Comparisons of Magnetic Standards, 1905-14 277 



0.0001// in horizontal intensity. We have been able to refer the standards of nearly all of 

 tlie chief magnetic observatories to these international standards (see Tables A and B). 



The arduous work being over, we can now give, on the basis of the three observatories 

 which show the smallest horizontal-intensity corrections, and for which the grade of relia- 

 bility a could be assigned throughout, a definition which will answer all requirements. The 

 International Magnetic Standards, if desired, may be taken to be the absolute magnetic 

 instruments at either one of the three observatories, Kew, Potsdam, or Wasliington 

 (Department of Terrestrial Magnetism of the Carnegie Institution of Washington), pro- 

 ^'ided the approximate corrections given in Table C be applied algebraically, remembering 

 that the signs of the corrections are positive for east declination and inclination of north 

 end of needle below horizon. 



Naturally the more of these observatories any one employs in determining the rela- 

 tions between his magnetic standards and the International Magnetic Standards, as above 

 arbitrarily^ defined, the more correct will be his results. WTiether the corrections at Kew 

 and Potsdam for 1910, given in Table (', also hold for 1914 depends upon whether any 

 change has been made at these observatories in the instruments or in their constants during 

 the interim. As alreadj^ stated elsewhere, the corrections, on absolute standards, for the 

 declination and inchnation may be in error by O'.l or 0'.2 and for the horizontal intensity 

 by about 0.0001//. Accordingly, no undue significance is to be attached to the circumstance 

 that the corrections for the Washington standards arc apparently the sjiiaUcst of the 3 observa- 

 tories, nor to the fact that the apparent zero correction of the Washington standard for horizontal 

 intensity happens to be the mean of the corrections for Kew and Potsdam. 



From other studies we have made we are confident that the mean result from a combina- 

 tion of the observatories in Table C wUl come as close to an absolute magnetic standard 

 as is possible by any magnetic or electric method for determining the magnetic elements. 



CONCLUDING REMARKS. 



In conclusion, we desire to say that we ourselves regard the past work rather in the 

 nature of a preUminary investigation, lea\'ing to the future and to others a more complete 

 examination into the various questions involved. It should be distinctly imderstood that 

 there is no recommendation in this report that observatories adopt our international mag- 

 netic standards, or the corrections of their instruments on these standards, as shown in 

 Tables A and B. If, however, any observatory is induced by our results to purchase its 

 instruments with especial attention to their design and construction, to make an inde- 

 pendent investigation of their accuracy, and to arrange for the scientific control of the 

 instnnnental constants, our purpose has been accompUshed. AYliile any desire to establish 

 universal magnetic standards is disclaimed, it has been found in a number of instances that, 

 when the observatory staffs have been led to investigate the causes of the rather large differ- 

 ences which were disclosed in the comparisons at their observatories, they have found 

 sources of error, the remedying of which has effected improvement. 



