Tracking the Course of Evolution


by Richard Cowen

NOTE: This is page 1 of a three-page document.

THIS ESSAY, written in 1999, is a chapter from my book History of Life, published by Blackwell Science, Boston, Massachusetts, 2000. © Richard Cowen. You may print out a copy for personal or educational use, and you may link to this site. Illustrations are missing from this web version of the chapter.
Cowen, R. 1994. History of Life. 2nd edition. 460 pp. Blackwell Scientific Publications, Cambridge, Massachusetts. This is a freshman-level textbook published by Blackwell Science. Copyright Richard Cowen 1994. Available from Blackwell Science, 238 Main Street, Cambridge, Massachusetts 02142, telephone 800-215-1000. Information and updates on the 3rd edition.
See also a separate essay devoted to the Cretaceous-Tertiary extinction, and for an outline of Richard Cowen's oral presentation.
At the Geology Department at the University of California, Davis, Richard Cowen tries to maintain other Web pages of interest:


Extinction happens all the time. Some species have small populations that depend on a particularly narrow range of food, or habitat, and are vulnerable to even small-scale ecological disturbance. Strictly speaking, if the world has fairly even diversity through time, existing species will become extinct about as often as new species evolve. Extinction is the expected fate of species, rather than a rarity.
However, there are time of extremely rapid extinction, just as there are times of extremely rapid evolutionary radiation. Sepkoski's data of diversity through time show dramatic drops in diversity (Figures 5.7, 5.11). Times of rapid extinction require us to ask whether some special extinction mechanism is at work, exactly the reverse of the questions that are raised by great evolutionary radiations.
We have to proceed in two steps. First, we have to ask whether there really are times of unusually rapid extinction. Then we have to identify the causes for those special events. Were they processes that were just extreme examples of normal (ordinary) Earth processes, or were they catastrophic events that were truly extraordinary?

Mass Extinctions

David Raup and Jack Sepkoski analyzed Sepkoski's data on the fossil record (Chapter 5). They identified extinction events that seemed to be very large, large enough to be called mass extinctions (Figure 6.1). Six mass extinctions have been recognized in a nonquantitative way for decades, and they and others have been studied in some detail:
  • At the end of the Ordovician
  • At the end of the Frasnian stage of the Late Devonian ("F-F")
  • At the end of the Permian period, ? a double event ("P-Tr")
  • At the end of the Triassic period
  • At the end of the Cretaceous period ("K-T")
These six mass extinctions are alike in some ways. They all represent the extinction of a significant component of global faunas and/or floras, and they are all relatively sudden in geological terms: that is, they occurred over a few million years at most, and in some cases much less than that.


Extinctions are followed by recovery, and the patterns of biological and ecological recovery are a new field of study. The larger the extinction, the longer the recovery time, and the more the global ecosystem changes across the event. The rise of the Modern Fauna begins in the Triassic, and the rise of mammals begins in the Paleocene. Groups sometimes show amazing radiations: all living sea-urchins are descended from two Paleozoic genera that survived the P-Tr extinction, for example.
During recoveries we sometimes see Lazarus taxa: plants or animals that disappeared at the extinction and are missing from the fossil record for a long time, sometimes millions of years. Corals, for example, are absent from the fossil record of the Lower Triassic, after the P-Tr extinction, and re-appear in the Middle Triassic. The existence of Lazarus taxa implies that somewhere there must have been refuges for scattered survivors; only when they radiate away from the refuges do they appear again in the rocks that we have collected.

Explaining Mass Extinction

Mass extinctions were global phenomena, so they have to be explained by global processes. The first that comes to mind is plate tectonics. Plate tectonic changes could alter global diversity patterns (Chapter 5). However, tectonic changes are relatively slow in geological terms, while mass extinctions stand out because they are relatively sudden. If we are to use plate-tectonic explanations, we have to add a plausible mechanism for a trigger that suddenly fires a deadly "extinction bullet."
If plate-tectonic arguments (with trigger mechanism) do not work, we must look to more extreme suggestions. Some plausible extinction agents that have been examined at various times are:
  • A failure of normal ocean circulation affects ocean chemistry enough to cause global changes in climate and atmosphere.
  • A rapid change in sea level affects global ecology and climate.
  • An enormous volcanic eruption.
  • An extra-terrestrial impact by an asteroid.
Only the last is extraterrestrial in nature. The K-T mass extinction certainly occurs at the same time as a major asteroid impact, so I shall discuss the evidence for impact, and its likely consequences, elsewhere. For the purposes of this essay, it is important to remember that an asteroid impact large enough to cause a global impact should leave behind physical evidence in the form of a sharply defined layer or spike of the element iridium, and/or impact-generated glass spherules called tektites, and/or quartz crystals with characteristic "shock marks," in the rocks associated with the extinction.
I shall deal with mass extinctions in time order, except for the largest of them all, the Permo-Triassic (P-Tr) extinction, which deserves a more extended discussion of its own.

