The early Eocene (ca. 55-48 Ma) encompasses one of the warmest intervals of the past 65 m.y. and is characterized by an unusually low equator-to-pole thermal gradient. Recent proxy studies suggest temperatures well in excess of 30 °C even at high latitudes, but confl icting interpretations derived from different types of data leave considerable uncertainty about actual early Eocene temperatures. A robust comparison among new paleotemperature proxies may provide insight into possible biases in their temperature estimates, and additional detail on the spatial distribution of temperatures will further resolve the early Eocene meridional temperature gradient. We use a suite of paleotemperature proxies based on the chemistry of bivalve shell carbonate and associated sedimentary organic matter from the United States Gulf Coastal Plain to constrain climate at a subtropical site during this key interval of Earth history. Oxygen isotope and clumped isotope analyses of shell carbonate and two tetraether lipid analyses of sedimentary organic carbon all yield temperatures of ~27 °C. High-resolution, intraannual oxygen isotope data reveal a consistent, large range of seasonal variation, but clumped isotope data suggest that seasonality is due primarily to precipitation, not to temperature. These paleotemperature estimates are 2-3 °C warmer than the northern Gulf of Mexico today, and generally consistent with early Eocene temperature estimates from other low and mid-latitude locations, but are signifi cantly cooler than contemporaneous estimates from high southern latitudes.

The sizes and shapes of marine organisms often vary systematically across latitude and water depth, but the environmental factors that mediate these gradients in morphology remain incompletely understood. A key challenge is isolating the individual contributions of many, often correlated, environmental variables of potential biological significance. Benthic foraminifera, a diverse group of rhizarian protists that inhabit nearly all marine environments, provide an unparalleled opportunity to test statistically among the various potential controls on size and volume–to–surface area ratio. Here, we use 7035 occurrences of 541 species of Rotallid foraminifera across 946 localities spanning more than 60 degrees of latitude and 1600 m of water depth around the North American continental margin to assess the relative influences of temperature, oxygen availability, carbonate saturation, and particulate organic carbon flux on their test volume and volume–to–surface area ratio. For the North American data set as a whole, the best model includes temperature and dissolved oxygen concentration as predictors. This model also applies to data from the Pacific continental margin in isolation, but only temperature is included in the best model for the Atlantic. Because these findings are consistent with predictions from the first principles of cell physiology, we interpret these statistical associations as the expressions of physiological selective pressures on test size and shape from the physical environment. Regarding existing records of temporal variation in foraminiferal test size across geological time in light of these findings suggests that the importance of temperature variation on the evolution of test volume and volume–to–surface area ratio may be underappreciated. In particular, warming may have played as important a role as reduced oxygen availability in causing test size reduction during past episodes of environmental crisis and is expected to inflict metabolic stress on benthic foraminifera over the next century due to anthropogenic climate change.

No abstract

Spatial variation in environmental conditions can elicit predictable size and morphological responses in marine organisms through influences on physiology. Thus, spatial and temporal variation in marine organism size and shape are often used to infer paleoenvironmental conditions, such as dissolved oxygen concentrations. Benthic foraminifera commonly serve as a tool for reconstructing past ocean oxygen levels. For example, benthic foraminiferal species assemblages, within‐ and among‐species patterns of test morphology, and geochemical analyses of carbonate tests are often employed to reconstruct past marine conditions. In this study, we measured the sizes and shapes of modern foraminifera representing four species that inhabit a steep oxygen gradient in the Santa Monica Basin on the Southern California Borderland with the aim of quantifying the influence of oxygen availability on foraminiferal morphology, both within and among species. Most foraminifera rely on aerobic respiration, but the four benthic foraminifera from the Santa Monica Basin do not show the predicted size and morphological responses to variations in dissolved oxygen concentrations based on first principles of cell physiology: Bolivina spissa shows no volume or volume‐to‐surface area response, Uvigerina peregrina increases in both test volume and volume‐to‐surface area ratio with decreasing dissolved oxygen concentrations, and both Bolivina argentea and Loxostomum pseudobeyrichi decrease in test volume, but only L. pseudobeyrichi shows a decrease in test volume‐to‐surface area ratio with decreasing oxygen concentrations as expected from physiological predictions. These findings imply that the morphological responses of individual foraminiferal species are not necessarily representative of the responses of other foraminiferal species within the community. Our findings further suggest that these species use physiological strategies such as depressed metabolic rates and alternative energy metabolisms to persist in low oxygen environments and, therefore, cannot be used in any simple way as paleo‐oxygen indicators. Should Proterozoic (1,000–542 million years ago) protists have possessed metabolic strategies similar to foraminifera, the sizes and shapes of protists in the fossil record may not usefully constrain ambient oxygen conditions during the appearance and initial taxonomic radiation of heterotrophic eukaryotes.

Energy availability influences natural selection on the ontogenetic histories of organisms. However, it remains unclear whether physiological controls on size remain constant throughout ontogeny or instead shift as organisms grow larger. Benthic foraminifera provide an opportunity to quantify and interpret the physicochemical controls on both initial (proloculus) and adult volumes across broad environmental gradients using first principles of cell physiology. Here, we measured proloculus and adult test dimensions of 129 modern rotaliid species from published images of holotype specimens, using holotype size to represent the maximum size of all species’ occurrences across the North American continental margin. We merged size data with mean annual temperature, dissolved oxygen concentration, particulate organic carbon flux, and seawater calcite saturation for 718 unique localities to quantify the relationship between physicochemical variables and among-species adult/proloculus size ratios. We find that correlation of community mean adult/proloculus size ratios with environmental parameters reflects covariation of adult test volume with environmental conditions. Among-species proloculus sizes do not covary identifiably with environmental conditions, consistent with the expectation that environmental constraints on organism size impose stronger selective pressures on adult forms due to lower surface area-to-volume ratios at larger sizes. Among-species adult/proloculus size ratios of foraminifera occurring in resource-limited environments are constrained by the limiting resource in addition to temperature. Identified limiting resources are food in oligotrophic waters and oxygen in oxygen minimum zones. Because among-species variations in adult/proloculus size ratios from the North American continental margin are primarily driven by the local environment’s influence on adult sizes, the evolution of foraminiferal sizes over the Phanerozoic may have been strongly influenced by changing oceanographic conditions. Furthermore, lack of correspondence between among-species proloculus sizes and environmental conditions suggests that offspring sizes in foraminifera are rarely limited by physiological constraints and are more susceptible to selection related to other aspects of fitness.