| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Book Review |
Department of Biology and Burke Museum, University of Washington, Seattle, Washington 98195 USA
Any new species must begin its existence as a small isolated population competing either against an abundant established biota or an alien environment or both. What gives it the edge to survive is likely to be unique for each species that makes it through those early challenges, but what they all share in common is the capacity to reproduce in excess and to compete successfully against whatever biotic or abiotic competitors they encounter. Given this straightforward statement about speciation, Jonathan Silvertown, a prominent plant ecologist from the Open University, UK, asks the simple question, "Why hasn't the ultimate competitor come along that can dominate all others and reign supreme in plant communities everywhere, or at least broadly across common landscapes?" While I've never seen the question of what causes and maintains biodiversity asked in exactly that way before, it makes sense when considered in light of the tremendous success some species have had when introduced into new communities or landscapes, often a continent away from their homeland.
The author thus sets the stage for an exploration of the biological basis for the invasion of alien plants that is transforming landscapes everywhere. In the process of this exploration, Silvertown leads the reader on a fascinating tour of the evolutionary and ecological research of the recent past that has addressed the nature of plant diversity, its origins and its maintenance, from the chalklands of Britain to the tropical forests of Panama; from the mountains of Japan to the Everglades of Florida. He starts with the big picture of plant phylogeny that is rapidly emerging from the study of DNA sequences. He moves from there to the Canary Islands to look at examples of plant groups that have radiated in a time and space amenable to interpretation in ways that only oceanic island archipelagoes are. From there he moves conceptually to the ecological scale to explore the interactions within and between species and their environment to weave a story of how the complexity of life in plant communities is maintained. He returns at the end to show how one species, Homo sapiens, has altered ecosystems in ways that have resulted in other species becoming "demons" when no longer in the communities in which they evolved. In the end it is clear that WE are the real "Demons in Eden."
The power of phylogenetic inference, combined with the rapidity with which molecular systematics has enabled phylogenetic history to be reconstructed, has led to a tremendous spurt in new interpretations of many old evolutionary questions, as well as ones never before considered. Silvertown is fascinated by the patterns he sees in the tree of flowering plants. He highlights some: Amborella trichopoda is sister to all other living angiosperms (some 300 000 species), most families have long subtending branches (interpreted to mean a long isolated history and relatively sudden diversification), species-rich families often have sisters that are species-poor. Is this evidence of "demon-hood" writ large, leaving its footprint on the entire phylogeny of flowering plants?
Efforts to quantify rates of diversification show that it is often difficult to substantiate such claims. For example, the tremendous radiation of flowering plants has long been assumed to be due to some intrinsic quality of flowering plants and that may be true. However, the earliest branches of the flowering plant tree are all, for the most part, species-poor groups, and statistical efforts to show a change in diversification rates have not detected a significant change in rates associated with the base of the flowering plant tree (Sanderson and Donoghue, 1994
). Maybe that just means the cause of the change in rates was something else. Surely, two sister families in which one is species rich and the other species poor (Poaceae and Ecdeiocoleaceae, in his example) must indicate something of significance? However, this would be so only if one accepts that "families" indicate something of significance. But families, like all ranks in conventional classifications, are arbitrarily defined and have no equivalence, except in the isolated cases where two families also may be sister groups. In conventional plant classifications, family is an important rank for organizing everything from herbaria to floras to plant systematics classes, so it is not surprising that the first explicitly phylogenetic classification of angiosperms (APG II, 2003
) has adjusted various groups so that the well-supported (long-stemmed) branches are recognized as families (Chase et al., 2001
). It is appealing to argue that there is some demonic process at work that results in one family radiating more than its sister. However, given the full range of patterns among sister clades (not just families) that can be identified on a tree as complex as that of the angiosperms, from sister clades that are highly asymmetrical to ones with equal numbers of descendant species, cherry-picking ones that stand out seems to offer little more than speculation and certainly does not constitute evidence of an underlying process. While these are interesting patterns and lead to fascinating questions, they might be more a function of how human perception tends to simplify complex patterns than they are of any underlying biological significance (Scotland and Sanderson, 2004
). Silvertown does better when he reaches his own turf in later chapters.
