The cluster of recent papers in Systematic Biology on phylogenetic diversity and conservation (e.g., Hartmann and Steel, 2006; Lewis and Lewis, 2006; Minh et al., 2006; Steel, 2005) has highlighted the growing importance of phylogenetic perspectives on biodiversity conservation problems. This growth is evident also in the recent Society of Systematic Biologists symposium on “phylogeny and conservation” at the 2007 annual meeting in New Zealand, with talks covering a broad range of topics integrating systematics and conservation disciplines. That symposium was cosponsored by the new “Biogenesis” core program within DIVERSITAS (the international organization for biodiversity science). Biogenesis promotes the links between evolution, phylogenetics, and all aspects of biodiversity conservation. This reflects DIVERSITAS' recognition that within-discipline research efforts are not enough, and that synergies and synthesis must be found across the disciplines concerned with biodiversity conservation (e.g., economics, land-use planning).
Such efforts may help to provide some much-needed synthesis. Although there has been a lot of interesting work exploring the links between phylogeny and conservation, this can be characterized as a bits-and-pieces story, rather than as a coherent picture. My own favorite example of this fragmentation of phylogeny–conservation research is that, for a long time, there seemed to be a whole parallel universe of publications on phylogenetic-diversity measures in economics journals such as Land Economics and Quarterly Journal of Economics. Stories were re-caste, definitions sometimes were changed, and, although new insights emerged, they did not necessarily migrate beyond the readers of those journals. Now, the work of Mooers, Steel, and others, in Systematic Biology and elsewhere, is beginning to reconcile these various bits of research.
Against this backdrop, the book Phylogeny and Conservation is timely and welcome. In their introductory chapter, editors A. Purvis, J. Gittleman, and T. Brooks argue that there is much progress still to be made in developing the useful synergies between conservation biology and phylogenetics. Promoting such synergies is therefore the major rationale for pulling together this collection of contributed chapters, based on papers presented at a meeting organized by the Zoological Society of London. Although there appear to be no economists among the 20 or so contributing authors, the range of specialist interests and chapter topics has succeeded in bringing lots of different perspectives and ideas together in the same place. As part of a Cambridge University Press conservation biology series, this edited volume is a well-placed effort to give these issues a high profile. This series sets out to present and review active areas of conservation biology, in a way accessible to undergraduates and above. (This book is described within as number 8 in the series, although the list of published titles shows 9 books published prior to this.)
The structure of the book supports the goal of review and cross-disciplinary integration. It nicely divides into four sections (“Units and currencies,” “Inferring evolutionary processes,” “Effects of human processes,” and “Prognosis”) that capture issues ranging from measurement to human impacts. This organization helps to drive home the flow of ideas from theory to conservation applications. The assortment of topics, all collected together in one place, means that there is a significant opportunity also to provide some real synthesis in support of the emerging synergies. Here, I'll first provide an overview of the book's content and then comment on the book's overall degree of success regarding synthesis and synergies.
The first section, on units and currencies, provides well-written reviews and discussion about some fundamental but still contentious issues, including how phylogeny informs us about biodiversity units extending from features of organisms to the species level. As phylogenetic background, Sinclair et al. summarize aspects of molecular phylogenetics, including sequence alignment and the basics of phylogenetic inference. This presentation necessarily is brief, but it helps to satisfy the book's goal of accessibility.
The chapters in this section go beyond theoretical arguments to consider how various units and currencies transfer to conservation assessments. Avise's chapter notes the many factors that may be used to set conservation priorities for species. He argues that, given all these considerations, phylogenetic criteria probably will not have much impact on current practice. This discussion reflects one of the strengths of this book—a healthy skepticism, with discussion of strengths and weaknesses, rather than advocacy.
Several of the chapters in this first section touch on species concepts and conservation. However, much of the discussion is limited to the phylogenetic species concept (PSC), with consideration of other concepts mostly just in the context of putting forward the idea of a simplistic pluralism of approaches. A book on phylogeny, of course, does not mean that the PSC has to be the species concept of choice. An evolutionary lineage concept (e.g., Mayden, 1997) may better allow for integration of a range of different kinds of evidence about species hypotheses, without simplistic pluralism. Nevertheless, the discussion of species concepts in this section is useful background, particularly in highlighting the fact that the choice of concept can affect conservation decision outcomes.
