The end is brutal for electrons hurtling at 99.9999999% of the speed of light through SLAC National Accelerator Laboratory’s two-mile-long beam pipe: a final slam into End Station A. In the late 1960s and early ’70s, such collisions broke apart protons and neutrons to reveal the elementary particles that make them up. The discovery won the experiment’s leader a Nobel Prize. “End Station A is this hallowed ground at SLAC,” said the physicist Timothy Nelson. Walking to the back of the warehouse, past piles of equipment, Nelson pointed at the skeleton of an old experiment, beyond the point where the tree-trunk-size pipe of the historic accelerator cut off.

It’s there, he said, that a soon-to-be-constructed experiment will see — or quickly rule out — one of the most popular new candidates for dark matter. Almost a century ago, the Swiss astrophysicist Fritz Zwicky described a galaxy cluster that appeared to rotate too fast to be held together by its visible mass. He proposed that invisible matter was lending its gravity to the situation. The evidence has grown and grown, and researchers now believe that 85% of the universe’s matter is hidden. But the mystery of dark matter’s identity has endured. For decades, researchers focused on two candidate particle types: weakly interacting massive particles (WIMPs) and axions.

These are the simplest formulations for dark matter, and each type of particle would also elegantly solve other physics mysteries. But after roughly 40 years of vain searches for these particles — searches that have almost entirely ruled out the chance that dark matter is made of ordinary WIMPs — physicists have become far more open-minded about what dark matter might be. Maybe dark matter is not simple at all.

It could, some suggest, comprise a whole family of particles, as does visible matter. “The most common hypothesis is that this is somehow simple.

Why on earth should we expect that?”

said Philip Schuster, a theoretical physicist at Stanford University. “That is not what nature has been trying to tell us for the last 200 years.” Move away from the WIMP or axion paradigms, and you probably have to give up on the idea that dark matter consists of a single type of particle.

Instead, new models involve a bevy of extraordinarily lightweight entities, sometimes called feebly interacting particles. Today’s dark matter hunters are especially excited by two categories of these particles: “light dark matter” and “ultralight dark matter.” Light dark matter can be thought of as lighter variations on WIMPs, while ultralight dark matter is made of even lighter axions.