WW1 US Prototypes

Automatic Land Cruiser – ‘Alligator’

ww1 British tanks American ww1 armor USA / British Empire 1915
Tank – None built

The USA was a latecomer to WW1. By the time they started sending men and machines to Europe to fight the Central Powers (Germany and Austria-Hungary), it was June 1917. By that time, millions of men had already been killed and the war on the Western Front had become a war of attrition in trenches in a shell-blasted landscape.

Prior to this date, however, parts of America had not been idle. Indeed, the first British work on tanks had used the American Bullock Creeping Grip track system, which formed the basis first of Colonel Crompton’s work and was eventually fitted to the vehicle commonly known as Little Willie – the world’s first tank.

What is less known is that the Bullock system was also planned for use by another retired British officer – this time in America, albeit at a time when the British were dropping the Bullock tracks in favor of their own system developed by Sir William Tritton and William Foster and Co. Ltd.

The British man concerned here is Alexander McNab and he was based in the heart of America’s arsenal – Hartford, Connecticut. A ship engineer by profession, he proposed a well shaped and well armed ‘tank’ which became known as the ‘Alligator’ – the most viable tank design to come from America in the whole war.


The first use of tanks in WW1 was by the British at the Battle of Flers-Courcelette on 15th September 1916 as part of the Battle of the Somme, and there was a quick reaction to the employment of this new mechanical weapon of war in the press around the world. Various newspapers, magazines, and artists, whether officially or even humorously, tried to envisage what these machines looked like based only on written reports, leading to some rather outlandish ideas of what a ‘tank’ looked like. However, it was not until November that year, when the first official photographs were passed by the censor and published in newspapers, that the public finally got to see these machines.

In this dark period between knowledge of their use and the first photos lay, amongst others, a serious article in Scientific American published on 7th October 1916. Serious because, unlike the majority of newspaper speculation which seemed (especially in America) to claim that the Holt tractor was the basis of the British tanks (it was not), Scientific American instead considered them to be based on the Bullock system. They were not based on that either, but they could not have known this at the time and given that the first tank, known as Little Willie’ or the Lincoln No.1 Machine, was indeed fitted with Bullock Creeping Grip tracks when it was first made, meant that this is a very forgivable error.

A Mk.I Female tank is seen in multi-colored camouflage and using a pair of wheels at the back. Of note are the wire nets over the top, intended to stop grenades from landing on the roof, and the leather helmet worn by the man standing out of the top. Although this was the first official photo of a tank, the first images of tanks did not reach the public until November that year.
Source: Imperial War Museum Q2488

Scientific American, in their article, presented what was to be a common image of the Alligator tank which they described as “a military tractor for use against the trenches”. They claimed that the vehicle had been designed as a response from the British to an unnamed “Western firm” and named the vehicle as an ‘Armadillo’. Given the rounded top of the Alligator and the lines of bolts holding it together, the name Armadillo is, despite it being the product of the magazine, perhaps a better name than the name provided by the designers.

When the article is referring to a “Western Firm”, it is unclear if it is referring to the Bullock Company’s work for the British in 1915 supplying lengthened versions of their Creeping Grip or something else. Certainly, it is possible that the Bullock work was being referenced, although it is notable that the company was actually based in Chicago, Illinois, in the east of the country.

Scientific American went on to state that the design for this vehicle was submitted to the British Naval Munitions Board in London some months prior to the actions by tanks that September. The armored tractor shown in Scientific American, named as ‘Armadillo’ in the artist’s rendering, was not the product of some artist’s febrile or absinthe-induced imagination like so many others, but one based on these Bullock Creeping Grip plans.

The Alligator, as imagined by the artists at Scientific American and based on the published drawings. Here it is shown patriotically and dutifully crushing ‘the Hun’ in their trenches. Of note is that the tank is shown with Maxim-type machine guns in the front in error. The lengthened Bullock Creeping Grip tracks are evident.
Source: Scientific American.
Plan and side view of the Alligator, as published in Scientific American. Note the high position of the driver in the center.
Source: Scientific American.

The Men

In understanding the origins of the Alligator, those behind it need to be considered. There are, in fact, two men involved in the story of the Alligator. The first and most important was Alexander McNab. The second was an American called Norman Leeds.

McNab was originally from Scotland and according to him, had served 12 years in the Royal Navy, finishing with the rank of Lt. Commander. As early as July 1913, he was demonstrating his skill as a marine engineer, with a patent application for an automatic circulator for a steam boiler, followed by another patent related to steam boilers in 1914, and a third in 1915.

From those patents and US census data, followed by his military census record of 1917 as well as various local newspapers, it is possible to determine that he was a British citizen born in 1876, meaning that he was 38 years old at the outbreak of the war.

In 1917, he had given his original occupation in the USA as an inventor and that he was, by that time, a marine engineer and was running the McNab Company (and McNab Indicator Company) which made nautical and engineering appliances, including his ‘iceberg detector’, amongst others.

He would eventually move around Bridgeport as his fortunes increased through the First World War, with addresses changing from Post Office Arcade (1915), where the McNab Indicator Company held large offices on the 1st floor (2nd floor in America), to Fairfield Avenue (1916), and Brooklawn Park (1917). By the time of his last patent in 1931, he was still in Bridgeport but was residing on Main Street.

The elegant Post Office Arcade in Bridgeport, Connecticut. The building survives to this day.
Source: Bridgeport Library

The second man was Norman Leeds. Leeds was the managing director of the Automatic Machine Company (A.M.C.) in Bridgeport, Connecticut. Born on 15th November 1871 in Manhattan, New York, Leeds was an American citizen also residing in Bridgeport (on Boston Avenue). Unlike McNab, however, prior to 1917, he had no prior military experience disclosed on his US military census card. He had, however, a prior career with Western Electric Company amongst others, until 1908 when he and a few others took a controlling interest in the Automatic Machine Company. Leeds was also the President of the Board of Construction and Supply in 1914 and both he and McNab were donors to various charitable causes in the area, in particular when war broke out.

Norman Leeds pictured in 1922.
Source: The Bridgeport Times

Together, these two men worked to seek potentially lucrative engineering work for the new war in Europe. This is no surprise, as that area was producing vast quantities of arms at the time. The nearby town of Hartford, for example, was where John Browning invented his automatic pistol machine gun and automatic rifle in 1917 and the location was home to the Colt Armoury, which made more than ½ million guns during the war.

Both men were skilled and knowledgeable in the boat industry, with A.M.C. producing, amongst other things, boat engines ranging from a single-cylinder motor producing just 6 hp all the way up to a 6-cylinder 150 hp unit as their common motorboat engines. By at least 1913, they were also offering engines up to 250 hp.

In July 1912, Leeds was already in the news, traveling around Europe in order to promote and sell his marine engines. Business was obviously good enough to sustain the enterprise through to the start of the First World War in 1914.

Advertisement for Automatic Machine Co. marine engines. Source: Motor Boating March 1911

The Design

Using the lengthened ‘Alligator’ type tracks from the Bullock Creeping Grip tractor, the vehicle was to be some 23’ 6” long (7.16 m) long, 10’ (3.05 m) wide, and 11’ (3.35 m) high. The front of the vehicle was noticeably pointed, with an upturned portion at the bottom and then meeting at a point to the two sides forming a piked-nose. Directly above this double-glacis was a fighting section consisting of a semicircular shape with three guns pointing forward and to the sides. The sides of the hull were vertical and the roof curved, creating an arched roof over the large interior. The rear of the machine was rounded off with a pair of guns pointing backward. On each side of the machine was a small sponson projecting outwards, fitted with yet another gun. Surmounting the whole lot was a low cylindrical structure for the driver and commander to see out of.

The fuel tank was designed to sit directly above the water tank and directly in front of the driver’s position which, rather like the later German A7V, was atop the vehicle. The driver was, therefore, sat directly above the gearbox and controlled the direction of the vehicle with a simple steering wheel. This position would provide an unobstructed view of the terrain ahead, but also created a huge blindspot at least the length of the vehicle directly in front of it. This would mean that the vehicle was dangerous to maneuver against obstacles close by and, should the front elevate to cross an obstacle slightly, the driver would see nothing but the sky above, making control of the machine difficult.

Crew-wise, there are no details at all, other than an obvious driver’s position atop the machine. Assuming two men in the elevated driver position (a commander and a driver), and at least one man per gun, this would mean a crew of not less than 9 men.

Development and Timeline of the Alligator

It was apparent to Leeds and McNab that the war engulfing Europe brought with it certain commercial opportunities. It is also clear that reports of this new war and the shocking numbers of casualties were something their respective engineering skills might be able to redress. The result was Leeds’ idea for a fully tracked and armored fighting machine to break the deadlock. His initial design work on this vehicle idea was completed on 9th July 1915.

Shortly thereafter, he consulted with McNab and some changes were made to the design, with this second version ready on 14th July, making this a sort of Anglo-American project. No drawings are known of the first Alligator design to which a comparison with the modifications done in conjunction with McNab could be made. It is not clear, therefore, how extensive or visible, if at all, any of the changes were.

The Alligator, as it appeared in Scientific American in October 1916. Despite the lack of public knowledge of what British tanks actually looked like, the vehicle, as shown, is very competent. Full length tacks, a clear cupola to command and steer from and well positioned armanet on the front and sides. Note the distinctive swirl-pattern of the Bullock Creeping Grip wheels at the front and rear of the track. Source: Scientific American.

There was no point in having a design for this weapon of war and having no means of selling it, so, Leeds tasked McNab with taking it to the relevant British and French authorities. This is at least part of why Leeds brought McNab into the project in the first place, although this is perhaps unfair to McNabs’ skills. They were both in the same industry, both qualified and skilled men and McNab had the advantage of being British, ex-Royal Navy and therefore more likely to be taken seriously by the British establishment, as well as able to leverage whatever contacts or knowledge he would have as to where to go with the concept.

McNab left New York on 17th July, arriving in Liverpool on 27th July 1915. Upon arrival, he went to see Colonel Holden, then head of the British Army’s Army Service Corps (ASC). McNab left Holden with a copy of the plans for the vehicle, even though he was unable to assist McNab. It seems that rather than digging further into the military establishment in Britain, McNab chose to get straight over to France instead.

He had only been in Britain until 15th August, meaning a stay of just 19 days. Two days after his arrival in Paris, McNab tried to elicit interest from the French military authorities in the vehicle in a presentation at the War Office in Paris. McNab was left with the impression that the French were interested in the idea of the vehicle and especially in placing an order for engines from the firm. Nonetheless, he left on the 20th, after just 3 days, and returned to London. On 23rd August, McNab met with General Moir (Comptroller of Munitions Inventions) who, according to McNab, was so interested that he sent McNab to get the plans back from Holden to show him. Why that course of events was necessary was utterly unclear as, apparently, on his sales trip, McNab must have only brought two copies of his plans with him – something of an oversight for a sales trip.

Following the conversation over the plans with Moir, an appointment was made for him to attend the Naval Armoured Car Division at Pall Mall and once more reported that the officers he spoke to were very interested. He thereafter returned to the USA.

The first trip had clearly pricked some interest and it spurred a second trip, which took place in September that year. By the 8th of that month, McNab, visiting London with his wife, even managed to witness the aerial bombing of London by a Zeppelin from their balcony at the Metropole Hotel. Traveling to France on this second trip, McNab was able to speak with Monsieur Corcas, the Secretary to Albert Thomas – the French Minister for War. M. Thomas was later to be a thorn in the side of the nascent French tank program, which was working on a 2-man tank from Renault (the Char Renault FT), as Thomas had wanted a bigger machine.

McNab was back in Bridgeport by the 24th, meaning that this second trip – like the first, was an all too brief affair. Seemingly, no more was heard of the matter and, with the failure to obtain either engine orders separately or together with their vehicle design, both men went back to their normal business but still seeking to profit from the war. The same month McNab returned from his second visit to Britain, he was acting as a promoter for the New England and Pacific Steamship Company – a company he founded in 1915 to ship goods from Bridgeport and New London to the Pacific Coast via the Panama Canal. He advocated strongly for this war as a “golden opportunity” for the US shipping industry to produce as much new merchant shipping as possible, both for commercial benefit and replace the losses of Allied shipping by German submarines.

Even though their efforts had been unproductive, both men were still successful in other respects. In February 1916, McNab was giving his title as ‘Vice-President of Marine Specialties Ltd.’ and had made yet another trip back to France, where he had engaged with the French military authorities over his becoming an advisor to their engineering corps. The advisor stint perhaps was a little bit of an overstatement by McNab, as he was back in the USA in March 1916 with his wife who had accompanied him to France suggesting a little more of a business trip combined with sightseeing than a formal appointment as a technical advisor.

When in September 1916 Leeds and McNab got to hear about the use of tracked armored machines on the Somme, it is therefore forgivable and understandable why these men might believe that their machine was the basis of the British work. They could not, and would not have known of the top-secret work which had already taken place a year beforehand to develop a machine better than theirs. British work had, in fact, started in February 1915 – several months before their own efforts.

Certainly, McNab remained closely involved in both his Bridgeport community as well as providing talks locally on the war. In March 1917, he was providing local talks on the war in the Bridgeport area, claiming to have been to France to study the war, although, given that even the Landship Committee (the body tasked with designing and buildings Britain’s first tanks) was denied access to the front line, his reconnaissance would likely have been fruitless. The idea that he would be directly visiting the front is also undercut by the fact he had brought his wife on the trip and was mainly reported to be in Paris.

In September 1916, following the announcement of the British use of this new weapon, there was obviously a lot of attention paid to the machines and, in response to this event, Leeds was claiming that it was he who had invented the tanks as used by the British. He pressed the fact that it was he, not McNab, who pushed for their ‘tank’ design and that he had commissioned McNab to go to Europe in 1915. The phrasing of his claim is significant because when it states “….McNab… to go to England and France and try to enlist the interests of the Allies in the invention, but that since then the English have adapted the idea by using an English engine”. In other words, whether or not the idea was to promote the whole design or just the engine within it, as the Director of a business supplying marine engines, his concern was with engine production contracts.

It could be taken from that statement that his primary goal was only to sell engines, but then why go to the effort of designing or promoting the vehicle around them? This claim is also the origin of the ‘Alligator’ perhaps being explained as to why the name is applied to the vehicle as Leeds states that his design was more to do with that type of tracked vehicle than the Holt caterpillar.

In November 1916, just two weeks before the first photos of British tanks were published, Leeds provided the most thorough account of the theory, purpose, and design of the Alligator vehicle.

