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

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

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

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

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

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.

b. Reactions to Vertical Colored Light. The same box was used as in the previous experiments, but the end was closed with a black cloth, and an electric light marked 32 c. but measuring only 22 c. was hung exactly over the middle of the box, 40 cm. from the bottom. By means of wires it was arranged that a plate of colored glass could be swung in such a manner that all of one half the box (the whole of one end) was illuminated with the desired color, while the other half was either left white or illuminated with another color. In this way there was a fairly sharp dividing-line between the two colors. The animals were observed at intervals of one minute for 40 minutes, and reset at the middle on the dividing-line every ten minutes as before. Table VII gives the results of the observations.

In this set of experiments it was possible to keep the groups intact except that in group 1 no. 13 had to be replaced by no. 46. If now we conceive the colors arranged in the order of the spectrum with black at one end and white at the other, and consider the black a lower stimulus than the white, we have the ascending series black, red, yellow, green, blue, white. Now since the animals have already been shown to be somewhat negatively phototactic, we should expect them to prefer a color of lower stimulus to one of higher. Turning to the sum totals in the table we find that the first color of each pair (which is always the lower stimulus) has the larger number of reactions in every case but one, the first pair of red-blue. As was stated above, it was impossible to secure colored light of the same intensity by means of the glass at our disposal, and in the present case the red was considerably brighter than the blue. Owing to the fact already mentioned that different intensities of white light seem to have no effect on the reactions it was thought that these differences in the intensities of the colored lights might be overlooked. Since the only thing that could be thought of to account for the anomalous behavior to the red-blue was this difference in intensity, another experiment was undertaken with the same animals under slightly different conditions. A glass aquarium about 40 cm. long by 20 cm. wide was covered with black cardboard and black cloth in such a manner that light could enter only through a space 5 cm. wide at the bottom of each end. Each of these ends was covered, the one with blue, the other with red glass, and 15 cm. from each end was placed an electric light marked 32 c. Later, however, it was found that one of these lamps measured 30 c. and the other 22 c. The red light was found to be much more intense to the eye than the blue, so the former was damped down with tissue paper until the two appeared to have the same intensity. The second pair of red-blue in Table VII gives the results of the observations under these conditions, and these are found to be in harmony with the rest of the table, i. e.,[236] the color giving the lower stimulus has the higher number of reactions.

TABLE VII. REACTIONS TO VERTICAL COLORED LIGHT

The most striking feature of the table is the marked predominance of the red over the white. Here the red reaches 73% of the total number of reactions, and inspection shows that this predominance is uniform not only through the groups but even for the individuals. The constancy of this reaction and the fact that it is so much more frequent than the one to the black as compared with the white, would lead one to expect that the red would have the higher percentage in the combination black-red. Such, however, is not found to be the case, although it does happen with one group. If the arrangement of our color-scale in accordance with increasing intensity of stimulus were correct, we should expect a gradually increasing predominance in reactions to colored light over those to white in the first five pairs. Instead of this we find that green and yellow stand nearest to the white, blue and black come next and are almost equal, while red is very much higher than any. In the pairs black-red and black-blue the red holds its predominance over the blue at about the same rate as in the second pair of the direct comparison, red-blue. The wide individual variations, however, in all these reactions to colored light, except perhaps in the case of red-white, indicate that there is nothing very regular, stereotyped, or mechanical about them. The most that can be said is that in a general way the red end of the spectrum furnishes a less intense stimulus to negative reaction than the blue.

A tendency to habit formation was noticed during the course of these experiments, and it is possible that this may have influenced the results somewhat. Many individuals apparently formed a habit of going to a certain corner as soon as they were reset at the centre. The positions in which they were set were varied and they were headed in different directions, but within a minute after they were released in the middle of the box they would be found in their favorite corner. This was especially the case with no. 38 in Table VI, and I think accounts in some measure for the persistent avoidance of the white. In no case did this continue throughout the whole series, but would sometimes be noted for two or three days at a time in the case of an individual. What were the controlling factors in this habit formation, the means by which orientation and recognition were effected, I was unable to determine.

3. Reactions to Objects

In no case did an animal give any sign of perceiving stationary objects in its path or of avoiding them in any way that could be referred to a visual stimulus. When the animal approached an obstruction there was no hesitation in the movement until the object was touched. Usually even when the antenna had touched the object the animal did not stop, but continued until the contact of the chelæ or even of the rostrum made further movement in that direction impossible.

