Chapter XXX: Section IV: , 1. The experiments of Table IX, C, repeat those of A with (10)

_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.

TABLE VII. COLOR ASSOCIATION. GENERAL DISTRIBUTION OF CHOICES

_Choices of all 6 boxes when food was placed in red, yellow, green,
or blue boxes_

_Food in Red Box_ _Food in Yellow Box_
_Animals_ R, Y, G, B, B'k. G'y. R, Y, G, B, B'k, G'y.
(B) 12 6 7 3 2 0 2 12 6 7 0 3
(C) 15 8 1 4 2 0 3 12 1 6 3 5
(E) 11 8 5 3 3 0 3 10 5 3 5 4
(F) 7 5 9 6 2 1 6 6 5 6 3 4
(G) 9 1 7 5 3 5 5 11 6 3 2 3
(H) 13 3 5 3 3 3 5 9 6 3 3 4
(I) 11 5 2 4 6 2 4 10 5 4 3 4
(J) 10 0 6 8 4 2 5 10 3 3 5 4

Total, 88 36 42 36 25 13 33 80 37 35 24 31

_Food in Green Box_ _Food in Blue Box_
_Animals_ R, Y, G, B, B'k, G'y. R, Y, G, B, B'k, G'y.
(B) 4 5 12 4 5 0 1 3 5 10 6 5
(C) 3 4 14 5 1 3 3 1 2 13 2 9
(E) 1 9 7 11 0 2 3 5 5 10 3 4
(F) 0 8 9 6 0 7 0 3 6 11 5 5
(G) 2 5 16 2 4 1 5 5 4 5 4 7
(H) 1 3 15 5 6 0 7 5 5 6 4 3
(I) 0 4 15 6 2 3 5 5 3 9 5 3
(J) 4 1 12 9 1 3 6 4 4 7 5 4

Total, 15 39 100 48 19 19 30 31 34 71 34 40

TABLE VIII. COLOR ASSOCIATION. DISTRIBUTION OF RIGHT CHOICES

_Choices from series 1 to series 5 in the case of red, yellow, green,
and blue boxes_

_Red Box_ _Yellow Box_ _Green Box_
_Animals_ 1, 2, 3, 4, 5, 1, 2, 3, 4, 5, 1, 2, 3, 4, 5,
(B) 0 1 2 4 5 1 2 3 2 4 1 2 2 4 3
(C) 2 2 4 3 4 1 2 3 3 3 3 2 3 4 2
(E) 2 1 2 3 3 1 0 3 2 4 0 1 1 1 4
(F) 0 1 1 2 3 0 2 2 1 1 0 1 2 3 3
(G) 0 2 3 2 2 2 2 1 3 3 2 4 3 3 4
(H) 1 1 3 4 4 0 2 2 2 3 2 3 4 3 3
(I) 1 2 3 3 2 1 2 2 3 2 1 3 3 3 5
(J) 1 2 2 2 3 1 2 1 2 4 1 1 3 3 4

Total, 7 12 20 23 26 7 14 17 18 24 10 17 21 24 28

_Blue Box_
1, 2, 3, 4, 5,
1 3 1 2 3
3 2 2 3 3
0 1 3 2 4
1 1 2 4 3
1 1 0 0 3
0 1 2 2 1
0 2 2 3 2
1 2 1 1 2

7 13 13 17 21

There is no evidence that the color-preference of the animals assisted them in choosing correctly, in fact, they were rather less successful in dealing with those colors for which they had previously shown decided preference,[212] since the whole number of right choices was less in the case of the green and blue boxes (85) than in the case of the red and yellow ones (92), and since there was a relative diminution in the rate of learning toward the last in case of the former boxes.[213]

To test the animals' ability to discriminate shades of colors in finding their food, two birds were used, with four boxes, each covered with a different shade of red paper, and two with the boxes covered with green paper. The brightness of the different shades was not measured, but to the eye it seemed to be equal in each of the cases. The food was placed in the box having the most nearly saturated color, and twenty-four trials in series of six, as before, were given each bird. The results were quite similar to those secured with different colors. With the red shades there were twenty-two choices of the best saturated shade to eight, ten, and eight, respectively, of the other three; and with green, twenty-one to nine, ten, and eight. The 43 correct choices were distributed from series 1 to series 4 as follows: 7, 11, 12, and 13, which shows learning as before. The relatively large number of right choices was probably due, partially to the fact that fewer alternative choices were possible since only four boxes were used, instead of six, and partially to the fact that the box containing the food may have been slightly brighter than the others.

Throughout these trials the position-element was a decidedly disturbing factor. When the animals were first learning to choose a box of a definite color, some would show a marked tendency to approach a receptacle occupying a certain position, and would persist in this from series to series. Others at first showed no special preference for certain positions, but, after happening to make a correct choice, they would return to that same place the next time, and thus miss the right box which had been changed for the new test.

