^{page 543 note 1} Archaeological Journal, lxiii. 106Google Scholar.

^{page 543 note 2} Ibid., liii.

^{page 543 note 3} Archaeologia, lxii. 81Google Scholar.

^{page 544 note 1} Victoria History of the Counties of England: Lincolnshire, vol. ii, p. 9Google Scholar, where references to the authorities will be found. See also The Architectural History of Lincoln Minster, by the Rev. Poole, George Ayliffe (Associated Architectural Societies' Reports, iv), Appendix of Documents, p. 36Google Scholar; and Recorded History of Lincoln Cathedral, by the Rev. J. F. Dimock (ibid., lx. 190).

^{page 544 note 2} Dugdale, , Mon. Angl. (1830 ed.), vi. 1270Google Scholar; Bradshaw and Wordsworth, Lincoln Cathedral Statutes, part ii, p. 1.

^{page 544 note 3} The outer face of the apse must have been only some 12 to 15 yards within the Roman wall.

^{page 544 note 4} Henry of Huntingdon, Hist. Angl. (Rolls Series), 212.

^{page 544 note 5} Giraldus Cambrensis, Vita S. Remigii (Rolls Series), vii. 19; John de Schalby (in the same volume), 194.

^{page 545 note 1} Giraldus Cambrensis, ibid., vii. 21, and preface, p. xix. ‘… sepultus est a fratribus in eadem ecclesia, in prospectu altaris sanctae crucis.’ (Ibid., vii. 22.)

^{page 545 note 2} ‘Cum ecclesiam Lincolniensem iam perfectam dedicaturus esset.’ Henry of Huntingdon, Hist. Angl. (Rolls Series), 216.

^{page 545 note 3} The western towers, however, do not appear to have been carried up higher than the level at which the original work now finishes, just below the twelfth-century arcade which runs across the west front immediately above the arches of the great lateral recesses.

^{page 545 note 4} We hope to make this the subject of a separate study when our investigation has been completed.

^{page 545 note 5} The late Precentor Venables records that these were discovered by Mr. Willson, T. J. in 1852 (Archaeological Journal, xliv. 196)Google Scholar. A conjectural sketch-plan, based on these remains at the east and west ends, was published by the Rev. Poole, George Ayliffe in illustration of his paper, The Architectural History of Lincoln Minster (Associated Architectural Societies' Reports, iv. 11)Google Scholar, but he, and Precentor Venables following him (Archaeological Journal, xl. 173Google Scholar), assumed a transept of the same width as the later transept, and a choir of only two bays in front of the apse.

^{page 546 note 1} From these and from some indications of smaller finds kindly given me by Mr. John Allan, the cathedral clerk of works, and Mr. Henry J. Davis, the master-mason, I made a conjectural plan for the Lincoln meeting of the Royal Archaeological Institute in July, 1909, which has proved to be not very far from the truth.

^{page 547 note 1} All the foundations shown are from my own measurements, except the north-west angle of the north transept where my plan shows the lines which are marked on the floor.

^{page 547 note 2} On pl. LXXVI, the existing plan of the west end, beyond the doorways to the Morning Chapel and Consistory Court, is drawn in detail. Actually existing eleventh-century work or (as regards those parts which have been cased) parts which may be presumed to be of eleventh-century date are distinguished by dark hatching. All later work, mediaeval or modern, is shown by dotted shading, without distinction of date.

^{page 547 note 3} Mr. Ayliffe Poole's paper contains a plan of these fragments (Associated Architectural Societies' Reports, iv. ii). A better plan by Mr. J. J. Smith illustrates Precentor Venables' paper in the Archaeological Journal, xliv. 194Google Scholar (plan no. 1), and was also published in The Builder, lii. (May 21, 1887) 755Google Scholar. It is curious that both plans are inaccurate; the former shows only two projections between the choir and apse, and the latter shows two on the north and three on the south. The fragments are also shown, again incorrectly, on a plan of the choir in Archaeologia, xlvii, p. 44, pl. iiGoogle Scholar.

^{page 547 note 4} In the original masonry bared during these excavations, the joints are generally from ¾ to 1 in. in thickness.

^{page 547 note 5} This is the projection which is omitted in Mr. J. J. Smith's plan.

^{page 548 note 1} These dimensions are as accurate as I have been able to fix them by careful measurement, but the hatches in the floor of the stalls, by which alone access is gained to these remains, are some distance to the west of the apse itself.

^{page 549 note 1} Those at u and v had already been discovered in 1900, during the construction of a wind-trunk for the organ, but they had not been marked on the pavement.

