Editorial Note.  The following article is the text of a lecture given by the late Hugh Alexander Boyd M.A. M.Lit. M.Phil. M.A.(Ed) H.DipEd. (1907-1996) at St. Mary Star of the Sea Secondary School on the occasion of Ballycastle Civic Week in August 1968. We are grateful to Mrs Pearl Boyd for her permission to print the article, and Mr Danny J. McGill for sourcing the article. (The Glynns Vol. 31, 2003)

italian paper

(Cover of Sunday Newspaper La Domenica del Corriere 1903)

“It is amazing how the most outlying places can, at times, link up with the great names of the world. Rathlin Island, a hilly fragment of land only some five miles long and one and a quarter miles broad at the most, is mentioned both by the naturalist Pliny in the First Century A.D , and by Ptolemy, the celebrated Egyptian mathematician, astronomer and geographer, in the Second Century A.D. in his illustrated description of Europe. Just as remarkable as the island’s link with Pliny and Ptolemy, however, is it’s link with the early development of wireless telegraphy through the work of Guglielmo Marconi.

Born at Bologna on April 25th 1874, Marconi was the second son of Guiseppe Marconi, an Italian country gentleman and Annie Jameson, the daughter of Mr. Andrew Jameson, of Daphne Castle in County Wexford and Granddaughter of John Jameson, founder of the Jameson Whiskey distillery.

Genius and Perseverance

Marconi was not the discoverer of electric waves: he was not even the first to suggest that they might be used for signalling  to long distances, but thanks to his genius and perseverance, it is due to him more than to any other single worker, that they are now one of the most important means of communication. Wireless has conferred inestimable benefits not least on the mariner by enabling him to keep in touch with the rest of the world and by helping him to safety in his course across the oceans.  Broadcasting too, a direct outcome of Marconi‘s work, has now taken rank among the necessaries of life.

Marconi was educated at the University of Leghorn and at the University of his native city Bologna, generally regarded as the oldest university foundation in the world. At Bologna he studied under Adolfo Righi, himself the author of important investigations on electric waves.

Marconi1 (Marconi  ca. 1895)

His first attempts to turn Hertz’s laboratory work to practical use for the transmission of signals to a distance were made at his father’s villa at Pontecchio, near Bologna and at an early stage in his experiments he effected a fundamental improvement, by employing elevated conductors or aerials with both his transmitter and his receiver in combination with metallic connections to earth, which at once gave him results far ahead of those obtained by other workers in the same field. For the detection of the electric waves at the receiving end he used the “coherer” of Branly and Lodge, on the improvement of which he worked between 1894 and 1896.

His First Patent                  Portrait of Guglielmo Marconi

(Marconi ca. 1896)

It was in May, 1896, that Marconi, then a retiring, modest young man of 22, unknown and almost without friends, first came to England, and it was in this year that he took out his first patent. Upon his arrival in London he lost no time in presenting his credentials to Sir William Preece, the eminent electrician and then engineer-in-chief of the General Post Office. Marconi was fortunate  to enlist the interest of Sir William Preece, who gave the young scientist substantial assistance. It was also at this stage that Marconi came into contact with a very remarkable man in the person of George Stephen Kemp, who, on leaving the Royal Navy, became laboratory assistant to Sir William Preece.

marconi_george-kemp

(Marconi and George Kemp ca. 1897)

Kemp was detailed by Sir Preece to work with Marconi in the first demonstrations of wireless telegraphy given on these islands to officials of the Post Office and other Government Departments. Successful tests in the City of London between St. Martins-Le-Grand and the Thames Embankment were then followed by transmissions across Salisbury Plain, where signals were received at a distance of two or three miles from the transmitter. These transmissions were, in fact, among the first experiments in which Marconi and his assistant were engaged .

Floated Company

A further successful experiment on Salisbury Plain later in the year 1897 brought sufficient public confidence to enable Marconi to float the Marconi Wireless Telegraph Company. In demonstrations carried out for the Italian Government in the same year at Spezia, signals were sent for a distance of twelve miles.  In 1898 Marconi erected permanent stations in the Isle of Wight, near the Needles. It was from the Royal Needles Hotel that the first paid-for wireless message was sent.

