* These writers hold that a given amount of practice is necessary to smooth the way for the operation of a native capacity whose efficiency is largely a function of the age of the animal. They hold that such a curve has two parts — the early state (2 days of practice) showing effects of practice; the latter stage (from this point on, where a slow but steady increase in accuracy is shown) showing the slow process of maturation. This conclusion is surely questionable and certainly confusing. The curve exhibits what every habit curve in animals shows, i.e., a rapid initial increase in efficiency and then a slow increase until the level of no further improvement is reached. Unless the writers are willing to admit that the latter part of every habit curve is illustrative of maturation, it seems useless to use such a terminology here. They may intend to imply this; but what such " maturation " would mean in the case of habit formation in the adult is not clear. The authors are not quite consistent in the use of the term anyway. In the beginning of the article one gets the idea that they mean what one commonly would mean by it, i.e., actual growth, increase in size and length of conductors, possible increase in the number of connections, increase in size and efficiency of the effector, etc., and possibly certain structural growth of the receptors. In the latter part of the article, they use it without giving the reader a clear idea of how they would define it. These growth processes may influence the rate of improvement in the func- tion whose accuracy they measured, but the writers ofTer no real evidence of it, since exactly the same results are obtained whenever habit curves are recorded.
142 STUDIES UPON INSTINCT in the field by the parents, possess songs which are char- acteristic. Until recent years it was supposed that the characteristic songs of birds were inherited, like instincts. Apparently this is not wholly true. It would seem from the work of Scott and Conradi that what the birds inherit is a strong tendency to sing, but that no characteristic song develops without training. What are inherited are throat formation and vocal apparatus suitable to produce the song of the race if the bird is thrown under the proper influences (habit). Scott isolated Baltimore orioles before they had heard any of the songs of their species. He ob- served the birds for a number of years. They became very gentle in their artificial environment. The early vocal reflexes were quite similar to those of the wild birds, e.g., the single call note, the peculiar rattling chatter, etc. The birds finally became good singers. During certain seasons of the year song was incessant. '' It was now a loud clear series of notes of great brilliancy, poured forth in such rapid succession as to be like that of the house wren {T. aedon) in the intervals, and lasting about as long as the warble of that bird. Except for the rattle, which was now and then a part of the repertoire, this song had nothing in it that reminded one of the song of the Baltimore oriole as heard in New York, Massachusetts, or at any other point where the birds occur." The birds dwelling thus in isolation developed a song of their own. The ex- perimenter next put a young brood of Baltimore orioles 6 days of age to live with 2 adult birds that had been brought up in isolation. This young brood began to sing at the proper age, but to sing the song of their 2 adult associates. This same author found very great modification in the songs of the various species of birds when some 15 to 16 different species were brought up to- gether and within hearing of one another's songs and the songs and call notes of their neighbors on the outside of the aviary. Some of the.birds were more resistant to such influences than others. The robin had what Scott called an invented song. The wood-thrush's song, while varying much from the normal, was not original. All of the cat- EFFECT OF SOCIAL INFLUENCE 143 birds showed mimicrj^ of the songs of other birds. In par- ticular he found a yellow-breasted chat learning to imitate the postman's whistle with such a degree of accuracy that the experimenter was often deceived. One of the red-wing blackbirds crowed constantly for two months in the year. The crow was in imitation of the crow of a bantam rooster. Conradi reared a young English sparrow in the same room with canaries. The regular sparrow chirp developed at the proper time, although the bird had never heard this chirp from the outside. Probably under the influence of his canary-bird environment, he lost the chirp and took on the peep similar to the peep of the young canaries. When the bird was about three and a half months old ' ' he constantly chimed in with the canaries in his own fashion, giving a low note followed by a few high notes, with now and then some slurring from a high to a low note, similar to the notes the canaries have in their overtures. He joined the canaries freely for a few days, when he became ill and was silent for a week." Two weeks later he again joined the canary chorus. In general Conradi says that the spar- row's efforts resembled the confusion of notes which oc- curred when all three of his adult canaries were singing at their best. These experiments were repeated upon other sparrows, in some cases even with more marked success. The sparrows so trained by the canaries were then removed to a place where they would be away from the canaries and under the influence of a group of sparrows which frequented that part of the building. They rapidly lost their canary notes and readopted a considerable part of the sparrow chirp. Nevertheless, their voices remained more musical than those of the untrained sparrows. When again- placed under the influence of the canaries they rapidly re- gained what they had lost.
