Ove JAMES .L. KELLOGG 
exposed face of the lamella. As on all ciliated surfaces, these _ 
particles are caught and held by mucus. It is a point of prime 
importance that the touch of foreign substances causes the dis- 
charge of mucus, and in quantities proportionate to their volume. 
A few particles cause a slight local flow where they touch, and 
many cause a copious discharge—apparently in all cases just 
enough to hold the stimulating bodies, and prevent their escape. 
This seems to be true of all collecting surfaces, whether gills, 
visceral mass, or mantle. Cilia on filaments forming the bottoms 
and sides of the grooves between folds (fig. 43), cause a move- 
ment of mucus and suspended particles upward toward the base 
of the lamella (fig. 48, mm and fig. 42, g), where a tract leading 
to the palps is met. On the outer faces of the gill folds (/,f), 
however, cilia currents trend in the opposite direction, toward 
the edge of the demibranch. On this edge also there is a groove 
in which particles are carried toward the palps, as is usual in 
lamellibranch gills. Thus on every lamella there are two sets 
of currents, one directed toward its base, and the other toward 
its lower edge. 
It is my conviction, based largely on the palp mechanism, that 
lamellibranchs (except, perhaps, the pardoxical, sand-eating 
‘Macoma) are able to take food into the digestive tract only when 
particles are brought to the gills a few at a time, and this belief 
is strengthened by the operation of the ciliary mechanism of 
the Pecten gill. Experimentally, it is easy to show that when a 
few scattered particles are brought to the gill surface the greater 
number of them will be drawn into the grooves, where, with the 
mucus holding them together, they are taken to the base of the 
lamella. From this region, they have repeatedly been seen to 
move to the oral groove and enter the mouth. The few particles 
falling on the folds are dragged off from them if there is a mucus 
connection with particles in the grooves; or, if not so connected, 
they move downward independently toward the free margin of 
the lamella. Being small in volume, they also are carried on 
this margin to the palps, and often on into the mouth. This I 
believe to be the mechanism in operation in feeding, when dia- 
toms—probably seldom, if ever, extremely numerous—are 
