I 



No. IV. 



REMARKS AND TABLES CONNECTED WITH ASTRONO- 

 MICAL OBSERVATIONS. 



DIURNAL VARIATION OF THE MAGNETIC-NEEDLE. 



The Tables Nos. IV., V., VI., contain the monthly averages of a series of mean 

 diurnal variations ; the two first have been deduced from the preceding Tables, Nos. 

 I. and II., and the third from observations on the position of the needle at Slave 

 Lake, which, in consequence of their length, are not inserted in this work. Table IV. 

 was extracted from Mr. Hood's Journal, where it is accompanied with these remarks : 

 " The following table contains the mean diurnal variations of the compass for four 

 months, at Cumberland-House. Many unavoidable interruptions prevented the 

 number of days in each month from being complete, and some irregularities have 

 been caused by the motion of the compass box. Those days are not included on 

 which the needle was affected by the Aurora. As in other places, the diurnal varia- 

 tion increases with the advance of summer, and the needle reaches the extremes 

 of variation at nearly the same hours. But the maximum is at the coldest period, 

 and the minimum at the warmest, which is the reverse, I believe, of the observations 

 which have been made in Europe and in the East Indies." 



Table V. is an abstract of my observations on the positions of the needle at Fort 

 Enterprise (contained in Table II.) which have been reduced to the variation shewn 

 by the same needle, on September 4, 1820. 



In Table VI. the positions of the needle observed at Moose-Deer Island, in 1822, 

 have been reduced to the mean variation obtained at that place in July, 1820. 



It should be remarked, that the whole of the observations made on the positions 

 of the needle at Fort Enterprise and Moose-Deer Island, have been used in forming 

 the Tables V. and VI. 



The appearance of the Aurora, and the disturbance it occasioned on the motion 

 of the needle at Fort Enterprise, were so frequent, that the mean monthly variation 

 must have been deduced from but few observations, if they had been rejected. 



The circumstance of the mean variation being least at midnight there, and at 

 Moose-Deer Island, was evidently caused by the frequent disturbance in the motion 

 of the needle which the Aurora occasioned ; for on those days when it was not visible, 

 the mean diurnal variation followed the course Mr. Hood had observed it to do at 

 Cumberland-House, being most easterly at the time of the first observation in the 

 morning, and least between three and four in the afternoon. The change in the 

 diurnal variation in these parts of North America seems to be governed by the same 

 law as that in England, as the decrease in easterly variation between the morning and 

 afternoon, is in fact a motion of the needle to the westward. 



