14 An Elizabethan Sui^fcy and Domesday Book. [ch. 



were 20 free and soiled and 18 servile. Thus the percentage of 

 originally servile homesteads was 47*4. 



Not counting the two manor sites, the rectory, or the town and 

 gild houses, there were mentioned in the Survey in both parishes 

 together 108 messuages, including 27 vacant and 81 built. Omitting 

 the two doubtful cases, the servile messuages were 48 per cent, 

 of the total number. 



Turning to Domesday Book, we find that a rough correspondence^ 

 exists between the relative numbers of freemen and sokemen on the 

 one hand, and of villeins, bordiers and slaves on the other hand, 

 mentioned in connection with a given hamlet in 1086, and the relative 

 numbers of free and servile messuages in the same locality as indicated 

 in the Survey of 1565. This appears from the following tables : 



1565. Messuages. 



Total, 115 persons. Per cent, of unfree, 42*6. 



The tables show that in 1086 there was a large proportion of unfree 

 persons in the parish of St Mary's, and a large proportion of freemen 

 in Twanton, especially (as the Survey of 1565 indicates) in East 

 Twanton. Now it is just in this eastern part of Twanton that the 

 houses, instead of being in comparatively closely-built rows, as in 

 other parts of Forncett, are scattered either in very small loose 

 clusters near Twanton, or Middleton, or Moorgate or Southgate 

 common, or are quite apart and isolated. 



1 For many reasons vi^e should not expect to find the correspondence exact, for some of 

 the sokemen of ioi^6 probably held bond land. See below, p. 12, n. 2. 



2 It is perhaps permissible to use the term * unfree ' as a convenient one, under which to 

 include villeins, bordiers and slaves, in contrast to * liberi homines ' and * sochemanni.' Cf. 

 Maitland, Domesday, 24-79. ^^ is not of course to be understood to imply that the villeins 

 of the eleventh century were serfs. 



