Search

International audience; To examine the petrogenesis and sources of basalts from the Kolbeinsey Ridge, one of the shallowest locations along the global ridge system, we present new measurements of Nd, Sr, Hf, and Pb isotopes and U-series disequilibria on 32 axial basalts. Young Kolbeinsey basalts (full-spreading rate = 1.8 cm/yr; 67 degrees 05'-70 degrees 26'N) display ((230)Th/(238)U) < 1 and ((230)Th/(238)U) > 1 with ((230)Th/(238)U) from 0.95 to 1.30 and have low U (11.3-65.6 ppb) and Th (33.0 ppb-2.40 ppm) concentrations. Except for characteristic isotopic enrichment near the Jan Mayen region, the otherwise depleted Kolbeinsey basalts (e. g. (87)Sr/(86)Sr = 0.70272-0.70301, epsilon(Nd) = 8.4-10.5, epsilon(Hf) = 15.4-19.6 (La/Yb)(N) = 0.28-0.84) encompass a narrow range of ((230)Th/(232)Th) (1.20-1.32) over a large range in ((238)U/(232)Th) (0.94-1.32), producing a horizontal array on a ((230)Th/(232)Th) vs. ((238)U/(232)Th) diagram and a large variation in ((230)Th/(238)U). However, the ((230)Th/(238)U) of the Kolbeinsey Ridge basalts (0.96-1.30) are inversely correlated with ((234)U/(238)U) (1.001-1.031). Samples with low ((230)Th/(238)U) and elevated ((234)U/(238)U) reflect alteration by seawater or seawater-derived materials. The unaltered Kolbeinsey lavas with equilibrium (234)U/(238)U have high ((230)Th/(238)U) values (>= 1.2), which are consistent with melting in the presence of garnet. This is in keeping with the thick crust and anomalously shallow axial depth for the Kolbeinsey Ridge, which is thought to be the product of large degrees of melting in a long melt column. A time-dependent, dynamic melting scenario involving a long, slowly upwelling melting column that initiates well within the garnet peridotite stability zone can, in general, reproduce the ((230)Th/(238)U) and ((231)Pa/(235)U) ratios in uncontaminated Kolbeinsey lavas, but low ((231)Pa/(235)U) ratios in Eggvin Bank samples suggest eclogite involvement in the source for that ridge segment

Major and selected trace element abundances of MORB dredged from the moderately slow spreading southern MAR (40–55°S), in the vicinity of the Shona and Discovery mantle plumes, are used to constrain their melting conditions. All samples plot to high Fe8 relative to the Na8–Fe8 global trend, the high Fe8 of N-MORB in particular being anomalous. Shona-influenced MORB plot at lower Na8 and towards higher Fe8 values along the global trend than associated N-MORB, consistent with deeper initiation of melting. Discovery-influenced MORB extend to higher Na8 and lower Fe8 compositions than associated N-MORB. Anomalous Na8 and Fe8, and Si8, Ca8 and Al8, of most Discovery-influenced MORB are interpreted to reflect increased modal clinopyroxene in their source regions. Calculated Fmean range from 15–17% for N-MORB, 16–19% for Shona-influenced MORB, and ~11–18.5% for selected, least-anomalous Discovery(+LOMU)-influenced MORB. Calculated Pinitial are fairly constant for N-MORB (~18±2 kbar), slightly greater for Shona-influenced MORB (~20±3 kbar), whereas for least anomalous Discovery(+LOMU)-influenced MORB calculated Pinitial group around ~16±2 and ~22 kbar. Calculated crustal thicknesses range mostly between 6–9 km, but for Shona- and a single Discovery(+LOMU)-influenced MORB location the range is 8–12 km. Anomalous compositions of some Discovery-influenced MORB are interpreted to reflect variable melting of pyroxenite veins, initially formed as small-degree (2–3%) melts of garnet lherzolite within the upwelling off-axis Discovery mantle plume, and subsequently entrained in the ambient spinel lherzolite beneath the ridge axis. Because of lower solidus temperatures relative to ambient spinel lherzolite, initial melting preferentially consumed the low abundance (2.5–3.5%) pyroxenite veins. Variable mixing between vein- and host mantle-derived melts led to the range of Discovery(+LOMU)-influenced MORB compositions. In the vicinity of transform off-sets, melting is restricted to a narrow pressure interval, and melts are ineffectively homogenized and focused towards segment centres. Beneath slower spreading ridges, such as the southern MAR, reduced heat flux further retards melt homogenization and focusing, resulting in high proportions (≤40%) of pyroxenite-derived melt in final magmas erupted up to ~30–40 km from ridge segment ends, i.e. retaining a strong pyroxenitic source signature. Electronic supplementary material to this paper can be obtained by using the Springer LINK server located at http://dx.doi.org/10.1007/s00410-002-0376-3.

