6 9 



and the diaphragm at the base of the branch lies just on its proximal side. There are no 

 definite joints in the region of the diaphragms, although the stem may be somewhat constricted, 

 a short distance on the distal side of a diaphragm, both of the main stem and at the base of 

 the branch. Zooecia not completely connate, the distal two-fifths or more being free, and sub- 

 conical. Orifice terminal, obscurely quadrangular. Basal wall of the zooecium not conspicuously 

 broader than long, the zooecia thus not having the compressed form characteristic of those of 

 A. convoluia. Polypides with 8 tentacles and a gizzard. 



The present species differs from A. convoluta in its smaller size, its repent habit, the 

 smaller diameter of its sterns and the shorter spirals formed by the groups of zooecia, and in 

 the fact that its zooecia are not completely connate. It is attached to its substratum, which is 

 sometimes a Sponge, by rootlets, which may be formed as a modification of an ordinary branch. 

 The branch thus modified usually assumes an irregular, sinuous outline, and gives off a new 

 rootlet here and there, a few diaphragms being developed. 



The zooecia (with retracted polypides) reach a length of about 450 — 500 u.. The vestibule 

 is nearly half the length of the zooecium and the collar is small. In some of the old zooecia 

 the free terminal part is marked by numerous circular striae; a condition which probably 

 indicates the approaching degeneration of the polypide. 



The discrimination of the species of Amathia is well known to be difficult; and there 

 is much uncertainty with regard to the synonymy. The present specimens appear to be 

 indistinguishable from the type-specimen (Brit. Mus., 87. 12. 9. 926) of A. distans, described by 

 Busk l ) from Bahia, Brazil. With that species it agrees in its straggling growth, in the restriction 

 of the spiral groups of zooecia to the distal half of the internode, in the fact that the zooecia 

 are only partially connate, and in the measurements. The internodes are described by Busk as 

 being 150 u. in diameter, and the zooecia are 400 — 500 u. in length. A South Australian form 

 has been described by MacGillivray -), under the same name. The species recorded by Waters 3 ) 

 from the Sudanese Red Sea, under the name of A. fortuosa, is similar to the 'Siboga' specimens 

 in the form of its zooecia, but its internodes seem to have a thin cuticle. I venture to doubt 

 the correctness of Mr Waters' determination, since the zooecia are shown as very short and 

 somewhat discrete; whereas in the original description of Tenison-Woods, the "great length of 

 the pairs of cell" *) is emphasized, and they are represented as completely connate; the spiral 

 being, moreover, much elongated and occupying most of the length of the internode. A. distans, 

 mentioned by Waters in the same paper, is figured (PI. XXIV, fig. 7) as having connate zooecia. 



The 'Siboga' specimens show considerable resemblance to A. pruvoti, described by 

 Calvet 5 ) from Banyuls-sur-Mer; but they differ from that form in the fact that the zooecia 

 are less completely connate. 



1) Busk, G., 1886, loc. cit. 



2) MacGiu.ivray, P. H., 1895, loc. cit. 



3) Waters, A. W., 1910, loc. cit. 



4) Tenison-Woods, J. E., 1880, "On the Genus Amathia", Trans. Proc. R. Soc. Vict., XVI, p. 89. 2 figs-on unnumbeied Plate. 

 In Tenison-Woods' fig. 6, the zooecia appear rather short. It is possible that the words "the great length of the pairs of cells in each 

 internode" were intended to refer to the length of the spiral group : — which would be more in accordance with the figure. 



5) Calvet, L., 1911, "Deux Esp. Nouv. Bry. Mediterranée", Arch. Zool. Exp. (5) VIII, Notes et Revue, p. I.IX, text-fig. on p. lx. 



69 



