Figures and Tables

top

Introduction

Earth’s silicate mantle is largely inaccessible to direct sampling yet bears the imprint of planetary formation and core segregation, melting to form the crust, and the recycling of crustal material through plate tectonics. Oceanic basalts form by partial melting of Earth’s upper mantle, and their concentration ratios of incompatible elements or radiogenic isotope compositions are considered to relate directly to that of the mantle source. The fundamental tenet underlying the use of radiogenic isotopes in such mantle-derived basalts is that they are in equilibrium with their mantle source (e.g., Hofmann and Hart, 1978

Hofmann, A.W., Hart, S.R. (1978) An assessment of local and regional isotopic equilibrium in the mantle. Earth and Planetary Science Letters 38, 44–62. https://doi.org/10.1016/0012-821X(78)90125-5

; Zindler and Hart, 1986

Zindler, A, Hart, S. (1986) Chemical geodynamics. Annual Review of Earth and Planetary Sciences 14, 493–571. https://doi.org/10.1146/annurev.ea.14.050186.002425

). Thus, the composition of the long lived isotopes of heavy elements in basalts and their upper mantle source should be the same.

For the most part the radiogenic isotope composition of Mid-Ocean Ridge Basalts (MORB) points to the depletion of incompatible elements (that is, those that preferentially partition into a silicate melt) in the mantle over much of Earth’s history (e.g., Hofmann, 2022

Hofmann, A.W., Class C., Goldstein, S.L. (2022) Size and composition of the MORB+OIB mantle reservoir. Geochemistry, Geophysics, Geosystems 23, e2022GC010339. https://doi.org/10.1029/2022GC010339

). Variations in the radiogenic isotope composition of MORB, however, indicate that there must be some degree of chemical heterogeneity in the mantle (Zindler and Hart, 1986

Zindler, A, Hart, S. (1986) Chemical geodynamics. Annual Review of Earth and Planetary Sciences 14, 493–571. https://doi.org/10.1146/annurev.ea.14.050186.002425

; Salters and Dick, 2002

Salters, V.J.M., Dick, H.J.B. (2002) Mineralogy of the mid-ocean-ridge basalt source from neodymium isotopic composition of abyssal peridotites. Nature 418, 68–72. https://doi.org/10.1038/nature00798

; Warren, 2016

Warren, J.M. (2016) Global variations in abyssal peridotite compositions. Lithos 248, 193–219. https://doi.org/10.1016/j.lithos.2015.12.023

; Hofmann, 2022

Hofmann, A.W., Class C., Goldstein, S.L. (2022) Size and composition of the MORB+OIB mantle reservoir. Geochemistry, Geophysics, Geosystems 23, e2022GC010339. https://doi.org/10.1029/2022GC010339

). These radiogenic isotope variations may be due to the presence of mantle material of either differing age, or differing concentrations of incompatible elements, resulting in variable radiogenic isotope compositions over time (or some combination of both factors). Furthermore, these chemical variations are often ascribed to the presence of discrete rock types in the mantle. Clinopyroxene is the dominant host of incompatible elements in the upper mantle, therefore, any mantle lithology relatively rich in clinopyroxene, such as pyroxenite, is likely to be enriched in incompatible elements relative to typical depleted mantle peridotite. Indeed, pyroxenites from orogenic peridotites often possess radiogenic isotope compositions that are distinct from peridotites from the same bodies (e.g., Warren, 2016

Warren, J.M. (2016) Global variations in abyssal peridotite compositions. Lithos 248, 193–219. https://doi.org/10.1016/j.lithos.2015.12.023

).

