TRANSACTIONS OP THE SECTIONS. 49 



or candle-light under the circumstances of the experiment, and is nevertheless so 

 easily acted upon by solar light, that when perfectly free from all admixture, its com- 

 ponent gases unite with explosion in the diffused light of a room. 



In order to eliminate the source of error of the retarding action of the hydrochloric 

 acid formed, it is only necessary that water saturated with the gaseous mixture should 

 be present. By this means the hydrochloric acid is removed from the gas at the 

 moment of its production, and thus a diminution of the volume of the gas is effected. 

 This diminution of volume is a direct measure of the amount of chemical action of 

 the light, and it is upon this fact that the method of measuring photochemical action 

 rests. According to this method, and by carrying out a great number of necessary 

 precautions, the following laws were arrived at : — 



1. The amount of chemical action is directly proportional to the time of insolation. 



2. 7%e amount of chemical action is directly proportional to the intensity of the 

 light. 



The law connecting the amount of action with the mass of the decomposing body 

 is not as yet completely established, but the results obtained seem to show that the 

 light suffers mere optical absorption, and is not in any way expended, and therefore 

 cannot be represented by any equivalent in chemical action. 



Many very interesting phsenomena connected with the action of solar light upon 

 mixtures of chlorine and hydrogen will be fully treated of in the ne.st communication 

 to the Association. 



In the prosecution of these researches the authors reserve to themselves the exami- 

 nation of all subjects arising out of this method ; amongst others, — 



1. The reflexion and absorption of the chemical rays. 



2. Polarization of the chemical rays. 



3. Examination of the arrangement of the chemical rays in the spectrum. 



4. Application of the method to meteorological observation. 



On the Manufacture of Iron by Purified Coke. 

 By F. Grace Calvert, F.C.S. 



After pointing out what were believed to be the causes of the inferiority of iron in 

 many works, apart from the varying qualities of the ores, the injurious action which 

 an impure fuel had upon the quality of the iron was particularly alluded to ; and the 

 necessity of removing sulphur from the coal or coke, in the blast-furnaces, before it 

 could be imparted to the cast iron during the process of smelting, was strongly 

 enforced. Mr. Grace Calvert then referred to several instances in whiclr the quality of 

 iron, by the application of the chloride of sodium in the blast-furnace, had been greatly 

 improved. 'J'hese improvements were described to have been eflected at a very small 

 cost by the following simple process. If the blast-furnace was worked entirely with 

 coal, chloride of sodium was added with each charge, in proportion to the quality of ore 

 and flux employed ; but a better result was produced if the coal was previously con- 

 verted into coke, and a very slight excess of the chloride was used in its preparation in 

 order to act on the sulphur of the coal and of the ore, should any be found therein; and 

 a greater improvement was manifested in the quality of iron, when only coke so prepared 

 was used in the blast-furnace. The coke so purified emitted no sulphurous fumes when 

 taken out of the coke-oven ; nor, when extinguished with water, did it give off the 

 unpleasant odour of sulphuretted hydrogen ; nor was any sulphurous acid gas liberated 

 during the operation of smelting iron in the cupola, or in raising steam in the loco- 

 motive boiler, by coke so prepared ; and it was stated that these decided advan- 

 tages were gained, in some cases, at an additional cost of only one penny per ton of 

 fue . 



Mr. Grace Calvert gave the results of a series of experiments which had been made 

 upon trial bars one inch square, cast from iron melted in the cupola, with coke pre- 

 pared by his process. He exhibited specimens of the iron so prepared, when the 

 closeness of texture and the absence of the ' honeycomb' appearance prevailing in the 

 iron cast with ordinary coke were clearly demonstrated. 



The mode of experimenting was described and the results were given very fully, and 

 it was shown that the average increase of strength was from 10 to 20 per cent. 



1855. 4. 



