In Higden's Polychronicon Gothia is also spoken of as lower Scythia, but among the provinces of Asia, although it is said that it lies in Europe; it has on the north Dacia and the Northern Ocean. But the geographical confusion in this work is greater; as already mentioned (p. 31), the countries of the Scandinavians are described together with the Insulæ Fortunatæ, Wyntlandia, etc., as islands in the outer ocean. The disagreement between Higden's text and his map gives us an insight into how little weight was attached at that time to the relation between maps and reality; they are for the most part merely graphic schemes. Probably Higden's map was partly copied from an older one, and the desirability of bringing it into better agreement with his text did not occur to him.
The so-called "Anglo-Saxon mappamundi" or "Cottoniana" (reproduced vol. i. pp. 180, 183), which is in the British Museum, occupies a position of its own among early mediæval maps. Its age is uncertain; it may at the earliest date from the close of the tenth century, but possibly it is as late as the twelfth [cf. K. Miller, iii., 1895, p. 31]. It exhibits no agreement with the text of Priscian (Latin translation of Dionysius Periegetes, see vol. i. p. 114), to which it is appended. Many of the names might rather be derived from Orosius, there is also great resemblance to Mela (cf. vol. i. pp. 85, ff.), and in some ways to the mediæval maps already mentioned, although the representation of the North is different. Probably an older, perhaps Roman (?) map formed the basis of it. Name-forms like Island, Norweci[166] (Norwegia), Sleswic, Sclavi, may remind us of Adam of Bremen, but they may also be older. This map is doubtless less formal than the pronounced wheel-map type, but it does not bear a much greater resemblance to reality, although the form of Britain, for instance, may show an effort in that direction. The peninsula which has been given the name of Norweci (Norway) has most resemblance to Jutland, and the name seems to have been misplaced. No doubt it ought rather to have been attached to the long island lying to the north, which has been given the names Scridefinnas and Island. The representation has great resemblance to Edrisi's map (cf. p. 203), where Denmark forms a similar peninsula, and Norway a similar long island, with two smaller islands to the east of Denmark, which is also alike. The "Orcades Insule" are given a wide extension on the Cottoniana map, and Tyle (Thule) lies to the north-west of Britain, as it should do according to Orosius. This map does not therefore indicate, any more than the others, any particular increase of knowledge of the North, and compared with King Alfred's work it is still far behind in the dark ages.