The Mid-Atlantic Ridge between 29°N and 31°30′N in the last 10 Ma

The segmentation of the Mid-Atlantic Ridge between 29°N and 31°30′ N during the last 10 Ma was studied. Within our survey area the spreading center is segmented at a scale of 25–100 km by non-transform discontinuities and by the 70 km offset Atlantis Transform. The morphology of the spreading center differs north and south of the Atlantis Transform. The spreading axis between 30°30′N and 31°30′N consists of enéchelon volcanic ridges, located within a rift valley with a regional trend of 040°. South of the transform, the spreading center is associated with a well-defined rift valley trending 015°. Magnetic anomalies and the bathymetric traces left by non-transform discontinuities on the flanks of the Mid-Atlantic Ridge provide a record of the evolution of this slow-spreading center over the last 10 Ma. Migration of non-transform offsets was predominantly to the south, except perhaps in the last 2 Ma. The discontinuity traces and the pattern of crustal thickness variations calculated from gravity data suggest that focused mantle upwelling has been maintained for at least 10 Ma south of 30°30′ N. In contrast, north of 30°30′N, the present segmentation configuration and the mantle upwelling centers inferred from gravity data appear to have been established more recently. The orientation of the bathymetric traces suggests that the migration of non-transform offsets is not controlled by the motion of the ridge axis with respect to the mantle. The evolution of the spreading center and the pattern of segmentation is influenced by relative plate motion changes, and by local processes, perhaps related to the amount of melt delivered to spreading segments. Relative plate motion changes over the last 10 Ma in our survey area have included a decrease in spreading rate from 32 mm a⁻¹ to 24 mm a⁻¹, as well as a clockwise change in spreading direction of 13° between anomalies 5 and 4, followed by a counterclockwise change of 4° between anomaly 4 and the present. Interpretation of magnetic anomalies indicates that there are significant variations in spreading asymmetry and rate within and between segments for a given anomaly time. These differences, as well as variations in crustal thickness inferred from gravity data on the flanks of spreading segments, indicate that magmatic and tectonic activity are, in general, not coordinated between adjacent spreading segments.     

A geomagnetic field reversal time scale back to 13.0 million years before present

The ages of reversals of the Earth's magnetic field have been dated accurately back to 3.4 m.y. ago. Between this time and the age of the Cretaceous-Tertiary boundary, dates for reversals have been calculated assuming a constant rate of sea-floor spreading in the South Atlantic Ocean. The presence of thick piles of lava flows in Iceland allows us to produce independent evidence for the ages of reversals back to 13.0 m.y. B.P. Because of the extreme regularity of extrusion of these lava flows, the measurement of their magnetic polarity allows us to correlate the lava flows which were extruded during the polarity intervals associated with sea-floor spreading anomalies. The measurement of many K-Ar ages on these lava flows also allows us to compare the ages of reversals assumed by the linear interpolation between the ages of 3.4 m.y. and the Cretaceous-Tertiary boundary at 66.5 m.y., with those suggested by the radiometric dates. We find that in general the assumption of constant spreading has been a good one, but suggest a small change in the ages of reversals, amounting to an increase of about 0.27 m.y. in ages of reversals between 8.5 and 13.0 m.y. ago.     

Local axial migration and spreading rate variations,East Pacific Rise, 31°S

Mapping and analysis of marine magnetic anomalies generated during the past 3.78 m.y. at the East Pacific Rise crest near 31°S reveals a history of ongoing small-scale migration of the spreading center. The axis first became curved and then broke when the curvature became too severe, forming a 10-km offset. The offset healed rapidly and the topographic axis of the rise is now continuous and essentially linear. Sea-floor spreading has occurred asymmetrically in this area with east and west flank rates of 86 and 77 mm/yr, respectively, since 2.41 m.y. ago. Total spreading rates show an overall decline from 176 to 145 mm/yr prior to the Jaramillo event, 0.9 m.y. ago. For the last 0.7 m.y. the total spreading rate has been 162 mm/yr.     

