previous next

Relative numbers at Gettysburg.

[We had expected ere this to have finished our Gettysburg series, but we are sure that our readers will be glad to have the two papers which follow on the numbers of the armies at that great battle — the second letter of our distinguished correspondent, the Count of Paris, and the able, exhaustive and conclusive paper of General Early, which seems to us to settle the question beyond all controversy.]

Letter from the count of Paris.

Chateau D'Eu, Seine Inferieure, March 23d, 1878.
Rev. J. William Jones, Secretary Southern Historical Society:
With the permission of the Adjutant-General of the United

States army, General Humphreys has kindly furnished me with a complete and authentic copy of the monthly return of the Army of Northern Virginia for the 31st of May, 1863. The inspection of that document settles at once the difficulties which I met with in the evaluation of the effective strength of Lee's army at Gettysburg, and which I had submitted to you. It explains the difference between Colonel Taylor's figures — which embraced only the enlisted men present for duty — and that given by General Humphreys, which comprises both officers and men present for duty. As the Federal reports always reckon the officers with the men, whenever a comparison is to be made between the forces of both armies it is the latter system which should be adopted. An error of nine in the aggregate of Rodes' division having been corrected by me, there is the same difference between the figures I give here and those of the original return. As some of these figures have been published, both by Mr. Swinton and by Colonel Taylor, but without the necessary explanations for their intelligence, I think it is no breach of confidence to give these figures and a few others with the required explanations:

On the 31st of May, the Army of Northern Virginia numbered 133,680 officers and men and 206 guns. Out of these 44,935 were absent and 88,745 present; the latter figure embraces 7,387 officers and men sick, 5,951 on extra duty and 948 in arrest. Lastly, there were present for duty 6,116 officers and 68,343 men, or, in the whole, 74,459. The division of this effective force between the different arms was as follows: General staff, 47; infantry, 69,418; cavalry, 10,292; artillery, 4,702. During the month of June this [11] force must have been increased somewhat by the regular operations of the draft, and by the return, both of sick men restored to health under the genial influence of the season and of the men recovering from slight wounds received a month before at Chancellorsville. If that increase is difficult to appreciate, there is another element which can be easily calculated — it is the reunion of three brigades which do not appear on the return for the 31st of May. These brigades were--first, Pettigrew's, nearly 4,000 men strong (before leaving in Virginia one of its five regiments); second, Jenkins' cavalry, and third, Imboden's mixed command, numbering together more than 2,500 men.

On the other hand the effective strength of the army was reduced by the three following causes: first, detachments; second, losses in fights; third, sickness, straggling and desertion. First, detachments: Corse's brigade of Pickett's division and one regiment of Pettigrew's brigade (about 800 strong) were sent to Hanover Junction (Virginia), and later Early left one regiment to escort the prisoners from Winchester, and two others to occupy that town. These forces can be reckoned at 3,500 men. Second, losses in fights: the losses at Fleetwood, Winchester, Middleburg, Upperville and Hanover (Pennsylvania) were 1,400. Third, sickness, straggling and desertion: the reduction of the army through these causes must have been very small. The marches of the army were in average neither excessive nor continuous; the weather was fine; the roads in good order; and I have the best authority to believe that Pettigrew's brigade, by example, which was less accustomed to hard marching than the rest of the army, reached Pensylvania with at least as many men present for duty as when it crossed the Rapidan. Early's division had some of the hardest marching before it reached the Potomac, and therefore it can be taken as a fair standard of comparison. Thanks to General Early we have the elements for that comparison. On the 31st of May his division, which was the smallest but two of the army, numbered 6,943 officers and men present for duty; on the 20th of June (see Southern Magazine, September, 1872, page 318, foot-note) this figure has dwindled down to 5,638. The difference is 1,305, but that decrease must be ascribed altogether to the three above mentioned causes, viz: first, the detachment of three regiments, left at or about Winchester, at least 850; second, the loss in battle at Winchester, 162; third, therefore the reduction by sickness, straggling and desertion is only 293, unless the division should [12] have received individual accessions between the 1st and the 20th of June. At the latter date the sick present were 343. It cannot be supposed that when General Early started he dragged his sick men behind the division; therefore these 243 must have become sick during the march, and, as this number is superior to the whole reduction, it will be admitted that the division had been somehow recruited after the 1st of June; but for the sake of simplicity, I shall take in the whole, both of the possible increase by the draft and the return of sick and wounded soldiers, and of the reduction by sickness, straggling and desertion, and consider only the difference between the two. That difference I have shown to be for Early's division 293, or less than four per cent. The proportion for the whole army could not be quite as large, and therefore should not be reckoned at more than 2,600. In that case the reduction by the three above mentioned causes would be 7,500; the increase by addition of three brigades, 6,500, and therefore the net decrease, 1,000, leaving the effective force under Lee in Pennsylvania and Maryland the 1st of July at 73,500 men. If we deduct the cavalry on both sides, we can say that the Southern general fought with 62,000 or 63,000 men and 190 guns the 80,000 or 82,000 men and 300 guns with whom Meade encountered him at Gettysburg.

Excuse the length of this, and believe me, dear sir, yours truly,

P. S.--Here is the calculation to which I allude in the last sentence: Effective force of Stuart, May 31st, 10,292+Jenkins' and Imboden's cavalry, 2,200==12,500; minus losses in fights, 1,200, and other losses, 200; remains 11,100. 73,500-11,100==62,400. To be deducted also 16 guns with Stuart on one side, and 27 with Pleasonton on the other.

General Early's reply to the count of Paris.

The Remarks on the numerical strength of both armies at Gettysburg, by the Comte de Paris, published in the April number of the Southern Historical Society Papers, contain some very serious errors which it becomes necessary to notice.