The veteran scientist, Lord Kelvin, sometime Professor of Natural Philosophy the University of Glasgow and a native of Belfast – his statue may be seen at the entrance to the Botanic Gardens in that city – accompanied by Lady Kelvin and the Poet Laureate, Alfred, Lord Tennyson, went to the station at The Needles to see for himself the working of the new system of telegraphy. Lord Kelvin was keenly interested and asked the young scientist if he might be allowed to send telegrams to some of his friends. He insisted on paying one shilling for each message in token of his belief in the commercial possibilities of wireless telegraphy.

640px-Lord_Kelvin_photograph

Kelvin, Lord William Thomson (1824-1907)

Historic Occasion This historic occasion took place on 3rd June, 1898, and caused a good deal of favourable newspaper comment. Kemp thus recorded the event in his diary:

“June 3rd, 1898. Gave a show to Lord and Lady Kelvin and Lord Tennyson, who sent and paid for their messages. One of these was for Sir William H. Preece, via Bournemouth by wireless and then by land lines. At 3.15pm I left for Waterloo Station, London.”

It is of great interest that immediately after this all-important demonstration on 3rd June, the next day (June 4th) Kemp set off from London to Ballycastle.

Kemp & Glanville

In May 1898, Lloyds Insurance of London financed an experimental wireless link to test signal reception at Ballycastle from Rathlin. Marconi made a preliminary survey but the work was given to his right-hand man, George Kemp, who hired Edward Glanville, a Trinity College Dublin graduate, to assist him. They in turn hired islander Johnny Cecil as a labourer.

Kemp was in Ballycastle conducting wireless experiments from Saturday, June 4th, to Monday, July 6th, and from Monday, July 25th, to Thursday, September 8th, 1898.

Significantly, Marconi did not accompany him to Ballycastle on 4th June, and probably for a very good reason for the young scientist was providing a service to Queen Victoria and the Prince of Wales (afterwards King Edward VII), both of whom were then at Osborne. The Prince had the misfortune to slip and fall heavily upon one knee, injuring it so severely that it was at first feared that his limb might be stiff for the rest of his life. His Royal Highness was compelled to remain on board the Royal Yacht, which was anchored in Cowes Bay.

Messages to Royal Yacht

The Prince who had already taken a deep interest in wireless telegraphy, thought that it might be possible to establish communications in this way between the yacht and Osborne House, where the Queen was staying. Marconi was naturally delighted to place his knowledge and skill at the service of the Queen. Two stations, one on board the yacht and the other at Osborne House, were soon established and in working order.

For the next 16 days while Kemp was proceeding with his experiments at Ballycastle. Marconi was on board the Royal Yacht, sending altogether about 150 messages by wireless telegraphy. Not one of these messages had to be repeated. The new system of wireless telegraphy attracted widespread attention and remarkable progress was made in the next few years.  Before the year 1898 had run its course, messages were being sent successfully between such places as Ballycastle and Rathlin Island and between the South Foreland Lighthouse and the East Goodwin Lightship. The wireless experiments between Ballycastle and Rathlin in the summer of 1898 and extending over the months of June, July, August and early September were the first such demonstrations for Lloyds, the London based shipping agents.

Eventful Summer

The summer of 1898 must have been a particularly eventful one in the life of Ballycastle. On July 30th the town was thrown into a state of excitement by the arrival of a couple of motor cars which had come round the coast from Lame via Cushendall, and put up for a time at the Marine Hotel. They subsequently left for the Giant’s Causeway, both cars had a full complement of passengers, most of whom were English tourists who had been staying at Henry McNeill’s. Ltd. Larne.

 

 

790745_1449820_800_530_phoenix_wagen_1899

(A Daimler ‘Phoenix Wagon’ 1899)

During their short stay the motors were surrounded by quite a crowd, who were taking a lively interest in them, for it was a mere thirteen years previously, that Gottlieb Daimler had patented the light high-speed two-cylinder petrol engine which would be much used in motor cars.

…..” The present lovely weather has attracted a considerable number of visitors and the town presents an animated appearance”…

so ran a contemporary report of the occasion!

I might also add that it was in this memorable year of 1898  that the Ballycastle Lawn-Tennis Club held its first annual tennis tournament.

Complaints from Lloyds

The wireless experiments at Ballycastle were the outcome of complaints by Lloyds about not being able to report steamers from Torr Head (Lloyds signalling station on the north-east corner of Ireland) in spite of the fact that the steamers were able to report to what is now known as Rathlin East Lighthouse. Ships passed close to this lighthouse, off Altacarry Point and even in a, fog they could signal their numbers, etc., by means of flags; but be this as it might, Rathlin had no means of conveying this information to Torr Head.

