In spite of our medical advances, one-third of all human deaths are caused by infectious disease. That could be part of the reason John Thompson is so upset over his newly logged research site.

Granted, Thompson's research directly concerns neither human disease nor logging. He is more immediately drawn to the interaction between Lithophragma parviflorum, a flower commonly known as prairie star, and Greya politella, a little gray moth, which, though seemingly abundant, few people have ever noticed. In fact, before Thompson and his colleagues could study it, they had to taxonomically describe and name it. Now the shifting menage of Lithophragma, Greya, and another flower, Heuchera grossulariifolia, lends itself to many intricate stories of evolution, speciation, and biodiversity--and epidemiology.

As a logging helicopter thud-thud-thuds over the rise, Thompson stands staring across a Ponderosa glade where in another week or so an abundant population of Lithophragma will bloom. But now, what is most apparent is the skid trail through the middle of the glade and the slash piles scattered about it.

The desecration is not disastrous. Compared to logging effects on the denser fir forests to the north, the result looks more like that of a skirmish than a war. The Litho will still bloom over much of the site. It's just that the logging here seems such a mockery.

Another of his twelve study sites, on the border of the Gospel Hump wilderness, recently had an ATV trail scoured through the middle of it. "When will it end?" he asks. * * *

"It is interesting to contemplate an entangled bank, clothed with many plants of many kinds, with birds singing on the bushes, with various insects flitting about, and with worms crawling through the damp earth, and to reflect that these elaborately constructed forms, so different from each other, and dependent upon each other in so complex a manner, have all been produced by laws acting around us." Profound as his conclusion to The Origin of Species is, Darwin had glimpsed only the surface of nature's genetic complexity currently revealing itself through phylogenetic studies and molecular techniques.

Among the evolutionary revelations of the past few decades is that species are about landscape. Using new molecular techniques, Thompson and his compatriots are tracing not only the reciprocal adaptation of genomes to other genomes--which defines "coevolution"--but intricately tuned genomic harmonies composed by their dispersal over the landscape--a structural counterpoint that he calls the geographic mosaic of coevolution. Thompson and others have seen that evolution and logging, evolution and sprawl development, evolution and industrial agriculture, are intimately related. However, commonsense though it seems after a certain exposure, demonstrating the effect of fragmentation and disturbance on evolution is a laborious process.

The study of coevolution is particularly difficult because species tend to coevolve not with just one other species, but with many. Thus, of the millions of species, only a few lend themselves readily to the sort of manipulation that will reveal the secrets of their most intimate associations. But a few discrete relationships have been teased out.

The beauty of the Greya-Heuchera-Lithophragma model is how distinct it is against the entangled bank of central Idaho, one of the very few large areas left in the continental United States that allows a study as broad and coherent as Thompson's.

Greya spends most of its life sitting on its host plant. The moths mate on the flowers, and the inseminated female, carrying pollen, ranges out a ways, from a few inches to ten meters, to deposit her eggs in new flowers. But first she sticks her proboscis into the flower's corolla, presumably to get nectar.

Then she turns around, sticks her abdomen deep into the corolla and cuts through the nectary into the flower's ovary with her ovipositor, the long organ through which she lays her eggs. Having laid them and, in the process, pollinated the flower, she rests, maybe a few seconds, maybe an hour, before flying on to the next plant.

Once hatched, the moth larvae feed on the flower's seeds, generally consuming about a quarter of them. As the seeds mature, the larvae chew through the ovary wall and fall to the ground, spending the rest of the summer buried in the ground beneath their dying hosts.

Late the next winter, the larva emerges from the soil and wraps itself in a new basal leaf from its host, in which it pupates. In late March or early April, the males start emerging from their leaf cocoons, followed several days later by the females.

Greya politella ranges from Montana to southwestern Colorado and from southern British Columbia to southern California. It feeds exclusively on flowers of the Lithophragma genus--except for spots in Idaho, where it feeds on Heuchera.

Greya's shift to Heuchera is immediately interesting because Greya is otherwise so predictable. What caused it to break its middle-class mold and try a new species? Opportunity and timing probably lead the list of reasons. Heuchera blooms just a little later than Lithophragma. In spite of their different appearance, they are closely related.

So perhaps one year, due to a random mutation, a group of moths emerges a little later than normal. Lithophragma has already bloomed, and the moths are hungry. They try the blooming Heuchera, and what do you know? It's pretty good--and the ovipositor fits fine! So why not?

Maybe the same thing happens next year. It wouldn't take much, says Thompson--just a few generations with natural selection favoring the mutant, and suddenly there's a shift in preference. Suddenly, also, this population is genotypically distinct from their forebears. And from the population just upriver, or just over the ridge.

Same species, distinct population. Or is it?

We rendezvous near the end of the road heading up the Main Salmon with graduate students Scott Nuismer and Kurt Merg, who have been working out here for several days. The back of our stationwagon is packed with greenhouse-reared Lithophragma. They unload these and replace them with other potted Lithos that have been here in the field with them.

The exchange is part of their investigation into polyploidy among the Lithophragma populations. Polyploidy, the development of multiple sets of chromosomes, has recently drawn interest as a major factor in plant evolution--and in the maintenance of biodiversity.

Nuismer explains that they place pairs of diploid and tetraploid plants out and then watch to determine which the local Greya prefer. Diploid Lithophragma, which have the normal two sets of parental chromosomes, and tetraploids, which have twice the normal sets, are physically indistinguishable, at least to the casual human observer.

As similar as they are morphologically, however, they are reproductively distinct. Although some genes might "squeak through," as Thompson puts it, generally if diploid and tetraploid populations do cross-pollinate, the result is a triploid, which is sterile.

