Search

Data on psychiatric disorders were collected on first-degree relatives and spouses of 70 patients with primary affective disorder and 75 patients admitted to a general hospital without psychiatric disorder, from the Jewish population of Jerusalem, Israel. Bipolar patients tended to have more bipolar relatives than unipolar patients, but unipolar patients had an observable morbid risk for bipolar disorder in their relatives. Ethnicity and age of onset did not appear to be related to transmissible factors in the prevalence of affective disorders in relatives, although age of onset was associated with morbid risk. Sex-linkage or genetic liability to affective disorder related to sex did not appear to be present. A spectrum of related disorders for this population was defined by those disorders which were more prevalent in relatives of affective disorder patients than in relatives of controls. Assortative mating was not found, but a modest degree of inbreeding appeared to be present. For purposes of testing liability-threshold models for genetic factors in the transmission of mood disorders, only the data on polarity seems suitable.

SummaryHysteria (Briquet's Syndrome) and sociopathy cluster in the same families instead of segregating as independent traits. Assortative mating between hysterics and sociopaths increases the observed similarity between relatives, but the familial association between sociopathy and hysteria remains after taking assortative mating into account. The Multifactorial Model of Disease Transmission with three thresholds related to severity and sex accounts for population and family data about sociopathic men, sociopathic women, and women with hysteria. The data were obtained for 227 first-degree relatives and for 800 subjects in the general population.Depending on the sex of the individual and its severity, the same aetiological process may lead to different, sometimes overlapping, clinical pictures. Specifically, analysis indicates that hysteria in women is a more prevalent and less deviant manifestation of the same process that causes sociopathy in women.

A 5 item sexual permissiveness scale devised by Reiss, and 4 personality inventory scales purporting to measure psychoticism (P), neuroticism (N), extraversion (E) and propensity to dissimulate (L for lie scale) were administered to 427 male and 436 female subjects aged between 20 and 50 years. It was found that men were more permissive than women on all 5 items; that items are intercorrelated in a pattern which could be predicted from the Guttman‐type properties of the original scale; that high P scorers were more permissive than low P scorers; that high L scorers were less permissive than low L scorers; and that extraverts were slightly more permissive than introverts. All the findings relating to personality and permissiveness were in good agreement with findings from previous studies of the relation between personality and sexual attitudes. Assortative mating was studied for 241 couples who formed part of the original sample. Correlations for permissiveness were all positive, and reached sign...

The heritabilitý of performance in thoroughbreds was studied using the Timeform Ratings for horses which raced during 1970. Timeform Ratings are based on the relative performance of all horses racing in a given year. 794 Individual three year old records, representing 96 sires which had five or more progeny, were used. The mean ratings for sires, dams and progeny were 128.5, 93.6 and 81.4 respectively; among the progeny the ratings for colts, fillies and geldings were 88.3, 79.1 and 77.1 respectively. Calculations made suggest that while stallion selection is effective in selecting sires with high performance ratings it does not appear that the selection of dams is very closely related to performance. There was no evidence of assortative mating. Paternal half sib and regression of offspring on sire analyses of the data yielded heritability estimates of 0.35 ± 0.11 and 0.74 ± 0.20 respectively. Further regression analyses on a reduced set of data (553 progeny records) of offspring on sire, offspring on dam and offspring on mid parent value gave heritabilities 0.56 ± 0.20, 0.36 ± 0.10 and 0.34 ± 0.08respectively. The possible reason for the substantially higher heritability estimates from the regressions of offspring on sire are discussed. It is concluded that the best measure of heritability of performance is 35%.

SummaryA group of 46 bipolar probands and their first degree relatives were studied. A high rate of affective disorder (19.6 per cent) was found, including both unipolar (13.2 per cent) and bipolar (6.4 per cent) types, with females predominating (3: 1). The presence of four father-son pairs suffering from affective disorder made the hypothesis of X-linked dominance untenable. Results compatible with polygenic inheritance were found, using both Slater's and Falconer's methods. There was no evidence for assortative mating or for increased total number of females (both well and ill) among first degree relatives. The probands and affectively ill first degree relatives who have died show an alarmingly high rate of suicide (46 per cent). Other forms of mental disorder, including alcoholism, were no more common than in the rest of the community.

