The bridge was not, however, carried out after this design,
but was made of separate wrought-iron girders composed of riveted plates.
Recurring to his first idea of this bridge, Mr. Stephenson thought that a
stiff platform might be constructed, with sides of strongly trussed
frame-work of wrought-iron, braced together at top and bottom with plates
of like material riveted together with angle-iron; and that such platform
might be suspended by strong chains on either side to give it increased
security. "It was now," says Mr. Stephenson, "that I came to regard the
tubular platform as a beam, and that the chains should be looked upon as
auxiliaries." It appeared, nevertheless, that without a system of
diagonal struts inside, which of course would have prevented the passage
of trains _through_ it, this kind of structure was ill-suited for
maintaining its form, and would be very liable to become lozenge-shaped.
Besides, the rectangular figure was deemed objectionable, from the large
surface which it presented to the wind.

It then occurred to him that circular or elliptical tubes might better
answer the intended purpose; and in March, 1845, he gave instructions to
two of his assistants to prepare drawings of such a structure, the tubes
being made with a double thickness of plate at top and bottom. The
results of the calculations made as to the strength of such a tube, were
considered so satisfactory, that Mr. Stephenson says he determined to
fall back on a bridge of this description, on the rejection of his design
of the two cast-iron arches by the Parliamentary Committee. Indeed, it
became evident that a tubular wrought-iron beam was the only structure
which combined the necessary strength and stability for a railway, with
the conditions deemed essential for the protection of the navigation. "I
stood," says Mr. Stephenson, "on the verge of a responsibility from
which, I confess, I had nearly shrunk. The construction of a tubular
beam of such gigantic dimensions, on a platform elevated and supported by
chains at such a height, did at first present itself as a difficulty of a
very formidable nature. Reflection, however, satisfied me that the
principles upon which the idea was founded were nothing more than an
extension of those daily in use in the profession of the engineer. The
method, moreover, of calculating the strength of the structure which I
had adopted, was of the simplest and most elementary character; and
wha