Beak of the Finch March
Atheist Society Lecture 12 March 2019
Robert Bender

The Galapagos and its finches

The archipelago is about 600 km west of Ecuador in the eastern Pacific and comprises many islands large and small, all of them volcanic. Some 2 million years ago, it is estimated, a small population of finches were blown out to sea from South America and landed on one of these islands. They eventually radiated to form 13 different species, varying hardly at all in plumage, but quite markedly in beak size and strength and body size.

From 1969 Peter and Rosemary Grant spent 25 years visiting the archipelago with their graduate students, studying the finches very intensively on several of the smaller islands, especially a very tiny one just north of Santa Cruz, Daphne Major.

Daphne is surrounded by steep cliffs, is very difficult to land on, and has never been occupied by humans, so it was pristine territory for the study of its birds.

In 1994 Jonathan Weiner, a prolific science writer, produced a book mainly devoted to the work of the Grants and their graduate students, The Beak of the Finch. Its main theme was that in an important respect, Darwin was wrong – he firmly believed that evolution occurred so slowly that it could only be recognised to have occurred at intervals of many thousands, perhaps millions, of years, and would not be detectable in the short term.

Charles Darwin collected a few finches while the Beagle was in this archipelago but was unaware they were all finches, as they varied so much in size and beak. On his return to England, his ornithology colleague, John Gould, classified them all as finches. Darwin’s writings about them have led to them being called Darwin’s finches as they supposedly set him thinking about evolution of species on islands, but they do not get any mention in the Origin of Species.

Darwin published his theory in 1859. Its subtitle "by means of natural selection" is deceptive. He believed it could not be demonstrated in real time, but only by analogy with artificial selection by domestic breeders. He mainly studied pigeons.

Similarly, his "preservation of favoured races in the struggle for life" was not demonstrated. Instead he demonstrated the preservation of varieties favoured by breeders. "Natural selection" long fell into disfavor, replaced by Mendelian genetics and mutations.

"Darwin’s finches" are not mentioned in the Origin, but get a paragraph in his Journal of the Beagle. It commented on the presence of 13 species with fine gradations of beaks, possibly adapted to "different ends" which he did not specify.

"The remaining land birds form a most singular group of finches, related to each other in the structure of their beaks, short tails, form of body and plumage: there are thirteen species, which Mr. Gould has divided into four sub-groups. All these species are peculiar to this archipelago; and so is the whole group, with the exception of one species of the sub-group Cactornis, lately brought from Bow Island, in the Low archipelago.

Of Cactornis the two species may be often seen climbing about the flowers of the great cactus-trees, but all the other species of this group of finches, mingled together in flocks, feed on the dry and sterile ground of the lower districts.

The males of all, or certainly of the greater number, are jet black; and the females, (with perhaps one or two exceptions) are brown.

The most curious fact is the perfect gradation in the size of the beaks in the different species of the Geospiza, from one as large as that of a hawfinch to that of a chaffinch, and (if Mr. Gould is right in including his sub-group, Certhidea, in the main group) even to that of a warbler.

Figures from the Origin of Species illustrating beak differences among the finches

The largest beak in the genus Geospiza is shown in Fig. 1 and the smallest in Fig. 3, but instead of there being only one intermediate species, with a beak of the size shown in Fig. 2, there are no less than six species with insensibly graduated beaks.

The beak of the sub-group Certhidea, is shown in Fig. 4. The beak of Cactornis is somewhat like that of a starling, and that of the fourth sub-group, Camarhynchus, is slightly parrot-shaped. Seeing this gradation and diversity of structure in one small intimately related group of birds, one might really fancy that from an original paucity of birds in this archipelago, one species had been taken and modified for different ends."

In 1876, 17 years after the Origin was published, Thomas Huxley delivered a lecture about the demonstrative evidence for evolution – it was all about the sequence of horse fossils which showed that over some millions of years horses had grown from the size of a spaniel to the animal we know today, and its teeth had modified along the way from the Eocene, around 50 millions years ago.

The changes in horse leg bones, toes and teeth could only be seen at very long intervals. This has been the standard view for over a century.