SynopsisA two-generation study of 46 families of convicted women felons showed that the daughters of sociopathic fathers had a significantly higher prevalence of hysteria than did the daughters of other fathers. The differences were significant both for daughters with hysteria plus sociopathy and for daughters with hysteria without sociopathy. The association was independent of assortative mating between sociopathic men and women with hysteria or sociopathy.

The assortative mating which has been found in the dimorphic Lesser Snow Goose, Anser caerulescens caerulescens, could be explained if birds chose mates according to the colour of their parents and/or sibs. Three distinct captive flocks were tested for colour preferences in terms of: 1) approach response, 2) association preferences, and 3) mate selection. 1) Approach responses. Young birds placed in a choice situation had a significant preference for birds of the parental colour. Sib colour when different from parental colour appeared to modify their choice. If parents were removed during adolescence the early colour preferences could be altered, the most recent associations determining the preferences. The pattern of response did not change in tests carried out at the age of 3 months and repeated at 11 months. The presence or absence of auditory cues did not alter the pattern of response. No differences were detected in the responses of gosling to maternal versus paternal colour. 2) Association preferences. In an open field situation, birds usually associate with their peer group at both one and two years of age. The degree of association with the peer group is less at two years than at one year. When birds associate with non-peer group birds they show a distinct tendency to associate with birds which are the same colour as their peer group. 3) Mate selection. In the flock which was raised as a single large group, with virtually no parental contact, pair formation did not depart from randomness in terms of colour, suggesting that non-random mate selection in Snow Geese is not an inherent property but is a function of their prepairing experience. In the flock which consisted of families where the foster parents were of one colour and the sibs of a different colour, and in which the foster parents were removed after one year, the pairing at two and three years of age was non-random; pair formation reflected preference based on sib/self colouring. It is concluded that familial plumage colour does influence colour preferences in terms of approach response, association preferences and mate selection. However, continuous association with birds of parental plumage colour is a prerequisite for this colour to influence mate selection. If the parent is removed (as happens in the wild) the colour preference may be altered but it is more likely that the preference will be maintained through association with birds of familial plumage colour. Thus, directly or indirectly, parental colour will influence mate selection.

The effectiveness of the assortative mating of selected individuals in increasing selection response was tested, using abdominal chaeta score inDrosophila melanogaster. Three paired comparisons were made. In two sets of lines with 10 matings per line, individual score was used for selection and as the basis for the assortative mating. In the third set with 20 matings per line an index of individual and family score, designed to maximize rate of response, was used.The intensity of selection was one in ten in all lines. Flies were raised in vials and individual pedigrees were kept.In all comparisons, assortative mating gave a greater selection response, this being partly due to a greater realized heritability and partly to a greater selection differential. The effect of the assortative mating was largest in the index selected lines. With random mating, the effectiveness of the index selection itself when compared to individual selection was in accordance with theory.In two comparisons, assortative mating increased the rate of inbreeding. The highest rate of inbreeding was observed with index selection and assortative mating, even though there were here twice as many matings as in the individually selected lines.In the individual selection lines, the effective population size was 7·4 pairs of parents, compared to the actual value of 10 and in the index lines 7·0 compared to 20. In the former, only one-half of the matings in the initial generations made any permanent contributions to the line and in the index lines only one-third. Within generations and lines, there was a significant positive correlation between the mean score of a family and its inbreeding coefficient.It is suggested that assortative mating is a method of increasing selection response in some situations. Its particular characteristic is that it becomes more powerful when the heritability is high whereas all of the other environmental aids to individual selection are more effective when the heritability is low.

A comparison was made of 104 married couples rated by friends as well adjusted and 70 couples rated by friends as proorly adjusted. Little or no relationship was found between marital adjustment and factors such as family authority patterns of childhood, sex favoritism on the part of parents, sibling position, nativity of parents, and occupation of father. The evidence suggested relationships between material adjustment and factors such as unequal intimacy of women with their parents, excess or deficiency of friendships with the opposite sex on the part of men, similarity of mates as regard to schooling, and assortative mating with respect to age.