In that article, Leeds described that it was he who, upon realizing that combat on the Western Front had ground to a halt, had conceived of a vehicle based on this Alligator-type tractor chassis, long enough to cross trenches and that the tracks would not be bothered by enemy barbed wire. Clad in armor and fitted with weapons, the machine would break the deadlock and bring victory for the Allies – at least in theory. With the knowledge of the vehicle in use in September 1916, but unaware of what the machine was that the British were using or when they had started their secret design work (before his own), his claim to the invention is understandable if incorrect.


No armor thickness is specified in the writing available from Leeds or McNab. However, the protection was going to be substantial, as Leeds was wanting armor capable of protection from enemy 3” (76 mm) guns. However, when it came to convincing the British or the French over the design, Leeds and McNab were quite happy for the end-user to determine the final armor protection.

Although the protection level is not specified, it is reasonable to assume that, at an absolute minimum, protection from bullets would have to be provided, meaning at least 8 – 12 mm of plating. The available images of the Alligator show that the body is riveted together throughout. With the heavily angled front, the Alligator design would actually provide some sharp angles to incoming fire to help deflect shells and bullets and the same was true for plunging fire on the roof. Overall, this rather crude machine was well designed in terms of a shape for ballistic performance in comparison to its contemporary British designs.

Construction and Deployment

Leeds proposed at least 1,000 such machines would be required – certainly a very healthy contract if he had to provide the engines. For use, he imagined them operating as a naval screen, protecting the soldiers who followed from enemy fire both with their armor and also by attracting the enemy fire to them.


The artists for Scientific American clearly drew Maxim-type machine guns in the vehicle. Belt fed, these guns were definitely not the same as the ones shown in the plans of the Alligator, and neither Leeds nor McNab mention exactly what weapons they were proposing. The drawings are unclear, but there are some options as to what the guns may be.

A close-up view of the front of the Alligator from Scientific American clearly shows Maxim-type machine guns fitted. Note the vision slits to the right of the machine guns.
Source: Scientific American
Digitally cropped from the drawings shown in Scientific American the armament for the Alligator clearly sits atop a skewed conical mounting and is more than a mere machine gun. What appears to be a small hand crank is visible in the right-hand image and the outline of what may be a magazine sat on the top left-hand side at the back. Source: Scientific American as modified by the author.

The first option is the Driggs-Schroeder 1-pounder gun. Like all Driggs-Schroeder guns, this used a rifled barrel where the twist progressively increased towards the muzzle. The 1-pounder guns all had a caliber of 1.445 inches (36.7 mm).

The gun had started life in 1889, with a request from the US Navy for a 1 pounder gun that could outperform existing designs and still be under 100 pounds (45.4 kg) in weight. The result was the Driggs-Schroeder 1-pounder Mark I, with a 40-caliber bore and firing shell at a muzzle velocity of between 1,313 and 1,800 feet per second (400 to 549 m/s, respectively).

A second design followed shortly thereafter, known as the Mark II, with a 50-caliber barrel and a muzzle velocity of 1,884 fps (574 m/s). It used the same shell with 140 grams of black powder as the propellant, as used in the Mark I gun.

A final version of the 1-pounder was developed specifically for light vessels, such as yachts, and used a shorter bore (33 calibers) and a lighter charge than the preceding guns. Made in one piece as forging, the light 1-pounder was 12 lbs. (5.44 kg) lighter than the Mark I, weighing in at 88 lbs (39.9 kg).

The total shell weight, including case and propellant, was 1.53 lbs. (0.69 kg), with the actual projectile weighing 1.06 lb. to 1.10 lb. (0.48 to 0.50 kg) including a 0.03 lb. (13.6 gram) burster charge in the armor-piercing shell. With no burster charge or fuse, the armor-piercing shell weighed 0.94 lb. (0.43 kg). This shot was capable of perforating up to 1 ¼ inch (32 mm) of steel at point-blank range and up to ¾” (19 mm) at 1,000 yards (914 m). These shells were developed for naval combat but were more than sufficient to deal with the tanks of WW1.

1 pounder Driggs-Schroeder gun on standard conical recoil mount. The small size of the gun is readily apparent.
Source: American Ordnance Company

A lower weight of gun was obviously a good thing to help keep the weight of a vehicle down, as was the reduced charge. This would result in reduced recoil forces, meaning any mounting could be smaller and lighter as well. Considering that the low-power 1 pounder was designed on a simple conical mount and bolted to the wooden deck of a yacht, this was an excellent choice of gun for a tank design in terms of dealing with enemy bunkers or penetrating the shield of a field gun, or even enemy armor.

Unfortunately, the small size of the gun came with a serious handicap – a small shell. For the Driggs-Schroeder guns, steel shells with a base fuse and small explosive filling (Armor Piercing High Explosive – APHE or ‘Semi-Armor Piercing’ – SAP) and common (High Explosive – HE) shells were available for all calibers. The small size of the 1-pounder HE shell, however, would mean a very weak performance for a round which would be needed in the anti-infantry role or to smash an enemy position.

1-pounder steel shot fired from the Driggs-Schroeder gun and photographed after having penetrated a 1 inch (25 mm) thick iron plate (right), the shell on the left is a 3-pounder Driggs-Schroeder after passing through a 3 inch (76 mm) thick iron plate. Note that neither shell has burst from the small explosive filler so was presumably fired unfused to demonstrate the robust steel body of the round. The contrast in size between the two shows the diminutive size of the shot from the gun selected for the Alligator.
Source: American Ordnance Company
1-pounder standard shell (left) compared to the 1-pounder low-power cartridge (right). Both fire the same 92 mm long shell but from different sized cases.
Image composited by the author.
Source of the original: American Ordnance Company
Steel armor-piercing shell (left) showing the base fuse and small high explosive bursting charge compared to the common shell (right) with the larger charge. Image composited by the author.
Source of original image: American Ordnance Company

The drawing is, however, different from the Driggs-Schroder guns in some important regards. Firstly, the barrel of the Driggs-Schroder gun appears to be thinner than that shown and, if the item on top is a magazine or feed-trough, then the breech-fed DS guns are not the ones drawn.

Another option is the Hotchkiss 37 mm revolving gun, another common, albeit somewhat ancient naval weapon (it first came out in 1871). This one, unlike the Driggs-Schroder guns, had both a vertical magazine on top and a cranking handle. With five 37 mm barrels, each 20 calibers long, the weapon was a means of delivering serious firepower both at sea and on land.

Hotchkiss 37 mm revolving cannon.
Source: Koerner

Described as a cannon, the Hotchkiss revolving gun is basically a large compound machine gun with 5 barrels rotated by a cranking handle which also fed rounds from a vertical hopper or magazine on the top, firing them in turn as the barrels rotate. Each trough can hold up to ten rounds and, fed by another operator, the firing soldier operating the gun can fire between 60 and 80 rounds per minute. If he has to feed the trough on his own the rate of fire is still substantial, at around 40 rounders per minute.

For naval use, the gun weighed 200 kg, although the light version for field use was 225 kg, as well as having a ‘powerful’ version weighing 475 kg. The light versions of this gun were able to deliver their Common 37 mm shells out to 4,473 m.

Ammunition for the 37 mm gun included Steel shot, Common shells, and Canister shells, all using a metal cartridge case. The Common (High Explosive) Shell was made from cast iron and was hollow, with this cavity holding an explosive charge. Other calibers of this gun included 40 mm, 47 mm, and 53 mm caliber versions.

The Steel Shot was pointed and used no explosive bursting charge, relying on its mass, velocity, and shape to penetrate light armor and ship’s decks.

Left to right shells are Common, Steel, and Canister. The shell on the far right is a common shell, showing how large the complete round is with its case. Source: Composite image created by author from Koerner.

Even the Hotchkiss cannon is not a perfect match for the drawing. The feed-trough for the ammunition is not quite right and, of course, these are not multiple barrels shown. So perhaps the guns drawn in the Alligator are neither of these options. Given, however, that both Leeds and McNab were marine engineers and had expertise in shipping, it is no surprise they might select a gun like the Driggs-Schroeder or Hotchkiss. At the end of the day, however, they were also perfectly happy for the end-client to select and install their own armament to suit their needs.

Exactly what the guns were aside, the plans clearly show three of these guns in the front of the tank forming an arc across the front and able to provide fire across nearly 180 degrees of fire. Less clear are the side and rear guns, although these two appear to be the same designs, with one in a small sponson on each side, in a manner very similar to how British tanks did actually use them and then two more covering the rear. This meant a grand total of 7 guns although, in reality, rear-facing weapons would be of little use and the absence of a machine gun would render the machine more vulnerable to enemy infantry swarming the tank.

For the purpose of comparison, if the Alligator was fitted with the Driggs-Schroder guns and a modest ammunition supply of just 50 rounds per gun, this would mean a total load of about 520 kg. If it was the light Naval Hotchkiss 37 mm for the same assumption, it would be 2,263 kg – more than 4 times the weight.


Perhaps the defining element of the Alligator is not the shape, the guns, or even the attempts to sell it, but the selection of engine and track. The track, as already discussed, was an extended form of the Bullock Creeping Grip but the engine was not from Bullock.

The engine – the primary purpose of the entire project, was located just slightly aft of the center down the length of the tank and centrally along the longitudinal axis. It is clearly shown in ‘The Iron Age’ of September 1916 to be a 4 cylinder petrol motor from the Automatic Motor Company delivering 100 hp. The transmission lay in front of the engine and was itself preceded in the vehicle by a large water tank.

The ‘Alligator’ type tracks from the Bullock Creeping Grip tractor were to be 16’ (4.88 m) between the centers of the main wheels at each end and this length allowed the Alligator to cross a gap up to 8’ (2.44 m) wide.

For the Alligator, the Bullock suspension was stretched forming 6 distinct sets of bogies, each containing a trio of road wheels supported by a horizontal bar between them. Right In the middle of the length of the suspension is a single wheel on its own, meaning a total of 19 road wheels. This single central wheel was connected to the bogies fore and aft by a single horizontal bar running along all 7 wheels. Also visible in the cross-section view are 7 return rollers on the top of the track run, keeping the track tight.

Close up of the suspension system on the Alligator.
Source: Scientific American
Bullock Commercial Tractor with 150 hp engine and short tracks. Source: Le Gros
The lengthened Bullock Creeping Grip tracks ordered by Colonel Crompton and fitted to the Juggernaut / No.1 Lincoln Machine.
Source: UK National Archives

The track too was an issue. Colonel Crompton, whose work led to Little Willie, was certainly in favor of the Bullock track on the rather sensibly pragmatic basis that it was the only one available at the time that worked reasonably well. Nonetheless, there is no mention from Leeds or McNab of any idea of changing the actual track on the Bullock suspension so it can be reasonably assumed that the Alligator’s tracks would be little more than an extended version of those fitted to the Juggernaut/No. 1 Lincoln Machine.

The selection of the Automatic Machine Company 100 hp 4-cylinder petrol engine would certainly have been an improvement over the standard engines available from Bullock, of which the largest was the 75 hp 4 cylinder with a 5” bore and 6.5” stroke (127 mm bore / 165 mm stroke) as used on the Creeping Grip ‘Giant’.

Automatic Motor Company 4 cylinder engine as installed in a motorboat.
Source: Motor Boat, December 1920

However, assuming for a moment a similar level of performance from the Alligator as could be achieved at best from the Bullock Creeping Grip ‘Giant’, this would mean a top speed of 1.06, 2.4, and 3.4 miles per hour (1.7, 3.9, and 5.5 km/h) in 1st, 2nd, and 3rd gear and just 1.77 mph in reverse. Whilst the 100 hp engine from A.M.C. was larger, so too would be the weight of the vehicle and soft ground smashed by shellfire littered with the detritus of war and barbed wire would only serve to slow the vehicle even more. Certainly, the speed would be slow.

The End

In the end, the design of the Alligator came to nothing. The British, for their part, had already decided well before McNab’s first trip in 1915 on the direction of their own studies. Regardless of any advantages or disadvantages the Alligator had, the work did not appear to influence British designs in at least so far as the tracks and body shape. The one piece which could be argued was derived might be those side sponsons that were to become the dominant and most recognizable of features of British tanks of WW1. These too, however, may also be explained by the British designers – men like Sir Eustace D’Eyncourt who was himself a naval architect and simply took inspiration from naval weapon mounts as well.

Whatever claim to the invention of the tank in whole or part by Leeds and McNab they do not seem to have engaged with the Royal Commission on the invention of tanks after the war to press their case. Perhaps it is not too surprising either – there was a wealth of inventors both genuine and fraudulent after September 1916 claiming to be the inventor or inspiration for the tank. They were also successful businessmen and had moved on from their foray into tanks.

The design, however, was actually rather well organized, providing substantial firepower directly to the front as well as coverage over the side. In some regards, the design even had a better fighting arrangement for the crew than was on the later British Mk.I, as the engine was further back and control could be by a single driver rather than a driver and gearsmen having to work together.

The biggest flaw in the idea was the tracks. Whilst the lengthened Bullock system was a good system, it simply was not as good as the Tritton system the British were eventually to adopt and the track layout was just too simple. No Alligator tanks were ever built.

McNab passed away on 6th March 1941 and his former colleague, Norman Leeds followed him three later, dying on 29th October 1944.

What-if view of the Alligator Automatic Land Cruiser in green camouflage. Illustration by Pavel Carpaticus Alexe, funded by our Patreon campaign.


