A Series of (Random) Unfortunate Events? Distributions of Extinction Events through the Cenozoic.

1
A Series of (Random) Unfortunate Events?
Distributions of Extinction Events through the
Cenozoic.
John Kloke, Pomona College, Claremont CA
91711 Jennifer Stempien, University of Colorado
at Boulder, Boulder, CO 80309 K. Rebecca Thomas,
Montgomery College, Rockville MD 20850
Introduction
Results
Estimates based on the fossil record suggest the
majority of animals and plants that have ever
existed on Earth are now extinct. Numerous
studies have focused on determining the spacing
and intensity of both mass and background
extinction rates through the entire Phanerozoic
(545 Million years ago to present) to suggest
potential causes of extinction and the effects of
extinction on evolutionary patterns. Some of
these studies have suggested a cyclic nature of
26 million years for background extinction
patterns for the past 250 million years (Raup and
Sepkoski, 1984), however this pattern has been
seen as controversial and no conclusive causative
factors have been suggested for a cyclic pattern.
Despite the controversy of a cyclic pattern for
background extinctions, present focus of
attention has been on investigating the intensity
of extinctions to one another and not if the
extinctions are distributed randomly across the
temporal scale.
Are extinction events for genera distributed
randomly during the Cenozoic?
Ho Extinction events are distributed randomly
with respect to time in the Cenozoic.
Table 1 Extinction rates for all genera in
dataset.
EPOCH AGE (mya) EXTINCTIONS / MILLION YEARS
Pleistocene 1.8 50.6
Pliocene 5.3 128.9
Miocene 23 69.3
Oligocene 33.9 43.3
Eocene 55.8 62.6
Paleocene 65.5 46.2
max 128.9 p 0
Materials and Methods
Dataset The number of genera that had their
last occurrence in each epoch in the Cenozoic (65
million years to 10,000 years ago) were acquired
from Sepkoski et al. 1981, available from the
following website http//strata.ummp.lsa.umich.e
du/jack/. Time scale The Geologic Time Scale,
1999, Palmer and Geissman, The Geological
Society of America, Boulder CO.
http//www.geosociety.org/ Method Extinction
rates per million years were calculated for each
epoch by dividing the number of genera that went
extinct by the length of the epoch (Table 1).
The maximum extinction rate among the epochs was
used as the test statistic. A distribution of
maximum extinction rates was generated by
randomly distributing 4059 extinction events
between 0 and 65.5 million years for 10,000
repetitions. This distribution was used to test
whether the maximum extinction rate found in the
data (Pliocene) was significantly different than
that expected by chance. This procedure was
repeated for genera within five separate phyla
(Table 2).
Table 2 Maximum extinction rates for genera by
phylum.
PHYLUM MAXIMUM EXTINCTION RATE (Epoch) P-VALUE
Arthropoda 11.7 (Pliocene) 0.0002
Brachiopoda 3.4 (Pliocene) 0.0042
Chordata 16.9 (Pliocene) 0.0074
Echinodermata 5.5 (Eocene) 0.7139
Mollusca 70.6 (Pliocene) 0
Conclusions

• We reject the null hypothesis that extinction
  events for all genera are randomly distributed in
  the Cenozoic.
• We reject the null hypothesis that extinction
  events for genera within Arthropoda, Brachiopoda,
  Chordata, and Mollusca are randomly distributed
  in the Cenozoic however, we cannot reject the
  null for Phylum Echinodermata.
• For all phyla in which extinction events were
  nonrandomly distributed, maximum extinction rates
  were in the Pliocene. This is consistent with
  individual case studies exploring diversity
  patterns in the Pliocene.
• Future studies may further explore these patterns
  and their potential causes at a variety of
  temporal, spatial and faunal resolutions.