Light Rifle, Part IV: The M1 Garand Learns To Rock And Roll

The Remington-developed T22E2 (top) and its ancestor, the T22. Image source:

This is the fourth part of a series of posts seeking to describe and analyze the 7.62mm Lightweight Rifle concept promoted by the Americans, and subsequently adopted by NATO in various forms. This series will cover development from before World War II to the present day, but will focus primarily on the period from 1944-1970, which constitutes the span of time from the Light Rifle’s conception until its end in the United States with the standardization of the M16. This article itself deals with the fully automatic variants of the M1 rifle developed before US focus shifted to the then-new, shorter .30 Light Rifle cartridge that led to the 7.62 NATO caliber. Therefore, all rifles covered in this article were chambered for the standard .30 M2 caliber, and the series of fully automatic M1 derivatives chambered for the .30 Light Rifle experimental round will be covered in a subsequent installment. In more than one way, this is the first “true” installment of the Light Rifle series, as the three preceding articles can be considered prologue material, though that does not reduce the importance of their subjects. My readers should also note that while I consulted a variety of sources to write this article, my narrative heavily relies upon Bruce Canfield’s magnum opus The M1 Garand Rifle, as well as R. Blake Stevens’ U.S. Rifle M14 from John Garand to the M21. Indeed, the title of this article is adapted from the third chapter of that latter book, as I could think of nothing superior.

You can read the other parts of the series by following the links below:

Until this point in the Light Rifle series, all of the development we have covered has been essentially background information. The relevance of projects like the Pedersen and Johnson rifles will become clearer as we go along, but until now we haven’t actually begun to talk about the U.S. Army’s decade-and-a-half long search for a select-fire full-power service rifle. This article marks our introduction to this program, but it’s also where a comfortable narrative about exactly what was going on that led to these weapons begins to break down. As we’ll see, Army Ordnance would become very familiar by the end of the Second World War with full-size select-fire infantry rifles, but despite their obvious flaws and the sheer impossibility of meeting every desire outlined by the Army, the program would carry on for another twelve years after the Japanese capitulation. Why did they do this? There is no clear or satisfying answer to this question. In my research, I’ve found that I’ve grown more, not less, frustrated by this and other questions, and I suspect readers, too, will find themselves baffled at the inability of Army Ordnance to recognize what should have been right in front of their noses.

With that said, let’s dive right in.

The first hints of an issue rifle-derived support weapon in fact predate the development of the Garand gas-operated selfloading rifle itself. In 1919, the War Department put together a Board to determine the ideal characteristics of a new semiautomatic service rifle, incorporating lessons learned from the recent Great War in Europe. In 1920, the Board reported their findings, including this segment:

A self-loading shoulder rifle capable of a high rate of fire for short periods of time, together with the increased accuracy due to the fact that aim will not be disturbed by hand manipulation of the bolt, would add greatly to the firepower of front line troops, and probably meet all requirements as to firepower now supposed to be solved by the inclusion of a Browning automatic rifle in each squad of infantry.

In other words, in 1920 the Ordnance believed that a new self-loading rifle could actually replace not only the bolt-action M1903 and M1917 service rifles, but also the fully automatic M1918 Browning Automatic Rifle, as well, such would be the increase in firepower. This dream of standardizing all weapons in the rifle squad once again on a single type would remain with Ordnance through the 1970s.

When the United States entered the second great war of the century in late 1941, her troops were thusly armed with a general issue selfloading rifle and one automatic Browning rifle per squad. However, the desire to standardize remained. As mentioned in the second installment of this series, Ordnance was interested in finding a replacement for the aging M1918, which later led them to tap Winchester’s lightweight G30R selfloading rifle as a potential solution. If a select-fire M1 could be produced instead, perhaps the goal of the 1920 Board could be met in the process, and one standard weapon issued to all troops.

Before the first half of 1942 had ended, the United States Army was also looking to improve the M1 rifle. In the last edition of this series, we covered some of the proposed changes, including adding to the rifle a folding stock, which could potentially combine the best aspects of both the powerful rifle and the handy carbine. Further, Ordnance had cottoned to the idea of a select-fire rifle, a weapon that in theory could provide the advantages of both the selfloading M1 and the fully automatic, but much heavier, M1918 Browning Automatic Rifle. As early as mid-1942, the stage was set for Ordnance’s “do-it-all” weapon, which officers of the Department hoped could replace the Army’s mix of M1 rifles, M1 carbines, M1918 automatic rifles, and, possibly, even the M1928 and brand-new M1 Thompson submachine guns.

However, converting the M1 to an effective fully automatic weapon was easier said than done, and no less than three houses raced against each other to solve this problem. Springfield Armory, spearheaded by the genius of John Cantius Garand himself, represented the Army’s own internal development of a select-fire M1, while Remington was also contracted to develop their own solution. Winchester, in the shadow of feeling increasingly spurned by the U.S. Army, would independently develop a third select-fire, detachable magazine M1. It is these three programs that are together the subjects of this post.


Springfield Development: The T20

By May of 1942, John Garand had already begun work modifying his rifle to fire either in either the semi- or fully-automatic fire modes. This weapon, which has no formal designation, was an existing M1 rifle that was modified to use a modified BAR magazine limited to 18 rounds capacity of .30 caliber M2 ball ammunition. Testing of this rifle revealed that the additional weight of the cartridge stack of 18 rounds caused it to rise too slowly, leading to the bolt overriding the cartridge and close on an empty chamber. There were three possible solutions: Shorten the cartridge stack front-to-back, which would require a change in ammunition (keep this in mind for later installments in this series), change the magazine – which Ordnance rejected as it was desired to use existing BAR magazines – or lengthen the rear of the receiver. It was this last solution that was chosen for the new rifle design, but for reasons unknown the project was delayed until early 1944.

