PARTICLES IN FOGS AND CLOUDS. 425 



melted at the time of observation. The instrument for observing them was not held in 

 the hand as usual, but was fitted to a support in the open air, and left for a considerable 

 time to cool, its condition being tested by placing a drop of water on the mounting of 

 the micrometer. As this rapidly froze, and remained frozen, there does not seem to be 

 any probability that the particles were thawed by their approach to the instrument or to 

 the observer. 



There are theoretical considerations pointing to the conclusion that fog particles will 

 not freeze till cooled far below the freezing-point, and 1 have shown in previous papers 

 that with artificially-made fog particles, even though cooled to 6° Fahr. before being further 

 cooled by expansion, there was not the slightest indication of freezing. But though both 

 theory and experiment pointed to this conclusion, yet I thought it as well to give the 

 result of observations made on the particles under the conditions existing in nature. 



This liquid condition of the condensed particles in our atmosphere, at temperatures 

 below the freezing-point, suggests the conditions under which the large and beautifully 

 formed crystals of snow are grown. These beautiful flakes, with their regular angles 

 and perfect uniformity, are evidently not aggregations of small frozen particles in the 

 way we may suppose a rain-drop to be built up of unfrozen drops. From their 

 extreme regularity and perfection, one would naturally expect they were formed under 

 conditions free from restraint, and with an abundant supply of nourishing vaporous 

 material on which to grow. Now, a cloudy condensation of liquid particles below 

 the freezing-point would give the conditions which seem most favourable for the growth 

 of large and regular crystals. Suppose some particles in the cloud, from some so-called 

 accident of size or other cause, to freeze before the others. At once these solid 

 particles are in a position to rob the liquid ones of their moisture, and becoming heavy 

 they begin to fall in what is to them a supersaturated atmosphere, and they soon grow in 

 perfect form to some size. 



If the above supposition be correct, it is possible, when the snow-flakes are small and 

 irregular, that they are formed by the aggregation of very small frozen particles due to 

 a very low temperature. From this some might conclude that large regular snow-flakes 

 could be formed only when the temperature was not very low. Any conclusion, how- 

 ever, on this point is evidently premature — 1st, because we do not know to how low a 

 temperature extremely small cloud particles may be cooled without freezing ; and, 2nd, 

 any observations, at present at our disposal, that have been made on snow-flakes observed 

 to fall at different temperatures, are of little value, as the temperatures were taken at 

 the surface of the earth, and not in the cloud itself, which in all probability would be at 

 a very different temperature from that of the air low down. 



VOL. XXXVII. PART II. (NO. 20). 3 S 



