As the Earth undergoes the largest mass extinction since the elimination of the dinosaurs 65 million years ago, one of the biggest unsettled questions is the phenomenon's effect on the intricate biological network upon which all life depends.

Some have argued, for instance, that the disappearance of even a few species starts a reaction down the chain of life, with animals types toppling like dominoes in a line until the last one -- that is to say, humans -- is felled. Naturalist Aldo Leopold expressed this point of view when he said the first rule of intelligent tinkering was "saving all the parts."

Yet recent research has indicated in order to have a functioning network, it might not be necessary to have all the parts. Curiously, some tantalizing evidence emerges from a decidedly inorganic network -- the Internet.

Alun Lloyd of the Institute for Advanced Study in Princeton, N.J. and Robert May of Oxford University in England, writing in the May 18, 2002, issue of the journal Science, said, "The study of communications networks ... has interesting parallels both with conventional epidemiology and with the ability of ecosystems to handle disturbances."

The two scientists have looked at how computer viruses propagate through a network and related it to the transmission of contagious diseases such as AIDS in humans and in ecological networks.

The landscape of the Internet is not flat and uniform. Some of its nodes, such as university servers, large corporations or America Online, have a lot of connections to other points. On the other hand, personal web pages will have relatively few connections.

In this landscape, AOL, Microsoft and other highly connected nodes are the mountains and the rest of us are out there in the desert. This has important implications for the spread of computer viruses, real viruses and the dominoes of extinction.

The first is if a virus attacks a relatively loosely connected node -- one of us out in the Internet desert -- the virus will spread very slowly. A few dominoes might topple, but the overall network will remain intact.

If, however, a virus climbs one of those mountains of connectivity, it spreads much more rapidly. So the spread of viruses is related to the landscape -- also called the topology in mathematical terms -- in which it finds itself. In a randomly connected network, each node will have a very few connections. In constructed networks such as the Internet, or ones that evolved without randomness, such as human society or ecological networks -- called, counter-intuitively, "scale-free networks" -- some nodes will be central to proper functioning while others will be less important.

You can see this process at work in the spread of AIDS and other sexually transmitted diseases. As epidemiologists have noted, "a few individuals (such as prostitutes) have very high numbers of partners."

In an interview, Lloyd told United Press International, "In terms of computer networks, one of the clear points that comes out of the analysis is that if you have this scale-free network, the most transmission can be traced to the most highly connected nodes. So this is a clear implication for how to prevent transfer of viruses --- you concentrate on the most highly connected nodes."

So if you are interested in stopping the AIDS epidemic, this evidence clearly argues for steps like providing clean needles to drug users and condoms to sexually active people.

On the ecological front, however, so many processes are involved that the conclusions are not clear-cut. Lloyd said, "It's not so easy to figure out in an ecosystem case. A food web may be one of those networks, so there are interactions in species where the nodes are the species and the links are that one species eats this one and competes with another one. The stability of the ecosystem might depend on these links."

This kind of model could have implications for establishing priorities in the protection of endangered species. "With some species," Lloyd said, "you could remove it quite easily and it might not have much of an effect. But a species with a lot of links might have a very large effect."

In a paper written about computer viruses a couple of years ago, IBM scientist Jeffrey Kephart anticipated interconnectedness might unearth important results.

"Lack of attention to this simple, fundamental question has seriously hampered our theoretical understanding of population biology and epidemiology," he said. "For example, in this heyday of HIV, we are admonished daily by educators about the dangers of promiscuous activity, yet until recently there were no quantitative theoretical studies of how the spread of disease depends upon the detailed network of contacts between individuals."

Ecologically, rather than a "saving all the parts" approach, research would tend to argue for protecting the "most highly connected nodes," or what are called in biology "keystone species."

Our knowledge about what constitutes a keystone species still is in its infancy. It has been argued that some top predators such as wolves fit this definition. It seems certain some less glamorous animals -- the recently controversial prairie dog comes to mind -- almost certainly should be considered critical. They feed so many other species, from snakes to hawks and many in between, that their importance is, if anything, understated.

But what about the California condor? Or Preble's meadow jumping mouse? Or the narrow-foot hygrotus diving beetle? It is pretty hard to argue they are highly connected nodes. The planet has gotten along without noticeable damage over the past 30 years with almost zero California condors. Do we simply consign them to extinction?

The answer to this question is usually, "No." But it gets us into the arena of values -- morality, if you prefer. The only reason free needles are not distributed to AIDS-susceptible drug users in the United State -- though most everyone knows this is an effective way to fight the epidemic -- is the misguided application of moral precepts.

The environmental movement generally has ignored morality, arguing that good science will replace value judgments to determine the best outcome of ecological controversies.

But science is a two-edged sword. Sometimes the scientific result and the morally correct result are not the same.