by Meave Leakey
As so often happens in science, somebody else was puzzling over the same quandary but through quite a different line of enquiry. The man who had also picked up on this phenomenon was our old friend and colleague, Thure Cerling, an easygoing geophysicist with a vast and eclectic knowledge accumulated over years of fieldwork that has taken him to all corners of the globe. Our association with Thure stretches all the way back to Richard’s first Omo expedition in 1968, and he is always a pleasure to work with in the field. He wrote to me in 1992 that he’d been “analysing carbon isotopes in fossil teeth fragments of herbivores. I have exciting evidence of a major change in the vegetation in the Siwaliks at about seven million years, and I want to see if I can detect similar changes in the isotopes at the same time in East Africa.” The Siwalik deposits are in Pakistan, and although the majority are older than those at Lothagam, some cover Lothagam’s sweeping age span. “Please do join us, as soon as you can!” I replied. “This is a perfect coincidence—we seem to be picking up a similar signal from the tooth morphology in many of the herbivores at Lothagam. But how on earth do you detect such a major change in habitat just from tiny little fragments of teeth?”
Taking me back to one of my first biochemistry lessons, Thure explained that all cells have a carbon base and that plants manufacture their carbon through a process of photosynthesis. Capturing carbon dioxide from the air and combining it with water, plants use energy derived from sunlight to split the oxygen and hydrogen in the water and reconfigure the atoms as oxygen and glucose (glucose being the simplest hydrocarbon, which is the building block of all cells). Dredging back to my days at the polytechnic, I did remember the basic formula for photosynthesis, CO2 + H2O + light energy → (CH2O)n + O2. But it’s not that simple! A number of intermediary steps are involved between throwing carbon dioxide, water, and light together and producing sugar and oxygen.
There are actually three types of photosynthesis adapted to different conditions, and as they have an important bearing on our story, they merit a bit of explanation. The most common type is the one I learned about in school—C3 photosynthesis. With less machinery (fewer enzymes and no specialised anatomy), this process is the most economical: C3 plants use the same enzyme, rubisco, to both latch hold of the carbon dioxide and process it during photosynthesis. But it turns out that rubisco is only efficient at getting hold of carbon dioxide in conditions of moderate heat and light. At higher temperatures, the enzyme also latches on to oxygen, and the reaction creates a different compound the plant does not need, causing the plant to waste precious energy.
C4 photosynthesis is an adaptation to avoid this predicament in hotter climates. C4 plants use a different enzyme to get hold of carbon dioxide, which it then delivers to the rubisco enzyme in an inner cell for the photosynthesis process. Because this enzyme is more efficient in carbon dioxide uptake, the plant can keep its pores open for a shorter time, thus cutting evaporation rates and saving on scarce water. A key difference in this process is that the intermediate step creates a four-carbon molecule as opposed to the conventional three-carbon sugars first created in the C3 pathway. In cooler climates, it is most economical for plants to use the C3 pathway, but above a certain temperature threshold, the additional expense of C4 reaps greater efficiencies, which makes it worthwhile. Thure also told me about CAM, a third photosynthetic pathway that is ideally suited to arid conditions because the plant is able to shut down operations completely during prolonged dry spells. But because CAM is restricted to fewer plant species, mostly succulents, it needn’t concern us here.
So how does all this biochemistry relate to the changes we were witnessing at Lothagam? Put very simply, we are what we eat. An herbivore that derives its basic building-block carbon from plant material will pick up the photosynthetic signature of the plant matter it ingests, and this is preserved in the tooth enamel, hair, and hooves. By measuring the relative composition of the carbon isotopes, we are able to discern the proportion of C3 to C4 plants in the animal’s diet. Because most of the trees and shrubs in sub-Saharan Africa today are C3 plants and the majority of grasses and hedges are C4, this ratio neatly translates to the proportion of leaves versus grasses ingested by the animal. It turns out that this signature is preserved perfectly in fossil tooth enamel. Thure had been analysing the teeth from modern mammals coupled with dietary information established by direct observation as a control group for comparing the fossils.
