At this point, I cannot reconcile the 3rd Cavalry Reconnaissance Squadron account with the 7th Armored Division account, neither of which mentions the other at Épernay.

Before going into the specifics of the events, it is important to note the following so as not to arrive at erroneous understanding of the events.

1. The Germans destroyed two C/31 tanks and not just one.
   This is clear by the fact that the MOS (Military Occupational Specialty) of those killed included two drivers (MOS 736) and two gunners (616). A tank crew had only one of each MOS. This is significant because the post-combat Graves Registration investigators appear to have assumed that only one tank was destroyed and that all those killed and those who survived came from a single tank crew.
2. The deaths came in separate events.
   While I do not yet know which men died in which events, it is clear that there were at least three events: (1) one C/31 tank, (2) the other C/31 tank, and (3) the death of Capt. Davis of A/31 who would not have been with C/31. The deaths of the 23 AIB men could have been separate from these events or could have been part of these events. As I do the research, I hope to clarify this matter.
3. Unknowns X-5, 6, 7 were really 4 men and not 3
   The men killed at Epernay were buried by the French in the town cemetery. When Graves Registration personnel disinterred the remains, they identified one grave as a group burial (later identified as Ensminger, Shaljian and Tucker) and assigned X-5, 6 and 7 to the Unknowns in that group. However, it was later determined that Bossman's remains were co-mingled posthumously. That is, the French buried Bossman next to the other three, but Graves Registration co-mingled Bossman's remains with the other three during disinterment and transfer to Champigneul. I have yet to figure out how this was later resolved since Graves Registration had only three X-numbers but four sets of remains. It appears that Bossman was finally cross-referenced as X-7. As I do the research, I hope to clarify this matter.
4. Delayed Reporting of A/31 Deaths
   At least one of the two 31st Tank Battalion companies (A/31) was detached from Battalion Headquarters and in a different task force. In addition, the three task forces were moving rapidly each day over long distances during this period. These two factors probably explain why the Morning Report entries for the A/31 casualties at Épernay do not show up until several days later and then show only as Missing in Action, even though the men who were with them knew that they were Killed in Action. The C/31 casualties at Épernay were recorded in the 28 Aug Morning Report, so that C/31 may have been in the task force attached to Battalion Headquarters. As I do the research, I hope to clarify this matter.
5. Chronology of Events
   I have yet to find a definitive chronology of when each unit lost men. I do have some times and locations. But, I have yet to fit together a chronology that includes the A/3rd Cavalry Reconnaissance Squadron casualties and those of 7th Armored Division. I hope to develop a chronology, but the records are fragmentary which makes that a difficult task. A major problem is that this was the first month in combat for 7AD and 3rd Cav, and they were still working out just what they should include in their monthly and daily reports. So, some reports are very specific while others do not help very much.

By far the most complicated deaths to understand are those of Capt. Davis and Cpl. Stephens. They were the only members of Company A, 31st Tank Battalion, lost that day. But their loss was not included in the Morning Reports until 4 days later when they are reported Missing in Action as of 28 Aug 1944 in the A/31 Morning Report of 1 Sep 1944. The Individual Deceased Personnel File of Capt. Davis is a PDF file of 209 pages -- the largest of any 7th Armored Division IDPF.

28 Aug 1944: A/31 Morning Report
The A/31 Morning Report for 28 Aug 1944 definitely places A/31 at Epernay and gives this Report of Events: "Left E of Provins at 0920 arrived near Epernay at 1500 via motor convoy distance traveled 40 mi". The only personnel entry is that 1st Lt. Foster assumed command. This is in spite of (a) Capt. Davis not shown as absent and (b) the officer strength of A/31 showing a field officer or Captain present. There was implicit acknowledgement that Capt. Davis was no longer in command. And yet he was still shown as present on the MRs of 28-31 Oct 1944. The Battle Casualty Report No. 20 for Capt. Davis was not filed until 12 Sep 1944, showing him as MIA, which was changed 22 Sep 1944 on Battle Casualty Report No. 30 to KIA. (See Davis IDPF p 107 para 2a.)

