To make the pistons move sufficiently steam-tight in these early imperfect cylinders, Watt contrived a packing formed of a collar of hemp, or tow, as represented in _fig._ 27. The bottom of the piston was formed of a circular plate of a diameter nearly, but not altogether equal to the interior diameter of the cylinder. The part of the piston above this was considerably less in diameter, so that the piston was surrounded by a circular groove or channel two inches wide, into which hemp or soft rope, called _gasket_, was run, so as to form the packing. The top of the piston was placed over this, having a rim or projecting part, which entered the circular groove and pressed upon the packing, the cover being pressed downwards by screws passing through the piston. The lower part of the groove round the piston was rounded with a curve, so that the pressure on the packing might force the latter against the inner surface of the cylinder. This packing was kept supplied with melted tallow, as already described, from the funnel, screwed into the top of the cylinder. The metallic edges of the piston were by this means prevented from coming into contact with the surface of the cylinder, which was only pressed upon by the stuffing or packing projecting beyond these. (80.) Improved methods of boring soon, however, relieved [Pg149] the engine from a part of these imperfections, and Watt writes to Mr. Smeaton in the letter above quoted as follows:— "Mr. Wilkinson has improved the art of boring cylinders; so that I promise, upon a 72 inch cylinder, being not further distant from absolute truth than the thickness of a thin sixpence in the worst part. I am labouring to improve the regulators; my scheme is to make them acute conical valves, shut by a weight, and opened by the force of the steam. They bid fair for success, and will be tried in a few days." The person here alluded to was Mr. John Wilkinson, of Bersham near Chester, who, about the year 1775, contrived a new machine for accurately boring the insides of cylinders. The cylinder being first obtained from the foundery with a surface as accurate as the process of casting would admit, had its inner surface reduced to still greater accuracy by this machine, which consists of a straight central bar extended along the axis of the cylinder, which was made to revolve slowly round it. During the operation of boring, the borer or cutter was fitted to slide along this bar, which being perfectly straight, served as a sort of ruler to guide the borer or cutter in its progress through the cylinder. In this manner the interior surface of the cylinder was rendered not only true and straight in its longitudinal direction, but also perfectly circular in its cross section.