This is a very difficult topic because different numbers of men straggled by different amounts.
If we assume that the Union forces and the Confederate forces straggled by about the same amount (i.e. 63% of theoretical PFD gets into line of battle, as per the Union 1st Corps), then the rough estimate is about 47,000 as against 55,000 Union troops on the field (i.e. including Morell and Franklin, but not Humphreys and Couch). In both cases this is just 73,000 (CS) and 87,000 (US) times 0.63.
Going into more detail means looking at individual divisions, brigades or even regiments, and that means trying to source the numbers where they claimed how many troops they had. This is particularly problematic when the numbers we have conflict with one another - for example Jackson's division is down in Carman as 1,784 infantry, but then Starke's brigade (1/3 of the division) claimed about 1,400-1,500 infantry at Antietam. This seems to be because Jackson's divisional strength was measured when they'd just arrived and all the stragglers been collected up, but Starke's brigade was measured when they were about to go into action (which is much closer to what the 63% number for 1st Corps represents).
On the 2nd September returns, per Hartwig, Jackson's division was (in PFD):
JR Jones | Grigsby | 1160 | |
| Johnson | 812 | |
| Warren | 1543 | |
| Starke | 1623 | |
| Arty | 440 | 6 Btys |
So 1,400 for Starke would indicate that his brigade was at 86% of nominal strength, while 1,784 infantry for the whole division would indicate that the division was at about 35% of nominal strength. This difference is pretty huge - if applied across the whole army it would mean the difference between claiming about 63,000 and claiming about 25,500.
67th Tigers went through the numbers in Carman and reassessed them a few years ago, including going to Carman's sources, and the rough numbers he gets are:
DH Hill: DH Hill claimed 3,000 infantry, but Carman calculated 5,449 infantry and 346 artillery.
Via Clemens we get:
Ripley's brigade: 1,349 infantry into action
Rodes' brigade: 3rd, 6th and 12th Alabama carried 840 into action, excluding at least the officers of the 12th. 5th Alabama was a small regiment, but no clue about the 26th; if the 5th and 26th average the same as the other three then it's ca. 1,400.
Colquitt's brigade, from Carman's correspondence:
6th Georgia: 300+ (and one company on picket not counted)
23rd Georgia: 485 men (not officers)
27th Georgia: 400 men (not officers)
28th Georgia: 250 men or 250-275 officers and men
13th Alabama: Carman guessed 310 based on a 2nd September strength of 612
Which implies that there were ~1,780 men plus officers, for a brigade strength of ~1,900
Garland's brigade:
Carman gives the brigade 756, but it was the strongest in the division on the 2nd September (though it did then take plenty of casualties at South Mountain) and the 5th North Carolina had ~625 men at Antietam. Adding officers and assuming the other four regiments are each about half the size of the 5th NC gives an estimate of about 2,100, but I'd say we could go with 1,700 to avoid overclaiming. Effectively this is assuming the other four regiments average 250 officers and men.
GB Anderson's brigade - Carman gives 1,174 and this seems fairly solid.
Divisional total:
1349+1400+1900+1700+1174 =
7,523 infantry, and 346 gunners.
McLaws: Carman gives
2,823 infantry and 138 gunners. This might actually make sense, on 2nd September McLaws disposed of 7,340 infantry and 419 gunners but they got hit hard at Crampton's Gap and suffered 1,907 claimed infantry casualties. If the actual casualties were higher and there was the usual straggling, or those casualties and a bit more than the usual straggling, then I suppose it's possible they could be down to about 3,000.
RH Anderson: The brigades except for Armistead (who remained in reserve) were sent to reinforce DH Hill, who estimated it at ca. 3,000-4,000. As we've seen DH Hill's estimated strength was really low for his own division, so it could be considerably higher. (The 22nd September return gives 5,324 PFD, and it suffered 1,278 casualties at Antietam; it'd probably be reasonable to say it had about
5,000 actually with the division once Armistead is counted.)
Walker: Carman calculated
3,764 infantry and 230 artillery.
DR Jones: DR Jones himself estimated "only 2,430 men" from a pre-campaign PFD of 8,953 infantry; he suffered about 1,000 casualties in the pre-Antietam fighting, but being down to less than 2,500 simply isn't credible. In particular this division includes Drayton's brigade, which is the one that had had almost no straggling until September 11th, so a lot of the hypothetical stragglers would have had to fall out of line and be swept up.
Carman estimated 3,311 infantry and 81 artillery, but going brigade by brigade:
Toombs: Toombs claimed over 400 muskets in the 2nd and 20th Georgia, while Carman estimated 357 officers and men in the same two regiments. Then 15th and 17th Georgia (Carman est. 281 muskets) joined them, and half the 11th Georgia (reported as 140 muskets in the five companies present).
