TRANSACTIONS OF THE SECTIONS. 79 



the light a much greater motion of the air than in the dark closet. Both these 

 circumstances would operate in producing a larger consumption of candle. 

 In a second trial with No. 2 composites the results were — 



Tn the dark 140 grains each candle. 



In the light 134 „ „ 



In a third, also with No. 2 composites, the results were — 



In the dark 131 gi-ains. 



In the light 129 „ 



In these two trials the flames were protected as far as possible from currents of 

 air, and in the third trial the temperatiu-e both in the light and in the dark was 

 nearly equal. 



The fourth trial was made on a bright sunshiny day with hard sperm candles, 

 which are less affected by variations of temperature than the composites. The 

 results were — 



In the dark (temp. 81°) 544 grains, 



or 136 grains per hour. 



In the light (temp. 84°) 567 gi-ains, 



or 142 grains per horn* nearly. 



It is evident that in this case the increase of temperature caused by the bright 

 sunshine led to an increased consumption of material. 



It will be seen that in the first and fourth trials there is a greater consumption 

 of material in the light than in the dark, and in the second and third trials the 

 consumption is greater in the dark than in the light ; but in any case the difference 

 is so small, amounting only to from 2 to 7 grains per hour, that it may fairly be 

 referred to accidental circumstances, such as differences in temperature, iu currents 

 of air, and in the composition and make of the candles, the final conclusion being 

 that the direct light of the sun or the diffused light of day has no action on the 

 rate of burning, or in retarding the combustion of an ordinary candle. 



On the Manufacture of Chlorine by means of perpetually regenerated Manganite 

 of Calcium. By Walter Weldox. 



Since the last Meeting of the British Association chlorine has begun to be manu- 

 factured extensively by a process which depends on the production and perpetual 

 regeneration of a compound no mention of which as yet exists in chemical litera- 

 ture. This process, besides thus producing and continually reproducing what Ihe 

 author believes to be a new compound, reduces by fully 80 per cent, the principal item 

 in the cost of the manufacture of chlorine, greatly increases the quantity of chlo- 

 rine which can be practically obtained from a given quantity of hydrochloric acid, 

 and, moreover, enables the manufacture of chlorine to be earned on without the 

 production of any offensive residue. 



What has hitherto been the ordinary process of manufacturing chlorine consists 

 simply in digesting with hj-drochloric acid ores containing peroxide of manganese. 

 The reaction which takes place, besides liberating chlorine, produces chloride of 

 manganese, which remains behind in solution after the chlorine has gone off', and 

 has hitherto been usually thrown away. There have been proposed and tried a 

 great number of processes for transforming this chloride into peroxide for use over 

 again, but the only one of them which has met with the slightest measure of prac- 

 tical success, prior to that which is the subject of this paper, is the one which is 

 known, from the name of its inventor, as Dunlop's process. Dunlop's process de- 

 composes the chloride of manganese by heating its solution, under a pressure of 

 from two to four atmospheres, with milk of carbonate of lime, and then, in the dry 

 way, transforms the resulting carbonate of manganese into a mixture or compound 

 of two equivalents of peroxide with one equivalent of protoxide, by subjecting it 

 for 48 hours to the action of air at a temperature of about 600° Fahr. The 

 product of Dunlop's process is a sufficiently satisfactory one, containing about 72 

 per cent, of MnOj ; but the process requires a very formidable amount of appa- 



