CABEREA 377 



and now in the British Museum, can both be recognized as figured specimens. Of these 

 99 . 7 . i . 855 (Straits of Magellan) is figured ( x 2, not natural size as stated) in pi. xxxviii, 

 fig. 1 of the B.M. Catalogue. The other, 99.7.1.856 (South Patagonia, 19 fm.) is, as 

 we have seen, the original of Challenger pi. xxxii, fig. 6/. It has not been possible to 

 determine which, if any, of these four specimens were used for figs. 2-7 in the B.M. 

 Catalogue, but as 99.7.1.855 is the original of fig. 1, and is a complete and well- 

 preserved colony, I select it as the holotype. 



Description of specimens. The holotype (Fig. 21 A), and other specimens from 

 Darwin's collection represent one of the extremes among the specimens attributed to 

 C. darwinii. They are indistinguishable from C. minima (Fig. 21 B) which is thus, as has 

 been suggested by various authors (e.g. Marcus, 1921a, p. 96), a synonym of C. dar- 

 winii. The colonies are small, compact and fan-shaped, branching freely (cf. Plate VI, 

 figs. 1, 2). The branches are rounded or flattened, rather than keeled, frontally. The 

 ovicells are at least as wide as long and are only slightly oblique. There are usually two 

 and occasionally three spines on the outer distal corner (sometimes very close together 

 and hiding each other), one inner spine near the base of the scutum (peduncular spine), 

 and often another inner one placed more distally (Fig. 21 A). The scutum is more or 

 less rounded, its proximal lobe is the longer, but is not particularly long. The cryptocyst 

 may be smooth or slightly granular. The frontal avicularia are not noticeably enlarged. 

 The setal grooves and setae of the vibracula are relatively short. 



The other extreme is represented by some of the Antarctic material of the National 

 Antarctic Expedition. The colony consists of long, stout, rather straggling branches. 

 The zooecia are large and long, those of the minima type being less than two-thirds 

 the length of the Antarctic ones. The vibracula are large, with long setal grooves and long, 

 stout setae. The zooecia face obliquely away from the mid-line of the branch, giving it 

 a slight keel (Fig. 21 C), which tends to be accentuated by the enlargement of the frontal 

 avicularia, which are placed on this line, and by the ovicells which are usually placed 

 obliquely so that their most prominent portion comes on to the projecting median portion 

 of the branch. The avicularium on the median zooecium at the bifurcation is commonly 

 the largest, though not differing in shape, nor very markedly in size, from the others 

 (Fig. 24 A). The cryptocyst is granular. The scutum has a long proximal lobe. The 

 spines in such specimens are commonly reduced to two, a stout one at the outer distal 

 corner and a peduncular spine near the stalk of the scutum, but the more typical arrange- 

 ment with two outer and one peduncular spine is also found, and in the figured specimen 

 the small spine distal to the peduncular spine is also present. The larger size of Ant- 

 arctic specimens of this species was also noticed by Kluge. 



Challenger specimens are intermediate between these two extreme types. In the 

 specimen from St. 148, Possession Island, Crozet Group (Plate VI, fig. 3), the branches 

 are moderately keeled (cf. Busk, 1884, fig. 6 d), and many of the frontal avicularia are 

 somewhat enlarged (Fig. 22 A). The zooecia and vibracula are large, the ovicells rather 

 oblique and the cryptocyst sharply defined and granular. A single outer distal spine is 

 usual, but there are two on some of the zooecia in the proximal part of the colony, which 

 does not otherwise differ from the other parts. 



