dentally discovered, for they are unconsciously injected into the answers while the mind of the writer is intent on some other idea. They represent the most fundamental reactions of the child. As representing such conclusions I may mention the various discoveries from studies of children's drawings; the expansion of the social consciousness as shown in remoter ideals chosen by older children; the types of geographical interests revealed in the things remembered from a journey described; the growth of the consciousness of cause and effect, as revealed in the reasons given for various statements; numerous sex and age differences, and the like. It is no exaggeration to say that some of the most important generalizations of child study have been discovered in this incidental way. Most of these things would escape one in the direct observation of the children, because his attention is on other things. But when they are embodied in permanent written statements which may be viewed in the aggregate and with deliberation, they reveal themselves with emphatic distinctness.

5. By no means the least of the merits of the questionnaire method is the excellent approach from a distant to a more intimate relationship with children. which it furnishes to young teachers or to those preparing to teach. I have found the use of the syllabus in attacking problems of child-mind a most excellent discipline, as well as a source of inspiration, to those preparing to teach. I have for several years required my students, after having carefully studied the method and the conclusions of a number of the most important questionnaire studies, to work out one complete study from the formulation of the questions to the drawing of the generalizations; and, in spite of the drudgery involved, they have never failed to grow enthusiastic as the conclusions began to emerge from the tabulated data. Such studies have invariably resulted in a closer and more sympathetic observation of children thru direct contact. Attention has been called to the fact that the questionnaire method simplifies its problems by attacking them from a distance (III, 1). Like long-range guns, it picks off the enemy's leaders at a distance, and thereby insures an easy victory at close range. It is just this feature which makes the method a good introduction to the study of children. By it so many of the native reactions of the child can be studied at long range that the really complex personality is rendered comparatively simple, and offers much simpler problems than it otherwise would, when one comes into direct contact with him.

IV. It would be sheer folly to claim that the method under discussion is the only method, or the best method, for studying children. Like all instruments of science, it has its limitations as well as its excellencies. There are many types of problems to which it cannot be applied. All investigations of physical growth, reaction time, sense discrimination, relation of physical and mental development, correlation of mental traits, fatigue, abnormal conditions, heredity, general and special training, and numerous others, are proper fields for the use of direct observation, laboratory tests, and other methods. On

the other hand, numerous contributions have been made to the psychology of race and sex, thru the use of the questionnaire. There are limitations of other kinds also. It is evident that our method is not applicable to those too young or too illiterate to express themselves easily in writing. Hence all studies of children below the age of seven or eight must be conducted along other lines, except in the case of questions requiring very simple answers. Rare or exceptional experiences or conditions-those not easily verified by comparison are best studied in other ways. All questions seeking only a statement of facts observed, or of personal opinion, may well be investigated. thru a syllabus; but those asking for expert testimony, or involving careful introspection, or susceptible to distortion thru personal bias, must be confined to data gathered, by syllabus or otherwise, from those alone whose ability is beyond question.

But, in spite of criticisms or commendation, it remains true that we must judge any instrument or process by its results. Judged by its results, whether in the high character of many of its users, in the variety of its topics, in the range of its application, in the soundness of its chief generalizations and in the effects of their application to pedagogical practice, or in the inspiration which it has given to scores of earnest workers in the field of genetic psychology, the questionnaire method of child study stands second to none.

LABORATORY TESTS AS A MEANS OF CHILD STUDY

MISS MABEL CLARE WILLIAMS, INSTRUCTOR IN PSYCHOLOGY, STATE UNIVERSITY OF IOWA, IOWA CITY, IA.

Our knowledge of the mind of the child may be advanced by several standard methods. Among them are: the method of simple observation of the individual child, followed by Preyer, Miss Shinn, and others; the statistical method, including anthropometry and the use of syllabi; the method of reminiscence, where the adult reviews the experiences of his childhood; the study of childrens' productions as an expression of the mental processes underlying them; the method of unclassified observations followed by Russell; and the direct experimental method.

Of all of these methods, the last-named is perhaps of the greatest value, not only scientifically, but practically. It may be employed alone, or it may be supplemented by, or used in connection with, some of the other standard methods of child study, the advantages of such combination being reciprocal. It is the purpose of this paper to outline some of the characteristics and advantages of this experimental method, some of the fields of application open to it, and a brief consideration of some of the requirements which the use of this method places upon the laboratory equipment and the experimenter.

A psychological experiment resembles an experiment in any other branch

of science in that it is an observation made under standard or controllable conditions. It differs from experiments in other fields, however, in that, as a rule, the processes experimented upon are not measured directly. Mental processes are conditioned by brain-processes; and by varying the stimulus to neural action, the brain-process is varied in a definite manner. Thus, a psychological experiment is an investigation of the mind thru "stimulation of the nervous system" (Baldwin). While it is true that a mental process cannot be measured, yet psychological measurement can be employed. The objection to the use of psychological measurement because the processes under investigation are too complex, is usually made thru ignorance of the conditions present. The experimental method discovers these conditions and reduces the complex process into its constituent elements.

One advantage of the experimental method is that it secures reliability of results. A record is kept of all the conditions which might influence the result, and they are many; of the exact method of procedure; and of the results. secured. In a carefully planned experiment, what can be known is known and provided for beforehand, and every source of error is eliminated in so far as that is possible.

A second advantage of the experimental method is that it may be repeated by any number of independent investigators. It is possible also to compare results of independent researches.

The experimental method permits of the changing of one condition at a time, in order that its effect and importance may be determined. In this manner relevant factors may be separated from irrelevant factors and the complexity of the problem reduced greatly.

