Coldwar American Prototypes

Assault Amphibian Personnel Carrier LVTPX-12

U.S.A. (1964 – 1982)
Landing Vehicle Tracked (LVT) – 15 Built

The United States Marine Corps (USMC) has, as core element of its role, the task of assaulting enemy-held coastlines. In order to fulfill this obligation, they required a form of transport that could get them from the landing ship off the coast to shore quickly and safely. It needed to be fast in the water as it was vulnerable to enemy fire with nowhere to hide and had to protect the occupants to get them onto the beach. It should also be able to deliver supporting firepower to support the Marines in their attack. All that sounds simple enough, but the combination of these demanding requirements was a complex juggling skill to balance the competing requirements.

A Modern Amphibian – 1964

The existing fleet of amphibians dated back to the Second World War and were obsolete. The need of the Marine Corps had not gone away so a new vehicle was needed. As a result the Bureau of Ships (a Department of the Navy) issued a preliminary specification on 23rd January 1964 for an Assault Amphibian Armored Personnel Carrier with a minimum requirement for a forward water speed of 8 mph (12.7 km/h) from tracks or 10 mph (16 km/h) from auxiliary power, a reverse (backing) speed in water for 3.5 mph (6.3 km/h). The vehicle was to carry enough fuel for 7 hours at 8 mph and to manage a forward speed on land of 30 mph (48.3 kmh).

The dimensions of the vehicle were not to exceed 26’ (7.9 m) long, a beam (width) of not more than 10.5’ (3.2 m), and a deck height of 8.5’ or less (2.6 m). Space was important and the inside of the vehicle had to provide a space of at least 14’ (4.3 m) long and 6’ (1.8 m) wide sufficient for a team of 25 marines with field equipment.

Weight was to be kept at a minimum but not at the expense of armor. The LVTPX-12 was to be an amphibious assault vehicle, so the gunner had to have an efficient view, and the protection had to be enough against 99% of 105 mm air burst shrapnel at 50 feet (15.2 m). Armament was to consist of either a 20 mm cannon or 7.62 mm machine gun mounted in a turret. The US Marine Corps followed this 1964 requirement with their own in March of that year for a new Landing Vehicle Tracked (LVT) to replace their LVTP-5’s.

Several ideas for construction were entertained, and the firm of Chrysler was one of them, and was awarded contract number 4777 for their work. Engineers at Chrysler looked afresh at all areas of amphibian vehicle design with work on the project divided into several phases, with Phase 1 running from June 1964 to May 1965. Three specific area required attention for the design: water speed, armor protection, and the transmission.


Various options for a layout with a single engine were considered including:

  • Single engine and transmission forward
  • Single engine and transmission aft
  • Single engine forward, transmission aft
  • Single engine aft, transmission forward
  • Double engines anywhere along the length
Ramp forward concept for the LVTPX-12, which was quickly abandoned. Source: Chrysler Corporation

Although, in theory, the troops could be unloaded through a front opening door, this exposed them to enemy fire, and consequently the choice, following advice from the US Marine Corps, was an engine and transmission forward design allowing the crew to disembark via a power ramp at the rear covered by fire from the vehicle.

Diagrammatic of the twin-engine LVTPX-12 proposal. Source: Chrysler Corporation

A double or dual engine design permitted more flexibility, being able to put the engines anywhere in the vehicle making the trim much easier to manage. The only engine narrow enough to fit was the GM6-71T but each one produced less power than a 12V71T, just 460 hp. A pair of GM6-71T’s though would produce 800 hp which was plenty, but came at a higher price as together they were heavier and were not as efficient.

The only advantage of the dual engines therefore was an ease of control and with that, the idea was dumped. When consideration was given to transmissions, either Mechanical, Hydrostatic, or Electric, only two engines were under consideration: the 12V71T or the 8V71T.

Table showing the different equipment systems that were under consideration.


More than 100 separate models of various types and modifications of the design were produced over the course of testing, including models of just the bow and stern respectively to find the optimum shape. A single box shaped model was made of a 5 road wheel layout to which various shaped box and stern sections could be attached for trials. All these various designs meant that the length and exact water-trim level varied each time. Only the width, at 10.3’ (3.14m) remained the same, although the trials were conducted with a ⅕ scale wooden model rather than a full size pickup. Some of the trials also involved a beam reduction of 20% with a much thinner vehicle to see if that would deliver the water speed improvements demanded. The narrow design caused a lot of problems, the wheels had to have very little travel, just 9” (229mm), and the whole wheel and suspension arrangement was much more complex. This also made handling harder and less efficient. There was insufficient width for two propellers, so only a single propeller could be used, which would mean a loss of ground clearance too. The marginal improvements in performance in the water were not worth the compromises required and the idea was dropped.

Side and cross sectional views of the narrow LVTPX-12 concept. 20 % narrower than the standard vehicle. Source: Chrysler Corporation
Partial model made for water testing lacking stern and bow. Source: Chrysler Corporation


Testing of the models was conducted at the Ship Hydrodynamics Laboratory at the University of Michigan, and involved not just the various LVTPX-12 configurations, but also the LVTPX-11, LVTPX-2 and LVTPX-7. The model tests showed that very small scale models gave unreliable results and no model smaller than ⅕ scale should be used for testing. They did however, confirm the final hull shape required and the most efficient method of propulsion in water.

Water Propulsion

Two options for propulsion of the LVTPX-12 were considered; propulsion in water by the tracks alone, and propulsion by other means, such as bow or stern propellers. Additional consideration was given to the use of hydrojets too.

For the tracks driving the vehicle in water, 5 different types of grousers ranging from 1.12” (28.5 mm) to 1.25” (31.8 mm) wide were tested to find the most efficient type.

The grouser tests showed that increasing the height of the grouser without increasing the width did not increase the efficiency of the vehicle in water, as it simply increased the friction instead. Wider grousers, likewise, did not improve performance either, although grouser number one was marginally better than all of the others. It was also found that, just like with the LVTP-5, the space between the hull and return portion of the track had to be as small as possible and blocked or the returning track would rob the forward track of motive power as it was effectively driving the vehicle in the opposite direction. An attempt of the LVTPX12 model to alleviate this with a completely covered side track made the problem even worse; large side plates did not help and neither did cutting holes in them. If a fender was to be used it would have to be small and only on the forward portion of the track and were essential to meet the design speeds wanted.

Precisely the opposite was true using the propellers to drive the vehicle. Even these small front fenders were harmful to water speed. Therefore, the LVTPX-12 would either be a track driven short-front fender design or a non-fendered propeller driven machine.

One peculiar finding was the use of a stern plane. Contrary to a first impression, the addition of a large flat stern plane placed flat along the back of the tracks, full width, actually improved performance in the water even though it actually interrupted the flow of water. The engineers at Chrysler did not know why this was but found that this had first been suggested in 1860 by Arthur Rigg in Great Britain to improve the efficiency of paddle wheels on ships.

Large flat sern plane. Source: Chrysler Corporation

Having looked at the stern plane, the engineers looked at bow vanes and the first attempt involved a design copying that on the LVTP-7. This plane was supposed to prevent submergence of the LVTP-7 at speeds of up to 10 mph, although even in a calm sea the real vehicle struggled to manage 7mph without serious swamping and even less in Sea-State 2 (2 foot waves). The same bow vane fitted toot the LVTPX-12 did not improve performance in the water at all and was actually a hinderance based on a loading of 59,000lbs (26.8 tonnes). The outcome was that a bow plane was pointless and the vehicle was simply better served with a boat shaped bow. Stern shape was even simpler. Because the only bow shape acceptable had to be boat shaped it extended the vehicle to the maximum 26’ (7.92 m), the stern shape could not be boat-shaped. Any angle less than 15 degrees was equally bad and the conclusion was simply to make it square to maximize space instead.

On the question of propellers, the single 27” (686 mm) Kort propeller was found to be highly satisfactory for water propulsion, but caused other problems. It had to be fitted as low as possible for maximum propulsive effort, but it reduced the ground clearance to just 16” (406mm), although a secure stowed position for it was provided. The significant advantage of a single propellor was that it could easily be used for steering the LVTPX-12, but during sharp turns increased the risk of capsize. To solve this capsizing issue, the solution was two propellers with one on each side. Although they would be vulnerable to damage as they projected outside of the width of the vehicle, they would not affect the ground clearance and even should both fail the vehicle would use its tracks to manage 6mph. Controllable steering for a twin propeller design would necessitate the use of controllable pitch propellers.

Rear view of LVTP12 designs with single 27” (686mm) Kort propeller. 26” (660mm) versions were also tested. Source: Chrysler Corporation
Double 29” (737mm) Kort propellers in the open and stowed positions seen from the rear. Source: Chrysler Corporation

A better form of steering for the LVTPX-12 was by means of electrically driven 7.5” (191 mm) diameter hydrojets fitted within the sponsons of the machine. These small hydrojets extended just 39” (991mm) along the inside of the sponson weighing just 87 lbs (39.5kg) each.

LVTPX-12 with the rear ramp and hydrojet positions at the bottom of the hull. Space for 26 men and three crew provided. Source: Chrysler Corporation

It was recommended, however, that later some kind of reactive steering should be incorporated within the track drive. For the initial recommendation from Chrysler, the choice was to use twin propellers and vary the pitch to steer the machine. This saved weight and space and allowed for 13 hours of water operation at 8mph.

Positions of one of the auxiliary drive units. Source: Chrysler Corporation

The Final Designs

After all of the development work, there were 5 main possibilities for the LVTPX12 recommended by Chrysler, all with the same basic dimensions, 26’ long by 10.5’ wide by 8.5’ high (7.9m x 3.2 m x 2.6m), and all of which met the requirements from the Navy, although the report from Chrysler was clear that only concepts 1 or 2 were ideal:

  • Concept 1: Track Propelled, 12V71T engine
  • Concept 2: Auxiliary Propelled, 12V71T engine
  • Concept 3: Auxiliary Propelled, 8V53T engine
  • Concept 4: Auxiliary Propelled, 8V71T engine
  • Concept 5: Auxiliary Propelled, twin AC-350C engine

The final designs were based upon a final vehicle weight (unladen) of 45,000 lbs (20.4 tonnes) although some of them went as high as 53,670 lbs (24.3 tonnes) and 26 feet (7.9 m) long. Bow fenders would wrap 150 degrees around the front sprockets with the ability to retract for operation on land. The design would have stern baffles with a contravane extending 4” (101.6 mm), also retractable for land use.

Side skirts were also to be added. They improved water speed and it did not matter what they were made of, just so long as they were smooth and extended to the level of the bottom of the hull. The type 3 bow shape with type 1 grouser were also to be used.

Original requirement was 8 mph (12.8 km/h), but the design was calculated to be able to achieve 10.7 mph (17.2 km/h), although testing only went as high as 9.55 mph (15.4 km/h).

Final hull profile of the LVTPX12. Source: Chrysler Corporation


The hull was to use a modern generation of high hardness steel developed by Chrysler Defense Engineering providing the ballistic protection required making the vehicle lighter than it would be with an aluminium hull, as it permitted the hull to be semi-monocoque. Consideration had been given to a dual-hardness steel hull, but although this could save 1090 lbs (494 kg) in weight, the costs involved were considered too high to be justifiable. The same went for the idea of using titanium or ceramics within the armor; the costs simply did not justify the small additional benefits.

Structural framework for the LVTPX12. Source: Chrysler Corporation

The top of the hull, including the cargo hatch, was to be made from military-grade steel (MIL-S-12560), 0.375” (9.5mm) thick with a nylon blanket backing. The sides, front and stern were made from BHN-500 steel, 0.31” (8mm) thick, and the bottom from military-grade steel (MIL-S-12560) too, 0.375” (9.5mm) thick, with structural elements made from either US Steel’s T1 high strength alloy or Cor-Ten low alloy steel as they were far more resistant to corrosion. The flooring inside the vehicle was to be aluminium paneling as were the fenders and external baffles

Protection had been required to defend against shell fragments and small arms fire only, and the use of a steel hull met these requirements. As for armament, Chrysler exceeded the requirements for either a 20mm or a machine gun by adding both. The 20mm Hispano-Suiza cannon and machine-gun were to be mounted coaxially, forwards on the hull in a small 360º rotating turret with 12º of depression. Just 325 rounds of ammunition for the cannon could be carried, but, along with 700 rounds for the 7.62mm machine-gun, provided adequate firepower for the vehicle. Spent casings from the cannon were ejected overboard via an ejection port, but spent cases from the machine-gun were simply to be collected internally. Additional firepower could be given by the mounted troops for whom small arms ports were provided.

LVTPX-12 ‘Track Propelled’ version. Source: Chrysler Corporation
LVTPX-12 ‘Screw [Propeller] Propelled’ Version. Source: Chrysler Corporation


During trials, the multiple small wheels had been shown to have far too little travel, so the vehicle to be built was to have 6 large rubber-tyred road wheels on each side instead. These were supported on torsilastic springs cantilevered out from the sides of the vehicle. Drive was delivered to the sprockets at the back, pulling the rubber padded single pin track. The front idler was compensated to account for track movement during amphibious driving and was used to tension the track too.


The transmission for the LVTPX-12 design was a complex problem. The engine would have to deliver in excess of 600h p (recommended to be 800 hp) through the tracks to power the vehicle, which would still only deliver a top speed of 16 mph (25.7 km/h) on land. Various solutions over different types of transmission were considered, all of which were going to be more suitable than using the transmission from a tank like the LVTP-5 had done.

Chrysler concluded that for the final design a new transmission should be developed by a Phase II contractor of a type recommended by Chrysler. Chrysler were prepared to develop this new transmission but not out of their funds. They were clear it would have to be a government-funded and owned project only.

Artist’s conception of the LVTPX12. Source Hunnicutt


The vehicle itself was to feature just three crew members. A driver positioned in the front left, an assistant driver in the front right and the commander acting as the gunner stood who centrally with his head in the small turret at the front. Seating at the back was provided for 25 troops on benches, although, had the LVTPX-12 been accepted for service, it is likely this area would have been adapted for a variety of other uses too from mortar carrier to recovery vehicle all on the same platform. Access for personnel was via either the large cargo hatch above the troop compartment, the large rear powered ramp, or crew hatches. Two hatches were at the front, with one each for the driver and co-driver, and there were two emergency escape hatches in the vehicle with one on each side of the hull.

LVTPX-12 model completed 1967 in an early form with boat-shaped front, which was later abandoned.


Chrysler’s model LVTPX-12 development team had been tasked with producing an amphibian vehicle capable of 8 mph and yet was calculated to be able to manage over 10 mph. In all areas, this design surpassed the LVTP-5, providing for improved armament and performance. The problems of making an amphibian APC were obvious during the development. A large amphibian was simply ill suited to both land and sea operations. Either too heavy for the sea or ill protected for the land. Big, bulky and cumbersome, these vehicles, packed with troops, were to be the amphibious assault vehicles for the US Marine Corps. In comparison with the LVTP5-A1, the LVTPX12 was considered by Chrysler to be a better design and still manageable within the budget constraints imposed by the military.

