146 



which is situated in the emargination of the bifurcation. The basal surface of the ovicell is 

 only slightly convex; and therefore projects but httle beyond the general level of the basal 

 surface of the branch. It is covered by very numerous pores; which, as is usual in Cyclostomes, 

 are more numerous than those of other parts of the colony. In the frontal view (fig. 1) the 

 distal end of the ovicell appears in the emargination of the branch. lts ooeciostome is a very 

 short tube, which stands vertically, so that the ooeciopore is easily seen in this view. This 

 opening may be described as circular, with one part of the circumference flattened. On its 

 distal side the ooeciostome is in close contact with the peristome of the outermost zooecium 

 of the series next beyond the bifurcation on the left side. In the other ovicells, the ooeciopore 

 is concealed by the corresponding peristome and is thus not visible in frontal view, and it is 

 less circular than in the specimen figured. The ovicell appears to be a modification of one 

 of the kenozooecia of the basal surface of the branch. 



The ovicells of Tervia irregularis have been figured by Waters x ). One of the figures 

 (fig. 5) given by this author, and stated by him to be the normal ovicell of the species, is 

 elongated, and is not unlike that shown in fig. 2 of this Report, except that it commences 

 just on the proximal side of a bifurcation and belongs mainly to one of the branches thus 

 formed. The other (fig. 6) is described as "a short abnormal one"; and is hardly longer than 

 broad; but it has the same relation to the bifurcation that is shown in my own figure. In 

 both cases the ooeciostome appears to be visible in basal view. A similar short ovicell is 

 figured by Jullien and Cal vet, 1903 (t. cit.). 



I have compared the forms described in the present Report with specimens of /. 

 irregularis Menegh. , in the Cambridge and British Museum Collections. Ovicells are ' not 

 present; but while the general arrangement of the zooecia is similar to that found in the 

 Oriental and Australian forms, the specimens differ in having their lateral zooecia developed 

 into strongly projecting wing-like outgrowths of the sides of the branch; each such outgrowth 

 being formed of several (as many as four or five) zooecia, which are in the main connate, but 

 have some of their constituent peristomes free at their ends. The pores are less numerous 

 than in the specimens here referred to the present species. The differences, both of the ovicells 

 and of the zooecia, seem to be sufficiënt to justify the separation of the forms described in 

 this Report from the Mediterranean species. 



M e a s u r e m e n t s , in y. : 

 Width of the branch, at the proximal end of fig. 1, 700 ; 

 Width of the same branch, immediately distally to a row of zooecia which runs transversely 



across the entire branch, just before the bifurcation, 1,550; 

 Diameter of the ooeciopore, 175; 

 Diameter of the zooecia, 150 — 200 ; 

 Diameter of the orifices, 150; 



Length' of the part of the ovicell shown in fig. 2, 2,550; 

 Width of the ovicell, at its distal end, 750. 



i) Waters, A. W., 1888, t. cit., PI. XIV, figs 5, 6. 



146 



