August 1 8, 1S87] 



NATURE 



379 



To this statement it is only necessary to add that Dr. E. Schunck 

 will preside over Section 13 (Chemistry), and Prof. A. Newton 

 over Section D (Biology). 



THE JAPAN EARTHQUAKE OF JANUARY 15, 

 1887.^ 



COOX after the occurrence of the earthquake of January 

 '^ 15 last, which caused considerable damage to pro- 

 perty in and near Yolvohama, the authorities of the Imperial 

 University directed the writer to visit the places which had been 

 affected by the shock, and to make a full report of all the cir- 

 cum=;tances. The results thus arrived at form the subject of the 

 present paper. Before proceeding with this, however, it seems 

 desirable to give some particulars respecting the principal shocks 

 which have been felt in the Empire since 1879. 



The earthquake of February 22, 1880, is the severest that 

 has been experienced in the Plain of Musashi during the last 

 ten years. The damage done to buildings was very much 

 greater than on the recent occasion. Its origin was in the Bay 

 of Tokyo. 



On October 25. 1881, Nemuro, in Yez^, was visited by a 

 somewhat destractive shock. Fissures were opened in the 

 ground, and the damage to property was not inconsiderable. 



The well-known Atami Spa and its neighbourhood were 

 convulsed on the morning of September 29, 1882, by a 

 sudden and severe movement, which damaged embankments, 

 destroyed an historical monument, and did sundry other 

 mischiefs. 



The earthquake of October 15, 1884, originated in the Bay 

 of Tokyo, and affected the Plain of Musashi. It overturned a 

 considerable number of chimneys, cracked walls, and broke 

 articles in museums and elsewhere. In Tokyd, the greatest 

 horizontal movement, in a soft ground, was 42 mm., or double 

 the amount observed on January 15 last. However, the 

 total damage, taking the whole affected area mto account, was 

 smaller. 



The seismic waves in the disturbance of October 30, 1885, 

 extended over the whole of Northern Japan and part of Yezo, 

 shaking a land area of 34,738 square miles. But, though of 

 great extent, they fortunately did little harm. 



On July 23, 1886, quite a destructive earthquake visited 

 Shinano and the neighbouring provinces, overthrowing several 

 houses, and forming fissures in roads and hill-sides. The 

 shock also stopped the flow of a hot spring at Nozawa. The 

 l>art most severely shaken was a mountainous district some 

 2000 feet above the sea, including the famous active volcano of 

 Asama, and many extinct craters. This case was an unusual 

 one, as most of the larger earthquakes in Japan extend along 

 the sea-shore. 



Next in the list comes the severe shock of last January. 



It thus appears that this Empire is visited by a more or 

 less destructive earthquake almost once a year, and that the 

 Plain of Musashi is affected in like manner at intei-vals of a 

 fexv years. 



The shock of last January' was of most unusual violence. It 

 originated near the coast, about 35 miles south-west of Toky5, 

 and the seismic waves propagated nearly 200 miles to the west 

 and north-east al mg the Pacific seaboard. On the north-west 

 they approached but do not quite reach the shore of the 

 Japan Sea. They shook, in all, about 32,000 square miles 

 of land area. 



In Tokyo the disturbance began at 6h. 5110. 59s. p.m., with 

 slight tremors. After thirty seconds from the commencement, 

 the greatest horizontal motion (21 mm.) was recorded. The 

 time taken to complete one to-and-fro motion of the ground was 

 2"S seconds. The maximum vertical motion was only i*8 mm., 

 being, as usual, very small compared with the horizontal move- 

 ment. The principal motion continued for more than two 



' Paper by S. Sekiya, Professor of ScLsraalogy, Imperial University. Japan. 

 Reprinted fro.n the J_ournal of the College of Science. Imperial University, 

 Japan, vol. i. part lii. _ The earthiuake, the distribution and effects of 

 which are described in this paper, is the shock which formed the subject of an 

 article in Nature for June 2 (p. 107), in which one of the autographic 

 records obtained by the author with Prof. Ewing's seismographs w.as repro- 

 duced ill fiusiiitile. The diagram i;i question, which showed a gr-ateit 

 horiz ntal movement of 7J millimetres, was one of those referred to near the 

 end of this paper as having been obtained on the stiff elevated soil where 

 the University is built, and where the amplitude of the m ition was little 

 more than one-third of the m >tion shown by seismographs of the same 

 construction on the lower alluvial soil. 



minutes, during which time no less than sixty distinct shocks 

 occurred. The maximum velocity and maximum acceleration, 

 which measure the overthrowing and shattering power of earth- 

 quakes, have been calculated from the above numbers, and 

 found to be resj^ectively 26 mm. and 66 mm. per second. These 

 numbers, considering the range of motion, are small ; or, in other 

 words, the oscillations of the ground were comparatively gentle 

 and slow, which serves to explain the fact that but little harm 

 was done to property in the capital. In Yokohama, Hipp's 

 seismograph registered a horizontal motion of 35 mm. 