C. _Form Tests._

In this experiment the six food-boxes were each of different form: triangular, square, oblong, hexagonal, circular, and elliptical. They were of the same capacity, and were covered with light-brown paper. As in the preceding experiment, the birds were tested for only four of the boxes, and were given thirty trials each. Six animals were used, and as they were not the same as those previously employed, the square box (which had always been used before) had no advantage over the others in attracting the birds at the beginning of the trials. The tests were given as in the preceding experiment, except that it did not seem necessary to change the position of each of the six forms before giving each test; it was thought sufficient to move the food-box, and, if a wrong choice had been made in the preceding test, also the box wrongly chosen. The results are shown in Tables IX and X.

TABLE IX. FORM ASSOCIATION. GENERAL DISTRIBUTION OF CHOICES

_Choices of all 6 boxes when food was placed in Tri., Sq., Hex.,
or Cyl. boxes_

_Food in Tri._ _Food in Sq._
_Animals_ Tri. Sq. Ob. Hx. Cyl. El. Tri. Sq. Ob. Hx. Cyl. El.
(U) 8 6 5 5 2 4 5 9 6 1 6 3
(V) 9 2 3 6 7 3 5 8 5 2 5 5
(W) 8 7 3 4 5 3 5 8 5 3 3 6
(X) 11 5 2 5 3 4 5 8 3 6 3 5
(Y) 11 4 5 4 2 4 4 9 5 6 3 3
(Z) 8 3 7 4 4 4 6 11 7 3 1 2

Total, 55 27 25 28 23 22 30 53 31 21 21 24

_Food in Hex._ _Food in Cyl._
_Animals_ Tri. Sq. Ob. Hx. Cyl. El. Tri. Sq. Ob. Hx. Cyl. El.
(U) 5 5 3 11 3 3 6 5 3 5 8 3
(V) 4 4 5 8 5 4 6 6 3 4 7 4
(W) 3 5 3 10 7 2 4 3 4 3 10 6
(X) 4 5 4 7 6 4 2 5 4 4 9 6
(Y) 3 4 2 8 6 7 3 2 5 6 9 5
(Z) 5 4 2 11 3 5 3 5 4 4 9 5

Total, 24 27 19 55 30 25 24 26 23 26 52 29

TABLE X. FORM ASSOCIATION. DISTRIBUTION OF RIGHT CHOICES

_Choices from series 1 to 5 in the case of Tri., Sq., Hex.,
and Cyl. boxes_

_Food in Tri._ _Food in Sq._ _Food in Hex._ _Food in Cyl._
_Animals_ 1, 2, 3, 4, 5, 1, 2, 3, 4, 5, 1, 2, 3, 4, 5, 1, 2, 3, 4, 5,
(U) 2 0 2 1 3 1 2 2 2 2 2 1 2 3 3 1 1 2 1 3
(V) 1 1 2 3 2 2 1 1 2 2 0 1 2 3 2 1 2 2 1 1
(W) 1 2 1 2 2 2 1 2 1 2 1 2 2 2 3 1 2 2 3 2
(X) 1 2 3 3 2 0 2 2 3 1 1 2 2 1 1 1 1 2 3 2
(Y) 1 2 3 2 3 2 2 2 3 2 1 1 2 1 3 0 2 2 2 3
(Z) 1 2 2 2 1 0 2 3 3 2 1 2 2 4 2 1 2 3 1 2

Total, 7 9 13 13 13 7 10 12 13 11 6 9 12 14 14 5 10 13 11 13

It will be seen that each animal chose the right box oftener than any other one box, but not oftener than all of them; also that there was a small increase in the number of right choices from series to series. No one of the four forms seemed better discriminated than the others if we may judge from the practical equality of right choices made in each case (55, 53, 55, 52) or from the similar increase in number of right choices from series to series; the hexagonal and cylindrical boxes received fewer choices in the first series than did the triangular and square, but this was exactly counterbalanced in the last series. The triangular box was more often confused with hexagonal and square, and the square with triangular and oblong, than with the others. For the hexagonal box the cylindrical was more frequently mistaken than were the other forms, especially the oblong; and with the cylindrical the elliptical was more frequently confused than were the others, especially the oblong. In this series of tests nothing new as regards general behavior or method of learning was observed.

TABLE XI. POSITION, COLOR AND FORM ASSOCIATION

_Total Right
Choices_[214] _Right choices from series 1 to series 5_[215]
1 2 3 4 5
_Position_ 57.9% 54.2% 55.6% 56.9% 63.2% 59.0%
_Color_ 35.3% 16.2% 29.7% 37.0% 42.7% 51.6%
_Form_ 29.8% 17.4% 26.4% 34.7% 35.3% 35.3%

If we compare the results obtained in these three experiments (see Table XI and Fig. 7), we shall see that the pigeons were governed much more by the position of the food-box than by either its color or its form, and that color was better associated than form. Position was a most important factor throughout, as was observed also by Porter[216] in the case of the English sparrow. Porter[217] also found that his sparrows could associate color better than form. In the position-tests the pigeons showed very little improvement from series to series (see table); almost all that the animals could learn was acquired at the beginning. The more difficult color- and form-trials, however, showed almost constant improvement, although we should have expected this to be greater in the latter case than it was. When judged entirely by the actual number of right choices in a given kind of tests, some of the birds made a very poor showing; but from the standpoint of increasing number of right choices they appeared in a wholly different light.

