Search

Late Jurassic--Cretaceous are-related volcaniclastic rocks from the southern Guerrero and western Mixteca terranes of Mexico were analyzed by U-Pb detrital zircon geochronology (laser ablation-multicollector--inductively coupled plasma--mass spectroscopy) to place constraints on the depositional history and provenance of the rocks. Pre--Middle ,Jurassic basement rocks and sandstone from the Upper Cretaceous Mexcala Formation were also analyzed to define the origin and provenance of the prevolcanic substratum, and the time of accretion of Guerrero composite terrane sequences. Data from the Taxco-Taxco Viejo, Teloloapan, and Arcelia assemblages indicate that the youngest (129-141 Ma) zircon fraction in each sequence was derived from local volcanic sources, whereas older populations (ca. 247-317, 365-459, 530-617, 712-878, 947-964, 1112-1188, 1350-1420, 1842-1929, 2126-2439, and 2709-3438 Ma) show sediment influx from varied sources, most likely through grain recycling. The major zircon clusters in these sequences match the populations recorded in the nearby Acatlan Complex. In contrast, the Huetamo sample is dominated by Lower Cretaceous (ca. 126 Ma) zircons of local volcanic provenance, and the Zihuatanejo sample contains zircon clusters (ca. 259, ca. 579, and ca. 947-1162 Mat comparable to major populations recorded in the underlying Arteaga Complex. A sample from the Middle Triassic--Middle Jurassic Arteaga Complex at Tzitzio contains zircon clusters (ca. 202-247, ca. 424, ca. 600, ca. 971, and ca. 2877 Ma) consistent with an ultimate derivation from both North American and South American sources. The sample from the Las Ollas suite contains comparable zircon populations (ca. 376-475, ca. 575, ca. 988-1141, and ca. 2642-2724 Ma). and it is interpreted to be part of the prevolcanic basement. In contrast, the youngest zircon cluster (ca. 105 Ma) in the Mexcala Formation coincides with the major volcanic events in the Taxco-Taxco Viejo, Teloloapan, and Arcelia assemblages, whereas the older clusters (ca. 600, ca. 953, ca. 1215, ca. 1913, and ca. 2656-2859 Ma) broadly match the major populations recorded in rocks from the Acatlan Complex. These new data combined with available geochemical and isotopic data indicate that the Taxco-Taxco Viejo arc assemblage developed on continental crust. The Acatlan Complex is the most plausible candidate. The Teloloapan and Arcelia arc assemblages were developed on oceanic crust as offshore arcs facing the Acatlan Complex. The Zihuatanejo terrane assemblages were developed on the Arteaga Complex, and evidence no influence from the Acatlan Complex. This suggests that these assemblages were formed farther away or in a restricted basin. The Guerrero composite and Mixteca arc successions are coeval with the Alisitos arc of northern Mexico and in part with the Nevada and Klamath ranges of the southwestern United States, and with the arc series from the Greater and Lesser Antilles and northern South America. Data indicate that during late Mesozoic time, southwestern North America was a site of intensive volcanism in a complex arc-trench system similar to that of the east Pacific. Our data are consistent with a diachronic accretion of the Guerrero composite terrane sequences, beginning during late Cenomanian time with the amalgamation of the Teloloapan and probably the Arcelia assemblages, and finishing at the end of Cretaceous time with the accretion of the Zihuatanejo terrane assemblages. Keywords: U-Pb detrital zircon, Late Jurassic-Cretaceous arc successions, Guerrero terrane, Mixteca terrane, southern Mexico, southwestern North America, Caribbean region.

