THE CELTIC SYSTEM 1 89 



In a survey of 32 Henry VIII we are told the areas of the town- 

 lands, and learn in addition how many peasant households each 

 contained. Two descriptions run as follows: — 



" Villata de Balnestragh. WilHam Dyxson tenet scitum man- 

 erii vocati Balnestragh super quod edificantur duo castra . . . 

 terra dominica continet Ix acras terre arabiUs et i acram prati 

 iacentes in villata de Balnestragh. . . . 



" Et [dicunt] quod sunt infra eandem villatam vii messuagia 

 cxv acre iii rode terre arabilis x acre communis pasture ac ii acre 

 more in occupatione Donaldi O'Daylye, Donaldi Holloghan et 

 aliorum. ... Et sunt iii cottagia. . . . 



" Et [dicunt] quod sunt in Villata de Ballerayne vi messuagia 

 cxxxiii acre terre arabiHs xx acre communis pasture et xx acre 

 more quas Mauricius O'Nayry, clericus, Ricardus O'Morrye et 

 aUi tenentes ibidem occupant. ... Et sunt x cotagia. . . . " ^ 



From these instances we may conclude that Irish units of 

 settlement were in size much Kke Scottish townships. Their 

 areas, averaging from one hundred to two hundred acres, were 

 perhaps a Kttle greater and their tenants may have been a httle 

 more numerous. From the point of view of the EngHsh midlands, 

 however, both forms of settlement coalesced into what may be 

 called the Celtic type. Instead of the large village we find the 

 hamlet; instead of extensive arable fields, the restricted areas 

 of the farm, the townland, or the petty township. Wherever in 

 England hamlets and small townships appear as the prevaiHng 

 type of settlement, Celtic influence is to be suspected. Within 

 the three-field area we have already seen such, notably in the 

 border counties, Herefordshire and Shropshire. If, however, 

 Celtic influence determined the form of settlement and the size 

 of the townships there, it did not prevent the superposition of a 

 three-field system upon the arable. Since such a system was not 

 Celtic, a further effort should be made to determine what was its 

 Celtic correspondent. 



We have seen that a salient feature of the ancient agriculture 

 of Scotland and Wales was the intermixture of the parcels of the 

 tenants. Known in Scotland as runrig or rundale, this feature 



^ Rents, and Survs., Ro. 934. 



