222 INTELLIGENCE IN LOWER VERTEBRATES 
augmented if the tube was placed in a different part of the 
aquarium or by feeding the fish with a few of the worms. 
That fish readily associate certain stimuli with food has 
been shown by several observers. Herrick found that 
catfish which at first would seize pieces of cotton which were 
brought in contact with its barbels would gradually cease 
to react to them. After a few trials the fish would make a 
movement toward the cotton, but it was soon checked. 
Finally the cotton could be rolled over the barbels or other 
parts of the body without eliciting any response. The 
fish prohably recognized the cotton through the sense of 
touch as it made little difference if, at any time, red cotton 
was substituted for white. Washburn and Bentley in their 
experiments on the creek chub showed that this fish is able 
to associate different colors such as red and green with food. 
The chub in one set of experiments was fed out of a pair of 
red forceps which were let into the water alongside of a green 
pair. Although a meal worm was placed in the red pair the 
fish often snapped at the green forceps first, but after a num- 
ber of trials the green forceps were ignored. When the fish 
. had been given a number of worms the forceps were put in 
empty. The fish continued to snap at the red pair and 
avoid the green regardless of their relative positions. Red 
and green forceps of different degrees of brightness were 
used and precautions were taken to eliminate the in- 
fluence of odor, but in the last 40 experiments in which no 
food was placed in either forceps and the relative position 
of the forceps exchanged after each trial the fish snapped at 
the red pair every time. It is probable, therefore, that the 
chub is able to distinguish colors as opposed to mere differ- 
ences in brightness—although the proof of this is not en- 
tirely conclusive—and it is very evident that it is able to 
form associations with a fair degree of rapidity. 
