The writer reviews the subject of weed control by chemicals in the growing crop mainly with reference to prairie conditions, and discusses it under the following heads: the herbicide, the sprayer, weather conditions, the crop, the experimental technique. A technique is described which permits the analysis of a large group of variable factors in the weed-control experiments carried on in Saskatchewan.An extensive program of weed research was undertaken in Saskatchewan in 1929, to determine the practicability of both methods of control by chemicals (control of perennial weeds in small patches, and control of annual weeds in the crop) under semi-arid conditions and with extensive farming methods. The experiments were conducted under the most variable conditions of soil and climate, and many chemical solutions were tested at varying concentrations and amounts of spray per acre. Treatments were made on stinkweed and wild mustard in wheat, oats, barley, rye, corn, sweet clover, western rye grass, alfalfa and brome grass. During the last three years the effects of the following chemicals have been studied: sodium chlorate, sodium dichromate, copper nitrate, copper sulphate, sulphuric acid, iron sulphate, sodium chloride and ammonium bisulphate. All of these, except the last, reduced the weed infestation of the treated crops; the chemicals are listed in order of effectiveness. The results showed, however, that many factors other than proper strength of solution and amount of spray per acre should be taken into account.The largest percentage of control of wild mustard in the crop of 1931 was obtained with weak solutions of sodium chlorate. The effects of this chemical on the crop are more severe than those of the others, and therefore its use should be further investigated before it can be recommended.The application of dry chemicals for control of weeds in the crop combine in general two farming operations—fertilizing of the soil and weed control. These methods offer little promise for Saskatchewan as they require cool, foggy, wet weather which is rarely found in the west; and because it has been found that, in order to be effective here, commercial fertilizers have to be drilled in with the seed and not broadcasted on the surface.It appears that, under Saskatchewan conditions, only a few weed species can be controlled successfully. In wheat, sulphuric acid will control common wild mustard, Indian mustard, wild radish, stinkweed, false flax, tumbling mustard and wild buckwheat. New shoots of Canada thistle are somewhat retarded in their development. Hare's ear mustard, ball mustard, cow cockle, lamb's-quarters, Russian pigweed, and the grasses, wild oats, darnel, etc., could not be controlled under the conditions of these experiments. It is thus important that chemical treatments be limited to fields where weed infestation consists mostly of susceptible weeds, because in controlling these the crop and resistant weeds are stimulated. At Watson, Sask., in 1931, when wild mustard or stinkweed was associated with either wild oats, lamb's-quarters or Russian pigweed, there was always an increase in the development of the latter weeds after treatment.In these experiments the cereals ranked as follows in resistance to sulphuric acid sprays:–resistant; wheat, oats: intermediate; barley, spring rye: susceptible; corn.