The instrument and data reduction techniques use in obtaining remote measurements of carbon monoxide during the 1979 Summer Monsoon Experiment are described. The form of the signal function (the variation of signal with altitude) and the impact of variation in the vertical distribution of carbon monoxide are discussed. Estimates of the experimental accuracy are made both by assessment of error sources through the use of numerical simulations and by comparison with concurrent measurements made by means of gas chromatography. It is found that the radiometric measurements tend to be about 9% lower than the direct measurements and to have a precision of about 8%.