Compositional evidence regarding the influx of interplanetary materials onto the lunar surface

Siderophilic element/Ir ratios are higher in mature lunar soils from highlands sites than in those from mare sites. We infer that the population of materials responsible for the early intense bombardment of the Moon had high ratios, and that the population responsible for the essentially constant flux has low ratios. No group of chondrites has siderophile/Ir ratios identical to those in the mare or highlands soils; CM chondrites are the most similar, and CM-like materials may account for a major fraction of Earth-crossing materials during the past 3.7 b.y.Siderophile/Ir ratios may be used to determine the amount of highlands regolith in soils or breccias from the mare-highlands interface areas (Apollo 15 and 17), and to infer the time of formation of highlands breccias whose sideropbiles originated in mature soils. Arguments are summarized against the viewpoint that the siderophiles in most highlands breccias originated in basin-forming projectiles. Differences in mature soil siderophile concentrations at Apollo 14 and 16 indicate a substantially greater concentration at the latter site immediately following the Imbrium event.Siderophile concentrations are used to estimate mean regolith depths at the landing sites; as relative values these are more precise than estimates based on seismic or crater observations. The longlived flux is calculated to be 2.9 g cm^–2 b.y.^–1 averaged over the past 3.7 b.y. A consideration of the relationship between mass fluence and time indicates that the mass flux decreased with a half-life of about 40 m.y. immediately following the Imbrium event.     

Procrustean science: Indigenous siderophiles in the lunar highlands,according to Delano and Ringwood

The best estimate of indigenous lunar siderophiles comes from 29 pristine lunar rocks, characterized by low siderophile abundances, plutonic textures, and high age. Delano and Ringwood's blanket rejection of these rocks, on the contention that they are impact melts, is not justified by the petrologic evidence. Contrary to their claims, gold in highland breccias is largely meteoritic and is unaffected by fumarolic volcanism, as shown by its correlation with Ir and noncorrelation with fumarolic T1 (r=0.896 and 0.272). Delano and Ringwood's approach, involving subtraction of an H-chondrite meteoritic component from highland breccias, ignores the variation of Ir/Au ratios in modern and ancient meteorites, and hence leads to spurious excesses of Au, Ni, and volatiles, and in some cases to physically meaningless, negative residuals. Their excess volatiles in highland crust relative to mare basalts disappear when the highland composition is based on pristine lunar rocks rather than under-corrected breccias. Contrary to claims by Delano and Ringwood, the Ni/Co trend in Apollo 16 samples cannot be explained by an indigenous component rich in Ni (150–200 ppm) and Co (30–45ppm); mixing lines show that much lower Ni and Co contents are required (e.g., 7 ppm each).Chondrites and lunar highland breccias show essentially parallel fractionation trends for the siderophile-element ratios Re/Ir, Au/Ir, Ni/Ir, Ni/Pd, and Os/Ir. Because the chondritic ratios were established in the solar nebula, it appears that the lunar ratios also reflect nebular processes, and have not been modified by planetary processes.Properly derived abundances for the lunar highlands show large, systematic depletions relative to terrestrial oceanic tholeiites, by the following factors: Ge 270, Re 230, Sb170, Zn150, Au60, Tl 50, Ag 48, Ni 42, Se 12. It would seem that the resemblance to the Earth's mantle is not quite as striking as claimed by Delano and Ringwood.     

Petrology,chemistry, and isotopic compositions of the lunar highland regolith breccia Dar al Gani 262

