WW1 2-Inch Trench Mortar

The British Army entered the First World War with very few mortars, and certainly none at the battalion level. As the stalemate of trench warfare set in and the effectiveness of enemy mortars became clear it was decided that trench mortars of various sizes would be needed.

Nicknamed ‘plum pudding’ or ‘toffee apple’ mortars after their projectile’s characteristic shape, the 2 inch Medium Mortar or 2 inch Trench Howitzer, was one of Britain’s first effective light trench mortars to be introduced.

tilt_h13_2_inch_mortar
Right-side view of the 2in Trench Mortar (Matthew Moss)

Trench mortars were the army’s most forward artillery, right up on the front line. These short range weapons were able to throw large, high explosive projectiles, short distances across No Man’s Land at the enemy trench system opposite. The 2 inch mortar was considered accurate out to 350 yards with a maximum effective range out to just under 600 yards.

Introduced in 1915, the 2 inch mortar was originally crewed by men taken from the battalion it was stationed with, along with some specialists from the Royal Artillery. However, with the introduction of the 3 inch Stokes mortar which was operated by the infantry themselves the 2 inch mortars became the sole responsibility of the Royal Field Artillery.

Mesopatamia 4
A young gunner loads a 2in spigot mortar bomb into his mortar (Imperial War Museum)

Mortar positions were often in secondary trenches just behind the infantry’s frontline. This was to help protect the infantry from potential counter-battery fire. The trench mortars were often deployed to sectors to provide counter battery fire against German minenwerfers or in the run up to an offensive or local action. A British Army report on artillery use, drawn up in February 1917, noted that “Owing to their liability to be destroyed by hostile artillery fire it may often be advisable to defer opening fire with these mortars till the last day of bombardment.” The mortars were also tasked with keeping gaps made in the wire clear and with supporting any feint attacks made by infantry during gaps in the bombardment running up to a larger offensive.

Llewelyn Wyn Griffith, a captain with the 15th Royal Welch Fusiliers and later a novelist, recalled in his war memoir:

“At night a trench mortar officer set his guns in a derelict trench about twenty yards behind the line and carried up his ammunition, heavy globes of iron with a little cylindrical projection like a broken handle. In the morning I moved the men from the bays between the trench mortars and their target, to lighten the risk of loss from retaliatory fire.”

Sometimes the width of No Man’s Land required saps to be cut extending out from the frontline so the mortar rounds didn’t fall short. The 50 lb lollipop-like projectile had a maximum effective range of 570 yards (depending on the size of cordite charge used), and could create a crater 5 feet deep and 14 feet wide. The ideal mortar position was a 6 feet wide by 9 feet deep sandbagged pit with the weapon’s heavy wooden bed at the bottom and room for the crew to load the mortar.

Mesopatamia 1
A British 2in Mortar position in Mesopotamia, note the ignitor’s breech is open (Imperial War Museum)

Crews could manage to fire approximately once every two minutes. Much slower than the lighter 3 inch Stokes Mortar and but faster than the heavy 9.45 inch Heavy Mortar. The mortar comprising of just its tube, bed, stand and ignition system weighed 320 lbs (145kg), not including the accompanying tools and the Temple silencer system which could be fitted (which weighed 47 lbs or 21 kg alone).

Typically manned by a 5 man mortar crew comprising of an NCO, gunners, and ammunition bearers. To operate the 2 inch mortar a cordite charge was first placed down the tube, the projectile’s shaft was then inserted on top of the charge, the projectile’s fuse was set and checked and a new blank cartridge chambered in the ignition system. The crew then got clear of the weapon and pulled the lanyard to fire the mortar. To reload the crew ran a clearing stick down the tube and then repeated the loading process.

DSC_0790.jpg
The business end, a view down the length of the mortar (Matthew Moss)

Interestingly, the 2 inch Medium Mortar, like the larger 9.45 inch Heavy Mortar used a cut-down rifle, which screwed into the ‘breech’ end of the mortar tube. This particular mortar has an 1894-dated cut down Lee-Enfield MkI as its ignition system, the cutdown rifle has a wooden insert in its magazine well but it still has its rear volley sight attached. This reusable system replaced the T-tube Friction ignitor, which was in high demand by Britain’s bigger guns. The Lee-Enfield-based system enabled the cordite propellant charges to be ignited by a blank .303 round instead. The rifle’s trigger was pulled with a lanyard from nearby cover. These cutdown ignitor rifle are sometimes confused for Obrez-style Lee-Enfields.

DSC_0787
A close up look at the 2in Mortar’s SMLE ignitor (Matthew Moss)

The weight of the cordite charge used dictated the range while a variety of different fuses were used with the projectiles, these screwed into the nose of the bomb. The sphere was about 9.3 inches in diameter with a 2 inch thread for the fuse at its head and a cup for the 22 inch long, 2 inch thick solid cast iron stick or spigot at its base. The sphere was filled with high explosive (Amatol or Ammonal). The high explosive bombs were painted white with a green or pink stripe around their middle.

They were often deployed in batteries of four with three Royal Field Artillery medium mortar batteries attached to each division. The mortars were predominantly tasked with cutting enemy barbed wire and destroying enemy trenches and forward positions, such a machine gun nests.