The first error which I will examine is contained in the following passage: “The total is the figure which is generally given in both armies where only one is given, the number of men on detached [13] service being liable to vary greatly from day to day.” By “detached service,” he evidently means “extra or daily duty,” which is a very different thing from detached service. With this understanding as to his meaning, his remark, that the number of men on such duty varied greatly from day to day, does not apply to the Confederate army. At a very early day it had been found injudicious and unsafe to employ negroes as teamsters and laborers for the army when it was in an active campaign, and when the conscript act became a law, and all able-bodied white men were made liable to military duty, it of course became necessary to detail from the ranks all the teamsters and laboring men required. The number of these was very considerable, in order to furnish drivers for the baggage and supply trains, as well as such men as were required for manual labor in the several staff departments; and the details were permanent and of course not liable to vary from day to day. It was owing to this fact that the number of men reported on extra duty in the Confederate army greatly exceeded, in proportion to strength, that reported on extra or daily duty with the Federal army. With the latter the men on extra or daily duty might be made available for a fight, whereas in the Confederate army the teamsters, whose presence with their teams was always necessary, were no more available in a fight than the mules they drove.

The next errors to be noticed are found in the following passage: “Through the operations of the draft the effective strength of each regiment had been increased after Chancellorsville. The regiments had received some recruits between the 15th and the 31st of May; some more came between the 10th and 1st of June. Von Borcke says that the regiments of cavalry were largely increased in that way, but I am not satisfied by such vague statements, and in order to prove the fact I propose to calculate the average strength of the regiments from the known strength of several corps, divisions or brigades a few days before the battle, as stated by reliable authorities, and mostly by official reports.”

The assumption that our army was increased in strength after Chancellorsville through the operation of the draft, or by recruits in any way, is without the slightest foundation in fact. Major Von Borcke's sketches are not at hand to refer to, but if he has made the remark attributed it to him, he is wholly mistaken. It is very far from my purpose to say anything in the slightest degree disparaging to that chivalrous foreigner, whose sympathy for our cause and gallant deeds in its defence have given him a place in the [14] heart of every true Confederate; but it did not come within his province to be familiar with the statistics of the army, or even of the cavalry with which he served. The cavalry was an arm of the service that was never recruited by conscripts, and in May, 1863, the only recruits that were obtainable from voluntary enlistment were the young men just arriving at the military age. As our cavalrymen had to furnish their own horses, and keep themselves mounted at their own expense, it was the practice to permit a large number to go to their homes during the winter and early spring months, for the purpose of recruiting their horses and obtaining new ones when they were dismounted. These men generally returned at the period for active operations, and in that way the cavalry was strengthened on the opening of a campaign. It is this fact, it is presumed, that Major Von Borcke refers to, or that led him into error if he has made the remark as broadly as the Comte de Paris states it. The opening of the cavalry operations prior to the Chancellorsville campaign, and that campaign had recalled to. the army all the available cavalrymen, and the returns of May 31st, must have shown the whole cavalry of the Army of Northern Virginia that was available for the approaching movement. If any raw recruits had been received after that time, they would have been worthless from the want of training and seasoning of the men as well as of their horses.

There is a very great misapprehension existing in the minds of persons outside of the Confederacy, and even among officers of the Confederate army, as to the number of men put into the army under the conscript law. In a report to the Secretary of War, dated the 30th of April, 1864, General John S. Preston, Superintendent of the Bureau of Conscription, says: “The results indicate this grave consideration for the government — that fresh material for the armies can no longer be estimated as an element of future calculation for their increase; and that necessity demands the invention of devices for keeping in the ranks the men now borne on the rolls.”

In a report made in February, 1865, General Preston gives a table showing the “number of conscripts enrolled and assigned to the army from camps of instructions since the act of Congress, April 16, 1862,” from which it appears that the whole number of men added to the army east of the Mississippi, in that way, up to that time, was 81,993, exclusive of some obtained under the operations of General Pillow in the States of Alabama and Mississippi. He estimates the number of volunteers who joined the army during [15] the same period, without passing through camps of instruction, at 72,292. Of course the greater number of these conscripts, as well as the volunteers, went into the army during the first year succeeding the passage of the conscript act; and hence there were very few to be obtained after the battle of Chancellorsville, and they consisted exclusively of men who had managed to evade the conscript officers, and the few arriving at the conscript age.

In a letter addressed by General Lee to the Secretary of War on the 11th of February, 1863, he says:

Sir — I think it very important to increase the strength of all our armies to the maximum by the opening of the next campaign. Details of officers and men have been sent from all the brigades of this army to collect deserters and absentees. By the return of last month, forwarded to the Department to-day, you will perceive that our strength is not much increased by the arrival of conscripts; only four hundred and twenty-one are reputed to have joined by enlistment, and two hundred and eighty-seven to have returned from desertion, making an aggregate of seven hundred and sixty-eight, whereas our loss by death, discharges and desertion amounts to eighteen hundred and seventy-eight. Now is the time to gather all our strength, and to prepare for the struggle which must take place in the next three months. I beg you to use every means in your power to fill up our ranks.

