Radio — Marconi Transmits Wirelessly

In plain terms

A spark cracks between two electrodes, an electrical circuit starts to vibrate, and that vibration travels off through the air: that is the whole principle of Marconi's radio. In 1895, on the family estate at Villa Griffone, the signal covered only 2.4 km. Six years later, on December 12, 1901, Marconi announced that he had heard, in Newfoundland, three Morse dots — the letter "S" — sent from Cornwall 3,500 km away, at a time when the physicists of the day, Lord Kelvin foremost among them, judged it impossible to exceed ~300 km because of the curvature of the Earth. What saved him was something he did not know about: an electrically charged layer of the upper atmosphere sends waves of this kind back down to the ground, and the signal bounced between that layer and the surface of the ocean. An honest reservation: no physical record of that listening session was preserved, and the reception of December 12, 1901 remains debated by historians.

Discovery

Technical explanation

1. Wave generation — Marconi's transmitter uses a spark-gap oscillating circuit: a high-voltage discharge (15–20 kV) between two electrodes ionizes the air and generates damped electrical oscillations in an LC circuit (inductance + capacitance). The oscillation frequency is determined by $f = \frac{1}{2\pi\sqrt{LC}}$f=2πLC​1​. These oscillations are coupled to an antenna that radiates electromagnetic waves.

2. Transatlantic propagation — Skeptics (including Lord Kelvin) maintained that radio waves, propagating in straight lines, could not cross the Earth's curvature beyond ~300 km. Marconi circumvented this obstacle without understanding it: the ionospheric layer (identified by Kennelly and Heaviside in 1902) reflects waves at frequencies below ~30 MHz. The Poldhu signal (~850 kHz, MF band) bounced between the ionosphere — principally the E layer (~90–150 km altitude), which carries nighttime MF propagation — and the surface of the ocean, crossing the Atlantic by multi-hop ionospheric propagation.

3. Reception — At Signal Hill, Marconi was not using Branly's filings coherer (a glass tube containing nickel-silver filings whose resistance drops from ~100 kΩ to ~1 kΩ in the presence of radio waves, reset after each pulse by a mechanical hammer — the tapper), but a mercury coherer known as the "Italian Navy" detector: a self-restoring device, listened to directly on a telephone earpiece, with no tapper and no paper-tape recording. The absence of any physical trace of the signal explains why the reception of December 12, 1901 remains historically debated.

4. Modulation and bandwidth — Marconi's system used on-off keying (OOK): the signal was either present (Morse dash or dot) or absent. The spark-gap transmitter produced a wideband spectrum (spark gap = noise across the whole MF band), limiting the number of simultaneous transmissions. This problem would be solved by Lee De Forest's vacuum-tube oscillator (Audion, 1906), which made continuous modulation possible.

Why it worked

Marconi was not a theoretician — he did not truly understand ionospheric propagation. His strength lay in empirical engineering: he systematically tested the effect of antenna height, grounding and power on transmission range. He discovered empirically that range increases with the square of antenna height (an approximate relationship for diffraction beyond the horizon), which led him to build ever taller antennas.

Maxwell's theoretical work (1865) and Hertz's experimental demonstration (1887) had proven the existence of electromagnetic waves, but no one had perceived their practical usefulness for communication. Hertz himself, when asked, declared: "I do not think that the oscillations I have discovered will have any practical application."

The limits of the apparatus were recognized from the outset. The spark-gap transmitter radiates across the entire width of the MF band, which caps the number of simultaneous links until the arrival of continuous modulation (Audion, 1906). And the empirical rule Marconi followed — range grows as the square of antenna height — holds only as an approximation of diffraction beyond the horizon: it does not describe the ionospheric path that actually carried the signal, a mechanism whose reflecting layer would not be identified until 1902. There remains the gap between the announcement and what is materially demonstrable: the listening session of December 12, 1901 left neither paper tape nor recording, the mercury coherer being listened to directly on a telephone earpiece; it is this absence of a trace, and not any challenge to the principle, that keeps the event in dispute.

Causal chain

Maxwell formalizes the equations of electromagnetism (1865) → Hertz proves the existence of radio waves (1887) → Branly invents the coherer (1890) → Marconi transmits wirelessly (1895) → transatlantic transmission (1901) → De Forest's Audion (1906) → commercial broadcasting (KDKA, 1920) → Armstrong's FM (1933) → radar (1935) → television (1936) → communication satellites (Telstar, 1962) → mobile telephony (1G, 1983) → WiFi 802.11 (1997) → 5G NR (2019)

Anecdote

On the night of April 14, 1912, the Titanic sinks after striking an iceberg. Radio operators Jack Phillips and Harold Bride transmit the distress signal CQD, then SOS, on 500 kHz for 1 hour 40 minutes. The Carpathia, 93 km away, picks up the call and rescues 710 passengers. The Californian, only 16 km away, does not hear it: its sole radio operator has switched off his set and is asleep. This tragedy leads to the Radio Act of 1912 and to the requirement for a permanent radio watch at sea — radio moves from the status of technical curiosity to that of a legal safety obligation.

Legacy and current data

The chain opened at Villa Griffone has never been broken: from WiFi 802.11 (1997) to 5G NR (2019), today's links still rest on an oscillator coupled to an antenna. Wireless telecoms now form an ecosystem of 1.8 trillion dollars (2024). What remains to be established lies at the other end of the story: for want of a physical recording, the transatlantic reception of December 12, 1901 cannot be authenticated after the fact — the debate concerns that particular link, not the 1895 experiments nor the British patent of 1896.

Sources

References verified during the August 2026 fact-checking audit: these are the pages
against which this bulletin's claims were checked.

1. Le premier système d'antenne de Poldhu, décembre 1901 — Community Stories
2. Milestone: Transmission of Transatlantic Radio Signals, 1901 — IEEE, Engineering and Technology History Wiki
3. The 'Italian Navy Coherer' Scandal Revisited — IEEE