by Rik Smits
Geschwind spotted a clue to the causes of left-handedness in two phenomena. The first was the fact that left-handedness occurs slightly less often in girls than in boys and the second was the strange and intriguing but fairly firmly established connection between left-handedness and autoimmune disease. To Geschwind the central question concerned the connection between these two facts. In some way or other, left-handedness had to do with the distribution of functions between the two halves of the brain. How autoimmune disease could be linked to that pattern of distribution was a complete mystery.
Geschwind began by tackling the simple question raised by all these mysterious findings: what is the main difference between girls and boys in their intrauterine development? The answer he found was testosterone, the male sex hormone. Boy embryos pretty much swim in it, producing it themselves, in contrast to girl embryos.
Boys are generally more susceptible to disorders of the immune system, so the next question was whether it might be possible to make a connection between autoimmune disease and testosterone. There too Geschwind succeeded. High doses of testosterone have an inhibiting effect on the thymus gland, which carries out important tasks that affect the development and precise regulation of the immune system. The next step was to make a connection between testosterone and the brain, where left-handedness originates. Here again the resourceful Geschwind came up trumps. It was already known that the left side of the brain starts to develop slightly earlier than the right, and now it turned out that high concentrations of testosterone reduce the speed at which the left brain develops, precisely in the period when that half is particularly vulnerable. This phase therefore lasts longer, which creates a dual risk: the left brain can more easily be damaged and there’s a chance it will be overtaken and outflanked by the right brain, which begins to develop slightly later. The result, he concluded, would be an increased risk of disorders such as dyslexia – and of left-handedness.
His theory seemed tidy and watertight, but one question remained: where do left-handed girls come from? Although girl embryos don’t make their own testosterone, girls are almost as likely to be left-handed as boys. And girls too suffer from the ailments that Geschwind linked to a disruption of the testosterone level in boys, if slightly less commonly. He found a solution in the fact that mothers also produce some testosterone as a by-product of all kinds of processes. This meant girl embryos could be exposed to a testosterone level that was higher than normal for them. So it was not so much a matter of exposure to testosterone as such, which differs a good deal between girls and boys, but of relatively high concentrations.
A cleverly constructed story, and there’s no doubt a good deal of truth in it. It’s also unique in that it’s the only theory to make a credible connection between the slightly higher incidence of left-handedness in boys and their slightly greater susceptibility to all kinds of autoimmune diseases, allergies and so forth. But it can’t be the whole story, because again it breaks down when it comes to monozygotic twins. How can we explain that when a monozygotic twin is left-handed, the other twin usually isn’t?
It might be possible to explain this in the case of boys. Boy embryos, after all, produce their own testosterone, and two out of three pairs of twins share the chorion or outer embryonic membrane but not the amnion, the inner embryonic membrane. As long as the excess of testosterone produced by one of them stays within that inner membrane, the other will be unaffected. But why would one of a pair of identical twins overproduce testosterone and the other not? It can’t be a result of genetic factors, since those are exactly the same in both cases. But then, so is the environment, because influences from outside the mother ought to affect both twins equally.
Geschwind has even more difficulty with twin girls. The testosterone that he suggests may be a cause of the various abnormalities in girls always originates in the mother’s body. How two genetically identical unborn females could respond differently to it is a mystery.
31
How Even Detrimental Characteristics
Can Survive
So that’s the position we find ourselves in after a century of research. We’re left pretty much empty-handed, since none of the three best explanations for the existence of left- and right-handedness is satisfactory.
General trauma theories have a long list of shortcomings, of which the most important is that they require us to assume that in newborns the world over, the incidence of brain damage is horrendously high, although we barely notice it in the course of people’s lives. The 2009 discovery that differences between left- and right-handers in the arrangement of features in the two sides of their brains can be far greater and more general than was once thought makes it harder than ever to assume that some kind of trauma lies behind all left-handedness.
Geschwind gets us a long way with his testosterone theory, but it doesn’t solve the twins problem. He’s therefore unable to explain the existence of left-handedness as such, although the suggestion that testosterone is a driving force behind pathological switches of hand preference seems perfectly sensible.
So we’re left with genetic explanations of the kind proposed by Marian Annett. When she presented her right shift factor it was no more than an abstract idea. Genetics has moved on since then and her factor has been fleshed out with some appealing biology. For example, it has been proposed that the shift is caused by a recessive gene variant for left-handedness on the x chromosome that has low penetration. The reason why it seems logical to expect to find the gene on the x chromosome is that it enables us to explain the slightly higher incidence of left-handedness in boys. Males have only one x chromosome, so its alleles are always dominant – there’s no copy of any of its genes to stand in the way. In boys the recessive left-handedness allele could therefore take full effect, whereas in girls it would need to be present on both x chromosomes.
