selves to be removed bodily from it before offering to fly. The two birds take turns at brooding the egg. The nest shifts occur sometimes with remarkable regularity, two hours being about the preferred time. A number of in- teresting movements in the matter of caring for the egg appear; it is constantly covered. They turn it round and round with the beak; they go to the water to wet the breast feathers to keep it moist; they shove one another aside when the shifts are made without exposing the egg for any length of time, etc. When the young bird appears, there is again a marked change in the type of response. The birds again take turns in feeding the j^oung one (of this, how- ever, we are not sure). Probably the most interesting set of changes is that in their " dispositions." As the young bird grows older the parents no longer attack the human intruder but leave the nest much more readily than during the brooding season. (VI) The noddy has an ex- ceedingly sharp bill and depends largely upon it in attack. When fighting on the ground it strikes vigorously with its wings. The blow of the beak of the tern is severe enough oftentimes to draw blood from the human hand. The beak is serrated on the inner edge, which enables it to hold quite firmly to an object. Two terns, in fighting, will strike vigorously with the beak and then lock beaks, and turn and twist, striking strong blows with the wings. It is no un- usual sight to see a noddy chasing a large man-o'-war bird from the island. He flies under the man-o'-war bird and strikes him on the breast until he soars up and leaves the island. These fights between the noddies and the man-o'- war birds are brought about by the latters' attempts to roost over the nests of the noddies. Fights among the noddies occur when a strange bird attempts to usurp the nest; to have intercourse with the female of the nest; to steal straws; or in general to encroach upon the neighborhood of a nest. They will also attack any strange birds that come to the island, with the exception of the hawks. (The young sooty, which is hatched upon the same island, exhibits an interesting case of death-shamming. When one, two, three, and four days of age it will not leave the nest upon the 118 STUDIES UPON INSTINCT open sand at the intrusion of a human being, but will lie in the nest 'perfectly motionless with head outstretched. Oftentimes when the little fellow is picked up and then put down again he will dart off rapidly to the shelter of a neighboring bush.) We have never noticed death-shamming in the adults nor a simulating of lameness (for the at- traction of the hunter, as is so often mentioned in the literature by the naturalists.) (VII) The noddy is migra- tory, wintering somewhere along the shores of Central America. Some time in April, they leave there in a body for Bird Key, Dry Tortugas, reaching that island often about the 27th or 28th of April. There is wide variation, though, as has been found lately, in the time at which they reach Bird Key. The time of arrival will vary as much as a month to six weeks. Some time in September, again the time is variable, but the 20th was once actually observed, they leave Bird Key and go again to their winter home. If the ability to home is instinctive, this bird exhibits a wonderful development of such an instinct. We have been enabled to get the noddy to return to its nest from all distances up to eight hundred miles over an open water pathway (Galveston to Bird Key). (VIII) While young the noddy gives a gentle peep exceedingly like that of the chick. The variety and range of its vocal accomplish- ments are exceedingly limited. The adult bird is almost silent, day and night. Occasionally as one approaches its nest it will give a hoarse hiss-like warning cry, and then again a purring, unbird-like kind of noise, during periods of feeding and sex activity. (IX) This bird shows some interesting unadaptive types of response. The most inter- esting one is its inveterate system of nodding to every bird that comes near it and to its mate. This nodding is often preliminary to a fight; to a shift at the nest between the mates; and to sex activity. The photograph, Fig. 28, shown below, gives some idea of the elaborateness of the process. Another form falling under this head is that of complete cessation of activity for long periods of time. This has already been mentioned under another head. It is possible that there are a large number of such acts ex- INSTINCTS OF THE NODDY TERN hibited during the mating period, but since we have not been able to observe this, no record can be given. (In a number of other birds elaborate dances are gone through with as well as certain peculiar antics which are described Fig. 28. Noddies in the Act of Bowing by Fisher.^ Another example of such peculiar responses may be found in the inflation of the cheek pouches exhibited by the man-o'-war bird.) (X) There are so many random acts exhibited by the young birds of this species that an enumeration or classification of them is impossible in this connection. (XI) Very great individual differences were noted in the persistency of certain of the birds in working at a problem, e.g., where entrance to the nest was denied. There are also very great differences in the boldness of the birds. At certain nests the birds will always fly up at the observer's approach during the entire nesting period; at others, the birds will remain in the nest even after very rough handling — often after they have been picked up and sent off on trips they will suffer themselves again to be ^ Birds of Laysan and the Leeward Islands, Hawaiian Group. Bull. U. S. Fish Commission, Vol. XXIII, 1903, pp. 767-807.
