This paper attempts to relate current knowledge of sea-level history in Beringia to the Broecker-van Donk “Termination” concept of climatic and sea-level history. The Einahnuhtan transgression is thought to represent Termination III, which according to Broecker and van Donk, took place about 225,000 y.a. The Kotzebuan transgression is thought to represent a positive fluctuation that modulated the generally falling sea level during the ensuing 100,000 yr. Sea level probably fell to about −135 m in the Bering Sea area during the maximum phase of the penultimate glaciation. The two Pelukian shorelines probably represent Termination II (about 125,000 yr BP in the Broecker-van Donk chronology) and one of the two positive fluctuations that modulated the generally falling sea level of early Wisconsinan time, about 105,000 and 80,000 y.a. according to Broecker and van Donk. Another positive modulation brought sea level to at least −20 m, about 30,000 y.a. Sea level evidently fell to between −90 and −100 m during the late Wisconsinan regression, but a substantial part of the outer Bering shelf remained submerged. Submerged shoreline features at −38m, −30 m, −24 to −20 m, and −12 to −10 m represent stillstands or slight regressions that modulated Termination I, the late Wisconsinan, and early Holocene recovery of sea level.

Bed I in Olduvai Gorge spans the interval from about 1.7 to 2.0 m.y.a., and all evidences of hominid activity at Olduvai are within the latter half of this period. Bed I was deposited in a closed basin approximately 25 km in average diameter, and it can be subdivided into five lithologically different but partly time-equivalent rock assemblages, or lithologic facies, each of which was deposited in the same geographic environment or closely related series of environments. These lithofacies comprise lake deposits, lake-margin deposits, alluvial-fan deposits, alluvial-plain deposits, and lava flows.

Alluvial-plain deposits form the lowermost part of Bed I, and they presumably interfinger northeastward with lake deposits in an area not now exposed. Discharge of the lavas diplaced the lake westward to a position it occupied through the latter half of the deposition of Bed I. The lake fluctuated greatly in salinity, level, and extent, and at times of low level it was highly saline and generally ranged between 7 and 10 km in average diameter. At times of high level it was as much as 15 km in diameter and was relatively fresh, at least along its southeastern margin. The lake-margin deposits accumulated on the zone of relatively flat terrain that was flooded at times of high level. Lake-margin deposits interfinger eastward with deposits of an alluvial fan of pyroclastic materials produced in an explosive phase of the volcano Olmoti. The climate at the time of Bed I was relatively dry although somewhat wetter than the present climate in the same region today.

Radiocarbon dates on molluses in marine facies associated with glacial deposits in northern Cumberland Peninsula indicate both main fiord (Laurentide) ice and local glaciers remained at their late Wisconsin maxima until ca. 8000 BP. Essentially continuous deglaciation followed; local corrie glaciers melted out by 7100 BP and by 5500 BP fiord glaciers had receded behind the present margin of the Penny Ice Cap. The Hypsithermal warm interval probably lasted from ca. 8000 to 5000 BP. Lichenometry and radiocarbon dates on peat and buried organic horizons delimit a detailed Neoglacial chronology. Of 46 outlet and corrie glaciers investigated, the oldest Neoglacial moraines are dated lichenometrically at 3200 ± 600 BP. Subsequent advances terminated immediately prior to ca. 1650, 780, 350, and 65 yr BP, the most recent of which marked the most extensive ice coverage during the Neoglacial. The highest occurrence of lateral moraines from late Wisconsin advances of local and Laurentide ice suggest that at the late Wisconsin glacial maximum, depression of snowline varied from 450 m below present at the coast to 350 m below present level in the vicinity of the Penny Ice Cap. Moraines, surrounded by glacial ice and lying above the present steady-state ELA, suggest that during the Hypsithermal snowline was up to ca. 200 m above its present elevation. A radiometrically controlled reconstruction of relative summer paleotemperatures for the postglacial derived independently of lichenometry agrees well with the lichenometric age dating of moraines. The data suggest that between ca. 1650 and 900 BP climatic conditions were unfavorable for glacier growth, whereas the period ca. 800-65 yr BP was one of general glacial activity. During the last decade permanent snow cover has been increasing in the area. Previously reported data on climatic trends in the Canadian Arctic based on palynological analyses are similar to the chronology reported here.

Moraines and rock glaciers in Front Range cirques record at least four, and possibly five, intervals of Holocene glacier expansion. The earliest and most extensive was the Satanta Peak advance, which deposited multiple terminal moraines near present timberline shortly before 9915 ± 165 BP. By 9200 ± 135 BP, timberline had risen to at least its modern elevation; by 8460 ± 140 BP, patterned ground on Satanta Peak moraines had become inactive. Although a minor ice advance may have occurred just prior to 7900 ± 130 BP, there is no evidence that glaciers or perennial snowbanks survived in the Front Range during the “Altithermal” maximum (ca. 7500–6000 BP), or during a subsequent interval of alpine soil formation (ca. 6000–5000 BP).

Glaciers were larger during the Triple Lakes advance (5000–3000 BP) than at any other time during Neoglaciation. Minimum ages of 4485 ± 100 BP, 3865 ± 100 BP, and ca. 3150 BP apply to a threefold sequence of Triple Lakes deposits in Arapaho Cirque. After an important interval of soil formation and cavernous weathering, glaciers and rock glaciers of the Audubon advance (1850–950 BP) reoccupied many cirques, and perennial snowbanks blanketed much of the area above present timberline; although the general Audubon snow cover had begun to melt from valley floors by 1505 ± 95 BP, expanded snowbanks lingered on tundra ridge crests until 1050–1150 BP, and glaciers persisted is sheltered cirques until at least 955 ± 95 BP. Following a minor interval of ice retreat, glaciers of the Arapaho Peak advance (300–100 BP) deposited multiple moraines in favorably oriented cirques.

