Search

The (f)O2 [oxygen fugacity] of crystallization for martian basalts has been estimated in various studies to range from IW-1 to QFM+4 [1-3]. A striking geochemical feature of the shergottites is the large range in initial Sr isotopic ratios and initial epsilon(sup Nd) values. Studies by observed that within the shergottite group the (f)O2 [oxygen fugacity] of crystallization is highly correlated with these chemical and isotopic characteristics with depleted shergottites generally crystallizing at reduced conditions and enriched shergottites crystallizing under more oxidizing conditions. More recent work has shown that (f)O2 [oxygen fugacity] changed during the crystallization of these magmas from one order of magnitude in Y980459 (Y98) to several orders of magnitude in Larkman Nunatak 06319. These real or apparent variations within single shergottitic magmas have been attributed to mixing of a xenocrystic olivine component, volatile loss-water disassociation, auto-oxidation during crystallization of mafic phases, and assimilation of an oxidizing crustal component (e.g. sulfate). In contrast to the shergottites, augite basalts such as NWA 8159 are highly depleted yet appear to be highly oxidized (e.g. QFM+4). As a first step in attempting to unravel petrologic complexities that influence (f)O2 [oxygen fugacity] in martian magmas, this study explores the effect of (f)O2 [oxygen fugacity] on the liquid line of descent (LLD) for a primitive shergottite liquid composition (Y98). The results of this study will provide a fundamental basis for reconstructing the record of (f)O2 [oxygen fugacity] in shergottites and other martian basalts, its effect on both mineral chemistries and valence state partitioning, and a means for examining the role of crystallization (and other more complex processes) on the petrologic linkages between olivine-phyric and pyroxene-plagioclase shergottites.

A spiked (with REE, V, Sc) martian basalt Y980459 composition was used to synthesize olivine, spinel, and pyroxene at 1200 C at 5 oxygen fugacities: IW-1, IW, IW+1, IW+2, and QFM. The high spike levels for REE were used for two specific reasons. First, we wanted to be able to analyze REE by both electron microprobe and ion probe. Second, we wanted the most important "Others" components, (i.e., those outside the pyroxene quadrilateral such as Al, Cr3+, Fe3+, REE3+, V3+, V4+, etc.) to be REE3+Mg (Si,Al)2O6. At the doped levels we used, the most important "Others" component is REE3+ in the M2 site coupled with Al in the tetrahedral site. The goal of this paper is to explain the significant increase in the value of D(sub V)(sup pyroxene/melt) with increased Wo content of the pyroxene. We compare augite (Wo approx. 33), pigeonite (Wo approx. 13) and orthopyroxene (Wo approx 3.8). We also show olivine for comparison. The crystal chemical factors which account for this remarkable increase of DV with Wo are twofold. First, with Ca in the M2 site (as in diopside, CaMgSi2O6) the site is large and 8-coordinated while Mg in the M2 site (as in enstatite, Mg2Si2O6) the site is smaller and 6- coordinated. Second, tetrahedral Al in the pyroxene chains provides charge balance and makes the M2 site larger and more compliant for the introduction of REE.

Cr is a ubiquitous and relatively abundant minor element in basaltic, planetary magmas. At the reduced oxidation states (

Interpreting the relationship between "enriched" olivine-phyric shergottites (e.g. NWA 1068/1110) and the "enriched" pyroxene-plagioclase shergottites (e.g. Shergotty, Los Angeles) is problematic. Symes et al. [1] and Shearer et al. [2]) proposed that the basaltic magma that crystallized to produce olivine-phyric shergottite NWA 1068/1110 could produce pyroxene-plagioclase shergottites with additional fractional crystallization. However, additional observations indicate that the relationship among the enriched shergottites may be more complex [1-3]. For example, Herd [3] concluded that some portion of the olivine megacrysts in this meteorite was xenocrystic in origin, seemingly derived from more reduced basaltic liquids. This conclusion may imply that a variety of complex processes such as magma mixing, entrainment, and assimilation may play important roles in the petrologic history of these meteorites. It is therefore possible that these processes have obscured the petrogenetic linkages between the enriched olivine-phyric shergottites and the pyroxene-plagioclase shergottites. As a first order step in attempting to unravel these petrologic complexities, this study focuses upon exploring the effect of fO2 on the crystallization history for an analog primitive shergottite liquid composition (Y98). Results from this work will provide a basis for reconstructing the record of fO2 in shergottites, its effect on both mineral chemistries and valence state partitioning, and a means for examining the role of crystallization on the petrologic linkages between olivine-phyric and pyroxene-plagioclase shergottites. A companion abstract [4] explores the behavior of V over this range of fO2.

