About 1867 William Robinson, then a recent graduate from college, entered actively upon the development of an automatic signal system for preventing accidents of various kinds on railroads. His attention was called to the subject by the consideration of certain railroad accidents which had occurred, and for the prevention of which there were no adequate means known. From this starting point he developed such a system, and in 1869, constructed an elaborate model illustrating the same, which he exhibited at the American Institute Fair in New York City, in 1870. This system was what is now known in the art as a "wire" or "open circuit" system; that is, there were circuit-instruments in proximity to the track which were actuated by the wheels of a car. The action of the wheels on a lever at one point closed the circuit through a relay, whose magnet was so arranged that the instant it was magnetized it attracted its armature and kept its own circuit closed. The circuit of the magnet which directly actuated or controlled the signal was under control of the relay, which operated to open and close the signal circuit directly. When the train or car proceeded to the proper point beyond, it actuated a reversing lever, thus opening the relay circuit and reversing the signal. In the model described the reversing lever operated to open the relay circuit by cutting off the battery therefrom by short circuiting. This model was in continuous and perfect operation throughout the duration of the fair. At the close of the fair Mr. Robinson had some of his descriptive circulars left over. These he immediately sent out to railroad companies at random. One of these circulars, at least, was as seed sown in good ground. It elicited an immediate response from Mr. William A. Baldwin, general superintendent of the Philadelphia and Erie railroad, with the result that Mr. Baldwin, who was an old telegraph operator and a very able and progressive railroad man, on looking into the system was so impressed with its practicability and importance that he at once arranged with Mr. Robinson to make an installation of the system on his road. This was in 1870. At that time Mr. Theodore N. Ely, now chief of motive power (1906) of the Pennsylvania railroad, was assistant superintendent of the Philadelphia & Erie, and, under direction of Mr. Baldwin, furnished Mr. Robinson with all the facilities and material necessary for prosecuting the work of installation. This installation was made at Kinzua, Pa., and after a little experimenting was soon in perfect working order, performing all claimed for it, and considered satisfactory by the railroad company. This was a normally open-circuit wire system, however, controlled by track levers, as above described, in connection with the model.