Plate kinematics: The Americas,East Africa,and the rest of the world

Euler vectors (relative angular velocity vectors) have been determined for twelve major plates by global inversion of carefully selected sea-floor spreading rates, transform fault trends, and earthquake slip vectors. The rate information comes from marine magnetic anomalies less than 5 m.y. old, so the motions are valid for post-Miocene times. Plate motions in a mean hotspot frame of reference have also been determined, and statistical confidence limits for all the Euler vectors estimated. Among the consequences of the global motion model is the conclusion that fast-spreading ridges (separation rates greater than 3 cm/yr) have plate motion nearly perpendicular to the strike of the ridge and magnetic anomalies. Four more slowly separating ridges have an average obliquity of spreading of almost 20°.For several plate boundaries, results that differ from previous studies are in agreement with geological evidence. The North and South American plates converge slowly about a pole east of the Antilles and near the Mid-Atlantic Ridge. The results for Africa versus Somalia imply slow east-west extension on the East African Rift Valleys. The pole for motion of Eurasia relative to North America is located near Sakhalin, in accordance with evidence from Siberia and Sakhalin.     

Evidence of widespread Cretaceous remagnetisation in the Iberian Range and its relation with the rotation of Iberia

A palaeomagnetic investigation has been carried out at 13 sites of Jurassic age in the Iberian Range (northern Spain). Two components of remanent magnetisation have been found at each site. A primary high-temperature component shows an average counterclockwise rotation with respect to the north of 33±2° clockwise about a vertical axis corresponding to the absolute rotation of the Iberian plate since the Jurassic. A secondary low-temperature component shows a systematic declination difference of 16±4° with respect to the primary component. This indicates that a rotation of Iberia must have occurred between the two acquisition times. Comparison of the magnetisation directions with previous palaeomagnetic data and with sea-floor spreading data, constrains the age of the remagnetisation between 95 and 125 Ma. The remagnetisation may be associated with the extensional phases in the Iberian Basin in the Early Cretaceous (Barremian–early Albian) or Late Cretaceous (Cenomanian). A principal characteristic of the remagnetisation is its widespread character in the Iberian Range.     

New constraints on the tectonic evolution of the eastern Indian Ocean

From marine magnetic anomaly studies, a fossil spreading ridge is identified beneath the Nicobar Fan in the northwestern Wharton Basin. Several north-south-trending transform faults offset this ridge left-laterally east of the 86°E transform fault. Our findings show that this ridge, which was part of the plate boundary between the Indian and Australian plates, ceased its spreading shortly after formation of magnetic anomaly 20 (～ 45.6m.y. B.P.). Since the breakup of Australia and Antarctica probably occurred sometime between 110 and 90 m.y. B.P., we suggest that the Indian, Australian, and Antarctic plates were moving relative to one another from about 90 to 45 m.y. B.P. A triple junction would have existed in the southeastern Indian Ocean during that period of time. At anomaly 19 time (～ 45m.y. B.P.), the junction became inactive, and Australia and India became a single plate. The northwest-southeast-trending Southeast Indian Ridge was formed by connecting the India-Antarctica spreading center with the Australia-Antarctica spreading center. Its activity has continued to the present time.     

A revised identification of the oldest sea-floor spreading anomalies between Australia and Antarctica

We propose that magnetic anomalies south of Australia and along the conjugate margin of Antarctica that were originally identified as anomalies 19 to 22 may be anomalies 20 to 34. The original anomaly identification has two troublesome aspects: (1) it does not account for an “extra” anomaly between anomalies 20 and 21, and (2) it provides no explanation for the rough topography comprising the Diamantina Zone. With our revised identification there is no “extra” anomaly and the Diamantina Zone is attributed to a period of very slow spreading (～4.5mm/yr half rate) between 90 and 43 m.y. The ages bounding the interval of slow spreading (90 and 43 m.y.) correspond to times of global plate reorganizations. Our revised identification opens up the possibility that part of the magnetic quiet zone south of Australia formed during the Cretaceous long normal polarity interval. Breakup of Australia and Antarctica probably occurred sometime between 110 and 90 m.y. B.P. The “breakup unconformity” identified by Falvey in the Otway Basin may correspond to a eustastic sea level change.     

