462 Scientific Proceedings, Royal Dublin Society. 



The specific heat of two forged bars of manganese steel, marked 

 6 A and 552, containing respectively 9*4 and 13*7 °/ Q of manganese 

 I determined by the method of mixtures. Both were found to 

 have a higher specific heat than iron, the 9^- per cent, specimen 

 being 0*1238 and the \Z\ per cent, specimen being 0*1264 — iron 

 having a specific heat of 0*114. Since these experiments were 

 made I have learned that Mr. Crichton Mitchell has determined 

 with great care both the specific heat and thermal conductivity of 

 manganese steel. 1 The specific heat of a 10 per cent, specimen 

 of manganese steel Mr. Mitchell finds, by the method of cooling, 

 to be 0*124 (1 + *0014 t), being 1*087 times that of iron ; this 

 closely agrees with my own result. The thermometric conduc- 

 tivity of the same specimen at 100° 0. Mr. Mitchell finds to be 

 .00272, whilst iron at the same temperature is *01274, so that 

 the presence of 10 per cent, of manganese in iron or steel lowers 

 its thermal conductivity one-fifth. 



The experiments about to be described were made on samples 

 containing from 1*10 to 21*69 per cent, of manganese. Some of 

 the specimens were in the form of ingots just as they were broken 

 after casting, others were test bars that had been forged and 

 then tested for breaking strain. The ingots were nearly cubical 

 blocks, on an average measuring three inches along the edge ; the 

 test bars were cylindrical rods, some \ inch in diameter, and from 

 6 to 9 inches long ; the ends had been enlarged for holding them 

 in the jaws of the testing machine, and as all the bars had been 

 broken in testing, the experiments were made on a broken half of 

 each bar. 



The method of experiment was as follows : — A magnetising 

 helix was constructed of sufficiently large diameter to contain the 

 ingots, and about twice their length, and another longer helix was 

 employed for the forged bars. Through the helix a powerful 

 current from a storage battery was passed for a definite number 

 of seconds, the strength of the current being measured by a 

 Thomson's graded galvanometer. The current was repeatedly 

 flashed on and off — a process that facilitates magnetisation — then, 

 interrupting the current, the sample was at once carried to a 

 magnetometer in an adjoining room and there tested. 



1 Trans. Royal Soc. Edin., vol. xxxv., part 4, 1890. 



