As emphatically he maintains that most observable migration over Heligoland is due east to west or west to east, though the birdstuffer Aeuckens, who supplied him with much of his information, told Seebohm that it was north-east in spring and south-west in autumn (45). Is it not perfectly evident that the geographical position of Heligoland makes it a convenient resting place for large numbers of migrants, for it is certainly true that large numbers are observed there, which pass southward and westward along the Baltic, crossing Schleswig-Holstein and the mouth of the Elbe, or coast south along Denmark, and cross the Elbe diagonally, _en route_ for the Dutch and French coasts and to a lesser extent the south-east coasts of Britain? Coasting undoubtedly exists; birds, day migrants especially, may be observed following coast-lines in steady flight, though a mile or less inland no passage is visible. On the North Norfolk coast I have seen little parties of swallows passing along the shore in spring, coasting slowly but steadily from east to west. All day long and almost every day for more than a week this steady flight was continued, though I never saw any passing within sight more than a few yards out at sea, nor any at all more than a few hundred yards inland. Evidence which cannot be refuted shows that this habit of coasting is general, though a deeply indented bay, an estuary or strait, is usually crossed, and by no means always at the narrowest point. The same careful observations prove that both narrow and wide river valleys are followed by migrating birds in greater numbers than are ever observed passing beyond the limits of these valleys. Seebohm's experience in Siberia led him to doubt the existence of routes, but his later studies of migration in autumn at Arcachon and in spring at Biarritz, caused him to modify his ideas. He found a gentle but continuous stream of migrants following the coast of the Bay of Biscay, arriving from over the Pyrenees on their northward journey, but moving "only within a mile or two of the coast." He contrasts island and coastwise migration; in the latter the travellers can rest at night or take short journeys during bad weather, but in the former they must await favourable conditions before attempting a perilous passage (45). On the other hand many birds undoubtedly pass over inland localities independently of any river valley or mountain range which might indicate a route. Even such typical coast-lovers as the maritime waders constantly cross or pass through inland England. They are heard at night, or met with resting or feeding on inland waters, or their bodies are found when, on a dark night, they have collided with telegraph or telephone wires.