September io, 1903] 



NATURE 



463 



as Carnegie research scholar, its publication having been 

 delayed by his appointment to the chair of metallurgy at 

 Montreal. The memoir covers thirty-six pages. The 

 burnt structure of very much overheated steel is shown to 

 be largely due to the partial melting which results from 

 heating the steel above a given temperature. This melt- 

 ing causes brittleness directly, and indirectly by the 

 admission of oxygen to the steel. According to American 

 metallurgists the latter stage would alone be called burn- 

 ing, but as the effect of partly melting the steel is quite 

 distinct from that of overheating below the zone of partial 

 fusion, the author would prefer to apply one word to the 

 whole of the changes that take place in this zone. If the 

 word burning is still employed, it should be remembered 

 that it is essentially a partial melting of the steel, though 

 often accompanied by oxidation. The following stages 

 have been recognised : — (i) overheating (below the point of 

 incipient fusion) ; (2) partial melting, called burning ; (a) 

 merely producing segregation of carbon in the joints ; (h) 

 accompanied with liquation and producing flaws ; (c) further 

 liquation and oxidation in the flaws, (i) Steel that has 

 merely been overheated can be completely restored by heat- 

 ing just above its highest recalescence point and allowing 

 to cool. (2) Steel in the stage (a) can be restored by suit- 

 able annealing; in the stage \h) forging would also be 

 needed ; and in stage (c) it would be restored with great 

 difficulty, if at all. 



The paper on the heat treatment of steel submitted by 

 Dr. William Campbell (New York) is a report on research 

 carried out by the author as Carnegie research scholar. It 

 forms a pamphlet of ninety-three pages. The steel used 

 contained 050 per cent, of carbon, 098 manganese, 0094 

 silicon, 0098 phosphorus, and 008 sulphur. The structure 

 of the steel used was found to depend upon the two con- 

 stituents present, namely, the ferrite and the pearlite. The 

 pearlite will certainly show the finest structure when the 

 steel has been heated to just above Ac,, or when it has been 

 transformed into martensite. Heating to temperatures 

 above this point will cause a coarsening of the structure. 

 The higher the temperature the coarser the structure. 

 Above Ac, the ferrite begins to diminish in size, due to its 

 being dissolved in the martensite. This will continue until 

 the whole of it is dissolved, when the change Ac2_3 is com- 

 plete. Then the finest structure of the whole will be found 

 where these two changes balance. This point is apparently 

 just below the point where Ac2_3 is complete. The best 

 finishing temperature is such that the bars leave the rolls 

 as near Ar2_3as possible. The bars would necessarily have 

 to be drawn from the furnace at a higher temperature, 

 which is about 740° C. in this case, allowing for a cooling 

 of, say, 40° C. or more during rolling. In comparing the 

 results obtained with those of pure carbon steel, the effect 

 of the manganese present must be taken into consideration. 



.'\n animated discussion followed the reading of these 

 papers on heat treatment, in which Messrs. Westgarth, 

 Ridsdale, Lange, Price-Williams, L. N. Ledingham, and 

 Hadfield took part. 



The probability of iron ore lying below the sands of the 

 Duddon Estuary formed the subject of a paper by Mr. J. L. 

 Shaw (Whitehaven). He adduces evidence to show that 

 there is a limestone area probably carrying large bodies of 

 ore, and advocates the putting down of exploratory bore- 

 holes. In the discussion Mr. G. J. Snelus gave further 

 particulars of geological interest. 



The paper by Mr. W. F. Pettigrew on coal as fuel at 

 Barrow-in-Furness contained much of interest. In that 

 district at the present time coal is obtained from Cumber- 

 land, Lancashire, and Yorkshire. As the prices at the pit, 

 the cost of carriage, and the quality of the coal from these 

 districts vary considerably, the author has carried out 

 several experiments to find the relative value of coal 

 obtained from the districts before mentioned, also from 

 various parts of Scotland and South Wales. Experiments 

 carried out with a locomotive showed that the sample of 

 Yorkshire No. i gave the best results. This coal has excel- 

 lent steaming qualities, is very clean, with an open clinker. 

