Eastern European Front Campaigns - Encyclopedia

GEOGRAPHICAL NAMES

"EASTERN EUROPEAN FRONT CAMPAIGNS. - Under this heading comes the general story of the campaigns of the World War which were fought between 1914 and 1917 on the front between the Baltic and the Black Sea. Till the summer of 1916, Rumania was neutral, and the theatre of war was limited on the S. by the northern extremity of that country. Thereafter, till the conclusion of the peace of Brest, the Russian and Rumanian fronts became one.

The story falls into three main parts, of which the first is considerably the most important. These are: - the open-warfare, free-manoeuvre campaigns from the outbreak of war till the establishment of a continuous trench line and the setting-in of trench-warfare conditions, along the whole front in Oct. 1915; the trench-warfare operations on the Russian front from that date to the peace of Brest; and the Rumanian campaigns of 1916 and 1917. The events of 1918 belong rather to the story of the Russian civil wars than to that of the World War, and may be summarized for the present purpose in two clauses - the occupation of the Ukraine, for its economic exploitation, by German and Austrian forces, and the maintenance of a cordon, requiring large numbers of troops, along the frontier of Bolshevik Russia to provide against the contingency of a new eastern front being built up by the Entente and the Soviet Government, or by either singly. As an active element in the operations of the World War, the eastern front closes its history with the battle of Riga in the autumn of 1917, and this event, therefore, is taken as the limit of the present article.

I. THE Theatre Of War The operative contrast between the eastern and the western theatres of war lies less in the greater distances and areas of the former than in the fact that there Nature's handiwork has not been greatly modified by man's, whereas in France and Belgium there is an intense network of main roads and railways, and in many parts a great industrial development that has covered the country with factories, mines, tramways and workmen's suburbs. Hence arises a peculiar distinction. Strategically, the western theatre is penetrable everywhere; tactically, it is in many parts so tangled that coherent operations are nearly impossible. In the east, on the contrary, it is strategy that is difficult and tactics that are simple.

The importance of area and distance must not of course be ignored. Without counting Rumanian territory the theatre measures 650 m. x 320 m. - a six weeks' march under peace conditions from flank to flank, and a three weeks' march from front to rear. This and the unfamiliar sound of the place and river names to western ears have tended to make the operations of the eastern front seem more difficult to understand. than they really are. In fact, the course of operations was largely dictated by geography, and the map, rightly read, shows the lines of geography to be drawn in bold, strong strokes. And even in point of distance, the E. - W. depth of the theatre is not more than 12 times the distance covered by the Germans in their 1914 sweep through Belgium and France, and only half that covered by the Grande in its march from the Rhine to Austerlitz in 1805. The picture of the operations of 1914-17, therefore, is not too large for comprehension, and the meanings of its parts are usually clear.

The broadest characteristic of the eastern theatre is its division into four well-defined regions. (a) The great central salient of Poland on and W. of the middle Vistula. (b) The Pripyat or Rokitno marshes, an area of 2 4 0 x 160 m. which, though largely reclaimed in modern times and therefore penetrable to a certain extent for tactical purposes, constitutes an almost insurmountable barrier to strategic movements on a large scale. Lying behind the Polish salient, these marshes, as it were, hollow out its base, leaving on either hand two avenues or corridors: - (c) the northern, connecting Petrograd and Moscow with northwestern Poland, and (d) the southern, connecting Kiev and S. Russia with Galicia and S.W. Poland. To the right and left rear of the salient (a) the two corridors (c) and (d) lie exposed on their outer flanks to hostile attack from E. Prussia and Galicia respectively, except in the portions nearer to their eastern entrances where the hostile frontiers curve away to the sea and to Bessarabia. Across the base of both corridors and in rear of the central marshes runs a, water barrier consisting of the western Dvina and the Dnieper lines, unbroken save for the narrow gap at the watershed traversed by Napoleon in 1812. This waterline marks the eastern limit of the theatre. Its western limits, which espouse the shape of the salient, lie inside the frontiers of Germany and Austria-Hungary and may be taken as the lake region of W. Prussia, the Oder and the Silesian and Carpathian mountains. This limiting line, in contrast to the eastern, has several gaps, of which the most important is that lying between the Silesian and the Carpathian mountains - which is the gate to Vienna, and, owing to the higher cultural development of Germany and Austria, is strategically more penetrable even where geographical obstacles exist.

Across the whole width of the theatre, cutting off the salient from the corridors and the marshes, runs an almost straight barrier of water, constituted by the Vistula and its tributary the San, from the Baltic to beyond Yaroslav, and by the Dniester from the lakes S.W. of Lemberg to the Black Sea. The only gap is between Yaroslav and the lakes of Grodek.

All railways connecting the salient with the interior of Russia, whether they approach by the northern corridor, the marsh or the southern corridor, converge on the Warsaw - Ivangorod portion of this waterline and thence make south-westward for Upper Silesia. Practically all railways from S. Russia to AustriaHungary, on the contrary, traverse the gap of Grodek - Yaroslay. The only line from Russia to the German Baltic lands enters E. Prussia at Wirballen at the broad entrance of the northern corridor; and similarly, at the other end of the theatre, a line from Bessarabia comes into the Bukovina system at Czernowitz. Apart from these two, the whole length of the northern corridor is traversed by three lines from Dvinsk, Polotsk and Orsha respectively ending at Warsaw and Ivangorod; the central marshes by one from Gomel which at Brest-Litovsk merges with the third of the northern lines; and the southern corridor by two from Kiev and Berdichev respectively which at Kovel become one, ending at Ivangorod. The significance of the various lateral lines connecting these approach lines is best judged by studying the map, and here it is enough to draw attention to (r) the line along the eastern base itself; (2) the line Baltic - Shavli - Vilna - Minsk with its accessory Vilna - Baranovichi - Rovno; (3) the line Kovel - Brest-Litovsk - Osowiec - LyckMemel (4) the line Ivangorod - Warsaw - Mlava - Danzig; (5) the line Skierniwice - Lowie - Wloclawek - Danzig. It should also be noted that, in the salient, no lines exist W. of Lodz and N. of Czenstochowa, and that in the northern corridor about Grodno and Augustowo the Prussian and Russian railways carefully avoid contact. Of the road system, it may be said, broadly, that first-class roads are not numerous, and that they group themselves, in the main, on the same axes as the railways. In the area N.W. of Lodz - Czenstochowa, however, roads to some extent mitigate the absence of railways, and about Augustowo the connexion with E. Prussia, which the railways avoid, is, as regards roads, intimate.

Within each of these broad divisions - the salient and the two corridors - other natural features exercised a considerable influence. The chief characteristic of the northern corridor is the practically continuous waterline which defends its flank from attack from E. Prussia. Leaving the Vistula at Novogeorgievsk below Warsaw, this line is formed by the lower Bug, the lower Narew, the Bobr, the lakes of Augustowo and Suwalki, the middle Niemen to Sredniki, the Dubissa, the Vinda y ski canal which crosses the low Shavli watershed, and the Venta prolonged by the Vindava to the Baltic. From the Niemen section to Novogeorgievsk almost every important crossing - there are not many - is protected by permanent fortification of some sort. Its most vulnerable section is that at which the E. Prussian frontier makes contact with Augustowo - Suwalki - Kovno - Grodno. South of this region, on the stretch Rozan - Lomsha, owing to the absence of railways and first-class roads, military operations were never principal, but always dependent upon either those of Suwalki and Augustowo or those astride the Warsaw - MlavaDanzig line. North of Kovno, at the broad entrance to the corridor, it was safe against all but secondary attacks, so long as Kovno held out and kept the attack toward Shavli.

Frontally, of course, the corridor was protected by the Vistula and its fortresses Ivangorod, Warsaw and Modlin or Novogeorgievsk (this last at the origin of the flank barrier just described), and behind this frontal defence were other successive lines - the middle Bug, the middle and upper Narew, the upper Niemen and its feeders, the Vilia system - not to mention partial harriers such as the Wieprz. But most of these rear barriers, in particular the Bug, tend in their upper course to turn southward, thus opening to an invader who stands N. of the San a series of successive gates along the inner edge of the corridor, by which penetration is possible to Bialystok or even to Baranovichi. Hence the special importance attaching, in the operations of 1914-5, to the lower San sector and the fortress of Brest-Litovsk.

The southern corridor, unlike the northern, lies partly on one side of the political frontier and partly on the other. No important natural barrier prevents either an Austrian irruption from the S. as far (roughly) as the line Lublin - Kovel - Sarni, or a Russian irruption through and past Lemberg (Lvov) to the Dniester. As has just been mentioned, the left wing of such an Austrian irruption has the opportunity of seizing the gates of the northern corridor; no reciprocal advantage offers itself to the Russians since the Dniester line is doubled by that of the Carpathians. But, in particular, the fact that the whole Lemberg region is within the Austrian frontier narrowed the corridor normally open to the Russians to a mere strip of country. To protect this from being cut off from behind, the Russians had constructed a triangle of fortresses Rovno - Dubno - Lutsk. At its front end, where it joins the northern corridor and the salient, Ivangorod, Brest-Litovsk, and minor river courses and marshes were relied upon to seal the region of Chelm and Vladimir Volynsk; in effect, a drive by the Austrians into that region if pressed too deep laid open its flanks to counter-attack both from Ivangorod and from Lutsk (Luck).

The geography of the interior of the marsh area needs little description. As above mentioned, much of it is tactically penetrable, but owing to the extreme paucity of communications, as well as to its physical difficulties, it is on the strategic plane essentially an obstacle and not a field of manoeuvre. Its outstanding geographical feature is its river system; the Pripyat itself runs W. - E., but it has numerous N. - S. tributaries notably on the S. side, and these tributaries sometimes form, with tributaries of the Dniester (flowing in the opposite direction), N. - S. waterlines of defence only broken at the watershed (Brody, for example) along which run the communications between Rovno and Lemberg.

In the forepart of the central salient, too, it is the waterlines that are the most important features. The course of the upper Warta; that of the Pilitca; the position of Lodz (or rather Lenczyska) at the divide of the Warta and Bzura systems; the course of the Nida meeting at its mouth the mouth of the Dunajec, one of the several Galician rivers which double the San obstacle; lastly, the upper Vistula itself which forms the southern boundary of the salient - all these were important.

Practically the whole of this region belongs to the W. Russian plain, and has marshy valleys, feeble undulations, and great forests, some of these last still existing in primeval density, others already broken up by man's clearings and settlements. The only hilly mass is the Lysa Goza in the Kielce region of the salient. On the contrary, the Lemberg - Brody - Buczacz portion of the southern corridor, and all country between the San or Dniester and the Carpathians, is almost wholly a country of deep-cut valleys and high plateaux.

The German reentrant opposed to the Polish salient is geographically similar to, but in point of human development very different from, that region. In Silesia, owing to its industrial character, the network of roads and railways is as dense as in western Europe. Without going west of Posen, no less than three complete lateral or circumferential railways join Upper Silesia to the trans-Vistula railways of E. Prussia. As, in face of these, no Russian lateral exists W. of Lodz it is easy to see how this region, in spite of its want of natural defences, was able to act as a curtain between the two bastions of E. Prussia and Galicia, facilitating quick transfers of the centre of gravity from flank to flank and itself (save at one critical moment) immune from attack because of the difficulty of approach.

Of these two " bastions," E. Prussia was the more important as menacing the whole length of the northern corridor, from front to rear. Whereas the Lemberg region only projects from the San - Dniester barrier, E. Prussia has its whole length at right angles to the Vistula. It is served by so many railways that either end of this length is utilizable for the offensive.

The principal directions which this offensive may take are - from the eastern end of the province towards Shavli, from the same towards Kovno and Grodno, and from Mlava towards the Narew and, if and when that obstacle is overcome, on Siedlce or Bialystok. We have seen that the first of these is inevitably a secondary or dependent operation. Between the other two the choice was always, for the German Command, difficult. Presuming the Narew forced, or Kovno taken, as the preliminary in either case, the one offensive leads close into the rear of the Warsaw - Ivangorod stronghold, while in the other the corridor is seized far back near its entrance; the choice therefore depended on how deeply the enemy was advanced in the Polish salient or how long the passive front of the " curtain " could be held, or what chance there was of cooperation from the lower San through the Bug " gate," and on other factors which had to be reckoned together on every occasion that an offensive was planned. But these two avenues (Kielce or Warsaw - Mlava, and Vilna - Kovno [or Grodnoj - Insterburg) equally serve for Russian offensives, and the defensive characteristics of E. Prussia were nearly if not quite as important as its qualities as an offensive base.

The main feature of military geography in E. Prussia is the chain of the Masurian lakes which, in a sickle from N. to S. and then westward, protects the interior against attack from the E. or the S.E. The tongues of land which separate the lakes represent only a narrow frontage which has actually to be defended, and have the effect also of gathering communications, plentiful in the interior, at a few points of exit. To the S. of the lakes a number of tributaries of the Bobr - Narew system continue the water barrier, as against eastern attack, to the Narew; to the N. of them the river Angerapp presents a similar barrier as far as the Pregel, beyond which river smaller streams continue the line of defence with some gaps to the Niemen. Behind the lakes, the next important N. - S. barrier is the line of the Alle which, rising in the central Masurian lakes, runs to the Pregel at Wehlau, whence from Tapiau to the Kurische Haff runs the Deime. Other partial barriers to an invader's westward progress exist but are of less importance. Finally there is the German section of the lower Vistula which, intricate at Danzig and fortified at Thorn and Graudenz, still bars access to Germany proper when E. Prussia has been conquered or evacuated.

Thus on the E. this province is singularly well protected. But it is to be noted (1) that the frontier, especially in the northern part, lies well in advance of the barrier, and that a policy of passive defence on the lake line forfeits a not inconsiderable region at the outset; and (2) that both the Insterburg - Johannisburg line and the Alle are turned by attack from the S., by Mlava and Soldau, where the westernmost part of the lake system dies away. At the centre of the " sickle," on the other hand, the density of the lakes is highest and they not only afford local protection to this part of the region, but also enable the defending army to shift its weight from E. to S.W. and vice versa without much fear of flank attack in doing so; while, on the Russian side, the paucity of communications in the foreground of these central lakes seriously impedes liaison between the northern or Gumbinnen and the south-western or Soldau groups of the invaders. Such shifts of the centre of gravity are, moreover, facilitated by the dense railway system lying behind the lakes. The frontier railway, which runs from Thorn, by Soldau, Johannisburg and Lyck (junction of the Russian Bialystok-Brest-Litovsk transversal), to Tilsit and Memel, lies outside all defensive barriers. But inside the barriers are some three other transversals, one being the Thorn-Insterburg-Wirballen section of the Berlin-Petrograd main line, and the others parts of a well-developed provincial system.

The military-geographical characteristics of the Lemberg region, the other potentially offensive base lying outside the Vistula-San-Dniester barrier, are less sharply marked and their influence is not so definite. Offensive possibilities lie in the direction of (a) Bessarabia, (b) Kiev, (c) Kovel, (d) the inner edge of the northern corridor, towards Brest-Litovsk. Of these, as in the case of E. Prussia, (a) is eccentric, except as a secondary element of (b); and (c) centres on a region which is ill-developed in communications, and therefore operations there are subsidiary to those on either flank. The important alternatives are therefore, speaking broadly, (b) and (d). In (b) Dubno and Rovno play the same role as Kovno in the N., and the results to be expected from a successful operation of this character are similar to, but smaller in scale than, the corresponding enterprise on the Niemen. (d) The operation, twice carried out and several times contemplated, offered many results and many risks, and its usefulness varied according to a number of factors like that of the corresponding operation from the N., with which, in fact, it was logically combined.

Defensively, the conditions of the Lemberg region were similar in some respects to those of E. Prussia. Waterlines opposed invasion from the E., while from the N. Lemberg was open. But the real obstacle value of the E. Galician watercourses,- Gnila Lipa, Ziota Lipa, Strypa, etc., - whose names were to become historic, is small, and, though N. of the Styr system and the uppermost streams of the Bug (Styr) have wide marshy valleys and are serious barriers, the watershed itself (DubnoBrody-Lemberg) is an open gate both for road and rail approach to the Galician capital.

The railway system of the Galicia theatre, though far inferior to that of Silesia and Prussia, included two complete laterals N. of the Carpathians, and at least one S. of them. From the interior of Hungary and Moravia, over the Carpathians, to the San-Dniester barrier there were eight approach railways between Teschen in the W. and Czernovitz in the E., and four of these pass the barrier at or near the Grodek gap, converging on Lemberg and Rava Ruska. In the latter region itself the railways lie chiefly radially from Lemberg. It is to be noted that on the whole front N. of Lemberg the Russian frontier region is destitute of approach railways.

Finally, the Carpathians (of which Galicia to the San, to Lemberg and to the Dniester, is simply a glacis) are not as the sea is to E. Prussia, a definitive barrier, but rather a wall with many gates for the passage of an invader into Hungary and Austria. The mountains themselves are rather Vosgian than Alpine, and their main passes are low enough to be practicable for railways. At the W. and E. ends, the mountains broaden out into the Tatra and massifs, but in the centre the mountain zone is at its narrowest, and it is exactly in front of this that the Grodek gap breaks the forward barrier and allows these railway approach lines to make for the Hungarian plain. West of the Tatra massif, the Troppau gap opens Moravia to an invader who has mastered Upper Silesia. (C. F. A.) II. THE Campaigns Of 1914 The Russian Plan of Campaign. - Two characteristics of the Russian Army were admitted on both sides as axiomatic, the relative slowness with which its total forces could be brought to bear and the numerically overpowering superiority of those forces when assembled and ready. Both these were summed up in the popular phrase of 1914, which likened the Russian Army to a steam roller. The axioms were not, however, independent. Only by waiting could the overpowering strength be realized, and by temporarily forgoing this numerical advantage, it was possible for the Russians to act with partial forces and provisional objectives, almost if not quite as promptly as the armies of the Central Powers. Instead, therefore, of the usual stages of couverture and full-power action there would be, or might be, three - couverture, rapid partial action, and delayed full-power action - and the application of the geographical factors to strategy varied accordingly.

In all alternatives, the inclusion of the central salient, either in the couverture system or in the deployment for the main action, was impossible. In other words, it was militarily evacuated from the outset. In the alternative of delayed full-power action, the couverture would guard the outer flanks of the two corridors and the Warsaw-Ivangorod-Lutsk front, while the main masses assembled further back. Flank-guard groups would prolong the defence of the corridors respectively in the Shavli region and to the S.E. of Dubno and Rovno. The line of detrainment for the main bodies would be, substantially, Kovno-Grodno-Bialystok-B rest, and (for the Southern armies) points behind Rovno. But the abandonment of so large a portion of Poland would only be necessary in the case of Germany's employing the major portion of her forces in the east. In that case, especially if it arose in winter, it was calculated that the Russian forces on the couverture line would have to retire fighting, giving up Warsaw and possibly Ivangorod, but holding firmly at all costs on the middle Niemen front and at Brest. If that case did not arise, then the couverture was strong enough to enable the main masses using the northern corridor to detrain further forward. In proportion as the arrangements for mobilization and concentration were improved in the years 1910-4, and in proportion also asit became more probable that Germany would elect to employ the bulk of her forces on her French front, not only this forward concentration but also preparatory offensives delivered from the couverture line came to be considered.

In all cases the main object which was to be sought when the forces were fully assembled was practically the same. It was the destruction of the Austrian armies in Galicia, the occupation of the Carpathian line, and eventually an advance into Moravia and Silesia by Troppau and the Oder head, turning Breslau. The exact form in which this ultimate offensive would be realized could not be foreseen until the Germans and Austrians had shown their hand; meantime, the problem before the Russian general staff was so to plan their couverture arrangements, their detrainments, and their now feasible preparatory offensives as to subserve this purpose.

Generally speaking, the couverture on the Narew-Bobr, that on the middle Niemen, and that in the Shavli region were disposed and directed to checking as long as possible any German attack on the flank of the northern corridor. It would be reenforced in situ to the strength of two armies and an independent group. If powerful German attacks developed it would offer an elastic defence, on one line after another, to protect at all costs the region of Bialystok-Grodno-Vilna during the troop movements in that area. If not, it was to take the offensive and, by conquering E. Prussia to the Vistula, definitely to secure the right rear of the future main effort. This conquest was to be carried out from the S. by Mlava, turning the lake barrier, by one army while the other pressed up against the front of the lakes and the Angorapp, so as to occupy the Germans and at any rate to prevent a rush upon Kovno and Grodno. The independent group about Shavli was to deal with minor enterprises of the enemy in its own area, and especially with landing threats on the Baltic coast as far as Riga. From that point inclusive, coast defence was entrusted to another army, with headquarters at Petrograd.

In the centre two armies, coming from the interior by the central and eastern railways of the corridor, were, if possible, to concentrate about Lublin and Chelm respectively; otherwise they were to divide, one going to the right of the defensive wing about Shavli, the other continuing S. to Brest and Kobryn. Supposing that this proved unnecessary, the two armies, from Lublin and Chelm respectively, were to take the offensive against the left of the Austrian armies in Galicia. The right of these meantime would be attacked by two other armies, advancing from Dubno and from Proskurov. These armies were given special precedence in their equipment, so as to be ready to act early. At Odessa, a minor army of reserve divisions was to be assembled to watch Rumania.

Defensive or offensive as the case might be, these preparatory engagements were all assumed to be in progress before the full concentration had been effected. Including the Petrograd army, only 28 out of a total of 37 active corps were comprised in the dispositions, and the reserve divisions formed on mobilization were not counted upon for immediate service. The remainder, in so far as no new complications occurred to tie them to their peace regions (e.g. Caucasus), would become successively available and constitute a mass of manoeuvre or a pool of reinforcements, according to the course of events.

On mobilization, accordingly, the allocation of troops was as follows: I. Army (Rennenkampf). Niemen, including Shavli. II., III., IV., XX. Gd., I. Corps. (As soon as relieved by reserve divisions [XXVI. Corps] at Shavli, XX. Gd. was to proceed to IV. Army.) First task: protection in front of Niemen line, on that line, or if necessary further back towards Vilna. Second task: advance to bind the German forces on the lakes and Angerapp.

II. Army (Samsonov). Narew. VI., XV., XVIII., XIII. Corps. First task: protection of Bobr-Narew-Bug line and reconnaissance into Mlava-Neidenburg region. In case of heavy German offensive, the region of Bialystok to be protected at all costs. Second task: invasion and conquest of E. Prussia via Mlava, turning the lakes. (These two armies had each several reserve divisions allotted.) IV. Army (Evert). Concentration area Lublin. Grenadier, XIV., XVI., XVIII. Corps.

V. Army (Plehve). Concentration area Chelm.°. V., Xvii., Xix., and XXV. Corps.

Both for attack of N. front of Austrian armies in Galicia.

III. Army (Ruzsky). Concentration Rovno-Dubno. IX., X., XI., XXI. Corps.

VIII. Army (Brussilov). Concentration S. and W. of Proskurov.

VII., VIII., XII., XXIV., III. Caucasian Corps.

Both for attack of N.E. and E. front of Austrians in Galicia. The I. and II. Armies formed the north-western front under Gen. Zhilinsky (succeeded after the first operations by Ruzsky), the IV., V., III., VIII. the south-western front under Gen. Ivanov, whose Chief of Staff was Alexeyev.

The VI. Army (Grand Duke Nicholas) was the title of the Petrograd force, the VII. (Nikitin) that of the Odessa troops.

(In the event of German offensives developing on a large scale, requiring the adoption of the rear line of rail-heads, the IV. Army was to be switched en route to the right of the I., and to it instead of to the VIII., the XXIV. Corps was to go. It would also become part of the north-western front.) The peace-time scheme, as thus outlined, was at once modified in the early days of mobilization, not so much in intentions as in allocations of force. No commander-in-chief of the whole was appointed before the war, as the Tsar was undecided as to whether to take command himself. At the outbreak of war the Grand Duke Nicholas, Commander of the VI. Army, was appointed. He had taken no part in drawing up the scheme, and his own ideas differed somewhat from it. He therefore formed a new scheme, or rather a modification of the basic scheme, whereby the Guard and I. Corps were dispatched to Warsaw (instead of to the I. Army) to form the nucleus of a IX. Army, and the VI. or Petrograd Army was reduced first to one corps, and then to reserve divisions only. The first corps to leave was the XVIII., originally intended for the IV. Army but now assigned to the IX. (replaced in the IV. by the III. Caucasian Corps taken from the VIII. Army). The XXII. followed towards the end of August, joining the I. Army in lieu of the Guard and I. Corps. Further, a number of the reserve divisions accumulating behind the I. and II. Armies were constituted a little later as a X. Army with the mission of connecting the I. and II. Armies - but too late to avoid the catastrophe of Tannenberg.

Mobilization and concentration proceeded rapidly. The cavalry divisions allotted to the Prussian front were detrained complete by the 7th day of mobilization, the infantry corps by the 13th day. On Aug. 14 the Grand Duke informed the French ambassador that the I. and II. Armies would open their offensive on the morrow, considerably sooner than was expected by the French, who only began their advance on that day.

The " preventive " offensive that was to lead to Tannenberg was thus launched on Aug. 14. Its objects were, partly, the accelerated fulfilment of the original plan of campaign (at the lowest, the active flank defence of the northern corridor, now being traversed by a IX. Army as well as the IV.); and partly, the desire to aid France by startling the German command into making detachments to the E.

Central Powers. - The problem of war on two fronts had for many years been anxiously studied in Germany and it had been generally accepted in principle that a simultaneous offensive E. and W. was impossible. In the time of the elder Moltke, the difficulty of defending the long, open eastern frontier, as compared with the relative ease with which the short, strong line Thionville-Strassburg could be held, had decided the great general staff in favour of choosing the east as the offensive theatre; and this plan held the field, with few modifications, until Schlieffen came into office as Chief of the General Staff and reconsidered the military position. He decided that the first offensive must be directed against France, but in such a way as to insure the quick and complete destruction of the French army, i.e. by using Belgian avenues for the envelopment of its left. His solution of the two-front war problem, therefore, was to prevent its happening: neither he nor his successor, the younger Moltke, seems to have dealt exhaustively with the case that actually arose, i.e. that of a prolonged contest in which the centre of gravity constantly required to be shifted from E. to W. and vice versa. An important factor, perhaps the ruling factor, in the decision was the assumption that it would be impossible to bring the Russian army to decisive battle; owing to its slow assembly, the distances to be traversed in order to reach it required a time allowance which the western defensive, at grips with the highly trained and efficient French army, could not insure for it. Moreover, with unlimited space behind them the Russians were regarded as having every chance of avoiding a decision for as long as they wished to do so, and the re-distribution of the Russian peace garrisons after 1910 (which pointed to the choice of the rear line Kovno-Bialystok-Brest as the probable line of entrainment) confirmed the conclusion. Two possible offensive directions were considered, that from the Mlava region against the Narew line, and that from the lake front by Wirballen and by Augustowo and Suwalki against Kovno and Vilna. These alternatives and their meaning have been alluded to already. The choice was a difficult one, hardly to be settled except ad hoc; it was to be the chief bone of contention between Falkenhayn and Hindenburg in the 1915 campaign. But even the second, and more promising, line of operations would not lead to the .nemy's rear if he abandoned all Poland at the outset, and concentrated between Kovno and Brest.

In fact such a course of action was provided for in the Russian concentration scheme. But the alternative preferred by the Russians was an offensive, or two offensives, carried out by the readiest portion of their forces, and their alternative naturally engaged the attention of the Central Powers in the years after 1910, when the war-readiness of the Russian army was evidently being improved with menacing rapidity. The defence against such an attack could not readily be combined by the two Central Powers because of the salient W. of the Vistula; on the defensive, therefore, Germany and Austria-Hungary formed two theatres, either or both of which might be the target of enemy offensives of uncertain power. Further, the entire peace forces of the Central Powers, taken together, were not equal numerically to the peace forces of Russia, and the adhesion of Turkey, and still more that of Rumania, to their side was problematical. If the bulk of the Russian forces concentrated on the forward line, then there were only two practical alternatives for the Central Powers: either (a) to concentrate as much as possible of the German army in the E. (relying upon the short and well-fortified defensive line of Lorraine and Alsace, doubled by the Saar and the Rhine, to hold up the French), and to take the offensive with 90 or 95 divisions, German and Austro-Hungarian, as soon as possible so as to catch the enemy in the act of detrainment; or else (b) to stand on the defensive, each in his own theatre of war, resigned to give up territory in order to gain time for the annihilation of the French.

But that annihilation effort would require at least four-fifths of the German mobilizable forces, if it were to be carried out in the short time that the conditions of the E. allowed, and in the case of Germany the territory that would have to be resigned was E. Prussia, bound indissolubly to the Hohenzollerns and to the Prussian army by ties of sentiment and tradition. Its abandonment was " unthinkable." Yet the force that could be spared to defend it was small indeed. The Reichstag had declined to sanction the creation of the three new army corps which would have eased the problem; and, in the event, one to two corps allotted in principle to the E. were taken at the last moment for the W. In short, the German army allotted to the E. was a minimum force. But it was not on that account authorized to give up any German ground.

The case of Austria-Hungary was more favourable to this extent, that nearly the whole force of the Dual Monarchy could be employed in the defence of Galicia, unless (as actually happened) offensive action was simultaneously undertaken in the Serbian theatre. On the other hand, Galicia would clearly be the enemy's principal target, and were he to leave mere flank guards against E. Prussia, there was little doubt that even in an accelerated offensive he could employ superior forces. Many Austrian authorities therefore favoured a withdrawal of the line of defence to the Carpathians, and probably the majority considered that nothing could be held in advance of the San - Dniester barrier. The problem then was difficult and obscure, and differences of opinion both within each country and between the two countries themselves were certain. Austria's strategy even in respect of her local problems depended largely upon Germany's, and no definite, binding convention appears to have been negotiated, either for the case of the offensive or for that of the defence. More, the interchange of views which did take place led to completely disjointed action. When the inner wing of the Austro-Hungarians was driving forward the offensive on Lublin and Chelm, the Germans in E. Prussia were under orders to retreat to the Vistula.

Conrad von Hotzendorff, the head of the Austrian general staff, was essentially active in temperament, and the wave of sentiment in favour of the undiluted offensive which swept through all European armies about 1912 strongly influenced him and his entourage. A scheme was prepared under which the left portion of the Austro-Hungarian army was to take the offensive from the lower San, northward on Lublin and Chelm, flank-guarded by an echelon directed on Vladimir Volynsk, while the right portion defended Lemberg against attack from the E. In cooperation with this left wing, a German army was to advance by Mlava on the Narew line, force this, and effect a junction with the Austrian advance about Siedlce. By this scheme it was hoped either to cut off a part of the Russian army and beat other parts in detail as they detrained - if the Russians were attempting to forward concentration - or to make good military occupation of almost the whole of Poland in the shortest time - if they were concentrating on the rear line Kovno - Brest. At the lowest, Conrad held, the protection of Galicia and of E. Prussia would be best assured by the offensive.

In how far Moltke agreed to this plan is doubtful. He had definitely committed himself to the Schlieffen scheme of putting France out of action before an eastern front came into existence, and though he had considerably altered its details, he had provided even less force for E. Prussia than Schlieffen had proposed. Such evidence as is available tends to show that Moltke agreed with the scheme as the operative idea of the eastern offensive that was to follow the decisive defeat of France (expected to have been sufficiently achieved by about the 30th day of mobilization), but not as a preventive offensive to be launched while the issue in France was still undecided. Conrad, on the other hand, was determined to carry it out the moment he was ready, hoping, as he said, that Moltke would not "leave him sitting in the ink too long." The scale of the operation for him was only that of a preventive offensive, carried out substantially by about 27 Austro-Hungarian and io German divisions from the San and from the Mlava region respectively. This force, if it caught the Russians in the act of concentration, would create " favourable conditions for later operations " on a large scale.

Moltke, on the contrary, gave the E. Prussian Army (VIII., Gen.-Oberst von Prittwitz and Gaffron) nine active and reserve divisions (I., Xvii., Xx. Active Corps, I. Res. Corps, 3rd Res. Div.), for both the lake and the Mlava fronts. Apart from a number of Ersatz and Landwehr formations, most of which were intended for the defence of Thorn, Graudenz and Konigsberg, this was all. In Posen province and in Silesia, there were only frontier guards of Landsturm, and the Landwehr and Ersatz garrisons of Breslau and Posen; as the salient facing these provinces was practically evacuated, no more was necessary, and indeed eight Landwehr regiments were grouped in Upper Silesia as a field force (the "Landwehr Corps," von Woyrsch) to accompany and guard the left of the Austrian offensive.

Thus, the first campaigns in the E. were distinct and without connexion of idea or of date. The battles of both being described elsewhere, it is sufficient here to outline the campaigns of Lemberg and Tannenberg in succession.

The first requirement of the Russian scheme of operations being free use of the northern corridor for the assembly of forces against the Austrian left, the troops disposed on the dangerous flank of the corridor were ready for action about ten days before the date set for the completion of the Lublin - Chelm concentration. In the original scheme, their mission was primarily defensive and in the second place offensive, but as early as Aug. 9 the Grand Duke determined to push forward both the I. and II. Armies on their offensive missions, in the hope of at once compelling the Germans to hold back forces destined for the W. On the 14th, their concentration completed, these armies moved out of the detrainment areas, the I. (II., III., IV. and XX. Corps) under Rennenkampf on the axes Kovno - Gumbinnen and Suwalki - Marggrabova, the II. (VI., XV., XXIII., XIII. and later I. Corps) under Samsonov on the axis Przusznysz - Soldau. Seven to eight cavalry divisions accompanied and preceded them. At many points on the frontier from Memel to Bialla and round to Mlava there had already been local engagements, especially on the axis Kovno - Gumbinnen, where Rennenkampf on the one side and von Francois (commander of the German I. Corps) on the other had both strong motives for activity, the Russians to thrust back the enemy's forces as far from the " corridor " as possible, the Germans to preserve the region between the frontier and the lakes as long as possible from occupation or pillaging. On the Mlava axis these episodes were fewer, for the Russian main bodies were more distant. The Germans were unable to prevent the enemy's mounted troops from ranging up to Soldau, but their Zeppelins reconnoitred the line of advance of Samsonov's main bodies.

