34 Staff-Commander E. W. Creak on the Magneticcvl [May 8, 



declination, the formula used by Captain (now Sir J. H.) Lefroy, in 

 the published volume of his " Magnetical and Meteorological Obser- 

 vations at Lake Athabasca,"* &c, was adopted. This renders the 

 several results directly comparable with other series obtained at 

 numerous stations on the -North American continent, and Van 

 Rensselaer Harbour, Greenland. 



The formula for -a whole month or longer iperiod is, 



v'fjk**-**)*- 



In this -yfsh represents the hourly scale readings, tyh the monthly mean 

 at the same hour, IS" the total number of the observations, 2 1 the sum 

 of all the squares. 



From the values of the mean hourly disturbance without regard to 

 sign, it is found that the disturbing force never ceases, and that in the 

 mean monthly values it decreases as the winter solstice is approached, 

 and increases rapidly towards the equinox. This points to an accord- 

 ance with the investigations by Mr. Broun, F.R.S., of the disturbances 

 of declination in high northern latitudes, " that there are maxima at 

 the equinoxes and minima at the solstices. "f 



In the hourly values the maximum mean disturbance occurs between 

 the hours of 9 a.m. and 3 p.m., and again from 9 p.m. to 2 a.m. ; the 

 minimum from 6 to 9 p.m. and 3 to 8 a.m. Comparing the results for 

 the whole period at Van Rensselaer Harbour and Discovery Bay, a 

 marked difference is apparent. 



At the first-named place the disturbing force was greatest during 

 the night and least during the day : at Discovery Bay the disturbing- 

 force was considerably greater during the day than the night. 



Another result which also deserves attention is that of the mean 

 maximum disturbance about noon, being preceded and followed by 

 smaller values. This is in accordance with the results obtained at 

 Yan Rensselaer Harbour, Lake Athabasca, Toronto, and Sitka. At 

 Discovery Bay there is also a very marked disturbance at 11 p.m. 



Having considered the mean disturbance without regard to sign, it 

 yet remains to be seen what was the direction of the disturbances 

 during the several hours of the twenty-four. The computed values 

 show that the easterly disturbances prevail over the westerly during 

 the night from about 7 p.m. to 7 a.m., and during the remaining hours 

 of the day the westerly disturbances predominate. 



Taking the difference between the mean easterly and mean westerly 

 disturbances, it is found that the easterly disturbance is both moderate 

 in amount and monthly change when compared with the westerly. 



Thus the ratio of the excess of westerly disturbance in December 



* See " Irregular Fluctuations," &c., pages 74-75 of that work, 

 f See " Trans. Koy. Soc. Edinburgh," vol. xxvii, Part IT, page 586. 



