Spectral analysis of deep-sea sediments indicates that the fluctuations in compositional parameters are not random fluctuations with time. Spectra show significant peaks representing periodicities in the data of 380, 1300, and 2600 years. Two of these periods are similar to periods reported in 14C fluctuations. Analysis of a paleotemperature curve from the North Atlantic shows that the characteristics of the fluctuations within interglacial and glacial stages of the climate are similar, and that the spectrum has a significant peak at 2600 years.

A simplified and more accurate version of the quantitative paleoenvironmental method proposed by Imbrie and Kipp (1971) is described, which is based on untransformed rather than transformed species per cent data. The method yields faunal indices (Ts, Tw, S), useful both as objective measures of paleontological properties and as estimates of Pleistocene sea-surface summer and winter temperatures (Ts, Tw) and salinity (S). Similarly, the oxygen-isotope method yields objective measurements of δ O18, useful stratigraphically and as indications of past changes in isotopic water composition and temperature.

Laboratory errors of the two methods have about the same magnitude relative to ranges observed in V12-122. Accuracy of faunal indices as estimates of oceanic conditions is evaluated by study of modern oceanographic data and sea-bed samples. Under favorable conditions, accuracy is apparently limited primarily by the degree of ecological control exercised by the estimated parameter. Accuracy of the isotopic paleotemperature estimates is limited primarily by uncertainty as to the magnitude of the water-correction term in the isotope equation, a value which combines global ice-volume and local evaporation-precipitation effects.

Curves of δ O13 and S in V12-122 record all or part of seven major climatic cycles, and display a fundamental periodicity of about 85,000 years. Ts and Tw curves show small but significant differences: two phase shifts in estimated temperature minima, and a long-term increasing trend.

Amplitude-frequency histograms of Tw and δ O18 indicate that only two percent of the time during the past 450,000 years have Caribbean temperatures been as warm and isotopic ratios as low as they are today.

A comparison of the magnitude of δ O18 change (2.2%.) during the shift from late-glacial to post-glacial times, with that of the faunally estimated change in average temperature (2.2°C), provides a basis for estimating the associated change in isotopic water composition (1.8%.) by back-calculation in the isotope equation. At least 0.4%. of this change in A is attributed to an evaporation-precipitation effect, and the balance (≤1.4%.) to an icevolume effect.

Isotopic and faunal methods monitor different responses to global climatic change. Used in conjunction, they provide deeper insights into the past than either could achieve alone.

Core Vema 28-238 preserves an excellent oxygen isotope and magnetic stratigraphy and is shown to contain undisturbed sediments deposited continuously through the past 870,000 yr. Detailed correlation with sequences described by Emiliani in the Caribbean and Atlantic Ocean is demonstrated. The boundaries of 22 stages representing alternating times of high and low Northern Hemisphere ice volume are recognized and dated. The record is interpreted in terms of Northern Hemisphere ice accumulation, and is used to estimate the range of temperature variation in the Caribbean.

Factor analysis of dissolution-resistant species of planktonic foraminifera in the >250 μm fraction of seabed and core samples from the tropical southeast Pacific yields five mappable varimax assemblages. The distribution of these assemblages in core tops is more related to the subsurface thermal structure than to the surface temperature. A transfer function relating the assemblages in core tops to the temperature at 200 m is applied to five cores. Cycles of estimated temperature in three cores are approximately in phase with climatic cycles in Atlantic cores spanning the past 200,000 yr, and display amplitudes on the order of 5°C. Two zones of increased calcium carbonate dissolution occur in these three cores. These zones occur in the transition from interglacial to glacial intervals. Temperature variations inferred are ascribed to fluctuations of the main thermocline, and the inference is made that during cold intervals the thermocline was displaced upward as the central water mass was areally diminished. A model of circulation dynamics suggests that such areal diminution would be associated with an intensified circulation. No significant fluctuations in the estimated 200 m temperature are inferred for a core in the Galápagos area during the last glacial, but a pronounced peak of high-latitude species Globorotalia inflata does occur. This peak is judged to reflect the removal of a low-oxygen ecological barrier.

A quantitative study of the distribution of Radiolaria in surface sediments of the North Pacific has demonstrated the feasibility of utilizing complete radiolarian thanatocoenoses as indicators of past oceanographic conditions.

In samples from 36 core tops from the region 33° N-54° N, 148° E-140° W, 87–96% of all radiolarians encountered could be assigned to one of 120 taxa recognized. Q-mode factor analysis of distributional data for the 57 most abundant species yields four independent assemblages. Three of these reflect near-surface oceanographic conditions, and the fourth responds to bottom conditions. Regression-developed transfer functions describe the relationship between the assemblages and oceanographic parameters.

Objective quantitative estimates of paleo-oceanographic conditions in the North Pacific can be made by analyses of radiolarian assemblages. With appropriate computation, transfer functions developed in a study of surface sediments can be used to estimate oceanographic conditions in cores containing late Pleistocene radiolarian faunas. Analysis of core V21-173 indicates that conditions as warm as the Holocene were rare during the past 800,000 yr, and that the region experienced marked near-surface temperature drops correlative with Caribbean and continental records for the past 250,000 yr. A major world-wide warm event at about 400,000 yr is also indicated.

