J. E. Burbank — One Phase of Microseismic Motion. 473 



with one marked exception : on October 16-17 a hurricane, 

 with pressure ahout 29*05 inches, was in the Gulf southwest 

 of Florida and the microseisms had a period from 5*0 to 5*8 

 seconds ; on the 18th, when the center had approached the 

 Florida coast and was passing inland, the period had decreased 

 to 3*5 seconds and the amplitude greatly increased. 



In general the period of the microseisms is from 3*0 to 3*5 

 seconds regardless of the part of the coast under strain. Peri- 

 ods greater than 3*6 seconds apparently occur only when the 

 low is of great extent and the center almost wholly over the 

 ocean. It would appear from this that the period of the micro- 

 seisms varies with the extent of the disturbed water area. 



In general, pressure changes due to high areas are too grad- 

 ual and widespread to produce microseisms of appreciable 

 intensity, although about 40 cases have been noted, nearly all 

 being cases in which a depression was closely followed by a high 

 area of marked intensity. 



In my earlier paper it was suggested that the microseisms 

 might be connected with the movements of large masses of 

 water set in motion by the wind accompanying the pressure 

 changes. This assumption is not borne out by a comparison 

 of the winds, normal to the coast line, and the microseisms 

 occurring during the period January 1 to June 30, 1910. 

 During this period there were strong microseisms on days when 

 there was little or no wind along the coast, and also days when 

 there were high winds without any well-marked microseisms. 

 In general, high areas are accompanied by winds when they 

 approach the coast, although they are rarely accompanied by 

 microseisms. 



Another point of interest is that the period of the micro- 

 seisms does not appear to be conditioned by the geological nature 

 of the part of the coast line over which the low is passing, as 

 all parts of the coast give essentially the same periods. It 

 seems probable that this period is a characteristic of the local- 

 ity in which the seismograph is mounted, although the change 

 of period during different microseisms is difficult to explain on 

 that basis. Klotz at Ottawa observed periods of 5 to 6 seconds 

 with occasional changes to 3 seconds. 



The above conclusions by no means preclude the probability 

 of microseisms being produced by the movement of lows and 

 highs wholly over the land area ; in fact it is extremely prob- 

 able that they do occur, and could be readily recorded by a 

 sufficiently sensitive seismograph, but it is evident that, at least 

 for the eastern part of the United States, the most marked 

 microseisms are those related to the variations of pressure 

 along the coast line. 



