There is an established ritual when an airline crashes. Swiftly, some slight component is produced and anointed as the author of the disaster. The May 25 crash of a McDonnell Douglas DC10 in Chicago evoked a classic response. According to the press, 200,000 kilograms (220 tons) of airplane and 275 people came to ruin because of the fracture of a bolt just "the size of a man's finger."

Two desires are satisfied by this ritual. Newspapers like drama along for-want-of-a-nail-the-kingdom-was-lost lines. And - assuming "human error" cannot be blamed - the aerospace industry prefers any account of an accident which avoids the idea that large, systematic deficiencies might exist in any of the costly machines it builds. But this time the ritual collapsed.

The Federal Aviation Administration began with a mild request that all DC10s be checked for the state of their engine-pylon flutter-bolts. These are indeed no bigger than a human finger - though rather easier to replace - and operators throughout the world, particularly Britain's Freddie Laker, whose transatlantic Skytrain depends on the DC10, were eager to be shown complying.

Then, four days after the crash, the FAA ordered all DC10s grounded, pending complete examination of the pylons structures which hold their engines in place. Evidence of metal fatigue was so grave that the administration had "no choice." Neil Callaghan of the FAA office in Chicago admitted that they were worried about something more than a bolt."We are worried about the whole airplane," he said.

About time, some might say. When a Turkish Airlines DC10 crashed outside Paris in 1974, killing 346 people, there was the same initial concentration on small bits of metal. There, it was an ill-designed lock which allowed a cargo door to blow off. But the Paris plane did not die because of a lost door, any more than the Chicago plane died because an engine fell off. In each case, catastrophic loss of control followed from the failure of a component that was not in itself essential.

This is the crucial link between the two events, and the evidence for saying that the DC10 has weaker defenses against disaster than the other "airbus" jets.

Paris, loss of the cargo door depressurized the cargo hold at 12,000 feet.High-pressure air in the passenger cabin above then collapsed the floor between the two compartments. The hydraulic power lines and controls, running from cockpit to tail, are carried under the floor in a DC10. They were severed, and the plane dived into the forest of Ermenonville.

At Chicago, a similar end followed from a different beginning. Pylonmounted engines are slung below and ahead of a wing. The weight of the whole assembly in the DC10 is fixed to the wing by a tough group of vertical bolts called the "tombstone." When the engine is thrusting forward, it tries to "swing" upon the tombstone, and will pitch up over the wing unless restrained by other fastenings at the rear.

The pylon assembly, holding a enormously heavy engine out in front of the wing, is also apt to "flutter" from side to side as the engine pours out its 18,000 kilograms (39,600 pounds) of thrust. These movements are restrained by horizontal bolts, or pins, at the point where the pylon joins the leading edge of the wing. One of them broke away from the left-hand engine pylon as American Airlines flight 191 passed the 8,000-year mark in its take-off run at O'Hara International Airport on May 25.

Flutter built up very quickly with the engine at takeoff power, and within seconds the vertical bolts at the rear of the pylon snapped. The engine tore loose, pitched over the top of the wing, and fell away. The DC10 was just lifting off the runway: the point at which the second pilot, watching the instruments, calls out "VR" to the captain. (VR is an abbreviation for "velocity for rotation": raising the nose for takeoff is "rotation" in pilot's language.) On this occasion, the second pilot said: "VR-damn." After that, the cockpit voice recorder went blank.

Airline spokesmen have been swift to explain that loss of one engine need not prevent a three-engined plane from taking off safely - which is true. They have been less ready to explain why the DC10 crashed nevertheless. But inquiries with crash investigators suggest that it happened because, as the engine reared up over the wing, it wrecked the power lines and hydraulic control lines carried just head of the wing's main spar. Without hydraulic power, a modern jetliner cannot move any of its control surfaces. The crew must have been quite helpless as the DC10 climbed for a few more crazy seconds, then rolled on its back and smashed into the earth.

Only detailed investigation will show whether loss of all control was inevitable. But already there are questions about the relative vulnerability of the DC10. According to the FAA-inspection has shown evidence of serious metal fatigue not just in "finger-sized" flutter bolts, but elsewhere in the DC10 pylon assembly. Metal fatigue is what one sees if one twists a paper clip backward and forward until it breaks: elaborate computer calculations are supposed to insure aircraft structures against it.

So far there has been no directive against the other main "jumbo jets" - the Boeing 747, and the Lockheed Tristar, which is the DC10's direct, and so far unfavored, rival in the marketplace. Both use pylon engines, but apparently without exhibiting dangerous metal fatigue. Both have the hydraulic lines and control lines, where they run along the wings, mounted aft of the heavy main spar: safer, if it is assumed that impacts will more often come from the front. And both have four separate hydraulic systems, each capable of flying the aircraft, where the DC10 carries only three.

