the stronger will be the current in the line; so that the extent of the motion of the arm, a, in the direction of its length, that is to say, the direct component of the motion of the pencil along the line of the arm, a, is attended by a corresponding change in the current traversing the line. If the pencil makes a long up and down stroke there will be a strong current in the line, if a short one there will be a weak current, and so on. A precisely similar arrangement is used to transmit the sidelong motion of the pencil along the line, L. [Illustration: Fig. 5.] The current from the line, L, flows at the receiving station through a powerful galvanometer, G, to earth. The galvanometer has a stout needle, one tip of which is connected to a duplicate pen, P, by a thread, t, which is kept taut by a second thread stretched by a spring, s'. The current from the line, L', flows through a similar galvanometer, G', to earth. The needle of G' is also connected to the pen, P, by a taut thread, t', stretched by means of the spring, s. Now, since the needle of each of these galvanometers deflects in proportion to the strength of the current flowing through its coil, the points of these two needles keep moving with the varying currents. But since these currents vary the motions of the sending pen, the receiving pen controlled by the united movements of the needles will trace out a close copy of the original writing. We give on another page a facsimile of a sentence written by Mr. Cowper's telegraph. [Illustration: THE COWPER WRITING TELEGRAPH.] The receiving pen is a fine glass siphon, drawing off aniline ink from a small glass holder. There are thirty-two coils, C, in each circuit, with a corresponding number of contact plates, c, so as to get accuracy of working. A few Daniell's cells are sufficient to operate the apparatus, and writing has been already sent successfully over a line 40 miles in length. The writing may be received either of the same size or larger or smaller than the original, as the case may be. At present the writing must not be too hurried, that is, unless the characters are bold and well formed; but further improvement will, of course, quicken the working of the apparatus. The engravings, Figs. 1 to 4, illustrate the actual apparatus. Fig. 4 is a plan of the sending instrument, with the writing pencil, a, the traveling paper, b, the light connecting rods or arms, d (which correspond to a in the theoretical diagram above), the series of metal contact plates over which these arms slide, the resistance coils connected to these plates, and the battery and line wires. It will be seen that each arm, d, is connected to its particular battery, and each set of contact plates to its particular line. Fig. 3 is an elevation of the sending instrument, in which a is the pencil as before, c c the contact plates over which the arms, d d, slide, f f the coils, and b the traveling slip of paper.