314 tukbinoliad^e. 



Mr. E. Q. Couch, however, says, " In the youngest state 

 the animal is naked, and measures about the fifteenth of an 

 inch in diameter, and about the thirty-second of an inch in 

 height. In the earliest state in which I have seen the 

 calcareous polypidom there were four small rays, which 

 were free or unconnected [i.e. without any wall] down to 

 the base ; in others I have noticed six primary rays, but 

 in every case they were unconnected with each other. 

 Other rays soon make their appearance between those first 

 formed ; they are mere calcareous specks at first, but after- 

 wards increase in size. The first union of the rays is 

 observed as a small calcareous rim at the base of the 

 polype, which afterwards increases both in height and 

 diameter with the age of the animal."* 



From a valuable series of observations made by Mrs. 

 Thynne, j~ it would appear that the Caryophyllia discharges 

 its ova in spring, which in about two days become rotating 

 infusorioid animalcules. In a week or two these affix 

 themselves, and develop tentacles and a disk, and gradually 

 grow to the size, and even far more than the size, of the 

 parent, with all the characteristic colours and marks, hut 

 without the least trace of a corallum. During the progress 

 of this condition, the individuals increase rapidly by 

 spontaneous fission, the separated portions immediately 

 becoming independent animals. It is difficult to suggest 

 any flaw in the evidence of identity ; but it is to be 

 regretted that the experiments terminated without any 

 sign of the development of a corallum. 



Double and even triple specimens are not uncommon ; 

 and I have seen at least two examples (one of which I now 

 possess) that are fourfold. J The appearance of such speci- 

 mens is exactly that of a branching coral ; and, strange to 



* Quoted in Johnston's Br. Zooph. i. 199. 



t Ann. N. H. for June, 1859. 



+ Such a specimen I have figured in my Dev. Coast, pi. v. fig. 5. 



