312 Mr. G. Owen on Condensation Nuclei produced in 



400° C.) than before in order that condensation might result 

 for an expansion of the same amount. 



That this eflfect was not due to the wire having on it some 

 ilirt which was driven off when the wire was raised to red 

 heat, is disproved by the fact that the wire had been made 

 red-hot before the hydrogen was passed into the apparatus. 

 In the light of later experiments this effect is very probably 

 due to the hydrogen having been rendered purer by the high 

 temperature of the wire causing the combination with hydro- 

 gen of the small quantity of oxygen in the impure gas. 

 For in later experiments, when every precaution was taken 

 to obtain hydrogen free from oxygen, it was found that the 

 platinum wire must be raised to temperatures above a red 

 heat before condensation-nuclei were produced. 



Pure hydrogen was obtained from pure zinc and dilute 

 pure HCl. The gas on its way to the expansion-apparatus 

 was passed through wash-bottles containing boiled distilled 

 water and a solution of potassium permanganate, then through 

 tubes containing red-hot copper, calcium chloride, and tightly- 

 packed cotton- wool. 



The distilled water in the expansion -apparatus and in the 

 cloud-chamber had been well boiled, and allowed to cool in 

 an atmosphere of hydrogen. Before every experiment in 

 hydrogen the wire was always heated for some minutes to a 

 bright yellow heat. This was found to be necessary in order 

 to obtain fairly consistent results. 



The results obtained in pure hydrogen are of a nature 

 similar to those obtained in air. 



As in the case of air, a relation was found to exist between 

 the expansion and the minimum temperature at which nuclei 

 are produced which are caught with that expansion ; and, 

 as before, the nuclei were found to increase in size and 

 number as the temperature was increased. But in the case of 

 hydrogen the results are obtained at temperatures some six 

 or seven hundred deorees hioher than in air — in fact, not 

 until the wire is luminous. 



The curves giving the relation between the expansion and 

 minimum temperature are roughly straight lines, and are 

 shown in diagrams 4, 5. The fact that some points are a 

 good deal oft' the curve is not so serious when one considers 

 the experimental difficulties, such as the difficulty of measuring 

 the temperature accurately when so high, and of distinguishing 

 between the density of one shower and another. The extreme 

 delicacy of the expansion method of detecting nuclei is 

 another source of difficulty. Owing to the results not being 

 yery regular, a large number of experiments were made to 



