Yale University Library 39002025316747 YALE UNIVERSITY ART AND ARCHITECTURE LIBRARY LOMBARD ARCHITECTURE MONUMENTS Abbazia di Albino — Milan Lombard Architecture By ARTHUR KINGSLEY PORTER Volume II — Monuments Abbazia di Albino — Milan * SS ^OSK^HI NEW HAVEN: YALE UNIVERSITY PRESS LONDON: HUMPHREY MILFORD OXFORD UNIVERSITY PRESS MDCCCCXVI COPYRIGHT, 1916 BY YALE UNIVERSITY PRESS Published October, 1916 Seven Hundred Fifty Copies Printed from Type LOMBARD ARCHITECTURE MONUMENTS Abbazia di Albino — Milan Lombard Architecture ABBAZIA DI ALBINO,1 S. BENEDETTO (Plate 1, Fig. 1, 2) I. The ancient abbey of S. Benedetto, or Vall'Alta (Vallis Alta), lies in the valley of the Luglio some kilometres above the town of Albino, from which it may be reached by a good carriage road. The monument, although of the greatest historical and artistic importance, is still entirely unknown, having escaped the attention, not only of Enlart, but also of the numerous other writers who have occupied themselves with the question of Cistercian architecture in Italy. The history of the monastery is but little better known than the church itself. The archives, which must have been exceedingly rich, were dispersed when the abbey was suppressed at the end of the XVIII century. A great number of documents found their way to the Archivio dello Stato at Milan;2 others are preserved in the Biblioteca Civica at Bergamo. There is a persistent tradition that still others are extant in the Ambrosiana and Castello at Milan, i (Bergamo). 2 These are gathered together in a parcel supplied with the provisional number Fondo di Religione 16/82-87 and labelled "Pergamene di San Benedetto di Vall'Alta, Bergamo." The bundle is divided into smaller packages disposed as follows: Package No. 81, 16 documents, 1240-1300 do No. 82, 32 do 1371-1380 do No. 83, 36 do 1331-1340 do No. 84, 10 do 1321-1330 do No. 8.5, 12 do XVI century do No. 86, 27 do 1421-1480 do No. 87, 17 do 1361-1370 LOMBARD ARCHITECTURE but my efforts to trace them in both of those localities have been without success.3 Some documents were copied and published by Lupi in the XVIII century, and the archives were seen and studied in their entirety by Pietro Gatti, priest at Abbazia, who published in 1853 a history of the abbey. In this the author promised to continue his researches with the publication of a Codice dei Diplomi, but he died in 1867 without having produced this work.4 It may be conjectured that the documents which Gatti evidently had in his hands, and which have now disappeared, have passed into the hands of private individuals either at Vall'Alta or at Borgo S. Caterina, whither Gatti subse quently removed.0 Until they be found, the little book of Gatti is the most important source for the history of the monument. An inscription in the church, placed there at the time of the restoration of 1843, and probably composed by the same Gatti, was, when I saw the church in July, 1913, almost illegible, and has doubtless since entirely disappeared. Gatti, however, gives a copy of it. The present priest, Sacerdote Mayer, possesses a photograph showing the Empire fa£ade built in 1843 and replaced in 1913 by a new construction in the Romanesque style. II. Although Gatti held in his hand a plentiful supply of original documents, his work leaves us somewhat in doubt as to the year of the actual foundation of the abbey.6 In one place,7 he states that it was founded in 1133, by Gregorio, bishop of Bergamo, but this notice is difficult to accept, since 3 Other important sources for the history of the abbey are: Libri Oensuum in the Archivio della Curia Vescovile at Bergamo; Celestino. — Del matrimonio e verginita di San Grata, and Relazioni su diversi inonasteri — two manuscripts which passed from the possession of Cavagnis to the Biblioteca Civica of Bergamo, where they are at present preserved, Nos. X 1.12 and \p 2.35. Part of the second manuscript, dealing with the abbey of Pontida, was published in 1876, by Alessandri. * Elogio funebre del Sacerdote Pietro Gatti letto dal Sacerdote G. C. Sirani, nel trigesimo celebrato nella chiesa parrochiale di Borgo Santa Caterina in Bergamo, il 2& Maggio, 1867. Bergamo, Natali, 1867. 5 The present priest of the church, Sac. Raimondo Mayer, told me that he had knowledge of certain documents relating to the abbey in the hands of private individuals of the place, and promised to obtain for me at least a copy of the same, but the promise was never kept. - It is quite impossible to connect our abbey with the basilica of S. Daniele, founded in 928, by Adelperto, bishop of Bergamo, as is evident from the following passages of the will of this prelate: In nomine Domini nostri Iesu Christi. Regnante domno nostro Hugo rex in Italia anno tercio mense november, indictione secunda. Adelbertus sancte pergamensis ecclesie episcopus. . . . Casis denique et rebus seu familiis juris mei, quas habere visus sum in vico et fundo qui dicitur Albinies, statuo et judico, ut presenti post meum discessum usque in perpetuum habeat presbyter et custos ille, qui pro tempore custos et officialis fuerit in capella et basilica ilia, que est constructa in curte ilia, que dicitur Albine, quam ego in honorem beati sancti Danielis consecravi. . . . (Hist. Pat. Mon., XIII, 899). 7 4. ABBAZIA DI ALBINO, S. BENEDETTO Gregorio did not become bishop until the following year, 1134.8 It is, more over, certain that the abbey was really founded by this bishop, who had been a monk of the Cistercian order, since Gatti cites a document of April, 1136, fourteenth indiction, in which Bishop Gregorio, indignus monachus et Sancte Pergamensis ecclesiae humilis episcopus, invests the monks sent him by St. Bernard with a good part of the lands which he possessed in Vall'Alta.0 The bishop goes on to state: ecclesiam edificavi, fratresque meos sub monastica regula victuros constitui. This document implies that the abbey was founded by Gregorio and that the church had already been at least partly constructed in April, 1136. It therefore follows that the foundation must have taken place immediately after the accession of the bishop to the episcopal throne in 1134, since it is inconceivable that the many formalities connected with the foundation and the obtaining of monks from St. Bernard could have been arranged in less than two years.10 In 1138 the pope, Innocent II, confirmed, with a bull dated May 5, all of their possessions to Oprando, abbot of Vall'Alta, and his monks.11 The church, although it is spoken of as built in 1136, could then have been only partially completed, since the edifice was not consecrated until May 24, 1142. This function was celebrated by the bishop Gregorio, assisted by Magnifredo, bishop of Brescia, and Giovanni, bishop of Lodi.12 The consecration of 1142 is also recorded in the inscription on the west wall of the church, thus given by Gatti:13 8 According to Vincenzo Coronelli, Rerum et Temporum Ecclesiae Bergomensis Synopsis, ed. Graevius et Burmannus, Thesaurus Antiquitatum et Mstoriarum Italiae, Vol. IX, pt. 7, p. 15, Gregorio succeeded Agino, suspected of simony, in the early part of 1134. He came in personal contact with St. Bernard at the council of Pisa, June 19, 1134. ¦j A few days afterwards the bishop made another donation to the monastery. io Gatti, 4-5. n Ibid., 5. 12 Condotta poi al suo termine la chiesa di S. Benedetto, il vescovo Gregorio assistito dagli onorevoli vescovi Magnifredo di Brescia e Giovanni di Lodi, col consenso ed autorita del sommo pontiflce Innocenzo, ne fece la solenne dedicazione nel mese (ai 24) di Maggio del 1142. Ed avendosi dato principio in nome della Santissima Trinita ai venerabili offizii della consecrazione, e fattosi discorso della dote della medesima chiesa, senza la quale, giusta i canoni, non si pud celebrare la dedicazione, e tenendo tutti volti gli occhi al venerabile vescovo di Bergamo, » cui quel luogo special- mente s'apparteneva, egli col consiglio dei canonici e dei nobili uomini, ponendo sopra l'altare un legno che teneva nella sua mano, per lignum quod in sua tenebat manu, super altare ejusdem ecclesiae positum, etc., fece alia chiesa investitura di donazione di altre possessioni, cioe del versante di nord del Monte Pelsino (Pizzo), serbato per6 » se il diritto di tagliarne legna a' propri usi, e della Valle Altina. L'istromento si legge al foglio 142 del libro vescovile Censuum tempore 7R. R. D. D. Joannis Barotii episcopi, qui profecto fuit magm ingenii vir. (Gatti, 5-6). 13 49. LOMBARD ARCHITECTURE CON SANTO PENSIERO L'ANNO 1136 IL VESCOVO GREGORIO DI BERGAMO FONDAVA L'ANNO 1142 I VESCOVI GREGORIO DI BERGAMO, MANFREDO DI BRESCIA E GIOVANNI DI LODI CONSACRAVANO PEI MONACI CISTERCENSI DI S. BENEDETTO QUI CONVENUTI A SALMEGGIARE L'ETERNO QUESTO TEMPIO CUI NEL VOLGENTE ANNO 1850 IL DEVOTO POPOLO COL SUDORE DEL SUO FRONTE AMPLIANDO ADORNAVA AD ONORE DI S. BENEDETTO ED ALLA MASSIMA GLORIA DI DIO Notwithstanding the donations of the bishop Gregorio, the monastery seems always to have been poor, and several donations of the XII and XIII centuries refer to the fact that the monks did not have sufficient revenues on which to live.14 Moreover, as time went on, the prosperity of the monastery and the number of monks seem to have diminished, as is shown by the following table deduced by Gatti from various documents seen by him: 1220, 15th December The Abbot, 6 monks, 16 lay-brothers. 1289, 12th October The Abbot, 5 monks, 9 lay-brothers. 1325, 19th May The Abbot, 8 monks, 12 lay-brothers. 1351, 30th May The Abbot, 4 monks, 6 lay-brothers. 1387, 14th June The Abbot, 2 monks, 1 lay-brother. (2 lay-brothers absent). 1408 The Abbot, 2 monks and 1 lay-brother. 1472 Two monks. 1518 A single monk. In the XV century the monastery, following the lot of many of its pros perous contemporaries, passed into commendam, and in the XVI, XVII and n Gatti, 15-16. ABBAZIA DI ALBINO, S. BENEDETTO XVIII centuries it fell into the state of decay usual in those times. Never theless, in 1772 the church was entirely repainted with frescos. In the Revolution the abbey was finally suppressed, and in 1813 a parish was established in the church. The growth of the latter necessitated in 1843 the enlargement of the church. A radical transformation was undertaken and the work was carried to completion only in 1850. Two new side aisles were added to the nave, originally of a single aisle. A new facade was erected, and the whole church baroceoized.15 When I visited the church in July, 1913, a new and even more radical reconstruction was in progress. The facade was being rebuilt on pseudo- Lombard lines, the barocco stucco and plaster had been removed from the interior, and the whole church was being restored in a style which the architect doubtless intended as Romanesque. It is unfortunate that the new side aisles added in 1843 were also being made over so as to appear integral parts of the ancient structure. III. The church at present consists of a nave three bays long, two side aisles, a choir of a single bay also flanked by side aisles (these side aisles are now walled off), an apse, a southern absidiole, a modern chapel replacing the old northern absidiole, and a campanile to the south of the choir. As we have seen, the side aisles of the nave are a modern addition, and, except in the choir, the edifice originally had only a single aisle. The choir is covered by a rib vault, of which the broad rectangular diagonals, some half a metre in width, are slightly segmental in elevation. The vault surface is, nevertheless, highly domed. There are no wall ribs, and the wall arches are approximately semicircular. It is a singular fact that this vault is pierced by two square- headed windows, apparently original. It is plain from the stereotomy of the intersection that one diagonal was completed first and the other subsequently added against it. The remainder of the nave is also covered with rib vaults, but of a very different character, since the diagonals have a torus section and the intersection is formed by a regular keystone which completes both arches is The church as thus made over is described by Gatti in the following words: Pilastri di ordine composito con basamenti e capitelli semplici reggono le volte del vasto tempio. II volto che si alza sopra la navata di mezzo, si sfoggia a padiglioni, ciascuno diviso in lunette da lunghi filoni, che salgono dal suolo co' pilastri e si incro- cicchiano nel centro delle arcate. . . . Nell'alto e sulle pareti, sugli scomparti e ne' frontoni sono dipinte figure di Santi, gloriette, rappresentazioni storiate, simboli religiosi e diversi ornati e festoni di finti stucchi e bassi rilievi benissimo sfoggiati e disposti con dolcezza di tocchi, con grazia di chiaro oscuro, e con vaghezza e varieta di pensiero e si bene rispondenti alle varie parti d'architettura, etc. The church as it was before the restoration he describes on page 5, as follows: La chiesa era ad una nave colle mura schiette e liscie ma tutte di vive pietre senza intonaco ne dentro ne fuori; le volte pure di pietre erano sostenute da forti pilastri, che sporgevansi nell' interno e nell' esterno delle pareti ; e la maniera di costruzione e precisamente secondo il gusto di quel secolo. LOMBARD ARCHITECTURE at the same time. The southern side aisle of the choir has an undomed groin vault. The two western vaults of the nave are modern and have been entirely rebuilt. In the east wall the choir vault is carried on corbels, but elsewhere there is a logical and continuous system consisting of three members, of which the central one is rectangular and the two outer ones are torical. In the choir, uncarved and characterless capitals serve to adjust the rectangular load of the ribs to the shafts, but in the nave the shafts are continued to form ribs, and no capitals are interposed. The masonry consists of rough ashlar, of which the stones are often not square and are separated by wide mortar-beds. The horizontal courses are frequently broken. The windows, widely splayed, were evidently intended to serve without glass. Above the heavy domes of the apses there is apparently laid solid masonry, so as to produce a conical form externally; there are, however, and probably always have been, wooden roofs over the main vaults, although it is evident that the eastern wall of the church has been raised. IV. The edifice is notable for the absence of decoration. The apse and southern absidiole are not supplied with either arched corbel-tables or pilaster strips, and the cornices are formed of a single moulding of the simplest char acter. In the interior there is a similar absence of all decoration and carved ornament. There are no bases ; capitals are either omitted altogether, or are replaced by imposts of the simplest description. V. It is evident from the edifice itself that the church consists of two distinct periods of construction. To the first belongs the choir, with its side aisles and apses. It was evidently intended to construct a church of three aisles which should be entirely vaulted, the side aisles by groin vaults, the nave by rib vaults of thoroughly Lombard type, with broad, rectangular diagonals. This plan was subsequently changed. A single-aisled nave was erected, and was covered with rib vaults, of which the diagonals had a torus section. The two epochs of construction differ so slightly from each other that it is not possible that they are separated by a great interval of time. The story told by the stones coincides perfectly with the documentary evidence, and it seems certain that the choir, its side aisles and apse, were erected 1134-1136, and formed the church which Bishop Gregorio in 1136 speaks of as constructed. Owing probably to lack of funds the work was temporarily suspended, and resumed in a somewhat less ambitious manner. In 1142 the entire edifice was finished and consecrated. The abbey of Albino, thus authenti cally dated, is a most important monument. Not only is it one of the earliest Cistercian abbeys in Italy, but it also furnishes the earliest example of a profiled rib vault south of the Alps. ABBAZIA DI SESTO CALENDE, S. DONATO ABBAZIA1 DI SESTO CALENDE,2 S. DONATO (Plate 1, Fig. 3, 4, 6) I. The frazione of Abbazia, in which is situated the ancient abbey church of S. Donato, lies about a kilometre distant from the commune of Sesto Calende. De Dartein3 has studied and illustrated the architecture of the church. The history of the abbey has been made the object of a special monograph by Spinelli in a work which is a serious contribution to the local history of Sesto Calende. The historians Robolini, Giulini,4 Bescape and Ughelli have all treated of the history of the monastery. II. The exact year of the foundation of S. Donato is unknown. Ughelli has published a bull of Pope John VIII, dated 874, in favour of Giovanni, bishop of Pavia, in which, among other goods, are confirmed to the latter "the monastery of S. Donato founded by your predecessor Bishop Luitprando, in the place which is called Scozzola."5 This bull, which is very badly printed by Ughelli, offers several difficulties. According to Ughelli, Bishop Luitprando or Liutardo died in 830/ but whence this information is derived I do not know, and I suspect that it is incorrect, since a Bishop Liutardo of Pavia is mentioned by Anastasius Bibliothecarius7 as a contem porary of Pope Nicholas I (858-867), and of the archbishop Giovanni of Ravenna. On the basis of this evidence historians have been divided as to what year and even as to what time the foundation should be ascribed, and have assigned it anywhere from 822 to 860. At any event, it is certain the foundation must have taken place in the second or third quarter of the IX century. 1 This frazione was formerly known as Scozzola. 2 (Milano). 3 383. * I, 274. s JOANNES EPISCOPUS Servus servorum Dei. Reverendiss. Joanni S. Ticinensis Ecclesiae, &c. in perpetuum. . . . Igitur postulante &. nobis tua reverentia quantum ea, quae ad stabilitatis integritatem, & ad profectum honoris sanctae fuse perti- nere noscuntur Ecclesiae . . . confirmamus, tibi, successoribusque tuis; . . . harumque tenore prsecipienies [sic = praecipientes], ut Monasterium S. Dorati [sic] fundatum a Luitprando Epis. decessore tuo in loco, qui dicitur Scogialo, cum omnibus rebus mobi- libus, & immobilibus secundum testamenti sui seriem collatis . . . te, successoresque tuos perpetuis temporibus j urisdictionem tenere, habereque decernimus. . . . Datum est hoc nono Kal. Septembris, per manum Leonis Episcopi missi, & Apocrisarii S. Sedis Apostolicae, imperante Dom. Carolo coronato magno imperatore. Et ut certius appareat hoc nostrum Privilegium, & inconcussum permaneat, sigillo nostro jussimus insigniri. Anno II. & post consulatum eius anno II. indict. XI. (Ughelli, I, 1085-1086). This bull has been reprinted in the Hist. Pat. Mon., XIII, 463. «Ibid., 1084. 7 De vitis Roman. Pontif., ed. Muratori, R. I. S., Ill, 255. LOMBARD ARCHITECTURE The monastery, although outside of the diocese of Pavia, depended upon the bishops of that city. In 1105 it is mentioned among the possessions confirmed to the bishop Guido by Pope Paschal II.8 The subsequent history of the abbey, so far as it is known, is of minor importance for the study of the architecture. In 1509 the monastery passed into commendam,9 and in 1533 the commendam was given to the Ospedale Maggiore of Milan.10 Subsequently Olivetani monks were introduced. From a passage of the acts of the pastoral visits of the bishops of Pavia, Spinelli11 has deduced that in 1566 the sacristy was placed at the east end of the south side aisle and was vaulted. The church was whitewashed between 1607 and 1613, but the final baroccoization did not take place until after 1616. It was, I presume, at the end of the XVIII century that the monks were suppressed and a parish established in the church, although I have found no explicit mention of the fact. In 1816 the church was removed from the jurisdiction of the diocese of Pavia and placed under that of Milan.12 III. The edifice consists of two distinct parts, a church and a narthex. The church comprises a nave four bays long, two side aisles, a raised choir flanked by side aisles, three apses and a crypt of three aisles four or five bays long. The narthex, of about the same width as the church, consists of two bays divided into three equal aisles, which do not correspond to those of the basilica. The campanile rises to the north of the choir. The narthex is covered with groin vaults, oblong in plan, highly domed and provided with transverse arches of which the extrados is so much loaded as to form a pointed curve. These vaults are provided with wall ribs, also with loaded extrados. The free-standing supports are columns notable for their rather exaggerated entasis. The responds comprise five members, of which the central one is semicircular or semioctagonal. The nave has been covered internally and externally with barocco stuccos, which make it exceedingly difficult to trace the original forms. The rect angular piers show a very pronounced entasis or inward lean on the side of the nave as well as on that of the side aisles. The crypt retains to a much larger extent its original character. The groin vaults are not so highly domed as those of the narthex, but have similar disappearing transverse arches. The facade of the narthex and part of the side walls of the same are constructed of large and carefully wrought blocks of ashlar somewhat crudely laid in courses frequently broken and separated by thick beds of mortar; in parts of the narthex, however, much cruder masonry, hardly superior to rubble, is introduced. The campanile and the remainder of the basilica s Monasterium S. Donati a Ticinensi quondam Episcopo in Scovilla fundatum. . . . (Ughelli, I, 1085). a Spinelli, 47. io Ibid., 228. mil. 12 Spinelli, 111. 8 ABBAZIA DI SESTO CALENDE, S. DONATO (where the latter has not been remade in the period of the Renaissance) are constructed of very much smaller stones — in the apse many bricks are intro duced — for the most part uncut, rather carelessly laid in a manner which suggests rubble construction. In addition to scaffolding holes, there are numerous scaffolding brackets. A distinct break in the masonry of the exterior walls makes it evident that the narthex was added after the nave had been completely finished. IV. The capitals of the narthex are of a curious type, without, so far as I am aware, analogy in Lombardy.13 The coarse carving possesses a certain barocco quality that recalls the capitals of S. Giorgio in Palazzo at Milan.14 The design is confused, and one hardly knows whether one is looking at leaves, interlaces or an all-over pattern. The execution is mechanical, and under cutting is avoided, yet the character of the leaf-forms seems rather advanced. One of the capitals15 of the free-standing columns has a row of uncarved acanthus leaves surmounted by two sets of volutes — a motive Carlovingian in origin, but here treated in the dry manner of the capitals of Fontanella al Monte.16 The abacus of this capital has a scale ornament, other abaci are decorated with rinceaux, or similar motives. Some of the bases are of Attic character and supplied with griffes. Others have a profile consisting of two square fillets separated by a scotia, in the centre of which is a torus. The capitals of the crypt, on the other hand, are uncarved or merely with corners splayed in the form of a leaf; the monolithic shafts are without bases. The campanile is decorated with arched corbel-tables grouped three and three resting on pilaster strips. On the west face these pilaster strips are grouped two and two, but here, like the belfry itself, they appear to have been rebuilt. The central apse is adorned with a cornice of blind niches in two orders. In two orders also are the windows. The north absidiole, still preserved, has a cornice of arched corbel-tables and small, widely splayed windows. Im bedded in the apse as second-hand material are several bits of Carlovingian carving. V. The narthex, as proved by its ashlar masonry and by the capitals analogous to those of S. Giorgio in Palazzo at Milan (1129) and Fontanella al Monte (c. 1130), is certainly a construction of the XII century, and may be ascribed with confidence to c. 1130. Of the remainder of the edifice, which appears homogeneous, in so far as it has not been remade in the barocco period, it is more difficult to determine the epoch. There is, it has been seen, conclusive internal proof that the nave is earlier than the narthex, that is to say, earlier than c. 1130. The masonry is precisely analogous to that of the is Plate 1, Figs. 3, 4. i* Plate 128, Fig. 5. is Plate 1, Fig. 3. ie Plate 93, Fig. 2. LOMBARD ARCHITECTURE neighbouring church of S. Vincenzo, consecrated in 1102. I therefore assign this portion of the edifice to c. 1100, with the exception of the columns and capitals of the crypt, which are probably taken from an earlier building. I experience, I confess, considerable reluctance in ascribing an edifice in many ways so primitive to so late a date, but the close analogy of the masonry to that of S. Vincenzo seems to leave no other alternative. The singular crude- ness of the masonry and the numerous primitive features of the edifice must be ascribed to extraordinary haste and carelessness on the part of the builders. ABBAZIA DI SESTO CALENDE, S. VINCENZO (Plate 2, Fig. 1) I. Spinelli is the only author who has noticed the existence of this monument, which lies in the fields a short distance from the abbey of S. Donato. II. In the acts of the pastoral visit of the bishop of Pavia of 1595, Spinelli found and quotes a passage which proves that this church was formerly a convent of nuns.1 In 1595 it had already ceased to have a separate exist ence owing to the calamities it had suffered in the wars, and its goods were given to the abbey of S. Donato. The church, however, continued to be officiated as a dependence of S. Donato until 1780, when it was finally suppressed. In the Ambrosiana at Milan, in that very valuable transcript of ancient documents known as the Codice della Croce, I had the good fortune to find an inedited document which throws light upon the early history of the convent, and happily gives us the year (1102) in which the church was consecrated.2 It may, consequently, be assumed that the abbey was founded somewhat before this date. The edifice appears to have been made over and baroccoized in the early XVIII century. On the wall is a destroyed inscription of 1729, and the altar bears the date of 1732. i Ad quam [ecclesiam sub titulo sancti Vincentii] et monasterium quodam ut dicitur ibi ad eamdem ecclesiam constructum permanebant moniales et iterum visa fuerunt quedam vestigia et fundamenta monasterii et ut item dicunt fuit mpnas- terium. . . . tempore bellorum vastatum fuit. (Spinelli, 145). 2 1 transcribe the most important part of this document: Anno ab incarnacione domini nostri Ihesu Christi milleximo centeximo secundo mense iulii indicione decima. Dum in Dei nomine intra clausa modoeciensis ecclexie bernardus romane ecclexie insignis cardinalis atque legatus domni apostolici pascalis, nee non et grosolanus uenera- bilis archiepiscopus ecclexie mediolanensis honeste tractarent de diuinis et humanis aduenerunt legati ermeline abbatisse sestensi monasterii suplicantes exconsecrationem prefati monasterii. (Codice della Croce, MS. Amb., D. S. IV, Vol. V, f. 20). 10 ACQUANEGRA SUL CHIESE, S. TOMMASO III. The edifice is extremely simple in character, and consists of a single- aisled nave and an apse. The upper part of the nave walls has been rebuilt in the barocco period, but with the exception of the apse, the church has evidently always been roofed in wood. Like the nave, the apse was somewhat raised in the barocco period. This apse, however, is the best preserved part of the edifice, and still retains its widely splayed windows, albeit walled up. The masonry is extremely rough, and recalls that of the choir of S. Donato. Uncut bits of stone, pebbles, bricks and a few roughly squared blocks of stone are crudely laid in courses, frequently broken and deviating widely from the horizontal. The mortar-beds are extremely wide, and there are numerous scaffolding holes. The herring-bone masonry of the nave is the result of late alterations. IV. The apse is decorated externally with arched corbel-tables supported on pilaster strips. The interior possesses fine frescos of the XVI century, one of which bears the date 1516. V. The documentary evidence seems conclusive that this church was consecrated in 1102. It is true that the masonry is singularly crude to have been executed at this epoch, and I confess that I have long debated whether it be not necessary to assume that the apse is the remains of an earlier chapel preserved in the church reconstructed in 1102. However, having observed that in the narthex of S. Donato, an edifice evidently constructed c. 1130, there is some masonry almost as crude as that of S. Vincenzo, I am forced to the conclusion that the masons of Sesto Calende were singularly careless and slipshod in their work. They were, perhaps, forced to this by the lack of good stone or brick for building. It therefore seems necessary to accept S. Vincenzo as an authentically dated monument of 1102. ACQUANEGRA SUL CHIESE,1 S. TOMMASO I. The mosaic pavement of Acquanegra has been illustrated by Matteucci and Venturi.2 For the history of the monastery the little book of Casnighi is of great value, and some important notices are contained in the inexact publication of Lucchini. II. The earliest known document relating to this monastery dates from November 9, 1101, and is a deed of Adalperono, bishop of Trent, investing the abbot of Acquanegra with the abbey of S. Maria della Gironda of Bozzolo.3 Since in this document the choir of the church is very precisely i (Mantova). 2 111,436. 