The concept of penetration is brought into play at this point because the effect of dominance as a result of the lack of a second x chromosome is far too strong. Boys are only slightly more likely than girls to be left-handed. Penetration is a refinement of the old Mendelian division into dominant and recessive gene variants. Over time it has turned out that dominant alleles are not always expressed in individuals in whom they are present. This is the case with polydactyly, for example, a condition in which a person has one or more fingers or toes too many. Among its causes is the possession of the dominant allele of a single gene, yet only two out of three carriers of that variant actually have additional digits. The allele is therefore said to have a penetration of 65 per cent. The main result of incomplete penetration is that dominant characteristics, like their recessive counterparts, may skip generations before suddenly re-emerging.
Penetration describes just how free a genetic variant is to exert its influence, but until it becomes clear how the brakes are applied this explains nothing. And since no one has determined conclusively which gene causes hand preference, such a theoretical allele is merely ‘in the frame’, to use police terminology – no more than a possible suspect against which there is as yet insufficient evidence to merit taking it in for questioning, let alone any hope of a conviction on grounds of convincing and legally admissible proof.
Even if we managed to adjust the penetration valve so cleverly that the presumed hand preference gene on the x chromosome came extremely close to producing the precise chances of inheriting left-handedness that exist in reality, we wouldn’t really have got very far, because a gene variant can never directly explain the distribution of hand preference. The fate of Darwin’s famous finches shows why.
In 1835, halfway through his famous journey on HMS Beagle, Charles Darwin caught and stuffed all manner of birds that lived on the Galápagos Islands, including a number of rather unsightly finches of various kinds that were extremely similar in appearance. The males were jet black, while the females had mousy, grey-brown plumage. The main visible difference between the species was the shape and size of their beaks. Only many years later, long after
he returned home to Down House in Kent, did Darwin realize the deeper significance of this variation. Under the influence of the highly variable foods available on the different islands, a single species of finch had given rise to six separate, specialized species. Where the food on offer took the form of large, hard seeds, individual birds with big nutcracker beaks were at an advantage and became dominant, until eventually all the finches on those islands had large beaks. On islands where the birds were dependent on small seeds that were tricky to extract, the opposite happened.
Three of the six species of finch that are found only on the Galápagos Islands. At the top left the small-beaked Geospiza párvula, at the top right Geospiza fortis and below them Geospiza magnirostris, whose name means ‘with a large beak’.
We now know that the shape of a bird’s beak is dependent upon two gene products, BMP4 and calmodulin. The more BMP4 an avian embryo produces, the broader and deeper its beak, whereas calmodulin makes a beak longer. The contrasting beak shapes of Darwin’s finches represented the first of many genetic divergences. Over the longer term the finch populations on the islands, isolated from each other, built up such a large number of genetic differences that they became separate species.
Although we generally think of evolution as a slow process, nature can sometimes take rapid strides. In 1982 several individuals of the finch species Geospiza magnirostris were blown across to the Galápagos island of Daphne by a severe storm. On that island, where large, hard seeds were available, the coarse-beaked finches flourished, far more so in fact than their less well endowed native cousins G. fortis. But in 2004 drought hit Daphne. It was so severe that hardly any large seeds could be found that year. Only cactus seeds were available, and they were difficult to work loose from the plant using a large beak. The population of newcomer finches declined precipitately and the previously disadvantaged individuals with smaller beaks took the lead once more. G. magnirostris died out. Even among the native species, individuals with the smallest beaks quickly gained ground and within a short time after the drought the average size of the birds’ beaks fell considerably. This meant not only that a new, different balance between species had arisen for the second time in a quarter of a century but that the characteristics of one species, under pressure of circumstances, had changed with what in evolutionary terms was lightning speed.
This is of course a very different matter from the emergence of a truly new species, but all that’s important here is that genetic variation instantly came into play among the finches when conditions altered. It’s a familiar pattern. Variation within its package of genes makes a species flexible and robust, increasing the likelihood that part of the population will be able to survive changes to the environment. What if all species of Geospiza on Daphne had been endowed with large nutcracker-style beaks? There would probably be no finches left at all on that island today.
The source of genetic variation is the creation and in some cases disappearance of alleles as a result of mutation. This is the motor of evolution: without mutation no new variation; without variation no selection, no adaptation to changing circumstances and eventually no living creatures. This means, however, that the relationship between carriers of different alleles is inherently unstable. The entire system, after all, is geared to making change possible. As a consequence of that instability, characteristics that aid successful reproduction will eventually spread through the whole population. Equally inevitably, everything that is disadvantageous will be selected against and eventually disappear. Since some genes are recessive, rare or have low penetration, this can sometimes take many generations, but truly detrimental inherited characteristics will not take long to become all but eradicated.