Alexander, J. 2015. Briefly Famous, The 1917 Caterpillar G-9 Tank and other American Tanks 1916-1918. Private Printing, USA
American Ordnance Company. The Driggs-Schroeder System of Rapid Fire Guns. The Deutsch Lithographing and Printing Company, Baltimore, MD, USA, 1896
Hills, A. (2019). Pioneers of Armour 2. Col. R. E. B. Crompton. FWD Publishing, USA
Koerner, A. (1879). The Hotchkiss Revolving Cannon. Private Publication, France.
The Farmer, 9th November 1914: ‘Norman Leeds’ parents in double funeral at Woodlawn Cemetery’
Bridgeport Evening Farmer, 14th June 1915. ‘To run freighters from Bridgeport to Pacific coast’
Bridgeport Evening Farmer, 20th September 1915: ‘See airmen fight over city roofs’
New Britain Herald, 20th September 1915. ‘London damaged by Zeppelin raid’
The Farmer, 24th September 1915: Connecticut shipyard owners have unusual opportunity to restore once famed business’
The Farmer, 2nd February 1916: Bridgeport Inventor now associate of French Army Corps’
Bridgeport Evening Farmer, 28th March 1916: ‘Briton’s won’t invade Germany asserts McNab’
The Farmer, 21st September 1916: ‘Norman Leeds claims invention of ‘tanks’ used by British Army’
The Farmer, 3rd August 1916: ‘M’Nabs indicator is advertised in wireless waves’
Harrisburgh Telegraph, 7th November 1916: ‘Automatic land cruisers was developed by American purely in an effort to sell engines’
The Farmer, 22nd March 1917: ‘St. John’s Men’s Club with hear Alex M’Nab’
Bridgeport Evening Farmer, 6th May 1916. ‘Plan additions at Post Office Arcade’s annex’
The Bridgeport Taimes, 17th April 1922. ‘Norman Leeds of community drive is optimistic’

Motor Boat Magazine, December 1920
US Patent US1103425 ‘Automatic-circulators for steam-boilers’, filed 18th July 1913, granted 14th July 1914
US Patent US1155832 ‘Boiler-circulator’, filed 19th November 1914, granted 5th October 1915
British Patent GB6228 ‘Improvements in or relating to circulators for steam boilers’, filed 26th April 1915, granted 9th March 1916
British Patent GB367608 ‘Improvements in Shock Absorbers’, filed 30th March 1931, granted 25th February 1932
McNab, A. (1920). Encyclopedia of Marine Appliances. The McNab Company, USA.
Pacific Marine Review, June 1920
Motor Boating Magazine, Vol.10, 1912
Motor Boating, February 1913
Motor Boat, Vol.17, 1920
Power Boat Magazine, Vol.36, 1925
Power Boat Magazine, Vol.21, 1921
The Iron Age, 7th February 1924: ‘Plans of New Companies’
The Iron Age. 28th September 1916. The Automatic Land Cruiser by W. E. Freeland.
The Shipbuilder and Marine-Engine builder, Vol.48. Obituary Commander Alexander McNab
United States Census 1900 Sheet 5A
United States Census 1910 Sheer 24A
United States Census 1920 Sheet 7B
United States Census 1930 Sheet 35B
United States Census 1940 Sheet 21A
US Military Census 1917 for Norman Leeds
US Military Census 1917 for Alexander McNab
Yale University. (1912). Quindecennial Record of the Class of 1895. Yale University Press. Connecticut.

‘Alligator’ specifications

Crew est. 9 (Commander, Driver, 7 gunners)
Propulsion A.M.C. 100 hp petrol
Armament up to end-user but shown with 7 guns of an unknown type
Armor up to end-user but desired protection from 3” (76.2 mm) enemy guns
For information about abbreviations check the Lexical Index
Cold War US Prototypes

Moon ‘Tanks’ and Lunar Combat

USA (1958-1967) Concepts – None Built

Source: US Army Future Weapons Office

Following the defeat of Germany in 1945, the two former allies and ‘superpowers’, the United States and the Soviet Union, began a decades-long confrontation and global competition which became known as the Cold War.

The United States, assisted by its allies, such as Great Britain, had developed a functional nuclear weapon in 1945, and was followed by the Soviet Union in 1949 (post-war, the USA refused to share this technology with the UK, which developed its own nuclear device in 1952).

A year later, in 1950, the newly formed United Nations went to war in Korea with the Chinese and Soviet backed North Korean forces in a 3-year long conflict which could have escalated into the use of nuclear weapons.

It was clear to the US military that, whilst conventional forces may face off against each other, nuclear weapons may also come to dominate the battlefields of the 1960’s and beyond, and also that, in the larger geopolitical rivalry with the Soviet Union, any edge may be decisive.

When, at the end of July 1955, the United States announced its intention to launch artificial satellites into orbit around the Earth, the Soviets saw this as a challenge to their own security. They then began their own space program, culminating in the launch of Sputnik I and II in October-November 1957, the world’s first artificial satellite and the start of what became known as the Space Race. This was a race not just to show some technical superiority over a rival power, but also had the potential to yield dominance of a fourth realm of warfare – space.

Dominance of space would mean control over enemy surveillance satellites, the ability to spy on anyone, anywhere, and also to mount weapons beyond the ability of an enemy to counter strike, but also meant some kind of base would be needed.

This would be a base not in the form of a giant man-made space station orbiting the Earth, but on the greatest natural satellite – the Moon itself. This was made official as a request in early 1958, when the US Secretary of Defense, Neil McElroy, ordered the Advanced Research Projects Agency (ARPA) to take “a close look at the moon”. This was followed on 2nd April 1958 by President Eisenhower declaring the need for establishing a single unified national space agency (the National Aerospace and Space Administration – NASA, was formed on 1st October 1958).

“Last year (1958), Air Force General Homer A. Boushey,… stated that a lunar military base someday will be vital to national security, that in one sense the Moon represents the age-old military advantage of [‘high ground.’] He is one of the many uniformed American leaders seriously thinking in terms of the military relation of outer space to the Earth and human events thereon. While there is perhaps no unified agreement in American military circles on the concept that ‘he who controls the Moon controls the Earth,’ nevertheless, there is well-grounded agreement that vital military keys to our future national defense lie in outer space.”

LTC Robert B Rigg, US Army, 1959

The Project Horizon Moonbase dug into the lunar surface, ‘Late 1965’. Source: US Army

The Moon, Earth’s nearest neighbor, measures some 3,476 km in diameter (equatorial) and orbits the Earth at a distance of between 356,400 km (perigee) to 406,700 (apogee) at an average speed of 1,022 km/s. With a mass of 7.342 x 1022 kg, just 1.2% of the mass of the Earth, it has a surface gravity of just 1.62 m/s2, just 16.5% of Earth’s gravity. The Moon has no functional atmosphere. There are gases to be found on the surface, but the density of molecules is so low as to technically form an exosphere. Whilst minute levels of oxygen, amongst other gases, have been found, the quantities that are so minute as to be of little or no use – certainly not enough for a person to breath or vehicle engine to use.

No atmosphere also means no filtering of the solar and cosmic radiation, with charged particles being attracted to the exterior of any suit or vehicle as a result of the lunar surface being well insulated, building up a charge on anything moving across it. Men or vehicles returning to a base would have to be ‘earthed’ (grounded) to remove this charge – yet another problem to overcome.

If these conditions are not harsh enough, the temperatures range from -130° C to 120° C at the equator and are even colder at the poles. The temperature variation in a small locale can be severe too. The Moon moves around the Earth at just 0.5° per second, meaning not only that days and nights are two-week-long, but also that the area in daylight from the Sun lying next to an area shaded may only be a few centimeters away from each other, but could be over a hundred degrees different in temperature.

The surface is heavily pitted with craters from asteroid and meteorite impacts that struck the crust of the Moon over billions of years. Those impacts led to the surface being covered with a fine dusty material called regolith, which can range in depth from as little as 3 meters deep to around 20 meters deep in places with numerous rocky outcrops. The surface is a difficult and barren zone and any vehicles conducting operations, or any men on the surface, would have to be able to both navigate, survive, and even fight in these conditions – no small feat. Navigation both on foot or by vehicle is, for example, complicated by the fact that the magnetic field is too weak for an effective compass to work (less than 200 nT [nanotesla] compared to Earth’s 22,000 to 67,000 nT) and that a man 1.82 m (6’) tall can see only 2 miles (3.2 km) at best and more usually just 1.5 miles (2.4 km) and could easily become disoriented and lost. Further, the lack of an atmosphere means that radio signals can not be bounced down over the horizon, so radio communication is limited to line of sight only. With this high vacuum, metals in contact with each other can weld themselves together and yet lubrication is complicated by the lubricants boiling off. Any metals should therefore have to consider solid lubricants (which do not sublimate), special surface anti-friction coatings and/or only be in contact with non-metals.

Unusual conditions required unusual vehicles and Project Horizon certainly had its share of them and this would require some careful planning.

It has also to be borne in mind that in 1959, the exact surface conditions on the Moon were unknown to the extent that a plan in 1959 called for detonating a 500 kiloton nuclear weapon on the surface to analyze the material thrown up and the seismic effects resulting from the blast. Thankfully, the idea of nuking the Moon for science was not carried out and the first actual lunar samples would, in fact, not come back to Earth until 1969 and 1970 with Apollo 11 (USA) and Luna-16 (USSR) respectively.

The most probable site identified was the Sinus Aestruum (Latin: ‘Bay of Seething Heat’) near the Sinus Medii (Latin: ‘Middle Bay’) (center of the map to the right of the Copernicus crater). From a 1:5,000,000 scale labeled Mercator-projection relief map ‘Lunar Earthside Chart LMP-1’ October 1970 of the Moon 50 deg. N/S. & 100 deg. W/E. Source: Lunar and Planetary Institute via US Air Force Aeronautical Chart Information Centre


Planning for a base of operations on the Moon would start in the first half of 1959 without physical samples of the lunar surface, but with an idea of the conditions based on scientific observation instead. Work on the project began with a feasibility study (Phase I), followed by a detailed development plan including how much it would cost (Phase II) through the rest of 1959 and the first months of 1960. All of the necessary development of technology and hardware needed would begin in 1960 (Phase III) with 4 years of solid work up to the start of lunar base construction in mid-1964. Phase III was to continue through to 1967, with the lunar base construction (Phase IV) finished by the end of 1966 and online (Phase V) by the end of 1967. The final phase, Phase VI, would then follow as capabilities for operations were progressively improved and expanded. All told, it was estimated to take nearly 10 years to have this base fully operational and, by that time (end of 1967), the base would be housing nearly 300 personnel.

The planning for Project Horizon fell to Lt. General Arthur Trudeau, who, at the time, was the Chief of Research and Development. He ordered the Chief of Army Ordnance to draw up the details in a 90-day planning exercise. The experts for this project worked for an agency known as ABMA – the Army Ballistic Missile Agency, and were led by none other than Dr. Wernher von Braun (23/3/1912 – 16/6/1977). Just as Trudeau tasked the Chief of Army Ordnance, who tasked Dr. Braun, Dr. Braun passed the task to Heinz-Hermann Koelle (22/7/1925 – 20/2/2011) as project lead, who brought in Dr. Georg Heinrich Patrick Baron von Tiesenhausen (18/5/1914 – 4/6/2018). Tiesenhausen had been an employee in the German missile program at Peenemünde (from April 1943) during WW2, where he had been a Section Chief in charge of ground support equipment and designed Pruefstand XII for the exhaust deflection system for the Schwimmweste [Eng: ‘life vest’] project (submarine-launched V2 rockets for targeting the USA) and had worked alongside Koelle, Braun, and others. He had not, however, been brought to the USA until 1953. Subsequent to the end of the war in 1945 and prior to being gathered up as part of Operation Paperclip, the Latvian-born scientist had been running a tank collection park for the British and working at a ship winch fabricator. The Project Horizon work was also to bring in every conceivable specialist at some point to have their say or express their views on the potential task at hand. This was an enormous project, despite the very narrow window in which to do it – certainly more than a simple paper exercise in lateral thinking, but a serious discussion over the challenges ahead.

Dr. Wernher von Braun (left) and Heinz-Hermann Koelle (right).
Source: NASA via Air and Space Magazine
Dr. Georg H.P. Baron von Tiesenhausen, the brains behind the lunar mobility program for NASA. Source: Huntsville Times via the Tiesenhausen family.

Lt. Gen. Trudeau kicked off the planning work in March 1959 and, by 8th June that year, the report was done and published as ‘Project Horizon: A U.S. Army Study for the Establishment of a Lunar Military Outpost’ – a plan costing US$6 bn. to US$7.7 bn. For reference, US$$6 bn. and US$7.7 bn. in 1959 dollars would be the equivalent of US$53.4 bn. to US$68.8 bn. in 2020 values. Whilst this is a huge sum of money, it is still less than the current US defence budget for 2020, which is US$686 bn., and the costs were going to be spread over a decade. It accounted for just under 1% of the total budget, not to mention this would be money spent by the government within the US on rockets and salaries, etcetera. Maintenance of a lunar base was estimated in 1960 as an immodest US$631m per annum, equivalent to US$5.5 bn. in 2020 dollars – still under 1% of the US annual defense budget. The 1960 viability report by the US Air Force (also considering their own lunar base) on a permanent lunar military outpost pointed out that its total 10-year cost was equivalent to the annual operating costs of the US Farm Subsidy Program; in other words, it was expensive but affordable.

Lt. General Arthur Trudeau, US Army Chief of Research and Development 1958-1962. Source:

The Project Horizon report would be the outline for how a base could be set up as a critical lunar system, for which a later decision would be taken on exactly how to militarise it and what for. The ability to conduct observations from a fixed platform onto the Earth was obviously valuable, but a base would go further than that, as can be inferred from the ominously titled 1959 report on the base plans “Lunar Based Earth Bombardment System” which could, according to a 1960 report, deliver weapons to Earth with an accuracy of 2 to 5 nautical miles (3.7 to 9.3 km), clearly indicating that nuclear weapons were going to be the weapons of choice.

Getting men, machines, and equipment there, setting up a base, and then protecting it from a potential adversary meant creating a supply system and potential weapons systems. All of this had to be integrated together and work in synergy due to the distance from potential help or resupply. Thus, all of the problems of such an outpost have to be considered together to understand the decisions made.

Getting There

The task of bringing all of the tools, equipment, and men to the Moon was to take 229 flights over a four-year period using the Saturn I and Saturn II rockets. Just getting men into Earth orbit would take 16 flights of the Saturn I and 6 of the Saturn II, with an additional 47 and 71 flights respectively to get the cargo into space. As well as the 300 personnel to be ferried back and forth between the Earth and Moon, using a clever rotation schedule to establish a 12-man outpost, approximately 245 tons of equipment and building materials had to be delivered to the Moon. This does not include the sustaining cost of maintaining the base either. A total of 61 Saturn I and 88 Saturn II launches would be needed over a 28 month period (a rate of 5.3 per month) to get set up, with 64 more to supply the base in year 1 – more than 1 per week. Later, larger and more efficient rockets, including ones propelled by nuclear detonation (predicted to be available from about 1970) would reduce the logistic burden of supplying the base. Bases for launching the Project were planned for either Brazil or Christmas Island, as locations near the equator were desirable.

“to be second to the Soviet Union in establishing an outpost on the Moon would be disastrous to our nation’s prestige and in turn to our democratic philosophy.”