Our research on valence state partitioning began in 2005 with a review of Cr, Fe, Ti, and V partitioning among crystallographic sites in olivine, pyroxene, and spinel [1]. That paper was followed by several on QUE94201 melt composition and specifically on Cr, V, and Eu partitioning between pyroxene and melt [2-5]. This paper represents the continuation of our examination of the partitioning of multivalent V between olivine, spinel, and melt in martian olivine-phyric basalts of Y980459 composition [6, 7]. Here we introduce a new, potentially powerful oxybarometer, V partitioning between spinel and olivine, which can be used when no melt is preserved in the meteorite. The bulk composition of QUE94201 was ideal for our study of martian pyroxene-phyric basalts and specifically the partitioning between pyroxene-melt for Cr, V, and Eu. Likewise, bulk composition Y980459 is ideal for the study of martian olivine-phyric basalts and specifically for olivine-melt, spinel-melt, and spinel-olivine partitioning of V as a function of oxygen fugacity.

Several studies, using different oxybarometers, have suggested that the variation of fO2 in martian basalts spans about 3 log units from approx. IW-1 to IW+2. The relatively oxidized basalts (e.g., pyroxene-phyric Shergotty) are enriched in incompatible elements, while the relatively reduced basalts (e.g., olivine-phyric Y980459) are depleted in incompatible elements. A popular interpretation of the above observations is that the martian mantle contains two reservoirs; 1) oxidized and enriched, and 2) reduced and depleted. The basalts are thus thought to represent mixing between these two reservoirs. Recently, Shearer et al. determined the fO2 of primitive olivine-phyric basalt Y980459 to be IW+0.9 using the partitioning of V between olivine and melt. In applying this technique to other basalts, Shearer et al. concluded that the martian mantle shergottite source was depleted and varied only slightly in fO2 (IW to IW+1). Thus the more oxidized, enriched basalts had assimilated a crustal component on their path to the martian surface. In this study we attempt to address the above debate on martian mantle fO2 using the partitioning of Cr and V into pyroxene in pyroxene-phyric basalt QUE 94201.

Using partitioning behavior of V between olivine and basaltic liquid precisely calibrated for martian basalts, we determined the redox state of primitive (olivine-rich, high Mg#) martian basalts near their liquidus. The combination of oxidation state and incompatible element characteristics determined from early olivine indicates that correlations between fO2 and other geochemical characteristics observed in many martian basalts is also a fundamental characteristic of these primitive magmas. However, our data does not exhibit the range of fO2 observed in these previous studies.. We conclude that the fO2 for the martian upper mantle is approximately IW+1 and is incompatible-element depleted. It seems most likely (although clearly open to interpretation) that these mantle-derived magmas assimilated a more oxidizing (>IW+3), incompatible-element enriched, lower crustal component as they ponded at the base of the martian crust.

Spinel is a minor but important phase in planetary basalts because its variable composition often reflects basalt petrogenesis. For example, complicated zoning trends in spinel can give clues to melt evolution [1], and V concentrations in chromite lend insight into magma oxygen fugacity (fO2) conditions [2]. Nickel and Co are two elements that are commonly used as a measure of melt fractionation, and their partitioning between olivine and melt is fairly well understood. Less clear is their partitioning into spinel, although [3] has explored Ni and Co systematics in experimental charges. This study documents Ni and Co behavior in early crystallizing spinel (chromite) from several planetary basalts in an attempt to compare our results with [3], and also gain insight into basalt evolution on the three planets.

