1914 Western European Front Campaigns

"1914 Western European Front Campaigns -8. - The story of the successive campaigns in Belgium and France during the World War, embracing the continuous struggle on the western front from Aug. 2 1914 to Nov. II 1918, is dealt with below under four main sections, representing the phases into which it naturally divides itself: - viz. the " open " warfare campaign of 1914, which ended without decisive victory to either side in the open field and left the armies " stabilized " on a continuous line from Nieuport to the Swiss frontier; the three years of trench-warfare campaigns, 1915-7; the great German offensives of March-July 1918, which, breaking the stability of the trench-warfare system, re-introduced a condition of semi-open warfare; and finally, the allied offensive which synchronized at its outset with the last German attack effort, and closed with the Armistice at II A.M. on Nov. i i 1918.

I. THE Campaign Of Aug.-Nov. 1914 France's Defence Problem. - During the years which followed the Franco-Prussian War of 1870-I, a guerre de revanche for the reconquest of Alsace and Lorraine was very frequently discussed in France. But it implied aggressive action against Germany, and those who judged with reason and not with sentiment knew very well that such action was impossible. In fact, Germany, at every international crisis that arose, asserted, sharply and menacingly, her readiness to accept a challenge, while France prudently yielded and avoided a conflict.

It may be observed that, under similar military systems - that is, under laws of universal military service - the effort of France could not surpass the effort of Germany, for France counted less than 40 million inhabitants against the 65 to 70 millions of Germany. Presuming equality of national effort, the vis viva of France in relation to that of Germany would be in the proportion of 4 to 7; and this calculation makes no allowance for the fact that in Germany all factors combined to exalt the army, to intensify war preparation, and to produce solid cadres and reserves, while in France the tendency of politics was to depreciate the army and to lower its quality, to minimize its training periods, and to depress its military spirit. It is not unfair to say therefore that the possible warlike power of Germany was at least double the possible warlike power of France.

If, next, we cast a glance at the theatre of the possible war, we see at once that France had no natural frontier with respect to Germany, and was obliged to make good this defect by a system of fortresses and entrenched camps - a form of defence which it is exceedingly difficult to maintain at such a level as to be capable of resisting at any moment an artillery that itself is constantly evolving in the direction of increased power. On the side of Germany, on the other hand (even leaving out of consideration her first-class fortresses, for which money was never lacking), there was a line of defence of the very first order, the Rhine - impossible to turn even if the neutrality both of Holland and of Switzerland were violated, for its flanks rest on the Alps and the sea. There are not in Europe two lines of defence of this value, and it was reinforced by a chaplet of entrenched camps. Supposing then that, in spite of the conditions mentioned, France somehow contrived an initial superiority, her penetrative effort could in no case pass the Rhine, while, in a German penetration of France, Paris was within measurable reach.

From the point of view of numbers, the French alliance with Russia might seem at first sight not only to redress the balance but even to weigh down the scales heavily in France's favour. The effective utilization of these numbers was, however, subjected in practice to grave limitations. The strategic conditions of what came to be called the eastern front are discussed in the article Eastern European Front Campaigns, and here it 1S only necessary to say that these conditions and in particular the paucity of railway communications and of rolling-stock in Russia - evident from a glance at any map of central and eastern Europe - left it within Germany's power to use by far the greater part of her forces in an initial campaign against France. These forces might, from the relation of the two populations concerned and the characters of their respective politics, attain a figure almost double those of France. In such a contingency, the French armies almost might be crushed under a very considerable numerical superiority; the Germans could sweep up to Paris; and there probably the war would end. Germany would rapidly bring back her armies to deal with Russia, aided both by the general E.-W. orientation of her railways and by their perfect technical preparedness. That the German plans did not always take this form, that the soundness of its principle was a matter of considerable controversy, within the German General Staff as well as in military publications, and that the numerical German superiority was not in fact attained, may be freely admitted. But, as the most dangerous alternative that France had to consider, this plan was found' to be taken on the logical basis upon which the military policy of the defence should be build up. Whether Germany's own allies would cooperate in such an invasion, and if so, to what extent, was doubtful. Austria's main effort would have to be made in the East; and as regards Italy, it was known that her obligation under the Triple Alliance would become operative only if one of her Allies was defending itself against attack. On the other hand, the Entente Cordiale between England and France had enabled the latter to concentrate her naval effort in the Mediterranean, and the details of possible cooperation on land, for some years before 1914, had been studied by the British and French general staffs in concert. But England reserved to herself complete freedom to decide for or against intervention if and when the case arose. It was not till Aug. 2 1914 that Sir Edward Grey engaged that the British navy would protect the Channel coast of France, and not till the actual violation of Belgian neutrality by Germany that the British Government declared war on their own account.

Armies require, for their operations, zones with fronts proportionate to their effectives, for it is obvious that they should neither occupy a space so vast as to deprive them of the density necessary for powerful action, nor on the other hand be so overcrowded that their component parts cannot each contribute at the proper time its share in the effort towards a common object. The proper width of these zones depends on the balance of many factors between two reasonable limits; but, above all, it is necessary that all the elements that are to be brought into action at the same time should have at their disposal enough routes, more or less parallel, leading to the objective, and that these routes should be approximately at deploying interval apart. Now if Germany attacked France without violating any neutralities, the available width between Switzerland and Luxemburg was practically the same as it had been in 1870. The frontier had changed its position. It is true that Strassburg and Metz were comprised no longer in France but in Germany, and to that extent Germany had gained. But, apart from the fortresses, this frontier was only a conventional line, devoid of strategic interest. The space available for the deployment of the armies, and the distance in a straight line from one neutral frontier to the other, had not altered. Further, not all this space was equally utilizable; the Vosges region, for instance, which was very unsuitable for military operations, formed a large part of it. Thus if the available zone of the Franco-German frontier had seemed somewhat cramped even in 1870, it was far more so - and was becoming impossibly so - for the much larger armies of 1914. If the Germans violated no neutrality, it was out of their power to bring into play the ensemble of their attacking forces, and this fact gave to the French army - presumed to be numerically much inferior - an immense advantage. With its fortresses of Belfort, Epinal, Toul and Verdun, the French front of contact was very strong, and moreover, organized both to resist any brusque attack of the nature of a " bolt from the blue" and to enable the French army on mobilization to concentrate close up to the frontier without fear of being disturbed. Such, at any rate, was the adopted French view, though the experience of the war, which brought into play destructive engines of a power formerly unimagined, suggests grave doubts as to its accuracy.

In any case, it was clear that, if the Germans wished to obtain in a short time a success decisive enough to put France out of action, or at least to cripple her sufficiently to enable a large part of their forces to be sent against Russia, the violation of one or another national neutrality was a necessity for her - that of Switzerland if it was decided to envelop the French right, that of Belgium and of Luxemburg if the French left was to be the object of the manoeuvre. It was not necessary to violate the neutralities of both flanks, but military opinion was divided as to which would be selected. Each had its partisans in the German General Staff, and neither possibility was ignored by the French. At first sight, the passage through Switzerland might seem the more difficult. But, examined more closely, it loses most of its difficulties. For, in effect, the operation would consist in slicing off such a corner of Switzerland as would give the necessary number of roads, railways and Rhine passages (which could, of course, be multiplied in the sequel). For this, the Jura region alone would be enough; and the Swiss forces, massed on the flank of the invaders, would be held in check by another army, presumably Austrian. For it must be borne in mind that the Swiss army could not have been mobilized and concentrated quickly enough to hold the line of the Rhine in sufficient strength to bar access, and that Switzerland possessed not a single fortress to support it. On the other hand, the probability of the Belgian route being the one chosen - as it was - was indicated by the systematic and prolonged German preparation of rail facilities on that front.

Thus, for many years, it had been regarded as certain that part of the German army of invasion would traverse either Belgium or Switzerland. But it was not possible to foresee the proportioning of forces that would be adopted by the Germans on the thus extended front, for the high development of their railway network, and the consequent flexibility of their concentration transports, gave every facility for changes of plan and variants. Further, it was naturally to be presumed that they would make efforts to secure in advance the agreement of the state whose territory they proposed to borrow, and the success or otherwise of these diplomatic moves would necessarily react on the proportioning of forces on the military front. And this was true whether the Germans sought by way of Belgium to reach Paris and deal France a mortal blow by capturing the capital, or by way of Switzerland to cut France in two. In either event, the French armies of the eastern frontier, once turned, would have no alternative but retreat.

The French General Staff naturally foresaw that, since the holding of Russia would devolve upon Austria, the French army would probably have to struggle alone against greatly superior German forces (possibly against a preponderance of almost 7 to 4, as previously stated). The first effort must be directed towards establishing a well-prepared and fortified front, proportioned to the forces available, and to locate this as near to the Alsace-Lorraine frontier, whence the enemy could launch a direct attack at any moment, as was possible without exposing the army to the risk of a surprise. In the next place, the possibilities of attack through both Belgium and Switzerland must be guarded against. To extend the line of battle sufficiently to secure both approaches, or even one of them, in an adequate manner, would involve such a weakening of the whole front as would enable the enemy to attack anywhere with a certainty of finding only a third to a quarter of the French forces against him. Any such plan was contrary to all the principles of war, and therefore unthinkable. A completely defensive line must extend from Dunkirk to the Vosges, the Donon, and thence, in order to guard the Swiss frontier as far as Bellegarde, to the Rhine; at the rate of one army corps to 7.5 km. of front, this line (750 km.) would require loo army corps, without reckoning reserves, to secure it - or more than four times the whole of the numbers available. If the violation of Belgium could be taken as certain the Swiss frontier need not be considered; the line would then run from Belfort to Dunkirk, a length of 550 km.; but 73 army corps - more than the entire French army - would even so be required for its defence. However, if the French staff acted on this assumption, the German army would undoubtedly abandon the Belgian plan and advance through Switzerland; in which case the French army would be turned on the right in such a way that the forces sent to secure the northern frontier could not possibly arrive in time to stop the enemy from crossing the plains of the Saone and coming down into the basin of the Seine. Such a disposition was therefore impossible. The French force must occupy the centre of the line, in Alsace-Lorraine, which was the part most threatened, and be ready to oppose on the north or the south according to the enemy's decision. Accordingly, the following dispositions were made. They are the key to the whole of the first period of the war.

The choice of the first line of defence, to be held against direct attack, was determined by the necessity of fixing it at a sufficient distance from the fortress of Metz, the outer defences of which almost touched the frontier, so that the zone of action of the entrenched camp extended into French territory in the Woevre plain, which could not be defended. The most advanced line of battle considered practicable was that of the Meuse slopes, which dominate the Woevre and continue into the neighbourhood of Verdun. This accordingly became an entrenched camp, which had to be raised to the highest pitch of efficiency as a counter to Metz, the " loaded pistol pointed at the heart of France." The line of battle was based on Verdun, and, consolidated by forts constructed on the Meuse slopes, passed in its natural course through Toul and then through the good defensive positions afforded by the left slope of the valley of the Moselle. Between Epinal and Mirecourt these heights lay further from the river, and accordingly the French line drew away from Epinal to arrive at a hill called the Cote de Vivine. Thus the entrenched camp of Epinal, on the Moselle, lay in the advance of the battle-line, and enabled an offensive flank movement against the left of the enemy attack to be made under the protection of its artillery. Behind this first very solid line, which could be held by forces very inferior to those of the enemy, a whole series of positions were available in case of a retreat, in the valleys of the rivers flowing south and north.

The front being rendered secure from direct attack, there remained the question of the two flanks, in the event of an invasion by way of either Belgium or Switzerland. Facing the road from Switzerland there was a great concave arc of positions commanding the plain of the Saone and based on entrenched camps, i.e. Belfort, Epinal, Langres and Dijon. In front of these stood Besancon, in a position to divide and delay the invading stream. Facing the Belgium road, with Verdun, which acted as a sort of left shoulder, as the starting point, was a returning arc marked by the entrenched camps Reims, Laon and La Fere. In advance of these positions Maubeuge played a part corresponding to that of Besancon on the other side.

The enemy must take a certain line to traverse either of the neutral countries, even supposing that he met with no opposition. This would allow time for the French reserves to form, and for the first-line army to concentrate along the Verdun-ToulEpinal line, to prepare positions facing N. on the left flank of an invader from Belgium or facing S. on the right of an invader from Switzerland, and to be ready to attack, in either event, at the favourable moment. Reason and prudence dictated these dispositions, in view of France's isolation, separated as she was from her Russian ally, and of the fact that Germany and Austria had the advantage of " interior lines." The dimensions of the two flanks - from Verdun to ReimsLaon-La Fere, and from Belfort to Epinal-Langres-Dijon, respectively - were in accordance with the resources provided by the French recruiting laws in force for some years after 1872; but they had become insufficient for those given by the law of 1889, which greatly increased the military sacrifices demanded from the nation and added considerably to the war effectives. The result was that the flank facing an attack from Belgium, instead of ending at the Laon-La Fere system, which was becoming useless, was prolonged as far as Paris, which, as an immense fortified camp, must, by the mere fact of its presence, play a supremely important part, as was seen in 1914. Compared with this great entrenched camp, Reims itself was quite secondary. As for the other flank, it could be continued indefinitely beyond Dijon, by means of the formidable defensive positions provided by the mountains of the Cote-d'Or, which commanded the plains of the Saone. No new fortifications had been considered necessary. The line of battle opposed to Metz could also, because of the increased military resources, be prolonged N. of Verdun, still keeping the edge of the Meuse slopes up to Stenay and even beyond, so as to constitute a threat to the German right, on condition, however, that this right did not extend into Belgium. In the event of a turning movement by way of Belgium, the French left was not to go beyond the Verdun shoulder, and thence would trend away into a defensive, refused flank.

An alternative policy, that of entering Belgium and by reinforcement enabling the Belgians to hold the line of the Meuse, was the subject of a good deal of consideration in French military circles before the war. It involved, however, an extremely complex problem. The attitude of Belgium on the question of admitting French and British forces had been one of marked reserve, although there had been pourparlers at different times between the Belgian and the British authorities; and in fact it was not till Aug. 4 1914 that Belgium asked for the support of England, France and Russia, undertaking herself the defence of the fortified places. Yet, even had Belgium accepted British and French aid earlier and a united Allied front been formed along the Meuse, the strategic situation thus created would have been very difficult, owing to a cause which was operative whether the French advanced to the Belgian Meuse or not. The line of the Meuse runs N.-S. between Mezieres and Maestricht, while the direction of the frontier between Mezieres and the Vosges is substantially E.-W. A German concentration in the region Aachen-Trier would therefore occupy a zone midway between these two lines, and could act in the direction of either as circumstances dictated. Thus, whether the French army, flung northward, was to go as far as the Belgian Meuse or only to the French territory adjacent, was certainly a question of very great local significance, because the Soo-km. line of defence from the angle to the sea, destitute of natural defences and weak in artificial, was distinctly inferior to the short, strong, well-fortified line Givet-Namur-Liege. But it did not alter the fact that the German forces concentrated between Aachen and Trier might, after perplexing the defence by demonstrations, fling their weight upon the line between Mezieres and Verdun, break it by means of superior numbers, and so gain a position not only between the separated halves of the French but also nearer to Paris than either.

These conditions, together with Belgium's hesitating attitude, practically imposed the defensive principles upon which the French General Staff must proceed. Obscured as they were by the dramatic events of Aug. 1914, - by the glorious insistence of Belgium, the French offensives into Alsace, Lorraine, and the Ardennes, the tidal wave of the German I. and II. Armies traversing the Belgian plain and northern France, - it was nevertheless on these principles that the German effort was shipwrecked. For in Sept. 1914 the breakwater of the defence was established solidly on the line, marked substantially by the Vosges, the natural defences of Lorraine, the Cote de Meuse, Verdun, the Montagne de Reims and the advanced defences of Paris, which strategic reasoning had already indicated as the basic line of defence for France in the given conditions.

These conditions include other alternatives than the one selected by the Germans; and it may be asserted that, given the fact of Belgium's resistance and of England's intervention, the course taken by the Germans was - as against the alternative of a violation of Swiss neutrality, which would have occasioned much less concern to England than that of Belgium, and even as against that of a frontal forcing of the Lorraine defence, which perhaps was not as invulnerable as it was believed to be - the course which was the least disadvantageous for France.

(H. BE.) The French "Plan r7." - The characteristic of all French plans of concentration up to those bearing the No. 16 was that they were all applications of the defensive principles outlined above, differing only in detail, and providing for an initial defensive phase of operations out of which an appropriate counter-offensive would arise when the occasion was ripe. From 1912 onwards, however, a new school of thought had begun to prevail in the French General Staff. The teachings of Colonel (afterwards General) Loiseau de Grandmaison, the constant improvement of the mobilization scheme in details, the sharper tone of policy and sentiment after the Agadir crisis, all combined to create a "younger school" in the staff which did not admit that the army was so inferior in power or war-readiness that the defensive need be assumed a priori, as had hitherto been the case. Military France, like the rest of military Europe, was caught by a wave of enthusiasm for the offensive per se; doctrines and text-books were revised, senior officers, and generalissimo-designate, having predilections, real or alleged, for the defensive, were got rid of; and as soon as it became clear that the process of mobilization and concentration had been sufficiently accelerated, " Plan 17 " was drawn up, with the immediate general offensive in full force as its keynote.

" Plan 17," issued to commanders-designate of armies and their chief-of-staff in Feb. 1914, was based on certain assumptions which may be summarized as follows: On the right wing, the hypothesis of a German invasion through Switzerland was assumed to be so improbable that only an echelon of three reserve divisions, and these available for active operations towards Belfort, was allotted to that flank. On the left flank, the problem was far more delicate and difficult, as it depended on whether or not Belgian neutrality would be violated, and, if so, how far N. the right of the German forces would extend. It was involved with two other questions, that of the attitude of Belgium and that of the strength of the German army; neither of these was answered very definitely, and the assumptions of the plan proved substantially incorrect. Belgian aid was not counted upon - indeed, in one important detail provision was made for the case of the Belgians not interfering with a German march-througharid the German army for battle purposes was assumed to contain only some 20 or 21 active corps, the reserve divisions, it was thought, not being available till after an interval, and then only for subsidiary functions such as sieges and railway guarding. The conclusion drawn was that the German right, in case Belgian territory was taken in, would extend to the limit of the Ardennes - i.e. the Belgian Meuse - at the farthest, if as far. But the hypothesis of a frontal attempt of the Germans to break through between Longwy and the Vosges, without touching Belgian territory, was the basis of the plan; and the measures to be taken. in case Luxemburg and the Belgian Ardennes came into the theatre of war were embodied in a "variant." It was supposed, in addition, that attempts might be made by the Germans in Lorraine or the Woevre to break into the French concentration areas in the first days of hostilities; and a very strong protective system (drawn back in the Woevre out of range of a sortie from the Metz outer defences) was provided against this emergency, the augmentations of the peace effectives brought about by the "Three Years Law" having made this possible. On to this protective system, constituted by one corps of each front-line army, the remaining corps were to graft themselves as they arrived, and the whole was to be ready for active operations on the 12th day of mobilization. It was assumed - correctly - that the Germans would attack, and - incorrectly - that their attack would be a simultaneous onset of fairly evenly distributed forces; and it was argued that a French offensive, debouching with startling rapidity, would create a situation with which the German military system was not elastic enough to deal.

These active operations, if Belgian territory remained untouched, were to be a general offensive of four armies with another immediately behind them, directed eastward from the Meuse below Verdun and northeastward from the Nancy-Vosges front, northward from Belfort; and, if Luxemburg and Belgium were infringed, an equally general offensive with all five armies in line, those of Alsace and Lorraine directed as before, but those of the Woevre and the middle Meuse northeastward and even northward according to the positions found to be occupied by the German right. In either case the central army, besides helping its neighbours as required, was to drive back all sorties from Metz and begin the investment of that place.

The dispositions of the plan were as follows: The I. Army (five corps, two cavalry divisions and army artillery) was to attack with its main body from the concentration area west of the Vosges in the direction Baccarat-Saarburg-Saargemiind; the right, VII. Corps and a cavalry division based on Belfort, to advance into upper Alsace, rouse the population to a revolt, and hold as large a German force engaged as possible; between the VII. Corps and the main body, a smaller force in the Vosges was to maintain liaison and by descents into Alsace to cooperate with the advance from Belfort.

The II. Army (five corps, two cavalry divisions and army artillery), grouped initially about Nancy and Luneville, on the left of the I., was to attack in the direction Chateau SalinsSaarbriicken. The improvised fortifications of Nancy in the first stage, and a group of reserve divisions issuing therefrom in the later stages, were to protect the left of this army against Metz;.

and the I. Army, developing its advance along the Vosges, was to guard the right, cooperating in the battle of the II. Army with all the forces not absorbed by the flank along the Vosges.

The III. Army (three corps, three reserve divisions, one cavalry division and army artillery) was to connect this " principal attack" in Lorraine with the other "principal attack" mentioned below, first by holding the Cote de Meuse between Verdun and Toul, next by repelling sorties from Metz and blocking up the west front of that fortress, and lastly by giving support to the attacks of the neighbouring armies.

The V. Army (five corps, two reserve divisions, one cavalry division and army artillery) had to deal with two alternatives, those of violation or non-violation of Belgian territory. In the latter case, it was to drive eastward from its concentration area N. of Verdun and the Argonne across the Meuse, dropping in its progress a flank-guard to watch the Belgian frontier; its objects were to defeat and drive northward all German forces encountered, and to storm or invest, according to circumstances, the fortifications of Thionville (Diedenhofen), guarded and assisted on its right by the III. Army. In the first alternative, it was to be so disposed that it could both attack northeastward on Neufchateau and Florenville in the Ardennes, and guard its left rear with a special detachment.

The IV. Army (three corps, one cavalry division and army artillery), concentrated behind the III., was the general reserve. It was destined to be used either on the right or on the left of the III. Army according to which of the two "p rincipal" attacks - Lorraine or Ardennes - needed additional weight. If the offensive of the V. Army was directed upon Neufchateau and Florenville, the IV. Army was to come in between the V. and the III., and fight its way in the direction of Arlon. Behind the right were to be three reserve divisions, ready to follow up the VII. Corps and take over the guard of the Rhine as it advanced. Behind the left, but not definitely allocated to the V. Army, were to be three more reserve divisions about Vervins, with a somewhat indeterminate mission. A corps of several cavalry divisions was to form about Mezieres in the first days of mobilization on the left of the protective system, and thereafter to operate eastward or northeastward into the Ardennes as required. Its supporting infantry was to occupy the bridges between Dinant and Namur if the Belgian Government did not do so.

This was the plan which was carried into effect when war came in August 1914. As early as Aug. 2, it was decided to act on the hypothesis of a German movement through the Belgian Ardennes, the seizure of Luxemburg by the German advanced guards on that day being a sufficiently suspicious indication. But during the following days the French General Headquarters were confronted with a mass of definite and indefinite information which it was hard indeed to appraise. On the protective line, apart from two severe local fights, at Mangiennes in the Woevre and Lagarde in Lorraine - the first a French, the second a German victory - there were no events and no important indications. To the N. of the left flank, want of liaison, and, it must be added, of mutual confidence, made it difficult for the French to gauge exactly what the Belgian army would do, and especially what was happening at Liege. That fortress was attacked on Aug. 5, and its capture (see Liege) was announced as a fait accompli on Aug. 7, yet for many days thereafter the gathering masses of the Germans between Aachen and the Ardennes seemed to make no move.

The British Expeditionary Force (four divisions and a cavalry division) was about to land in France, but it was not comprised in " Plan 17." A secret appendix to the plan, known to a few, provided for a hypothetical "Army W." landing from overseas and proceeding to the region of Valenciennes and Maubeuge, but the way in which this army (should it materialize) might best be employed could not be seen until the role of the French V. Army had become clearer. Meantime, it was to double the part of left echelon which was assigned to the French reserve formations about Vervins.

But meantime, the troop-trains were arriving in the concentration areas, and the broad " Plan 17 " had to be replaced by an operation order "No. 1." On the morning of Aug. 8, therefore, General Joffre, general-in-chief of the "Armies of the NorthEast," issued his specific instructions.

The French Offensive

The enemy, it seemed, had grouped his main forces in the region of Metz, in front of Thionville (Diedenhofen), and in Luxemburg, with some 12 divisions in Lorraine and Alsace and an undetermined force which included parts of Io divisions in the Liege and Ardennes regions. This main force (Metz - Luxemburg) seemed to be pointing westward, but might equally well swing southward, pivoting on its fortifications. The French armies were, consequently, to take the offensive - which was to be as foudroyante as possible - and with all forces in combination to seek to bring the enemy to decisive battle, resting their right flank on the Rhine. In order to ensure simultaneity and unison in the battle effort, it was laid down that the left wing armies might have to hold back, so as not to become involved in battle with German masses traversing the southern Ardennes or northern Woevre, or both, before the right wing had advanced and made effective contact with its opponents.

The I. Army (Gen. Dubail), composed as in "Plan 17," was, instead of merely cooperating with and flankguarding the II. (as in the plan), to become the main offensive element in Lorraine and Alsace. Its VII. Corps, with a cavalry division, was to break into upper Alsace at once from Belfort, to drive back all forces it met, and, progressively reinforced by the three reserve divisions from Vesoul, to gain ground towards the fortified barrier Strassburg - Molsheim, destroying bridges and blockingup bridgeheads on the Rhine as it advanced. The main body of the army, with a frankly E.N.E. direction, was to push towards the front Fenestrange (Finstingen) - Saarburg - the Donon, and to drive back its opponents on Strassburg and into lower Alsace.

The II. Army (Gen. de Castelnau), composed as in the plan, was now to play the part of auxiliary to the I. Its first objective was to be the front Delme - Salins - Dieuze, and its axis Chateau Salins - Saarbriicken. It was to flankguard towards Metz, and, moreover, to leave two of its five corps in the region of Toul at General Joffre's disposal.

The III. Army (Gen. Ruffey), constituted as in the plan, was disposed in the Woevre facing Metz, and was to be ready either to counter-attack any German forces emerging from the Metz region or to take the offensive northward, with its left on Damvillers, according to the situation. The two corps taken from Castelnau would probably be employed in concert with this army, either in repelling a counter-offensive from Metz or in a northward movement.

The roles to be given to the I V. and V. Armies were now defined more precisely. The IV. (Gen. de Langle de Cary) was to group itself between Argonne and Meuse, and the V. Army (Gen. Lanrezac) to condense between Vouziers and Aubenton, ready either (a) to attack any German army which traversed the Meuse between Mezieres and the line Damvillers-Montfaucon or (b) to cross the Meuse themselves for the Ardennes - Arlon offensive. The II. Corps, hitherto the left wing of the protective system and attached to the V. Army, was now added to the IV. Army and directed to hold firmly to the northern outliers of Verdun and the left flank of Ruffey's army. Beyond the flank of Lanrezac was the group of reserve divisions about Vervins; the cavalry corps operating E. of Mezieres and Montmedy was expected, if and when forced back over the Meuse, to take positions about Marienburg and Chimay. The role of " Army W." was as yet quite unsettled, as also was that of the Belgians.

Such was the order which initiated the "Battle of the Frontiers," the opening of the World War on the western front. The intentions may be, and have been, criticized, but they are clear. The general offensive of the French right wing, fixed for the 12th day of mobilization (Aug. 14), was directed N.E. and E.N.E. into the Rhine lands behind Strassburg and Molsheim, with a subsidiary effort in Alsace which would make good a front facing Molsheim - Strassburg - Neu Breisach and the upper Rhine fortifications, and, in case of success bringing the I. Army to behind Strassburg, besiege that fortress. The general offensive of the left was to: be timed to coincide with the decisive phase of the operations of the right, but placed according to the progress of the main enemy armies which were presumed to have their right flank not farther N. than Mezieres and their left flank on or in the fortified region Metz - Thionville.

But the possibility of arranging for the whole system to take its time from Dubail was made doubtful by a geographical factor - the Meuse. Had the areas in which the IV. and V. Armies were to act formed a single region, it would have been different, but the critical question was how to get these two armies over the Meuse at the exact moment determined by events in Lorraine, which might also be the very moment at which the German masses from Luxemburg themselves arrived on the river. It was this disturbing factor, quite as much as any events to the N. of Mezieres, which governed the development of the French scheme. The details of this development, so far as concerns the left wing, are extremely complex and must be studied in the documents reproduced in F. Engerand's Briey, Joffre's Preparation de la guerre et conduite des operations, Lanrezac's Le plan de campaigne francais, and the anonymous Le Plan XVII. (pub). Payot). Here only a summary can be given.

By Aug. 13, the eve of the day fixed originally for the general offensive, the order of Aug. 8 had ceased to apply integrally to the left wing. While Dubail and Castelnau were to advance on the front Donon - Saarburg - Saarbriicken, as previously indicated, and a new and stronger army of Alsace under Gen. Pau was to carry out that part of the scheme which the VII. Corps had just attempted with disastrous results (see Frontiers, Battles Of The: section Alsace), Ruffey, de Langle de Cary and Lanrezac were directed according toa new scheme which was independent in time as well as place of operations in Lorraine. It was now clear that the German northern group was stronger than had been supposed, but the evidence of its intention to cross the Meuse above Liege and sweep round through the Belgian plain seemed less convincing than the probability of its descending southward, and the French Command, after much interchange of views with Lanrezac, decided to push part of the V. Army northward into the region W. of Givet as a defensive precaution, and with the remainder and the IV. and III. Armies to carry out a series of preparations which would suit either of two hypotheses. If the Germans moving westward through the Ardennes were well advanced, they were to be struck by an offensive against the front, flank and rear, as soon as they were thoroughly involved in the crossing of the Meuse; if, as now seemed more probable, they were grouped with greater density in the northern and northeastern parts of the Ardennes, there would still be time for the V. and IV. Armies to advance before battle, not only over the Meuse, but also over the Lemoy and the lower Chiers. In that event the III. Army, which, with various mobile forces belonging to Toul and Verdun and the corps of the II. Army reserved to the general-in-chief by the order of Aug. 8, could assemble a considerable force, was to divide into a defensive group facing Metz and an offensive group which would aim northward, conforming to the right of the IV. Army, which would make good the lower Chiers; while the V. Army, holding defensively on the Meuse at Givet, was to reach the front Beauraing - Gedinne - Paliseul - Cugnon.

