Forced Movements, Tropisms, and Animal Conduct

With the same method Lillie tried to prove that the spermatozoa of Nereis and Arbacia are positively chemo¬ tropic to extracts of their own eggs.^*^ jje proceeded as foUows: A' suspension of Arbacia sperm, freshly made, was put under a raised cover slip and a drop of the super¬ natant sea water which had been standing over eggs (as inBuller’s experiments) was introduced under the cover slip. Observation with the naked eye showed that around this drop of egg-sea water iromediately a dense ring of spermatozoa formed and behind this a clear external zone was formed about 1.2 to 2 mm. wide. The dense ring then broke up into small agglutinated masses. In Lillie’s opinion the formation of this dense ring of spermatozoa at the periphery of the egg-sea water is the expression of a positive chemotropism of the spermatozoa for a sub¬ stance contained in the egg-sea water, the “fertilizin.” He assumes that the spermatozoa near the drop of egg- sea water all swim to the egg-sea water, leaving a clear space behind them. While this explanation of the ring formation might be true—if supported by a direct chemo¬ tropic method like Pf eff er’s—it can be shown that the ring formation is in ah probability due to an entirely different

phenomenon which has no relation to chemotropism or any other tropism. BuUer had already observed that the supernatant sea water of sea urchins contains a substance which causes the agglutination of spermatozoa.®® A drop of sea water in wliich egg:s had been deposited was placed upon a slide and a drop containing spermatozoa near it. On joining the drops a large number of small balls were formed in a very few seconds. When very numerous spermatozoa were present the balls became 0.1 mm. in diameter, containing many thousands of spermatozoa packed together in a dense mass.

Buller explains the phenomenon as being due to small bits of egg jelly floating in the sea water SO small that they will (like spermatozoa) pass through ordinary filter paper and, so transparent that one cannot directly see them. A few spermatozoa become attached to each piece of jelly, the presence of which may be inferred from the manner in which the small groups of sperma¬ tozoa move about. Owing to the length of the spermatozoon, although its head may be imbedded in a jelly particle, the tail may remain partly free. The little collections of spermatozoa thus mo-ve about hither and thither in no particular direction. When tWo such groups come by accident into contact they fuse. Certain of the spermatozoa adhere to both little masses of jelly and lock them together. The fused mass combines with other simple and fused masses, and so on.<=

The writer was able to show that when the jelly of the egg of Strongylocentrotus purpuratus is dissolved by an acid treatment the eggs when washed and transferred to sea water no longer give off agglntinating substances, while the acid sea water containing the dissolved jelly, when rendered neutral through the addition of alkali, will canse the agglutination of sperm.^^^ While all the jelly can he washed off with an acid treatment in the egg of purpuratus, the same is not true for the egg of Arhacia

c This explanation of the fusion of two clusters to a larger one is per¬ haps not correct. The writer is inclined to a&cril>e it to the adhesion or agglutination of the spermatozoa of two neighboring clusters with each other, due to a sticky surface on the sperm head. of Woods Hole. Here tlie acid treatment does not as a rule dissolve all the jell 7 , or possibly some new jelly may be given off by the egg. While Buller may be correct in assuming that micro¬ scopic pieces of the egg jelly form the center of these sperm clusters, the writer reached the conclusion that the dissolved mass of the jelly makes the surface of the spermatozoa transitorily sticky, so that if they impinge against each other they will stick together for some time, until the sticky compound formed by the jelly on the sperm head is dissolved by the sea water, which occurs after a short time.

This agglutinating effect of the egg-sea water upon the sperm of Arhacia gives rise to that ring formation which Lillie considers a proof of positive chemotropism. When a drop of egg-sea water is put into a sufficiently dense suspension of spermatozoa, the spermatozoa at the surface of the drop will agglutinate into practically one dense ring around it, and through the diffusion of some of the dissolved jelly through this ring numerous little clusters will form at the external periphery of the ring, and these clusters will fuse with the ring. In this way the clear region behind the ring originates. The process of fusion continues inside the ring with the result that the latter breaks up into numerous bead-like spherical clusters as Lillie described. In a former paper the writer has pointed out the analogy between the phenomena of transitory sperm agglutination (under the influence of egg-seawater) and surface tension phenomena, inasmuch as two small clusters upon coming in contact fuse into one larger one and inasmuch as elongated clusters break up into two or more spherical clusters.