LVTPX-12 possible prototype hull No.1 during trials. The ‘guns’ appear to be dummies with the cannon below the machine gun rather than coaxial. The lines are clean and smooth but the front has lost its boat shape. Photo:

The US military continued with development of the amphibian as fighting in Vietnam had shown the extreme vulnerability of these vehicles to mines in particular and the requirements of 1964 had changed from high speed beach assault to more emphasis on land operations. The LVTPX-12 design was therefore modified to a more modest vehicle and subsequently the LVTPX-12 was manufactured as a prototype first in September 1967 and later as a batch of 14 more prototypes by Food Machinery Corporation (FMC), which were finished by 1969. It was still the ‘Landing Vehicle Tracked LVTPX-12’, but not for long.

The engine and transmission had been relocated and it was only an LVTPX-12 in name. At least three prototypes and possibly as many as 10 were manufactured under the LVTPX-12 name, but are easily mistaken for the LVTP-7, as the design was modified in favor of a smaller size and weight and refined further. The finished vehicles were smaller, but the work on the original LVTPX-12 was not wasted. It produced valuable experience and lessons for the military.

One prototype vehicle which underwent tests around June 1969 was reported as being made from aluminium instead of steel as originally planned. This switch to aluminium is also confirmed by Hunnicutt, who states that the LVTPX-12s made for trials were made from 5083 aluminium, just like the M113 APC. The Chrysler prefered engine was changed for the smaller 8V53T diesel producing just 400 hp and the torsilastic suspension was changed to torsion bars.

The LVTPX-12 program might not have been successful in itself, but it continued into what was to become the LVTP-7 program instead. By this time, very little of the original form of the LVTPX-12 remained. Gone were the boat shaped front, the central front turret, and side egress doors, and the visual similarities with the LVTP-7 make identification and tracking of the vehicle from this point almost impossible.

Following successful trials of the LVTPX-12 at Aberdeen (Maryland), Yuma Proving Grounds (Arizona), Fort Greely (Alaska), and in Panama it was accepted for service, and the LVTPX-12 name was almost completely dead by 1969, when it was officially redesignated LVTP-7 in this new form and it entered service with the US Marine Corps in 1972, replacing the LVTP-5 completely by 1974.

LVTPX-12 3rd prototype seen during trials about 1970. Very little of the original LVTPX-12 remains. Source: US Marine corps


The fate of these prototypes is not known, although two did get modified into test beds of the LVTRX-2 recovery vehicle with a 30,000 lb (13.6 tonne) winch fitted to the vehicle roof. The weapons cupola was removed at that point. Other experiments were the LVTCX-2 as a Command Variant and LVTEX-3 as an Engineering variant. A final variant planned but never built was the LVTHX-5 with a turret mounted 105mm gun, but all of these had little to do with the original LVTPX-12 and were now firmly in the realm of the LVTP-7.

Planned but never built LVTPX-12 based LVTHX-5 with turret-mounted 105mm gun. Source: Hunnicutt

Strangely, although the LVTPX-12 was ‘dead’ by 1969, the name crops back up again in 1982 with Congress allocating money to design and construct 2 sets of hydropneumatic suspension units for the LVTPX-12 along with clear armor inserts, suggesting the vehicle was still serving, performing a continuing role for testing and evaluation.

LVTPX-12 10th Prototype during trials with the 4th Marine Division at Camp Pendleton (California) in about 1971. One of two prototypes was received there for familiarisation training. Source: USMC and Hunnicutt

At least one of the original prototype vehicles survives at the Allegheny Arms and Armor Museum, Pennsylvania.

Surviving LVTPX-12 prototype number 12. Source: Harold Biondo

Illustration of the Assault Amphibian Personnel Carrier LVTPX-12 produced by Jarosław Janas, funded by our Patreon Campaign.

Specifications LVTPX-12 Track-Propelled Version

Dimensions (L-w-H) 26 x 10.6 x 8.6 feet (7.9 x 3.2 x 2.6 meters)
Total weight, battle ready 51,990lb (23.6 tonnes, combat laden) with 10,000lb (4.5 tonnes) payload
Crew 3 (Driver, Assistant Driver, Commander) + 25 troops
Propulsion GM 12V71T 800hp
Maximum speed > 10 mph (16.1 km/h) in water, 30 mph (48.3 km/h) on land. < 5 mph (8 km/h) in reverse in water
Armament 1 x 20mm Hispano Suiza cannon and 1 x 7.62mm machine-gun
Armor Welded steel up to 8mm thick

Specifications LVTPX-12 Auxiliary-Propelled Version

Dimensions (L-w-H) 26 x 10.6 x 8.6 feet (7.9 x 3.2 x 2.6 meters)
Total weight, battle ready 53,670lb (24.3 tonnes, combat laden) with 10,000lb (4.5 tonnes) payload
Crew 3 (Driver, Assistant Driver, Commander) + 25 troops
Propulsion GM 12V71T 800hp, 8V53T 400hp, or 8V71T 530hp
Maximum speed > 10 mph (16.1 km/h) in water, 30 mph (48.3 km/h) on land. < 5 mph (8 km/h) in reverse in water
Armament 1 x 20mm Hispano Suiza cannon and 1 x 7.62mm machine-gun
Armor Welded steel up to 8mm thick

Specifications LVTPX-12 (LVTP-7) Prototype

Dimensions (L-w-H) 26 x 10.6 x 8.6 feet (7.9 x 3.2 x 2.6 meters)
Total weight, battle ready 48,500 lbs (22 tonnes)
Crew 3 (Driver, Assistant Driver, Commander) + 24 troops
Propulsion GM 8V53T 400hp diesel
Maximum speed 40mph (64.3 km/h) land 8.4mph (13.5 km/h) water
Armament 1 x 1 x .50 cal. (12.7 mm) Browning M2 Heavy machine gun
Armor Welded 5083 aluminium


Final Engineering Report on the LVTPX12, Vol.1 Technical Study. (1965 ). Chrysler Corporation, Detroit.
The Water Performance of Single and Coupled LVTP7’s, with and without bow plane extensions. (1980). Irmin Kamm and Jan Nazalwicz. Ship Research and Development Center, Office of Naval Research, Department of the Navy.
The Stability of the amphibious Craft LVTPX-12 in Waves and Surf. (1968). Robert Patterson. Massachusetts Institute of Technology, Department of Naval Architecture.
Evaluation of LVA Full-Scale Hydrodynamic Vehicle Motion Effects on Personnel Performance. (1979). William Stinson. Naval Personnel Research and Development Center.
Bureau of Ships Preliminary Specification for Assault Amphibian Personnel Carrier (LVTPX12), (1964). Bureau of Ships
US Marine Corps FY82 Exploratory Development Program. (1982). Marine corps Development and Education Command
LVTPX-12 Accepted. (November 1967). J.H. Alexander. Marine Corps Gazette Vol. 51, Issue 11
Special Test for LVTPX-12. (June 1969). Marine corps Gazette Vol.53, Issue 6.
Amtracs: US Amphibious Assault Vehicles. (1999). Steven Zaloga, Terry Handler, Mike Badrocke. Osprey New Vanguard No.30
Bradley: A history of American fighting and support vehicles. (1999) R.P. Hunnicutt, Presidio Press

WW1 American prototypes

Jehlik’s Armored Vehicle

US armor ww1 USA (1916)
None Built

Right from the early days of World War One (1914-1919), many people, military, political, and civil, saw the need for armored vehicles to break what had stagnated into the focus of the war: a long and brutal slugging match between the great powers across the continent of Europe. Until the revelation that the tracked ‘tanks’ would be the selected primary method of waging mechanical war across the shattered landscape of Europe, many of these same visionaries and inventors considered wheeled vehicles and many came to exactly the same outcome.

Specifically, they rightly concluded that large diameter wheels were better off-road in soft ground and for crossing obstacles than small wheels, as they had a greater surface area over which to spread the weight of the vehicle. Many of these designs therefore simply become a ‘big-wheel’ landship. often in the form of a pair of large front wheels with a small stabilising wheel or tail behind. The 1916 design from Anton J. Jehlick is reflective of this but he went one step further. Jehlick designed not just large-diameter wheels, but an enormous roller in the manner of a large diameter cylinder on its side. Jehlick produced one of the strangest looking of these big-wheel landships.

Cross-section view of Jehlik’s armored vehicle. Source: US Patent 1195680


Jehlik had an unusual background for an armored vehicle designer. He was not an engineer or soldier. He was, in fact, a pharmacist, known at the time as a ‘druggist’. He had graduated as such by 1899, married Bertha and had had a son around 1910. By the time of the outbreak of war in 1914 and prior to the US entry into WW1 in 1917, Jehlik was working as a pharmacist in Chicago and, as an educated man, no doubt saw an opportunity to consider the need for an armored vehicle for the war he would have seen reported in the newspapers of the day.


Jehlik described his ‘armored vehicle’ as being in the manner of a “self propelled armored vehicle, comprising a large, heavy, cylinder containing the driving engines, ammunition, guns, equipment, and men to operate the artillery, guns, etc., and a second look-out compartment, mounted above the cylinder”.

That description of the machine really does not do justice to the size of it. The cylinder alone was to be 20 to 30 feet (6.1 to 9.1 metres) in diameter and 50 to 100 feet (15.2 to 30.5 metres) wide. In order to be protected against any possible enemy fire, Jehlik seemed to have ignored the practicalities of how heavy armor plating is and suggested armor plating of the type used on battleships, between 6 and 8 inches (152 to 203 mm) thick.

The Purpose

Disregarding for a moment Jehlik’s experience or lack of experience in vehicle technology or military matters, his design was very specific about what it had hoped to achieve. His goal was to create a vehicle capable of travelling across open-ground at “a high rate of speed” with the large heavy cylinder crushing obstacles such as barbed-wire entanglements. The armor would protect the men and vehicle from enemy fire. The roller would crush a path for troops to follow, and the size of the vehicle enabled it to cross trenches.


The armament for the vehicle was concentrated in the aft section, facing backwards, and positioned over a pair of small trailing wheels at the back. These wheels were actually ovaloid in shape, in the manner of an American football on its side, with the axle through the sharp points on each end.

Each gun was mounted in “a series of semi-circular sponsons” arranged at 30 degree intervals around the circular arc of the aft end. Further armament was provided fore and aft by means of loopholes through which rapid-fire guns were positioned. The exact type and number of weapons considered is not stated in Jehlik’s patent application, although the images provided show 5 large guns in the aft end on rotating floor-mounts. None of the rapid-firing weapons (likely he means some kind of machine gun) is shown however. The position of the oval loopholes through both walls of the heavy drum roller at the front would indicate positions for up to 5 such guns.

Why the primary armament faces to the rear is unclear, although possibly, Jehlik was picturing the vehicle rolling through enemy lines and then firing backwards to harass the enemy. This would, of course, mean firing your guns in the direction of your own following forces which could be extremely hazardous.

Plan view of Jehlik’s armored vehicle showing the unusual arrangement of gun in the back. Note that the vehicle is facing to the right in this image. Source: US Patent 1195680


Only a single member of crew is shown by Jehlik in his design: the driver. Positioned well above the rotating cylinder and fighting area of the vehicle, the driver sits in an armored cab that would have provided an excellent, if vulnerable view of the terrain in front and behind him. Bullet guards in front and behind him were intended to stop troops shooting upwards into the cab. Assuming just this one man was needed to steer the vehicle and control the propulsion, it would have needed another man to command it, presumably stationed with the driver on top and at least 10 men inside the main space just to operate the guns. Assuming just two men per gun and one per machine gun would have been required, this could have been as much as 15 or more men. Access to the vehicle for these men was concentrated in just a single rectangular floor hatch in the centre of the fighting compartment. The reason for this hatch was to provide a defence for the men inside against enemy ingress. Just wide enough for one man to enter/exit at a time, this would prevent the enemy from climbing in, and if they tried, the machine could reverse over them with the roller. Just as this may have sounded good for defence of the machine, it was also a major problem for the crew. All of the men inside would have to exit the machine one at a time through this small hatch and would have to hope the machine was not going backwards at the time. This slow egress is even more of a problem when it is appreciated that Jehlik considered the roof space of the fighting chamber as the best place for the fuel tank – above the men manning the guns.


This very large machine was to be propelled by three four cylinder engines, although Jehlik was careful to mention that ”any number” of engines could actually be used in order to achieve the “high rate of speed” he wanted from the vehicle. The engine/s for the vehicle were shown rigidly connected to gearing to the main cylinder at the front. No provision appears to have been provided for a gearbox of any kind. Steering was to be provided by the pair of ovaloid trailing wheels under the back of the fighting space which could be rotated by more gearing, creating a rear-steering sensation for the driver. The petrol to power the engines was held in a single large tank on the ceiling.


Given the enormous size of the machine, Jehlik’s Armored Vehicle could easily be dismissed as an unworkable idea. This would ignore the purpose behind using a ‘big-wheel’ type machine and why so many people at the time, and even subsequently, came to the same conclusion of using large-diameter wheels. They are simply more effective at crossing obstacles and gaps than smaller wheels. The problem is that they need to be huge to put down enough surface area onto the ground to spread their load adequately and that size also means they become unwieldy to move and an easy target for the enemy.

For Jehlik, this problem is slightly alleviated by making his ‘big-wheel’ into a roller which significantly increases the bearing surface which could carry the weight of the machine, but this also adds additional problems. Making the roller wider makes it significantly heavier, as the ludicrous amount of armor he was proposing, far more than would ever be needed to protect against small arms, would have to be extended to cover the full width of the machine. The wider the machine, the more armor and thus the more weight carried. Bearing that in mind, the rather puny suggestion of just a trio of four-cylinder petrol engines would likely have left his machine completely immobile on anything other than a very hard surface. It is not known whether Jehlik ever sent his ideas to the Army or Government of any nation, and if he did, it certainly was not adopted by anyone. The idea was as impractical for him as it was for every other ‘big-wheel’ machine before or since. Even so, the work of men like Jehlik, submitting their ideas at a time of war for a brand new type of weapons, adds to the understanding of how armored warfare evolved from its crude and often ill-conceived beginnings.


Upon his death, Jehlik was described in a periodical of the day as being of the “best Bohemian blood” and his cemetery records provides a city of birth for ‘Antone J. Jehlik’ as Prague, in modern-day Czechia (Czech Republic). This means he had immigrated to the US as a child, where a name like Jehlička (a more common Czech form of the name) could be Americanized like so many other names were modified as new citizens arrived and became US citizens. Jehlik passed away on 28th September 1920 after being in ill-health for some time. He was laid to rest in the Bohemian National Cemetery, Chicago. No trace of his pharmacy business remains today and Jehlik’s design has been forgotten.

