TRANSACTIONS OF SECTION G. 713 



resistance to certain chemical agents, impermeability to water, solubility in 

 certain liquids, and electrical resistance and dielectric strength. 



The methods of testing employed at the present time consist in chemical 

 analysis and in the determination of the elongation and load at rupture and of the 

 sub-permanent set resulting from a given extension maintained for a given time. 



It is evident that the chemical tests can give no indication with regard to 

 many of the physical properties enumerated above, and it would seem that their 

 true function lies in the determination of causes, while the mechanical tests 

 should deal with the effects produced by these causes in the commercial product. 



The mechanical tests should then form the primary tests of both the manu- 

 facturer and the purchaser, and should, when necessary, be supplemented bv the 

 chemical tests, either as confirmatory tests or for the elucidation of the causes 

 of the defects indicated by the mechanical tests. In the case of raw rubbers the 

 chemical tests give little or no indication of the value of the product for indus- 

 trial purposes, and it is suggested that test pieces for such materials should be 

 prepared from the so-called 'Admiralty mixing,' consisting of 60 per cent, 

 rubber, 3 per cent, sulphur, and 37 per cent, zinc oxide, and be subjected to the 

 hysteresis tests hereinafter described. 



The author has designed a machine in which a specimen of rubber of 

 standard dimensions is loaded at a given rate to a given percentage of its 

 breaking load. The load is then removed at the same rate, and a graphical 

 record is obtained on the chart table of the machine of the extension and 

 retraction curves. 



Rubber possesses very considerable mechanical hysteresis, and a considera- 

 tion of the loop diagram obtained from the machine enables the following 

 physical quantities to be determined for any given test piece : — 



(1) The rate of extension with load. 



(2) The work done in extension. 



(3) The work done by the rubber in retracting. 



(4) The work expended in the rubber itself. 



(5) The sub-permanent set remaining after a given extension. 



The limits of the hysteresis loop may also be set in terms of extension 

 in place of load as already stated. The author finds that for a given rubber the 

 following laws hold good : — 



(a) The load per unit area of the initial cross-sectional area of the test 

 piece is constant for a given extension of the test piece and independent of the 

 cross-sectional area of the specimen within certain limits. 



(b) The work done in extension, in retraction, and in the rubber itself is 

 within certain limits proportional to the cross-sectional area of the test piece, 

 and is directly proportional to the length of the specimen with a given percentage 

 extension. 



On the completion of the first cycle of extension and retraction the specimen 

 may be subjected to a series of similar cycles the limits of which may be set 

 either by a given maximum extension or a given maximum load. 



For high-grade rubbers the areas of the loops for successive cycles become 

 constant after about the sixth loop, when the subsequent cycle loops are taken 

 up to the same maximum load as that for the first loop. 



The author finds that the extension for a given load limit increases with 

 each successive cycle, and that the rate of increase follows a logarithmic la\i 

 from the second cycle onwards. 



Applications of the hysteresis test are given to the determination of various 

 grades of rubber, of the quantity of rubber in a given mixing, of the de°ree of 

 vulcanisation, and of the deterioration due to age or high temperature. 



3. The Utilisation of Solar Radiation, Wind Power, and other Natural 

 Sources of Energy. By Professor Fessenden. 



4, Experimental Investigation of the Strength of Thick Cylinders. 



By G. Cook. 