120 STUDIES UPON INSTINCT caught and sent off. Many tricks, such as pecking at the toes, idly peeking at the nest material, tricks which might serve to identify the birds, were observed.
Instinctive activity of reptiles. — Very little* is known concerning the phylogenetic modes of response of the rep- tiles. We shall mention some of the more common types ol: response found in the tortoise. (I) Aspidonectes (the " soft-shell " tortoise) runs swiftly on the land and when in the water swims with powerful strokes. It is almost im- possible to overtake it by running. The essential form of the tortoise, etc., are too well known to require a separate description. (II) Aspidonectes is carnivorous, feeding upon crayfish and the larvae of large insects. It crawls or swims along, thrusting its snout into masses of aquatic vegeta- tion, picking up now and then a crayfish, etc., swallowing the booty whole and, when needed, using the forefeet to force the food down. It is rather interesting that fish often accompany this species of turtle in its wanderings after food. The turtle often stirs up and dislodges food which it cannot capture but Avhich the fish can take with ease. (Ill) This animal, at the onset of cold weather, buries itself in the mud, and hibernates until the warm spring days of April come. It buries itself in the mud in a peculiar way, by rocking the body from side to side and throwing the mud up in such a way that it settles on the carapace and covers it quickly from sight. It sinks deeper and deeper until it gets well below the frost line. It comes out in April in a very much weakened condition. (IV) This animal has a tendency to lie on floating objects. On approach, it invariably slides off into the water. When on a floating object, it always lies facing the water; conse- quently no time is lost in turning when danger approaches. At times it crawls out on a bank and basks in the sun. Lying in shallow water, which is of course always warmer, is another favorite mode of response. By this process the turtle obtains both rest and warmth. From Newman's de- scription of its activity, it would seem that this variety is diurnal. In the case of the logpnerhead turtle, during the period when the eggs are laid, the animal is usually noc- INSTINCTS OF REPTILES 121 turnal in its land habits. The general mode for catching the green and the loggerhead turtle is to walk around the shore lines of the small islands on moonlight nights. The turtles crawl out at night to lay their eggs and if seen quickly enough can be turned over on their backs in such a way that they are helpless. When large numbers are captured they are taken to the various marketplaces and sold. (V) Little is known of the process of mating, sexual activity, etc., in the case of the turtle. Aspidonectes begins to lay early in June. These animals are exceedingly wary in their choice of times and places for laying. This species apparently lays its eggs during the daytime. The animal comes to the surface many yards from the shore and care- fully reconnoiters before coming ashore. On reaching it the turtle comes out on the sand, extending the head at its full height and remaining motionless for some time. If at this time a slight movement or faint sound alarms the animal it dashes back into the water. When undis- turbed, it proceeds slowly and cautiously to the nesting ground. The female scratches out foot-holds for the fore- feet and excavates with the hind-feet, using the right and left foot alternately with a circular gouging movement. At intervals she pushes aside the accumulated earth with her hind-feet. In less than 40 minutes (single observation) the nest is completed and she commences to lay the eggs. After laying several eggs, she arranges them with the hind- feet and rakes in earth previously wet with water from the accessory bladders. The earth is generally packed in be- fore any more eggs are laid. The remainder of the eggs are laid and the earth is tramped upon quite firmly with the knuckles of the hind-feet, the right and left foot being again used alternately. The treading movements continue for some minutes. No attempt is made to cover up the traces of the feet. It has been said on good authority that the tortoise cannot be frightened away when making her nest. This apparently is not true with the various species studied by Newman. However, if laying has actually be- gun, the animal will hesitate long enough to cover up the eggs more or less completely when surprised. Yet again, 122 STUDIES UPON INSTINCT nests may be found with eggs completely exposed. The number of eggs laid ranges from nine to twenty-four.