“ Lloyds requested me”, wrote Kemp, “to fit a wireless station at Rathlin Lighthouse and another at Ballycastle, and I travelled on to Ballycastle at 1 p.m on Saturday, June 4th from which place I communicated with Lloyd’s agent, Mr. Byrne, at 11 a.m.  I studied the plans and surveyed the coasts of the North of Ireland and Rathlin Island.  The mast at the Lighthouse was 60 feet high and 30 feet from Lloyd’s hut.  I left Ballycastle with Mr. Wyse and inspected Rathlin Island returning at 6.10pm.”

Four Stages

In the course of the Wireless experiments at Ballycastle during the summer of 1898, four stages may be noted:

Stage I : A Short Stage

Stage I began on Friday, June 10th. when Kemp met Mr. Hough from Lloyds in Ballycastle at 11.00 a.m. They arranged to experiment at the coal store, now the Pier Pavilion with the aerial leading over the road to a small mast on top of the cliff, now the car park opposite Hillsea Hotel. The mast was that appertaining to the coastguard station.

This is generally believed to have been the first wireless installation ever set up in Ireland.

Next day Kemp and Hough went to Rathlin and upon their return Kemp fitted up the station in the coal-yard. The following Monday, Kemp started teaching Lloyd’s agent, Mr. Byrne, and his sons, the Morse code in the hope of getting their help until Lloyds sent someone to take charge of the station.

Four days later Kemp was in Belfast where he tried, unsuccessfully as it proved, to obtain masts for the Rathlin and Ballycastle stations. On June 22nd and June 23rd, 50 Obach cells for transmission were fitted at the coalyard station at Ballycastle Quay, and 50 were fitted at the Lighthouse Station on Rathlin. Mr. Byrne and his sons received further instruction from Kemp, this time in the working of the coal -yard station.

http://lisburn.com/books/ulster/ulster-images/ulster-guide-23.jpg

(Coal-Yard at Ballycastle Quay visible to the right of picture)

On 2nd July, wire and insulators arrived in Ballycastle from London “by the last train” (as Kemp describes it), and by July 5th half the wire, insulators and stores were taken to Rathlin Island and fitted up at the station there for transmission. Kemp instructed Signalman Dunovan and his two sons in the working of the station. Next day news came from London to the effect that Kemp was to take all the apparatus, “half a a ton of gear,” as he calls it from Ballycastle to Kingstown (Dun Laoghaire), for the Kingstown annual regatta.

So ended Stage I – a very short-lived one- in the fitting up of the two experimental stations at Ballycastle and Rathlin.  What happened at Kingstown does not concern us here, suffice it to say that the apparatus was employed to transmit reports of Kingstown Regatta to a Dublin newspaper- The Daily Express- from a steamer, “The Flying Huntress” in Dublin Bay.

Mr Marconi himself was present at these experiments, the first ever of messages being sent from sea to land by a vessel in motion. Upon the conclusion of the regatta, Kemp and a young man named Edward Edwin Glanville a native of Blackrock, an engineering student at Trinity College, Dublin, and employed as assistant to Marconi, set out by train from Dublin to Ballycastle.  This brings us to Stage II in our story of the early wireless experiments in Ballycastle.

Stage II : Aerial on A Spire

Glanville was put in charge of the Rathlin island station with instructions to transmit to Kemp at the Ballycastle end every day.  Kemp records in his diary…

“I received at various places and on the cliffs along the coast in the vicinity of Ballycastle and received the best results on an aerial connected to the Roman Catholic Church spire in the town, but as there was no house or room available there—and the Company would not let me use a hut—there was not much chance to make a speedy job, as there was delay in getting spares from Belfast.”

image (Church of St. Patrick & St. Brigid,  Ballycastle)

The chapel spire must then have presented a very new appearance; for it had been added to the building in 1891, thanks to the John Lawless bequest, the chapel itself was built in 1870. Kemp is at pains to state that these good results were the outcome of the courtesy and co-operation extended by the Parish Priest, the Very Rev. John Conway, V.F. Little wonder that later on in the course of the experiments in September when Marconi came to Ballycastle, he and Kemp showed their appreciation of this favour by calling on the Very Reverend gentleman, presumably at the parochial house on the day before they left Ballycastle for London.

