METEX: An expert system for metamorphic petrography

Classification of metamorphic rocks is normally carried out using a poorly defined, subjective classification scheme making this an area in which many undergraduate geologists experience difficulties. An expert system to assist in such classification is presented which is capable of classifying rocks and also giving further details about a particular rock type. A mixed knowledge representation is used with frame, semantic and production rule systems available. Classification in the domain requires that different facets of a rock be classified. To implement this, rocks are represented by 'context' frames with slots representing each facet. Slots are satisfied by calling a pre-defined ruleset to carry out the necessary inference. The inference is handled by an interpreter which uses a dependency graph representation for the propagation of evidence. Uncertainty is handled by the system using a combination of the MYCIN certainty factor system and the Dempster-Shafer range mechanism. This allows for positive and negative reasoning, with rules capable of representing necessity and sufficiency of evidence, whilst also allowing the implementation of an alpha-beta pruning algorithm to guide question selection during inference. The system also utilizes a semantic net type structure to allow the expert to encode simple relationships between terms enabling rules to be written with a sensible level of abstraction. Using frames to represent rock types where subclassification is possible allows the knowledge base to be built in a modular fashion with subclassification frames only defined once the higher level of classification is functioning. Rulesets can similarly be added in modular fashion with the individual rules being essentially declarative allowing for simple updating and maintenance. The knowledge base so far developed for metamorphic classification serves to demonstrate the performance of the interpreter design whilst also moving some way towards providing a useful assistant to the non-expert metamorphic petrologist. The system demonstrates the possibilities for a fully developed knowledge base to handle the classification of igneous, sedimentary and metamorphic rocks. The current knowledge base and interpreter have been evaluated by potential users and experts. The results of the evaluation show that the system performs to an acceptable level and should be of use as a tool for both undergraduates and researchers from outside the metamorphic petrography field. .

Isotopic studies of the Guerrero composite terrane, west-central Mexico: Implications for provenance of crustal rocks and genesis of ore metals

A variety of world-class mineral deposits occur in Mesozoic and Tertiary rocks of the Guerrero terrane. New Pb isotope analyses of various crustal units and ores from distinct subterranes of the Guerrero terrane are presented to trace metal sources in these deposits and infer source reservoirs. New Sr and Nd isotope results are provided to gain insight into the provenance of the crustal rocks from the Guerrero terrane. Triassic schist samples from the Arteaga Complex and Triassic-Jurassic phyllite and slate samples from the Tejupilco metamorphic suite contain radiogenic Pb (206Pb/204Pb = 18.701–19.256) relative to bulk earth models. Cretaceous sedimentary rocks of the Zihuatanejo Sequence are more radiogenic (206Pb/204Pb = 18.763–19.437) than samples from the Huetamo Sequence (206Pb/204Pb = 18.630–18.998). Tertiary intrusive rocks from La Verde, Inguaran, La Esmeralda, and El Malacate plot to the right of the average Pb crust evolution curve of Stacey and Kramers (206Pb/204Pb = 18.705–19.033). Ores from the La Verde and La Esmeralda porphyry copper deposits yield isotopic ratios (206Pb/204Pb = 18.678–18.723) that are generally less radiogenic than the host igneous rocks, but plot within the field defined by the sedimentary rocks from the Huetamo Sequence. Tertiary intrusive rocks from the Zimapan and La Negra districts in the Sierra Madre terrane plot above and to the right of the Stacey-Kramers reference line (206Pb/204Pb = 18.804–18.972). Lead isotope ratios of ore minerals from the Zimapan and La Negra skarn mines ( 206Pb/204Pb = 18.775–18.975) resemble those of the associated igneous rocks, implying a magmatic Pb input in the skarn deposits. New Sr and Nd isotope data on metamorphic rocks (87Sr/ 86Sr = 0.707757–0.726494 and 143Nd/144 Nd = 0.512109–0.512653) suggest that the basement of the Guerrero terrane originated from sources that had been derived from an old cratonic area. The narrow ranges and generally low 87Sr/86Sr ratios (0.704860–0.705755) and 143Nd/144Nd values (0.512765–0.512772) above that of bulk earth for igneous rocks from Inguaran, El Malacate, and La Esmeralda suggest a relatively low degree of crustal contamination. However, the isotopic values for the La Verde site (87Sr/86Sr = 0.708784 and 143Nd/144Nd = 0.512640) may indicate the involvement of a more evolved crustal component.