Evidence for mixing of magmas from both depleted and enriched sources in the mantle is seen in elemental variations (e.g., Rhodes et al., 1979

Rhodes, J.M., Dungan, M.A., Blanchard, D.P., Long, P.E. (1979) Magma mixing at mid-ocean ridges: evidence from basalts drilled near 22° N on the Mid-Atlantic ridge. Tectonophysics 55, 35–61. https://doi.org/10.1016/0040-1951(79)90334-2

), and radiogenic isotopes in erupted basalts (e.g., Stracke et al., 2003

Stracke, A., Bizimis, M., Salters, V.J.M. (2003) Recycling oceanic crust: Quantitative constraints. Geochemistry, Geophysics, Geosystems 4, 8003. https://doi.org/10.1029/2001GC000223

). Moreover, pioneering work showed how olivine-megacryst hosted melt inclusions preserve significantly greater chemical variability than their host lava, and that mixing must occur during crystallisation in the crust or shallow mantle (e.g., Sobolev, 1996

Sobolev, A.V. (1996) Melt inclusions in minerals as a source of principle petrological information. Petrology 4, 209–220. https://doi.org/10.1134/S0869591109030047

). More recently spatial variations in radiogenic isotopes preserved by magmatic minerals in both ancient and modern oceanic crust have been taken to indicate mixing and melt reaction (e.g., Jansen et al., 2018

Jansen, M.N., Lissenberg, C.J., Klaver, M., De Graaff, S.J., Koornneef, J.M., Smeets, R.J., MacLeod, C.J., Davies, G.R. (2018) Isotopic variation in Semail Ophiolite lower crust reveals crustal-level melt aggregation. Geochemical Perspectives Letters 8, 37–42. https://doi.org/10.7185/geochemlet.1827

; Lambart et al., 2019

Lambart, S., Koornneef, J.M., Millet, M.A., Davies, G.R., Cook, M., Lissenberg, C.J. (2019) Highly heterogeneous depleted mantle recorded in the lower oceanic crust. Nature Geoscience 12, 482–486. https://doi.org/10.1002/2015JB012762.

), dramatically reducing the chemical variability seen in erupted basalts. Nevertheless, that radiogenic isotope variations are still preserved, both globally and at the scale of individual ridge segments, indicates that magma mixing is incomplete.

The problem that remains is that the range of radiogenic isotope variations seen in erupted basalts appears to be biased towards enriched compositions. It is now well established that many MORB possess radiogenic isotope compositions that must be derived from a mantle rock type that is more enriched than typical depleted peridotite, likely pyroxenite (e.g., Salters and Dick, 2002

Salters, V.J.M., Dick, H.J.B. (2002) Mineralogy of the mid-ocean-ridge basalt source from neodymium isotopic composition of abyssal peridotites. Nature 418, 68–72. https://doi.org/10.1038/nature00798

; Warren, 2016

Warren, J.M. (2016) Global variations in abyssal peridotite compositions. Lithos 248, 193–219. https://doi.org/10.1016/j.lithos.2015.12.023

). As the mantle upwells beneath mid-ocean ridges it is inevitable that enriched, more fusible rock types, such as pyroxenites will melt first (e.g., Pertermann and Hirschmann, 2003

Pertermann, M., Hirschmann, M.M. (2003) Partial melting experiments on a MORB-like pyroxenite between 2 and 3 GPa: constraints on the prescence of pyroxenite in basalt source regions from solidus location and melting rate. Journal of Geophysical Research: Solid Earth 108, 2125. https://doi.org/10.1029/2000JB000118

). Although the abundance of pyroxenites in the mantle is thought to be low, around 2–5 % (e.g., Pertermann and Hirschmann, 2003

Pertermann, M., Hirschmann, M.M. (2003) Partial melting experiments on a MORB-like pyroxenite between 2 and 3 GPa: constraints on the prescence of pyroxenite in basalt source regions from solidus location and melting rate. Journal of Geophysical Research: Solid Earth 108, 2125. https://doi.org/10.1029/2000JB000118

), they appear to make a disproportionate contribution to melts forming the oceanic crust, perhaps because of their lower solidus temperature, higher melt productivity and higher elemental abundance (e.g., Shorttle et al., 2014

Shorttle, O., Maclennan, J., Lambart, S. (2014) Quantifying Lithological Variability in the Mantle. Earth and Planetary Science Letters 395, 24–40. https://doi.org/10.1016/j.epsl.2014.03.040.

