Logo IMG


Setting the Record Straight on the Refugia Hypothesis

Paul Baker

AMAZON EXPEDITIONS: My Quest for the Ice-Age Equator. By Paul Colinvaux. xx + 328 pp. Yale University Press, 2008. $32.50.

Anyone who has undertaken scientific research knows that pursuit of "the truth" is seldom straightforward. This has certainly been the case for ecologist Paul Colinvaux, as he describes in his book Amazon Expeditions: My Quest for the Ice-Age Equator. The story is part picaresque and part polemic, at once a tale of adventure and a firsthand account of the deconstruction of a "beautiful" theory.

The adventure story recounts the details of many expeditions over the course of half a lifetime by the author and his students to tropical South America, including the Galápagos Islands, Ecuador and Brazil. The purpose of their travels was to search for lakes containing mud dating from the last glacial maximum (the time when the ice sheets were at their greatest extent, about 20,000 to 25,000 years ago) or earlier. Such sediment, they believed, would contain clues about the nature of climate and vegetation during a key period of Earth's history.

Their goal turned out to be surprisingly elusive, especially in the Amazon basin, where throughout the ages migrating rivers have been very efficient at erasing traces of the distant past from the surface of the landscape. Of course, in the days before satellite imagery was widely available, finding lakes in the dense rain forest was not easy. Reaching these lakes with heavy coring equipment was even more challenging. Recounting these experiences with wry humor, ample ego and his signature enthusiasm, Colinvaux is at his best.

The beautiful theory he sought to disprove was the refugia hypothesis, put forward by Jürgen Haffer. In an article in the July 11, 1969, issue of Science, Haffer theorized that

during several dry climatic periods of the Pleistocene and post-Pleistocene, the Amazonian forest was divided into a number of smaller forests which were isolated from each other by tracts of open, nonforest vegetation. The remaining forests served as "refuge areas" for numerous populations of forest animals, which deviated from one another during periods of geographic isolation. The isolated forests were again united during humid climatic periods when the intervening open country became once more forest-covered, permitting the refuge-area populations to extend their ranges. This rupturing and rejoining of the various forests in Amazonia probably was repeated several times during the Quaternary and led to a rapid differentiation of the Amazonian forest fauna in geologically very recent times.

The refugia hypothesis thus spliced together disparate subjects, including paleoclimatology, biogeography and the concept of allopatric speciation due to vicariance (the prevention of genetic exchange by some sort of barrier), which had been recently amplified by Ernst Mayr. The hypothesis gained wide acceptance in the years that followed.

Eventually, though, Colinvaux's studies of the vegetation history of the Amazon discredited Haffer's theory. The history of the theory, Colinvaux's own refutation of it, and the proponents' counterattacks make both edifying and compelling reading.

But in setting the record straight on a flawed thesis, has Colinvaux gone too far the other way? He insists that the Amazon's lowland rain forest has persisted uninterrupted since well before the Pleistocene "in all the places where it grows now" and that "it was never fragmented nor replaced by different kinds of vegetation." Is this just a different flawed thesis? The answer is, simply, we don't yet know.

Although we are still far from understanding why the tropics have such great biodiversity, molecular phylogenetic studies, despite their imperfections, are starting to provide important clues. From what we know now, it is apparent that many plant and animal species of the neotropics originated prior to the ice ages that began roughly 2.5 million years ago. However, many other species first appeared during the Pleistocene (from about 2.5 million to about 11,500 years ago). Thus, the known ages of species origins in the neotropics do not exclude the possibility of significant Pleistocene diversification as put forward by Haffer.

Whether these species originated in isolated refugia is a separate question, as is the matter of whether such refugia even existed. In tropical Africa, much evidence indicates that Pleistocene aridity did bring about a great increase in the extent of savannah at the expense of rain forest. Scientists doing research there thus are less resistant to the notion of species-rich Pleistocene forest refugia. Their disputes focus on the ages and locations of refugia, not the very existence of the refugia.

Unfortunately, paleoclimatic studies of the neotropical Pleistocene have produced less-definitive results than have similar studies in Africa. Although there is little doubt that Amazon surface air temperatures were significantly colder during the many glacial stages of the Pleistocene than they were during the intervening interglacial periods, we still do not know the timing and magnitude of past precipitation changes in the Amazon region. The studies of Colinvaux and his collaborators established that at the time of the last glacial maximum the Amazon was wet enough to retain a rain forest along the equator. In other regions of tropical South America, however, there is evidence for great variations in precipitation, in some cases accompanied by evidence for significant changes in flora. Maximum drying occurred in the southern tropics during warm interglacial stages—not during cold glacial stages, as many scientists previously assumed. (Incidentally, Haffer did not stipulate in his original hypothesis that the glacial stages were the arid periods.)

At the very least, it seems likely that Pleistocene climate variations, coupled with known major changes in atmospheric CO2 levels, forced changes in the composition and nature of Amazon forest communities, bringing about the extinction of some species, necessitating migration of some species, and resulting in the isolation and evolutionary divergence of some species. Yet in the equatorial Amazon there is no evidence for the sort of fragmentation of the forested landscape during the Pleistocene that is envisioned by the refugialists.

In sum, the evidence is mixed. Colinvaux has his widespread forest continuity and significant glacial-stage temperature decreases; the refugialists have their Pleistocene diversification and a likely significant interglacial drying. Haffer and his supporters, along with Colinvaux and his, have advanced our understanding of the history of the tropics and the origin of tropical diversity. And now, Colinvaux has given us a most entertaining behind-the-scenes account of his challenging quest and worthy accomplishments.

Paul Baker is professor of geochemistry at Duke University. For many years he has studied the paleoclimate of tropical South America.

comments powered by Disqus

Connect With Us:


Sigma Xi/Amazon Smile (SciNight)

Subscribe to Free eNewsletters!

RSS Feed Subscription

Receive notification when new content is posted from the entire website, or choose from the customized feeds available.

Read Past Issues on JSTOR

JSTOR, the online academic archive, contains complete back issues of American Scientist from 1913 (known then as the Sigma Xi Quarterly) through 2005.

The table of contents for each issue is freely available to all users; those with institutional access can read each complete issue.

View the full collection here.


Of Possible Interest

Book Review: On the Origin of Origin Stories

Book Review: O Pioneer

Book Excerpt: Cypherpunks Write Code

Subscribe to American Scientist