Bibliot. Arabico-Hispana, tom. ii. p. 104. Casiri translates the original testimony of the historian Rasis, as it is alleged in the Arabic Biographia Hispanica, pars ix. But I am most exceedingly surprised at the address, Principibus caeterisque Christianis, Hispanis suis Castellae. The name of Castellae was unknown in the 8th century; the kingdom was not erected till the year 1022, a hundred years after the time of Rasis (Bibliot. tom. ii. p. 330), and the appellation was always expressive, not of a tributary province, but of a line of castles independent of the Moorish yoke (D'Anville, Etats de l'Europe, p. 166 - 170). Had Casiri been a critic, he would have cleared a difficulty, perhaps of his own making.