In the 1930s David Lack spent ten months in the Galapagos, studying the finches – as opposed to Darwin’s two weeks – and wrote a wonderful book about them, trying to explain the sequence by which one species became thirteen and how some islands came to have several finch species co-existing on them.

Gould’s original classification is still pretty much the accepted view – he divided the finches into seed eaters, cactus-flower nectar feeders, insect-eaters and warblers (feeding on small soft insects). So they have diversified into several different ways of life, exploiting different resources on the islands

DNA has of course enabled precise study of the family tree of this group of finches, to see which are more or less closely related, and how they relate to the finches of South America. Darwin put strong emphasis on variation within a species as the basis for natural selection. Some of the finch species are very variable, with larger or smaller bodies and deeper or shallower beaks, while others have very little variability. Within Gould’s groupings, the birds do differ markedly in size, for example in their body weight, most obvious in the Geospiza group that the Grants studied: fuliginosa, fortis and magnirostris.

Two of the thirteen species are resident and breed-ing on the island and two others visit occasionally from nearby Santa Cruz, but do not breed there. So the Grants could study the interaction of the two breeding species over this entire period. They are both ground species that pick up seeds fallen from plants growing on the island.

These plants are of several species, some producing small soft seeds, and one introduced weed producing spiny, woody, tough seed pods very difficult to crack open. Daphne Major is small, only 500 hectares. The Grants marked out 8 patches of 150 x 150 metres, within which to study the birds. They found the two species of ground finches ate seeds from 20 species of plants, a mix of big and small, easy and hard to crack open, soft and hard.

Peter Grant invented a tool to measure the force needed to crack open seeds of various plants and ranked the forces from 0.5 to 12. The ones ranked 12 could only be cracked open by heavy birds with deep, wide, strong beaks, while the small soft seeds were available to all the species of birds, but were small and each one contained only a small amount of nutriment so the big birds had to consume very many of them to get their daily needs.

Grant ranked the seeds in terms of the difficulty of extracting the nutritious kernel from the exterior husk, and they ranged from Portulaca, at 0.35 to Cordia lutea at 14.0. So a range of beak sizes and strengths would have different abilities to open and benefit from the seeds of the different plant species.

His team then marked out 50 grids of 1 x 1 metre and collected every seed within each square metre, to assess the relative abundance of each seed type

Darwin in 1835 observed on the few islands he visited, that the different finch species seemed to all be eating the same variety of foods. Lack in the 1930s observed the same thing, as he too was there in the wet season, but he suggested in his book that visiting in the dry season, when competition between the birds would be intensified, could be revealing. The Boags (a couple of Grant’s students) actually did that, and in 1977 when there was a very severe drought, most of the finches died of starvation as so little seed was set by the over-stressed trees, shrubs and cactus.

One of the fundamentals of Darwin’s theory is that in a variable species, which individual lives or dies is not a random selection. Some part of the variation spectrum will be better adapted to survive stressful events. And so it proved with these ground finches.

One of the seed species that appears much in the story is an invasive weed from North America, Tribulus terrestris, Caltrop or Puncture vine. The seeds grow in tightly bonded groups that must be broken apart into "mericarps", then forced open to reveal the nutritious seeds inside, 4 to 6 per capsule. The pods are woody and very tough, so only the birds with the biggest beaks can open them.

Geospiza magnirostris has a powerful beak, but is only an occasional visitor to Daphne Major. The Geospiza fortis with the biggest beaks could open the Tribulus seeds, but with some difficulty. Birds of the same species with smaller beaks were unable to open the seed pods. The difference between beaks that could and couldn’t was 0.5 mm. in depth

Of course it doesn’t help the plant to have its seeds eaten by birds rather than dropping to the ground and starting new plants. So a survey of the distribution of Caltrop found that on one side of the volcanic crater, where birds were abundant and ate most of the seeds, the plants had more and longer spines to protect the seeds, and put less energy into growing seeds, more into protecting the smaller number of seeds. The other side of the crater, with few birds, the plant was less protective, the seed pods had fewer and shorter spines, so the plant put more energy into growing seeds and less into protective devices.

