I 2 I 



Both lobes possess an ovicell near the distal extremity, and the longer lobe has a more 

 proximal one as well. The ovicells are all alike, rather broad and moderately swollen, but 

 not very distinctly outlined. The ooeciostome is in each case a very short tube, which rests 

 against the two inner zooecia of a series, on their proximal side. The ooeciopore is transversely 

 oval and is visible in the frontal view of the branch. 



The basal surface of the lobes is prolonged into a well marked marginal lamina, which 

 projects beyond the rest of the lobe and is closely adnate to the substratum. The marginal 

 lamina shows a series of porous areas, which are represented in fig. 9 (/) on the right side, 

 and are separated from one another by low but moderately broad ridges. 



The Polyzoa belonging to Dr A. Willey's Collection from Lifu and other localities 

 were described by Miss E. G. Philipps, and are preserved in the Universitv Museum of Zoology 

 at Cambridge. Two slides mounted by Miss Philipps were referred by her to I. australis\ 

 but they certainly do not belong to MacGillivray's species. Of these, one slide contains 

 several young colonies, on sea-weed; showing the primitive disc, which is surrounded by fine 

 denticulations, like those figured by Busk r ) in a form referred by him to Tttbulipora organisans 

 D'Orb. The marginal lamina of the specimens in question, instead of forming a broad zone round 

 the branch, is hardly present, but it is represented by a number of short lateral projections 

 which are separated from one another by rounded bays ~). The zooecia have long, slender, 

 curved peristomes, some of which are free and others connate. It is not impossible that these 

 specimens belong to the same species as the 'Siboga' specimens ; but, if so, it would have to 

 be assumed that the entire colonies hacl not developed beyond the condition which is indicated 

 in the central region of fig. 9 and that the projecting points of the marginal lamina represent 

 the porous areas of the 'Siboga' specimens not yet closed. In the absence of any evidence as 

 to the characters of the primitive disc of the 'Siboga' specimens, it woulcl, however, be unsafe 

 to assume that these Lifu colonies belong to the same species. 



The second slide referred by Miss Philipps to I. australis may, however, be placed 

 without hesitation in the same species as the 'Siboga' specimens, with which it agrees closely 

 in the form and arrangement of the zooecia, in the ovicells (of which two are present), in 

 the ooeciostomes, and in the marginal lamina. The specimen throws considerable light on the 

 morphology of the porous areas of the marginal lamina shown in fig. 9. The agreement with 

 the 'Siboga' specimens is complete in the more central part: but, at the end of one of the two 

 lobes present, the porous areas are represented by the open mouths of structures which exactly 

 resemble the incompletely developed zooecia which are usually found at the growing margin of a 

 Cyclostome colony. It is obvious from this specimen that the porous areas result from the closure 

 of the imperfectly developed zooecia by calcareous plates, in which pores are developed. The ridges 

 separating the porous areas are clearly seen to be the remains of the lateral walls of zooecia, the 

 porous plates having been developed at a level slightly lower than the edges of the septal walls. 



1) Busk, G., 1879, "Pol. Kerguelen", Phil. Trans., CLXV11I (Trans, of Venus Exp.) p. 6 (sep.). PI. X, figs 22 — 25. 



2) It may be noted that the occurrence of fine denticulations round the primitive disc and of pointed lateral projections at the 

 edges of the marginal lamina cannot be regarded as distinctive of any single species, since both characters are shown in Japanese specimens 

 (Owstün Coll.. 40. F.), in the Cambridge Collection, the adult condition of which is of a flabellate type resembling Tubulipora jlabcllaris. 



121 



SIBOGA-EXTEDITIE XXVIII a. IÓ 