This is just brilliant! I thought. A tiny fragment of fossil can now yield invaluable indicators of the animal’s diet. Similar analyses of carbonates in fossil soils indicate the dominant vegetation at the time, and the ratios of two different oxygen isotopes can tell us about the sources of drinking water the animal used. All these analyses can greatly add to our understanding of the climate and habitat millions of years ago. Once I had grasped the basics, I couldn’t wait to hear what Thure had found in Pakistan. I eagerly pored over his papers as I waited for him to join us in the field.
Thure’s research in Pakistan documented a dramatic shift in the biomass away from plants photosynthezing with the C3 pathway to a dominance of C4 plants some seven million years ago. We don’t know for certain what caused this change in vegetation, but we do know that the relative efficiencies of C3 or C4 pathways at different ambient temperatures is related to the concentration of carbon dioxide in the atmosphere: below a certain CO2 level, the rubisco enzyme simply isn’t very efficient at pulling carbon dioxide out of the air. If Thure’s observations of a sudden expansion of C4 biomass in Pakistan were repeated at Lothagam, this could indicate a warming event that was global rather than localised and could be the impetus that led our own ancestors to adapt to a changing environment. We know that CO2 levels in the atmosphere had been steadily decreasing during the Late Miocene. If Thure was right, I couldn’t help but wonder what this might mean for the havoc we are causing with our current runaway carbon emissions. Will this affect our global food supply? As a species, we are almost totally dependent on a terrifyingly small number of cash crops for the bulk of our food supply: just three cereals now provide about 60 percent of plant-based human energy intake. Over the last century, some 75 percent of plant genetic diversity has been lost. Out of some 250,000 to 300,000 known edible plant species, only 150 to 200 are used today. Metaphorically speaking, we don’t have very many eggs in our basket!
Thure finally joined us for a couple of days in the field at the beginning of August in 1992, and I excitedly put the whole field crew onto finding tooth fragments for him. Focusing on the time interval when we expected to detect a change, we searched systematically, starting at the base of the Lower Nawata and walking up-section through each time interval through the Apak. Thure made detailed notes on the geology and provenance while I collected the specimens and plotted them on the aerial photo. He was amazed at the field crew’s skill and how much they found in such a small time—after a day and a half, we had ninety-eight specimens! The sample provided identifiable fragments of the dominant herbivore taxa from each time sequence: elephants, rhinos, horses, giraffes, pigs, hippos, and antelopes. A fringe benefit of Richard’s new position was that he now had easy access to bones of modern herbivores from a wide range of habitats across Kenya’s network of parks—most notably from the mounting stacks of skulls and mandibles of elephants lost to poaching. A number of Richard’s wardens kindly contributed to our research with tooth fragments that significantly bolstered our control sample of modern herbivores.
Thure worked with John Harris on the analyses back in Utah, and we couldn’t wait to get our hands on the results. When they finally came in, they were fascinating. As Thure had suspected, they tallied neatly with what we were seeing in the morphology of the pig and elephant teeth. As in Pakistan, many of the species changed from dominantly browsing to dominantly grazing diets during the Nawata, and the analyses of the fossil soils showed a corresponding replacement of C3 plants with C4 grasses as the dominant vegetation. It was clear that there was a dramatic change in the
vegetation at this time. Subsequent work in South and North America has shown the same signal in horses, so it seems that this change was indeed global in nature. Such a widespread change in habitat at low latitudes would have completely altered the world. And, as new feeding opportunities opened up and old ones diminished or disappeared, this would certainly have driven both the mass extinctions and the evolution of numerous species we were witnessing at Lothagam. Our analyses showed that the animal groupings did not all undergo this transition at once; it was a phased progression instead.