Breed carried out some experiments upon the chick to determine the effect of social influence upon the develop- ment of the pecking instinct. In the one ease the chick was allowed to develop its pecxmg instinct in isolation; in another, the chick was allowed to develop the pecking reaction where it could watch the pecking of older birds. In 144 STUDIES UPON INSTINCT general it was found that the effect of social influence was practically nil in increasing the rate of improvement in accuracy of the pecking reaction. Curve IV-F, Fig. 29, p. 139, shows the growth of the reaction where tuition is allowed; curve IV-E where tuition is denied. In general the two curves are closely similar. Apparently both the dove and the chick are influenced to some extent by their mates, so far as the initiation of the drinking and pecking instincts is concerned, i.e., when two or three chicks out of a brood begin to either peck or drink, the act serves as a stimulus and causes the others to engage in the same activity.
V. Hereditary character of certain instinctive acts and traits. — Yerkes has so far made the only contribution bearing directly upon the heredity of instinctive acts and traits. He has recently tested the heritability of savage- ness, wildness, and timidity in certain strains of rats. He first tested the rats and graded them with respect to the presence of these traits. On regrading after an interval of several days he found that the judgment as to the grades of savageness or wildness remained almost constant.
Savageness was indicated by (1) biting; (2) exposing or gnashing the teeth; (3) jumping at hand or forceps; (4) squeaking. Wildness by (1) attempts to hide from view in cage or in hand; (2) random or excited running about in the cage or excited attempts to escape from the hand or forceps; (3) squeaking; (4) urination and defecation. Timidity by (1) attempts to avoid the experimenter; (2) by a kind of chattering or gnashing of the teeth; (3) by cowering and what looks like trembling; (4) urination and defecation. Timidity is possibly not sharply enough marked off from the other traits to measure it quantita- tively.
The wild rats possessing these traits were then mated with strains of tame rats which had been bred in the Harvard laboratory for at least 10 years. The offspring (Fi) from these crosses were tested, as likewise the progeny (F2) obtained by mating the individuals of F^ among themselves. Yerkes holds that his tests show conclusively WANING OF INSTINCTS 145 that savageness, wildness, and timidity are heritable be- havior complexes.^ This same author tested the heritabil- ity of the mode of whirling in the dancing mouse. The dancers may be classified into right, left, and mixed whirl- ers. In one line of descent (No. 400) he found the tendency to whirl to the left was heritable.
In view of observations of Franz that certain monkeys are left-handed and certain others right-handed, it would be interesting to test the heritability of the preferential use of right or left hand.
VI. Waning of instinct: loss through disuse, etc. — In the case of mammals one instinct common to many species exhibits the process of waning very clearly. This instinct in question is that of carrying the young from place to place in the mouth. The following quotation is taken from one of the author's early observations upon the white rat: "Within the last two days (when young were from 12 to 14 days of age) a change has come over the mother. Wlien the rats were 10 days old the mother would become frantic if one of them were taken beyond her grasp. At 14 days of age she was not at all worried when I took one of them out for an experiment. Indeed for two days past she has not received the young rats in her mouth when they entered the cage, but has allowed them to scramble through the straw to her as best they could."
While this instinct wanes as the general cycle of instincts changes, it reappears with all its vigor on the birth of the next litter of young. The process of weaning in nearly every case is a gradual one. We have under observation a young monkey (M. rhesus) nearly 18 months of age which is still nursing. There is no actual necessity for nursing longer than 18 weeks (even less).
There are no reliable observations which have been experimentally controlled on the loss of instinct through disuse. The case of the Scotch terrier attempting to bury food in the carpet and then finally giving over the attempt on all subsequent occasions has often been cited. ^ The ^ Yerkes' complete report has not yet appeared. ^ An interesting case of elaboration of this instinct to bury food appears in an observation made by Frost {Journal of Animal Be- 146 STUDIES UPON INSTINCT observation comes from William James. The observation, so far as it goes, is unquestionably true, but there is no evidence that the instinct is lost. Given the proper en- vironment and the proper conditions of hunger, and we should expect to see the instinct reappear in all its pristine vigor. In the further case cited by Mr. James, viz., that of the young goslings kept away from water, a similar lack of control in the experiment is apparent. Spalding, who made the test, states that when the goslings were reared in a kitchen away from water until they were several months old and then taken to a pond, they not only refused to go into tlie water but scrambled back to the bank the moment they were thrown into it. It is quite possible that if Spalding had taken goslings that had been reared in the water and tossed them in in this way, they would have done exactly the same thing. There is room for in- teresting work upon this question. That instincts are overlapped and obscured by later habits is unquestionable. Whether this process of obscuration results finally in the complete elimination of the instinct is certainly not proven by any observation we have in the field at the present time.
