CILIARY MECHANISMS OF LAMELLIBRANCHS 627 
face itself, hence errors may easily arise from this. A very 
fine black sand obtained in Puget Sound and elsewhere, was more 
often successful, sometimes revealing the presence of ciliation 
that carmine would not show at all. In demonstrating the 
fact that there is no selection or separation of food from other 
water-borne substances, but that the volume of material alone 
determines the courses that it will take, masses of diatoms 
gathered by a silk tow-net, were used. 
Broad currents were followed by means of hand lenses; minute 
streams, such as those in the grooves between palp folds, often 
being demonstrated with a compound microscope. Some of 
these latter, however, were clear enough to be seen by the un- 
aided eye. ‘Tissues, especially palps, were sometimes removed 
and pinned out in a dish for observation. Most of these studies 
were made without the aid of the binocular dissecting micro- 
scopes which have recently appeared, and have proved to be 
ideal aids in work of this nature. 
Difficulties. The apparent ease of observing currents may 
lead to carelessness. Water-borne particles often appear to be 
cilia-borne. There is a great amount of variation in the inten- 
sity of action on the same region in different individuals. The 
greatest care of the living material having been observed, even 
the mantle, in certain specimens, will show little or no movement, 
while it will be perfectly clear in others. The palps present 
many difficulties. In Unio, for example, material ultimately 
reaching the mouth from the distal oral groove, must pass 
through a notch to reach the lateral groove. Usually it will 
be seized here by outgoing currents on the palp margins and 
carried to the palp tips, and it is only now and then that it may 
be followed onward in the groove. 
Most of the currents to be described have oe determined by 
scores of observations, but of necessity, a few have been followed 
much less often. Possibly the errors are mostly those of omission. 
Dorsally and ventrally directed currents on palp grooves, for 
example, were clearly understood only after most of the forms 
had been examined, and it has not been possible to re-examine 
them all with a view of finding these currents. 
JOURNAL OF MORPHOLOGY, VOL. 26, No. 4 
