Discussions of Ocean Magnetic Work, 1905-16 435 



Table 99 contains the results of the sea magnetic observations in Gardiners Bay, 1909- 

 1915, all referred to mean of day. Details regarding conditions prevailing during the swings 

 of the Carnegie will be found on page 436. From the values of the magnetic elements in 

 the second, third, and fourth columns, it is found that during the period, 1909-1915, the 

 average changes were as follows: 



West declination increased 6! 2 per annum. ") 



North inclination increased 0!9 per annum. > (3) 



Horizontal intensity decreased 0.00053 c.g.s. per annum, j 



Table 100 contains the results of the land magnetic observations at Greenport, 1909- 

 1914, all referred to mean of day. From the values in the second, third, and fourth columns, 

 it is found that during the period, 1909-1914, the average changes were as follows: 



West declination increased 6 '. 2 per annum. ] 



North inchnation increased 1 '. 4 per annum. V (4) 



Horizontal intensity decreased 0.00052 c. o. s. per annum. ) 



The agreement between the average annual changes, (3) and (4), is very satisfactory. 

 Taking the mean of the two sets of values, (3) and (4) , the observed values in Tables 99 

 and 100 are referred to July 1, 1912 (1912.5). Thus the quantities given in the fifth, sixth, 

 and seventh columns of the two tables are derived. The remaining columns contain the 

 residuals resulting by the subtraction of the mean values for 1912.5 from the individual 

 values. The //^-residuals are expressed in terms of 7 = 0.00001 c. o. s. The probable 

 errors of a single result are given at the bottoms of the two tables. It will be seen that 

 the accuracy reached for the swing observations in Gardiners Bay is satisfactory. That 

 the probable errors of the horizontal-intensity observations are practically the same for 

 the land and sea work here discussed, may be accidental; it should be remembered, how- 

 ever, that the sea values of horizontal intensity are derived from two instruments (the 

 sea deflector and the sea dip-circle) and that the time consumed is about twice that spent 

 on the magnetometer shore-work. 



ABSENCE OF MAGNETIC DEVIATIONS ON THE CARNEGIE. 



It was explained in the description of the Carnegie (pp. 160-163) how every precaution 

 possible was taken in the construction of the vessel and with regard to the installations to 

 insure that, at the various places where the magnetic observations were to be made, there 

 would be no magnetic effects of the kind known as "ship deviations," of sufficient magni- 

 tude to be taken into account. In the construction of the auxiliary-power plant, however, 

 it was not found feasible to employ exclusively non-magnetic metal. Thus certain parts of 

 the engine (piston-rings, cam-springs, etc., see p. 162) had to be made of steel. About 3 per 

 cent of the total weight of metal in the auxiliary-power installation is magnetic in its 

 character, but, according to calculation, this 3 per cent could not cause any observable 

 magnetic effects even at the nearest instrument which was mounted inside the after observ- 

 ing-dome (PI. 9), 41 feet distant from the engine. However, some of the stores which 

 entered into the equipment of the vessel for her long voyages, for example, the tin cans 

 containing provisions, were also of a magnetic character, and that, too, of a variable extent 

 during a voyage. Such stores were stowed in the extreme after part of the vessel, at least 

 51 feet from the nearest observing-dome.' Then, again, there were the numerous magnets 



'Special tests made at the laboratory of the Department in Washington, May 7, 1916, showed that the maximum 

 quadrantal deviation in declination that might be expected as the result of a largest number of tin cans likely to be in the 

 provision storage-space on the Carnegie at any one time, at the distance of the nearest magnetic instrument, would be less 

 than 0?02 or less than 1'. An effect of this minor order of magnitude would be practically beyond the limits of accuracy of the 

 ocean observations, though it could be determined, were it worth while, from a number of swings in smooth water and in a 

 region of no local disturbance. 



