Evolutionary and preservational constraints on origins of biologic groups: Divergence times of eutherian mammals

Science

Volumn 283, Issue 5406, 1999, Pages 1310-1314

  Foote, Mike ^a   Hunter, John P ^b   Janis, Christine M ^c   Sepkoski Jr , J John ^a  

^a UNIVERSITY OF CHICAGO   (United States)

^b NEW YORK INSTITUTE OF TECHNOLOGY   (United States)

^c BROWN UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EVOLUTION;

ARTICLE; BIODIVERSITY; EVOLUTION; FOSSIL; NATURAL SELECTION; NONHUMAN; PHYSICAL ANTHROPOLOGY; PRIORITY JOURNAL; SURVIVAL; TAXONOMY;

NASA DISCIPLINE EXOBIOLOGY; NON-NASA CENTER;

ANIMALS; EVOLUTION; FOSSILS; MAMMALS; MATHEMATICS; MODELS, BIOLOGICAL;

EUTHERIA; MAMMALIA;

EID: 0033605215     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.283.5406.1310     Document Type: Article

References (87)

1. E. Zuckerkandl and L. Pauling, in Evolving Genes and Proteins, V. Bryson and H. J. Vogel, Eds. (Academic Press, New York, 1965), pp. 97-165; W. M. Fitch, in Molecular evolution, F. J. Ayala, Ed. (Sinauer, Sundertand, MA, 1976), pp. 160-178; M. Kimura, J. Mol. Evol. 26, 24 (1987).
2. E. Zuckerkandl and L. Pauling, in Evolving Genes and Proteins, V. Bryson and H. J. Vogel, Eds. (Academic Press, New York, 1965), pp. 97-165; W. M. Fitch, in Molecular evolution, F. J. Ayala, Ed. (Sinauer, Sundertand, MA, 1976), pp. 160-178; M. Kimura, J. Mol. Evol. 26, 24 (1987).
3. E. Zuckerkandl and L. Pauling, in Evolving Genes and Proteins, V. Bryson and H. J. Vogel, Eds. (Academic Press, New York, 1965), pp. 97-165; W. M. Fitch, in Molecular evolution, F. J. Ayala, Ed. (Sinauer, Sundertand, MA, 1976), pp. 160-178; M. Kimura, J. Mol. Evol. 26, 24 (1987).
4. B. Runnegar, Lethaia 15, 199 (1982); C. A. Wray, J. S. Levinton, L. M. Shapiro, Science 274, 568 (1996); F. J. Ayala, A. Rzhetsky, F. J. Ayala, Proc. Natl. Acad. Sci. U.S.A. 95, 606 (1998);
5. B. Runnegar, Lethaia 15, 199 (1982); C. A. Wray, J. S. Levinton, L. M. Shapiro, Science 274, 568 (1996); F. J. Ayala, A. Rzhetsky, F. J. Ayala, Proc. Natl. Acad. Sci. U.S.A. 95, 606 (1998);
6. B. Runnegar, Lethaia 15, 199 (1982); C. A. Wray, J. S. Levinton, L. M. Shapiro, Science 274, 568 (1996); F. J. Ayala, A. Rzhetsky, F. J. Ayala, Proc. Natl. Acad. Sci. U.S.A. 95, 606 (1998);
7. A. Seilacher, P. K. Bose, F. Pflüger, Science 282, 80 (1998). For animal phyla, the gap may be partly explained by a major change in preservability corresponding with the advent of skeletons.
8. A. Cooper and R. Fortey, Trends Ecol. Evol. 13, 151 (1998).
9. S. Kumar and S. B. Hedges, Nature 392, 917 (1998).
10. S. B. Hedges et al., ibid. 381, 226 (1996); M. S. Springer, J. Mammal. Evol. 4, 285 (1997).
11. S. B. Hedges et al., ibid. 381, 226 (1996); M. S. Springer, J. Mammal. Evol. 4, 285 (1997).
12. K. H. Wolfe et al., Proc. Natl. Acad. Sci. U.S.A. 86, 6201 (1989); W. Martin, A. Giert, H. Saedler, Nature 339, 46 (1989); P. R. Crane et al., ibid. 342, 131 (1989); W. Martin et al., Mol. Biol. Evol. 10, 140 (1993); K. J. Systma and D. A. Baum, in Flowering Plant Origin, Evolution and Phylogeny, D. W. Taylor and L. J. Hickey, Eds. (Chapman & Hall, New York, 1996), pp. 314-340; J. A. Doyle and M. J. Sanderson, Am. J. Bot. 84 (suppl.), 132 (1997).
13. K. H. Wolfe et al., Proc. Natl. Acad. Sci. U.S.A. 86, 6201 (1989); W. Martin, A. Giert, H. Saedler, Nature 339, 46 (1989); P. R. Crane et al., ibid. 342, 131 (1989); W. Martin et al., Mol. Biol. Evol. 10, 140 (1993); K. J. Systma and D. A. Baum, in Flowering Plant Origin, Evolution and Phylogeny, D. W. Taylor and L. J. Hickey, Eds. (Chapman & Hall, New York, 1996), pp. 314-340; J. A. Doyle and M. J. Sanderson, Am. J. Bot. 84 (suppl.), 132 (1997).
