The explanation for this probably lies in the fact that Japan has practised civilised arts for a longer period than many of the European and other countries mentioned in the table. In Japan, no doubt, the records of later years have been more perfect than they were in early times, but this, although so potent an effacer of what was probably the true state of natural phenomena in the case of Europe, has not quite obliterated the truth in Japan; for instead of an apparent increase of seismic energy since early times it shows a slight decrease. To draw up a table of earthquakes such as the one which has just been given, and then, after the inspection of it, draw conclusions as to whether there has been an increase or decrease in seismic energy, is, however, hardly a just method of reasoning. The earthquakes, taken as they are, for the whole of Japan, represent a collection of places some of which are 1,000 miles apart. When we consider that many earthquakes which occurred at one end of this line were never felt at the other end, in order to justly estimate the periodicity of seismic phenomena it would seem that we ought either to take some particular seismic area or else the whole world. The particular area which has been taken is that of Kioto in Central Japan, and the earthquakes which have been felt there are enumerated in the table. In order to show the variation in seismic activity of this district a curve has been plotted, fig. 35, with ordinates equal to the values given for the Kioto earthquakes during succeeding centuries. The upper points of these ascending and descending lines are joined by a free curve. The lower points are similarly joined. The points of bisection of ordinates drawn between these two curves are taken as points in a curve to show the true secular change in seismic energy. [Illustration: FIG. 35.—Curve of Seismic Intensity for Kioto.] By looking at this wavy line it will be seen that the intervals between maxima and minima are closer together in early times than they are later on. Thus, between the eighth century and the ninth century, points of maximum and minimum seismic efforts occurred at times a century apart, whilst later on, from the eleventh to the fifteenth century, they were at intervals of 300 years apart.