396 KINEMATICS OF MACHINERY.
that the revolutes of which d consists are represented respectively by the double cone 1 of a (having A A for its axis) and the plane- pair forming the sides of d itself (and forming a section of a revolute of which the axis is perpendicular to the plane of the paper).
The machine of Kiister, PL XXXI. Fig. 1, known to me only by description, is very similar to that of Duclos. Here the chamber, instead of being made a double cone with spheric sides, is a globoid or cylindric ring, of which the piston d is a sector. I have no details of the packing used where the piston d passes through the slot in the disc I, but it appears to be very defective.
Fig. 2, PI. XXXI. Wood's rotary steam-engine.* This machine is very nearly related to that of Duclos, but here the cross-block c receives a more important position in the train than that of a mere packing-piece. Its formation from two revolutes 3 and 4 crossing each other at right angles is obvious at once. The disc 6 has a slot profiled as in Fig. 281 for the motions of the piston d. The chamber a has for one of its revolutes the double cone upon the axis A A, for the other the bearings 2 of the driving- shaft I. I must mention that I have somewhat altered Wood's drawing. Instead of the cross-arm 4 this shows the two convergent arms 4' as in Fig. 3. With such an arrangement, however, the mechanism cannot move, for the piston, if it were constructed as shown at c, could not revolve in the chamber. For by the spindle 3, which Wood distinctly showed to be rigid, c is compelled to remain always in the plane of the axis 2, so that the breadth of the piston must vary periodically between the real width of the chamber and the width shown in the drawing. From the existence of this error, and others which may be discovered by a closer examination of the original drawing, it is evident that this machine, in spite of its representation and treatment by Bataille, had never actually been at work.
Our last illustration, Fig. 5, PI. XXXI., shows one of the latest productions of the inventive spirit which is kept alive by this rotary machine problem. It is the rotary engine of Geiss,f who has it at work at Gebweiler. The chamber a is hemispherical. The link 6, the continuation of which serves as a fly-wheel shaft,
- Bataille et Jullien, Machines a Vapeur, 1847, vol. i., p. 447.
t Propagation Industrielle, vol. v., 1870, p. 132.