14. K. H. Wolfe et al., Proc. Natl. Acad. Sci. U.S.A. 86, 6201 (1989); W. Martin, A. Giert, H. Saedler, Nature 339, 46 (1989); P. R. Crane et al., ibid. 342, 131 (1989); W. Martin et al., Mol. Biol. Evol. 10, 140 (1993); K. J. Systma and D. A. Baum, in Flowering Plant Origin, Evolution and Phylogeny, D. W. Taylor and L. J. Hickey, Eds. (Chapman & Hall, New York, 1996), pp. 314-340; J. A. Doyle and M. J. Sanderson, Am. J. Bot. 84 (suppl.), 132 (1997).
15. K. H. Wolfe et al., Proc. Natl. Acad. Sci. U.S.A. 86, 6201 (1989); W. Martin, A. Giert, H. Saedler, Nature 339, 46 (1989); P. R. Crane et al., ibid. 342, 131 (1989); W. Martin et al., Mol. Biol. Evol. 10, 140 (1993); K. J. Systma and D. A. Baum, in Flowering Plant Origin, Evolution and Phylogeny, D. W. Taylor and L. J. Hickey, Eds. (Chapman & Hall, New York, 1996), pp. 314-340; J. A. Doyle and M. J. Sanderson, Am. J. Bot. 84 (suppl.), 132 (1997).
16. K. H. Wolfe et al., Proc. Natl. Acad. Sci. U.S.A. 86, 6201 (1989); W. Martin, A. Giert, H. Saedler, Nature 339, 46 (1989); P. R. Crane et al., ibid. 342, 131 (1989); W. Martin et al., Mol. Biol. Evol. 10, 140 (1993); K. J. Systma and D. A. Baum, in Flowering Plant Origin, Evolution and Phylogeny, D. W. Taylor and L. J. Hickey, Eds. (Chapman & Hall, New York, 1996), pp. 314-340; J. A. Doyle and M. J. Sanderson, Am. J. Bot. 84 (suppl.), 132 (1997).
17. K. H. Wolfe et al., Proc. Natl. Acad. Sci. U.S.A. 86, 6201 (1989); W. Martin, A. Giert, H. Saedler, Nature 339, 46 (1989); P. R. Crane et al., ibid. 342, 131 (1989); W. Martin et al., Mol. Biol. Evol. 10, 140 (1993); K. J. Systma and D. A. Baum, in Flowering Plant Origin, Evolution and Phylogeny, D. W. Taylor and L. J. Hickey, Eds. (Chapman & Hall, New York, 1996), pp. 314-340; J. A. Doyle and M. J. Sanderson, Am. J. Bot. 84 (suppl.), 132 (1997).
18. M. J. Sanderson, Mol. Biol. Evol. 14, 1218 (1997).
19. G. U. Yule, Philos. Trans. R. Soc. London Ser. B 213, 21 (1924); D. G. Kendall, Ann. Math. Statist. 19, 1 (1948).
20. G. U. Yule, Philos. Trans. R. Soc. London Ser. B 213, 21 (1924); D. G. Kendall, Ann. Math. Statist. 19, 1 (1948).
21. D. M. Raup, Paleobiology 11, 42 (1985).
22. D. Normile, Science 281, 774 (1998).
23. D. M. Raup, Paleobiology 4, 1 (1978); ibid. 9, 107 (1983); ibid. 17, 37 (1991); _ and J. W. Valentine, Proc. Natl. Acad. Sci. U.S.A. 80, 2981 (1983); R. R. Strathman and M. Slatkin, Paleobiology 9, 97 (1983); M. E. Patzkowksy, ibid. 21, 440 (1995); M. Foote, ibid. 22, 141 (1996).
24. D. M. Raup, Paleobiology 4, 1 (1978); ibid. 9, 107 (1983); ibid. 17, 37 (1991); _ and J. W. Valentine, Proc. Natl. Acad. Sci. U.S.A. 80, 2981 (1983); R. R. Strathman and M. Slatkin, Paleobiology 9, 97 (1983); M. E. Patzkowksy, ibid. 21, 440 (1995); M. Foote, ibid. 22, 141 (1996).
25. D. M. Raup, Paleobiology 4, 1 (1978); ibid. 9, 107 (1983); ibid. 17, 37 (1991); _ and J. W. Valentine, Proc. Natl. Acad. Sci. U.S.A. 80, 2981 (1983); R. R. Strathman and M. Slatkin, Paleobiology 9, 97 (1983); M. E. Patzkowksy, ibid. 21, 440 (1995); M. Foote, ibid. 22, 141 (1996).
26. D. M. Raup, Paleobiology 4, 1 (1978); ibid. 9, 107 (1983); ibid. 17, 37 (1991); _ and J. W. Valentine, Proc. Natl. Acad. Sci. U.S.A. 80, 2981 (1983); R. R. Strathman and M. Slatkin, Paleobiology 9, 97 (1983); M. E. Patzkowksy, ibid. 21, 440 (1995); M. Foote, ibid. 22, 141 (1996).
27. D. M. Raup, Paleobiology 4, 1 (1978); ibid. 9, 107 (1983); ibid. 17, 37 (1991); _ and J. W. Valentine, Proc. Natl. Acad. Sci. U.S.A. 80, 2981 (1983); R. R. Strathman and M. Slatkin, Paleobiology 9, 97 (1983); M. E. Patzkowksy, ibid. 21, 440 (1995); M. Foote, ibid. 22, 141 (1996).
