PHYSICS. 213 



long. This is placed in the cavity in the sinker with the 

 bend upward, and of course the closed end downward. The 

 whole is supported by a swivel-link at top. When immersed, 

 the air in the tube will be compressed, and at a depth of five 

 and a half fathoms the water will rise in the open leo* to the 

 bend ; any further descent will cause it to flow over into 

 the closed leg. By calibrating and graduating this le^ the 

 amount of water which it contains when raised will, of course 

 indicate at once the depth to which it has descended. If 

 desired for use in water shallower than five and a half fath- 

 oms, the open leg may be made shorter than the closed one. 



MECHANICS. 

 1. Of Solids. 



In Mechanics Abbot has repeated and extended his exper- 

 iments to determine the velocity of transmission of earth 

 waves caused by explosions, in order to settle certain ques- 

 tions raised by Mallet. He concludes : 1st. That a high 

 magnifying power of telescope is essential in seismometric 

 observations. 2d. The more violent the initial shock, the 

 higher is the velocity of transmission. 3d. This velocity 

 diminishes as the general wave advances. 4th. The move- 

 ments of the earth's crust are complex, consisting of many 

 short waves first increasing, and then decreasing in ampli- 

 tude ; and with a detonating explosive, the interval between 

 the first wave and the maximum wave at any station is 

 shorter than with a slow-burning explosive. The seismom- 

 eter used was a dish of mercury, whose surface was watched 

 through a telescope, by which the beginning and duration 

 of the tremor could be determined. The velocity observed 

 was for short distances 8500 feet, and for long 5300 feet per 

 second. 



Cornu and Bailie have continued their experiments on the 

 determination of the density of the earth by the method of 

 Cavendish, and have considerably improved their apparatus. 

 They have quadrupled the force to be measured by increas- 

 ing the attracting spheres of mercury from two to four, and 

 by diminishing the distance through which the attraction is 

 exerted in the relation of -\/2 to 1. With these improve- 

 ments, the results are completely uniform, so much so that 



