384 



Mr. A. Mallock 



[May 2' 



in various metals and a few non-metallic substances from this cause. 

 The range of temperature covered by the experiments lay between 

 100° C. and the temperature of liquid air. I had hoped to go as low 

 as liquid hydrogen temperatures, but the difficulty of getting a 

 sufficient supply of the liquid has been in the way. 



The first series of experiments were on the temperature change 

 of Young's Modulus, and the slide (6) will show the form of 

 apparatus adopted for the purpose. 



It was necessary to be able to work on small test-pieces, both on 



Fig. 2. 



account of cost of such metals as rhodium, platinum, etc, and of the 

 convenience of having a moderate-sized lathe for the heating or 

 cooling liquids. 



Four temperatures were generally tried for each test-piece — viz. 

 100° C, ordinary temperature about 14° C, a freezing mixture of 

 alcohol and carbonic acid ( - 100° C), and liquid air ( - 17o C). 



The test-pieces were in the form of thin plates an inch and a half 

 long and about a quarter of an inch wide, with a thickness of one or 

 two hundredths of an inch. The lower end of the test-piece was 



