Radiation levels inside the damaged Three Mile Island nuclear power plant are vastly lower than expected, a finding that could save the plant's owners about $25 million.

The readings are so much lower, in fact, that they are raising scientific questions about the behavior of certain radioactive elements during the reactor malfunction there March 28.

A November probe by remotely controlled instruments into the Pennsylvania reactor's containment building was the source of the figures. It found gamma radiation, the most penetrating and dangerous type, was 0.7 rads per hour, a level 1/500th to 1/3000th of the estimates made in July by a Bechtel Corp. study. The study said 320 to 2,400 rads could be expected. A 500 rad/hour level is considered fatal.

Similarly, the report's estimate of beta radiation, which can be stopped by minimal protective clothing, was about 100 times too high.

The Bechtel study was considered authoritative until the actual measurements were made, and many estimates of cost, time and manpower needed for the cleanup were based on it. The latest estimates before this finding involve four to five years and at least $400 million to direct expenses alone.

The new readings mean that a high-pressure spray system to wash down all the interior walls of the containment building will not have to be used, according to Robert C. Arnold, senior vice president of Metropolitan Edison Co. in charge of the utility's cleanup and restart effort.

Workers will be able to enter much earlier than planned and will be able to work three to four hours before receiving their three-month exposure limit, Arnold said. The cleanup effort will take about three months less than expected, which translates into $25 million, he added.

"It's a very important finding, Arnold said. "It means the initial stages of the cleanup can be approached faster than we expected."

Bechtel engineers apparently over-estimated the amount of radioactive material deposited on the inner walls of the containment building when the reactor lost its coolant and overheated. "It shouldn't really be seen as an error," Arnold said. "It wouldn't have taken much cesium floating around to get the radiation to the lower levels they were postulating."

Radioactive cesium, a product of nuclear fission, had been expected to escape from the reactor coolant water into the air when the accident occurred. Instead, it apparently remained in the water that is now six feet high and awaiting removal in the containment building basement.

Robert Rider, one of 10 Bechtel employes who compiled the July study and now project engineer for the consulting firm's recovery work, said the radiation predictions were based largely on one gamma radiation instrument monitor that proved to be defective.

"It was the only one designed to monitor high levels," he said."The others were all off the scale." The error, he said, was compounded by the assumption that more of the cesium would come out of the water than proved to be the case.

"It's more soluble than we expected," he said.

Existing radiation levels come largely from the krypton gas floating inside the reactor building, the probe showed. Metropolitan Edison has asked the Nuclear Regulatory Commission for permission to vent the gas after filtering and recovery, but several alternative methods of purging the gas are under NRC consideration and no decision has been made.