STRUCTURE OF SBRIATOPORA, POCILLOPORA, ETC. 397 



tion to these zooids, as they had apparently been supposed 

 before he made his distinguished investigations to be water- 

 pores in communication with the canal system. He shows 

 that they are not of that nature, but describes them as 

 buds from which the large sexual polyps are developed, and 

 he actually watched the process of the growth of the one 

 form into the other in specimens which he kept living in 

 glass vessels. Singularly enough, however, I have been unable 

 to find intermediate forms on any of the specimens I have 

 examined between these minute polyps and the larger forms. I 

 find none of the small ones with traces of tentacles, and as far as 

 the very small quantity of material I have had at my disposal 

 goes, I have not been able to verify Prof. Lacaze Duthiers' 

 statement that the minute polyps are most abundant towards the 

 growing tips of the branches of the coral. 



I have failed to obtain a further supply of Corallium from 

 Naples this year, and therefore I publish this note in order that 

 attention may be drawn to the matter. Kolliker has described 

 in Haliseptrum and Virgularia the existence of certain unde- 

 veloped zooids which, occupying the lower and younger parts of 

 the stocks, produce the generative elements, and apparently as 

 growth proceeds become developed into fully-formed autozooids, 

 relinquishing at the same time their generative function. These 

 zooids have traces of the two dorsal mesenterial filaments only, 

 and otherwise, excepting in the possession of generative organs, 

 correspond to ordinary siphonozooids. The mode in which 

 they attain the remaining six pairs of filaments after ceasing 

 to perform the generative function is not known. The unde- 

 veloped zooids in Corallium rubnim are so abundant, and there are 

 so many of them constantly present, all in the same siphonozooid 

 stage, that we should apparently be justified in regarding this 

 Alcyonarian as polymorphic, like so many others composed of 

 two kinds of zooids, with this peculiarity, that any of the 

 siphonozooids are capable, like the sexual zooids of Haliseptrum, 

 of being developed into autozooids. 



A new form of Corallium (C. stylasteroides) , lately described 

 by Mr. Ridley^ of the British Museum, has much the appearance 

 of having distinct minute siphonozooids disposed regularly be- 

 tween the autozooids. Mr. Ridley agrees with me in considering 

 them as such. The occurrence of the two kinds of polyps amongst 

 Alcyonarians is much more common than was once supposed, 

 and is by no means confined to the Pennatulids. 



' " On tlie Arrangement of the Coralliidse, with Descriptions of New or 

 Rare Species," ' Proc. Zool. Soc.,' 1882, p. 225, PI. IX, fig. 2. 



