Premium
Periodicity in Extinction: The Role of Systematics
Author(s) -
Patterson Colin,
Smith Andrew B.
Publication year - 1989
Publication title -
ecology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.144
H-Index - 294
eISSN - 1939-9170
pISSN - 0012-9658
DOI - 10.2307/1941349
Subject(s) - extinction event , extinction (optical mineralogy) , paleontology , monophyly , biology , ecology , permian–triassic extinction event , zoology , clade , phylogenetics , demography , biological dispersal , population , biochemistry , sociology , gene
The case for periodicity in extinction, with a mass extinction every 26 x 10 6 yr since the mid—Permian, rests on analyses of the fossil record of marine families and genera. We have checked the echinoderm and fish family extinctions which make up °20% of the data, and find that only 25% of our sample is signal (extinctions of monophyletic groups correctly dated to stage). The 75% noise component includes pseudoextinctions of non—monophyletic groups, and spurious data from other sources. The signal/noise ratio is virtually the same in echinoderms and fishes. Conspicuous peaks in our total family data correspond to five of the eight periodic extinction peaks that have been identified, but these five peaks are a feature of the noise component, not of the signal. We have also checked a randomly selected 20% of the generic extinctions in echinoderms and fishes since the Permian. The proportion of our data reflecting plausible extinctions is much lower among genera (31%) than among families (47%), whereas the proportion that samples peaks of fossilization rather than of extinction is much higher among genera (32%) than among families (16%). As with our families, peaks corresponding to periodic extinctions are restricted to the noise component of the generic data, but here the signal component is too small to bear much significance. We question the expectation that studies of global extinction will be more reliable the lower taxonomic level investigated.