The first wave of Asian immigration that swept over Europe gave names to the great features of nature, such as the rivers, long before the wandering tribes that composed it settled down into fixed habitations, and gave names to their dwellings and their lands. The names thus given at the outset may be taken therefore to contain some of the most ancient forms of the Indo-European speech. And once given, they have in many, if not in most cases remained to the present day, for nothing affords such strong resistance to change as the name of a river. The smaller streams, variously called in England and Scotland brooks, becks, or burns, whose course extended but for a few miles, and whose shores were portioned out among but a few settlers, readily yielded up their ancient names at the bidding of their new masters. But the river that flowed past, coming they knew not whence, and going they knew not whither--upon whose shores might be hundreds of settlers as well as themselves, and all as much entitled to give it a name as they--was naturally, as a matter of common convenience, allowed to retain its original appellation. Nevertheless, it might happen that a river such as the Danube, which runs more than a thousand miles as the crow flies--being divided between two great and perfectly distinct races, might, as it passed through the two different countries, be called by two different names. So we find that while in its upper part it was called the Danube, in its lower part it was known as the Ister--the former, says Zeuss (_Die Deutschen_), being its Celtic, and the latter its Thracian name. So the Saone also was anciently known both as the Arar and the Sauconna--the latter, according to Zeuss, being its Celtic name. And Latham, (_Tacitus_, _Germania_,) makes a similar suggestion respecting the Rhine--"It is not likely that the Batavians of Holland, and the Helvetians of Switzerland, gave the same name to the very different parts of their common river." It does not follow then as a matter of course--though we must accept it as the general rule--that the name by which a river is known at the present day, when it happens to be different from that recorded in history, is in all cases the less ancient of the two. There might originally have been two names, one of which has been preserved in history, and the other retained in modern use. It is also to be observed, that in the case of one race coming after another--say Germans or Slaves after Celts--while the newcomers retained the old names, they yet often added a word of their own signifying water or river. The result is that many names are compounded of two words of different languages, and in not a few cases both signifying water.