Slike strani
PDF
ePub

known the cases involving normals are of no value. And this uncertainty affects the classes of simplex individuals, for we do not know how many to add to that class from the normals.

It should be noted here that the second generation is not divided definitely into defectives and normals, as would be expected from a simple Mendelian unit factor. It is true rather that all grades of mentality and many varieties of neurotic conditions are produced. The fact again would suggest the same explanation as that mentioned in the discussion of feeble-mindedness, that several factors, rather than one single unit, are present. One reason, however, for the variability here is that two defects, epilepsy and feeble-mindedness, are involved in almost all of the cases. Davenport and Weeks in combining these defects seem to assume that both are due to the absence of the same determiner; but Holmes points out that this cannot be true, for, if the causes for the two defects were the same, they would tend to appear equally from feeble-minded and epileptic parents. But actually epilepsy is much more likely to appear when one or both parents are epileptic than when they are feeble-minded. The two defects seem to be closely associated because feeble-minded parents produce epileptic as well as feeble-minded offspring; but it seems certain that they represent separate factors in inheritance.

It must be admitted that the value of this table for determining mode of inheritance is greatly weakened by the discrepancy between the number of conceptions and the number of offspring classified, amounting to about 40% of the total offspring. This is due partly to lack of information but chiefly to the large amount of infant mortality. If all the offspring were included, the proportion of the defectives would probably be increased, on the theory that the inferior rather than the superior die in infancy.

Insanity. Insanity is much more difficult to study than either feeble-mindedness or epilepsy for two reasons. First, there are various kinds of insanity. Some kinds are inherited and others are produced by environmental causes, and the latter have to be parated before the true nature of the inheritance can be ascertained. Then, of the inherited types of insanity, there are admittedly various forms; and, while it is possible that all these forms may be inherited in the same manner, it is quite conceivable that certain kinds may be inherited with much greater strength than others. The second difficulty is that insanity, even when inherited, does not show itself early in life. It is essentially a phenomenon of middle life, yet it may appear at almost any age, so that one never knows when the age of incidence is passed. Many members of the families studied may die before insanity would naturally show itself, and this fact makes any results unreliable.

David Heron made a study of hospital cases of insanity in which all forms were included, no attempt being made either to classify according to types of insanity, or to separate inherited from non-inherited cases. His results, given below, show little resemblance to the expected Mendelian ratios, though they do show close resemblance to results obtained by Pearson in his study of tuberculosis, and these are included as of interest for purposes of comparison.

1

TABLE XXI

Expected to have Recorded to have Diathesis
Matings

Diathesis Tuberculosis Insanity
1 Both sane
D(R)xD(R)

21%

21%
2 One Insane
(RR)xD(R)

29

24
3 Both Insane
(RR)x(RR) 100

57

50

25%

50

For the sake of comparison Heron assumes that the normals may be simplex. If they are not simplex, and insanity is a recessive trait, all offspring of cases i and 2 should be normal. However, as some of the cases included are probably noninherited, that fact might account for insanity in the offspring of normals. Probably also a number of cases of neurotics, treated by other authors as simplex, are counted as sane in this study, for the dividing line between sanity and insanity used here was hospital treatment for insanity. A better knowledge of the mental condition of relatives outside of asylums might

1 Eugenics Laboratory Memoirs, No. II.

Type of
Mating

increase the number that should be classed as probable hybrids. On the whole this study does not give material sufficiently detailed really to test Mendel's law, but it does show a high degree of heredity in insanity. The coefficient of correlation between parents and offspring was .57 and the fraternal correlation was between .45 and .55.

In addition to Heron's investigation, Rosanoff and Orr made a study of 206 matings with 1,097 offspring, using Mendel's formula as a guide in their classification. They attempted to exclude all cases of insanity not inherited. The results are given below.

[blocks in formation]
[merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][ocr errors][ocr errors][ocr errors][ocr errors]

The mental condition of the parents was not known in all cases and the authors frankly stated that they classified "b,” and "d," according to the character of the offspring. As these divisions contain the largest number of cases the table, of course, is not very conclusive. The results as classified do not depart widely from the Mendelian expectation, but the ages of the individuals are not given. If many of them are still young the normals are likely to be over-estimated. One interesting

1 Rosanoff and Orr,-A Study of Inheritance in Insanity, Eugenics Record Office, Bulletin No. 5.

Actual

Findings

Expectation

Theoretical

result of the study was that while normal mentality is usually dominant over defective conditions, some kinds of neuropathic conditions are dominant over other kinds. This indicates, according to the authors, that normal mental conditions are dependent upon a group of determiners instead of a single determiner. In a later preliminary report of a study of ten families, Dr. Rosanoff finds evidence in at least four families that insanity is not entirely recessive but tends to appear with varying neuropathic symptoms. This conclusion would be in agreement with the results for feeble-mindedness and for epilepsy.

It is probable that conclusive data cannot be obtained for insanity until records have been kept for more than one generation. Mental conditions of parents of patients are not always obtainable unless these families have been under observation by experts. And it is not always possible to classify individuals as to insanity until late in life. In addition to these complications, it is coming to be recognized that data for the inheritance of insanity is very greatly complicated by racial poisons. It is possible that the difficulties of classification may prove to be too great ever to give very exact results for heredity; but the evidence now is that insanity is strongly heritable and that it does not appear to be uniformly recessive. The material for all forms of mental defectiveness leads to the same conclusion. If their inheritance is Mendelian at all, they represent cases of multiple factors or at least of imperfect dominance.

Supplements to the Theory of Heredity. Multiple factors. The cases of human heredity reviewed here are sufficient to show that human heredity is complicated and does not follow any simple formula. Several theories have been proposed to explain the situation. The Mendelists believe that their analysis furnishes the ground work for all forms of heredity, though allowance has to be made for many variations and combinations. And their position is supported by microscopical analysis of cell divisions and combinations of chromosomes. In other words the microscopical' work is consistent with Mendelian suppositions and with experimental results, though it does not exclude other explanations.

In order to make Mendel's theory more generally applicable it has been suggested that so called unit characters are not stable and indivisible but are complexes, which under the right circumstances may hold together as a unit, but which under other conditions may split up and form new combinations. Many characters at first supposed to be simple units have been broken up by selected crossings, and therefore it is suggested that all characters are multiple. A possible explanation is that a character may become stabilized through long inbreeding and in that case act as a single Mendelian unit, as is frequently the case in plants. But when there is continual intercrossing, as with the human species, characters become less and less coherent and more easily yield new combinations of component factors, thus giving the complex results so often found in human inheritance. According to this theory there is no hard and fast line of distinction between continuous and discontinuous variations. It is a matter not of quality but of degree. At one extreme in the scale is the coherent unit character which follows the simple Mendelian formula, and from this extreme the factors involved may increase in number until they become so numerous as to form a blend.

This suggestion of complexity has much to commend it as an explanation of the behavior of hereditary characters. It has been proved already that certain characters are multiple, composed of three or four factors; and therefore it is reasonable to assume that others also are multiple, and may even be made up of many factors. It would be difficult to disprove the claim that blended inheritance is due to the combination of many factors. It has been shown that if six factors are present the original character will reappear only once in 4,096 cases. And in that case, if the character did reappear, it would be impossible to say whether it was a mere variation or whether it was the original character reappearing through segregation. In an extreme case of this kind, however, there seems to be no gain in trying to mold it to Mendel's formula. For practical purposes several multiple factors give a different form of inheritance. A draftsman is not enlightened by being informed

1 See G. N. Collins, Nature of Mendelian Units, Journal of Heredity, 5: 425.

« PrejšnjaNaprej »