A small but groundbreaking study of infants who received vaccines containing a mercury-based preservative has found that the levels of mercury in their blood were well within the federal safety limits.

The study, reported last week in the Lancet, medical journal based in London, also found that infants excrete the mercury much faster than expected, suggesting that it does not build up from one vaccination to the next.

The preservative, thimerosal, is no longer used in U.S. vaccines for infants younger than 6 months old, but the issue is important to parents of children who did receive thimerosal-containing vaccines as infants and are now autistic. Thousands of those parents have filed hundreds of damage claims or lawsuits against thimerosal's maker. A clause protecting Eli Lilly & Co., the manufacturer, from lawsuits was mysteriously slipped into the domestic security law signed by President Bush on Nov. 25.

The director of the Institute for Vaccine Safety at Johns Hopkins University, Dr. Neal Halsey, praised the study as "much needed and done quite well," although more work needs to be done, he said, especially on underweight babies and long-term follow-up.

FINDINGS 'VERY MUCH OFF-BASE'

But Sallie Bernard, director of Safe Minds, a parents group suing the vaccine industry, vehemently attacked it, calling its optimistic conclusions "very much off-base."

Mercury is unquestionably poisonous. At extreme doses, it causes tremors and madness. Children who accidentally get high doses tend to speak and walk later and have tics and lower intelligence, but not autism, medical experts say.

Small amounts, however, are common in soil and plants, in power plant fumes and in dental fillings. Fish are the largest source for humans, and a tuna sandwich may contain more mercury than a vaccine shot.

No scientific study has proved that thimerosal causes any ill effects, but at the urging of federal health officials, vaccine makers began eliminating it in mid-1999. The study began with that recommendation.

Thimerosal, which kills fungi and bacteria, is still used to preserve vaccines sent to the Third World, and the World Health Organization defends it.

The vaccines prevent common diseases there, so the benefits far exceed potential side effects. Also, vaccines there must survive dirtier storage conditions and are shipped in bottles containing a number of doses to save money.

HOW STUDY WAS CONDUCTED

The Lancet study, led by Dr. Michael Pichichero of the University of Rochester, tested the blood, urine and stool of 33 infants ages 2 months to 6 months, all of them seen by Rochester pediatricians injecting thimerosal- containing vaccines. They were compared with 15 infants seen at a clinic in Bethesda, Md., that used mercury-free vaccines.

In their first six months, children typically receive three doses of diphtheria-tetanus-pertussis vaccine, one or two for hepatitis B, and sometimes up to three for haemophilus influenza. Other vaccines, such as polio,

may not contain thimerosal.

The Environmental Protection Agency's safe level for mercury in children's blood is 5.9 parts per billion. That, Pichichero explained, is based on a study of children in the Faeroe Islands, southeast of Iceland, whose mothers ate whale blubber polluted with mercury and PCBs. When the mothers had 59 or more parts per billion of mercury in their blood while pregnant, their children scored lower on intelligence tests several years later. The EPA took one-tenth of that -- 5.9 parts -- as a safe level.

All but one of the infants in the group exposed to thimerosal had blood levels of 1 to 3 parts per billion; the one exception went to 4.1. In the mercury-free control group, only one baby had even a measurable level of mercury.

The babies had high concentrations of ethyl mercury in their stools, indicating it was being rapidly excreted. Its half-life in the blood appeared to be seven days, which surprised the investigators, Pichichero said, since methyl mercury, the kind found in fish, has a half-life of 45 days.

Quick excretion suggests that the thimerosal mercury would not build up in the brain from a child's two-month shots to the four-month and six-month shots,

causing cumulative damage, said Halsey of the vaccine safety institute. That conclusion was reassuring, he said.

"I give the investigators lots of credit for acting quickly and doing a study that a lot of people didn't want to do," he added. "Who wants to sit down and say to parents, 'We're giving your babies vaccines containing mercury'?"

But Bernard of Safe Minds called the study "small and mediocre" and said it was alarming that The Lancet had given it such prominence.

She complained that Pichichero had done studies for Eli Lilly and other vaccine makers, that 33 children were too few to be conclusive and that blood had been drawn several days after immunizations, so peak levels might have been missed.

Bernard also pointed out that the child with the 4.1 mercury level was of high-normal weight and had received only about 60 percent of the thimerosal that children received under typical immunization regimens of the 1990s.

That the child approached the EPA's safe level suggested, she said, that a smaller child receiving all thimerosal-containing shots might exceed it.

Pichichero acknowledged that he had done antibiotic studies for Lilly and vaccine studies for other companies, but he said that this one was paid for by the National Institutes of Health.

He agreed that the size of the sample was relatively small and said a follow-up study of 200 children was under way in Argentina, where thimerosal is still used. Some children as small as four pounds would be included in that study, he said, and blood would be drawn within hours of vaccination.

He did not disagree that, theoretically, an underweight child might exceed the EPA safe level.

"We recognize those shortcomings," Pichichero said, but he argued that they did not invalidate his conclusions. Mercury levels in the body take days to stabilize, he said, adding that damage to fetuses is greater than to infants and that neurological damage is thought to result from long-term high mercury levels, not brief peaks.