3 In nomine Domini Dei eterni. Die Sabati nono intrante mese novembris. Dum in Dei nomine dominus Adalperonus Dei gratia Tribentinus adesset episcopus in 11 LOMBARD ARCHITECTURE mentioned, it is evident that the edifice must have been constructed and in use at this period. Lucchini states4 that after examining fifty-two documents contained in the Libro dei Privilegii del comune di Acquanegra e Mosio he reached the conclusion that the monks had been put under the jurisdiction of Trent by the emperor (Barbarossa ?), who was favourable to them, and that they were subsequently spoiled and plundered by the papal party. A docu ment of 1104 referred to, but not published, by Casnighi5 makes it evident that at that epoch the monastery had already been in existence for some time. The same author also cites numerous other documents which prove the continued existence of a monastery of Benedictines at Acquanegra in the XII century." In the first half of the XV century the abbey passed into commendam,7 and in 1562 a vicaria perpetua was erected in the church,8 which in 1802 was reduced to the rank of a simple parish. In 1898 an act of atrocious barbarism was committed when the fine old Lombard basilica9 was torn down and replaced by the existing modern edifice. At this time was discovered the mosaic pavement which is now all that remains of the ancient monastery. III. Of the ancient church as it was before 1898, there is extant not even a description or a photograph. IV. The ancient mosaic pavement is preserved under the wooden floor of the new church, and can only be seen when the planks above have been specially removed — a fatiguing piece of work which requires much time. The ancient pavement covered an area at least as great as that of the existing church. It is in very unequal preservation, some parts being in good condition while others are much damaged. In the northern side aisle there was probably a representation of the signs of the zodiac — at least a crab, a ram and a Capricorn are extant in part, and it is reasonable to suppose the other signs were placed in the remaining squares into which the pavement was divided. Other fragments of the northern side-aisle pavement show animals which I believe to be purely fanciful. At the west end of the northern side aisle is one of the most interesting representations of the entire pavement. On an excellently drawn stallion rides the figure of a horseman apparently nude. In his left hand he holds the reins, and his right hand is placed against his cheek with a gesture that seems to indicate thought. From his head flutters in the breeze what I take to be the crest of a helmet, but the pavement is so much damaged that it may be in reality long hair. The inscription SINON is clearly legible and complete. Venturi10 calls this figure "Sidone sul cavallo," and says that it is a representation taken from the iEneid. He means to identify the horseman, I presume, with the Sinon of wooden-horse fame.11 I Ecclesia S. Thome apost. de Acquanigra. Ibi in choro ejusdem ecclesie etc. . Factum est hoc anno Domini Millesimo C. primo Indictione nona . . . (Lucchini, 95). * 96. 5 23. c Ibid., 26-27. - Ibid., 30. s lud., 31. fl Ibid., 36. io III, 436. n Mneid, II, line 77 f. 12 ACQUANEGRA SUL CHIESE, S. TOMMASO confess, however, this identification seems to me to offer several insurmountable difficulties. In the first place, the horse in the mosaic, far from being of wood, is extremely realistic. Secondly, it does not appear from Vergil's account that Sinon ever rode upon the wooden horse. Finally, in all mediaeval church art I know of no other representation of a subject taken from Vergil. I consider it much more likely that we have here a fragment of a representation of a scene from the Maccabees, such as Lombard artists loved to represent, and executed, for example, in the pavements at Casale Monferrato and Bobbio. Simon, the second son of Mattathias, called also Thassi, is a prominent char acter in the first book of Maccabees, and upon the death of Jonathan became the head of the nation, both as captain and high priest. During his brilliant reign he took part in many warlike enterprises.12 Our pavement doubtless represents a fragment of a scene from one of these battles. Near the figure of Sinon are placed various animals of a veritable zoological garden. Particularly well drawn is what looks like a kangaroo. Below is an immense dog. The fore parts, including the long ears, pointed nose and eyes, neck and forelegs, are seen in plan from above. The hind parts, the leg with claws and the skinny tail, are, on the contrary, shown in eleva tion. Below is the inscription [CERJBERVS. At the extreme western end of the side aisle and, in fact, under the door, may be read the inscription IDRA. The figure of the hydra herself is broken, but one of the dog-heads can still be clearly seen. The extant fragments of the pavement of the southern side aisle show a great deal of conventional ornament. The patterns here, as throughout the entire pavement, are most varied and interesting, and consist of rinceaux, Greek frets, or a sort of "T"-formed pattern, guilloches, parallel lines, and numerous other motives. Many of these ornaments are extremely similar to those found in the mosaic of the Campo Santo at Cremona. The pavement of the southern side aisle, in addition to these purely ornamental forms, shows also several animals. In one medallion is a fox looking at a chicken placed in the neighbouring medallion, as if Reynard saw with considerable enthusiasm a prospective dinner. We apparently have here a version of one of the folk stories which later took form in the Roman de Renart. Below is seen another episode of the same cycle, a fox hunting a hare. A fragment showing part of a goose may also have belonged to the same epic. Another scene represents two birds, perhaps geese, and two animals I am unable to identify, placed on the four sides of a large diamond, the interior of which has been destroyed. In the nave, the extant fragments represent interlaces of circles in the centre of which are seen two birds drinking out of a bowl, after the manner of the early Christian frescos in the catacombs ;13 an animal which I am unable to identify, and numerous bits of other destroyed subjects. 12 I Maccabees, passim. 13 This fragment has been illustrated by Venturi. • 13 LOMBARD ARCHITECTURE Technically this mosaic is excellent, and represents a great advance over that of Pieve Terzagni, in that much green is used in addition to black, white, yellow and red. The colouring of the pavement in the nave is particularly fine and delicate, perhaps the daintiest and most charming I have ever seen in a Romanesque mosaic. All crudeness has disappeared alike from the colour and the drawing, and the design is as exquisite as is the colour. V. The excellent technical quality of the mosaic at Acquanegra gives reason to believe, notwithstanding the naivete of the subjects, that we have here not a particularly early work. Numerous of the conventional figures show great analogy with the mosaic of the Campo Santo at Cremona, which was executed 1106-1117. There is documentary evidence that the church was in use in 1104, and it is probable that the edifice had been constructed not long before this. We may, therefore, ascribe the pavement of Acquanegra to c. 1100. ACQUI,1 CATHEDRAL (Plate 2, Fig. 2, 3, 4, 5, 6; Plate 3, Fig. 1, 3, 4, 5 ; Plate 4, Fig. 1, 4) I. Although an extremely important monument of Lombard art, the cathedral of Acqui remained completely unknown, save for a passing reference of Pellate,2 until I published an account of it in the American Architect in 1913. Only the mosaics, carried away to Turin in the middle of the last century, had attracted some notice. Aus'm Weerth3 published a drawing of them made by Vico when they were still in their original position in the crypt, and before they had been broken in pieces for transportation. This drawing is, therefore, of great value. The study of Fabretti is also of some help in interpreting the inscription. For the history of the cathedral of Acqui, the same writers — Savio, Moriondo, Biorci, Lavezzari and Blesi — are to be consulted as for the history of the church of S. Pietro.4 II. In the account of the early documents regarding the cathedral of Acqui, cited below in connection with the history of S. Pietro, it will appear there is every reason to believe that until 1023 the cathedral of Acqui was situated in the church of S. Pietro. In view of the serious documents which prove this, there is no reason to give heed to the late and unreliable chronicler, Fra Jacopo, when he tells us that Lodovico Pio (t840)5 was buried in the existing cathedral. i (Alessandria). 2 423. 3 18. * See below, p. 25. b Facta strage Sarracenorum. predictus imperator Ludovicus Pius de Roma venit in Lombardiam. et in civitate Aquis Lombardie infirmatur et moritur. et in capitulo ecclesie sancte Marie Maioris sepellitur. ubi stat ipsius sepulcrum. (Frate Jacopo da Acqui, Chromcon Imaginis Mundi, ed. Hist. Pat. Mon., V, 1524). In alia autem Ecclesia 14 ACQUI, CATHEDRAL The bishop Primo (989-1018) began the construction of the new cathedral church of Acqui on its present site.6 This church, continued by Primo's successor, Brunengo, must have been sufficiently advanced in 1023 so that services could be held in it, for the bishop Dudone in that year transferred to the new cathedral the canons from the old cathedral of S. Pietro, and began to celebrate mass in the new edifice.7 In 1034 Dudone was succeeded by S. Guide A document of 1042 speaks of the canonica of S. Maria at Acqui as already constructed.8 S. Guido embraced the task of completing the cathe dral with such enthusiasm that in after times he came to be considered as its sole builder. A pious reverence for the saint has doubtless played its part in somewhat exaggerating his role. None the less it is certain that he made very large donations to the diocese, and there is still extant the confirmation of one of these donations, made by the emperor Henry III, in 1040. 9 In 1067 S. Guido consecrated his cathedral. The most authentic source for this fact is an inscription in the contemporary mosaic discovered in the crypt of the cathedral at Acqui and now in the basement of the Museo Civico di Arte Applicata ad Industria at Turin.10 Although sadly mutilated, enough of the inscription is extant to leave no doubt as to its significance. According to Durand there was in his time a portion of this inscription still to be seen in its original position in the crypt; but, if so, it has since disappeared.11 The fact that S. Guido completed the cathedral in 1067 is confirmed by a number of other sources. The most ancient of these is the sculpture of the XII century now in the facade of the arcade of the episcopal palace facing the Piazza del Duomo, in which is shown a bishop, evidently S. Guido, holding a model of the church in his hand. In the life of S. Guido written by Lorenzo Calceato about 1260, the consecration of the cathedral by S. Guido in 1067 S. Mariae maj oris est corpus Ludovici pii Imperatoris. (Chronica Fr. Jacobi de Aquis, ed. Moriondo, II, 135). As a matter of fact, Lodovico Pio died on an island in the Rhine near Mainz. 8 See text cited below, p. 26. f See text cited below, p. 27. 8 canonica sanctae Mariae quae est constructa intra civitatem Aquensem. (Moriondo, I, 30). Cf. Biorci, II, 286. 9 Biorci, I, 189-190. 10 DNO WIDONE PONTIFICE VIRfO] PRVDENTISSIMO C[PLETVM EST W] IDONE [PER] OMIA LAVDABILI ET OBTO [observantissimo] O [ANNO AB INCARNATIONE DNI NRI] IHV X[PI ML]XMO VII. INDICE V n Un petit fragment est seul visible ; il contient deux mots latins qui seraient les noms d'un eveque d'Acqui du XII siecle. 15 LOMBARD ARCHITECTURE is very explicitly recorded,12 although the author makes two mistakes in his chronological notes, giving the sixth, instead of the fifth, indiction, and the emperor as Henry III instead of Henry IV. Calceato says nothing about the cathedral having been begun by Guido's predecessors, and although it is not stated that S. Guido began the construction, the text might be construed to imply it, and it probably was so interpreted by later authors. Calceato gives us two new details touching the consecration: (1) that S. Guido was assisted by the bishops Pietro of Tortona and Oberto of Genoa; and (2) that the consecration took place on November 13. The first detail is certainly correct.13 The second item appears to be also accurate, since the day of the consecration is given as the thirteenth of November in an ancient parchment missal cited by Biorci.14 It is therefore probable that later texts which give the date of the consecration as the eleventh of November are erroneous. The longer life of S. Guido, edited by the Bollandists and written long after the death of the saint,15 repeats the notice of the consecration in 1067 with the same chronological errors, and states that S. Guido transferred the cathedral from the church of S. Pietro to a loftier situation, built it from its foundations, gave it the new name of the Assunta, and provided it with a chapter of canons.16 It is evident that the pious author, in his religious zeal, 12 Divinae namque providentiae nutu, qua Christi miles diregebatur in omnibus, nobilissimam Christi matrem, ac reverentissimam suis expensis aedificavit ecclesiam, in qua nunc sedes Episcopalis est, & illam fecit solemniter consecrari a venerabilibus Episcopis Petro Terdonensi viro per omnia laudabili, & Alberto Januensi tertio idus novembris anno incarnationis Jesu Christi millesimo sexagesimo septimo inditione sexta domino Henrico III. Imperatore regnante. . . . (Yita B. Guidoni Aquensis Episcopi auctore Laurentio Calceato Aq. circa an. 1260 conscripta., Cap. XVI, ed. Moriondo, II, 99). 13 It is known that Pietro I was bishop of Tortona 1022-1068. See Moriondo, II, 100. Oberto, not Alberto, was bishop of Genoa in 1067. (Ughelli, IV, 844). i* . . . idibus hoc fit officium consecrationis hujusce Ecclesiae maj oris . . . (Biorci, I, 189-190). 15 This fact is witnessed by such passages as : . . . haec omnia approbante Impera tore Longobardorum, anno ab Incarnatione Christi millesimo sexagesimo quinto, con- stantibus de hisce omnibus etiam hodie privilegiis et scripturis authenticis. . . . Corpore ipsius . . . marmoreaque hie in area usque in praesens recondito, meritissime veneratur. i« Suscepto igitur Pastorali regimine, quod mira devotione simul et auctoritate regebat . . . universo patrimonio, quod habebat, Cathedralem imprimis ecclesiam, ante sub titulo S. Petri in dicta civitate nimis antiquatam, et aeris intemperie fere inhabilem, eminentiori dictae civitatis loco, ampliorem et ornatiorem sub invocatione Assumptionis Beatae Virginis, quam Advocatam excorde gerebat, ab imis fundamentis instituit proprio aere; illamque Archidiaconatu, Praepositura, Archipresbyteratu, et Canonica- tibus duodecim exornavit et auxit, prout hodie adhuc cernitur: accersitisque una Reverendissimis Dominis Episcopo Dertonensi et Oberto Episcopo Januensi, viris undequaque laudabilibus et dignissimis, earn devote consecrarunt, anno a Nativitate Domini millesimo sexagesimo septimo; Indictione sexta, Henrico Imperatore tertio. (Vita S. Guidonis, ed. Jean de Bolland, Acta Sanctorum, Junii die secunda, I, Cf. also Moriondo, II, 110. 16 ACQUI, CATHEDRAL has attributed to S. Guido certain acts really performed by his predecessors, Primo and Dudone. The office for the feast of S. Guido in the church of Acqui contains three different references to the construction of the church by the saint.17 Similar notices, derived from one of the lives of the saint, are repeated by the late and inaccurate Fra Jacopo da Acqui.18 In later times the tradition that S. Guido constructed the cathedral was universally accepted. It is, for example, recorded on two inscriptions on the western portal added to the cathedral in 1481. 10 Similarly in an inscription of 1655 in the choir, S. Guido is referred to as the founder of this basilica,20 and in an inscription in the episcopal palace, I know not of what time, but evidently late, he is said to have built the cathedral at his own expense.21 About 1177 the diocese of Acqui suffered a severe blow when, at its expense, was created the new see of Alessandria. A long struggle ensued between the rival cities. A bull of Innocent III of 1198, which ordered that 17 Ecce Sacerdos magnus, qui in diebus suis aedificavit domum, & exaltavit templum sacrum Domino paratum in gloria sempiterna. (Officium in festo B. Guidonis Aquensis Episcopi, ed. Moriondo, II, 104). Templum fecit templi cultor honoris eximii quo completo fine loeto cursum hujus seculi consummavit, & intravit in gaudia Domini (Ibid., 107). Ecce Sacerdos magnus, qui in vita sua suffulsit domum, & in diebus suis corroboravit templum &c. (Ibid., 108). Cf. Eccli., 1, 1. is [sanctus Guido] facit suis expensis et Ecclesie [sic] ecclesiam maiorem sancte Marie matris Dei quam mirifice ornavit. clericis officio libris divinis et paramentis. ecclesia quasi completa ... ad extremam horam devenit. Et . . . spiritum suum suo tradidit Salvatori. in ecclesia sua quam edificavit. in archa marmorea collocatur ubi devotis cottidie crebra fiunt miracula de quibus plenam fidem invenies in legenda sua que servatur in sacristia ecclesie predicte aquensis civitatis. (Frate Jacopo da Acqui, Chronicon Imaginis Mundi, ed. Hist. Pat. Mon., V, 1548). Cf. also Moriondo, II, 142. 19 HOC TEPLV . ASSVPTE . COSTRVXIT . WIDO . MARIE WIDO . VENlis COMES . AQ SANE . ET AQVN . EPS . HAC . PROPIO ERE . CoSTRVXIT . ET . DOTAVIT . ECCLEXIAM AD . HOREM . VIRGINIS . ET.IN.EA.' REQUIESCIT . V . F . Ao . DI (— viri facti anno domini) M . L X VII 17 LOMBARD ARCHITECTURE the episcopal seat of Acqui should be transferred to Alessandria, precipitated a war between the two towns.22 In 1202 Innocent III settled the dispute by uniting Alessandria and Acqui in a single diocese, of which Alessandria was the primary seat and Acqui the second. The bishop resided six months alternately in each,23 but in 1405 the two dioceses were again separated and Acqui regained the rank of an episcopal city, albeit with sadly diminished prestige. In 1479 the bells of the cathedral were transferred from the old to the new campanile.24 From this fact it may be inferred that the old Romanesque campanile was destroyed about this time, and that the new campanile, which still exists, was finished in the year 1479. Two years later the existing western portal was erected by Giovanni da Pillacorte, son of Antonio of Val Cavargna, Lugano.25 Pillacorte is possibly the name of the native town of Giovanni, but where it is I have not been able to discover. .Still another inscription records that the same portal was made in the time of Tommaso de Regibus, of Alba, bishop of Acqui, who also caused the adjoining episcopal palace to 20 D . O . M CIVITATI . IMMINENTIBVS . IAM . BELLORVM . PERICVLIS . S . GVIDO . EP1SCOPVS . HVIVS . BASILIOE . FVNDATOR NATALfV . NOBILITATE . INSIGNIS . PIETATE . CONSPICVVS . INCOPARABILI . CHARITATE . EXITIA . C^EDESQ . AVERTIT . VOVIT . VNIVERSVS . POPVLVS . AQVENSIS . NOVISSIME . TANTVM . INVOCAVIT . ALLOBROGES . INTERCESSOR . RETRAXIT . ITALOS . TEVTONES . ET . IBEROS . E . MVRIS . PROTECTOR . DEIECIT RARI . CIVES . IN . PVGNA . OMNES . INCOLVMNES . ORNATIORI . ARA . PR^ESTANTIORI . VRNA . VOTVM . EXOLVERVNT . MDCLV . 21 S . WIDO . DOMO . AQVESANA TEMPLVM . MAXIMVM . B . V . ASSVMPT^E . AERE . SVO CONSTRVXIT. (Biorci, I, 191). 22 G. Schiavinae, Annales Alexandrini, ed. Hist. Pat. Mon., XI, 92. 23 Lavezzari, 43, 45. 2*Eo anno [1479] die prima septembris presbiteri majores Aquensis Ecclesia? permutaverunt campanas de campanili veteri super novo luna existente in oppositione cum sole & sequenti die Bartolomeus Carpentarius, qui dictas campanas collo- cavit in dicto campanili novo, cecidit, & mortuus est. (Notae historicae repertae in quodam antiquo Codice Can. Gabrielis Chiabrerae Praepositi Capituli Aquensis, ed. Moriondo, II, 265). 18 ACQUI, CATHEDRAL be erected. This is entirely consistent with the preceding inscription, since it is known that Tommaso de Regibus died in 1484. The Maggiorino, who is invoked with S. Guido in the first part of the inscription, is reputed the first bishop of Acqui, and is revered as a saint. His statue and that of S. Guido are placed in relief on either side of the portal and distinguished by the inscriptions S. MAIORINVS and S. WIDO.26 In the west gallery of the cloister on the second floor is an inscription which states that Costantino Marenco27 built from the foundations the houses of the canons at his expense, March 24, 1495.28 On the west side of the 25 MCCCCLXXXI HOC. OP' IM PRESSIT DE PIL LACVRTE . IOHA . NES . QVE . TV LIT . ANTONIV VALIS . CAROA LVGANI . 26 MAIORINE PSVL . POPVLO QVOQ' . GRAT AQVESI . NVC . CV . VVIDONE SALVA COGE DO . NOCETES . HEc . PORTA . FACTA . TPR . D . TOME . DE . REGI BV DE . ALBA . EPI AQN . QVI . PALA CIV . VNA . CV DOMO. [CON] TIGVA FIERI . FECIT 27 For an account of this bishop, see Lavezzari, 91. 2s 4. CONSTANTINVS . MARENCVS . IS TIVS . AQVENSIS . ECCLEXIE . PASTOR . RESTORATORQ' . HAS . CHANONICH ALES . MANSIONES . SVO . ERRE . A . FON DAMENTIS . ERRESIT . . SVB . DOMINI .1.4.9.5. DIE . 24 . MARCH 19 LOMBARD ARCHITECTURE canons' house is another inscription from which it appears that Costantino Marenco reconstructed not only the houses of the canons, but also the cloister.29 At the close of the Cinquecento began the baroccoization of the church. Francesco dei Conti di S. Giorgio e Biandrate, bishop of Acqui (1585-1600), was the first who seriously undertook to change the church "da spelonca di ladri in casa di Dio."30 His principal achievement appears to have been the whitewashing of the nave.31 His successor, Camillo Beccio, added the western portal in front of the cathedral and the graceful colonnade of the episcopal palace. A manuscript of Blesi, written about 1614, but containing several later annotations and additions, gives an interesting description of the church as it emerged from these alterations.32 In 1618 the body of S. Guido was translated 29 R' D' IN . XPO . DNS . CONSTATIN' . MARENCHVS . CIVIS ET . EPS . AQVEN . AC COES . HAS EDES.CV CLAVSTRO P[ER] CANONICOR[VM] ALIOR[VM]Q[VE] HVP ECCLIE SACERDOTV RESIDENTIA . SVA IMPENSA . A FODAMETIS ERREXIT MCCCCLXXXXV 30 Biorci, I, 233-234. 3i Lavezzari, 255. 32 [La Cattedrale] quale fu fatta fabbricare dalla Santa memoria del Beato Guido vescovo e prottettore della medma Citta, con bellissima regola d'architettura, e giusta proporzione, tutta di pietre forti piccate a scalpello, e che la rendevano magnifica, e riguardevole, e sebbene fu bonissima l'intenzione di Monsignor Illmo S. Giorgio vescovo di far imbiancar da Chiesa le ha nondimeno scemato assai d'onorevolezza, et antichiti che dimostrava. Rende per6 riguardevole La medma Chiesa una piazza detta del Duomo. . . . E nella medma Chiesa un orotorio ossia Confessione, ovvero il Scurolo, il piii bello, e meglio disposto, e proporzionato, che io abbia a quest'ora visto in altra CittEl e talmente assicurato sopra una quantita di collonne di Pietra, che e molto lodato da chi giudiciosame lo considera. Ha di piu la Cattedrale preda la canonica congionta ad essa con Claustri, e stanze comode alii Sig" Cannonici, che quelle servono anco tal volta a Cittadini Privati di molta conversazione [sic, considerazione]. Aggiongono poi molto di magnificenza a qta Chiesa due scale commodissime, una nell'entrare che resta sulla detta piazza del Duomo, e l'altra che separa il Coro et Altare maggiore con l'ala a traverso del Corpo piu grande d'essa Chiesa. La quale fu dedicata dall'istesso Beato Guido alia Assonzione della Gloriosissima vergine Maria Nostra Signora, et in da Chiesa oltre i corpi del Beato Guido sud. e del Beato Majorino, sono altre relliquie. . . . Addizione. — Questa piazza che dice il nostro autore esser di competente grandezza, al presente si trova poco nieno che del tutto occupato da un antiporto fatto fare, e solame principiato avanti detta Chiesa dal medmo Revmo vescovo della Citta. Li claustri, e Canonica furono fatti edificare da Monsig. Costantino Marenco, Vescovo della Citta, ad effetto che li Sgi Canonici dovessero in quelle Stanze far residenza . . . ma di presente ne restano esclusi. . . . La Seguente memoria di Monsige Bicuti, e stata ritrovata in un manuscritto. Monsige vescovo Gio' Ambroggio Bicuti, ha primierame fatto lastricare 20 ACQUI, CATHEDRAL from the crypt into the upper church.33 From 1644 to 1655 was erected a new chapel of S. Guido in accordance with the vow made by the citizens thirty years before.34 The bishop Giovanni Ambrogio di Bicuti (1647-1675) at his own expense raised the pavement of the choir and lateral chapels, and adorned the choir with stuccos and paintings. He also remodelled the dome over the crossing.35 In 1709 the chapel of S. Guido was repaired. In 1845, in the course of restorations in the choir, there came to light in the crypt a mosaic pavement. The chapter of Acqui, moved by loyalty rather than by good judgment, tore up this mosaic from the spot in which it had been discovered, and sent it as a gift to the king at Turin, where for a long time it was stored away in the cellar of the royal palace. The project of Promis, who wished to place it in the library of the University of Turin, happily came to nought, for had he done so it would doubtless have perished with the other treasures of that institution in the lamentable fire. Recently the mosaic has been placed in the basement of the Museo Civico di Arte Applicata ad Industria. The cathedral of Acqui has suffered severely in the loss of its precious mosaic and in the barocco restorations of which it has been the unfortunate victim. However, it has had the rare good fortune to escape a modern restoration, and the Romanesque structure wherever it appears is still genuine and unfalsified. III. The edifice consisted originally of a nave six bays long, two side aisles, projecting transepts with eastern absidioles, a choir flanked by side aisles, three apses and a crypt extending beneath the transepts and choir. Numerous alterations have, however, somewhat modified this plan. The western bay of the southern side aisle has been blocked up by the walls of the XV century campanile.36 A complete set of side chapels was added to the church in the time of the Renaissance, destroying the old side-aisle walls, of which, however, some bits, with fragments of their corbel-tables, are still di Chiapponi quadrati L'Antiporto, et con sassi, e suoi cordoni fatto far fuori, et intorno di esso con bella architettura una cordonata, ossia Lastricatura, et indi ha fatto adornare di stucco, e Pitture il coro di essa Chiesa Cattedle nell'anno 1668, et fatto fare due Cantorie corrisponti et accomodar la Camera sopra la Capella della Ssma Annunciata. 33 1618. Corpo del nostro Protettore S. Guido si colloca nella Cappella maggiore della Confessione del Duomo. Convocati 27 Maggio, e 5 Giugno 1618. II sito e umido, si trasporta al disopra nella Cappella detta de' Santi. (Extract from the Municipal Archives, cit. Biorci, I, 203). 34 Biorci, I, 203. ss Ibid., II, 257. 36 The upper part of this campanile rests on the top of the clearstory wall. S. Guido's masonry must have been very substantial to enable it to bear all this extra weight. 21 LOMBARD ARCHITECTURE preserved under the aisle roofs. Finally, the absidioles of the transepts have been destroyed. The fact that they formerly existed is proved only by the lower story still preserved in the crypt. The nave has been entirely covered with stucco, so that it is exceedingly difficult to trace the original dispositions. However, parts of the bases of the original piers, still visible here and there, prove that the latter had a compound section consisting of a rectangular pilaster strip towards the nave, two half columns to the east and west, and a half column engaged on a pilaster strip towards the side aisles. In all probability the pilaster strip towards the nave did not support a system, but the second order of the archivolts, as at S. Maria Canale, Tortona (Plate 211, Fig. 5). The piers of the crossing had a similar section, but were heavier and had some extra members. The responds of the side aisles probably consisted of a simple pilaster strip, but this point cannot be definitely determined. The side aisles were doubtless covered with groin vaults. The existing vaults of the side aisles show great variation of form and perhaps — even probably — some of the original ones still survive, but so thick is the coating of intonaco that it is impossible to be certain. The transverse arches of the side aisles were, I believe, in one order. The nave was not vaulted, but roofed in wood. The ancient clearstory walls are still perfectly preserved above the modern vaults which cut across the old clearstory windows. These clearstory windows are peculiarly spaced, in that there are only three on each side, corresponding to six bays. Above the modern vaults the wall of the nave on the interior shows no signs of a system or of vaults, a clear indication that no vaults were ever erected. It is evident that the wall of the south transept has been raised. Externally there are two rows of arched corbel-tables (Plate 2, Fig. 5), of which the lower one corresponds to the symmetrical arched corbel-table of the other transept. Under the roof may still be seen the old arched corbel- tables of the south wall of the crossing, which originally were visible above the transept roof. The wall above and the second arched corbel-table are of inferior masonry. It is evident that the wall was raised at the time the barocco vaults of the transepts were erected. There is nothing to prove that the transepts were originally vaulted, and nothing to indicate how the crossing was treated, but it may be conjectured that the latter was surmounted by an octagonal cloistered vault. The choir vaults appear to be modern, but it is not improbable that they replace similar barrel vaults of the ancient edifice. The crypt still retains its original groin vaults, although the greater number of them have been made over or at least covered with intonaco. Especially under the north transept the original forms of the crypt vault may still be traced. It is evident that the vaults were not domed, but formed of bricks laid in courses generally approximately regular and separated by wide mortar- beds. The masonry is nevertheless very crude. There were disappearing 22 ACQUI, CATHEDRAL transverse arches but apparently no wall ribs. The church possessed no buttresses. The campanile, Gothic in style, is evidently later than the church. A proof of this, were other needed besides the style of the architecture and the character of the masonry, is to be found in the fact that the arched corbel- tables which originally crowned the south clearstory wall are still visible inside the campanile. The masonry throughout consists of crudely shaped pieces of stone of about the size of bricks, with an occasional brick (bricks are used especially in the archivolts and the upper portions of the edifice), laid in thick beds of mortar. The courses are in general horizontal but often are irregular. There are numerous square scaffolding holes. To the south of the cathedral are the graceful cloisters (Plate 3, Fig. 3), entirely rebuilt in 1495, but in part with the ancient columns and capitals of S. Guido. IV. The capitals of the nave piers have, for the most part, been so denatured in the barocco period that it is impossible to judge of their original style, though from the stunted proportions of some of them it is possible to infer that they were of the same type as the capitals of the upper gallery of the cloister. In the two western piers of the crossing, however, the old Lombard capitals are still preserved above the new stucco capitals of the Renaissance, having been spared because they were hidden by the gallery which runs around the church on top of the cornice. They are of a rudimentary bell type, with the angles slightly indented or foliated. The capitals of the crypt are all alike and of a similar type, but somewhat shallower. More interesting are the capitals of the cloister (Plate 3, Fig. 3). Those of the upper story recall the capitals of Lodi Vecchio, and are characterized by shallow, plain abaci, with grotesque heads or leaves in the angles, and some times a rosette in the centre of each face. One is a crude imitation of the classical Ionic. Those of the lower story and of the west gallery of the second story have broad, flat leaves, volutes in the angles, acanthus leaves stiffly carved under the volutes and a bulbous projection in the middle of each face, something between a volute and a leaf. They are evidently of the XV century (Plate 3, Fig. 4). In the west gallery of the cloister in the lower story are gathered together several carved fragments of interest. One is a capital like those of the upper story of the cloister (Plate 4, Fig. 1) ; another is a capital carved with a representation of the Resurrection (Plate 3, Fig. 5). At one angle is seen St. Peter with a key and a book. In the centre of the front face Christ rises from the tomb, naked, and with an inscribed halo. The haloed and draped figure to His right with clasped hands must be the Virgin. To the left stands the Magdalen carrying a vase. These rough figures must have belonged to a capital of the cloister of S. Guido of 1067, and are of great 23 LOMBARD ARCHITECTURE interest as the first extant serious figure sculptures in stone of an iconographic subject made in northern Italy after the famous altar (744-749) of Ratchis at Cividale (Plate 3, Fig. 2). However crude, these sculptures, therefore, mark a new era in plastic art. The central apse (Plate 2, Fig. 5; Plate 3, Fig. 1) possesses a cornice formed of arched corbel-tables in two orders and surmounted by an. ornament consisting of a saw tooth and two rows of zigzags like those of Fontanella and S. Pietro in Vallate. In the principal apse the pilaster strips are in two orders. Elsewhere the edifice is crowned by cornices consisting of arched corbel-tables rising from pilaster strips and surmounted by saw teeth. In the absidioles the arched corbel-tables are grouped two and two (Plate 2, Fig. 6), but elsewhere the pilaster strips are placed at much rarer intervals (Plate 2, Fig. 5; Plate 3, Fig. 1 ; Plate 4, Fig. 4). In the facade of the arcade of the episcopal palace facing the Piazza del Duomo are several fragmentary reliefs. One represents St. Lawrence with his grill, another St. Stephen, distinguished by a palm and a book, a third S. Guido in episcopal robes and holding a model of the church, the others St. Paul, St. Peter and St. Bartholomew — the latter, however, represented with a hatchet instead of with a knife. These sculptures appear to be of the XII century. The mosaic in the Turin museum is finely executed in black and white. The designs are for the most part conventional or grotesque, such as birds, fishes, a devil, an archer shooting a camel with an arrow, and a dragon. The only biblical subject is Jonah and the whale, represented in the early Christian manner, the whale being a winged and footed dragon, who holds Jonah's foot in his mouth.37 Curious above all is the representation of the flight of Icarus. Icarus with his hair arranged in a manner that recalls a cock's comb, and with a strange wing, is seen falling. Below is the inscription [V]OL[ITVS] : ICAR.38 It is noticeable that these mosaics are more evenly executed than the neighbouring ones coming from the church of S. Salutore. V. As an authentically dated edifice begun before 1018 and consecrated in 1067, the cathedral of Acqui is a most important monument of Lombard architecture. si Compare the relief in the cathedral of Verona (Plate 216, Fig. 5). ss Fabretti saw in this mosaic the letters MID, of which I can discover no trace. He also made out » ship which he said was misunderstood as a dead panther by Aus'm Weerth. 