Duplex .22-06 M1 Garand at RIA

During the 1950s, the US Army ran a series of programs trying to find a better solution for infantry rifles than firing single semiauto bullets. These projects (including SALVO, SALVO II, and SPIW) would include experiments with multiple barreled rifles, burst firing rifles, flechette firing weapons, and multiplex cartridges. That is, single cartridge cases with two or three bullets stacked inside. The idea was that a soldier would have a single sight picture, trigger press, and recoil impulse – but would have multiple chances to hit the target. This particular rifle is an M1 from the SALVO testing which is chambered for .22-06 Duplex, a .30-06 case necked down to .22 caliber with two 50-grain bullets fired at just under 3000 fps.


  1. At a guess, I’d say the ordnance .22-06 round is probably very close to the .22 Newton, .22 CHeetah (no, the capital “H” is not a misprint), and/or the somewhat larger-calibered 6mm-class wildcats such as the .228 Ackley or .230 Ackley.

    All of these were based on the .30-06 case, and generally delivered MVs ranging from 3100 to over 4000 FPS with bullet weights from 50 to 90 grains. (The lighter bullets delivered the highest MVs, of course.)

    It is reasonable to assume that any of them could have taken a “duplex” load of two bullets in the 45-grain range and delivered about 3000 FPS with them.

    The serial bullet idea isn’t even that new. During the American Civil War, the Shaler sectional bullet was used in the .58 Springfield rifle musket;


    The three sections separated on leaving the muzzle. Together, they massed about 500 grains, or the same as a single “.58″ (actually .575” unfired) Minie bullet.

    BTW, the only complete Shalers generally to be found today are ones taken from unfired rounds. Few have ever been found on the battlefields with all three parts still “together”. So apparently the Shaler slugs worked more or less as advertised.

    It occurs to me that if I wanted a really nasty “multiple-projectile” CQB round for a 7.62 NATO rifle, that still delivered a good whack out to 300 meters, something like the Remington .308 “Accelerator” round would be a good choice. Keep the .223 bullet, but replace the four-section plastic .308 sabot with one made of four pieces of aluminum.

    At close range, it would act a lot like a multiple flechette load. Oh, BTW, a lot of us here in the back country pronounce it “flesh-ETT”, just FYI.




  2. All these years have gone by since the last musket was issued and they are still chasing a replacement for the old “Buck and Ball” loadings.

  3. Never sure if we as taxpayers and non-military get the real deal as to what is issued, why, and what is not.

    Think the world of .30’06, but wouldn’t want to lump it on my back for miles in the snow…

  4. Regarding the WWII stat of 50,000 rounds per casualty, I’d be interested if that was from rifles only or rifles and machine guns. Machine guns and automatic rifles were used for suppression and for just-case-sweeps of likely sniping positions; some rifle shooting was no doubt used for the same end. Does the stat include supply ship cargo sunk by U-boats? On storming beaches, the Marines had a habit of straffing every Palm tree in sight. Every so many thousand rounds a Japanese sniper would fall out of one. And if there were no shortages of ammo and barrels from back home, it was ammo well spent. That should not impune the marksman of the GI’s, and while marksmanship could always be better, it sounds like a statistic engineered to prove a point that someone wanted to make.

    25-06’s have a reputation for burning out barrels, wonder if that was an issue with this clambering as well.

    • Jacob, that “spray and pray” approach would also have made the marines very visible to any Japanese machine gunner with a Type 99 LMG equipped with a scope… Or even worse, any of Imperial Japan’s woefully underpowered (but still very lethal) tanks or a Type 88 heavy AA gun.

      A reversed situation in a war manga had a Japanese anti-air gunner spray a tree with his dual-mount Type 98 AA guns after a Japanese military band member gets headshot while trying to pick up music sheets from a bombed truck. A US marine falls from the tree, chopped in half clean through the kidneys. Overkill much?

      If you were a marine in 1944 and you did sweep the trees on the beach, what would you do if you had encountered a Type 97 Shinhoto Chi-ha tank before your own tank support came? The tank is probably too close for you to call for a naval gun barrage… and that tank gunner probably shot your flame-thrower guy to bits…

      As for “duplex” bullets, I suppose those would probably be more useful to hunters or security guards at Fort Knox (no Goldfinger references, please!)…

      • Amphibious landings in the pacific left little to the imagination as to where the troops had landed. Any sniper in the trees on the beach knew where they were–does one wait for marines to start dropping then start a counter-sniping exercise? Or dump a few belts of 30 caliber into the trees, just in case, and be done with it?

