479 
paratively recently produced a spine of its own (c). This spine 
has its origin against the side of the spine a, and is not an in- 
dependent spine like d, for instance. It may here be men- 
tioned that in all cases of “ forked ” spines there is a secondary 
spine applied to the side of a primary spine, as is the case in 
Cephalodiscus hodgsont (“ Discovery ”” Reports, 1906, p. 51, 
and plate 4, fig. 21). 
The next tube in point of age is that marked e in fig. 9. Be- 
fore the tube had attained more than half its present ‘length, 
the polypide began to secrete a spine (f) by applying to the 
side of the tube first blunt mounds of test, and later steep 
deposits, of the shape of projectiles. Shortly after this another 
polypide settled by the side of tube ¢ and spine /, and secreted 
a tube g, with a spine of its own (d) applied obliquely to the 
tube. Later still, another polypide settled down between the 
spines f and a, and secreted a tube (/), which is still in a com- 
paratively early stage of construction, being short, and pro- 
vided with a delicate projecting peristome. The polypide of 
tube h had. however, at the time when the specimen was 
dredged, already produced a spine 7, which it began to secrete, 
not against the side of its tube as was the case with the poly- 
pides of tubes e and g, but against the side of the spine /, much 
as the polypide of tube b based its spine (c) against the side of 
the spine a. Thus it results that the four tubes are related to 
five spines; but one of the spines (@) was there in the first in- 
stance, and belonged primarily to an older tube, not shown in 
the figure, so that, as far as the group under consideration is 
concerned, there are as many spines as tubes. 
Fig. 10 represents a group of five tubes which I interpret 
as follows. The shortness of the tubes implies that they are 
only recently formed, and the uniform character of the tubes 
that the five polypides settled almost at the same time. Vite 
spine @ is probably a spine which was present before the advent 
of the five polypides in question, and three of the tubes are 
built against it, namely, b, c and d. The polypide of either 
c or d—possibly both—originated a new spine (¢) against the 
side of the initial spine (a); the polypide of 6 seems to have 
appropriated to itself the spine a, perhaps in conjunction with 
the polypide of c; the polypides of tubes f and g have pro- 
duced spines of their own (/ and 4), originating on the sides of 
their tubes. 
Fig. 11 is interesting as an example of a branch-end with 
very little of the common test between the tubes. The con- 
sequence of the deficiency of this material makes the greater 
part of the tubes a and 6 to stand out from the general tubarium 
in the form of “ peristomes,” which circumstance shows that 