Jehlik’s grave provides a birth date of 2nd May 1878, meaning he was just was just 42 years old when he died. He left his widow Bertha and a son who according to grave records was Eugene Franklin Jehlik. His death announcement in 1920 stated his son was 10, but Eugene’s grave shows a date of birth as 25th April 1918. The reason for this is unclear, but reporting mistakes are not particularly unusual. 1st Lieutenant Eugene Jehlik was killed on 28th November 1942 commanding an attack of M3 Lee tanks along a railroad track at Djedeida in Tunis, North Africa. He is buried at the Rock Island National Cemetery, Illinois.

Mortar and Pestle (appropriate for a pharmacist) monument for Anton J Jehlik at the Bohemian National Cemetery, Chicago. Source:

Illustration of ‘Jehlik’s Armored Vehicle’ produced by Mr. C. Ryan, funded by our Patreon Campaign.


Dimensions Roller Diameter: 20 to 30 feet (6.1 to 9.1 metres)
Roller Width: 50 to 100 feet (15.2 to 30.5 metres)
Crew ½ + up to ~15 men (Driver, Commander, 10-15 gunners)
Propulsion Three 4 cylinder petrol engines
Armament 5 large guns plus up to 5 rapid fire guns (machine guns)
Armor 6 to 8 inches (152 to 203 mm) steel


US Patent 1195680, Armored Vehicle, filed 10th January 1916, granted 22nd August 1916
Illinois State Board of Pharmacy. (1902). Annual Report. Vol.31
National Association of Retail Druggists (1920). Anton Jehlik. NARD Journal
A.J.Jehlik. Memorial ID 112057826.
1 LT E.F. Jehlik. Memorial ID 54973417.
Brig.Gen. Robinett, P. (2017). Armor Command: The Personal Story of a Commander of the 13th Armored Regiment, of CCB, 1st Armored Division, and of the Armored School during World War II. Arcole Publishing, USA
Atkinson, R. (2007). An Army At Dawn. Holt Paperbacks. USA

WW2 Italian Prototypes

Semovente Moto-Guzzi

Italy ww2 Kingdom of Italy (1921)
SPG – 1 Built

Mountainous northern Italy was treacherous tank country. Any war fought there on that terrain would be difficult. As well suited to defense as mountains were, any force which could deploy armor there, especially if they could also deploy some mobile artillery, would have a significant advantage. Combine this need with the late 1920’s assessment that the doctrine for new warfare should be highly mobile tactically making full use of light tanks combined with strategic mobility and Italy had a clear requirement for a vehicle able to achieve both goals.

Early work had involved testing the British Carden Loyd Mark VI light tank as well as the Mark V* light tank. Both had some advantages such as their small size and mobility but also some disadvantages in terms of the arrangement of automotive components.

Carden Loyd Mark V* fitted with Schwarzlose machine gun. Source: Ceva and Curami

The Carden-Loyd Mark VI entered service with Italy fitted with a single machine gun as the CV.29. The Mark V* was not assigned an official Italian service number but served as an inspiration for Ansaldo’s design for a 65 mm gun armed vehicle.

The Entry of Moto-Guzzi

The Italian firm Società Anonima Moto Guzzi, better known as just ‘Moto Guzzi’, is world famous for motorcycles and is Italy’s oldest motorcycle producer, having started in 1921. They even developed a completely enclosed armored motorcycle mounting a heavy machine gun. Of far more military use though they also worked on a fully tracked vehicle design. This design was a significant departure from anything seen before. It had two versions. The first was just a tractor which served as a test bed and the second was based on the same vehicle, but mounting a gun.

Semovente Moto-Guzzi 65/17. Source: Pignato


The basic structure of both vehicles was the same as the chassis were the same consisting of two independent track units mounted on arms and driven by an engine in the front. The driver and another crew member would sit at the back to command, drive and service the gun. The cabin seen in the test bed is a roll cage for protection from rolling over and it is unknown whether it was intended to be a permanent feature on either vehicle. Certainly, some sort of protection for the crew would have been required for the Semovente (a self propelled gun for infantry support) should it have been built.


The suspension was very unusual and consisted of two tracks completely independent of each other consisting of four wheels on a supporting bar, along with an idler and drive sprocket all suspended on three rotatable arms. As one track moves down, these arms rotated, ensuring that the body of the vehicle remains horizontal and the tracks move at different heights. The arrangement was unlike anything before in Italy and had the significant advantage of allowing movement across the side of a steep slope. The drawing from Moto-Guzzi suggests a side slope of approximately 45 degrees. Drive appears to be provided via the front sprocket as per the prototype vehicle. However, the exact arrangements of the fastenings for the wheels is likely to have differed on the envisaged production vehicle from the test bed.

The drawing of the Semovente shows four road wheels and the test bed also has four wheels but each wheel is paired so as to leave a gap down the centre for the track guide. A large external supporting bar ran from the front sprocket to the rear idler wheel on which were mounted small track support wheels.


Moto-Guzzi made a lot of motorcycles for the Italian Army and had a lot of experience in small high powered engines but it is not known what engine was used in the design. The cylindrical object at the front of the test bed is unknown but appears to be either the exhaust or possibly a fuel tank, which would have to have been completely rearranged on a production vehicle to permit the gun to be used.

The Moto-Guzzi designed vehicle chassis, during mountain trials, showed off its incredible ability to remain upright even when traversing a steep side slope. This flexibility is shown in the modified image from the outline drawing. Source: Ceva and Curami, and Pignato (modified) respectively
1: The Moto-Guzzi vehicle showing its ability to climb very steep slopes even in snowy alpine conditions. 2: The Moto-Guzzi traversing a steep side slope in snowy conditions. The ability of the tracks to move vertically keeps the body horizontal. 3: The Moto-Guzzi tractor during trials. Note the large logo on the cab. Source for all: Pignato
Enlarged view of the logo on the side of image 3. Source: Pignato


The Moto Guzzi, as a tractor, would (amongst other duties) be used for towing light guns. As a semovente, it would mount the 65mm Model 1913 mountain gun. This gun was made by the Royal Arsenal in Naples and was 17 calibers long. It was an ideal gun for use in the mountains as it could be broken down easily into 5 loads for transport. It fired the 65mm x 172R rounds just under 5kg in weight and ammunition for the gun was mainly high explosive, but a specific shrapnel shell was also available. Later, an armour piercing (AP) and also a hollow charge shell were available for it.

65mm L.17 Model 1913 Mountain gun. Source: Italian Ministry of Defence


The Semovente Moto-Guzzi was a very advanced design for the time. A design ideally suited to the rigours and demands of mountain warfare and carrying a weapon specifically designed for the job. It was not to be, however. The Moto-Guzzi design, either for work as a tractor or as a semovente, was discontinued by 1930. The Italians would have to look elsewhere for a suitable gun carrier to meet their unusual requirements.

Illustration of the Semovente Moto-Guzzi produced by Andrei Kirushkin, funded by our Patreon Campaign.


Crew 2
Armament 65mm L.17 Model 1913 mountain gun

Italian Artillery of WWII, Ralph Riccio
Iron Arm: The Mechanization of Mussolini’s Army, 1920-1940, John Sweet
Gli Autoveicoli da Combattimento Dell’Esercito Italiano V.2, Pignato and Cappellano
Italian Armoured Vehicles of World War Two, Nicola Pignato
La Meccanizzazione dell’Esercito Italiano, Ceva and Curami

WW2 Australian Prototypes

Grasshopper Light Tank

Australia Australia (1944)
Light Tank – None Built

Naming tanks is a complex business. The name is supposed to capture the essence of the vehicle, inspire the crews with confidence and the enemy with fear and, as a result, many such vehicles are named after large raptors, predators, and wild animals. There are also cases in which vehicles receive a less aggressive name, but one which perfectly encompasses the entire vehicle. One of these perfectly named vehicles is the Australian Grasshopper Light Tank designed in March 1944. It is also referred to in Army correspondence as “Grass-hopper”, although in the original 1944 letter to the Adelaide Office of the Inventions Directorate, the designer referred to it as the ‘“Grass Hopper”.

Designed by William (Bill) Hope Murray of Verdun, South Australia, the Grasshopper was quite simply a light tank fitted with twin helicopter rotors “somewhat on the lines of a helicopter” to enable it to fly clear of obstacles. In other words, this was not a flying tank, but a light tank capable of great leaping jumps – at least in theory.

The Design

The aerial aspects of the design were inspired by the Igor Sikorsky YR-4 and VS-300 ‘Helicopter’ designs, although the designer did not have much information to work from based on wartime military security. Murray had, however, pieced together information on the YR-4 as being 35 feet (10.7 m) long with three 19 feet (5.8 m) long rotor blades powered by a single 180 hp engine. Capable of less than 100 mph (160 km/h) and an maximum altitude of 14,000 feet. (4,270 m), Murray had some data, but had no information regarding payload, range, fuel consumption, etc. He did however, estimate the weight of the machine to be 6000 lbs. (2.7 tonnes) gross and capable of carrying a load of ⅓ of that (~0.9 tonnes).

Sikorsky R-4 helicopter in RAF Service as the ‘Hoverfly I’ in 1945 showing the relative size and scale of this small, light helicopter. Photo: wikimedia. The VS.300 (right) flown by Igor Sikorsky himself wearing his trademark Homburg. Photo:

For the VS.300, Murray had the specifications as 27 feet (8.2 m) long with a 90 hp air-cooled engine, capable of 80-100 mph (130 km/h to 160 km/h ). Whilst these specifications for the VS.300 and YR-4 are not completely accurate, it is clear that Mr. Murray had done his research well considering the constraints of wartime secrecy and the newness of helicopter technology.

In outlining his design, however, Murray skipped over significant pieces of information. His design was really more of an outline for what could be done with the helicopter technology to blend it with a light tank than a very specific and detailed design.

What can be discerned from his plans though is that the machine was to have a crew of at least 4 or 5 men. Two men (although there appears to be space for at least a third man) sat at the front behind the unique (for a tank) aircraft style cockpit windows. Presumably one of them was the pilot/driver. Above the position of the driver/pilot was the main gun position in a very unusual elevated semi-rotating position. The gun is not specified, but even a 2-pounder gun and ammunition would have been a big demand on the payload available. Although there is only one crew member shown in this turret, the commander would have also been in the turret, as this was the highest crew position and would likely have had to have a second crew man to man the main gun or else the commander would have to load, aim and fire the gun, all while commanding the tank. At the rear of the tank was another member of the crew in a bubble-shaped turret with excellent visibility from the glass or perspex bubble he was sat in.

Looking side on in the drawings provided, at least two unspecified weapons, likely machine-guns, were at his disposal, positioned unusually on top of each other. This might have meant that, in fact, Murray was picturing a quadruple mounting as a side view, but irrespective whether the turret was to mount 2 or 4 machine-guns, the positioning was very poor. It would be able to provide very little anti-aircraft fire when on the ground, as firing upwards was hindered by the rotors, very little fire backwards due to the position of the tail, and no fire at all to the front of the vehicle. Neither could it provide any fire downwards when flying.

Between the fore and aft turrets lay the engine for the rotor in a small vertical box structure. Below this was a large space in the hull for the engine of the tank which actually ran under both the fore and aft turrets. This was actually one of the better parts of the design as there was no complicated linkage to use a single engine, but it would have added significant weight the machine.

An artistic interpretation of the Grasshopper Light Tank based from the description and drawings. Drawing: Author’s own.

Murray had selected dual rotors for extra lift, giving the impression of a helicopter with counter-rotating rotors, although this is not the case. These would, he felt, provide ample mobility for independent attack surmounting obstacles and hopping past enemy lines. In deployment though, Murray seems to have not done as much research as he did for the flying elements of the design as he envisaged the tank being used in “large numbers [with] a fast fighter support” where it “would be a very dangerous weapon”. Obviously, with a large accompaniment of fighters, the need for the anti-aircraft part of the armament would seem superfluous. He wanted the design to be “as light as possible”, which would clearly mean little or no armour could be used, and he was unclear on what he meant by a ‘belly periscope’, although it is likely to do with trying to see where it was landing.

Composite of views of Murray’s Grasshopper tank design from Australian Army Inventions Directorate file 15430, 1944. Source: Author


The visibility, despite the huge windows, was very poor. The pilot would effectively be unable to see where he would be landing this large helicopter through the front windows and instead would have to use the large windows in the lower front portion of the hull. All of this glass effectively meant that this vehicle was going to be unarmored. Comparing protection and weight, bulletproof glass was, and still is, substantially worse than steel, therefore, to be even bulletproof, this glass was going to have to be heavier than the weight of bulletproof steel. For a ‘light tank’ this was a significant problem with the concept and one unaddressed by Murray in his letter. Assuming the ‘glass’ to actually be aircraft type perspex at the front and back, this would mean that the tank was not even bulletproof. Although this would mean that the weight would be very low, it was a significant flaw for a ‘tank’ meant for fighting the enemy head on. The rear turret would be useless when attacking forwards and almost useless for anti-aircraft use. The arc of fire was too constrained by the design and the crew member would have been idle the majority of the time on the ground or in flight.

The selection of two engines, one for flight and one for driving, did ensure mechanical simplicity of design, but it also added a huge amount of unnecessary weight. When on the ground, the vehicle had to haul around the aircraft engine which had no use on the ground and when trying to ‘leap’ or fly, the helicopter now had to lift the tank engine as part of the payload rather than as part of the drive for the rotors.

The whole of the rear compartment of the tank from the back of the aft turret to the rear of the hull was intended to be for fuel for both engines. This fuel tank would have been huge and also very heavy but was at least at the back of the vehicle where it was better protected from enemy fire. Projecting from the back of the tank was a large rotor tail made from hollow tubing with a skin around it and had no need for armouring or protection of any kind as it contained no fuel, mechanicals (other than the tail-rotor drive shaft) or men. It is not clear if this tail could be detached. One further point of note in the design which received no attention from the Army Inventions Directorate was the bomb doors. Positioned directly behind the driver in the bottom of that compartment was a ‘flap’ through which bombs could be dropped, which perhaps accounts for the second crew member in that location, who is possibly supposed to be the ‘bomber’.

Suspension for the vehicle is not mentioned, although given its light nature and contemporary designs could have been either springs or ‘Christie’ type shocks, but relies upon 6 road wheels per side. Some illustrations show the lead and rear road wheel as being larger than the other four but may have simply been an illustration issue. Either way, the central four wheels were relatively uniformly spread from each other but further from the front or rear wheels than they were from each other. It is not clear from the design at which end the transmission for the tank was supposed to go or from which end drive would be applied to the tracks.

The overall outside of the vehicle is extremely heavily curved with what appear to be the two crew access hatches located on either side, although how they would get in and out from their compartments is unknown.


The design, like so many others, was evaluated dispassionately by the Australian Army for its merits and defects and what it might bring to the Allied war effort. The assessment here was, as with many others, not a positive one. The Grasshopper tank simply had too many flaws to warrant further investigation, flaws like the fact that when in the air the machine had absolutely no means whatever to fire down at any enemy below it. The incredible vulnerability of such a slow moving, rather large and thinly protected vehicle passing slowly and low over enemy positions was a very tempting and vulnerable target for the enemy.