in the opinion of the writer no connection with positive chemotropism.*^ 4. The method of Pfeffer cannot well he nsed for larger organisms. Barrows has devised an apparatus which allowed him tO' test quantitatively the chemotropic reactions of Drosophila. The flies which are positively heliotropie were allowed to go to the light inside of a narrow hollow groove. At a certain spot of the groove two glass bottles were inserted with their openings oppo¬ site each other, one of which contained the substance to be tested for chemotropic efficiency, while the other served as a control. The number of flies which on their path were deviated by the bottle containing the substance to be tested were counted and their number compared with that going into the control bottle. The collection of odor¬ ous matter in the groove was removed by suction. In this way it was possible to ascertain that the flies are posi¬ tively chemotropic to ethyl and amyl alcohol, acetic and lactic acid, and to ether. The chemotropic effect of alcohol was increased through the admixture of traces of an ester, e.g., methyl acetate.

In describing the manner of reaction of these flies, Barrows makes the statement that when the odor is weak the fruit fly “attempts first to find the food by the method of trial and error, but as the fly passes into an area of greater stimulation, these movements give way to a direct orientation. This orientation is a well defined tropism response.” A similar statement had been made by d Lillie also assumes that it is the intensity OTadient which determines the direction of motion in tropistic reactions. This is not correct, since posi¬ tively heliotropie animals go to the light even if by so doing they have to go from strong into weak light (see page 50). The direction of motion in tropistic reactions is determined by differences in the mass of chemical substances on both sides of a symmetrical animal.

Harper for the heliotropisra of certain worms, namely that in strong light the animals move by heliotropism, in weak light by “trial and error.” These statements are as erroneous as the assertion that while a stone falls under the influence of gravity a feather finds its way down by the method of “trial and error.” Barrows and Harper overlook the role of mass action and reaction velocity. When an animal is struck on one side only by light or by a chemically active substance emanating from a center of diffusion, the mass of this substance or of the photochemical reaction product in¬ creases on this side. These substances react with some substance of the nerve endings and as soon as the mass of the reaction product reaches a certain quantity the automatic turning, the tropistic reaction, occurs. When the light is strong or when the animal is near the center of diffusion, this happens in a short time and the tropistic character of the reaction is striking, since the animal is quickly put back into its proper orientation if it deviates from it. When the light i^ weak or when the animal is at some distance from the center of diffusion it will take a longer time before this critical value of the reaction prod¬ uct is reached, and in this case the animal can deviate considerably out of the correct orientation before it is brought back into the right orientation.

Under the name of thermotropism M. Mendels¬ sohn 352-335 jjas described the observation that Paramcecia gather at a definite end of a trough when these ends have a different temperature. The organisms were put into a flat trough resting on tubes through which water was flowing. When the water in the tube had a temperature of 38° at one end of the trough, while the tube at the opposite end was perfused by water of 26° the organisms all gathered at the latter end. If then the temperature of the water in the two tubes was reversed the organisms went to the other end of the trough. If one end had the temperature of 10° the other of 25°, all went to the latter end. In this case we axe in all probability not dealing with a tropistic reaction but with a collection of organisms due to the mechanism of motion described for Paramcecium by Jennings. When these organisms come suddenly from a region of a moderate temperature to one of lower tem¬ perature the activity of their cilia is transitorily reversed, but owing to the asymmetrical arrangement of their cilia they do not go back in the old direction but deviate to one side. This can lead to a collection of Paramcecia such as Mendelssohn described.

The teleological way of analyzing animal conduct has predominated to such an extent that there has heen a tendency to connect aU animal reactions with the preser¬ vation of the individual and the species. Instincts are considered to be such reactions of the organism as a whole which lead to the nutrition of the individual, the mating of the two sexes, and the care of the offspring. If the tropism theory of animal conduct is justified it must he possible to show that instincts are tropistio reactions.

