77 



In the absence of any polyps, I could not do more than 

 give close attention to the mode of branching, the arrange- 

 ment of the twigs, and the disposition and character of 

 the spines. By comparing the Fteroe specimens with 

 others, and by following a method of exclusion, I have 

 convinced myself that it is referable to the well-known 

 Mediterranean species — Parantipathes larix (Esper), see 

 Brook (1889, p. 142). If this conclusion is correct, the 

 occurrence of this species to "the northeast of the Fseroe 

 Islands " is of great interest, which is increased by Professor 

 Hickson's record (1907) that Parantipathes larix was col- 

 lected by the "Huxle)'" from the north side of the Bay 

 of Biscay in August 1906. 



Eecognising the scientific responsibility of stating that 

 a well-known Mediterranean species occurs in such high 

 latitudes as north-east of the Faroes, I have carefully 

 considered all the Antipatharians in Brook's "Challenger" 

 Eeport and in subsequent publications, and I find that, 

 apart from ParoMtipatJies lai-ix, there is only one other Anti- 

 patharian which the Faroe specimen resembles in any 

 marked degree, and that is Taxipathes recta, Brook (1889, 

 p. 156), which is only known by a single specimen, obtained 

 off Ascension. That this species cannot include the Faroe 

 specimen might perhaps be inferred from the distribution, 

 but as that is a dangerous way of arguing, I may point 

 out that, while Tao:ipathes recta resembles Parantipathes 

 larix and our specimen in certain respects, e.g., in having 

 the same kind of delicate simple pinnules arranged in a 

 brush in six vertical rows, it may be left out of considera- 

 tion because of its extremely stiff rectangular branching, 

 its rigid and absolutely regular pinnules (only 18 to 21 

 to a centimetre, moreover), and its spinulation (with four 

 vertical rows of spines on one aspect). It need hardly 

 be said that the presence of even a few polyps would 

 have obviated even a minute's consideration of whether 

 the form in question was referable to the Schizopathinje, 

 among which Taxipnthes is included, or to the Antipathinse, 

 among which Parantipiathes is included. 



The real difficulty in regard to the specimen from the 