Gen. von Prittwitz, in spite of his small forces, was confident. He placed the I. Corps (Francois) facing E. on and in front of the Angerapp, the XX. Corps facing S. between Allenstein and Soldau, the XVII. and I. Res. Corps and the 3rd Res. Div. in the interior, waiting on events. To the left rear of the I. Corps, the Konigsberg main reserve - an Ersatz and Landwehr force numerically, but only numerically, equivalent to a corps - moved out N. of the Pregel to Insterburg. To the right of the XX. Corps was a frontier guard, also composed of Ersatz and Landwehr belonging to the fortresses of Thorn and Graudenz. On Aug. 14 v. Prittwitz, satisfied that no important threat was impending on his S. front, turned over the defence of that front to the Landwehr and Ersatz formations of Gen. von Unger, drew the XX. Corps to Ortelsburg, in readiness for an offensive to wards Johannisburg, and brought the remainder of his forces to the E. front behind and on the flanks of the I. Corps, the 3rd Res. Div. (reinforced by one brigade and a screen of Landsturm posts) holding the lake barrier.

On the i i th took place the first serious encounter of large forces. Von Francois still maintained a forward position on the Kovno railway at Stalluponen, barely five miles inside the frontier; he was determined to defend offensively, and he inflicted a sharp blow on the central columns of the enemy before the others became effective. But his left was driven in, and Prittwitz, whose intention was by no means to fight so far forward, ordered the combat to be broken off and the troops to retire to the Gumbinnen position. There, on the 10th and loth, the battle of Gumbinnen was fought. Claimed by both sides as a victory but in fact indecisive, since parts of each line gained successes or suffered failure, it ended in Prittwitz's ordering the battle to be broken off. To the astonishment of his corps commanders, he announced that he proposed to retreat over the Vistula.

A grave crisis had arisen. The Russian II. Army, seemingly quiescent on the Narew, had in fact been cautiously advancing on the Mlava axis, which was now defended only by secondline troops and, partially, by the XX. Corps - everything else, even mobile Landwehr brigades, having been brought over to the E. front by the order of the 14th. Such was the situation of the defences when, some time after noon on the loth, reports reached Prittwitz to the effect that four or more Russian corps were approaching Dllava and Ortelsburg. He had three alternatives - to disregard the threat, win an effective victory at Gumbinnen, and pursue the enemy in such a way as to impose caution on all Russian forces in advance of the sensitive point of the " corridor "; to leave a containing force about Gumbinnen, trust to the lake barrier, and bring back the bulk of the forces so as to strike the flank of the oncoming II. Army; or to fall back beyond the sweep of that army's manoeuvre. The first alternative was eagerly advocated by von Francois, but the other corps had met with little success in the battle. It is probable that no reasonable hope remained of winning a thorough victory on the 21st, and nothing less would serve. The second alternative was not, at that moment, considered and the third was adopted in its most extreme form, retreat beyond the Vistula. The I. Corps was to move by train to Bischofswerder and Gosslershausen, in order to bar the road to the Vistula, the XX. to fight for time, and the remainder to withdraw southwestward under cover of these corps. A factor in the decision was the activity of Russian cavalry which, in large and small bodies, was appearing in the interior of the province.

In the Kriegsspiel exercises of peace-time, this problem had often been fought out, and the idea of sacrificing E. Prussia - on paper - was familiar. But, as the elder Moltke observed in 1866, " in practice one does not abandon provinces." On the loth, apparently on the initiative of Lieut.-Col. Hoffmann of the staff of the VIII. Army, who sent a protest direct to supreme headquarters at Coblenz, Moltke communicated with Prittwitz by telephone, and urged him to try the alternative of a manoeuvre on interior lines. The army commander replied that this was impossible and that he might need reinforcements even to secure an escape to the Vistula. Thereupon Moltke relieved him of his command, opened direct telephone communication with von Francois and von Scholtz (XX. Corps), telegraphed to General-Oberst von Hindenburg, in retirement, to offer him the command, and summoned Ludendorff (deputy chief of staff II. Army) to act as chief of staff.

This situation, in fact, was less alarming than it had been on the 21st. Neither Rennenkampf nor Samsonov displayed any important activity; and Prittwitz recovered confidence, decided to hold the line of the Passarge against Rennenkampf, and began to work out a scheme of attack against the Russian II. Army. But the order of dismissal reached him that evening.

On the 23rd Hindenburg and Ludendorff arrived at Marienburg (H.Q. VIII. Army). Already, after a conversation with Moltke, Ludendorff had (apparently on a suggestion from Francois) fixed Deutsch-Eylau and eastward, instead of Goss lershausen, as the rendezvous of the I. Corps and ordered all available Ersatz and Landwehr units from Thorn and Graudenz to strengthen von Unger, thus beginning to prepare a group of two active corps and other troops to check Samsonov. The other forces lately engaged at Gumbinnen were to remain, temporarily, opposing Rennenkampf - all measures designed, evidently, to arrest the sense of retreat and panic. Not until the staff of the VIII. Army had reported the situation in detail was a clear idea of possibilities formed by the new leaders. In principle the plan was adopted of holding up Rennenkampf, maintaining the lake region against any break-in from Lomzha, and concentrating offensive effort on Samsonov. Both the newcomers and the staff already on the spot were in agreement as to this. But it remained to be seen whether, and even how, it was to be accomplished. On the evening of arrival, Hindenburg reported to Coblenz " assembly of army at XX. Corps and enveloping attack planned for Aug. 26," but next evening, developing the idea in some details, he added: " moral determined but not impossible things turn out badly." The intention was to disengage some, or even all, of the troops opposing Rennenkampf, and with them by a flank march close behind the lakes, to envelop Samsonov's right; to bring in the I. Corps and the nearest portions of von Unger's force against his left, and to hold him frontally with the XX. Corps. It was on this last that everything hinged. Short of a simultaneous effort by both Rennenkampf and Samsonov - the case feared by the Germans - pressure by the II. Army alone was of greater significance than that of the I. Army alone would be. Rennenkampf, however active, could only drive the I. Res. and XVII. Corps south-westward towards the Passarge (and the Konigsberg troops into their fortress) and work a passage down the rear side of the lakes to join hands with Samsonov, whose VI. Corps was made to diverge towards Ortelsburg for that purpose. Samsonov, on the contrary, could by an energetic advance bring three corps (XIII., XV., XXIII.) against Scholtz, and in case of success break into the midst of the new dispositions of his opponent. On the 23rd24th this seemed probable, for on those days he attacked the XX. Corps and forced it to swing back from the line Gilgenburg-Orlau to the line Gilgenburg-Hohenstein. At that moment the 3rd Res. Div. at Hohenstein and even the first arrivals of the I. Corps at D. Eylau were being drawn into the fight to assist von Scholtz. The arrival of the rest of the I. Corps, destined for the flank attack on the W., was delayed by misadventures; and this western attack (I. and Unger) was itself becoming imperilled by the advance of yet another Russian Corps, the I., from the IX. Army forming at Warsaw. Of the other German corps not one was disengaged for its southward march before the 24th.

On the 24th, however, the withdrawal from Rennenkampf's front began. It was carried out in the midst of an emigration en masse, main roads being so crowded with refugees that troops were marched in some cases entirely by tracks and by-roads. Russian cavalry parties were by now riding about the country as far as the Passarge.

The Angerapp line having been given up on the 22nd, the front of contact opposite the Russian I. Army now (24th) ran along the Deime and the lower Alle, astride the Pregel, to Allenburg, thence by Gerdauen to Angerburg at the N. end of the lakes. North of the Pregel, the Konigsberg force was slowly retiring on its fortress and had left Wehlau. South of it there were withdrawn, each in succession and covered by the rest, the XVII. Corps, which was directed on Bischofsburg-Ortelsburg; the I. Res. Corps, directed upon Seeburg; and finally the 6th Landwehr brigade from Lotzen, the key of the lakes, to the same region. Only the Konigsberg force, one cavalry brigade and some Landsturm, remained in front of the Russian I. Army.

Meantime, Samsonov continued his methodical advance, but very slowly - the VI. Corps on Bischofsburg via Ortelsburg; the XIII. on Allenstein; while the other two, followed in echelon to the left rear by the I. Corps, were sent against Scholtz (loth and 3rd Res. Div.), whose left was driven from Hohenstein. But already the two wings of the envelopment were being pre pared and directed according to the indications of Russian wireless messages sent in clear. On the W., the German I. Corps, with additional troops under Miilmann coming up on its right, attacked towards Usdau on the 26th. On the same day the Russian VI. Corps was met and defeated at Gross-BOssau by the oncoming eastern enveloping wing. Von der Goltz's Landwehr division, arriving opportunely from SchleswigHolstein, was added to Scholtz's threatened flank. From the 26th the battle was general. Strategy had done its part. By the 31st the destruction of Samsonov's army by double envelopment was complete, only the attached I. Corps echeloned back on the left being outside the ring and able to escape as a formed body. Samsonov himself fell, and 92,000 prisoners and 300 guns remained in the hands of the victors.

Meantime the German supreme command at Coblenz had taken a step which is generally regarded as having been fatal to Germany's success in the war. Moltke had recognized from the first that the strength of the VIII. Army was little above, if not below, the safety limit, and in the background there was Conrad's repeated demand for effective cooperation in the Siedlce scheme. Only after much hesitation was the IX. Res. Corps in Schleswig-Holstein taken to reenf orce the W. on the strength of Prittwitz's optimistic reports on the eve of the battle of Gumbinnen. Two days later came the crisis which led to Hindenburg's appointment, but at that moment the battle of the Frontiers was developing all along the line in the W. and Moltke did not suggest (nor did Ludendorff ask for) a reonforcement of the E. On Aug. 25, however, caught apparently in a wave of optimism which pervaded the armies of the W. after five simultaneous victories, Moltke decided to send no less than six army corps to the VIII. Army, not so much in order to reestablish a compromised situation there as to deal the offensive blow in the E. that was only waiting upon a decision in France. Two corps were to go from each portion of the western front, and the Guard Reserve and XI. Corps, being reported by their army commanders, after the fall of Namur, as free, were sent first, along with the 8th Ca y. Div. In the event, the other four were never sent, as the results of Tannenberg altered the balance of forces in the E. at the same time as a new crisis was arising in the French theatre.

These reenforcements arrived too late for the battle of Tannenberg, but began to be available in the first week of September. Meantime the VIII. Army Command had to decide whether to pursue immediately to the southward, forcing the Narew line and making rendezvous with Conrad about Siedlce, or to deal with Rennenkampf's army which still stood, inactive but threatening, on the Deime-Wehlau-Allenburg-Angerburg-Bialla line. The latter course was preferred, as was practically inevitable. The progress of the Austrian I. Army and Woyrsch (see below) in the Lublin region was evidently being neutralized by the advance of Ruzsky and Brussilov in E. Galicia, and Rennenkampf's inactivity could hardly continue. Moreover, he occupied a great part of E. Prussia and the call of the civil population for rescue from the Cossacks could not be ignored.

Rennenkampf's halt on the Deime-Angerburg line, when enemy forces were daily slipping away from him to take part in the destruction of Samsonov's army, was and is severely criticized, and exposed him to the reproach even of treason. Part at least of the causes of this passivity lay in the inherent slowness of Russian military practice - a slowness which equally characterized the unfortunate army of Samsonov, as we have seen. For the rest, it is to be noted that the Grand Duke was himself at Insterburg during the critical days. Such evidence as is available suggests that the intention of the Russian supreme command was not to press even Samsonov's offensive, still less the frontal advance, farther than it would go, but to give the whole campaign a wider sweep by means of the new IX. Army assembling at Warsaw and intended to move on Thorn and Posen,' turning the Vistula barrier from the south.

1 The I. Corps of this army was not placed at Samsonov's disposal till Aug. 26, the Guard not at all. One cavalry division was actually taken from Samsonov.

From the German point of view, although information was no doubt lacking as to the large undisclosed reserves moving in the " corridor," it must have been clear that the defeat of Rennenkampf would effectively answer any renewed threat from the S. by endangering the Grodno-Kovno artery. In the conditions of the moment this defeat could best be ensured by attacking his left wing, and in the first days of Sept. the VIII. Army with the corps from the W. were disposed accordingly on a long line from Preussisch-Eylau to E. of Willenberg: in order from left to right Guard Res., I. Res.; XI., XX., Xvii., I. Corps and 3rd Res. Div. Von der Goltz with his own division and another made up from Unger's and Miihnann's forces (called 35th Res. Div.) watched the southern front on both sides of Mlava. The Konigsberg force still held the Deime line. On his side Rennenkampf had already brought up two of his reserve divisions from the Niemen for the siege of Konigsberg, and he now strengthened his left from botn active and reserve formations assembled about Grodno. As had been the case at Tannenberg, the forces were numerically almost even. On neither side was any important condensation of force at particular points effected, and the resultant battle, known as the battle of the Masurian lakes, or of Angerburg, was practically " linear." The idea pursued by Hindenburg was to press the Russian right, as far S. as Angerburg, with four corps, to break out of Lotzen (the key of the lakes, which had been kept throughout) with the XVII. Corps while the I. Corps and 3rd Res. Div. advanced from their Tannenberg positions eastward along the frontier railway. These 22 corps were intended to roll up the left of Rennenkampf and press northward, with an echelon to the right against the fresh enemy forces reported detraining about Grayevo. The battle began on Sept. 7 and on the 8th was general. But the lake barrier this time favoured the Russians. The German XVII. Corps made only slow progress in advancing from the pass of Lotzen, and most of the I. Corps was soon drawn north-eastward. The balance, however, passing S. of the lakes along the axis Johannisburg-Bialla, made marked progress, and on the night of the 9th-roth Rennenkampf decided to take down his front by successive fractions from right to left, and retire into the Mariampol region whence he had come. The battle then became one of tactical incidents, with all the local vicissitudes of a general chase. At the end, thanks to the traditional rearguard aptitudes of the Russian soldier, Rennenkampf's army had flowed away to safety, leaving the bulk of the VIII. Army congested round Vladislavov and Eydtkiihnen with the I. Corps E. of Vilkovishki and the 3rd Res. Div. at Suwalki. Goltz's southern cordon had meantime extended eastward as far as Marggrabowa.

The battle of the Masurian lakes freed E. Prussia, and the victors gleaned a harvest of some 30,000 prisoners in manifold combats amidst woods and lakes. But it was not a Tannenberg, and already events elsewhere were in progress which involved the VIII. Army in a general eastern front campaign.

As has been said above, Conrad had determined to carry out the offensive in the region Lublin-Chelm, where the Russians were concentrated, though without definite assurances of cooperation from E. Prussia. In the offensive, the forces to be employed formed two armies - the IV. Army (Auffenberg), consisting initially of the II., VI., IX. and newly formed XVII. Corps, and four cavalry divisions; and (detrainment area Yaroslav-Przemysl) the I. Army (Dankl), I., V., X. Corps and two cavalry divisions (detrainment area middle and lower San).

East of Lemberg it was intended to place two armies, the II. and III. But owing to the belief that the war crisis would be limited and localized as a campaign against Serbia, the II. Army was assembled initially on the Danube, and could only be brought N. by degrees. At the outset it was represented in Galicia only by the Army-group Kovesz (XII. Corps and some extra divisions S.E. of Lemberg and on the Dniester), but the IV. and VII. Corps were being disengaged from the Serbian front and sent up gradually. The III. Army (Brudermann) K and N.E. of Lemberg consisted of the XI., III. and XIV. Corps and some other divisions, of which the XIV. Corps was presently taken to form the Army-group of Archduke Josef Ferdinand and placed N. of Lemberg to maintain liaison between the IV. and III. Armies, intervening as required by either.

On the left of the I. Army, along the N. side of the upper Vistula (i.e. in the Polish salient) an Army-group under von Kummer, formed of Landsturm troops, and to the left of Kummer, the German Landwehr Corps of Woyrsch, were to advance in the direction of Sandomir and Ivangorod respectively, driving back such Russian mounted forces as remained in this region. These formed an echelon protecting the left rear of the I. Army, but were primarily intended to form a rallying-point for an insurrection in Poland. This hope was not realized, or realized only to a small extent, and the " Polish Legion " that was formed in fact consisted largely of Galician Poles.

The Archduke Friedrich was commander-in-chief, with Conrad as chief of staff and effective director of operations. The campaign which ensued constituted in reality a chain of battles and as such is described elsewhere. Here it need only be summarized very broadly. Apart from the movement of Kummer and Woyrsch, who started early, in order to be in position at the date of the general advance, the campaign opened on Aug. 20. Prior to that date, the Austrian cavalry divisions had made many attempts to ascertain the Russian movements in the " southern corridor " and the adjacent parts of Bessarabia, but without obtaining much information. The Russian masses were in fact still in the stage of rail transport, and their mounted troops, trained to fire action and favoured by the country, easily kept the screen intact. The Austro-Hungarian offensive was therefore in its first stages carried out according to the a priori scheme.

The objective of the I. Army was Lublin, that of the IV. Chelm; they therefore aimed at the concentration centres of the IV. and V. Russian Armies respectively, and the conditions of this concentration led to a series of encounter battles in which the Austrian left was constantly echeloned forward, with the result on the other side that the Russian V. Army's tended to strike south-westward rather than southward, and so in turn exposed a flank to the Austro-Hungarian IV. Army. This army, again, depended for security on its right upon the Armygroup (Josef Ferdinand), which was itself attracted now to the N. for intervention in Auffenberg's battle, now to the E. to protect Brudermann's exposed left. On the one side, therefore, an advance in echelon, on the other successive detrainments, produced a battle of marked day-to-day fluctuations. The I. Army in a series of combats collectively called the battle of Krasnik reached the line S. of Chodel - Borzechow - Turobin by Aug. 26, against increasing Russian resistance especially on the left nearest Lublin, where it was found necessary to bring Kummer and Woyrsch E. of the Vistula in order to strengthen the forces aiming at that place while the right advanced to Krasnosta y. On Dankl's left, meanwhile, Auffenberg was advancing into the area between the Wieprz and the Huczwa, and on Aug. 26 the battle of Komarow began. In this, between Aug. 26 and Sept. i the Austrian IV. Army broke the Russian V. Army into two fractions, the more important of which, halfsurrounded, only escaped through a maladroit withdrawal of that part of the Austrian army which had seized its line of retreat. The withdrawal of the one Russian fraction to Chelm and the other to Hrubieszow on the Bug, with heavy losses, constituted a signal victory, and would have had great results but for events in E. Galicia.

There, in accordance with the prevailing doctrines and also in order to keep Russian influences as far as possible from the Ruthenian capital, Brudermann III. Army and Bohm-Ermolli II. Army (in reality Army-group Kovesz) had been sent forward to carry out an offensive defence, although in the one army Josef Ferdinand's group was limited in its range by its liaison task,' and in the other the IV. and VII. Corps were still on their way to the theatre of war. This numerically weak offen 1 In fact, it was wholly absorbed in the battle of Komarow.

sive encountered the Russian III. and VIII. Armies in full force - as has been mentioned above, these armies had been given priority in equipment and otherwise - and was brought to a standstill in the battle of Zloczow (Aug. 26-27) fought on the line upper Bug - Zlota--Lipa. On Aug. 29-30 a new battle, defensive this time, was accepted and lost on the Gnila Lipa (battle of Przemyslany) and the III. Army fell back on Lemberg itself, which the supreme command thereupon decided to give up. It was evacuated on Sept. 2.

Thus Conrad was confronted with new problems. His left army (I., Kummer, Woyrsch) was already close upon Lublin, the victorious IV. Army pushing towards Chelm with its main body and Hrubieszow with its lesser half. In the region of Sokal and Rawa Ruska only cavalry activity had occurred, and Ruzsky's right wing was trending to the S.W. in the Lemberg direction. The beaten III. and II. Armies were assembled in good order on the strong line of the Grodek lakes (near Wereszyca), while no important attack had developed on the Dniester. There were, substantially, three courses open - to pursue the northern offensive, trusting to distance and water to make interference with the right flank impossible during the necessary time; to take down the whole northern front and come back to the Vistula - San - Dniester position; and to use the advantageous position of the IV. Army for a manoeuvre on interior lines against Ruzsky's right flank. In principle, he preferred the first course, and as we have seen, he invited Hindenburg's cooperation in the still valid Siedlce scheme. But Hindenburg declined, as Rennenkampf had not yet been dealt with, and opposition in front of the I. Army had visibly stiffened. The second alternative had obvious advantages and disadvantages; in the existing conditions, the disadvantages which had weighed heavily in peace-time - that E. Galicia was thereby abandoned - no longer applied since that region was now lost, and the preservation of the only available armies of the Dual Monarchy was of the highest importance. Nevertheless, Conrad chose the manoeuvre on interior lines, as the VIII. German army had done. It may be that Tannenberg contributed to the decision.

The germ of this idea appeared in the orders for Sept. 2, in which the IV. Army was ordered to suspend its offensive and change its front from N. to S. in readiness for a south-westward attack towards Lemberg, or for a south-eastward retreat towards the San. At the same time the lines of communication of each army were shifted westward, so that the base of the system became the region between Cracow and the Carpathians. The effort of the I. Army to gain ground northward was not given up, so that in effect, at this date, the supreme command had not made up its mind. In the orders for Sept. 4, on the other hand, the choice was definitely made in favour of a IV. Army offensive in the Lemberg direction, though the I. Army, Kummer and Woyrsch, were still left with their mission unchanged.

On the 6th, the complicated manoeuvre of the IV. Army was completed, but in its southward progress it had developed considerable opposition on the E. flank, while the W. and centre passing by Rawa Ruska and Niemirow met little or none. The result was that the army practically swung into line with the III. instead of striking from N. to S. against the assailant of that army. On the 7th, therefore, Conrad changed his plan again. The Austrian leader now proposed to take down the northern front by degrees, to use the IV. Army as a fixed pivot between Rawa Ruska and Magierow and to swing up the II. and III. Armies against Brussilov. This plan came to nothing. Russian pressure increased on the front of Woyrsch, Kummer, Dank', and the Russian V. Army, beaten at Komarow, resumed the offensive against the group of divisions under Josef Ferdinand which had been left by the IV. Army to protect its rear. Finally, Ruzsky's right, augmented by a process of regrouping which had been going on at the same time as that of the AustroHungarians, emerged in great strength on and beyond Auffenberg's left, N. of Rawa Ruska. There was no surprise, as marked indications of such a move had been discovered in the southward advance of the northern army. But when the Russian V. Army, joining the general offensive, began to drive into the weakly held gap between Dankl's right (Krasnosta y) and Auffenberg's left rear (group Josef Ferdinand about Lasczow), Conrad gave up the battle altogether and ordered a retreat to the line of the San and the Carpathians. The various forces along the Dniester retreated to the Carpathians, the II. Army to the region of Sambov, the III. to that of Przemysl, the IV. to Yaroslav, and the I. with Kummer and Woyrsch to the lower San (Sept. ii-13).

On this line, however, no stand could be made. Already on the 14th the Russian IV. Army, strengthened from the assembling IX., had been able to force a passage of the San near its mouth. The Austrians thereupon resumed their retreat southwestward (followed up in the later stages very cautiously by the Russians) and stood on Sept. 22 on the line of the Visloka, the Vislok and the Carpathians. Przemysl was left to be defended by its garrison. On the 26th the retreat came to an end on the line of the Dunajec-Tarnow-Gorlice-Usczie-Ruskie-Carpathians. But at that date, the German IX. Army was beginning to assemble in Upper Silesia. The eastern front had come into being.

In the last stages of the Marne battle Moltke had been succeeded, in effective direction of the German operations, by von Falkenhayn. Possibly because he had held, and for a time continued to hold, the office of war minister, certainly from judgment and temperament, Falkenhayn took a broad view of the eastern front problem from the first. The war, after all, had become a war on two fronts instead of two successive single-front campaigns, as had been hoped, and it would have to be conducted accordingly. This involved, first, a more intimate cooperation between the German and the Austro-Hungarian forces than had existed hitherto; secondly, the necessity of keeping the Austro-Hungarian army, in spite of its heterogeneous composition and known deficiencies, in a fighting condition similar to that of the German forces working with it; and thirdly, constant reconsideration of eastern plans, whether German or Austrian or joint, in the light of the situation on the western front; that these three were interdependent the first united operations clearly showed.

The immediate problem was to fulfil the second requirement without neglecting the third. This meant, in concrete form, the reestablishment of the Austro-Hungarian army without bringing over forces from the west. At that moment - midSept. - the battles of the Aisne were developing northward into Picardy and Artois. The " race to the sea " was in progress and the chance of decisive victory in the W. had not been lost on the Marne. On the other hand, it was clear that the AustroHungarian army had not only lost Galicia but had suffered very heavily in casualties and material, and was shaken by its experiences. The retreat to the Dunajec had on two occasions come near to disaster - in the early stages when the IV. Army's left flank was exposed and out of touch with the I. Army, and in the later stages when strong Russian efforts were made to drive the armies off their S. W. direction by enveloping the left flank of Woyrsch and Kummer. After reviewing various alternatives offered by geography and the railways, he came to the conclusion that to press the advance of the VIII.Army on KovnoGrodno, i.e. to pursue the victory of the Masurian lakes, would not serve, and decided to form a " South Army " in Upper Silesia as a direct support to the Austrian left. At first it was intended that this should be a small army, practically no more than a reenforcement of Woyrsch, but within a few days Ludendorff's proposal to transfer the bulk of the VIII. Army to South Poland, with its implication of a serious counter-offensive campaign, was accepted. The object of Falkenhayn in agreeing to this was, by enabling the Austro-Hungarian army to reassert itself in the offensive, to gain time for achieving a decisive result in the west. The theatre in which risks were taken was, as before, E. Prussia. Hindenburg's victories had altered the situation there, and a sort of pursuit could still be maintained by a small force for some time, before the inevitable reaction set in and Rennenkampf came on again. Moreover, the barrier of the lakes and the Angerapp was now being seriously fortified, and it was to be expected that Rennenkampf could be brought to a halt on that line if not in front of it. On the Mlava side, no repetition of Samsonov's offensive seems to have been feared. But as a precaution one of the 62 newly raised reserve corps was sent to E. Prussia, and two more cavalry divisions were extricated from the west. The forces of E. Prussia under von Schubert retained the title VIII. Army. Those in South Poland were designated the IX., under Hindenburg.

The Grand Duke Nicholas, meantime, was pursuing more and more vigorously the idea which was first evidenced in the creation of the IX. Army behind Warsaw. This army had been absorbed in the fighting against Dankl, but by now the more distant active corps as well as numerous reserve divisions were detrained and ready. Reenforcements had to be provided to enable Rennenkampf's I. and X. Armies to check and drive back the probable pursuit on the middle Niemen, and to reconstitute the shattered II. Army on the Narew. But even with these demands to be satisfied, enough remained for the constitution of an offensive group between Warsaw and Ivangorod. With this group he meant to transfer the centre of gravity to S.W. Poland, making Warsaw-Czenstochowa and Ivangorod-Beuthen the principal axes of his advance. Accordingly, in the last days of Sept. and Oct. r, the Russian army in front of the Austrians began to be reduced. 1 And a formidable mass - the " steam-roller " for which the world waited - gathered behind the middle Vistula. Meanwhile, lighter forces, keeping level with the advance S. of the upper Vistula, had advanced beyond Kielce, Petrikov and Lodz.

The Austro-German offensive thus struck the Russians in the act of regrouping. Its plan was: - the German IX. Army and part of the Austro-Hungarian I. Army, N. of the upper Vistula, to advance, driving back all forces met with, to the line of the Vistula above and below Ivangorod, and there to form the pivot of a sweep of the Austro-Hungarian IV. (Josef Ferdinand) and III. (Boroevic) Armies which should advance to the San, relieve Przemysl, and then strike northward and north-eastward. The II. Army (Bohm Ermolli) in the Carpathians and the left of the I. Army (Dankl) on the Vistula about Zawichost were to conform to the movement as it developed. Danger of counterattack upon the extreme left of the IX. Army from the Warsaw bridgehead was provided against partly by causing the various frontier guards of Posen, Hohensalza and Thorn to advance into Poland, partly by echeloning out a mixed force called Frommel's corps - chiefly cavalry - on the middle Pilica.

Moving out from the concentration area in Upper Silesia on Sept. 28, and joined on its right by the left of the Austrian I. Army from Sept. 30, the German IX. Army reached the line Klimontow (Austrian I. Army), Opatow (Woyrsch and XI.), Ostrowiec (Guard Res.), Szydlowice and Ilza (Xvii.), W. of Opoczno and S. of Rawa (Frommel). At that date the AustroHungarian I., IV. and III. Armies had also begun their advance; and reached the Wisloka, while in the Carpathians the II. Army and Hoffmann's Corps to the E. of it began to dislodge the various bodies of the Russian VII. Army that had established themselves in and beyond the passes. Along the whole front only light troops of the enemy were met, and the advance continued during the following days. But, almost simultaneously, the Austrian IV. and III. Armies were brought to a standstill on the San barrier and at the gap of Chyrow which gave access to the Dniester, and both Mackensen's XVII. Corps and Frommel's mixed force advancing north-westward, came into contact with the heavy Russian forces now debouching from Warsaw.

This growing intensity of the fighting S. and S.W. of Warsaw deflected the advance of the German IX. Army northward, causing a corresponding extension of front of the Austrian I. Army, which now passed wholly to the N. of the Vistula, its left centre facing Ivangorod. On Oct. to, the battle was gen The III. Army (now commanded by the Bulgarian Radko Demitriev), minus several of its units, was employed in besieging Przemysl; the VII. had come up from Bessarabia and taken over the Dniester front from about Stryi eastward. Its designation was shortly afterwards altered to that of " Dniester Group," but in 1915 a new VII. Army was formed in the same region.

eral along the whole front from Blonie W. of Warsaw, by Kalvarya S. of that city, along the Vistula and the San to Przemysl (relieved on the 9th) and thence across the Chyrow gap to the Carpathians. Here and there both sides sought to force the water barrier. In most cases no foothold was obtained, but where a bridge-head could be established, or where it existed as at Warsaw and Ivangorod, effort was concentrated.

By the 14th the assembly of Russian forces about Warsaw and Ivangorod was so great that no less than three army staffs were required to direct operations - in order from right to left the I. (brought down from Kovno region), the reconstituted II. and the IV. (from the San): on the left of the IV. was the IX., on the left of this the V., while the III., VIII. and VII. (Dniester Group) held the front of the San, Chyrow and the Dniester foreground. The process continued on the following days; the V. Army, taken out of the line S. of the Vistula, was put in between the II. and IV., the IX. was brought up to Ivangorod, and more and more Russian troops passed the bridge-heads, while the thinned lines of both sides contended on the SanChyrow - Turka front without material changes, and the opposed detachments of the Russian Dniester Group and Hoffmann and Pflanzer-Baltin fought local battles on the various routes between the mountains and the Dniester.

On Oct. 17, Ludendorff, already warned of the strength of the enemy's Warsaw armies by events and by a captured order, advised Hindenburg to retreat. The want of success on the Chyrow front indicated that the scheme for which the German IX. Army had been brought to the Vistula had failed, and the IX. Army and Da.nkl's I. Army were now exposed to the convergent attack, from Warsaw, from Ivangorod, and from Zawichost, of five hostile armies, while Josef Ferdinand, Boroevic, Beam Ermolli, and the forces eastward were pinned.

The retreat after a last attempt to gather a striking force on the Pilica for a blow against the Russian II. and V. Army - made at the expense of thinning the front of the Austrians before Ivangorod - set in on the l ist, and spread from left to right as far as the Vistula above Zawichost. The San - Turka line, on the other hand, continued to be held by the Austrians, fighting being concentrated principally upon the right of the II. Army, where a break-through was narrowly averted on the 27th. During the next days, the lost ground was regained; and progress was made between the Carpathians and the Dniester by the smaller forces operating there. But on Nov. 2, operations were suspended on the whole front S. of the upper Vistula.

During this period, the E. front of E. Prussia had been subjected to attack, as had been expected. Rennenkampf, advancing from Kovno and from Ossoviec as well as frontally, had pressed back the VIII. Army (von Schubert, later von Francois) to Kibarty and to the W. of Lyck. Francois, sanguine in temperament, defending his own corps district, inspired by a personal order from the Kaiser to protect E. Prussian territory, and conscious that the work in the lake defences was incomplete, was determined to hold his forward position to the last possible moment. Falkenhayn, objective in mind and uneasy in spirit, reinforced him with the new XXV. Res. Corps, which retook Lyck and Grayevo, threatening Suvalki from the south. The front then became quiet, for the Russians had no serious offensive intention. Their I. Army was already on its way to Warsaw when the German counter-advance took place, and the X. Army left to flank-guard the northern corridor was reduced in strength to 13 divisions, as compared with some 47 in Poland and 30 from Zawichost to Turka. On the Mlava front, held by von Zastrow with a Landwehr Corps called the XVII. Res., all was quiet in the period of the Vistula - San operations.

