244 bulletin: museum of compaeative zoology. 



absence of an epithelial laj'er on the inner surface of the test of the 

 ventral side of the oral disc. I have searched in vain for such an 

 epithelium. I regret the necessity of leaving these points, important 

 to a full understanding of the morphology of this most interesting 

 creature, still obscure. 



We may now examine a little more attentively the other structures 

 belonging to the branchial sac. What I have called the dorsal lamina 

 is really not a lamina at all, nor has it the languets that usually take the 

 place of a lamina when such a structure is wanting. Here we have two 

 irregular, approximately parallel, bands of somewhat thickened epithe- 

 lium, with a narrow interval between them (Fig, 9, d. /.). These are, 

 as compared with the organ in typical ascidians, relatively short, they 

 being but a little longer than the combined length of the ganglion and 

 neuro-hypophyseal gland (Fig. 8, d. L). 



The peribranchial band (Fig. 9,^. b. h.) is also of an unusual character. 

 It consists of a thickened ridge of epithelium, continuous on each side 

 with the corresponding band of the dorsal lamina; the peribranchial 

 ridge being, however, less clearly defined than the dorsal bands. These 

 bands are so irregular in both outline and definition as to defy exact 

 representation in a drawing. 



An area of uniform, thinner epithelium occupies the angle between 

 the diverging peribranchial bands, and in tliis are situated the gang- 

 lion, and gland with its duct. The ganglion is anterior and dorsal 

 to the gland, and from the three large nerves given off from it i. e. 

 a pair extending forward, and a single median one extending backward, 

 is somewhat triangular in form. The nerves of the pair are much, 

 larger than the single posterior nerve. This is correlated with the fact 

 that it is the anterior nerves which supply the oral disc. These nerves 

 can be traced forward along the "horizontal membrane," each giving off 

 branches which go to the arms and musculature of the disc. They are 

 large and elaborately branched, thus showing that the regions supplied 

 by them are Avell enervated. 



A detailed study of the nerves and their terminals, particularly the 

 sensory terminals, would in all likelihood yield interesting results. 

 The gland (Fig. 9, s. gl., Plate 3) is nearly spherical, and as above 

 indicated is situated ventral to, and behind the ganglion. On its ventral 

 surface is a ridge extending somewhat diagonally fore-and-aft. This is 

 in all probability a portion of the duct, though I have not made out 

 with certainty a collection between it and the large thin-walled dorsal 

 tubercle (Fig. 9, d. t.). Tliis tubercle, or hypophysis funnel, opens forward 



