Fluctuating asymmetry (FA) is the unsigned difference between the left and right sides for some trait on an organism with no average difference between the sides. Because FA is assumed to reflect differences in development but not in genetic effects, it is often used as a measure of the developmental stability of an organism. The genetic properties of developmental stability, such as the heritability and correlations among characters, are of some interest, because the evolutionary future of developmental stability in populations depend on these quantities. As such the heritability of FA has often been measured. This paper notes a substantial bias in the estimation of the heritability of developmental stability; FA is a trait with very low repeatability and as such is only a weak estimation of the developmental stability of an organism. The biases are correctable, though, from a knowledge of the coefficient of variation of FA in the population being measured. This paper demonstrates why the heritabilities of fluctuating asymmetry are always so small in well designed experiments and suggests more accurate measures of the genetic properties of canalization. Furthermore, the correlation of FA of different traits in the same individuals is always low, which is inconsistent with the mechanisms proposed for the development of FA. Because of the low repeatability of FA, these correlations are expected to be low.