^{page 220 note 1} The main authorities are The Book of Glasgow Cathedral, ed. Eyre-Todd, George (Glasgow, 1898)Google Scholar; MacGibbon and Ross, Ecclesiastical Architecture of Scotland, ii, 160–203; and P. Macgregor Chalmers, The Cathedral Church of Glasgow (Bell's Cathedral Series, 1914). In the present century, apart from Chalmers's book, very little seems to have been published except for Watson's, T. L.Architectural History of Glasgow Cathedral (Glasgow, 1910)Google Scholar and The Double Choir of Glasgow Cathedral (Glasgow, 1901; 2nd ed. 1916). An official Guide by C. A. Ralegh Radford, to be published by H.M. Stationery Office, is in preparation.

^{page 220 note 2} The western towers were unfortunately demolished in the middle of the nineteenth century. Dunblane Cathedral, though in appearance comparably well preserved, had no roof to the nave for three centuries, and its present form owes much to extensive modern restorations.

^{page 220 note 3} Registrum Episcopatus Glasguensis (Maitland Club, Edinburgh, 1843)Google Scholar; Diocesan Registers of Glasgow (Grampian Club, 1875).

^{page 220 note 4} It may be noted that we now have a draft list of the dignitaries of the cathedral in Fasti Ecclesiae Scoticanae Medii Aevi (ed. D. E. R. Watt, St. Andrews, 1959). Mr. Norman F. Shead, of Glasgow, is preparing an edition of the Acta of the bishops down to 1258, which is apparently the first such collection attempted for any Scottish diocese.

^{page 220 note 5} Honeyman, John, The Age of Glasgow Cathedral (Glasgow, 1854)Google Scholar; ‘Notes on the Oldest Part of the Crypt of Glasgow Cathedral’ in Transactions of Glasgow Archaeological Society, N.S., i (1890), 5–12; and his remarks in Book of Glasgow Cathedral, pp. 228–9. Also Macgregor Chalmers, Cathedral Church of Glasgow, pp. 14–16; MacGibbon and Ross, Eccles. Arch. of Scotland, ii, 165–72; T. L. Watson, Architectural History, pp. 12–13, and Double Choir, pp. 22–35.

^{page 220 note 6} The charter cited on p. 221, n. 1 describes it as ‘[ecclesia] exilis et angusta’.

^{page 220 note 7} Chronicle of Holyrood (Scottish History Soc., 1938), pp. 30, 119.

^{page 221 note 1} Registrum Episcopatus Glasguensis (Maitland Club, 1843), i, 66–67. This charter of King William the Lion refers to the church of Glasgow as ‘in hiis diebus nostris igne consumpta’. The charter is undated, but on internal evidence it cannot be earlier than 1189.

^{page 221 note 2} ‘Jocelinus … cathedralem ecclesiam suam quam ipse novam construxerat, pridie nonas Julii die dominica anno episcopatus sui xxiiij dedicavit’ (Chronica de Mailros, Bannatyne Club, 1835, p. 103). Jocelin was a notable benefactor of the city, as well as of the church of Glasgow.

^{page 221 note 3} Especially the plinths and bases of the responds in the aisles, the lowest parts of the external walls and buttresses and the whole of the south door. See the drawings in MacGibbon and Ross, ii, 169–70.

^{page 221 note 4} Willelmus, dictus de Bondington … ecclesiam suam Glasguensem miro artificio lapideo aedificavit, et multis bonis ditavit et ornavitʼ (Scotichronicon, ed. Goodall, ii, 92).

^{page 221 note 5} In medieval times called ‘ecclesia inferior’; the term ‘crypt’ is rare. We speak here of an enlarged lower church; it is important to realize that Jocelin's eastern arm seems also to have had a two-storeyed arrangement, though a much less extensive one (see below, p. 223).