The distributions of the radiolarian assemblages in the Northeastern Pacific Ocean were determined and correlated with the average summer temperature of the near surface waters of this region. These assemblages were compared with those in three sediment cores taken beneath the Transition Zone waters. This comparison indicates that the assemblage off Oregon at the last maximum cold interval (24,000 yr B.P.) was like that now found off southern Alaska. The correlation of the radiolarian assemblages with temperature gives an estimate of 11°C for the average summer temperature at that time. This is approximately 4°C cooler than present day conditions in the area. Superimposed on the general warming trend that began 24,000 y.a., there are minor oscillations in the assemblages which correspond to estimates of temperature change of about 2°C in the Pleistocene and about 1°C in the Holocene. In the Holocene, these minor warm intervals appear to be approximately synchronous with advances in mountain glaciers.

Quantitative paleo-environmental analyses of planktonic foraminifera in 182 samples covering the past 130,000 years in North Atlantic deep-sea core V23-82 yield time series interpreted in terms of changing surface-water conditions. An absolute chronology is estimated by linear interpolation between levels dated by 14C or by stratigraphic correlation with other radiometrically dated climatic records. Significant events include: (1) rapid warming at 127,000 YBP, marking the start of the penultimate North Atlantic and European interglacial; (2) 124,000 YBP temperature maximum (Eemian); (3) 109,000 YBP cooling, correlated with the beginning of the last European glaciation (Würm), and representing a temporary cooling of the North Atlantic; (4) severe cooling 73,000 YBP, marking the start of the last full glacial regime in the North Atlantic; (5) short warm intervals within the last glacial regime dated at 59,000 YBP, 48,000 YBP, and 31,000 YBP; (6) rapid termination of the last glacial interval at 11,000 YBP; (7) a 6000 YBP hypsi-saline, followed by lowering salinity values presumably associated with decreasing flux of Gulf Stream waters over the core site.

A 9300 yr-old zone of disseminated volcanic ash in North Atlantic sediments between 45° N and 65° N provides a time-synchronous reference layer against which we have compared the stratigraphic level of deglacial warming of ocean surface waters. In the Atlantic north of 45° N the most prominent feature of this warming is the replacement of low-carbonate glacial marine sediment containing only a single species of polar Foraminifera by calcareous oozes containing a diverse temperate fauna and flora. The local terminations of glacial conditions marked by this change are not synchronous at these latitudes, but range from 13,500 yr B.P. or older in the southeast near Great Britain to 6,500 yr B.P. or younger in the northwest near Greenland. Regionally, these local warmings trace the progressive westward and northward retreat of polar water from the North Atlantic. Since the withdrawal of polar water from the North Atlantic coincides with the northward shrinkage of temperate-latitude continental ice sheets, it is the best oceanic analog to continental deglaciation.

Faunal, floral, lithologic, and isotopic parameters showing evidence for a sudden deglacial warming may not be time-synchronous; those parameters are subject to a range of environmental controls and may thus respond differently to the causal mechanism for global warming.

Today, below 2500 m, benthonic foraminiferal faunas in the North Atlantic are dominated by a few species. Faunal composition changes slowly with increasing depth and decreasing temperature. Surface sediment and down-core counts of benthonic foraminifera reported by Phleger, Parker, and Peirson (1953) in the reports of the Swedish Deep-Sea Expedition have been supplemented by additional bottom sediment and piston core samples. Present-day benthonic foraminiferal assemblages from the deeper portions of the North Atlantic appear to be controlled more by the distribution of bottom water types than by bathymetry. In most piston cores, the assemblages vary greatly during the last 150,000 yr, suggesting depression and elevation of faunas at the core site through a depth range of several hundred meters. This would indicate that bottom water characteristics have shifted back and forth in this interval of time and, therefore, that bottom circulation partakes in the well-documented shifts recorded for surface waters of the North Atlantic. It appears that dense water, similar to present-day North Atlantic Deep Water, was produced over a wide area north of 45° N during cooler intervals and that it spread widely at depth.

By He-U dating of corals from elevated Pleistocene reef tracts on Barbados, we have extended back to the Middle Pleistocene the high sea stand chronology previously deduced by Th230-U dating. Six samples from the first major reef tract complex older than the 200,000-yr complex gave ages of 350,000 ± 25,000 yr B.P. Two corals from the crest of Second High Cliff, an unusually large escarpment occurring approximately midway in the terrace sequence, gave concordant ages of 480,000 and 500,000 yr. Unrecrystallized corals from older reefs gave ages ranging back to 650,000 yr.

The results date episodes of high sea stands at 350,000 and 500,000 yr B.P.

Calculations confirm that the uplift of Barbados during the past 130,000 yr has been at nearly constant relative rates in the Clermont and Christ Church standard traverses, and that sea levels responsible for Barbados terraces I (82,000 yr B.P.) and II (105,000 yr B.P.) attained approximately the same level which was 20–25 m below the level represented by Barbados III (125,000 yr B.P.).

Preference for the correlation of Barbados III with the prominent first interglacial 18O peak in stage 5 is stated once again. Further, correlation with Eemian and Pangaion of the European pollen record is suggested.