Noble gases in five shales have been examined in the traditional protocol for trapped gas analysis and also in a new protocol in which the shales are sealed in air prior to analysis. The second protocol eliminates the possibility of gas loss in the preanalysis laboratory vacuum exposure characteristic of the traditional protocol. This experiment is designed to further test the ''shale hypothesis'' which stipulates that atmospheric noble gases occur in the same proportions as ''planetary gases in meteorites and that the factor of approximately 23 deficiency of air Xe relative to planetary Xe is made up by Xe stored in shales or other sedimentary rocks. Previously available data do not support this hypothesis but could be criticized as possibly seriously underestimating shale Xe contents because of experimental limitations. The procedure used in this stuidy circumvents these limitations, and the results still do not support the shale hypothesis. From these results it appears quite unlikely that sedimentary rocks can account for the hypothesized missing Xe and, unless some other atmospheric reservoir can be found to contain it, it correspondingly appears that the best working hypothesis is that atmospheric noble gases are simply not planetary. If so, either the chemical history of the earth must account for preferential storage of Xe in the earth's interior or solar nebula process responsible for noble gas incorporation in solids were significantly different for meteorites and the earth.