As 1800's came to a close, the United States trailed Europe in radio development. Led by Marconi, Europeans were reporting continuing progress, while in the U.S. the best that various individuals could do was to try to piece together enough technical information in order to replicate some of those advances. Although in this account Professor Green was able to send signals over a significant distance, he was merely duplicating the work of Marconi and others, and does not report making any significant modifications or advances over their original designs. He isn't even able to match the distances achieved by some of his European counterparts. Still, this is one of the first reports of success on U.S. soil in relatively long-distance radio signaling.
American Electrician, July, 1899, pages 344-346:



    The apparatus for demonstrating the effectiveness of telegraphing by means of Hertz waves is comparatively simple. In the experiments recently conducted at the University of Notre Dame all the essential parts of the sending and receiving stations were found in the regular equipment of the electrical and physical laboratories, except a few details which were constructed in the workshops by the students. Fig. 1
    It has been known for several years that electric waves are propagated from an insulated conductor when this conductor is rapidly charged and discharged in such a way as to produce a series of very sudden disturbances in the surrounding ether. The waves are said to travel with about the velocity of light, but their length is much greater than the length of light waves, and this length depends on the character of the spark produced at the discharging terminals. These terminals, as used in wireless telegraphy, are always spherical in shape, with great variations as to size.
    These electric waves of various lengths may be refracted or reflected like the waves of heat and light. They pass through many substances but are absorbed or reflected by metals.
    The parts required for a complete sending station are an induction coil capable of producing a stream of sparks from one to ten or more inches in length. A smaller coil will of course answer if the receiving station is located close to the sending apparatus--a suitable primary or secondary battery to operate the coil and a discharger or oscillator. The oscillator may be two large metal spheres placed about half an inch apart on insulating supports. The space between them is sometimes filled with a heavy petroleum oil, but this is not essential. Additional spheres or plates may be connected with each of the large spheres between which the discharge takes place. These additions increase the capacity of the oscillator and change the character of the spark produced.
    A simpler and more easily constructed oscillator is the vertical wire with a metal sphere at its upper end, in connection with two small spheres about an inch in diameter. This is the apparatus Mr. Marconi used. The lower end of the vertical wire is attached to one of the spheres and the other is fastened to a steam or water pipe.
    An easy way to make this form of oscillator is to drill holes in two small metal spheres and put these spheres on the adjustable points, with which all induction coils are supplied, at the terminals of the secondary windings. Fig. 2
    The vertical wire should be attached to one of the terminals of the secondary coil, and the ground wire to the other. (See Fig. 1). Then adjust the distance between the spheres, and the tension on the vibrating interrupter spring until a stream of thick, white, noisy sparks is produced when the primary switch is closed.
    For the receiving station there is needed a coherer with a decohering tapper, a sensitive relay, and a sounder or a Morse recorder.
    The coherer is, of course, the essential part of the receiving station. It may be made from a small glass tube, of, say, one-eighth of an inch internal diameter, and it may be about an inch and a half or two inches long. Into the ends should be fitted brass rods, and these plugs of metal should have a space between their ends near the middle of the tube, which space can be increased or decreased in length by sliding out and in one of the metal plugs. To complete the coherer a small quantity of filings of metal, such as silver or nickel, or a mixture of these metals, is put in the space between the plugs in the glass tube. Filings from the ordinary five-cent piece will answer very well.
    This contrivance has the property of changing its resistance when it is acted on by the waves. A space in such a tube one-fourth of an inch in length, loosely filled with filings in their normal condition, will have a resistance of several megohms, but when the waves from the sending station strike the vertical wire attached to one of the terminals of the coherer the resistance of the coherer falls to about ten ohms. The coherer is put in circuit with a high-resistance relay, and a dry battery of one or two cells. When no waves are passing, the coherer has so high a resistance as to virtually open this circuit; but when the action of the waves begins, the decrease of the resistance closes the circuit, and the armature of the relay is drawn down and closes the circuit on a sounder, or on whatever we wish to add to the receiving set. The resistance of the coherer remains at this low value even after the waves have ceased to strike the vertical wire, but it becomes very high again when the filings are shaken. A little tapper like that on an ordinary electric bell is made to strike the side of the tube for this purpose. The relay in circuit with the coherer may close a circuit to operate such a decohering tapper or the filings may be shaken by tapping the tube with a lead pencil or by other mechanical means. Fig. 3
    From the diagram, Fig. 3, it will be seen that the minute oscillating current produced in the vertical wire set up by the passing of the waves has the choice of two paths to earth; through the coherer, C, or through the relay, R. To prevent its passing to the relay and leaking across through the metal, well insulated choking coils A & A are placed in the circuit with the relay. The rubber-covered spools of a four-ohm telegraph sounder will do for these coils. The small current is then forced to go through the filings to ground, and this causes them to cohere as long as it is passing, even though they are shaken. It is well to put some non-inductive resistance (such as the lamp shown in the diagram) across the points of the vibrating tapper and the tongue of the relay to prevent a spark there. It is especially important that the spark be prevented at the tapper, as it is so close to the filings that if considerable sparking occurred here it would set up waves which would cause the filings to cohere to some extent and prevent the decohering action of the tapper. The great difficulty in the first experiments at Notre Dame was to make the filings decohere quickly. If this action is not prompt it is impossible to make dots and dashes so that they can be distinguished. Fig. 4
    Fig. 4 shows the coherer tube mounted on hard rubber supports on the top of the metal cover of an ordinary buzzer. An extension from the armature of the buzzer reaches through the cover and strikes the tube.
