3<J8 BULLETIN OF THE UNITED STATES FISH COMMISSION. 



lu the oystei' the partitiou between the two openings in the mantle is permanent, 

 the mantle folds being united. Fig. 89, PI. xciii, represents a view of this part of the 

 mantle seen from behind. The mantle folds are fused at hri). Above, the upper part 

 of the cloacal ehaiiiber is exposed where it communicates with the exterior, and the 

 rectum (r) is seen to open into it. Below the fusion, the l)ranchial chamber is seen, 

 together with the posterior ends of the gills {og and ig), which hang in it. 



In Mytilm this region is more modified. The cloacal opening, which is small, 

 appears from the exterior to be much more definite (Fig. 87, PI. xciii, -to) and is sur- 

 mounted by a tougli ring, which, like the whole mantle edge in this region, is colored 

 by a deep-brown pigment. Tentacles are here absent from the mantle edge. Ventral 

 to this, the mantle folds do not shut in a complete branchial orifice. Closing the upper 

 part of the space between the mantle folds on the inside, is a thick, pigmented mem- 

 brane, a part of the mantle {hr m). This is also shown in Fig. 88, which is a pos- 

 terior view of Fig. 87. In this figure the extreme posterior ends of the gills mark the 

 partition between the epibranchial chamber above, whose opening to the exterior is 

 seen at co, and the branchial chamber below. Thus far there is little indication of 

 a development of much tissue in connection with the siphon-like openings other than 

 that usually present in the mantle edge. 



If we imagine that the two separated openings in Ostrca imdMytilns are made 

 definite — that is, that the branchial passage is separated from the ventral mantle edge 

 by a second fusion — and that they are protruded as tubes, we will have in the main 

 the condition in the siphoned forms (Fig. 06, PI. xciv, sn). A vertical, longitudinal 

 section thnmgh the posterior region of the body of Venus (Fig. 93, PI. xciii) will show 

 the relations of most of these parts. Below, and to the right in the figure, is seen the 

 branchial chamber {hr c), in which hang the gills (ig). The upper i)art of the gills, 

 forming the floor of the epibranchial chamber [ej) c), is seen in the section, and the 

 openings of the water tubes into the latter chamber are indicated. Bounding this 

 part of the epibranchial chamber above is the posterior adductor (^;«). The epibran- 

 chial chamber opens into the base of the cloacal sij)hon, which also receives the end of 

 the rectum (r) from above, it having come down over the adductor. This basal por- 

 tion of the siphon may perhaps be called the cloaca, though it is small. The lower 

 or branchial sii)hon opens into the branchial chamber, but at its base there stretches 

 across its whole upper part a membrane {hrm), occupying the same position as that 

 already referred to in Alytilus. If we should take an unsectioned specimen and throw 

 back the mouth folds so as to get a view of the base of the lower siphon from the 

 branchial chamber, we would liave the membrane shown as in Fig. 90, PI. xciii, hr m. 

 Exteiuling down from the posterior end of the gills, it covers all but the low^er part of 

 the base of the branchial siphon seen below in the figure. It does not extend straight 

 across from one side to the other, but presents the appearance of a deep notch extend- 

 ing upward. On either side of it the bases of the thick siphon walls may be seen. 



One might naturally sujipose that fr(mi its position this branchial membrane 

 would allow Mater to enter the branchial chamber, but that any back flow would 

 apply it over the base of the branchial siphon ; this would prevent the water from 

 escaping, thus acting as a valve. If the animal is able or has occasion to use this 

 fold in such a Avay, I have not observed it. Mactra solidissima, the sea clam, much 

 like Venus in anatomical points, has a branchial membrane that apparently completely 

 covers the branchial siphon at its base. But if an individual be quickly taken out of 



