BIOSOCIAL CORRELATES OF STATURE IN A BRITISH NATIONAL COHORT

Analyses of height variation using the 1970 UK national cohort study (12,508 children at age 10 and 5470 at age 16) found clear evidence that children of higher socioeconomic status (as measured by social class, crowding, tenure, type of accommodation, income and receipt of government financial assistance) were on average taller than children of lower socioeconomic status but there was little or no difference in average stature between children living in urban or rural areas. Significant differences in height remained for most of the variables after removing the effects of father’s social class suggesting that reliance on social class per se to explain height variation is inadvisable.

The genetic structure of English villages: surname diversity changes between 1976 and 1997

OBJECTIVE

To examine the change in the coefficient of relationship by isonymy (Ri) over a 21-year period.

RESEARCH DESIGN

Ri was calculated from the surnames of males in the registers of electors for 15 villages near Oxford in 1976 and again in 1997. The within-residence component reflected the number of adult men of the same surname resident at the same address while the between-resident component of Ri was calculated by counting only one name of any surname at each address listed in the registers.

RESULTS

Ri tended to decrease during the 21-year-period in 14 of the 15 villages and the between-resident component of Ri also tended to decrease in 14 of the villages. The within-residence component decreased in a statistically non-significant proportion (11 of the 15 villages).

CONCLUSIONS

This study, in keeping with some but not all other surveys, showed that there was a tendency for random isonymy to decline in the 20th century. The reduction in the between-residence component can be accounted for by migration into the villages.

The hierarchical genetic structure of an urban town, Kidlington, Oxfordshire, examined by the coefficient of relationship by isonymy

The surnames of the 3443 males registered to vote in Kidlington in 1977 yield a Coefficient of Relationship by Isonymy of 0.000564 {Ri = sigma(n(n-1))/2 N(N-1), in which n = the number of men of each surname and N = sigma n}. Those of the four wards separately average 0.000722. However, if one includes only one male of any one surname in each residence, the values are, respectively, 0.000534 and 0.000535. That is, the only structure seen between the two levels is in the influence of men of the same surname resident in the same house. An analysis of relationship by residence on the same street yields a value of Ri somewhat higher than that for the ward as a whole, however, suggesting that even within a ward there may be a tendency for the house of relatives occasionally to lie close together.

Changes over 100 years in degree of isolation of 21 parishes of the Lima Valley, Italy, assessed by surname isonymy

Changes over 100 years (1887-1986) in degree of isolation of 21 parishes of the Lima valley, Italy, were assessed using surname analysis. Crow and Mange's inbreeding coefficients and Lasker and Kaplan's repeated pair values were calculated using 8026 marriage records; temporal changes were assessed by dividing birth cohorts into 4 time periods of 25 years each: 1887-1911, 1912-1936, 1937-1961, and 1962-1986. Analysis was carried out at 2 hierarchical levels: the population of the valley as a whole and the valley's subdivision into 21 parishes. The relationship between population size and level of isonymy during the breakdown of isolates was investigated. The results show that there is a small difference in inbreeding coefficients between the first 2 periods at either hierarchical level of analysis and a substantial decrease in marital isonymy during the study period is mostly due to the change in male random isonymy. Furthermore, the Fn value at the higher hierarchical level almost coincides with the mean F value at the lower hierarchical level, indicating that over time the parish remained the fundamental reproductive unit. Regression analysis showed that geographic isolation became increasingly important in differentiation among the parishes in population size and in levels of inbreeding. The marked deviation from equilibrium between drift and migration that characterizes the breakdown of isolates of almost all the rural populations is an important disturbing factor in assessing the relationship between level of inbreeding and population size. Comparison over time allows us to better describe the evolutionary forces at the basis of the changes in genetic structure of a population.

