NOTES ON THE CLIMATE OF DIMBULA. 57 



heavy and continuous rain again falls from the north-east, 

 and continues till the middle of November, after which 

 showery weather may be expected till the end of December. 

 January is a very uncertain month, but as a rule heavy 

 showers fall about the middle of the month. After these 

 have ceased little rain can be expected till the burst of the 

 " little south-west" towards the end of April, and after this 

 a month of, on the whole, fine weather brings us back to our 

 starting point. 



The amount of cloud and humidity of the air may be 

 expected to vary directly as the rainfall, and the close cor- 

 respondence between the three comes out very markedly in 

 the subjoined diagrams, which I have added to the observa- 

 tions as making them more easily understood. As will be 

 seen at a glance, all three correspond almost exactly in 

 their rise and fall, corresponding with the increase or 

 decrease of the readings. 



There seems no regularity in the connection between the 

 season and the temperature. If we take the mean between 

 the highest and lowest temperatures as our mean tempera- 

 ture we find that it increases with the northward progress 

 of the sun from January till April, May being stationary 

 at the same point as April . Thence to the end of the year 

 it seems to be independent of the sun, and to vary inversely 

 as the rainfall, in direct opposition to the earlier portion of 

 the year in which it varied directly at it. 



As a rule, the higher the temperature during the day the 

 lower it is at night, owing to radiation. March, April, and 

 May are however exceptions to this, probably from the 

 nearly verticle position of the sun in the case of March and 

 April, and in April again and May from increase of cloud 

 affecting the minimum, while there is little difference in the 

 maximum temperature. 



The readings of the exposed thermometer give a very 



H 



