44 DISCOVERY REPORTS 



of which a row of small pores occurs, through which the gastropore and the lower part of the dactylopores 



communicate. 



Although at present the full range of variation in the species of the genus Conopora is not known, it 

 seems most probable that the specimen from the Discovery material represents a hitherto undescribed 

 species, the features of which may be summed up as follows : 



Conopora pauciseptata n.sp. 

 Colony flabellate (?). Stem and main branches with cyclosystems in three main rows along the 

 anterior and lateral sides. Surface vermiculated, with numerous small nematophores. Cyclosystems 

 complete only terminally on branchlets, elsewhere embedded and incomplete, with an adcauline 

 diastemma in the dactylopore circle. Gastropore with a feebly broader lower chamber separated from 

 the funnel-shaped narrower upper chamber by a circlet of small pores, through which the dactylopores 

 communicate with the gastropore. Dactylotomes about half as long as the wall of the upper gastropore 

 chamber. Dactylopores in the (incomplete) cyclosystems of stem and branches number 3 to 1 1 

 (average about 7*07). 



ZOOGEOGRAPHICAL REMARKS 



Up to the present species of the genus Errina have only been reported from antarctic and sub-antarctic 

 (antiboreal) waters. The Discovery Expedition has added one species of a closely related genus, 

 Errinopsis. It is of interest at this point to review the antarctic and sub-antarctic data as a whole and 

 in connexion with the comparatively rich Discovery collections, and to compare them with data from 

 other regions. 



According to our recent knowledge (Broch, 1942) the genus Errina must be regarded zoogeo- 

 graphically as a southern genus, the overwhelming majority of specimens deriving from sub-antarctic 

 waters, and it seems reasonable to assume that the genus must have originated in this region. 



It is very difficult to determine the species of this genus. Their distinguishing characteristics are in 

 many cases seemingly of minor importance, and Hickson (1912), moreover, observed great variation 

 in his comparatively abundant collections of the species E. novae-zealandiae, which he accordingly 

 split into several 'fades'. However, like previous investigators of Stylasteridae, Hickson did not take 

 into account those subtle characters which have proved to be of the greatest importance as a specific 

 fundamentum divisionis in several cases, and his ' facies ' seem to be more in the nature of casual growth 

 forms of colonies, probably caused by ecological conditions. 



It is accordingly of interest to note that the ' small ' characteristics of the species E. antarctica are 

 obviously stable. Colonies, the surfaces of which have in places been more strongly eroded, may at first 

 give an aberrant impression. However, branch ends intact with undisturbed surface and intact spines 

 exhibit all the typical characters and do away with every doubt as to specific identity. 



E. antarctica is characteristic of the antarctic and sub-antarctic regions (Fig. 12). According to 

 Ridley (1881) the northernmost find and the type locality was made by the Alert expedition in the 

 Trinidad Channel (Madre-de-Dios archipelago in south-west Chili, about 50° S) at only 45 fathoms 

 depth, still the shallowest locality recorded. 



The species is remarkably eurybathic. According to present data it is found at depths of some 

 500 m. (the two deepest Discovery hauls of E. antarctica were from St. 1948, 490-610 m., and St. 2200, 

 532-512 m.). The Belgica expedition brought home some specimens of a Stylasterid recorded by 

 von Marenzeller (1903) as E. gracilis, which, as stated above, must be regarded as a synonym of 

 E. antarctica. Von Marenzeller did not give the exact depths of the four localities south-east of Peter I 

 Island, whereas Hickson (1912) says that the depth 'is probably between 500 and 600 metres'. I am 



