542 MAYNARD M. METCALF, 



special mention. A nerve of considerable size arising from the anterior 

 end of the ganglion, on the right side, bends down around the great 

 blood sinus which encloses the brain, and seems to be distributed to 

 a part of the epithelium of the ciliated funnel The fibres run to the 

 non-ciliated cells of the upper part of the funnel and here the base- 

 ment membrane of the epithelium is interrupted and the outer ends 

 of the cells are irregular. (Compare with Hcrdmania, Plate 36, Fig. 37.) 



In an earlier paper ^) I stated that there are no ventro-lateral 

 outgrowths from the brain in the solitary Salpa cylindrica. This is 

 not strictly accurate. There are no small-celled ventro-lateral ganglia, 

 but the large-celled ganglia are present as a number of large cells, 

 not gathered into two groups, but stretching across the whole ventral 

 surface of the brain, between it and the chambers of the gland. This 

 is more nearly their arrangement in most of the Ascidians. 



The chain form of this species shows decidedly different features. 

 The neural gland has no chambers. The ducts of the gland, however, 

 are present, though the one on the left is very much shorter than 

 the one on the right. Even the right one is shorter than in any other 

 species, not reaching to the ganglion. It is interesting to note that 

 in this species, which has no enlarged chambers in the gland, the 

 whole of the rudimentary ducts are lined by cubical epithelium in- 

 stead of the usual squamous epithelium. This causes the somewhat 

 enlarged apertures of the ducts to appear at first sight like accessory 

 ciliated funnels. Careful study of well preserved material shows, 

 however, that no cilia are present. I have traced the development of 

 the neural gland in the chain individuals of S(dpa cylindrica and find 

 it agrees in the early stages with that in Cyclosalpa pinnata. At first 

 the pharynx wall is pressed against the ventral surface of the brain. 

 (Cf. Plate 39, Fig. 72, a drawing of C. pinnata.) Later, when it be- 

 gins to withdraw from the brain, there are, for a time, two circular 

 areas where it still adheres to the brain. (Cf. Plate 39, Fig. 73, a 

 drawing of C. pinnata.) Soon, however, these also pull away from 

 the brain, but not until the adjacent regions of the pharynx wall 

 have become pulled out into tubes. These tubes persist in the adult, 

 but have no discoidal enlargements at their inner ends. In the young 

 buds the pharyngeal epithelium is composed of cubical cells. As the 

 pharynx enlarges the cells over the greater portion of its surface 

 become flattened into pavement cells. In the developing ducts of the 



1) Metcalf, 1893«. 



