4 Professor Dewar [Jan. 24, 



While continuing to work intermittently at his great discovery, 

 Joule employed himself in the following years in elaborate investiga- 

 tions bearing uj)on the point of maximum density of water, specific 

 gravity, and atomic volumes. An illustration of his method of deter- 

 mining maximum density was given by means of two large cylinders 

 filled with water and connected by a narrow channel in which was 

 placed a floating indicator. It was shown that the slightest variation 

 in density of the water of either cylinder — variations far beyond the 

 scope of the most delicate thermometer — set up currents which were 

 immediately detected by the movement of the indicator, and that by 

 this means it was quite possible to ascertain the exact temperature at 

 which water attained its maximum density. 



Joule's determinations of atomic volumes were marvellous at the 

 time they were made, and were still interesting. Illustrations of his 

 work in this direction were given by means of a solution of sugar, 

 which was seen to occupy j)ractically the same space as was occupied 

 by an amount of water exactly equivalent to that combined in the 

 carbohydrate. The carbon hypothetically combined with the water 

 to form the sugar appearing to make no sensible difference to the 

 volume ; and in contrast to this was seen the enormous difference in 

 volume brought about by dissolving two equal portions of soda car- 

 bonate, one portion being ordinary hydrated crystals and the other 

 portion being anhydrous, in equal volumes of water. 



Joule's last great research was carried out conjointly with Sir 

 William Thomson, and occupied nearly ten years of laborious enquiry. 

 Its chief object was to prove that in compressing a gas the amount of 

 heat produced is equivalent to the work done, and independent of the 

 mere fact of the approach of the particles. But Joule was desirous of 

 amplifying the enquiry, and in fact the work might be divided into 

 three sections : (1) the study of gases passing through narrow apertures'; 

 (2) the velocity attained by bodies passing through the air ; and (3) 

 the temperature ultimately attained by such moving bodies. With 

 respect to 2 and 3, it was shown that a body rotating in the air at the 

 rate of about 150 to 180 feet per second increased in temperature by 

 nearly 1° F., and that this increase of temperature was definite for a 

 given velocity, and independent of the size of the moving mass and the 

 density of the gaseous medium. With regard to (1) the relation of 

 gaseous pressure and volume to temperature, the researches of Eegnault 

 had already shown that the simple law of Marriotte and Boyle could not 

 stand by itself; and Joule sought to modify it by the study of gases 

 passing through very small tubes or porous bodies. The investiga- 

 tions were carried out at Manchester on a large scale, and were assisted 

 by a Government grant. Steam engines were employed to maintain a 

 current of gas at a constant temperature and pressure through loug 

 coils of pipe placed in water tanks. They proved that any difference 

 of temperature in the gas brought about in its passage through the 

 porous body must be due to work done by it, and that this difference of 

 temperature varied for different gases, according to their constitution. 