Many river systems in western North America retain a fluvial strath-terrace record of discontinuous downcutting into bedrock through the Quaternary. Their importance lies in their use to interpret climatic events in the headwaters and to determine long-term incision rates. Terrace formation has been ascribed to changes in sediment supply and/or water discharge produced by late Quaternary climatic fluctuations. We use a one-dimensional channel-evolution model to explore whether temporal variations in sediment and water discharge can generate terrace sequences. The model includes sediment transport, vertical bedrock erosion limited by alluvial cover, and lateral valley-wall erosion. We set limits on our modeling by using data collected from the terraced Wind River basin. Two types of experiments were performed: constant-period sinusoidal input histories and variable-period inputs scaled by the marine [[delta].sup.18]O record. Our simulations indicate that strath-terrace formation requires input variability that produces a changing ratio of vertical to lateral erosion rates. Straths are cut when the channel floor is protected from erosion by sediment and are abandoned--and terraces formed--when incision can resume following sediment-cover thinning. High sediment supply promotes wide valley floors that are abandoned as sediment supply decreases. In contrast, wide valleys are promoted by low effective water discharge and are abandoned as discharge increases. Widening of the valley floors that become terraces occurs over many thousands of years. The transition from valley widening to downcutting and terrace creation occurs in response to subtle input changes affecting local divergence of sediment-transport capacity. Formation of terraces lags by several thousand years the input changes that cause their formation. Our results suggest that use of terrace ages to set limits on the timing of a specific event must be done with the knowledge that the system can take thousands of years to respond to a perturbation. The incision rate calculated in the field from the lowest terrace in these systems will likely be higher than the rate calculated by using older terraces, because the most recent fluvial response in the field is commonly downcutting associated with declining sediment input since the Last Glacial Maximum. This apparent increase in incision rates is observed in many river systems and should not necessarily be interpreted as a response to an increase in rock-uplift rate. Keywords: fluvial features, landscape evolution, modeling, river terraces, sediment supply.

Stratal patterns of the Middle Ordovician Hagan K-bentonite complex and associated rocks show that the Black River-Trenton unconformity in the North American midcontinent formed through the complex interplay of eustasy, sediment accumulation rates, siliciclastic influx, bathymetry, seawater chemistry, and perhaps local tectonic uplift. The unconformity is diachronous and is an amalgamated surface that resulted from local late Turinian lowstand exposure followed by regional early Chatfieidian transgressive drowning and sediment starvation. The duration of the unconformity is greatest in southern Wisconsin, northern Illinois, and northern Indiana, where the Deicke and Millbrig K-bentonite Beds converge at the unconformity. On the basis of published isotopic ages for the Deicke and Millbrig beds, it is possible that in these regions erosion and nondeposition spanned a period of as much as 3.2 m.y. Two broad coeval depositional settings are recognized within the North American midcontinent during early Chatfieldian time. (1) An inner shelf, subtidal facies of fossiliferous shale (Spechts Ferry Shale Member and Ion Shale Member of the Decorah Formation) and argiliaceous lime mudstone and skeletal wackestone (Guttenberg and Kings Lake Limestone Members) extended from the Canadian shield and Transcontinental arch southeastward through Minnesota, Wisconsin, Iowa, and Missouri. (2) A seaward, relatively deep subtidal, sediment-starved, middle sheff extended eastward from the Mississippi Valley region to the Taconian foreland basins in the central and southern Appalachians and southward through the pericratonic Arkoma and Black Warrior basins. In the inner shelf region, the Black River-Trenton unconformity is a composite of at least two prominent hardground omission surfaces, one at the top of the Castlewood and Carimona Limestone Members and the other at the top of the Guttenberg and Kings Lake Limestone Members, both merging to a single surface in the middle shelf region. The inner and middle shelves redeveloped later in approximately the same regions during Devonian and Mississippian tine.

Comparisons of the physiognomy of leaves from modern vegetation of known climates with that of Eocene and Oligocene fossil leaf assemblages from middle latitudes of western North America indicate paleoaltitudes comparable or higher than those at present. Using canonical correspondence analysis, a multivariate statistical approach that includes nonlinear relationships between characters and environmental parameters, we relate leaf physiognomy of fossil samples to modern vegetation of known environmental parameters, including moist enthalpy of the atmosphere, a thermodynamically conserved quantity of the atmosphere that varies with altitude. By estimating enthalpy values for both lowland leaf assemblages and coeval interior leaf assemblages, altitudes for the interior assemblages can be estimated to [+ or -]890 m. In northeastern Washington and southern British Columbia, the high Eocene altitudes indicate subsidence, or collapse, of an area undergoing crustal extension and may reflect an immediately preceding period of uplift that triggered the extension. Similarly, other areas that have undergone Cenozoic crustal extension appear to have been at least as high as they are at present. Three sites from the Rocky Mountains also indicate elevations at least as high as at present, and therefore suggest subsidence, either resulting from cooling of a hot upper mantle or erosion and isostatic compensation of surrounding terrain. High altitudes during Eocene and Oligocene time in western North America appear to have been normal, even in areas such as the Green River basin, and therefore cast doubt on the commonly inferred late Cenozoic uplift of that region.