Instinctive activity of fish. — There has been very little work done in the w^ay of systematic and continuous observation of fishes. The most striking work which has •yet appeared upon the instincts of fish was reported by Jacob Reighard upon the breeding behavior (V) of the dogfish {Am.ia calva Linnceiis). During the last week in April the male of this species fixes on the nesting location. This is usually a place in a quiet bay or inlet well grown with water plants. He constructs a saucer-like excavation 30 to 90 cm. in diameter and from 10 to 20 cm. deep. The bottom of the excavation is covered with the fibrous roots of water plants, which form a spongy mass. At times the bottom is of gravel, sand, or black loam. Occasionally one finds in it water-soaked stems or leaves of cattails or other similar water plants. The sides of the nest are usually of rootlets and growing plants. The fish usually selects a relatively plantless area surrounded by a wall of growing plants and from this area removes the few young shoots. The nest is constructed entirely by the male without assist- ance from the female. The method of constructing the nest is to fan away the vegetation and ooze so as to expose the underlying material, and to rub down and bite away the scant vegetation. The nests are built usually at night, each nest being the property of an individual male. After the nest is constructed the male guards it for a period of usually 24 to 36 hours, but such a period may be extended to six days or possibly longer. At times the male will leave the nest. If females do not appear, he finally abandons the nest. The sexes seem to be able to detect one another at considerable distances (smell?), and the females apparently seek the nests prepared by the males. After the female reaches the nest, spawning may be ob- served. The male is very active at the time of spawning. He circles around the female in such a way as to meet her head on and travels towards her tail. In passing over her body he bites her on the snout and sides of her head and body as far back as the middle. At times during this INSTINCTS OF FISH 123 period both animals leave the nest. On each return the male reengages in the circling movement. The female usually lies quietly upon the nest. After 15 to 20 minutes of these circling movements the female is suffi- ciently stimulated to deposit the eggs. The male then approaches the female and lies by her side. This marks the completion of spawning. Spawning is done inter- mittently and usually at night. Apparently more than one female may spawn in the nest with an individual male. Then begins the guarding of the eggs, also by the male. The male lies at times directly over the nest, while at other times he is concealed in the natural openings of the ad- jacent vegetation. At these times one can approach within a very short distance of the nest. Usually as the observer approaches the male moves towards him, but not usually beyond the limits of the nest. If fright- ened, he dashes away. After a time he returns slowly and quietly to the nest. When the male guards the nest he lies, for the most part, quietly, or with only slight movements of his fins, but at intervals he moves over the nest, and thus by.the movement of his fins keeps the eggs free from sedi- ment which would otherwise soon smother them. He also keeps away other animals that try to occupy the nest.
Fierce fights between the males are described by Whitman and Ecleshymer. The male also has to keep off the minnows and sunfish. The eggs hatch in from eight to ten days. When the larv^ are about 12 to 13 mm. in length they leave the nest. Observations have been made upon behavior dur- ing nest forsaking. The movements of the swarm of larvae when leaving the nest suggest that it follows the trail of the male by scent. ^ Special forms of instinctive response in fish. — Mention ^ To test this matter a rough model of an adult Amia was made and covered with black rubber cloth. A freshly killed male fish was then trussed up and attached to the end of a stick so that it could be moved about in the water. The male was then taken to a nest where the larvoe were about ready to leave and made to pass several times over a course leading from the nest to a distance of about a meter. The larvae soon followed this course and assembled about the male. When the model was substituted they paid no attention to it.
124 STUDIES UPON INSTINCT should be made of some highly specialized instinctive modes of response in fish which have been studied in detail by Sumner, Mast, and by several German investigators (Hess, Frisch, etc.)- These responses are of such a character as to make the surface of the animal appear visually like the background upon which it rests. While present to some ex- tent in other fish, it can best be observed in the flatfish.