The drought led to a great decrease in the abundance of seeds for birds to eat. 1 x 1 m. grid counts done several times a year showed that the 10 grams per square metre of early 1976 rapidly fell to only 2 grams per square metre by late 1977. Soil temperature measures showed that from 11 a.m. to 3 p.m. the surface was too hot, up to 50o, for the birds to stand on the ground hunting for seeds.

Parents were unable to feed their chicks, so no chicks survived. There was also not enough moisture to support the usual insect population, to the insectivorous birds starved to death as well.

This in turn led to a far greater than usual death rate of the birds, as their food supply failed. There had been ~1,400 Geospiza fortis on Daphne Major in 1975, but during the drought, as well as many adult birds dying, breeding failed totally and virtually no chicks fledged. The number of birds decreased to under 200, so only 1 in 7 survived.

Grant’s team captured and measured the birds before and after the drought. They found there had been strong selection pressure against birds with smaller beaks, and in favour of birds with larger, stronger beaks, able to pry open the toughest big seeds.

The distribution had been almost a normal distribution with mirror-image tails at both ends, and a mean beak depth of 9.77 mm. After the drought the distribution was far more skewed, with a long tail at the small end of the distribution and a peak at the larger end. The mean value had shifted to 9.94 mm, a mere 0.17 mm difference, but those bigger birds that survived produced offspring with bigger beaks. So the overall population had moved in a single year towards birds having deeper, wider, stronger beaks, in response to a single year’s drought. Natural selection in action, as Darwin wrote in his Origin of Species, though he was unable to demonstrate it with data like that collected by the Grants.

One of the issues that needed to be settled was whether beak size is heritable by the next generation. One of the Grants’ research assistants, Jamie Smith, went off to do his own research in a Canadian island in Vancouver Bay, Mandarte, where an endemic sparrow with much variability in the birds’ sizes enabled him to perform an experiment. He took eggs from the nests of small sparrows and placed them in the nests of big sparrows, and vice versa. He wanted to learn whether the beak sizes of the chicks would develop to match those of their natural parents, or their adoptive parents as a result of the parents feeding practices. He found the adopted birds grew to very strongly resemblance their natural parents, so their body and beak size was strongly determined by genes, and not by their nurture. An interesting study in the nature vs nurture debate.

The shift in seeds was also towards greater hardness of the seeds, disabling the smaller birds from getting access to them. There was a progressive increase in the mean size of birds surviving on the island. Those with strong breaks, able to break open the toughest seeds, survived. Smaller birds with weaker beaks perished and left no descendants. This size shift was reinforced by sexual selection, as the females favoured birds with bigger beaks as mates, so the big-beaked birds got to pass on their genes and had big-beaked offspring.

Peter Grant wrote a paper about his team’s observation of the two ground finches’ feeding behaviour on the Tribulus seeds and the consequent advantage to the bigger birds when those seeds became the bulk of what was available to eat.

In 1982-3 there was a very strong El Nino bringing severe flooding to the island.

This did something really interesting to the relative abundance of seed types. The Tribulus and Cactus plants were soon smothered by scrambling vines and set little seed, while the smaller plants set seed several times within the year, able to respond rapidly to the sudden excess of water. The chart shows the relative abundance of the large seeds (white columns) and small seeds (black columns). During the 1977 drought year the large seeds were far more abundant than small seeds, which created strong selection pressure favouring birds with large, strong beaks. Following the El Nino year, the large seeds were scarce and the small seeds super-abundant, which created very strong selection pressure favouring small birds that could quickly obtain their daily food needs from small seeds. So the situation changed very markedly after the El Nino year.

The next question is: why don’t the birds just go on growing bigger, each time a drought succeeds a previous drought? The answer came in 1982-3 with the El Nino floods. Vines grew over the Tribulus and cactus bushes and prevented them from setting seed, so there was a different shift in relative abundance of seed sizes. The abundant seeds were now small soft ones. Small birds could easily get their daily needs hopping about on the ground, larger birds had to work harder to fill their bellies.

The constant rain caused many birds to abandon nests, so there was a considerable death rate of chicks. By the end of the flooding year, the size distribution of birds had shifted back the other way, with natural selection favouring smaller birds against larger ones. What the Grants had learned was that evolution is going on every day, but that it does not thrust in one direction only. It is a compromise between forces operating to make birds’ beaks smaller and other forces operating to make them larger – an oscillation in the short term.

