Colonel Devin detached a squadron to Ellis' and Richards' Fords, accompanied by Captain A. Moore, aide-de-camp, who surprised the pickets of the enemy at Richards' Ford, capturing officers and some 35 enlisted men.
On reaching the Rapidan, at 5 p.m., two squadrons of cavalry were pushed across, which soon drove off the few mounted men of the enemy who were keeping a lookout for our approach. The cavalry was immediately followed by Griffin's and Sykes' divisions, which forded the stream, notwithstanding it was 3 feet deep, with a very rapid current. Both divisions were bivouacked on the right bank of the Rapidan by midnight.
The Third Division (Humphreys') had been left at the Rappahannock, to cover the passage of our trains and assist in taking up the bridges. This duty was not accomplished until late in the night, at which time Humphreys commenced his march to rejoin the command, but, owing to the darkness of the night and the ignorance of his guides, he lost his way, and did not reach the Rapidan until after the departure of the column.
Before daylight on the 30th, two squadrons of cavalry were sent out, with instructions for one to push on in the direction of Chancellorsville, the other to take the road to the United States Ford, after crossing Hunting Creek; both to drive in the enemy's pickets, if encountered and pursue them as far as they could, reporting results. Just as the column of infantry was ready to move, I received a report from Colonel Devin that he had driven in the enemy's pickets on the United States Ford road, and had pursued them for several miles until he had encountered the enemy in force drawn up in line of battle to the number, he should think, of at least a brigade. Having no reason at the time to doubt the authenticity of this report, I concluded the enemy had become apprised of our movement and were prepared to dispute the opening of the United States Ford, a point of vital importance to our operations. I therefore ordered Sykes' division, in advance, to proceed at once to the United States Ford, and halted Griffin, after passing Hunting Creek, to await the development of Sykes' movement.
Soon after making these dispositions, the cavalry reported their occupation of Chancellorsville, after a slight skirmish with a small force of the enemy. Being satisfied from this that there could be no enemy at the United States Ford, I pushed Griffin on to Chancellorsville, which place he occupied at about 11 a.m., and where I was rejoined by Sykes about 1 p.m. As soon as Chancellorsville was occupied, I directed Colonel Devin with his whole cavalry force to send out a strong picket on the Plank road, and to send another party out on the Banks' Ford or River road, to feel for the enemy, and ascertain how much of our front was uncovered.
About 3 p.m. I received a report from Colonel Devin that he had driven in the enemy's pickets at the Banks' Ford road, and had pursued them until he could see their line of battle, and from the wagons visible he concluded they were about to evacuate the position. I immediately directed General Griffin to advance a brigade to support the cavalry, and, if practicable, to drive in the infantry and uncover Banks' Ford.
About an hour afterward, Griffin reported to me that he was with his brigade in the presence of a superior force of the enemy, and that he would require support if he had to maintain his position. I then ascertained, to my surprise, that the cavalry, instead of going down the Banks' Ford road, as they reported and led me to believe, had gone