Using size analyses of ancient sandstones, thin-section-to-sieve and sieve-to-thin-section empirical conversion equations have been derived, via linear regression analysis, for several graphical size distribution statistics: median, graphic mean, inclusive graphic standard deviation, graphic standard deviation, inclusive graphic skewness, graphic skewness, and graphic kurtosis. The correlation coefficients for each of these are respectively .958, .944, .705, .786, .567, .234, and .316. Conversions of high accuracy are possible only for the median and graphic mean. The thin-section-to-sieve conversion equations for these two parameters are, median: Y(sieve)= .121 4 + 1.030 . X(thin-section) mean: Y(sieve) = .227 + .973 . X(thin-section) Thin-section-to-sieve empirical conversion equations, of generally high accuracy, have also been derived for the following cumulative percentiles: 2, 5, 9, 16, 25, 36, 50, 64, 75, 84, 91, 95, and 98. These may be used in the conversion of whole distributions.