During the latter part of the exhibition I learned that the McNaught and the Hopkinson indicators were in common use in England; that one or both of these were to be found in the engine-rooms of most mills and manufacturing establishments, and that if the Richards indicator were properly put on the market there would probably be some demand for it, although at existing engine speeds the indicators in use appeared to be satisfactory. A special field for its employment would doubtless be found, however, in indicating locomotives. I felt sufficiently encouraged to set about the task of standardizing the indicator, and during the winter of 1862-3 made a contract with the firm of Elliott Brothers, the well-known manufacturers of philosophical apparatus and engineering and drawing instruments, to manufacture them according to my plans. This was my first attempt to organize the manufacture of an instrument of any kind, and I set about it under a deep sense of responsibility for the production of an indicator that should command the confidence of engineers in its invariable truth. I found that the opportunity I had enjoyed for studying the subject had been most important. The daily use of the indicator which I had brought to the exhibition was an invaluable preparation for this work. I decided, first, to increase the multiplication of the piston motion, by means of the lever, from three times to four times, thus reducing by one quarter the movement of the piston required to give the same vertical movement to the pencil, and, second, to increase the cylinder area from one quarter to one half of a square inch. The latter was necessary in order to afford sufficient room for springs of proper size, and correct reliable strength in their connections. The first problem that presented itself was how to produce cylinders of the exact diameter required, .7979 of an inch, and to make an error in this dimension impossible. This problem I solved in the following manner: At my request Elliott Brothers obtained from the Whitworth Company a hardened steel mandrel about 20 inches in length, ground parallel to this exact size and certified by them. Brass tubes of slightly larger size and carefully cleaned were drawn down on this mandrel. These when pressed off presented a perfect surface and needed only to be sawed up in lengths of about 2 inches for each cylinder. Through the whole history of the manufacture that removed all trouble or concern on this account. The pistons were made as light as possible, and were turned to a gauge that permitted them to leak a little. The windage was not sufficient to affect their accuracy; a thickness of silk paper on one side would hold the pistons tight; but they had a frictionless action, and the cover of the spring case having two holes opening to the atmosphere, there could be no pressure above the piston except that of the atmosphere. [Illustration: