Why do we still vary?

By Razib Khan | October 19, 2011 1:33 am

I notice that last summer Karl Smith asked “Why Are There Short People?” His logic is pretty good, except for the fact that the fitness variation seems to be much starker in males than females (there is some evidence I’ve seen that shorter women can be more fertile, though that’s balanced by the fact that larger women seem to be able to manage gestation better). In any case, height seems to be a fitness enhancing trait which is highly heritable, and yet the variation in height remains!


Karl’s readers offered some reasons. What do you think? Mind you, something which immediately comes to mind is that the logic presented for why everyone should be tall and vary only a touch is logic. Not all the assumptions need to hold. For example, has the advantage to height been invariant at all times and places? I have posited for example that the fact that humans became smaller after the Ice Age may have something to do with increased morbidity and declining mortality, where agricultural settlements “hugged” the Malthusian boundary more consistently than hunter-gatherers. In this sort of environment smaller individuals may have gained a fitness advantage because they required fewer resources to make it through the inevitable “starving times.”*

But more generally the fact that height seems widely distributed across the genome brings to mind the role that pleitropy might play. Genes of small effect in height may have larger effect in many other traits which are constrained from shifting in allele frequency too much. In other words, can it be that the G-matrix is somehow maintaining this quantitative trait? I’d be curious what those of you with a quantitative genetic background might have to say….

* Another obvious issue is that larger individuals may be more favored energenetically in cold environments. As the world warmed that constraint was released.

MORE ABOUT: Genetics, Variation

Comments (38)

  1. Peter Ellis

    For example, has the advantage to height been invariant at all times and places?
    In a similar vein, has height always been as heritable as it is today? I suspect that for much of our evolutionary history, height was more dependent on nutrition during development than on genetics. In times of plenty, we’re uncovering variation that may not have previously been visible to selection.

  2. Eurologist

    You may want to ponder the fact that today, in Germany, men with the surname Schneider (Taylor) are significantly shorter and weigh less than men with the name Schmidt (or similar; Smith).

    There were niches for everyone.

  3. miko

    1. Pleiotropy.
    2. Variation in selection in time/space.
    3. I’m assuming height is much less heritable when nutrition is limiting. The current hypothesized sexual selection for tall males may be based in choosing better-fed males in our poorly-fed past…it is only in our hyper-fed present that the genetic variation for height is uncovered and could potentially be selected for. However, height heritability still varies greatly among different populations.
    4. While on average women state a preference for taller men, I think the evidence that taller men actually have more children is limited. I could not find any peer-reviewed article on this (I tried for about 5 minutes)… there is one study mentioned in the magazine New Scientist based on a small sample of men from the US Navy, but I couldn’t find a real paper.
    5. Have there been any studies on whether short men who develop disproportionately large biceps gain back some fitness advantage? Something I often wonder in Massachusetts.

  4. 4runner

    Recollections of the tall girl wallflowers at my eighth grade dance pop into mind.

  5. JL

    Shorter people need less food. I remember reading that Holocaust survivors tended to be small.

  6. miko

    Found the paper: Mueller & Mazure “Evidence of Unconstrained Directional Selection for Male Tallness” Behavioral Ecology and Sociobiology, Vol. 50, No. 4

    They claim that tall men have more children because they are more likely to have second families. Oddly, they found no link between height and socioeconomic success in this study, which should raise eyebrows. However, it is a military sample and therefore not typical.

  7. Bean Soup

    Even with evolutionary pressure pushing upwards. Until height hit a brickwall where-after increased height had a negative influence on fitness- there will always be some people shorter than others.

    If generation 1 had an average height of 5ft.
    Generation x had an average height of 6ft
    Generation 2*x had an average height of 7ft

    Whilst the average height is still going up through evolutionary pressure (selection for people with taller genes- and mutations for taller genes)- there would still be people shorter than others. The bottom 5% may average 4ft in generation 1 and 5ft in generation x.

    Unless there was some really strong pressure on the shorter people (and there isn’t) there will always be some shorter than others.

    Of course- when the limit of “perfect height” is reached- slowly the number of shorter people would be removed from the population.

  8. Markk

    I would definitely challenge the “height is advantageous” issue directly. For example there is a popular model that one of the reasons that long distance running is dominated by east Africans is that they are smaller with a smaller heat load and better heat loss characteristics over longer distances. Check some old “Science of Sport” blog entries for discussions.

    Being able to run farther faster has been touted as a unique human trait for decades in the popular press, so the fact that smaller individuals have an advantage here makes the premise very hard to take very seriously.