Royal Army Ordnance Corps playing cards 1916
Men of the Royal Army Ordnance Corps playing cards on a dump of trench mortar ammunition during Battle of the Somme (Imperial War Museum)

Captain Griffith described a battery of 2 inch mortars opening fire on enemy lines:

“A pop, and then a black ball went soaring up, spinning round as it went through the air slowly; more pops and more queer birds against the sky. A stutter of terrific detonations seems to shake the air and the ground, sandbags and bits of timber sailed up slowly and then fell in a calm deliberate way. In the silence that followed the explosions, an angry voice called out in English across No Man’s Land, ‘YOU BLOODY WELSH MURDERERS.’

The 2 inch medium mortar entered service in spring 1915 and remained in use into 1917 with British and Empire troops. It was used on the Western Front and in Mesopotamia. Over 800 were ordered initially with 675,000 bombs, many of the mortars were made in railway and agricultural machinery workshops, allowing larger factories to focus on more complex weapons. The 2 inch mortar was superseded by the larger bore Newton 6 inch mortar later in the war. Some of the remaining 2 inch projectiles were re-purposed as makeshift anti-tank mines, buried in no man’s land in anticipation of possible German tank attacks.

If you enjoyed the video and this article please consider supporting our work here.


Specifications:

Barrel Length: 3 feet (90cm)
Overall Weight: ~340lbs (154kg)
Projectile Types: High explosive & smoke
Projectile Weight: 51lbs (23kg)
Effective Range: 100-570 yards (90-520m)


Bibliography:

Primary Sources:

Field Artillery Notes No.7, US Army War College, (1917) (source)

‘Artillery in Offensive Operations’ GHQ Artillery Notes No. 4 January/February 1917 (source)

‘History of the Ministry of Munitions’, Volume XI, Part I Trench Warfare Supplies (1922)

Up to Mametz, L.W. Griffith,  (1931)

Newsreels:

The Battle of the Somme, 1916, Imperial War Museum (source)

With the Forces in Mesopotamia, 1917, Imperial War Museum (source)

Secondary Sources:

British Artillery 1914-1919. Field Army Artillery, D. Clarke, (2004)

Tommy, R. Holmes, (2004)

The 1866 Benson-Poppenburg Breechloading Rifle

In early 1865, in the wake of the Danish-Prussian War which had shown how effective breechloaders could be, Britain’s Board of Ordnance began to explore retrofitting Britain’s muzzle-loading Pattern 1853 Enfield Rifle Muskets with a breech-loading cartridge conversion. Along with this interim solution the Ordnance Department also began the search for a breech-loading rifle designed from the ground up. Dozens of designs were examined from engineers and gunsmiths from across Britain, Europe, and the United States. One of these came from Johann von der Poppenburg, a Birmingham based engineer. Poppenburg’s rifle was tested along with 24 others during the initial phase of testing. The Ordnance Department’s Breech-Loading Rifle Committee were largely unimpressed by the rifles submitted and selected only four to progress, Poppenburg’s design was not included.

dsc_0483
An 1866 Benson-Poppenburg Rifle (Matthew Moss)

Poppenburg patented his first breech-loading design in February 1865 (#421) with an American patent following in October (US #50,670). It was this system which was first submitted to the trials, the rifle while described as ‘Poppenburg’s principle’ was made or at least submitted by Messers. Benson and Co., also of Birmingham.
Poppenburg’s patent describes a system that could be loaded with either loose powder and a projectile – with a percussion cap igniting the charge held in the ‘charge-chamber’ or with a paper cartridge which was detonated when pierced by a needle. It was the latter, more modern, option which was chosen for submission to the British trials.
The submitted design used a needle fire action, which ignited a paper cartridge by piercing through the paper and powder to ignite a copper cap in the base of the projectile. Poppenburg patented this cartridge design was on 3rd April 1865 (#932), it lapsed three years later and became void in April 1868. The action was hinged to the right, with a hollow breech chamber swinging out to allow a cartridge to be loaded into it. The estimated unit cost to produce these rifles, for quantities over 5,000 rifles, was £3 each. The needle fire action and hinged breech proved “too complicated, and liable to accident for a military arm” according to the Trials report.

poppenburgj-18650214-421-copy
Poppenburg’s needle fire breechloader, patented February 1865 (courtesy of Research Press)

Interestingly, in October 1866 Poppenburg also patented specific system for a breech-loading conversion (#2580). The system used a vertically hinged breech block which locked using a rack and pinion system attached to a lever. It does not appear that this system was tested by the Trials Committee. This action may have been developed following the failure of his more complex action and the adoption of the brass-cased .577 round. This patent lapsed and became void in October 1869.

The October 1866 patent (#2580), appears to be the last patented solely under Poppenburg’s name. Subsequently patents were granted jointly between Poppenburg and John Solomons Benson. This may have been due to the cost of applying for and maintaining patents, which in the 1860s could cost over £45 for three years of protection. Today that’s the equivalent of over £5,000 or nearly $7,000. Both Benson and Poppenburg were based in Birmingham, Britain’s leading centre for small arms manufacturing at the time. In a patent notice, dated 22nd December 1866 (#3382), Benson is listed as a merchant while Poppenburg is described as a mechanical engineer. It may be that Benson provided the financial backing for Poppenburg’s breech-loading system, this was an arrangement that was common at the time.

dsc_0483111
left and right profiles of the 1866 Benson-Poppenburg (Matthew Moss)

In 1866-7 Benson and Poppenburg submitted a number of rifles for testing in the Prize Competition launched by the War Office to find a new breech-loading rifle. The system submitted was radically different to Poppenburg’s earlier needle fire designs which used hinged breeches. The patent for the new system was granted jointly to Benson and Poppenburg on the 22nd December 1866 (#3382).

