Julv 3, 1915. 
LAND AND i^ATER. 
advanced, as his further bringing forward 
of heavy munitions continued, and the next blovy 
at Lemberg was delivered against the 
Upper Dniester, notably at Zurawnow. That 
blow failed. The Russians again held their 
line, while they were acting with the railway 
junction of Lemberg exactly as they had 
acted with the salient of Przemysl. Not until 
everything had been withdrawn, and not until no 
gun or box of stores or railway wagon or loco- 
motive could fall into the enemy's hands did they 
further withdraw. They are now pursuing pre- 
cisely the same method, covering for the moment 
the junctions of Tarnopol and of Busk; though 
we have said that the covering of these two im- 
portant junctions will not be permanent, for the 
enemy has still a railway system to serve him and 
to bring up his still remaining superiority of 
munition. 
Now all this retirement before a superiority 
of concentrated fire, proceeding as it does, 
with regularity, deliberation, and upon a set 
plan, never (save at one moment, on the 
Dunajetz, at the beginning) suffering from con- 
fusion, and — one may say it without exaggera- 
tion — undertaken as to each step, though not as to 
the whole movement, consonant to the Russian, 
and not to the Austro-German will — never forced 
— has had a second characteristic which general 
opinion has largely missed. The Russian retire- 
ment has throughout, or at least since the Wislok, 
preserved full contact with the enemy ; and to pre- 
serve full contact in such case means that you both 
desire to and can make your advancing enemy 
lose as many men as possible. 
This rather technical point gains its mean- 
ing from the fact that a retreat deliberately 
undertaken may always, or nearly always, hrecui 
contact with the enemy if it chooses. 
Of this we see an excellent example in the 
German retreat through Russian Poland last 
October. 
It takes some considerable number of hours to 
organise a large army for any particular move- 
ment. A commander, therefore, who decides to 
fall back, and who is only concerned with saving 
his army can usually, if he chooses, cover his 
retirement with comparatively thin rearguards, 
and keep the mass of his force unmolested. 
It is self-evident that there must be many ex- 
ceptions to such a rule, and history provides us 
with many. The great retreat from Mons and 
the Sambre is acutely present as a modern in- 
stance. But as a general rule, a commander who 
has undertaken to retreat can take his choice 
between retreating disentangled from his enemy 
and retreating in contact with his enemy — using 
the word contact to mean the contact of the 
masses, and not only of rearguards, which will, of 
course, always exist under any circumstances. 
The Russian retreat before Napoleon in 1812 is a 
clear instance of the former. 
Now if a commander chooses the second of 
these two alternatives, particularly if he chooses 
it during a very slow, methodical, and deliberately 
organised retirement, it always means two things. 
It means, in the first place, that he is the 
master of his own actions and that his retreat is 
conducted at his own pace and without peril. 
It means, in the second place, that his object" 
is to inflict the greatest possible loss on the enemy 
in the course of that retreat. I do not think 1 
am misjudging the Russian effort, nor taking 
desires for realities, when I say that so very 
gradual a retirement remaining throughout fully; 
and (if one may use the word) " densely " in con- 
tact with the corresponding enemy advance, was 
based upon a deliberate calculation to inflict a 
maximum loss upon that enemy advance, and has 
succeeded in inflicting it. 
Consider mere distance and see how true this 
is. In the first rush, under the first overwhelming 
superiority of the enemy heavy artillery fire, the 
whole Russian line falls back some eighty-five 
miles in less than a fortnight. Then it holds its 
line for a full fortnight more, until Przemysl is 
evacuated. Then, on the evacuation of Przemysl, 
it falls back not more than some fifteen miles and 
holds the line of the Dniester and positions in 
front of the Grodek line, and holds these without 
peril for three weeks. Then, after a complete and 
successful evacuation of all material from Lem- 
berg, it falls back behind Lemberg. still maintain- 
ing itself upon the Dniester to the south. That is, 
it allows its central and northern portion to fall 
back about sixteen miles, but keeps its southern 
half still well thrust forward. Note this care- 
fully, for it is a proof of the ease with which the 
Russian retirement proceeds. The Russian line 
for nearly a week after the evacuation of Lemberg 
is in an S like this. If it were in any haste or 
Lemberg* 
\ 
-^Ax- 
M 
^:rS' 
>M^ 
\ 
''i. 
<v4 
\ 
w 
\ 
i 
^ 
\ 
^F 
HI 
confusion such a scheme would be perfectly im-< 
possible. It would be clearly exposed to disaster 
between the points A and B south of Lemberg. 
But because there was no confusion or any peril iti 
held that odd salient at C just as long as sufficed 
to inflict the maximum of losses upon the enemy as 
he tried to force the Dniester at D D. Then, and 
then only, did it fall back and straighten the 
whole line out in the direction B-E. 
One may sum up and say that so far the Rus- 
sian retreat has, since the rally on May 13-16 upon 
the San, betrayed a certain motive and accom- 
plished a certain object. The motive was the 
saving of an army from superior fire and the pre- 
serving of its line intact. The object was the 
infliction of the maximum of loss upon the enemy, 
during the retirement. The line has been main- 
tained, the loss has been inflicted. 
,What further task lies before the enemy ? He 
set out to divide the Russian army into two or 
more portions, which he could deal with sepa- 
rately. In the last two days of April and the first 
5» 