BIBLIOGRAPHY BIBLIOGRAPHY Allen, J., "The Associative Processes of the Guinea Pig," Jour.
Comp. Neu. and Psych., 1904, XIV, 293. Beery, C. S., "An Experimental Study of Imitation in Cats," Jour.
Comp. Neu. and Psych., 1908, XVIII, 1. Breed, Fred S., " The Development of Certain Instincts and Habits in Chicks," Behavior Monographs, Ser. No. 1. Breed, Fred S., and Shepard, J. F., "Maturation and Use in the Development of an Instinct," Jour. Animal Beh., 1913, III, 274. CoNRADi, E., " Song and Call Notes of English Sparrows When Reared by Canaries," Am. Jour. Psych., 1905, XVI\ 190. Craig, Wallace, "Behavior of Young Birds in Breaking Out of the Egg," Jour. Animal Beh., 1913, II, 296. " Observations on Doves Learning to Drink," Jour.
havior, 1913, p. 145). The squirrel at times buries nuts in separate caches, and on later occasions revisits these separate caches and assembles all in one hiding place. This was observed while snow completely covered the ground. The finding of the separate hiding places seems to be almost instantaneous with the squirrel.
BIBLIOGRAPHY 147 Franz, S. I., " Observations on the Preferential Use of the Right and Left Hand by Monkeys," Jour. Animal Beh., 1913, III, 140. James, William, Principles of Psychology. New York, Henry Holt, Lashley, K. S., and Watson, J. B., " Notes on the Development of a Young Monkey," Jour. Animal Beh., 1913, III, 114. Newman, H, H., " The Habits of Certain Tortoises," Jour. Camp.
Peckham, G. W. and E. G., Wasps, Social and Solitary. Boston, Reighard, Jacob, " The Natural History of Amia Calva Linnaeus," Mark Anniversary Volume, 1903. Scott, W. E. D., " Song in Birds," Science, 1901, XIV, 522.
" Data on Song in Birds," Science, 1902, XV, 178.
"The Inheritance of Song," Science, 1904, XIX, 154, "Rearing Wild Finches," Science, 1904, XIX, 551.
Sherrington, G. S.. The Integrative Action of the ISlervous System.
New York, Scribner's, 1906. Small, W. S., " Notes on the Psvchic Development of the Young White Rat," Am. Jour. Psych.] 1899, XI, 80. Sumner, F. B., " The Adjustment of Flatfish to Various Back- grounds," Jour. Exp. Zool., X, 409. Watson, J. B., " The Behavior of Noddy and" Sooty Terns," Carnegie Wheeler, William Morton, "Ants, Their Structure, Development and Behavior." New York, Columbia Univ. Press, 1910. Yerkes, R. M., The Dancing Mouse. New York, Macmillan, 1907. " The Heredity of Savageness and Wildness in Rats," Jour. Animal Beh., 1913, III, 286. Yerkes, R. M., and Bloomfield, Daniel, " Do Kittens Instinctively Kill Mice?" Psych. Bull., 1910, VII, 253.
CHAPTER V CONCERNING THE ORIGIN OF INSTINCTS Introduction.— Early differentiation of parts. I. Heredity: the con- cept of unit characters. — Unit characters. — Mendel's experiments. II. Origin of diversities in organisms. — Introduction. — Darwin's conception of variations. — Continuous variation due to direct action of the environment upon the developing organism. — The non-inheritance of continuous variation. — Discrete variations or mutations. — Frequency of mutations. — Cause of mutations. — The Darwinian conception of selection. — Changes in the concept of natural selection. — Effect of natural selection upon mutations. — Ability to form habits enables the animal to supplement a faulty inheritance. — Some special forms of adaptations, (a) Protective resemblance. (6) Warning coloration, (c) Sexual dimorphism. HI. Inheritance of acquired characters. — Lamarck's laws. — Re- cent experiments upon the inheritance of acquired characters. — The experiments of Kammerer. — Some evidence negative in char- acter.— The direct adaptation theory. — Summary.