Paleomagnetism of the Xigaze ophiolite and flysch (Yarlung Zangbo suture zone, southern Tibet): latitude and direction of spreading

The Xigaze ophiolite (29.2°N, 89.5°E), which outcrops in the Yarlung Zangbo suture zone, represents the remnants of an oceanic lithosphere formed in middle Cretaceous times between the Lhasa block to the north and the Indian plate to the south. In an attempt to define the paleo-orientation and latitude of the spreading center at which it has been created, a paleomagnetic study has been done on three sites in volcanics and overlying (or interbedded) radiolarites forming the upper part of the ophiolite sequence and also on seven sites in the Xigaze Group flysch which stratigraphically overlies the volcanics to the north. In each site, hand-blocks carefully oriented both with sun and magnetic compass have been sampled. The paleomagnetism data, combined with structural data on the ophiolite dolerite intrusives, allow a partial reconstruction of South Eurasia at the time of formation of the Xigaze ophiolite. The paleolatitude of accretion and deposition of the Xigaze ophiolite and overlying sediments is found to be 10–20°N. Both ophiolite and basin have encountered a 85 ± 20° anti-clockwise rotation. The corresponding ridge was close to the southern margin of the Lhasa block and was oriented N175 ± 25°.     

Eastern Pacific spreading rate fluctuation and its relation to Pacific area volcanic episodes

Sea-floor spreading rates from four locations along the Nazca-Pacific plate boundary and one along the Juan de Fuca-Pacific plate boundary show variations over the past 2.4 m.y., with decreasing rates prior to the Jaramillo to Olduvai time interval (0.92–1.73 m.y. ago) and increasing rates since then. Other Pacific area volcanic phenomena in mid-plate and convergent-boundary settings also show minima about 1.3–1.5 m.y. ago and a maximum at present and another maximum about 5 m.y. ago: extrusion rates along the Hawaiian Ridge; volcanic episodes associated with calc-alkalic provinces of western Oregon and Central America; temporal variations in the SiO₂ content of Aleutian ash layers; and the number of deep-sea ash layers. These phenomena may fluctuate in response to changing spreading rates. During times of more rapid spreading increased shear and melting along lithospheric boundaries may occasion increased volcanic activity, whereas during times of less rapid spreading volcanic activity may be less intense.     

Unroofing history of a suture zone in the Himalaya of Pakistan by means of fission-track annealing ages

The uplift history of the Swat Valley and Hazara region of northwestern Pakistan has been established using 22 fission-track dates on apatite, zircon and sphene. A major fault, the Main Mantle Thrust (MMT) strikes east-west across the Swat Valley, separates regions of markedly differing fission-track age regimesm, and may be a suture zone separating an extinct island arc terrane on the north from the Indian plate to the south. Fission-track ages ranging from about 55 to 58 m.y. for sphene, 18 to 53 m.y. for zircon, and 9 to 17 m.y. for apatite were obtained from the region north of the MMT. To the south the fission-track age ranges are 20 to 25 m.y. for sphene, 17 to 26 m.y. for zircon, and 16 to 23 m.y. for apatite. Disparate zircon and sphene ages on each side of the MMT imply different cooling histories for each side of the fault prior to 15 m.y. Similar apatite ages on both sides of the fault imply similar cooling histories during the past 15 m.y. This may indicate that faulting ceased by 15 m.y. Mean uplift rates have been derived from the fission-track data using mainly the mineral-pair method. Uplift rates in the region north of the MMT increased from 0.07 to 0.20 mm/yr during the period 55 to 15 m.y. South of the fault, uplift rates averaged in excess of 0.70 mm/yr for the period 25 to 15 m.y. During the past 15 m.y. uplift across the MMT in the Swat Valley showsno discontinuities, ranging from 0.16 mm/yr in the south to 0.39 mm/yr in the north. A plausible interpretation for the fission-track uplift data has the MMT verging to the south with overthrusting taking place at a depth between 3.5 and 6.0 km, juxtaposing two terranes that were originally separated by a substantial, but unknown distance. In this model, regional uplift followed cessation of faulting just prior to 15 m.y.     

