SCIENCE'S COMPASS BOOKS: PLANT ECOLOGY To Emerge or Not to Emerge Susan J. Mazer w e often regard the life cycle of seed plants as straightforward and continuous. They germinate from seeds, grow, flower, export and re- ceive pollen, and produce new seeds. But an additional phase, seed dormancy, often lasts longer than the cumulative du- ration of all the other steps. This period, extending from seed maturity to germination, is the subject of Carol and Jer- ry Baskin's comprehensive overview, Seeds. Over the last three decades, the authors' names have be- come synonymous with de- tailed studies of seed germina- Subsequent chapters on seed banks and on the causes of intraspecific variation in germination behavior might ordinarily es- cape the attention of theoretical population biologists, but overlooking them would be perilous. Researchers have been slow to include parameters that reflect realistic seed bank behavior in mathematical mod- els of plant population dynamics. This is not surprising given the levels of inter- and intraspecific variation in the conditions necessary for germination, the proportions tion and dormancy. Seeds stands out among the numerous volumes on germina- tion because it is satisfyingly pervaded by the Baskins' understanding that, like any other trait, seed dormancy evolves. Recur- ring throughout the book is a clear evolu- tionary message: determining what makes seeds germinate (or not) requires knowl- edge of the ecological conditions that they normally encounter. By culling the results of thousands of studies covering over 3500 species, the Baskins show that the require- ments for breaking dormancy and for ger- mination can usually be unequivocally interpreted as adaptations that protect seedlings from exposure to the most pre- dictable environmental hazards. The authors begin by reviewing the ba- sic components of successful and unam- biguous germination experiments. They then offer detailed descriptions of the cur- rently recognized classes of seed dorman- cy. (Researchers working on species whose seeds have no special germination require- ments may be surprised to find that there are seven major classes of dormancy, the subdivisions within which yield a total of 18 more or less distinct behaviors.) Fol- lowing this introduction, three chapters survey hundreds of studies on the environ- mental factors known to break each kind of dormancv. Filled with brief summaries of primary research, these chapters are very dense. The author is in the Department of Ecology, Evo- lution and Marine Biology, University of Califor- nia, Santa Barbara, CA 93106, USA. E-mail: rnazer@lifesci.ucsb.edu of seeds germinating in re- sponse to particular environ- mental cues, and the number of times a seed may cycle be- tween dormant and non-dor- mant states. Although the prospect of assigning values to such parameters is intimidat- ing, realistic estimates are crit- ical for accurate predictions of population trajectories. Even if the Baskins' volume will pro- vide little comfort to those who develop or interpret popu- lation dynamic models of species-kith dormant seeds, it clearly identifies the pa- rameters (and the ranges of their values) that must be considered. Two well-organized chapters offering a geographical perspective document the tremendous variation in the germination ecology of sympatric species. These cover tropical, subtropical, temperate, and arctic and some aspects of their dormancy- breaking requirements. The authors' analy- ses of these tables reveal several intriguing patterns. For example, higher percentages of species with dormant seeds are found in plant communities that experience cyclical periods of cold or little precipita- tion. These and other patterns raise many tantalizing questions, but, curiously, no statistical analyses are reported to confirm their significance. Readers would have to return to the original tables to extract data for rigorous analysis; the data are ripe for such mining. In fact, multivariate analyses of the data in Seeds would help tease apart the associations between climate, vegeta- tion type, life form, and dormancy type. I was sorry to find that that the volume lacks such analyses. zones. For each of the zones' major vege- tation types, the Baskins discuss the ger- mination requirements of the different plant life forms (trees, shrubs, vines, and herbaceous species). In a separate chapter, they offer similar digests of the dormancy characteristics of plants with specialized life cycles (including parasites, orchids, aquatic plants, and halophytes). The book's final chapter begins to pro- vide the synthesis the authors, and proba- bly most readers, seek. Here the Baskins use one of the volume's major contribu- tions-the tables that collectively list hun- dreds of species, their dormancy types, The results of the authors' examination of the evolutionary history of dormancy are largely inconclusive. Because there are no unambiguous morphological indicators of most kinds of dormancy, the brief sum- maries of paleoclimates and the fossil his- tory of seed morphology provide little in- sight into the ecological factors involved in its evolutionary origin. It seems likely that most types of dormancy have evolved in multiple lineages, but determining the sequences and directions of dormancy evolution within seed plants will require detailed phylogenetic analyses that were beyond the scope of this volume. Those considering Seeds as the focus of, say, a graduate course should 1 note a few omissions. Perhaps the most glaring is the lack of sum- maries for all but the final chap- ter. The vast amount of informa- tion presented makes for dull reading when there are no over- riding themes with which to glean its significance. A glossary would also have been helpful, es- pecially given that the authors hope the volume will be used by newcomers to seed ecology. In fact, the authors hope to "(1) pro- vide people who are just begin- ning to learn about seed germination a comprehensive overview of seed germina- tion ecology, biogeography, and evolution and (2) give active researchers in the field a sense of what we know and do not know about these aspects of seed biology and thus what research needs to be done on them in the future." These aren't the usual 2 goals of a reference volume, and I doubt ' that this book will be read cover-to-cover 5 by even the most avid seed enthusiast. $ Nevertheless, it should be on the bookshelf 5 of any ecologist, evolutionist, or theoreti- 3 cian who needs to know h0.w seeds can be e P coaxed to get on with the rest of their life. a 15 JANUARY 1999 VOL 283 SCIENCE www.sciencernag.org