^{page 221 note 6} We are greatly indebted to H.M. Ministry of Public Building and Works, and in particular to Mr. S. H. Cruden, F.S.A., for providing facilities to examine and to photograph the remains, and for giving permission to reproduce official photographs. The Carnegie Trust for the Universities of Scotland has made a generous grant towards the cost of our colour block which is gratefully acknowledged; the block has been made from a colour transparency by Mr. George Alden, and is reproduced here by permission of the editor of The Scottish Field, who has also kindly supplied pl. lx. We are indebted to the Warburg Institute for pl. lxv a, the Courtauld Institute for pl. lxiv b, the National Buildings Record for pl. lxiv a, the Pilgrim Trust for pl. lxv b, and the Oxford University Press for pl. lxvi. Professor W. J. Smith, F.S.A., kindly supplied the lantern slide on which pl. lxii is based, as well as giving some essential information on matters within his experience. Mr. J. G. Paul, who at the time was a member of the staff of the Geography Department in Glasgow University, gave generously of his skill in preparing exact measurements of the voussoir, and provided the drawing for fig. i. We are grateful also for the advice and help of Mr. T. S. R. Boase, Dr. George Zarnecki, Dr. William Urry, Mr. Robert Cowper, Mr. Hugh Forbes, and Prebendary E. E. F. Walters. But none of those mentioned is responsible for the views expressed here.

^{page 221 note 7} Our list does not claim to be exhaustive; there may be other twelfth-century fragments among the stones preserved in the lower church, but there are none of comparable importance. The early tomb-slabs in the lower church have been excluded, as not being structural features. They deserve separate treatment on another occasion.

^{page 222 note 1} A measured plan is given in Watson, Double Choir of Glasgow Cathedral, 2nd ed., p. 24.

^{page 222 note 2} We refer to the stones situated south of the shaft. Similar tooling is found on the chamfered edge of the bench-table below, on the wall-face below the bench-table, on the plinth in situ, and on the detached plinth (it is clearly seen in pl. lxiii b). This last is certainly unrestored, and serves as a standard of comparison. There are a number of diagonally tooled stones in the vicinity which must belong to a later period of building. Whether they are re-used twelfth-century stones, or modern restorations, is uncertain.

^{page 222 note 3} A vertical section of the whole column is given in Honeyman, ‘Notes’ (ut supra) fig. 2. The scale is not stated, but is about ½ inch to 1 foot. The section in MacGibbon and Ross, ii, 172, is inaccurate.

^{page 222 note 4} Honeyman, ‘Notes’ (ut supra), p. 9; see also MacGibbon and Ross, ii, 170. Since the merits of his work have gone largely unrecognized, we should say here that Honeyman's writings show a rare knowledge of medieval archaeology outside Scotland, as well as in his own country.

^{page 222 note 5} The rediscovery of this fragment is due to Mr. Richard Stones. The diagonal tooling is perfectly preserved on the front and sides; and so is part of the original mortar at the point where the rear part of one side was bedded into the wall.

^{page 223 note 1} This rib, like the other ribs in the compartment, has a profile of twelfth-century type, and must have been reconstructed. The abacus, which has no trace of a mortar-bed left on it, is large enough to carry three such ribs.

^{page 223 note 2} Honeyman, ‘Notes’.

^{page 223 note 3} For spurred square plinths cf. Chichester Cathedral retro-choir (c. 1200); the nave of Great Bedwyn church (late twelfth century); and the Norman crypt of York minster (c. 1180, illustrated in Browne, J., Metropolitan Church of St. Peter, York (1847), pl. xxiiiGoogle Scholar). These examples are more elaborate than the simple keeled triangular spur of Glasgow. We must add that Honeyman (‘Notes’, p. 9) was convinced that the plinth now in situ at Glasgow originally had spurs which were later cut off; but the evidence of this is very dubious, though in other respects the two plinths are certainly identical. The base mouldings of both are in section very like nos. 16 and 17 in Clapham, A. W., English Romanesque Architecture after the Conquest (1934), p. 119Google Scholar, fig. 39, dating from c. 1180 and c. 1181–8. For diagonal axe-tooling see Clapham, op. cit., p. 116. The keel-shaped shaft is, of course, a commonplace of the transitional period. The octagonal abacus is a late twelfth-century feature, see Clapham, op. cit., p. 100. The trefoil foliage has been thought a later addition to the capital (Honeyman, ‘Notes’, p. 8); but this is an unnecessary assumption.

