But it was shown, long since, by direct experiment, that the area of a bird's wing is considerably in excess of what is required for the purpose of flight. Dr. J. Bell Pettigrew, more than fifty years ago, to test this matter, cut off more than half of the secondary wing feathers of a sparrow, parallel with the long axis of the wing. He first clipped one, then both wings, and found that in both cases flight was apparently unimpaired. He then removed a fourth of the primary feathers--the outermost quills--and still the flight was unimpaired. At any rate the bird flew upwards of thirty yards, rose to a considerable height and alighted in a tree. Thirty yards, however, is a short flight even for a sparrow. But it is enough to show that flight, if not _sustained_ flight, was possible after this mutilation. Not until more than one-third of the quills along the whole length of the wing were removed, did the flight become obviously laboured. And he found that what was true of the sparrow, was equally true of the wings of insects. Though these experiments demonstrate, in a very unmistakable manner, that flight with a greatly reduced wing area is possible, we have no evidence that this reduction would make no difference to the length of time the bird could remain on the wing. And this is a very important matter. An aspect of flight which has now to be considered is that of birds which fly in troops. Some species always travel thus, others only on occasions. Rooks and gulls afford instances of this, when, during windy weather, or for other reasons, they congregate and fly round and round in great circles, at a considerable height. Small wading-birds, like ringed plovers and dunlin, commonly fly in "bunches." The last named furnish a singularly interesting sight when thus travelling; for their evolutions are so amazingly timed. As if at a given signal every bird in the troop will change its course at the same moment, and in the same direction, so that now one sees a flickering mesh-work of grey, and now a shimmering as of snow-flakes, as first the grey backs, and then the white breasts are turned towards one. But flights such as this are to be seen only during the autumn and winter months. For during the breeding season these little flocks are broken up and distributed far and wide. But there is yet another reason. They wear a totally different dress--the courtship or breeding plumage. Herein the upper parts are of a rich chestnut hue, streaked with black, while the under parts are black. Even more fascinating to watch are the autumn troops of starlings on the way to their roosting places. Hundreds at a time, not to say thousands, take part in these flights. Now they rush onward, in one great far-flung sheet, and now they close up into a great, almost ball-like, mass: and now they thin out till they look like a trail of smoke. But always they wheel and turn and rise and descend, not as separate bodies, but as one. How are such wonderful evolutions