The positioning of the weapons was also criticised. In particular, the two guns in the turret at the back were very poorly located. Intended, as anti-aircraft guns these guns would be completely useless in the air as well as on the ground firing upwards or to the rear as they would be firing through the tail/tail rotor, or the rotors above. Whilst methods for timing machine gun fire through a propeller had been developed during WW2, there was simply no way to time the fire of these weapons through these rotors. Any attempt to use the weapons during flight could, therefore, lead to the crew shooting down their own vehicle.

Although a leaping tank could be very useful tactically and solve many problems, the idea was simply impractical and technically impossible at the time.

This supposed light tank was simply going to be too heavy to be useful as a light tank, too big, and too slow. As a helicopter, it had neither the range nor the altitude and speed to be useful and was not going to be able to be armored sufficiently for ground work nor protected enough for aerial work.

Layout wise, the elevated position for the main gun was desirable, although it was in a limited traverse turret limiting its usefulness and, without a second crew member in it, would be very difficult for the commander to use whilst commanding the tank. The location of the bombs and bomb doors too ensures that, should the tank hit a landmine, the complete destruction of the vehicle and loss of crew was virtually guaranteed.

The road wheel design seems to have been inspired from that of the Valentine, with either 6 uniformly sized road wheels or larger fore and aft wheels with four smaller ones in between, but would have ran on quite thin tracks. Thin tracks increase ground pressure from the vehicle and this design is very wide meaning it would be more vulnerable to becoming grounded out when crossing rough terrain.


The assessment was not completely dismissive however, the general concept was seen as a very desirable one. There would be no need for light bridging equipment to get the tank over trenches or ditches, rocky escarpments or walls. It could leap heavily forested sections, rivers, very soft ground and, most importantly, enemy minefields and wire entanglements. So desirable is the overall concept of a ‘leaping’ vehicle that such ideas continue to be entertained by various military forces, albeit with the same level of success that Murray’s Grasshopper had.

Bill Hope Murray, service number S76738, had already enlisted in the Australian Army (4th Battalion Volunteer Defence Corps) in April 1942, aged 45 (he was born January 1897), which meant that he was 47 when he submitted his design. He concluded his Army service in October 1945 with the rank of Lance Sergeant having served his nation in the war. His design might not have helped to win the war, but his service played its part, and even at home in Australia, he and his wife raised money for the war effort. Post-war, Murray went back to his life in Verdun and eventually retired as the architect for Public Buildings as well as operating a family farm. The patriotism that spawns such ideas should not be discounted though, neither the desire to help which underpins so many of these well intentioned, albeit flawed tank designs.

Bill Hope Murray. Source: Butler

Illustration of the ‘Grasshopper’ Light Tank produced by Yuvnashva Sharma, funded by our Patreon Campaign.


Dimensions ~8.2m (27’) long
Crew 4 – 5
Weight 2.7 to 3.6 tons
Armor Bullet Proof


Australian Army Inventions Directorate file 15430, 1944
Murray, Bill Hope at
The Disused Grunthal Gold and Copper Mine, 3rd August 2013 –
Reg Butler, Hahndorf Memorial Institute – ‘50 Years On’ – Golden Jubilee of the Hahndorf Institute Extensions 1957-2007

WW1 American prototypes WW1 Austro-Hungarian Armor

Kempny’s Armored Automobile

Austria-Hungary/USA (1916-18)
Armored car – Blueprints Only

World War One had started much along the lines of previous wars. Political saber-rattling, followed by posturing, declaration of war and mobilization. Despite the growth in industrial potential across Europe at the turn of the century and the perfection of the machine gun as a practical weapon of war, the armies of Europe in 1914 went to war in much the same way as they had done in the previous century and yet were quickly faced with a new reality. Their men were easy prey to the rapid-firing effects of the machine guns.

There had been numerous ideas before the war for armored machines, but there was little impetus to develop one until the slaughter of WW1. That fate had befallen an Austrian called Gunther Burstyn, who had patented a very crude form of armored vehicle before the war but had done little with it. Another Austrian, Karl Kempny, far less well known or remembered, was living in Cleveland, Ohio, USA during the war. Kempny was not the visionary that Burstyn was, but was certainly quick to see the potential of armor. In 1916, he submitted his own ideas for an armored vehicle carrying heavy armament but still mounted on wheels. Future armored power was going to be best deployed on tracks, not wheels as envisaged by Kempny.

Divided Loyalties?

Little is known of Karl Kempny and any attempt to research the man online is sadly frustrated by a hockey player of the same last name playing for Cleveland. What is known of him, therefore, comes only from his patent applications. His name was given as Karl Kempny and he described himself as a subject of the Emperor of Austria, albeit living in Cleveland, Ohio, USA at the time. Whilst WW1 had started in the summer of 1914, and Austria-Hungary had been involved in military action right from the start, it was not until 1917 that the United States had come into the war. It was not, in fact, until 7th December 1917 that the US actually declared war against Austria-Hungary, even though it had already done so against Germany that April. At the time that the patents were submitted, therefore, between 20th November 1916 and 1st February 1917, there was no state of war between the USA and Austria-Hungary for Kempny to worry about. What is more interesting though is that this Austrian citizen was granted two patents for military designs in 1918 (including this armored automobile) at a time when the US was at war with his home country. To whom was the design intended then? Was Kempny, filing in 1916, suggesting his design was for use by Austria? If so, then he did not file an application for it there. It seems more likely that Kempny, a first-generation immigrant from Austria, not yet naturalized as a US citizen, filed his patent in his new adopted country for use either by them or for commercial purposes. Whilst Austria might have a claim on Kempny via ancestry, it would appear his vehicle is more appropriately assigned as an American one.

The Patents

As alluded to in the preceding paragraph, there was more than one patent. In fact, Kempny submitted three patents, two in 1916, and one in 1917, all for military equipment. The first, titled ‘moveable shield’, was one of dozens of wheeled, armored shields being suggested by a myriad of inventors, commentators, and military men throughout the First World War. Almost without fail, the designs were crude, clumsy and found no use. A man-propelled shield which was thick enough to be bulletproof was simply too cumbersome and heavy for even a small number of men to move. And that is before consideration is given to moving it over the tortuously muddy conditions of the battlefields of WW1 on the Western Front or the often vertigo-inducing mountainous terrain of the Southern (Italian) Front. Despite its flawed utility, his shield was nonetheless granted a patent in July 1917.

During the war, he filed his application for his armored automobile that December, followed three months later in February 1917 with a design for a bulletproof helmet. The helmet is certainly a novel design and one really has to wonder if Kempny was even serious with it given the design. Ludicrously tall and covered with spikes, the helmet consisted of a protective dome over the top of the head over which a taller helmet was fastened by means of springs. As if that was not impractical enough, the outside of this design was then clad all round the outer surface with spikes. All of that weight, precariously perched on top of the wearer’s head, was secured by just a single thin chin strap, meaning that as soon as the wearer might run or duck for cover, this spiked affair on top of his head would simply fall off and either impale him, another nearby soldier, or just get stuck in something. Truly, there can not be any helmet design which was less practical or realistic and perhaps that is why Kempny stopped submitting patents. He was just wasting his money on pure fantasy silliness.

Kempny’s ludicrous design for a bullet-proof helmet. US Patent 1251537

The design between the shield and the helmet though certainly has some elements of fantastic and impractical thinking, but also of some common sense and is worthy of some consideration.

Armored Automobile

Filed in December 1916, the design was not approved until October 1918, just before the end of hostilities. His design was specifically intended as a vehicle for repelling attacks by enemy infantry but also for mounting rapid-fire guns in bullet-proof mounts. The overall layout is clearly that of a standard truck with an engine at the front, directly over the front axle, mounting a pair of steered-wheels. A further axle at the back was also fitted with a pair of wheels.

The body of the vehicle was essentially a large rectangular prism, flat vertical sides and rear and a flat horizontal roof. The front though was different. A large rounded section angled steeply backwards, going from above the engine to the roofline with a large horizontal viewing cupola halfway up. This cupola was for the driver to see out of and appears to have been located centrally behind the engine. A second cupola, fully rotatable, was mounted behind the point where the angled front met the roof and would provide the vehicle commander with all-round vision. Located centrally and at the front, the driver should have had good visibility of the ground in front of the vehicle, but he would have been unable as Kempny drew on a large curved shield extending from the front of the vehicle and up to a level above that of his cupola. Thus, the driver’s view ahead would be severely limited. The purpose of that large curved section at the front was to primarily force down barbed down as the vehicle approached but it also served as armor for the front of the vehicle, deflecting bullets away from the men inside.

Access to the vehicle was to be via a single large rear hatch with vision provided by the cupolas and by various vision slots in the side of the hull and in the sponsons.

Kempny’s Armored Automobile as shown on US Patent 1282235. The removable socket-type sword bayonets sticking out of the side make a fearsome if somewhat useless impression.

No mention is made of armor except it would presumably have been armored to at least the level of being reasonably well protected against a service rifle. This would mean protection in the region of 8 mm or so of steel. As far as crew goes, there would need to be at least 4 men inside, a driver, a commander, and one man per gun. There is a lot of space inside the body and one use Kempny envisaged involved the removal of weapons and use as simply an armored lorry. This would suggest enough space for half-a-dozen or so more men even when armed.


The first and most obvious weapon on the vehicle are the spikes. These are actually sword bayonets mounted in rows along the side of triangular extensions attached to the side of the vehicle with the intention of making it harder to approach/climb when stationary and also to scythe through enemy troops when mobile. Thankfully, Kempny decided that these bayonets should be able to be folded away when not in use, or else the number of enemies they would be killing would surely only have been outweighed by the numbers of its own men, passers-by, and animals which would have been cut limb from limb as it went by. Despite the appearance of having a large cannon in each of the sponsons sticking out of the side, the Kempny design was to rely instead upon a pair of ‘rapid fire guns’ which could be machine guns or a cannon of some description with one in each sponson. Each gun was mounted on a rotating pedestal providing fire to the front, sides, and even to the rear. This type of mounting in an armored car, a sponson projecting from the side, was most likely the result of seeing exactly the same manner of armament carried on the first British tanks which were receiving a lot of press coverage at the time. As these were projecting from the side, it would mean the vehicle would be able to deliver fire straight ahead as well as to the sides. It would also affect lateral stability, as significant weight would be placed outside the wheelbase.

Kempny’s Armored Automobile as shown on US Patent 1282235. The interior layout shows the emphasis of the design on those large side sponsons for the armament.

A Lithuanian Connection?

One small added mystery to the identity of Karl Kempner comes from the signatories to his armored automobile patent, acting as witnesses: Stanley Stanslewicz, and A.B. Bartoszewicz. Bartozewicz was also a witness on his shield patent and appears to be Apdonas B. Bartoszewicz (also known as Apdonas B. Bartusevicius) who ran a Lithuanian-language publishing company in Cleveland which included the printing of the newspaper Santaika (Peace) in 1915 and which changed name to Dirva (Field) in 1916. The fact that Bartozewicz witnessed two of Kempny’s designs suggests that they knew each other reasonably well, although the nature of the relationship is unclear. Perhaps they were related or business partners, or that Bartozewicz was a notable person locally, we may simply never know. Nothing today remains of Kempny’s legacy and even Bartoszewicz is almost forgotten. Only his name remains on a building in Cleveland.

The physical legacy of Bartoszewicz in Cleveland.Source


Kempny’s shield added nothing new to the multitude of such designs and met with much the same fate. His helmet is memorable because it is simply such a totally impractical concept. His armored car however, is a different story. It was never built, never saw combat, and made no effect on the pursuit of the war so could easily be dismissed, but this would be wrong. His vehicle’s design clearly shows a popular mindset amongst designers at the time and just how little was understood about the true conditions at the front. Designs which could only operate on good surfaces and not the mud of Flanders are common, a complete misunderstanding of the conditions despite plenty of photographs available.

Yet, despite that misunderstanding, Kempny did foresee a multi-purpose vehicle, one suitable for carrying men and goods as much as for combat, a vehicle with weapons mounted in sponsons projecting from the side in the same manner as was used on tanks and an appreciation of the problems of barbed wire.

Kempny wanted to simply crush it down and roll over it, things which were tried and failed. The influence of the British tanks of 1916 can even be seen in the design, yet overall the design was still a retrograde one.

It is not known who, if anyone, may have seen Kempny’s design at the time and it is unlikely that it had any influence on following designs, especially the wholly impractical idea of the sword bayonets on the side, but Kempny’s design illustrates the time well – a no doubt well-meaning amateur designer, a first generation immigrant to the US trying to have his voice heard during the maelstrom of war. Whilst his design for an armored automobile went nowhere, received no orders, and was never built, Kempny’s armored automobile provides an insight into how the war was still being seen on the home front at the time.

Illustration of Kempny’s Armored Automobile produced by Mr. C. Ryan, funded by our Patreon Campaign.


Crew est. est. 4 driver, commander, 2 x gunners) + ~ 6 men
Armament multiple rows of sword bayonets, 2 x rapid-fire guns
Armor Bulletproof
Engine a ‘suitable motor’


US Patent 1234174 ‘Moveable Shield’, filed 20th November 1916, granted 24th July 1917
US Patent 1251537 ‘Bullet Proof Helmet’, filed 1st February 1917, granted 1st July 1918
US Patent 1282235 ‘Armored Automobile’, filed 18th December 1916, granted 22nd October 1918
‘Dirva’, Ohio History Central

WW1 British prototypes

Macfie’s Landship 1916-17

British Empire (1916-17)
Design Only

Many of the early ideas for armored land warfare which were suggested to break the stalemate of trench warfare in WW1 were impractical, outlandish, or otherwise beyond the technology of the day. Indeed, for a new type of warfare, a new type of weapon was needed and several nations had come to this conclusion at around the same time. With any new technology, there are also those ideas that were, in hindsight, totally useless, and likewise, those whose potential was not exploited. One design and one man which were not exploited for their potential were the 1916-7 landship and its designer, Robert Macfie.

As an early proponent of tracked warfare, the American-born Robert Macfie had managed, by the end of 1915, to achieve little more than making sure that the official British work on Landships would be track-based rather than wheel-based. His own work had been ignored, sidelined and then either stolen or copied. His career in the military was a flop and his commission had been cancelled, so on the face of it, Robert Macfie should have stopped working on tracked vehicles and focussed his attention elsewhere. However, Macfie was a stubborn man, and stubbornness can lead to both success and failure in life in equal measure. Not content with the multiple rebuttals for his work up to that point, Macfie had one final tracked vehicle endeavor up his sleeve, the culmination of all of this development work up to that point.

A New Beginning

Almost a year after his initial design and 12 months after his military career had ended, Macfie had submitted a second landship design with improved features. The new and improved vehicle shared some features of the 1916 design, such as having multiple tracks, but it would not have the complex nose-mounted track-bogey of the previous vehicle. Instead, the vehicle was to use a permanently elevated lead track described as:

“a self-propelled vehicle, of an endless portable track mounted at the front of the vehicle in such a position (for example sloping forwardly and upwardly) as to provide a driving means which can engage a steep bank or like obstacle, and means for coupling the said track to the engine of the vehicle”


Along with the permanently fixed and elevated lead bogey, the general shape was also different. The sides were parallel for the portions along the main pair of tracks, which were at the back, but after this, the shape became a wedge pointing forwards with the elevated track at the front point. A pair of large sponsons were located at the back, one on each side. Traction was provided by means of four tracks. Two at the back for propulsion, a third track located ahead of those on a steerable bogey and which provided the steering for the vehicle, and the fourth set at the front elevated at about 60 degrees to assist in climbing.