We have insisted in previous chapters that animals indifferent to light can he made strongly positively or negatively heliotropic by certain chemicals or vice versa ( e.g., the experiments on certain fresh water crustaceans with acids or alcohol and caffein). We know that the body itself produces at various periods of its existence definite hormones and such hormones can act similarly as the acids or the caffein in the experiments on crustaceans, since it makes no difference whether such substances as acid are introduced into the blood from the outside or from certain tissues of the animal’s own body. We know through F. Lillie’s observations that in the blood of the male cattle embryo substances circulate which inhibit the develop¬ ment of secondary sexual characters of the female embryo, and we know through Steinach’s experiments that the in¬ termediate tissue from the sexual gland of one sex when introduced into the castrated organism of the opposite sex may impart to the latter the sexual instincts of the

former. Hormones produced by definite tissues, there¬ fore, influence the instincts. We want to show that this influence is due to a modification of tropistic reactions by the hormones. Alating in certain fish, like Fundulus, consists in the male pressing that part of its body which contains the opening of the sperm duct against the corresponding part of the female body. The latter responds by pressing back, and the pressure of the body is maintained by both sexes through motions of the tail. During this mutual pressure or friction both sexes shed their sexual cells, sperm and eggs, into the water, and since the openings of the cloaca of the male and female, through which the sex cells are shed, are brought almost in contact with each other, sperm and eggs mix at the moment they are shed. This act of mating is due to a stereotropism which exists only during the spawning season and which is supposedly due to cer¬ tain hormones existing at this time in the animal. The existence of such hormones is also indicated by certain colorations which develop and exist in the male during this period. This stereotropism is to some extent specific since it is exhibited by the contact between the two sexes. The specificity of this stereotropism is of importance and needs further experimental analysis, but that it is in reality a type of common stereotropism is evidenced by the fact that if during the spawning season we keep females isolated from males in an aquarium the females will go through the motions of mating and shed the eggs every time they come in contact with the glass walls of the aquarium. When they are kept permanently is olated from the male they repeat this non-specific purely stereotropic mating throughout the season. The eggs which they shed they quite frequently devour.

These manifestations of a highly developed stereo- tropism in the segments of the reproductive organs are probably widespread in the animal kingdom. The late Professor "Whitman told the writer that male pigeons when kept in isolation will try to go through the motions of mating with any solid object in their field of vision, e.g., glass bottles, and even with objects which give only the optical impression of a solid, namely, their own shadow on the ground.

In ants, the winged males and females become intensely positively heliotropic at the time of mating. Copulation occurs in the air, in the so-called nuptial flight. At a cer¬ tain time'—in the writer’s observation toward sunset, when the sky is illuminated at the horizon only—the whole swarm of males and females leave the nest and fly in the direction of the glow. The wedding flight is a heliotropic phenomenon®®’’ presumably due to substances produced in the body during this period. After copulation the female loses its wings and also its positive heliotropism.®' It becomes now intensely stereotropic. When kept in a dark box with pieces of cloth in folds the wingless female will now be found in the folds where its body is as closely as possible in contact with the solids. This positive stereotropism leads the queen to begin a subterranean existence which marks the founding of a new nest. Helio- tropism and stereotropism are, therefore, the controlling factors in mating and the starting of a new nest in these ants.®®’’

V. L. Kellogg ®®® has made observations which show that the nuptial flight in bees is also due to an outburst of positive heliotropism as in the ant. a It has already been mentioned that artificial removal of the wings of the fruit fly will also abolish its heliotropism. In the course of some experiments on the sense-reactions of honey¬ bees, I have kept a small community of Italian bees in a glass-sided, narrow, high observation hive, so made that any particular bee, marked, which it is desired to observe constantly, can not escape this obser¬ vation. The hive contains but two frames, one above the other, and is made wholly of glass, except for the wooden frame. It is kept covered, except during observation periods, by a black cloth jacket. The bees live contentedly and normally in this small hive, needing only occasional feeding at times when so many cells are given up for brood that there are not enough left for sufficient stored food supplies. Last spring at the normal swarming time, while standing near the jacketed hive, I heard the excited hum of a beginning swarm and noted the first issuers rushing pellmell from the entrance. Interested to see the behavior of the community ixn the hive during such an ecstatic condition as that of swarming, I lifted the cloth jacket, when the excited mass of bees which was pushing frantically down to the small exit in the lower comer of the hive turned with one accord about face and rushed directly upward away from the opening toward and to the top of the hive. Here the bees jammed, struggling violently. I slipped the jacket partly on; the ones covered turned down; the ones below stood undecided; X dropped the jacket completely; the mass began issuing from the exit again; I pulled off the jacket, and again the whole community of excited bees flowed—that is the word for it, so perfectly aligned and so evenly moving were all the individuals of the bee current—up to the closed top of the hive. Leaving the jacket off permanently, I prevented the issuing of the swarm until the ecstasy was passed and the usual quietly busy life of the hive was resumed.