On Aug. 14, the offensive in Lorraine opened. Its progress was slow, but not marked by any untoward incidents up to Aug. 19. It was independent of events to the left of the Moselle except in so far as sorties might emerge from the S. and S.E. fronts of Metz, against which contingency the defensive group of the III. Army, the two reserved corps near Toul, and the echeloned left wing of the II. Army, were an adequate safeguard. On the other wing, however, obscurity still prevailed. Though Lanrezac was becoming more and more uneasy as to his left flank, and the Belgians, standing on the Geete line, called for support, nothing positive as to the German dispositions revealed itself, but on the evening of the 15th the veil was, partly at least, torn away. The part of Lanreza :.'s army which was stationed in the angle of the Sambre and Meuse became engaged with a large force of the enemy at Dinant. This consisted in reality only of cavalry and light infantry, but was estimated by the French commander, Franchet d'Esperey, as an army corps at least. At the same time, information came in tending to show that the Germans in the Ardennes included sixteen divisions. Moreover, the impression was formed both by Franchet d'Esperey opposing them and by the French cavalry commander skirting their southern flank, that the Dinant Germans were flankguarding a much more considerable force engaged in passing the Meuse below Namur; and Lanrezac energetically insisted on the fact that such large enemy forces could not conceivably be intended to operate entirely on the right bank of the Meuse. A remarkable absence of troops, at the same time, was reported by the French aviators reconnoitring the Arlon region. Thereupon Joffre formed a new plan. The V. Army, except one corps (already attached to the IV.), and its two reserve divisions, reinforced by one of the reserve corps on the Moselle and by forces from Algeria, newly arrived, was to join the forces already in the angle of the Sambre and Meuse, and to cooperate with the Belgians and the British - both of which armies now for the first time figured in the scheme of operations - in attacking the front and outer flank of the German " northern " forces, while the IV. Army was to prepare to debouch from the front Sedan - Montmedy in the direction) of Neufchateau against the " southern " enemy group (formerly presumed to be the " main " one) advancing from Luxemburg on the front Sedan-Damvillers; and the offensive portion of the III. by Etain and Jametz, was to be ready to march on Longwy, to break into the rear of this force. But this manoeuvre was merely sketched out by preparations, and next day the veil descended again. The supposed movement of German masses over the Belgian Meuse was unconfirmed, and indeed denied; and without for the moment devoting more attention to the details of the cooperation to be obtained between the three distinct and independent commands W. of the Meuse (especially as Sir John French indicated Aug. 21 as the earliest date at which the British could come into action), Joffre's headquarters merely sent the cavalry corps withdrawn from the Ardennes to get into touch with the right of the Belgian positions on the Geete and - as a precaution against minor inroads through the Belgian plain into the industrial region of Lille - began the transfer to Arras of some ill-equipped territorial divisions, which, under General D'Amade, had hitherto watched the Italian frontier. It was to the proposed offensive of the IV. and especially of the II. Army that Joffre devoted his principal attention. For, on the estimate which had been formed of the German strength - which, however, was radically incorrect because it ignored the presence of reserve corps in immediate proximity to active corps - it seemed to certain of the directing brains at Vitry-le-Francois that the more forces the Germans placed west of the Belgian Meuse the slighter would be the resistance to be expected about Neufchateau and Longwy and the better the chances of cutting the enemy in two by the offensive directed on these points.

From day to day the situation developed without becoming quite clear (for the Germans veiled their dispositions with the utmost success), and Joffre held stubbornly to the conception of his Neufchâteau-Longwy offensive. Lanrezac's anxieties and those of the Belgians increased, but they served only to confirm the impression that the drive into the Ardennes would, if properly timed and directed, reap a great harvest; and the declaration of Sir John French that he would not be able to begin operations till Aug. 23, instead of Aug. 21, caused the scheme of a combined operation N. of the Sambre to recede still farther into the background. On Aug. 20, Joffre, estimating that all the German forces destined for the Meuse had by that time passed out of the region Audun-leRoman, Arlon, Luxemburg, gave the orders which launched the IV. Army into the Ardennes and the III. on Virton and Longwy.

The period of nuances was at an end. On that very day, on the one flank, Dubail's and Castelnau's offensive, which had penetrated to Morhange, Saarburg and the Donon, met defeat (see Frontiers, Battles Of The: section Lorraine). Castelnau drew back hastily towards Nancy-Luneville; Dubail, in spite of the exposure of his long right flank in the Vosges (which Pau's methodical advance from Belfort had done little to shorten), took down his left more steadily; but almost in a moment both were back in their concentration areas, followed by the eager enemy. On the other flank, the German masses facing the Belgian army front, hitherto screened by their cavalry, had at last declared themselves on Aug. 18, and the Belgian army, threatened with separation from Antwerp, yet most unwilling to give up the expectation of British or French support on its S. flank, was falling back from one position to another. Its decision to close up northward and fall out of the main operations was, it must be recognized, put off to the last possible moment, but the disconnectedness of the Allied movements left no alternative. For at that date Sir John French was not ready; and in the angle of Sambre and Meuse, Lanrezac, a prey to new and not ill-founded anxieties regarding his liaison with de Langle de Cary, was, with Joffre's approval, standing fast till Aug. 23, the date British cooperation should arrive.

In spite of the anxieties and disappointments caused by these events, joffre held firmly to his intention. On the morning of Aug. 21 the executive order for which the IV. and III. Armies were waiting was sent. For the situation was now clear, and the plan of breaking through between the German manoeuvre-masses and their fixed pivot, which in one form or another had been consistently followed in the period of obscurity, seemed destined now to have its reward. But there was one fundamental miscalculation. The old error which had led the professional soldiers of Napoleon III.'s day to regard the Prussian citizen-army as a " sort of militia," had reappeared in the form of a contempt for " reserve " formations. It was a mere matter of calculation that Germany's resources permitted her to create such formations; but that they should figure in the masse de choc was regarded as incredible. Yet it was true; and thus, instead of meeting a battle-army of 42 to 50 divisions with an array of 73 French, 6 Belgian and 4 British, as anticipated, the Allies encountered in reality one of 77 divisions, i.e. an equivalent instead of a much inferior force. This was especially important as bearing on the prospects of success in the Neufchateau and Arlon directions. Strength was encountered where weakness was expected, and the relatively small numerical superiority of the attack did not suffice.

The story of the battles of Longwy and the Ardennes, of Charleroi, and Mons, will be found in detail in the article Frontiers, Battles Of The. Here it must suffice to say that the French offensive into the Ardennes and towards Virton-Longwy-Audun-le-Roman met with general failure, and in some places with disaster; that the German II. and I. Armies, swinging on Huy as a pivot, swept down upon the French V. Army at Charleroi and the British Expeditionary Force at Mons, and bore them back; and that on Aug. 25 at 22 :00 hours (10 P.M.) Joffre's orders were issued for a general retreat. The German plan of campaign had prevailed, and the German Command had the initiative in its hands.

At this point, then, the story of the operations is most conveniently told from the point of view of that side which dictated their course.

The German Plan of Campaign.

For the Germans, a war against France was essentially part of a two-front war. The resources of the country not being equal to simultaneous offensives against France and Russia, the choice had to be made between (a) standing on the defensive against France while seeking a decision in battle in the East, (b) waging a defensive war on both fronts, and (c) striving to crush France while standing on the defensive in the East. Of these (b) was held to be excluded by the presumed impossibility, for an industrial state, of enduring a long war, as well as by obvious military objections; (a) was never completely excluded, and had until some ten years before the war been the fundamental war-plan of the German General Staff; while (c) had in those last years obtained general acceptance, owing to the difficulty, for Germany, of waiting till the slow-moving Russians could be brought to action and defeated in a battle of the first magnitude. Whether, in view of the increased strength of the defensive on the one hand and the increased war-readiness of Russia on the other, the adoption, once more, of alternative (a) was not the best policy for Germany in the circumstances of 1914, is an open question; but, in fact, (c) was maintained and carried into effect.

But this increased readiness of Russia made it imperative for the Germans to protect East Prussia by a force at least sufficient to offer a step-by-step defence of that province and also, with their main armies reduced to that extent, to obtain a decision of the war in the West at the earliest possible moment, so as to release the greater part of the forces which had gained it for service in the East. The proportioning of means to the two theatres, therefore, was a very difficult problem, admitting of many a priori solutions, which might bring either victory or ruin.

The solution that found most adherents was that of Count von Schlieffen, chief-of-staff of the German army, in the first years of the present century. On assuming office, he had both restudied the draft plans of campaign and the tactical doctrines in vogue, and he had come to these conclusions: (a) that an offensive of maximum power, carried so far as to put France out of action definitively, was the only way to secure freedom of action in the East; (b) that this offensive, to secure the result aimed at - nothing less would suffice - must be developed on so broad a front as to grip and out-wing the most northerly and the most westerly points of France's defensive dispositions; (c) that a maximum density must be sought for on the right wing, even at the cost of exposing Lorraine and the Rhine lands to invasion.

The first of these considerations led to the acceptance - contrary to all the traditions of the German army - of the principle that not only active, but reserve, Ersatz and every other category of soldier must be effectively used. Schlieffen even proposed an intimate mixture, practically an amalgamation, of active and other elements, and aimed at putting into the field - in case of a single front war, it is true - no less than 114 divisions against France. The second consideration led to the idea of a swing through Belgium and northern France far wider than that which was actually carried out. The route of the outer flank, which in spite of its extension beyond all probable French defences was to have a defensive echelon following on, was to touch Dunkirk, Abbeville, Rouen and pass round by Chartres, far to the W. of Paris - which would be invested automatically - so as to march in upon Auxerre and Troyes from the east. The line Ghent-Maubeuge-Thionville was to be reached on the 22d and the line Amiens-Rethel-Thionville on the 31st day of mobilization; that is, not hurry but certainty and power were to be the executive rules.

The third consideration, however, led to an even more remarkable result than the second. Of the 114 divisions no fewer than 101 were to operate N. and W. of the Thionville pivot, Lorraine, Alsace and the Rhine lands being committed to 13, of which two were allocated to the fortresses. The leitmotiv of extreme density on the right occurs in all Schlieffen's drafts and schemes. ." Macht mir nur den rechten Fliigel stark " were his half-conscious dying words. In a second scheme, based on the two front war, which provided for a defence force in East Prussia approximately equal to that which von Moltke actually placed there in 1914, the total force was lessened to that extent, but the ratio of about seven divisions N.W. of Thionville to one S.E. of that pivot was maintained.

When von Moltke, the younger, succeeded Schlieffen, the above plans were gradually blunted, first because the idea of making the active army a simple kernel for soldiers of all categories was accepted only in part, and secondly because the growing war-readiness of the French army, the fever of offensive spirit that had obviously seized it, and after 1913 its very high peace-strength, made it increasingly likely that the French would open the war with a determined offensive into Lorraine and the Rhine lands. In these circumstances so drastic a depletion of the forces to the left of Metz as that contemplated by Schlieffen did not commend itself to Moltke, who found a compromise in allocating one-quarter of the whole available force, instead of one-eighth, to the defensive (or defensive-offensive) front, and holding large quantities of empty rolling-stock on the Rhine in readiness to transfer a proportion of this quarter to the right wing as soon as circumstances should allow this to be done. The amplitude of the swing was, however, undeniably diminished thereby.

In one point, Schlieffen and Moltke were agreed - the necessity of pushing out beyond the line of the Belgian Meuse. In both schemes therefore the quick seizure of Liege and a deployment foreground beyond that fortress figured as an indispensable preliminary to the operations proper.

The German plan, to which effect was given in August 1914, provided as follows: (a) A protective system was formed all along the line, consisting, not as had been expected of complete formations, but of single brigades of infantry (with a proportion of other arms), brought up from their peace stations without waiting to receive and equip their reservists. These brigades took over from the local troops the positions of the line that their respective army corps were to occupy, and their reservists rejoined by parties. (b) There was a concentration of the remainder of each corps, after mobilization in the usual way, behind its own representatives in the protective system. This concentration of the active corps was completed by the 14th day of mobilization (Aug. 15). (c) Concentration of reserve corps, as a rule immediately behind or to a flank of the corresponding active corps, was completed by the 16th day (Aug. 17). (d) Concentration of other formations, Ersatz divisions and mixed Landwehr brigades, was completed from the 11th to the 17th day. (e) The six advanced, peace-strength brigades of the corps intended to assemble about Aachen were employed as a striking force under General von Emmich, which without waiting for siege artillery was to attempt to storm Liege at once, if the Belgians did not agree to let the Germans pass. The German ultimatum to Belgium was handed in late on the evening of Aug. 2 (rst day of mobilization), and required an answer within twelve hours. There is reason to believe that, on Belgium's refusal, a proposal was made to Holland to allow the use of the Maestricht tongue as a gateway into the Belgian plain, but, whether this be so or not, it was only on the evening of the 3rd day of mobilization that the striking force crossed the frontier.

The order of battle, and allocation of the German forces N. of the pivot, was as follows: I. Army (General-Oberst von Kluck), five corps, assembled behind the Maestricht tongue (JiilichKrefeld area); II. Army (General-Oberst von Billow), seven corps, including one attached from I. Army, assembled facing the Liege frontier (Aachen - Malmedy - Euskirchen); III. Army (General-Oberst von Hausen), four corps, assembled in the area St. Vith - Wittlich - Bittburg; IV. Army (Duke Albrecht of Wiirttemberg), five corps, assembled in the area Luxemburg (seized Aug. 2)-Trier-Diekirch - Wadern; V. 'Army (Wilhelm, German Crown Prince), five corps, assembled in the area Metz - Thionville - Saarbriicken.

Thus twenty-six active and reserve corps (52 divisions) - with a number of Landwehr brigades to follow - were allocated to the five armies of the moving wing, of which nearly half were to cross the Meuse between Namur and the Dutch frontier as soon as the way was clear. The remainder were to traverse the Ardennes from E. to W. in echelon from the right (III. Army) and to prepare to wheel gradually S.W. in proportion to the progress of the I. and II. Armies on the other side of the Meuse.

The 2nd Cavalry Command was of three divisions in front of the I. and II. Armies; the 1st Cavalry Command of three divisions in front of the III.; the 4th Cavalry Command of two divisions in front of the IV. They were individually responsible to the Supreme Command, except when from time to time placed at the disposal of certain armies. Similarly, the armies were in principle directly subordinate to the Kaiser's headquarters, i.e. to General-Oberst von Moltke, Chief of the General Staff, unless temporarily paired, as was the case at the outset, with the I. and II. Armies, of which Billow was in general charge.

The organization of the forces E. and S.E. of Metz was somewhat different. Rupprecht, Crown Prince of Bavaria, was appointed not only Chief of the VI. Army, but " Commander-inChief in the Rhine lands," a title with wider implications than that of an army commander, and his task included the protection of the left flank of " the Army " or " the main forces." He had under him his own VI. Army (five corps, of which four were Bavarian), the VII. Army of General-Oberst von Heeringen (three corps) and the 3rd Cavalry Command (three divisions), in addition to the war garrisons of Metz, Strassburg, and the upper Rhine defences (Neu Breisach, Istein). These 16 active and reserve divisions were to be supplemented a few days later by 62 Ersatz divisions, and mixed Landwehr brigades, which were grouped in some cases in "Landwehr Commands." In all, the formations assigned to the western theatre of war comprised 68 active and reserve divisions, 62 Ersatz divisions, 17 2 Landwehr brigades and 10 cavalry divisions.

The First Operations of the Germans

The first operations to be carried out were: (r) The seizure of Liege and of as large a foreground as possible beyond; (2) the securing of the left flank of " the Army," and the attraction of as many French divisions as possible to Lorraine and Alsace, by the living and passive forces under Prince Rupprecht. Both these essentials having been provided for, the five armies (I. - V.) were to proceed at once to the main task, which was to be a great " wheel through Belgium into France, pivoting on Thionville - Metz," in which wheel the II. and I. Armies were to govern the pace. The II. Army was to swing on the arc Liege - Wavre, the I. following in echelon on the arc Tongres - Brussels, flankguarding towards Antwerp, whither, it was presumed, the Belgian field army would retire. No very distant objectives were fixed. When Liege and the region W. of the Meuse had been occupied, and the cavalry divisions of the I. Cavalry Command had fixed the situation of the Belgian army, of the British forces expected to land at Ostend, and of the French forces which might be pushed into Belgium from the S., specific orders could be given. Meanwhile, the problem was to deploy the two highly condensed armies of Kluck and Billow on the W. side of the Meuse, in spite of Liege and of the Belgian field army. On this, as the I. and II. Armies were to give the time to the rest, the whole movement depended; but while the III., IV. and V. Armies awaited developments, the 2nd and 4th Cavalry Commands were to push through the Ardennes and towards Damvillers, in order to clear up the situation in the Meuse valley from Namur to St. Mihiel - a mission which naturally brought about a series of conflicts with the French cavalry, and, above Stenay, with the French protective infantry system in the Woevre. In point of information, this cavalry activity probably yielded only confirmations of the obvious, but it was invaluable in veiling the army movements when later these were begun.

The Liege operation is described under Liege. There was no surprise, Belgium having mobilized her available forces at an early date in the European crisis; and five out of six brigade attacks on the night of Aug. 5-6 failed. One, however, penetrated into Liege, and for some days neither the German nor the Belgian, British or French Commands seem to have been able to form a clear view of the situation, as the fort-ring held out. Reconnoitring patrols of the German 2nd Cavalry Command, which passed the Meuse near the Dutch frontier simultaneously with parts of the attacking force, learned nothing of the Belgian dispositions, and the main body remained near its bridge. On Aug. 8 the situation at Liege was clear enough to enable the cavalry to push westward, and in the days following it established the front of the Belgian field army as lying approximately on the Geete line from S. of Jodoigne to Diest. But on Aug. 12, the attempt of the Germans to work round to the left of this position was checked in a severe local action at Haelen. The Liege forts, meantime, had to be reduced by super-heavy siege artillery, one by one, and it was not till Aug. 15 that the masses of the German I. and II. armies were able to begin to cross into the area comprised between Liege and the Geete. The resistance of the Liege forts had put back the start of the great wheeling movement for four days.

Thus the difficult and essential preliminary operation of seizing a bridgehead beyond the Meuse did not pass off quite according to programme. It had often been alleged that the Germans had obtained a long start in their concentration by means of surreptitious mobilization. That this was not actually the case has been shown by a study published in 1920 by the Belgian General Staff; but even had such a mobilization taken place, it would not have helped to solve the problem of Liege. All that infantry could contribute to its solution, the first six peace-strength brigades contributed. The rest was a matter of siege artillery, and it must be regarded as a serious flaw in Moltke's plans that this artillery did not put in an appearance on Aug. 6 instead of on Aug. 10.

On the other flank, the operations in Lorraine and Alsace, which were to maintain the "pivot" of the wheel against French attack proceeded more favourably than had been anticipated - too favourably, as the sequel was to show. The problem was difficult, and a priori reasoning could not assist its solution materially. For here, much more than on the Belgian flank, events depended upon the independent will of a great enemy army which was equally capable of the offensive or of the defensive-offensive; and although the tendency of French military thought in the last years before the war had evidently been towards the former, it was not clear how far this tendency had actually gone. Prince Rupprecht's task included not only the maintenance of the Metz - Thionville pivot but the attraction of as large a force of the French as possible; and if their old policy of defensive-offensive still held good, then, in order to set up this attraction, he must attack. If not, although attack would not be necessary, a much larger proportion of the total force than that which Schlieffen had thought sufficient would have to be allotted to the front E. of Metz. For Schlieffen had reckoned with certainty on a defensive-offensive policy in his opponents, and had argued therefrom that no deep inroad of large forces into the Rhine lands would in fact be attempted by the French, when a tidal wave of seventy to a hundred German divisions was advancing through northeastern France. But the situation was quite different if the French offensive in full force was to be launched at the outset, while the tide in. the N. was still dammed up. For these reasons, Prince Rupprecht was given about one-quarter of the total available forces, and instructed: (a) to take the offensive over the Lorraine frontier in order to draw upon himself as much as possible of the enemy's forces, in the event of these being handled according to the old policy; (b) to prepare a great defensive system, Metz - Lower NiedSaar - Vosges - Strassburg, against the contingency of an immediate powerful offensive of the French; (c) in the latter event, to suspend the offensive initiated in accordance with (a), and to draw back from line to line till the Saar was reached, where a standing defence would be made in conjunction with counter-attack on the French left from the Metz - Nied system. In the least likely alternative of only minor forces of the French being met with in Lorraine, Prince Rupprecht might utilize the Metz system to cover the transfer of part of his forces W. of the Moselle for cooperation with the V. Arm y. Upper Alsace was to be held against light forces, but evacuated step by step if seriously invaded. The VI. Army with the 3rd Cavalry Command was assembled accordingly in Lorraine, the VII. Army (minus two expected Italian cavalry divisions) in Alsace; and mobilized and civil labour on a large scale was employed in creating the " Nied " position, which ran from where the newest Metz works touched that river to its mouth N.W. of Saarouis.

The French I. and II. Armies, ready for operations as soon as the Germans, spared the latter the necessity of testing their intentions by advancing on Aug. 14 in great force. Accordingly, the German VI. Army drew back from position to position, while the VII. Army, which had defeated the first inroad of the French VII. Corps at Miilhausen, likewise drew back gradually before the army of Pau, dispatching part of its forces by Zabern to the upper Saar to assist the VI. Army's defence, and preparing with the remainder to hold the line Schirmeck - Molsheim - Strassburg firmly. But this withdrawal, after making all preparations for the offensive in the first place, was only unwillingly and half-heartedly carried out by Rupprecht's headquarters. And when, on Aug. 17, news arrived from Moltke that the French advance into Lorraine was, after all, not their main offensive, Rupprecht determined to fight on the line of his rear-guards on Aug. 20. Thus, just at the moment when the French had decided to push their offensive without further hesitation, the Germans changed their policy into one of flank attack. The battles of Morhange (MOrchingen) and Saarburg followed (see Frontiers,Battles Of The: section Lorraine), in which the Germans were victorious.

A hot pursuit was initiated, and with that disappeared for the time being all possibility of drawing off any part of Prince Rupprecht's forces for the benefit of the " Army." Moltke's rolling-stock on the Rhine waited in vain. The Commander-in-Chief in the Rhine lands had brilliantly performed both his tasks, but he had become a commander of two ardent armies who could see only the enemy in front of him.

The German Advance through Belgium and France

Once the Liege foreground had been secured, and the pivot of ThionvilleMetz been made firm, there was no need, in Schlieffen's opinion, for overhaste in the development of the main operation, which required principally power. Moltke, on the other hand, sought to remedy the reduction of power on the striking wing by urging it to speed. This was not easy, for the administrative and technical marching arrangements required for drawing the five or six corps of the I. Army through the narrow tunnel between Liege and the Dutch frontier were complicated, while Billow's II. Army displayed no great energy in crossing the Meuse above Liege. Thus the Belgian main army was left undisturbed on the Geete - the most forward position which it could safely adopt in view of the risk of being outflanked by a German movement through Dutch territory - till Aug. 19, when, threatened with severance from Antwerp, it began to retire westward with but little fighting. Even then, however, the task of the German I. Army was by no means clear. The possible arrival of a British Expeditionary Force on its outer flank was a standing menace, and until the true right flank of the system, Billow's II. Army, had begun its advance in earnest its protective echelon (I. Army) could not operate to any purpose. On the evening of Aug. 19 - the 18th day of mobilization - the outer flank of the wheel had only reached the line Thildonck - Tirlemont (I. Army) - Sart Ribsart - Mehaigne (II. Army), while the armies traversing the Ardennes were at Marche (III.) and Bastogne - Arlon (IV.) respectively. The V. Army was just beginning to draw out of the Thionville area towards Longwy. On Aug. 20, learning that a large French army had been assembled in the Charleroi region, Billow prepared its encirclement by beginning to wheel-in not only the II. Army's right wing but the I. Army (then under his orders) as well; but as the cooperation of the III. Army moving on Dinant was intended, that army, as well as the I., had to be given time, and the concentric blow upon Lanrezac was fixed for Aug. 23, while the left wing of the II. Army was to attack Namur from both banks of the Meuse. Had these measures been carried. out, the " vast wheeling movement" would, on the 22nd day of mobilization, have occupied the line Nivelles - Charleroi - DinantNeufchateau - Arlon - Thionville instead of the line Ghent - E. front of Maubeuge - Sedan - Thionville, as Schlieffen had planned for that day. But the fact of being a few days in arrear was of small importance relatively to the fact that, instead of continuing to swing out, the German line would be beginning already to roll in. The essential principle of the movement, that of outflanking not only the momentary position, but also any and every position of the enemy's left wing, was sacrificed. The British Expeditionary Force, unlocated. lay outside instead of inside the scope of the wheel.

But during Aug. 20-23 events crowded upon one another at almost every point of the German line. The left wing of the II. Army stormed Namur, with the powerful aid of the super-heavy artillery (this time deployed from the outset), while the right became engaged front-to-front with Lanrezac (see Frontiers, Battles Of The: section Charleroi). The I. Army, advancing in echelon to the battlefield indicated by Billow, but seeking still to preserve both its rearward echelonment and its power of extending outwards, came into contact, with its left wing, with the British Expeditionary Force in the neighbourhood of Mons (see Frontiers, Battles Of The: section Mons). Meanwhile, the German III. Army, drawing out of the W.S.W. direction in order to cooperate in the battle proposed for Aug. 23, had lost touch with the situation of Billow's army (apparently through the neglect of the latter to keep Hausen posted as to events on the Sambre), and the IV. Army in the heart of the Ardennes had already begun to wheel forward and southward in order to protect Hausen's left when it received the full weight of the French IV. Army's offensive from the Chiers. Lastly, the V. Army, on which a defensive attitude was at first imposed by the Supreme Command, managed to convert its defensive, flushed by success in the battle of Longwy, into a flank offensive which threatened to create a gap between the five armies and the Thionville-Metz pivot, on which they were to wheel. (See Frontiers, Battles Of The: section Ardennes.) Thus, at almost every point, local situations and the initiative of army commanders and troops turned the smooth and regular tide into a series of eddies. On the German side, as on that of the Allies, the northern half of the battle of the Frontiers was a chain of fierce local battles which only a very strong Higher Command could take in hand, either to straighten the links or to reforge them in a different pattern.

From that point to the battle of the Marne, the contest is less between schemes, less even between armies, than between the capacities of the two Supreme Commands.

At the head of the German armies was a man in indifferent health, by nature kindly rather than insistent, one for whom responsibility was rather a burden than a source of strength. He carried, moreover, the strain of watching and attempting to direct affairs on the eastern front. His opponent was a different man and differently placed. Essentially authoritative in temperament, sound in health, and concerned with one task only, Joffre was a commander in the full sense where Moltke was a responsible adviser only. Leaving the details of planning to his staff, and in particular to General Berthelot, Joffre devoted himself entirely to the role of commanding. His personal activity in the days after the battle of the Frontiers is astonishing, but it is essentially of the moral and not the operative kind - deciding, encouraging, punishing - the role of King William I. in the war of 1870-I, and one which the Emperor William II. was unable to sustain in 1914. Add to the factors weighing against Moltke, the prestige and confidence of his army commanders, most of whom had won victories, none sustained defeat, whereas no French subordinate general had obtained an important tactical success, and it is evident that the higher control was necessarily firmer on the side of the French than on that of the Germans.

The prevailing impression on the German side after the battles along the frontiers was that a decisive victory had been won, and that the next phase was to be one of exploitation. The consequences of this impression, which soon penetrated to General Headquarters, were: (a) the decision to send six corps (two from each portion of the line) to the East Prussian front; (b) freedom of action granted to the V. Army to cut loose from contact with Thionville and join in the general pursuit by a movement round the N. of Verdun; (c) noninterference with Prince Rupprecht's pursuit in Lorraine; and (d) a new orientation of the I., II., III., IV. and V. Armies, which, abandoning the " wheel," were to advance in line in a general S.W. direction, with the I. Army heading for the lower Seine, the II. for Paris, the III. for Château-Thierry, the IV. for Epernay, and the V. for Vitry-le-Francois, the last-named flankguarding against Verdun and the first preserving a defensive echelon on its right.

