THE ALCYONARIA OF THE CAPE OF GOOD HOPE AND NATAL. 561 



situated in the rind, and extended only a short distance into the coenenchyma. The 

 canals, twelve to twenty in number, occupy the greater part of the central coenenchyma. 

 Numerous connecting capillary tubes are present, and the course of these is as follows : 

 Capillary tubes pass inwards from the polyps to join other tubes which form an encircling 

 network around the canals of the central coenenchyma ; from this encircling network, 

 other still smaller tubes pass into and terminate in the walls of the canals. PI. III. 

 fig. 15 illustrates this structure better than any verbal description.* 



The structure of the ccenenchyma, as a whole, resembles in many respects that of 

 Alcyonium. The external part of the polyp-bearing region contains more spicules than 

 any other portion of the colony. They are mostly clubs, the heads of which project 

 from the surface ; comparatively few spindles occur. The spindles measure from 0'204 

 by 0-068 to 0'493 by O'OSl mm., and from 0'204 by 0'068 to 0'289 by 0'017 mm. 

 The clubs from the same part measure as follows : From 0'187 by 0'085 to 0'357 by 

 0-102 mm., and from 0'212 by 0'051 to 0'323 by 0'102 mm. The spicules are also 

 abundant in the rind of the trunk, and are chiefly clubs. They have the following 

 dimensions: From 0'102 by 0'142 to 0'238 by O'OGS mm. The centre of the 

 trunk has comparatively few spicules : these are mostly clubs. Their dimensions range 

 from 0-17 by O'OGS to 0'204 by 0'068 mm. 



Plate 111. fig. 15 shows the presence of a number of gonads or possibly embryos 

 in the canals of the ccenenchyma. Unfortunately, I have not a sufficient number of 

 this interesting new species to enable me to make a detailed study of these gonads 

 or embryos. 



Locality, etc. : P.F. 10,993. Off Umhloti river mouth (Natal). Depth, 39 to 40 

 fathoms. By large dredge. Sand and shells (hard ground). Date, Dec. 21, 1900. 



This form differs from Metalcyonium clavatum, Pfeffer, by a sharper differentiation 

 into polyp-bearing and stalk portions, by the absence of spicules in all parts of the 

 polyp with the exception of the calyx, and by the spicules in the rind of the trunk 

 being clubs, not slender rods. 



it also differs from Metalcyonium capitatum, Pfeffer, which has a distinctly flattened 

 form, and polyps present on the basal part of the colony. Metalcyonium capitatum 

 has also spicules in the pinnules and a crown of eight rows in the polyp head. The 

 spicules of the rind are straight rods, not clubs, and the colour of the colony is light 

 orange. 



My species is readily distinguished from Metalcyonium patagonicum, May, by its 

 form and by the absence of spicules in the upper part of the polyp. 



Metalcyonium molle, Burchardt, is a form apparently without spicules, but it has not 

 been well described. My species may come nearest Metalcyonium novarce, Kiikenthal 

 (from the Cape of Good Hope), but the latter shows the following differences : 



* While this paper was passing through the press, I have learned that in Alcyonium digUatum "the fine lines 

 which look like capillary tubes really consist of strings or rows of cells" and "have no lumen." The structures 

 described above are probably similar. 



