tower we have already seen what was done, and obviously it was one of the two towers that had fallen.  But what of the other of these?  What suggestions remain to show which it was?  It is well known that a central tower had been erected as part of the original plan, and also that a new upper part was being added to this same tower about the middle of the thirteenth century.  This new portion eventually rose above the roofs to the level of the top of the square parapet, about the base of the octagonal spire, the spire being a still later addition.  Now the heightening of this tower—­perhaps with already the idea of a future spire in view—­would raise many questions.  Experience would already have taught the builders that the early central towers of many other churches were incapable of carrying their own weight.  This being so, much less would it do to suppose that it could bear the addition of new weight upon the old piers; for though to all appearance sound, the cores were of rough rubble work, not solidly bedded and not properly bonded with the ashlar casing.  So the question arises, did they remove the whole or part of the old central tower and piers, or were they saved this trouble by the structure having shared the fate of many others like itself, which fell, and so made way for new work?  Another tower had fallen besides the one to which attention has already been drawn; and as there appears to be nothing to show that this other was the north-west tower, we must see what evidence there is concerning the central tower.  That it was added to we already know.  But documentary as well as structural evidence comes to our aid.  The first is supplied by the records of Bishop Neville’s episcopate; the next by the researches of modern archaeology.  Professor Willis has shown in his remarks upon the structure of the piers at the time of the collapse of the mediaeval tower and spire in 1861, that these had not been rebuilt at a date later than the twelfth century.  But Mr. Sharpe [6], writing to Professor Willis seven years before the occurrence, indicates his discovery—­from a close examination of the structure then existing—­that before the upper part of the central tower was rebuilt in the thirteenth century the earlier arches at the crossing which were to support it had been taken down, and probably a large part of the piers carrying them.  And that, though the twelfth-century voussoirs were re-used others of a fine grained stone were inserted among them to strengthen the arches, or as a substitute for some of the rougher sandstones that could not be used again.  By this means, then, the original form and detail of the twelfth-century arches was preserved, so that the drawings representing the measured studies of the building, which were Sir Gilbert Scott’s principal authority upon which to base his restoration of this portion of the tower, were made from work which had already been once rebuilt.  But why was this part of the church rebuilt, and by whom?  Two alternative suggestions for the reason have been offered.