Drawings of Macfie’s December 1916 design, from his 1917 US Patent application.
Drawings of Macfie’s December 1916 design, from his 1917 US Patent application.

Unlike the 1916 design, there was no provision here for lowering armored shields over the tracks to protect them or to create a mobile fort. With this new and improved layout, Macfie clearly felt the old and complex shields were now superfluous as:

“The [new] general arrangement, moreover, renders it more easy to provide effective shields for the portable tracks”


Having first submitted this design in December 1916, Macfie submitted an amendment six-months later in May 1917, making it clear that the body he had outlined could actually be made in any desired manner, armored, and armed with machine guns and other weapons such as artillery. The power-plant for the vehicle was specifically omitted just as before because the patent and intention were more concerned with the overall layout of a vehicle and the use of steering bogies and raised front tracks.

Mounted in the sponsons at the back, the weapons would have been able to provide fire across a wide arc on both sides of the vehicle. Despite the angled shape of the design, these sponsons would still not be able to fire directly to the front. In keeping with the ‘tanks’ which had, by this time, appeared in the popular press, side-mounted sponsons would be able to fire down the length of a trench as the vehicle crossed it.


This landship from December 1916 was much more clearly thought out than his January 1916 version and much less complex in terms of gearing. Although Macfie did not file a patent application in the United States for the January 1916 landship, he did file one for the December 1916 landship, filing it in September 1917. That application was filed not just in his name, but in conjunction with Traction Development Limited. When, less than a week later, he also filed the design for a patent in France, it was only in the name of the company, suggesting that Macfie may have sold the rights to the design, perhaps because he was in financial difficulty.

This design was, compared to his other designs, the best of the bunch, with a more practical and less complicated layout, armament mounted in sponsons, and a special nose track for climbing which was simpler than his 1916 design. Even so, it met with no more success than his other ideas. The patent applications in France, Britain, and the USA would indicate that Macfie was seeking some other potential markets for his ideas, but with a functional ‘tank’ already in operation on the Western Front by this time, it is hard to see why anyone would go for a totally new and untested design. As such, Macfie was left, by the end of the war, having not produced a single functional vehicle. He was awarded a pittance by the post-war inquiry into the invention of tanks and shortly after returned to the USA, no doubt a bitter and disappointed man. Robert Macfie died on 9th February 1948 in New York, aged just 67 years old, having lived long enough to see the tanks he helped to originate become the dominant land weapon of the age.

Illustration of Macfie’s 1916-17 Landship Design, produced by Mr. R. Cargill, funded by our Patreon Campaign.


Armament Machine guns, artillery or other
Armor Bulletproof


Hills, A. (2019). Robert Macfie, Pioneers of Armour Vol.1. FWD Publishing, USA 
British Patent GB124450 ‘Improvements in or relating to Motor-Vehicles’. Filed 3rd January 1916, Accepted 3rd April 1919
US Patent US1298366 ‘Motor Vehicle’. Filed 4th September 1917, Accepted 25th March 1919
Proceedings of the Royal Commission on Awards to Inventors: tank 1918-1920

Robert Macfie (Pioneers of Armour)
Robert Macfie (Pioneers of Armour)

By Andrew Hills

The foundations and principles of modern armoured warfare did not appear out of a vacuum, and nor did the machines of WW1 and WW2. Their development was full of false starts, failed ideas, and missed opportunities. Robert Macfie was a pioneer in aviation at the turn of the century followed by work with the Landships Committee on tracked vehicles to break the stalemate of trench warfare. Although his tank designs never saw combat the work he started was carried on by other pioneers and helped to usher in a dawn of armoured and mechanised warfare.

Buy this book on Amazon!

WW1 French Prototypes

The Boirault Machine

ww1 French Tanks France (1914-20)
Prototype – 2 Built

Probably the most unlikely looking of all tank designs is the famous Boirault machine or, to give it its full name, the Fortin Automobile ‘la Machine Boirault écrase barbelé’ (Eng: The Barbed Wire Crushing Boirault Machine). Usually used as an example of bad design or for mockery, these machines were an ingenious and inventive means of crossing trenches and broken ground and were not the product of some random inventor. These designs came from the highly respected and very well experienced French Engineer Louis Boirault. Boirault was an undenied expert in railways in his position in the state rail system. Specifically, he was an expert in railway couplings, and between 1900 and the 1920s, having had amassed over 120 patents to his name, mostly for railway connected matters.

Monsieur Louis Boirault (right) during his work with the French National Rail System before the war. French National Library Reference #6928904

When, in August 1914, Germany declared war on France, few could have envisaged the mostly static grinding butchery that the war would become. Most armies were grossly unprepared for the mass slaughter brought about by the industrialization of Western Europe, with machine guns and artillery becoming the predominant weapons of war rather than the rifle or lance. It was not long, however, before the armies settled into lines of defense with each side unable to inflict a decisive breakthrough and defeat the other. By the end of 1914, with casualties mounting rapidly (over 300,000 men by the end of the year,) inventive minds across Europe were awaiting a breakthrough, so it is no surprise that a man such as Boirault should turn his engineering prowess to the difficult question of how to take the war to the enemy across open ground cut with trenches, guarded by belts of barbed wire, and covered by machine gunfire. An armored machine was the inevitable outcome, but wheeled vehicles could not cross such ground and, in 1914, there were few tracked machines available in France on which to base such a vehicle. Whilst the French designs eventually ended up using the Holt tractor chassis (a chassis rejected early in 1915 by the British) there was, with the war ground to a halt, in 1915 a period of experimentation looking at methods of attack by machine. Monsieur Boirault’s idea would stand out amongst many others as the most original.

Origins December 1914 to February 1915

In December 1914, Louis Boirault and his company (the Boirault Company) started considering the question of crossing no man’s land (the area between the front lines of the opposing armies) and proposed a solution to the French War Ministry. With few other options, the War Ministry authorized production on 3rd January 1915, and under the eye of a commission to supervise the invention led by Paul Painleve (Minister for Inventions), construction began. By February 1915, his considerable experience with railways led to his design ideas being presented to the French military, which were evaluated along with various other, often odd ideas, offering solutions to the same problem.

Theory of design for the machine as shown in French Patent FR513156(A) 1919.

An Inventive Design

Instead of wheels, he envisaged tracks. Not the sort of tracks used in some agricultural vehicles of the age, but much more akin to what he knew: train tracks. Obviously, you cannot lay train tracks for a train to attack the enemy, so the answer is to simply bring them with you. With two parallel rails attached horizontally by 4 steel girders, this machine used 6 sets of these frames to carry the machine. The ‘rail’ parts and the ‘sleeper’ parts spanning between them across the frames would carry the load of the vehicle on the soft ground and as the machine advanced they would simply be picked up behind and carried back over the top of the vehicle to be laid in front once more. This is exactly the same principle as other track layers but the single-track nature and sheer size of the tracks in question make it look more unusual than it actually is as an idea.

Theory of frame-related movement. French Patent FR513156(A) 1919
Boirault Machine during testing. Source: Vauvallier via the Touzin Collection

Each frame was simple and rugged and obviated the problems found with other tracked vehicles of the time. There was no track sag (where the tracks drop away from the body of the vehicle or running gear), there was no lateral slippage (tracks slipping off the wheels sideways), and, most importantly, the six frames required just 12 pins to attach them together, which reduced the chances of a pin failing because they could be made very substantial, like a railway car coupling. These three problems that plagued early tracked vehicles were ‘solved’ so to speak, in one simple move, but the price of this solution was also, eventually, to be its greatest failing.

The giant frames on which the machine ran. Also, note the complex arrangement of drive chains. Source:
The first Boirault machine showing the angularised nature of the frames nicely. Source: Granier

The machine inside this 6-frame-track run was an unusual rectangular based pyramid with the edges of the pyramid constructed from the same type of heavy steel girders used in the frame sections. Where these pieces met the frame at the bottom (2 contact points per side) and at the top of the triangular sides (one per side) there was a heavy steel roller along with a large square section ‘doughnut’ forming a path through which the frame sections would run. This was a very rigid and robust system for which there would be no movement at all of the track. Effectively, it was as if a train’s wheel was being held firmly onto a track and taking all of the weight. Within this pyramidical structure, at the heart of the machine, lay the Aster petrol engine. This engine drove the tracks by means of steel drive chains but was underpowered for the task and capable of propelling the machine at just 1.6 km/h (1 mph). It is worth noting that at least one contemporary source states that the 80 hp petrol engine used came from the Filtz agricultural tractor.

Under this engine, and within the body of the machine, a crew would have to be accommodated somewhere, along with fuel, and, at some point, some kind of offensive weaponry. The drive was at the top of the machine. Whilst this had the huge advantage that the drive could not get clogged with wire or mud, it also meant that the crew sat underneath all of this, making it harder to see, and raised the center of gravity for the machine. The bottom of the pyramidical vehicle can be seen in photographs to have a curved bow and stern section and would slide along the frame on its wheels crushing down any obstacles such as barbed wire staves in its way.

Two views of the Boirault prototype no.1 during testing in 1915. Source: Scientific American


The design, despite being functional and able to move under its own power, was unsuitable for war. The French military authorities found the machine to be robust and ingenious but the faults were obvious. Firstly, it was huge, and perhaps it was because the size and slowness of it some officers christened this ‘Appareil Boirault’ machine with the tongue-in-cheek moniker of ‘Diplodocus militaris’ – a military dinosaur (the Diplodocus was a large long-necked herbivore living in the late Jurassic period about 150 million years ago). Measuring 8 meters long, it was still narrow enough as a machine (just 3 meters wide) to be transported by rail, but this would have to be done partially disassembled because, at 4 meters high, it would not clear any bridges. The height was also a considerable problem, it would be impossible to hide from enemy observation and very vulnerable to enemy fire. With the engine at the top, it would become crippled quickly, although the lack of conventional track plates would at least render these huge frames safe from damage by concentrated machine-gun fire on them. At just 1.6 km/h (1 mph), the machine would be unable to evade fire, and being easily targeted would have to carry a large weight of armor which would make it even less mobile. Clearly, this was all impractical even though these frames enabled it to cross trenches up to 3-4 meters wide. The greatest fault though was steering. With just a single track and no means of altering power to one side or another or hydraulic adjustment on the machine, the machine would only be able to move in a straight line. Whatever direction it was pointed in when it set off would be its direction, forwards or backward, making it even easier to hit as a target and also rendering it likely to be thrown off course by undulations in the ground.

If that happened, the two-man crew (presumably a driver, and a commander who would have to operate whatever weapon might have been mounted eventually), would have to exit the vehicle and laboriously use jacks to lift one side of the vehicle from the ground to bring it back on course: a suicidal endeavor in the middle of no man’s land.

According to the historian Alain Gougard, these faults were published in an official report on the subject on 17th May, and consequently, on 10th June 1915, this first machine was abandoned. The historian Francois Vauvallier, however, gives the dates of the test not in February 1915, but on 10th April 1915, although it is possible that testing actually spanned that period. He also provides the date of abandonment of the design as 21st June 1915, not the 10th, but either way, by the end of June 1915, it was officially abandoned.

Boirault machine crushing its way through a barbed wire entanglement during 1915 trials. Image: Steven J. Zaloga

A New Hope – A Second Trial

All was not lost for Monsieur Boirault though. This was still the middle of 1915 and the French Army still did not have an effective tank. In context, the British had yet to perfect the all-steel tank track, so it can be understood that this was still very early on in the development of tank warfare. Boirault was stubborn in his insistence as to the viability of his design and he had been correct that the design worked. The resulting efforts by him pressing his case meant that the French Government created a new commission to oversee his work and authorized him to produce an improved version of this machine. Howard states that this improved machine was scheduled for trails to take place in November 1915, but this second phase is not mentioned at all by Vauvallier.

The enormity of the machine is apparent in this image. The chain drives from the engine going upwards to the drive gears can also be seen as can be ballasted weights on the lower part of the vehicle. Source:

This second trial, organized on 4th November 1915 and taking place on the 13th, showed once more that the machine moved. This time it was laden with 9 tonnes of ballast to simulate arms and armor which would be fitted and albeit still very slowly, merrily plowed through a barbed-wire entanglement 8 meters deep and over a two-meter wide trench.

The primary modification to the machine appears to have been in the steering. Previously, a single large external jack was used, but this November trial demonstrated an internal system of smaller jacks operated from inside the machine. It is not clear when these were fitted, but given the criticisms of the machine from the spring trials and the lack of obvious visual changes, the addition of these smaller internal jacks is a logical assumption. Either way, turning was very slow, had to be done from a halt and was limited to a maximum of 45 degrees.

The modifications proved insufficient to overcome the inherent problems of the design and once more the machine was rejected.


With the first machine having proved his theory at least technical sound, if not very practical, Boirault used his company and experience to produce a second vehicle. This new machine was substantially different from the first machine incorporating the same ‘over the top’ single track principle but substantially smaller and with a much more compact track run. Vauvallier states that this machine was already underway in either design or construction by the time of the November 1915 trials of the first machine.

A New Design


The essential principle of traction remained the same. A track run going completely around the machine made from six steel frames with the engine and fighting compartment located enclosed within the track run. Each frame was more compact this time, no longer the huge open rectangular frame of the first machine with four horizontal spars. This time the frames were square and made from four steel girders braced together at each corner by a triangular plate and with two rectangular ‘feet’ projecting from each side providing for additional grip on the ground. On the inside of the frame, running vertically through it, was a single steel girder connecting each of the center couplings together and connected from that was a further brace to each corner of the frame. Those four external girders, together with the triangular corner bracing, provided the ground contact area and also left a large octagonal space on the outside of the frame. Each frame segment was coupled to the preceding and following frame by a heavy steel pin in the connecting corners and also by a flexible coupling in the center, providing a very solid means of connecting each ‘link’ in this giant-sized track to each other. This massive type of track construction rendered the track impervious to damage from machine-gun fire and also providing protection for the cabin within, but also a robust system unlikely to be damaged by any obstacles it might contact on the ground. These massive frame tracks were so substantial in fact that it is hard to imagine the damage to them being caused by anything short of a direct hit by a field gun or shell.

