We have considered the relative position of the unfree holdings, of the domanial land around which they were grouped, and of the class which had to put the whole machinery of the manor into action. But incidentally we had several times to notice a set of men and tenements which stood in a peculiar relation to the arrangement we have been describing: there were in almost every manor some free tenants and some free tenements that could not be considered as belonging to the regular fabric of the whole. They had to pay rents or even to perform labour services, but their obligations were subsidiary to the work of the customary tenants on which the husbandry of the manorial demesne leaned for support. From the economic point of view we can see no inherent necessity for the connexion of these particular free tenements with that particular manorial unit. The rent, large or small, could have been sent directly to the lord's household, or paid in some other manor without any perceptible alteration in favour of either party; the work, if there was such to perform, was without exception of a rather trifling kind, and could have been easily dispensed with and commuted for money. Several reasons may be thought of to explain the fact that free tenements are thus grouped along with the villain holdings and worked into that single unit, the manor. It may be urged that the division into manors is not merely and perhaps not chiefly an economic one, but that it reflects a certain political organisation, which had to deal with and to class free tenants as well as servile people. It may be conjectured that even from the economic point of view, although the case of free tenants would hardly have called the manorial unit into existence, it was convenient to use that class when once created for the grouping of villain land and work: why should the free tenants not join the divisions formed for another purpose but locally within easy reach and therefore conveniently situated for such intercourse with the lord as was rendered necessary by the character of the tenement? Again, the grouping of free tenants may have originated in a time when the connexion with the whole was felt more strongly than in the feudal period; it may possibly go back to a community which had nothing or little to do with subjection, and in which the free landowners joined for mutual support and organisation. It is not impossible to assume, on the other hand, that in many cases the free tenant was left in the manorial group because he had begun by being an unfree and therefore a necessary member of it. All such suppositions seem _prima facie_ admissible and reasonable enough, and at the same time it is clear, that by deciding in favour of one of them or by the relative importance assigned to each we shall very materially influence the solution of interesting historical problems.