The decision Defense Secretary Caspar Weinberger made to cancel the Army's divisional air defense gun, DIVAD, was a good one, and we salute him. The weapon did not work, everyone knew it, and had he decided in its favor or even temporized, he would have created a circus issue that neither the administration nor Congress needs right now. As it is, the secretary has to have helped himself with Congress, regaining some lost ground.

Internally, it cannot have been easy for Mr. Weinberger to have decided as he did. Major weapons systems are rarely killed. There is always a great deal invested in such programs, partly in dollars but also in careers. The decisions are all the harder when they come as late in the procurement process as this one. Mr. Weinberger blocked about $3 bilion in further expenditures on DIVAD, but in the process he wrote off $1.8 billion already spent. That is a lot to walk away from.

One lesson is probably that the services ought to have a better reason than the Army did in the DIVAD case to telescope their normal procurement procedures. They should test before they buy. To save time on DIVAD, it was decided to do both at once. Partly because DIVAD, like so many modern weapons systems, was so complex, that turned out to be the wrong thing to do. The good news on this front is that the tests -- and the department's new testing office -- did finally catch the weapon up. But earlier would have been easier.

Mr. Weinberger says there is still an urgent need for a weapon to do what DIVAD was supposed to: protect against the threat posed by the latest Soviet helicopters, which are thought capable of popping up and firing accurately at U.S. tanks and front-line troops from points several miles away. One interesting question is how elaborate a weapon the Army will decide it needs. The secretary suggested it might be possible to modify one of several existing weapons. That would be quite a come-down from DIVAD, but very possibly a reasonable thing to do.