The Dry Creek archeologic site contains a stratified record of late Pleistocene human occupation in central Alaska. Four archeologic components occur within a sequence of multiple loess and sand layers which together form a 2-m cap above weathered glacial outwash. The two oldest components appear to be of late Pleistocene age and occur with the bones of extinct game animals. Geologic mapping, stratigraphic correlations, radiocarbon dating, and sediment analyses indicate that the basal loess units formed part of a widespread blanket that was associated with an arctic steppe environment and with stream aggradation during waning phases of the last major glaciation of the Alaska Range. These basal loess beds contain artifacts for which radiocarbon dates and typologic correlations suggest a time range of perhaps 12,000–9000 yr ago. A long subsequent episode of cultural sterility was associated with waning loess deposition and development of a cryoturbated tundra soil above shallow permafrost. Sand deposition from local source areas predominated during the middle and late Holocene, and buried Subarctic Brown Soils indicate that a forest fringe developed on bluff-edge sand sheets along Dry Creek. The youngest archeologic component, which is associated with the deepest forest soil, indicates intermittent human occupation of the site between about 4700 and 3400 14C yr BP.

Quaternary loess deposits containing charcoal and in situ organic matter constituents from Banks Peninsula, New Zealand, were subjected to various physical and chemical treatments before radiocarbon assay. A stepwide procedure was used in which each component was radiocarbon dated before and after the treatments were applied. The criterion adopted for judging the effectiveness of a treatment is an increase in the radiocarbon age of loess layer. On this basis the oldest and therefore the most reliable date was obtained from the intra-loess charcoal and its humic acid extract. By comparison, the radiocarbon ages of organic matter constituents of the whole or partitioned loess were much younger, and are considered to represent only average ages for the particular layers concerned. No advantage was gained by dating the carbon-enriched clay-humus fraction as opposed to whole loess samples. There is an apparent conflict between our dates and current interpretation of loess stratigraphy and chronology in the South Island of New Zealand, which requires further investigation. This uncertainty apart, the upper layers of multiple loess deposits are clearly much older than earlier reports indicate, and it follows that the correlation of these deposits with late-glacial events both within and beyond New Zealand is even more tenuous and unreliable than previously thought to be the case.

Reference samples of three prominent pumice units of Glacier Peak tephra collected east of the volcano within a distance of 100 km are similar petrographically to units described by earlier workers. Glass shards isolated from these samples were analyzed by electron microprobe to determine the content of Ca, Fe, and K. Resulting data, plus those published for two other references samples, provide a basis for attributing certain outlying tephra layers from 14 locations in eastern Washington, Idaho, Wyoming, and Montana to eruptions of Glacier Peak. Ten of the samples have properties of both Glacier Peak tephra and Mount St. Helens set J tephra, but proportions of Ca:Fe:K in glass shards indicate that 9 of the 10 outlying samples came from Glacier Peak, whereas one is assigned to Mount St. Helens set J. The remaining six outlying samples, all from southeastern Washington, contain cummingtonite phenocrysts and are chemically similar to some parts of Mount St. Helens tephra sets that are older than 12,000 BP.

We have extended the fallout areas for each of two members of tephra-set W, erupted from Mount St. Helens about 1500 ad , by several hundred kilometers beyond the limits mapped in 1975. We traced one member (We) east into Idaho, and the other (Wn) northeast into British Columbia. After using stratigraphic and petrographic observations to assign more than 100 tephra samples to set W, we found 26 of these, selected for chemical analysis, to be closely similar in content of Ca, Fe, and K in glass shards. But improved homogeneity was evident when the 26 sampling localities for tephra W were segregated geographically, east vs. northeast of the volcano. When Ca:Fe:K proportions were plotted on a ternary diagram, there was no overlap of the plotting areas for these two groups of tephra W samples. Without such data, tephra layers We and Wn are currently separable only from stratigraphic and geographic information. Partial glass analysis is also an aid, along with stratigraphic position and petrographic characteristics, in distinguishing tephra W from associated tephra layers. These include tephra layers T and Yn from Mount St. Helens, as well as older tephra layers from Mount Mazama and Glacier Peak.