Thus, for example, bird F (Table VII) made only 33 right choices in a possible 120, yet their arrangement is significant, being, from series 1 to series 5, respectively, 1, 5, 7, 10, 10. It is probable that there would have been still greater improvement had the tests been continued; perhaps the animal would have become as proficient in finding its food by depending upon the color of the receptacle usually containing it, as by relying upon the position of the box in the group.

V. SUMMARY

1. Respiration in pigeons is sensitive to various stimuli, and since its alterations of rate, amplitude, etc., can be easily recorded pneumo-graphically without frightening the animals, it may well serve as a process through which to study their mental life.

2. By repetition meaningless stimuli, for example, pistol-shots, quickly lose their disturbing influence; whereas the breathing remains sensitive to those of a significant character, such as the noises made by other birds.

3. Reaction to light of moderate intensity consists principally in an immediate quickening, the amount varying with the color; since a direct correspondence was found between color-preference and breathing-rate, it would seem that here agreeable feeling involves increased breathing activity.

4. Visual, acoustical, probably tactual, and certainly organic data, are the principal sensory factors of the associations of pigeons.

5. The animals readily form useful associations by a method of "trial and error," or the selection of successful movements which were at first accidental.

6. Apparently a pigeon does not learn by merely seeing a new act performed by another pigeon; yet there are instances of simple ("instinctive") imitation, and "trial and error" learning is not wholly independent of social conditions, since it proceeds much more satisfactorily if the animal is trained at least within hearing distance of other pigeons.

7. When a habit is being formed, the "period" required for the first test is usually very long, but learning proceeds quite rapidly during the next few trials; later it is more gradual, but it continues till the act becomes thoroughly familiar.

8. Associations are fairly permanent, and some remain practically unaltered for at least six weeks. Modification is easily accomplished, however, on the basis of new experience.

9. Pigeons differ widely both as to the ease with which they acquire associations and also as to their permanence. Difference in activity seems the chief reason for this.

10. While these birds seem mentally inferior to English sparrows and to various mammals which have been tested in a similar manner, they are capable of numerous ready adjustments. They discover circuitous labyrinth passages, they learn to manipulate latch apparatus when adapted to their natural habits and conveniently placed, and they easily reach their food by depending upon the position, color, or form of the box containing it. But the process is apparently simple association throughout. There is no evidence of higher mental activity--no looking the situation over and acting accordingly, no "reasoning" in the proper sense of the word, but only blind movements, some of which are retained and become highly specialized, merely because successful.

FOOTNOTES:

[Footnote 180: D. Ferrier: The Functions of the Brain, p. 111, London, 1886.]

[Footnote 181: A. Hill: Can Birds Smell? Nature, vol. 71, pp. 318, 319, 1905.]

[Footnote 182: W. Mills: The Nature and Development of Animal Intelligence, pp. 248, 250, New York, 1898.]

[Footnote 183: Hachet-Souplet: Examen Psychologique des Animaux, pp. 33-38, Paris, 1900. See also Riverside Natural History, vol. 4, pp. 240, 241, Cambridge, 1888.]

[Footnote 184: Orientation chez le pigeon-voyageur, Revue Scientifique, vol. 13, pp. 352-359, 1900.]

[Footnote 185: Orientation du pigeon-voyageur, Revue Scientifique, vol. 2, pp. 417-420, 453-457. 1904.]

[Footnote 186: Grundlinien zur Erforschung des Helligkeits- und Farbensinnes der Tiere, p. 102, Prague, 1888.]

[Footnote 187: Some Notes on the Psychology of Birds, Seventh Annual Report of the New York Zoölogical Society, p. 154, 1902.]

[Footnote 188: Ueber die Begleiterscheinungen psychischer Vorgänge in Athem und Puls, Philosophische Studien, vol. 18, pp. 7-14, 1901.]

[Footnote 189: Animal Intelligence, pp. 8-12, 31-36, 51-55, New York, 1898.]

[Footnote 190: Nature of Animal Intelligence and Methods of Investigating It, Psychological Review, vol. 10, pp. 262-274, 1897.]

[Footnote 191: An Experimental Study of the Mental Processes of the Rat, American Journal of Psychology, vol. 11, pp. 135-164; vol. 12, pp. 206-210, 1900-1901.]

[Footnote 192: Mental Life of Rhesus Monkeys in Captivity, American Journal of Psychology, vol. 13, pp. 97-148, 180-210, 1902.]

[Footnote 193: A Preliminary Study of the Psychology of the English Sparrow, American Journal of Psychology, vol. 15, pp. 313-346, 1904.]

[Footnote 194: For a more complete report of this special part III, see the writer's paper, Respiration and Emotion in Pigeons, Journal of Comparative Neurology and Psychology, vol. 15, pp. 494-513, 1905.]

[Footnote 195: If shallowing accompanies quickening, the respiratory activity may be no greater than before; but since depth alterations were seldom observed in these trials after the first day of experimentation, the rise in rate may be taken as a fair measure of the influence of the stimulus.]

[Footnote 196: P. Zoneff und E. Meumann: _op. cit._, pp. 57, 58.]

[Footnote 197: R. MacDougall: The Physical Characteristics of Attention, Psychological Review, vol. 3, pp. 162, 176, 177, 1896.]

