44 Jolm Eliot — On the occasional Inversion of the Temperature [No. 1, 



In the preceding discussion it has been shewn that the temperature 

 conditions and changes at the hill stations are usually different from 

 those of the plain stations. For example, ordinary anti-cyclonic weather 

 gives increased day and night temperature at the hill stations, and hence 

 increases the mean temperature and only affects the diurnal range very 

 slightly, whilst in the plains it gives increased day and decreased night 

 temperature, and hence increases very largely the diurnal range of tem- 

 perature, whilst it only slightly affects the mean daily temperature. Again 

 stormy weather in the mountain districts of Northern India gives de- 

 creased day and night temperature and hence a much lower mean temper- 

 ature than usual with little change in the diurnal range of temperature. 

 The same weather in the plains gives decreased day and increased night 

 temperature, and hence the diurnal range of temperature is largely 

 diminished, whilst the mean temperature is very slightly affected. Hence 

 the important conclusions, 



1st. — That the chief weather changes and conditions in Northern 

 India during the cold weather affect the temperature in 

 entirely different ways in the plains and hills. In the 

 former they modify the diurnal range of temperature chiefly 

 and in the latter the daily mean temperature. 

 2nd. — That the monthly means of temperature or of daily range of 

 temperature are in consequence not comparable for the hills 

 and plain stations, and that similar variations from the nor- 

 mal imply different conditions and actions in the two cases. 

 3rd. — Hence the nature and causes of these changes and variations 

 of the vertical temperature relations cannot be properly 

 estimated and investigated by comparing monthly means, 

 but by comparison of the actual temperature conditions 

 prevailing in each particular state or type of weather. 

 Hence typical cases have been selected in tbe previous portion of 

 the paper and the same principle is adopted throughout. 



We are now in a position to give a simple explanation of the high 

 night temperatures at the hill stations observed during fine clear 

 weather in December and January. 



In ordinary anti-cyclonic weather in January in the Punjab plains 

 the temperature ranges from an average maximum of 72° to an average 

 minimum of 36°, giving thus a mean diurnal range at such periods of 

 36°. The hill stations in Upper India are at an elevation of about 7000 

 feet above the sea or 6000 feet higher than the neighbouring plain 

 stations. The rapid increase of temperature in the plains during the 

 morning gives rise almost entirely to convection currents. As the air 

 is very dry, it may be assumed that in rising and expanding it will cool 



