A laboratory test program was undertaken to determine the effects of variation in concrete constituents and proportions on the response of a nuclear water-cement content gage. A total of 14 separate test series were evaluated in the program. The evaluations included study of the effects of the following variables on the nuclear gage's determination of cement and water contents: air content, hold time, Class C and F fly ash, ground granulated blast furnace slag, maximum size of coarse aggregate, limestone coarse and fine aggregates, basalt coarse aggregate, and temperature. Testing has shown that the cement content probe is sensitive to all materials containing calcium atoms; therefore, the gage must be calibrated with exactly the same materials as will be used on the job in question. While the water probe is, in theory, only sensitive to hydrogen atoms, there are some effects from other materials; therefore, calibrations using job materials are also required. With proper calibration, the cement gage is capable of determining cement content of fresh concrete to within approximately 10-20 lb/cu yd (6-12 kg/cu m). The water gage is capable of determining water content to within approximately 2-4 lb/cu yd (1-2 kg/cu m). From the predicted standard errors in water and cement contents, an error in calculated water-cement ratio of as much as 0.03 may be expected. As entrained air decreases concrete density, which appears to have an effect on the gage response, it is necessary to control air content with the gage within very close tolerances (+ or - 1%) in both lab calibrations and field work.