  9. Bob Dole

    “Genetics of human height” (2011)

    “Another limitation of (earlier) DNA chip genotyping is their focus on SNPs as the most important form of genetic variation. While this may be largely valid, there is accumulating evidence that other variation can affect traits. For example copy number variants (CNV) are a class of DNA size polymorphisms where stretches of DNA sequence – 10s to 100s base pairs in length – are relatively duplicated or deleted between individuals. These have been shown to influence risk of schizophrenia and autism ( [Sebat et al., 2007] and [Stefansson et al., 2008] ) but their role in height remains to be investigated as many of the SNP chips used were not geared toward detecting CNVs.”


    “Genetic risk sum score comprised of common polygenic variation is associated with body mass index” (2011)

    “There is also evidence to support the influence of copy number variation (CNV) on BMI. In the Willer et al. (2009) meta-analysis, when examining CNV by SNP-CNV linkage disequilibrium, they found 10-kb and 45-kb deletion polymorphisms upstream of NEGR1 with the 45-kb deletion flanked by their two most associated BMI SNPs. The recent advent of SNP arrays designed for CNV detection may reveal additional genetic associations with BMI. Epigenetic variation, although more widely researched in syndromic obesity such as Prader–Willi, may also be linked to common obesity. Finally, G×G interactions have yet to be included in risk prediction of body composition. Twin studies support the role of non-additive genetic effects although most study designs have limited ability to detect them (Maes et al. 1997; Flint and Mackay 2009).”


  10. I asked my undergraduate men the other day how many of them choose women based on height. I had a handful out of around 100.

    When I asked the women the same question, roughly two-thirds (out of 150) said they choose based on height.

  11. Karl Zimmerman

    I can think of a number of potential disadvantages to being tall.

    1. There is the obvious requiring of more resources, which you mentioned.

    2. Although the jury seems out, extreme height seems to cause health problems, and moderately tall height (above 6 foot 2 for men) arguably could cause increased morbidity. Most of the issues would likely crop up late in life and not really change reproductive fitness. But I can think of a few which would. For example, a taller person will always hit the ground when they trip with more force than a shorter person, meaning there should be a slightly elevated risk of a taller person dying in accidents, even in societies where people tend not to live very long.

    3. From what I have read, newer studies suggest there isn’t as strong a correlation between height and perceived attractiveness for men as was thought. Women like men taller than themselves, but at the extreme end, legs tend to be out of proportion with torsos, which isn’t seen as attractive. Certainly, height doesn’t seem to be sexually selected for in women, however. IIRC, slightly-below-average women tend to have the most luck. This makes some sense, if you assume women themselves engage in the sexual selection. The very short will often have complicating problems lowering fertility. But merely short women have a very wide range of men who are taller than themselves to choose from. In contrast, women who are taller (especially taller than the average man) probably have more limited choices in partners, be it due to their own pickiness or men passing them over due to perceived inadequacies.

    Of course, the last hypothesis is very western-centric, and probably doesn’t have any bearing in societies with arranged marriages at all. If modern western sexual selection for male height is merely culture, and recent, than height is solely selected for by reproductive and overall physical fitness.

  12. Jason Malloy

    Taller men have higher fitness, but Recent papers from at least two different developed nations have reported that the selection pressure on height is in the direction of shortness for women.

    Daniel Nettle also reported some evidence of stabilizing selection in male height with fitness hits for both very tall and shorter men; both are associated with certain kinds of health problems.

    Anyway stabilizing selection, balancing selection, frequency-dependent selection, etc are usually fairly serviceable explanations for persistent variation.

  13. Chris T

    John Hawks – I wonder what the revealed preference of both genders is in your class?

  14. Jason Malloy

    Here’s another paper with a Finnish sample tying tradeoffs of female height into a life history model:

    “These findings among Finnish women are thus compatible with tradeoffs between reproduction and growth, by showing a compromised adult height at the cost of early age at menarche and first birth. However, in these women, natural selection favored those women who traded their stature for young motherhood.”

    This fits in with other data; there has been selection for both shorter height and earlier motherhood in modern females.

  15. Cathy

    When women say they select based partially on height, there are limitations. I’m rather short so I’d be intimidated by anyone who was seven feet tall. The man I married is pretty much at my upper limit of non-intimidation, just about a foot taller than I am. That really needs to be taken into account into such surveys. And as a general rule, men may be similarly intimiated by women who are significantly taller than they are.

