A remarkable feature of these casualty lists is the wide variation at times from the usual proportionate number of killed to wounded.
This is due, quite often, to delay in making out the nominal list after the battle.
If the first sergeants hand in their company list of casualties promptly to the adjutant or colonel at the close of the action, there will, evidently, be less men reported as killed than if there is a delay of several days.
In the latter case, many will have a large proportion of the mortally wounded die within a few days after the battle, the ratio of the number of killed to the wounded would be changed considerably by delay in the reports.
In some actions and in some campaigns it was difficult to make prompt reports of casualties.
In some actions a division would be under arms for several days, momentarily expecting an attack.
The nature of the fighting also affected the ratio of the killed and wounded.
In a hot fight at close quarters, or in an assault, the proportion of killed is naturally large; at long range, or in the second line, or while engaged in “supporting battery,” the proportion of killed is less than the common ratio.
But casualty lists will fail to give an intelligent idea of the extent of the loss unless the number, or probable number, of men engaged is kept in mind.
The average American regiment of infantry in the last war, while in active service, numbered about 400 muskets; and, unless the number taken into action is definitely known, it will be safe to assume, in examining the casualties, that the number engaged was not far from that amount.
Newly organized regiments, fresh from their rendezvous, often took from 700 to 800 men into a battle; but, if their first battle did not occur until after several months of campaigning, they would take only about 500 men into action.
Then, there were regiments which became so depleted by battles, marches, campaigns, and disease, that they often went into battle with less than 200 effectives.
Some of these depleted commands were restored to an effective strength by accessions of recruits; or, by transfers of men from regiments returning home, these transferred men having unexpired terms of enlistment.
Even then, the regiment, thus recruited, would seldom number over 400 effectives.
From personal observations at the time, and subsequent studies of official returns, an effective strength of 400 appears to have been the most common.
In many of the instances specified in the subjoined table of greatest casualties, the number actually engaged will be found in the list of maximum percentages, pages 28-34.
The heavy artillery regiments have in some instances here been classed by themselves, their larger organizations requiring, in a fair statement of losses, that their casualties be kept separate from those of the small and depleted infantry commands.
These heavy artillery regiments were not called upon to take the field until the spring of 1864, their first experience under fire occurring at Spotsylvania
and Cold Harbor, in which actions each of these regiments had nearly 1,800 men engaged.
But the bloody vicissitudes of Grant
's campaign soon reduced these splendid commands to nothing but skeletons of their former selves.
The casualties in the cavalry are also given separately in these lists.
Their losses occurred mostly in cavalry battles,--cavalry fighting cavalry, with no infantry near.
In many cases hi le losses are not large enough to warrant classing them with the heavier losses of the infantry, and, so, they are given separately.
The cavalry losses in particular actions are not so remarkable as those of the infantry; but, the mounted regiments were in action so much oftener, that the aggregate of casualties in one of their campaigns, or raids, would equal those of an average infantry regiment.
The style of fighting which prevailed in the cavalry service during the Civil War
was new and peculiar.
The wooded countries in which they operated prevented any charges by large bodies of mounted troops.
The cavalry used their sabres but little; they fought dismounted,