264 CARNEGIE INSTITUTION OF WASHINGTON. 



is not valid. Since most of the special lines installed for the study of earth- 

 current phenomena are necessarily limited in extent, the object of the experi- 

 ments described was to investigate the nature and order of magnitude of 

 some of the spurious effects which may function under these conditions. 



In one group of experiments a method similar to that used by Des Coudres, 

 in his study of the E. M. F. produced by the action of gravity in salt solu- 

 tions, was used to investigate the possible presence of an effect due to dif- 

 ference of pressure at the electrodes. It was found that a tube filled with 

 soil and provided with an electrode at either end showed a component of the 

 total E. M. F. which when the tube was vertical tended to form a cathode 

 at the lower electrode regardless of which electrode was involved. The 

 order of magnitude of this effect was found to be sufficient to account for 

 certain observed phenomena which appear inconsistent with physical prin- 

 ciples, provided the effect exists as a general phonomenon in nature. For 

 example, it would account for what has been called a vertical earth-current 

 of density much larger than that of either the horizontal earth-current or 

 the vertical air-earth current. It would also account for the fact that many 

 observers have found the "vertical earth-current" to be nearly free from 

 diurnal variation. 



Inasmuch as our knowledge of the interrelations between earth-currents 

 and terrestrial magnetism must come from a study of the variations involved, 

 it is important to know to what extent observed earth-current variations 

 are genuine. One phase of this question was investigated at Washington (as 

 indicated on page 254 of annual report of the Director of the Department of 

 Terrestrial Magnetism for 1917) by means of two pairs of similar electrodes, 

 each pair being in the same vertical, one pair being inclosed in an insulating 

 tube and the other being free, the corresponding electrodes of the two circuits, 

 however, being at the same level. Continuous measurements of P. D. and 

 temperature difference were made on these circuits extending over a period 

 of about 8 months in 1916 and 1917. The results show that most if not all 

 of the diurnal variation which some observers have ascribed to a vertical 

 earth-current is very probably due to variations in temperature difference 

 between the electrodes. 



The spurious effects introduced by temperature-difference variations were 

 found to be greatly increased and reversed in sign when the soil in contact 

 with the upper electrodes was frozen. 



While the results are strictly applicable only to the actual installation 

 employed during the experiments, they show that the effects which may 

 result from temperature difference at the electrodes of an earth-current line, 

 and from variations in this difference, may, for short lines, be of the same 

 order of magnitude as the quantities to be measured and with which they 

 are associated. In the light of these experiments it appears that some of the 

 phenomena which various observers have ascribed to a true earth-current 

 must be largely caused by such temperature-difference effects. Although 

 proper emphasis is usually placed on the employment of non-polarizable 

 electrodes for earth-current measurements, it appears not always to have 

 been appreciated that with the electrodes customarily used this does not 

 prevent the introduction of temperature-difference effects. 



Results of dip-of-horizon measurements made on the Galilee and Carnegie, 1907-1917. 

 W. J. Peters. Terr. Mag., vol. 2.3, 47-60 (June 1918). Washington. 



Since all astronomic positions at sea depend upon sun-altitudes or star- 

 altitudes measured from the horizon, it is evident that the precision of the 

 determinations of the geographic positions for the magnetic stations at sea 



