HYDROIDA 21 



from the "Michael Sars" 1902, when the species was found in the warm Atlantic waters to the south of the 

 Wyville-Thomson-Ridge (Broch 1903). There is in this case a possibility that the animal has been carried 

 with Arctic currents from northern regions; if so, the currents have conveyed the animal at an early 

 period of life to the new locality, where it has been able to subsist and develop further. The other 

 locality, in the Trondhjemfjord was previously recorded, and the Trondhjem Museum was in possession 

 of a defective specimen which was said to have been taken in the fjord ; but particulars were wanting 

 till the researches of the fjord in 1911 brought to light, from the depth of 200 m., near Tautra, and 

 thus in the midst of the Atlantic water layers of the fjord, a well developed female polyp. This find 

 gives the more surprise, because there is left no opening for the possibility that the specimen should 

 have been carried to this place from Arctic water layers. L,ike other Arctic hydroids, such as 

 1'nbnlaria rcgalis Boeck, Corymorpha groenlandica (Allman), and Stegopoma plicatile (G. O. Sars), 

 Myriothcla phrygia thus thrives very well in the region of the Trondhjemfjord which is otherwise 

 characterized by Atlantic Lop/ioAetia-reeis. This cannot, however, be turned to account as a proof 

 against the Arctic character of the species; in the Trondhjemfjord all these species must be looked 

 on as Arctic relicts. 



Family Tubulariidae. 



"Hydroids forming colonies, or solitary, with large oviform or globose nematocysts in the ecto- 

 derm. The tentacles of the full-grown polyp are filiform and simply constructed with no central cavity; 

 they are arranged in two main circles, a proximal whorl round the broad basal portion of the polyp 

 and a distal whorl round the mouth. The basal whorl of tentacles is supported by a ring-shaped 

 mesogloeal cushion round the basal portion of the polyp. The polyps are radially symmetrical. The 

 supporting lamella shows a very simple structure". 



Most authors refer to this family also Branchiocerianthus, which I reckon as the type of a family 

 of its own, the Branchiocerianthidae. This family is distinguished from the Tubulariidae partly by bilaterally 

 symmetrical polyps, partly by the peculiar anatomical structure of the polyps. In fact, the structure of the 

 polyps of the Branchiocerianthidae differs greatly from that of all other hydroids (comp. Stecho w 1909). 

 There occurs a supporting lamella of complicated structure, and the polyp is provided with numerous, 

 prominent radial canals. They are distinguished from the Tubulariidae also in their inner anatomy by almost 

 lacking any trace of mesogloeal tissue at the base of the proximal whorl of tentacles, owing to the 

 radial canals. These points of difference are, indeed, of such importance that they fully justify the sepa- 

 ration of Branchiocerianthus into a family of its own, the Branchiocerianthidae beside the Tubulariidae. 



Kiihn (1913) divides the Tubulariidae into two subfamilies, Tubulariinae and Corymorphinae. 

 The former he defines by the characters "Koloniebildend, Periderm gut entwickelt", while the latter 

 is defined as "Solitar, Periderm hautig oder riickgebildet, Wurzelhaare". To this is to be observed that 

 a species as Tubularia cornucopia Bonnevie is a typical Tubularia in spite of its forming no colonies. 

 The filamentary appendages of Corymorpha must be compared with similar phenomena in Myriothcla, 

 some species of which have rhizoids, while others such as Myriothela Cocksi (Vigurs), attach thems- 

 elves by a plateformed perisarc; when this is considered as nothing else or no more than a criterion 



