Not so much a bang as a lot of lava

The cover of this small book shows a T. Rex skull evidently roaring in defiance at Dante's inferno. All very evocative of the controversy that surrounds the mass extinction of the dinosaurs at the close of the Cretaceous Period, about 65 million years ago. Vincent Courtillot's book explores this particularly hot issue, as well as the apparently repeated re-setting of the fossil tape recording the history of life on this planet following the condensation of Earth from interstellar gases about 4.5 billion years ago.

It is probably safe to say that the majority of readers already "know" that a sizeable meteorite impact at the end of the Cretaceous Period was responsible for the extinction of the dinosaurs (among many other types of animal). The tabloids tell us this reasonably regularly, as do television programmes, in addition to the international discussions that have taken place in recent years to decide on an appropriate strategy to deflect the next huge meteorite that might threaten our own existence. Meteorites and comets are out there - the surface of the Moon is liberally sprinkled with impact craters (we all saw Hale-Bopp's impact with the surface of Jupiter); and in recent years much work has indicated the presence of the "smoking gun" - a timely impact crater in the Caribbean, an iridium anomaly, shocked quartz: the geophysical litany is indeed impressive.

So do we need this book? It is indeed reminiscent of David Raup's The Nemesis Affair (1986), which reviewed the evidence for an extraterrestrial impact and tried to tie the periodicity of mass extinctions on Earth to the eccentric orbit of "Nemesis", a hypothetical companion star to our sun.However, this book does differ. It attempts to stand back from the debate and dispassionately assess the evidence of the impact theory. I stress the word "attempt" because it is clear that Courtillot is quite vocally sceptical about the evidence for any purely impact-driven extinctions. He is, after all, a volcanologist.

The author of Evolutionary Catastrophes is a very senior French geologist who has been involved in the study of the structure (tectonics and volcanism) of the Earth for a number of decades. He is in a position to reflect on the degree of progress that has been made in recent years as part of our attempt to try to understand more precisely what caused, or perhaps triggered, such cataclysms. The book is, I believe, intended to be an accessible general review and contribution to the public understanding of science and represents an updated English translation of Vie en Catastrophe , which was published in 1995.

Dust-jacket comments tend to draw me like a magnet when I pick up any new book. This is a silly habit, but I am always interested to see whether they contain comments from colleagues. I was, for example, intrigued to see Raup, an advocate of impact models, who has contributed much to the debates in this field, informing me that the book represents a "... superb job!... obligatory reading for all professionals involved in the controversies surrounding the causes of mass extinction".

The book opens (much as did Raup's) by considering the history of debate about the ages and nature of the history of the Earth, and the traditional philosophical divide that separated scientists in the late 18th century as much as it does today. For example, across the 18th and 19th century, dominant scientists such as Cuvier, Buckland and Agassiz advocated periods of quiescence interrupted by major disruptions in the history of life - such major changes allowed us to divide up the fossil record quite effectively. In contrast, at the same time Lamarck, Hutton and Lyell advocated continuity rather than disruption in the history of life. We then learn about the age of the Earth and charted changes in the diversity of life over time - indicating that there appear to have been a series of crises or extinction events of differing intensity over time. The extinction that marked the disappearance of the dinosaurs (the K-T event) is the most recent event in geological time, and because it is nearest to the present day is likely to have the greatest amount of evidence. For this reason, the K-T is discussed first by examining a few of the lines of evidence that can be used (how accurate is the evidence concerning the time of extinction - was it instantaneous, or spread over thousands or millions of years?), and then by exploring some of the hypotheses that have been put forward (whether they can be consistently applied to dinosaurs, ammonites, foraminiferans, etc). What the author is partly drawing you towards is a realisation that one person's catastrophe can be somebody else's gradual event.

Courtillot then hits us with much of the weight of evidence that followed Walter Alvarez's (1980) identification of remarkably high levels of the element iridium in a thin band of clays that seemed to mark the K-T boundary layer. The iridium concentration could, to his mind, be explained only if its source was extraterrestrial - and contained within a vaporised meteor. From this brilliant induction comes a scenario relating the extinction at the end of the Cretaceous to the disruption and/or collapse of global ecosystems that resonated remarkably well with the nuclear-winter debates of the time.

Since that time, a formidable wall of supporting evidence has accumulated in the form of impact-droplet distribution around the globe, shocked-quartz grains, soot and charcoal deposits, and finally the identification of the deep impact structure at Chicxulub. With this apparently overwhelming case building in favour of an extraterrestrial cause for the extinction, the author then leads the reader on a circuitous path in his preferred direction - towards an intrinsic, rather than extrinsic model for extinctions, by focusing on periods of intense volcanic activity in the history of the Earth.

During the early 1980s, in the face of the evidence that was building for the impact scenario, a number of workers persisted in considering volcanism as a potential cause of mass extinctions. They focused mainly on the evidence from the Deccan of India - vast laval outpourings (measured in hundreds of thousands of cubic kilometres) that emerged at the precise time of the K-T event. What appears from an extended discussion of these events is that many of the anomalies that have been tied exclusively to the impact scenario (the iridium anomaly, in part at least, melt droplets and even, he suggests, shocked quartz) can be explained by explosive mantle-plume-mediated volcanism; this can be tied to India's moving (tectonically) across an Indian Ocean "hot-spot" (representing the top of one of these plumes) at the close of the Cretaceous period.

As a result, Courtillot builds a plausible case for an alternative scenario for the end - Cretaceous extinction, which linked climate and ecosystem collapse to the emergence of flood-basalts at this time in Earth's history. What was the cause, and what is the degree of support for his view? He argues that several (seven) of the significant mass extinctions in earth's history coincide with times of intense volcanic activity; he links these latter with the Earth's rotation, stochastic turbulence in its liquid core (reflected in magnetic polarity changes over time) and subsequent disruptions in its mantle. By contrast, only one event (by his reckoning), the K-T event, can be linked clearly to an extraterrestrial impact - by 7-1 you see where his bet lies.

Read and enjoy this book, even if you do not believe it all. It represents a pleasantly argued counterpoint to the rather shrill and abrasively dismissive "impacticist" views.

David Norman is director, Sedgwick Museum, University of Cambridge.