132 ANNUAL RECORD OF SCIENCE AND INDUSTRY. 



tus for showing the increase of weight in combustion where 

 the products are gaseous. Upon eacli scale pan of a balance 

 is placed a candle. Above it is a glass gas chimney, having 

 gauze at bottom, and containing several large pieces of caus- 

 tic soda, suspended to the stirrup. One of the candles being 

 lighted, in a very few minutes that side of the balance pre- 

 ponderates. In six minutes there is a difference of weight 

 of a gram, and in fifteen minutes of more than three crams. 



A lecture was delivered in March before the London Chem- 

 ical Society, by T. E. Thorpe, on the theory of the Bunsen 

 lamp, in which he gives first a bit of history relative to its 

 origin, and then discusses the results of Mallard as to the 

 velocity of inflammation in meters per second of various 

 mixtures of coal-gas and air, of Blochmann on the composi- 

 tion of the gas at various parts of the flame, and of Heumann 

 and Frankland on the cause of the disappearance of lumi- 

 nosity in the flame. 



Midler has proposed a simple apparatus for determining 

 approximately the density of gases, as a lecture experiment. 

 A well-closed half-liter flask, through the cork of which pass- 

 es a glass tube drawn out to a point, is taken, and water is 

 boiled in it until all the air is expelled from its interior, when 

 the point is sealed. After cooling it is carefully weighed. 

 The difference is the weight of the contained air, corrections 

 being made for the tension of aqueous vapor and for tem- 

 perature. Replacing the stopper by a second having two 

 tubes, and again expelling the air from the flask, it is succes- 

 sively filled from a gasometer with the gases to be weighed, 

 and their weights determined in succession. The method is 



quite accurate. 



NON-METALLIC. 



Schobig, at Kammerer's suggestion, has experimented to 

 determine the effectiveness of a solution of potassium per- 

 manganate for the purification of hydrogen gas for analytical 

 uses. He finds it highly satisfactory, the impurities pres- 

 ent sulphur, arsenic, antimony, phosphorus, and carbon 

 all being completely removed. The hydrogen itself is oxid- 

 ized by the solution, but only to a slight extent. The hy- 

 drogen thus purified the author finds will reduce silver. 



Corne lias observed that if to an aqueous solution of an 

 iodate there be added a few drops of water in which phos- 