Project Horizon, June 1959

The Base

Buried into the lunar surface, the base would be partially protected from damage by solar radiation and falling meteoroids. It would be built from a series of tubes connected together in much the same manner as many nuclear shelters on Earth, forming functional modules in which work would be conducted. The primary difference would be the source of power, and for this, it was planned to use nuclear reactors to provide sufficient electrical power for all of the heating, lighting and equipment. An initial 5 kW nuclear reactor would be landed to provide power for the construction camp, followed by a 10 kW unit and a 40 kW nuclear unit for the construction of the rest of the facility. At that time, the 5 kW reactor would be dedicated to just powering the life-support. Lacking radiation shielding (to save weight), these would be placed into craters blasted out from the lunar surface if natural ones could not be found, and then covered with lunar soil to a depth of 12 feet (3.7 m). For normal operations, all power was to come from the 40 kW reactor, with the 10 kW unit held for emergency backup use. Despite the extreme cold in places, spare heat would still have to be released from the reactors and this was to be done with metal radiators fabricated from scrap, such as from cargo containers. Poking out above the surface, these radiators would dump the excess heat through thermal radiation, but also ensure a steady source of radiation on the surface, meaning men would have to stay away. At 50 feet (15.2 m) from the radiator, the dosage for a man within that zone would be 300 millirems per week. For reference, background radiation on Earth is around 100 millirem per year. Such a dose for the soldier within that distance would be just over 3 times the US Nuclear Regulatory Commission (NRC) sets as permissible for someone working around nuclear material. However, this was safe for the lunar surface, as the two plants would be located over 100 meters from the living quarters, about 30 meters apart. A final 60 kW nuclear power unit would also be added later, meaning no less than 4 different nuclear power units for the base to sustain life and conduct its military purposes.

The V-shaped arrangement of trenches containing the cylindrical housing modules is put together with the multi-purpose lunar construction vehicle. To the right of the image and just in the distance, can be seen the radiators for the nuclear reactors sticking out of the lunar surface. Source: Project Horizon Vol.II

Vehicles – Construction

One of the first vehicles to consider was a construction vehicle which could perform a variety of duties for the initial set-up phase of the outpost. Bearing in mind that gravity on the Moon was just 16.5% of Earth’s, this would, at first, appear to make life easier, as a 1,000 kg load on Earth would weigh just 165 kg on the Moon, but the low gravity created problems for the vehicles as well as the stability of them. On Earth, stability is based in large part on ‘normal’ gravity. With the mass distribution as it is, even if a regular bulldozer could be brought to the Moon, the weight distribution would be incorrect, making the vehicle unstable.

Vehicles would have to be weighted in order to maintain an effective power-to-traction ratio and that weight would not be brought from Earth – it would be added locally on the lunar surface. Much of the lunar vehicle work was to fall to Tiesenhausen and, as a matter of record, it should be noted that he became head of lunar mobility systems at NASA in 1963. Tiesenhausen is also primarily responsible for the lunar rover as it finally appeared as well several years later, but in terms of Project Horizon, there was primarily a concern for a vehicle for construction duties.

Weighing just 4,500 lbs. (2.04 tonnes) as its basic structure, additional mass the vehicle would need to become stable would simply be provided by means of ballasting it with lunar material, bringing its operational weight up to 9,000 lbs. (4.08 tonnes). Operating on four 4’ (1.22 m) open-spoke metal wheels, no tyres were to be used, but the speed would be low – just 1.5 mph up to 5 mph (2.4 to 8.0 km/h) so the cushioning of a rubber tire was not needed. The wheels would have diamond-shaped grousers to gain purchase on the lunar surface, and with a mass of 2,040 to 4,080 kg would, in effect, mean a weight of just 16.5% of that, making traction easier. Powered by a pair of 4 hp electric motors (rechargeable from the nuclear power-generated electricity supply from the outpost), the entire vehicle would be just 15’ (4.57 m) long, 6’ (1.83 m) wide, and 6’ (1.83 m) high and could be controlled directly through a pressurized cab where a man could work without a suit, or via radio (subject to the limits of radio range already discussed). On Earth, such a vehicle would require 20 to 25 hp per ton to operate off-road but, with the reduced gravity of the Moon, the electrical power units (8 hp total) were sufficient to meet the equivalent engine-power on the Moon and provide an operating range of 50-150 miles (81 to 241 km) although, as previously stated, anyone operating that far from a base would run a serious risk of becoming lost.

Lunar construction vehicle. Source: US Army

It was fitted with a variety of attachments, such as a U-shaped bulldozer blade, robotic arms, a crane boom, and a power take-off for a variety of other equipment which may be needed, such as an auger. This vehicle was calculated to be able to carry up to 480 cu. ft. (13.6 m3) per hour over 250 feet (76.2 m) with its 43 cu. ft. (1.2 m3) front bucket and to bulldoze up to 750 cu. ft. (21.2 m3) in the same time over the same distance. The crane would be able to lift 4,500 lbs. (2,042 kg) up to 2 feet (0.61 m) from the surface at up to 10 feet (3.0 m) from the vehicle. The vehicle would also be suitable to tow heavy loads, such as moving the habitation modules into the trenches dug for them, or supply sleds. Those would be made from leftover cargo containers which would be cut up to form rudimentary supply sleds and also to produce shielding for the habitation modules to protect from radiation and meteorite impacts. The towing capacity of the construction vehicle would be around just 50% of the total vehicle weight, just 2.04 tonnes at best.

Vehicles – Lunar Transport

To avoid the soldiers in suits having to walk around on the surface too much, a transport vehicle was going to be required. Consisting of a skeletonised frame made from light-weight metal, it was going to be compact – just 6’ (1.82 m), square, with the seat and controls exposed – avoiding the need for a cab, although a two-man cab could be used if required. A small load deck, in the manner of a miniature pickup truck, would be at the back. Also electrically powered, this 3-axle vehicle would weigh just 2,000 lbs. (907 kg), and be able to move a load of men and equipment between 50% and 300% of its own weight, depending on the nature of the surface.

Each axle would have a wheel at each end and each would have its own 1 hp electric motor for a total tractive power of 6 hp. Suitable batteries would provide a range of 50 miles (80 km) or 10 hours of operation before needing to be recharged. Interestingly, the prospective design for this vehicle is missing from the available copies of Project Horizon.

A foreign analysis of other types of lunar vehicles was published for the US Air Force in 1967 from an original document by D. St. Andreescu of Romania. Andreescu was a Major General and a member of the Commission on Aeronautics in Romania and the article the US Air Force translated in 1967 was first published in 1963. Andreescu’s work had involved looking at conflict and exploration in space at least as early as 1957, so it is no surprise that his work would garner attention from the US.

Rocket-type lunar lander from the front page of Andreescu’s 1963 article. Source: Andreescu

The Foreign Technology Division (the US Air Force Branch responsible for identifying and analyzing foreign technical material) investigated what ‘the other side’ was looking at in terms of potential lunar vehicles. Originally titled ‘Vehicule Lunare’ [English: Lunar Vehicles] in the original Romanian article by Andreescu, the report considered that a Moon-based command post was, just like the USA had concluded, a logical step in the steady conquest and use of space. Andreescu considered a few different types of vehicles and types of power systems for lunar propulsion, including tracks, wheels, a large rolling spherical body, helicoidal screws, rockets, electric motors, walking machines, flying machines, and even leaping or ‘bounding’ vehicles. In doing so, he appears to have been drawing upon ideas or designs which predated this 1963 work, although it is unclear where he got them from. They may have been his own ideas from before or a collection of ideas from a variety of international sources – it just is not known as he provided no references. There are, however, two vehicles of particular interest from Andreescu. One, a giant tracked ball virtually identical to the vehicle shown in the Project Horizons plan, and a small tracked luna mobile. Whilst the former vehicle does appear in the Project Horizon report, the second does not – it only appears in a 1967 US Air Force translation of his paper. Of the multiple vehicles Andreescu discussed, therefore, it is those two which are of the most interest in terms of Project Horizons.

Spheres and Tracks

The ‘tank-like’ vehicle ran on wide tracks supported by four large road wheels and a drive or jockey wheel at either end. It is not clear which end the power was supplied to the tracks, as it is not shown in the available image of the vehicle. No supporting rollers are shown. The track is, however, covered across the top by a large dust shield to prevent clouds of regolith being thrown up as it moves. The body of the vehicle is flat-sided, with a shallow sloping front surface, which on a tank would be the ‘glacis’. The rear is more sharply angled. On top of the vehicle is shown what appears to be an exhaust stack for venting gas and/or heat, much like a ship’s funnel over the back end, with a large dome with multiple large rectangular windows at the front of the roof. Inside are shown a trio of astronauts.

Completing the vehicle are a pair of headlamps at the front, a flapped-horizontal slot and a circular feature (perhaps a light) on the ‘glacis’. Projecting from the front, just below those headlamps, are a pair of supporting arms for what appears to be a bulldozer blade.

It would be powered by electric motors running on compressed (pressurized) hydrogen provided by a turbogenerator and recharged by use of nuclear power generators. This power supply would propel the vehicle on the lunar surface at up to 40 km/h for 24 hours of continuous use between recharges. The heat irradiated from the passengers on-board would boil off the hydrogen fuel source and this heat was then recaptured by a lithium hydride recuperator. This power supply would weigh just 1,200 kg – more than 200 kg less than an equivalent diesel engine and without the problems of liquid fuels in a lunar environment.

With a crew of three in the bubble on top, this 1.2-ton lunar vehicle uses the heat from the humans as part of the propulsion system. This image is identical to the one featured in the 1967 US Air Force translation but does not appear in the 1958 Project Horizon plan. Source: Andreescu
Note: surrounding text from image has been digitally removed for the sake of clarity but the vehicle image has not been altered.

If the tracked lunamobile is rather mundane as a tracked vehicle, albeit one for the Moon, then it is the spherical vehicle idea from Andreescu which is perhaps the most curious. This was a vehicle with a spherical hull and held upright by means of “hydroscopic” (gyroscopic) stabilisation, ensuring the 3 or 4 men inside would stay correctly orientated in relation to the lunar surface. Running on a single circumferential track, the spherical lunamobile was thought to be useful for exploring the surface due to its shape, which would prevent it from becoming stuck in a crater. A second track ran circumferentially around the vehicle in the horizontal plane, meaning that, even if the vehicle toppled over to one side, it could move by being driven by both tracks either out of the obstruction or until the gyroscopes righted the vehicle.

Power was supplied by a large circular solar cell acting like a parasol above the vehicle, which could be angled to receive maximum solar radiation, although no other details are provided.

A gyroscopically-stabilised tracked ball vehicle with overlapping circumferential track, this highly impractical vehicle was pictured in some form to be useful for exploring the surface. This image is identical to the one featured in the 1967 US Air Force translation and also appears in the 1958 Project Horizon plan. Source: Andreescu

The curious matter about this vehicle, described in General Andreescu’s report from 1963, is that this vehicle features at all. It is, in fact, identical to the spherical lunar vehicle seen in Volume III of the Top Secret Project Horizons plan, a plan which was still secret through this time. Looking at the two views of this vehicle in these two reports, it is hard not to see that one party must have known about the other and, as Horizon was 4 years before Andreescu’s report, it would seem at first glance that Andreescu took inspiration from it for that image. The Project Horizon work was, for decades, Top Secret, so it is highly unlikely he could have seen the tracked spherical lunamobile. Instead, the logical conclusion is that the spherical tracked vehicle predated the Project Horizon in some non-classified form from which ABMA and, in due course, Andreescu, drew inspiration. Whether this was something serious or some science-fiction writing is not known at this point. Certainly, if ABM was copying from some preexisting idea from Andreescu, then the tracked and rather practical lunamobile could reasonably have been expected to have featured in Project Horizon.

The spherical lunar vehicle as pictured in the 1959 Project Horizon (left), and the 1963 Andreescu Romanian report (right). The similarities between the two (other than the horizontal track running in the other direction) are undeniable. Source: US Army and Andreescu (via US Air Force) respectively.

The use of the spherical exploration machine in the Project Horizon plan comes up during discussion of combat too, so it is fair to say that where concepts of combat on a lunar surface are concerned- they would, in the eyes of the planners, inevitably involve vehicles and probably tracked ones at that.


With a very credible Soviet space program and the sense of a race to the Moon, it was part of the planning activities of the program that there was a potential for the base to be attacked by the Soviets. It would, afterall, be a base for nuclear weapons and needed guarding. In considering combat on the lunar surface, the environmental considerations of temperature, pressure (or rather, the lack of pressure), and the low gravity made for a complex combat situation.

“…sole possession of a lunar outpost… would provide a military advantage in case of terrestrial hostilities. Defense must be provided to deter attack to obtain such an advantage”

Project Horizons Vol.I

Development of various elements of equipment to meet the needs of defending the men and equipment on the outpost was going to be divested to the various expert technical agencies. For example, nuclear weapons systems would be left to the Atomic Weapons Laboratories at Picatinny Arsenal, electronic systems would be in the hands of Diamond Ordnance Fuze Laboratory (DOFL), and missile/rocket design would go to the US Army Ordnance Missile Command (AOMC). On the surface of the Moon, vehicles were going to be needed to move around, build facilities, and provide security and these were to be left in the hands of the Ordnance Tank-Automotive Command in Detroit with responsibility for “developing and testing both combat and transport vehicles to meet special requirements”.

“If the moon and other planets are explored and possibly colonized, the world could eventually see a second evolution of weaponry and protection therefrom. Visualise starting with a weapon capable of penetrating thin skinned vehicles. The vehicles then get thicker skin. The weapons then attain a greater penetrating capability. The vehicles get even thicker skin until the weight and cost thereof becomes insurmountable. The weapons attain longer ranges etc., etc., etc.,. This proceeds through the mortar, howitzer, gun and tank stages until eventually you have missiles, antimissiles and nuclear weapons…..”

US Army Directorate of Research and Development, 1965.

Just as the comment above related to ‘tanks’, it was part of Project Horizon’s initial assessment that the use of vehicles in a potential attack or defense at the outpost was a credible concern. Although they could not carry armor in the sense of a conventional tank due to the difficulties in getting such a vehicle to the Moon, it was certainly conceivable such vehicles may have some ballistic protection beyond just retaining a pressurized atmosphere within the vehicle. Thus, any weapons planned for the outpost should also have had an anti-materiel ability to perforate a vehicle and depressurize it or cripple the automotive parts.

