Weeks pass. We are sitting next to the fireplace in MacArthur’s living room, with notes and graphs spread out on a coffee table.
Wilson: So far so good. The numbers of bird and ant species do go down as islands get smaller and farther from the mainland. We’ll label the two trends the area effect and the distance effect. Let’s take them both as given for the moment. How do we know that they prove the equilibrium model? I mean, other people are almost certainly going to come up with a rival theory to explain the area and distance effects. If we claim that the results prove the model that predicted them, we will commit what logicians call the Fallacy of Affirming the Consequent. The only way we can avoid that impasse is to get results that are uniquely predicted by our model and no one else’s.
MacArthur: All right, we’ve gone this far with pure abstraction—let’s go on. Try the following: line up the extinction and immigration curves so that where they cross and create the equilibrium, they are straight lines and tilted at approximately the same angle. As an exercise in elementary differential calculus, you can show that the number of years an island takes to fill up to 90 percent of its potential should just about equal the number of species at equilibrium divided by the number going extinct every year.
Wilson: Let’s look at Krakatau.
Krakatau is the small island between Sumatra and Java that had been wiped clean of all life in the great volcanic explosion of August 27, 1883. Scientists from several nations, principally the Netherlands and Indonesia, then a Dutch colony, began to visit the reduced remnant of Krakatau within a year of the event. They managed to keep a spotty but serviceable record of the return of birds, plants, and a few other organisms to the bare volcanic slopes. The basic equilibrium model we developed predicted that the birds in particular, for which the best data of all were available, should reach equilibrium at about thirty species. Upon approaching that level, the fauna should be losing one established species by local extinction each year while acquiring one new species by immigration. The data gathered by the early researchers indicated that the bird fauna did indeed appear to be leveling off at approximately thirty species. But the turnover recorded was one species every five years, not one each year.
Was the model really off fivefold, or was the discrepancy due to sampling errors? There was no way to tell. At this point we saw the need of replicate data sets in order to advance equilibrium theory in a serious way. By 1965 I set out to devise such an experimental system in the Florida Keys, using the insects and other arthropods of the smallest islands. That is another story, an unusual, rather bizarre, adventure of field biology, to which I will return in the next chapter.
As MacArthur and I progressed on the island biogeography project, the loose confederation of young population biologists continued to grow. In late July 1964, five of us met at MacArthur’s lakeside home at Marlboro, Vermont, to discuss our personal research agendas and how they might contribute to the future of population biology. Joining MacArthur and me were Egbert Leigh, a young mathematician with a special interest in the structure of plant and animal communities, later to join the Smithsonian Tropical Research Institute as research scientist; Richard Levins, a theoretical population biologist of contemporary renown who later joined the faculty of Harvard’s School of Public Health; and Richard C. Lewontin, the rising star of theoretical and experimental genetics, who was to come to Harvard as Agassiz Professor of Zoology in 1973. In close touch with several of us but not present at the lakeside retreat were Slobodkin and Leigh Van Valen, a paleontologist and general evolutionary biologist at the University of Chicago.
For two days between walks in the quiet northern woodland, we expanded upon our common ambition to pull evolutionary biology onto a more solid base of theoretical population biology. Each in turn described his particular ongoing research. Then we talked together about the ways in which that subject might be extended toward the central theory and aligned with it. Besides island biogeography, in which MacArthur and I were now well advanced, I saw myself as adding the study of ants and other social animals to the enterprise. An animal society is a population, I argued, and it should be possible to analyze its structure and evolution as part of population biology. My student Stuart Altmann and I had already, early in 1956, discussed the idea of finding common principles to explain primates and insect societies. We had even used the term “sociobiology” to describe the effort. But we had had little intuition on how to proceed, and our collaboration had advanced no further. I hoped that the combined thinking of this new group, the “Marlboro Circle” as I have come to call it, would provide me with clues. The others were encouraging in their remarks, but few clues were forthcoming. William Hamilton’s article on kin selection and altruism, which was to be a keystone of sociobiology, was published that year, but neither I nor the others had yet seen it.
How to proceed with the sociobiological and similar overlapping agendas? There emerged from our freewheeling talk the notion of pooling our work. We would produce a series of essays under the single pseudonym “George Maximin,” in imitation of the French mathematicians who have been publishing since the 1930s under the name Nicolas Bourbaki. Maximin was named not in honor of the Roman soldier-emperor but after the point of greatest minimum in optimization theory; George was an arbitrary first name added. With Maximin we thought we could achieve the twin goals of anonymity, with its freedom from ego and authorial jealousy, while acquiring license to be as audacious and speculative as the group decided.
