Compositions and textures of melt rocks from the upper part of the Chicxulub struc-ture are typical of melt rocks at other large terrestrial impact structures. Apart from variably elevated iridium concentrations (less than 1.5 to 13.5 t 0.9 ppb) indicating nonuniform dissemination of a meteoritic component, bulk rock and phenocryst compositions imply that these melt rocks were derived exclusively from continental crust and platform-sedi-ment target lithologies. Modest differences in bulk chemistry among samples from wells located 40 km apart suggest minor variations in relative contributions of these target lithologies to the melts. Subtle variations in the compositions of early-formed pyroxene and plagioclase also support minor primary differences in chemistry between the melts. Evidence for pervasive hydrothermal alteration of the porous mesostasis includes albite, K-feldspar, quartz, epidote, chlorite, and other phyllosilicates, as well as siderophile element-enriched sulfides, suggesting the possibility that Chicxulub, like Sudbury, may host important ore deposits.

INTRODUCTION

This report presents detailed petrographic descriptions and chemical analyses of igneous-textured rocks from the Chicxulub structure in Yucatan, Mexico. A suite of observations including diagnostic evidence of shock metamorphism (Sharpton et al., 1992), isotopic signatures (Sharpton et at., 1992; Swisher et al., 1992; Blum et al., 1993; Krogh et al., 1993), and geophysical constraints (Sharpton et at., 1993) provide compelling arguments that the Chicxulub structure is a buried multiring basin formed by hypervelocity impact and is the source of ejecta distributed worldwide at the Cretaceous-Tertiary (K-T) boundary, 65 m.y. ago (Alvarez et at., 1980). Nevertheless, some workers dispute its impact origin and continue to proffer opinions that the Chicxulub structure is a volcanic sequence of Late Cretaceous age (Meyerhoff et al., 1994). These opinions are based in part on early well-log descriptions of andesite and bentonitic breccia-now recognized as a sequence of impact-melt rock and suevitic breccia (Sharpion et al., 1992). The stratigraphic sequence involved in the impact event includes ~2.5 km of platform sedi-ments over crystalline basement of conti-nental affinity (Lopez Ramos, 1975; Sharpton et at., 1994a). With estimates of its diameter ranging from 180 to 300 km, (Hildebrand et al., 1991; Sharpton et at., 1993), Chicxulub is clearly one of the largest and best-preserved impact structures on Earth. Consequently, it provides a unique opportunity to observe, on a variety of scales, the effects of processes involved in the formation and evolution of large impact-melt sheets, as well as the relation between these main melt volumes and globally dis-persed ejecta.

At present, however, samples of melt rock from within the structure are limited, and it is unclear whether any of them represent material from a continuous melt sheet. Our samples were obtained from drill cores recovered from Petr6leos Mexicanos explor-atory wells Chicxulub I (Cl) and Yucatan 6 (Y6), located 40 km apart near the center of the structure (Sharpton et at., 1993). Spe-cifically, they comprise material from the Cl-Nb interval, 1393-1394 m below sea level (bsl), and from intervals Y6-N17 (1295.5-1299 m bsl) and Y6-N19 (1377-1379.5 m bsl). Although initial studies of samples from these and the adjacent C1-N9 intervals have been published (Hildebrand et at., 1991; Kring and Boynton, 1992; Swisher et al., 1992; Sharpton et al., 1992; Blum et at., 1993; Koebert et al., 1994), a coordinated evaluation of whole-rock major and trace element chemical analyses together with compositional variations among the principal liquidus phases has not been published previously. Here we present additional geochemical and textural characterizations of those Chicxulub melt rocks currently available to us in order to provide further constraints on their formation and evolution.

