22 Mr. Sidney G. Brown [Jan. 30, 



This lag in the instrument renders it difficult to hold a ship 

 dead on her course, and the path, as a consequence, is sinuous, and 

 may oscillate, even in a calm sea, as much as V each side of the 

 correct heading. 



As a ship has usually to steam entirely by the readings of the 

 compass any error is serious. For instance, if there is an error of 

 :; . and the ship is steaming at 16 knots, she will move one English 

 mile off her course every hour. It is obvious how necessary it is to 

 have absolutely correct readings. Lord Kelvin was the first to 

 seriously study the errors of the magnetic compass. He started in 

 1871, and in 1876 produced his well-known instrument. Although 

 it was a great advance on any compass in the British Navy, he had 

 the greatest difficulty to get them to adopt it ; finally, in 1879, he 

 proposed to place an instrument at the disposal of the Admiralty 

 at his own expense. This offer was accepted. In spite of this, it 

 was only through the acquaintance of influential Naval officers, 

 particularly of Captain (now Lord) Fisher, that the compass was ever 

 adopted. 



In 1889, eighteen years after the commencement of his experi- 

 ments, and long after it was in common use in commercial ships, he 

 received official notification that his 10-inch compass was to be 

 adopted in future as the standard of the Navy. It is fortunate that 

 we have an alternative method of securing a north-seeking property 

 in the Gyro-Compass, an instrument of much greater accuracy than 

 the magnetic, and with none of these errors, for if deviations do 

 occur they are known deviations, and can therefore be allowed for. 



Evans and Smith, in 1861, were the first to discover how impor- 

 tant it was to mount the needles on the card so that the moments 

 of inertia of the moving system should be the same about all 

 directions — that is to say, that the system should be in dynamic 

 balance — otherwise the rolling of the ship would cause deviations in 

 the reading. I have lately discovered that another deviation may 

 be brought about, not by an oscillation in one direction, but by the 

 card being set wobbling ; the needles and card would then have a 

 force applied trying to carry the moving system round in the direc- 

 tion of tbe wobble. 



I have a magnetic compass here to demonstrate this. It consists 

 of a heavy brass disc mounted on a vertical frictionless spindle. The 

 needles are fixed to the disc, and the whole movable system is carried 

 on a pendulous mounting, as in the Gyro-Compass. The disc and 

 needles are in correct static and dynamic balance. 



Swinging the pendulum in any one direction produces no devia- 

 tion, but by causing it to swing in a circular conical path, thus giving 

 a wobble to the plate, we have a serious deviation in the reading of 

 the compass ; the error is permanently maintained against the earth's 

 attract ion so long as the circular motion of the pendulum persists. 



I have also carried the compass round in a horizontal circular 



