48 John Eliot — On the occasional Inversion of the Temperature [No. 1, 



As there is little or no difference, so far as can be judged, in the 

 radiating power of the earth's surface at Ludhiana, Roorkee, Bareilly, 

 and Grorakhpur as compared with Lahore, Delhi, Agra, and Allahabad 

 in January to account for the greater cooling of the earth's surface and 

 the superincumbent air, it is almost certain that the mass of air descend- 

 ing from the hills is warmed by the action of compression in descending, 

 and that this is one, if not the chief, factor in giving a smaller fall of tem- 

 perature and slightly increased night temperature at the stations nearest 

 to the hills when compared with the more distant ones. Hence it is 

 clear that the descending air does not contribute towards the cooling of 

 the plains of Northern India during the night but actually tends to dimi- 

 nish it. 



The efficient factors in the rapid cooling of the air in the plains of 

 Northern India at such periods are : — 



1st. — Absence of cloud and other conditions favouring rapid radiation 

 from the earth's surface. 



2nd. — Absence of air motion, and more especially of downward 

 convection currents, so that the same mass of air remains 

 in contact with the earth's surface. 



The first ensures the rapid cooling of the earth's surface and the 

 second of the air immediately above the earth's surface. 



A brief explanation will serve for the temperature conditions in 

 Upper India during and after stormy weather. The most important factor 

 appears to be snow fall in the hills and rain in the plain districts. The 

 condensation takes place largely, if not almost entirely, in the upper return 

 current of the north-east monsoon circulation and hence at a great ele- 

 vation. The falling rain and snow carry down with them the temperature 

 of their place of origin and hence tend to cool considerably the whole mass 

 of air through which they fall. The amount of the cooling will evidently 

 depend greatly upon the amount and period of the rain and snowfall. In 

 the hill districts, the temperature falls steadily throughout a long snow 

 storm, and the lowest temperatures are usually recorded just before the 

 weather begins to clear. In the plains, the day temperatui'e falls in con- 

 sequence of cloud and rainfall and the action of rainfall referred to above. 

 But the cloud canopy causes terrestrial radiation to proceed very slowly 

 at night. The effect of the cloud in diminishing radiation is so large that 

 the night temperatures are hence at such periods considerably higher 

 than usual. Hence stormy weather in January and February depresses 

 temperature largely throughout the whole day at the hill stations, and 

 in fact tends to give them a temperature nearly equal to the permanent 

 temperatures of a stratum considerably above their level (i. e. of the stra- 

 tum in which condensation occurs). Whereas in the plains the chief 



