TRANSACTIONS OF SECTION G. 827 



material. The results of such investigations tend, further, to place our specifica- 

 tions upon a sounder basis, and at times to remove unnecessary restrictions from 

 manufacturers. 



The principal causes of engineering breakages were classified in three groups 

 — as arising from defects inherent in the material, defects produced during the 

 process of manufacture of the machine or structure, and defects arising during 

 the life of the finished object. Some of the more important cases, such as impure 

 material, metal which has been burnt or overheated, badly forged or annealed, 

 or subjected to improper mechanical treatment, are referred to, and the manner 

 in which these factors can be traced by subsequent investigation are indicated. 

 In illustration of the principles thus laid down the investigation of four typical 

 cases of breakage as carried out by the engineering and metallurgical depart- 

 ments of the National Physical Laboratory were described. In three of these 

 cases, viz., those of a large gun-tube, of a locomotive crank-pin, and of a large 

 shaft, the full text of the Laboratory reports was reproduced, with some addi- 

 tional comments; the fourth case — that of the failure of a boiler-plate — was 

 only referred to as having been already fully published. 



In the case of the large gun-tube, which fissured after an unduly short 

 service, mechanical tests show signs of weakness, and the microscopic evidence 

 leads to the conclusion that the failure was due to the contamination of the steel 

 with an excessive amount of slag enclosures. The boiler-plate just mentioned 

 was found to have failed because it was exposed to unduly severe local deforma- 

 tion during the process of construction. The locomotive crank-pin was found to 

 consist of satisfactory steel which had been case-hardened and subsequently 

 quenched. The microscopic evidence showed that the hardening process employed 

 had been too severe, and had resulted in the formation of a number of hardening 

 cracks, some of which had extended through the entire mass of metal under the 

 working stresses. In the case of the large shaft (12 inches in diameter) the 

 failure was found to be due to the fact that while the exterior of the shaft, for a 

 depth of about an inch, consisted of fairly satisfactory material, the interior oi 

 the steel showed a very coar.se, weak structure, possessing very unsatisfactory 

 mechanical properties. The failure is in this case regarded as due, in all proba- 

 bility, to insufficient forging of the steel during the manufacture of the shaft. 



In conclusion it was claimed that modern methods of investigation, if com- 

 pletely carried out imder favourable conditions, have a considerable prospect of 

 success in arriving at the cause of a given breakage. The favourable conditions 

 principally insisted upon are the fullest po.^^sible information as to all relevant 

 circumstances and the careful preservation of all portions of the fractured part 

 from rust and dirt. 



5. The Electrical Conductivity of certain Light Aluminium Alloys an 

 affected by Exposure to London Atmosphere.^ By E. Wilson. 



TUJ<:SDAr, SBPT£!\fB/'JR 8. 

 The following Papers were read : — 



1. The Laws of Flight.- By F. W. Lanciiester. 



The Theory of Stistentation. 



The law of the pressure reaction for speeds such as may be usefully employed in 

 aerial flight has been established experimentally, and is : — 



'1) On a body of stated gemnetrical form in yiven presentation the pressure 



' Published in the Mectrician, September 11. 

 ^ Published in Engineering, September 25. 



