other with Central Kentucky, where the occupation of Kirby Smith had established for him a second base.

Munfrodville did not offer the same advantages, for, although much stronger natural position, yet the same in taking it he gave up his communications with Tennessee, and rendered those with Kirby Smith less secure against a force operating from the Ohio River, supposing Louisville to be secure to us. At Munfordville his communication with Kirby Smith must have been along the Louisville turnpike and thence across to Bardstown, while at Glasgow it would be along the old Lexington road through Summersville and Lebanon, or through Columbia and Lebanon or Liberty, by all of which roads I have moved large bodies of troops. Besides, at Munfordville he would have been in a much less productive region than at Glasgow. These considerations taken in connection with the risk he would run by advancing farther into Kentucky, made it at least reasonably doubtful whether he would not halt at Glasgow. The fact that hits his purpose was to penetrate still farther into Kentucky and that he had designs on Louisville was only known only known when it was ascertained that he had left Glasgow and through correspondence which was captured subsequent to that time. But supposing it had been reasonably certain that Bardstown was his destination, it was not to be assumed that he would go by the way of Munfordville; on the contrary, it is undoubtedly true that but for the bait which was offered to him in the garrison at that place he would not have gone there at all, for the simple reason that without any object whatever it would have taken him off the direct and excellent turnpike from Glasgow to Bardstown, and thrown him on another road not so good and 12 miles longer. This brings me naturally to the question of the relief of Munfordville.

The foregoing explanations show that I could not have reached Munfordville in advance of the rebel force even if it had been desirable to leave it between me and Nashville. The first information received at Bowling Green that Munfordville was attacked or threatened was on the 14th, and the report was that it had been captured, though that was not certain. On the same day the last of five of my divisions arrived at Bowling Green, and on the same day, as was afterward ascertained, the main body of the rebel army marched from Glasgow, 18 miles from Munfordville, with the roads. If I had moved forward at once I could not have reached Munfordville in less than four days, for, considering that I must march on one road, it would, for the rear of my column, have been equivalent in time to a march of 60 miles; and in the presence of an enemy whose position was not know the march could not well have a been made more rapidly. As for re-enforcing the garrison by the first of my divisions which arrived at Bowling Green, even if the necessity of it could have been known it would have been out of the question, for those divisions would have been thrown into the midst of the whole rebel force; a folly which it appears the enemy actually anticipated, and prepared to reap the fruit of.

But I propose to inquire also what necessity there was for such relief, and on what grounds it could reasonably have been expected that I would furnish it. It is apparent from a study of the map, and the evidence shows, that the possession of Munfordville was not essential to Bragg's army in a strategical point of view. At least three other preferable routes were open to him, whether his object was to attack Louisville directly or to advance into Central Kentucky for other purposes: First, the shorter and better road from Glasgow to Bordstown and thence on to Louisville; second, the old Lexington road to Leb