^{page 549 note 2} The floor level of the north transept is the same as that of the nave.

^{page 550 note 1} It should be remarked that, whereas the other chamfered plinths found at T, U, and v are at the same level, which is also about the level of the chamfered plinth on the outside of the west front, the level of this plinth at Á is about 1 ft. lower.

^{page 551 note 1} du Gard, Roger Martin, L'abbaye de Jumièges (Montdidier, 1909), p. 74 and pl. iiiGoogle Scholar.

^{page 551 note 2} Cathédralc de Bayeux: Reprise en sous-œuvre de la Tour Centrale par M. Flachat. Description des travaux par MM. H. de Dion et L. Lasvignes (Paris, 1861Google Scholar).

^{page 551 note 3} So also in the western bays of the north aisle.

^{page 551 note 4} The top of the remaining masonry bared in these western bays is 9½ to 10 in. below the level of the floor.

^{page 552 note 1} Mr. Allan had previously told me that Ý, Ź had been bared 1903 for the late Mr. J. J. Smith, formerly clerk of works to the cathedral under the late Mr. J. L. Pearson, R.A.

^{page 552 note 2} The heights from the top of the remaining masonry to the floor level are, at v′, 9 in.; at z′, 1 ft.; and at Ý, Ź, 1 ft. 11 in.

^{page 553 note 1} On pl. LXXVII my plan of the eleventh-century church is reproduced in red, over the plan of the existing church reduced from the admirable plan made by Mr. E. J. Willson, F.S.A., in the thirties of the last century, and now in the possession of the Society.

^{page 553 note 2} Using the term ‘Norman ’ in its proper sense, and not as a nickname for a ‘period’.

^{page 553 note 3} Porée, Chanoine, L'église abbatiale de Bernay, in the volume of the Congres archéologique de France tenu à Caen, 1908, pp. 588–614Google Scholar.

^{page 553 note 4} Roger Martin du Gard, , L'abbaye de Jumièges, pp. 31–2.Google Scholar

^{page 553 note 3} Gout, Paul, Le Mont-Saint-Michel (Paris, 1910)Google Scholar.

^{page 553 note 4} Bouet, G., Analyse architecturale de l'abbaye de Saint-Étienne de Caen (Caen, 1868Google Scholar), and in the Bulletin Monumental, vols. xxxi and xxxiiGoogle Scholar; Serbat, L., in the volume of the Congres archéologique de France tenu a Caen, 1908, pp. 21–50Google Scholar.

^{page 554 note 1} It is not, of course, suggested that any of these three churches influenced the plan of Lincoln. They are introduced simply as later illustrations of the ‘type’ of Saint-Etienne, Caen—the type which Lincoln so closely followed.

^{page 554 note 2} Rhein, Andrè, L'eglise abbatiale de Cerisy-la-Forêt, in the volume of the Congrès archéologique de France tenu à Caen, 1908, pp, 545–87Google Scholar.

^{page 554 note 3} Lefèvre-Pontalis, E., in the volume of the Congrès de Caen, pp. 242–6.Google Scholar

^{page 554 note 4} Besnard, A., Monographie de l'église et de l'abbaye Saint-Georges-de-Boscherville (Paris, 1899)Google Scholar.

^{page 554 note 5} The Church of Edward the Confessor at Westminster, by the Very Rev. Robinson, J. Armitage, D.D., F.S.A., in Archaeologia, lxii. 81–96Google Scholar.

^{page 554 note 6} Willis, R., The Architectural History of Canterbury Cathedral (London, 1845)Google Scholar.

^{page 554 note 7} Evans, Sebastian, jun., Excavations at St. Augustine's Abbey, Canterbury, in Archaeologia Cantiana, xxvi. 1Google Scholar; Archaeological Journal, lxiii. 106Google Scholar, with plan by Mr. Harold Brakspear, F.S.A.

^{page 554 note 8} This, though probable enough, is not absolutely certain of Mont-Saint-Michel, Saint-Étienne, Caen, or Canterbury Cathedral. The precise plan of the terminations of the choir aisles of Jumièges is also uncertain (see Archaeologia, lxii. 96Google Scholar), and the plan of those of Westminster is an inference only. Sainte-Trinité, Caen, also had three parallel apses, but the plan was not quite of the normal type.

^{page 555 note 1} In the Archaeological Journal, liii. 10Google Scholar, I stated (following the published plans) that the choir aisles of Bernay finished square, both externally and internally. This is a mistake. In August, 1910, I had the pleasure of collaborating with M. le Chanoine Poree in the excavation of the end of the south choir aisle, which was found to be apsidal both externally and internally. My plan will be published in a forthcoming number of the Bulletin Monumental.

