CAULIBUGULA, KINETOSKIAS AND CAMPTOPLITES 433 



Caulibugula Verrill, 1900 



1. Caulibugula tuberosa Hastings. 



Caulibugula tuberosa Hastings, 1939, p. 340, text-figs. 279 A-C. 

 Station distribution. New Zealand: St. 934. 

 Geographical distribution. New Zealand (Hastings). 



Kinetoskias Daniellsen, 1868 



1. Kinetoskias pocillum Busk. 



Kinetoskias pocillum Busk, 1881, p. 7, pi. i, figs. 2, 5; 1884, p. 45, pi. viii, fig. 2; Harmer, 1926, 

 p. 468. 

 Station distribution. Not represented in the Discovery collections. 

 Geographical distribution. Off Valparaiso, 3953 m. ; off Brazil, 58-6-732 m. (Busk). 



Camptoplites Harmer, 1923 



General. Camptoplites was introduced by Harmer (1923, p. 300) and redefined by him 

 (1926, p. 452). The abundant material before me suggests one modification of the de- 

 finition. I do not think the zoarium is usually stalked. The stalk of C. lunatus consists 

 of a bundle of the axillary rootlets, which are applied to the basal surface of the branch 

 and so run together throughout their length. In the other species the axillary rootlets 

 are free from the surface of the zoarium, and are attached singly to the substratum, 

 and it is clear that a stalk, if present, would have to be formed in a different way from 

 that of C. lunatus. Ancestrulae of several species are known (Figs. 39 A, B, 40, 41 A, B, 

 42 A, D). They agree in having a tubular proximal portion, ending in root-like structures 

 by which the ancestrula is attached (see p. 437). As the growth of the colony proceeds 

 this attachment is supplemented by axillary or marginal rootlets which pass to the 

 substratum quite independently of the ancestrula, of the zooecia, and of each other (Figs. 

 41 A, 42 D). There is nothing in any of these instances to suggest that a stalk would 

 subsequently have been formed. On the other hand, it is easy to imagine the delicate 

 ancestrula, and probably also some of the other basal zooecia, breaking away leaving 

 the colony attached by a number of independent rootlets, just as all the adult colonies 

 (except that of C. lunatus) examined by me have been. It is hard to believe that if a stalk 

 were characteristic of the genus no trace of it would have been found in my material. 



In 1926 Harmer added to his earlier definition the statement that the connecting 

 process at a bifurcation gives off two distal branches and a rootlet (e.g. Fig. 52 A). 

 Examination of the Antarctic material shows that the constant feature is not the forma- 

 tion of the two distal branches or runners (though these are present at some or all 

 bifurcations in the majority of the species), but the separation of the connecting process 

 itself, 1 at least in those bifurcations that have produced an axillary rootlet, as a separate 

 chamber from which the rootlet originates when present (Fig. 46 B). The axillary 

 chamber can usually be seen to be connected with one of the neighbouring zooecia by 

 a conspicuous rosette-plate (e.g. the large one on the left in Fig. 47 H). In most species 

 the rootlet grows symmetrically and from the basal surface, but in C. reticulatus, 



1 Compare Bugula longissima Busk (p. 43 1 above). 



