39 o DISCOVERY REPORTS - 



The type specimen of this species is a complete, mounted colony (87. 12.9. 128), and 

 agrees well with Busk's description and figures, as do some fragments (99.7.1.892 

 Busk Coll.) from the same station. A large colony in spirit and some mounted pieces 

 (87. 12.9. 129 and 99.7. 1 .891 respectively) also from Challenger St. 167 were labelled 

 by Busk as Caber ea rostrata, but they are characteristic specimens of C. zelandica. 



The specimens from Tristan agree very closely with the type. The only differences 

 detected are the slightly greater length of the pointed distal lobe of many of the scuta 

 (Fig. 25 A), and the somewhat shorter mandible and correspondingly less prominent 

 cell of the giant avicularia (cf. Fig. 25 A and B). 



A specimen from New Zealand (St. TN 134), which otherwise resembles the type, 

 has a little projection from the outer distal corner of the opesia which sometimes meets 

 the scutum. 



One of two young colonies from St. 1187 (Tristan, 18 November 1933) has developed 

 far enough to show that it belongs to C. rostrata, and there are several young colonies 

 from New Zealand (St. TN 134, 31 August 191 1). The ancestrula and first zooecia are 

 hardly to be distinguished from those of C. darwinii. Frontal avicularia appear sooner 

 and are more conspicuous than in C. darwinii, and the scuta differ in shape. In C. 

 rostrata the scuta on the first zooecia are less unlike those typical of the species and 

 the transition to the normal form is completed by about the fifth zooecium. 



The form recorded by Marcus as C. rostrata differs from Busk's species, see below. 



The specimen from La Perouse, Sydney, attributed by Waters (1887) to C. rostrata, 

 has been lent to me by the Manchester Museum. It has no giant avicularia, and the 

 vibracula have a rather small chamber, leaving a great part of the basal surface of the 

 zooecia uncovered. The groove is short and does not extend beyond the chamber. The 

 setae are unfortunately all missing. The few complete scuta resemble those of C. rostrata 

 in general shape but are a little larger, and the cryptocyst is narrow as in that species, 

 but the specimen is clearly distinct from C. rostrata. 



9. Caberea sp. 



Caberea rostrata Marcus, 1921a, p. 95, text-fig. 1 a, b (not C. rostrata Busk). 



Station distribution. Not represented in the Discovery collections. 



Geographical distribution. Juan Fernandez (Marcus). 



Professor Marcus has kindly lent me a slide of the material from Juan Fernandez 

 recorded by him (1921a, p. 95) as Caberea rostrata, and I have also examined the un- 

 mounted material for which I am indebted to the Riksmuseum, Stockholm. It differs 

 from typical C. rostrata in the shape of its giant frontal avicularia in which the chamber 

 expands proximally, spreading over the base of the scutum of the neighbouring zo- 

 oecium ; in the scutum which has a larger blade with both the distal, and the proximal 

 lobes rounded ; and in having a slightly wider cryptocyst. The first two points are clearly 

 shown in Marcus's figure, but the cryptocyst, as Prof. Marcus has himself pointed out 

 to me, was not very accurately represented by his artist. I regard Marcus's form as 

 distinct from C. rostrata. Comparison with C. darwinii (see p. 387 above) shows that it 

 is also distinct from the forms of that species known to me. 



