and theoretical ratios of ionization for fast charged particles in argon relative to air which may be partly responsible for the higher values of cosmic-ray ionization in the lower 5 atmosphere measured by Neher using argon filled chambers. In order to pursue this question further we are planning with the cooperation of the NYU group to expose two identical thin walled chambers filled with argon and air, respectively, to the predominantly cosmic ray field on a large lake. If both chambers are properly saturated the resultant ratio of ion currents per atmosphere should be helpful in investigating this discrepancy. If the experimental value of 1.65 - 1.70 is verified then further study will be necessary to explain why present stopping power theory fails to predict this value. Since the experimental value of 1.65 was obtained in all cases using thick walls it is possible that the discrepancy may be related to a fast particle wall effect, although Johnston’? claims this is unlikely. It is clear, however, that the interaction of high energy particles with a thick- walled chamber is still not perfectly understood and is worthy of continued study since such chambers are still being used on balloon flights for high altitude cosmic ray studies.’ * Theoretical Investigation of Gamma Ray Transport Our present theoretical calculations of dose rates for various source distributions in the soil and air, based on the dose rate buildup theory of O'Brien et al,** is apparently adeguate for the calculation of the total dose rate from distributed sources. However, it does not allow a calculation of the angular distribution or energy spectrum of the incident flux at the detector. The knowledge of these distributions would be valuable in interpreting the response of various types of instrumentation to different kinds of source distributions and source energies as well as for calculating the actual energy deposition in various organs of the body. At present these distributions can be calculated only by time consuming Monte Carlo methods. Because of the machine time required these methods have only been applied to a few particular problems and only for a few source energies. In particular Monte Carlo calculations have substantiated the experimental prediction of an interface effect for a point source near the boundary of two media of very different

Select target paragraph3