46 " NINTH REPORT. 



sound like near thunder preceded the shock b}- a second or two of time, and 

 increased in intensity so that the falling of heavy masonry buildings a feAV 

 rods awa}'' was not audible. Unable to keep his feet he sat down upon the 

 ground, and not only felt but distinctly saw the ground thrown into violent 

 waves, "as though composed of soft jelly.'' These waves appeared to ad- 

 vance atong the ground, and induced in him a feeling of nausea akin to sea- 

 sickness. When the shocks had passed all masonry structures had been 

 leveled to the ground, and over each hung a cloud of i)ink plaster particles and 

 dust. Above the town in the park a horseman noticed that a peculiar 

 rustling of the leaves upon the trees preceded the first sounds by a brief in- 

 terval of time as though resulting from an earlier treinor. 



By many the shocks were described as in places gyratory or twisting in 

 their nature, and monuments which were built up of sections revealed an 

 increasing amount of rotation for those higher blocks in the structure which 

 had not been completely detached. 



• Over large areas the surface of the ground was rent by numerous fissures, 

 large or small, and some of these had great extent. It was noticed that these 

 fissures followed in their direction the lines of the ranges of hills. Sometimes 

 they gave the impression of having opened and later closed under great 

 pressure, as the ground was raised in a furrow. If sandy, the ground ap- 

 appeared as though a steam plow had passed over it, tearing up the surface 

 and throwing heavy clods in every direction. Posts were sunk deeper into 

 the ground and were surrounded by a cup-shaped depression. In many 

 instances monuments and even houses were similarly projected into the 

 sandy ground so that only the tops and roofs remained visible. 



In addition to the numerous cracks, crater-like pits appeared in the ground. 

 These were usually about six feet across, though sometimes more, and through 

 them jets of sand and water were thrown to a height of seven or eight feet, 

 and probably much higher. Mixed with the sand were fragments of peat, 

 coal, resin, half-petrified pieces of timber, and a black earth at the time un- 

 known in the district. The same materials also welled up through some of 

 the fissures. The large amount of sand thus brought to the surface was spread 

 aroimcl the orifices in flat domes, and where these were most numerous the 

 entire face of the country was flooded, and after an interval blanketed with 

 a layer of quicksand in which cattle floundered and were held fast. The local 

 streams were swollen suddenly and raised from two to ten feet, though they 

 settled back to their former levels shortly after. The Brahmaputra ad- 

 vanced as a wall of water ten feet in height. 



In the Garo and Khasi hills the numerous land slips within the weathered 

 sandstone rocks developed wide spread fans of sand at their bases. The 

 rivers of this section are ordinarily a series of deep pools separated by rocky 

 rapids, though in flood-time they changed to raging torrents. Following 

 the rains after the earthquake of 1897, the pools were found to be filled up 

 with sand, the rapids ol)literated, and the streams flowed over the sandy 

 floor of a broad and shallow channel. In this hill country were found the 

 most interesting of the geological changes. Although only a single zig-zag 

 journey was made through the country, three large earthquake faults, hun- 

 dreds of great fissures and no less than thirty lakes were found to have re- 

 sulted from the earthquake. One of these lakes was more than a mile 

 across. 



The largest of the earthciuake faults, known as the Chedrang fault, adhered 

 to the course of a meandering but otherwise straight river, and was thus 

 followed for a distance of twelve miles. The vertical displacement or throw 



