The success of Paradigms Lost, Casti's 1989 survey of science's most compelling questions, made this successor and its title nearly inevitable. Unfortunately, the new book fails both as a complement and as a contrast to the earlier work. Like its predecessor, it transforms multifaceted scientific inquiry into the motif of an adversarial courtroom battle--a device that, though useful for framing the discussion and possessed of some entertainment value, inevitably produces a distorted picture of the evolution of scientific thought. Scientific progress is continually punctuated by breakthroughs that vault new areas of inquiry into the foreground and relegate others to the background realm of apparently resolved questions. Because of these shifts in scientific thought, not all the topics important in 1989 merit the same level of attention in 2000--and yet Casti insists on revisiting them. As a result, although he devotes a dull chapter to recapitulating old arguments about artificial intelligence, for example, he admits in those pages that ""nothing of eyebrow-raising substance has really changed in the AI debate since the mid-1980s."" The book's strongest chapter is its last, which discusses the peculiarities of the quantum mechanical world. Casti's fascinating discussion of new insights into the wave-particle duality of matter and energy might make readers wish he had written a book about new paradigms for a new millennium instead of this sometimes contrived and oft-contorted sequel. Illustrations. (Mar.) FYI: Also in March, Wiley will release the second volume of Casti's survey of math in the last century, Five More Golden Rules: Gordian Knots, Secret Codes, and the Importance of Being Nonlinear--More Great Theories of 20th-Century Mathematics, with illustrations ($27.95 256p ISBN 0-471-32233-4).