58 Ocean Magnetic Observations, 1905-16 



SHORE MAGNETIC WORK. 



Mention occurs on page 15 of shore magnetic observations being made at every port 

 visited by the vessel. Specimens of the usual land observations will be found in Volume I, 

 pages 30-41. The corrections for the various instruments on the adopted standards will be 

 found on pages 76-77. 



The results of the shore observations and descriptions of stations are given on pages 



105-114. 



DETERMINATION OF GEOGRAPHIC POSITION AT SEA. 

 General Methods. 



It would avail little to strive for the highest accuracy in the values of magnetic elements 

 at sea if the corresponding geographic position were not well determined at the same time. 

 There are regions in the Pacific, as well as in other portions of the globe, where the magnetic 

 declination and inclination vary almost as rapidly as the geographic coordinates expressed 

 in the same units. It is therefore of vital importance to secure the closest approximation 

 possible to the true geographic position corresponding to the time of the magnetic results, 

 particularly at sea, where the ship's position is changing continuously. If it were possible 

 to determine simultaneously both geographic coordinates at the middle of the magnetic 

 observations, or at mstants whose mean would correspond to the middle of the magnetic 

 observation, the problem would be comparatively simple and the desired accuracy might 

 be readily secured. 



Opportunities for simultaneous determinations of both coordinates have sometimes 

 occurred for several days in succession, when either the Moon or some very bright star, 

 visible in daylight, was favorably situated with regard to the Sun about the middle of mag- 

 netic observations. Such opportunities are never missed, as the geographic position thus 

 determined is final, if the chronometer rate assumed is found to have been satisfactory. 

 Usually, however, as the Sun only is available in the day, and stars only during twilight, 

 since the horizon is lost in the later darkness, the geographic coordinates must be deter- 

 mined, in succession, as the Sun changes in azimuth and as the ship sails from point to point. 

 The customary procedure, therefore, has been to observe the Sun's altitude in the morning, 

 at noon, and in the afternoon, and the altitudes of stars at dusk. The ship's changes of 

 position during the day were determined from the course or courses sailed and the distances 

 recorded by a taffrail log. Whenever feasible the altitudes of three different stars were 

 observed, to eliminate, as far as possible, both instrumental and observational errors. 



The geographic position at the time of magnetic observations was determined by computing 

 the changes in latitude and longitude from the preceding position fixed by astronomic 

 observations, then noting the differences in the latitude and longitude of the following 

 astronomic station, as computed by the course and distance run (the dead reckoning), and 

 as given by the astronomic observations. These errors of run in latitude and longitude 

 were distributed over the distance run proportionally to the time elapsed. Such distribu- 

 tion of the error of dead reckoning is based on the assumption that the causes producing the 

 error, be it current, drift, leeway, or bad steering, are constant throughout the period con- 

 sidered. 



Thus, on April 14, 1908, there were magnetic-declination observations (see p. 47) 

 about 6 a. m., ship's time, followed by two altitudes of the Sun about 9 a. m., two latitude 

 observations at noon, and two altitudes of the Sun in the middle of the magnetic observa- 

 tions. The altitudes were taken in each case by observers D.C.S. and P.H.D. Finally, 

 altitudes of Canopus, Jupiter, and Rigel were measured at twilight, thus completely con- 

 trolling the changes in latitude and longitude throughout the day. 



