AGE DETERMINATIONS

To determine the eruption ages of submarine lava clasts drilled at Sites 977 and 978, laser ⁴⁰Ar/³⁹Ar age determinations were performed on plagioclase, sanidine, amphibole, and biotite phenocrysts from volcanic rocks covering the spectrum from basalt to rhyolite. Age determination results are presented in Table 4.

Eleven plagioclase phenocrysts (0.134-0.387 mg) from dacite 7525 were dated by single-crystal ⁴⁰Ar/³⁹Ar analysis. Apparent ages range from 5.5 ± 1.2 Ma to 8.5 ± 0.6 Ma. Isotope correlation returns a scatterchron with an age of 6.5 ± 0.4 Ma, and an initial ⁴⁰Ar/³⁹Ar ratio of 293 ± 4 (mean squared weighted deviate [MSWD] = 3.8; Fig. 5A). If the oldest crystal is excluded, the subpopulation yields an isochron age of 6.4 ± 0.3 Ma (MSWD = 1.63) and a mean apparent age of 6.1 ± 0.3 Ma (MSWD = 2.45), which is considered here the best estimate of the eruption age of dacite 7525.

Nine plagioclase phenocrysts, one sanidine phenocryst, and one biotite phenocryst (0.250-0.800 mg) from rhyolite 7521 were analyzed by single-crystal fusion. With apparent single-crystal ages ranging from 9.39 ± 0.18 Ma to 9.98 ± 0.30 Ma, the plagioclase analyses define an isochron with an age of 9.46 ± 0.11 Ma (MSWD = 1.32; Fig. 5A), and yield a mean apparent age of 9.55 ± 0.49 Ma. Their initial ⁴⁰Ar/³⁹Ar ratio is determined as 307 ± 12, identical to the atmospheric ratio of 295.5 within 1- error limits. A step-heating analysis of a single biotite phenocryst from the same sample gave an errorchron of 9.41 ± 0.10 Ma (MSWD = 6.88) and an integrated age for individual heating steps of 9.49 ± 0.05 Ma (Fig. 5B), which is identical to the plagioclase ages within error limits. A single sanidine phenocryst yields the most precise age estimate for the eruption of rhyolite 7521. Its apparent age of 9.25 ± 0.02 Ma overlaps within error limits with the results obtained from plagioclase phenocrysts, but not with the poorly defined biotite errorchron.

Plagioclase phenocrysts from rhyolite 7522 (10 single-crystal analyses; 0.190-0.611 mg) give apparent ages ranging from 9.37 ± 0.07 Ma to 9.99 ± 0.12 Ma. The data are isochronous at 9.51 ± 0.14 Ma (MSWD = 2.71) and have a slightly elevated initial ⁴⁰Ar/³⁹Ar ratio of 341 ± 31 (Fig. 5A). The step-heating analysis of a biotite phenocryst gives an isochron age of 9.29 ± 0.10 Ma, a near-atmospheric initial ⁴⁰Ar/³⁹Ar ratio of 296 ± 12, and an MSWD of 1.64. The age spectrum is flat but does not represent a plateau region (sensu stricto) within 1- error limits (Fig. 5B). The integrated age is 9.25 ± 0.05 Ma. The best estimate of the eruption age of rhyolite 7522 is again derived from a single sanidine crystal analysis. Its apparent age of 9.29 ± 0.02 Ma agrees with the biotite step-heating results, but is significantly younger than the poorly defined plagioclase data.

Seven amphibole phenocrysts (0.060-0.261 mg) were dated from basalt 7523 (Fig. 5A). With single-crystal apparent ages ranging from 8.8 ± 2.9 Ma to 11.8 ± 1.6 Ma, the analyses yield an isochron age of 9.95 ± 0.64 Ma and an initial ⁴⁰Ar/³⁹Ar ratio of 294 ± 10 (MSWD = 0.94). Isochron and mean apparent ages (9.90 ± 0.40 Ma) are identical within error limits, but analytical precision is low because of small sample masses, low potassium content (i.e., insufficient neutron dose), and high nonradiogenic argon contents.

Plagioclase phenocrysts from dacite 7524 (six single-crystal analyses; 0.162-0.563 mg) give apparent ages ranging from 11.1 ± 1.3 Ma to 13.8 ± 0.3 Ma (Fig. 5A). This population yields a scatterchron age of 12.5 ± 0.9 Ma with an initial ⁴⁰Ar/³⁹Ar ratio of 303 ± 13 (MSWD = 7.1). The mean apparent age (13.0 ± 0.3 Ma) is poorly constrained as well; its high MSWD (6.1) indicates the presence of different age populations in the crystal assemblage analyzed. Excluding two crystals with apparent ages of 13.5 ± 0.2 Ma and 13.8 ± 0.3 Ma drastically decreases the MSWD (0.36) and yields a mean apparent age of 12.1 ± 0.2 Ma. It is not clear whether the observed scatter in the plagioclase population of 7524 and 7525 reflects loss of radiogenic ⁴⁰Ar from some of the apparently younger crystals because of alteration or thermal overprinting, or whether it indicates the presence of older xenocrysts in these magmas.

In summary, the single-crystal and step-heating laser ⁴⁰Ar/³⁹Ar analyses of plagioclase, sanidine, biotite, and amphibole phenocrysts from lava clasts in Cores 161-977A-60X-63X and 978A-47R indicate that the eruptions of these basaltic to rhyolitic magmas occurred between 6.1 (dacite 7525) and 12.1 Ma (dacite 7524). At Hole 977A, the eruption ages of the volcanic rocks studied, systematically increase with depth and thus conform to the stratigraphy (i.e., from Core 60X [9.25 Ma], to Core 61X [9.29 Ma], to Core 62X [9.9 Ma] and Core 63X [12.1 Ma]). A clast from Core 60X was assigned to Zone NN7 (~10.6-12.2 Ma) within the Serravallian, based on the microfossil assemblage scraped from this sample (Comas, Zahn, Klaus, et al., 1996). This provides further support for a mid-Miocene age for the base of Hole 977A. Sample 161-978A-47R-1, 0-6 cm (7525), also has an age similar to that of microfossils from the same unit (6.52 Ma; Comas, Zahn, Klaus, et al., 1996). These observations are consistent with deposition of the volcanic clasts shortly after their eruption and suggests that at least some of the pebbles were derived from local volcanic structures, such as Al-Mansour Seamount, Yusuf Ridge, and Maimonides Ridge (?).