•Prof. Allman on the Hydroida. 



mh 



ovum. With this change, the diagram of the alternation would 

 'stand as follows, where T represents the trophosome : — 



©••&c. 



&c. 



&c. 



&c. 



&c. 



Now many instances are known in which Medusae originating 

 as buds from a polypoid trophosome give rise by gemmation to 

 other Medusse, both sets of Medusse being also in all probability 

 capable of sexual reproduction. Sarsia, HybocodoUy Clavatella 

 may be cited as examples of this phenomenon; and the diagram 

 expressing it would exactly resemble that just given, with this 

 difference, that while the medusoid elements are, so far as we 

 yet know, similar in Sarsia, &c., they are heteromorphic in the 

 other. Fritz Miiller* has recorded a case in which an 8-tentacled 

 Cunina gave origin to buds which became developed into another 

 form of Cunina having its parts disposed in accordance with the 

 number 12, instead of 8. This would therefore be a case of true 

 heteromorphic budding. The bud-producing Medusse, however, 

 had not in this instance been traced to a polypoid trophosome. 



Though the phenomena as understood by Haeckel would thus 

 present nothing really at variance with the fundamental principle 

 of the law of alternation of generations, it must be admitted 

 that they are very exceptional. We cannot lose sight of the 

 anomalous fact that the medusoid elements in each period are 

 not only dissimilar, but, according to HaeckePs view, are both 

 properly sexual. A question, however, here suggests itself. Is 

 it necessary to adopt HaeckeFs interpretation of the phenomena ? 

 I believe not; on the contrary, I am strongly of opinion that 

 the Geryonia is not a sexual Medusa at all, and 1 believe that 

 the same may be asserted of the Cunina. Some years ago f, I 

 insisted on the non-sexuality of those gymnophthalmic Medusse 

 which, like Obelia, Eucope, Thaumantias, &c., carry their genera- 



* Wiegm. Arch. 1861. f Annals of Natural History, November 1859. 



32* 



