﻿Intelligence and Miscellaneous Articles, 479 



after explosion. The usual precautions were taken to avoid the 

 influence of the surrounding temperature, and to measure the quan- 

 tity of heat produced in the vessel *. The observations, properly 

 calculated, have shown that the powder had always a temperature 

 higher by o, 18 to 0°'45 than that of the drops before the explo- 

 sion. By this method it was impossible to avoid a very violent 

 agitation of the liquid, produced by the contact of the bursting tears ; 

 the results thus obtained are therefore liable to important errors. 



In a third series of experiments the drops were placed in a sort 

 of truncated cone of cardboard, the axis of which was almost verti- 

 cal ; the base of the cone was uppermost, and was closed by a sheet 

 of caoutchouc. The points just projected through this sheet far 

 enough to be seized with a pair of nippers and broken. The vitre- 

 ous powder fell into a very thin cylinder of brass, suitably placed 

 below the cone, and containing some grammes of oil of turpentine. 

 The temperature of the drops before bursting was observed, as well 

 as that of the turpentine before and after the fall of the powdered 

 glass. The apparatus was protected, of course, against the effects 

 of the surrounding temperature ; and the observations were calcu- 

 lated so as to furnish the difference between the temperature of the 

 fragments of glass and that of the drops which produced them. All 

 the results indicate an increase in the temperature of the glass on its 

 bursting ; the results, which are in better agreement than those of 

 the preceding series, are comprised between o, 26 and o, 35. 

 Control experiments were made with the view of proving that the in- 

 crease of temperature could not be due either to the mere fall of the 

 glass into the oil, or to some capillary action between the liquid and 

 the powder of the solid. 



All the experiments made authorize the following conclusions : — 



(1) At the time of the explosion of the drops the glass powder 

 which is produced has a higher temperature than that of the drops 

 themselves at the time of the explosion. 



(2) A series of six experiments (third method) made with eighteen 

 drops having a mean weight of 4*8 grms. indicated a mean excess 

 of 0°-30. 



This heating probably derives its origin from the molecular 

 motion which accompanies an explosion ; and 1 think it can be 

 brought into connexion with the phenomena produced when metallic 

 wires, having been elongated (without exceeding the limit of elasti- 

 city), suddenly resume their original length. Mr. Joule has shown j* 

 that wires become cooler when drawn out, and then heat again when 

 they resume their original volume. In a recent memoir % M. Edlund 

 has published similar results, obtained by drawing out wires of dif- 

 ferent metals, and then allowing them to cool. 



* The details of these experiments will be found in a memoir in an early 

 Number of the Archives des Sciences Physiques et Naturelles de Geneve. 



t Phil. Trans, for 1858. Phil. Mag. vol. xv. p. 538. 



j Ann. de Chim. et de Phys. vol. lxiv., and Pogg. Ann. 1865. Sir W. 

 Thomson has given a formula, deduced from the mechanical theory of heat, 

 by which these variations of temperature may be calculated. 