Introduction. — In the preceding chapter we dealt with certain activities on the part of the animal which do not have to be learned. We intimated there that such activities or instincts appear because of the fact that the animal is born with certain systems of arcs ready to function in serial order the moment the appropriate stimulus appears (as has been brought out, many such systems are not ready to function until definite intervals after birth). We wish now to consider especially the origin of such structures. It can readily be seen, since instincts are but the func- tioning of certain preformed structures, that any consid- eration of the origin of the special structures underlying them must inevitably involve a consideration of the origin of the whole organism. It is not our purpose to enter the field of experimental evolution except for the purpose of bringing out certain facts which appear to be helpful in envisaging our own problems. One such group of facts which especially concerns the behaviorist comes from em- bryology.
EARLY DIFFERENTIATION OF PARTS 149 Early differentiation of parts. — Embryology teaches us but little of the early differentiation and growth of par- ticular reflex arcs. In regard to the larger divisions of the nervous system the case is different. The investiga- tions of Harrison, Lewis, Spemann, Hooker, and others show that the gross development of neural structure is V.H Fig. 31. Normal Brain of Frog Embyro V.H., fore brain; M.H., midbrain; Comm. post., posterior com- missure; Ep., epiphysis.
determined early in the course of embryonic development and progresses to a large extent independently of the changes which occur in other organs. Spemann 's experi- ments bring this out with great clearness. He cut out small pieces from the floor of the medullary plate of the frog embryo and replaced them in a reversed position so that the end which before had been anterior was now posterior. The embryos continued to develop; the neural 150 CONCERNING THE ORIGIN OF INSTINCTS grooves closed, and eventually brains were formed. But these brains showed marked abnormalities. In many cases two sets of optic cups were produced and the normal rela- tion of parts within the brain was much disturbed. The structures produced were those which the reversed tissue Pl.chdr.
V.H.
VH.
oc.
M.H.
Comm.post. Lab.
Ep.
Fig. 32. Brain Formed After the Incision of a Portion of the Floor of the Medullary Plate The dotted line indicates the boundary of the disturbed area.
The chief relations of these areas are reversed; a second pair of optic cups has been produced behind the labyrinth, and the thalamus, Zw. H., lies behind the posterior commissure. (Both cuts after Spemann, Zool. Jahrb., 1912, Taf. 1.)
would have produced if undisturbed, but their positions were now reversed.
" The gross form relations of the grafted pieces naturally undergo a certain amount of modification at their boundaries but the recog- nizable portions retain even here their peculiar tendencies to develop- EARLY DIFFERENTIATION OF PARTS 151 ment. The finer differentiation proceeds, not only in the middle of the reversed piece but even at the borders where the cells have been brought into new and abnormal relations. For some time there is a distinct boundary along which the different tissues may be distin- guished by their appearance. Thus it follows that both the gross relationships and the finer differentiation of individual parts of the brain are already determined in the open medullary plate."
Fig. 32 shows such a brain. For comparison a normal brain is also given.
Not only have the various parts of the nervous system this early independence of development but the extra- neural structures as well. Goldfarb destroyed the spinal cord of the newt in the lumbo-sacral region together with the sensory ganglia of the nerves of this region, thus cut- ting off entirely the nerve supply of the hind legs. Re- generation occurred in such animals just as in normal ones, growth and differentiation of the regenerating foot taking place in some cases after the complete degeneration of all nerve elements of the leg. Growth of neural and extra- neural structures thus seems to be relatively independent. It is thus clear that one result of a great part of the work in experimental embryology has been the establishment of the fact that many of the final details of the structure of the adult are already determined in the early cleavage stages of the egg before any differentiation can be recognized. It would seem from this that we can safely assume that what is true of the gross anatomical structure of the brain and nerves is also true of the microscopic structures, of the posi- tion, number, and possible interconnections of the neurones, i.e., of the reflex arcs. If this is true it means that the special systems of arcs underlying instinctive activity are predetermined in some way in the egg. It is probable, furthermore, that at the birth of the animal or soon after- wards all possible nervous connections are already estab- lished and that all later development — all adjustments of the animal to changes in its environment by habit forma- tion involve only changes in resistance through various inherited arcs. Thus the possible habits which an organ- ism may acquire are limited by its nervous structures.
