264 DR E. G. COKER ON 



specimens of iron and steel. These differences may be caused by the hardening effect 

 of the drawing, minute cracks in the wires, want of roundness, and the like. It there- 

 fore appeared probable that experiments on the lines indicated by physicists would be 

 of some service, and it was with this idea that the investigation was commenced. 



The chief difficulty in the accurate investigation of the torsional properties of metal 

 bars lies in the lack of suitable apparatus for the work ; and after reviewing the chief 

 machines available for measuring strain and applying torque — to all of which there 

 seemed some objection — it was resolved to design and construct special appliances for 

 the work. 



Attention was first directed to the design and construction of a self-contained instru- 



O 



ment for measuring strains, which should be sufficiently accurate to measure strains of 

 one second of arc ; and after some experiments an instrument was constructed which 

 satisfied these conditions.* A modification of this was used in the work of this paper, 

 and is described in Section II. 



In most machines for applying torque, the construction is such that the weigh-lever 

 can only be used for torsion in one direction, and the ends of the specimen are fixed, 

 so that it is impossible, for instance, to apply a bending moment and torque, or a 

 tension and torque, together. 



A machine was therefore constructed to allow of torque in either direction, and also 

 permit of the application of a uniform bending moment and a pure torque, to give a 

 combined stress. A separate device was constructed for giving the combined stresses 

 of tension and torsion. 



IT. — Description of the Apparatus. 



1. Instrument for Measuring Strains. 



In making measurements of small strains it is a great advantage to use an instru- 

 ment which will read directly, and which is self-contained and wholly supported on the 

 specimen under test, thereby avoiding external scales, the positions of which with respect 

 to the specimen may be changed by a disturbing element, such as slipping of the grips, 

 applied bending moment, and the like. In order to meet these conditions, an instrument 

 was designed for the purpose of these experiments, and is shown in sectional elevation 

 by fig. 1 , and in side elevation by fig. 2. It consists of a graduated circle A mounted 

 upon a chuck plate B, provided with three centring screws adjustable by hand. Upon 

 the Vernier plate J, an arm carries an extension K, upon which is secured a frame X, 

 carrying a thick wire P. The movement of the wire is observed by a reading micro- 

 scope carried in the sleeve R of an arm S mounted upon the short cylinder C, which 

 latter is gripped upon the test bar by screws L. 



The reading microscope has an eye-piece T provided with a glass scale, and a right- 



* Cokek, " On Instruments for Measuring Small Torsional Strains. 1 ' Phil. Mar/ , December 1899. 



