So long ago as 1886 Mr W. Brewster maintained that the breadth of the fly-line varied according to the character of the country which was being crossed. The migrating column, he said, might be hundreds of miles in length, "a continuous but straggling army," which only became a "solid stream" when travelling through some narrow pass (8). This solid stream or army passage is, however, seldom observed when the birds are crossing continents, especially if they are traversing a wide area in which food is equally plentiful for miles on either side of the direction of flight. The consolidation of their numbers appears only to take place when, either on account of the indifferent food-supply or of unsuitable weather conditions, the speed is accelerated. In America Mr Cooke proves that the Mississippi Valley is undoubtedly utilised as a fly-line by a large number of species, but by no means all, and his evidence, though proving the use of routes, is that these are seldom constricted pathways but broad areas crossed in a generally coincident direction by the birds which make use of them. This main fly-line is however formed in America as in other places by the convergence of subsidiary streams, and it is these tributaries, as Herr Herman points out, which have in many instances led to error; they have been mistaken for main routes. The main route may be compared to the trunk of a tree, the birds following the roots from the area in which they have been nesting or wintering, and at the end of the journey splitting off in various directions, like the branches, to their temporary winter or summer homes. The contrast in the method of travelling of different species or of the same species under different conditions, may be realised by taking two examples. Firstly, Mr Eagle Clarke's experiences at the Eddystone and Kentish Knock Lightship, when birds passed during the daytime at varying elevations, sometimes close to the waves, in twos or threes or scores, and at night in large numbers. The other is an observation of a "bird wave" by Mr P. Cox, during a snow storm in 1885 at Newcastle, New Brunswick. The birds passed eastward in a column about twenty-five yards wide, some just above the trees, others hardly visible, but the bulk in a massed column directly over the margin of the shore, and not over the river or meadow on either side. The movement was continuous for about two hours. Dr I. A. Palmén was the great upholder of routes in the Old World, but his routes were largely speculative; they were founded on a considerable knowledge of migratory birds, but not sufficient to cover the vast area mapped out (39). Until a very large band of workers, working on similar lines all the world over, accumulate a sufficient mass of evidence as to which birds do or do not pass their various stations, with the times at which they appear, accurate knowledge of the routes of birds is impossible.