The nature of the Great Valley basement, whether oceanic or continental, has long been a source of controversy. A velocity model (derived from a 200-km-long east-west reflection-refraction profile collected south of the Mendocino triple junction, northern California, in 1993), further constrained by density and magnetic models, reveals an ophiolite underlying the Great Valley (Great Valley ophiolite), which in turn is underlain by a westward extension of lower-density continental crust (Sierran affinity material). We used an integrated modeling philosophy, first modeling the seismic-refraction data to obtain a final velocity model, and then modeling the long-wavelength features of the gravity data to obtain a final density model that is constrained in the upper crust by our velocity model. The crustal section of Great Valley ophiolite is 7-8 km thick, and the Great Valley ophiolite relict oceanic Moho is at 11-16 km depth. The Great Valley ophiolite does not extend west beneath the Coast Ranges, but only as far as the western margin of the Great Valley, where the 5-7-km-thick Great Valley ophiolite mantle section dips west into the present-day mantle. There are 16-18 km of lower-density Sierran affinity material beneath the Great Valley ophiolite mantle section, such that a second, deeper, 'present-day' continental Moho is at about 34 km depth. At mid-crustal depths, the boundary between the eastern extent of the Great Valley ophiolite and the western extent of Sierran affinity material is a near-vertical velocity and density discontinuity about 80 km east of the western margin of the Great Valley. Our model has important implications for crustal growth at the North American continental margin. We suggest that a thick ophiolite sequence was obducted onto continental material, probably during the Jurassic Nevadan orogeny, so that the Great Valley basement is oceanic crust above oceanic mantle vertically stacked above continental crust and continental mantle.

Models integrating geodynamic and stratigraphic processes show that some gross features of Phanerozoic North American cratonic strata can be explained with dynamic topographies generated by subduction and cycles of supercontinent aggregation and dispersal. A three-dimensional finite-element model is used to calculate mantle flow beneath North America during Phanerozoic time in response to episodes of subduction at cratonic margins and two cycles of supercontinent formation and breakup. Dynamic topographies calculated by the flow models are used as input to a stratigraphic model that also includes background subsidence, eustasy, denudation, clastic and carbonate deposition, compaction, and isostasy. These models successfully reproduce aspects of the Sloss sequences; the best matches were obtained by combining two wavelengths of dynamic topography with second-order eustasy. Long-wavelength dynamic topography generates first-order stratal cyclicity. Periods of erosion were shorter when North America was over a dynamic topography low than when it was over a high. Long-wavelength dynamic topography also explains the absence of Mesozoic strata on the eastern portion of the craton. Characteristic stratal patterns are shown to result from subduction-related dynamic topography, although sensitive to sediment supply and other subsidence mechanisms. Aspects of Upper Cretaceous stratal patterns may be explained by the effects of Farallon plate subduction. Generally, strata deposited in a dynamic topography depression have low preservation potential because the topography is reversible. Thus, ancient subduction-related dynamic topography is most likely to be represented by unconformities.

A magnetic polarity stratigraphy and a corresponding paleomagnetic pole position are reported from 113 sampling sites representing 3000 m of Upper Triassic continental sedimentary rocks that crop out in the Dan River-Danville basin of North Carolina and Virginia. Characteristic magnetizations isolated by thermal demagnetization for either the hematite-bearing red siltstones or the interbedded magnetite-bearing gray to black mudstones of the Leakesville Formation are indistinguishable in mean direction and pass reversal tests. The magnetic polarity sequence consists of 11 magnetozones that vary from [approximately equal to]100 m to 800 m in thickness and can be uniquely correlated within biostratigraphic constraints to magnetochrons E9n to E14n of the Newark geomagnetic polarity time scale. According to this correlation, the sampled section is the age equivalent of the uppermost Stockton, the entire Lockatong, and the low-ermost Passaic formations of the Newark basin, and represents [approximately equal to]7.5 m.y. of deposition. The late Carnian Dan River-Danville paleopole is located at 55.4 [degrees] N 100.1 [degrees] E ([A.sub.95] = 1.9 [degrees]), which is not significantly different from paleopoles reported from essentially coeval rocks in the Newark basin. Considering that the Dan River-Danville and Newark basins are [approximately equal to]600 km apart, the close agreement of the coeval paleopoles argues strongly for the overall tectonic coherence of these rift basins with respect to each other and, most probably, with respect to cratonic North America. Discordant latest Triassic paleopoles from the southwestern United States, which have tended to be attributed to fast apparent polar wander for North America in the Late Triassic and predict anomalously high paleolatitudes for eastern North America, are best accounted for by a large net clockwise rotation of the Colorado Plateau.

