166 FRAGMENTS OF SCIENCE. 



which we can throw obvious light in London. Causing 

 our luminous beam to pass through a large flask filled 

 with the air of this room, and charged with its germs 

 and its dust, the beam is seen crossing the flask from 

 side to side. But here is another similar flask, which 

 cuts a clear gap out of the beam. It is filled with un- 

 filtered air, and still no trace of the beam is visible. 

 Why ? By pure accident I stumbled on this flask in 

 our apparatus room, where it had remained quiet for 

 some time. Acting upon this obvious suggestion I set 

 aside three other flasks, filled, in the first instance, with 

 mote-laden air. They are now optically empty. Our 

 former experiments proved that the life-producing 

 particles attach themselves to the fibres of cotton-wool. 

 In the present experiment the motes have been brought 

 by gentle air-currents, established by slight differences 

 of temperature within our closed vessels, into contact 

 with the interior surface, to which they adhere. The 

 air of these flasks has deposited its dust, germs and all, 

 and is practically free from suspended matter. 



I had a chamber erected, the lower half of which is 

 of w<~>od, its upper half being enclosed by four glazed 

 window-frames. It tapers to a truncated cone at the 

 top. It measures in plan 3 ft. by 2 ft. 6 in., and its 

 height is 5 ft. 10 in. On February 6 it was closed, 

 every crevice that could admit dust, or cause displace- 

 ment of the air, being carefully pasted over with paper. 

 The electric beam at first revealed the dust within the 

 chamber as it did in the air of the laboratory. The 

 chamber was examined almost daily ; a perceptible 

 diminution of the floating matter being noticed as time 

 advanced. At the end of a week the chamber was 

 optically empty, exhibiting no trace of matter competent 

 to scatter the light. Such must have been the case in 

 the stagnant caves of the Paris Observatory. Were our 