Undated, but on or before 7 Sep 1945: "Check List for Disinterment of Unknowns"
The disinterment of X-8 (PDF pp 13-14 and 139-140) from the Épernay town cemetery shows his OD pants had markings of Capt. Davis: "In waistband of trousers was found the name 'ALVIN E. [sic actually F.] DAVIS' A serial number was found also but was not very legible. It was thought to be 'O-362602' [which was the ASN of Capt. Davis]. Also found in waistband of pants were the numbers '111654'."

7 Sep 1944: Burial at Champigneul
His remains were buried at the Champigneul #1 temporary U. S. military cemetery at 1900 on 7 Sep 1944 (PDF pp 151-2 and copy 149-150). The Report of Burial, completed the next day, made no reference to the clothing marks included in the disinterment checklist. The "Cause of Death" surmised from the conidition of his remains by the 3041 QM Graves Registration Company team was "KIA, Burned in Tank".

19 Sep 1944: 3041st Quartermaster Graves Registration "Supplemental information covering bodies, Unknown X-7, Unknown X-8"
The memorandum (PDF p 10 and 141) shows that their understanding of the situation was highly erroneous. Paragraph 3 reference "two bodies listed as (X)" when in fact there were five bodies (Ensminger, Shaljian, Tucker, Bossman, Davis). Paragraph 4 states that (a) the Unknowns were all burned beyond recognition and (b) they were all from a single tank from which survivor T/5 Henry Bruzdowski/Boizdowksi of C/31 escaped, completely unaware of the fact that any A/31 tank had been hit and that in fact two C/31 tanks had been hit. Also recovered from the same cemetery at the same time were Stephens, Holtz and Kos.

2 May 1946: Initial Identification
The initial identification of X-8 Champigneul as Capt. Davis was made on or before 2 May 1945 (PDF p 11). "Identification was established by clothing marks and investigation by Third ARMY."

15 May 1945: 7th Armored Division Headquarters challenge of the identification
7th Armored Division Headquarters wrote a 2-page challenge of the identification, with 5 inclosures (PDF pp 107, 109 plus inclosure 3 on p 114, inclosure 4 on p 115, inclosure 5 on p 10). The challenge was clearly based on the "Supplemental information" memo's erroneous characterization of the events and tanks and units involved. Here are the images of the main document and inclsoures 3-5. (Inclosures 1 and 2 are the Report of Burial documents for Capt. Davis and X-8.) Click on the thumbnail to see the full image.

Memo Page 1 of 2

Memo Page 2 of 2

Inclosure 3

Inclosure 4

Inclosure 5

This is very much in line with the description given to me 28 Sep 2002 by 3/A/31 veteran Joie Gjernes who said the Capt. Davis was the A/31 commanding officer and that he was killed while his tank was crossing a bridge. He was killed by a shell that hit the tank in the turret where Capt. Davis was standing, and the shell pretty much cut Capt. Davis in half.

The 7AD challenge seems based primarily on two apparent discrepancies in the records supplied by Graves Registration. First (para 2d), 7AD HQ cites the fact that the Report of Burial of X-8 (PDF pp 149-150) shows no clothing information. It refers to the supplemental information sheet (PDF p 10), but that too has no clothing information. Graves Registration did not sent 7AD the Disinterment Checklist that gave the information about X-8 wearing clothing marked as Capt. Davis. Thus 7AD's challenge was valid but only because they had not received all the information from Graves Registration.

The second basis of 7AD's challenge was based on yet another flaw in what they reported: the supplemental information sheet (PDF p 10) seriously mischaracterized what had happened and confused a C/31 survivor with an A/31 death and tank. Once again, 7AD's challenge was valid but only because of the faulty Graves Registration supplemental information.

So, Graves Registration's identification of X-8 Champigneul as Capt. Davis was correct. But 7AD's challenge of that identification was also correct, based on the incomplete and flawed information provided to 7AD by Graves Registration.