).

An alternative possibility is that melt transport and crystallisation is controlled by the magma chemistry itself. Recent experiments indicate that magmas from enriched mantle rocks retain a higher melt fraction during cooling and ascent, largely because plagioclase crystallisation is suppressed to lower temperatures in the Na- and H₂O-rich, but Ca- and Al-poor, liquids derived from enriched source lithologies (Neave et al., 2019

Neave, D.A., Namur, O. Shorttle, O., Holtz, F. (2019) Magmatic evolution biases basaltic records of mantle chemistry towards melts from recycled sources. Earth and Planetary Science Letters 520, 199–211. https://doi.org/10.1016/j.epsl.2019.06.003

). The preferential survival of enriched melts during cooling and crystallisation biases the compositions of erupted basalts towards those from enriched mantle sources (Neave et al., 2019

Neave, D.A., Namur, O. Shorttle, O., Holtz, F. (2019) Magmatic evolution biases basaltic records of mantle chemistry towards melts from recycled sources. Earth and Planetary Science Letters 520, 199–211. https://doi.org/10.1016/j.epsl.2019.06.003

). In contrast, melts from depleted mantle peridotite will start to crystallise at higher temperatures in the lower crust (or mantle). If depleted melts crystallise at greater depths in the lower oceanic crust or mantle then this raises the intriguing possibility that their radiogenic isotope signatures are not simply erased by mixing but rather are not represented by the melts that arrive at the surface.

top

Rationale

What is needed is a radiogenic isotope system that is highly susceptible to differing degrees of incompatible element enrichment in the mantle, and magmatic minerals that preserve a robust record of the chemistry of magmas that crystallised in the lower oceanic crust,

The lead (Pb) isotope system, involving the decay of uranium (U) and thorium (Th) to the isotopes of Pb, provides an extremely sensitive tracer of chemical variability in the mantle. Uranium and Th are highly incompatible (relative to Pb) and therefore likely to be depleted in mantle rocks that have experienced melting. Depleted mantle rocks will, consequently, possess low U/Pb and Th/Pb ratios and evolve to relatively unradiogenic Pb isotope compositions over time. In contrast, enriched mantle rock types, such as enriched melts from prior melting trapped in the mantle, or recycled crustal material, with high U-Th/Pb ratios, acquire relatively radiogenic Pb isotope compositions over time that can be readily traced in the melts that form the oceanic crust.

Mid-Ocean Ridge Basalts (MORB) commonly contain calcic plagioclase and magnesian olivine crystals that are physically entrained in more evolved and enriched host lavas from which they cannot have crystallised (e.g., Rhodes et al., 1979

Rhodes, J.M., Dungan, M.A., Blanchard, D.P., Long, P.E. (1979) Magma mixing at mid-ocean ridges: evidence from basalts drilled near 22° N on the Mid-Atlantic ridge. Tectonophysics 55, 35–61. https://doi.org/10.1016/0040-1951(79)90334-2

). Recent work, using the volatile contents of olivine- and plagioclase-hosted melt inclusions, indicates that crystallisation of plagioclase megacrysts often occurs in the lithospheric mantle (Drignon et al., 2018

Drignon, M.J., Nielsen, R.L., Tepley III, F.J., Bodnar, R.J. (2018) Reequilibration processes occurring in plagioclase‐hosted melt inclusions from plagioclase ultraphyric basalts. Geochemistry, Geophysics, Geosystems 20, 109–120. https://doi.org/10.1029/2018GC007795.

