AMERICAN INSTITUTE. 611 



Tlie notch of the link admits steam during .43 of the stroke, 

 and expands it to 1.97 of its initial volume, and opens the port 

 three-eighths of an inch, or .03 of its full width. If I open my 

 steam port to half its full width I cut off at exactly half stroke, 

 and expand to double the initial volume, and get an effect slightly 

 greater than the link in this notch, besides the advantage of a more 

 sudden and full opening, and a cut-off that avoids wire-drawing. 



The 8th notch admits steam during .36 of the stroke, the open- 

 ing being five-sixteenths, or a quarter of the port, and the release 

 at 17| inches in a 22 inch stroke; the steam is therefore expanded 

 to 2.27 times its initial volume, and the w^ork done during ex- 

 pansion is .3.2 of the initial w^ork; making the total work 1.82, 

 which is 7 per cent more than the effect of expanding to just 

 double the initial volume. But when we allow for the wire-drawl- 

 ing of this throttled admission it will appear probable that this 

 advantage is more than balanced by it. 



There is no notch for a shorter admission; showing that Rogers 

 did not deem it useful to provide for an earlier cut-off by the 

 valve. But of course the throttle, or regulating valve, wall pro- 

 duce a partial expansion, if there be separate passages for steam 

 to the two steam chests. This effect will be greater in the case of 

 a port that is widely opened and closed suddenly, than in the case 

 of a narrow and gradual ojoening and closing. 



Wherever this notch is used, the work is light; w^e may there- 

 fore, in the half-stroke movement contract the throttle; in which 

 case the pressure in the steam chest will fluctuate; being ad- 

 mitted all the time, and let out only half the time, it will be 

 greatest a little after the beginning of the stroke, and least at the 

 moment of suppression; and the dimension of pressure between 

 these points is the effect of expansion, and the piston receives that 

 effect. The piston starts under a relatively high pressure, with a 

 full opening, and no wire-drawing at the port; but the narrow 

 opening of the link so wire-draws the steam that the pressure 

 upon the piston is very much less than that in the steam chest, 

 as the indicator diagrams show; and this difference is so great, 

 and so clearly attributable to the narrow opening, that I confi- 

 dently expect a more effective expansion wath this hnlf stroke 

 cut-off than is obtained by the link in the shortest admission 

 which was deemed useful in this instance of it. 



Let us now consider the necessity of an admission later than 

 half-stroke. I have often heai'd engineers say that they want 

 full stroke when they ascend a heavy grade, or start a train; and 

 I do not remember to have met with one who w^as not of that 