These other planes, in short, have greater defense in depth against such an accident than appears to be the case with the DC10. Neither is as likely to lose an engine through pylon failure: in each case, if it were to happen, the airplane would be more likely to remain controllable. Aircraft safety is largely about having defenses in depth. Comparison after Paris showed that both the Tristar and the 747 had deeper defenses against floor collapse (indeed, the 747's controls were almost immune to such an event), and the available evidence from Chicago reinforces the message.

Most aircraft accidents result from unhappy conjunctions of human error, bad weather and perhaps sloppy design of some particular instrument or control. That description applies reasonably to all the crashes so far on the Boeing 747 and the Lockheed Tristar. But twice now a DC10 has fallen out of a clear blue sky because of catastrophic failure of the whole control system, which left the pilots helpless. And grace alone has averted at least two other disasters, which might have pushed the DC10 toll past 800.

In June 1972, in an incident which clearly foreshadowed the Paris tragedy, an American Airlines DC10 lost a rear cargo door over Windsor, Ontario. The cabin floor collapstd, severing most of the "plumbing." Only because the aircraft was rather lightly loaded, and because the captain displayed extraordinary skill, the aircraft got down without anyone being fatally injured.

And in November 1975, 139 people narrowly escaped when an engine of an overseas National DC10 disintegrated during takeoff at Kennedy Airport in New York. The pilot managed to stop the airplane before hitting the end of the runway - despite the fact that hydraulic failure robbed the plane of about half its braking power. The undercarriage collapsed, and the fuselage was enveloped in flames. But the passengers were all airline employes, practiced in evacuation procedures, and no one was seriously hurt.

After the Windsor close call an engineer named F. D. Applegate, who worked for a DC10 subcontractor, wrote his now famous memorandum predicting that the cargo door would blow open, that cabin floors would collapse and that "I would expect this to usually result in the loss of the airplane." At Paris, it did. After New York, the National Transportation Safety Board, which conducts the crash investigations, concluded that the DC10 engine was "susceptible to catastrophic failures."

The DC10 was born out of one of the most savage marketing conflicts in civil aviation history. In the 1960s, it became clear that big fan-jet engines would make possible a new generation of "wide-bodied" airbuses. Boeing established a commanding lead in the long-haul section of that market with the remarkable 747. Two ailing firms were left to struggle for the medium-haul business: Lockheed, which was trying to return to the civil market after years of over dependence on defense, and Douglas, once the world's greatest builders of commercial aircraft, reduced to chaos and penury by the eccentricities of its founding family.

Douglas was taken over by the aggressive military aircraft builders McDonnell and, late in 1967, the new McDonnell Douglas Corp. announced that it was going to try to catch up on the lead of nearly one year that Lockheed had established in the race to get orders for a three-engined airbus. This was a contest of "paper aircraft," in which both sides made larger and larger promises: when Lockheed slashed its price by a million dollars a plane, McDonnell Douglas countered with a half-million cut and other attractions.

Neither firm had decided which make of engine to favor for its theoretical airliner, so rival engine manufacturers entered the field with further promises to both airframe firms and airline companies. What nobody really expected - or wanted - was that both paper planes would actually be built. For both Lockheed and Douglas to make a proft would require 500 trijets to be sold.

At first the Lockheed Tristar seemed to have a decisive lead in orders. But then United, the largest airline in the western world, chose McDonnell Douglas's DC10. Thus the paper chase had been inconclusive: both aircraft would have to be built, with success likely to go to the plane that got into airline service before the other. Under McDonnell management, the motto at the old Douglas works became "fly before they roll" - that is, have the DC10 flying before the Tristar could even be rolled out of its hangar. And not only did they wipe out Lockheed's initial lead, they produced the DC10 nine months earlier.

This remarkable speed was a major factor in the DC10's superior market performance. There are now 274 DC10s in service compared to 163 Tristars. McDonnell Douglas was on the verge of making a profit with the DC10, while Lockheed has a long way to go.

The promise of the new, big jets was that they might be safer than any aircraft seen before. It may be doubted whether McDonnell Douglas allowed time enough for all the technologies to mature that could make that promise true.McDonnell Douglas designed, with its subcontractors, a highly inadequate system of cargo door locks. Both the Tristar and the 747 could claim to have included new and substantial safety features to go with their massive increases in size and power. In the case of the DC10, for all its commercial success, it is not easy to find substantial improvements on its smaller and simpler predecessors.

The world aviation industry would be brought close to ruin by taking the DC10 out of service. So successful was its sales campaign that many airlines throughout the world, like Lufthansa and Swissair, could scarcely manage without it. Small firms like Britain's Laker might be financially demolished if the DC10 could not fly. On average, there are 750 DC10 flights every day, operated by 41 airlines around the world. If the plane were to be taken out of service, the industry would have to find a million extra seats a week. And the truth is that the airplane would have to be a great deal more lethal before the aerospace business could be made to face up to such a possibility. CAPTION:

Picture, A newly ground United Airlines DC10 sits on the apron at Dulles International Airport. U.S. carriers operate 138 of the wide-bodied jetliners. By James A. Parcell - The Washington Post