ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 719 



steel rails which have fractured or duly worn in use, as well as on 

 other rails which have shown especial durability. His conclusions are 

 corroborative of other investigators' results and show the iinportauce of 

 giving to steel rails a fine-grained structure throughout — not merely as 

 a surface layer. The rail should be fine-grained even at the centre 

 of the head, and practically amorphous at the surface. Reliance upon 

 chemical analysis is quite untrustworthy. 



S. S. Martin's * experiments point to the same conclusion, and he 

 argues that since no change of structure can result from finishing below 

 the critical point, a rail must be rolled as near the critical point as 

 possible to get the best structure, or in the case of rail steel between 

 700° C. and 725° C. 



A. Sauveur,f after discussing theories of present methods, points out 

 that, in order to confer a fine-grained structure upon steel rails, there 

 seem to be three courses of manufacture open : — (1) To shorten the 

 crystallising period, i.e. the time during which the rail is allowed to 

 cool undisturbedly above the critical temperature (say about 700° C). 

 (2) To cause a part of the crystallising period to occur previous to the 

 final pass through the mill. (3) To finish the rail at the temperature 

 most desirable for easiness and speed of manipulation in rolling, and 

 then to reheat it to a temperature slightly above the critical, a treat- 

 ment which would result in the breaking up of the pre-existing coarse 

 structure and replacing it by a much finer one. 



S. S. Martin | also shows by the comparison of fine micro-sections 

 the difference between hot and cold sawing upon steel. It follows that 

 in studying the structure of rails care should be taken that the polished 

 sections are sufficiently removed from the hot sawn side to be un- 

 affected by the action of the saw ; otherwise seriously misleading 

 observations might result : a coarsely crystalline rail might be made to 

 appear fine-grained. 



Effects of Strain on the Crystalline Structure of Lead.§ — J. C. 

 W. Humfrey experimented with some exceptionally favourable lead 

 crystals. The effect of tensile strain was to produce slip-bands, and it 

 appeared that, when a slip had been produced in any part, there was 

 a tendency for it to continue there rather than in other parts of the 

 specimen. Thus the effect was rather to localise the strain. If the 

 originally uniformly oriented crystal showed signs of recrystallisation 

 after straining, it was found that moderate heating (up to 100° C.) very 

 much facilitated the process. Experiments were specially carried out 

 to determine whether the recrystallisation, which is apparent imme- 

 diately after re-etching a severely strained crystal, is a direct and 

 instantaneous effect of the strain, or is a growth which occurs during 

 the interval of time that has elapsed during the straining and the 

 examination. The author's opinion is clearly in favour of the latter 

 hypothesis, and he considers it ought to be classed with the progressive 

 growth demonstrated by Ewing and Eosenhain in their observations 

 on the crystals of ordinary lead after straining. 



* Tom. cit., pp. 191-6 (4 microphotos); and Iron Age, Dec. 26, 1901. 



t Tom. cit., supra, pp. 197-202. 



% Tom. cit., pp. 245-7 (5 microphotos). 



§ Proc. Roy. Soc, lxx. (1902) pp. 462-4. 



