83
country have not been formed with the intention of wheeled vehicles being used on them. Their surfaces are uneven and irregular, and little skill has been shewn in the choice of route so as to avoid hills or to get the best possible gradients.
There are many rivers which, if properly tended, would form excellent means of transport, but in some cases these have been neglected and in others treated in an erroneous manner. The Tone-gawa, the largest river in Japan, has a bar across its mouth on which there is not sufficient water to allow the native junks to pass over it. Inside the bar there is a considerable depth of water, and the river is navigable for small craft for more than 100 miles. The Shinano-gawa, the second largest river in the country, has 6 feet of water on its bar, and there is little doubt that this might be deepened with ease were proper means taken to effect this. It has been allowed to break through its original confines until it is in some places two or three times its proper width, and is so dammed back by shallows that in floods the water overflows the banks and spreads over hundreds of square miles of rich cultivated country. For how many hundred years this natural process of washing away the banks and widening the river has been going on without check, or for how long it has been allowed to flood the adjacent lands, I am not in a position to say, but a step was recently taken with the avowed intention of remedying the latter evil, which however has proved unsuccessful. Instead of keeping such an enormous river, which is equal in volume to that of the Rhine, in the course which nature ordained for it, and taking the natural and more easy method of training its banks, regulating its width and inclination, and, if necessary, straightening its course, the Japanese conceived the idea of cutting another and separate channel to the sea fur the purpose of carrying off the flood waters—a great part of which has been already executed—but the works are now stopped. The design was erroneous in so far that the abstraction of the flood waters would probably result in a further shallowing of the natural course of the
