Discovering The Last Lost World

To produce a mighty book you must choose a mighty theme ... to include the whole circle of the sciences, and all the generations of whales and men and mastodons. Herman Melville, Moby-Dick

Career paths are notoriously unpredictable, and I never imagined that mine would lead me to focus on prehistoric megafaunal extinction. Beginning in 1948 at Rancho del Cielo, a cloud forest just within the Tropic of Cancer in eastern Mexico, I collected birds for ornithologist George M. Sutton and other vertebrates for the University of Michigan's Museum of Zoology. The best part was the forest of tall, dense trees rising to 100 feet: sweet gum, many species of oak unknown north of the border, Magnolia, Podocarpus, redbud, a few scattered palms, and a wealth of tank bromeliads, a home for frogs and lungless salamanders. Had I not suffered a handicap from a bout with polio in 1950, I might never have turned my attention to peat, rich in fossil pollen, being studied by botanists in the postglacial lakes around Ann Arbor. Then, in 1955, I learned from ecologist Ed Deevey at Yale University how to extract, identify, and count fossil pollen. From these counts one could learn what happened to plants after the glaciers melted away.

For Ed, the biogeography of the Pleistocene (the last ice age, 1.8 million years ago to 10,000 years ago), with all its glacial and interglacial changes in climate, was the key to understanding modern plant and animal distributions. Ed took cores of organic sediment from lake beds and counted samples of the fossil pollen, spores, and copepods (minute aquatic crustaceans) they contained. He could date these remains by

Willard F. Libby's then-new radiocarbon method; scientists could now refer to "Libby time" (roughly the last 40,000 years, the period for which radiocarbon dating is most effective). Magically, the fossil pollen record in sediments cored from New England lakes told of the comings and goings of treeless tundra and of spruce, fir, jack pine, and other trees as the climate warmed, the glaciers melted and on occasion readvanced, and eventually the ice-margin boreal vegetation yielded to today's deciduous forest. It even gave clues to the fate of animal species during this period. For example, fossil pollen counts plotted in percentages as a diagram associated with bones of mastodons indicated that they vanished around the time that, according to the fossil pollen counts, spruce gave way to pine. Some paleontologists thought that the change in tree cover from boreal conifers to temperate hardwoods might help explain mastodon extinction. Then they dropped the extinction question and returned to their primary interests, vertebrate anatomy, evolution, and geochronology. Nobody bothered to study the extinctions. Somehow the Pleistocene megafauna, big as it was, remained out of sight and out of mind.

In the winter of 1956, my wife and I found ourselves raising our children in a tenant farmhouse we rented from Anatole Cecyre, a French Canadian dairy farmer outside Chateauguay, Quebec. I commuted to a postdoctoral fellowship at the Université de Montréal, working on a pollen record of late-glacial climatic change and offering a seminar on Quaternary biology held jointly with McGill University.

Identifying and counting pollen grains can become monotonous. In a break from the microscope one day, I skimmed through George Gaylord Simpson's monumental Classification of Mammals, a long list of genera organized taxonomically. Malcolm McKenna and Susan Bell of the American Museum of Natural History have recently revised this classic tome. They recognize 5,158 mammalian genera, of which 4,075, about 80 percent, are extinct (McKenna and Bell 1997). The large number of mammal extinctions is to be expected, because the list embraces the end of the Mesozoic (the era that began about 250 million years ago) and the entire Cenozoic (the era that began about 65 million years ago and continues to today).

As a diversion that snowy subzero weekend, a diversion that fit right into the seminar on Quaternary biology, I began to plot all the late-Quaternary megafaunal extinctions listed by Simpson against those that had taken place earlier in the Cenozoic. After two days I was stunned by what I found. In the Miocene (starting 24 million years ago) and the Pliocene

(starting five million years ago), many mammals of all sizes turned over (i.e., evolved and went extinct). But at the end of the Quaternary, the pattern of extinctions in North America became very strange. It was the large terrestrial mammals that disappeared, and those for which radiocarbon dates were available did so suddenly.* Large marine mammals, on the other hand—the whales, dolphins, and pinnipeds—had been hard hit by extinctions in the Miocene and Pliocene but survived the Quaternary virtually intact. So did most of the continental small mammals (the shrews, moles, rats, and mice).

