Indian Academy of Sciences

PERSPECTIVES

Endogamy, consanguinity and community genetics

A. H. BITTLES* Centre for Human Genetics, Edith Cowan University, 100 Joondalup Drive, Perth, WA 6027, Australia The population of India is composed of many thousands of subpopulations, divided by geography, language, religion and caste or biraderi (patrilineage) boundaries, with endogamous marriage the norm. The net effect has been the creation of multiple genetic isolates with individual mutation profiles, but to date the clinical consequences of this highly complex differentiation have been largely ignored. In contrast, the topic of consanguinity continues to attract attention among medical and population geneticists, clinicians and social scientists. The significant progress made in India in improving childhood nutritional status and combating infectious disease means that genetic disorders have assumed everincreasing importance. In populations where consanguineous marriage is widely practised, recessive genetic disorders will continue to gain greater prominence in the overall spectrum of ill health. At the same time this increase will in part be negated by urbanization and the move to smaller family sizes, which predictably will result in a decline in the prevalence of consanguineous unions. Developing an understanding of these changes will require a wide-ranging and multidisciplinary investigative approach for which community genetics is ideally suited. mutism were collected across four geographical subdivisions, and for 13 Hindu and two Sikh castes and 17 Muslim biraderis (patrilineages) (Census of India 1921, 1923). From a genetic perspective the population of India is unique in its size and level of subdivision, with 15 major languages and six main religions. Within the majority Hindu community there are an estimated 3000 major castes, 1055 scheduled castes and 572 schedules tribes (Bhasin et al. 1992), and the Muslim population of over 130 million is similarly subdivided into Sunni, Shia, Ismaili and Dawoodi Bohra communities, and biraderis that are based on traditional social and occupational divisions (Shami et al. 1994). The net result is that the national population of 1050 million (PRB 2002) is composed of some 50,000 to 60,000 essentially endogamous subpopulations (Gadgil et al. 1998). Although mitochondrial DNA data have suggested that the founding female population of India was small (Roychoudhury et al. 2000), analysis of modern Indian populations has confirmed the continuing high level of gene differentiation (Majumder et al. 1999), with marked divisions according to caste ranks (Bamshad et al. 2001). While there is evidence of some female gene flow between castes (Bamshad et al. 1998), male gene flow across ethnic boundaries appears to have been negligible (Bamshad et al. 1998; Bhattacharya et al. 1999). Studies of this nature are, however, still in their infancy and in many cases they depend on quite broad population subclassifications. A hint of the possible levels of complexity of genetic subdivision was provided by the 1901 Census of India, which enumerated 101 Brahmin subgroups in the state of Mysore alone, although no information was provided as to whether or not marriage was permissible or had occurred between these different subgroups (Census of India, Mysore 1903). India also is subdivided into two major regions with respect to a preference for or avoidance of consanguineous marriage, a subject that strongly attracted the attention of J. B. S. Haldane (Haldane 1963). The interest expressed by Haldane was instrumental both in the inclusion of a subsurvey on consanguinity within the 1961 Census of India (Roychoudhury 1976) and in numerous subsequent studies into the prevalence of both consan91

Introduction
It can reasonably be claimed that India has long led the world in the field of community genetics. Beginning with the Census of India 1871 and continuing until the 1931 Census, information on the prevalence of blindness, deafness, insanity and leprosy were routinely collected and analysed on state and national bases, with additional information obtained on diseases specific to certain areas, e.g. filariasis in Travancore. In particular years and states the data also were collected with caste as basis. Thus in the 1921 Census of Punjab, which at that time comprised the present-day Indian states of Punjab and Haryana and the Pakistan province of Punjab, data on prevalence of deaf-

*E-mail: [email protected].

