Marry far and breed tall strong sons

By Razib Khan | July 7, 2011 12:34 am

ResearchBlogging.orgThe Pith: When it comes to the final outcome of a largely biologically specified trait like human height it looks as if it isn’t just the genes your parents give you that matters. Rather, the relationship of their genes also counts. The more dissimilar they are genetically, the taller you are likely to be (all things equal).

Dienekes points me to an interesting new paper in the American Journal of Physical Anthropology, Isolation by distance between spouses and its effect on children’s growth in height. The results are rather straightforward: the greater the distance between the origin of one’s parents, the taller one is likely to be, especially in the case of males. These findings were robust even after controlling for confounds such as socioeconomic status. Their explanation? Heterosis, whether through heterozygote advantage or the masking of recessive deleterious alleles.

The paper is short and sweet, but first one has to keep in mind the long history of this sort of research in the murky domain of human quantitative genetics. This is not a straight-forward molecular genetic paper where there’s a laser-like focus on one locus, and the mechanistic issues are clear and distinct. We are talking about a quantitative continuous trait, height, and how it varies within the population. We are also using geographical distance as a proxy for genetic distance. Finally, when it comes to the parameters affecting these quantitative traits there are a host of confounds, some of which are addressed in this paper. In other words, there’s no simple solution to the fact that nature can be quite the tangle, more so in some cases than others.

Because of the necessity for subtlety in this sort of statistical genetic work one must always be careful about taking results at face value. From what I can gather the history of topics such as heterosis in human genetics is always fraught with normative import. The founder of Cold Spring Harbor Laboratory, Charles Davenport, studied the outcomes of individuals who were a product of varied matings in relation to genetic distance in the early 1920s. This was summed up in his book Race Crossing in Jamaica:

A quantitative study of 3 groups of agricultural Jamaican adults: Blacks, Whites, and hybrids between them; also of several hundred children at all developmental stages. The studies are morphological, physiological, psychological, developmental and eugenical. The variability of each race and sex in respect to each bodily dimension and many basis vary just as morphological traits do. In some sensory tests the Blacks are superior to Whites; in some intellectual tests the reverse is found. A portion of the hybrids are mentally inferior to the Blacks. The negro child has, apparently, from birth on, different physical proportions than the white child.


Because of the fears of miscegenation in the early 20th century scholars had a strong bias toward finding the data to confirm the assumption that admixture between divergent human kinds resulted in a breakdown and depression in trait value in relation to both parental lineages. Today this is not so. Rather, I would argue that the bias is now in the opposite direction, at least in the West. My friend Armand Leroi wrote Meet the world’s most perfect mutant seven years ago. Who is the most perfect human according to Armand? She is Saira Mohan, a model of Indian, Irish and French ancestry. Armand concludes:

If deleterious mutations rob us of it, they should do so with particular efficacy if we marry our relatives. Most novel mutations are at least partly recessive, and inbreeding should accentuate their negative effects. Many weird genetic disorders come from Pakistan and Saudi Arabia, where there is a strong tradition of first-cousin marriage.

Conversely, people of mixed ancestry should show the benefits of concealing recessive mutations. And this, I suspect, is the true meaning of Saira Mohan: half Punjabi, quarter Irish, quarter French and altogether delightful. She, too, is a mutant – but a little less so than most of us.


Thandie Newton masking recessive alleles

This is entirely in keeping with the dominant ethos of the global elite, which aims for a panmixia of genes in concert with an alignment of a particular set of cosmopolitan post-materialist memes. But, as I pointed out to Armand there are also cases where crosses between genetic backgrounds may have deleterious consequences. For example, a European specific allele in African Americans may have a negative fitness interaction with the predominant African genetic background of this population. I am not implying here that science is fiction, a construction of our biases and preconceptions. But the dominant cultural narrative framework does put pressure upon how we interpret science, and all the more so in domains which require a level of statistical subtlety and personal candor.

Of course now that we can see exactly how individuals are mutant at the level of the genome Armand’s supposition can actually be tested. That is, we can see how many deleterious recessive alleles are in fact masked in people of hybrid origin. That at least may plug one of the fuzzy spots in our picture of how genetic backgrounds interact in humans.

