.S7A' M'lUJAM SIEMENS, F.R.S. 



399 



uiiv was tested to nearly luO tons per square inch. In his own 

 practice he generally applied 80 or 1)0 tons as the weight which 

 steel wire, for telegraphic purposes, ought to bear. But between 

 the two limits between the material that would bear a breaking 

 strain of 28 tons, and elongate 25 or 30 per cent, before breaking, 

 and the highly wrought material which would bear 80 or 90 tons 

 to the square inch there were a great many steps, all of which 

 applicable under circumstances such as practice must indicate. 

 K<T links, he should consider that a material capable of bearing 

 from 45 to 50 tons to the square inch would be the best material. 

 It would be sufficiently yielding to excessive strain ; it would 

 yield 10 per cent, before rupture took place, and at the same time 

 it could be loaded to certainly 15 tons to the square inch with 

 safety. In constructing long bridges, the difference was so great, 

 whether using a material that could be loaded safely and practically 

 with 15 tons, or a material safely bearing a strain of only 5 tons, 

 which was the limit now imposed upon iron, that there could be 

 no question, in the long run, which was the right material for 

 such large structures as were referred to in the paper. As re- 

 garded compressive strain, steel must be looked upon as potentiated 

 wrought iron, bearing a strain of compression equal to that of ex- 

 tension ; but it must be borne in mind that the smaller scantling 

 of the steel under compression necessitated a change in construc- 

 tion with a view to giving the requisite amount of lateral support ; 

 it was therefore not sufficient to make a design in iron and reduce 

 the scantlings in substituting steel, as had been the practice to 

 some extent. 



