A naval tank designed to swim into battle
The Type 2 Ka-Mi was a purpose-built Japanese amphibious tank rather than a standard light tank fitted with improvised flotation. Its sealed fighting hull, detachable bow and stern pontoons, propellers, rudders, rubberized openings, bilge arrangements, and internal release controls formed an integrated water-to-land system. The vehicle was intended to deliver armored firepower across reefs, lagoons, beaches, and small harbors where conventional tanks depended on suitable landing craft or prepared ports. Sources
This was primarily a naval requirement. Special Naval Landing Forces needed organic armor for offensive island operations, and Mitsubishi is generally identified with development and production. Allied technical intelligence later documented the vehicle in detail, showing that the finished machine involved much more than waterproofing a Type 95 Ha-Go. Sources
The tactical idea was coherent. Ka-Mi could approach afloat with pontoons attached, reach the beach, release flotation sections from inside the hull, and continue inland as a tracked light tank. A 37 mm gun and machine guns could support landing troops against infantry, light positions, and soft vehicles. Five crewmen included an engineer whose work reflected the demands of engines, water propulsion, sealing, and bilge control. Sources
The strategic problem was timing. As Ka-Mi became available, Japan moved from offensive amphibious expansion toward defensive island warfare. In static defense or local counterattack, the vehicle lost much of the advantage that justified its complexity. Thin armor and a light gun then faced Allied tanks, anti-tank weapons, artillery, naval fire, and air power on increasingly unfavorable terms. Sources
Pontoons, propulsion, crew, and the cost of amphibious specialization
The Ka-Mi’s hull and appendages were designed around buoyancy and transition to land. Bow and stern pontoons were divided into sections and held by release mechanisms operated from inside the tank. Twin propellers and rudders provided water movement and steering; seals protected openings; a bilge pump addressed leakage; and some configurations included additional air or spray-control fittings. Once ashore, the crew could discard the pontoons rather than fight with the full flotation structure attached. Sources
A five-person crew was comparatively generous for a Japanese light tank but reflected the mission. The commander and gunner occupied the turret, the driver and hull machine-gunner worked forward, and an engineer handled machinery associated with amphibious operation. The turret carried the Type 1 37 mm tank gun with a coaxial Type 97 7.7 mm machine gun, while another machine gun sat in the hull. U.S. technical intelligence recorded 132 rounds of 37 mm ammunition for the examined configuration. Sources
Protection remained light, generally in the range needed against small arms and fragments rather than modern tank and anti-tank guns. The six-cylinder air-cooled diesel and four-forward-speed transmission supported land movement, while the transfer system also interacted with water propulsion and pumping. Wartime technical figures place road speed around 23 mph and water speed around 6 mph for the examined vehicle, but operational conditions, load, sea state, and individual condition would have changed actual performance. Sources
Amphibious capability added maintenance burdens. Seals, pontoons, clamps, propeller shafts, rudders, corrosion, and bilge equipment all demanded attention beyond that of a conventional light tank. The design was sophisticated in mission terms, but sophistication did not remove the basic compromise: armor and armament had to remain limited enough for the complete system to float. Sources
Production, operators, and the boundary with other naval amphibians
The safest production statement remains approximately 180–184 Type 2 Ka-Mi tanks. Secondary summaries often choose 182, 184, or a rounded number near 200, while the detailed wartime technical manuals do not settle the total. The project therefore preserves a range with uncertainty and leaves exact annual and factory acceptance figures open. Sources
Mitsubishi is usually credited as principal developer and manufacturer, but the complete industrial chain is not documented in the cited sources. The division of manufacture among plants and subcontractors for hulls, pontoons, propulsion equipment, engines, and seals remains uncertain. Production complexity is a reasonable inference from the documented hardware, not a substitute for labor-hour or cost records. Sources
Japan was the sole operational user. The cited accounts associates Ka-Mi with Special Naval Landing Forces from 1942 to 1945. Captured tanks photographed or tested by Australian and American personnel were intelligence subjects, trophies, or museum artifacts; they do not create foreign operators. Unit-by-unit allocation remains one of the most important gaps because campaign-level references often name an island without identifying the receiving naval formation or quantity. Sources
