356 Mr. J. J. Hood on the Rate of 



permanganate, potash, oil-of-vitriol, and calcic chloride ; a 

 small mercury exit-trap was connected with / to prevent the 

 entrance of air. When the current of gas had swept through 

 the apparatus for about two hours the trap was disconnected, 

 the stopper a inserted rapidly, and the screw-clip h closed. 

 The apparatus was then exhausted by the Sprengel-pump till 

 the level of the water in the manometer stood at 500 millim., 

 and the screw-clip g closed. The absorbable gas was admitted 

 by opening the screw-clip g f and allowing the gas to enter till 

 the gauge fell to zero, then closed. 



The apparatus so charged was allowed to stand for twenty 

 hours in order that the gases might become thoroughly mixed. 

 At the end of this time the potash solution was placed in the 

 side chamber /; and when the temperature of the water-bath 

 had been kept at 16° C, the temperature at which all the 

 experiments were made, for about half an hour, by raising the 

 stopper b, the absorbent was introduced, an operation taking 

 about four seconds. 



As any fluctuations in the temperature of the water-bath 

 produced alterations in the manometric readings, constant 

 attention had to be given to this. By maintaining the tem- 

 perature of the room about ^° C. below that at which an 

 experiment was made, and using a large bath, fluctuations 

 were scarcely discernible, although a delicate thermometer 

 divided to tenths of a degree was employed. Any alterations 

 in the barometric pressure likewise influenced the manometer ; 

 indeed so sensitive was the apparatus in this respect, that it 

 was quite impossible to perform an experiment during a gale, 

 as every gust of wind was accompanied by a depression of the 

 gauge by two, three, and sometimes by even six millimetres. 

 All experiments were discarded when the barometer showed 

 any alteration between one hour before and one hour after the 

 experiment As, however, the most important observations 

 were made during the first five minutes from the start, baro- 

 metric changes could introduce no appreciable error. 



It was found necessary before beginning an experiment to 

 force the water in the gauge to the top of the scale in order 

 to moisten the inside of the tube ; otherwise the neglect of 

 this precaution rendered the results untrustworthy, owing to 

 the inside of the tube becoming dry by long standing, and 

 thereby causing irregularities in the ascent of the water. 



Although in each case the exhaustion of the apparatus was 

 carried to 500 millim., this pressure was never indicated at the 

 end of an experiment, owing to the fact that a portion of the 

 o-as was absorbed during the running-in of the potash This 

 was not under control : but only those experiments are here 