The retreat had been foreseen in time for the German IX. Army to make elaborate preparations for delaying the enemy's advance along the south-westerly railway lines by which, evidently, his intention was to reach Upper Silesia and the Moravian gap. In the course of the retreat the demolitions planned as well as the evacuation of stores and supplies, were carried out, if not completely, at any rate sufficiently for their purpose. But both Hindenburg's and Conrad's headquarters realized that they had now to deal with the full effort of the enemy. The " steam roller," after breakages and delays, had started. By Oct. 31 the German IX. Army had gone back to the line Syeradz - Szczercow, Novo Radomsk, Wloszczowa, Chechiny, the Austro-Hungarian I. Army to Kielce, Opatow, R. Opatowka. On Nov. I-2 the latter was driven back from the Opatowka line, necessitating the withdrawal of the IV. and III. Armies from the San and the abandonment of the offensives in progress on the E. of the Stary Sambor region. A few days later the Russians had again invested Przemysl and were advancing to the Dunajec. The centre of gravity, however, was no longer S. of the Vistula.

The crisis brought out, in the three men who had to deal with it, Conrad, Falkenhayn and Ludendorff, the characteristic quality of each.

Conrad proposed to Falkenhayn that no less than 30 German divisions should be brought over from the W. at once, bringing Hindenburg's strength up to about 53 divisions. Forces in the Carpathians and in E. Prussia were to be economized, and the bulk of the Austro-Hungarian and German Armies were to seek. decisive victory in battle in Poland. Now that the Russians had gathered, and gathered so far W., it would be possible to bring this about without fear of their retreating into the limitless interior of their own country. In short, the war could now be won in the E. It could also be lost, for unless some such decision were attempted, Conrad held that it would be necessary to retreat to the Danube. Falkenhayn, on the other hand, was becoming convinced - especially by the experience of Ypres, that the war would be a protracted trial of endurance, and must be handled on the principles adopted by Frederick the Great in the latter part of the Seven Years' War, viz. a wary, economical defensive, with offensive sorties on every favourable opportunity or necessary occasion, but no staking of all upon a throw. If Conrad was the Lee of the Central Powers, Falkenhayn was their Johnston. Had the Southern Confederacy possessed a Grant, the parallel would be complete, for Ludendorff met the problem as Grant would have met it, by a strategy that was at once objective and grandiose. Hindenburg was now commander-in-chief of the German eastern front, and his headquarters could deal with the situation as a whole. Ludendorff's plan was to transfer the bulk of the IX. Army by the Silesian railways to W. Prussia (Thorne - Hohensalza region), whence by a sudden advance through the north-western part of Poland, he could strike upon the right or right-rear of the enemy's system. To reenforce the offensive mass, the E. front of E. Prussia was to be stripped almost bare of troops and the country in front of the lakes and the Angerapp deliberately evacuated and broken up. 1 The S. front of E. Prussia (Zastrow's and other formations) was to participate in the offensive by advancing on Plock, Ciechanov and Przasnysz, with the mission of flankguarding the main attack on the E. side of the Vistula, and of keeping the Russian I. Army busy on the axis Mlava - Ciechanov; and to ensure that these forces should not be drawn away to the E., they were placed directly under General Headquarters. To fill the place of the IX. Army in S. Poland, Woyrsch's Landwehr Corps was reenforced, and by agreement with Conrad, Bdhm Ermolli with the bulk of the Austro-Hungarian II. Army, was brought on rail from the Carpathians to the upper Warta, while to the left of Bdhm the " Posen " 2 and " Breslau " corps of Ersatz and Landwehr were brought forward on the KalischSieradz line. To prepare for the worst, arrangements were made for destroying the mines of Upper Silesia. By all these drastic measures, Ludendorff expected to obtain a partial success that would suffice, without at present calling upon Falkenhayn, to provide the mass of divisions asked for by Conrad. At the moment at which the plan was put into effect, more was scarcely possible. The continuance of the retreat, especially on the front of the Austro-Hungarian I. Army which was taking 1 Von Francois resigned his command in indignation and was replaced by von Below.

This was the second reserve of Posen. The first, as Bredow's division, was already on the field.

the weight of the Grand Duke's attack on the line Kielce R. Opatowka, brought the enemy ever nearer to Cracow and Upper Silesia, and the destruction of bridges and railways on the IX. Army front could only have a temporary effect. Moreover, new dangers threatened both the eastern and the southern fronts of E. Prussia.

On Nov. 5, the Austrian I. Army had retired behind the Nida, Zastrow's advanced forces were retiring on Mlava, Below was preparing to meet a new thrust of the Russian X. Army (Sievers). On the 8th Ludendorff asked Falkenhayn for 6 to 8 more divisions as soon as possible, and for more later. The crisis, and with it the hope of decisive victory, was becoming more acute. On the 10th the regrouping was complete, except in the centre, where Bbhm Ermolli was not yet on the scene. Here, cavalry alone held the country to the N. and N.E. of Kalisch. The Posen and Breslau Corps were beginning their advance from Kalisch and Kempen respectively. Woyrsch (Ldw. Corps, ist Gd. Res. Div., 35 Res. Div.) was in front of Czestochova; from Zarki to Wielun the Austro-Hungarian I. and IV. Armies had fallen back concentrically on Cracow, in front of which they now stood; the XI. Corps covered W. Galicia; the III. Army had taken over the front of Bohm as well as its own and stood on the line Virempna - Dukla Pass - Uszok Pass, and Pflanzer-Baltin, his offensive suspended, was at Verecze, Okormezo, S. of Delatyn, R. Pruth. But the offensive group (IX. Army under Gen.-Oberst von Mackensen) was ready - the XI. and XVII. Corps astride the Warta where it enters Germany, the XX. at Hohensalza, the I. Res., XXV. Res. and 3rd Gd. Div. between Hohensalza and Thorn.

At that date the Grand Duke's Armies were thus disposed X. on E. Prussian eastern front, I. (8 divisions) on E. Prussian S. front (Plock to Mlava), with advanced troops approaching Soldau, Rypin and Lipno and one corps S. of the Vistula about Wloclawek; II. (6 divisions and a cavalry mass), W. of Lodz, advancing on Kalisch, with the II. Corps between Kutno and Lencyzka as a protective echelon; V. (8 divisions) nearing the Widawka river, cavalry approaching the Upper Warta; IV. (6 divisions) between the Pilica and Jendnziejow, pointing towards Beuthen; IX. (8 divisions) in the angle of the Nida and the Vistula; III. following up the Austro-Hungarian retreat towards and beyond the Dunajec, VIII. and Dniester Group on the Carpathian front; XI. (newly formed) besieging Przemysl.

On Nov. 11 the advance of the German IX. Army began. On the 12th at Wloclawek, parts of three corps quickly overwhelmed the corps of the Russian I. Army there, and drove it over the Vistula. The next phase was a concentric advance on the Russian II. Corps, right echelon of the II. Army, which held a position from Kutno to Lenczyca; out of this position it was driven with heavy losses on the r 5th, losing at Lenczyca the gate between the Bzura and the Ner (Warta) waterlines. Then, while part of the German army pushed forward down the Vistula to intercept any assistance that might come from the I. Army, the XXV. Res. and XX. Corps from Kutno and Lenczyca, with the XVII. and XI. Corps from the Warta valley, advanced on Lodz, the manufacturing centre of Poland.

The battle of Lodz, which began on Nov. 17, is described elsewhere. In its intensity, its vicissitudes and its significance, it was the Ypres of the eastern front. In it took place the epic incident of the break-through, envelopment and final selfrescue of the XXV. Res. Corps and 3rd Guard Division. No battle of the World War shows such varied, involved and difficult tactical situations. Here we are concerned with the results only. From the 19th the Posen Corps and Frommel's cavalry were actively engaged on the left of the XI. Corps, thus connecting the battle of Lodz with the fighting which went on all along the line to Cracow, where the Austro-Hungarian I. and IV. Armies contended without defeat or victory against the thrust of the Russian IV. and IX. Armies. Further E., the Austro-Hungarian XI. Corps and III. Army engaged, equally without decisive results, the Russian III. and VIII. Armies. But Ludendorff had undeniably won his " Teilerfolg," for the Russian onset on all parts of the line S. of the Lodz area was partly or wholly suspended in order to assemble all possible forces for the prevention of disaster on the right wing. Pressure was relaxed also on the two fronts of E. Prussia as the uncommitted reserves of the attack were taken away. In his regrouping the Grand Duke was successful; ' a continuous line of battle was formed by Dec. 6 from Ilow on the Vistula, W. of Lowicz, E. of Lodz, W. of Petrikow, W. of Novo Radomsk, and so to the Cracow battle-field. But the cost had been heavy, and the Russians were unable, then or thereafter, to resume the tidal advance on Silesia and Moravia. With the formation of this long continuous line from N.W. of Warsaw to S. of Cracow began a new phase of the struggle, in which the battle of Lodz merges into the battle of Lowicz, and that of Cracow develops, on its southern side, into the battle of Limanova-Lapanov.

It has already been mentioned that Ludendorff had on Nov. 8 asked Falkenhayn for 6 to 8 divisions to be sent at once from the W. and more later. At that date Falkenhayn was still contemplating an attempt to revive the battle of Ypres, and had not reconciled himself to position warfare. On the r8th, before the decision had fallen at Lodz, Falkenhayn in agreeing to send 6 divisions had at the same time expressed his belief that it would even so be impossible to bring Russia to admit defeat, and that the outcome - certainly desirable in itself - would only be to relieve Austria-Hungary by the reconquest of the VistulaSan - Dniester line, and perhaps of Lemberg also. But a week later, under the influence of Mackensen's victory, he said that success in N. Poland might decide not only the Galician question but the whole war. He thought this might be achieved by building up yet another striking force E. of the Vistula, where the Russian I. Army was continually giving up divisions for the battle of Lodz. Ludendorff, on the contrary, saw no prospects in such a piecemeal building up of strength which the Russians could answer pari passe. Power and surprise combined, he held, were essential. At this moment, there were in Germany 9 new divisions under training, but the awful wastage of the lives and energy of their predecessors at Ypres had convinced Falkenhayn that it was necessary to avoid cutting short their training and to give them more experienced leaders before committing them to battle. Ludendorff, in spite of the achievement of the XXV. Res. Corps at Lodz, seems to have concurred in this view. Thus the reinforcement reduced itself to a gradual incoming of 8 divisions from the W. (II., III. Res., XIII. and XXIV. Res. Corps) which, with the I. from the now relieved E. Prussian front were all absorbed in the frontal battle about Lodz and Lowicz, save one which was sent to assist the Austrian IV. Army S. of Cracow. The battle of Lowicz began and continued as a front-to-front battle in which each side sought to condense enough force for a blow, now here and now there. It ended, in mid-December, with a general withdrawal of the Russian line to a winter-position, which ran along the Bzura and Rawka to Rawa, and thence southward, crossing the Pilica E. of Tomaszew and following the upper course of that river, 5-15 km. E. of it, till near Jendnziejow it reached the Nida, to follow it to the upper Vistula.

South of Cracow, in a country of hills where manoeuvre was possible and open flanks frequent, advance and counter-advance alternated during the month of December. In conforming to the general retirement of the Allied forces in October, the Austrian I. and IV. Armies had gathered about Cracow, and during November they had maintained their front against the Russian IX. Army (battle of Cracow). At the end of the month, however, the enemy had developed a strong attack S. of the Vistula, which reached the line Wyelica - Sieprow - Droginia, and threatened by turning the fortress from the S. to make the desired breach for passage into Moravia. This danger was averted by a regrouping of the Austrian IV. Army, which enabled an attackforce to be assembled on the right wing about Mzana Dolina and Dobra, almost in the mountains. On the 3rd this force attacked northward, bringing the Russians' advance at once to a standstill, and forcing them to make new dispositions. The fighting was prolonged and heavy. On Dec. 8, forces of the Russian VIII. Army, condensed on the western flank of that army, be gan in turn to attack the flank of the Austrian attack-group, which had gained ground northward as far as Lapanow and Rajbrod. At Limanova, the scene of this flank attack, three dismounted Austrian cavalry divisions had to meet the onslaught of more than an army corps. At the same time, the centre of the Russian III. Army farther N. assumed the offensive again, and threatened at Lapanow, to break the Austrian main body in two. But resistance at Limanova continued till the Austro-Hungarian III. Army, defending the Carpathians with varying fortune, had managed to assemble a group on its left which struck in on the flank of the Russian forces about Limanova (Dec. I I). Thereby the battle of Limanova - Lapanow was decided. A last Russian force which was seeking to reach the flank of this Austrian counter-offensive was itself engaged in flank by other forces of the Austrian III. Army, and the Russians withdrew along the whole W. Galician front to R. Dunajec - Krzostek - Krosno - Lisko (Dec. 14-16). A few days later the Russians launched a fresh offensive which in the battle of Jaslo (Dec. 21-25) drove back the inner flanks of the Austrian III. and IV. Armies to the line Zaklicyn on Dunajec - GorliceUscie - Ruske - Koniecza, and pressed the front of the former back to some places behind the Carpathian line. Here, and farther E., the operations were entering on the phase known as the Battle of the Carpathians, which will be dealt with later. But from Tilsit to Gorlice, the campaign of 1914 closed in " stabilization." At this period, according to Falkenhayn, the combatant strengths on both sides were: Ios,000 Germans and 320,000 Russians E. of the lower Vistula (E. Prussian fronts); 525,000 Germans and Austrians and 847,000 Russians between the lower and the Upper Vistula; 525,000 Austrians (including 1 German division), and 521,000 Russians between the Upper Vistula and the Rumanian frontier. In sum, I,155,000 Germans and 1,688,- 000 Austrians (of whom 502,000 were German-Austrian).

Acknowledgments are due to General Y. Danilov for certain information as to the Russian plan of campaign and strategic deployment. (C. F. A.) Campaign Of January-September, 1915 By the third week of December, 1914, the struggle in the central salient had died down to a trench-warfare contest, in which the remaining energy of the troops was devoted to consolidating gains or to preventing the opponent from doing so. The situation of Ypres was reproduced in that of the eastern front at the end of the battle of Lowicz. But there was the important difference that on both flanks there was still room to manoeuvre. On the N. flank, the region of Plock, Mlava and Myszyniec was open, and the Russian army's position, in front of the Angerapp and the lakes, reached for the third time as the result of the battle of Rominten Heath (Nov. 13-16), rested its flanks on no very secure obstacles. On the S. flank, the line was continuous from Cracow to the Carpathians, but thence eastward the position was fluid. The Grand Duke, therefore, determined to assert his offensive will and power, and, confiding in the hardiness of his men, for whom winter was less terrible than for the enemy, began to group his forces with greater density on the flanks. The first signs of this tendency appeared in the counter-stroke of Jaslo, which nullified the reverse of Limanova-Lapanow and initiated the battles of the Carpathians. The second consequence was the reinforcement of the X. Army, and the re-formation, under a new army staff (XII.), of an offensive mass on the Narew.

At the outset, in the latter part of Dec. 1914, this new policy seems to have aimed at tactical results only, but in Jan. the offensives maturing on the outer flanks became evidently strategic. Interpreting the experience of the previous campaigns, the Russian headquarters could see not only the insecurity of their northern corridor, which must continue until E. Prussia had been cleared to the Vistula, and the similar but lesser risk to their left flank, but could also judge that the conquest of E. Prussia and the invasion of Hungary would be very heavy blows to the heart of the war-sentiment in Germany and Aus tria-Hungary. Reinforcements were constantly coming in, and it seemed that what the Russian headquarters chose to adopt as their plan they could impose upon the enemy. One factor, however, was already causing anxiety, that of munitions. Although the ammunition expenditure on the eastern front was on a much lower scale, both then and thereafter, than that in the W., yet even so it was far greater than had been foreseen; and Russia, with her low industrial development and her difficulties in communication with the outer world, was less well equipped than either her allies or her two opponents to meet the strain. Later, the shortage was to become disastrous and tragic; at present it was an additional argument for transferring operations to those parts of the line where trench warfare had not set in. It was not regarded as a reason for suspending the offensive, but rather for choosing for it those areas where conditions favoured human manoeuvring-power.

On the other side, the problem of 1915 was, like those of 1914, viewed differently by the three men concerned, Falkenhayn, Conrad and Ludendorff. The first named, after a moment of enthusiasm in the Lodz period, had returned to his normal method of conducting the war as a war of endurance, with limitations on particular acts of it. One of those limitations in the present instance was the necessity sooner or later of opening a way to Turkey by seizing at least part of Serbia. Another, and the principal, was the necessity of holding firm on the western front. German strategy was now paying the penalty for having doubled its fighting front there by bringing in Belgian territory. Throughout 1915, the year in which Russia was the principal theatre, just as in 1914 when it was only secondary, we find Falkenhayn working with extremely narrow margins of free strength. At a time when Germany alone possessed some 160 to 170 divisions, the adoption or rejection of operative schemes of the highest importance was made to depend on availability or otherwise of four, six or ten of them. Yet there was no remedy for this, short of a considerable surrender of occupied territory in the W.; and in the war of endurance, as conceived by the Falkenhayn school, occupied territory is an asset not to be sacrificed for the sake of a showy, but indecisive, tactical victory. The principle of working from situation to situation was, with Falkenhayn, fundamental, and in the winter of 1914-5 his projects in the east did not go beyond the formation of a German " South Army " under General von Linsingen to aid the Austrians in the Carpathian struggle. In this the motive was direct stiffening and not manoeuvre - in fact, only half of this army (4 divisions) was German. To find these divisions, the German chief had to postpone sine die his Serbian project, to which he attached very great importance; but the condition of the Austro-Hungarian army in the bitter winter fighting of the Carpathians left him no alternative, especially as the prevention of a Russian break-through into Hungary was a condition precedent of any Danube operation.

Conrad von HOtzendorff, for his part, was sanguine as ever, and the plight of Przemysl - undergoing its second and more terrible siege - continually spurred him to activity. While meeting, with local counter-offensives, the growing Russian pressure on the Carpathian front, he proposed, first an offensive in the centre of the Polish salient on Radom (scarcely a promising direction), and then a resumption of the old scheme of an Austrian and Prussian rendezvous near Siedlce. Neither was accepted by Falkenhayn, and Conrad then proposed the direct relief of Przemysl by means of a great offensive from the Carpathian line. It was for this offensive and this purpose that the German South Army was formed and, later, Bohm Ermolli's II. Army brought back from Poland. Substantially, then, Conrad, unlike Falkenhayn, was eager for battle as such. But, like Falkenhayn, he had no manoeuvre - in the true sense of the word - to propose, that was in the given conditions practicable and worth the supreme effort.

At Field-Marshal von Hindenburg's headquarters, on the other hand, the idea of manoeuvre was always uppermost. Its basis was the fixed conviction that it was possible not merely to lame but to destroy Russia's fighting power on the field of battle.

To achieve this result against superior numbers, manoeuvre was the only way; and by the term " manoeuvre " Ludendorff understood the preparation and sudden delivery of a destructive blow by locally superior force upon that part of the enemy's system which was the key of the whole. In the present case, this key position, Ludendorff held, was the Russian X. Army in the foreground of the Masurian lakes. In this quarter, and also between Mlava and Myszyniec, Russian offensives were maturing as early as mid-Jan., and in any case Hindenburg's headquarters had to consider the question of a preventive offensive in E. or W. Prussia or both. But the operative aim became higher as soon as it was known that Conrad meant to attempt the relief of Przemysl by winning through to the San and the Dniester. The four new army corps completing their training in Germany were asked for, for the purpose of a winter offensive which should not only anticipate that of the enemy, but also, in conjunction with Conrad's effort, " decide the whole war." These four corps (Xxxviii. - Xli. Res.) were Falkenhayn's cherished reserve, with which he meant to parry any great crisis that might arise out of the " Winter Battle of Champagne " then in progress, and himself to attempt a decisive offensive in France. 1 From the contemplated blow in Prussia he expected no more than the temporary and local disablement of the enemy, so that he did not think it necessary to coordinate the effort closely in date or direction with Conrad's advance. Nevertheless, " with a heavy heart," as he says, he surrendered the four corps to the east, though at first - till the Champagne crisis cleared in March - he reserved the right to withdraw them again. Actually, the XXI. Active Corps of Alsace-Lorrainers was sent from the French front, the XLI. Res. Corps taking its place there; the other three, with the XXI., went to E. Prussia at the end of Jan. and constituted the new X. Army (General-Oberst von Eichhorn).

The two operations with which the campaign of 1915 began in the W. were not, in the strict sense, coordinated, though their combined effect, owing to geographical conditions, was expected to be the destruction, according to Ludendorff, or the prolonged paralysis according to Falkenhayn, of Russia's offensive power.

Owing to the relatively low development of Hungarian lateral railways - the Galician laterals were in the hands of the Russians - it was not feasible for Conrad to form a really important offensive mass in the eastern Carpathians and the Bukovina, as Hindenburg did in the region of the Masurian lakes, without great loss of time. The struggle therefore resolved itself into surgings of frontally-opposed tides, the one seeking to break into the Hungarian plain, the other to rescue Przemysl. Although, the lines being for the most part discontinuous, tactical and local outflanking efforts, for the time and place decisive, were constantly made by both sides, there was no systematic attempt at strategic envelopment on either. At one moment indeed (Feb. 20), Pflanzer-Baltin's army group, victorious in Bukovina, sought to wheel in on the rear of the battle-field of the German South Army; and at a later stage the Russian Dniester forces were heavily reenforced for the purpose of driving Pflanzer-Baltin away and so gaining the flank of Linsingen. But in the main the opposed tides affronted each other and were broken, each in turn. In W. Galicia, the Russian offensive of Jaslo came to a standstill in the first days of January, and for the next three months nothing of importance took place W. of Gorlice. Here the Russian III. Army (Radko Dimitriev) and the Austro-Hungarian IV. Army (Archduke Josef Ferdinand) were opposed. In the middle Carpathians, where Brussilov's VIII. Army was opposed by Boroevic's III. Army and by the left wing of the widespread Pflanzer-Baltin group, the year opened with the evacuation by the defenders of the important Uszok Pass, under a local threat of envelopment. The Dukla Pass and the adjacent mountain region had already been lost, and from the Uszok the withdrawal spread east to the 1 The German contingent of the South Army had been formed from local reserves already in the east. Its staff was formed from that of the II. Corps.

Volocz and Wyszkov Passes. In the eastern Carpathians and Bukovina the Russian Dniester group (Nebel) pushed back the light forces which Pflanzer-Baltin had in the foreground of the mountains, but in the last week of Jan. the arrival of the Austrian XIII. Corps from Serbia gave Pflanzer-Baltin enough forces to enable him on the 31st to begin the reconquest of the lost ground. Meanwhile, the right of Boroevic's III. Army had held on, in spite of the loss of the Uszok and the Dukla Passes, and it was now reinforced. After covering the assembly of the German South Army about Munkacs, this wing was to constitute the striking force of Conrad's offensive for the relief of Przemysl, the South Army (including Hofmann's Austrian Corps facing the Volocz Pass) following it in echelon on the right.

The offensive began on Jan. 23, and as usual in this part of the eastern theatre, met at first only light forces of the Russians. The whole Austrian line, from E. of the Dukla to the Wyszkow Pass, moved forward, the left wing of Pflanzer-Baltin conforming. The Uszok, Volocz and Wyszkow passes were retaken by the South Army, and Boroevic's striking force reached and passed the upper San (line Czeremcza - Baligrod - LutowiskaBorynia - Smorze) by Jan. 31. But the Russians had already answered by accelerating their projected offensive against the centre and left of Boroevic (front MezOlaborcz - Konieczna) and especially southward and south-westward from the Dukla. From this point the battle was a contest of will-power and manpower. The inactive fronts were stripped of more and more divisions. Early in Feb. Bohm Ermolli's headquarters returned from Poland to their old place on the right of the III. Army (front Lupkow Pass - Uszok Pass), and on the other side Letchitsky's IX. Army headquarters were withdrawn from the Nida for the Dniester theatre. Between the end of January and the end of April the strength of the opposed forces in Poland west of the Vistula were approximately halved. In the event the Grand Duke Nicholas not only succeeded, during the first three weeks of Feb., in checking (and in forcing back somewhat) the Austrians on the Upper San, but considerably enlarged his gains S. and S.W. of the Dukla Pass, taking Mezolaborcz and the Lupkow Pass, and penetrating the Laborcz and Ondava valleys. On the other hand the German South Army made its way forward, very slowly, astride the Munkacs - Stryi railway.

Further E., the counter-offensive campaign of Pflanzer-Baltin, begun on Jan. 31, was successful in clearing all Bukovina and the Carpathian foreground as far as the Pruth on the right and the Dniester in the centre, but its left, attempting to intervene in the rear of Linsingen's opponents, was involved in heavy fighting about Krasna on the Lomnica, and in the last week of Feb. the heavy counter-attacks of the assembling Russian IX. Army drove the centre from its forward position on the Dniester. By mid-March, Pflanzer-Baltin had been forced back still farther to a line marked by the upper Lomnica - Solotvina (on the Bistrica) - Czernelica - Horodenka - Snyatin - Czernowitz, on which operations came to a standstill. These operations were however of secondary importance in which only some io% of the whole forces of each side were concerned.

The real crisis, which culminated in March, was on the front between the upper San and the head of the Ondava valley, N.E. of Bartfeld (Bartfa). As in Feb., the right of the Austrians sought to force a way to Przemysl - now in extremity - and the right of the Russians to enlarge the bridge-head in front of the Dukla and Lupkow Passes. The fighting was again intense, for the Austro-Hungarian II. Army had been reinforced for a last effort; but in the main its advance on Przemysl was definitely stayed by the middle of the month, while the Russians in the Dukla region made continuous, if slow, progress. The German South Army progressed along the railway to Tuchla, but at this stage of the battle its advance had not and could not have any great result, and its left was held up for weeks before the strong positions known as Zwinin and Ostry, covering Koziowa. Finally, on March 20, sure of the imminent surrender of Przemysl (which in fact fell on the 23rd) the Russians launched all along the front of the Austrian III. Army new attacks which, fed by troops released from the blockade of Przemysl, drove that army back to the line S. of Zboro-Kurima-Strzopko-S. of ViravaWola Michowa. At the same time BOhm's forces on the upper San front were compelled to fall back to the starting line of Jan. 23, whence they were withdrawn, in a state of exhaustion, to a line generally behind the mountain crest. The right, still in front of the recaptured Uszok Pass, was transferred to the control of the South Army.

Three weeks longer the battle lasted, but without material change, though both sides were iio,000 to 120,000 stronger in combatants than they had been in January. In the area of the Austrian III. Army two fresh German divisions, grouped as the " Beskiden Corps," arrived to stiffen the defence. In its new positions the Austrian II. Army held its own. The South Army maintained its ground also from N. of the Uszok Pass to Tuchla, and stormed at last the positions of the Zwinin (April 9) and Ostry (April 25) and Koziowa. To the E., Pflanzer-Baltin's right wing and centre, reinforced by German mounted troops, regained its positions on the Dniester and held off a new attack which Lechitzky mounted against its outer flank between Czernowitz and Usciebiskupice on the Dniester. By April 20, however, the Battle of the Carpathians was at an end, after three months of continuous mountain fighting, in temperatures sometimes as low as -22 deg. F.

The apparent effect of these battles was to give the Russians more secure possession of a bridge-head S. and S.W. of the Dukla Pass which they could not use, and to waste the remaining warenergy of the Austro-Hungarian army in attempting to relieve a fortress which certainly contained fewer men than the number sacrificed in the attempt. But in reality the indirect consequences of the battle were of much greater importance than the direct. In the Carpathians, no less than in the Masurian winter battle presently to be described, the Central Powers had managed to snatch the initiative before the Russian offensives had got under way, and thus put back the date and place of those offensives so far that the break-through into Hungary proved impossible. For the third time the " steam roller " had been brought to a standstill. Moreover it was showing signs of wear. Manpower had been unsparingly expended by the Russian command in its determination to break through; the trained officers and under-officers of peace-time were reduced to a skeleton, and the supply of munitions and even arms was becoming a very grave problem. In the majority of cases, it had been the Russians who attacked and the Austrians who defended the strong mountain and hill positions, and, though specific figures are not known, all the evidence available points to the Russian losses having been far greater than those of the Austrians and Germans. In sum, the Russians needed a pause even more than their opponents.

The Masurian Winter Battle and the First Battle of Przasnysz.- The plan of campaign formed by Ludendorff for E. Prussia, as already mentioned, aimed higher than the simple preventiveoffensive for which Falkenhayn had " lent " the four new army corps. His line of reasoning, differing from Falkenhayn's, was, and remained, this: - the war will be decided by military victory in the W.; but this victory will not be possible till after the definitive defeat of Russia, because the degree of numerical and material superiority required for the double task of breaking through the strong trench-system of the W. and exploiting the break-through in an open-field campaign was not attainable till Germany could devote practically every battle-worthy man and gun she possessed to the western theatre. Meantime, nothing was gained and much lost by using up reserves in repetitions of the battle of Ypres. Whether, in Feb. 1915, the time was ripe for such a blow as Ludendorff contemplated is however more doubtful. Both on the Mlava-Myszyniec front and on the E. front of E. Prussia the Russians were well in advance of the natural barriers protecting the northern corridor. Victory W. and N. of those barriers could only lead to a limited exploitation unless the barriers could themselves be carried in the tactical pursuit. Victory on the barriers themselves, on the contrary, would give an unlimited field for strategic exploitation inside them. In the situation of Feb. 1915, then, an effort to inflict a completely disastrous defeat on the Russians required two successive efforts, or successive maxima in a continued effort; hence a double allotment of force would have to be made. A large part of the required divisions could have been found from the army reserves of the central salient, or by thinning the line itself there, had it not been for the formation of the German " South Army," which, raising the number of divisions absorbed in the AustroHungarian front from 1 to 5, left only limited possibilities of drawing on the IX. Army and Woyrsch, for the benefit of the E. Prussian Army, which ever since Nov. had been on a very low footing. Woyrsch and Mackensen were in fact able to provide six free divisions. For the rest, if nothing could be spared from France, the eight new divisions were the only available reenforcement. Hence Falkenhayn's well-founded scepticism as to the scope of the E. Prussian offensive, and hence also Ludendorff's regret, after the event, at having parted with so much of his local reserves for the bolstering-up of a Carpathian attack.

The secret augmentation of the E. Prussian forces from the figure of so divisions (8 of which were Landwehr and Ersatz) in mid-Jan. to that of 24 in the first week of Feb. was itself no small task, and had it not been for a very fierce diversionary attack by the IX. Army at Bolimow, in the angle of Bzura and Rawka, on Jan. 31-Feb. 2 - memorable as the first occasion on which gas-shell were employed on a large scale - it is doubtful whether it would have been accomplished, for the assembly had to be made under cover of a thin screen of mounted troops, and by hypothesis the opponent himself was preparing to attack. The plan itself was comprehensive, and suggests that Ludendorff had not given up hope of being able to extract more divisions from Falkenhayn. It consisted in three main elements: - (r) the destruction, by means of breaking through and envelopment combined, of all enemy forces lying between Lyck and Tilsit, (2) the attempt to carry the Bobr line with a rush so as to break into the " corridor " south of Grodno, and (3) an advance on the Mlava-Willenberg front, in conjunction with (2), so as to bind the Russian I. Army while the X. was being destroyed and the Bobr forced.

The German forces were divided into three armies: - the VIII. (Otto v. Below) of 7 divisions (including one of the new corps), which, after covering the whole eastern front during the assembly was to form an attack front on its right wing (Johannisburg); the X. (v. Eichhorn) of 71- divisions, including the XXI. active and the other two new Reserve Corps assembled between the Niemen and the Lakes; and the Army Group Gallwitz, ten divisions, of which six came from central Poland, holding the southern front from the Orzyc to the lower Vistula.

The scheme of the German offensive, though it was to be carried out over much the same ground as the September battle, differed considerably from the plan of that battle. The winter trench-line represented the halt of the Russians after the Rominten Heath battle, in front of the Lakes-Angerapp barrier. It ran N. to S. from the Schorellen Forest by Darkehnen, E. of Lotzen to W. of Johannisburg, where it began to curve away to join the southern front. The right wing therefore presented to the Germans better chances of envelopment than Rennenkampf's right had shown in Sept., and it was on this flank that Ludendorff meant to make the chief effort. But the most significant difference was that it was now intended to treat the attack on the Russian S. flank as a break-through and not an envelopment problem. For this reason, not only was an attack-group formed behind Lake Spirding but von Gallwitz, guarding the S. front, was to occupy the Russians on the Narew and prevent them from assembling large forces against the S. side of the VIII. Army attack. Moreover the attack was to aim at seizing crossings of the Bobr at and near Osowiec. Tactical cooperation in the encirclement of the Russian forces north of Lyck was the primary but by no means the principal task of the VIII. Army's attack-group. If the power and speed of the X. Army's blow from the N. proved as great as was hoped, the exact position of the anvil on which it crushed the Russians was of secondary importance compared with the seizure of Osowiec and the Bobr by brusque attack in the Liege manner. On this, and on the progress made by the XX. Corps (Gallwitz's left) by Myszyniec on Lomzha, would depend the strategic, as against the tactical results of the whole enterprise.

The " Winter Battle of Masuria " therefore may be regarded, if not as the first great battle of the latter-day type, at any rate as in a transitional style. Although an open flank existed and was utilized to produce the tactical envelopment or " Cannae " of pre-'war theory, yet the effective victory was intended to be gained from a break-through, tactically difficult, but aimed in a strategically favourable direction.