These documents are to be found in the final report of the Provost-Marshal General of the United States ( “messages and documents, War Department, part 3, 1865-‘66” ), pages 122,128 and 131, and were printed from the originals in the “Archive office.” I have my own official returns for the entire year 1863, being the office copies which were retained, and the return for January 31st, 1863, shows that 52 joined my division by enlistment during that month, being within less than one of one-eighth of the number received in the whole army for the same time. My return for February shows 45 received by enlistment for that month, while the loss by death, discharges and desertion was 305. My return for March shows 96 received by enlistment, while the loss by death, discharges and desertion was 231. There was no monthly return for April by reason of active hostilities progressing at the end of that month; and my next return for May 31st shows 60 received by enlistment for the two preceding months, while the loss by death during that period was 327, a considerable portion being in battle, and by discharges and desertions, 327 for the same period, making a total loss of 754. The next monthly return was for July 31st, and that shows 77 received by enlistment during the months of [16] June and July, they being received after the return from the Gettysburg campaign, and the loss by death for the same period was 344, being mostly in battle, and by discharges and desertion it was 160. So that the recruits by enlistment, during the whole period, from the 1st of January to the 1st of Agust, 1863, did not amount to half the loss by discharges and desertion, leaving that by death out of the question. Three of my brigades were from Virginia, North Carolina and Georgia, the States from which conscripts for the Army of Northern Virginia were principally received. My returns show that very nearly the whole of the recruits received in the division were from those States, the greater number being from North Carolina. So if recruits were received to any extent by the Army of Northern Virginia between the 31st of May, 1863, and the time that army crossed the Potomac, my division returns would indicate the fact. That they do not do so is very conclusive evidence that the assumption of the Comte de Paris is wholly unwarranted.1 [17]

His method of estimating our strength by taking my regiments as the average for the whole army is not warranted by the facts of the case. The return for May 20th, as given by Colonel Taylor, shows present for duty in the entire army at that date, in the infantry, 55,261 officers and men. This does not include the general officers and their staff. I had nineteen regiments in my division included in that return, and the number of officers and men present for duty, excluding general and staff officers, was 6,421. There were certainly one hundred and sixty-nine regiments and battalions present in the army at that time, and there may possibly have been another. The average for my nineteen regiments would be 338, and this multiplied by 169 gives 57,122, being in excess of the number actually present for duty 1,861. Multiply by 167, the number of regiments assumed by the Comte to have been at Gettysburg, and it gives 56,446, an excess of 1,185. But three of my smallest regiments were left behind in the Valley, and taking their joint strength (773 present for duty) on the 20th of May, there would be left 5,648, giving an average of 353 for the sixteen which were present on the 20th of May and also at Gettysburg. Multiply 353 by 169, and it gives 59,657, an excess of 4,396 above the number actually present; and by 167 and it gives 58,951, an excess of 3,690 above the number actually present. The return of May 31st, as now correctly given, shows present for duty in the infantry, including all officers, a total of 59,457. In my division there were present for duty at the same time 6,943, giving an average of 365 to the nineteen regiments present. This multiplied by 169 gives 61,685, an excess of 2,128 above the number actually present; and by 167 gives 60,955, an excess of 1,498 above the number actually present. Taking off 919, the strength of the regiments detatched, and the other sixteen had 6,024, giving them an [18] average of 376, which, being multiplied by 169, gives 63,544, an excess of 4,087 above the number actually present, and by 167 gives 62,792, an excess of 3,335 above the number actually present. One of my regiments, the Thirty-first Virginia, was absent and not embraced in the returns of May 20th and 31st, but had returned on the 3d of June, and was embraced in the returns of June the 10th and 20th; so I had only seventeen regiments at Gettysburg, instead of eighteen as the Comte supposes. The strength of the Thirty-first Virginia in present for duty, on the 20th of June, was 280. Adding the strength of this regiment to that of the sixteen present on the 31st of May, and counting the seventeen as eighteen, and the average thus obtained would be about that of the regiments for the whole army. I had a small battalion of two companies, but as two of my regiments wanted a company each, I have not counted it, and as that battalion was detached permanently before the 20th of June and is not embraced in the return of that date, its strength is included in the 919 deducted for the strength, on the 31st of May, of the regiments that were left behind. The Comte's mode therefore of estimating the strength of our infantry, by taking the average for my regiments as the average for the whole number, is not correct, though he arrives at very nearly its strength when it crossed the Potomac by mistaking the number of my regiments. I estimate that we had 169 regiments and battalions at Gettysburg, of which six were battalions, and I think there can be no doubt that that was the precise number of infantry organizations there, not including in them the battalion employed as a provost guard at army headquarters, and the battalion of two companies from my division employed in the same way at corps headquarters.

The Comte makes no allowance for decrease in our infantry after it crossed the Potomac, and hence he gives as its strength at Gettysburg what it probably was on crossing the Potomac. He is entirely mistaken in assuming that I had a battery attached to one of my brigades. This was not the case — I had a battalion of four batteries which accompanied my division, and that is to be counted with the artillery of the army. He is equally mistaken in saying that Imboden had a few hundred infantry with him. Imboden had had three regiments of infantry with him on an expedition into Northwestern Virginia in the spring, to wit: the Twenty-second Virginia of General Sam. Jones' command, the Twenty-fifth Virginia of Johnson's division, and the Thirty-first Virginia of my division, all of which had returned to their respective [19] commands. He had the Sixty-second Virginia regiment, called mounted infantry, but it was armed precisely like the rest of his command, which consisted of a regiment and a battalion of cavalry, with a battery attached.

The Comte arrives at the conclusion that we had at the battle 66,639 present for duty of all arms, of which 52,571 was infantry, 4,190 artillery and 9,878 cavalry, and a total present of 75,783. The discovery of the error made by Colonel Taylor and Mr. Swinton, in ommitting to count the officers present for duty on the 31st of May, shows that the total of officers and men present for duty at that date was 74,451, of which 6,099 were officers and 68,352 enlisted men. The officers include those of all grades, and among them were 935 chaplains, quartermasters, commissaries, surgeons, assistant surgeons, and ordnance and signal officers, who did not belong to the fighting department. As one brigade of five regiments that was counted in the returns of May 31st and three regiments of my division were left in Virginia, to replace which was another brigade of four regiments, two regiments that had been with Imboden, and perhaps two other regiments in Davis' newly formed brigade, it May be assumed that the number of men thus added was about the number in the brigade and regiments that were left behind — that is, 74,451 officers and men for duty may be assumed as the basis of the calculations to be made to arrive proximately at the strength of our army when it reached Gettysburg. Of course the difference between that number and 68,352 makes a considerable difference in the estimates. As we were going away from the section from which we could be reinforced, the idea of the Comte de Paris that conscripts were hurried on to overtake us and fill our ranks, is to be entirely discarded; the only real additions made to the army were the cavalry brigades of Jenkins and Imboden.