Journal of Genetics, Vol. 81, No. 3, December 2002

A. H. Bittles guineous and affinal marriage (Chakravarti 1968; Roychoudhury 1976, 1979; Singh and Tyagi 1987). As indicated in figure 1, based on the 19921993 National Family and Health Survey (IIPS 1995), unions between biological kin are uncommon in the northern, eastern and northeastern states because of a general prohibition on consanguineous marriage in the majority Hindu population. The prohibition is believed to date back to approximately 200 BC and forbids the marriage of a man with the daughter of his fathers sister or of his mothers sister, or of his mothers brother (Kapadia 1958; Sanghvi 1966; Rao 1984). Among Hindus of North India, the descendants of Indo-European-speaking populations who are thought to have entered the Indian subcontinent around 2000 BC, pedigrees are examined over an average of seven generations for males and five generations for females to ensure avoidance of a consanguineous union. By comparison, uncleniece marriage and first-cousin unions between a man and his maternal uncles daughter (mothers brothers daughter) have a long tradition in South India (Sastri 1976). Because of their traditional status consanguineous unions are regarded as customary for the peoples of southern India, i.e. those living south of the Narmada river. Cross-cousin marriage was recognized in the Hindu Marriage Act of 1955, and the legality of uncleniece unions was subsequently confirmed in the Hindu Code Bill of 1984 (Appaji Rao et al. 2002). As a result, consanguineous unions increase in prevalence in the states south of the Narmada, with the highest rates reported in Andhra Pradesh, Karnataka and Tamil Nadu (table 1). Kerala is an exception, in part because of the strict avoidance of consanguineous marriage among members of the Christian Syrian Orthodox church and, as indicated

Figure 1. Map of India, indicating prevalence of consanguineous marriage from 0% to 20+% by state reported in the 19921993 National Family and Health Survey (IIPS 1995). The numbering of the states corresponds to that in table 1.

92

Journal of Genetics, Vol. 81, No. 3, December 2002

Endogamy, consanguinity and community genetics below, by lower levels of consanguinity in a long-established local Muslim community. There also is evidence that the levels of consanguineous marriage within the Hindu community have decreased since the 1961 Census of India (Roychoudhury 1976), although high levels of consanguineous marriage have continued among tribal groups (Joseph and Mathew 1991). No comparable northsouth division exists in the Muslim population of India. Although consanguineous marriage is common in all Indian Muslim communities, the Quran contains no specific guidance that could be interpreted as encouraging consanguinity (Hussain 1999). Indeed, according to one of the hadith, a recorded pronouncement of the Prophet Muhammad, cousin marriages were better discouraged. On the other hand the Prophet married his daughter Fatima to Ali, his paternal first cousin and ward. Thus, for Muslims, cousin marriage could be interpreted as following the sunnah, i.e. the deeds of the Prophet. Variations are seen in the levels of consanguineous unions contracted in different branches of Islam and between specific communities. While 43.4% of marriages in the Shia community of Lucknow in North India were consanguineous (Basu 1975), the comparable figure was less than 10% among the Mappillas of Kerala, the descendents of Arab traders who settled in the southwest of India from the eighth century AD (Bittles and Hussain 2000). In fact, within the Indian subcontinent the clearest division in rates of Muslim consanguineous marriage is seen between Islamic communities in India and their Pakistani coreligionists (Hussain and Bittles 1998; Bittles and Hussain 2000), which emphasizes the important influence of local and regional customs in the arrangement of marriage contracts. The net result of the various religious preferences and prohibitions with respect to consanguinity are summarized in table 2. While consanguineous unions were reported in all religions, at national level the highest rates were observed in the Muslim and Buddhist communities and the lowest among Sikhs and Jains.

The genetic effects of consanguineous marriage

It can be argued that all humans are to some extent inbred, as evidenced by a simple calculation of the impossibly

Table 1. Region North 1. 2. 3. 4. 5. 6.

Consanguineous marriage in India by state, 19921993. State Consanguineous marriage (%) 4.3 1.0 0.8 8.0 0.9 1.3 4.1 7.5 5.0 5.7 5.0 3.9 1.7 2.1 2.7 0.5 1.5 1.9 10.6 4.9 21.0 30.8 29.7 7.5 11.9 Mean coefficient of inbreeding () 0.0023 0.0004 0.0003 0.0049 0.0006 0.0006 0.0025 0.0044 0.0032 0.0035 0.0030 0.0029 0.0010 0.0013 0.0018 0.0002 0.0009 0.0010 0.0066 0.0029 0.0131 0.0212 0.0180 0.0042 0.0075

Delhi Haryana Himachal Pradesh Jammu and Kashmir* Punjab Rajasthan

Central East

7. Madhya Pradesh 8. Uttar Pradesh 9. Bihar 10. Orissa 11. West Bengal 12. 13. 14. 15. 16. 17. 18. Arunachal Pradesh Assam Manipur Meghalaya Mizoram Nagaland Tripura

Northeast

West

19. Goa 20. Gujarat 21. Maharashtra 22. Andhra Pradesh 23. Karnataka 24. Kerala

South

All-India *Data were collected only in Jammu region.