 

Stage III : White Lodge image

(White Lodge/ Kenmara House, Ballycastle)

The third stage of the experiments followed when Kemp managed (as he tells us)

….“to get the loan of a small bedroom in a lady‘s house on the cliff, and the loan of a jib of a crane in the pier yard (now the car park) which served me for a lower-mast.”

This house, as Marconi subsequently explained to Kemp, belonged to  a Mr.Thomas Magregor Greer, M.A., T.C.D., Solicitor of Ballymoney.  At the time of the wireless experiments it was rented by Mr. Greer’s brother-in-law, Mr. Talbot Reed, 1, Hampstead Lane, Highgate, in the City of London. The house, known as White Lodge, is currently the property and residence of Colonel H. A. Allen, D.S.O.

Tragedy On Rathlin

A sad tragedy occurred on Sunday, August 21st in the course of the third stage of the experiments when young Glanville, out for a walk on Rathlin on that afternoon accidentally fell over a cliff and was killed. This sad accident was quite unconnected with the experiments. The people on the island had often seen Glanville, who was interested in geology, climbing over the cliffs and this was no doubt the cause of the accident. The verdict at the inquest was accidental death. but the jury, presided over by Mr, J. P. O’Kane,(father of Mrs. Boylan), added the following rider—“That we beg to tender our deepest sympathy with the parents of the deceased, and also with Mr, Kemp and the other members of the staff of the Wireless Telegraph Company, with whom the deceased worked so cordially and we desire to place on record our sorrow at such a tragic ending to so promising a career, connected as it was with one of the most important discoveries of the century.”

Mr. Glanville’s body was taken to the mainland by the s.s. “Glentow” and from thence by rail to Dublin, for burial.

Stage IV: Colonel Allen’s House

We now come to the fourth and last stage of the early wireless experiments in Ballycastle, seventy years ago almost to the day. On August 24th Kemp proceeded to erect a new mast in a field 104 feet to the top of the sprit and 104 feet from the window of a child’s bedroom, which was loaned to him, at the northern side of what is now Colonel Allen’s house. Thus two different bedrooms in this house were used in the course of Stage III and Stage IV of the early wireless experiments in Ballycastle. marconihouseS

(White Lodge/Kenmara House, Ballycastle)

Next day Kemp states that he finished the station and adjusted the receiver and inker. He instructed Mr. Byrne in all the details of the transmitter and requested him to follow (when he received the dots and dashes from Kemp) on the inker. He sent messages to, and received messages from, Mr. Byrne until 1 p.m. on that day, left the station on Rathlin in the charge of Mr. Dunovan and sons, and returned to Ballycastle.

Satisfactory Experiments

This trip from the island to the mainland must have been something in the nature of an adventure for Kemp, as it took four hours to cross what he describes as “that very dangerous piece of water” as a result of which, he records, “I caught a terrible cold.”

The experiments were now apparently proving very satisfactory, next day messages were sent and received from 10 a.m, to 6.30 p.m., mostly red each way. Ten ships were reported and Lloyds’ agent (Mr. Byrne) sent a report to Lloyds concerning the day’s work which had been carried out in a dense fog. The following day, August 27th, two more ships were reported to Lloyds.

Marconi’s Arrival

Kemp must have been a somewhat sick man when next day – Monday, August 29th, the eve of the Lammas Fair – Mr. Marconi arrived in Ballycastle by the 6.15 p.m. train. One is tempted to wonder how the 24 year old scientist contemplated the scene as the engine driver sounded the whistle at Kilcraig and as the narrow gauge train neared its journey’s end by rounding the sharp Ballylig curve, passing Broombeg Wood, (now known as Ballycastle Forest) and  finally reaching its destination at the Ballycastle railway terminus.

image

(Ballycastle Narrow Gauge Railway 1880 – 1924)

The rough passage from Rathlin two days previously had evidently proved too much for Kemp, because he records that he…”had to take to bed suffering from Neuralgia and fever.” He continues.. “The weather was very windy and wet during the day and I was forced to keep the window open to enable me to transmit, and this increased the violent cold that I caught in the boat.” Conditions were no better by August 28th. “Weather,” writes Kemp, “still very wet and wind blowing a gale. I had to remain in bed, taking medicine which reduced the fever, but made me very weak.” ….

As Marconi made his way from the station to the Antrim Arms Hotel, almost certainly by way of the Poor Row or Station Street, I wonder if he delayed at the stalls erected on the Diamond in readiness for the the Fair next day or was he more concerned with his scientific pursuits?

