78 METHDOS AND RESULTS OF TORONTO OBSERVATIONS. 



enclosed in a box and suspended by a thread of unspun silk. The 

 magnet carries a mirror which reflects a finely divided scale fixed 

 some distance off, the scale being read by means of a telescope 

 which is securely fastened to a stone pillar. In this way small 

 changes in azamuth of four or five seconds of arc can be immediately 

 detected. 



The changes in the horizontal component ot the force are meas- 

 ured by an instrument invented by M. Gauss many years ago. In 

 this instrument the magnet is suspended by two threads separated 

 by an arbitrary interval, the circle to which the upper ends of the 

 threads are attached is then turned until the torsion of the threads 

 compels the magnet to take up a position as nearly as possible at 

 right angles to the magnetic meridian, any increase of force will then 

 pull the marked end of the magnet towards the north ; whilst if the 

 force decrease, the torsion of the threads pulls the marked ends 

 southwards again. As in the declinometer the needle carries a mir- 

 ror which reflets the fixed scale by means of which the amount of 

 change is measured. 



Changes in the vertical component of the force are measured by 

 a magnet suspended by means of knife edges on agate planes and 

 therefore only free to move in the vertical plane, this needle is 

 mechanically balanced so that at the normal force the magnet shall 

 be as nearly as possible horizontal, an increase of force will cause the 

 north end of the magnet to dip, the angle through which it moves 

 being measured by means of micrometers. 



The photographic instruments are placed in an underground 

 chamber with a view of exposing them as little as possible to change 

 of temperature. As in the instrument used for direct reading, each 

 instrument has attached to it a small mirror and immediately below 

 a fixed mirror is attached to the slate bed on which the instruments 

 are placed. The light from a gas jet passes through a slit into a 

 collimating tube, the image of the slit passes then through a lense 

 and is thrown on to the two mirrors and by them reflected through 

 a semi-cylindrical lense which focuses the light into two bright 

 points which are projected on cylinders carrying sensitised paper and 

 fed forward by clock work. The spot from the fixed mirror gives a 

 straight or base line, and that from the mirror attached to the 

 magnet exhibiting the direction and amount of the movements of 



