40 THE VOYAGE OF H.M.S. CHALLENGER. 



fined to the anterior or superior surface of a branch. In much branched species, in 

 which the stem and main branches are relatively thick, the zooids situated on these 

 portions of the coral! um have a rounded outline, with the tentacles radiate and equidis- 

 tant. The zooid is short and the peristome and tentacles are the only portions of it which 

 project beyond the ccenenchyma. Such zooids are, however, usually not numerous. The 

 majority are situated in single linear series on the medium sized branches, branchlets, and 

 pinnules. In these parts the zooids are somewhat elongated in the transverse axis, so 

 that the long axis of the zooids corresponds with the axis of the branchlet or pinnule on 

 which they are situated. The stomodaeum is elongated in the opposite direction, the 

 mouth occupying the sagittal axis. Usually the shape is more or less rectangular, and the 

 difference in length between the sagittal and transverse axes is not great. The elongation 

 is, however, sufficiently important to bring about a change in the position of the tentacles, 

 by which the radial arrangement is lost. In most cases the tentacles become arranged 

 in three pairs, forming two longitudinal rows of three each, parallel to the axis of the 

 branch. There is a tentacle at each end of the sagittal axis as usual. The other pairs 

 consist of a tentacle on each side of the mouth, the two pairs being close together in 

 elongate polyps. These may be spoken of as the lateral pairs of tentacles ; they limit 

 the long axis of the zooid. The two pairs of lateral tentacles are always inserted into 

 the peristome. The sagittal tentacles, on the other hand, appear to vary somewhat in 

 position, and arise partly from the body-wall. Thus in a side view of a row of zooids 

 on a pinnule, the middle tentacle of each zooid appears to arise from a point nearer 

 to the axis than the others. In young zooids the bilateral arrangement of the tentacles 

 is often not well marked, and all seem at first to share the radiate outline of those situated 

 on the thicker portions of the corallum. With increase in size a more or less well- 

 marked bilateral arrangement of the tentacles is brought about. In extreme cases the 

 tentacles form two straight rows, one on each side of the median transverse axis. 

 Perhaps with regard to their relations to the axes of the zooid, the rows of tentacles 

 would be more correctly defined as "anterior" and "posterior" instead of longitudinal. 

 It is to be noted that in all the species observed having a zooid referable to this type, 

 the elongation of the body in the transverse axis is not pushed so far as to isolate the 

 tentacles of a row. In most cases they are quite as close together as they would have 

 been if arranged radiately. In transverse vertical sections the mouth is seen to open on 

 a prominent oral cone, from the base of which the ectoderm courses out horizontally for 

 a little distance and then becomes rapidly depressed towards the axis, quickly rising again 

 to commence the outline of the next zooid. In some cases the zooids are more isolated, 

 but are rarely more than one diameter apart. The mesenteries in this genus are ten in 

 number, all of which behave in precisely the same manner as those of Cirripathes. The 

 relative breadth of the primary mesenteries necessarily depends on the shape of the zooid 

 and the length of the mouth. The reproductive organs are developed on the transverse 