; Bennett et al., 2019

Bennett, E.N., Jenner, F., Millet, M.-A., Cashman, K.V., Lissenberg, C.J. (2019) Deep roots for mid-ocean-ridge volcanoes revealed by plagioclase-hosted melt inclusions. Nature 572, 235–239. https://doi.org/10.1038/s41586-019-1448-0

), while crystallisation of olivine may occur at shallower levels (Bennett et al., 2019

Bennett, E.N., Jenner, F., Millet, M.-A., Cashman, K.V., Lissenberg, C.J. (2019) Deep roots for mid-ocean-ridge volcanoes revealed by plagioclase-hosted melt inclusions. Nature 572, 235–239. https://doi.org/10.1038/s41586-019-1448-0

), in the lower oceanic crust. For both the plagioclase and olivine megacrysts, their major element chemistry points to crystallisation from incompatible element depleted magmas (Sobolev et al., 2007

Sobolev, A.V., Hofmann, A.W., Kuzmin, D.V., Yaxley, G.M., Arndt, N.T., Chung, S.L., Danyushevsky, L.V., Elliott, T., Frey, F.A., Garcia, M.O. Gurenko, A.A. (2007) The amount of recycled crust in sources of mantle-derived melts. Science 316, 412–417. https://doi.org/10.1126/science.1138113

; Neave and Namur, 2022

Neave, D.A., Namur, O. (2022) Plagioclase archives of depleted melts in the oceanic crust. Geology 50, 848–852. https://doi.org/10.1130/G49840.1

) typical of those expected from the melting of depleted mantle peridotite.

top

Methods

Here we present high precision Pb isotope data (Table 1) for sulfide and melt inclusions trapped in an olivine megacryst in a single MORB sample from the FAMOUS ridge segment, and for plagioclase megacrysts and host basaltic glass from several localities in the Pacific and Atlantic oceans (most of which are thought to have crystallised in the lower crust or mantle) (see Supplementary Information for a full description of the methods used, and major element data for all phases; Table S-1). Plagioclase and sulfide have relatively high Pb concentrations and are amenable to precise isotope measurement. Additionally, they possess low U/Pb ratios, so there is little ingrowth of Pb subsequent to crystallisation. Furthermore, for plagioclase, diffusion rates of Pb are often sufficiently low that the isotope composition is not significantly modified at magmatic temperatures (Cherniak, 1995

Cherniak, D.J. (1995) Diffusion of lead in plagioclase and K-feldspar - an investigation using Rutherford backscattering and resonant nuclear-reaction analysis. Contributions to Mineralogy and Petrology 120, 358–371. https://doi.org/10.1007/BF00306513

).

			206Pb/204Pb	+/− (2 s.d.)	207Pb/204Pb	+/− (2 s.d.)	208Pb/204Pb	+/− (2 s.d.)	[Pb], ppm	An, %
FAMOUS, North Atlantic
ARP1973-010-003	36° 83.7' N	33° 24.8' W
matrix			18.514	0.002	15.506	0.003	38.090	0.008	0.231
glass 1			18.669	0.004	15.523	0.004	38.173	0.010	0.175
glass 2			18.581	0.005	15.528	0.003	38.198	0.008
melt inclusions in olivine			17.447	0.030	15.381	0.033	37.094	0.008	0.240
sulfide 1			17.595	0.024	15.404	0.023	37.258	0.052
sulfide 2			17.748	0.027	15.431	0.029	37.429	0.058
spinel			17.345	0.002	15.375	0.005	36.968	0.012	0.355
plagioclase			18.054	0.002	15.433	0.003	38.038	0.008	0.244	82
CH31 DR01-76	36.50° N	33.16° W
matrix			18.446	0.008	15.462	0.010	38.033	0.032	0.346
plagioclase			18.041	0.002	15.414	0.003	37.604	0.008	0.150
ARP1974-014-013	36.85° N	33.25° W
plagioclase			18.257	0.006	15.427	0.006	37.797	0.019	0.104	87
ALV 518-3-1	36.82° N	33.26° W
glass			18.786	0.002	15.538	0.003	38.351	0.008	0.368
plagioclase			17.649	0.005	15.408	0.004	37.430	0.016	0.223	91
South Atlantic
EW9309 12D	49.76° S	8.03° W
glass			18.348	0.002	15.504	0.003	37.828	0.008	0.344
plagioclase			17.606	0.002	15.362	0.003	37.250	0.008	0.213	81
Garrett fracture zone, East Pacific
GN4-11	13° 26.9' S	111° 29.5' W
glass*			17.903	0.014	15.444	0.018	37.327	0.040	0.271
plagioclase			17.493	0.011	15.320	0.010	36.994	0.028	0.024	94
GN12-10	13° 24.2' S	111° 56.5' W
glass*			17.655	0.014	15.363	0.018	37.004	0.040	0.294
plagioclase			17.480	0.002	15.325	0.003	36.899	0.008	0.093	94
Iceland, Central rift zone
Borgarhraun	65° 50.35' N	17° 1.07' W
matrix			18.085	0.002	15.438	0.005	37.663	0.008	0.246
plagioclase			17.458	0.005	15.337	0.005	37.119	0.008	0.137	85