The extent to which gene-linked characteristics will be found that have no effect at all on reproductive success is unpredictable. Such traits ought really to cease to exist because there is no force ensuring their spread. The fact that they do sometimes occur is a result of what’s known as genetic drift, the aimless floating around in the genome of neutral alleles. It happens as follows.
The intractable law that says favourable alleles spread through an entire population while harmful alleles disappear is a statistical one. It holds true for large populations as a whole, as measured over many generations. We can imagine such a population as an ocean of successive generations of individuals. Seen from an aeroplane, high above cloud level, the ocean looks like a broad, shining, virtually smooth surface. That is the average view. But from a ship on the water the sea may be calm and flat as a mirror, or an inferno of heaving waves higher than a house, or a washboard of raging white crests. In such large systems as seas and worldwide populations, considerable deviation from the norm can occur, usually at limited times and locations.
One such possible deviation is the spread by chance, within a small subpopulation, of a gene variant that has no effect. From there it can pene trate further into the population. This is what we mean by genetic drift. It’s sometimes been claimed, precisely because the percentage of left-handers is so constant, that left-handedness has become anchored in the population by such a route. This is nonsense. The fact that a characteristic that has no effect neither becomes more common nor disappears as a result of selective pressure does not make it stable. It can disappear as easily and as much by chance as it appeared. That could happen here and not two communities away, or vice versa.
Instead of stability, one would expect characteristics that came into being through genetic drift to cause particularly large differences between subpopulations. Suppose for instance that a ship full of colonists of whom only 2 per cent happened to be left-handed arrived at a remote island and the travellers settled there. If left-handedness originally arose because of genetic drift, we would expect in theory that in the absence of selective pressure that percentage would remain constant. Many centuries later left-handedness would still occur in only about 2 per cent of the population of the island. Either that or the rare phenomenon of left-handedness would have disappeared altogether as a result of chance. This is not what happens.
If the characteristic distribution of left- and right-handedness cannot be deduced from existing, recognized biological mechanisms, then it is sensible to turn the matter on its head and begin by defining the characteristics of its distribution before looking for a mechanism that fits.
For all the mystery, there are one or two things that we do know for certain. We can be sure, for example, that there is an inherited aspect to hand preference and that the basic probability of its manifestation is an unchanging 10 per cent or thereabouts. We also know that left-handedness in a parent doubles the likelihood that a child will be left-handed. Oddly, every characteristic that conforms to these stipulations turns out to be self-stabilizing, as long as the product of the basic probability of its occurring multiplied by the influence of both parents remains below 100 per cent. Beyond that, the characteristic will spread unstoppably across all individuals.
Take our ship full of colonists on a remote island again. When they start to reproduce they form pairs that will sometimes consist of two right-handers, sometimes of a right-hander and a left-hander and sometimes of two left-handers. How many pairs of each composition there will be depends on the size of the group and the percentage of left-handers on board. But if every child that results has a 10 per cent chance of being left-handed and if that probability doubles with each left-handed parent, then after a while the colony will consistently produce left-handedness at a rate of about one in ten, no matter how many left-handers arrived with the ship and irrespective of whether the group grows, shrinks or maintains its original numbers. It doesn’t even make any difference if there was only one left-handed person on the original ship, or only one right-hander, or 50 per cent of each. In all cases, after a handful of generations, 12.7 per cent of the population will be left-handed, a percentage that from then on will not change at all.
It barely matters at all whether a trait of this type is advantageous or disadvantageou
s in evolutionary terms. Which is nice, because one of the puzzling aspects of the existence of left-handedness alongside right-handedness has always been that no one could think what positive effect on reproductive success might ensure that left-handedness continued to exist. If hand preference were a matter of the emergence of gene variation under selective pressure, then there would have to be a positive effect of some kind. It might perhaps be something like the advantage of having sickle cell anaemia in regions where malaria is prevalent, but no one has ever been able to think of anything concrete. Now that we know that left- and right-handedness cannot be dependent on a duo of alleles, we are liberated from the requirement to come up with this elusive advantage. Evolutionary quality is of hardly any significance in the case of self-stabilizing traits.
To understand this we need to look at another human characteristic, one that has nothing to do with hand preference but does exhibit the same kind of stable distribution. That characteristic is sexual orientation.
How left-handedness stabilizes in an initial population of 10,000 people, of whom in the first case 2% and in the second case 98% are left-handed, given that the basic likelihood of left-handedness is 10% and each left-handed parent doubles that likelihood. The average number of children per couple is two, so the population neither declines nor increases. Within five or six generations the stable distribution of 12.7% to 87.3% is reached.