The mass extinction at or near the end of the Ordovician has no iridium spike associated with it, and seems to be closely linked with a major climatic change. A first pulse of extinction happened as a big ice age began, and the second occurred as it ended. Some paleontologists feel that as we collect fossils from more regions of the world, this "mass extinction" may turn out to have been a comparatively minor event.

A mass extinction took place, probably in several separate events, at the boundary between the last two stages of the Devonian, the Frasnian and Famennian (the F-F boundary). There was a major worldwide extinction of coral reefs and their associated faunas, and many other groups of animals and plants were severely affected too. Iridium anomalies, shocked quartz, and glass spherules are reported from China and Western Europe at or near the F-F boundary. However, there are also indications of climatic changes, and major changes in sea-level and ocean chemistry, at the same time. Carbon isotope shifts indicate that global organic productivity was changing rapidly before the boundary.
George McGhee favors an impact scenario. There are three cautions. First, McGhee makes it clear that he always has favored this scenario; second, the geological evidence requires that there were several closely-spaced but medium-sized impacts over perhaps two or three million years, rather than the one tremendous impact that occurred at the K-T boundary; and third, the evidence is incomplete in terms of our understanding of timing, of world geography at the time, and there are difficulties in going from evidence to interpretation. Kun Wang suggests that global ecosystems were already stressed when an impact occurred. There is no "magic marker" of impact phenomena at the extinction event, as there is at the K-T boundary, and that makes the F-F boundary very difficult to work with.

The Triassic-Jurassic boundary marks a turnover of groups on land and in the sea. In 1999 a team of geologists reported that a gigantic eruption took place around this time, in a volcanic episode that marked the first major plate tectonic activity that began to split the Atlantic Ocean (Chapter 5). Critics pointed out that the eruption seemed to have occurred slightly after the boundary, so could not have caused any event at the boundary. To complicate the issue, shocked quartz has been discovered very near the boundary in northern Italy (the large Manicouagan impact crater in Québec, Canada, was formed at some uncertain date around this time). Several craters dated to the late Triassic line up, as if they had been formed by an incoming body that fragmented at the last moment. However, it's not clear whether the craters were formed simultaneously, or if so, whether they were formed at the same time as the extinction. It's not clear that the extinction was large. More research is needed: in fact, it's not clear what the evidence is, let alone the answer!

An extinction at the Eocene-Oligocene boundary affected marine organisms more than terrestrial animals. It occurred at a time of major climatic change, not obviously linked with any single major impact or any major volcanic event. There were impacts around this time. One struck the east coast of the United States, scattering tektites from a crater that now lies under Chesapeake Bay. Another struck central Siberia, with debris scattered perhaps to Europe. These were clearly separate events. Although it is premature to link them to the extinctions, impacts are not ruled out as contributing to the E-O event.

The extinction at 250 Ma, the end of the Permian, is the largest of all time: the "Mother of Mass Extinctions" according to Douglas Erwin. The extinction was used by John Phillips 150 years ago to define the end of the Paleozoic Era and the beginning of the Mesozoic (Figure 5.8). An estimated 57% of all families and 95% of all species of marine animals became extinct. The Paleozoic Fauna was very hard hit, losing especially suspension feeders and carnivores, and almost all the reef dwellers. The Permo-Triassic (P-Tr) extinction is a major watershed in the history of life on Earth, especially for life in the ocean; the K-T extinction is small in comparison (Figure 5.11).
The P-Tr extinction was rapid, probably taking place in less than a million years, and possibly much faster than that. Although it was much more severe in the ocean, it affected terrestrial ecosystems too. A prolific swamp flora in the Southern Hemisphere had been producing enough organic debris to form coals in Australia, but the coal beds stop abruptly at the P-Tr boundary. No coal was laid down anywhere in the world for at least 6 million years afterward. A large change in carbon isotopes occurred across the P-Tr boundary, which signifies an important and global drop in photosynthesis that lasted a long time.
There is no evidence for an impact at the P-Tr boundary. The continental collisions that formed Pangea in the Permian would account for a major drop in diversity (Chapter 5) but not for a sudden, enormous mass extinction. Perhaps most important of all, the Permian extinction coincides with the largest known volcanic eruption in Earth history.

CONTINUE to page 2

Return to top of page   |   TRACKING HOME   |   PRESENTATIONS