We have considered the valence-state partitioning of Cr, Fe, Ti, and V over crystallographic sites in olivine, pyroxene, and spinel from planetary basalts. The sites that accommodate these cations are the M2 site (6-8 coordinated) and M1 site (6 coordinated) in pyroxene, the M2 site (6-8 coordinated) and M1 site (6 coordinated) in olivine, and the tetrahedral and octahedral sites in spinel. The samples we studied are basalts from Earth, Moon, and Mars, which have fO2 conditions that range from IW-2 (Moon) to IW+6 (Earth) with Mars somewhere between at IW to IW+2. In this range of fO2 the significant elemental valences are (from low to high fO2) Ti4+, V3+, Fe2+, Cr2+, Cr3+, V3+, V4+, and Fe3+. V2+ and Ti3+ play a minor role in the phases considered for the Moon, and are probably in very low concentrations. V5+ plays a minor role in these phases in terrestrial basalts because it is probably in lower abundance than V4+ and it has an ionic radii that is so small (0.054 nm, 6- coordinated,[1]) that it is almost at the lower limit for octahedral coordination. The role of Cr2+ in the Moon is significant, as Sutton et al. [2] found that lunar olivine contains mostly Cr2+ while coexisting pyroxene contains mostly Cr3+. Hanson one vacancy only accommodates one V4+. Thus more vacancies are required in V4+ substitutions into olivine. In the Moon V3+ is much more abundant than V4+ [7, 8]. Thus in lunar chromite V3+ follows Cr3+, whereas in Earth V4+ (which is much greater in abundance than V3+) follows Ti4+ (eg in ulv spinel). We could go on in this vein for some time but space limitations do not permit us to do so.

We have been developing an oxygen barometer based on the valence state of V (V(2+), V(3+), V(4+), and V(5+)) in solar system basaltic glasses. The V valence is determined by synchrotron micro x-ray absorption near edge structure (XANES), which uses x-ray absorption associated with core-electronic transitions (absorption edges) to reveal a pre-edge peak whose intensity is directly proportional to the valence state of an element. XANES has advantages over other techniques that determine elemental valence because measurements can be made non-destructively in air and in situ on conventional thin sections at a micrometer spatial resolution with elemental sensitivities of approx. 100 ppm. Recent results show that fO2 values derived from the V valence technique are consistent with fO2 estimates determined by other techniques for materials that crystallized above the IW buffer. The fO2's determined by V valence (IW-3.8 to IW-2) for the lunar pyroclastic glasses, however, are on the order of 1 to 2.8 log units below previous estimates. Furthermore, the calculated fO2's decrease with increasing TiO2 contents from the A17 VLT to the A17 Orange glasses. In order to investigate these results further, we have synthesized lunar green and orange glasses and examined them by XANES.

The highly siderophile elements (HSE: including Re, Au, Ir, Os, Ru, Pt, Pd, Rh) are strongly partitioned into metal relative to silicates. In the terrestrial planets these elements are concentrated in metallic cores. Earth s mantle has sufficiently high abundances of the HSE (~0.008 times CI abundances) that it has been hypothesized approximately 0.1-0.5% of the mass of the Earth was added following the last major interaction between the core and mantle [e.g. 1]. The additional material added to the Earth and Moon has been termed a late veneer , and the process has often been termed late accretion [2]. The timing of the dominant late accretionary period of the Earth and Moon is still poorly known. The abundances of HSE in the lunar mantle could provide important constraints on when the late veneer was added. The material that ultimately became the silicate portion of the Moon was likely stripped of most of its HSE prior to and during coalescence of the Moon. Consequently the initial lunar mantle likely had very low concentrations of the HSE. Unlike Earth, the generation of permanent lunar crust by ~4.4 Ga prevented subsequent additions of HSE to the lunar mantle via continued accretion. Thus, if a substantial portion of the late veneer was added after 4.4 Ga, the lunar mantle should have retained very low HSE concentrations. Conversely, if the late veneer was mostly added prior to 4.4 Ga, HSE abundances in the lunar mantle may be roughly similar to abundances in the terrestrial mantle.

As our contribution to the new "Oxygen in the Solar System" initiative of the Lunar and Planetary Institute and the NASA Cosmochemistry Program, we have been developing oxygen barometers based largely on behavior of V which can occur in four valence states V2+, V3+, V4+, and V5+, and record at least 8 orders of magnitude of fO2. Our first efforts in measuring these valence proportions were by XANES techniques in basaltic glasses from Earth, Moon, and Mars. We now address the behavior of V valence states in chromite in basalts from Earth, Moon, and Mars. We have been looking for a "V in chromite oxybarometer" that works with data collected by the electron microprobe and thus is readily accessible to a large segment of the planetary materials community. This paper describes very early results that will be refined over the next two years.

