EARTHQUAKE PHENOMENA. 415 



If a common barometer be moved a few inches up and down by 

 the hand the column of mercury will be found to oscillate up and 

 down in the tube in directions opposite to the motions of the instru- 

 ment, the range of the mercury depending upon the velocity and 

 range of motion of the whole instrument. A barometer fixed to the 

 earth, therefore, if we could unceasingly watch it, would give the 

 means of measuring the vertical element of the shock-wave: and if 

 we could lay it down horizontally, it would do the same for the 

 amplitude or horizontal clement. This we cannot do; but the same 

 principle may be put into use by having a few pounds of mercury, 

 and some glass tubes bent into the form of JL, sealed close at one 

 end, and open at the other; the bore being under two-tenths of an 

 inch in diameter, and each limb about fifteen inches long. We shall 

 also require some common barometer tubes of the same calibre: the 

 open end being turned up like an inverted syphon, and equal in bore 

 to the rest of the tube. — (See Fig. 5.) The ^ tubes are used for 

 the horizontal, the others for the vertical elements. 



To fit the | tubes for use fill each partly with mercury, and so 



""■ " Sealed end. 



adjust it that a column of five inches in length shall be 

 in each limb of each tube, when held as in Fig. 2; the 

 limb a b horizontal, and the vertical column being sup- Fig2 

 ported as in a barometer. Tie four of these tubes so 



prepared together, back to back, so that if one hori- 



zontal limb face the north, the others shall face east, 



south, and west, respectively, as in Fig. 3. "In this position secure 

 them all down upon a broad, stout board, 

 that can be itself fixed to a surface of 

 rock, or other fixed surface of the earth. 

 An index or marker must now be pre- 

 pared for each tube; for one of these, cut' 

 a common piece of card two inches long wo 

 by rather less than two-tenths of an inch 

 wide, nick it partly through along a cen- «;< 

 tre longitudinal line, and double it down the long way, so that the 

 two segments shall stand at rather less than right angles to each other; 

 cut a cylindrical slice of cork one-eighth of an inch thick, of a di- 

 ameter such that it will go easily into the tubes; attach the bit of 

 cork with glue or sealing-wax to the end of one wing or segment of 

 the folded card, leaving the other free, and thrust the whole into the 

 horizontal limb of the tube until the cork just touches the mercury, 

 and so for the others. This marker is shown at rather more than full 

 size in 



Cork. Free side. 



Attached side. 



The edges of the card, having a certain amount of elastic extension, 

 must slightly grip the inside of the tube. 



It will now be found, if horizontal motion be given to the system 