The hypothesis put forward for discussion is that, among normal persons, differences in “intelligence” (i.e., innate general cognitive ability) are determined by a large number of genes, segregating in accordance with Mendelian principles and each producing effects that are small, similar, and cumulative. From these assumptions, with a due allowance for dominance and assortative mating, formulae are derived for the correlations to be expected between siblings, parents and offspring, and remoter relatives. Data from school surveys yield coefficients which are in almost complete agreement with the theoretical figures thus deduced. An attempt is also made to analyse the variance of the assessments obtained, and to estimate the contributions of genetic factors and non-genetic factors respectively. Appreciable differences are found between the results obtained from raw measurements based directly on intelligence tests and those obtained from revised or adjusted assessments based on all the available evidence. With the former nearly 25 per cent. of the variance is apparently due to non-genetic factors; with the latter barely half that amount.

Summary In this paper, various systems of non-random mating are considered for both monoecious and dioecious diploid populations. The effect of positive assortative mating on the rate of approach to homozygosity is investigated, at a single chromosome locus in the absence of mutation. The difficulty of setting up an amenable theoretical model for negative assortative mating is discussed. It is found that for a monoecious population, positive assortative mating increases the rate by a factor between 1 and 2 compared with random mating. For a dioecious population, assortative mating may either increase or decrease the rate of approach to homozygosity, depending on the mating system and the divergence of the sex numbers from equality. A population whose birth-death events occur in succession (a model introduced by Moran), is known to approach homozygosity at twice the rate of a population with non-overlapping generations, and this result is shown to hold under positive assortative mating. The probability of fixation of an allele is obtained, and depends on the initial gene frequencies.

1. A test is described for the development of sexual isolation between a wild and a derived population of D. melanogaster adapted to a new diet, containing EDTA. Other experiments had shown that adaptation to the new diet involved genetic changes in all chromosomes. Also fitness was reversed on the alternative diets under crowded competitive conditions.2. In three replicated trials flies from each population were used to establish paired cage populations, supplied with the medium to which each was adapted, and the pairs of cages were joined to allow restricted immigration between them. The experiment was run for about twenty-five generations.3. After fifteen and twenty-five generations, flies were collected from each cage to provide eggs which were cultured on the alternative diets to determine how far the members of pairs of populations differed from each other and from the foundation population. There were striking differences between the sub-populations and the parent populations, attributable to immigration between the former. Judged by the differences in performance between the sub-populations, genetic differences persisted but these were minor compared with the differences between the parent populations.4. Tests of preferential mating on the part of flies from paired sub-populations were entirely negative.5. Fourteen generations of selection for positive assortative mating failed to provide evidence of sexual isolation between the two basic populations, adapted to different diets.6. From these and other experiments it is inferred that sympatric divergence is improbable in a species like D. melanogaster.

A multi-locus model for complete positive assortative mating is discussed. For a two-locus model, if the gene frequencies for the two loci are different, as they are likely to be, it is shown that in equilibrium the population is not composed of only two homozygous types, as is usually thought. The limiting distribution will have three homozygous genotypes depending upon the initial gene frequencies. If there are m-loci such that gene frequencies at all loci are different, there will be (m+1) such homozygous genotypes present in the equilibrium population, one in each phenotypic group.

The result of a method of breeding involving the formation of pairs according to some system of assortment, rather than upon a purely random basis, is readily seen to be of great importance, not only for the foundation of racial diversities, but also for the conservation of genetic stability (Romanes, I906; Pearl, I907a; Wright, I921). Actual instances, however, of assortative mating, occurring under natural circumstances, have been little studied. And the matter is of interest beyond its strictly genetic bearings; for it is probable that through sexual coupling selective with respect to somatic features, there may in different species be achieved various automatic, adaptive, consequences non-genetic in character, but nevertheless significant racially (Crozier, I918). In this respect selective pairing of protozoans and of metazoans may differ greatly as to their implications. The selective combination of gametes (cf. Jones, I920) is a question quite distinct from that of assortment of individuals, and the two should not be confused. Among metazoans relatively few cases of normal selective mating have been recognized, although Jennings (I920, p. 193) remarks that the propensity for like to mate with like is probably in some degree quite a general phenomenon. In Paramecium and allied ciliates, which have been most extensively investigated, there is a well-defined tendency toward conjugation between individuals resembling one another in size and in fission-rate, and to this extent at least structurally and physiologically akin (Pearl, Io07a; Jenning, I9II, I920; Watters, I912). This is in large part due to the fact that the mutual fitting of two individuals, requisite for conjugation, is mechanically possible only when these individuals Vol. XL V. No. 2.

