Book Reviews: Ecological

Null Models in Ecology. Gotelli, Nicholas J. and Gary R. Graves. 1996. ISBN 56098-657-3 (paper US$30) 368 pp. Smithsonian Institution Press, P.O. Box 960, Herndon VA 22070
- Although collecting empirical data is an important function of science, the greatest advances in our understanding of the world are made by falsifying well-formulated null hypotheses. The science of ecology generates mountains of empirical data, but falsifiable hypotheses have been few and far between. Null models are statistical predictions or computer simulations against which empirical data can be tested for evidence of pattern, i.e. null models can provide falsifiable null hypotheses. Gotelli and Graves provide an intellectual history and philosophy of null models in Chapter 1, review their application in select fields of community ecology (species diversity, relative abundance, niche overlap, temporal niches, size ratios, co-occurrence, species-area relationships, biogeography, and food webs) in Chapters 2-10, and close with an Epilogue which summarizes the pitfalls in null model construction that should be considered when critically reading the literature or formulating a null model of your own. The book provides a skillfully written, thoroughly referenced, and remarkably evenhanded analysis of the literature in one of the most contentious fields of ecology. It should be read by every serious ecologist.

Before proceeding with a review of what this book is about, I should mention what the authors chose to omit. To quote the Preface:

We chose to limit our discussion of null models to community-level processes. Consequently, we have not discussed null models of population dynamics, animal behavior, landscape ecology, ecosystem modeling or phylogeny. We have also omitted purely statistical issues such as bootstrapping and jackknifing. Although we occasionally illustrate the probability equations used in these analyses, this is not a cookbook or workbook of null models...A second null model book needs to be written, one that contains null model software that would allow researchers to analyze data more easily using these tests. For the time being, null model analyses are accessible only to those with some programming expertise...

Even within these self-imposed limits, Gotelli and Graves find no shortage of literature to review. For those seeking an introduction to mathematical modeling applications in ecology, I suggest A Primer of Ecology (Gotelli 1995; ISBN 0-8793-270-4); those interested in an applied null modeling text should, I suppose, patiently await Null Models II.

In Chapter 1, the authors define their subject, tackle the history and controversy surrounding the use of null models, and, in so doing, demonstrate their own skill at fairly representing adversarial positions. By their definition: "A null model is a pattern-generating model that is based on randomization of ecological data or random sampling from a known or imagined distribution. The null model is designed with respect to some ecological or evolutionary process of interest. Certain elements of the data are held constant, and others are allowed to vary stochastically to generate new assemblage patterns. The randomization is designed to produce a pattern that would be expected in the absence of a particular ecological mechanism." Such models make a clear distinction between the pattern observed and the process or processes which may have generated the pattern. Null models also rely on the principles of parsimony and falsification, and thereby maintain the possibility of no effect. Critics of the null model approach charge that falsification is meaningless if the null hypothesis is not properly constructed, and that "reliable null hypotheses may be impossible to construct, because we cannot generally deduce the nature of expected patterns that would emerge in the absence of any given biological process." Rather than throw the baby out with the bath water, Gotelli and Graves argue that null models have forced both experimental and theoretical ecologists to articulate and evaluate their hypotheses, thereby increasing the rigor and lucidity of ecological thought.

Chapters 2 through 10 share a structure in which the authors review and critique important contributions to the development of the subject, highlight the strengths and weaknesses of each position in the ongoing debate, and close with a paragraph entitled "Recommendations." With 45 pages of citations, Gotelli and Graves are comprehensive in their review of the literature through 1993, providing a coherent and accessible introduction to the field for any student of ecology. Their noteworthy recommendations point to specific exemplars in the literature, suggest the most profitable directions for future research, and provide food for critical thought.

