SUBJfAKIXE TELEGRAPHY 229 



without withdrawing one drop of the fluid. When messages 

 were being received on board the ' Great Eastern/ they simply 

 caused the slight necessary oscillations in the marine galvano- 

 meter (an invention of Sir William Thomson's, dating June 

 1857), which were insufficient to disturb the insulation test. 

 When the signals were being sent from the ' Great Eastern/ the 

 rush of current in and out of the cable would have disturbed 

 the galvanometer unduly, so it was shunted; that is to say, 

 a part of the current was derived by a little sliding arrangement 

 at the end of each word the slide was moved and a perfect 

 insulation test made. These various practical improvements 

 can only be understood by professional men, but the leading 

 idea of Mr. Willoughby Smith's plan may be grasped by all. 

 The arrangements worked as well in practice as they were 

 admirable in theory. Fortunately no fault occurred. 



When a fault does occur, stopping the cable is a very trying 

 and hazardous proceeding. It can only be done gradually. 

 The ship is perhaps running at six miles per hour, or a mile in 

 ten minutes. She will not lose her impetus for a considerable 

 time, even if the engines are reversed ; and when the ship is 

 stopped, the cable cannot be instantly checked if it were, the 

 strain would rapidly become far too great for it to bear. The 

 twelve or fifteen miles which lay straight on the inclined water- 

 plane, as before described, would quickly fall into the common 

 catenary curve, of which the whole weight would have to be 

 borne ultimately by the cable at the ship's stern ; for when the 

 cable ceases to sink the resistance of the water ceases to buoy 

 it up. The strain caused by a flat catenary of this length is 

 enormous ; thus, if in a depth of two miles only ten horizontal 

 miles intervened between the ship's stern and the point where 

 the cable lay on the ground, the strain due to the catenary 

 would, with the Atlantic Cable, be fourteen tons. In practice, 

 therefore, the cable is generally restrained by such a force as is 

 thought safe, and then allowed to run out until it lies in a 

 catenary short enough to produce only this small strain, or if 

 the cable must be held, the ship must go astern over the cable. 

 When the foul flakes occurred during the 1866 Atlantic ex- 

 pedition, the ' Great Eastern 7 was stopped in two minutes after 

 the signal was given, and only 130 fathoms of cable paid out 



