56 THE COMPRESSIBILITIES OF THE AIORE IMPORTANT 



PREPARATION OF MATERIALS. 



The materials were all carefully prepared with a view to the peculiar 

 nature of the process to which they were to be subjected. They were 

 carefully examined for accidental flaws, and the metals were generally 

 hammered in order to make them as compact as possible and then if pos- 

 sible carefully turned in the lathe to fit the jacket, so that but little com- 

 pressible liquid would be needed to surround them. 



When sufficiently pure rods of metal could not be obtained in com- 

 merce, pure materials were cast in suitable form, using as a mold a piece 

 of smooth gas pipe, sawed longitudinally, so that the casting might be 

 easily removed from the mold. This was done with zinc, cadmium, tin, 

 lead, thallium, arsenic, antimony, and bismuth. The cylinders were in 

 each case satisfactory and upon superficial smoothing showed no signs of 

 cracks or flaws. The brittle substances arsenic, antimony, and bismuth 

 were smoothed by filing instead of by turning. 



Magnesium was turned from a bar of pure commercial material. The 

 calcium was very kindly sent us by Professor Edgar F. Smith, of the 

 University of Pennsylvania. It was not perfectly pure, containing nearly 

 1 per cent of silicon and some iron, but was pure enough to give a suffi- 

 ciently accurate idea of its compressibility. Its density was determined by 

 F. N. Brink, and found to be 1.54. 



The copper was turned from a compact hammered copper rod. The 

 results with this metal agreed closely with those obtained by Brink, 

 described in another place. 



Of silver two samples were used. The first sample was in the form of 

 fine rods about 3 mm. in diameter, being about 99.5 per cent pure. In 

 this form much water had to be used in order to fill the interstices in the 

 jacket. Accordingly, the results are less trustworthy than those obtained 

 with the second sample, and although approximately confirming them, are 

 not worthy of a place in the table. The latter sample was very pure, 

 having been prepared especially for this research, together with specimens 

 of gold, palladium, and platinum, through the kindness of Baker & Com- 

 pany, of Newark, New Jersey. This firm prepared cylinders of these 

 four metals in very satisfactory condition, and loaned us the large speci- 

 mens of costly material without charge. The ingots of precious metal 

 were all cast, hammered, and drawn, and finally turned to fit one of the 

 glass jackets. They were all of a very high degree of purity, a fact 

 which was shown by the specific gravities. The cylinder of gold weighed 

 149.861 grams; its loss of weight in water at 20 was 7.770 grams; there- 

 fore the specific gravity of the gold was 19.29. The palladium weighed 

 94.657 grams; its loss of weight in water at 20 was 7.783 grams; there- 