  16. Jason Malloy

    RE: gender preferences for height. I’ve linked to this study a lot in the past for a number of revealed preferences (using actual contact data from an Internet dating site, not survey preferences):

    Height matters for both men and women, but mostly in opposite directions. Women like tall men (Figure 5.4). Men in the 6’3 – 6’4 range, for example, receive 65% more first-contact e-mails than men in the 5’7 – 5’8 range. In contrast, the ideal height for women is in the 5’3 – 5’8 range, while taller women experience increasingly worse outcomes. For example, the average 6’3 tall woman receives 42% fewer e-mails than a woman who is 5’5 …

    Regarding height, we find that men typically avoid tall women. The probit estimates strongly indicate that this is a relative effect, such that men do not want to meet taller women than themselves. According to the logit estimates, on the other hand, men generally prefer shorter to taller women, irrespective of their own height. Women’s preferences over height are the exact opposite of men’s preferences. According to the probit estimates, women have a strong aversion to men who are shorter than themselves, while the logit estimates imply that regardless of their own height, women prefer to meet tall men. [Heh heh – JM]

    According to table 5.5 the male height advantage levels off at about 6’6, but doesn’t reverse even up to 6’10.

  17. Environmental factors have swamped genetic factors in height for the last century or so, at least, if not longer.

    Also, to some extent selective advantage is a yes or no question – either height increases survival or reproduction chances, or it doesn’t. Moreover, if there are multiple targets of selection at once, selection on the most important traits will dampen selective effects on all other traits.

    If height provides a slight advantage, while intelligence and metabolism provide huge advantages, and height is independent of intelligence and metabolism, then selection for height will be greatly suppressed (although not entirely eliminated) in favor of selection on intelligence and metabolism, because short people with the right intelligence and metabolism genes will be more likely to survive than tall people without those traits, even if all other things being equal, height is an advantage.

    @16 – very interesting point, especially if men and women have the same genes for height. If short has selective advantage in women and tall has selective advantage in men, then there is a balancing out of the effects at every generation.

  18. Environmental factors have swamped genetic factors in height for the last century or so, at least, if not longer.

    no, for individual variation it is the reverse. environmental factors are now much weaker than they were in the past. you’re talking about a secular rise in median height, which is due to environment….

  19. dcwarrior

    in America, there don’t seem to be large numbers of people being removed from the genome based on genetic factors, and most people survive to have kids, tall, short, whatever. I understand genetic traits can have an advantage, but in today’s US population, wouldn’t you expect there would be a lack of a strong selection pressure in any direction? Isn’t today’s differential genetic success based more on cultural factors, such as willingness to have kids and having kids at a younger age?

  20. ackbark

    It may be an effect of female selection.

    I saw somewhere that while men everywhere largely agree on who is and isn’t an attractive woman, who women find attractive has no apparent general agreement and varies widely.

    It’s as if women are designed specifically to have various interests in mate selection as a way of maintaining genetic variety, and height is a part of this.

    Old, toothless and pus-riddled is about the only thing women everywhere might agree is unattractive, and even then. . .

  21. Laurie

    Height is likely to be influenced by a variety of genes. An individual possessing all the alleles for greater height is likely to be dysfunctionally tall, if not physiologically then reproductively since who wants a partner a foot taller than them? So there would always be some selective pressure working against all the alleles for greater height.
    This hypothesis could be investigated by looking at the data for individuals who leave no offspring. If very tall people are over represented it would support my hypothesis.

  22. Justin Giancola

    16. I find that study rather misleading or short-sighted. 5’3 to 5’8 is in the taller range of average for women that I see – as there are oodles of women in the 5′ to 5’2 range. They give such a smaller number for guys that seems sort lame to me – these are the magic select few perfect men?!; if they’re going to be like that wouldn’t 5’4 to 5’7 be more appealing to a lot of guys? It’s unlikely all those men are even that height anyway as men often lie and wear shoes with a heel of an inch or two. So I’d say men from 6′ to 6’4 would seem more fitting. Both sex groups are the taller range of average – again with oodles of guys here in the 5’8 to 5’10.

    Wondering why the average man doesn’t go after 6′ women is kind of dumb – like the average woman goes after a 7′ guy? (not calling you dumb)

    Also isn’t there the stereotype of men likely the “leggy” chicks? “legs for miles”, models…blah blah, these are all alluding to tall-er. Why are women always wearing shoes to make them look taller?

  23. Charles Nydorf

    “Genes which have small effect in height may have larger effect in other traits which are constrained from shifting in allele frequency too much.” This sounds perfectly reasonable. Will someone please accept Razib’s challenge and formulate this in a quantitative model?

  24. “…why everyone should be tall…”

    Practically everyone is tall. Even pygmies. Look at Ardipithecus or Praeanthropus.

  25. While height may be an advantage in case of competition between individuals, it could be a disadvantage in competition between different kin groups, because the group with smaller members could produce more of them.

    I also like pleiotropy and balanced effects based on gender.

  26. Eurologist

    “I saw somewhere that while men everywhere largely agree on who is and isn’t an attractive woman, who women find attractive has no apparent general agreement and varies widely.

    It’s as if women are designed specifically to have various interests in mate selection as a way of maintaining genetic variety, and height is a part of this.”

