ABSTRACT. We show that smoothed versions of the high-resolution calibration curve should be used when 14C ages are calibrated with large (> -3014C yr) measurement errors (represented by standard deviation am) or are mixtures of elements of variable age (natural sample error with standard deviation a,,). The degree of smoothing should agree with the standard deviation of total sample error, o,, the square root of the quadratic sum of om and a,,. However, in most cases, a, is not well known, especially due to difficulties in quantifying aQ. We present an inverse method that gives a measure of mean a, for different materials that are widely used in (conventional) '4C dating. Calculations with large (>100) data sets of wood, charcoal, ombrotrophic peat and minerotrophic peat/gyttja samples indicate that a, of such materials is generally much larger than previously assumed, mainly because of large values of a,,. This means that particularly in organic deposits, strongly smoothed calibration curves should be used where medium-term 14C variations (wiggles) are completely straightened. This has especially major consequences for calibrating 14C histograms for natural 14C variations. We conclude that 14C histograms consisting of samples of organic deposits do not require correction for medium-term 14C variations and that uncalibrated 14C histograms need not be as suspect as is usually believed.