Scientists have discovered a drug that prevents the memory loss connected with drinking alcohol, according to a report released today.

The drug is the first to show such a clear effect on human memory and will "open whole new lines of research" into both memory and alcoholism, said Markku Linnoila of the National Institute of Alcoholism and Alcohol Abuse, a co-author of the paper that will appear in Science magazine next week.

Researchers say that the drug--a European antidepressant called Zimelidine-- may or may not lead to a substance that will reverse some effects of drinking. But they expect it to help them explore the brain system that controls memory deterioriation and other behavior, including violence, sometimes associated with alcoholism.

"We do not make any claims that this drug will reverse the effects of alcohol, but that doesn't stop us from trying" to find keys to alcoholism, Linnoila said.

Researchers said that although the drug reversed memory deterioration while a persons was drinking, it is not clear whether that kind of memory loss is the same as the long-term deterioration of memory in chronic alcoholics.

In previous experiments with animals, the new drug was found to have other powerful effects besides boosting memory.

In animals, it ended their preference for alcohol over water when a choice was given. In experiments with 16 humans, Edward Sellers and Claudio Naranjo of the Addiction Research Foundation in Toronto found a similar, but smaller, reduction in the motivation to drink. This finding eventually may be useful in treating alcoholism.

Zimelidine, according to the current hypothesis, works by affecting a key natural chemical called serotonin, which is believed to help store memories chemically in the brain and acts as a mediator in other behavior. It is found at abnormally low levels the brains of suicides and alcoholics.

In the new study, researchers put 10 volunteers through a battery of tests after giving them as many as seven cocktails' worth of alcohol and either a pill of Zimelidine or a placebo, a sugar pill.

Those who drank and had the placebo scored badly on memory tests as well as tests of balance and visual tracking.

But those who were given Zimelidine with their alcohol performed 65 to 80 percent better than their counterparts on some memory tests, such as one requiring them to remember a list of words after hearing it only once. Balance and visual tracking remained poor.

The researchers participating in the work were Linnoila, Monte Buchsbaum of the University of California at Irvine, Herbert Weingartner and Matthew Rudorfer of the National Institute of Mental Health.