532 MR JOHN ALLAN BROUN ON THE HORIZONTAL FORCE 
the bifilar temperature coefficient, as they are reduced to the mean bifilar for the 
month, which mean is corrected to a constant temperature; this error, it may be 
supposed, will obey some law depending on the mean temperature, or the amount 
of its variation during each month, which will have a single period, and be 
included in the term a, sin (#+¢,) of the equation (1) of sines; whereas we may 
expect, if the observations are sufficiently numerous, that the term a, sin (26+ ¢,) 
will be little affected. On this ground I have discussed the observations of abso- 
lute intensity made at Toronto, the Cape of Good Hope, and Hobarton: those 
which merit the greatest consideration, on account of the number of observations, 
are from the Toronto Cbservatory. 
47. Annual Period from Observations of Absolute Intensity, Toronto.—General 
SaABINE has discussed the Toronto monthly means for the years 1845 to April 
1849 in the Transactions of the Royal Society, 1850 (p. 205), and the means for 
the years 1845 to 1852 in the Introduction to the Toronto Observations, vol. ii. 
(p. xc); in the latter he has however omitted the observations for 1852, on 
account of a considerable change in the monthly values after April 1852. I 
have not been able to see the propriety of this omission, as,the changes in other 
years for an interval of two months (as May to July, June to August, September 
to November, 1850) are even greater than those from March to May 1852. In 
order, however, that the result may not depend on the rejection of the observa- 
tions for this year, I shall consider the means both inclusive and exclusive 
of 1852. 
48. The following are the monthly means (1845-52) with secular change 
eliminated at the rate of 0-00035 per mensem, mnus the lowest mean (that for 
October), the monthly means 1845-51 according to General SaBrne’s discussion ; 
and (I have added) the monthly means for Makerstoun (from the bifilar) for 
the eight years 1842-49, all reduced to the units | 
TABLE XV.—ANNUAL PERIOD FROM OBSERVATIONS OF ABSOLUTE INTENSITY AT 










TORONTO. 
PLACE, Years. | Jan. | Feb. | Mar. |April.| May. | June.| July. | Aug.| Sept.| Oct. | Nov.| Dec. 
Toronto, . . |1845-52| 5-4] 1-1} 9:1) 1-7 |13-0| 9°6| 7:9) 8-2) 3:1| 0:0) 3°7 Gr 
Makerstoun, . | 1842-49] 5:3) 8-9| 3:1] 2°5| 6:2| 88| 6:6] 3:3] 0-4] 0:0| 2°5| 56 
Toronto, . . | 1845-51 | 4:2) 0-8) 9-3] 8-5 |15-0 /15-5 |11-0| 9°3| 3-°7| 0-0] 3°7| 6°6 









49. If we compare the results at Toronto 1845-52 with those at Makerstoun 
1842-49, it will be obvious that the most marked difference is to be found in the 
€ 
high value for March at Toronto; there can be no doubt, however, thatthe — 