The major event in the Quaternary was the extinction of the large terrestrial mammals. Extinction also doomed their endemic species of parasites and commensals. (Commensals are species that benefit from accompanying other species without necessarily harming them.) For example, near-time extinction of internal parasites of ground sloths (Schmidt, Dus-zynski, and Martin 1992) or reduction in the number of species of cow-birds, magpie-type corvids, and dung beetles can be accepted as secondary, given the apparent dependence of these species on large mammals (Steadman and Martin 1984). Along with mammoths and ground sloths, the late-Quaternary extinctions also involved avian scavengers (such as condors), commensals (such as the Thick-knees, or Stone Curlews, of the Old World and tropical America), or guardians of the big mammals that eat their external parasites, such as tickbirds, which fly away if alarmed, alarming their host (Steadman and Martin 1984).

In the last 10,000 years, after most of the extinctions of the big mammals on continents, many small mammals, birds, reptiles, and land snails vanished. Small animal extinction and dwarfing, as well as extinction of some large mammals and birds, happened on oceanic islands, such as those in the West Indies and New Zealand, and on those in the Mediterranean. Thousands of small islands, especially in the remote Pacific, saw extinctions of small animals, especially birds and endemic species of land snails. The deep water surrounding these islands precluded any connection to each other or to the mainland, even when the sea level dropped

*John Alroy (1999, 2001) has looked at the fossil record in much greater detail. His analysis shows that late-Pleistocene extinctions in North America are quantitatively unlike any of the changes seen earlier in the Cenozoic. Only in the late Pleistocene is heavy extinction focused so strongly on large mammals.

Throughout this book, I call vertebrates large if they weigh over 45 kilograms (100 pounds), the size of a small adult human or an adult pronghorn. Some scholars define large mammals as those weighing over 1 kilogram (Alroy 1999, 2001). Just where the boundary is located does not alter the overall pattern: in near time on the continents, far more large mammals went extinct than small ones.

Figure 1. Map showing sequence of extinctions. Arrows indicate direction of human dispersals; numbers indicate order of human settlement. Adapted from Martin 1970, © American Institute of Biological Sciences.

by 400 feet or more, as it last did during the height of continental glaciation around 18,000 years ago.

In contrast, islands on the continental shelf, including Britain, Sri Lanka, Java, Sumatra, and Trinidad, were connected to continents when the sea level dropped.* Their faunas were much less vulnerable to prehistoric loss. Often much smaller than their continental relatives, the animals that evolved on oceanic islands included the dwarf mammoths of Santa Rosa and San Miguel islands off the California coast; in the Mediterranean, the dwarf elephants of Crete and the dwarf hippo of Cyprus; in the West Indies, the dwarf ground sloths of Cuba and Hispaniola (Haiti and the Dominican Republic). The tropical islands of the remote Pacific, some of them quite small, hosted 2,000 taxa of flightless rails. Apparently depending on when humans (often accompanied by Pacific rats) first arrived, many of the island endemics suffered prehistoric extinction. Pacific

*Islands surrounded by shallow water of the continental shelves, less than 120 meters (400 feet) deep, would not emerge long enough to evolve highly endemic species. These shelf islands disappear as rising interglacial sea levels shrink and eventually drown them. Islands artificially formed by impoundment of rivers experience comparable extinctions of larger animals and artificial increases in smaller ones. In contrast, deep-water islands could be colonized only by species surviving water transport or by ancient detachment from a continent such as Gondwanaland. They support faunas that are impoverished but rich in endemics, such as Jamaica's extinct giant rodents, Madagascar's extinct giant lemurs, and Sulawesi's extinct dwarf elephants.


Eon (543 mya to the present)


Era (65 mya to the present)

Quaternary (1.8 mya to the present)

Holocene (10,000 years ago to the present)

North American Land Mammal Ages

Was this article helpful?

0 0

Post a comment