164 



NATURE 



\_yuly I, 1875 



compared together, and the result of all these comparisons 

 tends to impress the reader with the fact that we have in 

 this volume a series of observations of the magnetic decH- 

 nation of a thoroughly accurate and trustworthy nature. 



The following passage from Mr. Broun's magnetic 

 diary may be quoted as exhibiting the sources of error to 

 which magneticians are exposed, as well as the care be- 

 stowed in avoiding them — 



"1855, Dec.4d.9h. A sudden vibration of Grubb's mag- 

 net through thirty scale divisions was observed, and the 

 difference of Adie's and Grubb's instruments, which had 

 previously been — o''o5, became suddenly -j- 3''5o. It was 

 supposed that either the suspension thread was breaking, 

 or that a spider had got within the box. 



" Dec. 4d. 22h. The boxes were removed, and an 

 exceedingly small spider was discovered and removed. 

 This was the only occasion in which a spider succeeded 

 in entering Grubb's declinometer boxes between 1852 

 and 1870. Every care was taken when the boxes were 

 removed, before replacing them, to hold them for some 

 time over the flame of a lamp, so that spiders, even invi- 

 sible to the naked eye, must have been dislodged or 

 destroyed." 



It remains now to give our readers a summary of the 

 most important results obtained by Mr. Broun from the 

 reduction of his observations. 



In the first place, the secular variation is found to be 

 irregular, but the observations seem to indicate that after 

 a certain interval the acceleration or retardation of the 

 secular movement has equal values. This interval is 

 estimated at 10*5 years. In order to find the annual 

 period, the variations which form the secular and decen- 

 nial inequalities have been eliminated. The observations 

 then indicate a twofold inequality, one of which corre- 

 sponds to a single oscillation in a year, with a minimum 

 in March or April, and a maximum in September or 

 October, while the other represents a double or semi- 

 annual oscillation with maxima in March and September. 



Mr. Broun was also led to suspect a. period of forty-four 

 months, which was repeated four times successively in 

 his observations, although no cause is known which 

 could produce an inequality of this duration. 



The next inequality noticed is the twenty-six day period, 

 which Mr. Broun is inclined to attribute to solar action 

 with more confidence than the longer period of ten or 

 eleven years. Our readers will remember that the period 

 was re-discovered by Dr. Hornstein, director of the 

 Prague Observatory. Mr. Broun thinks that there are 

 traces of a double oscillation of the twenty-six day period. 



Coming next to the important solar dijirnal variation, 

 the chief features of which are tolerably well known, Mr. 

 Broun finds this to consist of one marked maximum and 

 one marked minimum of easterly declination in each 

 month of the year, and of one or more secondary maxima 

 and minima. 



The principal maximum occurs in the six months of 

 April to September at about 7 A.M., and the principal 

 minimum about twenty minutes past noon in the same 

 months. Nearly the inverse of this happens in the four 

 months of November to February, The results obtained 

 by Mr. Broun appear to him to indicate the action 

 of opposite forces belonging to the two hemispheres, 

 which mainly destroy each other in March and October 

 at Trevandrum, but one of which is preponderant in 



the other months of the year ; and of these forces he 

 remarks that those of the northern hemisphere seem to 

 have a greater effect on the variations of the whole globe 

 than those of the southern hemisphere. 



The daily range was a minimum in 1856 and a maximum 

 in i860. It is a minimum in March and October, and a 

 maximum in August and December. 



In considering the lunar diurnal variati0n, Mr. Broun 

 begins by showing that the results relating to the varia- 

 tion to be obtained by him are really due to the lunar 

 action, and not to any portion of solar perturbation re- 

 maining uneliminated. 



The following very singular results have been ob- 

 tained : — 



1. The mean lunar diurnal variation consists of a 

 double maximum and minimum of easterly declination 

 in each month of the year. 



2. In December and January the maxima occur near 

 the times of the moon's passages of the upper and lower 

 meridians ; while in June they happen six hours later, the 

 minima of easterly declinations thus occurring near the 

 times of the two passages of the meridian. 



3. The mean of the ranges of the lunar diu rnal varia- 

 tion shows (like the solar diurnal range) a minimum in 

 1856 and a maximum in i860. 



4. The action of the moon on the declination needle is 

 greater in every month of the year during the day than 

 during the night. 



5. There appears to be a remarkable change in the 

 lunar action connected with the rising and setting of the 

 sun, especially with the former. 



We now come to a part of the reductions where we feel 

 compelled to differ from the eminent magnetician as to 

 what may be termed the scientific policy which he has 

 pursued. We allude to the question of disturbances. 



There can, of course, be no doubt that a strictly mathe- 

 matical discussion of a series of observations will indicate 

 the various periods of action of the influential forces. We 

 know that this method served to indicate many impor- 

 tant astronomical periods long before the mechanical 

 nature of the astronomical forces was recognised. 



We might, for instance, take a body of meteorological 

 observations and treat them in a strictly mathematical 

 manner, and we should no doubt be led to a yearly and 

 to a daily period, even if we were not acquainted with the 

 existence of the sun. But who would pursue this method ? 

 We take advantage of the knowledge derived from other 

 sources of the exact length of these two periods to begin 

 with, and do not think of endeavouring to obtain these by 

 means of the observations themselves. 



Furthermore, in meteorology, with the general consent 

 of all engaged in it, we have gone even further than this. 

 There is unquestionably a distinct daily and yearly fluc- 

 tuation of the meteorological elements brought about by the 

 sun, but besides this there are other phenomena ultimately 

 due to the sun, though not in the same way, which 

 meteorologists have agreed to consider apart by them- 

 selves. 



We allude to cyclones, which, when examined sepa- 

 rately, are found to obey very different laws from those 

 which regulate ordinary atmospheric changes. Thus 

 these laws have been discovered by agreeing to separate 

 certain observations which were unmistakably abnormal. 



