TRANSACTIONS 01' THE SECTIONS. 51 



On the Explosive Character of a Mixture of Magnesium and Potassium, 

 Chlorate. By P. Braiiam. 



On Hydrogen Peroxide and some Uranium Compounds, By T. Fairlet. 



On the Tliermo-cliemistry of Oxygen. By T. Fairley. 



On some Candles altered by long Exposure to Sea-ivater. 

 By Professor J. H. Gladstone, F.B.S. 



Mr. Latimer Clark had sent the author some specimens of candles recovered 

 from the wreck of a vessel sunk off the Spanish coast in 1702. They had remained 

 submerged till 1875, a period of 173 years. 



The wick has rotted away, leaving scarcely any trace of its existence, while the 

 fatty portion has become a friable heavy substance of a dull white colour. The 

 candles bore evidence of having been made by dipping, for the concentric layers 

 were easily separated from one another. Beth the outer and inner portions still 

 contain some of the fat apparently unchanged ; they are unctuous to the touch and 

 have a fatty odour. The fat can be easily separated from the rest by treating the 

 altered candles with ether. 



After combustion there remained a strongly alkaline white ash, consisting of 

 carbonate and chloride of calcium and sodium, with traces of magnesium and 

 potassium. From the details of the analysis it appears that the fat has been con- 

 verted in great measure into calcium and sodium salts, doubtless by the slow 

 replacement of the triatomic group C ,11- in the stearino by 3 atoms of the metal, 

 with the simultaneous production of glycerine. Though the calcium in sea-water 

 is far less abundant than the sodium, it appears to have had a much greater effect ; 

 and it is of course impossible to say whether the one salt has not been made by 

 double decomposition from the other. The author pointed out as a most interesting 

 fact, that whereas the fat has been in contact with a practically unlimited amount 

 of sea-water for 173 years, and a chemical cbange between them has been possible, 

 the double decomposition has proceeded so extremely slowly that the reaction is 

 only about half completed at the present time. 



On tlie Application of a new Unit of Light to the Examination of Coal-gas. 

 By A. Yernon Harcottrt, M.A., F.B.S. 



After pointing out the variations to which the unit of light now in use — a sperm 

 candle burning 120 grains of sperm per hour — is liable, and the irrelevance, for the 

 purpose of estimating the value of illuminating gas, of tests affected by any other 

 portions of the force radiating from aflame besides those which produce vision, the 

 author explained the application of the new unit of light to the examination of coal- 

 gas. By making a mixture in a small gasholder of the most volatile spirit from 

 American petroleum, which distilled at 50° C, with ordinary air, in the proportion of 

 one of the liquid to 600 of air, or seven of the vapour to twenty of air, he prepared 

 a gas which was scarcely at all soluble in water, and was permanent at ordinary 

 temperatures. and pressures. This gas was burnt at a pair of burners, corresponding 

 to the two candles commonly used in photometry, each consisting of a brass tube 

 surmounted by a plate through which the standard gas issued at the rate of half a 

 cubic foot per hour, through an opening one quarter of an inch across. The 

 illuminating power of the gas and the rate of burning were so adjusted that each 

 burner gave the average light of one candle. Photometric results obtained with 

 the same sample of coal-gas showed that successive observations made with standard 

 gas gave closely corresponding results. 



