2 Beppicker—On the Influence of Magnetism on the Rate of a Chronometer. 
1d. 
Two observations gave rise originally to the suspicion of the possibility of a 
magnetic influence, which, therefore, may here also serve to define the question 
more precisely. It was, firstly, the well-known fact that chronometers frequently 
change their rates when transported from one place to another, especially from 
shore on board ship, and vice versd ; and, secondly, the discovery that the steel 
parts in the work of a chronometer, especially the spiral spring, and the cross- 
-bar and corresponding steel rim in the balance, were frequently magnetic, and 
thus obviously subject to the same influences which act on a compass-needle. 
As to the first point, I did not seek completeness, and it is not my intention to 
reproduce in full all the observations I have made, but rather to give some striking 
instances only, and the attempts at their explanation. 
The most extensive observations of land and sea rates are by George Coleman 
(No. 4).* He publishes (as an appendix to George Fisher’s Paper, No. 5) a number 
of Tables, giving the land and sea rates of fourteen pocket- and forty box-chrono- 
meters. Besides these values—which frequently extend over different voyages, 
and go as far back as 1802—the table contains general remarks as to the per- 
formance of the chronometers, and the information as to whether the ship 
carrying the chronometer was iron- or copper-fastened. Excluding those box- 
chronometers, which are pointed out as indifferent or unreliable, there remain 
twenty-four instruments of iron-fastened and nine of copper-fastened ships, and 
it is perhaps remarkable, and of some importance, that six cases of losmg have 
been observed in those first twenty-four, and four such cases in the other nine 
chronometers. It is, however, impossible to draw general conclusions from the 
Tables, as they are rather imperfect, and of some importance only by the con- 
siderable number of chronometers. Some cases, however, deserve mention: 
for instance box-chronometer No. 1, which went five and a-half years at the 
rate of 5° gaining on shore, and 5°3 gaiming on board; or box-chronometer 
No. 33, which “‘ varied precisely the same quantity (losing) and the same way 
two succeeding voyages,” the difference being here 0°4 to 0°5. [There is 
a misprint in this case; the two rates on shore should be 0°7 gaining and 0*:2 
losing, as seen by the remark.] So we can conclude from Coleman’s Tables— 
which may deserve a closer discussion—that a difference between land and sea 
rates frequently takes place, but that it does not always lie—not even in 
the same chronometer—in the same direction, though the gaining sense on 
board ship seems to prevail. A special connexion between the changes of rates 
and the ship’s fastening is, in Coleman’s Tables, hardly perceptible. More striking 
* The figures after an author’s name refer throughout to the ‘Chronological List of Papers 
quoted” on pp. 52-56. 
