298 



FORMS OF TEST PIECES. 



From these figures it will be seen that the results are 

 much more uniform when the local contraction is left out. 

 This bar was more than usually variable. 



Mr. Wicksteed has investigated this point, by making 

 experiments on four sets of bars, three all alike in each 

 set. All the bars were of the same diameter but of diffe- 

 rent lengths. Mr. Wicksteed, in dividing the local 

 extension from the extension on the parallel portion of the 

 test piece, calls these two respectively, the strictional 

 and the proportional extensions. The following are the 

 results of the tests.* 



From an inspection of these results, it will be clear 

 that the extension measured on the whole length of 

 a bar, of whatever ratio of length to diameter, is 

 very misleading in the case of bars of ductile materials 

 giving local contraction. When the elongation was 

 measured on the total length, the percentage elongation 

 varied from 50 to 30 per cent., being, of course, 

 greatest in the short bars. The proportional or 

 parallel extension, when the local effect has been 

 eliminated, is sensibly uniform for all bars. Apart from 

 the dimensions of the test piece, there must of necessity 

 be small differences in the extension due to want of 

 uniformity of the metal itself. 



It would, therefore, appear to be reasonable, for pur- 

 poses of comparison, when the test pieces are of various 

 sizes and dimensions and proportions, to quote in the 

 report the proportional as well as the general extension of 

 the bar ; and, in addition, the percentage of reduction of 



* Industries, 1890, II., p. 299. 