Polymorphism in adult color pattern is found in many Lepidoptera species, particularly in mimicry complexes where mimetic coloration confers protection from predators or in cryptic forms where concealing coloration serves the same purpose (Ford, 1957, 1964). Other factors promoting discontinuous diversity, usually leading to a balanced polymorphism, are the maintenance of a normal sex ratio, heterozygous advantage, relative abundance of morphs in relation to learning behavior of vertebrate predators, and differential mating preferences (Ford, 1965). Negative assortative mating is one of several systems that could maintain in addition a high level of heterozygosity within a population without the heterozygote necessarily being superior in fitness to either homozygote. However, preference for mating with unlike partners has only rarely been encountered in nature. Disassortative mating is reported between morphs of the adult tiger moth, Panaxia dominula (Sheppard, 1952, 1953), between certain mutant lines of Drosophila melanogaster (Rendel, 1951), and in wild populations of the polymorphic whitethroated sparrow, Zonotrichia albicollis (Lowther, 1961; Thorneycroft, 1966). Frequency-dependent disassortative mating has also been discovered in laboratory colonies of six Drosophila species (see Ehrman, 1969 for review). All of these reported situations involve a positive contribution of this type of mate selection to the maintenance of a balanced polymorphism in the population. The present paper reports a situation in a polymorphic species where assortative and disassortative mating both work against the maintenance of a balanced (stable) polymorphism. The neotropical nymphalid butterfly Anartia fatima Fabricius shows a striking color dimorphism in the adult. This study presents data that indicate the polymorphism is balanced, and offers evidence that both negative and positive assortative mating contribute significantly in a negative way against the maintenance of the two phenotypes in natural populations in Costa Rica. Selective forces are outlined that may contribute to counterbalancing this negative effect of mate selection. These two sets of opposing selective pressures influence the frequencies of the forms at different altitudes, geographic localities, and seasons in this neotropical area.

In order to substantiate evidence and explore further details of frequency dependency between third chromosome arrangements in Drosophila pseudoobscura populations, multiple-choice mating experiments involving karyotypes of Arrowhead (AR) and Pikes Peak (PP) gene arrangements were designed to measure relative mating propensity with frequencies of karyotypes varied in one sex or in both sexes. Ten strains each of AR and PP crossed inter se as in previous experiments contrast in being fastest (75% mated) and slowest (20%) mated, respectively, in homokaryotype no-choice matings of 30 min. duration. Twenty pairs of strain-cross F1 flies, 6 days old, were introduced unetherized into small Plexiglas chambers and observed for 30 min. Flies had been marked to distinguish karyotypes by puncturing a wing on the day before the mating tests. Two contrasting types of both sexes were tested each time to measure relative propensities, total mating frequency, and any assortative mating. Ratios of karyotypes and condition...

An exact test is given for randomness of mating, and it is shown that in a population ofPanaxia dominula there is no evidence that mating was not random in any of 15 years.

Studies of the salivary gland chromosomes of many strains of Drosophila pallidipennis, with various cross-tests, has revealed that of the five rodlike chromosomes only one autosome (D) has undergone important inversions and the X chromosome exhibits special behavior (such as little variation of the strong parts and weaker pairing), and that the group is a complex rather than a single species. There are two separate Mendelian populations, the hybrids of which display more or less strong degrees of breakdown in gene coadaptedness (the minimum being complete F1 male sterility). Whether they are full species (through complete assortative mating when mixed) remains to be verified. One of these forms (centralis), occupying Central America and Colombia, is monomorphic for a CE gene arrangement of the D chromosome. In the other form (true pallidipennis) the CE gene arrangement occurs only in Peruvian populations and is completely lacking in Bolivia, Venezuela and Brazil, where local populations are mono- and polymorphic for rearrangements, each of which can only result from an inversion in the CE chromosome. It is suggested that selection is in the process of eliminating the CE arrangement in pallidipennis. This interesting situation, combined with the very high quality of breeding and chromosomal cytology of the flies, should prove valuable for studies in developmental and population genetics.

Disruptive selection with negative assortative mating as well as with opportunity for random mating was carried out for sternopleural bristles in Drosophila melanogaster. In the latter system, the divergence in score between the progeny of high and low females altered little after the fifth generation and was, at the end of 14 generations, much less than expected theoretically. An explanation is suggested for this result, which is paralleled by those of several other workers, in terms of the mating behavior of the females.