This implies about 750-780 men minus officers, or a bit more with officers (so ca. 825).
Garnett: Carman gave 261 officers and men in the brigade, arguing that the brigade was recruited from near where Lee crossed the Potomac and briefly deserted. This would imply something like 80% of the brigade had either done that or was elsewhere.
Kemper: Carman estimates 443 on the field by adding regimental reports and guessing for the 28th VA, but others reported the brigade at 500-600 muskets (which would imply 600 or so with officers added, indicating the 28th VA was a strong regiment). 600 would also indicate straggling of about 50%.
GT Anderson - Carman estimates 597, but see Hood/Evans...
Drayton: Carman estimates 465, excluding 11th Georgia (detached and one wing fought in Toombs' brigade, 140 men counted there). Carman was told by a veteran the 15th SC had 300 men, but he used 200 in his calculation. Carman's estimate of 100 for the 51st Georgia is given no basis but we'll use it. The Phillips Legion infantry was also present, but Carman doesn't include it.
Correcting the 15th SC from 200 to 300, and adding the Phillip's Legion (ca. 135) plus adding officers would be ca. 700 men plus about 80 officers, so 780.
Jenkins: Carman quotes a newspaper to give 755 but in correspondence with the brigadier he's told 1,250 officers and men plus the 1st SC (106). So 1,356.
This totals to 825+261+600+780+1356 infantry, for
3,822 infantry (not including GT Anderson) plus
81 artillery.
Hood/Evans:
Carman says Evans' brigade straggled really badly, down to 284 infantry and 112 gunners for 396 all told. That's down from 1,058 on 2nd September, though there were 216 casualties so it means "only" about 53% straggling, or 47% of the strength in line.
Hood's brigade and Whiting's brigade are estimated at 2000 infantry (Hood gave this) and 304 artillery, and this does not include officers (so ca. 230 officers). 2230 infantry means about 62% of theoretical strength in line.
GT Anderson's brigade: 597, and Evans claimed it in his report. That's about 43% of theoretical strength which is another bad straggle.
Total for these four brigades considered here is
3,111 infantry and 416 artillery.
Jackson: As covered above, the number Carman gives is 1,784 infantry and 310 artillery, but this is just after the arrival on the field. If we take Starke's number of 1,400-1,500, take the low number (1,400) and correct the other three brigades to match then we multiply their September 2 strength by 0.86, to give:
Grigsby 1,000
Johnson 700
Warren 1,330
Starke 1,400
Total
4,430 infantry and 310 artillery
Ewell:
Carman gives 3,904 infantry and 223 artillery, but for once there's full reports in the OR. Ewell moved more slowly than the Stonewall division and took care to keep his straggling under control (he paused south of Boteler's Ford to collect up stragglers before crossing), and Lawton and Early's brigades match very closely to 22nd September PFD plus casualties; the other two don't.
It looks like there's an artillery battery buried in the reports for Hays and Trimble (as Ewell's division reported artillery in brigade totals) and the numbers for Hay and Trimble are also a bit vague - Hays' entire brigade is reported as the force Hays took to the wheatfield, and Trimble is just "out of less than 700 men carried into action". Assuming the four brigades behaved similarly you get about 1,000 in both (1078 instead of 550 for Hays and 1029 instead of 700 for Walker) but this includes the artillery. This would give a total of about
4,400 infantry and 223 artillery, and would mean that there were 73% of the infantry in battle - what we'd expect from a division that controlled straggling especially well.
AP Hill: Carman gives 2,231 infantry for the three engaged brigades, plus 337 artillery. The three engaged brigades were Branch, Archer and Gregg (2nd September total 3,887, so 57.3% of 2nd September strength engaged), with Pender and Field (2nd September total 2,932) not engaged; if Pender and Field straggled as much as Branch, Archer and Gregg, that gives an extra 1,683 infantry for a total of
3,914 infantry and 337 artillery on the field.
Total strengths of the infantry divisions:
DH Hill 7523 + 346
McLaws 2823 + 138
Anderson 5,000 + 0
Walker 3,764 + 230
DR Jones 3,822 + 81
Hood/Evans 3,311 + 416
Jackson/Stonewall 4,430 + 310
Ewell 4,400 + 223
AP Hill 3,914 + 337
Total infantry
38,987
Carman gives the cavalry directly at
4,500, and the total artillery at
3,629. This would mean a grand total of
47,116 effectives
This estimate has large error bars on it, and could be off by a few thousand either way. There were also quite a lot of men who have to have been just behind the lines, as taking the September 22 return and adding back casualties gives just under 60,000 for the whole army - a difference of 13,000 men.