Calculating components of the coefficient of relationship

The coefficient of relationship by isonymy is Ri=sigma(n(n-1)/2(N(N-1)) in which n is the number of persons of each surname and N=sigma n. Dividing Ri into two components, one for the contribution of co-residence (family size) and the other for diversity of surnames among residences is achieved by letting a1 represent sigma n(n-1) for all residents, a2 represent sigma n(n-1) after eliminating all but one individual of any name at any residence, and b1 represent 2N(N-1) for all residents. Then the component for inter-residence diversity is a2/b1 and the component for relationship by co-residence (including the interaction) is (a1-a2)/b1. By the same logic it is possible to calculate separately the interaction component, but the additional information seems of limited importance.

Census versus sample data in isonymy studies: relationship at short distances

The degree of isonymy is greatly influenced by whether or not it includes the extent of isonymous relationship resulting from persons of the same surname living in the same residence. In a sample of 313 male citizens in one English village and 386 males in another village, isonymy within the same residence averaged 0.86. Total relationship by isonymy was 60% and 48% understated in samples of one individual of any surname at each residence in the two villages, such as would result from use of telephone directories as a source. In these villages isonymy between occupants of contiguous houses was also elevated. Only a small fraction of the total isonymy could be attributed to different residences on the same street, so this has little effect on the coefficient of relationship for the whole village or region. The larger the population, the less the bias from ignoring within-household relationships.

Repetition of the same pair of surnames in marriages in Albanian Italians, Greek Italians, and the Italian population of Campobasso Province

The isolation of a population as a result of any boundary leads to a kinship mating pattern, the extent of which can be measured by the frequency of repeated pairs of surnames in actual marriages compared with that in random pairings. The repeated pairs within surname lineages (RPw) method can be used to assess random repetitions and the endogamous or exogamous behavior of a population. The RPw method was applied to data from grandparent surnames of children living in 45 Albanian Italian and 13 Greek Italian villages of southern Italy and Sicily and in 22 Italian villages of Campobasso Province (central Italy). The total mean RPw was 0.02782 in Albanian Italians, 0.01993 in Greek Italians, and 0.03427 in the Italian-speaking population. When RPw was subdivided into its components and compared with random and marital isonymy, the low level of inbreeding shown by the two southern Italian ethnic minorities and by the Italian population of Campobasso Province could be accounted for by the subdivision of the populations.

Further note on possible changes in the geographic distribution of ABO and RH blood groups in Great Britain

We previously reported the tendency for the clines of blood group distributions in England, Scotland, and Wales to become flatter as the result of internal migration. The data came from a one-week birth cohort followed for 16 years. A further seven years and a finer geographic analysis yield mixed results. The Rh- clines did become flatter, but the clines for ABO blood groups A and O showed the opposite tendency during the additional years.

Relationship of people across an international border based on an isonymy analysis across the German-Danish frontier

Using lists of names of male personal telephone subscribers, isonymy was calculated within and between 29 contiguous areas in a north-to-south line extending 210 km south of Grindsted, Denmark. Each area shared some surname(s) with every other area. Isonymy was high across both the present and past borders of Denmark with Germany and was consistently lower in areas beyond 160 km south of Grindsted. Relative isonymy between areas was also smaller on average south of the present border than north of it, and smaller still for pairs of areas spanning the border. This is partly accounted for by decreases in isonymy with distance, but the slope of the logistic regression on distance is greater for the northern moiety than the southern one. Most of these findings can be traced to the influence of common surnames ending in 'sen', the distribution of which tends to correlate highly with isonymy. Such surnames tend to be of recent origin and to be very frequent and hence highly polygenic. Thus much of the heterogeneity is explained by surname history rather than genetic heterogeneity.