(As an aside, it is worth noting that to my knowledge, no pictures survive of the original select-fire John Garand prototype. Both Canfield and Stevens misidentify the Winchester prototype designed by Sefried – discussed below – as Garand’s initial prototype. Credit goes to our friend Daniel Watters for noticing this.)

It should be pointed out that all the attributes desired of the new rifle were, simply, impossible to achieve simultaneously. The Army – representing collectively the different forces of Ordnance, Springfield Armory, troops in the field, the Infantry Board, and others – wanted a rifle with select-fire capability, that was shorter (by virtue of a folding stock), lighter, still controllable and with a low enough rate of fire to replace the M1918 BAR, that used standard .30 caliber ammunition and standard BAR magazines, accepted existing rifle grenades and a flash hider, fired semi-automatically from a closed bolt and fully automatically from an open bolt, and – while satisfying all of those requirements – retained at least 85% commonality with the existing M1! Needless to say, creating a weapon that met all of these requirements at once was nothing but a pipe dream, much less doing so within a timeline that would have the rifle ready for combat during the war.

By September of 1944, construction of the first prototype of the new rifle had begun at Springfield Armory’s Model Shop, and the rifle was designated T20.  The T20’s receiver was lengthened vs. that of the M1 by .3125 inches, and was adapted to use 20-round BAR magazines. In addition, the roller lug from the M1E3 replaced the solid camming lug of the original M1, and full auto fire was provided for by a connecting bar on the right side of the receiver which actuated an extension of the sear. A selector lever positioned at the rear right of the receiver in a cutout in the rifle’s stock raised or lowered the connecting bar to allow either semi- or fully-automatic fire. As per the initial Ordnance requirement, the original T20 design fired semiautomatically from a closed-bolt, and fully automatically from an open bolt. A new gas cylinder lock incorporated an integral parasol-shaped muzzle brake to control the considerable recoil forces of the new rifle, and special grooves were machined into the weapon’s barrel to help dissipate heat. It should be noted that Canfield and Stevens disagree on whether this last feature was incorporated from the beginning into the T20, or was a workaround incorporated into the T20E1 after the release of the Aberdeen test report of the T20 to Springfield Armory.


The original T20 rifle. Note the lack of a clip release latch on the side of the receiver, and the parasol-shaped muzzle brake. Image source: Stevens


In October of 1944, the rifle had been proofed and function tested at Springfield Armory, and was sent to Aberdeen Proving Ground in Maryland for testing. However, the evaluators soon found that not only did the open bolt feature not significantly improve the gun’s cook-off characteristics, but that it suffered from poor timing. Depending on where the bolt was in its travel when the shooter released the trigger at the end of a burst, the action might come to rest in either the open or closed position. Because of these problems, the open/closed bolt requirement was dropped.


Other side of the original T20 rifle. Note the selector switch above the trigger guard. The T20’s select-fire mechanism and selector switch would be retained all the way through development, and eventually used in the M14 service rifle. Image source: Stevens


The Aberdeen report on the T20 enumerated the various flaws of the new rifle, but responded favorably to the basic concept, which resulted in the creation of a second type, designated T20E1. This rifle disposed of the open/closed bolt feature in favor of closed-bolt only operation, incorporated minor improvements to the receiver/trigger group interface, and featured an improved magazine retention design. It was also at this point that the requirement for a folding stock was dropped from the program, and a number of other requirements relaxed. The T20E1 also incorporated changes to allow it to use the M81 and M82 telescopic sights, or a grenade launching sight, and the muzzle brake was modified to allow fitment of a bayonet. Unlike the T20, the T20E1 used a pattern of magazine not interchangeable with the BAR, and had no provision for a bolt hold open besides the follower of the magazine itself. Springfield Armory engineers also added a non-removable, adjustable sheet steel bipod to the weapon.


The T20E1 rifle. The parasol-shaped muzzle brake has been modified to accept a standard bayonet, and a non-removable sheet metal adjustable bipod has been added. Image source: Stevens


In late January of 1945, the T20E1 had its go at testing at Aberdeen. Perhaps giving us a rare insight into one of the primary personalities driving the project, Col. Studler reflects in his post-war report Record of Army Ordnance Research and Development, 1946:

During the period of 22 through 26 January 1945, the T20E1 Rifle was tested at the Ordnance Research Center, Aberdeen Proving Ground. This model was complete in all details, and the test results, other than the failures to feed, were exceptional. These failures to feed were caused by the bolt bearing surfaces in the barrel being soft.

Studler continues, somewhat despite himself, to describe the changes that were subsequently made to the “complete in all details” T20E1:

  • Induction harden the breech end of the barrel.
  • Increase length of the bipod to permit greater command height.
  • Redesign the gas cylinder and gas cylinder lock screw assembly to permit ready attachment and removal of rifle accessories.
  • Improve the handguard to prevent charring.

Regardless, the T20E1 rifle was considered a success, and 9 additional rifles were fabricated by April of 1945, and sent to Ft. Benning for troop trials. Studler was in fact so impressed with the new weapon that he authorized the creation of 100 limited production T20-series weapons upon the conclusion of the Ft. Benning trials, as well.


Different muzzle brake designs tested for the T20E2 rifle. The design of the rifle’s brake was absolutely essential to the concept, as otherwise the weapon would be totally uncontrollable in fully automatic mode. Image source: Stevens


At Ft. Benning, both the U.S Army Infantry Board and the U.S. Marine Corps Equipment Board tested the eight T20E1 rifles sent to them by the Model Shop. According to Studler’s Record cited above, testers found the weapons to be simple and gave them overall favorable marks. The Benning tests did, however, outline some changes that would be made to the next iteration of the design:

  • Magazine to be usable in the B.A.R.
  • Magazine catch to be altered in shape to eliminate the possible hazard of being accidentally depressed or damaged during rough handling.
  • Provision for means of retaining the operating slide in the rear position in order that the rifle might be cleaned in the prescribed manner.

