MAGNETIC DECLINATION AT BOMBAY. 
385 
at this rate to reduce the mean numbers at the foot of Table XX. to the common epoch, 
the 1st January of the mean year, we have 
Table XXL 
Months | 
Janu- 
ary. 
Febru- 
ary. 
March. 
April. 
May. 
June. 
July. 
August. 
Sep- 
tember. 
Octo- 
ber. 
Novem- 
ber. 
Decem- 
ber. 
Year. 
Mean for mean 1 
year J 
Correction 
29-863 
■000 
29-882 
-•25 1 
30-302 
-•503 
30-252 
-•754 
30-417 
-1-006 
30-776 
-1-257 
31162 
-1-509 
31-416 
-1-760 
31-687 
-2-012 
32-094 
-2 263 
32-177 
-2-515 
32-751 
-2-766 
Means corrected. 
29-863 
29 631 
29-799 
29-498 
27-411 
29-519 
29-653 
29-656 
29-675 
29-831 
29-662 
29-985 
29-682 
Annual varia- j 
tion j 
+•181 
-•051 
+•117 
-•184 
-•271 
- -163 
- -029 
- 026 
- -007 
+ -149 
- -020 
+ -303 
The last line of numbers is far from presenting the decidedly systematic and regular 
appearance that is found by General Sabine (see Philosophical Transactions, 1863, 
pages 291 to 295) from the similar treatment of the observations of Kew and Hobarton; 
let us therefore examine, as the similar results for those places and for St. Helena and 
the Cape of Good Hope would suggest, whether (failing a regular periodical progres- 
sion from month to month) there be any semiannual inequality of Declination at 
Bombay. Combining together the months April to September and the months Janu- 
ary to March, and October to December, so as to eliminate secular change, we get the 
following results. 
Table XXII. — Showing the mean values of Easterly Declination for April to September, 
and for January to March, and October to December in each of the years 1859, 
1860, and 1862 to 1865 ; and the excess of the mean for the former half year over 
the mean for the latter in each year. 
Years 
1859. 
1860. 
1861. 
1862. 
1863. 
1864. 
1865. 
April to September 
21-246 
24-001 
30-472 
33-897 
36-226 
39-867 
January to March, and October to December 
21-656 
23-989 
30-532 
34023 
36-680 
40-189 
Excess of April to September over October to March 
-•410 
+•012 
-•060 
-•126 
-•454 
-•322 
Mean excess 
— 0 ; -227 
The signs of the differences in five of the six years are negative, whilst the difference 
for the sixth year has a slight positive value. The inference to be drawn thence is that 
there is some probability that the north end of the declination-magnet points at Bombay 
0'-227 * more to the westward in the months April to September than in the months 
J anuary to March and October to December ; and the same result exactly is arrived at 
if all the disturbed observations are included in the calculation of the mean values of 
declination. I cannot regard the evidence as conclusive, however, for two reasons,— first, 
because the observations at the four stations Kew, Hobarton, St. Helena, and the Cape 
of Good Hope agree in showing that at those places the north end of the declination- 
* The fact that this is exactly the magnitude of the semiannual inequality at St. Helena renders the cor- 
rectness of the inference more probable. 
3 F 
MDCCCLXIX. 
