The topography of the vicinity of Knoxville may be briefly described as follows.

On the north bank of the river a narrow ridge is found, extending from a point about 2 1/2 miles east of Knoxville to Lenoir's. It has an average base of about 1 1/2 miles in width. At Knoxville the width is about 1 mile. The ridge is cut through at short intervals by small streams; two of which, First and Second Creeks, run through the town of Knoxville, at a distance from each other of about threefourths of a mile. The main part of the town is built upon that portion of the ridge bounded on the northwest by the valley, on the southwest by Second Creek, on the southeast by the Holston River, and on the northeast by First Creek. It has the appearance of a table, elevated about 150 feet above the river, and about 100 feet above the valley. Again, Third Creek is found about seven-eighths of a mile below Second Creek, forming a second similar table. A depression in the ridge, about the same distance east of First Creek, forms still another table, upon which is built East Knoxville. This elevated ground is called Temperance Hill. Eastward from this the ridge is more broken, until it disappears and other ridges spring up.

This last division is known as Mabry's Hill, and is the highest ground, by some 20 feet, to be found on the north side of the river, within cannon range of Knoxville. Commencing at Third Creek and going eastward, these tables may be numbered 1,2,3, and 4. A succession of ridges, all parallel to the one alluded to above, are found at short distances apart as one goes back from the river, the most important of which is at an average distance of 1 mile from the one upon which Knoxville is situated, and is that which the enemy occupied. As near as can be ascertained by a pocket level, its elevation is about 15 feet more that of table No. 1 at its highest point.

In selecting lines it was my opinion that tables Nos.1,2,3,and 4 should be occupied. In this opinion the general commanding concurred, except regarding No. 4. It was a question whether it was most prudent, with our small force, to attempt to occupy No. 4, or to limit our occupation in force to Nos.1,2, and 3, simply holding No. 4 with a strong out-post. It seemed to me that we could (as we did) construct lines on No. 4. and also on No. 3; those on No. 4 to be thrown so far to the front that they could not be made available against us in the event of their capture by the enemy, and to be held by us with obstinacy, only leaving them when absolutely compelled to do so; in which case the lines on No. 3 would form a defense, and the enemy, after a stubborn fight, in which he would have suffered, would only have occupied a position, from which, in the former case, it would have been necessary to dislodge a mere out-post. In other words, after crippling him all we could in his successful assault upon No. 4, we would still have just as good a defense as if we had occupied only No,3 int he first place.

But, on the other hand, by extending our lines to include No. 4, we made them weak in numbers at any one point, and we were consequently exposed to greater danger of being successfully assaulted along our whole position. This difficulty was partially obviated by the successful construction of a dam across First Creek, by which an impassable water obstacle of three-fourths of a mile in length was made immediately in front of and parallel to the crest of No. 3, which enabled us to greatly weaken our numbers (of infantry) in that part of the line.