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 Mackensen's two armies on June 16-19 (seeLEMBERG, BATTLES ROUND, Sectionll.) and the successful two- fronted battles of Linsingen's South Army about Stryj and Drohobycz, in which his left, facing north, held off the counter- attacks of the Russian XI. Army, while his right, by intervening in the flank of Pflanzer-Baltin's opponents (IX. Army, Let- chitsky), made them retreat to the Dniester (May 3 i-June 15).
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 dissolu- tion, 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 Tanev and the region of Rawa Ruska, the retreat took a northerly direction and thtfs 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, un- changed, 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, with- out 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 geo- graphical barriers, 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, to- ward Vladimir Volhynskiy, E. of the Bug. Only the South Army and the VII. Army would remain to drive the Russians remaining S.'of Brody outofE. 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 Woyrsch extending, in place of it, to the Vistula and reconstituted it about Rawa 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 evacu- ation 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 reenforce 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 ti 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