All errors are 2 s.d. mean. Concentrations in ppm weight.
Pb isotopic ratios relative to NBS 981 Pb standard which yields an external precision of ²⁰⁶Pb/²⁰⁴Pb = 16.9428 ± 24, ²⁰⁷Pb/²⁰⁴Pb = 15.5007 ± 28 and ²⁰⁸Pb/²⁰⁴Pb ratio of 36.7276 ± 75 (2 s.d., n = 59).
* Pb isotope data for glass samples from Wendt et al. (1999)Wendt, J.I., Regelous, M., Niu, Y., Hékinian, R., Collerson, K.D. (1999) Geochemistry of lavas from the Garrett Transform Fault: Insights into mantle heterogeneity beneath the eastern Pacific. Earth and Planetary Science Letters 173, 271–284. https://doi.org/10.1016/S0012-821X(99)00236-8.

top

Results

Sample ARP1973-010-003 is a porphyritic, picritic basalt from the FAMOUS segment (36° 50' N) on the mid-Atlantic ridge, with abundant (>20 %) olivine megacrysts (Fo_89-91) in a glassy to crystalline groundmass. Chromium spinel and sulfide occur as inclusions in olivine and groundmass. Spinel is thought to be a relict high pressure phase (Sigurdsson and Schilling, 1976

Sigurdsson, H., Schilling, J.G. (1976) Spinels in Mid-Atlantic Ridge basalts: chemistry and occurrence. Earth and Planetary Science Letters 29, 7–20. https://doi.org/10.1016/0012-821X(76)90021-2

), and volatile contents of melt inclusions in olivine indicate crystallisation in the lower crust (Laubier, pers. comm). Melt inclusions themselves possess both depleted and enriched trace element compositions (Laubier et al., 2012

Laubier, M., Gale, A., Langmuir, C.H. (2012) Melting and crustal processes at the FAMOUS segment (Mid-Atlantic Ridge): New insights from olivine-hosted melt inclusions from multiple samples. Journal of Petrology 53, 665–698. https://doi.org/10.1093/petrology/egr075

). The Pb isotope composition of the glass and crystalline matrix (blue circles) is in the range of other enriched MORB samples typical of this ridge segment (green circles) (Fig. 1). However, spinel, sulfide and melt inclusions in olivine (an aggregate of hand picked inclusions) possess Pb isotope compositions that are substantially less radiogenic than the host glass, indicating that these phases crystallised from a depleted melt in the lower crust. The question then is to what extent these unradiogenic Pb compositions are typical in the lower crust?

Full size image

Plagioclase megacrysts from samples from FAMOUS, Garrett fracture zone, South Atlantic and Iceland, all possess Pb isotope compositions that are less radiogenic than their host lava (Table 1, Fig. 2a,b). For plagioclase from the North Atlantic substantially so, but even for the Garrett fracture zone lavas (amongst the least radiogenic MORB globally) the Pb isotope composition of plagioclase is less radiogenic than the host lava, and close to the composition expected for the bulk silicate Earth.

