LIEUT.-COLONEL SABINE ON TERRESTRIAL MAGNETISM. 
117 
We have next to consider the more important question of the steadiness with 
which the needles may have maintained their magnetic condition during a voyage of 
so many months, and under such numerous and various trials. When there are many 
needles, all of generally steady magnetism, their intercomparison affords on the whole 
a not unsatisfactory mode of discovering the periods when any one amongst the 
number may have sustained an accidental loss, and of obtaining an approximate cor- 
rection for it. I have already shown in No. II. of these Contributions*, the steadiness 
of Nos. 5. 7- 8. 9. 11. 12. and 13. of Sir Edward Belcher’s needles, from October 
1838 to March 1839, by means of the observations made at Panama at both those 
dates. I have also noticed in the same paper that the intercomparison between 
March and November 1839, had shown that No. 8. was apparently more subject than 
the other needles to small occasional losses, and that I deemed it therefore less fit 
than the others for carrying on a chain of magnetic determinations. Subsequent 
experience with this needle has confirmed this early indication, as will be more fully 
shown in the sequel. 
For the present investigation we shall therefore employ only Nos. 5. 7. 9. 11. 12. 
and 13. 
A similar opportunity to the one above noticed (at Panama), of evidencing the 
general steadiness of these needles, was afforded by Sir Edward Belcher’s return to 
Singapore in December 1841, having previously visited that station in the October 
of the preceding year : the agreement in the respective times of vibration at those two 
periods is shown in the following Table : — 
Periods. 
Designation of the Needles. 
5. 
7. 
9. 
11. 
12. 
13. 
S 
s 
s 
s 
s 
s 
October 1840 
466-9 
532-8 
433-2 
469-1 
397-0 
390-5 
December 1841 
465-8 
532-3 
434-0 
467-8 
397-9 
390-4 
We are therefore warranted by the observations at Panama and Singapore, in regard- 
ing the usual condition of these needles to be that of steady magnetism, subject 
nevertheless, as all magnets appear to be, to occasional loss of force from accidental 
causes, the nature and operation of which are not perfectly understood. 
The ratio of the squares of the times of vibration of two needles, or the difference 
of the logarithms of the squares, at stations at which they were both used, should 
be a constant quantity (within the limit of errors of observation), if both needles have 
continued steady; consequently a loss of magnetism occurring in either will be shown 
by an alteration in the ratio exceeding in amount the ordinary errors of observation ; 
if the ratio diminishes, the loss has taken place in the needle, which for the purpose 
of comparison is regarded as unity ; if it augments, the loss is in the needle with which 
it is compared. A simultaneous loss in both needles to an equal amount will not 
* Philosophical Transactions, 1841, Part I., p. 13. 
R 2 
