REPORT ON THE TUNICATA. 95 



which form a delicate network, and iu some places the strands of this net arc curiously 

 curled (PI. X. fig. 13). 



I have found in the horizontal membrane (a part of the wall of the branchial sac) a 

 number of small scattered pits, sometimes placed singly, and sometimes united in twos 

 and threes (PI. X. fig. 15). These arc just visible to the eye (PI. X. fig. 9, hr.s.), and look 

 like small openings, but they are really small imperforate sacs, the walls of which are 

 thinner than the rest of the membrane ; and they may possibly l)e the representatives 

 of the perforations or stigmata in the ordinary Ascidian branchial sac. Over the general 

 surface of the membrane is a layer of squamous epithehum (PI. X. fig. 16), the cells 

 of which are rounded or polygonal in form ; while on the margins of the little pits the 

 cells become nearly rectangular in surface view (PI. X. fig. 17), and fusiform in 

 profile (PI. X. fig. 18). Possibly this altered epithelium bordering the depression 

 may represent the line of ciliated cells round the edges of the stigmata in the normal 

 branchial sac. 



The nervous system, the subneural gland, the dorsal tubercle, and the neighbouring 

 muscle bands are shown in fig. 11. The ganglion, which is placed upon the anterior 

 dorsal part of the visceral mass (PI. X. fig. 10, n.g.), is triangular in shape (PI. X. fig. 

 11, ii.g.), and has nerves prolonged from each angle. The subneural gland is a large 

 elliptical mass which extends from the front of the ganglion to half-way up the dorsal 

 tubercle (PL X. fig. 11, s.gl.). It is apparently the spherical thin-walled body taken 

 for a sense - organ by Moseley. The dorsal tubercle has a sj)Oon - shaped opening 

 leading into a narrow duct which is lost on the lower surface of the subneural gland. 

 The neighbouring muscle bands are arranged in a symmetrical manner round the 

 ganglion, as shown in fig. 11. 



A somewhat irregular opaque line is visible on the anterior surface of the bod}^ 

 showing through the clear test (PI. X. fig. 6, en.). It runs from near the branchial 

 aperture in an oblique course ventrally, and to the left side. This is produced by an 

 elevation on the inner surface of the anterior half of the branchial sac, which I believe 

 represents the endostyle of other Tunicata. If this be the case, it does not correspond 

 to the short endostyle described by Moseley as placed upon the nucleus, and in my 

 specimen I failed to find any trace of an endostyle in that position. This may be 

 one of the points of difference if these two specimens belong to distinct species of 

 Octacnertms. In Moseley 's form only the posterior part of the endostyle may have 

 been retained, while in my form only the anterior half is found (see Fig. 11, p. 93, end 

 and end'). 



The visceral mass lies loosely in the interior of the dorsally placed projection of the 

 body (PI. X. fig. 9), and can be easily withdrawn from it. It has occurred to me 

 that possibly Moseley's specimen had been distorted while coming up in the trawl, and 

 that the visceral mass had become displaced. If that be not the case, then the two 



