eitter: octacnemus. 247 



is obviously somewliat to the right and ventral side of the branchial 

 sac. It is certainly not in the position occupied by it in the animal 

 described by Moseley. Herdnaan gives us no information on this point 

 for his specimen, and I therefore conclude that he did not see it, and 

 assume this to be another particular in which the species studied by 

 him and myself differs from that studied by Moseley. 



The position of the anus in this species is of special interest since the 

 location of it would throw some light on the question of the atrium. 



The position and character of the gonads are indicated in Figures 3 

 and 4, Plate 2. The ovary is a rather voluminous mass applied closely 

 to the posterior border of the digestive tract. In two of the specimens 

 the ova are numerous and apparently near maturity. Tliey are quite 

 spherical, and measure about .32 mm. in diameter. The characteristic 

 ascidian " test " cells are present and make a layer of considerable thick- 

 ness, though it is not uniform over the entire egg. The testis, much 

 less voluminous than the ovary, is situated at the left end of the ovary, 

 closely applied to it, and also to the digestive tract. It is of a lighter 

 color than the ovary, and is divided into numerous small rounded lobes. 



I am unable to find an oviduct, and believe that none exists. The 

 ova probably escape by dehiscence. What appears to be a sperm duct 

 runs forward fur a short course closely applied to the ventral intestinal 

 wall (consequently not visible on any of the figures). The branched 

 strand shown in Figures 3 and 4, Plate 2, crossing the concavity of the 

 ovary toward the right, appears to be a mantle fold, pi-obably serving as 

 a ligament to hold the ovary in place. The possibility of its belonging 

 to the blood vascular system naturally suggests itself; but it certainly 

 has nothing to do with this system. I have, however, seen nothing of 

 either heart or blood vessels. No " liver " or excretory organ appears 

 to be present, nor has the chyliferous organ been found. 



From the fact that the Octacnemus patagoniensis of Metcalf seems to 

 propagate by budding, I have naturally looked with care for evidence 

 of such a mode of propagation here ; but none has been found. I do 

 not believe it occurs, and this it seems to me is one Aveighty reason for 

 holding that the species studied by Metcalf should be regarded as 

 generically distinct from the animal named Octacnemus by Moseley, 



We may turn now to the question of the wider affinities of the Octac- 

 nemidae. The present investigation makes it obvious, as Metcalf had 

 already furnished ample reasons for believing, that they are not related 

 to Salpa, but to the simple or colonial ascidians. Herdman's sugges- 

 tion that their relationship is with Salpa has been so generally accepted 



