The first opt-ration in the use of this form of comparator is to level the main base it (Pig. 9), tIten sliding the micruscope-plate D from end to end of the steel tubular guides, having the microscope adjusted so as to be in focus upon the surface of mercury contained in a shallow trough, over which the microscope passes, the curvature due to flexure of the guides is determined, and may be compensated for by counter-weights at the neutral points of sup port, a and n'. In order to test this right-line path of the microscope-plate horizontally, the method of the 'stops' is employed, or, another method, which is that of tracing a fine line the entire length of a standard bar upon its upper surface, and, reversing the liar, tracing another line very near the first, and at an equal distance apart at each end ; then, if this distance is uniform between the two lines the entire length, it is safe to assume that the path of the plate is a straight line horizontally, and at the middle the amount of curvature, if any, and if uniform, is readily determined. This method is used by Prof. Rogers with complete success. The 'stop method' is to compare a line-measure or an end-measure bar. on each side of the center line of motion of the microscope-plate, using one microscope, and comparing this fixed length with the constant quantity before referred to, which is the distance between the stops. Should the path be a curved one, the distance between the defining lines upon the bar will appear greater on one side than on the other, in proportion to the amount of curvature exist ing,. The length of the standard, being the length of chords of circles of different radii, seems, by comparison with the stops. to be different in length at each position, caused by the different distances from the center of curvature—about 18 in. in this instance—over which the microscope passes when placed in these two positions. By means of the proportion of similar triangles thus formed, the length of the radii may be very accurately determined. By placing different standards on one side of the line of the stops, they may be, by being com pared with a constant quantity, compared also with each other.
"Another method for comparing two or more standards is to place two microscopes, one on each of two microscope-plates, upon the guides, at a distance determined by the length of one of the standards, and by replacing this one by a second, the coincidence of the lines in the eye-piece micrometer, or their variation, showing at once their relation. The microscopes
may be placed horizontally in this same fixed relation, using the method invented by J. Homer Lane, and which has been successfully used in the office of the United States Coast Survey at Washington.
The subdivision of these standards of length, which is effected by the use of this same process—the microscope-plate sliding between fixed stops. This is accomplished in the fol lowing way : A yard, for instance, is to be subdivided into three equal parts, or into three separate feet. We divide the whole length by trial into three parts, then, by setting the stops so that the microscope-plate may move very nearly the distance represented by-the first one of the three parts, by readings of the eye-piece micrometer carefully taken at each end of the path of motion of the microscope, and using the finely ruled lines by which these three parts are defined, we obtain the length of this subdivision as compared with our constant quantity ; then, by sliding or moving the bar along under the microscope until the second part is in place, the same operation is again performed, and so for the third, thus determining the relation of each with respect to this temporary or arbitrary standard ; then, by adding the differences between these separate parts and the constant length, and taking the mean or average of these differences, from which we subtract each difference, gives us the correction to be applied to each part in order that it shall be exactly one third the total length, or, as in case of a yard bar, giving us exactly 12 in., or the standard foot. The foot may then be subdivided in the same manner into 12 equal parts, establishing the standard inch, and, further, to th, or even to TATEr of an in." (See Trans. Am. In.q. Mech. Engrs., vol. iv.. 1882.) Measuring-Machine: see Leather -Working Machines and _Measuring Instruments, Mechanical.
Micrometer: see Measuring Instruments, 'Mechanical.
Middlerg's Pumper : see Milling Machines, Grain.
Kill. Grain : see Milling-Maehines, Grain.