426 Bigelow — Meteorological Elements of the United States. 



the pyrheliometer, and the simultaneous temperature at the 

 surface of the earth in the temperate zones, must be treated 

 with the utmost caution, and the entire effect of the temper- 

 ature due to the currents of the atmosphere, prevailing in a 

 given region, must be fully eliminated before the function of 

 equivalent solar heat energy can be known. It is this necessity 

 of separating the effects of circulation from those of radiation 

 which constitutes the unusual difficulty attaching to this re- 

 search, and which has made it necessary to secure very accu- 

 rate data from the observations before a beginning could be 

 made upon the real analysis of the problem. 



4. It is seen that the amount of the annual variation of the 

 temperature is about one degree Fahrenheit on either side of 

 the normal in the long period, and another degree in 

 the short period, making two degrees amplitude or four 

 degrees range, per day for the year, Multiplying this 

 amplitude by 365, we have 730° F. or 405° C. as the accu- 

 mulated temperature between the extreme years. These 

 numbers can be compared with the numbers of the solar prom- 

 inences and the magnetic field as given in our tables, which 

 are the annual sums. The range of the vapor pressure is 

 about 0*040 inch of mercury in the long-range period and 

 0*020 inch in the short-range period. That of the barometric 

 pressure is 0*015 inch for the long period and 0*040 for the 

 short period. Some preliminary studies show that the varia- 

 tions in certain seasons, as the winter or the summer, may be 

 very large in restricted areas, such as the Lake Region, and it 

 will be our next endeavor to carry this analysis one step 

 further in the direction of the periodic variations of the 

 seasons. As stated, the complexity of the function is very 

 great on account of the circulation, but there is good ground 

 for pursuing the research. 



5. In my studies on the magnetic field it was found that a 

 periodic action takes place, having 26*68 days duration, and 

 that a certain type curve emerged from the variations within 

 that period. On attempting to match this curve with the 

 temperatures and pressures of the United States, and other 

 data, it was found that the phenomenon of inversion appeared 

 persistently. A very brief account of this research was 

 summarized in Bulletin No. 21, U. S. Weather Bureau, and 

 on Table 21 the number of the direct and the inverse types 

 found by trials for the magnetic field, the temperatures and 

 the pressures, were recorded. Taking the magnetic field, 

 1841-1894, we found for the ephemeris the curve given in 

 fig. 4 reproduced from Chart 19 of that Bulletin. It is quite 

 probable that a better conformity can be secured by executing 

 more exact computations. For the prominences the intensity 