Journal of Genetics, Vol. 81, No. 3, December 2002

93

A. H. Bittles
Table 2. Consanguineous marriage in India by religion, 19921993. Consanguineous marriage (%) 10.6 23.3 10.3 1.5 4.3 17.1 8.7 Mean coefficient of inbreeding () 0.0068 0.0141 0.0068 0.0009 0.0024 0.0107 0.0053

Religion Hindu Muslim Christian Sikh Jain Buddhist Others

cousin (F < 0.0156) differ only to a minor degree from those observed in the general population. In a large majority of clinical studies only the effects of consanguinity in the current generation are considered. As indicated above this may underestimate the actual level of homozygosity in an individual, but the decision is taken largely on practical grounds as in many communities the family pedigrees show complex multiple pathways of consanguinity that are difficult and very time-consuming to interpret.

The preference for consanguineous unions

The highest rates of consanguineous marriage in South India are usually reported in traditional rural areas and among the poorest and least educated groups. However, close kin marriage is commonplace even in Brahmin communities (Srinivasan and Mukherjee 1976), and it may be strongly favoured among major land-owning families as a means of ensuring the maintenance of their estates. The only communities in which consanguinity appears to be specifically avoided are those with origins in North India, and which continue to follow the traditions of that region (Appaji Rao et al. 1998). Both social and economic reasons are given for the popularity of consanguineous unions (Bittles et al. 1991; Bittles 1994). It is believed that consanguineous marriage helps to strengthen family ties, and at the same time health or financial uncertainties that could arise through marriage with a partner from another family or community will be avoided. Premarital arrangements are greatly simplified in consanguineous unions, and the relationship of a couple, especially the bride, with their in-laws (and family relatives) is expected to be more congenial. Economic considerations also are important and, in communities where either dowry or bride-wealth payments are the norm, the arrangement of a marriage within the family reduces or even abolishes the associated financial costs (Badaruddoza and Afzal 1995). Although the most common form of consanguineous marriage in all major societies is between first cousins, the importance of customary influences is apparent from variations in the specific types of first-cousin marriage contracted. While marriage to mothers brothers daughter is the strongly preferred form of consanguineous union among South Indian Hindus (Rao and Inbaraj 1977), all four types of first-cousin union, i.e. to fathers brothers daughter, to fathers sisters daughter, to mothers brothers daughter, and to mothers sister daughter, are arranged in South Asian Muslim communities.

large number of human ancestors implied if all matings had occurred with non-kin in past generations. The term consanguinity is used to describe unions between couples who are known to share genes inherited from one or more common ancestors. Globally, the most common form of consanguineous marriage is between first cousins, who are predicted to have 12.5% of their genes in common, and so on average their progeny will be homozygous at 6.25% of gene loci, equivalent to a coefficient of inbreeding (F ) of 0.0625 (Bittles 2001). Thus comparable values for uncleniece or double first-cousin progeny are F = 0.125, and for second cousins F = 0.0156. At the population level the mean coefficient of inbreeding () can be calculated according to the formula = p iF i , where the summation is over the proportion of individuals pi in each consanguinity category Fi. The highest value reported for an Indian community was 0.0449 in Pondicherry, where 20.2% of the marriages were uncle niece unions and 31.3% between first cousins (Puri et al. 1978). Composite figures for different Indian communities and comparable data on other populations worldwide have been compiled and are available in Bittles (1998) and at the website http://www.consang.net. In many well-established communities there is a long, unbroken history of consanguineous unions. Under these circumstances the cumulative level of inbreeding may be significantly higher than the value calculated for a single generation. A correction can be applied to account for the effects of ancestral inbreeding using the formula F = ()n (1 + FA), where FA is the ancestors inbreeding coefficient, n is the number of individuals in the path connecting the parents of the individual, and the summation is taken over each path in the pedigree that goes through a common ancestor. For most purposes a marriage is regarded as consanguineous if it has been contracted between spouses who are related as second cousins or closer (F = 0.0156), since the levels of homozygosity in marriages beyond second 94

The influence of consanguinity on fertility and mortality

An overall assessment of the genetic influence on prenatal losses within consanguineous unions is difficult, as

