No wonder, to begin with, that speculations on fossil vertebrates stimulated some of the best pieces of evolutionary biology. The quality of the fossil record, especially the wealth of information we may often recover from teeth or skeletal remains, together with the subterraneous but generally unconfessed interest for a large and diverse taxon to which we ourselves belong, set the background for some of the most enjoyable and stimulating readings in our ®eld. Think, for instance, of George Gaylord Simpson's Tempo & Mode in Evolution (Simpson, 1944), a book of such an importance as to justify a jubilee (Fitch & Ayala, 1995) 50 years after it appeared in print, or of his later The Major Features of Evolution (Simpson, 1953). No wonder, then, that Robert L. Carroll's book is `dedicated to the work of George Gaylord Simpson, who brought palaeontology into the evolutionary synthesis and established the study of large-scale evolution on a strong scienti®c footing.' Carroll's book, however, is nothing of a celebrative or historical book. It is a sound, well-argued monograph on evolution, using ± as much as Simpson also did ± fossil vertebrates as the main, but not exclusive model for describing and interpreting evolutionary patterns over the scale of geological time. Carroll deals at length with several of the most controversial points in palaeobiology and evolutionary biology at large: topics obviously familiar to any reader acquainted with the evolutionary biology literature of the last couple of decades, but all too often approached through a few selected examples and in terms of philosophical controversy. This was true even of the valuable stimulating pages of the late Antoni Hoffman (1989). Here, on the contrary, the reader is offered a broad range of well-documented examples, and Carroll's conclusions often verge on the pluralistic: in the Preface, he anticipates the general conclusions of his study, which will `differ from those of Darwin, the modern evolutionary synthesis, and the theory of punctuated equilibrium in being much more pluralistic', which is repeated at the end: `[P]atterns, rates, and controlling forces are much more varied than had been conceived by either Darwin or Simpson.' (p. 390). Carroll must be congratulated on his superb ability to provide the reader with a digest of precious information on genetics and developmental biology and in carefully distinguishing between the problems where these disciplines may already help understanding the patterns from the fossil record and other still grey areas, where a similar understanding is likely to be developed, but some key information is still lacking. I dare to say that Carroll succeeded in grasping and summarizing genetics and developmental biology for palaeobiologists better than most geneticists and developmental biologists have been able to do with palaeobiological evidence. A die-hard tradition in palaeobiology is ®lling pages with statistics of ®rst occurrences, extinctions and the like. As poignantly demonstrated by Andrew Smith and the late Colin Patterson (e.g. Smith & Patterson, 1988; Smith, 1994), all those statistics are prone to very strong in uence from subjective taxonomic decisions. In Carroll's book there is still a place for that kind of numerical exercises, but not too much. Instead, the author embraces the cladistic method of counting changes of character states along the branches of a cladogramme: a beautiful example of this kind of analysis provides the backbone for his analysis of the evolution of mosasaurs. What about macroevolution? Carroll soberly rules out the existence of evolutionary mechanisms peculiar to it, but goes on commenting that `On the other hand, macroevolutionary patterns cannot be directly extrapolated from microevolution because there are additional external forces that are undetectable over short time scales and whose force, frequency, and duration are unpredictable. These include plate tectonics, with corresponding changes in climate and routes of dispersal, and mass extinction' (p. 392). To sum up, if Simpson's The Major Features of Evolution `was the latest attempt by a vertebrate palaeontologist to provide a uni®ed picture of evolutionary theory' (p. 60), now it has been replaced, in many ways, by Carroll's Patterns and Processes of Vertebrate Evolution. His last chapter includes a section entitled Agenda for the Future: `Counting taxa is never suf®cient to solve the important biological aspects of these events.' (p. 393); we need to differentiate between extinctions and pseudoextinctions, to look for processes underlying the patters, to incorporate into palaeobiology the new spectacular advances in developmental biology. But, to be sure, there is still scope for getting important lessons from the past.