686 



NATURE 



[November io. 1923 



Letters to the Editor. 



\Tht Editor does not hold himself responsible for 

 opinions expressed by his correspondents. Neither 

 can he undertake to return, nor to correspond with 

 the writers of rejected manuscripts intended for 

 this or any other part of Naturk. No notice is 

 taken of anonymous communications.} 



The Relation between Solar Activity and 

 Atnnosplieric Electricity. 



Dk. Chree in his reply (Nature, September 8, 

 p. 3()i) to my communication on solar activity and 

 atmospheric electricity (Nature, August ii, p. 203) 

 first makes reference to the status of the question as 

 to the effect of sun-spot activity on the secular change 

 of the earth's magnetism. Those who have in- 

 vestigated this question have reached apparently 

 contrary conclusions according to the phenomenon . 

 examined, the data used, and the method employed 

 by the individual investigator. It would require too 

 much space to enter into detail as to the reasons for 

 the discordant results. Let it suffice here to state 

 that Dr.- Chree and I have investigated different 

 phenomena. Thus Dr. Chree, on the basis of the 

 Kew data alone, concluded, contrary to Leyst, that the 

 secular change of the magnetic declination did not vary 

 markedly, if at all, with sun-spottedness. 



Such a restricted investigation could, of course, not 

 be accepted as settling the broad question as to 

 whether any appreciable change in the direction of 

 magnetisation of the earth may be related to solar 

 activity changes during a sun-spot cycle. I, on the 

 other hand, confined my investigation solely to the 

 question whether there was an appreciable change in 

 the earth's intensity of magnetisation which might be 

 associated with change in solar activity during the 

 sun-spot cycle. Instead of relying upon the data 

 from one station alone, I used the intensity data from 

 eight stations distributed around the globe, namely : 

 Kew, Potsdam, Pola, Bombay (Colaba and Alibag), 

 Honolulu, Sitka, Cheltenham (Maryland), and Porto 

 Rico. Regarding the various questions which must 

 be considered in investigations of this character, the 

 interested reader may be referred to my paper on the 

 subject,* at the conclusion of which the intention was 

 stated of making a still more comprehensive examina- 

 tion, as soon as additional data were available. 



With regard to the difference in the method or 

 formula used by Dr. Chree and myself when investi- 

 gating a possible relationship between solar activity 

 and atmospheric electricity, let me state briefly the 

 assumptions involved. Dr. Chree in his paper* 

 adopts a formula which assumes that there is strict 

 synchronism between the phenomena of sun-spotted- 

 ness and atmospheric electricity, and that for the same 

 sun-spot number, during the first and second halves 

 of the . cycle, for example, the atmospheric-electric 

 element investigated should have precisely the same 

 value. In my formula (Nature, August 11, p. 203) 

 1 introduced a term, provisionally called a time- or 

 cycle-term, which was intended to take into account, 

 approximately, a possible a-cyclic effect in atmo- 

 spheric electricity during a sun-spot cycle such that 

 the atmospheric-electric element considered, barring 

 other disturbing causes, would not have precisely the 

 same value for the same sun-spottedness. With the 

 aid of this additional assumption, which does not 

 appear unreasonable in view of similar effects in other 

 geophysical and cosmical phenomena, an improved 



» Terr. Mag. and Atmos. Eltct., vol. 23 (1918), pp. 1-22 and 61-68. 

 • Proc Phys. Soc., London, vol. 35, part 3, Apnl 15, 1923, p. 132. 



NO. 2819, VOL. 112] 



mathematical representation was obtained and higher 

 values of the correlation coefficient were derived than 

 those which Dr. Chree had found. No futile attempt 

 was made to get an exact representation by unduly 

 multiplying the number of unknowns to oe deter- 

 mined by the method of least squares. The desire 

 was merely to obtain, in accordance with the best 

 practice, a sufficiently satisfactory representation of 

 the observed facts with the fewest possible unknownn ; 

 the general concordance in the derived unknowns 

 from widely separated stations would appear to l>e 

 ample justification of the formula employed. 



It must be realised that no method of applying an 

 a-cyclic correction, due to an undiscovered cause, can 

 be made perfect ; however, when more extensive data 

 for several sun-spot cycles are available, no doubt 

 improvements may be made. In this cf.^-^-v.,... ;♦ 

 may be remarked that Dr. Chree's metho 

 ing a-cyclic corrections to the observed nia^..- „ ..„,1 

 electric diurnal variations has not yet been generally 

 adopted. However, no great refinement in mathe- 

 matical method is requisite to show, even f' i 

 at present available, that a definite i- j> 

 between solar activity and atmospheric electricity i» 

 sufficiently plausible to merit careful attention. Some 

 of the evidences have already been cited in my previoo 

 communication and reference has been made to .t 

 later and more complete pap>er.* 



Dr. Chree directs attention to some low values of 

 the atmospheric potential-gradient at the Ebro 

 Observatory ; by reference to the obser\-atory 

 bulletins it will be found that recent low values, 

 especially during the period June -October 1922, 

 were not unnoticed by the Obser\'atory, and that 

 possible artificial disturbing causes were investigated, 

 as the result of which some changes have been made. 

 The Observatory will doubtless make such additional 

 tests and redetermination of reduction factor as may- 

 be requisite in the circumstances. This later in- 

 formation from the Ebro Observatory had not 

 been received at the time of my previous communi- 

 cation, in which data only to 1921 inclusive were 

 utilised. (The date for No. 11 in Table 2 of my 

 previous communication should be 1921-5, instead 

 of 1921-1.) 



I am glad that Dr. Chree is helping to keep ali\ e 

 an interest in the highly important question as to 

 possible variations in atmospheric electricity which 

 may have to be associated with changes in solar 

 activity. We may rest assured that until this question 

 is definitely settled no complete theory of the origin 

 and maintenance of the earth's electric charge can 

 be definitely formulated. My main purpose appears 

 to have been accomplished, namely, to bring back into 

 the literature a question for reinvestigation which was 

 actively discussed more than a half-centur\' ago and 

 then dropped for want of sufficiently accurate data of 

 the requisite extent. It is hoped that the renewed dis- 

 cussion will contribute towards the multiplication of 

 atmospheric-electric stations where every possible 

 care will be taken to ensure continuity of strictly 

 comparable data for as long a period as possible. 

 Among other precautionan.- measures, more frequent 

 and more extensive controls, than is at times the case, 

 of the factor for reducing observed potential -gradients 

 to an infinite plane, are requisite. 



Louis A. Bauer. 



Department of Terrestrial Magnetism, 



Carnegie Institution of Washington, 



October 5. 



* It is expected that this paper may appear in the December, 1923, isiue of 

 Terrestrial Magnetism ani Atmospheric Electricity, when it is hoped that, in 

 addition to other data, those for 1922 at Kew and Eskdalemuir will be 

 available. 