With these changes incorporated, the order for production of 100 new rifles was given to Springfield Armory in late April/early May of 1945, and on May 17, 1945, nine days after the capitulation of Germany, the Ordnance Technical committee recommended limited production and procurement of 100,000 of the finalized T20 variant, now designated T20E2, to be used against the Empire of Japan. While there is no known evidence of this being formalized, it is very likely that had the war continued, the T20E2 would have been procured (and standardized) as the U.S. Rifle, Cal. .30, M2.


The T20E2 rifle, very nearly standardized as the Rifle, Caliber .30, M2. While it did not include open/closed bolt select fire capability, it did add a fully automatic fire mode and magazine compatibility with the M1918 automatic rifle, as well as an effective muzzle brake. The use of two much more significant weapons resulted in the stillbirth of this improved Garand. Image source:


The T20E2 was compatible with a unique grenade launching device that fit over the muzzle brake of the rifle, and which was cleverly designed to shut off the brake’s ports, allowing normal grenade launching in the same manner as the existing M1. This same arrangement was used to provide the T20E2 with an effective flash hider, although with either device mounted the brake would be rendered ineffective at reducing the rifle’s recoil. As per the Benning recommendation, the magazine designed for the T20E2 was compatible with the existing BAR, but the reverse pairing was not possible.


The grenade launcher for the T20E2, and its interface with the rifle’s substantial muzzle brake. Note that the launcher blocks the brake’s ports, rendering it ineffective. Image source: Stevens


So serious was Ordnance about adopting the T20E2, that when, in June of 1945 the first 10 rifles were completed, they were immediately sent to Raritan Arsenal for “preparation of Notes of Materiel”, that is, to create a field manual for the weapon. The resulting document was War Department Technical Bulletin TB 9X-115, dated June 28, 1945, and it is replicated in full in Stevens’s book. Despite this expedited timeline, series production of the T20E2 was not anticipated to begin until January of 1946. However, Operation Downfall, the anticipated invasion of Japan, was set to begin at the end of 1945, and was estimated to last at least until 1947. In mid-1945, the T20E2 was set to be the primary weapon of Operation Downfall, and so it was essential to get the rifle ready for action as soon as possible.

On August 6th, 1945, another secret U.S. weapon was used against Japan. Dropped from the B-29 Superfortress heavy bomber Enola Gay, the uranium-235 gun-type fission bomb dubbed “Little Boy” destroyed most of the Japanese city of Hiroshima. Three days later, another B-29, Bockscar, dropped the plutonium-239 implosion-type fission bomb “Fat Man” on the city of Nagasaki. With this, and the commencement of the Soviet invasion of Japan earlier that day, the Japanese Emperor Hirohito decided to surrender to the Allies, accepting the terms of the Potsdam Declaration. On August 15, 1945, Hirohito gave a radio address announcing the surrender, and on September 2, 1945, the Japanese Empire signed its formal surrender aboard the USS Missouri, finally ending World War II.

Just like that, the Atomic Age had begun, and the adoption of the T20E2 suddenly evaporated. That month, the Ordnance Technical Committee shelved the adoption of the rifle, although it did not end the program. In fact, the T20E2 rifles would continue to serve as testbeds for the weapons that would eventually lead to the M14 throughout the late 1940s and early 1950s.


The mechanism of the T20E2, which was carried over to the T44 prototypes (which were themselves based on re-used T20 receivers), and eventually the T44E4 that was standardized as the M14 in 1957. Image source: Canfield



Remington Development: The T22, T23, T24, and T27

In addition to Springfield Armory, Remington Arms Co. was in September of 1944 given the second of two contracts for a new select-fire, folding stock “paratroop” rifle derivative of the M1. Given the requirement for 85% commonality with the existing M1 rifle, the leadership at Remington decided that the most expedient course of action would be to design a rifle that retained as much of the original design as possible, and unlike those at Springfield Armory, Remington’s developments would all retain the standard M1 receiver length. Remington’s project was given the designation “T22”.

As work proceeded in the first few weeks at Remington, two distinct methods of facilitating fully automatic fire became apparent. The first, like the Springfield T20, was to add an extension to the rifle’s sear, which would be tripped as the operating rod reached its frontmost position in the receiver, after the bolt had been locked. The second was championed by Remington engineers Kenneth J. Lowe and Crawford C. Loomis, and utilized a hammer release stud. Switching the selector to fully-automatic caused a specially shaped piece would disengage the hammer from the rifle’s sear, while simultaneously engaging the hammer release surface with a stud on the hammer. Upon the return of the bolt to battery after firing the first shot, the forward surface of the operating rod would contact a connecting bar, forcing the hammer release surface out of alignment with the stud on the hammer, and releasing the hammer, causing fully automatic fire that would continue until the release of the trigger.

To test both of these designs, Remington modified two existing M1 rifles, which became the T23 with the independent hammer release, and the T24 with the independent sear release. The tests showed to Remington the value of the independent hammer release fire control group, and at this point it became obvious to Remington and Ordnance both that the requirements for the project were unrealistic. On a meeting between Col. Studler, Major E. G. Cooper, and Kenneth Lowe on November 1st and 2nd of 1944, the requirements specified in the contract were modified:

  • Weight – not to exceed 12 pounds, less magazine.
  • Eliminate the folding stock requirement, and use the standard Garand stock maintaining the standard Garand overall length.
  • Use the standard Flash Hider but so modified as to function also as a muzzle jump reducer or stabilizer.
  • Use B.A.R. 20-round magazine modified to suit.
  • Use modified design of Bren Bipod.

