AR-40 AR-39 AGE SPECTRA OF APOLLO-15 IMPACT MELT ROCKS BY LASER STEP-HEATING AND THEIR BEARING ON THE HISTORY OF LUNAR BASIN FORMATION/

We have obtained 26 high-resolution (16-51 steps) Ar-40/Ar-39 age spec tra using a continuous laser system on submilligram fragments of recry stallized melt and single-crystal plagioclase clasts from 12 Apollo 15 impact melt rocks collected at the Apennine Front where the Imbrium a nd Serenitatis basins intersect. These melt rocks represent a wide ran ge of compositions and at least half a dozen different impacts. Six of the melt rocks have reproducible, intermediate-temperature plateaus o ver 40% or more of the Ar-39 released; the plateaus are interpreted as crystallization (impact) ages and much of the non-plateau behavior is attributable to recoil. Samples 15294,6,21, 15304,7,69, 15314,26,156, 15357,15, and 15359,12 have mean Ar-40/Ar-39 plateau ages that are st atistically indistinguishable and fall within the narrow range 3852 +/ - 14 (2sigma) Ma to 3870 +/- 12 Ma with a weighted mean of 3865 +/- 5 Ma. Sample 15356,9 has a mean plateau age of 3836 +/- 11 Ma and may re present a distinctly younger impact. A seventh sample (15314,30,158) h as a peculiar but reproducible double plateau; a low-T one at 3873 +/- 9 Ma, which we think records the crystallization age, and a high-T on e of 3831 +/- 10 Ma, which we interpret as an experimental (Ar-39 reco il) artifact. Four of the remaining melt rocks (15308,9, 15414,2,37, 1 5436,2, 15445,253) have complex Ar-40/Ar-39 age spectra that indicate that they either formed in or were disturbed by impacts that occurred less-than-or-equal-to 3850 Ma but did not completely reset the K-Ar is otopic system. Sample 15414,3,36 is different. Its spectrum may repres ent release mainly from clasts that were not well degassed in a melt e vent at 3870 Ma; the melt phase contains little potassium. Because mos t of the Apennine Front material must be coeval with or predate format ion of the Imbrium Basin, it seems likely that the Imbrium impact is n o older than 3870 Ma and probably no older than 3836 Ma. So far there is no convincing evidence in the lunar record for melt-producing impac ts, such as basin formation, older than about 3.9 Ga.