We present new 238 U– 230 Th– 226 Ra– 210 Pb, 235 U– 231 Pa, and Nd, Sr, Hf, and Pb isotope data for the slow- to ultraslow-spreading Mohns, Knipovich, and Gakkel Ridges. Combined with previous work, our data from the Arctic Ridges cover the full range of axial depths from the deep northernmost Gakkel Ridge shallowing upwards to the Knipovich, Mohns, and Kolbeinsey Ridges north of Iceland. Age-constrained samples from the Mohns and Knipovich Ridges have ( 230 Th/ 238 U) activity ratios ranging from 1.165 to 1.30 and 1.101 to 1.225, respectively. The high 230 Th excesses of Kolbeinsey, Mohns, and Knipovich mid-ocean ridge basalts (MORB) are erupted from ridges producing relatively thin (Mohns, Knipovich) to thick (Kolbeinsey) oceanic crust with evidence for sources ranging from mostly peridotite (Kolbeinsey) to eclogite-rich mantle (Mohns, Knipovich). Age-constrained lavas from 85°E on the Gakkel Ridge, on the other hand, overlie little to no crust and range from small (∼5%) 230 Th excesses to small 238 U excesses (∼5%). The strong negative correlation between ( 230 Th/ 238 U) values vs. axial ridge depth among Arctic ridge basalts is controlled not only by solidus depth influence on 238 U– 230 Th disequilibria, but also by variations in mantle source lithology and depth to the base of the lithosphere, which is expected to vary at ultra-slow spreading ridges. Small 231 Pa excesses (65% excess) in age-constrained basalts support the presence of eclogite in the mantle source for this region. Conversely, the ultraslow-spreading Gakkel Ridge basalts are homogeneous, with Sr, Nd, and Hf radiogenic isotopic signatures indicative of a long time-averaged depleted mantle source. The Gakkel samples have minimum ( 226 Ra/ 230 Th) ratios ranging from 3.07 to 3.65 ± 3%, which lie along and extend the global negative correlation between 226 Ra and 230 Th excesses observed in MORB. The new 230 Th– 226 Ra data support a model for global MORB production in which deep melts record interaction with shallower materials. This scenario requires either mixing with shallow-derived melts, or melt-rock reaction with shallower rocks in the lithosphere or crust.

He-Pb-Nd-Sr isotope systematics in basalts from the Mid-Atlantic Ridge (MAR) from 65°N to 78°N are reported for mapping the zone of influence of the Jan Mayen and Iceland mantle plumes in the Arctic. The geographical variation and the two distinct trends observed in the He-Pb-Nd-Sr isotope space clearly indicate that the boundary between the zone of influence of the “low 3He/4He” Jan Mayen plume and the “high 3He/4He” Iceland mantle plume is in the vicinity of the Spar FZ. Modeling indicates that the dispersion of the Jan Mayen plume is not purely radial, but extends preferentially northward, probably because of decreasing spreading rate and the cascading of the buoyant mantle plume across the Jan Mayen fracture zone (FZ) due to the large denivellation (change of level) of the base of the lithosphere caused by the large age offset (∼20 Myr). The incompatible parent and daughter (PD) element concentrations and their ratios for the basalt population from the Jan Mayen plume are highly coherent with the Pb-Nd-Sr isotope ratios and show essentially the same geographical pattern in spite of large variations in the mean degree of fusion and extent of fractional crystallization. In contrast, over the southern Kolbeinsey Ridge, unusual decouplings are observed between He-Pb-Nd-Sr isotopic systematics and incompatible element variations, as well as inferred melting conditions. These decouplings are best explained by a modified version of the dispersion model of the Iceland plume by Mertz et al. [1991] which was based on more limited isotopic data, and the fluid dynamic models of Ito et al. [1999]. In addition to binary mixing of the Iceland plume with the depleted asthenosphere a combination of (1) defluidization of the Iceland mantle plume occurring deeper than the dominant zone of dry decompression melting for enhancing He and heat transport along the southern Kolbeinsey Ridge and (2) fractional melting accompanying the northward dispersion and decompression of the Iceland mantle plume is required in order to explain the difference in wavelength between the He gradient and the Pb-Nd-Sr isotope gradients observed in mapping the dispersion of the Iceland plume along the MAR. Note that the Tjornes TZ does not act as a dam against the northward dispersion of the Iceland plume, contrary to what the rare earth element variation, by itself, previously suggested. Finally, over the Knipovich Ridge the large scatter and lack of any systematics between the Pb-Nd-Sr isotope and related parent/daughter ratios along this immature, discontinuous, and shear-dominated ridge, running parallel and close to the Svalbard continental break (74°–79°N), suggest the involvement in the melting of randomly distributed continental mantle lithosphere schlierens present in the depleted upper mantle source of the Knipovich Ridge basalts.