Reinvestigation of Quaternary sediments in West Feliciana Parish, southeastern Louisiana, and adjacent Wilkinson County, southwestern Mississippi, has resulted in revision of previous terrace stratigraphy of this portion of the Gulf Coastal Plain. Plant-macrofossil and pollen assemblages incorporated in fluviatile terrace deposits in the study area are reexamined in light of the current stratigraphic understanding. Macrofossils identified as white spruce (Picea glauca), tamarack (Larix laricina), and northern white cedar (Thuja occidentalis), recovered from these terrace deposits along with fossil remains of distinctly southern plant species, were initially interpreted as the result of dynamic intermixing of aggressive boreal species within a southern forest during the early Wisconsin (Brown, 1938). Failure to distinguish chronologically separate fossiliferous deposits resulted in the conceptual “mixing” of northern and southern plant species which came from two distinct fluviatile terrace sequences. Terrace 2 is now believed to be a fluviatile and coastwise depositional terrace of Sangamon Interglacial age; deposits of terrace 2 contain a distinctly warm-temperate plant assemblage. Fluviatile terrace 1 dates from 12,740 ± 300 to 3457 ± 366 BP and is now considered to be related to late glacial and Holocene aggradation and lateral migration of the Mississippi River (the local base level for streams in the study area); basal portions of terrace 1 contain fossils of white spruce, tamarack, and many plant species today characteristic of the cool-temperate Mixed Mesophytic Forest Association. Terrace 1 fossil deposits occur in fluviatile terraces along tributary streams of the Mississippi River at elevations 15 to 30 m above the maximum recorded historic flood stage of the Mississippi River. The plant macrofossils represent remains of species that grew at or very near the site of deposition; they were not “rafted in” by floodwaters of the Mississippi River. We present quantitative data for plant macrofossils and pollen that support our hypothesis that at least local cooling along the Blufflands of Mississippi and Louisiana promoted southward migrations of mixed mesophytic forest species and certain boreal species along this major pathway during late Wisconsin continental glaciation.

The Champlain Sea occupied the Champlain Valley from about 12,500 to 10,000 yr BP. Following an initial maximum limit of inundation, isostatic crustal rebound caused the sea's gradual regression, which is documented by the parallel alignment of tilted shoreline features at successively low elevations along a north-south profile. Two new radiocarbon shell dates, 11,665 ± 175 (QC 200), elevation 95 m, and 10,300 ± 180 (QC 199), elevation 47 m, date early and late Champlain Sea deposits, respectively. From the elevation (ASL) and invertebrate fauna of littoral deposits, three environmentally distinct phases of the sea were recognized. Early Champlain Sea Transitional phase deposits at high elevations are characterized by a mixed association of fresh and euryhaline marine ostracodes. Frigid-subfrigid climates and fluctuating salinities of this period possibly reflect intermixing of the fresh waters of Lake Vermont with incoming marine waters. Hiatella arctica phase faunas indicate similar climatic conditions but significantly higher salinities (polyhaline). Deposits from the final phase of the sea, the Mya arenaria phase, were found at low elevations just above the present level of Lake Champlain. A predominantly cold-temperate, mesohaline fauna characterizes this period. The influences of Lake Algonquin drainage, warm Gulf Stream water and perhaps the retreating Laurentide Ice Sheet are discussed as possible causes for the observed faunal and environmental changes.

Two species of marine Foraminifera have been identified in the Quaternary pluvial sequence of the Estancia Valley, central New Mexico. Besides being an unusual occurrence of marine organisms in an inland body of water remote from the marine environment, the Foraminifera yield useful paleolimnological and paleoclimatological information about two major late Wisconsin pluvial lakes. In addition, the geographic location of the Estancia Valley relative to the closest marine environment requires foraminiferal introduction by avian means. Paleontologic evidence verifies the freshwater nature of the pluvial maxima of Late Lake Estancia (18,000-10,500 BP) and Lake Willard (8500-6000 BP) but early lake conditions were considerably different. The occurrence of foraminifers Cribroelphidium selseyense and Protelphidium orbiculare, both extant species, in the sediments from the early part of each pluvial stand indicates that initial lake development was characterized by a salinity range of 25 to 35‰. At these times lake depth approximated 6 m. By comparison, the freshwater maximum of Late Lake Estancia attained a minimum depth of 90 m. The modern holarctic distribution of the Foraminifera could suggest a mean August lake temperature of 10°C during the developmental stages of Late Lake Estancia and Lake Willard. Because of the shallow-water nature of the lake basins it is likely that this temperature was reflective of mean August air temperature. Therefore, a lowering of mean August air temperature of 9.7°C from that of the present is possible.

Evidence is presented for a more extensive ice cover over South Georgia, the South Orkney Islands, the South Shetland Islands, and the tip of the Antarctic Peninsula. Ice extended across the adjacent submarine shelves to a depth of 200 m below present sea level. Troughs cut into the submarine shelves by ice streams or outlet glaciers and ice-scoured features on the shelf areas suggest that the ice caps were warm-based. The South Shetland Islands appear not to have been overrun by continental ice. Geomorphological evidence in two island groups suggests that the maximum ice cover, which was responsible for the bulk of glacial erosion, predates at least one full glaciation. Subsequently there was a marine interval and then a glaciation which overran all of the lowlying peninsulas. The Falkland Islands, only 2° of latitude north of South Georgia, were never covered by an ice cap and supported only a few slightly enlarged cirque glaciers. This suggests that the major oceanographic and atmospheric boundary represented by the Antarctic Convergence, which is presently situated between the Falkland Islands and South Georgia, has remained in a similar position throughout the glacial age. Its position is probably bathymetrically controlled.