division was placed in supporting distance, one regiment being stationed in the rifle trenches on the south bank east of the railroad. A gun from the works on the left of the road was also ordered to be placed in the battery at this point, to command the approach by the railroad embankment on the opposite side, but the enemy's sharpshooters had advanced so near the river that the order was countermanded, the preparations already made being deemed sufficient.

The enemy placed three batteries on the hills from which our skirmishers had been forced to retire, and maintained an active fire upon our position until dark, doing no damage, however, so far as has been reported. Our batteries replied from both sides of the river, but with so little effect that the two on the south bank were ordered to cease firing.

Light skirmishing took place along the line. It was not known whether this demonstration was intended as a serious attack or only to cover the movement of the force that had crossed at Kelly's Ford, but the lateness of the hour and the increasing darkness induced the belief that moving would be attempted until morning. It was believed that our troops on the north side would be able to maintain their position if attacked, and that in any case they could withdraw under cover of the guns on the south, the location of the pontoon bridge being beyond the reach of a direct fire from any position occupied by the enemy.

As soon, however, as it became dark enough to conceal his movements the enemy advanced in overwhelming numbers against our rifle trenches, and succeeded in carrying them in the manner described in the reports of Generals Early and Hays. It would appear from these reports and the short duration of the firing, that the enemy was enabled to approach very near the works before being seen.

The valley in our front aided in concealing his advance from view and a strong wind effectually prevented any movement from being heard. It was essential to the maintenance of the position under these circumstances that sharpshooters should have been thrown forward to give early information of his approach, in order that he might be subjected to a fire as long as possible, but it is not stated that this precaution was taken. The breaking of the enemy's first line and the surrender of part of it, as described in the reports, also contributed to divert attention from the approach of the second and third, and enable them to press into the works. No information of the attack was received on the south of the river until too late for the artillery there stationed to aid in repelling it; and it does not appear that the result would have been affected, under the circumstances, by the presence of a large number of guns.

The artillery in the works at the south end of the bridge was relied upon to keep it open for the retreat of the troops, as it could sweep the crest of the opposite hill at a short range. The darkness of the night and the fear of injuring our own men who had surrendered, prevented General Early from using it. The bridge, however, seems to have remained accessible to the troops on the left up to the last moment, as Lieutenant-Colonel Tate, with a few men, crossed just before it was fired by order of General Early.

The suggestions above mentioned afford the only explanation I am able to give of this unfortunate affair, as the courage and good conduct of the troops engaged have been too often tried to admit of question.

The loss of this position made it necessary to abandon the design of