P. W. Paget, one of his technical assistants, has put it on record that Marconi had little interest in anything outside wireless. In any event, he may have been too deeply concerned about the death by accident of his assistant, Glanville, only a week previously to bother too much about the fair. Certainly he remained indoors in the Antrim Arms Hotel all the evening.

On Lammas Fair Day

On the following day, (Lammas Fair Day) Kemp called up Rathlin Island station and found that their sensitive tube had been broken. He records … “I told them to stop for a few days. The weather was still very wet and windy and I spent the remainder of the day packing apparatus and transporting it to the Antrim Arms Hotel. I told Mr. Byrne that he would have to get a station built for carrying on further work, as the present room must be given up because Mr. Greer of Ballymoney, the owner of the property was coming back…. I subsequently tried to go to Rathlin but found that no boat had been there since August 25th when I crossed.”

On the second day of the Lammas Fair, Kemp tried to get a boat to take Marconi and himself to Rathlin, but no one would venture to cross. Instead, the scientist and his assistant went to Fair Head, whence they saw Rathlin, Torr Head, the Mull of Kintyre and the two islands at the mouth of the Clyde, Sanda and Ailsa Craig.

One of the Largest

The Lammas Fair held that year  was one of the largest held in the district for years. Buyers from Belfast, Derry and Armagh and across the Channel attended, some even before the day of opening. It was thought that it would have been the most successful for years, but unfortunately from eleven a.m to seven p.m a continuous heavy downpour of rain set in and damped those attending and practically spoilt the day. The second day compensated by being gloriously fine, but the rain of the previous day acted on the attendance.

There was a great show of sheep, cattle and horses and prices were as follows: Bullocks, first class £14 to £17; second class £11 10s to £14; third class £9 to £11; heifers, first class £12 to £16 10s; second class £10 to £12; third class £7 to £10.  Milch cows £14 to £16 and £9 to £11.  Sheep from 17s 6d to 42s.  A large sale was made in this class; lambs 18s to 33.  Bullocks 27s per cwt; middling class 23s 6d per cwt.

There was a splendid show of Cushendall ponies. The Islay fish trade was most successful, ling, cod, etc., being in abundance. All the lots were sold from 3s 6d to 7s 6d a bundle. In the fruit market there was a great supply. The lots were chiefly brought by wholesale traders from Ballymena and Belfast. Such were the chief characteristics of the Ballycastle Lammas Fair of seventy years ago.

Crossing to Rathlin

 

On Thursday, 1st September, Mr. Marconi and Kemp started from Ballycastle at 9 a.m. and crossed to Rathlin in one hour with, as Kemp describes it, “a fair wind and large sail.” They visited the lighthouse, beside which the aerial mast was erected; some of the cement blocks inscribed “Lloyds” and used to hold the stays of the mast may still be seen there.  As a memorial of the early wireless experiments here, it is surely possible for one or more of these to be brought to Ballycastle and placed somewhere for all to see in the new promenade or municipal gardens as described in Mr. Fergus Pyle’s article on “Civic Week in Ballycastle” in the “Irish Times.”

(Rathlin East Lighthouse – Present Day)

Marconi and Kemp found that the ridge of Lloyds’ land bore north and south and that the station at Ballycastle bore south-south-west. Kemp induced Mr. Dunovan and his sons to pack up the apparatus while he took Mr. Marconi to Badlyconagan to see the cliff where Glanville lost his life. Thus by early September, 1898, the experiments came to an abrupt end at Ballycastle. Whether or not the accidental death of Glanville on Rathlin had anything to do with this it is impossible to say.

Thanked for Co-operation

Mr. Marconi and his assistant returned to Balllycastle at 2 p.m. “pulling and sailing in one and a half hours.” Upon their return they visited the Very Rev. John Conway, P.P., V.F., the proprietors of the Water Mill (presumably Messrs. Alex. and John Nicholl), and the landowners, in the area (presumably the agent to the local estate). This was evidently to thank each of the parties for their help during the experiments.

Next day, September 2nd, Mr. Marconi left for London and Kemp took down the mast that he had erected (104 feet North at Greer’s house at the top of the Quay)and returned all the stores to the Antrim Arms Hotel. Six days later, on September 8th, Kemp left Ballycastle for London, travelling via Belfast and Fleetwood.

