1881.] H. F. Blanf ord — Relations of Cloud and Rainfall to Temperature. 75 



but which, being of general and not merely temporary interest, the author 

 had thought desirable to recast as the subjects of' a special paper for the 

 Society. 



The first of these was a discussion of the circumstances which mainly 

 determine those marked variations of temperature, that characterize the 

 corresponding seasons of different years in India. Adverting to a paper 

 recently communicated to ' Nature' by Mr. Douglas Archibald, in which 

 Mr. Archibald had shown some reasons for inferring that the cyclical 

 variations of rock-temperature, brought to light by Prof. Piazzi Smythe, 

 were determined by corresponding cyclical variations of cloudiness in the 

 atmosphere, the author remarked that this view was in part identical with 

 that which he had originally suggested in a paper read before the Society 

 in June 1875 ; wherein he had endeavoured to show that the temperature of 

 the lower atmosphere on the land surface, in India, depends more on the 

 quantity of cloud and rainfall, than on variations of the solar intensity. 

 Evidence bearing on this subject, drawn from the meteorological observa- 

 tions of the last 6 years, was then brought forward. The temperature of the 

 air and ground as observed at Calcutta in the first five months of 1879 was 

 contrasted with the corresponding temperatures of 1880, and the difference 

 shown to accompany marked variations in the cloud and rainfall. Similar 

 evidence was obtained when the temperature of the N. W. Provinces in the 

 hot months and rains respectively of the years 1877 and 1879 were compared, 

 the variations of the two seasons being in opposite directions in the two 

 years compared ; and these were shown, in like manner, to have accompanied 

 very striking variations in the cloud proportion and rainfall. 



The action of the different agencies which most powerfully influence 

 temperature at different seasons was discussed briefly, with the result that, 

 only in November and December, is the effect of cloud to raise the tempera- 

 ture above the normal value. At all other seasons, the effect is the opposite, 

 and hence abnormally cloudy and rainy years are abnormally cool years. 



The second subject discussed was the variations in the density of the 

 lower and higher strata of the atmosphere, as shown by a comparison of 

 the barometric pressures at hill stations with those on the plains. It was 

 shown that, in many cases, the density of the lower strata of the atmosphere 

 was below the average, when the barometric pressure as a whole was in 

 excess of the average, and vice versa ; indicating that the higher strata must 

 have an anomaly of the opposite character to that of the lower. It was 

 also shown that a similar opposition of conditions is of annual recurrence 

 at the setting in of the rains, and that it might therefore he probably 

 traceable to some definite play of physical causes. Then adverting to a 

 former discussion, in which it had been shown that the principal cause 