3 He/ 4 He ratios in Easter Microplate basalt glasses show clear evidence of the effects of a mantle plume. The East Rift of the microplate between 26 and 28°S, identified by La/Sm, Sr and Pb isotopes and ridge crest elevation as the region of maximum plume influence, has 3 He/ 4 He ratios spanning the entire range from 7.5 to 11.7 R A . The Easter Microplate is the only section of the entire East Pacific Rise that is associated with a known ‘hotspot’ track (mantle plume) and has elevated 3 He/ 4 He ratios. Although most of the West Rift basalts contain MORB helium (8.0–8.7 R A ), the basalt closest to the East Rift has an elevated 3 He/ 4 He ratio (11.3 R A ), consistent with a significant plume component. The diversity in isotopic signatures also indicates that homogenization of isotopic anomalies does not occur, even in this region of ‘super-fast’ spreading. The overall 3 He/ 4 He/ 206 Pb/ 204 Pb and 3 He/ 4 He 87 Sr/ 86 Sr trends have positive correlations, although the high 3 He/ 4 He, 87 Sr/ 86 Sr and 206 Pb/ 204 Pb component is not a single value but rather a broad band. The apparent ‘decoupling’ between the He and Sr isotope distribution is modeled in the context of a plume source-migrating ridge sink. During channeling of the plume toward the ridge, helium is preferentially lost from the center of the channeled plume, resulting in lower He/Pb and He/Sr concentration ratios in the high 3 He/ 4 He component. Mixing trajectories in He Sr isotopic space between a LILE depleted asthenosphere and a variably degassed plume component provide a reasonably good fit to the data and may explain the isotope systematics of plume-ridge interactions in the context of modern theories of plume dynamics.

[1] The geochemical variations in basalts from Iceland and the Reykjanes Ridge have for a long time been attributed to mixing between the depleted mid-ocean ridge basalt (MORB) asthenosphere source and the enriched Iceland mantle plume (Schilling, 1973; Hart et al., 1973; Sun et al., 1975). In contrast, the occurrence of some Iceland basalts with relatively depleted incompatible trace element ratios and low Pb and Sr isotope ratios compared to normal MORB (NMORB) have been interpreted to indicate that the Iceland mantle plume transports both geochemically enriched and depleted material from the lower mantle into the upper mantle and that the depleted MORB asthenosphere source does not contribute significantly to the generation of the Iceland basalts (Thirlwall et al., 1994; Thirlwall, 1995; Hards et al., 1995; Kerr, 1995; Kerr et al., 1995; Fitton et al., 1997; Hardarson and Fitton, 1997; Hardarson et al., 1997; Kempton et al., 1998, 1999; Nowell et al., 1998). As a follow-up to this controversy, we show on the basis of new Nb, Zr, Y, La, and Sm elemental abundances and Hf isotope ratios reported here, and published trace element and isotope data for Iceland and the North Atlantic MOR (50°–79°N), that the apparent distinction between Iceland basalts and MORB can readily be accounted for by a three-component mixing model involving two incompatible trace element enriched components, one with relatively high, the other with low 206Pb/204Pb ratios, and the usual surrounding depleted MORB mantle source as the third component. The radiogenic 206Pb-rich component represents the hot Iceland mantle plume source, while the enriched but low-206Pb/204Pb EM1-like component most likely represents entrained subcontinental lithospheric material embedded in the North Atlantic depleted MORB source. The incompatible element and isotope ratio variations do not require a depleted Iceland plume (DIP) component, nor do they exclude the presence of a MORB depleted asthenosphere source component for some of the Iceland basalts, particularly in recent time.

Basalts dredged from the Galapagos Rift Zone between 85°W and 100°W were analyzed by electron microprobe to determine the chemistry of the glass exteriors and included phenocrysts, microphenocrysts and quench minerals. The basalts come from both “normal” mid-ocean ridge segments and from ridge segments that cross the Galapagos Platform. The basalts fall into two chemical and geographical groups. Group A basalts come from outside the central region of the Galapagos Platform (i.e., outside 89–92.5°W) and are chemically similar to basalts from “normal” ocean ridge segments. Group B basalts come from the center of the Galapagos Platform (89–92.5°W) and are enriched in incompatible elements like “plume-influenced” basalts from the Mid-Atlantic Ridge. The spinel, olivine, plagioclase and clinopyroxene phenocrysts in both groups of basalts are low-pressure, equilibrium phases, but the chemical difference among basalts from within each group indicates high-pressure fractional crystallization is also responsible for the chemical evolution of some of these basalts. Presently, no crystallization or partial melting model can relate the chemistry of the two groups of basalts and the compositional influence of a large-ion-lithophile elements and water-rich mantle beneath the Galapagos Platform is a viable alternate hypothesis. The eruption temperature of magmas from the “normal” ridge segments, as determined by olivine-liquid thermometry is 1217±10°C, suggesting steady-state conditions but on the Galapagos Platform the eruption temperatures are lower and more variable than on the “normal” ridge (1186°C±30°C) suggesting a more complex plumbing system and the absence of a steady-state magma chamber beneath the platform.

A simple radiochemical neutron activation analysis scheme has been developed for the determination of chlorine in silicate rocks. The method involves a 15-min thermal neutron irradiation of rock powder followed by a quick separation of 38 Cl as AgCl, and Ge(Li) spectrometry. Chemical yield, normally ranging between 95% and 100%, is monitored gravimetrically through the recovery of AgCl. The procedure has been tested on several geochemical standards to assess its accuracy and precision. The values obtained for standard rocks agree with the literature values. At the 100-ppm level, the analytical precision for chlorine is within ±5% (2σ).