The new orders were issued on Aug. 27, after the battle of Landrecies-Le Cateau had accounted for all undisclosed British forces and established the feebleness of the French cordon to the left of them, while the III. Army was well S. of Rocroi and the IV. bordered the Meuse from Sedan to Stenay. They were not executed with the certainty and confidence of an exploitation. The V. Army, in the act of letting go its connexion with Thionville-Metz, had on Aug. 25 exposed its left flank to a very sharp offensive of the defensive portion of the French III. Army now styled " Army of Lorraine," under General Maunoury, whose progress had been stopped only by orders from Paris. The French III. and IV. Armies, quickly rallied from their Ardennes-Longwy defeats, gave ground only slowly and with frequent counter-strokes (see Frontiers, Battles Of The: section Ardennes), and the German III. Army was continually drawn southward, off its line of advance, to assist the IV. in real or supposed crises. Thus a great gap opened up between the III. Army and the II.; and the latter, uneasy as to its left flank, gradually drew away into a southerly direction, while Kluck's I. Army continued S.W. on Amiens. Almost immediately thereafter, the I. Army began to come into contact with French forces, distinctly superior in number and in quality from those hitherto met on this outer flank. While driving these back in various minor actions, and expanding ever westward in so doing, it was suddenly checked and caused to swerve southward by demands for assistance from the II. Army, which, unsupported on its left by the III., found itself counter-attacked, with a vigour that had not been observed since Charleroi, by Lanrezac (see Guise, Battle Of). This crisis, like similar crises on a smaller scale in the area affecting the IV. and III. Armies, passed away after a time, but the disintegration of the German mass-movement had now reached a climax. Apart from regulating special questions between armies as they arose, the German Higher Command had not intervened in the conduct of operations since its instruction of Aug. 27. On the night of Aug. 31, the I. Army, in the vain hope of seizing the left flank of the British or of Lanrezac, had pushed its left far to the S. to the Aisne below Soissons, while its right was in the Lassigny hills and W. of Montdidier and even farther north. The II. Army had not progressed beyond the Guise-St. Quentin battlefield, its front facing due S. between Essigny-le-Grand and Vervins; the III. Army on the contrary was on the upper Aisne on both sides of Rethel, the IV. astride the northern Argonne between Semuy and Buzancy, the V. wrapping itself round the N. side of Verdun while still maintaining considerable forces in the Woevre facing the E. front of that fortress, and the Cote de Meuse. Two corps had left for the eastern front, belonging not to the subsidiary armies in Lorraine but to the striking wing. One corps had been left to face the Belgians in Antwerp, one and a quarter corps to besiege Maubeuge, other detachments here and there to guard lines of communication or to invest small French forts such as Givet. The only new forces on their way to the West were the two divisions of the IX. Res. Corps and certain Ersatz brigades, all of which were needed to support the dangerously weak cordon of the III. Res. Corps facing Antwerp and to be on the spot in case of a Russian and British landing at Ostend, rumours of which at this time filled western Europe.

In Lorraine, the pursuit from the battlefields of Morhange and Saarburg had led the German VI. and VII. Armies in a southerly direction, substantially on Rambervillers and Charmes. Forced to condense into two main groups by the fort of Manonviller - a work condemned as useless by peace-time critics - and by the forest of Parroy, they had exposed their right flank to counter-attack by the restored army of Castelnau, which held the fortifications N. and E. of Nancy and the north flank of the so-called Trouee de Charmes. This French counter-stroke not only gravely imperilled Prince Rupprecht's army - for Manonviller resisted long enough to act as an anvil to Castelnau's hammer - but deprived the German Command in Lorraine of its initiative. With that loss, it forfeited all real power of holding larger French forces in its front; and though the German Supreme Command, in the same confident general instructions of Aug. 27 which initiated the southwestward pursuit of the I.-V. Armies, ordered the Bavarian Prince to break through the French line in the direction of Neufchateau on the upper Meuse, it soon appeared that Joffre had the " inner line." He could take troops from Lorraine for other service, while his opponent could only continue costly holding attacks that did not hold.

On Sept. I the German Supreme Command gave up the conception of a general southwesterly pursuit, which, by its incidents, had not only lost its direction but brought the armies into a very irregular array and resumed the original conception of the wheel pivoting on Thionville, or rather, in the new situation, on the troops of the V. Army facing the N. side of Verdun. By now, however, with losses and detachments, the frontage of such a sweep was reduced by the front of a whole army, if not two armies. The appearance of French active and reserve forces N. of Paris made it clear that a protective echelon - such as had always been prescribed and rarely formed by the I. Army - would have to follow the rear of the army on the outer flank, and, moreover, the gap between the II. and III. Armies must be closed. The new general instructions, therefore, prescribed that the II. Army should steadily drive the French in a southeasterly direction, followed in echelon by the I., which was to watch Paris and break up the communications leading. thither. But almost immediately after the I. Army, still well in advance of the II., received this order, one of its corps, exploiting a local advantage, crossed the Marne at Chezy and Château-Thierry, and Kluck determined to support it rather than withdraw it. The Supreme Command made no protest, all the more so as he reported evidences of real dissolution in the ranks of the retreating enemy. Kluck pushed on. The echelon facing Paris was maintained, but it was growing thinner and thinner. On Sept. 4 the Supreme Command, in increasing uneasiness, limited the offensive front still more. Not only was the I. Army to stand fast between the Oise and the Marne, but the II. Army was to wheel outwards and fill the space between the Marne and the upper Seine. The III. Army, now become the operating wing, was to march on Troyes and to the E. thereof, continuing in close liaison with the IV. and V. while the Lorraine armies were to renew their attempt to break through the upper Moselle front.

The final phases of the battle in Lorraine represent the endeavour of exhausted forces to carry out their part in this scheme. The central and western portion came to nothing, for although Kluck began at 23: oo hours (II P.m.) on Sept. 5 to counter-march his army so as to fill the space between the Oise and the Marne, now guarded only by the last relic of his echelon, and Bulow gained ground between the Marne and the Seine as far as Montmirail and the marshes of St. Goud, General Joffre had, on the afternoon of Sept. 4, issued the command to his armies to face about and attack.

The Preparation of the Counter-Offensive

While on the German side we see the battles of the frontiers followed by a high spirited chase in which the driver was able to keep little more than the semblance of order in his team, on the French side the picture is one of an astonished and confused, but in no sense a routed citizen-army, too clear-sighted to believe itself betrayed and yet too ignorant of the ensemble to see where miscalculations had led to disaster. In the hands of one who had specialized in the art of inspiring confidence, whose silence even was imperturbable, and whose career had been spent not in technical subtleties of operations but in varied branches of administration, it had every chance of early recovery, provided that it was handled according to a definite policy and not exposed to incidents. This definite policy was laid down in a general order of Aug. 25, which began with the phrase: " As it has proved impossible to carry out the projected offensive, the next operations will be regulated so as to constitute on the left, - by means of the IV. and V. Armies, the British army, and new forces from the eastern front, - a mass capable of resuming the offensive while the others hold up the enemy for the required time." Following this cool and convincing statement, the detail paragraphs prescribe a retreat to the line Braye-sur-Somme-Ham for the British, to Vermand-Moy for the offensive portion of the V. Army, to La Fere-Laon-Craonne-St. Erme for the defensive wing of that army, to the middle Aisne for the IV. Army and to the ArgonneVerdun line for the III. " From this situation the offensive will be resumed," said the order. On the left of this line a barrage against cavalry inroads was to be formed between Picquigny and the sea, and either between Domart and Corbie or between Picquigny and Villers-Bretonneux there was to be formed a new army, soon to be designated the VI., and composed, as regards its staff and several of its divisions, of the troops which had just defeated the left of the German Crown Prince as it sidled past Etain. This was to be ready for action on Sept. 2, and the direction of its offensive would be either St. Pol-Arras or ArrasBapaume, according to the position of the extreme right of the Germans. The British would attack on Bertincourt or Le Catelet, according to the situation, the V. Army wing on Bohain. The right of the V., the IV., and the III. were to defend the line laid down and eventually to attack from it. To the I. and II. Armies went the laconic instruction, " the role of these armies is, to endure." But the pressure of the German pursuit in its first freshness did not admit of the British coming to a halt on the line ordered, and when the elements of the new VI. Army began to assemble about Amiens the battle had passed far to the south of them. Similarly, with the V. Army, the battle of Guise, which may be considered as a section of the proposed general offensive, led to small results because the British element was wanting on its outer flank. The controversies which have arisen as to the rapidity of the British retirement from the battlefield of Le Cateau to behind the Oise need not here be discussed, for it is more than doubtful in any case whether the state of the French army, in its ensemble on Aug. 31, justified the risk of incurring final defeat. Be this as it may, Joffre put aside all temptations to exploit the local successes at Guise on the Meuse, and in a new general order of Sept. r laid down that the VI. Army and the British having insufficiently checked the enemy's turning movement, the whole system must pivot about its right continuing to retreat, until the left of the V. Army should be free from the menace of envelopment. Then the armies would take the offensive, this time utilizing the position of the III. Army, protected by Verdun on the N. and the Cote de Meuse on the E., to strike the chief blow. The position from which the offensive would be resumed was now well back from Paris, which was to be left to itself (though Joffre suggested that its troops might cooperate in the general offensive), the V. Army behind the upper Seine (Nogent), the detachment Foch (IX. Army) and the IV. Army behind the Aube and S. of Vitry, the III. Army, augmented by defence troops borrowed from the Cote de Meuse and possibly by troops from the Lorraine front, N. of Bar-le-Duc. The British and the VI. Army were to constitute with Gallieni's local troops a Paris group which should hold the Seine from Melun to Juvisy, and the E. and N.E. fronts of Paris.

Thus was prepared the initial situation of the battle of the Marne. The scheme as outlined at first underwent many modifications, due to the ardent initiatives of Gallieni in Paris, and of Sarrail, commander of the III. Army, S. of Verdun, as well as to other causes. These are discussed in the article Marne, Battle Of The. Here it is not necessary to analyse too closely the form projected for the battle. Essentially, the fact to be retained by history is that a great army, in retreat after failure, could be energized, ordered to turn about, and launched to the attack, by a modern commander-in-chief whose influence must filter through a complex hierarchy before reaching the fighting soldier. Many had believed this to be an impossibility, and they were proved wrong. The operative scheme of the battle of the Marne and even its apparent barrenness of specific military results, are of insignificant importance compared with the fact that the battle of the Marne was actually fought. (X.) The " Race to the Sea." - The establishment of the German defensive on the line of the Aisne, prolonged across the plain of Champagne, which ended the Marne battle, did not put an end to the Anglo-French offensive. The front between Compiegne and Verdun was stabilized here and there, but the battle of movement continued at the free extremity, that is, to the W. of Compiegne, and beyond. This new offensive has improperly been called the " Race to the Sea." In reality it was not a question of reaching the coast as quickly as possible, so as to obtain there an absolute protection against turning movements. If it had been so, the shortest line for the Allies, and the easiest to hold, would have been that of the Somme, from Compiegne to Montdidier and Amiens. To the estuary of the Somme, this line does not measure much more than zoo km., while the line from S. to N., which was that of the actual " Race to the Sea," ended N. of the Yser and was nearly double that length, presenting features of very various nature, among which some entirely lacked defensive value.

The truth is that the offensive, which was throughout the policy of the French Command, did not stop at the Marne victory. On Sept. I r, when the VI. Army (Maunoury) arrived at Compiegne, the Commander-in-Chief gave the order for this force to place immediately as many troops as possible on the right bank of the Oise. On Sept. 17 he indicated his plan by ordering the formation, on the left wing, of a force capable of parrying a flanking movement by the enemy as the best precaution to be taken. But Sir John French has stated that on the very next day (Sept. 18) General Joffre informed him that he was developing a new plan which aimed at attacking and enveloping the German right flank. The enemy, moreover, showed by his method of occupying the ground that the initiative no longer rested with him.

At the extreme end of the Allied line on the Oise the valley of the Aisne cleaves its way through, a forest-clad massif, cutting it in two S. of the Aisne; the larger part of this mass consists of the forest of Compiegne and the northern part is the forest of Laigne. If the Germans, very skilled in turning forests to military account and manoeuvring in them, had retained any hope of resuming the offensive against the Allied left flank, they would have occupied the forest of Compiegne in order to make it the starting point of their turning movement and force the Allied left to retreat towards Paris. But they abandoned that front; nor did they retain the forest of Laigne, on the plateau to the N. of Attichy. The offensive impulse in the World War was thus on the side of the Allies and they kept it until the fatal day of Russia's defeat on the eastern front. Until that time every attack which resulted in the gain of ground came from the Allies, who, save for a few occasions, methodically pushed back the enemy from one entrenched position to another. The Germans had later to defend their right flank, more and more threatened as it became more and more prolonged. And, as the best way of arresting the progress of the Allies would have been to strike at their offensive, they tried without ceasing to outflank them, while they were resisting in front.

Thus the " Race to the Sea," viewed as a whole, consisted in establishing an offensive Allied flank against the German right, and, this flank being always unsupported at one extremity, in German efforts to seize this extremity from two directions. These attempts, on either side, taking place farther and farther from Compiegne, appeared as a " Race to the Sea "; but in reality neither side was deliberately making for the coast. The S. - N. direction taken by the line of contact was not sought by either opponent; it resulted from the balance of forces.

It is evident that, whatever resources on either side were furnished by reserves and new formations, the extension of the front of contact over a length of 200 km. was only possible on condition of leaving much thinner forces on the stabilized front than had been required for the battle of movement, from Aug. 20.to Sept. 12. In proportion as the lines extended to the N., transferences took place, depleting the line from Belfort to Compiegne, as had already been done in the Vosges and in Lorraine, to enforce the regions where the struggle was being carried on in open country.

As regards such transferences, the advantage was with the Germans, because they occupied the interior of the angle whose apex points to Compiegne, while the Allied troops had to be moved around this point.

On the German side, the manoeuvre brought 18 new army corps into the line. On the Allied side it resulted in the transference of the II. Army (de Castelnau) between the Oise and the Somme; the formation of the X. Army (Maud'huy), N. of the Somme, and the VIII. Army (d'Urbal), which included the Belgian army, brought back from Antwerp; and lastly in the shifting of the British army into Flanders. To these transfers and new creations must be added various formations constituted on the spot, of which the most important was a group of territorial divisions placed under the orders of General Brugere. Some of these formations had already joined in the offensive after the battle of the Marne, notably at Amiens, which they had cleared of the German units scattered through Picardy and Artois. This ensemble was placed under the command of General Foch, but was only brought to completion by degrees. The successive steps will now be described.

The first offensive action against the German right began on Sept. in accordance with the order given by General Joffre. It was carried out by the VI. Army, with one additional corps on the right bank of the Oise. It immediately encountered energetic resistance on the Aronde, a small tributary prolonging the depression of the Aisne valley on the opposite side of the Oise. On the northern bank of this stream wooded heights extend between Compiegne, Lassigny and Noyon, and surround Ribecourt - names which all became famous during the war.

The VI. Army had a difficult task in the subduing of them, and could hardly have achieved it, threatened as it was with an attack in flank, without the help of the II. Army (de Castelnau), which, detraining in the Clermont - Beauvais area, had to cover 30 or 40 km. in order to outflank the German right. At the same time the cavalry and the territorial divisions of General Brugere extended the movement towards the north.

The resulting battles were prolonged until the end of Sept. with alternations of success and reverse, through Lassigny, Roye, and Chaulnes, as far as Peronne. From Peronne to Lassigny, where the wooded hills ceased, the terrain consisted of undulating plains, where no line of battle could be found. It was thus the balance of forces that determined the front of contact, which was gradually fortified on either side.

On Sept. 24 the French retook Peronne and lost it again. This little place, in a hollow, offering no possibility of outlook or of action outside its walls, had no military value. The positions which should have been occupied in the circumstances were the heights of the Somme above Peronne which formed a very considerable obstacle. The great value of this line, especially facing eastward, owing to the command of the country in that direction given by the heights, did not escape the German staff, and for a long time their efforts were directed towards preventing the Allies from securing the heights, by the defence of improvised fortifications, at some distance in front of them. Later it will be seen how these lines were linked up with that of the Ancre.

It will now be shown how, after the preliminary fighting, the German front became established between the Aisne and the Oise. A salient is always a weakness. The front was fixed from E. to W. along the Aisne; and the flank resolving itself into a new front running N. and S. the German line exhibited a right-angle salient pointing towards the forest of Laigne. It was very largely to smooth out this salient that the line was traced behind the forest in such a way as to form a great arc instead of a point. South of the portion of the line of the Somme between Peronne and Ham it was necessary to connect this arc with the fortified line S. of Peronne; the line Chaulnes - Roye - Lassigny was thus strongly indicated as the connecting line.

On Sept. 23 there was fighting near Lassigny, on Sept. 25 near Roye, and on Sept. 29 in the same places and also at Chaulnes, where the Allies were repulsed, as they were at the salient itself. On Oct. i the fighting-line extended to near Arras. It will be seen how it came to be fixed on the N. of Peronne.

The old fortress of Arras, which was no longer more than half fortified, but whose citadel had been maintained in good condition, was a point d'appui for the Allies. The interval, 40 km. wide, between Arras and the Somme, provided some features which were favourable to the establishment of a line. First, near Arras, there was the little valley of the Crinchon. The stream itself is unimportant, but its banks afford positions which are good in default of better ones. Next, a connecting line had to be ensured over about 15 km., from N. to S., across the undulating plateau, to the course of the Ancre, which forms a deep ravine both above and below Albert. The river bends in a S.W. direction as far as Corbie on the Somme.

Thus the line traced by the depression favourable to a line of resistance forms a series of zigzags: the Ancre near Albert, the Somme from Corbie to Peronne, the Somme from Peronne to Ham. It was because of this peculiarity that the Germans, when they were defending the line of the Ancre, opposite Albert, could not make use of the section of the Somme between Peronne and Ham, because it was 20 km. to the rear. Their solution consisted therefore of drawing a line through the Corbie salient, behind which lay Combles, transformed into a magazine and supply depot. This brought them in front of the section of the Somme between Peronne and Ham.

During the " Race to the Sea" the fighting round Albert and Arras (see Artois, Battles In) began at the end of Sept. and on Oct. I respectively. The X. French Army then came into the line. The Germans were already strongly entrenched on the Thiepval plateau, opposite Albert, where they were to pile up the defensive works which the British were to capture, one by one, two years later.

On Oct. 2 -3 the X. Army (Maud'huy) made an effort to seize the German flank at its northern extremity by moving forward to Douai, where there had been for some time a detachment of the territorial army, which did not succeed in maintaining its hold. But the Germans opposed with heavy forces of infantry and cavalry. The French were pressed back; the enemy occupied Lens and made a vigorous but unsuccessful attack on Arras. For a long time afterwards the Allies' line left Lens in possession of the Germans, lying farther westward on a line from Arras towards Bethune. Arras was included indeed, but the important positions of Vimy and Notre Dame de Lorette, of which the Germans would not lose hold, were cut off. Their importance consisted in their facing N.E. and later their capture necessitated long and painful effort.

The German cavalry was in all the country which lay beyond in the N.; it occupied Ypres and Bailleul, and sent out patrols still farther forward. The situation was very difficult for the Allies, and its improvement was an urgent matter. The VIII. Army (d'Urbal) had its base at Dunkirk, but it was still inadequately constituted. The French cavalry and General Brugere's territorial divisions were maintaining an arduous and very fatiguing struggle. At that moment the British army, which could now be withdrawn from the Aisne, as that front was strong and solid and could be held by a small force, was summoned N. Reinforcements were also brought from the eastern area, and lastly the Belgian army, no longer of any use in Antwerp, came to take its place in the ensemble. The story of the siege of Antwerp and of the escape of the Belgian field army from the place is told under Antwerp.

At the time of the fall of Antwerp Sir John French and General Joffre met at Doullens, on Oct. 8. The British II. Corps arrived on Oct. 12 near Bethune and Aire, and was able to attack the enemy's flank in a combined operation with the X. French Army. The III. Corps detrained at St. Omer on Oct. 12; the I. followed. Finally the IV. Corps, supporting the Belgian army in its retreat, would rejoin the rest.

It is evident that if the organization and transport of the new army corps which were to swell the forces of Sir John French had been accomplished more quickly, and the N. of France occupied earlier, that region would have been more easily freed from the enemy invasion, and the blow at his flank aimed at by the Command would have resulted instead of the " Race to the Sea." But it must not be forgotten that Great Britain created this new army ab ovo - a stupendous military effort - and that armies cannot be improvised. Until about Oct. 8, the Germans had only cavalry in the regions around to the north of Lille. It was then strengthened by the arrival of a strong army corps, which came from the neighbourhood of Antwerp by way of Courtrai.

The Lille Question

On Oct. 8 the enemy arrived before Lille. The fortress closed its gates and resisted with the few troops that it possessed. The Germans did not trouble to attack it in earnest. They contented themselves with bombarding the town from a distance and demolished about a thousand houses. On Oct. 13, after five or six days of bombardment, Lille yielded. This short resistance was not useless, for it enabled the British army to take up its position behind Lille and to consolidate it. The question of Lille gave rise to much discussion in France. The arguments cannot be examined here, but a few words must nevertheless be said on the subject. If it had held out longer, it might have formed part of the battleline, like Reims, in which case it would inevitably have been destroyed, and with it the industrial towns of Roubaix and Tourcoing. Should the sacrifice have been made ? The British front would have gained about 15 km. on the E. but the general line would have passed along the Yser on the N. and through Arras on the S. just the same. It would, moreover, have been difficult to find a good connecting line to link up Lille and Arras, across country very illadapted for defence, especially in the southern half. If the defence of Lille had been included in the general scheme of operations from the beginning, a very strong garrison must have been left there, proportional to the extent and number of works of the entrenched camp. It would have had to be an immobilized force whose task was to immobilize an enemy force also, as at Maubeuge. But it would not, any more than Maubeuge, have hindered the enemy's turning movement. Thus, to assume the resistance of Lille, until the arrival of the British army at the beginning of Oct., is to assume that it could have sustained a siege and bombardment from the last week in Aug. onwards - that is, for over 40 days. The fate of other isolated fortresses, such as Liege and Antwerp, leaves no doubt that this would have been impossible.

On Oct. 12 a junction was effected between the Belgian army, retreating towards Ypres with the French marine fusiliers, and the II. British Corps (General Smith-Dorrien) in the region between Bethune and Bailleul, along the canal and the Lys, while the III. Corps (General Pulteney) marched on the left of the II. towards the line running from Armentieres to Ypres, to occupy the heights of Messines, later so fiercely disputed. On Oct. 13 the British cavalry, which held the left of the advance, took the heights which stretch between Cassel and Messines (Mont Noir, Mont des Cots, etc.) and which are of great military importance. On Oct. 14 the II. Corps entered Bailleul. The same day General Byng's cavalry division arrived at Ypres, with a French territorial division.

Definition of the Front of Contact. - Although the struggle never ceased and a great German offensive was immediately prepared in Belgium against the left of the line, at Ypres, on the Yser, the " Race to the Sea " must be considered to have ended on Oct. 15 1914, the date on which the Allies occupied the whole front from Ypres to the sea. At that date the left, on the Yser, was at Nieuport, and possessed a bridgehead at Lombartzyde in advance of the line of the river and near its mouth. The Yser front passed through Dixmude. To the S. of that town it is delimited partly by the course of the river, which spreads out like a fan a little farther to the S., partly by the Ypres canal, after passing the ancient fort of Krocke. Then it goes beyond the canal to the villages situated farther E. on the stream of the St. Jean, in order to arrive at the heights which surround the basin of the Yser and its tributaries, reaching them at Gheluvelt, on the road from Ypres to Menin. It held the crest and the heights as far as Messines. The French were on the left, from Nieuport to Dixmude; the English on the right, on either side of Ypres: the Belgians were between the two, their feeble effectives occupying only a small portion of the front.

From the beginning the Yser front was well selected and lent itself well to defence, not so much owing to the obstacle formed by the Yser, as because the country behind, much cut off by water-courses, was very ill-adapted for an offensive by large masses. At the same time, this part of the front, which was entirely defensive, had not its full value until it was flooded. The dykes were burst open on Oct. 22, the floods then gained ground little by little as they rose higher. The effect was not obtained very rapidly, because the whole volume of water in the basin is not great. It did not hinder the enemy from crossing the river below Dixmude on Oct. 26, and penetrating thence to a depth of 4 km. till he was stopped by the lines in the rear, and especially by the railway embankment. But two days later the. Germans were forced to retreat, partly by the Allied counterattacks, partly by the water, which was spreading and rising. The part of the line which passed in front of the Ypres canal near the little river St. Jean could not be made very solid, but it was necessary in order to form a link with the Gheluvelt heights. But the true line of battle, which overlooked a large stretch of country to the E., was only attained in the last phase of the war, with the capture of the entire line of heights in the direction of Staden, at least as far as Westroosebeke, the point where the ridge is crossed by the high road from Ypres to Ghent. Also, during the whole of the first period of the struggle, the British army was always very vigorously .attacked around Ypres and at Ypres itself. The most important point d'appui was the portion of the heights lying directly to the S. of Ypres, Wytschaete and Messines, - that is, the eastern end of the line of the Monts de Flandre, which run from Messines to Cassel and appear again farther on to the N. of St. Omer. These hills are of immense importance in the whole defence of Flanders. This was seen in the last phase of the war, when they fell into the hands of the enemy as far as Bailleul, and it took very hard fighting and the greatest energy to arrest his progress in this direction and to stave off still more fatal consequences.

Between the promontory of Messines and the cliff formed by the heights of Notre Dame de Lorette and of Vimy, above the plain of Lens, stretched a sector with a front of nearly so km. which afforded no continuous line of any length, and included only isolated strong-points, sometimes in a position favourable to the Allies, sometimes to the enemy. Those most useful to the Allies were the ridges of Messines and Vimy. The first was taken by the Allies on Oct. 15 1914; the second was at first occupied by the French, and later their success was repeated by the British. The French front line passed 2 km. to the west of Notre Dame de Lorette, the key to the position. The attack on this point began on Oct. 20; but the conquest of the whole ridge necessitated efforts which were renewed up to the middle of April 1917.

In the plain, about midway between Messines and Arras, is a point relatively stronger than the rest, owing to the canals and marshes which protect its approaches. This is the little town of La Bassee, occupied and strongly fortified by the Germans; and for a long time it delayed the progress of the British army. The difficulty of the advance towards Vimy was very largely owing to the part played by La Bassee in preventing the British in the plain of Lens from taking part in it, in striking at the rear of all their attempts in the neighbourhood of Lens, especially those which aimed at outflanking the town on the north. But hard as were the local struggles in the MessinesVimy sector, no great action ever developed there until 1918: first, because as regards an Allied action a great offensive in the direction of Lille would have been very risky unless it was led up to by a more strongly protected movement either to the S. in the region of Arras, Peronne and Ham, or in the N. towards Belgium; secondly, because as regards a German action, the sector to be conquered to carry out a flank offensive against the Allied left was that of Ypres and the Monts de Flandre.

Such was the real meaning of the German project Nach Calais!, which cost the army of William II: so dearly. For the Germans, to engage themselves in the Messines - Vimy sector, in order to reach St. Omer, would have been to run into the j aws of a pair of pinchers formed on the N. by the Monts de Flandre and on the S. by the ridge of Vimy - Notre Dame de Lorette, extending to the S. of Bethune and Brie. It would have meant the formation of a salient, which would be more vulnerable the farther it was pushed in the direction of Calais. The result of all this was that almost up to the end of the war the front passing through Armentieres and Arras underwent very little change.

It has already been related how the front was determined first between Arras and Albert, then S. of Albert in front of the Somme between Peronne and Ham, and lastly as far as the salient opposite the forest of Laigne. The attack on this very strongly fortified part of the German front was only begun in the early part of Dec. 1914.

At that time the French still held Thiepval wood on the enemy slope of the Ancre. Farther S., some fluctuations occurred on the plateau, in the region of Chaulnes and Roye. But there was no great action other than the Anglo-French general offensive, which developed between July 1916 and March 1917.

All this leads to the conclusion that the last half of Oct. 1914 may be considered to mark the end of the " Race to the Sea," and the establishment of equilibrium, except for local fluctuations, along the whole length of the immense line of contact. (H. BE.) Trench Warfare, 1 915 - 17 While weary British troops were handing over their lines in the salient to their French comrades at the close of the first battle of Ypres (see Ypres And The Yser, Battles Of) and they had time to think of other things than the grim struggle that had just ended, it dawned upon them that the war in the west had entered upon a new phase. The trench barrier had been completed from the North Sea to the Swiss frontier, and a war of movement and manoeuvre, for which most of them had been trained, had become impossible. This was a development which had not been foreseen by the military world of Europe and it took it by surprise. For the Germans this surprise was less unpleasant in the winter of 1914-15 than for the French and the British. When their second great effort to win the war in the west had failed they had decided to adopt a defensive policy in France and Belgium while they attacked in the east. Therefore they could regard the difficulties of attack upon trench lines and the restrictions of manoeuvre with some complacency, while they were better prepared for the new type of warfare than were either of their opponents, more particularly the British. The fortifications of the French and Russian frontiers had compelled the Germans to study closely the art and science of siege warfare and to make preparations for such warfare. They had assimilated the experiences of the Russo-Japanese War and had learnt from them to their profit. Therefore trench mortars, hand and rifle grenades, searchlights and pistol lights, and the possibilities of mining, were to them no novelty; and having the plant ready for the manufacture of the material required for siege warfare, and considerable stocks of such material already in store, it was not difficult for them to meet quickly the demands made by their soldiers in the trenches on the western front.

The British army was in very different case. Its chiefs before the war believed that it would be employed only in a war of movement, and the study of siege warfare had been confined to a small body of engineers, amongst whom it was largely theoretical. All the money to be found for the Army Votes had been used in the training and equipment of the Expeditionary Force, and there was none left for the provision of the material required for the attack of fortresses. The cost even of hand grenades, with which experiments had been made when the reports of British observers in Manchuria came home, was held to be prohibitive. For these reasons the winter of 1914-15 was for the British infantry in France and Flanders a period of unmitigated suffering. The ebb and flow of attack and defence had left the Germans almost everywhere along the British front in possession of the higher and drier ground, while in the low-lands of Flanders water was found at a few feet and even sometimes a few inches below the surface; under incessant rain, parapets melted away unless held up by sand-bags constantly replaced, but the British army had not sand-bags in sufficient quantities. The construction of communication trenches was all but impossible, and material had to be brought up to the front lines by parties floundering through the mud at night. Lying in sodden trenches, before an enemy who possessed weapons which they had not, the British infantry endured at this time a longer and in many respects a severer test of their constancy than in some of the worst crises of the war. The strain upon the French was less severe because a small part only of the French army had to endure the incomparable mud of Flanders, and from the first it was better provided with H.E. shell, while the greater military resources of France made it possible to meet the new conditions more readily.

