33 6 



Profs. J. J. Thomson and R. Threlfall. [May 6, 



potash., and D with a solution of pyrogallol, also well boiled. The 

 tube connexions are then replaced, and the bottle E, whose volume is 

 large compared with that of the other parts of the apparatus, is then 

 exhausted as far as possible by means of a water-pump. A stream of 

 nitrogen is allowed to flow into the tubes, which are exhausted by 

 connecting them with the bottle. After exhausting and refilling 

 several times the tube is left exhausted, and D in connexion with 

 the nitrogen supply. The tap between and D is then cautiously 

 turned, as soon as this is down the contents of D flow over into C. 

 As the potash solution is denser than the pyrogallol, a very perfect 

 mixing of the fluids takes place automatically in C. As we never 

 succeeded in getting the mixed solution colourless, it is necessary to 

 preserve some of it as a standard for comparison. This is done by 

 connecting B and C with the nitrogen supply together, and opening 

 the clamp between them ; the liquid then flows into B till it stands at 

 the same level there as in C. The capacities of the tubes are arranged 

 so that there are sensibly the same quantities of liquid in B and C, 

 this can be done by raising or lowering B. The clamp between B 

 and C is now closed, C put in communication with the exhausted 

 bottle, and D in communication with the gas supply.* The pressure 

 of the gas in the exhausted bottle is observed : suppose it is p. The 

 tap between C and D is then opened, so as to allow a slow stream of 

 the gas to pass from D up through into the bottle ; as soon as a 

 sufficient deepening of the colour has taken place in the liquid in C, 

 the tap is turned off: and the new pressure p in the bottle noted. 

 Knowing p and p' and the capacity of the bottle, we can calculate the 

 quantity of gas which has flowed through the liquid in C. The stream 

 of gas passes very slowly, and it is assumed that all the oxygen it 

 contains has been absorbed by the liquid in C. We now require to 

 know how much oxygen is required to produce the same change in 

 colour, this is done by comparing the colour with that of the liquid in 

 B. A is a pipette divided into cubic centimetres, and dipping below 

 the surface of water contained in a beaker, the top of the pipette is 

 connected with the delivery-tube sealed into and running down B. 

 This tube is very fine inside B, and ends in a very fine point. In 

 order to make the comparison of colour, the upper part of the tube is 

 exhausted, and the clamp connecting it with the pipette slowly 

 opened, a stream of air will pass up from the pipette into B. This 

 process is stopped when the operator judges the colour of the liquid 

 to be the same in B and C. The level of the water around the pipette 

 is brought to the same level as that of the water inside, and the 

 quantity of air taken is read off. From this we can calculate how 

 much oxygen is required to produce the same change of colour as that 



# In the diagram the bottle is at the wrong end. 



