TRANSACTIONS OF SECTION G. 



747 



a mercurial tliermometer. It is believed that the temperatures are quite accurate 

 except those above G00° F. Above 600° the thermometer behaved irreffularly. The 

 bars were broken in a small special testin|^-miichiiie of the manometer type, tlie 

 pressure on the diaphraj^m being balanced by a mercury column. 



Rolled bars of yellow brass, Muntz metal, and Delta metal were tried, and the 

 results on these are quite regular. Some bars of cast brass also pave very fairly 

 regular results. The bars of gun-metal gave results of less regularity. This is 

 due, in part at all events, to the fact that some of the bars cast first proved 

 unsound, and new bars had to be cast to replace them. At some future time the 

 autlioi' hopes to try a series of gun-metal bars all cast at the same time. 



The results were plotted in a diagram, and show that in all cases the decrease 

 of strength follows a regular law without any such sudden loss of strength as was 

 shown in the Admiralty experiments. Even at temperatures of G00° to C50° all 

 the bars had still a not inconsiderable tenacity. 



The ultimate elongation of the bars in the two-inch test length was measured, 

 and is given in the table. There is a peculiarity in the influence of temperature 

 on the ductihty of the bars. In most cases the ultimate elongation diminishes 

 with increase of temperature. AVith Muntz metal the decrease is regular, and 

 there is still considerable elongation before fracture at a temperature of 650°. 

 With yellow brass (rolled) the decrease is more rapid, and there is very little 

 elongation before fracture at temper.atures above 500°. Cast brass behaves in the 

 3ame way. The elongations of the gun-metal bars were very irregular, and at 

 .emperatures of 600° and upwards the elongation was extremely small. 



On the other hand, in the case of the Delta metal bars the elongation increased 

 regularly with increase of temperature. 



The contraction of area was also measured. This follows generally the same 

 law as the elongation at fracture, but the contractions of area are more irregular 

 than the elongations. 