[Footnote 198: Thorndike: _op. cit._, pp. 13-15.]

[Footnote 199: Small: _op. cit._, vol. 12, pp. 236, 237.]

[Footnote 200: Small (_op. cit._, vol. 11, p. 146) states that in his rats "the persistence of useless motor habits is striking" and "explainable by the supposition that the movements are touched off automatically."]

[Footnote 201: _Op. cit._, vol. 12, p. 214.]

[Footnote 202: _Op. cit._, p. 28.]

[Footnote 203: The Instincts, Habits and Reactions of the Frog, Harvard Psychological Studies, vol. 1, pp. 591-593, 1903.]

[Footnote 204: Small: _op. cit._, vol. 12, pp. 230, 231.]

[Footnote 205: _Op. cit._, pp. 318, 319.]

[Footnote 206: C. L. Morgan: Animal Behaviour, pp. 179-193, London, 1900; Animal Life, and Intelligence, p. 453, Boston, 1891.]

[Footnote 207: Thorndike: _op. cit._, pp. 54, 56, 57, 60, 61; The Mental Life of the Monkeys, Psychological Review, Monograph Supplement, vol. 3, pp. 318, 319, 1901. Kinnaman: _op. cit._, pp. 198-200.]

[Footnote 208: Small: _op. cit._, vol. II, p. 160.]

[Footnote 209: W. Mills: Nature of Animal Intelligence and Methods of Investigating It, Psychological Review, vol. 10, pp. 262-274, 1897.]

[Footnote 210: _Op. cit._, p. 335. See also C. L. Morgan: Introduction to Comparative Psychology, p. 232, London, 1900.]

[Footnote 211: Cornish states (Animals at Work and Play, p. 30) that hunters near the Caspian are able to decoy partridges by use of brilliant colors.]

[Footnote 212: See the writer's paper, Respiration and Emotion in Pigeons, _op. cit._, p. 502.]

[Footnote 213: One of Porter's sparrows was less successful with yellow and red than with blue and green. He says: "This may be partly explained from the fact that she was more afraid of these." _Op. cit._, pp. 338, 339. See also E. L. Thorndike: Instinctive Reactions of Young Chicks, Psychological Review, vol. 6, pp. 283-284, 1899.]

[Footnote 214: Tables V, VII, and IX.]

[Footnote 215: Tables VI, VIII, and X.]

[Footnote 216: _Op. cit._, p. 338.]

[Footnote 217: From the tables (_op. cit._, pp. 330-339) it seems that the right choices for position, color, and form, were respectively, 40%, 58% and 20%. The comparatively small number of correct position choices was probably due to his using ten boxes instead of six, as in the other two series. My results given in Table XI were secured under almost exactly comparable conditions. Compare results of Kinnaman in case of the Rhesus monkey, _op. cit._, pp. 130, 131, 134, 141, and 177.]

REACTIONS OF THE CRAYFISH

BY J. CARLETON BELL

The crayfish has long been the typical Crustacean for anatomical and physiological investigations, but it is only recently that its reactions to sensory stimuli have been made the object of experimental study. The purpose of this paper is to describe the reactions of the animal to certain sensory stimuli under experimental conditions, and to estimate the relative importance of these stimuli in the life of the organism.

I. REACTIONS TO VISUAL STIMULI

Huxley[218] states that crayfish avoid direct sunlight, hiding under stones during the day, and becoming active in the evening. On the other hand, they are attracted like moths to fires lighted on the bank at night, and may be scooped out by hand. Abbott,[219] giving an account of the burrowing crayfish, _Cambarus diogenes_, states that it is very difficult to observe the animals at work, since all their digging is done at night. It would seem from the account of Miss Hoppin, quoted by Garman,[220] that the blind crayfish, _Cambarus pellucidus_, is not altogether insensitive to light, for, reporting on the fauna of the caves of Missouri, she says that the crayfish are all found near the entrance to the cave, where there is considerable light. In the dark recesses there are only little white fishes. Blind fish and crayfish are also taken from the wells in the neighborhood, where the crayfish are found only in wells that are rather shallow and light; the fish, on the other hand, are only obtained from deep, dark wells.

According to the above accounts it would appear that the crayfish is negatively phototactic to direct sunlight or diffuse daylight, but positively phototactic to a light at night, and moreover, that light may influence the behavior of the animal even when the eyes have ceased to function.

The directive influence of light upon the movements of the crayfish has never been experimentally studied to my knowledge. Dearborn[221] thinks that light has no effect upon the animals. Yerkes[222] and Towle[223] have shown that Daphnia move toward the light. Bethe[224] finds that Carcinus is negatively phototactic, and also shows a tendency to hunt out corners. When the eyes are varnished with lampblack, the phototaxis disappears, but the tendency to seek out corners still remains. Bethe says that he has observed the same phenomenon in the crayfish. Keeble and Gamble[225] discovered that _Hippolyte varians_ responds positively to light under all conditions, and Palæmon is just as markedly negative. _Macromysis_, however, reacted now positively now negatively, depending on the background. A black (absorbing) background called forth a positive response, while a white (scattering) background produced a negative reaction. Spaulding,[226] in studying the habits of the Hermit Crab (_Eupagurus_), found that it is strikingly positively phototactic. When animals are placed in an aquarium, one half of which is shaded, none of them are ever noticed inside of the dark line. Herrick[227] notes that lobsters are nocturnal, and avoid the light when placed in a tank, and Bateson[228] says that prawns and shrimps lie hidden during the day, and are active only at night. Parker,[229] in a study of Copepods, finds that the females have a strong positive phototaxis for light of a low intensity, while males show a weak negative phototaxis. To light of over 100-candle power at a distance of 10 cm. or to direct sunlight the female Copepods are negative, while the reaction of the males does not seem to be altered.

