AND THE CUBIC COMPRESSIBILITY OF MERCURY. 45 



APPARATUS FOR DETERMINING THE LINEAR COMPRESSIBILITY 



OF COPPER AND IRON. 



The differences having been thus determined, it remained to determine 

 the absolute linear compressibilities of copper and iron --a matter far more 

 difficult than the preceding. The method used for this purpose was essen- 

 tially the same in principle as that used by Amagat. The apparatus, how- 

 ever, was somewhat simpler and less liable to accidental errors. The 

 metal whose compressibility was to be determined was turned into a thin 

 rod, 2.6 meters long. The change in length of this rod was measured 

 with great care upon subjection to pressure. For this purpose the rod 

 was confined in a steel jacket of peculiar construction. 



The apparatus is depicted in section in fig. 7. The jacket for contain- 

 ing the rod which represents the extreme ends almost in full size, and 

 more of the arrangement on a reduced scale, consisted of a steel cylinder 

 () nearly 3 meters long, 23 mm. outside diameter, and 7 mm. inside 

 diameter. The steel cylinder was surrounded by a thin iron cylinder 

 (L L) of 8 cm. inside diameter to contain water for maintaining a constant 

 temperature. Screwed into the inside of the L L was the heavy iron 

 rod (K) which acted as a support for the fixed point A. The w r ater in 

 the jacket was stirred by the up-and-down motion of metal bars bent in 

 the shape of horse-shoes, placed 0.3 meter apart. The stroke of this com- 

 plicated oscillating apparatus was about 5 inches. 



Not only was the whole apparatus kept as nearly as possible at a con- 

 stant temperature, but the work was done in a room the temperature of 

 which was regulated to within 0.1. The room was heated by a gas-stove 

 having a hydrogen-barometric electrical regulator* attached for control- 

 ling the supply of gas. The air in the room was kept in constant motion 

 by means of a large electric fan. 



On the upper end of E was screwed the nut D, through the center of 

 which passed the bolt C. To insure a tight joint the threads of this bolt 

 were tinned. This bolt was pierced through the center for almost its 

 whole length by a small hole in which rested loosely the steel needle B. 

 The lower end of the bolt bore a rigid but very sharp platinum point (F) 

 about a centimeter in length. 



Obliquely through the lower end of the steel cylinder was drilled a hole 

 for the passage of an insulated copper wire (M) which was attached to 

 the lower end of the metal rod (R) to be measured. This channel was 

 made pressure-tight by the use of marine glue. 



*Regand and Fouillard, Zeitsch. wiss. microscop., 20, 138 (1903) ; also Richards 

 and Mark, Proc. Am. Acad., 41, 119 (1905). 



