468 
BULLETIN OF THE BUREAU OF FISHERIES. 
material present by means of the groove mentioned. Grave states that in one 
experiment with oysters after three hours the stomach was so filled with “sediment” 
that he could not estimate the number of diatoms. It would seem that his experi- 
ments went to prove that lamellibranchs feed in heavy suspensions of silt, but that 
they do not accomplish this by a selection of one kind of material from another. 
9. In the tests with carmine there was evidence of rejection of material. Sooner 
or later many of the mussels ceased to ingest carmine. The adults rarely took any. 
Rejection, however, was not made by any selective action of the palps. From the 
observations it appeared that the rejection was due to a stimulation (perhaps by 
the sense of taste) of the alimentary canal by the carmine, followed by a closure of 
the valves. Material was thus excluded from the mantle chamber. 
10. Material passed through the alimentary canal in from one to five hours in 
mussels from 2 to 25 mm. long, the length of time being roughly proportional to the 
size of the mussel. 
11. In the mussels about 0.2 to 0.25 mm. long, which have just dropped from 
the fish, the ingested particles are, of course, very small — only a few micra in size. 
The material is essentially the same, however, as that for the larger juveniles — 
Protozoa, diatoms, and minute particles of detritus. Mussels of a length of 1 mm . 
were found to ingest Euglenas measuring about 60 by 18 micra when elongated and 
about 25 when contracted. Specimens 2 mm . long swallowed red Euglenas 160 
by 35 micra. Mussels from 3 to 4 mm. long ingested sand grains as large as 30 
micra across. 
12. The food of fresh-water mussels from the moment of falling from the host, 
throughout life, consists (aside from what is probably a relatively small amount of 
dissolved material) of microscopic animal and plant forms and debris or detritus 
resulting from the decay and disintegration of such forms. Along with this material 
everything else small enough to be admitted to the esophagus, not active enough to 
escape, and not possessing what might perhaps be called a “chemical or disagree- 
able taste” (as carmine no doubt does), is also ingested. Even these last sub- 
stances are often ingested as carmine, as stated above. From this heterogeneous 
material the alimentary canal digests and absorbs what it can and the rest passes on. 
Entire diatoms with color, contents, and nucleus are found in the rectum (fig. 18) 
and in the feces. Broken and partly disintegrated plant forms, however, are found 
in abundance in the alimentary canal. Some of these are of sufficiently fresh 
appearance to warrant the assumption that they were entire when ingested. 
Protozoa, since they lack the cellulose walls, are no doubt more easily handled. 
In the experiments with the red Euglenas it was shown that these forms disinte- 
grated in the alimentary canal, giving the appearance of having been acted on by the 
digestive fluids. It would appear that a very important food element is the organic 
remains in suspension in the water, especially abundant in ponds or rivers in which 
a luxuriant plant life is constantly going through a process of disintegration. 
13. In case it is desired to rear young mussels from the time they drop from 
the fish it would appear to be necessary, as far as food is concerned, only to arrange 
ponds, uncontaminated by sewage or stock, and place in them some of the common 
water plants and algae. The requisite diatoms, Protozoa, etc., will appear and 