My own division was certainly as good a one as any in that army, and having been trained under Stonewall Jackson, it was as well enured to marching and the hardships of an active campaign as any. Whatever ratio of decrease, therefore, occurred in that division may safely be assumed as the ratio of decrease for the whole infantry of the army. No troops were detached from Hays' and Gordon's brigades, and no additions were made to them between the 31st of May and the 20th of June. They jointly numbered 4,016 for duty on the 31st of May, and 3,447 on the 20th of June, showing a loss of 569, of which 163 was for loss in action. [20] Their loss then from other causes than casualties in battle was a little over ten per cent. By an oversight in my article in the last December number of the Papers, the loss between the 10th and 20th of June was stated at twelve per cent., when it should have been ten per cent.

My return for June 20th showed 5,643 for duty, including five chaplains, and my return for July 10th at Hagerstown showed 4,144, giving a loss of 1,449, of which 1,181 was in battle, leaving a loss of 318, a little over five and a half per cent., from other causes than casualties in battle. My aggregate present on the 20th of June was 6,476, and on the 10th of July it was 4,791, being a loss of 1,685, from which the loss in battle being deducted leaves 504, or a loss of very nearly eight per cent. from other causes than casualties in battle on the aggregate present. The greater part of this doubtless resulted from leaving the sick behind, or sending them to the rear. As it took us only three days to march from Gettysburg to Hagerstown, at which latter place we arrived on the 7th, there had been time for all the men with the trains to join the division. In fact a return made on the 8th showed 261 less for duty, and 408 less in the aggregate present on that day than on the 10th. I may assume-therefore, that there was a loss of five and a half per cent. in my division from the 20th of June to the beginning of the battle, and that there was the same ratio of decrease in the rest of our infantry-during the same period. To show the likelihood of there being at least as much loss in Longstreet's and Hill's corps as in Ewell's, I quote from General Kershaw's report the following statement: “Tuesday, June 16th, the brigade marched to Sperryville; 17th, to Mud run in Fauquier county. These two days were excessively hot, and on the 17th many cases of sunstroke occurred.” General Hill started from the heights of Fredericksburg on the 15th, I believe, and his march had to be rapid to join Longstreet's corps, and hence the probability is that the loss in his corps exceeded the ratio in my division.

Take as the full strength of the infantry, May 31st59,457
Deduct for chaplains, quartermasters and other non-combatant officers786
Off ten per cent5,867
Probable strength of infantry on reaching the Potomac52,804
Deduct 5 1/2 per cent. after that time2,904
Probable strength of infantry at Gettysburg48,900
Add for cavalry6,000
For artillery4,000
Probable strength in all arms at the battle59,900

Major McClellan, Stuart's Adjutant-General, says that there was, at the beginning of the campaign, less than 6,000 for duty in the three brigades of cavalry that were with Stuart when he crossed the Potomac, there being about 4,500 in the two brigades of Robertson and Jones. He further says that the losses in action in these three brigades, which bore the brunt of the battle of Fleetwood, and the cavalry fights near the Blue Ridge, and from hard service and deficiency of forage, had reduced them to less than 4,000 when he crossed the Potomac; and he thinks to about 3,500. General Fitz. Lee thinks they were under 4,000 strong at the battle. This loss was not unreasonable, as will be seen when we come to notice that in the Federal cavalry. Jenkins' brigade, which was not embraced in the returns of May 31st, was about 1,600 strong before it crossed the Potomac, and White's battalion, which belonged to Jones' brigade, did not exceed 200. 6,000, therefore, will cover all the cavalry we had available for the battle. The artillery numbered 4,702 on the 31st of May, and some of it was very evidently left in Virginia with Corse's brigade, as the return for July 20th shows more present for duty in the artillery at that date than on the 31st of May. Some therefore must have rejoined the army by the former date, and very probably some that had been left with Jenkins' brigade near Suffolk had come back. We had 252 pieces with the infantry, as shown by a statement furnished me by General Pendleton, and allowing 15 men to a piece, which would be a superabundance, would give 3,780 men. Add 220 for the officers, giving nearly one to a piece, and we have 4,000, which certainly covered the artillery force with the infantry. There were 16 pieces of horse artillery with the cavalry, the men for which were returned with the cavalry, and as part thereof. They are included in the 6,000 allowed for that arm.

We had therefore not exceeding 60,000 men of all arms for duty at Gettysburg. In this estimate I do not include the cavalry brigades of Robertson, Jones and Imboden, which did not arrive in time to take part in the battle, and should not be counted as part of the force available for it. If they are to be counted as a part of our force at Gettysburg, then the 8,000 men under French at Frederick, which were employed in protecting Meade's communications [22] to the rear, and threatening ours, and Couch's force, a part of which was marching to Meade's assistance, and between a portion of which and Stuart's cavalry there was a conflict at Carlisle, on the 1st of July, should be counted as parts of Meade's force.

The loss in the aggregate present in my division, exclusive of losses in action and the regiments left behind, was fifteen per cent. from the 31st of May to the 20th of June, and after that near eight per cent. Deduct the same per cent. from 88,754, the aggregate present in the whole army on the 31st of May, and there would be less than 70,000 as the aggregate present at Gettysburg, without making any deduction for Robertson's and Jones' brigades.