The extremely substantial tracks of the second machine are obvious here as is the heavy riveted construction of the armored body of the tank. Source: French National Archives

An important change to the connections in this track system though was that the connections at the corners and the coupling had a small degree of lateral flexibility. Rather like the railcar coupling Boirault was so familiar with (and an expert in the design of), each frame could twist slightly against the other providing the steering movement for the vehicle. The wholly useless jacking system of the first machine was totally abandoned in favor of this design which not only ensured the crew could stay enclosed in armor, but that steering could also be effected on the move. What is unclear is exactly how this ‘flexing’ of a frame was to be carried, but as the original jacks on the floor of Machine One were hydraulic, it is a logical extraction that these jacks moved to either the rear, front or roof of the machine could be used to push on one side of a frame as it moved over it, in order to effect the movement. This system certainly worked but provided for a marginal turning movement and when this machine was later tested the radius of a turn to be 100 m.


The old machine used a pyramid-shaped cabin with the curved bottom section holding the ‘cab’, although it was open to the elements and with the engine and gearing above them. Essentially, this setup with engine-over-crew was retained on the second machine but totally gone was the tall pyramid. Instead, a new, rhomboidal shaped cabin was built with a pointed bow and stern. Constructed from heavy armor plate riveted to a steel frame, the gearing was still at the top, along with the drive chains, but the engine had been moved lower down into the rear of the machine.

Another view of the incredibly substantial tracks for the tanks and the ‘feet’ sticking out from the sides of each frame (track section). The drive gearing can be seen on top of the machine too. The two men in the photo appear to be Monsieur Boirault (left) and possibly the French Minister for War (right). Source: French National Archives.

This had the advantage of substantially lowering both the profile of the tank but also its center of gravity so it was less prone to toppling sideways. Access to this machine was by means of a large rectangular door on each side which had a curved top opening backward. In each door, a portal was fitted into which presumably a machine gun could be mounted, although no armament was specified.

Rearview of the second machine showing the drive chains descending to the engine and gearbox in the rear. The thickness of the armor is also apparent and substantial in keeping with the rest of the machine. Source: French National Archives via Vauvallier.

One photograph from trials of this machine seems to indicate a weapon mounting on the nose section of the machine which could be interpreted as the sort of weapon like a 75mm cannon required for breaking up an enemy position and later used on the Schneider CA-1, although this is speculation.

Completed second tank from Boirault showing the side door (open) and possibly a cannon mounted in the nose. Here it is negotiating the crossing of a trench at an angle. French National Library Reference #530164880

Trials and Tribulations

The spring trials of 1915 showed that the first machine had significant problems with height, speed and steering. Amendments made by the end of the year led to further trials of it in November with amendments to the steering system. These had still proven inadequate and that machine was abandoned. The principles employed though remained, and the second machine had a new steering system using the same type of ‘frame track’. When this second machine was finished, it was subjected to trials by the Army on 17th August 1916 at Souain-Perthes-les-Hurlus, in North-Eastern France. This was probably also the site of the 1915 trials of the first vehicle.

Here, despite the obvious improvements to the machine and its heavy armor and construction combined with the ability to cross a trench 1.8 meters wide, the low speed (under 2 km/h) meant that it failed to impress the Army. General Henri Gouraud (4th Army) however, was more sanguine on the matter despite the obvious problems. He was very impressed by the ingenuity of the design and the robust nature of the design. Narrower than the first machine it was still able to plow effectively through belts of barbed wire creating an avenue for any infantry to follow which was more than 2 meters wide – something which was in critical demand at the time but which was now also solved by new designs using a new robust pair of tracks.

Seen during testing in August 1916 the second Boirault tank strikes an imposing figure. The man in front is probably Monsieur Boirault himself. Source: French National Library Reference #53016523
A crowd of curious officers and men look at the mechanical marvel in front of them during tests which were presumably meant to be secret. One can only imagine what they are thinking. French National Library Reference #53016497

Postscript to the War

In a report dated 20th August 1916, Gen. Gouraud indicated he would have preferred further trials of an improved machine but these, sadly, were not to pass, and the design was abandoned by the military in favor of new tank designs which were becoming available to them. This was not, however, as thought by many historians, to be the end of the Boirault vehicles or Monsieur Boirault’s interest in this type of traction technology.

After the war was over, he continued his development with the submission of a patent in France (number FR513156) for ‘Appareil roulant pouvant etre employe comme pont automobile’ (Rolling apparatus that can be used as a motor bridge) on 1st April 1919. The patent was granted on 28th October 1920 and published the following February.

Drawings of Boiraults adventurous ideas for multiple machines forming a huge continuous bridge from French Patent FR513156(A) of 1919

The patent was nothing short of optimistic, envisaging hundreds of such frame-machines like his 1915 design with the pyramidical center acting together to form a type of rail bridge would be employed to replace missing bridges, create entirely new bridges, and even off-load full-size steam locomotives from a ship. Each frame-machine could use multiple vehicles within them to literally create a mobile moving bridge although quite how these were to actually function outside of a paper design is unclear but given that his first machine can be accepted as a ‘tank’ in some aspects, then this 1919 patent can also be accepted, by extension, as probably the most unusual tracked bridging vehicle imagined. This design was never committed to construction and it remained entirely on paper as a patent, perhaps simply just to stop someone else taking the ideas on which he had worked so hard through the difficult years of 1915 and 1916. One final note from the design is shown in Figure 8. Figure 8 shows a 12-frame machine carrying what appears to be a very heavy field piece of artillery. This enormous single-engine carrying this gun was representing a potential single engine of war or transportation replacing a series of frames using smaller engines within them. It is hard to quantify whether this additional Boirault design is a tank, a bridge, a transporter or something of all three, but certainly, the vision and ingenuity of Monsieur Boirault seemed boundless with this idea.

‘Figure 8’ from French Patent FR513156(A) of 1919 showing a huge 12-frame vehicle replacing 4 individual vehicles transporting what appears to be a large caliber gun.


The Boirault tanks were undoubtedly the most unusual tanks built and tested during the First World War and came around at a time when the technology relating to AFVs was in its infancy. The problems faced by the first machine were mostly overcome, save for speed and turning, but given the enormous progress between the two designs, it is reasonable to imagine that, given more time, this machine could genuinely have been put into production to create breaches in the enemy wire. It is perhaps sad from an engineering viewpoint that he did not continue his unique style of vehicles as tanks or attempt, post-war, an even more adventurously sized vehicle.

What is clear from his work, however, is that these machines worked. They achieved everything he had expected of them and at the time they were designed were the only machines available to the French capable of crushing the wire in no-man’s land and crossing trenches. They should not be looked upon as a failed design or something ridiculous but as a triumph of the power of engineering skills married to the imagination to come up with an innovative solution to a problem killing tens of thousands of men. Monsieur Boirault was a pioneer who helped to spur the military in France to produce tanks for the war effort. Regardless of the fact that his machines were never used in combat, the path to the design of a workable tank was a difficult one for all nations. A difficult journey in which he played his part, and should, therefore, be recognized as such, a visionary, a patriot, and engineering visionary.

Illustration of the Fortin Automobile ‘la Machine Boirault écrase barbelé’ (Eng: The Barbed Wire Crushing Boirault Machine) produced by Yuvnashva Sharma, funded by our Patreon campaign.

Specifications (Prototype No. 1)

Dimensions (L-W-H) 8 x 4 x 3 meters
Total weight, battle-ready 30 tonnes (+9 tonnes ballast later)
Crew ~2 (Commander, Gunners)
Propulsion 80hp Aster petrol engine from a Filtz agricultural tractor (petrol)
Speed 1.6 km/h

Specifications (Prototype No. 2)

Dimensions (L-W-H) 6-8 x ~3 x ~ 2-3 meters
Total weight, battle-ready 30-40 tonnes
Crew ~4 (Driver, Commander, Gunners x 2)
Speed 2 km/h
Armament est. two machine guns and one cannon
Armor ~25mm


French Patent FR513156(A) Appareil roulant pouvant etre employe comme pont automobile, filed 1st April 1919. Accepted 28th October 1920. Published 9th February 1920
Gougaud, A. (1987). L’aube de la Gloire – Les Autos-Milirailleruses et les Chars Francais pendant la Grande Guerre. Societe Ocebur
Granier, V. (1919). Les etapes successives de l’arme victorieuse: Le tank. La Science et la Vie No.44
Vauvallier, F. (2014). The Encyclopedia of French Tanks and Armoured Fighting Vehicles: 1914-1940. Histoire and Collections.
Zaloga, S. (2010). French Tanks of World War 1. Osprey Publications

WW1 British prototypes

Macfie’s Landship 1916

UK United Kingdom (1916)
Design Only

It has been over a century since the guns of WW1 fell silent. A war best known for mostly static trench warfare on the Western Front in France and Belgium as the great Empires of Britain, France, and German slugged out a 4 year-long brutal slaughter. In the immediate aftermath of this war, one of the things done to return to normality for the British was to assign credit to the designers, inventors, and engineers who invented, designed and built many of the key weapons which had led to ultimate victory over Germany. This, of course, included the question over who invented the ‘tank’ and involved dozens of separate claimants to the title. When it was all done, some names, like Tritton, d’Eyncourt, Churchill, and Swinton became very well known for their part. Several men were to receive scant attention and credit for their part in the process and one of them was Robert Macfie. Robert Macfie is a virtually unknown name even to people with an extensive reading or knowledge of armored warfare, yet he fills in a gap in the evolution of tracked warfare at a time before tanks even existed in anything like the form we know them today. A strong advocate for tracked armored vehicles even before the existence of the Landships Committee, Macfie was to design a series of landships for the British war effort, although post-war these were almost completely forgotten.

The Man

Born on 11th November 1881 in San Francisco, the American-born son of a sugar baron with financial interests in the Caribbean and Hawaii, Macfie took an interest in military matters outside of the family sugar business. Aged just 17 or 18 years old, he enrolled in the Royal Naval Engineering College, at Davenport, England, studying naval design. After this, he went back to help with the family’s sugar business before settling in Chicago in 1902.

Around this time, he began an interest in aviation and was back in Britain by 1909 building his own aircraft and testing it at Fambridge in Essex. It was during his endeavors in the then-brand-new field of aviation that he met Thomas Hetherington, a man later connected with landships in his own right.

His attempts at getting into the aviation business, however, were not a success. He was back on the family sugar plantations in the years before the outbreak of war and it was there that he became acquainted with the Holt agricultural tractor. When war was declared in August 1914, Macfie and his knowledge of tracked vehicles returned to Britain once more. He immediately sought out his contacts from his aviation days advocating for the use of Holt-based tracked vehicles, advocacy which persuaded him to enlist in the Royal Naval Volunteer Reserve (R.N.V.R.) in October 1914.

With his experience, he was brought to the first meeting of the Landships Committee on 22nd February 1915. That though was his only attendance at the committee. After that, he was sidelined to a project working for the Royal Naval Air Service (R.N.A.S.) on a tracked truck. When that project fell apart, in acrimony with the firm building the vehicle, Macfie was essentially a man without a project. His commission was terminated in December 1915 and his military career was over. Just two weeks after the end of his commission, he filed for a patent innocuously titled ‘Improvements in or relating to Motor-vehicles’. At this time he was still giving his occupation as that of a Mechanical Engineer living at 3 Kingsway, London. The purpose of the design was a completely new style of landship, unlike anything before and the culmination of his work and theorizing into the problems associated with tracked armored vehicles to that point. Specifically, the design was described as an “improved vehicle is particularly suitable for use as an armoured car in warfare”


The 1916 landship from Macfie was mounted on a pair of bogies carrying portable track (or tracks) which was held separate to the body. This allowed the bogies to turn independently of the body of the vehicle, and of each other, in order to facilitate steering and with the front end of the vehicle pivoted on the lead bogie. The bogies could also turn together which would allow for sharper turns. The front bogie was mounted on a pivot along a horizontal beam which allowed it to move in the vertical dimension. This additional degree of movement enabled the lead tracks to remain in contact with the ground even when coming into contact with a steep incline, such as a riverbank or parapet.

Macfie does not mention what sort of track system was to be used, but he was an avowed fan of the Holt system. Prior to the war, he had gained experience of Holt tractors and the drawings he provided are suggestive of this system as well. It could be reasonably expected therefore that any tracklayer system he was considering would be similar to or based upon the Holt system.

The Holt track system was produced in various lengths like this short section but following the same principle. It was slow and prone to allowing the track to drop away from the wheels when unsupported. Nonetheless, it was a very popular system prior to the war and the basis for numerous designs both practical and otherwise. Source: Author’s own

This vertical movement of the lead bogey was controlled by means of a large screw jack attached to a rotating mounting within the front part of the vehicle. Turns of the nut on this screw jack, mounted on the floor of the compartment, pushed more of the vertical threaded screw downwards which pushed down the rear of the front bogey. With the pivot in front of this position, this action served to raise the front of the tracks. Elevation range appears to be in the region of 0 to +45 degrees.

What sounded good in theory was not necessarily practical or possible with the technology and materials of the time though. To add extra complexity, Macfie (a tried Naval engineer) also suggested making the lower part of the body of the vehicle watertight so it would float. Able to traverse open water or rivers, Macfie envisaged the use of a propeller mounted at the back powered from the engine. This was an improvement on his 1915 design, as this propeller was able to be stowed vertically, whereas it was fixed in the ‘down’ position in the 1915 design.

The vehicle was to be driven by “an internal combustion engine or other motor driving wheels around which the portable track passes”

Macfie’s design of January 1916 from the British Patent.


The overall shape of the machine was that of a large box with a pitched roof and pointed front. The sides were flat and vertical topped with an angled roof where the body was over the tracks of the rear bogie. At the front, the roofline dipped down to join the pointed front of the vehicle, forming a wedge shape. In front of the leading edge of the body of the vehicle was a large vertical wire cutter made from overlapping triangular blades. This device was very similar in appearance to that trialed on the Italian Pavesi Autocarro Tagliafili (Pavesi Wire Cutting Machine) and the French Breton-Pretot Wire Cutting Machine, although there is nothing known to suggest a link between any of those designs and this one.

Ahead of the wire cutter, attached to the lead bogey, was an armored box with a pointed face that was adjustable.


When stationary, the landship served the role of a fortification. Within the body of the landship was a large set of shields that could be lowered to cover the rear bogey, and that pointed box at the front of the lead bogey could likewise be lowered. This lowering method for the body and nose was intended to protect the tracks from enemy fire when the vehicle was stationary.


No specific armament was mentioned in Macfie’s 1916 patent application other than allowing for the use of “fire-arms or guns from the interior of the vehicle”. Based on his 1915 design and the prevailing thought of the time, armament was likely limited to a pair of machine guns on each side and a series of loopholes along the sides for troops within to fire from.


No specific crew is mentioned, but a review of the design shows the need for a single driver at the front and at least another in the rear to control the rear bogey. A third man may have been required to operate the wheels controlling the elevation of the lead bogey meaning at least 2 or 3 men would be required to steer the vehicle. With a commander and at least one gunner for each machine gun, that would make a minimum crew of around 7 men with space in the back for maybe 20 or so men.


