In forming all my plans I have looked to the perfection of the surface on which the carriages are to run, as the great and ultimate desideratum; and in the detail of construction of this last operation, without which all the previous labour is comparatively wasted, I have always contemplated introducing all the perfection of materials and workmanship of which it is capable. With a view to improvement on this point, Mr. Brunel considered it would be advantageous if two important changes were made. He proposed, in the first place, to abolish the use of stone blocks for the rails to rest on, and to substitute timber; and, secondly, to apply the support uniformly and continuously along the whole length of the rails, instead of only at intervals. The first of these changes, namely, the substitution of timber for stone, was not wholly new, for transverse wood sleepers were often used in exceptional situations; but it was the general opinion that stone, where it could be applied, formed the best support for the rails,[51] and the exclusive employment of timber was considered a great innovation. The other principle, that of ‘continuous bearing,’ was similar to that of the old wooden and stone tramways; and, even as applied to iron rails it had been extensively used before, as Mr. Brunel mentions in his report of August 1838 (see Appendix I. p. 535.) Mr. Brunel designed for this continuous bearing a peculiar shape of rail, which, from the form of its section, acquired the name of the ‘bridge rail.’ The rail was bolted down to the longitudinal timbers, and the timbers of the two rails were connected together at intervals by cross-pieces, called transoms, bolted to them; these served to keep the two rails at a proper distance apart. The longitudinal timbers lay on gravel or other ‘ballast,’ which had been found to form the best foundation, as being firm and solid, easy of adjustment, and allowing free drainage.[52] Mr. Brunel, however, thought there would be difficulty in giving the longitudinal baulks a sufficiently solid bearing on the gravel below them. A similar difficulty had already been experienced with the heavy stone blocks used on other railways. As a remedy for this, Mr. Stephenson caused each block to be lifted and dropped several times on its place, so as to consolidate the ballast below. The same thing could not be done with a long wooden baulk, and Mr. Brunel therefore contrived another mode of overcoming the difficulty. Piles were driven into the ground between the rails, and their heads bolted to the cross-transoms, the object being to hold the timber framework firmly down. The gravel was then rammed hard under the longitudinal baulks, to give the consolidation desired. The result, however, of this mode of construction was far from successful, and the state of the road, when run over by the trains, was in many places very defective.[53]