On August 2, 1917, Norwegian engineer Nicolay Fougner launched the Namsenfjord, an 84-foot, 400-ton vessel that became the first self-propelled concrete ship designed for ocean travel. The U.S. government immediately invited Fougner to study the feasibility of building concrete ships in American shipyards. The logic was simple: steel was going into tanks, guns, and conventional warships faster than foundries could produce it. Concrete required only cement, sand, gravel, and steel reinforcing bars - materials available in quantity. The tradeoffs were equally simple. Concrete hulls needed to be thick, which meant either a bigger cross-section that hurt hydrodynamics or less interior space for cargo. Labour costs were high. Operating costs were higher. But when you need ships and have no steel, you pour what you have.

Between the wars, concrete shipbuilding lapsed into obscurity. Steel was plentiful, conventional methods were cheaper, and nobody particularly wanted to operate a ship that traded cargo capacity for hull thickness. Then came 1942 and another steel shortage. The U.S. government contracted McCloskey and Company of Philadelphia to build 24 self-propelled concrete ships at a shipyard at Hookers Point in Tampa, Florida, that employed 6,000 workers at its peak. Additional contracts went to California firms for concrete barges. In Britain, 200 concrete barges were commissioned to carry petrol, constructed on the London dockside and lowered into the Thames by crane. Few of the World War I ships had seen action in time, but the World War II fleet served - concrete barges supported American and British invasions in Europe and across the Pacific during 1944 and 1945.

The largest surviving collection of concrete ships sits at Powell River, British Columbia, where ten ferrocement vessels float as a breakwater known simply as the Hulks. A lumber mill has used them for decades to shield its log booming grounds from the swells of the Strait of Georgia. Elsewhere on the continent, nine sunken concrete ships form the Kiptopeke Breakwater in Chesapeake Bay, Virginia. Off Sunset Beach near Cape May, New Jersey, the wreckage of a vessel commissioned in 1919 and sunk in 1926 remains visible from shore. The SS McKittrick, launched in 1921, became a gambling ship off Coronado, California, before running aground on New Year's Eve 1936 - its wreck periodically reappears during strong storms. On Shipwreck Beach in Hawaii, a concrete gasoline barge built for the Navy in 1942 is frequently misidentified as a Liberty ship. Each wreck is a monument to the same improvisation: you build with what you have.

The phenomenon was not confined to North America. During the German occupation of Greece, the Wehrmacht built 24 concrete cargo vessels at the Perama shipbuilding area in Piraeus for delivering supplies to the Greek islands. After the war, many were repurposed as piers and breakwaters. Hitler ordered 50 more for oil transportation from Romania and the Baltic front, including specialized hospital ships that evacuated wounded soldiers along rivers. Japan built four concrete ships during the war, two of which still serve as a breakwater in Kure, Hiroshima. On Iwo Jima, American forces grounded several concrete ships on the west beach to create a breakwater, most of which were later destroyed by typhoons. In Britain, the Purton Hulks - a collection of vessels beached to prevent coastal erosion - includes eight ferro-concrete barges. From France to Florida, from Piraeus to Powell River, concrete ships endure in ways their builders never expected.