The Newark Supergroup currently includes nine stratigraphic groups, each of which applies to part or all of the rock column of only one or a few basins. Because the group nomenclature within the Newark Supergroup is neither inclusive nor parallel in its concepts, nearly half of the strata within the Newark Supergroup lacks any group placement. A new system is proposed herein that (1) establishes unambiguous group boundaries, (2) places all Newark Supergroup strata into groups, (3) reduces the number of group names from nine to three, (4) creates parallelism between groups and three major successive tectonic events that created the rift basins containing the Newark Supergroup, and (5) coincidentally provides isochronous or nearly isochronous group boundaries. These proposed groups are (1) the Chatham Group (Middle Triassic to basal Lower Jurassic sedimentary rocks), (2) the Meriden Group (Lower Jurassic extrusive volcanic and sedimentary rocks), and (3) the Agawam Group (new name) (Lower Jurassic sedimentary rocks above all early Mesozoic igneous intrusive and extrusive rocks). This new rock classification system makes use of the fact that a discrete interval of synchronous or nearly synchronous volcanism and plutonism occurred throughout the early Mesozoic rift system of eastern North America. The presence or absence of volcanic rocks provides a powerful stratigraphic tool for establishing regional groups and group boundaries. The presence of sedimentary rocks injected by diabase dikes and sills, in the absence of extrusive volcanic rocks, places Newark Supergroup rocks in the Chatham Group. The presence of extrusive volcanic rocks, interbedded with sedimentary rocks injected by dia-base dikes and sills, places Newark Super-group rocks in the Meriden Group. The presence of sedimentary rocks lacking both extrusive volcanic rocks and diabase dikes and sills, places Newark Supergroup rocks in the Agawam Group. Application of this new regional group stratigraphy to the early Mesozoic rift basins requires revision of the stratigraphy of several basins to make formation boundaries match group boundaries.

To understand the patterns of lithofacies, marine faunas, organic-carbon enrichment, isotopes, and trace elements deposited in the early Turonian Western Interior seaway, we conducted circulation experiments using a three-dimensional, turbulent flow, coastal ocean model driven by GENESIS, a climate model developed at the National Center for Atmospheric Research (NCAR). Circulation and chemical evolution of the seaway waters are computed under the following initial and boundary conditions: (1) paleobathymetry according to a new interpretation of the lithostratigraphy and biostratigraphy; (2) temperatures and salinities of the Boreal and Tethys oceans and adjacent drainage basins based on isotopic data, atmospheric temperatures, and precipitation-evaporation magnitudes computed by GENESIS; and (3) mean annual wind stresses over the seaway computed by GENESIS. Results show that the seaway exported freshened water much like Hudson Bay today. Runoff from eastern drainages exited the seaway as a northern coastal jet; runoff from western drainages exited as a southern coastal jet. Both jets simultaneously drew in surface Tethyan and Boreal waters, creating a strong counterclockwise gyre occupying the entire north-south extent of the seaway. The curious stratal and faunal variations of the early Turonian deposits arise from this gyre and its associated water masses.

The Pacific plate moved northwest relative to North America since 42 Ma. The rapid half rate of Pacific-Farallon spreading allowed the ridge to approach the continent at about 29 Ma. Extinct spreading ridges that occur offshore along 65% of the margin (Lonsdale, 1991) document that fragments of the subducted Farallon slab became captured by the Pacific plate and assumed its motion prior to the actual subduction of the spreading ridge. This plate-capture process can be used to explain much of the post-29 Ma Cordilleran North America extension, strike slip, and the inland jump of oceanic spreading in the Gulf of California. The Pacific and North American contact zone lengthened with each successive plate capture event, underpinning the parts of western North America directly inland with a strong plate undergoing Pacific relative motion. We suggest that much of the post-29 Ma continental tectonism is the result of the strong traction imposed on the deep part of the continental crust by the gently inclined slab of subducted oceanic lithosphere as it moved to the northwest relative to the overlying continent. The plate-capture hypothesis is distinctly different from theories involving shallow slab gaps. Kinematic problems associated with shallow slab-gap models cause us to question them. This conclusion is consistent with seismic refraction interpretations that suggest there is an inclined layer with high velocities like that of basalt or gabbro at the base of the continental crust beneath much of the Californian margin and the documented reduction of slab-pull forces and density associated with young subducting slabs. Thermal and rheologic modeling suggests that coastal California was a strong zone at all depths allowing it to be firmly linked to Pacific motion. Our model shows that deformed regions such as the basin and range and borderland provinces developed in predicted weak parts of the crustal section, but they have been incompletely linked to the deep plate across the ductile middle and lower crustal layer.