From the big phylogenetic picture, Silvertown turns his attention to one of the hot spots of speciation, the Canary Islands. Oceanic islands have held the attention of evolutionists ever since Darwin made his observations on the diversity he observed on the Galápagos Islands. True oceanic islands are formed by emergence of land above sea level where none has been before. They often can be dated geologically and have a relatively recent "starting point" for occupation by resident flora and fauna. By comparison, continents are huge and old. The history of diversification of life on continents is complicated by both of those facts. Other islands are but fragments of continents and do not have the recency of history that oceanic islands may have, even if they are as isolated. They, too, have an important part of the biodiversity story to tell. The Canaries, for example, are only a couple hundred kilometers from the coast of Africa. New Caledonia is just as isolated, but by virtue of its history as a continental fragment, it contains old lineages of plants, including Amborella trichopoda, mentioned earlier as the sister to all other extant flowering plants.
Silvertown presents an appealing scenario of diversification by plant groups lucky enough to make it to these islands when they were young and not very diverse biologically. The nature of tropical volcanic islands, with altitudinal relief creating a mosaic of temperature and rainfall around the island, presents a diverse array of habitats available in close proximity. When there is an archipelago of such islands, each contains the same mosaic of conditions found on the others. If they are sufficiently distant from one another, successful migration between them is rare, thus evolution is presented with replicate experiments in adaptation into those habitats. Studies of numerous plant groups now have been done on the Canaries. If colonization among islands is rare, then successful colonization of the islands in the first place must be rarer (success in colonization is a product of rate of migration and likelihood of establishment). This is borne out by the many studies that show single ancestors for groups that have many species in the Canaries. However, Silvertown takes a leap of faith in suggesting that the rate of migration may not be all that rare (after all, there are a lot of plant groups represented and many of them have been there quite long) and that the single ancestry of so many endemic groups must, perforce, require an ancillary explanation. He postulates that the arrival of any subsequent migrants will be unsuccessful due to the presence of the now-abundant descendants of the first successful colonists. He cites a study of olives, where estimated age of arrival on each island coincides reasonably well to the age of the islands, suggesting that migrants must arrive relatively early and, thus, move among islands frequently. I am not sure I buy this as a general explanation, but it may well be the case in some groups. There are many reasons why any migrant may be unsuccessful; the likelihood for any seed to survive to maturity once dispersed is small. One could as readily postulate that having potential mates and an opportunity to avoid inbreeding might favor later migrants.
Silvertown makes the novel case that the demon-like reproductive ability of early colonists enables the early arrivals to spread and prevent subsequent arrivals from establishing. However, he then goes on to a more conventional explanation for the "adaptive radiation" of the successful colonist into a series of species each successfully occupying a different habitat in the mosaic described (p. 36): "natural selection would have built on this small beginning and honed the drought tolerance of the invader's descendants with each new generation." In so doing, he ignores an important element of oceanic island diversification that may be relevant to his later thesis on anthropogenic invasive plants. He misses an important implication of the relatively unrestricted reproductive potential of the successful island colonists relative to their continental ancestors. Rapid population growth means that a greater proportion of the offspring survive, permitting mutations to survive that would have been culled through stabilizing or modal selection in a more biotically crowded environment. Thus, a greater genetic variance may exist among descendants. This release from strong selective constraints may ultimately be the key to the pattern of rapid radiation in oceanic island endemic groups. Ultimately, competition will build among conspecifics, and selection will hone new populations to their preferred environment. Character displacement may play a role in maintaining reproductive isolation, even though close relatives in oceanic island clades often have no intrinsic barriers to mating (Carr and Kyhos, 1986
; Motley and Carr, 1998
).
Silvertown concludes his chapter on Canary Island radiations by trying to draw a line from the rapid radiations on oceanic islands to the putative pattern of sudden radiation of families in angiosperm phylogeny. While it is true that many groups that have radiated on oceanic islands exhibit what might be called starburst phylogenies and are quite distinct from their continental relatives, it is simply an incorrect reading of the phylogenetic trees of angiosperms to draw a line from this pattern in endemic island groups to the pattern he described for flowering plant families, most of which exhibit a readily resolved branching order at their bases indicating their diversification over an extended period of time.
In Chapters 4 through 7, Silvertown moves into the ecological basis for the maintenance of biodiversity in plant communities. Here he is on firmer footing, and he is able to cite scientific studies (some of this his own work) to support his line of reasoning. The book takes the form of a travelogue, with Silvertown taking the readers to plant communities around the world to see how various mechanisms help maintain diversity in natural settings and hold potential demons in check.