Many species can hybridise, such as horse and donkeys, or lions and tigers, if their speciation was fairly recent and their genetic make-up is still fairly similar. But with most species, such as horses and donkeys, the hybrids are sterile, so there can be no third generation.

The Grants and their students had leg-banded every bird and knew who was mating with whom. After the flood-year they found the number of cross-species matings rose significantly, and these produced many offspring, that survived better than the pure-bred birds. Over several years the number of hybrids continued to increase, and they have speculated that under the right condition, two species can return to being one species, an "unspeciation" process. The offspring of course have few other hybrids to mate with so they engage in back-crossing, but if the hybrid matings become frequent enough there would be many birds with the genes of both species, and separating species to say which bird belongs to each would become quite difficult.

In 1948 J. B. S. Haldane published a short paper on the rate of evolution. He proposed a common measure based on a percentage change in the size of some organ, such as a skull, or arm bone or tail. He called his unit the "Darwin"

One Darwin – 1% change in a measurement per million years. He saw this as normal, based on evidence of several fossil species: horses and dinosaurs. He thought evolution was very slow, not visible short-term. "Rates of one Darwin would be exceptional in nature."

In the Galapagos drought the change rate was 25,000 darwins. 1% in 40 years. In the flood 6,000 darwins. 1% in 166 years. Evolution really occurs very fast.


On the island of Trinidad, off the north coast of South America, and in nearby Venezuela, there are steep, fast-flowing mountain streams, with pools separated by waterfalls. In each pool there are populations of small fish, guppies. The gravel that forms the beds of the pools is made of multi-coloured tiny pebbles, making a spotted, multi-coloured background for these guppies. The guppies are also spotted in multiple colours. If their spots are small and pale, they can blend in with the background pool floor. If their spots are large, bright and garish, they can stand out and be easily visible. Many of the pools contain guppy-eating predators. The pools highest up the hills tend to contain a single predator species, a Rivulus that tends to gobble a single guppy each 5 hours on average, usually very young ones. Going down the mountains, the pools steadily increase in number and variety of predatory guppy-eaters until at the bottom of the mountains some pools contain seven species of predators, including cichlids, which tend to eat more guppies per hour than the Rivulus, and target mainly the larger, older ones. So the intensity of predation pressure grows as one goes down the mountain to lower elevations. As one descends from the highest to the lowest pools, the nature of selection pressure changes and intensifies.

Guppies need to not only avoid being eaten, but also need to find mates and reproduce, to leave another generation of guppies. To find a mate, they need to stand out from the background, achieved by having larger and more garish spots on their flanks

In the upper pools, Endler’s colleague, David Reznick found that many of the guppies had bright large splotches of colour on their flanks, so as to stand out from the background and be more visible to potential mates. So the guppies have two opposed adaptation pressures – to blend into the background to avoid being eaten, and to stand out from the background to be more successful at mating and reproducing.

David Reznick studied these streams and their guppies. What he found was that the highest streams, with a single predator, had garish guppies with large spots. It looked as though the guppies had adapted to a relatively predator-free situation there and opted to advertise their mating potential rather than opting to avoid detection by predators. The problem was to explain why such a pattern had developed.

Endler and Reznick drew a template of a guppy, dividing it into "pixels" and mapped the spots on every guppy they captured and photographed. They found each individual is unique, its pattern of spots unlike that of every other fish. They also found the fish stayed in the same pools in which they had hatched.

In the streams near the bottom of the mountains, with many predators to avoid, including voracious cichlids, he found the guppies had many very small pale spots, and blended far better into the background, so they had adapted to their situation by avoiding garish colours and favouring survival over reproduction.

To see how long it might take for this varied situation to develop, he set up some artificial ponds at his university in California, captured guppies in Trinidad, created mixed populations of interbred guppies, so each pool had a mix of guppies with many small pale spots and a few large garish ones and everything in between. He let them live in predator-free ponds for several generations. Then he introduced predators to these artificial ponds – some had just the Rivulus, and others had up to all seven predator fish, reproducing closely the situation in their ponds in Trinidad.