PETROGRARAPHY

Melt-rock textures of the three core in-tervals are distinctly different (Fig. 1), most notably in the size and abundance of undi-gested clasts, variations in color, grain size, and porosity of the matrix, and evidence of alteration. Clasts in Y6-M7 constitute 35% of the rock and show a bimodal size distribution dominated by single mineral fragments and polycrystalline domains of highly deformed, recrystallized quartz and feldspar less than 1 mm in length, with larger fragments up to 4 mm (Fig. 1A). Subhedral, stubby to skeletal pyroxene prisms (10-70 mm long) enclosed in quartz (some bordered by anhydrite) form coronas surrounding quartz clasts and pervade the interiors of more highly disrupted granitic domains, with aggregates up to 1 mm. In some extreme instances, a glomerophyric cluster of pyroxene is the only visible remains. Such pyroxenes are confined to individual quartz and quartz-rich granitic fragments and are notably absent around feldspar. Micrographic intergrowths of pyroxene, magnetite, and vermicular feldspar form clotlike domains in the matrix and probably represent melted but unassimilated terro-magnesian basement-clast components. The matrix comprises subhedral to euhedral microphenocrysts of pyroxene and plagioclase ranging from 5 to 15 mm in length, set in a porous, cryptocrystalline mesostasis (Fig. lB). Minor phases include magnetite; flmenite; apatite; sphene; sulfides; a hydrous, iron- and magnesium-rich aluminosilicate; and trace amounts of barite and halite. Anhydrite constitutes 8% of the thin section, mostly as veins and cavity fillings.

Samples from the Y6-N19 interval reveal a melt matrix breccia (Fig. 1C) containing 2-11 cm angular to subrounded melt clasts of at least two texturally distinct types. The dominant melt clast type is very similar to the surrounding matrix, and in some cases the boundary between them is difficult to discern. This material is also essentially similar to Y6-N17, consisting of 5-15-mm-long, subhedral to euhedral pyroxene and plagioclase in a cryptocrystalline quartzofeld-spathic mesostasis showing variable poros-ity. Minor constituents include magnetite, itmenite, apatite, sphene, zircon, sulfides, and a rare earth element (REE)-rich phase. In some regions, lath-shaped pyroxene and plagioclase microphenocrysts show a well-developed trachytic texture interfingering with regions of more randomly oriented grains, which may also be aligned but in a direction oblique to the plane of the thin section. These alignments appear to be flow foliations reflecting turbulent mingling of melt. In contrast to Y6-N17, undigested silicate basement clasts in Y6-N19 are typically larger (up to 8 mm diameter) and show clear examples of planar deformation features. Also, in addition to veins and cavity fillings, there are undigested but recrystallized angular fragments of anhydrite. As in Y6-N17, pyroxene intergrown with quartz commonly mantles partially digested quartz and granitic fragments.

The other melt clast type in Y6-N19 appears to be derived from a granitic or granodioritic gneiss protolith that was not disaggregated but in which most of the silicate mineral constituents were melted. These clasts are predominantly anhedral quartz and feldspar domains (up to 4 mm), which deformed plastically around isolated fragments of undigested shocked quartz, and elongate, irregular dense regions that appear opaque in transmitted light. Reflected-light and backscattered-electron images re-veal that these dense regions are melt domains composed of cryptocrystalline pyroxene with a vermicular intergrowth of feldspar and minute oxides. In some thin sections, these dense, elongate regions are roughly aligned and may reflect a relict foliation of ferromagnesian minerals in the protolith. Other opaque regions consist of anastomosing networks of skeletal ilmenite, intergrown with sphene. The interstices of the silicate domains include brownish, fluidal textured regions, some showing spherulitic textures typical of devitrification. The anhedral quartz domains are commonly surrounded by pyroxene prisms up to 75 mm in length, whereas the feldspar domains, which have nonstoichiometric compositions ranging from An48Ab490r3 to An26b46Or28 are mantled by a similarly nonstoichiometric but more potassic composition of An2Ab100r88.