Metamorphic investigation of mid-P aluminous gneisses from the central Grenville Province

This is a metamorphic study of mid-P anatectic aluminous gneisses from the Manicouagan and lac du Milieu areas of the central Grenville Province. The rocks are derived from hydrothermally altered felsic protoliths and were metamorphosed at granulite facies conditions during the Grenvillian orogeny. The samples come from three locations separated by several tens of kilometers and exhibit a wide range of textures and bulk compositions. However, they all have the same peak mineral assemblage: garnet + biotite + quartz + K-feldspar +/- plagioclase +/- sillimanite with retrograde cordierite in some, and show evidence of partial melting and melt loss. In terms of mineralogy and bulk composition, the samples were divided into two groups, sillimanite-rich and sillimanite-poor, with a high and low Alumina index in the AFM space, respectively. Phase equilibria modeling in the Na₂O–CaO–K₂O–FeO–MgO–Al₂O₃–SiO₂–H₂O– TiO₂–O (NCKFMASTHO) system using Thermocalc constrained the P–T field of the peak mineral assemblage at 800–900ºC and 6–11kbar, with melt solidification in the range of 800–865ºC and 6–8kbar. The presence of sillimanite inclusions in garnet, and of only scarce, retrograde cordierite, is consistent with moderate dP/dT gradient ‘hairpin’ P– T paths, which were similar between the three locations. This study also investigated the role of Fe3+ on phase stability in mid-P aluminous systems. Fe³⁺ is problematic because although it is incorporated in the NaCKFMASTHO system, it is rarely measured in modeled minerals and rocks and its value is generally assumed. Biotite may contain significant amounts of Fe³⁺, and these were analysed by Mössbauer spectroscopy in selected samples, where they were found to be low (0-4%). In addition, the effect of increasing the bulk Fe³⁺ in the mid-P portion of phase diagrams was modeled. This increase added new minor phases and changed the phase proportions, as well as shifted phase boundaries to a small degree, but P–T paths remained largely unaffected. Finally, the two methods commonly used in phase equilibria modeling to account for melt loss were compared. In some cases there were major differences in the topologies between the ‘melt reintegration’ and ‘adding water’ methods, but the former method is the most consistent with the rock data, and should be the method of choice.

Representative analyses of garnet, plagioclase, orthopyroxene, ilmenite, hornblende and biotite for pre-tectonic mafic and metapelitic rocks in Gjelsvikfjella

A metamorphic petrological study, in conjunction with recent precise geochronometric data, revealed a complex P-T-t path for high-grade gneisses in a hitherto poorly understood sector of the Mesoproterozoic Maud Belt in East Antarctica. The Maud Belt is an extensive high-grade, polydeformed, metamorphic belt, which records two significant tectono-thermal episodes, once towards the end of the Mesoproterozoic and again towards the late Neoproterozoic/Cambrian. In contrast to previous models, most of the metamorphic mineral assemblages are related to a Pan-African tectono-thermal overprint, with only very few relics of late Mesoproterozoic granulite-facies mineral assemblages (M1) left in strain-protected domains. Petrological and mineral chemical evidence indicates a clockwise P-T-t path for the Pan-African orogeny. Peak metamorphic (M2b) conditions recorded by most rocks in the area (T = 709-785 °C and P = 7.0-9.5 kbar) during the Pan-African orogeny were attained subsequent to decompression from probably eclogite-facies metamorphic conditions (M2a). The new data acquired in this study, together with recent geochronological and geochemical data, permit the development of a geodynamic model for the Maud Belt that involves volcanic arc formation during the late Mesoproterozoic followed by extension at 1100 Ma and subsequent high-grade tectono-thermal reworking once during continent-continent collision at the end of the Mesoproterozoic (M1; 1090-1030 Ma) and again during the Pan-African orogeny (M2a, M2b) between 565 and 530 Ma. Post-peak metamorphic K-metasomatism under amphibolite-facies conditions (M2c) followed and is ascribed to post-orogenic bimodal magmatism between 500 and 480 Ma.

