A QUEST FOR QUAKEPROOF BUILDINGS.

Ten feet wide by about 25 feet high, the narrow concrete-block wall stands in the center of a cavernous laboratory. Heavy steel cables run from the top of the wall to the floor. Its surface is peppered with small, silver sensors. Dozens of black wires run from them and are pig-tailed together on the concrete floor.

The brief countdown echoes through the lab: ''Five, four, three, two, one.'' Abruptly, the wall begins to shake. It snaps back and forth wildly. The heavy cables pop. The mortar between the blocks turns to dust, puffing into tiny white clouds. After 15 or 20 seconds, the shaking subsides. Cracked and battered but triumphant, the masonry structure remains standing.

It is a remarkable performance: The wall has just been shaken with forces roughly equal to what buildings in San Francisco experienced during the devastating 1906 earthquake, says Hugh D. McNiven. Dr. McNiven directs the Earthquake Engineering Research Center (EERC) operated by the University of California (UC) Berkeley, where this test took place.

An ordinary concrete-block wall would have been reduced to rubble. But this one has been strengthened with three inexpensive iron reinforcing rods. It is just such nuts-and-bolts details that determine whether a building stands or falls during a major earthquake - details that could prove crucial to a good portion of the US population.

According to the latest statistics, 15 percent of the nation's population has settled in areas where earthquake faults are most active. Another 20 percent reside in regions where seismic activity is moderate. A majority, 57 percent, live in locations of minor earthquake activity, but only 8 percent have chosen places where the seismic activity is negligible.

Furthermore, a disproportionate amount of building is going on in seismically active regions. About 35 percent of the roughly $230 billion being spent annually on construction is occurring in areas of high or moderate seismic activity.

Even people living in safe areas could be financially affected by a great earthquake elsewhere in the country. Four years ago the Federal Emergency Management Agency (FEMA) evaluated the consequences of four potential California earthquakes. It concluded that somewhere between 3,000 and 23,000 lives could be lost and property damage could range as high as $69 billion in a single incident.

Despite the uncertainty of these figures, there is general agreement that great earthquakes represent the largest potential for social disruption, short of war.

So much so that the National Security Council has ordered FEMA to reevaluate the potential of California quakes not just for direct damage but also for indirect economic loss to the rest of the United States economy due to disruption of the state's crucial financial and manufacturing capability.

Most Americans consign earthquake activity to California. But, according to seismologists, areas such as Seattle, Salt Lake City, St. Louis, Charleston, S.C., and Boston are also in seismically active zones.

Between 1811 and 1812, Missouri was shaken by three devastating quakes greater than magnitude 8 on the Richter scale. And nearly a century ago, Charleston, S.C., trembled with a destructive and still unexplained earthquake. These affected areas 10 times the size of the great California quakes, explains Dr. James Beavers of Martin Marietta Energy Systems Inc., who is organizing a conference on the subject.

New England is also subject to frequent, small temblors that so far are not well understood. In 1755, Boston was hit by an substantial quake that did considerable damage. According to Robert Whitman of the Massachusetts Institute of Technology (MIT), a repeat of this quake would kill 50 to 100 people and do extensive damage, particularly to the older brick and masonry buildings in the city.

Earthquakes also occur frequently in many other countries. In the last 20 years, Yugoslavia has suffered earthquake-related economic losses that amount to 1.5 percent of its annual gross domestic product, reports Yugoslav scientist Jakim Petrovski. And Japan is spending an estimated $1.7 billion to prepare Tokyo for a large quake anticipated within the next 10 years.

Skyrocketing projections of the cost of potential earthquake damage, coupled with the growing sophistication of civil engineering, has led to increased efforts to engineer buildings, bridges, pipelines, and other structures to withstand even the severe earthquakes that infrequently occur.

As disciplines go, earthquake engineering is very young. It was begun in the 1950s. The first international conference on the subject was held in 1956. Two hundred people attended. The most recent gathering occurred late last month in San Francisco and attracted 1,600 experts from around the world.

Still, enough progress has been made so that the experts, in a recent report by the Earthquake Engineering Research Institute, maintain that implementing the engineering procedures that have been developed could reduce loss of life in California quakes by 90 percent and cut property damage by 25 percent or more.

This does not mean that all the outstanding questions about earthquake proofing structures have been answered. Considerable uncertainties remain.

Advances have been hindered by several factors. For one thing, major earthquakes are extremely rare and unpredictable. So real-world tests of current building practices come infrequently. And there is no substitute. Unlike automobiles, airplanes, or most other products, it is not feasible to build and test a prototype of an office building, freeway overpass, or similar structure.

So civil engineers have been forced to adopt a number of less straightforward methods, computer modeling coupled with tests on scale models and various parts of a building. The world of the earthquake engineer is one of massive machines that crumble concrete and twist steel beams to determine their breaking points.

The masonry wall test at EERC is an example. The wall was built on a device called a shaking table. Berkeley's table is typical of those that have been set up around the world. It is a 20 foot slab of concrete that weighs 20 tons and can hold up to a 20-ton structure. The slab is floated on a cushion of air and hydraulic pistons shake it horizontally and vertically. These are controlled by a computer, so the pattern of ground shaking can be varied to simulate different types of earthquakes.

Sitting next to the shaking table is a 1/5-scale model of a seven-story concrete building built and subjected to 68 shaking tests as part of an ambitious $12 million joint US-Japanese program, which is just ending. This is an engineering model, basically just the walls and floors built meticulously to scale. It is chipped in innumerable places and veined with cracks that have been carefully traced with different-colored felt pens. This is an exact duplicate of a full-size building constructed in a laboratory in Japan. The building was erected next to a wall designed to be extremely rigid. A series of massive hydraulic pistons were connected between the wall and the building. These were used to shake the wall, but in ultra-slow motion.

