DEPARTMENT OF TERRESTRIAL MAGNETISM. 281 



With the apparatus described, observations have been made for some 

 months, and perhaps the most interesting feature of these observations is the 

 importance which they assign to the ions of mobihty smaller than those which 

 aie ordinarily supposed to figure in measurements of the conductivity. With 

 the potential of 250 volts applied to the subsidiary attachment, ions of mobil- 

 ity greater than about 0.1 cm. per second per volt per centimeter should be 

 cut out during the intervals of application of the potential. It frequently 

 results that at night, and during periods of rain, the conductivity mounts to 

 an abnormally high value, and that the ions mainly responsible for this phe- 

 nomenon have a mobility at least as small as about 0.1 cm. per second per volt 

 per centimeter is borne out by the fact that the apparent conductivity is not 

 greatly reduced during the periods of application of the potential of 250 volts. 

 The existence of a very high conductivity as a result of ions of very small 

 mobility necessitates, of course, the presence of a very large number of these 

 ions. It is not improbable that the very high ionic densities found on the 

 occasions referred to are associated with a reduction of the rate of recombina- 

 tion of the ions as a result of their combination with nuclei. This would result 

 in an increase in the number of ions present in the steady state. It is of inter- 

 est to notice that the new steady state would take some considerable time to be 

 set up. For, suppose the unipolar conductivity were of the order of 10^ 

 E. s. u., as is sometimes the case. If the mobility of the ions responsible for 

 this conductivity were less than 0.1, there would be at least 60,000 ions per 

 cubic centimeter. If the rate of formation of ions were as great as 10 per 

 cubic centimeter per second, nearly 2 hours would be necessary for the pro- 

 duction of these ions, even if there were no recombination. 



Supplementarv report on atmospheric-electric observations made aboard the Carnegie 

 from May 17, 1916, to March 2, 1917. W. F. G. Swann. 



The general scheme of the observations is similar to that described in Volume 

 III, of Researches of the Department of Terrestrial Magnetism, but it was 

 possible to secure a greater number of observations on the diurnal variation 

 than heretofore. 



Table 7 (p. 282) gives the dates of arrival at and departure from various 

 ports, and the observations are grouped according to the parts of the cruise 

 specified by these ports. The figures in parentheses give the numbers of 

 days upon which observations were made, and the definition of the quantities 

 in the table is as follows : 



n+ and ?j_, the numbers of positive and negative ions per cubic centimeter. 



X+ and X_, the conductivities for positive and negative ions. 



v+ and i'_, the specific velocities of the positive and negative ions. 



X, the potential gradient. 



i, the air-earth conduction ciuient density. 



R, the number of pairs of ions produced per cubic centimeter per second 



in a closed copper vessel of 27 liters capacity. 

 Ra. Em., the radium-emanation content of the air. 



The means for all the observations are given in the line next to the last, and 

 in the last line are given the means for all the observations already pubhshed 

 for Cruise IV of the Carnegie through March 1917. The numbers in both 

 cases are uncorrected for diurnal variation, and correspond to a mean time of 

 about 9'5. 



In obtaining the mean value of the radium-emanation content from Easter 

 Island to Buenos Aires, a few exceptionally large observations taken near the 

 coast on approaching Buenos Aires, and obviously influenced by the prox- 

 imity of land, have been omitted. 