Behavioral influences on the evolution of human genetic diversity

Unique aspects of human behavior account for special features of human evolution. For instance, the best data on the mating patterns of undisturbed hunting-gathering populations are those of Norman Tinsdale for Australian Aborigines. Among some 574 linguistically distinct tribes, he collected information on 755 marriages from the period prior to significant contact with Europeans. One can extrapolate from these data on 1510 individuals. In an average tribe of some 500 individuals, 62-65 of them would have had one parent who was a member of a different tribe and 7 or 8 of them would have been offspring of a parent from a distant tribe, not an adjacent one. Such rates of intergroup marriage, generation after generation, would have produced considerable gene flow, a pattern that likely has occurred during most of human prehistory. The pattern of descent is trellis-like, not one of successive fissioning. Some genetic diversity between tribes must have existed, but most genetic variation would have been within rather than between tribal groups. One implication of this pattern is that ethnically defined groups are not a suitable basis for studying human genetic diversity. Geographically stratified random sampling of the species would be more likely to ensure an unbiased estimate of genome variability.

Influence of social class on the correlation of stature of adult children with that of their mothers and fathers

Studies of parent-child correlations in stature require data which can be viewed as random samples of some general population and which are large enough to allow partition of the variable and evaluation of non-genetic and genetic influences. In a sample of 4336 individuals drawn from a cohort of all persons born in England, Scotland and Wales in 1 week in 1958, the correlation of statures of the males with their fathers, the females with their fathers, the males with their mothers and the females with their mothers were 0.36, 0.43, 0.41 and 0.47 respectively at age 16 of the offspring and 0.41, 0.41, 0.47 and 0.46 respectively at age 23. Allowance for the occupational social class of the fathers lowers the correlations, but in no case by more than 5%. Allowance for the occupational class achieved by the offspring by age 23 has little effect on the correlations.

Effect of interregional migration on geographic variability in biological and social traits in Great Britain

Over 19% of individuals born in England, Scotland, and Wales during March 3-9, 1958, resided in 1981 in a region different from the region of their birth. This internal migration among the 11 regions increased geographic homogeneity for one genetic variable (ABO blood group). Cramer's V for mother's A, O, and B or AB blood group decreased from 0.0504 to 0.0476. Mother's Rh+/- blood group was not significantly different among regions of place of birth or subsequent place of residence of the offspring. Variability among the regions increased by migration from region of birth to region of residence 23 years later for the social class of male head of household (Cramer's V increased from 0.0815 to 0.0877) and for years of schooling completed (V increased from 0.107 to 0.129). Stature behaved more like the social variables (mean square deviation among regions increased from 371 cm2 to 481 cm2 in females and from 426 cm2 to 471 cm2 in males), but body weight tended to become more uniform among regions (mean square deviation decreased from 220 kg2 to 178 kg2 in females and from 315 kg2 to 260 kg2 in males).

Inbreeding coefficients from the surnames of grandparents of the schoolchildren in Albanian-speaking Italian villages

Data on grandparental surnames were obtained from children in 45 Italo-Albanesi villages in nine provinces of southern Italy and Sicily. Concordance of surnames (isonymy) and inbreedding by village were estimated for each province and on the total sample. Total mean isonymy is 0.0251. The weighted mean inbreeding coefficient, and its random and non-random components are 0.0063, 0.0024 and 0.0039, respectively. Isonymy values are similar to those of rural Italian villages except that Alpine and some Appennine villages appear to be more isolated and inbred.

A spatial analysis of 100 surnames in England and Wales

Spatial patterns are described and analysed for the 84 most common surnames in England and Wales, as well as 16 others selected for various reasons. At least three-quarters of the surname frequencies show spatial structure and are heterogeneous over the area of study. While they do not exhibit clines extending over the entire area of study, they do divide into four characteristics patterns. Spatial autocorrelation, while significant, is relatively low; similarity in surname frequency does not extend much beyond 100 km. Correlograms could be clustered to yield groups of surfaces denoting partial clines, isolation by distance, and differentiation at far distances. A method for detecting zones of rapid change found 21 such zones, mostly near the periphery of the study area. These boundaries do not indicate barriers to gene flow, but appear to be patterns brought about by historical factors. There are diffusion patterns between areas that differ greatly in surname composition, such as Wales and central England. There is little evidence of long-distance movements involving several surnames. At least three characteristic migration patterns, east-west and north-south diffusion and local dispersal, were found.

