Ecosystem Function in Heterogeneous Landscapes is the product of a Cary conference held at the Institute for Ecosystem Studies in Millbrook, New York, in 2003. Like other edited conference volumes (Likens 1989, Canham et al. 2003, etc.), it is long (489 pages) and broadly based, with the topics of its 24 chapters ranging from the theoretical and the synthetic to applications such as conservation planning.

I have to admit a certain prejudice against conference volumes of this sort, which tend to be lengthy and cannot avoid disparities of style and focus. One obvious manifestation of this disparity is the great variation in the number, quality, sophistication, and usefulness of the figures used to illustrate the different chapters. My preference would be a “distilled wisdom” version with half the length of this volume and a synoptic approach that eliminates both repetition and contradiction.

Although the jacket blurb hints at a groundbreaking synthesis, the book is more a compilation of current thinking and the present state of research in a number of related fields, including ecosystem science, landscape ecology, and conservation biology. The format, however, makes it difficult to compare chapter material, and it is not easy to follow the thread of a single concept throughout the volume. A flowchart or map of the relationships among the contents of the chapters might have been a useful addition.

That said, I commend the editors for their introductions to each section, which identify the common themes in all of the contributions. This is essential, given the broad range of ecosystems under discussion, from freshwater and oceans to cities and forests. The major theoretical component, which the editors emphasize, is the significance of the configuration of landscape elements, not merely the composition of the landscape, for the functioning of these heterogeneous systems: the flows of ecological processes are facilitated or inhibited by the landscape's configuration.

Although it is useful to have much of the current thinking on this topic collected in a single volume, I was left with the impression that a lot more thinking still needs to be done. Perhaps the most important challenge for this area of study is to develop conceptual and analytic frameworks that take into account the common elements of the broad range of systems studied. The spatial and temporal heterogeneity of ecological systems “nearly always affects processes and functions in ecosystems, and in diverse ways” (p. 414). The strength of this book is that it presents this diversity comprehensively, but diversity is also the book's weakness, inasmuch as general principles that unify the “diverse ways” are not presented.

Also missing from this collection are the theoretical tools needed to facilitate the research, even though these are available elsewhere. The concept of networks is mentioned in more than one chapter (pp. 39–41, 455–456), but it is never fully developed, and its parent concept, graph theory, does not even appear in the index.

This is a simple yet powerful body of theory in which spatial units are depicted as points (or nodes or vertices), with the connections and relationships between them depicted as lines (or arcs or edges); the points may have quantitative or qualitative characteristics, and the lines may have directions and other properties, such as rates. Graph theory has already been used in a variety of ecological studies (Dale 1977, Ricotta et al. 2000, Urban and Keitt 2001, Proulx et al. 2005, among many), and the subjects covered in this book would have benefited from its application. Discussions sometimes touch on the ideas of graph theory (for example, in reference to “point” processes at particular locations, with flows of energy, materials, and information between them [figure 2.1]), but it could be argued that the raster-based conceptual model depicted in figures 2.1, 2.2 and 2.3 will overly constrain developments in this area. A more open and flexible conceptual model will be important to the study of the function of heterogeneous landscapes, allowing all of the effects of the mechanisms of configuration, not just compositional differences, to be fully realized.

The book's contributors are to be credited for taking on such a difficult and critically important topic. Nonetheless, the current state of research in this field leaves room for considerably more work. I look forward to future developments in this area and to more distillation of the collective wisdom. The effects of spatial heterogeneity cannot be ignored, and they must be accounted for in ways appropriate to the system under study.