The sea qualities which the subchaser displayed, and the development of listening devices which made it possible to detect all kinds of sounds under water at a considerable distance, immediately laid before us the possibility of direct offensive operations against the submarine. It became apparent that these listening devices could be used to the greatest advantage on these little craft. The tactics which were soon developed for their use made it necessary that we should have a large number of vessels; nearly all the destroyers were then engaged in convoy duty, and we could not entertain the idea of detailing many of them for this more or less experimental work. Happily the subchasers started coming off the ways just in time to fill the need; and the several Allied navies began competing for these new craft in lively fashion. France demanded them in large numbers to work in co-operation with the air stations and also to patrol her coastal waters, and there were many requests from stations in England, Ireland, Gibraltar, Portugal, and Italy. The question of where we should place them was therefore referred to the Allied Naval War Council, which, at my suggestion, considered the matter, not from the standpoint of the individual nation, but from the standpoint of the Allied cause as a whole. A general survey clearly showed that there were three places where the subchasers might render the most efficient service. The convoy system had by this time not only greatly reduced the losses, but it was changing the policy of the submarines. Until this system was adopted, sinkings on a great scale were taking place far out at sea, sometimes three or four hundred miles west of Ireland. The submarines had adopted the policy of meeting the unescorted ships in the Atlantic and of torpedoing them long before they could reach the zones where the destroyer patrol might possibly have protected them. But sailing great groups of merchantmen in convoys, surrounded by destroyers, made this an unprofitable adventure, and the submarines therefore had to change their programme. The important point is that the convoys, so long as they could keep formation, and so long as protecting screens could be maintained on their flanks, were virtually safe. Under these conditions sinkings, as already said, were less than one-half of 1 per cent. These convoys, it will be recalled, came home by way of two "trunk lines," a southern one extending through the English Channel and a northern one through the so-called "North Channel"--the latter being the passage between Ireland and Scotland. As soon as the inward-bound southern "trunk-line" convoys reached the English Channel they broke up, certain ships going to Plymouth, Portsmouth, Southampton, and other Channel ports, and others sailing to Brest, Cherbourg, Havre, and other harbours in France. In the same fashion, convoys which came in by way of the North Channel split up as soon as they reached the Irish Sea. In other