poppenburgj-18661222-3382-copy
Poppenburg’s third breechloading design, co-patented with John Benson, patented December 1866 (courtesy of Research Press)

Benson and Poppenburg’s new rifle had a breech which opened horizontally with a ‘tubular breech-block’ which slid to the rear when a hinged lever was liftedand pulled backwards. To open the breech the rifleman first depressed a small catch on the left side of the breech cover, once depressed the breech block could then be pulled back by the hinged lever. This movement also actuated the rifle’s T-shaped semi-circle extractor allowing the rifleman to remove the spent case. A new cartridge could then be loaded and the breech closed and the striker was then pushed forward with the thumb to cock the weapon. Once the hinged lever was pushed forwards again the breech block moved forward, closing the action, and locked with a pair of lugs cut into the receiver (described as the ‘breech-shoe’ in the patent) and at the rear by the catch.

dsc_0505
The Benson-Poppenburg with its breech open, note the t-shaped semi-circular extractor (Matthew Moss)

Depressing the breech release button with the striker cocked will de-cock the action and in theory allow a round to be carried in the chamber. The example pictured in the accompanying photographs may be a slightly more refined version of the rifle submitted as it differs from another rifle, said to be a trials gun, which more closely resembles the December 1866 patent.

 

The rifle with its breech closed (left) and open (right)

At least four rifles (with some differences in design between them) were provided for testing, the War Office’s April 1868 Report on Breech Loading Arms found that three of the rifles submitted were shorter than the required length while a fourth was too long – with the maximum overall length allowed being 51 inches. Examples of both full-length rifles, with 32 inch barrels, and carbine models with 23½ inch barrels exist (both of these lengths are significantly shorter than the Snider-Enfield’s barrel length). The trials rifles appear to have been sighted out to 1,100 yards and were chambered in a .577 calibre cartridge (probably the Boxer cartridge selected officially in 1866). At least two probable trials example were also chambered in a .450 cartridge. From a survey of the remaining examples it seems that the serial numbers for the rifles range up to at least 239.

dsc_0489
Right side view of the rifle with its breech open, note the breech lever projecting upwards and the breech block and cocking assembly back (Matthew Moss)

The Benson-Poppenburg was unsuccessful during the trials, being rejected from both the Prize Competition and the Breech Action Selection Trials. With the Committee’s report stating that despite the rifles having “several good ideas embodied in their breech action”, they “appear to have been hastily manufactured and the inventions are as yet in an incomplete state”. The specific reasons given for this were that the rifles were of unsatisfactory overall lengths. It seems they were submitted in a rush, in an ‘incomplete state’, with the report also noting that the extractors on two of the rifles submitted destroyed cartridges during extraction, probably ripping the base from the case.
The Committee’s report explained that its rejection from the separate Breech Action Selection Trials was due to issues: “if dirt or sand enters the shoe of this rifle it causes misfires, and even prevents the bolt from entering the aperture in the block.” They also noted that “The working of the breech mechanism is slow.”

The British Army’s extensive trials eventually resulted in the selection of Jacob Snider’s system, adopted in April 1866 to convert existing Pattern 1853s and the selection of Friedrich von Martini’s action and Alexander Henry’s barrel, which when combined as the Martini-Henry was formerly adopted in March 1871.

Addendum:

Mathieu Willemsen, curator of the Netherlands’ Military Museum, was kind enough to share some information about The Dutch Army’s trials with the Poppenburg in 1868. The Dutch trialled a version similar to that tested by the British but chambered in 11x42mmR. The rifle’s action has a more angled external appearance than the example we have examined but works along the same principle.

51036513_2209579459265619_6545350938708672512_n (1)
1868 Benson-Poppenburg trials rifle, note the longer breech lever (courtesy of Mathieu Willemsen)

The rifle was found to be rapid firing but suffered from some issues with fouling and failed a pressure test. Later testing with a smaller calibre round was also carried out but the rifle was not adopted. We hope to have a chance in the future to examine a Dutch trials rifle for comparison.

If you enjoyed the video and this article please consider supporting our work here.


Bibliography:

Treatise on the British Military Martini, I. Skennerton, (1995)

Reports of a Special Committee on Breech-Loading Rifles (1869)

Abridgements of the Specifications Relating to Fire-Arms and Other Weapons, Ammunition, and Accoutrements, Commissioners of Patents, (1870)

‘Poppenburg’s Projectiles’, Newton’s London Journal of Arts and Sciences, (January, 1866)

Experiment and Trial, M. Willemsen (2012)

Various British Patents:

‘Breech Actions, Hinged-Chamber’, J. von der Poppenburg, UK Patent #421, 14th Feb. 1865

‘Projectiles and cartridges for central-fire breech-loading fire-arms and ordnance’, J. von der Poppenburg, UK Patent #932, 3rd Apr. 1865

‘Breech Actions, Hinged Breech-Block’, J. von der Poppenburg , UK Patent #2580, 6th Oct. 1866

‘Breech Actions, Sliding Breech-Block’, J.S. Benson & J. von der Poppenburg, UK Patnet #3382, 22nd Dec. 1866

‘Breech Actions, Hinged Breech-Block’, J.S. Benson & J. von der Poppenburg, UK Patent #1950, 15th June, 1868

‘Improvement in breech-loading fire-arms’, J. von der Poppenburg, US Patent #50670, 24th Oct. 1865, (source)


Special thanks to our friends at the Hayes Collection for letting us examine their rifle, and thanks to David Minshall over at the Research Press for his help researching Poppenburg’s numerous patents.

