very unsatisfactory.  E.L.  Thorndike[1] has demonstrated the ability of fishes to learn a labyrinth path.  In his report no statement of the time required for the formation of habit is made, but from personal observation I feel safe in saying that they did not learn more quickly than did the frogs of these experiments.  Norman Triplett[2] states that the perch learns to avoid a glass partition in its aquarium after repeatedly bumping against it.  Triplett repeated Moebius’ famous experiment, and found that after a half hour’s training three times a week for about a month, the perch would not attempt to capture minnows which during the training periods had been placed in the aquarium with the perch, but separated from them by a glass partition.  Triplett’s observations disprove the often repeated statement that fishes do not have any associative processes, and at the same time they show that the perch, at least, learns rapidly—­not so rapidly, it is true, as most animals, but more so in all probability than the amphibia.

   [1] Thorndike, Edward:  ‘A Note on the Psychology of Fishes,’
   American Naturalist. 1899, Vol.  XXXIII., pp. 923-925.

   [2] Triplett, Norman:  ‘The Educability of the Perch,’ Amer. 
   Jour.  Psy., 1901, Vol.  XII., pp. 354-360.

The only quantitative study of the associative processes of reptiles available is some work of mine on the formation of habits in the turtle.[3] In the light of that study I can say that the turtle learns much more rapidly than do fishes or frogs.  Further observations on other species of turtles, as yet unpublished, confirm this conclusion.

   [3] Yerkes, Robert Mearns:  ’The Formation of Habits in the
   Turtle,’ Popular Science Monthly, 1901, Vol.  LVIII., pp.
   519-535.

For the frog it is necessary to measure and calculate the improvement in order to detect it at first, while with the turtle or chick the most casual observer cannot fail to note the change after a few trials.  In connection with the quickness of the formation of associations it is of interest to inquire concerning their permanency.  Do animals which learn slowly retain associations longer? is a question to which no answer can as yet be given, but experiments may readily be made to settle the matter.  I have tested the frog for permanency, and also the turtle, but have insufficient data for comparison.

3. Sensory Data Contributing to the Associations.—­Among the most important of the sensory data concerned in the labyrinth habit are the visual impressions received from the different colored walls, the slight differences in brightness of illumination due to shadows from the partitions and the contrast in form of the two sides of the labyrinth resulting from the use of the partitions, and the muscular sensations dependent upon the direction of turning.  The experiments proved beyond question that vision and the direction