Cainozoic history of southern New Zealand: An accord between geological observations and plate-tectonic predictions

Sea-floor spreading data from the Southwest Pacific have recently been used to predict the Cainozoic geological history along the Indo-Australian/Pacific plate boundary. Geologic and sedimentologic data pertaining to this plate boundary where it crosses southern New Zealand, as the Alpine Fault, are summarised and discussed. It is concluded that there is a close accord between the plate-tectonic predictions and South Island Cainozoic geological history. In particular, (1) no Cainozoic plate boundary traversed the New Zealand region prior to 38 m.y. B.P. (late Eocene); (2) transcurrent movement on the Alpine Fault took place largely between ca. 30 m.y. B.P. (middle Oligocene) and ca. 10 m.y. B.P. (late Miocene); and (3) the period 10 m.y. B.P. to present corresponds to a phase of oblique compression, continental collision, and mountain building along the Alpine Fault sector of the plate boundary. There is a close correlation between the sites and histories of Cainozoic sedimentation and this tectonic timetable.     

Interception losses of rainfall from Cashew trees

The rainfall interception losses from Cashew trees were quantified, based on the records of 105 selected storms within the range 25.0 mm, occurring in a humid tropical region at Kottamparamba, India.

The storage capacity of the Cashew trees was worked out as 0.8 mm and the throughfall coefficient as 0.391. The trees under observation were 15–20 years of age with a leaf area index of 1.0–1.25.

About 31% of the storm rainfall for storms 25.0 mm was intercepted by the Cashew trees and lost to the atmosphere.

The measured interception losses from the trees were compared with the estimated interception losses using the analytical model of Gash (1979). The predicted interception losses from the Cashew trees were within ± 10% for storms with total rainfall 10.0 mm and within ± 22% for storms with a rainfall of 10.1–25.0 mm.     

Flexural uplift of a lithospheric slab near the Vema transform (Central Atlantic): Timing and mechanisms

The Vema Transverse Ridge (VTR) is a prominent, long and narrow topographic anomaly that runs for over 300 km along a sea floor spreading flow line south of the Vema transform at 11° N in the Atlantic. It rises abruptly about 140 km from the axis of the Mid-Atlantic Ridge (MAR) in 10 Myr old crust and runs continuously up to 25 Myr old crust. It reaches over 3 km above the predicted lithospheric thermal contraction level. It is absent in crust younger than 10 Myr; thus, the uplift of the VTR must have ended roughly 10 Ma. The VTR is interpreted as the exposed edge of a flexured and uplifted slab of oceanic lithosphere that was generated at an 80 km long MAR segment. Based on satellite gravimetry imagery this MAR segment was born roughly 50 Ma and increased its length at an average rate of 1.6 mm/yr. Multibeam data show that the MAR-parallel sea floor fabric south of the VTR shifts its orientation by 5° to 10° clockwise in 11–12 Myr old crust, indicating a change at that time of the orientation of the MAR axis and of the position of the Euler rotation pole. This change caused extension normal to the transform, followed between 12 and 10 Ma by flexure of the edge of the lithospheric slab, uplift of the VTR at a rate of 2 to 4 mm/yr, and exposure of a lithospheric section (Vema Lithospheric Section or VLS) at the northern edge of the slab, parallel to the Vema transform. Ages of pelagic carbonates encrusting ultramafic rocks sampled at the base of the VLS at different distances from the MAR axis suggest that the entire VTR rose vertically as a single block within the active transform offset. A 50 km long portion of the crest of the VTR rose above sea level, subsided, was truncated at sea level and covered by a carbonate platform. Subaerial and submarine erosion has gradually removed material from the top of the VTR and has modified its slopes. Spreading half rate of the crust south of the transform decreased from 17.2 mm/yr between 26 and 19 Ma to 16.9 mm/yr between 19 and 10 Ma, to 13.6 mm/yr from 10 Ma to present. The slowing down of spreading occurred close in time to the change in ridge/transform geometry, suggesting that the two events are related. A numerical model relates lithospheric flexure to extension normal to the transform, suggesting that the extent of the uplift depends on the thickness of the brittle layer, consistent with the observed greater uplift of the older lithosphere along the VTR.     