The flounders ordinarily lie on the bottom and the skin assumes a color and pattern so nearly like that of their environment that it is frequently difficult to see them. On a black bottom they become dark; on a white bottom, pale. Whether similar changes occur with monochromatic light is not so well established. Mast states, however, that the fish when placed on a yellow bottom appear yellow; on a blue bottom, bluish; on a red bottom, reddish, etc. All of these changes in the skin are regulated through the eyes. If the bottom is finely mottled the pattern in the skin as- sumes a fine grain; if coarsely mottled, it assumes a coarse grain. But there is no evidence indicating an actual repro- duction of the configuration of the background. If, after the skin has become adapted to a given bottom, the fish are moved to a different bottom, they tend to return to the original. The reflex mechanism by means of which such changes are accomplished has not been ascertained (see Results of the experimental study of instincts. — The study of the instinctive development of young animals reared in captivity has been exceedingly fruitful. The results so far accomplished may be considered under one of the following heads: I. The initial performance of some instinctive acts.
II. SeriaJ unfolding of instinctive repertoire of animals.
A. Guinea pig, B. Rat, C. Monkey, D. Sooty tern.
III. Quantitative stud}^ of improvement of instinctive functions.
IV. Modification of instinct by means of social influences.
FIRST PERFORMANCE OF ACTS 125 V. Hereditary character of certain instinctive traits.
VI. Waning of instinct; loss through disuse, etc.
I. The initial performance of some instinctive acts. — The current notion of instinct is that it is an act which takes place (without previous exercise or training) when- ever the proper stimulus is first presented. The common notion further states that the instinct, even upon its first appearance, is just as perfect as during the later per- formances. Under the present heading we wish to take up some initial instinctive performances and view them entirely apart from the consideration of their perfectness or ac- curacy. We shall delay such discussions until p. 138 is reached. Certain scientific writers are inclined to doubt even the existence of instincts. To illustrate this type of conception we cite the work of C. S. Berry. Some years ago he reported that the young of the Manx cat learn to catch and kill mice by imitating the parents. He con- eludes that: " Cats are credited with more instincts than they really possess. It is commonly reported that they have an instinctive liking for mice and that mice have an instinctive fear of cats. It is supposed that the odor of a mouse will arouse a cat and that the odor of a cat will frighten a mouse. My experiments tend to show that this belief is not in harmony with the facts. When cats over five months old were taken into a room where mice were they did not show the least sign of excitement. A cat would even allow a mouse to perch upon its back without attempting to injure it, nor did the mouse show any fear of the cat. I have seen a mouse smell at the nose of a cat without showing any sign of fear."
This work of Berry's did a good deal of violence to the common notion of instinct. The later work of Yerkes and Bloomfield seems to reestablish our faith in instinct. They made their tests upon two litters of common cats. The animals were carefully fed upon milk, beef, usually cooked, and fish. They were housed in a room free from mice. In the first week of life the kittens showed no special interest in mice. Shortly after they gained their sight (12 days of age), they were again tested, but the presence of the mice did not elicit the instinctive response sought for. When slightly over 4 weeks of age, 3 of the first lit- ter of 4 failed to exhibit the instinct, as before. One of 126 STUDIES UPON INSTINCT the kittens, now, however, exhibited a type of behavior quite in contrast with that of the other 3. She noticed the mouse soon after it had been placed in the cage, as it moved near her, and quickly seized it, growling the while. The mouse escaped and the kitten gave chase but failed to recapture it before it had climbed to the top of the cage.
Five days later the kittens were again tested. Nos. 1, 2, and 3 acted as before; No. 4, as in the above test, made efforts to catch the mouse. Two days later the tests were repeated. Kittens 1, 2, and 3 showed a marked interest in the mouse, but behaved as in the other tests. No. 4 exhibited almost a complete repertoire of movements used by the adult cat in catching and killing mice. The mouse was pursued, caught, worried, killed, and partially eaten. The tests were continued upon the remainder of the kittens belonging to the same group as No. 4, and upon 4 kittens belonging to a second litter. All the animals tested, at slightly varying ages, exhibited the characteristic in- stinctive behavior of No. 4. The instinct to kill commonly appears at about the end of the second month. It may appear as early as the first month. Since Berry worked with kittens 5 months old it is just possible that his negative results were due to the fact that the instinct had waned through disuse.