Abstract— Lunar meteorite Dar al Gani 262 (DG 262)—found in the Libyan part of the Sahara—is a mature, anorthositic regolith breccia with highland affinities. The origin from the Moon is undoubtedly indicated by its bulk chemical composition; radionuclide concentrations; noble gas, N, and O isotopic compositions; and petrographic features. Dar al Gani 262 is a typical anorthositic highland breccia similar in mineralogy and chemical composition to Queen Alexandra Range (QUE) 93069. About 52 vol% of the studied thin sections of Dar al Gani 262 consist of fine-grained(100 μm) constituents, and 48 vol% is mineral and lithic clasts and impact-melt veins. The most abundant clast types are feldspathic fine-grained to microporphyritic crystalline melt breccias (50.2 vol%; includes recrystallized melt breccias), whereas mafic crystalline melt breccias are extremely rare (1.4 vol%). Granulitic lithologies are 12.8 vol%, intragranularly recrystallized anorthosites and cataclastic anorthosites are 8.8 and 8.2 vol%, respectively, and (devitrified) glasses are 2.7 vol%. Impact-melt veins (5.5 vol% of the whole thin sections) cutting across the entire thin section were probably formed subsequent to the lithification process of the bulk rock at pressures below 20 GPa, because the bulk rock never experienced a higher peak shock pressure. Mafic crystalline melt breccias are very rare in Dar al Gani 262 and are similar in abundance to those in QUE 93069. The extremely low abundance of mafic components and the bulk composition may constrain possible areas of the Moon from which the breccia was derived. The source area of Dar al Gani 262 must be a highland terrain lacking significant mafic impact melts or mare components. On the basis of radionuclide activities, an irradiation position of DG 262 on the Moon at a depth of 55–85 g/cm³and a maximum transit time to Earth <0.15 Ma is suggested. Dar al Gani 262 contains high concentrations of solar-wind-implanted noble gases. The isotopic abundance ratio ⁴⁰Ar/³⁶Ar < 3 is characteristic of lunar soils. The terrestrial weathering of DG 262 is reflected by the occurrence of fractures filled with calcite and by high concentrations of Ca, Ba, Cs, Br, and As. There is also a large amount of terrestrial C and some N in the sample, which was released at low temperatures during stepped heating. High concentrations of Ni, Co, and Ir indicate a significant meteoritic component in the lunar surface regolith from which DG 262 was derived.     

Report of the Working Party on the classification of the lunar igneous rocks

Abstract— A report is presented for a possible revised classification of lunar igneous rocks that still uses the division of Moon rocks into mare and highland types. It subdivides the mare rocks into basalts depending on TiO₂ content and glasses depending on colour, and subdivides the highland rocks principally into KREEP basalts and into coarse‐grained igneous rocks comparable to and using terrestrial igneous rock terminology.     

On the petrology and structure of a gravitionally differentiated moon of fission origin

In a previous paper, it was shown that the basic properties and the developmental history of a gravitationally differentiated Moon of fission origin match those known for the Moon. In the first part of this report, the models of a differentiated Moon are critically reviewed based on second order considerations of some of the chemical systems used to develope the earlier models and based on new lunar data. As a result, slightly updated models are developed and the results indicate that a Moon of fission origin has a feldspar rich crust (≈70% Or_0.8Ab_5.3An_93.9 with ≈30% pyroxene and olivine) reaching an average depth of ≈65 km. A KREEP rich layer is located at the interface of the crust and the upper mantle. The upper mantle consists of peridotite (≈80% Wo₁₀En₇₀Fs₂₀ and ≈20% Fo_75–80 with ≈3% Al₂O₃ and ≈ 2% TiO₂) and reaches a depth of 300–400 km. Below 300–400 km lies a dunite (≈Fo₉₅) lower mantle. A simple model for the distribution of K, U and Th (and by inference, KREEP) in the differentiated Moon model is developed using a distribution coefficient of 0.1 for the three elements. This coefficient is derived from published data on the distribution of U in Apollo 11 basalts. The simple model successfully accounts for the observed K, U and Th contents of the various mare basalts and upland rocks and yields a heat flow of 21 erg cm^?2s^?1 for the Moon. A model for the fine structure of the peridotite upper mantle of the model Moon is developed based on the TiO₂ and trace element variations observed in the various mare basalts. It is proposed that the upper mantle is rhythmically banded on the scale of 10's of km and that this banding leads to local variations of a factor of ±3 in the K, U and Th content, _-10 ⁺⁵ in the TiO₂ content and _-∞ ⁺² in the olivine content of the peridotite. It is also proposed that this banding leads to large scale horizontal inhomogenuities in the composition of the upper mantle. It is also shown that the formation of the primitive suite of upland rocks is easily explained by the cumulation of plagioclase, which carried varying amounts of pyroxene, olivine and melt with it, during the peritectic crystallization of the last 20% of the differentiating Moon. It is found that the 100 Mg/(Mg+Fe) ratios of the mafics and the An contents of the plagioclases of the rocks are controlled by several factors, the most important of which is the ratio of melt to crystals which together formed the various upland rocks. The inverse relationship between the An contents and the Mg contents of the upland rocks is a direct consequence of the differentiation sequence proposed. The results and models presented in this paper further support the hypothesis that the Moon formed as a result of fission from the proto-Earth.     

