146 



TKNSIOX TKSTIXU. 



On inspecting these two columns, it will be seen that 

 the second set is the more -reyidar, and probably the more 

 accurate. It will, therefore, be well to neglect the first set 

 and make use of the second. 



We want the mean extension, say, per 1 ton of load. 

 This can be found either, as above, by taking the mean of 

 the 4 ton extensions, and multiplying by 2. Thus, the 

 mean extension per ton of load is 



0-0005 x 2 = 0-0010 in. ; 



or, better still, the method advocated by Dr. Kennedy can 

 be used. In this the increments are reckoned on greater 

 differences of load. In doing this, the greatest possible 

 range should be taken. Thus, the extension at the load 

 half way down the list should be subtracted from that at 

 the last load, and so on, taking the successive pairs up the 

 list. Suppose an interval of 4 tons be taken, the details of 

 the process will be as follows : 



0-0090 0-0085 0-0080. 0'007f> 0'0070 

 0-0050 0-0045 0-0040 0'0036 O'OOSl 



0-0040 



0-0040 



0-0040 



0.0039 



0-0039 



0-0065 

 0-0026 



0-0039 



0-0060 

 0-0021 



0-0039 



Take the mean of these 



0-0055 

 0-0016 



Q-0039 



0-0040 

 40 

 40 

 39 

 39 

 39 

 39 

 39 

 39 



0-0040 



o-ooso 

 0-0011 



0-0039 



0-0045 

 0-0005 



0-0040 



0-00394 in., 



and the extension per ton of load will be one-fourth of 

 this ; that is, 



- 0-000985 



which result is not quite the same as that first obtained. 

 This will be taken as the correct result. 

 Now, the modulus of direct elasticity, 



E = ii 



