18 



In plainer words, I suspect my friend's etymology has gone astray, and 

 has carried off his topography on its back. My objections to his theory are 

 inj,nifold — negative as well as positive. 



First, I think that he has quite overlooked the evidence of Roman ideas 

 in the construction of Wall Hills camp. That castrametation, although simpler 

 in form than Risbury and Capler, not to mention the elaborate entrenchment 

 on the Herefordshire Beacon, is yet manifestly constructed on a Roman plan, 

 of which the right angle is the basis, and is of definite proportions, the length 

 being to the breadth as 3 to 1. The pool at the N.E. end for water, and the 

 fact that the site is so chosen as to allow its defenders to communicate by signal 

 with the Herefordshire Beacon on the one hand and with the Roman station at 

 Circutio on the other, are also manifest indications that it belonged to a time 

 when the Beacon heights were defended although the Roman station was still 

 standing. That there ever was a British town here is wholly unproved. 

 Wherever it is certain that British towns existed, we find their names still pre- 

 served more or less purely. Here there is none, and the inference is that there 

 never was any such town, although it may nevertheless have been, as my friend 

 thinks, "occupied" as a camp "up to the time of the latter Saxon invasion." 



Secondly, if we leave topography for etymology we are led to the same 

 result. The name, Wall Hills, is compounded of two words, only the latter 

 of which is undoubtedly Saxon, and as that is the generic term, and one of 

 frequent occurrence, nothing can be drawn from its use except that the 

 Saxons became at some time possessed of the hills, upon which they found 

 "walls." The inquiry must turn upon the specific word in the name, which 

 is wall. I think it will be found that this word always points out a site on 

 or near to a Roman work, whether a fortification or the road which that forti- 

 fication is meant to command. Wall's End (Northumberland), is the end of 

 the wall of Severus. Wal-ford (Hereford), Bower Walls (near Bristol), Wall 

 HiUs (near Thornbury), and those which my friend cites, at Suckley, Cradley, 

 Alfrick, Eastham, Orleton, &c., are all situated near Roman camps or roads. 

 If the word were Saxon, we ought to find it most commouly in the Saxon 

 districts, and not, as my friend himself remarks, most frequently in the very 

 strongly British district of Herefordshire, and the frontier Angle and Wiccii 

 land of West Worcestershire. Yet I know of no single example in Wessex to 

 set against the many in the marches. 



3. I derive the word toall through the Britons from vallum. It should 

 be remembered that the Romans came first in contact with the Belgse of S.E. 

 England, whose posterity (the cockneys) still sound the v as w. Phis may not 

 have been a blunder in Roman times. I find the Romans, when adapting into 

 their tongue British words commencing with gw — which according to rule in 

 some cases drops its initial — representing the lu by a v. Thus Gwrtheym and 

 Gwrthefyr appear in Latin as Vortigem and Vortimer. 



The proper Saxon word is die, or (as now spelt) dike, which is still used 

 in Yorkshire and in lowland Scotland to mean a stone wall. 



