156 



Mr. J. B. N. Hennessey. 



[Dec. 16, 



claim cannot be advanced for Delira. Farther, it may be objected 

 that the distance between the two stations is not sufficiently great, 

 being only some nine miles. To this can be urged, that in view of the 

 latitude and the sun's declination, the direct rays to the sun from the 

 two stations, about noon, travelled nearly seven miles apart from one 

 another through the envelope of the earth's atmosphere : whether this 

 be sufficient or not, is perhaps not so readily apparent. But as a 

 matter of fact, the time at our disposal did not permit of our observing 

 anywhere but at Mussooree and Dehra, so that the choice lay between 

 accepting Dehra or dispensing with a second station. Notwithstanding 

 the drawbacks to Dehra individually, its results appear less liable to 

 distrust if in accord than if in conflict with those at Mussooree : for 

 the latter case visible causes are not wanting, but the former is difficult 

 to account for without assuming special conditions. To this may be 

 added, that distinct mutual visibility between the stations, and the 

 ability at Mussooree to look nearly a mile above Dehra, are not without 

 advantages.] 



6. The observations with B at Dehra were taken by Mr. W. H, 

 Cole, M.A., those with A at Mussooree and Dehra by myself.* The 

 procedure prescribed in the " Admiralty Manual of Scientific Enquiry,' 5 

 pp. 129 — 130, was exactly followed (as was done in 1869), whereby 

 the change of reading in 60 seconds is obtained alternately in sun (©) 

 and shade ( x ) at intervals of 30 seconds (beginning and ending with 

 a O observation ; f hence in a series of measurements there result 

 (71 + I) of the latter to n ef X . Ordinarily the average result from a 

 series is found from 



n+L n 



where R stands for the mean radiation, and the brackets [ ] denote 

 summation. We may, however, exhibit the successive values of radia- 

 tion for 60 seconds by writing 



1 r(O, + 2+ x 1 ) + (Ol+© 2+ X 3 )+ . . +(©£±©211+ x»)] 



n 



l ience R-B 1 =l ( Oi + ®±ti_ _JL [O] »+i ) ■ 



n I 2 n+l J 



This difference is very small, especially if the series be long continued, 

 so that n is large. Accepting the mean value Rj., its individual results 

 may be conveniently reckoned as if obtained at the mean of the begins 

 ning and ending times of successive shade observations ; thus the pth 

 result, or 



# Excepting on 1st and 2nd November, when Mr. H. W. Peychers observed, 

 f See Table IX for example. 



X Corrections to 32° Fahrenheit being understood. 



