Mr. \V. 

 Table H. Octant Observations at Milverton, Somerset, 1890. 



On the 13th October, 1890, I made a series of octant observations ; 

 but, as I donbted whether the sky was really clear (t.e , as the clear 

 sky of Dacca in the cold weather), I made a series of sunstrips, 

 under : 



Table J. 



In the first four observations, the sky was apparently, and dont 

 less really, clear ; in the three latter observations, some slight 

 invisible cloud over the Sun produced great changes in the sunstrips. 



From the exact coincidence in the readings in the four first obser- 

 vations, at Dacca and Milverton, I think it follows (1) that there 

 was no material difference in my candles at Dacca and Milverton ; 

 (2) that the chemical action of the Sun at the same* altitudes was the 

 same at Dacca and Milverton. 



It is also clear that the number of really fine hours of sky in 

 England (i.e., when it can be compared with the Dacca cold- weather 

 sky) is very small perhaps not a score in the year. And further 

 that, in a great many apparently clear skies in England, there is 



