RESULTS OF COMPARISONS OF MAGNETIC STANDARDS, 1915-1921. 



By J. A. Fleming. 

 EXPLANATORY REMARKS. 



This report contains, in continuation of the report for 1905 to 1914 in Volume II 

 (pp. 211-278), the results of the various intercomparisons of magnetic standards ob- 

 tained by the observers of the Department of Terrestrial Magnetism from 1915 to 1921, 

 inclusive, the world over, at magnetic observatories and in the field. Preliminary 

 summaries of the results of some intercomparisons in recent years by others and exhibit 

 of the relation of these results to those of the Department are also given. As hereto- 

 fore, it has been found that, for one reason or another, magnetic instruments may differ 

 among themselves by quantities far exceeding their observational errors. Sometimes 

 these differences can be referred to imperfect values of the instrumental constants, at 

 other times they are to be ascribed to causes inherent in the instruments themselves. 



When, therefore, a general magnetic survey of the globe is to be conducted on a 

 common and consistent plan, it becomes a matter of importance to know how far instru- 

 mental constants and reductions to standards, as determined at one place, can be relied 

 upon in other places where the magnetic elements are considerably different, or what 

 changes may be expected during strenuous field campaigns, such as must be carried out 

 in more or less unexplored countries. If, furthermore, the magnetic results obtained 

 by various organizations, using instruments of greatly different construction, are all to 

 be reduced to a common basis, it becomes increasingly important to have the requisite 

 data at hand for the proper correlation of all work. 



Accordingly, whenever opportunity was presented during 1915 to 1921 in the course 

 of field work to obtain such correlation data, the observers of the Department have 

 carried out, with the cooperation of the staffs of the various observatories visited, series 

 of intercomparisons of magnetic instruments. Whenever circumstances permitted, the 

 method of intercomparisons of magnetic instruments described in Volumes I (pp. 19-20) 

 and II (pp. 211-212), was followed. Only occasionally did it happen that, for lack of 

 time or other reason, the full program had to be curtailed. The method as prescribed 

 in the general directions for magnetic measurements given observers of the Department 

 is shown in the following paragraphs: 



"Disturbed sites are avoided for the intercomparisons of instruments, but this can not always 

 be done in the field, as for example, in the ocean work where islands, or ports, often afford the only- 

 opportunity for the desired comparisons. If the preliminary examination has shown the existence 

 of pronounced local magnetic disturbance, and if another site is not available, it is arranged that, at 

 the same station, the magnetic systems of the various instruments are in the same horizontal plane. 

 Should this procedure not be possible, then the height of magnet from a suitable reference point, 

 e. g., from the top of a stake driven into the ground, is carefully noted and determinations are made 

 at each station to find the necessary corrections for the various levels in which the intercomparisons 

 had to be secured. With these precautions, it has been found that results of sufficient accuracy 

 for field work can be obtained. 



"Generally but two stations are required, which, unless already named, as may be the case 

 at observatories, are designated A , B. For observatory work B is the auxiliary station and A the 

 regular observing-pier; at some observatories different piers or stations are used for the various 

 elements and intercomparisons for each particular element must be made accordingly. The azi- 

 muth lines for both stations are preferably referred to the same determination of azimuth, especially 

 when no exchange of stations can be effected. Whenever possible both stations arc placed in the 

 same azimuth line and the same mark is used at each, thus assisting in the avoidance of extraneous 



395 