1915

While the soldiers in front were enduring and slowly learning to mitigate the horrors of trench warfare, those behind were planning. There were not lacking active brains in Paris and in London who saw that the assaults upon the ever-widening barriers of barbed wire and the evermore serried lines of trenches must prove a slow and bloody business. These brains sought eagerly for a way round the barrier which would lead to a speedier and less costly victory, and so began the controversy between " Easterners " and " Westerners " which endured while the war lasted. There were signs early in 1915 of preparation for an Austro-German attack upon Serbia; and Mr. Lloyd George, inspired by his friends in France, proposed at once to save a weak Ally and to attack the weakest link in the opposing chain by transferring the bulk of the British army to the Balkans and reinforcing it with the New Armies as they became ready for the field. Both the French and the British commanders-in-chief hotly opposed this proposal. The lines of communication through Serbia were long and difficult, and it was very doubtful whether they could be made to maintain an adequate force. The transference of the British army to the Balkans must in any event have taken many months, during which it would have been condemned to an inactivity of which the enemy on the inside of the circle, with not only shorter but better communications, would certainly have profited. The Balkan enterprise was therefore condemned, and the immediate outcome of the controversy on the British side was the starting of the Dardanelles adventure.

Joffre's plan was simpler. He desired to attack on the western front at the earliest possible moment and with all possible force. It was argued at the time, and has repeatedly been argued since, that it would have been better to have awaited the development of the British army and the increase of the Allied artillery and the improvement of their stock of munitions, and to have employed the interval in gaining advantages over less formidable foes in other theatres of war. To these arguments Joffre's answer was that the Germans were in occupation of a great part of France, that near Noyon they were only 50 miles from Paris, on the Somme they were barely 20 miles from Amiens, the main junction connecting the British and French in Flanders and Artois with the remainder of the French troops, while farther N. they were little more than 40 miles from Calais. At the beginning of 1915 he was assured of a definite numerical superiority over the Germans in the W., but the Germans had not nearly reached the limits of their man-power, and they might at any time call a halt on the Russian front, and by reversing the process which they had carried out after the first battle of Ypres bring back troops to France. A successful German attack at any one of a number of parts of the western front might gravely cripple the Allied armies, for German guns might be brought up to within range of Paris or of Calais, or the enemy might again occupy Amiens. The French commander-in-chief maintained that the security of the western front must be a paramount consideration in Allied strategy, and that to secure the position in the W. it was necessary to drive the Germans farther back.

Sir John French was in general agreement with Joffre's views. He at first desired a combined naval and military attack upon the coast of Belgium, but on receiving representations from Lord Kitchener that neither the men nor the munitions required for this operation could be made ready in time, he abandoned this proposal and set himself whole-heartedly to cooperate in Joffre's plans. These plans comprised a grand attack by the British army N. of the La Bassee canal and by the French northern group of armies under Foch on the front between the La Bassee canal and Arras. The hope was that this attack would give the Allies the Vimy Ridge and compel the Germans to evacuate Lille. In order that Foch might have the forces necessary for such a battle it was agreed between the commanders-inchief that the British should relieve the French troops, who had occupied the Ypres salient when Haig's men had been withdrawn from it after the first battle of Ypres. Sir John French had at the end of 1914 received one more regular British division, the 2 7th, made up by Kitchener from foreign garrisons, while another, the 28th, arrived early in January. The 1st Canadian Div. disembarked in France in the second week of February. This gave him a total of 1 3 infantry and 5 cavalry divisions, besides a number of selected territorial battalions. These reinforcements allowed him to form his command into two armies, the first under Sir Douglas Haig and the second under Sir Horace Smith-Dorrien, and at the same time to extend his front gradually N. into the Ypres salient.

A further integral part of the preparation for the spring campaign consisted of a number of attacks intended to divert the enemy's attention from the Artois front and to gain experience in the tactics of trench warfare. One of these attacks was carried out by the French during the winter against the salient of St. Mihiel, when they learnt that the position (which appears from a map to be almost untenable) could with the aid of trenches and barbed wire be made very formidable. Other attacks were made in Champagne on either side of the Perto and in Flanders. In the latter of these British troops cooperated by a minor attack opposite Wytschaete, which had very little success. In fact on the whole the Germans gained the chief success in this preliminary sparring, for in the middle of Jan. r915 they made an attack upon the lines which had been held during the battle of the Aisne by the British army, N. of that river near Soissons, and drove the French back to the S. bank.

The British army, which like other European armies, had been trained to believe in the supreme virtue of attack, had since the beginning of the war, with the exception of the short period of the advance to the Marne and the Aisne, been compelled to fight defensively, and Sir John French desired to give it a wider experience of attack upon entrenchments. For this purpose he had been carefully saving up artillery ammunition by strictly limiting the amount to be expended in the routine of trench warfare, and he calculated that he would have sufficient at his disposal to allow him to engage in a considerable battle early in March. He entrusted the execution of this battle to Sir Douglas Haig's I. Army, which was directed to attack the German lines near Neuve Chapelle, an operation which it was hoped would result in a favourable position for the further battle which was to be undertaken later in cooperation with Foch. The battle of Neuve Chapelle opened on Mar. ro with what was in those days held to be a very heavy bombardment. This bombardment was followed by an infantry assault, which was at first successful, and carried the village of Neuve Chapelle, but was soon brought to a standstill by the enemy's machineguns. In this battle the British first experienced the difficulty of bringing up reserves at the right time through roads blocked by the debris of battle and over ground scarred by trenches and torn by shell-fire. On the whole, however, both the French and the British were impressed by the result of the bombardment, which was held to promise great things when there should be more guns and more ammunition. Preparations for further battles therefore went forward.

The Germans, guessing what was on foot, were not slow to interfere with these preparations. On April 17 a portion of Smith-Dorrien's II. Army attacked and gained a footing upon Hill 60, an important feature on the S. face of the Ypres salient. The Germans, counter-attacking, promptly regained possession of the hill and followed up this success by a far more serious enterprise on April 22. By then Foch's preparations for the attack on the Vimy Ridge were far advanced, and to obtain the troops he needed for the coming battle he had greatly weakened the French forces left in the N. portion of the Ypres salient. Against this portion the Germans launched waves of poison gas discharged from cylinders, which completely overwhelmed the French troops, who had no protection against this utterly unexpected barbarity. The flank of the ist Canadian Div. was completely exposed, and for a short time a definite breach was created in the Allied line. Fortunately the Germans were unprepared for the extent of their success, and had not the troops at hand to take immediate advantage of it, while the gallantry of the ist Canadian Div. and the 28th Div. held the enemy at bay until reinforcements could be brought up. The first complete British territorial division to arrive in France, the 46th North Midland Div., was fortunately in the neighbourhood, and these and other reinforcements, some contributed by Foch from the reserves he was holding in readiness for his battle, sufficed to save Ypres and to patch up a new line.

But the second battle of Ypres, if it did not give the Germans all the success they might have attained had they been ready to follow up the first success gained by the employment of a method of attack which no civilized nation had conceived to be possible in modern war, at least gained a great part of their purpose by weakening the forces and exhausting the meagre supply of ammunition which the Allies were accumulating for their projected battle.

That battle began on May 9 with an attack by the I. British Army under Sir Douglas Haig on either side of Fromelles, and an attack by the I. French Army, commanded by General d'Urbal but under the direction of Foch, on a much wider front which extended from the Scarpe to the N. of the Souchez. The British attack made very little progress and it was soon evident that the preliminary bombardment had not sufficed to destroy sufficiently the enemy's defences, or, what was of greater importance, to overcome his machine-guns. Sir John French, however, felt himself bound by his agreement with Joffre and Foch to keep the enemy occupied on his front as long as possible, and the battle of Fromelles became merged in the battle of Festubert by bringing the front of attack slightly S. into what it was hoped would prove more favourable ground. The experience of Festubert was, however, hardly more favourable than that of Fromelles, for again the German machine-gunners checked all real progress, while the lack of artillery ammunition became more and more felt. The situation was made more difficult in this latter respect by renewed German attacks in the Ypres salient upon the II. Army, now under the command of Sir Herbert Plumer. The situation of the British army in Flanders was somewhat eased by the arrival of five more divisions of territorials, followed by the first of the New Army divisions, the 9th, but it was lack of shell rather than lack of men which forced Sir John French to stop his attacks, and the battle of Festubert petered out on May 25.

Meanwhile, on the British right, Foch was making encouraging if slow and very costly progress. In the battle of Souchez the villages of Ablain St. Nazaire, Carency, Neuville St. Vaast and Thelus were carried by d'Urbal's men, who fought their way doggedly up the W. slopes of the Vimy Ridge. The French found the German machine-guns to be the chief obstacle to progress and the prime cause of casualties, particularly in the villages, in which the enemy's machine-gunners fought indomitably from cellar to cellar and in a certain elaborate series of works which became known to fame as the Labyrinth. The French had by this time embodied the flower of the manhood of the nation in their army, and the splendid gallantry with which Foch's regiments fought their way forward in the battle of Souchez, enduring tremendous losses but ever gaining ground, if only at the rate of a score or two of yards per day, towards the crest of the Vimy Ridge, was never surpassed in the whole long war. When the British ceased their attacks on May 25, Foch had made enough headway to encourage him to believe that he could gain the whole. ridge, and he determined to continue the battle alone. But for this he wanted more troops, and an extension of the British front, which would set free Frenchmen, while both he and Joffre were fearful lest the New Army divisions preparing to leave England should be sent to the Dardanelles. The French commander-inchief therefore wrote, at the end of May, a letter to Kitchener which contains the key to the developments of the following months. It ran as follows: " The retreat of the Russian army, consequent upon the temporary failure of its offensive, will doubtless allow the Central Powers to withdraw, at least for a time, a certain number of army corps which they will be able to use on another front, but it is probable that the greater part of these will be required to meet the situation created by the entry of Italy into the war.' The situation of the Russians, who will be for some time to come unable to undertake a decisive offensive, and the difficulties of ground which the Italian 1 The Italian troops crossed the frontier at midnight May 24-25.

theatre of war presents until the Italian armies are able to debouch into the plains, show clearly that the principal effort of the Allies must be made in France. The developments of Arras 1 have proved that it is tactically possible to break the German front, but that for success an even more powerful effort is required, and that it is necessary to attack simultaneously at a number of points. France has at the present time 2,200,000 men engaged on her N.E. front and has reached the limits of her man-power. She can maintain her armies at their actual figure but she cannot increase them. The solution of the problem is then in the hands of England. If she sends us her New Armies we shall be able to make not only an English and a French effort but an English effort and two French efforts simultaneously at the most favourable moment and with great power. The British forces, reinforced by the New Armies, will be given zones of operation proportionate with their strength. They will retain their present zone, extending it on both flanks to the N. of Ypres and to the S. of La Bassee canal, and they will take over the zone to the S. of Arras towards the Somme. They will thus lie on either side of the French X. Army, which will hold the front of attack which it has organized. The alternation of French and British troops has always given the best results. The principal effort of the English armies will be directed between the left of the French X. Army and the La Bassee canal and will extend to the N. of the canal; thus it will be linked up with the French attack in the neighbourhood of Arras.

" Of course, if the Germans are compelled to fall back before all the British forces are in line, their whole strength will be directed towards exploiting the success won in the direction of Antwerp and of Brussels. We owe it to ourselves and to our Allies to make a great effort now. It is therefore at the present time of the highest importance that the New British Armies should be dispatched as rapidly as possible and at dates and under conditions fixed beforehand, so that definite plans of operations may be drawn up in agreement by the two commanders-in-chief. I am firmly convinced that our action will be decisive, if it is combined and coordinated." This letter shows that Joffre had not, at the end of May, much hope that Foch would capture the Vimy Ridge, but the latter's fiery spirit would not permit him to admit failure while the prize seemed to be within his grasp, and he persuaded the commander-in-chief to let him continue the battle. So the battle of Souchez, which had begun on May 9, dragged on until July 13. Twice Foch's men won their way to the crest of the ridge, only to be driven back, and at last Joffre called a halt and decided to prepare for the greater effort of which he had written to Kitchener. In that letter Joffre had indicated that he proposed to renew the attempts on the Arras front to take the Vimy Ridge, while the British army fought on the left of the French X. Army, but he had said nothing about the second French effort. This he designed to be the principal blow to be delivered in Champagne, to the E. of Reims. To obtain the French troops for this campaign he required the British army to extend its front to the right and left, and also to relieve de Castelnau's VI. Army on the Somme front. The remainder of the 'summer was occupied by these changes, which became possible as the New Army divisions arrived from England. The 2nd and 3rd of these reached France, disembarked at the end of May, and 6 more before the end of July, so that by then the 4 divisions and the cavalry division of Mons had in z i months increased to 28 divisions and 5 cavalry divisions. These reinforcements enabled a British III. Army to be formed under Sir Charles Monro, which took over some 17 miles of front from a point to the S. of Arras as far as the Somme, eventually extending its lines to the S. of that river. Haig's I. Army prolonged its right S. of the La Bassee canal to the neighbourhood of Lens, and thus found itself facing the open plain of Loos, while Plumer's II. Army relieved the remaining French troops in the Ypres salient and brought its left into contact with the right of the Belgians.

While these movements were in progress, vast preparations were taking place on the battle-front. Guns, trench mortars, shells and military stores of all kinds had accumulated in hitherto undreamt-of quantities. Joffre and his staff had it in their minds at this time that they were engaged in operations in the nature of a huge siege, and that the essential was to blow a practicable breach in the enemy's lines through which the infantry could be poured to the assault. A study of the previous battles of trench warfare had convinced them that with sufficient guns and sufficient ammunition this was possible. The output of the 1 The battle of Souchez.

French factories had been increased enormously, and though the British Ministry of Munitions had hardly yet begun to be productive, still the supply of heavy guns and shells for the British army had been greatly increased and it was equipped to reply effectively to the German gas. For these reasons the hopes which Joffre had expressed in his letter to Kitchener were very generally shared in the Allied armies. The one fly in the ointment was that there had been a renewal in the British Cabinet of the controversy between the " Easterners " and the " Westerners " and Mr. Churchill had pressed with all his eloquence and skill for a decisive campaign which should open the road to Constantinople. The result was a compromise, and three of the New Army divisions had gone to the East. Some compensation for this was obtained by the arrival of two more New Army divisions in France, and by the formation of a Guards' division, which had been made possible by the creation of new battalions of Guards and the replacement of those already in France by other battalions. This gave Sir John French an additional army corps, which he kept in his hands as a reserve.

Joffre opened his autumn campaign on Sept. 25. In addition to his great attack in Champagne, Foch with the French X. Army attempted once more to storm the Vimy Ridge, while Haig's I. Army attacked between the La Bassee canal and Lens, in conjunction with a secondary British attack to the N. of the canal and demonstrations on the front near Armentieres. The general plan of the N. battle was that Foch, having carried the Vimy Ridge, should advance E. to the S. of Lens, while Haig pushed E. through Loos to the N. of Lens and joined hands with Foch beyond that town. In this way the Allies would avoid the mass of ruined buildings and miners' cottages which composed the town and would form admirable refuges for the German machinegunners. The tactical methods to be employed on both the S. and N. battlefields were similar in conception, that is to say, there was to be a great rush forward of the assaulting infantry as soon as the attack had been adequately prepared by the artillery, and the reserves were to follow hard after the first-line troops. Only in the matter of the preparation of the attack was there any essential difference in the methods to be employed by the British and the French. The former had determined to replace an intense bombardment by a discharge of gas from cylinders similar to that from which they had suffered in the second battle of Ypres.

Of the British share in the general plan, the demonstrations in front of Armentieres proved quite ineffective, while the attack N. of the La Bassee canal failed. On the front opposite Loos, however, the discharge of gas surprised the Germans and overcame the resistance of their first-line troops, though it caused some casualties among the British themselves and made it difficult to direct the attack. Despite this, the first wave of British infantry passed through and beyond Loos, when it appeared that they only needed the prompt arrival of the reserves to secure a very considerable success. Sir John French had, however, kept these back until he saw how the battle progressed, and when they arrived late in the evening the New Army division which formed two-thirds of the reserve found great difficulty in making their way through the confusion of the battlefield under conditions entirely strange to them, and were unable to confirm the first success won, so that, though Loos was held, German counter-attacks recovered a considerable part of the ground gained, and in particular the important Hill 70, which dominates Loos on the north. Worse still, Foch's attack on the Vimy Ridge failed almost completely and the Germans remained in possession of the crest. This alone condemned the N. attack to failure, for the British front of battle was not large enough to insure a breach in the enemy's defences sufficiently wide to be exploited successfully. The remainder of the battle of Loos, which lasted until Oct. 15, resolved itself into the repulse of a number of fierce German counter-attacks, which ended in mutual exhaustion, with the British in possession of the salient extending round Lens to the La Bassee canal.

The great battle in Champagne was an even more severe disappointment, because more had been expected. The first news from the field aroused high hopes. The first two German lines of defence were carried on a wide front, and many prisoners and guns were captured, while on the fourth day of the battle the third German line, which was believed to be the enemy's last system of defence, was for a time breached near Sainte Marie; but again the solution of the problem of bringing up reserves at the right time and in good order was not found, while the enemy's reserves, which came up fresh through country which had not been fought over, arrived in time to fill the breach. The battle of Champagne ran on into Nov., developing into a series of struggles for tactical points of importance, but ended with no material change in the position won by the French at the end of the fourth day of the battle.

So the campaign of 1915 closed on the western front, with the Allies still asking themselves how it was possible to get through the trench barrier and drive the Germans from France and Belgium. The great bombardment followed by the assault in mass had failed, and some other method of attack was required.

Ere the year closed a new problem had developed in the Near East, which had its repercussion on the western front. German, Austrian and Bulgarian forces had overrun Serbia, and the Allied Governments had decided to send a relief expedition to Salonika. Part of the troops required for this new enterprise came from the Dardanelles, but more were needed and these had to be supplied from the western front. Five divisions of French troops under the command of General Sarrail were ordered off, and were accompanied by the three British divisions of the III. Army which had been holding the line S. of the Somme, the 27th, 28th, and 22nd. It was also decided to spare the two divisions of Indian infantry which had proved such a timely reinforcement at the time of the first battle of Ypres, the horrors of another winter of trench warfare in Flanders, and they were sent off to Mesopotamia. This reduction of the British army was made good by the arrival in France before the end of the year of the 2nd and 3rd Canadian Divs. and the 16th Irish Div., and of the 21st, 23rd, 24th, 25th, 26th, 36th and 38th New Army Divs., so that the strength of the British army in France and Belgium stood at 36 divisions and 5 cavalry divisions, a combatant strength of about 750,000 men. The Allies on the western front had therefore at this time about 3,100,000 men, opposed to 2,000,000 Germans. These figures gave rise to anxious discussion as to what superiority of numbers was necessary to ensure success under conditions of trench warfare, and whether and how the Allies could obtain the necessary superiority.

While these discussions were going on important changes took place in the Allied Command. Joffre had hitherto been nominally chief of the French General Staff. He was now definitely appointed commander-in-chief of the armies of the N.E., with de Castelnau as his chief-of-staff. In the British army Sir John French was brought home to command the troops in Great Britain and was succeeded by Sir Douglas Haig, with General Kiggell as his chief-of-staff, while Sir William Robertson, who had been chief-of-staff to French, became chief of the Imperial General Staff at the War Office.

Operations in 1916. - In Dec. 1915 the first serious attempts obtain unity of action between the Allies took place, and a conference of commanders-in-chief and chiefs-of-staff of the British, French, Belgian, and Italian armies, attended also by representatives of the Russian and Japanese armies, was held at Joffre's headquarters. At this conference it was agreed to attack the enemy as early in 1916 as possible, sufficient time being allowed for the training of the New Army divisions that the British army was receiving, and for the reequipment of the Russian armies. These plans, however, never matured, because the enemy anticipated them; and it is therefore unnecessary to refer to them further.

During Dec. 1915 and Jan. 1916 the Germans developed considerable activity along the front and made local attacks at Nieuport on the North Sea coast, against more than one point in the Ypres salient, at Givenchy on the Vimy Ridge front, on the Somme, and in Feb. on the Alsace front. The majority of these attacks took the form of experiments in various methods of bombardment which the Germans wished to test in view of a greater effort which they were planning. There had not been wanting signs that the enemy were preparing an attack on the Verdun sector; and considerable anxiety having been expressed as to the adequacy of the French defences on that part of the front, General de Castelnau was sent thither by Joffre, and he ordered certain improvements, but the time was lacking to give full effect to his recommendations (see Verdun). In fact, the Allied defences at this period of the war were notably inferior to those of the Germans. The Allies had spent the greater part of 1915 either in carrying out the vast preparations necessary for an attack in trench warfare or in attacking, and had little energy or labour left for the elaboration of defences. The'Germans on the other hand had been on the defensive throughout the year, and they enjoyed the great advantage of being able to employ upon their entrenchments forced labour from Belgium and the occupied provinces of France, and the large number of prisoners they had captured on the Russian front. Further, they had immediately behind their zone of battle the forests of Alsace and of the Ardennes, which provided them with an almost unlimited amount of timber. The Allies could only find labour either at the expense of the fighting troops or of the munition factories, and the British army was forced to import the greater part of the timber it required. It was not until much later in the war, when elaborate arrangements were made for the provision of native labour, and for the exploitation of the French forests, that conditions became at all equal in these respects.

It was the superior strength of their defences which enabled the Germans, while inferior in numbers on the whole front, to concentrate sufficient force for a great attack upon one part of that front. That attack opened on a front of nine miles on the banks of the Meuse opposite Verdun on Feb. 21, 5 German divisions attacking 2 French divisions. The enemy at once gained startling success, penetrating the French defences, and on the fourth day of the battle capturing Fort Douaumont, one of the chief of the outlying works of the fortress of Verdun. This success was greater than any yet gained by the Allies in attack, though their relative superiority in men on the battle-fronts had been far greater than that of the Germans at the beginning of the battle of Verdun. The Germans won their successes mainly by the skilful handling of their medium and heavy guns, of which they had assembled a great number for the attack. They had before the war made a much closer study of the use of heavy and medium howitzers, both for field and siege warfare, than had either the British or the French, and they had numbers of gunners highly trained in their use, while careful experiments in bombardment, carried out before the battle, bore good fruit. It is to these causes that the tactical success won by the Germans in the first days of Verdun may be attributed. At Mons 5 German divisions had attacked 2 divisions of the British II. Corps, but the Germans had not had time to bring up their heavy artillery and at the end of a day's fighting had done little more than drive in the British outposts, while at Verdun, with the help of a mass of heavy guns, 5 German divisions had overcome 2 French divisions and gained such a position as menaced seriously the French fortress. The result was, therefore, a tactical victory for the German artillery.

Joffre dealt with the crisis promptly. De Castelnau was sent again to Verdun and arranged with Langle de Cary, who commanded the group of armies of the Centre, a command which included Verdun, as to the disposal of the reinforcements which were hurrying to the battle-field and as to the methods of defence. Joffre also sent with these reinforcements General Petain, who arrived on Feb. 26 and assumed command in the battle-zone. The German attempts to gain ground beyond Douaumont were repulsed, and the French commander-in-chief had time to look round and survey the whole position. He had at once realized that the enemy was in. deadly earnest. " C'est la bataille," was one of his first remarks when he heard of the German attack. He had immediately requested Haig to relieve his X. Army on the Arras front, and had asked Kitchener to hasten the dispatch of British reinforcements to France. Two more New Army divi sions and i Territorial division had reached France from England in January, while 3 more New Army divisions and 3 more Territorial divisions from England were to be in France before the end of May. The Dardanelles Expeditionary Force was being rapidly reorganized and refitted in Egypt by Sir A. Murray, and 2 British, 4 Australian, and i New Zealand divisions were expected to come from that country to the western front before the end of June. Haig had lost in Jan. one more division, which had gone to Salonika, but reinforcements would give him by the middle of the year an increase of no less than 15 divisions to his strength in Dec. 1915.

Both commanders-in-chief were agreed that the principles on which the plans of battle in 1915 had been drawn up required modification. It was seen that the analogy of the great siege did not hold, that something more was required than to blast a great breach in the enemy's lines and then to launch a great assault. The something more was the defeat of the enemy's reserves, which came up fresh and in good order to meet troops when the assault had been thrown into some confusion. It was agreed, therefore, that the first object of battle should be to draw in and exhaust the enemy's reserves, and that until that object had been achieved no decisive success could be expected. So long as the enemy continued to attack Verdun, it would, on this principle, be to the advantage of the French to endure these attacks provided always that the enemy gained no success which would affect seriously the strategical position on the whole front, and provided that the exhaustion of the French man-power was not excessive. Joffre therefore proposed to fight defensively at Verdun as long as possible, but to be ready to strike back as soon as the situation there appeared to him too dangerous, or as soon as the French army was approaching the limits of endurance. He therefore asked Haig not only to relieve the French X. Army but to prepare to attack N. of the Somme on a front as wide as the resources of the British army would permit, and undertook to support that attack with a French attack S. of the river on a scale which would depend upon the effect of the battle of Verdun upon the French army.

The Germans had at this time arrived at a very similar theory of battle. The chief of their General Staff, Falkenhayn, has said that he believed the strain of the war upon France to be such that a break might occur if the strain could be increased. This was not an unduly optimistic appreciation of the position if it be remembered that in the previous May Joffre had told Kitchener that France had then reached the limits of her capacity to expand her military forces. Materially Falkenhayn was not far wrong, but he understood the psychology of his enemies no better than other Germans, and he failed to appreciate the spirit of France. The maxim of Verdun - " on ne passera pas " - became for France an inspiration as potent as the influence of Jeanne d'Arc. The object of the Germans in the battle of Verdun was to bleed France white, while the object of the Allies during that period was to wear down the military power of Germany as the preparation for striking the coup de grace. So upon both sides the theory of the war of exhaustion developed.

During March, April and May the struggle for Verdun continued, the Germans in their several attacks gaining sufficient ground to encourage them to make a new effort now on the right bank of the Meuse, now on the left. During all this fighting the reputation of two of the defenders of Verdun increased steadily, and in May General Petain succeeded to the command of the group of armies of the Centre, General Nivelle taking his place in command on the actual front of battle.

Between Jan. and July the British strength in bayonets and sabres grew from 450,000 to 660,000, and there was a more than corresponding increase in artillery and aircraft. This enabled Haig to fall in completely with Joffre's wishes, and as soon as the relief of the French X. Army was completed he set about preparing for the great attack on the front N. of the Somme. The growth of the armies made necessary the creation of a IV. Army under the command of Sir Henry Rawlinson. The III. Army was now commanded by Sir E. Allenby, and Sir Charles Monro, who had gone out to supervise the evacuation of the Dardanelles, returned to command the I. Army. British troops now held a continuous line from the Yser canal to the Somme, and were actively preparing to take upon themselves more of the brunt of the war on the western front, a burden of which nearly seveneighths had hitherto fallen on their French comrades. This preparation entailed enormous labour, for it not only involved the accumulation of immense stores and piles of munitions of all kinds on the selected front of attack, the construction of miles of roads, railways and trenches and many other preliminaries of a great attack in trench warfare, but also the completion of the training of hitherto untried troops. The British army was, in fact, in process of becoming a national army. The old regulars were now little more than a small leaven of the whole lump, and though the new troops arrived in France with considerable knowledge of their duties as cavalrymen, infantrymen, gunners and airmen, they had had little opportunity of learning to work together as part of a great machine. The problem of training was the more complicated because of the great variety of new weapons and methods which had been developed since the outbreak of war; and to teach each part of the army the powers and limitations of every other part, and the whole to work together in combination, was a very heavy task. To this task Haig at once addressed himself, and formed behind the lines schools of instruction in every branch of trench and battle warfare, while careful arrangements were made for the training in attack of divisions when out of the line. All this busy work behind the lines did not mean any cessation of activity on the front, and during these months of preparation a long series of raids into the enemy's trenches were planned and executed, raids which gave the new troops valuable experience and kept the enemy on continual tenterhooks.

In the latter part of May the German Crown Prince redoubled his attacks on the Verdun front, and on the list the Germans stormed the Mort Homme hill on the left bank of the Meuse, for which they had been struggling for weeks. Petain at once called Joffre's attention to the gravity of the situation, and pressed for an early beginning of the counter-offensive on the Somme, but Joffre was anxious to give Haig as much time as possible for the training of his troops, and made Petain endure yet longer. Fire from the Mort Homme hill had long impeded German progress on the right bank of the Meuse, but with the hill in their possession the enemy in the beginning of June began to press hard on that bank and on the 7th captured Fort Vaux. Thereupon Petain renewed his representation to Foch, who in consultation with Haig decided that the battle of the Somme should begin on July i. The preliminary bombardment was begun a week earlier, on June 24, the day on which the Germans, after capturing Fort Thiaumont, stormed the village of Fleury and attained the farthest point in their progress towards Verdun. While this preliminary bombardment was in progress, a bombardment so intense that the guns could be heard in England, no less than 70 raids were carried on the British front between Ypres and the Somme, and gas was discharged into the enemy's lines from 40 different points. The troops which were about to attack the German trenches from Gommecourt to the Somme included the flower of British manhood, and no more splendid body of men has ever gone forward to battle. British aircraft had already gained the ascendancy over the enemy's airmen, and the British army was now well equipped with machine-guns, trench mortars, bombs, gas-projectors, and all other new appliances which experience of trench warfare had shown to be necessary; the tunnellers had proved themselves to be more than a match for the enemy's mines, and for the first time in the war there was a sufficiency of heavy and medium guns and an assurance of an adequate if not abundant supply of shells. Expectation therefore ran high.