Lithological mapping of central part of Mare Moscoviense using Chandrayaan-1 Hyperspectral Imager (HySI) data

We conducted spectral analysis of central region of the Mare Moscoviense area on the far side of the Moon using the Hyperspectral Imager (HySI) data from the Chandrayaan-1 mission in an effort to identify and map the major lithological units present in the area. Various spectral band parameters, namely, band curvature, band tilt and band strength have been used for lithological discrimination based on the nature of the spectral profile. These band parameters essentially measure the shape, position and strength of the absorption feature near 1000 nm arising due to electronic transition of Fe²⁺ in crystallographic sites of major rock forming silicates. Spectral band parameters have been used for generating rock type composite image. Based on spectral studies and rock type composite image as obtained using band parameters, five major compositional units have been identified: highland basin soils, ancient mature mare, highland contaminated mare, buried unit with abundant low-Ca pyroxene (LCP), and youngest mare unit. In the present study, a multispectral approach in the form of spectral band parameters has been adopted for analysing the HySI hyperspectral data from Chandrayaan-1 mission. Present study clearly shows that the spectral band parameters obtained using selected HySI channels could efficiently be used to discriminate and delineate the major litho-units present across the central part of Mare Moscoviense and the same approach can thus be used for lithological mapping of other parts of lunar surface using HySI data.     

Apollo 12 ropy glasses revisited

Abstract— We analyzed ropy glasses from Apollo 12 soils 12032 and 12033 by a variety of techniques including SEM/EDX, electron microprobe analysis, INAA, and ³⁹Ar-⁴⁰Ar age dating. The ropy glasses have KREEP-like compositions different from those of local Apollo 12 mare soils; it is likely that the ropy glasses are of exotic origin. Mixing calculations indicate that the ropy glasses formed from a liquid enriched in KREEP and that the ropy glass liquid also contained a significant amount of mare material. The presence of solar Ar and a trace of regolith-derived glass within the ropy glasses are evidence that the ropy glasses contain a small regolith component Anorthosite and crystalline breccia (KREEP) clasts occur in some ropy glasses. We also found within these glasses clasts of felsite (fine-grained granitic fragments) very similar in texture and composition to the larger Apollo 12 felsites, which have a ³⁹Ar-⁴⁰Ar degassing age of 800 ± 15 Ma (Bogard et al, 1992). Measurements of ³⁹Ar-⁴⁰Ar in 12032 ropy glass indicate that it was degassed at the same time as the large felsite although the ropy glass was not completely degassed. The ropy glasses and felsites, therefore, probably came from the same source. Most early investigators suggested that the Apollo 12 ropy glasses were part of the ejecta deposited at the Apollo 12 site from the Copernicus impact Our new data reinforce this model. If these ropy glasses are from Copernicus, they provide new clues to the nature of the target material at the Copernicus she, a part of the Moon that has not been sampled directly.     

“New” lunar meteorites: Impact melt and regolith breccias and large‐scale heterogeneities of the upper lunar crust

Abstract— We have analyzed nine highland lunar meteorites (lunaites) using mainly INAA. Several of these rocks are difficult to classify. Dhofar 081 is basically a fragmental breccia, but much of its groundmass features a glassy‐fluidized texture that is indicative of localized shock melting. Also, much of the matrix glass is swirly‐brown, suggesting a possible regolith derivation. We interpret Dar al Gani (DaG) 400 as an extremely immature regolith breccia consisting mainly of impact‐melt breccia clasts; we interpret Dhofar 026 as an unusually complex anorthositic impact‐melt breccia with scattered ovoid globules that formed as clasts of mafic, subophitic impact melt. The presence of mafic crystalline globules in a lunar material, even one so clearly impact‐heated, suggests that it may have originated as a regolith. Our new data and a synthesis of literature data suggest a contrast in Al₂O₃‐incompatible element systematics between impact melts from the central nearside highlands, where Apollo sampling occurred, and those from the general highland surface of the Moon. Impact melts from the general highland surface tend to have systematically lower incompatible element concentration at any given Al₂O₃ concentration than those from Apollo 16. In the case of Dhofar 026, both the bulk rock and a comparatively Al‐poor composition (14 wt% Al₂O₃, 7 μg/g Sm) extrapolated for the globules, manifest incompatible element contents well below the Apollo 16 trend. Impact melts from Luna 20 (57°E) distribute more along the general highland trend than along the Apollo 16 trend. Siderophile elements also show a distinctive composition for Apollo 16 impact melts: Ni/Ir averaging ?1.8x chondritic. In contrast, lunaite impact‐melt breccias have consistently chondritic Ni/Ir. Impact melts from Luna 20 and other Apollo sites show average Ni/Ir almost as high as those from Apollo 16. The prevalence of this distinctive Ni/Ir ratio at such widely separated nearside sites suggests that debris from one extraordinarily large impact may dominate the megaregolith siderophile component of a nearside region 2300 km or more across. Highland polymict breccia lunaites and other KREEP‐poor highland regolith samples manifest a strong anticorrelation between Al₂O₃ and mg. The magnesian component probably represents the chemical signature of the Mg‐suite of pristine nonmare rocks in its most “pure” form, unaltered by the major KREEP‐assimilation that is so common among Apollo Mg‐suite samples. The average composition of the ferroan anorthositic component is now well constrained at Al₂O₃ ?29–30 wt% (implying about 17–19 wt% modal mafic silicates), in good agreement with the composition predicted for flotation crust over a “ferroan” magma ocean (Warren 1990).     

