EXPLOSIVENESS OF NITRE. 9 



facts and inferences ; and moreover, I endeavored to illustrate the subject by refer- 

 ring to the explosion so frequently produced by blacksmiths, through the forcible con- 

 tact with moisture, of incandescent iron struck by a hammer. It has been ascertained 

 that globules of oxide of iron, as they fall in a state of fusion from a vs^ire ignited in 

 oxygen, do not at first produce any commotion in water. This arises from the gener- 

 ation of a protecting atmosphere of rarefied aqueous vapor, which renders contact 

 with the liquid water impossible. Widely different would be the result, were the liquid 

 suddenly forced into contact with the globule by a blow from a hammer, as above 

 mentioned. Analogous causes operate when globules of the most volatile liquids 

 or solids are retained for a time in the cavity of an incandescent metalhc ladle, 

 meanwhile evaporating much more slowly than if the temperature were less. In 

 any one of these instances an explosion would follow from a contact being coerced 

 between the heated surface and the liquid. When a hammer is employed as above 

 described, mechanical force produces that contact, which, in the explosive union 

 of incandescent oxide of potassium with water, is caused by intense chemical 

 affinity. 



11. The explosion produced by Hayes, as above mentioned, and that which took 

 place in my laboratory, as well as the explosive reaction of oxide of potassium 

 with water, gave a practical confirmation to the inference, that the meeting of 

 water with the base of incandescent nitre could cause tremendous results. Sub- 

 sequently, in the winter of 1845-6, I found that when nitre, by the flame of a 

 hydro-oxygen blow-pipe supplied with atmospheric air and oxygen, is heated to 

 incandescence, and then quickly submerged in water previously situated beneath 

 the containing ladle, a sharp explosion ensues. I found, nevertheless, that when 

 thrown, under like circumstances, upon molasses or sugar, the effects were those 

 of deflagration rather than explosion. Yet, latterly, I have fallen upon con- 

 trivances, by which pulverized sugar and nitre may be made to explode. The 

 first expedient which succeeded, was that of pouring melted sugar upon the face 

 of a hammer, so as to make a disk of commensurate size. Such a disk, if it should 

 not adhere, is easily made to do so by slightly moistening the face of the hammer. 

 Some nitre was put into a thin shallow platina capsule, situated over a small anvil, 

 near one of its edges, so that the bottom of the capsule might be reached obliquely 

 by a hydro-atmospheric blow-pipe flame. Under these circumstances, the nitre 

 having been heated until its potash began to be volatilized, was struck with the 

 sugar-faced hammer. A smart detonation was the consequence. This experiment 

 may fail sometimes from the blow not being properly given ; from the nitre not 

 being sufficiently hot ; or the capsule being ill situated. The explosion of ful- 

 minating mercury by a hammer fails sometimes, from the blow not being so given 

 as to produce a due degree of parallelism between the surfaces. 



12. Another method of producing explosive reaction is as follows : — Nitre and 

 sugar being coarsely powdered, let disks of paper, about three inches in width, be 

 prepared. Place one of the disks upon an anvil, and cover it with a stratum of 

 sugar. Then cover the sugar with a stratum of nitre, placing over this another of 

 the disks. Heat a flat iron bar, wider than the disks, to a welding heat, and 

 quickly withdrawing it from the fire, and holding it above the paper, strike it down 



2 



