Ceraunocandela furcataThunder Candelabra

Only on mornings when water can still be heard in the gully does the cup-shaped flower rise a few centimeters above the mud. Its anthers appear to touch the stigma during the same opening, so most flowers likely set seed by self-pollination. Before the mud had fully dried, however, I saw signs that a small hoverfly had carried pollen once between plants. The flower is not the main means by which the colony moves; it is better understood as a small insurance against sameness where erosion continues.

When I lifted plants exposed by the current, the ends of the main roots were not mere torn fragments but pale, thickened bodies. Inside the thin cortex were minute points resembling dormant buds, and the bodies rounded readily in muddy water. Because they lodge together at bends in rills and behind small stones, a week later the young rosettes appear in a line of nearly equal height along the fresh failure edge.

On dry days, this plant almost refuses to become a specimen. The leaves are less grey-green than mud-coloured, half buried, and were probably dismissed in notebooks as seedlings of small Asteraceae or Brassicaceae already known from the badlands. Flowering ends within two or three days after rain, while the slopes most visible from tourist paths are also the ones compacted and drained into relative stability. On ground that no longer moves, the evidence for this plant disappears with it.