It is, however, when the Comte de Paris comes to estimate Meade's force that he commits the greatest errors. It is a fact to be noted that he does not once refer to any official returns of that army, when it was a very easy thing for him to obtain them, and the return for the 30th of June, the day before the battle began, ought to furnish the very best evidence of Meade's force at the battle, but he resorts to the vague declarations of Federal officers, though he refuses to take the estimates of Confederate officers as to our strength in the absence of any return later than the 31st of May. This does not speak very well for his impartiality. When he ascertains what the Federal officers state as their present for duty, he insists that they mean thereby the aggregate present, including all men on extra duty, sick and in arrest, and then cuts down that number at a most extravagant rate. He says: “Whenever Federal officers gave what they called their effective strength, the figures represented always all the men present and not only those present for duty.” This was not only not the case generally, but it was not the case when he was connected with the Army of the Potomac. McClellan, in his report, page 11, gives the strength of that army at various periods — that for the 20th June, 1862, six days before the Seven Days battles began, being given as follows:

for arrest or confinement.Aggregate.
1862--June 204,665101,16049610,54144320117,226


Now, will the Comte pretend to say that McClellan intended by this that his effective strength was 117,226 on the 20th of June? In his testimony before the Committee on the Conduct of the War, McClellan said: “The largest number of men I had for duty at any time on the Peninsula was 107,000 men;” and in reply to the question: “How many available men did you estimate that you had at Harrison's Bar?” he said: “I think I had about 85,000 or 90,000 men at Harrison's Bar.” The same statement in his report that has been referred to, shows that on the 10th of July, 1862, when he was at Harrison's Bar, he had, present for duty, 3,834 officers and 85,715 men, total 89,549 for duty, and an aggregate present of 106,466. The Comte, therefore, is slightly mistaken in this respect, and the fact will abundantly appear from the various returns of McClellan contained in the same volume with his testimony, which are certified by the Adjutant-General. Upon this unwarranted assumption, the Comte takes the figures stated by Butterfield and Meade as the present for duty as the aggregate present, and then cuts them down by deducting thirteen per cent. for the men on extra duty, sick and in arrest. This is directly in the teeth of the return for the 30th of June, 1863, which I have been able to procure through the kindness of a friend in Congress, and to which return I will refer again when I come to estimate Meade's force.

The Comte is again grieviously at fault when he says: “The Federal regiments were certainly not stronger than the Confederate ones. The reason is, that by the operation of the draft, however limited, the old regiments in the Southern army were at certain times refilled by recruits, while on the Union side, whenever a new call of volunteers was made it was by the creation of new regiments. It is a well known fact that as soon as a regiment left for the army it ceased to recruit itself.”

He seems to think there was very great efficiency in the conscript act in keeping our regiments filled. Now, there were something over 500 regiments and 100 battalions of infantry, and smartly over 100 regiments of cavalry in the Confederate service, besides a great many battalions and batteries of artillery, as will be seen by reference to Colonel Jones' roster, which is imperfect in not giving all the regiments we had. Say we had 700 regiments in all to keep up, and 81,993 conscripts divided among them would give about 117 to a regiment, which would not refill it often. Add the 72,292 volunteers, and it would give only 154,285 men that were available [24] for recruiting all the Confederate armies east of the Mississippi river, after the 16th of April, 1862, up to February, 1865. Let us see how it was on the other side. The Comte seems to be unaware of the fact that, on the third day of March, 1863, an act of the United States Congress was approved, which provides for conscription, though generally designated the “Enrolment act.” On the 17th of March, 1863, the Bureau for Enrolment and Conscription was organized under Brigadier-General James B. Fry as Provost-Marshal General (see his report, page 13), and on the 1st of May, 1863, an order was issued giving it the superintendence of the entire volunteer recruiting system (same page): After the 3d of March there were no more calls on the States except for “emergency men.” The Provost-Marshal General, in his report (page 2), says:

One million one hundred and twenty thousand six hundred and twenty-one (1,120,621) men were raised, at an average cost (on account of recruitment, exclusive of bounties) of nine dollars and eighty-four cents ($9.84) per man; while the cost of recruiting the one million three hundred and fifty-six thousand five hundred and ninety-three (1,356,593) raised prior to the organization of the Bureau was thirty-four dollars and one cent ($34.01) per man.

Of the 1,356,593 raised before the organization of the Bureau, 1,146,189 were for three years, as shown by a table on page 160 of the report, 18,000 of them being called for for the navy; and if all that number went into that service, there were left 1,128,189 three years men for the army. On page 57 he says that in the summer of 1863, 956 volunteer regiments, 7 independent battalions, 61 independent companies, and 158 volunteer batteries were in the service. There were then less than 1,000 regiments, including those in the regular army, for the 1,128,189 three years men to be divided among, which would give over 1,128 men to a regiment. From the beginning to the close of the war, there was not quite 600,000 men put in the Confederate army in any way, which would give less than 1,000 each that the Confederate regiments received from the beginning to the close of the war. Of course it follows, as a necessary consequence, that in June, 1863, the Federal regiments were greatly larger than the Confederate regiments were at that time, unless we had rendered hors de combat a great many more of them than they had of us. Besides the troops put into the field before the passage of the Federal conscript act, it appears from the Provost-Marshal General's report (page 53), that 13,971 [25] militia were furnished by New York, and 32,104 by Pennsylvania in June, 1863, upon a call “by the President for troops to meet the emergency created by the rebel invasion, which culminated in the battle of Gettysburg.” These militia men, who were admirably armed, equipped and clothed, were certainly as good as any conscripts that the Confederate government could have sent forward to recruit our army after it started. I will here state that it appears (page 149) that of the 1,120,621 men furnished by the Bureau of Enrolment, only 168,649 men were actually drafted into the army, leaving 951,972 who were raised by voluntary enlistment by that Bureau; and of course they were put into old organizations. It is not unreasonable to suppose that some of them were put into the service before the battle of Gettysburg, as that Bureau began its operations for raising volunteers in May, 1863. The Comte has therefore jumped to his conclusion that “the Federal regiments were certainly not stronger than the Confederate ones.” His statement, that “the figures given by Meade and Butterfield do not show, as has been alleged by Dr. Bates, all the men borne upon the rolls; nor, I think, as Confederate writers have asserted, only the men present for duty on the battle-field, but all the men who at the morning call were not reported absent, whatever may be their occupation at that time; the men known as having fallen off the ranks not being generally reported absent at once, to give them a chance to join without losing their pay, the usual stragglers were in fact embraced in that figure,” --is calculated to excite a smile from any military man, and would no doubt elicit an indignant protest from General Meade if he were alive. Of all men about an army, the most worthless was a straggler, for he was always up to get his share of the rations, but never present to do his share of the fighting. The deserter was infinitely better, for by absenting himself he ceased to be a burthen on the commissariat of the army, and rendered fully as much service as the straggler. No military man of one grain of sense would be likely to count him as a part of his “effective strength in battle.” In using the terms “effective strength,” and “present for duty,” Generals Meade and Butterfield knew the full import of the terms they used, as is conclusively shown by the report of the 30th June, 1863, supervised by the one and signed by the other. The following are abstracts from the returns of June 20th and 30th and July 10th, 1863: [26]

