A Blog by Ed Yong

A life in the trees is a longer one

An assortment of tree-living mammals

In The Descent of Man, Darwin talked about the benefits of life among the treetops, citing the “power of quickly climbing trees, so as to escape from enemies”. Around 140 years later, these benefits have been confirmed by Milena Shattuck and Scott Williams from the University of Illinois.

By looking at 776 species of mammals, they have found that on average, tree-dwellers live longer than their similarly sized land-lubbing counterparts. Animals that spend only part of their time in trees have lifespans that either lie somewhere between the two extremes or cluster at one end. The pattern holds even when you focus on one group of mammals – the squirrels. At a given body size, squirrels that scamper across branches, like the familiar greys, tend to live longer than those that burrow underground, like prairie dogs.

These results are a good fit for what we already know about the lives of fliers and gliders. If living in the trees delays the arrival of death, taking to the air should really allow lifespans to really take flight. And so it does. Flight gives bats and birds an effective way of escaping danger, and they have notably longer lives than other warm-blooded animals of the same size. Even gliding mammals too tend to live longer than their grounded peers.

These trends make sense when you think about ageing in the light of evolution. Imagine a creature that never ages, say, a Tolkienesque elf. Even this potential immortal could succumb to a predator or a disease or an accident. The more time passes, the greater the odds that one of these external calamities will claim its otherwise never-ending life. This creature will therefore have a “statistical lifespan”, an age by which it will probably have been killed.

Now think about the genes of this hypothetical immortal. A mutated gene that harms the creature early on in life, when it’s still having sex and bearing young, has little chance of being passed onto the next generation. Such mutations will be weeded out by natural selection. However, mutations that harm individuals after their statistical lifespan are a different story because their carriers will probably have been killed before they experience any ill effects. Accumulating under the radar of natural selection, these late-acting genes are the ones that contribute to ageing.

There are many evolutionary explanations of ageing but almost all share this basic concept. And note that a central part of this concept is the threat of dying from predators, diseases and so on. If such threats are minimised, then an animal’s statistical lifespan increases, more late-acting genes are exposed to natural selection and more will be weeded out. Based on this theory, you’d expect that species that can escape external threats, whether by flying off or by hiding in the branches of trees, can evolve to age more slowly and live longer. And that, of course, is exactly what Shattuck and Williams have found.

In fact, they showed that the link between long life and a treetop existence holds true for all but two groups of mammals. The first are the marsupials, including kangaroos, koalas and wombats, and the second are the euarchontans, a group that includes treeshrews, flying lemurs and primates, like monkeys, apes and ourselves. At first, this seems odd, for some scientists had thought that a life in the trees could actually explain why primates have such relatively long lives. But to Shattuck and Williams, these are the exceptions that prove the rule.

Both marsupials and euarchontans have experienced a long and persistent evolutionary history in the treetops. For other mammal groups, the occasional member may have evolved to negotiate trunks and branches, but Shattuck and Williams say that the marsupials and euarchontans started off in the trees. Their members have repeatedly descended back to the ground. For the euarchontans in particular, this long tree-based history may have meant that all primates are long-lived for their size.

Of course, if the species that returned to terra firma faced a greater threat from predators, evolutionary theory would predict that their lifespans would eventually contract. But ground-living primates have other defences. They’re typically large and they live in big social groups (larger ones that tree-dwelling primates). These qualities may have provided them with the protection they needed to retain the relatively long lives that first evolved in the branches. For the moment, this is just a hypothesis, and it will need to be tested further.

Reference: Shattuck, M., & Williams, S. (2010). Arboreality has allowed for the evolution of increased longevity in mammals Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.0911439107

More on life histories:

Twitter.jpg Facebook.jpg Feed.jpg Book.jpg

12 thoughts on “A life in the trees is a longer one

  1. very cool! i’ve heard of the lengthy lifespans of fliers, but not of tree-climbers. and interesting hypothesis about marsupials and the euarchonta. nice.

  2. The African naked mole rat may be a supporting case for this hypothesis. Living underground and being eusocial they have incredibly long lives for a mammal their size. According to Wikipedia the naked mole rat is the longest lived rodent (reaching up to 28 years) compared to the African Pygmy Mouse which squeaks by at 2 yrs.

  3. I can only read the abstract – is it typical longevity records or maximum longevity records being discussed? If it’s the former, the issue may be niche selection rather than aging. It makes sense that a species that chooses a defense-maximizing niche, like living in trees, flying, being cryptic, or developing body armor, will live longer than another species that occupies a different kind of evolutionary niche, like maximizing growth rates or litter size.

  4. If it’s the natural log, that point at the top of the graph is something that lives 24 years, and masses 120 kg or pounds; a chimpanzee?

  5. I asked Scott.

    That’s humans! You can see the same datapoint in Fig 4. The numbers are 60.2 kg and 1469 months, or 122.4 years. We’re the largest residual in this comparison. There are a few primates that are not far behind us though – Cebus, Hylobates, and Leontopithecus, for example. Hope that helps.

    Cebus = capuchins; Hylobates = gibbons; Leontopithecus = lion tamarins.

  6. Humans was my first guess, but I couldn’t get the units to work out. (I can’t get the PDF.) The ordinate, then, is log base 10 of grams, and the abscissa is log base 10 of months.
    I notice that the terrestrials appear to group along lines paralleling the main regression, below it. Do they share a common, particularly risky, environment? I suppose particular risks of juvenile mortality would throw off the trends.

  7. I’m not an expert on this topic but I happen to recall that the evolution of senescence is a theoretically slippery issue. I’m not doubting that there is a relationship between life in the trees and longevity, but demonstrating this in the context of the evolution of senescence requires something more than thinking about extrinsic mortality (which the authors refer to in the abstract and text), it requires a consideration of age-dependent extrinsic mortality. GC Williams main prediction that “extra morality” will cause senescence to evolve more quickly is imprecise (as shown most recently by Hal Caswell in a 2007 letter in Trends Ecol Evol), since the selection gradients acting on age-specific mortality (which shape the pattern of senescence) are unaffected by extrinsic mortality that is not age-specific. While the strength of selection decreases with age, the question is whether additional extrinsic mortality will cause the strength of selection to decrease even quicker (i.e., resulting in increased senescence)–something which Caswell (and Peter Abrams 1993 in Evolution) showed to be wrong. More to the point, increased “extrinsic mortality” across all age classes doesn’t influence the evolution of senescence in a straightforward manner, but age-dependent extrinsic mortality does. In the future, it would be interesting to look at how mortality rates change with age, and how these age-specific rates are correlated with arboreality.

Leave a Reply

Your email address will not be published. Required fields are marked *