342 Magnetical Experiments and Observations 
The paper to which we now refer, contains the detail of four 
distinct series of pendulum observations ; the first at the Royal 
Observatory, Greenwich, in May 1824, prior to the voyage; 
the second and third at Port Bowen in Prince Regent’s Inlet 
where the Hecla and Fury wintered; and the fourth, which is 
indeed composed of two distinct sets, at the Royal Observatory 
on the return of the expedition. 
The result of these experiments is in the highest degree satis- 
factory. The difference in the two Greenwich sets of observations, 
after an interval of eighteen months, under a different pressure, 
and after the pendulum had been exposed to a temperature of 
47° below zero, is only that of two- tenths of a vibration in twen- 
ty-four hours; and the two series at Port Bowen give a still 
nearer approximation. In the former place, the observations 
were made in a room selected for the purpose by the astronomer 
royal ; in the latter, in a snow-house, ingeniously constructed. 
By comparing the mean from the two series at each station, the 
author finds, for the ellipticity of the earth, which agrees 
well with a number of other independent observations, and is not 
very wide of the general deduction by Laplace, which is 
although it differs widely from the means deduced by Captain 
Sabine, which is It must, however, be carefully remem- 
bered, that this ellipticity of Captain Sabine is obtained by an 
accommodation of results, and is by no means directly deduci- 
ble from his observations. It is merely that ellipticity which gives 
the least errors ; and if we had any reason to believe that the 
earth was a spheroid of uniform ellipticity, the result obtained by 
Captain Sabine might be admissible ; but is it not probable that 
different arcs have really different ellipticities ? and if so, the 
mean obtained by encreasing the number of vibrations by Jive 
in one place, and decreasing them by the same number in ano- 
ther, in the space of twenty-four hours, must be considered un- 
satisfactory, if not erroneous. It is, in fact, assuming a uni- 
formity of figure, which is at variance with all the best recorded 
experiments, amongst which those by Captain Sabine himself 
are justly included. We admire the accuracy of his experi- 
ments, but object altogether to his deductions ; and we sincere- 
ly recommend to Lieutenant Foster, in all his future experi- 
ments, to observe the same rigid adherence to his experimental 
