HYDROIDA 



49 



there are iu the colonies also mnnerous blastostyles, which arc wholly devoid of rudiincuts of tentacles, 

 and accordingly only forming a stalk in the same way as is indicated by the diagnosis of Bonnevie. 

 Bnt there always occnrs a small polyp, and a "complete atrophy" of the polyj) is in no case demon- 

 strable; nor is it obvions from the imperfect figure of Bonnevie (1899, Tab. I, Fig. 3). — The great 

 variation of the blastostyles of the original specimens determined b\- liergh, shows us that the 

 species stated by Bonnevie may be founded on a colony where the blastostyles carrying tentacles 

 were reduced to a minimum in number. But this does not give sufficient rea.son for nuiintaining it 

 as a peculiar species beside Ilydnictiiiia carica. 



Hydractinia carica is an Arctic litoral species, which, in boreal waters, proves able to penetrate 

 along the West coast of Norway as far as Bergen (Text-fig. M). It has been found elsewhere only 

 in high arctic regions, in the Kara vSea, in the Mnrman Sea, and at Spitzbergen. It has also been re- 

 corded by J;iderholni (1909) from Davis Strait. 



Gen. Bougainvillia Lesson. 



Upright colonies with branched hydrocanli clothed with a perisarc. The polyps are fusi- 

 form with the tentacles placed in a whorl below the conically pointed oral portion. The polyps are 

 naked without the slightest attempt at formation of pseudohydrotheca round their proximal portion. 

 The gonophores are placed on the stems of polyps fully developed or reduced (blastostyles). In the 

 latter case, the unbranched polyp stem will sometimes rise from the hydrorhiza instead of from the 

 h\-drocaulus. 



A closer inquiry into the distinguishing characters and their systematic value, shows us that there 

 is no reason to distribute the species of Boiigai)ivillia on the three genera. Boiigai?ri'il/ia^Dicoryu(\a.\u\ Hctcro- 

 cordylc. Sufficient cause for distinguishing between Bougainvillia and the two other genera is not at all 

 given by the fact that the former has medusoid gonophores, while the two other genera have st\-loid go- 

 nophores. In mentioning the genera earlier treated, I have sufficiently explained the insignificance of tiiis 

 criterion as to classification. Then renuains the other distinguishing mark that the gonophores of Bougain- 

 villia develop on the stems of some fullgrown pohps, while in Dicoryiir and Hcterocordylc they are devel- 

 oped on the stems of reduced polyps (blastostyles). The genus Hydractinia show's us a full parallel to this 

 condition of things. But becau.se of the many and close connecting links between one extreme and 

 the other, it has been agreed that the stronger and weaker .specializing into nourishing jiolyps and 

 blastostyles cannot be employed as a generic character. When, on the other hand, this has been done 

 with Bougainvillia and Dicorync ami f[ctcrocordyh\ the only reason must be that, between the few 

 species known of these genera, nuxst of the links occurring in Hydractinia are wanting. This lack, 

 in fact, is not a sufficient i-eason to elevate, in one case, the character to an importance which is denied 

 in another case, even within the same family. It is not right to base the genera Bougainvillia, Di- 

 corync and Ifcfrrocordylc on characters that must be used with discretion as specific characters 

 in Hydractinia. Further, as to the distinguishing nuuk between Dicorync and Hcterocordylc, it is 

 still more diminutive. Hcterocordylc conybcari A 1 1 m a n , tlie only species known of the genu.s, is so like 



The Ingolf-ExpcJilion. V. 6. / 



