What is a Mass Extinction?

One of the main background themes in my history of life series was that extinctions will always happen – they’re a natural part of the biosphere’s evolution. But I never really explained what a mass extinction is. For example, the case of mayflies, who emerge as sexually mature adults simultaneously and live between a few minutes to a week before dying en masse is obviously not a case of mass extinction, even though there is a lot of death involved.

One intuitive definition for a mass extinction is by one of the revolutionisers of their study, Jack Sepkoski, who wrote the following in Sepkoski (1986):

A mass extinction is any substantial increase in the amount of extinction (i.e. lineage termination) suffered by more than one geographically wide-spread higher taxon during a relatively short interval of geologic time, resulting in an at least temporary decline in their standing diversity.

This is relatively vague. Any event where there is a sudden global spike above the background extinction rate is termed a mass extinction. This background rate is the regular cycle of extinction (balanced out by the regular cycle of speciation). For example, the average rate for mammals is 0.2 – 0.5 extinctions/1 million species and 1 million years. For an event to count as a mass extinction, it must quickly (geologically) kill off more than than that number (no idea how many mammals there are, it’s an easy calculation), all around the world, and it must also affect other taxa in the same way. Otherwise, it’s just a local disaster (e.g. a volcano farting a cloud of CO2 and killing everything living around it).

But given just how many ways there are for mass extinctions to be caused (see the basics of the Ordovician, P-T, K-T and Holocene extinctions), this definition falls rather short, as would any other definition. A too-broad definition will erroneously include other phenomena, while a too-narrow definition will miss out on certain events that should be called mass extinctions (the Big 7 mass extinctions are undoubted, but there are many excursions above the background extinction rate – when should they count?). So instead of attempting a definition, a better way is to make a list of the common points shared by those undoubted mass extinctions. This is what Levinton (2001) did in chapter 7 of the book, and I summarise here:

  • The number of extinct taxa is much higher than regular extinctions.
  • The event happens very suddenly and quickly in geological time.
  • The event affects several taxa across the phylogenetic and geological age spectrum.
  • The event affects several ecosystems and biomes, not necessarily in the same way.
  • The event has a geographically wide reach, often global, but not necessarily so.
  • The cause of the mass extinction differs from the cause of the background extinction, qualitatively and/or quantitatively.
  • The cause of mass extinctions is independent of the cause of background extinctions, although the cause of the background extinctions may exacerbate the mass extiction.
  • The recovery from mass extinctions proceeds by the spreading of previously non-dominant taxa (e.g. mammals after dinosaurs) or of new taxa (e.g. archosaurs after the P-T), or by restructuring of ecosystems (e.g. post-Ordovician).

If an event fulfills these criteria, then it is a mass extinction.

This is but an introduction into the topic. For much more info on mass extinctions and an overview of Phanerozoic mass extinctions and candidate mass extinctions, see this paper:

Bambach, R. (2006). PHANEROZOIC BIODIVERSITY MASS EXTINCTIONS Annual Review of Earth and Planetary Sciences, 34 (1), 127-155 DOI: 10.1146/annurev.earth.33.092203.122654


Levinton JS. 2001. Genetics, Paleontology, and Macroevolution.

Sepkoski JJ. 1986. Phanerozoic overview of mass extinction. In: Raup DM & Jablonski D (eds.). Patterns and Processes in the History of Life.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.