18 MR JOHN A1TKEN ON 



are drawn through the apparatus. On the pipe, E, at a point about I m. from the 

 combustion chamber, there is a branch pipe, G, which leads through the stop-cock, H, into 

 the test-flask, /. This flask is provided with an air-pump, K, and a filter, L. By means 

 of this part of the apparatus a sample of the air passing from the combustion chamber 

 can be tested when desired. To make this test, the flask, I, is first cleared of all dust 

 particles by means of the pump, K, and filter, L. After the air is free from nuclei, the 

 stop-cock, H, leading to the pipe from the combustion chamber is opened, and a sample 

 of the products passing in the pipe are drawn into the flask, I. The stop-cock, H, is then 

 dosed, and the air in the flask expanded by means of the pump, when the density of the 

 condensation observed indicates the number of nuclei in the air from the burning 

 hydrogen. For observing the amount of condensation in the flask, I, the dark lantern, R, 

 was used, the light being focussed on the contents of the flask by means of the 

 condensing lens, S, whilst the cloudy particles were observed through the lens, T. 

 When making these experiments the room should be darkened, as the phenomena are 

 then most easily observed. 



At first the results were unsatisfactory : there was always considerable clouding, but 

 the density was far from being constant. Efforts were, therefore, made to get everything 

 as perfect as possible. Pure redistilled zinc and the purest sulphuric acid were used 

 for generating the hydrogen, and, to prevent sulphur compounds from coming over with 

 the gas, cupric sulphate* was added to the sulphuric acid, which was used very weak, 

 and mixed with water and cooled before being put in the flask, A. The flask, A, in 

 which the hydrogen was generated was kept cool by a bath of water, M. The hydrogen 

 after leaving the generating flask was passed through the wash-bottle, N, filled with 

 solution of lead nitrate. The gas then passed through a plug of cotton-wool, O, 

 saturated with a strong solution of lead nitrate, then through a dry cotton-wool filter, P. 

 It then passed through a small but very tight cotton-wool filter, Q, to regulate the 

 pressure of the gas, as without this obstruction the light was apt to go out, owing to 

 irregular pressure due to the bubbling of the gas. The gas was burned at the platinum 

 jit, B, which was made red-hot with a blow-pipe before each experiment to thoroughly 

 cleanse it, and prevent nuclei being driven off by the heat. The combustion chamber 

 was kept cool by surrounding it with a wet cloth. 



We shall now turn to the cmestion of filtering the air. This looks a simple matter, 

 but it was found to be one of the most troublesome parts of the experiment. In the 

 apparatus a largo cylindrical metal box, C, filled with cotton-wool, was used, with a pipe 

 leading off it at the end. This pipe was at first connected with the combustion chamber 

 bv means of a short length of india-rubber tube. When the pure hydrogen was burned 

 in this apparatus the flame was scarcely visible, and quite invisible if there was any 

 daylight in the room, even if the combustion cylinder was surrounded with a black sur- 

 face. Watching the burning hydrogen on one occasion, suddenly a little bright spark 



* It was necessary to be constantly adding cupric sulphate to the acid in .1. If this was not done, con- 

 siderable condensation took place in the products. 



