It is well known that the ScS wave appears very distinctly in the seismograms of deep-focus earthquakes recorded near the epicenter. Contrary to our expectation, we have scarcely observed the sScS wave. The amplitude decrease of the seismic wave due to the reflection at the surface is first examined by means of the observational data, but it is proved that the scattering of waves caused by the reflection can not so effective as to make the sScS wave so weak. Thus the author has tried to interprete this phenomenon, “defect of S wave”, as an effect of a certain layer in the earth's crust. In some parts of the crust, which are of course shallower than the depth of focus, -possibly these exist just near the focus or in a very shallow place-there must be a special surface where the seismic waves are subjected to scattering according to the angle of incidence. For, the particles of the earth's crust are considered to be allowable to give relative tangential movements between the just upper and lower sides of this surface. In the second part of this paper some calculations are done concerning this problem. The boundary conditions which have to be satisfied at this surface are assumed as follows:- Thus the amount of the scattering of elastic waves due to the existence of such a surface is calculated for two cases when the incident waves are distorsional and dilatational respectively.
As the measurement of the earth current has not so small difficulties and only a few stations are constantly recording the phenomena, it is impossible to investigate fully the distribution of the earth current over the globe. But roughly speaking, in an average state, the earth current is directed towards the equator in the meridional plane of the globe. The variation in its east-west component is more conspicuous than that in its north-south component. To explain these states of the earth current various theories have been proposed without a full success. In this paper the author considers the rôle of the thunderstorm upon the earth current and shows some relations existing between them in its average state. Owing to the luck of data available, this may perhaps be mere a suggestion in the explanation of the earth current. Recent investigations of the polarity of thunderstorm by Klydnograph show us that the bottom of thundercloud has a negative charge making under it the great potential gradient opposite to the fine weather potential gradient. This state necessarily causes the upward electric current of considerable strength by the point discharge from the tops of trees, grasses and even sea waves. Though the duration of the thunderstorm is very small compared with the total hour of an year, the greatness of the point-discharge electric current contributes most strikingly to the average state of the vertical electric current near the earth's surface. In the equatorial region the earth's surface receives an excess of negative charge by the disturbed weather, while an excess of positive charge at the polar caps. This conclusion is mainly due to the unequal distribution of thunderstorm over the globe, hence of the upward point-discharge electric current. (Table 2) Then the unbalanced distribution of the charge over the globe should contribute to the occurence of the electric current in the semi-conducting medium, the earth, in order to make the charges distribute equally over the globe. This is the reason proposed by the present author of the generation of the earth current. The average state of the earth current coincides in its direction and in its variation (very roughly in its intensity, too) with the observations. Hence if the earth current is mainly due to the thunderstorm activity, in its diurnal variation, too, this relation should be hold very approximately. Namely the earth current must point to the most thundry region at hours which are very favourable for the occurence of thunderstorm at that region. The observations of the earth currents at Kakioka and Toyohara Observatories show us that the earth currents point generally to the tropical islands such as Java, Borneo etc and the direction changes gradually westwards as the time advances from 9h. (1. m. t.) to 17h. indicating the possibility of the existence of the mutual relations between the earth current and the thunderstorm activity in the tropical island (Fig. 3)
This paper contains the following results: (1) The undulation of water with period of eleven minutes has been recorded by a tide-guage, installed sometimes at few places in Lake Hamanako. Comparing the observed amplitude of oscillation with our theory we found that they are not in agreement at the northern parts of this lake, near the mouth of Lake Inasahosoeko. He has discussed them as the mouth-effect of Lake Inasahosoeko, northern small part of this lake. (2) The principal cause of the seiches of eleven minutes is due to the Westerly gales blowing after the passage of cycloue, and the period of eleven minutes is difficult to explain by the undulation of water in a bay with narrow mouthes.