ALBERT

Description: Viking XRF analyses are compared with those for terrestrial and lunar basalt samples, and eucritic meteorites (of possible Mars origin). The comparison indicates depletion of Ca relative to Si in the Mars regolith. It is suggested that carbonate formation during a warmer, wetter epoch early in Mars' history could have been responsible.

Description: The Apollo highlands rock collection includes more than 100 'pristine' fragments that survived the intense meteoritic bombardment of the ancient lunar crust with unmixed, endogenously igneous compositions. The geochemical anomaly manifested by the 'ferroan-anorthositic suite' (FAS) appears to reflect a geochemical, and probably also a genetic, bimodality among the ancient lunar cumulates. Early models that purported to account for this bimodality as a product of a single magma have been discredited. The model of the present paper implies that the Mg-suite rocks formed by a comparatively normal variety of basaltic fractional crystallization (FC) shortly after the era of magma ocean (MO) crystallization and FAS genesis.

Description: Alkalic suite pristine nonmare rocks are distinctly enriched in plagiophile elements such as Na and K, as well as generally incompatible elements, despite modes and textures more characteristic of typical crustal cumulates (most commonly anorthosites) than of the basaltic KREEP rocks that appear to account for the bulk of the lunar crust's total complement of incompatible elements. Our studies indicate that the 77115 troctolitic clasts of Winzer et al. is actually a troctolitic anorthosite (or anorthositic troctolite), probably best classified as a member of the alkalic suite. We managed to obtain a thin section with pyroxene and olivine, and analyzed a 13.4-mg chip by instrumental neutron activation analysis (INAA). The results of this analysis are briefly discussed.

Description: The Serenitatis Basin is the one lunar basin from which we confidently identify a suite of samples as pieces of the impact melt sheet: the distinctive Apollo 17 noritic breccias. Recent studies of the Sudbury Complex indicate that its 'irruptive' is almost entirely of impact-melt origin, making it the closest terrestrial analog to the Serenitatis melt sheet. Any attempt to model the evolution of the Moon's crust should be compatible with the relatively well-understood Sudbury Complex. However, the Sudbury-Moon analogy might be a misleading oversimplification, if applied too rigidly. The cause of evolutionary differences between the Serenitatis impact melt and the Sudbury impact melt is discussed.

Description: In the literature on cratering, one frequently finds allusions to the 'sheet' of impact melt that supposedly forms in the aftermath of any large impact. The presumption seems to be that either (1) most of the impact melt is formed virtually at the surface, or else (2) most of the melt inevitably collects into a near-surface, roughly sill-shaped 'sheet,' soon after the impact. However, in basin-scale impacts on the mature Moon, neither of these scenarios appears likely. Differentiation of impact melts on the Moon is hampered by similarity in density vs. the porous lunar crust, and by the adverse melt/'displaced' matter ratio that results from the lunar g.

Description: The MAC88104/105 meteorite, a lunar highlands regolith breccia, is described. The rock and a number of its component clasts are characterized. One of the clasts is considered to be a rare pristine nonmare rock containing extraordinarily Fe-rich (Fo40) olivine; the other has silicate compositions that extend the range of the Mg-suite in the direction of the high-mg* end of the ferroan-anorthositic suite. For the major element composition of the crust, the highlands meteorites confirm that the two low-Th central nearside sites, Luna 20 and Apollo 16, are approximately representative. The high Al2O3 composition indicated for the upper crusts supports the magmasphere hypothesis. For the trace-element composition of the crust, the highlands meteorites indicate that the central nearside Apollo and Luna sites are in several respects grossly unrepresentative.

Description: It is pointed out that the implication of the popular giant impact model of lunar origin (e.g., Hartmann and Davis, 1975; Cameron and Ward, 1976; Stevenson, 1987) is that any depth-related silicate differentiation within the impactor (and/or the earth) at the time of the impact must be partly inherited by the preferentially peripheral matter that forms the moon. This paper presents calculations of the magnitude of the net differentiation of the protolunar matter for a variety of elements and scenarios, with different assumptions regarding the geometries of the 'sampled' peripheral zones, the relative proportions of the earth-derived to impactor-derived matter in the final moon, and the degree to which the impactor mantle had crystallized prior to the giant impact. It is shown that these differention effects constrain the overall plausibility of the giant impact hypothesis.

Description: Data for 10 siderophile elements in all of the known lunar meteorites except for Y793274 are reported. Bulk compositional data for Ni, Ge, Cd, Re, Os, Ir, and Au were obtained from radiochemical neutron activation analysis (Warren et al., 1986). Data for Fe, Co, and Zn, and additional data for Ni, Ir, and Au were obtained by instrumental neutron activation analysis (Kallemeyn et al., 1988). Except for the case of Au in Y791197, there was good agreement between the results obtained by the two methods. The differences observed between the lunar-meteoritic regolith samples and central nearside highlands regolith samples are discussed.

Description: Launch mechanisms for lunar and martian meteorites have been investigated, by integrating physical modeling constraints, geochemical cosmic-ray exposure (CRE) constraints, and petrologic constraints. The potential source region for lunar meteorites is remarkably small compared to the final crater volume. CRE constraints indicate that most launches start at depths of less than or equal to 3.2 m, and cratering theory implies derivation of suitably accelerated objects from a subvolume with diameter only about 0.3 x the final crater diameter. The shallow depth provenance is probably related to shock-wave interference, enhanced by the lunar regolith's extremely low compressional wave velocity. CRE constraints alone imply that four to five separate launch events are represented among the eight well-studied lunar meteorites. Most of the lunar meteorites are regolith breccias, which tend to show only limited compositional diversity within any kilometer-scale region of the Moon. Several others are polymict breccias, which also show relatively subdued compositional diversity, compared to igneous rocks. The observed diversity among these samples in terms of abundances of mare basalt and KREEP, and in Mg/(Mg + Fe) ratio, implies that among eight well-studied lunar meteorites only two potential source craters pairings are plausible: between Asuka-881757 + Y-793169 (most probable) and between Y-793274 + EET875721. Altogether, these eight lunar meteorites apparently represent at least six separate source craters, including three in the past 10(exp 5) years and five in the past 10(exp 6) years. CRE constraints imply that SNC meteorites are launched from systematically greater than lunar meteorites. SNCs are also systematically bigger, and all nine well-studied SNCs are uncommonly young (by martian standards) mafic igneous rocks. Comparison between Viking and Apollo results reveals that rocks the size of common meteorites are remarkably scarce in the martian regolith, probably due to pervasive weathering. A plausible explanation for these trends is that most old, small, and shallow rocks on Mars have been weakened by a two-stage process of brecciation followed by pervasive weathering, to the point where they seldom survive the stresses of spallation off the planet. The scarcity of source-crater pairing among the lunar meteorites implies that these objects can be launched from craters much smaller than previously estimated and tends to support the suggestion of Rabinowitz (1993) that present-day cratering rates for the Earth-Moon region may be higher than previously estimated.

Description: The fact that pristine KREEP basalts generally have bulk-rock molar MgO/(MgO+FeO) ratios lower than average for the lunar crust despite their extraordinarily high incompatible element contents is discussed. Pristine KREEP basalts also have primitive Ni contents compared to mare basalts with comparable REE contents. It is suggested that these facts can be explained through mixing between ancient KREEP precursor materials and primitive Mg-rich melts. Finite-difference models of this mixing, followed by anorthosite assimilation and fractional crystallization provide satisfactory fits to the composition of these basalts. It is suggested that this is a further confirmation of the magmasphere hypothesis.