Interpretation of Holocene glacial deposits in the Southern and Central Rocky Mountains has been hampered by (1) a heavy reliance upon relative-dating criteria, many of which are influenced by factors other than age; (2) the assumption that glacial advances in high-altitude cirques can be correlated directly with alluvial deposition in far-distant lowlands; and (3) the assumption that glacial advances have necessarily been synchronous throughout the Rocky Mountain region and the world. Although Holocene glacier fluctuations in the Front Range are believed to reflect changes in regional climate, the Front Range chronology does not have particularly close analogs in other parts of North America. Better-dated local sequences are needed before the hypothesis of global synchroneity can be adequately evaluated; until synchroneity has been proven, long-distance correlations and worldwide cycles of recurring glaciation will remain unconvincing.

Previous attention has been called to the morphology of the glaciated Appalachian Plateau, including periglacial phenomena (Coates, 1970; Conners, 1969). This paper deals with an unusually well-developed hierarchy of small landforms in the Great Bend area of the Susquehanna River. Essential properties of these features include: (1) concentration in N-S valleys, (2) till composition, (3) concavo-convex form, sometimes ending in a hill on the valley floor, (4) alternation with steep, truncated bedrock spurs, (5) a col in the interfluve at their head. These characteristics could be explained by the following sequence of events. A prior fluvial landscape was eroded with tributary streams forming lateral valleys that head in cols along the divides. The main stream flowed south between interlocking spurs. Ice then widened the valley, leaving truncated spurs and a straightened stream. Deposition of locally derived till filled tributary valleys, similar to till shadows (Coates, 1966). In periglacial conditions, while nivation was widening the tributary valley heads, solifluction in the unstable till was forming the concavo-convexities that alternate with the truncated bedrock spurs. The hierarchy of forms range from minor convexities to small hills on the main valley floor. Thus, the features are primarily of periglacial origin, but owe their development and position to prior subaerial and glacial events. The cycle may have occurred during more than a single glacial episode. The recession of bedrock spurs and the valley floor convexity of the till features have caused a reversal in stream sinuosity of the main valley.

Recent pollen and macrofossil data from the Southeast is consistent with a displacement of boreal forest species by over 1000 km during full-glacial time. Data from west of the Appalachians suggests a displacement of some 600 km. Thus boreal forests were developed in a broad area south of the ice margin. Few deciduous forest elements persisted in that region. The displacement appears to have been azonal. There is good evidence to suggest a significant mid-Wisconsin interstadial (23,000-36,000 BP) characterized by a more temperate biota.

Remains of dead bristlecone pine (Pinus longaeva Bailey) are found at altitudes up to 150 m above present treeline in the White Mountains. Standing snags and remnants in two study areas were mapped and sampled for dating by tree-ring and radiocarbon methods. The oldest remnants represent trees established more than 7400 y.a. Experimental and empirical evidence indicates that the position of the treeline is closely related to warm-season temperatures, but that precipitation may also be important in at least one of the areas. The upper treeline was at high levels in both areas until after about 2200 B.C., indicating warm-season temperatures about 3.5°F higher than those of the past few hundred years. However, the record is incomplete, relative warmth may have been maintained until at least 1500 B.C. Cooler and wetter conditions are indicated for the period 1500 B.C.-500 B.C., followed by a period of cool but drier climate. A major treeline decline occurred between about A.D. 1100 and A.D. 1500, probably reflecting onset of cold and dry conditions. High reproduction rates and establishment of scattered seedlings at high altitudes within the past 100 yr represents an incipient treeline advance, which reflected a general climatic warming beginning in the mid-19th century that has lasted until recent decades in the western United States. This evidence for climatic variation is broadly consistent with the record of Neoglacial advances in the North American Cordillera, and supports Antevs' concept of a warm “altithermal age” in the Great Basin.

Late-glacial and postglacial pollen stratigraphy and radiocarbon chronology of a marine core from the continental slope and a core from the western Olympic Peninsula, ca. 110 km apart, are compared. Divisible into four pollen assemblage zones (L, P-1, P-2, and P-3), the cores exhibit a succession of correlative zonal prominences: grass-sedge (L), pine (P-1), alder (P-1-P-2 boundary), and hemlock (P-3). Volcanic ash of Mt. Mazama provenance is also correlative in zone P-2. Quantitative relationships of the pollen in the cores (relative and absolute numbers and pollen influx) are dissimilar, however, and are attributed to the influence of the Columbia River pollen load reaching the locale of the continental slope core compared with the local pollen rain influencing the Olympic Peninsula core site.

The amplitude of glacial/interglacial temperature changes in the Caribbean Sea, Gulf of Mexico, and Equatorial Atlantic, and a generalized faunal paleotemperature curve for the Caribbean Sea for the last 125,000 yr have been determined by using a quantitative micropaleontological model. This model is based on a direct comparison of Pleistocene foraminiferal assemblages with Recent ones whose geographic distribution have been correlated with modern ocean surface temperatures. The results of such an analysis indicate a glacial/interglacial surface temperature variation of 5°C for the Caribbean Sen, 3–4°C for the Gulf of Mexico, and 5–6°C for equatorial waters off the west coast of Africa. Generalized paleotemperature curves derived from faunal and isotope data for the Caribbean indicate nearly identical temperature oscillations during the last 125,000 yr.