It is not necessary that all the observations for finding the errors, at the beginning or end of the interval, should be made on the same day, nor that the Moon's distances should be all taken from the same object. When the observations, at the beginning or end of the interval, are made on different days, set down all the times at Greenwich with their respective errors opposite to them, find the sum of the times, and also that of the errors; these sums being divided by the number of times, or errors, will give the required epoch and the corresponding error of the Chronometer. For example, let the times at Greenwich, and corresponding errors of a Chronometer for mean time at that place, be as follows:

Here the mean of the times at Greenwich is March 6, at 11h. 24m. and the mean error, or that which answers to the mean Greenwich time is 2h. 6m. 50s.; and by again finding the error of the watch for some subsequent time, the rate may be deduced as before.

The same degree of accuracy is not to be expected in this method of settling the rate of a time keeper, as may be obtained by altitudes of the Sun or Stars taken on shore; it is, however, as has been before observed, the only method that can be employed at sea, and the Navigator, who carefully practises it, will seldom find 5 miles of error in the longitude, as given by a tolerably good Chronometer, during a passage of any length, for in this case we do not depend upon the exact going of the Chronometer, for a long period, but merely from one set of lunar observations to another.

Sometimes lunar observations and the bearings of some well-known point of land may be combined, to obtain the error and rate of a Chronometer, and should be had recourse to when no other means offer.

EXAMPLE.

1844. July 20d. 1h. 30m. M. T. Lat. 24° 29′ S., lon. 60° E.. The following app. alts. of O and, and app. dist. of west of C, were deduced from observations to obtain the error of Chronometer, No. 858. The Greenwich Date employed to correct the elements from the N. A. was 19d. 21h. 30m. Time shewn by No. 858 at moment of observation 9h. 29m. 45s., elevation 15 ft.

By two more sets of sights of the O and, and 3 sets of Antares east of D, the following results were obtained. the above.

2 6

and * Set 1 is

Lunar.
d. h. m.
19 21 31 51

S.

h. m. s.

0 2 6

21 36 44

0 2 4

21 40 09

2 20 06

20

2 22 18

2 24 16

2 26 35

2 31 27

2 33 39

0

2 12

Gwh. Date

20

0 1 53 Error of
No. 858

2 10.5 slow for M. T. Gwh.

h

EXAMPLE.

To find the Error of Chronometer by

the bearing of the Land, &c.

The Quoin Pt. bore E. N. E.

1844. August 10d. at App. Noon. by compass, the variation of which was ascertained to be 24 points westerly. The obs. lat. was 35° 9 S.

True bearing N. E. N.

x of lat.

Long. of Quoin pt.

Long. of Ship

20 and course 31 pts. give 16-5 dep. to z of long. 0° 20′

in time 1h. 17m. 8s.

19 37 E.

= 19 17

By the morning sights 19' East of the merid. of observation at Noon the M. T. at Ship, was (less 12h.)

The merid. 19' to the West of the A. M. Observations would have the time earlier by the amount of that ex long. viz.

8h. 49m. 57s.

- 1 18

999

The interval between the dates of the two errors is 20d. 19h. 30m. equal to half hours (by which divide the loss) 139-5·14

14 × 48 the half-hours in a day give 6-72s. as the daily rate, which is losing, as both errors are slow, and the last is the greater.

To find the Error for Mean Noon at Greenwich on the 10th August, being the nearest noon.

On the Management of Chronometers.

Unless particular care be taken of Chronometers, it is not to be expected that such delicate pieces of mechanism can continue to go with regularity; it may, therefore, be of service to those who have not had much experience in the use of time-keepers to attend to the following remarks:

1. A Chronometer should be wound up at regular intervals, it being very improper to let one that is generally wound up between 8 and 9 o'clock in the morning run till noon. Great care should be taken to avoid circular motion; therefore, when winding up a Chronometer, it must be kept steady, and the key only turned. The number of turns of the key should be counted, that the last few may be wound gently so as not to butt too sharply.

2. Chronometers should be placed so as to be as little exposed as possible to sudden shocks, from the sea striking the ship, or from the shutting of doors, &c.: they ought not to be exposed to a current of air; and nothing magnetic should be allowed to remain near them. They should be placed as near the centre of motion as possible. A small table, secured to the deck (not to a bulkshead), with a stancheon on each side, to prevent people from rolling against it, will be found convenient: the table may have a well in it, capable of holding the requisite number of watches, with a sufficiency of dry sawdust or cotton around them, to break the force of unavoidable concussions upon deck or elsewhere, or a box upon the table may be used instead of a well in it. Watches should always be kept in the same position that they are placed in when first they come on board. As soon as they are placed they should be compared.

3. It is very improper to make a practice of taking a Chronometer on deck, when observing altitudes, merely to find the longitude; for, besides the risk of accidents, it is hardly possible to carry about a Chronometer without giving it too much circular motion. Any sudden change of temperature ought also to be avoided: it is therefore proper to take the times of observing the altitudes by a common watch, and find the difference between it and the Chronometer, immediately before or after the observation: then this difference being applied to the mean of the times of observation by the watch, will show the time by the Chronometer answering to the mean altitude.

4. If a Chronometer be allowed to run down, it will not commence going again, after being wound up, until it gets one or two pretty quick quarter turns in a horizontal direction, its face being upwards. After being set going, a Chronometer will sometimes keep the same rate it had before it was let down; this, however, is uncertain, and no dependance can be placed on it, until a fresh rate and error be obtained.

5. When a Chronometer is carried to or from the Ship, it should

traversing on its gimbles. Great care should be taken by a person carrying a Chronometer, that it gets no sudden jerks, or quick circular motion. If the rate of a Chronometer can be properly ascertained, when it is on board, it ought not to be taken on shore for that purpose. Commander Richard Owen, in his "Essay on Chronometers," says that "it may be taken for an absolute maxim in general practice, that the rate of a Chronometer obtained on shore will not be the rate that it will keep when it is removed to a vessel.”

6. Change of temperature generally produces the greatest alteration in the rate of a watch. In a paper by Mr. Fisher, published in the "Philosophical Transactions," for 1820, page 196, it is stated that "Chronometers generally go faster on board of a Ship than on shore." Mr. Fisher ascribes this to the magnetic effect of the iron in the Ship on the steel part of the balance of a Chronometer. This is a subject that deserves farther attention from all intelligent Navigators.

7. The conclusion to be drawn from the foregoing remarks is this: As soon as a Chronometer is in the Navigator's hands, every opportunity should be employed to ascertain its error and rate. The Time Balls at our large sea-ports were established for this object. Headlands and lights, when in soundings, and lunars, on the ocean, must all be resorted to. In addition to these, the prudent Navigator will not neglect to put into practice the good old maxim, "Latitude, look out, and LEAD."

Table A is given for the method of finding the rate of a Chronometer, by equal altitudes of a Fixed Star, observed on different days. It will also answer for the method by transits of the Fixed Stars. If the acceleration be wanted for any time exceeding 60 days, it may be found by adding together the accelerations answering to the days in the Table that make up the given time.

TABLE A*, Acceleration of the Fixed Stars, in Mean Time.