8-10 Cluis. Harding — Memorandum on the diurnal Variation of [No. 4, 



Books" and have dealt with the observations obtained from that source 

 in a manner which as far as I can see is open to very little objection, and 

 which is probably the best possible method under the circumstances. 



I have meaned all the noon observations for an absolute reading, and 

 for the means of the other 4-hourly periods have worked on the absolute 

 differences which are obtainable from the various observations recorded : for 

 example, working all the obtainable differences of 4 p. m. observations on 

 noon observations for the same day, and finally meaning these differences, 

 and aj^plying the mean difference to the noon mean already obtained for a 

 4 p. M. mean : — so on for all the other 4-hourly periods. The method has 

 been to take noon as a centre and work on differences at each side of it. 

 The midnight observation has thus been obtained, by working backwards 

 and forwards from noon, and the difference of the two results is only 

 •002 in. 



The results for the 4-hourly j^eriods are as follows, using all the sea 

 data between 20° and 22" N. and 80° to 90° E. 

 4 am. 8 a. m. Noon. 4 p. m. 

 29-996 30-056 30041 29987 



8 p. M. 

 30029 



—•026 



+ •034 +-019 



--035 



+ •007 



Midt. Mean. 



30 024 30 022 



+ ^^"^ iDaily M. 



I have computed Bessel's constants from the 4-hourly values, and the 

 expression is as follows : — 

 sc = 30-022 + -0136 sin (?^. 15° + 320° 5') + 0311 sin (n. 30° + 160° 42') 



+ •0002 sin (?i. 45° + 90°). 

 From the constants, the following hourly values for the sea observations have 

 been computed : they are given side by side with the hourly values of sea 

 observations deduced by Mr. Blanford and with those of Calcutta. 



Hour. 



Calcutta. 



Sea [C. H.] 



Sea [Mr. Bl 





30011 



30022 



30022 



Midt. • 



+ -003 



+ -002 



+ ^016 



1 A. M. 



— •006 



— •Oil 



— •ooi 



2 „ 



— -015 



— ^023 



— •021 



3 „ 



— 021 



— ^028 



— •037 



4 „ 



— -024 



— ^026 



— •042 



5 „ 



— •020 



— ^016 



— ^033 



(3 „ 



— •005 



•000 



— 013 



7 „ 



+ ^021 



+ -018 



+ -010 



8 „ 



+ •oso 



+ 034 



+ ^028 



9 „ 



+ ^072 



+ ^043 



+ ^037 



10 „ 



+ -078 



+ ^043 



+ -035 



11 „ 



+ -062 



+ ^035 



+ -024 



