248 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 



that it will probably be a long time before the matter can be set straight 

 in general zoology ; but it should be recalled that the suggestion was 

 made with such reservation as would be expected from so careful and 

 experienced a zoologist as Herdman, on a point concerning which there 

 was at the time such imperfect knowledge. " On the whole," are his 

 words, " I regard this form as being allied to SaljM." It would seem 

 that this conclusion was based chiefly on the supposed correspondence 

 of the visceral mass to the "nucleus" in Salpci ; and the supposition 

 that the endostyle of Odacnemus is located in the floor of the cavity, 

 taken to be branchial, within the oral disc. Had Herdman found the 

 true branchial sac with the stigmata and endostyle, it is quite certain he 

 would not have suggested the kinship of the animal to Salpa. With 

 the information at his command, his conclusion was justified. 



As to exactly what genus among the simple ascidians Odacnemus has 

 most in common it is not yet possible to say. Certain it is, though, that 

 there is nothing to support the conjecture of Moseley that it is related to 

 Cystinr/ia. Metcalf's suggestion that it is related to the Clavelinidae 

 has perhaps as much in its favor as any that can now be made. I would 

 point out, however, that the branchial sac, in particular, suggests the genus 

 Hypohythius of Moseley. This genus alone shares with Clavelina and 

 some of its nearest allies, the character of having a branchial sac without 

 folds or internal longitudinal bars ; besides this its stigmata are irregular 

 in size and distribution. In this latter particular it seems that Odac- 

 nemus resembles Hypohythius quite decidedly. The stigmata of Odacne- 

 mus are perhaps too few in number to warrant the assertion that they 

 are irregular in both respects. They are certainly so as to size ; and 

 there are in Odacnemus neither folds nor internal vessels or papillae. 

 The simplicity of the digestive tract of Hypohythius and its close approx- 

 imation to the side of the branchial sac are likewise points of resemblance 

 to Octacnemus. It must be noted, however, that the stomach-intestine 

 0^ Hypohythius rests on the dorsal side of the branchial sac, while in 

 Octacnemus it is ventral and dextral. I do not think it worth while to 

 make much of the comparison between these two genera, our knowledge 

 of both being still too imperfect, but one other point may be referred to. 

 Moseley's original description of Hypohythius indicates that its oral sur- 

 face is decidedly flat, and that the atrial orifice is far to one edge, if indeed 

 not beyond, this disc. With 0. 2^(it<xgoniensis in mind it is not difficult 

 to imagine a disc like that of Hypohythius to be a starting point for the 

 production through modification, of a tentaculated disc, first like that of 

 0. patagoniensis, and finally like that of 0. hythius and 0. herdmania. 



