24 



NATURE 



[March 3, 1910 



dimensions of core, the time taken by the current to reach 

 any particular fraction of the full value will vary as the 

 square of the length. 



This result of the theory was of fundamental import- 

 ance. It was also, at the time, of particular interest, for 

 the project was then beginning to be mooted of connecting 

 England and America by wire. The only experience avail- 

 able as to speed of signalling was on short cables, and 

 in passing from them to a line 2000 miles long the " law 

 of squares," as it was called, seemed at first to give little 

 prospect that signalling across the Atlantic could be accom- 

 plished at a speed that would be commercially practicable. 

 To lay the cable it was coiled on board two ships of war, 

 the British battleship Agamemnon and the United States 

 frigate Niagara. On August 5, 1857, the shore end was 

 landed at \'alencia, and the Niagara began to pay out, 

 the intention being that her section should be laid first and 

 the Agamemnon should continue the work after making 

 a splice in mid-ocean ; but the paying-out gear was very 

 crude ; the brake for maintaining a proper tension in the 

 cable was difficult to regulate, and after 300 miles were 

 laid there was a mishap at the brake and the cable parted 

 in 2000 fathoms. The ships returned to Devonport, the 

 cable was stored for the winter, new machinery was 

 designed, and some 700 miles of fresh cable were manu- 

 factured against the next attempt, to be made in the 

 following year. 



Thomson had joined the expedition at the request of his 

 brother directors, and was on board the Agamemnon. He 

 came back full of ideas as to both the electrical and the 

 mechanical sides of the great problem. On the mechanical 

 side he had worked out, for the first time, the theory of 

 the forces concerned in the laying and lifting of deep-sea 

 cables ; this was published almost immediately after his 

 return. Let me give }"ou a brief sketch of the results of 

 this theory. 



A cable paid out from a ship going at uniform speed 

 does not hang as a catenary, but takes the form, as it 

 sinks, of a straight line stretching at a uniform slope from 

 the ship's wake to the point far in the rear at which it 

 touches the bottom. This is because each part of the cable 

 in sinking through the water attains almost immediately a 

 constant velocity of descent against the resistance which 

 the water opposes to its motion. Imagine a ball, heavier 

 than water, to be dropped from a ship. It will, after 

 sinking a foot or two, attain a practically uniform velocity, 

 and keep that until it reaches the bottom. Imagine, now, 

 a ship to drop a series of such balls, at regular intervals, 

 while she steams ahead at a steady speed. At any instant 

 the depth through which each ball has sunk will be pro- 

 portional to the time which has passed since it was 

 dropped, and therefore to the distance run by the ship, 

 and hence a line joining the successive balls will be a 

 line of uniform slope. The continuous cable behaves in 

 this respect like the row of balls, but with this important 

 difference. Each ball sinks vertically ; it has no tendency 

 to do anything else ; but the cable tends, not only to sink, 

 but to glide along the direction of its own length, just as 

 a rope resting on an inclined plane tends to glide down it. 

 A certain amount of such gliding is desirable, indeed 

 necessary, for it secures that the cable will be laid with a 

 suflficient percentage of slack to accommodate itself to sihy 

 inequalities on the bottom, and to provide for the possi- 

 bility of its being raised should that be required. It is 

 the function of the paying-out brake to apply just so much 

 retarding force as will allow the right amount of this 

 gliding to take place, and not too much. As cables are 

 actually laid, there may be 10 or 12 per cent, of slack, 

 and this means a considerable velocity of gliding motion. 

 In a cable of the type which was afterwards successfully 

 laid across the Atlantic, the straight line had a slope of 

 about I in 8| — in other words, with a depth of two miles 

 there were seventeen miles from the ship to the place where 

 it touched bottom. On the gliding motion down this long 

 slope the frictional resistance of the water is an important 

 factor; it reduces ver}- much the retarding force needed 

 at the brake. If it were simply a question of holding 

 the cable from gliding down the slope at all, the retarding 

 force would be equal to the weight, in water, of a length 

 of cable equal to the depth. In fact, however, it is about 

 half that, the other half being accounted for by the fric- 

 NO. 2105, VOL. 83] 



tional resistance the cable experiences in gliding down the 

 slope. 