The attack of the VIII. Army began on Feb. 7, that of the X. Army on Feb. 8, in the midst of snowstorms which, during the battle, changed to rain - the worst conditions for the carrying out of the scheme and notably its strategical part, which depended on the marshes of the Bobr being frozen hard. In sum, the X. Army drove the Russians southward without intermission from the first day. By the 10th the northern portion of the Russian line was being taken down with all speed, and by the 12th the German X. Army stood on a line from Ludwinow to Rominten Heath at right angles to the VIII. Flank guards were put out toward Pilwiszki and Mariampol against intervention from Kovno, but neither then nor later did anything more serious than threats by light forces develop on that side. Meantime, however, the VIII. Army's attack (XL. Res. Corps and parts of the I. Corps) from Johannisburg Heath and Lotzen on Lyck was brought to a standstill in front of Lyck by the fierce resistance of the III. Siberian Corps, which not only suspended the advance eastward, but led the German forces that were to the S. of it to swing north-eastward on Rajgrod so as to envelop the Lyck position. The expected Russian counter-attacks from Lomzha and Osowiec proved too feeble - being absorbed chiefly by the advance of a division of the XX. Corps farther W. - to interfere seriously with this tactical manoeuvre. But thenceforward the Osowiec portion of Ludendorff's scheme was doomed. The battle became the purely tactical " Cannae." As such, it was brilliantly successful. By Feb. 14 Lyck had fallen and the VIII. and X. Armies had made good a semicircular position from Rajgrod, by Raczki and Seyny to the N.E. corner of Augustow forest. In the forest the Russians (no longer able, for want of routes, to withdraw with speed) fought with desperation to gain time for orderly withdrawal to Grodno, the one remaining avenue of escape. But by the 18th, forces of the XL. Res. Corps from Rajgrod reached the Bobr about Krasnybor, and, on the other wing, part of the XXI. Corps from E. of Seyny drove down at all risks, parallel with the Niemen and within range of the guns of Grodno, to Lipsk, thus closing the ring round four Russian divisions left in the forest. In this extraordinary situation, the German X. Army slowly completed the destruction of the encircled Russians, who resisted for several days and made fierce efforts to break the ring, while small German forces, fighting back to back with the encircling troops, held off relief attacks from Grodno and the Bobr. Finally - but some days too late for the realization of Ludendorff's plan - the remnant of the four divisions in the forest surrendered. In all, this astonishing victory gave the Germans 110,000 prisoners, over 300 guns, and a vast quantity of stores which the Russians could ill spare.

Even before the end, Ludendorff had attempted to extricate enough forces from the W. and N.W. portions of the ring to form the attack on Osowiec and the Bobr. He reconstituted the management of the mixed-up armies as best he could by putting all forces W. of Augustowo under Below (including the XX. Corps) and all engaged in and N. of the forest of Augustowo under Eichhorn. But most of the troops destined for this were involved in the forest battle, and the Osowiec groups had to be made up chiefly out of the troops that had been crowded out of the line as the wings converged. Of the XX. Corps only one division was available, and this had advanced no farther S.E. than Stawiski and Lipniki since it moved from its concentration area three weeks before. The other division was engaged on the Omulew river, and was connected to Lipniki by a thin screen of Landsturm. In sum, it was impossible with exhausted and scattered troops to force the now sodden marsh-valley of the Bobr or to reduce Osowiec. Hindenburg therefore ordered the attacks to be discontinued.

Moreover, the position of the X. Army, far ahead of regular supplies, had become untenable, and as soon as the battlefield had been cleared it began to withdraw, just in time to secure good conditions for meeting a Russian counter-offensive from Grodno and Olita. There the Grand Duke, " by stamping his foot on the ground " - as it seemed to his opponents - had called into being a new X. Army.

This counter-offensive penetrated through the Augustowo forest, almost to Augustowo, and, to the N. of the forest zone, it reached and passed Seyny and Simno (March 5-7). But, thinking that at Simno it had found the flank of the German defence - i.e. miscalculating the promptness of the German decision to regroup on a rear line - the Simno force swung in to the S.E. toward Lozdzieje (March 8), exposed its own outer flank to counter-attack from Eichhorn's left, which stood between Simno and Kalwarja, and on March 9 fell upon the flank and rear of the Russians, at the same time as the frontal defence in and north of the forest turned to counter-attack. The Russians thereupon withdrew behind the Niemen again. The German X. Army now returned to its prepared line Augustowo-KrasnopolKalwarja-Mariampol-Pilwiszki-Szaki.

But the real crisis of the second half of Feb., which lasted till mid-March, lay not on the Niemen, but on the front of the new German VIII. Army and more particularly on that of Gallwitz. Here with his XII. Army (Plehve) the Grand Duke had all along intended to make the main effort of his Russian offensive, as geographically dictated; and the advances of Gallwitz and of the German XX. Corps, as diversions and flankguards for the Masurian battle, had merely put back the Russian preparations in time and place. Anger at the disaster to the X. Army, and fears for the safety of the "corridor" at its sensitive point N.E. of Osowiec, caused the Grand Duke to divert forces from the XII. Army to form the new X., but without affecting the mission of that army, which accordingly took the offensive against Gallwitz about the same time as the struggle in Augustowo forest came to an end. At the same date the attempts of the German VIII. Army against Osowiec and the Bobr line were dying out, and the division of the XX. Corps north of Lomzha was pinned by heavy counter-attacks from that place, while the other division of that corps was making head on the Omulew against similar efforts from Ostrolenka, and the Landsturm screen between them was holding its ground with difficulty against other attacks from Novograd. The crisis, from the German point of view, was so grave that even in Ludendorff's memoirs, written four years after the event, satisfaction in the " Cannae " of Augustowo is almost completely smothered in the remembrance of anxieties, makeshift reinforcements, and critical decisions concerning the S. front. All energy on both sides was now focussed on this front.

In the winter of 1914-5, light forces of the Germans had been advanced, originally as an element of the battle of Lodz, a considerable distance S. of Strassburg and Mlava, and the reinforcement of these troops to the strength of an army group had taken place on this forward line. Gallwitz had then advanced, in conjunction with the Masurian offensive, deep into the concentration zone of the Russian XII. Army (Feb. 13). In a few days he had reached the line Plock-Racionz-Przasnysz. But by about the 24th, Plehve's interrupted concentration was sufficiently near completion for him to advance. Pressing the front of Gallwitz on each main route, he developed his greatest strength in the Orzyc and Omulew valleys. In the latter, the division of the German XX. Corps above mentioned engaged the Russian advance in a series of combats which in the event were undecisive; but in the Orzyc region the Russian blow upon Przasnysz succeeded in driving back three divisions under v. Morgen (I. Res. Corps) with very heavy losses (Feb. 25-27). The whole centre and left of the German line then fell back, pursued by the Russians, to the line Radzonovo-Mlava-Chorzele. On and about this line fighting remained severe till about March 19, kept alive on the German side by successive reenforce ments from the X. Army, and by assumption of responsibility for the Omulew front by the VIII. Army, which enabled Gallwitz to group his forces more closely on the Chorzele and Mlava fronts.

The crisis died away in local attacks in the latter half of March. The Russians were becoming weak in munitions; the Germans continued weak in men. The last fluctuations of the battle brought the Germans from Chorzele close up to Przasnysz.

Thus the E. Prussian offensive of the Russians closed, in the same way as the Carpathian offensive was soon to close, with little gain and great loss of leaders and of irreplaceable ammunition. The Germans, on the other hand, like the Austrians, had failed to achieve their strategic purpose. The general results were thus, for both sides, negative, in spite of the accomplished " Cannae " in Masuria - itself an exhausting effort.

On both Prussian and Galician fronts a pause of some weeks was imposed by the weariness of both sides. The latter part of March and early April in the N., and the last half of April in the S. were devoted to discussion and formation of plans. At this stage the distribution of force was, according to an Austrian headquarters statement of April 20, as follows, in rifles and carbines: East Prussian fronts: 263,000 Germans of X., VIII., and Gallwitz Armies, and 508,000 Russians of the X. and XII.

Armies. Frontage 380 km.

Central salient: 245,000 Germans and Austrians of IX., Woyrsch, and Austro-Hungarian I. Armies, and 436,000 Russians of the II., I., V., and IV. Armies. Frontage 275 km.

Upper Vistula to Beskidengebirge: 108,000 Austrians (IV. Army) and 10o,000 Russians (III. Army). Frontage iio km.

Carpathians and Bukovina: 385,000 Austrians and Germans (III., II., S. Army, Pflanzer-Baltin) and 496,000 Russians (Viii., Xi., and IX. Armies). Frontage 387 km.

In sum, there were i,001,000 Germans and Austrians to 1,540,000 Russians.

According to the distribution table given by Falkenhayn for the end of that month, 366,000 German combatants faced 640,000 Russians between the Baltic and the Vistula; 184,000 Germans and 54,000 Austrians, in all 238,000, were opposed to 407.000 Russians in the Polish salient; and 89,000 Germans and 610,000 Austrians, total 699,000, stood on the W. Galicia, Carpathian and Bukovina fronts against 7 20,000 Russians. In sum, 1,303,000 soldiers of the Central Powers to 1,767,000 Russians. The difference between the two sets of figures is partly accounted for by the fact that artillery personnel is included in the second and not in the first set: but whether taken separately or together, the figures throw a strong light on the state of the Russian army on the verge of the tremendous campaign of summer, 1915. It will be noticed that the total of 1,767,000 combatants is approximately the same as the mean monthly strength with the colours in peace (1,700,000). At this period no considerable forces were maintained in any but the eastern European theatre, so that, in effect, practically the whole of Russia's resources in men had been absorbed in maintaining the formations existing in peace and some 35 reserve divisions created on mobilization.

It will be noticed also that under the imperative needs created by the two-front war the German forces in the East had trebled, as compared with the strength at the time of the Masurian lakes battle in Sept., but that the Austro-Hungarian forces, though far above the nominal figure of Sept. 1914, were well below their mobilization figure. Hitherto, it must be remembered, the policy of " winning the war in the East " had not been accepted by Falkenhayn, and the German increases represented simply defensive and counter-attack requirements, and in particular the relief of pressure on the Austro-Hungarian armies. Correspondingly, German ideas and execution began from this date to predominate over Austrian. But no effective united command was ever created. German interferences in Austrian operations and operative methods, imperatively necessary to the common cause, but very often tactless, were constantly resented by Conrad and by most Austrian leaders; and moreover great divergencies of policy developed between the two imperial Governments in respect of Poland, Italy, and the Balkans.

Falkenhayn neither then nor thereafter accepted the principle that a decision could be obtained in the East. But his ideas had undergone a change since he conceded the eight new divisions to the eastern theatre " on loan." The French attempt to break through the Champagne lines had failed. A large number of German divisions were being reorganized on the basis of three infantry regiments instead of four, and the forces thus obtained were grouped in new handy divisions of veteran troops, which gave greater freedom in the play of reserves. He had abandoned, after detailed study, his Jan. prospect of a break-through on the Albert - Arras front, and therewith all offensive plans in the western theatre, while Conrad had refused to agree to his renewed proposal to force a way through Serbia for munitions for Turkey, though the peril of a Dardanelles break-through was becoming more and more evident. On the other hand, indices collected both on the Carpathian and Prussian fronts pointed to a growing shortage of material on the Russian side, as well as to a decrease of efficiency owing to losses in leaders and pre-war soldiers. Falkenhayn further thought it possible to keep both Italy and Rumania neutral, at least for a long time. All things considered, he came to regard a very heavy blow on the Russian front as necessary, possible, and desirable; and on Conrad's reviving, on April 7, the old scheme of combining blows from the lower San, and from the S. front of E. Prussia, with a rendezvous near Siedlce, he agreed, not indeed as to the plan, but as to the principle. It was still only a " sufficiently " heavy blow that he intended to deliver, but the limitation implied in the adverb was considerably relaxed. Eight divisions (Guard and X. Corps, XLI. Res. Corps, and two of the new divisions) were to be brought over from the western front, this time simultaneously and for use as an army. Of this army (XI.) Mackensen was appointed chief, with Colonel v. Seeckt as his chief of staff, Prince Leopold of Bavaria succeeding Mackensen at the head of the IX. Army. To cover the withdrawal from the W., sharp local actions were initiated at different points on the trench-line. One of these, involving ten or more divisions, is known to history as the Second Battle or " gas attack " of Ypres.

The theatre of Mackensen's operations was to be the country between the upper Vistula and the mountains (Dunajec - GorliceTarnow), where the front of contact was in much the same position as it had been at the end of the battle of Jaslo. It was held by the Russian III. Army (Radko Dimitriev) on the one side and by the Austro-Hungarian IV. Army (Joseph Ferdinand) on the other, both being relatively weak. Supposing surprise to be effected, a mass of eight first-class divisions, supported by the troops already on the front and by artillery on a scale never before seen in the East, had every prospect of breaking through. Falkenhayn took many precautions to secure his surprise, and in the main with success, although the Russians and their Allies were well aware that a blow was impending at some point of the eastern front. The troop trains were sent by roundabout routes, false rumours were circulated, and Conrad himself was not informed of Falkenhayn's decision till the movements of concentration had begun. Hindenburg, whose jurisdiction only extended to the left of Woyrsch's line, was instructed to make demonstrative attacks at different points. One of these, the raid of v. Lauenstein's group into Courland, had an important sequel, and will be discussed later. The significance of the other two, an attack at Suwalki by the X. Army and a gas attack near Skierniewice by the IX. Army, was only momentary. In direction the attack was partly frontal, and it has been criticized for that reason. But a prime factor was the necessity of relieving the situation for the Austrians on the Carpathian front as soon as possible; and, besides in all probability compelling the Russians to retire in the southern part of the central salient, a drive N.E. and E. from the front Gorlice - Tarnow would make the Russian positions in the Carpathians untenable at least as far as the Lupkow pass inclusive. Falkenhayn went further, and proposed to involve the Russians even more thoroughly in mountain difficulties by retiring the right of the III., and the II. and South Armies. To this, however, Conrad would not agree; and Mackensen's blow lost part of its effect through this refusal.

Reinforced by the Austrian VI. Corps already on the front, and placed in general charge of the Austrian IV. Army as well as of his own, Mackensen was himself subordinated to Conrad's headquarters, though in fact no major decision could be taken without Falkenhayn's agreement. On the Nida front the Austrian I. Army, and in the Beskidengebirge the Austrian III. Army, stood on the flanks of the two attack armies, and in case of success would be carried along as supports. On May 1 (see Dunajec-San, Battles Of) Mackensen's artillery preparation began. The scale of artillery and trench-mortar strength - hardly higher than that of a quiet sector in France in 1918 - was, for the East and for 1915, overwhelming. At night, as a final diversion, an Austrian division crossed the Dunajec a little above its mouth and established two bridge-heads. On May 2 Mackensen's attack was launched between Woynicz on the Dunajec and Mlalastov (S.-S.E. of Gorlice). The troops of Radko Dimitriev gave ground, fighting stubbornly. By the 6th they had retired with heavy losses beyond the Wisloka; and the Austrian III. Army, taking up the attack in echelon rightward, had regained the Dukla Pass. By the 9th Mackensen had forced the Wisloka, Boroevic was at the evacuated Lupkow Pass, and even Linsingen's left was advancing. On the 11th, on the other flank, the Russian IV. Army evacuated the Nida position, pivoting on Kielce. Operations were fluid, and it was Falkenhayn's and Conrad's problem to maintain them so.

Falkenhayn's intention was to ensure this by making the operation continue as a tactical one, with as little regrouping as possible outside the limits of the battle that was in being. For this reason he rejected a proposal of Conrad to reinforce, at Mackensen's expense, the Pflanzer-Baltin group (now called VII. Army), which by reason of its position might be enabled thereby to reach the rear of the Russian southern wing. He ignored the relief offensives started by his opponent against the front of Pflanzer-Baltin and elsewhere, and he even sought to utilize the attack upon Pflanzer-Baltin as a means of setting in motion the German South Army and the still stable portion of the Carpathian front, E. of the Lupkow Pass. But at first he had no intention that the effort should go in the slightest beyond its tactical limit, which he fixed as the San-Dniester barrier. Conrad agreed. Both leaders were anxious to disengage large forces for use against Serbia or Italy or both.

As foreseen, the rush of the Gorlice offensive came to a stand on the San-Wisznia line. The Grand Duke had, under cover of his relief offensives, collected adequate forces on the III. Army front and was prepared to hold it firmly. By the 14th Mackensen had taken a total of 140,000 prisoners and more than loo guns, and had reached the line Tarnobrzeg on Vistula (link with Opatowka line)-Nisko on San-Sieniawa (Austrian IV. Army); SieniawaJaroslaw-Radymno (XI. Army); Magiera and Chyrow region (III. Army); Stary-Sambor (II. Army). But along the lower San, in the bridge-heads of Jaroslaw and Radymno and the fortress of Przemysl, the Russians were ready to fight again, on the alert, in prepared positions, and had by demolitions of all sorts made the supply problem difficult for the Germans and Austrians. At that date Brussilov's VIII. Army and Shtcherbachev's XI. on its left were intact; Szurmay's and the left of Linsingen's were only beginning to advance; while Pflanzer-Baltin was on the defensive along the Pruth except at Kolomea where he still held a bridge-head. Moreover, Italy was on the point of declaring war (as she did on the 24th) and Rumania's intentions were impenetrable. On the western front, the French and British had opened their relief offensives of May 9 (battles of Carency and Festubert). The Dardanelles was under military as well as naval attack, and the Turkish and Balkan problems, always obscure, had thereby become acute as well.

Nevertheless, during the fourth week of May, Falkenhayn finally determined to carry on the Galician offensive and even to extend it. It appeared, from Mackensen's reports, that the shortage of munitions on the Russian side, already observed here and there, was general, and that it was possible in consequence to keep the offensive alive till it had secured a decision " sufficient for our purposes," in Falkenhayn's own words. Fresh troops were drawn from the West - in spite of the crisis north of Arras. Hindenburg was invited to press the advance of Woyrsch's army group - which had already begun on the 12th to move forward on the left of the Austrian I. Arm y and was in front of Radom by the 16th - up to the Vistula below the San confluence. As in the Vistula-San operation of October 1914, the threat of turning the San line by Josefow was thought to be an effective means of weakening it against frontal attack. Ludendorff, however, declared this operation to be impossible, in spite of the offer of fresh divisions - his mind was already set upon a more grandiose scheme. Falkenhayn thereupon gave the incoming divisions (21 from France and 2 from Poland) to Mackensen, and on June 3 that general received instructions to push the XI. and IV. Armies over the San barrier, south of the Tane y -, in coOperation with an eastern advance of the Austrian II. Army (now comprising what was left of the III. after Boroevic's departure for Italy), which should " finally" beat the enemy still remaining south of the Dniester in front of the South and VII. Armies. Hindenburg was merely " to take any chance that offered itself anywhere of profiting by the enemy's shortage of munitions." In sum, then, the scheme was simply a prolongation eastward of the Gorlice-Tarnow effort by means of a fresh engagement of reserves. No new operative idea was involved. But the decision to continue the battle was in itself an operative decision of the first importance, and, in view of the general war situation, a very bold one.

Mackensen meanwhile, partly urged by his own fighting spirit, partly compelled by Russian counter-attacks, had been involved in constant fighting on and for the San line. The Austrian IV. Army was strained to the utmost in holding on to the positions it had gained on the middle San (below Sieniawa) and in front of the link Nisko-Tarnobrzeg (or " San angle position ") which joined Radko Dimitriev's front to that of Evert on the Opatowka. The right of the XI. Army was similarly held up by the Russian positions about the Radymno bridge-head, and Przemysl interposed a formidable obstacle between that army and the advancing Puhallo group (the relic of the Austrian III. Army, which included also the German Beskidenkorps). But the left of the XI. Army stormed the Jaroslaw bridge-head and, crossing into the Lubaczowka valley, pressed the right rear at the Radymno bridge-head farther up the San. On May 24 a general assault carried this line, and the Russian centre, its right still holding the " San angle " position and the San below Sieniawa, fell back to the line of the Wisznia, the Grodek lakes, and the Wereszyca. Practically at the same time, the right of the XI. Army, Puhallo, and the left of the II. Army closed upon Przemysl from the N., W., and S.; after severe fighting the fortress fell on June 3, as described under Przemysl. Farther E., the right of Bohm-Ermolli's and the South Army, advancing in the last ten days of May, reached the line Weliko Bloto (" great marsh ") on the Dniester-E. of Drohobycz-S. of Stryj-Dolina, making connexion at Jasien with the left of the VII. Army, which was holding, still with success, the Pruth line.

The Russians, however, failing as were their resources, reacted powerfully. The Grand Duke's instructions were that " for political reasons, it is imperative to hold " the Opatowka-SanGrodek line " at all costs," and he carried them out by a series of heavy counter-strokes. First on the lower San against the Austrian IV. Army, then on the Pruth against Pflanzer Baltin, and lastly against Linsingen on the Stryj front, offensives on a large scale were delivered in the latter half of May and the first week of June. New masses were drawn from an army at Odessa which was to have cooperated in the attack on Constantinople. Even Woyrsch's advance, far away on the Kielce-Radom railway, was opposed by stubborn defence and sharp local counter-attacks. But in the last resort the Grand Duke's forces were inadequate for prolonged defence. The long exposed flank of the northern corridor compelled him to keep fairly large forces inactive on the Narew, the Bobr, and the Niemen; and the Lauenstein operation in Courland (described below) made a continual drain on his northern resources. But above all, the failure of munitions led to enormous losses, both in counter-attacks and in rearguard operations. By June 16 the Russians had lost, in the battles of Gorlice-Tarnow, the San, Stryj and the Pruth, no less than 392,000 in prisoners alone, besides 304 guns. The last acts of this phase were the forcing of the Grodek lines by Ma.ckensen's two armies on June 16-19 (see Lemberg, Battles Round, Section II.) and the successful two fronted battles of Linsingen's South Army about Stryj and Drohobycz, in which his left, facing north, held off the counterattacks of the Russian XI. Army, while his right, by intervening in the flank of Pflanzer-Baltin's opponents (IX. Army, Letchitsky), made them retreat to the Dniester (May 31 - June i 5).

The Bug and the Narew Campaigns. - Although the Russian retirement in E. Galicia was not, as Conrad imagined for a moment after the fall of Lemberg on June 22, a retreat in dissolution, it was definitively a retreat on the largest scale. Once the gap between the San and Dniester had been forced, neither was tenable by the defence. Very soon, therefore, the Russians on the Dniester were taking down their line from right to left, to re-form on the positions offered by one or another of the N. - S. tributaries of that river. From the San, the Tane y and the region of Rawa Ruska, the retreat took a northerly direction and thus there came up again the same possibilities, risks and alternatives for the Austrian offensive as those of Aug. and early Sept. 1914. The conditions were, however, partly changed. The Russians and Austrians alike had lost most of their peace-trained leaders and their offensive energy. Instead of the general clash of an encounter battle, it was now a case of retreat and of a follow-up, upon which delay was imposed by the necessity of restoring demolished communications, and caution by the risk of counter-attack striking the pursuit at a weak spot as it opened out fan-wise towards Lublin - Chelm and towards Sokal. Such a counter-attack did in fact bring the German XI. Army into momentary peril between July 7 and July 12.

The prospect of a slow advance of indefinite depth made it imperative for Falkenhayn and Conrad, and especially for the former, to reconsider the position. Hitherto the German leader had proceeded from one limited objective to another, all along the same general direction. Now - at the beginning of July - the choice had to be made between initiating a far larger operation and calling a halt to consolidate gains.

There were in reality two decisions to be made, one of principle and one of method. On the principle of continuing the offensive against Russia, Falkenhayn's opinion was, fundamentally, unchanged, and he foresaw new dangers in France owing to the impending appearance there of 12 British new army divisions, considered as heralding an attack. But deciding, on the evidence, that the great French offensive would not take place till Sept., and relieved of fears for the Italian front - impressed also, without doubt, by the repeated counter-strokes of the Russians - he decided on June 28 to initiate a new eastern offensive effort.

The second decision, as to the form and direction of this effort, was more difficult and controversial. Apart from Conrad's proposal, made once again, to strike from two directions against Siedlce, there were two schemes under consideration. One was from Ludendorff on behalf of Hindenburg; the other from von Seeckt, representing Mackensen. In the sequel, Falkenhayn accepted the latter, with additions of his own.

In Section I. of this article, mention was made of the geographical harriers, both flank and transverse, of the " northern corridor," and it was noted that the tendency of the latter was to turn southward in their upper courses, so that a series of gateways existed along the inner flank of the corridor. Seeckt's proposal, first made as early as June 15, was to wheel the two Mackensen armies sharply northward, pivoting at about the mouth of the San, to the line Ivangorod - Wlodawa, with, as flank-guard against dangers from the Luck direction, the Austrian II. Army, which should advance, in echelon from the left, toward Vladimir Volhynskiy, E. of the Bug. Only the South Army and the VII, Army would remain to drive the Russians remaining S. .of Brody out of E. Galicia. The Austrian I. Army on the other side of the Vistula was to conform by pushing the enemy back to about Josefow, and, itself crossing there, to come into line to the S.E. of Ivangorod, thereby allowing Mackensen (I., IV., and XI. Armies and Beskidenkorps) to condense on his right wing and drive forward on the Bug, with on his right, beyond the river, a deep echelon which could pull out and outflank the enemy's left wherever it was found. To this scheme it was open to Falkenhayn to add a similar enveloping element on the northern flank.

But, in accepting the plan, Falkenhayn and Conrad modified it considerably. The situation in E. Galicia did not seem to them to justify the plunge of the II. Army northward on Vladimir Volhynskiy. They therefore reserved this army, as heretofore, for operations in the Brody direction, and instead withdrew the I. Army from the central salient - Wovrsch extending, in place of it, to the Vistula - and reconstituted it about Rama Ruska with orders to line the Bug as a flank guard in proportion as Mackensen progressed. It was during this regrouping that the Russian counter-attack of July 7, above mentioned, was delivered. A serious objection to Seeckt's scheme was, in Falkenhayn's eyes and probably in Conrad's also, the fact that the II. Army would have become involved in the marshes of the Pripet region N. of Vladimir Volhynskiy. Both Seeckt and, incidentally, Ludendorff considered the difficulty of this country to be exaggerated, and Falkenhayn admitted after the event that this was so. In any case much would have depended upon the scale of the operations E. of the Bug, and this was just the unknown factor in the problem.

Falkenhayn therefore limited the Mackensen operation to the area between the Vistula and the Bug, thus turning some, but not all, of the transverse barriers by their inner gates. Reckoning upon obstinacy in the command and slowness of the machinery of his opponent, he considered that it would suffice to come in upon the rear of the Russian centre during its presumed evacuation of the central salient, at some point between Siedlce and Brest-Litovsk. But he was aware that the centre of gravity of the whole Russian line now lay opposite Mackensen, who would be called upon to make a purely frontal advance through country that was destitute of railways and would certainly be devastated. He therefore intended to deliver an additional blow from the other wing generally in the same direction; that is, to reinforce Gallwitz to such strength as would enable him to force, in succession, the Russian XII. Army's Przasnysz lines and the Narew barrier, and so to descend upon the same region from the other side, N. of the middle Bug. Thus he expected to obtain the maximum result that was possible, and within a time-limit set by the forthcoming French offensive in Champagne and by the Bulgarian peasants' harvest.

Ludendorff, on the other hand, aimed at the " annihilation " of the Russian armies and thereby the certainty of winning the war. He argued that Mackensen's movement on the left of the Bug would be a slow frontal drive; that a Gallwitz offensive toward the Narew would be brought to a standstill, or at the least reduced to the condition of Mackensen's, very little beyond the Narew; that Byelostok could not be reached with certainty by an offensive from the VIII. Army front (Osowiec), such as had been projected in the Masurian campaign, though he and Falkenhayn were agreed as to this being, ideally, the decisive point; that Kovno and Grodno effectively held the middle Niemen line; and that, in effect, the only practicable envelopment was one which, starting from the N. of Kovno, swept round and invested that fortress and swung in by Vilna toward Molodechno and Minsk. The cross-barrier of the Vilya, and that alone, was sufficiently far back from the present Russian front to ensure the cutting-off of the entire Russian army in Poland, Polyesie, and southern Lithuania. To complete the " Cannae," he proposed that the Mackensen group of armies should place its centre of gravity on, and even E. of, the Bug, as laid down in Seeckt's original plan.

To understand the significance of this proposal and the arguments for and against it, it is necessary to realize the new position of affairs on the extreme left of Hindenburg's front. At the close of the Masurian winter operations the X. Army leaned to the left on the lower Niemen, rather east of Tilsit. In March there had been some advances and retreats on both sides but no substantial change in the situation. A raid on Memel, beyond the left flank of Eichhorn's Army, by a small body of Russian militia from Libau (who were expelled after doing some damage) was the only incident of importance N. of the Niemen till, in mid-April, Hindenburg received orders to deliver feint attacks in order to divert attention from the forthcoming Gorlice offensive. He chose, for this purpose, the region N.E. of Tilsit, and formed a mobile army group of infantry and cavalry divisions under General von Lauenstein. In this quarter the Russians had only small forces, and the advance could be carried out in three separate columns, thus covering an enormous front. In all, 3 infantry and 3 cavalry divisions were sent out on April 27, by Memel toward the Vindava, by Tauroggen on Shavli (Schaulen, Szawle), and by Yurburg on Sredniki and the Dubissa line. A small raiding body, in conjunction with light naval forces, took possession of Libau early in May.

Lauenstein's movement was unexpected, and his left column penetrated to Mitau before the reaction set in. The others made good Shavli and the line of the Dubissa, and during May and June a series of fierce battles on a small scale took place all along this line. The Russians brought up considerable reinforcements under the V. Army staff, and the first object of Lauenstein's enterprise thus attained marked success. But, like other wide extensions of front in the war, as soon as serious infantry fighting opened, manoeuvre began to call upon reserve resources for its maintenance. Two infantry and 2 cavalry divisions were added to the German force, which became the "Army of the Niemen" under Otto von Below, Scholtz succeeding this officer as the head of the VIII. Army. Thus, at the end of June, when the plan of future operations was being settled, the ground was prepared for the manoeuvre advocated by Ludendorff. From Shavli, with flank guards set out successively towards Riga and Dvinsk, the Niemen army could, after being made sufficiently strong to defeat the Russian V. Army assembled in front of it, turn Kovno and reach the Vilya line long before the Russians in retreat from western Poland could do so. On the other hand, so grave a peril would clearly bring into existence a new Russian army of relief in the Riga - Dvinsk - Petrograd region, and this army would make short work of a few flank-guard divisions facing Riga, Jakobstad and Dvinsk. One necessary condition of Ludendorff's plan, therefore, was heavy reinforcement of the Niemen army; another the reduction of Kovno, so as to clear a direct and safe line of communications Insterburg - Vilna and to bring the X. Army into action E. of the Niemen. From Falkenhayn's point of view, however, the eccentricity of the whole manoeuvre was its gravest drawback. He doubted whether so distant an operation would affect the situation of Mackensen, but especially whether it would not become just that plunge into the unlimited interior of Russia which, with his time-limit fixed, he dreaded above all. Operations N. and E. of Kovno were permissible, in his opinion, only for hunting down an army already in dissolution, not as a preliminary to the battle that was to bring about that dissolution.

Such, in sum, were the elements of a controversy between Falkenhayn and Ludendorff, which in the course of the summer created a serious breach between the Supreme Command and the commander-in-chief East, and undoubtedly handicapped the operations, for Falkenhayn never swerved from his intention to close down the campaign as soon as an " adequate " result had been achieved, and Ludendorff on his side returned to the charge at every opportunity, with the result that the few available reserves were handled without singleness of purpose.

The Ludendorff plan, first proposed as early as June 7, was discussed fully at a conference on July 2, in the presence of the Emperor William, who, bound by the practice of the German army either to follow the counsels of his sole and responsible adviser or to dismiss him, chose the former course.

It was decided therefore that Mackensen, after completing his wheel-up, should advance with all possible energy against his immediate opponents between the Vistula and the Bug, with the reconstituted Austrian I. Army protecting his right flank by making good the line of the upper Bug as he advanced; and that Gallwitz's army group, reinforced, should break through at Przasnysz and on the Narew. When Gallwitz's operation, with its immediate relief to Mackensen, should have been completed, then Falkenhayn was prepared to allow an extension of the offensive to the middle Niemen region.

On Mackensen's front the wheel-up was completed in the midst of a heavy Russian counter-attack, and the advance that was to follow was involved in great difficulties from the outset. His three armies - from left to right, the Austrian IV. and the German XI. and Bug Armies (the last newly formed under Linsingen) - had not moved appreciably when Gallwitz's attack was delivered. The Russians had massed considerable forces to deny access to the inner gates of the corridor, and under cover of their activity had already begun the evacuation of the central salient. There all the old line had been already given up S. of Inowlodz on the Pilica, and, on Mackensen's intention becoming evident, the retreat was continued to the line of the Vistula itself, where, however, the foreground of Ivangorod and, especially, the great entrenched positions west of Warsaw continued to be held in force. The German IX. and Woyrsch Armies in front of this line, now constituted as a group of armies under Prince Leopold (probably in order to give Falkenhayn a force independent of both Hindenburg and Conrad), had been weakened and could do little more than follow up, boldly on the right but very cautiously on the left where the Warsaw positions and Novogeorgievsk imposed respect.

When Woyrsch reached the region of Ivangorod (July 21) so little progress had been made on the Mackensen front that Conrad proposed that Woyrsch should cross the Vistula above that fortress, so as to intervene in rear of Joseph Ferdinand's opponents. This movement, which would have thrown the axis of Woyrsch, and eventually that of the IX. Army also, away from the region of the middle Bug and put an end to all hopes of cutting off the Warsaw group of the enemy, was opposed by Falkenhayn and also by Mackensen, and Woyrsch received orders to cross the Vistula below Ivangorod, as he did on the night of July 28-29 near Muciejowice. The IX. Army meanwhile felt its way forward to the Warsaw lines and the S. front of Novogeorgievsk.

