1886.] on New Applications of the Mechanical Properties of CorJc. 437 



Metals wlien subjected to pressures wliicli exceed their elastic 

 limits, so that they are permanently deformed, as in forging or wire- 

 drawing, remain j)ractically unchanged in volume per unit of weight. 



I have here a pair of common scales. To the under sides of the 

 2)ans I can hang the various specimens that I wish to examine ; under- 

 neath these are small beakers of water which I can raise or lower by 

 means of a rack and pinion. Substances immersed in water lose in 

 weight by the weight of their own volume of water ; hence if two 

 substances of equal volume balance each other in air, they will also 

 balance when immersed in water, but if their volumes are not the 

 same, then the substance having the smaller volume will sink, because 

 the weight of water it displaces is less than that displaced by the 

 substance with the larger volume. To the scale on your left hand is 

 suspended a short cylinder of ordinary iron, and to the right hand 

 scale a cylinder of ordinary copper. They balance exactly. I now 

 raise the beakers and immerse the two cylinders in water, you see the 

 copper cylinder sinks at once, and I know by that, that copper has a 

 smaller volume per pound than iron, or, as we should commonly say, 

 it is heavier than iron. I now detach the copper cylinder and in its 

 j)lace hang on this iron one, which is made of the same bar as its 

 fellow cylinder, but forced, while red hot, into a mould by a pressure 

 of sixty tons per square inch and allowed to cool under that pressure. 

 The two cylinders balance, as you see. Has the volume of the iron 

 in the comj)ressed cylinder been altered by the rough treatment it 

 has received ? I raise the beakers, immerse the cylinders, the balance 

 is not destroyed, hence we conclude that although the form has 

 been changed the volume has remained the same. I substitute for 

 the hot compressed cylinder one pressed into a mould while cold, and 

 held there for some time, with a load of sixty tons per square inch ; 

 the balance is not destroyed by immersion, hence the volume has not 

 been altered. I can repeat the experiments with these copper cylinders 

 and the result will be found the same. Extension also is incapable of 

 appreciably altering the density of metals. I attach to the scales 

 two specimens of iron taken from a bar which had been torn asunder 

 by a steady pull. One specimen is cut from the portion where it had 

 not been strained, and the other from the very point where it had been 

 gradually drawn out and fractured. The specimens balance, I 

 immerse them, you see the balance is not destroyed ; hence the 

 volume of the iron has not been changed appreciably by extension. 



But cork behaves in a very different manner. I place this 

 cylinder of cork into just such a brass tube as served to restrain the 

 indiarubber and apply pressure to it in the same way : you see I can 

 readily compress the cork, and when I release it it expands back to its 

 original volume : the action is a little sluggish on account of the 

 friction of the cork against the sides of the tube. In this case, 

 therefore, a very great change in the volume of the material has been 

 easily effected. 



But although solids evidently do not change sensibly in bulk, 



