Although Tavistock, in the first scutage, appears to have paid on the anomalous assessment of ten and a half knights its payment on fifteen in the two succeeding ones may fairly be taken as evidence that this was its _servitium debitum_.[65] Its abbot, however, made no reference to that _servitium_ in his return, and--by an exception to the regular practice in the case of church fiefs--we find him charged, not on the fees, (1) 'quos recognoscit', (2) 'quos non recognoscit', but on those which were enfeoffed 'de veteri', and 'de novo' just as if he were a lay tenant. As his fees 'de veteri' were sixteen, this figure recurs in successive scutages, until in 3 John we find him contesting as to one knight ('unde est contentio') who, doubtless, represented the difference between fifteen and sixteen. The case of Westminster presents considerable difficulty, the entries relating to its payments of scutage being very puzzling. The abbey's fees lay chiefly in Worcestershire and Gloucestershire--especially Worcestershire--and it is under this county that we find it ultimately (_i.e._ from 1168 onwards) assessed at fifteen fees, an assessment which the abbot himself seems to have claimed, in the first scutage, as the right one. Taking then the _servitium debitum_ of all the church fiefs, at their earliest ascertainable assessment, we obtain this result: Bishops 458-1/2 Heads of religious houses 318 Capellaria de Bosham 7-1/2 ------- Grand total 784[66] Far more difficult is the calculation of the _servitium debitum_ from the lay fiefs. The list which follows is constructed from the evidence of the _cartae_ and the rolls, and, though substantially correct, is liable to emendation in details. It only comprises those fiefs the _servitium_ of which I have been able to ascertain with certainty or probability.