Variation in accumulation rate as a function of depositional latitude has been determined for North American limestone, dolostone, sandstone, gray-black shale, red shale, and evaporite rocks from tabulations by Ronov and co-workers, from data in lexicons of United States and Canadian stratigraphy, and from isopach and lithofacies maps published in 1975 by Cook and Bally. Phanerozoic sediment is similar to that on other continents, consisting of 63% terrigenous clastic facies (23% sandstone; 40% shale), 36% carbonate, and 1% chert, evaporite, and other minor rock types. Latitudinal variation in accumulation rate is influenced primarily by past positions of foreland and intracratonic basins. Additional variation related to paleoclimate is recorded as strong dependence of limestone, dolostone, red shale, and evaporite deposition on latitudes of accumulation, a relation reflecting the importance of ambient temperature and meteoric precipitation on the formation of various authigenic phases. Poleward increase in dolostone/total carbonate ratio in both hemispheres implicates the importance of syndepositional diagenesis in the formation of most dolostone sequences. The ratio of sandstone/total terrigenous clastic sediment exhibits little variation with latitude, a constancy indicating that most sand and mud is derived from older sediment rather than from the weathering of crystalline rock. As a result, terrigenous lithofacies record little climatic influence in the proportional delivery of mechanically versus chemically derived weathering products to basins of accumulation.

The Newark basin is a half graben, bounded on its northwestern margin by a system of mostly normal-slip faults, many of which are reactivated Paleozoic thrust faults and were active at least sporadically during the deposition of the entire basin fill. As a result of along-strike variations in displacement on the border-fault system, subsidence increased from the lateral edges toward the center of the basin and from the hinged margin toward the border-fault system. Some intrabasinal faults also formed during the deposition of units currently preserved in their hanging walls. Along-strike variations in the density of intrabasinal faults probably reflect changes in the dip angle of the border-fault system; variations in the amount of extension accommodated on these intrabasinal faults resulted in the formation of a transfer fault. Along-strike variations in fault displacement probably produced the transverse folds in the hanging walls immediately adjacent to border faults and major intrabasinal normal faults; some folds formed syndepositionally. Progressively younger strata onlap 'basement' rocks along the hinged margin of the basin, and younger splays of the border-fault system propagated into the footwall; the basin therefore grew deeper, wider, and longer through time.

Two stratigraphic sequences were sampled through the early Anisian Anton Chico Member of the Moenkopi Formation in northeastern New Mexico. Two polarities of magnetization are present: a normal-polarity interval succeeded by a reversed-polarity interval and followed by a short normal- and reversed-polarity couplet. Detailed thermal demagnetization (10 to 17 steps) was employed to separate magnetic vectors. The secondary magnetization is largely a direction similar to the present-day and/or axial-field directions. Demagnetization above 520 [degrees] C reveals a near-horizontal characteristic magnetization. The lithology of the stratigraphic sequence is such that the reversed polarity is contained almost exclusively within coarse sandstone lithologies, and hence, the reversed-polarity characteristic magnetization direction is rarely completely separated from the secondary magnetization. Because of this, the paleopole was calculated from only the lower normal-polarity portion of the section. The pole, calculated from the samples of two localities, is located at 121.4 [degrees] E, 43.2 [degrees] N, ([[Alpha].sub.95] = 5.3). This Middle Triassic paleopole is in good agreement with published Triassic paleomagnetic poles for cratonic North America and statistically overlaps the Early Triassic paleopoles. The large-scale relative motion during the Triassic between the magnetic pole and the North American plate is constrained by this study to have begun after early Middle Triassic time; it suggests that as much as 10 [degrees] of apparent-polar wander occurred between the late Anisian-Ladinian (late Middle Triassic) and the middle to late Carnian (early Late Triassic), a time interval of between 11 and 14 m.y.

