HYDROIDA II 73 



of the hydrotheca. All the sareothecae are adcaulinally split, and have a slightly dentate margin. The 

 hvdrotheca is broad, and lias a very slightly curved opening margin; its basal part has a somewhat 

 thicker wall than the distal two-thirds, and the boundary between the two portions is apparent as a 

 fine line abcladially directed obliquely forward towards the opening; hydrothecal ribs or septa lacking. 

 The internodinm shows two to four inner ribs at the adcladial side of the hydrotheca; one such, which 

 is three-branched, and markedly prominent, forms a boundary between the proximal sarcotheca and 

 the internodinm. 



The gonothecse' are situate on the stem at the apophyses. The young gonotheca shows distally 

 a peculiar oblique termination, the one side of which gradually projects further forward, suggesting 

 that the fully developed gonotheca would be furnished with a roof-like upper lip. Phylactogonia 

 lacking. 



Material : 



"Ingolf vSt. 127 66°33' N., 20°05' W., depth 44 fathoms, 5,6° 



Iceland: Tistil fjord, 66°43' N., i4°53' W., — 78 — 



Ritchie (1912 p. 226) refers this species to Cladocarpus, but with a query, as he did not find 

 gonangia in his specimens. These agree entirely with the present colonies (fig. XXXYI| which, from 

 the comparatively short, broad hydrothecae, can hardly be 

 confused with other species from northern waters. The col- 

 onies in question show that the species should be placed 

 under Halicornaria. The gonangia occur in several stages of 

 development, but none fully developed. The series of phases 

 represented in the colonies investigated are entirely parallel 

 to what we find in Cladocarpus integer ( G. O. Sars); the 

 oldest stage also seems to show that it developes, as does 

 the species mentioned, an "upper lip" which will dome out 



Fig XXXVI. Halicornaria campanulata from "In- 



more or less over the opening; further investigations must golf , st I2 „ Hydrothecate inte modium from a 

 determine, how far the development proceeds. hydrocladium in side and front view. (X 601. 



• Phylactogonia are altogether lacking in the colonies here concerned, and the species must 

 consequently be regarded as a Halicornaria; there is, however, the possibility that it may later prove 

 to be a primitive Cladocarpus. The otherwise close resemblance of the species to Cladocarpus integer 

 would also seem to point in the same direction; the last-named species has hitherto likewise been 

 regarded as a Halicornaria, but, as is further explained below, it is as a matter of fact a primitive 

 Cladocarpus, whose phylactogonia do not always or everywhere attain development. We cannot 

 therefore altogether disregard the possibility that Ha/icon/aria campanulata may also under normal 

 conditions develope more or less regularly occurring phylactogonia. But as long as this has not been 

 shown to be the case, the species must remain in the genus Halicornaria. 



Ritchie (1912 p. 227) could not state the locality of origin of the species nearer than "from 

 the neighbourhood of Iceland", which from a bio-geographical point of view is very meagre inform- 

 ation indeed. The localities now recorded are the more surprising, since we should a priori suppose 



The [ngolf-Expedition. V. 7. IO 



