WOLF-RAYET 104The pinwheel
and the signal

What the array is looking at

WR 104 is a binary star roughly eight thousand four hundred light years away in Sagittarius, catalogued as V5097 Sagittarii. The primary is a Wolf-Rayet star of type WC9: a massive star late in its life that has blown away its outer hydrogen and is now shedding several Earth-masses of itself every year in a wind moving at over a thousand kilometres a second. Its companion is a hot main-sequence star of type B0.5, orbiting about two astronomical units away — closer than Earth to the asteroid belt.

Where the two winds meet they shock, compress, and cool, and in that narrow collision front carbon condenses into soot. The pair keep turning, so the soot is flung outward in a curve rather than a straight line. One orbit lays down one turn of the spiral. The result is a plume of glowing dust more than two hundred astronomical units across, discovered in 1999 by the Keck Aperture Masking Experiment and immediately nicknamed the pinwheel.

A clock you can read from eight thousand light years away

The orbit takes 241.5 days. Because the dust is manufactured continuously and thrown outward at a steady speed, the spacing between the arms is a direct record of that period — the pinwheel is a mechanical clock, and the spacing of its arms tells you how fast it is running. Track the pattern for eight months and it returns to where it started.

This is also why the system was so useful to astronomers long before anyone worried about what it might do. Very few objects in the sky hand you their orbital period as a picture.

The Death Star problem

When a Wolf-Rayet star reaches the end, it can collapse into a type Ic supernova. If it is still spinning fast enough, that collapse can drive a gamma-ray burst — an emission so concentrated that it emerges as a beam along the star's rotation axis rather than spreading in all directions.

The trouble with WR 104 was geometry. The pinwheel appeared almost perfectly face-on, and in a system like this the dust spiral traces the orbital plane. Face-on spiral implies the orbital axis — and plausibly the stars' spin axes — points more or less at us. The press did the rest, and for twenty-five years WR 104 was the Death Star.

Peter Tuthill, who discovered the pinwheel, spent much of that time publicly picking his own scenario apart: whether the star will spin fast enough by the time it goes, whether high metallicity in our galaxy brakes Wolf-Rayet stars below the threshold, and how tightly a burst would actually be collimated. His own conclusion was that the chain of requirements is long and most of the links are uncertain.

March 2025: the angle moves

Grant Hill measured the system spectroscopically at Keck, using three instruments to track the stars' velocities and solve the orbit directly rather than inferring it from the dust. The orbital plane turned out to be tilted at least thirty to forty degrees out of the plane of the sky, possibly as much as forty-five.

That is not face-on. Whatever WR 104 eventually does, Earth is not lined up with it in the way the story assumed.

The part nobody can explain

A tilted orbit should produce a tilted spiral. Seen from an angle, a pinwheel should look like an ellipse.

It doesn't. The dust plume still appears face-on. The orbit and the thing the orbit is supposed to be drawing are pointing in different directions, and the physics that would reconcile them has not been worked out. Hill's own suggestion is that additional mechanisms must be shaping the plume.

It is, at the time of writing, an open question about a real object: the geometry of WR 104 does not add up.