Dating of the devils hole calcite vein.

et al. (2) present a long, continuous, highresolution climate record extracted from a calcite vein in Devils Hole, Nevada. If their dating is accurate, the record appears to contradict the Milankovitch hypothesis (3). Dating of subsamples of the vein was done with the use of well-established principles of 230Th dating (4) and with recently developed techniques for precise measurement of 230Th and 234U by thermal ionization mass spectroscopy (5). Careful analytical work has produced a precise 230Tf chronology (2). The accuracy of the chronology (1) depends on the assumption that the calcite behaved as a closed system and that no 230Th was incorporated during calcite growth. The first assumption appears to hold because initial 234U/238U ratios of subsamples are generally the same, even for the oldest subsamples. The second assumption also appears to hold on the basis of the following argument. The 230Th/232Th value of thorium initially incorporated in the calcite was estimated; by multiplying this value by the measured 232Th content of the calcite, the initial 230Th content was calculated. Thus, the initial 230Th in each sample from Devils Hole was found to be negligible (2), which apparently confirmed the second assumption. Although conventional wisdom argues that the calculated ages are accurate, there is a question with regard to the fate of 23Mth produced by the radioactive decay of 234U dissolved in the water in Devils Hole. While the detailed chemical behavior of 230Th in groundwater is not well understood, it has low solubility in most natural waters and it is likely that 23oTh produced in Devils Hole water was adsorbed onto the walls of the hole. We offer a simple model with an idealized reservoir geometry and assume that 230Th is instantaneously adsorbed onto the walls. This mechanism would invalidate the assumption of no 23&fl incorporation and offset 230Th ages to artificially high values. Such an offset would be insignificant in most cases; however, at Devils Hole it may be important because of extremely slow calcite growth rates. In our model, a "hole" is bounded by two parallel infinite half-planes and filled with water containing dissolved 234U. Calcite precipitates on both walls and incorporates 234U from the water as it precipitates. 234U in the water decays to 230Th, which is adsorbed instantaneously onto the walls. At the surface of each wall, the ratio of adsorbed 230Th atoms to 234U atoms incorporated in the calcite is