Heckler & Koch G41

In 1981, Heckler & Koch introduced what would be their last infantry rifle that used their tried and tested roller-delayed blowback action, the HK G41. In October 1980, following NATO’s smalls arms and ammunition testing during the late 1970s, a meeting of NATO Armament Directors, agreed to standardise to the 5.56x45mm round favoured by the United States since the mid-1960s. Standardisation Agreement (STANAG) 4172 saw NATO standardise on the Belgian/FN SS109 ball round. At the same time Draft STANAG 4179 proposed adopting US 30-round M16 magazines as the standard 5.56 magazine pattern, while this proposal wasn’t ratified the M16’s magazine became the de facto standard.

At this time Heckler & Koch were engaged in a major engineering project to develop the G11 caseless ammunition-firing individual weapon. Their main offering for the 5.56x45mm rifle market at the time was the HK33, a rechambered version of the 7.62x51mm G3 developed by Tilo Moller, which was introduced in 1965. The HK33, however, used a proprietary HK magazine and was not compatible with the M16’s magazines. In 1977, as the NATO trials began and it became clear that 5.56x45mm would be adopted, HK began to develop what would become the G41. In 1979 with initial development completed HK submitted 18 G41s for testing with the West German Army. It wasn’t until 1981 that HK introduced the G41 onto the market.

dsc_0862a
Left and right profile views of the G41 (Matthew Moss)

While continuing to use the same roller delayed blowback operating system as the G3, HK33 and MP5, the G41 embodied a number of improvements. While still using a stamped metal receiver it utilised 1mm thick high tensile steel rather than the 1.2mm thick steel used by the HK33. This helped to lighten the receiver. The new rifle also used a lighter bolt assembly, paired with a new recoil spring which comprised of five wound strands around a central coil, rather than a single coil, which had a longer stroke. This acted to lower the felt recoil. The G41, however, had a higher rate of fire at around 850 rounds per minute compared to the 750 rounds per minute of the HK33.  Some of the G41’s bolt geometries were reworked and a new extractor was added.

The G41’s lower receiver was redesigned to allow the rifle to feed from STANAG magazines rather than HK’s earlier proprietary magazines. The cocking lever and forward assist were taken from the HK21A1 (XM262) general purpose machine gun, developed for the US SAW trials.

dsc_0041a
HK G41 (top) and HK33 (bottom) field stripped (Matthew Moss)

It also had a new more triangular polymer foregrip and added a plastic dust cover to the ejection port, a NATO pattern optics mount (meeting STANAG 2324) replaced HK’s claw-mount system, and a spring-loaded folding carrying handle near the centre of balance was added. Importantly it also added a last round hold open device and a bolt release catch, on the left side of the lower receiver.

The usual thumb serrations on the side of the bolt, for pushing the bolt home, were replaced by a prominent forward assist, similar to that found on the M16A1 and other HK weapons such as the HK21 light machine gun and the PSG-1 sniper rifle. HK sales literature described it as a ‘low noise’ forward assist and the manual describes the “quiet cocking of the weapon” – essentially riding the cocking handle back into battery and then pushing the forward assist to lock the action, the system is not as ‘low noise’ as advertised.

dsc_0876
Right side of the G41, note the addition of a forward assist and dust cover (Matthew Moss)

Another important feature of the rifle was the inclusion of a three-round burst setting alongside semi and fully automatic. The G41 could mount a standard G3 bayonet, fit an M16 bipod and had a flash hider designed to enable it to fire NATO standard rifle grenades. The 40mm HK79 under barrel grenade launcher could also be mounted to all variants of the G41, simply swapping it out for the polymer forend. HK referred to this set up as the G41-TGS or ‘Tactical Group Support system’.

9
Spread from a 1985 HK promotional product brochure showing the various G41 configurations (HK)

The G41 came in a number of variants with designations A1 to A3. The base rifle had a fixed buttstock and its rifling had 1 turn in 7 inches with a right-hand twist, in a 18.9 inch barrel. The A1 had a 1 in 12” twist barrel and fixed buttstock. The A2 had a collapsing, single position stock and 1 in 7” inch twist rifling, while the A3 had 1 in 12” inch twist rifling. The 1 in 7” rifling was optimised for the new SS109, while the 1 in 12” optimised for the US M193 round. There was also a shortened G41K model which had a collapsing stock and a 15 inch barrel available with both rifling types.

One of the main issues with the G41 was its weight. Despite efforts to lighten the sheet metal receiver, it weighed more than its predecessor the HK33. According to measurement data compiled by researcher Nathaniel F, unloaded the G41 weighs in at 4.31kgs or 9.5 lbs, this is a full pound heavier than the HK33. A contemporary M16A2 weighed 3.39kg or 7.5 lbs while the Spanish CETME L, a similar stamped receiver rifle chambered in 5.56×45, weighed 3.72kg or 8.2 lbs. The rifle eventually adopted by the Bundeswehr, the HK G36, weighed 3.13kg or 7.3 lbs.  The G41K with its collapsing steel stock wasn’t much lighter, weighing 4.3kg or 9.5 lbs, according to HK sales literature. Another potential issue may have been reliability with the move to the STANAG magazine rather than the optimised proprietary HK magazines may have introduced some issues.

dsc_0040
The bolts of the HK G41 (top) and HK33 (bottom), note the redesigned extractor, forward assist serrations on the bolt carrier and the G41’s thicker but shorter recoil spring (Matthew Moss)

