In part II, I will discuss today's range practice with my Uberti reproduction of a Colt New Model Army revolver, often called the Colt 1860 Army today. I wrote a few weeks ago about comparing Johnston and Dow combustible cartridges with Kerr cartridges in my Colt (see the article HERE), but that shooting was all done from a rest to take the human factor out of the equation as much as possible in order to focus on the ammunition. In this practice session I used only Kerr combustible cartridges, and all shooting was done offhand in order to test and improve the accuracy (not precision) of my shooting. In addition to normal shooting, I also shot a table of fire using the "snap shooting" technique described in Captain Stanhope Blunt's Instructions in Rifle and Carbine Firing for the United States Army from 1889.
Given the fairly rudimentary sights on Civil War revolvers (especially Colts!) and their lack of adjustability we have to ask how important they were in actual combat, as opposed to in target shooting. Captain Blunt suggested ignoring them completely and learning to aim instinctively, as if pointing your finger, saying:
832. Owing to the unsteady support that the hand gives to the weapon the methods of aiming previously prescribed for the rifle and carbine cannot be advantageously followed; this is especially true of the practice mounted, where the motion of the horse and the very limited time available for the delivery of the fire permit neither the steadiness nor deliberation so requisite for success with the other arms.
833. The best results will then be obtained by following the method of snap shooting; for which the pistol should be held raised and then quickly projected at the mark and fired without pause or any effort to align it upon the object, the action being somewhat similar to that employed in throwing a missile from the hand and from the same raised position of the arm. (Blunt 1889 p. 309.)
Blunt gave the following directions:
838. The clasp of the thumb and second and third fingers should be firm, the first finger being on the trigger and the little finger underneath the end of the handle. If the clasp is too high up on the handle the muzzle will be elevated; if too low, the muzzle will be depressed. The clasp should not be so tight as to communicate tremor to the pistol, yet sufficiently firm to sustain, when firing with ball cartridges, the force of the recoil. After the discharge the position of raise pistol will be resumed.
839. These motions will at first be executed rather slowly, the instructor correcting the positions if necessary, and the motions quickened as the soldier acquires the habit of leveling or projecting instinctively the pistol in the same man-ner that the forefinger would be pointed at an object. (id. p. 311.)
I strongly suspect that this method, or something very like it, was the norm in combat during the period, and, indeed, beyond, as it presages the later "point shooting" technique taught by Fairbairn, Applegate, Sykes, and others during WWII. I do not practice this technique as often as I should, but plan to focus more heavily on it in future. Normally, when practicing snap shooting I don't determine a String Test measurement (see below), instead opting to draw a 14 inch circle around the aiming point and simply calculating the number of hits within that circle since it's not a target technique, however, today I neglected to prepare any targets with circles, so I chose to use the string Test, and was pleasantly surprised by the results.
My ammunition was combustible cartridges with nitrated paper shells containing 25 grains of Schuetzen 3F powder under a Kerr bullet and dipped in a mixture of beeswax and tallow taken from the 1861 Ordnance Manuals. To learn more about how I make combustible cartridges, read the article HERE.
I judge the accuracy (as opposed to precision) of my shooting using the 19th-century method called the String Test, a system vastly superior to just looking at the group size since it takes both group size and the distance from the group to the intended mean point of impact in a single figure. To learn how to use this system and why all historical shooters should do so, read the article HERE.
My Colt NMA, ammunition, and accouterments. |
Shooting Conditions: Lytle Creek Range, bright and sunny, 42 degrees, wind 9 mph (gusting to 22 mph) from 10:00, 52% humidity, barometer 30.08 inHg.
All shooting was done offhand at 15 yards using a full sight while aiming off to adjust fire. Table of Fire Five was done using the above-mentioned snap shooting method. Note that this article begins with Table of Fire Four because the first three Tables of Fire were done with my Colt conversion revolver, as discussed in part I of today's range report.
Table of Fire Four. |
6 rounds, string measurement 17 inches.
String Test: 2.8 in./rd.
Table of Fire Five: Snap Shooting. |
6 rounds, string measurement 19 inches.
String Test: 3.2 in./rd.
Note how the line of shots track down as I sought to overcome the first high shots, but how I drifted left in so doing.
String Test: 1.9 in./rd.!!!
String Test: 2.5 in./rd.
String Test: 2.6 in./rd.
6 rounds, string measurement 14.5 inches.
String Test: 2.4 in./rd.
Conclusions:
I don't normally write blog posts about routine practice sessions unless I am conducting some kind of comparison or experiment, but I thought this session was sufficiently interesting to warrant posting it because the Colt NMA is new for me (this is my first actual practice session with it) and because of the snap shooting trial explained above.
String Test: 2.4 in./rd.
Conclusions:
I don't normally write blog posts about routine practice sessions unless I am conducting some kind of comparison or experiment, but I thought this session was sufficiently interesting to warrant posting it because the Colt NMA is new for me (this is my first actual practice session with it) and because of the snap shooting trial explained above.
As I have said before, I believe (although I can't prove it) that snap shooting of one sort or another was far more common in combat than any kind of aimed target-style shooting. What I found particularly interesting today was the rather excellent String Test I achieved, which reflects a lot closer hitting than I usually get. Although I have never done a String Test with snap shooting before, believing that since it's combat shooting all that counts is the ability to hit a man's chest somewhere, I can still see from previous experiences that the groups I got with my Remington New Model Army were larger than the group I got today with the Colt--and believe me, as someone who vastly prefers the Remington, that's hard to write. I attribute this to the better balance of the Colt in my hand, a factor that really seems to matter a lot with snap shooting. I plan to practice this a lot more.
The other shooting results pleased me as well. This Colt shoots closer to point of aim thand does my Colt Navy, which means that I don't have to guess in aiming off, which leads to better consistency, which is the key to a good String Test result. My personal standard of acceptable accuracy is a minimum of 3 in./rd. at 15 yards, and with my Remington I have been consistently getting sub-2 inch results, which I consider to be not only good, but probably at or near the limit of the accuracy of the piece. I consider Colts to be inferior weapons in almost every way, so to come close to that kind of result today (albeit with only one sub-2 inch result) is extremely gratifying.
Still, the only easy day was yesterday, and the only score that's good enough is one better than any you've done before, so I have a lot more practice to do.
No comments:
Post a Comment