96 



IN VERTBBBA TB MORPHOLOG T 



medusan, then, it is evident that the medusa form is a secondary modifica- 

 tion — a specialized reproductive organ, which in the Narcomedusse and 

 Trachymedusee has become so important that the ancestral polyp form is 

 practically suppressed in the life-history. On this view it must be sup- 

 posed that organisms so similar as the medusai of the Tubularian and 

 Campanularian polyps have been developed entirely independently of one 

 another, a view which carries with it many difficulties, and that the medu- 

 soid buds represent stages in their evolution. 



It seems more probable, however, that, leaving the Hydrarise out of the 

 question, in all the other groups the medusa was the parent form. This 

 is borne out by the fact that in the Narcomedusse, which, with their broad 

 pouch-like extensions of the gastric cavity, are the most primitive of all 



the craspedote medusae, there is no fixed 

 polyp form. It has been shown, how- 

 ever, that the Narcomedusse and Trachy- 

 medusae in their development pass 

 through a stage which may be considered 

 to represent the polyp form, and if, while 

 in this form, non-sexual reproduction 

 should have taken place, the buds re- 

 sembling the immature form which gave 

 rise to them, a polyp colony would result, 

 some of the buds of which might con- 

 tinue their development and become 

 medusae. By this view the difficulties 

 presented by the similarity of the medusae 

 throughout all the groups where they 

 occur are overcome and the medu.soid 

 buds are regarded as imperfectly devel- 

 oped or degenerate medusae. Further- 

 more this view is rendered more tliau 

 probable by the development of Cunoe- 

 tantlia and the allied Cunina. The 

 former while in the embryonic polyp 

 form actually does bud (Fig. 49), the 

 buds resembling the original embryo which gave rise to them, and all the 

 buds, the parent embryo included, later develop into medusae. In Cunina^ 

 however, the parent embryo which gives rise to buds undergoes no further 

 development, only the buds continuing on their course of growth to medu- 

 sae. In this case a true and typical alternation of generations occure and 

 points out a simple explanation of the alternation which is found in the 

 AnthomedustB and Leptomedusae. In these the polyp colonies are the re- 

 sults of non-sexual reproduction of a larval medusa, and some only of the 

 individuals so formed continue their development to medusae. 



The relationships of the Hydrocorallinae to the other groups a;-e not yet 

 quite demonstrated. It would seem, however, from the medusa-bud which 



Fig. 49. — Budding Larva of 



Gunoatantlia ocionaria (after 

 MoCrkadt, from Brooks). 

 a = egg larva. 

 lb = budded larvae. 



