The Monkey's Voyage

by Alan de Queiroz

  This description may make Heads sound about as legitimate a scientist as a flat-Earth advocate, but that is far from the case. His papers appear in well-respected scientific journals, and he recently wrote a book on the biogeography of the tropics, published by the University of California Press. His antidispersal views are definitely extreme, but, in fact, they’re not far out of line with what many biogeographers started thinking in the 1970s and 1980s, when the field, while taking advantage of cladistics and incorporating continental drift, also made a turn down an intellectual cul-de-sac. In Heads’s case, at least, it’s clear where the strange thoughts came from.

  REBEL AMONG REBELS

  Most of the vicariance scientists mentioned in Chapter Two—Lars Brundin, Gary Nelson, Colin Patterson, Donn Rosen, and others—worked within typical scientific institutions, in particular at large natural history museums. They did the things that institutional scientists are supposed to do, such as conducting original research, going to scientific meetings, mentoring graduate students, serving as editors of scientific journals, and publishing peer-reviewed papers. They may have been intellectual rebels, but they were rebels working within the system.

  Then there was Léon Croizat, who not only did most of his work outside of the system, but also reveled in his status as a nonconformist, an iconoclast, an individualist rather than a cog in the wheel.

  3.1 Rebel, iconoclast, and opponent of Darwin’s ideas: Léon Croizat, the originator of panbiogeography, in Caracas in 1974. Photo by Jonathan Baskin.

  From the start, Croizat’s career trajectory didn’t resemble that of a typical biologist. He was born to French bourgeoisie parents in Turin, Italy, in 1894, but, by the time he was a young man, the family had fallen on hard times and he was scrambling to make a living. Fearing the fascists, he fled Italy in the early 1920s with his wife and two children and eventually settled in Massachusetts, where he worked at the Arnold Arboretum at Harvard, first mapping the grounds and then as a technical assistant. Although he wasn’t formally trained as a scientist, Croizat had been interested in biology from an early age. In the early 1930s, he started publishing botanical papers, at first only horticultural articles on cacti and other succulents, but later also technical papers on plant taxonomy, especially on the spurge family (Euphorbiaceae). In a couple of those papers he criticized the work of another botanist at the arboretum—someone well above him in the hierarchy—and got himself into hot water. The director of the arboretum, E. D. Merrill, apparently served as Croizat’s protector in that case, but following Merrill’s dismissal (his “defenestration,” in Croizat’s words), Croizat also lost his job. That episode may have been the first significant run-in Croizat had with another scientist, but it wouldn’t be the last.

  Unable to find another job, in 1947 Croizat immigrated to Venezuela, where he and his first wife divorced. He then married a woman who eventually became the head of a successful landscaping business in Caracas. That was a fateful turn of events for Croizat. Through his voracious reading of the scientific literature in the library at Harvard, he had developed radical thoughts on biogeography and evolution, and his second wife’s business allowed him to focus full time on expanding and self-publishing these views. Untouched by the usual channels of scientific criticism, and unfettered by editors, he poured out a series of books in the 1950s and 1960s that are impressive for their breadth of knowledge and their sheer bulk; the three-volume Panbiogeography alone is more than 2,700 pages long, and all told his self-published books run to more than 5,000 pages.17

  These books are, in a word, unconventional, and, in small doses, they provide a refreshing change from the usual dry science texts. Croizat’s writing is personal and culturally wide-ranging (he was fluent in more than half a dozen languages), and he constantly challenges authority, from Charles Darwin and Alfred Russel Wallace to George Gaylord Simpson and Ernst Mayr. As a very typical example, consider his sarcastic outburst against William Diller Matthew, the father of the New York School, and his herd-like disciples, from volume 1 of Panbiogeography:

  Gone forever are the days of that melanchonious soul, the individual thinker, carrying the weight, and challenge, of individual ideas when shaving in the morning and donning—fancy that—a night-gown at bed-time; the present and the future do belong to “collegiate” undertakings, mass-education, rosy vulgarization, two chickens in every pot and two cars in every garage. Matthew is the man of the times; he sees big, he sees rosy, he sees easy, he does not tire. So structured, and so precisely chiselled out in a messianically authoritative handicraft, Matthew’s “zoogeography” proves of course impervious to the pinpricks of reason.