I prefaced the review of a paper on marital distance and height with some history of science and a reflection of how contemporary values influence the generation and interpretation of knowledge because there’s a lot of confusing material in the literature on correlations between genetic distance and trait value. There is the result that marriages between 3rd cousins seem the most fertile in Iceland. Is this because of a balance between genetic incompatibilities and expression of recessive diseases? Or perhaps the answer lies in social dynamics, insofar as people who come from related lineages are more likely to weather difficult times in their relationship? It’s one study from Iceland. But of course the minority who vociferously argue against racial amalgamation and admixture on moral/normative grounds will focus upon this specific positive empirical finding in the literature. Now, Iceland is ideal for many human genetic studies because it  has excellent records and is culturally homogeneous. But at the end of the day Iceland is still Iceland.

And today Poland is still Poland. I say that because this study tracks thousands of Polish youth over the years. Here’s the abstract:

Heterosis is thought to be an important contributor to human growth and development. Marital distance (distance between parental birthplaces) is commonly considered as a factor favoring the occurrence of heterosis and can be used as a proximate measure of its level. The aim of this study is to assess the net effect of expected heterosis resulting from marital migration on the height of offspring, controlling for midparental height and socioeconomic status (SES). Height measurements on 2,675 boys and 2,603 girls ages 6 to 18 years from Ostrowiec Świętokrzyski, Poland were analyzed along with sociodemographic data from their parents. Midparental height was calculated as the average of the reported heights of the parents. Analyses revealed that marital distance, midparental height, and SES had a significant effect on height in boys and girls. The net effect of marital distance was much more marked in boys than girls, whereas other factors showed comparable effects. Marital distance appears to be an independent and important factor influencing the height of offspring. According to the “isolation by distance” hypothesis, greater distance between parental birthplaces may increase heterozygosity, potentially promoting heterosis. We propose that these conditions may result in reduced metabolic costs of growth among the heterozygous individuals.

As you may know, height is substantially heritable. That means that ~80-90% of the variation in the trait within the population in developed nations is due to variation in genes. This has some validity even within families. Tall parents tend to given rise to tall offspring, though there is a variation around the expectation. In other words, siblings differ in height, in part because of environmental factors, but also in part because siblings differ in their genetic endowments from their parents. So naively one can model this like so:

Height ~ Genetic endowment + Environmental contingencies

The genetic endowment is a function of the mid-parent value in standard deviation units. That means you average the standard deviations of the parents from the sex-controlled mean. Let’s give a concrete example. Imagine a male who is 5’8 inches, and a female who is 5’7 inches. The standard deviation for height is ~3 inches, with the American male mean being 5’10 inches and female being 5’4 inches. That means that the male is -2/3 standard deviations below the mean, and the female is 1 standard deviation above the mean. The expectation for their offspring then will be 1/3 standard deviation above the mean (5’11 for males, 5’5 for females). But because of the variation in the nature of genetics and environment, there’s actually going to be a standard deviation of ~3 inches for the offspring (e.g., ~70% chance that the male will be between 5’8 and 6’2). There is also the reality that because environmental factors aren’t heritable the offspring should regress somewhat back to the population mean all things equal, though in the case of height not too much because it is so genetically influenced.

A few years ago I played this game with libertarian pundits Megan McArdle and Peter Suderman, who announced their engagement. Megan and Peter are both 6’2. I estimated that the expected value is that any son of theirs would be 6 feet 3.6 inches, and any daughter 5 feet 9.6 inches. How can it be that their sons should be taller than either of them? Remember that Megan is much taller than Peter in standard deviation units in relation to her sex.

Now how would expectation be altered if Megan McArdle and Peter Suderman were full-siblings? (they are not full-siblings, this is a thought experiment!) At this point even if you had never taken college genetics you might be wondering whether it makes sense to calculate an expectation for the height of the offspring of two full-siblings. You know very well that there are much more serious genetic issues at hand. Going back to the relation above, you might update it like so:

