New York, NY: Wiley. xix, 719 p. £ 70.00 (1995).

This is the third volume of the new edition of the four volumes sized “Distributions in statistics.” Similar to the first two companion volumes, this revised second part of “Continuous univariate distributions” (CUD), for the review of Vol. 1 see

Zbl 0811.62001, has been expanded more than twice in size (from 306 to 719 pages). The distributions covered in this volume are: Extreme Value, Logistic, Laplace, Beta, Uniform, $F$-, $t$-, noncentral $\chi\sp 2$-, noncentral $F$-, noncentral $t$-distributions, distributions of correlation coefficients , and lifetime distributions. The chapter on Quadratic forms has been postponed to the projected revised edition of “Continuous Multivariate Distributions.” The length of all chapters has been substantially increased (about doubled). As in the first volume of CUD, the first chapters are more or less organised according to definition, genesis and historical remarks, moments, estimation of parameters, relations to other distributions. The $F$-, $t$-, and $\chi\sp 2$- distributions are of less interest in modelling, so estimation is almost no topic in these chapters.
Many new examples of applications of distributions are included. Also numerous results relating to approximations are included. The number of references increased almost threefold (to over 2100 items). They are given at the end of the chapters.
In summary, it can be said again: This second edition of “Distributions in statistics” will serve as the primary source for statistical distributions for a long time.