The results of the first attack were a heavy disappointment. The main attack was delivered by Rawlinson's IV. Army between the Ancre and Maricourt, about a mile N. of the Somme, where they joined hands with the French VI. Army commanded by General Fayolle, who was to attack astride the river under the general direction of Foch. This main attack was combined with a subsidiary attack by the IV. Army N. of the Ancre, and yet another by the III. Army upon the Gommecourt salient. Upon July 1 both the subsidiary attacks and the whole left of the main British attack failed completely and with heavy loss, but the right of the main attack and the whole of the French attack made such good progress as to warrant the continuance of the battle. The British failures were in the main due to want of experience in the artillery. More than two-thirds of the British batteries engaged had been created since the outbreak of war, and at that time they did not possess sufficient ammunition to gain such practical experience in intense bombardment as the Germans had given their gunners before the battle of Verdun. The chalky soil of the Somme hills lent itself to the construction of deep dugouts, of which the enemy had a great number, and to prevent his men from coming out of these in time to meet the British infantry special and very accurate methods of artillery preparation were required. In default of these the infantry yet again found themselves checked on the greater part of the front by the deadly German machine-guns, and it was mainly through the devoted valour of the infantry through the first days of the battle that the gains made by the right of the IV. Army were confirmed and extended. The first successes won by Foch's men along the river itself were greater and were obtained at less cost, partly because the Germans, overrating the effect of the battle of Verdun, had not expected attack from that quarter, and partly because of the better preparation of the attack by the French artillery, the French having a far larger number than the British of trained gunners for the expansion of that arm. The lesson of the first days of this great battle is that the creation of new armies during the course of a war is an even more intricate and difficult business than had been imagined, even by those who, knowing some of the difficulties, had undertaken the creation of Kitchener's armies with devotion and enthusiasm.

One of the results of the events of the first days of the battle was that Sir Douglas Haig decided to divide his fighting front between two armies. He directed Rawlinson with the IV. Army to exploit the advantages won on the right, and formed on his left a V. Army under Sir Hubert Gough which was to keep the enemy busy on its front and act as a pivot to the IV. Army. The battle was fought out in three phases, the first being the struggle up the slopes from the valleys of the Ancre and of the Somme to the S. crest of the Somme plateau, the second the struggle for possession of the plateau, the third phase consisting of the advance down the N. slopes. The first phase was consummated by a brilliant night attack by the IV. Army on July 14, and on July 17, the British and French troops N. of the river were abreast of the French S. of the Somme, who had for some time been established opposite to Peronne. The whole front of attack could again move forward together. The second phase constituted a long series of fierce struggles, the Germans bringing up more and more troops and disputing every yard of ground, so it was not until Sept. 9 that the British, with the French on their right, were able to look down upon the N. slopes of the plateau and the plains beyond.

By this time two of the objects for which the battle had been fought were gained. The Germans, forced to transfer troops to the Somme, had to relax their pressure on Verdun. The French retook Fort Thiaumont on June 30, while throughout July they slowly regained part of the ground which had been won from them, and on Aug. 17 drove the enemy out of Fleury. Verdun was no longer in danger, and Petain and Nivelle were able to plan at leisure counter-attacks on a more extensive scale. The second object of the battle of the Somme, the exhaustion of the enemy's reserves, was being obtained as surely. When the battle began, the front attacked by the British was held by 6 German divisions, that attacked by the French by two. In the 2 months 36 German divisions had been engaged on the British front, 18 on the French. In the 6 months of Verdun the Germans had employed 43 divisions in battle, so that their defence on the Somme was far more exhausting than their attacks at Verdun.

At the end of Aug. the failure of Falkenhayn's plans was publicly admitted by his supersession by Hindenburg, with Ludendorff as his chief assistant. The latter, after visiting the fields of Verdun and the Somme, found the German position on the western front to be one of great gravity, and the chief problem confronting him to be how to stop " the progressive falling off " of the German fighting power. The situation of the Allies had improved marvellously since June, when men were wondering how long it would be before the Germans entered Verdun. Not only had Verdun been saved and the Germans been forced to fight desperately on the defensive, but the Italians had driven back the attacking Austrians in the Alps and had then passed themselves to attack on the Isonzo. On the Russian front Brussilov had won great victories on the Bukovina, and Rumania had entered the war, too late certainly to profit by Brussilov's success, but none the less adding apparently another ally to those who confronted the Central Powers. The whole machinery of the Allies was, for the first time, simultaneously at work, and Joffre's strategy appeared to be triumphant.

It was in these encouraging circumstances that the third phase of the battle of the Somme began on Sept. 15. The attack of that day was made famous not only by the successes won, which were considerable, but by the fact that tanks then made their first appearance in battle. There has been much controversy as to the wisdom of this step (see Tanks). The experts have maintained that the value of this invention was discounted by premature use, that it should have been kept in reserve to surprise the enemy when large numbers of the new weapon were ready, and that it should have been first used on ground more favourable than a shell-torn battlefield. It was decided to employ tanks in the Somme battle for two reasons. Firstly, so much having been gained at great cost, the moment seemed to have come to press the enemy with every available means. The chief obstacle to the progress of the infantry continued to be the German machine-guns, and tanks were reputed to be the ideal means of overcoming machine-guns. If it would have been foolish not to have pressed the advantage won, it would have been criminal to have withheld from the sorely tried infantry the protection and aid which was at hand. The second reason was that experience was required in the use of tanks in battle. It was necessary to learn both how the tanks would comport themselves when put to the highest test, and how they would work in combination with infantry and artillery. The effective cooperation of infantry, tanks and artillery undoubtedly went a long way towards winning the war in 1918, but it is a legitimate belief that this cooperation would not have resulted unless experience had been gained in 1916.

Despite the employment of tanks, and despite the splendid valour of the infantry of the New Armies, the resistance of the enemy was not broken in the third phase of the battle of the Somme. The days were growing shorter and the weather became uncertain, while the enemy, drawing troops from all parts of the front to prevent his line from breaking, fought with fine courage. By Nov. 17, when rain and mud put an end to the battle, the Germans had engaged no less than 127 divisions. The enemy's reserves had indeed been worn down; in the valley of the Ancre he was hemmed by Gough's V. Army into an awkward salient, but the weather had broken and it was too late to reap the harvest on the battle-front. The first fruits of the Somme were garnered elsewhere.

On Oct. 24 Nivelle began an attack on the right bank of the Meuse, and on the following day recaptured Fort Douaumont. This conspicuous success was followed by the recapture of Fort Vaux on Nov. 2. The battle ended in a complete victory for the French, 6 French divisions overcoming 7 German divisions at surprisingly small cost. Nivelle and Mangin, who commanded the army corps engaged, became the heroes of France. The victory was largely due to the skilful handling of massed artillery, and the Nivelle method became famous. Its fame was extended when, on Dec. 14, a second attack won an even more brilliant success, which made Verdun quite secure and brought in 11,387 prisoners and 115 guns, again at small cost. When the statesmen of Paris and London compared the results of these two battles at Verdun, which had resulted not only in important gains of ground but in the capture in a few days' fighting of more than 17,000 prisoners, with the slow bludgeon work of the Somme in which the British army in four and a half months had captured 38,000 Germans at a tremendous price, they began to think that they had at last discovered the man for whom they were looking so anxiously.

Operations in 1917. - On Nov. i i 1916 Joffre assembled his second conference of commanders-in-chief to consider plans for the following year. It was agreed that the Germans were in great difficulties on the western front, and that the situation of the Allies was more favourable than it had ever been. The fighting strength of the British army had grown to about 1,200,000 men, and it was known that considerable further reinforcements would reach France during the first few months of the year. The fighting strength of the French army had been increased by the incorporation of native troops to about 2,600,000, so that, including the Belgians, the Allies disposed of about 3,900,000 men against about 2,500,000 Germans.

Joffre declared that the French army could maintain its strength for one more great battle, but that thereafter it must progressively decline, as France had no longer a sufficient number of men of military age to replace losses. He therefore warned Sir Douglas Haig that during the coming year the burden must fall more and more upon the British army, a position which the British commander-in-chief readily accepted. Germany had recently created a number of new divisions, some of which had been employed against Rumania, but it appeared probable that the transfer of these divisions to the western front would be delayed if Russia was able to be active on the eastern front, and of this the great improvement in the supply of munitions for the Russian army held out promise at this time. It was also agreed that, in view of the probable decline in strength of the French army later in the year and the promised reinforcement of the British army, the relative superiority of the Allies on the western front would be greater in the spring of 1917 than at any time which could be foreseen with certainty. For all these reasons it was decided to take the earliest possible opportunity of pressing the advantage won by the battle of the Somme, and of continuing the process of exhausting the enemy's reserves as preparation for an effort which should be decisive. All the armies of the Entente were to be ready to attack in the first fortnight of Feb., the British army between the Vimy Ridge and Bapaume, the French armies between the Somme and the Oise; and the French attack was to be followed soon after by another in Champagne to the W. of Reims. It was further understood between Joffre and Haig that these attacks would, if necessary, be followed by further attacks by the British army in Flanders. During the winter the British army was to do its utmost to press the enemy on the Somme battlefield, and to prevent him from recovering from his embarrassment there.

The underlying ideas of this plan were primarily that the policy of exhausting the German reserves should be resumed at the earliest possible moment, and secondly that the utmost effort should be made to complete the work begun on the Somme. The commanders-in-chief believed the situation to be such that victory could be won in 1917, but they were under no illusion as to the possibility of ending the war by one great blow to be delivered in the spring. Joffre followed up the results of this conference by issuing general instructions embodying the decisions reached, and in these instructions he directed that the first British and French attacks, that is to say, those to be delivered between Vimy and Bapaume and between the Somme and the Oise, were to be ready by Feb. 1.

No sooner were Joffre's plans completed than a series of intrigues against the French commander-in-chief came to a head.

A number of officers of the French General Staff regarded with dismay a proposal to give more and more of the task of consummating victory to the British army and less and less to the French army. They found many supporters among the politicians in Paris, and these were reinforced by others, who feared that the " war of exhaustion " and the process of wearing down the enemy's reserves would end in exhausting France before it exhausted Germany. The cry therefore went up that it was time to have a change in the Higher Command. Foch, whose bloody assaults upon the Vimy Ridge had not been forgotten, was held to be too much of a " hammer and tongs " fighter, and he was placed on half-pay, while the state of the defences of Verdun before the German attack began was brought up against Joffre. So he was given a marshal's baton and an honorific position in Paris, and Nivelle reigned in his stead.

The new commander-in-chief at once made a drastic change in Joffre's policy and plans. He proposed to increase the weight of the French attack; and in order that he might obtain the French troops necessary he proposed that the British should relieve the French VI. and XX. Armies and extend their front S. across the Somme as far as the Amiens-Roye road. In return he proposed that Haig should modify his plans for keeping up the pressure on the Germans on the Somme battlefield during the winter, and that the date of the combined attacks should be postponed until Mar. 15. This meant a delay of six weeks in launching the attacks planned by Joffre, and the enemy would be given time to recover from the effects of the Somme. But it became clear, as Nivelle's plans developed, that there was to be an even more complete change of plan than this. He proposed to apply on a great scale the methods he had employed with such success at Verdun, and to return to the policy which had been discarded after the failure in Champagne in 1915. He intended, by skilful employment of a great mass of artillery, to overcome the enemy's resistance in his front lines, and then to pour in to the assault a great reserve which should break through the trench barrier completely and so change the whole strategical position on the western front.

Early in Dec., 1916, there had been a change of Government in Great Britain, and Mr. Lloyd George had become Prime Minister. He had made up his mind that the Somme had been a costly failure, and was eagerly looking for some method of winning the war which should be speedier and less costly in life than that of a " war of exhaustion." He therefore welcomed a general who promised a short, sharp and decisive battle, which would be over, one way or the other, within a comparatively short time. So at a conference held at Calais at the end of Feb. 1917, it was agreed between the British and French Governments that the British army should be placed under the general direction of Nivelle for the forthcoming operations. This decision violated a fundamental principle of military organization. A general of division is not, while still in command of his division, placed also in command of an army corps which includes other divisions, for the good and sufficient reason that if his attention is absorbed by the details of one unit or of one part of the front, he cannot simultaneously give proper attention to the other units or to other parts of the front. The right course would have been to have given Nivelle the general charge of the whole western front and to have appointed another commander-in-chief for the French army. The results of this mistake soon became apparent. On the last day of the Calais Conference news arrived from the British V. Army that there were signs of a German withdrawal in the valley of the Ancre. Some time before, the British airmen had discovered that a great new system of defensive works had been constructed by the enemy covering Douai, Cambrai and St. Quentin, the system which became known to the Allies as the Hindenburg line; and not long after the report from the Ancre came in there were indications that the Germans were preparing to retire from the whole of the Somme battlefield into this line. But Nivelle, not being in close touch with happenings on the British front, did not believe in a German retreat, and issued to Haig orders which were not compatible with the changed situation, and in certain matters went far beyond the agreement reached in Calais. This led to friction, which was adjusted at a further conference in London. By then it had become apparent that the Germans were in retreat on the whole front between Arras and the Aisne near Vailly. The Germans, relieved from pressure on their front during the winter, had prepared for their retreat systematically and brutally. The whole country which had been in their occupation W. of the Hindenburg line was devastated, villages were burnt, roads and railways destroyed, fruit trees cut down and everything of any value was removed, and mines which exploded at a touch were prepared with diabolical ingenuity. In these circumstances a rapid pursuit became impossible, and the Germans were able to delay the advance of the Allies by rearguards, while they removed their heavy artillery and established their main bodies in the Hindenburg lines.

This manoeuvre, planned and successfully carried through by Ludendorff, effected a great change in the situation to the benefit of the Germans. Not only did it materially shorten their front and thereby enable them to increase their reserves, but their troops exchanged the battered defences of the Somme battlefield with its awkward salients for the strongest lines which had yet been built upon the western front. Further, the enemy had withdrawn from a considerable part of the front which Nivelle had intended to attack, and this made necessary a further postponement of his battles, but he still adhered to the main features of his plan. In the altered circumstances grave doubts arose in the minds of some of the senior French generals as to the feasibility of this plan, and when these came to the ears of the French War Minister, M. Painleve, he assembled a Council of War on April 6, on the very eve of the offensive, at which criticisms of the plan were presented by certain of the commanders who were to take a leading part in its execution. Nevertheless, the French Government decided not to interfere with Gen. Nivelle. It is difficult to conceive of a more unfortunate prelude to a great battle. However, these doubts and hesitations of the leaders were not known to the rank and file of the army or to the French people; and when, on April 9, the spring campaign began by an attack by Allenby's III. Army on the enemy's lines E. of Arras, and by the Canadian corps with one brigade of the 5th Div. on the Vimy Ridge, and met with an immediate success, hopes soared high. The French public was deeply impressed by the rapid capture of the Vimy Ridge, which had for so long resisted Foch's attacks, and great things were expected when the French army advanced.

The second of Nivelle's blows was delivered by the group of armies of the centre, now under Franchet d'Esperey, against the German front in the neighbourhood of St. Quentin; but Franchet d'Esperey was here in contact with a part of the main Hindenburg line, and he had neither the time nor the means to prepare effectively for an attack upon their formidable defences. The operations of the centre group of armies, which had been intended to be an important part of Nivelle's programme, dwindled therefore into little more than a demonstration, which took place on April 14 and had no material results. Nivelle's main battle, which took place on the front between Reims and Anizy, began on April 16. It had been planned that the assaulting troops should on the first day of battle break through the first three German lines. The attack was made by Mangin's VI. Army and Mazel's V. Army, with Duchesne's X. Army and a mass of cavalry in reserve ready to exploit their success. Antonine's IV. Army struck in to the E. of Reims on the 17th. The left of Mazel's attack failed almost entirely; and elsewhere, though the first German line was captured, little progress was made beyond it. The dream of a rapid rupture of the enemy's front had to be abandoned, and a fresh plan of battle had to be formed.

One of the first results of the failure of Nivelle to realize his hopes was that he had to request Haig to press his attacks to the E. of Arras with all possible vigour, so as to keep the largest possible number of Germans occupied in that quarter. This entailed a prolongation of the battle of Arras into a period when gains became small and were only purchased at great price. None the less Haig decided that the situation made it necessary to support the French with all his power, and he fought on till May 17, by which time the British front was established some 4 m. to the E. of Arras and in the plain to * the E. of the Vimy Ridge. While Haig was thus battling in the N., Nivelle on the Aisne front had brought his X. Army into his front line, and by slow and bitter fighting had won his way up the Chemin des Dames ridge, of which he captured the eastern portion. Early in May it was quite evident that there was no prospect of such a break-through as had been planned, and on the 15th the French Government replaced Nivelle by General Petain, while General Foch, recalled from semi-retirement, became chief of the staff in Paris. Petain's first task was to wind up the operations on the Aisne front, and the battle ended definitely on May 20.

The spring campaign had proved a failure in comparison with what might have been, and still more in comparison with what Nivelle had promised, but its results were far from being insignificant. The German retreat in March, which was a direct consequence of the battle of the Somme, had at last brought about the attainment of one of the objects for which Joffre had been striving for so long. The Allies had now more elbow room on one of the most vital parts of their front, that which covered directly the roads to Amiens and Paris. Had the Germans in March 1918 started from the positions which they held in Feb. 1917, and had their attacks progressed at the same rate, they would have entered Amiens on the second day of the battle, which would have ended with the German guns bombarding Abbeville and communication between the French and British armies severed. It is therefore not too much to say that the retreat which was forced upon the Germans by the battle of the Somme saved the Allies in the following year. But how much greater might the results have been if the plan formed by Joffre and Haig in the previous Nov. had been followed - if the Germans had been pressed on the Somme battlefield during the winter, and if they had been attacked early in Feb. before their plans for retreat had been completed. Despite all the difficulties with which the successful conduct of that retreat by the Germans had confronted them, the Allied armies had in the battles of April and May captured 62,000 prisoners, 446 guns, and i,000 machineguns, and had gained positions of the first importance; 57 divisions had been compelled to fight upon the French front and 99 on the British front. Had Nivelle been content to follow Joffre's example, and to prepare methodically for the exhaustion of the German reserves without overtaxing the endurance of the sorely tried French army before attempting to break through the enemy's lines, he might have claimed a conspicuous success for his first campaign. But the hopes which he had roused had been extravagant, and the dejection when they were not realized was correspondingly great. The dejection was increased by the news of the Russian revolution, and by exaggerated reports of the losses in the Aisne battles; and it was hardly alleviated by America's entry into the war, for it was well understood that American troops could not be ready to take their places in the firing line during 1917. The immediate consequence of this dejection was the outbreak of a series of mutinies in the French armies, which so affected the moral of the French troops that Petain found it necessary to appeal to Haig to keep the enemy engaged while he restored the confidence of his men.

If the attention of the Germans was to be occupied by the British armies it was necessary that they should be forced to fight. Upon any part of the British front S. of the point where it bends S. from the Belgian frontier N.W. of Lille it was possible for the Germans to repeat their manoeuvre of March and avoid a battle by retiring into another system of defences, for in doing this they would be merely abandoning a portion of French territory which was of no great value to them, while they might by this method economize sufficient troops to enable them to fall upon the French. On the Belgian front they could not fall back without risking their hold upon the ports of Ostend and Zeebrugge, which were to them important bases for their submarine campaign, and without endangering the security of the chief aerodromes from which their air attacks upon England were made. For these reasons Haig decided to press the enemy with all his available means upon the Belgian front, and on June 7 he began this campaign with the battle of Messines. This battle was most skilfully and thoroughly prepared by Sir Herbert Plumer, and was fought and won by his II. Army. The battle began with the explosion of a number of huge mines, the secret of which had been preserved by constant and devoted watchfulness on the part of the miners, who had tunnelled beneath the enemy's lines many months previously and awaited patiently the opportunity for their use. The effect of these explosions, combined with a very skilfully planned bombardment of massed guns of all calibres, was such that, except in the right of the attack in the neighbourhood of Messines, the infantry, for once, had little to do. The whole of the Messines-Wytschaete ridge was captured at comparatively very light cost, and the Ypres salient, a name of ill omen for the British army since Oct. 1914, disappeared.

The strength of the British army in combatant troops was now at its greatest. Haig had 64 divisions and 10 cavalry divisions under his orders, and a mass of heavy artillery, tanks and aeroplanes. It was well that this was so, for the army was to be called upon to endure greatly while Petain and his men got their second wind. Nor was it only the situation on the western front which called for resolute action, for the condition of the Russian army was far more critical than that of the French army. Kornilov had, on July 1, begun an offensive, and if he was supported and encouraged by success in the W. it was still possible that the Russians might continue to be a powerful factor in the war.

With these heavy responsibilities on his shoulders, Haig began on July 31 the second battle of Ypres, with a great attack by the V. Army, which had been moved N. for the purpose. This attack was combined with a subsidiary attack by the French IV. Army under General Antoine on the British left. The object was to gain possession of the Passchendaele ridge, so as to be able to sweep with gun-fire the plains beyond it toward Zeebrugge and Ostend. This achieved, a combined naval and military attack, which had been secretly prepared in England, was to be made on the Belgian coast, which it was hoped would gain possession of the ports and so relieve the British Admiralty of some of the many anxieties caused by the German " U " boats. In preparation for the landing the British took over from the French the lines on the Belgian coast near Nieuport and moved other troops up to the coast behind these lines. The position at Nieuport, which consisted of a narrow strip of ground, with the Yser at its back, was not easy to hold against determined attacks, and before the British preparations for defence were completed the Germans attacked and captured the lines E. of the river. This was an inauspicious beginning, but worse followed, for the weather broke immediately after the battle began and then followed a rainfall unprecedented for August.

The plan of battle was to deliver a series of blows, each with an objective limited by the support which the artillery could give without changing position. It was believed that the experience of Messines and of Verdun had shown that this would allow the infantry to reach their objective without heavy loss. Ludendorff, however, met this method of attack by a new method which he called the elastic system of defence. He made no attempt to hold his front lines in strength, but withdrew the bulk of his infantry from the zone which would be most heavily bombarded and relied mainly upon machine-guns in concrete " pill-boxes " to break up the British infantry attack, and upon counter-attacks by the troops whom he had held back. But it was less this method of defence than the mud of Flanders which prevented British progress. The opening battle of July 3 1 gave the British possession of the whole of the Pilken Ridge, of the German first line of defence between Nordshoote and Klein Zillebeke, a front of 10 m., and of most of the German second line, but it was not until Sept. 20 that the enemy's third line was penetrated, and not until Oct. 4 that the British were established on the high ground between Broodseinde and Becelaere. The difficulty of getting guns and ammunition forward through the slough of mud prevented the delivery of a rapid succession of blows, each with a limited objective, as had been planned, and in the event a more terrible strain was imposed upon the British troops than in any other battle of the war. As in the case of the battle of the Somme, the first fruits of the third battle of Ypres were reaped elsewhere than on the battle-front. The Germans, forced to send more and more troops into the fiery furnace which blazed in the Ypres ridges, were compelled to leave the French alone, and Petain had time to restore the confidence of his army. Part of his method was the delivery of very carefully prepared attacks on a comparatively small front, supported by a great mass of artillery which should leave the infantry little more to do than to occupy the ground won. The first of these attacks was delivered by Guillaumat's II. Army on the Verdun front, and was completely successful, ending with the French in possession of all the ground which the Germans had won in 6 months fighting in 1916. This was followed by a more important attack delivered on Oct. 23 by Maitre's VI. Army, which gave the French the whole of the Chemin des Dames ridge, and resulted in the capture of 11,00o prisoners and 200 guns. Then and not till then Petain expressed himself as satisfied that his immediate purpose was achieved.

The British troops, struggling in the mud of Flanders, could not be told the reasons which had called for a supreme effort from them, and the terrible struggle through the mud, unrelieved by any conspicuous success, was a heavy strain upon them. As events turned out it would probably have been wiser to have brought the third battle of Ypres to a close immediately after the French had won the Chemin des Dames, but at that time the British were within a short distance of the crest of the Passchendaele ridge, while information received at G.H.Q. showed that the strain upon the German army had been far greater and that there had been a very appreciable lowering of the moral of the German troops. Haig had yet another blow in preparation. The continued bad weather and the slowness of the progress had caused the abandonment of the project of landing on the Belgian coast, and all hope of driving the Germans from the Belgian ports had gone, but there still appeared to be an opportunity of profiting from the exhaustion of the German reserves before the winter gave them a period for recovery, as it had after the battle of the Somme. A final reason for continuing the struggle was that on Oct. 24 an Austro-German attack had been launched in Italy, and at Caporetto had broken through the Italian lines. It was therefore of importance to keep up the pressure upon the Germans on the western front. So the third battle of Ypres was continued, until the ridge and village of Passchendaele were captured on Nov. 3.

A fortnight later Byng's III. Army attacked the German front opposite Cambrai. This battle opened a new era in trench warfare. One of the outstanding difficulties which the trench barrier had created was that it had hitherto eliminated one of the chief resources of generalship, surprise. The time and labour required to prepare for a great bombardment, and the accumulation of the huge stores of material of war on the selected front of battle, made it impossible to conceal intentions from the enemy. But at Cambrai these difficulties were overcome by using a great number of tanks, brought up secretly to take the place of the bombardment in breaking the enemy's defences. The attack was made upon one of the strongest parts of the Hindenburg system, but the tanks successfully broke through, and the surprise was complete. At Messines the guns had left nothing for the tanks to do, and in the third battle of Ypres they had been defeated by the mud of Flanders, but at Cambrai they came into their own. One thing alone was lacking as far as their part in the battle went. The cooperation between the tanks and the artillery in the later stages of the attack was not complete, so that numbers of tanks fell easy victims to the German guns, a lesson of which advantage was taken in 1918. Of greater importance was the fact that 6 French and 5 British divisions had been transferred to Italy to help the Italian army to stem the disaster of Caporetto, so that Haig had not the troops to complete and extend the first successes won at Cambrai. It is a typical illustration of the advantage which their central position conferred upon the Germans that several of the British divisions which would have been invaluable at Cambrai had not reached the Italian front at the time when the Austro-Germans were checked on the Piave and the battle of Caporetto came to an end. So the German counter-attacks won back a good part of the ground which Byng had gained in the first advance, and the battle of Cambrai ended on Dec. 7 in one more disappointment for the Allies.

The campaign of 1917 on the western front had been fatally hampered by the change of plan which had been made by Nivelle when he succeeded Joffre. That change had permitted Ludendorff to prepare for and carry through the retreat into the Hindenburg line, and had postponed the date of the Allied offensive from Feb. ,, the date fixed by Joffre, until April 9a delay of nine precious weeks. As Joffre, had anticipated, it had been necessary for the British army to bear the brunt of the fighting, but it would have done so under very different conditions if the Germans had been hustled back into the Hindenburg line, as they were in Sept. 1918, and if Messines had been fought at the beginning of April, and the third battle of Ypres had begun on May 30 instead of July 31. The battle of Cambrai might then have synchronized with the last offensive of the Russian army, and the combined effect might well have been such as to have saved that army from dissolution, for adequate French and British reserves would have been available in France to support Byng's attack, and the war might have ended victoriously in the autumn of 1917. As it was, the battles of 1917 showed clearly that the solution of the problems of trench warfare at which Haig and Joffre had arrived was correct. It was first necessary to exhaust the German reserves and then to strike a surprise blow or series of surprise blows. Cambrai had shown how surprise might be achieved. But all this experience, which had been purchased at great cost, had been acquired too late to be put to profit in 1917, owing to the fatal delay in opening the campaign of that year. The collapse of Russia was definite and complete, and the Germans were transferring their divisions from E. to W. as rapidly as their railways could carry them. The French divisions had since the middle of the year been gradually reduced in strength, as France had no longer the men to replace the losses in the ranks, and now Petain found himself compelled to cut down the number of his divisions. The British army was not receiving from home the men to fill the gaps caused by the bloody fighting of Passchendaele; and Haig, early in 1918, was compelled to follow the example of the French and reduce the strength of his divisions, while II British and French divisions had been removed to Italy. True, American troops had reached. France, but it was improbable that they would be able to take their place in the line of battle before the middle of the summer. In April 1917 there were in France and Belgium 64 British, 108 French and 6 Belgian divisions, or 178 in all, opposed to 128 German divisions. At the end of the year there were 59 British, 98 French and 6 Belgian divisions, a total of 163, opposed to 175 German divisions. Further, the British and French divisions were considerably weaker at the end of the year than they had been at the beginning, though this was offset, to some extent, by a corresponding reduction in the size of the German divisions. Most important of all, there were still large German reinforcements, which might amount to as many as 40 divisions, and did, in fact, amount to 32, ready to come across from the Russian front. The Allies could only obtain reinforcements in the shape of formed divisions by withdrawing troops from Palestine and Salonika, and to this their statesmen were opposed. So fate decreed that at the very time when the Allies had at last found out how to breach the trench barrier, they were thrown willy-nilly on the defensive, and had to prepare to meet the greatest effort which Germany had yet made in the west. (F. B. M.) III. German Offensive, 1918 The military situation of the Central Powers at the end of the year 1917 and the beginning of 1918 has been thus described by Ludendorff in his Memories of the War:- " Throughout the latter half of 1917 I had strained every nerve to bring about the results that had now been attained, sparing myself no more than I spared others. The western front had held, the Italian army was defeated and the Austro-Hungarian armies in Italy were inspired with new courage. The Macedonian front was holding out. In the east the armistice negotiations were finished and the way to peace lay open to the diplomatists. Negotiations at Brest Litovsk were to begin about Christmas. There was a prospect of our winning the war. Only in Asia Minor had there been any hitch, but the great events in Europe had pushed into the background." Under the influence of this hopeful outlook the German Supreme Command decided in favour of a decisive battle in the western theatre of war in the spring of 1918. This decision was justified in the first place by the collapse of Russia. Fortune had favoured the Central Powers there, apart from their own military successes. Yet these might be regarded by them as having made good fortune deserved. The problem of a " war on many fronts " had been clearly comprehended, and the principle of first completing the work that had to be done in the east had been consistently maintained. But the fortunes of war are seldom all on one side - as Hindenburg and Ludendorff were to discover. The peace negotiations in the spring of 1918 dragged on so long in Trotsky's hands that not only did a new appeal to arms, though a brief one, become necessary, but the final result was merely an " armed peace." This involved keeping strong German forces tied up in the east to secure the treaty and profit by it, and prevented the best use being made of these forces in the decisive battle of the war on French soil. But the fact remains that the chief presupposition - indeed, the indispensable one - on which the Supreme Command founded their project was the breaking up of the enemy in the east. By the end of March 44 divisions had gone to the west, followed in April and May by 15 other divisions, among which were 3 of cavalry.