Before any of these movements were under way - largely indeed with the intention of helping them to get under way - the Gallwitz group, reinforced from the central salient by 4 divisions to a strength of about 15, had opened its offensive on July 13-16 by breaking through the Russian XII. Army' s trench-lines at and west of Przasnysz (see Narew, Battles Of The). On the night of the 17th Gallwitz stood within range of Ostrolenka on the left and the N. defences of Novogeorgievsk on the right. But a new and more severe effort was needed for the forcing of the Narew line itself. Russian counter-attack forces arrived in time, and it was only on Aug. 8 - more than 3 weeks after the offensive began - that the Gallwitz group, now styled XII. Army, had made good a line E. of the river defined by Serock - Wyszkow (on the Bug) - E. of Ostrow - R. Ruz, the last named being occupied by the right of Scholtz's VIII. Army which had advanced in sympathy. The right of the German XII. Army meantime, W. of the Narew and facing S., was holding its own, not without considerable difficulty, against repeated counter-attacks issuing from the Novogeorgievsk defences, where the Grand Duke maintained large mobile forces up to the eleventh hour - and indeed beyond it.

In these 3 weeks Mackensen's right, the Bug Army, had been engaged (see Brest-Litovsk, Battles Of) by the Russian XIII. Army, at the halt on almost every line of E. - W. streams available. It had fought on the line Grabowiec - Grubieszow from July 19-21, on that of Chelm - Annopol from the 21st to the 31st, and along the Ucherka river and at Sawin in the first days of August. The XI. Army, with better conditions, had advanced first astride and then east of the Huczwa, and by Aug. 6 had reached Lubartow - Sawin; while Joseph Ferdinand had - without the suggested flanking assistance from Woyrsch - reached the line Novo ,Alexandryn-Lubartow. In the centre Woyrsch had extended his Muciejowice bridgehead and was passing all his forces over the Vistula for the advance on Siedlce-Lukow, and under this threat the Russians had entirely evacuated the left of the Vistula. Warsaw city fell on the 5th, though the German IX. Army was unable to force the river - there a kilometre broad - till the 8th. Ivangorod was evacuated on the 5th. Thus the German front had assumed a still more pronounced N.E. direction than at the beginning of the Mackensen manoeuvre; owing to its battle and route conditions, Linsingen's Army was back instead of forward of the alignment, and the Russians had retreated clear of the dangerous central salient to a line marked by the Liwiec, the Bystrica and, facing Mackensen, the middle Wieprz, the Swinka and the Ucherka. The Austrian I. Army, occupied principally with flank-guarding Linsingen along the Bug, had advanced its right to Vladimir Volhynskiy but no farther. On the other flank of the Russian retreat Gallwitz was firmly held for the time being. In other words the Russians - handled with great skill by General Alexeiev, commander-in-chief of the N.W. front, were successfully effecting their retreat to that line (Kovno-Grodno-Brest-Litovsk-W. of Kovel-LuckDubno) which had been already in peace-time regarded as the line of safety for deployment. In territory, they had abandoned no more than they would have been prepared to give up gratuitously in their pre-war concentration scheme.

But this in itself was, after a year of warfare, a confession of defeat. The enormous losses of that year in men and material - losses such that the great army of peace-time with all its resources had practically ceased to exist and the stocks of arms no longer sufficed to equip even the men in action, let alone new formations, with rifles - left no doubt that as a dominant factor in the war Russia was out of the reckoning. In the light of after events, the decision to continue the struggle after the loss of the San line in June is seen to be the first step to the Russian Revolution. Yet, on purely military grounds, it was justifiable on the assumption that the French effort to break through the Champagne front would succeed. Only this confidence in victory in September, indeed, can explain the stagnation on the Western front from April to August (broken only by the May battle in Artois), enabling Falkenhayn to withdraw some 12 divisions for the Eastern operations.

By August it was evident that the chances of cutting off any considerable formed army of the Russians in the Kielce region was at an end, and again there came up on the German side the controversy between Falkenhayn and Ludendorff as to what the operations were intended to achieve. Falkenhayn held firmly to the view that the Russian army must be beaten before any wide enveloping movement was undertaken to surround its debris. Writing after the war, he maintained the same opinion, only reproaching himself with not having compelled G.H.Q. East to give Gallwitz 20 divisions instead of 14. And certainly, if prisoners and booty were considered, he had in fact inflicted what by all military standards was a " sufficient " or " decisive " blow - for by the middle of Aug. the Russian losses in prisoners alone had reached the figure of 750,000 since May r, nearly 50% of their combatant strength as it had been at the end of April. But the time-limit was close at hand, and the withdrawals of forces to France and Serbia, delayed as long as possible, had now to be begun. The weeks remaining must, according to Falkenhayn, be devoted to inflicting as much additional loss on the Russians as was possible by frontal pressure coupled with flank attacks on the middle Niemen and east of the Bug, i.e. in the immediate vicinity of the frontal fighting, and possibly raids by light forces on the communications behind Kovno and Brest-Litovsk. At a suitable date the operation would be closed down, and the best line of defence taken up as a winter front.

Ludendorff, on the contrary, considered that the actual annihilation of the Russian armies was the only " sufficiently decisive result " that would give freedom of action in the West, and with renewed insistence - which went as far as a personal appeal by the Field-Marshal to the Kaiser - demanded the reinforcement of his left (Niemen army) with a view to quick swooping down on Vilna and Molodechno and the closing of the " corridor. " The axis Orany-Lida, originally suggested, was now too near for the required effect, but the principle was the same, and the movement would originate from a more favourable situation of the Niemen army than that existing in June. Preparations for the attack on Kovno by the X. Army were already well advanced, and Ludendorff considered that even at this stage complete success would be possible.

At this period the fighting on the Vindava-Schavli-Dubissa line had definitely turned in favour of the German Niemen army, the Russian V. Army receiving little or no further reinforcements when Mackensen's and Gallwitz's attacks developed. Below was progressing beyond the line named in each of the three directions Mitau-Riga, Poneviesh-Dvinsk, Keidamy-Wilkomir, and about Aug. r his various columns, totalling about 71 inf. and 51 cavalry divisions, were approximately on the line River Aa-R. Musha-E. of Poneviesh-Keidamy. To the southwest, the German X. and Russian X. Armies were still making war in the same fashion as in March, the Germans based on the Suwalki-Schali lines, and the Russians on their Kovno-Niemen-Grodno fortifications, making periodical thrusts in the region between. But the last important Russian thrust was delivered early in May, as a " relief offensive" toward Schali; and the German reaction became a methodical advance toward Kovno and Olita, which at the time here considered brought their left almost up to their opponent's stronghold. Behind the German advanced line preparations had been made for the siege of Kovno, an essential part of the scheme which Ludendorff still advocated.

It was evident that the scheme of bringing Below and Eichhorn down upon Vilna and Molodechno, and capturing Kovno in time, would call for the reinforcement of either or both, and, on this ground principally, Falkenhayn preferred to continue the campaign on the same lines as before, though a little later he conceded to Hindenburg freedom to dispose as he chose of the forces in his own area and to Mackensen freedom to pass to the E. of the Bug. Conrad, meantime, was planning an operation in East Galicia with the II., South and VII. Armies.

Thus the last phase of the tremendous campaign consists of 4 parts: (a) the frontal drive of (right to left) the Bug Army, the XI., Woyrsch, IX., XII. and VIII., (b) the attack on the north flank and the rear of the " Corridor " by the German X. and Niemen Armies, (c) the N.E. swerve of the Bug Army and the A.-H. I. Army, and (d) the autumn campaign in E. Galicia. All these were carried out without any great regrouping or reinforcement, and indeed, as regards (a) the forces concerned, were gradually reduced in order to form the army for the Serbian front and to increase the reserve in France. In the case of the operations in E. Galicia, the Russians followed a clear purpose and the parts of their efforts were coordinated. But elsewhere, under the tremendous pressure of the row of hostile armies stretching from Lomza. to Wlodawa and Vladimir Volhynskiy, the only general policy was that of gaining time at the expense of ground and of avoiding envelopment at all costs, and the day-to-day situations were met as best they could be. On the German and Austrian side the offensive energy of the troops was beginning to approach its limit, except as regards troops N. of Grodno, so that it may be said that the allied left and the Russian left alone retained the capacity for fresh achievement, while the rest were wearing each other out at an increasing rate.

The central campaign, between the Bobr and the Bug, may best be summarized by recording the battlefields of each of the German armies in succession.

Protected on its right by the Austrian I. Army, the Bug Army fought and won the battles of the Ucherka (Aug. 7-12) and of Wlodawa (Aug. 13-17), and in concert with the XI. Army continued its advance northward along the Bug against Brest-Litovsk. Meantime, the crossing of the Bug was authorized in so far as concerned the establishment of bridgeheads; and in carrying out orders with this object the German subordinate leaders became involved in fighting E. of Wlodawa, which inevitably formed the starting-point of an offensive against the eastern communications of Brest-Litovsk. By Aug. 21, then, the greater part of the Bug Army was engaged on the line of the Kapajowska from its mouth to Switiaz lake inclusive, well inside the region of the great marshes; the remainder (Beskidenkorps only), still west of the Bug, was nearing the outworks of Brest.

To the left of the Bug Army, the XI., already being reduced for the forthcoming Serbian campaign (for the conduct of which its staff was presently withdrawn), moved forward correspondingly against the W. of Brest. On Aug. 19 its left had reached Janow on the Bug below the fortress, while the Beskidenkorps stood at Koden on the same river above it. To the left of the XI. Army, again, the Austrian IV. Army at that date lined the Bug between Janow and Niemirow; and beyond Joseph Ferdinand, already N. of the river, was Prince Leopold with Woyrsch's and his own armies, which, as soon as they had debouched from Ivangorod and Warsaw, had made rapid progress, as the Russian centre retreated at the fastest possible pace to escape while Gallwitz and Mackensen were still being held off. The German IX. and Woyrsch Armies stood, on Aug. 19, N. of Niemirow, facing the line of the Pulwa and the Nurzec on which the Russians were preparing to make a stand.

Meantime Gallwitz, in his bridgehead position in the angle of the Bug and Narew, had overcome the Russian counterattacks, but not before their purpose of keeping open the railways and roads for the retreat of the Warsaw and Ivangorod forces had been achieved. The battles of Ostrow (Aug. 8-10) and Tschishew-Sambrow (Aug. 11-12) and the advance in the direction of Bielsk which ensued were thus similar in character to the operations of the IX. and Woyrsch's Armies, viz.: a direct pursuit where an envelopment had been hoped for. At the date of Aug. 18-19, Gallwitz stood between the Nurzec and the upper Narew, facing Biala, where the Russians were prepared.

The rightmost troops of the XII. Army, viz. those which in the battle of the Narew were facing south against counter-attacks from Novogeorgievsk and the strong points of the lower Bug, had now been combined with the leftmost troops of the IX. Army for the siege of Novogeorgievsk, in an army group under von Beseler, the captor of Antwerp; and the siege, pressed with energy, was nearing its close. On the 10th the place, with a large garrison, surrendered. On Gallwitz's other flank, the right of the VIII. Army had conformed to his advance and was taking the direction of Byelostok; its centre had mastered Lomza and Wiszna on Aug. to; and its left was again, as in Feb., battering Osowiec, which fell to the superheavy artillery on the 2 2nd. Kovno, as will be seen, had already fallen on the 18th, to the attack of the German X. Army.

Throughout these pursuit operations large numbers of prisoners continued to be taken by the Germans, and the Russian fortress artillery swelled enormously the total of captured guns. At Novogeorgievsk some 85,000 men and 700 guns were taken. Shortly it was to be the turn of Brest-Litovsk and Grodno, though these places were not defended after the withdrawal of the battle-lines outside them.

The later stages of the frontal pursuit may be very briefly dealt with. The general direction of the Woyrsch, IX. and XII. Armies was eastward. From Aug. 19-24 Woyrsch and the IX. Army were engaged in mastering the Pulwa-Nurzec line, on which the Russians delayed their opponents long enough to cover the evacuation of Brest-Litovsk against interference from the N.W. or N. From the 25th to the 31st these two armies were involved in a fresh series of combats in and about the " primeval forest " of Byleovitsa. Meantime the XI. and (till its withdrawal) the .Austrian IV. Armies, with the Beskidenkorps of the Bug Army, had attacked Brest-Litovsk concentrically from the W. and S., and the last Russian rearguards had been driven out of the evacuated stronghold on the 26th. The Germans and Austrians then continued the pursuit eastward, where the operations of the Bug Army and the Austrian I. Army (presently to be described) came into line with theirs in the early part of Sept. The XII. Army drove the Russians from the Bielsk posi tions on the 26th, from the Swislocz river a few days later, and from the Naumka-Wereczya line on Sept. 4, at which date the IX. Army and Woyrsch had at last debouched from the Byelovitsa forest towards the Jasiolda river.

In general, the effort of the Bug, XI., IV., Woyrsch and IX. Armies in the earlier stages of pursuit had tended to crowd the Russians into the area round Brest-Litovsk, and at a certain stage in this process the Bug Army had been authorized to push through the marshes E. of the river so as to reach the line of communications Brest-Litovsk-Kobrin-Pinsk. At the same time the Austrian I. Army about Vladimir Volhynskiy advanced to Kovel, and thence eastward (see Autumn Campaign in East Galicia p. 907) while from Kovel its cavalry worked up through the marshes northward to join the swinging right wing of the Bug Army. But that army, although it drove the retiring and diminishing forces of its opponent N.E. from the Kapajowska to Kobrin, was unable to reach that point before the Russians evacuating Brest-Litovsk had flowed past it. The Russian rearguard stood to fight on a line N.W.-S.E. through Kobrin, but, the Austro-German Cavalry Corps of General von Heydebreck from Kovel arriving on their flank, they soon fell back to the oblique line of the Dnieper-Bug canal, where they were temporarily secure against all but frontal pressure. Thus in this quarter too the pursuit became a direct one. The Russians were driven by the Bug Army and by what remained of the Austrian XI. and Austro-Hungarian IV. Armies - the whole now commanded by Linsingen - out of the canal lines in the battle of Horodec (Aug. 31-Sept. t) and out of the defences of DrohiczynChomsk (Sept. 4-6). But Linsingen's offensive, more and more hampered by poor communications, came to an end with the occupation of Pinsk on Sept. 16, and positions were taken up here which remained unchanged till the end of the war.

With the almost simultaneous capture of Brest-Litovsk, Bielsk, and Grodno (the last named fell to the German VIII. Army on Sept. 2-3), the Germans obtained possession of that line across the northern corridor which had usually been regarded as the Russian stabilization line. Falkenhayn, however, took full advantage of the shortening of front which resulted from the directions taken by his armies. and then at last Ludendorff's scheme came into play. Such an operation as Ludendorff contemplated, or at least one from the middle Niemen, Falkenhayn had been willing to agree to from the first; and as the occasion approached he relaxed his hold on Hindenburg's dispositions, stipulating only for the observance of his general directions and for the release of certain divisions for the West. In practice he approved the attack on Kovno. Ludendorff promptly took advantage of this, and the intended wheel-in upon the rear of the " corridor " was already in progress before the fall of Grodno and BrestLitovsk. On Aug. S the X. Army was able to begin the siege of Kovno. Ten days later the fortress was in its hands - even earlier than at Novogeorgievsk, Osowiec, and Brest-Litovsk. On condition of strengthening either the Niemen army or the left of the X., therefore, Ludendorff's plan had become feasible, if feasible at all, while masses of the enem y were still south of Brest-Litovsk, on the Pulwa and the Nurzec, about Bielsk and Byelostok and Grodno. At that date, Aug. 18, the Niemen army had pushed its left columns close up to the Riga-Uxkiill bridgehead on the Dvina, and to Friedrichstadt on that river, whence its centre and right ran southward along the Jara and Sventa to the north side of Kovno. It was still very strong in cavalry, but some of its transport had been taken for the armies pursuing through the devastated areas to the South.

Nevertheless, no serious advance was made to the westward from Kovno for more than a week, and even then part of the X. Army swerved full to the south against Olita to open an advance in the direction of Orany, and also to help the VIII. Army in cutting off Grodno, now a pronounced salient. At this late stage Ludendorff himself had doubts of the efficacy of the westward movement, and for a moment contemplated taking the direction favoured by Falkenhayn, viz.: Orany, Lida, Baranovichi. Not only was this the shortest route to the enemy's heart - the shortest, that is, as measured by the time necessary for concentrations and for rebuilding routes - but it offered hopes of driving a large mass of the enemy into the marsh region round Slonim, where the avenue of escape was narrowest (whereas at the latitude of Vilna - Molodechno the corridor broadens out considerably). However, he chose, in the end, to follow the current scheme of operations, as offering " annihilation " of the enemy as a prize, though admittedly that prize might escape him. On the 28th, therefore, with the expressed or implied consent of Falkenhayn, the X. Army was ordered forward on Vilna, with centre of gravity on the north wing, north of the Vilya. Reinforcements were collected from the troops lately besieging Novogeorgievsk and from the VIII. Army, which, after the fall of Grodno, would evidently be crowded out of the line. The Niemen army was directed to press up to the Dvina bridgeheads and, especially, up to Dvinsk, to cooperate with its left wing in the operations of the X. Army north of the Vilya, and to prepare a mass of cavalry to break through the thin line of the Russians near Swentsiany and seize or destroy the railways at Molodechno and Minsk.

The last great battle of the campaign, known as that of Vilna-Molodechno, began after the Grodno episode had been closed on Sept. 9. At that date Linsingen was advancing on Pinsk, Woyrsch and Prince Leopold driving the enemy slowly from one river-line to the next, over the Ja.siolda, in the direction of Slonim; and Gallwitz and the remnant of the VIII. Army were pressing slowly forward up the Niemen in the same direction. The Niemen army was, by its activity between Riga and Dvinsk, forestalling and perhaps diverting the attack of new Russian forces which were coming up from the Baltic provinces. From Wilkomir, north of the Vilya, to Orany, the X. Army engaged the very heavy forces that the Russians had collected for the last effort to hold the flank of their corridor - the final act of command of the Grand Duke Nicholas before the Tsar took over the control from his able hands. The German offensive progressed slowly, like all offensives against the Russian flanks in this campaign, but after some days it was judged that the forces on the Dvina and amongst the Dvinsk lakes had obtained sufficient security for the left flank, and on Sept. ii the German cavalry divisions broke through the cordon west of Novo Swentsiany and made for Swentsiany and Molodechno. On Sept. 14 the horsemen reached and broke the Vilna - Molodechno line at Smorgon. At Wilejka and farther north at Glubokoye they cut the vital Lida - Plotsk line. A party even reached the Minsk - Orsha line at Smolewice.

This last crisis was also the most dramatic. The first wave of cavalry was followed by others till about seven divisions were collected about Wilejka, Smorgon and Molodechno. But, recovering from their first surprise, the Russians quickly sent troops from Vilna and from Minsk, as well as from the southeast of Dvinsk, to clear their intercepted lines of retreat. These had to be recovered at all costs, for, while the forces retiring before Gallwitz, Leopold and Woyrsch still had the lines focussed on Baranovichi at their disposal, these could not help the northern masses, and it was in the north, towards Vilna, that the centre of gravity lay.

Thus a race to build up forces about Smorgon, Molodechno and Wilejka set in. The Russians, having the better communications and consequently the larger forces, won it. They drove back the German cavalry, after a continuous skirmish of five days, to the west of Smorgon and the northwest of Wilejka. Two days later the first infantry divisions arrived on the German side from the left of the X. Army. The detour of these troops along the north bank of the bending Vilya had enabled the Russians, moving on the shorter line, to reopen their line of communications; and, with this, the battle of Vilna became, like the battles farther south, a slow frontal drive. Thereupon Falkenhayn ordered operations to be broken off and more divisions to be withdrawn for other theatres, and fixed in general the line to be taken up as a winter line. The concluding operations of the campaign, mostly completed in early October, consisted in the methodical advance of all armies to this line, which, so far as the Hindenburg, Leopold, and Linsingen groups were concerned, ran from Tuckum, on the gulf of Riga, past the south side of Riga and parallel to the Dvina to Novo Alexandrovsk, and thence southward by Lake Drisvyaty and Lake Naroch, Smorgon, Krewo and Baranovichi to Pinsk, south of which point Linsingen's right came into touch with the left of the Austrian operations in East Galicia.

In East Galicia the pursuit of the Russian VIII. and IX. Armies, after the GrodekLemberg break-through in June, had been left by Conrad and Falkenhayn to the Austrian II. Army, the German-Austrian South Army, and to Pflanzer-Baltin. Although the first impressions of the victors in that battle had been that the Russian armies remaining in East Galicia were incapable of more than retreat and rearguard fighting for a long time to come, in fact it cost the Austrians and Germans much fighting and manoeuvring to establish themselves on the line of the upper Bug and the Zlota Lipa; and Pflanzer-Baltin was at one time subjected to a heavy counter-attack by General Lechitsky's Army, for in this quarter the Russians had an ample supply of reinforcements in their Odessa army. Towards the end of July, however, the fighting in Galicia died down.

Towards the end of August, as a part of the same final offensive act which produced the battle of Vilna - Molodechno in the other flank, Conrad initiated a campaign which was intended to confirm the separation of the northern and southern groups of the enemy and to clear the latter out of Austro-Rumanian territory definitively. The thinness of the defensive cordon in the Pripet marshes, revealed by the lack of serious opposition to the movements of Puhallo's I. Army on and beyond Vladimir Volhynskiy, and the advance of Heydebreck's Cavalry Corps across the swamps and forests to Linsingen's Drohiczyn battlefield, led the Austrian command to make its effort on the north side of the Lemberg - Brody watershed. Profiting by the general shortness of the line between the Bug and Vistula, Conrad withdrew the Archduke Joseph Ferdinand and the IV. Army from the Brest-Litovsk field of operations in the last few days of August, concurrently with the withdrawal, mentioned p. 906, of the German XI. Army for Serbia. During the gradual withdrawal of the IV. Army, Puhallo began his offensive from the line Vladimir - Volhynskiy - Kovel in a south-easterly direction. The fighting which followed is described in the article RovNO, Battle Of. The incoming of the Austrian IV. Army on Puhallo's left, on the one side, and the arrival of reinforcements for Ivanov's Viii., Xi. and IX. Armies, on the other, led to the battle spreading along the whole front from the Pripet to the Pruth. In sum, the Austrians, after advancing from Kovel to the rivers Goryn and Putilowka N.W. of Rovno, and from the Zlota Lipa to the Galician Sereth, were checked and driven back by a counter-attack group formed by Brussilov's VIII. Army in the region of Rajalowka. The rest of the Russian front taking up the movement, the Austrians were driven back from the Sereth to the Strypa, and from the Horyn - Stubiel line to the upper Styr and Stokhod, while the centre held practically all its gains. From the fourth week of September the battle, after some further fluctuations on the left, became a stabilized trench-warfare conflict which dragged on till mid-November, when both sides settled down in their winter lines. These ran from the Pripet along the Styr and the Kormin and thence past Dubno to Zborow and so along the Strypa. From the Strypa mouth to the Sereth mouth, the Austrians retained positions north of the Dniester, and from that point Pflanzer-Baltin's front substantially followed the frontier to Rumanian territory E. of Czernowitz. Thenceforward up to the opening of the great Russian offensive in 1916 the only important operations which took place in East Galicia were the relief offensive known as the " New Year battle " (see Strypa-Czernowitz) initiated by the Russians in the hope, which was not realized, of calling off Austrian troops from Montenegro, and the Russian capture of the Dniester bridgehead of Uzcieszko on March 19 - a diverting attack in aid of the spring offensive of the north.

(C. F. A.) IV.. Russian Front, 1916-17 Operations in Russia and East Galicia, 1916 and 1917. - About the end of 1915 and the beginning of 1916 the rival belligerents in the World War were confronted with the necessity of making vital decisions. For the Central Powers and their allies the past months had been rich in results. In the Balkan Peninsula Bulgaria's entry into the alliance, and the conquest of Serbia and Macedonia, had opened the way to Constantinople and Asia Minor. The Allied army in the East had tried in vain at Salonika to bring about a change in the state of affairs. The Entente troops had been withdrawn from Gallipoli. Even the bloody battle in East Galicia and on the Bessarabian frontier at the New Year had had no effect upon the general situation. Against Italy, and in the French theatre of war, the armies of the Central Powers had successfully maintained their position.

The chief of the Austro-Hungarian General Staff, Gen. Conrad von Hbtzendorff, proposed to clear up the situation in the Balkans as far as possible. Rumania must be forced to give up her ambiguous attitude by an ultimatum, supported by the presence in South Hungary of a powerful force of troops. Montenegro and at least the north and centre of Albania must be occupied by the Central Powers. These measures having been taken, an offensive, prepared in the meantime, on Salonika would end the campaign in the Balkans. But the chief of the German General Staff, Gen. von Falkenhayn, had since late autumn, 1915, remained with his plans in the West in the French theatre of war. He pronounced against an offensive at Salonika on several grounds, and his view of the political, military and technical difficulties of such an undertaking could not be waived aside. The German Gen. von Seeckt also upheld Falkenhayn in this, on the strength of a conference with the Bulgarian Army Command. While the plan of an attack on Salonika was rejected in this manner, pressure on Rumania was now likewise deemed unnecessary, since the military successes of the Central Powers had meanwhile caused a more conciliatory attitude at Bucharest.

As regards the Austro-Hungarian Army Command's plans for dealing with Montenegro and Albania, Falkenhayn tried to postpone these indefinitely. But Conrad clung to his point and carried out his intentions, more or less against the will of his German colleagues, whereupon a most acute personal quarrel broke out between the two generals, lasting nearly a month.

This quarrel, in the course of which Gen. Conrad had the satisfaction of seeing his troops take the Lovchen (Lovicen) and subdue Montenegro, obviously laid no promising foundation for their common decisions in the future.

The idea of bringing about a decision in the war by a campaign against Kiev or Odessa in the spring of 1916 seems to have engaged political rather than military circles in Vienna and Berlin. In the latter the Russian operations in 1915 only strengthened the old conviction that the Russian armies - thanks to the illimitable area of operations and the skill of the Russian leaders in retreat - would always slip their heads out of the noose again, and that any further advance of the Central Powers towards the east could only result in an inconvenient extension of the front. The war, according to the view of both the General Staffs, could only be won against the western opponents. Conrad proposed a combined offensive against Italy. An annihilating blow delivered against this enemy would have been not only in accord with his personal feelings and those of his armies, but was worthy of consideration on many other important grounds. The tension on the Italian front was increased for the Austro-Hungarians by every new defensive battle; the Italian menace to Trieste became more intolerable week by week. On the other hand, Italy was easier to overthrow than France - or England, for that matter; and, as often before in history, the fate of the Rhine might be decided in the plain of the Po. Falkenhayn did not refute these arguments; but he was doubtful whether, in the first place, it would be possible to force Italy to break with the Entente, in view of her dependence on England, and, in the second, whether even if, contrary to expectations, Italy's overthrow should be brought about, the Western Powers would take the loss of this Ally so very much to heart. Falkenhayn was convinced that the decisive campaign could be fought only in the French theatre of war. Conrad held to the other solution, but declared himself willing to place a few particularly good fighting corps at the disposal of the German Higher Command for use in France. This offer was declined by Falkenhayn both on military grounds and as a matter of prestige. He proposed as an alternative that his allies should take over, in addition to the 400 km. of front which they were defending between the Bessarabian Pruth and the Pripet (Prypec) against the Russians, a further portion of the Lithuanian front stretching towards the north. In this way it would become possible to set free more German troops for the attack on Verdun. But Gen. Conrad could not bring himself to accept this purely passive role, and the result of this difference of opinion was that the two empires of central Europe divided their forces, the one proceeding to the attack in France, the other to the Venetian mountains.

For the execution of these attacks, forces that had been set free in the Balkans were brought up and others from the Russian theatre. The German eastern troops were, between Oct. 1915. and Feb. 1916, reduced from 56 to 45 or 47 inf. divs., not to mention the exchange of other fighting troops for less serviceable units. Heavy artillery and technical supplies were also withdrawn and sent to France, but these could be adequately replaced, thanks, to the mechanical power of German industry.

The Austro-Hungarian eastern front in March 1916 was so organized as to have 6 divs. less than at the close of the fruitless October campaign in 1915. To balance this, however, a series of regts. and batts. were brought up from other divs.., so that the Austro-Hungarian eastern armies gave up,, in all:,. 120 batts. for the attack on Italy. The drafts for the infantry in this fighting force were supplied mostly from home at regular intervals, the drafting reserve being overfilled owing to the slight losses entailed by the war of positions. Out of this superfluity of men the regts. formed 5th and 6th Batts. Thus there could be no question of numerical weakening on the Austro-Hungarian eastern front. Far more heavily weighed the fact that the best and most reliable troops had been picked for the Italian attack, including nearly all the German Austrians and a great proportion of the Magyars. The eastern armies were seriously weakened thereby on the moral side; and the militia-like character, which the Austro-Hungarian army had begun to take on in the Carpathian battles in the spring of 1915, now became particularly apparent in the east. Still more severely felt was the withdrawal of the whole of the heaviest artillery, and a considerable portion of the medium-heavy, to the Italian theatre, and the considerably smaller share of technical supplies which had been assigned to the eastern front when these were divided.

In the beginning of March there were about 40 Austro-Hungarian and 46 German divs. on the Russian front. Of these, 42 German and 2 Austro-Hungarian held the front (Pinsk) between Riga and the Pripet and were under the German Higher Command; the other half of the fighting forces, in the south portion of the front, was under the orders of the Austro-Hungarian Army Higher Command (Teschen). Each section had a breadth of 400 km. The Austro-Hungarian divs. were on an average 14 batts. strong, the Germans only nine. The inferior rifle-shooting of the Germans was abundantly compensated by their superior equipment in artillery and fighting material of all sorts. The entire rifle strength of the forces of the Central Powers on this front amounted at this time to rather more than a million. It would be safe to estimate the Russian front at double that strength. The Russian Higher Command, controlled since autumn 1915 nominally by the Tsar but actually by his chief-ofstaff, Gen. Alexeiev, could draw on its drafting reserve to the fullest extent. In the spring of 1916 the regiments, in spite of the gigantic losses suffered in the last campaign, had been replenished for some time. Immediately behind the army front were enormous masses of reserves, and all the recruiting depots were full. Half of the world's munition factories were straining to supply equipment for the Tsar's armies. A number of En tente officers were instructing Russians in the western methods of attack. Along with all this the greatest efforts were made to raise the soldiers' moral. By March 1916 about 130 inf. divs. and 40 ca y. divs. stood on the Russian front, the inf. divs. consisting of 16 batts. - almost double the number of the German. This did not include the draft formations standing in readiness immediately behind the fighting reserve. The rifle strength of the front might safely be estimated at 24 millions. Her allies might well hope that Russia, in spite of her defeat in 1915, would come up to expectations in the general offensive planned for the summer.

The German attack on Verdun in Feb. 1916 brought the Russians into action earlier than was expected. Like Italy, who was now making her fifth attack on the Isonzo, the empire of the Tsar was expected to lose no time in doing its utmost for the relief of France. Russia had already transferred her centre of gravity to the area N. of the Pripet before this demand reached the Higher Command. On this section, that is, opposite the German front, were 80 out of 130 Russian divisions. Since the beginning of March Hindenburg's general staff (Kovno) had located a concentration of Russian troops at Smorgon, Dvinsk (Diinaburg) and Jakobstadt. On the other hand, the attack of the II. Russian Army under Gen. Smirnov on both sides of the Narocz lake on March 18 took the Germans somewhat by surprise. After a preliminary bombardment, such as had not yet been seen on the eastern front, this army flung itself upon the German XXI. Army Corps commanded by Gen. von Hutier. It was assumed from orders of the supreme commander of the Russian west front (Gen. Ewerth), which were captured by the Germans, that the Russians meant more by this attack than a mere relief offensive. While Gen. Litvinov's I. and Gen. Plehve's V. Armies were holding the weak German forces occupying the trenches at Widsy. Dvinsk, and Jakobstadt, Smirnov was to force a way through in the direction of Vilna-Kovno and then to wheel northwards and so drive the German wing to the sea.

The " Narocz Offensive " led at first to considerable successes for the Russians. The attack delivered between the Narocz and Wiszniew lakes by Gen. Balujyev with 4 army corps pressed the weak German forces backwards some miles between March 18 and 21. Though the groups attacking farther to the N. were not so fortunate, the Russians might yet hope for success. Then, to the rescue of the Germans, came a sudden thaw. This, indeed, increased the difficulty of bringing up the reserves which they had scraped together so painfully, but incomparably worse was the plight of the attackers in this melting of snow and ice. Their second and culminating attack on March 26, according to the German reports, was choked literally " in mud and blood." Towards the end of March the Russian spring offensive of 1916 died away, without ever getting beyond the local success on Lake Narocz. Their losses were estimated by the Germans at 150,000 men, while the Germans sacrificed not more than 15,000.

On April 28 1916 the troops of the German X. Army under Gen. von Eichhorn snatched from the Russians the greater part of that tract of country which they had captured during the March battles in the confined area of Lake Narocz.

