designs of and by processes invented by Prof, Rowland, who was constantly at hand to direct every movement. This machine is in a dark vault under the laboratory. When a "grating" is being made, it runs night and day. The vault is locked, and no one is allowed to enter it, for the machine is so sensitive that the temperature of a human body would disarrange it. When a new diamond point is being tested, as is now the case. Prof. Rowland will permit a few people to visit it. Sir William Thomson, the Earl of Rosse, Lord Rayleigh, Prof. Ball, Astronomer Royal of Ireland, the late Prof. Helmholtz, of Berlin, Prof. Mascart, of Paris, and Prof. Lemström, of Sweden, are among those to whom this courtesy has been extended. The motive power of the machine is a hydraulic engine. The water is kept at a constant height in a tank near the roof, to insure unvarying speed. It is driven by a belt attached to a solid brass driving wheel on the machine. A crank is turned by the same on the other end of the shaft. This crank moves the carriage that conveys the diamond point back and forth over the surface of the "grating" or plate. This carriage rests on two steel ways, which are flat on top and slanting slightly outward, so that there are three points on one way or rail on which the carriage rests. These "ways" are ground so as to make them as nearly accurate as possible. But they can not be made perfect, for Mr. Rowland tested them with a microscope and found that they were "out"—that is, not exactly perfect—by one fifty-thousandth of an inch. He did not attempt to improve them.
One of the most difficult problems that Prof. Rowland and Mr. Schneider have to solve is to find a diamond point that is exactly right. Some are too blunt, some have one defect, some another, and it generally takes from two to eight months to find an available diamond.
As the diamond carriage moves exactly in the same line backward and forward every time, the metal plate or grating beneath must move slightly when the diamond makes a stroke. These tiny grooves must be exactly the same distance apart, and as there must be from ten thousand to forty-eight thousand parallel grooves or lines made within the space of one inch, it is readily seen that the lateral movement of the metal plate is very small. At every stroke of the diamond, the carriage carrying the plate is moved by means of a steel screw. It is the only absolutely exact screw ever made. The "ways" mentioned above, when tested by the microscope, are one fifty thousandth part of an inch "out" of the exact, but the strongest microscope can find no flaw in the exactness of the screw. In order to manufacture this screw, it was necessary to make it under water, which was kept at a certain temperature. If it had been made in the air, or the tempera-