With moving objects the case was quite different. Here the condition and disposition of the individual animal seemed to be the deciding factors. Often when the animals were trying to climb out of a shallow pan in which they were kept in the experimenting-room, raising a finger or holding out a pencil would be sufficient to make them stop or even start back into the pan. Nor was this response occasioned by any change in the intensity of light, such as that caused by a shadow falling on the animal, for they would react to a movement made on the opposite side of them from the window. In fact, no. 56, the most active in response to moving objects, seemed to react more vigorously to a motion made on the opposite side than when it was made between him and the light. Whenever a person came near the aquarium he and one or two others would take an attitude of defence, and would "face about" to correspond to any movement the person made toward one side or the other. When in the pan mentioned above, any movement of a person within two or three yards of him usually called forth a reaction on his part, and if the pan were placed on the table and the person moved slowly round it, the animal turned with the person, making a complete circuit of the pan.

Reaction to a smaller moving object, however, was not so marked. A black object, 20×8×8 cm., was suspended above the middle of the pan so that if set swinging it would just pass over the top. When it was pulled to one side the animal responded slightly, but after the first swing he seemed to pay no more attention to it. When the operator stepped out from behind the screen, the animal was as keen in its response as before. The experiment was now tried of allowing the object to approach from one direction while the operator moved to a position at right angles to its line of movement. Without hesitation the animal moved so as to keep fronting the operator, without paying any attention to the movement of the smaller object, although this was much nearer.

These observations on the reactions of the crayfish to stationary and moving objects are in line with the conclusions of Plateau[238] and Exner[239] drawn from observations on other Arthropods. It is Exner's belief that the compound eye is a visual apparatus which is almost worthless for detecting the forms of objects, especially if these objects are stationary, but that it may furnish a very keen perception of moving objects.

II. EXPERIMENTS WITH SOUNDS

Hensen[240] stated that Palæmon and Mysis reacted to sounds made by striking a thin, resonant board floating on the surface of the water, or by tapping the walls of the aquarium or of the room. Beer[241] repeated Hensen's experiments, but denied that the Crustacea reacted to sounds, and claimed that their movements were due to visual and tactual stimuli. Prentiss[242] confirmed Beer's results on Palæmonetes, and noted that the reactions were only slightly diminished by the removal of the otocysts, but that removal of the antennæ and antennules caused their almost complete cessation. More extended experiments were made on the fiddler crab, Gelasimus pugilator, which is on land a good deal of the time, and Prentiss's conclusions are: "(1) The reactions formerly attributed to sound-stimuli are nothing more than tactile reflexes. (2) The otocyst has little or no part in calling forth these reactions. (3) There is no direct evidence to prove that decapod Crustacea hear, and until such evidence has been obtained, we are not warranted in ascribing to the otocyst a true auditory function."

The experiments performed on the crayfish in this connection all resulted negatively and go to confirm Beer's and Prentiss's conclusions. Rapping upon a board floating in the water, and tapping the sides of the aquarium did not cause the slightest apparent reaction in the animals under observation, even though the vibration of the water could be plainly perceived by the sense of touch in the hand. When a rather large electric bell was sounded just over the surface of the water some reactions were observed which were evidently due to the movements of the hammer, but there was nothing which could be referred to the sound-stimuli. If the bell was held against the sides of the aquarium, or in the water near the animals, the vibration could be plainly felt by the fingers, yet no reactions on the part of the crayfish were observed. A metal snapper making a crack like a small pistol-shot was tried both in and out of the water but with no success in producing a reaction. A large hand tuning-fork, when held with its base pressed firmly against the glass walls of the aquarium, gave a deep rich tone of great volume, or when lightly touched to the glass produced a shrill, piercing, penetrating sound which was extremely sharp and disagreeable. Here again the vibrations of the water were quite perceptible to the hand at a distance of 10 cm., yet in neither case was there a sign of a reaction. Finally two electric tuning-forks, one of 256, the other of 512 vibrations per second, were tried on the animals taken one by one, and especial attention was given to the regular movement of the little thread-like appendages which keep up the current of water to the gills, with the idea that perhaps their rate of movement might be affected. In no case was there the slightest movement that could be referred to vibration, although here again the tactile stimulus was very perceptible to the finger. None of these experiments, then, give any indication that the crayfish reacts to vibratory stimuli which to the human ear produce sound.

