A monograph aimed at providing a delineation of currently available modelling approaches and inferential methods for nonlinear repeated measures, whilst making the material accessible to a wide audience. "?gives a very well-written account of the field over the past few decades, focusing mainly on US work and including much of the author's own, plus a glimpse of the future. It fairly reflects that literature over the years in dwelling at length on certain computational methods for maximum likelihood estimation. There is a leaning toward biopharmaceutical applications, this being a field in which the authors are acknowledged authorities. There are no exercises, but enough detail of the methodology is given, together with helpful guidance on available software, to enable the keen novice to try his hand. ?--M. J. Crowder, Biometrics, September 1997