24 ACQUI, S. PIETRO ACQUI,1 S. PIETRO (Plate 4, Fig. 2, 3, 5 ; Plate 5, Fig. 1) I. With the exception of a very brief reference in an article by Pellate, the architecture of the church of S. Pietro at Acqui had never been referred to in print until I published, in 1913, a short account in the American Architect. For the history of the edifice the classic work of Moriondo is indispensable, and the historians of Acqui, Biorci and Lavezzari, have also contributed much that is important. Savio, with his fine scholarship and extraordinary perspi cacity, has solved several puzzles of ecclesiastical history bearing directly upon the vicissitudes of S. Pietro. Finally I found much valuable information in the book of Blesi, a manuscript copy of which was shown me by Canonico Vincenzo Maccid.2 II. Although one may feel some hesitation in following Biorci3 in believ ing that the diocese of Acqui was founded as early as the III century, there is still reason to suppose that it dates from a venerable antiquity. In the cemetery of S. Pietro was found a Christian inscription of the year 432,4 which probably indicates that the cathedral church existed on the site of S. Pietro as early as the V century. Savio states that before 1023 the bishops of Acqui were buried at S. Pietro.5 This assertion is perhaps a little sweeping, but it is certain that at S. Pietro were buried many of the early bishops, including the first three, Maggiorino, Massimo and Severo;6 the problematical bishop Tito;7 Ditario (t 488) ;8 Gotofredo, the eleventh, and Arnaldo, the thirteenth, bishops of Acqui.9 In the time of Pedrocca there were to be seen in S. Pietro statues of SS. Maggiorino and Tito.10 It is, moreover, clear that before 1023 the cathedral was known as S. Pietro, for it is cited under this title in three documents of 891, 978 and 996.11 The conclusion is therefore justified that until the year 1023 the cathedral was situated at S. Pietro. i (Alessandria). 2 To the latter I am indebted also for the use of his unique library during the three weeks that I was at Acqui, as well as for much valuable information and innu merable acts of courtesy and kindness. 3 I, 91. * Ibid., I, 107-108. 5 Savio, 27-28. 6 Catalogue of Pedrocca, ed. Savio, 10. 'His epitaph was found in 1753 at S. Pietro. (Biorci, I, 95). s Biorci, I, 111. - Pedrocca, ed. Savio, 10; see also Lavezzari, 254-255. io Savio, 17. n in Episcopatu Aquensi, videlicet in honorem D. Petri Apostolorum principis dedicato . . . (Diploma of Wido, 891, ed. Moriondo, I, 2). . . .jam dicto Epis copo in honorem S. Petri dicato. . . . (Diploma of Otto II, 978, ibid., 7). The same phrase occurs in a diploma of Otto III of 996, ibid., 15. 25 LOMBARD ARCHITECTURE The statement in certain old chroniclers that the monastery of S. Pietro was founded in the time of the Lombard kings merits no faith, and probably originated in a tradition of a reconstruction of the cathedral church in the VI or VII century. This notice is first given by the fabulous chronicler Jacopo da Acqui, who assigns the reconstruction to about 595 A. D. and states that the altar was consecrated by angels.12 This is doubtless the ancient manuscript from which Della Chiesa derived his information13 and from which the tradition has been spread by numerous other writers. It deserves no greater belief than many of the other fabulous accounts of the uncritical Fra Jacopo. From the ancient catalogue of the bishops of Acqui written by Pedrocca early in the XI century and published by Savio,14 we learn that Bishop Primo built a cathedral church; that he first established a canonica in it and that he built another church of S. Pietro without the walls of the city. It appears furthermore that this bishop was the first of all the bishops of Acqui to be buried at S. Maria Maggiore.15 Now Primo occupied the episcopal throne of Acqui, according to Savio, from 989-1018. It seems probable, therefore, that what happened was this. Primo commenced to build a new cathedral on a new site, that is to say, the cathedral of Acqui as it stands to-day. Not only did he move the cathedral from S. Pietro to the new church, but he reformed his clergy as well, and established for the first time in the diocese of Acqui a chapter of canons regular. He also rebuilt the old cathedral church of S. Pietro, doubtless with the idea of founding there a new monastery. The extraordinary building activity of Primo may be best understood in the light of the history of the see of Acqui at the end of the X century. Acqui had been one of the few dioceses of northern Italy which had remained faithful to the emperor, Henry II, throughout the long period of wars which closed in 1014. This fidelity was richly rewarded. In a diploma of 1014 granted by Henry II in favour of the bishop of Acqui, there is omitted the clause almost constantly used in the earlier diplomas in favour of the diocese, referring to 12 In civitate Aquis de Lombardia est factum monasterium monachorum sancti Petri cum maxima et honorabili devotione in suburbio civitatis. et habetur ibi. quod maius altare illius ecclesie fuit ab angelis consecratum. et ibi multa iacent corpora Sanctorum. (Frate Jacopo da Acqui, Chronicon Imaginis Mundi, ed. Hist. Pat. Mon., V, 1449). Published also by Moriondo, II, 134. 13 Noi habbiamo ritrovato in alcuni libri antichissimi, scritti a penna da i sudetti Re [lombardi] esser stati fondati in essa Provincia molti monastery in honore di S. Pietro: tra gl'altri il Monasterio di Acqui . . . (Della Chiesa, 52). i*10. is Primus vener. Epus qui sedit ann. XXVIII M III. et D. XVI. 6. X Kal. aprilis et cessavit episcopatus M. Ill et dies VI. Hie Ecclesiam episcopalem funditus edificavit et Canonicam primum constituit, et aliam foris muros in honorem Apostolorum Principis, Aquensis Episcopii defensoris, doctoris et magistri. Hie etiam Heribertum Mediolanensis Eccie. Episcopum consecravit. Requiescit ad S. Mariam Maiorem. (Ed. Savio, 10). 26 ACQUI, S. PIETRO the poverty of the church of Acqui. It appears that, thanks to the donations, not only of the emperors, but also of other pious persons (such as, for example, the marchese Guglielmo and Riprando, who in 999 made notable gifts to the diocese of Acqui), the bishops found themselves in conditions of extraordinary financial prosperity. Hence it was that a remarkable series of building enter prises was undertaken.16 The tradition that the church of S. Pietro was begun by Primo was recorded in an inscription in the episcopal palace seen and cited by Biorci.17 Primo was succeeded by the bishop Brunengo, who occupied the episcopal throne for the brief space of four years (1018-1022), and by Dudone (1023-1033). The catalogue of Pedrocca says of the latter bishop that he was the first to celebrate mass at S. Maria, the new cathedral church, that he transferred thither the canons of S. Pietro and that he estab lished a monastery in the latter church18 — all this at Christmas, presumably in the first year of his episcopate. The catalogue of Pedrocca is confirmed in regard to the foundation of the monastery at S. Pietro by Dudone by a document of S. Guido of 1041, which explicitly mentions that the monastery of S. Pietro was founded by Dudone.19 It is still further confirmed by a legal document of 1224 in regard to a lawsuit over the parish rights of S. Pietro, claimed by both the canons of the cathedral and the Benedictine monks of S. Pietro. The lawyer of the monks asserted that in the church of S. Pietro there had anciently been secular canons and not monks, and his adversary replied denying that there had been canons two hundred years before, but admitting that there had been two hundred and twenty years before; that is to say, he claimed that the monastery of S. Pietro was established between 1004 and 1024, which coincides perfectly with the date of foundation (1023) given by Pedrocca in his catalogue.20 Still further confirmation, were any 16 Biorci, I, 177. it Ibid. i8 Dudo qui et Petrus bonae memoriae Episcopus. . . . Nativitatis Domini primam ad S. Mariam, antiquitus episcopalem ecclesiam, missam celebrare, Sanctique Petri Ecclesiae Canonicos ad istam transvexit, et de ilia Monasterium fecit. (Cit. Savio, 10). io In nomine Domini Dei nostri & Salvatoris Jesu Christi. Wido favente divina dementia sanctae Aquensis Sedis Episcopus. . . . Sit ergo notum omnibus Ecclesiis Dei, omnique Laj eorum fidelium conventui, quod Monasterium sancti Petri, quod in suburbio civitatis Aquensis, in qua auctore Deo sedem Pontificalem, quamvis indigne habemus, situm scilicet a D. Dudone bona; memoriae antecessore nostro Episcopo, in primis aliquid de rebus etc. . . . Actum anno incarnationis Domini Nostri Jesu Christi millesimo quadragesimo primo, indictione secunda [sic] ; Enrico regnante anno ejus secundo. . . . (Ed. Moriondo, I, 28). 20 Exposuit Magister Otto Sindicus, & procurator monasterii S. Petri Aquis . . . quod prsedictum monasterium est circumdatum habitationibus hominum undique. . . . Item ponit, quod in Ecclesia S. Petri fuerunt canonici seculares, & non monaci. Ad quod respondet, quod non concedit a ducentis annis infra, sed a ducentis viginti annis infra concedit. . . . Item ponit, quod ipse Episcopus [Dudo] jacet ad prsedictum monasterium. (Moriondo, I, 177). 27 LOMBARD ARCHITECTURE needed, is given by the fact that Dudone was buried at S. Pietro, that is, in the church of the monastery he had founded.21 The cathedral chapter and monks of S. Pietro, thus closely associated in origin, had many rights which were not clearly defined, and which resulted in subsequent times in endless litigation. As early as 1108 these lawsuits were in full progress.22 The subsequent history of the abbey was comparatively uneventful, and the documents seem to indicate that the monastery enjoyed for some centuries a prosperous and tranquil life. The commendam was established sometime after 141 5. 23 In 1663 the episcopal visitor seems to have been little edified with the condition in which he found the monastery, to judge from the inedited acts of the proceedings.24 The bishop Carlo Antonio Gozzani (1625-1721) dispersed the monks entirely and divided the church into two parts, one of which he desecrated and the other of which he dedicated to S. Maria Vergine Addolorata, after redecorating it in the barocco style.25 It is in this condition that the edifice has come down to our own time, having had the rare good fortune to escape modern restoration. III. The church originally consisted of a nave, two side aisles, a choir flanked by side aisles, three apses (Plate 4, Fig. 2), a crypt and an octagonal campanile (Plate 5, Fig. 1) rising over the southern side aisle of the choir. 21 In the face of all this trustworthy evidence it is impossible to lend faith to the rather vague tradition that fhe cathedral was built by S. Guido. The work of S. Guido must have been merely to complete and consecrate it. See above, pp. 16-17. 22 Moriondo, I, 45. Cf. ibid., 72, 89, 90. 23 Lavezzari, 255. 2* Delli disciplini di S. Pietro del do luogo. Si prouedino qto pra tosto quelli fratelli della loro regola di Milano, et procurino ad ogni suo puotere di ben essiguirla, et agiutino il Rettore della Chiesa la festa in insegnare la vita Christiana. Non si facciano piu bacanali dell'entrate o elemosine della Compania di San Spirito ma li diuidano a poueri confonne al decreto gnale di qsta visita. (Visita Apos- tolica, inedited document of 1663). 25Blesi's description of this church is not without interest: La chiesa di S. Pietro e posta nel Borgo maggiore della Citta, qual dall'istessa Chiesa, si chiama Borgo di S. Pietro, e molto antica, ^_tengo che sii la piu antica Fabbrica che si trovi oggidi nella Citta, et a vederla solame ne pub restar certo chi si voglia. Fu fatta fabbricare dal re de' Longobardi, et con la Chiesa maggiore, et Cattedrale, et dalli niedmi Re fu data, et assegnata entrata di qualita. Dopo che dal Beato Guido per accrescimento del culto Divino, et maggior comodita de' Cittadini et onorevolezza della Citta fu fatta edificare l'altra della quale abbiamo detto di sopra, fu data alia Religione de' Padri di S. Benedetto, con l'entrata, e Titolo d' Abbazia, ma essendo anco dai Padri lasciata e stata conferta a diversi Abbati Secolari. Finalme essendo vaccata nel tempo che monsige Costacciara era vescovo della nostra Citta, procur6 Egli come desiderossissimo di augmentare l'entrate del vescovato d'unirla a quello, et per effettuare tale unione, fidendo molto piu nell'amicizia del Sig. Francesco. ... In questa Chiesa si vedono alcune pietre antiche di marmi, e d'altra sorte con i nomi di quelli antiehi Romani. Avanti l'ultima amplificazione della Citta, che segui del 1400, restava detta Chiesa fuori delle mura, ma ora resta nella Citta. . . . 28 ACQUI, S. PIETRO The edifice has, however, been much changed. The barocco church of the Addolorata has been built in the western bays of the old nave, and the rest of the building has been transformed into houses, warerooms and stores (Plate 4, Fig. 5). The ancient architecture is best preserved in the three apses and in the clearstory walls of the flanks (Plate 4, Fig. 2, 5). The choir, which was lower than the nave, must have been vaulted; — in fact, some traces of the groin vault which covered the north side aisle of the choir were still extant when I studied the church. The nave was not vaulted, as may be clearly seen from the ancient walls still preserved above the vaults of the existing church, and it is evident also that there was no system in the nave. It is possible that the side aisles in the nave may have been vaulted, but I believe that they were not. The piers were octagonal, with plain, square capitals, and apparently no bases (at least none are visible). Several of these piers are still to be seen in the north arcade. Of the ancient crypt nothing is extant save some few fragments of old masonry sufficient to prove its existence. Otherwise the cellars under the church are entirely modern. The clearstory (Plate 4, Fig. 5) had large, round-headed windows, doubtless glazed. They were placed at great intervals, as in the cathedral. A curious feature of the church is the polygonal form given externally to the absidioles (Plate 4, Fig. 2 ; Plate 5, Fig. 1). With the exceptions of the apses of VI century churches in Ravenna and of the sacristy of SS. Felice e Fortunato at Vicenza I know of no other example of a similar construction in northern Italy. The masonry is formed of uncut brick-shaped stones, with a few bricks roughly laid in thick beds of mortar (Plate 4, Fig. 3). Some attempt is made to keep the courses horizontal. In the clearstory the masonry is even rougher than in the apses. Horizontal courses are abandoned, and the wall is frankly rubble. IV. The nave piers are without capitals, and the octagonal supporting member slides into the rectangular load merely with a sort of bevel. The clearstory walls are decorated with arched corbel-tables grouped two and two (Plate 4, Fig. 5), and the side-aisle walls had also the same decoration, as may be seen from the few fragments which still survive in the south wall near the campanile (Plate 4, Fig. 5). The absidioles were decorated with blind arches resting on a podium (Plate 4, Fig. 2, 3), but in the central apse (Plate 4, Fig. 2) the pilaster strips separating the arches were in three cases interrupted by windows, thus forming arched corbel-tables grouped two and two. The existing cornice of the central apse (Plate 4, Fig. 2), formed of flat corbel-tables, dentils and saw teeth, is obviously an addition of the XIII century. The original cornice of the absidioles, however, consisting of a single cavea, is preserved. The clearstory walls are, and perhaps always have been, without a cornice. The campanile (Plate 5, Fig. 1) has pilaster strips at the angles, rising ?rom a podium crowned by a string-course of saw 29 LOMBARD ARCHITECTURE teeth. The buttress that marks the end of the south wall is decorated with an ornament consisting of two vertical zigzags (Plate 5, Fig. 1). V. The primitive style of the architecture is completely in accord with the documentary evidence that the church was begun by Primo between 989 and 1018, probably c. 1015, and was already finished when Dudone established the monastery in 1023. AGLIATE DI CARATE BRIANZA,1 BATTISTERO (Plate 5, Fig. 5, 6; Plate 6; Plate 7) I. Practically all the works mentioned below in connection with S. Pietro of Agliate treat also of the baptistery, which forms an interesting complement and almost a part of the church itself. II. There are no historical documents which throw light upon the archaeology of the baptistery, besides those already mentioned below in connection with the basilica. The baptistery, however, enjoys one great advantage over the church in that it escaped restoration in the fatal campaign of 1890-1895. We here find ourselves, therefore, face to face with a document, mutilated, it is true, and much consumed by age, but without the modern falsifications which so largely destroy the scientific value of S. Pietro. III. The baptistery is a severely simple but irregular structure, in plan (Plate 6) a nine-sided polygon, two of whose sides have been replaced in the lower story by an apse. The walls are of enormous thickness (Plate 6; Plate 7), and the edifice displays surprising asymmetries in the placing of its windows (Plate 6) and other details. It is covered at present by a cloistered vault, but this has been remade, as is shown by the character of the masonry, after the original construction of the church and probably in the XII century. The monument suffered in the restoration of the XVIII century, when the windows were walled up, the west wall in part rebuilt, the doorways altered and the interior covered with intonaco. The walls, like those of the basilica, are constructed of rubble (Plate 5, Fig. 5), consisting of stones, pebbles and pieces of brick laid loosely in a mass of mortar. However, this masonry is more advanced than that of the church (Plate 5, Fig. 7). The stones are laid more carefully, there is more of an approximation towards horizontal courses, and the mortar-beds are narrower. The masonry of the upper part of the exterior wall above the arched corbel- table is better and later than that below (Plate 5, Fig. 6). Here, the hori zontal courses are distinctly maintained, and the stones are carefully fitted i (Milano). 30 AGLIATE DI CARATE BRIANZA, S. PIETRO together (Plate 5, Fig. 5). This masonry, executed in imitation of the earlier masonry below, doubtless belongs to a restoration of the XII century. IV. The ornament of the church consists merely of the cornice. Above, there is a row of semicircular niches, precisely like those of the basilica, except that there are no pilaster strips (Plate 5, Fig. 5). It is singular that below this is placed a row of arched corbel-tables (Plate 5, Fig. 6). These arched corbel-tables are supported on terra-cotta brackets (Plate 5, Fig. 6) of a type characteristic of the XII century. Moreover, arched corbel-tables of this form were not used in Lombardy before c. 1035. The explanation is that in the XII century the vault was restored, and that in consequence the exterior cornice was also repaired, the old semicircular niches being retained, but built over, and an arched corbel-table being added below them. V. The baptistery is usually believed to be contemporary with the church, that is, of the third quarter of the IX century.2 However, as has been seen, notwithstanding strong analogies of style which prove that the two buildings can not be separated by any great interval of time, the masonry clearly indicates that the baptistery is somewhat later than the church. The baptistery may therefore be assigned to c. 900. AGLIATE DI CARATE BRIANZA,1 S. PIETRO (Plate 5, Fig. 2, 3, 4, 7; Plate 8; Plate 9) I. The ancient basilica of S. Pietro at Agliate is deservedly one of the best known monuments of Brianza, and, indeed, of Lombardy. Although situated in a small and remote village, the basilica early attracted the attention of archaeologists. De Dartein, who was the first to study the edifice seriously, completely misunderstood its chronology. In 1874 appeared the study of Mongeri with valuable historical notices. To Cattaneo2 belongs the credit of having pointed out the significance of the basilica as a dated example of the Lombard style of the IX century. The notices of Mongeri, Landriani and various other writers who have referred to the monument incidentally, have added little that was new. In 1895 appeared the monograph of Corbella, a work of real value, notwithstanding certain errors of chronology. In recent years the church has been studied by Rivoira.3 Most of the authors who have written upon the monument have illustrated it with drawings for the most part not of unimpeachable accuracy. II. The little that is known of the history of the church has been given 2 De Dartein, I think, is the only archaeologist who has assigned the baptistery and basilica of Agliate to a period subsequent to the year 1000. i (Milano). 2 218. 3 196. 31 LOMBARD ARCHITECTURE careful attention by Corbella, and may hence be summarized briefly here. To judge from certain early Christian epitaphs,4 a church existed at Agliate at least as early as the middle of the VI century. At all events, there is excellent reason to believe that a chapter of canons was established in the church in the IX century by the famous Ansperto, archbishop of Milan (868-881).° The style of the architecture fully justifies the conclusion that the existing church was built at the time the chapter was founded. During the Middle Ages the pieve of Agliate acquired considerable importance. In the XIII century, according to a manuscript of Goffredo da Bussero cited by Corbella,6 S. Pietro of Agliate enjoyed jurisdiction over fifty-seven churches and seventy-one altars. Its importance must have begun to decline soon after, for in a sort of tax-list of 1398, published by Magistretti, mention is made of this church as being officiated by eleven canons, but as having only twenty-two dependent chapels. In 1568 S. Carlo Borromeo contemplated removing the prepositura of Agliate to Carate.7 The church was at this time in bad physical and financial condition, as is known from a passage in the 'Acts' of Federigo Borromeo.8 A certain Marchese Guido Cusani and a certain Don Pietro Tonsi offered to restore the church,9 but that such a restoration was ever actually executed, or that it affected the architectural forms of the monument, there is no evidence. At any event, Federigo Borromeo in 161910 deliberated anew upon remaking the church in better form, and upon suppressing the canons. Fortunately this unhappy idea was never put into execution, and the mutilations perpetrated upon the monument at this epoch appear to have been confined to the destruction of the ancient stone ambo, which was supplanted by a new one in wood.11 Until the second quarter of the XVIII century the church seems to have escaped restoration, but at length the inevitable happened. The edifice was made over in the barocco style by the prevosto Curioni, and the consecration was solemnly celebrated in 1731.12 A new pulpit and a new ambo were made, the roofs of the nave and side aisles were rebuilt, the sacristy between the church and the baptistery was erected, a new balustrade and stairway in the barocco style were provided for the choir, the pavement was raised half a metre, the old windows were walled up and new ones opened, and the walls covered with intonaco. Fortunately the evident antiquity of the monument seems to have secured for it a certain amount of respect even in the XVIII < Corbella, 30-31. s This notice is preserved by Giulini, I, 330-331 : Uno scrittore delle vite de' nostri arcivescovi, la di cui opera da me si conserva manoscritta, narra che la canonica di san Pietro nel luogo d'Aliate capo di una delle nostre pievi, e stata fondata da Ansperto medesimo. Io non so a qual fondamento egli abbia appoggiato la sua asserzione; non- dimeno, poiche quell'autore e antico gia di tre secoli, non £ da sprezzarsi tal notizia in una cosa, la quale per se non patisce alcuna difficolta. 6 46. t Corbella, 58. « Ibid., 69. 1 1bid, io Ibid., 60. ii Ibid., 64. ^-Ibld., 63-65. 32 AGLIATE DI CARATE BRIANZA, S. PIETRO century, and it was spared the complete devastation which fell to the lot of so many Lombard monuments of the Middle Ages. In 1838 the church lost the dignity of pieve13 to become a simple parish. In 1874 the agitation for the restoration of the monument began. An ambitious project was drawn up which provided for (1) the demolition of the campanile and the construction of a new one; (2) the remaking of the roofs of the church throughout; (3) the cleansing of the walls from intonaco; (4) the reopening of the ancient windows looking from the crypt into the church and the closing of the windows added in the XVIII century; (5) the removing of the barocco balustrades and church- furniture from the choir; (6) the lowering of the pavement to the original level; (7) the restoration of the two absidioles and their altars.14 In 1875, however, but few of these projected changes were carried out, and the activity of the restorers was limited to executing merely the most urgent repairs to the exterior of the apses. In that same year, 1875, the church was declared a national monument. In 1893 the restoration was seriously taken in hand under the direction of the architects Luca Beltrami, Gaetano Moretti and Luigi Perrone. The actual work executed followed closely the project of 1874. In 1893-1894 the portal on the south side of the basilica was reopened; the roofs were entirely remade; the windows on the north side of the nave were reopened and restored, and an attempt was made to replace their ancient painted decorations. The roof of the northern absidiole was lowered; the southern portal of the facade was restored and a symmetrical one opened in the north side aisle. The entire lower wall of the facade was restored, and a new central portal erected. The most mischievous act of the restoration was the placing of two columns in the middle of the eastern intercolumniation on either side of the nave, in the mis taken belief that two such columns had been removed in the XVIII century. In 1894 the frescos of the triumphal arch and the choir vault and those on the north wall of the nave came to light, beneath the intonaco.15 In 1896 the old campanile had already been destroyed, and plans for a new one were under discussion. This has since been completed and is a peculiarly offensive example of what modern restoration can do at its worst. An inscription within the church records the lamentable restoration in these pompous words: L'AUGUSTO DESIDERIO QUI MANIFESTATO DA UMBERTO I E MARGHERITA DI SAVOIA DI RITORNARE QUESTA VETUSTA BASILICA ALLE PRISTINE FORME DALLE SECOLARI VICENDE MANOMESSE FU ESAUDITO COL CONCORSO DEL GOVERNO, DEL COMUNE, DEI TERRIERI E COLLA DIREZIONE DELL'UFFICIO REGIONALE PEI MONUMENTI DELLA LOMBARDIA DALL'ANNO MDCCCXC AL MDCCCXCV. a Ibid., 68. ulbid., 70. " Arch. Stor. Lorn., 1894, 241. 