Three groups of spinels have been identified in dredged basalts from the Mid-Atlantic Ridge in the Azores region (30–40°N): (1) magnesiochromites with 0.4–0.5 Cr/(Cr + Al) are most common and characteristic of olivine tholeiites of the region; (2) titaniferous magnesiochromites are found in an olivine basalt with alkali affinities, of local occurrence and evolved in relatively high fugacity of oxygen; (3) chromian spinels with 0.23 Cr/(Cr + Al) occur in unusual high-Al picrites of local occurrence and possible high-pressure origin. Spinels are restricted in occurrence to the least fractionated lavas, with FeO * /FeO * + MgO ratio less than 0.575 and with Cr content greater than 350 ppm. A close relationship between Al content of liquidus spinel and Al content of magma has been observed for basaltic types. High-Al spinels deviating from this relationship, such as those found in picritic lavas from the Mid-Atlantic Ridge, may have crystallized at high pressure. The use of spinels as geobarometers in magmas of a restricted compositional range seems a promising prospect. There is no evidence of systematic variation in spinel chemistry of occurrence along the Mid-Atlantic Ridge, such as could be related to different mantle sources of the basalts, plume versus non-plume or binary mantle mixing.

The origin of the Tuzo Wilson Seamounts, 50 km south of the Queen Charlotte Islands, has been ascribed by various workers to either the Pratt–Welker mantle plume, which has formed the Pratt–Welker seamount chain, or the formation of a new segment of the Explorer–Juan de Fuca spreading ridge system. Abundances of major and trace elements in dredged alkali basalts from Tuzo Wilson and Bowie seamounts (360 km northwest of Tuzo Wilson Seamounts) are typical of alkaline volcanism on ocean islands associated with mantle plumes, but 87Sr/86Sr ratios (0.70252–0.70258) fall within the range of mid-ocean ridge basalts (MORB) from the Explorer and Juan de Fuca ridges. Geochronological and chemical data from the Pratt–Welker, Bowie, and Tuzo Wilson seamounts suggest that the Tuzo Wilson Seamounts are in an early stage of development as a result of activity of the Pratt–Welker mantle plume but that contributions from both a depleted and an undepleted mantle source are necessary to reconcile trace-element and Sr isotope values. Modelling of rare-earth behaviour during partial melting indicates that neither the Tuzo Wilson nor Bowie basalts could be generated from a mantle source similar to that of the Explorer or Juan de Fuca MORB, unless recent metasomatism has enriched the seamounts' source region in incompatible elements.

F, Cl and Br contents of tholeiitic volcanic glasses dredged along the Mid-Atlantic Ridge from 53° to 28° N, including the transect over the Azores Plateau, are reported. The halogen variations parallel those of 87 Sr/ 86 Sr, La/Sm or other incompatible elements of varying volatility. The latitudinal halogen variation pattern is not obliterated if only Mg-rich lavas are considered. Variations in extent of low-pressure fractional crystallization or partial melting conditions do not appear to be the primary cause of the halogen variations. Instead, mantle-derived heterogeneities in halogens, with major enrichments in the mantle beneath the Azores, are suggested. The Azores platform is not only a ‘hotspot’ but also a ‘wetspot’, which may explain the unusually intense Azores volcanic activity. The magnitude of the halogen and incompatible element enrichments beneath the Azores appear strongly dependent on the size of these anions and cations, but independent of relative volatility at low pressure. The large anions Cl and Br behave similarly to large cations Rb, Cs and Ba, and the smaller anion F similarly to Sr and P. Processes involving crystal and liquid (fluid and/or melt), CO 2 rather than H 2 O dominated, seem to have produced these largescale mantle heterogeneities. Geochemical ‘anomalies’ beneath the Azores are no longer apparent for coherent element pair ratios of similar ionic size. Values of such ‘unfractionated’ coherent trace element ratios provide an indication of the mantle composition and its nature before fractionation event (s) which produced the inferred isotopic and trace element heterogeneities apparently present beneath the North Atlantic. The relative trace element composition of this precursor mantle does not resemble that of carbonaceous chondrites except for refractory trace element pairs of similar ionic size. It is strongly depleted in halogens, and to a lesser extent in large alkali ions Rb and Cs relative to refractory Ba. These relative depletions are comparable within a factor of 5 to Ganapathy & Anders’s estimates for the bulk Earth, with the exception of Cs. There is also evidence for removal of phosphorus into the iron core during its formation. With the exception of San Miguel, alkali basalts from the Azores Islands appear to have been derived from the same mantle source as tholeiitic basalts from the ridge transect over the Azores Platform but by half as much degree of partial melting. The Azores subaerial basalts seem to have been partly degassed in Cl, Br and F, in decreasing order of intensity. A working model involving metasomatism from release of fluids at phase transformation during convective mantle overturns is proposed to explain the formation of mantle plumes or diapirs enriched in larger relative to smaller halogen and other incompatible trace elements. The model is ad hoc and needs testing. However, any other dynamical model accounting for the 400 -1000 km long gradients in incompatible trace elements, halogens and radiogenic isotopes along the Mid-Atlantic Ridge should, at some stage, require either (1) some variable extent of mixing or (2) differential migration of liquid relative to crystals followed by re-equilibration (or both), as a diffusion controlled mechanism over such large distances is clearly ruled out, given the age of the Earth.

