These statements are taken from the Saxon Chronicle, Florence of Worcester, Simeon, and other authorities, under the years quoted. For the sake of illustration I have added in the Appendix a list of Anglosaxon towns, whose origin we have some means of tracing. ----- The evidences of Eádweard’s activity are yet more numerous. The following burhs or towns are recorded to have been built by him. In 913, the northern burh at Hertford, between the rivers Mimera, Benefica and Lea: a burh at Witham, and soon after another on the southern bank of the Lea. In 918, he constructed burhs, or fortresses, on both sides of the river at Buckingham. In 919 he raised the burh on the southern bank of the Ouse at Bedford. In 921 he fortified Towchester with a stone wall; and in the same year he rebuilt the burhs at Huntingdon and Colchester, and built the burh at Cledemouth. The following year he built the burh on the southern bank of the river at Stamford, and repaired the castle of Nottingham. In 923 he built a fortress at Thelwall, and repaired one at Manchester. In 924 he built another castle at Nottingham, on the south bank of the Trent, over against that which stood on the northern bank, and threw a bridge between them. Lastly he went to Bakewell in Derbyshire, where he built and garrisoned a burh. A large number of these were no doubt merely castles or fortresses, and some of them, we are told, received stipendiary garrisons, that is literally, king’s troops, contradistinguished on the one hand from the free landowners who might be called upon under the _hereban_ to take a turn of duty therein, and on the other from the unfree tenants, part of whose rent may have been paid in service behind the walls. But it is also certain that the shelter and protection of the castle often produced the town, and that in many cases the mere sutler’s camp, formed to supply the needs of the permanent garrison, expanded into a flourishing centre of commerce, guarded by the fortress, and nourished by the military road or the beneficent river. It is also probable enough that on many of their sites towns, or at least royal vills, had previously existed, and that the population whom war and its concomitant misery had dispossessed, returned to their ancient seats, when quiet seemed likely to be permanently restored. It cannot be doubted that those who were already congregated, or for the sake of security or gain did afterwards collect in such places, were subject to the authority of the burhgeréfa or castellan, and that thus the burh by degrees became a Palatium or Pfalz in the German sense of the word. In truth _burh_ does originally denote a castle, not a town; and the latter only comes to be designated by the word, because a town could hardly be conceived without a castle,—a circumstance which favours the account here given of their origin in general.