28. D. M. Raup, Paleobiology 4, 1 (1978); ibid. 9, 107 (1983); ibid. 17, 37 (1991); _ and J. W. Valentine, Proc. Natl. Acad. Sci. U.S.A. 80, 2981 (1983); R. R. Strathman and M. Slatkin, Paleobiology 9, 97 (1983); M. E. Patzkowksy, ibid. 21, 440 (1995); M. Foote, ibid. 22, 141 (1996).
29. D. M. Raup, Paleobiology 4, 1 (1978); ibid. 9, 107 (1983); ibid. 17, 37 (1991); _ and J. W. Valentine, Proc. Natl. Acad. Sci. U.S.A. 80, 2981 (1983); R. R. Strathman and M. Slatkin, Paleobiology 9, 97 (1983); M. E. Patzkowksy, ibid. 21, 440 (1995); M. Foote, ibid. 22, 141 (1996).
30. We evaluate divergence times rather than branching topology. We are not concerned with determining whether the relative order of hypothesized origins is consistent with observed first appearances [M. A. Norell and M. J. Novacek, Science 255, 1690 (1992); J. P. Huelsenbeck, Paleobiology 20, 470 (1994); P. J. Wagner, ibid. 21, 153 (1995); R. Hitchin and M. J. Benton, ibid. 23, 20 (1997)]. Rather, we are interested in cases in which the absolute timing of postulated origins and of observed origins are already known to disagree.
31. We evaluate divergence times rather than branching topology. We are not concerned with determining whether the relative order of hypothesized origins is consistent with observed first appearances [M. A. Norell and M. J. Novacek, Science 255, 1690 (1992); J. P. Huelsenbeck, Paleobiology 20, 470 (1994); P. J. Wagner, ibid. 21, 153 (1995); R. Hitchin and M. J. Benton, ibid. 23, 20 (1997)]. Rather, we are interested in cases in which the absolute timing of postulated origins and of observed origins are already known to disagree.
32. We evaluate divergence times rather than branching topology. We are not concerned with determining whether the relative order of hypothesized origins is consistent with observed first appearances [M. A. Norell and M. J. Novacek, Science 255, 1690 (1992); J. P. Huelsenbeck, Paleobiology 20, 470 (1994); P. J. Wagner, ibid. 21, 153 (1995); R. Hitchin and M. J. Benton, ibid. 23, 20 (1997)]. Rather, we are interested in cases in which the absolute timing of postulated origins and of observed origins are already known to disagree.
33. We evaluate divergence times rather than branching topology. We are not concerned with determining whether the relative order of hypothesized origins is consistent with observed first appearances [M. A. Norell and M. J. Novacek, Science 255, 1690 (1992); J. P. Huelsenbeck, Paleobiology 20, 470 (1994); P. J. Wagner, ibid. 21, 153 (1995); R. Hitchin and M. J. Benton, ibid. 23, 20 (1997)]. Rather, we are interested in cases in which the absolute timing of postulated origins and of observed origins are already known to disagree.
34. n-j if p ≠ q The last equation is a correction of equation (A18) from (9). Let P(≥ n, t, a) be the probability that diversity is greater than or equal to n lineages at time t, given that it is equal to a lineages at t = 0. Then (equation presented) Let t and T be two points in time such that t < T, and let the group have diversity equal to one lineage at t = 0. Let P(n, t, s, T) be the probability that diversity is exactly equal to n lineages at time t, given that the group survives at least until time T. Let P(n,t,N,T) be the probability that diversity is exactly equal to n lineages at time t, given that diversity is exactly equal to N lineages at time T. And let P(n,t ≥ N, T) be the probability that diversity is exactly equal to n lineages at time t, given that diversity is greater than or equal to N lineages at time T. Then, by the rules of conditional probability, P(n,t,s,T) = P(n,t,1) · P(s,T - t,n)/P(s,T,1); P(n.t.N,T) = P(n,t,1) · P(N,T - t,n)/P(N,T,1); and P(n,t, ≥ N,T) = P(n,t,1) · P(≥ N,T - t,n)/P(≥ N,T,1) The corresponding expected (mean) diversities at time t are given by (equation presented) and (equation presented)
35. M. Foote, Paleobiology 23, 278 (1997). Preservation rate, r, per lineage-million-years, implicitly incorporates preservation and recovery.
36. C. R. Marshall, ibid. 16, 1 (1990).
37. A. R. Solow and W. Smith, ibid. 23, 271 (1997).
38. -rS. If we demand a minimal probability of 0.5 for a hypothesis of missing diversity to be plausible, then r must be less than or equal to -ln(0.5)/S.
39. We simplify the problem using the well-known positive correlation between origination and extinction rates and constrain the net growth rate (p - q) to be equal to ln(N)/T, thus forcing p to be equal to q + ln(N)/T. For discussion of this correlation, see S. M. Stanley, in Causes of Evolution, R. M. Ross and W. D. Allmon, Eds. (Univ. of Chicago Press, Chicago, IL, 1990), pp. 103-127.
40. M. J. Novacek, Nature 356, 121 (1992). This phylogeny is similar to that of J. Shoshani and M. C. McKenna, Mol. Phylogenet. Evol. 9, 572 (1998).
41. M. J. Novacek, Nature 356, 121 (1992). This phylogeny is similar to that of J. Shoshani and M. C. McKenna, Mol. Phylogenet. Evol. 9, 572 (1998).
42. M. J. Benton, J. Mol. Evol. 30, 409 (1990); R. L. Carroll, Vertebrate Paleontology and Evolution (Freeman, New York, 1988); C. M. Janis, K. M. Scott, L. L. Jacobs, Eds. Evolution of Tertiary Mammals of North America (Cambridge Univ. Press, Cambridge, 1998).
43. M. J. Benton, J. Mol. Evol. 30, 409 (1990); R. L. Carroll, Vertebrate Paleontology and Evolution (Freeman, New York, 1988); C. M. Janis, K. M. Scott, L. L. Jacobs, Eds. Evolution of Tertiary Mammals of North America (Cambridge Univ. Press, Cambridge, 1998).
44. M. J. Benton, J. Mol. Evol. 30, 409 (1990); R. L. Carroll, Vertebrate Paleontology and Evolution (Freeman, New York, 1988); C. M. Janis, K. M. Scott, L. L. Jacobs, Eds. Evolution of Tertiary Mammals of North America (Cambridge Univ. Press, Cambridge, 1998).
45. R. C. Stucky and M. C. McKenna, in The Fossil Record, M. J. Benton, Ed. (Chapman & Hall, London, 1993), vol. 2, pp. 739-771.
46. A number of Cretaceous fossils have been interpreted as modern eutherians, but none of these interpretations is without controversy. Claims that the Cretaceous eutherians Batodon [M. J. Novacek, Contrib. Sci. Los Angeles Co. Mus. 283, 1 (1976); L. Krishtalka and R. M. West, Milwaukee Publ. Mus. Contrib. Biol. Geol. 27, 1 (1979) ] and Paranyctoides [ R. C. Fox, Spec. Pub. Carnegie Mus. Nat. Hist. 9, 9 (1984) ] are lipotyphlans have been regarded as weak [P. M. Butler, in The Phylogeny and Classification of the Tetrapods, M. J. Benton, Ed. (Clarendon, Oxford, 1988), pp. 117-141). Cretaceous "zhelestids," recently claimed to be ungulatomorphs (26), basal ferungulates (sensu Kumar and Hedges), or archontans (4), may instead be archaic eutherians allied with either Prokennalestes (39) or zalambdalestids [ P. M. Butler, Biol. Rev. Cambridge Philos. Soc. 65, 529 (1990)].
47. A number of Cretaceous fossils have been interpreted as modern eutherians, but none of these interpretations is without controversy. Claims that the Cretaceous eutherians Batodon [M. J. Novacek, Contrib. Sci. Los Angeles Co. Mus. 283, 1 (1976); L. Krishtalka and R. M. West, Milwaukee Publ. Mus. Contrib. Biol. Geol. 27, 1 (1979) ] and Paranyctoides [ R. C. Fox, Spec. Pub. Carnegie Mus. Nat. Hist. 9, 9 (1984) ] are lipotyphlans have been regarded as weak [P. M. Butler, in The Phylogeny and Classification of the Tetrapods, M. J. Benton, Ed. (Clarendon, Oxford, 1988), pp. 117-141). Cretaceous "zhelestids," recently claimed to be ungulatomorphs (26), basal ferungulates (sensu Kumar and Hedges), or archontans (4), may instead be archaic eutherians allied with either Prokennalestes (39) or zalambdalestids [ P. M. Butler, Biol. Rev. Cambridge Philos. Soc. 65, 529 (1990)].
48. A number of Cretaceous fossils have been interpreted as modern eutherians, but none of these interpretations is without controversy. Claims that the Cretaceous eutherians Batodon [M. J. Novacek, Contrib. Sci. Los Angeles Co. Mus. 283, 1 (1976); L. Krishtalka and R. M. West, Milwaukee Publ. Mus. Contrib. Biol. Geol. 27, 1 (1979) ] and Paranyctoides [ R. C. Fox, Spec. Pub. Carnegie Mus. Nat. Hist. 9, 9 (1984) ] are lipotyphlans have been regarded as weak [P. M. Butler, in The Phylogeny and Classification of the Tetrapods, M. J. Benton, Ed. (Clarendon, Oxford, 1988), pp. 117-141). Cretaceous "zhelestids," recently claimed to be ungulatomorphs (26), basal ferungulates (sensu Kumar and Hedges), or archontans (4), may instead be archaic eutherians allied with either Prokennalestes (39) or zalambdalestids [ P. M. Butler, Biol. Rev. Cambridge Philos. Soc. 65, 529 (1990)].
49. A number of Cretaceous fossils have been interpreted as modern eutherians, but none of these interpretations is without controversy. Claims that the Cretaceous eutherians Batodon [M. J. Novacek, Contrib. Sci. Los Angeles Co. Mus. 283, 1 (1976); L. Krishtalka and R. M. West, Milwaukee Publ. Mus. Contrib. Biol. Geol. 27, 1 (1979) ] and Paranyctoides [ R. C. Fox, Spec. Pub. Carnegie Mus. Nat. Hist. 9, 9 (1984) ] are lipotyphlans have been regarded as weak [P. M. Butler, in The Phylogeny and Classification of the Tetrapods, M. J. Benton, Ed. (Clarendon, Oxford, 1988), pp. 117-141). Cretaceous "zhelestids," recently claimed to be ungulatomorphs (26), basal ferungulates (sensu Kumar and Hedges), or archontans (4), may instead be archaic eutherians allied with either Prokennalestes (39) or zalambdalestids [ P. M. Butler, Biol. Rev. Cambridge Philos. Soc. 65, 529 (1990)].
50. A number of Cretaceous fossils have been interpreted as modern eutherians, but none of these interpretations is without controversy. Claims that the Cretaceous eutherians Batodon [M. J. Novacek, Contrib. Sci. Los Angeles Co. Mus. 283, 1 (1976); L. Krishtalka and R. M. West, Milwaukee Publ. Mus. Contrib. Biol. Geol. 27, 1 (1979) ] and Paranyctoides [ R. C. Fox, Spec. Pub. Carnegie Mus. Nat. Hist. 9, 9 (1984) ] are lipotyphlans have been regarded as weak [P. M. Butler, in The Phylogeny and Classification of the Tetrapods, M. J. Benton, Ed. (Clarendon, Oxford, 1988), pp. 117-141). Cretaceous "zhelestids," recently claimed to be ungulatomorphs (26), basal ferungulates (sensu Kumar and Hedges), or archontans (4), may instead be archaic eutherians allied with either Prokennalestes (39) or zalambdalestids [ P. M. Butler, Biol. Rev. Cambridge Philos. Soc. 65, 529 (1990)].
51. W. W. de Jong, Trends Ecol. Evol. 13, 270 (1998).
52. M. Foote and D. M. Raup, Paleobiology 22, 121 (1996).
53. S. M. Stanley, ibid. 11, 13 (1985); L. M. Van Valen, Evol. Theory 1, 1 (1973); ibid. 7, 127 (1985).
54. S. M. Stanley, ibid. 11, 13 (1985); L. M. Van Valen, Evol. Theory 1, 1 (1973); ibid. 7, 127 (1985).
55. S. M. Stanley, ibid. 11, 13 (1985); L. M. Van Valen, Evol. Theory 1, 1 (1973); ibid. 7, 127 (1985).
56. J. D. Archibald, Science 272, 1150 (1996); L. A. Nessov, J. D. Archibald, Z. Kielan-Jaworowska, Bull. Carnegie Mus. Nat. Hist. 34, 40 (1998).
57. J. D. Archibald, Science 272, 1150 (1996); L. A. Nessov, J. D. Archibald, Z. Kielan-Jaworowska, Bull. Carnegie Mus. Nat. Hist. 34, 40 (1998).
58. J. Alroy, Science 280, 731 (1998).
59. 0 (all standard errors based on bootstrap resampling of these 41 species). Because of the small number of species and the uncertainty in absolute dating, there is uncertainty in our estimate of extinction rate. But because the relation between estimated extinction rate and estimated preservation rate is linear, the value of q would have to be more than an order of magnitude lower than our estimate in order to affect our conclusions substantially. Taxonomic rates this Low are incompatible with rates estimated for Cenozoic mammals (14, 24, 25) and for even the most slowly evolving groups of animals (25).
60. Because of the difference between Cenozoic and Cretaceous preservation rates, we do not recommend using Cenozoic occurrences to place confidence limits (15) on stratigraphic ranges for mammal taxa that may extend into the Cretaceous [see also (10)]. Moreover, confidence limits are more difficult to estimate for higher taxa than for single species, because the probability of group preservation per unit time is potentially greatly affected by changes in diversity [R. Bleiweiss, Geology 26, 323 (1998); C. R. Marshall and R. Bleiweiss, ibid. 27, 95 (1999) ].
61. Because of the difference between Cenozoic and Cretaceous preservation rates, we do not recommend using Cenozoic occurrences to place confidence limits (15) on stratigraphic ranges for mammal taxa that may extend into the Cretaceous [see also (10)]. Moreover, confidence limits are more difficult to estimate for higher taxa than for single species, because the probability of group preservation per unit time is potentially greatly affected by changes in diversity [R. Bleiweiss, Geology 26, 323 (1998); C. R. Marshall and R. Bleiweiss, ibid. 27, 95 (1999) ].
62. M. Messer et al., J. Mammal. Evol. 5, 95 (1998).
63. It is also likely that preservation rate fluctuates over time, because sea level and other factors change. Modeling of time-heterogeneous preservation (14, 24), however, shows that fluctuating preservation rate is not likely to distort substantially either the overall probability of species preservation or our estimates of preservation rate.
64. M. S. Springer et al., Nature 388, 61 (1997); M. J. Stanhope et al., Mol. Phylogenet. Evol. 9, 501 (1998); M. J. Stanhope et al., Proc. Natl. Acad. Sci. U.S.A. 95, 9967 (1998).
65. M. S. Springer et al., Nature 388, 61 (1997); M. J. Stanhope et al., Mol. Phylogenet. Evol. 9, 501 (1998); M. J. Stanhope et al., Proc. Natl. Acad. Sci. U.S.A. 95, 9967 (1998).
66. M. S. Springer et al., Nature 388, 61 (1997); M. J. Stanhope et al., Mol. Phylogenet. Evol. 9, 501 (1998); M. J. Stanhope et al., Proc. Natl. Acad. Sci. U.S.A. 95, 9967 (1998).
67. P. D. Gingerich, Mol. Biol. Evol. 3, 205 (1986).
68. R. J. Britten, Science 231, 1393 (1986).
69. J. H. Gillespie, Proc. Natl. Acad. Sci. U.S.A. 81, 8009 (1984); Genetics 113, 1077 (1986); Mol. Biol. Evol. 3, 138 (1986); M. M. Miyamoto and W. M. Fitch, ibid. 12, 503 (1995).
70. J. H. Gillespie, Proc. Natl. Acad. Sci. U.S.A. 81, 8009 (1984); Genetics 113, 1077 (1986); Mol. Biol. Evol. 3, 138 (1986); M. M. Miyamoto and W. M. Fitch, ibid. 12, 503 (1995).
71. J. H. Gillespie, Proc. Natl. Acad. Sci. U.S.A. 81, 8009 (1984); Genetics 113, 1077 (1986); Mol. Biol. Evol. 3, 138 (1986); M. M. Miyamoto and W. M. Fitch, ibid. 12, 503 (1995).
72. J. H. Gillespie, Proc. Natl. Acad. Sci. U.S.A. 81, 8009 (1984); Genetics 113, 1077 (1986); Mol. Biol. Evol. 3, 138 (1986); M. M. Miyamoto and W. M. Fitch, ibid. 12, 503 (1995).
73. M. Goodman, Prog. Biophys. Mol. Biol. 37, 105 (1981).
74. D. M. Raup, in Rates of Evolution, K. S. W. Campbell and M. F. Day, Eds. (Allen & Unwin, London, 1986), pp. 1-14.
75. If molecular rates are faster during evolutionary radiations, this may reflect two end-member mechanisms that have been suggested previously: (i) Adaptive change within species is faster during the rapid occupation of new adaptive zones (41), and many molecular substitutions are selective rather than neutral (35, 36); or (ii) a punctuational model applies to molecular evolution, with change concentrated at lineage splitting; thus, rapid speciation during evolutionary radiation (41) causes faster molecular rates (37). These mechanisms are testable with data from extant groups for which fossil diversity, genealogical relationships, and fossil divergence times are reliably known: Each pair of living species is separated by a divergence time, a number of cladogenetic events (patristic distance), a morphological distance, and a molecular distance. If morphological divergence provides a measure of adaptive change, the first hypothesis implies a positive partial correlation between molecular and morphological distance, with divergence time and patristic distance held statistically constant. The second hypothesis implies a positive partial correlation between molecular and patristic distance, with divergence time and morphological distance held constant.
76. M. C. McKenna and S. K. Bell, Classification of Mammals Above the Species Level (Columbia Univ. Press, New York, 1997).
77. Occurrence data are from a compilation of faunal lists and synonymies, extensively supplemented with observations by J.P.H. This database is heavily influenced by the rich fossil record of North America, but also includes the known faunas of South America, Europe, Asia (Mongolia, India, and western Asia), and Madagascar. Biostratigraphic correlation for species outside North America was based largely on information in the references consulted for faunal lists. For North American species, these sources were supplemented by the following: J. A. Lillegraven, Contrib. Sci. Los Angeles Co. Mus. 232, 1 (1972); _ and L. M. Ostresh Jr., Geol. Soc. Am. Spec. Pap. 243, 1 (1990); M. B. Goodwin and A. L. Deino, Can. J. Earth Sci. 26, 1384 (1989); R. R. Rogers, C. C. Swisher III, J. R. Homer, ibid. 30, 1066 (1993); C. C. Swisher III, L. Dingus, R. F. Butler, ibid., p. 1981; F. M. Gradstein et al., in Geochronology, Time Scales and Global Stratigraphic Correlation, W. A. Berggren, D. V. Kent, M.-P. Aubry, J. Hardenbol, Eds. (Society for Sedimentary Geology (SEPM), Tulsa, OK], pp. 95-126. To determine whether a species is known from a single horizon, it is generally necessary to have only a relative biochronology. We did not base our relative chronology solely on faunal associations of mammals, but also used local and regional stratigraphic relations, radiometric dates, and correlations with marine sequences based largely on regional strandline stratigraphy, as derived from sources cited above. North American localities were arranged into 14 resolvable stratigraphic intervals, and the localities of other continents were resolved to between 1 and 3 intervals, depending on the region. We treated each of these intervals operationally as a horizon. This approach is conservative insofar as it lumps distinct horizons into one unit, overestimating the proportion of species confined to single horizons and thus underestimating preservation rate. Our approach is also conservative insofar as we generally considered species with uncertain stratigraphic ranges as single-horizon taxa. To estimate average stratigraphic ranges of North American species, for which the biochronology is best understood, we assigned absolute ages to the 14 stratigraphic intervals, mainly using correlations with ammonite zones and radiometric dates. See supplemental data at www.sciencemag.org/ feature/data/985988.shl.
78. Occurrence data are from a compilation of faunal lists and synonymies, extensively supplemented with observations by J.P.H. This database is heavily influenced by the rich fossil record of North America, but also includes the known faunas of South America, Europe, Asia (Mongolia, India, and western Asia), and Madagascar. Biostratigraphic correlation for species outside North America was based largely on information in the references consulted for faunal lists. For North American species, these sources were supplemented by the following: J. A. Lillegraven, Contrib. Sci. Los Angeles Co. Mus. 232, 1 (1972); _ and L. M. Ostresh Jr., Geol. Soc. Am. Spec. Pap. 243, 1 (1990); M. B. Goodwin and A. L. Deino, Can. J. Earth Sci. 26, 1384 (1989); R. R. Rogers, C. C. Swisher III, J. R. Homer, ibid. 30, 1066 (1993); C. C. Swisher III, L. Dingus, R. F. Butler, ibid., p. 1981; F. M. Gradstein et al., in Geochronology, Time Scales and Global Stratigraphic Correlation, W. A. Berggren, D. V. Kent, M.-P. Aubry, J. Hardenbol, Eds. (Society for Sedimentary Geology (SEPM), Tulsa, OK], pp. 95-126. To determine whether a species is known from a single horizon, it is generally necessary to have only a relative biochronology. We did not base our relative chronology solely on faunal associations of mammals, but also used local and regional stratigraphic relations, radiometric dates, and correlations with marine sequences based largely on regional strandline stratigraphy, as derived from sources cited above. North American localities were arranged into 14 resolvable stratigraphic intervals, and the localities of other continents were resolved to between 1 and 3 intervals, depending on the region. We treated each of these intervals operationally as a horizon. This approach is conservative insofar as it lumps distinct horizons into one unit, overestimating the proportion of species confined to single horizons and thus underestimating preservation rate. Our approach is also conservative insofar as we generally considered species with uncertain stratigraphic ranges as single-horizon taxa. To estimate average stratigraphic ranges of North American species, for which the biochronology is best understood, we assigned absolute ages to the 14 stratigraphic intervals, mainly using correlations with ammonite zones and radiometric dates. See supplemental data at www.sciencemag.org/ feature/data/985988.shl.
79. Occurrence data are from a compilation of faunal lists and synonymies, extensively supplemented with observations by J.P.H. This database is heavily influenced by the rich fossil record of North America, but also includes the known faunas of South America, Europe, Asia (Mongolia, India, and western Asia), and Madagascar. Biostratigraphic correlation for species outside North America was based largely on information in the references consulted for faunal lists. For North American species, these sources were supplemented by the following: J. A. Lillegraven, Contrib. Sci. Los Angeles Co. Mus. 232, 1 (1972); _ and L. M. Ostresh Jr., Geol. Soc. Am. Spec. Pap. 243, 1 (1990); M. B. Goodwin and A. L. Deino, Can. J. Earth Sci. 26, 1384 (1989); R. R. Rogers, C. C. Swisher III, J. R. Homer, ibid. 30, 1066 (1993); C. C. Swisher III, L. Dingus, R. F. Butler, ibid., p. 1981; F. M. Gradstein et al., in Geochronology, Time Scales and Global Stratigraphic Correlation, W. A. Berggren, D. V. Kent, M.-P. Aubry, J. Hardenbol, Eds. (Society for Sedimentary Geology (SEPM), Tulsa, OK], pp. 95-126. To determine whether a species is known from a single horizon, it is generally necessary to have only a relative biochronology. We did not base our relative chronology solely on faunal associations of mammals, but also used local and regional stratigraphic relations, radiometric dates, and correlations with marine sequences based largely on regional strandline stratigraphy, as derived from sources cited above. North American localities were arranged into 14 resolvable stratigraphic intervals, and the localities of other continents were resolved to between 1 and 3 intervals, depending on the region. We treated each of these intervals operationally as a horizon. This approach is conservative insofar as it lumps distinct horizons into one unit, overestimating the proportion of species confined to single horizons and thus underestimating preservation rate. Our approach is also conservative insofar as we generally considered species with uncertain stratigraphic ranges as single-horizon taxa. To estimate average stratigraphic ranges of North American species, for which the biochronology is best understood, we assigned absolute ages to the 14 stratigraphic intervals, mainly using correlations with ammonite zones and radiometric dates. See supplemental data at www.sciencemag.org/ feature/data/985988.shl.
80. Occurrence data are from a compilation of faunal lists and synonymies, extensively supplemented with observations by J.P.H. This database is heavily influenced by the rich fossil record of North America, but also includes the known faunas of South America, Europe, Asia (Mongolia, India, and western Asia), and Madagascar. Biostratigraphic correlation for species outside North America was based largely on information in the references consulted for faunal lists. For North American species, these sources were supplemented by the following: J. A. Lillegraven, Contrib. Sci. Los Angeles Co. Mus. 232, 1 (1972); _ and L. M. Ostresh Jr., Geol. Soc. Am. Spec. Pap. 243, 1 (1990); M. B. Goodwin and A. L. Deino, Can. J. Earth Sci. 26, 1384 (1989); R. R. Rogers, C. C. Swisher III, J. R. Homer, ibid. 30, 1066 (1993); C. C. Swisher III, L. Dingus, R. F. Butler, ibid., p. 1981; F. M. Gradstein et al., in Geochronology, Time Scales and Global Stratigraphic Correlation, W. A. Berggren, D. V. Kent, M.-P. Aubry, J. Hardenbol, Eds. (Society for Sedimentary Geology (SEPM), Tulsa, OK], pp. 95-126. To determine whether a species is known from a single horizon, it is generally necessary to have only a relative biochronology. We did not base our relative chronology solely on faunal associations of mammals, but also used local and regional stratigraphic relations, radiometric dates, and correlations with marine sequences based largely on regional strandline stratigraphy, as derived from sources cited above. North American localities were arranged into 14 resolvable stratigraphic intervals, and the localities of other continents were resolved to between 1 and 3 intervals, depending on the region. We treated each of these intervals operationally as a horizon. This approach is conservative insofar as it lumps distinct horizons into one unit, overestimating the proportion of species confined to single horizons and thus underestimating preservation rate. Our approach is also conservative insofar as we generally considered species with uncertain stratigraphic ranges as single-horizon taxa. To estimate average stratigraphic ranges of North American species, for which the biochronology is best understood, we assigned absolute ages to the 14 stratigraphic intervals, mainly using correlations with ammonite zones and radiometric dates. See supplemental data at www.sciencemag.org/ feature/data/985988.shl.
81. Occurrence data are from a compilation of faunal lists and synonymies, extensively supplemented with observations by J.P.H. This database is heavily influenced by the rich fossil record of North America, but also includes the known faunas of South America, Europe, Asia (Mongolia, India, and western Asia), and Madagascar. Biostratigraphic correlation for species outside North America was based largely on information in the references consulted for faunal lists. For North American species, these sources were supplemented by the following: J. A. Lillegraven, Contrib. Sci. Los Angeles Co. Mus. 232, 1 (1972); _ and L. M. Ostresh Jr., Geol. Soc. Am. Spec. Pap. 243, 1 (1990); M. B. Goodwin and A. L. Deino, Can. J. Earth Sci. 26, 1384 (1989); R. R. Rogers, C. C. Swisher III, J. R. Homer, ibid. 30, 1066 (1993); C. C. Swisher III, L. Dingus, R. F. Butler, ibid., p. 1981; F. M. Gradstein et al., in Geochronology, Time Scales and Global Stratigraphic Correlation, W. A. Berggren, D. V. Kent, M.-P. Aubry, J. Hardenbol, Eds. (Society for Sedimentary Geology (SEPM), Tulsa, OK], pp. 95-126. To determine whether a species is known from a single horizon, it is generally necessary to have only a relative biochronology. We did not base our relative chronology solely on faunal associations of mammals, but also used local and regional stratigraphic relations, radiometric dates, and correlations with marine sequences based largely on regional strandline stratigraphy, as derived from sources cited above. North American localities were arranged into 14 resolvable stratigraphic intervals, and the localities of other continents were resolved to between 1 and 3 intervals, depending on the region. We treated each of these intervals operationally as a horizon. This approach is conservative insofar as it lumps distinct horizons into one unit, overestimating the proportion of species confined to single horizons and thus underestimating preservation rate. Our approach is also conservative insofar as we generally considered species with uncertain stratigraphic ranges as single-horizon taxa. To estimate average stratigraphic ranges of North American species, for which the biochronology is best understood, we assigned absolute ages to the 14 stratigraphic intervals, mainly using correlations with ammonite zones and radiometric dates. See supplemental data at www.sciencemag.org/ feature/data/985988.shl.
82. Occurrence data are from a compilation of faunal lists and synonymies, extensively supplemented with observations by J.P.H. This database is heavily influenced by the rich fossil record of North America, but also includes the known faunas of South America, Europe, Asia (Mongolia, India, and western Asia), and Madagascar. Biostratigraphic correlation for species outside North America was based largely on information in the references consulted for faunal lists. For North American species, these sources were supplemented by the following: J. A. Lillegraven, Contrib. Sci. Los Angeles Co. Mus. 232, 1 (1972); _ and L. M. Ostresh Jr., Geol. Soc. Am. Spec. Pap. 243, 1 (1990); M. B. Goodwin and A. L. Deino, Can. J. Earth Sci. 26, 1384 (1989); R. R. Rogers, C. C. Swisher III, J. R. Homer, ibid. 30, 1066 (1993); C. C. Swisher III, L. Dingus, R. F. Butler, ibid., p. 1981; F. M. Gradstein et al., in Geochronology, Time Scales and Global Stratigraphic Correlation, W. A. Berggren, D. V. Kent, M.-P. Aubry, J. Hardenbol, Eds. (Society for Sedimentary Geology (SEPM), Tulsa, OK], pp. 95-126. To determine whether a species is known from a single horizon, it is generally necessary to have only a relative biochronology. We did not base our relative chronology solely on faunal associations of mammals, but also used local and regional stratigraphic relations, radiometric dates, and correlations with marine sequences based largely on regional strandline stratigraphy, as derived from sources cited above. North American localities were arranged into 14 resolvable stratigraphic intervals, and the localities of other continents were resolved to between 1 and 3 intervals, depending on the region. We treated each of these intervals operationally as a horizon. This approach is conservative insofar as it lumps distinct horizons into one unit, overestimating the proportion of species confined to single horizons and thus underestimating preservation rate. Our approach is also conservative insofar as we generally considered species with uncertain stratigraphic ranges as single-horizon taxa. To estimate average stratigraphic ranges of North American species, for which the biochronology is best understood, we assigned absolute ages to the 14 stratigraphic intervals, mainly using correlations with ammonite zones and radiometric dates. See supplemental data at www.sciencemag.org/ feature/data/985988.shl.
83. S. M. Stanley, Macroevolution (Freeman, San Francisco, 1979).
84. W. B. Hartand et al., A Geologic Time Scale 1989 (Cambridge Univ. Press, Cambridge, 1990).
85. E. Gheerbrant, J. Sudre, H. Cappetta, Nature 383, 68 (1996).
86. R. C. Fox and G. P. Youzwyshyn, J. Vertebr. Paleontol. 14, 382 (1994).
87. We thank J. Alroy, R. J. Asher, J. Flynn, D. Jablonski, C. R. Marshall, R. R. Rogers, P. J. Wagner, and J. R. Wible for discussion, and R. H. De Simone, D. Jablonski, C. R. Marshall, D. M. Raup, and two anonymous referees for reviews. This research was supported by NSF (grant EAR-9506568) and NASA (grant NAGW-1693).