I70 TRANS. ST. LOUIS ACAD. SCIENCE. 



globe. A galvanometer of suitable strength may perhaps answer 

 the purpose as a more convenient measuring instrument. I will 

 give it a trial this year and report on it at some future time. 



As an important addition to our knowledge of atmospheric elec- 

 tricity, I consider the discovery, resulting from my observations, 

 of its threefold periodicity — the daily, the annual, and the secu- 

 lar (perhaps better called the cyclical). The daily one has been 

 partially guessed at by some observers without proof positive, 

 while the two others were not even thought of. The daily waves 

 of positive atmospheric electricity are not so constant that they 

 would show themselves every day : meteorological influences — 

 especially temperature, humidity, and direction and force of wind 

 — modify it so often, that the regular order of electricity appears 

 even sometimes reversed ; but in the course of a week, a month, or 

 a year, these conflicting influences counterbalance each other, so 

 that the steady undercurrent of periodicity becomes always visi- 

 ble and unmistakable. Thus we see in Table 3, in which the 

 mean of the six daily observations is given for each year and for 

 the twelve years successively, at the different hours of observation, 

 two maxima and minima of electricity, with two intermediate 

 points. The first maximum appears at 9 a.m., the first minimum 

 at 3 p.m. ; the second maximum at 6 p.m., the second minimum 

 at 9 p.m. The daily periodicity is thus sufficiently proved ; a trial 

 of 12 years with about 25,000 observations leaves no doubt on 

 that point. 



The second periodicity is the annual, exhibiting itself in the 

 regular, gradual increase in quantity or intensity of electricity 

 during the six colder months of the year, culminating generally 

 in January ; and of a gradual decrease of electricity during the 

 six warmer months, with its lowest point generally in July. 

 Slight deviations between two consecutive months will take place 

 in different years, but in the mean of the twelve years strict regu- 

 larity is preserved, as Table 1 will show. The existence of the 

 annual electricity is thus as clearly proved as the daily. But a 

 point, which I raised in a former paper in regard to it, is not yet 

 decided. I asked myself, if the regular increase of electricity in 

 the colder season (and vice versa) was caused directly and ex- 

 clusively by the change in temperature, or by some internal con- 

 nection between atmospheric electricity and terrestrial and solar 

 magnetism, of which I consider atmospheric electricity an emana- 

 tion and offspring. If the latter hypothesis should be correct, our 