''It took us more than a week to run a 15-second earthquake,'' explains Robert Hanson of the University of Michigan, the cheerful bespectacled engineer acting as the US coordinator of the program.

Each time the pistons push the building a short distance, a battery of sensors record how the building responds. From this a computer calculates how far each portion of the building would move in the next fraction of a second in an actual quake and instructs the pistons to push the building into this position.

Perhaps the most significant result of this program is validation of the scale-model shaking-table tests.

It ''demonstrated that scale models work much better than I expected,'' Dr. Hanson says. When properly done, model tests accurately estimate the overall response of a building, although it will differ in some specific details. This is important because many civil engineers have questioned the validity of such tests.

One of the pioneers in the field, Vitelmo Bertero of UC Berkeley, sees another implication. Dr. Bertero says that ''it has clearly shown that reinforced concrete can be used successfully for earthquake resistance.''

There have been several cases where modern concrete structures have failed in moderate earthquakes. This has raised questions about concrete's use in high seismic zones. If several provisions in the earthquake code are tightened up, it will take care of this problem, the engineer says.

The Japanese and US engineers involved agree that these tests generally validate the adequacy of the engineering codes established in the two countries, although they may lead to some revisions.

''All of these things are basically confidence-building measures. All of us think the current codes are pretty good. But we need proof,'' says MIT's Whitman.

Basically, the engineering fraternity says it feels that the large buildings designed and built to the seismic codes in the last decade will ride through major earthquakes without collapsing. In the case of many modern skyscrapers, they calculate that the forces inflicted by large wind storms are actually greater than those expected from an earthquake.

In fact, emphasis in the research is moving from the structural to the nonstructural portion of the building.

''Now that buildings no longer collapse, we can see what happens to the architectural systems. We've discovered that the nonstructural damage to a building can add up to 75 percent of the building's replacement cost,'' explains Henry J. Lagorio, professor of architecture at UC Berkeley. This has sparked considerable research into upgrading the design of windows, utility piping, and partitions, among other things.

The cost of all this additional engineering is high. It can add anywhere from 3 to 30 percent to the cost of a new building. As a result, many new structures being built in earthquake zones in the Midwest and Eastern US are not being built to withstand ground shaking.

''Engineers know they need to consider earthquakes,'' Dr. Beavers explains. ''But many owners do not want to do it because of the additional cost. And in most parts of the country they are not obligated to have it done because it is not in the building code.''

While many buildings appear to be seismically underdesigned in the East, there is a feeling that structures in California tend to be overdesigned.

''Most people don't understand that seismic risk has two parts: what nature does and what the consequences are,'' explains Haresh C. Shah of Stanford University. ''In this part of the world we overdo. The earth really shakes, but we are very prepared. If you go back East ... the hazard may not be so large, but people have become very complacent. As a result, the risk there can be even larger than it is here.''

Much of the work going on now is designed to reduce to price tag for earthquake protection.

Dr. Bertero, for instance, says he believes that the tests in the cooperative US-Japan program will make it possible to construct earthquake-resistant buildings less expensively. In concrete buildings, for instance, he argues that the tests have proved that certain provisions, such as putting heavy beams under the floors right next to the walls, can be relaxed and that this will reduce the cost of construction significantly.

One innovative cost-cutting approach to earthquake protection has generated considerable expert interest lately. This is nothing less than mounting buildings on large rubber shock absorbers, a concept called base isolation.

''An analogy would be a spring and shock absorber system on a car,'' explains Ronald Mayes of Dynamic Isolation Systems, one of the concept's proponents.

Several dozen buildings, a number of bridges, and two French-designed nuclear power plants have been built using this concept in various parts of the world. Its first use in the United States will be a county courthouse in southern California, now under construction. So far, none of these structures has been subject to a major earthquake. But tests on shaking tables and computer modeling have convinced many engineers that it has considerable merit.

In essence, this involves putting a series of large shock absorbers between a 3-to 20-story building and its foundation. These absorbers are made of thin layers of natural rubber sandwiched between steel plates. This makes them very flexible in the horizontal direction and very rigid in the vertical direction. As Dr. Mayes explains, lateral ground shaking is the most destructive. So the rubber absorber allows the ground to move back and forth while cutting down the force transmitted to the building to 1/5 or 1/10 what it otherwise would be.

''Base isolation can lead to cost reductions of up to 30 percent in building design because the reduction in force exerted on the structure means that less expensive building methods can be used,'' Dr. Mayes says.

As good as it sounds, there is one major source of uncertainty that makes the experts view even promising new ideas like this with considerable caution. Actual recordings of how the ground shakes in major earthquakes remain extremely rare. Ordinary seismographs, which measure the intensity of earthquakes thousands of miles away, are too sensitive to capture the violent ground motions near a quake's epicenter. Special detectors are required.

According to Gerald Brady of the US Geological Survey (USGS), records from the 1971 San Fernando earthquake in California doubled the amount of data on ground shaking. And the Imperial Valley quake in 1979 was the first to be thoroughly instrumented and has given even more precious information.

But these were both moderate events. And as Bruce Bolt, head of the UC Berkeley seismology department, points out, the big question is how ground motions in the great earthquakes differ from those in smaller and more frequent tremors. For this reason, he argues that the country can't afford to improperly record any future major California quake.

Roger Borcherdt of the USGS agrees: ''If the instrumentation is in place for the next big earthquake, we could have data for a real breakthrough in our understanding of building failure.'' Such a program would run between $10 million and $20 million a year and is currently under discussion within the federal government.