SREL Reprint #1838





Tropical Mass Extinctions and the Scientific Method

John B. Dunning Jr.

Institute of Ecology, University of Georgia, Athens, GA 30602

Kangas (1992) argues that there is a paradox concerning the mass extinctions projected from current rates of tropical deforestation. He states that there is wide acknowledgment that the vast majority of species in the tropics are undescribed, thus the projected mass extinction will be dominated by the loss of unknown species. Kangas proposes, however, that this generates an unfaisifiable hypothesis: because we cannot prove the extinction of species that we do not know to exist, we cannot study the extinctions that we know to be occurring. Kangas concludes that the tropical mass extinction "will be a terrible loss," but it will be a loss that we cannot use the scientific method to study.

Kangas would be right if the extinction process can be studied only by specifically identifying each threatened species and then documenting its extinction. In fact, we do not need to know the identities of the threatened species. Instead, we can study the extinction process if we know the expected number of undescribed species that would go extinct with the loss of a particular area of tropical forest. If we knew the mean and variance of the expected number of undescribed species in 100 ha of tropical forest, for instance, then we could generate predictions as to how many undescribed species might be destroyed by the loss of specific tracts of forest. These predictions then could be tested as falsifiable hypotheses on the extent of extinction in a particular region.

SREL Reprint #1838

Dunning, J.B., Jr. 1994. Tropical mass extinctions and the scientific method. Bulletin of the Ecological Society of America 75:44-45.

To request a reprint