1905.] The Effects of Momentary Stresses in Metals. 501 



The general result that I have obtained is that iron and copper wires 

 may be stressed much beyond the static elastic limit and even beyond 

 their static breaking loads without the proportionality of stresses and 

 strains being substantially departed from, provided that the time 

 during which the stress exceeds the elastic limit is of the order of 

 1/1000 second or less. 



The wire was in each case of No. 10 gauge, and about 30 feet long ; 

 it was hung in a vertical chase in a wall, the upper end being firmly 

 fixed in a block of iron, weighing about 20 Ibs., the ends of which were 

 built into the wall. This block carried a vertical steel rod, at any 

 point of which could be clamped the contact-making device for 

 measuring the momentary extension. The construction of this 

 sufficiently appears from the figure. The light hard steel point A is 

 fixed to the wire at a certain distance, usually 20 inches, from the upper 

 end. The wire having been drawn taut preparatory to the experiment, 

 the insulated spring S is pushed up by the micrometer screw until 

 contact is made with the point as shown by the deflection of the 

 galvanometer. The spring is then withdrawn by the amount of 

 extension expected ; the blow is delivered and the galvanometer shows 

 whether contact between the point and the spring has occurred or not. 



By using a sensitive ballistic galvanometer without any resistance 

 in series with it, it was found quite easy to determine the instantaneous 

 extension of 20 inches of wire correct to 1/1000 of an inch, that amount 

 of difference in the position at which the spring is set converting a big 

 throw of the galvanometer into no deflection at all. In a few cases a 

 second point was added with a similar contact spring close to the upper 

 end of the wire, in order that any displacement of the wire relative to 

 its supports might be detected. But I found that if the wire was soft 

 soldered into a bolt about 3 inches long screwed into the block, this 

 precaution was unnecessary. 



The falling weight was a cylindrical piece of steel weighing 1 lb., 

 with a hole drilled along its axis which was a loose fit on the wire. 

 The wire was kept taut by a spring balance attached to its lower end, 

 the tension in which could be varied from 20 to 200 Ibs. The stop 

 struck by the falling weight was a metal sleeve slipped over the wire 

 and soldered to the "lower end. It was made as light as possible so 

 that the velocity of the weight should not be much diminished at 

 impact. The velocity at impact was calculated from the height of fall, 



being taken as - - ^/2^ where M is the mass of the weight, and m 



that of the stop. 



From time to time the permanent extension on the 20 inches was 

 measured by again pushing the spring into contact with the wire when 

 under steady tension. No analyses were made of the materials, 

 because, at the present stage, all that I desire to do is to compare the 