Early in the life of the first T22 prototype, tests at Remington revealed that the rifle needed significant revision. The fire control mechanism gave unreliable and erratic performance. Like the engineers at Springfield, Remington engineers too found that a fully automatic M1 had a tendency to overheat. The magazine was initially retained only in the rear, which resulted in magazine tipping, causing misfeeds. Other problems manifested alongside these, and the T22’s design was modified, becoming the T22E1 on January 13, 1945.


The Remington Arms Co. T22 rifle. This rifle used a hammer release mechanism designed by Remington, unlike the T20. Image source: Stevens



The T22E1 incorporated a host of significant improvements to the Remington design. The magazine issue was solved by supporting it both front and back, a new pattern of magazine with a heavier spring was designed for the weapon that was compatible with the BAR, but that provided a persistent last-round hold open via a catch in the receiver, revised and improved feed ramps were incorporated into the barrel, cooling slots were cut in the handguards, and a bipod was added, based on the British Bren light machine gun bipod. Then, again on January 29th, the design was modified further:

The arm should fire from a closed bolt in both full and semi-automatic operation;
Instead of locking the gun open on the follower, it is requested that the follower operate a separate latch which would retain the bolt in the open position after the last shot;
Improve the effectiveness of the muzzle depressor and ignore its deficiency as a flash hider, if such results.

Finally, by April 2nd, 1945, the T22E1 was ready for testing at Aberdeen Proving Ground. Kenneth Lowe and Crawford Loomis personally delivered the prototype to the testing facility, where it was evaluated by firing six thousand rounds of .30 M2 AP ammunition over four days as part of routine testing. The personnel at Aberdeen were impressed, according to Stevens called the rifle’s performance “excellent”. Further, the rifle’s new magazine pattern functioned perfectly with the Browning Automatic Rifle, satisfying in part the need to share commonality with the existing automatic rifle type.


The T22E1 rifle, with improved, ribbed magazine, also compatible with the M1918 BAR. Note the Bren-style bipod, and large cup flash hider modified to act as a muzzle brake. Image source: Stevens


On May 4th, 1945, Remington was represented by Lowe, Loomis, and G. O. Clifford at a meeting in Washington with representatives from Ordnance (including Col. Studler), Springfield Armory, and Aberdeen Proving Ground to discuss the T20E1, T22E1, and the findings thus far as to the proper configuration of the new rifle, whichever was chosen. It appears that at some point in these proceedings, it was decided to dispose of the superior Remington magazine, and revert to the existing M1918 BAR magazine.

Development continued at Remington from this point until after the end of the war. The improved T22E2 prototype was completed by late October, 1945, and submitted for testing before the end of that month. Unsurprisingly the test results showed that the rifle’s cyclic rate was outpacing the BAR magazine’s ability to feed rounds into the action, and this resulted in an inordinate number of closures of the bolt on an empty chamber; interestingly, though, this was the only type of malfunction reported!

Further Aberdeen trials of an improved model of T22E2 were scheduled for July 22nd, 1946, and these reportedly gave excellent results. Subsequently, Springfield Armory was given a contract to fabricate fifty T22E2 rifles for field tests, modified with push-through selectors and Remington-designed muzzle brakes (these were used instead of the T20E2 brakes because of a shortage of those components). Oddly, at one point the new rifles were specified to fire from the open-bolt position in both semi- and fully-automatic fire modes, however it was found that this was a poor match for the somewhat sticky Garand action, as a failure of the bolt to feed a round could result in an unintended discharge.


The T22E2 rifle, with Remington-designed muzzle brake, and revised push-through selector lever. Note the reversion to using M1918 BAR magazines, a change that caused problems with this type. Image source: Stevens


The T22 program concluded with Lowe’s report Light Automatic Rifle Caliber .30 (Paratroop Rifle) Models T22, T22-E1, T22-E2, and T22-E3, submitted on October 15, 1946. However, in February of that year, Col. Studler awarded to Remington a contract to design a select-fire kit that could be installed in existing 8-shot en-bloc clip loading M1 rifles, sending to the company an example of such a rifle designed by John Garand utilizing the sear release mechanism of the T20. Studler’s reasons for requesting this kind of weapon were that existing rifles could be retrofitted to train troops for light automatic rifles, and that fully automatic capability, even with an anemic 8-shot magazine, would still be an improvement in firepower for the squad. Design study for this rifle, named “T27”, began on July 15 of 1946, and the rifle was demonstrated on September 20, 1946, more than a year after the end of the war. The T27 utilized the same mechanism as the T22E2 (with which the components were interchangeable), and was a technical success, requiring only basic field-stripping to give the access needed to modify the rifle’s components for select-fire. Minor improvements made to the Remington mechanism as part of the T27 program were rolled into the T22’s design, as well, resulting in a rifle designated “T22E3”, which was never built.