Journal of Genetics, Vol. 81, No. 3, December 2002

Endogamy, consanguinity and community genetics the information available is usually dependent on maternal recall, which often proves to be inaccurate (Wilcox and Horney 1984). The outcomes also may be dependent on nongenetic factors, such as maternal age, birth order and birth interval, factors that have seldom been incorporated into the study design. Where data have been available, primary infertility appeared to be reduced and there was little general evidence of increased numbers of miscarriages or stillbirths (Bittles 2001). Thus a meta-analysis conducted on 30 populations showed that at levels of consanguinity from F = 0.0156 to F = 0.125 the mean number of live births was higher in consanguineous than nonconsanguineous unions, and for first-cousin marriages (F = 0.0625) the fertility differential was significant at P < 0.0001 (Bittles et al. 2002). Most studies in India have indicated that early postnatal mortality is higher in the progeny of consanguineous unions, owing to the expression of deleterious recessive genes. Consanguinity-associated deaths are largely concentrated during the first year of life (Hussain et al. 2001), and multiple deaths have been reported in specific consanguineous families in proportion to the level of parental genetic relatedness (Bittles et al. 1991). Following a hypothesis by Sanghvi (1966), there is a body of opinion which suggests that through time the practice of consanguineous marriage would have led to the effective elimination of recessive lethal alleles from the gene pool. As with investigations into the association between consanguinity and prenatal losses, many earlier studies were hindered by a lack of control for important sociodemographic variables, and in some cases the sample sizes were unrealistically small to adequately test any hypothesis. Where appropriate clinical and laboratory investigations have been conducted there has been no convincing evidence of a consanguinity-associated cleansing of the gene pool. There has been a marked downward revision in the estimates of excess consanguinity-associated mortality through time (Bittles and Makov 1988), which principally results from better sampling techniques, including partial control for nongenetic variables. The most recent representative mortality estimate is based on a multinational study of over 600,000 pregnancies and live births, in which 10 of the 38 populations studied were from India and nine were from Pakistan (Bittles and Neel 1994). The analysis showed that, from approximately the sixth month of pregnancy to a median age of 10 years, deaths in first-cousin progeny exceeded mortality in nonconsanguineous progeny by an average of 44/1000 births. Even this figure may, however, be exaggerated as few if any of the constituent surveys in the analysis had explicitly included control for potential confounding sociodemographic factors, e.g. maternal age, maternal education, birth order and birth interval. The mean coefficient of inbreeding for the total population of India in the 19921993 National Family and Health Survey (IIPS 1995) was = 0.0075 (table 1). The data of Bittles and Neel (1994), where consanguinityassociated mortality to approximately 10 years of age averaged 44/1000 births at F = 0.0625, would suggest that on a national basis consanguinity would be a contributory factor in 5.2/1000 deaths. To place this figure in partial perspective, the rate of under-five-year mortality in India over a comparable study period was 109.3/1000 (IIPS 1995).

Consanguinity and morbidity

A comprehensive listing of genetic disorders that have been diagnosed in the progeny of consanguineous couples is available at the website http://www.consang.net. The diagnosis of birth defects in India may overlap with and reflect late foetal and neonatal survival rates, and the data that are available were almost exclusively collected in the major cities, whereas the majority of the population is rural. For these reasons the reported prevalence rates may be unrepresentative of the numbers of affected cases in the country as a whole. Many different types of genetic disorders have been reported to be more common among consanguineous progeny, for example congenital disorders (Centerwall and Centerwall 1966; Asha Bai et al. 1981; Agarwal et al. 1991), including neural tube defects (Kulkarni et al. 1989; Jain et al. 1993) and congenital heart defects (Jain et al. 1993; Badaruddoza et al. 1994; Gnanalingham et al. 1999). Autosomal recessive hearing loss disorders (Chen et al. 1997) and visual defects such as early-onset retinal dystrophies (Rahi et al. 1995), primary congenital glaucoma (Panicker et al. 2002) and anophthalmos (Hornby et al. 2001) also are present at increased prevalence. However, there is a paucity of national or even state-level data for all genetic types of disorder, which makes derivation of credible prevalence rates virtually impossible. A topic that has so far received little attention is the possible effect of consanguinity on adult-onset disease. Little comment can usefully be made because of the overall lack of information but, for example, a preliminary prediction that breast cancer appeared to be more common in females born to consanguineous parents in Pakistan (Shami et al. 1991) has recently been supported by studies on early-onset cases, i.e. patients under 40 years (Liede et al. 2002). Also, in a French-Canadian isolate a significant association between consanguinity, expression of the APOE-4 allele and late-onset Alzheimers disease has been identified (Vzina et al. 1999). Given the extremely high prevalence of premature coronary heart disease in South Asian males, an urgent detailed investigation of the possible role of recessive genes as predisposing disease factors would appear to be warranted. 95