In his work on _Carcinus_, Bethe obtained retraction of the eye-stalks by suddenly throwing a strong light on the eye by means of a mirror. "Usually the eyes were quickly drawn in and protruded again, sometimes several times in rapid succession, like a man blinking under a sudden, strong light." When a dark object, the size of the hand, was moved just over the water, the eyes were seldom retracted, but the antennules were usually drawn in. Lemoine[230] observed that in Astacus retraction was due to touch alone, and that no light, however strong, was able to bring about such a reaction. Gulland[231] takes just the opposite view with reference to Astacus, stating that there are no setæ of any sort on the eye-stalk, and therefore it is insensitive to touch, but is withdrawn only because the animal sees the object by which the stimulus is given. If a curved needle is used, and the stimulus is applied from behind, no retraction follows. Dearborn,[232] however, working with Cambarus, agrees with Lemoine in saying, "Withdrawal of the ophthalmites into their sockets occurs only on contact with some hard object,--not from any light-stimulus of an ordinary sort." I may say in passing that in none of the following experiments on Cambarus was there ever a sign of retraction due to stimulation by light, the retraction always taking place in response to a touch-stimulus.

Lyon,[233] in his study of compensatory movements of the eye-stalks, found that when the eyes were painted with lampblack, the crayfish showed a reduction of about 10% in the compensatory movements when rotated in vertical planes, but the compensation remained the same for rotation about the dorsi-ventral axis. On rotation in the dark the compensatory movement of the eyes was found to be from 5° to 8° less than in the light.

EXPERIMENTAL

In the investigations to be described, 58 crayfish of the species _Cambarus affinis_ were made use of, and for identification the animals were marked on the back with white enamel paint, the males receiving the even numbers from 2 to 64, the females the odd numbers from 1 to 51.

1. _Reactions to White Light_

The questions proposed for investigation were, (_a_) How does the crayfish react to diffuse daylight; (_b_) to reflected sunlight; (_c_) to direct sunlight; (_d_) to artificial light of different intensities? (_e_) What is the influence of previous conditions of exposure to light upon the reactions of the animal? (_f_) Do changes of temperature affect the reactions?

A wooden box, 80 cm. long, 25 cm. wide, and 20 cm. high, painted black on the inside, and constructed so as to hold water, was covered with a heavy black cloth to exclude the light from above. The front end of the box was of glass, thus admitting the light from the end. In all the experiments except those with direct sunlight, this glass end was covered with black cardboard in which a hole 10 cm. long and 5 cm. high had been so cut that the light was admitted at the middle of the bottom of the glass. The direct sunlight was admitted through the whole of the glass end. At the rear of the box a piece of black cardboard was so arranged that an aperture was afforded for observing the animals without admitting any appreciable amount of light, and this aperture could be readily closed by a slide when not in use.

The method of experimentation was to place the animal in the box about 20 cm. from the glass end, and observe whether it went toward or away from the source of light. The animals were experimented on in two groups of five each, and one hundred observations were made on the individuals of each group with each intensity of light, that is, twenty observations on each animal. In order to check the influence of the orientation of the animal at the time of exposure to the stimulus, the following four positions for placing the animal were chosen: (1) Head toward the light; (2) Head away from the light; (3) At right angles to the light with right side toward it; (4) At right angles with the left side toward the light. Thus five observations were made on each animal of each group in each position, exposed to each of the different intensities of light.

Seven different intensities of light were employed, and the results have been arranged in eight sets, as follows: I. Diffuse daylight in dry box, _i. e._, the animals were taken out of their ordinary medium, water, and were exposed to the stimulus of diffuse daylight in the air. The reactions under these conditions, however, were so slow and so unsatisfactory that the test was abandoned after the first group, and thus the second group has nothing to show for itself under this head. The remaining seven sets of observations were made on animals placed in 10 cm. of water at 15° C. II. Diffuse daylight. III. Reflected sunlight. The box was placed near a window on a clear day, and the sunlight was thrown in horizontally by means of a mirror. IV. Direct sunlight. On a clear day the box was placed in such a position that the sun shone in directly and illuminated the front half of it. V. 9-candle-power incandescent electric light. This lamp was marked 16 c., but it had been used a great deal, and on being tested with a Lummer-Brodhun photometer showed only 9 c. VI. An incandescent electric light of about 50 c. This lamp was marked 100 c., but had been used considerably and was slightly smoked. Unfortunately it was broken before there was any opportunity to test it. Judging from the fact that another 100 c. lamp of the same manufacture, in slightly better condition, measured 64 c., the estimate of 50 c. seemed a safe one. VII. The incandescent electric light alluded to above, which measured 64 c. VIII. An arc light which varied in intensity from 150 c. to 250 c.