 and low percentage of ash. the Welsh coal was also good 

 when tried, and equal in all respects to the Yorkshire coal, 

 and would no doubt give even better results if properly fired, 

 which was not the case during the trials, the men having 



had practically no experience with this kind of coal. The 

 Cumberland coal was good, particularly one sample, but 

 this was not found suitable for locomotive purposes. The 

 other sample of Cumberland coal gave fairly good results, 

 but it is a dirty coal, and necessitates the frequent cleaning 

 of fires. The Lancashire samples were in some cases very 

 good steaming coal, with a moderately low consumption, 

 but several samples gave very bad results, and were quite 

 unfit for locomotive purposes. The Scotch coals tested were 

 fairly good, but in most cases a very heavy consumption was 

 recorded. They are quick burning coal and dirty, but with 

 an open clinker, which did not interfere in any way with 

 the steaming. The consumption was from 20 to 40 per 

 cent, higher than the Yorkshire coal. 



Mr. C. H. Ridsdale (Middlesbrough) read a lengthy paper 

 on the diseases of steel. In it he collated various types of 

 defects, and traced them to their origin. 



Mr. H. Ehrhardt, of Diisseldorf, contributed a paper de- 

 scribing a process for making weldless steel pipes and 

 shells by which rings up to 8 feet in diameter and 10 feet 

 in length are manufactured. 



The regulation of the combustion and distribution of the 

 temperature in coke oven practice was dealt with in a paper 

 by Mr. D. A. Louis. Illustrations were given to show the 

 design and character of the Brunck and v. Bauer coke 

 ovens, two ovens of new design. 



The influence of silicon on iron was dealt with in a paper 

 by Mr. Thomas Baker. He prepared a series of alloys of 

 silicon and iron with traces only of other elements, and 

 studied the micro-structure and physical properties of each. 

 Although the addition of silicon to iron increases the elastic 

 limit and tenacity of iron, such increase is only obtained 

 by loss of ductility, which loss, provided the material has 

 been well annealed, is very small until the silicon reaches- 

 3 per cent., after which it becomes very great, the ductility 

 almost becoming zero with 4 per cent, silicon. The alloys 

 gradually increase in hardness with the addition of silicon, 

 and after exceeding 5 per cent, silicon require great skill 

 and care in machining in order to avoid fracture of the bar. 

 As the percentage of silicon increases the permeability fOr 

 low magnetic fields increases, and the coercive force and 

 hysteresis loss decrease. Prof. T. Turner (Birmingham) 

 was the chief speaker in the discussion. 



The proceedings concluded with the customary votes of 

 thanks to the reception committee, and an invitation, 

 tendered by Mr. Kirchhoff, of New York, on behalf of the 

 American societies, that the Institute should meet in the 

 United States next autumn was accepted. 



In connection with the meeting an elaborate programme 

 of visits and excursions was arranged, including the works 

 of the Barrow Haematite Steel Co., the Askham blast 

 furnaces, the Hodbarrow mines and sea-wall, the naval 

 construction works of Vickers, Sons and Maxim, the Fur- 

 ness Railway locomotive works, the North Lonsdale iron 

 works, and to Lake Windermere, Grasmere, and Blackpool. 

 The social functions included a conversazione given by the 

 Mayor, a ball by the reception committee, a garden party 

 by Mr. Victor Cavendish at Holker Hall, and an illuminated 

 jHe at Furness Abbey. 



UNIVERSITY AND EDUCATIONAL 

 INTELLIGENCE. 



Saturday, October 31, has been fixed for the holding of 

 a convocation of the University of Oxford for the purpose 

 of electing a Chancellor of the University in the place of the 

 late Marquis of Salisbury. 



Arrangements for next term have been published in con- 

 nection with the Oxford University School of Geography. 

 Nine lectures a week by different members of the staff will 

 be given in various branches of geographical science, and 

 practical instruction to supplement several of the courses 

 of lectures has been arranged. A geographical scholarship 

 of the value of sixty pounds is to be competed for on 

 October 14, and candidates must have taken honours in one 

 of the final schools of the university. Courses of instruc- 

 tion are now given also in preparation for the university 

 certificate in surveying, and to meet the requirements of 

 students reading for the university diploma in education. 



NO. 1767. VOL. 681 