Journal of Genetics, Vol. 81, No. 3, December 2002

A. H. Bittles

Discussion
While caste endogamy remains largely unchanged throughout India, data on trends in the prevalence of consanguineous marriage in predominantly Hindu South India are somewhat contradictory, with some studies suggesting a decline (Audinarayana and Krishnamoorthy 2000) and others that there has been no recent change (Bittles et al. 1993). In the Muslim population of India there has been no evidence of a reduction in consanguineous marriage during the last 40 years (Hussain and Bittles 2000). Nonetheless, it is probable that increased urbanization and the gradual shift to smaller family sizes will impose constraints on consanguineous marriage in future generations. In this respect, a reduced prevalence of uncleniece marriages would appear to be especially likely because of unacceptable age differentials between the potential partners (Radha Rama Devi et al. 1982). The excess risk that an autosomal recessive disorder will be expressed in the progeny of a consanguineous union is inversely proportional to the frequency of the disease allele in the gene pool (Bittles 2001). For this reason, during the last decade many disease genes that are rare in the general population have been identified and their chromosomal locations mapped by studying highly inbred families with multiple affected members. As previously discussed, in India the population is subdivided into many thousands of endogamous communities that through time have evolved into distinctive breeding pools. Whether or not a mutation will appear in all communities or be restricted to a single subcaste or biraderi will be dependent on the origin and the age of the community. At least three major migrations into the Indian subcontinent have been identified (Gadgil et al. 1998; Roychoudhury et al. 2000), but across historical time there almost certainly were other, smaller migrations involving one or several subpopulations. Four basic classes can be defined with respect to the age of mutations (table 3). For example, mutations that occurred over 100 generations ago may conceivably be found in all Hindu castes, whereas those of more recent origin probably have restricted distribution and may be unique to specific subcastes. The smaller the community the greater the probability that founder effect and genetic drift will exert a significant influence on the distribution patterns of specific mutations. Therefore, even in the absence of preferential consanguineous marriage, genetic isolation often results in an increased frequency of community-specific genetic diseases. This fact is commonly overlooked, with the consequent general assumption that where an autosomal recessive disease is present in a family or community at high frequency consanguinity is necessarily implicated. The situation is somewhat different in the Muslim population of India, because of the conversions from 96

Table 3. Period Modern Historical Pre-historical Ancient

The time scale of human mutations. Generations 10 > 10100 > 1004000 > 4000 Years 250 2502500 2500100,000 > 100,000

Hinduism that occurred from the twelfth to the nineteenth centuries AD (Hussain and Bittles 2000) and the various waves of invasion by peoples from Central Asia. The 1921 Census of Punjab also indicated the occurrence of inter-biraderi marriages and hence some gene pool admixture. As a result, it seems probable that the levels of genetic differentiation between biraderis would be less pronounced than those between castes. Nonetheless, genomic studies in Pakistan have demonstrated highly significant differences between Muslim biraderis at both autosomal and Y-chromosome loci (Wang et al. 2000). The presence of multiple discrete subpopulations creates substantial practical difficulties in the compilation of national disease registers, and in the organization of screening programmes for specific genetic disorders. On the positive side, once a specific mutation has been identified within a consanguineous pedigree, case ascertainment in other family members can be greatly simplified (Ahmed et al. 2002; Modell and Darr 2002). This applies within communities that are effective genetic isolates, whether or not consanguinity is involved. Thus in a rural South Indian community it was possible to rapidly identify the causative mutation for the autosomal dominant disorder familial adenomatous polyposis coli, and to efficiently counsel and test all persons at risk (Savithri et al. 2000). It should, however, be stressed that genetic heterogeneity may persist even where a recessive disorder is known to be common within a particular inbred subpopulation, with no guarantee that all affected members are homozygous for the same mutation. Furthermore, in many consanguineous isolates there is the possibility that two or more mutant alleles are segregating within a family, which greatly complicates diagnosis and genetic counselling (Bittles 2001). The total burden of disease in a population can increase significantly as populations progress in economic terms, with ameliorative treatment of formerly lethal genetic disorders placing an ever-increasing demand on family resources (Bittles 2001). Thus, although consanguineous marriage may remain culturally desirable, a major shift in balance between the social and economic benefits associated with intrafamilial marriage and adverse health outcomes can be predicted. These changes are already under way in India, and new diagnostic, counselling and treatment skills need to be rapidly developed in conjunction with appropriate community education programmes.

