The variation of percent gypsum with depth in gypsiferous soils requires a special calibration curve for each layer in order to obtain accurate results in soil-water studies using a neutron probe. To overcome the tedious work of the field calibration for each layer, a correction factor for percent gypsum and bulk density was suggested. The effects of gypsum and bulk density on the neutron probe calibration curve were studied in the laboratory, using a special drum that can be bolted together in two halves, and in the field. The effect of bulk density determined on core samples was negligible for the soil studied in the laboratory, whereas it was significant for one of the field calibrations carried out on gypsiferous soils. An increase in the slope of moisture content on a volume basis vs. count ratio with increasing gypsum content determined by precipitation with acetone method of the soil was observed in the laboratory calibration. A simple method for correction of the calibration curve for gypsum content was adopted to obtain a specific curve for each layer. The adopted method requires the gypsum fraction to be estimated for each layer and then incorporated in the calibration curve to improve the coefficient of determination. The field calibrations showed improvements of the determination coefficient by introducing gypsum fraction, calcium carbonate fraction determined by the calcimeter, and bulk density in one calibration in addition to count ratio using moisture content on a volume basis as a dependent variable in multilinear regression analysis. The presence of gypsum in the soil increases count ratio in that some of the water is water of hydration, unavailable to plants. The effect of bulk density is not as clear in the case of organic soils and when hydrogen exists in the soil in forms other than free water