The ability to estimate oxygen fugacities for mare basalts and to extend these observations to the lunar mantle is limited using bulk analysis techniques based on buffering assemblages or the valence state of iron. These limitations are due to reequilibration of mineral assemblages at subsolidus conditions, deviations of mineral compositions from thermodynamic ideality, size requirements, and the limits of the iron valence at very low fO2. Still, these approaches have been helpful and indicate that mare basalts crystallized at fO2 between the iron-w stite buffer (IW) and the ilmenite breakdown reaction (ilmenite = rutile + iron). It has also been inferred from these estimates that the lunar mantle is also highly reduced lying at conditions below IW. Generally, these data cannot be used to determine if the mare basalts become increasingly reduced during transport from their mantle source and eruption at the lunar surface and if there are differences in fO2 among mare basalts or mantle sources. One promising approach to determining the fO2 of mare basalts is using the mean valence of vanadium (2+, 3+, 4+, 5+) determined on spots of a few micrometers in diameter using synchrotron x-ray absorption fine structure (XAFS) spectroscopy. The average valence state of V in basaltic glasses is a function of fO2, temperature, V coordination, and melt composition. Here, we report the initial results of this approach applied to lunar pyroclastic glasses.

The absolute and relative abundances of the highly siderophile elements (HSE) present in planetary mantles are primarily controlled by: 1) silicate-metal partitioning during core-mantle differentiation, 2) the subsequent addition of HSE to mantles via continued planetary accretion. Consequently, constraints on the absolute and relative abundances of the HSE in the lunar mantle will provide unique insights to the formation and late accretionary history of not only the Moon, but also Earth. Determining the HSE content of the lunar mantle, however, has proven difficult, because no bona fide mantle rocks have been collected from the moon. The only materials presently available for constraining mantle abundances are lunar volcanic rocks. Lunar basalts typically have very low concentrations of HSE and highly fractionated HSE patterns. Because of our extremely limited understanding of mantle melt partitioning of the HSE, even for terrestrial systems, extrapolations to mantle compositions from basaltic compositions are difficult, except possibly for the less compatible HSE Pt and Pd. Primitive, presumably less fractionated materials, such as picritic glasses are potentially more diagnostic of the lunar interior. Here we report Os isotopic composition data and Re, Os, Ir, Ru, Pt and Pd concentration data for green glass (15426,164) and orange glass (74001,1217). As with previous studies utilizing neutron activation analysis, we are examining different size fractions of the spherules to assess the role of surface condensation in the generation of the HSE abundances.

Previous studies have demonstrated the usefulness of major and minor elements in silicate phases to understand differences among basaltic systems and the influence of different planetary environments on basalt chemistry (e.g., Papike [1]). Intriguing data displays presented by Papike [1] include a plot of Mn vs. Fe (atoms per formula unit, afu) for pyroxene and olivine and a plot showing the anorthite content of plagioclase from different planetary basalts. Here we combine portions of these plots (Fig. 4) and provide all new data for olivine and plagioclase.

Major, minor and trace element analysis of silicates has allowed for the study of planetary basalts in a comparative planetary mineralogy context. We continue this initiative by exploring the chemistry of plagioclase feldspar in basalts from the Earth, Moon, Mars and 4 Vesta. This paper presents new data on plagioclase from six terrestrial basalt suites including Keweenawan, Island Arc, Hawaiian, Columbia Plateau, Taos Plateau, and Ocean Floor; six lunar basalt suites including Apollo 11 Low K, Apollo 12 Ilmenite, Apollo 12 Olivine, Apollo 12 Pigeonite, Apollo 15 Olivine, and Apollo 15 Pigeonite; two basaltic martian meteorites, Shergotty and QUE 94201; and one unequilibrated eucrite, Pasamonte.

Spinels from Apollo 12 Olivine basalts have been studied by Electron and Ion microprobe techniques. The zoning trends of major, minor and trace elements provide new insights into the conditions under which planetary basalts form. Additional information is contained in the original extended abstract.

The systematics of vanadium in olivines from the Earth, Moon and Mars allows for the comparison of planetary basalt origin and igneous setting and process. Additional information is contained in the original extended abstract.

