326 On the Chemical Evidence 



cumstance the fire had been attributed. In some districts of 

 the North of England, martial pyrites are found mixed with the 

 coal in such abundance as to render the coal absolutely un- 

 saleable ; but persons are employed to pick oiit these pyrites 

 and remove them to a proper situation for the manufacture of 

 green vitriol. This species of coal, v^hen in contact with water, 

 is apt to be decomposed, or rather to decompose the water, and 

 thus burn spontaneously. An account of several other cases 

 of spontaneous combustion may be seen in my Chemical Essays, 

 voki., pp. 222— 231. 



Bryan Don kin, Esq., Chairman of the Committee of Me- 

 chanics at the Society ofArts^-wSiS next called. 



This gentleman stated that he considered Mr. Wilson's pro- 

 cess to be much safer than the old one ; that fixed oil emits 

 neither inflammable vapour nor gas till it has arrived at a 

 temperature far beyond that at which sugar will take fire ; that 

 the lead-pipe arising from the oil-vessel would not have been 

 melted by the oil vapour, even if it had been heated to more 

 than 600°, and that the boiling of sugar in the ordinary mode 

 is attended with moredanger than the boiling of it by means of 

 the oil process. 



Observations. — In answer to a question from Mr. Stephen, 

 this gentleman replied, " I rather think no permanent gas would 

 be given out under 600°.'' When Mr. Donkin spoke of per- 

 manent gas, I conclude he meant to be understood of perma- 

 nently inflammable gas ; because carbonic acid, which is a per- 

 manent gas, is given out from fish-oil at a temperature much 

 below 400°. It is necessary, however, to remark, that at the 

 time of the trial neither Mr. Donkin, nor any of the chemists 

 employed by the plaintiffs or by the defendants, seem to have 

 made any experiments to ascertain the means by which a fixed 

 oil might be rendered capable of giving out inflammable gas. 

 Since the trial, I have made an experiment which convinces 

 me, that fish-oil must be submitted to a heat sufficient to de- 

 compose it, before a particle of inflammable gas can be pro- 