There is also related discussion of the idea that, for many conservation planning assessments, one can side-step species concepts and species counting and use phylogenetic patterns directly through the PD (“phylogenetic diversity”; Faith, 1992) measure. This discussion remains timely, given emerging new contexts for such considerations. For example, large-scale DNA barcoding provides a wealth of new data for conservation planning applications, and PD-based assessments could allow us to side-step contentious species designations (e.g., Faith and Baker, 2006).
That said, there is more attention in this section, and indeed in the whole book, to the reverse question—whether conservation of phylogenetic diversity can be well covered by a conventional focus on the species level. Thus, phylogenetic information is seen here as the limiting, difficult-to-obtain information, and it is hypothesized that simply counting species tells us all we need to know about PD. As the editors' introductory chapter foreshadows, a key theme for the book (reflected in the early chapters by Avise et al., Purvis et al., Rodriguez et al., Mooers et al., Moritz et al., and others) is whether using PD actually makes a difference (changes the results) for conservation planning, relative to conventional use of species counts.
Mooers et al. review previous case studies, presenting the evidence supporting the idea that we are losing much more PD (or “evolutionary heritage”) than if extinction were random among species. It is satisfying that this early chapter highlights the idea that extinctions may be phylogenetically clumped, so de-coupling species loss and the loss of PD or evolutionary potential (see also Forest et al., 2007). No doubt future work will continue to extend these kinds of analyses, as in the recent work (Yesson and Culham, 2006) showing that phylogenetic dispersion of species that are resistant to climate change will imply that a large amount of PD will persist, in spite of species losses.
Many of the previously published papers investigating possible congruence of PD and species assessments have suffered the weakness of reliance on a single data set. In contrast, the chapter by Rodrigues et al. provides a fresh perspective through the use of computer simulations of phylogenies and species' geographic distributions to examine this issue. Their results are interpreted as implying that only in extreme cases do sets of localities maximizing species richness act as poor surrogates for capturing PD. However, this chapter may be more useful in inspiring further simulations than in delivering definitive answers. One limitation of their simulated phylogenetic trees is that they in effect ignore all possible cases where character divergence (longer branches leading to some species) would mean that PD gives a different answer. But a more subtle, and pervasive, bias may be found in their simulated geographic distributions of species. Typically, a given locality had a full third of all species, with about half of the species quite widespread. Further, the simulation design in effect guaranteed that every clade most likely had one or more very widespread species. Under these conditions, it is not surprising that selecting localities to maximize number of species ensured also the representation of PD. As soon as one or more widespread species of a given clade were sampled, all its deeper ancestral branches were captured, and PD was therefore well sampled. Pending more general simulations, it therefore seems wise to view their skepticism about the utility of PD skeptically.
“Units and currencies” is all about what we might like to count for conservation assessment. In the second section, chapters by Smith et al., Moritz et al., and others describe why evolutionary processes are important considerations in conservation, reinforcing the idea of moving beyond conservation of pattern (e.g., lists of species conserved). In this context, conservation of PD is seen as one way to retain evolutionary potential (see also Moritz et al., 2002).
In this section, Smith et al.'s detailed study takes a phylogenetic approach to understanding how African ecotones may be important regions for future divergence and speciation. They advocate strategies that identify surrogates for key processes, set conservation targets for these, and integrate these goals into regional conservation planning. However, processes in these chapters are not just alternative targets for conservation action. Also discussed is the need to understand evolutionary processes as responses to present challenges such as global climate and land-cover change. One intriguing suggestion is that we may be able to produce at some stage a null model that helps indicate which lineages represent unusual evolutionary processes.
In another chapter, Jones et al. show how tracing change in geographic range size over evolutionary time is relevant to present-day conservation planning. An interesting discussion concerns the possibility that there is heritability of geographic ranges. This will be an important property, for example, if it means that descendents of taxa with small range sizes will also have small range sizes.
The final two sections of the book follow logically from the previous discussion of pattern and evolutionary processes, to address “effects of human processes” and “prognosis.” This is perhaps the most diverse section of the book. The thought-provoking chapter by Brooks et al. returns to that core theme of species' range size, linking it to the principle of complementarity (the idea that the conservation value of a locality depends on what additional, not-already-represented, components of biodiversity it contributes). Brooks et al. argue that “the key to measuring biodiversity value of areas lies in considering the range sizes of the species they hold.” The chapter goes on to address the implications for setting global conservation priorities, including consideration of global hotspots and key biodiversity areas.
While that discussion highlights links from phylogeny to conservation decision making, other chapters explain how phylogeny helps us to understand patterns of human-induced threats to biodiversity. The chapters by Purvis et al. and Bennett et al. provide compelling arguments that phylogenetic and evolutionary considerations must play a role in any attempts to understand patterns of extinction risk among species.
