Evolution With Relaxed Natural Selection
In my post “A Very Simple, But Common Mistake,” I argued the following:
If you find a feature in a species that is common or even ubiquitous, one possible reason for this is “natural selection.” I do not use the term “natural selection” here (and below) in the modern tautological sense of having relatively more descendants on average, but in the original sense of a relatively higher chance of survival, basically to and through fertile age. “Sexual selection” is one other reason why a feature might be common in a species or even ubiquitous. I use the term in the narrow sense of a relatively higher chance of finding a mate after surviving to the relevant age.
While these are plausible reasons why a feature may become common or ubiquitous, it is false to conclude that for any such feature they must be the explanation. Skipping from a feature being common to “natural selection” as the cause or else “sexual selection” is a fallacy. If mechanisms A, B, C, D, E, F, etc. can lead to an outcome, you cannot conclude from the outcome that it must have been A or else B.
One other mechanism I described was what I called “random selection.” For purely random reasons some features that do not confer any advantages or disadvantages for natural or sexual selection become more or less common from one generation to the next. Over time this creates a random walk, which can lead to the feature becoming common or even ubiquitous. But it might also become rare or even disappear.
However, there are other possible mechanisms that I will explore in this and further posts. The case I will focus on here is that a feature becomes common because there is no natural selection anymore in some regard. I use the term “feature” in a broad sense, not only for an innovation, something new, but also for a qualitative or quantitative modification of what is already there. My main example is body size. It can go up or down on average, but that does not imply any new structures per se.
If a feature becomes common with no natural selection any longer in some regard, then this is not an adaptation to anything in the environment. That is also so with random selection. So the associated conclusion that any feature that is common has to be an adaptation to something in the environment falls flat. That may be the case, but you cannot conclude it from a feature being just common if there are other possibilities that do not lead to an adaptation. And there are.
— — —
“Foster’s Rule” (named after the biologist J. Bristol Foster) sums up the observation that species on islands may either grow larger than otherwise or smaller. The first is known as “insular gigantism” and the latter as “insular dwarfism.” There are many examples for this, eg. giant swans 2 meters tall and pigmy elephants with a shoulder height of less than 1 meter that both once lived on Sicily. The curious thing here is that some species grow larger and others smaller in such a situation. And it is the species that is usually small that grows, and the species that is usually large that shrinks.
Of course, it is possible that this phenomenon is an adaptation to something in the environment. There are explanations along these lines, eg. that a smaller size is better to cope with less food on an island. However, it does not matter what the total amount of food is, which is surely less on an island than on a continent, but only on how much there is of it per capita. And there is no reason to believe that this has to be different on an island and worse. In addition, there is the awkward contradiction that for some species the same constraint leads to larger body sizes.
One plausible cause for Foster’s rule might be that in territorial species those that grow larger take away the territory from those who do not, and hence are able to reproduce with relatively more success. That would be an explanation via sexual selection. While it is a plausible route, it cuts against the fact that there is also insular dwarfism and not all respective species are territorial in a relevant sense.
Another explanation is that there was a “population bottleneck.” That is similar to random selection or actually a special case. I will explore this in a future post. If the initial population was a selection of specimens from a larger population, they could have been larger or smaller than average for some reason. That might have to do with how they reached the island. However, it is not clear why this would work sometimes in favor of larger and sometimes of smaller sizes with this regularity. Also larger specimens of a large species or smaller specimens of a small species might end up on an island. And then it is also not obvious whether there were any such large or small specimens to begin with. Did pigmy elephants and giant swans live also outside of Sicily and only went there? If not, then much of the explanation must have to do with what happened on the island.
— — —
The most plausible explanation for Foster’s Rule is that it results from no natural selection any longer for body size on an island while there was natural selection on the continent. That can only apply over a certain range because there must be some physical constraints: a general body plan can only be scaled up or down to some extent. For example, miniaturization would also imply shrinking the parts, eg. brains, which would affect the species materially from some point on. And being very big might lead to problems pumping the blood around. At such extremes, there would be natural selection, however only in a rather lame sense that a species cannot go beyond any bound in either direction, not in the sense that it adapts to anything specific in the environment.
One explanation for relaxed natural selection is that perhaps not all species on the continent make it to an island, and so large predators in particular are mostly absent. This is not completely true as there are also giant species of eagles or owls. Still it is conceivable that a species is driven to a small or to a large size ,under the circumstances on the continent because of predation. A small size would make it harder to spot, and a large size to hunt. So if predation lets up, those in a species that are either larger or smaller than usual on the continent do no longer suffer from a relative disadvantage.