    We employed in all our experiments an eight-inch induction coil of American make, with a heavy Apps vibrating circuit breaker. The current for the primary was furnished by a battery of six five-ampere storage cells. The double-pole jack-knife switch on the base of the coil served to open and close the circuit at first, but later a heavy telegraph key with special contacts was used. The coil is shown in Fig. 2.
    An ordinary one hundred and fifty-ohm telegraph relay was employed in the receiving set for distances up to one-half of a mile, then a similar one of fifteen hundred ohms' resistance, with a lighter armature, was substituted for the ordinary relay, and it gave good results at distances of about two miles. Later a more sensitive relay was made by adding platinum contact points to the needle of a portable galvanometer of the D'Arsonval type. This responded more readily than either telegraph relay. Five small dry cells were used to operate relay, tapper and sounder. Fig. 5
    We first suspended an ordinary No. 14 rubber-covered wire about ten feet long from the ceiling of the physical laboratory and attached its lower end to one of the binding posts at one of the terminals of the secondary of the induction coil; the other terminal was connected to a steam pipe near by. These binding posts also carried the adjustable pointed rods, on the ends of which were placed the polished brass spheres one inch in diameter. The tension of the vibrator springs was adjusted so that a spark about an inch long passed between the spheres when the circuit was closed. A similar wire about six feet long was suspended in an adjoining room with its lower end connected to one end of the coherer. A wire from the other end of the coherer was attached to a steam pipe. When the circuit was closed on the induction coil the filings in the tube appeared to be arranged in strings or chains as when they are strongly magnetized. Their resistance fell to about six ohms. A very vigorous shaking was required to cause them to resume their normal condition of high resistance. Cohesion took place in the tube of filings, but not so strongly, when the ground connection was removed from the coherer. Fig. 6
    The receiving apparatus was then placed in a building about one hundred feet distant, and ground connections were made to a steam pipe. The same short vertical wire was used and the relay closed every time the coil was worked. The windows and doors of both buildings were closed. Next a wire about twenty-five feet long, with a metal sphere at its upper end, was suspended from the roof of Science Hall. The lower end was brought through a window and attached to the induction coil, and ground connection was made from the coil to a steam pipe. A similar wire and sphere were suspended from a window in the top floor of a three-story building across the campus, at a distance of about two hundred yards from the sending station at Science Hall. The lower end of this wire was also brought through a window and attached to the coherer; the ground connection was made to a steam pipe. The impulses in this case were received as strongly as when we were working from one room to the next with the short wires. The receiving apparatus was next placed at a building about five hundred yards from Science Hall. The vertical wire was about twenty-five feet long and ground connection was made to a water pipe. The coherer responded readily at this increased distance. Its normal resistance, as measured with a Wheatstone bridge, was very high, more than a megohm. The tapper was then disconnected to prevent its shaking the filings, and when the circuit was closed on the induction coil at the sending station the resistance of the coherer instantly fell to ten ohms and remained at this low value until shaken.
    The induction coil was next placed at the foot of a steel flag pole about one hundred and twenty-five feet high. The vertical wire and sphere were suspended from the top of the pole on a wooden bracket, which held the wire about ten feet from the pole. Ground connection was made to the base of the pole. The receiving apparatus responded at distances up to two miles when impulses were sent from the station at the flag pole.
    In the last experiments at Notre Dame the vertical wire and sphere were suspended from a high church tower and the induction coil was placed about thirty feet up in the tower. The lower end of the vertical wire was brought through a window and attached to the coil. Ground connection was made to a water pipe. The active length of wire here was about one hundred and fifty feet. The impulses were received strongly at a distance of three miles, about one mile of this space being occupied by the city of South Bend. Another trial to Mishawaka, a distance of about six miles across the country, resulted in failure.
    Trials were made in Chicago among the steel buildings and overhead wires in the business district. When the sending station was located at the Polk street railroad station and the receiving apparatus at the "Tribune" building, a distance of about three-fourths of a mile, the relay failed to respond when the coil was operated. In this case a line drawn from the sphere at the top of the wire at the sending station to the similar sphere at the top of the wire at the receiving station would be intercepted by many electric light, telephone and telegraph wires running in every direction. These wires apparently absorbed the waves, for in another trial from the Monadnock Block, to the "Tribune" building, there was a clear space between the spheres at the tops of the wires. The receiving apparatus responded strongly. The distance in this second experiment was about half that at which the first trial was made.
    Later the sending apparatus was set up at the life saving station at the mouth of the Chicago river, and the receiving set was placed on a tug. One wire and sphere were suspended from the end of a pike pole lashed to the railing on the lookout of the station, giving an active length of about forty feet. Ground connection was made by attaching a wire to an iron pipe placed in the water. The other wire and sphere were suspended from a short mast on the tug, giving an active length of about thirty feet. Ground connection was made by attaching a wire to a piece of iron hung out over the stern of the tug. The impulses caused the sounder to operate very strongly until a distance of about two miles out in the lake was reached, when the sounder ceased to respond.
    It seems from these trials that it is much easier to operate over water than over land. This result is also indicated by the great distances Mr. Marconi has attained over water, while Ducretet, with the best instruments that skill can produce, has succeeded in sending messages only about five miles over the city of Paris, where there are no overhead wires to interfere. Our experiments with crude apparatus show that it is quite easy to reach short distances, but it is quite difficult to adjust all the parts to make the dots and dashes accurately.