6 7 



as hearing two or three groups of zooecia, instead of a continuous spiral. But the groups in 

 question may readily be regarded as parts of a single spiral. That this view is correct is proved 

 hy 126. A., in which, in one and the same branch, some of the internodes bear a single spiral, 

 of the form typical in the present species, while in other internodes the spiral is more or less 

 broken up, in various degrees. Thus in some cases the groups are separated from one another 

 by intervals of bare stem, while in others successive groups touch at their bases, being manifestly 

 parts of one spiral in which, here and there, two zooecia diverge from their bases, thus dividing 

 the spiral into discrete groups. In some of the branches of the last order, in which the limit 

 of growth has been reached, all the internodes bear these short discrete groups of zooecia. 



The correctness of this conclusion has been confirmed by an examination of Busk's type- 



* 



material in the British Museum (87. 12.9.935,936), from the 'Challenger' Station 188, S. of 

 New Guinea. The spirit-specimen (936) has the same general appearance as that of the 'Siboga' 

 material. A large proportion of the internodes, both in older and in younger parts of the colony, 

 have the spiral broken up into 2 or 3 groups ; but the reference of the specimen to the present 

 species is indicated by the fact that in other parts the spiral is continuous and quite typical. 

 The condition figured by Busk as A. scmispiralis may thus probably be regarded as one which 

 is sometimes assumed by A. convoluta, under circumstances unfavourable to growth in full vigour. 

 Waters l ) has described a similar specimen from Torres Straits. In the absence of figures I am 

 unable to express any opinion on the question whether " Serialaria [Vesicularia'ï) scmispiralis" 

 Kirchenp. ~) belonged to the same species or not. 



Amathia convoluta Lamouroux and Serialaria convoluta Lamarck 3 ) were described in 

 the same year (18 16); neither author giving a reference to the other. Lamouroux stated, 

 however (p. vi) that in 1 8 1 2 he had named a part of Lamarck's collections for him ; and 

 Tenison-Woods, in the Memoir (p. 96) cited above, has used this as one of the reasons for 

 preferring Lamouroux' names to those of Lamarck, in cases where the same names were used. 



MacGillivrav describes the spiral, in Victorian specimens, as being continuous but 

 "interrupted at the division of the branches". The figures given by this author seem to show 

 that the 'Siboga' specimens belong to the same species, although the groups of their successive 

 internodes are apparently always separated by a slight interval. It is clear that Busk, in the 

 'Challenger' Report, had wrongly ascribed the specimens recorded as A. spiralis to that species. 

 Mr Kirkpatrick *), who has for many years been in charge of the Polyzoa in the British Museum, 

 has pointed out that A. spiralis Busk, is probably A. convoluta Lamx; and MacGillivrav") 

 has come to the same conclusion. 



Pergexs 6 ) has called attention to a series of unpublished Plates, by Desmarest and 

 Lesueur, of which the originals are said to belong to the Museum at Havre. The MS. describing 

 these Plates was deposited by the authors, in 1829. in the Library of the Muséum de Paris, 



1 Waters, A. W., 1887; — see ref. in synonymy. 



2) Kirchenpauer, G. H., 1869, "Museum Godeffroy. Cat. IV", Hamburg, p. xxxiv (Samoa). 



3) This appears to be A. spiralis Lamx. 



4) Kirkpatrick, R., 1888, "Pol. Port Phillip", Ann. Mag. Xat. Hist., (6) II, p. 19. 



5) MacGillivrav. P. H., 1895, t. cit.. p. 133. 



6) Pergexs, E. : — see synonymy. under A. crispa. 



67 



