290 
MAGNETIC SURVEY IN NORTH AMERICA. 
to 1*224, being 2*45 inches and 3*00 inches respectively; but in order to secure an 
independent check upon the results by these bars, a third deflector of 3*6 inches in 
length, was also employed, at one distance of deflection only, at many of the stations 
of observation. It was found that the values of the absolute intensity given by the 
3*6 inch bar agreed very satisfactorily with those given by the 3*0 inch bars, and in 
taking the general mean at each station, an equal weight has been allowed to the 
results by each bar. The observations were made in the manner described in the 
instructions for the use of portable instruments by Lieut. C. J. B. Riddell, R.A. 
The separate parts of each experiment were reduced to a common value of the hori- 
zontal force by corresponding observations with a portable bifilar magnetometer, 
which was previously placed in adjustment at all the stations where it is not other- 
wise stated ; but as a small instrument mounted in the open air is exposed to acci- 
dents, and to be much affected by atmospheric changes, it has happened in some in- 
stances that breaks and interruptions have occurred in the series, and when the 
several parts of the experiments were not completed at once, (as at Dunvegan and 
Isle a la Crosse,) they could not be reduced to a common scale division ; at some few 
stations, where time did not permit of the adjustment of a differential instrument, 
or other circumstances rendered it inconvenient, the bifilar was not employed. 
“Table XXXVIII. contains the corrected data of each experiment, viz. — 
“ 1. The angles of deflection at each distance, corrected to a temperature of the 
deflecting bar of 50°, and reduced to a mean value of the differential instrument. 
“ 2. The time of one vibration corrected for the rate of the chronometer, for torsion, 
and for arc, reduced to a common temperature of 50°, and to the mean value of the 
differential instrument. The uncorrected data are given in a separate Table, No. 
XXXIX. 
“ Corrections for Temperature . — The following Table contains the values of the 
coefficient ( q ) for changes of temperature for each bar, as obtained by different expe- 
riments at Toronto. 
Table XXXI. 
1845. 
No. 30. 
No. 31. 
No. 17. 
January 
and 
February 
< 
'•000226 
•000267 
•000218 
•000225 
•000205 
•000254 
•000218 
•000224 
•000266 
•000226 
•000218 
•000362 
•000378 
•000371 
•000374 
Means. . 
•000228 
•000234 
•000371 
“ For bars 20, 23 and 29, employed at Athabasca and M c Kenzie’s River, I have 
assumed a value for q , viz. *00023. 
I 
