On the Gonostyles of hvo Antarctic Siphonophora. 19 



III. Scotia Collections. — Note on the Gonostyles of tioo 

 Antarctic Siphonophora. By Professor J. Arthur 

 Thomson, M.A. [Plate I.] 



(Received 14th November; read 28tli November 1904.) 



Mr W. S. Bruce has been good enough to entrust to me 

 three interesting and somewhat puzzling specimens collected 

 by the Scottish Antarctic Expedition off the South Orkneys. 

 They seem to be the separated gonostyles of some large 

 Siphonophore colony, and as such they certainly deserve to 

 be recorded, since we have very little knowledge of Siphono- 

 phora from the far south. In the report on the " Southern 

 Cross " collections, Mr E. T. Browne mentions the necto- 

 calyces of a Diphyid, and a specimen of an Agalmid 

 {Halistemma) , about 6 to 10 inches long ; in the Antarctic 

 Manual, Mr A. E. Shipley refers to an abyssal Discalid, 

 Disconalia pectyllis. 



Specimen A. was obtained in July 1903 in Scotia Bay, 

 South Orkneys, on the surface of the water, in a hole which 

 had been cut in the ice. Tiie depth of water at that place 

 was 20 to 30 fathoms ; the temperature 29° F. Specimen B. 

 was dredged from among mud and pebbles in December 

 1903 in Scotia Bay, South Orkneys; the temperature 31° F. 

 Specimen C. was dredged from 9 to 10 fathoms in May 1903 

 in Scotia Bay. It is in several respects very different from 

 A. and B., and I have therefore referred in the title to two 

 Antarctic Siphonophora. Other specimens, which I have 

 not seen, were found in summer, floating on the ice-free 

 surface of the bay. Specimens B. and C. were allowed to 

 die slowly in sea-water, and were then preserved in formol. 

 Their general structure is quite clear, but such delicate 

 organisms require more elaborate treatment if they are to be 

 used for histological analysis. Nevertheless the sections 

 that have been made reveal many interesting details. 



Specimen A. measures 4 inches in length by 07 inch in 

 maximum diameter. It narrows gradually to one end, where 

 there is a definite opening at the top of a small conical 

 elevation (see Fig. I., 0.). The other end is blunt and blind. 