^{page 555 note 2} Plans of all these east ends (except, of course, Lincoln) are illustrated in my paper in the Archaeological Journal, liii. 17, pl. iiiGoogle Scholar. I may take this opportunity of correcting another mistake in this paper. The plans of Cerisy-la-Forêt and Lessay show the ends of the choir aisles as they now exist, but there can be no doubt that both have been altered from their original form, which was apsidal internally.

^{page 555 note 3} This slight excess of the width of the choir over that of the transept and nave was probably due to something in the architectural disposition of the side walls of the choir. At Cerisy the choir is 2 ft. wider than the nave, but this is due to the arcaded treatment of the side walls of the choir (see Congrès de Caen, pl. opp. p. 566). The remains at Lincoln, however, do not suggest this particular arrangement.

^{page 555 note 4} St. Albans also had a width of about 31 ft.

^{page 555 note 5} This is based on the ascertained width of the north choir aisle.

^{page 556 note 1} According to Professor Willis, (op. cit, p. 65).Google Scholar

^{page 556 note 2} The choir roof would doubtless finish eastward with a gable over the arch opening into the apse, and the roof of the apse itself would abut against this gable at a lower level, as at Cerisy, Lessay, and Saint-Georges-de-Boscherville.

^{page 556 note 3} Pugin, and Le Keux, , Specimens of the Architectural Antiquities of Normandy (London, 1827)Google Scholar, pls. xviii, xix.

^{page 556 note 4} Besnard, A., op. cit, p. 58 (coupe longitudinale).Google Scholar

^{page 556 note 5} At Saint-Georges-de-Boscherville the apse is covered with a ribbed semi-dome.

^{page 556 note 6} At Cerisy there is a second shaft to the west, but this receives the arch of the triforium arcade of the straight bays of the choir (Congres de Caen, elevation opp. p. 566).

^{page 556 note 7} So also at St. Albans (according to Buckler's plan) and Peterborough. So also in the ambulatory plans of Winchester and Norwich.

^{page 557 note 1} The apse of Cerisy seems to have been set out in this manner.

^{page 557 note 2} The standard thickness of 4 ft. 9 in., the depth of an external wall-arcade, and the depth of the internal arcade of Cerisy, make up the actual foundation thickness of 7 ft. 4 in. of the Lincoln apse.

^{page 557 note 3} Pugin, and Le Keux, , op. cit., pls. xviii, xix.Google Scholar

^{page 557 note 4} Besnard, A., op. cit., pls. opp. pp. 58 and 66.Google Scholar

^{page 557 note 5} Photograph in Congrès de Caen, opp. p. 570.

^{page 557 note 6} This is about 1 ft. more than the length of the choir of Cerisy, which has only two bays, the bay-width at Lincoln being much less.

^{page 557 note 7} Bouet, G., op. cit., p. 18.Google Scholar The place indicated is now inaccessible without removing the roof covering, but Bouet's record of facts (from which it is sometimes necessary to distinguish his theories) so accurate in respect of this church that his statement may be accepted for the present, in place of the current surmise that the choir of Saint-Étienne was only two bays in length.

^{page 558 note 1} This explanation is favoured by many French archaeologists. See, for example, R. Martin du Gard, L'abbaye de Jumièges, p. 109 seq. Cf. de Lasteyrie, R., L'église de Saint-Philbert-de-Grandlieu (Mémoires de l'Académie des Inscriptions et Belles-Letires, xxxviii), p. 69Google Scholar.

^{page 558 note 2} There is some evidence that the builders of the nave of Saint-Étienne, Caen, thought of a groined vault over the double bay, though the intention was abandoned.

^{page 558 note 3} Archaeologia, lxii. 98Google Scholar.

^{page 558 note 4} So at Saint-Nicolas, Caen.

^{page 558 note 5} Mr. Arthur G. Wallace gives me the internal length of the transept of Jumièges as 123 ft. 1 in.

^{page 559 note 1} Professor Willis (op. cit., 65) gives the transept length as 127 ft. As this did not agree with my own measurements (which were not taken specially for this purpose), M. Ferdinand Huard, the Caen architect, has very kindly taken this and some other measurements of Saint-étienne purposely for me.

^{page 559 note 2} Op. cit., 65.

^{page 559 note 3} At Cerisy (according to the plan in Ruprich-Robert, L'architecture normande, pl. liv, fig. 7) the lengths of the transept arms are about the same as the length of the choir to the corresponding point.

