464 MR MILNE ON THE DRYING UP OF THE 



perceive how the frost on the night of the 26th November, acted so as to produce 

 this effect. It is stated in the letters above quoted or referred to, that, during that 

 night, there was a strong gale from the east ; a very unusual circumstance when 

 the temperature is so low as 20°. Now, it is well known that cold, when accom- 

 panied by wind, acts very differently than when the air is calm. Bodies exposed 

 to a refrigerating breeze, will be cooled much more rapidly, than when exposed to 

 air of the same temperature in a state of repose. The rate of cooling is, in fact, 

 increased, exactly in proportion to the velocity of the wind. Professor Leslie 

 has shewn, that, whatever is the ordinary rate of cooling of a body over which a 

 stream of colder air of a certain velocity is blowing, this rate of cooling will be 

 doubled if the velocity of the wind be doubled, and quadrupled if the wind blow 

 with four times its previous violence ; or, in other words, it will be reduced to 

 the temperature of the air in one-half or one-fourth of the time which it would 

 otherwise require. " We thence gather (says Sir John Leslie), that even a mo- 

 derate wind will quadruple the waste of heat, and that a vehement hurricane is 

 capable of increasing the rate of dissipation perhaps fifteen or twenty times. 

 Hence also the keen impression of frosty winds on our feelings, and their prodi- 

 gious effects in chilling the surface of the ground. We thus perceive in a strong 

 light the vast utility of shelter."* 



The effect, therefore, of the wind on the night of the 26th, must have been 

 to cool down to its own temperature, the objects and places exposed to it with 

 great rapidity. If, at 10 p. m., the temperature of this wind was 26°, we may 

 assume that, by this time, the surface of the ground which it swept over in fuU 

 force, had been cooled down to at least 29°, — a degree of cold at which water 

 in drains, streamlets, morasses, and shallow pools, would be entirely consoli- 

 dated. But it is obvious that this effect would, by that time, have been pro- 

 duced only in exposed or unsheltered places. In the glens and valleys, especially 

 those which were wooded, the surface would not be so rapidly cooled ; and if 

 they lay in a north and south direction, there would be additional means of 

 shelter. In proof of this observation, if any were wanting, I might refer to the 

 state of the thermometer at Drumlanrig (in a low sheltered district), and at Lead- 

 hills, situated about 1000 feet higher. On the night of the 26th the thermometer 

 at LeadhiUs was 27°, at Drumlanrig 34°. On Tuesday morning, at Leadhills it 

 was 29°, at Drumlanrig 30° ; at noon, at Leadhills 31°, at Drumlanrig 35°. 



We thus see two distinct and separate reasons wh}'^ the rivers should, on the 

 night of the 26th November, have had their sources frozen up and arrested, whilst 

 their currents in the main channels continued to flow with scarcely diminished 

 rapidity. In the first place, the mere difference of height subjected the sources 

 to a greater degree of cold, than prevailed in the lower parts of the river's course, 



* Experimental Inquiry into Heat, p. 284. 



