1891.] Heat produced by compressing Liquids and Solids. 141 



ingots were prepared and filed flat on one face, so that they could be 

 tied one on each side of the flat junction, which was used without 

 any tube. It is difficult in this way to secure at all a satisfactory 

 contact between the sample and junction, and we are of opinion that 

 the method originally used as described in the beginning of the paper 

 will prove more trustworthy. The rise of temperature in metals is 

 so small that, to obtain a readable deflection, it would be necessary to 

 use a more sensitive junction than can be made of platinum-iridium 

 alloys, as the galvanometer cannot be made much more sensitive 

 without becoming too slow to permit of true readings. Considerable . 

 care was taken to secure specimens of metals in a state of approxim- 

 ate purity, but, the results having proved of so little value, the 

 methods used need not be here described. 



Rise of temp. 



O 



Aluminium O'l 81 



Magnesium O'ISl 



Zinc 0-062 



Silver 0-047 



Tin 0-125 



Several other metals were tried, but the deflection was too slight 

 to allow of accurate observation.. 



Notwithstanding the large amount of work involved in these ex- 

 periments, study of the results shows but too clearly that this research 

 can only be regarded as preliminary. Many interesting problems 

 present themselves for solution, of which we think the following are 

 worthy of mention : 



1. The effect of pressure in causing solidification should be fol- 

 lowed up and tested with different substances at temperatures little 

 removed from their melting point. In the case of tertiary butyl 

 alcohol it may be noted that, though the crystals formed were too 

 small to be clearly distinguished, they appeared to be of a different 

 habit from those formed under ordinary circumstances. 



2. The effect of temperature in causing differences in the amount 

 of heat developed. It might be necessary to compare the substances 

 at different temperatures so as to have them all in the same physical 

 state. 



3. The relation of compressibility to heat of compression. 



This is but an indication of the few out of the many lines of 

 research suggested by the work we have done. Some of these seem 

 to afford good hope of yielding new knowledge of the constitution of 

 matter. 



In conclusion, we wish to record our gratitude to the Government 

 Grant Committee of the Royal Society for affording us the means of 



