It is to be observed, that the wheels H are of such a thickness, as to engage only one of the five type circles at once, and their distance from each other is such, that they take the same circle in the one series as they do in the other. Now, by moving the back axis a small quantity endwise, it is obvious that the wheel-II can be brought to act upon any of the five circles, or be placed in such a position as to be clear of them all. It is for this purpose that the head I, Fig. 2, comes through the frame of the machine ; for by means of this the axis can be moved on end, and by proper marks upon it, it may be set to any of the five circles. In these positions it is confined by a semicircular clip, which enters grooves turned round on the axis, and deprives it of longitudinal mo tion, unless when the clip is raised. This can be done by a nut coming through the back of the frame at K, Fig. 1. It has a short lever on the inside of it, which, when the nut is turned round, raises up the clip, and releases the axis while it is set to the re quired circle, and the clip being let fall into the pro per groove, confines it from any farther motion. In order that all the circles may stop at the exact point, when the figure is at the highest, and consequently when the surface of the figure will be horizontal, an angular notch is made on the inside of the figure circles, in the intermediate spaces between each fi gure ; and at the lowest point of the circle e, Fig. 1, a moveable pin is fitted into the fixed axis, with a spring, which gives it a constant pressure down wards. The end of the pin is formed spherical, and well polished, so that when the circle is turned round, it is forced into its hole in the axis ; but when ano ther notch in the circle presents itself, the pin presses out into it, and retains the circle with a moderate force in its proper position, until the raising of the tympan, as before described, overcomes the resistance of the pin, and turns the circle round By this con trivance, the types always arrange themselves into a straight line, after being turned round, without which the impression would have a very disagreeable and irregular appearance. The tympan E, Fig. 1, is composed of two parts : a solid brass plate, against which a few folds of cloth are placed and secured by the second part, which is a brass frame, covered with parchment, and attached to the former by four screws, two of which appear at f f in Fig. 2. The brass plate of the tympan is fastened to the leaf L, Fig. 1, projecting from the axis, by means of six screws. Two of these, only one of Which, h, can be seen in the figure, tend to throw the tympan from the leaf, while the other four, which are arranged one on each side of the two former, draw the tym panum and leaf together. By means of these screws ,thus acting in opposition, the tympan can be adjust ed so as to fall exactly parallel upon the type, and communicate an equal pressure to all parts of the paper, which is held against the tympan by means of a frisket of parchment, stretched on a frame which surrounds the tympan, and is moveable on joints at A-, k, Fig. 2. The frisket is cut through, as repre sented by the shaded parts in Fig. 2, in order to ex pose the paper where it is to receive the impression of the figures, and the N° before the figures. and
also the impression of the date, year, and place. The type for these are formed in and fastened down upon the surface of the brass cover a, the piece containing the day and month being chan ged every day. 111 order to find the proper position which the paper should occupy upon the tympan, two fine pins are fixed to project from it, and are re ceived into holes made in the brass cover: Two dots are printed upon the note from the copper plates, and the pins being put through at these dots, ensures the figures, Sec. coming on their proper places.
The manner of using the machine is as follows : Suppose the back axis put so far on end as to be de tached from all the circles ;—the figure circles ar ranged by hand, so that the blanks are all upper most; and the proper stereotypes put in for the date ; the back axis is then first set, so that its wheels may take the first of the five circles towards the right hand, and, by moving the handle down almost to touch the type, and returning it up again, the pallet moves the wheels H, and turns the two right hand circles, bringing up figure I. The clerk now inks the type with a printer's ball, opens the frisket sheet L, Fig. 2, on its hinges, and places the note. (already printed in the copper-plate press), against the tympan, the proper place being determined by the two pins and the dots printed on the note, as be fore mentioned. He now shuts up the frisket sheet, in order to confine the paper and to keep it clean, except in the places where it is to be printed ; then, by pressing down the handle F, the impression is given ; and on lifting it up again, it moves the circles and brings up figure 2 The note is now removed, a fresh one put in, and so on, the figure always changing every time. During this operation, the two right hand circles act as units, and advance one each time. When 9 are printed in this manner, and 0 comes up, the handle is moved twice succes sively without printing, which brings up a blank and then 1. The back axis is moved, to act upon the second circle from the right hand, which now be comes the units, the first circles representing tens; by moving the handle a, without printing, figure 1 in the second circle comes up, making 11, the next time 12, and so on to 19. The first circle is now put forwards by hand, bringing up 2 and 0, on the second 20, then moving the handle to pass the blank, produces 21, 22, Sc. to SO, when the first circle is again advanced, bringing up 4 ; in this manner the bu siness proceeds to 99. The back axis is now shifted to the third circle, which becomes units, the second tens, and the first hundreds ; the 0 and blanks of which are advanced to bring up 1 0 ; is brought up in the second ; and the machine itself brings up 0 in the third ; after printing this, it changes to 101. The process now continues through the successive hun dreds in the same manner as before, till 999. The back axis is now shifted to the fourth circle, and the three first must be advanced by hand when they re quire it. At 9999 the back 'tx:s is shifted to the fifth circle, and will serve to 999,999, beyond which it is not required to print.