Following NATO’s decision the early 1980s saw a large number of countries looking to replace their ageing 7.62x51mm battle rifles. Sweden began to look for a 5.56x45mm rifle to replace its licensed version of the G3, the Ak4, in the late 1970s. HK could initially only offer the HK33 but the G41, tested later, was also rejected by the Swedes in favour of FN’s FNC. Italy sought to replace the BM59 with a more modern rifle and HK entered into an agreement with Luigi Franchi which saw them offer both the original HK configuration and the develop their own, slightly modified version, the Franchi mod. 641, but the Beretta AR70/90 was selected. Similarly, in 1984 Spain decided to adopt the indigenously developed CETME L. In 1986 the HK G41 was also submitted to the Irish Army’s trials to replace the FN FAL, it was beaten by the Steyr AUG. Initially West Germany had hoped to procure up to 20,000 HK G11 rifles per year, with a total of 224,000 in service by 2003.

dsc_0921
HK’s G11 and G41 (Matthew Moss)

The collapse of the Soviet Union and the subsequent reunification of Germany saw Federal budgets stretched and the G11 programme was subsequently abandoned entirely. The Bundeswehr still needed a suitable rifle to replace the G3 and in the 1990s sought a lighter weight rifle. HK felt their HK50 project, in development since the mid-1970s was a better bet than the heavier G41, and following Bundeswehr trials the G36 was subsequently adopted in 1997. Sadly, I have not been able to get a hold of any of the trials reports from the nations that tested the G41, so can not say with certainty why the countries mentioned above rejected HK’s rifle.

7
Graphic from HK’s manual for the G41 (HK)

From photographs of members of the Turkish Gendarmerie special operations group training at the Foça Commando School, dating from the early 2010s, it appears that Turkey either purchased a number of G41s or Turkey’s state-owned defence manufacturer, MKEK, produced an unknown number under license.At some point in the 1980s the British Army also tested a small number G41s with serial numbers #11131, #11832 and #11833 remaining in UK collections.

Denmark’s elite Jaegerkorpset and Froemandskorpset used the G41 for a time and Argentina’s special forces, including the Grupo de Operaciones Especiales, have also been photographed with both HK G41s and G41A2(collapsing stock) fitted with the TGS package comprising of the HK79 under barrel grenade launcher.

Argentine commandos with HK G41
Argentina’s Grupo de Operaciones Especiales on parade with G41s and the G41-TGS, grenade launcher package (source)

The G41 represents the last evolution of HK’s infantry rifles using the roller delayed blowback action. It comes from a period when HK were developing what they hoped would be the next generation of small arms technology and with the collapse of the G11 programme and the lack of interest in the G41 the company faced financial uncertainty throughout the early 1990s. HK’s move away from the roller delayed blowback action to a more conventional gas operated rotating bolt system, combined with lightweight polymers, in the G36 proved to be more successful than the ill-fated G41.

If you enjoyed the video and this article please consider supporting our work here.


Specifications (standard G41 rifle model):

Length: 39in (99cm)
Weight (unloaded): 4.31kgs or 9.5 lbs
Barrel Length (not including flash hider): 17.7in (45cm)
Action: Roller-delayed blowback
Calibre: 5.56x45mm
Feed: 30 round STANAG magazines
Cyclic Rate: ~850rpm


Bibliography:

The World’s Assault Rifles, G.P. Johnston & T.B. Nelson, (2016)

Die G11 Story, W. Story, (1993)

Full Circle: A Treatise on Roller Locking, R. Blake Stevens (2006)

The 5.56 Timeline, D. Watters, (source)

1985 HK Brochure on the G41 Series (via SAR Archive)

HK G41 Owner’s Manual (via SAR Archive)


Our thanks to the collection that holds this rifle for their kind permission to examine and film it. Please do not reproduce photographs taken by Matthew Moss without permission or credit. ©The Armourer’s Bench, 2019.

Prince’s Breechloading Rifle

In February 1855, London gunmaker Frederic Prince patented an intriguing breech-loading system. Prince offered his rifle to the Board of Ordnance for testing where it outshot the then-standard Enfield 1853 Pattern rifle musket during trials at the School of Musketry at Hythe in 1855. However, the Ordnance Department refused to consider adopting the new breechloading system believing it to be too complex and expensive to manufacture. It would be another nine years before the British Army took breechloading seriously.

dsc_0354side by sidew
Prince’s Breechloader, with its action closed and open (Matthew Moss)

Prince’s system used a sliding barrel to open up the breech to allow the loading of a paper cartridge, once the breech was closed the percussion lock was then capped. Once the hammer was brought back to full cock the rifle was ready to fire. In order to load the rifle the weapon was placed on half cock, the ‘bolt handle’ was then unlocked by pulling back the locking piece which protruded from the base of the trigger guard. The bolt handle was then turned slightly to the right disengaging the two lugs, which locked the breech, and then the bolt could be pushed down a short ‘L’ shaped channel. This pushed the barrel assembly forward several inches, opening the breech and allowing the rifleman to load a paper cartridge.

dsc_0362
Right side view of Prince’s action (Matthew Moss)

Once loaded the bolt handle was pulled rearward again, and turned to the left again to re-engage the locking lugs. The bolt locking piece was then pushed back into a recess in the bolt handle to secure it. The bolt handle, along with the lugs inside the receiver, act to keep the breech block locked during firing.

dsc_0368
View of the inside of the action with the barrel pushed forward, note the pair of locking lugs and the rubber seal on the breech plug (Matthew Moss)

Hans Busk, the prominent Victorian rifle proponent, described Prince’s action, in his book The Rifle, and How to Use It (1861) as: “one of the simplest and handiest breechloading rifles that has yet been tried.”

