140 DUFFIELD CASTLE. 



radiating from London over the surface of the island, was com- 

 pleted in its main features. Some of the cross roads were made 

 at a later date ; but there can be no doubt that a road leading 

 from Wirksworth and its lead mines (which were long before then 

 in the hands of the conquering forces) to that great artery, the 

 Rykneld Street, for the Southern ports, was at that time con- 

 structed. 



The bed of a river, especially one so near to high ranges, and 

 so subject to sudden floods as the Derwent, is liable to many and 

 remarkable changes when utilised for various commercial purposes. 

 Dams and weirs at varying places materially affect its flow, and 

 bring about complete changes in fords. Thus, before the cotton 

 mills of Mil ford were erected, there was a ford at the upper side 

 of Milford, now altogether indistinguishable. Some eighteen 

 centuries ago, the Romans first forded the Derwent on their way 

 to Wirksworth, and finding a fordable place (already perhaps dis- 

 covered and used by the Celtic tribe settled on the brow above it) 

 determined to make this their permanent passage. The great 

 lapse of time, the formation of new weirs at Milford and Little 

 Eaton, the removal of old fishing dams in other places, the 

 embanking against floods, anil the prevalence of systematic drain- 

 age, have all combined to obliterate many of the traces of the 

 Roman ford, and to destroy the possibility of crossing at the place 

 where the Roman legionaries, and the slaves of the Roman lead 

 merchants were at one time so constantly passing. 



We have invented no plausible theory of a Roman ford at the 

 spot immediately opposite the castle, in order to fit in with the 

 now proved Roman settlement of such long duration above it ; 

 for, ten years before the unearthing of the castle, when resident 

 in the neighbourhood, we had marked this site as a Roman ford, 

 for the simple reason that the Roman cross-road to the Rykneld 

 Street seemed to run straight for it. Place-names, and other 

 indications, had pointed out that this road ran on high ground 

 from Wirksworth, dipped down from Knaves Cross into Black, 

 brook, and thence ascended to the ridge of the Chevin. An 

 unusually dry summer showed us, in what seemed then an unmis- 



