FORMATION OF SMALL CLEAR SPACES IN DUSTY AIR. 259 



cells. This tendency of the dust in the residual air to settle is increased by the 

 load of water deposited on it by the moist air. 



The amount to which our lungs are protected by heat and evaporation can 

 scarcely be solved in a physical laboratory, and will be best determined by 

 anatomical examinations of lungs which have lived under different conditions of 

 temperature and moisture. 



A Thermic Filter. 



Having observed that the dust particles tended to move away from hot 

 bodies and to attach themselves to cold ones, I made some experiments 

 on the subject to study the movements of dust particles when placed between 

 hot and cold surfaces. Most interesting results were obtained by placing near 

 the hot platinum wires, already referred to, a piece of glass or a plate of metal, 

 and getting the dust deposited upon it. One arrangement of the experiment 

 is to place the glass with its plane vertical and transversely over the wires, 

 at such a height that its lower edge almost touches the wires, and fill the 

 box with dust by blowing up some calcined magnesia or other fine powder. 

 After all the currents have settled, and while the air is still full of dust, the 

 electric current is turned on and the wire heated. A well-marked dark plane 

 at once rises over the wire, and in its upward passage it is cut transversely 

 by the glass plate. After the plate has been left for some time with the air 

 current streaming over its surface, it is found to have a very beautiful impres- 

 sion of the dark plane imprinted on it. The warm air, in streaming upwards 

 over the surface of the glass, deposited its dust on it, and the fact of there being 

 no dust in the dark plane is recorded by a well-defined line of clear glass, 

 the deposit of dust on each side of the clean line being thickest just along 

 the edge, and thinning away on each side. These impressions of the dark plane 

 may be made permanent by causing the dust to be deposited on a plate newly 

 coated with black varnish, and used while the varnish is still soft. 



Tt is not necessary to put anything on the surface of the glass to cause the 

 dust to adhere, as it attaches itself to a clean surface of glass with considerable 

 firmness, but some adhesive substance on the plate enables the impression to 

 stand rougher treatment. Impressions of the dark plane have also been made 

 with charcoal dust deposited on opal glass. These black impressions are, of 

 course, " negatives " of the magnesia ones, the plane in the former case being 

 white, surrounded by black dust. The charcoal dust was securely fixed by first 

 coating the glass with a thin solution of gum, which was dried before the 

 dust was deposited on it, and the dust fixed by breathing on the surface. 



If in place of putting a plate vertically over the wires, we place two plates 

 vertically — one at each side of the wire — we then get the dust deposited on the 

 plates, thickest opposite to the wires and thinner higher up. Arrangements 



