EEPORT ON THE TUNICATA. 
OT^ 
li O 
The Ascicliozooids are visible all over the upper surface of the colony, except on the 
projecting margin. Their branchial apertures form minute depressions which show as 
dark dots (PL XXXVII. fig. 9). The thorax is very much larger than the abdomen, and 
is placed vertically in the colony (PI. XXXA II. fig. 10). It usually extends about 
three-fourths of the way across from the upper to the lower surface. The abdomen 
does not extend behind the thorax, but lies at right angles to it on the dorsal edge 
(see PL XXXVII. fig. 10). This condition of parts is not at all related to that found in 
the family Botryllidse, where the abdomen is also placed alongside the thorax in place of 
extending behind it. In the present species, and some other Didemnidse, the abnormal 
position of the abdomen has probably been produced by the thinness of the colony, which 
must have been derived from a much thicker mass, and then became gradually converted 
into an incrusting film. As the colony became thinner and thinner, the Ascidiozooid 
would gradually come to have the abdomen bent at a greater and greater angle to the 
original antero-posterior axis of the body, until it reached the extreme condition found 
in the present species, where it does not extend beyond the posterior end of the thorax, 
but runs outwards at right angles to it. Any further bending after this angle had been 
reached would be of no advantage. 
In the Botryllidse, the alimentary canal lies on one side of the body, not on the 
dorsal edge, and is, moreover, placed close alongside the branchial sac, so as to be covered 
by the same layer of mantle. In the present species, on the other hand, the alimentary 
canal is widely separated from the thorax, not only by the mantle, but also by a large 
projection of the common test which lies in the angle betw^een the oesophagus and the 
posterior end of the dorsal edge of the branchial sac (PL XXXVII. fig. 10). 
The test is so hard and brittle that it feels like a layer of solid calcareous matter. 
Sections show that it is very full of spicules, which are specially abundant on the lower 
surface of the colony (PL XXXVII. fig. 10, l.s.), and form a very dense band in which 
no test matrix is visible. The spicules are more closely placed in this region of the 
colony than in any other Leptoclinid which I have examined. The part of the test 
where the spicules are least crow^ded is the central layer, and they gradually become 
denser as either surface is approached. On the upper surface, however, they are not 
nearly so abundant as on the lower (PL XXXVII. fig. 10). 
In the central layer of the colony the test cells and the individual spicules are most 
clearly seen (PL XXXVII. fig. 11). The test cells (t.c.) are small and are generally 
rounded or fusiform ; a few of them have developed small vacuoles, but no true bladder 
cells are present. The spicules vary greatly in shape. Some are of the regular stellate 
form, with tapering rays and sharp apices (PL XXXVII. fig. 11, sp'), while others have 
the rays thicker and blunter. In many cases the rays are reduced to rounded knobs 
(PL XXXVII. fig. 11, sp.), so that the spicule as a whole may become a mammilated 
sphere. A few of the spicules are asymmetrical, and have one half of the surface 
(ZOOL. CHALL. EXP. PART XXXVIII. 1886.) Pp 35 
