Paternal Age and the Force of Mortality

Looking into how paternal age affects life history evolution, I took our data on survival curves of Herero males and calculated the strength of selection, that is the derivative of fitness with respect to survival, as a function of mean paternal age.  The computation is just equation 16.3 on page 116 of Alan Rogers’ textbook that I mentioned in a previous post.

The graph shows the strength of selection on mortality by age in two model populations.  In the young fathers version, the birth rate (to males) is uniform from age 20 to 30,  In the old fathers model the birth rate is uniform from age 30 to 50.

Image

No surprises here save one.  While selection for survival should extend male lifespans by 10 to 20 years in the case of old fathers, selection for survival before the age of reproduction is much weaker in the the case of older fathers.  A prediction is that adolescent and young male death rates should be higher in old father societies because selection is weaker.  I never realized that.

Hamilton’s theory does not describe  human life history very well, as Rogers shows in his Figure 16.1 and discusses in the text.  Human female fertility ceases long before the theory predicts that it should and humans live much longer.  The reconciliation certainly has to do with kin selection or indirect selection.  For example Kris Hawkes pushes the “grandmother hypothesis” according to which females cease reproduction and instead work for their daughters’ children.  If she is right this grandmother effect selected for the prolonged human lifespan, and the long lifespan of males is a side-effect of selection for long life in females.

Hamilton’s theory also fails to predict the mortality peak in infancy and early childhood.  Here again indirect selection ought to be brought in.  A frail infant with no prospects “ought” to die early since the death frees up the mother for the next child, an important effect since the human female reproductive span is short and every year counts.

Rogers has worked much of this out, incorporating kin selection into life history theory, in a paper several years ago.  The paper is, to put it mildly, somewhat difficult to read.  Good luck.

About these ads
This entry was posted in Uncategorized. Bookmark the permalink.

21 Responses to Paternal Age and the Force of Mortality

  1. typal says:

    Rodgers P117 “If aging occurs primarily by mutation-selection balance, then the forces affecting aging are weak: selection on these genes is no stronger than mutation, and mutation is a very weak force. Because these directional forces are weak, genetic drift is likely to be important. We should expect great and unpredictable variation in rates of senescence among characters and among populations.”

    Chimps die in wild at 40-50 and are selected for early reproduction. So why can can captive ones sometimes live well into their seventies and still be in good shape ? here

    • harpend says:

      typal quotes Rogers: “If aging occurs primarily by mutation-selection balance, then the forces affecting aging are weak: selection on these genes is no stronger than mutation, and mutation is a very weak force. Because these directional forces are weak, genetic drift is likely to be important. We should expect great and unpredictable variation in rates of senescence among characters and among populations.”

      This seems dated given recent measures of mutation rates: I will ask Alan tomorrow if he still believes that.

      • typal says:

        ” A prediction is that adolescent and young male death rates should be higher in old father societies because selection is weaker”

        That would apply to natural causes presumably. But wouldn’t most young males’ deaths be inaccessibly extrinsic: accidents, homicide & suicide. These are kind of ‘testosterone poisoning causes of death’. (I’m not sure high testosterone doesn’t affect vulnerability to disease)

        In polygyny there is necessarily some kind of selection because some males are going to be excluded from reproduction.Older men in the Gambia talked about in GC’s post have have got something that the men who end up with no woman at all haven’t got.: According to Alvergne et al it’s ‘ “pro-social behavior which reflects sociability, assertiveness, activity, dominance and positive emotions.” ‘. Discussed HERE. Those traits correlate with higher testosterone. So maybe in a polygynous society a higher risk of death when young is a price worth paying for success later in life .

    • rich lawler says:

      My sense is that lots of the mortality is extrinsic in wild chimps–they die because of disease, fights, hunting, etc. Many of the theoretical models, such as mutation-selection balance, ignore (out of convenience) extrinsic mortality and focus on the degree to which animals die “intrinsically” for example due to a lethal, late-acting mutation. Thus I think the discrepancy is due to the extrinsic versus intrinsic mortality, and which source of mortality is taken into account within a particular model

      • rich lawler says:

        Oh, and I obviously meant to also write that captivity is likely to reduce extrinsic mortality quite a bit, since there is a steady food supply, few diseases, reduced aggression, etc.

      • typal says:

        There is significant amount of extrinsic mortality remaining in humans (mainly car accidents before age 30). In fact human mortality in the US actually declines around 30 for that reason. The oldest living person would likely be wearing a nappy and being fed with a spoon. The oldest living chimp has got a few grey hairs. Chimps of both sexes reproduce young and can live a long time after prime reproductive age, if protected from extrinsic factors. So I think the example of chimps tend to support an idea of Michael Rose: populations with an earlier last age of reproduction stop aging earlier and live longer.

      • harpend says:

        That’s not quite right: the P(x) curves in the Hamilton senescence model are just that, descriptions of extrinsic mortality. Change the environment and the P’s change.

  2. typal says:

    Missing link HERE

  3. dantae says:

    This probably reflects my lack of sophistication with the nuts and bolts of understand population genetics, but I am having trouble figuring out how this could be.