troops.present.present for duty equipped.Present and Absent.
for duty.aggregate.infantry.cavalry.artillery.
Total Commissioned Officers.Enlisted Men.Aggregate.Commissioned Officers,Enlisted Men.Commissioned Officers.Enlisted Men.Commissioned Officers.Enlisted Men.aggregate.
June 20th, 1863.General and staff, provost guard, engineer brigade, &c2002,4672,6673,048  252  4,714
Artillery reserve1183,1083,2263,46622321  962,7874,459
1st Army Corps7189,1759,89311,7196508,317  2059317,584
2d Army Corps85610,51911,37512,7448199,6313791353819,962
3d Army Corps80311,84912,65213,98475010,504  1765122,459
5th Army Corps6159,68810,30311,8685388,582    16,870
6th Army Corps99414,43015,42417,41895313,1025119301,00824,467
11th Army Corps5859,94910,53412,0635429,0784551563017,184
12th Army Corps5208,1888,7089,9614967,802  1167714,585
Cavalry5669,62610,19212,162      20,417
June 30th, 1863.General and staff, provost guard, &c1822,3982,5803,031      4,125
Artillery reserve1042,4642,5682,74523312  722,1393,138
1st Army Corps7599,59610,35512,1576878,716  2159817,502
2d Army Corps96812,08813,05614,37392711,4363791453722,317
3d Army Corps83111,79912,63013,88179610,451  1965822,403
5th Army Corps83712,37413,21115,10279711,157  854721,365
6th Army Corps1,03114,67915,71017,62598613,5305119331,00624,036
11th Army Corps6039,97310,57612,0965498,6486461562917,374
12th Army Corps5418,0568,5979,8165217,672  1238414,574
Cavalry5669,62610,19212,162      20,417
July 10th, 1863.General and staff, provost guard, &c1572,1112,2682,648      4,112
Artillery reserve1122,8622,9943,14837219  672,1613,778
1st Army Corps4294,3634,7926,0913723,509  1851514,645
2d Army Corps6207,6258,2459,3895836,8992791050321,147
3d Army Corps89613,50614,40215,77977911,543  351,07929,311
5th Army Corps6689,40110,06911,8356368,241  843220,383
6th Army Corps99113,65414,64516,35192612,32051102795523,380
11th Army Corps4126,4836,8958,206  5351157015,729
12th Army Corps5768,5039,07910,4005458,022  1238516,983
Cavalry61211,23011,84213,717  56710,2422067223,216


There are in the returns a great many columns with various headings to show those present sick, on extra duty, and in arrest, and so with the absent, as well as for alterations. All the figures under these various heads are not given in the transcript furnished me, but enough is given to show all the present for duty, and the aggregate present, as well as the aggregate present and absent. Opposite the cavalry in the returns for June 20th and 30th is this remark: “Taken from last return received, May 31st, 1863.” Opposite the artillery in the return for July 10th is this note: “Brigade of regular batteries, aggregate 595, omitted in last report of June 30 (on account of loss of previous returns and absence of the officer who could replace them), included as gain in this report.”

Hooker in his testimony (page 162) says that, at Fairfax Courthouse, Stahl's cavalry, numbering 6,100 sabres, was added to his cavalry — which was about the 16th or 17th of June.

As the cavalry for duty on the 31st of May numbered 10,192, the addition of Stahl's increased it to over 16,000, from which are to be deducted the losses in action, &c.; but as the return for July 10th showed 11,842 for duty in that arm at that date, it must have numbered considerably more than 12,000 for duty at Gettysburg. The brigade of regular batteries, out of an aggregate of 595, must have numbered at least 500 for duty according to the ratio in the other artillery, and that ought to be added to the present for duty at the battle. Lockwood's Maryland brigade joined the Twelfth corps on the morning of the 2d of July, and Stannard's Vermont brigade was added to the First corps on the same morning: of this fact I am positively assured by the Comte de Paris in a letter to me, and Bates also states it. But the fact is very apparent that they were not included in the returns of those corps for the 30th from the returns themselves. Butterfield and Bates show that they numbered 2,500 each, making 5,000 for the two, and that number should be added. We shall thus have--

Number for duty by report of 30th June99,475
Lockwood's and Stannard's brigades5,000
Addition to cavalry, say2,000
Brigade of regular batteries500
Total for duty106,975

Meade certainly had at least that number for duty at Gettysburg, though all of it might not be regarded as effective for a fight. His report, however, shows what was the actual force of infantry [29] and artillery equipped and ready for the fight on the 30th, under the heading of “Present for duty equipped.” At the foot of the transcript, which is given on the regular printed form, is this printed note: “Under the heading ‘Present for duty equipped’ only those will be given who are actually available for the line of battle at the date of the regimental reports” --that is, it includes none but line officers and men who actually go into the fight. For June 30th the number so present and equipped in the cavalry is. not given, but it is given in the return for July 10th, and then amounted to 11,045. It can therefore be safely assumed to have been 12,000 at Gettysburg. The numbers under that heading then are as follows:

