The fertility clinic: a bird's-eye view of our future

This article refers to:

Introduction to ‘The fertility clinic: a bird's-eye view of our future’

‘The Universe makes rather an indifferent parent, I'm afraid’, said Dickens’ kindly Mr Jarndyce in Bleak House (Dickens 1865, p. 75). Humans have evolved to understand and intervene in the unsentimental processes of nature—with some unfortunate and unintended consequences. Back to nature or on to the future?

From 1948–68, New Zealand farmers applied vast quantities of DDT to control grass grubs (not realising until the 1970s that rye grass monoculture was to blame and DDT actually made the problem worse). When the insecticide was banned in 1968, the Department of Scientific and Industrial Research (DSIR) looked at biological control using starlings. We began a study in 1970, using 500 nest boxes, and were surprised to find dead adults in the boxes. DDT killed many birds, so we asked the Ministry of Agriculture and Forestry (MAF) to check corpses—they all had DDT residues. Starlings in that area still have, after 40 years, and four times more than in any other country (Eens et al. 2013), but not at levels that would kill them. But these sudden deaths of apparently healthy birds remained a mystery until we found a pair linked together with their claws in their opponent's eyes. Normally, one bird would grip first and live to fly away. A claw into the brain via an eye socket left no mark: the perfect murder (Fig. 1).

Figure 1 Corpses of two starlings that killed each other. ‘[W]hen two army corps may mutually annihilate each other in a second, probably all civilised nations will recoil with horror and disband their troops’ (Alfred Nobel, 1892).

When we went to publish our observations, the editor said: ‘Nature red in claw—put that in front of your title; everyone knows Darwin, even if that's all they can remember apart from “survival of the fittest”.’

‘But their claws don't get red,’ we objected. ‘That's why MAF vets couldn't tell what was killing them.’

‘Doesn't matter,’ he said. He was a great editor, and had accepted the paper for his journal, so we took his advice. The title became: ‘Nature red in claw: how and why starlings kill each other’ (Flux & Flux 1992).

But why? In Dickens' day, mothers admonished their squabbling brats with:

Birds in their little nests agree;

And 'tis a painful sight,

When children of one family

Fall out, and chide, and fight.

‘Love between brothers and sisters’ – Isaac Watts (1715, Song XVII)

Now such fighting is considered normal, a way of establishing a pecking order to form a stable social structure. Fighting to the death, however, is unusual (except among humans). In game theory, it pays only as a final resort, a last chance for future reproduction. The starlings in our study area had increased to 1500 pairs and could get a box only by fighting for it. All these fights were male v. male or female v. female. In Darwinian terms, the victor was bound to be a fitter mate; so it didn't pay to help your partner to defend the box—better to team up with the winner.

A major problem remains. Starlings are quite capable of nesting in odd places—in trees, holes in rocks or rabbit burrows. But they can also dig into clay banks or road cuttings to make their own ‘boxes’, and there were plenty clay banks in the study area. Why fight to the death when you could just make a nest somewhere else? It seems that starlings, like many colonial nesters, must nest within the colony or miss out altogether (Flux 2003).

Wynne-Edwards (1962) suggested that this was a form of population control (like territory defence) that keeps wild animals safely below the food limit. There is good support for this in the way that populations of domesticated animals, selected over generations for docility and tolerance to crowding, expand to the food limit and then starve (Flux 2001).

The way in which self-limitation of populations evolved has been much debated. Wynne-Edwards (1986) opted for group selection—groups with altruists fared better than those without—and Darwin (1871, p. 166) had already suggested this mechanism:

Although a high standard of morality gives but a slight or no advantage to each individual man and his children over the other men of the same tribe, an increase in the number of well-endowed men and advancement in the standard of morality will certainly give an immense advantage to one tribe over another.

‘But just because Darwin said it, doesn't make it true’, say critics of group selection (e.g. Barash 2008). They prefer kin selection and inclusive fitness as mechanisms. Others see selection acting at several levels—between genes, individuals, races or species (Eldredge 1985)—and the reality of group selection is now generally accepted (Wilson 2012). Competition is the key, and can be a remarkably strong force.

In a study of how to remove rabbits from islands, competition from hares proved to be twice as effective as introducing predators, or even myxomatosis, one of the most lethal viruses known, which kills 99.9% of rabbits (Flux 1993, 2008). However well adapted an individual or tribe may be, competition between species determines which species survive; and it explains why most species are extinct.

As with much science, irrelevant theoretical sidelines were more interesting than the job in hand. We were supposed to be helping farmers to understand and intervene in the unsentimental process of pasture damage by turning loose these murderous starlings on the grass grubs. But did they eat grass grubs? And if so, did they eat enough to prevent damage? You can't buy starlings at Wrightsons (an agricultural supplier) to release on your farm. Even if you could, they regularly fly 30 km to feed, except when nesting, when they are restricted to a 600 m circle. The trick is to put up nest boxes where you want damage control and hope the starlings fatten their chicks on grass grubs. Hence our study involved killing chicks to see what they had been fed.