    Not only that, but women’s attraction towards men changes during the menstrual cycle, with attraction towards more masculine and athletic types around ovulation, and more “caretaker” types at other times.

  27. “Genes which have small effect in height may have larger effect in other traits which are constrained from shifting in allele frequency too much.” This sounds perfectly reasonable. Will someone please accept Razib’s challenge and formulate this in a quantitative model?”

    For a few-locus system, it could be reasonable for directional selection on stature to be balanced by selection on the pleiotropic effects of the genes that explain the additive variance in stature. Directional selection means that stature and fitness covary; if pleiotropic traits also covary with fitness, the net additive variance in fitness could be zero even with substantial additive variance in stature.

    For stature, the additive variance is spread across too many loci for this model to be plausible. We presently know of more than 300 loci that contribute to the additive variance, and together they account for only a fraction of the additive variance. There must be many, many more, maybe even so many that new or recent mutations comprise an appreciable fraction of the additive variance. There should be plenty of additive variance that can respond to directional selection on height.

  28. One overlooked concept is that sexual selection can actually act to increase genetic variability. So if height is a sexually-selected trait due to female choice, this can lead to increased genetic variation due to selection for modifier loci that have beneficial mutations that hitchhike on loci contributing to the selected trait. I think there are a few models out there (by Pomiankowski and by Petrie) that show how female choice creates genetic variability in the selected trait. Basically, models that incorporate biased-mutation versus those that don’t can change the direction of the variance we expect to see in traits under selection (Hanna Kokko reviews this in her 2006 paper on ‘unifying models of sexual selection’).

    Also, a minor point: directional selection (acting on the mean) can occur independently–and often does occur independently–of selection acting on the variance, since the moderate amounts of phenotypic variation observed in traits under directional selection out in nature (not in fruit flies or corn) is proof-positive that it must.

    In response to John Hawk’s comment, I think Gunter Wagner has already modeled this in terms of morphological evolution in his paper on “the influence of variation and developmental constraints on the rate of multivariate evolution…” (in J. Evol Biol 1988). This is known as the ‘corridor model” where variation is allocated in one direction but constrained in another, thus limiting what variation gets expressed when traits are linked. Something like that.

    wow, this was one pedantic post…sorry.

  29. #28, me like. no apology needed!

  30. I’m a little over 6’4″, and I find being tall pretty useless other than for seeing over other spectators at golf tournaments. It used to be a good clue to how well fed you were as a child, but fortunately that’s becoming less meaningful. Personally, I think tall genes exist to fool other people into thinking you are from a wealthy family. We’re used to people using nurture to try to fool other people about their nature, but I think this is the mirror image. The main advantage to being tall was that other people assumed you must come from a well-to-do family.

  31. Charles Nydorf

    Thank you, John Hawks and Rich Lawler!

  32. twl

    //When I asked the women the same question, roughly two-thirds (out of 150) said they choose based on height.//

    Who amongst you is NOT forced to date sub-standard mates because you can’t attract a tall guy? Put up your hands now!

  33. rob

    Maybe within population variation height works like schizotypy (or whatever we call that now) and intelligence maybe work. There can be lots of variation despite directional selection because genes break faster that selection can remove them. Kinda makes sense, tons of genes influence the trait, no common alleles have a huge effect. Probably the general rule for quantitative traits.

  34. Empirical evidence here that tall men are more likely to be married, even adjusting for age and income etc. :


  35. Mark Houston

    Maybe the question everyone should be asking is – WHY ARE THERE TALL PEOPLE. After all, Christopher Ruff and Tom Samaras would both laugh at the thought that being small is odd.

  36. silylene

    Sometimes increasing height is not always a positive selection. I can think of two examples:
    – In WW1, shorter infantry men had a higher survivability than taller men, with a strong correlation. I read a nice paper on this about 25 yrs ago. The authors speculated that being tall increased one’s risk of dying in trench warfare.
    – Height also corresponds better to leg length than trunk length (there is more variation in leg length than trunk length). Dienekes just wrote a short article on why shorter legs are advantageous to populations who live in mountainous regions. http://dienekes.blogspot.com/2011/10/short-legged-neandertal-mystery-solved.html This fits well with some of my observations also of shorter legged people living in Bolivia and Peru, or even in Greece compared to leggy people from the the flatlands such as Netherlands or the African plains.

  37. Justin Giancola


    No way 😮 … I theorized along these lines years ago! awesome.

    I wonder if they’ll look into the musculature of the calf as well – as well as attributes of pelvic and foot bones.

  38. Tomasz R.

    Height is not necessairly good. Taller heigh means a lot of growth hormone which increases chances for getting cancer. Most long-lived people are not tall.


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


See More


RSS Razib’s Pinboard

Edifying books

Collapse bottom bar