1921 REVIEWS OF BOOKS 447 as they necessarily are, a valuable guide to twelfth-century practice. Except where a reference to the British Museum Catalogue of Seals is thought sufficient or no sufficient indications remain, we are always told whether the seal is (or was) pendent by a strip or by a tag. The strip seal occurs only in fifteen or sixteen cases, and only three of these are later than 1160, though one of the exceptions (no. 343) is definitely dated after May 1188. A superior and inferior limit of date are often not ascer- tainable, but no. 511, marked 'late twelfth century', is shown not to be earlier than 1184 by the attestation of Kalph Fitz Simon of Ormesby, and the same is perhaps the case with no. 288. The direct light upon social and economic conditions derivable from charters is often disappointing in amount, nor are these above the average in that respect, but a practised eye will detect much that escapes ordinary attention, and Mr. Stenton's long and fruitful introduction is full of little additions to our knowledge of English society in the twelfth century in the eastern midlands. The cumulative result is to confirm generally the evidence of Domesday Book that the five counties in question preserved pre-Conquest conditions in a less modified form than either the northern or southern Danelaw. It is more than confirmed, indeed, for the almost complete absence of the word ' manor ' from these charters in itself proves that the compilers of Domesday applied a terminology to this district which was to some extent artificial. The region of the Five Boroughs was not, it is true, quite uniform in this respect, for Leicestershire with its preservation of the English agrarian unit of the virgate alongside the Scandinavian bovate and its more developed manorial economy had apparently more in common with Northamptonshire than with its northern neighbours, especially Lincolnshire, which was the most unchanged of the five counties. Even in Lincolnshire the manor was present in fact, if not always in name, but the outstanding features are the village mainly inhabited by substantial free peasants under more than one lord and so possessing no common organization but that of the township, and the soke, large or small, which might in some cases be called honour or manor without being quite the same thing as those names connoted elsewhere, though in the smaller sokes the differences tended to be minimized as time went on. The hundreds of twelve carucates survived, though with diminished functions. It is noteworthy that at least ninety of these hundreds in Lincolnshire were coincident with villages (p. lxvii). In a district where freemen were still so numerous the cash nexus was much more prominent than in more thoroughly feudalized parts of England. There are some indications that week- work was not always required from the villeins in this quarter. A serf and his family could, indeed, be sold without his tenement, but the fact that serfs did homage to their lords is a warning not to attach too much weight to later legal definitions of unfreedom, and justifies the inference that in the twelfth century ' the original independence of the inferior party had not yet passed beyond the limit of memory '. Mr. Stenton has some useful remarks on the charters which transfer the services of men who can be proved to have been freemen ' with formulae appropriate to the conveyance of serfs '. His instances could