Such a unique combat environment precluded the use of conventional ammunition and weapons like rifles. This is not, as may be thought because the ammunition would not work in an oxygen-free atmosphere. It would work (albeit with some known problems with primer activation in a vacuum) – modern ammunition has its own oxidizer self-contained, so a rifle would indeed work. In fact, the lack of moisture and oxygen actually helps to preserve plastic explosives and prevent corrosion infuses, etcetera, although non-plastic explosives like TNT would degrade with the temperature extremes. Such conventional firearms were, however, designed to fire in Earth-gravity situations, so the sights would be almost useless on the Moon. Add to this that the reduced gravity on the Moon would seriously affect the trajectory of any bullet fired on the surface and that other features of a conventional rifle, like a stock, were not required because a man in a suit would not be able to carry the weapon easily in such a way and what is left is the need for a bespoke weapon – albeit it one which might use conventional-types of ammunition. However, that too involved a lot of weight-wastage through a casing. Newton’s Third Law of Motion states that for every action there is an equal and opposite reaction. In other words – recoil, and this recoil would be harder to manage for the soldier in a low gravity environment. With no effective atmosphere either, any missile fired would not be subject to drag from friction, like it would on Earth, and would increase the range of a projectile by some 6 times or more. The same was also true of the explosive out-gassing behind a recoilless weapon, making them extremely hazardous to friendly vehicles, buildings, or personnel behind. The same applied to explosively producing fragments/debris, as it would travel further and faster than on Earth. Given that penetration is a function of the square of the velocity of the projectile, even bits of regolith kicked up by a blast created an enormous hazard for men and machines.

The effect on-target of the projectile or directed energy device differed substantially from terrestrial combat too, as substantial damage could be done to a target simply by exposing it to the vacuum of space, piercing the suit or vehicle skin.

An attack on the outpost from the Soviets could not come without warning. Either the launch would be detected or the launched-craft picked up by Earth-based monitoring. If a craft could be landed secretly, the outpost would also have a short-range radar set with a range of up to 2 miles (3.2 km) and backed up by an active infra-red system with a range of 400 – 500 yards (366 – 457 m) range.

Planning assumption for outpost defense related to a roughly platoon-sized attacking force, but defense had to consider combat in 4 situations:

  1. Space-suited personal combat
  2. Lunar surface vehicle combat
  3. Moon-orbiting space vehicles
  4. Non-orbiting space vehicles

Project Horizon considered various modes of killing or incapacitating an enemy force approaching the perimeter of the outpost (scenarios 1 and 2 above). This perimeter defense had to be able to provide adequate means of killing both suited men and also vehicles and should be fireable directly or remotely. Options for such a weapon were considered and evaluated for both performances but also whether they were technically feasible.

These included

  1. Beamed electromagnetic radiation
  2. Flamethrower
  3. Blast weapons
  4. Fragmentation weapons
  5. Radiant energy

Defending against orbiting craft was more complex than a surface defense and, whilst an orbiting enemy could be tracked using Hercules, Zeus, or Hawk radar, it would be difficult to engage as available surface to air (or in this case surface to orbit) missiles were well beyond the 1,000 lb. (454 kg) weight limit imposed and a new bespoke weapon was needed. This would be expensive and with a long development time. An electron accelerator-based weapon was considered and would work by directing beams of gamma and/or neutron radiation at a craft to kill the crew and damage the craft. Just like the missile idea, however, this was far too heavy, 6,000 lbs. (2,722 kg) on Earth for such a system and still over 1,000 lbs. (454 kg) equivalent for a lunar-based system. In conclusion, Project Horizon rationalized that no effective defense could be provided against an orbiting enemy vehicle and the men on the outpost could, at best, dig into the surface to obtain protection until a suitable missile defense system became light enough to install there.

Personal Combat

With a set of unique problems from the environment (lack of atmosphere and exposure to solar radiation), equally unique weapons would be needed for troops to use if combat was to take place between US soldiers and a (presumably Soviet) force landing there.

Spacesuits would have to include an element of armor protection, along with the ability to self-seal following perforation, as a simple weapon that made even a small hole in a suit could prove lethal. More holes meant a greater chance of killing the opponent with the environment rather than the terrestrial method of killing the opponent by means of trauma. Weapons that optimized the delivery of high-velocity fragments were therefore considered the most desirable.

May 1959 design for a ‘Lunar Clothing System’. It owes more to Buck Rogers than a practical suit. The ice-skate-looking devices on the boots are soles to help stop the astronaut from sinking and presumably he would want some gloves too, although these are not shown in the drawing. Source: US Army

The suit proposed by Project Horizon included a hard suit for the body, made from titanium approximately 4 mm thick for standard protection and up to nearly 10 mm thick for maximum protection covering the stomach and chest, with a composite inner layer for resisting radiation. The same would go for a vehicle, with a titanium metal structure of approximately the same thickness to provide some resistance to fragments and also to maintain pressurization. Later, in analyzing how much protection 3.70 – 9.25 mm of titanium could actually provide on the lunar surface, it was calculated that a simple 6 grain (0.38 gram) fragment at 3,800 ft/s (1,158 m/s) or a 17 grain (1.10 gram) fragment at 2,400 ft/s (732 m/s) could pierce the suit or vehicle.

In terms of personal firepower, Project Horizon considered the various options and limitations which had to be considered, such as having to be self-sustaining.

The assessment was clear that the best weapon for defeating an enemy force either suited on-foot or in a vehicle of any description would be a Claymore-type mine. The standard Army type coming into service was the T48E1 mine, weighing just 1.36 kg and containing 675 5.56 mm diameter steel balls in a plastic resin. Once detonated, these balls, which were arranged in a slightly concave shape, would be fired out at a speed of 3,700 to 3,900 fps (1,128 to 1,189 m/s) across a 60-degree arc. On Earth, the mine’s effective lethal radius was around 200 feet (61 m), but on the Moon, with no atmosphere to slow down the balls and low gravity, the effective lethal range was increased to 2,500 feet (762 m), although the spread of the balls at that range would be extreme.

The Claymores could be set as a perimeter defense to be triggered upon command or by a tripwire, but they could also be handheld. This was not as hazardous as it may sound as, due to the vacuum of space, there was no blast hazard to the man detonating it on the end of a pole 5 to 6 feet (1.52 to 1.83 m) long. With a suitable shield surrounding the back of the mine to prevent mine fragments from being fired at the soldier wielding this weapon, in total, it was under 1 kg in weight. This made it simple, cheap, and easy to use for all combat purposes against men or machines.

T48E1 Claymore Antipersonnel mine. Source: US Army
Claymore mine detonated to kill approaching enemy forces on the lunar surface. Note the presence of the tracked spherical lunamobile in the image. Source: US Army.
Handheld claymore-type directional fragment weapon. This odd-looking device was ideal for use against men and vehicles on the lunar surface. Source: US Army
Artist’s impression of the handheld claymore device in use by a US Soldier against an unidentified adversarial force on the lunar surface. Source: US Army.

Whilst the attack scenarios involving orbiting were complex, the judicious use of detonating a nuclear weapon nearby was possible. Detonating a 10MT nuclear device in space, launched from the outpost for protection would, for example, have a lethal radius of around 320 km, although it did mean having to station nuclear weapons there and obviously accepting the risk of killing yourself with your own weapon’s radiation (unlike a detonation in Earth’s atmosphere, a nuclear blast in a vacuum primarily produces intense radiation rather than blast or thermal energy release). Even should the outpost detonate a weapon in lunar orbit to defend against an attacking craft, the lack of atmosphere on the surface meant that there would be no attenuation of the intense radiation, creating a hazard for the outpost.

Proximity to the blast was one of the notable flaws of the Army’s new M-28 ‘Davy Crockett’ nuclear weapon system. Often referred to as a recoilless rifle system, the M-28 was, in fact, a type of spigot mortar (120 mm) that could throw the 23 kg Mk.54 nuclear warhead weighing around 2 km. With a yield of 0.01 to 0.03 kT of TNT, the warhead on Earth would cause severe damage over a 10 m radius and moderate blast damage out to around 500 m (surface detonation), and up to a kilometer as an airburst. On the Moon, with no atmosphere to attenuate the explosion, damage could spread a lot further but so could the potential range of the warhead.

The Davy Crockett, fired at a maximum angle of 45 degrees, could, from the Moon, reach a lunar altitude of just over 1,200 m – high enough to be a threat to a descending enemy craft but not high enough to engage an orbital target and both with serious personal safety issues.

M-28 ‘Davy Crockett’ Short Range Spigot system. Source: US Army
The variation in terrestrial and lunar trajectories for the M-28 ‘Davy Crockett’ are shown starkly in this graph where the range on Earth is just 1,800 m compared to on the Moon at 15,500 m.
Source: US Army

An alternative to the M-28 was also considered, one which was more accurate, lighter, and simpler, but with a shorter range. This was the Dumbo II, which was still on the drawing board and consisted of little more than a launching tray on legs with a rocket-propelled bomb loaded into it. On the lunar surface, this device would have a range of just 4,400 yards (just over 4,000 m) with an accuracy of 40 yards (37 m). It was, in the understatement of the decade, “desirable” to increase the range. With a range improvement to 17,700 yards (just over 16,000 m), it would increase the weight from 130 lbs. (59 kg) to 160 lbs. (72.6 kg).

Dumbo II short-range nuclear weapon system. Source: US Army

In 1965, this subject of lunar personal weapons was examined once more by the US Army in the unusual title report “The Meanderings of a Weapon Oriented Mind When Applied in a Vacuum such as on the Moon”.

The cover-art of the June 1965 Army report features a weapon carrying vehicle on multiple legs with two space-suit clad soldiers. One is showing off with a pair of weapons in use at the same time followed by the less well armed soldier behind. Such artwork on an Army report was no doubt very exciting to see, but did not add a great deal of technical credibility to the reader and is more in keeping with science fiction rather than the plausibility of lunar combat. By the time of the report, all matter of lunar exploration had been handed over by the Army to NASA. Source: US Army

With the problems of cold welding over close-fitting and non-low friction metal parts, no lubricants, the temperature extremes in which it would have to operate and the issue of recoil from conventional-type weapons, solutions to a projectile weapon were perhaps harder to overcome in some regards than for a vehicle. The laser (‘Light Amplification by Stimulated Emissions of Radiation’), a concept of creating a focussed and intense beam of light, had been theorized for some time but did not exist before May 1960, when Theodore Maiman made the first laser (a ruby-laser) at the Hughes Research Laboratory in California. To theorize a laser weapon was beyond the ability of engineering in 1965 when the paper was written. It was not until 1967, with Co2 lasers of over 1,000 watts, that the first lasers were powerful enough to cut through even a sheet of steel 1 mm thick (TWI- The Welding Institute Cambridge, England), and the US Army paper actually suggested a lead time of at least 20 years. Although proposed in 1965, therefore, as a solution, laser-based systems were just pie in the sky, leaving recoilless weapons or very-low recoil weapons as the next best probable alternative.

Projectiles considered were rockets of various descriptions, balls (spheres), flechettes, and even rocks, which, based on the low gravity environment, would travel 2.73 times further than on Earth. For example, fired from the shoulder of a 6’ (1.8 m) tall man 5’ (1.5 m) from the ground at 3,000 f/s (914 m/s), a projectile would go 8,190 feet (2,500 m) when fired horizontally. Rather hazardously, however, was that due to the lack of atmosphere, if the projectile were fired upwards above the horizontal plane, it could go enormous distances. At 45 degrees, for example, a projectile launched at 3,000 f/s (914 m/s) (well short of the escape velocity of 2,400 m/s), would travel 320 miles (515 km), having reached an altitude of 80 miles (129 km) above the lunar surface.

The lack of ‘drop’ on the round added the severe hazard of projectiles in flight for large distances and long periods of time but had one significant advantage – the lack of anything other than the most rudimentary of sights being required. This is because a projectile would drop just 60 mm or so for every 100 m. Given a maximum line of sight engagement range of maybe 1.5 km, a projectile would, at worse, drop just 90 cm.

The lack of atmosphere also meant that wind resistance was effectively zero, so the shape of the projectile did not matter in terms of an efficient ballistic shape. Although ‘armor’ was not probable, it was objectively going to be easier for a spacesuit to resist perforation by a blunt object than a pointed one. Considering that vehicles were also a possibility, a pointed projectile with a high sectional density was preferred to aid in perforating the skin on those as well. The report concluded with a series of prospective weapon types summarised below. Noteworthy is that none of them are as expensive or perhaps simple and dangerous to men and vehicles as the Claymore-on-a-stick proposed in Project Horizon.


Exo-activity on the lunar surface (or even in space, as a ‘space-walk’) had not been done in 1959, so some of the suit-ideas and the science of space-suits were not fully developed by the time. There were, however, clear goals for a practical suit and the advantage of the low gravity meant that a 300-pound (136 kg) spacesuit would weigh just 50 pounds (22.7 kg) equivalent on the lunar surface.

Nonetheless, wearing a suit is a tiring task and normal wear would be limited to just 8 continuous hours, although 12 hours was acceptable. With a degree of caution, a man may operate in his suit for up to 24 hours and, in a dire emergency, 72 hours.

Referred to as a “metallic body conformation suit”, the outer surface was to be shiny (like all good 1950’s sci-fi) in order to reflect heat, but tough enough to resist the impact of micro-meteorites.

Any combat taking place in such a suit would be awkward and obvious to any attacker.


In 1959, there was a serious lack of crucial data on which to make some decisions – such as where to even put the base. Earth-based observations of the Moon could show the ‘seas’ – large flattish areas with few impacts, but the best available map being made by the Army Map Service at the time was just 1:5,000,000 scale and the contours were 2,000 feet (610 m) apart, meaning what looked flat could actually be quite a steep slope and landing on a cliff was ill-advised. An improved version of the map, due in around 1960 using better methods of stereo-image analysis, would be 1:1,000,000 scale with contours just 300 meters apart, with the aim of getting accuracy on surface features down to 150 or even 100 m by 1962. It is obvious, therefore, that with so little known, any outpost might find itself on impossible terrain. Any such landing without a better idea of the terrain was impossible in practical terms.