  On the other hand, Croizat is often maddeningly vague and absurdly repetitive. A passage that begins as refreshingly opinionated frequently deteriorates, within the space of a few pages, into a tedious and venomous rant. For instance, in Space, Time, Form, the book that sums up his views (an 881-page summing up), he refers to his nemesis Darwin as “a very unhappy thinker,” “congenitally not a thinker,” “not a clear, cogent thinker,” “essentially not a thinker,” “not born a thinker,” “anything but a thinker,” “definitely poor as a thinker,” and “quite limited intellectually” (italics in original), along with many other descriptions to the same effect. Reading Croizat, one starts to think not only that he could have used a good editor, but that he needed a better filter in his brain as well.

  Style aside, though, Croizat had some things to say that would eventually permeate the field of historical biogeography. An obvious one was his constant harping on the dispersalists. He was convinced that Darwin, far from having finally put biogeography on the right path, had instead created a science based on a set of unsupported and unsupportable assertions (that was the context for most of those Darwin-as-an-unhappy-thinker quotes). Darwin and Wallace and, eventually, Matthew, Simpson, Mayr, and many others thought that species and higher taxa had “centres of origin” from which they had often spread out, generating wider distributions. For instance, Matthew, probably betraying some form of Eurocentrism, believed that most groups had originated in the Northern Hemisphere and subsequently dispersed to the south. Dispersal, in this view, was often of the long-distance, chance sort, and gave rise to isolated populations that eventually evolved to become distinct species. As we have seen, oceanic islands like Hawaii were key examples for the dispersalists; since these islands had originated and persisted as remote and isolated bits of land, native lineages, such as the honeycreepers and the silverswords, must have reached them by long-distance, overwater dispersal. Not surprisingly, this school of thought, beginning with Darwin and his experiments, emphasized “means of dispersal,” which dictated how easily different kinds of organisms could move about the globe and surmount barriers such as oceans. According to the dispersalists, there was a good reason why South America and Africa share quite a few genera and even some species of plants, but almost no land vertebrates: plants, with seeds that can float or be blown by the wind or survive a long journey on a raft, have a much easier time crossing the Atlantic than do lizards or frogs or rats.

  Croizat wasn’t buying any of this. “Centres of origin,” “chance dispersal,” and “means of dispersal” were all dirty phrases for him, reflecting how Darwin and his congenitally limited intellect had initiated the biogeographic equivalent of the Dark Ages. In Croizat’s view, Darwin had gone about things entirely backwards, coming up with a dispersalist theory of biogeography and then looking for facts that supported it. Croizat seemed to hold to the Baconian view that knowledge about the workings of nature would emerge naturally, without the bias of already having a pet theory in mind, if one simply compiled enough facts (assuming, of course, that one was “born a thinker” and could decipher the meaning of those facts). Croizat claimed that his views on the history of life had come about in just this way.

  3.2 Croizatian tracks for the ratite birds (ostriches, rheas, kiwis, and relatives). Enclosed areas rep
resent the group’s modern (including recent historic) distribution. Circles are fossil localities. Lines connecting areas are panbiogeographic tracks, interpreted as reflecting an ancestral Gondwanan distribution later fractured by continental drift. One can picture the Atlantic and Indian Ocean tracks getting shorter and shorter as one goes back in time (and the ocean basins contract) until finally the tracks are reduced to nothing and the ancestral, largely continuous distribution is seen. Redrawn and modified from Craw et al. (1999).

