because he cannot be sure that in breeding for one group of good traits he will not introduce an equal number of harmful traits, or that in attempting to breed out certain bad traits he will not lose others that are essential, and furthermore, what may be particularly desirable in this generation might prove to be harmful ten generations hence. Practically the only safe rule for us to follow at present is to breed from average good stocks and to avoid anything that is positively bad. If we had to make a definite set of rules for breeding they would be somewhat as follows:

Fifth: With certain limitations as to type and to time.

These rules apply, however, to the breeding of large groups for racial development in the mass rather than to small groups or communities. There exist within every race certain barriers to the natural or easy application of these rules, which are geographical and social. The effects. of the former we see in valley populations hindered from easy access to the outer world by abrupt mountain ranges, and in the population on some parts of the seacoast, especially upon islands. The social barriers are equally evident in their action, being determined by the limitation in marriage selection imposed upon the members of the various social classes. As an example of the latter we might cite the tendency of institutional inhabitants or workers to intermarry as, for instance, among the deaf and dumb who are thrown into each other's society by the nature of their defect. The stable non-migratory population tends to multiply its negative traits by inbreeding,—for instance, if we may assume that weakmindedness is a negative trait, of which it has certain characteristics, if not all, we know that its occurrence increases by inbreeding, whereas it may diminish by outbreeding. On the other hand positive traits, such as hemophilia, tend to diminish in a stable population, but increase by migration and outbreeding

through the development of new foci. Examples of the heredity of feeble-mindedness or of anti-social mental traits are found in the histories of the Jukes family and of the Ishmaelites in Indiana, and of the inheritance of good traits in the Edwards family with which you are all familiar. Those who maintain that environment or nurture plays an important rôle in the development of the individual would argue that had the members of the Jukes family received the same training both at home and abroad with the same good nourishment that the members of the Edwards family had, they would not have continued to be as actively anti-social as they were. This position of the euthenists would seem, however, to be ill taken, because there are no experiments or statistics that tend to substantiate it, whereas there are many that contradict it. The study of the Kalikak family in New Jersey should serve as an argument against this position.

Note: The writer does not pretend to any originality in this paper or to any special knowledge of the subject, other than what he has been able to obtain from spare-time reading, but he has tried to give a brief historical review and a summary of recent opinions on Heredity in order to provide a ready source of reference for the members of the Society, and to do this he has drawn largely from Mendel's Principles of Heredity by Bateson of the University of Cambridge, Parenthood and Race Culture by Saleeby, the writings of Karl Pearson and his colleagues of the Francis Galton Laboratory of the University of London, and Heredity by Watson of the University of Edinburgh, of which latter excellent book much of this paper is an abstract. No. 1. Heredity. J. A. S. Watson, Edinburgh.

DISCUSSION.

DR. C. F. HAVILAND (Middletown): Dr. Waterman's paper clearly shows the complexity of the subject of heredity. When we realize that in the tenth generation each individual has 1,024 tenth grandparents, it is evident every person is made up of innumerable convergent elements and that there is a constant tendency to an average mean. But we know that in every generation there are individuals who depart widely from the mean, either advantageously or disadvantageously. And as Dr. Waterman has pointed out, every human attribute, whether normal or abnormal, is

in a sense inborn, in that potentiality must exist before actuality. We have organism, environment and potentiality either for adaptation of organism to environment or lack of such potentiality, in which case there is no adaptation and the organism perishes. The equation of course holds good for the human organism but here the adaptation is accomplished through a highly differentiated nervous system, containing infinitely greater potentialities than any other living organism. The human mind in seeking ever better adaptations, endeavors to understand the laws of heredity, but with the theory of psycho-physical parallelism, is merely making manifest a potentiality of human organism.

It is evident that in any specific adaptation the final result is the product of intrinsic and extrinsic factors, and while there is a limit to possible variations there is, nevertheless, a wide range of possibilities as regards the relative potency of each type of factor. But it is idle to talk of the relative importance of heredity and environment. Both are important, and as one supplements the other neither can be practically considered, except in its relation to the other. As Professor Cattell of Columbia once said, "Heredity is what a man may be,-environment is what he is." That is, the human unit represents ancestral possibilities, as revealed and modified by environmental conditions.

That hereditary endowment has something to do with disease has long been recognized. The "diathesis" and "predisposition" of our medical forbears indicates their recognition of the fact, while the knowledge that heredity taint is important in the production of nervous and mental disease is commonplace.

Early theories of heredity give little help in suggesting practical means of control of evil inheritance. It remained for Mendel's theories, so well set forth in Dr. Waterman's paper, to offer an opportunity for practical work. If such theories be correct inheritance is merely a matter of precise selection and reconstruction of parental qualities. And so far as control of parental qualities can be exercised, so far can inheritance be directed. Were complete analyses of unit characters possible, inheritance could be predicted with mathematical certainty, assuming the absence of adventitious extrinsic factors.