Chemical and mineral composition of rocks from the Philippine Sea

The book is devoted to geology of the Philippine Sea floor. This region is studied most extensively among other marginal seas of the Pacific Ocean. Rocks of the sedimentary and basalt layers within this sea have been studied during five legs of D/S Glomar Challenger. International geological expedition on board R/V Dmitry Mendeleev carried out according to the Project ''Ophiolites of Continents and Comparable Rocks of the Ocean Floor''obtained unique collection of rocks from the second and third layers of the ocean crust in the Philippine Sea. The book provides detailed petrographic and geochemical description of igneous and sedimentary formations from the Philippine Sea and compares them with rocks of the continental ophiolite association. An analysis of structure and history of the ocean crust formation in the region is based on all known geological information. The main periods of tectonic movement activation and nature of their manifestations within the sea are shown.

Composition of sedimentary rocks from the Burun-Tas Formation, Bol'shoi Lyakhov Island (New Siberian Islands)

Graywackes and shales of the Bol'shoi Lyakhov Island originally attributed to Mesozoic were subsequently considered based on microfossils as Late Proterozoic in age. At present, these sediments in the greater part of the island are dated back to Permian based on palynological assemblages. In the examined area of the island, this siliciclastic complex is intensely deformed and tectonically juxtaposed with blocks of oceanic and island-arc rocks exhumed along the South Anyui suture. The complex is largely composed of turbidites with members displaying hummocky cross-stratification. Studied mineral and geochemical charac¬teristics of the rocks defined three provenances of clastic material: volcanic island arc, sedimentary cover and/or basement of an ancient platform, and exotic blocks of oceanic and island-arc rocks such as serpentinites and amphibolites. All rock associations represent elements of an orogenic structure that originated by collision of the New Siberian continental block with the Anyui-Svyatoi Nos island arc. Flyschoid sediments accumu¬lated in a foredeep in front of the latter structure in the course of collision. Late Jurassic volcanics belonging to the Anyui-Svyatoi Nos island arc determine the lower age limit of syncollision siliciclastic rocks. Presence of Late Jurassic zircons in sandstones of the flyschoid sequence in the Bol'shoi Lyakhov Island is confirmed by fission-track dating. The upper age limit is determined by Aptian-Albian postcollision granites and diorites intruding the siliciclastic complex. Consequently, the flyschoid sequence is within stratigraphic range from the terminal Late Jurassic to Neocomian. It appears that Permian age of sediments suggested earlier is based on redeposited organic remains. The same Late Jurassic-Neocomian age and lithology are characteristic of fossiliferous siliciclastic sequences of the Stolbovoi and Malyi Lyakhov islands, the New Siberian Archipelago, and of graywackes in the South Anyui area in the Chukchi Peninsula. All these sediments accumulated in a spacious foredeep that formed in the course the late Cimmerian orogeny along the southern margin of the Arctic conti¬nental block.

Age derminations of rocks from the Read Mountains, Antarctica

The Shackleton Range can be divided into three major units: (1) The East Antarctic Craton and its sedimentary cover (Read Group and Watts Needle Formation), (2) the allochthonous Mount Wegener Nappe (Mount Wegener Formation, Stephenson Bastion Formation, and Wyeth Heights Formation), and (3) the northern belt (basement: Pioneer and Stratton Groups, sedimentary cover: Haskard Highlands Formation (allochthonous?), and Blaiklock Glacier Group). The northern units are thrust over the southern ones. The thrusting is related to the Ross Orogeny. The Mount Wegener Nappe, which appears to be a homogeneous tectonic unit, consists of a Precambrian basement (Stephenson Bastion Formation, Wyeth Heights Formation?) and a Cambrian cover (Mount Wegener Formation). Some questions are still open for discussion: the position of the Haskard Highlands Formation (trilobite shales) may be erratic or represent a tectonic sliver, the relation of the former Turnpike Bluff Group, the origin of the crystalline basement west of Stephenson Bastion and others.

(Table) Argon isotope ratios and argon ages of rocks obtained from the Chernorud zone, western Baikal area

