How can we account for the occurrence of regular 'shareholding' among the freeholders? Two possibilities have to be considered: the free character of the tenements may be newly acquired and the 'shareholding' may be a relic of a servile past; or, on the other hand, the freehold character of the tenements may be coeval with the 'shareholding,' and in this latter case we shall have to admit the existence of freeholds which from of old have formed an element in the village community. In the first of these cases again we shall have to distinguish between two suppositions:--Servile tenements have become free; this may be due either to some general measure of enfranchisement, a lord having preferred to take money rents in lieu of the old labour services, and these money rents being the modern equivalent for those old services, or else to particular and occasional feoffments made in favour of those who, for one reason or another, have earned some benefit at the lord's hand. To put it shortly, we may explain the phenomenon either by a process of commutation such as that which turned 'workland' into 'molland,' or by special privileges which have exempted certain shares in the land from a general scheme of villainage; or, lastly, by the existence of freeholds as normal factors in the ancient village community. Let us test these various suppositions by the facts recorded in our surveys. At first sight it may seem possible to account for the freehold virgates by reference to the process which converted 'workland' into 'molland.' We have seen above that if a lord began to demand money instead of work, the result might, in some cases, be the evolution of new tenures which gradually lost their villain character and became recognised as genuine freeholds. And no doubt one considerable class of cases can be explained by this process. But a great many instances seem to call for some other explanation. To begin with, the mere acceptance of rent in lieu of labour did not make the tenement a freehold; servile tenements were frequently put _ad censum_[748], and it seems difficult to believe that many lords allowed a commutation of labour for rent to have the effect of turning villainage into freehold. Another difficulty is found on the opposite side. What force kept the shares together when they had become free? Why did they not accumulate and disperse according to the chances of free development? It may be thought that custom, and express conditions of feoffment, must have acted against disruption. I do not deny the possibility, but I say that it is not easy to explain the very widely diffused phenomenon of free shareholding by a commutation which tended to break up the shares and to make them useless for the purposes of assessment. Still I grant that these considerations, though they should have some weight, are not decisive, and I insist chiefly on the following argument.