PHOSPHORUS, SULPHUR, AND SELENIUM. 



31 



THE APPARATUS. 



In the course of this work, which was prolonged over two years, five 

 glass jackets were used. These were of two types, represented in the 

 accompanying diagrams. Jacket v was the one already used for the alkali 

 metals, and jackets vn, vm, and ix were of the same type, but slightly 

 larger; jacket in was of a different and somewhat less advantageous 

 -form, devised before jacket v. The pressure of the high column of mer- 

 cury on the right-hand side in this form made the danger of leakage 

 greater than in the other ; but when the stopper was properly lubricated, 

 no trouble was experienced. 



Jacket in was always lubricated around the glass stopper with 10 mg. 

 of rubber-paraffin lubricant; jackets v, vm, and ix always with 3 mg., 

 and jacket vn always with 4 mg. The different extent of lubrication and 

 the difference in the glass caused but a very slight difference in the behav- 

 ior under compression of these jackets. Because the lubricant was always 

 the same in kind and quantity with any one of the jackets, even this slight 

 difference is entirely 'eliminated from the final results. 



Data concerning glass jackets filled with mercury. 



Before giving the data for each of the elements, the behavior of the 

 jackets alone under pressure must be detailed. The figures given in the 

 accompanying table were found by the method already described in the 

 paper on the compressibility of the alkali metals. In brief, the empty jacket 

 leaving been weighed, it was carefully filled with mercury, and weighed 

 again. The difference gave the values in the first column of the figures in 

 -the table. This mercury 'had been so adjusted as to stand slightly above 

 the fine platinum point in the narrow limb of the apparatus. Pressure 

 -was now applied, by placing the jacket in the barrel of a Cailletet compres- 

 sion apparatus, until contact was just made at this point this pressure 