Paleomagnetic evidence of northward movement of the Pacific plate in deep-sea cores from the Central Pacific Basin

Seven deep-sea sediment cores recovered in the central equatorial Pacific collectively span a magneto- and biostratigraphically determined age interval ranging from about 0.1 to 21 m.y. B.P. Measured values of paleomagnetic inclination and their systematic variation with depth in these cores denote relative motion between the central Pacific lithosphere and the magnetic field of the earth. Assuming that the position of the earth's dipole field remained essentially parallel to the present spin axis during the interval, the data provide evidence of a marked decrease in the northward rate of plate motion from about 11 cm/yr to about 6 cm/yr at approximately 12 m.y. B.P. This date of change of motion as well as the northward direction and overall average rate of about 8 cm/yr throughout the last 21 m.y., agree reasonably well with results of other studies of the tectonic history of the Pacific plate and ridge system. More significantly, however, these preliminary results demonstrate the usefulness of the paleomagnetic record in deep-sea sediment cores spanning sufficiently long intervals of time as an aid in reconstructing plate motions.     

A new approach to the Nd residence time in the ocean: the role of atmospheric inputs

Concentrations of rare earth elements (REE) and Nd isotopic ratios were analyzed for seawater, filtered suspension and sediment trap samples collected in the tropical Atlantic Ocean (EUMELI program, EUtrophic, MEsotrophic and oLIgotrophic sites, 20°N, 18°–21°W). This is the first REE/Nd dataset on solution and different-sized particles collected at the same site. We present direct evidence of the Nd isotopic exchange between particulate lithogenic fraction and seawater without significant mass transfer. This exchange is probably one of the main factors that simultaneously constrains the Nd concentration and isotopic ratio budget. We propose a new approach to estimate the residence time of Nd in the ocean (τ_Nd) based on isotopic exchange: 200 yr < τ_Nd< 1000 yr. The exchange requires a partial dissolution of lithogenic Nd. We estimate that the fraction of soluble Nd proportion in atmospheric dust is of the order of 20% based on the isotopic ratios. We suggest that the partial dissolution of atmospheric fallout is probably one of the main REE sources of the ocean.     

Asymmetric fracture zones and sea-floor spreading

An asymmetric pattern is observed in the orientation of minor fracture zones about the axis of the Mid-Atlantic Ridge at five sites where relatively detailed studies have been made between latitudes 22°N and 51°N. The minor fracture zones intersect the axis of the Mid-Atlantic Ridge in an asymmetric V-shaped configuration. The V's point south north of the Azores triple junction (38°N latitude) and point north south of that junction.The rates and directions of sea-floor spreading are related to the asymmetric pattern of minor fracture zones at the sites studied. Half-rates of sea-floor spreading averaged between about 0 and 10 m.y. are unequal measured perpendicular to the ridge axis. The unequal half-rates of spreading are faster to the west north of the Azores triple junction and faster to the east south of that junction. The half-rates of sea-floor spreading calculated in the directions of the asymmetric minor fracture zones are equal about the ridge axis within the uncertainty of the direction determinations.A discrepancy exists between minor fracture zones that form an asymmetric V about the axis of the Mid-Atlantic Ridge, and major fracture zones that follow small circles symmetric about the ridge axis. To reconcile this discrepancy it is proposed that minor fracture zones are preferentially reoriented under the influence of a stress field related to interplate and intraplate motions. Major fracture zones remain symmetric about the Mid-Atlantic Ridge under the same stress field due to differential stability between minor and major structures in oceanic lithosphere. This interpretation is supported by the systematic variation in the orientation of minor fracture zones and the equality of sea-floor spreading half-rates observed about lithospheric plate boundaries.     

