J. AITKEN ON THE NUMBER OF DUST PARTICLES IN THE ATMOSPHERE. 13 



condensation can take place on it ; but a high degree of supersaturation is not possible 

 while the larger particles are present, and it is not till the air is freed from them that the 

 higher degree of supersaturation necessary to cause condensation on the smaller ones is 

 possible. 



In order to determine whether the showers observed with the high degree of super- 

 saturation used in the tests were due to the presence of extremely small particles, it 

 was necessary for us experimentally to determine whether the size of the particles has 

 practically any influence on the degree of supersaturation required to cause them to 

 become centres of condensation. 



In order to get an answer to this question, the apparatus was arranged very much as 

 described for testing the filtering powers of cotton wool, that is with the test receiver A, 

 connected on the one side with the air-pump, and on the other with the tube E, through 

 which air could be drawn either from the filter or the gasometer. Through the india- 

 rubber stopper in the receiver A was passed a third tube, to the lower end of which was 

 attached a burette provided with a stopcock. The burette was fixed in an inverted 

 position to the projecting end of the tube by means of a small piece of india-rubber 

 tubing. The burette dipped into a very tall vessel full of water. This vessel was so 

 arranged that it could be easily raised and lowered, or fixed at any desired height. The 

 burette and vessel of water were used as an air-pump for producing small and known 

 amounts of expansion of the air in the receiver. The ordinary air-pump could not be used 

 for this purpose, as it is not suitable for making small and correctly measured amounts 

 of expansion, and also because, in this experiment, the air which is taken out of the 

 receiver at each expansion has to be returned to it, as no air from the outside can be 

 admitted during the test. 



Supposing the apparatus arranged as described, the following is the manner adopted 

 for testing the condensing powers of the different particles : — The air in the receiver and 

 connecting tubes is first purified by pumping air through the filter. All connection is 

 then cut off with the outside air by closing the stopcock F between the receiver and the 

 filter. No stopcock is necessary on the exit tube, as the air-pump stops the passage of 

 air in that direction. The burette stopcock is then opened, and the tall vessel of water is 

 lowered till the water in the burette falls 2 c.c. The support under the vessel is then 

 fixed at this level, after which the vessel is again raised to its original height and kept 

 there till the water in the burette has recovered its original level. The burette stopcock 

 is now closed, and the vessel lowered to its previous position. The apparatus being thus 

 adjusted, the air to be tested is drawn into the receiver through the pipe E from the 

 gasometer. After a short time, to allow any change of volume to adjust itself, the con- 

 nection between the receiver and the gasometer is cut off, and the testing begun by 

 opening the burette stopcock, when the water in it at once falls, and draws out 2 c.c. of 

 air from the receiver. This expansion gives rise to condensation, the small fog particles 

 being easily seen with the aid of a magnifying glass. After sufficient time has been 

 given for the particles to settle, the tall vessel of water is raised, and the 2 c.c. of air are 



