•508 Scientific Proceedings, Royal Dublin Society. 



-I have carefully compared them with numerous specimens of 21. 

 Murrayi (Quelch) in the British Museum. The conclusion I have 

 come to is that the specimens must be considered to belong- to 

 Quelch's species. The tips of the branches are, it is true, some- 

 what broader than in most of the type specimens, and the 

 ampullae are somewhat smaller ; but both of these characters may 

 be considered to be local variations, for some of the tips of the 

 branches of the type specimens are quite as broad as the broadest 

 of the tips of Professor Haddon's, and the diameter of the ampullae 

 varies considerably on the same branches of both. The interest of 

 the spirit specimens forwarded to me by Professor Haddon centred 

 in the examination of the contents of the ampullae. Q-uelch very 

 reasonably suggested that the ampullae were the receptacles for 

 large ova or embryos as in the Styladeridce, but the first series 

 of sections I made showed that Quelch had been misled. The 

 ampullae were found to contain very primitive but, nevertheless, 

 complete free swimming medusae, and the ova are small and 

 a,lecithal as in M. plicata. I found medusae in all stages of deve- 

 lopment, and was able to prove that they originate by an actual 

 metamorphosis of the dactylozooids, a condition that it is hardly 

 necessary to state has not been found in any other Ccelenterates. 

 This remarkable species of Millepora also shows that very rare 

 combination of a free medusa with a migration of the sexual cells 

 from the coenosarc into the medusa, a condition that is only known, 

 so far as I am aware, in Monobrachium parasiticum (Wagner). 



These, and many other facts that I was able to bring forward, 

 showed that the medusa of Millepora is not reduced or degenerate, 

 but primitive in its simplicity. The condition of the ova in M. 

 murrayi engaged my attention for some time, but I was unable 

 i;o discover any points either in their structure, mode of origin, 

 or development, that showed differences between this species and 

 If. plicata. 



I was able to make out nearly all the stages in the fragmenta- 

 tion of the germinal vesicle, and the subsequent arrangement of 

 ihe nuclear fragments that I have described in the other species. 

 In fact I was able to confirm all the statements and conclusions 

 of my previous paper. I have nothing to add to our previous 

 knowledge of the coenosarcal canals and polyps of Millep)ora. The 

 •different species of this genus seem to offer very few variations 



