STUDIES IN ATMOSPHERIC ELECTRICITY 



155 



air over the ocean. When nuclei and large ions are 

 present, it has been shown that for equilibrium condi- 

 tions the equation 



q = an2 + -n^H^n + rjjNn (2) 



applies where Nq represents the number of uncharged 

 nuclei per cc, N represents the number of large ions of 

 one sign (positive and negative ions assumed to be 

 equally numerous), n represents, as stated above, the 

 number of small ions of one sign per cc (positive and 

 negative also assumed to be equally numerous), tjq and 

 T)l are the combination coefficients between the small 

 ions and the uncharged and charged nuclei, respectively, 

 and a is as defined above. This equation reduces to 



and finally to 



q = an'^ + 2 TjjNn 



q = an^ + wNah 



(3) 



(4) 



(2) where Na is the total number of condensation nuclei, 

 charged and uncharged, per cc. The value of the com- 

 bination coefficient oo between the small ions and the 

 nuclei is given in the following relation 



where 



CO = 77i[2/(R + 2)] 



R = No/N = 7)i/tJo 



(5) 



(6) 



From equation (4), assuming q = 1.4 I, where I is the 

 number of ion pairs per cc in the atmosphere, ex = 1.6 

 X 10-6, and using N^ = 1776 and n = 522 from table 3, 

 it is found that the value of co comes out to be 1.0 x 

 10-6. From (5), assuming a value of 5 x 10-6 for t)i 

 (3) the value of R is 8 and, since Nq + 2N = N^, Nq/Na 

 = 0.80. This value of Nq/Na is only slightly greater than 

 the value found for Washington (4), the latter being 0.75. 

 The value of N/Na accordingly comes out as 0.1 from 

 which one would deduce a value of 178 per cc for the 

 number of large ions of one sign over the ocean. 



The cause of ionization over the ocean has been dis- 

 cussed by Swann (5). His discussion involved a question 

 concerning the large-ion content of the air over the 

 oceans. He derived a relation between the ratio of ioni- 

 zation over land and ocean and the ratio of ion content 

 over land and ocean, which was expressed in the follow- 

 ing equation 



(Nl ^ "L>/"s = (il/^s^ ^^^ C^) 



where the subscripts L and S refer to land and ocean 

 values, respectively, and the other notations are as given 

 previously. In arriving at this relation it was assumed 

 that there are no large ions in the air over the ocean and 

 that the value of a is the same over the ocean as that 

 over land. It was further assumed that ng = nL, and that 

 qg = 1.6 I and qL = 6.1 I. The number of large ions per 

 cc over land, on this basis, was found to be about equal 

 to the number of small ions over land, that is, Nl = ul. 

 In reconsidering this matter, on the basis of equation (3), 

 assuming that 2 tj^ = 6 a and that there are no large 

 ions over the ocean, it is found that 



("l + 6NLnL)/ns = q^/qg 



(8) 



In applying this equation, it seems necessary, in 

 view of the large variation in the values of the elements 

 nL, Nl, and qL from place to place, to choose values 

 appropriate to some particular land station. At Wash- 

 ington, D. C. hl and Nl have been measured over a 

 long interval of time. The value of qL has been closely 

 estimated from ionization measurements with a thin- 

 walled chamber. For this station the average value of 

 nL niay be taken as about one-third the average value of 

 ng, whereas the average value of qL appears to be about 

 seven times that of qg. From these values it is found 

 that the number of large ions in the air at Washington 

 is about ten times the number of small ions at this sta- 

 tion. This ratio, though large compared with the value 

 obtained by Swann, is less than one-half the ratio actually 

 found, on the average, to exist at this station. During the 

 warm season (6) of the year, nL = 198, Nl = 4010, or 

 Nl^l = 20.1, while during the cold season of the year, 

 nL = 169, Nl = 6010, or a ratio of Nl to nL of 35.6. 

 The above values were obtained in measurements from 

 October 1932 to September 1933 inclusive. 



Since it was found, from estimates made earlier in 

 this paper, that the large ions over the ocean are not ab- 

 sent but probably average around 178, it seems neces- 

 sary to reconsider the whole matter of ionization over 

 the ocean and allow for the presence of such large ions. 

 Equation (3) may be assumed to hold over land so that 



and, in a similar manner over the ocean. 



(9) 



'Is = ""s 



2 ^sVs 



(10) 



To simplify these equations, let us assume that 2 t^l = 

 6 a = 2 Tjs, that ng = 3nL = 3Ns, and that qL = 7qs; then 



Ql/is 



I 2 

 ("L 



6nT Nt )/(n| + 6n<;N, 



'L"L 



S^S^ 



(11) 



from which it follows that Nl = 31 nL. This ratio of the 

 number of large ions to the number of small ions at 

 Washington is more in keeping with that found, the aver- 

 age ratio from the two seasonal values given above being 

 about 28. The estimate of the value of each element in- 

 volved, including that of the large-ion content over the 

 ocean, appears to have been reasonably correct; other- 

 wise the resulting check would not have been so favor- 

 able. 



On the basis of the mean value of condensation nu- 

 clei and of small-ion content of the air it is possible, as 

 pointed out in the previous paragraph, satisfactorily to 

 account for a reduced number of small ions through their 

 destruction by the condensation nuclei. When, however, 

 one examines the variation in the nuclei content and the 

 corresponding ion content from leg to leg of the various 

 cruises, such an explanation is not so satisfactory. From 

 the results given in table 3 it is seen that the average nu- 

 clei content of the air varies considerably from one leg 

 of the cruise to another. The small-ion content, on the 

 other hand, remains much more nearly constant. This 

 suggests that the nuclei were not always equally effective 

 in the destruction of the small ions. It accordingly ap- 

 pears that, on the average, as the nuclei content of the 

 air increases, the average effectiveness of a nucleus for 



