"The Plains Zebra (Equus quagga) is, in the author's initial assessment, a compelling example of disruptive camouflage — the striped pattern breaking up the animal's outline against the dappled light and movement of the savannah. Each individual's pattern is unique, which the author finds both remarkable and, in a herd of several hundred, impractical to document. The quagga (Equus quagga quagga), a Southern African subspecies with stripes confined to the front half of the body, was hunted to extinction. The last wild individual died circa 1883. The author notes this without additional comment, having found that additional comment does not improve it."
The flies were the first thing.
The Colonel had been watching the herd for forty minutes, noting the stripe patterns, writing his camouflage theory with the quiet confidence of a man repeating something held by every naturalist he had read, when he noticed that the flies circling the herd's periphery were not landing on the animals. They moved around them, through the air immediately adjacent, but did not settle on the striped coats. He noted this as: "Anomalous insect behaviour around herd. Flies present but not alighting. Investigating."
He investigated.
The research note was from a colleague's letter received at the last camp — a summary of recent experimental findings that he had read once and filed. He found it in the second bag. He read the relevant passage again.
Flies, the research established, were strongly deterred by striped surfaces. Horses in striped blankets received measurably fewer biting insects than horses in plain coverings beside them. Multiple species of biting fly avoided striped patterns. The deterrence was real, specific, and consistent across experimental conditions.
The Colonel read this.
He looked at the herd.
He looked at his field note, where he had written "compelling example of disruptive camouflage" three times in slightly different phrasing.
He crossed it out.
"The stripes," he wrote, "are for flies. The camouflage theory, which the author has held for twenty years and which is held by most naturalists the author has read, is incorrect. The author notes that being wrong in the most universally held way has a certain dignity to it, though he is not entirely convinced of this."
He looked at this sentence. He left it in.
The sneeze came from somewhere in the middle of the herd, which was the worst possible location for a sneeze to originate.
The Colonel had been watching a dominant male conduct his full display — the head-tossing, the tail-flagging, the sideways movement that he wrote as "prancing with intent" — toward a mare who was demonstrably not watching. Then one animal sneezed. Not dramatically. A standard sneeze. The animal nearest it startled. That animal's movement startled its neighbour.
What followed was not a stampede in the fullest sense — no predator, no sustained flight — but a surge: three hundred animals moving simultaneously, the stripe patterns blurring as the herd compacted and turned, each individual's unique marking dissolving into a collective visual noise that the Colonel wrote as "precisely the effect that makes individual selection by a predator theoretical."
It resolved in thirty seconds. The herd stopped. The male resumed his display. The mare continued to ignore him.