Terrestrial iron formations preserve remnants of life on Earth and may serve as analogs for identifying evidence of biologic activity in martian rocks. We report on the petrography, mineralogy and trace-element abundances of carbonates of the Gunflint banded iron formation. Additional information is contained in the original extended abstract.

Mn-oxides are promising biomarkers because microbes greatly accelerate Mn-oxide formation rates and produce distinctive oxidation states. Mn minerals in terrestrial caves form subaerially and could conceivably be present in the subsurface of Mars. Additional information is contained in the original extended abstract.

Secondary Ion Mass Spectrometry (SIMS) data for Ni, Co, Cr, and V in olivine in martian basalts is compared to data from lunar and terrestrial basalts. We use experimentally-derived and published D values to calculate as-yet unsampled, olivine-bearing, non-cumulus melt compositions. Additional information is contained in the original extended abstract.

Pyroxene chemistries from the Earth, Moon, Mars, and 4 Vesta show indications of planetary parentage. Additional information is contained in the original extended abstract.

Orthopyroxene chemistry in unique, metamorphic diogenite GRO 95555 is compared to orthopyroxene in a suite of diogenites.

Experiments involving the composition of the groundmass of EETA79001 Lithology A constrain the Cr content of the melt and the oxygen fugacity, and suggest that overgrowth of olivine and pyroxene from the groundmass onto xenocrysts has occurred.

This volume contains papers that have been accepted for presentation at the Workshop. Topics considered include: On the sample return from Vesta by low-thrust spacecraft; Astronomical evidence linking Vesta to the HED meteorites; Geologic mapping of Vesta with the Hubble Space Telescope; A space mission to Vesta; Asteroid spectroscopy; The thermal history of asteroid 4 Vesta, based on radionuclide and collision heating; Mineralogical records of early planetary processes on Vesta.

We are evaluating the trace-element concentrations in the pyroxenes of Pasamonte. Pasamonte is a characteristic member of the main group eucrites, and has recently been redescribed as a polymict eucrite. Our Pasamonte sample contained eucritic clasts with textures ranging from subophitic to moderately coarse-grained. This study concentrates on pyroxenes from an unequilibrated, coarse-grained eucrite clast. Major-, minor-, and trace-element analyses were measured for zoned pyroxenes in the eucritic clast of Pasamonte. The major- and minor-element zoning traverses were measured using the JEOL 733 electron probe with an Oxford-Link imaging/analysis system. Complemenatry trace elements were then measured for the core and rim of each of the grains by SIMS. The trace elements analyzed consisted of eight REE, Sr, Y, and Zr. These analyses were performed on a Cameca 4f ion probe. The results of the CI chondrite normalized (average CI trace-element analyses for several grains and the major- and minor-element zoning patterns from a single pyroxene grain are given. The Eu abundance in the cores of the pyroxenes represents the detection limit and therefore the (-Eu) anomaly is a minimum. Major- and minor-element patterns are typical for igneous zoning. Pyroxene cores are Mg enriched, whereas the rims are enriched in Fe and Ca. Also, Ti and Mn are found to increase, while Cr and Al generally decrease in core-to-rim traverses. The cores of the pyroxenes are more depleted in the Rare Earth Elements (REE) than the rims. Using the minor- and trace-element concentrations of bulk Pasamonte and the minor- and trace-element concentrations from the cores of the pyroxenes in Pasamonte measured in this study, we calculated partition coefficients between pyroxene and melt. This calculation assumes that bulk Pasamonte is representative of a melt composition.

Considerable attention has been paid recently to the primitive achondrites because they may form a link between chondrites and more differentiated achondrite meteorites. A recent paper by Miyamoto and Takeda addresses the thermal history of lodranites Yamato 74357 and MAC 88177 as inferred from chemical zoning of pyroxene and olivine determined by electron microprobe analyses. Their results suggested that interstitial melt was present and then extracted. We have taken the analysis of Lodran-type meteorites one step further by incorporating the techniques of Electromagnetic Pulse/Wavelength Dispersive Spectroscopy (EMP/WDS) compositional imaging and scanning ion mass spectroscopy (SIMS) analysis. Orthopyroxene in Lodran is strongly zoned in CaO, Al2O3, TiO2, and Cr2O3 within the last 10-30 microns from the grain boundaries. The rims are reversely zoned in Mg-Fe, exhibiting Mg enrichment, and compositions change from a fairly uniform Wo3En94 within the grains to Wo1En96 at the rims. CaO drops from 1.6 to 0.6 wt% and Al2O3, TiO2, and Cr2O3 exhibit similar depletions. MnO is fairly uniform throughout the grains at around 0.5 wt%. Olivine is also reversely zoned with respect to not only grain boundaries but also to fractures within the grains, giving many olivine grains a complex, patchy zoning pattern. Some of the core-rim trace-element systematics for orthopyroxene are illustrated. Because of the rather narrow zoned rims in Lodran orthopyroxene and the low trace-element abundances, it is difficult to clearly resolve the trace-element systematics. Nevertheless it is evident that the cores are enriched in the incompatible trace elements Ce, Nd, Dy, Er, Yb, Y, and Zr relative to the rims.

