Quantitative spectrophotometric measurements of cathodoluminescence (CL) intensities of calcite cements from the Mississippian Lake Valley Formation are compared with Mn and Fe electron microprobe analyses of the same calcites. CL intensities of these calcites show excellent positive correlation with Mn concentrations at approximately constant Fe/Mn ratio and within narrow ranges of Fe contents (r = 0.76 to 0.97). Our data suggest that Fe/Mn ratios may control the maximum CL intensity of calcite. Below this maximum, at constant Fe/Mn ratios, CL intensity can vary significantly and is controlled mainly by absolute concentrations of Mn and Fe. CL intensities are therefore a function both of absolute concentrations of Mn and Fe and of Fe/Mn ratios. The data further suggest that CL intensities in calcites with low Fe concentrations (< 1,000 ppm) are more sensitive to Mn activation than to Fe quenching. CL spectra on the calcites show no evidence of rare earth element activators. Visual estimates of CL intensities of these calcite cements compare moderately well with intensities measured with the spectrophotometer. Spectrophotometrically measured CL intensities of calcites in any single visually estimated category (bright, moderate, dull, non-luminescent) overlap with those from adjacent categories. Thus spectrophotometric measurements are distinctly superior to the visual estimates. Our data indicate that calcites with more than about 100 ppm Mn will luminesce even in the presence of moderate amounts of Fe. In the absence of detectable Fe, calcites with about 150 to 650 ppm Mn have moderate CL, and those with Mn contents greater than about 700 ppm have bright CL. In the presence of detectable Fe, the amount of Mn required to impart moderate or bright CL increases with Fe content. Furthermore, Fe does not extinguish Mn-activated CL up to about 10,000 ppm Fe, the maximum of our data. These results are in general agreement with the electron microprobe study of Grover and Read (1983), in partial agreement with the data of Frank et al. (1982), and in poor agreement with the data of Fairchild (1983). Considering the poor precision of the electron probe at low element concentrations (< 100 ppm), our data are reasonably consistent with the published ion microprobe and atomic absorption data (Mason 1987; ten Have and Heijnen 1985; Richter and Zinkernagel 1981) which imply that Mn concentrations in the 30 to 50 ppm range are sufficient to activate cathodoluminescence in calcites.