HYDROIDA 



one and the same species can be distinguished by different types of gonophores; thus the female 

 gonophore of Laomedca flexuosa Hincks is heteromedusoid, while the male gonophore is styloid. 

 This strongly defined sexnal dimorphism is most interesting, and the question is obvious whether the 

 case is a solitary one, or if in other species other types of gonophores, of those stated by Kiihn, 

 might perhaps be found united in one and the same species. On that account I have (1915) more 

 closely examined the development of the gonophores of, among others, the Tubularia indivisa Lin. 

 and the Tubularia regalis Boeck, two species in which a marked sexual dimorphism is found pro- 

 nounced in the external characters of the gonophores. The examinations resulted in the startling result 

 that the female gonophores of both species are eumedusoid, though not equally high in medusoid 

 organisation, whereas the male gonophores of both species are heteromedusoid and of entirely the 

 same organisation and development as the gonophores of the genus Lampra Bonnevie (1898) (comp. 

 Broch 1915). It is owing to the gonophores that iMmpra is separated from Coryrnorpha, the latter 

 genus producing free medusae or having gonophores eumedusoid. Acting on this principle of division, 

 as far as the species of the genus Tubularia are concerned, we should have to separate the male 

 Tubularia indivisa and Tubularia regalis as a new genus, while the female individuals belonging to 

 the same species would retain their places 1 . 



The outcome of the searching studies of the last years shows, indeed, that in proportion as a 

 greater number of species have been examined, the more evidently the characters of the gonophores 

 appear as criteria of species. The increasing knowledge of the gonophores makes ever clearer that 

 the gonophores alone cannot form the base of any division of genera, but at most serve as second- 

 arily corroborative. 



The free medusae show throughout a greater abundance of forms developed than the polyps, 

 which are more conservative. No doubt, the medusae present a series of phenomena of adaptation 

 and, accordingly, display several characters of convergency, to which the system atists are inclined to 

 attach a greater phylogenetic value than is their virtual due. Nowhere, I dare say, the adaptations 

 to habits of life and the accommodations to varying physical conditions play such a part as with the 

 pelagic organisms, to which the slightest variation of salinity and of temperature causes great 

 changes of viscosity of the surrounding water, decisive of the adaptation of their suspension organs. 

 Therefore, the hydroid systematist should not lay too much stress on the statements of the medusoid 

 specialist as to the systematic grouping of the free-swimming generation ; phylogenetically the char- 

 acters of the nurse generation are of the greatest interest. A similar state of things urges itself with 

 the sessile gonophores. It is a circumstance of vital importance to the maintainance of the species, 

 that the generative individuals are able to accommodate themselves quickly to the peculiar conditions 

 of life to which the species is subjected. What I observed in my treatise on the Stylasterida (1914), 

 is fully applicable (or even more so) to the case under consideration: "Owing to their conservatism in 

 development the polyps are of the most important phylogenetic interest. The gonophores, the genera- 

 tive individuals, on the other hand, might almost be said to be a playball in the hands of chance 

 biological conditions and thus phylogenetically have much less interest". 



1 The possibility of a fusion of four species is precluded, as in Tubularia. indivisa and in Tubularia regalis c? and 9 

 occur in the same colony. 