  The facts in this case were the facts of distribution. More specifically, Croizat refined a method that involved placing points on a map to indicate each region where members of a group—a genus of sunflowers or a family of beetles, for instance—were found, and then connecting the dots to form a distributional line that he called a track (see Figure 3.2). Examining many such cases, he found that the tracks for unrelated groups often overlap; for example, the tracks for baobab trees and ratite birds, among others, cross the Indian Ocean, connecting Africa and Australia. The line representing the overlap between multiple tracks Croizat called a fundamental, or generalized track. This approach of examining individual tracks, and from them identifying fundamental tracks, was the essence of Croizat’s panbiogeography, and the overall pattern of fundamental tracks was the key to his view of biogeographic history. He claimed that the “patterns of geographic distributions of plants and animals—whatever their ‘means’ [of dispersal]—are absolutely congruent, as a fact of nature, within a minimum of fundamental tracks and centers.”

  To Croizat, the observation that there are relatively few fundamental tracks led to an obvious conclusion: the processes that have given rise to distribution patterns must be general ones that have affected diverse organisms, with completely different “means of dispersal,” simultaneously. Long-distance dispersal is, by definition, a random process, so it couldn’t possibly account for these repeated geographic patterns. Lizards on rafts, seeds blown by hurricanes, snails on birds’ feet—these things would have produced a chaotic cobweb of individual tracks, not a small number of shared, fundamental tracks. That unhappy thinker Darwin may have believed that chance dispersal was a key to biogeography, but the facts clearly showed otherwise.

  In place of the notion of groups spreading out from centers of origin via long-distance dispersal, Croizat saw vicariance almost everywhere. In his view, the history of a group typically involved a phase of spreading out by means of normal, garden-variety dispersal, followed by a phase in which the group’s geographic range remained more or less static. Mobilism and immobilism, he called the two phases. During the immobilist phase, some external change, such as tectonic movement or a rise in sea level, often broke the overall range of the group into fragments. Populations in the isolated fragments would then be free to evolve into distinct forms. In other words, diversity was generated by the splitting of an ancestral geographic range, that is, by vicariance. Fundamental tracks crossing oceans reflected the shared ancestral ranges of many groups and their later sundering by a common process of fragmentation, tectonic or otherwise. For instance, a fundamental track crossing the Indian Ocean showed that various groups, such as the baobabs and the ratites, had once lived on a contiguous African-Australian landmass that had been broken apart by the opening of the ocean.

  Vicariance, of course, was not a new idea in the 1950s, going back as it did to the biblical literalists and perhaps even to the ancient Greeks, not to mention Darwin, Wallace, and the other “dispersalists,” who all believed to some degree in the importance of range fragmentation. What distinguished Croizat was what one might charitably call the “purity” of his vision: he saw vicariance as all-important, and long-distance dispersal as insignificant. As a consequence, “means of dispersal” were not even worth studying. Darwin’s experiments with floating branches and seeds in bird guts, his calculations based on the speed of ocean currents, his conjectures about transport on icebergs, they were all a colossal waste of time. For Croizat, the diversification of living things was driven by geologic and climatic processes that fragmented ancestral ranges. “Earth and life evolve together,” he famously wrote. That’s pretty much all there is.

  Even at the time, there were some serious weaknesses in Croizat’s argument. Two things, in particular, come to mind. First, his claim that there are very few fundamental tracks was misleading. The tracks of individual lineages on New Zealand, for instance, run all over the place—to New Guinea, New Caledonia, South America, Australia, Tasmania, and Southeast Asia, among other places. Certainly, if all New Zealand lineages had tracks connecting up to, say, southern South America plus Australia, that would argue strongly for origins by vicariance (via Gondwanan breakup) rather than oceanic dispersal, but that is simply not the case. Similarly, tracks for the Hawaiian Islands run to various parts of the Americas, Asia, Australia, New Zealand, and other Pacific islands. And the same pattern holds for virtually any sizable area on the planet. Croizat and his followers raised the reasonable point that a diversity of tracks is actually expected because landmasses are often made up of pieces with different geologic origins; for example, New Caledonia is an amalgamation of several different island arcs, and thus should contain lineages with different tracks. However, if any number of fundamental tracks can be accommodated under Croizat’s worldview, it becomes unclear whether the number of such tracks is a valid argument in his favor and against long-distance dispersal. If the signatures left by both vicariance and chance dispersal are a spider’s web of tracks, how do we choose one explanation over the other?