Contraction in the western North American Cordillera, exclusive of accreted terranes, occurred over a remarkably uniform duration of 110 [+ or -] 15 m.y. between approximately 165 and 55 Ma. The amount of contraction, as much as 285 km, was similar from southeastern British Columbia to northern Nevada and Utah, although irregularly partitioned between hinterland, fold and thrust belt, and Rocky Mountain foreland. Time-averaged contraction rates are 1.5 to 2.5 mm/yr. The contraction, however, appears to have occurred in pulses with cumulative duration of 75 m.y., implying rates during active periods of 2.5 to 3.8 mm/yr. Displacement rates as high as 6 mm/yr on some thrust faults suggest episodicity within pulses. Episodic contraction on 10- to 30-m.y. intervals corresponds well to episodicity in relative plate motion, suggesting rapid response of intracontinental kinematics to plate-boundary interactions. The relationship is nevertheless indirect, possibly influenced by the position and number of plates, partitioning of displacements across accreted terranes, and crustal structure.

Promontories and embayments along the late Precambrian-early Paleozoic Appalachian-Ouachita continental margin of southeastern North America are framed by a northeast-striking rift system offset by northwest-striking transform faults. Inboard from the continental margin, basement fault systems have two sets of orientation; one is northeast parallel with rift segments, and the other is northwest parallel with transform faults. Late Precambrian castic and volcanic synrift rocks overlie Precambrian basement rocks along the Appalachian Blue Ridge. Lower Cambrian sandstone at the base of a transgressive passive-margin succession oversteps the rift-fill successions and basement rocks, defining the time of transition from an active rift to a passive margin along the Blue Ridge. Locally thick Early Late Cambrian and older sedimentary rocks fill downthrown blocks of the intracratonic Mississippi Valley-Rough Creek-Rome graben system and Birmingham basement fault system. These basement fault systems, which indicate northwest-southeast extension like the Blue Ridge rift, are overstepped by Upper Cambrian strata. The northwest-striking Southern Oklahoma fault system is interpreted to be a transform fault that propagated into the continent from the Ouachita rift. Early and Middle Cambrian rift-related igneous rocks along the fault system and adjacent Precambrian basement are overstepped by Upper Cambrian sandstone. The differences in age of rift-related rocks suggest a spreading-center shift at the beginning of the Cambrian Period from the Blue Ridge rift to the Ouachita rift southwest of the Alabama-Oklahoma transform fault. From Early to Early Late Cambrian, a small component of extension propagated northeastward to form the intracratonic fault systems northeast of the transform fault, but most of the extension of the Ouachita rift was transformed along the Alabama-Oklahoma transform fault to the Mid-Iapetus Ridge outboard from the Blue Ridge passive margin.

Twenty-two numerical experiments using a multi-layer, numerical model of turbulent flow in shallow seas better define fluid circulation and sediment transport paths in the Cretaceous Interior Seaway of North America. Each experiment consists of a different combination of paleogeography, paleobathymetry, Coriolis acceleration, boundary tidal amplitudes, wind stresses, and bed friction. The paleogeographies and paleobathymetries represent three seaway configurations: a seaway of intermediate size and depth (200 m) during a transgressive peak (T5) in late Albian time; a seaway of maximum size and depth (400 m) during peak transgression (T6) in early Turonian time; and a seaway of minimum size and depth (100 m) during a peak regression (R8) in the Campanian. Values for the other key model parameters are: [M.sub.2] co-oscillating boundary tides of 0.1 to 0.2 m range at the Arctic Ocean boundary and 0.5 to 1 m range at the prot-Gulf of Mexico boundary; Coriolis accelerations corresponding to 30[degrees]N, 45[degrees]N, and 60[degrees]N latitude; Chezy friction factors ranging from 31 to 70 [m.sup.1/2][s.sup.-1]; and average winter winds and two winter storms computed by the community climate model at the National Center for Atmospheric Research for paleogeographic conditions during the late Albian. Results of the numerical experiments, and comparison of these results with geologic observations, allow the following conclusions. (1) Circulation in the seaway was generally storm-dominated. (2) Typical winter storms crossing the seaway from west to east generated 0.3 m [s.sup.-1] shore-parallel, geostrophic currents on the shelves north of Arizona, at first flowing weakly to the north, but later during the storm, flowing strongly to the south. (3) Extreme storms could have produced 0.8 m [s.sup.-1] currents over the shelves, and 0.3 m [s.sup.-1 currents in 200 m of water. (4) Co-oscillating tides propagated into the seaway as progressive Kelvin waves, and therefore tidal ranges in the seaway were mesotidal to macrotidal along the southeastern margin but microtidal everywhere else. (5) Significant deep-water wave heights and periods of the fully developed wave field are predicted to have been 5 to 6 m high and 10 s, respectively. The northwestern shore of the seaway would have experienced these storm waves approaching from the north, and thus net sediment drift should have been to the sotth. Limited available field data support these conclusions.

