114 KENNEDY EXTEXSOMETE1S. 



one another, and the horizontal bases of the triangular 

 frames slide past one another to an extent equal to the 

 stretch of the bar between the clips B', B". To one of these 

 bases is attached a small plate G, which is capable of l>ein^- 

 made to slide along the bar, its movement being con- 

 trolled by a screw adjustment. On the top of this 

 plate is a small cupped disc. A similar cupped disc 

 is carried by the other base, to which it is attached 

 by means of a light plate spring. The cups are 

 arranged so as to be about a quarter of an inch apart, they 

 are on the same level, and their centres lie on a straight 

 line at right angles to the direction of the test bar, and 

 about half way along. These cups move with respect to 

 one another a distance equal to the stretch of the specimen, 

 and this distance is magnified a hundred-fold by the lever 

 M. Corresponding to the two cup discs, there are fixed 

 into this lever two hardened steel points, each of which 

 rests in one of the hollows ; the distance between these 

 two points forms the short arm of the magnifying lever ; 

 the long arm is 25 in. in length, so that the ratio of the 

 arms is twenty-five to one-quarter, or one-hundred to one. 

 The outer extremity of the lever terminates in a fine point 

 which moves over a graduated scale carried by the rod N, 

 which is attached to the clip B". The long arm of the lever 

 has its weight balanced by a counter-weight attached to a pro- 

 longation of the short arm beyond the suspending points. 

 The scale is graduated into inches and tenths. These tenths 

 of an inch each represent, in stretch of the specimen, one 

 hundredth of one tenth, or one-thousandth of an inch. It is 

 therefore quite easy to read to thousandths of an inch, and 

 as tenths of the smallest divisions can be estimated by eye, 

 it is possible to read to the nearest ten-thousandth. When 

 calibrating the instrument, the correctness of the measure- 

 ments of the lever arms should not be relied upon, but a 

 direct calibration should be effected by means of a vernier 

 caliper or a measuring machine. When a vernier caliper is 

 used, one clip of the extension gear is fixed to a rigid support. 

 The main bar of the caliper is also fixed to a rigid support, 

 and the second clip made to grip the slide of the vernier. 

 When the apparatus has been fixed in this way, by 

 turning the screw of the vernier caliper its slide is made 

 to move with respect to the bar, and, consequently, one 

 clip of the extensometer relatively to the other one. In 

 this way the pointer of the extensometer is made to move 

 over its scale. At the same time the exact distance moved 