In intensities V and VI the lamp was placed 5 cm. from the glass end of the box to allow the interposition of a heat-screen consisting of an alum solution in a flat glass jar 5 cm. thick. Reckoned in candle-metres, therefore, the intensity of the illumination at the surface of the animal in V was 144 c. m., and that in VI was about 800 c. m. In VII two heat-screens were used, and between these was placed a lens of considerable but not accurately determined curvature, so that it is impossible to express the intensity in candle-metres. In VIII the light was so variable that such an expression would mean nothing.

Unfortunately it was impossible to keep the two groups constant throughout the whole series, owing to the death of two individuals in each group during the experimentation. Group 1 was composed of nos. 1, 3, 4, 8, and 9, of which 1 and 8 were replaced by nos. 13 and 42 respectively. Group 2 was begun with nos. 23, 27, 32, 34, and 38, and the vacancies caused by the death of 23 and 32 were filled by nos. 21 and 36. The following table exhibits the reactions to the different intensities of light, + indicating an orientation toward the source of light, - an orientation away from the light, and ± an indifferent orientation, which usually means no movement at all.

TABLE I. SUMMARY OF REACTIONS TO WHITE LIGHT

Group 1 Group 2 Totals
+ - ± + - ± + - ±
I. 23 48 29 23 48 29
II. 49 47 4 19 81 68 128 4
III. 40 60 30 70 70 130
IV. 47 52 1 36 64 83 116 1
V. 51 49 30 70 81 119
VI. 39 61 39 61 78 122
VII. 28 72 28 71 1 56 143 1
VIII. 35 63 2 37 60 3 72 123 5

312 452 36 219 477 4 = 531 929 40 = 1500

The following table gives the average time required for orientation for each group. The time of each animal in seconds was noted with a stop-watch from the instant the animal was placed in the box until a definite orientation was assumed with reference to the light. If no orientation followed within three minutes, the result was called indifferent.

TABLE II. AVERAGE TIME OF ORIENTATION

Intensities
I II III IV V VI VII VIII Ave.
Group 1 (144) 52 11 4-1/2 9 8-1/2 11 19 16-1/2
Group 2 9 3 4 4 5 4 5 5

Inspection of these tables shows that when the animals were taken from the water and placed in diffuse daylight in the air (I), their movements were so sluggish that in twenty-nine cases out of one hundred there was no orientation within the three-minute limit. Moreover, in the seventy-one cases where there was definite orientation the average time was over two minutes (144 seconds). While, therefore, the conditions were quite different from the normal environment of the animal, it is interesting to note that of the cases where orientation did take place the negative reactions were more than twice as many as the positive. In II, where the animals were under the same conditions of diffuse daylight but in the water, a wide variation in the reactions of the two groups is noted. In group 1 they are about equally divided between positive and negative, while in group 2 there is the largest proportion of negative reactions in the whole series. It will be observed that the time for group 1 is extremely long compared with the other averages, and this doubtless indicates a general sluggishness and lack of sensitiveness to stimuli in the animals, which might to some extent account for the difference in reaction. If we consider the totals of both groups for each intensity, we are led to conclude that there is no appreciable difference in the reactions of crayfish to diffuse daylight, to sunlight, or to artificial light within the limits here employed. A slight exception to this is found in VII, where the 64 c. lamp with the lens caused a somewhat more uniform negative reaction. The action of direct sunlight in IV is rather remarkable in that with the lowest proportion of negative reactions in the whole series we also observe the shortest time-average, indicating that the animals are the liveliest and most sensitive. This would seem to indicate that while the animals are in general somewhat negatively phototactic to all light-stimuli of moderate intensity the action of direct sunlight tends to reduce the negative phototaxis to a minimum. If we consider the totals of the two groups separately we observe that group 1 has only 57% of negative reactions while group 2 shows 68%. This is rather in accordance with what we would expect from the general time-average, which is over three times as much for group 1 as for group 2. But although we may in a general way connect rapidity of orientation with a large percentage of negative reactions, it will not do to carry it to individual cases, for it was observed that no. 27 showed 83% of its reactions negative, yet its total time-average was 10 sec., the highest in its group.

In general, then, we conclude that the crayfish is negatively phototactic in the proportion of about two to one. This apparently contradicts the statement made by Huxley that crayfish "are attracted like moths to fires lighted on the bank at night." For surely if this were the case some such tendency would have been observed in these experiments. On the other hand, there is no such marked and definite response to light as in the case of Daphnia or Hippolyte or Palæmon or the Hermit Crab. The action of the stimulus is by no means mechanical and constant, but there is wide variation in individuals.

As was mentioned in the description of the method of experimentation, four different positions for placing the animal were chosen with the idea that the initial position of the animal with respect to the light might have some influence on the direction of its movement. To determine what this influence might be, a careful record was kept of the orientation with reference to each one of these positions, and the following table gives a summary of these observations. In the table position I is where the animal is placed with its head toward the light; position II, with head away from the light; position III, at right angles to the light with the right side toward it; position IV, at right angles with left side to light.