33 LOMBARD ARCHITECTURE In 1895 the restoration proceeded with the total reconstruction of the north wall and the north absidiole — a piece of sheer vandalism which nearly destroyed the archaeological importance of the monument. The bifora of the crypt opening into the church were remade, and the new choir-rail, stairway and ambo were added, as well as a new altar in the northern absidiole and the rail of the same chapel. III. The church consists of a nave seven bays long, two side aisles, a crypt (Plate 9), a choir and three apses (Plate 8). The apses are surmounted by half domes, the choir is covered by a barrel vault, the two compartments of the side aisles preceding the absidioles have groin vaults (Plate 5, Fig. 2), and the crypt is also groin-vaulted. The nave and side aisles have been sup plied with a modern roof, presumably an imitation of the ancient one, traces of which were found during the recent restoration.16 The archivolts are in a single order and unmoulded. The voussoirs are alternately stones and bricks, a mannerism which recalls Roman technique. The columns and capitals are formed of a heterogeneous collection of pilfered material, mostly Roman fragments, several with inscriptions. The groin vaults of the side aisles of the choir (Plate 5, Fig. 2) have wall ribs and are very slightly domed. Although daubed over with modern plaster, they were not, I believe, actually rebuilt in the restoration. The vaults of the crypt are similarly slightly domed and supplied with disappearing transverse arches. The masonry has been completely covered on the interior by a coat of modern plaster. On the exterior, however, it can still be studied. It consists of a mass of pebbles and bricks laid irregularly in a heavy mass of mortar (Plate 5, Fig. 3). There are no cut stones. In the original parts of the edifice little attempt is made to place the stones and bricks in horizontal courses. The north wall, where this is done, is modern. In the facade and in the eastern gable are placed two windows in the form of a Greek cross, perhaps the earliest example of this characteristically Lombard feature. Below the Greek cross window in the facade are still clearly visible the remains of a large semicircular arch. It has been supposed by Sant' Ambrogio that this opened originally into a western apse, similar to the one at S. Giorgio of Valpollicella. Unfortunately, the restoration has so denatured the facade that it is impossible now to say for what this arch served. However, from a drawing made before the restoration and published by Corbella,17 it seems probable that this was merely a great lunette like the two at the sides, introduced in the XVIII century to lighten the church. The windows of the apse and clearstory are all large, like those of S. Vincenzo of Milan (Plate 135, Fig. 2). They were doubtless originally filled, in the early Christian manner, with stone tracery. 16 Corbella, 65. 17 71. 34 AGRATE CONTURBIA, BATTISTERO IV. There is but little original ornament in the church. The most important and characteristic decorative feature is the cornice of the main apse (Plate 5, Fig. 7). A series of semicircular niches is constructed below the eaves of the roof in the wall, where this is thickened by the springing of the half dome internally. These niches are separated into groups of three or two by shallow pilaster strips, which rise from the ground to the roof. This cornice is without analogy. The niches, it is true, are found at S. Ambrogio (Plate 117, Fig. 5), S. Eustorgio (Plate 127, Fig. 4) and S. Vin cenzo in Prato at Milan (Plate 137, Fig. 4), but in these cases the pilaster strips support what looks very much like an arched corbel-table, encircling the niches. It 'is altogether probable that at Agliate similar corbel-tables would have been added had not the crude quality of the materials employed in the masonry made such a construction exceedingly difficult. The absidioles are without decoration of any kind. The capitals of the crypt are extremely crude in character (Plate 5, Fig. 4), but seem to have been undoubtedly sculptured for their present position. V. Until 1890 S. Pietro of Agliate was a well preserved, homogeneous and extremely important example of the style of the third quarter of the IX century, and as such has been recognized by all archaeologists with the single exception of De Dartein. Although nearly everything possible was done to destroy the archaeological value of the monument in the restoration, the church still retains enough of its original form to make it of no little significance for the history of Carlovingian architecture in Lombardy. AGRATE CONTURBIA,1 BATTISTERO (Plate 10, Fig. 3, 5) I. Agrate Conturbia, a little hamlet at the foot of the Alps, has attracted the attention of several students of Lombard antiquities because of the notable baptistery there preserved. To the indefatigable Mella belongs the credit of first having called it to the attention of archaeologists. De Dartein2 subse quently studied it, and the monument has since been mentioned incidentally by several other writers. In the Museo Civico at Novara there is a drawing which shows the monument as it was before restoration. Before the southern portal stood a Renaissance portico which has now disappeared, and the north side of the baptistery was entirely masked by houses built against it. II. The baptistery belonged to the neighbouring church of S. Vittore, a fact which justifies the conclusion that the latter once enjoyed the rank of i (Novara). 2 401. 35 LOMBARD ARCHITECTURE pieve. In a diploma of 976,3 the basilica Sancti Victoris constructa infra Castro agregade is mentioned as sub regimine et potestatem of the episcopal church of Novara. In the early XVII century the fortunes of the church of S. Vittore must have fallen to a low ebb, since the bishop Bescape mentions that he restored it to the rank of parish.4 III. The plan of the edifice is very irregular, doubtless owing to the fact that parts of an earlier building were retained. In general it forms an octagon, on each of whose sides in the lower story are built out, in the thickness of the wall, niches rectangular in plan, except that the two adjoining the entrance on the east side (Plate 10, Fig. 3) are approximately elliptical. Externally the wall of the lower story is almost circular. The structure is covered with an octagonal cloistered vault. In the centre of the interior is the depressed octagonal font to which three steps descend. All the windows were evidently intended to serve without glass. The masonry of the exterior walls belongs to two distinct epochs. In the lower story, on the north and west sides, there is preserved a piece of wall in rubble masonry (Plate 10, Fig. 5). Pebbles of irregular size are laid in courses more or less horizontal and embedded in a thick mass of mortar. It is evident that this wall once formed part of a circular edifice. It is broken at present by a pilaster strip of brick and stone, and is crowned by a cornice, but both of these features are additions of the XII century. The masonry of the remainder of the edifice is somewhat irregular ashlar with cut stones of widely varying sizes, rather carelessly fitted together, the interstices being often filled up with bricks (Plate 10, Fig. 5). Bricks are also in many cases employed to form the archivolts of the upper gallery and the arched corbel- tables. This masonry shows much variation in character and is often little better than rubble, but is always different from the more ancient rubble masonry, being of better quality and formed of stones angular rather than round. Scaffolding holes are numerous, large and prominent. IV. The decoration is simple and characteristic. The cornices of the two eastern niches (Plate 10, Fig. 3) are formed of arched corbel-tables supported by slender pilaster strips. The cupola, on the other hand, has a cornice consisting of simple arched corbel-tables without pilaster strips (Plate 10, Fig. 3). Below this there is inserted on each face of the octagon a group of three blind arches supported on columns (Plate 10, Fig. 3). The capitals are formed of blocks running back into the wall and carved into a bracket-like form. Such blind galleries are characteristic of the Lombard style of the XII century and are found in numerous churches of Milan, Pavia, 3 Hist. Pat. Mon., I, 246. ^Agratum ... a nobis parochia erecta est: licet iam ibi rotundum templum esset pro baptisterio plebano nescio quo modo paratum. (Bescape, 108). 36 ALBUGNANO, S. PIETRO AL CIMITERO etc. The doorway (Plate 10, Fig. 3) is a severely plain rectangle, surmounted by a simple lunette. V. The edifice possesses so little architectural character, in both its earlier and later portions, that we have no guide other than the masonry itself for assigning it a date. The rubble masonry of the earlier part of the edifice (Plate 10, Fig. 5) is evidently later than that of the basilica (Plate 5, Fig. 3) or baptistery (Plate 5, Fig. 5, 6) of Agliate, since the courses are more horizontal, the stones better laid, the mortar-beds thinner. On the other hand, it is distinctly less advanced than that of S. Vincenzo of Galliano (Plate 99, Fig. 1), where a certain number of rectangular blocks are intro duced among the pebbles, where the horizontal courses are even better main tained, where the stones are more skilfully fitted together, and where the mortar-beds are thinner. It is evident, therefore, that the older part of the baptistery of Agrate Conturbia is later than c. 900, the probable date of the baptistery of Agliate, and earlier than 1007, when S. Vincenzo of Galliano was consecrated. Since, however, the character of the masonry approaches that of Agliate more closely than that of S. Vincenzo, we may assign the earlier portions of Agrate Conturbia to c. 930. The later portions are undoubtedly of the XII century, as is shown by the character of the ornament. All the motives here used, however, persisted for a considerable period, and hence offer no criterion for determining the date with precision. The masonry in its mixture of ashlar and rubble, offers striking analogy with that of the not distant narthex of S. Donato at Abbazia di Sesto Calende, an edifice there is reason to believe was erected c. 1130. Since, therefore, the masonry at Agrate Conturbia is very similar, but slightly more primitive, we may assign the XII century portions of the latter edifice to c. 1125. ALBUGNANO,1 S. PIETRO AL CIMITERO (Plate 10, Fig. 2) I. The position of the little town of Albugnano, perched on one of the loftiest crests of the Monferrato, and to be reached only by an extremely difficult carriage road leading from Castelnuovo d'Asti, is so inaccessible that, notwithstanding its proximity to the well known and much visited abbey of Vezzolano, the charming little church of S. Pietro, when I found it in 1910, was still entirely unknown. In 1911, however, a photograph of it was published by Bevilacqua-Lazise.2 II. I know of no documents which throw light on the history of the church. i (Alessandria). 243. 37 LOMBARD ARCHITECTURE III. The edifice consists of a nave of a single aisle, and an apse. The nave is roofed in wood, and the interior is without interest. The masonry is formed partly of ashlar of rather indifferent quality, and partly of bricks incised and of regular size, laid in horizontal beds. The mortar- j oints average about fifteen centimetres in width. A curious feature is a cross-hatching on certain of the stones. IV. The decoration is in flat and double arched corbel-tables, with shafts and pilaster strips. The apse cornice is carved with an interlace, and the cornice on the south wall is in part decorated with a rinceau. The apse windows (Plate 10, Fig. 2) are in three unmoulded orders. The great interest of the building, however, lies in the fact that one of these windows still retains its original perforated stone tracery (Plate 10, Fig. 2). This is one of the few extant instances in Lombardy of a form of decoration which was once undoubtedly common. Analogous tracery is found in Apulia, at S. Gregorio of Bari (Plate 10, Fig. 1), in the cathedrals of B arietta and Ruvo (Plate 10, Fig. 4) and also at S. Maria in Cosmedin at Rome. V. The church of S. Pietro, both in its masonry and in its decoration, is strikingly similar to the neighbouring abbey of Vezzolano, by which it appears to have been influenced. The construction may, therefore, be assigned to c. 1185. ALMENNO S. BARTOLOMEO,1 S. TOMMASO (Plate 10, Fig. 6, 7; Plate 11, Fig. 1, 2, 3) I. The circular church of S. Tommaso in Limine, locally known as S. Tome, lies in the fields about a kilometre from the present town of Almenno S. Bartolomeo, and about the same distance from the church of S. Giorgio of Almenno S. Salvatore. It is one of the most discussed of all Lombard monu ments. In the third quarter of the XVIII century Lupi,2 the venerable historian of Bergamo, described the church at length and published accurate drawings on a large scale, including an elevation, plan and section. So excellent are these engravings that they are still of the greatest value as showing the form of the church in the XVIII century, before it had suffered from modern restorations. The ancient cupola and other features which have since disappeared are clearly shown. In 1828 the church was discussed at length by the Sacchi brothers.3 The historian Ronchetti,4 who wrote in the first half of the XIX century, has contributed several historical notices of value. In 1823 Seroux d'Agincourt5 published small-scale drawings of the i (Bergamo). 2 I, 209. 3 36. * V, 89. 5 IV, Plate 24, Fig. 16-18. 38 ALMENNO S. BARTOLOMEO, S. TOMMASO church. The large-scale plan, section and elevation of Knight6 appeared in 1843. These drawings show the cupola crowned by a pseudo-Lombard aedicule. Of great value also are the large-scale plan, section, elevation and plate of details published by Osten,7 from which it is evident that in the middle of the XIX century the inscriptions of 1704 and 1752 were still in place in the portal. In 1866 more drawings of the church were issued by Hiibsch.8 Locatelli published in 1879 a description which supplements all these drawings and gives precise information as to the condition of the church as it was before the recent restoration. A new plan and section were published in the Grande Illustrasione,9 and still others by De Dartein. The little monograph of Fornoni, which appeared in 1896, is a convenient resume of what had before been written relating to the monument. II. Nothing is known of the early history of S. Tommaso. Two docu ments of 1346 and 1347, cited by Ronchetti,10 refer to two episcopal visits made to the nuns of S. Tommaso of Almenno. It is therefore evident that the church belonged to a convent, but of this convent nothing further is known.11 In 1403 certain Ghibellines took refuge in the church, but were besieged and driven out by fire. Fornoni12 conjectures that in consequence of this the edifice was abandoned. At all events it appears to have been in bad condition in 1672. According to an inscription formerly in the portal, but now placed inside the church, the edifice was struck by lightning in 1704 and subsequently restored. According to a second inscription, now also removed from its position and placed inside the church, another restoration was carried out in 1753.13 The unfortunate edifice was again struck by lightning in 1885, and sustained damage which either necessitated or formed a pretext for the restoration begun in 1892, the hardest blow which the monument has had to suffer. At this time the church had long been closed for worship, and had been used as a sto're-room for agricultural implements. The project of restoration included the remaking of the pavement and the lantern, and the replacing of the tiles on the roof by slates. Actually the works executed under the sanction of the Minister of Public Instruction (for the edifice is a national monument) were considerably more extensive. In 1892- 1893 two bases and a capital were removed from the lower story and replaced by new ones, and the exterior walls were thoroughly restored. In 1893-1894 the lantern was rebuilt on the authority of two colonnettes and a portion of the old cornice discovered in the course of the restoration, and in the gallery in the interior three columns, one base and two capitals were made anew. The 6 I, Plate XVII. t Plates XLIII, XLIV, XLV, XLVI. s Plate LIV, Fig. 6-13. » V, 984. io V, 89. n Ibid., I, 58. 12 13. is These inscriptions have been published by Locatelli (III, 186). The last time I visited the church (1913), one was leaning against the north wall near the stairway, and the other, upside down and only partly legible, served as a doorstep. 39 LOMBARD ARCHITECTURE groin vaults were patched up where they were cracked, as was also the dome. The southern portal and two columns of the lower story were restored. In 1895 the western portal was renovated.14 During the restoration, I know not in what year, the intonaco was stripped off the walls of the interior.15 It has been supposed by some authorities that the church of S. Tome is the baptistery of a destroyed basilica, but such is manifestly not the case. In addition to the fact that it is known that the church belonged to a convent, Lupi16 has published a document which proves that it was only in 1175 that the bishop and canons of Bergamo ceded to the inhabitants of Limine the right of administering baptism. The Duomo Vecchio of Brescia proves that a circular church need not necessarily be a baptistery. III. The church consists of two parts: an octagonal nave, surrounded by a side aisle, which is surmounted by a gallery; and a rectangular choir terminating in an apse (Plate 11, Fig. 1). The apse has a half dome; the choir, side aisles and gallery are groin-vaulted, while the nave is surmounted by a dome. The groin vaults of the side aisles and gallery have much depressed wall ribs (Plate 10, Fig. 7; Plate 11, Fig. 2). While the arches in the main arcade are stilted, as are also the transverse arches, the groins are depressed in elevation and sometimes curved in plan; moreover, they do not intersect in the middle of the compartments. The line of the groins is worked to a sharp point towards the springing. These vaults are constructed of rubble and were erected with the aid of a solid centering in wood. This is proved by the fact that their soffit still bears the traces of the boards of the centering impressed on the plaster (Plate 10, Fig. 7). The groin vaults of the gallery, instead of being horizontal, lean against the nave, being in this respect analogous to those of S. Fedele at Como. This inclination, given them partly to buttress the dome, partly to avoid raising the roof so high as would other wise have been necessary, gives rise in . the actual construction to several distortions. The capitals of the responds are placed at a lower level than those of the columns separating the gallery from the nave, and consequently the transverse arches springing from a higher level on one side than on the other, produce a singular curve in elevation, the irregularity of which is increased by the fact that its highest point is thrown nearer the outer than the inner wall. The resulting distortion in the vault surface itself is somewhat minimized by the loading of the transverse arches. The wall ribs are very much depressed in elevation, so that their crowns rise to a height much lower than that of the transverse and arcade arches. The groins, though broken, tend to intersect at a point which is nearer the inner than the outer edge of the vault. The transverse arches thus seem to sink into the surface of the i* Arch. Stor. Lomb., anno 1894, 254. i-'Arte e Storia, I Ottobre, 1894. 16 II, 1281. 40 ALMENNO S. BARTOLOMEO, S. TOMMASO wall on the outer side, and would be totally submerged were not the surface hollowed out, as it were, to receive them. The groin vault of the choir is approximately square in plan. The groins are somewhat depressed in elevation, the other arches are semicircular; never theless, there is considerable doming, and traces of the solid board centering are evident. The walls of the rotunda are enormously thick (c. 1.15 metres), and the two stairways giving access to the galleries, and the six semicircular niches of the ground floor are merely voids in this mass of solid masonry. The technique of the construction recalls that of Roman buildings, a core of well laid rubble being faced on both sides with a coating of ashlar masonry (Plate 11, Fig 1). This masonry is formed of rather roughly dressed blocks showing great variation in size, but skilfully laid in courses in which the horizontal direction is generally well maintained. Scaffolding holes, penetrating some times the entire thickness of the wall, occur at frequent intervals. Above the gallery arcade is a row of segmental relieving arches, an unusual feature in Lombard construction. A small clearstory, consisting of windows in the form of a Greek cross alternating with oculi, is pierced in the dome (Plate 11, Fig. 1). In the north wall a graceful triforium window lightens the gallery (Plate 11, Fig. 1). IV. S. Tommaso presents decorative features of extraordinary interest. The capitals are evidently of two distinct epochs. Most of those of the ground story (Plate 11, Fig. 3), together with their columns, have evidently been pilfered from an earlier edifice. This is clear not only from their style, earlier than that of those of the gallery, but from the fact that inverted capitals are used as bases, and a base with grilles in one instance is used as a capital. It is clear that the old columns were a little too short for the position in which the builders wished to use them. They were accordingly pieced out by a variety of expedients. New abaci were inserted above the inverted base already mentioned, and above one of the capitals ; fragments of columns were placed under the inverted capitals which served as bases. Of the aisle responds, some of the capitals and bases are pilfered, others are original. The half columns form an integral part of the masonry of the walls (Plate 11, Fig. 3). The pilfered capitals are of several types. Four are cubic (one was carved at a subsequent epoch with an interlace), nine are of a curious type, somewhat resembling a cubic capital, in which the curve of the outer edge is concave instead of convex (Plate 11, Fig. 3), and with four simple flat leaves incised on each corner. One has eagles, the earliest and crudest example of this motive that I have seen ; one has sirens ; one is ornamented with a double row of crude uncarved acanthus leaves, which recall the capitals of Caen, and one has a kind of interlace. The bases are of Attic type with griffes ; the shafts have a marked entasis (Plate 11, Fig. 3), and one is octagonal. 41 LOMBARD ARCHITECTURE Besides these capitals there are several more or less fragmentary capitals and bases lying about the building, for the most part the originals of those replaced by the restorers. Two now placed on either side of the altar in the gallery I suppose to have been those of the lantern found during the restoration. They are distinctly crude, rather than primitive, in style. Another capital of broad-leaved Corinthianesque type in the gallery has been placed in the wall near the restored capital which replaces it, and a third capital with ram's head is also near the modern copy. A fourth capital, now in a niche on the ground floor, is a copy of one of the ancient ones in the gallery, and the original base of this column is near its original position. These fragments were all made for the existing edifice, i.e., are not pilfered. The capitals of the gallery (Plate 11, Fig. 2) also were undoubtedly made for their present position, and are superb examples of Lombard decorative art. With a single exception they are of the Corinthianesque type. In one the volutes are replaced by rams' heads, in another by the figures of the Evangelists.17 The exception is the one sculptured with what I take to be scenes from the history of Tobit (Plate 10, Fig. 7). On one corner we have the figure of Tobit, the blind old man, with a long beard and patriarchal garments. To the left Raphael, the archangel, with staff in his hand, leads back to his father Tobias, whom the sculptor has represented not as a boy but as a bearded man.18 On the other faces are depicted the husband of Sara, killed by the evil spirit19 Asmodaeus, in the form of two dragons, and finally a scene which probably represents the union of Tobias and Sara.20 Tobias here is shown as a boy, very small, beardless and with long hair, who seems to cling to his powerful bride. One of these gallery capitals of block Corinthian type is unfinished. The exterior of the edifice is characterized by small, arched corbel-tables surmounted by flat corbel-tables and supported on pilaster strips, or on shafts which terminate in capitals, cubic or of interlaced or broad-leaved type (Plate 11, Fig. 1). The principal portal, which has been restored, is in many orders, as is the window above it. The impost of the portal is adorned with crude and much broken figure sculptures. Of the subjects I can recognize only St. Bartholomew, characterized by his knife. By the same hand is the relief of a monk bearing a candle over the lunette of the southern portal. The rectangular choir and the apse, obviously later than the main body of the edifice, are adorned with a very rich cornice of brackets, saw teeth, and double arched corbel-tables (Plate 10, Fig. 6). The latter motive also recurs over the little doorway (now walled up) in the south wall (Plate 10, Fig. 6). In the church are two inferior frescos of the XV century. " This capital and the one with scenes from Tobit appear to be by the same hand as the ambo of the Madonna del Castello. is Tob., xi. 19 Tob., iii. 20 Tob., vii. 42 ALMENNO S. SALVATORE, MADONNA DEL CASTELLO V. It is evident that we have in S. Tommaso three distinct eras of construction: first, the pilfered capitals and bases coming from an earlier building, not improbably on the same site; second, the rotunda itself; and third, the choir, added subsequently. It is to the main structure of the building that it is easiest to assign a date. The vaults of the gallery are less finely executed than those of S. Fedele at Como (Plate 64, Fig. 2), an edifice of c. 1115. They are, however, of a more advanced type. The builders have attempted a more difficult experiment than those of S. Fedele dared to under take, for they have renounced the easy expedient of doubling the number of supports in the outer perimeter of the annular gallery which was to be vaulted. Notwithstanding the complex problem they thus set themselves, they carried the execution through with great skill and mastery of the technical obstacles to be overcome. Such vaults could not have been erected before the year 1125. On the other hand, they still contain distortions and crudities which do not allow us to place them later than the year 1150. The style of the capitals is advanced. The two with figure sculptures are by the same sculptor who executed the ambo at the Madonna del Castello (c. 1130). Moreover, the masonry is much better than that of the church of S. Giorgio, erected c. 1120, and the rich mouldings of the western doorway and window betray a late date. We may therefore conclude that the present rotunda was erected about the year 1140. The pilfered capitals must consequently be somewhat earlier than this. They are all contemporary with each other, and were doubtless taken from some one single building. The cubic capitals are of the type familiar at S. Abondio of Como and other edifices of the XI century. The grotesques also are crudely executed and appear to be not later than the year 1100. On the other hand, the capitals with leaves cannot be earlier than this epoch, so that the pilfered materials may be with confidence ascribed to c. 1100. As for the choir, the cornice is evidently of c. 1180, a date which agrees with the details of the other parts of the structure. ALMENNO S. SALVATORE,1 MADONNA DEL CASTELLO (Plate 11, Fig. 4, 6) I. Some distance beyond the church of S. Giorgio lies a curious monu ment known as the Madonna del Castello. It has been published by De Dartein2 and more recently has been made the subject of a little monograph by Fornoni. For the history of the monument the historians of Bergamo, Calvi3 and Ronchetti4 as well as Lupi,5 should be consulted. i (Bergamo). 2 61. si, 33; II, 364. * II, 169; IV, 34. 5 11,327,653. 43 LOMBARD ARCHITECTURE II. The will of the priest Giovanni, dated April 9, 975, mentions the priests, deacons, subdeacons and officials of the church of S. Maria and S. Salvatore near the castle at Limine.6 It is, therefore, evident that at the end of the X century the church enj oyed considerable importance and was officiated by a chapter. Another document of 1058, cited by Ronchetti7 and Lupi,8 shows that at this date the church belonged to the jurisdiction of the bishop of Bergamo, but in 1169 it was given to the cathedral chapter9 by the bishop Guala. Other documents of 107310 and 1228" mention the priest of this church as being appointed by the chapter of Bergamo. On the western portal of the church is the date 1578, which doubtless records the epoch in which the western, or ante-church, was added. But the construction probably occupied twelve years, for according to Calvi12 the consecration was not celebrated until June 4, 1590. Calvi also relates13 that in 1613 a miraculous image was translated into the church, and he intimates that the edifice was at this time again enlarged. This statement, however, is in contradiction with his own assertion of a consecration in 1590, and with the date of 1578 over the portal of the church. Moreover, his sources for the date of 1613 are contradictory, and one inscription, he says, places it in 1611. It therefore seems probable that there is an error in this second part of his account, and that the enlargement of the church between 1578 and 1590 was really made to accommodate the image which was translated then, and not in 1613. The campanile was restored in 1894, as is evident from the inscription which it bears: R 1894. III. The singular plan comprises a modern ante-church, a nave of two bays covered with a modern barrel vault (Plate 11, Fig. 4), two side aisles, also with modern barrel vaults, a groin-vaulted choir (Plate 11, Fig. 4) of a single double bay terminating in a square east end and flanked by two side aisles, and a crypt. Heavy rectangular piers (Plate 11, Fig. 4) separate the nave and side aisles, but these were made over in the time of the Renaissance. They are ornamented with a simple moulding forming the impost of the rectangular archivolt (Plate 11, Fig. 4). The choir is sharply deflected, and indeed the whole monument abounds in irregularities of every description. The groin vault of the choir (Plate 11, Fig. 1) is not domed, but is supplied with wall ribs, of which the lower parts were cut away when a sort of screen- wall carried on an architrave and Ionic columns set transversely was added in the XVI century (Plate 11, Fig. 4). These Ionic columns, which are so disposed that the central one on each side comes squarely on axis, have been 6 . . . presbyteris diaconis vel subdiaconis & officiates Ecclesie Sancte Dei Gene- tricis Marie & Domini Salvatoris que est edificata intus castro eodem Lemenne . . . (Lupi, II, 327). 7 II, 169. s II, 653. = Ronchetti, IV, 34. See text cited below, p. 46. io Ronchetti, II, 183. n Ibid., IV, 34. 12 II, 264. is I, 33. 