There remains a form of support which has not yet been considered, and that is diversionary movements or feints by the fleet to draw the enemy's attention away from the landing place. This will naturally be a function of the covering battle-squadron or its attendant cruisers and flotilla. The device appears in Drake's attack on San Domingo in 1585, an attack which may be regarded as our earliest precedent in modern times and as the pattern to which all subsequent operations of the kind conformed so far as circumstances allowed. In that case, while Drake landed the troops a night's march from the place, the bulk of the fleet moved before it, kept it in alarm all night, and at dawn made a demonstration with the boats of forcing a direct landing under cover of its guns. The result was the garrison moved out to meet the threat and were surprised in flank by the real landing force. Passing from this simple case to the most elaborate in our annals, we find Saunders doing the same thing at Quebec. In preparation for Wolfe's night landing he made a show of arrangements for a bombardment of Montcalm's lines below the city, and in the morning with the boats of the fleet began a demonstration of landing his marines. By this device he held Montcalm away from Wolfe's landing place till a secure footing had been obtained. Similar demonstrations had been made above the city, and the combined result was that Wolfe was able to penetrate the centre of the French position unopposed. Such work belongs of course to the region of tactics rather than of strategy, but the device has been used with equal effect strategically. So great is the secrecy as well as the mobility of an amphibious force, that it is extremely difficult for an enemy to distinguish a real attack from a feint. Even at the last moment, when a landing is actually in progress, it is impossible for the defenders to tell that all the troops are being landed at the one point if a demonstration is going on elsewhere. At Quebec it was not till Montcalm was face to face with Wolfe that he knew he had to deal with the whole British force. Still less from a strategical point of view can we be certain whether a particular landing represents an advance guard or is a diversionary operation to mask a larger landing elsewhere. This is a special difficulty when in the case of large operations the landing army arrives in echelon like the Second Japanese army. In that instance the naval feint was used strategically, and apparently with conspicuous effect. The Russians were always apprehensive that the Japanese would strike for Newchuang at the head of the Gulf of Pe-chi-li, and for this reason General Stakelberg, who had command of the troops in the peninsula, was not permitted to concentrate for effective action in its southern part, where the Japanese had fixed their landing place. Admiral Togo, in spite of the strain on his fleet in effecting and securing the