TABLE III. INFLUENCE OF POSITION ON LIGHT REACTIONS

Position I Position II Position III Position IV Totals
+ - ± + - ± + - ± + - ± + - ±

Group 1 85 108 7 61 135 4 74 114 12 92 95 13 312 452 36
Group 2 39 133 3 48 126 1 57 118 75 100 219 477 4

124 241 10 109 261 5 131 232 12 167 195 13 531 929 40

Since the animals have been shown to be somewhat negatively phototactic, we should expect that position II, with the head away from the light, would show the largest number of negative reactions, and this is what we find if we take the sum of both groups. But by the same course of reasoning we should expect position I, with head toward the light, to yield the smallest number of negative reactions, a condition which prevails neither in the sum nor in either of the groups. On the whole we can only say that difference of position seems to have remarkably little influence on the orientation of the animals.

In his work on the eye of the crayfish, Parker[234] called attention to the migration of the pigment in the retinular cells under the influence of light. The question now arose, what influence, if any, does this pigment migration exert upon the reactions of the crayfish to light? The time required for pigment migration in the eye of the crayfish has never been determined to my knowledge, but from the work of Parker[235] on Palæmonetes it was thought that confinement in the dark for about an hour would be sufficient to bring about a retraction of the pigment. Accordingly, group 1 was kept in the dark for one hour, group 2 for one hour and a half, before experimentation. A further test was made to observe the effects of pigment expansion, both groups having been exposed for one hour and a half to the rays of a 32 c. incandescent electric light at a distance of 40 cm. The apparatus used in the reaction-tests was the box described above, with the 64 c. light as a stimulus. As to the method of observation, each group was placed in the centre of the box at right angles to the horizontal rays of light, and the position of each animal was accurately noted at intervals of one minute for one hour. In reporting the results, all the observations of animals in the half of the box nearest the light are denominated positive, those in the half farthest from the light negative. The results are given in Table IV, where line I indicates the reactions after confinement in the dark, line II those after exposure to the light.

TABLE IV. INFLUENCE OF PREVIOUS CONDITIONS UPON REACTIONS TO LIGHT

Group 1 Nos. 13 3 4 9 42 Totals
+ - + - + - + - + - + -
I 54 6 60 5 55 48 12 60 167 133
II 59 1 60 60 9 51 55 5 183 117

Group 2 Nos. 21 27 31 36 38 Totals
+ - + - + - + - + - + -
I 5 55 5 55 7 53 10 50 4 56 31 269
II 38 22 60 13 47 45 15 16 44 116 184

Let it be said at once that these results do not offer an altogether satisfactory basis for an answer to the question proposed above. Some of the animals would take up a position during the first ten minutes and remain in it for the rest of the hour. Whether the position taken was due solely to the light, or was owing to thigmotactic influences, or whether it depended on the way in which the animal was released, are questions which cannot be answered, and for this reason the conclusions to be drawn from the table are tentative. If we examine the table we find that the totals of both groups agree in manifesting a decrease in negative results for line II, after exposure to the light, as compared with line I, after confinement in the dark. This is what we would expect from negatively phototactic animals. When taken from the dark the pigment is retracted, and the sensitive retinal substance is exposed to the direct action of a rather strong light. The negative tendency of the animal we should expect to find accentuated. The decrease in negative reactions is especially marked with group 2, which was shown above to be much the livelier of the two, and all the individuals except no. 27 share in the change. In group 1 the decrease is not so striking, and is observed to be due to two individuals solely. Inexplicable is the preponderance of positive over negative reactions in the results for group 1.

All of the experiments thus far described were carried out in water at approximately 15° C. The question naturally arises, what will be the result of raising or lowering the temperature upon the reactions of the animals to light? Unfortunately the experiments anent this question are fragmentary and incomplete, but the results will be given for what they are worth. The same apparatus and the same intensity of light (64 c.) were used as in the preceding paragraph. The results, presented in Table V, are arranged in three sets, as follows: The line marked I represents the results obtained from group 1 at a temperature of 5° C. The animals were placed in the box one at a time, as in Table I, and their orientation noted. They were set at right angles to the rays of light, five times with the right and five with the left side toward the source of the stimulus. No observations were made upon group 2 at 5° C., and those on group 1 are so few as to have a questionable value. Line II gives the reactions of both groups of animals in water at 25° C., and in this set the animals were placed in all four of the positions indicated for Table III. Line III presents the reactions of the animals in water at 25° C. by the method outlined for Table IV, _i. e._, each group of animals was placed in the centre of the box, and observed at intervals of one minute. To obviate the objection of the animals remaining in one spot, they were reset every ten minutes in the middle of the box, at right angles to the entering rays.

