GEOGRAPHICAL DISTRIBUTION 483 



water in the South Atlantic Ocean and of the temperatures at 2000-2500 m. throughout 

 the area show that the Antarctic convergence corresponds to a marked change in 

 conditions at all depths (cf. discussion of abyssal species, p. 484). 



The Antarctic and sub-Antarctic abyssal records (right-hand column) are all from 

 depths of more than 2000 m. A few of the records of these species outside the area 

 (left-hand column) are from depths between 1000 m. and 2000 m. (see Table 3 A), 

 and are thus not truly abyssal. They appear, however, to be part of a deep-water fauna 

 best represented by the abyssal symbol. 1 Only four of the abyssal species extend into 

 depths of less than 1000 m. and two of these are only known from deep water. The 

 four species are Farciminellum hexagonum, known from Marion Island in the sub- 

 Antarctic region and abyssally farther north; Himantozoum sinuosum, known from the 

 islands of the south Indian Ocean and abyssally in the Antarctic ; Kinetoskias pocilhim, 

 known abyssally off Valparaiso and in less than 733 m. off Brazil ; and Cornucopina conica, 

 known abyssally off the Crozet Islands and from the Malay Archipelago both abyssally 

 and in 924 m. (see Table 3 A). 



Within the sub-Antarctic area the few species known from the South Pacific region 

 are, with the possible exception of the unnamed Caberea (see p. 390), also known from 

 the South Atlantic region. The South Atlantic and South Indian regions have some 

 species in common, but most are restricted to one or other of the two regions. Here, 

 too, the tabulation of varieties and species independently makes little difference, 

 Notoplites elongatus being the only instance where the species is in one zone and the 

 variety in the other. 



The islands of the South Indian Ocean are on the border-line hydrologically (see 

 Deacon, 1933, p. 193). Marion Island, Prince Edward Island and the Crozet group are 

 to the north of the Antarctic convergence, which passes through Kerguelen and leaves 

 Heard Island well to the south. Hydrographic conditions in the region are complicated 

 (see Deacon, 1937, pp. 31, 34, 35), but Heard Island is clearly Antarctic hydrologically. 2 

 Nevertheless, the Polyzoan fauna of Heard Island appears to be related to that of the 

 other islands of the group. The Cellularine Polyzoa of the whole group of islands com- 

 prise (1) species not known elsewhere (except Farciminellum hexagonum found abyssally 

 further north), (2) species also found in South Atlantic sub-Antarctic localities, (3) 

 species common to the Antarctic and the South Atlantic or South Pacific regions of the 

 sub-Antarctic, (4) species found in the Antarctic but not in the South Atlantic or 

 South Pacific regions of the sub-Antarctic. Half the species are not known elsewhere 

 (group 1). Of the remainder the majority are sub-Antarctic species unknown in the 

 Antarctic (group 2), while group 4 only comprises two species, Himantozoum sinuosum, 

 known abyssally in the Antarctic, and Bugula longissima. Moreover, the most charac- 

 teristically Antarctic genus in this report, namely Camptoplites (see p. 492), has not 



1 A few of the non-abyssal species extend into depths of a little more than 1000 m., but such records have 

 not been discriminated. 



2 The data published by Howard (1940, p. 43) afford further evidence of the Antarctic character of 

 Heard Island. Forthcoming Discovery Station Lists will give hydrological data for more stations in this 

 region. 



