Generic Tree Flora of Madagascar. George E. Schatz (foreword in English by Peter R. Crane and Peter H. Raven; in French by Albert Randrianjafy and Philippe Morat). Royal Botanic Gardens, Kew, United Kingdom, and Missouri Botanical Garden, St. Louis, 2001. 477 pp., illus. $39.95 (ISBN 1900347822 paper).

The literature on the botany of Madagascar is quickly becoming as rich as the flora itself. For anyone with the least interest in this island 400 kilometers east of Mozambique, this is a wonderful development. While the richness and uniqueness of the Malagasy flora and fauna have long been appreciated, it may have been the French naturalist Philibert Commerson who conveyed these ideas most elegantly when he declared that Madagascar was the “promised land” for naturalists, so rich that Linnaeus would have to devote 10 editions of his Systema Naturae to the island (Lasègue 1845).

How rich is rich? It is estimated that 8500 to 12,000 species of vascular plants, 70 to 80 percent of them endemic, occur in Madagascar (Humbert 1959, Dejardin et al. 1973, White 1983). To place this in perspective, the island is slightly smaller in area than the state of Texas, which has 4800 species of vascular plants, a mere 8 percent of them endemic (Correll and Johnston 1970).

The most ambitious project to catalog the vascular plants of Madagascar has been Flore de Madagascar et des Comores, which was initiated in 1936 by Henri Humbert shortly after he became professor of botany and director of the Laboratoire de Phanérogamie at the Muséum National d'Histoire Naturelle in Paris. Modeled on his immediate predecessor's Flore Générale de l'Indo-Chine (1907–1951), which coverd another region colonized by France, Flore de Madagascar has dominated floristic work in Madagascar for almost 70 years. Issued in fascicles by plant family, its treatments are comprehensive, with keys to genera and species, lengthy descriptions, full synonymy, illustrations (including maps), citation of specimens examined, and lists of vernacular names. Perhaps 60 to 70 percent of the vascular plant species of the island have now been treated. Earlier fascicles are out of date, and there is no key to the 189 families that were projected originally to comprise the finished flora; however, for systematic botanists trained to recognize plant families and familiar with the extensive but scattered literature on plant identification, the lack of such a key is not an impediment.

For further assistance, some of us have turned to René Capuron's modest but insightful Essai d'Introduction à l'Étude de la Flore Forestiére de Madagascar (1957). This mimeographed forest manual, virtually unknown outside of Madagascar, treats 95 families and about 450 genera. Those without systematic training or access to Capuron's forest flora, however, have had nowhere else to turn until now. The publication of Generic Tree Flora of Madagascar thus opens up broader access to the plants of Madagascar and gives not only botanists but also anthropologists, conservation biologists, ecologists, foresters, and zoologists an important tool for identifying Malagasy trees.

Defining a tree as “a woody plant at least (4–) 5 m tall, and/or with at least one vertical stem attaining 5 cm in diameter at breast height,” George Schatz proceeds to treat 107 vascular plant families and 490 genera (161 endemic, 329 nonendemic) native to Madagascar. (These genera are represented in Madagascar by 4220 species, 96 percent of them endemic.) This is a major subset of the vascular flora, which the most recent fascicle of Flore de Madagascar indicates includes 224 families (but see below). After very brief introductory remarks, Schatz immediately cuts to the chase with a series of keys that emphasize vegetative characters. This is followed by several lists of genera exhibiting distinctive characters (presence of spines, color of exudate, presence of glands, etc.). By far the greatest portion of the book is then devoted to treatments of families and genera. The volume concludes with a glossary (partially illustrated) of botanical terms and with indexes to scientific names, vernacular names, and illustration sources (including some original illustrations). Each of the family treatments begins with a statement concerning global distribution and extent. A morphological description is provided for those families (54 of 107) that have two or more native or naturalized genera in Madagascar. References are cited to relevant fascicles in the Flore de Madagascar and, if available, to revisions and monographs of constituent genera. Dichotomous keys to genera are elaborated. Each generic entry in the tree flora includes not only the author and place of publication but also synonymy as it applies to Madagascar. Global distribution is indicated (for the nonendemic genera), and estimates or statements concerning the number of species are made. Following a morphological description based principally or wholly on Malagasy material, distribution within Madagascar is given in the context of the five bioclimatic regions recognized for the island: humid, subhumid, montane, dry, and subarid. Vernacular names are also recorded; these are not Malagasy names for genera, but rather Malagasy names for species in a particular genus.