    Consider two populations:

    In pop A reproduction uniformly occurs between ages 20-30

    In pop B reproduction uniformly occurs between ages 40-50

    Assume a mutation that gives a 1/10th risk of death at age 10
    and that 50% of individuals in populations A & B have this mutation and that everyone is the same age

    Due to this mutation, by age 11 the mutation will only be present in ~47% of the population

    Then we let the populations reproduce with all living individuals having uniform reproductive success. In the next generation at birth the allele frequencies will be the same in both populations. The rates of decline of the mutation allele frequency will decline at the exact same rate per generation (but not per year).

    Isn’t the force of selection against this mutation the same?

    Can you find a way to explain the phenomenon in a more intuitive way, or is this one of those cases where verbal models like mine are just going to break apart?

    • harpend says:

      We are in the same boat here. What one does with a problem like this is to stare at the math for a long time and turn the understanding of the math into words. Often takes a long time.
      I have another method: I will collar Rogers today and get his version of a verbal explanation of the result.

      • dantae says:

        Any word from Roger?

      • harpend says:

        No, we both have had some health stuff at home and one or the other of us has always been rushing off somewhere when we have met. I will take another shot at collaring him today.

        Must be something like this: you buy a fancy car that you plan to trade in in either (1) two years or (2) five years. How much does a minor ding in year 1 bother you? More in case (1), less in case (2) since you will surely have more after five years anyway. Something like that, but I am merely blowing smoke.

  4. typal says:

    A significant subset of Sumatran Orangutan males don’t get to mate until they turn 30 years old and bulk up: HERE. The Borneo subspecies doesn’t do this. From what I can gather it seems the very oldest orangutans are Sumatran. However the world record oldest orangutan was a male. Now if it is males who are selected for fighting the force of mortality in a human population, shouldn’t we expect that in such a population very long lived men would be overrepresented, (even if men don’t live longer than women). The oldest lived man was Christian Mortensen at 115 . He was Danish . The oldest black man ever black man ever was Johnson Parks who reached age 113. Black men don’t seem to be long lived, relative to black women.

  5. Greying Wanderer says:

    “For example Kris Hawkes pushes the “grandmother hypothesis” according to which females cease reproduction and instead work for their daughters’ children. If she is right this grandmother effect selected for the prolonged human lifespan, and the long lifespan of males is a side-effect of selection for long life in females.”

    This would fit my observed experience. It’s almost as if humans are designed to have two prior generations working to ensure the survival of the next. Thinking about it creatures with young who take a very long time to mature don’t get to have as much practise at parenthood as creatures who can raise a litter every year and so have more scope for mistakes before maturity so i wonder – do elephants or similarly long-to-mature critters have grandmother elephants?

    Thinking aloud.

    Might provisioning come into this also? Humans in chimp-mode where the females can feed themselves and their offspring and the males fight to impregnate but otherwise don’t bother with the resulting offspring can’t work (it seems to me) except in environments where the females can provision themselves. I was thinking any move out of environments like that would require the males to start providing for their young and be the spark for traits like pair-bonding and a lot of ancillary traits like if the males were going to be around the young a lot more then traits that reduce aggression around the family unit

    http://www.guardian.co.uk/science/2011/jan/06/sniffing-women-tears-sexual-desire

    (i’m assuming here that children’s tears have the same effect but no-one’s checked – could be wrong.)

    However could the grandmother effect be an alternative to this at least at first? i.e. human group moving into an environment where mothers need help feeding their young, two solutions: males gradually evolve in a provider direction and/or grandmothers evolve to stay alive longer – or more accurately longer lived grandmothers confer a reproductive advantage on their daughters and daughters in law and pass on the longevity? Trying to imagine both i can see both happening over time but the longer-lived grandmother change being potentially much quicker and easier than the male provider effect?

    • rich lawler says:

      I just saw a paper published in Science showing that the ‘grandmother effect’ also seems to work in killer whales. That is, for a long time, folks knew that some cetaceans had a long post-reproductive lifespan, but this paper demonstrates that offspring with a surviving grandmother have a much better survival probability. Something like that.

      Another thing that characterizes humans (but not other primates) is the tendency to have overlapping offspring, such that elder siblings can care for younger ones, thereby freeing-up parents (or grandma) to care for the littlest, most helpless offspring.

      • Greying Wanderer says:

        I’m imagining a sequence:
        - grandmother effect
        – female longevity
        — male longevity as a side effect
        —- older males now have an opportunity to compete not just for the females of their generation but the next as well

  6. typal says:

    Chimps have menopause, and they (females at least ) seem to live far longer than orang-outangs which don’t have menopause. That’s like humans – the 9 oldest people everare female Females stop reproduction early so they stop aging earlier?

  7. typal says:

    Men are twice as likely to die as women of same age, from any cause. Here start at 0:01:18.

    There seems to be a cross over in testosterone levels “African Americans have a clear testosterone advantage over Euro-Americans only from puberty to about 24 years of age. This advantage then shrinks and eventually disappears at some point during the 30s (1). The pattern then seems to reverse at older ages (Ellis & Nyborg 1992; Gapstur et al. 2002; Nyborg 1994, pp. 111-113; Ross et al. 1986; Ross et al. 1992; Tsai et al. 2006; Winters et al. 2001).” Here

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s