340 
THE VOYAGE OF H.M.S. CHALLEHGEE. 
The internal longitudinal bars in the branchial sac are wide (PI. XLV. figs. 7, 8, ^./.). 
The larger transverse vessels, which bound the rows of stigmata, are provided with 
delicate horizontal membranes, which are exactly like the alternate smaller transverse 
vessels which cross the meshes (PI. XLV. fig. 7, tr. and tr'.). 
The stigmata in the fully developed sac are long and narrow (PI. XLV. fig. 7), and 
much larger than in the other species of the genus ; but in the young sac they are 
shorter and more rounded, and closely resemble those of Goodsiria placenta and Goodsiria 
peduncidata. Figures 5 and 6 show the condition of the stigmata in the younger sacs, 
and figure 6 also shows, in the very young Ascidiozooid, an arrangement of the internal 
longitudinal bars which suggests the presence of three rudimentary folds. This interesting 
indication of a relationship to the branchial sac of Goodsiria placenta is totally lost as the 
Ascidiozooid grows older. The stigmata are very regular in the adult sac, and the 
ciliated cells are well marked (PI. XLV. fig. 8). 
The endostyle is well developed. Its course is straight. In a transverse section of 
tlie ventral part of the body (PL XLV. fig. 16) the structure and relations of the endostyle 
are well seen. Its ventral-most part is continuous with the mantle (m.), and so separates 
the right and left halves of the peribranchial_ cavity (PL XLV. fig. 16, p.hr.c.). The cells 
on the floor of the endostyle bear very long cilia, which in a transverse section project as 
a large tuft (PL XLV. flg. 16, en.), while the sides bear shorter and less conspicuous cilia. 
The oesophagus is short (PL XLV. flg. 17, oe.). It runs backwards and ventrally to 
open into the small globular stomach (st.). There are usually about six well-marked 
folds upon the right side of the stomach. A transverse section (PL XLV. fig. 18) shows 
in addition a single large fold, which projects far into the interior, nearly dividing it 
into two cavities. The intestine is wider than the oesophagus (PL XLV. fig. 17, i.). 
It runs ventrally for a short distance from the stomach, and then turns anteriorly and 
then dorsally in a wide loop, running parallel to the stomach and oesophagus tiU it 
reaches the dorsal edge of the body, where it again turns anteriorly to become the short 
rectum (PL XLV. fig. 17, r.). The anus (a.) has a reflected margin. The stomach is 
provided with a short caecum projecting from its left hand side, and not visible in figure 
17. It curves towards the intestine in a semicircle (PL XLV. fig. 19). 
The polycarps are not very conspicuous, as they are deeply buried in the mantle 
(PL XLV. fig. 13). When dissected out they are seen to be ovoid or flask-shaped bodies 
(PL XLV. fig. 14), with short projecting ducts. They are hermaphrodite, the same 
polycarp containing both ova and spermatic vesicles (PL XLV. fig. 13, o. and t.v.). The 
oviduct and vas deferens are occasionally seen cut in sections in the form of small tubes, 
the former slit-like and the latter circular in section (PL XLV. fig. 13, o.d. and v.d.). 
A few endocarps are also present projecting from the mantle into the peribranchial 
cavity. Tliey are of the usual irregular shapes, with corrugated outlines (PL XLV. fig. 15). 
Although I have not found any buds in the colonies examined, I am inclined to think 
