386 DISCOVERY REPORTS 



two-thirds or three-quarters as long as in the type and Ijima and Ikeda's specimens of 

 pelagicus. The latter and Sasaki's specimens seem to differ quite markedly from thielei. 

 It thus seems that thielei is quite distinct from the type and other specimens oi pelagicus, 

 though in one character it is not unlike Thiele's pelagicus. The status of the latter is 

 very uncertain. In the light of this evidence it is quite impossible to do otherwise than 

 treat A. thielei as a distinct species. 



Concerning the general affinities and status of these forms, I need say very little at 

 present. They plainly present us with a problem of the greatest interest. In the first place, 

 the external "choroidal" gelatinous coating, double pallial aperture, multicuspid teeth 

 and telescopic eyes suggest high specialisation . The remarkably developed teeth, however, 

 seem to accord little with the very weak jaws (the reverse situation in fact to that in the 

 Cirromorpha) and simple suckers. The suckers, teeth, vaginae (.?) and gills suggest 

 relationship with the Eledonellidae. The funnel-organ is, however, Octopus-\\kt. 



Family OCTOPODIDAE 

 Sub-family ELEDONINAE 



Forty-four specimens were obtained by the ' Discovery,' all of which were in very good 

 condition. This rich haul, which far surpasses any series obtained previously, provides 

 us with an admirable opportunity of reviewing this characteristic Antarctic group. 



In a work, shortly to be published, I hope to discuss in full the systematic relationship 

 of all the forms that have been placed in Eledone and the status of the sub-family. I am 

 of the opinion that Naef was correct in his suggestion that the genus may be polyphyl- 

 letic (1923, p. 716). A final verdict on this subject is not at the moment possible. I con- 

 tent myself with pointing out the following facts : 



(i) The Antarctic species so far described differ very markedly from the Mediter- 

 ranean forms, and certainly do not belong to the same genus. I consider that the name 

 Eledone should be kept for E. moschata and cirrosa and that charcoti, Joubin (aurorae, 

 Berry syn.), turqueti, Joubin, antarctica, Thiele, harrissoni and adelieana, Berry, and 

 polymorpha (p. 390) should all be placed in Graneledone, Joubin. The status of Berry's 

 albida is not for the moment clear, though I believe it should be placed with E. rotunda, 

 Hoyle, in a distinct genus or sub-genus. T. gunteri (p. 392) is placed in a different 

 genus, principally on account of its highly degenerate radula. 



(2) The new species now described bring up the number of Eledonid species to 

 sixteen (irrespective of doubtful forms), if we accept Berry's suggestion (1918) that his 

 aurorae is identical with charcoti and keep media, Joubin, as a distinct species (cf. Joubin, 

 1924, p. 38). Of these nine are from Antarctic or sub-Antarctic waters, three more are 

 from southern waters. The remaining four are from the north Atlantic. The group is 

 very largely absent from tropical waters, and is poorly represented in the Indo-Pacific 

 region. Its metropolis and probable place of origin is the littoral of the Antarctic 

 continent and deeper water further north, with a smaller offshoot in the north Atlantic 

 and Mediterranean. 