Related Japanese amphibious vehicles require strict boundaries. The larger Type 3 Ka-Chi was another amphibious tank; Type 4 Ka-Tsu was principally an amphibious cargo or assault-support carrier; Type 5 To-Ku remained a projected heavy amphibious concept. They belong in the wider Japanese naval amphibious story but does not add to Ka-Mi production or be described as simple Ka-Mi marks. Sources
From intended landings to captured defensive equipment
The intended combat sequence was launch or swim, approach shore, land, release pontoons, and fight inland. In practice, Japan’s loss of maritime and air initiative often left Ka-Mi tanks as garrison equipment or conventional light tanks. Allied photographs from Rabaul show both waterborne trials and rows of captured vehicles without flotation pods, visually demonstrating the difference between the vehicle’s intended operating cycle and its late-war fate. Sources
Ka-Mi could be dangerous to infantry, unarmored positions, and light vehicles, but its 37 mm gun and thin protection were inadequate for sustained combat against late-war Allied medium tanks and combined-arms formations. Bazookas, anti-tank guns, artillery, tank guns, naval fire, mines, and close assault all threatened it ashore. Amphibious sophistication did not make the tank a heavy breakthrough weapon. Sources
Campaign associations in the secondary literature is described as leads until unit records are found. Photographic evidence supports captured vehicles at Rabaul, and other institutional material places Japanese amphibious armor in the Philippines. It does not by itself establish the first combat use, the number present in every campaign, or detailed loss causes. Sources
The fairest comparison is with other amphibious armored systems. Ka-Mi was more tank-like and compact than many tracked landing vehicles, while American LVT families benefited from far larger production and an integrated amphibious logistics system. Sherman DD tanks carried medium-tank firepower and armor but used a different flotation concept. Ka-Mi remains an ingenious specialist design whose battlefield opportunity narrowed as the war turned against the navy that created it. Sources
Primary technical strength, operational gaps, and image status
The source base is strongest for physical design because wartime Allied technical intelligence describes the hull, pontoons, releases, water propulsion, crew, engine, armament, ammunition, armor, speed, and range of an examined vehicle. It is weaker for exact production totals, Mitsubishi subcontracting, naval unit distribution, first combat use, and losses. Those gaps remains unresolved rather than filled from repetitive secondary summaries. Sources
The five images document three different evidence categories. Australian War Memorial photographs show a captured Ka-Mi coming ashore after post-surrender trials, another afloat during those trials, and numerous captured vehicles near Rabaul without flotation pods. A U.S.-government-associated photograph shows a vehicle with flotation sections under testing, while a modern licensed photograph shows a museum survivor with pontoons. None of the photographs establishes an offensive combat landing. Sources
Item-level AWM pages identify accession, date, location, subject, and public-domain status for the three Australian images. Some photographic identifications remain uncertain. Sources
Designation details, subcontracting, unit movements, campaign service, and the complete survivor count remain unresolved in the cited evidence. Sources
Photographs and image credits
Background reading and references. Each photograph has its own source and credit.
.jpg){kind=link}
Sources and further reading
Background reading and references. Each photograph has its own source and credit.
- MID 1945 — Military Intelligence Division, War Department. Japanese Tank and Antitank Warfare. Special Series No. 34, 1 August 1945.
- TM-E 30-480 1944 — U.S. War Department. Handbook on Japanese Military Forces, Technical Manual TM-E 30-480, 1 October 1944, Chapter IX, Weapons, Section IV: Tanks and Armored Cars.
- AWM 099057 — Australian War Memorial collection record 099057.
- AWM 099055 — Australian War Memorial collection record 099055.
- AWM 096634 — Australian War Memorial collection record 096634.
- HistoryNet — Paraag Shukla. “Japan’s Type 2 Ka-Mi Amphibious Tank.” HistoryNet.
- National WWII Museum 2008.354.445 — The National WWII Museum Digital Collections, object 2008.354.445.
- Wikimedia Ka-Mi float — Wikimedia Commons, File: Ka-Mi float.JPG.
- Wikimedia Kubinka / Alan Wilson — Wikimedia Commons, File: Type 2 Ka-Mi - Patriot Museum, Kubinka (38218661842).jpg; photograph by Alan Wilson.