At the conference held on March 18 1916 the Allies had fixed July 1 for the opening of the great general offensive on all fronts. For this the Russian Supreme Command was now making ready. By the end of May all their preparations pointed to the probability of their decisive attack again being made N. of the Pripet marshes, and again on the German front. Of the 130 Russian divs., comprising over 24 million rifles, as to which the AustroHungarian and the German intelligence service had accurate reports, 74 to 77 - or less than two-thirds - were in the northern section. On the side of the Central Powers there were at the same time on the eastern front 831 inf. divs. and about 20 cay. divs., each ca y. div. counting almost as many rifles as one regt. of inf., and often fewer. Altogether these amounted to 600,000 fighting men for the Germans and the same number for the 0 Austro-Hungarians. The distribution of forces was the same as in the beginning of March.

In the middle of May the Austro-Hungarian offensive against Italy had started, meeting at the beginning with great success. Once more the Russians were faced with the necessity of relieving their hard-pressed allies, and at least preventing any further transference of Austro-Hungarian fighting forces to the Italian front. Now the preparations for the Russian attack were not yet complete. Also it was evident that active relief to the Italians could only ensue from an attack, not on the German, but on the Austro-Hungarian eastern front - that is, between Pinsk and the Bessarabian Pruth. The Russian Supreme Command were not easily induced to depart from their original plans or to attack prematurely before July 1. In the end, however, they had to yield to the pressure of the Allies. Gen. Brussilov. supreme commander of the Russian " south-west front," with the Quartermaster-General, Gen. Dietrich, as the real source of energy at his side, received the order to advance to the attack from Rovno down to Bessarabia. The very first assault, made with attack groups that had been got together at haphazard. brought Brussilov great and unexpected success on both wings in the battles of Luck and Ocna, although the defenders were not unfavourably situated as regards numbers. Thereupon the Russian Supreme Command decided to refrain from the great attack on the German front altogether and transfer the centre of gravity of their operations to the southern section. The advances on the Russian side during the next three months, at Riga. Jakobstadt, Dvinsk, Smorgon and Lake Narocz, were therefore undertaken only at odd moments, without any successes worth mentioning, and must be treated simply as demonstrations. But meanwhile Brussilov had snatched from the Central Powers large portions of Volhynia and East Galicia and the Bukovina.

On the other hand, the objective of the Russian Supreme Command in the three battles at Baranovichi had a close connection with the operations at Luck. In the first battle, on June 13 and 14 1916, the attacks led by Gens. Ragosa and Lesch failed completely. Gen. Woyrsch maintained the upper hand over the Russian grenadiers with his Silesian Landwehr. The Germans lost 150 men, the Russians 7,000. In the second battle, on July 2-14, the Russians put in 16 of their divs. against the 21 German and 2 Austro-Hungarian divs. holding the section Gorodishche (Horodyszcze)-Baranovichi. The Russian main blow fell on the Austro-Hungarian XII. Corps under Gen. von Henriquez, and forced it back to the second position. German battalions were hastily scraped together to reinforce their hard-pressed allies. There were critical hours and critical days. But on the last two days of the battle the greater part of the ground captured by the Russians was torn from them again. East of Baranovichi Gen. Ragosa's troops were fated to achieve only unimportant local successes. The defenders lost - in. dead, wounded and missing - iSo officers and 8,000 men; the attacking Russians many times this number.

For the third time the battle of Baranovichi blazed forth on July 25 1916, this time as an introduction to the great Russian general attack N.W. of Luck. Once more the Russians flung themselves against the Gorodishche section, but were driven back by the Germans after a fierce three days' fight.

The .Russian offensive in the beginning of June 1916 brought the attention of the Central Powers with a jerk to the eastern front, where all at once the situation had become extraordinarily tense, and the anxiety became all the greater with the reflection that the results in the other theatres of war had not come up to their expectations. The Verdun undertaking had cost the Germans heavy sacrifices without making them masters of the fortress, and it was but a small consolation to know that the French had bled even more than they. On the Somme an English-French attacking force of prodigious size and fighting strength was massing itself. In the Venetian mountains at Asiago, the Austro-Hungarian corps, though it was still attacking, had lost much of its momentum since May 25. A pause in the fighting at the end of that month had given the Italians an opportunity of flinging powerful masses of troops on to the hard-pressed mountain front, thus averting catastrophe for the time being.

The turn of events in the east called for new decisions, and a few days after Luck the two chiefs of the General Staffs of the Central Powers met in Berlin to form these decisions. The idea of leaving the eastern front - in Falkenhayn's words- " to look after itself " was, it may be assumed, only theoretically discussed. Neither did such resolutions come under consideration, either then or later, as those executed in 1914 by the Central Powers when they shook off the enemy by one mighty move backwards and thus again deprived him of the initiative. The scarcity of food alone, under which the peoples of the Central Powers were already beginning to suffer heavily, made it imperative to cling to every foot of fruitful soil in Volhynia or East Galicia at all costs. On the other hand, the situation was so grave on all other fronts that for the moment any assistance proposed for the eastern front must be of a modest order. The commanders of the armies fighting against Russia indeed attempted, even within their own areas, to keep their forces together for use as units rather than to use them to fill up gaps. Gen. von Linsingen, for example, made frequent efforts in the area of his own group of armies at Luck to concentrate strong forces for counter-attack. But the strength of these attackgroups, in most cases, very soon exhausted itself against the numerical superiority of the enemy. Similar attempts were made several times in East Galicia and also in the Carpathians. Mention should be made in this connexion of a plan formed in the beginning of July 1916 to form a XII. Army out of the German and Austro-Hungarian troops in East Galicia and to attack with it on both sides of the Dniester. This idea certainly promised success; but the divs. selected for the purpose were, in view of the new increase of the Russian attacks, in most cases diverted to some particular danger-spot on the wide-spreading defence front; and the construction of the XII. Army, together with the task to be entrusted to it, had to be given up. There was nothing for it but to persist in the method practised since the middle of June, and contest every inch of ground in dogged local defence-battles. And even this mode of warfare was conditional on a considerable expenditure of force. Between the beginning of June and the end of Aug. about 17 German divs. had to be brought over from France and 8 to 10 Austro-Hungarian divs. from Italy. In addition, the front to the N. of the Pripet transferred a large portion of its regts. and divs. to the southern section, receiving in exchange only worn-out troops.

Since the beginning of July 1916 the Russians had also withdrawn strong forces from Kuropatkin's and Ewerth's fronts to add them to Brussilov's. Finally, at the beginning of Sept. the area S. of the Pripet, with 71 divs., had 20 divs. more than the northern section. The attacks during the summer offensive of 1916 cost the Russians enormous bloodshed. Great as were the results, the sacrifices far outweighed them. The Russian Supreme Command remained true to the methods practised in the Carpathians. It is quite impossible to point to any great conception underlying the operations of the Russian Command in these battles. They worked on purely local considerations and prospects, and often did not even make use of these, as for instance immediately after the first great blows delivered at Luck, when they gave their opponents time to close a gap of 50 km. which had been made. More than once did the Russian Supreme Command let slip an opportunity of a mortal blow.

The great crisis on the eastern front, lasting several months, reacted strongly on the relations between the armies of the Central Powers. The Austro-Hungarian troops had, from the very first Russian attacks, shown considerably less power of resistance than the German. The Austro-Hungarian armies fighting at Luck and Ocna had, within a few days, left a quarter of a million prisoners in the enemy's hands. Even in peace-time the conditions in the polyglot Dual Monarchy were less favourable by far than those in the German Empire for a display of military power, and the unexpectedly long duration of the war increased the difficulties enormously. It should also be remembered that in the first year of the war the Austro-Hungarian military forces had had a considerably larger drain on their men than the German. At the end of 1 9 15 only a small remnant of the forces deployed at the beginning of the war was left at the front. The rest were dead, wounded or prisoners. In the quiet period before the Russian summer offensive of 1916 the training of the drafting reserve was certainly better organized than in the first year, when recruits had on occasion to be sent to the front after a month's training. But between the young, systematically trained peace-time forces, full of heroic self-sacrifice, with which the war started, and the Landsturm troops of the later campaigning years, some of them physically and morally unsound to begin with and many of them far too old, there could be no comparison. This was particularly the case with a considerable proportion of the Slav and Rumanian forces, on whom the great national crisis could not act as a spur but rather as the reverse, as was not infrequently proved. Under these difficult conditions the lack of good regular officers was most keenly felt. The flower of these had been left on the battle-fields of 1914.

In consequence of the internal weakening of the AustroHungarian army in the east - which was not noticeable in anything approaching the same degree where the army was opposed to its " hereditary enemy," Italy - a rule was made that on every point of the battle-front where the Russians were using great pressure German units should be flung in. In this way, from the beginning of July, the whole Austro-Hungarian section was interspersed with German troops. This system of " stay-boning," as it was sarcastically called, naturally brought with it a powerful increase of German influence in the combined army. It also happened that the Austrian leading provoked frequent criticism on the part of the German commanders. Immediately after the first Russian assault at Luck, for instance, the commander of the IV. Army, Archduke Joseph Ferdinand, was relieved of his command on the explicit demand of the German General Staff. Added to this, between the new commander and his Austrian subordinate commanders intermediate posts were interposed and filled by German generals, who alone exercised direct power of command over the troops. As the number of German forces on the Austro-Hungarian front increased, the ambition of the Germans to get the principal commands into their own hands became more and more evident. Immediately after the beginning of the Russian offensive, the area commanded by the German Gen. Linsingen, which began on the Pripet, was extended to the boundary of Galicia. At the same time Falkenhayn proposed to entrust Field-Marshal Mackensen, who was in Bulgaria, with the supreme command of all the allied troops fighting S. of the Pripet. Conrad von Hotzendorff was opposed to this arrangement, but offered to confer on Mackensen the command of a group of armies in East Galicia. This Falkenhayn declined.

In July Falkenhayn made the proposal to recall Field-Marshal von Hindenburg from Kovno and appoint him supreme commander between the Pripet and the Dniester. To this plan Conrad agreed, though without seeing any particular meaning in it. As a matter of fact Falkenhayn's proposal was made more on personal than on practical grounds. The chief of the German General Staff had from the start few friends but many enemies. Since the failure of the attack on Verdun, Emperor William had begun to be besieged with complaints against the man who had his particular confidence. The Imperial Chancellor also urged that Falkenhayn should be replaced by Hindenburg, with a vigour quite unusual with him. The summer battle made the situation more acute. A depression fell over Germany, the army lost faith in the Supreme Command, and louder and louder became the clamour for Hindenburg.

Falkenhayn, though realizing that his relations with Hindenburg and Ludendorff had been somewhat strained for more than a year past, felt obliged to fall in with the general opinion. He therefore proposed - assuredly more or less against his inward conviction - that Hindenburg should receive the appointment alluded to, that of supreme commander from the Pripet to the Dniester. This he followed up a few days later, under pressure from all sides, by offering to place the Field-Marshal in command of the whole eastern front from the Baltic Sea to the Carpathians. For some such unity of command there was urgent need on military grounds. Indeed the proposal had received a passing consideration in Nov. 1914, when the Archduke Frederick was to have held the command with Ludendorff as his chief-of-staff. But Conrad was opposed to this solution, arguing that the Russian assaults were not to be stopped by new commanders but by strong battalions, that the non-German peoples and troops of the Austro-Hungarian Monarchy would regard any too conspicuous manifestation of German influence as a burdensome tutelage, and finally that in the southern section of the eastern front so many specifically Austrian interests were at stake - in particular with respect to the danger from Rumania - that this section ought not to be withdrawn from the immediate influence of the Austro-Hungarian Higher Command. But, however worthy of consideration Conrad's objections may have been, it is only human to suppose that personal grievances also played their part. The solution proposed by Falkenhayn was tantaihount to making a clean sweep of the Austrian General Staff from the command of the war in the east. Yet, the leader of the Austro-Hungarian army was practically alone in his opinion. Even in those Viennese circles most jealously concerned to uphold Austrian prestige there was a pressing demand to have the question of the command straightened out on German lines, since Conrad's leadership no longer inspired full confidence. In principle the old Emperor, Francis Joseph, agreed also. Emperor William, for his part, proposed a compromise to his allies. Hindenburg was certainly to have supreme command from the Baltic Sea to Lemberg, by far the greater portion of the eastern front, but the southern portion on both sides of the Dniester and in the Carpathians was to be placed under the command of the Austrian heir-apparent, Archduke Charles Francis Joseph, who later became emperor. The Archduke had been fetched from the Italian front to take over the command of the XII. Army, and now, since this army had not come into being, was temporarily commanding a group of armies on both sides of the Dniester. To protect German interests the German Gen. von Seeckt, formerly Mackensen's chief-of-staff, would be attached to the Archduke. This solution was accepted by the Austrians and acted upon in the beginning of Aug. 1916. It was the prelude to the establishment of a Supreme War Command (Oberste Kriegsleitung) over all fronts.

The impulse to create a general supreme command of this nature, to apply to all fronts, emanated from Sofia and Constantinople. Germany was agreeable to the proposal, which also gained ground rapidly in Vienna's political circles, but here again it was Conrad who, in defiance of his superior commander, Archduke Frederick, sharply opposed the idea, even threatening to resign. Emperor Francis Joseph's personal intervention alone was able to overcome this opposition. On Sept. 16 1916 the agreement on the Oberste Kriegsleitung was signed in the German headquarters at Pless in Prussian Silesia. According to these arrangements, shortly after acknowledged as binding by the Bulgarian and the Imperial Ottoman Army Commands, the German Kaiser became responsible for the higher leading of operations in general without disturbing the relations between the allied sovereigns and their fighting forces. The Kaiser was assisted by the chief of the General Staff of the German armies in the field, who before every important decision was to consult the chiefs of the allied General Staffs and, as far as possible, bring them to an agreement. This done, the German chief of the General Staff would issue orders, binding on all, in the name of the Oberste Kriegsleitung. In order to meet Conrad's particular misgiving the German Kaiser bound himself, in a special supplementary note which was kept secret from Sofia and Constantinople, that the integrity of Austria-Hungary should be as carefully protected as that of Germany.

The institution of the Oberste Kriegsleitung was undoubtedly an advance on the method of dealing with each case as it arose, which had been practised since the beginning of the war. But the undertaking, set on foot with such difficulty, still lacked one thing to complete it - the creation of a common political and domestic policy. This was never achieved. On the contrary, the forces involved in these matters fell farther and farther apart the longer the war lasted, particularly after the change of Government in Austria-Hungary. To make matters worse, when the young Emperor Charles took over the Austro-Hungarian command, alterations were made in the most important parts of the agreement in consideration of his position as sovereign, so that in the end the old methods employed in the first two years of the war came back into use.

The agreement on the Supreme War Command had been signed on the part of the Germans by Field-Marshal von Hindenburg as new chief of the General Staff. On Aug. 29 1916 Falkenhayn had left the Supreme Command. It had long been only a question of when the Kaiser would be forced to yield to the storm raised by Falkenhayn's critics; the immediate cause of his dismissal was Rumania's declaration of war on Austria-Hungary on Aug. 27 1916. Up to the last hour, in spite of the well-founded warnings of Austria-Hungary, Falkenhayn had been unable to believe that Rumania was on the point of coming in, and had perpetually reassured the Kaiser to that effect. When the event happened the Kaiser was thunderstruck, and Falkenhayn's fall followed. The German nation and its allies greeted the new men, Hindenburg and Ludendorff, with the utmost confidence.

The first task to fall upon the new command was the organization of the Rumanian campaign (see next section). Meanwhile the defensive battle against Russia had to be carried on. This constantly flamed up again along the whole front from Luck to the Carpathians till the end of Oct., although the Russian attacks had fallen off in strength and determination, and no more successes worth mentioning were gained by them. The relief offensive, too, which the Russians undertook between Nov. 28 and Dec. 13.1916 in the Wooded Carpathians against Kovess' and Arz's armies, to relieve Rumania, hard pressed in Wallachia, was without results and could save neither Bucharest nor Focshani. Neither did success attend the Russian surprise attack on Jan. 23 1917 on the Aa at Riga, great as were the prospects of success on the first day. The German position was indeed rushed, but the defenders' reserves, brought up in haste, restored the situation.

The enormous drain on Russia's forces in the summer, and the difficulties of her interior political situation, had sapped the marrow of her army. The armies of the Central Powers and their allies had come through their difficult crisis. As in the end of 1915 and the beginning of 1916, so now they experienced a great relief in the east.

The Austro-Hungarian chief of the General Staff, Conrad, who since Nov. 11 1916 had been a field-marshal, was once more proposing to take the opportunity of attacking Italy. The attack was to begin in the spring of 1917 and was to be carried out by an equal division of Austro-Hungarian and German forces. But the new German Oberste Kriegsleitung in the middle of Jan. rejected the plan for the time being. They relied on the ruthless submarine war, begun in Feb. 1917, to bring their enemy to his knees. Gen. von Arz, who had succeeded Conrad as chief of the Austro-Hungarian General Staff in March 1917, received the decision of German Headquarters with unqualified approval. The Russian Revolution, which broke out in the middle of March, was extraordinarily favourable to the military situation of the Central Powers. It could not, of course, be seen as yet whether Russia's armies would permanently withdraw from the list of enemies, which now included America. But for the moment the crash was so enormous that it must be months before the Russian High Command could consider offensive operations. The Central Powers now left nothing untried that could hasten the process of disorganization among their enemies. This purpose was above all to be served by an extensive peace propaganda, which was to be carried to the Russian trenches - though Ludendorff's consent to this step had not been easy to obtain. Meanwhile the war fell practically fast asleep, as was natural. On April 4 1917 the troops under Prince Leopold of Bavaria, who had taken over the eastern front between Riga and the Carpathians from Hin denburg in the autumn of the previous year, captured the small Russian bridgehead, Stochod, at Tobol in Polyesie, by a coup de main, on which occasion the Russians gave themselves up as prisoners in swarms. On the political circles of the Imperial Powers this action produced a most disturbing impression, and the troops were now ordered to suspend all hostilities against the Russians unless they should provoke them.

In preparing the line to be taken with regard to propaganda in the trenches, the Austro-Hungarian Government would have liked to make use of the Petrograd catch-words, " Peace without annexation or indemnity " and " the right of nations to self-determination." But the dominating influence of the German Supreme War Command, which was not prepared to give up its Balkan aspirations in a hurry, prevented this. Nevertheless it was hoped, especially when the Workmen's and Soldiers' Councils became a stronger political factor in Petrograd in the beginning of May, that a way might be found to force Kerensky's Government to agree to an armistice and consent to open peace negotiations. This hope was to prove deceptive. When at last an envoy from Prince Leopold of Bavaria succeeded in penetrating beyond the Russian trenches to Gen. Dragomirov, commander of one of the armies on the north front, he was met with an absolutely unequivocal refusal. Meanwhile, at numerous points of the front, a local truce had been declared. But in June a remarkable change was noticeable on the Russian side. Kerensky, relying on Gen. Brussilov and numerous Entente military missions, succeeded gradually in converting a considerable portion of the army to the idea of carrying on with the war to make the world " safe for democracy," and in restoring their fighting spirit.

By the end of July the Army Commands of the Central Empires had reason enough to count upon a revival of the war in the east. Indeed, the Russian Supreme Command, apart from the concentration of troops at Riga, Dvinsk and Kriewo 1 in the Courland-Lithuania section, had assembled two powerful attack groups in East Galicia. The one, a division strong, stood N.E. of Brzezany opposite the Austro-Hungarian II. Army (BdhmErmolli); the other, 15 inf. and 2 ca y. divs. strong, was piling itself up adjacent to the first and opposite the German Southern Army under Gen. von Bothmer. 2 These powerful attack-masses were charged to overthrow the enemy and to take Lemberg.

The Army Higher Commands of the Central Powers did not look on idly at the Russian preparations for attack. By June 27 the German Emperor, in concert with the Austro-Hungarian General Staff and Prince Leopold of Bavaria, was ready with the order to counter-attack in case of a Russian attack in East Galicia, and to throw back the enemy beyond the frontier of the Austrian Empire. Besides the allied troops already on the spot, 5 divs. could be brought up from the west and 3 to 4 divs. from that section of the eastern front which was not threatened.

The anticipated Russian attack was launched on July 1 1917 on both sides of Brzezany. The Western Powers had supplied their Russian allies with artillery, munitions and war supplies of all sorts in abundance. The battle of Brzezany lasted six days with only slight interruption. The Russians made only slight gains in fighting the German Southern Army E. and S. of Brzezany, and these were for the most part wrested from them again. N.E. of Brzezany, in the village of Koniuchy, they were more successful. Here they had brought into the fight a Czechoslovak brigade against Austrian battalions of Slavonic speech, large sections of which surrendered, with the result that the Austro-Hungarian line was driven back 1 At Kriewo the Russians actually attacked with 10 inf. divs. on July 21 and 22, forcing back the German front 2 km. deep along 4 km. of front.

The designations " Austro-Hungarian II. Army," " German Southern Army," etc., refer only to command and staff. The troops were thoroughly mixed on the whole of the eastern front.

some 4 or 5 km. on a front of io km. On the evening of the second day of the battle, however, the Russian blow was countered by the German troops. Since, to the immediate N. of the battle-field and N.W. of Zborow, the first echelons of the German divs. rolling up for the counter-attack had arrived, and the Russian attack had so lost its force, Prince Leopold of Bavaria now supposed the danger to be averted.

Far more unpleasant was the effect upon the command of the heavy set-back to the Austro-Hungarian III. Army under Gen. von Tersztyanszky at Stanislau only a few days later. Here Gen. Kornilov, the ambitious commander of the Russian VIII. Army, had advanced to the attack at 7 A.M. with an attackgroup thrown together anyhow. On the following day he already held in his hand the key to the enemy's position - the Jutrena Gora height dominating Stanislau. Tersztyanszky hoped at first to have to withdraw the north wing only, but the Russians pushed the Austro-Hungarian regiments back so vigorously that by July i 1 the whole of the III. Army had to be withdrawn behind the Lomnica. The town of Stryj, and the East Galician petroleum district, Drohobycz-Boryslaw, on the possession of which the continuance of the submarine waf very largely depended, were in the utmost danger, and Prince Leopold of Bavaria was forced to let 3 inf. and 1 ca y. divs. of the units rolling up for the counter-blow be diverted to the III. Army.

When the Russians again attacked at Kalisz and made progress there, Prince Leopold and his chief of the General Staff, Col. Hoffmann, were confronted with the difficulty of deciding whether in the given case the counter-blow at Zborow, already being prepared, should be given up, and help sent in haste to the sore-pressed Gen. von Tersztyanszky. The Prince resolved to adhere to the original plan. He proved to be right. The attacks of Kornilov's troops lost their sting as rapidly as those delivcred at Brzezany by the Russian VII. Army. Aided by German reenforcements, Gen. Kritek, who relieved Tersztyanszky in the command of the III. Army, was able by the 16th to prove his troops' newly established powers of resistance in counterattacks at Kalisz.

Meanwhile, between the upper Sereth and the railway line between Lemberg and Tarnopol, immediately W. of Zborow, 8 inf. divs. (including the I. and II. Guard Divs.) and one combined cavalry div. were deployed for the counter-blow along 25 km. of front behind the divisions of position. The German Gen. von Eben was in command on the battle-field. The intention was on the first day to make a hole in an easterly direction in the south wing of the Russian VIII. Army which stood opposite, and then to wheel to the S.E. and grip the massed Russians of the VIT. Army standing on either side of Brzezany, in the N. flank and in rear.

This idea underlying the battle of Zborow (July 19-26 1917) was carried out according to plan. Early on the 19th the German and Austro-Hungarian forces drove the Russians from the Zlota Gora height, N. of Zborow, under the eyes of Prince Leopold of Bavaria. Simultaneously the German Guard, reinforced by a line division, broke through the Russian front immediately S. of the Sereth. Only in places did the Russians offer resistance. Their retreat frequently degenerated into precipitate flight. While the Guard Div. in the following days drove down on Tarnopol, the divs. brought forward from the 2nd line pressed after in a S.E. direction. The Russian masses at Brzezany were soon swept into the general retreat. By the 22nd the German Souther' Army was able to take up the pursuit, also from the N. wing. On the 23rd the III. Army followed S. of the Dniester, and was able on the next day, after several fights, to push out to beyond its old positions at Stanislau. On the 25th the German Guard took Tarnopol in presence of the German Emperor, and on the 26th the heights to the E. of it, thus assuring an adequate protection to the S.E. blow by the other allied forces. The S. wing of the II. Army was already beyond Trembowla, and the Southern Army beyond Buczacz. They had rapidly broken the Russian resistance.

A few days after the defeat of Zborow the Russian command passed out of Brussilov's hands into those of Kornilov. The Russian General Staff reports of those days give a tragic picture of the condition of the VII. Army and the S. wing of the VIII. Army at that time. It was clearly out of the question to maintain a hold on East Galician soil in this area. The decision was therefore made at Mohilev to withdraw the VII. Army and those parts of the VIII. Army pursued by the enemy behind the river Zbrucz on the frontier.

More cheering were the reports received by the Russian Supreme Command from the VIII. Army now commanded by Cheremissov, which was retiring S. of the Dniester. This army could, with some hope of success, be charged to hold Czernowitz and as much as possible of the Bukovina. Further relief was expected from the results of the Russian-Rumanian attack in the valley of the upper Susita in Rumania, which had commenced on July 23 and was accompanied by demonstrations along the whole Transylvanian east front. This attack had really succeeded by July 27 in forcing back the numerically very weak defenders to a not inconsiderable distance. But the fate of the Bukovina was nevertheless sealed. Between the Dniester and the Carpathians Cheremissov's troops several times put up a good resistance; in fact, the Austro-Hungarian III. and VII. Armies (the latter under Gen. von KOvess) had even to deliver counter-blows. But on the morning of Aug. 3 1917 the Russians, threatened on the N. and the S., had to surrender Czernowitz, and soon to retire from the Bukovina into the frontier area. The AustroHungarian III. Army pursued between the Pruth and the Dniester and the VII. S. of the Bukovina.

East Galicia had also been swept clean of the Russians, apart from the area N.E. of Tarnopol which had remained untouched by the offensive. Advanced detachments of the German Southern Army had set foot on Russian soil on the middle and lower Zbrucz. But here the soldierly characteristics of the Russian people, which had survived even the unnerving influence of the revolution, came to the fore again. The Russians not only cleared the E. bank of the Zbrucz, but roused themselves in an amazing manner to renewed resistance E. of Czernowitz and in the southern part of the Bukovina. It was undoubtedly to their advantage that the enemy in his rapid advance had come dangerously far away from his railways. The Austro-Hungarian III. Army at Czernowitz for instance was 120 km. removed from its main detraining station lying W. of Stanislau. This was particularly serious in view of the meagre means of transport supplied to the Austro-Hungarian troops. Had the armies been crossing a less fertile area the pace of the offensive must soon have slowed down very considerably. As it was the troops could subsist largely on the resources of the country and the rich booty left behind by the Russians. But now, on the frontiers of East Galicia and the Bukovina, the advance of the Imperial forces was arrested.

For some weeks past the Allied Higher Commands had been considering the idea of combining with the East Galicia offensive an attack on the Rumanians in Moldavia, which should drive them behind the Pruth, thus gaining a particularly useful defence section in which troops could be economized. Accordingly, on Aug. 6 1917 Mackensen advanced to the offensive against the Rumanians N. of Focshani. The battle of Marasesti ended unfortunately for the forces of the Central Powers. In view of this, and of the difficulty of obtaining fresh drafts in the Bukovina, the Central Powers abandoned the idea of occupying Moldavia for the present, and dropped it completely when, at the end of Aug. and the beginning of Sept. the Isonzo battle led to the combined Oct. offensive against Italy. Ludendorff lays stress, in his memoirs, on his own reluctance to give up the Rumanian campaign.

In the Bukovina and on the Transylvanian-Rumanian front minor operations lasted until the middle of September. Local attacks and counter-assaults were distributed on both sides. Then gradually the fighting died down.

German Capture of Riga. - Meanwhile, in the extreme N., the Germans had won a fresh victory over the Russians. During the second half of Aug. they had been quietly preparing to cap ture Riga. By order of Gen. von Hutier, supreme commander of the German VIII. Army, 6 divs. were placed in readiness for crossing the Dvina opposite i?xkiill, to the S.E. of Riga. Other forces were to follow. Altogether there were 14 divs. available for the undertaking, including the Guard and other units brought from East Galicia.

The crossing at Ixkiill was carried out most punctually on Sept. 1 1917. By now 3 bridges had been built. The Russian XII. Army (Parski), 20 inf. divs. strong, made only a slight resistance, and by the 2nd had evacuated all the positions S. of Riga. On the following day the 2nd Guard Div. and the 1st Res. Div. were able to enter the ancient Baltic trading-centre, the one from the east, the other from the west. The Russians now evacuated the whole N. bank of the Dvina up to beyond Friedrichstadt. On the 4th the German infantry reached Hinzenberg railway station, 4 o km. N.E. of Riga. The permanent position was now formed along a line drawn from Uxkiill to Hinzenberg and thence westwards to the sea. Only the German cavalry now pursued the enemy, who first came to a stand 20 to 40 km. E. and N. of the German line.

The occupation of Riga needed to be supplemented for the Germans by the capture of the Baltic islands, Osel, Moon and Dag, and this was duly achieved in the middle of October. For the first time in the war, on the side of the Central Powers, the navy was present in some strength to assist in the operations of the land army. The landing corps consisted of the German 42d Inf. Div. and the Cycle Bde., and was commanded by Gen. von Kathen. The spot selected for the landing was Tagga Bay on the N.W. corner of Osel Island. While Adml. Erhard Schmid's German squadron penetrated through the Domesnas straits, after silencing the coast batteries, the torpedo boats went round Osel in a northerly direction, in order to bring their guns to bear on the mole connecting Osel with Moon and to cut off the retreat of the Russian troops on Osel. From the N. they were to press on into the Moonsund. On Oct. 13 the German troops landed in Tagga Bay. The enemy, about one div. strong, tried to effect their escape, some southwards to the Sworbe Peninsula, others over the mole to the island of Moon. By evening on Oct. 16 the whole of Osel was in possession of Gen. Kathen. Ten thousand Russians were taken prisoners, among them one divisional and three brigade staffs. On the 18th Lt.-Gen. von Estorff, advancing over the mole, occupied the island of Moon, and on the 21st Dago had also been taken by the Germans. In the waters of Moon it came to fighting engagements between German and Russian ships, in the course of which the Russian battleship " Slava " was set on fire.

On Nov. 7 the Bolshevist Revolution broke out in Russia. On Nov. 9 the congress of the " Workmen's and Soldiers' Council," meeting at Petrograd, issued its proclamation of peace " to all." In vain did Kerensky and Kornilov attempt to give matters a different turn. An army corps sent by them against Petrograd on Nov. 12 was defeated at Tsarkoye Selo. On the 10th the Council of People's Commissaries gave instructions to the new Russian Supreme Commander, Dukhonin, to offer an armistice to all the belligerents. As Dukhonin hesitated to carry out the order he was replaced by Ensign Krylenko. On Nov. 28 the troops of the Imperial forces on the eastern front intercepted a wireless message in which Lenin and Trotsky invited the earliest possible preliminary arrangements for the armistice and peace negotiations. On Dec. 2 the armistice negotiations between the Imperial forces and Russia were begun. at Brest-Litovsk, at Prince Leopold of Bavaria's headquarters. The only questions which caused serious difficulty were that of the Baltic islands, which the Russians wished the Germans to evacuate at least in part, and that of the transference of German troops to the west. On the first point the Germans refused to give way; on the second they compromised. After a formal 10 days' truce had been agreed to on Dec. 5 and the armistice had set in on the Rumanian front on the loth, the cessation of hostilities for one month on all the Russian fronts against the Central Powers was declared on Dec. 15. On Dec. 22 the peace negotiations of Brest-Litovsk began. (E. G.-H.) V. Rumanian Campaign, 1916-7 Operations in Trans y lvania and Rumania, 1926-7. - After the unexpectedly great results of the Russian summer offensive under Brussilov in 1916 the conviction gained ground in Rumania that the moment had now come for her intervention on the side of the Entente Powers. In accordance with the policy pursued since the Balkan Wars of 1 9 12 -3 of harvesting from international quarrels the greatest practicable advantage with the least possible sacrifice, Rumania now hoped to be able to realize the desire, cherished by the entire nation, for the enlargement of the kingdom by the incorporation of the districts of Austria-Hungary inhabited by her nationals.

The Austro-Hungarian Supreme Command had foreseen most clearly the intervention of Rumania, while at German Headquarters the danger did not appear so imminent; and the Hungarian Government - fearing unrest among their own population and in the hope of not destroying the last chances of maintaining peace with Rumania - avoided taking in the threatened frontier districts the measures necessary in the event of war breaking out. Thus it happened that Transylvania was quite inadequately defended from the military point of view against the Rumanian attack, and the country was hardly prepared at all as a theatre of war. On account of the pressing need for all effective units on the Russian and Italian fronts, the Austro-Hungarian Supreme Command could transfer to Transylvania in the beginning of Aug. only the seriously reduced 61st Div., the 51st Honved Div., and the 82nd Inf. Regt., troops for which an urgently needed rest had been intended. The 11th Honved Ca y. Div., disentrained in Transylvania, was at once constituted the southern wing of the VII. Army. The removal of the war-worn 39th Honved Div. began in the last days of Aug. 1916. In addition there were in Transylvania, at the end of Aug., 8 newly formed inf. batts., 2 Landsturm batts., 10 communication batts., 3 mining batts. (armed coal-miners from Petrozseny), 9 " alarm " batts. (march batts. not yet fully trained), about 5,000 frontier police, 3 Landsturm squadrons and 9 field batteries, which were formed into newly created larger units whose formation was, with many changes, only completed in Oct. 1916. All the troops above mentioned, in the areas between the Hungarian frontier of the Bukovina and the Danube at Orsova, were from Aug. 13 onwards under the newly created Austro-Hungarian I. Army headquarters, under Gen. Arz von Straussenburg, at Klausenburg (Kolozsvar). Preparations were also made for the thorough destruction in the passes, of the roads and railways leading to Rumania.

