There ought to be no uncertainty in the reports of casualties. Yet, while the general result is clear, precision in detail is now hardly attainable. General Beauregard's report gives his loss as 1,728 killed, 8,012 wounded, and 959 missing; making an aggregate of 10,699. Of the reported missing, many were killed or wounded. These numbers are the aggregate of losses reported by brigades. They cannot include casualties at division, corps, or army headquarters, happening either to the generals commanding, or to the officers on their staff, or to enlisted men on duty there. And while batteries were attached to brigades, the cavalry was a wholly independent command, not attached or reporting to brigades or divisions; two regiments were not attached to any corps. Their casualties cannot be included in brigade reports. Colonel Johnston, after much examination, "finds a possible variation of 218 more casualties, principally in missing, that might be added to General Beauregard's report." The generally accepted official report of the National loss is: in Grant's army, 1,437 killed, 5,679 wounded, and 2,934 missing, making a total of 10,050; in Buell's army, 263 killed, 1,816 wounded, and 88 missing--making a total of 2,167. The two armies aggregated 1,700 killed, 7,495 wounded, and 3,022 captured--making total, 12,217. The War Department, in the printed collection of battle reports, does not give the casualties of the two armies separately, but gives the aggregate, 1,574 killed, 7,795 wounded, and 2,794 missing--making a total of 12,163. The "Medical and Surgical History of the War" makes the loss 1,735 killed, 7,882 wounded, 3,956 missing--making a total of 13,573. The loss of the Army of the Ohio, as given above, is the report of General Buell on April 15th. Six days later, the Medical Director of that army made to General Buell a tabulated statement of killed and wounded in each regiment, brigade, and division engaged, which makes the number 236 killed and 1,728 wounded. All these estimates are based upon the same material--upon the field reports. As the revisers of the reports for publication have had the best opportunity for deliberate examination and for comparison of the reports with muster-rolls, their estimate of casualties is perhaps the most trustworthy.