222 SCIENTIFIC RECORD FOR 1884. 



roughly north and south and extending radially in or under the northern 

 slope of Epomeo. {Am. Jour. Sci., xxvi, 475.) Palmieri attributes it to 

 the collapse of subterranean cavities, probably caused by the abstrac- 

 tion of matrer carried away in solution by the hot springs. Fuchs says 

 {Min. u. petr. Mitth., 1884, 185) it was neither volcanic nor a collapse 

 of a cavity, but only a rearrangement of the materials of the hill taking 

 place at a moderate depth. 



On August 10, 1884, between seven and eight minutes after 2 p. m., 

 a decided earthquake occurred in the Middle and Eastern States, a brief 

 preliminary notice of which was given by the writer. {Am. Jour. Sci., 

 XXVIII. 242.) It extended along the coast from Baltimore, Md., to Port- 

 laud, Me., and westward to the Alleghauies. The origin was evidently 

 not lar from I^ew York City. The shock was more violent than any 

 which had been felt in that region for a considerable time. 



Part second of the seventh volume (1884) of the Transactions of the 

 Seismological Society of Japan is occupied by John Milne in the dis- 

 cussion of 387 earthquakes observed in North Japan from October, 

 1881, to October, 1883. It is accompanied by 123 maps of the areas 

 affected by siugle earthquakes, and one general map shaded to repre- 

 sent the distribution of volcanic and seismic activity in Japan, and also 

 15 pages containing 66 figures of the tracings made by various record- 

 ing instruments. As results of the discussion the author finds that 84 

 per cent, of the earthquakes originated under the ocean or on the eastern 

 seaboard ; that the winter intensity is nearly three and a half times as 

 great as the summer intensity ; that there is a general coincidence be- 

 tween the maximum of earthquakes and the minimum of temperature; 

 and that there were 11.2 per cent, more earthquakes at low water than 

 at high water. Sixteen of the earthquakes occurred simultaneously in 

 separated areas, not being felt in the intervening districts. He also 

 finds that the indications of exactly similar instruments may vary con- 

 siderably at stations only a few hundred feet apart. Whence he concludes 

 that the amplitude and period of the vibration constituting an earth- 

 quake are very largely dependent on the character of the soil and other 

 local circumstances, and that therefore the direction of vibration in an 

 ordinary earthquake has usually no immediate relation to its direction 

 of propagation. 



Thomas H. Streets, M. D., U. S. IST., gives {Ain. Jour. Sci., xxv,361) 

 a list of earthquakes compiled from the weather statistics kept at the 

 United States naval hospital at Yokohama, Japan. It includes all 

 shocks, appreciable without instruments, that occurred in a i^eriod of 

 four years beginning with 1878. They numbered 124, and, classified by 

 seasons, were: Winter, 41; spring, 35; summer, 27; autumn, 21; show- 

 ing a greater frequency in winter than in summer, as has been given 

 by other lists. If March be substituted for December as a winter month, 

 the contrast is still more striking, as we then find 51, or 41 per cent, of 

 the whole number, occurring in those three months. February and 



