674 



MR J. A. BROUN ON THE DIURNAL VARIATION OF THE 





produces a secondary minimum near 5 a.m., and this minimum has its repre- 

 sentative in an inflexion near the same hour in the curves for May to November. 



10th, It might be suggested that the minimum or inflexion near 5 a.m. in the 

 six months of April to October is due to the same cause as the minimum in the 

 remaining six months at 7 a.m., and that there is here also a shift of the critical 

 hour.* In any case there cannot be a doubt of the absolute inversal of the mean 

 movements at 7 a.m. during the months of March and October, a phenomenon 

 which has been observed at no other station. 



These conclusions as to maxima and minima are deduced from curves which 

 represent the mean movement corresponding to the middle of each month. I 

 have had to point out, on different occasions, that deductions from mean values 

 may be wholly inaccurate, since it sometimes happens that the observations re- 

 present laws which are very different, and the mean may represent none of them. 

 It will be desirable, then, in this instance, to endeavour to determine in what way 

 the law of variation really changes from day to day, especially in the months 

 when the change is marked. I shall, however, consider previously the range of 

 movement of the declination needle. 



Table II. — Ranges of the Monthly Mean Movements in Table I., and Monthly Means of all the 

 Daily Ranges during the twelve years, 1853 to 1864. 



Months. 



Range of Mean 

 A 



Mean of Ranges 

 B 



Ratio 

 B 

 A 



B 



m 



January, 



2-06 



3 : 09 



1-50 



2-78 



February, 









1-48 



2-54 



1-72 



2-29 



March, 









0-79 



2-07 



2-62 



1-86 



April, . 









1-67 



2-35 



1-41 



2-12 



May, . 









2-90 



3-23 



Ml 



2-90 



June, . 









3-06 



3-40 



Ml 



3-06 



July, . 









3-06 



3-45 



1-13 



3-06 



August, 









3-64 



4-02 



M0 



364 



September, 









3-31 



3-68 



Ml 



331 



October, 









1-27 



2-25 



1-79 



2-03 



November, 









2-14 



2-70 



1-26 



2-43 



December, 









2-33 



3-04 



1-30 



2-74 



Ranges of the Mean Movements of the Declination Needle near the Magnetic Equator. 



It is evident from the quantities A and from the curves, Plate XL., that there 

 is a tendency to an extinction of the mean movement in March and October, and 

 especially so in the former month. The range of the mean movement is a 

 minimum in these two months, and it is a maximum in August and December. 



The range is greater in August than in December. This fact, and the exist- 



This point will be examined afterwards when the lunar effect has been eliminated. 