Under the above title a paper has recently appeared by Mr. A. D. Darbishire in the ‘Roy. Soc. Proc.,’ vol. 81, B, p. 61 et seq ., giving further experimental evidence with regard to the inheritance of certain characters in peas. The paper is an interesting one, but the method adopted is not, I venture to think, capable of answering the problem which the author set himself. It has been supposed by some Mendelians that the theory of inheritance summed up in the “law of ancestral heredity” was in some way invalidated by investigations such as Mr. Darbishire’s and that opinion consciously or unconsciously seems to be expressed in the paper just referred to. The law of ancestral heredity is embraced in the following statements:- (i) In a population breeding without assortative mating the regression line for offspring on any ancestor is linear. (ii) The correlations between offspring and the successive grades of ancestry form a progression diminishing geometrically as we ascend to distant grades; and (iii) The general relation of an individual to his ancestry can be closely expressed by the multiple correlation formula.

Two hundred and sixty-three Indians aged six and above were studied in Oklahoma for blood types, hemoglobin types, and physical traits; 53% were Seminoles and the remainder were admixed with Creek or other populations. The results indicate that they resemble the Florida Seminoles in most of their serologic traits. In physical traits they are comparable to the group studied by Krogman two decades ago. Compared with the Florida Seminoles, the Oklahoma sample are slightly taller and heavier, significantly lighter in skin color, and have a lower incidence of sickle cell gene. By both serology and morphology the Oklahoma group are most similar to Florida Seminoles, slightly less similar to other Indian groups, and still less to White and Negro populations. The Oklahoma sample of women showed a non-significantly greater fertility than those of Florida. Some indication of positive assortative mating for skin color was found in both groups.

The concept of intelligence, and the attempt to measure intelligence by standardized tests, have of late furnished a target for vigorous attack. The objections urged are partly practical and partly theoretical. Yet few of the critics show a clear or correct understanding of what the term really designates or of the reasons that have led to its introduction. Two misconceptions have become widely current.

It is important to know whether within a given group of animals or plants mating is or is not at random. Statistical tests for randomness are given. They are applied to the data of Ride where there is no dominance, and to those of Philip in which the frequency of heterozygosis amongst dominants was deduced from test matings.

The variation shown by the Australian Commercial population of P. tuberosa in respect of date of ear emergence, seed weight and seedling weight has been analysed, and the interrelationships among the variables characterized. All three characters show appreciable additive genetic variation, in the sense that approximately seven generations of artificial selection would be sufficient to push the population mean for each trait beyond the range shown by introduced ecotypes.Apart from the positive association between seed weight and seedling weight due to maternal influence, the three variables are to a large extent genetically independent in this locally adapted interbreeding population. It is therefore probable that the negative ecotypic correlations between seedling growth rate and date of ear emergence, and between seed weight and date of ear emergence, would rapidly be dispelled under random mating in a synthetic population.Under open-pollination, date of ear emergence has been shown to be subject to phenotypic assortative mating of degree ρ = 0·78, approximately 53% of the variation being additive genetic. Variation in seed weight within the strain is extensive, with a heritability of 0·79. Neither character shows evidence of important genotype × years interaction.Variation in seedling weight involves an appreciable genotype × environment interaction component, and has a heritability of only 0·17 which includes variation due to genetically determined maternal effects. The correlation between the seed weight of an ovule parent and the ‘ true’ mean seedling weight of the derived maternal half-sib group is of the order of 0·57. It has been estimated that the correlated response per generation in seedling weight, due to selection for seed weight, is 0·54 times that expected from direct selection.

When the scores obtained with the WISC battery of tests from children attending a psychiatric clinic were compared with those obtained from normal children, they showed a greater dispersion. Attempts to explain the discrepancy in terms of selection, non-normality, or emotional and environmental factors, proved unconvincing. An application of Bartlett's formulae for the sampling theory of ability suggested that the difference might be due to genetic factors arising from pronounced assortative mating.

