When arranged as in Fig. 1, the scale is balanced only when the center of gravity of the structure is vertically above the middle line of the wire, and the support of the scale must be leveled in the direction of the beam, so as to cause the center of gravity to take this normal position.  After the scale is thus leveled, if from any cause whatever, such as shifting the scale on a table, or shifting the table itself, the scale support is thrown out of level, the center of gravity of the poise and beam is shifted from the vertical line above the support, and its moment immediately becomes greater than the torsional resistance, and the beam tips out of balance, and cannot be used as a correct scale until the support is again leveled.

[Illustration:  KENT’S TORSION BALANCE.  Fig 2.]

In spite of all the foregoing facts, it was reserved for the “Encyclopedia Britannica,” in its ninth edition, to use the following as the result of its condensed wisdom: 

“In the torsion balance proper, the wire is stretched out horizontally, and supports a beam so fixed that the wire passes through the center of gravity.  Hence the elasticity of the wire plays the same part as the weight of the beam does in the common balance.  An instrument of this sort was invented by Ritchie, for the measurement of very small weights, and for this purpose it may offer certain advantages; but clearly if it were ever to be used for measuring larger weights, the beam would have to be supported by knife edges and bearing, and in regard to such applications therefore (as in serious gravimetric work), it has no raison d’etre."

[Illustration:  KENT’S TORSION BALANCE.  Fig 3.]

This would seem to settle the whole case, for if the encyclopedia says it has no reason to be, then, like the edict of the Mikado, it is as good as dead, and if that is the case, “Why not say so?” On the contrary, the torsion balance seems very much alive.  But as it is not very generally known, perhaps the early history of this form of balance, briefly sketched, may prove of interest.

One of the first forms of the torsion balance which met the disapproval of the “Encyclopedia Britannica” was attended with the difficulty that the pivoted wires were attached directly to the bifurcated ends of the beam, and could not be tensioned without bending these ends unless the beam was made so heavy as to interfere with its employment in delicate weighing.

[Illustration:  KENT’S TORSION BALANCE.  Fig 4.]

The next step was the substitution of light forms stiffened by the wires being tensioned over them.  This was the invention of Professor Roeder, recently deceased.  The next step was the common counter scale, and then that form of letter scale in which one of the bands acts as a fulcrum and the other as a pivot.

After Professor Roeder’s death, Dr. Alfred Springer, of Cincinnati, continued perfecting this invention, and with marked success—­scales not intended for anything but the weighing of the ordinary articles of a grocery store working so accurately that up to 50 lb. two grains would turn the balance.