Mechanical disintegration seems to be the most effective way of damaging the cells in potato tubers so that their juice can be expressed. If disintegration is too intense the pulp is intractable; a grater makes a uniform pulp with suitable texture. With existing continuous equipment it is necessary to press in two stages to get to 40% dry matter and further batch pressing is needed to get to 50%, but experience suggests it should be possible to make a press that would press to > 50% in one operation. Much less energy is needed to press out water rather than evaporate it. but about one-sixth of the tuber dry matter and more than one-half of the N is in the juice. Surplus, diseased, or damaged potatoes can be preserved for use as stockfeed by ensiling or drying. With the help of fungistatic agents such as ammonia or propionic acid, and if permanent preservation is not sought, it is not necessary to dry the tubers completely, and partial drying would cost less. We have therefore studied the practicability of pressing out fluid to the necessary extent instead of evaporating it off. Only small amounts of fluid can be pressed from sliced potatoes. The obvious methods for destroying the capacity of the tuber cells to retain fluid are: heating, chemical poisoning, freezing, and mechanical damage. Heating swells the starch so much that little water can be pressed from the cooked mass and chemical treatments were not studied because problems were envisaged in removing the agents from the products after pressing.Freezing is the traditional first step in making chuno on the altiplano of South America, where tubers are frozen in the open air at night and trodden during the day after thawing. After freezing at -i5°C and thawing, as much juice can be pressed from sliced tubers as can be pressed from mechanically disintegrated tubers, the thawed mass is sufficiently coherent to make pressing easy, and the juice removes less protein than juice made after mechanical disintegration. However, in spite of these real advantages, freezing does not seem to be as practical on a large scale as pulping because of the amount of energy consumed in freezing, which would be only partially recoverable.As tubers are damaged to increasing extents, the amount of juice that can be pressed out under the same conditions increases, but the pulp becomes an increasingly intractable paste that can be retained in a press only by wrapping it in filter cloth or in some similar way, which would probably not be practical on a large scale. The problem can therefore be formulated: to find a degree of damage that releases sufficient juice without producing pulp that is uncontrollable in a press. Experience gained in extracting protein from leaves is of limited assistance because the fibre in a leaf pulp gives it a texture that simplifies subsequent handling. If, as is probable, the conserved material is to be used as cattle fodder, the problem could be side-stepped by mixing some fibrous ruminant fodder, e.g.