HENRY of Huntingdon’s Historia Anglorum, the earliest source of the wellknown and popular stories of Cnut and the waves and Henry I’s death from ‘a surfeit of lampreys’, enjoyed a wide circulation during the middle ages. It has survived reasonably complete in thirty-six known medieval manuscripts — none of them autograph — and was used by various medieval chroniclers, both during the author’s lifetime and later. The Historia received attention, too, in the sixteenth and seventeenth centuries, when several manuscript copies were made and parts of the text were first put into print. It was in 1596 that the narrative of the Historia was first published, by Sir Henry Savile, and in 1668 Luc d’Achery printed one of the portions omitted by Savile, the letter ‘De Contemptu Mundi’, which was to be re-edited in 1691 by Henry Wharton. Apart from Henry Petrie’s edition in 1848, which printed most of the narrative up to 1066, using Savile’s edition with fresh collations from four manuscripts, and Thomas Forester’s translation, published in 1353, for which a new collation was used in the final historical book, there was no other edition of the Histaria Anglorum until the Rolls series volume of 1879, and there has been none since.

The Rolls series edition, by Thomas Arnold (1823-1900) (second son of Dr Arnold of Rugby and younger brother of the poet Matthew Arnold), represented an enormous advance on the earlier work. Arnold described thirtyseven manuscripts, most of which had been usefully listed by Thomas Dyffus Hardy in 1865. Drawing together the evidence of a dozen or more of these, and relying heavily on the critical work of Felix Liebermann, Arnold laid out a scheme of the five ‘editions’ as he called them, through which, following Liebermann, he considered Henry’s text to have passed during the two and a half decades from 1129 to 1154. These ‘editions’ were seen by Liebermann and Arnold as belonging to 1129, 1135, 1139, 1145 and 1154.

Well as it has stood the test of time, Arnold’s edition has some severe inadequacies. In the first place, Arnold did not add to the corpus already in print. He printed only one of the four items in the book ‘De Summitatibus Rerum’ — this was the letter already printed by d’Achery and by Wharton, the ‘De Contemptu Mundi’.