The T27 rifle, designed by Remington Arms Co. to provide a field kit to retrofit existing M1 rifles for select-fire capability. This improved on the T22E2’s select-fire mechanism, and its design was reincorporated into the T22 program resulting in the T22E3. Note that the rifle does not have a detachable box magazine, retaining the 8-shot en-bloc clip system of the M1. Image source: Stevens



Winchester Development

Legendary Winchester CEO Edwin Pugsley was aware of the development that was happening at Springfield and Remington (both still in their infancy), and felt that if the company did not produce its own select-fire M1 derivative, it would be unable to secure future rifle contracts. Further, Pugsley determined that if Winchester could solve the problem of making a successful select-fire infantry rifle ahead of the other two initiatives, then it could win the contract and have a substantial lead over its competitor, Remington. Pugsley put then-Winchester engineer Harry H. Sefried II in charge of the project. In a memo dated June 22nd, 1944, the first example Sefried’s M1 variant utilizing a 20-round box magazine was documented having fired 135 rounds, with two failures to feed. Sefried’s design utilized a modified, cut-down M1918 BAR magazine, a front-mounted magazine latch operating from a modified set of M1 Garand lockwork with the follower and follower arm for the en-bloc clip system removed, and a sear trip lever actuated by a connecting arm activated by the return of the operating rod to battery.


The Winchester select-fire Garand prototype. This rifle is often misidentified as John Garand’s original select-fire prototype, but it was in fact designed by Harry H. Sefried II. Image source: Stevens


By July, Pugsley had informed the Chief of Ordnance’s Small Arms Division, Colonel Rene R. Studler, that Winchester was pursuing independent development of a select-fire M1-derived rifle. However, Studler’s attitude towards Winchester at this point seemed cold and uninterested. In the transcript of a telephone conversation between Studler and Pugsley in August of 1944, Pugsley asks Studler why Ordnance won’t test their rifle design, and the Colonel responds that the design isn’t finished enough for tests at Fort Benning – saying nothing of Ordnance testing the gun themselves. Studler then told Pugsley that more features, such as a bipod and flash hider, should be added to the design, while not expressing any interest in Winchester’s mechanism itself. Further reinforcing the idea that Studler was dismissing Winchester, he changes topic to the M2 Carbine (which was designed without Winchester’s involvement, or even knowledge), and refuses to send them a sample for testing, but offers to demonstrate it at Springfield Armory instead. The signals received by Pugsley must have communicated to him that Ordnance was uninterested in Winchester’s project, preferring to pursue its own T20 and Remington’s contract.


Harry Sefried’s select-fire M1 design, field-stripped. Note the sear extension and the forward-mounted magazine release, both highlighted by arrows. Image source: Stevens


Still, Pugsley was dedicated to perfecting the select-fire Garand ahead of Springfield Armory and Remington, and on November 21st, Pugsley demonstrated the select-fire Garand, Winchester Automatic Rifle, and M2 Carbine (which Winchester was selected to produce) prototypes at Aberdeen Proving Ground. This demonstration revealed that when firing .30 cal M2 Armor-Piercing ammunition, the soft fronts of the magazine could be dented by the severe recoil impulse of the rifle, combined with the hard bullet tips of the AP cartridges. The magazines used in this test were modified M1918 BAR magazines, and were not designed to resist the much more extreme recoil impulse and rate of fire (955 rounds/min versus 350-450 for the M1918) of the modified Winchester M1. Winchester’s earlier testing had uncovered this problem, but they did not know that the Army planned to use M2 AP exclusively in theater until that point. Prior to learning of this, Winchester engineers had suggested that treating the M1918 magazines as disposable could circumvent the problem. However, their testing thus far had been with lead-cored M2 Ball ammunition, where hardened-steel-cored M2 AP would dent a brand new magazine sufficiently to stop the cartridge stack from rising with only a few shots.


An image from Sefried’s patents on the Winchester select-fire Garand. Note that the rifle described here is identical to the one pictured above. Image source:


As an amusing aside, the lead project engineer Harry Sefried retold a story about Winchester’s select-fire Garand to Charles E. Petty, who authored an article for the 1983 March/April issue of American Handgunner:

One of his first projects at Winchester was to develop a modified M-1 capable of full-automatic fire. Sefried, Roehmer, Pugsley and Williams took the model to Aberdeen Proving Ground to demonstrate it to John Garand and a host of army brass. In honor of his position, Garand was given the first opportunity to fire the gun. Before Sefried could explain the light trigger pull and 1,000-round-per-minute cyclic rate, Garand raised the rifle to his shoulder, touched the trigger and. .. you guessed it! While the assembled dignitaries scrambled for cover, Sefried grabbed Garand and held on. He says it was the high point of his career-“to have John Garand by the ass.”

After the demonstration, Col. Studler continued to play coy with Winchester. In mid-December of 1944, Pugsley inquired of the Colonel whether Ordnance would still be interested in their select-fire M1, on which work was still well underway. Studler, who was surely unimpressed by the Winchester gun with its magazine issues, low total round count to that point, extremely high rate of fire, and lack of features, responded noncommittally on December 20, 1944:

Concerning your letter of 12 December 1944 and in particular the matter of the Browning Automatic Rifle magazine your conclusions as to the effect of this item on functioning agree with ours. It seems to be quite evident that an improvement in the magazine is required for correct functioning of this weapon.

In answer to your last paragraph, we will of course be glad to arrange for tests of any further developments which you may accomplish.

Interestingly, readers will note that at this point only the initial Springfield T20 prototype had been produced, and was itself having problems. Despite Studler’s non-answer, Winchester continued to work on the firearm through 1945, as evidenced by a June report sent by Winchester to Colonel Studler. Most likely, given Pugsley’s tenacity for a contract, Winchester development of this weapon ended with the end of World War II, however the company will re-enter our story later on in the series.


To Conclude

Thus, we have seen that at the end of World War II, or shortly thereafter, Ordnance had in its grasp no less than four rifles (T20E2, T22E3, and Winchester’s select-fire M1 and the Williams-designed G30R discussed in Part II) that offered what they were looking for in a next-generation infantry weapon. These rifles were chambered for standard .30 caliber ammunition, and any one of them could have been tested, adopted, and produced before the end of the 1940s. However, as we’ll see Ordnance would choose “perfect” over “good enough”, as a future installment will cover the fateful decision to abandon the .30 M2 caliber in favor of an entirely new round. This, combined with the collapse of another program, and the “life support” level of funding that the Garand improvement program would receive post-war, would cause a 12 year delay in fielding a weapon of this kind.



