Compressibility of Glass 369 



the higher the pressure ; and there is no doubt that the higher 

 the pressure is the greater is the accuracy of the observation. 

 The only way in which the pressure affects the reading of the 

 micrometer is that when it is sufficiently high it produces a 

 microscopic distortion of the tube, which throws the point of 

 tlu nnl very slightly out of focus. This is remedied by a slight 

 touch of the fine adjustment screw of the microscope. 



The general result of these experiments is that the linear 

 compressibility of the glass experimented on is 0*96, and its 

 cubical compressibility 2-92 per million. 



Grassi 1 gives as the means of his observations at pressures 

 up to ten atmospheres Glass, 2-25 ; crystal, 2-804 and 2-8584. 

 The agreement between the two is very close. 



We have then for the apparent compressibility of water in 

 glass at 2-5 C. 48-68 per million per atmosphere. Adding 

 2-92 for the compressibility of glass, we have for the absolut-' 

 compressibility of water at 2-5 C. 51-60. Similarly at 12-5 C. 

 w have 45-39 for the apparent, and 48-31 for the real com- 

 pressibility. Grassi gives the following values for the true 

 compressibility of water at various temperatures: At 1-5 C. 

 -51-5; at 4-1 -49-9; mean 50-7. At 10-8 C. 48-0 ; 



: ;/4 47-7; mean 47-8. My results agree very closely 

 with these. 



Before concluding, I would call attention to a very curious 



phenomenon, which I have never seen noticed, namely, the 



peculiar noise which accompanies the relief of pressure in a 



mixture of ice and water. In comparing the piezometer K, 



\, in meltin- ice with the manometer at 12-5 C., I proceeded 



gradually from lower to higher pressures. When the pressure 



which was relieved was 100 or 120 atmosphere^. I thought 



ticed a slight n< I-HMD- tin- pressure higher the 



noise became more and more dUtim t. until when tin 



ved was ovr 200 atmospheres, it was distinctly audible 

 at a distance of five 01 he noise produced 



by bending a piece <f tin backwards and forwards, and is 

 in.ukedlv ini by accelerating the relief, just as the noise 



1 ' (1851) [3], 3*. P- 474- 



B. III. 