Macfie’s January 1916 design was amended in April 1916, improving the design in several areas. Firstly, the amendment clarified details of the elevation mechanism for the front bogey making it clear that it could be provided with a locking system to hold it in an elevated position even during steering. The wire cutters which had been drawn in January were explained as well. Macfie was in doubt as to the importance of the machine to be able to cut its way through “barbed wire networks”. Any wire encountered by this vehicle would be guided by the shape of the nose which was drawn as a box with a pointed front but described as either ‘pyramidical’ (square-based pyramid pointing forwards) or conical. Powered by the engine, the blade would then slice through the wire, a significantly better concept for cutting wire than the somewhat feeble attempts considered by the official landships work as demonstrated the summer before-hand at the R.N.A.S. Depot at Barlby Road, which relied on the engine power pushing the vehicle through the wire.


Robert Macfie had failed in his attempts to get the Landships Committee to adopt his original design. He had, however, achieved considerable success in the one Landships Committee meeting he was at. He had convinced them of the benefits of a tracked vehicle, albeit basing his ideas on the Holt chassis at the time. This more adventurous design from 1916 though was not a success. It was significantly more advanced in both what was being envisaged and in technical terms than the first vehicle but, without a direct line of contact to the Committee, this design went nowhere. It is not clear if Macfie even tried to submit this idea to the relevant authorities or not, but it would not have mattered anyway. By the time he submitted his design in January 1916, the famous quasi-rhomboid-shaped British landship had already been settled upon.

Despite the lack of success with this design, Macfie would try once more to design what he felt was a tracked vehicle, but neither of these designs featured in the post-war enquiry into the invention of the tank. Macfie died in New York, USA in 1948.

Illustration of Macfie’s 1916 design, produced by Mr. R. Cargill, funded by our Patreon campaign


Crew est. 7 men (driver, steersman, commander, machine gunners x 4) + 20 soldiers
Armament Machine guns
Armor Bulletproof


Hills, A. (2019). Robert Macfie, Pioneers of Armour Vol.1. FWD Publishing, USA. (Available on Amazon)
British Patent GB124450 ‘Improvements in or relating to Motor-Vehicles’. Filed 3rd January 1916, Accepted 3rd April 1919
US Patent US1298366 ‘Motor Vehicle’. Filed 4th September 1917, Accepted 25th March 1919
Proceedings of the Royal Commission on Awards to Inventors: tank 1918-1920

Robert Macfie (Pioneers of Armour)
Robert Macfie (Pioneers of Armour)

By Andrew Hills

The foundations and principles of modern armoured warfare did not appear out of a vacuum, and nor did the machines of WW1 and WW2. Their development was full of false starts, failed ideas, and missed opportunities. Robert Macfie was a pioneer in aviation at the turn of the century followed by work with the Landships Committee on tracked vehicles to break the stalemate of trench warfare. Although his tank designs never saw combat the work he started was carried on by other pioneers and helped to usher in a dawn of armoured and mechanised warfare.

Buy this book on Amazon!

WW1 American prototypes

William H. Norfolk’s War Weapons

USA ww1 USA (1915-16)
None Built

When the United States entered World War 1 (1914-1919) on 2nd April 1917, it did so without any tanks or conventional armored vehicles outside of a few armored cars and trucks. Artillery was either horse-drawn or towed by unarmored lorries and infantry assaults would have to take place without armor protection. Whereas America’s allies, Great Britain, France, and Italy, had all quickly realized the butcher’s bill which followed unprotected infantry attacks meant a need for some armored vehicle, the US entered the war with none of that experience. That is not to say that there were no designs and suggestions in existence for such weapons though. One designer who submitted a variety of war weapons was William Norfolk of San Pedro, California.

Norfolk’s powered ‘dirigible’ float for deploying a net (left) and deployed from a building onshore (right). Source: US Patent US1181339

The Mine and Submarine Destroyer

In light of the raging conflict on mainland Europe, William Norfolk submitted a design for what was effectively a type of net. It was designed to counter enemy naval vessels and torpedoes. Filed on 25th August 1915, Norfolk submitted his idea for a cable-net deployed by means of a powered float driven electrically. This float could be steered from shore or even a ship and would tow out behind it a long cable-net designed to ensnare an enemy ship or torpedo. The net would be prevented from sinking by virtue of a series of buoyant floats and could even be fitted with magnets to make sure the net would attach to an enemy ship or torpedo. He must have been confident as to the utility of such a device, as he filed a patent for it in Canada on 6th November 1916 as well. However, what may have seemed like an innovative idea resulted in no further development.

Trench Artillery

The Mine and Submarine Destroyer net patent was granted to Norfolk on 2 May 1916 (US Patent). Sometime between then and September 1916, Norfolk turned his attention towards the war on land. Characterized by lines of trenches covered with belts of barbed wire and covered by machine-gun fire, no-man’s land was deadly for exposed men. Whilst the amount of information coming back from the Western Front was heavily censored in the media (primarily newspapers and newsreels), there was no concealing the scale of the losses and the primary reasons for them. The British had started their formal Landships program in February 1915, but this was still secret, including the development of the first characteristic quasi-rhomboid shaped ‘tanks’ at the end of that year. This secrecy continued through to September 1916 with the first tank deployment on the Western Front, but even then it was some time before a clear idea of what these machines really looked like became public knowledge.

Knowing this, it can be said with some certainty that Norfolk’s concept for breaking this stalemate and the static war was not inspired by the development of the British or anyone else. What he produced was, in fact, very similar to a plan by the British in 1940 for a trench digging assault machine. That machine, known under the codename of Cultivator Number 6, was very similar to Norfolk’s and perhaps indicates that Norfolk’s idea was perhaps not quite as ‘off-the-wall’ as it may have appeared at first glance.


Norfolk’s machine, like the future Cultivator Number 6 a quarter-century later, was a subterranean assault machine. It did not go underground but used the ground as its armor. The means of advance was simple in concept, mounted on wheels with traction from the front pair, the machine was driven by an engine and was faced with a full width cutting face consisting of what could be described as a very wide track with cutting teeth. Driven by a separate motor, this ‘cutting-track’ ran from the bottom upwards, progressively digging away the face of the soil and throwing it into a hopper (identified as point 46 on Canadian Patent CA174919) and from there onto an outwardly facing conveyor belt which threw the soil off to one side. In this manner, the machine not only dug a wide trench as it headed towards the enemy, but also created a berm along one side of the trench which would further conceal the vehicle from enemy fire. It is important to note that the height of the machine above the ground could be varied by adjusting the pitch of the cutting face so it could self-dig down up to a maximum depth of being level with the ground. No dimensions are given for this digging machine but based on an estimate of the wheel (item 74) as 1.5 to 2 m in diameter it would have an estimated height of around 3 m or so for the whole machine – certainly a very deep trench although it could, if needed, operate with a portion above ground in order to make use of its machine guns.

Norfolk’s Trench Artillery machine of 1916. Source: Canadian Patent CA174919


While the general layout may seem straightforward, the rest of the design, including the armament, was anything but straightforward or conventional.

Firstly, the primary armament was a ‘disappearing gun’ mounted on a central turntable on a triangular mounting. This unspecified caliber of gun was to be loaded under the cover offered by the machine and would then rise up and fire, destroying enemy strongpoints. The armored casemate was also meant to carry a series of machine guns mounted through circular loopholes along each side, although the type and number were not mentioned. Importantly, it should also be noted that the casemate had no protective roof – a significant flaw for a weapon below ground level and exposed to shrapnel and debris from above.

The final weapon system, for lack of a better description, consisted of a pair of catapults. Along the sides of the casemate, at the level of the bottom of the frame, were two ‘arms’ connected to a driven gear. Each arm was held down in the horizontal position during movement but, when required to be used, could be driven upwards-acting around the driven gear, propelling what appears on the patent diagram to be a large disc. Each ‘disc’ is described as an ‘Enfilading Machine’ and these machines were subject to a later patent application by Norfolk.

Each ‘Trench Artillery’ machine carried a pair of catapults with a single Enfilading Machine at the end of each one. When the machine closed on the enemy lines, it could activate these catapult arms either together or independently and these would quickly lift the Enfilading Machines up to the surface and onto the ground in front of the machine.

Norfolk’s Trench Artillery machine of 1916 from above showing the position of the main gun in its ‘disappeared’ position. The two large objects in the back corners are not wheels but the tops of the Enfilading Machines carried on the Trench Artillery machine. Source: Canadian Patent CA174919

The Enfilading Machines

This complicated facet of the design was so involved that Norfolk submitted a completely separate patent for the Enfilading Machine in its own right. The date for that patent application is February 1916, whilst the Trench Artillery Machine is September 1916 (Canada) and no trace of a filing in the USA. The Enfilading Machines, therefore, predate the Trench Artillery machine, which served as much as a launching platform for the enfilading machines as an armored war-machine in its own right.

Just as the Trench Artillery machine predated the Cultivator machine of World War 2, this Enfilading Machine predated another WW2 project known as the Great Panjandrum. Just like the Panjandrum, the Enfilading Machine was based on the principle of an unmanned wheel rolling towards the enemy. The Enfilading Machine though, was significantly more complex than the Panjandrum, which was little more than a barrel full of explosives on two rocket-propelled wheels. Norfolk’s idea was an entire weapon system in itself, consisting of a pair of traction wheels spaced slightly apart but on a common axle. Mounted between these two wheels was a frame to which was attached a trailing wheel for balance (fitted with ‘spurs’ for traction), but also an electric motor to drive the machine forwards, delivering power to the axle and wheels respectively. Around the periphery of each wheel was a pair of concentric circles, each made from 64 recessed tubular chambers. These 128 chambers were actually short barrels for what was a single shot charge firing a single cylindrical shell or bullet perpendicular to the direction of travel of the wheel. Across both sides of the machine, this meant 256 shots to be fired out to the sides. Ignition was electrical and triggered by means of a timer.

The outline of the Enfilading Machine (facing left to right) showing the pair of concentric shot-chambers around the periphery of the wheel. Source: US Patent 1227487 of 1916

Other weaponry for the Enfilading Machine was in the form of spherical exploding balls (shells) which were mounted into recess cavities in the outer face of each wheel, with 24 on each side for a total of 48. Each shell was detonated by a rather crude burning fuze ignited when it was launched by means of explosives. This was supposed to project the shell out to the sides, although the patent drawing shows them being launched in, at least, pairs at a time on each side. Just like the shot-chambers, to launch the spherical bombs chambers these were triggered electrically by means of a timer. The use of the timer suggests that it could be ‘programmed’ to travel a set distance before detonating some or all of its weaponry to the sides.

Cross-section of the Enfilading Machine showing the rectangular launching chambers for the pairs of spherical bombs to be launched from each side. Source: US Patent 1227487 of 1916


The Enfilading Machine was an interesting design in its own right and predates the Grand Panjandrum by a quarter of a century. The Grand Panjandrum proved impossible to control and was significantly wider and simpler than this Enfilading Machine, which was a serious flaw in its design. The concept was clearly not fundamentally bad, launching a remote demolition or assault weapon was, and still is, a viable tactic but the execution of the idea was completely unworkable. The machine was far too complex with too many weapons and working parts and mechanisms for a disposable weapon. It was far too narrow to avoid simply flopping over on its side on anything other than a perfectly flat surface and the single, heavy bearing surface from the two wheels would simply be hopeless in anything other than hard ground, as it would otherwise just sink and become stuck. The final criticism of the Enfilading Machine is the armament, which was too much and too weak. Considering the use of trenches rather than exposed troops, anything other than a direct landing of the wheel into a trench would produce nothing more than a lot of bullets fired into thin air and bombs landing harmlessly outside of a trench. The single, large high-explosive charge of the Panjandrum was simply a far better idea and a more effective weapon. One final note in favor of the Enfilading Machine though might be the trailing wheel. Acting as a counterbalance to help keep it on track, it has to be considered whether such an addition to the Grand Panjandrum might have helped rectify its flaw where it would lurch off to one side, becoming a potential hazard for the forces launching it.

Norfolk’s Enfilading Machine attacking a pair of flimsy barbed wire fences. In reality, barbed wire often formed very large and dense entanglements far removed from what appears to be a fence more suitable for restraining cattle than for warfare as depicted here. Source: US Patent 1227487 of 1916

For the Trench Artillery machine, a conclusion is equally nuanced. The concept of a giant digger approaching below the ground surface towards the enemy was clearly viable. The Cultivator No.6 proved this was possible, but where the Cultivator was tracked, the Trench Artillery was wheeled and used small wheels at that. Just like the Enfilading Machine, it would have become hopelessly stuck in anything other than very hard ground and the vulnerability of such a machine to shrapnel shells exploding above it is also patently obvious too. Once more, the idea was not completely unworkable but the solution offered was.

Neither the Enfilading Machine nor the Trench Artillery machine should be ignored or written off as a crazy idea though. Both have some merit and, in 1915-1916, they provide an interesting insight as to one of the possible solutions being considered to the problems of trench warfare. In some ways, the ideas are less crazy than some official projects which were attempted by the French or British and really present a picture of how the war was being viewed outside of the front where technical solutions to problems of machine guns and wire were being presented. Neither machine was ever built and far more sensible and better-considered ideas did prevail. However, a failure to consider even some of these flawed ideas does a disservice to men like Norfolk, his ideas, and to properly appreciate how difficult it really was to develop tanks as they first appeared on the battlefields of France in 1917.

Post-Script for Norfolk

William Norfolk had no luck with his military designs but he did submit one further patent, albeit not a military-related one. In 1930, he submitted an idea for a crack-filling and sealing device. It is not known what became of Norfolk, but perhaps with his crack-sealing invention, he found some success with his innovations.

Illustration of William Norfolk’s Trench Artillery Machine based on the design of 1916 produced by Yuvnashva Sharma, funded by our Patreon campaign.


US Patent US1181339 Mine and Submarine Destroyer, filed 25th August 1915, granted 2nd May 1916
Canadian Patent CA174919 Trench Artillery, filed 21st September 1916, granted 6th February 1917
Canadian Patent CA176438 Mine and Submarine Destroyer, filed 6th November 1916, granted 17th April 1917
US Patent US1227487 Enfilading Machine, filed 23rd February 1916, granted 22nd May 1917

Cold War Italian Fake Tanks Fake Tanks

Progetto M35 Mod. 46 (Fake Tanks)

Italy (1946)
Medium Tank – Fake

The popular online game World of Tanks (WoT) published and developed by Wargaming (WG) has many tens of thousands of players and a wide variety of historical and semi-historical armored vehicles to play. It also has a few ‘fake’ tanks as well, that is, tanks that never existed in either drawings or material. The Progetto M35 mod.46 Medium tank is one from the latter category. The tank is very handsomely represented with a 3D model, but it is a fake, as the tank never existed. However, the vehicle in-game is not wholly fabricated, as it has a minuscule basis in fact.

WoT Representation

In WoT, the Progetto M35 mod.46 is, as might be expected from its name, represented as a project dating from 1946 for a 35 tonne (hence the ‘M35’) medium tank. There is even a short ‘history’ provided:

“Conceptualization of a draft design developed at the request of General Francesco Rossi who believed that only light vehicles weighing up to 35 tons would be effective in a new war. Such an innovative design was not approved; development was discontinued when Italy joined the Standard Tank project.”

