508 Proceedings of the Royal Society of Edinburgh. [Sess. 
Finally the specimen was submitted to a process of reversals at 105° C. 
and retested at that temperature, when the following results were obtained : — 
Magnetising current in amperes . 
0*2 
0-4 
0*6 
0-8 
1-0 
1-5 
Magnetometer deflection 
38 
79 
127 
183 
240 
397 
It will be seen that a magnetisation curve yielded by the specimen at 
105° C. following upon the application of the process of reversals at 105° C. 
lies everywhere below the curve yielded by the specimen at 105° C. follow- 
ing upon the application of the process of reversals at 15° C. The former 
curve is the true magnetisation curve of the material corresponding to 
105° C. If the latter curve is taken, the errors introduced into the results 
are very considerable. 
If the temperature of a test-specimen is changed and a test carried out 
without first applying the process of reversals at the new temperature, the 
magnetisation curve obtained depends not only on the new temperature 
but on the thermal change effected. The magnetisation curve yielded by a 
test following upon a process of reversals carried out at the new tempera- 
ture depends only on the temperature of the specimen. 
In view of the above facts it seemed of importance to repeat Hopkinson’s 
work on steels. Accordingly a series of specimens containing varying 
amounts of carbon were obtained from Messrs Armstrong, Whitworth & 
Co. Ltd. These steels form a suitably graded series, extending from mild 
steel, through medium carbon steel, to high carbon steel and cast iron. 
The following table gives their compositions : — 
Table I. 
Compositions of Steels tested. 
Description of Material. 
Percentage of Composition. 
C. 
Mu. 
Si. 
P. 
S. 
Cast iron 
3*15 
0T5 
0T3 
High carbon steel 
1-64 
0T3 
0-85 
0*02 
0*02 
Medium carbon steel . 
0-80 
0-20 
0*075 
0*012 
0*02 
Low carbon steel . 
0-30 
0-60 
Trace 
0*025 
0*03 
Soft iron 
0-06 
0T0 
The specimens were supplied in the form of cylinders 20 cms. in length 
and 0’9 cms. in diameter. Previous to being tested each specimen was 
carefully annealed from 900° C. in a Fletcher gas furnace, care being taken 
