Sixteen relatively large veins and one irregular felsic patch in gabbro were selected for chemical analysis. The vein material was carefully separated from the host rock, crushed in an agate mortar, and analyzed by standard X-ray fluorescence (XRF) techniques for major oxides and some of the abundant trace elements (Table T17). The same powders were then analyzed by inductively coupled plasma-mass spectrometry (ICP-MS) for the remaining trace elements, and H2O and CO2 were measured using a CHN analyzer. The sample preparation and all analyses were carried out at the GeoForschungsZentrum, Potsdam, Germany.
Twelve of the analyzed veins are classified as felsic, three are plagioclase + diopside, and one is plagioclase + amphibole. Seven of the felsic veins (Samples 176-735B-90R-3, 38-44 cm; 90R-4, 55-58 cm; 90R-4, 84-88 cm; 124R-1, 111-115 cm; 135R-3, 68-72 cm; 130R-3, 52-58 cm; and 138R-4, 59-63 cm) contain 5 modal% or more of quartz and three (Samples 176-735B-150R-7, 93-100 cm; 157R-7, 1-5 cm; and 161R-7, 78-84 cm) have traces of quartz in the groundmass. One of the plagioclase + diopside veins (Sample 176-735B-120R-2, 99-104 cm) contains ~15 modal% of chlorite. Sample 176-735B-202R-7, 99-101 cm, is a felsic patch rather than a discrete vein. It consists chiefly of light green, very fine grained, poorly crystallized material with a few large zoned crystals of green and brown amphibole.
Because the veins have a rather simple mineralogy dominated by plagioclase, they exhibit a relatively narrow range of composition. Silica contents range from 53 to 74 wt% and reflect the abundance of quartz. Except for two diopside-rich veins (Samples 176-735B-120R-2, 99-104 cm, and 123R-6, 140-146 cm), SiO2 and Al2O3 show a strong negative correlation, reflecting both the relative proportions of quartz and plagioclase and the more sodic nature of the plagioclase in the quartz-rich veins. Both Fe2O3 and MgO are low in all but the diopside-rich veins and correlate positively with the abundance of amphibole. CaO is relatively high in many of these veins, commonly between 2 and 5 wt% and up to more than 12 wt% in the most diopside-rich specimen. K2O is very low in all the veins, ranging from 0.01 to 0.43 wt%. This reflects the nearly complete absence of K-feldspar and the sparse presence of mica in these rocks. A few flakes of yellowish brown to reddish biotite are present in Samples 176-735B-135R-3, 68-72 cm; 138R-4, 59-63 cm; and 157R-7, 1-5 cm, the veins with the highest K2O. Likewise, Rb is very low, generally <1 ppm. The highest Rb is 4 ppm, and that is in the sample with the highest K2O. Despite the presence of a few grains of titanite in most of these veins, TiO2 contents are also very low (maximum = 0.27 wt%). Samples 176-735B-135R-3, 68-72 cm, and 157R-7, 78-84 cm, have a few grains of apatite, just enough to produce very small increases of P2O5 compared to the other veins. All of the analyzed samples contain some CO2 (0.29-2.60 wt%), suggesting the presence of carbonate. No carbonate was identified optically in these veins, but some finely disseminated material might be present in the groundmass. However, there is only a weak correlation between CO2 and CaO contents, and the samples with the highest CO2 show no particular enrichment in CaO. This suggests that the CO2 resides in some other phase, perhaps in sparse clay minerals, zeolites, or poorly crystallized matrix materials that are present in a few samples. Except in Sample 176-735B-120R-2, 99-104 cm, which has abundant chlorite in the matrix, water contents are relatively low, mostly <1 wt%.