It should be noted that one enemy weapon possibly had a tremendous amount of influence on the Army’s search for a select-fire infantry rifle during this period. The German Fallschirmjägergewehr, or FG-42, was a select-fire weapon in the German standard 7.9x57mm cartridge, and had proven to be surprisingly lightweight and controllable, even on fully automatic. The weapon, which was designed for Luftwaffe paratroops building upon experience in the Battle of Crete, operated from a closed bolt in semi-automatic mode and an open bolt in fully automatic mode. According to several secondary sources, the FG-42 was received by US testers with great enthusiasm, and the timing of at least one of those tests was fairly conspicuous. Given the specific requirements of the U.S. program, and even the name given in the initial solicitation, which was for a paratroop rifle, it seems highly probable to me that the FG-42 gave the Ordnance Department all the justification it needed to set such ambitious requirements. After all, if the Germans could produce such a weapon, why couldn’t the United States? And if such a weapon could be produced, why shouldn’t all U.S. troops be armed with it? The above remains entirely my own speculation, but it’s possible – probable, even – that the FG-42 paratroop rifle represents the true first light rifle, and that it was this German weapon that so thoroughly convinced the Army that a select-fire, full-caliber lightweight automatic rifle could be made a practical reality.

4632-SA.A.1 (1)

An early-model FG-42 paratroop rifle. There is not enough direct evidence to conclusively determine that it was this weapon that convinced Ordnance to pursue the path of a select-fire, full-caliber infantry rifle, but in this author’s opinion that is highly probable given the circumstantial evidence. Note the date of the photo, 9 August 1944. Image source



Nathaniel F

Nathaniel is a history enthusiast and firearms hobbyist whose primary interest lies in military small arms technological developments beginning with the smokeless powder era. In addition to contributing to The Firearm Blog, he runs 196,800 Revolutions Per Minute, a blog devoted to modern small arms design and theory. He is also the author of the original web serial Heartblood, which is being updated and edited regularly. He can be reached via email at


  • Riot

    Good article, btw you have “Bred bipod” in one of the bullet points.
    Seems to be another case of dictated/limited magazine design causing problems – remmingtons being discarded even though it was compatible seems daft.
    How common was the BAR, and therefore its magazine, at that time?

    • Fruitbat44

      My understanding is that the BAR was issued at one per infantry squad, what us Brits would call a section, in the US Army which would shortly increase to two per squad in time for Korea. The USMC had in 1944(?) adopted a thirteen man squad of a squad leader and three four-man fireteams, each with three rifles and a BAR. So I guess the answer would be, pretty darn common.

      • Riot

        I’ve been looking a bit, seems when they entered the war it was one per 12 man squad for the US army and two for the marines. Army copied the marine corp at some point. Though there were production shortages apparently.
        USMC had 513 BARs per division upon entering and 867 in 1945.

        That is a lot of magazines.

        • Tom

          Even if the magazines were not exactly common it was a magazine that US Ordance knew worked and knew how to make. Given those factors and compatibility with the existing BAR would be major plus points for using the BAR magazine rather than adopting a new design.

          • Riot

            But allowing improvements to a magazine that was already noted as being flawed should have been considered if the changes in manufacture were small.

          • UnrepentantLib

            It sounds like the main problem, not raising the bullet stack quickly enough, might have been solved just by installing a stronger spring.

          • I actually think the only justification was commonality. Engineers at all three organizations recognized that much better patterns of magazine could be produced, but the cry from Benning was always “why doesn’t it use BAR mags?”

      • gunsandrockets

        A prewar (for the U.S. at least) issue of Infantry Magazine had an article about the development and intended employment of the M1918a2 model of the BAR.

        At that time the Army intended to group multiple BAR all placed in a weapon squad integral to the rifle platoon organization. That’s why the M1918a2 had some curious features like the removable buttstock monopod, for firing on fixed lines of fire. Of course during the war, more typical organization moved the BAR to the rifle squads and equipped the weapon squad with a bazooka and M1919a4 Browning belt fed machine-gun on a light tripod.

    • Thanks for the correction. Over the course of the war, IIRC, they made about 100k BARs, and there was one per squad. So reasonably common.

  • pilum57

    That should be Operation Olympic as the invasion of the Japanese home islands not Overlord.

    • Thanks for the correction. My mistake!

      • Secundius

        @ Nathaniel F.

        It might have been another “Game Changer”, if Japan had introduced the Second of its Atomic Bomb’s in produced in 6 August 1945…

        • jcitizen

          A prominent Japanese businessman was researching atomic power, when the Nippon government noticed the huge electric capability of his generation stations attempting to make nuclear pile grade uranium. They shut his operation down and diverted the power to factories for war production. Thank God they didn’t understand one thing about what this futurist was doing in his lab!

  • Mike


  • Kristoff

    Wow, the US really had a thing for BAR mags. Standardisation and commonality I get it. “Our rifle is working great, General!” “Good, now how ’bout some BAR magazines?”

    • Secundius

      @ Kristoff.

      Try either BMG Parts & Accessories or Numrich Gun Parts. BAR magazine’s came in 20, 30, and 40-round capacities, the latter two might be harder to find…

      • Kristoff

        Can you imagine 40 rounds of 30.06 hanging off your gun? Whew.

  • Joseph Smith

    Another great piece, thanks TFB!