^{page 223 note 4} It is right to point out here that the courses of the wall do not run at the same level on the north and south sides of the shaft, and do not correspond with the courses of the shaft itself. It seems impossible to explain this without assuming a certain amount of demolition and rebuilding; but some disturbance would be inevitable in the process of constructing the arch in a 5-foot-thick wall.

^{page 223 note 5} Honeyman, ‘Notes’, p. 7; for the same suggestion in more detail, see Watson, Double Choir, p. 28.

^{page 223 note 6} Some details are more clearly seen in the sketches by Honeyman (Age of Glasgow Cathedral, opp. p. 9) than in a photograph.

^{page 224 note 1} The moulding between the volutes (see pl. lxiii c, d) seems to be a variety of the ‘pollarded willow’ fluting found in western England in the late twelfth century. The closest parallel is perhaps at St. Mary's Church, Shrewsbury. Pl. lxiv b shows a detail of the capital at the south-west end of the nave (c. 1200; photograph kindly supplied by Professor Zarnecki). See also Francis Bond, Introduction to English Church Architecture (1913), ii, 499–500.

^{page 224 note 2} A brief criticism is desirable of the colours in the plate; it implies no reflection on the work of the printer, since some departure from the original shades is unavoidable. The plaster surface and upper mortar-bed should be a brownish ivory rather than putty-coloured; the greens and reds should everywhere be more subdued; and at the bottom of the left-hand face the black border line should have no green tinge, and the green colour is too prominent in the triangular figure above the line.

^{page 224 note 3} Distances measured by the scale on such drawings are accurate only when taken along the direction of the scale, or along axes inclined at 120° to it.

^{page 224 note 4} See the transverse section in MacGibbon and Ross, ii, 185.

^{page 224 note 5} For this information, and the photograph, we are indebted to Professor W. J. Smith, who witnessed the excavation in 1916.

^{page 224 note 6} Unfortunately, the design on the soffit is too broken to be of any value in this discussion.

^{page 224 note 7} The St. Albans Psalter, ed. Pächt, Dodwell, and Wormald (1960), p. 101.

^{page 226 note 1} The St. Albans Psalter, pp. 278–80.

^{page 226 note 2} For these and other illustrations see pl. 135 of The St. Albans Psalter.

^{page 226 note 3} Clapham, op. cit., p. 146.

^{page 226 note 4} Tristram, E. W., English Medieval Wall Painting: The Twelfth Century (1944), pp. 22–23Google Scholar, 105, 120–1, and pls. 25 and 87; Clapham, A. W., English Romanesque Architecture after the Conquest, pp. 146–8Google Scholar and pl. 35. The dating at Durham depends upon the fact that the Galilee Chapel was built about 1175.

^{page 226 note 5} e.g. B. Rackham, Ancient Glass of Canterbury Cathedral (1949), pl. 111 (1190–1200).

^{page 226 note 6} For slightly earlier examples see the capitals on the north-west door of Lincoln Cathedral (ante 1148), ill. in Francis Bond, Introduction to English Church Architecture, i, 468.

^{page 226 note 7} Bond, op. cit., 504, and cf. 492.

^{page 226 note 8} Browne, Church of St. Peter, York, pl. xxvi.

^{page 226 note 9} It has been suggested, as a compromise view, that the painting may have been painted about the middle of the twelfth century on the stonework of the church of 1136. But there seems to be no real need for this hypothesis.

^{page 226 note 10} This is illustrated by the presence of a stone in the upper bed of mortar, no doubt to pack a gap between it and the next voussoir.

^{page 227 note 1} Mr. Paul describes his method of measurement as follows. ‘The stone was measured on its pedestal. Measurements were made using steel rulers and a plumb-bob. Since the roughness of the stone surface precluded any more accurate measurement, readings were taken to the nearest inch. Each face was measured by clamping a steel ruler along one side, and taking offsets to the stone surface from the ruler edge every inch of length. These measurements were taken on each side and along the centre line of each face, giving a clear plan in each case. Then diagonal measurements were taken across each face with a beam compass, to give checks for the shape. The measurements were then plotted at full scale on graph paper, giving the outline shape of each side, and then the whole was fitted together in the usual way to give an isometric projection drawing of the stone at full size. This was then reduced to the required size. The measurements for radius of curvature were made on the full-size drawings by fitting railway curves. This gave radii of 28 and 20 inches respectively. This was checked by measurement and calculation from the formula radius = where x is half the length of the chord, and a the perpendicular distance from the surface of the stone to the centre of the chord. These calculations give radii of 31·5 and 20·6 inches respectively.’

