If a nonsmoker gets lung cancer, his or her parents' cigarette habit may be to blame.

According to a study in today's New England Journal of Medicine, nonsmokers who grew up with two parents who smoked are twice as likely as other nonsmokers to develop lung cancer.

This is the first study to find an association between exposure to someone else's cigarette smoke during childhood and the development of lung cancer later in life.

The study estimates that about 17 percent of all lung cancer cases in nonsmokers are caused by exposure to others' cigarette smoke during childhood and adolescence.

According to figures from the Environmental Protection Agency only about 6 percent of those who get lung cancer are lifelong nonsmokers. They account for about 10,000 new cases of cancer each year. Another 16 percent, or about 25,000 cases, smoked at some time but had quit. About 83 percent, or 130,000 cases, occur in smokers.

"This adds to the evidence that involuntary smoking is harmful," said Peter Greenwald, a National Cancer Institute researcher who was an author of the report. "For the first time it provides evidence that exposure as a child increases the risk of lung cancer later in life."

Previous studies have indicated that children of smokers are more likely to get bronchitis and other respiratory tract infections, but had not been able to trace cancer risk because of the long time between exposure to smoke and development of cancer symptoms, Greenwald said.

To trace that risk, this study used what is called a case control technique. It looked at 191 nonsmokers who had lung cancer and 191 randomly chosen nonsmokers who did not. The healthy nonsmokers were selected by screening the files of the New York State Department of Motor Vehicles.

The researchers then surveyed the subjects to find out how many had parents or spouses who smoked.

They found that people who lived with two smoking parents for 12 years or more had twice the rate of lung cancer than those with nonsmoking parents. People with parents who smoked for a shorter period of time and people with spouses who smoked were not at increased risk, the study found.

Lung cancer researchers said the increased lung cancer rate in people exposed to smoke as children was consistent with other studies which have found an increased rate in people exposed as adults.

Their study, however, found that people with smoking spouses had no increased risk of lung cancer. The researchers said they found this puzzling because many larger studies have found that spouses of smokers run between a 10 percent and 30 percent increased risk of lung cancer.

"I would go by the accumulation of evidence of a number of studies that there is an increase in risk," Greenwald said.

Researchers said children may be more susceptible to other people's cigarette smoke because their lung tissue is still developing, but they said there is no proof for this theory.

"For many kinds of cancer, tissues are more susceptible during their most rapid growth phases during childhood, but for other cancers that's not the case," said Ronald M. Davis, director of the Centers for Disease Control's Office on Smoking and Health, who was not involved with the study.

A Tobacco Institute spokesman, Thomas J. Lauria, said the study's finding that spouses were not at increased risk reinforces the cigarette industry's opposition to restrictions on smoking. But Greenwald said he thought that was a misinterpretation of the findings.

"The Tobacco Institute tends to try to distort everything to deny risk," he said. "The take-home message {of the study} is . . . it's best that parents don't smoke. {If they do smoke}, it's best that they try to avoid smoking in closed places like the car where kids are forced to breathe in a lot of smoke."