274 KEPOKT— 1889. 



As also pointed out by Mr. Howe, ' Silica is often mistaken for silicon : 

 who knows how far it is responsible for this metalloid's bad name ? ' 

 This was actually noticed by Mr. Turner in test bars of steel containing 

 comparatively low percentages of silicon — that is, under "5 per cent. — 

 much of the silicon present being in the oxidised condition. 



Also, according to M. Gautier, there is a difference between steel made 

 with silicon only and that with silicon and manganese, i.e., between a 

 product made by adding ferro-silicon (carbon, silicon, and iron alloy), 

 and that with silicon-spiegel (carbon, silicon, manganese, and iron alloy) ; 

 and he mentions the following interesting experiment by his then colleague, 

 M. Pourcel. 



In a porcelain tube were placed two receptacles, one holding steel 

 made by adding ferro-silicon only, and the other steel by an alloy of 

 silicon-spiegel. A current of chlorine was passed until all the iron was 

 removed in the state of chloride. It was then seen that in the first 

 receptacle there remained a network of silicate of iron preserving the 

 original formation of the piece, whilst steel by silicon and manganese 

 alloy left no residuum. Also, that such steel with no manganese was" 

 red-short, lacked fluidity, and possessed other defects. The writer has, 

 however, not noticed such difference in the material now described, 

 which in its molten state pours well, the ingots forging easily, and up 

 to 2 per cent, silicon the ductility in the testing-machine being very 

 good. 



However, as suggested by Mr. Howe, possibly silicon does enter into 

 different combinations in steel, some promoting, some impairing ductility 

 and malleability. In favour of this is the fact that so many well-known 

 scientists and metallurgists have utterly condemned in forged steel the 

 employment of silicon, even if present in small amounts. Such strong 

 opinions would not be expressed without good grounds, and a reasonable 

 explanation for the apparent discreiDancy noticed by different observers 

 seems to some extent to be in the direction named. At any rate the 

 samples described in this paper, and containing up to 2 per cent., present 

 a remarkable ductility and toughness both in the bending and tensile 

 specimens. 



The writer wishes it to be understood that he does not claim that 

 silicon should take the place of carbon. Smaller quantities of carbon 

 produce the requisite hardness and different tempers required in the in- 

 dustrial application of steel, and in fact silicon alone does not produce a 

 steel that will harden by water-quenching, thus in this respect resembling 

 manganese steel. Still it is a somewhat remarkable fact that a steel 

 (specimen 0) coutaining 1'60 per cent, of a metalloid ordinarily so much 

 distrusted stretched 35-10 per cent, (on 2"), with 54'52 per cent, reduc- 

 tion in area, and a test bar from the same material tested by Professor 

 Kennedy gave 24-30 per cent, (on 10"), with 58'30 per cent, reduction in 

 area. Also, had not the specimen D, tested by Professor Kennedy, broken 

 in the threads (the diameter of the bar over the threads being only -93 

 against •898 of the tested part of the bar : too small a difference with hard 

 steel in the holding part), no doubt his test would have confirmed the 

 writer's that a material with eveu 2-13 per cent, of silicon will elongate 

 36-50 per cent, on 2" (equivalent to about 27 per cent, on 10"), with 

 59-96 per cent, reduction in area. So that whilst it may not be advisable 

 to use silicon as a hardener in making steel, it is important to have it 

 proved that the brittleness noticed in ordinary so-called silicon steel is 