Journal of Genetics, Vol. 81, No. 3, December 2002

Endogamy, consanguinity and community genetics A multidisciplinary research team in North America has recently recommended that for neonatal and early-childhood investigations the progeny of first-cousin marriages should be treated in a manner comparable to the treatment of children of nonconsanguineous parents (Bennett et al. 2002). This advice also applies in India, although once again the enormous complexity of the population subdivisions may make any form of all-encompassing recommendation difficult to sustain. While India still has major problems in dealing with infectious diseases and nutritional disorders, genetic disease already presents as a significant, although largely underestimated, problem. The time is now appropriate to determine the prevalence of the major genetic diseases in different parts of the country, and in the various major communities, and to invest in the training of specialist medical, scientific, nursing and counselling staff.
Acknowledgements and dedication This paper is dedicated to the memory of Professor Vulimiri Ramalingaswami, F.R.S., and is based on the Inaugural Professor V. Ramalingaswami Lecture delivered at the Symposium on Community Genetics in Developing Countries, held in the Indian Institute of Science, Bangalore, on 16 January 2002 with generous support from the Sir Dorabji Tata Trust. Professor Ramalingaswami first came to international attention though his outstanding work on iodine deficiency disorders in the Himalayan region of India. His breadth of knowledge and administrative skills were acknowledged by his appointment as Director-General of the Indian Council of Medical Research, and in his subsequent secondment to UNICEF in New York. Throughout his long and highly distinguished career Professor Ramalingaswami led by personal example. His deeply compassionate nature, sincerity and great personal charm will always be remembered by his colleagues and by those who were fortunate to be counted among his friends. Excellent technical support in the preparation of the manuscript was provided by Sheena Sullivan. Asha Bai P. V., Jacob John T. and Subramaniam V. R. 1981 Reproductive wastage and developmental disorders in relation to consanguinity in South India. Trop. Geogr. Med. 33, 275280. Audinarayana N. and Krishnamoorthy S. 2000 Contribution of social and cultural factors to the decline in consanguinity in South India. Soc. Biol. 47, 189200. Badaruddoza and Afzal M. 1995 Effects of inbreeding on marriage payment in North India. J. Biosocial Sci. 27, 333 337. Badaruddoza, Afzal M. and Akhtaruzzaman 1994 Inbreeding and congenital heart disease in a North Indian population. Clin. Genet. 45, 288291. Bamshad M. J., Watkins W. S., Dixon M. E., Jorde L. B., Bhaskara Rao B., Naidu J. M., Prasad B. V. R., Rasanayagam A. and Hammer M. F. 1998 Female gene flow stratifies Hindu castes. Nature 395, 651652. Bamshad M. J., Kivisild T., Watkins W. S., Dixon M. E., Ricker C. E., Rao B. B. et al. 2001 Genetic evidence on the origins of Indian caste populations. Genome Res. 11, 1574 1585. Basu S. K. 1975 Effect of consanguinity among North Indian Muslims. J. Popul. Res. 2, 5768. Bennett R., Motulsky A. G., Bittles A. H., Hudgins L., Uhrich S., Lochner Doyle D., Silvey K., Scott R. C., Cheng E., McGillivray B., Steiner R. D. and Olson D. 2002 Genetic counseling and screening of consanguineous couples and their offspring: Recommendations of the National Society of Genetic Counselors. J. Genet. Counsel. 11, 97119. Bhasin M. K., Walter H. and Danker-Hopfe H. 1992 The distribution of genetical, morphological and behavioural traits among the peoples of Indian region, pp. 1435. Kamla-Raj, Delhi. Bhattacharya N. P., Basu P., Das M., Pramanik S., Banerjee R., Roy B., Roychoudhury S. and Majumder P. P. 1999 Negligible male gene-flow across ethnic boundaries in India, revealed by analysis of Y-chromosomal DNA polymorphisms. Genome Res. 9, 711719. Bittles A. H. 1994 The role and significance of consanguinity as a demographic variable. Popul. Dev. Rev. 20, 561 584. Bittles A. H. 1998 Empirical estimates of the prevalence of consanguineous marriage in contemporary societies. Working Report no. 74, Morrison Institute for Population and Resource Studies, Stanford University, Stanford, USA. Bittles A. H. 2001 Consanguinity and its relevance to clinical genetics. Clin. Genet. 60, 8998. Bittles A. H. and Hussain R. 2000 An analysis of consanguineous marriage in the Muslim population of India at regional and state levels. Ann. Hum. Biol. 27, 163171. Bittles A. H. and Makov E. 1988 Inbreeding in human populations: assessment of the costs. In Mating patterns (ed. C. G. N. Mascie-Taylor and A. J. Boyce), pp. 153167. Cambridge University Press, Cambridge. Bittles A. H. and Neel J. V. 1994 The costs of human inbreeding and their implications for variations at the DNA level. Nature Genet. 8, 117121. Bittles A. H., Mason W. M., Greene J. and Appaji Rao N. 1991 Reproductive behaviour and health in consanguineous marriages. Science 252, 789794. Bittles A. H., Coble J. M. and Appaji Rao N. 1993 Trends in consanguineous marriage in Karnataka, South India, 1980 1989. J. Biosocial Sci. 25, 111116. Bittles A. H., Grant J. C., Sullivan S. G. and Hussain R. 2002 Does inbreeding lead to decreased human fertility? Ann. Hum. Biol. 29, 111130.