  • TechnoTriticale

    Seems like the concept of a select-fire “light” rifle is close to incompatible with the .30-06.

    It only worked in the BAR because the thing weighed 24 pounds. Even so, in that photo of the 1917 Congress Heights demo (seen on the BAR wiki page), you can see that the lads are really having to lean into the BARs for that offhand live fire.

    • I didn’t have a problem with the BAR offhand, but the weight and slow ROF really helps a lot.

      You’ll note most of these programs (T20, T22, FG-42, AR-10) revolve around the use of extremely large muzzle brakes. It’s possible to tame the recoil of a weapon like that, but with the recoil of a round like .30-06, you need a seriously powerful brake, meaning serious flash and blast.

      • TechnoTriticale

        re: …extremely large muzzle brakes.

        And the ostensible civilian version of the BAR, the Colt Monitor, was often factory-configured with a copious Cutts compensator.

        • Secundius

          @ TechnoTriticale.

          Other then the FBI, There’s NO RECORD of anyone else using the Colt Model 1921/8 Monitor…

          • TechnoTriticale

            re: NO RECORD of anyone else using the Colt Model 1921/8 Monitor

            I didn’t say there was, although saying “often SHOWN factory-configured” might have been less susceptible to misinterpretation. The point was that Colt thought it necessary to include a compensator even to offer it for the civilian market.

            And true, none of the current Ohio Ord. M1918 reproductions include a comp, but then they are semi only.

      • iksnilol

        Also shorter barrel with that muzzle brake. To increase the pressure in the brake (and thus make it more effective).

    • Riot

      The “light” rifle also suffers from the lightness causing a lack of control because certain components being light tends to lead to a high rate of fire on automatic.
      It exacerbates the problem that you have to deal with.

  • I. P. Freeley

    >Why did they do this? There is no clear or satisfying answer to this question.


    When TFB gets awarded their first Pulitzer, I’d like my name to be mentioned as a researcher.

    • Erm, well, Col. Studler made a salary according to his rank, so I doubt he was particularly motivated by money.

      • It has been suggested that Studler had some family wealth as well. Studler’s 1934-1939 tour of European weapon manufacturers would have been difficult on his Army salary alone.

  • TDog

    Great article. And not a bad supposition on your part regarding the FG-42. The US military has had a Wehrmacht fetish for decades….

    • The first serious tests of the FG-42, so far as I can tell, occurred in 1946, but as you can see from the photo, at least some level of evaluation occurred before the end of the war.

      There’s a lot of circumstantial evidence that the FG-42 provided the “shining light at the end of the tunnel” that Ordnance was trying to achieve with the Lightweight Rifle program. A big piece of evidence is that another program, which resulted in the M60, was directly built on the FG-42, and then you have the conspicuous similarity of the requirements (especially open/close bolt operation) to the FG-42, but what’s got me about 99% convinced of this is that name: “Paratroop Rifle”.

      • Secundius

        @ Nathaniel F.

        Actual Field Testing, started after the Fall of Monte Cassino in late May 1944. But Prototype Testing of the T52A3, first of the M60 Prototypes, started in February 1952…

      • FWIW: The Ordnance Research Center’s Foreign Materiel Branch released a memo report on the FG42 “Type E” in June 1944.

        In 1946, Aberdeen ran a “Type G” through the standard light automatic rifle tests.

      • Jrggrop

        Did the US Army ever try to just make a copy of the FG 42, rechambered for .30-06, like their attempts at making an American version of the MG 42?

        • Magazine-fed? Not so far as I know. There was the T44 Light Automatic Machine Gun:

        • Riot

          A .30-06 MG42 was commissioned but the team “forgot” that .30-06 and 7.92 are different lengths.

          • UnrepentantLib

            If Ordnance was impressed with the FG-42 as a paratroop rifle, you have to wonder why they didn’t consider an improved Johnson LMG for the role. Of course, it was invented by a Marine, so there is that difficulty.

          • There’s a pretty strong undertone in a lot of sources that Johnson became a persona non grata because of his type A personality.

          • Riot

            I like the johnson – but it cannot achieve the fg-42s, essentially LSW goals, of accurate semi-auto and capable automatic.
            The design is much more suited to automatic rifle (and much better at it than the bar)

          • Miguel Raton

            The Johnson LMG being a 14# rifle, was a much more portable solution than the 19# BAR, while still being controllable on FA due the added wt and in-line stock vs. the M14/M15 impossible dream. It’s tragic that Johnson was obstructed at every turn by the pinheads running Army Ordnance at the time; if he hadn’t been, the Johnson LMG might have at least had a functional double-stack magazine instead of the awkward single-stack mag he felt forced to use after he’d been sabotaged in one of his tests.

          • Riot

            Well you also have the lewis gun which was batted away multiple times, only to be (terribly) copied in the form of the m60 after it had been adopted by the british and later copied by the germans for the fg42.

          • The folks at GM-Saginaw Steering Gear were tasked with too much when they tackled the T24. Basically, they were trying to convert a
            captured MG42 to .30-06, all while introducing replacement assemblies,
            and attempting to reduce the cyclic rate by roughly half. I suspect that the latter goal caused the most problems.

            would have been better to have split it into a spiral development. One
            party would get a basic MG42 in 7.92x57mm with the instruction to reduce
            the cyclic rate as far as possible while retaining reliability. Another
            party would get a MG42 with the goal of building a straight conversion
            to .30-06. Even if they couldn’t reduce the cyclic rate, a successful
            project could have given troops a way to convert captured MG42.
            Afterwards, you could then see if the lessons from the cyclic rate
            reduction project could be adapted to the .30-06 conversion. Once this
            was wrapped up, you could then look at how to reverse-engineer the design for
            US production.

