IN THE YEARS following World War II, teenage boys

developed a passion for hot rods--firebreathing screamer machines with jacked-up rear ends, chrome headers and multiple four-barrel carburetors. Inevitably, many of the kids who bought hot rods couldn't really afford them. And there were other drawbacks. Hot rods were impractical and difficult to keep running. When they did run, they often caused accidents, and were a menace to their own drivers as well as others.

But there was one thing about hot rods you couldn't deny: they were really neat. Sometimes something is so neat you've just got to have it, whether it makes any sense or not.

According to an exceptional new book, The Nuclear Barons, much the same kind of rationale brought us nuclear power plants. The authors, British journalist Peter Pringle and former Australian government official James Spigelman, argue that making electricity from atoms never made much sense, either economically or technologically. Yet utility executives and government officials the world over have longed for it, like boys dreaming of hot rods. They've sunk hundreds of billions of dollars into it, then often looked the other way when their dream machines produced more grief than power.

Nuclear Barons begins with a familiar history of atomic weapons development in the United States and in other nations, then goes on to establish the connection between weapons projects and the development of commercial nuclear power. It is here that it is most interesting.

Postwar military officials urged commercial uses for the atom both to get the public behind nuclear power as a positive, not destructive, force, and to help justify their research budgets before Congress. Big companies, like General Electric and Westinghouse, competed throughout the 50s to adapt military reactors for commercial uses. They were motivated not so much by market analysis as by the desire for engineering prestige. Even at that time there were indications that nuclear power stations would be overly expensive to build and operate. But logic was yielding to atomic age frenzy around the globe. Nuclear advocates were predicting everything from unlimited electricity to irraditated foods that never spoil to cures for cancer, rehumatism and lumbago.

In this atmosphere, some outlandish proposals were taken seriously. Washington subsidzed the Savannah, a nuclear-powered cargo ship, since moth-balled, and even sank $1 billion on nuclear airplane research, which bogged down when no one could solve the problem of lead shielding around the plane's reactor. It weighed so much the plane couldn't take off.

Pressure on utilities to "go nuclear" was considerable. Reactor suppliers made technical promises that couldn't be kept; for instance, despite the fact that the largest operating reactor produced only 200 megawatts of electricity, utilities were assured there would be no problem "scaling up" the designs to produce 1000 megawatts. Pringle and Spigelman chronicle how in 1964, Jersey Central Power and Light opened a station in Oyster Creek, New Jersey, that supposedly proved the "commercial viability" of nuclear power. It was the first reactor built without government subsidies. But it later turned out Oyster Creek's reactor had been delivered by GE at an artificially low price--the atom industry's first "loss leader."

Nuclear power's biggest breakthrough came in 1966, when GE sold two reactors to the coal-and hydro- power-rich Tennessee Valley Authority. The contract stunned utility executives everywhere, and they scrambled to make nuclear arrangments of their own. Yet when they asked for deals like the one TVA made, Pringle and Spigelman say, they were told such prices were no longer available.

As reactor orders began to escalate in the years after Oyster Creek, prices also escalated--dramatically. American Electric Power, a midwest utility, found the reactors it ordered during post-TVA lightheadedness came in at three times the promised cost (and were finished several years late). AEP president Donald Cook, who had championed the nuclear plan, later said, "We are delighted to have two nuclear plants. We are also delighted not to have three." Overruns became a nuclear tradition. The Seabrook plant, bid at less than $1 billion, will cost $5 billion or more to finish. And completing a nuclear station, utilities found, didn't necessarily mean generating marketable electricity. Nuclear plants had so many problems and needed so much maintenance, they seldom ran at full power. (Today nuclear facilities run at 60 percent of capacity, on average.)

Nuclear Barons also looks at, but not closely enough, the issue of net power. In the 1960s the gaseous-diffusion factories used to "enrich" uranium ore into reactor fuel were turning out to be voracious consumers of electricity themselves--each one consuming as much power as New York city, according to Pringle and Spigelman. Uranium preparation today still consumes a great deal of electricity, although there are differing estimates of just how much, or how it affects the net power actually produced.

Foreign companies too have been taken in by nuclear power. According to Nuclear Barons, Britain decided to build expensive gas-cooled rectors simply to have something different from the American and Canadian water- cooled designs. These reactors took more than a decade to construct and scarcely ever function. Electricit,e de France, the state-owned French utility, leapt into nuclear power, committing itself to an "all nuclear" company motto. India dedicated its meager resources to reactor construction, Indian government officials speaking of "international prestiage" while ignoring the fact that coal-fired plants would bring far greater benefits to their people. In 1974 the International Atomic Energy Agency reported a projected demand for 140 nuclear power plants in the Third World. IAEA's "market survey" predicted, among other things, that Bangladesh would build 10 major nuclear generators. At the time Bengladesh's power grid could not have handled the output of a single reactor.

The authors attribute most of nuclear power's appeal to the lure of "Big Science." They write, "Utility executives had been trapped into going nuclear by a seductive combination of manufacturer's propaganda, fascination with new technology, and questions of coporate prestige." It is here that Pringle and Spigelman are most convincing, where so many critiques of the nuclear industry falter. They understand that human foibles and well-intentioned blunders, not sinister conspiracies, lie at the heart of the nuclear mess. It may seem hard to believe that so many intelligent and well-informed industry and government officials all fall for nuclear power. But unfortunately, modern hierarchies actually favor the most expensive, impractical and unreliable way to do things. Expensive projects make for big budgets, and empires. Likewise, the more complicated a project, the harder it is for outsiders to question.

In the end, what's most troubling is that the kind of irrational decision-making described in The Nuclear Barons may soon see an upsurge. Nuclear power construction has been declining for the last few years, as utilities (and their investors) finally came to their senses about the atom's combination of high cost, great dangers and relatively meager benefits. But now President Reagan and Energy Secretary James Edwards have committed the country to a "revitalization" of nuclear power. One of the few non-defense items increased by the Reagan budget was nuclear funding.

If Reagan's policies hold, the technology that still supplies only about 4 percent of the energy consumed in this country will continue to draw the bulk of government energy funding, tie up billions in badly needed utility investment capital, and monopolize our best engineering minds.

But after all, it's Big Science.