306 BULLETIN OF THE 



This line is perpendicular to the direction of the object ob- 

 served. If it be projected on a chart, and lines parallel to it 

 be drawn on either side at a perpendicular distance equal to the 

 uncertainty of the altitude, and others at a distance in longitude 

 from these equal to the uncertainty of the chronometer correc- 

 tion, the position of the ship will be within the belt delineated; 

 and this is as valuable a determination as if the latitude alone, 

 or the longitude alone, had been found. 



Again, from an altitude of the same, or another object, another 

 line of position may be found and projected on a chart; and one 

 of the lines being shifted for the run of the ship (including known 

 currents) in the interval, the intersection of the two gives the 

 ship's position botli in latitude and longitude. This intersection 

 is best determined when the azimuths for the two observations 

 differ 90 \ It may be found by computation as readily as by 

 projection. 



If the object is near the meridian, it is better to assume two 

 or more longitudes and compute the corresponding latitudes. 

 This, however, was not proposed by Capt. Sumner. 



This method* was first published by Capt. Thomas H. Sumner 

 of Boston in 1843. His conception of the problem was purely 

 geometrical. The sun, or any other body, at a particular instant 

 is vertical at a place on the earth's surface, whose latitude is the 

 declination of the body, and whose longitude is its hour angle at 

 the prime meridian; and the body will be at the same altitude at 

 all points of a small circle, whose pole is where the body is verti- 

 cal, and whose polar radius is the complement of the altitude. 

 An altitude of an object, when the latitude and longitude of the 

 place of observation are unknown, simply determines the position 

 of such a circle, or a limited portion of it depending on the 

 accuracy with which the latitude, or the longitude, is known. 



This method was strongly commended by some officers of the 

 XJ S, Navy and before 1851 formed a part of the course of naviga- 

 tion at the Naval Academy In a few years it was very generally 

 used in the Navies of the United States and Great Britain, and 

 has been introduced in a more refined form in the best works on 

 navigation ; but it is not much known in the merchant services of 



* A new and accurate method of finding a ship'.s position at sea by pro- 

 jection on Mercator's Chart, by Capt. Thomas H. Somner ; Boston, 1843. 



