motion is not confined to its general shape, but is specially noticeable in its feet and legs. It is not strictly web-footed like a duck or gull; the toes are distinct but are broadened out or lobed, whilst, in the same way, the tarsus or leg is flattened and broad. The grebe does not paddle through the water with alternate strokes, but, at any rate when travelling quickly, rows itself along, the legs striking out sideways and not beneath the plane of the body. Set far back, the legs are further furnished with special rotatory muscles; in the forward swing the bird "feathers under water," turning the tarsus and lobed foot so as to present as little resistance as possible to the water, cutting it with the narrow cutting edge In the back stroke the lobed toes are turned, spread so as to give a bigger surface; foot and tarsus grip the water. So freely do bone and muscle work that the bird can easily raise a foot above its back; often, when a grebe is lazily swimming on the surface, head resting on the back, it will raise and shake a foot above its back.
In one big and generally correct work on British birds, published not very long ago, it is definitely stated that the grebes leave fresh water for the sea in November and return in February. This, I believe, is the case in Norfolk, and even in Cheshire there is a certain amount of movement, especially amongst birds of the year, but the meres are only deserted entirely when they are frozen. There are always a few birds about in December and January if the water is open, and often there are many together, for the grebe is gregarious in winter. Some of the Cheshire birds go to the coast in late autumn, and there is always a noticeable increase of returning birds in flocks.
In the same book it is stated that "the wing beat is not unlike that of a duck, and they strike the water on alighting in similar fashion." This is exactly what they do not do. Anyone who has watched ducks alighting,