44 ALMENNO S. SALVATORE, S. GIORGIO mistaken by archaeologists for part of the original construction. At the east end of the side aisles are flat niches in the thickness of the wall. The crypt is covered by groin vaults with transverse ribs, which disappear towards the springing. It is greatly to be regretted that the walls internally and externally have been covered with a thick smudging of intonaco so that it is absolutely impossible to determine the character of the masonry (Plate 11, Fig. 4). IV. The capitals of the crypt are of the style of the Roman decadence, probably of the IV century, and are pilfered. In the north side aisle there is an ambo (Plate 11, Fig. 6) supported on four stump columns with capitals of Corinthianesque design. The sculptures, representing the emblems of the four Evangelists, are heavy and ponderous, similar to, but evidently more advanced than, those of S. Giacomo of Bellagio (Plate 22, Fig. 1, 2). The wings are awkward, and lack grace of line. The eagle has an unduly prominent breast and a widely spread tail; the lion is quaintly drawn and represented as roaring. In the curls of the angel and other details the influence of Guglielmo da Modena is evident. These sculptures are like those of Bellagio in that the animals seem to be suspended in mid air, and the feet hang limply down, supporting no weight. At Almenno, however, the sculptor appears to have tried to represent flight, whereas at Bellagio the wings are folded. The Almenno sculptures are, moreover, superior in the details of the technique, the general animation of the figures, and in the composition. V. So few fragments of the ancient church remain — really only the vaults of the crypt and choir — that it is almost impossible to ascribe a date, especially since the masonry can not be examined. The choir vault, however, is of a type which suggests the XII century; and the conjecture is, therefore, permissible that the remains of the ancient edifice are contemporary with the ambo. The latter, from the style of its capitals and sculptures, more advanced than those of S. Giacomo at Bellagio (c. 1115) and showing the influence of Guglielmo da Modena, must have been executed about the year 1130. ALMENNO S. SALVATORE,1 S. GIORGIO (Plate 11, Fig. 5, 7, 8, 9) I. About a kilometre from the village of Almenno lies the notable church of S. Giorgio. The architecture has been described and illustrated by Osten2 and by De Dartein.3 II. Lupi has printed a document from which it appears that the bishop Guala, in 1169, gave to the chapter of S. Alessandro at Bergamo the Madonna i (Bergamo). 2 XLVII, XLVIII. 3393. 45 LOMBARD ARCHITECTURE del Castello and the church of S. Gregorio at Almenno. I believe that the venerable historian of Bergamo has here made a mistake, or else his manu script has been misprinted, and that it was a question, not of the church of S. Gregorio, but of S. Giorgio.4 Certain it is, in any event, that our church has been called S. Giorgio at least since the XIV century, since at that epoch it was adorned with frescos of the titular saint.5 On the west porch is the inscription ANNO 1774, which doubtless indi cates the epoch in which this feature was added to the church, and the upper part of the facade remade. III. The church consists of a nave (Plate 11, Fig. 8) three bays long, with wooden roof; two side aisles, also with wooden roof; a rib-vaulted choir (Plate 11, Fig. 9) of a single bay, flanked by two groin- vaulted side aisles now walled off, and a single apse (Plate 11, Fig. 7). Over the eastern end of the northern side aisle rises a Renaissance campanile (Plate 11, Fig. 7). The nave (Plate 11, Fig. 8) is separated from the side aisles by rect angular piers with no capitals or bases, and only a very simple moulding at the impost, to mark the springing of the arches. The archivolts are in a single unmoulded order, but of light and graceful construction. The rib vault of the choir (Plate 11, Fig. 9) has diagonals of rectangular profile about half a metre in width. They describe in elevation, not a semi circle, but a segment of a circle; the wall ribs describe in elevation an egg-shaped curve. However, the vault itself is somewhat domed, and the ribs are allowed to project from its surface much more at their crown than at their springing. We have here, evidently, a very interesting experiment in the 4 . . . Ea propter ego Guala Dei Gratia Sancte Pergamensis Ecclesie Episcopus dono cedo atque per hanc presentis privilegii paginam confirmo ut a presenti die & hora deveniat in jure & potestate Ecclesie Beati Alexandri que constructa est extra prope civitatem Pergami ubi ejus sanctum requiescit corpus videlicet eapellam unam cum parochia sua que constructa est infra villain de Lemine ad honorem Dei & Beate Virginis Marie. Et aliam Ecclesiam que constructa ex extra predictam villam in salecto [=salicto] ipsius loci ad honorem Dei & Sancti Gregorii. . . . (Lupi, II, 1254). The chronological notes of this document are erroneous. January 3 fell on a Wednesday in the year 1168, which would seem to be indicated by the year of the Incarnation 1169. There is a further confusion in the indiction which, in 1168, would be the first, not the second. The year 1170, on which the third of January fell on a Friday, corresponds neither with the year nor with the indiction. 5 Above the fine figure of S. GOTARDV on the second pier from the west of the southern arcade is an inscription giving the date of the fresco: . DIS' LVIN' DE CAROLIS DE LEMEN . VI FECIT . FIERI . HOC . OPVS .M.C.C.C.L.X.X.X.I.I. On the west respond of the same arcade is a similar inscription of the same date, now in part illegible. 46 ALMENNO S. SALVATORE, S. GIORGIO construction of rib vaults and one full of significance for the history of architecture. The builders, for aesthetic reasons, wished to avoid excessive doming of the vault. Not being able, however, to obviate this entirely in a vault constructed without solid centering, they minimized its effect by depressing the intrados of the diagonals into a curve more segmental than that of the vault itself. Since the vault is erected over an area nearly square in plan (6.39 x 5.90 metres), the wall ribs project but slightly, and like the diagonals are loaded at the crowns. The vault is probably constructed of ashlar, but is covered with intonaco, so that the stereotomy can not be studied. The stone courses of the diagonals, however, can still be seen, and it is evident from the arrange ment of the stones at the intersections, that one diagonal was erected first as a complete and self-sustaining arch, and that against this were later placed the two half arches of the second diagonal. Shafts are provided for the diagonals in the eastern piers, but on the west side they are carried on corbels. The side aisles of the choir have been walled off to form separate rooms, but the ancient groin vaults are still intact. These vaults were very oblong in plan, and are so highly domed that they look more like longitudinal barrel vaults than groin vaults. They were undoubtedly erected with solid centering, since the imprint of the boards of the centering is still clearly visible in the plaster. These vaults are constructed of rubble. The piers separating nave and choir have systems consisting each of a broad pilaster strip (Plate 1 1, Fig. 8). The pavement slopes sharply towards the west. The masonry of the church is most peculiar. The lower part of the side- aisle walls and the nave piers and arches (Plate 11, Fig. 8) are constructed of ashlar. In the piers this ashlar is smooth and well laid, but in the side-aisle walls, even internally, it is somewhat rough, and externally it is decidedly rough, the blocks being unsmoothed and treated almost like rusticated work. It is probable that it was the intention of the builders to polish off these blocks, but that for some reason the walls were left unfinished. The choir (Plate 11, Fig. 7) and the east bay of the nave (Plate 11, Fig. 5) on the south side are constructed of smooth blocks, but this ashlar is somewhat crude and has irregular courses and wide joints. The clearstory walls of the nave (Plate 11, Fig. 5) and the upper part of the side-aisle walls (Plate 11, Fig. 5) are of rubble in which herring-bone pebbles are much used. This rubble masonry can not be altogether the result of later rebuilding, since it is quite mediaeval in character, and in the nave XIII century frescos are placed upon the rubble wall (Plate 11, Fig. 8). IV. The supports of the choir are Lombard compound piers (Plate 11, Fig. 8), of which the bases have, for the most part, been covered up, leaving only a few mouldings visible, not enough really to determine their character. The capitals are of Corinthian type, with a single row of uncarved leaves, or 47 LOMBARD ARCHITECTURE of broad-leaved type, or with curious handle-like volutes, and show very skilful execution. They are quite analogous to those of S. Fedele at Como (Plate 63, Fig. 6, 7; Plate 64, Fig. 5). The apse (Plate 11, Fig. 7) is adorned with a row of blind arches supported on pilaster strips, with engaged shafts. Above is a cornice ornamented with a saw tooth, and there may well have been originally a row of arched corbel-tables above this, for the apse roof has been lowered in modern times. The few windows are widely splayed, and evidently never had glass. The nave imposts are without capitals (Plate 11, Fig. 8); in the west wall, however, are two responds which have capitals, one of which is cubic, the other of the curious wreath variety familiar at Fontanella al Monte. Without doubt the interior walls were anciently covered entirely with frescos, of which many notable fragments are still extant. These fragments comprise at least three layers, of which the latest is dated 1388 by two inscriptions. It is to be regretted that no critical study of these important monuments of pictorial art has yet been made. There are clear traces of frescos also on the exterior walls. V. Although there appears to be a distinct break in the masonry between the nave and the choir, the style of the two portions of the edifice is so similar that the former can not be more than a year or so later than the latter. The severe, almost bare character of the nave is extraordinary, and suggests the XI century. The character of the two capitals in the west wall, however, the lightness of the arches, and the skilful stereotomy, show that it could not have been erected before the XII century. Moreover, the rib vault of the choir is obviously not an early experiment, since it shows refinements which the builders could have introduced only at a period when they had thorough mastery of the technique of construction. This vault is less highly domed and more skilfully executed than the vaults of S. Savino at Piacenza (Plate 186, Fig. 1), a church consecrated in 1107. The capitals present analogies with those of S. Fedele at Como (c. 1115). Finally, the masonry in its mixture of rubble and ashlar recalls the narthex of Abbazia di Sesto Calende (c. 1135). We may, therefore, conclude that the monument is essentially a homogeneous structure of c. 1120. AOSTA,1 CATHEDRAL (Plate 12, Fig. 1, 2, 4, 5) I. The literature of the cathedral of Aosta is more conspicuous for quantity than quality. The church as a monument of architecture was first studied by De Lasteyrie in 1854 in a work hardly worthy of the famous archaeologist's reputation. In 1857 Aubert published an article upon the i (Torino). 48 AOSTA, CATHEDRAL mosaics, and in 1860 a large illustrated work on the valley of Aosta, which included a sumptuous illustration of the cathedral pavement. In 1873 the mosaics were further studied by Aus'm Weerth,2 and in 1880 Berard returned to the thesis of Aubert that the pavement was a work of the XV century, bringing forward new documents of some importance. This thesis was in 1891 again reaffirmed by Due, who incidentally published several new texts. The article of Leclere on the Challant family, which appeared in 1907, contains some important notices bearing upon the cathedral. In 1911 Toesca published, under the auspices of the Italian government, a catalogue of the works of art of Aosta. The publication is pretentious and contains numerous half-tones of great value. The text, however, is perfunctory and disappointing. In the same year Monneret de Villard published a little handbook on Aosta in the Bonomi series, containing excellent half-tones of the mosaics. Finally should be mentioned the study of the architecture by Commendatore Rivoira.3 II. Berard has published a notice taken from an ancient martyrology of the cathedral, which speaks of Gontran (t 593), king of the Franks, as the restorer of the cathedral.1 Berard says that Gontran was a king of Burgundy in the middle of the VI century, but with evident error, since Godomar, the last king of the first kingdom of Burgundy, fell in 534, and no Burgundian king of the name of Gontran is known to history.5 The restorer of the cathedral of Aosta must be Gontran, the pious grandson of Clovis, renowned for his orthodoxy and mild character. The documents of the Middle Ages are silent in regard to the history of the monument, and it is only in the XV century that we begin to have notices of the changes and alterations made in the cathedral. The keystones of the vaults of the nave are sculptured with the coat of arms of the family of Challant. De Lasteyrie6 has conjectured that these escutcheons belong to Francois I, who died in 1421 and was buried in the cathedral. Leclere,7 on the other hand, assigns the vaults of the nave to Georges de Challant, who became a canon of the cathedral chapter in 1460. As a matter of fact, the shields prove only that the vaults must have been erected in the XV century by some member of the Challant family. There is extant a notice of the consecration of the church on August 24,8 but the year, unfortunately, is not 2 17. 3 230. i QVINTO KL' APRILIS EODEM DIE APVD CABILONE CIVITATE GALLIARV DEPOSITIO BTI GONTRANNI REGIS FRANCORUM INSTRAVRATORIS HVI ECCLIE (Berard, 146). 5 Lavisse, Histoire de France, Paris, Hachette, 1903, II, pt. 1, pp. 121 f ., 138, 146. Lavisse et Rambaud, Histoire G6n6rale, Paris, Colin, 1905, I, 100. 6 7. 7 137. 8 Extractus Anniversariorum ecclesiae Augustae, ed. Hist. Pat. Mon., V, 648: Augustus 24. Bartholomei apostoli et dedicatio ecclesie cathedralis Auguste. 49 LOMBARD ARCHITECTURE given. It is possible this consecration may have been that solemnized after the vaults had been erected, or it may refer to one of the consecrations which must have taken place in the XI and XII centuries. Francois de Challant, whom De Lasteyrie suspects of having vaulted the nave, was buried in the cathedral. The tomb, which was placed in the choir, was of great magnificence and its construction was the occasion for erecting in the choir a solanum.9 The misinterpretation of this word solanum as referring to the mosaic pavement of the choir has misled several archaeologists into assigning the latter to the XV century, whereas the style clearly shows the pavement must be of the XII century. Just what solanum means is not clear ; it may refer to the vaults, to the ambulatory or to a raised platform for the tomb. The word, so far as has been shown, does not occur elsewhere in mediaeval Latin. In 1460 the cloisters were begun by Georges de Challant,10 whose name appears with those of the other donors sculptured in relief on the capitals of the arcade. On June 5, 1518, the campanile was struck by lightning. It must have been old at this time, since it had been previously restored by one of the canons who was also master of the fabbrica.11 The atrium was adorned with frescos by Giovanni Goubaudelli, who died in 1525 apparently, although he is recorded as having made a bequest to the church dated 1535.12 Besides this we know that an important reliquary was donated by Francois de Challant.13 The necrologies of the cathedral14 abound in records relating to the various - Pro Magistro imaginum. In nomine Domini, Amen. Per hoc praesens publicum instrumentum ad universorum notionem deducatur quod cum illustris et magnificus Dominus Franciscus Comes Challandi, nuper ordinaverit et proposuerit fieri facere in ecclesia Augustense, auxilio divino mediante, quoddam magnificum opus valde sump- tuosum videlicet eius sepulcrum et solanum chori dictae ecclesiae, et ipsius magnifici operis onus fiendi in se suscepit discretus vir magister Stephanus Mossettaz, burgensis Augustae, hinc est quod constituti in capitulo praedictae ecclesiae . . . venerabilis viri, etc. . . . die vigesima prima mensis ianuarii, anno Domini millesimo quatercentesimo vigesimo nono etc. (Due, 6-7). Anno Domini millesimo quatercentesimo trigesimo quarto, die decima octava mensis iulii . . . magister Stephanus Mossettaz confessus fuit habuisse a Domino Comite Chal landi plenam solutionem de septem centum et quinquaginta florenis parvi ponderis in deductionem operis solani ecclesiae et suae sepulturae . . . (Due, 9). Item fecit fieri suis expensis solanum supra chorum. Jacet in confessione. . . . (Due, 10). io The inscription has been published by Aubert, 216. n Venerabilis domini Hugoni Ferrenchi de Curia maiori canonici Auguste, magisti fabrice huius ecclesie, qui reparauit aulam nostram capitularem et campanille, ante consumationem fulguris que euenit. 1518. die 5a iunii. (Hist. Pat. Mon., V, 645). 12 Venerabilis dominus Iohannes Goubaudelli de Liniaco in ducatu Barri, Tulensis diocesis, canonicus Auguste, et magister fabrices benemeritus . . . ornauit atrium ecclesie nostre picturis variis, sumptibus propriis. (Ibid., 631-632). is /bid., 640. i* Ibid. 50 AOSTA, CATHEDRAL chapels, at least thirty-one in number. The notices are strikingly confused and contradictory, but it appears that the earliest of these chapels of which the date can be established were those of S. Tommaso and S. Maria Magdalena, founded by the bishop Nicolao (1281-1300).15 The western portal was rebuilt 1522-1526, and in 1848 the facade was adorned with mediocre statues. III. The church has been almost entirely modernized and retains of its original architecture, at least that is visible, only the crypt (Plate 12, Fig. 2), the campaniles (Plate 12, Fig. 4), three capitals of the ambulatory, and the mosaic pavement. These fragments, however, are sufficient to show that the monument in the Romanesque period possessed remarkable architectural forms. The ambulatory (Plate 12, Fig. 5) and the twin towers flanking the choir (Plate 12, Fig. 4) savour of dispositions common enough in ultramontane churches but rare in Italy. The ambulatory at present (Plate 12, Fig. 5) presents superficially all the characteristics of the Gothic style of the XV century. It is evident that it is not part of the original construction, for the crypt stops with the wall of the choir, and does not extend beneath the ambulatory. The crypt was enlarged in the XII century, and it is probable that at this time the ambulatory was added. The irregularity of the existing soffits gives some reason to suppose that in the XII century the ambulatory was covered with compartments of wedge-shaped transverse ribs almost like triangular barrel vaults and with trapezoidal groin vaults (Plate 12, Fig. 5). I can find no evidence that the radiating chapels (Plate 12, Fig. 5) are ancient, or that the Romanesque edifice was supplied with any such absidioles. In fact, it is impossible to prove absolutely that the vaults, and, indeed, the whole ambulatory, be not of the XV century, since modern plaster completely covers the masonry. It seems, therefore, that Commendatore Rivoira ventured a somewhat hazardous conjecture when he brought this church forward as an important example of a Romanesque ambulatory. The crypt (Plate 12, Fig. 2) is covered with groin vaults with transverse ribs which disappear towards their springing. The western bay of this crypt is obviously later than the eastern bays, since not only the capitals differ, but the structure of the vault itself is different. In the eastern part the vaults (Plate 12, Fig. 2) are undomed and have transverse arches which completely disappear at the springing, but are loaded in a very exaggerated way at the crowns. One which I measured projects 60 centimetres (Plate 12, Fig. 2). In the side walls the wall ribs were in two orders, and the responds appear to have consisted of five rectangular members without capitals. The traces 15 Selecta e libro anniversariorum ecclesiae cathedralis Augustanae, ed. Hist. Pat. Mon., V, 548. Compare, however, ibid., 627. A great wealth of perplexing details in regard to the chapels of the cathedral is supplied in the records published in the Hist. Pat. Mon., V. 51 LOMBARD ARCHITECTURE of the solid board centering are still preserved in one vault. The masonry is rubble of the roughest kind, only the responds and the ribs being executed in a crude sort of ashlar of stone or brick. The western part of the crypt, on the other hand, has groin vaults similarly undomed and with transverse ribs, but these transverse ribs do not die away altogether at the springing and are but slightly loaded at the crown. Moreover, the vaults themselves are far better executed than those of the eastern part of the crypt. The campaniles are of widely different design (Plate 12, Fig. 4). The southernmost is unbroken by windows, save bifora in the face of the two upper stories. Below are arched corbel-tables grouped two and two by pilaster strips. The northern campanile, on the other hand, has two sets of bifora in its three or four upper stories, and is adorned with arched corbel-tables of the customary type. Both campaniles are crowned by pinnacles and turrets evidently later than the towers themselves. The southern campanile is supplied with an eastern apse, and it is evident that the northern campanile as well must originally have had a similar excres cence. The undomed groin vaults of the lower stories of the campaniles are the original ones of the XI century. They had continuous responds consisting of three rectangular members (three have been made over), and were characterized by loaded wall ribs and much depressed diagonals. It is clear that the church must originally have had five apses, three aisles and no transepts.16 The piers of the nave, rectangular in section, may be the original ones, and original also appears to be the south side-aisle wall, with pilaster strips so disposed as to suggest that the arched corbel-tables were grouped two and two. The north side-aisle wall and clearstory, on the other hand, appear modern. The ambulatory is supplied with flamboyant buttresses, doubtless added when the vaults were remade in the XV century. North of the church are the lovely flamboyant cloisters, unfortunately in part destroyed to make room for a modern chapel. IV. The ornamental details of the church, like those of most of the monuments of the Val d' Aosta, are singularly lacking in character. The cubic capitals of the ambulatory appear to be of the usual Lombard type. Since, however, similar capitals were added to the choir in the XV century, it is, in the present state of the edifice, impossible to be certain that the whole ambulatory does not date from this time. The capitals of the campanile are similarly completely without architectural character. The capitals of the eastern portion of the crypt are formed of ancient bases, uncarved blocks of stone, etc. (Plate 12, Fig. 2). Among the pilfered materials are notable two fine Roman Ionic capitals (Plate 12, Fig. 2). Only one capital is origmal, and this has chamfered corners adjusting the rectangular load to an octagonal io Toesca (7) is wrong in saying that the Romanesque church had transepts. 52 AOSTA, CATHEDRAL shaft. It is exactly analogous to the capitals of the piers of S. Pietro at Acqui. The capitals of the laterj western portion of the crypt, on the other hand, are of Corinthianesque type, or else are adorned with string patterns and are skilfully executed in fine marble with sure technique and deep undercutting. The absidiole of the southern campanile is adorned with blind niches. The mosaic of the choir is the decorative feature of the church which commands the greatest interest. The subject recalls that of the mosaic pave ment of S. Savino at Piacenza. In the centre is the Year, ANNVS, haloed — significant detail — bearing in his hands the Sun, SOL, and the Moon, LVNA. About him is a circle formed by medallions with the works of the twelve months. January, IAN with two heads, stands between two doors, the one open, the other closed. February, FEBRV|ARIVS, warms himself at the fire; March, MARCIVS, prunes a tree; April, APRILIS, is represented holding a flower in either hand and with a bird's nest; May, MAIVS, is on horse-back ; June, I VNI VS, mows ; July, I VLIVS, binds the grain in sheaves ; August, AVGVSTVS, threshes it; September, SEP|TEMB|ER, treads the grapes; October, OCTOjBER, sows; November, NOV, carries a load of wood on his back; December, DECEMBER, slaughters the swine. At the angles of the quadrangle are four figures pouring water from jars, representing the four rivers of Paradise. The inscriptions naming two of them, FIZON, GION, are still extant, but those of the other two, Euphrates and Tigris, if they ever existed, have disappeared. I suspect, however, that they were not repeated, since already represented in the upper mosaic. Here we see first EVFRAjTES represented as a nude woman pouring water from a jar. The head of the bull of St. Luke placed near this figure tells us that in the thought of the artist these four rivers signified the four Evangelists. Then follows the TIGRI|S, near which may be seen part of the lion of St. Mark. This mosaic contains other grotesques and symbolic figures; among the former an elephant, ELEFANS, and a CHIMERA with two heads (both, however, possibly have reference to the bestiaries) ; among the latter three of the symbols of the Evangelists, but a unicorn is substituted for the fourth (St. Matthew). In the cloisters is a carved plaque (Plate 12, Fig. 1) which probably belonged to some of the church-furniture. The central motive is a whirl which has a Carlovingian look, although it must be remembered that similar whirls persisted at Pieve Trebbio until the XII century. Around the outer border of the whirl is a row of crockets, and in the outer corners were four animals. The two uppermost, of which one has been almost entirely destroyed, were lambs of God, holding each a cross. The two lower ones represented stags taking the poison from serpents, according to the bestiary story. These animals are drawn in excellent proportion, but the details, such as the eyes and mouth, are very crude. 53 LOMBARD ARCHITECTURE V. The core of the campaniles, with their absidioles, the bays of the side aisle adjoining them, the choir and the eastern part of the crypt, the piers of the nave and parts of the southern wall are undoubtedly fragments of a basilica of the early part of the XI century. The only original capital of the crypt is perfectly analogous to the capitals of the piers of the nave of S. Pietro at Acqui, a dated monument of c. 1015-1023. The blind niches of the absidiole of the southern campanile recall those of the same church of S. Pietro at Acqui, and those of S. Vincenzo at Galliano, a surely dated monument of 1007. The rectangular piers also recall those at Galliano. The southern wall was probably adorned with arched corbel-tables grouped two and two, like those of S. Pietro at Acqui and other churches of the early part of the XI century. This portion of the cathedral of Aosta may, therefore, be assigned with great confidence to c. 1010. I do not hesitate to ascribe to the same epoch the carved plaque of the cloisters, which probably belonged to the church-furniture made for the XI century basilica. The motives, it is true, are still Carlovingian, but there is a largeness in the design, a freedom in the execution, and a certain verve in the technique which announce the transition towards the developed Lombard forms. This is apparent upon comparing it, for example, with the carved archivolt from the cathedral of Ferentino (Plate 12, Fig. 3). Moreover, the introduction of a bestiary story is without analogy in Carlovingian work, and is one of the earliest examples of a subject of definite iconographic purport in Lombard art. About a century later the crypt was extended to the westward, the mosaics of the choir executed, and possibly the ambulatory added. The mosaic is very analogous to that of S. Savino at Piacenza, which is known to date from 1107. We may, therefore, with confidence ascribe the Aosta mosaic to c. 1110, a date which accords well with the style of the capitals of the western bays of the crypt. The northern campanile entire and the upper part of the southern have been much made over and restored in the XIII and succeeding centuries. But the lower part of the southern tower with its arched corbel-tables grouped two and two appears to preserve essentially unaltered its XI century form. AOSTA,1 S. ORSO (Plate 12, Fig. 6; Plate 13, Fig. 1, 2, 3; Plate 14, Fig. 1, 2, 3; Plate 15, Fig. 1, 3) I. The collegiate church of S. Orso, although it has been mentioned incidentally by a number of writers upon Lombard antiquities, has never been given the careful study which this really very important monument deserves. As early as 1860, Aubert2 described the cloisters and published an engraving i (Torino). 2 225. 54 AOSTA, S. ORSO which shows the monument as it was before the existing ugly grill was erected. The history of Due, the publication of which was begun in 1901, contains some historical notices of value, especially in regard to the pastoral visits to the church. Ten years before, Ceradini had published a little book in which he undertook to give the subjects of the cloister sculptures. His work is not altogether successful, for many subjects are passed over in silence and others are misidentified. However, as a first attempt at this difficult task, the book merits praise, and is far superior to the iconographic study of Venturi,3 which is well illustrated but abounds in mistakes and inaccuracies. Equally unsatis factory, except for the illustrations, is the pretentious but inaccurate catalogue compiled by Toesca under the auspices of the Italian government, and published in 1911. The little handbook of Monneret de Villard, also published in 1911, contains excellent half-tones. Sumptuous photographs have been published by Martin. The architecture of the church has been touched upon by Commendatore Rivoira.4 II. According to the Pita Sancti Vrsi our church was in existence during the lifetime of the saint, or about the middle of the VI century.5 This life, it is true, is not of great weight as an historical source. Internal evidence shows that it was written by a monk of Aosta who lived long after the death of the saint. The writer shows himself perfectly familiar with the local geography, but his only sources for the events which he narrates seem to have been vague traditions of the place passed down by word of mouth. His scant matter is padded out with texts from the Scriptures and platitudes upon the virtues of the saint.6 Nevertheless the tradition that S. Orso was a priest in the church 3 III, 72. ^231-232. 5 Savio, 84. 