between Metallic Masses having different Temperatures. 439 



went over not only the metals which could not be conveniently 

 tried in the other way, but also those of which I already had 

 bars. 



21. Pursuing the relations of the metals to COLD LEAD, I 

 found, in the first place, that the position of platinum is not 

 very different from that of tin. The mass used weighed about 

 7 oz., and was kindly lent to me by Dr HOPE ; its form prevent- 

 ed its being used as a bar, and its small thickness and angular 

 corners did not fit it for retaining a high temperature, or per- 

 forming well the part of a block. When held in a vice and 

 heated by a spirit-lamp, the vibrations of a cold lead bar were 

 very active. From some experiments made with iron, tin, and 

 platinum, at a temperature of 350, I conceived that they stood 

 as vibrators in the order just named : I had not then, however, 

 fallen upon a method of operating upon small masses with ac- 

 curacy ; and subsequent experiments, often repeated, with a small 

 thick mass lent me by Professor JAMESON, and heated till water 

 boiled on the surface (as above explained) led me to the con- 

 clusion, that, in the other mode of experimenting, platinum had 

 been placed too low, as might have been anticipated, and that 

 it is at least equal to iron in vibrating power. It appears at the 

 same time that there is little difference between platinum, iron 

 and tin. 



22. From numerous experiments with antimony and bismuth, 

 these metals when heated appear to have no vibratory action 

 with cold lead. This experiment was tried at a great range of 

 temperatures, and, notwithstanding the low melting point of 

 bismuth, I raised it by a particular arrangement to a temperature 

 of 350 without obtaining any vibration. 



23. The next experiments were made with silver. Even 

 with very unfavourable apparatus it appeared remarkably active 

 as a vibrator. A small block of heated silver placed in a hot 

 vice, gave indications of being at the very top of the series of 