Prince’s patent for the system was granted on the 21st February 1855. Between 1854 and 1859, Frederic Prince patented no less than eight improvements to small arms ranging from actions to manufacturing processes and even improvements in metallurgy. Prince was a gunmaker working in London, for a time in partnership with William Green with premises on New Bond Street.

princef-18550221-386-b
Prince’s 1855 patent for his action (courtesy of Research Press)

One of the key elements to Prince’s system was his self-contained paper cartridge, protected by two patents dating from January 1855 (#33 and #173). Prince’s cartridges were relatively conventional in their design; made up of a paper tube with a wad between the bullet and powder. The paper was treated with a mixture of sulphuric and nitric acids, the process toughened the paper but also, according to Busk, caused it “to be entirely consumed in the barrel”, in theory leaving no debris behind, once it was ignited by the flash of the percussion cap. This had the result of greatly speeding up loading, in a similar way to the cartridges used by the continental needle guns but not going so far as to include the primer inside the cartridge.

dsc_0375
Prince’s breechloaders used a back action, here we can see Robert S. Garden’s makers mark on the lock plate (Matthew Moss)

Prince’s February 1855 patent describes a conical plug at the rear of the breech with a pair of locking lugs which locked into the walls of the breech. The patent also mentions the possible use of rubber gas seals to obturate during firing and prevent gases being vented. A subsequent patent (#3036 22nd December 1856) protected the concept of having a hollow, concave plug which could either be circular or hexagonal.
In his book Busk recounts that during the trials at the British Army’s School of Musketry at Hythe, Prince’s rifle was reportedly able to fire six rounds in just 46 seconds and a total of 120 rounds were fired in just 18 minutes by Prince himself. On another occasion, using a small bore version of the rifle, Prince was able to demonstrate how accurate his rifle was putting 16 rounds onto a small piece of notepaper, a grouping of 1 ¾ inches, at a range of 100 yards during a demonstration at the Victoria Regimental Practice Ground. The trials at Hythe saw it fired against the Enfield Pattern 1853 rifle musket where it put 48 out of 50 rounds on target at 300 yards compared to the Enfield’s 47.

dsc_0398
view of the Prince’s breech, note the two cuts either side of the chamber to lock into the locking lugs once the breech is closed (Matthew Moss)

The School of Musketry’s annual report for 1855-56 makes some interesting observations. The report states that testing took place in July 1855 and noted that the Pattern 1853 performed better at 600 yards, no doubt because of its longer barrel, while the Prince “had a slight advantage at the shorter distance”. The report also notes that Prince’s rifle required “a greater angle of elevation than the rifle musket 1853, especially at the longer distances, which proves that the latter has a more horizontal trajectory”, the more parabolic trajectory of the Prince rifle is likely due to its shorter barrel and possibly its five groove rifling. In terms of rate of fire the official report the Pattern 1853 was said to be capable of 35 rounds in fifteen minutes whereas Prince’s rifle managed 72. This casts some doubt on Busk’s account of the trials, which of course may refer to a different test, but it does support the rapidity of Prince’s action. One final interesting observation from the report is that in operating his rifle Prince administered “copious lubrications of saliva” to the action which as a result “worked easily throughout”. The report concludes with the suggestion that Prince’s rifle should “be subjected to a prolonged trial before an opinion can be expressed as to its efficiency for infantry.”

dsc_0379
Left side view of Prince’s rifle (Matthew Moss)

While Prince’s rifle performed admirably the War Office refused to order a batch for further testing, perhaps feeling his system was too complex or too expensive to manufacture, or perhaps not robust enough for military service. Another important factor to consider is that in 1855, the British Army had just two years earlier formally adopted the muzzle-loading 1853 Pattern rifle musket and was of course engaged in the Crimean War.

dsc_0394
Underside view of the Prince’s bolt and bolt travel track, note the reinforcement around the track to strengthen the stock (Matthew Moss)

Prince’s patent is undeniably an ingenious breech-loading system. It is a testament to the belief in the design that in 1858, three years after it had first been rejected, a group of prominent London gunmakers including Manton, Wilkinson, Samuel Nock, Parker Field, and Tatham petitioned the Department of the Master-General of the Ordnance to reconsider their decision in a testimonial, published in The Field magazine in April 1858, arguing that they wished “to see the most effective weapon in the hands of our soldiers” and describing Prince’s rifle as “the best we have seen”. Several of these gunmakers, including Manton and Wilkinson, went on to produce rifles based on Prince’s system.

A survey of surviving examples of rifles using Prince’s action shows that a large number of contemporary gunmakers made rifles based on his system, although how many exactly were made remained unknown.

The rifles were produced by gunmakers including Prince’s own company – Prince & Green, as well as Wilkinson’s, E.M. Reilly, Robert S. Garden, Manton & Sons and Hollis & Sheath (later Hollis & Sons). According to De Wit Bailey the London Armoury Company also took on a manufacturing license for Prince’s action in 1861, it is unknown if any were ever produced before the company collapsed in 1866.

dsc_0358
Right side of the Prince (Matthew Moss)

