13 



OX COMPARING AND REDUCING MAGNETIC OBSERVATIONS. 85 



both as observed and after correction for non-cyclic variation, are the 

 elements compared. The sums of hourly inequalities in declination are 

 generally greater, and in horizontal force generally smaller at Kew than 

 at Greenwich. In the comparison of diurnal range attention may be 

 called to the comparatively small magnitude of the values of a — b, in 

 declination, and their general similarity in different months at Greenwich 

 and Kew. The corresponding values of a — 6 in horizontal force are larger 

 at both places, and larger at Greenwich than at Kew, but otherwise do 

 not accoT-d altogether badly. The diurnal range in declination as made 

 cyclic is, on the whole, greater,(column, K — G) at Kew than at Greenwich, 

 on the average by 0''40:=21 of the horizontal force unit of the table, which 

 is precisely the amount by which on the average the diurnal range in 

 horizontal force, as made cyclic, is less at Kew than at Greenwich (column 

 K — G for horizontal force). In this table the monthly sums of inequalities 

 and monthly values of diurnal range are, for both places, necessarily 

 derived from the I'espective monthly diurnal inequalities, but the quarterly 

 and yearly values are formed directly from the monthly values (not from 

 the quarterly and yearly values of diurnal inequality), the Kew numbers, 

 with which the Greenwich numbers are compared, having been so formed. 

 If derived from the quarterly and yearly diurnal inequalities, the numbers 

 would have had a diminished value, since the grouping of monthly in- 

 equalities, in which the points of maximum and minimum vary somewhat 

 in different months, necessarily flattens the curve to some extent. For the 

 purpose of making compai'ison, either method would serve, but it must be 

 similar at both places. The method adopted is the better one of 

 the two. 



Table YII. contains a comparison of the mean non-cyclic increment on 

 quiet days, in declination and horizontal and vertical force, at Greenwich 

 and Kew, founded on the observations of the six years 1890 to 1895. 

 The Kew numbers are those given in Table I. of the 1896 Report. The 

 non-cyclic increment in declination and horizontal force is at the two 

 places of the same character, both as regards variation of sign in declina- 

 tion in different months of the year and in the close correspondence in 

 magnitude in horizontal force throughout the year. The individual 

 months of positive and negative values are also very similar in number. 

 In vertical force there is by no means the same accordance in value at the 

 two places : in January the Kew value is 48 c.g.s. less than the Green- 

 wich value, and in February 49 c.g.s. greater, and similar discordances 

 occur in other months. The months of positive and negative values agree, 

 however, in number better than, remarking the differences in value, might 

 be expected. It may be here explained, with reference to all the tables that 

 deal with the non-cyclic variation. Tables VII. to X., that the Greenwich 

 values are not likely to be influenced by any incomplete correction for 

 temperature. The diurnal range of temperature in the magnet basement 

 (of which an automatic record is kept) is small, and the difference between 

 the mean temperature of the successive midnights, or successive noons, 

 both in the monthly and yearly grouping, never amounts to more than a 

 small fraction of a degree. Further, the persistently large values in hori- 

 zontal force in different months at Greenwich (Table VII.) are corroborated 

 by the similarly large values observed at Kew. 



It may be further noted that the character of the values of a — b, in 

 Table VI., both in declination and horizontal force, is generally such as, 

 considering the position of the points of maximum and minimum in the 



