On a grassy, eastward-facing hillside in Riverside County, between a citrus grove and a dusty basin, fenced in by a sagging string barrier and a couple of crooked stakes, grow half a dozen wild radish plants. Standing as they do among foxtails and wild sunflowers, the radishes, three feet high with pink or white blossoms and green seedpods, look more like vigorous flowering weeds than garden vegetables. But, says Norman Ellstrand, a UC Riverside evolutionary biologist who has studied them for the last three years, the seeds from these undistinguished pods may turn out to have major consequences in ecology and sociobiology.
Until now, some scientists considered evolution to be almost solely the work of natural selection--the "survival of the fittest" theory that organisms best adapted to their environment survive more readily than ill-adapted ones, thereby naturally weeding out undesirable traits from a particular species. Ellstrand, however, has conducted a relatively simple experiment with his radish plants that suggests that evolution may in fact be more random than previously thought. The implication is that human behavior may not be, as some ecologists and sociobiologists believe, inexorably fixed in our genes and unchangeable.
It is quite understandable, Ellstrand says, why in a chaotic world many people are drawn to the concept of natural selection. For one thing, it is virtually the only aspect of evolution most people learn in school. But most important, he says, may be that natural selection is comforting to many people in that it offers an "organized, deterministic, predictable way of looking at the world. We are better than dogs and dogs are better than goldfish and goldfish are better than protozoa. Natural selection rules like God." And its "highest manifestation" is man.
This is not to say that Ellstrand doesn't believe in natural selection. "Obviously it works or we wouldn't be here," he says. It's that contrary to the semi-mystical powers attributed to it by some people, natural selection doesn't so much bring out the very best in us as it allows us to just get by. "This is why," Ellstrand says, "we have bag ladies and diseases and things like that."
Ellstrand, 34, has a high-pitched voice, an informal manner and a droopy-nosed smile reminiscent of comedian Robin Williams. He seems almost too mild-mannered to go around smashing evolutionary icons. Although trained as an evolutionary geneticist, he teaches population ecology at UC Riverside, where his weed-covered experimental plots look so little like research sites that a developer once inadvertently razed one to build a condominium.
Ellstrand didn't begin his radish experiments with the intention of challenging sociobiological theory. In 1980, he says, one of his graduate students was "thrashing around for a Ph.D. project. She was interested in flower-color variation and what the consequences were with regard to pollination biology." So Ellstrand sent her to look at wild radishes, which have flowers that routinely come in many different hues. The student brought back research that showed that, at the biochemical level, wild radishes made good experimental plants because they had highly variable genes.
Ellstrand found that interesting but thought nothing more of it until a year later, when he was lecturing on the importance of gene flow (the movement of genetic information from one population to another) as an evolutionary force. "And we were getting frustrated because there was no direct data on what gene flow was in plant pollinations," Ellstrand says. "I said there must be some way of measuring this directly." What was needed was a plant with highly variable genes. Then it struck him--the wild radish. As the purpose of the experiment was to measure gene flow, which in this case meant nothing more than the average distance pollen travels to fertilize each plant, Ellstrand sent his graduate students out into the field to locate and tag every radish plant in the study area out to a radius of 1,000 meters. When the plants matured, the seeds were collected, and Ellstrand's research associate, Janet Lee, performed a sophisticated biochemical technique to determine their genetic composition.
Because Ellstrand already knew the genetic contribution of the mother, he could use a simple technique to determine which plant in the study area was the father. And that, in turn, told him the minimum distance the pollen had to travel to fertilize the blossoms.
In the past, Ellstrand says, biologists generally assumed that pollen travels an average of one to two meters. "And it was anticipated that the gene flow at 100 meters would be 1% or less." Yet, Ellstrand's experiment showed that the pollen travel at 100 meters was closer to 20%. For 1,000 meters it was 9%.