Next, we may presume (for evidence is lost, especially under the later Roman work) there was a track towards the centre of Norfolk. Next there was some great road going northward east of the Pennines, following the dry land which skirts the Fens and reaching the great fertile plain of York, and so on northward through Durham up to the crossings of the Tyne. Where this original main track went we cannot say. We know the trace of the Roman road which followed it. We may presume that the divagations and modifications of this road of the Dark Ages and the Middle Ages, which ultimately built up our main road to the north, reverted in some degree to the original track. But the whole thing is guess-work. One thing seems fairly certain: this eastern road to the north (the twin to the great north-western road by Chester to Lancashire) must have split about half way to York, one branch making directly to the plain of York itself, the other obviously running along the inevitable ridge which points right north through Lincolnshire to the Humber. There is here no bridge possible. It is not too broad for a ferry. But though the Roman road, superseding the earlier trackway, went on northward, it is a fair guess that the original trackway stopped at the river. Of cross roads we have fragments, of course, in the Pennines, but we know nothing of their history. It is clear that the main cross communications between the peopled area of the Yorkshire Plain and that of the Lancashire Plain must have gone over by Shipley--the obvious gap in the chain. But more we cannot tell. That is the natural way, and there was, so to speak, no avoiding it. What was mainly used further south we cannot tell. It was a tangled land. There is no clear and certain trace of cross communications which must have existed across the Midlands south of Trent. We do not know what great patches of wood may here have determined the windings of an original road. There are no serious obstacles (it is high land and dry, with no marshes or large watercourses), but there was less reason for continual traffic here from east to west than there was for traffic from north to south; therefore there was less “potential” than was created by the traffic on cross communications further south.