Given name relationships support surname 'genetics': a note and correction

Analysis of a further four samples of first names in the index of marriages registered in England and Wales in the first 3 months of 1975 support the claim that there is no significant difference of 'between' versus 'within' registration district Ri. Since given names show none of the localisation seen in surnames, the surname geography is ascribable to genetic rather than cultural factors of personal naming. The correct formulations for coefficient of relationship by isonymy are given.

Cultural factor in the geographic distribution of personal names: pseudogenetic analysis of first names used to estimate the cultural component of coefficients of relationship by isonymy

By applying to given names formulations for analyzing the genetics of surname distributions, under certain assumptions one can separate the genetic components from the cultural components of surname distributions. Geographic distributions of surnames regularly yield larger coefficients of relationship or kinship within local populations than between them: for instance, Ri = 75 x 10(-5) within a local area in England but the Ri of those villages with all of England and Wales is 42 x 10(-5). On the contrary, the first names in an English and Welsh sample give essentially the same pseudocoefficient (based on first names) within registration districts (Ri = 354 x 10(-5) as between districts (Ri = 370 x 10(-5). Thus the decrease with distance of the coefficients based on surnames can be ascribed to the genetic component according to the Malécot principle, assuming that the first names are chosen in the same way as the surnames originated and consequently that the cultural component of surname distributions is no more localized than the distribution of given names (in this sample not at all).

Inbreeding coefficients from isonymy in the Italian-Greek villages

There are four Italian-Greek communes in one area of Calabria and nine in an area of Apulia. The communes are too small to give good information on marital isonymy for each, but in the Calabrian area the weighted mean inbreeding coefficient and its random and non-random components are F = 0.01442, Fr = 0.00450 and Fn = 0.00997, respectively. The weighted means in the more populous Apulian area are significantly lower, F = 0.00423, Fr = 0.00379, Fn = 0.00045.

Relationships estimated by isonymy among the Italo-Greco villages of southern Italy

Surnames of parents and grandparents were collected from 1993 children in the primary schools of the thirteen Italo-Greco communes that lie in two areas, four communes in Reggio Calabria in the "toe" of Italy and nine in Lecce in the "heel." The coefficients of relationship by isonymy show almost no relationship between the two areas. The smaller area in Reggio Calabria Province has consistently larger coefficients of relationship between communes than the larger area in Lecce Province. The difference can be ascribed to greater accumulated random isonymy in the smaller area. These populations are not genetic isolates, but each area shows a degree of cohesiveness with respect to surnames that suggests that they are genetically somewhat distinct. Contiguous pairs of communes tend to have higher coefficients of relationship than pairs of communes separated by intervening communes.

Lack of an association between ABO and Rh blood group polymorphisms and stature, body weight, and BMI in a cohort of British women

Analysis of variance shows no significant associations between stature, weight, or body mass index (BMI) and ABO or Rh blood group phenotypes among a sample of mothers in England, Scotland, and Wales whose children were born during March 3-9, 1958. Social factors are significantly associated with stature and weight; the effects of social class of the women's fathers, regions of birth of the women, their ages, whether their education continued beyond age 16 or not, and the total number of births were separated out by regression analysis. The adjusted residual regression of ABO and Rh phenotypes were not significantly related to reported stature, weight, or BMI.

A model for the study of sex-linked influences through the study of different classes of cousin pairs

Higher positive correlations in certain classes of cousin pairs compared with others would imply sex-linked genetic factors. Thus Y-linked influences would raise correlations in patrilateral male cousins. X-linkage would raise correlation (to various extents) in male and mixed sex pairs where each male is related through his mother, for instance. Among female first cousins, however, patrilateral pairs have the highest degree of kinship and cross cousins the lowest in respect to X-linked genes.

