Autochrome was one of the very first practical method of colour photography (the similar Jougla Omnicolore was introduced a few months previously) and was first marketed by the Lumière brothers in 1907, becoming widely used before the advent of subtractive colour film in the 1930s.
Autochrome used an additive method, with a glass plate coated in microscopic grains of potato starch that were dyed red-orange, green, or blue-violet, underneath a panchromatic photographic emulsion. The plates needed longer exposures than contemporary black and white photography (about thirty times more). Although there was no need for any special camera, there was a need for a yellow filter to correct the over-sensitivity to blue light, and a tripod was needed for the long exposures.
The resulting positive image was quite dark and needed bright light for viewing. Autochrome plates could be viewed with a special stand called a diascope, that offered a brighter image, or they could be projected with a magic lantern but they needed a very powerful light. Unfortunately, Autochrome images on glass plates were difficult for artistic photographers to exhibit and so most autochromes were taken by amateur photographers.
The Autochrome process produces images that are considered very attractive and dreamlike. Stereoscopic Autochromes were available, and combined both colour and depth.
Despite being much more expensive that monochrome plates, the Lumière factory was making 6,000 autochrome plates a day by 1913, in a range of different sizes.
From 1932, film versions of Autochrome became available, but faced competition from new subtractive colour films such as Kodachrome and Agfacolour Neu. Autochrome film was finally discontinued in 1955.