Lunar meteorites from Oman

Abstract– Sixty named lunar meteorite stones representing about 24 falls have been found in Oman. In an area of 10.7 × 10³ km² in southern Oman, lunar meteorite areal densities average 1 g km^?2. All lunar meteorites from Oman are breccias, although two are dominated by large igneous clasts (a mare basalt and a crystalline impact‐melt breccia). Among the meteorites, the range of compositions is large: 9–32% Al₂O₃, 2.5–21.1% FeO, 0.3–38 μg g^?1 Sm, and <1 to 22.5 ng g^?1 Ir. The proportion of nonmare lunar meteorites is higher among those from Oman than those from Antarctica or Africa. Omani lunar meteorites extend the compositional range of lunar rocks as known from the Apollo collection and from lunar meteorites from other continents. Some of the feldspathic meteorites are highly magnesian (high MgO/[MgO + FeO]) compared with most similarly feldspathic Apollo rocks. Two have greater concentrations of incompatible trace elements than all but a few Apollo samples. A few have moderately high abundances of siderophile elements from impacts of iron meteorites on the Moon. All lunar meteorites from Oman are contaminated, to various degrees, with terrestrial Na, K, P, Zn, As, Se, Br, Sr, Sb, Ba, U, carbonates, or sulfates. The contamination is not so great, however, that it seriously compromises the scientific usefulness of the meteorites as samples from randomly distributed locations on the Moon.     

Lunar meteorite Queen Alexandra Range 94281: Glass compositions and other evidence for launch pairing with Yamato 793274

Abstract— Lunar meteorite Queen Alexandra Range 94281 is remarkably similar to Yamato 793274. Pairing in the conventional Earth-entry sense is difficult to reconcile with the 2500 km separation between the find locations for these two samples. Nonetheless, both of these regolith breccias are dominated by very-low-Ti (VLT) mare basalt, the pyroxenes of which feature exsolution lamellae on a remarkably coarse scale (typical lamella width = 0.5–1 μm) by mare standards. The pyroxenes also show similar compositional variations (e.g., Fe# vs. Ti# trends, which confirm parentage from VLT mare basalt). Plots using Al₂O₃ or FeO as a tracer of the highland component indicate indistinguishable internal mare-highland geochemical mixing trends. The same two distinctive glass types dominate the mare glass populations of both breccias. Glass type YQ1 features 0.37–0.63 wt% TiO₂, 10–17 wt% MgO, and 9–11 wt% Al₂O₃. Glass type YQ2 features higher TiO₂ (0.99–1.22 wt%), which is inversely correlated with MgO (12.6–13.8 wt%), and nearly constant (8.8 wt%) Al₂O₃. All of these similarities suggest that Y-793274 and QUE 94281 are a launch pair, which we designate YQ. Most of these similarities also extend to another mare-breccia meteorite, Elephant Moraine 87521. However, the EET 87521 mare basalt is unusually V-poor (～88 μg/g), whereas the YQ mare component contains ～166 μg/g. Queen Alexandra Range 94281 features a variety of textural domains. Discrete patches of dark matrix material appear to represent clods of mature regolith that have been mixed with a coarser, relatively immature material. Interior to a frothy fusion crust are areas of massive glass that probably formed as a splash coating on QUE 94281 when it was still on the Moon. The coarse YQ and EET 87521 pyroxene exsolution features imply relatively slow cooling in either a very shallow sill or an unusually thick (ponded) lava and/or later annealing within a cryptomare. Mare pyroclastic glasses, including the two YQ varieties, are systematically MgO-rich compared to crystalline mare basalts. This disparity may be a consequence of limited survival of graphite—the main fuel for explosive volcanism—during formation of the mare source regions as magma ocean cumulates. Graphite (2.2 g/cm³) survived preferentially in regions that avoided extensive early melting and thus remained MgO-rich. An apparent bimodality in the TiO₂ contents of mare volcanics, especially the pyroclastic glasses, also seems a plausible consequence of petrogenesis by remelting of magma ocean cumulates. Cumulates deposited after the magma ocean evolved to ilmenite saturation had vastly higher TiO₂ contents than cumulates deposited shortly before. The YQ regolith's subequal proportions of mare and highland matter are consistent with derivation from a terrain close to a mare-highland boundary. However, a similar mixture might also develop through vertical mixing in a cryptomare or a region of thin mare coverage. Thus, unfortunately, the YQ bulk composition is not a very useful clue to the identity of the source crater.     