There seems to be no good reason for the somewhat acrimonious discussions which have from time to time arisen between adherents of Galton's biometric theory and adherents of Mendel's mosaic theory. One theory attacks the problem in mass and the other in the individual. They offer no essential contradictions, and the body of facts collected by each method supplements the other.

As yet relatively little is known as regards unit characters in man. The very complexity of the matter will render progress slow. But sufficient facts have been determined to indicate that congenital cataracts, certain digital malformations and probably eye color are transmitted according to the Mendelian law. Furthermore a former colleague, Dr.

Rosanoff, has published results of an investigation he made, which seem to show that hereditary mental disease and allied neuropathic conditions are transmitted according to Mendel's theory. One of the first facts that appeared in his study was that different forms of hereditary mental disease did not constitute independent hereditary characters, but were closely related to imbecility, hysteria and so-called psychopathic states. In other words, he failed to find that clinical distinctions, so far as the matter of nervous heredity was concerned, were of essential importance. What he did find were manifestations of neuropathic makeup in closely related persons, expressed in such different forms as mental defect, epilepsy, hysteria, eccentric personalities, dementia praecox, manic-depressive insanity, etc. But a neuropathic makeup was not held as the basis of all mental and nervous cases studied. Cases of paresis and alcoholic conditions were excluded as probably being predominatingly exogenous in origin, while Huntington's chorea was found to be plainly an independent Mendelian character. But cases of paresis, alcoholism and other types of mental and nervous disabilities of exogenous origin must be further studied with regard to the rôle played by a possible Mendelian "determiner," for after all but a relatively small percentage of syphilitics become paretics, and relatively few victims of chemical poisoning and bacterial invasion show mental symptoms.

Rosanoff found a number of instances of neuropathic children born of normal parents, but not a single case of a normal child born of parents both of whom were neuropathic. This neuropathic makeup can be regarded as recessive to normal. Formulating Mendelian expectations accordingly, we have:

I. Both parents being neuropathic, all children will be neuropathic. 2. One parent being normal, but with neuropathic taint from one parent, and the other parent being neuropathic, half the children will be neuropathic and half will be normal, but capable of transmitting the neuropathic taint to their progeny.

3. One parent being normal and of pure normal ancestry, and the other parent being neuropathic, all children will be normal, but capable of transmitting the neuropathic makeup to their progeny.

4. Both parents being normal, but each with neuropathic taint from one parent, one-fourth of the children will be normal and not capable of transmitting neuropathic taint to their progeny; one-half will be normal, but capable of transmitting neuropathic taint and the remaining onefourth of the children will be neuropathic.

5. Both parents being normal, one of pure normal ancestry and the other with neuropathic taint from one parent, all children will be normal, but one-half will be capable and one-half incapable of transmitting neuropathic taint.

6. Both parents normal and of pure normal ancestry, all children will be normal and incapable of transmitting neuropathic taint.

As a matter of fact the study of the pedigrees of several hundred cases showed actual findings in extremely close relation with theoretical expectations, especially when it is considered that many children died in infancy, before nervous inheritance is manifested, numerous families have but a single child, and there was a margin of error due to incomplete pedigrees.

Thus neuropathic taint apparently follows the same law of transmission as the dwarf peas in Mendel's early experiment.

It is true we know too little of unit characters to make broad generalizations in reference to human mating. Unit characters may not only influence each other, but there may be compensatory associations of unit characters. But demonstrated facts do show the unwisdom of a mating between two neuropathic or defective individuals. Yet during the past year three instances have come to my attention, where such action was promoted, in one case by a selectman, and in the other two cases by a clergyman. The kindly motive was to legitimatize illegitimate offspring, the social consequences, of course, being ignored.

Regardless of theory, common experience indicates that evil extremes in mating should be avoided, and our sterilization laws are directed to such end. But while theoretically desirable, it may be questioned if much dependence can be placed upon their operation in the face of present unconcerned public opinion and religious opposition. Furthermore, sterilized individuals remaining in the community are frequently active disseminators of venereal disease, and constitute great moral hazards. Were persons for whom sterilization is proposed, permanently segregated in institutions, society and posterity would be equally served. Great as the financial cost would be, it would be no greater than the present cost of allowing such persons to remain in society, while future generations would reap the benefit through the gradual elimination of at least the grosser types of defective humanity.

DR. WATERMAN: I am grateful to you all for your attention and to those who have discussed the paper. I concluded my paper with some warnings and some references to the biological foundations of the theory of heredity, which I omitted in reading the paper, but which will be available when published.