Recent U/TH and 14C measurements by Bard et al. (1990) on corals from Barbados have indicated that 20,000 years ago the activity of 14C in the atmosphere was approximately 50% higher than the modern activity. They ascribed this increase in activity to enhanced 14C productions due to more intense cosmic radiation as a result of a weaker terrestrial dipole magnetic field. We propose to test this origin for the 14C enhancement by reconstructing the deposition of another cosmogenic radionuclide, 36C1. A similar enhancement of the 36C1 would demonstrate that both nuclides were responding to production fluctuations. We propose to sample atmospheric 36C1 preserved in fossil packrat (Neotoma) urine in an arid site in western Nevada. This site contains a midden that has been 14C dated to cover the time interval 29,000 to 3,000 years before present. If strong correlation of the 36C1 and the Barbados 14C records is found, the 36C1 data might help to reconstruct atmospheric 14C activity during the period 30,000 to 20,000 years B.P., for which other data are scarce. We will also compare 36C1/C1 ratios in midden samples from western Nevada and other widely spread localities with meteoric 36C1 ratios in the same localities in order to establish the relationship between the rat urine and soil-water 36C1. In addition to addressing the problem of secular variations in cosmogenic nuclide production, the record of variation in meteoric 36C1 may provide a valuable dating agent for goundwater and soil water.
