226 On the strength of iron bolts, &c. [April, 



when 10=71.0 I 2 . This place is marked * in the table. 

 Remarks on keys, hold-fasts, fyc. 



Put 6= the breadth in inches, 

 tZ=the depth in inches, 

 w= weight in tons, 

 Z= length of bearing in inches ; then the breadth should never 



w / 37 wl\%' 



be made less than ——• and the section bcP—,37 vo I, or «?=\ — 7 — / 



As an example, suppose a bar 1 inch square to support 8 tons was 

 fastened by a key ; required the breadth and depth ? 



w l 



w =8.—l=l and-^ = -^— = — = b or the breadth required, 



/.37m>K 4 



" ~ V — b — / *" v/8 - 88 — 2 ' 98 inch e^ the depth required. 



To support the accuracy of this table, a set of experiments was com- 

 menced, but the results from them were so unsatisfactory, that they 

 were not continued. But during the proof of three bridges in which 

 bolts of from 1^ in. to 2£ in. were used, with various lengths of bear- 

 ing, and pressures of from 20 to 1 5 tons, the dimensions marked in 

 the table were found sufficiently strong in every instance ; but the 

 diameter of the bolt thus given could not be reduced much, or what 

 was the same thing, the length of bearing could not be decreased with 

 out a risk of failure. 



«^_ The best Swedish iron bolts did not sustain a greater pressure 

 ^eY^ than the ordinary English bolt iron, (rolled, not hammered.) The 

 Swedish iron when strained in excess bent, and became dented as in 

 the marginal figure : the side a was bulged or rose half as much as b 

 was indented or bent, on the other side ; when the bolts were 'formed 

 of English bolt iron (unhammered), numerous cracks opened on the 

 convex surface of the bolts at a and c c, when the indentation at b 

 amounted to ,'„ of the diameter of the bolt ; the bolt failed by these cracks 

 meeting each other, and the centre part of the bolt was drawn out. 



The bars, which these bolts connected, were calculated to sustain 9 

 tons per square inch of section, and the eyes 7 tons, but when the 

 whole were proved by a tension ird greater than the calculated 

 strength, the eyes broke more frequently than either the bars or bolts. 



The following table, for which we are also indebted to Captain 

 J. Thomson, Engineers, will serve as a practical continuation of the 

 observations on roofing, in the last number of the Journal. 



