Drawing from articles appearing in several national publications about the dioxin contamination found in Midland, Michigan, in l983, a case study examined how the publications handled the phenomenon of expert disagreement concerning the nature of dioxin. Specifically, it examined how the publications handled expert disagreement about (1) the way dioxin gets into the environment, (2) whether it causes cancer in humans, (3) whether it causes birth defects or miscarriages in humans, (4) the significance for human health of studies showing dioxin has severe and acute health effects in animals, and (5) the implications of regulation of the substance. The study discovered that some of the publications obscured the existence of disagreements among scientists or made no attempt to account for it, while others attempted to explain it by suggesting that one side or another was irrational or psychologically disturbed, that the science used by one side was incompetent, or that the science used was corrupted by politics, economic pressures, or a commitment to corporate power. The study concludes that journalists can benefit from the review of studies in disciplines other than journalism showing that expert disagreement is a normal aspect of science that can be accounted for in a number of ways. (FL)