The concentrolion and isotopic composi1ion of helium and hydrogen have been measured in submarine basalt glasses from the Mid-Atlantic Ridge between 52° and 73°N and in recent volcanic rocks from Iceland. Observations or “plume-type” 3 He/ 4 He ratios ( R greater than the mean MORB value of 8 R A ) show that the effect of the Icelandic mantle plume on helium isotope rn1ios is observable along the entire length of the Reykjanes Ridge ( R = 11−16.1 R A ) and to 70°N on the Kolbeinsey Ridge ( R = 10.3−12 R A ). These two ridges are unique among all mid-ocean ridge segments so for studied in possessing high 3 He/ 4 He ra1ios along their entire length. Northof 70°N, basalts associated with the Jan Mayen province (based on major element chemistry) have 3 He/ 4 He ratios somewhat lower than the MORB average ( R = 6.8−8.0 R A ) and similar to other alkali-enriched provinces such as the Azores Platform. In detail, however, the 3 He/ 4 He distribution does not correspond to trace element variations. A maximum ratio of 16.1 R A is present near 60°N on the Reykjanes Ridge: north of this point, toward Iceland, the ratio decreases to a minimum of 12.7 R A at 63°N. The 3 He/ 4 He ratios increase again on Iceland, reaching a maximum in south-central Iceland. In the area between 60° and 63°N, the 3 He/ 4 He ratios tend to be inversely correlated with La/Sm and 87 Sr/ 86 Sr, which suggests that the isotopic heterogeneities beneath Iceland and the Reykjanes Ridge are long lived (> 10 8 years). The concentration and hydrogen isotopic composition of the water in this suite of basalts display a good positive correlation with trace clement data (e.g. La/Sm). Between 61°N and 63°N on the Reykjanes Ridge, the water concentration in the basalt is high (0.22–0.41 wt.%) and its isotopic composition is enriched in deuterium (δD = −61 to −74‰; with one extreme value to −50‰), relative to MORB levels (0.2 wt.% and −77‰). South of 61 °N, both the concentration and isotopic composition of water are normal MORB values (0.11–0.25 wt.% and δD = −71 to −91‰). The Mohns Ridge basalts have elevated water concentrations (0.37–1.23 wt.%) and an isotopic composition (−44 to −60‰) which is unique among all MOR basalts so far studied. It remains to be seen if thesehigh δ D values represent a “primary” mantle water, distinct from the MORB source, or “secondary” water, the result of recycling during subduction.

Forty-two fragments of young, fresh basalts, dredged from the Explorer Ridge, Paul Revere Ridge (Fracture Zone), and Dellwood Knolls, have been analysed for 34 major, minor, and trace elements, and 87Sr/86Sr ratios have been determined in seven of the fragments.The Explorer Ridge basalts have major element compositions similar to most mid-ocean ridge basalts (MORB), particularly to the iron-rich basalts of the southern Juan de Fuca Ridge. The basalts exhibit variability in trace element content, largely attributable to crystal fractionation at low pressure. With respect to MORB the incompatible minor and trace elements are weakly to strongly enriched in the Explorer samples, and are most strongly enriched in basalts from Explorer Deep. Adjacent ridge segments erupt basalts with variable incompatible element concentrations and ratios. This could be an indicator of a chemically heterogeneous mantle source, or may be the result of intermittent injection of enriched magmas from a hotspot beneath Explorer Deep. 87Sr/86Sr ratios are similar for both basalt types, and values are typical of most MORB.Based on their very different rare earth element patterns and 87Sr/86Sr ratios, the two Dellwood Knolls appear to have different mantle sources, one typically depleted and one chemically and radiogenically enriched.

Supplement is available with entire article on microfiche. Order from American Geonhvsical Union, 1909 K Street, N.W., Washington, D.C. 20006. Document J75-002; $1.00. Payment must accompany order. Rare-earth (RE) abundance patterns in 26 basalts occurring across normal segments of the Reykjanes ridge, 60°–53°N; the mid-Atlantic ridge, 29°S (Deep-Sea Drilling Project leg 3); and the East Pacific rise, 2°–19°S, including the Nazca plate, are reported. Without exception, all the RE patterns were found to be characteristically light RE depleted, reflecting to a first-order approximation, a mantle source with uniquely defined RE content. Computer multiphase and eutectic, equilibrium-partial melting models show that the reconstituted RE pattern of the mantle underlying these regions must be more depleted in light RE than the basalts generated. This is so whether a lherzolite (spinel free or bearing), garnet peridotite, or plagioclase peridotite is considered for the primary mantle source. The source is thought to be the low velocity layer apparently depleted in light RE and other large ionic lithophile elements. A lherzolite phase assemblage is found to be preferable for the depleted low-velocity layer source under normal midocean ridge segments at a depth greater than 30–38 km. Above 30–38 km, equilibration of normal ridge basalts with a plagioclase peridotite cannot be ruled out. Rare-earth results across these three normal ridge segments further emphasize (1) the spatial and temporal RE uniformity of the depleted low-velocity layer source, (2) the widespread geographic coverage of this mantle source, (3) its RE composition contrast with mantle source regions, such as those beneath Iceland, the Azores, and the Afar, which have been assumed to represent rising plumes (or blobs), and finally (4) the passive nature of volcanism along normal ridge segments occurring in response to the spreading as a simple ‘healing’ process.

