Page:BirdWatchingSelous.djvu/331

report of a gun, the whole troop burst suddenly out of the trees, which were on the outer edge of the plantation, flew a little way over the heath (I caught them against the fading red of the sky), wheeled round, returned, and shot into them again. There was a little cawing as they got back, but this soon sank, and again there was silence. Then, in a moment, there was the same sudden rush of wings, and the whole black cloud shot, like one bird, into the open sky, wheeled again, and shot back, as before. This occurred nine times in succession, at intervals of not longer, I should think, than three or four minutes. In the later rushes the birds circled several times—flying out again, each time, over the moor—before resettling in the trees. After the last time they settled in a different part of the plantation. Immediately before two of the rushes out, I heard a loud 'caw,' in rather a high-pitched tone, from a single rook, which seemed to be the signal for the exodus, whilst, almost immediately afterwards, there was another single note of quite a different character—deeper and more guttural—from either the same or another bird still in the trees, which seemed to call the rest back again. A well understood signal-note indeed, would be the easiest way of accounting for these sudden and extraordinarily simultaneous flights and returns, but it was only twice out of the nine times, that this explanation seemed tenable. On other occasions, the caw, at starting, seemed only one of many, or did not correspond so exactly, in point of time, with the sudden flight out, as the theory seemed to require, whilst the deep 'quaw,' which seemed to be made by one particular rook, who always stayed behind, and which I had at first thought called the