It was a question of considerable importance whether, and to what extent, the participation of the Austro-Hungarian forces in the coming decisive battle would be possible. The Italian army, though not destroyed by the short autumn campaign of 1917, had been so thoroughly beaten, and was so unstrung morally, that it might reasonably be neglected as a military factor for the next few months. The offensive there had also had the important result of drawing off II French and English divisions from the French theatre of war to Italy. The duration of Italy's collapse was, of course, uncertain. It would presumably end automatically when the Central Powers dropped their menacing attitude. All the more was it important to maintain this. But it was not considered necessary to keep German fighting forces in the Italian theatre of war to this end. The moral of the AustroHungarian army had risen so markedly since the success of the last offensive that it seemed equal to carrying out this task without German support. The 6 German divisions in Italy were accordingly withdrawn during the winter, and were in the French theatre of war by the end of March. The German Supreme Command would have liked to use the Austro-Hungarian army to still greater advantage in the general scheme by bringing over a number of good fighting divisions and some heavy artillery to take a direct part in the forthcoming battles in France. This .had been agreed upon in principle at a conference between Ludendorff and General von Arz on Nov. 3. The matter was further discussed in writing up to the beginning of Jan. 1918. But in the meantime other influences were at work, opposing the designs of the two Supreme Commands. According to Gen. von Arz the Austrian Emperor and Empress were averse to sending Austrian troops to fight on French soil against the French; and the non-German nations of the Austro-Hungarian Monarchy, supported by the Social Democrats, were strongly opposed to taking part in the war in the west. The German Supreme Command, on being informed of this, sent Gen. von Cramon, their representative with the Austro-Hungarian army in the beginning of 1918, " a definite order to insist upon a binding declaration." Arz replied that no Austro-Hungarian divisions would be available until peace had been concluded with Russia and Rumania, but that artillery could be sent, though it would be deficient in munitions. This offer was accepted.' General von Cramon succeeded in persuading Hindenburg and Ludendorff, in spite of the doubts which these proceedings had aroused in them, to make another request for the cooperation of Austro-Hungarian divisions in the western theatre of war. But Cramon's intervention with General von Arz in the meantime did not succeed. General von Arz explained to him in the end in strict confidence that it would not be agreeable to those in high places if infantry were sent to the west. As a last resource now the German Supreme Command tried to stir up its ally to an attack in Italy. On March 15 1 Von Cramon, Unser OesterreichischUngarischer Verbundeter, p. 6. The number of heavy batteries sent was 46.

Hindenburg implored General von Arz to arrange for an immediate offensive by the Austro-Hungarian armies in Italy, to relieve the German army in its difficult decisive battle. After some hesitation Arz replied on March 27 that he would rally all the means at his disposal and deliver a blow against Italy at the end of May which should completely break her.

Germany was thus left to carry out the decisive battle on French soil on her own resources. There were three enemies to be reckoned with: England, France and America. The British had borne the chief burden of the fighting during the latter half of 1917, when the great battle in Flanders had towered over all other events in significance. In spite of the lost ground in the Ypres sector, and the unavoidably heavy loss in fighting power, it could in the end be registered as a German victory in so far as the English had failed to achieve their strategic aim, the destruction of the German submarine base in Flanders. For the estimation of future prospects, at least as important as this happy issue was the apparent failure of the British attack system - with its unreserved employment of masses in a battle of materiel, and its methodical conduct of the offensive as a series of thoroughly prepared attacks with objectives limited in space and, once chosen, rigidly adhered to. It was believed that this system revealed a lack of capacity for operative manoeuvring inherent in British leadership and in the British army. The tank battle at Cambrai in Nov. was looked upon as a further proof of this. Should the Germans succeed in bringing about a war of movement again in the west, their conviction was that they would prove themselves superior to the British.

Contrasted with the powerful effort that the British had put forth to gain the victory in Flanders, the ends for which their French allies were striving in the second half of 1917, after the failure of the great Aisne offensive, were apparently more modest. The local attacks to which they confined themselves at Verdun and later in the Laffaux corner turned out favourably for them, it is true, and inflicted considerable losses on the Germans. But on the whole their cautious strategy led to the deduction that the moral depression of the French nation and the army, which had set in after the battle of the Aisne, and was not hidden from the Germans, had not been overcome. Not that the German Command was likely to regard the spirit of France as permanently paralysed; on the contrary it was considered certain that the French army, in the following spring, would enter the struggle for final victory completely refreshed and stronger than before. In comparison with. England she was the militarily stronger opponent, more skilled in strategy and tactics, and more dangerous.' When it had to be decided upon which of the two the German blow was next to fall, a success over the British therefore suggested itself as being more easily and certainly obtainable. Added to this there was the consideration - decisive from a political standpoint - that the principal enemy, England, would probably be more inclined for peace when she herself had suffered a crushing defeat. In this respect the estimated value of the respective opponents had altered considerably from that made by Falkenhayn, which had passed muster two years earlier. The war in which England was fighting with her own forces on the European continent was, since the battle of the Somme, no longer a " side show." She was, on the contrary, now conducting it with all her available forces, with the utmost tenacity and with her own weapons. For the rest it might be assumed that when the German " hammer blow " fell on the one enemy, the other would not stand idly looking on, but would either directly assist his ally or proceed to a relief offensive. One hammer blow would not suffice.

A general battle was therefore launched. Ludendorff dwelt on this in making his report to the Kaiser on Feb. 13 1918 at Schloss Homburg, when he said: " The battle in the west which the year 1918 will bring presents the biggest military problem ever set before an army. France and England have grappled with it in vain for 2 years.. It must not be imagined that we are going to have another such offensive as in Galicia or Italy. It will be a stupendous struggle, beginning in one place and continuing in another, and will take up a long time." As regards the relative strengths of the two sides, the German High Command cherished no illusions as to any marked numerical superiority for their own forces. The strength of the German army in the west was brought up to 194 divisions by the addition of divisions brought up from the eastern and Italian theatres. The Entente forces in France in Feb. 1918 were estimated at 167 divisions. If the 1 i French-English divisions in Italy, which were easily available, were added to these, there remained only a slight superiority in the number of divisions on the German side. In artillery the German western army was not even quite as strong as its opponents. Ludendorff based his decision on the theory that the totals of the two fighting forces would balance each other. A factor which counted for much with the Germans was the physical condition of the army. An offensive attack best suited the character of the nation and the tradition and training of the troops. It was the more powerful form of warfare. Germany owed to it all her previous tangible successes. The ordinary citizen could see, through all his heartfelt longing for peace, that his efforts could only be rewarded when Germany had overthrown her enemies. Here and there, it is true, the same disintegrating influences which were undermining the war spirit at home could be seen at work in the army. But the influence of the good elements, which far outweighed the rest, stamped the whole as an excellent body of men. Their " will to win " was not indeed inspired purely by victory for its own sake. The attack was longed for also as a deliverance from the terrible battering which they had endured for years with resignation and with courage.

It now became supremely important to find out the precise moment at which the American forces would actively intervene. In a review of the situation drawn up by the German Supreme Command in the winter of 1917-8 it was stated: " The United States are forming an army of about 50 divisions, of which three only have as yet landed in France. One of these is at the front to be trained. The two others are in need of more training behind the front. By the spring of 1918 the American forces in France may reach a strength of about 15 divisions. The mass of the divisions will only be suitable for use on quiet fronts. Only the 3 divisions now in France may be expected to take part in a spring offensive. The corps of officers is not yet trained for war on a large scale. On these grounds the independent use of large American units in. difficult positions will be out of the question at present. The drafting of reserves and the arming and equipment of the American troops are good. Training is still inadequate. But the first regiment put in at the front fought well during a German attack, and it is therefore to be expected that the American soldier, after more training and experience, will prove himself a worthy opponent." In another calculation, made in Dec. 1917, the Supreme Command estimated the whole of the American forces that had been landed in France up to the spring at 450,000 men at most.

A larger number was not to be expected on account of the lack of shipping for transport. The mass of this army could not be ready for an attack by the spring of 1918. The value of the Americans at first would therefore lie in their power to set free EnglishFrench divisions on quiet fronts. As a matter of fact this calculation of the American strength was too generous. The total number of Americans landed in France up to the end of March 1918 has been stated by the American Secretary of War at just under 370,000. Of these only 144,000 were included in the 5 fighting divisions. The fact that the Germans did not at once realize the full extent of the increase in American transports, from April onward, in response to the urgent demands of England and France, does not actually affect their review of the situation at the end of 1917 and the beginning of 1918. When Hindenburg and Ludendorff resolved on the decisive attack they were entitled to hope for so crushing a victory over the English and the French by beginning operations early that the palm of victory could not be snatched from them again even by very considerable masses of American troops, whose intervention would only become effective in the later war of movement. Supposing - as a basis for the actual starting-point of the offensive - the Germans to be in possession of the line Doullens - Amiens by the beginning of April, which was within the realm of possibility, the annihilation of the British army might be completed within that month. Meanwhile the battle against the French would have broken out in full force. During the months of May and June a decisive defeat had to be inflicted on them also. If that succeeded, the Americans - whose troops were for the most part not yet sufficiently trained for a war of movement - would merely be swept into the general debacle. Entirely inexperienced in the leading of great masses, they would hardly change the decrees of fate.

The Germans had of course to take into account the fact that even a complete military victory by the Central Powers in 1918 on the continent would not end the war, so long as the will of Lloyd George in England was unbroken. Again, should the Entente by a great effort succeed in readjusting the situation on the continent, the war of starvation could be carried on with greater effect in proportion as the submarine menace diminished. The question was therefore whether the Central Powers, after subduing their enemies on the continent of Europe, could still hold out economically. The opening of the Ukraine had come so late that it was extremely doubtful whether its resources would be available in time to stave off the threatened economic collapse of the nations of the Quadruple Alliance. Ludendorff had no illusions upon this point, being convinced that it was absolutely essential to have his military offensive in the west accompanied and supported simultaneously by a political offensive on a large scale on the English home front. This would be directed toward bringing about the fall of Lloyd George and persuading the English nation to accept rather Lord Lansdowne's efforts in the direction of peace. It was for the political leaders to call into being and carry out a propaganda offensive of this nature. The commander-in-chief could only demand it - and this he did.

As early as the middle of Jan. 1918, Ludendorff had handed in to the Imperial Chancellor, with a strong personal recommendation, a memorandum for a German political offensive drawn up by Colonel von Haef ten. But this urgent warning to the political leaders of the State met with no response. The politicians were unmoved. Once more, on June 3 1918, Ludendorff made another passionate appeal to the Chancellor to undertake a political offensive against the English home front, again sending a memorandum by Colonel von Haeften, which this time included a detailed plan of campaign. But it was unavailing.

The question arises here whether the German Supreme Command would not have done better, at a moment when they were militarily strong, to attempt their utmost to induce the political heads of State to prepare the way for peace. Ludendorff's published memoirs show that it was never opposed to efforts aiming at an honourable peace that would safeguard the existence of the German Empire. But all the attempts in this direction made by the political leaders found the Allied Governments unresponsive, and were regarded merely as signs of internal weakening in the Central Powers. Ludendorff was to see for himself, shortly before the beginning of the great spring offensive, how little the attitude of the Entente statesmen had changed and how hopeless and damaging the renewal of any such attempt would prove. According to a credible report from a neutral country, Washington's readiness to enter into official peace negotiations depended upon the following preliminary conditions: the unconditional evacuation of northern France and Belgium; the payment of reconstruction expenses; Alsace and Lorraine to be made independent; the annulling of the treaty of Brest Litovsk, just concluded in the east; reference of all eastern questions to a peace conference to be summoned by the Entente; and a complete change of the Government system in Germany on lines to be laid down and enforced by President Wilson later. A commanderin-chief who, in the spring of 1918, should have pressed the political leaders to pave the way for peace negotiations under such conditions, without having tried for a decision on the field, would have been cursed by his fatherland.

Plan for a Break-through at St. Quentin. - Suggestions for an offensive had been made by the higher command of Prince Rupprecht of Bavaria's army group after the English offensive in Flanders had died down in the beginning of Nov. 1917. The suggestions culminated in the proposal to launch the main attack from the Armentieres-La Bassee front in the direction of Hazebrouck against the right flank and rear of the British, on the assumption that they would certainly concentrate their forces in Flanders in the coming spring for a renewed break-through XXXII.-32 operation in the direction of the German submarine base. The Allies would then obviously be in a difficult operative position. The mass of their fighting forces would be crowded up on the extreme N. wing of the whole western front. To bring up strategic reserves would take time. On their left flank and rear lay the sea. For the British in particular, strategically less trained as they were, it would not be easy to deploy their closely packed masses in the direction of the right flank and to cover their threatened communications, all the more so as a large part of their non-mobile fighting material was rigidly fixed. Tactically the prospect of breaking through the front was a good one, since the attackers would be faced by few positions technically very strong. It was, however, recognized that the country would be a difficult one in which to follow up the attack, which would be hemmed in between two commanding ranges of hills, and still further confined to the left by the La Bassee canal. It would therefore have to advance mainly in the wet Lys depression, and the ascent to the Bailleul and Hazebrouck heights would have to be carried by fighting. On account of the wet ground the operation would probably have to wait until the middle of April. General Ludendorff fully acknowledged the advantages of the proposed operation, but laid stress on the serious difficulties presented by the ground, and above all on the point that the attack, being dependent on the weather, could not be made early enough. He considered that an attack in the region of St. Quentin offered better prospects. When the line of the SommePeronne-Ham had been captured, the attack could proceed in a N.W. direction, resting its flank on the Somme, and might succeed in rolling up the British front. The higher command of Crown Prince Rupprecht's army group held to its own point of view, however, that the attack on the line Armentieres-La Bassee in the direction of Hazebrouck - code name " St. George "- was to be preferred on tactical and strategical grounds to any other offensive setting in farther to the south. Their reasons were thus stated in a memorandum of Nov. 20: " In consideration of the general political situation and the appearance on the scene of the Americans, the attack should clearly be made as soon as possible. On the other hand a decisive effect can only be attained if the objective, i.e. the mass of the British army, is securely united in massed groups in Flanders. This condition of security can only exist when the British in Flanders are preparing to attack. Our offensive can only set in when this has become a certainty. The British must attack in Flanders again in the coming year. They are forced to do so by our submarine base. We may therefore count on it with certainty and make a strategic use of the situation. Side by side with these considerations arises that of the difficult nature of the ground in the Lys depression, which makes it imperative not to attack too early. From previous experience and observation it would appear that considerable difficulty may be expected with the ground and the water up to about the middle of April.. .. The British in Flanders have similar ground conditions to consider. If they proceed to the offensive, our attack at Armentieres-Estaires would presumably be possible also. We should do best therefore - both as regards the operative effect and the state of the ground - to wait until the British attack in Flanders. It will then be necessary at first for us to give way before the enemy offensive in Flanders and - so far as possible - on the French front also. If we accept the defensive battle, we shall have to tie up such strong forces in the process that we shall not be powerful enough for an attack. We can without hesitation afford to retire as far as the line Vladsloe-W. of Roselare-Werwicq, as the submarine base will still be covered." Opposition was also raised in some quarters to an attack from the La Bassee canal front to the corner of Bullecourt, on the ground that it would come up against a strongly fortified system, and that no rapid result at the start could be expected. The enemy would, it might be assumed, gain time for bringing up his reserves from the north and the south. " Unless the circumstances change considerably in our favour there is a danger of the operations resulting in a pocket being formed in the front and not in a decision in the war of movement." Tl?.e idea of an attack delivered from the II. Army's front - code name " Michael " - was criticized as follows by the higher command of Prince Rupprecht's army group: " Decisive operations by the II. Army can only aim at a breakthrough of the enemy front and the attainment of the best possible results in the war of movement against the enemy's reserves. The Somme - an unusually powerful obstacle - will serve as a support for the left flank. The main idea of an operation on the II. Army front must therefore be to break through the enemy front in the first place, in order to protect the left flank against the French and roll up the enemy front toward the north. The operation would then be continued in the area between the Somme and the Pas de Calais against the enemy forces there, as a war of movement with a N.W. direction. The enemy would have his back to the sea. There would be a prospect of a decisive victory if the operation were pushed far enough forward. The progress of the operation in detail after the successful break-through would depend upon the measures taken by the enemy, and cannot be foreseen. Operations of this order presuppose strong forces, considerably stronger than would be necessary for the " St. George " scheme. The advantages of this scheme are that in the II. Army area operations are possible at all seasons; that the enemy positions - excepting those S. of St. Quentin - are not strongly built up and are but thinly occupied at the moment; that the question of strong enemy reserves need hardly be considered, as the British will make their attack in Flanders, and the French are not likely to make theirs against the II. Army. If the French should prepare an attack at St. Quentin, the German attack would have to set in farther to the north. The disadvantages are that the operations would lead through the ruined tract of country ` Alberich,' 1 and would involve crossing the wide area of the Somme battle, strewn with positions and craters; that the II. Army front runs from N.W. to S.E., while the direction desired for the main operation is N.W. It therefore becomes considerably more difficult to roll up the enemy front towards the N.W. after the successful break-through.

" The attack would first have to be made in a W. direction as far as the Somme, and could only later develop toward the N.W. A certain amount of time would necessarily be wasted before the operation of movement came into swing. This would give the enemy an opportunity of bringing up his reserves, the network of railways being favourable for the purpose." Although General Ludendorff and the chief of the general staff of Rupprecht's group of armies, General von Kuhl, were agreed on the main point that the offensive should be directed against the British, the chief of the general staff of the German Crown Prince's army group, Count von der Schulenburg, held at first a directly opposite view, considering an attack on the French to be the better policy. " England, with her dogged self-confidence," he said, " is not likely to end the war on account of a partial defeat of her army. She will be more inclined for peace when the power of the French is broken by a heavy defeat." Count Schulenburg's proposal was " to attack in the Argonne and to the E. of it, and simultaneously to carry out a strong attack from the St. Mihiel region in a W. direction. The objective of the attack would be Verdun, and, if possible, the destruction of the portion of the French army enveloped by the attack. The wooded, indistinct character of the deployment area would make it easier to hide the preparations for attack. The attacks would have a good chance of success if managed as a surprise. The French would never get over the loss of Verdun. If the seizure of the fortress were combined with a decisive victory over a portion of the French army, which would mean depriving the French of the possibility of a really promising offensive in 1918, the French nation and its army would be swept by a great wave of depression: "The British are certain to attack in Flanders if we attack the French at Verdun. A French offensive may be predicted with equal certainty if the British are attacked. If the Supreme Command is not in a position to execute a big attack, and at the same moment fight a defensive battle in another place, there remains the possibility of evading the enemy attack on the threatened front by retreating. This could be carried out by the VII., I. and III. Armies, and also to a limited extent in Flanders presumably, but not E. of the Argonne or on the V. Army's front." The armistice concluded with Russia on Dec. 15 made a considerable difference in the general situation. Russia no longer counted as a military factor, and the balance of power in the western theatre of war had now readjusted itself in favour of the Germans in consequence. All the reports received pointed to the conclusion that the Entente Powers would for the present limit themselves to a strategic defence, and would refrain from a great offensive until strong American forces became available. This was all the more probable because the effect of the submarine war so far could not apparently be considered so successful 1 "Alberich " was the code name for the destruction of the ground surrendered on retreat to the Siegfried positions in the spring of 1917.

as to force Great Britain to undertake the destruction of the German submarine base in Flanders at an early date. This change in the situation removed the principal presupposition on which Gen. von Kuhl's proposed offensive at Armentieres - La Bassee in the direction of Hazebrouck was based. The close massing of the British main forces in Flanders in the coming spring could not be relied on. It was far more likely that the Allies would distribute their reserves behind the front and place them in readiness round important railway junctions. It could not be denied that the operative conditions for a break-through in the St. Quentin region might also be unfavourably affected. The possibility of a French relief offensive had still to be faced.

In these circumstances Ludendorff refrained from laying down any definite direction for the attack against the British for the time being, reserving his decision until he could see how the situation developed. On one point only did he insist - the moment of the offensive must be fixed as early as possible on account of the Americans. With this in view the Supreme Command issued an order on Dec. 27 1917 for the preparation of several attacks on different parts of the front. The preparations were to be pushed forward so as to be complete by the end of March.

Count Schulenburg's original proposal - an enveloping operation at Verdun - had not yet been rejected, but was for the present only to be treated as a rough draft, the German Crown Prince's group using it as a foundation for an offensive from Champagne and the W. of Verdun on Clermont, and Duke Albrecht's group for an attack over the Meuse, S. of Verdun (code name " Castor and Pollux ").

On one point the Supreme Command was now quite clear. The offensive must not take the form of a battle of materiel, such as the Allies had over and over again attempted, invariably without results. German aims would not be furthered by an offensive which condemned the forces to months of strain. The break-through must be made to lead up to a decisive operation in the open field. This could only be done if the enemy's trench system were overrun so rapidly that the reserves he had brought up could not arrive in time to intercept the blow behind the dinted-in front. A prompt and complete success was only conceivable at the moment of the surprise. This could only be attained by observing the strictest secrecy with the troops concerned in the attacks. Each army must be convinced that the attack which it was preparing was intended to be the one actually selected. To this end all preparations for attack, in respect of laying out communications, shelters, aerodromes, etc., were to be spread over the whole army front as far as possible. The placing of the troops in readiness was to come later, and was to be undertaken outside the selected battlefield by various large groups which could be quickly and secretly formed up for deployment in different directions by train and by night marches. Another feature was to be the deception of the enemy, who was to be perplexed by the semblance of an attack on the whole army front (artillery registration and so forth), by partial actions with limited objectives, and by feints of great attacks in other places. All the attack preparations were to be carefully observed on the ground and from the air, to see that they were not attracting attention. Although it might be impossible to prevent the enemy from discovering the direction of the attack in time, it might reasonably be hoped to keep him in uncertainty as to the movement and the scope of the attack, and the form which it was to take. Success therefore depended very considerably on the most rapid execution of the attack itself.

On Jan. 24 the Supreme Command decided which of the attacks should be carried out. The choice fell on the " Michael " operation in the zone of the Xvii., Ii. and XVIII. Armies. At the same time the XVIII. Army was ordered to be transferred to the German Crown Prince's group of armies with the Omignon brook for its northern boundary.

In preparing for the Michael attack, the XVIII. Army's direction was to the N.E. of Bapaume, the II. Army's to the N. of the Omignon brook, and the XVIII. Army's to the S. of the Omignon brook on both sides of St. Quentin. The XVII. Army was to prepare a simultaneous attack S. of the Scarpe (code name " Mars South "), and the VII. Army another S. of the Oise across the front of the Crepy group (code name " Archangel "). The Michael operation was to take place about March 20, the Mars South and Archangel attacks a few days later after the regrouping of the necessary artillery and mine-throwers. The aim of the Michael attack was to be a break-through of the enemy front as far as the Somme on the line Ham-Peronne, and an advance, in conjunction with the Mars attack, on the right back of the Somme through Peronne-Arras. The Archangel attack had merely to make a diversion, and try to seize the heights E. of the Oise-Aisne canal.

The Supreme Command also gave instructions that preparations for the attack over the Lys depression at ArmentieresEstaire (St. George I.) and towards the Ypres salient (St. George II.) by the VI. and IV. Armies were to be pushed forward so as to be completed by the beginning of April. The idea of an enveloping attack at Verdun (" Castor and Pollux ") was allowed to drop, as the chief of the general staff of army detachment C. had held out only very moderate hopes of success for an offensive S. of Verdun across the Meuse. On the other hand, in the event of a great French attack in Champagne, there was some thought of letting the III. Army fall back within certain limits, while the I. Army delivered a flanking counter-attack.

Ludendorff thus held fast to his plan of directing his offensive blow against the British. In choosing the Michael instead of the St. George operation, he was influenced chiefly by the fact that it would be independent of seasons and weather conditions and could therefore be carried out earlier. The tactical attack would, moreover, fall on a particularly weak spot in the enemy's front.

The idea of a diversion on a large scale, to take place either before or simultaneously with the Michael operation, was abandoned, as it seemed necessary to use all the available forces for carrying out successfully the one great blow as planned. On the other hand the Supreme Command arranged for deceptive measures to be taken at various points on the armies' front, e.g. a lively artillery battle on the St. George and Archangel fronts; partial actions on the I. and III. Armies' fronts, particularly at Verdun; a long-range artillery battle on the Lorraine front. These commenced in some cases in the beginning of March, and were continued in the days immediately preceding the battle and until after it was well started.

On March io Hindenburg sent out a definite order fixing the morning of March 21 for the attack. According to this order the centre of gravity of the operations lay in the XVII. and II. Armies both at the beginning and during the later development. After achieving the first great tactical aim - the cutting off of the British in the Cambrai sector, the offensive was to be carried N. of the Omignon brook in the direction of Albert-Arras and beyond, where the British front was to be dislocated by the VI. Army. The XVIII. Army was only required to cover the left flank S. of the Omignon brook, and to this end to take possession of the Somme and of the Crozat canal. Its deeply echeloned right wing could be extended northward to Peronne in case of necessity. An additional order from the German Crown Prince's army group in the meantime paved the way for the idea of a new move by the XVIII. Army by contemplating the possibility of its advance over the Somme and the Crozat canal. General von Hutier, commanding the XVIII. Army, at once grasped this idea --it had probably occurred to him before - and, in a document handed in to the army group on March 15, proposed as the XVIII. Army's task " as soon as the Somme and the Crozat canal had been crossed, to draw upon itself the French reserves designed for the support of the British and beat them, and to break the communications between the British and the French.. .. The sooner the army reached the line Chaulnes-Roye, the more chance would it have of meeting the French while they were still deploying, and the better the prospect of bringing about the war of movement." The Higher Command of the army group passed on the proposal with the additional note: "the more the French counter-offensive spends itself on Rupprecht's army group, the more effectually will the proposed operation hit the French. The enemy will be quick to recognize its decisive meaning and the threat to his capital. We may, therefore, expect a very strong resistance, and on that account the operation must be launched by powerful forces." Ludendorff's attitude toward this proposal is not known. It would appear from a conversation over the telephone with General von Kuhl on March 20 that he had already weighed the possibility of accommodating himself to the idea if circumstances so shaped themselves, since he now intimated his intention of fixing the centre of gravity for the advance of the XVII. Army in the direction of St. Pol, and that of the II. Army in the direction of Doullens-Amiens - in case the XVIII. Army should meet with strong French opposition on the line BrayNoyon if not earlier.

The peace training of the German General Staff was based on the strategic and tactical principles of its former chief of many years' standing, Count Schlieffen. Although a declared champion of the Cannae idea, the Count had also definitely accepted the break-through in his reflexions and teachings, though only under the conditions of a war of movement. In choosing the spot for the break-through Count Schlieffen considered the tactical weak spot within the enemy's lines to be of the first importance as a clue. At the same time there must be the possibility of following up a successful assault and break-through by an operation in a useful direction. He therefore considered a simultaneous attack against the whole enemy front to be the best means of breaking through, as by this their forces would be tied up, the reserves engaged and the shifting of troops to another place prevented. When Ludendorff in 1918 was faced with the problem of the break-through, it was not under the conditions of a war of movement. The outward appearance of the war had fundamentally changed in the war of positions which had lasted for years. The defence was established along the whole front in a modern field position constructed according to a technique based on experience. The German attacking force was therefore confronted everywhere with the task of overcoming the opposition of the enemy in its tactically strong positions. For this he needed - at whatever point the break-through was attempted - the means employed in siege warfare, in particular a large quantity of heavy artillery and flame-throwers. For all that, however, the enemy positions were not everywhere equal in their tactical power of resistance. The ground, constructional work, density of occupation, formation of reserves, and value of the defence troops, showed many points of difference, and admitted of the sorting out of strong and weaker portions of the front. Ludendorff formed his decision quite in the spirit of Schlieffen's teaching. He, too, spied out the enemy's weakness. For the choice of the English front the leading political motive was probably primarily responsible, but the decision also happened to be in agreement with the military considerations. For although the British and French troops, in respect of their power of resistance, in tactical defence might be assessed at equal value, the British were inferior to the French in the skilled handling of masses, in the art of defensive battle and in power of strategic manoeuvre. The weak points within the 140-m. British front in question, Armentieres-La Bassee and St. Quentin, had undergone an exhaustive critical examination by the Higher Command of Rupprecht's army group. If Ludendorff decided to attack at St. Quentin it was because he would be hitting the enemy at his weakest point.

Military critics have raised the objection that Ludendorff let himself be swayed too much by tactical considerations and neglected the demands of strategy, seeing his own task from the very beginning only as a battering performance, consisting of a succession of independent hammer-blows. There is nothing to support such views. Rather is it evident that Ludendorff, here as always, was basing his strategy on the prospect of a promising tactical battle in complete accordance with the Schlieffen ideas. The attack delivered from Lens in the direction of St. Pol, as recommended by the French Gen. Buat, was extremely difficult tactically and did not offer any guarantee of a prompt initial success. But on this everything depended. Moreover, Ludendorff, in placing the attack where he did, had had visions of one great definite strategic aim, to break through the front of the British army on its S. wing, cut it off from the French, and by pressing on its right flank and attacking it from the front cause it to waver and fall to pieces and force it back to the coast. What was that but a " Cannae operation," in which a natural obstacle - the sea - took the place of Hasdrubal? A second natural obstacle - the Somme - was to serve the German left wing, advancing in deeply echeloned formation, as a protection against a French flank attack. The difficulties presented by the unavoidable traversing of the ruined Alberich area, and the Somme battlefield with its craters, were fully recognized, particularly as regards shelter and the bringing up of fresh drafts. These difficulties would diminish, however, as soon as the operation reached out into the country as yet untouched by battle, to the W. of this zone. Only in case that did not succeed would the disadvantages of this wilderness as a permanent stopping-place become evident.