The origin of the shock was in a narrow band of country 

 running from west to east in the province of Sagami, parallel 

 to the coast, at a distance from it of about seven miles. It 

 emanates from the western or mountainous parts of the pro- 

 vince, passes through the southern foot of Oyama (4125 feet 

 above the sea-level), and reaches the Ray of Yokohama in a 

 total distance of about 30 miles. I believe the most pro- 

 bable cause of the shock to have been faulting or dislocation of 

 the earth's crust along the band ab^ve named. This inference 

 is supported by the fact that the parts of the country through 

 which the western half of the band parses consists of rocks of 

 different geological formations, interwoven in such a way that 

 their junctions present lines of weakness favourable to earth- 

 snaps. The topographical features of the district — high moun- 

 tains on the north, and comparative low plateau and sea-shore on 

 the south — also lend strength to this conclusion. Unequal dis- 

 tribution of loads on the earth's surface tends to facilitate bending 

 and folding of the rocks. 



It is along the above-named axis or band that the effects were 

 most striking. They were mainly confined, however, to a 

 small breadth on either side of it, so that places as little as 

 two or three miles to the north or south experienced a well- 

 marked diminution of seismic energy. This is not the first 

 instance in the history of the severer shocks in which the de- 

 structive effects have been practically limited to a small area 

 near the origin. 



More especially on the hilly or western portion of the origin, 

 land-slips and cracks were numerous. The cracks mostly took 

 place in banks, hill-sides, or other situations favourable for their 

 formation. The writer counted no fewer than seventy-two in a 

 distance of seven miles, the largest measuring a foot wide and 

 five hundred feet long, and all of them running parallel to the 

 axis of origin, which is also parallel to the general contour of 

 the country. Several wells became turbid. In some of artesian 

 character the water permanently decreased ; in others it in- 

 creased. There is a ferry across the large river Banyu where 

 it is crossed by the axial band ; but the water was so agitated by 

 the shock that for some time afterwards the boat could not be 

 used. The water in one of the rivulets on the west became 

 muddy. The shock was severely felt on board of vessels in 

 Yokohama harbour, the people in many of them rushing on 

 deck under the impression that they had been nm into. The 

 effects upon the.se vessels were doubtless cau-ed partly by motion 

 communicated through the cables, and partly by agitation of the 

 water due to movements of the sea-bottom. The earthquake 

 was preceded by the usual warning roar or ruml)ling, as of 

 distant cannon, emanating apparently from the western part of 

 the origin-band. In that district, too, the after-shocks on the 

 same night were five in number, while in Tokyo there was only 

 one. There were four tremors near the origin during the night 

 of the i6th. 



Dwelling-houses in country towns and villages are always 

 built of wood. Their frame-work is of timbers from four to 

 seven inches square, cros'-ing one another at right angles. The 

 uprights are placed about three feet apart, and stand on rows of 

 squared stones or boulders, the intervening spaces being filled 

 with bamboo-laths, on which is laid the mud-plaster that forms 

 the walls. Tiles and straw are principally used for the roof- 

 covering. In the district near the origin these wooden houses 

 sh lok with great violence. Several of them were more or less 

 twisted, cracked, or unroofed. Sliding doors, covered with 

 paper or of wood, which serve as shutters, partitions, and win- 

 dows in Japanese houses, broke and were shot out of their 

 grooves. The joints between the frames were in some cases 

 badly loosened. Although there are thousands of wrecked 

 houses, in the district of origin, on the verge of falling down, 

 and looking as if a strong breeze would be enough to blow them 

 over, the buildings of this class nevertheless withstood the vio- 

 lence of the earth movements so far as to escape actual demoli- 

 tion. The writer saw only two small rotten hovels which had 



