(31)

bottom, the first to resist compression, the latter extension are much too small. It has been found by experiment that the strength of all girders whether of cast or wrought iron depends upon the strength of the top and bottom, the difference between the two materials requiring a different arrangement of the strength; in the case of cast iron which gave a greater resistance to compression than to extension the bottom flange of a girder is made of a greater sectional area than the top; the properties of wrought iron in respect to compression and extension being reversed, the position of the largest flange (so to speak) in girders formed of this material is also reversed. I have before me the records of the experiments made to ascertain the strength of wrought iron tubes before the section of the Conway tubular bridge was decided, and the following are some of the dimensions and results. In the final experiments the tube or girder was 75 feet in length between the supports, it was 4 ft. 6 in. in depth (about one-seventeenth of the span), 2 ft. 8 in. in width, the sectional area of the plates forming the top of the tube was 24 square inches and that of the bottom of the tube in the first experiment 8 4-5ths inches the breaking weight was 35 1/2 tons, the lower flange or bottom giving way. The sectional area of the bottom of the tube was gradually increased to 2 1/2 inches (the top remaining constant) when it appeared there was an equilibrium between the respective capabilities of the top of the tube to resist compression and the bottom to resist extension. The breaking weight in this case was 86 1/2 tons. Compare the above with the dimensions of No. 9 design as given with the dimensions of the experimental tube, in the table below.

From the above it will be seen that the sectional area of the top of the girder or tube of No. 9 design is less (with a greater span) than one-third of the experimental tube, and the sectional area of the bottom of the tube is one-eighth less than that of the bottom of the experimental tube as first tried. With these dimensions it is

not too much to assume that the breaking weight of the tube or girder in the design would be not greater than one-fifth of the breaking weight of that recorded in the first experiment on the tube, say 7 tons, and as a beam or girder should never be loaded beyond, I will say for iron, one-third of the breaking weight, any greater proportional weight destroying its elasticity, each girder in the design would carry 2 1/3 tons, or the three girders 7 tons. Girders for this bridge must perform a much greater duty than that.

I come next to No. 10, which is a variety of the American lattice bridge. Here the principle of this mode of construction has been departed from by diminishing the depth of the truss at the abutments, and from the circumstance of the upper string being made somewhat in the bow form and of greater dimensions than the lower string or tie, it is presumed that it was intended to act to a certain extent as a suspension girder which is not its duty and for which it is undoubtedly insufficient. I do not mean to say that the girders are insufficient in strength, but if the principle of this kind of construction had been carried out, they might have been formed with a less expenditure of material. It is to be observed also as far as would appear that every alternate girder only of the floor is trussed : that such trussing of girders 12 x 12 with a bearing of 23 feet is requisite there can be little question, there is therefore defect of strength in the immediate ones. But the most serious defect in the design is the evident insufficiency of the abutments. A simple inspection of the drawing, and comparison of the massive superstructure with its supports I think will show to any person that the latter are very far from being in due proportion or sufficient. I observe that the designer of No. 12 in his preliminary remarks states that the river is gradually receding from the East bank, if so doubtless it is encroaching in a corresponding degree on the West bank, and the configuration of that bank seems to favour the presumption; should that therefore be the case, the West abutment of No. 10 should be shortly isolated and placed in a very precarious condition even if it met its proper duty from the first, hence for this and the foregoing explained defects this design is not a suitable one. I may observe also in conclusion that there is not any specification attached to this design, and wanting that and sufficient details on the drawing, it cannot be clearly ascertained how the different parts are to be connected: for instance it does not appear how the vertical struts 1' 6" x 9" crossing the intersection of the diagonal struts are to be framed, nor how the props