This paper was published in additional pages of “Annual Report of the Central Meteorological Observatory of Japan-Magnetic Observations for Years 1917 and 1920, ” and will reappear in “Japanese Journal of Astronomy and Geophysics, Vol. III.” An abstract in Japanese is written in this journal (this Vol. p. 173). Chief conclusion in this paper is proof of proportionality of the square of mean range and the mean variability. Some discussions about the mean of the squares of range and about natures of hourly range are added. Theories proved in this paper may also be used to discuss variations of any natural phenomena.
In the preceding paper, (Journ. of the Meteor. Soc. of Japan, Ser. 2, Vol. III, No.3), by drawing a serie of isochronal lines using the results of seismic observations made at the various meteorological stations of our country, the author has found that the velocity (V1) of condensational seismic wave propagating in the direction perpendicular to the arc. of the Japanese Island is greater than that (V2) of the wave propagating along it. These velocities of condensational wave V1 and V2 cstimated from isochronal lines are 9.6 and 6.5km. per sec. respectively. But the data used in the preceding paper are confined to the results of observations of earthquakes which have their epicentres along the Pacific coast. In the present paper, he dealt with five semidestructive carthquakes which occurred on the Japan Sea, side. Of these five quakes, the destructive shock of North Tazima occurred on May 23, 1925 is most notable one. Using these data and with the method deseribed in the preceding paper he has drawn the following conclusions:- (1) In carthquakes having their epicentres in the Pacific Ocean, velocity of seismic condensational wave propagating along the arc of our island is smaller than propagating perpendicular to it. But in all the earthquakes having their epicentre in the Japan Sea, the difference of the velocity in these two directions is smaller than that in the Pacific quakes. (Figs. 1, 2 and 3) (2) Velocity of the condensational seismic wave propagating along the coast of the Japan Sea is greater than that propagating along the coast of the Pacific Ocean. From these two results. we may conclude that the Pacific side of the Japanese Island is stretched and the Japan Sea side is compressed. (3) According to the results of observation of Tams and Angenheister, the velocity of the Seismic surface wave is greater when it propagates along the ocean bed than that along the continent by 25 percent. But it seems to me that the same conclusion may hold good even in the case of the condensational wave passing through upper strata of the earth's crust. This result may in accord with the thcory of isostasy as well as with Wegener's theory of continental drift. Moreover, from above results it may be considered that the depth of the discontinuity surface which is generally estimated to be 60km. is different according as it lies under the continent or under the ocean. Namely, under the ocean bed the discontinuity surface is shallower than that under the colltinent, but the confirmation of this conclusion is left to further studies. (4) From the results, that our great island is stretched at the Pacific coast and consequently compressed on the Japan Sea side, it follows that there may exist many large cracks perpendicular to the are of the island and the Kii Straits running from north to southward between Sikoku and Kii Peninsula is one of these cracks. When condensational seismie wave passes along the straits there oceurs retardation of velocity owing to its less elastic constant. This phenomenon is also observed when condensational wave passes along the Inland Sea from east to westward by the earthquakes which occurred at the coast of the Pacific Ocean. (5) The so-called “Lauf-zeit”curve for the condensational wave, consists of the three sets of curves for the following three kinds of waves:- P: Mohoroviêié wave, Pa: P-wave passing along the are of Japanese Island, Pp: P-wave passing perpendicularly to the are of the Island. (Fig. 4 and 5) (6) Seismic long wave (L) may positively be composed of S-wave as was stated by A. Mohoroviêié and afterwards confirmed by Mr. T. Isikawa of this Observatory. This fact is also verified in my papers.
The author tested the breaking limit of gummy balloons made in Japan. Those having weight of 23gr. break at about the diameter of about 75cm. While the diameter suitable to the asccnt of 150m/min is about 54cm. Leakage of hydrogen into the air from the balloon with this diameter is about 4gr. per hour. By equating the bouyan_??_y with the resistance of during the ascent of balloon the author obtained an equation where D is the diameter of the balloon, V its ascending velocity assumed to be a constant, W+ω is its weight with hydrogen, ρ is the density weight of the surrounding air and K is the constant of resistance. Solving this equation he obtained the value of D, with corresponding ρ which is a known function of height for average atmosphere. So that the diameter of the balloon, at any hight can be given. Hence the height at which the balloon comes to have the diameter of breaking limit can be obtained. He, however, not yet check the formula with the data of aetual observation.
The author compared the seasonal amount of precipitation observed at Tu in Japan for 43 years with sunspot number. He found that the maximum or minimum of sunspot number occurred in the year of, or one year before or after the maximum or minimum phase of precipitation. The correlation coefficient and ratio between the amount of precipitation and sunspot number were given as in the following table This shows that the correlation is not linear but is of some parabolic relation. (Original in Japanese pp. 194-197)