June 11, 1909] 



SCIENCE 



927 



well. Calculating with my own measures 

 of the mobilities and specific numbers, it 

 appears that the determination of the 

 specific number of the small ions from 

 the indications of the Bbert instrument 

 must be from two to four times too 

 great. As for the remaining part of the 

 discrepancy, having used Dolezalek elec- 

 trometers in my own observations, I may, 

 perhaps, be prejudiced in thinking that 

 the metal leaf electroscope of the Elbert 

 apparatus is an unreliable appliance for 

 use in such determinations; in any case 

 the matter must be made the subject of a 

 special enquiry, but in the meantime I 

 have the utmost confidence in my own 

 measures. 



With regard to the other ions, from the 

 very limited series of observations which 

 I have as yet made of the intermediate 

 ones, in air in its natural state, what I 

 have previously called relatively strong 

 ionization is represented by about one 

 thousand per cubic centimeter, while for 

 the relatively weak ionization the number 

 is about two hundred. 



For the specific number of the large ions, 

 a series of 117 observations gives 5,500 as 

 the maximum and 600 as the minimum, the 

 mean for the positives being 1,914 and for 

 the negatives, 2,228. 



The numbers given, with the exception 

 of those for the intermediate ion, are the 

 results of measures with air drawn directly 

 into the testing apparatus without the in- 

 tervention of any pipes; later observations 

 give much higher values for the specific 

 number of the large ions in air led through 

 a considerable length of piping. 



It is now well known, since Lord Kel- 

 vin's memorable work on the subject, that 

 a potential difference exists between the 

 earth's surface and the upper layers of the 

 atmosphere. In the electrical field, which 

 is thus indicated, the ions in the air move 

 more or less steadily in a vertical direction, 



the negatives ordinarily traveling upwards, 

 the positives downwards to the earth. 

 Such a movement constitutes a vertical 

 electric current in the air, the magnitude 

 at any time depending on the air's specific 

 conductivity and the value of the potential 

 gradient at the moment. The specific con- 

 ductivity is represented by the sum of the 

 continued product of the specific number, 

 the mobility, and the charge for each class 

 of ion. An instrument designed by Dr. 

 Gerdien, in which an electroscope is used 

 as in the Ebert apparatus, has been uni- 

 versally employed for such determinations 

 as have been made of this important quan- 

 tity. It measures the sum of the conduc- 

 tivities due to each type of ionization, and 

 the calculation of the result from observa- 

 tions with the apparatus is not affected by 

 the discovery of a new class of ions. The 

 complexity of the natural ionization, how- 

 ever, prevents the instrument being used 

 to accurately determine the specific num- 

 ber of the small ions. The average value 

 of the specific conductivity of the air in 

 other parts of the world, as given by the 

 Gerdien apparatus, is about 10"* in electro- 

 static units.^^ The magnitude of this 

 quantity can be calculated from the meas- 

 ures of the mobilities and specific numbers 

 of the ions, and the average specific con- 

 ductivity of the air at Sydney, so de- 

 termined, is only about one tenth of the 

 value just stated. Here again there is a 

 considerable discrepancy between my own 

 and other measures which has yet to be 

 investigated. 



"With increasing knowledge we can look 

 forward to developments of importance to 

 meteorology in connection with ionic ob- 

 servations ; just now it is doubtful, I think, 

 if valuable effort is not being wasted as a 

 result of over-confidence in the present 

 state of the art. 



" Gerdien, Oessell. Wiss. Oottingen, Nachr., 

 Math-Phys. Klasse, 1, p. 77, 1907. Dike, Terr. 

 Magn. and Atmoa. Elect., September, 1908. 



