The Pennsylvanian-Permian Vegetational Transition: A Terrestrial Analogue to the Onshore-Offshore Hypothesis STOR William A. DiMichele; Richai'd B. Aronson Evolution, Vol. 46, No. 3 (Jun., 1992), 807-824. Stable URL: http://links.jstor.org/sici?sici=0014-3820%28199206%2946%3A3%3C807%3ATPVTAT%3E2.0.CO%3B2-L Evolution is currently published by Society for the Study of Evolution. Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/teiTns.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http: //w w w .j s tor. org/j oum al s/s se vo 1. html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is an independent not-for-profit organization dedicated to creating and preserving a digital aichive of scholarly journals. For more information regarding JSTOR, please contact [email protected]. http: //w w w .j s tor. org/ WedSep 1 12:53:14 2004 EwluXion. '16(3), 1552, pp. 807-834 THE PENNSYLVANIAN-PERMIAN VEGETATIONAL TRANSITION: A TERRESTRIAL ANALOGUE TO THE ONSHORE-OFFSHORE HYPOTHESIS WILLIAM A. DIMKHELE AND RKHARD B. ARONSON Department of Paieabiology, National Museum of Natural History, Smithsonian Instiimion, Washington, DC 20560 USA Abstract. —An analysis of 68 floras from the Pennsylvanian and Early Permian of Euramerica reveals distinct patterns of environmental distribution. Wetland assemblages are the most com- monly encountered floras from the Early and Middle Pennsylvanian. Floras from drier habitats characterize the Permian. Both wetland and dry-site floras occur in the Late Pennsylvanian, but floristic overlap is minimal, which implies strong environmental controls on the distributions of the component species. Drier habitats appear to be the sites of first appearance of orders that become prominent during the Late Permian and Mesozoic. Higher taxa originated in physically heterogeneous, drier habitats, which were geographically marginal throughout most of the Penn- sylvanian. They then moved into the lowlands during periods of climatic drying in the Permian, replacing older wetland vegetation. This pattern is analogous to the marine onshore-oflfehore pattern of origination and migration. The derivation of Mesozoic wetland clades from the Permian dry- lowland vegetation completes the parallel. The similarities of the marine and terrestrial patterns suggest that the combination of evolutionary opportunity, created by physical heterogeneity of the environment, and migrational opportunity, created by changing extrinsic conditions, may be un- derlying factors that transcend the specifics of organism and environment. Key words—MacTOtvahilion, onshore-offshore hypothesis, paleoecology, Pennsylvanian, Permian, upland flora. Received July 2, 1990. Accepted October 29, 1991. Botanists have long regarded semixeric to dence is relatively recent and continues to xeric habitats as sites favoring the survival accrue. The best documented Paleozoic ex- of morphological novelties (Stebbins, 1952; amples are the conifers (Scott and Chaloner, Axelrod, 1972). In contrast, swamps and 1983; Lyons and Darrah, 1989), eariy fili- associated wetlands have been character- calean fems (Scott and Galtier, 1985), cy- ized as refugia where archaic taxa and com- cads (Mamay, 1976, 1990; Leary, 1990), munities survive for long periods of geo- peltasperms (Kerp, 1988di), and gigantop- logic time (Knoll, 1985; DiMichele et al., terids (Mamay, 1986, 1988). Others that also 1987). The persistence of archaic species and appear to fit the pattern include the cordai- communities in the wet lowlands is a gen- talean gymnosperms (Chaloner, 1958), the eralization drawn from the vast majority of Ginkgoales (Mamay, 1981), the Noeggera- paleobotanical data (for Paleozoic data see thiales (Leary and Pfefferkom, 1977), and Davies, 1929; Oshurkova, 1974, 1978; glossopterids (Retallack and Dilcher, 1981). Havlena, 1970; Phillips etal., 1985). On the Early seed plants, although they may have other hand, the inference of evolutionary originated in Late Devonian wetlands (Gil- innovation in extrabasinal, peripheral, or lespieetal., 1981), radiated subsequently in upland areas (Mapes and Gastaldo, 1987, more xeric habitats (Rothwell and Scheck- discuss terminology) developed primarily ler, 1988; Bateman and Rothwell, 1990; Re- as an explanation for the absence of ex- tallack and Dilcher, 1988). A variety of pected intermediate forms linking the high- enigmatic forms, including genera with un- er taxa found in lowland floras (e.g., Arnold, certain affinities also occur first in what ap- 1947; Axelrod, 