Many Prince’s patent rifles were made for civilian sporting and target use. The surviving rifles tend to have barrel lengths of between 25 and 31 inches and most have either three or five groove rifling. The rifles were made in various calibres from the British army’s preferred .577 to much smaller rook and rabbit hunting guns in .24 and .37 calibre. Other larger calibres include .500 and .90 inch bores. With the variety of makers the sights, stocks and fittings found on the rifles vary greatly from simple dovetailed leaf rear sights to more complex ladder sights. There is some variation in the shape and orientation of the locking lugs on the breech plug, this may indicate some experimentation by gunmakers to find the most efficient shape and angle. There is also variation in the style of the barrel bands which held the barrel to the stock, most have a single barrel band that loops over the barrel but there are several examples which have bands split in half and do not surround the barrel, one example has a set of two of these. All of the guns, however, have back action locks – in order to leave more room around the breech to ensure strength.

princef-18590128-259
Prince’s later 1859 patent for his under-hammer siding breech action (courtesy of Research Press)

Following the rejection of his system by the military Prince developed another breechloading action, patenting it in January 1859 (#259). The new rifle combined a percussion cap magazine inside the rifle’s stock, below the barrel, with an under-hammer lock which was connected to a vertically sliding breech block. It is unclear if this design was ever produced, tested or sold commercially.
In refusing to adopt Prince’s breech-loading system it can be argued that the British Army passed on an opportunity to leap ahead of its rivals. The system was undoubtedly fast and accurate in action, it is possible the Ordnance Department felt the system was too complex and its sliding barrel was not robust enough for service conditions. However, the Prince rifle was a single victim of a wider trend between 1842 and 1865, the Board of Ordnance and later Ordnance Department examined and trialled dozens of breech-loading rifles during the period but did not feel it necessary to adopt one until they had been overwhelmingly proven in the field.

In early 1864, the armies of Europe were shocked by the decisive victory the Prussian Dreyse needle guns helped to bring about during the Danish-Prussian War. In 1865 the British began to seriously look for a breech-loading replacement of their Enfield 1853 Pattern rifle muskets. Following trials of various submitted designs Jacob Snider’s cartridge conversion was selected and in September 1866, the Snider rifle was introduced becoming Britain’s first breech-loading service rifle.

If you enjoyed the video and this article please consider supporting our work here.


Bibliography

Patents:

22nd December 1854, Patent #2705
3rd January 1855, Patent #33
23rd January 1855, Patent #173
21st February 1855, Patent #386
16th November 1855, Patent #2590
22nd December 1856, Patent #3036
28th January 1859, Patent #259
15th July 1859, Patent #1679

Sources:

The Rifle and How to Use It, H. Busk, (source)

Pattern 1853 Enfield Rifle, P. Smithurst (2011),

English Gunmakers: The Birmingham and Provincial Gun Trade in the 18th and 19th Century, D.W. Bailey & D.A Nie (1978)

Guns Dictionary: Guns & Gunmakers, J. Walter (source)

Annual Report on the Instruction & Experiments of the School of Musketry, 1856


Special thanks to our friends at the Hayes Collection for letting us examine their rifle, and thanks to David Minshall over at the Research Press for his help researching the Prince’s various patents.

Live Fire: Browning M1919A4

Today’s episode is the last video of 2018, so we thought we’d end the year with a bang, literally. Earlier this year Matt had the chance to get behind an original Browning M1919A4 so we’ve put together a video showing the classic belt-fed machine gun in action with some slow motion footage thrown in!

DSC_0450
Browning M1919A4 (Matthew Moss)

This M1919A4 was built in 1944 at GM’s Saginaw Steering Division plant, in Saginaw Michigan. It was one of nearly half a million M1919A4s built during World War Two. In the video Matt explains a little of the gun’s history and how it worked.

S1270015_Moment
M1919A4 with its feed cover open (Matthew Moss)

This M1919 has been rechambered from the original .30-06 to 7.62x51mm NATO and uses M13 disintegrating links rather than a cloth belt or M1 disintegrating links. My thanks to Chuck and his buddy over at GunLab for letting me put several belts through his gun, it was a lot of fun.

We’ll have a full, in-depth, episode on the Browning M1919 in the future.


Thanks to everyone for watching, liking, subscribing and commenting on our videos this year, we can’t tell you how much we appreciate all the support we have received. I’m very pleased to say we reached 3,000 subscribers before the end of the year, very pleased that our community is growing! We have much more to come in 2019, and we’ll be back with regular videos in January.

Call of Duty: WWII’s Sterling SMG

During a recent discussion over on the HF Twitter page, I was informed to my surprise that the Sterling submachine gun had been added as a DLC weapon to Call of Duty WW2. I thought it would be interesting to take a look at the model used in the game and see how historically accurate it is. I recently finished writing a book about the Sterling and have done some research into the theories of the Patchett prototypes seeing some action during the war.

The model that Sledgehammer Games, the developer, have used appears to be a mix of the early prototypes and the later production Sterlings. In terms of historical accuracy the gun should be correctly referred to as the Patchett Machine Carbine – after its designer George Patchett. It only began to be called the Sterling, after the company that manufactured it in the 1955.

COD Sterling
Developer’s model of the COD: WW2 Sterling SMG (courtesy of Activision/Sledgehammer Games)

The model appears to share some similarities with the original Patchett prototype, including the step in the welded together receiver – the result of using left over Lanchester machine carbine receiver tubes, which was also built by Sterling. The position of the stock hinge point also appears to be in the correct place (it was later moved forward when the stock was modified). However, it appears to be feeding from a much later curved commercial pattern Sterling magazine (you can tell by the zigzag outline on the rear of the magazine and of course the curve – although seemingly not quite as curved as the real thing.) In reality the Patchett prototypes fed from Sten magazines, it wasn’t until after the war that Patchett designed his excellent 34-round magazine.