Height ~ Genetic endowment + Environmental contingencies – Incest decrement

Even stipulating viability of the offspring, any child of full-siblings would exhibit all the problems that Armand alludes to above. It seems likely that whatever potential their parents might impart to their offspring, the combination of their genotypes would be highly deleterious, because near kin carry the same recessives. The paper above posits the inverse effect, where outbreeding results in greater outcomes than are to be expected based on the mid-parent trait value. In this telling, height is a proxy for health and development. This seems biologically plausible in the case of humans. Individuals who marry those genetically dissimilar impart gains of fitness to their offspring by virtue of elevated heterozygosity. So now we create a new relation:

Height ~ Genetic endowment + Environmental contingencies + Magnitude of outbreeding

In pre-modern societies individuals tended to marry those close to them geographically. Even if cousin marriage was not normally practiced, over time clusters of villages would form networks of de facto consanguinity. In the 19th and especially 20th century much of this in the extreme cases abated in Europe because of better transport. L. L. Cavalli-Sforza documented this in Consanguinity, Inbreeding, and Genetic Drift in Italy. Modern roads resulted in a radical drop in inbreeding in mountainous regions of the country. Some researchers have argued that this shift resulted in an increased level of height, intelligence, and health, among European populations.

With that, here’s a nice map from Consang.net:

Going back to the paper, after controlling for socioeconomic status they found that:

1) The increased marital distance predicts taller height than expected, especially in boys.

2) This effect is most noticeable in boys who already have parents who are relatively tall.

3) Finally, greater marital distance seems to be correlated with greater height in the parents!

The last is actually a possible reason why there’s no reason to appeal to heterosis at all. This might simply be a function of assortative mating of tall individuals who are more mobile. In the paper the authors go at length about sexual selection, greater mobility of individuals who are taller, etc. But whatever the reason, this shows exactly the care which must be taken with these sorts of results. It is known for example that taller individuals seem to have higher I.Q.s, leading some to assert that the genes which control height and I.Q. variance must be the same (some of them almost certainly are if there are many loci of small effect). But, it turns out that this height-I.Q. correlation disappears within families (tall siblings are no smarter than short siblings), implying that the correlation might be a function of assortative mating.

As for why there may be a sex difference, the authors suggest that heterosis may manifest at different points in the developmental arc of children. Females mature somewhat faster than males. This may be so, the sexes differ and such. But my own preference is that the original results merit a deeper and expanded examination before we posit an evolutionary story (that’s not possible in a scientific paper which needs a discussion, but I’m proposing an ideal world of knowledge generation and refinement!). The empirics need to be firmed up before we scaffold it in theory. Poland is Poland, and if you troll through enough data sets there’ll be millions of correlations which are publishable. And yet we are living in the age of information, so we had better get going in sieving through it. At the end of the paper the authors go in a direction which I think might yield some interesting finds in the future:

One possible limitation of our study and explanation of the results may come from the fact that we used geographical distance between parental birthplaces as the only approximate measure of offspring heterozygosity. Further studies should focus on more direct examination of individuals’ allele diversity and its influence on physiological processes. Of particular interest would be investigation of a possible relationship between the level of basal metabolic rate and individual’s heterozygosity both in general term as well as heterozygosity of specific locus. Such suggestion seems to be supported by previous studies which indicate that the variation in energy expenditure at rest is determined by substantial genetic component (Bouchard et al., 1989; Bouchard and Tremblay, 1990) and heterogeneity of gene loci (Jacobson et al., 2006; Loos et al., 2007). More studies in this regard may be crucial for a better and profound understanding of the Homo sapiens metabolism and energy budget.

Because of the advances in genomics, as well as the proliferation of social science data sets (thanks to corporations and government) I hope that we can begin breaking out of the habit of being led about by the nose by our norms in more areas of human genetics than just the study of Mendelian diseases! That’s a hope. I’m not saying I’d bet money on it.

Citation: Sławomir Kozieł, Dariusz P. Danel, & Monika Zaręba (2011). Isolation by distance between spouses and its effect on children’s growth in height American journal of physical anthropology : 10.1002/ajpa.21482

Image Credit: Caroline Bonarde Ucci.

  • Domino

    “Modern roads resulted in a radical drop in inbreeding in mountainous regions of the country. Some researchers have argued that this shift resulted in an increased level of height, intelligence, and health, among European populations”
    That seems dubious. One of the tallest peoples of the world live in the Dinaric Alps. Not the most accessible place.