The lunar Mg suite, which includes dunites, troctolites, and norites, could make up 20-30% of the Moon's crust down to a depth of 60 km. The remainder is largely anorthositic. This report focuses on norites because we have found that the chemical characteristics of orthopyroxene are effective recorders of their parental melt compositions. Many of the samples representing the Mg suite are small and unrepresentative. In addition, they are cumulates and thus are difficult to study by whole-rock techniques. Therefore, we decided to study these rocks by SIMS techniques to analyze a suite of trace elements in orthopyroxene. The 12 norite samples were selected from a recent compilation by Warren who attempted to select the best candidate samples from the standpoint of their pristine character. Our present database includes greater than 300 superior Electromagnetic Pulse (EMP) analyses and greater than 50 scanning ion mass spectroscopy (SIMS) analyses for 8 Rare Earth Elements (REE), Zr, Y, and Sr. The Mg#s for the parental melts calculated from Mg#s in orthopyroxene show that most melts have Mg#s in the range of 0.36-0.60. This compares with a range of Mg#s for lunar volcanic picritic glass beads of 0.4-0.68. Therefore, although the cumulate whole-rock compositions of the Mg suite can be extremely magnesian, the calculated parental melts are not anomalously high in Mg. A chemical characteristic of the Mg-suite norites that is more difficult to explain is the high KREEP content of the calculated parental melts. The REE contents for the calculated norite parental melts have REE that match or exceed the high-K KREEP component of Warren. Therefore, mixing of a KREEP component and a picritic melt cannot, by itself, explain the high estimated REE contents of the melts parental to norites. Advanced crystallization following KREEP incorporation, especially of plagiclase, may also be required.

In order to constrain the thermal histories of chondritic meteorites, a detailed study of the microstructures of low-Ca pyroxenes produced experimentally and in types 4 and 5 ordinary chondrites was carried out. Cooling experiments on synthetic MgSiO3 at cooling rates between 2 and 10000 C/hr from the protopyroxene stability field into that of orthopyroxene (OPX) were performed and the products of these experiments were annealed for a variety of annealing times. There are clear microstructural differences between samples which were cooled and those which were subsequently annealed. A comparison of the microstructures observed in the experimental samples with those in H4-5 ordinary chondrites shows that they cannot have experienced a single stage cooling history, as proposed for the autometamorphism model.

A petrographic and mineralogical study of a number of carbonaceous chondrite clasts in the Kapoeta howardite has been carried out. Most of the clasts have mineralogical and chemical properties which link them to the CM carbonaceous chondrites. Some clasts contain chondrules which often have well-developed fine-grained rims, but many have been extensively brecciated. PCP-rich objects are common and pentlandite and pyrrhotite also occur. Calcite has also been found. The remainder of the clasts are extremely fine-grained and appear to be closely related to CI carbonaceous chondrites. In these clasts magnetite framboids are common and finegrained sulfides and magnetite occur disseminated throughout the matrix.

The very low Ti green glasses from the Apollo 15 site have intrigued scientists for over 20 years. Their primitive composition has been used to understand magmatic processes and the structure of the moon. The compositional variability observed in the Apollo 15 glass population has long been a point of debate. Initial studies did not recognize the compositional diversity in the glasses. Stolper et al. documented the major element variability and concluded it could not be produced by magmatic processes and therefore concluded that these glasses must be of impact origin. Subsequent studies confirmed a volcanic origin for the glass population and attempted to elucidate magmatic processes to account for its compositional variability. Models that have been proposed for these glasses include the following: (1) the crystallization of single or multiple phases (olivine, pyroxene, Fe metal, immiscible sulfide); (2) the incompatible behavior of Ni and Co during multiple phase crystallization at extremely low fO2; and (3) magma or source mixing. All of these models have problems. Type (1) models appear not to be consistent with recent trace element studies on the glasses; model (2) is dependent on the debatable incompatible behavior of Ni and Co, and, in models of type (3), the origin and nature of mixing models are somewhat unconstrained. This study compares the Apollo 15 green glasses with the very low Ti picritic glasses from other landing sites.