Cl-Nb (Fig. 1D) is distinct from the Y6 samples with respect to both the virtual ab-sence of unmelted clasts and the coarser grain size of the matrix. The matrix is dominated by an intersertal arrangement of subhedral to euhedral pyroxene up to 0.7 mm and plagioclase showing a range of crystal morphologies from skeletal, swallowtail, and box-work outlines to lath-shaped prisms up to 1 mm in length. Some of these pyroxene and plagioclase phenocrysts are twinned and slightly zoned toward their margins. The mesostasis is a porous intergrowth of sodic and potassic feldspar and quartz, showing spherulitic texture in some regions. Minor phases include magnetite; apatite; sphene; pyrite; chalcopyrite; chlorite; epidote; calcite; and a hydrous, iron- and magnesium-rich aluminosilicate. The matrix also contains angular or rounded clasts of much finer grained melt rock (Fig. 1D). The minerals in these melt fragments are essentially the same as those in the host, although the smaller pyroxene (less than 100 mm) and plagioclase (less than 50 mm) phenocrysts impart a more granular texture and no potassic feldspar was observed in the mesostasis, which is less porous.

CHEMISTRY

Whole-rock major element compositions (Table 1) are similar to medium- to high-K calc-alkalic andesite to dacite (Gill, 1981). Results for Y6-N17 generally agree with those published elsewhere (Hildebrand et al., 1991). As expected from the variegated lithology of the Y6-N19 breccia, these sub-samples exhibit some compositional variability, but on average are significantly lower in SiO2 and Na2O and higher in GaO than either Y6-N17 or Cl-Nb.

Trace element concentrations are also similar to those of andesites, with the only significant departure being anomalous Ir enrichments in several of the specimens. Concentrations in two fragments of C1-N1O and duplicate splits of C1-N1O-2 and Y6-N19-R are identical within analytical uncertainties except for Ir and Au; the Cl-Nb analyses also show heterogeneity for Cr and Co (Ta-ble 1). The Y6-N19 subsamples span nearly the total range of variation among specimens from the three core intervals for many elements. The Cl-Nb samples exhibit modest enrichments in Co, Zr, Mt Ta, and heavy REEs (HREEs), and lower Sr relative to the Y6 specimens.

Pyroxene phenocryst compositions in our samples from all three core intervals are exclusively augite and, predictably, lie within the range of augite core compositions (En40-55Wo38-50Fs7-20) in andesites (Gill, 1981). The coarser grains of the C1-N1O matrix show an iron-enrichment trend (Fig. 2A), with modest, corresponding increases in Na2O, TiO2, and MnO. These variations also characterize the extent of core to rim zoning within individual phenocrysts, the increase of Fe occurring abruptly near crystal margins. Compositions within the finer-grained melt clasts in C1-N1O form a relatively tight cluster with an average composition of En49Wo42Fs9. Our analyses of augite microphenocrysts in Y6-N17 and Y6-N19 yield an average (En43Wo45Fs12) consistent with those of Kring and Boynton (1992), but contrast with the fassaitic compositions reported by Cedillo et al. (1994). Compared to those in Cl-Nb, augites in Y6-N17 and Y6-N19 are generally lower in SiO2 and molar Mg/(Mg + Fe), and higher in Na2O, TiO2, and MnO. Apart from slightly higher 5i02, there are no significant compositional differences between augite microphenocrysts in the groundmass and those bordering undigested quartz clasts.

Although the feldspar mineral assemblage as a whole shows considerable chemical variability, plagioclase is the only feld-spar present as a phenocryst. Consequently, those early-formed plagioclase crystals that have not suffered extensive alteration (Fig. 2B) define a more restricted range of variation (andesine to labradorite) and thus are compositionally as well as texturally distinct from feldspars in the surrounding mesostasis (Fig. 2C). With decreasing An content coarser plagioclase phenocrysts in the Cl-Nb matrix show a cor-responding monotonic decrease in MgO, and an initial Fe-enrichment trend that attains a maximum at An50, followed by a de-crease in FeO. Plagioclase phenocrysts within the finer-grained melt clasts of Cl-N1O tend to be more calcic, relatively constant in MgO, and higher in FeO, with an Fe-enrichment maximum at An56. Analyses of Y6-N17 and Y6-N19 are generally higher in K20 and FeO and lower in MgO than those of CI-N1O and are consistent with the average composition of groundmass plagioclase in Y6-N17 published previously (Kring and Boynton, 1992).