Two definite things came out of this study: (1) as regards method, it shows quite clearly the possibility of being able to note the first appearance of an instinct and the variation in the time of its appearance among the same litter of animals; (2) it shows that the initial per- formance of an instinctive act may be fairly complete even though the act as a whole is enormously complex.
One of the most interesting of all responses, performed only once in the life of the individual, is that exhibited by the young bird in gaining egress from the shell. The technique of its method has long been known and may be described somewhat as follows: " The chick chips the egg a little at a time with its bill, and as it does so it turns around inside the egg shell, the axis of its rotation being precisely the long axis of the egg. The consequence of this turn- FIRST PERFORMANCE OF ACTS 127 ing is that the tip of the bill, chipping the shell as it goes, describes a very exact circle around the end of the egg, and thus the large end of the shell is cut off.* The time which it takes to finish the cutting is very variable. When the opening has been extended nearly or quite around the egg, the chick pushes and separates the two sections of the shell, tearing in two any shreds of membrane which may have been holding them together. Thus the little bird effects its escape." (Craig.)
It has recently been shown by Craig that the young' pigeon opens its egg in the same manner as the chick. The activity of the yonng bird actually under observation is extremely interesting. We quote from this author: "No. 46, April 19, 1911. 8;30 a.m. My wife found egg chipped and called me to see it. While we watched it the young chipped the egg about one-third the way round (120°) in about ten minutes. The young made a strong movement, then rested for several seconds, then made another strong movement, and so on. Each movement seemed to consist ( 1 ) of pushing out the large end of the egg, i.e., pushing lengthwise of the egg; (2) thrusting the bill through the shell, the bill coming just far enough to break through^, in fact many times not breaking through but only bulging the shell; ( 3 ) turning around a few degrees. We could see this turning through the hole in the shell; also we could see that each thrust of the bill appeared a little beyond the previous one. With the egg large end up, the turning was anti-clock-wise."
In the drinking and eating responses of many animals we should expect to find a high degree of completeness even in the initial performance. Chickens when left to develop naturally in the vicinity of water and food usually find the water by accidentally pecking at the dish containing it. The contact of the water on the beak sets off the rest of the mechanism. At times, apparently, drinking may be hastened by imitating other chicks. When the intra- organic stimulation becomes sufficiently intense, which is brought about by keeping the chick from water until the third day, almost any form of extra-organic stimulation will set off the act of drinking. Under this condition neither imitation nor fortuitous pecking at objects in the water are necessary to obtain the response. The dove does not instinctively give a drinking response to the sound.
3 Breed, however, denies that egress is accomplished mainly by pecking. He maintains that lifting movements are chiefly responsible.
128 STUDIES UPON INSTINCT of water or sight, nor to the touch of it on distal parts of the body. The young dove, like the chick, gets its bill into water probably chiefly by pecking, either at objects in the water which attract its attention, or in imitation of the old birds. When studied in detail it was found that for the first three or four times the bird has accidentally to peck at the water and to get it inside the beak before the drink- ing response is set up. But on the fourth time and always thereafter the response was perfect. Certainl}^ in the case of the terns the feeding responses are quite indefinite at first, and on the part of many birds it is wholly a passive kind of process; the bird merely opens the beak and w^aits until the food is dropped in.
The young guinea pig is born in a highly developed state. Even at the end of the first day the animal is able to eat grass, bread and milk, and nibble at a carrot. It is able to run around and can find the breast of the mother unaided. This is quite in contrast with the case of the wdiite rat. The young rat is extremely helpless and the mother takes the active part in feeding.
In the case of many vertebrates there seem to be no in- herent methods of responding actively to different food stuffs. The young chick will peck at almost any small ob- ject. The young tern, shortly after hatching, will peck at its own toes, at small bright objects, and at fecal matter, and will even attempt to swallow matches. Even up to 10 and 12 days after being fed by hand upon minnows, they have to be watched constantly to prevent the swallow- ing of shells, seaweed, and other objects.
II. Serial unfolding of instincts in young captive ani- mals.— We shall attempt to give in this connection a summary of the development of a few animals that have been carefully observed in captivity.