“…The dependence of iv1/2 on a2C°i s shown in Figure 4. These results are given in terms of the fractional deviation of iv1/2 from W1/2,o» thus A U>1;2 u'l/2 W 1/2,0 ¡¿1/2,0 (30) These results indicate that when nonlinearity is slight (a2C°< 0.1) the effect on wA¡2 is about the same as the effect on the second moment, as would be expected since these are so closely related for a Gaussian peak. At large concave nonlinearity A", is larger than that in the variance.…”