About three hours later there was a similar performance and failure to issue from the quickly unjacketed hive. On the next day another attempt to swarm was made, and after nearly an hour of stmggling and moving up and down, depending on my manipulation of the black jacket, most of the bees got out of the hive^s opening and the swarming came off on a weed bunch near the laboratory. That the issuance fro‘m the hive at swarming time depends upon a sudden extra-development of positive heliotropism seems obvious. The ecstasy comes and the bees crowd for the one spot of light in the normal hive,. namely, the entrance opening. But when the covering jacket is lifted and the light comes strongly in from above^—my hive was under a skylight—^they rush toward the top, that is, toward the light. Jacket on and light shut off from above, down they rush; jacket off and light stronger from above than below and they respond like iron filings in front of an elwtromagnet which has its current suddenly turned on.

Finally there are indications of the role of ohemo- tropism in mating. It has been observed for a long time . that if a female butterfly is kept hidden from sight in a not too tightly closed box, male butterflies of the same species will be attracted by the box and settle on it. The female apparently gives off a substance to which the male is positively chemotropic. All these observations should be worked out more systematically. The data suffice, however, to indicate that what the biologist and psycholo¬ gist call instinct are manifestations of tropisms.

The fact that eggs are laid by many insects on material which serves as a nutritive medium for the offspring is a typical instinct. An experimental analysis shows again that the underlying mechanism of the instinct is a positive chemotropism of the mother insect for the type of sub¬ stance serving her as food; and when the intensity of these volatile substances is very high, i.e., when the insect is on the material, the egg-laying mechanism of the fly is auto¬ matically set into motion. Thus the common housefly will deposit its eggs on decaying meat but not on fat; but it will also deposit it on objects smeared over with asafoetida, on which the larvae cannot live. Aseptic banana flies will lay their eggs on sterile banana, although the banana is only an adequate food for the larvae when yeast grows on it. It seems that the female insect lays her eggs on material for which she is positively chemotropic, and this is generally material which she also eats. The fact that such material serves as food for the commg genera¬ tion is an accident. Considered in this way, the mystic aspect of the instinctive care of insects for the future generation is replaced by the simple mechanistic concep¬ tion of a tropistic reaction. In this case natural selection plays a r61e since species whose females would too fre-

quently lay their eggs on material on which the larvae cannot thrive would be hable to die out. As an illustration of the role of tropisms in the instinc¬ tive self-preservation the -writer wishes to apologize for selecting an example which he has used so often in pre¬ vious discussions, namely the role of heliotropism in the preservation of the life of the caterpillars of Porthesia chrysorrhosa.^^’’ This butterfly lays its eggs upon'a shrub, on which the larvae hatch in the fall and on which they hibernate, as a rule, not far from the ground. As soon as the temperature reaches a certain height, they leave the nest; under natural conditions this happens in the spring when the first leaves have begun to form on the shrub. (The larvae can, however, be induced to leave the nest at any time in the winter, provided the "temperature is raised sufficiently). After leaving the nest, they crawl directly upward on the shrub where they find the leaves on which they feed. If the caterpillars should move down the shrub they would starve, but this they never do, always crawl¬ ing upward to where they find their food. What gives the caterpillar this never-faihng certainty which saves its life and for which the human being might envy the little larva? Is it a dim recollection of experiences of former generations, as Samuel Butler would have us believe ? It can be shown that this instinct is merely posi¬ tive heliotropism and that the light reflected from ^e sky guides the animals upward. The oaterpUlars upon waking from their winter sleep are violently positively heliotropic, and it is this heliotropism which makes the animals move upward. At the top of the branch they come in contact ■with a growing bud and chemical and tactile influences set the mandibles of the young caterpillar into activity. If we put these caterpillars into closed test tubes which lie