152 CONCERNING THE ORIGIN OF INSTINCTS I. Heredity: the Concept of Unit Characters Unit characters. — These embryological studies from the standpoint of our very special interests teach us little more than that the structure which must later develop into the anatomical mechanism of instinct is already present in the fertilized egg of the given species. They teach us little about the process by means of which one egg will produce a mammal, another a bird, or of the manner in which one chick forms white feathers and another barred. Or put in another w^ay, how it comes about that one mammalian egg will develop into a form which will possess the structures involved in catching and killing mice, and another egg will develop into a form which will gnaw down trees and con- struct a dam. Nevertheless, in spite of our ignorance of the way in which such differences are produced, it has been found possible to determine beforehand, from a considera- tion of the characters of the parents, many characters which will appear in the new-born animal. This may be illustrated by numerous examples. Possibly an illustra- tion of the inheritance of combs in the domestic fowl will be best for our purpose. Three types of combs are com- monly met with. The most common type is the single comb, which is high, relatively narrow, and deeply toothed (Fig. 33 — A). The second type, the pea comb, is smaller, less notched, and is ridged lengthwise (Fig. 33 — B and C). The third type is the ros^ comb, flat, broad, triangular, and covered with numerous papillae (Fig. 33 — D). When fowls bearing pea combs are crossed with others bearing single ones their progeny all have pea combs. In the next gen- eration obtained by interbreeding these hybrids the single comb appears again in one-fourth of the birds. The re- maining three-fourths have pea combs. The same thing occurs when rose and single combed birds are crossed. Their progeny will have rose combs and when interbred will produce rose and single combed birds in the propor- tions of three to one. When rose and pea combed birds are crossed the results are, in the first generation, quite different. The progeny have neither rose nor pea combs, UNIT CHARACTERS but an entirely new type which consists of an irregular fleshy knob at the front of the head, the back of the head being rather bare, the so-called walnut comb (Fig 33 — E), When walnut combed fowls are interbred they give walnut, pea, rose, and single combed progeny in the proportions of 9:3:3:1. There are some irregularities in the proportions, and the mechanism by which they are brought about is too Fig. 33. Various Types of Combs in Fowls A, single comb; B, pea comb: cock; C, pea comb: hen; D, rose comb (bantam) cock; E, walnut comb in young cock. (After Bateson: Mendel's Principles of Heredity, Cambridge, 1909. Permis- sion to use this cut was kindly granted by the press of Gustav Fischer. ) complicated to be entered into here. The important fact for our present purpose is that when the progeny of any cross are interbred among themselves, they produce in their progeny not only their own type, but also that of both their parents, and in quite definite proportions. The char- acters do not lose their identity in hybrids but reappear unchanged in later generations and appear, moreover, in definite proportions which may be determined empirically. The example of the fowls illustrates Mendel's law of segre- gation and heredity, the principle of which may best be 154 CONCERNING THE ORIGIN OF INSTINCTS shown by a brief summary of some of its discoverer's own experiments.
Mendel's Experiments: Mendel bred different varieties of com- mon garden peas and studied, among other characters, the form and coloration of the seeds. When he bred plants from varieties bearing smooth seeds with others from varieties bearing wrinkled seeds, he found that all the seeds produced from this cross mating were smooth. The character of wrinkledness seemed to have been lost entirely. But the next generation, grown from these seeds, when interbred among themselves gave 7,324 seeds, of which 5,474 were round and 1,850 were wrinkled; a proportion of about 3 to 1. The character of wrinkledness had not been lost completely but had only withdrawn from observation. Each of the two characters retained its own in- dividuality in the cross. In the next generation from these plants Mendel found that wrinkled seeds produced only wrinkled ones; the smooth seeds gave rise to plants of two sorts. One-third of them gave plants which produced only smooth seeds even after several generations of close inbreeding; the remaining two-thirds produced both smooth and wrinkled seeds and again in the proportion of 3 to 1. They were in all respects like the first generation hybrids.
If we suppose that the two characters are absolutely independent of each other, that either one or both may be present potentially in the organism, and that the presence of one, roundness, hides the presence of the other, these proportions may be explained. The first generation hybrids receive from one parent the capacity to be round, from the other the capacity to be wrinkled. The round- ness prevails in this generation, but the thing which causes wrinkledness, the determiner, persists unchanged though hidden. "When the adults of this generation produce new germ cells, these germ cells may receive either determiner for roundness or that for wrinkledness, not for both. Since all the hybrids produce both egg cells and pollen grains containing both determiners in equal numbers, all possible pairing of the determiners may occur in the fer- tilized eggs, i.e., individuals may be formed having either two determiners for roundness; one determiner for round- ness and one for wrinkledness; or two for wrinkledness. Those which contain any determiner for roundness will be round, so that only one in every four pairings will be wrinkled. These relations can best be shown by the con- sideration of a simple diagram.
MENDELIAN INHERITANCE "■ _ B o n o n VI vn Fig. 34. Diagram Illustrating Mendelian Inheritance