Geologists have never agreed on what the Pinnacles are. Not on the rock, which is plain dune limestone, lime ground out of seashells on the Western Australian coast and blown inland as sand. The argument is about the shape: why any of it stands up in columns at all. Thousands of them do, in bright yellow sand two hours north of Perth, most of them between your knee and your head. One reading has rainwater cutting vertical pipes down through the limestone until the pipes widened and merged, leaving the rock between them standing. Another says some of the pillars are not what was left over but the hardened infill of those same pipes. A third camp reads them as calcrete casts that formed around the roots of trees long gone. From the raised lookout on the drive you get the only view that takes the field in whole, pillars running away in every direction until they hit a wall of green heath. The line between the two is startlingly hard: bright sand one step, dense scrub the next.
The rest of it has to be walked. A few rows in, the fine sand gives beneath every step and the taller columns cut your line of sight back to the track. Close up, no two are alike: blades, blunt stumps, pillars leaning at odds with their neighbours. In the last hour of light every one of them lays down a shadow that keeps stretching. The park has no gate and no lighting, so the desert stays open long after the visitor centre shuts, and once your eyes adjust the galactic core comes up behind the columns as black silhouettes. None of it was on show for long: drifting sand buried the whole field and only gave it back a few centuries ago. One shape here does have a settled answer. The pillars wearing a flat dark cap on a narrowing stalk are capped with calcrete, a crust that hardened at an old land surface. It shields the softer limestone under it, so the sides wear back faster than the top.