halves of the engine-shaft will always have the same angular velocity ; but in case the pressure on the teeth of the two rack-pinions on one side of the drum should fall below that on those of the two rack-pinions on the other side, the spur wheel will move slightly on the shaft until the rack-pinions receive equal pressure again. By this apparatus equal pressure on the teeth of rack and pinions is at all times insured. The author was convinced of the necessity for such a device by watching it when the span was being turned; for several times during each quarter rotation the little pinions on the spur-wheel would make a sudden movement of such magnitude as to indicate a considerable variation in the spacing of the rack-teeth.

In designing draw-spans with high towers, especially long, double-track ones, there is an important matter that is sometimes overlooked, viz., the tendency of the end of the unloaded arm to rise when a moving load is on the other arm. For single-track bridges the only harm that this would do would be to pound the end bearings; but for a double-track bridge it would certainly some time cause a serious disaster by the derailment of an oncoming train when the other track on the other arm is covered by another train. Before designing the 520-ft. draw-span for the East Omaha bridge, the author looked up this matter as well as he could, having heard of trouble being experienced from rising ends on a double-track draw-span but little shorter than the one then contemplated. The results of the investigation were rather contradictory, so the design was made with three features that were conducive to resist the raising of the ends, viz., extra-deep trusses at both inner and outer hips; stiff, continuously riveted top chords between these points; and an end-lifting apparatus capable of raising the ends one and a half inches. This was the best at that time which the author could do to avoid the difficulty; but at the same time he figured upon using later a holding-down apparatus in case the necessity for same should ever arise. As explained in Chapter XII,[p. 135-136]  this  span  has  at  present  only  a single track at the middle, and the highway cantilevered floors   are   not   yet   put  on.   Observation  proves  that,  with  one