Northwest Africa 482: A crystalline impact‐melt breccia from the lunar highlands

Abstract— Northwest Africa 482 (NWA 482) is a crystalline impact‐melt breccia from the Moon with highlands affinities. The recrystallized matrix and the clast population are both highly anorthositic. Clasts are all related to the ferroan anorthosite suite, and include isolated plagioclase crystals and lithic anorthosites, troctolites, and spinel troctolites. Potassium‐, rare‐earth‐element‐, and phosphorus‐bearing (KREEP) and mare lithologies are both absent, constraining the source area of this meteorite to a highland terrain with little to no KREEP component, most likely on the far side of the Moon. Glass is present in shock veins cutting through the sample and in several large melt pockets, indicating a second impact event. There are two separate events recorded in the ⁴⁰Ar‐³⁹Ar system: one at ～3750 Ma, which completely reset the K‐Ar system, and one at ?2400 Ma, which caused only partial degassing. These events could represent, respectively, crystallization of the impact‐melt breccia and later formation of the glass, or the formation of the glass and a later thermal event. The terrestrial age of the meteorite is 8.6 ± 1.3 ka. This age corresponds well with the modest amount of weathering in the rock, in the form of secondary phyllosilicates and carbonates. Based on terrestrial age and location, lithology, and chemistry, NWA 482 is unique among known lunar meteorites.     

Geochemistry and petrology of lunar meteorite Queen Alexandra Range 94281, a mixed mare and highland regolith breccia,with special emphasis on very-low-titanium mafic components

Abstract— Queen Alexandra Range (QUE) 94281, a lunar meteorite recently discovered in Antarctica, is a glassy-matrix, clast-rich regolith breccia containing a mixture of mafic, volcanic-glass and gabbroic constituents and a diverse set of highland constituents. In thin section, the clast assemblage is dominated by coarse mineral debris from a shallow intrusive or hypabyssal setting, or from deep within a thick mare flow. Abundant coarse-grained pyroxene clasts have fine-scale exsolution lamellae and compositions similar to pyroxenes of known lunar very-low-Ti (VLT) basalts and other lunar meteorites of basaltic composition. Pyroxene compositions follow Fe-enrichment extending to hedenbergite, which is associated with fayalite and cristobalite, indicating slow cooling. We refer to the protolith of the crystalline VLT component as VLT gabbro. Fragments of pyroclastic glasses that have high Fe and low Ti concentrations, similar to the pyroclastic green glasses known from Apollo samples, are common. Lithic clasts include abundant subrounded, glassy to cryptocrystalline, aluminous (～17–30 wt% Al₂O₃) KREEP-poor melt breccias of highland origin and a variety of other feldspathic impactites. On the basis of composition of our subsamples, QUE 94281 consists of ～54 wt% mafic or “mare” components and 46 wt% feldspathic or “highland” components. The bulk composition of QUE 94281 is similar to that of Yamato (Y) 793274, but QUE 94281 has slightly greater concentrations of some siderophile elements and slightly lower concentrations of those elements contributed mainly by mafic constituents. Differences in siderophile element concentrations are consistent with longer surface exposure of QUE 94281. Minor differences in trace element variations of subsamples of the two meteorites suggest subtle differences in the composition of their highland constituents. Nonetheless, the overall similarity of compositions supports the possibility that they were ejected from the same source region on the Moon. The crystalline VLT component of QUE 94281 differs from those known from Apollo 17 and Luna 24 VLT lithologies and from that of basaltic breccia Elephant Moraine (EET) 87521. The VLT-gabbro component and the ferroan VLT volcanic glasses in QUE 94281 have compositions that may be petrogenetically related by derivation from a common picritic parent composition, represented by an ultramafic glass found in QUE 94281.     

