Geographic Position at Sea 59 



Forms.— As in the magnetic observations and computations, forms are also used in the 

 work of determining the ship's position. They not only lead to a systematic record of the 

 observations and computations, but are also great aids to the computer and reviser. Three 

 forms are used: one for the dead reckoning, one for the longitude, and one for the latitude. 

 This method of navigation lends itself readily to the subsequent operations of correcting 

 the positions for both the error in the course and distance sailed and for the error in the chro- 

 nometer rates used at sea. Latitude observations were usually made by three observers 

 with different sextants, following the usual practice at sea of noting the maximum altitude 

 at noon. Sun-longitude observations were usually made by at least two observers with 

 different sextants, six altitudes in rapid succession being measured by each observer. 

 When three stars were available at twilight observations, two or three altitudes of each one 

 were measured in succession, followed by an equal number of altitudes of the same stars 

 taken m the reverse order, so that the mean altitude of each star corresponded very closely 

 to the same instant. In order to expedite the work, readings were entered directly on 

 the forms by a recorder, who also noted the tunes. Both before and after observations, 

 the time-piece used was compared with the chronometer selected as a standard, and as all 

 the chronometers were also intercompared daily it was possible to determine the longitude 

 by each one separately. This, in fact, has been done in the office revision. 



Before leaving port the chronometer rates were determined by comparison with stand- 

 ard-tune signals, or by local time-determinations with sextant and artificial horizon at 

 intervals of seven to ten days. The rates thus determmed were accepted and used until 

 the next port was reached, when the errors resulting from the accepted rates were deter- 

 mined and the preceding longitudes were corrected accordingly. Every longitude of a 

 magnetic station at sea, therefore, depends upon the controlled rates of from tliree to five 

 chronometers. 



The value of the dip of the sea horizon received considerable attention. When pos- 

 sible, back altitudes through the zenith to the opposite horizon were measured, but the 

 application of this method was restricted practically to altitudes of the Sun in the equatorial 

 regions. On Cruise III of the Galilee the dip of the horizon by the Pulfrich dip-measurer was 

 applied to each altitude observed m daylight. Occasionally also on this cruise altitudes 

 were attempted with the gyroscopic octant constructed by Ponthus and Therrode, but the 

 operation was found to be so difficult, owmg to the motion of the ship, that this method was 

 soon abandoned. On a larger and steadier vessel this instrument might be more practical. 

 It may be stated here that the many checks on the altitudes measured during Cruises 

 II and III of the Galilee, and the observations with the Pulfrich dip-measurer, gave no indi- 

 cation of abnormal refraction or apparent dip of the horizon beyond the limits of precision 

 of the instruments used. However, it must be admitted that sextant observations on the 

 Galilee, where checks on the altitudes were available, were made mostly upon the deep sea 

 and in regions where there was no very large difference between sea-temperatures and air- 

 temperatures. 



Sextant index-corrections were determined every few days by star-methods or Sun- 

 methods. 



Specimen observations for the determination of geographic position will be found given 

 in connection with the work of the Carnegie (see pp. 226-230). 



Accuracy of Positions at Sea. 



Accuracy of geographic positions is dependent on so many factors that it is quite impos- 

 sible to define it by exact figures based on any one investigation of numerical results. The 

 first consideration would naturally be the magnitude of the probable error of the measured 

 altitudes, and, if the observation were a meridional one, this probable error would be the 

 probable error of the resultant latitude at the instant of observation. But as it rarely 

 happens that this instant corresponds to the time of a magnetic observation, the observed 