It is widely accepted that basaltic magmas are products of partial fusion of peridotite within planetary mantles. As such they provide valuable insights into the structure and processes of planetary interiors. Those compositions which approach primary melt compositions provide both a clearer vision of planetary interiors and a starting point at which to understand basaltic evolution. Within the collection of lunar samples returned by the Apollo and Luna missions are homogeneous, picritic glass beads of volcanic origin. These glass beads provide a unique perspective concerning the origin of mare basalts, the characteristics of the lunar interior, and processes culminating in the early differentiation of the moon. In this presentation, we report our ion microprobe derived trace element data from all picritic glasses previously identified. We place this trace element data and literature isotopic and experimental data on the picritic glasses with the framework of mare basaltic magmatism.

Eucrite, howardite, and diogenite members of the achondrites are considered, by many, to be genetically related. Therefore, each provide a piece of the puzzle for reconstructing magmatic processes on the eucrite parent body (EPB). The relationship between eucrites and diogenites can be viewed within the context of two distinctly different models: (1) fractional crystallization; and (2) partial melting. In fractional crystallization models, eucrites and diogenites represent a complementary continuum of planetary fractional crystallization products in which the diogenites represent crystal accumulations during the crystallization of eucritic magmas at shallow to deep levels in the EPB. Alternatively, experimental studies may be interpreted as indicating eucrites represent peritectic partial melts of a primitive, chondritic EPB mantle. Within this type of model, the diogenites are also generally considered to be cumulates; however, their petrogenetic relationship to the eucrites is less clear. Sack et al. proposed that the olivine diogenites represent residua from the partial melting events that produced eucritic liquids. Initial trace element studies of orthopyroxene (OPX) are consistent with this model. However, this trace element modeling of the olivine diogenites is nonunique. As a further test of these models, we did the following three things: (1) analyzed OPX from cumulate diogenites to compare with the olibine diogenite data; (2) improved ion microprobe analytical techniques for the analysis of elements critical to our interpretations; and (3) selected more relevant Kd's for OPX-eucritic melt.

As a part of our research to better understand magmatic processes in the Eucrite Parent Body, we have initiated an ambitious program of study of major, minor and trace elements in orthopyroxene from diogenites. This paper reports preliminary results for major and minor elements in orthopyroxenes for a suite of 13 diogenites: Aioun El Atrouss, ALH 84001, ALH A 77256, EET 87530, Ellemeet, Garland, Ibbenburen, Johnstown, Manegoan, peckelsheim, Roda, Shalka, and Tatahouine. A companion paper by Shearer et al. reports new trace element data for ALH 84001, ALH A 77256, Ibbenburen, and Tatahouine. We have presently collected over 800 high quality pyroxene microprobe analyses for Si, Al, Ca, Na, Mn, Fe, Mg, Cr, and Ti. The chemical systematics observed for these orthopyroxenes reflect original magmatic mineral/melt partitioning plus later trapped liquid/mineral equilibration, subsolids, exsolution, and mineral/mineral metamorphic reactions. We have therefore avoided, at this point, any attempt to use statistical analysis to group (e.g. factor or cluster analysis) these orthopyroxenes chemically.

A detailed understanding of the properties of silicate minerals in chondrules is essential to the interpretation of chondrule formation conditions. This study is further work in a series of petrologic studies of chondrules in the least equilibrated LL chondrite, Semarkona (LL3.0). The objectives of this work are as follows: (1) to understand chondrule formation conditions and nebular processes; and (2) to use the data as a basis for understanding the effects of metamorphism in more equilibrated chondrites. FeO-rich pyroxene in the chondrules described shows complex zoning behavior. Low-Ca clinopyroxene, orthopyroxene, pigeonite, and augite are all observed, in various associations with one another. Coexisting olivine phenocrysts are also FeO-rich and strongly zoned. Compositional and zoning properties are similar to those observed in boninites and are interpreted as resulting from rapid cooling of individual chondrules.

