58 Dr. van Rijckevorsel on Irregularities in the Yearly 



does the barometer behave ? Do the winds, do the magnetic 

 elements, show something pointing to a common origin ? " 

 The answer is most certainly a positive one, as I hope to 

 show now. 



On the diagram (Plate II.) the uppermost curve, marked T, 

 shows the temperature of every day of the year for the Helder 

 in the Netherlands. The next curve, marked H, shows the 

 horizontal, and the following one Z the vertical component of 

 the earth's magnetic force as shown by the registering instru- 

 ments at Utrecht. The curve marked R the rainfall at the 

 same station. The curve marked P shows the barometric 

 pressure at Greenwich ; the last curve, D, the magnetic 

 declination at the same station. 



T shows the mean of 50 years' observations, H of 33 years 

 between 1857 and 1896, Z of 30 years in the same period. 

 For the magnetic data it is not always possible to use all the 

 years for which values are available ; for if in any case a 

 serious break occurs in a series, which cannot be safely 

 bridged over by fictitious values, or if a discontinuity occurs 

 such as may be occasioned by a slight alteration in the instru- 

 ment or in the adjustment of the scale, it is, as a rule, necessary 

 to reject that whole year — this for more than one reason, but 

 chiefly on account of the secular variation which is so irregular. 

 It is unnecessary, however, to explain this now at any length. 



E, is the result of the 40 years 1856-95, P of the 18 years 

 discussed by Mr. Glaisher *, viz. 1841-58, and lastly "D of 

 the 30 years 1865-94, of which 1 owe the three years which 

 have not been published in the publications of the Observatory 

 to the courtesy of the Astronomer-Royal. 



Of course not all these series are of the same quality; some 

 are decidedly not long enough ; but although a large amount 

 of material is still being computed, I am at present only 

 able to show T what there is in this diagram. The 19 years 

 for air-pressure are decidedly not enough, and at the same 

 time all those who are interested in terrestrial magnetism can 

 form a judgment for themselves as to the intrinsic worth, of 

 the data for the vertical component. But for my present aim 

 the material is about sufficient. 



All the curves have been smoothed down in the same 

 manner as explained in my first paper on this subject. For 

 the magnetic curves, however, a preliminary operation was 

 thought to be advisable, because the effect of the secular 

 variation throughout the year would otherwise tilt the curves 

 and render them less easily comparable with the others. This 

 very simple operation was to take the difference between the 

 * 9th and 10th Reports of the British Meteorological Society. 