Structural-petrologic and isotopic-geochronologic data on magmatic, metamorphic, and metasomatic rocks from the Chernorud zone were used to reproduce the multistage history of their exhumation to upper crustal levels. The process is subdivided into four discrete stages, which corresponded to metamorphism to the granulite facies (500-490 Ma), metamorphism to the amphibolite facies (470-460 Ma), metamorphism to at least the epidote-amphibolite facies (440-430 Ma), and postmetamorphic events (410-400 Ma). The earliest two stages likely corresponded to the tectonic stacking of the backarc basin in response to the collision of the Siberian continent with the Eravninskaya island arc or the Barguzin microcontinent, a process that ended with the extensive generation of synmetamorphic granites. During the third and fourth stages, the granulites of the Chernorud nappe were successively exposed during intense tectonic motions along large deformation zones (Primorskii fault, collision lineament, and Orso Complex). The comparison of the histories of active thermal events for Early Caledonian folded structures in the Central Asian Foldbelt indicates that active thermal events of equal duration are reconstructed for the following five widely spiced accretion-collision structures: the Chernorud granulite zone in the Ol'khon territory, the Slyudyanka crystalline complex in the southwestern Baikal area, the western Sangilen territory in southeastern Tuva, Derbinskii terrane in the Eastern Sayan, and the Bayankhongor ophiolite zone in central Mongolia. The dates obtained by various isotopic techniques are generally consistent with the four discrete stages identified in the Chernorud nappe, whereas the dates corresponding to the island-arc evolutionary stage were obtained only for the western Sangilen and Bayankhongor ophiolite zone.

Table 1: U-Pb analyses of zircons and monazites of paragneisses which belong to the high grade metamorphic basement of Wilson Terrane

Samples of high grade metamorphic basement rocks of Wilson Terrane cropping out in the Deep Freeze Range and on Kay Island were collected during GANOVEX VI to study their isotopic evolution. The age and origin of granulite facies gneisses and of their migmatite host rocks are especially of interest for the interpretation of the geological and tectonic development of North Victoria Land. Another important research aspect is the influence of the polyphase metamorphic evolution on the isotopic systems of whole rocks and minerals like zircon, garnet, orthopyroxene, amphibole and feldspar.

U-Pb SHRIMP zircon dating of high-grade rocks from the Upper Allochthonous Terrane of Bragança and Morais Massifs (NE Portugal); geodynamic consequences

Bragança and Morais Massifs are part of the mega-klippen ensemble of NW Iberia, comprising a tectonic pile of four allochthonous units stacked above the Central-Iberian Zone autochthon. On top of this pile, the Upper Allochthonous Terrane (UAT) includes different high-grade metamorphic series whose age and geodynamic meaning are controversial. Mafic granulites provided U–Pb zircon ages at 399±7 Ma, dating the Variscan emplacement of UAT. In contrast,U–Pb zircon ages of ky- and hb-eclogites, felsic/intermediate HP/HT-granulites and orthogneisses (ca. 500–480 Ma) are identical to those of gabbros (488 ± 10 Ma) and Grt-pyroxenites (495 ± 8 Ma) belonging to a mafic/ultramafic igneous suite that records upper mantle melting and mafic magma crustal underplating at these times. Gabbros intrude the high-grade units of UAT and did not underwent the HP metamorphic event experienced by eclogites and granulites. These features and the zircon dates resemblance among different lithologies, suggest that extensive age resetting of older events may have been correlative with the igneous suite emplacement/crystallisation. Accordingly, reconciliation of structural, petrological and geochronological evidence implies that the development and early deformation of UAT high-grade rocks should be ascribed to an orogenic cycle prior to ≈500 Ma. Undisputable dating of this cycle is impossible, but the sporadic vestiges of Cadomian ages cannot be disregarded. The ca. 500–480 Ma time-window harmonises well with the Lower Palaeozoic continental rifting that trace the VariscanWilson Cycle onset and the Rheic Ocean opening. Subsequent preservation of the high heat-flowregime, possibly related to the Palaeotethys back-arc basin development (ca. 450–420 Ma), would explain the 461 ± 10 Ma age yielded by some zircon domains in felsic granulites, conceivably reflecting zircon dissolution/ recrystallisation till Ordovician times, long before the Variscan paroxysm (ca. 400–390 Ma). This geodynamic scenario suggests also that UAT should have been part of Armorica before its emplacement on top of Iberia after Palaeotethys closure.