Kemp regarded the Ballycastle experiments as very successful demonstrations in spite of (as he put it)… “the most peculiar instructions”  ever given to him. In 1897 the whole of the G.P.O’s skill was put on to a similar job, but in the case of the Ballycastle/Rathlin experiments, carried out under the auspices of Lloyds, he was sent without any assistance and he had to instruct all those who helped him.

Not Until 1905

Apparently the relations between Kemp and Lloyds were not as friendly as those between Kemp and the G.P.O. At all events the Marconi system of wireless telegraphy between Ballycastle and Rathlin was not brought into use until 1905. It replaced the old parallel system between Ballycastle and Rathlin, purely G.P.O. affair, and was distinct from the experiments described, which, as I have explained, were carried out under the auspices of Lloyds.

Despite the somewhat adverse criticisms of Kemp in his relationship with Lloyds, the Ballycastle/ Rathlin experiments must, nevertheless, have had some definite significance in the development of wireless telegraphy.  Within two years, in 1900, Marconi had taken out his famous patent No. 7777 for “tuned or syntonic telegraphy”

marconi-clipping-photo

In fact Marconi had transmitted signals to such a distance—over 200 miles—as to convince him that the electric waves, instead of being projected into space, as some prophets averred would be the case, would, as it were, cling to the surface of the earth, the curvature of which would, therefore, be no bar to the attainment of long ranges. Accordingly, he determined to make an attempt to send signals across the “Atlantic, and for that purpose proceeded to erect a powerful station at Poldhu, in Cornwall, with a similar station at Cape Cod, in the United States of America

Wrecked by Storm

The masts and aerial at Poldhu were wrecked by a storm in September, 1901, and though they were repaired by the end of November a similar mishap at Cape Cod threatened to delay the test for several months. To save time Marconi went to Newfoundland and installed his instruments in a disused barracks on Signal Hill, St. John‘s. He had intended to support his aerial by a balloon, but as this was blown away, he substituted a kite which, with great difficulty owing to the strong wind, was kept at a height of about four hundred feet.

Marconi_at_newfoundland

(Marconi watching associates raising the kite used to lift the antenna at St. John’s, Newfoundland, December 1901)

It had previously been arranged that at fixed times the Poldhu station should send out the signal for the letter ‘S’ on the Morse code—three dots—and on December 12th, 1901, both he and Kemp using a self-restoring coherer, repeatedly heard in their telephone the three clicks which showed that the electric waves had traversed the 1,800 miles separating St. John’s from Poldhu.

This momentous incident was recalled by Mr. P. W. Paget, Mamoni’s first technical assistant in these words:—

“I was with him at Signal Hill, Newfoundland, when he received the first wireless signal, the letter ‘S’, ever transmitted across the Atlantic. It was from the station at Poldhu, Cornwall. He showed no excitement, calmly handing over the earphones first to Mr. George Stephen Kemp, who was also assisting him, and then to me, with the words  ‘Can you hear this?’  He was never unduly elated and never unduly depressed.  When the twenty masts, erected at Poldhu for the Transatlantic transmission collapsed in a gale, Marconi looked at the wreckage and said quietly to me  ‘Well, they will be built again.’ That was all. Marconi had few hours of sleep. He had little interest in anything outside wireless.”

Commercial Basis

Soon afterwards preparations were begun for the establishment of wireless telegraphy between England and America on a commercial basis. We are reminded here of those words of Longfellow:—

The heights by great men

reached and kept

Were not attained by

sudden flight,

But they, while their

companions slept,

Were toiling upward

in the night.

 

“At that time” (1901), wrote Marconi, “and for long afterwards, certain important sections of the technical Press in this country were against me, and spared no efforts in their determination to discredit both me and my work on long-distance wireless communication. From the first, however, ‘The Times’ declared its belief in me, and was swift and forceful to rebuke those who persisted in a policy of disparagement. Radio has made very great strides since 1901, and yet I often look back to those early days and remember with deep gratitude what a wonderful encouragement of support it was to me to know that a great newspaper like ‘The Times’ had faith in me, even as regards my work relating to the utilization of electric waves for world wide communication.”

In the words of his contemporary, Rudyard Kipling, Marconi was one of those who could

“meet with triumph and disaster and treat those two imposters just the same.’