The concentrations of Ni, Cu, Zn, Ga, Ge, and Se in five chondrule rims in the CO3 chondrite ALH A77307 (3.0) using the synchrotron x-ray fluorescence (SXRF) microprobe at Brookhaven National Laboratory were determined. The data show that the trace element chemistry of rims on different chondrules is remarkably similar, consistent with data obtained for the major elements by electron microprobe. These results support the idea that rims are not genetically related to individual chondrules, but all sampled the same reservoir of homogeneously mixed dust. Of the trace elements analyzed Zn and Ga show depletions relative to CI chondrite values, but in comparison with bulk CO chondrites all the elements are enriched by approximately 1.5 to 3.5 x CO. The high concentrations of the highly volatile elements Se and Ga and moderately volatile Zn (1.5 to 2 x CO) in rims show that matrix is the major reservoir of volatile elements in ALH A77307.

Results are presented from SIMS analyses of major, minor, and trace elements in lunar picritic glass beads obtained from the Apollo 14 (A-14) and Apollo 17 (A-17) landing sites. The results obtained indicate that the glasses from each site differed significantly in their Ba/Sr and light-REE/heavy-REE ratios. The differences between the A-14 and the A-17 glasses appear to be the result of a higher KREEP component in all the A-14 glasses, indicating that the mantle source of the A-14 site is intrinsically different from the A-17 mantle source.

Results are presented on major- and trace-element abundance analyses of Apollo 15 pyroclastic green glasses from groups A, B, C, D, and E, carried out using electron- and ion-microprobe techniques. The diagrams depicting Sr, Zr, Ba, and Nd vs Co variations indicate the presence of a high-Co trend in groups A and D and a low-Co trend in groups B and C. Group-E glasses were found to be significantly enriched in Sr, relative to the other four glass groups. Chemical data of this study were integrated with previous data to evaluate various magmatic processes that have been proposed in the past to explain chemical variations in the lunar green glass. Results of calculations using a source mixing model suggest that the Apollo 15 green glasses represent multiple eruptive events from three chemically distinct but compositionally variable source regions.

In Apollo 16 regolith breccias, Wentworth and McKay (1988) identified a suite of minute (less than 120 microns) 'komatiitic glass beads'. The wide major element compositional range, and ultra-Mg-prime character of the glasses suggest a variety of possible origins from complex impact processes to complex volcanic processes involving rather unusual and primitive magmatism. The extent of trace element depletion or enrichment in these glasses appears to be correlated to the siderophile character of the element (ionization potential or experimentally determined silicate melt/Fe metal partition coefficients. The ultra-Mg-prime glasses are depleted in Co relative to a bulk Moon Mg/Co exhibited by many lunar samples (volcanic glasses, basalts, regolith breccia, estimated upper mantle). The low Co and high incompatible element concentrations diminish the possibility that these glasses are a product of lunar komatiitic volcanism or impact, excavation, and melting of a very high Mg-prime plutonic unit.

Results are presented from trace element analysis, by ion microprobe techniques, of individual glass beads representing seven compositionally distinct types of picritic glass beads from the Apollo 14 landing site. The picritic glass beads at the A-14 exhibited a wide range of primary magma compositions and a lack of petrogenetic linkage (via crystal fractionation) to crystalline basalts. The wide range of major and trace element characteristics of the picritic glass beads is consistent with derivation from mineralogically distinct sources which consist of varying proportions of olivine + orthopyroxene +/- clonopyroxene +/- ilmenite +/- plagioclase +/- KREEP component.

Results are presented of petrological and chemical analyses of ten Apollo 17 breccias, showing that two of these consist predominantly of highland material, seven are mare-dominated, and one is a welded volcanic glass deposit; all were formed at or near the Apollo 17 site, and all contain both mare and highland components. The data are indicative of the Apollo 17 breccias formation from immature source regolith. The breccias are considered to be formed locally after an eruption of basalt and orange glass at the site. Since the formation of the breccias, the regolith at the Apollo 17 site has become more mature, and the orange glass abundance has been somewhat decreased by mixing. One of the sample may contain a previously unreported volcanic glass type.