Nd and Sr isotope variations along the Galapagos spreading centre support Morgan's preferential subcrustal flow channel connecting the Galapagos mantle plume with the northeastward migrating ridge axis. Decoupling of La/Sm and Nd isotope variations near 91° W further suggests that basalts are derived from smaller degrees of melting and are deeper beneath this elevated part of the ridge.

Nd and Sr isotopic compositions and Rb, Sr, Sm and Nd concentrations are reported for madupites, wyomingites and orendites from the Pleistocene volcanic field of the Leucite Hills, Wyoming. All Leucite Hills rocks have negative eNd signatures, indicating derivation or contribution from an old light rare earth element (LREE) enriched source. In this respect they are similar to all occurrences of high potassium magmas so far investigated. But Sr isotopic variations are comparatively small and 87Sr/ 86Sr ratios are unusually low for high-K magmas (0.7053–0.7061, one sample excluded). These values suggest that the light REE enrichment of the source was not accompanied by a strong increase in Rb/Sr. Wyomingites and orendites are isotopically indistinguishable which is consistent with chemical and petrographic evidence for their derivation from a common magma series depending on emplacement conditions. Basic to ultrabasic madupites and more silicic wyomingites/orendites are distinct in their Nd isotopic variations (madupites: eNd= −10.5 to −12.3; wyomingites/orendites: eNd= −13.7 to −17.0) despite similar Sm/Nd ratios and complete overlap in 87Sr/86Sr. Selective or bulk assimilation of crustal material is unlikely to have significantly affected the Nd and Sr isotopic compositions of the magmas. The measured isotopic ratios are considered to reflect source values. The distinct isotopic characteristics of madupite and wyomingite/orendite magmas preclude their derivation by fractional crystallization, from a common primary magma, by liquid immiscibility or by partial melting of a homogeneous source. Two isotopically distinct, LREE enriched and slightly heterogeneous sources are required. Heterogeneities were most pronounced between magma sources from each volcanic centre (butte or mesa). The relationship between the madupite and wyomingite/orendite sources and their evolution is discussed on the basis of two simple alternative sets of models

The width of geochemical and residual elevation anomalies along the Mid-Ocean Ridge System decrease as both the distance between migrating ridge segments and related hotspots and the local geochemical dispersion in the anomalies increase. These observations provide further evidence that flows of mantle plumes in the upper mantle are deflected by migrating ridge sinks once they override hotspots.

The trace element composition of basalts dredged at the axis of the Mid-Atlantic Ridge between 12°N and 17°N by the R/V “Akademik Boris Petrov” demonstrates the presence of a high-amplitude geochemical anomaly, centered around 14°N and extending at least 300 km along the strike of the Rift Valley. The anomaly does not fit easily into any of the models that have been proposed: it may reflect the upwelling of an embryonic mantle plume or of a passive mantle domain responsible for or associated with a triple junction, possibly marking a recent shift into the area of the South American/North American plate boundary.

There is a growing Easter microplate between the Pacific and Nazca plates. The eastern and western boundaries of the microplate are actively spreading and propagating north and south, respectively. The northern and southern boundaries are broad transform zones. Most of the interior of the microplate contains the sort of oblique sheared structures characteristic of zones of transferred lithosphere between propagating and failing spreading centres.

A neutron activation technique for determining gallium and germanium in igneous rocks is described. After irradiation, germanium is separated from interfering activities by extraction with carbon tetrachloride from 9 M HCl solution. Gallium is extracted from 6 M HCl solution into isopropyl ether. Germanium and gallium activities are determined by counting the radionuclides 77 Ge and 72 Ga, respectively. Chemical yields are determined by re-irradiation of separated solutions. About 12 man-hours are necessary to process 6 samples, excluding re-irradiation. Precision of the analysis for both elements is ±2–3% for concentrations encountered in typical igneous rocks. The detection limits for a thermal neutron flux of 4 × 10 12 n cm -2 s -1 are about 0.0005 ppm for gallium and 0.01 ppm for germanium with an error of ±50%.

The relations of geoid data from radar altimeter measurements (GEOS 3) to water depth and geochemical variations along the Mid-Atlantic Ridge are investigated. Eight geophysical lines across the ridge in the Atlantic Ocean were constructed approximately normal to the local ridge trend. Slopes of geoid per kilometer of topographic relief show an asymmetry between the two sides of the ridge that generally is not matched by topography, which has better symmetry between the two flanks. Subtracting a regional field, based on previous work, produces residual geoid anomalies whose source must be in the earth's outer 600 km. The residual geoid anomalies of the North Atlantic show striking correlation with the axis of the Mid-Atlantic Ridge, but those in the South Atlantic do not. Instead, major positive residual geoid anomalies appear over the Rio Grande and Walvis ridges. Determination of slope S, residual geoid per million years crustal age, reveals that S increases with greater residual geoid anomaly and with higher (shallower) crestal ridge topography. The slope of 22 cm/m.y. west of the Azores high (near 38°N) suggests that the anomalously high crestal residual geoid and depth anomalies may not be entirely of thermal origin, which may normally produce slopes only of about 8 cm/m.y. as found near 28°N away from shallow ridge topography. To account for this difference in slopes simply by a variation in mantle temperature would require a difference of about 940°C between the two sites, which appears excessive. Also, the decrease in topographic height with crustal age away from the Azores high is much greater than predicted by theoretical cooling curves. Correlation of residual geoid anomalies with depth, geochemical variations along the length of the Mid-Atlantic Ridge, and the chemistry of the Rio Grande Rise and Walvis Ridge suggest that many regions of anomalous residual geoid and crestal depth values may be in large part a function of crustal thickness and density variations induced by plume intrusion and volcanism in addition to their transient thermal effects.

