1896.] BEITISH HIDEOIDS AND MBD08-E. 479 



taken, and found a few specimens agreeing with Agassiz's 

 description of Margellium gratum, both in the number of marginal 

 tentacles, and in the number of clusters of nematocysts on the 

 oral tentacles. I think this removes all doubt concerning the 

 ideutity of the two species. 



In the Valencia specimens the clusters of nematocysts develop 

 in the following order : — The earliest stage has each oral tentacle 

 terminating in a single cluster of nematocysts. A second cluster 

 appears near the first cluster upon a short stalk ; the tentacle then 

 appears bifurcated, each branch terminating in a round cluster of 

 nematocysts. Two more clusters, each on a short stalk, make 

 thoir appearance, one on oach side of the tentacle, about the middle 

 of its length. Tliis was usually the appearance of the oral tentacles 

 in all the large specimens taken at Port Erin and at Plymouth. 

 At Valencia, in a few of the largest specimens, with four or five 

 tentacles in each perradial group, I observed on each oral tentacle 

 a second pair of lateral clusters of nematocysts, below the first 

 pair. One specimen had a fifth cluster on a short stalk situated 

 midway between the two terminal clusters. This agrees wilh the 

 development of the clusters described by Agassiz in Margellium 

 grntum. Another specimen, liowever, showed a variation in 

 development, by possessing three terminal clusters and only a 

 single pair of lateral clusters. 



Allman (1859) has described and figured a Calyptoblastic hydroid, 

 Laomedea tenuis [ = Lejyiosci/phics tenuis, Hinclts (1868)], which he 

 found at Stromiiess. The hydroid has gonothecse each containing 

 a medusa. Allman has not given any description of the medusa 

 inside the gonotbeca, and from the figure it is impossible to 

 identify it, chiefly on account of its being at a very early stage in 

 development. Allman found inside the jar containing this hydroid 

 a number of young medusae which he believed to be closely related 

 to the genus Lizzia. Prom the description given of these medusae I 

 believe they are probably an early stage of Margellium octopunctatum. 

 Allman regards these free-swimming medusae as the medusa of the 

 hydroid in the jar. This observation has never been confirmed, 

 and if it be true, then a case is established in which a Calypto- 

 blastic hydroid produces Anthomedusse. 



Allman does not state that he has seen a single medusa leave tbe 

 hydroid nor show in any way that the medusa inside the gonotheca 

 resembles the free-swimming Lizzia. I do not think that there is 

 sufficient evidence to prove that the medusa came from the hydroid, 

 and I hesitate to accept the statement until the observations have 

 been confirmed. It is quite possible that the young Lizzia 

 entered the jar along with the sea-water. 



Distribution. America — Massachusetts Bay, Agassiz. Norway, 

 Sars. Prance — -Wimereux, Giard. 



Scotland — Shetland Islands, Forbes. St. Andrews, M'Intosh. 



England — Plymouth, Garstang; Allen. Powey, Peach, 

 Palmouth, Vallentin. Isle of Man, Browne. 



Ireland — Valencia Island, E, T. B, 