The charge was led by the horses and the proboscideans, with the horses the first to change to a largely C4 diet. The pigs were the slowest to fully exploit the new C4 dietary resource. In spite of the apparent changes in suid molars that would have favoured a grazing diet, pigs remained mixed feeders through the Nawata Formation. Not all animals became grazers, however. Extensive forests and woodlands remained, and some animals, like the deinotheres, continued to exploit these habitats. Among the species of rhinoceros represented at Lothagam, two strategies were successful—we find the earliest occurrence of the modern white rhino, a grazer, as well as the earliest occurrence of the black rhino, which remained a browser. This complex ecological partitioning of diets is best illustrated by modern bovids that show the whole range of feeding strategies. Hypergrazers such as the oryx, hartebeest, buffalo, and kob are at one end of the spectrum, and hyperbrowsers at the other extreme include kudu and duikers. Mixed feeders such as impala and gazelle successfully exploited both food types.
The implications of this change in habitat and the timing of it are mind-boggling. The massive faunal turnover occurred at the end of the Miocene between seven and five million years ago. This corresponds exactly with the time that the human lineage is estimated to have split from that of the apes. Could the change to a new “C4 world” in Africa have provided the initial stimulus for bipedality?
* * *
I HAVE ALWAYS been convinced that the main driver for such a momentous change must be first and foremost related to feeding. Any slight variations that help give an individual animal a competitive advantage in finding food, defending against predators, and gaining a mate are strong selective pressures in evolution because individuals must reproduce for a species to continue. And those variations related to feeding are paramount because individuals must obviously first survive to adulthood before they can breed. These selective pressures act most strongly in times of stress and hardship because an animal that cannot feed efficiently will not live long enough to pass on its genes nor will it breed successfully during prolonged food shortages.
Charles Darwin did not believe that natural selection could be witnessed in real time as he was convinced that changes visible to the human eye must accumulate over many generations beyond a human life span. But in the famed “Enchanted Isles,” where the beginnings of Darwin’s revolutionary theory of evolution were spawned, the extent that climatic fluctuations exert selective pressures on a population is crystal clear. In the Galápagos, there is a little island called Daphne Major. Here, the drama of the struggle of life over death is played out daily on a stage that is pared down to the minimum essentials.
On this lava-strewn crater of an island, there is a population of finches small enough to count and tag but large enough to provide a statistically robust sample. The finches are absurdly easy to catch because they have no fear of humans, and they have highly variable beaks that are easy to measure within a fraction of a percent. The finches’ ability to forage successfully depends almost entirely on the shape and size of the beaks that nature endows them with, and the beak a particular bird is endowed with is highly hereditary. Then there are the seeds that the birds eat, which vary in size and toughness but belong to few-enough species that researchers can easily recognise and count exactly what type of seeds individual birds are eating. They can also relatively easily calculate the density of seed mass on the island at any one time. Through continuous meticulous research in this near-perfect natural laboratory since 1973, Peter and Rosemary Grant and their team of finch watchers have witnessed both natural and sexual selection unfold before their very eyes in what is quite possibly the greatest show on earth. They were able to prove through solid, unassailable data, what Darwin could only impute by logic. Little idiosyncrasies, such as a tiny variation in the shape and size of a bird’s beak, can literally be the difference between life and death, survival and extinction. In Darwin’s own words, “the smallest grain in the balance, in the long run, must tell on which death shall fall, and which shall survive.”
The Grants’ study of the Galápagos ground finches would later be eloquently described by Jonathan Weiner in his Pulitzer Prize–winning book, The Beak of the Finch, a riveting and crisply detailed account of evolution in action. As I read the Grants’ scientific articles while I was pondering the complete transformation of Lothagam’s long-ago climate, what struck me most was the irrefutable correlation they found between the size and shape of the finches’ beaks with climatic fluctuations.
The Grants first arrived in the Galápagos during the wet season, and the finches were enjoying four good years of decent rainfall and plentiful food. The finches concentrated on two dozen species of seed and spent half of their foraging time eating seven favourite soft seeds and fruits. This confusing picture is what threw Darwin off the scent some 140 years earlier, for he also arrived during the wet season. If all the finches are eating the same food, why does the large ground finch, Geospiza magnirostris, have a large deep beak and the small ground finch, Geospiza fuliginosa, have one about half as deep and half as long? And why is there a medium ground finch with a medium beak, Geospiza fortis, occupying a middle zone? If Darwin is correct, all these variations should eventually zero out into an “average” middling bird with an average, medium beak?