Three families (Euphorbiaceae, Fabaceae, and Rubiaceae) account for one-third of the genera treated in Generic Tree Flora of Madagascar. Here Schatz's contribution is especially noteworthy, since of these three families only one has been described in Flore de Madagascar (in one part of a projected twovolume treatment of the Euphorbiaceae). Neither the Fabaceae nor the Rubiaceae has been treated in the same series, but soon after Generic Tree Flora of Madagascar was published, a monograph of the Malagasy Fabaceae by Du Puy and colleagues (2002) appeared.

Circumscriptions of vascular plant families are presently in a state of flux, changing with the accumulation of new molecular data. This presents enormous challenges to anyone writing a flora such as this one. Schatz made the wisest choice that he could and adopted the family concepts of the Angiosperm Phylogeny Group (Stevens 2003), which are not static. Consequently, serious users of the tree flora will have to begin penciling in corrections as soon as they purchase a copy of the book. As an example of how complicated this may become, the genera Androya, Buddleja, and Nuxia are placed in the Buddlejaceae in Generic Tree Flora of Madagascar. The same genera were treated as Loganiaceae in Flore de Madagascar. Now, in the Angiosperm Phylogeny Group system (Stevens 2003), Androya is allied with Myoporaceae, Buddleja is placed in Scrophulariaceae, and Nuxia is in Stilbaceae. Scrophulariaceae, but not Myoporaceae or Stilbaceae, is mentioned in Generic Tree Flora of Madagascar. Myoporaceae and Scrophulariaceae, but not Stilbaceae, are projected to be among the families treated in Flore de Madagascar.

Generic Tree Flora of Madagascar departs from Capuron's forest flora in associating 3000 vernacular names with the 490 genera treated. Capuron (1957) felt that while such names might give useful clues to identification, they just as easily could confuse, since one vernacular-name could apply to more than one species. His example was the term hazomena, which was applied to any woody plant with reddish wood. Capuron (1957) cited three unrelated genera with this common name; Schatz lists nine! For most of us, Malagasy is not a familiar language, and providing lists of Malagasy names without explanation (as is also done in Flore de Madagascar), while it has some merit, is not especially useful. Apart from the possibility of confusing unrelated taxa, intriguing information can be lost by simply compiling lists of names. As an example, vavaporetaka, which is given in Generic Tree Flora of Madagascar as one of many common names of Melanophylla Baker (Melanophyllaceae), is in fact applied only to M. alnifolia Baker, and as reported by Dan Turk in A Guide to Trees of Eastern Madagascar and Ranomafana National Park (manuscript on file at the library of the California Academy of Sciences), it means “beak of the poretaka,” which is a local name for the Madagascar brush warbler (Nesillas typica). The inside of this bird's mouth is said to be yellow, the same color as the wood of M. alnifolia. This admittedly has little impact on identifying or understanding the biology of these trees, but its omission does shortchange the linguist, the ethnobotanist, or the anthropologist who may want to delve into the folk taxonomy of the different ethnic groups residing in Madagascar.

There is no doubt that Generic Tree Flora of Madagascar is an extremely important tool for identifying the Malagasy flora and that it is likely to have a beneficial and lasting impact on the study of plants (and animals) in Madagascar. One hopes that as our knowledge of vascular plant phylogeny stabilizes and as scores of scientists and students use and test the keys and descriptions in Schatz's tree flora, these changes, corrections, and comments will be conveyed to the author. Important and useful works tend to have long lives and many editions, and Generic Tree Flora of Madagascar is certain to be one of them.