Here’s a photo of the Patchett’s original tool room prototype that I took last year while researching:

Patchett prototype
Patchett’s Original Toolroom prototype (Matthew Moss)

Note how they even replicated the slanted brazed on rear sight that was added after the first trials. The game developers, however, added a metal guard tab just in front of the ejection port – something that wasn’t added until later and they also gave the gun markings on the magazine housing that mimic the later commercial Sterling markings.

The game model also has the Sterling’s helical grooves on its breech block, something the early prototypes did not have. It seems the developers mashed together the Patchett prototype with later production Sterling L2A3/Mk4s.

Did the Patchett See Action During WWII?

image1
A grainy photograph, sadly lacking provenance, that appears to show members of the Free French SAS with two Patchett prototypes during Operation Amherst, April 1945 (source)

While the early Patchett prototypes may have seen action in late 1944 – 1945 with one or two prototypes possibly making it into the hands of specialist troops there is no solid evidence to support this. There is a tantalising grainy photograph of what is believed to be members of the Free French SAS on operations in the Netherlands in April 1945 (during Operation Amherst). The photo above shows what appear to be two Patchetts during a meeting with local resistance members. There is also an uncorroborated story that one prototype was carried by Lt.Col. Robert Dawson, officer commanding No.4 Commando, during Operation Infatuate but there is no documentary evidence to support this. I discuss these and several other pieces of evidence that support the idea that the Patchett/Sterling saw action in my new book on the Sterling.

I have written a book for Osprey’s Weapon series looking at the development, use and significance of the Sterling, it’s available now, you can find out more about it here.

Shotgun sight Sterling SMG Prototype

Before its adoption by the British Army in 1954 the Patchett Machine Carbine, later better known as the Sterling submachine gun, was extensively tested all over the world. The Patchett went through nearly a decade of testing, evaluation and refinement. It was tested by British troops around the world, from West Germany to Africa, from the middle east to Malaya.

Today, we’re going to examine a unique Patchett/Sterling prototype assembled in Malaya. The gun we’re examining is officially a MkII Patchett Machine Carbine, but as the guns are better known as the Sterling we will refer to it as such from here on out. This prototype has been specially adapted with a shotgun style rib sight to help aiming in jungle conditions.

DSC_0113
Right side profile of the jungle rib sight Patchett prototype (Matthew Moss)

It was in Malaya that the specially adapted but short-lived prototype improvement emerged. As early as December 1952, British troops were testing the gun during operations against communist insurgents in Malaya. The harsh jungle conditions were a challenge for any weapon but an early report testing a single prototype noted that the weapon performed well but one of the issues identified was that the rear aperture sight was found to be “smaller than was desirable” and the report suggested that the aperture be widened to 0.098 inches 2.5mm – the same as the Owen gun. The report also noted that the front sight “did not stand out well in relation to the front sight protectors”.

It seems that when a batch of 75 trials guns arrived in 1953, a number of them were specially adapted in theatre. It was hoped that the shotgun-style rib sight fitted along the length of the receiver would aid snap shooting in the jungle. It was intended to enable users to engage fleeting targets quicker and improve ‘first shot hit’ probability in thick jungle and heavy rainstorms.

96329
British troops patrolling the Malayan jungle, 1957 (National Army Museum)

During operations in Malaya and Borneo, many scouts and point men carried shotguns such as the semi-automatic Browning Auto-5. Shotguns were favoured during jungle operations because of the ease with which they could be quickly and instinctively aimed and their exceptional close-range firepower.

The modification saw the complete removal of the standard front and rear sights and the razing on of a rib sight running along the length of the top of the gun from the muzzle to the rear sight. It appears an armourer cut down and removed the front and rear sight assemblies and used them as mounting points. The first few inches of the rib are stippled to minimise glare and a brass front sight bead has been added to help sight acquisition.

DSC_0232
Left side profile of the jungle rib sight Patchett prototype (Matthew Moss)

The simpler sight rib also helped with another issue that was identified during early jungle testing, it removed the problem of the sights getting clogged with mud. It is unknown just how many were adapted but at least three are known to survive. The jungle-specific modifications were short-lived and not formerly adopted because the rib sight offered poor longer range accuracy.

Here are some more detail photographs of the rib sight prototype:

 

 

 

With the adoption of the Patchett as the L2A1, in 1954, a list of modifications based on trials recommendations was drawn up in June 1953, one of the suggestions was the enlargement of the rear sight aperture to 0.1, (2.5mm) 0.15 (3.8mm) or 0.2 inches (5mm). In August 1953, the infantry board decided that the 100 yard aperture would be 0.15 (3.8mm) in diameter while the 200 yard would be 0.1, (2.5mm). The spacing of the rear sight protectors was also subsequently widened to 0.55 inches (14mm). With these changes made the Sterling saw service in the jungles of Malaya and Borneo for over a decade during the Malayan Emergency and Indonesian Confrontation.

If you enjoyed the video and this article please consider supporting our work here.


Bibliography

Primary Sources:

‘Operational Research Section, Singapore, Technical Note No.5 – Technical Notes on Initial Trials of the Patchett Carbine in Malaya’, Maj. R.St.G. Maxwell, 1th December, 1952, Royal Armouries Library

‘Minutes of a Meeting held at the war office on Friday 7th August, 1953, to decide whether the Patchett sub-machine gun be introduced into the Service as a replacement for the Sten sub-mahcine gun’, Royal Armouries Library



I have written a book for Osprey’s Weapon series looking at the development, use and significance of the Sterling, it’s available now, you can find out more about it here.