140 REPORT— 1897. 



results in the pendulum having a period of 15 seconds — that is to say, 

 it is reached when the pendulum makes one complete swing or one back 

 and forward motion in 15 seconds. To make this adjustment the pivot 

 against which the boom abuts may be moved in and out until the desired 

 period is approximately obtained, after which the front screw of the stand 

 may be raised or lowered until the adjustment is completed. To observe 

 the period the observer presses with his hand against the side of the 

 column. This sets the boom in motion. He then goes to the end of the 

 instrument, and, looking downwards through a plate of glass beneath the 

 lamp, watches the rectangular plate on the end of the boom and notes 

 with a watch how many seconds it takes for the boom, as it slowly moves 

 across the scale of millimetres fixed in the top of the clock-box parallel to 

 the slit in the same, to complete a back and forward motion. For various 

 reasons it seems that in all forms of horizontal pendulums this quantity 

 will not remain constant for any great length of time. It therefore must 

 be noted, say, once a week, and if any marked change has taken place the 

 instrument should be readjusted. For stations founded on rock the pendu- 

 lum may be adjusted to have a period of 18 seconds ; but with a pendulum 

 having this sensibility in a station on alluvium, the diurnal motion may 

 exceed the width of the slit in the clock-box, and with changes of weather 

 and the seasons the wandering of the pendulum to one side or the other 

 will be so great that readjustments will be continually required. 



The boom is to be brought into a central position by turning one or 

 other of the two back screws in the bed-plate. 



The Sensibility of the Instrument. — The distance between the two 

 back screws of the instrument is 150 mm. The front one of these has 

 0-5 mm. pitch, so that one complete turn of this would tilt the 

 stand through an angle the tangent of which would be measured by 

 3 X -r5~o = a^o- -^y nieans of a lever fitting the head of the screw, rather 

 than giving it a complete turn, it may be turned 1°, 2°, or any other 

 fraction of a complete turn that may be desired, this quantity being 

 indicated by a pointer attached to the screw which moves over an arc 

 gi-aduated in degrees. For example, assuming that the boom has a period 

 of 18 seconds, and we find by several trials that a 1° turn of the 

 test-screw corresponds to a deflection of the outer end of the boom of 

 5 mm., as shown on the scale opposite the slit in the clock-box, and 

 assuming, further, that we can read displacements on the photogram of 

 1 mm., under these circumstances we can measure tiltings the angular 

 values of which would be 



11 11 1 



X X „--r^ X - = - 



2 150 360 5 540000' 



and because 1 sec. of arc=l/206265, it follows that 1 mm. deflection of 

 the outer end of the boom corresponds to a tilt of 0"-38. 



If we read deflections to within half a millimetre, to do which there is 

 no difficulty, the sensibility of the instrument is doubled. For the object 

 in view this is not required, and if a deflection of 1 mm. is obtained for a 

 tilt of 1" to 0''-5, this will he. sufficient. 



Clock-box. — This, which can be run on rails in and out of the instru- 

 ment-case, has a cover which is removed to wind the clock and put new 

 paper in the roll. Once a day, when the lamp is filled and trimmed, and 

 the watch is wound, this cover is removed, and the 3 or 4 feet of paper 