References
Agarwal S. S., Singh U., Singh P. S., Singh S. S., Das V., Sharma A., Mehra P., Chandravati, Malik G. K. and Misra P. K. 1991 Prevalence and spectrum of congenital malformations in a prospective study at a teaching hospital. Indian J. Med. Res. B94, 413419. Ahmed S., Saleem M., Modell B. and Petrou M. 2002 Screening extended families for genetic hemoglobin disorders in Pakistan. New Engl. J. Med. 347, 11621168. Appaji Rao N., Savithri H. S., Radha Rama Devi A. and Bittles A. H. 1998 Consanguinitya common human heritage? The effects on the health and well-being of Indian populations. In The Indian human heritage (ed. D. Balasubramanian and N. Appaji Rao), pp. 1121. Universities Press, Hyderabad. Appaji Rao N., Savithri H. S. and Bittles A. H. 2002 A genetic perspective on the South Indian tradition of consanguineous marriage. In Austral-Asian encounters (ed. C. Vanden Driesen and S. Nandan), pp. 326341. Prestige Books, New Delhi.

Journal of Genetics, Vol. 81, No. 3, December 2002

97

A. H. Bittles
Census of India 1901, 1903 Volume 24, Mysore Part-1 Report (ed. T. Ananda Row), p. 454. Office of the Superintendent of Government Printing, Bangalore. Census of India 1921, 1923 Punjab and Delhi, Part-I Report (ed. L. Middleton and S. M. Jacob), volume 15, pp. 328330. Civil and Military Gazette Press, Lahore. Centerwall W. R. and Centerwall S. A. 1966 Consanguinity and congenital anomalies in South India: a pilot study. Indian J. Med. Res. 54, 11601167. Chakravarti M. R. 1968 Consanguinity in India. Z. Morphol. Anthropol. 60, 170183. Chen A., Wayne S., Bell A., Ramesh A., Srikumari Srisailapathy C. R., Scott D. A. et al. 1997 New gene for autosomal recessive non-syndromic hearing loss maps to either chromosome 3q or 19p. Am. J. Med. Genet. 71, 467471. Gadgil M., Joshi N. V., Manoharan S., Patil S. and Prasad U. V. S. 1998 Peopling of India. In The Indian human heritage (ed. D. Balasubramanian and N. Appaji Rao), pp. 100 129. Universities Press. Hyderabad. Gnanalingham M. G., Gnanalingham K. K. and Singh A. 1999 Congenital heart disease and parental consanguinity in South India. Acta Paediatr. 88, 473474. Haldane J. B. S. 1963 The implications of genetics for human society. In Genetics today: proceedings of the 11th International Congress of Genetics (ed. S. J. Geerts), pp. 91102. Pergamon Press, The Hague. Hornby S. J., Dandona L., Foster A., Jones R. B. and Gilbert C. E. 2001 Clinical findings, consanguinity and pedigrees in children with anophtalmos in southern India. Dev. Med. Child Neurol. 43, 392398. Hussain R. 1999 Community perceptions of reasons for preference for consanguineous marriages in Pakistan. J. Biosocial Sci. 31, 449461. Hussain R. and Bittles A. H. 1998 The prevalence and demographic characteristics of consanguineous marriages in Pakistan. J. Biosocial Sci. 30, 261279. Hussain R. and Bittles A. H. 2000 Sociodemographic correlates of consanguineous marriage in the Muslim population of India. J. Biosocial Sci. 32, 433442. Hussain R., Bittles A. H. and Sullivan S. 2001 Consanguinity and early mortality in the Muslim populations of India and Pakistan. Am. J. Hum. Biol. 13, 777787. IIPS 1995 National Family and Health Survey, India, 199293. International Institute for Population Sciences, Bombay. Jain V. K., Nalini P., Chandra R. and Srinivasan S. 1993 Congenital malformations, reproductive wastage and consanguineous mating. Aust. N. Z. J. Obstet. Gynaecol. 33, 3336. Joseph S. and Mathew P. M. 1991 Consanguinity studies in the Mudugars of Attappady, Kerala. J. Cytol. Genet. 26, 14. Kapadia K. M. 1958 Marriage and family in India (2nd edition), pp. 117137. Oxford University Press, Calcutta. Kulkarni M. L., Mathew M. A. and Reddy V. 1989 The range of neural tube defects in southern India. Arch. Dis. Child. 