The magnetic characteristics of returned lunar samples and 5heir implications for regolith processes

Magnetic observations yield information about the amount and nature of the magnetic phases present in a sample. They reveal that the predominant magnetic phase in the lunar samples is metallic iron which is sometimes alloyed with nickel and cobalt. In the mare basalts less than 0.1% of metallic iron is present, whereas in the non-mare crystalline rocks several percent of iron has been found in some samples. The soils have approximately 0.5% of iron, which is fine grain, rather pure iron occurring in impact glass. In the recrystallized breccias and the igneous rocks the iron is coarser. Systematic minor variations in metallic iron content in the soils reveal soil maturity trends. Mixing between highland and mare soils can be traced with the Fe²⁺ content. Mare soils differ from highland soils in having a higher value of reduced remanence. The magnetic characteristics of the Apollo 14 breccias are not consistent with the progressive metamorphism of a common starting material. Shock welding in the range of tens of kbs can account for the characteristics of some of the ‘unmetamorphosed’ breccias. Greater shock accompanied by recovery can account for the magnetic characteristics of the ‘recrystallized’ breccias.     

“New” lunar meteorites: Implications for composition of the global lunar surface,lunar crust,and the bulk Moon

Abstract— New data for lunar meteorites and a synthesis of literature data have significant implications for the interpretation of global Th data and for the Moon's bulk composition. As presently calibrated (Prettyman et al. 2002), the Lunar Prospector gamma‐ray data imply that the average global surface Th = 1.58 μg/g. However, that calibration yields implausibly high concentrations for the three most Th‐poor documented sampling sites, it extrapolates to a nonzero Lunar Prospector Th, ?0.7 μg/g, at zero sample Th, and it results in a misfit toward too‐high Th when compared with the global regolith Th spectrum as constrained using mainly lunaite regolith breccias. Another problem is manifested by Th versus K systematics. Ground truth data plot consistently to the high‐Th/K side of the Prospector data trend, offset by a factor of 1.2. A new calibration is proposed that represents a compromise between the Th levels indicated by ground truth constraints and the Prettyman et al. (2002) calibration. Conservatively assuming that the Th versus K issue is mostly a K problem, the average global surface Th is estimated to be ?1.35 μg/g. The Moon's remarkable global asymmetry in KREEP abundance is even more pronounced than previously supposed. The surface Th concentration ratio between the hemisphere antipodal to the Procellarum basin and the hemisphere centered on Procellarum is reduced to 0.24 in the new calibration. This extreme disparity is most simply interpreted as a consequence of Procellarum's origin at a time when the Moon still contained at least a thin residual layer of a global magma ocean. Allowing for diminution of Th with depth, the extrapolated bulk crustal Th is ?0.73 μg/g. Further extrapolation to bulk Moon Th yields ?0.07 μg/g, which is nearly identical to the consensus estimate for Earth's primitive mantle. Assuming chondritic proportionality among refractory lithophile elements implies Al₂O₃ of approximately 3.8 wt%. The Moon's bulk mantle mg ratio is only weakly constrained by seismic and mare‐basaltic data. KREEP‐and mare‐free lunaite regolith samples, other thoroughly polymict lunar meteorites, and a few KREEP‐free Apollo highland samples manifest a remarkable anticorrelation on a plot of Al₂O₃ versus mg. This trend implies that an important component of the Moon is highly magnesian. The bulk Moon is inferred to have an Earth‐like oxide mg ratio of ?87–88 mol%. The close resemblance between the bulk Moon and Earth's primitive mantle extends to moderately volatile elements, most clearly Mn. Unless major proportions of Cr and V are sequestered into deep mantle spinel, remarkably Earth‐like depletions (versus chondrites) are also inferred for bulk Moon Cr and V.     

