As soon as it was found to be working perfectly and accomplishing all claimed for it, Mr. Robinson, who aimed to be the most severe critic of his own work, entered systematically into a deeper study of the system from the standpoint of a railroad man, with a view of finding the weak points in it, if any existed. He soon discovered the following serious defects, which are inherent in all normally open circuit or wire systems of automatic signaling, without exception. Such systems are extremely limited in their functions, and _may, under certain circumstances_, show a SAFETY signal when the danger actually exists which they are designed to avert, as in the following cases: _First_: A train enters regularly upon the section and sets the signal at danger; the train breaks in two, the forward part passes off the section, reverses the signal and shows ALL CLEAR behind that portion of the train remaining on the section; and a following train, lured on by the false signal ALL CLEAR, dashes into the stalled portion of the preceding train left standing on the section. This is extremely liable to happen on sharp curves and grades, where breaks are not of uncommon occurrence. _Second_: A train may enter within the section from the opposite end or from a siding, thus blocking the track, while the signal, not having been affected, shows ALL CLEAR as before, a false signal again. _Third_: If a line wire break or other connection be interfered with accidentally or maliciously, or the battery fail from any cause, the signal will invariably show ALL CLEAR, under every train passing over the section, a false signal again. Mr. Robinson at this early date recognized the above serious objections as inseparable from open circuit system of signaling, apparently, before these defects were recognized by any one else, and at once entered upon the solution of the problem presented, of eliminating these objections by producing a signal system which would meet all the requirements of safe and efficient railroading. He reasoned, first, that to accomplish this result every car and every pair of wheels in the train must have controlling power over the signal throughout every inch of the block section, and second, the signal should go to danger by gravity, the electric current being used to hold it at safety. Could these two results be accomplished? Could the rails be used in any way to carry the primary current in a reliable manner? Manifestly not by any open circuit means, for the reason that sections of rails of even moderate length, on open circuit, would form a good ground, especially in damp or wet weather, thus keeping the circuit closed continuously and preventing any operation of any kind. He at once cast aside this open rail circuit idea as fruitless, and having previously, in 1869-70, used the short circuiting principle in his model, as above stated, he concluded that this principle presented the only possible solution of the problem.