DEPARTMENT OF TERRESTRIAL MAGNETISM. 



325 



It is to be remarked that only on one occasion during the whole cruise was 

 a negative potential gradient observed, although observations were made 

 frequently while it was raining. Usually during rain the potential gradient 

 was very high, often exceeding the range which the electroscope would measure 

 but it was always positive. On the one occasion when a negative potential 

 gradient was observed the sky was nearly covered with clouds, but there was 

 no rain. One very striking thing shown by the results is the large values for 

 the conductivity and for q. The mean values of the conductivity found over 

 land up to the present is not much greater than 2 X 10~^, while q is less than 

 1.20. It is not probable that these large values of the conductivity can be 

 due to large amounts of radioactive emanations in the atmosphere, for no 

 clear relation between the radioactivity of the atmosphere and the conductivity 

 was found. In view of the fact that, as shown in the paper, the errors in 

 the measurements are such as to decrease the value found for the conduc- 

 tivity, it is deemed safe, taking everything into consideration, to say that, 

 over the oceanic regions where we observed, the average value of the conductivity 

 is not less than 3.26 X10~*, q is not greater than 1.2, and that the mean value of 

 potential gradient near the surface of the water is of the order of magnitude of 

 120 volts -per meter. During the passage from Falmouth to New York the 

 observations of the radioactive content of the atmosphere formed a fairly 

 complete set. The mean value for this cruise, of the activity, expressed in 

 Elster and Geitel units, is 12.3, and the nature of the deposit on the wire was 

 such that the activity decayed to half value in about 40 minutes. 



Table 2. — Mean resulting values. 



It has been attempted to discover any relation^ which may exist between 

 the various atmospheric-electric elements or between these and the various 

 meteorological factors. As a rule, the relations which have been found agree 

 with those which have been previously known to exist on land. In most of 

 the passages, both the potential gradient and q decrease with increase of the 

 conductivity, and in the final mean this relation is shown very clearly. In 

 the various portions of the passage from Tahiti to New York, large values of 

 the conductivity correspond to small values of the relative humidity and vice 

 versa. In the first half of the cruise, from New York to Tahiti, this relation is 

 not clearly indicated in the separate portions, but it is revealed in the final 

 means from New York to Colombo, and from Manila to Tahiti. There is also 

 a clear relation between the conductivity and temperature, increase of tem- 

 perature corresponding to the increase in conductivity. It is probable that the 

 relation here is somewhat indirect, and is bound up with the effect of the tem- 

 perature on the moisture conditions in the air. Increase of the conductivity 

 is accompanied by little change in the absolute humidity. It was thought that 

 possibly solar radiations might in some way affect the conductivity at the 

 surface of the Earth, so that the observations of the cloudiness of the sky 

 were grouped according to the conductivities. There does not appear to be 