Most of the trace elements, except for Ba, Sr, Zr, Y, Zn, Ni, and Ga, have very low concentrations. Ba averages ~30 ppm but is significantly higher in the two siliceous quartz-plagioclase veins (Table T17). Sr ranges from 35 to 194 ppm (40-197 ppm by ICP-MS) and shows a good positive correlation with CaO except in the diopside-rich samples. Zr values determined by XRF (33-1720 ppm) are significantly higher than those determined by ICP-MS (3-30 ppm). The differences suggest that zircon was not completely digested during preparation for ICP-MS analysis. Because the XRF values correlate best with the observed modal distribution of zircon, we believe that they are the most accurate. If zircon was not completely dissolved during sample preparation for the ICP-MS analyses, other elements such as Hf, Pb, U, and Th, which typically concentrate in zircon, are probably also underrepresented and thus are not reported here. Yttrium values have a narrow range of variation (19-136 ppm) and show no correlation with Zr. Concentrations of Zn range from <10 to 358 ppm, and the few high values probably reflect small amounts of secondary sulfides that are present in some of the veins. Ni values are relatively high (up to 87 ppm) in some felsic veins composed chiefly of plagioclase and quartz. This suggests that small amounts of olivine may have been incorporated into the vein sample during preparation.
Because the geochemistry of the veins is strongly controlled by the vein mineralogy, we produced geochemical maps of four veins in order to show the zoning and distribution of elements within the veins. The geochemical maps were produced using the Cameca SX-100 microprobe at the GeoForschungsZentrum, Potsdam, Germany, using both energy-dispersive and wavelength-dispersive spectrometers. Spot spacing was 30 µm, small enough to sample all but the finest crystals.
Sample 176-735B-124R-1, 111-115 cm, is a quartz-plagioclase vein, ~2 cm wide, with a myrmekitic texture. Element maps of this vein reveal sharp, distinct vein walls and strong zoning within the vein (Fig. F8). The vein walls are clearly delineated in Figure F8 by the presence of calcic plagioclase, indicated by high Ca and Al. Along the vein walls are very narrow bands of amphibole characterized by moderate levels of Al, Ca, and Si, followed inward by a bands of sodic plagioclase about 2 mm wide. The myrmekitic intergrowths of quartz and sodic plagioclase (clearly visible in the plot of Si) form two parallel bands, each about 5-7 mm wide. These are followed inward by narrow felsic bands, 1-1.5 mm wide, without quartz. The center of the vein is composed of greenish brown amphibole with a few very small grains of ilmenite.
Sample 176-735B-123R-6, 140-144 cm, illustrates the compositional zoning in a typical plagioclase + diopside vein (Fig. F9). The vein margin (upper left) is sharp and marked by a significant change in plagioclase composition. The boundary in the lower right is against epoxy. Small amounts of diopside are present along the vein wall in the upper left, but most of the vein is filled with sodic plagioclase (An0.9-22.5). The distinct band, 1-2 mm wide, near the center of the vein consists of diopside and minor plagioclase in a matrix of black to dark brown aphanitic material. Overall, the plagioclase in this vein is quite uniform in composition, with very little zoning, although there is a very narrow band along the margin with slightly elevated Na.
Sample 176-735B-161R-7, 77-83 cm, is an excellent illustration of a coarse-grained, plagioclase-rich vein with a strong hydrothermal overprint (Fig. F10). Again, the margin of the vein is sharp against calcic plagioclase and clinopyroxene in the host rock. The large plagioclase crystals within the vein are subhedral to euhedral and strongly zoned, particularly at the edges, clearly becoming more sodic. In thin section, the cores are pitted and corroded, whereas the rims are clear and smooth. The spaces between the grains may be open or filled with minor amphibole and quartz.
Strong zoning of plagioclase is also observed in Sample 176-735B-120R-2, 99-104 cm (Fig. F11). Subhedral crystals up to 5 mm long have relatively calcic cores and sodic rims. The large clinopyroxene grain in the upper right part of the map has a narrow rim of chlorite, and hairline cracks in the vein plagioclase are also filled with chlorite. The central part of the vein is filled with dark brown to black aphanitic material, probably a clay mineral.