In the early summer of 1858 the cable squadron was 

 again ready to put to sea. New paying-out brakes had 

 been devised. Thomson had succeeded, with much diffi- 

 culty, in getting systematic tests of the conductivity estab- 

 lished during the manufacture of the additional 700 miles. 

 Most important of all, he had invented a new signalling 

 and testing instrument which was to make Atlantic tele- 

 graphy commercially practicable. This was the mirror 

 galvanometer, the first description of which is found in his 

 patent of 1858. 



We have no time to linger over the story of the cable 

 of 1858. This time the two ships, after encountering a 

 storm of great severitj', in which the coiled cable suffered 

 severe damage, met in mid ocean, spliced the cable, and 

 began to pay out simultaneously, the Agamemnon steam- 

 ing towards Ireland and the Niagara towards Newfound- 

 land. The cable broke when only six miles were paid out. 

 Again the ships met to make a fresh splice, and again the 

 cable failed when some eighty miles had run out. A third 

 attempt promised better, for some 200 miles were laid, 

 when again the cable broke, this time at a place where 

 it had been injured in the storm. The ships returned to 

 Queenstown ; Bright, Thomson, and the other leaders, dis- 

 appointed but not discouraged, advised the Board to order 

 a fresh attempt. Their advice was taken. The ships once 

 more met at the mid-ocean rendezvous, and this time 

 success crowned their efforts. On August 5 both ships 

 completed their task, and the ends of the cable were 

 brought to land. 



Scarcely had the enthusiasm awakened by this great 

 event begun to subside when it was apparent that all was 

 not well. The Irish end of the cable had been handed over 

 to Mr. Whitehouse, who attempted with little or no success 

 to establish communication by means of his own signalling 

 instruments. It was only when the galvanometer of 

 Thomson was resorted to, with a simple Daniell battery to 

 send the current, that messages were transmitted. The 

 Board, dissatisfied with Whitehouse's action, directed 

 Thomson to take complete charge. Various important 

 messages passed, but the tests showed that the insulation 

 of the cable had broken down ; a bad fault developed, 

 which had doubtless been intensified, if not produced, by 

 the high-tension induction coils used by Whitehouse. The 

 signals grew more and more feeble, and in a few weeks 

 the cable altogether ceased to speak. 



It never spoke again, and not until 1865 was the attempt 

 made to lay a new Atlantic cable. By that time much 

 had been accomplished. It was in the intervening years 

 that the work of establishing standards for electrical 

 measurement was undertaken by a committee of the British 

 Association. The committee was appointed at the instance 

 of Thomson, and he took a prominent part in its work. 

 Besides • this, the cable engineers were busy, and were 

 gaining experience from lines laid in other places. Methods 

 of systematic testing were devised ; a type of cable was 

 designed which was better adapted than before to bear 

 the strain of laying, and especially the much severer strain 

 of picking up, and material improvements were made in 

 the paying-out machinery. 



In i86.; the Great Eastern was available for laying the 

 cable. Thomson, along with Cromwell Varley, went as a 

 consulting expert on behalf of the company. Twelve 

 hundred miles were successfullv laid, and then a fault 

 showed itself ; picking up was begun, but in manoeuvring 

 the ship the cable parted in deep water. Attempts were 

 made to recover it by grappling ; three times it was hooked 

 and brought part of the way to the surface, but the shackles 

 used to couple up successive lengths of the grappling rope 

 were too weak to stand the strain. Grapnel, rope, and 

 cable were lost, and the ship returned with the task un- 

 finished, but with everyone now full of confidence, not only 

 that a sound cable could be laid, but that the lost cable 

 could be found and lifted. 



In 1866 the thing was done ; an entirely new cable was 

 laid with complete success, and then the Great Eastern 

 with her consorts proceeded to the lost end of the cable 

 of 1865, and began once more to fish in water more than 

 2000 fathoms deep. A fortnight passed before the watchers 

 at Valencia saw any sign ; then the spot of light began 



