H. JAMES-CLARK ON THE AFFINITIES OF SPONGES. 319 



In a very few minutes the two resultants were totally separated and divergent from each 

 other at a sharp angle, and in less than half an hour after the last time noted they had 

 assumed the proportions of the other members of the colony. Shortly after the investiga- 

 tion of the phase just described, the last stages of self-division of another body, belonging 

 to the same colony, were observed, and thus the group, which within two hours before con- 

 sisted of five individuals, was increased to eight (fig. 7). It seems to be a rare occurrence 

 that so many bodies remain long together, since it very seldom happens that more than four 

 or five ffig. 8) are found in a colony ; and now and then, in such instances, I have seen an 

 individual drop off and swim away. When we meet with them settled down upon some 

 point, amidst others which have scarcely any stem, and those which are seated on very short 

 peduncles, it becomes perfectly clear that they are there for the purpose of secreting a new 

 support from the posterior end. 



§ 7. Salpingoeca, 1 nov. gen. (S. gracilis, nov. sp.) 

 (Plate IX., figs. 38, 39.) 



The difference between this genus and Codonceca has already been pointed out. It might 

 well be compared to a stemless Codosiga (§ 6) enveloped in a sheath. I have met with 

 three quite diverse species of this genus, of which that under present consideration and 

 another (S. amphoridium, nov. sp. § 9) are fresh-water denizens, and the third (C. marinus, nov. 

 sp. § 8) is a marine inhabitant. S. gracilis (figs. 38, 39) was found upon only one occasion, 

 and then in an old aquarium, which could not be said to be in a perfectly healthy condi- 

 tion, although its contents were by no means putrid. 



The body is yellow and has a cylindrical shape, about four times longer than broad, nar- 

 rowed and rounded behind, and rounded-truncate in front. Like Codosiga it bears a filmy, 

 membranous, colorless collar (b), which is attached to the extreme edge of the frontal area(/r), 

 and arises to a height which is equal to two thirds the length of the body. The outline of the 

 collar is cylindrical, as a general thing, and truncate at the distal end, but still is subjected to 

 various degrees of momentary change. Unless it be by means of the vibrations of the fla- 

 gellum, there is no other immediate agent which can be supposed to move the body up 

 and down in its sheath. There is no visible movement in itself, like creeping, to be ob- 

 served, and moreover the body progresses so quickly, when changing its place in the calyx, 

 that it becomes evident that it is not due to any reptant mode of transposition. When 

 withdrawn (fig. 38) into the basal, tapering portion of its calyx, the collar (b) does not 

 extend beyond the rim (c 1 ) of the latter, but on the other hand the body occasionally moves 

 so far in the opposite direction (fig. 39) that nearly the whole of the collar (b) projects 

 outside of the dormitory. 



The flagellum is a delicate filament, which arises from the axial point of the front and 

 projects a short distance beyond the edge of the collar. It presents a constantly undulating 

 aspect, and vibrates from base to tip. 



The mouth, we are obliged to presume, as we did in regard to Codosiga, lies somewhere 

 about the base of the flagellum. Abundant digestive vacuoles were observed, as well as 

 loose particles of food, in various parts of the body ; but at no time were we so fortunate 

 as to see the introception of nutritive material, or the ejection of fiscal matter. 



1 oakmyi, a trumpet ; okcu, to inhabit. 