The experimental method is of special value in testing the validity of theories; in fact, it is the only rigid method of proving or disproving an hypothesis. A disputed point in a discussion may be settled once for all by means of a crucial test.

New problems are being discovered constantly by means of the experimental method. In the evolution of an experimental research new factors. continually present themselves, and what seemed at the beginning a single specific problem is shown to be a complex series of problems, all perhaps awaiting solution by the experimental method.

So much for the experimental method in general. What are the possibilities of this method when applied to the study of the mind of the child? There are many fields open to it. Among the most fertile of these are anthropometry, and developmental and educational psychology, in both their normal and abnormal aspects.

Anthropometrical measurements may be of three types-physical, psychophysical, and psychological. Experiments in any one of these directions lead to a more thoro understanding of the individual child.

The study of the physical condition of the child is necessary for a proper estimate of his mental condition. Normal and abnormal states of nutrition,

general nervous tone, balance, etc., should be observed; but more direct attention should be directed toward a determination of the condition of the organs of special sense. The child depends upon these for knowledge of his environment. Frequently defects are present which, if neglected, may mean permanent injury. The statements of parents, teacher, or even of the child himself, as to his physical well-being, do not have the significance of laboratory tests. The reliable means of determining the actual physical status of the child is the experimental method.

Or the tests may be psychophysical in character. In fact, by far the greater amount of psychological research is upon psychophysical problems. Under this head come tests of sensations of all kinds, voluntary motor ability, reaction times, perception of space. An example of a psychophysical test may be cited as a type of many others. The problem is to measure the child's power to discriminate the pitch of tones, this being one factor of his musical ability. The experiment is made with a series of specially tuned tuningforks. These range from 1 to 30 v. d. above a given standard pitch of 435 v. d. With them the smallest pitch difference that the child can detect is determined. A pair of forks is selected, the standard always being one of the. pair, and these are set in vibration and held to his ear at an interval of about three seconds. Each fork is sounded about three seconds. The child is required to judge whether or not the second fork is higher or lower in pitch than the first. The test begins by sounding the standard and 30, the standard and 23, the standard and 17, etc., until a mistake is made. At this point ten trials are made on the same pair of forks, and ten trials on either side, until a place is reached in which, say, 80 per cent. of the trials are correct. This is regarded as the threshold. The value of this test is evident. It affords a definite and reliable measure of the musical ability of the child. It enables the experimenter to advise his parents or teacher whether or not he will be able to profit by musical instruction. When the threshold is low, the study of music may be strongly urged. When the threshold is high, it is often. unjust to require the study of music. The results of this test are not superficial findings; the experiment goes into details and has a distinct bearing upon the theory of hearing.

Finally, observations yielding valuable results may be made upon more strictly mental processes, such as discrimination and choice in complex reaction experiments, the affective processes, and other modes of mental activity.

The great emphasis which anthropometrical measurements are receiving at the present time has aided in establishing the fact of individual differences. Every person differs from every other one in both physical and mental traits. This fact is an illustration of the wide biological law of variation. Variability, the tendency to be unlike, is one of the fundamental properties of living matter. The study of individual variations in mental traits forms a department of anthropological research known as individual psychology, and the experi

mental method is employed in determining the individual differences in mental traits. For instance, laboratory tests reveal the presence of individual types of action, imagery, feeling, etc. From a practical standpoint the individual psychology of the child is more important than that of the adult, for it is frequently necessary to modify his treatment or instruction with reference to his personal traits. It is possible to distinguish those children who vary within normal limits about a definite standard or mode from the more extreme cases of variability, and then, if necessary, to subject the latter to a special course of instruction. The consideration of these more or less abnormal cases is not only important in itself, but frequently the normal is best understood by comparing it with the abnormal.

Another problem in this study of individual mental traits is to discover their interrelations or correlations. That is, we have what is called correlational psychology. For instance, what relation does one mental trait bear to another mental trait, or what is the relation of a mental and a physical trait? Does training in manual dexterity facilitate the study of mathematics? Or does skill in reading Latin avail in meeting practical everyday problems? The presence of correlations is for the most part revealed by the application to the experimental data of mathematical formulæ borrowed from the field of applied mathematics in biology. So far as the work of discovering correlations of mental traits has progressed, the main trend of the results is rather negative than otherwise. Many mental processes which have hitherto been supposed to be functionally related may bear no functional relation to each other. For instance, practice in memorizing one set of facts may not make it any easier to memorize a different set of facts. We are still grossly ignorant of this whole subject of correlations and await with interest the results of experimental research along this line.

Another field, in addition to the general one of anthropometrical investigations, is the application of the experimental method to developmental problems. What characteristic resemblances and differences does experiment reveal in the several developmental periods or stages of growth-the periods of infancy, childhood, boyhood, puberty, adolescence, maturity, and senescence? This problem can be approached in a thousand different ways, and it is useless to specify experiments in detail. Nearly all the anthropological and standard psychological tests lend themselves for the study of developmental changes. This field has scarcely been touched upon as yet, and, as is the case in all new fields, the problems inherent in it are not fully recognized. The study of the development of mind as manifested in the child is obviously important as an end in itself, apart from any reference to problems in pedagogy and adult psychology, but such study serves the triple purpose of furthering our knowledge of the mind of the child, of the adult mind, and aids in solving educational problems. Each stage of development has its own physical and mental characteristics, its crises, its seasons of rapid and of slow growth, its instincts, its predominance now of one trait, now of another.