He found the populations of guppies developed to very close resemblance of the Trinidad populations very fast, within a few generations, so the ponds with the single predator had mainly garish big-spotted guppies, and the ponds with many predators had mainly guppies with many tiny very pale spots.

The ponds with single predators that targeted very young guppies had fish populations that grew faster to escape from the size range targeted by the predator, and began reproducing later in life, once they had got past the vulnerable size.

The ponds with many predators that targeted big adult guppies had developed populations that grew more slowly, to avoid growing too fast into the vulnerable size, and they reproduced earlier in life, while still below the size range targeted by their predators. All this evolution in the ponds occurred over about 11 years, extremely fast evolution. Darwin would have been very surprised by all this.

One website I found had a set of three photos and lists of predators from three pools, to show the nature of the selection pressure that influence guppy spot patterns. Pool 1 is high in the mountains, with a single predator, pool 3 is down near sea level, with the three selected predators and a different adaptation to this situation by the guppies.


There is a small group of finch-relatives called Crossbills, most of them native to North America. Their bills do not meet, but cross over and look as though they have been twisted out of shape.

David Lack in his book on Darwin’s Finches, drew attention to the issue of Crossbills and their different foods and adaptation of species to their food source.

Lack pointed out that study of Crossbills would be worthwhile, as they seemed more closely adapted to their major food sourse than their relatives on the Galapagos.

They feed on the seeds in conifer cones, their crossed bills being well adapted to prying cones open before they naturally open to release the seeds. The birds thereby gain food denied to other animals that have to wait for the cones to open. The three North American species are the two-barred crossbill, Loxia leucoptera that feeds on the soft cones of the larch, the heavy-beaked Parrot crossbill, Loxia pytyopsittacus that feeds on the much harder pine cones, and the Red Crossbill Loxia curvirostra that feeds on spruce cones. Back in 1947, David Lack, author of the classic book on Darwin’s Finches, suggested that this group of birds would repay study of how adaptation varies with diet.

One of the big issues in evolution is whether organs adapted to special uses are only useful when they are fully developed, or are useful to some degree while in process of evolving towards the forms we see today.

In 1990 Craig Benkman and Anna Lindholm conducted an interesting experiment on Red Crossbills to investigate this question. They trimmed the beaks of some Red Crossbills until they were no longer crossed, but the upper met the lower when the beak closed, then watched the birds feed, and continued watching as the beaks slowly grew back to their natural crossed length.

When the beaks were no longer crossed, the birds at first had to wait for cones to spring open to gain access to the seeds. But even with a millimetre of regrowth, some birds, with some success, and more slowly than usual, managed to pry open the spruce cones and get some seeds that way. As their beaks steadily grew back to the crossed shape, the birds steadily improved their skill at prying open cones, and gained a growing amount of their food this way until when the beaks had fully regrown they reverted to their normal practice of gaining all their food prying open cones.

This demonstrated that as the crossbills were evolving, at each stage the crossed bills were marginally more useful to them in prying open cones than they had been when the crossings were smaller. There was no stage at which the developing organ was of no adaptive value to the birds.

Benkman and Lindholm scored a two-page paper in Nature for their efforts.


Dolph Schluter, noe of the Grants’ graduate students, studies stickleback fish, in five lakes in western Canada which each have a pair of species. In each case the two species have found a way to avoid competing with each other. One has become a bottom-feeder (benthic) while the other exploits the water column above the lake floor (limnetic). Their diets contain a suite of food items that do not overlap, so they specialise in different foods.

In all five lakes similar adaptations have emerged to enable the two fish species to live well in their chosen segment of the water column. The bottom feeders (Benthics) are bigger and fatter, have a wider gape when they open their mouths and eat bigger prey. They suck in water through their mouths and expel it through their gills. Inside the gills are "gill rakers", structures that filter food out of the water as it passes through the fish. These gill rakers can have different forms, being either short and stiff, or long and flexible. The different forms filter out different things, to some extent sorting them by size. The benthic bottom-feeders have short and stiff rakers that filter out relatively large prey.

The limnetic fish occupying higher segments of the water column are all smaller, more slender, have narrower gapes and feed on smaller prey. Their gill-ratkers are long and flexible, just right for collecting small prey items.