This was a fundamental part of the death knell for such ideas. It had not been the costs that killed Project Horizons perhaps as much as politics and practicality. It was not seen as being unfeasible or even unaffordable but perhaps just not necessary. The idea of launching at least one new Saturn rocket a week for years on end was one thing the Army could live with, but with the formation of NASA in October 1958, the writing was on the wall for the Army. NASA would take over all of the space work from the Army and Project Horizon, in particular, was handed over to their control by General John Medaris in March 1960. NASA had no interest in this project and it was thus left to die. The rest of the Army’s role in space was also handed over by the first half of that decade and the Army regained its terrestrial focus.

No weapons base on the Moon meant no need for weapons against men and machines, no lunar combat, and no ‘Moon-tanks’.

In 1967, with the signing through the United Nations of the Outer Space Treaty (OST), all signatories (including the USA and the Soviet Union) agreed not to place nuclear weapons in space (Article IV) either in orbit or on a celestial body (like the Moon), and that the Moon, in particular, was to be used exclusively for peaceful purposes. It also prohibits the establishment of military bases, installations, and fortifications on the Moon (Article IV), although military facilities for peaceful purposes were not prohibited. The treaty did not end ideas for a Moonbase, like Project ARES in the 1990s, but it did preclude space-based nuclear weapons. A further treaty, known as the Agreement Governing the Activities of States on the Moon and Other Celestial Bodies (‘The Moon Treaty’) signed in 1979 and effective from 1984, added to the Outer Space Treaty by prohibiting the creation of military bases although neither the United States nor the Soviet Union (or now Russia) have ratified that treaty.

Article IV
State Parties to the Treaty undertake not to place in orbit around the Earth any objects carrying nuclear weapons or any other kinds of weapons of mass destruction, install such weapons on celestial bodies, or station such weapons in outer space in any other manner.

The Moon and other celestial bodies shall be used by all State Parties to the Treaty exclusively for peaceful purposes. The establishment of military bases, installations, and fortifications, the testing of any type of weapons, and the conduct of military maneuvers on celestial bodies shall be forbidden. The use of military personnel for scientific research or for any other peaceful purposes shall not be prohibited. The use of any equipment or facility necessary for peaceful exploration of the Moon and other celestial bodies shall also not be prohibited.

United Nations Office for Disarmament Affairs: Treaty on Principles Governing the Activities of States in the Exploration and Use of Outer Space, including the Moon and Other Celestial Bodies 1967. Underlining added by the author. Full text at

No national sovereignty rules in outer space. Those who venture there go as envoys of the entire human race. Their quest, therefore, must be for all mankind. And what they find should belong to all mankind.
President Lyndon Johnson in ‘The President’s News Conference
at the LBJ Ranch (29th August 1965).
Collected in Public Papers of the
Presidents of the United States:
Lyndon B. Johnson: 1965 II (1966), 944-945.

Project Horizon was, perhaps, more of an embarrassment just a few years after it had been completed. The new NASA-controlled program was not interested in nuclear weapons on the Moon and the Defense Department fought to prevent its release to the public. In a 3 year struggle (1959-1962), the project was subject to numerous requests for release and, after no less than 10 separate security reviews by 3 different intelligence agencies, it was finally revealed at the end of the year. The project was effectively obsolete before it had even been finished, regardless of future changes, but it can be said that the project did form an important mental step in the long process of getting men to the Moon and, for Tiesenhausen – for what would eventually convey astronauts on the lunar surface with the Apollo 15 mission in July 1971.

Tiesenhausen’s work finally reached its fruition in the form of the lunar rover (Lunar Roving Vehicle – LRV) on the Apollo 15 mission. It took the simple option of tires over tracks. Source: NASA

A small 1.2-ton construction vehicle meant for the lunar surface. It would have been useful not only for constructing the base but also for constructing defensive positions. Illustration by Pavel ‘Carpaticus’ Alexe, funded by our Patreon campaign.


Andreescu, D. (1958). Incotro, spre Marte sau spre Venus. Ed. Militara MFA. Bucharest, Romania
Andreescu, D. (1959). Va scriu din Gara Cosmica. Ed. Militara MFA. Bucharest, Romania.
Andreescu, D. (1959). Racheta Arma Si Vehicul. Ed. Militara MFA, Bucharest, Romania.
Andreescu, D. (1963). Vehicle Lunare. Stiinta Si Tehnica No. 9, 1963. Romania.
Andreescu, D. (1967). Lunar Vehicles. Foreign Technology Division FTD-HT-66-132. US Air Force systems Command. Translated from the original in Stiinta Si Tehnica No. 9, 1963.
Bernstein, L. (2004). The Army and Space 1958-1984. Army Space Journal, Fall 2004
Directorate of Research and Development. (1965). The meanderings of a weapon oriented mind when applied in a vacuum such as on the Moon (U). Directorate of R&D, Future Weapons Office, Illinois.
Georg von Tiesenhausen (Dr. Donald Tarter Video Interviews) Interview date 1/1/1987.
Godwin, R. (2019). The Forgotten Plans to Reach the Moon – Before Apollo. Air and Space Magazine
Hall, R. (1992). The Origins of U.S. Space Policy: Eisenhower, Open Skies, and Freedom of Space. BMD Technical Information Center, Washington D.C., USA
Huntsville Times. Obituary: Georg H. P. Baron von Tiesenhausen. 8th June 2018
Kipp, J. (1988). The Problem of Space in Soviet Operational Art. Soviet Army Studies Office, Fort Leavenworth, Kansas.
Lang, S. (2017). SMDC History: Project Horizon: ABMA explores a lunar outpost. US Army.
Merstallinger, A., Sales, M., Semerad, E., & Dunn, B. (2009). ESA-STM-279 Assessment of cold welding between separable contact surfaces due to impact and fretting under vacuum. ESA Communication Production Office, The Netherlands.
Military Lunar Base Program (C) or S.R. 183 Lunar Observatory Study (U). Vol. I. Study Summary and Program Plan. ARDC Project No. 7987 April 1960. Air Force Ballistic Missile Division, California.
NASA Public Affairs Office. (1986). Biography of Georg F. Von Tiesenhausen.
National Space Society: Fischer, (D). (2010). 1959 – Twelve Men on the Moon.
Peenemunde Interview Project. NASM.1999.0038. Von Tiesenhausen, Georg. Interview conducted 22/1/1990.
United States Congressional Serial Set. Issues 12750-12751. (1972). Senate Documents 92nd Congress 1st Session January 21 – December 17, 1971, Vol. 1-2. United States Government Printing Office, Washingon D.C., USA. Page 67-572 O-71-6
US Army. (1959). Project Horizon Volume I: Summary and Supporting Considerations. United States Army, Washington D.C.
US Army Ordnance Missile Command. (1960). A Lunar Exploration Program Based Upon Saturn-Boosted System. RCS-ORDXM-C-1004.
NASA. (2014). Lunar Temperatures from Lunar Reconnaissance Orbiter.
Reiffel, L. (1959). A Study of Lunar Research Flights Vol. I. Defense Documentation Center for Scientific and Technical Information, Air Force Special Weapons Center.
Riggs, R. (May 1959). Outer Space and National Defense. Military Review Vol. 29, No. 2,
P.21-26. Sharp, T. (2017). Atmosphere of the Moon Redd, N. (2017). What is the Moon made of?
Space World. (1987). Last of the Rocket Team: A Conversation with Georg von Tiesenhausen.
US Army. (1959). Project Horizon Volume II: Technical Considerations and Plans. United States Army, Washington D.C.
US Army. (1959). Project Horizon Volume III: Military Operational Aspects, United States Army, Washington D.C.
US NRC. Personal Annual Radiation Dose Calculator.
US National Security Council Action No. 1846, 22nd January 1958 (NSC 5814/1 ‘Preliminary U.S. Policy on Outer Space 18th August 1958’)
United Nations Office for Disarmament Affairs: Agreement Governing the Activities of States on the Moon and Other Celestial Bodies.
United Nations Office for Disarmament Affairs: Treaty on Principles Governing the Activities of States in the Exploration and Use of Outer Space, including the Moon and Other Celestial Bodies.
Washington Post. 11/10/1962. Moon Study is Released by Pentagon.
Washington Star. 25/9/1962. Moon Study Gets Security Treatment.

WW2 Australian Prototypes WW2 British Prototypes

Gerrey Machine Gun Motor Vehicle

Australia Australia/United Kingdom (1939)
None built

Budgong Gap may not be the sort of world-famous location associated with great architecture or magnificent structures of the ancient world. Nor is it a place with any association in the world of armored fighting vehicle manufacturers, yet this somewhat obscure location, lying nearly a 3-hour drive south of Sydney, New South Wales in Australia, does have one claim to armored vehicle fame – the Gerreys. World War 2 broke out for Great Britain on 3rd September 1939, when it declared war on Germany after the Wehrmacht had invaded Poland. Australia followed suit, with Prime Minister Robert Menzies announcing that Australia was also once more at war. Within a month, the Gerreys had submitted a design for their own tracked and turreted weapon of war. Looking like a tracked motor car with a turret, the Gerrey design is perhaps yet another of those well-intentioned designs submitted in wartime but is also one of, if not the first of Australia’s homegrown armored vehicle designs of the war.

The Gerreys

The Gerrey family were ranchers/farmers from a very rural part of Australia and had been established in the area from at least the turn of the century.

The two inventive members of this family were Bernard Bowland Gerrey and James Laurence Gerrey. Although it is not known what their relationship was, it is surmised that they were brothers. Both men gave their occupations in Budgong Gap in 1939 as Timber Hauliers. Bernard had previously been in the National Press in 1931, relating of a possible sighting of the wreckage of the Southern Cloud, an Avro 618 which was lost in bad weather in March that year over densely forested wilderness. There would be no fame or reward for Gerrey for that foray into the public eye and the plane was not found until 1958.

The Design

In October 1939, when the Gerreys (both Bernard and James) submitted their design, the vehicle was intended to fulfill a single simple objective – to provide a “motor propelled armoured vehicle” with a turret “in which a series of automatic machine guns are mounted in superposed groups”.

Shaped like a giant boot, the vehicle effectively had the appearance of a tracked saloon car with a giant cylinder sticking up on the back of it. Access to the vehicle was via a pair of large rectangular doors on the right-hand side of the vehicle. The left side is not shown in the drawing, but it can be assumed that these doors were duplicated on the left side as well. The first door was directly accessing the cab area of the ‘car’ part of the vehicle, roughly halfway along the side. The second door was at the back of the vehicle, below the cylindrical turret. Judging from the size of the doors, it would indicate enough space inside for perhaps as many as three men, although a crew of two is perhaps more reasonable as only one man would be needed to drive the vehicle and another to operate the weapons. One final note on access is what appears to be a series of 5 or 6 parallel steps on the right-hand side (and presumably the left as well to match). These steps were on the sides of the vehicle running vertically just behind the door to the driver’s position and running up to the roofline. There appears to be a third hatch shown on the side view just above and behind the side door as well, although it is neither mentioned in the text of the patent application nor in the plan view drawing.

Outline of the Gerrey Machine Gun Motor Vehicle, as modified from the Patent image by the author.

The other very notable feature of the design is the very elegantly sculpted cowls for the pair of headlamps on the wings of the vehicle over the tracks. Such a design feature provided zero military advantages and was clearly inspired by a civilian style of automobile instead.

A crew of two would also match the plan view of the vehicle, which shows just a single opening in the front for the driver and no such opening or weapon on the left of the cab to indicate the need for a second crewman in the ‘car’. Likewise, in the turret, the plan view clearly shows a tractor-style seat for the weapon operator to sit on, indicating just a single man in the turret. At most, therefore, a crew of three could be hypothesized with the third man presumably languishing inside the passenger side of the ‘car’, occupied perhaps with managing a radio.

Both the 1931 report of a possible aircraft sighting and their employment as Timber Hauliers in 1939, as well as the rugged terrain in which they lived, indicate that they should have had at least a working knowledge of vehicles off-road, such as log-hauling tractors. This may account for the flat style of tracks used, as they have the appearance of the type more associated with industrial plant-like crawler tractors and even tracked cranes than the sort of tracks associated with military vehicles, which usually have a well-defined and raised leading front edge for the track.

Indeed, the flat style of track selected is not usually a problem on industrial machines, as the front and back of the bodywork rarely project the way they do on this vehicle. The projection over the back of the vehicle would, for this design, severely limit the steepness of a slope that could be climbed and likewise, at the front, the projecting bodywork would foul on the ground reducing trench crossing ability. This is perhaps the most surprising weakness of the design of the machine from the Gerreys.


The ‘car’ shape of the body of the vehicle is misleading, as the means of propulsion is decidedly not-car-like. Instead of running on rubber-tired road wheels, like a normal passenger motor car, this vehicle instead used a pair of full-length and rather narrow tracks. The Gerreys described the means of propulsion for these tracks as consisting of an “internal combustion motor of ordinary type and associated with the usual driving and change speed gear”. However, despite showing the vehicle on tracks, the Gerreys also suggested that this type of turret and armament system could be mounted upon “any kind of transport vehicle”.

The side view available from the patent indicates 20 or so large flat links per track with power delivered via a toothed sprocket located at the back. Each link is overly large for such a small vehicle with a large pitch and would indicate that the vehicle would have serious limitations on its top speed. Likewise, the positioning of the drive sprocket and idler, both in contact with the surface, would create a most uncomfortable ride even if there was any suspension shown or described, which there was not.

The engine, as shown in the plan view, lay at the front, under the bonnet, as would in the case of an ordinary passenger motor car, with the radiator in front of it. In order to protect the radiator from enemy fire, a set of hinged and moveable plates were fitted to the front, which could be opened in order to access the radiator and also allow more air in to improve cooling. Although the lines of the front of the car suggest a bonnet that could be opened in order to access the engine, there is none showed, which would mean a difficult time for anyone trying to do maintenance on the vehicle.


The primary weapons for the Gerrey design were a series of automatic machine guns. All of them were connected together in the wall of the turret and provided with a slot or other opening through which the barrels could protrude. All of the machine guns could be operated by a single crewman using a simple one-pull trigger, firing all of them at the same time. This wall of fire was no doubt an impressive thought. Assuming perhaps a rate of fire of 600 rounds per minute from each of the unspecified machine guns, it would be essential that each one was fed by a belt, or else the only job of the gunner would be changing magazines. With three rows of machine gun pairs fitted within the large cylindrical turret, a total of at least 6 machine guns (and as much as 18) are shown with their barrels projecting from small loopholes. Six machine guns firing 600 rounds per minute would be 3,600 rounds per minute fired in the general direction of an enemy through a narrow opening, allowing for very limited vertical movement of the guns.

