After the spokes have been driven in they are shaved off by the spokeshave to their proper form, Fig. 14; and the lengths being measured from the nave, the outer tenons are cut, sometimes square, sometimes cylindrical, but leaving the back shoulder square to abut on the felloe with greater firmness. In the manner of tonguing there is a great deal of difference of opinion amongst wheelwrights; that the tongues in size should be slightly in excess of the hole or mortise to receive them, is a generally received idea, but a difference of opinion exists as to the length. But certainly it seems more reasonable to cut the tongues a little _shorter_ than the length of the hole in the felloe to receive them, because when the tyre is put on it shrinks in cooling and draws up all the joints. Now suppose the tongues are a little longer than the holes, the shrinking of the tire causes it to press on these; and as they are firmly fixed at the nave, there is no escape for them, and the result is that the spoke is seriously crippled. It is fair to say that there are many good practical men who work in this latter way, and with to all appearance good results, but it is evident that the principle is not a good one. One of the difficulties in making light carriage wheels is to get the spoke tightly into the felloe without splitting it, and the manner of accomplishing this is more successfully done, not by making the tongue on the spoke so large that it will fill the hole in the felloe of itself, but by making the tongue rather smaller and slitting the edges after it has been driven in, and then wedging up with small wedges, just in the same manner that a joiner would wedge up the mortises and tenons of a door. When the wheel is so far progressed with it is laid on the ground, and the felloes are ranged round it in the order they are to be fitted. It is of the greatest importance that the holes should be bored in the felloe in an exact radial line from the nave; if this is not done, the spoke will have to be strained out of the straight line in order to get it into the hole; this will put an undue pressure upon it, and it is very likely that before the wheel has been long in use the spoke will break off short at the felloe. The exact position of all the mortises and joints should, therefore, be worked out on a full-sized drawing, and this being accurately done, there will not be much danger of going wrong. As the construction of a wheel is somewhat analogous to the arch, it is considered that by giving the felloes a number of joints the strength of the wheel is very much increased. Whether this be so or not must be left to the theorists to determine, as we have no trustworthy results from the various experiments under this heading.