Ge and Ga abundances in 19 petrologically and geochemically well-documented Hawaiian volcanic rocks were determined simultaneously by radiochemical neutron activation analysis. The data demonstrate a close affinity of Ga for AI and Ge for Si over the entire range of the nephelinitic, alkalic and tholeiitic lava series studied. Ga and Ge become progressively enriched in the final products of crystallization of the alkali series. No systematic pressure effect on the Ga/Al and Ge/Si ratios could be detected over the pressure range at which nephelinites, alkali olivine basalts and tholeiites have presumably been generated. Ga/Al ratio varies by less than 62% within the three Hawaiian series, and decrease slightly toward residual melts of the alkali series. Ge/Si ratio varies generally by less than 20% within the three Hawaiian series, and can, within experimental errors, be considered independent of degree of partial melting and extent of fractional crystallization effective in the generation of these lavas. We suggest that the Ge/Si ratio in basalts may be a useful mantle source indicator and should help mapping mantle heterogeneities.

Continuous fluxing at a decreasing rate of mantle Cl and Br by volcanism throughout the Earth's history can account for the mass of Cl and Br, and the Cl/Br ratio, of the oceans.

Short-wavelength geochemical anomalies in the basaltic layer 2A occur along the South Atlantic mid-ocean ridge opposite Ascension, St Helena and Tristan hotspots, further evidence that preferential subcrustal flow channels develop between mantle upwelling zones (hotspots) and migrating ridge axes. Geochemical variability suggests that the plume along these channels would be schlieren-like rather than continuous.

Chlorine and bromine variations in basalts erupted along the Reykjanes Ridge axis and Iceland follow a predicted first-order model, reflecting volatile degassing at water depths shallower than about 500 m, superimposed with changes due to the presence of two chemically distinct mantle sources and their mixing. Iceland mantle ‘plume’ contains three to four times more pristine Cl and Br than the asthenosphere depleted in large ionic lithophile elements.

A continuous chain of volcanic islands and seamounts extends on the Nazca plate eastward from the Easter and Sala y Gomez islands to the islands of San Felix and San Ambrosio, off the coast of Chile (Easter line), in a latitudinal band about 200 km wide. A similar chain of seamounts and islands probably extends west of Easter Island on the Pacific plate up to Pitcairn Island, which was recently active. The hypothesis that Easter Island marks the site of a fixed mantle hot spot responsible for these chains is tested here. Easter Island consists predominantly of basalts but also contains terms of a differentiation series including trachytes and rhyolites. The Easter Island basalts have higher contents of Fe, Ti, Ba, and Zr and higher 87Sr/86Sr and 206Pb/204Pb ratios than ocean ridge basalts; they also have light rare earth enriched patterns. Thus they are similar to basalts which elsewhere (i.e., Iceland and the Azores) have been ascribed to mantle plumes. Rocks from Sala y Gomez, San Felix, and San Ambrosio and various seamounts along the Easter line have similar petrochemistry. But consideration of the spreading history of the Nazca plate (with the fossil Galapagos ridge as an active spreading center up to about 10 m.y. ago and spreading occurring subsequently from the east Pacific rise) excludes the possibility that the Easter line chain could have been formed by one mantle hot spot. At least two hot spots had to be active, one of them located at the intersection of the fossil ridge with the Easter line. These hot spots must have moved with respect to each other. Recent (

Ge/Si and Ga/Al variations emphasise the distinct chemical nature of the mantle present beneath Iceland as compared with the Reykjanes Ridge. Generation of basalts at greater depth beneath Iceland than the ridge is suggested.

Strontium isotope variations in basalts from the Mid-Atlantic Ridge and the Azores Islands support the Azores mantle plume model. The similarity of isotope ratios in Mid-Atlantic Ridge tholeiites from the Azores platform and Azores Islands alkali basalts implies that these different magma types are derived from the same mantle source but under different conditions of origin.

Rare earths, Fe and Ti concentrations in basalts dredged along the Galapagos spreading centre reveal that a juvenile mantle plume is welling up beneath the Galapagos Islands and mixing with the depleted asthenosphere to the north-east and north-west beneath the ridge. Unusually extensive shallow-depth, crystal fractionation is also apparent on the ridge between 85 and 87°W. The ‘magnetic telechemistry’ hypothesis is further clarified.

Geochemical evidence confirms that there is current active spreading on both the East and West Rift zones of the Easter microplate, and northwards propagation of the East Rift with the active tip at 25° S. The East Rift is influenced by lateral flow of plume-derived material from a hotspot located beneath either Sala y Gomez or Easter Island. Spreading rate variation does not seem to be influential in controlling the composition of erupted basalts.

Strontium isotope ratios in basalts demonstrate two distinct mantle sources under Iceland and the Reykjanes Ridge.

