J. AITKEN ON THE NUMBER OF DUST PARTICLES IN THE ATMOSPHERE. 19 



according to certain conditions, such as the time during which the gas has been burning 

 in the room, manner of collecting air from the Bunsen flame, &c. 



Many will perhaps think that the numbers of particles as given in the table are far 

 too high, and that there must be a mistake somewhere. It may be mentioned that the 

 evidence in support of the correctness of our conclusions is similar to that of the chemist's, 

 when he puts, say, a solution of sodium chloride into a solution of silver nitrate, and from 

 the weight of precipitated silver chloride tells the amount of silver that was in the 

 solution. If there had been no silver in the solution, then there would have been no 

 precipitate ; and the amount of precipitate is in proportion to the quantity of silver 

 present. So it is in testing for dust. If there is no dust in the air, no drops are thrown 

 down, and the number of drops is determined by the number of particles present, for each 

 drop has a dust nucleus. 



I have not yet entered carefully into the consideration of the exact value of the figures 

 in the table. While the described method of counting the particles gives a very fair 

 estimate of the number in the air when it arrives at the test receiver, and the numbers 

 entered in the table are the numbers calculated from the particles counted there, no 

 allowance has been made for the dust which may have settled in the gasometer and con- 

 necting pipes. The probability, therefore, is that the figures are rather under than over 

 estimates of the particles in the air tested. As showing the tendency the dust has to 

 settle in the apparatus, it may be mentioned that about one half of the dust particles 

 settled out of the air, if it is allowed to remain in the gasometer for one hour. In the 

 development of the apparatus for this investigation, this is one of the points to which 

 special attention has been given, and a method has been devised by which a very short 

 time will be given for settling before the particles are counted. It is possible the dust 

 might settle more slowly if the vessel containing it were dry, and mercury used instead 

 of water for displacing it. This, however, with many other points, remains for considera- 

 tion. It may be as well to note here, that the dust settles out of the air not only on 

 account of its weight causing it to fall, but very much is settled out by difference of tem- 

 perature at the different parts of the vessel. Wherever the air meets a surface colder 

 than itself it parts with some of its dust to that surface, for a reason which has been 

 explained in a previous communication. 



Most of us were quite prepared to find that whenever the dust particles in our atmo- 

 sphere were counted, the figure would be a very high one ; but I imagine that the figures 

 here given are such as few, if any expected, and they increase our admiration of the 

 Old Testament writers' selection of dust as a type of the infinitely small and the 

 innumerable. 