The mesostasis of Cl-Nb (Fig. 2C) in-cludes alkali feldspar and plagioclase ranging from oligoclase to pure albite. An example of the textural relations of these feldspars to a euhedral plagioclase phenocryst (Fig. lE) shows that albite forms at the expense of the calcic host, which in turn is surrounded by anhedral K-feldspar inter-grown with quartz, epidote, minute opaque minerals, and a cryptocrystalline alumino-silicate that appears to be a devitrification product of glass. Feldspar compositions in the mesostasis of Y6-N17 are highly variable (Fig. 2C); however, with the exception of albite, our analyses indicate that they arc nonstoichiometric. These anhedral, cation-deficient phases fill the interstices of the andesine and augite microphenocrysts, some of which protrude into the ubiquitous drusy cavities (Fig. lB). Thermodynamic consid-erations together with textural relations between early-formed phases and thetorous mesostasis suggest to us that, as in Cl-Nb, the albite results from secondary alteration (Schuraytz and Shaipton, 1993). Similar compositional and textural relations characterize feldspar variations in the Y6-N19 mes-ostasis, although the variations in porosity are more extreme.

DISCUSSION

Except for anomalous Ir enrichments in several specimens attributed to nonuniform dissemination of the projectile (Shaipton et al., 1992; Schuraytz and Sharpton, 1994), our analyses suggest that the melts were derived exclusively from continental crust and platform-sediment target lithologies, with no evidence of a significant mantle or oceanic crustal signature. These results are sup-ported by Sr, Nd, 0, and Os isotopic studies on the Ci-NiG samples with regard to both the continental affinity of the target rocks (Blum et al., 1993) and the heterogeneous distribution of up to 3% meteoritic contam-ination (Koeberl et al., 1994). Considering current constraints on excavation depth (15-25 km) of the Chicxulub impact event (Sharpton et al., 1994b) and the potential lithologic diversity within this volume, the observed chemical variability is rather small, in keeping with the gross compositional homogeneity of melt rocks from other terres-trial impact structures, such as Manicouagan (Grieve and Floran, 1978) and West Clearwater (Simonds et al., 1978). However, given that these few specimens represent an inordinately small sampling of the upper ~100 m of known melt rock, the limited compositional range should be regarded as tentative, as should comparisons with smaller structures where the upper part of the melt sequence has been eroded. The small variations in bulk composition (e.g., 5i02, CaO, Na2O, Sr, and HREEs) suggest that, compared to those from C1-N1O, the melt rocks from Y6 assimilated a greater proportion of platform-sediment target rocks relative to silicate basement. Compo-sitional differences among augite and plagioclase phenocrysts (the principal silicate liq-uidus phases) also imply primary variations in melt chemistry. The inverse correlation between clast abundance and matrix grain size (cL Figs. lA and 1D) indicates a substantial difference from site to site in the thermal regimes of the melts; this difference, together with the compositional differences, suggests that the Cl-Nb samples were derived from a zone of deeper melting and protracted cooling.

Although the phenocrysts preserve clear evidence of igneous crystallization, it appears that secondary mineralization due to percolation of hydrothermal fluids through the porous mesostasis was an integral process in the evolution of these rocks. All our specimens show some level of alteration, al-though 40Ar/39/Ar determinations indicate that Cl-Nb is least affected (Sharpton et al., 1992). Kring and Boynton (1993) argued that evidence of hydrothermal alteration in Y6-N17 is limited to quartz and possibly anhydrite veins. Despite their claims to the contrary, the melt-rock groundmass is pervasively affected by alteration. Even the least altered samples from C1-N1O contain anhedral albite (greater than Ab99), K-feldspar, quartz, epidote, chlorite, and yet-to-be determined phyllosilicates, as well as pyrite and chalcopyrite. The pyrites are significantly en-riched in Co, Ni, Au, As, and Sb (Schuraytz and Sharpton, 1994; Schuraytz et al., 1994) and show oscillatory zoning (Fig. 3) similar to pyrites observed in hydrothermal ore deposits (Fleet et al., 1989), indicating episodic variations in the composition of circulating fluids over the course of sulfide-mineral growth.