Maximin died an early death. He was an ill-conceived Frankenstein monster. By mid-August MacArthur was expressing serious doubts in letters to me. He argued that we should each take credit and responsibility for his own ideas. Slobodkin disliked the concept from the start. Maximin, he said, would look to others too much like a cabal. I had to admit that down deep I shared these misgivings. Personal idiosyncrasies doomed Maximin. MacArthur was particularly confident of his own powers and inclined to work unimpeded. He seemed to believe that he could generate ideas singly or in groups whenever the spirit moved him. Slobodkin for his part was turning against the idea of unifying theories and heavy dependence on mathematical modeling. I myself was temperamentally ill suited to Maximin, preferring to work alone or at most with a single partner. So the program faded, and for the most part the conspirators went their separate ways. We never met as a group again. But a lot was gained from Maximin’s ghostly spirit. I cannot speak for the others, but I believe we all carried away a new confidence in the future of evolutionary biology, and in ourselves.
By the end of the year MacArthur and Slobodkin were growing apart. Slobodkin, Robert wrote me in a letter, “is in an antitheoretical mood.” Nature defeats theory, Slobodkin was widely quoted as saying at the time. In August MacArthur pulled out of the biology textbook project we had planned three years earlier. Slobodkin by this time had produced little, and I had not done much better, having become distracted in the meantime by a half-dozen other projects. As a result the book soon followed Maximin into oblivion. We just stopped mentioning it. In 1966, when MacArthur published a short introductory text for freshman courses with Joseph Connell, Slobodkin condemned it with a slashing review. He opposed the very philosophy of science it represented. MacArthur in turn bridled at what he considered gratuitous hostility. He believed he had been misunderstood by retrograde thinkers. “I think I can tell why there are potatoes in the field and where they lie,” he mused to me, “but these people say no good, they want to know the size and shape of the potatoes.”
None of this had any effect on my collaboration with MacArthur. I believed deeply in the power of reductionism, followed by a reconstitution of detail by synthesis. In December 1964, I suggested that we write a full-scale book on island biogeography, with the aim of creating new models and extending our mode of reasoning into as many domains of ecology as we could manage. Robert agreed at once. He was enamored of the subject by this time, and had begun to call himself a biogeographer instead of an ecologist. In this domain e
xisted in most readily definable state the patterns he wanted to discover. When he later brought out a book under his own sole authorship, in 1972, he titled it Geographical Ecology.
Off and on in the two years following the Marlboro meeting, MacArthur and I assembled the pieces of an expanded theory of island biogeography. We explored the implications of the balance of species in the colonization of islands, lakes, and other isolated habitats. From published data we traced the course of the recolonization of Krakatau and other devastated islands. We examined the general qualities of the niche and the forms of evolution by which species adapt to dispersal and competition. We considered from the bottom up, species by species, the means by which animals and plants are most efficiently packed together to create diverse communities.
When our book, The Theory of Island Biogeography, was published in 1967, it met with almost unanimous approval in the scientific journals. Some of the reviewers declared it a major advance in biology. A quarter-century later, as I write, it remains one of the most frequently cited works of evolutionary biology. The Theory of Island Biogeography has also become influential in conservation biology, for the following practical reason. Around the world wild lands are being increasingly shattered by human action, the pieces steadily reduced in size and isolated from one another. Nature reserves are by definition islands. The theory serves as a useful tool in conceptualizing the impact of their size and isolation on the biodiversity they contain. Some parts of the formulation made in 1967 have been discarded by later authors, justifiably, and other parts greatly modified. Later researchers have added powerful new insights and definitive data sets unavailable to us at the time. I do not think it an exaggeration to say, however, that MacArthur and I accomplished most of what we set out to do. We unified, or at least began to unify, biogeography and ecology upon an internally consistent base of population biology.
In the 1960s and 1970s a new wave of population biologists trained in both ecology and mathematics passed through Ph. D. programs in the United States, Canada, and England. They gained the respect of the molecular and cellular biologists, and they were well funded for a while, before the academic recession of the late 1970s and 1980s. They shared the ambition and optimism of their immediate predecessors in the Marlboro Circle. I was able to play a role in this next step, more as a result of my residence at Harvard than of any special talent. My course “Evolutionary Biology,” begun in 1958, was relabeled “Population Biology” in 1963 and focused more on basic theory. At first I thought that I had failed by pushing model construction too far at the expense of natural history. An undergraduate complained in the Crimson Confidential Guide, the uncensored and often scathing student review, that the course was a dull exercise in numerology. So it might have seemed to some, but I came to realize later that many of the students were greatly influenced by my presentation, and a few were drawn into population biology as a career. They include some of the current leaders in the field: William Bossert, Joel Cohen, Ross Kiester, Jonathan Roughgarden, Daniel Simberloff, and Thomas Schoener. In 1971 Bossert and I collaborated on a short, self-teaching textbook, A Primer of Population Biology, which remains popular more than twenty years later.
In the spring of 1971 Robert MacArthur experienced abdominal pains during a field trip to Arizona. Returning home to Princeton, he learned that he had renal cancer. The affected kidney was promptly removed, and he was placed on chemotherapy. Too late: the surgeon told him that he had only months or at most one or two years to live. Robert thereafter conducted his life with even greater intensity than before. He completed his final book, Geographical Ecology. He journeyed to Arizona, Hawaii, and Panama for more field work, and while at the university he continued to guide his students. He began a new round of theoretical research, this time with Robert M. May, a brilliant Australian physicist who soon thereafter joined the Princeton faculty. Under MacArthur’s influence May converted to biology and developed into one of the world’s most influential ecologists. He subsequently moved to Oxford University as a Royal Society Professor.
