RESULTS

A single basalt glass sample and plagioclases from three samples were analyzed for major elements using the Cameca CAMEBAX electron microprobe at Brown University. The electron beam was set at 15 nA beam current and 15 kV accelerating voltage. Analyses were performed on polished thin sections and the results are shown in Table 1. The pillow-rim glass is from the aphyric Formation L. Its composition is tholeiitic and is very similar to the shipboard whole-rock analysis of the same flow (Table 1). Plagioclase analyses were performed on basalts from sparsely plagioclase-phyric (Formation B), plagioclase and pyroxene-phyric (Formations G and H), and aphyric (Formation L) basalts. Plagioclase phenocrysts are euhedral to subhedral and may display evidence of disequilibrium with the surrounding melt (rim corrosion and embayments). The relatively high anorthite content of phenocryst cores from Formations B (An₈₄) and H (An₈₆) and the lower anorthite contents of the rims and groundmass laths (An_73-60) also indicate that some of the phenocrysts are not in equilibrium with the melt and they could therefore be defined as xenocrysts.

Age determinations on three whole-rock samples were performed at Oregon State University using standard ⁴⁰Ar-³⁹Ar incremental heating techniques (Duncan and Hargraves, 1990). Samples were irradiated at the Oregon State University TRIGA reactor for 6 hr, and neutron flux was monitored by FCT-3 biotite (28.04 ± 0.12 Ma; Renne et al., 1994). Ar isotopic composition of whole-rock basalts were determined using either an Associated Electrical Industries MS-10S mass spectrometer or a Mass Analyzer Products (MAP) 215-50 mass spectrometer attached to a low-blank, all-metal extraction line.

Individual ages for each ⁴⁰Ar-³⁹Ar step were calculated after corrections for background, mass fractionation, isotopic interferences, and atmospheric argon content. Step ages and Ar isotopic measurements for each sample are reported in Table 2. Plateau ages were calculated using the procedure described in Dalrymple et al. (1987), in which step ages were weighted by the inverse of their variance. Weighted linear regressions of the step compositions in each of the isotope correlation diagrams yielded a slope that is proportional to the age (isochron), with the inverse of the y-intercept giving the initial ⁴⁰Ar-³⁶Ar composition at the time of crystallization.

Three whole-rock analyses yielded consistent "plateau" and isochron ages of approximately 81 Ma with calculated initial ⁴⁰Ar-³⁶Ar values close to atmospheric (295.5). The lowest uncertainties (± 1.3 and ± 1.2 Ma) are from the plateau ages from the two samples analyzed with the low-blank MAP 215-50, although the isochron ages display much higher uncertainties. Because the ages for all three samples are close to 81 Ma, a 1- to 2-m.y. uncertainty can be assigned on the assumption that the MS10 analysis overestimates the uncertainty, primarily because of the higher system blank.