The Girdle of the Deep

The First Telegraph Wire Across the Atlantic Floor

On August 16, 1858, clerks in the wooden cable house at Knightstown on Valentia Island, Ireland, began tapping out a ninety-eight-word congratulatory telegram from Queen Victoria to President James Buchanan in Washington. Across 1,660 nautical miles of the North Atlantic seabed, operators at Bay of Bulls in Trinity Bay, Newfoundland, watched their receiving instruments for sixteen hours and thirty minutes before the final word arrived—a transmission rate of roughly ten minutes per word. Even more baffling to nineteenth-century telegraphers used to overhead landlines, where crisp battery clicks traveled hundreds of miles nearly instantaneously, every electrical pulse sent into the Atlantic cable emerged on the far shore as a long, sluggish swell of current that took seconds to rise and seconds more to drain away.

The culprit was the physical geometry of an insulated wire submerged in conductive salt water. To survive four kilometers down on the abyssal Telegraph Plateau under four hundred atmospheres of hydrostatic pressure, the 1858 Atlantic cable used a core of seven twisted copper wires weighing 107 pounds per nautical mile, coated in three layers of gutta-percha—the natural thermoplastic latex sap of the Malaysian Palaquium gutta tree—wrapped in tarred hemp and armored with eighteen strands of seven-wire iron rope. On an overhead pole line, air separates the copper wire from the ground by twenty feet, keeping electrical capacitance negligible. On the ocean floor, less than four millimeters of gutta-percha separated the central copper conductor from the surrounding Atlantic Ocean, which acted as an infinite grounded outer electrode. The entire 3,000-kilometer cable behaved as a single cylindrical Leyden jar—a distributed resistor-capacitor ($RC$) network.

Three years before the ships HMS Agamemnon and USS Niagara spliced the cable in mid-ocean, the Glasgow physicist William Thomson (later Lord Kelvin) derived the differential equation governing electricity inside a submarine line. Because series copper resistance $R$ per unit length and shunt gutta-percha capacitance $C$ per unit length dominated magnetic inductance at low signaling rates, every voltage pulse had to charge the dielectric wall locally before current could advance down the wire. Voltage $V(x,t)$ along the cable obeyed the exact partial differential equation Joseph Fourier had formulated for heat conduction through a solid iron bar, yielding Thomson's "law of squares" for the characteristic signal arrival and decay time $\tau$:

Because total resistance scales linearly with length ($R_{\text{tot}} = R L$) and total capacitance also scales linearly with length ($C_{\text{tot}} = C L$), doubling the length $L$ of an undersea cable multiplied signal distortion and delay fourfold. A ten-times-longer cable smeared pulses out one hundred times wider, causing consecutive Morse dots and dashes to bleed into an unreadable continuous smear known as inter-symbol interference.

The Atlantic Telegraph Company's chief electrician, a Brighton surgeon named Wildman Whitehouse, publicly rejected Thomson's law of squares as academic theory. Convinced that pushing signals across two thousand miles of ocean simply required brute electromagnetic force, Whitehouse refused to rely on low-voltage battery signaling. Instead, he secretly wired a pair of five-foot-long induction coils energized by a bank of high-current Grove nitric-acid cells into the Valentia Island transmitter, hammering the Irish end of the cable with alternating spikes approaching 2,000 volts in an effort to drive heavy electromechanical print relays in Newfoundland.

Under four kilometers of near-freezing water, the 2,000-volt discharges arced across microscopic air bubbles and metallic impurities embedded in the thin gutta-percha insulation. Each high-voltage spark carbonized a tiny pinhole in the latex sheath, allowing seawater to contact the copper core and leach current directly into the seabed. As the insulation fault worsened through late August 1858, Whitehouse ordered operators to apply even higher voltages. On September 1, 1858, after carrying just 732 messages over twenty-seven days, the insulation failed completely and the cable went dead.

When a reorganized syndicate chartered Isambard Kingdom Brunel's 692-foot iron steamship Great Eastern to lay a new cable from Valentia Island to Heart's Content, Newfoundland, in July 1866, Thomson sat on the board and dictated the physical specifications. The 1866 core used high-purity copper weighing 300 pounds per nautical mile and 400 pounds of gutta-percha, slashing unit resistance $R$ and capacitance $C$. In the receiving stations, Whitehouse's 2,000-volt coils were banned in favor of low-voltage Daniell batteries operating below fifty volts, paired with Thomson's mirror galvanometer—a tiny magnetized steel needle glued to the back of a two-grain silvered glass mirror suspended on a single silk fiber inside a copper coil. Nanoampere currents arriving from across the ocean twisted the mirror by fractions of a degree, sweeping a reflected beam of paraffin-lamp light across a graduated paper scale so clerks could read eight words per minute without stressing the insulation. Two weeks after landing the 1866 shore end at Heart's Content, the Great Eastern steamed back to mid-ocean, grappled the broken 1865 cable from a depth of 3,700 meters on its thirtieth drag, spliced on fresh line, and gave the Atlantic two working telegraph channels.