The Epilogue raises some neglected problems in the construction of null models: i.e. the construction of source pools, species taxonomy, sexual dimorphism and geographic variation of body size, anthropogenic extinctions, and data quality. In essence, it is difficult or impossible for the quality of the model output (predictions) to exceed the quality of the input (model structure and parameterization). Rigorous attention must be paid to the assumptions underlying model structure and the data used in parameterization for the model predictions to be meaningful. Now that second- and third-hand data are available with unheard of ease over the Internet, and the software hurdles to writing and running models have been lowered to the point where everyone can enter the race, it is likely that we will see increases in both the use and misuse of the null model approach. It is the authors' hope that "the suggestions in this Epilogue and the recommendations at the end of each chapter will at least serve to increase the quality of future null model studies." I believe they will. - Jonathan P. Frye, McPherson College, McPherson, Kansas

The Natural History of Pollination Proctor, Michael, Peter Yeo, and Andrew Lack. 1996. ISBN 0-88192-352-4 (hardcover US$42.95), ISBN 0-88192-353-2 (US$24.95 paperback). Printed and typeset in Great Britain by The Bath Press, available from Timber Press, Inc., The Haseltine Building, 133 S.W. Second Ave., Suite 450, Portland, OR 97204. - The forerunner of this book, Proctor and Yeo's 1973 The Pollination of Flowers, No. 54 in the New Naturalist series, was motivated by a desire to draw together the European literature on pollination ecology. For the new edition, which includes literature published as recently as 1995, Michael Proctor, an Honorary research Fellow in the Department of Biological Sciences at the University of Exeter and a Fellow of the Royal Photographic Society, and Peter Yeo, a former taxonomist at the University Botanic Garden, Cambridge, have been joined by Andrew Lack, lecturer in biology at Oxford Brookes University. Lack is well-known to ecologists for his work on competition for pollinators in Centaurea, pollination in British chalk grassland and sand-dune communities, as well as several studies of pollination in tropical species.

The book contains 16 chapters, beginning with the history of pollination biology (among the best chapters in the book), the structure of flowers and inflorescences (for "the general reader"), the main groups of pollinators, including sections on insects, birds, bats, and other vertebrates, and wind and water pollination. Subsequent chapters are devoted to orchid pollination, pollination involving deception, and pollination involving egg laying by pollinators in flowers. Breeding systems and crop pollination are covered next, while the last three chapters deal with pollination through time, pollination in plant communities, and "Flowers, Genes, and Populations." This last includes sections on male vs. female function of flowers, sexual selection, and genetic consequences of mate choice that in my view should all have gone into the "breeding system" chapter. The bibliography comprises 1192 references of which 65% are more recent than 1973.

The Natural History of Pollination is well written and covers a lot of ground. It contains much more information (in the sense of specific examples or case stories) than any other available book on pollination. Moreover, pollination biologists, who have loved the book's predecessor for its outstanding black-and-white photographs of flowers, will find that the new book has (even) more pictures; there are an additional eight color plates, with four to eight photos each, all of exceptional quality. The black-and-white drawings illustrating insect mouthparts or flower structure in some cases are reduced rather drastically to fit the 17.5 x 11.3 cm type area of the book, but are still perfectly readable. (Two small quibbles: Fig. 7.19 of a chalcid wasp with a pollinium of musk orchid attached to each front femur is incompletely labeled. It shows a "pm", which must mean pollen mass, and a "vd", which I have yet to figure out. In Fig. 7.24 of a fresh and a triggered Pterostylis rufa, photos a and b have been switched.)