In the boreal forests of northern Japan, the interplay of environment and intrinsic reproductive capacity keep a few species from dominating. High latitude forests often have relatively few species, but what's to prevent one from dominating? At lower elevations, the understory bamboo Sasa nipponica spreads by rhizomes and forms a waist-high thicket that completely dominates the forest floor. Nothing can grow in its shade, not even seedlings of overstory trees. Higher on the slopes, firs dominate and form a uniform and dense cover. Why don't these forests become monocultures of bamboo and fir? Each pays for its competitive edge early in life by foregoing seed reproduction in favor of vegetative growth, but when reproductive maturity sets in, the cost is significant and ultimately leads to the demise of local dominance. Bamboo is perhaps unique in being monocarpic with the capacity to dominate plant communities, yet all individuals reproduce and die simultaneously. Thus, tree seedlings can establish and other species thrive for a while, until the bamboo returns. On the upper slopes, harsh winters mean that the firs are near their limit of survival and when they reach reproductive maturity, the cost of producing cones and seed, along with the increased exposure that size brings, results in die-off and the appearance of "fir waves" that spread across the mountain slopes. Surely it is true that all seed plants must bear this cost. But is it possible that this is a general phenomenon, or is it idiosyncratic for a few species in a few places?
From the boreal forest we travel to the tropical forests of Panama. Tropical forests are renown for their incredible species diversity. Environmental constraints are few, and there are no apparent demons that dominate until reproductive maturity extracts its toll. So, what then, holds any species in check, or conversely, how do so many species coexist. Silvertown presents four hypotheses that have been considered in this case: (1) the environment that seems so homogeneous is really subdivided by different species that have evolved to take advantage of heterogeneity that ecologists cannot discern, (2) all the species are really so alike that coexistence is random and a more or less equal chance of reproductive success maintains a diverse flora, (3) frequency-dependent selection in the form of "enemies" (pathogens and predators) prevents any species from dominating and permits many species to survive, so long as they stay scattered and infrequent, and (4) a trade-off between competitive ability and dispersal ability permits such a diverse assemblage to coexist, so long as there is sufficient small disturbance (e.g., wind-thrown trees providing openings in the canopy) to permit weak competitors to continually find a place to become established. It is too bad that Silvertown does not also present the "centrifugal" theory of species diversity (Heard, 1991
), but he sticks pretty closely to theories for which experimental evidence is available, so ignores this and other minority views. The initial studies of fine-scale patterning and diversity on Barro Colorado Island failed to show evidence of either niche partitioning or the "enemies" hypothesis and seemed to favor the "dispersal limitation hypothesis." However, subsequent studies have started to lend greater support to the "enemies" hypothesis, and the latter conclusion provides an eventual link to species invasions that comes later in the book.
Anyone who has traveled much and looked at patterns of vegetation on the landscape knows that the tropical forests of Panama are not going to tell the whole story of plant diversity. In the Pacific Northwest, where I live, the Pacific slopes of the coastal mountains are uniformly clothed in a forest that varies little with slope or aspect; only elevation brings significant changes in species composition. However, on the interior slopes, difference of a few tens of meters, when accompanied by changes in aspect from north exposure to south, can result in dramatically different community composition. This is true in much more subtle communities, too, as Silvertown shows with his own work on niche differentiation in meadow communities in England, where species sort out on moisture gradients more subtle than we see in the Cascades. One last experimental study (The Park Grass Experiment at Rothamstead in the UK), the longest running study of its kind, shows that nutrient limitations permit a greater diversity of species to coexist in some plots and that regular addition of nutrients, especially nitrogen, can result in a few species dominating and eliminating other, more stress-tolerant species. Nitrogen limitation is a common characteristic of many plant communities around the world, thus nutrient stress also contributes to maintenance of diversity in plant communities. Thus Silvertown leads us through a series of examples from rather different ecosystems and communities around the world to show that biodiversity is maintained by several factors that intersect in ways that prevent any species from dominating to the exclusion of all others.
After traveling from Japan to Panama to England, Silvertown brings us to Florida, a place run amok by many invasive plant species and also brings us to the ultimate point of this book—how does what we know about the maintenance of biodiversity in natural communities help us to understand plant invasiveness? It may be that continental plant communities are not as niche saturated as generally thought and recent immigrants from other continents find opportunities akin to what a chance immigrant to an oceanic island might find. It appears that many successful invasive species are so successful because they have escaped predators or pathogens found in their native ranges, which were left behind when introduced to their new habitat. The occasional success in controlling invasive plants (e.g., St.-Johns-wort, Hypericum perforatum, in western North America) by introducing predators from their homeland lends support to this as a contributing factor in many instances. Also, importantly, increasing levels of nitrogen-containing air pollution has the potential to reduce diversity in many communities, as better competitors displace stress-tolerant species. Sometimes combinations of these work in concert, as in the Pacific Northwest where the invasive leguminous shrub, Scot's broom (Cytisus scoparius) increases soil nitrogen levels through its nitrogen-fixing capacity, thus altering soil chemistry to the detriment of native species.
