'The Transactions of the Bombay Geographical Society. From January 1863 to December 1864. (Edited by the Secretary.) Volume XVII.' [‎7] (168/524)

BOMBAY TIDAL OBSERVATIONS FOR 1861.
The corrections in Professor Orlebar’s formula* for variations of parallax
and declination are only empirical, and consequently, though made to re
present with considerable accuracy the observations of 1846-47, from
which they were derived, they may vary very considerably when applied
to more recent observations. Certain limits of instrumental error must
of course be always admitted. Mr. Montriou says, the gauge registers
the tides generally correct to within 5 minutes of time ; and in extreme
cases only w r ere the times of spring tides uncertain to 10 minutes, of
neap *tides to 25 minutes. On examining the differences between
the observed and computed times of high-water however, I found them
to be as follows
Diffs. from 0m. to 6m. at U.
6—12
12 — 18
18 — 30
30 — 45
45 — 60
60 — 90
90 —120
120 —150
150 —180
180 +
If the gauge is still to be
minutes on an average, these
three-fourths of the observations
error. Whence does this arise ?
50 Ohs. at L.
T. 32.
total 82
40
41
91
48
48
96
66
80
146
46
61
107
32
35
67
29
25
54'
12
5
37
2
4
6
0
2
2
0
2
2_
trusted generally to within even 10
differences present some difficulty :
lie beyond the limits of instrumental
1. Are the observations unworthy of confidence — especially for tides
attending the transits between four and eleven hours ?
2. May not the formulce fail to represent the true times of high-
water with sufficient accuracy ? Or—
3. May not small instrumental errors, and occasionally others of
eren considerable amount, combine with erroneous results of the
formula* to give rise to such a variety of differences ?
The smallness of the algebraic sum of the differences, as ah
leady stated, and which harmonises so well with an alteration of
the mean establishment by about 10 minutes, seems to indicate
that if there are errors in the observations, they are at least not very