EXPEDITION AXTARCTIOUE BELGE 



of some spécimens, and then the pedicel is not continuous in the same plane, which would suggest 

 injury, but this is not borne out by other nodose species. As some of the pedicels hâve thèse 

 internodes and others are continuous we hâve hère ample grounds for banishing the genus 

 Gonypodaria of Ehlers (') as some zoœcia on the same colony would be Barcntsia, and others 

 Gonypodaria. In the B. gracilis, var. nodosa Lomas ( 2 ), in the same way, some stalks are entire 

 and others ncdulated ; and doubtless other generic names are superfluous. 



The structure of the internode is similar to that ofthe « barrel-shaped cylinders », and 

 figures are given of sections of thèse cylinders, as they differ from published descriptions of 

 other species. The walls of the pedicel are a firm, thick, dark chitin continuing over the 

 upper part ofthe barrel, and although the membrane continues down the side of the barrel the 

 wall is without any chitinous deposit [ex). Within this is a broad contractile layer (cl) folio wed 

 by a layer of large cells (et) within which is the parenchym tissue. 



At the junction ofthe pedicel and the barrel there is on each side a chitinous projection, 

 which will form the équivalent of the rosette plate, but doubtless also serves for muscular 

 attachments, and this is no doubt what Ehlers figures on PL II, fig. 22, while similar 

 contractions also occur in the muscular internodes of B. discreta. This contraction is added 

 to the drawing of fig. 17c from another section. The same structure is shown by Miss 

 A. Robertsox ( ! ) in fier figure of Gonypodaria ramosa Robertson. 



Pediccllina and Barcntsia hâve been found in the northern seas, Australia and S. America, 

 but hâve only been mentioned within the tropics by Kirkpatrick who described P. laxa from 

 Torres Straits. However I hâve had material sent to me, by my nephew Ernest J. H.YVaters 

 of Sydney, collected by him from Lizard Island off Queensland, in which there are spécimens of 

 Barcntsia gracilis Sars. In the same material, besides Scrupocellaria cervicomis Busk, Loxosoma 

 occurred, and I believe it has not been previouslv found within the tropics and possibly not in 

 the southern hémisphère. 



It may be well to call attention to the, fact that the « spirally ringed flexible joint» 

 uniting the polypides to the pedicel is subject to variation, and there being hère no thick 

 chitinous walls the contractions cause the rings. In 011e case the junction (fig. 6/) with the 

 hard chitinous wall ofthe pedicel is very similar to that of Ascopodaria fruticosa Busk (Challenger 

 Report, pi. IX, fig. 2). 



Habitat. — Tristan d'Acunha ioo-i5o fat h. (B.) ; Tizard Bank, China Sea 27 fath. 

 (Kirkpatrick) ; Ile Hoste, Baie Orange, Cap Horn 26 fath. (40 met.) (Jull.). 



Exp. Antarct. Belge. 



X" g5i. Fixés sur Prienc cancellata (Lamx.), N° 118. Baie du Torrent, Ile Londonderry, 

 Canal Français, Magellanes Chili. Dec. i8 th , 1897. 



Note. — Since the remarks on the number of species (p. 10) were printed there has been 

 reason for considering that the number of fossils is only 10 as subsequentlv shown on pp. 18, 19. 



(1) E. Ehlers, Zur Kenntniss der Pedicellineen. (Abh. d. k. Gesellsch. d. Wissensch. zu Gottingen, vol. XXXVI, 

 p. 144.) 



(2) Proc. Lit. Phil, Soc. Livapool, vol. XL, p. 190, pi. III, fig. 2, 1886. 



(3) Proc. California Acad. of Se, vol. II, X T ° 4, pi. XVI, figs. i5, 16. 