At the Sudbury structure, Ontario, the breccias and melt bodies within the Onaping Formation exhibit many textural and mineralogical similarities to the Chicxulub rocks described above, including extensive alteration of plagioclase to albite, secondary epidote, chlorite, and minor, but ubiquitous sulfide mineralization (Muir and Peredery, 1984). Muir noted that sulfides in the sub-layer of the Sudbury Igneous Complex are similar to those occurring throughout the Onaping Formation, whereas Peredery regarded the majority of these sulfides to be due to secondary replacement. The origin of the metals in these sulfides is unknown; however, Allen et al. (1982) cited secondary silicate and clay mineralization within the Onaping as evidence of impact-induced hydrothermal alteration. Although the relation between sulfide genesis within the Onaping Formation and other parts of the Sudbury Igneous Complex is not completely understood, potential similarities between sulfide mineralization in the Onaping and in melt rocks from Chicxulub may signal the possibility of more extensive strategic resources elsewhere at the Chicxulub structure.

ACKNOWLEDGMENTS

We thank D.W. Mittlefehldt and S.-R. Yang for con-sultation and use of analytical facilities at the NASA Johnson Space Genter; A. Treiman, R Anderson, and an anonymous referee for critical reviews; and S. Hokanson and D. Rueb for graphics support. The Lunar and Planetary Institute is operated by Universities Space Re-search Association under contract (NASW 4574) with the National Aeronautics and Space Administration. Lunar and Planetary Institute contribution 834.

