The identification of places and of persons in Mr Waters' treatise shows extraordinary knowledge; but both Mr Eyton and Mr Waters had the provoking habit of making important assertions without giving their authority. I expressed a wish in the _Academy_, at the time, that Mr Waters would give us some clue as to his sources of information, but as he did not think fit to do so, we have to test his statements as best we can for ourselves. Now we learn from him on p. 36 that 'Walter fitz William', a tenant at South Willingham, was 'brother of Simon mentioned above', namely of 'Simon fitz William (ancestor to the Lords Kyme)'. This is impressive until we discover that the actual words in the survey (as indeed in Hearne's text) are 'Walt[erius] fil[ius] Walt[eri]i' (fo. 11 _b_).[7] To an expert such a test as this will prove significant enough. But to turn from an actual misreading of the text to cases in which are incorporated interlineations, not part of the original text, but written in later times, we find Mr Waters--like other antiquaries who had followed Hearne's text--stating that 'Ranulf [Meschin] is twice styled in the Roll Earl of Lincoln, but there is no record of his creation, and no other authority for possession of the earldom' (p. 8). The difficulty vanishes when we discover that this supposed style was a mere interlineation made by a much later hand.[8] So again we read on p. 30: Richard, Earl [of Chester], has 6 cars. in Barnetby-le-Wold, where [William], the constable of Chester, is his tenant [as his father was Earl Hugh's in Domesday]. But on turning to Mr Greenstreet's facsimiles, we find that the survey had nothing about 'the constable of Chester', the words 'constabularia [_sic_] Cestrie' being only a faint interlineation by a later hand. And even where a reference to the true text does not at once dispose of the matter, these statements of Mr Waters are, on other grounds, open at times to question. He assumes, for instance, that Hugh fitz Ranulf, who occurs as a landowner in the survey, was a younger son of Ranulf Meschin, afterwards Earl of Chester (p. 12). No such son would seem to be known; and this assumption, moreover, does violence to chronology. For the pedigree it involves is this: Roger (1) Lucy (2) Ranulf fitzGerold = = Meschin | |-------..... | | . William Ranulf, Hugh de Roumare, Earl of fitz Ranulf Earl of Lincoln Chester Now William de Roumare was not old enough to claim his inheritance from the King till 1122, and his half-brother, Ranulf, was some years younger than he was, as the words of Orderic imply in 1140. Consequently Hugh, the youngest brother, can have been only a boy in 1122. How then could he, as Mr Waters alleges, have held a fief in right of his wife so early as 1115 or thereabouts?