III. ROTATION EXPERIMENTS

It has been found that the higher vertebrates, on being rotated on a turn-table, exhibit all the symptoms which accompany the sensation of dizziness in man. The question arises, to what degree and in what manner do invertebrates respond to rotation? Schaefer,[243] the first to take up this question, denied on rather meagre observations that Crustacea respond in any way to rotation on the turn-table. Kreidl[244] showed that this statement was altogether too sweeping, that Palæmon reacts very definitely to rotation by running in the opposite direction. Bunting[245] tried the crayfish, but all the rotation experiments resulted negatively, so she was led to confirm Schaefer's statement so far as the crayfish is concerned. Bethe[246] found that Carcinus behaved in a very definite manner on being rotated, that during the rotation the animals ran in the opposite direction to that in which they were turned, and as soon as the motion ceased they began running in the other direction. Finally Lyon,[247] while agreeing with Bunting that adult crayfish do not react to rotation, discovered that young animals two or three centimetres long react very prettily to the movement, going in a direction opposite to the turn. To confirm and if possible extend these observations on the crayfish was the purpose of the following experiments.

It was soon found that a great deal depended on the method of experimentation. None of the experimenters mentioned above gives any detailed description of the manner in which the experiments were carried out. One is left uncertain whether the animals were placed on the periphery of the turn-table or over the centre, whether in the former case they were set with their heads toward the centre or away from it, or placed at right angles to a radius, or whether they were merely set down in any chance fashion and whirled about. The same indefiniteness exists in most of the accounts as to how fast they were turned, and whether the experiments were performed in the air or in the water. Finally it is not stated whether the rotation was always in the same direction, or whether its direction was alternated.

The turn-table used in the following experiments was one that had to be turned by hand, so that it was impossible to regulate the speed accurately. The crank, however, was not attached directly to the rotating board, but was connected with it by means of a gearing so that one turn of the crank produced about ten turns of the table. This gearing gave a steadying effect to the motion so that the speed could be kept tolerably constant. A circular pan, about 15 cm. in diameter at the bottom with the sides slightly sloping outward, was set so that its centre coincided with the axis of the rotating table. It was in this pan that all the experiments were tried. Through various preliminary experiments to determine the most favorable speed, it was found that a rotation rate of over one turn of the table per second produced such a strong centrifugal force that unless the animals were set exactly over the centre they were swept off against the side of the pan in such a manner that it was difficult to decide whether the rotation as such had any effect upon their movements. It was finally decided that the best results were obtained from a rate of approximately one rotation in two seconds.

It soon became evident that when the larger and more sluggish crayfish were merely dropped in the pan and rotated there was no particular reaction. This was true whether the animals rotated were in the air or in the water. The smaller and more active crayfish, however, showed a decided tendency to run either with or against the direction of the rotation, especially when the experiments were carried on in the water. In no case was there any tendency to go in the opposite direction when the rotation ceased, except in so far as the animals were carried along by the water. To get a quantitative expression for these tendencies a more delicate method of experimentation was resorted to. If there was a tendency on the part of the active animal to move either with or against the rotation, such a tendency might also be supposed to exist in the sluggish animal, only in the latter the inertia was sufficiently strong to prevent its appearance. If, however, the animals should be set radially to the periphery of the pan, the tendency to go with or against the rotation would be exhibited in the direction in which they turned out of the radial position. For it was found that no animal would remain in that position for any great length of time. Two groups of animals were used for these experiments, five animals in each group, and the first group was selected from the smallest and most active animals, the second from the largest and most sluggish. Each animal was set in two positions, position I, with the head toward the centre, position II, with the head away from the centre. Each animal was given ten trials in each position, and the number of times it turned in a direction with the rotation is set down in the + column, the number of times it turned against the rotation is indicated in the - column. In general from 5 to 15 turns were necessary for the orientation of the animal, though sometimes the number ran up to 30 or 40. Each trial was made in the opposite direction to the preceding one, in order to avoid the formation of any habit in turning. All these experiments were carried out in water, the depth of which in the pan was about 4 cm. In order that there should be no difference between the velocity of the water and that of the pan, the table was rotated a few times before the animal was put in. As a check a series of experiments of 5 in each position was performed in the air on the more active group.

The following table shows the results of these experiments in rotation:

TABLE VIII. REACTIONS TO ROTATION