However, there are some problems also with this explanation. Suppose in a species on the continent, all else equal, the larger specimens (or the smaller depending on the species) suffer from more attrition until maturity than the smaller (respectively, larger) specimens. If you take that away with no predators any longer, then as a first stab you should only have that they are now are on a par with equal chances of survival. But that means that sizes should spread out, not that they drift in one direction. Those with the original size would not be at a disadvantage.
You could try to fix this if those on an island with a previously unusual size now have an advantage, eg. for natural or sexual selection. However, why would islands quite regularly have an environment that works in favor of one direction for some species, and in the other for others? Or why would finding a mate sometimes work more easily with a larger size and at other times with a smaller size? That would be a strange coincidence. If size were now unimportant, it could also drift in the other direction, with even larger large species and smaller small species. However, as far as I can see that does not happen.
— — —
Despite these objections, I would accept the explanation via no natural selection for body sizes anymore (at least over a range delimited by physical constraints) as essentially correct. But then this means that the result is not an adaptation to anything on the island. That’s my main point here: You can find a feature that is common or even ubiquitous, and it is not the result of natural or sexual selection for it. Those might play a role here, but in principle they don’t have to. You could have equal chances of survival and equal chances of finding a mate independent from body size (at least in some range), and Foster’s Rule could still work out or rather because of it.
I think what you basically see here is what eugenicists call “dysgenics,” ie. that something is no longer selected for that used to be selected for. Any such changes would have to result from mutations that can now remain in the population while they could not do so on the continent. As they accumulate, the mutations change average size for various species. The term “dysgenics” carries negative connotations, but as for size, none are warranted. Not even the direction is fixed, it could go either way. But why in this pattern: Small species grow, large species shrink?
— — —
Growing a specimen of a species is rather simple. Let me explain the main part for body size in humans, namely for the extremities: arms and legs. My medical training was long ago, but I hope I get this roughly right.
Bones may seem like inert matter, but they are living tissues. There are cells here that build the hard stuff, which is mostly salts, and in between collagen. These cells react to how a bone is used. They reinforce it in the right directions, but not where this is not necessary or less so. A bone can in this way grow, but mostly only become thicker.
However, to make it longer, the process works differently. You have a zone of cartilage close to the ends of the long bones where the joints are. On both sides there is bone: the long part and the cap that is part of the joint. The cartilage expands, but the extra length is turned into bone and so the zone remains narrow. That’s where longitudinal growth, not just thickness comes about. Until some point in time, this is how it works, but then the cartilage is completely ossified, and growth grinds to a halt. The length of your arms and legs does not change any longer when you are an adult, only when you are a child or an adolescent.
The whole process is controlled by the growth hormone somatropine, which is produced in the anterior pituary gland that sits at the bottom of your brain. Also other hormones are involved that are produced in the liver: Insulin-like growth factors. Somatropine does not only induce the above process that lengthens the limbs, but als growth in other body parts. For example, the skull has a similar growth zone with the fontanelles. Yet, the legs are a major part of how tall someone is. There is some regular slow growth over time, however, much of what happens comes in growth spurts where humans grow faster for some time.
My guess for why you don’t have only continual growth is that with growth spurts you can have a certain buffer. Suppose a growth spurt works out more slowly than it should be, which might result from insufficient nutrition, parasites or diseases. You can then lengthen the time and still get the same result, at least if there is time left to the next growth spurt. In this way you can reach a target with a certain precision. Building in several such buffers at different times makes also sense because any disturbance will only lead to a limited shortfall. If you suffer from too many problems, it can still be too much, and it then results in “stunted growth.” But to a certain extent you can make up for some problems. If you had only continual growth, there would be no leeway to catch up again. What is lost is lost.
— — —
To target a small size, the tricky part is to stop growth with precision over the different growth spurts. To target a large size, the point is instead to make sure you get at least a certain contribution out of them. There is an asymmetry here: In the first case, there should be mechanisms that focus on not overshooting a maximum for growth spurts, in the second case, mechanisms that focus on not undershooting a minimum. In a small species growing too big is a more serious problem than falling somewhat short. For a large species being too small is a more serious problem than being somewhat too big. There should be a preponderance of mechanisms on one side for a species. Basically, you have an inherent size dictated by the body plan. But for some species body size is pulled downwards, for others upwards.
