282 
]\IR. C. T. R. WILSON ON CONDENSATION OF WATER VAPOUR 
Tlie apparatus was charged with a fresh sujaply of air each day. 
The results of all the measurements made with the small apparatus on the rain- 
like condensation are o;iven in the table. 
The expansion can be measured with quite as great accuracy as in the larger 
apparatus, but the whole number of drops when condensation first begins is incon¬ 
veniently small in the small apparatus. This makes the measurements of the critical 
value of more troublesome. 
Each determination of this requires a large number of observations to be made, 
being always made less or greater according as the pi’evious expansion had brought 
down a shower or not. In this way it was possible, finally, to obtain the least value 
of necessary in order that a shower should result within about 2 millims. 
It will again be noticed that the result obtained is with the limits of experimental 
error independent of the temperature between 15'5° and 18'5° C. 
We may summarise the results of the measurements made with both machines 
upon the expansion required for rain-like condensation in the following statement. 
When saturated air free from all extraneous nuclei is suddenly expanded, rain-like 
condensation takes place if the ratio of the final to the initial volume exceeds 1’252. 
This is true if the initial temperature is between 15° and 28° C. 
It will be noticed that this expansion is less than that used by Aitken in the 
experiments already referred to. The difference between the result of making the 
stroke of his pump slow or quick was therefore evidently due to the expansion not 
being quick enough to give the theoretical lowering of temperature. 
Phenomena ohserred ivith Expansions greater than that re<iuired to produce Rain- 
lihe Condensation. 
If a series of expansions be made of constantly increasing amount, the following 
phenomena are observed. 
The drops are, if the expansion be only slightly greater than is sufficient to pro¬ 
duce condensation, only few in number. More of them are seen if the expansion be 
somewhat greater, but even when is as great as 1‘37, the condensation still 
takes the form of a shower of drops, which settle within a few seconds. To the eye 
there is no marked difference in the density of the shower over quite a wide range 
of expansions. 
If, however, Uj/rq be increased from 1'37 to 1’38, the increase in the number of 
drops is so great that there is no longer any resemblance to a shower of rain, but a 
fog results, taking a minute or more to settle. 
* [Note added July 2'2nd, 1897.—When expan.sion results in a fog, it is of course necessary to get rid 
of all traces of it before proceeding to a fresh observation. This was done by repeated expansions of 
moderate amount, as in the removal of the original dost particles.] 
