It was solemnly maintained by the Oxford School and its German masters that these great Roman towns, one after the other, were first utterly destroyed by the Pirates of the North Sea, then left in ruins for generations, and then _re-occupied_ through some sudden whim by the newcomers! It needs no historical learning to laugh at such a fancy; but historical learning makes it even more impossible than it is laughable. Certain rare towns, of course, decayed in the course of centuries: the same is true, for that matter, of Spain and Gaul and Italy. Some few here (as many in Spain, in Gaul and in Italy) may have been actually destroyed in the act of war. There is tradition of something of the sort at Pevensey (the old port of Anderida in Sussex) and for some time a forgery lent the same distinction to Wroxeter under the Wrekin. A great number of towns again (as in every other province of the Empire) naturally diminished with the effect of time. Dorchester on the Thames, for instance, seems to have been quite a large place for centuries after the first troubles with the pirates, though today it is only a village; but it did not decay as the result of war. Sundry small towns became smaller still, some few sank to hamlets as generation after generation of change passed over them: but we find just the same thing in Picardy in the Roussillon, in Lombardy and in Aquitaine. What did _not_ happen in Britain was a subversion of the Roman municipal system. Again, the unwalled settlement outside the walled town often grew at the expense of the municipality within the walls. I have given Huntingdon as an example of this; and there is St. Albans, and Cambridge. But these also have their parallels in every other province of the West. Even in distant Africa you find exactly the same thing. You find it in the northern suburb of Roman Paris itself. That suburb turns into the head of the mediæval town—yet Paris is perhaps the best example of Roman continuity in all Europe. The seaports naturally changed in character and often in actual site, especially upon the flat, and therefore changeable, eastern shores—and that is exactly what you find in similar circumstances throughout the tidal waters of the Continent. There is not the shadow or the trace of any widespread destruction of the Roman towns in Britain. On the contrary there is, as much or more than elsewhere in the Empire, the obvious fact of their survival. The phenomenon is the more remarkable when we consider first that the names of Roman towns given above do not pretend to be a complete list (one may add immediately from memory the southern Dorchester, Dover, Doncaster, etc.), and, secondly, that we have but a most imperfect list remaining of the towns in Roman Britain.