5 I transcribe from the Bollandists those passages of the life of the saint which seem to bear upon the history of the church or the sculptures of the cloister: In nomine sanctae & indiuiduae Trinitatis, hie subter insertu est, qualiter beatissimus Vrsus Con fessor Christi, & Sacerdos Dei excelsi, natione Scotus, Patronus contitularis Ecclesiae, ciuitatis & burgi Augustas, nocte ac die Christo famulans, curam gerebat officij sui in Ecclesia, quae in honore S. Petri Principis Apostolorum aedificata, & dedicata est, extra ciuitatem quae nuncupatur Augusta. . . . Fuitque in illo tempore in eadem vrbe Prassul quidam, nomine Ploceanus: sed, vt res gesta apertissime probat, & illius loci ciues asser- runt, ob perfidiam suam niniiamque crudelitatem, non Potifex pastoralis curse, sed sub veste ouis lupus rapax, & crudelissimus inuasor, atque tyrannus potius extitit, quam Pontifex vel pastor. . . . Circa ecclesiam vero S. Petri, vbi vir Dei Vrsus fungebatur officio, tanta inundatio fuit, vt nullus earn ingredi, aut egredi valeret. . . . Restat autem adhuc & aliud miraculum, quod per ipsum Domini virtus operatur est, quod mihi visum est non praetermittere, quod tamen hominibus loci illius notissimum est. . . . Cumque aues superiiis nominatas aleret, & resideret ante ianuam ecclesiae, vbi sedul6 Deo seruiebat; aspiciens vidit iuuenemt quemdam equum domini sui equitantem, & amarissime flentem: qui cum assidue hie idem custos equorum ante iam dictum templum transiret, numquam tamen ad orationem descendit, nee caput suum Sanctorum reliquiis inclinauit. Tunc beatissimus Vrsus hasc intuens frequenter agentem, vocauit eum ad se, cui & dixit: Die, inquit, mi fili . . . & omnia secreta tua mihi fac manifesta. Tunc ejulans 55 LOMBARD ARCHITECTURE of S. Pietro, which was subsequently renamed in his honour, merits faith, and the life of the saint, dubious as it is as an historical source, still gives sufficient reason to believe that the church of S. Pietro existed as early as the VI century. According to Due,7 S. Orso was a collegiate church from the earliest times. It is evident, however, that the historian of Aosta has here made an ille, ait ad virum Dei: Heu mihi, Domine, hodie cum transirem per hanc viam, ablatus est a me equus domini mei optimus, quern nimio diligebat amore, unde nimio coarctor timore: quid faciam nescio. Unde non aliud peto, nisi tuam beatissimam orationem pariter et consolationem. Tunc idem vir Dei in Domino confisus, cum magna fiducia dixit adolescenti: Vade, fili mi, ingredere, in "hanc domum Domini, in qua antea non fuisti: flagita misericordissimum et amatorem omnium benignissimum Deum de salute tua: postea vero venies ad me, et quicquid necessarium fuerit pandam. Quibus ille jussis libentissime obediens, ad virum Dei concito gradu re versus est. Dixit ad eum S. Ursus: Die mihi juvenis, quanti tibi in custodiendo, curam gerehdo, commendati sunt equi? Respondit: Sex. Dicit ei S. Ursus: Quantos ad pabula dimisisti, quando nunc, quern quaerere coepisti? Aut quis tibi talem dedit equum tam pulchrum, tamque formosum et honestum? Tunc pavef actus puer ille, videns se captum in interrogationibus S. Ursi, prosternens se ad pedes ejus cum lacrymis dixit: Miser ego, ipsum, quem equito, et quaero lugens, ipsum sedeo. . . . Tunc demum castigavit eum S. Ursus. . . . Eodem tempore . . . Ploceanus, ut dictum est, in eadem urbe Pontifex esse videbatur. Contigit autem ejusdem Ploceani ministerialem quemdam incurisse crimen. Unde valde pertimescens, praesidium fugefecit ad ecclesiam S. Petri, ubi S. Ursus custos erat. Cumque more solito in ecclesiam fuisset ingressus, vidit eum juxta altare stantem, cui et dixit: Quidnam, inquit, peccasti, fili? Cumque puer singulas retulisset admis- sionis suae culpas, coepit eum rogare, ut apud dominum suum pro se intercederet. Egrediensque S. Ursus venit ad Episcopum in civitatem, et prosternens se ad pedes ejus dixit: Mi domine Pater, quidam ex pueris vestris sciens se graviter deliquisse in vos, confugium fecit ad S. Petrum: unde peto, ut in ejus amore, ad quem confugit, absolutionem criminis ei concedatis, et vestrae domesticitatis concessionem. Ploceanus vero fallens in virum Dei dixit: Vade, f rater, die ei, ut securus cum gratia veniat ad praesentiam nostram, nulla pertimescens mala. Tunc vir Dei ad puerum reversus, dixit: Vade fili mi, quia ex hac re impedimentum tibi nullum erit. Ploceanus enim Episcopus clam quibusdam sibi astantibus dixit: Nisi eum, cum exierit de ipsa ecclesia, in qua nunc latitat, mihi vinctum praesentaveritis, ejus sententiae vos subj acebitis. Qui jussa complentes, ante eum adductus est. Qui frendens et tabescens in eum, tanquam cru- delissima bestia, tamdiu a dorso et ventre flagellatus est, quousque expirare crederetur. Tunc capite ejus tonso, jussit tyrannus viscosum bitumen super ilium fundi, ut vix evaderet usque ad virum Dei morte vicina ... cui et dixit: Quid, inquit, facere in me voluisti, Pater mi, ut me . . . egredi de hac domo Dei juberes? . . . Tunc beatissimus Ursus . . . dixit . . . "Vade cito fili mi, die Ploceano: Notum tibi sit, quia post paucos dies tolletur anima tua a te, et a daemonibus, ut dignus es, strangulaberis, et a tetris spiritibus susceptus sepelieris in infernum; ut quibus nunc usque servire studuisti, ab his recipias meritum servitutis. . . . Tu ergo, fili mi, praepara iter tuum, quia et tu migrabis cito de hoc mundo pergens altercaturus cum eo. Ego autem subsequens ero iter vestrum, ut sim audientiae vestrae assistens, cum in ratione steteritis ante tribunal Judicis magni. Mira res, ac verissima, quia quicquid in ilia hora praedixit, in utroque postea rei probavit eventus. Eadem vero nocte, in qua migraturum Episcopum praedixit, projectus est de stratu suo a daemonibus in humum, et sic miserabiliter expiravit. (Vita S. Ursi, ed. Jean de Bolland, Acta Sanctorum, Februarius, I, 946). 7 1, 225. 56 AOSTA, S. ORSO error, for there is extant a charter of the bishop Anselmo of 923, establishing a chapter of canons in the church.8 It is therefore certain that before 923 no canons existed. The chronological notes of this charter offer some difficulty, for while the eleventh indiction corresponds to the year 923, Rodolfo did not become king until 924. However, the fact that the chapter was founded by Anselmo is not open to question,9 since it is confirmed by a passage in the necrology of the chapter10 and by a martyrology cited in the acts of the pastoral visit of 1419.11 In 1032 one of the canons made a donation to the chapter, from which we learn that the latter must have consisted of at least six members, since so many are mentioned individually by name.12 Another donation of 1040, made by Conte Umberto I to the canons of S. Orso, has been published by Carutti.13 It seems to have been a vague tradition of this donation which inspired a curiously erroneous passage in a XVI century necrology of the cathedral, in which the Marchesi of Monferrato are named as founders of the monastery or priory of SS. Pietro e Orso.14 In 1132 at the prayer of the bishop of Aosta, Ariberto, Pope Innocent II reformed the chapter and established canons regular of the order of St. Augustine at S. Orso. The papal bull on this subject is still extant.15 On the capital of the cloister of the church (Plate 14, Fig. 2), moreover, there is an 8 Quapropter ego anselmus largiente diuina dementia episcopus augustensis ecclesie et comes. Notum esse uolo omnibus sub Xpo principe militantibus, quod anno ab incarnatione domini nostri Ihu Xpi DCCCCXXIII, indictione XI, pro amore Dei et remedio anime mee et animarum parentum meorum et item pro remedio anime domni regis Rodulfi. dono donatumque esse in perpetuum uolo ecclesie sancte Marie et sancti Iohannis sanctique LTrsi ad communem uictum suorum canonicorum predicto rege Rodulfo laudente . . . hoc est quasdam terras, etc. (Hist. Pat. Mon., VI, 28). 9 See, however, Savio, 84. i° XVII kal. Febr. Ob. Anselmus Episcopus Augustensis qui nostram construxit ecclesiam. (Necrologium insignis collegii canonicorum Sancti Petri et Ursi, ed. Hist. Pat. Mon., V, 519). n . . . Interogati quis fundavit ipsum prioratum, Responderunt unamiter quod bone memorie dnus Anselmus quondam episcopus augustensis, quia ita describitur in eorum libro vocato martiligio [ = Martyrologio] ligato cum cathena ferrea in pulpito in medio dicti capituli. (Due, 113). 12 Hist. Pat. Mon., I, 497. is Archivio Storico Italiano, Anno 1878, quarta serie, II, 348. n . . . anno 1040, sedente in cathedra Beati Grati felicis recordationis Arnulpho episcopo, necnon regnante et principiante in valle nostra Auguste Sallassorum Humberto primo comite Maurianensi, filio illustris Beroldi de Saxonia, qui Humbertus eodem anno legauit capitulo nostro dominium loci Derbie pro duabus partibus, legata alia tertia parte cenobio seu priori Sanctorum Petri et Vrsi, quod dudum fondauerunt, et dotaverunt bone memorie Marchiones Montisferrati . . . (Extractus Anniversariorum ecclesiae cathedralis Augustae, 1 Nov., ed. Hist. Pat. Mon., V, 656). is Hist. Pat. Mon., I, 769. The bull is dated: Datum Placentie XIII kalendas decembris indictione X. Incarnationis Dominice anno MCXXXIII. Pontificatus Domni 57 LOMBARD ARCHITECTURE inscription recording this same fact.16 This inscription has been interpreted by Venturi17 to indicate that the cloisters were erected in 1133, and by Toesca to prove that they were erected before 1133. As a matter of fact, it does neither the one nor the other. Since the inscription is manifestly contemporary with the other inscriptions in the cloister, which were evidently part of the original construction, it proves that the cloisters were erected after 1132. On another capital of the cloister is a sculptured representation of the foundation of the chapter of canons regular. On one face St. Augustine, SCS. AVGVSTINVS| EPISCOPVS, with crosier, gives his blessing to the reverencing figure of Arnolfo, the first prior, ARNVLFVS. PRIMV. PIOR.,18 who is presented by S. Orso, QVI REDDIT . SCS . VRSVS. On the other side the bishop Ariberto, ARBERTVS . EPS., approaches St. Augustine and gives his '' blessing and approbation to the new order, as is evident by the inscription on the book: BENE DIGIT.PRIORS As an historical source this sculpture has considerable importance. It confirms, in the first place, by a nearly contemporary document, the fact that the chapter of canons regular was founded in 1132. In the second place, it demonstrates that the cloister was erected soon after this foundation, for the capital formed an integral part of the construction of the cloister, and it is inconceivable that it could have been sculptured and added after the construc tion of the latter. The cloister must, therefore, have been built after 1132. Nevertheless, the fact that the foundation of the chapter was given so prominent a pictorial representation gives great plausibility to the conjecture that the cloister was erected immediately after and in connection with that foundation. That the bishop Ariberto played a prominent part in the foundation of the new chapter is also indicated by the fact that he was buried at S. Orso when he died in 1139.19 Innocentii Pape II. anno III. This date has been interpreted 1133 by the editors of the Hist. Pat. Mon., erroneously. Savio (90) has shown that it should be interpreted November 19, 1132. io ANNO. AB INCARNATIOE . DNI M. C. XXX. III. IN H. CLAUSTRO. REGVLARS VITA INCEPTA. EST 17 III, 71. is That Arnolfo was the first prior of the reformed chapter is known also from a passage in the necrology of the chapter: prid. idus Aug. Ob. Arnulfus primus prior S. Ursi et Episcopus Augustensis (Necrologium insignis collegii canonicorum Sancti Petri et Ursi, ed. Hist. Pat. Mon., V, 531). io Due, I, 15 f. 58 AOSTA, S. ORSO According to a tradition handed down by the historians of Aosta, there was an inscription above the vaults of the existing church which stated that the campanile was built in 1151.20 The style of the tower gives reason to suppose that this tradition is correct, rather than that other which assigns the construction of the campanile to Goutier d'Ayme, and to the year 1131.21 The reformed chapter of S. Orso speedily acquired importance and wealth. Rich donations were made to the canons in 1136 by Guido, bishop of Ivrea.22 These were followed by a bull of privileges and immunities of Innocent II in 1136,23 and by other bulls of indulgence of Innocent II in 1142,24 and of Lucius II in 1144.25 In 1146 another donation was made to the chapter,26 and in the same year Eugenius III granted the canons a bull of privileges.27 Other papal bulls in favour of the monastery were granted by Adrian IV, c. 1159,28 by Alexander III in 1161,29 and two by Lucius IV in 1164. 30 In addition, there are extant two donations in favour of the church, without date, but doubtless of c. 1134.31 A curious feature of the internal discipline of the chapter was the presence of many female conversae, frequently mentioned in the necrology. In 1400 the chapel of S. Erasmo was founded by the canon Oddo.32 In the pastoral visit of 1419 it was ordered that the roof of the church should be repaired, since it menaced ruin.33 It was doubtless at this epoch that the existing vaults were erected. In the same pastoral visit complaint was made of the condition of the cloister, and it was ordered that here, too, a new roof should be erected.34 Apparently, however, this order was not carried out, since it was repeated in 1427. 35 This order must have occasioned the reconstruction of the cloister in its present form. With the exception of the addition of the modern grill, which greatly impedes the enjoyment of the beauty of the cloister, the monument has happily escaped modern restoration. III. The edifice, although considerably altered, evidently possessed originally a nave (Plate 13, Fig. 2) of seven bays flanked by two side aisles. 20 Toesca, 119. 21 ibid. 22 Hist. Pat. Mon., I, 773, 774. 23 Ibid., 776. 2* Ibid., 784. 25 ibid., 785. 26 Ibid., 789. ^ Ibid., 790. 28 ibid., 819. 29 Ibid., 822. so ibid., 930, 933. si ibid., VI, 218, 219. 32 III nonas Iunii. Ob. venerabilis Oddo canonicus, qui fundauit capellam S. Erasmi 1400 . . . (Necrologium insignis collegii canonicorum Sancti Petri et Ursi, ed. Hist. Pat. Mon., V, 526). 33 . . . reformetur solare desuper per longitudinem ecclesie quia in pluribus locis minatur ruynam. (Due, 113). 34 Subsequenter accessimus ad claustrum quod reperimus de novo fuisse coopertum. Ordinavimus tamen reparari et bituminari in ipso clsustro et in capitulo in locis caducis et ruinosis. (Ibid., 122). 35 Et item coopertura claustri [reparerentur] ejusdem prioratus secundem exigen- tiam defectus quia propter hujusmodi defectum perspeximus picturas ejusdem claustri de vita sancti ursi esse deletas (Ibid., 148). 59 LOMBARD ARCHITECTURE The crypt (Plate 15, Fig. 1) is placed only under the choir, and does not extend beneath the side aisles. The original piers were probably rectangular, but the rectangular section of the existing piers (Plate 13, Fig. 2) is evidently the result of alterations executed in the XV century, when new masonry was built around the original supports so as to strengthen them to bear the weight of the new vaults. The segmental curve of the arches of the main arcade (Plate 13, Fig. 2) shows that such an alteration has been made.36 Super ficially, the edifice with its Gothic vaulting and flamboyant choir-screen, appears to have been entirely rebuilt in the XV century. As a matter of fact, however, in addition to the crypt, the core of the old X century basilica is still extant, although covered by intonaco. Above the Gothic vaults are still preserved the old clearstory walls constructed of rubble, in which large, round stones are laid in thick beds of very poor mortar. The large, round-headed clearstory windows are of the type of those of S. Vincenzo at Milan. Upon the walls are still to be seen notable remains of frescos which I take to be the original ones of the X century. The crypt (Plate 15, Fig. 1) is covered by very crude groin vaults, which are for the most part original, although repaired in the XVIII century. There are no capitals, and the transverse ribs die away towards the springing. The vaults are not domed. The supports consist of crudely hewn monoliths, square or cylindrical in section. To the south of the church lie the beautiful cloisters (Plate 12, Fig. 6), the arcades of which on three sides still retain their original columns of the XII century. The archivolts and bases, however, were made over in the XV century, as was the entire eastern gallery, and in the same epoch were added the rib vaults resting on corbels, which cover the ambulatory. The imposing campanile is a characteristic monument of the local style of the Val d' Aosta. It is constructed of rough ashlar masonry, and is evidently a homogeneous work of about the middle of the XII century. It terminates in an octagonal spire with four angle turrets, a motive which shows the influence of the transitional clochers of the Ile-de-France. IV. None of the ornament — if any ever existed — of the XI century basilica is extant with the exception of the frescos of the clearstory walls preserved above the existing vaults.37 Among the subjects represented are Elijah, the apostles Andrew and John, the cities of Patras and Ephesus, angels, soldiers, etc. Above runs a Greek fret like that of Spigno, in which birds are represented at intervals. The technique of these frescos, with their broad strokes of black, is extremely similar to that of the frescos of Spigno. It has frequently been written that the cloisters are Provencal in style. 36 The tradition current at Aosta that the side aisles were added in the XV century is disproved by a study of the masonry. (Due, I, 225). 37 These frescos have been illustrated by Toesca, 88. 60 AOSTA, S. ORSO As a matter of fact, the sculptures are thoroughly Lombard and were, as will be seen, executed by a Lombard master. The archivolts of the arcade, which are the only part of the cloister which really shows Provencal influence, are of the XV century. A peculiarity of the style is the fact that many of the shafts of the colonnettes are square or polygonal. The capitals, especially those which are sculptured, form a most interesting study. Since the subjects of many of them have never been explained, I shall give a list of them in order, beginning at the north-west angle and proceeding systematically to the south and east around the cloister. (1) The north-west angle capital is ornamented with broad leaves and has no figure sculptures (Plate 12, Fig. 6). (2) Coupled capitals, the inner one of which shows the birth of Jacob and Esau (Plate 13, Fig. 3). In a bed with many turned knobs (a most interesting representation for the history of furniture) lies REBECCA. By her side, with folded hands, sits a midwife, OBSTETRIX. From Rebekah emerge two twins, first ESAV and then IACOB, the latter holding his brother's heel (Gen., xxv, 25). The outer capital represents the deception of Isaac. On a similar bed lies the patriarch YSAAC, who feels IACOB, on whose hands and about the smooth of whose neck is the goatskin, while REBE|CCA brings savoury meat, two good kids, on a platter. Meanwhile ESAV shoots a stag with his cross-bow (Gen., xxvii, 1-29). (3) (Plate 15, Fig. 3) IACOB lies on the ground, or, rather, is on all- fours, face downward, asleep; with his left hand he supports a ladder, SCALA, on which are seen two angels, ANGLI, ascending and descending; to the left is the Lord, DNS (Gen., xxviii, 11-16). On the west face of the same capital IACOB is seen seated. He holds in his lap an object difficult to identify, but which is perhaps intended to be a bag of money, symbolizing the wealth he has accumulated with Laban. On the south face REBECCA is seen seated.38 On the east face is a seated male figure, with beard, holding in his hand a sort of fleur-de-lis. There is no inscription, but there can be no doubt that the sculptor here wished to represent Jesse and his rod, the fruit of the union of Jacob and Rebekah. (4) Leah, LIA, and Rachel, RACHEL, are seen seated. The latter is decidedly the younger and better looking, and is characterized by long, flowing hair. Next to Rachel is seen Jacob, IA[COB], who is engaged in lifting the great stone, which looks like a cover, from the well, PVTEVS (Gen., xxix, 2-10). On the other faces of the capital are represented the flocks of Laban, swine, sheep, a camel, calves. The outer capital is entirely occupied with similar flocks, goats, swine, sheep and cattle.39 ss Illustrated by Venturi, III, Fig. 62. 39 Illustrated by Venturi, III, Fig. 60. 61 LOMBARD ARCHITECTURE (5) The meeting of IACOB and ESAV, who are seen embracing each other (Gen., xxxiii, 4). Beyond Jacob are LIA and RACHEL and three sons of Jacob, FILII (Gen., xxxiii, 1-3), whom the sculptor has put to symbolize twelve, owing to lack of space. On the other side of Esau are his handmaidens and household, ANCILLE . FAMILIA ESAV, all moving to the left and apparently all girded together by ropes. Then the household of Jacob, FAMILIA IACOB, similarly with rope girdles, and holding whips or flails in their hands. In both households the beardless figures have longer shirts and are perhaps intended to represent women. Jacob's household is preceded by camels and sheep, TVRMA CAMELORVM; GREX. PECORVM (Gen., xxxiii, 1-16). (6) The outer capital shows LABA[N] who comes to the tent and draws aside the curtains. Within sits RACHEL, feigning that it is with her after the custom of women. Below emerges the head of the image, l|DO|LU (Gen., xxxi, 33-35). Next is represented LIA sitting bolt upright, her hands crossed upon her lap. On the east face of this capital is one of the male children of Jacob, clothed in a curious woolen cloak, which the sculptor seems to have adopted to indicate the twelve sons of Israel. On the inner capital are three similar figures, doubtless other sons of Jacob, and on the west face of this capital is the figure of a woman, unnamed, who may be either Rachel or Dinah, Jacob's daughter (Gen., xxx, 21). 40 (7) IACOB is seen wrestling with the angel, ANGS (Gen., xxxii, 24). On the other face is seen Jacob, now called by his new name, ISRAHEL (Gen., xxxii, 28), holding a staff, and accompanied by LIA and RACHEL. (8) On the inner and outer capitals are eight more sons of Israel, which, with the four shown in No. 6, make twelve altogether. The four on the outer capital are distinguished by the inscriptions, SIMEON, IVDAS, IOSEP (this Joseph carries his coat of many colours hung from a pole across his shoulders) and RVBEN. (9) The south-west angle capital is decorated with broad leaves, but has no figure sculptures. (10) Four birds of no iconographic significance. (11) Capital ornamented with interlaces and grotesque goats' heads. On the abacus is the inscription in rhyming hexameters : MARMORIBVS — VARUS . HEC . EST . DISTINCTA . DECENTER: FABRICA . NEC . MINVS . EST . DISPOSITA . CONVENIENTER . (12) On each angle is depicted the figure of a seated man with close- fitting shirt and short skirt. Two of the figures are beardless, two have beards. 40 This capital is illustrated by Venturi, III, Fig. 58. 62 AOSTA, S. ORSO Each man takes hold with either hand of plants that grow from pots placed in the centre of each face.41 This capital has no iconographic significance. The five following capitals are all symmetrical, and have on each face a medallion with the bust of a prophet, who is haloed and holds a scroll: (13) On the west face, *f< ZACHARIAS, with the inscription which over runs the scroll, IRATVS. E. DNS | POP VLO SVO (Zechariah, i, 2). On the north face, Haggai, AGGEVS, with the inscription which overruns the scroll, EGO MOVEBO.CELVM (Haggai, ii, 21, 22, in Vulgate). On the east face, SOPHONIAS, with the inscription which overruns the scroll, LAVDA.FILIA SYON (Zephaniah, iii, 14). On the south face, MALACHIAS, with the inscription that overruns the scroll, MALEDICTVS . DOLOSVS (Malachi, i, 14). (14) On the south face, ?r-YSAYAS, with the inscription EGREDIETUR (Isaiah, xi, 1). On the west face, DANIHEL, with the inscription which overruns the scroll, ASPICIEBAM . IN VISV NOCTIS . (Dan., vii, 13). On the north face, ¦* IEZECHIEL, with the inscription, PATRES . COME- DERVNT . VVAM . ACERBAM (Ezekiel, xviii, 2). On the east face, ¦*¦ HIEREMIAS, with the inscription which overruns the scroll, HIC '-. EST: DEVS iNOS (Baruch, iii, 36-37). (15) On the west face, NAVM, with the scroll, SOL . ORTVS . E . (Nahum, iii, 17). On the north face, MICHEAS, with the inscription which overruns the scroU, PERIIT . SCS . DE TERRA (Micah, vii, 2). On the east face, IONAS, with the inscription which overruns the scroll, DE . VENTRE . INFERI: (Jonah, ii, 2). On the south face, ABACVC, with the inscription that overruns the scroll, VE . QVI . POTV' . DAT AMICO . SVO (Habakkuk, ii, 15). (16) On the south face, 4" OSEE, with the inscription that overruns the scroll, ET . ERIT . QVASI OLYUA . GL[ORI]A . EIVS (Hosea, xiv, 6; in Vulgate xiv, 7). On the west face, 4"ABDIAS, with the inscription that overruns the scroll, P[ER]DA . SAPIENTES . DE ]EDVMEA (Obadiah, 8). On the north face (Plate 14, Fig. 1), *r- AMMOS, with the inscription that overruns the scroll, DNS . DE SYON . RVGIET (Amos, i, 2). On the east face, Joel, *¦ IOEL, with the inscription, . PLANGE . QVASI . UIRGO (Joel, i, 8). (17) On the west face, BALAAM, with the inscription that overruns the scroll, MORIATVR . ANIMA| MEA . MORTE . IVSTORVM (Numbers, xxiii, 10). On the north face, NAT AN, with the inscription that overruns the scroll, DNS . TRANSTVLIT . |PECCATVM . TVVM . N . MORIERIS "Venturi, III, Fig. 57; Toesca, 117. 63 LOMBARD ARCHITECTURE (II Samuel, xii, 13). On the east face Moses, MOYSES, with the inscription that overruns the scroll, CANTEMVS . DO|MINO (Exodus, xv, 1). On the south face, HELYAS, with the inscription that overruns the scroll, FACIAMVS HIC.TRIA TABERNACVLA (Matt., xvii, 4). (18) This capital represents the life of S. Orso (Plate 14, Fig. 3). First, on the west face, is seen the saint, S[ANCTVS] VRSVS, giving alms to the poor, PAVP[ER]ES, who kneel naked before him. Above their heads hangs a great stone with a ring in it, -possibly intended to signify the privations under which they suffer. The sculptor has followed a different legend of the saint from the one that has come down to us,42 and there are several details of this capital which it is impossible to interpret. In the next scene we see the saint, S[ANCTVS] VR]SVS, who strikes with his cane a fountain, FONS, from which emerge three streams of water. Above is the church, ECCL [ESI] A, subsequently erected to commemorate the miracle. Finally, we see the saint, S[ANCTVS] URjSVS, seated and holding a book in his right hand, his left hand raised in exhortation. He is approached by a servant of the wicked bishop, Ploceano, riding upon the horse he has stolen, ARMIGER.ERRANS] EPI . |CV . PALAFREDO. The saint receives the confession of the penitent transgressor, and, after having imposed a penance upon him, goes to the bishop to implore the latter to forgive his servant. In the following scene the wicked bishop is shown seated on his throne and holding a crosier. The saint, holding his cane, kneels before him, S[ANCTVS] VR|SVS| RO|GANS| P[RO] ARMIGERO . |EPM . PLOCEANVM. The bishop feigns to grant the saint's prayers, and the servant joyfully leaves his sanctuary, only to be taken by the underlings of the bishop and cruelly tortured. The saint, in consequence, pronounces a curse against the wicked bishop. In fulfilment of the saint's prayers, that very night the bishop, EPISCOPVS, was cruelly tormented in his bed by devils, DIABOLI. Near by are two crows, CORVI, who are present at his agony as birds of ill omen. Finally, his throat is cut by the demons, a scene with which terminate the sculptures of the capital, HIC . IUGVLATVR. On the abacus of the capital is inscribed in rhymed hexameters an invective against the wicked bishop, which is a delightful example of the f orcef ulness of mediaeval Latin : * ECCE . DI [= diem] . SCM . Q[I] A . FALLERE . N . TIMVISTI . DEMONIIS . ES . PDA . MISER . Q [I] A SIC . MERVISTI ; PRESVLIS . EXEMPLO . S VBEANT . NIGRA . TARTARA . LVSI ; QVI . NOS . T[= item] . PVGNANT . CECA . FORMIDINE . FVSI . (19) The raising of Lazarus.43 Above is an inscription, much defaced, 42 Quoted above, pp. 55-56. *3 Illustrated by Toesca, 117. 64 AOSTA, S. ORSO which appears to be a paraphrase of John, xi, 39. 44 Christ between the Alpha and Omega, touches the tomb and commands Lazarus to come forth: LAZARE. VENI.FO|RAS.DE SEPVLCRO: (John, xi, 43). Lazarus is seen awakening in the tomb, behind which are Mary and Martha and three other persons (John, xi). On the south and east faces of this capital are represented two apostles, SCS SYMON and S ANDREA. (20) Christ, IHC NAZARE|NVS, a fine figure, although unfortunately broken, occupies the most prominent position in this capital. His feet are anointed by MARIA MAGDALENA, whom MARTA tries to restrain (John, xii, 3). On the north and east faces are IACOBVS . ALFEI and BARTOLOMEVS. (21) This is a capital representing the foundation of the chapter regular described above.45 On the north face is shown SCS . PETRVS . AP[OSTO]L[V]S, with two keys. (22) See Plate 14, Fig. 2. This is a capital with a conventional anthemion ornament of the Modenese type. On the abacus is the inscription recording the foundation of the chapter regular cited above.46 (23) The capital at the south-east angle has broad leaves of purely conventional type. (24), (25), (26) These three capitals, with purely conventional ornament, were made when the cloister was reconstructed in the XV century, more or less in imitation of the XII century capitals. They are much larger than the others. (27) The north-east angle capital is ornamented with grotesques. (28) On this capital are seen the three children of Israel in the smoking, fiery furnace. The fire is being poked by an executioner, and an angel flies above. On another face is the image which Nebuchadnezzar, the king, had set up, and the king himself, with effeminate, beardless face, holding his sceptre languidly in his hands (Daniel, iii). (29) The Annunciation. The Virgin is seated and holds a distaff in her hand. Curiously enough she is crowned. By this detail the sculptor doubtless wished to recall her royal lineage, since, on the other face, he has placed two of her ancestors. The one who plays the violin is undoubtedly David, the other, who holds in either hand a lily, is possibly Jesse. (30) The Nativity. Mary lies in a bed like those already described. A woman holds the curtains. Joseph, seated, puts his hands to his face. The child is seen in the manger with the ox and the ass. (31) The three Magi, all crowned and bearing vases, stand before Herod, who is also crowned and holds a sceptre. 44 I believed that I could read the letters DIX . EI . IHC M . ET MARTA IASSO NARINV and above in smaller letters: DVS V DEI SALVE but I am unable to explain them. « p. 58.' 46 p. 58. 65 LOMBARD ARCHITECTURE (32) Joseph, in bed, asleep, is warned by the angel to fly.47 There follows the flight into Egypt. (33) A capital of conventional Corinthianesque pattern. (34) This capital, which has been strangely misunderstood, represents the stoning of Stephen. Stephen, tonsured and naked, is being stoned, and sees in his vision the glory of God represented by an angel and a ladder (Acts, vii, 55-60). Saul, tonsured and with characteristic features, is seated and holds a pile of stones in his lap. Beside him stands another person. Both Saul and his companion are casting stones at Stephen. (35) This capital seems to represent a simple genre scene of monastic life. All the figures are tonsured and evidently represent canons of S. Orso, with the exception of one who is bearded, and hence is doubtless a lay brother. The latter draws water from a well. The two canons prepare the repast, bringing both jugs of wine and platters. (36) This capital represents four grotesque birds, with human heads. (37) This capital represents the fable of the crane and the fox.48 The crane, having invited the fox to dine, provides for the repast a narrow-mouthed jar. The fox, returning the compliment, invites the crane to dine out of a shallow dish. (38) and (39) are both conventional. In addition to these capitals there are in the cellar of the Museo Civico di Arte Applicata ad Indus tria at Turin three capitals which evidently came from the cloister. The first, which bears the number 2609, is of the same type as the one with the inscription described above (22). 