^{page 559 note 4} This arch would, of course, be immediately under the vault of the gallery.

^{page 559 note 5} At Caen this opening is 14 ft. gin. wide; at Lincoln it is 15 ft. 5 in. wide.

^{page 560 note 1} The continuation of the inner face of the choir aisle wall at Q in relation to the angle of the pier at the west end of this aisle, found at F′ (plate LXXV), indicates a projection for this pier which is not enough for the three shafts which are found in this position, and also on the backs of the crossing piers, at Saint-Étienne, Caen. Following this indication, I have shown all these piers (plate LXXVII) with two shafts only, without the central shaft.

^{page 560 note 2} Demaison, L., Date de l'éqlise de Saint-Remi de Reims in Travaux de l'Academic de Reims, lxxi. (1883) 298–308Google Scholar.

^{page 560 note 3} Lefèvre-Pontalis, E. and Jarry, E., La cathédrale romane d'Orléans, in the Bulletin Monumental, lxviii. 372Google Scholar.

^{page 560 note 4} Jumièges, Bayeux, Saint-Étienne and Saint-Nicolas, Caen, Cerisy, Saint-Georges-de-Boscherville, Canterbury, Winchester, and Ely.

^{page 560 note 5} The plan can only be interpreted in this manner, though it may be objected that the vaulting of such excessively oblong bays would present serious difficulty. The difficulty was met in many of these early unribbed vaults by the employment of a low segmental curve for the arches, of longer span, while the shorter spans were covered by semicircular arches, perhaps stilted. The segmental curve for the arch from north to south under this gallery at Lincoln, developed in this manner, would indeed be very low, but it would be very little lower than that of some of the arches in the central crypt of Winchester Cathedral, and not so low as the curve of the upper part of the outer arch between the ambulatory and the eastern chapel of this crypt; and the curve would be very similar to that of the diagonal ribs of the vaults of the choir aisles of Durham (for these, see Journal of the Royal Institute of British Architects, 3rd ser., vi. 298, fig. 10Google Scholar).

^{page 561 note 1} As at Saint-Georges-de-Boscherville. At Winchester the cylindrical piers have a pilaster and half-shaft on the back to receive the aisle vault. In both these cases, the gallery only extends over the extreme bay of each arm of the transept.

^{page 561 note 2} The shaft which rises from the abacus of the gallery pier at Winchester, and stops abruptly at the gallery floor, cannot be explained in this manner, and it would seem to be merely decorative.

^{page 561 note 3} Congrés de Caen, pp. 145 and 154.

^{page 561 note 4} Mr. C. R. Peers has found evidence that at Christchurch (Hampshire) the transept galleries extended up to the crossing piers. His account of this church will be found in the forthcoming vol. v for Hampshire in the Victoria County History.

^{page 561 note 5} The thickness of the wall at the arcade piers of the nave of Saint-Étienne, Caen, is (as stated above) 4 ft. 8–9 in., and the thickness at the crossing piers is about 5 ft. 2 in.

^{page 561 note 6} The total internal length of eleventh-century Lincoln, from the inner face of the apse to the inside of the west wall (actually measured to the back of the fourteenth-century wall-arcade on the west wall), was about 310 ft. The total external length, from the outer face of the apse to the outer face of the west front, was about 334 ft. 4 in.

^{page 562 note 1} i. e. the third to the eighth bay, inclusive.

^{page 562 note 2} This point, one bay west of the crossing, is a likely place for a slight change of this kind.

^{page 562 note 3} They are so drawn on my plan, except as regards the wall-shafts next the nave, which at Caen are alternately simple half-shafts, and half-shafts on the face of pilasters. For the plan of the Caen piers, see Journal of the Royal Institute of British Architects, 3rd sen, vi. 291Google Scholar, fig. 2, iv.

^{page 562 note 4} The two westernmost bays of the present nave are narrower than those eastward, but the difference is very much greater than that indicated by the second bay on the north side for the eleventh-century work.

^{page 563 note 1} I have been obliged to ignore the foundations found at Ý, Ź, of which I frankly confess I cannot offer any satisfactory explanation.

^{page 563 note 2} See the plans in Pugin and Le Keux, op. cit, pls. vii and xxiv.

^{page 563 note 3} The exact width between the lines of these two jambs on the west front is 38 ft. 7½ in. The width of the nave, as indicated by the existing remains on either side of the bay between the towers, was 28ft. 9in.; adding to this two walls of the assumed thickness of 4 ft. 9 in. each, we have a total width to the outside of the nave walls of 38 ft. 3 in.