Being particularly interested in Melastomataceae, I was surprised to find a flower of Tibouchina (Fig. 10.1) under the chapter heading "Deception and Diptera: Sapromyiophyly." Of course, most of the chapter deals with Aristolochiaceae, Araceae, and orchids that are fly-pollinated, but it starts out by discussing deception in melastomes, Cassia, and other bee-pollinated pollen flowers. Pollination by pollen-collecting bees capable of extracting pollen from anthers by high-frequency vibrations transmitted to the stamens is treated at length in chapter 6 (pp. 179-180) under the header "flowers providing pollen as the sole reward." Cross reference to this sections would have been helpful. Also, it is by no means clear that Melastomataceae and Cassia are good examples of deceit. What constitutes deceit in pollination is a difficult subject, partly because of our incomplete understanding of insect sensory systems or learning processes and partly because net gains and losses in mutualisms shift constantly, which is what drives their evolution. Proctor, Yeo, Lack describe as deceit the fact that Melastomataceae and some other plants have two types of stamens, one of which is assumed to provide food while the other dusts visiting insects with "functional" pollen. This hypothesis goes back to the Fritz Mueller (1873, 1883) and Hermann Mueller (1881, 1883) although they are rarely credited for it. Wolfe et al. (1991) tested the hypothesis in Solanum rostratum, a buzz-pollinated species with dimorphic anthers held to be feeding and fertilizing anthers, respectively, and found that pollen from each set of anthers reached the stigma in equal amounts. A further example of how difficult it is to neatly assign pollination systems to either the deceit category or the reward category is provided by Proctor, Yeo, and Lack's treatment of brood-site pollination in two places: first in chapter 10, under the subtitle "brood-site imitation" (pp. 295-305) and then again in chapter 11 under "brood-site pollination" (pp. 311-320). Surprisingly, Pellmyr and Huth's (1994) paper on how overexploitation is prevented in one of the best studied brood-site pollination system is not cited in the long section on the Tegeticula - Yucca interaction.

The coverage of the European literature is excellent; a few obvious omissions are the series of papers by Y. Nyman on the pollination of Campanula (see Nyman 1993) and of A. Erhardt on that of Dianthus (one of his early papers is cited, however). Because of the authors' effort to include recent research foci, the treatment of breeding systems and incompatibility has changed considerably from the previous incarnation of this book, and I found this section (occupying 29 pages) particularly well-balanced. Phylogenetic viewpoints, however, are absent. Thus, macroevolution is alluded to only obliquely, for example, where it is stated that the androdioecious species of Datisca are derived from the dioecious ones, but the critical phylogenetic assessment of that hypothesis by Swensen et al. (1994) is not cited. Bee keeping, crop pollination, and pollination in ecological communities, on the other hand, receive ample treatment as is appropriate in a book conceived for the New Naturalists rather than plant systematists.

Clearly, The Natural History of Pollination is invaluable as a source reference and for its succinct summary of knowledge on the pollination and breeding systems of plants. It is also an exceptionally beautifully illustrated book. - Susanne Renner, Department of Biology, University of Missouri, St. Louis

Literature Cited

Mueller F. 1873.
(letters). In: A. Moeller (editor), Fritz Mueller. Werke, Briefe und Leben. vols. I-III. Fischer Vlg., Jena.
-. 1883.
Two kinds of stamens with different functions in the same flower. Nature 27: 364-365.
Mueller, H. 1881.
Two kinds of stamens with different functions in the same flower. Nature 24: 307-308.
-. 1883.
Arbeitstheilung bei Staubgefaessen von Pollenblumen. Kosmos 13: 241-259.
Nyman, Y. 1993.
The pollen-collecting hairs of Campanula (Campanulaceae). II. Function and adaptive significance in relation to pollination. Am. J. Bot. 80: 1437-1443.
Pellmyr, O, and C. J. Huth. 1994.
Evolutionary stability of mutualisms between yuccas and yucca moths. Nature 372: 257-260.
Swensen, S. M., Mullin, B. C., and M. W. Chase. 1994.
Phylogenetic affinities of Datiscaceae based on an analysis of nucleotide sequences from the plastid rbcL gene. Syst. Bot. 19: 157-168.
Wolfe, A. D., Estes, J. R., and W. F. Chissoe III. 1991.
Tracking pollen flow of Solanum rostratum (Solanacaee) using backscatter scanning electron microscopy and x-ray microanalysis. Am. J. Bot. 78: 1503-1507.

