[en] The objective of this study was to determine if neutron activation analysis technique can be used to determine the shell content of a sand-shell mixture by measuring the amount of calcium. To accomplish this aim, samples of sand and shell, obtained from different sources, were artificially blended in the laboratory in various proportions and subjected to neutron bombardment by use of a californium-252 source. The gamma of specific energy of emission was then counted and analyzed electronically. The nuclear count was then plotted against the known shell content for development of calibration curves for each combination of sand and shell tested. Statistical analysis of the data indicates a strong linear relationship exists between nuclear count (gamma) and percent shell. Test results for 30 sets of sand-shell mixtures indicated a maximum standard error for any single set to be + or - 7.0% of the calibration curve at the 95% confidence level. The mean standard error for the 30 sets tested was + or - 4.2%. Sands from different sources may result in widely different nuclear counts after activation. This effect can be negated by generating a separate calibration curve for each source of sand used