1967; Stidd, 1980). pear to be moisture-stressed habitats of the The earliest evidence of most major vas- Early Permian {Glenopteris: Sellards, 1908; cular plant groups (orders approximately) Yakia: White, 1929; RmselUes: Mamay, suggests that they originated or diversified 1968; Wattia: Mamay, 1967; callipterid in habitats that were periodically moisture- complex: Kerp, 1988i; Kerp and Haubold, stressed and often peripheral to major ba- 1988). sinal wetlands. For many groups this evi- In this paper we examine the Pennsyl- 807 808 W, A. DiMICHELE AND R. B. ARONSON vanian-Permian vegetational transition as a of lithologies, or characteristic of a time unit possible analogue of the marine onshore- were rejected from the data base because ofFshore hypothesis. Many of the genera, the degree of species cooccurrence could not families and orders that originated in mois- be assessed reliably. ture-limited habitats during the late Paleo- Eighty taxa, identified to genus, or in some zoic were part of a vegetation that replaced cases to generic groups such as "walchians," the archaic wetland floras of the Carbonif- were included in the data base. Although erous tropics as the lowlands dried out. The using genera reduces ecological precision, it new taxa, inhabitants of the new seasonally avoids the inconsistencies in many species- dry lowlands, became the ancestors of the level taxonomies, and provides a conser- dominant early Mesozoic wetland vegeta- vative estimate of distributional patterns. tion. This pattern of lineage and community replacement may be a terrestrial parallel to Environmental Data the "onshore-offshore" patterns detected in The habitat from which a fossil plant as- the Phanerozoic marine record. Major mor- semblage was drawn can be described in a phological innovations in benthic inverte- variety of ways, often in great detail. For brates first appeared in physically hetero- the purpose of this analysis we were con- geneous, disturbed nearshore environments cerned with just two attributes: local mois- and then expanded offshore into more sta- ture hmitation and physical location rela- ble deep-water environments (Jablonski et tive to basinal lowlands. Unfortunately there al., 1983; SepkoskiandSheehan, 1983; Sep- are significant terminological difficulties as- koski and Miller, 1985; Bottjer and Jablon- sociated with the description of these fea- ski, 1988; Jablonski and Bottjer, 1990; Sep- tures, resulting in part from their being con- koski, 1991). Despite possible differences in founded (e.g., all lowlands are wet, all factors mediating the habitat shifts, the sim- uplands are dry). Such problems are dis- ilarities between the marine and terrestrial cussed and partially clarified by Havlena patterns are striking. Both patterns may re- (1970) and Mapes and Gastaldo (1987), but flect a fundamental relationship between considerable confusion remains. evolution and environment. We have avoided the term "stress" due to the ambiguity of its usage (see DiMichele etal., 1987). Swamps, for example, are high- DATA AND ANALYSIS ly "stressful" in the sense of their edaphics; Floristic Data physico-chemical properties of the sub- Roras were compiled from 68 North strate exclude nearly all plants and create a American and European localities, part of sharp, nongradational boundary with sur- a continuous tropical landmass during the rounding habitats (heavy metal-rich sub- late Paleozoic (Ziegler, 1990). The floras strates have the same properties, see Brad- range in age from Early Pennsylvanian to shaw and McNeilly, 1981 and references Late Permian. A "flora" in the sense used therein). Seasonally dry habitats also are here is a collection made from a single de- "stressful," but in a different sense. A gra- positional environment at a single site. A dient exists from habitats or microhabitats site can vary from a meter square prospect with constantly available moisture to those pit to the roof shale of a coal exposed in one that are moisture "stressed." The evolu- strip mine. Floras from depositional envi- tionary response of plants to edaphic stress ronments suggesting a high potential for al- is quite different from their response to lochthonous origin were excluded from con- moisture limitation. Swamps, for example, sideration due to potential mixing of are not centers of evolutionary innovation. vegetational types from edaphically distinct Over geological time they tend to accu- parts of the landscape. Temporal hetero- mulate species that can tolerate