284 CARNEGIE INSTITUTION OF WASHINGTON. 



The average annual changes in the values of the magnetic elements, given 

 in table 2, were obtained by using the results of observations made on the 

 Carnegie and Galilee in the vicinity of the intersections of their various tracks. 

 The method employed in computing the tables is the same as that used in 

 discussing average annual changes for the southern Atlantic.^ The charts 

 referred to are U. S. H. O. No. 2406 for 1920, and B. A. No. 3777 for 1917. 



Terrestrial magnetism. J. P. Ault. North American Almanac for 1923, 64-73. 



This article was prepared at the request of the North American Almanac 

 Company, on the general subject of terrestrial magnetism. It includes a 

 brief history of the beginnings and development of the science of terrestrial 

 magnetism, and briefly discusses some of the theories which attempt to 

 explain the origin and manifestations of the Earth's magnetic phenomena. 

 A short description of the Carnegie and her cruises, and of the land work of 

 the Department of Terrestrial Magnetism of the Carnegie Institution of 

 Washington, is also given. 



The Carnegie and her cruises, with special reference to Cruise VI. J. P. Ault. 



This illustrated lecture, giving a general and somewhat popularized account 

 of the work of the Carnegie, was presented on the following occasions : Wash- 

 ington Arts Club, December 15, 1921; Association of American Geographers 

 at the December meeting in Washington; Men's Club of Foundry M. E. 

 Church of Washington, February 17, 1922; College Women's Club of Wash- 

 ington, April 12, 1922; Washington Society of Engineers, October 18, 1922. 



Rotation by magnetization. S. J. Barnett. 



Since the report of 1921 a great deal of progress has been made in improving 

 the apparatus and in determining and reducing the corrections. Three extra 

 coil-systems have been installed to compensate the residual parts of the Earth's 

 field not annulled by the main compensating system, and three variometers 

 have been installed to control them. This and a more nearly symmetrical 

 arrangement of the magnetometer have made it possible to reduce uncertainty 

 as to eddy-current effects. The rotor journals and bearings have been 

 improved, and the rotors have been balanced as far as practicable. Improve- 

 ments in the driving mechanism have been made, and the motor corrections 

 have been more thoroughly studied. The process of demagnetization of 

 the rotors has been greatly improved, making the elimination of several 

 extraneous effects less difficult, and other improvements have been made. A 

 long-suspected systematic error in the 1917 magnetometer observations, 

 causing the results to differ considerably from those obtained by the method 

 of electromagnetic induction in 1914 and 1915, has been found. All methods 

 now agree. Rotating any ferromagnetic substance investigated at one 

 revolution per second is equivalent to placing it in an axial magnetic field 

 with intensity —3.55X10"''' gauss, within about 1 part in 10. This result 

 would be expected if a single type of negative magneton, with ratio of angular 

 momentum to magnetic moment equal to m/e, one-half that of a ring of 

 electrons, were responsible for ferromagnetism. Abraham showed in 1903 

 that a superficially charged Lorentz electron in rotation has the moment 

 and the momentum giving this ratio. If a ring of electrons is chiefly responsi- 

 ble for ferromagnetism, positive magnetons also participate, as stated before. 



An account of the work up to April 1922 was presented to the American 

 Physical Society at the Washington meeting by S. J. and L. J. H. Barnett. 

 (See Phys. Rev., vol. 20, pp. 90-91, July 1922.) 



' See Annual Report of the Director of the Department of Terrestrial Magnetism, Year Book 

 of the Carnegie Institution of Washington for the year 1920, pp. 310-312. 



