December 30, 1922J 



NA TURE 



879 



way of the heat-treated over the untreated metal. 

 One illustration will suffice. In the case of mild steel, 

 containing 0-25 per cent, carbon and 0-7 per cent, 

 manganese, the yield point was raised by the heat 

 treatment from 22-2 to 30-5 tons per square inch ; the 

 maximum stress from 33-5 to 42-8 tons ; while the 

 Brinell Ball Hardness Numbers rose from 145 to 197. 

 Needless to say, all the metals have been subjected 

 to careful chemical analysis, and both the treated 

 and untreated specimens have been studied photo- 

 micrographically, horizontal and longitudinal sections 

 having been prepared of all the metals save the cast 

 irons. This was rendered desirable in view of the 

 fact that all the wrought irons and steels had been 

 rolled. The longitudinal sections were taken at 100 

 diameters magnification, this being regarded as par- 



better shock test results obtained with the treated 

 material. It is calculated that in one of the mild 

 steel specimens the number of grains per square inch 

 is 820,000, while, when heat-treated, including quench- 

 ing, the ferrite grains number about 5 million per 

 square inch. This gives an idea of the closeness of 

 the structure and the greater homogeneity produced 

 by suitable heat treatment. 



The necessary bars having been prepared, the com- 

 mittee were now faced with numerous problems connected 

 with their despatch to various parts of the world, namely 

 to Plymouth, Auckland, Colombo, and Halifax (Canada). 

 One of the most difficult of these was the method of 

 marking the bars. In view of the possibility in some 

 instances of very severe corrosion, there was a distinct 

 probability that any ordinary marking would be obliter- 



:&j. 



,y 



*< 



Fig. 3. — Mild steel (with 0-7 per cent, manganese). 

 Longitudinal section X 100 (treated Quo" water : 700 water). 



ticularly suitable for examining the elongation of crystal 

 grains due to rolling ; the transverse sections were 

 photographed at 600 diameters. Fifty-four beautiful 

 reproductions of the photomicrographs are given in 

 the paper, and four of these are reproduced in these 

 columns through the courtesy of the Institution of 

 Civil Engineers. 



It is to be anticipated that the microstructure 

 of the metals will play an important part in their 

 powers of resistance to corrosion. Correlation of the 

 micrographs and mechanical tests reveals several 

 interesting features. In so far as the wrought irons 

 are concerned, the heat treatment, by reducing the 

 grain size, distinctly improves the shock test figures. 

 The effect of heat treatment on the carbon steels 

 has been, in the main, in the direction of preventing 

 the marked separation of ferrite and pearlite, such 

 as exists in the bars as rolled, and thus to produce 

 a more homogeneous structure. This is well illus- 

 trated in the accompanying photographs. 



This serves, in a laree measure, to account for the 



Fig. 4. — Mild Steel (with 0.7 per cent, manganese). 

 Transverse section X 600 (treated 900 . water: 700 ° water). 



ated. It was intended that, when exposed to corroding 

 influences, the two ends of the bars should be firmly 

 embedded in concrete in a special frame erected for 

 the purpose. Although a precise record would be 

 kept of the position of each specimen in the frame, 

 which would serve as some protection against mixing, 

 there was the further danger that badly corroded bars 

 might fall out of place and their identity be lost. 

 The difficulty was eventually overcome by an ingenious 

 system suggested by Mr. Maurice F. Wilson, a member 

 of the committee, and now the chairman. The 

 method consists in having one, two, or three holes 

 drilled through the plates at both ends where they 

 will be preserved by the enveloping layers of concrete. 

 The holes are drilled in different positions ; those at 

 one end give what is termed the " classification letter " 

 and indicate the type of metal, whether, for example, 

 it is Swedish iron or cupriferous steel. At the opposite 

 end the holes indicate the number of the bar. 



In order to determine the effect of strain and of 

 contact of dissimilar metals a few bars were bent at 



NO. 2774, VOL. I IO] 



