Bicycle helmets aren't required by local laws, but they should be part of the standard riding gear for all cyclists, especially small children riding on hard surfaces and commuters threading their way through traffic.

"Close to half a million Americans are seriously injured every year in bicycle accidents; more than a thousand die from their injuries and something like three-quarters of the deaths are due to head injuries," said Thomas Balderston, author of a new study by the Washington Area Bicyclist Association and the Snell Memorial Foundation of Sacramento, Calif.

The right kind of helmet can protect bicyclists against such injuries, Balderston said.

But bicycle riders should shop carefully to be sure the helmet they select will provide that protection. Of 24 helmets tested during the study, only 10 passed the laboratory impact tests devised by the Snell Foundation, a nonprofit research group established in 1957 after sports car driver William H. Snell died from head injuries in a racing accident.

"Since there is no organized lab-testing program for bicycle helmets in this country (other than testing done by or for manufacturers), the organizers of this study arranged for the Snell Foundation to test the helmets in this report, using the method developed for the Snell motor-vehicle helmet tests (but with less severe impacts)," Balderston said in a report in the March issue of Bicycling magazine.

Working with Balderston on the study was Randy Swart, who helped organize the testing of the helmets for comfort, fit and coolness. For those tests, Swart arranged for a panel of Washington-area bicycle riders to wear the helmets and rate them for comfort, ventilation and the ease with which they could be put on and taken off.

For the impact test, the foundation researchers strapped the helmet to a head-shaped form containing recording instruments. The helmet then was dropped onto a steel anvil from one meter (3.28 feet), five feet and six feet. When the helmet hit the anvil, the instruments recorded the impact.

"The human head can tolerate about 300 Gs, so that is the maximum limit permitted by most helmet standards," Balderston said.

The G force is a measurement of deceleration or acceleration. The tests sought to pinpoint the drop distance at which a helmet would allow the force on the head to exceed 300 Gs.

Fourteen of the helmets failed the test.

Here are the brand names of the helmets that failed, in order of the best protection provided to the worst, as measured by the foundation lab tests: Skid-Lid II, Skid-Lid, Griffin, Schwinn, Hanna-Pro (old), Cooper SK600, Pro-tec Firefly, Pro-tec PTH4000, Pro-tec PTH3000, Norcon AP-2, Premier, Kunoh, Cortina Brancale and Sportivo.

Kevin Montgomery, president of Skid-Lid Manufacturing Co., which makes the Skid-Lid helmets, objected to the Snell Foundation tests on the grounds they were "grossly oversimplified." Montgomery contended that Skid-Lid construction does provide protection.

Six of the other helmets tested passed the impact test from a height of six feet. In order of protection provided, they were the Fury, Bell Biker, Bell Prime, Bailen (new), Bailen (early) and Bell TourLite.

Four other helmets passed the five-foot test. They were, in order of protection provided, the MSR, Hanna-Pro (new), Supergo and Cooper SK2000.

As a general rule, the helmets that did the best in the impact tests had hard protective contruction, usually plastic, and thick foam interiors. The ones that failed typically had less foam padding.

The average price for most helmets tested was $40 to $45. One exception was the Fury, a lightweight motorcycle helmet that had the best showing in the impact test. It has an average price of about $72.

Balderston cautioned against buying a helmet solely on the basis of the impact tests, however.

"Consider this data as guides to assist your selection," he wrote in his report. "The Snell Impact Test, which in our opinion is a meaningful measure given the current scientific thinking . . . , is one criterion you may choose to use. But don't ignore the other data.

"As practicing cyclists we know that many cyclists do not wear their helmets due to dissatisfaction with overall comfort, fit and coolness. Please consider the results of the lab tests within the total cycling environment."