586 ANNUAL KECOPwD OF SCIENCE AND INDUSTRY. 



among themselves in the rate with which they cause the 

 generation of heat on being exposed to air upon the surface 

 of fabrics, shavings, or other materials; but all are more or 

 less liable to this result when spread out in thin films, or in 

 any other minute state of division. The so-called " drying 

 oils" are particularly susceptible to such atmospheric influ- 

 ences ; the dryhig itself consisting in the conversion of the 

 oil into a kind of rosin by the action of the air. A greater 

 or less amount of heat is evolved by the process, which, under 

 favorable circumstances, may produce spontaneous combus- 

 tion. Such "drying" is not similar to that where cloth, re- 

 moved from water, is said to dry; since, in this case, it is 

 merely the evaporation of the liquid portion, instead of the 

 conversion of it into a different substance, by the absorption 

 of oxygen. The emission of heat from a painted surface is 

 not generally noticed, since it is comparatively slight in quan- 

 tity, and is carried away by the atmosphere as fast as form- 

 ed ; but accumulated in any way it becomes very evident. 

 Thus a heap of old greasy rags soon becomes so hot in the 

 central part that the temperature can not be borne by the 

 hand. 



A number of experiments were made by the use of paper, 

 cotton, and wool slightly impregnated with different kinds 

 of oil and exposed to the air under similar conditions ; and, as 

 a result, it was found that the temperature rises from 25 to 

 200, according to the variety of the oil, the amount of sur- 

 face in contact with the air, and the duration of the exposure. 

 In one experiment with an initial temperature of 80, the ma- 

 terial of the mass soon rose to 275, and in another, where it 

 Avas 70, it rose to 223. The escape of heat in these experi- 

 ments was purposely prevented, a condition that might ensue 

 naturally, by the compression of pliable substances, or cov- 

 ering them with some non-conductor. Even in cases where 

 an actual spontaneous ignition did not take j^lace, the pres- 

 ence of a highly volatile, inflammable oily vapor was devel- 

 oped, which at once ignited at a considerable distance from 

 the oiled substances on the approach of a flame. 



Great care, therefore, is urged in reference to the accumu- 

 lation of any oiled substances (either refuse or manufactured 

 articles), and every precaution should be taken Avhere the 

 VOoms in which such articles are contained are exposed, in 