Melting of β-quartz up to 2.0 GPa and thermodynamic optimization of the silica liquidus up to 6.0 GPa

Melting relations of β-quartz were experimentally determined at 1.0 GPa (1900±20 °C), 1.5 GPa (2033±20 °C), and 2.0 GPa (2145±20 °C) using a new high-pressure assembly in a piston–cylinder apparatus and substantial differences were found with data previously reported. The new melting data of β-quartz were combined and optimized with all available thermodynamic, volumetric, and phase equilibria data for β-cristobalite, β-quartz and coesite to produce a P–T liquidus diagram for silica valid up to 6.0 GPa. Using the new optimized thermodynamic parameters, the invariant point β-cristobalite+β-quartz+liquid and β-quartz+coesite+liquid were determined to lie at 1687±17 °C and 0.457 GPa, and 2425±25 °C and 5.00 GPa, respectively.     

Granoblastic olivine aggregates in magnesian chondrules: Planetesimal fragments or thermally annealed solar nebula condensates?

Granoblastic olivine aggregates (GOA) have been discovered in some Type I magnesian chondrules within carbonaceous chondrites by Libourel and Krot [Libourel, G., Krot, A.N., 2007. Evidence for the presence of planetesimal material among the precursors of magnesian chondrules of nebular origin. Earth Planet. Sci. Lett. 254, 1–8], who proposed an origin from pre-existing planetesimals. Amoeboid olivine aggregates (AOA), generally considered as aggregates of solar nebula condensates and found within similar carbonaceous chondrites, display similar equilibrium texture, though on a finer scale. For these reasons, we conducted experiments to determine if annealing of olivine required time scales appropriate to planetesimal or nebular heating. Pressed < 43 µm and < 63 µm San Carlos olivine powder (Fo_88.4) was isothermally heated at temperatures ranging from 1350–1550 °C for 1–100 h. The 100 h runs yield olivine aggregates with well-developed granoblastic texture at all temperatures, manifest as a network of randomly-oriented and sutured olivine grains with 120° triple junctions. Individual olivine grains are 4–6 sided and polygonal by 1450 °C and equigranular texture is developed at high temperature (1500–1550 °C). Melting of olivine commences at 1450–1500 °C and aids in ‘ripening’ and suturing (grain coarsening and grain boundary migration). Textural equilibrium is clearly met at 1550 °C. A planetesimal origin cannot be ruled out; however, the experimental evidence reveals that granoblastic texture can be reproduced in an interval not inconsistent with heating times for nebular objects. GOA may have experienced higher degrees of thermal processing than the finer-grained AOA. If the precursors were the same, grain coarsening would have to be accompanied by modification to bulk and isotopic compositions. However, the precursors could have been olivine condensates formed later than AOA. Annealing may have been a widespread process operating in the primordial solar nebula responsible for thermal processing and formation of GOA prior to their incorporation into chondrules.     

Palaeomagnetic data from Ediacaran (Vendian) sediments of the Arkhangelsk region, NW Russia: An alternative apparent polar wander path of Baltica for the Late Proterozoic–Early Palaeozoic