Females from the inbred strains BALB/Ibg and DBA/Ibg and females from the reciprocal crosses between these strains, which were fostered by BALB/Ibg and DBA/Ibg males, were tested for mating preference between BALB/Ibg and DBA/Ibg males. The tests included the use of both the standard and the tubetechnique methods. In these strains, it was found that the main determinant of female mating preference was the genotype of the foster father. Females reared by DBA/Ibg males had a tendency for negative mating preference, whereas females reared by BALB/Ibg did not. The genotype of the mothers had some effect on the degree of the preference. Females reared by BALB/Ibg mothers had somewhat lower preference for BDA/Ibg males than did daughters' of BDA/Ibg females. The effect of the female's own genotype on mating preference was found to be only minor in a comparison done between these strains and their F 1 crosses which had been raised by foster parents of the same genotype.

A modification is presented of Reeve’s (1953) method of estimating heritability from progeny tests when the parents are selected and mated assortatively.

Among human populations, the selection of a spouse is conditioned by many factors involving not only natural but cultural determinants. The social regulation of marriage, common to all human societies, narrows the range of potential spouses in a way which interferes with random mating. At the genotypic level, this nonrandomness is referred to as inbreeding (Li, 1961; Wallace, 1968) and is a function of the frequency of consanguineous marriages. At the phenotypic level, nonrandom mating is referred to as assortative mating (Falconer, 1961; Wallace, 1968) and is usually ascribed to the exercise of "choice" in spouse selection, though genetic relationships may be a factor as well (Crow and Felsenstein, 1968).

Summary. Expressions (2·1), (3·1), (4·1), (5·1) are found for the composition of Mendelian populations subjected to partial self-fertilisation, inbreeding, assortative mating, or selective fertilisation, and equations (2·2), (3·2), (4·2), (5·2) derived for the effect of selection on such populations. The effect of selection is greatly increased by inbreeding and self-fertilisation.

Females of two inbred strains ofMus musculus domesticus, C57BL/Ibg and DBA/Ibg, were allowed to choose between two males, one of each strain, who were restrained within their cages. Females of both strains preferred to spend more time with, and to mate with, the males of the opposite strain rather than with males of their own strain. No major strain difference for the degree of preference was found. Preference was also unchanged between trials. There is a general lack of correlation between trials which is explained in terms of the nature of the variance in inbred strains. Female mating preference may have an evolutionary significance in reducing inbreeding.

sorting apparatus, dropping out the less gifted at an early stage and carrying along the more gifted to higher levels. In the third place, the failure of the upper classes to add their full quota to succeeding generations has served to create vacancies to be filled by mobility from lower levels. As Fisher says, intelligence has in our society been associated with social success but with biological failure. The total effect of these various processes would seem to be dysgenic. Democratic individualism destroyed a society of castes and privileged classes only to set up conditions leading to biological stratification. The best blood has moved upward at rates which may well have left the lower strata genetically impoverished. Assortative mating renders each level more and more homogeneous in biological constitution. The differential birth rate thus appears to be definitely selective. Even if this is a substantially correct picture of the present situation, we should not fail to note that we have little accurate information as to the rapidity of this selective action. Moreover, the recent reduction of the birth rate in all classes serves to reduce whatever selection there may be, while its f-urther reduction among the less successful elements of society might well eliminate selection altogether.

(1968). The effect of assortative mating on the genetic composition of a population. Eugenics Quarterly: Vol. 15, No. 2, pp. 85-97.

(1968). Assortative mating by educational attainment in relation to fertility. Eugenics Quarterly: Vol. 15, No. 2, pp. 98-112.

1. Approximately 25000 mice have been produced over twelve generations of selection in nine lines. The experimental design involved a 3 × 3 factorial arrangement of direction of selection with mating system. The primary character measured was 6-week body weight.2. Consistently high phenotypic correlations between mates have been achieved, positive in the assortative lines, negative in the disassortative lines. Correlations were low and inconsistent in direction in the random-bred lines. These correlations have had very little, if any, effect in redistributing the genetic variance as estimated from the variance component analysis; the expected higher variances in the assortative lines and expected lower variances in the disassortative lines have not appeared, thus leaving heritability unaffected.3. Selection differentials likewise show no consistent advantage for the assortative lines, so that the progress from selection has been virtually identical in all three mating systems in each direction.4. Assortment of mates, either positively or negatively, for characters of even moderate heritability appears to have little influence on the outcome of selection. On the other hand, selection has been singularly effective in modifying the mean 6-week weight, with progress markedly greater in the downward direction; indeed it appears that the lower limit, may already have been approximately attained.5. Environmental effects operating in the various generations have affected all lines in remarkably consistent fashion.