WoT Wiki extract.

This ‘history’ is a half-truth at best.

Progetto M35 mod.46 as represented by Wargaming in its World of Tanks game. Source: Wargaming

In-Game, the Design is as follows


The engine for the Progetto M35 mod.46 in the WoT game is given as a 652 hp ID36S 6V CA engine. Although the manufacturer’s name is not provided, the Italian firm of Isotta Fraschini did make a series of engines known as ID-36. These were 9.72 litre marine diesel engines with 6 cylinders arranged in a ‘V’ shape (hence the 6V in the name for a V6 engine) and producing 500 hp. Measuring just 92.5 cm high, 92 cm wide, and 137.2 cm long, this engine weighs just 890 kg. In WoT, the engine module weight is given as 1,200 kg, more than the actual engine. With an output of 652 hp, the engine in-game it is also much more powerful than the real engine, although static-engine versions of the ID-36 are available which produce in excess of 700 hp, like the Fire-pump version (725 hp)

Isotta Fraschini ID-36 series marine diesel engine. The particular model produced 500 hp. Source: mfc-ve on

The engines have only been around since the early 1980s, although the company itself dates back to the early years of the 20th century. Whilst the engine is neither for tanks and was not available in 1946, the engine is essentially genuine. They are still in use today for motorboats for example, as they are valued for their compact size and reliability. Their most notable use is in the Italian Lerici-class minesweeper ships of the Italian Navy. Other versions of this engine with 8 and even up to 16 cylinders are available producing up to 2200 bhp. The ‘CA’ added to the end of the WoT module in-game is simply to denote Carro Armato (tank use), although as already stated this engine was never used for tanks.

Suspension and Tracks

The suspension for the WoT Progetto M35 mod.46 is given in the game as ‘Progetto M35 mod.46’ suspension, although what sort of suspension this is open to question. With six evenly spaced road wheels on each side and a noticeable offset between the wheels on the left and right, it appears to be suggesting the adoption of torsion bar suspension for the tank. No such mention of this type of suspension or any other type of suspension is mentioned by General Rossi so this choice is entirely fictional/speculative on the part of WoT.

Secondly, the choice of tracks for the model is very odd too, as, with three rectangular rubber pads across each link, the tracks bear an uncanny resemblance to the British ‘hush puppy’ type of tracks as used on the British Centurion tank. There is no evidence that Italy ever operated a Centurion tank or the ‘hush puppy’ tracks for it either. Further, those types of tracks were not introduced on the Centurion until the 1960s in an effort to reduce the damage to paved roads. Therefore, even if Italy ever did get some of these tracks for some purpose, they would clearly be unsuitable to model on a tank from 1946.

Front view of the tracks on the Progetto (left) bear a striking resemblance to the British ‘hush-puppy’ (right) tracks. Source: WG WoT & Mark Nash


In WoT, the Progetto M35 mod.46 is shown using what is described as a 90/50 T119E1 main gun. This is a 50-caliber long 90 mm gun with a cylindrical muzzle brake/blast diffuser. The gun is a very interesting choice, as the history of the T119 gun makes it clear that it is entirely inappropriate as an option for this design.

For a start, the gun is American, not Italian. The T119 gun originated from the development of the US T42 Medium Tank which did not even reach the wooden mockup stage until March 1949. When it did, it was fitted with the M3A1 90 mm gun, but this was considered substandard and had to be improved with revised specifications for an improved pressure breech capable of withstanding 47,000 psi (324 MPa) instead of 38,000 psi (262 MPa). It was this revised 90 mm gun which became the T119.

T119 90 mm tank gun fitted with a single baffle muzzle brake. Source: Hunnicutt’s Patton

This T119 gun was able to fire the 90 mm ammunition of the M3A1 90 mm gun, but not the other way around, as it was a higher pressure (the cases were even modified to prevent an accident loading on the lower pressure gun with the higher pressure rounds).

The T119 gun fired the T33E7 AP-T shell (mounted in the T24 case) at 3,000 ft/s (914 m/s) as well as the M71 HE round (in the T24 case). At 177.15 inches (4,500 mm) in length, the T119 had a length of 50 calibers.

Manufacture of the T119 90 mm gun was not even authorized by the Ordnance Technical Committee for production at Watervliet Arsenal until 20th October 1948. This gun was still considered ‘new’ and experimental (hence the ‘T’ designation) in January 1950, when it was modified into the T119E1 and finally became the T125 gun (later standardized as M36) as part of the development of the M56 Scorpion (then the ‘Carriage, Motor, 90 mm Gun, T101). This T119 gun was originally fitted with a single baffle muzzle brake, but this was later replaced with a cylindrical blast deflector by the time it was mounted on the T42 Medium Tank. The gun on the Progetto M35 mod.46 is certainly a real gun, but it is neither an Italian gun nor in existence at the time of the vehicle. This is before even taking into account considerations of when a brand new and experimental American gun could even have got to Italy and certainly not an autoloader for that gun for Italy.

Other points of consideration for the Progetto M35 mod.46 include the armor. Data given by WoT states that the hull armor is supposed to be 60 mm thick frontally with 30 mm on the sides and rear for the hull, and 80 mm, 60 mm, and 25 mm on the turret front, sides and rear respectively. These figures are not based on any design but are purely a function of balance for the game.

Francesco Rossi

Having dissected the tank as claimed by WG, it is important to consider the man, General Franceso Rossi, claimed as the source and what he really wrote. General Rossi is certainly a real person. Born on 6th December 1885, Rossi was a professional soldier who was a Lieutenant Colonel by 1926. Through the 1930s, he rose through the senior ranks with appoints in Rome as Chief of Military Transport and then as the Commanding officer of various artillery regiments. By 1939, he was the Commanding Officer of an Artillery Corps and then Intendant of the Italian 1st Army. Through World War Two, he continued his rise going from the Commander of II Corps to Deputy Chief of the Army General Staff in March 1941. He was made a Lieutenant General in October 1942 and, in March 1943, Deputy Chief of Staff to the Chief of the Supreme General Staff of the Italian Royal Army (Regio Esercito – RE). It was in this capacity that Gen. Rossi was to play a crucial role during the Italian Armistice of September 1943 (the Armistice of Cassibile).

General Franceso Rossi. Source:

The Source of the Claim

The origin for the WoT claim is from a book written by Gen. Rossi, published in 1946, entitled “La Ricostruzione dell’ Esercito” – the reconstruction of the army. Written in 1946, this paper predates the Paris Peace Treaty of February 1947 and was outlining how a new Italian Army should be organized and the sort of equipment it needed. WW2 had been utterly devastating for Italy with a large but ill-prepared and usually rather poorly led army suffering severe defeats at the hands of the British and Americans. Germany, Italy’s ally in WW2, was not a particularly gracious ally at times either and, following the armistice with the Allies in September 1943, Italy basically collapsed into a civil war with some of the military remaining loyal to the Axis and the rest joining the Allies. This second half had suffered harsh reprisals from the Germans, who from then on had acted as an occupying power. Such a split in Italy required a lot of mending after the war. In this sense, Gen. Rossi’s short book was very well-timed. The Army was totally broken by the war and was still operating a few Italian vehicles left over from the war along with a mishmash of tanks and armored cars provided by the British and Americans. A total reorganization was certainly required. It is worth bearing in mind though, that Article 54 of the Paris Treaty of February 1947 strictly limited the Italian military to not more than 200 heavy and medium tanks and, through Article 61, a total of 250,000 personnel (Army and Carabinieri combined). The likelihood of authorizing the expensive and time-consuming development of a home-grown tank by Italy was simply neither likely nor realistic. It is hard to imagine that Gen. Rossi, from his senior position in the Italian military, would not have been aware of the parlous state of the Italian economy and military post-war.

Article 54 of the Paris Peace Treaty 1947 limiting Italy’s tank strength. Source: US Dept. of State

In his book, “La Ricostruzione dell’Esercito” Gen. Rossi wrote:

Italian Original:

“Accenno anche alle caratteristiche che dovrebbe avere un carro armato di produzione nazionale, unicamente per completare la visione dei mezzi meccanici, per il caso sia giudicato possibile ed opportuno, come io ritengo, procedere a studi ed anche all’approntamento del prototipo.”
“Carro armato veloce, ben corazzato, non mastodontico, perchè resti nei limiti consentiti dalle nostre ferrovie e dalle nostre opere d’arte, ma tale da tener testa ai più progrediti carri esteri: peso dalle 30 alle 35 tonn., cannone di calibro intorno ai 75 mm, motore di 5-600 H.P. di tipo appositamente ad iniezione per la minor facilità di incendio del gasolio rispetto alla benzina.
Dal carro armato potrà trarsi il cannone semovente, utilizzando lo stesso scafo per un cannone da 90, od un obice di calibro maggiore”

– La Ricostruzione dell’Esercito, 1946

English translation:

“I mention the characteristics a national production tank should have solely to complete the vision of the mechanic vehicles, if it is considered viable and appropriate, as I think, proceed to studies and the preparation of a prototype.
Fast tank, well armored, not too big and heavy [like an elephant], provided it stays within the limits allowed by our railway and artwork [bridges, tunnels, etc.], but able to stand up to the most advanced tank of foreign countries: weight between 30 to 35 tons, cannon of a calibre around 75 mm, 500/600 HP engine specifically of injection type due to lower risk of fire compared to a gasoline engine.
From the tank, a self-propelled gun might be derived using the same hull for a 90 mm cannon or a howitzer of a larger caliber”

General Rossi’s book and dust jacket cover binder, 1946. Source: Author

Despite the obviously weakened state of the Italian economy in 1946, Gen. Rossi was still hoping, perhaps vainly, for a new nationally produced tank at least to the level of the production of a prototype. To this end, he outlined the features it should have.

Firstly, powered by a fuel-injected diesel engine (due to the lower fire risk than a petrol engine) producing between 500 and 600 hp. The vehicle had to be quick, able to keep up with the most advanced foreign tanks. At the time of writing, the primary foreign tanks Rossi was likely familiar with would be the American Sherman, British Cromwell, Russian T-34-85, or even the German WW2 Panther with top speeds under ideal conditions of about 48 km/h, 64 km/h, 38 km/h, and 55 km/h respectively. Quite how much Gen. Rossi might have known about the most modern tanks from Britain, American, and Russian though is questionable, but he would certainly have been familiar with at least these WW2 tanks.

Weight-wise, Rossi was very clear, a tank of between 30 and 35 tonnes in weight and of sufficiently modest dimensions to be transported by rail. Armor-wise, the tank was supposed to be well armored yet not too large, hardly a thorough description but then that is because this was not a design – it was a concept of what tank Italy needed for a new army.

At 35 tonnes, this would still be heavier than the heaviest tank Italy produced during the war, the 26-tonne P.26/40 and around 10-tonnes lighter than the German Panther. The weight range given actually closely matches that of the American M4 Sherman. This is not the only similarity either. The gun called for by Gen. Rossi was one of a caliber of 75 mm or thereabouts. The British Cromwell was using the QF 75 mm gun, the American M4 used the M3 75 mm gun or the 76 mm M1A1 series. The British Comet had the 77 mm HV, whilst the German Panther had used the 75 mm KwK 42. Which, if any of these, Gen. Rossi might had been considering is unknown – perhaps he was considering an Italian gun in that caliber range, but he was clear on what he considered a suitable caliber – 75 mm or thereabouts. Bigger guns, like a 90 mm piece, were destined to be on a tank destroyer preferably based on the same chassis.

One of the more unusual elements of General Rossi’s book completely unrelated to its content is that the pages were actually printed and folded before being bound meaning that the majority of pages were still bound along at least one outer edge. Source: Author

That then, is literally ‘it’. There is no design, no model or plans and not a lot of specifics. This was 1946 too, so options were very limited for Italy. Gen. Rossi may have wished for a new tank to be produced in Italy- it would, afterall, be very good for Italian industrial rebuilding as well as for an independent army, but in 1946 this was wishful thinking. WoT’s “such an innovative design…” claim is simply false. There is no design and none of the features he mentioned were in any way innovative.

There was also no need at all for a new and expensive tank for Italy, especially a tank which, after all, would offer nothing that existing available and cheaper designs did not already offer. By the end of the 1940’s, the Italian Army had tanks and tank destroyers which matched what Gen. Rossi had been calling for in the form of Sherman tanks of various types armed with 75 mm, 76 mm and 105 mm guns, Sherman Fireflys armed with the British 17 pounder gun, and the American-supplied M36 Jacksons as tank destroyers armed with a 90 mm gun – a tank destroyer based on the chassis of a Sherman tank, just as Gen. Rossi had wanted back in 1946.


The Progetto M35 mod.46 is a fake. Not a completely made-up-from-nothing fake, but without doubt still a fake. The call from Gen. Rossi for a new tank made it clear that the 90 mm gun was not for this tank, but for a different vehicle. Not only that, but the 90 mm gun selected by WoT was simply not possible to be fitted to a tank in 1946, let alone one in Italy. The tracks, assuming they are ‘hush puppy’ tracks are neither Italian nor available in 1946. The engine certainly is a real thing, but it was not used in tanks and was not around in 1946. All this predated the attempts to develop a single tank as a ‘standard panzer’, sometimes known as the ‘Europanzer’ project.

Whatever Gen. Rossi might have been considering as a tank is unclear, but certainly what he wrote cannot be described as a design. The vehicle, as represented in the WoT game is simply not possible and purely invented.

Illustration of the Progetto M35 Mod. 46, produced by Ardhya Anargha, funded by our Patreon campaign.


Agarossi, E. (2000). A Nation Collapses: The Italian Surrender of September 1943. Cambridge University Press, UK
Data Sheet ‘Motore termico/ciclo Diesel/a quattro tempi/6 cilindri a V a 90: Isotta Fraschini Motori
Dunstan, S. (1980) Centurion. Ian Allen, England
Estes, K. (2016). M50 Ontos and M56 Scorpion 1956-1970. Osprey Publishing, England
Hunnicutt, R. (1971). Pershing: A History of the Medium Tank T20 Series. Feist Publications, California, USA
Hunnicutt, R. (1984). Patton: A History of the American Medium Tank . Presidio Press, California, USA
Isotta Fraschini. (1985). Industrial Diesel Power for Military Applications by Isotta Fraschini (advert)
Pettibone, C. (2010). The Organization and Order of Battles of Militaries in World War II, Volume VI – Italy and France. Trafford Publishing, USA
Rossi, F. (1946). La Ricostruzione dell’Esercito. Editrice Faro. Rome, Italy.
Symth, H. (1948). The Armistice of Cassibile. Military Review, 28(7). Command and General Staff College, Kansas, USA
US Bureau of Naval Personnel. (1990). Manual of Navy Enlisted Manpower and Personnel Classifications and Occupational Standards. US Dept, of the Navy
US Dept. of State. (1947). Treaties of Peace with Italy, Bulgaria, Hungary, Roumania and Finland. US Dept. of State, Washington D.C., USA
World of Tanks Wiki
Biography of Lt. General Rossi