          • Riot

            I’ve never heard of any attempt to reduce the cyclic rate.
            It also seems an incredibly stupid thing to do – the MG42s selling point is the high RoF – a target of 600 rpm would not choose the 42 as a starting point.

          • They nearly tripled the mass of the bolt and increased the aperture size of the muzzle booster from 0.425 to 0.6″. This knocked the cyclic rate back to just over 600rpm.

        • Secundius

          @ Jrggrop.

          NOT in .30-06! Bullet has a Overall Length of 85mm, almost twice the size of the 7.92×33 kurz Overall Length of 48mm. They did with the 7.62x51NATO though…

          • The T44 was chambered for .30-06, and is an FG-42 with an MG-42 belt-feed mechanism slapped on it.

          • Actually, Bridge Tool & Die Works kept the T44 MG prototype in the native 7.92x57mm.

          • Iiiinteresting, then multiple online sources have it wrong. None of the photos are big enough for me to tell clearly, either.

          • Springfield claims the T44 in their collection is 7.92x57mm. Note the report title: “Government Contract W-36-034-ORD-7638 – Notes on Materiel. Gun, Machine, Light, Cal. 7.92mm T44. December, 1946.”

          • Miguel Raton

            Ah, “online sources.” There’s your problem… 😉

          • The FG42 was chambered for the standard 7.92x57mm cartridge, not the 7.92x33mm. However, at least one FG42 prototype chambered for the 7.92x33mm was discovered.

  • gunsandrockets

    How did the M2 carbine fit into all this activity? Seeing as the M2 was actually produced and saw some service during WWII, I wonder how it fit into the US Army decision matrix for the full-auto Garand development.

    • M2 Carbine was developed by Inland, and it did influence the process modestly. Col. Studler in fact brings up the T4 (which became the M2) in his phone conversation with Pugsley, and mentions that there’s a lot of interest in full auto individual weapons at that point.

      One of the supreme goals of the Army during this time was to reduce the number of guns in service. They attempted to do this in almost every way you can imagine, from dreamland-type ideas about the T20E2 replacing the M1/M2 Carbines, to making an M3 Grease Gun chambered for .30 Carbine.

  • Rock or Something

    I might point out, the U.S. arguable had an Americanized “FG-42” by this time in WWII, the Johnson LMG. It was already in service in limited production with a few select Marine Corps and “Special Operations” units. They even looked similar, although I believe the Johnson LMG was a few pounds heavier then the FG42.

    But of course, it handled much differently then the M1 Garand and didn’t use a BAR magazine, so maybe the Army just wasn’t as keen on a FG42 style rifle after all. (I’m also the guessing there is truth to the rumor that the Ordnance department’s general disdain for Johnson was true.)

    • jcitizen

      It would have been a trick to convert his rotary magazine to a belt fed or BAR magazine rifle. I’ve hefted the Johnson and was shocked at how light it was; but I’ve never had the opportunity to heft an FG-42.

      • There were belt-fed conversions of the Johnson LMG tested after the war.

        • jcitizen

          Interesting! If I’d just get out my “Small Arms of the World” and review this history, I would have known that – but reading this fine article and your posts is much easier – [now where is that book? I need to go find it!] =)

      • roguetechie

        actually no it would not have been a trick.. It would have been two pins and a subassembly swap….

        • jcitizen

          Thanks! That was amazing!

        • jcitizen

          Okay, I checked the history on forgotten weapons, and it looks like a LOT of modifications were necessary, like a bolt bounce spring and new bolt, and a new gas port drilled. This is more significant than a simple swap. This was just one of the modifications, and many more were needed to make either the box magazine or belt fed version work.

      • Miguel Raton

        I think you’re confusing his rifle w/ his LMG. The Johnson rifle used the rotary mag [hence the “pregnant guppy” profile] – it had the nice benefit of being able to be topped up w/ m1903 Springfield 5rd stripper clips. The Johnson LMG, otoh, used a lo-o-o-ng single-stack stick mag hanging off the left side, much like the FG42 [one wonders how the spy train that fed the info on Johnson’s developments back to the Reich worked… Material for a novel there? ;-)] They faced off @ Monte Cassino, reportedly.

        Anyway, Johnson resorted to the long single-stack mag when Ordnance sabotaged his efforts to make his LMG mag-compatible w/ the BAR by supplying him w/ defective BAR mags during development. The frustration paid off: Ordnance didn’t have to accept the Johnson LMG, & our guys kept using the BAR when a better piece of kit might have been available “if only…”

  • Don Ward

    Well done Nathaniel. Although at this rate, you should have begun your Light Rifle series with the opening line “Call me Ishmael”.

    • “Well, Prince, so Genoa and Lucca are now just family estates of the Buonapartes…”

  • midnitelamp

    was there ever an attempt to produce a belt fed BAR?

    • That is the M240.

      • UnrepentantLib

        The Swedes produced a prototype belt fed modification of their M/1937 BAR, but it had some flaws. The belt feed mechanism came up from the magazine well and they had problems when they came to the last few rounds in the belt.

        • FN and US Army Ordnance had also each played with belt-fed BAR variants before WW2. FN even filed US patents.

          • gunsandrockets

            I remember seeing a picture of the T23 LMG in a book written by Johnson. I think Bill Ruger worked on the 1943 project.

          • Miguel Raton

            And if you look at the patent drawings upside down, you can see the M240 [FN MAG], kinda like one of those old pics in MAD magazine… 😉

  • Aptly noted that the brakes closely resemble the AK-74 brake. My guess is if ultimate recoil reduction is your goal, you necessarily end up with something like that.

    Glad the article was thought-provoking!