164 



Mr. J. B. N. Hennessey. 



[Dec. 16, 



Cause (&.) In 1869, mean of sun's meridional zenith 



distances (at Mussooree) 44 0, 6 



In 1879, mean of sun's meridional zenith 



distances (at Mussooree) 47°'3 



Difference 2°7 



hence, to make the results of 1869 strictly comparable with those of 

 1879, the former (or Dj) must be increased for (a) and decreased for 

 (b). We might estimate these corrections empirically, proceeding on 

 the basis afforded in Table VIII of the defect of radiation correspond- 

 ing to + 1 hour to + 2°"1 of sun's zenith distance. It appears, 

 however, sufficient to note that the effects of (a) and (b) are of con- 

 trary signs, and that the excess of (b) may be set down roughly at 

 about five units, whence D r would be reduced to about twenty, of 

 which the principal portion, if not the whole, may be ascribed to 

 intrinsic or residual variation. Hence, it may be asserted with 

 confidence, on the evidence adduced, that the solar or residual radia- 

 tion in 1879 was less than in 1869 by so much as 2 per cent. If we 

 imagine this result to hold good for the twelve months, some notion 

 may be received of its consequences by conceiving that, in point 

 of heat- rays at least, the sun was- abolished one entire week in 1879. 



[To this may be added, that since the larger radiation of 1869 cor- 

 responds more nearly than the smaller radiation of 1879 to a period of 

 maximum sun-spots, this result confirms, so far as the evidence goes, 

 the hypothesis of greater solar energy at maximum than at minimum 

 epochs.] 



19. It will be seen in the column of mean results for the day, 

 Table VII, that considerable differences appear on a comparison 

 inter se ; at the same time that the several individual results for the day 

 (in Table V), of which the means are composed, exhibit comparative 

 harmony. This leads to the conclusion that all the results for each 

 day contain one or more constants ; but it is by no means a simple 

 matter to show by experiment* how these constants may be sub- 

 divided between intrinsic or residual and instrumental causes ; as to 

 local causes, they may be disregarded at least at Mussooree. That 

 instrumental causes do exist I am strongly inclined to believe, though 

 probably they eliminate one another in the mean from a large 

 number of days ; but supposing them to be real, these causes are in all 

 likelihood connected with the preliminary adjustment of the fluid, by 

 which a sufficient quantity of the latter is " cast off " into the surplus 

 bulb above, so as to make the head of the column play up and down 



* Some experiments have been made, but they are not sufficiently numerous to 

 justify a conclusion. 



