In the popular mind, Darwin's great discovery was that apes were our ancestors. To Stephen Jay Gould, that's not the half of it. The Darwinian ideas that people really have a hard time embracing, he maintains, are the implications of natural selection -- a theory that explains the development of Homo sapiens without reference to any sort of divine plan or vision of progress.

Woven through much of Gould's writing, and at the heart of his new "Full House," is an insistent demand that we "cash out" the deepest implications of Darwin's insights -- and begin to comprehend that our species, far from being the pinnacle of some inevitable trend in nature toward greater complexity, is simply a tiny accident occurring on a minor side-branch of the evolutionary tree.

In his 1991 "Wonderful Life," which is a sort of companion to "Full House," Gould used the example of the Cambrian explosion of species found in the fossils of the Burgess Shale to demonstrate that "contingency" -- accident, happenstance, the particular way that events unfold -- plays a central role in determining the fate of species. Rewind the tape of events to play evolution out once more, Gould argues, and the odds are against anything like Homo sapiens developing. We're here because we're here -- not because we had to be here.

"Full House" turns from the fossil record to the nature of statistics. Gould takes up a series of apparently unrelated and seemingly abstruse questions -- from the disappearance of the .400 batting average in recent decades to the likelihood of his own surviving an episode of stomach cancer -- and weaves them into an impassioned critique of the progressive view of evolution.

Ever since Plato, Gould argues, we've tried to understand events by identifying linear trends based on shifting averages. But in evolutionary terms, there are no averages -- just individual variations. If we focus on the spectrum of variations rather than averages and trends, we are less likely to be led astray by minor events taking place at statistical extremes -- and less likely to conclude that we are the culmination of a trend toward complex life. On Gould's graph of the distribution of complexity among life forms (see illustration below), humanity exists at "the right tail," not the top of the heap.