The Grants got their first inkling of why this was not happening when they came back in the dry season. Now the birds were devoting only one-thirtieth of their time to the seven species that they had favoured in the wet season. On a “struggle index” of seed hardness and size, the average seed now ranked 6 compared to 0.5 three months earlier (the toughest seeds to crack approached 14). And the volume of food had decreased by 84 percent. With less to go around, the finches had all become specialists. Cactus finches now dined almost exclusively on cactus, and the large G. magnirostris focused on big heavy seeds. Even within a single species, the medium G. fortis ground finch, the pattern was the same: each bird according to its beak. Fortis finches with relatively big beaks focused on big seeds, medium individuals ate medium seeds, and small-beaked birds ate the smallest seeds.
Then the study got really interesting. During the 1976 wet season, Daphne Major received 137 millimetres of rain, and the resulting density of ten grams of seed per square metre sustained a population of more than a thousand fortis finches and three hundred cactus finches. But in the following year, there was an epic drought, and only twenty-four millimetres of rain fell in the whole year. The cactus finches bred following the first rainfall, but not a single fledgling survived past three months. The fortis did not breed at all. The seed density dropped dramatically to three grams of seed per square metre. There was very little food, and between eleven a.m. and three p.m., the black sunbaked lava was too hot for the birds to forage. By January 1978, there were fewer than two hundred finches alive, a grim survival rate of one in seven. The Grants found that the oldest and the biggest birds had lived while the younger generations were practically wiped out. Surviving fortis were, on average, 5 to 6 percent larger than their dead brethren. Before the drought, the fortis population’s average beak was 10.68 mm long and 9.42 mm deep. After the drought, the survivors had an average beak 11.07 mm long and 9.96 mm deep. Half a millimetre or less is the seemingly infinitesimal difference that decided their fate. One drought and one generation were enough to show that natural selection was much swifter and stronger than Darwin himself had ever imagined.
But Daphne Major had more ric
h surprises to offer the Grants. There was another key feature to the roster of survivors: many more males made it than females because males are on average bigger than females. When at last the clouds gathered and fifty millimetres of rain fell in a single torrential deluge, a breeding frenzy began. But each female had six males to choose from, and choose she did. The females deliberately picked the biggest birds with the biggest beaks as mates. Their offspring had even larger and deeper beaks. Sexual selection accentuated the power of natural selection during the drought. For the next four years, this trend towards ever-larger birds with bigger and deeper beaks was perpetuated.
Then El Niño arrived in 1982. Every three to six years, the warming of the waters of the Pacific Ocean wreak havoc on worldwide weather patterns, and the Galápagos sit in the eye of this upheaval. On Daphne Major, sheets of water sluiced down the steep slopes of the volcano, and more rain fell than any time in living memory. The bare, depleted lava burst into leaf and flower, yielding such a bumper crop that there was twelve times as much seed as the year before. But it was too wet for the arid-loving cactus, so the big-seed crop crashed as the small-seed supply boomed. In such a time of plenty, this didn’t matter overly much to the finches, who feasted indiscriminately. Their numbers climbed by over 400 percent, soaring to more than two thousand individuals by June.
The following year, only fifty-three millimetres of rain fell and a paltry four millimetres barely dampened the earth the year after that—and now the big-beaked fortis were the ones in trouble. The finches had overshot the island’s carrying capacity, and small seeds were in greater supply than big ones. Imagine a room with grains of rice strewn across the floor. Give two teams a bowl and an implement to gather their dinner. You would sleep on a far fuller tummy if you were on the team armed with tweezers rather than pliers. And so it was for the finches. The biggest birds, needing the most calories to fuel their big bodies and having the least dexterity in their big beaks, were now at a disadvantage. Big-beaked big birds were dying, and small-beaked small birds were flourishing. And more males were dying than females. Nature had neatly and ruthlessly reversed the trend of the previous years. The “normal” distribution of the size of the finches and their beaks swung back again according to the whims of a fickle climate.