Seasonally Dry Tropical Forests Bullock, Stephen H., Harold A. Mooney, and Ernesto Medina, eds., 1995. ISBN 0-521-43514-50 (cloth US$ 95) 450 pp. Cambridge University Press, 40 West 20th St., New York NY 10011 - Tropical dry forest habitat has been the most exploited by humans, with much of it having been developed or converted to agriculture and pasture. Dry forests occupy more area than wet forests, yet very little of the dry forest remains, and only a small portion of that is actively being conserved. Public and scientific attention has been long directed toward the imperiled rainforests, but only recently has the plight of tropical dry forests been brought to the attention of the scientists and others concerned with conservation of biodiversity. Seasonally Dry Tropical Forests provides the educated reader with a wealth of information about these widespread and important habitats. This comprehensive volume is mostly the result of a symposium held at the Estacion de Biología Chamela in Jalisco, Mexico. Chapters are authored by authorities from six countries, and most of the chapters consider data from a diversity of tropical dry forest sites. The book covers topics at many levels: autecological (drought responses, ecophysiology), population (phenology, reproduction, herbivory), community (biogeography, diversity, ethnobotany), and ecosystem (climate, gas emissions and global effects).

A number of chapters describe forests in different geographic locations in great detail: Murphy & Lugo examine Central America and the Caribbean; Sampaio examines caatinga in Brazil; Menaut, LePage & Abbadie review the African savanna and dry forest; Rundel & Boonpragob discuss the seasonal monsoon climate in Thailand, contrasting it with India. Other chapters cover particular topics, such as Graham & Dilcher's review of fossil plants in N. Latin America and Bye's discussion of ethnobotany in Mexico. Several chapters deal with biodiversity: Gentry compares 31 forests with data from transects discussing taxonomic trends and endemism; Ceballos surveys vertebrates of Mexico and Central America; and Medina compares life form diversity in dry forest with other forests. Some of the chapters are more mechanistic and physiological: Holbrook, Whitbeck and Mooney review drought responses of trees, including effects on structure, physiology, and phenology; Martinez-Yrizar reviews biomass production and primary productivity, including litterfall and litter decomposition; Jaramillo and Sanford discuss nutrient cycling; and Cuevas reviews underground phenomena (soil respiration, roots, microbes, and their relationships); Matson and Vitousek discuss nitrogen trace gas emissions from their Mexican study site and contributions to the global balance. The final chapter by Maas on conversion of tropical dry forest reviews effects of deforestation on many ecosystem processes.

I found two chapters to be exceptional: Bullock's on plant reproduction, and Dirzo & Dominguez on plant/herbivore interactions. Bullock examines mating patterns of plants in a comprehensive survey of diverse dry forest sites. He includes an interesting figure graphing the index of self-incompatibility for different sites and different plant genera. Extended flowering is less common in dry forest plants than wet forest plants, and flowering is shorter in duration and more synchronous within and among species. Dirzo and Dominguez examine the effects of animals on plants, including folivory, seed predation, and effects of herbivory on plant population biology (fruit production, seeds and seedlings, fruit maturation, dispersal and establishment. They demonstrate that, in general, single measures of herbivory underestimate leaf area loss by 50% compared with long-term measures in which leaves are marked and re-measured. This differs among species and is especially dramatic for species with high levels of damage. There is really something for everyone in this book. The chapters I found to be the most interesting and exciting happen to be in my own interest areas, and I suspect that might be true for most readers! Overall, the volume seems well written and edited, with each chapter offering new insights. I think this book will be of great interest to not only tropical biologists, but to any biologist working in communities where there is seasonality in rainfall and deciduous vegetation. This book is an important reference for anyone interested in biological diversity, and is especially useful because of the multiple comparisons among distant dry forest sites. I recommend it to professors, graduate students, and undergraduates. It will be useful in courses in plant ecology, biogeography, tropical botany or ecology, and conservation biology. - Suzanne Koptur, Department of Biological Sciences, Florida International University, Miami, FL

Return to: | Table of Contents || Book Reviews Home Page || Previous Review || Next Review