While reading Demons, I couldn't help comparing it to two recent books by John Terborgh (Requiem for Nature) and E. O. Wilson (The Future of Life). Both explore man's role in decimating biodiversity. Wilson sugar-coats the future with platitudes that no one who observes the situation closely could easily swallow, whereas Terborgh presents a much more pragmatic and, hence, cynical perspective, in which neither a top-down, nor bottom-up approach to preserving biodiversity is likely to succeed. Both of those books focus on the number one source of the problem of biodiversity loss—habitat destruction. However, neither of these books address the problem recognized as second only to habitat destruction as a threat to biodiversity—alien invasive species—and the biological basis of what may become a more critical problem in the future. Demons in Eden provides a needed accounting of the biological basis of invasiveness and how it is likely to affect natural communities in the future.
Silvertown attempts to explain plant diversity, from a centimeter scale in the chalk downs of England to the flowering plant "tree of life." Along the way, he explains hypotheses, introduces the scientists involved, and describes selected studies and data that illuminate the synthesis as he hopes others will see it. His bias enters occasionally, but largely it is kept in check or hidden carefully enough that I didn't often see it. I must admit that when I started reading this slim volume, I was prepared to be skeptical, even cynical, about its premise and whether the author could weave together concepts from so many diverse fields into a compelling story that would be palatable to an educated lay audience and not so superficial as to be meaningless to a professional biologist. If there is anyone likely to be cynical, it is a phylogeneticist like myself reading an ecologist's explanation of biodiversity. This may have been exacerbated by the book starting with the phylogenetic perspective before moving to the more ecological, where I might be more easily persuaded by arguments with which I am less familiar. However, by its end, I was taken in and my marginal notes became enthusiastic, instead of cynical.
This book is a quick read and Silvertown's style is engaging. I highly recommend it for biologists looking for a way to better comprehend the alien invasion happening all around us. As I write this, I am already laying out in my mind a capstone course planned around the layout of Silverstone's book for our seniors. Its 10 chapters, along with readings from the primary literature, which is well referenced in the footnotes with suggestions for further reading, will fit perfectly in our 10-week terms.
FOOTNOTES
101 1 Demons in Eden: The paradox of plant diversity. J. Silvertown. University of Chicago Press, Chicago, Illinois, USA, 2005; 192 pp., 8 color plates, $25.00 cloth, ISBN 0-226-75771-4. 
2 olmstead{at}u.washington.edu
LITERATURE CITED
Angiosperm Phylogeny Group (APG II).. 2003. An update of APG classification for the orders and families of flowering plants. Botanical Journal of the Linnean Society 141: 399-436.[CrossRef]
Carr G. D. Kyhos D. W.. 1986. Adaptive radiation in the Hawaiian silversword alliance (Compositae–Madiinae). II. Cytogenetics of artificial and natural hybrids. Evolution 40: 959-976.[CrossRef][ISI]
Chase M. W. Fay M. F. Savolainen V.. 2001. Higher-level classification in the angiosperms: new insights from the perspective of DNA sequence data. In T. Stuessy, E. Horandl and V. Mayer [eds.] Plant systematics: a half-century of progress (1950–2000) and future challenges 157-176 International Association for Plant Taxonomy, Vienna, Austria.
Heard S. B.. 1991. The centrifugal theory of species diversity. Bulletin of the Ecological Society of America 72: 13.
Motley T. J. Carr G. D.. 1998. Artificial hybridization in the Hawaiian endemic genus Labordia (Loganiaceae). American Journal of Botany 85: 654-660.[Abstract]
Sanderson M. J. Donoghue M. J.. 1994. Shifts in diversification rate with the origin of angiosperms. Science 264: 1590-1593.
Scotland R. W. Sanderson M. J.. 2004. The significance of few versus many in the tree of life. Science 5658: 643.
Terborgh J.. 1999. Requiem for nature Island Press, Washington, D.C., USA.
Wilson E. O.. 2002. The future of life Alfred A. Knopf, New York, New York, USA.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