21 Feb 1949: Dental Match Confirmation
Graves Registration in Europe sent a memo to the Quartermaster General in Washington (PDF p 128) about the reinvestigation of the case and confirmed the original identification. Paragraph 3 specifically states "Inasmuch as the enclosed tooth chart for this deceased [i.e., X=8 Champigneul] is in agreement with Form 371 for Capt. DAVIS, it is recommended that present identification be accepted ..."

25 Mar 1949: Summary of Identification
A 1949 "Identification Check List" (PDF p 91) summarized the identification status. The original identification was based on the trouser markings, as noted in the Checklist for Disinterments. 7AD HQ challenged the identification, and a reinvestigation took place. The reinvestigation confirmed the identification as Capt. Davis: "Favorable comparison of tooth chart accomplished upon reprocessing, with Army dental records for Capt. DAVIS substantiates identification as originally established."

Cpl. Virgil S. Stephens
I do not have any accounts of his death by veterans. His MOS 616 means he was the gunner (the bow gunner) of a tank crew. I do not know if he was in the crew of the tank of Capt. Davis. The only mention of him I have thus far found in the IDPF of Capt. Davis is in the "Supplemental information" sheet (PDF p 10) that shows him as one of the identified American soldiers who had been buried at the civilian cemetery at Épernay.

The same 1 Sep 1944 A/31 Morning Report that shows Capt. Davis as Missing in Action as of 28 Aug also shows Cpl. Stephens as Missing in Action as of 28 Aug. And the same 7 Nov 1944 A/31 Morning Report that changed the status of Capt. Davis is Killed in Action as of 28 Aug also shows the status of Cpl. Stephens changed to Killed in Action as of 28 Aug.

The 7 Sep 1944 Report of Burial (PDF p 40) shows that the burial team surmised his cause of death as "KIA, Head severed". When he was disinterred from the Champigneul, France, temporary US military cemetery for repatriation (PDF p 6), the condition of his remains was documented as "Skeleton, multiple fractures with extremeties missing, R/clavicle, and scapula". So, his head was apparently present.

My current conjecture is that he was the bow gunner of Capt. Davis' tank. He may have been sitting with his head above the lower hatch when the shell that killed Capt. Davis took off the head of Virgil Stephens just before it hit the turret in which Capt. Davis was standing.

The IDPF of Capt. Davis of A/31 has survivor accounts from the tank of Lt. Shaljian (Davis IDPF pp 120-123). Since only two crew members survived, it seems likely that one of the C/31 gunners killed (either Ensminger or Sroka) was a member of this crew.

The post-war (1949) book The 3rd Cavalry Reconnaissance Squadron (Mecz.) in World War II, 9 August 1944 to 9 May 1945 gives the following account (PDF pp. 93-94) of events 28 Aug 1944 at Épernay:

The problem with this account is that it makes no mention of 7th Armored Division, and the 7th Armored Division After Action Reports (7AD, CCB, 31st Tank, 33rd Engineers) make no mention of 3rd Cavalry. Furthermore, the Individual Deceased Personnel File of Cpl. Cumbo has no information about where he or how he was killed and shows his remains were collected by Graves Registration from Provins, which is a bit over 50 miles south of Épernay so that it makes death at Épernay very unlikely.

In the extensive folder of 3rd Cavalry Group combat interviews (National Archives RG 407 Entry 427 Box 24106 [in 2004] Folder 321), there is a copy of the Group's August 1944 After Action Report. The following account is from page 2 of that report.

At this point, I cannot reconcile the 3rd Cavalry Reconnaissance Squadron account with the 7th Armored Division account, neither of which mentions the other at Épernay. The 7AD After Action Report places entry to the town at 1800 while the 3rd Cav report places entry at 1700. So, it is possible that both are accurate and that neither knew of the other.

T/5 George Miranda was wounded on 28 Aug 1944 and died the next day, 29 Aug 1944, at the 35th Evacuation Hospital. I researched his death in case he was a casualty at Épernay, but I think it is most likely that he was wounded somewhere along B/33's route from Saint Loup to Avize, France, which is the route shown by the B/33 Morning Report. The 2d and 3rd platoons of B/33 were attached to 31st Tank Battalion at this time. However, the B/33 Morning Report locations do not place them at Épernay.