So the middle ground has been abandoned, and the fish have retreated into specialisms to avoid competing with each other.

Other nearby lakes that have only a single species of stickleback show no sign of this sort of competitive divergence, are all about the same size – in between the two sizes of the lakes with two species, and share the same prey items. This suggests that when there is a competition for food between two species, they will tend to specialise, and also to adjust their size to suit their specialism, and develop non-overlapping diets and a non-overlapping size range, just as was seen by the Grants on Daphne during the 1977 drought. If there is just a single species, this selection pressure is absent and a single size range with a single dietary regimen develops. Again, just as was seen by the Grants – where Geospiza fortis had an island to itself, and was not in competition with Geospiza fuliginosa, its size range settled in between the ranges that developed when there were two species.

Dolph Schluter set up artificial ponds at his university in British Columbia to be stocked with a single species from these two-species lakes, stocked with the full variety of food items found in the real lakes. The ponds stocked with the smaller fish species showed they grew larger over time and occupied the middle-ground abandoned in the two-species lakes. The ponds stocked with the larger benthic species also reverted towards the centre, in the absence of competition. Natural selection at work again, showing what happens when a selection pressure is introduced, or when it is removed, from an ecosystem.

Soapberry bugs

The soapberry tree, Sapindus saponaria is native to Texas in north America. It produces a near-spherical fruit, which is targeted by the soapberry bug, Jadera haematoloma. It pokes its sucking beak tube through the fruit’s flesh to reach the seeds, injects a chemical to liquefy the seeds, then sucks them out through its tubular beak. The length of its sucking beak is well adapted to the size of the Soapberry fruit, which has a radius of 6 mm. The bug’s beak is 6.5 mm long, just right for reaching the seeds, that are 6 mm from the fruit’s surface.

There is a population of the bugs in Florida where it feeds on a different tree, Cardiospermum corindum, which has a radius of 12 mm. The local soapberry bugs have grown their beaks to 9 mm to adapt to this larger fruit. This suggests that bugs with beaks that were too short failed to thrive, while those with longer ones outcompeted them and left more grandchildren.

In the last half century, three new species of East Asian trees have been introduced to these parts of USA, and the soapberry bugs have exploited their fruits for the seeds. One is in Florida and two in Texas. The Florida introduced species is the Flat-podded Golden Rain Tree – it doesn’t have the balloon-like shape of the Balloon vine fruit, so the seeds are much closer to the fruit surface.

Their fruits have radii different from those of the native trees, and the soapberry bugs are rapidly adapting the lengths of their beaks to be more appropriate to the new food sources. The Florida species is now exploiting the Flat-podded Golden Rain Tree, Koelreuteria elegans, which has a fruit with a radius of only 2.8 mm.

The soapberry bugs exploiting it have so far managed to reduce their beaks to an average of 7 mm. and have clearly not stopped the shortening process. The Texan bugs are adapting to Asian trees with slightly plumper fruits, so have needed to lengthen their beaks to reach the seeds inside them. As all populations of bugs (or anything) are variable, some have slightly longer or shorter beaks, and the food source favours the bugs at one end or other of the variable continuum. In Florida it has favoured the short end, in Texas the long end, so there has been in a mere 40 years a shift in beak length.

Scott Carroll’s team published charts showing their measurements of the beaks of Soapberry bugs they found feeding on the different tree species. The fruits have non-overlapping size distributions, the insect beaks have size distributions that just overlap, so the shortest ones feeding on the native fruits overlap with the longest ones feeding on the exotic fruits, but the adaptation process is not over yet, and further beak shortening should be expected.

Apple Maggot Flies

Benjamin Walsh (1808-1869) was an American entomologist. He observed that a maggot fly, Rhagoletis pomonella, usually fed on Hawthorn fruit, haws. In Walsh’s time it was adapting to the introduced apple – but restricted itself to a few varieties.

It has spread across USA and Canada as apple farming has spread, and now reached the west coast. It has become a serious pest.

The interesting issue now is: has the Hawthorn maggot fly split into two species? The Apple maggot fly doesn’t interbreed with it, even when they have overlapping populations

The orthodox view is that for a species to split into two species, it needs a geographical isolated population to start differing from its parent species.

