76 OBSERVATIONS OF THE MAGNETIC INTENSITY 



plained in a former paper, read before the society.* The dip was determined 

 with a six inch dipping circle by Robinson, which yielded quite as satisfactory 

 results as the instrument by Gambey, used in my observations in the United 

 States.f The needles for the statical method of Professor Lloyd were also by 



ROBINSON.J 



Two needles were ordinarily used in the observations for horizontal intensity, 

 a cylinder of the Hansteen model and a bar, designated respectively as C and 

 A in the memoir on horizontal intensity just referred to. The correction for 

 temperature, and also, in general, the mode of observing there recorded were 

 employed. To determine the time of beginning and ending of the oscillations, 

 however, five sets of observations were taken,§> and the usual mode of deducing 

 the mean, by comparing the five corresponding observations at the commence- 

 ment and close of each series, was adopted. A pocket chronometer, by French, 

 was used to observe the duration of the oscillations, and its rate during the ob- 

 servations was ascertained by comparison with an observatory clock, when such 

 means was at hand. This watch had been selected in reference to its quality 

 of bearing change of position without considerable change of rate, and stood the 

 trial to which it was exposed reasonably well. It is my impression, however, 

 that when more perfect instrumental means are used in the determinations, 

 greater care will be required in regard to those for ascertaining the time. The 

 observed correction for the rate of the chronometer is duly applied in the tables 

 of results. As all the observations were made between the same arcs of vibra- 

 tion, a reduction to indefinitely small arcs is not required. || The correction for 



* American Philosophical Society's Transactions, vol- v., N. S., Art. xviii. 



t Ibid. Art. viii. 



X They had been heated in boiling water, to discharge as much of the magnetism as could be 

 done by this temperature, according to the recommendation of Mr. Christie. I supposed, from 

 the result of calculations made while in Paris, that these needles lost their magnetism rapidly, but, 

 on farther examination, find that such was not the case, and that they lost but a small portion of 

 their force during more than a year, as will be found stated in a subsequent part of this paper. 



§ Besides eliminating errors of observation, this has the farther advantage of correcting errors of 

 division of the dial plate, as noticed by Professor Forbes, in his " Account of some Experiments 

 made in different Parts of Europe on Terrestrial Magnetic Intensity," &c., &c. Edinburgh 

 Transactions, vol. xiv., part I., p. 5. 



II The semi-ares of vibration for the cylinders were from 0° to 2°, and for the bar, from 4° to 2°. 



