492 yiM. Bunsen and Schischkoff's Chemical 



cap of the glass tube c, which is about 1 metre long and 2*5 

 millims. wide. The cap h contains a stop having a small circular 

 orifice with a sharp edge, through which the grains of j)owder fall 

 in a fine stream like that of an hour-glass, when the vulcanized 

 tube is brought into the position represented in fig. 2. The 

 tube a, therefore, in being fitted on b, is held downwards in order 

 to prevent the powder from prematurely falling into the glass 

 bulb d. This glass bulb d, which is gently heated externally 

 by means of a lamp, serves for the combustion of the powder, 

 which takes place regularly in the form of a rhythmical flame if 

 the stream of powder is fine, and falls continuously. Occasional 

 interruptions, which are easily remedied by gentle motion of the 

 india-rubber, are not injurious to the experiment. 



The solid residue remains almost completely in the bulb d, 

 and in its tubular continuation, while the smoke is deposited 

 without loss in the tube e e^, which is about 25 millims. in dia- 

 meter, and 1 to 2 metres long, and the gaseous products escape 

 into the air at e,. These gaseous products cannot be collected 

 by means of a delivery tube at the extremity e^, for as soon as 

 such a tube dips under a liquid, the pressure of even the small- 

 est column causes the flame to reverberate from the bulb d, 

 through the tube c, into the vulcanized tube, and thereby ex- 

 plode the whole mass of the powder. Such explosions in masses 

 of from 15 to 20 grms. of powder, are very violent, but are quite 

 without danger, even for the operator in charge of the tube a ; 

 for the vulcanized tube is immediately burst without off"ering any 

 resistance. In such explosions, which also occur if the tubular 

 continuation of the bulb is stopped up, the glass part of the 

 apparatus is not injured. 



In order, therefore, to collect the gaseous products for ana- 

 lysis without danger of such an explosion, we used the tube//, 

 connected with the aspirator h, and collected the gases in the 

 tubes ss,, which could be closed by means of clamps, and then 

 hermetically sealed before the blowpipe. 



The residue obtained in the bulb d, fig. 2, from the combus- 

 tion of about 20 grms. of powder, formed a semi-fused, yellowish- 

 gray compact mass, which dissolved readily in water, leaving a 

 little charcoal. It was analysed according to the following me- 

 thods : — 



1 . Determination of the unburned Charcoal. — 7 grms. of sub- 

 stance dissolved in hot water gave 00682 grm. of charcoal col- 

 lected on a weighed filter, which corresponds to 0'97-i per cent, 

 of charcoal. 



2. Determination of Sulphide of Potassium. — The filtrate from 

 the above was digested for two days with ignited oxide of copper 

 in a stoppered cylinder, with occasional agitation, until the 



