January 12, 1905] 



NA TURE 



259 



commissioner, remarlis that he considers " By far the most 

 important experiments witnessed by the Commission were 

 those made by Keller, Leleux and Company at their works 

 at Livet. " 



It is a little diflficult to realise upon what grounds the 

 above conclusion was arrived at. Putting aside the 

 speculative calculations of M. Keller and descending to 

 experimental facts, it appears that the commission saw 

 smelted several tons of pig-iron, as a rule remarkably high 

 in manganese (1-5 per cent, to 4 per cent.), and hence of 

 limited commercial interest, and as it is evidently not 

 thought by the conunission that the electric furnace is 

 to become a serious competitor with the blast furnace, the 

 specified exceptional value of these results from an industrial 

 point of view is not quite clear. 



As regards steel, only one not very satisfactory and un- 

 tested heat was made (see pp. 77-78), yet upon such evidence 

 the report states that this process is capable of producing 

 steel equal to the best products of Sheffield's crucibles. 

 Such premature conclusions based on such scanty data are 

 not calculated to carry conviction to the experienced metal- 

 lurgical mind. 



The commission also describes a series of experiments 

 made by M. H^roult at La Praz works. The analyses of 

 the steels obtained appear quite satisfactory, but this process 

 is hardly capable of competing with the ordinary open- 

 hearth furnace even from the rosy point of view taken by 

 the commissioners based on costs calculated (in all good 

 faith) by the patentee. 



From a British point of view Kjellin's induction process 

 deserves the most serious attention in view of (under certain 

 conditions) its probable competition with the crucible steel 

 process. 



Analytically, mechanically, and micrographically this 

 steel leaves nothing to be desired, but unfortunately 

 chemical and tensile tests, and the indications of the micro- 

 scope, have a limited value in determining the working 

 capabilities of tool-steel. 



In his "conclusions" on p. 115 of the report, Mr. 

 Harbord states that " Steel, equal in all respects to the 

 best Sheffield crucible steel, can be produced, either by the 

 Kjellin or H^roult or Keller processes, at a cost consider- 

 ably less than the cost of producing a high-class crucible 

 steel." 



The above statement, so sweeping and involving issues 

 of profound industrial import, should have been made only 

 as the result of a series of exhaustive working tests. For 

 such, in the report, the reviewer has sought in vain. 



It is true that a series of tests of turning tools made 

 from Kjellin and H^roult steels has been carried out at 

 Woolwich by Mr. H. F. Donaldson, but the results are 

 •quite inconclusive, because of the steels employed hardly 

 one was fit for turning tools. 



Cold sett steel, carbon 08, cold chisel steel, carbon 09, 

 tap-steel, carbon 11, and drill steel, carbon 1-2, have all 

 been set to do the work of a comparative turning tool steel 

 of carbon i jS per cent. 



The natural consequence is that in the Woolwich results, 

 where " E " means equal to the ordinary Woolwich turn- 

 ing tool steel of carbon 1-38, and " NE " means not equal 

 to that steel, we find in the report, pp. 87 and 88, five 

 " equals " and no less than fourteen " not equals." 



As to whether Kjellin electric steel is or is not equal to 

 crucible steel time alone can show. The conclusion of the 

 commission may be accurate, but it is certainly not based 

 on any scientific evidence worthy of the name. 



Such evidence on a commercial scale can be conclusively 

 obtained only by at least two comparative years of shop 

 practice, employing all kinds of tools, and recording the 

 average wear and waste of the steels as evidenced by the 

 ratio between the work turned out and the annual cost of 

 the tool steels purchased. 



In the micrographic section of the report the reviewer 

 notes with regret a recrudescence of the use (in this con- 

 nection) of the meaningless and unscientific term " grain " 

 in describing the allotrimorphic crystals of ferrite. 



These crystals, although usually lacking idiomorphic 

 external faces, nevertheless present that internal molecular 

 symmetry associated with individual crystals, and hence 

 should be classed as such. 