Effects of social class differences and social mobility on growth in height, weight and body mass index in a British cohort

By longitudinally linking follow-up restudies of the National Child Development Study it has been possible to examine not only the well-known association of social class with the size of the child, but also with longitudinal growth, and, in addition the effect of social mobility on growth. The relation of type of occupation of the male head of household to height and weight of the child is seen at all ages (7, 11 and 16) but class influence on growth from 7 years onward is minimal. Social mobility is a significant factor especially in relation to stature but is not significantly related to growth after age 7 so the effect of underlying conditions on the children precedes the change of type of occupation by their fathers.

Geographical distribution of common surnames in England and Wales

Analysis of the geographical distribution of 84 common surnames indicated that nearly all surnames are not distributed at random, but show distinctive geographical patterning. Factor analysis also revealed that there were similarities between surname distributions. The first two factors highlighted primarily Welsh surnames while the third and fourth factors were generally north of England names, mainly those ending in 'son'.

Geographic distribution of surnames in Britain: the Smiths and Joneses

The geographic distribution of the Smiths and Joneses among all persons married in England and Wales in 3 months of 1975 showed well defined frequency clines comparable to those of blood group distributions. Polynomial regressions of relative frequencies of these surnames on latitude and longitude account for 12% and 36% of the variation among registration districts in frequency of Smiths and Joneses, respectively.

Surnames in five English villages: relationship to each other, to surrounding areas, and to England and Wales

The surnames of all adult residents of five neighbouring communities in the fenlands of Cambridgeshire are compared with each other and with those of the 165,533 persons married in England and Wales in the period January to March 1975. Among the villages the average coefficient of relationship by isonymy (Ri) is 76 × 10−5. The villages nearer together may have a tendency to higher values of Ri: the correlation of Ri with the natural log of distance between villages is −0·49, p = 0·07. The surnames of the five villages give a weighted average Ri with the whole area about 25 miles of 54 × 10−5 and with a zone 25–40 miles away of 45 × 10−5, whereas Ri with all England and Wales is 42 × 10−5. Rare surnames show a much sharper gradient and contribute approximately twice as much to the coefficient in the 25–40 mile zone and six times as much within 25 miles as found with all England and Wales. Moderately frequent surnames and even common surnames show the same gradient, but to a lesser degree. In the part of Cambridgeshire studied, the present distribution of surnames indicates a slight but appreciable local isolation, with the degree of relationship decreasing from among local villages to that between the villages and all England and Wales. This pattern is consistent with the theory of genetic inbreeding based on distance but there is considerable variability in individual instances.

Secular trends in relationship as estimated by surnames: a study of a Tyneside parish

From the surnames of 21 132 individuals baptised in the Tyneside parish of Whickham between 1577 and 1758, the extent to which individuals of different periods bear the same surname shows (1) a coefficient of correlation of +0.93 between data on males and females; (2) the highest rates between individuals of the same 26-year period, lower between adjacent periods and still lower between periods more separated in time; (3) an apparent tendency for the closeness of genetic interrelationship to decrease from the late 16th and early 17th centuries, and to level off with fluctuations through the mid-18th century. This drop was concurrent with military campaigns and epidemics which affected the parish, but preceded the industrial revolution with its possible effects on labour migrations.

Anthropometric measurements and Darwinian fitness

The relationship between anthropometrics and three measures of Darwinian fitness—number of surviving children, number of living siblings and marital status—was sought in a population practising no contraception. The pattern suggestive of stabilizing selection was evident for one dimension, destabilizing selection for another dimension, and directional selection for yet another. The dimensions studied were those least intercorrelated one with another. Stabilizing selection for human physical characteristics may not be a universal phenomenon.