Retained Vigour of Youth

marconi telephone article-2175365-0060D69600000258-672_306x459                                              Marconi at his desk in Electra House, Moorgate, London in 1934

To the end of his life Marconi retained the vigour of youth, which was also apparent in his personal appearance. He died on 20th July, 1937 at the age of sixty three. Kemp pre-deceased him by four years. Marconi described Kemp as his first assistant, collaborator and friend, and would always be regarded as a pioneer of the wonderful science of wireless telegraphy.

When Marconi was about to be admitted to the honorary degree of Doctor of Civil Law in the University of Oxford, he was referred to as “the magician who found a means of transmitting signals from shore to shore and from ship to ship.”

Granted that the association of this locality with wireless telegraphy in its infancy is due entirely to the first demonstration for Lloyds so as to facilitate shipping, may have been of little significance, as compared, let us say, with the Poldhu – Newfoundland transmissions, nevertheless, Kemp’s installation of apparatus in Ballycastle in June, 1898 -apparatus that would have included a Righi spark gap and an induction coil, a Morse inker and relays was almost certainly the first of its kind in Ireland.

It is true that Preece was already conducting his own experiments wlth wireless by the induction method, which made use of long parallel wires, and that subsequently such a method was established between Ballycastle and Rathlin and was a purely Post Office affair. The service (sparking coil, coherer, etc.) established in 1898 at the request of Lloyd’s Shipping Agency terminated-1t would appear rather abruptly- in September 1898 because of objections on the part of the Post authorities. Post Office engineers carried out tests of their own by an induction method. This involved a heavy gauge wire carried on poles and down into the sea at Murlough Bay. In 1905 or 1906 the Marconi Company was again asked to install its system because, by that time, it had overcome opposition and was well established. But be all this as it may, the Ballycastle Rathlin trials undoubtedly stand out in the history of wireless communication and in states unborn and accents yet unknown, wireless discoveries and developments will still be broadening their influence upon the lives of men.

54445-first-radio-conv-1930

(First Radio Conversation 1930)

Footnotes to Marconi and Ballycastle. By D.J.McGill. B.Sc.,

Hugh A Boyd’s lecture reprinted above with the permission of Mrs Boyd is an account of the activities of the Marconi Company in Ballycastle and Rathlin Island. The locations are accurately identified and the events of Mr Marconi’s short visit are well documented. My article is intended to correct some popular misconceptions that have developed over the years about Marconi and Ballycastle. I have also added some notes from my research into Marconi and the history of ‘wireless’ to supplement Mr Boyd’s lecture.

A legend has grown up around one of Hugh Boyd’s coal-yard cottages in the townland of Tornaroan. This is one of two remaining sites along the shore where locally mined coal was stored prior to shipping 1. At first the yard was open, but after ‘losing much coal’, Boyd decided to enclose the yards with high walls. The site is now called ‘Marconi’s Cottage’ and here the Antrim County Council erected a commemorative plaque to Marconi. It is variously related that Marconi set up experiments there, making the first ever transmission over water to Rathlin Island, even that he owned the cottage and lived there. There is no evidence whatsoever connecting Marconi, or his company, to this site. He never worked there, nor stayed there, and there is no account of him going there during the stormy few days he visited Ballycastle.

It is a matter of record that Marconi made only one visit to Ballycastle, which is related in Mr Boyd’s lecture. His visit, over a four-day period from the evening of August 29th 1898 to September 2nd 1898, is fully documented in George Kemp’s diaries. Marconi’s close assistant George Kemp kept detailed records of his work, which are kept in the Marconi Company’s archive.

It is also important to distinguish between experimental work and the many demonstrations that Marconi was asked to set-up. The Ballycastle – Rathlin demonstration of wireless telegraphy was in response to a request from Lloyd’s.2

With regard to Tornaroan, a letter from an official Marconi Company historian, Mr Bell, states that after examining Kemp’s note-books, he was certain that there was no work done at the coal-yard cottage in Tornaroan3 . Mr Bell further stated that it was the later work of the Post Office in 1905 at that location, which has been mistakenly attributed to Marconi. A fact the company pointed out to Antrim County Council at the time they put up the plaque.

