392 PROCEEDINGS OF THE AMERICAN ACADEMY 



recoil, which was obviously near the greatest recoil the machine 

 will ever suffer, a horsehair was next tested. It was -ti/oo ^^ ^" 

 inch in diameter, it stretched thirty per cent, and broke at one 

 pound. 



" 3. Specimens were subjected to 1,000,000 pounds compression. 



"4. Delicate structures, such as eggs and nuts, were tested in com- 

 pression. 



" The results of these and of many other proof experiments demon- 

 strate the efficiency of this testmg machine. Its action as a whole does 

 not end its usefulness, for its separate parts may be adapted to other 

 modes of testing. It is evident, for example, that the bed and plat- 

 form, with the four supporting chambers, could be removed and built 

 in as one of the arch stones in a great arch, where the pressure at that 

 point would be indicated by the scale beam , and by a slight modifica- 

 tion of the connections, there might be shown the position of the 

 resultant line of pressure under either a still or a moving load. Were 

 the same parts buried in the rear of a retaining wall, they would 

 measure the thrust ; and the effect of that thrust would be shown if 

 they were built into the lower course of that wall. 



" The gauges in this machine which measure the pressure on the 

 specimen holders, and that in the straining press, constitute in them- 

 selves a very promising form of steam gauge. As they stand, they 

 are capable of indicating from one pound to the square inch to 3, GOO 

 pounds, without straining any part beyond the safe limit of elasticity. 

 The need of an accurate steam gauge which will not degenerate is 

 illustrated by the fact that the United States Board appointed to study 

 the causes of the bursting of steam boilers reported that its results were 

 entirely unreliable, because uo steam gauge could be found on which 

 dependence could be placed. 



" It only remains to indicate in what way and to what degree the 

 testing machine is conducive to human welfare. 



" It lessens the risk of life and the cost of construction, by condemn- 

 ing every dangerous part and exposing each excess of material. Struct- 

 ures may have various faults : (1) They may be too weak, and there- 

 fore liable to give way at all points. (2) They may be strong enough 

 in some parts but weak in others, where they are ready to break. (3) 

 They may be everywhere too strong, in which case the weight of useless 

 material must be subtracted from the load they ought to bear. In 

 the first instance, the structure is dangerous and too cheap ; in the 

 second, it is dangerous and in certain places too cheap ; in the third, it 