Rare earth pattern zonation about the Afar Triangle, in tholeiitic basalts erupting along the Red Sea Trough and the Gulf of Aden, suggests the presence of a primordial mantle plume rising beneath the Afar and overflowing in a star-like fashion into the soft asthenosphere.

An epithermal neutron activation analysis (ENAA) scheme involving a single-step bromide precipitation coupled with Ge(Li) spectrometry has been developed for determining trace amounts of Br in silicate rocks. The precision (2 sigma) of the method estimated from replicate analyses of rock standards is within +-10% for concentrations of Br exceeding 0.1 ppM. The observed values for Br in W-1 and JB-1 are very close to the reported values in literature. Commercial grade aluminum foil was used as sample container and flux monitor for Br analysis.

There is a progressive secular decline in light-RE enrichment from Tertiary to Present, similar to that previously observed on the Faeroes 50–60mybp. Light-RE depleted oceanic type patterns appear on Iceland only within the last 0.7my. These results suggest pulsating mantle plume activity with time and provide further support to the blob model previously proposed for the origin of the Faeroes-Iceland-Greenland Ridge. Available Sr and Nd isotopic data are consistent with the model.

New major and trace element data are reported for a suite of basalts dredge sampled from the southern MAR (40.6–52.5°S) during cruise 9309 of the R/V Maurice Ewing and cruise 32 of the S.A. Agulhas (51–54.5°S). Samples range from aphyric to moderately porphyritic with plagioclase and olivine the dominant phenocryst and microphenocryst phases. Clinopyroxene occurs rarely as a phenocryst phase. Bulk rock and quench glass samples have major element compositions ranging from moderately primitive (Mg#=68.5) to evolved (Mg#=41.1), with a comparable range in incompatible (Zr=54–232 ppm; Nb=0.8–21 ppm) and highly compatible trace element (Ni=51–289 ppm; Cr=39–949 ppm) abundances. Incompatible trace element ratios (e.g., Zr/Nb=5.9–69; Y/Nb=0.88–29; (La/Sm)N=0.48–2.9) indicate both enriched and depleted MORB occurring along this section of the MAR, agreeing with along-ridge radiogenic isotope ratio variations delineating compositional influences of the off-axis Discovery, and a so-called LOMU-component, and ridge-centered Shona mantle plumes (Douglass et al. 1999; Douglass and Schilling 2000). Major- and trace-element variations, particularly decreasing CaO/Al2O3 ratios and relatively constant Sc abundances with decreasing Mg# suggest crystallization of olivine, plagioclase, and clinopyroxene in most of these basalts prior to eruption. Liquid-lines-of-descent, derived from forward modeling of various reasonable parental magma compositions at pressures between 1 atm–8 kbar, confirm clinopyroxene as an early crystallizing phase, along with olivine and plagioclase, and its pivotal role in the compositional evolution of these basalts. Pressures deduced using algorithms of Grove et al. 1992 and Michael and Cornell 1998 indicate crystallization beneath ridge segments unaffected by mantle plumes occurred predominately at depths of 3–6 kbar (10–20 km). Crystallization beneath ridge segments affected by the Discovery and Shona mantle plumes occurred over a larger pressure range (1 atm–~7 kbar), but predominantly at crustal depths (1 atm–3 kbar), possibly due to more constant magma fluxes and higher temperatures in the crust and lithosphere in these regions favoring the existence and longevity of subaxial magma chambers.

We report on a Pb-Nd-Sr isotope and rare earth study of Mid-Atlantic Ridge (MAR) basalt glasses collected across the equatorial fracture zones from 7°S to 5°N (65 stations). The 1600-km-long profile reveals two mixing zones in the mantle that are isotopically distinct but cover the same range of (La/Sm)n ratios (0.3–2), with a gradational boundary between the Romanche and the Chain fracture zones. The potential mantle temperature profile inferred from Na2O content is also quite distinct. The north zone is dominated by a major, La/Sm and HIMU type Pb isotope anomaly centered at 1.7°N±300 km, which is flanked by two zones mildly radiogenic in Pb but depleted in light REE. A kinematic and evolutionary model describing the dispersion and interaction of the Sierra Leone plume with the asthenosphere and the MAR in the last 75 m.y. is proposed for this zone, which includes St. Paul and St. Peter's Rocks. In contrast, over the south zone the isotope/geochemical profiles are well correlated at all length scales and opposite in sign from the inferred potential mantle temperature profile and mean percent fusion. Broad negative gradients are observed between the Romanche and the Charcot fracture zones, superimposed by spikelike anomalies at the intersection with the eastern part of the Romanche and Chain transform faults, where cold plate edge effects prevail. The heterogeneous mantle model of Sleep [1984] and Langmuir and Bender [1984] is applicable to this zone, that is the volatile and radiogenic Pb-rich lumps are preferentially melted during mantle decompression and passively sampled. The lumps may reflect the early dispersion of the St. Helena or Ascension mantle plumes under a thick lithosphere, followed by redistribution due to intense shearing, continental lithosphere delamination, and secondary mantle convection. The presence of a depleted asthenosphere unpolluted by plumes along the 400-km-long MAR segment between the Charcot and Ascension fracture zones is also apparent in the data.