As the vehicle, as drawn, would be using weapons compatible with British supplies, it would suggest weapons like the Browning machine gun and a bullet caliber of .303 or 7.92 mm, like that fired from the BESA machine gun. Belts for the Browning machine gun came in lengths of 500 rounds with two belts to a crate. Assuming the crates or boxes were dispensed off, the weight of the ammunition alone to serve these weapons is a significant burden the Gerreys failed to take into account, as one belt alone weighed in the region of 7 kg. One minute of sustained fire therefore would demand:

((No. of guns x rate of fire) / rounds per belt) x weight of belt

((6 x 600) / 500) x 7 kg = 50 kg of ammunition per minute

That is 50 kg of ammunition per minute, so even a modestly useful combat load of just 10 minutes-worth of fire would mean carrying half a tonne in ammunition alone. For anything other than perhaps anti-aircraft work, the six machine guns were simply adding the weight of the guns, complexity of the mechanism, and weight of the extra ammunition for no particular benefit. Given that the design of the Gerrey vehicle would preclude the anti-aircraft option due to the limited elevation of the guns, it has to be considered that a far more useful and practical arrangement of the firepower from this vehicle would have been served with just a pair of machine guns.

It was, however, these superimposed rows of automatic weapons, all operated by a single person, which was the crux of the patent invention, as they felt that this style of turret was both novel and important. Thus, the Gerreys patented something novel but inherently unusable.

Plan view taken from British Patent GB537405 showing the engine and radiator at the front and the protective covers over the front grill. Note the vehicle is right-hand drive, as would be expected from a British or Australian vehicle. The single-seat for the gunner and the array of automatic weapons in front of the seat can be seen in the turret, shown facing backward. Source: British Patent GB537405
Plan view of the layout of the Gerrey’s machine gun motor vehicle taken from the patent and cleaned by the author to remove annotations and interior. The radiator shutter has also been added. Source: British Patent GB537405 as amended by the author.


Few specifics of the armor on the vehicle were described in the patent application, other than to say that the chassis was relatively lightly armored. From this, it can be inferred to be protection against small arms fire rather than the type of protection that would be needed against anti-tank shell fire.

The turret was sat recessed slightly into the body of the vehicle, which would reduce the chances of it being jammed by enemy fire and prevent splash from entering the vehicle through the gap.


The vehicle submitted by the Gerrys was a simple one, well within the technical abilities of the day to produce, but it was also redundant and somewhat naive. Even assuming the engine would have had sufficient power to propel the machine off-road, it was poorly laid out in terms of using space and was therefore going to be heavier than it needed to be. The same is true of the mechanism in the turret – heavily burdened with large gearing more akin to a heavy drive system than one for simply moving and firing machine guns.

They may be excused for some lack of knowledge when it comes to tank design, but it is hard to explain how they might see this vehicle moving across the sort of terrain they would have been familiar with. Overall, the design is crude and relatively poor, showing a great deal of naivety when it came to military vehicle design, with poor use of space and layout. The inherent problems of command and control between the driver and gunner and the excessive multitude of weapons would make any serious operation of the Gerrey vehicle problematic, to say the least, and likely one which would not be able to be operated efficiently under the stresses of combat. Indeed, only the US Army of the era might have been interested in a machine with such a preposterous number of machine guns, but regardless of the military faults, it is the tracks that are the biggest surprise.

For a pair of men who clearly have experience outdoors and almost certainly had seen or used vehicles in that terrain, it is easy to imagine the inspiration for the shape of the tracks they selected. What is less clear, however, is why the overhangs were not identified as a problem.

Nothing became of the design from the Gerreys, which is not really much of a surprise. As skilled as they may have been as timberers or as rugged outback pioneers, their ideas, whilst well-meaning, were simply too naive, too crude, and not fully thought-through – they simply were not what Australia or Britain needed in the war. This design, unofficial as it was, at least started the multitude of ideas coming from Australia for its own armored vehicle programs. Forgotten perhaps almost as soon as it was patented, the vehicle is little more than a short misstep in Australia producing what eventually became a very competent independent tank program. What became of the Gerreys is not clear, but they seem to have dropped their military ideas, submitted no more patents, and likely returned to what they knew best.

Illustration of the Gerry Machine Gun Motor Vehicle. Illustration by Yuvnashva Sharma, funded by our Patreon campaign.


British Patent GB537405, ‘Improvements in Armoured Motor Vehicles provided with Machine Guns’, filed 16th October 1939, granted 20th June 1941.
Kangaroo Valley Voice July 2008: An insight to life back then.
New South Wales Parliament, Proceedings of the Legislative council Votes and Proceedings Volume 6
The Sun (Sydney) 11th July 1931 ‘Southern Cloud Search’
Northern Times (Carnarvon) 16th July 1931 ‘Wreckage of Plane Discovered’
Sidney Morning Herald 25th October 2019 ‘From the Archives, 1958: The Southern Cloud mystery solved after 27 years

WW1 US Prototypes

Miller, DeWitt, Robinson SPG

USA USA (1916)
SPG – None built

World War One brought about numerous technical innovations to break the stalemate of static warfare which had rapidly become the defining characteristic of the war. Then, as now, it was artillery that was the key to defeating enemy defenses. The need to move large caliber guns to the front was fundamental to any army trying to achieve a breakthrough. Although the USA was not at war in 1916, this was a conflict watched keenly around the world as the fighting developed and was widely reported. Stanley Glonin̈ger Miller from St. Paul, Minnesota, a manufacturer by trade, Dorcy Olen DeWitt, also from St. Paul, who worked for the Crex Carpet Company as a machinist, and Myron Wilber Robinson, from New York City and also a manufacturer, submitted a patent application on 21st February 1916, ostensibly as an ‘Improvement in Belt-rail tractors’ for military purposes. What they actually designed was one of the world’s first tracked self-propelled guns.

The information for the design is held squarely within patent applications filed in the UK, Canada, and United States by those three men. These three men knew each other, as they all worked at the Crex Carpet Company. DeWitt was a machinist and employee, Miller was a Vice President, and Robinson was the President of the firm.

The firm itself bears some scrutiny, as it took harvested and dried wiregrass and wove it into twine and later into wicker products. The company had previously been the American Grass Twine Company which, in 1903, was rebranded as ‘Crex’, taken from the Latin name for the grass used, Carex Stricta. Woven into mats and carpets and wicker products, Crex was a profitable market-leading company for a short time and was even listed on the New York Stock Exchange in 1908.

Large factory floor space was needed to turn this dried tough grass into a workable material and machine looms would run, turning it into matting and carpet and eventually into wicker. By the time of the outbreak of the World War, the wicker industry from grass was waning. It was being replaced with wicker made from paper which had been invented in 1904 and, being cheaper to produce and easier to work with, rapidly ate away at Crex to the point where, in 1917, the firm had all but ceased to exist. Wicker was gone from its products and its decline only ended in 1935, when it finally went bankrupt.

This is relevant to the design from Miller et al. as, at the time it was drawn up, these men, who knew a thing or two about machinery and engineering processes, were looking for a new and profitable enterprise to which they could turn their energies.

Quite what inspired the design they came up with is not clear. It could well be a function of seeing tracked harvesting machines at work collecting their raw grass product. After all, this was the inspiration for Robert Macfie to look at Holt tractors in the UK in 1915 using his sugar plantation experiences.

With a war raging in Europe, it cannot have been made in isolation and, yet, the time of application for the patent is somewhat remarkable. January and February 1916, just months after the British had ordered a Top Secret new weapon into production – the tank. There was absolutely no way in which these men could possibly have known of that development so this was an advance made in isolation, a case of convergent evolution where the same solution comes about as a result of the same pressures.

The patents in question were filed in the UK on 18th February 1916, but the Canadian filing for it was even earlier, on 20th January 1916. All this was at a time when the United States was not even engaged in World War One, but in which these men could not have been unaware of one of the key problems encountered – how to get large artillery guns and other material to the front.


Traveling on what they called a ‘belt-rail’, which would be recognized today as a caterpillar-type track, the machine was to be able to traverse irregularities and undulations of the ground, soft or broken, and small obstacles to get where it needed to go. One of the key features in doing so was to keep the center of gravity for the vehicle as low as possible to reduce the chance of it overturning.


The vehicle was divided into two sections. The first took the form of the mobile tractor frame mounted on tracks and which was fitted with the engine and gearing. The second part of the vehicle was a structural framework that pivoted to the tractor frame. This part was fitted with guiding wheels that controlled the steering of the entire vehicle.

The primary frame was rectangular in shape and made from two longitudinal steel beams. Slung perpendicular between those two beams was a series of bracing beams to which the tractor units were connected.

Above the track units was a low slung platform on which the load of the vehicle sat.

Plan view of the vehicle showing three narrow track units supported by five supporting beams. The steering wheel can be seen at the front right of the track section and the large framework for the four drum-style wheels in front of that. Note that the image has been cleaned digitally to improve clarity. Source: UK Patent GB102849
Side view of the tractor showing the engine at the extreme rear, the very low-slung track units, and the four-wheel steering system at the front. Note that the image has been cleaned digitally to improve clarity. Source: UK Patent GB102849


In the patent drawings, three sets of tracks are used, but the description is clear that any number of track units could be fitted to a framework in this way. Power to those tracks was delivered via a very simple worm gear from the output shaft. This worm gear drove a large tooth gear that powered the tracks. The power for that worm gear came from an internal combustion-type engine.

Side view of the engine. It is clear that it projects over the rear of the tractor on a metal framework. The gearing from the engine and the engine is all exposed, but it is hard to imagine it would not be enclosed by some sort of cover had the vehicle even been constructed, as it would soon have fouled with dirt. Note that the image has been cleaned digitally to improve clarity. Source: UK Patent GB102849

The tracks were formed from interconnected metal links with a V-shaped grouser and were considered sufficiently different from existing tracks to warrant another patent application, submitted on the same day like that for the tractor. UK patent GB104135 for the tracks shows these interconnected thin one-piece links connected together by steel pins and using a built-in track guide in the center to hold the hold to the wheels preventing lateral movement. This is notable as, in 1916, the form of track being used was a simpler plate attached to a shoe, with the shoes being connected together and dragged around the vehicle by the drive sprocket. Early tanks, such as the British Mark I or French FT, used this shoe method. Those tanks also had separate plates which were fitted close together but did not intermesh. The design from Miller et al. wanted the edges of each link to intermesh with the preceding and following links. For a design in February 1916, seven months before tanks were even first used and entered the public imagination, this was an advanced system of track for a vehicle. It is worth noting that, although the British patent for this link was filed in February, the US patent for the tracks was filed on 10th January 1916.

The 1916 track links show a deep V shape for obtaining traction in the ground and an interconnecting link shape connected by steel pins. In the center of each link is a pronounced guide horn for keeping the track from sliding sideways on the wheels. Note that the image has been cleaned digitally to improve clarity. Source: UK Patent GB102849
From the patent for the tracks submitted at the same time as the tractor, more details of the design are clear, showing how the wheels ride on the inside of the links. This seems obvious in a 21st-century world, but in 1916, this was not the normal means by which tracks operated. Note that the image has been cleaned digitally to improve clarity. Source: UK Patent GB104135

Vertical movement of the front of the vehicle was controlled by hydraulic cylinders which served to prevent lateral movement but permit vertical movement whilst ensuring that the wheels stayed pressed onto the ground.

The similarity of this idea to the British use of wheels on the back of the Mark I tank in 1916 is very striking here. The Mark I used a system of springs to push the wheels down for the dual purpose of steering and to help raise the nose of the tank to climb obstacles. There is no mention of obstacle climbing assistance for the Miller et al. design, but the use of a system to keep the steering wheels pressed into the ground is very much the same.

On the Mark I tank, these were found to be superfluous and really a bit of a hangover from the original ideas of 1915, slaving tractors back to back, and were quickly abandoned. It is not necessarily the same situation with the Miller et al. design, as the wheels are at the front, substantially wider, and also more numerous. However, should Miller et al. have selected a second steerable track unit to be mounted in place of those wheels or a mechanism to vary drive to the tracks to provide the steering, this would have been a better steering solution for the vehicle.


No armament is specifically mentioned in the patent for the vehicle, other than to say there was sufficient space for “a gun”. The drawing, however, clearly shows a large-caliber mortar or howitzer on a mounting which appears to be shown capable of rotating on its base. Mounting a gun in this manner would have been a significant advantage for an Army of the age as, in 1916, there were no heavy guns mounted on tracked self-propelled carriages. Heavy guns, instead, had to be hauled around on old-fashioned wheeled limbers by horses, or trucks. This was a slow process which meant they were hard to move and slow to get into position on broken ground. They would then have to be set up in place to fire and could only fire from that position. If the gun had to be moved even a relatively short distance, it would have to be limbered back up, moved, dismounted, and set up all over again. This situation was even worse for large-caliber guns, which often had to be shipped in multiple pieces due to the size and weight of the elements of the gun and carriage.

With a self-propelled chassis, this was not the case and several armies, notably the Italians, placed field guns on heavy trucks to create a mobile artillery force. Whilst that system could indeed move guns around fairly quickly, what they could not do was move very well off-road and the maximum load carried was just 5 tonnes or so – limited by the strength of the truck frame and tires.

By using tracks in this design, Miller et al. would be able not only to move around on or off-road more easily but also carry a far larger (and heavier gun) if they wished to. A gun such as the British Ordnance BL 9.2” howitzer of the era weighed over 5 tonnes just for the gun alone, without including ammunition. A platform like this would have been able to mount such a gun and ammunition and the men to crew it and move it around. It might not have been fast but it would be a far quicker alternative method of moving the gun used to that point.

Seen as a whole vehicle carrying a hypothetical heavy gun of some description, the lack of wheels on the gun mount and elevation would indicate that it was envisaged that such a platform could be used to fire from Note that the image has been cleaned digitally to improve clarity. Source: UK Patent GB102849

Even if a gun was not being carried, this platform system would have been adequate for men, supplies, ammunition to be carried relatively simply, although it must be borne in mind that there is no armor and no protection from the elements for the men or load being carried.


The design from Miller, DeWitt, and Robinson was never built, it received no orders and the hopes of these men to profit from this design turned to nothing. When they submitted their design, Great Britain had already been at war since 1914 and, in 1917, the USA also joined in. Spring 1916, when they submitted this design, coincided with the British work on their new war invention, the tank, using a quite different system of track.