Relationship between positive cerium anomaly and adsorbed water in Antarctic lunar meteorites

Abstract— Concentrations of adsorbed water in single mineral grains of Antarctic lunar meteorites were determined with micro infrared (IR) spectroscopy. A relationship was found between the mineral's ability to adsorb water and the extent of Ce anomaly in rare earth element (REE) patterns precisely determined by the isotope dilution method using a thermal ionization mass spectrometer. Asuka (A) 881757, a lunar meteorite from the mare basalt without Ce anomaly, showed no trace of IR absorption due to adsorbed water. On the contrary, Yamato (Y) 791197-109, Y-86032-98, Y-86032-95, Y-791197-115 and Y-82192-55A from the lunar highland exhibiting positive Ce anomaly showed IR absorption due to adsorbed water in some of their minerals. The detected water would be of terrestrial origin, because it was not structurally bound and easy to exchange judging from the spectral band shape. The contrast in concentration of adsorbed water between the lunar highland and the mare basalt is derived from a difference in the density of microfractures in mineral grains. Average concentrations of adsorbed water in the lunar highland meteorites were 3.8 mg/cm³ for pyroxene and olivine, and 1.7 mg/cm³ for plagioclase. This contrast between minerals is noteworthy because it has been known that Ce anomaly of pyroxene and olivine is larger than that of plagioclase both for Antarctic lunar meteorites and some lunar rocks. Furthermore, more adsorbed water was detected for minerals in meteorites that exhibit larger Ce anomaly. The present observations demonstrate that the extent of Ce anomaly correlated with the concentration of adsorbed water, which suggests that active mineral surface resulting in adsorption of water could be a trace of interaction forming Ce anomaly. Terrestrial weathering on Antarctica and REE fractionation on the Moon are discussed for possible origins of the Ce anomaly.     

Energetic lunar nighttime ion events

The Rice University Suprathermal Ion Detector Experiment regularly observes ion events normally ranging from 250 eV q^?1 to 1000 eV q^?1 all through the lunar night. These ion events occur most often 2 to 3 days prior to the sunrise terminator. There is also a secondary activity peak 3 to 4 days after local sunset on the Moon. The events are normally of 4 hr or less in duration and the integral flux is 10⁶ ions cm^?2 s^?1 ster^?1. This article discusses the character of these events and presents the preliminary findings of a detailed study begun on this subject.     

Interpretation of lunar heat flow data

Lunar heat flow observations at the Apollo 15 and 17 sites can be interpreted to imply bulk U concentrations for the Moon of 5 to 8 times those of normal chondrites and 2 to 4 times terrestrial values inferred from the Earth's heat flow and the assumption of thermal steady state between surface heat flow and heat production. A simple model of nearsurface structure that takes into account the large difference in (highly insulating) regolith thickness between mare and highland provinces is considered. This model predicts atypically high local values of heat flow near the margins of mare regions - possibly a factor of 10 or so higher than the global average. A test of the proposed model using multifrequency microwave techniques appears possible wherein heat flow traverse measurements are made across mare-highland contacts. The theoretical considerations discussed here urge caution in attributing global significance to point heat-flow measurements on the Moon.     

The phase ratios of the color index: Mapping of two regions of the near side of the Moon

From the ground-based colorimetry performed for two surface regions of the near side of the Moon, images of the phase ratio of the color index C(600 nm/470 nm) have been built for the phase angles between 2° and 95°. It has been found that for phase angles smaller than α ∼ 40°–50°, the color index of the highlands grows with the phase quicker than that of the mare regions. For larger phase angles, α > 50°, a reverse situation is observed. The laboratory data on the spectrophotometry of the lunar samples confirm the peculiarities found in the phase dependence of color. The influence of multiple scattering on the phase dependence of the color of the mare and highland regions of the Moon are discussed.     