Infantry — Officers 5,286
 Enlisted men 71,922
Add for Lockwood's and Stannard's brigades 5,000
Total infantry 82,208
Artillery — Officers194 
 Enlisted men6,498 
Add for regular batteries500 
Total artillery 7,192
Cavalry  12,000
Total “Present for duty equipped”  101,400

Thus we get the actual fighting force available, after eliminating all the general and staff officers, provost guard, engineer brigade, signal corps and guards and orderlies, at over 100,000 officers and men. In my estimate of our own strength, I have only taken out the staff officers, who, under no circumstances, were required to get under fire, and left in all general officers and their staff officers, including engineer officers, as well as the non-commissioned staff officers. By an examination of the returns for the reserve artillery and the corps, it will be seen that besides the 2,580 at army headquarters, there are 2,803 officers and men reported for duty who are excluded from the statement of the “Present for duty equipped” in Meade's army.

No amount of figuring by the Comte de Paris, and no hocus pocus with his figures by General Humphreys, can evade the conclusive proof of the official return of the 30th of June, which bears Meade's signature.

Add for Lockwood's and Stannard's brigades, the increase in the [30] cavalry from the 31st of May, and the brigade of regular batteries to the 112,988, and the aggregate present would be smartly above 120,000.

In order to show how fallacious is the Comte's theory that there was a decrease of the number for duty in the Army of the Potomac on the march, it is only necessary to compare the returns of June the 20th and 30th together. At the former date, the return shows, in the seven corps, a total present for duty of 78,889, whereas at the latter date the return shows a total present for duty in these same corps of 84,135, being an increase of 5,246. The only evidence of any addition to these corps in the way of new troops is in regard to the addition of two brigades of Crawford's division to the Fifth corps, and the increase in the present for duty in that corps is only 2,908, in the aggregate present 3,234, and in the aggregate present and absent 4,495; whereas there was a total increase, in the present and absent of the seven corps, of 5,560. It appears that there was an increase in the Second corps of 2,355 in the aggregate present and absent, and an increase of the present for duty of 1,681. In the Eleventh corps there was an increase in the aggregate present and absent of 190, and in the present for duty of 42. There must, therefore, have been some additions to the Second and Eleventh corps in the way of recruits, or new organizations attached to them, of which no account is given. In each of the other corps there was a small decrease in the aggregate present and absent, and in all of them, except the Third corps, there was an increase in the number present for duty, showing that the additions to them in the latter respect were from the return of convalescents or others to duty. In the Third corps there was a decrease of 22 in the present for duty. Now, when the returns show a gradual increase in the numbers present for duty, and the aggregate present also, from from the 20th to the 30th of June, though the army was moving all the time, that increase being independent of any recruits or addition of new troops, the Comte de Paris has undertaken a task simply impossibe in attempting to show that there was a decrease of thirteen per cent. in the numbers reported for duty on the 30th of June, or stated to have been present for duty on the 28th, in so short a space of time. In order to succeed, he must first show that false returns were made out by both Hooker and Meade.

The return for May 31st showed 10,192 present for duty in Pleasonton's cavalry, and there was added to it Stahl's cavalry of 6,100 sabres, making the whole about 16,300, and this the Comte reduces [31] to 10,440 at the battle, thus disposing of near 6,000, while he is only willing to allow for a loss of 1,100 in battle in Stuart's cavalry, and 1,606 more from other causes. Now, if Pleasonton's cavalry had been reduced by the casualties in battle and the wear and tear of the campaign, when the government furnished new horses to the dismounted men, from 16,300 to 12,000 (the figure at which I put it at Gettysburg), is it unreasonable to assume that Stuart's cavalry had been reduced in the same ratio during the same period — that is, from 10,292 to 7,500, thus giving Stuart 4,000 in the three brigades with him, and 3,500 with Robertson and Jones?

The Comte de Paris must not be surprised if he is suspected of not treating this question of numbers with the impartiality that is demanded of a historian.

General Fitz. Lee, as shown by the first part of his very clever article on the battle of Gettysburg, in the April number of the Papers, has permitted himself to be misled by Federal officers as to the numbers on their side at the battle. In a note referring to Colonel Taylor's estimate of the strength of the two armies, he says: “The Federal force is overestimated. Their total of all arms was about 90,000. General Humphreys puts, in a letter to me, the Federal infantry at 70,000, inclusive of 5,000 officers.”

By reference to the abstracts I have given, the accuracy of which he can verify, if he thinks proper, by inquiry at the Adjutant-General's office, General Fitz. Lee will see that in the seven corps of the Army of the Potomac, there were, on the 30th of June, 5,286 officers and 71,922 enlisted men, making a total of 77,208 “Present for duty equipped” --that is, ready to go into a fight; and when Lockwood's and Stannard's brigades were added on the morning of the 2d July, there were 82,208 officers and men in the infantry available for duty in the line of battle. This should satisfy him that his other estimates, founded on testimony similar to that adduced on this point, in regard to the force available to oppose an advance by us after the close of the fight on the 1st, are fallacious. By reference to the return of July the 10th, he will find that the Eleventh corps had still 6,895 officers and men for duty, and the First corps 4,792, after the losses not only of the first day, but also of the second and third, though there had been no additions to either corps after the battle, and Stannard's brigade, which joined the First corps on the second day, had departed because of the expiration of its term of service.