Full size image

top

Pb Isotopes in Oceanic Basalts

The results here are consistent with previously reported Pb isotope data for plagioclase and host lava from a single sample on the Mid-Atlantic Ridge (Bryce and DePaolo, 2004

Bryce, J.G., DePaolo, D.J. (2004) Pb isotopic heterogeneity in basaltic phenocrysts. Geochimica et Cosmochimica Acta 68, 4453–4468. https://doi.org/10.1016/j.gca.2004.01.016

). Although ²⁰⁴Pb data are not reported in that study, differences in ²⁰⁶Pb/²⁰⁷Pb and ²³⁸U/²⁰⁷Pb are consistent with the plagioclase having crystallised from a melt with a low U/Pb ratio (relative to the host lava) (Bryce and DePaolo, 2004

Bryce, J.G., DePaolo, D.J. (2004) Pb isotopic heterogeneity in basaltic phenocrysts. Geochimica et Cosmochimica Acta 68, 4453–4468. https://doi.org/10.1016/j.gca.2004.01.016

). Likewise, strontium isotope data for plagioclase crystals in MORB shows that the plagioclase often possesses a less radiogenic ⁸⁷Sr/⁸⁶Sr isotope composition relative to host glass (Lange et al., 2013

Lange, A.E. Nielsen, R.L., Tepley III, F.J., Kent, A.J.R. (2013) The petrogenesis of plagioclase-phyric basalts at mid-ocean ridges. Geochemistry, Geophysics, Geosystems 14, 3282–3296. https://doi.org/10.1002/ggge.20207

). This indicates that the plagioclase crystallised from a melt with a mantle source with a low Rb/Sr ratio, and therefore, like Pb, a source that has experienced ancient depletion of incompatible elements. Taken together, the results here, and elsewhere, indicate that both plagioclase and olivine megacrysts crystallise from melts sourced by depleted mantle material with an unradiogenic Pb isotope composition that is potentially widespread in the upper mantle and lower oceanic crust.

top

The Pb Isotope Composition of the Silicate Earth and Ocean Crust Formation

Our data indicate that both plagioclase and olivine megacrysts in oceanic basalts commonly crystallise from melts sourced by depleted mantle material with an unradiogenic Pb isotope composition that is complementary to the radiogenic Pb isotope compositions seen in erupted basalts. In the past, the radiogenic Pb isotope composition of many oceanic basalts, the so called ‘Pb paradox’, has been attributed to segregation of Pb into Earth’s metallic core (e.g., Allègre, 1969

Allègre, C.J. (1969) Comportement des systemes U-Th-Pb dans le manteau superieur et modele d’evolution de ce dernier au cours des temps geologiques. Earth and Planetary Science Letters 5, 261–269. https://doi.org/10.1016/S0012-821X(68)80050-0

) or loss as a volatile element (Albarède, 2009

Albarède, F. (2009) Volatile accretion history of the terrestrial planets and dynamic implications. Nature 461, 1227–1233. https://doi.org/10.1038/nature08477

). While the results here do not preclude Pb removal to the core, or volatile loss, they do show that a substantial proportion of Earth’s unradiogenic Pb is potentially present in the mantle, and that Pb is simply not sampled by erupted basalts.

Spatial variations in radiogenic isotopes through the oceanic crust potentially reveal how melts are delivered to the surface, and the nature of oceanic crust formation. From an elemental and isotopic perspective it is generally thought that mantle melts from diverse sources (both incompatible element depleted and enriched) are delivered to the base of the crust, from where mixing and crystallisation during ascent act to reduce the primary chemical variability (e.g., Maclennan, 2008

Maclennan, J. (2008) Concurrent mixing and cooling of melts under Iceland. Journal of Petrology 49, 1931–1953. https://doi.org/10.1093/petrology/egn052

; Lambart et al., 2019

Lambart, S., Koornneef, J.M., Millet, M.A., Davies, G.R., Cook, M., Lissenberg, C.J. (2019) Highly heterogeneous depleted mantle recorded in the lower oceanic crust. Nature Geoscience 12, 482–486. https://doi.org/10.1002/2015JB012762.