The structure of the human population of the Isles of Scilly: inferences from surnames and birthplaces listed in census and marriage records

Investigation of the commonality of surnames among the isles of Scilly using data covering the period from 1726 to the present, yields coefficients of relationship by isonymy which are compared with measures of relationship based on marital migration rates. The coefficients of relationship, calculated from lists of surnames in the marriage registers, and from lists of male surnames from the 1851 and 1871 censuses, are higher than any values yet reported for other British populations. However, the coefficient of relationship between the islands as a whole and the mainland of England and Wales is similar to the values reported for other English populations. Comparison of coefficients of relationship by marital isonymy show an overall decrease in the frequency of such marriages over time and a slight tendency to avoid marriages between close relatives. The lack of correlation between coefficients of relationship by random isonymy and marital migration rates derived from the census data is believed to be due to fluctuations in marriage patterns over time, the marital migration rates accounting for single generational patterns, while the coefficients of relationship by isonymy reflect the accumulated breeding structure. The chief influences on the coefficients of relationship appear to have been the founder effect, coupled with a long history of inter-island marriage but a high rate of endogamy from the islands as a whole.

Communality of surnames: a measures of biological interrelationships among thirty-one settlements in upper Val Varaita in the Italian alps

The upper part of the Alpine valley of the Varaita (population about 1,596) has three communities situated in a triangle about 6 km on a side but divided into frazioni (clusters of houses) that form a "Y" along the road and river in the valley floor. The coefficient of relationship (Ri) based on the communality of surnames between pairs of frazioni and between pairs of the communities shows very high levels of interrelationship (up to 0.26 compared with 0.5 for brothers). It tends to be higher between frazioni of the same community than between communities. It is highest among five pairs of contiguous frazioni within 1 km of each other. It tends to be highest between frazioni of the community least influenced by tourism and migration. It is not dmonstrably affected by the division of one community into two parishes. It is about twice as high between contiguous communities as between the two communities with an intervening one. The high values represent the long history of the same surnames in the valley and the accumulation of relationship through high levels of valley endogamy.

Relationships among the Otmoor villages and surrounding communities as inferred from surnames contained in the current register of electors

The interrelationships of 17 villages and a town near Otmoor have been investigated by means of the coefficients of relationships by isonymy: Ri = sigma(Si1Si2)/2N1N2 in which Si1 and Si2 are the respective frequencies of the ith surname in communities 1 and 2 and N1 and N2 are the respective sample sizes. The coefficients calculated from all adult males listed in the current registers electros range from 0 to over 0.008. The coefficient of relationship within the town and between the town and villages average 0.0005. Those between villages average 0.0007. The correlation between the coefficient of relationship by isonymy wth the frequency of marriages between parishes is +0.75 but this high correlation rests largely on a few very closely related pairs of parishes and largely on the contribution to Ri of relatively rare surnames. The method can easily be applied to wide surveys, and even to historic data, to explore the development of intercommunity relationships in the period since surnames became general.

Physical anthropology: the search for general processes and principles

Physical anthropology consists of two interdependent types of study: (1) the biological history of man and (2) general biological processes in man (such as mechanisms of evolution and growth). Popular interest may focus on the former, the fascinating story of the origin of man and of specific people, but the latter affords physical anthropology potential practical value in respect to medicine, dentistry, public health, and population policy. The study of general processes is the study of human beings in particular situations, not for what we can learn about these particular populations but for the sake of generalization about mankind anywhere in comparable situations. This is, of course, the purpose of experimental science in general, but in anthropology the method is usually comparative. Long ago the study of the growth of the two sexes and of children in different countries was started on a comparative basis as was the study of the so-called secular change in adult stature. By 1911 Franz Boas had compared the changes in stature and head form of children of several different immigrant groups in the United States. There have since been comparative studies of the amount and distribution of body fat (but not yet adequate comparative measurements of the relation of tissue components to diet and to diseases). Demographic patterns, inbreeding, outbreeding, and their effects are other general problems. The Human Adaptability Project of the International Biological Program promises studies of human response to heat, cold, altitude, and other conditions on a wide international basis. If supported, these could turn physical anthropology's search in a useful direction. The functional biology of people of even out-of-the-way communities will be compared with each other. These studies can yield general statements concerning human response to types of ecological situation including such sociocultural conditions as those of hunting-gathering tribes and urban slums.