FORMATION OF SMALL CLEAR SPACES IN DUSTY AIR. 271 



is, that if after the powder is heated it is cooled quickly, and again heated 

 before there is time for it to absorb gases, the same semi-fluid appearance 

 is again produced while heating. Further, if the powder, instead of being 

 heated in a closed vessel,, is placed in a cup, so that the under side of 

 the powder is kept hot, while the top is cooled by radiation, so long as 

 these conditions are kept up the powder retains its fluid-like properties, 

 moving about on the slightest tilting of the cup, and conducting itself in a way 

 very suggestive of the spheroidal condition, but without any generation of 

 vapour to give rise to the irregular movements seen in liquids. It seems 

 possible that something of the spheroidal condition may receive its explanation 

 in this repulsion between hot and cold surfaces. This repulsion may be illus- 

 trated by placing a hot and a cold surface together. A piece of cold glass, for 

 instance, slides about in a remarkably easy way on a hot surface of glass. 



Many practical applications of this attraction and repulsion will no doubt 

 be found. It might be easily applied to the condensation of those fumes from 

 chemical works which at present are allowed to pollute the air. But perhaps 

 the application of most general interest would be towards the prevention of 

 smoke, or rather the prevention of the escape of smoke into the atmosphere. 

 Whatever interest, however, it may have in this way, it is clear it can never 

 meet with general adoption, save under compulsion, as it will effect no saving 

 in fuel, such as would result from more perfect forms of combustion. 



I have, however, made some experiments in this direction, and find that by 

 placing a tall metal chimney over a very smoky paraffin lamp, surrounding 

 this chimney with another tube slightly larger, and causing the products of 

 combustion to rise up the centre tube, and descend through the annular 

 space between the two tubes, the soot is all taken out, and nothing but a white 

 vapour is see a escaping. On examining the tubes after they have been in use 

 some time, the inside surface of the inner one is found to be slightly coated 

 with soot, while its outer surface is perfectly clean and bright, not a speck of 

 dust on it, and the inside of the outer tube, which is only a short distance from 

 it, is thickly coated with soot. This arrangement, however, is too complicated, 

 save for special purposes. 



It has been already stated that the reason why so much soot collects in 

 chimneys is that the gases are hotter than the sides of the chimney. In cases 

 where the gases are allowed to escape at a high temperature advantage might 

 be taken of this tendency. If we simply cooled the smoke in the presence of 

 plenty of depositing surface, much of its soot would be trapped out, and the 

 escaping smoke made less dense. The amount that might be trapped in this 

 way will depend on the extent to which the gases could be cooled. 



For works with large chimneys this plan evidently could not be adopted, 

 and in their case the purification would require to be down at the bottom of the 



VOL. XXXII. PAET II. 2 Y 



