Nor did this rule apply only to cases where the transports were protected by a mere escort. It held good even in the exceptional cases where the military force was accompanied or guarded by the whole available battle strength of the enemy. We have seen how in 1744 Norris was prepared to follow the French transports if necessary with his whole force, and how in 1798 Nelson organised his fleet in such a way as to contain rather than destroy the enemy's battle-squadron, so that he might provide for an overwhelming attack upon the transports.

Exceptions to this as to all strategical rules may be conceived. Conditions might exist in which, if the enemy's battle-fleet accompanied his transports, it would be worth our while, for ulterior objects of our own, to risk the escape of the transports in order to seize the opportunity of destroying the fleet. But even in such a case the distinction would be little more than academical; for our best chance of securing a decisive tactical advantage against the enemy's fleet would usually be to compel it to conform to our movements by threatening an attack on the transports. It is well known that it is in the embarrassment arising from the presence of transports that lies the special weakness of a fleet in charge of them.

There is, however, one condition which radically differentiates comparatively small expeditions from great invasions and that is the power of evasion. Our experience has proved beyond dispute that the navy alone cannot guarantee defence against such expeditions. It cannot be sure of preventing their sailing or of attacking them in transit, and this is especially

the case where an open sea gives them a free choice of route, as in the case of the French expeditions against Ireland. It is for this reason that, although an adequate navy has always proved sufficient to prevent an invasion, for defence against expeditions it must be supplemented by a home army. To perfect our defence, or, in other words, our power of attack, such an army must be adequate to ensure that all expeditions small enough to evade the fleet shall do no effective harm when they land. If in numbers, training, organisation, and distribution it is adequate for this purpose, an enemy cannot hope to affect the issue of the war except by raising his expeditions to invasion strength, and so finding himself involved in a problem that no one has ever yet solved for an uncommanded sea.

Still, even for expeditions below invasion strength the navy will only regard the army as a second line, and its strategy must provide in the event of evasion for co-operation with that line. By means of a just distribution of its coastal flotilla it will provide for getting contact with the expedition at the earliest moment after its destination is declared. It will press the principle of making the army its objective to the utmost limit by the most powerful and energetic cruiser pursuit, and with wireless and the increased ratio of cruiser speed, such pursuit is far more formidable than it ever was. No expedition nowadays, however successful its evasion, can be guaranteed against naval interruption in the process of landing. Still less can it be guaranteed against naval interference in its rear or flanks while it is securing its front against the home army. It may seek by using large transports to reduce their number and secure higher speed, but while that will raise its chance of evasion, it will prolong the critical period of landing. If it seek by using smaller transports to quicken disembarkation, that will decrease its chances of evasion by lowering its speed and widening the sea area it will occupy in transit. All the modern developments in fact which make for

defence in case of invasion over an uncommanded sea also go to facilitate timely contact with an expedition seeking to operate by evasion. Nor must it be forgotten, since the problem is a combined one, that the corresponding developments ashore tell with little less force in favour of the defending army. Such appear to be the broad principles which govern an enemy's attempts to act with combined expeditions in our own waters, where by hypothesis we are in sufficient naval strength to deny him permanent local command. We may now turn to the larger and more complex question of the conduct of such expeditions where the naval conditions are reversed.

By the conduct, be it remembered, we mean not only their defence but also their support, and for this reason the starting-point of our inquiry is to be found, as above indicated, in the contrast of combined expeditions with convoys. A convoy consists of two elements—a fleet of merchantmen and an escort. But a combined expedition does not consist simply of an army and a squadron. It is an organism at once more complex and more homogeneous. Its constitution is fourfold. There is, firstly, the army; secondly, the transports and landing flotilla—that is, the flotilla of flat-boats and steamboats for towing them, all of which may be carried in the transports or accompany them; thirdly, the "Squadron in charge of transports," as it came to be called, which includes the escort proper and the supporting flotilla of lighter craft for inshore work; and lastly, the "Covering squadron."

Such at least is a combined expedition in logical analysis. But so essentially is it a single organism, that in practice these various elements can seldom be kept sharply distinct. They may be interwoven in the most intricate manner. Indeed to a greater or less extent each will always have to discharge some of the functions of the others. Thus the covering squadron may not only be indistinguishable from the escort and support, but it will often provide the greater part of the landing

flotilla and even a portion of the landing force. Similarly, the escort may also serve as transport, and provide in part not only the supporting force, but also the landing flotilla. The fourfold constitution is therefore in a great measure theoretical. Still its use is not merely that it serves to define the varied functions which the fleet will have to discharge. As we proceed it will be seen to have a practical strategical value.

From a naval point of view it is the covering squadron which calls first for consideration, because of the emphasis with which its necessity marks not only the distinction between the conduct of combined expeditions and the conduct of commercial convoys, but also the fact that such expeditions are actually a combined force, and not merely an army escorted by a fleet.