REPORT OF MR. ROBERT B. LINES.
461
upon the line, setting the clock-work in motion and bringing the semaphore to arrest, which is the position of the diagram. The button a returns, of course, to its normal position, there is no current upon the line after the signal is given, and the " control ” signals remain quiet.
Now, if an atmospheric current comes upon the line, it discharges itself in the earth, influencing, of course, both the "control” signals at the station and the magnet/’of the semaphore (Plate VIII., Fig. B), both of which, however, are protected from damage by "lightning-arresters.” The magnet /turns the signal up to " line clear.” The object now would be to advise the station-master, who may not notice the momentary signals given by the lightning upon his " control,” of the
dangerous position of the semaphore, as he is advised, under tko present three-wire system, of its proper position at arrest; # namely, by sounding continuously his " control-signal.” The same amount of attention which he now gives to the " control ” to assure himself that the semaphore is in the proper position, Would, of course, suffice if the "control” were used to warn him of danger.
Suppose the commutator of the semaphore, instead of, as at present, closing an extra " control ” circuit when the arm ls turned to arrest, were, when the arm is turned to " line clear ” to break the direct connection of the magnet f with the ground and to bring in circuit an extra battery, y 1 , at the same time reversing the direction in ivhich its current should pass around one of the coils of the magnet / and throwing the current through that coil, by way of a resistance equal to that