What then possibly happens on the island is that the relevant mechanisms slowly deteriorate via mutations and work less well, and this should fall more on those mechanisms that the species focused on before. For a small species this means control on when to stop growing becomes laxer, ie. its average size increases. This might not be just one mutation, it could be many that work in the same direction and shoot holes in various relevant mechanisms. And for a large species, forcing growth on becomes less stringent, it hence shrinks in average size, again possibly not only because one mechanism deteriorates, but many.
The crucial point here is that all this would happen even in the absence of natural or sexual selection for body size (at least for some range until you hit hard physical constraints). It is possible that natural or sexual selection make the process faster or drive it further or less far, but it is entirely conceivable that it would work in the same way also without natural or sexual selection. So, in principle it is possible that a feature, either smaller size or larger size, becomes common on an island without any natural or sexual selection for it. In this case, it is also not an adaptation to anything. The shift could occur even with equal chances of survival for larger, respectively smaller, specimens and equal chances of finding a mate. The conclusion that only natural or sexual selection can be behind this development is then false.
Hence, we have now two other candidates that first have to be excluded to narrow the explanation down to only natural or sexual selection: random selection and no natural selection anymore for a feature that used to be selected for. Actually, there are more as I will explain in further posts.
— — —
Foster’s Rule is not the only phenomenon of this kind. There is also deep-sea gigantism, eg. with huge octopuses, where a similar route seems plausible. And there is still another example: Various species of mammals have larger brain sizes in an urban versus a rural environment. The explanation put forward is that this must be natural selection for something, eg. the urban environment is more challenging and so species need larger brains to cope with it. But a different take would be that it is the opposite: relaxed natural selection in an urban environment for body size. It does no longer matter whether a specimen has a somewhat larger brain, and then mutations see to it that it happens if they affect mechanisms more that previously limited growth.
You can also speculate whether something like this may also be behind other phenomena. How about those human populations that were longer under relaxed natural selection because of better modes of production, eg. as herders and agriculturists? Those humans that remained hunter-gatherers until recently are typically rather short by comparison. And being taller with relaxed natural selection might probably also lead to larger brain sizes all by itself without any natural or sexual selection for it.
It would be naive to equate this with average IQ. But those who make this connection also claim that those populations that were longer in such an environment —especially a longer history of agriculture — with relaxed natural selection tend to score higher for IQ as far as I understand it. If it works in this way, it would be wrong, though, to go on a search for how living in an agricultural society led to more natural selection for higher IQ. It would be the other way around, this would happen because of relaxed natural selection, and it would not an adaptation to anything. Correlations with average IQ might be a consequence of and not a cause for earlier development, and conclusions for prosperity built on a causal link could be mistaken if a longer history with agriculture drives both. That may be wrong, but the point would have to be addressed before jumping to conclusions.
Also under the assumption that there is a connection with IQ, the conclusion is especially funny for eugenicists. All human populations would grow taller with relaxed natural selection, especially in modern industrial societies, grow larger brains, and also have rising IQ because of relaxed natural selection. Higher IQ would then be a prime example for “dysgenics,” and the “eugenic nightmare” completely pointless. Put people in an environment with no natural selection for body size any longer apart from physical constraints, and they will grow on average, have larger brains, and a higher IQ.
I make no claim that it is so. However, any eugenicist who laments supposedly declining IQ in modern industrial societies with relaxed natural selection would have to rule this possibility out before making wide-ranging claims. I have not seen a single one address this point. But that would have to be the first thing to clear up: There is “dysgenics,” but in which direction does it work? It could simply be the good luck that we humans are a small species that has leeway on the upside and grows under these circumstances.
In a further post, I will explain another possible mechanism that can speed this process up, even by a lot. Basically, if you have several mutations that are initially only present in different subpopulations and whose effects boost each other when they come together, then you can have very fast directional shifts when these subpopulations mix, in principle over just a few generations or even a single one. This leads to another awkward conclusion for eugenicists who are usually also fans of segregation: Mixing humankind more would be the way to go if you want to raise average IQ.
But enough food for thought for now.
— — — — — — — — — — — — —
PS
This post is loosely related with my series on Thomas Malthus’ “Essay on the Principle of Population,” first published in 1798. You can find an overview of all posts here that I will keep updated: “Synopsis: What’s Wrong with the Malthusian Argument?”