witli tlieir longitudinal axes at right angles to the window they will all migrate to the window end where they will stay and starve, even if we put their favorite leaves into the test tube close behind them. These larvse are in this condition slaves of the light. The few young leaves on top of a twig are quickly eaten by the caterpillar. The light which saved its life by mak¬ ing it creep upward where it finds its food would cause it to starve could the animal not free itself from the bondage of positive heliotropism. After having eaten it is no longer a slave of light but can and does creep down¬ ward. It can be shown that a caterpillar after having been fed loses its positive heliotropism almost completely and permanently. If we submit unfed and fed caterpillars of the same nest to the same artificial or natural source of light in two different test tubes the unfed will creep to the light and stay there until they die, while those that have eaten will pay little or no attention to the light. Their positive heliotropism has disappeared and the ani¬ mal after having eaten can creep in any direction. The restlessness which accompanies the condition of starva¬ tion makes the animal leave the top of the branches and creep do-wnward—^which is the only direction open to it— where it finds new young leaves on which it can feed. The wonderful hereditary instinct upon which the life of the animal depends is its positive heliotropism in the unfed condition and the loss of this hehotropism after having eaten. The chemical changes following the taking up of the food abolish the heliotropism just as COg arouses positive hehotropism in certain Daphma.

Mayer and Soule have shown that negative geotropism and positive heliotropism keep the caterpillars of Danais plecdppus on its plant (the milk-weed). The chemical nature of the leaf starts the eating reactions, hut “once the eating reaction be set into play, it tends to continue, so that the larva may then be induced to eat substances which it would never have commenced to eat in the first instance.” These few examples may suffice to show that the theory of tropisms is at the same time the theory of instincts if due consideration is given to the role of hormones in producing certain tropisms and suppressing others. A systematic analysis of instinctive reactions from the view¬ point of the theory of tropisms and hormones will prob¬ ably yield rich returns. As an example we may quote the fact that diurnal depth migrations of aquatic animals, consisting in an upward motion during the night and a downward motion during the day, are in all probability determined by a periodic change in the sense of heliotropism.^®®<

When a muscle is stimulated several times in succes¬ sion, the effect of the second or third or later stimulation may be greater than that of the first. A consistently anthropomorphic author should draw the inference that the muscle is gradually learning to react properly. What seems to happen is that the hydrogen ion concentration is raised by the first stimulations to a point where the effect of the stimulation becomes greater. When the stimula¬ tions continue and the hydrogen ion concentration be¬ comes still greater, the response of the muscle declines and finally becomes zero; the hydrogen ion concentration has now become too high. The writer observed that when winged plant lice of a Cineraria were taken directly from the plant, they did not react as promptly as after they had gone through several heliotropic experiments. There is nothing to indicate that this is a case of “learning,” since it may also be the result of a change in the hydrogen ion concentration or of some other reaction product. It may also be the result of some purely mechanical obstacle to rapid locomotion being removed.

We can speak of learning only in such organisms in which the existence of associative memory can be proved. By associative memory we mean that mechanism, by which a stimulus produces not only the direct effects determined by its nature, but also the effects of entirely different stimuli which at some former period by chance attacked the organism at the same time with the given stimulus. Thus the image or the odor of a rose may call up the memory of persons or surroundings which were present on a former occasion when the image or odor of the flower impressed us. Brain physiology shows that this type of associative memory is the specific function of definite parts of the brain, e.g., the cerebral hemispheres which exist only in definite types of animals. We see also that certain species among vertebrates, insects, Crus¬ tacea, and cephalopods possess associative memory, while to the knowledge of the writer no adequate proof for its existence has ever been given for worms, starfish, sea urchins, actinians, medusae, hydroids, or infusorians.*'®* Claims for the existence of such memory in these latter groups of animals have frequently been made, but such claims are either plain romance or due to a confusion of reversible physiological processes with the imeversible phenomena of associative memory. The less a scientist is accustomed to rigid quantitative experiments, the more ready he is to confound the reversible after effects of a stimulus— e.g., the after effects due to an increase in hydrogen ion concentration—^with indications of associa¬ tive memory. Learning is only possible where there exists a specific organ of associative memory, the physical mechanism of which is still unknown.

