Yearly Variations of Terrestrial Phenomena and Solar Activities. 1 î> 



show often marked sinuousity. Still, it is possible in most cases 

 to divide the area of the diagram into three sectors of which the 

 middle one extending along the line x—y is crowded with either 

 O or X . 



The distribution of the stations belonging to each class on the 

 earth's sm*face is shown in Fig. 11. Stations of Types land II 

 are distinguished by the m^arks O and x respectively. Though 

 the density of the station is small, especially in the southern 

 hemisphere, I have tried to draw a series of lines dividing the 

 entire earth's surface into a number of hypothetical zones alternate- 

 ly belonging to Types I and II. The boundary lines of the zones 

 thus assumed are shown in Fig. 11 with thick lines. It will be 

 noticed that one of the poles of the zones falls somewhere near the 

 northern part of Siberia. It is interesting to observe on the other 

 hand that, if we suppose the distributions of land and water on the 

 earth's surface were to be roughly represented by means of the 

 zonal harmonics of the fourth order, the most suitable solution 

 seems to conform rouglily with the division above obtained. This 

 suggests that the relation of the means annual pressure with the 

 solar activity has something to do with the distribution of land 

 and water and arouses the suspicion that the relation may be 

 brought about by the fluctuation of the heat radiation of the sun. 

 In order to advance the investigation in this line, it is necessary 

 to consult the following results with respect to the temperature. 



2) Temper atw^e. 



In the case of temperature, the results were generally similar 

 as in the previous cases, but somewhat more complicated such 

 that it was found in some cases difficult to divide the area of the 

 diagram into three sectors by two lines of demarcations with no 

 considerable sinuousity. If seemed more suitable to distinguish 

 four types, instead of two, in the case of temperature: 



Type I. in which the years of maximum temperature corres- 

 pond to the small values of the specific hemispherical difference, 

 ^=\N~S\I {N+S), in analogy with Type I in the case of 

 pressure. 