On one point indeed, and that the most vital, did Ludendorff's procedure differ from Schlieffen's form of strategy. The German attack was directed, not against the whole enemy front, but against a limited section only. The perfectly obvious reason for this was that the fighting forces and battle requirements came nowhere near being sufficient for such an undertaking. The question is, whether it would have been possible and desirable to carry out the Schlieffen idea in a general sense, if not literally. The aim of the attack on the whole front was to engage all the enemy's forces, particularly his reserves, who might otherwise in due course intercept and choke off the break-through, just as it became ripe for operative development. Ludendorff saw the danger of this quite clearly, and sought to avert it by feint attacks on as many parts of the front as possible, by threats of a great attack and by partial actions on a small scale. These measures undoubtedly had a great temporary effect. Petain refused to send more than 3 divisions to Haig's hard-pressed front on March 24, on the ground that Ludendorff's main attack was to be at Reims, where the artillery battle had already begun. But the effect of these measures lasted only a short time, of course, and were limited as to material. The enemy's reserves were not absorbed, but could still, though after much delay, be moved and brought up to the decisive battlefield. It would certainly have been more in accordance with the Schlieffen idea if, alongside of these feint attacks and reaching beyond them, a serious diversion had been undertaken shortly before the main attack began. It would have had to be carried out by a strong but strictly limited number of troops, to give promise of a prompt initial success and to have a limited objective. For this purpose, so far as the British front was concerned the " St. George " operations across the Lys depression were not posble, on account of the season and weather conditions. In Flanders the circumstances were similar. There remained only the VI. Army front between the La Bassee canal and Arras. It may be questioned whether the attack, which was tactically very difficult here, would have succeeded to a sufficient extent in its object of tying up the strong enemy reserves. There were, in any case, important reasons for the decision not to make a preliminary diversion at this point in order to have a more powerful force to put in to the decisive battle itself.

On the French front things were essentially different. It was of the utmost importance that if the proposed operation were to succeed it should be secured from a strong flank attack by the French, and not be brought to a standstill by a relief offensive on a large scale. It is known that Ludendorff had intended the VII. Army to execute a diversion. But this was to take place after the great offensive had begun, and could not therefore have the effect of drawing off strong reserves of the enemy in a wrong direction and holding them fast there. The diversion would also be too closely in touch with the main attack as regards space. Several places had been proposed for a German diversion on the French front, such as the Chemin des Dames region and Champagne. Verdun was less suitable on account of the large force that would be required. The front of Duke Albrecht's army group in Alsace (Breuschtal) also seemed suitable. If the attack were made there the French reserves would be far away from the critical point of the coming decisive battle.

As far as the actual number of good attack divisions was concerned there were ample forces available. The spring offensive opened on March 21 with 62 divisions. Up to the close on April 5, 92 divisions had been put in, and even so not all the divisions available for attack purposes had been used up. This powerful mass would probably have been even more effective if it had not been used exclusively and directly for the break-through operation itself, but had been devoted in part to putting the French reserves into fetters at another place. General Buat even goes so far as to say that only by a series of diversions, delivered simultaneously or in rapid succession in different places, could the operative success of the break-through have been guaranteed. He admits however - and herein lies the point of the decision - that this method was impracticable for Ludendorff owing to lack of sufficient forces. For it was a question not only of having in readiness the required number of divisions but the massing of artillery, flame-throwers, airmen, munitions, motor columns and numerous other necessities of war. Of these the Supreme Command had not enough available for the furnishing of a powerful diversion immediately before or at the time of the main attack, if this was delivered on the scale planned. A diversion was therefore only conceivable in the event of a reduction of area or material in the main attack. But any reduction of this sort would have lessened the chance of a great and rapid victory on the spot selected, the indispensable condition of the undertaking.

The conclusions are that not only was the application of Schlieffen's theoretical ideal form for an operative breakthrough - the attack on the whole front - out of the question for Ludendorff, but the attempt to conform to the underlying idea in a modified form by executing a diversion on a large scale was not to be recommended in the spring of 1918, since the forces were insufficient. In practice the disciple was forced by existing circumstances to fall short of the master's theoretical standard laid down in time of peace.

But is it, as Buat thinks, the fact that the idea of the operative break-through was foredoomed to failure? Strategy is a system of makeshifts. This fundamental saying of Moltke's was fully appreciated by Count Schlieffen. Ludendorff had to act in accordance with it. If the enemy's reserves could not be tied up in another place and kept away from the scene of the main battle, and if, therefore, they might be expected to turn up there sooner or later, the work of the battle and the execution of the operative scheme would certainly be made more difficult. Whether the attacker, in spite of this, would have the strength to achieve not only a tactical victory, but the complete strategical success, which involved the destruction of the enemy, was the supreme question which only the god of battles himself could answer. If the reward due to his passionate efforts was withheld, he would have to moderate those efforts and reconcile himself to closing down the offensive and accepting the battle of materiel and its attendant overstraining of his forces. The first great hammerblow would then at least have had the effect of a diversion. It would have accomplished its aim according to the Schlieffen idea of operations by shattering and absorbing a large portion of the enemy fighting forces. It was then for the command to show its skill in launching, as quickly as possible, a new attack to bring about a decision with the yet unused forces from another wellselected and prepared position.

This then was the problem of the break-through as it presented itself to the German Supreme Command.

The Great Battle

As the Germans had not the means for equipping all the divisions on the western front equally, and as the essential value of the different divisions also varied for reasons connected with drafting, they were forced to limit themselves to a certain number of those which seemed most suitable for the purpose of the attack. These were the so-called mobile divisions. Altogether 52 of these divisions were made available for the beginning of the Michael operation. Besides these there were ro divisions in the line directly taking a part in the attack, making in all 62 divisions. They were distributed as follows: XVII. Army, 15 mobile divisions and 2 divisions in the line; II. Army, 15 mobile divisions and 3 divisions in the line; XVIII.

Army, 19 mobile divisions and 5 divisions in the line. The Supreme Command had kept 3 mobile divisions for the time being at its own direct disposal. These were brought up to the region of Douai shortly before the beginning of the offensive so as to be more readily available for carrying out the Mars attack, for which they were originally intended.

The bringing up of the attack formations that were destined at once for the offensive began at the end of Feb. and was completed according to plan. On March ro the bringing up of munitions began. During the last few nights the artillery, flamethrowers and divisions were formed up for deployment. On the morning of March 21 came the attack, delivered simultaneously by all three armies on the whole front - over 47 m. wide - from Croisilles to La Fere. It came as a surprise to the British III. and V. Armies. Contrary to the expectations cherished, the offensive made less progress in the first days in the case of the XVII. Army and the right wing of the II. Army than on the left wing of the II. Army and particularly in the case of the XVIII. Army, on account of the powerful British resistance. The cutting-off of the Cambrai salient failed because of the enemy's timely evacuation. In consequence of this the Supreme Command on the afternoon of March 22 ordered the XVII. Army to extend the success of the II. Army by an attack aimed particularly in the direction of Bapaume, and to prepare for the attack on both sides of the Scarpe (Mars) with strong forces to the N. of it.

When it became evident, on the morning of March 23, that the XVIII. Army and the left wing of the II. Army were advancing unchecked towards the W. and would in all probability reach their nearest objective, the Somme, on that very day, an order was issued for continuing the operation, as soon as the line Bapaume - Peronne - Ham should have been won. " The XVII. Army will attack with strong pressure in the directions Arras - St. Pol, the left wing in the direction of Miraumont. The II. Army will take the direction Miraumont - Lihons; the XVIII. Army will take the direction Chaulnes - Noyon and will send strong advanced troops through Ham." The three divisions held in reserve were now given to the XVII. Army.

Through this order the whole operation was pushed a long way to the left. The XVIII. Army, which was originally to have extended its front northward to Peronne on reaching the Somme, thereby releasing forces from the II. Army to carry on the attack N. of the river, had now instead to cross the Somme and advance its right wing in a slightly S.W. direction toward Chaulnes. There were thus portions of the II. Army left S. of the Somme as well as the XVIII. Army. The Somme was therefore no longer used as a support against a French flank attack, for the offensive now took a N. direction along the whole front, N. and S. of the river, with the operative aim of separating British and French.

This most fateful decision of March 23 arose from the tactical consideration of exploiting the XVIII. Army's comparatively easy success to the benefit of the general battle situation, by making a rapid forward push. Had the XVIII. Army stopped at the Somme and the Crozat canal, as was originally intended, extending with its right wing only northward to Peronne, its fine initial success would have had no effect on the advance of the attack farther to the N., which had up till then not quite come up to expectation. If, on the other hand, it had carried its attack across the Somme and the canal toward the W., in conjunction with the S. wing of the II. Army, the enemy, who was still holding out against the XVII. Army and the N. wing of the II. Army, would have been threatened on his right flank. From the strategical standpoint the decision was even more difficult and more vital. The fundamental idea of the Michael operation had from the first always been that of beating the British and the British only. The French were only to be held off from intervention by flank action. To this end the whole of the XVII. and II. Armies were to find a field for operations to the N. of the Somme. The reinforcements sent by the Supreme Command were also to follow in this direction, being mainly disposed in echelon in rear of the II. Army's left wing for the purpose of taking over the flank protection down the Somme from Peronne. At a later stage there were probably some portions of the XVIII. Army similarly engaged N. of the Somme. This whole strategical idea would have fallen to pieces if the strong natural obstacle of the Somme had been relinquished as a support at this point. It was clear that the left wing of the armies, entrusted with the offensive solution of its task on the far side of the Somme and the canal, would very soon not only draw upon itself considerable enemy forces, but would gradually have to prepare for a counter-offensive steadily increasing in strength. It had therefore to be reinforced from the reserves, which thus, as well as a portion of the II. Army, were no longer available for use in the direction followed by the main operation. But despite these apparent disadvantages the decision must be approved from the strategical standpoint also. The position of the XVII. Army and the north wing of the II. Army, as it was on the morning of March 23, made it doubtful whether the strong enemy resistance here could be broken in time to arrive at operations in the open field at all before the arrival of enemy reinforcements. There was a danger that not only the British but the French might throw strong forces on to the battlefield N. of the Somme, and so block the break-through in or close behind the British trench system. This was made easier by the contact with the positions E. of Arras, which had been maintained. The Somme - at Peronne or farther to the W. - would then serve the defender just as well for a safe support as the attacker had hoped it would serve him. It would also form an excellent obstacle for the front farther on up to Ham. As the direct intervention of the French, according to the way in which things turned out, had to be reckoned with, everything depended on hindering them from carrying it out systematically. The French must be caught up into the whirlpool of destruction. But this could only be done by forcing their Bray - Noyon front and attacking impetuously in the open.

In the days that followed Gen. Ludendorff held stubbornly to his operative aim of separating the British and French. The distribution of the reserves that had been brought up later was organized accordingly. The centre of gravity of the XVII. Army's advance, originally directed toward St. Pol, was now, on March 24, shifted more to the S. toward Doullens. The course of the battle on the whole front, up to March 25 inclusive, justified the expectation of achieving its ambitious aim. As the XVII. Army had pushed its way through the whole system of enemy positions and had advanced with its S.wing to beyond the Ancre, it too began to operate in the open field. The situation was now such as to warrant the attempt to dislodge the enemy front, both at Arras and farther to the N., by frontal attacks; and the Mars attack on both sides of the Scarpe was fixed for March 28.

The first faint doubts as to the possibility of carrying out the main operation to its full extent might have been aroused by the experiences of the XVII. Army on March 26. Its S. wing, on which everything depended, gained very little ground beyond the Ancre. As, however, in the meantime the II. Army's right wing had achieved the difficult crossing at Albert, there was hope that the XVII. Army's advance would also quickly get into its stride again. On all the rest of the front the brilliant progress of the offensive so far, particularly the impetuous forward push of the II. Army in the direction of Amiens and that of the XVIII. Army toward Montdidier, raised expectations of a continuation full of promise. Ludendorff proposed to bring about the separation of his opponents by a gradual concentration of the II. and XVIII. Armies against the French, and to this end the Somme below and at Amiens had to be reached and also the Avre. But the far-reaching aims of the Supreme Command were not to be realized. On March 27 the XVII. Army's offensive came to a standstill, and the next day brought the failure of the Mars attack on both sides of the Scarpe. At this point, therefore, the operation against the British was finally abandoned. Ludendorff decided to attack their front as soon as possible in a different place, and ordered the immediate preparation of an attack on the VI. Army's right wing on the Lys front in the direction of Hazebrouck. It would, however, probably be 8 or ro days before this could begin. Otherwise the continuation of the Michael operation, as it had turned out, seemed to promise success only in the direction where there was still movement, that is, on the S. wing of the II. Army and with the XVIII. Army. As the intervention of the French so far gave the impression of being precipitated, it was concluded that the opponent had not yet fully organized his forces. The point was to keep him from doing so now. The cooperation of the II. Army's N. wing, now held fast on the Ancre, in the forward wheel toward the Somme below Amiens, could indeed no longer be counted upon. It seemed, therefore, all the more urgent to get possession of Amiens, the strategically important railway junction, by the quickest means, and also to cross the Avre. The centre of gravity of the offensive was therefore laid exclusively on the inner wings of the II. and XVIII. Armies in the next few days, all reserves being switched off in that direction. However, no real progress could now be made in the direction of Amiens. A last attempt on April 4 broke down before the enemy resistance, which had visibly increased.

The close of the Great Battle left the Germans in possession of a narrow salient stretching far out toward Amiens. This position had its dangers, which necessitated perpetual watching, and kept strong forces tied to the spot. On April 24 the II. Army tried to improve its positions between the Somme and the Avre by a partial attack, which after a passing success at VillersBretonneux ended in a recoil. After this for a long time no important battle actions took place on this section of the front.

The Michael operation had not achieved the full operative success, but had nevertheless dealt the British a heavy blow and crippled their fighting power for a long time to come. More than 40 British divisions were seriously affected, and also about 20 divisions of the French army which had been drawn in.

In the light of subsequent criticism, the question arises whether General Ludendorff's leading idea would not have had more chance of being realized if the decision of March 26 had limited the objectives aimed at in one direction or the other. Persistence in the double intention of dividing the opponents and simultaneously dislodging the British at and N. of Arras by frontal shock had the effect of dissipating the still available attack energy of the reserves, and made it impossible to focus on a single object the largest possible part of the forces still capable of a great effort. In view of the general outlook the only limitation worth considering was one which would have facilitated the progress of the Michael operation by temporarily renouncing the Mars attack and the proposed frontal attacks farther N. which went with it. The danger then would have been that the British, not being threatened from the front, would throw all their available reserves on to the battle-field as it now stood, and also release .forces from their front for the same purpose. It would be the task of the XVII. Army and the. portion of the II. Army fighting on the Ancre to continue their attacks and so draw these forces upon themselves, preventing a flanking attack against the German main operation in the direction of Amiens. It was for the moment less important to gain much ground in the direction of Doullens, provided that the decisive blow on Amiens on both sides of the Somme were kept going. Supposing that the 5 divisions which had been put into the attack on both sides of the Scarpe had been used in the continuation of the Michael operation with the XVII. Army and on the right wing of the II. Army from March 27 on, it would have been possible by this time to shift a number of reserves toward the S. to the decisive wing. A portion of these reserves did gradually find their way to the wing S. of the Somme a few days later, together with a number of divisions which had been engaged in the previous fighting. But they arrived too late, and the offensive had meanwhile come to a standstill. The conclusion is that events might have shaped themselves more favourably, from an operative standpoint, if the decision of March 26 had limited itself to the separation idea, laying increased stress on this at the cost of the intention to break up the whole British front at the same time.

Finally, there is no doubt that the Germans, by their last attempt to get possession of Amiens, put too great a strain on their available forces. However strong the grounds for this, it should have been of supreme importance to the Germans to avoid the wearing effect on their fighting force of a battle of material. This would have been easier to accomplish if a decision had been made by the end of March to close down the Michael operation. Instead, the attack on April 4 placed the inner wings of the II. and XVIII. Armies in the salient over against Amiens on both sides of the Avre in so difficult a fighting position that, whatever the result, the reserve strength which was still coming in had to be committed and was used up.

German Attack

April, May, and June. - Closely connected with the Michael offensive, which came to a standstill on April 5, was the VII. Army's Archangel attack which followed. Between April 6-8 the right wing of this army threw back the enemy from its positions S. of the Oise through Amigny and Coucy le Château to beyond the Ailette. The ground gained made an improvement in the difficult rearward communications of the XVIII. Army's left wing. Following immediately upon this came the resumption of the great operations by the Georgette attack on the Lys front. General Ludendorff had had this attack in view since the end of March, and had prepared it at first as a diversion only. When it became clear in the beginning of April that the Michael offensive would not lead to a complete operative success, the Georgette attack was extended in its scope and aims to an operation for forcing a decision. It was proposed to break through the British-Portuguese front in the direction of Hazebrouck-St. Omer and then to continue the operation through St. Omer-Bethune and as far to the S. as possible. The VI. Army was to attack on the front Armentieres and the La Bassee canal with its centre of gravity on Hazebrouck; the left wing at first only to wheel in on the general line Aire-BethuneLa Bassee canal; the centre to push through toward Hazebrouck and the heights W. of it and to seize the canal crossings between St. Omer and Aire; the right wing to take possession of the commanding heights to the E. and S. of Godewaersvelde and then to take the direction of the barrier of heights at Cassel. The IV. Army, attacking one day later from the line HollebekeFrelinghem, was to attach itself with a strong left wing. Armentieres was to fall by envelopment. Beyond all these there was a proposal for the IV. Army to attack in Flanders from the Houthoulst Forest in the direction of Poperinghe, with the object of cutting off the Ypres salient.

The enemy's situation seemed favourable to the Germans. The British had hardly any more fresh reserves to draw on, so that the only reserve to be considered consisted of divisions worn out by fighting. The Portuguese stationed on the Lys were not credited with any great power of resistance.

Seventeen divisions from the VI. Army and 4 from the IV. Army were placed in readiness, and the necessary artillery was brought up, some of it being obtained by regrouping from the Michael front. In the course of the operation 14 more divisions, mostly from the zones of the XVII. and II. Armies, were put in. The attack by the VI. Army, beginning on April 9, took the opponent at first by surprise. On this same day the whole stretch of the Lys at Sailly was conquered. But the battles of the next few days, though successful, were obstinate and costly, and already it appeared doubtful whether the attack would develop into a break-through. The left wing had not succeeded in taking Festubert and Givenchy or in reaching the canal. On the other hand there could be no question of stopping the offensive immediately, as the inner wings of the VI. and IV. Armies were still fighting in a difficult tactical position. Their position improved, however, with the seizure of the Neuve Eglise and Baillcul heights, but in general the gains of ground were only local. The advantages of the initial surprise were forfeited and the opponent found time to organize his resistance more and more thoroughly. The hope of being able to set the interrupted operation in motion again by a surprise assault on the Belgian front, delivered by the IV. Army from the Houthoulst Forest, vanished when it became known on April 16 that the enemy was eluding the carefully built-up attack from the Ypres salient by slipping away behind the Steenbeek. The IV. Army higher command considered that on account of the difficult ground the attack across the Steenbeek had no chance of success unless it were newly organized, and postponed the execution from day to day. As since April r7 the French had been established at Wytschaete the arrival of further French reinforcements had to be reckoned with.

On April 20, therefore, General Ludendorff ordered the beginning of the offensive. With a view merely to improving the tactical situation of the inner wings of the VI. and IV. Armies the attack on Alt. Kemmel was carried out on April 25. The piecemeal capture of Festubert and Givenchy did not succeed.

On May r Ludendorff came to the decision to place Rupprecht's army group and also the XVIII. Army on the defensive for the time being. The Georgette operation had, apart from destroying the Portuguese, undoubtedly inflicted another heavy blow on the British army. Its losses in the defeats of March and April might be estimated at not less than half-a-million men. The fact that Foch was forced to send about 18 French divisions and 6 cavalry divisions to Flanders suggested that the British alone were not in a position to resist the pressure put upon them. It was also an important point for the German Supreme Command that it had for many weeks had the lead in the western theatre of war and had forced the opponent to stand on the defensive. Yet it could not record any operative success in this new place. Then, too, the Michael and Georgette offensives had used up a large number of forces-113 divisions - and this fact weighed heavily. Taken in conjunction with the difficulties about drafts there was no doubt that the balance of forces was gradually becoming unfavourable to the Germans.

It is indeed questionable whether the German command had it in its power to raise this strategically unsatisfactory result to the level of a striking success. With the forces actually available, and those that were put in, it would hardly have been possible, even if certain errors in the subordinate command had been avoided. The greater part of the divisions used did not belong to the mobile divisions, which had been trained and equipped for the attack, and others were worn out by fighting. There was, therefore, a certain lack of the necessary freshness and tenacity in attack. If the German Supreme Command had decided at the end of March to stop the Michael offensive and desist for the time being from the attack on the Archangel front, there would have been fourteen more unused divisions available at the beginning of the Georgette attack. With this additional strength considerably more pressure could have been exercised, particularly by the IV. Army, to the N. of Armentieres and N.E. of Ypres.

With the situation as it stood at present the difficulties in the way of forcing the decision of the war before the Americans made themselves felt were growing. In. spite of this Ludendorff remained unshaken in his aim, clearly recognizing that the Germans could now only achieve a success through their own initiative and by working against time. All the clever advice that subsequent criticism felt obliged to offer Ludendorff is met by the objection that by none of it could the victory of the Germans have been achieved. If the Allies were now allowed time, and were able at a self-chosen moment to use their fighting force, with its ever-growing superiority in personnel and material, f+or their own final blow, the war might be given up as lost at once.

The necessary forces were lacking for an immediate fresh German offensive. During the next few weeks it was imperative that the mobile divisions, some of which had been overtaxed, should be allowed to rest and freshen up again. By May 27 the German reserves had been brought up to 8r divisions again, exclusive of the transport movement from the east. Of these 58 had been resting.

The direction which the operations were to take had now to be decided. The French and British now formed a united front, and the former plan of beating each separately was no longer in question. At the end of April there were from ro to 12 French infantry divisions and 6 cavalry divisions established in Flanders. In front of the German XVII. Army at Doullens was the French X. Army with from 5 to 6 divisions. At Amiens and S. of it stood a group of from 12 to 14 fresh French divisions. As Foch had also released about 20 divisions by putting in territorials and Americans and economizing on numerous sectors, he now had at his disposal a reserve of over 60 French divisions. About half of these he kept to defend the coast and at Amiens, the other half being apparently distributed in readiness before the fronts of the remaining army groups. An offensive to force a decision against those sectors of the front held almost exclusively by the French from the Somme to the Swiss frontier promised the Germans a far-reaching operative success, at whatever point it might be attempted. The one sharply-defined objective in this connexion was Paris. But on the way there an encounter with the French army, prepared to defend it to theuttermost, was certain; and a defensive battle for them would have various advantages. On the other hand if a German attack should sooner or later find itself stuck fast on the way to Paris or in Champagne - as might almost certainly be predicted - the Germans would be in an unfavourable position for operating, with their line bent more or less far forward toward the S.W. or S. There seemed more prospect of success in resuming the offensive on the Michael - or Flanders - front, where the objectives were not fixed so far away. After all, the Germans had covered half the distance from St. Quentin to Abbeville in March well within a week. If they could succeed in doing this again with the same bulk and expansion, they would throw the enemy forces opposing them into the sea. But even with less success at first they might hope so to cramp their opponent's freedom of movement that his power of prolonged resistance would weaken, and he would be completely crushed by renewed hammer-like blows. The German Supreme Command therefore sought to gain their strategical aim as before in an attack on the northern part of the enemy's front. Clear on this point, it was again confronted, as in the spring offensive, with the choice between carrying out the operations in Flanders or farther S. against the Arras - Amiens front.

Acting on the suggestion of the Higher Command of Prince Rupprecht's army group, General Ludendorff in the beginning of May decided on the Flanders attack. The determining factor was the knowledge that an attack from the Michael front in the direction of Doullens would tactically be extremely difficult, depending for its success on a simultaneous side-attack from the region of Bethune, for which the forces were not adequate. But as there were still strong French reserves in Flanders at the time, Ludendorff decided not to lead the attack against PoperingheHazebrouck until a diversion in another place had drawn off considerable portions of this reserve from the Flanders front, leaving it weakened in consequence. At the suggestion of the Higher Command of the German Crown Prince's army group, a diversion - offensive within a limited area from the VII. Army front on the Chemin des Dames across the Aisne as far as the Vesle - code name " Blucher," was given preference over an attack by the I. Army in Champagne, E. of Reims, because its clearly defined aim offered the promise of a line suited both to a prolonged defence and a continuation of the attack. The VII. Army's task was limited to carrying the offensive over the Aisne sector on both sides of Soissons and over the Vesle as far as the heights to the S.W. and S.E. of Soissons and S. of Fismes, while the right wing of the I. Army was to accompany the attack westwards past Reims and nearly up to the Ardre. There was also an idea of letting the XVIII. Army push forward its left wing across the Oise to split up the counter-offensive.

By choosing the Blucher attack Ludendorff was again faithful to his principles in selecting a markedly weak spot in the enemy's front. There were at this time in the front line only 6 French divisions - of which 2 were worn out and r weakened by illness - and 2 British divisions which had been defeated in March and April. As regards reserves, there were supposed to be 2 fresh French divisions and 2 that had previously been beaten in the region between Compiegne and Reims. As against these the VII. Army had 29 divisions, the I. Army 4 divisions at its disposal. The Supreme Command reserved the right of taking back strong artillery forces from Rupprecht's army group and the powerful fighting divisions drawn back behind their front. This time, too, the attack was to break out as a complete surprise, and be relieved by opportune feints on different parts of the other armies' fronts. Owing to the comprehensive preparations the Blucher attack could not be launched until nearly the end of May.

It began on May 27 and succeeded beyond all hopes. By the evening the VII. Army's centre had reached the Vesle on both sides of Fismes, the wings holding back somewhat. The morning of the 28th brought an emphatic reminder from Ludendorff that the object was to get possession as quickly as possible of the high ground W. of Braisne, S. of Fismes and Bazoches and N.W. of Reims. The right wing was to advance by means of a sharp attack to a line on the high ground between the Oise and Aisne canal and the Aisne in a W. direction. The successes of the 28th enabled fresh orders to be given for advancing the objectives of the centre and left wing to Fere en Tardenois-the heights S. of Coulanges-the S. front of Reims. If the opponent evacuated the territory between the Aisne and the Oise, the XVIII. Army was to draw up forces on the S. bank of the Oise about Noyon and to gain ground in the direction of Compiegne. On May 29 the successes on the VII. Army's front were spreading rapidly, and orders were given at noon for the attack by the left wing of the XVIII. Army and the VII. and I. Armies to be continued in the direction of Compiegne-Dormans-Epernay and the block of hills between the Vesle and the Marne, S. of Reims, taken as a protection against Chalons. The progress of the VII. Army corresponded to these instructions. The I. Army, which had had difficulties to contend with, received an order on the morning of May 30 to reinforce its right wing from the centre, and shoot out its fighting line to the S. and S.W., thereby facilitating the envelopment of Reims. The VII. Army meanwhile had reached the Marne on the 30th with its centre, and on the following day gained a good deal of ground in the direction of Villers Cotterets, but the two wings of the attack did not seem able to make any further progress. The Supreme Command was for the moment inclined to send portions of the VII. Army over the Marne to push forward on Epernay, with the idea of getting the attack of the I. Army into swing again. But this scheme was dropped in consideration of the state of the troops and the strength of the enemy's resistance. The I. Army was to have a rest, and then be required only to undertake partial actions with limited objectives and to capture Reims. The Supreme Command was now anxious to get the centre of gravity fixed on the W. front of the VII. Army in the direction of Villers Cotterets and La Ferte Milon, in order to attract powerful French forces. Reenforcements were therefore sent up from the zones of the other army groups. But the VII. Army attack in a W. direction made no more progress to speak of in the beginning of June, as the French had established themselves here in great force by hurrying fresh troops on to the scene.

The continuation of the offensive now depended on the progress of the attack launched on June 9 from the S. front of the XVIII. Army on the Matz brook and the Aisne above Compiegne (code name, " Gneisenau "). But this attack by the XVIII. Army did not have the anticipated success, which would have justified the immediate opening of the Flanders offensive, but came to a standstill on April ii. A blow delivered by the VII. Army against Villers Cotterets to the S.W. of Soissons on the following day also failed, and as the immediate continuance of partial actions by the I. Army did not look like succeeding either, a lull set in along the whole new front of the German Crown Prince's army group in the middle of June.