        At any rate, that was the practice, and the point is not to say if that was the best tactic or not, rather the point is that the 50,000 to 1 statistic sounds like every GI had to shoot 50,000 rounds through their M1 rifle to hit one enemy, but the 50,000 rounds included things like strafing trees, and firing to make the enemy take cover. Bemoaning the hit probability assumed that one was dealing only with aimed fire at enemies in sight.

        • Sorry about the confusion. I was trying to imagine a scenario where there are no snipers in the trees but machine-gunners hiding in disguised underground concrete bunkers half-a-kilometer away having gotten smart enough not to sit in trees in the first place. I would imagine that the Imperial Japanese Army would be crazy enough to put a straw dummy in a tree with a fake rifle of wood and bamboo and let him fall out when the marines shoot his position. Think of all the disappointment that would cause.

          If I were on the defending team, I would make the attacking party paranoid as possible by playing the old “I know that you know that I know that you are playing ‘tag’ with bullets” game. Sure, they would defoliate the beach, but did they get any of my guys? If nobody falls dead from a tree after the initial fusillade, what do the attackers do, call in the big guns and the planes? And if the big guns and the air strikes don’t get any results, then what? Do you nuke the island and have done with it then?

        • One of my HS teachers was a veteran of Korea. He said that countersniping generally was his job in his unit. Once a sniper was localized, the SOP was him using his Ma Deuce and a 250-round belt.

          I’ve always been pretty sure, based on “anecdotal” evidence, that the figures for rounds per EKIA were largely due to MG fire, not rifle fire.

          The same held true to an even greater degree for other armies. The Wehrmacht considered the GPMG (MG-34 or MG-42) the actual “killer” in the infantry squad; the riflemen were there mainly to give it flank protection and cover it in a bounding advance.

          As for the Red Army, they had entire divisions that were basically armed with nothing but SMGs (PPSh-41s, for the most part), backed up by LMGs and some Maxim HMGs. They had those 71-round drums for a reason; the “Pay-Pay-Shay” could burn through one in a few seconds, and the infantry were trained to use them in both short and long bursts a lot like a shotgun at close range. They didn’t care how many rounds it took to drop an enemy soldier as long as he went down and stayed that way.

          As for the modern day, SALVO may have “proven” that aimed fire was no more effective than a “sustained firescreen”, but one not-so-obvious side effect of its research is that while the 5.56 round in its various iterations is certainly an effective one-shot killer at CQB ranges (defined as under 200 meters), beyond that it sheds enough velocity and energy that it’s more of a wounder than a killer. That was considered acceptable back in the Sixties, as the belief was that a wounded man was a logistics problem for the enemy.

          The trouble is, the enemies we’ve actually ended up having to fight don’t see it that way. A wounded man tends to just keep on fighting, and as such more efforts- and more rounds- have to be expended to actually make him stop fighting, i.e. kill him.

          I think this may go a long way toward explaining the “extravagant” expenditures of ammunition needed to neutralize a single enemy soldier since the 1960s. More the enemy’s mindset than anything else, coupled with a ballistic characteristic of our rifle rounds that wasn’t considered a handicap when they were originally conceived.

          As I said, just a guess, take it for what it’s worth.



          • “As for the Red Army, they had entire divisions that were basically armed with nothing but SMGs (PPSh-41s, for the most part), backed up by LMGs and some Maxim HMGs.”
            But it was never intentional. Due to German offensive in 1941 big quantity of self-loading rifles was lost, so the fire-arm plants was ordered to produce as many fire-arm as possible – because Mosin rifle and PPSh production was well-known this weapon system were produced in big numbers. The pre-war штат (i don’t known English name – document describing how unit is organized, what and how many equipment it have) for Rifleman Division – штат 4/400-416 (used from 5 April 1941) states that three Rifleman Regiments consist of:
            188 officers, 437 NCO, 2557 soldiers
            and should have been equipped with following weapons:
            441 automatic pistols
            313 sub-machine guns
            1301 rifles & carbines
            984 self-loading rifles [category separate from “rifles & carbines”]
            116 full-auto rifles & LMGs
            54 MMGs
            6 AA machine guns
            3 AA machine guns, 12.7mm-caliber
            27 mortars 50mm
            18 mortars 82mm
            4 mortars 120mm
            12 cannons 45mm
            6 cannons 76mm
            As you can calculate Rifleman Regiments were supposed to have more than 7 times more rifles (both repeaters and semi-auto) & carbines that sub-machine guns

  5. Bought the ex-wife a Savage in 25-06 about 15 years ago. It has seen quite a few rounds fired through it, granted only 1500 to 2000 at most, but it shows no signs of burning out the barrel. This was also with factory loads that are not very hot.

