There is a discussion of the War Department test in Major Julian Hatcher's book
Textbook of Pistols and Revolvers, written in 1934 and now available in reprint. I was just going through my copy, and the chapter on stopping power. As Hatcher is very careful to give his sources, I was able to find online and download a summary of the test results contained in the following book by Colonel Louis La Garde, U.S. Army Medical Corps, who was one of the two officers involved in the test. The other was Colonel John Thompson, inventor of the Thompson Sub-Machinegun. You can download the 1916 edition of the book from archive.org,
https://archive.org/details/gunshotinjuriesh1916laga, with the discussion beginning on page 69.
Back when it was the Department of War, a study was commissioned on the effects of various calibers, so they purchased a few cows to shoot.
They shot ten rounds of .22 into the side of a cow, and it only made it turn around to see what all the noise was about.
The following were used in the course of the tests, with the smallest calibre of weapon being the .30 Luger.
The Board fired altogether into ten cadavers, sixteen beeves and two horses.
One of the results of that study was the determination that it would take a 3 inch shell to stop someone if they were shot in the arm or the leg, so you'd have to be hauling around a cannon.
For shattering the long bones in the arms and legs, the most efficient cartridge was found to be the British .476, the largest cartridge used in the test. The .45 Colt also performed well, the smaller .38 and .30 caliber cartridges did not perform adequately in the view of the testers.
Soldiers are trained to aim for the center of mass because it's effective. Trying to target limbs in the middle of a gun fight is more likely to get yourself killed.
The following quote appears on page 69 of the book. The service rifle involved would have been the .30-40 Krag firing a round-nosed full-jacketed bullet.
Colonel Winter and Captain McAndrew, Medical Corps, U. S. A., have related the following incident to the author which bears upon the failure in stopping power of our service rifle: In 1907 a Moro charged the guard at Jolo, P. I. When he was within 100 yards, the entire guard opened fire on him. When he had reached within 5 yards of the firing party he stumbled and fell and while in the prone position a trumpeter killed him by shooting through the head with a . 45-caliber Colt's revolver. There were ten wounds in his body from the service rifle. Three of the wounds were located in the chest, one in the abdomen and the remainder had taken effect in the extremities. There were no bones broken.
The Krag had a 5-round magazine located on the side of the receiver, so it can be assumed that none of the soldiers firing had time to reload. If the guard was as large as 10 men, 50 shots would have most that could have been fired. Ten hits shows a very high level of accuracy on the part of the soldiers shooting.
One other factor taken into account by the testing group was range.
Revolvers and pistols being short-range weapons, 75 yards were
agreed upon as the extreme range, 37 1/2 yards as the medium range, and near the muzzle as close range. Simulated velocities were used for the first two ranges.
I will have to see about extracting this summary of the report from the book and posting it somewhere. The entire book looks to be interesting reading as Col. La Garde includes a considerable amount of data from casualties in World War One.