If we had delayed a single day, our retirement might have resembled that of the 2nd and 3rd Armies, which were in so awkward a plight at Mukden. For the reasons explained in my third volume, the 2nd Army was, on March 7, almost surrounded on flanks and rear. Great efforts were necessary in order that we might extricate ourselves from the position in which we were placed without being utterly defeated. But these efforts were not made, and the situation of our whole force on March 7, 8, and 9 became worse, and the danger of a considerable part of the 2nd Army being surrounded by Nogi’s troops still more imminent. Comparing the condition of our men with that of the Japanese on March 7 and 8, as well as the positions occupied by the two forces on the 8th, and taking into account the moral superiority of the Japanese, I should have given up hope of a victorious issue from the battle on the 7th and 8th, and have arranged for a retirement to Tieh-ling before the army became disorganized. The future historian will probably accuse me of having held on too long. I did not give the order to retire till March 10, and according to events and the opinion of my staff, the order should have been given a day earlier. If we had retired on the 9th, the army would probably have fallen back in complete order without losing anything (except wounded); indeed, we might have taken with us a fairly large number of prisoners and captured guns and machine-guns. In my report upon the battle of Mukden to His Majesty the Tsar, I acknowledged that I was primarily responsible for our reverse, and admitted that I should have more accurately gauged the difference between the men of the two forces and the qualifications of the commanders, and that I should have been more careful in making my decisions. Hoping against hope to defeat the enemy, despite the disastrous operations of the 2nd Army, between March 2 and 7, I gave the order to retreat too late. I should have abandoned all hope of eventual victory at Mukden a day sooner than I did, and our withdrawal would have been effected in good order. Thus, the general conclusion regarding the battles of Liao-yang and Mukden could, in my opinion, be expressed as follows: If we had retired from Liao-yang a day later than we did, the result would have been much the same as at Mukden; if we had retired from Mukden a day sooner, the result would have been much the same as at Liao-yang.[28]