^{page 227 note 2} We assume in all this that the face ADFE (see fig. 1) was not, when in situ, itself splayed inwards from the plane of the wall. There is no a priori reason against this (see, for example, the voussoirs in an arch of about 28 inch radius above the south door at Dalmeny, finely illustrated in Dr. James Richardson's Medieval Stone Carver in Scotland, pl. 32). The real objection is that if ADFE was in fact inclined inwards, the soffit would be brought round towards making a right angle with the face of the wall, and its direction could not then be reconciled with the evidence of decreasing curvature between DF and CG. The contours of the rough back and side faces of the stone suggest, on the whole, that ADFE was in a plane parallel to the surface of the wall.

^{page 227 note 3} See p. 221 and n. 3.

^{page 228 note 1} Exeter Cathedral affords an interesting parallel. The twelfth-century nave was gutted in the time of Bishop Grandisson (1327–69), who left standing only the side walls up to the sill level of the aisle windows, and inserted a new arcade, clerestory, and fenestration (H. E. Bishop and Prideaux, E. K., The Building of Exeter Cathedral (1922), pp. 25–26Google Scholar, 65–66).

^{page 228 note 2} Unless otherwise stated, measurements of the vaulted area are given from centre line to centre line of the vaulting shafts.

^{page 228 note 3} W. H. St. John Hope, Architectural History of the Cathedral Church and Monastery of St. Andrew at Rochester (1900) [cited as ‘Hope’], pp. 40–52. This is still the standard work on Rochester Cathedral; its conclusions need modification in the light of observations by later writers including A. W. Clapham, English Romanesque Arch. after Conquest, pp. 24 and 45–46, and Webb, G. F., Architecture in Britain: The Middle Ages (1956), pp. 79–80.Google Scholar

^{page 228 note 4} Hope, pl. 111; cf. Hürlimann, Martin and Meyer, Peter, English Cathedrals (Thames & Hudson, 1950), pls. 26 and 27.Google Scholar

^{page 228 note 5} Hope, pl. 11.

^{page 229 note 1} Hope, pp. 76–77. There is no evidence that the wooden screen of the early thirteenth century stood in this position.

^{page 229 note 2} Cf. Archaeologia, xlix, 331. Hope's reconstruction must be modified as suggested by Clapham (above, p. 228, note 3).

^{page 229 note 3} British Museum, MS. Cotton Nero D II, fol. 127b: ‘sepultus a parte boriali predicte basilice inter fundatores confundator, sicut Saul inter prophetas’ (quoted in Hope, p. 50; cf. Wharton, H., Anglia Sacra, i (1691), 346–7Google Scholar).

^{page 229 note 4} Hope, pp. 52–57.

^{page 229 note 5} Hope, p. 48, prints a list of benefactors from B.M. MS. Cotton Vespasian A XXII. Cf. Victoria County History: Kent, ii, 125.

^{page 231 note 1} Glasgow is remarkable, for a cathedral which possessed the relics of an important saint, in the number of its bishops who were buried elsewhere, including even great benefactors of the see. The evidence is available in Dowden, J., Bishops of Scotland (1912) pp. 294–352.Google Scholar

^{page 231 note 2} For John's burial see Dowden, op. cit., p. 296, and for Jocelin's, p. 299.

^{page 231 note 3} See J. E. Lloyd, History of Wales, i, 205. It seems that in the twelfth century ‘clas’ was popularly supposed to be the first element in the name ‘Glasgow’ (K. Jackson in Studies in the Early British Church, ed. N. K. Chadwick, pp. 311–12).

^{page 231 note 4} Royal Commission on Ancient and Historical Monuments in Wales and Monmouthshire: Caernarvonshire, ii, no. 681; Archaeologia Cambrensis, vol. c (1949), 256–61.

^{page 232 note 1} Dumfriesshire and Galloway Natural History and Antiquarian Society: Transactions, 3rd ser., xxxiv, 147 and 183–5.