Spontaneous Combustion on Weed Plats Sprayed with a Solution of Atlacide

Chlorates, especially sodium and calcium, in recent years have been used extensively as chemical aids in weed control. With them has also come an increased fire hazard. Not infrequently does one read of accidents where someone has been seriously injured as a result of his clothing becoming ignited after spraying weeds with chlorates. The Utah Agricultural Experiment Station, in cooperation with the State Department of Agriculture, has made extensive studies since 1929 on the possible use of chlorates in weed control. Since 1929 a large amount of both sodium and calcium chlorates have been used with little difficulty so far as fires have been concerned. However, on September 22, 1932, a fire, which might have been serious, broke out as a result of their use. This fire was purely a case of spontaneous combustion. Because of the possible seriousness of the problem, it appears desirable to submit a brief note giving an account of the incident. During the forenoon of September 22, 1932, three plats (^3 square rod) of Canada thistle (C. arvense L.) were sprayed with a solution of Atlacide. The solution was made by adding 2 pounds of Atlacide to each gallon of water. This solution was applied at the rate of i gallon to the square rod. The chemical was applied between 10 and ii a.m. Since it was an experiment which had been previously planned for three replications of each treatment, three separate plats were sprayed. These were treated within a period of 15 or 20 minutes of each other. At about 12:30 p.m. a fire started on one of the plats, about 3 minutes later a fire started on the second plat; and about 20 minutes later on the third. In each case fire seemed to occur in only one localized area on the plat from which it spread until it was extinguished. These weed plats were located in a bluegrass (P. pratensis) pasture, and at the time the chemical was applied there was a rather heavy undergrowth of dry bluegrass. The fire started in this dry undergrowth. Two of these burned areas are indicated in Fig. i. Plats apparently burned in the order of dryness of the undergrowth rather than in the order of their treatment. Since there was a sudden change in temperature and humidity just prior to the time the burning began, it was thought that this