SOLID AND LIQUID PARTICLES IN CLOUDS. 319 



special instruments, agrees with the result of our last observations made on the 

 Rigi. 



While on the Rigi this year I tried to find whether the density of cloudy condensa- 

 tion depends on the number of dust particles present, or on the number of water particles. 

 The question was, however, much more difficult of solution than I expected, owing to 

 the great rapidity with which the constituent parts of the passing clouds changed. To 

 have answered the question satisfactorily in the case of the clouds tested, three observers 

 would have been required, one observer noting the density of the cloud, one counting the 

 dust particles, and the third observing the water particles, — all three making their 

 observations at the same moment. I had, however, to depend on my observations alone, 

 and so far as they go they point to the conclusion that the density of a cloud depends 

 principally on the number of w T ater particles. Wherever the water particles fell at the 

 rate of about 100 drops per square mm. per minute, the limit of visibility in the cloud 

 was about 30 yards, and as the limit of visibility increased the rate of fall decreased. 



There were too few opportunities of testing the effect of the number of particles of 

 dust on the density of the air in a cloud, as on most occasions the numbers were far too 

 variable to offer satisfactory results. But on comparing the density of a fog and a cloud 

 when the same number of drops fell, it would appear that the number of dust particles 

 has a much smaller effect than the number of water particles. For instance, in a fog 

 last winter, as stated in a previous paper, drops were observed to fall at the rate of 

 30 per minute per square mm., and the limit of visibility at the time was about 100 

 yards. Now this is not very far from the limit of visibility observed on the Rigi when 

 the rate of fall was the same. But on the Rigi there were only a few thousand dust 

 particles per c.c, while in the fog there were about 50,000. It would thus appear that in 

 cloudy condensation the thickness depends chiefly on the number of water particles, 

 and only in a secondary way on the number of dust particles. The observations are, 

 however, as yet too few to warrant a definite conclusion. 



Although in all the cases of foggy or cloudy condensation investigated by me the air 

 was saturated, though it may have appeared to be dry, I do not, however, wish it to be 

 understood that there is no such thing as a dry fog, only that in my experience I have 

 not observed one. 



We see from the observations made with the fog-counter, that whenever a cloud is 

 formed, it at once begins to rain, and the drops shower down in immense numbers, 

 though small in size. These drops fall into the air under the cloud, where they 

 evaporate if the air is dry, and the distance they fall will depend on their size 

 and the dryness of the air underneath. So that on a summer day, with white clouds 

 passing overhead, it is really raining, but the drops being very small, they evaporate in 

 the air under the cloud long before they reach the earth. It seems probable, therefore, 

 that much of the melting of clouds is produced in this way, the particles falling from the 

 saturated air in which they were formed and dissolving in the drier air underneath. 