49 The second, without any number, is sculptured with representations of Adam and Eve and the serpent.50 The third capital, No. 2608, appears to represent the expulsion from Paradise.51 Adam and Eve are dressed as peasants with pointed hoods. Adam is bearded and bare-footed. He carries a cane, or, perhaps,. an instrument of agriculture, in his left hand. He appears to be standing' still, and to expostulate with the angel, pointing at Eve as the guilty one. The angel gestures with his left hand. Eve wears shoes and anklets. On the other faces of the capital are a goat and two sheep, representing the flocks of Adam and Eve. From the style of the sculptures of the S. Orso cloister, it is evident that the latter are closely related to the work of the sculptor who executed the, ( pulpit at Isola S. Giulio. This is clear from numerous analogies. In both we find the same massive, impassive faces, with the same heavy lower jaw; in both the use of the same hard marble; in both the same superlative skill 47 Toesca, 118, illustrates this capital. « Illustrated by Venturi, III, Fig. 59. 49 Illustrated by Venturi, III, Fig. 81. so Illustrated by Venturi, III, Fig. 63. si Illustrated by Venturi, III, Fig. 64. 06 ARSAGO, BATTISTERO in conventional ornament; and in both the same curious draperies, which, in some cases, look as though they were made of leather. S. Giulio and S. Orso both carry the same peculiar kind of cane. The feathers of the birds are treated exactly alike. Many of the shafts at Aosta are covered with flutings in diapered patterns, precisely as at Isola. However, the work at Isola is finer, and appears to be the prototype of that at Aosta. The numerous analogies of the Aosta cloisters with the Milan-Pavia-Lodi school, and with that of Guglielmo da Modena, appear to be the result, not of direct influence, but of influence exerted through the medium of the sculptures at Isola. It is therefore entirely probable that the sculptor at Aosta, if not a pupil of the Isola sculptor, was at least strongly influenced by him. V. The architectural character of the crypt and of the core of the church leave no doubt that in them we have relics of the church of 923. The cloister was erected in the years immediately following 1133. The style of the campanile confirms the tradition that it was erected in 1151. ARSAGO,1 BATTISTERO (Plate 15, Fig. 4, 5) I. The baptistery of Arsago has been published and illustrated by De Dartein,2 and by Pareto. The recent notice by Serafino Ricci contains a list of the authors who have referred to the monument incidentally. II. With the exception of an inscription recording a restoration in 1874, now preserved inside the baptistery, there are no historical documents relating to the monument. When I visited Arsago in 1909, I found the interior of the baptistery completely blocked up by scaffolding, as shown in the photograph (Plate 15, Fig. 4), but the restoration was not in active progress. When I returned in July, 1913, however, work had long been finished. The campanaio told me that the restoration lasted from c. 1900 to 1911. III. The plan of the baptistery of Arsago is peculiar, and, I believe, without analogy in northern Italy. The edifice (Plate 15, Fig. 5) is octagonal, with walls of enormous thickness, but lightened in the ground floor by a series of niches (Plate 15, Fig. 4), all rectangular except the easternmost, which is semicircular. These niches open off the interior of the nave, but are not expressed externally, being merely constructed in the thickness of the wall. In the second story the walls are lightened by a gallery (Plate 15, Fig. 4) covered with groin vaults, and by an eastern niche, all carried in the thickness of the walls. i (Milano). 2 395. 67 LOMBARD ARCHITECTURE The central area is surmounted by a sixteen-sided cloistered vault, which has almost the character of a dome. The octagon is worked to a sixteen-sided plan by means of arched squinches in two orders placed just above the gallery, and in the vault itself the re-entrant angles are smoothed out so that the plan becomes almost circular. There is a system in each angle (Plate 15, Fig. 4) supporting corbel-tables at the level of the gallery. Above the gallery and below the vault is a small clearstory (Plate 15, Fig. 5) of oculi and windows in the form of a Greek cross, or round-arched. Apparently no timber is used in the construction of the roof, which is formed of stones laid directly upon the extrados of the vaults (Plate 15, Fig. 5). In the middle of the central area is a depressed font with two steps. The vaults of the gallery, trapezoidal in plan, and with low transverse arches, are so highly domed that they resemble barrel, rather than groin, vaults. There are no wall ribs, but the wall arches, though depressed, rise to a much higher level than the main-arcade arches. Unfortunately these vaults were all remade in the recent restoration. The masonry (Plate 15, Fig. 5) is quite different from that of the church and consists of large, rectangular blocks, laid, however, in courses of which the horizontality is not infrequently broken. The mortar-beds are about 15 centimetres in breadth. IV. The capitals of the main arcade of the gallery are small uncarved blocks ; those of the system and gallery responds are without abaci, and sculptured with grotesques or simple leaf patterns. The clearstory is ornamented externally with a series of semicircular blind arches of a simple character. The cornices are formed of the usual corbel-tables without pilaster strips. V. The baptistery is nearly contemporary with the church (c. 1120), as is shown by the character of its capitals and arched corbel-tables. The masonry, however, is quite different, in that much larger blocks are employed (Plate 15, Fig. 5; Plate 16, Fig. 1). The explanation of this is undoubtedly partly to be found in the fact that the baptistery was a vaulted edifice, where greater strength was required in the walls than in the wooden-roofed basilica. Nevertheless it is reasonable to suppose that the church, the more essential edifice, was erected somewhat before the baptistery. The latter, therefore, may be assigned to c. 1130. ARSAGO,1 S. VITTORE (Plate 15, Fig. 2; Plate 16, Fig. 1) I. The first author to call the attention of archaeologists to the pieve of Arsago was Giulini,2 who printed a brief description of the church (which he i (Milano). 2 1, 358. 68 ARSAGO, S. VITTORE ealled S. Maria del Monticello) in connection with the historical events of the year 892. The monument was subsequently studied and illustrated by De Dartein. II. Of the history of the church nothing is known. The earliest notice of it which I find is in a sort of tax-list of 1398, published by Magistretti. From this we learn that the church was officiated by a chapter of eight canons, and enjoyed jurisdiction over sixteen chapels. III. The edifice consists of a nave (Plate 15, Fig. 2) of four double bays, two side aisles and three apses. The wooden porch which De Dartein mentions as preceding the facade has disappeared, doubtless in the restoration of 1892. The building is a simple basilica without vaults save for the half domes of the apses. The supports are alternately columns and square piers (Plate 15, Fig. 2), and there is no system. The side aisles are much higher than the level of the crowns of the arches of the main arcade. The triforium space becomes therefore disproportionately lofty (Plate 15, Fig. 2). The clearstory (Plate 16, Fig. 1) is formed of good-sized, round-arched windows, which, like the other windows of the church, were evidently glazed. The masonry (Plate 16, Fig. 1) consists of small, brick-shaped stones of irregular shapes and unsquared, laid in courses for the most part horizontal, an occa sional large block being inserted usually to form a sort of quoin at the angles. The mortar-joints vary from 1 to 4 centimetres in thickness. The campanile (Plate 16, Fig. 1) rises to the north of the church, and is so placed that its southern wall corresponds with the northern wall of the side aisle. It leans violently to the north in its lower stories, but returns towards the vertical in its upper part. It is illuminated by oculi, rather irregularly placed. The masonry, like that of the church with which it is contemporary, is supplied with numerous scaffolding holes. The stonework of the belfry is obviously somewhat later than that of the rest of the structure, and it is evident that the original belfry has been walled up. It is possible that there were originally two stories of bifora. The bells have recently been removed from the belfry and placed on top of the tower. IV. The interior of the edifice (Plate 15, Fig. 2) preserves nothing of its ancient ornament except the capitals, the walls having been covered with plaster and painted with modern frescos of the most deplorable taste. The two eastern capitals have been restored. The second pair are of a curious Corinthianesque type, which recalls more strongly the capitals in the crypt of the cathedral at La Scala (Salerno) than anything I have seen in northern Italy. The northern capital of the second bay from the west is of a more usual Corinthianesque type with uncarved acanthus leaves, closely imitated from the antique. The remaining capitals are pilfered Roman. The bases are at present covered by the pavement. The archivolts are of a single order, unmoulded, and the piers are without so much as an impost moulding. 69 LOMBARD ARCHITECTURE Externally the edifice and campanile are adorned with the usual cornices of arched corbel-tables and saw teeth (Plate 16, Fig. 1), but for the most part are severely simple. The capitals of the upper story of the campanile seem to be contemporary with those of the church, and were doubtless retained when the belfry was rebuilt. V. 4 The architectural forms and the plan of the church recall strongly the basilicas of Verona erected during the XII century, such as S. Giovanni in Valle, S. Pietro in Castello, etc. S. Giovanni in Valle (Plate 218, Fig. 4), which was rebuilt after the earthquake of 1117, has not only a plan which is nearly identical with that of the pieve of Arsago, but has several capitals which present the closest analogy with those of our monument. Moreover, the masonry at Arsago, while crude, is still much superior to that of S. Vincenzo at Abbazia di Sesto Calende (1102). The use of arched corbel-tables without pilaster strips except at the angles also accords well with the style of the first quarter of the XII century. We may, therefore, with considerable confidence, ascribe the pieve of Arsago to c. 1120, with the exception of the belfry, which has been remade, perhaps in the XVIII century. ASTI,1 S. ANASTASIO I. In the cellar of the existing Collegio Nuovo at Asti are the remains of the convent of S. Anastasio. The church itself, a structure of the XVI century, was recently destroyed to make room for the new school buildings, but the fragments of ancient architecture extant in its crypt were scrupulously preserved. These remains, which have been carefully studied by Brayda and Bevilacqua-Lazise in a monograph on the crypts of Asti, are of considerable importance for the history of art. The handbook of Bevilacqua-Lazise in the Bonomi series contains excellent half-tones of the capitals of S. Anastasio. Several notices important for the history of the convent have been contributed by Savio. II. The church was in existence as early as 792, since it is mentioned in a permutation of that year.2 According to a catalogue of the bishops of Asti, written in 1605, but containing notices of much earlier date, Alderico, bishop of Asti, who founded in 1027 the monastery of S. Giusto at Susa, made a donation to the nuns of S. Anastasio.3 According to Savio, who edited this text, the donation in question was made in 1029. The convent must, i (Alessandria). 2 Gabotto, Asti, 3. 3 Aldericus ep. ast. Manfredi comitis Sabaudiae et Secusiae marchionis frater sub Io. XX, 1027, inter cetera pietatis officia monasterium S. Iusti Secusiae fundavit et dotavit. Monalibus S. Anastasij civitatis Astens. bona S. Cristophori etc. (Ed. Savio, Vescovi, 111). 70 ASTI, S. ANASTASIO therefore, have been already in existence in the second quarter of the XI century. The same catalogue further informs us that Pietro I, bishop of Asti, in 1042 made a further donation to the nuns of S. Anastasio, and reformed them according to the rule of St. Benedict.4 The catalogue attributes further donations in this same year to Bishop Odo I,5 but Savio omits Odo I from his list of the bishops of Asti and assigns the donation to Pietro II and the year 1043.6 Further donations were made to the convent, one by Bishop Anselmo I in 1068, another in 1096 by Odo III.7 In 1070 Asti fell into the hands of the contessa Alaxia, who in 1091 burned the city to the ground.8 It is probable that the church of S. Anastasio was destroyed by this fire and rebuilt immediately afterwards, since the style of the later portions of the edifice is that of the last years of the XI century. Other fires occurred in Asti in 1145 and 1155, but appear not to have injured our monument. In the XVII century the church of S. Anastasio was recon structed, with reversed orientation, in the barocco style. The ancient crypt, however, remained accessible. When, about 1907, the barocco church made way for the existing school, the foundations of an earlier church of the XI century came to light, and these, together with the crypt, have, as has been said, been carefully preserved.9 In addition to the foundations already mentioned, three capitals belonging to the Lombard edifice came to light during the destruction of the XVII century church, and are preserved in the courtyard of the neighbouring Museo Alfieri. III. From the fragments which remain of the Romanesque church it is evident that the monument consisted of a nave of three double bays, two side aisles, of which the northern was much wider than the southern, a choir, probably separated from the side aisles by a solid wall, three apses and a crypt. The system of the church was alternate, like that of S. Savino at Piacenza (Plate 183). From the section of the piers it is clear that the nave was covered with rib vaults, since the system must have consisted of at least three shafts. The intermediate piers were quatrefoiled, and the aisle responds comprised five members. 4Petrus I ep. ast. sub eodem Benedicto [VIII], 1042, . . . partem castri Brado- lensis, multaque alia Monialibus S. Anastasij contulit, easdemque ad regularem observantiam B. Benedicti redegit. Sedit menses octo. (Catalogue of Bishops of Asti of 1605, ed. Savio, Vescovi, 111). 5 Otho I. ep. ast. sub eodem Benedicto, 1042, Monasterio S. Anastasii donationibus apud Valpianum et Montanarium ac Vafenaria a suis predecessoribus factis multas et copiosissimas adiecit. . . . Sedit annum. (Ibid., 112). e Savio, 137. 7 See text cited below, p. 73. A document of 1182 relating to the church has been published by Gabotto e Gabiani, 71. s . . . ab ea tota succensa fuit. (Chronicon As tense, ed. Muratori, R. I. S., XI, 141). »Bolletino d'Arte, II, 1908, 232. 71 LOMBARD ARCHITECTURE The crypt, which terminated in a polygonal east end, and was divided into three aisles by two rows of seven columns, extended not only under the choir but into the first bay of the nave. It shows two distinct eras of con struction. The three middle pairs of columns are Carlovingian and doubtless the remnant of an earlier church which terminated in an apse placed much to the westward of the later eastern limit of the church. In the XI century the crypt was much extended by the addition of three bays to the eastward and three to the westward. This crypt is now entirely covered by domed groin vaults with disappearing transverse arches. The Carlovingian vaults are less domed and rise to a lower level than do those of the XI century. It is evident also that the level of the pavement of the crypt in the Carlovingian epoch was lower than that of the Lombard pavement, for the bases of the Carlovingian columns are buried. IV. The six capitals of the Carlovingian epoch in the crypt have been studied and carefully illustrated by Bevilacqua-Lazise. Two, numbered 1 and 4 in Bevilacqua-Lazise's monograph, and illustrated on page 17 of his handbook, are evidently pilfered and taken from some earlier edifice. No. 1 is a typical work of the Roman decadence, and may be assigned to the IV century. No. 4 shows Byzantine tendencies in the bulge of the bell and in the crisp, sharp form of some of the acanthus leaves, and may be assigned to the early V century. The other four were evidently worked for their present position. No. 5, illustrated on page 18 of the handbook, is very similar to a capital at S. Vincenzo in Prato at Milan (Plate 137, Fig. 3), and is characterized by volutes and a single row of Byzantinesque acanthus leaves. No. 3, also illustrated on page 18, shows strong points of contact with another capital of S. Vincenzo (Plate 137, Fig. 1). Capital No. 2, illustrated on page 19, is so similar to a capital of the crypt of S. Giovanni at Asti (Plate 16, Fig. 3) that it must be by the same hand. The capitals of the Lombard period in the crypt are strikingly similar to those of the crypt of Modena, being characterized by graceful swirls of foliage, intermingled with grotesque motives, or by volutes and uncarved acanthus leaves. Those of the upper church, now in the Museo Alfieri, are adorned with birds, grotesque animals grouped two and two by a single head which forms the volute, rinceaux and anthemia. They show very close analogy with the capitals of S. Savino at Piacenza. V. Bevilacqua-Lazise has assigned the crypt of S. Anastasio to between 770 and 793. The style of the capitals, however, indicates a somewhat later epoch. One, as has been seen, shows strong points of contact with the capitals of the crypt of S. Vincenzo in Prato at Milan, an edifice of c. 830, and another appears to be by the same hand as the capital in the crypt of 'S. Giovanni at Asti, which is authentically dated 885. We may, therefore, assign this portion of the crypt to c. 860. As for the later portion of the edifice, the analogy of the capitals of the crypt with those of the crypt of the cathedral of Modena 72 ASTI, S. GIOVANNI (1099-1106) and that of the capitals of the upper church with those of S. Savino at Piacenza, a monument consecrated in 1107, justifies the conclusion that the edifice was rebuilt after the fire of 1091. ASTI,1 S. GIOVANNI (Plate 16, Fig. 3) I. The interesting but fragmentary remains of the crypt of the church of S. Giovanni, which serves at present as baptistery to the cathedral of Asti, have been called to the attention of archaeologists and admirably published by Bevilacqua-Lazise. The little handbook of the same author in the Bonomi series contains half-tones of the capitals even better than those in the more elaborate monograph. II. The history of the church of S. Giovanni is wrapped in considerable obscurity. Bevilacqua-Lazise plausibly conjectures that the Carlovingian remains which at present exist did not originally, as to-day, form a crypt, but were part of a basilica which was, indeed, at one time the cathedral of Asti. Of the early history of the cathedral building of Asti little that is definite is known, despite numerous documents which refer to the material possessions of the episcopal see. According to Bevilacqua-Lazise, the cathedral has been placed in its present site since c. 800. 2 A diploma of January 11, 885, mentions that the episcopal archives at Asti had been destroyed by fire.3 This fire is mentioned also in the catalogue i (Alessandria). 2 Little faith is merited by the statement of the catalogue of the bishops of Asti of 1605, that in 1090 the cathedral was entitled S. Aniano: Oddo II ep. ast. sub eod. Urbano, 1090. Privilegia et donationes monialium S. Anastasij ast. diplomate confirmavit in castro veteri ipsius episcopi, ubi tunc erat ecclesia cathedralis sub invocatione S. Aniani martyr, dicata, cuius sacrae reliquiae in basilica S. Sixti hodie asservantur. (Ed. Savio, 112). According to Savio (143) it was Ottone III, not Odo II (1008-1098) who made, not in 1090, but in 1096, the donation in question to S. Anastasio. 3 Jn nomine, sancte et indiuidue trinitatis Karolus diuina fauente Clemencia Jmperator Augustus. . . . Reuerentissimus Episcopus et Archicancellarius noster nostre innotuit celsitudine quod peccatis exigentibus. casu inprouiso accidente thesaurum sancte Astensis Ecclesie cui ioseph episcopus preesse dignoscitur igne crematum fuerat in quo uaria. Jnstrumenta cartarum oblationes uidelicet et donationes imperatorum ducum Comitum aliorumque sancte ecclesje fidelium que pro diuini cultus amore. eidem Ecclesie contulerunt eedem igne combuste sunt Super quo idem lituardus. venerabilis Episcopus summusque consiliarius noster sumissis petitionibus nostram exorauit magni- tudinem quatinus pro dei amore et remedio anime nostre seu coniugis ac prolis necnon pro debita ueneratione eiusdem Ecclesie que constructa est in honore sancte marie semper uirginis et sancti Secundi ubi eius humatum corpus quiescit nostre Auctoritatis munificentia omnia uariarum instrumenta cartarum eiusdem ecclesie confirmare 73 LOMBARD ARCHITECTURE of the bishops, where it is stated that the bishop Lituardo reconstructed the church.4 The text of the catalogue implies that there had been a previous fire of the cathedral of Asti and this Savio places — I know not on what authority — in 820. The text is erroneous in assigning the reconstruction of the cathedral to the bishop Lituardo who, according to Savio,5 never existed. There can, however, be no doubt that in 884 the cathedral was destroyed by fire. Its reconstruction was doubtless terminated in 894, when the construction of the cloister was begun by Staurace (892-899). The catalogue of the bishops,6 indeed, credits Audace with this work, but, according to Savio, the bishop Audace sat 904-926, so that there is evidently a confusion of name. In 899 the same bishop Staurace, or Stauracio, instituted a chapter of thirty canons regular,7 supplanting the fourteen sacerdotes qui sunt Custodes mentioned as officiating the church in 876.8 In 909, under Audace, the cloister was finished.9 It is probable that the cathedral, as rebuilt c. 885, continued in use until the present imposing structure was erected in the Gothic period. The old building was then turned into a baptistery, and in the XV century was rebuilt at a higher level, some fragments of the old church being preserved in the crypt. III. From the scant fragments that remain it is impossible to determine the plan of the ancient church. The four extant columns, placed in a sort of rectangle, may well have belonged to the arcades. The groin vaults which the columns at present support, notwithstanding their crudity, are modern, and were probably erected when the church was made over in the XV century. IV. Of the capitals, one is of Byzantine Corinthian type and is undoubtedly taken from an earlier edifice of the VI century. The other two (Plate 16, Fig. 3) are of a formal voluted type. The volutes are stiff and angular; a single flat leaf, on which the veins are incised, is placed under each angle, and on each face is a medallion. On one of the capitals two of these medallions show, crudely sculptured, the heads of saints. (It may be con jectured that these saints are S. Secondo and S. Maria, the patrons of the dignaremur . . . Datum III idus ianuarij Anno incarnationis domini nostri yhesu xpisti. D.CCCLXXXIIII Jndicione tercia. Anno imperij Jmperatoris Karoli IIII. (Ed. Assandria, II, 175). 4Lytuardus ep. ast. sub Hadriano III ann. sal. 884 Imperatoris Caroli III archicancellarius cathedralem ecclesiam, denuo incendium passam, pro viribus instauravit, ad pietatem pluraque alia praestans. Sedit an V. (Catalogue of Bishops of Asti of 1605, ed. Savio, Vescovi, 110). 5 126. 6 Audax ep. ast. sub Leone V. an. sal. 894 regularis observantiae studiosiss. claustra eccl. cathedralis extruenda curavit. . . . Sedit ann. IV. (Ibid., 110). 7 Gabotto, Asti, 47. 8 Ibid., 14. o Stauratus ep. ast. sub Sergio III, 909, canonicorum claustris supremam addidit manum. (Catalogue, ed. Savio, 110). 74 ASTI, S. PIETRO cathedral). Since few other sculptured capitals of the Carlovingian period are extant, this capital is of great importance for the history of art. V. As for the date of the capitals, it is not possible to doubt that they were executed in the last quarter of the IX century. It is, therefore, entirely probable that they formed part of the edifice rebuilt after the fire of c. 885. Bevilacqua-Lazise, it is true, assigns them to the VIII century, but with evident error, since they are without analogy to authentic capitals of this epoch, such as those of S. Salvatore at Brescia (Plate 35, Fig. 1, 2, 3, 4; Plate 36, Fig. 2, 5), S. Pietro in Ciel d'Oro at Pavia (Plate 177, Fig. 2), and S. Giorgio at Valpollicella (Plate 198, Fig. 4).10 On the other hand, the S. Giovanni capitals show close relationship in design with a capital of S. Satiro of Milan (Plate 132, Fig. 5), executed c. 875, and in the technique of the veining of the leaves with two capitals of 903 in the crypt of S. Savino at Piacenza (Plate 186, Fig. 2, 3) and with a capital of the crypt of Agliate (Plate 5, Fig. 4) of c. 875. We may therefore accept with confidence the capitals of the crypt of S. Giovanni as authentically dated monuments of c. 885. ASTI,1 S. PIETRO (Plate 16, Fig. 4) I. The baptistery of S. Pietro at Asti, although illustrated by Osten,2 has remained comparatively little known. The recent study of Cipolla3 is singularly unsatisfactory, and quite unworthy of the great archaeologist. II. In 806 S. Pietro was already a pieve, and was donated to the episcopal church of Asti.4 In a diploma of 886 mention is made of a certain petrus ar[c~\hipresbiter Custus et rector AeCClesie sanCtj petrj sita Cjujtate aste? Nothing further is known of the history of the edifice. III. The baptistery lies to the south of, and adjoining, the Renaissance church of S. Pietro, and is a simple octagonal structure with a side aisle. The nave (Plate 16, Fig. 4) is surmounted by a dome in which a clearstory was originally pierced. The side aisles have groin vaults (Plate 16, Fig. 4) reinforced externally by vigorous buttresses. These groin vaults have been apparently much restored, but originally appear to have had disappearing io Notwithstanding a certain resemblance in the tufts introduced in the middle of the leaves in both cases. i (Alessandria). 2 Plates V, VI. Osten's drawings show two grotesque reliefs which have now disappeared. 3 Appunti, 54. 4 Assandria, II, 222, 5 Gabotto, Asti, 22. 75 LOMBARD ARCHITECTURE transverse arches. They are distinctly trapezoidal in plan. The masonry is rather rough, and consists of bricks, of irregular lengths, laid in horizontal courses separated by thick mortar-beds. IV. The capitals in the interior (Plate 16, Fig. 4) are all of a ponderous and heavy cubic type, roughly blocked out and obviously unfinished. The neckings were for the most part intended to have bead-mouldings, but these in many cases were never finished. The bases (Plate 16, Fig. 4) are without griffes in the columns of the arcades, but are supplied with griffes in some of the responds. Those of the main arcade have a plain Attic profile (Plate 16, Fig. 4). The plinths, originally square, have been cut down to an octagonal form in certain cases (Plate 16, Fig. 4). The archivolts of the main arcade are ornamented with a roll-moulding (Plate 16, Fig. 4), an extraordinary thing in Italy. Decorative use is made of the banding of red bricks and white stone in the supports (Plate 16, Fig. 4). The exterior is ornamented with an arched corbel-table on the cupola, but the side-aisle walls were bare. The sculptures of the Madonna (Plate 16, Fig. 4) and St. Catherine on two of the capitals of the interior are a later addition, as is also the relief of Christ surrounded by the four Evangelists and by eight saints, at present inserted in the wall. V. From the heavy proportions of the capitals, the mouldings of the archivolts and the polychromatic masonry (Plate 16, Fig. 4), it is evident that we have here a monument of the last half of the XII century. However, the original vaults appear to have been similar to those of S. Tommaso at Almenno, and probably not much more advanced. We may therefore assign the edifice to c. 1160. ASTI,1 S. SECONDO I. The church of S. Secondo has been published and illustrated by Bevilacqua-Lazise in his study of the crypts' of Asti, and also by the same author in a handbook on Asti in the Bonomi series. The church itself is entirely Gothic, but the Carlovingian crypt is of some importance for the history of Romanesque art. II. Of the early history of the building practically nothing is known. There is a tradition, apparently authentic, that the church was erected on the site where the saint suffered martyrdom, doubtless at a very early epoch. From a series of rather confusing documents it is deduced that until about the year 800, the cathedral was situated at S. Secondo.2 In a document of 1202 there is mention of a priest and canon of S. Secondo,3 and canons are i (Alessandria). 2 Bevilacqua-Lazise, 16. s Gabotto e Gabiani, 164. 76 AVERSA, CATHEDRAL mentioned again in another document of 1214.* It is therefore certain that in the XIII century there was a chapter of canons connected with the church. III. In the crypt are preserved four extremely interesting capitals of the VIII century. The vaults have been entirely remade, and indeed the church preserves nothing else of interest for the history of Lombard architecture. IV, V. The capitals of the crypt are even cruder than the crudest fragments ("B") of the Chiesa d'Aurona at Milan (Plate 114, Fig. 1), which date from c. 950, the lowest point of the decadence of the X century. They are evidently very much cruder than capitals of the third quarter of the VIII century, such as those of S. Salvatore at Brescia (Plate 35, Fig. 2, 3, 4; Plate 36, Fig. 5), or of the crypt of the Rotonda in the same city (Plate 31, Fig. 1, 2, 3, 4). Bevilacqua-Lazise assigns them to the VII century. In view, however, of their relationship with the capitals of the Chiesa d'Aurona and of the total lack of Byzantine feeling which they display, I should consider it more likely that they were executed c. 950. AVERSA,1 CATHEDRAL (Plate 16, Fig. 2; Plate 17, Fig. 1, 2, 3, 4) I. The cathedral of S. Paolo at Aversa has attracted the attention of several general historians of Italian architecture, but has been given careful study only by Schultz2 and Rivoira.3 For historical notices the work of Parente should be consulted. II. The cathedral was not founded earlier than the XI century; for, although the city of Atella existed on the site of the ancient Aversa from an early period, it was without great importance until given new life by the Normans. An inscription of late date formerly in the campanile and quoted by Parente,4 states that the city was founded by the prince Rainolfo who, according to Parente,5 ruled from 1030 to 1047. It Was the third successor of Rainolfo, Riccardo I, who, according to Parente, founded the episcopal see of Aversa, in 1053. According to others, the foundation took place in 1047. The point is much controverted.6 Riccardo I ruled from 1051 to 1078, and was succeeded by his son, Giordano I (1078-1090). A contemporary inscrip tion still extant over the now walled-up northern portal (Plate 17, Fig. 3) * Ibid., 219. i (Caserta). 