    And Africans, who are perhaps the same height as Euros, are not noted for their road-building skills. Some Sudanese groups have average heights exceeding 6 ft, and I can see on your map that Sudan is one of the hot spots for consanguinity.

  • AG

    Well, this give males incentive to mate with exotic foreign women.

  • Alex

    @Domino:
    I agree with your skepticism. There is a lot that needs to be controlled for to make such a claim. However, on your specific point, you could posit the explanation that in this case extreme drift or positive selection for added height explains the tallness of people in the Alps. In very isolated populations height could wander from the human average in either direction, and in some cases it could end up with people being taller. Although I can’t think why, people in mountainous regions can be much taller than the human average (whatever that average is), this is seen in the Ethiopian plateau I believe. So maybe there is some selective advantage to this. Anyone have a thought on why there might be selection for tallness in mountainous regions?

    For the Africans appearing normal height, although there is much consanguinity there is actually much more variation in the genome of Africans compared to the average non-African, as non-African genetic diversity only represents a subset of that outside of Africa. I haven’t thought this all through very well and would appreciate further clarification from others, but I suppose that a high amount of consanguinity in a population with high genetic diversity could lead to more genetic variation in individuals than a population that has low consanguinity but has little variation in both population anyway, so even population admixture doesn’t introduce much variation. You have to consider the variation that exists between the populations as well as the amount of admixture between them. If two neighbouring populations share all the same genetic diversity, then it doesn’t matter if you marry someone from ‘your’ or ‘their’ population, the labels are meaningless.

  • Grey

    “In pre-modern societies individuals tended to marry those close to them geographically. Even if cousin marriage was not normally practiced, over time clusters of villages would form networks of de facto consanguinity.”

    I’m wondering if changes from the extreme ends of the endogamy-exogamy spectrum could be very dramatic. If the human norm, through geography, poor transport etc, was very high consanquinity then the first few generations of greater exogamy might show very large drops with continuing but diminishing returns on later generations.

    It could potentially work the other way too, for example take an initially relatively exogamous population in a steel town which closes down and the people there decay into an underclass. One aspect of the welfare underclass is the collapse of marriage and monogamy with women having children from multiple fathers and no-one really knowing who they’re related too, or rather it becoming too complicated to calculate except in the closest degrees.

    Effectively after a few generations like this (and generations in these kind of contexts are closer to 15 years than 25) everyone living within each segment of those kind of closed environment is having children with (at least) cousins. People from that kind of background do seem to be getting shorter but i put it down to nutrition.

    So i wonder if the percentage change in consanquinity when first moving from either of the two extremes could be dramatic.

    .
    “Height ~ Genetic endowment + Environmental contingencies – Incest decrement”

    Removing a decrement by greater exogamy sounds more plausible on the face, especially if some of those trial populations in Poland were from once very remote and rural populations.

    If it’s true and if the above point re dramatic percentage changes in the first few generations is also true then given some of the dramatic changes post-WWII there ought to be examples of this all over the place (outside the original industrial nations) particularly among what you might call the new elite i.e. not the old aristocracy but the newly emergent larger middle class in places like India and China. If the above is true they ought to be getting taller with possibly larger jumps among the first few generations – although difficult to separate this out from nutrition.

    .
    There’s a tie-in here with the idea of aristocratic in-breeding. Although the peasants existed in much greater numbers in total each village’s marrying pool may have been quite small as a product of an inability to travel. So although the aristocracy had smaller total numbers their marrying pool might actually have been much larger.

    .
    Another thought that comes to mind is a possible connection to the Flynn effect.

    .
    “That seems dubious. One of the tallest peoples of the world live in the Dinaric Alps. Not the most accessible place.”

    If this is a general rule, and maybe it isn’t, then one possible cause for exceptions would be peoples who had developed in one environment getting pushed into mountains by invaders.

    Another possibility might be particular diets outweighing any in-breeding effect i.e. a diet unusually high in protein. Maybe the Dinaric Alpine people live exclusively on milk?