The apple maggot fly lives in the same areas as the Hawthorn maggot fly.

Field studies have shown that the Hawthorn maggot flies eat haws, and they live their life cycle in and under Hawthorn trees.

Apple maggot flies eat only apples and live their life cycle in and under apple trees.

All the "speciation event" needed was a new host plant.

"This shift, which occurred approximately 150 years ago in the north-eastern United States, provides a unique historical context in which to examine the relationship between host specialization and speciation in action, in real time, in our own backyards."

Feder’s genetic research found that 11 of 29 allozymes had already become different in the two groups of maggot flies, so it is close to speciating or it already has speciated, in only 150 years.

Tuskless elephants

Elephants in East Africa have been targeted by poachers eager to make money from selling the ivory tusks for the past half century, especially during a civil war in Mozambique, a very lawless period. In the late 1960s and early 1970s, about 10-20% of all wild elephants were being slaughtered for their tusks each year. While the war lasted, over 90% of Mozambique’s elephants were slaughtered, mainly for their tusks, sold to finance the war.

Some elephants have genes that produce large tusks, much prized by poachers. Others have genes that yield small tusks or even no tusks. Poachers targeting the ones with big tusks created strong selection pressure in favour of small or no tusks, as these elephants survived the marauding poachers, and lived long enough to produce another generation.

So the genes for small or no tusks are now far more common. Andrew Dobson, a professor of ecology at Princeton has studied the phenomenon.

A 2008 paper published in the African Journal of Ecology noted that the number of tuskless female elephants in Zambia’s South Luangwa National Park and adjacent Lupande Game Management Area had increased from 10.5 percent in 1969 to 38.2 percent in 1989—the peak of the previous ivory wars—largely as a result of illegal hunting for ivory. A 1991 elephant conservation plan in Uganda reported a higher-than-normal percentage of tuskless elephants in Queen Elizabeth National Park and singled out poaching as the main cause. Whereas a normal level of tusklessness in an elephant population is somewhere between 3 percent and 4 percent, according to the Ugandan report, a 1989 survey of Queen Elizabeth National Park revealed tusklessness in the elephant population to be between 9 percent and 25 percent.

"Elephants carry a sex-linked gene for tusklessness, so in most populations there are always some tuskless elephants," says Poole. "Because males require tusks for fighting, tusklessness has been selected against in males and very few males are tuskless. For African elephants, tuskless males have a much harder time breeding and do not pass on their genes as often as tusked males."

In heavily poached populations, says Poole, the ratio of tuskless animals in the population increases as poaching continues.

"Whereas baseline tusklessness in a population might be 4 percent, over time as more and more tusked elephants are killed, the percentage may increase to 60 percent in the older animals," she explains. "When this group breeds with tuskless females, 50 percent of whose daughters are tuskless, you begin to see the gene for tusklessness spreading in the population. You can see this in almost any population that has experienced a wave of heavy poaching, in Gorongosa [in Mozam-bique], for example, or Selous [in Tanzania]."

So poachers are really conducting an experiment in evolution of the elephant. Exactly the same experiment is going on among big game fishermen, who select the biggest marlin to kill for their trophies and thus create selection pressure for marlin to be smaller fish.

Finches and creationism

John Endler, the guppy watcher, told Weiner a story about his travels:

"Not long ago on an airplane I talked for an hour with someone about what I do, and never once mentioned the word evolution. It’s very easy to do, you know. Darwin himself doesn’t use the word evolution in the whole of the Origin. You just talk about what happens, and how you can study what happens: changes over many generations…

The whole time on the plane, my fellow passenger was growing more and more excited. ‘What a neat idea! What a neat idea!’ Finally, as the plane was landing, I told him the neat idea is called evolution. He turned purple."

And Rosemary Grant has a similar story: "I’ve done exactly the same thing - and never let on it was evolution – and got exactly the same response. I described our work on Daphne to a Jehovah’s Witness. And he followed along, and said ‘Oh, how fascinating.’"

"Asked intelligent questions," says Peter.

"And I never plucked up enough courage to say, ‘Well, you know, what all this means…".