Beijerinck's contagium vivum fluidum

At the agricultural school in Wageningen, Adolf Mayer showed in the mid-1880s that tobacco mosaic, or leaf-spot, disease was contagious: sap pressed from sick plants and pushed into healthy ones in capillary tubes gave them the disease within weeks. His colleague Martinus Beijerinck, born in Amsterdam in 1851, watched the experiments, and neither man could find a microbe to blame. Beijerinck, who later taught at Delft, took up the disease again in 1887, still without result.

In the work he published in 1898, Beijerinck passed juice from diseased leaves through porcelain filters of the Pasteur-Chamberland type, which hold back bacteria, and the clear filtrate still infected healthy plants. A small drop injected with a syringe could infect many leaves and branches, and sap from those could infect an unlimited number of new plants, so the agent reproduced itself in the living plant. Because filter experiments could still be doubted, he also laid infected juice on thick agar plates. The infectious substance diffused well into the agar, leaving bacteria and other particles on the surface.

Only growing organs whose cells were still dividing could be infected, he found, and only there did the agent multiply. No multiplication could be shown outside the plant, although a bacteria-free filtrate kept its virulence for more than three months, and heating the filtrate to 90°C destroyed it. In his report to the Royal Academy of Sciences in Amsterdam he concluded that the disease was not caused by microbes but by a contagium vivum fluidum, a living contagious fluid, which he called a virus.

Others read similar evidence differently. Dmitri Ivanovsky in Russia had found in 1892 that the tobacco mosaic agent passed a filter, but he kept looking for a microbe and in 1903 claimed to have grown it in an artificial medium. Also in 1898, Loeffler and Frosch, former students of Koch, filtered the agent of foot-and-mouth disease of cattle and took it for a very small microbe. Beijerinck died in 1931 with his idea still unverified. Tobacco mosaic virus particles were isolated in 1935 and characterized as a nucleoprotein in 1937.