1856.] Report on the Magnetic Survey. 21 



At heights of from 10,000 to 12,000 feet, we got nearly always 10 

 (the last number of the scale) if the papers were exposed twelve 

 hours ; this allowed us to take a shorter time of exposition and to 

 shew the variation at intervals of 3 hours during the day. 



Electricity. 



The most violent electric discharges take place immediately 

 before the rains or at their first beginning. 



In May I found the electricity on isolated peaks of 10,000 to 

 12,000 feet, in clear days, five times greater than in the plains. 



Optical phenomena of the atmosphere. 



The plains, as well as the mountains of Sikkim offered a great 

 variety of interesting optical phenomena, of which the following 

 may be mentioned in a few words. 



The blue colour of the sky is in the plains of India much darker 

 than in higher (Northern or Southern) latitudes, but the darkness 

 of the sky does not increase with the height in the same ratio as 

 in Europe. At heights of 10,000 to 12,000 feet, the absolute bright- 

 ness of the sky is even greater than at the same height in Europe 

 between 45° and 47° of North latitude. 



The highest temperature we observed on a black bulb thermo- 

 meter lying on black wool was 74° C. = 165° Eht, July, Darjiling, 

 7,200 ft. English. 



During our stay at a greater height, we had never an entirely 

 clear insolation after llf a. m. 



The second colouration of the snow after sunset had not been 

 hitherto observed in tropical climates (see Humboldt's Cosmos? 

 vol. IY). 



We had some difficulty in observing this phenomenon, since at sun- 

 set it is generally very foggy in Sikkim ; but on two occasions, parti- 

 cularly June 2nd from Phulloot, it was as plainly visible and as well 

 defined as I ever saw it in the Alps. Besides this, I was told by 

 Dr. Campbell, that after the rainy season it is very often to be seen 

 extending over all the snowy peaks, and visible a considerable time 

 after sunset. 



The chemical action of light, determined as formerly described in 

 our "new researches in Alps," was found in maximo to be number 



