While on this subject I may as well dispose of the connecting-rod, although the other changes were made subsequently, and I do not recollect exactly when. The following shows the rod and strap as they have been made for a long time. The taper of the rod, giving to it a great strength at the crank-pin neck to resist the transverse fling, was, I presume, copied by me from a locomotive rod. The rounded end of the strap originated in this way. I had often heard of the tendency of the cross-head-pin straps to spread. This was in the old days, when these pins were not hardened, indeed were always part of the iron casting. The brasses, always used without babbitt lining, would wear these pins on the opposite acting sides only. Brass, I learned afterwards, will wear away any pin, even hardened steel, and not be worn itself. When this wear would be taken up, the brasses would bind at the ends of their vibration, coming in contact there with the unworn sides of the pin. To relieve this binding it was common for engineers to file these sides away. All I knew at that time was that the straps would yield and spread. It occurred to me to observe this deflection in a spring brass wire bent to the form of a strap. The pressure being applied on the line of the pin center, the deflection appeared to take place mostly at the back, and so I stiffened it. Since the introduction of the flats on the pin, which prevent the exertion of any force to spread the strap, this form seems to be rather ornamental than useful. [Illustration: Connecting-rod and Strap.] To this strap I added a wiper for lubricating the cross-head pin automatically. The drop of oil hung from the center of a convex surface provided above the wiper. The latter was inclined forward, and its edge partook of the vibration of the connecting-rod. On the backward stroke this edge cleared the drop. At the commencement of the forward stroke it rose to take it off. A note of the change then made by me in stop-valves will conclude the record of these changes. The valve and its seat had always been made of brass. The latter was fitted in a cast-iron chamber, and, expanding more than the iron, was apt to work loose. I disused brass entirely, employing a cast-iron valve in the cast-iron seat. These always remained perfectly tight, showing the additional cost and trouble of brass to be unnecessary.