14 J. A1TKEN ON THE NUMBER OF DUST PARTICLES IN THE ATMOSPHERE. 



returned to the receiver. In a short time after, when the air is again saturated, the 

 process is repeated, the tall vessel lowered, 2 c.c. of air drawn out, and a shower of fine 

 particles produced in the receiver. 



Eepeating this process a number of times, it was found that on each succeeding 

 expansion the denseness of the condensation became less and less. At first the particles 

 were numerous, very small, and fell very slowly, then after a few expansions they decreased 

 in number, while they increased in size, and fell more quickly. After a considerable 

 number of expansions had been made, condensation entirely ceased, not a drop being- 

 seen to fall. When this condition was attained, the tall vessel of water under the burette 

 was lowered so as to cause an increased expansion. The vessel was lowered so much 

 that on the burette stopcock being opened the water in it fell 5 c.c, arrangements for 

 this having been previously made. When the stopcock was now opened, a dense shower 

 of drops made its appearance in the receiver. An expansion of 5 c.c. thus produced 

 condensation in air which had no nuclei for an expansion of 2 c.c. Continuing the test, 

 the vessel of water was raised, and the 5 c.c. of air returned to the receiver, the stopcock 

 shut, and again opened after a time, when another but less dense shower fell. After this 

 was repeated a few times, no condensation took place with an expansion of 5 c.c. The tall 

 vessel of water was then lowered still further, this time to an extent to cause the water 

 in the burette to fall 10 c.c. When this was done, another shower made its appearance, 

 but not nearly so dense as the first one which fell when the expansion was increased to 

 5 c.c. After a very few 10 c.c. expansions, the condensation stopped. After this con- 

 dition of matters was attained, an expansion of 150 c.c. was made by means of one stroke 

 of the air-pump. When this was done, scarcely a drop made its appearance. 



We see from this experiment that a very slight degree of supersaturation will cause 

 condensation on some of the dust particles in the air, but that the degree of supersatura- 

 tion which is sufficient to cause some particles to become active centres, is yet insufficient 

 to cause condensation to take place on others. An expansion of 2su produced a super- 

 saturation sufficient to cause condensation on more than one half of the particles, but it 

 required a higher degree of supersaturation to produce condensation on the others, many 

 of them requiring the supersaturation produced by an expansion of ^ before vapour 

 condensed on them. 



It may be concluded from this experiment that we have here distinct evidence that the 

 condensing powers of the dust particles is affected by their size No doubt the com- 

 position of the particles has an effect on the degree of supersaturation necessary to make 

 even the small particles active, yet the affinity between the material of the particle and 

 water vapour does not satisfactorily account for the different degrees of supersaturation 

 required to make the different particles active. Though a small particle which has an 

 affinity for vapour may have the same condensing power as a larger particle which has 

 no affinity, yet the affinity alone does not explain the necessity for supersaturation to 

 make any of them active ; this, so far as we know, can be explained only by the sizes. 



We see from this experiment that an expansion of ^ is nearly, perhaps quite sufficient 



