TABLE V. ASSOCIATION OF POSITION: GENERAL DISTRIBUTION OF CHOICES
| Choices of boxes 1 to 6 when food was placed in boxes 2, 3, and 4 | ||||||||||||||||||
| Food in box 2 | Food in box 3 | Food in box 4 | ||||||||||||||||
| Boxes | Boxes | Boxes | ||||||||||||||||
| Animals | 1, | 2, | 3, | 4, | 5, | 6. | 1, | 2, | 3, | 4, | 5, | 6. | 1, | 2, | 3, | 4, | 5, | 6. |
| (B) | 1 | 18 | 4 | 6 | 1 | 0 | 1 | 1 | 8 | 7 | 8 | 5 | 0 | 0 | 2 | 21 | 4 | 3 |
| (C) | 2 | 20 | 6 | 1 | 1 | 0 | 0 | 9 | 19 | 2 | 0 | 0 | 0 | 7 | 2 | 15 | 6 | 0 |
| (E) | 4 | 18 | 2 | 3 | 2 | 1 | 0 | 6 | 18 | 3 | 3 | 0 | 2 | 6 | 2 | 15 | 3 | 2 |
| (F) | 5 | 20 | 1 | 2 | 2 | 0 | 1 | 6 | 18 | 3 | 2 | 0 | 0 | 2 | 5 | 14 | 8 | 1 |
| (G) | 0 | 18 | 5 | 2 | 2 | 0 | 0 | 3 | 13 | 5 | 7 | 2 | 0 | 0 | 3 | 17 | 9 | 1 |
| (H) | 0 | 22 | 4 | 3 | 1 | 0 | 1 | 0 | 18 | 4 | 6 | 1 | 0 | 4 | 8 | 15 | 3 | 0 |
| (I) | 4 | 18 | 3 | 1 | 3 | 1 | 0 | 1 | 23 | 6 | 0 | 0 | 0 | 6 | 7 | 15 | 2 | 0 |
| (J) | 1 | 17 | 7 | 5 | 0 | 0 | 3 | 0 | 21 | 3 | 2 | 1 | 1 | 4 | 3 | 16 | 6 | 0 |
| Total, | 17 | 151 | 32 | 26 | 12 | 2 | 6 | 26 | 138 | 33 | 28 | 9 | 3 | 29 | 32 | 128 | 41 | 7 |
TABLE VI. ASSOCIATION OF POSITION: DISTRIBUTION OF RIGHT CHOICES
| Choices from series 1 to series 5 in the case of boxes 2, 3, and 4 | |||||||||||||||
| Box 2 | Box 3 | Box 4 | |||||||||||||
| Series | Series | Series | |||||||||||||
| Animals | 1, | 2, | 3, | 4, | 5. | 1, | 2, | 3, | 4, | 5. | 1, | 2, | 3, | 4, | 5. |
| (B) | 4 | 4 | 5 | 2 | 3 | 2 | 2 | 1 | 1 | 2 | 4 | 4 | 4 | 5 | 4 |
| (C) | 5 | 4 | 2 | 5 | 4 | 4 | 4 | 4 | 4 | 3 | 3 | 3 | 2 | 2 | 5 |
| (E) | 3 | 4 | 4 | 3 | 4 | 4 | 4 | 2 | 5 | 3 | 3 | 2 | 2 | 4 | 4 |
| (F) | 5 | 4 | 3 | 5 | 3 | 5 | 3 | 3 | 4 | 3 | 0 | 2 | 4 | 4 | 4 |
| (G) | 3 | 4 | 4 | 3 | 4 | 2 | 2 | 4 | 2 | 3 | 3 | 3 | 2 | 4 | 5 |
| (H) | 4 | 4 | 4 | 5 | 4 | 4 | 3 | 4 | 5 | 2 | 3 | 3 | 4 | 3 | 2 |
| (I) | 1 | 4 | 3 | 5 | 5 | 4 | 4 | 5 | 6 | 4 | 2 | 1 | 3 | 4 | 5 |
| (J) | 4 | 5 | 4 | 3 | 1 | 3 | 4 | 6 | 4 | 4 | 3 | 3 | 3 | 4 | 3 |
| Total, | 29 | 33 | 29 | 31 | 29 | 28 | 26 | 29 | 31 | 24 | 21 | 21 | 24 | 30 | 32 |
| Average for the three boxes, 26, 27, 27, 31, 28. | |||||||||||||||
The method of learning in these position tests was the same as that noticed in previous experiments, namely, building upon chance successes. When first admitted to the large box containing the row of small ones at the farther end, the animal accidentally found the receptacle containing the food, and later associated the movements involved in reaching that position with various sense-impressions of the box, especially those experienced upon entering—certain tactual impressions of the small entrance compartment, sound of the lifting door and sight given of the interior of the large box.
While the results clearly indicate that pigeons readily learn the position of objects, nothing is proved as to "counting." Some experimenters speak of similar trials as "number-tests," just as they do of "form-tests," but this is probably going too far. To investigate counting in animals, experiments should be arranged which minimize spatial responses. These tests certainly show that pigeons can discriminate positions readily, especially toward the ends of the group, but little more is certainly indicated. Porter[210] says: "If we do not find in birds the power to count, we have in their nice sense for the location of a member of a series ... something of that preliminary number-sense which Ribot describes as belonging to children and savages."
B. Color Tests
To investigate the animals' ability to utilize colors[211] in finding their food, I employed the same apparatus as before, except that six boxes were used throughout and each was covered with paper of a different color: red, yellow, green, blue (Bradley's standards, except red, RO being substituted), gray, and black. The boxes covered with black and gray paper were employed merely to complete the group of six. The same method as before was employed, except that the board to which the boxes were attached was left stationary at the end of the large box, and also that the position of all six boxes was changed irregularly for each test.
The general behavior of the animals at the beginning of these tests was quite similar to that shown in the preceding experiment; but it was soon evident that colors occasioned them far more difficulty than positions. The general distribution of choices is given in Table VII. It will be seen that the proper box was usually chosen more often than any one of the empty ones, but never oftener than the other five combined, as occurred in the position tests; also that in the case of each color there were instances in which another color was as often, or more often selected. Yet it is clear that colors may serve as valuable sense-data for these animals. In the first series of six tests (see Table VIII) there were few right choices or none, but in each succeeding series the number increased. The learning process was evidently of the same type as before observed (selection, in this case gradual, of chance but useful movements), and involved visual data largely.