At the end of July a convention was concluded at Pless between the German, Austro-Hungarian and Bulgarian Army Commands for common action in case of hostile Rumanian intervention, and to this Turkey shortly afterwards became a party. The plan of campaign was laid down in its main outlines and the contingents to be provided by the separate States agreed upon. The Germans allotted 5 to 6 inf. and 1 to 2 ca y. divs., which, however, the German Supreme Command in expectation of coming events did not yet wish to place in reserve in Transylvania, as they would certainly be missed at critical points on the other battle-fronts. They therefore confined themselves for the time being to sending German construction troops to strengthen the existing railway lines in Transylvania, S.E. Hungary and northern Bulgaria for the advance of larger bodies of troops. The German detachment under Kauffmann, already placed in northern Bulgaria with a view to the creation of a GermanBulgar-Turkish army on the Dobruja frontier, was reinforced, and had heavy artillery, mine-throwers and flying formations.

Bulgaria placed on the Dobruja frontier the III. Army, commanded by Gen. Toshev, consisting of the 1st, 4th and one-third of the 6th Inf. Divs.,' one ca y. div., the garrison of Varna, and a part of the Kauffmann detachment, while the Danube was guarded from Tekija (opposite Orsova) to the mouth of the Vid by the Bulgarian 12th Inf. Div., and from there eastwards as far as Ruschuk by the German Kauffmann detachment. At 1 A Bulgarian inf. div. had 3 bdes. of 8 batts., i.e. 24 batts. in all.

Sistova there was a heavy Austro-Hungarian bridging train with its complement of men, some heavy batteries, and the Austro-Hungarian Danube flotilla. The supreme command over all the troops in Bulgaria for operations against Rumania was exercised by Mackensen, with headquarters at Tirnovo.

It was considered most probable that Rumania, when she struck, would begin by invading Transylvania, in order to gain possession of the country and to use it as a base from which completely to shatter the Carpathian front, which the AustroHungarians, as it was, were only maintaining with difficulty, while comparatively weak forces only would be employed against Bulgaria. On this assumption it was proposed to surprise and overrun the Rumanian positions in the Dobruja with the GermanBulgar-Turkish forces, in order to penetrate to the narrowest part of the district between the Danube and the Black Sea. The strongest possible forces would then be collected, and held ready at Sistova, where the heavy Danube bridging train was already placed, for a forward push in the direction of Bucharest. In Transylvania the Austro-Hungarian forces were to hold up the Rumanians in the mountains on the frontier if possible, but at latest on the position prepared on the upper Maros and the Little Kukiillo (Kleine Kokel), until the attacking troops being concentrated meanwhile could be marched up. These, and Mackensen's troops to be held ready at Sistova, would then proceed to the reconquest of Transylvania and the overthrow of Rumania. In this case the unusual happened, and the actual operations in their main lines were successfully executed as had been proposed in the discussion of war plans at Pless.

Rumania had pushed forward her mobilization, and by continuously reenforcing the troops on the Transylvanian border had so nearly completed their concentration that operations could begin immediately on the declaration of war, which was handed in at 9 P.M. on Aug. 27. It was intended first to conquer Transylvania. For this purpose strong forces were to push forward from the E. over the mountains on the frontier, and advance westwards through the valleys of the Kukiillo, the N. Kukiillo and the Maros. The calculations included a simultaneous push forward of the Russian front adjoining on the N., whose advance would he greatly facilitated by the offensive of the Rumanian army S. of the chain of the Carpathians stretching from Hungary into the Bukovina. The forces which penetrated the passes on the Transylvanian southern front were then to hold these by means of positions in the nature of bridgeheads, and to join the forces of the main offensive from the E. as these advanced.

On the Rumanian side expectation of an easy victory prevailed. In conformity with the plan of operations there was a concentric advance. The I. Army (Gen. Culcer), with about 4z inf. divs. and 3 ca y. bdes., and a stronger group on the E., advanced through the Roter Turin pass on Hermannstadt (Nagy Szeben); and with a weak group on the W. over the Vulkan and Szurduk passes in the line Petrozseny - Hatszeg (Hateg). The Orsova group (about one reinforced div.) on the Danube defended the left flank and the rear communications of the portions of the I. Army fighting in Transylvania against any advance from the Banat. The II. Army (Gen. Grainiceanu), with about 4 divs. and 4 ca y. bdes., operated from the Torzburg to the Ojtoz pass through all the defiles leading into the Kronstadt (Brasov) basin and the Haromszek. The IV., or Northern Army (Gen. Presan), with about 4 divs. and i ca y. bde., operated N. of the II. Army and in connexion with the Russian Carpathian front through the Uz, Gyimes, Bads and Tolgyes passes into the basins of the Csik and the Gyergyo. The III. Army (Gen. Averescu), with about 4 inf. divs. and a ca y. bde., faced Bulgaria on the Dobruja frontier in strong, well-fortified positions, and was to maintain the defensive. On the stretch of the Danube from Turnu Severinu to the mouth of the Alt stood protecting troops, one div. strong. In the district S. of Bucharest the Rumanians assembled a group of several res. divs. and other new formations for disposition as reserves.

The Rumanians crossed the frontier on the night of Aug. 27-28 over all the passes into Transylvania, driving back the weak Austro-Hungarian defence troops in numerous small engagements, not without suffering appreciable losses at many points. The Rumanian advance was substantially delayed by the destruction of roads and bridges effected by the Austro-Hungarian frontier troops, and especially by the bad roads of the mountain country.

By Sept. 3 the Rumanian Orsova group reached the lower course of the Cerna, and the western group of the I. Army occupied the important coal area between Urikany and Petroseny and had driven back the ineffective Landsturm and miners' battalions of the 144th Inf. Bde. over the saddle of Merisor. The eastern section of the I. Army, which had penetrated by the Roter Turm Pass, occupied positions S. of Hermannstadt, without attempting to take the town though it was garrisoned only by a weak Landsturm detachment. They were plainly apprehensive that by occupying Hermannstadt they would have to extend their bridgehead-like formation beyond capacity.

The II. Rumanian Army deployed cautiously in Burgenland and in the Haromszek, allowing their columns to close up, and receiving fresh reinforcements. The IV. Army forced their way, in continuous fighting touch with the 61st Inf. Div., through the narrow mountain valleys, and on Sept. 3 their advanced troops reached the eastern edges of the basins of the Gyergyo and the Csik. Meanwhile the first troops sent by the Central Powers were rolling up towards Transylvania. Gen. von Arz was instructed not to use the forces assembling on both army wings to strengthen the covering troops with them as they arrived, but first to concentrate them and hold them ready for wider action. In view of the expected continuation of the Rumanian advance he directed the 39th Honved Inf. Div. and what eventually, after many changes, became the 89th Inf. Div., to the district half-way between Szasz Regen (Reghina-Sas) and Klausenburg (Kolozsvar); the 187th Inf. Div. and 3 German ca y. refits. of the 3rd Ca y. Div. were to be disentrained at Maris Illye; the 1st Austro-Hungarian Ca y. Div. S. of this place between 1-latszeg (Hateg) and Karansebes; the ist Royal Hungarian Landsturm Hussar Bde. at Tovis. The first of the two German General Commands to arrive, Lt.-Gen. von Morgen, took over the command of the 61st and 71st Inf. Divs., the ist Landsturm Hussar Bde., the newly arrived 39th Honved Inf. Div. and the 89th German Inf. Div., while under Lt.-Gen. von Staabs were placed the 51st Honved Inf. Div., the 187th Inf. Div., the ist Ca y. Div. and the 3rd Ca y. Div., together with the covering troops at Hermannstadt, Hatszeg and Mehadia. The very slow progress of the Rumanians made it possible for the incoming divs. of the Austro-Hungarian army to move forward their disentrainment stations. Accordingly the following disentrainment arrangements were made: the 39th Honved Inf. Div. at Szasz Regen, the 89th Inf. Div. at Marosujvar, the 1st Cay. Troops Div. and the 3rd Ca y. Div. (which had been united in the Schmettow Ca y. Corps) at Mediasch and Elisabethstadt, the 187th Inf. Div. at Piski with a regt. intended for Hermannstadt at Alvinez.

Since the Rumanian group pushing northwards over P.etroseny might endanger the transport of further reinforcements on the Maros Valley railway, the bulk of the 187th Inf. Div. was directed against Merisor, in order, in conjunction with the Austro-Hungarian 144th Inf. Bde., and strengthened by the 3 first arriving German Jager batts. of the Alpine Corps, to throw back the Rumanian Mountain Corps over the frontier; and this task was accomplished between Sept. 14 and 22.

The Schmettow Ca y. Corps, linking up on the E. with the 51st Honved Inf. Div. standing directly N. of Hermannstadt, was posted on the heights N. of the Alt as far as Fogaras (Fagara). The Alpine Corps, which was only one div. strong but consisted of excellent troops, equipped for mountain warfare, was disentrained with the main body at Miihlbach. The German 76th Res. Div., which was on its way, was to be assembled at Karlsburg (Gyula Fehervar). The Austro-Hungarian 143rd Inf. Bde., which had been stationed at Hermannstadt, was moved behind the N. wing of the I. Army, and there formed into the 72nd Inf. Div. These measures, taken by the I. Army Command, on the one hand averted the menace to the Maros Valley railway at Piski, and on the other established the operative basis on which the battle of Hermannstadt was afterwards fought.

Bulgarian Offensive in the Dobruja. - Meanwhile events of far-reaching importance had taken place in the Dobruja. On Sept. i the III. Bulgarian Army crossed the Rumanian-Bulgarian frontier. The aim of the operation was the conquest of the Dobruja. After the capture of the bridge-heads of Turtucaia and Silistra the advance was to be made by the CernavodaConstantsa railway to the narrowest part of the territory lying between the Danube and the Black Sea. The fortress of Turtucaia consisted of a girdle of 15 forts on the S. bank of the Danube, which were connected by strongly built field positions. While very great care had been bestowed on the technical development of the place during the 3 years of preparation, the armament, consisting of only about zoo guns, including the field artillery, was inadequate. Artillery fire against Turtucaia began on Sept. 3; in the comprehensive attacks following on Sept. 4-5 and carried out by the 4th and sections of the 1st Bulgarian Inf. Divs. and the German detachment under Hammerstein, the bridge-head was stormed. The capture of this place by a coup de main was an admirable feat of arms. Only a very small portion of the garrison of the place, the 15th and 17th Rumanian Inf. Divs., which suffered heavy and bloody losses, escaped. Many soldiers were drowned in trying to swim the Danube, across which there remained no bridge. Twenty-one thousand men and 400 officers, including 3 brigade commanders, together with the whole armament, were captured.

While the remainder of the Bulgarian 1st Inf. Div. pushed forward by Akkartynlar and the 1st Ca y. Div. by Kurtunar, the 2nd Bde. of the 6th Inf. Div. and the garrison of Varna attacked the Rumanian 19th Div. on the plateau N. of Dobric (Hagi-Oglu) on Sept. 4 and threw them back northwards.

In contrast to Turtucaia the bridge-head of Silistra was in a state of unpreparedness. It fell on Sept. 9 into the hands of the cavalry of the 1st Div. after a short bombardment directed against the Rumanian cavalry.

While the Bulgarian III. Army was pressing forward successfully on the whole front, the retreating Rumanians were reenforced by the Russian Expeditionary Corps under Lt.-Gen. Zajanczkowski, which consisted of the Xlvii. Corps with 3, and later 4, inf. divs., among them the ist Serbian Div. (formed from Austro-Hungarian deserters) and the VI. Ca y. Corps. The Bulgarian III. Army put their main weight in the advance in the space between the Danube and the Dobric-Medzidie (Hagi-Oglu-Megidia) line, while E. of the railway on the right wing only sections of the 1st Ca y. Div. drawn from the centre of the army front operated.

On Sept. 15 the Rumanian-Russian fighting forces, which attempted to offer resistance on the line Lake Markeanu-Teke Deresi-Karalij-Kara Omer-Mangalia, were attacked by the Bulgars and compelled to retreat along the whole line. The III. Rumanian Army, reinforced by hurriedly-brought-up Russian and Rumanian units, prepared to fight again on the position Rasova-Copadin-Topra.i Sari-Urtukiiij, which immediately protected the Cernavoda-Constantsa railway and had been partially prepared in time of peace. The attacks executed by the Bulgarians on Sept. 19-20 did not penetrate the line this time. Instead, the III. Bulgarian Army Command were compelled to withdraw their troops some kilometres, to wait for the bringing up of munitions and the arrival of sections of the Bulgarian 12th Div., and the VI. Turkish Corps (25th and 15th Divs.). But the counter-attacks undertaken by the Rumanian eastern wing on Sept. 22 were repulsed by the recently arrived 25th Turkish Div., and the Bulgarian-Turkish front was again established on the line N. of Amuzacia.

In the Dobruja generally operations for the time being came to a standstill.

In Transylvania the IV. Rumanian Army advanced from the basins of the Gyergyo and the Csik through the Maros valley, then over the GOrgeny and Hargitta mountains, and continuously pressed hack the 61st Inf. Div., subsequently reenforced by the 1st Landsturm Cay. Bde. It was feared that it would all too quickly reach the inner region of Transylvania, with its excellent communications. In that case it would threaten the rear of the 71st Inf. Div. which occupied positions on the W. bank of the Alt (Oltu) between Fogaras and Reps, and farther N.E. to Homorod and Okland, at the weak angle where the front of the I. Army from a direction W. to E. bent sharply from S. to N. Generally, too, it would deprive the covering troops of the possibility of protecting according to plan the picked attacking troops coming up to the front. The commander of the east front, Lt.-Gen. von Morgen, therefore, planned to make a surprise attack on the Rumanian IV. Army on its emergence from the Gergeny and Hargitta mountains, using for the purpose the concentrated strength of newly arrived units. He proposed to attack either from the area N. of Szasz Regen (Reghina-Sas) southwards or from the upper course of the Great and Little Kukiillo valley in a N. E. direction, and by pressing on the Rumanian communications to prepare an annihilating defeat for them. This plan, however, was not approved in higher quarters; it was determined merely to strengthen the E. front by hurrying up the Austro-Hungarian 72nd Inf. Div., and a more active conduct of the defence was recommended. Both Supreme Army Commands adhered to the original plan of concentration and to the idea of striking first at the inactive enemy S. of Hermannstadt.

Before daybreak on Sept. 15 the Rumanian II. Army crossed the Alt between Fogaras and Reps in several columns, for the most part without bridges or river transport, and advanced farther N. from Barot, through Homorod - Okland - Draas towards Katzendorf. In order to make a mobile defence possible the 71st Inf. Div. had left only weak covering troops (about 3,000 rifles) on the 60-km. front, placing the main force in readiness in the district Petek-Mehburg. The weak defence naturally had to give ground before the far superior weight of the Rumanian attacking columns. But in the afternoon the main force of the div. made a surprise attack, advancing southwards through Palos and struck the 6th Inf. Div., marching as the most northerly column of the II. Army, in flank and rear. The surprise and confusion of the Rumanians were so great that the II. Army, which had only just crossed the Alt, ceased to advance, and remained inactive for a week.

The Rumanian IV. Army, on the other hand, continued to advance steadily, thereby compelling the command of the I. Army to support the Landsturm Cay. Bde. by 4 newly formed Bosno-Herzegovinian inf. batts., which really belonged to the unit of the 71st Inf. Div. In addition the 39th Honved Inf. Div., N.E. of Maros Vasarhely, was pushed up to the front; the 89th Inf. Div. was advanced to Maros Vasarhely, and an inf. bde. of the 37th Honved Inf. Div., coming up without artillery, was placed in readiness at Teke, N.W. of Szasz Regen. The newly arrived Austro-Hungarian VI. Corps Command took over the command of the N. wing (72nd Inf. Div., half the 61st Inf. Div. and half the 37th Honved Inf. Div.), while the I. Res. Corps Command retained command of the Landsturm Hussar Bde., the 39th Honved Inf. Div., the 71st Inf. Div., supported by the 19th Mountain Bde. of the 61st Inf. Div., and the 89th Inf. Division.

At the end of Sept. the IV. Rumanian Army in the N., with the reenforced 14th Div., had reached Deva in the Maros valley and Kasva in the Gorgeny valley, and with sections of the 8th Div., was already pressing at Kibed on the Kukiillo position. With the reenforced 7th Div. the west of Szekely-Keresztur was reached, where the 19th Mountain Bde., already much weakened, could only defend itself with difficulty against the overwhelming pressure, while the 71st Inf. Div. on their left wing had definitely to give way.

On the evening of Sept. 17 Gen. Erich von Falkenhayn, with the staff of the newly formed German IX. Army, arrived at Deva, and took over the command of Gen. von Staabs' troops, and of all the reenforcements coming into this district. His commission was, in conjunction with the I. Army, to throw the enemy out of Transylvania, and for this purpose, while masking the Vulkan and Szurduk passes, to surround the enemy posted at Hermannstadt, with a double ring, and beat him. Gen. von Falkenhayn first ordered Lt.-Gen. Sunkel, commanding the 187th Inf. Div. in the neighbourhood of Petroseny, who was about to push the Rumanians back to the frontier passes, after reaching this line to send all the troops he could spare from his div. and the Alpine Corps towards Hermannstadt; he ordered the assembly of the 187th Inf. Div. at Reussmarkt, of the Alpine Corps at Sinna, and the disentrainment of the 76th Res. Div. at Markt-Schelken. Finding by a reconnaissance in the direction of the Roter Turm Pass that the road was practicable for mountain troops without wheeled transport, he decided to direct the Alpine Corps by way of Cindrelu and Prezbe towards the Roter Turm, in order to hinder the retreat of the Rumanians by this route, while the 187th Inf. Div., the 51st Honved Inf. Div., the 76th Res. Inf. Div., and sections of the Schmettow Ca y. Corps, were to attack W. and E. of Hermannstadt in the direction of the northern outlet of the pass. This was not indeed a double encirclement of the enemy, as had been ordered by the Supreme Command, for which the forces of the eastern wing, where only a few squadrons could be made available, were insufficient. It was, however, a far-reaching enveloping movement against the one passable rearward communication of the enemy, in cooperation with an energetic attack on the front, of which the object was to destroy the group composed of the 2nd and 13th Rumanian Inf. Divs., under Gen. Popovici commanding the I. Corps.

On Sept. 22 Gen. Popovici attacked, but only attained success southward of Cornaticlu against the extremely thinly held positions of the 7th Ca y. Bde. of the 1st Div., being everywhere else completely repulsed. The expected continuation of the Rumanian attack on Sept. 23 did not take place, and it was possible to issue orders for the projected battle. By Sept. 25 the XXXIX. Res. Corps with the 187th Inf. Div. were able to be assembled at the foot of the mountains S.W. of Hermannstadt, the 51st Honved Inf. Div. to the N.W., and the 76th Res. Inf. Div. to the N.E. of the town, while the Alpine Corps was to be within a day's march of the Roter Turm Pass. The general attack in the direction of the pass was to begin on Sept. 26; the Alpine Corps was to endeavour to reach the E. side of the pass in order there also to block the bridle-tracks leading over the mountains. The Schmettow Ca y. Corps might, in the event of further pressure by the enemy, give way with its right wing, but with its centre on the Alt and its left wing towards Fogaras it was to hold its ground obstinately, and, in addition, if the operations proceeded according to plan, to arrange to push forward from the N.E. over the river towards the entrance of the pass. The I. Army Command was asked, as soon as possible, to place the 89th Inf. Div. in readiness at Schassburg (Segesvar).

The Rumanians standing at Hermannstadt did not interfere further with the preparations for the attack; on the other hand the 11th Rumanian Div. stationed at the Szurduk Pass attacked again on Sept. 25, and regained possession of Petroseny. The 144th Inf. Bde., reenforced by two German battalions and two batteries, held the heights N. of the place. The IX. Army Command did not contemplate further reenforcement, but the unattached staff (i.e. without troops) of the German 301st Infantry Div. was sent there, under the direction of which were placed the 144th Inf. Bde. and the Austro-Hungarian 2nd Mountain Bde., which had arrived on the 28th; and with these forces the Rumanians were again compelled to give up the extremely valuable coal basin.

On Sept. 26, favoured by beautiful autumn weather, the attack began, and it continued with undiminished violence against the obstinate defence of the Rumanians until the evening of Sept. 28. It ran, on the whole, the course intended by Gen. Falkenhayn.

The Alpine Corps had already reached Roter Turm, Riu Vadulin and Caneni with their advanced troops on the road to the pass, early in the forenoon of Sept. 26. The Rumanians indeed now thoroughly realized the magnitude of the danger which threatened them, and delivered the most violent counterattacks from N. and S. against the detachments of the Alpine Corps. These troops might be temporarily pushed back at one point or another, but the road over the pass now lay continuously under German fire, which inflicted heavy losses on the Rumanian columns still attempting to break through. The Alpine Corps, however, did not succeed in reaching the E. bank of the Alt, and sections of the Rumanians were thus able to escape the threatened encirclement, and to cross the western spurs of the Fogaras Mountains.

The three divs. of the XXXIX. Res. Corps made a concentric attack between Orlat - Hermannstadt and the heights to the E. of it. It was only with difficulty that they at first gained ground, and not till Sept. 28 did they succeed in breaking the Rumanian resistance; but then, in consequence of the heroic endurance of the Alpine Corps, which made both escape and the bringing up of reinforcements from the S. impossible, the Rumanian defeat became a complete collapse. This collapse was precipitated when on Sept. 27 the 3rd Ca y. Div. had succeeded, with two regts. of light horse, in crossing the Alt and narrowing S.W. of Porumbacu the circle of fire round the Rumanians, while the 18th Hussar Regt. of the 3rd Ca y. Div. at Chertisiora secured the front towards the E., whence there had been since Sept. 27 increasing indications of an advance by the Rumanian II. Army.

As the bringing up of reinforcements through the Roter Turm Pass from the S. failed, the Rumanian Command were compelled to set the II. Army on the march towards Hermannstadt to relieve the seriously threatened group under Gen. Popovici. The II. Rumanian Army executed their movements slowly and with difficulty, and since an advance on the shortest line in the Alt valley from the N. over Fogaras from the AgnethelnHenndorf district might easily have been threatened on the flank, the Rumanian Army Command thought themselves first compelled to secure freedom of movement N. of the Alt river. The sections of the 71st Inf. Div. in the forward positions were therefore first pressed back, and then the 6th Ca y. Bde. of the 1st Ca y. Div. standing N. of Klein Schenk were thrown back westwards. Meanwhile, the Austro-Hungarian I. Army Command had sent from Schassburg to Henndorf the greater part of the 89th Inf. Div., one infantry regt. and one light fieldhowitzer detachment going off by rail as army reserve to Salzburg (N. of Hermannstadt). Pushing between the 71st Inf. Div. and the Ca y. Corps, they made on Sept. 28 a successful attack in a southerly direction, and so put the brake on the advance of the Rumanian troops N. of the Alt. The I. Army of Gen. Arz had to withdraw steadily westwards under the superior weight of the IV. Rumanian Army, and as the seriously weakened 19th Landsturm Mountain Bde. especially had great difficulty in withstanding the continued Rumanian attacks in the direction of Schassburg, the I. Army Command considered it necessary to withdraw the 71st Inf. Div. to the Little Kukiillo (Kokel). Gen. Falkenhayn urgently dissuaded them from this move, as it would expose his eastern flank to an unbearable threat. He also expressed his doubt as to the ability of the I. Army, when once it had been pressed back behind the line of the Maros and the Little Kukiillo, to maintain that position permanently with its present forces. Thereupon the withdrawal of the southern wing of the I. Army was delayed.

On the afternoon of Sept. 28 the Rumanians again attacked the 1st Ca y. Div. N. of the Alt, and pressed them back to the heights E. of the Haarbach; the reserve, not required at Hermannstadt, was hastily sent with heavy motor wagons through the Haarbach valley to the aid of the heavily engaged Schmettow Cavalry Corps. Meanwhile, however, the fate of the Rumanians in the Roter Turm Pass was sealed, the attacking troops of the XXXIX. Res. Corps ceasing to meet with serious resistance in the early morning of Sept. 29. Those who were not able to escape through the forests over the mountains fell a sacrifice to the inexorable onslaught. The bulk of the Rumanian I. Army was destroyed. Three thousand prisoners - a relatively small number - were taken, but the whole of the artillery and the whole train fell into the hands of the victors.

It was now necessary rapidly to take new decisions for fighting the II. Rumanian Army, the threat of whose approach was im minent. In accordance with the instructions received, the IX. Army was to gather all its strength, and to deliver an enveloping attack from the S. against the southern wing of Rumanian main forces pushed forward W. of Fogaras: Falkenhayn intended to relieve the Alpine Corps for this purpose by the 5rst Honved Inf. Div., to assemble the 76th and 187th Inf. Divs. on the heights of Scorei on both sides of the Alt, and then to push forward in an easterly direction, an enveloping attacking movement in the Fogaras mountains being assigned to the Alpine Corps. But the rapid and violent push of the Rumanians in the space between the Haarbach and the Alt on Sept. 29 entailed changes in the plan of operations. The relief of the Alpine Corps had to be given up, as involving too much time; instead, the 51st Honved Inf. Div. and the 76th Res. Div. were to reach as rapidly as possible the Alt valley S. and N. of Avrigu and the 187th Inf. Div. Cornatielu in the Haarbach valley. The seriously weakened Cavalry Corps was to attach itself for the forward movement to the N. wing of the 187th Inf. Div. Of the I. Army, the 89th Inf. Div. and the strongest possible sections of the 71st Inf. Div., under the command of Lt.-Gen. von Morgen, were asked to attack in the direction of Bekokten. The beginning of the attack was proposed for Oct. 1.

To the surprise of their enemy the Rumanians did not continue the attack N. of the Alt on Sept. 30, but withdrew a little from the Cavalry Corps. With this object they attacked S. of the Alt and drove back the 18th Hussar Regt. westwards of Chertisiora. The weariness of the troops and the almost impassable state of the roads, owing to the rain which had set in, delayed the forward movements, and it was agreed to begin the attack on Oct. 2. The Rumanians did not take advantage of the loss of time this entailed on the German-Austrian side, but entrenched themselves in the positions they had reached.

The unification of the command in Transylvania was established by placing the Austro-Hungarian I. Army from Oct. r under the operative control of Falkenhayn.

On Oct. 2 began the advance of the XXXIX. Res. Corps, the Schmettow Ca y. Corps and the I. Res. Corps. South of the Alt the Rumanians offered no resistance, but retreated according to plan before the German advanced troops. North of the Alt, after strong forces had been brought up by way of Gross Schenk in a westerly direction, the advance also began. The 89th Inf. Div. attacked in the direction of Bekokten, and at first obtained a great success, but was thrown back to its point of departure by a Rumanian counter-attack. The 71st Inf. Div. had not been able to join in this attack because its artillery was not yet in position on account of the softness of the chalky roads after rain. Lt.-Gen. von Morgen thought the situation of these two divs. so endangered that he intended to withdraw them as far as the sector Henndorf - Jakobsdorf. On Oct. 3, however, this idea was abandoned, as the enemy themselves had withdrawn eastwards. Owing to this movement touch with the Rumanians became extremely loose, which made it exceedingly difficult for the Austro-German Command to discover betimes the measures taken by them.

On their side the Rumanians had obviously abandoned as early as Oct. 2 the idea of continuing the offensive. Impressed by the annihilating defeat at Hermannstadt and recognizing the impossibility of attacking in a tactically unfavourable situation the IX. Army, rapidly advanced eastwards, they had decided to withdraw betimes in order to defend the frontier passes. In order to secure the time necessary for the threading of the marching columns into the passes of the Geisterwald, the Hargitta and the GOrgeny mountains, the Rumanians undertook a series of forward pushes: on Oct. 1 S. of the Alt, and on Oct. 2 against the 89th Inf. Div. Against the I. Army these attacks continued until Oct. 5, and during them the Rumanians, especially on Oct. 3, obtained a fresh success against the 10th Landsturm Mountain Bde. and the Landsturm Hussars.

Though the IX. Army Command could not yet fully discern the intentions of the enemy, the puzzling behaviour of their opponents seemed no reason for delay, and the advance was I therefore pushed forward with the utmost speed. The 3 divs. of Lt.-Gen. Staabs were taken into the space S. of the Alt, the I. Res. Corps was to reach the passages over the Alt at Comana and Heviz, while the Ca y. Corps was to reach the N. wing of the IX. Army.

The Rumanians retired on the whole E. front, without being brought into action by the pursuing troops of the Austro-Hungarian I. Army and the German IX. Army. It was only after dusk on Oct. 4 that the XXXIX. Res. Corps was able to bring the Rumanians to a stand on the western slope of the Geisterwald in a prepared position behind the Sinca brook. The corps were ordered to attack early on Oct. 5, while the 76th Res. Div. was to advance along the high road to Volkany, the 51st Honved Inf. Div. to Vledeny, and the 187th Inf. Div. over the mountains, enveloping the enemy on the N., in the direction of Krizba. The 8th Mountain Bde., just arrived at Hermannstadt, was ordered to follow the XXXIX. Res. Corps forthwith in the Alt valley. It was intended to allow them to advance W. of the KOnigstein towards the road Kronstadt - Campolung.

Morning mists and the time taken by the enveloping movement of the 187th Inf. Div. in roadless mountain country delayed the beginning of the attack on Oct. 5. In order to lose no time Lt.-Gen. von Staabs ordered the 76th Res. Div. and the 51st Honved Inf. Div. to attack alone in the forenoon; they soon captured the Rumanian positions, the Rumanian 4th and 3rd Inf. Divs. suffering heavy and bloody losses in their violent counter-attacks. But when the enveloping movement of the 187th Inf. Div. became effective the Rumanians began their retreat into the Kronstadt basin with the utmost haste, at the cost of a great part of their artillery. Close upon them followed the victorious divs. of Gen. von Staabs. Meanwhile the advance guard of the 89th Inf. Div. had reached Comana on Oct. 5, after, by quick action, succeeding in putting out the fire which the Rumanians had set to the bridge. After a stiff pursuing action the 71st Inf. Div. took Reps, but, N. of Heviz, met with strong resistance from the enemy which was only broken down on the morning of the 6th. The 89th Inf. Div. which had been brought up here, was marched through Heviz in front of the 71st and directed over the Bogat saddle towards Foldvar. The Schmettow Ca y. Corps had assigned to it the task of throwing back the 2nd Rumanian Ca y. Div. over the line Mehburg - Palos. But the Rumanian Horse escaped attack by a hasty retreat towards the N.E., and established temporarily contact in the Upper Alt valley between the two Rumanian armies, which were diverging more and more.

On Oct. 6 the divs. of the XXXIX. and I. Res. Corps in their marching lines sought to reach the western outlets of the defiles of the Burgenland. The attack on the 3rd, 4th and 6th Divs. of the Rumanian II. Army, crowded together around Kronstadt and entangled with one another during the retreat, was fixed for Oct. 7. The 76th Res. Div. was to reach the TOrzburg Pass by way of TohanuluTSrzburg. Kronstadt was the goal fixed for the 51st Honved Inf. Div., advancing by Feketehalom, while the 187th Inf. Div., attacking to the N. of it, was to wheel inwards, its flank protected from the N.E. in order to envelop Kronstadt and the entrance to the pass S.E. of it. On Oct. 7 the 89th Inf. Div. was to reach FOldvar and the 71st Inf. Div. Miklosvar. Of the Cavalry Corps the 3rd German Ca y. Div. was to push forward through Barot towards Mikoujfalu, to hinder Rumanian movements of troops in the Alt valley; the 1st Ca y. Div., pressing forward towards Szt. Egyhazas-Olahfalu, was to bar the retreat of the rear sections of the Rumanian troops still on the SzekelyUdvarhely - Csiksereda road.

On the early morning of Oct. 7 the vanguard of the 76th Res. Div. emerging from the mountains at Tohanulu was caught by the Rumanian artillery fire, and could penetrate no farther. The main body had therefore to make a wide detour by Zernesti against the Rumanian left flank, and a pause was made for the arrival of the heavy artillery. Thus this div. could make no further progress on the 7th. But the 51st Honved Inf. Div. and the 187th Inf. Div. rapidly approached Kronstadt, meeting, however, with violent resistance from the Rumanians on the N. and W. sides of the town, so that it was not until evening that the vanguard of the 187th succeeded in penetrating into the northern part of the town, where an obstinate street fight raged all night. Next morning the 51st Honved Inf. Div. also won their way in and stormed the heights S. of the town.

In consequence of the enveloping movement through Zerneste and the threat exercised by the 8th Mountain Bde. approaching W. of the KOnigstein it became possible for the 76th Res. Div. on Oct. 8 to seize Torzburg and the heights on either side of it, together with the entrance to the Torzburg Pass. The advance against the pass was continued, and, in addition, a detachment was pushed forward through the Klein Weidenbach valley towards the Tomos Pass in order to bar the Rumanian retreat here. Although this div. failed to reach the road, its appearance in threatening proximity caused a panic-like flight of the troops and transport hastening southwards.

Meanwhile the Rumanians tried to hold up the German advance N. of Kronstadt, and, with reenforcements hurried up partly by rail from Sepsi-Szt. Gyorgy, delivered violent counterattacks against the E. wing of the 187th Inf. Div., standing at Szentpeter, which was hard pressed till the attack of the 89th Inf. Div. from the N. struck the Rumanians unawares.

