MECHANICS AND USEFUL ARTS. 43 



point varying from nothing to about half-stroke. The valve is lifted by the 

 machinery in proper time to commence the stroke, and is held suspended by 

 a coiled spring until the rush of steam is sufficient of itself to shut it. It is 

 easy to see, from the nature of the crank motion, that the piston, at the com- 

 mencement of the stroke, moves very slowly, and as the crank in its onward 

 motion successively assumes positions in which the piston moves faster, the 

 rush of steam through the opening is proportionally increased. The valve 

 may be described as a thin ring, which, by the aid of a circular fixture held 

 \vithin it a little above its seat, completely stops the passage when down, but 

 allows the steam to flow under its lower edge whenever it is lifted. At 

 whatever point in the stroke the motion of the piston becomes sufficiently 

 rapid to produce a strong rush, or a slight "wiredrawing" of the steam in its 

 passage through the orifice at that point, the action upon the upper edge of 

 this ring will overcome the resistance of the spring, and the valve will be 

 "sucked" down and closed tight. In practice the valve consists of two rings, 

 one placed concentrically within the other ; a fixed ring is fitted between 

 them (supported by suitable projections from the seat below), and the opening 

 for the passage of steam is consequently an annular rather than a circular one. 

 Several devices are connected with this to render its action more perfect and 

 quiet, one of which is a piston attached to the valve-stem, which piston works 

 loosely in a small cylinder above, to prevent it from rising too rapidly or too 

 far. We describe this as a late invention and a decided novelty, although there 

 appears to exist an inherent and radical defect which prevents its regulating 

 perfectly, although it may possibly work as nearly so, as the governor balls 

 ordinarily do. The fault is that as the rush of steam is always proportional 

 to the motion of the piston, the valve always shuts when the piston has 

 attained a certain velocity ; when, therefore, by reason of diminished work or 

 an increase of boiler pressure, the valve is dropped at an earlier point in the 

 stroke, it is proved that the piston must have acquired a high velocity too 

 early in the stroke, and that consequently its mean velocity must be very 

 considerably increased, and the engine makes more revolutions in a given 

 time. It is an evil of the same kind as attends the workMg of a conical pen- 

 dulum or ordinary governor, but the economy of regulating by the " cut-off " 

 rather than by a throttle-valve is sufficiently great to invite attention to this 

 method, which is also much recommended by its simplicity and cheapness. 



Clark's Patent Boiler Feeder and Indicator. This invention consists of a 

 short horizontal metallic tube of say 3 feet in length and 2 inches in diameter, 

 suitably attached to the outside of the boiler, or to a wall near by. The 

 height at which the tube is placed should be the same as that at which it ia 

 desired to maintain the water-level in the boiler. One end of the tube com- 

 municates with the upper or steam part of the boiler, the other end with the 

 water part ; when the water in the boiler is at the* proper level the tube will 

 be one half filled with water and one half with steam. A small cold water 

 pipe passes lengthwise through the tube; one end of this water pipe ia 

 plugged tight, the other end is furnished with a metallic cup covered with 

 rubber, forming a diaphragm. On this diaphragm rests a plunger rod 

 attached to a lever, the latter connected with the pump throttle. "When the 