Grain coarsening maps: A new tool to predict microfabric evolution of polymineralic rocks

Polymineralic rocks undergo grain coarsening with increasing temperature in both static and deformational environments, as long as no mineral reactions occur. The grain coarsening in such rocks is complex because the different phases influence each other, and it is this interaction that controls the rate of grain coarsening of the entire aggregate. We present a mathematical approach to investigate coupled grain coarsening using a set of microstructural parameters, including grain size and volume fraction of both second phases and matrix mineral in combination with temperature information. Based on samples from polymineralic carbonate mylonites that were deformed at different temperatures, we demonstrate how the mathematical relation can be calibrated for this natural system. Using such data sets for other lithologies, grain coarsening maps can be generated, which allow the prediction of microstructural evolution in polymineralic rocks. Such predictions are crucial for all subdisciplines in the earth sciences that require fundamental knowledge about microstructural changes and rheology of an orogen at different depths, such as structural geology, geophysics, geodynamics, and metamorphic petrology.

Grain coarsening in contact metamorphic carbonates: effects of second-phase particles, fluid flow and thermal perturbations

Under contact metamorphic conditions, carbonate rocks in the direct vicinity of the Adamello pluton reflect a temperature-induced grain coarsening. Despite this large-scale trend, a considerable grain size scatter occurs on the outcrop-scale indicating local influence of second-order effects such as thermal perturbations, fluid flow and second-phase particles. Second-phase particles, whose sizes range from nano- to the micron-scale, induce the most pronounced data scatter resulting in grain sizes too small by up to a factor of 10, compared with theoretical grain growth in a pure system. Such values are restricted to relatively impure samples consisting of up to 10 vol.% micron-scale second-phase particles, or to samples containing a large number of nano-scale particles. The obtained data set suggests that the second phases induce a temperature-controlled reduction on calcite grain growth. The mean calcite grain size can therefore be expressed in the form D 1⁄4 C2 eQ*/RT(dp/fp)m*, where C2 is a constant, Q* is an activation energy, T the temperature and m* the exponent of the ratio dp/fp, i.e. of the average size of the second phases divided by their volume fraction. However, more data are needed to obtain reliable values for C2 and Q*. Besides variations in the average grain size, the presence of second-phase particles generates crystal size distribution (CSD) shapes characterized by lognormal distributions, which differ from the Gaussian-type distributions of the pure samples. In contrast, fluid-enhanced grain growth does not change the shape of the CSDs, but due to enhanced transport properties, the average grain sizes increase by a factor of 2 and the variance of the distribution increases. Stable d18O and d13C isotope ratios in fluid-affected zones only deviate slightly from the host rock values, suggesting low fluid/rock ratios. Grain growth modelling indicates that the fluid-induced grain size variations can develop within several ka. As inferred from a combination of thermal and grain growth modelling, dykes with widths of up to 1 m have only a restricted influence on grain size deviations smaller than a factor of 1.1.To summarize, considerable grain size variations of up to one order of magnitude can locally result from second-order effects. Such effects require special attention when comparing experimentally derived grain growth kinetics with field studies.

Electron microprobe investigations of metamorphic reactions and mineral growth histories, Kwoiek area, British Columbia

The Kwoiek Area of British Columbia contains a pendant or screen of metamorphosed sedimentary and volcanic rocks almost entirely surrounded by a portion of the Coast Range Batholith, and intruded by several dozen stocks. The major metamorphic effects were produced by the quartz diorite batholithic rocks, with minor and later effects by the quartz diorite stocks. The sequence of important metamorphic reactions in the metasedimentary and metavolcanic rocks, ranging in grade from chlorite to sillimanite, is:

1. chlorite + carbonate + muscovite → epidote + biotite

2. chlorite + carbonate → actinolite + epidote

3. chlorite + muscovite → garnet + biotite

4. chlorite + epidote → garnet + hornblende

5. chlorite + muscovite → garnet + staurolite + biotite

6. chlorite + muscovite → aluminum silicate + biotite

7. muscovite + staurolite → garnet + aluminum silicate + biotite

8. staurolite → garnet + aluminum silicate

Continuous reactions, occurring between reactions 5 and 7, are:

A. chlorite + (high Ti) biotite + Al₂O₃ (from plagioclase?)→ garnet + staurolite + (low Ti) biotite + O₂

B. muscovite (phengitic) → garnet + staurolite +muscovite (less phengitic) + O₂ (?)

Detailed electron microprobe work on garnet, staurolite, biotite, and chlorite shows that:

(1) The garnet porphyroblasts are zoned according to a depletion model, called the Rayleigh depletion model, which assumes equilibrium between the edge of a growing garnet and the minerals which are unzoned, notably biotite, chlorite, and muscovite, but which assumes disequilibrium within the garnet.

