LIETJT.-GENERAL SABINE ON TEERESTEIAL MAGNETISM. 



471 



Land Stations at which the valms of the magnetic elements have been determined with 

 sufficient accuracy to justify their being regarded hereafter as Primary SxAXiONa 



Note. — It is intended to cmtinue this Table in subsequent numbers of the Contributions. 



Table XIII. 



• The observations which were made at Port Louis between April 12th and August 23rd, 1842, to determine 

 the Inclination with needles whose poles were reversed and the results obtained from the mean of the eight 

 positions of the circle and needle, afford a favourable opportunity of judging of the accordance attainable in such 

 results when the observations are made by skilled and careful 0')servors, and when suitable instruments are 

 employed. Sixty determinations were made with the Dip Circle of the ' Erebus ' and its three needles R 4, R 6, 

 and R 7, the observer being Lieut. A. J. Smith of the ' Erebus,' relieved occasionally by Mr. T. E. L. Mooke of 

 the ' Terror.' The results are detailed in the Vl.th No. of these Contributions, pp. 101-103. The arithmetical 

 mean of the sixty residts is —52' 25'-06, and the probable error of a single result is+ l'-52. It wiU be 

 remembered that at the time those observations were made, the instruments which were used had already been 

 employed for more than three years on a service of no ordinary exposure, and had been frequently disembarked 

 for observations on land or on ice. The general accordance of the results, and the very smaU amount of the 

 probable error of a single result, bear strong testimony to the care and skUl of the observers, as well as to the 

 improvement which took place in the English Dip Circles and Needles, in consequence of the pains taken 

 by the participators in the Magnetic Survey of the British Islands in 1837 and 1838, of whom Sir Jaues Ross 

 was one. [Dr. Lloyd in the VII. th Volume of the Reports of the British jlssociation, page 99, note 1 ; and 

 Table III. in the same volume, p. 57.] The detailed statement of the observations at the Falkland Islands was 

 published in the Vl.th No. of these Contributions almost immediately after its receipt in England (in 1844), 

 and afforded to aH who were desirous of profiting by the instruction it conveyed a knowledge of the degree of 

 accuracy which might be expected by the employment of the English Dip Circles and Needles, when placed in 

 the hands of properly trained and careful observers. Even the small probable error of + 1'-52 was doubtless 

 due in great measure to " magnetic disturbances " and to the effects of " horary variation, solar and lunar," 

 the influence of which can only be eliminated by corrections supplied by the continuous record of the magne- 

 tometers of an observatory. The probable error of the mean of the sixty observations with the circle and needles 

 of the ' Erebus ' at Port Louis is +0'-197. 



Those who are interested in the accuracy with which observations of the Magnetic Dip may be made will 

 recognize with interest and satisfaction that +l'-52, the probable error of a single result derived from the 60 

 observations at Port Louis, is almost precisely the same as that ( + 1'-50) obtained by the mean of 282 inde- 

 pendent results at the Kew Observatory (Proceedings of the Royal Society, March 1861) by several observers 

 employing Dip Circles and Needles of the British pattern ; in which results, as in those at Port Louis, no correc- 

 tions for disturbing influences, derived from the continuous records of an observatory, were introduced. 



3u2 