REFERENCES CITED

• Allen, C.C., Gooding, J.L., and Keil, K, 1982, Hydrothermally altered impact melt rock and breccia: Contributions to the soil of Mars: Jounnal of Geophysical Research, v.87, p. l0,083-10,101.
• Alvarez, L.W., Alvarez, W., Asaro, F., and Michel, H.V., 1980, Extraterrestrial cause for the Cretaceous-Tertiary extinction: Science, v.208, p. 1095-1108.
• Blum, J.D., Chamberlain, C.P., Hingston, M. P., Koe-berl, C., Mann, L. B., Schuraytz, B.C., and Sharp-ton, V. L., 1993, Isotopic comparison of Kr boundary impact glass with melt rock from the Chionilub and Manson impact structures: Nature, v.364, p.325-327.
• Cedillo P., F., Clacys, P., Grajales N., S. M., and Alvarez, W., 1994, New mineralogical and chemical con-straints on the nature of target rocks at the Chicxu-lub crater, in Conference on new developments regarding the KT event and other catastrophes in Earth history [abs.]: Houston, Texas, Lunar and Planetary Institute, LPI Contribution 825, p. 20-21.
• Fleet, M. E., MacLean, P.S., and Barbier, J., 1989, Oscillatory-zoned As-bearing pyrite from strata-bound and stratiform gold deposits: An indicator of ore fluid evolution, in Keays, R. R., et al., eds., The geology of gold deposits: The perspective in 1988: Economic Geology Monograph 6, p. 35&362.
• Gill, S. B., 1981, Orogenic andesites and plate tectonics: Berlin, Springer-Verlag. 390 p. Grieve, R.A.F., and Floran, R. 5., 1978, Manicouagan impact melt, Quebec 2. Chemical interrelations with basement and formational processes: Journal of Geophysical Research, v.83, p. 2761-2771.
• Hildebrand, A.R., Penfield, G.T., Icring, D.A., Pilk-ington, M., Camargo 1., N, Jacobsen, S. B., and Boynton, W.V., 1991, Chicxulub crater: A possible Cretaceous-Tertiary boundary impact crater on the Yucatan peninsula, Mexico: Geology,. 19, p.867-871.
• Koeberl, C., Sharpton, V.L., Schuraytz, B.C., Shirey, S.B., Blum, S.D., and Mann, L. F., 1994, Evidence for a meteoritic component in impact melt rock from the Chicxulub structure: Geochimica et Cos-mochimica Acta, v.58, p. 1679-1684.
• Icring, D. A., and Boynton, W.V., 1992, Petrogenesis of an augite-bearing melt rock in the Chicxulub structure and its relationship to Kr impact spherules in Haiti: Nature, v.358, p.141-144.
• Kring, D. A., and Boynton, W.V., 1993, KJT melt glasses: Nature, v.363, p.503-504.
• Krogh, T. E., Kamo, S. V, Shatpton, V. V, Marin, V F., and Hildebrand, A.R., 1993, U-Pb ages of single shocked zircons linking distal KY ejecta to the Chicxulub crater: Nature, v.366, p. 731-734.
• Lopez Ramos, F., 1975, Geological summary of the Yucatan Peninsula, in Nairn, A. E. M., and Stehli, F. G., eds., The ocean basins and margins, Volume 3-The Gulf of Mexico and the Caribbean: New York,, Plenum, p.257-282.
• Meyerhoff, A. A., Lyons, S. B., and Officer, C. B., 1994, Opinion: Chicxulub structure: A volcanic sequence of Late Cretaceous age: Geology, v.22, p. 3-4.
• Muir, T L., and Peredery, W.V., 1984, The Onaping Formation, in lrye, E. G., et al., eds., The geology and ore deposits of the Sudbury structure: Ontario Geological Survey Special Volume 1, p. 139-210.
• Schuraytz, B.C., and Sharpton, V. L., 1993, Chicxulub Kr melt complexities: Nature, v.362, p.503-504.
• Schuraytz, B.C., and Sharpton, V.L., 1994, Siderophile-element distribution in Chicxulub melt rocks: Forensic chemistry on the K-T smoking gun, in Conference on new developments regarding the K-T event and other catastrophes in Earth history [abs.]: Houston, Texas, Lunar and Planetary Institute Contribution 825, p. 106-108.
• Schuraytz, B.C., O'Connell, S B., and Sharpton, V. L., 1991, Iridium and other trace element measurements from the Cretaceous-Tertiary boundary, ODP Site 752, Broken Ridge, Indian Ocean, in Weissel, 3., Pierce, 3., Taylor, F., Alt, S., et al., Pro-ceedings of the Ocean Drilling Program, scientific results, Volume 121: College Station, Texas, Ocean Drilling Program, p.913-919.
• Schuraytz, B.C., Lindstrom, D.S., Martinez, R.R., Sharpton, V. L., and Mann, L.E., 1994, Distribution of siderophile and other trace elements in melt rock at the Chicxulub impact structure [abs.]: Houston, Texas, Lunar and Planetary Science XXV, p. 1221-1222.
• Sharpton, V.L., Dalrymple, G.B., Marin, L.E., Ryder, G., Schuraytz, B.C., and Urrutia-Fucugauchi, S., 1992, New links between the Chicxulub impact structure and the Cretaceous-Tertiary boundary: Nature, v.359, p. 819-821.
• Sharpton, V.L., and nine others, 1993, Chicxulub multiring impact basin: Size and other characteristics derived from gravity analysis: Science, v. 261, p. 1564-1567.
• Sharpton, V.L., Mario, L.F., and Schurayttt, B.C., 1994a, The Chicxulub multiring impact basin: Eval-uation of geophysical data, well logs, and drill core samples, in Conference on new developments regarding the K-T event and other catastrophes in Eanh history fabs.]: Houston, Texas, Lunar and Planetary Institute Contribution 825, p. 108-110.
• Sharpton, V.L., Marin, L.F., and Schuraytz, B.C., 1994b, Constraints on excavation and mixing during the Chicxulub impact event [abs.]: Houston, Texas, Lunar and Planetary Science XXV, p. 1255-1256.
• Simonds, C.H., Phinney, W.C., McGee, P.F., and Cochran, A., 1978, West Clearwater, Quebec impact structure, Part I: Field geology, structure and bulk chemistry: Lunar and Planetary Science Conference, 9th, Proceedings, v.2, p. 2633-2658. Swisher, C.C., III, and 11 others, 1992, Coeval ages of 65.0 million years ago from Chicxulub crater melt rock and Cretaceous-Tertiary boundary tektites: Science, v.257, p. 954-958.