64, 210214. Liede A., Malik I. A., Aziz Z., de los Rios P., Kwan E. and Narod S. A. 2002 Contribution of BRCA1 and BRCA2 mutations to breast and ovarian cancer in Pakistan. Am. J. Hum. Genet. 71, 595606. Majumder P. P., Roy B., Banerjee S., Chakraborty M., Dey B., Mukherjee N., Roy M., Thakurta P. G. and Sil S. K. 1999 Human-specific insertion/deletion polymorphisms in Indian populations and their possible evolutionary implications. Eur. J. Hum. Genet. 7, 435446. Modell B. and Darr A. 2002 Genetic counselling and customary consanguineous marriage. Nature Rev. Genet. 3, 225229. Panicker S. G., Reddy A. B. M., Mandal A. K., Ahmed N., Nagarajaram H. A., Hasnain S. E. and Balasubramaniam D. 2002 Identification of novel mutations causing familial primary congenital glaucoma in Indian pedigrees. Invest. Ophthalmol. Vis. Sci. 43, 13581366. PRB 2002 World population data sheet. Population Reference Bureau, Washington, DC. Puri R. K., Verma I. C. and Bhargava B. K. 1978 Effects of consanguinity in a community in Pondicherry. In Medical genetics in India (ed. I. C. Verma), vol. 2, pp. 129139. Auroma Enterprises, Pondicherry. Radha Rama Devi A., Appaji Rao N. and Bittles A. H. 1982 Inbreeding in the State of Karnataka, South India. Hum. Hered. 32, 810. Rahi J. S., Sripathi S., Gilbert C. E. and Foster A. 1995 Childhood blindness in India: causes in 1318 blind school students in nine states. Eye 9, 545550. Rao P. S. S. 1984 Inbreeding in India: concepts and consequences. In The people of India (ed. J. R. Lukacs), pp. 239 268. Plenum, New York. Rao P. S. S. and Inbaraj S. G. 1977 Inbreeding in Tamil Nadu, South India. Soc. Biol. 24, 281288. Roychoudhury A. K. 1976 Incidence of inbreeding in different states of India. Demogr. India 5, 108119. Roychoudhury A. K. 1979 The incidence of affinal marriages in different states of India. Demogr. India 8, 211216. Roychoudhury S., Roy S., Dey B., Chakraborty M., Roy M., Roy B., Ramesh A., Prabhakaran N., Usha Rani M. V., Vishwanathan H., Mitra M., Sil S. K. and Majumder P. P. 2000 Fundamental genomic unity of ethnic India is revealed by analysis of mitochondrial DNA. Curr. Sci. 79, 1182 1192. Sanghvi L. D. 1966 Inbreeding in India. Eugen. Quart. 13, 291301. Sastri K. A. N. 1976 A history of South India: from prehistoric times to the fall of Vijayanagar (4th edition), p. 66. Oxford University Press, Madras. Savithri H. S., Venkatesha Murthy H. S., Baskaran G., Appaji Rao N., Kool D., Edkins E., Wang W. and Bittles A. H. 2000 Predictive testing for familial adenomatous polyposis in a rural South Indian community. Clin. Genet. 58, 5760. Shami S. A., Qaisar R. and Bittles A. H. 1991 Consanguinity and adult morbidity in Pakistan. Lancet 338, 954955. Shami S. A., Grant J. C. and Bittles A. H. 1994 Consanguineous marriage within social/occupational class boundaries in Pakistan. J. Biosocial Sci. 26, 9196. Singh B. and Tyagi D. 1987 Studies on inbreeding in India. Hum. Sci. 36, 159174. Srinivasan S. and Mukherjee D. P. 1976 Inbreeding among some Brahman populations of Tamil Nadu. Hum. Hered. 26, 131136. Vzina H., Heyer E., Fortier I., Quellette G., Robitaille Y. and Gauvreau D. 1999 A genealogical study of Alzheimer Disease in the Saguenay region of Quebec. Genet. Epidemiol. 16, 412425. Wang W., Sullivan S. G., Ahmed S., Chandler D., Zhivotovsky L. A. and Bittles A. H. 2000 A genome-based study of consanguinity in three co-resident endogamous Pakistan communities. Ann. Hum. Genet. 64, 4149. Wilcox A. J. and Horney L. F. 1984 Accuracy of spontaneous recall. Am. J. Epidemiol. 120, 727733.

98

Journal of Genetics, Vol. 81, No. 3, December 2002