OCTOPODIDAE 387 



The identification of the various forms represented in this collection, and their re- 

 lationship to previously described species, have proved a very difficult matter. They do 

 not readily fall into line with the specific distinctions proposed by other workers, nor 

 am I satisfied that those distinctions represent natural groups. The initial difficulties 

 towards a clear understanding of the Antarctic Eledones will become clear, if the fol- 

 lowing facts are borne in mind. 



(i) In general the parts usually described in systematic papers are very variable and, 

 unless a large number of individuals is available from which a clear statistical expression 

 can be obtained, descriptions such as "web deep," "arms short," etc., are valueless. 



(2) The character of the hectocotylus is a valuable feature. 



(3) The epidermal sculpture is very liable to be effaced and is subject to modification 

 by the contraction of the skin. 



(4) Though very little is known on the subject, it is plain that the proportions of arms, 

 web and eyes alter during development, so that the young stages differ materially from 

 the adults. 



Now the descriptions of previously known forms of Eledone from this area are difficult 

 to utilise, either because they are based on a single or a few specimens, or because the 

 latter are small, or because females only were available, or finally because they were based 

 on characters of very dubious value. 



The species in question are as follows : E. charcoti. The type is a rather small female 

 specimen. Joubin subsequently figured (without description) two smaller examples 

 (1914). Massy (1916) described in some detail the anatomy of this form from fifteen 

 adults and five young. She did not, however, describe the variation in the external 

 diagnostic features. Hoyle (1912) described the colour, hectocotylus and radula, using 

 two specimens from the South Orkneys, but without giving other external diagnostic 

 characters. In 1917 Berry described a single male from off Queen Mary Land under the 

 nameoi " Moschttesaiirorae." Odhner( 1923) recorded some examples from South Georgia 

 without full description. The same incompleteness rules in our knowledge of E. tiirqueti. 

 This was originally described from Wandel Island off Danco Land by Joubin (1906) 

 from a single small female specimen measuring 42 mm. over all, and in 1914 he cited 

 (with a few notes and two figures) three small specimens and one much larger example 

 from King George Island, South Shetlands. In 1916 Massy described the anatomy of 

 four specimens (three females, one male) from Rio de Janeiro and McMurdo Sound, 

 Ross Sea. 



Turning now to the Antarctic species described by Berry (1917), we find that M. 

 albida (Wilkes Land) is known from one female, adelieana is represented by a single 

 adult female and two young and harrissoni by three females and one other specimen. 



From these remarks it will be seen at once that the available descriptions are either 

 based on very small series, or are of uncertain application. The determination of the 

 specimens obtained by the 'Discovery' is therefore of no little difficulty. 



In the forty-four specimens available I find that four distinct forms can be recognised : 



(i) A broad-bodied form with rather a narrow head and small eyes. The web varies 



