Physicists sometimes say that a beam of light traveling through space is like a “great smoky dragon.” One can know much about where the light comes from (the dragon’s tail) and where it is seen (the dragon’s head), yet still know precious little about the journey in between (the dragon’s mysterious, nebulous body). As light travels from source to detection, it can behave as either a particle or a wave—or, paradoxically, both states or neither state. Now an experiment using laser beams shot at satellites in low-Earth orbit has confirmed that this bizarre detail about the nature of light holds true across record-breaking distances.

Quantum physics, the best description yet of how all known particles behaves, suggests that reality is fuzzy and uncertain at its most basic levels. For instance, the surreal quantum effect known as superposition essentially allows electrons, atoms and other building blocks of the universe to each exist in two or more places simultaneously.

Another strange quantum phenomenon is particle-wave duality. Whereas Isaac Newton thought light was made up of particles, his contemporary the Dutch scientist Christiaan Huygens argued that it consisted of waves. Eventually, researchers performing the so-called double-slit experiment demonstrated that Newton and Huygens were both right—photons of light could behave as both particles and waves.

The double-slit experiment involves shining a single light source through two adjacent slits in an opaque plate, and onto a detecting screen. If an experimenter closes one slit, the light passing through the other forms a bar on the screen—as if the light were behaving like a stream of particles. But if the experimenter leaves both slits open the light will not form two such bars. Instead it generates a series of bright and dark bands on the screen, as if waves of light scattered through the slits and interfered with each other. The bright bands indicate where light waves reinforced each other, and dark bands where they canceled each other out. Remarkably, this interference pattern will materialize even if photons are projected at the slits one at a time.

The way quantum physics explains these confounding results is that the instruments used to detect light determine its state as a particle or a wave. Describing this situation, in 1983 the American physicist Jonathan Wheeler coined the now-famous “great smoky dragon” comparison.

To examine how light “chooses” to become either a particle or wave upon its detection, Wheeler conceived of the delayed-choice experiment. In it, an optical device called a beam-splitter offers a single photon two paths to take. At the end of each path is a detector. If the photon behaves like a particle, it has an equal chance of taking either path and being seen by either detector. If the photon instead behaves like a wave, it will take both paths simultaneously and register in both detectors.

When the experiment only incorporates one beam-splitter, a single photon will take either path and just one detector will see it. This suggests the photon made the “choice” at the beam-splitter to behave like a particle. However, a beam-splitter can also act in reverse to merge two photons into one; if the experiment has the paths converge at a second beam-splitter before channeling them to a detector the result will be the interference pattern from the photons acting as waves and reacting to each other. This holds true even when the second beam-splitter is introduced in the split second after light passes through the first one—but has yet to reach the detectors.

Scientists have successfully carried out both versions of this experiment in the decades since Wheeler proposed it. Its results make sense if the photon “delays” making the choice to become a particle or a wave until it actually gets detected. The alternative would suggest that the photon could somehow decide to become a particle if it encountered one beam-splitter but then change its decision and become a wave if it ran across a second beam-splitter.

Historically, all delayed-choice experiments have been performed on Earth. But now scientists are increasingly conducting quantum experiments involving lasers shot across the vastness of outer space. Quantum physicist Paolo Villoresi at the University of Padua in Italy and his colleagues wanted to verify if the dual nature of light still held true even across the distances between the ground and satellites in low-Earth orbit. “As Galileo—who did most of his work at the University of Padua—said, ‘we have to prove the laws we know in new contexts,’” Villoresi quips.

Using the Matera Laser Ranging Observatory in Italy, Villoresi and his colleagues performed Wheeler’s delayed-choice experiment by firing green laser pulses at the Beacon-C and Starlette satellites, which reflected the photons back at the observatory. At their farthest, the satellites were 1,771 kilometers (1,100 miles) away from the observatory.

“Distance matters,” explains astrophysicist Brian Koberlein at the Rochester Institute of Technology in New York, who did not take part in this research. “In a single lab, you could argue that maybe in some way the experimenters are affecting the outcome. But over larger distances, there isn’t a clear way to affect outcomes.”

Instead of having the photons travel down one of two separate paths of equal lengths, the scientists measured two different aspects of each photon—how each one oscillated in space, and whether it took a shorter or longer path to the detectors. Their results confirmed light’s curious quantum behavior over distances tens to hundreds of times greater than previously shown, Villoresi says. The team’s findings appeared Oct. 25 in the journal Science Advances.

“This work further confirms that quantum mechanics really is the description of the ‘great smoky dragon,’” Koberlein says. “It may be strange, but it is logically and mathematically consistent.”

Aside from testing the quantum qualities of light across unprecedented distances, Villoresi notes that quantum physics experiments conducted across space could help lead to satellite-based telecommunications networks protected by nigh-unhackable quantum cryptography. By clarifying the fundamental properties of photons during such experiments as was done in this study, “there may be direct applications for larger bandwidths in quantum communications,” Villoresi says. Indeed, great smoky dragons may someday carry secrets in their jaws.