2,6 DISCOVERY REPORTS 



contrary, we can say that the more extensive our material has become, the more indefinite are the 



limits which earlier seemed to point in the direction of local 'species'. 



Nevertheless, we must make one reservation as to the high-arctic U. encrinus already mentioned on 

 page 270. A study of its occurrence (text-fig. 5) indicates that it must have penetrated into the Polar 

 deep and into the cold bottom water of the Davis Strait and Baffin Bay from the Atlantic Ocean. In 

 the Polar Sea it has been found in water as shallow as 180 m., and it is therefore astonishing that the 

 species apparently has not been able to pass beyond the submarine Lomonosov mountain range into 

 the western part of the Polar Basin. 



Geographical distribution seems to indicate that U. encrinus in reality is an ecologically distinct 

 group. It must have invaded its two distinct habitats from the more temperate Atlantic bottom waters. 

 The western habitat in Davis Strait and Baffin Bay is barred to the north by shallow thresholds from 

 the central (both eastern and western) Polar Deep, and communication with the eastern north 

 Atlantic-Arctic deep sea, north of the Scotland-Iceland-Greenland ridges, round the southern part 

 of Greenland seems unlikely in the present state of our knowledge. It is, therefore, natural to assume 

 that penetration was originally via the Atlantic bottom water into the Davis Strait, and that a secondary 

 invasion occurred over the submarine ridge in the Faroe Channel. This double origin in Arctic waters 

 of U. encrinus favours considering it as an ecologically distinct, geographical variety, and not as a 

 genetically separate species. 



A study of Jungersen's (191 6, 191 9) papers indicates that the morphological differences between 

 U. lindahli and U. encrinus are more fluid in the Davis Strait, where no prominent submarine ridges 

 accentuate the biophysical border lines as do the submarine ridges from Scotland past the Faroes and 

 Iceland to East Greenland. But in spite of this the gap between the Arctic and the Atlantic groups is 

 so conspicuous that at present it is more convenient to treat U. lindahli and U. encrinus as separate 

 species. 



It is striking that U. encrinus is much more coarse and luxuriant than the largest Antarctic specimens 

 of U. lindahli hitherto captured. If the first is an extreme geographical race, the question again arises, 

 why the Antarctic specimens do not attain such large dimensions as the Arctic. Are conditions of life 

 subjected to greater fluctuations and, therefore, less favourable in Antarctic waters, than in the much 

 more sharply circumscribed Arctic deep sea? 



PENNATULIDAE 



Pennatula rubra Ellis 1764 



(Synonymy before 1910, see Kiikenthal and Broch 191 1) 



P. rubra Kiikenthal & Broch 191 1, p. 382, text-figs. 182-7. 



P. rubra Kiikenthal 1915, p. 91, figs. 98, 99. 



P. rubra Hickson 1916, p. 183. 



St. 272. Off Elephant Bay, Angola, 200-230 m. 9 specimens (three of them identified by Mrs MacFadyen) from 



13 to 25 cm. length. 

 St. 2633. 13° 11-5' S, 12° 44-1' E (off mouth of Elephant Bay, Angola) 104-? (sounding 91-104 m., but a note 



states that the net was fishing clear of the bottom at the end of the haul), i specimen, length 1 1 cm. 



The specimens from St. 272 vary much in length, and so does the length of the stalk, which generally 

 measures one-half to one-quarter of the entire specimen. It is of interest to note that the larger 

 specimens are much more inflated than the smaller ones, which are rather contracted, although not so 

 strongly as the specimen from St. 2633, which has a total length of 1 1 cm. only. Evidently the length 

 of the specimen depends to a certain degree on the degree of contraction, and a living Pennatula rubra 

 doubtless shows a great faculty of inflation by imbibing water. This influences the colours also. The 