Prolonged tests on Kjellin steel of all carbons, compared 

 NO. 1837, VOL. 71] 



with similar crucible steels, have been inaugurated at the 

 University College of Sheffield, and the erection of a Kjellin 

 furnace capable of making one ton of steel per day is under 

 consideration. 



Without in any way compromising one's industrial atti- 

 tude as to the exact capabilities of the respective methods 

 devised by Messrs. HiSroult, Keller and Kjellin, one can 

 cordially congratulate these gentlemen on the scientific 

 ability displayed in the development of their several methods, 

 all of which, within their legitimate spheres, are un- 

 doubtedly of great metallurgical value. It is the more 

 necessary to say this because such value is liable to be dis- 

 counted by the hasty and ill-digested conclusions drawn by 

 the Canadian commission. J. O. Arnold. 



LONDON FOG INQUIRY, 1901-3.' 

 'T'HE Meteorological Council have issued their final report 

 on the above inquiry, which had to be terminated at 

 the end of the winter 1902-3 as the London County 

 Council were unable to make any further contribution to 

 its cost beyond the 250/. originally assigned. A short 

 account of the chief results obtained by Captain Carpenter 

 from the observations of the winter igoi-2 has already 

 appeared in these columns (vol. Ixvii. p. 548). During the 

 succeeding winter records of the duration and intensity of 

 fog were continued at forty-six stations in and around 

 London, and in addition to this the scope of the inquiry was 

 extended to include a detailed study of the distribution of 

 air temperature over the London area. With this object 

 thermometer screens and dry bulb thermometers were issued 

 to thirty fire brigade stations, and daily observations of the 

 air temperature were made at fixed hours. 



The material so accumulated has been utilised to deter- 

 mine so far as possible the physical causes most active in 

 producing fog in each case. The guiding principles adopted 

 in the classification are those suggested in an article by 

 the secretary to the Meteorological Council which appeared 

 in Nature (vol. Ixiv. p. 649) at the time when the 

 inquiry was started. The majority of our fogs were 

 found to be due to radiation from the earth's surface during 

 calm nights. Others, among them the most persistent fog 

 of the winter, were caused by the passage of warm air over 

 a previously cooled surface, while a third group were 

 identified as *' cloud " fogs. A certain number ot fogs could 

 not be included in any of the above categories. They 

 appeared to be mere accumulations of the products of com- 

 bustion in an almost calm atmosphere, and as such were 

 termed " smoke " fogs. Full particulars of the thirty-nine 

 most serious fogs of the winter are given in an appendix. 



Among the chief results of the inquiry must be reckoned 

 the establishment of a workable scale for the estimation 

 of fog intensity by different observers, based on the extent 

 to which traffic is impeded by land, river, and sea. 



Comparison of the fog statistics from the various stations 

 confirms Captain Carpenter's results. With a few possible 

 exceptions which need further investigation, there is no 

 evidence to show that, in London, geological formation has 

 any influence on liability to fog. Again, as was to be 

 expected, the fog frequency on the river and in the parks 

 is very high, but the evidence does not support the view- 

 that the fog there found drifts far into the neighbouring 

 districts. 



With regard to the main purpose of the inquiry, greater 

 precision in fog forecasts, Mr. Lempfert points out that 

 a first step would be the establishment of a night service 

 at the Meteorological Office. As the majority of fogs are 

 caused by nocturnal radiation, and the intensity of this 

 radiation depends largely on the accident whether the sky is 

 free from cloud or not, it is manifest that forecasts issued 

 at the suggested hour of 5 a.m. would have a much greater 

 chance of proving correct than the present ones, which are 

 based on observations taken at 6 p.m. on the previous even- 

 ing. As most fogs become thick soon after sunrise, several 

 hours' warning could still be given, though the hour would 



1 Report of the Meteorological Council upon an Inquiry' into the Occur- 

 rence and Distribution of Fogs in the London .Area, during the Winters of 

 igoi-2 and 1002-3. with Reference to Forecasts of the Incidence and 

 Duration of Fogs in Special Localities, to which is appended the Report 

 by R. G. K. Lempfert, M.A. on the Observations of the Winter 1902-j. 



