212 Canadian Record of Science. 



jecting tongues of lead are left each 1 in. long and J in. 

 wide; these are turned up inside the pipe and upon their 

 ends rests loosely a perforated circular piece of lead repre- 

 sented in the figure by the dotted line. Into the top of F 

 is fitted a rubber cork through which a glass tube carry- 

 ing a glass tap passes. (A well-greased wooden plug carry- 

 ing a glass tube provided with a rubber tube and screw 

 pinch-cock will answer.) This is connected by a rubber 

 tube with a wash-bottle as represented, and the. wash-bottle 

 with a delivery-tube. Into H are screwed two screw-eyes to 

 which is fastened a loop of cord which in turn passes 

 through a loop on the end of a cord passing over the two 

 pulleys and down to the cleat. 



To set the apparatus in action, about 2 litres of dilute 

 sulphuric acid, 1 in 10, is placed in the Winchester, and 

 100 grms. of iron sulphide in pieces about the size of a bean 

 in the lead cylinder. The latter is lowered into the acid 

 and the glass tap slowly opened until a current of gas of 

 the rapidity desired is obtained. To put the apparatus out 

 of action, the tap is closed and the lead cylinder raised 

 until its bottom edge just touches the surface of the acid. 

 If acid stronger than 1 in 10 is used ferrous sulphate 

 crystallizes out in such quantity as to render the cleaning 

 of the generator difficult. There are several brands of iron 

 sulphide on the market, some of which are quite unsuitable 

 for this generator as they are very dense and require 

 stronger acid for their decomposition than that recom- 

 mended above ; a clean porous variety (sold in large 

 lumps) is to be preferred. 



The apparatus described above was employed for three 

 months in the Chemical Laboratory of McGill College and 

 gave the greatest satisfaction ; the reason that its use was 

 then discontinued being its replacement by a much larger 

 " machine " working upon the same principle, and which 

 for more than two years has given entire satisfaction. 

 That the "machine" works economically may be seen 

 from the fact that during the session of 1893-94, with about 

 ninety students working in the laboratories, only about 