The spectacular evolutionary history of terrestrial vertebrates is uncharacteristically poorly documented between the Paleozoic subequatorial exposures of the Permo-Carboniferous of North America and Europe, and the much higher latitude, Upper Permian deposits of central Russia and southern Africa. We report here that the discovery of the reptile Macroleter in Oklahoma provides the first direct vertebrate evidence of biochronological correlation between continental sediments from the Upper Permian of North America and Russia. The presence of the reptile Macroleter, a member of a major group of Upper Permian amniotes known previously only from central Russia, in North America improves dramatically our understanding of this early phase of amniote evolution, and also provides evidence, of terrestrial tetrapod faunal interchange between North America and Russia in Late Permian time. Keywords: paleobiogeography, paleontology, Permian, terrestrial reptile.

We present detailed isotope stratigraphies for Carboniferous time based on brachiopod shell calcite from the midcontinent region of North America. Evidence for shell calcite preservation includes (1) preservation of shell microstructure, (2) lack of cathodoluminescence, (3) low Si, Al, Fe, and Mn contents, (4) Na, Sr, and S contents comparable to those of modern brachiopod shells, and (5) [[Delta].sup.13]C and [[Delta].sup.18]O values higher than those of associated cements and matrix. The Carboniferous [[Delta].sup.13]C record for North America is characterized by three isotopic stages. The earliest stage, C1, follows a 2.0[per thousand] increase in Kinderhookian time (early Tournaisian), from 1.5[per thousand] to 3.5[per thousand], and includes a brief and perhaps local late Kinderhookian excursion to 5.4[per thousand]. The [[Delta].sup.13]C values remain stable at 3.5[per thousand] to 4[per thousand] during stage C1, then decrease about 1[per thousand] near the Meramecian-Chesterian boundary (Visean) to 2[per thousand]-3[per thousand] (stage C2). Stage C2 ends with a 1[per thousand]-2[per thousand] increase (C2-C3 transition) between middle Chesterian and early Morrowan time (Serpukhovian-Bashkirian). Stage C3 values remain mostly between 3[per thousand] and 4.5[per thousand] upsection to Virgilian strata (Gzhelian). Increases in [[Delta].sup.13]C probably reflect global increases in sedimentary organic carbon burial and suggest that p[CO.sub.2] declines in the earliest and middle Carboniferous strata. The middle Carboniferous [[Delta].sup.13]C shift of B. Popp, T. Anderson, and P. Sandberg, an [approximately]3[per thousand] increase in European sections, occurs in North America (C2-C3 transition) but is limited to [approximately]1.5[per thousand]. This 1.5[per thousand] increase was probably caused by increased organic carbon burial, whereas the additional [approximately]1.5[per thousand] shift in European sections likely reflects changes in ocean circulation patterns associated with the closing of the Equatorial seaway. based on the timing of the [[Delta].sup.13]C divergence between North America and Europe, the isolation of the Paleotethys began in late Chesterian time (Serpukhovian). The [[Delta].sup.18]O record can also be separated into the three stages. There is a 3[per thousand] increase during Kinderhookian-Osagean time (Tournaisian), corresponding to the Devonian to Carboniferous transition to stage C1, a 3[per thousand] decrease during Meramecian-early Chesterian time (Visean; C1-C2 transition), then a 2[per thousand] increase in late Chesterian-early Morrowan time (Serpukhovian-Bashkirian; C2-C3 transition). The [[Delta].sup.18]O values then fluctuate between -1[per thousand] and -3[per thousand] (C3 stage) upsection to the Virgilian strata (Gzhelian). If global, the 2[per thousand] to 3[per thousand] [[Delta].sup.18]O shifts are compelling evidence for cooling and glaciation in Early Mississippian time, warming and deglaciation in Late Mississippian time, and a return to cool, glacial conditions in earliest Pennsylvanian time. The general correlation between [[Delta].sup.13]C and [[Delta].sup.18]O shifts suggests that cooling is associated with drawdown of atmospheric C[O.sub.2].