The reason for the rule on mercury releases into the environment is not mercury in coal smoke, streams and lakes or fish. It's the purported neurological illness in children whose mothers eat fish high in mercury. With mercury emissions declining, and no evidence of the illness justifying current rules in the United States, why contemplate expensive, tougher rules on mercury releases?

The federal Environmental Protection Agency calculated a "benchmark dose" (one that begins to cause neurological illness) for mercury in blood based on levels in women of the Faroe Islands who eat whale meat. Those island women who had 85 parts per billion (ppb) - more often reported as micrograms per liter or µg/L - or more of mercury in their blood were subject to have babies who scored lower on a neurobiological test (Boston Naming Test). Because 85 ppb is a statistical estimate of the "benchmark dose," the real threshold could be higher or lower.

That real threshold was calculated to be within the range of about 27 ppb above or below 85 ppb, with a confidence level of 95 percent, meaning that the level that begins to cause neurological damage could be as low as 58 ppb or as high as 112 ppb. EPA adopted the lower (safer) "benchmark dose" of 58 ppb and then introduced a safety factor of 10 to set the "safe level" at 5.8 ppb.

Setting such a low "safe level" of mercury in blood is a two-edged sword. It is intended to protect but is used by activists to alarm. The government clearly doesn't know which side to take, and sometimes it's the activist. A government-published scientific journal, Environmental Health Perspectives, asserted in 2005: "A recent study from the Centers for Disease Control found that as many as 637,233 American children are born each year with mercury levels of more than 5.8 µg/L, the level associated with brain damage and loss of IQ."

Yet CDC data showed otherwise. The CDC's Third National Report on Human Exposure to Environmental Chemicals in 2005 said of mercury, "data for the period 1999-2002 show that all women (3,637 tested) of childbearing age had levels below 58 μg/L, a concentration associated with neurologic effects in the fetus." The CDC apparently is asserting both that 637,233 children are in danger of mercury poisoning and that none are.

Another large study in the Seychelles Islands with even higher mercury levels than women in the Faroe Islands was not used in the calculations of safe levels. Undoubtedly it was because no effects of mercury on children's health were found.

Americans rightfully expect U.S. studies on health problems they face from mercury. And U.S. studies found mercury levels are too low in blood of Americans to show damage to health. In the Faroe Islands study on which mercury rules are based, the average level in blood was 22.9 ppb; in American women of child-bearing age the average for 1999-2002 was less than 1, nowhere near as high as the levels that caused health problems in the Faroe Islands women.

The lack of a health problem is emphasized not only in the low levels in blood, but the fact that those levels are dropping. Mercury emissions in the United States decreased 48 percent from 1990 to 1999. After an EPA report to Congress in 1997 called for further study of mercury dangers to Americans, CDC began a survey of blood mercury levels in young children (1-5 years old) and women (16-49 years old) in 1999. Mercury levels declined. The average for women in 1999 was 1.2 ppb; for 1999-2000 the average was 1.02, for 2001-2002 it was 0.83.

Even in this curious way of averaging there was at least a 30 percent reduction in four years, and the decreases occurred in every category - male and female children, and Hispanic, white and black women. Equally significant, the percentage of women with blood mercury above the "safe level" of 5.8 ppb was cut in half, from 7.8 to 3.9 percent. Still, CDC was not certain; the decline in four years was declared statistically insignificant.

But an EPA scientist seemed to accept the decline in the following question presented in a 2005 conference; "Does the decline reported in blood mercury between the 1999/2000 release and the 2001/2002 release reflect the ratio of coastal to non-coastal residences or other study design considerations?" Apparently the CDC had seen enough though; blood mercury has not been measured - or at least not reported - since 2002. Progress, it seems, has its limits.

A health problem from mercury has not been demonstrated in the United States. We ought not argue about costly fixes for a problem that is non-existent, where the potential for problems is diminishing, and certainly not based on data borrowed from small island populations with whom we have little in common.

University of Georgia Professor Emeritus R. Harold Brown is an Adjunct Scholar with the Georgia Public Policy Foundation and author of "The Greening of Georgia: The Improvement of the Environment in the Twentieth Century." The Georgia Public Policy Foundation is an independent think tank that proposes practical, market-oriented approaches to public policy to improve the lives of Georgians. Nothing written here is to be construed as necessarily reflecting the views of the Georgia Public Policy Foundation or as an attempt to aid or hinder the passage of any bill before the U.S. Congress or the Georgia Legislature.