( 178 ) 



has been arranged in such a way that first this space can be exhausted 

 and filled with pure hydrogen gas, which is necessary to keep 

 the liquid hydrogen perfectly clear later on. The liquid hydrogen is 

 again conducted into this space in the way of Comm. N°. 94/ and 

 103 PL I fig. 4 ; the evaporated hydrogen escapes at Hg to the 

 hvdrogen gasholder. The hydrogen glass is surrounded by a vacuum 

 glass E c with liquid air, which in its turn is surrounded by a glass 

 Ed with alcohol, heated by circulation. 



By these contrivances and the extreme purity of the helium we 

 succeeded in keeping the apparatus perfectly transparent to the end 

 of the experiment, after 5 hours. Protection with liquid hydrogen is 

 necessary to reduce the evaporation of the helium to an insignificant 

 degree notwithstanding that the silver coatings of the vacuum glass 

 have been removed. That it ended in a narrower part, and the helium 

 thermometer reservoir not was placed at the lowest point, was because 

 it was possible that only an exceedingly slight amount of liquid was 

 formed. The vacuum glass was made transparent up to the cock in 

 order to enable us to see any mist that might appear and if on the 

 other hand much liquid was formed, to prevent the lower part from 

 getting entirely filled without our noticing it. The latter has actually 

 been the case for some time, and would not have been so soon 

 perceived, if the walls had been silvered further. But if the glass is 

 not silvered, the conduction of heat towards the helium is much 

 greater, and without protection with liquid hydrogen the helium that 

 was formed, might have immediately evaporated. 



In the preparation of the vacuumglasses x ) Mr. 0. Kesselring, glass- 

 blower of the laboratory, has met the high demands put to him, 

 with untired zeal and devotion, for which I here gladly express my 

 thanks to him. 



§ 3. The helium. As to the chemical part of the preparation of 

 this gas I was successively assisted by Mr. J. Waterman, Mr. J. G. 

 Jurling, Mr. W. Meyer-Cluwen and Mr. H. Filippo Jzn. Chem. Docts., 

 who collaborated with Mr. G. J. Flim, chief of the technical depart- 

 ment of the cryogenic laboratory. To all of them I gladly express 

 my indebtedness for the share each of them has had in the arrange- 

 ment, the improvement, and the simplification of the operation. More 

 particularly to Mr. Filippo for his careful analyses and the effective 

 way, in which the last combustion over CuO with addition of 

 oxygen, and avoidance of renewed contamination by hydrogen was 

 carried out by him. 



!) [There was one of each in reserve before the beginning of the experiment. 

 Only one of the reserves had to be used]. 