Early on Oct. 9 the victory of the IX. Army was complete. The beaten troops of the 3rd, 4th and 6th Rumanian Divs. retreated hurriedly through the passes, so that, supported by the loth, 21st and 22nd Inf. Divs. brought up for the purpose, they might undertake the defence of their country against the pursuing German and Austro-Hungarian divs. in fortified positions on the frontier prepared during peace.

Gen. von Falkenhayn in his pursuit tried to cross the mountains simultaneously with the Rumanians, and by a fresh distribution of his army, the I. Res. Corps with the 76th Res. Div. and the 8th Mountain Bde. attacked over the TOrzburg Pass in the direction of Campolung. Through the encircling movement of the 8th Mountain Bde. the pass was soon successfully opened, and the 2 2nd Inf. Div. which had arrived to support the seriously exhausted Rumanian 4th Inf. Div. was repulsed. But the attack of the I. Res. Corps was held up by the stronglyfortified positions N. of Campolung.

The XXXIX. Res. Corps had orders to push forward through the Tomos Pass with the 51st Honved Inf. Div., and through the Altschanz Pass with the 187th Inf. Div. towards the line Sinaia - Isorele. The 51st Div. did indeed succeed in storming the summit of the pass, but could not penetrate the 21st and 10th Rumanian Inf. Divs. in their strongly constructed positions. The 187th Div. had a similar experience against the Rumanian 3rd Inf. Division.

The 89th Inf. Div. had to attack through the Tatarhavas and Bodza passes. After reaching the basin lying S. of the frontier, it was held up by the main body of the Rumanian 6th Inf. Div. and by separate regts. of the 3rd, 15th and 22nd Inf. Divisions.

As the German Supreme Command urgently demanded that the strongest possible infantry and cavalry forces should be directed towards Ocna, to control the communications from there northwards by rail, road and telegraph, the 71st Inf. Div. was put under the command of Gen. Count Schmettow, commanding the Ca y. Corps, who led the div. in forced marches to the Ojtoz Pass. On the summit of the pass the div. overran a position held by the Rumanian 2nd Ca y. Div. and forced their way over the frontier. Recognizing their peril the Rumanians rapidly pushed up the 38th Inf. Bde. and sections of the 7th, 8th and the newly formed 15th Inf. Divs., and after long engagements with many vicissitudes prevented the 71st Inf. Div. from reaching its goal.

The 3rd Ca y. Div. assembled first in the basin of Kezdivasarhely, where the ist Ca y. Div., which had pursued the Rumanian 7th Inf. Div. up to the Uz Pass, had also been brought up. As the employment of cavalry on the route by way of Ocna into Moldavia had become impossible, the ist Ca y. Div. established communication in the forest-clad mountains, with their lack of roads, between the 89th and the 71st Inf. Divs. The three regts.

of the 3rd Ca y. Div. were later on stationed between the Torzburg and the Tvmos passes as the Transylvania Cavalry Brigade.

At the Roter Turm Pass the Rumanians - the remainder of the 13th and 23rd Inf. Divs. and the 2nd mixed Brigade of the 18th Div. - had discontinued their attacks against the Alpine Corps, reinforced by the 10th Mountain Bde. At Petroseny the i ith Rumanian Inf. Div. had again been pressed back to the frontier, whereupon the 2nd Mountain Bde. was shifted to the Roter Turm Pass. This was subsequently merged with the 10th Mountain Bde. in the 73rd Inf. Division.

In the Austro-Hungarian army the VI. Corps, with the 39th Honved Inf. Div., reached the frontier in the Uz valley and with the Gist Inf. Div. and the ist Landsturm Hussar Bde., in the Trotus valley advanced far over the frontier and, after fighting with varying success against the Rumanian 7th Inf. Div., occupied positions on the height of Sulta. On the N. wing the XXI. Corps with the 72nd Inf. Div. reached the Bekas Pass, and with the 37th Honved Inf. Div. the TOlgyes Pass. Thus Transylvania, six weeks after the invasion of the Rumanians, was again freed from the invader.

New plans had now to be agreed upon, in order to beat the Rumanians in their own country. Naturally the centre of gravity of the operations against Rumania lay in the first instance in Falkenhayn's IX. Army. His attempt to push forward on the shortest line to Bucharest with the troops he had in hand in the pursuit over the passes S. of Kronstadt had not succeeded. The Rumanians now defended themselves much more obstinately, and the German and Austro-Hungarian troops, wearied with their rapid operations, and with their war establishments weakened, had suffered temporarily in buoyancy from this victorious career. Events moved slowly also on the Roter Turm Pass, from which, after crossing the mountains, the main push directed towards Bucharest ought to have been supported by an advance of the reenforced Alpine Corps through Pitesti. The pursuit on all the many passes radiating from the Kronstadt basin had dissipated strength, and made the assembly of a strong main force impossible. New forces had to be brought up. These rolled up in Transylvania in the middle of Oct. - the 8th Bavarian Res. Div., the 11th and 12th Bavarian Inf. Divs. and the 6th German Ca y. Div.; towards the end of Oct. two further German inf. divs. (the 41st and 109th) and the 7th Ca y. Div. were to follow. Moreover, the Austro-Hungarian Higher Command intended to transfer the Austro-Hungarian 3rd and 10th Ca y. Divs. to Transylvania, but these would first have to be equipped and organized for employment in the intended offensive.

The 8th Bavarian Res. Div. was sent to the Transylvanian E. front to reinforce the I. Army. The 12th Bavarian Inf. Div. was placed by Falkenhayn under the I. Res. Corps on the Torzburg Pass, the 11th Bavarian Inf. Div. was to attack over the Szurduk Pass with the 144th Inf. Bde., and the group of Lt.-Gen. von Krafft at the Roter Turm Pass was strengthened by 2 Bavarian inf. regts. and 2 German Landsturm regts. At the T om6s, Torzburg, Roter Turm and Szurduk passes the attacks were to be continued, and wherever a gap was first effected Falkenhayn intended to bring up the mass of ca y. and the two later arriving inf. divs. to open up the remaining passes southward and in conjunction with Field-Marshal Mackensen's troops, to push forward towards Bucharest.

Both the Supreme Army Commands agreed to this plan. But the Higher Command at Teschen maintained in this connexion that it was desirable for the main pressure to be directed on the line Kronstadt - Bucharest. There the strongest opposing Rumanian and Russian opposing forces were to be expected; moreover, here they had to reckon with a threat of a Russian relieving offensive, urgently asked for by the Rumanians, coming from Moldavia in the general direction of Csik-Szereda. Falkenhayn therefore rather favoured a push through the Szurduk Pass, where, owing to the smaller width of the mountain chain, the Wallachian Plain would be most quickly reached.

On the E. front, meanwhile, the headquarters of the Army Front Commander, Archduke Charles Francis Joseph, in the arrangement of the commands, was moved from East Galicia to Grosswardein, as from Oct. 13, and the German IX. and Austro-Hungarian I., VII. and III. Armies were placed under him.

After the battle of Kronstadt the Rumanians were entirely reduced to the defensive. On the Transylvanian front they limited their activities to attempts to win back the lost frontier heights commanding important roads of invasion. The Rumanian Army Command also tried to induce the Russians to relieve the Rumanian troops in the Dobruja and on the Transylvanian E. front in order thus to set free forces for the defence of Wallachia.

On the Danube front the Rumanians on Oct. i had crossed the river at Rahova (S. of Bucharest) with a div., and had temporarily gained a firm footing. German and Bulgarian troops, rapidly assembled, compelled the Rumanians to return to the N. bank, the latter suffering severe losses, as the AustroHungarian Danube monitors had shot to pieces the Rumanian pontoon bridge. Rumanian forward pushes against the Bulgarian III. Army brought no success. On Oct. 19 an attack by Gen. Toshev's Army (Bulgarian 1st, 4th, 6th Divs., and sections of the 12th Inf. Div., ist Ca y. Div., Turkish VI. Corps, with the 15th and 25th Inf. Divs., German 217th Inf. Div.), broke through the Russo-Rumanian front on their E. wing, and drove the opposing army far over the Cernavoda - Constantsa railway, Rumania thereby losing her only rail connexion with the sea.

While the bulk of the Bulgarian III. Army followed only as far as the line Lake Tasaul - Bazanliia - heights of Kualnik - Danube S. of Topal, and settled themselves for the defence on this shortest line between the Danube and the sea, the reinforced ca y. div. pursued the retiring Rumanians and Russians as far as the line Sariuri - Sarighol - Docuzaci. Gradually the Russians again slowly pushed forward southward against the new position of the Bulgarian III. Army. The Rumanian troops were withdrawn in Nov. from the Dobruja into Wallachia. Of the Russians there were in the Dobruja the VI. Ca y. Corps, the Xlvii. and IV. Siberian Corps, with 6 inf. divs. and i ca y. div. in all, which were placed under the command of the newly formed Russian Danube Army (Gen. Sakharov).

In the new defensive position of the Bulgarian III. Army, which was by this time under the command of Gen. Neresov, there remained the 4th and the combined 6th Inf. Divs., then the ist Ca y. Div. The Turkish VI. Corps stood for the time being at Medzidie in reserve. The other troops in the Dobruja and northern Bulgaria, together with the expected Turkish 26th Div., were collected in the district around Sistova, and were placed in readiness for crossing the Danube as the new Danube Arm y under the command of Gen. Kosch.

On the Transylvanian S. front the obstinate struggle for the passes was continued. The I. Res. Corps succeeded in reaching a point just N. of C.inpolung after the arrival of the 12th Bavarian Inf. Div. and with the assistance of the enveloping movement in the mountains of the 8th Mountain Bde. on the W. wing. At that point irruption into the basin of Campolung was barred by a new strongly constructed position in which the newly broughtup Rumanian 12th Inf. Div., in addition to the 22nd Inf. Div., offered the most obstinate resistance.

At the Roter Turm Pass Lt.-Gen. von Krafft intended to force an exit from the mountains by enveloping on two sides, with the 2nd Mountain Bde. eastwards with the 10th Mountain Bde. westwards, and the Alpine Corps in the centre. The attack began on Oct. 16. After easy initial successes the weather broke on Oct. 18, and this circumstance, together with hastily executed Rumanian counter-attacks, prevented complete success.

South of Petroseny the group of Lt.-Gen. Kneusel, with the 1th Bavarian Inf. Div., the 144th Inf. Bde., and the 6th Cay. Div., began the attack in numerous columns through the Szurduk and Vulkan passes and over the heights to the west. In spite of the fall of snow the advance began on Oct. 23.

News had been received that, under pressure of the preceding attack by Krafft's group and the I. Res. Corps, the Rumanians had deflected against these reinforcements which had been sent up, and that it would therefore be easier to break through. At first, indeed, complete success attended the attacks of the Kneusel group. The troops, forcing back the Rumanian nth Inf. Div., had worked their way to the foot of the mountains N. of Targu Jiu, and were to fight their way out to the plain on Oct. 27. At this point, however, a counter-attack by the hurriedly summoned Rumanian 21st and 22nd Inf. Bdes., and a regt. of the 1st and 3rd Inf. Divs., struck the W. wing. After losing many in prisoners and guns the German detachments had again to withdraw to the frontier heights, whither the Rumanians pursued them only with skirmishing detachments. In spite of the defeat he had suffered Falkenhayn held fast to the idea of a break-through by way of Targu Jiu, and directed the newly arriving troops (41st and 109th Inf. Divs. and 7th Ca y. Div.) to Petroseny. After the experience just gained the most thorough preparations were to be made for this operation, which was to begin on Nov. II, Lt.-Gen. Kiihne of the LIV. Res. Corps being chosen for the command of this strengthened group.

Urged by the Rumanian Army Command, the Russians relieved the Rumanian troops facing the Austro-Hungarian I. Army, beginning from the N., and pushed the southern boundary of their IX. Army in the middle of Nov. to a point just N. of the Gyimes Pass. Simultaneously with the III. Ca y. Corps and the XXXVI. Corps, they attacked in this new sector the AustroHungarian XXI. Corps on Nov. 6. In expectation of this Russian push forward the army front command had placed in reserve the Brudermann Ca y. Corps (3rd and 10th Ca y. Div.) in the district Olah-Toplica-Gyergyo-Szt. Miklos, the 10th Bavarian Inf. Div. brought southwards from the VII. Army in the district around Csik-Szereda, and the bulk of the 8th Bavarian Res. Div. at Kezvivasarhely. The Russian forward movement obtained small successes on both sides of the TOlgyes and Bekas Pass. After the bringing up of the 10th Bavarian Inf. Div. and the 3rd Ca y. Div. the situation was once more restored.

In the Ojtoz Pass also, where the 71st Inf. Div. and the 1st Ca y. Div. were once more placed under the I. Army as from Oct. 29, and in the Trotus valley, the Rumanians, partly mixed with Russian units, attacked on Nov. 5 without obtaining noteworthy successes.

The Break-through at Targu Jiu. - According to plan, the attack of the group of Lt.-Gen. Kiihne began on Nov. I i S. of the Szurduk and Vulkan passes. They were to force their way into the Wallachian plain before the approach of winter made mountain operations impossible. Simultaneous attacks on all the other passes of the Transylvanian S. front and at Orsova were to distract the attention of the Rumanians and divert their reinforcements from the principal theatre of attack, an intention which was successfully accomplished.

Protected by the 41st Inf. Div. on the W., with the 109th and 301st Inf. Divs. E. of the river Schyl, on the W. flank by sections of the 6th Ca y. Div. and the 9th Regt. of the Hungarian Landsturm, the troops fought their way out of the mountains in an obstinate struggle lasting from Nov. II to 14, and on Nov. 15 reached Targu Jiu. The Rumanian nth Inf. Div., seriously weakened, retired to the heights S. of the town, where it again gave battle with rapidly brought up new forces of about the strength of two divisions. In the Kiihne group the Schmettow Ca y. Corps (6th and 7th Ca y. Divs.) was brought along the road over the pass and placed on the W. wing for the envelopment of the enemy; the nth Bavarian Inf. Div., hitherto in reserve, was placed on the front E. of Targu Jiu, while the 301st Inf. Div. acted as covering troops on the east.

On Nov. 16 the Kiihne group attacked once more. On Nov. 17 the Rumanians, in spite of the most courageous defence, were decisively beaten. The road into Wallachia lay open. The pursuit was undertaken without delay. With the right wing (6th Ca y. Div. and behind that the 41st) in the Jiu valley through Craiova, the centre (109th and nth Ba y. Inf. Divs.) towards Slatina, and the left wing (301st Inf. Div.) in the direction of Dragasani, the group swerved eastwards and made rapidly for the Alt. On Nov. 21, Craiova, the capital of Wallachia, was reached. The rapidly attacking vanguard of the 6th Cay. Div. succeeded on the 23rd in seizing the bridge E. of Caracalu, which had remained undamaged, over which the main body of this division on Nov. 24 and the 7th Cay. Div. on the 25th crossed the Alt, in order to push on against the Vede sector.

The Rumanians, repulsed from Targu Jiu (i i th and 17th Inf. Divs. and parts of other divs.) placed themselves after the destruction of the bridges on the E. bank of the Alt between Slatina and Dragasani, in order to bar at this point an advance by Lt.-Gen. Kiihne's troops. Farther N. too, opposite the group of Lt.-Gen. von Krafft, the Rumanians had evacuated the W. bank of the Alt, so that the German troops were able to occupy Rimnik Valcea on Nov. 25. The attempts of the 41st and nth th Ba y. Inf.. Divs. on Nov. 25 and 26 to cross the Alt at Slatina failed, in spite of the support of some squadrons of the 7th Ca y. Div., which had already come into action from a S.E. direction. The 109th Inf. Div. was now sent in support of the Ca y. Corps by way of Caracalu, and was soon followed by the iith Ba y. Inf. Div. and the 115th Inf. Division.

In consequence of the rapid break-through at Targu Jiu the retreat of the Rumanian Orsova group, 3 regts. of the 1st Inf. Div. with artillery, was cut off. Held in front by violent attacks on the part of the group of Col. Szivo, they were shut in on the rear by detachments of the Kiihne group. In a series of engagements in which at one time they threatened the rear communications of the Kiihne group, this Rumanian group went down along the Danube, until, completely surrounded at the mouth of the Alt, they laid down their arms before their pursuers on Dec. 6. Ten thousand men and 40 guns fell into the hands of the much weaker Szivo group.

In front of the group of Lt.-Gen. von Krafft, reinforced by the newly arrived 216th Inf. Div., the Rumanians also could not hold their own on the E. bank of the Alt in spite of the participation of the 7th and parts of the 8th Inf. Divs.; they retreated as far as Curtea d'Arges and behind the Topolog sector, where they offered a temporary resistance.

The I. and XXXIX. Res. Corps (under which latter the 89th Inf. Div. in the Bodza Pass had been placed) maintained undiminished pressure on the Rumanian groups opposed to them. With the aim of building up a further reserve of the army front, the 187th Inf. Div. was relieved by the approaching AustroHungarian 24th Inf. Div., and placed in readiness in the Haromszek. The Ojtoz group now under the command of Gen. von Gerok, of the XXIV. Res. Corps, was on Nov. 12 again placed under the IX. Army Command. On the E. front the Russians continued the relief of the Rumanians as far as the road over the Ojtoz Pass.

On the side of the Central Powers the Army Command now thought the moment had arrived for the Danube Army in position at Sistova to cross the Danube and push forward towards Bucharest, in order, in conjunction with the approaching IX. Army, to effect the complete conquest of Wallachia. The Danube Army consisted of the 217th German Inf. Div., the 1st and 12th Bulgarian Inf. Divs., the combined Ca y. Div. of Maj.-Gen. Goltz, German and Bulgarian Landsturm troops, German and Austro-Hungarian heavy artillery, the 26th Turkish Inf. Div. and Austro-Hungarian pioneer formations. At 4 A.M. on Nov. 23, favoured by thick mist, and supported by the Austro-Hungarian Danube monitors and the German motor-boat flotilla, the transport across the river of the 217th Inf. Div., unnoticed by the enemy, was successfully accomplished without delays. Zimnica was occupied. Then the ist Bulgarian Inf. Div. and the Landsturm formations crossed; the resistance of Rumanian detachments brought up was rapidly conquered.

On Nov. 24 the bridge-head was widened, and the construction of a pontoon bridge by the Austro-Hungarian pioneer group of Maj.-Gen. Gaugl was begun, and finished in the afternoon of Nov. 25 at 6 o'clock. The remaining troops were now brought over the bridge in unbroken sequence, and the advance was begun; on the left wing the ca y. div. towards Alecsandri, on its right the 217th Inf. Div., then the 12th and 1st Bulgarian Inf. Divs. The Turkish 26th Inf. Div. followed as Army Reserve behind the left wing. Rapidly advancing, and quickly breaking the resistance of the Rumanian 18th Inf. Div. and the ist and 2nd Ca y. Divs. sent against them, the heads of the columns had already on Dec. 1 reached the Argesu, S.W. of Bucharest. But in this hurried forward movement the Danube Army, after establishing only slight contact by means of cavalry at Rosi de Vede, again lost touch with the main body of the IX. Army, held up on the Alt sector; their left flank lay open.

The Rumanians recognized the opportunity offered them of falling on the rashly advanced Danube Army. They endeavoured, with their I. Army, to keep the Kiihne and Krafft groups as far to the W. as possible, and also made violent frontal attacks across the Argesu on the isolated Danube Army, and on Dec. 2 from the N.W., completely encircling the left wing of the Danube Army, with the Rumanian 1st Ca y. Div., then parts of the 2nd, 5th, 9th and 19th Inf. Divs. The Danube Army was thus placed in an extremely critical position. Rapidly brought-up Landsturm battalions, a few pioneer companies and the 26th Turkish Inf. Div., advancing in the second line, compelled a pause in the Rumanian enveloping movements. This Rumanian manoeuvre, which only failed of success because it was not executed with sufficient energy, was coincident with violent pushes carried out by the Russians on the Carpathian front, from the Tartar Pass southwards to the Ojtoz Pass and on the Dobruja front, and also with attacks by Gen. Sarrail's Army on the Salonika front, by which it was hoped to relieve the hard-pressed Rumanian Army and to snatch from the Central Powers the advantage developing in this area. Yet all efforts were in vain.

The right wing of the IX. Army, which had been placed from Nov. 30 under the army group command of Field-Marshal von Mackensen, was brought up with the utmost haste. The iogth Inf. Div., advancing northward on Nov. 27 and 28 on the E. bank of the Alt, had at last succeeded in compelling the Rumanian I. Army to abandon the Alt sector. The 41st and 301st Inf. Divs. could then cross the river at Slatina. The pursuit towards the E. was conducted in the following groups: along the projected. Craiova - Bucharest railway the 11th Bavarian Inf. Div.; behind that the 115th Inf. Div.; N. of the Bavarians the 109th, 41st and 30lst Inf. Divs. The Schmettow Cay. Corps had ridden in advance of the right wing. Thus the IX. Army approached the seriously threatened left wing of the Danube Army. On Dec. 2 parts of the Cay. Corps, and on Dec. 3 the 11th Bavarian and 109th Inf. Divs., swerving southwards, were able to participate in the battle. The Rumanians, now themselves enveloped, turned back with heavy losses to Bucharest. For the Danube Army the crisis was over.

While on Dec. 2 and 3 the main body of the Schmettow Ca y. Corps and the 109th Inf. Div. covered the road to Bucharest, Lt.-Gen. von Krafft at the same time, with the 216th, 73rd and 30lst Inf. Divs., struck the remnant of the Rumanian I. Army on the middle course of the Argesu, and pushed forward with the Alpine Corps and the 2nd Mountain Bde. towards Tirgoviste, which, after the capture of Campolung, the I. Res. Corps was also approaching.

Attacks by the just arrived Russian 40th Inf. Div. and the 8th Ca y. Div. on Dec. 4 and 5 against the Bulgarians on the S. wing of the Danube Army gave no results. The violent attacks delivered by the Russians against the Austro-Hungarian Army and against the Bulgar-Turkish Dobruja front in the beginning of Dec. were also continuously repulsed.

On the evening of Dec. 5, after successful engagements, the Danube Army stood E. of the Argesu and S.W. of Bucharest, and the IX. Army in close touch N. of the town as far as the Prahova valley. Since it was doubtful whether Bucharest would be defended as a fortress, heavy artillery and all the means of attack were placed ready to hasten its capture. In the night of Dec. 5-6 cavalry of the Schmettow Corps rode up towards the N.W. front, and found the works blown up and ungarrisoned. The Rumanians evacuated their capital almost without fighting. On the night of Dec. 6 the troops of the Danube Army and parts of the S. wing of the IX. Army entered Bucharest, while on the same day Falkenhayn's N. wing captured Ploesci, and with it the important petroleum area, where English hands had previously rendered the boring apparatus useless for a considerable length of time. Two days later, as the result of rapid enveloping movements carried out by Lt.-Gen. Morgen's group, the 4th Rumanian Div., left stranded in the mountains, were surrounded in the district N. of Ploesci, and were taken prisoners. The road to the S. now also lay open to Lt.-Gen. Staabs' group. The 51st Honved Inf. Div. was able to occupy Sinaia.

Pursuit to the Danube - Sereth Line. - Field-Marshal von Mackensen now received the order to push forward with his army group (III. Bulgarian Army, Danube Army and IX. German Army) to the shortest line of communication between the sea and the Carpathian front, that is the Danube mouth - GalatzSereth to Ajudumiu - Trotus river. The IX. Army was to advance between the mountain river and the projected railway line Bucharest - Urziceni - Foreivechii - Tecuciu; the Danube Army between this line and the Danube; the Bulgarian III. Army to advance in the Dobruja.

On Dec. 12 the IX. Army threw the Rumanian I. and II. Armies, reinforced by the Russian IV. Inf. and VI. Ca y. Corps, out of a fortified position on both sides of Mizil, and on Dec. 15 took Buzeu. There the 89th Inf. Div., which had advanced south-eastward from the Bodza Pass into the Buzeu valley, joined Falkenhayn's army. The Danube Army on Dec. 14 won the way through the Jalomitsa sector against the Russian VIII. Inf. Corps and III. Ca y. Corps. On Dec. 17 the two armies faced a Russo-Rumanian position running along the line from the lower course of the Calmatuciu by Foreivechii along the heights W. of Rimnicu-Sarat.

Meanwhile the Bulgarian III. Army had begun to clear the Dobruja and, meeting with little resistance, had soon reached the Danube estuary; turning towards the eastern bridge-head from Braila at Niacin it transferred the Turkish VI. Corps to the Danube Army.

At Christmas the IX. and the Danube Armies broke through the enemy positions, and threw the Russians and Rumanians back northwards. On Jan. 4 1917 the Danube Army captured Braila, and pressed forward as far as the Sereth; on Jan. 8 the IX. Army took Focsani and the country N. of it as far as the Putna. On the S. wing of the army front the Archduke Joseph launched to the attack against the Russians in the last days of Dec. the Gerok group with the 218th Inf. Div., the 1st Ca y. Div., the 71st Inf. Div. and the 187th Inf. Div. The S. wing fought their way through the extensive wooded mountain district, and effected a junction with the IX. Army. South of the Ojtoz road the attack came upon the Russians who were relieving the Rumanians. The Rumanian 15th Inf. Div. was again thrown into the action. The attack of the Gerok group only succeeded in winning a little more ground here. Fall of snow and sharp frost made further operations impossible.

The actual line won was fairly near the sector it had been intended to reach. On the side of the Central Powers it was decided to go into permanent positions here. There was reason to be satisfied with the success of the campaign. Transylvania was liberated; a country rich in resources, Wallachia, had been conquered; the Rumanian Army had been thoroughly beaten, and had for the most part ceased to be a factor in the fighting for a long time to come. The Russian Army, instead of giving the hoped-for support, had had in addition to take over another 4 00 km. of front. The Russians, too, were glad after a year too full of fighting to be able to rest. Besides the IX. Army (16th Inf. Div. and 2nd Ca y. Div.), extending from the Bukovina to the Casinu valley south-eastward of Ocna, the Russian new IV. Army stood here from Racosa on the Susita to Suraia E. of Focsani (6th Inf. Div.); from there eastwards to the Black Sea the VI. Army (92 inf. divs. and 3 ca y. divs.). Of the Rumanians only from 5 to 6 divs., reinforced by Russian troops, remained on the front as the Rumanian II. Army, between the Russian IX. and IV. Armies. The remnant of the Rumanian Army, saved with difficulty, was transferred to the district between Jassy and Targu Frumos to recuperate. A French mil itary mission undertook to reorganize the army, and to give it a thorough education based on the principles of the conduct of modern warfare. This task it had finished by the summer.

In the spring of 1917 events took place of the most far-reaching significance for the conduct of the war in the East: the deposition of the Tsar, the outbreak of the revolution in Russia and the beginning of the collapse of the Russian army. As on all parts of the eastern front, so in Rumania, the Russian infantry had no more desire for fighting; the Russian artillery, left to carry on alone, were threatened by the infantry; indeed it came to regular battles between the two arms. It was only with difficulty that the numerous officers of the Western Powers distributed among the higher commands could prevent the collapse of the eastern front. The fighting value of the Russians did indeed improve a the time of the Kerensky offensive of June 1917, but the improvement was not a lasting one. The Rumanian troops remained untouched by all these happenings. Indeed it seemed as if Rumania's fighting strength increased in proportion as her ally became less reliable.

In the second half of July the reorganized Rumanian I. Army was placed between the Russian IV. and VI. Armies from a point E. of Nemoloa§a to S. of Tecuciu on the Sereth front.

In connexion with the operations in East Galicia the Central Powers intended to strike a decisive blow against the Russians and Rumanians in Rumania, in order to shake the whole Carpathian front and if possible to gain Moldavia. The operation planned across the Sereth at Nemoloap was to begin in August. The preparations for this were in train when the Rumanians anticipated the attack.

On July 25 the Rumanian II. Army, with the IV. and II. Corps, and the Russian VIII. Corps on the N. wing of the Russian IV. Army, broke through the weak front of the 218th Inf. Div. and the 1st Ca y. Div. and threw them far beyond Soveja back into the mountains, the wing division of the IX. Army being thereby surrounded on the N. and N.W. by the Russian VIII. Corps. Even though there was little need to fear Rumanian advance against Kezdivasarhely in the rear of the I. Army, on account of the width and impassable nature of the mountains, there was all the more danger that, after the capture of the Mt. Odobeshti (Odobe§ti), the whole front of the IX. Army, which covered the sphere of the earlier Danube Army and was commanded by Gen. Kosch, might be rolled up from the north. This was obviously the intention of the Rumanians and Russians, but the troops in carrying out the operation did not strike hard enough. Precious time was thereby lost. On account of the want of roads direct support of the 218th Inf. Div. was hardly possible. It was only slowly that one regiment of the 117th Inf. Div., and then half the 37th Honved Inf. Div., which had been set free from the N. wing of the I. Army, could be brought up. The 217th Inf. Div. was supported by single regiments and battalions of 5 different divs. of the IX. Army, and the attack was thus barred.

In the counter-operation planned by the Central Powers it was intended to take up again the original plan of penetrating far into Moldavia. For this purpose the IX. Army was to conduct the main attack from Foc*ani W. of the Sereth in the direction of Ajudu Nuou, and simultaneously to cover this attack by the construction of a bridge-head on the E. bank of the Sereth in the direction of Tecuciu. A second push was to be delivered by the Gerok group from the Ojtoz valley on Onesci. By this means the Rumanian II. Army, which had advanced into the basin of the Soveja, was to be cut off.

For the attack which was to start from Focpni the following were placed in readiness under the command of Lt.-Gen. von Morgen (I. Res. Corps): the 12th Bavarian Inf. Div., 76th Res. Inf. Div., and the 89th Inf. Div., to be followed in second line by the 216th Inf. Div. As army reserve there stood at Focpni the 212th and 115th Inf. Divs. On Aug. 6 the attack began, and had indeed the desired success on the first day in a N.W.

direction. The attempt to cross to the E. bank of the Sereth, however, failed.

The Russian Corps which were attacked (the VII. and behind that the XXX.) put up a surprisingly obstinate defence. It was only after throwing in the army reserves that the German I. Res. Corps succeeded in overrunning the Susita sector. Moreover, the 5th and 9th Rumanian Divs. of the Rumanian I. Army also came forward to face the attacking Germans, and caused considerable delay, especially at Marasesti (Marasheshti), by their violent, deeply echeloned counter-attacks.

On Aug. ro the VIII. Corps with 3 (partly combined) divs. reinforced the attack of the Gerok group on both sides of the Ojtoz valley. They attacked the Rumanian IV. Corps (6th and 7th Inf. Divs.), and gained ground as far as just S. of Ocna and Grozesci. But on account of the obstinate resistance of the Rumanians the objective, Onesci, could not be reached.

Left of the I. Res. Corps the XVIII. Res. Corps, reinforced by the Alpine Corps, once more in action, had meanwhile joined in the attack with their left wing, and after heavy engagements had taken Panciu N. of the Susita. On Aug. 15 the S. wing of the Gerok group (218th Inf. Div. and sections of the 117th Inf. Div., half the 37th Honved Inf. Div. and the 8th Mountain Bde.) and the 217th Inf. Div., standing on the left wing of the XVIII. Res. Corps, also joined the attack and slowly drove the Rumanians out of the basin of the Soveja.

A bridge-head on the W. bank of the Sereth threatening the German flank, held by the Rumanian 5th Div., was stormed by the 216th Inf. Div. of the I. Res. Corps on Aug. 14, severe losses being inflicted on the Rumanians. The further attempts of the I. Res. Corps, under which was placed the newly arrived 13th Rifle Div., to advance over the line Marasesti - Panciu, failed through Russian and Rumanian counter-attacks.

In consequence of the events in East Galicia and in the Bukovina, where the Russians were driven back to the old boundary of the Empire, a regrouping of troops and new distribution of the armies in Moldavia was effected. The troops of the Russian IV. Army were withdrawn to the N. to the VII. Corps, and the Russian IV. Army Command took over from the IX. Army Command the sector on the Transylvanian E. front as far as the Slanic valley. The Rumanian I. Army also took over the sector held earlier by the Russian I. Army, so that the two Rumanian armies now stood side by side.

On Aug. 28 the XVIII. Res. Corps, with the 216th Inf. Div. and the Alpine Corps, attacked from the line Panciu - N. edge of the Mt. Odobeshti in a N.W. direction, to gain the upper course of the Susita. After stubborn engagements lasting for many days against the Rumanian II. Corps, Jresci and the heights S. of the Susita were captured, upon which practically the old line, as it stood before the Rumanian attack, was reached. On Sept. 3 attacks from the German side were again suspended.

At the beginning of Sept. the Rumanians with the IV. Corps conducted a series of violent attacks against the advanced positions of the VIII. Corps, especially against the 225th Inf. Div. standing just S. of Ocna, but they were bloodily repulsed.

On the side of the Central Powers, after this unsuccessful enterprise, the troops which could be spared (the Alpine Corps, the 13th Rifle Div., the 1 r 7th Inf. Div. and much heavy artillery) were withdrawn for transfer to other theatres of war. The remaining units again went into permanent positions. On the Rumanian side the fruitless attacks ceased. They had suffered heavy losses in killed and wounded, and important loss in prisoners and material. The newly formed Rumanian divs., instructed by the French, had succeeded in defending their country from complete conquest. The battle of Marasesti, as it was called by the Rumanians, is the most famous page of the Rumanian Army in the World War.

On Dec. 5 the commander-in-chief of the Russian S.W. front, Gen. Shtcherbachev, asked for an armistice. On Dec. 7 the negotiations began at Focgani under the presidency of Lt.-Gen. von Morgen; representatives of all the participating armies took part, and they were concluded on Dec. Io.. (R. K.)