Mr. Kent has devised a modification of Dr. Springer's system, which is shown in Fig. 3. It is applied in those varieties of the torsion balance in which there are two parallel beams, connected by either four or six wires. The wire, F, carrying the secondary beam, E, and poise, H, instead of being carried on an independent support, rigidly attached to the base, as above described, is attached directly to a moving part of the balance itself, and preferably to the two beams. In Fig. 3, T T T are trusses over which are tightly stretched the wires, B B B. A A' are two beams rigidly clamped to the wires; _t_ is another truss with stretched wire, F F¹. The upper wire, F', is attached by means of a flexible spring and standard, S, to the upper beam, and the lower wire is attached either directly or through a standard to the lower beam. The secondary poise, H, is rigidly attached to the truss, _t_. The secondary beam, E, is also rigidly attached to the truss, and acts as a multiplying beam. The secondary structure thus completely fills two functions: First, that of multiplying the angle of rotation and thereby increasing the apparent sensitiveness of the scale, and, second, that of overcoming the effect of change of level. The secondary beam may be dispensed with if a multiplier is not needed, and the secondary truss, _t_, with its standard and counterpoise, H, used alone to counteract the effect of change of level. Fig. 5 shows a modification of this extremely ingenious arrangement.--_Engineering_. * * * * * LINK BELTING. [Footnote: From a paper read before the "Technischen Verein" of New York, May 28, 1887.] By CHAS. A. SCHIEREN. The old saying that "there is nothing new under the sun" may well be applied to leather link belting. It is generally believed that these belts are of recent invention, but that is an error. They are over thirty years old. Mr. C.M. Roullier, of Paris, experimented that long ago with small leather links one and one-half inches long by three-quarters of an inch wide. These links had two small holes at equal distances apart, and were joined with iron bolts, which were riveted at the ends, thus making a perfectly flat surface, and in that way forming a belt entirely of leather links. Mr. Roullier's idea was to economize; he therefore utilized the material left over from the manufacture of flat belting. He perfected his belt and came to this country in 1862, when he patented the article here and tried to introduce it. At first it produced quite a sensation, and many tests were made, but it was soon found that Roullier's belts were not suited to running our swift motion machinery, and they were therefore abandoned as impracticable. Mr. Roullier then introduced his invention into England, where he met with some success, as his belt was better suited to English slow motion machinery.