"I keep getting dragged farther and farther into polyploidy," says Thompson. "We're not trying to move in that direction. It's really the data--the results keep pulling us to ask these questions about shaping plant-animal interactions."

Those questions are not just about speciation, but about how the interactions shift and diverge over both time and space, drawing together two themes, evolution and ecology, that simply cannot be considered separately, at least in Thompson's mind.

"The history of evolution and biodiversity is fundamentally a history of the evolution of species interactions," he argues in an essay in a recent Science. Not only do those interactions form the web of communities, but also of "landscapes." The shifting mosaics of interactions on that grand scale are essential to maintain diversity and ecological health.

Lots of otherwise self-contained populations are connected by movements of genes among them across the landscape. This, together with a process called genetic drift, by which small populations may lose some genes by chance, results in species interactions shifting in one direction or another, depending on how much movement of genes actually occurs and how differently natural selection acts in different populations.

If such indeed is the way interactions between species have been maintained over evolutionary time, one realizes how the wild geography of nature has been changed by humans, landscapes fragmented into isolated patches, resulting in the separation of populations and their genetic interplay.

* * *

As we head back downriver, we leave Nuismer and Merg planted patiently on the slope above the road, waiting for Greya.

Merg is building on work done here along the river by Kari Segraves, a former graduate student of Thompson's, now at Vanderbilt, who traced the phylogeny of regional pollinators feeding on diploid and tetraploid populations. Unlike its more famous relative the yucca moth, Greya is not the sole pollinator of its host plants. Merg is interested in how the evolution of polyploidy diversifies the interactions between species and how new species are generated because of decreased gene flow between populations by the pollinators.

Interesting as it seems, Merg's strategy, which involves tracking pollinators with a fluorescent dye, would mean little without the context of Segraves' work and the overall context of Thompson's endeavor.

As nature has yielded answers to surface questions, the deeper ones show themselves to be largely unanswerable without a great amount of scientific conversation and collaboration. Thompson's current laboratory ensemble and its alumni have developed an intriguing voice, intricately layered and harmonious, mixing with a growing chorus worldwide exploring the complex new score presented by co-evolutionary composition.

Thompson refers to the laboratories of Doug and Pam Soltis, just down the hall from his, without whose expertise Saxifragaceae phylogeny much of his work would not have been possible.

More specific to the geographic mosaic, Thompson edited a special supplement this year to the American Naturalist on the geographic mosaic that brings together much of the thinking worldwide.

Among other explorations of the geographic mosaic, some of the most dramatic work is by Craig Benkman at New Mexico State. Benkman works on the evolutionary interactions between conifers and crossbills. Long thought to be a single species, the crossbills are now believed to be a complex of species, each specialized to a particular conifer species--and each with a different bill morphology adapted to the cone of that conifer

May Berenbaum and Art Zangerl have shown that the parsnip and parsnip webworms differ among populations, with the parsnips defending themselves with varied mixtures of chemicals and the corresponding populations of moths matching counter-defenses.

Curt Lively of Indiana University has shown that snails and their trematode parasites differ from lake to lake in New Zealand and that higher rates of infection favor sexually reproducing snails.

Jeremy Burdon in Australia has traced the gene-for-gene coevolution between wild flax and flax rust.

What these scientists and others are demonstrating is how varied and vital populations are in the web of life. Thompson pushes this argument further. The continued evolution of interactions and the health of the ecosystems on which they depend require the complexity of large areas.

This notion is hard to swallow not only for real estate developers and politicians, but also for some ecologists. Thompson found himself in an argument recently with one of the world's leading community ecologists, who refused to accept the idea that community dynamics could be affected by rapid and ongoing evolution. The ecologist readily accepted that evolution was important over the long term, but thought of it mostly as a slow process.

He accused Thompson of using a molecular biology argument (the worst kind of reductionism!), of arguing that one should always go down to the hierarchy as far as possible in order to find an answer.

That's not fair at all, says Thompson. Rather, you take the argument to the appropriate level--which means that you sometimes move up the hierarchy, from the community level to the geographic level, for example, and incorporate the known important processes, such as ongoing evolution.

In spite of such resistance, Thompson finds himself lately being invited to address groups increasingly applied in nature. As the collective grasp of biological complexity becomes more sophisticated, his and others' arguments about the evolution of specialization and organisms' capacity for rapid adaptation are reaching more accepting ears. For example, though he had to turn it down, he was recently invited to the International Congress on Biological Control. The invitation was particularly significant in light of the recent observation that, in spite of careful screening, an insect introduced to prey on an introduced plant had moved to a native host, in effect saving a system in order to kill it.

Likewise, the failure of various pest control strategies in agriculture can be blamed on the ignorance of evolution. Much of the current enthusiasm for genetically engineered resistance fails to take into accound rapid evolution of counter-defenses.

Coevolutionary relationships and their capacity for rapid evolution are of intense interest also to epidemiologists. The story of infectious disease is one of evolution. HIV and the development of resistance by infectious organisms to antibiotics illustrate that evolution is not simply an epochal event, but an immediate and ongoing one.

On the way downriver toward the main highway, little red flags marking the boundary for a wider road are reminders that Thompson and his allies have not yet been invited to conferences of regional planners or road-builders. The new road will cover many Heucheras --and their interplay with Greya--which Thompson refers to sadly as that "really beautiful evolutionary event."

He recounts a recent discussion with students in his evolutionary ecology class and how dismayed he was by their resignation and cynicism. There's no stopping multinational mergers and population growth, they observed. Everything's going to end up looking like Manhattan. These are folks in their early twenties, says Thompson, where all the idealists should be.

But Thompson is old enough and scientist enough to persist. We just have to keep working, he says, so "we have the answers in the event the politicians ask for them."