188 MANTON COPELAND 
both with and without its tentacles, and it soon became evident 
that the animal was in some way testing the stimulating material 
by means of this organ. By using great care in the application 
of the stimulating fluid to the foot processes of snails deprived 
of their tentacles, the proboscis reaction was seen to occur when 
the siphon clearly entered the fish juice, whereas when the siphon 
was not depressed or swung far enough to reach the extract, 
the reaction did not take place. In one instance, for example, 
the extract was twice applied to the foot when the siphon was 
well raised in the water. Each time the foot process contracted, 
and the siphon was swept downward in the direction of the 
extract; but not reaching it, there was no protrusion of the pro- 
boseis. On the third trial the siphon was depressed farther, 
entered the juice and the proboscis reaction followed. Although 
sea water and carmine sometimes caused a contraction of the 
foot process, there was no well marked downward movement 
of the siphon such as occurred in the tests with fish juice. 
The dorsal surface of the siphon was next tested. When 
touched with cotton, placed in the end of a pipette and flooded 
with fish extract, the siphon was bent backward and upward 
toward the cotton. This movement was often accompanied 
by a lifting of the tentacles and anterior end of the foot, as if the 
snail were stretching upward to reach the stimulating material. 
To a pure tactile stimulus the snail usually responded by con- 
tracting the siphon somewhat, but much less tendency was 
shown to bend it upward. In ten trials the proboscis was ex- 
tended but twice, and in one of these cases at least an excess of 
fluid was applied. Another individual did not respond to 
stimulation of the dorsal surface of the siphon, but showed ten 
proboscis reactions in the same number of trials after stimulation 
of the ventral surface of the organ by the same method. When, 
however, cotton flooded with sea water was applied to this region, 
either no response was noted or there was a tendency shown to 
draw the organ away from the cotton. Since the ventral wall 
of the siphon is split its entire length, it was impossible to de- 
termine by this procedure whether the proboscis thrust was 
evoked by stimulation of the under part of the siphon, or of some 