Lockwood addresses the global dilemma of homogenization—where different places start to look taxonomically the same. The chapter puts forward the evidence for “a clear taxonomic footprint to the homogenization process” and argues that phylogeny, at least in combination with other factors, can explain variation among species in invasiveness. The chapter's conclusion—that phylogenetic clumping of extinction and phylogenetic clumping of invasiveness work together to produce homogenization—sets an agenda for future research.
An interesting question, considered only briefly in that chapter, is what effect taxonomic homogenization might have on the functioning of ecosystems. Given that some recent work suggests that high phylogenetic diversity within localities boosts functioning (Webb et al., 2006), it will be interesting to see how this intersects with homogenization. The background information in this section of the book will create more interest in the emerging work elsewhere on community-scale phylogenetic diversity studies (e.g., Webb et al., 2002; Proches et al., 2006).
Following the chapters in this section that describe how phylogeny might predict extinction risk and invasiveness, it is fitting that the final chapter, by Barroclough and Davies, addresses “phylogenetic forecasting” from a complementary point of view—the prediction of future speciation patterns.
So far, I have outlined the range of issues covered by the different chapters, but it seems useful, given the goals of the book, and the need for synthesis, to consider how well all of these bits fit together. This book is very useful in capturing in one place the current state-of-play. However, I think that this collection of papers only lays the foundation for, rather than really achieving, the needed synthesis. Perhaps this was a missed opportunity for a carefully edited volume.
The gaps in synthesis are well illustrated by the treatment in the book of the PD measure. The long history of such measures (going back to the “taxonomic distinctiveness” recommendations of IUCN, 1980) has been dominated more by debates about definitions than by applications. Although the widely accepted measure, PD (based on the length of the phylogenetic tree spanned by a set of taxa; Faith, 1992), was proposed some 15 years ago, the associated terminology and definitions surprisingly still have not gained much standardization (for discussion, see Faith 2006). For example, a recent Systematic Biology paper (Lewis and Lewis, 2005) introduces yet another new term, “exclusive molecular phylodiversity,” to describe what amounts to the PD contribution of a set of taxa defined by some shared characteristic—the rationale being that “PD” is only applicable to taxon sets defined by localities. Others correctly see PD as general, as in the large number of published studies examining the PD of those species that are extinction prone (see the recent example by Yesson and Culham, 2006).
This edited volume is a missed opportunity to reconcile such inconsistencies. Through the course of the book, the term “PD” is sometimes used to indicate “phylogenetic diversity” in a general sense, is sometimes used to abbreviate “phylogenetic distinctiveness,” and is sometimes used to indicate the term as defined by Faith (1992). Furthermore, in some chapters PD instead is called “IEH,” “EH,” or “evolutionary history.” To be fair, the overall inconsistency does not mean that individual chapters fail to clarify definitions. For example, Mooers et al. present a nice example illustrating how PD of a locality or region must be measured back to some common root if comparisons are to be made with other sets of taxa. Rodrigues et al. also highlight this proper interpretation.
The inconsistent terminology is not just a cosmetic problem: it weakens some of the links that need to be made to conservation planning and assessment. In such applications, PD calculations include PD-complementarity (e.g., how much of the phylogenetic tree is pruned away when we lose one or more species; for discussion see Faith et al., 2004), which forms the analogue to conventional species-based complementarity. PD-complementarity means that all of the usual biodiversity priority-setting and planning methods conventionally applied at the species level (e.g., Margules and Pressey, 2000) apply equally at the phylogenetic or “features” level (e.g., see the planning software of Faith and Walker, 1996).
The book addresses the principle of complementarity but does not provide any clear picture about how the calculations work. For example, the term “complementarity” is linked to “dissimilarity” (e.g., Moritz et al.), but dissimilarity values are not the same as complementarity values (see Faith et al., 2004). In another chapter, Sinclair et al. provide the reader with references to three published examples on the use of phylogeny and PD to set priorities on taxa/localities. However, all three of those published examples are problematic in ignoring PD-complementarity. For example, one of the studies obtains PD values for sets of taxa by adding individual taxa PD scores together, in effect multiple-counting shared branches.
In other chapters, it is not clear whether there is proper consideration of the role of PD-complementarity. Avise's chapter seems to reinforce long-standing misconceptions about how rankings of species might be used, in saying that “species receiving higher scores are deemed more worthy of immediate conservation attention.” Static phylogenetic diversity scores are added to other criteria scores, but all of this obscures the point that conservation of one species can change the phylogenetic score (the PD-complementarity value) of any other species.