Killing animals is unpleasant (although noisy, scratchy, flea-laden starlings can be hard to love) and can be justified only by getting as much information as possible. Starlings lay 2–8 eggs in a clutch, mean 4.5. If we killed only chicks from small clutches (2–4) we would be selecting for high clutch-size in our population. How fast is evolution?

This was an ideal set-up to answer the question in an unconfined wild population, something that hadn't been tried before. The rate of evolution depends on selection pressure (the more chicks killed, the faster clutch size increases) and is opposed by gene flow (unselected starlings coming to nest in our boxes from other areas dilute the effect). Both of these we could measure, and the study design had one enormous advantage: the selected birds and the unselected ‘controls’ would be nesting side by side under exactly the same environmental conditions.

This was the golden age of science. In New Zealand, DSIR was directed by scientists whose mandate was the advancement of knowledge, not obeisance to New Zealand's economy. They realised that we draw from, and contribute to, a world pool of information, freely available and unpatented. Technology could be managed; science was (despite CP Snow) an art. They said go ahead.

Under the new unsentimental processes introduced by science managers, staff loyalty has been lost (Flux 1979). Who today works an 87 hour week (the starlings had to be fitted in with night work on hare behaviour) when paid for 37? Capital outlay was 500 nest boxes at 15 cents each; the farmers would find out if starlings controlled grass grubs, and we could see how long it took for starlings to change their clutch size. The physical outlay was rather higher—we had to climb a total of 3000 ft up ladders every second day for 3 months to look into the boxes. By the end of the 10 year study, that was Everest from sea level 40 times. At our fertility clinic, 1700 adult starlings and 4500 chicks were banded, and data filled a filing cabinet. The final publication occupied two pages (Flux & Flux 1982).

The results? At about 50% selection pressure, and 70% gene flow, starlings would take 28 generations to lay one more egg. And farmers would have to be altruists, because grass grubs were taken mainly out of the nesting season when the birds were on other farms. Few farmers are altruists, so they turned to organophosphate insecticides instead.

But the real value was in realising how evolution worked. Clutch size is under multi-gene control (how many is unknown, but mouse litter size is controlled by 80 genes) and is not highly heritable (h₂ = 0.3). Body weight is twice as heritable, and linked to clutch size, which is why poultry breeders find it easier to select for size to produce more eggs. In starlings this worked backwards: selecting for clutch size had resulted in heavier starlings.

In case starlings were being selected to lay five small eggs instead of four larger ones, all eggs were measured. Egg size varied with age, and between years, possibly in relation to food supply; but there were no significant trends, nor change from the original British stock (Flux 2006). On being handled, some starlings deserted their nests, so we were incidentally selecting for tameness; and tame starlings are larger. (This is true for many animals, and is why farm animals are generally bigger than wild types. It holds for humans too—gentle giants and stroppy Hitlers.) Some starlings are ‘cuckoos’ and lay eggs in other starlings’ nests. These enlarged clutches avoided culling, so we were selecting for this antisocial behaviour as well. It's hard to drive evolution in the direction you want without unexpected side effects.

‘Survival of the fittest’ gives a misleading picture of evolution. To most people it means the biggest, strongest, fastest or fattest. But in evolutionary terms it is the genes carried by the animal or plant that leave the healthiest descendants in following generations; and that doesn't mean just the first generation. Conditions might change, a new predator or parasite might move into the area; the successful individual has to have enough genetic diversity. Evolution is full of dead ends: if your long wings are not needed any more, you will be eliminated by your short-winged compatriots before you can evolve short ones. Hence the best strategy is to hedge-bet, keep your head down and stay away from avant-garde fashion.

One way of hedge-betting came to light when we measured how fat starling chicks were (Thompson & Flux 1991). Heavy chicks were not always the fat ones; even in the same nest, the distribution of fat was not uniform. There seemed to be some spoilt brats. It was well known that starlings adjust brood size to feeding conditions by allowing the smallest runt to die. But fattening one or two chicks in a brood at the expense of the rest could give them an advantage if the weather at fledging was bad, while in good weather slim birds could escape predators better (Flux & Thompson 1995).

Starlings are ideal subjects for studies of climate change because they nest synchronously (and with tighter synchrony when nests are closer to each other [Evans et al. 2009]). Over the first 17 years of our study, the start of laying drifted 2 weeks later—a very significant shift, so far unexplained (Flux 1987). A Polish colleague wrote to say this paper was being ignored because birds were expected to breed earlier with climate warming; and had we more data that he could analyse? We had, he did, and the result was very curious (Tryjanowski et al. 2006): starlings at Belmont laid early in warm and cold years, but later in average years. And in a remarkable parallel, starlings in Northland and Southland nested earlier than those in central New Zealand (Bull & Flux 2006). Food supply is probably the reason: in warm years invertebrates thrive; in bad years or cold regions more lambs die, providing a feast of maggots; so in normal years there is relatively less food.

