Notices respecting New Books, 225 



ends of rectangular bars which slide in right-angled V-shaped 

 grooves. The measuring-bar in the one head-stoclc is moved by a 

 screw with a graduated head, one division of which corresponds to an 

 advance of the bar through l-5000th of an inch ; the other is moved 

 by a screw of 20 threads to an inch, on whose head are cut 200 

 teeth, which work with a second screw, the head of which is divided 

 into 250 equal parts ; so that one division of the head of this 

 second screw corresponds to an advance of the measuring-plane 

 through a distance of one millionth of an inch. It is not, however, 

 safe to rely on the unaided sense of touch for judging of so miuute 

 a distance ; the pressure of the measuring-plane against the end of 

 the object under measurement is regulated by its sufficiency to 

 support the weight of a small metal disk, interposed between the 

 object and the plane ; it is found that this disk (or gravity-piece) 

 will fall between them when their distance has a certain amount, 

 and ^^'ill be supported when that distance is diminished by a mil- 

 lionth of an inch. 



It is plain, from the above description, that the construction of 

 these machines depends upon an extreme accuracy in the work- 

 manship ; the method would manifestly fail unless the measuring 

 planes were truly plane and accurately at right angles to the axes of 

 the sliding bars. To Sir Joseph Whitworth belongs the honour of 

 devisnig the means of obtaining the needful extreme accuracy of 

 workmanship, as well as the modes of applying it to the construction 

 of these measuring-machines. If two pieces of metal are machine- 

 planed, and one of them is smeared with a film of red ochre and 

 oil, on placing it carefully on the other, the bearing-points become 

 apparent, and can be reduced by scraping, until a contact can be 

 attained which is sensibly at all points of the surface ; this process 

 of scraping, as distinguished from polishing, admits of being per- 

 formed with extreme exactness. ]Now, if three pieces (A, B, and 

 C) are taken, and the process of scraping performed on them till 

 A coincides with B, B with C, and C with A, the surfaces of all 

 three must be planes ; and when the operation has been performed 

 with all the exactness of w^hich it is capable, they are said to be 

 true planes. Strictly speaking, each of the pieces has an indefinitely 

 great number of evenly distributed bearing-points, which lie on a 

 surface differing from a true plane, by an amount insensible to any 

 test that has hitherto been applied. 



These " true planes " once formed, there is comparatively little 

 difficulty in forming others ; and numerous applications can be 

 made of them, e. g. the formation of the sliding bars and Y-shaped 

 grooves in the measuring-machine. Take three nearly rectangular 

 bars (a, h, c) ; two of the opposite faces of each can be scraped until 

 they coincide accurately in all positions with one true plane on which 

 they rest, and a second true plane placed on them. Each bar no\T 

 has two parallel faces, and in all three the distances of these faces is 

 the same. This being done, let a third face of each (viz. x of a, y of 6, 

 and zoi c) be wrought till they stand the following test: — The bars 

 resting on one of their parallel faces on a true plane, oc can be brought 

 into perfect contact with z and likewise with y;h is, now turned over 



Phil. Mag. S. 5. Vol. 4. No. 24. Sept. 1877. Q 