A special case of PD-complementarity, PD-endemism, reflects the PD unique to regions or sets of taxa (see Faith, 1992; Faith et al., 2004). These values naturally will contrast with values for total PD of the same region or set of taxa. For example, Mooers et al. measure the total PD for a country, and refer to it as the country's “evolutionary heritage.” A nice complement to those comparisons would be based on the PD-endemism value for each country. PD-endemism scores are now well established for evaluations of global hotspots (e.g., Sechrest et al., 2002).
The chapter by Moritz et al. uses the term “PD-endemism” to refer to a region's PD value, defined by those taxa present in that region. This value may have been calculated only for those species unique to the region, thus creating a link to “endemism.” However, such a calculation still differs from the alternative calculation of the amount PD that is actually unique to the region (the total PD of the set of endemic species is not the same as the PD-endemism of that set of species). This is not to say that one of the two measures definitely is better, but the book does not make such alternatives apparent to the reader.
Overall, the failure to provide more synthesis regarding PD concepts and calculations may limit the book's utility as background for evaluating new developments. For example, without the appreciation of the role of PD-complementarity for species priority setting, there will continue to be little take-up of the more complex versions of PD-complementarity that incorporate probabilities (e.g., probabilities of extinction, as in the probabilistic version of PD introduced by Witting and Loeschke, 1995). These methods currently are relevant to the unresolved debates about the methods used in a new, high profile, global program for species priority setting. The EDGE program (“evolutionarily distinct, globally endangered”; Isaac et al., 2007) is soliciting support for conservation action for threatened and evolutionarily distinctive species. It uses a simple, static, scoring approach, and debates now focus on whether, by ignoring the existing framework integrating PD, complementarity, and extinction probabilities, the program may be misallocating scarce conservation funds (Faith, 2007a; Faith, 2007b). Perhaps it is the general shortfall in synthesis that accounts for why the old Witting and Loeschke approach has been considered only after EDGE was up and running, rather than in its development phase.
With these concerns in mind, I return to a core question posed by the book: whether PD makes a difference in conservation planning. One conclusion (as summarized in the editors' overview chapter) is that PD will make a difference only in certain restricted cases. But lessons from other, analogous, work on biodiversity measures, if taken on board, might have tempered this conclusion.
First, I note that much of this evaluation of PD focuses on whether total PD correlates with the number of species. It is surprising that there has been so much work on this topic over the years—after all, a plot of PD versus species number produces a fundamental curve much like that for species versus area (the well-known species-area relation, SAR). The lessons from SAR show that the existence of the curve does not help a lot—we generally cannot just count up the total area amount (e.g., in reserves) and expect that to be an adequate indicator of the number of species captured. Furthermore, the business of reserve selection is mostly concerned about marginal gains in the number of species (the complementarity values of potential new reserves), and looks for better estimates of such gains than just hectares. So it is also for PD—even if we have the curve showing general correspondence with species number, the decision-maker will find that potential PD gains from conservation actions may not be indicated well by gains in species numbers. Such a decoupling of PD-complementarity and species-complementarity recently was documented by Forest et al. (2007).
Second, I note that the book's perspective on this issue reflects only work examining whether the PD value correlates with the species richness value of a taxon subset, within that given taxonomic group. But the broader lessons from biodiversity conservation planning indicate that the real issue in using a measure for biodiversity conservation is surrogacy—the prediction of patterns for many other taxonomic groups. The most important difference that PD may provide in conservation planning may be through improvements in surrogacy, relative to just counting species. For example, the phylogeny for one taxonomic group may reflect historical processes/relationships among geographic areas that are shared by other taxonomic groups, so that observed differences among areas in PD-complementarity values also reflect the values that would have been calculated for the other groups. Such a gain in surrogacy was part of the original rationale for using PD (Faith, 1992; see also the chapter by Moritz et al.), but this surrogacy hypothesis remains largely untested. Surrogacy value will be an important part of any justification for conservation planning using PD and DNA-barcoding trees, or for estimating any general loss of PD (and thus evolutionary potential) from impacts such as climate change.
Putting aside the question of PD's role in conservation planning, these considerations highlight again the importance of phylogeny in providing information about both pattern and process for biodiversity conservation. This book does well in addressing its stated goal of exploring a wide range of synergies between the rapidly developing disciplines of conservation biology and phylogenetics.