(5) A fifth point, almost as important as these first four, the necessity for the providing of crossings. The great arterial road reserved to the internal combustion engine would be, for people who had to cross it, an obstacle a great deal worse than a railway. Our forefathers protected in all sorts of fashions the road crossing the railway at a level crossing--by insisting on gates and an attendant, by compelling the road, if possible, to pass above the railway upon a bridge, and so on. More attention was paid to this point in England than in any Continental country, and we benefit by the results of that care to-day. But the arterial road would be far more dangerous. It would have a continual stream of very rapid vehicles in both directions, and the scheme had better not be envisaged at all if the cost of providing for cross traffic is not faced. The problem is by no means an easy one. It means, necessarily, embankments for bridges, or tunnelling, at _every_ crossing, and these will have to be more numerous than the road crossings: they will have to serve rights of way and private approaches as well. I think it will be found, when the scheme is first attempted, that this obstacle will prove the most serious of all. It is for experts in the science (of which I know nothing, and allusion to which I have therefore kept carefully out of this essay) to decide what these details of surface, width, foundation, etc., mean in practice: their expense and character. They know from experiments made what materials and foundation may be best, what minimum width suggests itself (I have occasionally heard the minimum width of 100 feet suggested); but whatever the detailed practice, when the experts set to work on the new motor roads it must be with these five main provisions before them. There are minor considerations. You have, with the new traffic, to consider a gradient somewhere between the old road gradient and the railway gradient. There, again, it is for experts to determine what the maximum useful gradient should be. The trouble in our present road system is that in any trajectory you will have one or two places where the new traffic is perilous. There are even exceptional points in England where it is almost prohibited by excessive gradients. Another point in connection with such great arterial roads is the capital one of exit from the great urban centres. It is of little use to relieve traffic, to diminish the strain and expense of energy connected with it, and the peril, and all the rest of it, between two urban centres if the exit and entry from and into each are blocked.