Unfortunately Ballycastle was not the location for the first transmission over the sea, nor did it host the first commercial wireless traffic. The truth of the matter is that the first public demonstration of wireless telegraphy over the sea, took place across eight miles of open water in the Bristol Channel in 1897. The demonstration was for the Royal Navy. A similar demonstration for the Italian navy was conducted that year, over twelve miles of open water.

who-invented-radio-1

The world’s first paid ‘radiogram’ was sent from the world’s first permanent wireless transmitter at the Royal Needles Hotel on the Isle of Wight to Marconi’s station at a house called ‘Sandhills’ on the cliffs at Bournemouth. No less personage than the prominent scientist, Lord Kelvin accompanied by Alfred Lord Tennyson sent three telegrams to friends and paid a shilling each for the privilege. This notable commercial landmark occurred on Friday June 3rd 1898, the day before George Kemp arrived in Ballycastle.4

The Marconi stations at South Foreland and on the East Goodwin Lightship, near the mouth of the river Thames pre-dated those at Ballycastle and Rathlin. In fact lives were saved after a collision between a German steamer and the East Goodwin Lightship. A distress signal was sent and the lifeboat service were able to launch. This was the first recorded incident of ‘wireless’ saving lives.

The first major lifesaving event involving Marconi’s ‘wireless telegraphy’ was when two liners the ‘Republic’ and the ‘Florida’ collided in fog off New York in 1909. A total of 1,650 people were saved by the ‘Baltic’ after it’s ‘Marconi operator’ received their distress signal.5 One would have thought that this would have led to more installations on ships, but in 1910, nine years after the success of his transatlantic transmission, there were still only 60 ships in the world with wireless equipment.

In 1912 the ‘Titanic’ was lost. The tragedy was that the ‘Californian’ was in sight of the ‘Titanic’ when she sent out distress signals, but did not receive them because her sole ‘Marconi man’ had just gone off watch. The ‘Californian’ sailed over the horizon unaware of the catastrophic events just a few miles away. After the ‘Titanic’ disaster the Board of Trade insisted that all ships carry wireless equipment, and Marconi was hailed a hero.6 Proof if needed that it is disaster that forces innovation.

But, consider this, Marconi would not sell his wireless installations, they could only be leased and manned by his own company operators. The cost of this arrangement must have affected ship-owners’ decisions about whether or not to install one of Marconi’s ‘wireless stations’ . This business approach could be seen as ‘over protective” of his technology, even exploitative when you consider that most of the patented equipment he used to build his ‘stations’ was freely available on the open market. A Marconi wireless station which was developed from Heinrich Hertz’s discovery of electromagnetic radiation (hence Radio), was assembled using the ‘Coherer’ invented by Branley, (and refined by Sir Oliver Lodge), the ‘Thermionic Valve” or ‘Diode’ invented by Sir Ambrose Fleming and powered by Poulsen’s arc generators.  Sir Oliver Lodge is acknowledged as the first person to demonstrate wireless signals. He transmitted a signal from Lewis’s department store to the clock tower of Liverpool University’s Victoria Building in 1894. It is also a fact that Sir Oliver filed his patent for ‘selective tuning’ before Marconi.7 After Lodge, Lord Rutherford transmitted a signal over 3/4 of a mile at Cambridge in l895 and Marconi followed with his first demonstration at Tonybee Hall in London on December 12th 1896.

image

 

 

marconi plaqueS

 

 

 

marconi memorial closeup

Further Footnotes:

1. See Glynns Vo1 17.

2. Correspondence dated 15/10/1963 from the Marconi Company to Mrs H.O Hamilton,.

3. Correspondence as (2). Mrs Hamilton, according to the correspondence, had been present on Rathlin Island with Marconi. She was an aunt of John Humphries of the Quay, who’s grandfather owned the coal yard where Kemp set up the first wireless station in Ballycastle. The yard was later covered over and used as a garage and then more famously it spent years as the home of Doherty’s ‘Housey Housey’. I wonder what Marconi would think of the electronic amusement machines in there now ?

4. November 1897. The 125 foot mast at ‘Sandhills’ was used to exchange messages with the Island station and vessels in Poole Bay and the Solent. ‘Heritage’ Roger Guttridge, Bournemouth Evening Echo (Dec 12th 1991).

5. ‘Eureka’ Ed Edward De Bono,Thames and Hudson London (1974).

6. ‘Since the first wireless message , which has been sent from New York to Chicago in 1909, the (American) Federal Government has already ordered steamships to carry radio equipment on board for safety’s sake’. After the Titanic they also insisted that all ocean going vessels carry sufficient lifeboats for all passengers. ‘The Almanac of American History’ Arthur Schelesinger Jr, Bison Books Corp (1983).

7. Biographical Dictionary of the History of Technology Eds. Lance Day and Ian McNeil, Routledge, London, (1996).