Ship Deviation-Coefficients 



89 



One value of R has been deduced for the sea-dip-circle position which appUes to the 

 three cruises. It is i?= —0.0010 c. g. s. 



Table ZA.—Horizontal-Intemily Demation-Constants and Parameters for Sea-Dip-Circle Position (Deflection Observations) } 



A specimen computation of the horizontal-intensity deviations for April 14, 1908, at 

 the sea-deflector position, will be found in Table 39, page 92. 



General Remarks. 



On comparing the values of the parameters, group by group, for any one instrument- 

 position, changes will be found for which no complete explanation can be given. They 

 may be ascribed partly to dynamic effects, partly to real changes that have occurred in 

 the magnetism of the ship because of one course having been held approximately throughout 

 the periods of the groups. 



Some of the parameters at the sea-dip-circle position are deduced both from observa- 

 tions of incUnation and horizontal intensity. The differences between these two deter- 

 minations may be referred to instrumental deviations partly, and partly to dynamic effects. 



The deviation-equations for sea deflectors 1 and 2 show, very clearly, the existence of 

 instrumental deviations. The latter may be caused by small impurities in the metal parts 

 or by lack of exact centering of the card in the compasses which had to be used with these 

 deflectors. In view of their small magnitude, they may always be treated as part of the 

 ship deviations. 



Starboard Angle at the Three Positions of the Gaulee Instruments. 



The starboard angle, a, is defined, in treatises on magnetism of ships, as the direction 

 of the resultant of the forces producing semicircular deviation in the compass. It is 

 determined from the equation 



tan a = CV-B'tf (37) 



It lies in the horizontal plane passing through the instrument, and is reckoned positive 

 from the ship's head aroimd by starboard "" • ^ . . ^ ,■ . 



the above equation becomes 



Expressing Cj' and Bt in terms of parameters, 



tan a = 



l(cta„7+|) 



' cZ^-P 



(38) 



From this equation it is evident that if / and c are not zero, the starboard angle, a, is 

 not constant as the vessel saUs around the world. 



'On the greater portion of Cruise I, the deflection method was not available (see pp. 21-22). 

 tions were taken from a table based on the analysis of each separate swing. 



The deviation-corree- 