). Thus, the actual extent of isotope heterogeneity in the mantle is lost in the final melts erupted as MORB, the composition of which represents some average of those different mantle melts. However, if melts from depleted mantle rocks tend to crystallise in the lower crust, then that crust may, at least in some circumstances, be isotopically stratified. With MORB and gabbro from the shallowest part of the oceanic crust being derived from enriched mantle sources, while intrusive rocks deeper in the crust originate from depleted mantle lithologies. In this case, not only are the melts delivered to the oceanic crust derived from isotopically distinct sources, but those melts can also traverse the crust while maintaining a high degree of chemical integrity.

The question remains as to what extent depleted melts with unradiogenic Pb isotope compositions contribute to the formation of the oceanic crust, rather than just the deep roots of oceanic magmatism in the mantle. Indeed, are the samples here more widely representative? Gabbroic plagioclase from the upper ∼500 m of oceanic crust at the Atlantis Bank on the Southwest Indian Ridge (ODP Hole 735B) preserves a relatively uniform Pb isotope composition (Kempton et al., 1991

Kempton, P.D., Hawkesworth, C.J., Fowler, M. (1991) Geochemistry and isotopic composition of gabbros from layer 3 of the Indian Ocean crust, Hole 735B. In: Von Herzen, R.P., Robinson, P.T., et al. (Eds.) Proceedings of the Ocean Drilling Program, Scientific Results 118, Ocean Drilling Program, College Station, TX, 127–143. https://doi.org/10.2973/odp.proc.sr.118.118.1991

) that is indistinguishable from nearby MORB. The work of Holm et al. (2002)

Holm, P.M. (2002) Sr, Nd and Pb isotopic composition of in situ lower crust at the Southwest Indian Ridge: results from ODP Leg 176. Chemical Geology 184, 195–216. https://doi.org/10.1016/S0009-2541(01)00364-3

however, suggests that, at greater depths in this crustal section, gabbroic plagioclase possesses a systematically less radiogenic Pb isotope composition, consistent with the observations here. The presence of plagioclase-bearing depleted gabbros in the lower crust or mantle might also explain, at least in part, the so called “ghost plagioclase” signature seen in many oceanic basalts (e.g., Sobolev et al., 2000

Sobolev, A.V., Hofmann, A.W., Nikogosian, I.K. (2000) Recycled oceanic crust observed in ‘ghost plagioclase’ within the source of Mauna Loa lavas. Nature 404, 986–990. https://doi.org/10.1038/35010098

). Basalts and melt inclusions characterised by Sr, Ba and Eu anomalies, have been attributed by some, to the presence of plagioclase-rich cumulates in the oceanic crust, through which these melts have percolated (e.g., Maclennan, 2008

Maclennan, J. (2008) Concurrent mixing and cooling of melts under Iceland. Journal of Petrology 49, 1931–1953. https://doi.org/10.1093/petrology/egn052

). Finally, some erupted MORB do show unradiogenic Pb isotope compositions akin to those seen here (e.g., Wendt et al., 1999

Wendt, J.I., Regelous, M., Niu, Y., Hékinian, R., Collerson, K.D. (1999) Geochemistry of lavas from the Garrett Transform Fault: Insights into mantle heterogeneity beneath the eastern Pacific. Earth and Planetary Science Letters 173, 271–284. https://doi.org/10.1016/S0012-821X(99)00236-8

), attributed to the preferential melting of depleted mantle stripped of enriched rock types by previous melting. Elsewhere, however, this unradiogenic signal appears to be lost either by mixing during ascent through the crust, or else by the tendency of melts from depleted sources to crystallise in the lower crust. In either case the actual composition of the depleted mantle is concealed.

top

Acknowledgements

We would like to thank Frances Jenner and Al Hofmann for their thorough, thoughtful and constructive reviews, likewise the feedback and editorial handling of Horst Marschall.

Editor: Horst Marschall

top

References

top

Supplementary Information

The Supplementary Information includes:

• Mineral and Glass Major Element Compositions
• Sample Preparation for Pb Isotope Chemistry
• Pb Isotope Chemistry and Mass Spectrometry
• Table S-1
• Supplementary Information References

Download the Supplementary Information (PDF)