A survey of the selenochemistry of major,minor and trace elements

Average data for igneous and/or metaigneous rocks and soils from seven lunar sites are presented. There are compositional similarities between Apollo 11 and Luna 16 eastern maria, Ap 12 and 15 western maria and between Ap 16 and L 20 highlands. Subtle differences do exist between the paired mare sites and the two highland sites and striking differences between the eastern and western maria. Chondritic normalized REE (rare earth element) patterns for igneous rocks and soils from all sites range from 7-350 generally with negative Eu anomalies. Anorthositic gabbroes to anorthosites, presumably highland material, exhibit a positive Eu anomaly. The REE patterns or Sr isotopic ratios suggest two lava flows each for the L 16 and Ap 14 sites, at least four lava flows for the Ap 11 and 12 site and about six for the Ap 15 site. Paucity of lunar andesites suggests rather limited lunar chemical differentiation. Norite-KREEP is a prominent component at Ap 12, 14 and 15, less at Ap 11 and 16 and L 16 and apparently very low at the L 20 highland site. Derivation of lunar soils can be best explained using multi-component mixing systems. Characterization of meteoritic impacting bodies is also observed in addition to a steady state veil of 1.9% carbonaceous Cl like material in soils. Interelement correlations impose constraints on the primitive composition of the Moon and on magmatic processes like selective volatilization.Paper dedicated to Prof. Harold C. Urey on the occasion of his 80th birthday on 29 April 1973.     

Four lunar mare meteorites: Crystallization trends of pyroxenes and spinels

Abstract— We studied crystallization trends of pyroxene and spinel in four Antarctic meteorites known to be derived from mare regions of the Moon: Y-793169 and A-881757 (YA meteorites) are unbrecciated igneous basalts, EET 87521 is a fragmental breccia, and Y-793274 is a regolith breccia. All have relatively low bulkrock TiO₂ content, and the YA meteorites are uncommonly ancient. Our electron probe microanalysis (EPMA) data indicate that the YA meteorites and the dominant mare components of Y-793274 and EET 87521 conform to a general trend for Ti-poor (low-Ti and very low-Ti) mare basalts. Their pyroxenes show a strong correlation between Fe/(Fe + Mg) (Fe#) and Ti/(Ti + Cr) (Ti#), both ratios typically increasing from core to rim. These trends presumably reflect local crystallization differentiation of interstitial melt. Previous studies (M. J. Drake and coworkers) have suggested that the detailed configurations of such Fe# vs. Ti# trends may reflect the bulk TiO₂ contents of the parent magmas (basalts). As a more systematic approach to this problem, we plot bulk-rock TiO₂ as a function of the Fe# = 0.50 intercept of each rock's pyroxene Fe# vs. Ti# trend. We call this intercept the Fe#-normalized Ti#. Based on our data for EET 87521, the YA meteorites, and Apollo 12 basalts 12031 and 12064, plus literature data for several other Ti-poor mare basalts, we find a strong correlation between Fe#-normalized Ti# and the bulk TiO₂ content of the parent basalt. This correlation confirms that fragmental breccia EET 87521 is nearly pure very low-Ti (VLT) basalt and that the YA meteorites, for which bulk-rock TiO₂ results scatter due to unusually coarse grain size (A-881757) or scarcity of available sample (Y-793169), are pieces of an uncommonly Ti-poor, but not quite VLT, variety of low-Ti mare basalt. Extrapolating from this correlation, the dominant mare component of regolith breccia Y-793274 is probably of VLT affinity. Besides the normal mare pyroxene trend of strong correlation between Fe# and Ti#, Y-793274 includes two additional pyroxene compositional trends, both showing a wide range of Ti# despite relatively constant (and low, by mare standards) Fe#. The most magnesian of these trends consists of a single clast with a mode of orthopyroxene + MgO-rich ilmenite. These two trends are of uncertain origin. Possibly one or both represents the highland component of this regolith breccia, although, unlike most highland pyroxenes, these appear relatively unaltered by impact brecciation and metamorphism. Compositions of spinels in the coarse-grained A-881757 show an extraordinary distribution: chromite and ulvöspinel components vary among grains but are nearly constant within grains. Despite its old age and unusually coarse grain sizes, mineralogical evidence (i.e., heterogeneity within both pyroxene and spinel; typical pyroxene exsolution scale very coarse by mare standards but exceeded by the pyroxenes of EET 87521 and Y-793274) indicates that A-881757 was cooled only slightly more slowly than typical mare basalts and may have formed near the center of an uncommonly thick lava flow. Both of the VLT basaltic lunar meteorite breccias, EET 87521 and Y-793274, are composed dominantly of pyroxenes with exsolution coarser than normal for mare basalts. Possibly VLT basalt flows tend to be systematically thicker, and thus more slowly cooled, than more Ti-rich flows.