There is no record as to when the Norman church was built, but Robert de Lymesey, the bishop, is said in 1088 to have used a large quantity of silver, which he took from the church at Coventry, in extensive buildings in Lichfield; and Roger de Clinton is declared to have exalted the church as well in building as in honour, so he may have erected, or helped to erect, the Norman cathedral. Nothing whatever remains above ground of this building, which was probably taken down gradually while the cathedral which now stands in its place was being erected. Before this was done, however, there was added a rectangular chapel to the east of the Norman apsidal choir, which, with it, must have extended nearly to the end of the seventh bay of the present choir. Nothing is known of it beyond the fact that the foundations were discovered and examined by Professor Willis, who decided that it was probably built late in the twelfth century, and that its existence, if it was ever finished, must have been short. Very early in the thirteenth century the first part of the present building was begun by erecting a rectangular choir just outside the walls of the Norman choir, which must have been then removed. This new Early English choir (including the presbytery) extended from the central tower to the end of the seventh bay of the present choir. It will be found that the eastern portion of this was subsequently removed, but the western half still remains, and can readily be distinguished from the Decorated part. The high altar of this period must have stood just to the west of the space between the fifth piers, thus leaving the space between the fifth and sixth piers as a processional path between the two side aisles; while against the eastern wall were four altars, one at the end of each aisle and two between them. At the same time that this choir was built, was also built on the south side the sacristy, with the room adjoining it: these both remain. The next alteration took place about 1220, and was the erection of a new south transept in place of the Norman one, which, as has been said, almost certainly had no aisle; then followed in about twenty years the north transept and the chapter-house, with its vestibule, all these buildings being in the Early English style, though the difference in their times of erection is clearly marked in their details. Two royal licences to dig Hopwas stone for "the new fabric of the church at Lichfield" in 1235 and 1238 are evidence that work was going on about this time; and the desire of Henry III. (which is more fully set out in the description of the interior of the transept) to have a roof at Windsor like that of Lichfield gives an almost documental certainty to the architectural theory that the transepts were built at the time above stated.