A series of frustrating delays in readying the shuttle Discovery for flight has sharpened a complex debate within the space agency over whether some of its tough post-Challenger launch-preparation rules could be withdrawn without reducing safety.

Officials at the National Aeronautics and Space Administration say they have begun reversing certain new procedures that seem to have added bureaucracy and delay without enhancing safety or quality. They plan to reassess other requirements that many believe fall into that category.Officials emphasize that this does not mean they will sacrifice safety to meet a schedule.

"The problem is not that the safety objective is there," said Thomas Utsman, director of shuttle-management operations and No. 2 official at Kennedy Space Center. "But you have to look at what's buying you safety and what isn't."

"You can't be too safe," said astronaut Charles F. Bolden Jr., a Marine colonel who heads the astronaut-support team here assisting in launch preparations.

"But sometimes you can be too conservative," he said.

The debate has been smoldering for some time, as delay followed delay. The shuttles will have been grounded for nearly three years by the next launch - if it takes place in late September or early October, which many officials believe is the earliest possible time.

Utsman said quality and safety teams have agreed to some reductions in the stringent new rules. For example, instead of having to clear an entire area during certain assembly operations, a way was found to allow nearby crews to continue working on other jobs. In another case, the number of quality inspections on shuttle heat-shield tiles was cut back.

But it was what many consider a piddling leak discovered 10 days ago in Discovery's orbital-steering system that brought the frustrations to the surface.

The fact that the leak made it to the launch pad before it was found, despite telltale signs months earlier and despite the mountain of tests, inspections and signatures it had to get through, raised questions about the value of all these new "hoops."

"Adding extra people and paper work didn't prevent these screw-ups," said a key congressional aide, referring not only to the steering-system leak but to a mistake made earlier in July in pressurizing a solid-rocket-booster joint.

And the fact that a leak engineers say is less than the size of a human hair could threaten a major delay for the colossus of the shuttle-launch effort - the U.S. space program's top priority - raised questions about how far the agency can go toward caution before it reaches a point of diminishing returns in terms of a viable program.

"This casual attitude about delay, delay, delay - you can't afford a two-month delay" because of its effect on planetary and other payloads scheduled to fly next year, said another key congressional source, referring to the steering-system-leak problem. "You don't want to go back to the old ways, but I think the pendulum has to swing back a bit."