TABLE V. REACTIONS TO LIGHT AT DIFFERENT TEMPERATURES

Group 1 Nos. 13 33 4 42 9 Totals
+ - + - + - + - + - + -
I 7 3 4 6 2 8 2 8 8 2 23 27
II 16 4 14 6 11 9 9 11 13 4(±3) 63 34(±3)
III 19 41 13 47 39 21 17 43 27 33 115 185

Group 2 Nos. 21 27 31 36 38 Totals
+ - + - + - + - + - + -
II 12 8 5 15 12 8 15 5 13 7 57 43
III 12 48 12 48 31 29 42 18 44 16 141 159

Judging from lines II and III we may say that there is a tendency toward a decrease of the negative phototaxis with an increase in temperature. It is true that group 1 in line III maintains the average, 62% negative reactions, but the others are much lower than this, line II even going over to positive phototaxis in both groups. In line III the animals of both groups were extremely active during the first ten minutes, rushing about from one end of the box to the other, pushing each other back and forth, and in general exhibiting great restlessness. Some of the animals when first put into the water reacted with a sort of cramp reflex, which was followed in a few seconds by intense activity. After the first ten minutes the animals began to grow more quiet, and in twenty or thirty minutes they had become quite sluggish, scarcely moving out of the position in which they were reset. During the period of restlessness the males showed marked sexual activity, rushing up to the females, pushing them about, seizing them, and trying to turn them over in spite of their vigorous resistance. One of the males, no. 36, did succeed in turning a female on her back twice, although she struggled violently to escape,--a thing which the female never does in the ordinary sexual act. The rise in temperature, therefore, seemed to stimulate the males to sexual activity, but not the females.

2. _Reactions to Colored Light_

No observations have ever been made, so far as I know, on the reactions of the crayfish to colored light. Lyon, in his work on compensatory eye-movements, found that rotation in blue light gave a compensatory movement only slightly less than that in white light, while in red light the compensation was only a little larger than in darkness. In some animals the interposition of an opaque object between the eye and the source of light caused an elevation of the eye 1° or 2° toward the vertical. Red glass acted like an opaque object, blue glass produced no effect, _i. e._, blue light had the same effect as white light. To observe whether the same thing applied to movement reactions was the object of the following experiments.

_a. Reactions to Horizontal Colored Light._ The same apparatus was used as in the previous experiments, viz., the dark box with light from the 64 c. lamp entering horizontally at the end. Across half of this end were placed pieces of colored glass of a saturated blue, green, yellow, and red. The colored light obtained by this means was not spectrally pure, but it was the nearest to it that could be obtained. A more serious objection is that the intensities were not the same, the red and the yellow being very appreciably brighter than the blue and the green. In addition to observations with these colors, a piece of black cardboard was introduced in the same position as the glass, thus cutting off the light from that half of the box. This, to preserve the uniformity of the series, is denominated black. The animals were placed in the centre of the box, on the line separating the white from the colored light, and were observed at intervals of one minute for forty minutes, the position of each animal being accurately noted. At the end of every ten minutes the animals were reset at the centre of the box. The following table gives a summary of the results for each individual. Here again it was impossible to keep the groups constant owing to the death of individuals during the progress of the experiment.

TABLE VI. REACTIONS TO HORIZONTAL COLORED LIGHT

Group 1 Group 2
Animals 13 37 41 42 44 33 4 9 43 46 Sum 21 27 31 36 38 52 Sum Sum
Total
Blue 12 9 21 30 18 90 23 2 24 13 37 99 189
White 28 31 19 10 22 110 17 38 16 27 3 101 211

Green 15 9 23 36 17 100 12 19 3 28 40 102 202
White 25 31 17 4 23 100 28 21 37 12 98 198

Yellow 20 10 31 28 19 108 17 11 28 37 35 128 236
White 20 30 9 12 21 92 23 29 12 3 5 72 164

Red 26 9 32 20 17 104 2 30 16 29 36 113 217
White 14 31 8 20 23 96 38 10 24 11 4 87 183

Black 24 27 12 27 26 116 15 18 26 25 5 89 205
White 16 13 28 13 14 84 25 22 14 15 35 111 195

In this table the colored lights are arranged in the order of the spectrum from blue to red. On the hypothesis that blue light has practically the same effect upon animal reactions as white light, while red is about the same as darkness, we might expect that the reactions would be about equally divided between the blue and the white, and that there would be a gradually increasing difference in number as we go down the table, reaching the maximum with the last pair, black-white. This, we see, however, is not quite the case. In both groups the white has a slightly larger number of reactions than the blue, while the pair green-white shows numbers more nearly equal. In the sum totals the yellow shows a greater preponderance over the white than any other color, and the black and white are very nearly equal. Group 1, it is true, shows a fairly regular ascending scale in reactions to the colored lights with the exception of the red, and the same might be said of group 2 if it were not for the very low number of reactions to the black and the exceptionally high showing of the yellow. On the whole, however, the differences are so small and the individual variations are so large that we can only conclude that for these conditions colored light has little or no effect on the reactions of the animals.

In the foregoing experiment the light came from a broad spiral coil inside the bulb of the lamp, and the distance from it to the edge of the glass was so small compared with the length of the box that there was no sharp dividing-line between the colored light and the white, but rather a wedge-shaped block of lessening saturation of the color, and this wedge, having the point toward the light, took up the whole of the box at the extreme farther end. Thus the imaginary central line dividing the white light from the colored departed farther and farther from the reality as the rear of the box was approached. To obviate this difficulty and to get a check on the previous work, the following series of experiments was undertaken.