at its junction with the stalk, the stem at first enlarges 