(2) The staurolite porphyroblasts are also zoned, and from their zoning patterns reactions A, B, and 5 are documented. Progressive reduction of iron with increasing grade of metamorphism is also inferred from the staurolite zoning patterns.

(3) During a late period of falling temperature garnet continued to grow and the biotite and chlorite reequilibrated. The biotite, chlorite, and garnet edge compositions can vary from point to point in a given thin section, indicating that the volume of equilibrium at the final stage of metamorphism was only a few cubic microns.

(4) The horizon within the garnet that grew at maximum temperature can be identified. The Mg/Fe ratio of this horizon, if the garnet composition is a limiting composition in the Al₂O₃ - K₂O - FeO - MgO tetrahedron, increases systematically with increasing metamorphic grade. Biotite and chlorite compositions also show a general increase in Mg/Fe ratio with increasing metamorphic grade, but staurolite appears to show the reverse effect.

(5) The Mg/Fe ratio at the maximum temperature horizon of the garnet porphyroblasts is a function of its Mn content as evidenced from the study of five garnet-bearing rocks, collected from one outcrop area, with the same assemblage but with differing proportions of minerals.

An important implication of zoned minerals is that the effective composition of a system in a phase lies on the join between the homogeneous minerals (if there are two) and not within three-or- four-phase fields when a zoned mineral, such as garnet or staurolite, is present in the assemblage.

Study of the three aluminum silicates found in the Kwoiek Area showed that a constant pressure change in polymorphs from andalusite to kyanite to sillimanite took place with increasing temperature. This transition series is best explained by the metastable formation of andalusite.

Photographic materials on pages 15, 121, 160, 162, and 164 are essential and will not reproduce clearly on Xerox copies. Photographic copies should be ordered.

In search of a hidden long-term isolated sub-chondritic 142Nd/144Nd reservoir in the deep mantle : implications for the Nd isotope systematics of the Earth

Here we search for evidence of the existence of a sub-chondritic 142Nd/144Nd reservoir that balances the Nd isotope chemistry of the Earth relative to chondrites. If present, it may reside in the source region of deeply sourced mantle plume material. We suggest that lavas from Hawai’i with coupled elevations in 186Os/188Os and 187Os/188Os, from Iceland that represent mixing of upper mantle and lower mantle components, and from Gough with sub-chondritic 143Nd/144Nd and high 207Pb/206Pb, are favorable samples that could reflect mantle sources that have interacted with an Early-Enriched Reservoir (EER) with sub-chondritic 142Nd/144Nd. High-precision Nd isotope analyses of basalts from Hawai’i, Iceland and Gough demonstrate no discernable 142Nd/144Nd deviation from terrestrial standards. These data are consistent with previous high-precision Nd isotope analysis of recent mantle-derived samples and demonstrate that no mantle-derived material to date provides evidence for the existence of an EER in the mantle. We then evaluate mass balance in the Earth with respect to both 142Nd/144Nd and 143Nd/144Nd. The Nd isotope systematics of EERs are modeled for different sizes and timing of formation relative to ε143Nd estimates of the reservoirs in the μ142Nd = 0 Earth, where μ142Nd is ((measured 142Nd/144Nd/terrestrial standard 142Nd/144Nd)−1 * 10−6) and the μ142Nd = 0 Earth is the proportion of the silicate Earth with 142Nd/144Nd indistinguishable from the terrestrial standard. The models indicate that it is not possible to balance the Earth with respect to both 142Nd/144Nd and 143Nd/144Nd unless the μ142Nd = 0 Earth has a ε143Nd within error of the present-day Depleted Mid-ocean ridge basalt Mantle source (DMM). The 4567 Myr age 142Nd–143Nd isochron for the Earth intersects μ142Nd = 0 at ε143Nd of +8 ± 2 providing a minimum ε143Nd for the μ142Nd = 0 Earth. The high ε143Nd of the μ142Nd = 0 Earth is confirmed by the Nd isotope systematics of Archean mantle-derived rocks that consistently have positive ε143Nd. If the EER formed early after solar system formation (0–70 Ma) continental crust and DMM can be complementary reservoirs with respect to Nd isotopes, with no requirement for significant additional reservoirs. If the EER formed after 70 Ma then the μ142Nd = 0 Earth must have a bulk ε143Nd more radiogenic than DMM and additional high ε143Nd material is required to balance the Nd isotope systematics of the Earth.