We’re awash in neutrinos. They’re among the lightest of the two dozen or so known subatomic particles and they come from all directions: from the Big Bang that began the universe, from exploding stars and, most of all, from the sun. They come straight through the earth at nearly the speed of light, all the time, day and night, in enormous numbers. About 100 trillion neutrinos pass through our bodies every second.

The problem for physicists is that neutrinos are impossible to see and difficult to detect. Any instrument designed to do so may feel solid to the touch, but to neutrinos, even stainless steel is mostly empty space, as wide open as a solar system is to a comet. What’s more, neutrinos, unlike most subatomic particles, have no electric charge—they’re neutral, hence the name—so scientists can’t use electric or magnetic forces to capture them. Physicists call them “ghost particles.”

To capture these elusive entities, physicists have conducted some extraordinarily ambitious experiments. So that neutrinos aren’t confused with cosmic rays (subatomic particles from outer space that do not penetrate the earth), detectors are installed deep underground. Enormous ones have been placed in gold and nickel mines, in tunnels beneath mountains, in the ocean and in Antarctic ice. These strangely beautiful devices are monuments to humankind’s resolve to learn about the universe.

It’s unclear what practical applications will come from studying neutrinos. “We don’t know where it’s going to lead,” says Boris Kayser, a theoretical physicist at Fermilab in Batavia, Illinois.

Physicists study neutrinos in part because neutrinos are such odd characters: they seem to break the rules that describe nature at its most fundamental. And if physicists are ever going to fulfill their hopes of developing a coherent theory of reality that explains the basics of nature without exception, they are going to have to account for the behavior of neutrinos.

In addition, neutrinos intrigue scientists because the particles are messengers from the outer reaches of the universe, created by violently exploding galaxies and other mysterious phenomena. “Neutrinos may be able to tell us things that the more humdrum particles can’t,” says Kayser.

Physicists imagined neutrinos long before they ever found any. In 1930, they created the concept to balance an equation that was not adding up. When the nucleus of a radioactive atom disintegrates, the energy of the particles it emits must equal the energy it originally contained. But in fact, scientists observed, the nucleus was losing more energy than detectors were picking up. So to account for that extra energy the physicist Wolfgang Pauli conceived an extra, invisible particle emitted by the nucleus. “I have done something very bad today by proposing a particle that cannot be detected,” Pauli wrote in his journal. “It is something no theorist should ever do.”

Experimentalists began looking for it anyway. At a nuclear weapons laboratory in South Carolina in the mid-1950s, they stationed two large water tanks outside a nuclear reactor that, according to their equations, should have been making ten trillion neutrinos a second. The detector was tiny by today’s standards, but it still managed to spot neutrinos—three an hour. The scientists had established that the proposed neutrino was in fact real; study of the elusive particle accelerated.

A decade later, the field scaled up when another group of physicists installed a detector in the Homestake gold mine, in Lead, South Dakota, 4,850 feet underground. In this experiment the scientists set out to observe neutrinos by monitoring what happens on the rare occasion when a neutrino collides with a chlorine atom and creates radioactive argon, which is readily detectable. At the core of the experiment was a tank filled with 600 tons of a chlorine-rich liquid, perchloroethylene, a fluid used in dry-cleaning. Every few months, the scientists would flush the tank and extract about 15 argon atoms, evidence of 15 neutrinos. The monitoring continued for more than 30 years.

Hoping to detect neutrinos in larger numbers, scientists in Japan led an experiment 3,300 feet underground in a zinc mine. Super-Kamiokande, or Super-K as it is known, began operating in 1996. The detector consists of 50,000 tons of water in a domed tank whose walls are covered with 13,000 light sensors. The sensors detect the occasional blue flash (too faint for our eyes to see) made when a neutrino collides with an atom in the water and creates an electron. And by tracing the exact path the electron traveled in the water, physicists could infer the source, in space, of the colliding neutrino. Most, they found, came from the sun. The measurements were sufficiently sensitive that Super-K could track the sun’s path across the sky and, from nearly a mile below the surface of the earth, watch day turn into night. “It’s really an exciting thing,” says Janet Conrad, a physicist at the Massachusetts Institute of Technology. The particle tracks can be compiled to create “a beautiful image, the picture of the sun in neutrinos.”

But the Homestake and Super-K experiments didn’t detect as many neutrinos as physicists expected. Research at the Sudbury Neutrino Observatory (SNO, pronounced “snow”) determined why. Installed in a 6,800-foot-deep nickel mine in Ontario, SNO contains 1,100 tons of “heavy water,” which has an unusual form of hydrogen that reacts relatively easily with neutrinos. The fluid is in a tank suspended inside a huge acrylic ball that is itself held inside a geodesic superstructure, which absorbs vibrations and on which are hung 9,456 light sensors—the whole thing looking like a 30-foot-tall Christmas tree ornament.

Scientists working at SNO discovered in 2001 that a neutrino can spontaneously switch among three different identities—or as physicists say, it oscillates among three flavors. The discovery had startling implications. For one thing, it showed that previous experiments had detected far fewer neutrinos than predicted because the instruments were tuned to just one neutrino flavor—the kind that creates an electron—and were missing the ones that switched. For another, the finding toppled physicists’ belief that a neutrino, like a photon, has no mass. (Oscillating among flavors is something that only particles with mass are able to do.)

How much mass do neutrinos have? To find out, physicists are building KATRIN—the Karlsruhe Tritium Neutrino Experiment. KATRIN’s business end boasts a 200-ton device called a spectrometer that will measure the mass of atoms before and after they decay radioactively—thereby revealing how much mass the neutrino carries off. Technicians built the spectrometer about 250 miles from Karls­ruhe, Germany, where the experiment will operate; the device was too large for the region’s narrow roads, so it was put on a boat on the Danube River and floated past Vienna, Budapest and Belgrade, into the Black Sea, through the Aegean and the Mediterranean, around Spain, through the English Channel, to Rotterdam and into the Rhine, then south to the river port of Leopoldshafen, Germany. There it was offloaded onto a truck and squeaked through town to its destination, two months and 5,600 miles later. It is scheduled to start collecting data in 2012.

Physicists and astronomers interested in the information that neutrinos from outer space might carry about supernovas or colliding galaxies have set up neutrino “telescopes.” One, called IceCube, is inside an ice field in Antarctica. When completed, in 2011, it will consist of more than 5,000 blue-light sensors (see diagram above). The sensors are aimed not at the sky, as you might expect, but toward the ground, to detect neutrinos from the sun and outer space that are coming through the planet from the north. The earth blocks cosmic rays, but most neutrinos zip through the 8,000-mile-wide planet as if it weren’t there.

A long-distance neutrino experiment is taking place under several Midwestern states. A high-energy accelerator, which generates subatomic particles, shoots beams of neutrinos and related particles as much as six miles deep, beneath northern Illinois, across Wisconsin and into Minnesota. The particles start at Fermilab, as part of an experiment called the Main Injector Neutrino Oscillation Search (MINOS). In less than three-thousandths of a second, they hit a detector in the Soudan iron mine, 450 miles away. The data the scientists have gathered complicates their picture of this infinitesimal world: it now appears that exotic forms of neutrinos, so-called anti-neutrinos, may not follow the same rules of oscillation as other neutrinos.

“What’s cool,” says Conrad, “is that it’s not what we expected.”

When it comes to neutrinos, very little is.

Ann Finkbeiner’s latest book, A Grand and Bold Thing, is about the Sloan Digital Sky Survey, an effort to map the universe.