460 
BULLETIN OF THE BUREAU OF FISHERIES. 
of the mantle. It must be assumed that the carmine was, so to speak, distasteful 
to them, and therefore excluded. This selection was not brought about, however, 
by the palps, but by closure of the valves, usually only after carmine particles had 
touched some portion of the alimentary canal. 
FOOD. 
JUVENILE MUSSELS. 
Considerable knowledge of the materials ingested by fresh-water mussels, 
especially in the juvenile stages, was secured in the course of the observations 
described above relative to the mechanism of ingestion. A survey of the experi- 
ments detailed on the preceding pages shows that, in general, mussels from the 
very earliest moment at which they fall from the fish begin to ingest particles of 
debris and both animal and plant organisms. A number of the mussels experi- 
mented with were embedded and sectioned and photomicrographs made in certain 
instances. These will now be discussed in detail. 
Figure 4 (opp. p. 451) shows a section across the stomach of a mussel 1 mm. 
long. Stephanodiscus, probably Odontidium, and other diatoms appear in it. 
Fragments of other plant forms and debris may also be noted. In other sections 
through this same stomach there were found, besides debris numerous specimens 
of Cocconeis, many Stephanodiscus, several filamentous diatoms probably Odon- 
tidium, and many cells of green algae no doubt from colonial forms such as Volvox 
and Pleodorina. There were also certain yellow bodies, which appeared to be 
pollen grains of some sort. 
Figure 24 (opp. p. 459) is that of a section of the stomach of another mussel 
1 mm. long. The head of the style may be observed projecting into the stomach. 
A portion of a filament of what appeared under the oil immersion lens to be Odomr 
tidium sp. can be seen, together with debris and other fragments. In other sec- 
tions through this same stomach the pollen grains of the ragweed ( Ambrosia artemisii- 
folia) were identified. This ragweed was abundant around the edges of the ponds 
that supplied the water for the rearing troughs from which the mussels were taken. 
Pollen grains of other plants, debris, and fragments of diatoms were also abundant 
in the sections. 
Figure 25 is a section through the stomach of a mussel about 3 mm. long. The 
anterior end of the style appears. Two or three specimens of Stephanodiscus, two 
of the desmid Cosmarium, and probably Odontidium may be seen here. 
Figure 26 is a section through the intestine of the same specimen. Stephano- 
discus, an unrecognizable green alga, some pollen grains, and filaments that may be 
Odontidium, appear. 
In other sections from this specimen, in addition to the forms just mentioned, 
there were found the large diatom Coscinodiscus, Cocconeis, a partly digested Pan- 
dorina colony, Volvox, Scenedesmus, Melosira, and many unrecognizable frag- 
ments of filamentous diatoms, cells of green algae, pollen grains, and debris. 
In the stomach contents (obtained by dissection) of a mussel 1 mm. long were 
found debris, a Navicula 43 micra long, Cocconeis, Cosmarium, and fragments of 
filamentous diatoms. In other small mussels were found broken or partly digested 
