CHARLES ROBERT DARWIN changed the world. In a single, superbly documented work, he completed humankind's Copernican drift away from the center of the universe, simultaneously offering the infant life sciences a binding naturalistic schema that has vigorously withstood the century-and-a-quarter of scientific change since its first publication. Darwin's magnum opus was, of course, the oddly-titled On the Origin of Species by Means of Natural Selection, or the Preservation of Favoured Races in the Struggle for Life -- oddly titled because the sage accomplished all this without even discussing the origin of species!

Which is hardly surprising, in retrospect. Darwin, like many of his contemporaries, labored under the impression that all life was constant change, and most change was progressive. History, that of life as well as that of humanity, proceeded through "insensibly graded series" of minute changes. And the fossil record as Darwin knew it seemed in part to bear him out. Certainly there was change through time as successive forms appeared in successive layers of rock. There was indubitably progress, too, if we regard Englishmen as higher than trilobites. Yet the fossils did not -- and still do not -- progress in those "insensibly graded series."

Since the publication of the Origin, most evolutionists have accepted without overmuch inspection the idea that evolution proceeds slowly and at a constant pace, each lineage of organisms being constantly modified by changing forces of selection to produce progressive changes in sosedly optimal directions. Broadly, this Darwinian notion may be called "phyletic gradualism," a term coined by Niles Eldredge and Stephen Jay Gould in 1972. They defined the process with a series of four tenets:

(1) New species arise by the transformation of an ancestral population into its modified descendants;

(2) This transformation is even and slow;

(3) The transformation involves large numbers, usually the entire ancestral population;

(4) The transformation occurs over all or a large part of the ancestral species' geographic range.

Hence, as Darwin suggested, the fossil record for the origin of a new species should consist of a long sequence of continuous intermediate forms linking ancestor and descendant. Any breaks in the record are therefore due to that record's imperfections rather than to the actual nonexistence of such contiguous forms.

Reflecting on these implications, we could say that there may be no such thing as a fixed species -- that "species" is simply our name for a group isolated within the timespan in which we happen to find it. As an example, there might be no real Homo sapiens; there may instead be only modern humanity as we know it today, this being but one of Darwin's "infinitely numerous transitional links" -- which links, unfortunately for the gradualist model, are usually missing. Even among organisms of which we have far more complete fossil records than we do for hominids, as in the case of many hard- shelled marine invertebrates, we find stepwise alterations rather than gradual transitions. This nonuniformity of change, this missing of links, appears virtually to be universal in the fossil record -- a flaw in the Darwinian prediction that such charlatans as creationists delight in waving about.

In response to these apparent anomalies in the Darwinian synthesis, Eldredge and Gould have proposed an alternative model for evolutionary change, one that affirms the existence of the species as a true evolutionary unit with a birth, life and death rather than as a transitional form among "infinitely numerous transitional links." The new model, for which Gould coined the name "punctuated equilibria," is defined with another four tenets:

(1) New species arise by the splitting of ancestral lineages;

(2) New species develop rapidly;

(3) A small sub-population of the ancestral form gives rise to the new species;

(4) The new species originates in a very small and remote part of the ancestral species' geographic range.

Hence we might guess that in any local fossil record containing both ancestral and descendant species, there ought to be a sharp break between an ancestral species and its descendant; each species itself will remain essentially constant in form throughout its existence. This fossil break marks the migration of the descendant from its peripheral turf into the midst of the ancestral range, and, since such speciation events occur rapidly in small, isolated populations, we will rarelcover the actual transitional event in the fossil record.

Viewed in this light, the frequent breaks in the fossil record where Darwin expected to find all those infinitely graduated links are actually real markers of species change. Here Darwin's dilemma is neatly eliminated. By abandoning the requirement that evolution proceed gradually under the influence of perpetually changing forces of natural selection, we bring our model of evolutionary mechanics more into accord with evidence presented both in fossils and on a variety of fronts in the living world.

Contrary to many misinformed accounts, the "punctuated equilibria" model does not eliminate evolution by natural selection. The selective process, indeed, becomes the primary stabilizer of the very real "boundaries" between species that seem so persistent among fossils and in living populations. Once a new form is established in an ecological niche, the exigencies of that niche -- in oter words, natural selection -- refine and define the form and behavior of the species until it achieves a more or less unchanging character through perhaps millions of years. "If you really have built a better mousetrap," as Eldredge puts it, "for heaven's sake don't fool around with it."

Rather than responding to constant slight environmental changes with constant slight genetic changes as in the old Darwinian model, species more often appear to respond to environmental change by actually following their ecological niches. Many complex Pleistocene ecological communities, for example, moved back and forth virtually in toto during the advances and recessions of the glaciers, but they did not change their component species, plant or animal, in so doing. This appears to be the pattern for almost all surviving species. Any species that cannot follow its proper niche via feet, wings, seeds or whatever tends to die out when times get tough rather than to "adapt." (Confusion can arise when fortuitous, isolated events of speciation permit descendant species to re-invade the ranges of recently extinct ancestors, often creating in retrospect a semblance of progress that is actually far from gradual.)

While still far from universally accepted as an evolutionary mechanism, the theory of punctuated equilibria seems to be one of those rare insights -- like Darwin's own -- that workably link a wide range of hitherto intractable phenomena. Time Frames is Eldredge's long-awaited recounting of the theory's development and implications, pros and cons. Lucid and entertaining, the book combines tales of field and laboratory research, history and politics, in one of the most intriguing scientific detective stories of recent years.