This extremely attractive book suffers from a problem that seems to have affected many recent conference proceedings emanating from North America: an excessive delay in publication. The meeting was held at the Royal Tyrrell Museum of Palaeontology in Drumheller, Alberta, in June 1986, and it has taken nearly five years to appear. In the interim, a much larger and more comprehensive book on a similar topic (The Dinosauria, edited by D. B. Weishampel, P. Dodson, and H. Osmolska, University of California Press, Berkeley, 1990) was conceived, written, and published, and indeed for only a slightly higher price, despite being three times the size! Nevertheless, the aims and contents of the two books are different. Dinosaur Systematics contains 22 chapters, covering all major dinosaur groups, and presented by 24 authors, mostly North American, but four British, one German, one Chinese, and one Australian. Although the chapters at first sight seem to be a motley collection of windows on the systematics of dinosaurs, there is a key theme: subspecific variation and morphometric approaches to its study.
Carpenter and Currie, in an introductory essay, show how dinosaur palaeontologists have changed their views to and from on interpretations of individual variation. To early workers, up to about 1860, such variation was expected, but the later Victorians favoured excessive splitting: every slight variation in a tooth or bone was justification for a new species of genus name. Today, the trend is to recognize the possibility of great morphological variation within a species, and to synonymize many earlier taxonomic names. But how can we assess which is the best approach?
Weishampel and Chapman look at material of the Late Triassic prosauropod dinosaur Plateosaurus: 22 species and nine genera reduce to one genus and perhaps two species. The technique of study was to use Principal Components Analysis to sort 52 Plateosaurus femura on the basis of 11 exclusive measurements. The first Principal Component represented size, and the specimens are said to separate into two clusters, although no test of separation is given, and the two 'clusters' overlap. Nevertheless, two morphs, possibly male and female are suggested.
Colbert provides a useful, but less quantitative, over-