    .
    “And Africans, who are perhaps the same height as Euros”

    Are they?

    http://en.wikipedia.org/wiki/Human_height

  • Alex

    Don’t forget that we often underestimate the amount of admixture that has been occurring even before the modern concept of ‘globalisation’. Human populations have probably always been mixing genetically, through trade, warfare and all other means under the sun. With the relative lack of genetic diversity in humans compared to many other species, including our sister species the chimpanzee, due to our recent origin and low mutation rate, it makes me quite skeptical about the importance of admixture for causing phenotypic differences. Although we are a global species we have less genetic diversity than many species that live within much more confined areas, like a rainforest, or a lake.

  • http://blogs.discovermagazine.com/gnxp Razib Khan

    ok, a few points

    That seems dubious. One of the tallest peoples of the world live in the Dinaric Alps. Not the most accessible place.

    this is a single variable model. it’s just stupid to conceive of it as so. that was kind of the point of my post, you have be really careful and not turn the model into a ‘toy’. so you’re comparing two different situations. what you need to do is compare the same region over time. there are probably some genetic potential differences in height across populations. this seems clear for the pgymies of central africa from the most recent genetic work.

    And Africans, who are perhaps the same height as Euros, are not noted for their road-building skills. Some Sudanese groups have average heights exceeding 6 ft, and I can see on your map that Sudan is one of the hot spots for consanguinity.

    you should work on your map-reading skills. or improve your knowledge base. sudan is not culturally homogeneous, as someone who follows the news might be cognizant of. even ignoring point #1 (there are other variables than just inbreeding), the northern sudanese muslims will probably have higher cousin marriage rates because of the norms among arabs and aspirant arabs.

    For the Africans appearing normal height, although there is much consanguinity there is actually much more variation in the genome of Africans compared to the average non-African, as non-African genetic diversity only represents a subset of that outside of Africa. I haven’t thought this all through very well and would appreciate further clarification from others, but I suppose that a high amount of consanguinity in a population with high genetic diversity could lead to more genetic variation in individuals than a population that has low consanguinity but has little variation in both population anyway, so even population admixture doesn’t introduce much variation. You have to consider the variation that exists between the populations as well as the amount of admixture between them. If two neighbouring populations share all the same genetic diversity, then it doesn’t matter if you marry someone from ‘your’ or ‘their’ population, the labels are meaningless.

    i’d need to look at the numbers for this, but i don’t think the high variance within african matters so much for inbreeding. remember, inbreeding is a problem because near kin carry many of the same negative recessive alleles. it doesn’t matter matter if you have a more varied set or not. there may be some dampening effect though if the population has higher heterozygosity, period, but i wouldn’t bet on it.

  • Alex

    I wonder how much genome wide-levels of heterozygosity vary across humans. As you mention Razib, potentially important for this article. It’s effectively a measure of how inbred human populations already are, isn’t it?

  • http://www.isteve.blogspot.com Steve Sailer

    Dan Seligman suggested in Fortune in the 1970s or 1980s that this can help explain why Irish-Americans used to be so much taller than Italian-Americans, but this is becoming less true. Ireland’s a pretty flat country and Irish-Americans didn’t immigrate as villages, the way some Italian hill villages got transplanted to American urban neighborhoods.

  • Sameet

    Very good article. But in one of the paragraphs, you mention Pakistan and Saudi-Arabia. I can say that it might be more than that. In many south-Indian communities first-cousin marriages are a norm. In many religions, first cousin marriages are almost a ‘requirement’. So though the argument makes sense in terms of genetics, it may or may not hold true when it comes to nature.

NEW ON DISCOVER
OPEN
CITIZEN SCIENCE
ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Gene Expression

This blog is about evolution, genetics, genomics and their interstices. Please beware that comments are aggressively moderated. Uncivil or churlish comments will likely get you banned immediately, so make any contribution count!

About Razib Khan

I have degrees in biology and biochemistry, a passion for genetics, history, and philosophy, and shrimp is my favorite food. In relation to nationality I'm a American Northwesterner, in politics I'm a reactionary, and as for religion I have none (I'm an atheist). If you want to know more, see the links at http://www.razib.com

ADVERTISEMENT

See More

ADVERTISEMENT

RSS Razib’s Pinboard

Edifying books

Collapse bottom bar
+

Login to your Account

X
E-mail address:
Password:
Remember me
Forgot your password?
No problem. Click here to have it e-mailed to you.

Not Registered Yet?

Register now for FREE. Registration only takes a few minutes to complete. Register now »