    • others here will have more exact figures than I can remember, but you are looking at a barrel life requirement of the order of 10K rounds for a service rifle.

      I guess that the .25-06 makes sense as a flat shooting rifle for someone who’s not into heavy recoil.

      For anyone who doesn’t mind the additional recoil, a .270 win, will consistently achieve higher muzzle velocity, with a heavier bullet that has as good or better ballistic coefficeint than the .25-06.

      and a .280 Remmington will more than match the .270, but do it at around 2/3rds of the chamber pressure of the .270 (40,000 PSI compared to around 65,0000PSI for a .270)

      I dread to think what a .22-06 was like with barrels, or on a hot day.

      • “a .270 win, will consistently achieve higher muzzle velocity, with a heavier bullet”
        Why? The .270 Win and .25-06 have similar case capacity (4.3 cm^3 vs 4.26cm^3) and maximum pressure (65000PSI vs 63000PSI) so I except that lighter bullet should have bigger muzzle velocity than heavier, why it is not true?

        • 31.67mm^2 for the basal area of the 6.35mm, compared to 36.108mm^2 for the 6.78mm diameter.

          That’s a 20% increase in basal area for the 65K PSI to act upon, even assuming that all else (eg resistance to gas flow through the different bore sizes and heat loss to the barrel) is equal, that’s a worthwhile difference.

          feel free to check the loading manuals or Barnes COTW for velocities with different weight bullets.

  6. so, if it was such a serious improvement, why did nothing came off it? barrel wear? ammo price? or just clueless decision makers?

    • 7.62 NATO duplex ammo (70 grains per bullet) was issued in the mid-1960’s and used in Vietnam. It was then dropped. There does not seem to be any indication it helped in the real world, and going to 5.56 made it a moot point.

      • About the same time, there was the 4.32mm Serial Bullet rifle from AAI, based on the XM 70 SPIW prototype;


        It fired either a flechette or a “duplex” round with two 30-grain bullets. They also experimented with three-shot burst limiters.

        It occurs to me that a duplex-bullet round plus a 3-shot burst would send six slugs downrange. The question is how many of the six would it take to put one enemy soldier down and keep him that way?



        • If a headshot, just one. If a shot to his lungs, heart, or kidneys, perhaps ALL of them! And don’t get started on aiming at the bad guy’s [unmentionables].

          • Just bear on mind that fellow on other side of range thinks the same of you; you can actually shake each other hands :-))).

      • Probably for insufficient separation. Same problem allegedly suffered Colt’s duplex of ACR era.
        To get this thing done properly it needs (besides of fluid dynamics modelling) appropriate amount of wind tunnel time. Early and controllable (repeatable) separation is the key.

  7. Great topic and some illuminating comments. It would be extremely interesting to know how effective a semi auto lower capacity magazine gun like this might be compared to a conventional assault rifle configuration if other factors like reliability and sights were equal. I could see triplex or duplex loads having some use with deer cartridges used on pigs or other medium sized vermin. Cost might be an issue though I guess.

  8. It looks I missed boat in this discussion; at least I can read other’s comments….

    I consider multiple projectile cartridge to be a pertinent and yet to this day un-materialised project. The idea behind it is sound, but implementation is difficult. In my own view, this combination of light projectile and high velocity is not really the best way how to carry on with it since light projectile has limited terminal effectiveness and rapidly looses speed.

    Also, the key to the success is early separation. In late Colt’s duplex round this apparently even did not occur due to efficient air-flow around ogival shape which was preventing it. To have it work, it need some necessary amount of negative pressure on back of each projectile in combination with acceptable air flow, meaning low drag. Hence, each individual part needs to have suitable mass to carry on with momentum.

    Other than that this is a phenomenal way how to add to hit probability. It can be further combined with sub-caliber version, out of same casing. It just needs more work and optimization.

  9. Just to bounce off the G11 remark…..

    In this case, the (otherwise technically brilliant) conduct was dashing blindly ahead of conceptual thought. It created (new and unnecessary) problems, which bogged viability of project inevitably down. Even if resurrected today, it would not mean any major new contribution. Just look what happens to rifle at the end of salvo; it kicks like mule – enough to say.

    This is what happens when you let technicians go ahead of science.