MacArthur was still reasonably strong as the fall term began at Princeton in 1972. He was coughing frequently as the cancer spread into his lungs, but he was still able to come to his office for short periods to talk to students and friends. In early October his health declined rapidly. By this time I had joined several senior American evolutionary biologists—James Crow, Darlington, Hutchinson, and Eugene Odum—to nominate him for the National Medal of Science. With news that he had only a very short time to live, we redoubled our efforts. Robert sent word through Hutchinson that the nomination was welcome, and he was “pleased that my friends think well of me.” Geographical Ecology had also just been published, and he awaited the first reviews.
On a Monday afternoon, October 30, John Tyler Bonner, chairman of Princeton’s Department of Biology, dropped by my office while visiting Harvard. He told me that MacArthur’s condition had deteriorated badly and the end could come in hours or in weeks. The matters most on Robert’s mind at this point beside his family were, he reported, the National Medal and the reviews of his book. I dropped everything and inquired about both matters. No progress in the committee office at the National Science Foundation on the medal. But two back-to-back reviews of Geographical Ecology had just appeared in Science, one by Thomas Schoener and the other by Scott Boorman, both important young population biologists. I called Katherine Livingston, the reviews editor, who said she would send copies directly to Robert.
They arrived too late. The next morning I tried to telephone Robert at his home. A nurse with an unidentifiable foreign accent said he was sleeping. I called again at two in the afternoon, and this time he came on. His voice was thin but level. He coughed frequently, and twice he had to stop to get his position changed in order to continue. I was relieved to find his mind clear and composed. I asked, Had he seen the Science reviews? Not yet. I fished out the manuscript of the one by Boorman (who was at that time studying under my direction) and read it. The text was long, detailed, and laudatory. Robert was fascinated, and stopped me several times to discuss technical points raised. Boorman is clearly bright, he said. Was the review by Schoener as good? I assured him it was. I’d seen the manuscript, which after exploring the general methodology of model building declared Robert’s book to be the key synthesis in the field. He said, Good, it’s better than I got from Slobodkin for my elementary biology textbook.
Had I heard more about the National Medal of Science? I had not, except that eighteen people had been nominated and the awards would be announced sometime after the November 7 presidential election. Robert was disappointed. I sensed that he was worried about his place in biology. We then moved on to gossip and miscellaneous news. Our conversation remained normal in content and tone, with no serious digression into his physical condition. We talked as though he had years to live. He grew tired and quiet. I began to do most of the talking, afraid to let him go. I nattered on about the arrival next term of my fellow entomologist Bert Hölldobler to assume a professorship at Harvard; the opening of the new laboratory wing of the Museum of Comparative Zoology; and Lewontin’s political demonstration at the Chicago meetings of the American Association for the Advancement of Science and his widely publicized resignation from the National Academy of Sciences. We drifted on to a recent proposal to exterminate the kiskadee as a pest bird species on Bermuda. Robert mumbled assent as I went along.
At last Robert said we had talked enough and should stop now. We agreed to stay in touch. At dinner, Betsy later reported, he was calm and happy. He spoke with particular pleasure about the favorable Science reviews. In the early hours of the next morning he died without distress, in his sleep. Today I can imagine no more inspiriting intellect or steeper creative trajectory cut so short with such a loss to others. I wish he might have known in those final days that his place in the history of ecology was secure. I owe him an incalculable debt, that for at least once in my life I was permitted to participate in science of the first rank.
*This account was first presented in my essay collection Biophilia (Cambridge, Mass.: Harvard University Press, 1984).
chapter fourteen
THE FLORIDA KEYS EXPERIMENT
WHERE COULD WE FIND MORE KRAKATAUS?
That question dominated my thought for months after MacArthur and I published our first article on island biogeography in 1963. We had conjured a plausible image of the dynamic equilibrium of species, with new colonists balancing the old residents that become extinct, but we could offer very little direct evidence. There are few places in the world where biologists can study the approach to equilibrium on a large scale. Krakatau-sized events, the sterilization of islands the size of Manhattan or larger by volcanic explosions, occur at most once a century. Another hundred years might then be needed, once the smoking tephra cooled down, to observe the full course of recolonization. How might we get data more quickly, say within ten years?
I brooded over the problem, imagined scenarios of many kinds, and finally came up with the solution: a laboratory of island biogeography. We needed an archipelago where little Krakataus could be created at will and their recolonization watched at leisure.
My dream embraced more than the search for new experiments in biogeography. I was driven by a more general need to return to the field, to enjoy once again the hands-on kinesthetic pleasures of my youth. I wanted to remain an opportunist, moving among, seeing, and touching a myriad of plants and animals. I needed a place to which I could return for the rest of my life and possess as a naturalist and scientist.
Naturalist 25th Anniversary Edition Page 23