The manifestations of associative memory are gener¬ ally discussed by the introspective psychologists, who as a rule are not familiar with or do not appreciate the methods of the physicist. There have been made repeated attempts to develop methods for the analysis of associa¬ tive memory, among which thus far only one satisfies the demands of quantitative science, namely Pawlow’s method. As is well known even to the layman, eating causes a flow of saliva. The quantity of saliva excreted

by the parotid (one of the salivary glands) in the dog can he collected and measured. The earlier physiological workers had observed that in a dog which had often been used for the study of the influence of eating upon the flow of saliva, the saliva began to flow whenever the prepara¬ tions for feeding were made before the eyes of the dog, even when no food was given. Pawlow made use of this fact to study quantitatively tlie “strength” of such asso¬ ciative phenomena, which he terms “conditioned re¬ flexes” (to escape the terminology and interpretations of the introspective psychologist).®” A fistula®' of the duct of the parotic gland allows the saliva to flow outside the cavity of the mouth. This fistula is connected with a long manometer which by a special air chamber arrange¬ ment gives a considerable change in the height of the meniscus for the secretion of as httle as one drop of saliva. The variations of the height of the column of liquid in the manometer are observed outside of the room where the dog is. For each dog which is to serve for such experiments the meal is preceded by a certain signal, the sounds of a metronome of definite rhythm, or a definite musical sound, or a definite optical signal, and so forth, which is to form the special conditioned reflex for this dog. After a certain number of repetitions the association is established and from now on the flow of saliva commences from the dog’s parotid when the typical signal is given. It was found that the quantity of saliva excreted by the signal changes in a definite sense and quantity when the signal varies or when other conditions accompanying the signal vary.

a The writer is indebted for the details of Pawlow's method to a short review by Dr. Morgulis.cs^^ sss Thus in one dog “hy persistent training a conditioned reflex has been established to the stimulation with 100 oscillations per minute of the metronome. The stimu lation of intermittent sounds of such frequency called forth 6 to 10 drops of saliva every time. The interval between successive oscillations was then modified, the moment of the disappearance of the conditioned salivary reflex indicating the lowest limit of ditferentiation. With¬ out going into any details of this most interesting investi¬ gation or quoting actual data, I will say that the dog could sharply distinguish the shortening of the interval by less than 1/40 to 1/43 of a second. Indeed with the well-developed reflex to the stimulation of 100 beats per minute a change of the rate to either 96 or 104 beats was immediately reacted upon by a marked diminution or even complete cessation of the flow of saliva.”

This example will give an indication how sensitive is this method of measuring the effect of a memory association. It is not our purpose to give the details of Pawlow’s results—they have only been published in Russian and are therefore not accessible to the writer—^but to show that the influence of an associative memory image is as exactly measurable as, e.g., the direct illumination of the eye; and moreover that what we call a memory image is not a “spiritual” but a physical agency. We there¬ fore need not be surprised to find that such memory images or “conditioned reflexes” can vary and multiply the number of possible tropistic reactions.

We have mentioned in the previous chapter that the stereotropism in the mating instinct includes apparently an element of species specificity inasmuch as naturally only males and females of the same species mate. The late Professor Whitman has shown hy experiment that f.liiR specificity is, in pigeons at least, not inherited hut the effect of memory images (a “conditioned reflex” in the sense of Pawlow). Whitman took the eggs or young of wild species, giving them to the domestic ring-dove to foster, with the result, that the young reared by the ring¬ doves ever after associated with ring-doves and tried to mate with them. Passenger pigeons when reared by ring¬ doves refuse to mate with their own species but mate with the species of the foster parents.®®* This shows inciden¬ tally that racial antagonism is notdnherited but acquired.

We have mentioned the fact that the mating iastinct is determined by tropisms aroused by specific internal secretions, and that in isolated male pigeons any solid body can arouse the mating reaction. Craig raised male pigeons in isolation so that they never came in con¬ tact with other pigeons until they were adult. One pigeon was hatched in July and isolated in August. Througliout the autumn and early winter this bird eooed very little. But about the first of February there began a remarkable development of voice and social behavior. The dove was kept in a room where several men were at work, and he directed his display behavior toward these men just as if they belonged to his own species. Each time I put food in his cage he became greatly excited, charging up and down the cage, bowing-and-cooing to me, and pecking my hand whenever it came within his cage. From that day until the day of his death, Jack continued to react in this social manner to human beings. He would bow-and-coo to me at a distance, or to my face when near the cage; but he paid greatest attention to the hand—^naturally so, because it was the only part with which he daily came into direct contact. He treated the hand much m if it were a living bird. Not only were his own activities directed toward the hand as if it were a bird, but he received treatment by the hand in the same spirit. The hand could stroke him, preen his neck, even pull the feathers sharply, Jack had absolutely no fear, but ran to the hand to be stroked or teased, showing the joy that all doves show in the attentions of their companions.