  10. I heard about the duplex principle being suggested around the era of Vietnam “I didn’t personally, I’m not old enough” But I mean, I read about it somewhere- In regards two bullet cartridges being fired from a M16, the idea being increased hit probability via two bullets being better than one so to speak. Kind of what the AN94 tried to do mechanically, as is my understanding. Anyway, I thought about putting say three fletchettes “available in certain U.S states apparently, for novelty 12g ammo” inside a .30 cal-Twenty two cal internal diameter sabot “which you can also get in the states” in a staggered fashion to facilitate stacking the fletchettes flights, loaded into 7.62x33mm ammo for use in say a .30 carbine “for the purposes of creating a spread of shot, per shot type thing in order one at least might hit at range if the round was fired under stress etc” Fascinating… Hey! How about metal storm type rounds, but three in say a 12g sized cartridge- Loads as per from a normal magazine but fires three. All good, he he 🙂

    • So the duplex principle was actually around earlier than “Nam” was my point, of interest if you will…

  11. As it happens, I’ve been reading up on SALVO I, SALVO II and SPIW (Special Purpose Individual Weapon) projects. They did actually issue duplex ammo for NATO for the M14 in the early 60d but it seems to have disappeared without a ripple.

    The duplex idea “works” given certain parameters. 1) You just want to hit the target, not necessarily do a lot of damage. Duplex rounds are obviously less massive but traveling at around the same velocity so their individual joules on target are just those of a the light round. In terms of damage, having two smaller round hit the same target near each other has more stopping power and lethality than a single unitary round (hence the “salvo” idea of simultaneous firings in with gave the SALVO projects their names) if only one hits you have only a light round hit. 2) They tend to be optimal only in middle ranges. In near ranges, the sub-rounds are to close together to give an increase in hits while at long range they diverge to much. 3) Duplex rounds have as much recoil as whatever unitary round they replace so they don’t help keep on target as would firing two lighter rounds individually.

    These days, with up gunning in response to improving body armor, I think the idea is a dead one, except possibly the idea of striking the same spot in armor in rapid succession which turns out to be (to me anyway) surprisingly effective if you can pull it off.

    I think what killed the SALVO and SPIW flechette weapons (beyond the Rube Goldberg piling on of everything short of the kitchen sink) was that the M16 turned out so good a weapon as to make any possible improvements using radical technology simply not worth the effort. It is reasonably light, highly accurate, highly controllable, had reasonable punch and with with a three round burst, stood a good chance of putting two rounds in the same target just like SALVO was attempting to do by more elaborate means.

    In any case, I think the entire idea of trying to replace marksmanship with technology was flawed both in its conception and the driving presumptions. Conceptually, the idea that greater accuracy and more projectiles would equal more hits was on shaky ground. The studies in the 50s showed that 1) The difference in hit rates between elite marksmen and ordinary soldiers in ordinary combat (as opposed to formal or informal take-your-time-sniping) was shockingly small. And that 2) the controlling variable in hit probability was really the body area the enemy choose to expos and how long they exposed it. In other words, nothing a friendly soldier did or did not do affected hit rates as strongly as what the enemy did. (Think of it this way, if the enemy stays down in the foxhole, you can’t shoot him even with a magically accurate rifle, on the other hand, if he just stands there 10ft in front of you, you can’t possibly miss with any ordinary weapon.) Even if new tech could turn every soldier into an elite marksman, the actual effectiveness would not increase.

    The oft repeated assertion that it took 50,000 rounds to kill and enemy soldier sounds increasingly to my ears like military folklore, one of those things that crops up in every field when some assertion is so oft repeated no one checks the original source. For one thing, I’ve seen French posters dating from the end of WWI making the exact same 50,000 rounds assertion. At best it started out as some sort of synthetic statistics that can’t be taken as literally true e.g. “life expectancy” statistics are a synthetic single number expression of the health of a population. When you read that some time in the past or some poor country now has a average “life expectancy of 37 years” that doesn’t mean people 38 and older rarer than hens teeth. It’s just a single number representation created by researchers to to combine lots of data into a single representative number. Yet a lot of people take it literally, especially when reading about life expectancy in history.

    I expect the 50,000 rounds per enemy started the same way, some attempt to boil down to a single number just on how much to expenditure of weapons, of all kinds from bayonets to A-bombs, it took kill an enemy soldier, Probably somewhere along the line, they converted crewed weapon rounds, artillery round, aerial ordinance etc each into an equivalent number of rifle rounds (probably based on cost or resource expenditure) then used that as the military versions of “average life expectancy.” Then the number got broadcasted to those who didn’t know it was synthetic, and who took it literally and tried to make super rifles to compensate.

    It’s pretty obvious in studying all kinds of fire fights on land, sea and in the air with all types of weapons, that the vast majority of rounds expended are suppressive fire so we would expect that the hit ration would be ridiculously low because most of the time, we aren’t shooting to hit, where shooting to deny area and mobility and if the enemy isn’t plain stupid or epically unlucky, he’s safe but neutralized.

Leave a Reply

Your email address will not be published.