The Blucher attack was not looked upon as an operation to force a decision, but rather as a diversion. The fact that the original limited objectives were exceeded in consequence of the unexpectedly favourable course taken by the attack is not incompatible with the leading idea, which aimed at holding and destroying the greatest possible number of the enemy's forces. This aim was fulfilled. The total number of the French engaged in the defence against the Blucher offensive was estimated in the beginning of June at over 40 infantry and 3 cavalry divisions. On the Marne the wavering French lines were only saved by fresh American divisions. This time the Germans spared their troops by avoiding costly independent battles of only local importance. At the same time the development of the tactical success should not have brought about an unfavourable operative situation at the close of the offensive. But this was exactly what happened. As the wings had not succeeded in getting the region of Compiegne and Reims, with its hilly country, under their control, a new sack-like bulge had been formed in the German position, with the flanks bent far back, to maintain which strong new forces would have to be put in. The rearward communications of the VII. Army were particularly unsatisfactory owing to the lack of railways. This unfavourable operative situation could perhaps have been avoided if from the beginning strong pressure had been used to push forward the right wing along the Oise to the Marne, and the I. Army to Reims, at the cost of some of the excessive amount of ground gained in the centre. Later attempts to work up the operation from the centre were impaired by the gradually failing fighting strength of the troops. According to Ludendorff a subordinate command also failed to carry out a swift and powerful advance through Soissons. The advance of the VII. Army's right wing along the Oise, which began later, was inadequately provided with the means of attack, and the Gneisenau attack by the XVIII. Army W. of the Oise, which followed, was apparently even less well prepared; neither could retrieve the results which could presumably have been obtained without much effort at the very beginning of the offensive, had the forces on the VII. Army's right wing been differently grouped. The disadvantageous strategic situation at the close of the attack corresponded with the tactically difficult fighting position in which the tired-out German divisions were confronted with the active defence offered henceforth by the French and Americans.

The German Of f July. - The effect of the Blucher attack very soon made itself felt on the front of Rupprecht's army group, where no little relief was felt. The enemy's fighting activity diminished, and a portion of French reserves intended for the support of the British vanished from the scene. All the same, this degree of slackness on the enemy's part did not seem to the German Supreme Command to warrant the carrying out of the planned and prepared attack on the Flanders front (code name " Hagen ") for the present. They hoped first, by carrying out another of the diversions on the front of the German Crown Prince's army group, to rob the British of the last of their French support, and even in certain circumstances to force them to send direct help themselves to their hard-pressed ally.

On June 14 Ludendorff arranged for the German Crown Prince's army group to attack with the VII. Army across the Marne, E. of Chateau-Thierry and between the Marne and Reims (code name " Marneschutz "), and with the I. and III. Armies between Reims and the Argonne (code name " Reims "). July io was the date aimed at for the beginning of the offensive. About io days later, after a rapid regrouping of the necessary artillery and so on, the Hagen attack was to be executed by Rupprecht's army group in Flanders. Ludendorff in his Memoirs gives the reasons for selecting new points of attack on the French front as follows: " The greater part of the enemy's reserves were to be found within the curve formed by the XVIII. and VII. Armies in the direction of Paris, only weak forces being left between ChâteauThierry and Verdun. The Supreme Command intended this time also to attack the enemy where he was weakest." The underlying idea of the VII. Army's far from simple operation on both sides of the Marne toward Epernay was inspired by the desire to escape from the tactically constrained position brought about by the pocket on the Marne. By flattening out the left flank of the pocket by eastward pressure, not only would the army's rearward communications be widened, but the danger which perpetually threatened the right flank would be diminished. While the necessity of supporting this flank on the sector from Villers Cotterets to the Marne during the attack was pointed out to the VII. Army, the forces needed for such support were not placed at its disposal. It is not by any means clear on what grounds the decision was taken to extend the diversion very considerably toward the east by calling in the III. Army. It was probably in the hope mainly of splitting up the enemy defences along as wide a front as possible. General Buat points out that the offensive, if successful, would have opened up brilliant strategic possibilities, such as extending the successful advance in the direction of Bar le Duc and rolling up the whole defensive position in the Argonne and toward Verdun.

A shortening of the German position from Chateau-Thierry through Chalons to St. Mihiel would have been infinitely valuable.

On the other hand the 44-m. front required such large forces to occupy it that, contrary to the original intention, it was necessary to fall back on some of the divisions set aside for the Hagen attack in rear of the Rupprecht group's front. The inevitable consequence was the postponement of the date provisionally fixed for the Flanders attack to the beginning of August. Also the preparations for the combined Marneschutz - Reims attack proved so extensive and took up so much time that the date had to be put forward to July 15 at the cost of preserving secrecy. Once more the execution depended upon surprise.

Through carelessness and treachery the German plans became known to Foch to a great extent during the first half of July. He had found time for adequate preparation of his defence, which was skilfully adapted to meet the German conduct of the attack. In this way the offensive in the case of the I. and III. Armies came to a standstill everywhere in front of the French main position. The VII. Army succeeded, after a successful crossing of the Marne, in shattering the main line of defence in several places. But here, too, far-reaching results were unobtainable. On July 15 the German Supreme Command ordered the cessation of the attack for the III. Army, and on the 16th refused to allow the I. Army to continue after an attempt had proved vain. The VII. Army continued to advance with great difficulty until July 17 in some places. The offensive blow had in fact completely failed, because it fell upon an enemy who was not taken by surprise but was prepared to offer a resistance as obstinate as it was skilled.

Ludendorff lost no time in drawing his conclusions from the unexpected turn in the general situation, and immediately ordered the withdrawal of the right attacking wing of the VII. Army behind the Marne. He was determined to regroup his forces with all possible speed for the Hagen attack in Flanders. Hardly had the necessary steps been taken, however, when, on the morning of July 18, Foch's flank attack fell on the insufficiently supported W. front of the VII. and IX. Armies. Owing to the disproportionate initial success of this attack, the execution of the Hagen attack had to give place to the pressing necessity of putting fresh forces into the VII. Army and bringing it back behind the Vesle. This was the turning point in the conduct of the war in 1918, and at the same time in the whole of the World War. The German offensive had met first with a sudden interruption, then with its final close, owing to the initiative of the opponent. From this time onward the German Supreme Army Command was subject to the strategical law of the enemy.

If we pass in review the many plans of attack entertained by the German Supreme Command - the Flanders attack, for instance, was, if circumstances permitted, to be followed by an offensive on Paris or Amiens - the impression is easily formed that the leader of the German operations was no longer pursuing a definite operative aim, as at the beginning of the spring offensive and for some time afterward, but had as his sole object the shattering of the enemy by independent hammer-blows delivered one after another at tactically advantageous points. But all these hammer-blows represented not the end but the means by which the final decisive operation should be prepared, facilitated and brought to a successful issue with the highest degree of certainty and the least effort. One misgiving undeniably arises in considering this method. The limitation of the available attacking forces and fighting material made it impossible to make the individual blows follow each other so quickly that the enemy would have no time to recover between each, to a certain extent, to make good the losses he had suffered, to prepare his defence against fresh blows, or even recover so far as to proceed to counter blows himself. The question is therefore whether a different procedure, after the close of the May - June offensive at latest, might have had more chance.

Possibly the necessary forces for the Hagen attack in Flanders, which was to have been the decisive operation, might have been mobilized by the middle of July if it had been decided to sacrifice the whole of the territorial gains, which were exhausting and difficult to maintain permanently, made up till then in the Michael and Blucher offensives, by a retreat movement on a large scale by which the fronts of the XVII., II., XVIII. and VII. Armies would return to their starting positions. With these forces a new and overwhelming surprise attack in a totally different form might have been sprung on the enemy, which would prevent his throwing his released forces straight on to the Flanders front and there avoiding a crash. What Hindenburg and Ludendorff had achieved with unexampled skill in Nov. 1914, immediately after the great retiring movement through S. Poland, by advancing from a newly selected position to a flank attack on the right wing of the Russian main army, could have been repeated in the summer of 1918 on French-Belgian soil when the general situation was strategically favourable.

In view of the complete failure from the outset of the attempt to realize their daring and far-reaching projects, there has been too much of a tendency to accuse the German Supreme Command of misjudging the situation, overestimating the working value of their own instrument of war and underestimating the enemy. Yet it was under no illusions as to the difficulties that it was essential to overcome. It was clearly recognized that time was now, more than ever before, a factor on the side of their enemies. The British had regained their strength, the French were not yet sufficiently shaken, and the Americans were bringing unexpectedly large masses of troops to France with amazing rapidity. To set against these factors the German commander-in-chief could count on no more reinforcements of any kind. No additional force for attacking purposes could be extracted from the divisions left in Russia and Rumania, which had already given up all their men under the age of 32. The drafting reserve from home was becoming more and more meagre, bringing in only 28,000 men for the infantry in June as against 44,000 in May. It was composed mainly of returned lightly wounded men, and those who had recovered from sickness. The average strength of the German battalion in the field had sunk by the middle of July from the original 850 men to 660. The actual front strength was even considerably lower than this number. The process of disintegration within the army caused by the prevalence of revolutionarily minded elements, did not escape the notice of the Supreme Command. Taking all in all, there was no doubt that the fighting value of the troops was no longer on the same level as at the beginning of the spring battle.

In spite of this there was no compelling reason to doubt the adequate striking power of the jagged though not blunted weapon of the German army, so long as its intentions and plans were kept absolutely secret, as before, and the blow was once again aimed at one of the enemy's weak spots.

Unfortunately for the Germans, their method of attack had now lost its magic effect on the enemy, who had found times and means to organize his resistance accordingly. The July offensive had been made with the greatest circumspection and thoroughness of preparation, just as before; and the attacking forces were certainly not deficient in courage or endurance. But an essential factor in the success of the undertaking was lacking - no commander-in-chief can do without luck. Hitherto the luck had been generally with Ludendorff. Now, at the decisive moment of the World War, it deserted him, and went over to his opponent - his equal in determination and will-power, and now his superior too in strength. If Ludendorff had had the luck at Reims in July 1918 which attended him at Tannenberg, history might have acclaimed him the greatest commander of all time, because he had remained true in spite of everything to himself and his belief in his star. But this was not to be. (W. F.) IV. THE Allied Offensives Of 1918 The fourth and last phase of the war was ushered in by the failure of the German Champagne-Marne attack of July 15 and the success of the Allied counter-attack at Soissons on July 18, the results and far-reaching consequences of which came as a surprise to German and Allied commanders alike. Von Hertling, the German Chancellor, has written: " We expected grave events in Paris for the end of July. That was on the 15th. On the 18th even the most optimistic among us understood that all was lost." To make clear the significance of that statement it is necessary to review briefly the condition of the opposing armies previous to the Soissons attack.

It was indeed well known to both sides that the German army was nearing the end of its offensive strength, but just how nearly German moral had been drained neither side fully appreciated until later for, superficially, it was good. The French army, on the other hand, was well known to be at its lowest ebb of moral. The French soldier, since April 1917, had ceased to be a war machine unit who could be depended on blindly to follow his leader and had assumed a certain independent thinking role. Discipline had slackened and orders to attack or defend stubbornly had lost their force unless the soldier wished to attack or defend stubbornly. Petain considered, early in July, that, although he had a number of rested divisions in reserve, he had not a single division which could be relied upon to push home and exploit an attack successfully. Of such as he had the Moroccan Div. was rated the best. The superiority of the British moral had been offset by the numerical weakness of their battalions, and, although they were holding their own doggedly, their confidence in their allies had suffered a severe strain and grew still further impaired as the lower units became intermingled. The American army was, of course, as yet an undetermined factor.

It was therefore with a certain amount of reasoned justification that Ludendorff, aware of these conditions, to which moreover he added an amazing underestimate of the strength of the American effectives in France, conceived that one more push directed against the French army would put it into headlong flight and thus pave the way for a similar stroke against the British. The adverse factors in the Champagne-Marne project were, first, that the method of attack, the so-called " Riga model," now lacked the element of surprise, since the methods of concentration for it were now too well known to make concealment Epossible, and its success was further discounted because Petain had discovered the tactical means of effectively stopping such an attack; second, that Ludendorff had overcentralized his command. No army group commander or army commander was called on or permitted to exercise judgment or decision; he could only carry out the plans devised by Ludendorff and his staff by methods similarly devised and prescribed. In the lower ranks of officers this benumbing influence was, if anything, still more strongly felt. Meanwhile, synchronously with his success in thus centring the power of decision in his own hands, Ludendorff had become preoccupied with a multiplicity of problems which had no immediate relation to the conduct of the army on the western front. Germany's allies, her own internal questions, the Russian and Near East situations, were all constantly taking his time and distracting his attention from the western front, although nothing on that front could be done without his dictation. A further cause of weakness was that propaganda, among soldiers and civilians alike, had been overdone. Although the soldier indeed still responded to propaganda, it was only to the most extreme statements. Therefore in order to stiffen in the men the will to fight more resolutely in the attack planned for July 15 they were told that the French were already beaten and exhausted; that the British were ready to go out of the war; that the American army could not get to France, and that, even if it could, it could not fight; that the Champagne-Marne attack was to be the " peace-assault " which would end the war if successful, although as a matter of fact the utmost which Ludendorff really expected of it was that it would pave the way for a similar attack on the Lys. When therefore this attack of July 15 failed, and the French army showed itself anything but a beaten force, and when, three days later, the supposedly nonexistent Americans established alike their presence and their fighting ability by marching through the German lines S. of Soissons in a fashion which compelled the evacuation of the whole Marne salient, the scales dropped from the eyes of the German soldier. To him the war was now lost; it was time to go home. Thereafter, curiously enough, while no longer crediting his own official propaganda, the German soldier became most receptive to Allied propaganda, and looked to it for the truth.

On the' Allied side conditions and conceptions were in general more correctly adjusted. The French leaders knew the weaknesses of their opponents, but were also cognizant of their own, and they were more successful psychologists in dealing with their own men. The British army, having again been recruited up to fighting strength, felt that it had nothing further to fear from the worn-down German army so long as the French line held. The American army, supremely confident in every rank, longed only for the opportunity to disprove the belittling judgments of its opponents and to remove the doubts of its allies as to its fighting capacity. Further, the Allied High Command had the supreme merit of being not only in capable hands but, to a rare degree, decentralized. The attention of its leaders was not distracted from its own field by the necessity of solving distant problems in politics or diplomacy, and was backed in all the principal Allied countries by statesmen who in every way supported and aided the military chieftains, without, on the whole, unduly interfering with the conduct of the armies. Although the Supreme Command was nominally in the hands of Marshal Foch, he was a coordinator of efforts rather than a dominant military commander; and in fact he lacked the staff which would have been necessary for such a control as that exercised by Ludendorff over the German armies.

The plan for the Soissons counter-attack of July 18 was not a new one. As early as the German offensive on the Aisne (May 27), it had been proposed by a member of Gen. Petain's staff, had been approved by his chief, and its details had been worked out. Marshal Foch had likewise favoured it, although hesitatingly, because of the attitude of the authorities in Paris. The difficulties of execution at that time lay in finding divisions of " attack-class " for the spear head at the crucial point. By the middle of July, with the increasing number of American divisions, which had gained and were gaining battle experience, that difficulty disappeared. On July r 1 Petain, on Pershing's insistence, again urged this plan upon Marshal Foch, purposing now to make it immediately after the long-awaited German Marne attack. Foch gave it his approval, not indeed with any hope of gaining thereby any decisive advantage, but rather regarding it as a desirable counter-stroke to the German assault.

The striking success of this counter-attack, which in two days gained and held control of the German communications in the Marne salient, and thereby compelled its evacuation, brought to the Allied leaders, as it did to many in Germany, the discovery that the tide of victory had already turned. On July 24 Foch arranged a meeting of the commanders-in-chief at Bombon, to discuss the means of following up this success and of preserving to the Allies the initiative thus unexpectedly gained.

The chief misgiving of the French Government at this time, now that Paris had been rendered secure through the driving of the Germans back from the Marne, was the lingering apprehension that the enemy might still drive a wedge between the French and the British armies at Amiens. It was therefore particularly welcome that Field-Marshal Haig should propose an attack on the Amiens salient to be made by the newly formed Australian corps, now in that sector and desirous of making the attack, together with the Canadian corps, which had not yet been engaged in the year's battles. At this meeting also it was decided to assign to the American army the reduction of the St. Mihiel salient as its first distinctive operation, but meanwhile to employ this new army in completing the reduction of the Marne salient. All the commanders-in-chief at this meeting expressed themselves as favouring a continuation of offensive action, yet still with the idea of keeping the German army busy, of wearing it down, of seizing favourable occasions and localities for attacks to gain prisoners and material and reconquer useful bits of territory, rather than with any thought of systematic plan for ending the war by victory before winter.

During the two weeks following this conference the German army was forced back slowly from the Marne salient, now become a mere pocket, which was, however, held stubbornly because in this area there had been captured from the French, in May, vast quantities of munitions and military supplies of all kinds and of materials which were urgently needed in Germany, but which there had not been time or facilities for removing. In addition a vast amount of German material had been brought up for the maintenance of the army on the Marne front, and for the July 15 attack. Much of this was irreplaceable, and the German army had to fight to gain time to remove as much of it as possible.

Ludendorff, who had been present with his army during the Champagne-Marne drive, was not especially disheartened at its result and had gone to Flanders hoping to recoup his failure in Champagne by hastening the preparations for his offensive next in contemplation in the Lys salient. It was there that he received news of the Soissons reverse. He immediately realized the threatening consequences to his armies of this Allied counter-thrust, and returned to Avesnes to arrange the necessary withdrawal from the salient.

Materially this retrograde movement did not seriously compromise the German army, since, except during the penetration by the three assault divisions (two American and one Moroccan) S. of Soissons on July 18 and 19, the withdrawal was made slowly and in good order, inflicting as heavy losses on the attackers as the Germans themselves suffered. But Ludendorff soon recognized that the Lys offensive would have to be indefinitely postponed and the troops destined for it used in easing the situation in the Marne salient, where the Allied forces - French, British, American and Italian troops - were now pressing vigorously from all points. What Ludendorff apparently failed to gauge correctly at this time was the resultant damage to the moral of the German army; neither did he yet, seemingly, share the conviction, which had now been brought home to the German people and to Germany's allies, that all hope of ending the war by a German victory was gone, and that the only question left impending was whether it would end by a compromise or by the utter defeat of the Central Powers. Had the German High Command faced at this time the logic of the situation and made a decision to do, after the " Second Marne," what most German officers have since agreed should have been done after the first battle of the Marne, namely, to retire to the line of the Meuse and re-form, subsequent history might have differed materially from the actual events.

As proposed by Haig on July 24, the Australian and Canadian corps on Aug. 8 attacked side by side the German salient facing Amiens, supported by a French corps on their right and a British corps on the left. This attack was one of the most brilliant and tactically interesting episodes of the war, and showed Ludendorff again that the much disparaged tanks were, on ground suitable for their employment, a potent factor in a surprise attack. Although the sector against which the assault was launched was held by first-class troops, the German divisions were overrun and virtually annihilated as organized units. This attack dealt a stunning blow to the pride of the German High Command, a deadly one to the weak moral of the troops, and produced a corresponding exhilaration in the British army, all ranks of which could now clearly see that a complete and final inversion of roles had taken place.

The shock was felt throughout Germany and reacted strongly upon the Government. The unsuccessful Marne attack, with the subsequent withdrawal from the Marne salient, although manifestly a lost battle, had, nevertheless, been one initiated by the German High Command on a battle-field of its own choice. The battle of Amiens could not be so interpreted. The Allies had here initiated the attack and it had been completely successful. Ludendorff correctly names Aug. 8 as the " Black Day " of the German army in the war. So grave was the crisis felt to be that a conference of army leaders and members of the German Cabinet was called to meet at Spa on Aug. 13. It was there agreed that further prosecution of Germany's war aims was hopeless, and that a peace would have to be negotiated at the first favourable opportunity, that is, at the first turn in the military situation even temporarily favourable to Germany. That looked-for turn never came. Under the persistent Allied attacks the German army reserves steadily dwindled, munitions and supplies lessened, and moral evaporated.

The day following the Amiens success Foch decided not to put the American army - which now had some 1,250,000 men in France - in on the Vesle, where the situation was virtually stabilized, but to assign it at once the task of reducing the St. Mihiel salient (see WoEVRE, Battles In).

The battle of Amiens was followed up by a French attack between the Oise and the Aisne on Aug. 20, which forced the German line back on Chauny. Still more serious for the enemy was the attack by the British III. Army, on Aug. 21, N. of the Somme, on the line Bapaume - Peronne, which brought another crisis. By the end of Aug. the military situation had become sufficiently defined to enable the Allied leaders to look beyond a mere driving of the German army back to its strongly fortified lines of the previous winter, popularly known as the Hindenburg line, and to make plans for its rupture in a way to reap the largest strategical as well as tactical fruits of victory.

For this the British army, now fully restored in man-power and in high moral, and the American army, untouched by war weariness or reverses, inspired by an almost religious fervour of belief in the righteousness of the cause in which it was fighting, were of necessity regarded as the chief Allied weapons. The French army was tactically a trained and skilled army, but could no longer count on any large reserves of man-power to replace losses, and the general feeling among the French that their country had already been " bled white " in the war led to the not unreasonable contention by Government and people that, while France must still do her share to the end, her army must from now on be spared as much as possible, since in any event French losses in man-power would far exceed that of any other nation in the war.

Foch, therefore, determined on two main offensives: the British, supported on their right by the French, were to break the Hindenburg line in the direction of Cambrai - St. Quentin; the Americans, after completing the reduction of the St. Mihiel salient, which had been assigned as their first task, were to break through the German lines of defence N. of Verdun, supported on their left by French armies, and to advance in the direction of Mezieres. In other words, the German line in northern France, constituting as it did a huge salient, was to be attacked in the simple orthodox manner by pinching in the two flanks. Of these two the Americans had possibly the harder task, for the Verdun front was well adapted to and thoroughly organized for stubborn defence, and, inasmuch as the railway communications through Sedan - Mezieres were essential to the German army so long as its front lay W. of the Meuse river, the Verdun front, only 50 m. in front of this railway line, was bound to be defended with all the vigour and skill still remaining in the German army. Connecting these two attacks, the French army was to continue its operations to throw back the Germans beyond the Aisne and the Ailette. Such was the Allied plan formulated in Foch's directives of Sept. 3.

By the end of Aug. the German High Command ordered the evacuation of the Lys salient, and it was completed Sept. 6. On Sept. 2 the attack of the British III. Army N. of the Somme was extended northward, E. of Arras, to include the I. Army reinforced by the Canadian corps; and as a result the whole German army fell back to the so-called Hindenburg line, which the Germans themselves designated the Siegfried Stellung. There they hoped to gain time to reorganize the depleted units. This withdrawal, and the accumulating evidences of increasing demoralization in the German army, made it evident to Allied military leaders that offensive operations on a still larger scale could be safely initiated; and Marshal Foch, in a conference with the British and Belgian commanders-in-chief at Cassel Sept. 9, arranged for a fourth offensive, on the extreme northern part of the western front, to force the Germans back towards Ghent. On Sept. 12 the American I. Army, as previously agreed upon, attacked and captured the St. Mihiel salient.

During the latter part of Sept. German H.Q., harassed and preoccupied by the crucial events which were taking place in other theatres, either gave insufficient heed to the precariousness and difficulties of the German military situation on the western front or were too stunned by their sudden and general reversal of fortune everywhere to be able to grasp and cope with them. Ludendorff, it is true, had two lines in rear reconnoitred: one from the Dutch frontier - Bruges - Valenciennes; the other Antwerp - Brussels - Namur - the line of the Meuse; but neither line solved the problem, nor could it be held in the face of a victorious pursuing enemy. The desideratum was to find a secure position for the army's winter respite from active operations, and time for the resting, reorganization and recruitment of the armies. It was possible to accomplish this only by a timely withdrawal, to the line of the Meuse at least, if not to the frontier. But Ludendorff still clung to the idea of holding every foot of French territory until the last possible moment.

On Sept. 26 the final Allied offensive, prepared by the directives of Foch, began. The American army under Pershing and the French IV. Army under Gouraud attacked on the Verdun and Champagne fronts (see Meuse-Argonne, Battle Of). On Sept. 28 the Belgians, supported by a French army under Degoutte and the British II. Army under Plumer, attacked the line from the coast southward beyond Ypres (see Ypres And The Yser, Battles Of, Part iv.). On Sept. 27 the British III. and I. Armies, including the Canadian corps, had attacked on a front of 13 m. in the direction of Cambrai, and on the 29th the British IV. Army under Gen. Rawlinson, after a heavy bombardment lasting two days, attacked the St. Quentin sector.

The American and Belgian attacks had the advantage of coming as a complete strategic surprise; but, in the case of the Meuse-Argonne front, the depth of the fortified zone behind the front lines enabled the German reserves to be brought up and increasingly strong resistance to be made. Both British attacks were made against strongly organized positions held by the best troops the Germans still had; but on the front of the British I., III. and IV. Armies the enemy was already virtually on his rearmost prepared line, the attack was not unexpected, and both opposing armies appreciated thoroughly the consequences of victory and defeat. If driven from the Hindenburg line the weakened German army must thereafter fight in the open. The contest was therefore bitter to the point of desperation, but, even with the aid of the elaborate system for defence afforded by the long-prepared Hindenburg line, the struggle proved unequal, and the German army was forced back with heavy losses, to begin its retreat through the open country.

The German High Command had not appreciated the risk of accepting battle on the Hindenburg line, or else had overestimated either the strength of the position or the remaining fighting capacity of the troops. Once the line was broken, however, they awoke to the situation. On Sept. 28 Ludendorff and Hindenburg agreed that the end had come; on the 29th the Foreign Minister was informed of the army's desperate plight; on Oct. i Hindenburg and the Kaiser went together to Berlin, and on Oct. 4 the first peace offer was sent to President Wilson.

On the battle-front the Allies were not permitting events to lag, and this same day renewed efforts were made on all fronts. The French V. Army under Berthelot had advanced from the Aisne and on Oct. 6 reached the Suippe. On Oct. 7 Foch ordered the attack on the right flank, extended to include the heights E. of the Meuse. On Oct. 8 the British I. and III. Armies renewed their attacks, and in three days drove the Germans back beyond the line of the Selle river - Le Cateau.

Between the two sectors of the Allied main right and left flank offensives lay the strong defensive German positions W. and S. of Laon. On Oct. 9 these positions were abandoned by the enemy, and the whole German line between the Scheldt and the Aisne began its retreat. By Oct. io the American I. Army had penetrated to the last line of German defences on its front, the Kriemhilde Stellung, and cleared the Argonne forest, while on its left the French IV. Army reached the Aisne.

On Oct. io a new directive of Foch gave more distant objec tives to the armies; the Northern Flanders Group was to advance toward Belgium; the British armies, debouching from the front Solesne - Vassigny, were to push both in the direction of Mons and toward Avesnes; on their right the French I. Army was to push up the Oise; while the French and Ameiican armies between the Aisne and the Meuse were to continue their northward movement. The Marshal defined the purposes of these converging attacks to be to force the Germans back on the rough Ardenne forest, where communications were lacking and a modern army would have difficulty in maintaining itself.

On Oct. 12 the French X. Army of Mangin, on the left of the V., reinforced by an Italian corps, passed the Aisne and occupied the Chemin des Dames. On Oct. 14 the army group under King Albert renewed its attack on the front from the Lys to Dixmude. The Germans were unable to hold; Lille had to be abandoned, and, under the combined pressure of this and the British attack, the whole German line N. of Cambrai rolled backward in disorder, toward the Scheldt, closely pursued. The Americans also attacked on the 14th, with important gains W. of the Meuse..

On Oct. 17 the British IV. Army and the French I. Army attacked the hastily improvised German line between Le Cateau and the Oise. On the 10th the III. Army attacked the line of the Selle, supported by the I. Army astride the Scheldt. Both attacks succeeded.

The conditions of the fighting are best understood by referring to the German reserves. From 69 divisions in reserve when the attack of Sept. 26 was begun the German army had been reduced by Oct. 15 to 26 divisions in reserve, of which only 9 were rested. Of the divisions in line many were unfit for combat, but could not be replaced.

Toward the end of Oct. the dissatisfaction in Berlin and elsewhere with the Government had become intense. The people felt that the war had been mismanaged and that they had been deceived. In an attempt to appease them Ludendorff was dismissed on Oct. 25, and his place as quartermaster-general was taken by Gen. Groener. But it was too late to save either the Cabinet or the Monarchy, for with the disaster to the army from the vigorous Allied attacks of Nov. i and following days the Kaiser was forced to abdicate on Nov. 9.

On Nov. i the Allied armies began their final drive. On the right the American I. Army on the Meuse-Argonne front effected a clean break through the German lines, and began an active pursuit which was only stopped by the Armistice on Nov. z r. On the left of the American armies the French IV. Army was equally successful. Farther N. the British I. Army attacked the line of the Rhonelle river and completed the evacuation of Valenciennes. This attack was followed on Nov. 4 by a general attack by the British I., III. and IV. Armies, on a 30-m. front, from Valenciennes to the Sambre, N. of Oisy. In spite of serious natural obstacles, having to force the crossings of the Sambre on the right and to penetrate the forest of Mormal in the centre, the line was advanced 5 m. on Nov. 4. This battle finally broke the German power of resistance, and the German army began a retreat along the entire northern front, though it still offered stiff resistance to the British I. Army on Nov. 5 and 6.

On Nov. 9 the important railway centre and fortress of Maubeuge was taken and the II. Army crossed the Scheldt on its entire front. On Nov. i i the Canadian 3rd Div. captured Mons. Farther N. the Belgian army stood before Ghent.

On Nov. 9 Foch had telegraphed all commanders-in-chief: " The enemy, disorganized by our repeated attacks, yields on the entire front. I appeal to the energy and initiative of the commanders-in-chief and of their armies to render decisive the results gained." On the same day the German delegates presented themselves at Rethondes to ask terms for an armistice, which were accepted on Nov. i i.

During the last week's fighting the outbreak of the revolution in Germany, interrupting as it did the service of communication and the forwarding of supplies, had combined, with the pressure on the front, the depleted and disorganized condition of the troops, and the absence of any available reserves to replace broken and worn combat units, to render further resistance on the part of the Germans an impossibility. Had the Armistice not been concluded a great debacle would have been the result.

Preparations had been made to extend the attack on Nov. 14 to include the Lorraine front E. of Metz, an attack which the German army was as little prepared to meet as it was to resist the advance of the whole Allied line to the north. But this proved unnecessary to secure the Allied war aims. (A. L. C.)