The standard deviation [sgrave] is known to be an absolute measure of Gaussian (G) distributions, because it (or multiples of it) always determines a constant fraction of the material. However, this is not true for distribution functions other than Gaussian. The weight fractions corresponding to weight standard deviations (±[sgrave]w) of Schulz-Flory distributions depend on the polymolecularity index xw/xn, approaching a limiting value of W(xw± [sgrave]w) = 86.5 for xw/xn→∞. The use of weight standard deviations is meaningless for generalized logarithmic normal (LN) distributions if xw/xn52. The weight fractions W(Xn± [sgrave]n) around the number-average degree of polymerization of G and LN distributions first go through a maximum before decreasing with increasing xw/xn. The weight fractions W(xn± [sgrave]n) decrease steadily with higher xw/xn.