  Second, the fact that unrelated organisms with different means of dispersal share the same track doesn’t necessarily argue against long-distance dispersal; it is possible for rare, chance dispersal events to collectively produce a clear pattern. As a case in point, an island close to a continent would likely be colonized by many groups from that continent. Similarly, if a strong, directional ocean current flows between two landmasses, as is often the case, dispersal might give rise to many tracks connecting those two areas. Every one of these dispersal events could be something rare and unpredictable—rats on a tangle of vegetation or seeds inside the floating carcass of a bird—but, having followed the same path across the sea, the lineages involved would share the same fundamental track. What is haphazard at one level can be ordered at another: throw a dandelion seed, a beetle, and a monkey into the air in random directions during a hurricane, and they’ll all end up sailing away from you with the wind.

  In short, the pattern of fundamental tracks doesn’t necessarily have an obvious interpretation. The “facts” of distribution are not as easy to read as Croizat made them out to be.

  Until the mid-1970s, Croizat’s work was somewhat widely known among botanists, but had hardly been cited at all by zoologists. This lack of attention was significant because most of the influential biogeographers of the time—including dispersalists like George Gaylord Simpson, Ernst Mayr, and P. J. Darlington—were zoologists. Croizat and his defenders have argued that he was the victim of a “conspiracy of silence,” citing the fact that Simpson, Mayr, and others had read parts of his books (and, in Simpson’s case, had even corresponded with Croizat), but did not refer to him in their own publications. The counterargument is that these dispersalists had such a low opinion of Croizat’s work that they simply didn’t think it was worth bringing up. Mayr wrote privately that Croizat had a “totally unscientific style and methodology,” and Simpson went even further, dubbing Croizat “a member of the lunatic fringe.”

  In any case, in the 1970s, Croizat’s panbiogeography did become more widely discussed, thanks to a zoologist who was much more receptive to his ideas than the dispersalists were. That zoologist was none other than Gary Nelson, the American Museum ichthyologist and hub of the cladistic vicariance movement.

  Nelson first read Croizat in the 1960s while in graduate school, where, in his words, he “pondered for a time” the three volumes of Panbiogeography. However, that initia
l rumination doesn’t seem to have led anywhere. It was perhaps only after reading Lars Brundin’s monograph on the chironomid midges that Nelson began thinking about Croizat more deeply. Brundin, while noting (without explanation) that Croizat’s approach was “not wholly sound,” was nonetheless impressed by his “blazing sermon,” a sermon that emphasized range fragmentation and blasted the dispersalists. Given Nelson’s respect for Brundin, those words may have spurred him to take another look at Croizat.

  By the early 1970s, Nelson had soaked up Croizat’s message and had become convinced that the man was something of an overlooked visionary. In a 1973 paper, he wrote enthusiastically of Croizat’s work and indicated how it might dovetail with the emerging field of cladistic vicariance biogeography. He was especially impressed with Croizat’s single-minded search for general patterns, reflected in the discovery of fundamental tracks shared by diverse kinds of organisms, but thought that the methods of panbiogeography could be improved by taking into account the evolutionary relationships within each group. Instead of just connecting the dots of distribution, as Croizat had done, one could, in effect, analyze sets of evolutionary trees (actually, to be technically accurate, cladograms) of many different groups superimposed on distribution maps; using trees would provide more exact information about how those groups might have become fragmented in the course of their evolution. Basically, Nelson envisioned adding another piece to the vicariance revolution by merging Croizat’s search for generality with the explicit cladistic methods of Hennig and Brundin. The vicariance school would then contain both the big-picture approach of Croizat and the precision of cladistics.