(1) _Feodum militis integrum._ The origin of the knight’s fee is obscured by a network of conflicting theories. A thread of connection is sometimes traced between it and the mysterious five-hide unit of Anglo-Saxon times; other authorities would ascribe its introduction into England to a definite act of some great personage—either William the Conqueror, according to Selden, who founds on a well-known but untrustworthy passage in Ordericus Vitalis, or Ranulf Flambard, according to Freeman, Stubbs, and Gneist. It seems probable that the Normans, here as elsewhere, pursued their policy of avoiding an open rupture with the past, and that the Conqueror adapted as far as possible the existing system of land tenure to his own needs. There is little doubt, in light of the evidence accumulated by Mr. Round in his _Feudal England_, that William I. stipulated verbally for the service of a definite number of knights from every fief bestowed by him on his Norman followers. A knight’s fee or _scutum_ thus became a measure of military service, and of feudal assessment; _servitium unius militis_ was a well-known legal unit. But a difficult problem arises when it is asked what definite equation, if any, existed between land and service. Three answers have been given: (_a_) A definite ratio exists between amount of service and extent of ground. In other words, the knight’s fee contains a fixed area of land; every five hides sent one warrior, thus preserving the old Anglo-Saxon unit.[365] (_b_) The ratio lies not between service and extent, but between service and value. An estate of £20 annual rental sends one knight to the king’s wars; the normal knight’s fee contains 20 librates of land.[366] (_c_) Other authorities deny that any proportion exists at all: William the Conqueror exacted from each of his grantees precisely as much or as little knight’s service as he saw fit. ----- Footnote 365: C. Pearson, _Hist. of Engl._, I. 375, note 2. Footnote 366: J. H. Round, _Feudal England_, 295. -----