The study area is situated along the Zolotica river in NW Russia, located within the Kola–Dvyna Rift System in the Baltic Shield that developed during Meso and Neoproterozoic times. A 9-m thick section made up of shallow marine sediments of Upper Ediacaran age was sampled in this locality. Two volcaniclastic levels from the middle part of the section yielded an age of 556 Ma. (U/Pb SHRIMP-II on zircons). Two magnetic components were successfully isolated, component A (Decl = 157.1, Incl = 68.0, ₉₅ = 1.9°, N = 575 in situ) carried by magnetite and component B (Decl = 120.3, Incl = − 31.7, ₉₅ = 3.9°, N = 57, bedding corrected), carried by haematite. While component A is thought to represent a younger overprint direction, the in situ direction for component B on the other hand, is dissimilar to any expected younger direction and is considered to be primary magnetisation in origin, acquired during or soon after deposition of the sediments in the Late Ediacaran. The corresponding palaeomagnetic pole for component A in situ is located at Lon = 55.4°E, Lat = 31°N, A₉₅ = 2.7° and for component B at Lon = 110°E, Lat = 28.3°S, A₉₅ = 3.8°, N = 57. Combined with other palaeomagnetic poles of the same tectonostratigraphic unit an alternative apparent polar wander path for the Late Proterozoic–Early Palaeozoic of Baltica is proposed. Such an alternative path shows that after the mid Cryogenian (750 Ma), the poles that were situated over South Africa (p.d.c.) moved to the east until they reached Australia during the Late Ediacaran (555 Ma) where they remained approximately stationary until the beginning of the Cambrian (545 Ma). Finally, they moved to the northwest until they reached the Arabian Peninsula in the Early Ordovician. Palaeolatitudes indicate that Baltica situated near the equator from the Cryogenian through to the Ediacaran moving gradually to the south at c. 1 cm/yr. During the Late Early Ediacaran, the plate suddenly began to drift northward at c. 8 cm/yr and in the boundary with the Cambrian it was positioned in low to intermediate latitudes. Finally, Baltica began to move back to the south at c. 13 cm/yr until in the Early Ordovician, reaching intermediate to high southern latitudes.     

High precision glacial–interglacial benthic foraminiferal Sr/Ca records from the eastern equatorial Atlantic Ocean and Caribbean Sea

Glacial–interglacial variation in the marine Sr/Ca ratio has important implications for coral Sr thermometry [J.W. Beck et al., Science 257 (1992) 644–647]. A possible variation of 1–3% was proposed based on ocean models [H.M. Stoll and D.P. Schrag, Geochim. Cosmochim. Acta 62 (1998) 1107–1118]. Subsequently, studies have used fossil foraminifera to test this prediction [P.A. Martin et al., Geochem. Geophys. Geosyst. 1 (1999); H.M. Stoll et al., Geochim. Cosmochim. Acta 63 (1999) 3535–3547; H. Elderfield et al., Geochem. Geophys. Geosyst. 1 (2000)]. But whether some component of foraminiferal Sr/Ca variation can be uniquely ascribed to seawater Sr variation is still not clear. To address this question, we developed cleaning and analysis techniques and measured Sr/Ca ratios on individual shells of the modern benthic foraminifer Cibicidoides wuellerstorfi. We showed that different size shells have different Sr/Ca ratios; however, samples with shell sizes of 355–500 μm appear to have normally distributed Sr/Ca ratios (1σ=1.8%). For multi-shell measurements (with estimated errors of 0.12–0.39%), the ratio varied by as much as 7.2±0.5% during the last glaciation for two Caribbean records at the same site and by 3.7±0.5% over the past 40,000 yr for one record from the Sierra Leone Rise in the eastern equatorial Atlantic. The two Caribbean records are very similar indicating that the behavior of shell Sr uptake was identical locally and that the shell Sr/Ca ratio faithfully reflects the local environment. The Atlantic record differs from the Caribbean records by as much as several percent. Thus, the foraminiferal Sr/Ca changes cannot be solely due to changes in seawater Sr/Ca unless the glacial deep ocean had spatial variation in Sr/Ca well in excess of the modern ocean. Certain similarities between the three records do exist. Notably, the rate of change of Sr/Ca is similar between 9 and 0 ka (−0.25%/kyr) and between 25 and 16 ka (+0.16%/kyr). This suggests that during these intervals, benthic foraminiferal Sr/Ca was affected by similar large-scale variables. One of these variables may be the average marine Sr/Ca ratio; however, comparison with model predictions [H.M. Stoll and D.P. Schrag, Geochim. Cosmochim. Acta 62 (1998) 1107–1118] suggests other factors must also be considered. The discrepancies between the two sites may be related to the different water mass histories for the Caribbean and eastern Atlantic. Our results suggest that variation of the seawater Sr budget only partially contributed to C. wuellerstorfi Sr/Ca records, while other significant factors still need to be quantified. At present we cannot confidently determine past seawater Sr/Ca variation from our foraminiferal records.