I will not continue the discussion with him of the propriety and [32] feasibility of an attempt to take possession of the heights at the close of the first day's fight. He admits that “of course, after the arrival of his chief, all responsibility was taken from Ewell in not ordering the troops forward — it was assumed by and is to be placed on General Lee.” That is what I have always thought, and the statement of Colonel Taylor that “General Lee witnessed the flight of the Federals through Gettysburg and up the hills beyond;” of General Heth, that he applied for and obtained permission from General Lee to attack while Rodes was engaged; and of General Pendleton, that General Lee arrived on the field about two P. M., and gave instructions for posting some artillery so as to enfilade the enemy's line before it began to fall back, settles the question of his presence beyond all dispute. Ewell is therefore relieved from the responsibility for not ordering a general advance, and it rests on General Lee, according to General Fitz. Lee's own admission. General Lee's fame can stand the ordeal of all the criticisms of all those who were not present, and can therefore form no just estimate of the obstacles to an advance on our part that presented themselves on the occasion. The order to Ewell contemplated the use of only his own troops then at hand, to carry the hill, if he found it practicable without bringing on a general engagement. He was on the low ground at the foot of the hill, and could neither see the enemy nor form any estimate of his strength, while General Lee had a much better view from Seminary ridge, and he ordered none of Hill's troops to advance. Ewell could not do so when the Commanding-General was present. If he had gone forward with his less than 8,000 men that were available before the arrival of Johnson, he could not “have shattered the Twelfth corps--possibly portions of two others;” and as our position was perfectly in view from Cemetery hill, and all our movements could be seen, when we commenced ascending that hill, Buford with his 2,500 cavalry might have swept around the town on our right, released the several thousand prisoners we had taken, and destroyed our trains, as there would have been nothing in our rear to oppose him.

When Johnson arrived, which was after six P. M., the opportunity for taking the heights without a desperate and uncertain struggle had passed, as Generel Hancock's statement makes very apparent.

Those who are still disposed to carp at the operations of the first day, can turn their batteries on General Lee, if they think proper; but it is very easy to imagine what would be his reply if he were alive.


Roster of infantry, A. N. V., at battle of Gettysburg, by General J. A. Early.

The infantry of the Army of Northern Virginia as it was reorganized just before the commencement of the Pennsylvania campaign of 1863 and as it remained up to the 1st of May, 1864.

1 Note.--In a communication to the Secretary of the Southern Historical Society from the Comte de Paris, dated March 23d, he states that he has now obtained a full copy of the returns of the Army of Northern Virginia for May 31st, 1863, by permission of the Adjutant-General of the United States army, but it does not appear that he has either applied for or obtained a copy of the return of Meade's army for the 30th of June, 1863, or at any other time — a fact which does not argue that diligent and impartial research which should characterize one who assumes the role of an historian.

In making his deductions from the return of May 31st, 1863, as he now has it, be again falls into some errors which it is proper to notice. He says: “Early's division had some of the hardest marching before it reached the Potomac,” &c. In this he is mistaken. The march from Fredericksburg to the vicinity of Culpeper Courthouse had been very deliberate, occupying from the 4th to the 8th of June, inclusive. From the vicinity of Culpeper Courthouse to Winchester, a distance of about fifty miles, the division had marched in four days from the 10th to the 13th, inclusive. After being engaged around Winchester the afternoon of the 13th, the 14th and the morning of the 15th, having taken a day's rest, It moved to Shepherdstown on the Potomac, a distance of between thirty and thirty-five miles, by the 20th. It had thus occupied ten days in reaching the Potomac from the vicinity of Culpeper Courthouse, a distance of about eighty miles-one day and parts of two others being occupied In the operations around Winchester. Longstreet's corps left Culpeper Courthouse on the 15th, and Hill's left the heights of Fredericksburg on the same day, and, as they crossed the Potomac on the 25th, after Longstreet's corps had done some extra marching to support Stuart's cavalry, it follows that both corps did much severer marching before crossing the Potomac, than my division or any other part of Ewell's corps had done. The weather was also more sultry during the period of their march than it had been during ours.

The Comte now finds, by comparison of the returns of my division for the 31st of May with that of the 20th of June, that there was a loss of only 293, after deducting for the three regiments left behind, and the losses in action, which he states to be less than four per cent. He is here again mistaken. 293 is a little over four per cent. on 6,943, the entire strength of my division for duty on the 31st of May, and deducting the 850, which he allows for the three regiments left behind, it amounts to very nearly five per cent. on the residue of the division. But the fact is, that in the return of June the 20th was Included the Thirty-first Virginia regiment, which was not included in the return of the 31st of May, as I have above explained. That regiment numbered for duty, on the 20th of June, 280, which, being added to the 293 assumed by the Comte to be my total loss, makes a loss of 573 in the portion of the division included in both returns, being very nearly nine and a half per cent.

As there were no detachments made from Hays' and Gordon's brigades, and no additions to either, I have taken those two brigades to ascertain the ratio of decrease, in the absence of the return of the Thirty-first Virginia for the 31st of May, and of the three detached regiments and battalion for the 20th of June.

In those two brigades the decrease, exclusive of loss in action, was a little over ten per cent., and hence, as the marching they had done was not as severe as that done by Longstreet's and Hill's corps before they crossed the Potomac, I have assumed ten per cent. as the ratio of decrease in the whole army.

It is a little curious that, as the Comte thinks the loss in our army must have been very small from sickness, straggling and desertion, on account of the very fine weather (another fact about which he is greatly mistaken, as will be recollected by those who had to endure, without shelter, the heavy rains and cold nights we frequently had), he should make the decrease in Meade's army so excessive for the four days preceding the battle of Gettysburg.

It is true, as the Comte says, that when there were but eight divisions in the army there were but two smaller than mine, but when the number was increased to nine, mine became and remained more than an average one in size.

2 So in the copy.

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 United States License.

An XML version of this text is available for download, with the additional restriction that you offer Perseus any modifications you make. Perseus provides credit for all accepted changes, storing new additions in a versioning system.

hide Display Preferences
Greek Display:
Arabic Display:
View by Default:
Browse Bar: