If strict Darwinists were to be believed, the history of life is an uninterrupted sequence of species turnover, with species incessantly appearing and disappearing as creatures either adapt to changing physical environments and ever-shifting competition, evolve into new species, or go extinct. No coordination is involved - species appear and disappear independently of each other through time. However, some palaeontologists have found evidence that the history of life is not always an unremitting process of change - evolution sometimes takes breaks of a few million years. Communities it seems, like individual species, show periods of stasis in the fossil record. Within the overall biodiversity increase through geological time sit prolonged phases of steady-state biodiversity (Rosenzweig 1995, 52). The marine invertebrate, land vertebrate, and plant fossil records share a common pattern at regional and continental scales - long intervals of comparative stability broken by bouts of swift change.
The notion of stasis and change in the fossil record is traceable to such early 'geologists' as Baron Georges Cuvier (1769-1832) and his pupil Alcide Dessalines d'Orbigny (1802-1857). Cuvier (1817) recognized several great revolutions of the Earth's surface that ravaged nearly all the globe, with species surviving in a few isolated places acting as a source for repopulat-ing the planet. However, d'Orbigny believed that world-wide upheavals annihilated all life, and he documented no fewer than 28 catastrophes in the fossil record (d'Orbigny, 1840-7). This fundamental pattern of biotic history is so pervasive that it has acquired many names, though they come from different theoretical foundations. Examples include stepwise evolution and extinction (Simpson 1944; Benson et al. 1984), typostatic and typogenetic/typolytic phases of evolutionary cycles (Schindewolf 1950), biomeres (Palmer 1965), and, by extension to communities, punctuated equilibrium (Eldredge and Gould 1972). The consensus view is that environmental change causes these patterns.
In the early twentieth century, Herdman Cleland observed a lack of change in a range of fossils (brachiopods, corals, molluscs, echinoderms, and trilobites) in Early Silurian to Middle Devonian shales in Ontario, New York State. Some 50 years later, using case studies from the Permian deposits of Texas and Oklahoma, Everett C. Olson (1952, 1958, 1980) recognized communities that persisted through well-defined intervals of rock strata. He used the term chronofauna for such long-lasting assemblages of vertebrate taxa that recurred through long stratigraphic intervals, and he attempted to reconstruct food webs for the constituent species (Olson 1952).
Since Olson's pioneering studies, other palaeontologists have unearthed many other examples of fossil vertebrate faunas displaying a measure of continuity through time. Each continent has its 'Land Mammal Ages' defined by long-lasting suites of taxa in particular chronostratigraphic intervals, which allow correlation over wide areas (Woodburne 1987; Janis et al. 1998). The success of such schemes attests in broad terms to the persistence of at least some components of mammalian faunas and the communities to which they belonged (DiMichele et al. 2004). Indeed, the pulse of Cenozoic mammal communities displays a 'syncopated equilibrium', with long-lasting and stable chronofaunas separated by rapid turnover episodes involving radical reorganizations of terrestrial ecosystems (Webb and Opdyke 1995). The result is a large-scale and long-term succession of terrestrial ecosystems. The same process has operated throughout the Phanerozoic (Sheehan 1991; Sheehan and Russell 1994). Biotic crises abruptly ended long periods of stability in major varieties of dominant organisms. After each crisis, rapid diversification and ecological reorganization ushered in a new period of stability.
Later work than Cleland's shows that other fossil marine faunas register similar patterns of assemblage persistence. Arthur Boucot (1978, 1983) identified long-lasting marine invertebrate assemblages while researching the biostratigraphy and evolution of marine faunas. He noticed that invertebrate assemblages did not demonstrate patterns of continuous turnover; instead, they fall into coherent temporal units - ecologic-evolutionary units (EEUs) - each significantly different from others. Moreover, within EEUs sit smaller units, or communities, recognized by recurrent patterns of composition (Sheehan 1996). Boucot (1983) defined 12 EEUs for the entire Phanerozoic, based upon level-bottom, marine ben-thic organisms. Within each EEU, the varied community groups preserve their generic integrity from beginning to end of the time interval, although species belonging to more endemic and more stenotypic genera tend to evolve through phyletic gradualism.
Carl Brett and Gordon Baird (1995) quantified Cleland's finding, identifying 14 intervals of between 3 million to 7 million years' duration, during which at least 60 per cent of the species living together in the same environment lasted with little change within benthic marine biofacies of the Silurian and Devonian of the Appalachian Basin. The intervals ended with a bout of rapid turnover lasting a few hundred thousand years in which old species die out and new ones appear. Brett and Baird coined the term 'coordinated stasis' to describe such times of minimal evolutionary change. Brett et al. (1996) provided explicit criteria for comparing data for different times and different regions: periods of coordinated stasis should last more than 1 million years, during which time about 60 per cent of species should persist and show little morphological change. Fewer than 40 per cent (and typically fewer than 20 per cent) of species should cross the bounding intervals, which should be no more than one-tenth the duration of the static intervals. Speciation and extinction are concentrated in the intervals of rapid turnover.
Examples of coordinated stasis are legion. I shall describe two in detail here - Late Carboniferous coal beds in the eastern United States and Pliocene mammal faunas in East Africa - to convey the quality of the evidence.
During the Late Carboniferous (Pennsylvanian), global climate was general cool, with phases of very high rainfall in the tropics supporting rainforests and vast peat swamps that were to become the coal beds of Europe and the eastern USA. Upper Carboniferous rocks appear to reflect, in part, glacial periodicity and may preserve orbital forcing in the Milankovitch frequency bands (Algeo and Wilkinson 1987). Fossil remains of the plants from peat swamp (mire) forests are preserved as 'coal balls' (petrified peat), as compression-impression fossils in mudstones and sandstones, or as spores and pollen (DiMichele et al. 2004). The plant fossils occur in multiple coal beds and in the intervening rocks, so allowing the study of temporal changes in plant composition under repeated, common environmental conditions. Moreover, the incremental collection of fossil plant and sporepollen samples within coal beds make vegetation dynamics resolvable at timescales of less than 100,000 years. Furthermore, vegetation dynamics is examinable at many sampling horizons (coal beds) and so through numerous glacial-interglacial cycles and in response to both background and large-scale extinctions (Phillips et al. 1985). Tom L. Phillips and co-workers have researched the Late Carboniferous ecosystem in detail. Part of their work examined more than fifty coal beds, representing more than 10 million years, which revealed the following basic patterns (DiMichele et al. 2004):
1 Within any one coal bed, multiple, recurrent plant communities are recognizable through statistical analysis. These communities reappear in successive coals, identified by placing abundance of the dominant elements in approximate rank-order; minor taxa vary widely in abundance.
2 A major extinction eliminated nearly two-thirds of the species at the Middle-Late Pennsylvanian boundary, about 306 million years ago. A short pulse of global warming and drying in the tropics seems to have been the cause (Phillips and Peppers 1984; Frakes et al. 1992).
3 After the extinction came a brief interval of high variability in dominance patterns (Peppers 1996). Peat-forming landscapes then reorganized, and groups previously in low abundances, particularly opportunistic tree ferns, rose to dominance by replacing the former dominants that had succumbed to the climatic changes. Thus, the ever-wet peat-substrate species pool reorganized internally.
The same pattern of vegetation persistence registers in Late Pennsylvanian peat-forming environments in southeastern parts of the Illinois Basin (DiMichele et al. 2002). A parallel change in dominance patterns occurred in tropical flood-basin floras at around the same time, although the reported taxonomic resolution is at the level of families and classes
(Pfefferkorn and Thomson 1982). Furthermore, patterns similar to those found in coals have been documented in floras from flood-basin sedimentary rocks (sandstones and mud-stones) lying between coal beds.
The African late Cenozoic fossil record offers some of the highest resolution fossil evidence obtainable on mammalian community structure through time. Deposits along the lower Omo Valley, southern Ethiopia, include a sequence of almost 800 m of sediments spanning roughly 4 to 1 million years. A carefully documented collection of more than 40,000 fossil specimens from the Shungura Formation furnishes cases of faunal change at several timescales, with a temporal resolution of about 103 years in parts of the sequence (Bobe et al. 2003). Research by René Bobe and his colleagues has established ecological patterns for bovids, suids, and hominids and other primates. Major climatic and environmental changes occurred in Africa during time spanned by the Shungura Formation, and some of these changes seem to have produced changes in the mammalian fauna (Bobe et al. 2002). Species turnover, based on first and last occurrences, is low overall between 3.5 and 2.0 million years ago; the only marked turnover event occurred at 2.85 million years ago, corresponding to the onset of Northern Hemisphere cooling. The relative abundances of the major mammalian taxa (4,820 specimens) reveals a period of stability between 2.8 and 2.5 million years ago (five stratigraphic sample levels) that is followed by a cyclical pattern of shifting taxonomic dominance over 100,000-year intervals up to 2.0 million years ago (Bobe et al. 2002). Statistical tests indicate that the interval of stasis is unlikely to have arisen through sampling error, and provides firm evidence for ecological stability over several hundred thousand years, when global climates were becoming cooler and more variable. Bobe et al. (2002) speculate that the palaeo-Omo River system, with its large drainage area, helped to buffer the lower riverine floodplain and gallery forest habitats from the impact of larger-scale climate change. This effect persisted for several hundred thousand years, until around 2.5 million years ago, when external changes finally penetrated the local system and destabilized the ecological communities of the lower Omo Valley. The Shungura sequence provides additional evidence for faunal stability and ecological persistence. The two most abundant species of the dominant family, the Bovidae, occur together throughout the interval from about 3.5-2.0 million years ago, with Aepyceros shungurae (an early impala) and Tragelaphus nakuae (similar to the bongo) alternating in first and second place (Bobe and Eck 2001). This persistent association may indicate that both species had similar tolerance limits for the wooded and moist environments of the Pliocene-Pleistocene lower Omo River. After 2.0 million years ago, during a period of increased environmental change, T. nakuae became extinct, and A. shungurae became a less conspicuous element of the Omo bovid fauna.
Was this article helpful?