184 _ 
indicated in text-fig. 1, F, by heavy dotting. The cells com- 
posing the dark rim are pigmented cells, similar to those so 
widely spread in the ectoderm of C. mgrescens ; they occur 
only in the marginal region of the dorsal surface of the shield 
(see text-fig. 3, pg.), and in fig. 4, plate 2, and text-fig. 1, F, 
are being viewed through the partially-transparent ventral 
wall of the shield. The cells when seen in bulk with the naked 
eye or with a low magnification appear blackish, but when 
examined with high powers of the microscope they are brown 
or yellow, in polypides which have been killed in formalin ; 
but in sections stained with Erlich’s haematoxylin and eosin 
of polypides killed in osmic acid solution the pigmented cells 
are red, while in polypides killed in Flemming’s fluid and in 
those killed in Perenyi’s fluid the cells are colourless. 
The pigmented cells stand high, and among them are colour- 
less refractive beads (text-fig. 3, 7.b.) similar to those first 
known in the end bulbs of the plumes of C. dodecalophus, but 
which investigation of C. nigrescens shows may occur on almost 
any part of the body. These I take to be the globules of the 
test material which have been accumulating in the cells, and 
coalescing, in the same manner as the globules of mucus in a 
goblet cell. On this assumption, the dorsal surface of the edge 
of the shield is in C. gitlchristi the part of the body which is 
more particularly concerned with the secretion of the tubarium. 
Plumes.—The plumes of C. gtlchristt are six pairs in number. 
Occasionally only ten or eleven plumes can be found, but this 
may be due either to the accidental loss of one or more of the 
plumes, or, more usually, to one or both plumes of the sixth 
pair being retarded in their development. The plumes of 
Cephalodiscus develope successively, in pairs, and not simul- 
taneously ; and the sixth pair, or one of the pair, may have 
failed to develope beyond an early stage at the time when the 
polypide was killed. Whether the plumes in question would 
have become larger had the polypide been allowed to live on 
is doubtful, because such a polypide is of full size, and possesses 
mature gonads. It is much more likely that the rudimentary 
plumes, sometimes no more than pinnule-less, club-shaped 
bodies situated near the bases of the plumes of the fifth pair, 
are permanently arrested. 
The tip of the plume-axis is only in rare cases (text-fig. 2 
B) swollen in a manner reminding one of the terminal bulbs of 
C. dodecalophus and C. hodgsom. The axis in the great major- 
ity of instances terminates in the manner shown in text-fig. 2, 
A and C; the tip is only slightly thicker than one of the pin- 
nules, and it shows no transparent, highly refractive beads. 
A moderately well-extended plume measures about ‘9g mm. 
in length, a well-extended plume r1mm. The maximum 