It would be 1917 before the British got their own tracked gun carrier, the Gun Carrier Mk. I. With a maximum payload of 7 tonnes, the Gun Carrier Mk. I allowed for heavy guns to be moved across broken ground with the added advantage of being able to load and unload field guns via a ramp at the front. No such ramp was provided for by Miller et al.’s design but it is nonetheless an advanced design and the tracks, in particular, were substantially more advanced as a design than those used on British tanks, although making them resilient enough for use is a different thing to designing them.

British Gun Carrier Mk.I. of 1917. Source: Wiki

Little can be found of the three men responsible for the vehicle, Dorcy Olen DeWitt, Myron Wilbur Robinson, and Stanley Glonin̈ger Miller. The US Census of 1910 and 1920 provides few details, but DeWitt is known to have been born on 23rd May 1880 and died on 15th June 1964. Myron Robinson, the President of the Crex Company and likely the team lead for this design, is more obscure. It is known that he was born on 11th August 1881 and was from New York but little more than that. The Crex Carpet Company went bankrupt in 1935 with just US$24.90 in the bank. The third man, Stanley Glonin̈ger Miller, is yet more obscure and all that can be confirmed about him at this time is that, in 1917, he held an associate membership of the American Society of Mechanical Engineers. The men were amateurs in that they were not military men or tracked vehicle experts, but they clearly knew about engineering and designed one of the first tracked self-propelled guns.

The vehicle would assuredly have been slow, the steering system inadequate, and the gearing system somewhat over-simplistic, but there is no denying the advanced design of the tracks and the theories being considered in mounting the gun.

The Miller, De Witt, Robinson SPG with the front wheels lowered down. This would have been one of the first SPGs in the world. Illustration by Yuvnashva Sharma, funded by our Patreon campaign.

Thank you to for supporting us in writing this article. Check out their free tank games on their website.


UK Patent GB102849 Improvement in Belt-rail Tractors. Filed 21st February 1916, granted 4th January 1917
UK Patent GB104135 Improvements in Beltrail Tractor Tracks, Filed 21st February 1916, granted 21st February 1917
Canadian Patent CA195323 Tractor. Filed 20th January 1916, granted 21rd December 1919
US Patent US1249166. Caterpillar Tractor Track. Filed 10th January 1916, granted 4th December 1917
Holmes, F. (Ed.). (1924). Who’s Who in New York City and State. Who’s Who Publications Inc. New York City, USA
The American Society of Mechanical Engineers Yearbook 1919. New York, USA.
Nelson, P. (2006). Crex: Created Out of Nothing. Ramsey County Historical Society Magazine Vol. 40 No. 4, Minnesota
United States Census 1910. Beloit Ward 3, Wisconsin Sheet A11

Cold War British Prototypes


United Kingdom (1956)
Heavy Tank Destroyer – Design only

Cerebos was a project designed by the 7th Tank Technical Officers (T.T.O.) Mechanical and Gunnery AFV design exercise held at the British Royal Armoured Corp (R.A.C.) School of Tank Technology (S.T.T.) in 1956. In the study, the designers were tasked with coming up with a heavy tank destroyer using guided anti-tank missiles as its primary offensive weapon. It had to be able to operate on the front lines of a European conflict, have relative immunity from Soviet guns at combat ranges, and a very high chance of scoring a direct hit and killing any Soviet vehicle of the day.


The concept of a heavy and super-heavy missile vehicle had already been on the minds of British AFV designers for a few years during the early part of the Cold War. The Anti-Tank Guided Missile (A.T.G.M.) was a relatively new technology in an era when tank guns were still relying on ranging machine guns for calculating the distance to the target. The ability to effectively engage a tank at twice the effective range of such a gun and to effectively track and guide the missile to the target was highly desirable. This fact, combined with the huge leaps in armor penetration capabilities from shaped-charge (SC) technologies used in High Explosive Anti-Tank (HEAT) type warheads, especially compared to ‘conventional’ anti-tank ammunition of the period, made many think the era of the conventional armored tank was over. This was simply because, using conventional armor technologies, no tank could hope to survive against HEAT warheads such as the French SS.10, Soviet AT-1 Snapper, and later the Mosquito or Swedish Bantam. In order to stop such weapons, steel armor would need to have been over 500 mm thick, which in turn would have led to impractical machines. One result of this technological shift away from conventional armor was a generation of very lightly armored main battle tanks like the German Leopard. Whilst this shift was recognised early in Western nations, despite projects like the British Conqueror and some American heavy tank/tank destroyer projects, it took longer to be recognised in the Soviet Union, at least in the eyes of the West. Tanks like the IS-3 and T-10 loomed large in the imagination and nightmares of Western planners along with some incorrect assessments of the armor of a new generation of Soviet medium tanks. This meant that new means of countering this Soviet armor were needed.

The debate over the end of the tank has been waged since almost the very beginning of the weapon. For each new anti-tank weapon, a new defense innovation was found and, conversely, for each new step-up in armor, a new weapon to defeat this armor was found. In this way, to a broad extent, the evolution of anti-tank weapons very much reflected the evolution of tank armor. Within this context, there are few evolutionary leaps that were as profound in tank terms as this first decade or so after the end of WW2. The A.T.G.M. had gotten to the point where it was closer to forcing the tank into obscurity than ever before and, were it not for the vast fleets of tanks in Soviet service that remained an active threat forcing NATO to maintain its own significant fleet of tanks, armored warfare may have taken a very different route.

In the meantime, all nations were still churning out regular tanks expected to fight other tanks and so, much like the Second World War, tank destroyers were still being developed and built with the sole aim of breaking up enemy tank formations at long range. For the British, the appearance of heavy Soviet armor and the prospect of large enemy armored formations posed a particular threat. Many of those vehicles were virtually immune to the UK’s best tank-guns then in service and in such large numbers that even if they could match Soviet armor with British firepower they could still be overwhelmed.

There was little the British could do to counter the enormous numerical advantage of the Soviet forces in Europe but there was something which could be done about the guns and this fed into the motivation behind the development of the Royal Ordnance L7 105 mm rifled gun and eventually the L1 120 mm rifled gun too. Despite some heavy Anti-Tank concepts in the UK, the 7th T.T.O. Course opted instead for an A.T.G.M.-based Anti-Tank platform over a gun-based solution. The weaponry for this option consisted of a version of the Malkara missile, and this, it was felt, would provide the offensive power required to counter the Soviet threat. It also provided the additional benefit that the avoidance of a turret allowed all available protection to be focussed on the hull instead and all for less weight than a conventionally armed and armored gun-tank.


This was the context and logic behind the Cerebos, a turretless guided-missile tank destroyer with heavy armor. It was intended to operate on the front lines, have enough protection to withstand strikes from enemy tanks using conventional guns, and ideally use the chassis of a vehicle already in service as a platform. It was desired to have a missile able to destroy the heaviest Soviet vehicles then known in service or considered to potentially enter service. An ideal rate of fire of four rounds per minute was requested, with a minimum of two rounds per minute, with two missiles ready to fire at any time.

The model of Cerebos shows a well-shaped front which was heavily armored and the vertical launching pods for the missiles. The missile shown on the ‘stick’ is merely illustrative as they were launched internally and vertically not from this elevated position.


The goal was to reuse, as far as possible, the hull of an existing vehicle and Cerebos did just that and was based around a heavily modified Centurion tank. This meant a high degree of commonality of parts between Cerebos and the standard battle tank of the British Army of the day, which would reduce the logistical burden of the vehicle. The modifications, though, were extensive. Instead of the sloped glacis of the Centurion, Cerebos used a steeply angled ‘pike’ type nose, similar in style to that on the Soviet IS-3 tank. The driver sat along the centreline of the tank with a forward observation window cut directly out of the armor. The commander sat directly behind him, and the loader sat even further back on a swivel chair that allowed him the freedom of movement to assemble the missiles.

The missile bin had to be as equally protected as the vehicle itself and yet maintain a potential 360° arc of fire. This was somewhat problematic, as adding a conventional missile rack on the top of the vehicle would add not only excessive weight but would also result in a large and conspicuous target that would be vulnerable to small arms fire, shell splinters, etc. It would also be heavy, requiring dedicated hydraulics just to operate. To overcome these issues, the designers had the missile bins located inside the hull of the vehicle in a vertical arrangement, with 5 additional missiles stowed vertically running alongside the left and right sides of the inner hull. On firing the missile, the silo roof would fold open in two triangular parts. The weapon was then fired and guided on to its target by the commander. Once the missile was away, a new one was selected and attached to what amounts to a ‘potter’s wheel’ type base. This base rotated 360 degrees in the missile chamber, with the four fins being added from a separate supply located in front of each missile. This might seem odd as an idea, but the fins were the part of the missile which increased their storage volume and this semi-assembly of the missile attaching the fins meant that a larger number of missiles could be stowed inside the tank.


Cerebos was based on the Centurion but it was better protected from enemy fire than the Centurion. Sporting heavy frontal armor with a glacis plate 120 mm thick angled back at 65° and a lower front plate 120 mm thick angled at 55°, the Cerebos was felt to be well-enough protected to be able to take any reasonable enemy fire which might be forthcoming from the Soviet tanks of the day. In more conventional UK armor terms, the sides were still quite weak though, with just 25 mm on the upper sides (at 8°) tapering to 20 mm (at 10°) on the lower hull sides. The roof and rear were 25 mm thick, just enough for protection from small arms fire and shell bursts. The belly plate, just 20 mm thick, was sufficient to provide some protection from landmines but the focus of armor was on the front, facing the enemy, making the best use of the weight allowance available for maximum effect.


Power for Cerebos was provided by a 9-liter Jaguar 90° V8 petrol engine delivering 350 b.h.p. at 3,750 rpm connected via a Merritt Brown 6-speed (4 forward and 2 reverse) gearbox. Drive was delivered, just like the Centurion – to the rear sprockets. This engine was expected to permit the 21-ton (21.3 tonnes) Cerebos to achieve a top speed of 28 mph (45 km/h) and operate for a maximum range of 220 km at 14 mph (22.5 km/h).


The primary armament proposed for Cerebos was a Manual Command to Line-Of-Sight (M.C.L.O.S.) type anti-tank missile that looked somewhat like a slightly smaller and sleeker Malkara missile, measuring 5 ft. (1.5 m) long and 10 inches (254 mm) in diameter. Unlike the High Explosive Squash Head (H.E.S.H.) warhead on the Malkara, this 20 lb. (9 kg) warhead was a shaped charge High Explosive Anti-Tank (HEAT) type. The total missile weight was expected to be 85 lb (38.5 kg) and these would be launched vertically from within the missile tube. Once assembled with its fins, it was ready for launching and this could be done whilst a missile was already underway as the targeting was being carried out by the commander with missile assembly taking place independently.

A pair of launchers and 12 missiles (two already assembled and ready to fire, with another ten stowed) could be carried. Although no performance data for these missiles was given, it can be estimated from the diameter of the warhead and the performance of contemporary missiles to achieve a penetration of approximately five times its diameter, which would equal about 750 mm of armor plate – more than sufficient to defeat any known Soviet tank in service at the time.

The maximum range for the missile was just as impressive as the anti-armor performance expected – far exceeding the range available from a conventional tank gun. Cerebos was to be able to engage targets at ranges of up to 6,000 yards (5.4 km), although the missiles did have a minimum safe distance as well – 500 yards (460 meters). With a flight-speed of 350 feet per second (107 m/s), the missiles had a potential maximum flight time of about 50 seconds. For ease of stowage, the missiles were kept without their fins. The gunner would have to assemble the bare missile, attach the fins individually by means of the snap-on fasteners and then load a missile into the missile bin. This whole process was estimated to take not more than 2 minutes per missile. This would mean (assuming two were already loaded) that up to 4 missiles could be fired in a 4-minute window.

Secondary armament for Cerebos was primarily for self-defense and consisted of a single Browning .30 caliber (7.62 mm) machine gun remotely operated from within the hull with a 360° degree arc of fire and provided with 4,250 rounds of ammunition. Six No.36 smoke dischargers were provided, with 3 per-side, and the crew was provided with grenades and small arms.

Side view of Cerebos showing the 5 unassembled fin-less missiles on one side of the compartment. Another 5 were along the right-hand side as well. With two in the bay ready to go, Cerebos had 12 missiles. Note the curvature at the front is unintentional and merely a result of the curvature of the original paper on which the plan was printed and bound.


For its time and era, the wings being clipped on was nothing new and this type of missile-build-before-launch concept was also to be added into the FV4010 heavy missile vehicle, as the later fold out missiles and overall lighter materials were still some years away. Two flaws not raised in the original documentation but more observable with hindsight are the lack of a telescopic mast or periscope allowing firing from the reverse side of slopes and the poorly placed second cupola that had much of its view blocked by being located behind the first. Other issues are the commander acting as the missile gunner, guiding it to its target, placing undue stress, and preventing him from monitoring the battlefield. The Cerebos was no more than a design project and never built, however, many of the ideas and features later appeared on the Malkara launching FV4010.

The Cerebos heavy missile tank destroyer. The vehicle is in travel mode, with the missiles safely within the hull. Notice the very heavily sloped pike-shaped front. Illustration by Yuvnashva Sharma, funded by our Patreon campaign.

Bovington Tank Museum Archives, STT section, Cerebos box

Cerebos Specifications

Dimensions (L-W-h) 21ft 5.5 inches x 9ft 10 inches x 8ft 4 inches (6.53 x 3.00 x 2.54 m)
Crew 3 (commander/gunner, driver, loader)
Propulsion Jaguar 9 liter 90° V8, 350 bhp
Speed (road) 28 mph (45 km/h)
Ground Clearance 17 inches (0.43 m)
Track Center Distance 8 ft. 4 inches (2.54 m)
Length of Track on Ground 14 ft. 7 inches (4.45 m)
Normal ground pressure 8.4 psi (57.92 kPa)
L/C ratio 1.75
Vertical obstacle crossed 3ft 8 inches (1.12 m)
Gap crossed 7ft (2.13 m)
Armament Manual Command to Line-Of-Sight (MCLOS) ATGM
0.5/12.7 mm HMG.
Maximum, Minimum Missile Range 6000 yards/5.4 km, 500 yards/457 meters
Missile Velocity 350 fps (107 m/s)
Ammunition 12 High Explosive Anti Tank Missiles
Armor Front: 120 mm @ 65 degrees
Sides: 25-20 mm
Rear 25 mm
Bottom 20 mm