Another conserved quantity is electric charge. Protons carry a unit of positive charge, electrons carry the same amount of negative charge, and neutrons carry no charge at all. Charge conservation requires that at no time during an experiment is the net charge anything other than what you started with. And that's as true for particle accelerators on Earth as for supernova explosions in distant galaxies.

Armed with the conservation laws of momentum, mass-energy, and electric charge, you're more or less where Pauli was in 1930. Back then, life was simpler. Particle physicists were not yet talking about quarks, muons, gluons, or Higgs bosons. What they did discuss was a subatomic process called beta decay, in which a proton and an electron spontaneously fly apart, accompanied by unbalanced momentum and a loss of mass-energy. Had the conservation laws lost their grip on nature? Or could the existence of an unforeseen and undiscovered particle resolve the conundrum?

Discoveries in physics often emerge from one's confidence in competing ideas. Rather than dismantle the foundations of physics, Pauli postulated that the escaping proton and electron (both later shown to have come from a decayed neutron) were not the sole products of the decay. His additional particle was to have no charge, some momentum, and vanishingly small, possibly zero, mass-energy.

Turns out, the key to beta decay was not the neutrino but its antimatter counterpart, the antineutrino. A decaying neutron yields a proton, an electron, and an antineutrino. Under the dictates of additional conservation laws, unknown to Pauli and his contemporaries, that's just what you'd expect. Two of those laws decree that no process can change the net numbers of heavy particles (baryons) and light particles (leptons). If your experiment starts with one baryon (a proton or a neutron), it must end with one baryon. That means a neutron can morph into a proton. And if it starts with zero leptons (an electron or a neutrino), it must end with zero leptons.

Wait a minute. Beta decay starts with zero leptons but ends with two leptons: an electron and an antineutrino.

Not to worry. The antineutrino is not simply a light particle; it's an anti--light particle. So in the particle count, an electron and an antineutrino cancel, resulting in zero net leptons. The laws of conservation triumph yet again.

Pauli died in 1958. Fortunately for him, he lived just long enough to see Cowan and Reines detect his "undetectable" particle. Today neutrinos remain among the most challenging subatomic particles to catch, even though everybody and everything is steeped in trillions of them. Problem is, they interact so rarely with other kinds of matter that you need enormous, clever traps to boost your chances of detecting any at all.

Nearly all the evidence for neutrinos from space comes from detectors buried deep underground, which hold enormous quantities of odd liquids surrounded by unusual hardware. These underground "telescopes" can catch intrepid neutrinos from any direction, even those that have passed all the way through Earth from below. In one detector the neutrinos enter a tank filled with 600 tons of chlorine-laden dry-cleaning fluid. Every so often, in a kind of reverse beta decay, one of the passing neutrinos changes a resident neutron within a chlorine atom into a proton, thereby changing the chlorine to radioactive argon. The presence of an argon atom serves as a tracer of the neutrino's visit. Other creative designs track the flash of blue light emitted by the particle products of neutrino interactions. Those tanks are filled with ultrapure water or a mixture of baby oil and benzene.

My favorite setup, though, is a not-yet-finished neutrino observatory called IceCube. Its "tank" is a cubic kilometer of clear, dense Antarctic ice, in which the investigators will suspend a lattice of sensors, lowered through deep holes melted by a hot-water drill.

Unfortunately, Pauli didn't live long enough to see how populous the particle zoo would become--how many categories and subcategories and families and flavors particle physicists would postulate and discover in the decades that followed his death. Nor could he have imagined that neutrinos themselves would land in the middle of one of the greatest astrophysical conundrums of the twentieth century.

In March 1964, in the journal Physical Review Letters, the late American astrophysicist John N. Bahcall published his calculations showing that vast quantities of neutrinos should continually flee the Sun as the nuclear furnace in its core transforms hydrogen into helium. In a tandem paper, Bahcall's friend and colleague Raymond Davis Jr. described an experiment he was building in the disused Homestake Gold Mine in Lead, South Dakota. In search of evidence for solar neutrinos, he would place a tank of chlorine-rich liquid deep belowground. As usual, the encounters between neutrinos and atoms would be exceedingly rare; Bahcall calculated that the experiment should record about ten neutrinos a week. But even those few neutrinos would reveal what was going on in the Sun's center, thus eliminating the need ever to visit the place.

For years, however, only about three of the ten neutrinos showed up. That gap became the tenacious "solar-neutrino problem." Some physicists copped an attitude, suggesting that astronomers didn't fully understand how the Sun manufactures energy—knowledge that underpins much of modern astrophysics. Shaken but not stirred, Bahcall was so sure the Sun was not misbehaving that he committed much of his career to demonstrating why. Meanwhile, Davis continued to refine his measurements. And the solar-neutrino problem endured.

Normally, physicists hand the laws of physics to astrophysicists, and it's those laws that guide questions and constrain answers. But every now and then, astro folks teach the physics folks a thing or two about how the universe works. Indeed, Bahcall was right all along. The missing neutrinos were there. They just weren't the kind of neutrinos that would turn chlorine into argon. Apparently, without telling anybody, they had left the Sun with one identity—the one the experiment was designed to detect—but reached Earth in a different guise, requiring a different experiment to be detected at all.