The usual (Victorian?) image of a pair of starlings producing half a dozen young, four of which die leaving a pair to replace the parents, is also wrong. Recruitment is strongly skewed in favour of a few Genghis Khans. Our study of lifetime productivity, measured as the number of daughters returning to the area to nest in future years, is a typical pattern. One female had five daughters, three had four, 12 had three, 44 had two and 135 had one; the remaining 850 females left no descendants at all. The successful females did not seem to differ from the failures in any way. If you think about it this is obvious: starlings today are just the same as when Darwin (or indeed Aristotle) watched them; if the only successful individuals were the big ones, say, they would by now have evolved to be the size of crows.

Individual productivity also varies widely. Of two females that nested for six years in the study area, one laid 27 eggs but fledged only three chicks, all of which returned to breed. The other female laid 33 eggs and fledged 22, but none returned. Consider the role of chance in the outcome: one southerly storm killed 60% of the young nestlings, and they were the best young, early hatched and heavy. The poor nestlings of late nesters were still being brooded by the parents and all survived. In 1965, Miriam Rothschild wrote: ‘Blessed are the meek—that is the not too successful—for they shall inherit the earth. One of the unpalatable truths about natural selection is that it imposes a certain mediocrity’ (Rothschild, 1965, n.p.).

In addition to physical evolution we have to consider cultural evolution. Starlings can learn new tricks, such as taking nectar from flax or building an open nest in a tree fern, by watching other birds. This information can be remembered for at least a year and is soon copied by other starlings. Gulls have learned to nest on buildings in several countries since 1960. House sparrows started taking dead insects from car radiators by flying up under the grille of parked cars in Britain, America and New Zealand almost simultaneously (Flux & Thompson 1986). Puffins on the island of Staffa in Scotland (but not, so far, elsewhere) come ashore when there are enough tourists watching them to scare off puffin-eating gulls.

Cultural evolution spreads fast—far more rapidly than is possible by natural selection. It is especially important in humans, where access to knowledge now outstrips our physical and mental abilities to cope. The level of wisdom, however, seems not to have increased since the Greek or Roman times. In Alexandria 2300 years ago, Eratosthenes correctly calculated the weight of the Sun, Earth and Moon—as the WUV advert says, it makes you think.

What about the statement: ‘Humans have evolved to understand and intervene’? The words are in the wrong order. Humans have evolved to intervene before understanding, which is why there have been too many ‘unfortunate and unintended consequences’ to bother listing. Even when they do understand, 70% go ahead anyway if there is a fast buck to be made: ‘If we don't do it someone else will, so why not?’ Other unfortunate consequences have followed well-meaning but short-sighted interference by teachers, doctors and missionaries. The most unfortunate of all will prove to be the loss of natural population control methods, and their substitution by religions promoting sanctity of life (human only) and love thy neighbour (human only). We don't learn from history, despite being the only animal with written records. Previous civilisations failed after over-exploiting their resources (Wright 2004; Diamond 2006); ours will too.

Back to nature or on to the future? The answer seems to be both at once.

Back to nature is the direction chosen by Greenies. They are altruists prepared to leave something for future generations. In most societies (of any vertebrates) they account for 5%–10% of the population, and this appears to be a genetic trait. Hence the Green vote is unlikely to increase even when their warnings of environmental doom come true.

In ecological terms, based on the energy coming from the Sun, the Earth is capable of supporting 2 billion people. The 7 billion present now are eating into capital, be it fossil fuels or nuclear power. Alternative energy sources such as wind or tides are also finite; build enough tide turbines and the tides slow down. But the available capital is enormous. It could feed, using artificial light and hydroponics, a thousand times the population with existing engineering techniques, provided people were content to be the only species on Earth and live their entire lives in continuous cities layered one above the other. Think battery hens and virtual reality. Layered cities underground date back a thousand years in Turkey, and permanent underground living has recently been demonstrated in Austria. The only problem is human frailty— someone forgets to turn on the power to level seven and a few billion people die. Any survivors would be tempted to turn off the rest of humanity.

Does Darwinism offer any insight on the likely outcome? Selection has a holiday while populations are in logarithmic expansion, because there is little competition. All sorts of odd individuals, tribes and cultures survive. Starlings killed each other when their limiting resource (nest boxes) ran out, but they were still within the food limit and not heading for extinction. Humans are heading to extinction because they became domesticated 11,000 years ago and lost their mechanisms to regulate population. Recall that the starling study was tied in with work on hare behaviour? Six hares in a 3 ha enclosure bred successfully, but after 3 years there were still only six hares. A colleague put 11 rabbits (two females) in a 5 ha enclosure; they bred like rabbits, in 3 years there were 690, they starved and the population crashed (Gibb et al. 1978). Why the difference? Hares are normal wild animals: rabbits were domesticated from AD 600–1200 before running free over Europe, and have not re-evolved their population control system. The prognosis for domesticated (civilised?) humans is bad, even if they get a worldwide fertility clinic operating right now.

Starlings, rabbits and hares, adapted to the monoculture pastures we provide, may grieve our passing; but natural selection eliminates the stupid, and the universe will not intervene. As Stephen Crane (1988[1899], p. 36) put it:

A man said to the universe:

‘Sir, I exist!'

‘However,' replied the universe,

‘The fact has not created in me

A sense of obligation.'

Disclosure statement

No potential conflict of interest was reported by the authors.