2 II, 189. 3 274-276. * I, 253. b I, 118. 6 See Parente, I, 57. Cappelletti, XXI, 434, assigns the foundation of the new see to 1049. 77 LOMBARD ARCHITECTURE proves that the actual construction of the cathedral was begun by Riccardo I, and completed by Giordano I: . PRINCEPS IORDAN' RICHARDO PRINCIPE NATVS . QVAE PAT INCAEPIT . P VS HAEC . IMPLENDARE CAEPIT . Exactly what it was that the father, Riccardo, began between 1051 and 1078 and that the son, Giordano, completed between 1078 and 1090 is not stated in the inscription, but there can be little doubt that it refers to the building itself and not merely to the portal. The latter is a simple construction without great adornment, and it is inconceivable that the liberality of two princes could have been expended on merely this comparatively insignificant doorway. The historian Summonte, whose work was published in 1601, records that he saw over the great western portal (sii la porta maggiore) of the cathedral of Aversa the following inscription: Vultu iocundo Roberto dante secundo Pulchra fit haec extra satis intus & ampla fenestra.7 The inscription itself implies that it was originally placed in a window, so that it is probable that at the end of the XVI century it had already been displaced from its original position, and it doubtless disappeared entirely soon afterwards in the reconstruction of the edifice undertaken about this period. The inscription is nevertheless very important documentary evidence, for it proves that a window of the church was constructed by Roberto II, who ruled from 1127 to about 1135. This fact in turn implies that the cathedral, finished before 1090, was reconstructed in the early years of the XII century. Such a reconstruction so soon after the completion of the edifice could only have been occasioned by a disaster, and it is in fact known that in 1134 or 1135 the city of Aversa was burned.8 The conclusion is therefore justified that the cathedral, finished before 1090, was destroyed by fire in 1134, and was subsequently restored by Roberto II. This conclusion is further confirmed by the internal evidence of the building itself, which gives clear indications of a reconstruction in the XII century. How radical was the rebuilding under taken in 1134 is proved by the fact that the restoration was not finished until 1160, when the relics of the saints were translated back into the church.9 In 1255 we hear of a solemn consecration of the cathedral, which implies that a new disaster and a new restoration must have overtaken the edifice in the course of 95 years.10 In 1349 the building was much damaged by an earthquake and 7 Summonte, I, 490. « Rivoira, 274. 9 Nella chiesa maggiore di Aversa vi sono infinite reliquie de' santi trasferiti l'anno 1160. (MS. del Calefati, f. 411, cit. Parente, II, 436). io In nomine Domini amen. amen. Anno ab incarnatione Domini millesimo ducentesimo quinquagesimo quinto die Jovis tertio mensis Junij quintae decimae 78 AVERSA, CATHEDRAL was subsequently restored.11 That the damage was serious is proved by the fact that Innocent VI, in 1352, granted an indulgence to all those who should aid in the reparation of the edifice.12 The condition of the building as it was in 1468 is shown by a painting of that date in the church of S. Sebastiano al Duomo. Parente,13 who has studied this fresco, states that the cupola and campanile of the Duomo are clearly shown as they were before being trans formed. In 1592 the altar of the church was remade.14 The edifice suffered severely in the numerous earthquakes with which this unhappy region has always been afflicted.15 In 1694 the cupola was ruined by one of these shocks.16 The worst blow to the church, however, was the baroccoization begun 1703- 171517 and completed in 1857.18 III. The cathedral of Aversa at first gives the impression of being an entirely barocco edifice, but on close examination it becomes evident that the northern portal (Plate 17, Fig. 3), the ambulatory (Plate 17, Fig. 1, 2) and some portions of the central tower are mediaeval, although the whole is so covered with intonaco that it is exceedingly difficult to study the original forms. The ambulatory (Plate 17, Fig. 1, 2), which is of very large size, being about twenty feet wide, includes two rectangular and five trapezoidal bays as well as three eastern absidioles. It is entirely vaulted, the semicircular absidioles with half domes, the ambulatory itself with rib vaults which are among the most extraordinary extant in Italy (Plate 17, Fig. 1, 2). The profile of the ribs of these vaults is rectangular, and the construction is extremely massive and ponderous. The diagonal ribs spring from capitals placed at a Indictionis, Pontificatus SSmi. in Xpo. Patris Domini Alexandri Papae quarti, anno primo, dictus Dominus Alexander Papa veniens Aversam, assistentibus sibi undecim cardinalibus, Domino Joanne Caietano Cardinali, Matheo Archiepiscopo in Tronti, Episcopo Placentano [sic], Episcopo Morbonensis [sic], Episcopo beatae memoriae [sic], ad reverentiam Dei et beatissimae semper Virginis et Beatorum Apostolorum Petri et Pauli, consegravit altare, quod est in pede crucis, et concessit talem indulgentiam, ut omni anno in die consegrationis et omnibus advenientibus inibi de aliquibus bonis offerentibus, unum annum 50 dies de iniuncta salutari poenitentia relaxavit; . . . item ex speciali gratia concessit et statuit, ut circa ipsam ecclesiam S. Pauli omni anno fiat mercatum de festo Apostolorum Petri et Pauli usque ad octavas eorumdem. . . . item omnibus qui benefecerint opibus Ecclesiae S. Pauli qualibet die, hinc ad V. annos centum de iniuncta sibi poenitentia relaxavit; item in die consegrationis predictae omnibus euntibus ibi vere poenitentibus et confessis et benefacientibus ab ipso die usque ad octavas Apostolorum Petri et Pauli praedictorum, illam dedit indulgentiam, quae datur euntibus ultra mare. Deo gratias. (Parente, I, 375-376, publishes from a late and obviously incorrect copy). n Parente, II, 425. 12 (indulgentias trium annorum cuicumque ex universitate civitatis Aversae, qui eleemosinam erogaverit ad reparationem ecclesiae civitatis ejusdem a terremotus concussione pro majori parte subversa funditus, et diruta). 13 II, 80. « Parente, II, 438. is Ibid., II, 425. i» Ibid., II, 428. vlbid., II, 429. is/biU, II, 439. 79 LOMBARD ARCHITECTURE lower level than those of the wall arches at their outer perimeter, and the transverse arches from capitals at a level higher than that of the capitals of the outer wall arches, while the arcade arches spring from the highest level of all. The vaults are not excessively domed, but the diagonals intersect far from the centre of the compartment, since they are straight in plan. They are somewhat distorted in elevation but not sufficiently so to bring the crowns to coincide with the point of intersection. Of the actual construction of the vault surface itself, the thick coating of intonaco makes it impossible to speak. A curious expedient is a wedge-shaped form given the transverse arches (Plate 17, Fig. 1), which are made much wider at the outside than at the inside edge. The excessively trapezoidal shape of the ambulatory compartment is thus somewhat reduced. The piers, of enormous solidity, are supplied with separate members for each of the ribs and the two orders of the arcade arches. The central cupola appears to be a Gothic structure of 1349, since it is adorned with an applique decoration of trilobed and pointed arches. Com- mendatore Rivoira states that it bears the traces of the fire of 1134, but if he saw such traces they have since disappeared. IV. The decoration of the church, like the structure, is sadly mutilated. Enough remains, however, to leave no doubt that in the cathedral of Aversa we have two distinct eras of construction.19 To the earlier belongs the northern portal (Plate 17, Fig. 3), the windows of the absidioles, a doorway in the south ambulatory (Plate 17, Fig. 2) and four windows of the central tower. All these fragments are marked by common characteristics, — hood-mouldings of very classic character usually adorned with egg-and-dart, bead or rope motives, the frequent use of consoles and brackets, spiral-fluted columns, capitals of distinctly Corinthianesque form, though executed with a certain stiffness and dryness that savours almost of the X century (Plate 17, Fig. 4), flat mouldings, and the use of marble. In this earlier edifice were also employed pilfered Roman capitals, one of which is now imbedded in the exterior of the north absidiole. This decoration, thoroughly Neapolitan in its character, is in such strong contrast to the style of the capitals used in the ambulatory, that it evidently must be pilfered material used second-hand in the new edifice. Thus the old portal was used to form a new northern entrance; old windows were utilized in the absidioles and central tower ; an old doorway was placed in the southern ambulatory, and even certain old capitals were used in the ambulatory. This old material, however, was pieced out with new material which shows an entirely different style of decoration. The new capitals of the ambulatory are Norman in character, though somewhat influenced, it is true, by Lombard and 19 According to Parente (II, 439), fragments belonging to the cathedral of Aversa are now preserved at Naples in the Biblioteca Borbonica, the Chiesa dell'Immacolata, etc. 80 AVERSA, CATHEDRAL southern Italian models. They are executed in stone, not marble. Some are very spare in their ornament, a characteristic which recalls the Norman style. Others have monsters coupled together with a single head which forms a volute, and below one or more rows of very stiff acanthus leaves (Plate 16, Fig. 2).20 Others are covered with scales. The arches opening into the absidioles are in two orders, and in two orders apparently were the arches of the choir arcade. The apses externally are adorned with arched corbel-tables. The bases are now for the most part hidden, but enough remains visible to make it clear that they were so deeply undercut as to have an almost Gothic character. Probably they were also supplied with griffes. V. The date of the earlier part of the church is determined approxi mately by the inscription on the portal. It was begun by Riccardo after he founded the episcopal see in 1053 and before he died in 1078. It was completed by his son, Giordano, before 1090. We therefore know that it was begun after 1053, that it was in construction in 1078 and was presumably finished before 1090. Indeed, the style is precisely such as we might expect to find in Campania at this epoch. This early church, like others of the same time and in the same region, was doubtless a wooden-roofed basilica. When, however, this edifice was burned in 1134, it was rebuilt, 1134-1160, in an entirely new style, not improbably by Apulian workmen. A rib-vaulted ambulatory was introduced, an extraordinary feature in Italy. The structure of the vaults themselves is nevertheless Italian rather than French, for precisely similar vaults are found in the slightly earlier church of S. Flaviano at Montefiaseone (Plate 151, Fig. 5). The capitals recall, for the most part, Norman models, although one of them is almost precisely similar to a capital 20 Rivoira has pointed out the strong technical resemblance of this capital of the ambulatory with one of the Badia of Venosa, in the province of Potenza. In view of the fact that Venosa, Acerenza and Aversa are almost unique among Italian churches in being supplied with an ambulatory and radiating chapels, there can be no doubt of the very close relationship of the three buildings. This relationship has been used by Rivoira as an argument for assigning Aversa to 1080, but his argument remains without force until the date of the abbey of Venosa is demonstrated. True it is that Venosa was founded by Robert Guiscard, who was there buried in 1085: [Robertus Wiscardi] sepultus est apud Venusiam in Coenobio Monachorum, quod ibidem ipse adhuc vivens construxerat. (Richardi Cluniacensis Chronicon, ed. Muratori, A. I. M. E., ed. A., XII, 109). But what proves that the existing church of La Trinita is the edifice erected by Robert and not a reconstruction of the XII century? Venturi (III, 504) inclines to the belief that such was the case. Schultz (I, 321-322) cites verbatim, and Lenormant (34) and Enlart refer to an extant inscription recording a consecration by the pope Nicolas II in 1159. The point can not be decided until an exhaustive study has been made, not only of the abbey of Venosa, but of the Romanesque architecture of the Basilicata, and as yet neither the one nor the other has been given serious attention. Notwithstanding the studies of Schultz (I, 317) and Lenormant (51 f.), the chronology of the cathedral of Acerenza remains likewise entirely uncertain. 81 LOMBARD ARCHITECTURE of the XII century in the Palazzo Arcivescovile of Verona. There is, there fore, good reason to believe that in the cathedral of Aversa we have a monument of 1134-1160, in which are used many fragments of an older building of the second half of the XI century. BADIA DI VERTEMATE,1 S. GIOVANNI (Plate 18, Fig. 1) I. The desecrated priory of S. Giovanni is situated about a kilometre from the commune of Vertemate, in the frazione known as Badia or Abbadia. The monument was known to Barelli,2 who published a plan, and to De Dartein, who studied and sumptuously illustrated it.3 Sant' Ambrogio has contributed observations on the history of the edifice,4 and historical notices of value may be found in the works of Giovio, Tatti and Giulini. II. The XVI century historian of Como, Giovio, has left us a long and detailed account of the foundation of the monastery, which he says was of the Cluniac order and had already been given in commendam long before his time. "It was founded by a certain Milanese, Gerardo, a noble, who, together with his friends Lanfranco and Amizone, had been received as a monk in the abbey of Cluny in France, by the abbot Hugo. In the course of time a certain prior of that order, whose name was David, head of the oratory of S. Paolo in Lombardy by permission of the abbot, Hugo, brought Gerardo back with him from France to Italy. Now it so happened that Gerardo, on his way to Milan, turned aside to Vertemate, where at that time lived certain noble knights by whom he was hospitably received as a guest. Gerardo noticed that to the east of Vertemarte there was a great solitary plain, and conceived the idea of founding there a monastery. Therefore he laid bare his project to his friends, and obtained from them a place to found the monastery, and immediately built amongst the brambles a little edifice supported on poles and covered with straw. Not long afterwards when he chanced to be wander ing about that solitude he found at the corner of a certain hill the ruins of an old castle and fortification. This place also was given to him by the same knights in the year of our Lord 1084, in the month of April, in order that a monastery of the Cluniac order and a basilica of S. Giovanni Battista might there be founded. To this Rainaldo, at that time bishop of Como, gave his formal consent, for that place was in his diocese. Soon afterwards Anselmo and Pietro, a subdeacon of the church of Como, joined Gerardo and aided in no mean fashion in the construction of the monastery. At the same time i (Como). 2 Not. Arch., 22. 3337. 4 Archivio Storico Lombardo, 1905, 217. 82 BADIA DI VERTEMATE, S. GIOVANNI Gerardo founded a convent for nuns at Cantu, in honour of S. Maria, and he summoned afterwards from Cluny his companion Amizone. Soon afterwards he died having chosen Pietro as his successor, a choice which was approved by the abbot of Cluny. The new church had not yet been dedicated because Pietro preferred to wait, inasmuch as many bishops were at that time infected with the taint of simony which was rife under the emperor Henry IV. At length the pope Urban II in the year of our Lord 1095 set out for France and held a council at Piacenza, in which there was much deliberation con cerning those who had bought ecclesiastical dignities, and those who had been ordained in the Gibertine schism. For Giberto, archbishop of Ravenna, was antipope under the name of Clement, having been created head of the schism against Gregory VII by the emperor Henry. When he left Piacenza, Urban came to Milan, and there Pietro, prior of Vertemate, obtained from the pope permission that Odone, bishop of Imola, might consecrate the church; and Odone dedicated with due rites the new basilica of S. Giovanni Battista at Vertemate, on the thirtieth day of December of the same year. There were present at the dedication a great number of clergy and laity of the city and diocese of Como. Landolfo of Carcano, who had recently usurped the episcopal throne of Como by a decree of the emperor Henry, was not present, however, since he had been anathematized by Pope Urban, after due trial. Beside this the pope granted privileges to the monastery of Vertemate in which the enemies of that institution are anathematized and indulgences granted to those who should aid and reverence the priory. When this monastery was later destroyed by the Comaschi, the prior Giorgio da Alzati commenced to restore it before the year 1480, but he died before he was able to finish the work; for that good man had planned to restore the abbey (which was almost deserted) to pros perity and to reform the discipline. Immediately after his death, however, it was given in commendam to a secular clerk who took no pains to finish that which had been begun."5 5 Monasterium S. Ioannis Baptists apud Vertemate Ordinis Cluniacensis, iam diu commendatum, a quodam Gerardo mediolanensi, nobili equitum genere nato, fundatum fuit, qui cum sociis Lafrancho et Amizone in gallico cluniacensi coenobio ab Ugone abbate in monachum receptus est. Procedente vero tempore, quidem eius Ordinis prior, cellae D. Pauli in Lombardia praefectus, cui nomen erat David, Gerardum e Gallia in Italiam reduxit, Ugone ipso abbate permittente. Forte vero Gerardus, cum medio- lanensem regionem peteret, Vertemate divertit, ubi turn nobilissimi quidam equites habitabant, a quibus hospitio benigne susceptus est. Porro Gerardus, animadvertens vertematense territorium ad orientem plagam vasta solitudine protendi, de condendo ibi monasterio cogitavit. Quare, consilio suo equitibus suis exposito, fundandi coenobii locum impetravit, ac statim inter vepres tuguriolum perticis suffultum et paleis opertum extruxit. Haud ita multo post, solitudinis illius cuncta perlustrans, in collis cuiusdam angulo veteris arcis, munitionisque vestigium reperit. Hunc itaque locum ab equitibus ipsis pariter accepit, anno Domini octogesimo quarto supra millesimum, mense aprili, ut ibidem Cluniacensis Ordinis monasterium cum basilica D. Ioannis Baptistae fundaretur, 83 LOMBARD ARCHITECTURE Hitherto nothing further has been known of the foundation of the priory of Vertemate beyond what is contained in this passage of Giovio. Tatti6 conjectured rightly that the historian wrote having under his eye authentic documents, but notes that the date for the foundation of the priory, April, 1084, is erroneous, since the bishop Rainaldo died in the preceding January. I am glad to be the first to point out that in the collection of the documents of Cluny published by Bernard is contained an incorrect copy of the deed of foundation which proves that Tatti's conjecture is correct, and that the foundation really took place in 1083. This document bears the date of the year of the Incarnation 1084, which corresponds to 1083. The seventh indiction, however, is erroneously given for the sixth.7 cui rei Rainaldus tunc comensis episcopus, quod hie locus in eius erat diocesi, assensum praestitit. Deinde Gerardo additus est Anselmus et Petrus comensis ecclesiae subdiaconus, qui novi monasterii fabricam haud medriociter adiuverunt. Per idem tempus Gerardus ipse muliebre coenobium apud Canturium honori D. Marias condidit, evocavitque post- modum e gallico cluniacensi monasterio contubernalem suum Amizonem, ac paulo post moriens praedictum Petrum successorem elegit, quod et cluniacensis abbas approbavit. Nondum autem nova dedicata erat ecclesia, quod, ut canonice fieret, idem Petrus solerter intendebat, quando simoniaca labe non pauci episcopi eo tempore infecti erant, Henrico IV imperatore sacerdotia venundante. Tunc Urbanus II pontifex maximus, vergente anno Domini nonagesimo quinto supra millesimum, in Galliam proficiscens, concilium apud Placentiam habuit, in quo magna consultatio facta est de his, qui sacerdotia emerant, quique in schismate Gibertino ordinati fuerant. Fuit autem Gibertus Ravennas archiepiscopus antipapa, nomine Clemens, quem praedictus Henricus contra Gregorium VII creaverat huius schismatis caput. Placentia discedens Urbanus Medio- lanum devenit, ubi Petrus ipse vertematensis prior a pontifice quendam Oddonem imolensem antistitem suscepit, qui Vertemate profectus novam basilicam D. Ioanni Baptistae de more dedicavit, tertio calendas ianuarias eiusdem anni. Interfuere dedica tion! innumeri civitatis et dioecesis comensis clerici ac laici. Abfuit unus Landulfus de Carchano, qui nuper comensem episcopatum sibi ab Henrico imperatore decretum invaserat, quem ideo pontifex Urbanus, audita causa, damnavit. Caeterum vertematense monasterium privilegiis munivit et ornavit, quibus in primis eius adversariis execrationem interminatus est; eidem vero obsequium praestantibus delictorum veniam dedit. Hoc coenobium, a Comensibus olim dirutum, novissimus prior Georgius de Alzate restituere inceperat, ante annum Domini millesimum quadragentesimum octuagesimum [MS. V. MCCCCXXX] ; sed, cum morte praeventus fuisset, tantum opus imperfectum mansit. Nam coenobium ipsum, monachis rarum, frequentem reddere et ad regularem observan- tiam reducere vir bonus cogitaverat, quod statim post eius obitum seculari clerico commendatum fuit, qui nihil minus quam inchoatum opus perficere euravit. (Giovio, 223-225). 6 II, 258. 7 Anno ab incarnatione Domini nostri Jesu Christi millesimo octogesimo quarto, mense aprili, inditione septima. Ecclesie et monasterio quod est constructum in honore beatissimorum Petri et Pauli, in loco qui dicitur Cluniacus, nos in Dei nomine Otto et Vuazo . . . offerimus predicte ecclesie Sancti Petri Cluniacensis, id est castrum unum juris nostri, quod est in predicto loco et fundo Vertemate, ad locum qui dicitur Castrum Vetus, cum propinquiore fossato usque in medium fundum ubi monasterium est constructum in honore Sancte Crucis [sic]. . . . (Bernard, IV, 765). 84 BADIA DI VERTEMATE, S. GIOVANNI The high reputation enjoyed by the new priory is witnessed by an undated letter assigned by Bernard8 to c. 1070, but which must, in reality, be later than 1083, in which Oberto, count of the Canevese, and Ardicio, baron of Castelletto, complain of the disorders occasioned by the base character of the prior of Castelletto, and request his removal and the substitution of Garnerio, prior of Vertemate. The abbey of Vertemate is included among the possessions confirmed to the abbey of Cluny by Urban II, in 1095,9 but, oddly enough, is not mentioned in a bull of the same pope of 1088.10 A prior of Vertemate appears in a Cluniac document of c. 1150, published by Bernard,11 and the priory is mentioned in an unpublished document of 1136.12 In the year 1125 the town of Vertemate was destroyed by the Comaschi. "While the Comaschi were returning to their city by the public highway the Vertematensi came upon them hurling javelins and insulting them with opprobrious epithets, and strove with all their might to prevent the Comaschi from passing through their territory. The entire army of the Comaschi noted all this in secret, but they said little, and bode their time to take vengeance with deeds not words, and resolved that the Vertematensi should pay the penalty and rue their act within a month. The latter kept on insulting and the former continued to mutter between their teeth. After a few days the Comaschi returned over the same road and canae to the same spot. There the Vertematense infantry was standing on the road, and hurled spears and javelins and shot arrows. Thus they prevented the Comaschi from passing. The Vertematense troops, armed in proof, prepare to fight hand to hand, sword against hard iron. The foot-soldiers on both sides began the fray. The Comaschi cavalry saw this, unfurled their banner, shouted, and rushed to the battle. They compelled their adversaries to give ground, and hurled many back on their fortifications. The Vertematensi then fled in fear and took refuge in the town near the castle. They fought from the town and sought to defend it. Then the Comaschi foot-soldiers overthrew the defenders of the gate. Then they sought fire and tried to burn the town, and they succeeded in kindling a conflagration. The entire castle and town were burned and many cattle perished. The Comaschi killed with the sword the foot-soldiers, the cavalry, the women, the strong, the weak, the young and the old together. But even more perished in the fire. Then the Comaschi cavalry were touched with pity at the death of so many, and kept circling about the walls of the castle, and drew out their enemies from the flames, and saved as well as they were able the Vertematensi themselves and their household possessions. One s IV, 540. 9Tomassetti, II, 158. io Ibid., 121. n V, 505. 12 Bonomi, Dip. Sti. Ben., Brera MS. AE, XV, 33, f . 34. 85 LOMBARD ARCHITECTURE hundred and twenty in all were killed by the sword but many more perished in the fire."13 It has been supposed by certain writers that the basilica of S. Giovanni was destroyed in this sack of Vertemate by the Comaschi. It is certain, however, that such could not have been the case. Not only is the architectural style of the church of the XI and not of the XII century, but the anonymous poet in his detailed description of the destruction of the town would not have failed to record any damage to the priory of S. Giovanni, had such been done. Moreover, it is expressly stated by Giovio that the abbey had been built on the site of a ruined castle some distance from the town of Vertemate, and to this day S. Giovanni lies a good kilometre from that commune, which there is no reason to suppose has ever changed location. In the deed of foundation 13 Dumque iter ad Cumas illis via publica donat, Inque revertentes simul adsunt Vertematenses, Jactantes jaculos, & turpia verba ferentes, Transitus hinc illis ne sit, pro posse laborant. Denotat hie illos clam tunc exercitus omnes, Pauca tamen dicunt, ad jurgia verba rimittunt: Poenam solvetis sed in isto mense gemetis. Tunc insultabant, illi post terga fremebant. Postque dies paucos semitam redeunt per eandem, Ad loca quae dudum fuerant disposita tendunt, Inque via pedites tunc stabant Vertematenses, Et jaciunt hastas, jaculos, funduntque sagittas: Sic iter impediunt, nequeunt transire volentes. Vertematensis miles protectus in armis, Cominus ense parat duro contendere ferro. Committunt bellurn pedites communiter omnes: Hsc equites cernunt, continu6 signa resolvunt, Voces emittunt, sic ad certamina tendunt, Obstantes pellunt, vallo pluresque revolvunt. Dant trepidi tergum, stant in villa prope castrum, De villa certant, illam defendere temptant. Tunc validi pedites sternunt in limine stantes, Inde petunt ignem, certant succendere villam, Acceduntque rogum; comburitur igne peremptum Castrum cum villa, nee non animalia multa. Et pedites, equitesque simul, pariter mulieres, Fortes, infirmi, juvenes simul, & seniores, Ferro mactantur, sed plures igne cremantur. Tunc equites flentes tantorum morte dolentes, Continu6 circiim discurrunt undique castrum, Ardentes retrahunt, illos & ab igne tuentur, Illis & vitam conservant, & suppellectilem. Centum viginti (sunt plures igne cremati) Sunt interfecti, sed sunt magis igne cremati. (Mediolanensium in Comenses Bellum, ed. Muratori, R. I. S., V, 439). 86 BADIA DI VERTEMATE, S. GIOVANNI the monastery is said to be not in the castle of Vertemate but in the old castle — Castrum Vetus. The abbey would therefore have been well out of harm's way when the town was burned. Even churches situated within the walls of cities were customarily spared in times of pillage, and in this case the quarrel of the Comaschi was with the Vertematensi and not with the monks. The same considerations give reason to believe that the monastery escaped also when the town of Vertemate was destroyed by Barbarossa in 1162,14 and by the Comaschi a second time, c. 1260.15 About 1287 it is true the abbey was destroyed, but it is explicitly stated by Giovio that the church itself escaped destruction. "At that time . . . the Comaschi completely destroyed the famous monastery of S. Giovanni Battista near Vertemate, except the basilica."16 About 1404 the town of Vertemate was again burned,17 but the monastery was not injured. In 1480, as has been seen, the abbey was given in commendam. The description of the church in the 'Acts' of the visits of the bishop Ninguarda is as follows: "On Monday, the thirteenth of July, the most reverend bishop of Como, pursuing his general visitation, came in person to the church of S. Giovanni Battista, called a priory, near the town of Vertemate. This church is preceded by an atrium which is almost entirely ruined, and, though ancient, the structure retains a certain air of elegance. Within there are three aisles, that is a nave and two side aisles, all in good condition. The side aisles have sixteen canopies with altars beneath them."18 In the Cluniac catalogue of Marrier, published in 1614, we read: "The priory of S. Giovanni of Vertemate is united with the priory of Cernobbio. According to the definition of 1367, there ought to be here a prior and six monks and alms are given to all those who ask them."10 14 Galvanei Flammae, Manipulus Florum, CLXXXVII, ed. Muratori, R. I. S., XI, 641. 15 Giovio, 48-49. 16 Per ea tempora Lutherio Ruscha comensis populi et Bonacursio e Vicedominis a Monticulo Communis pra3toribus, Comenses insigne D. Ioannis Baptistae coenobium apud Vertemate, Ordinis cluniacensis, salva basilica, funditus everterunt, quod eius loci monachi in eorum perniciem curn hostibus, superiore vigente bello, conspirassent. (Giovio, 55). 17 Giovio, 80. is Die Lunae 13 mens. lulij R.mus DD. Episcopus Comensis Visitationem generalem prossequendo, accessit personaliter ad Ecclesiam S.cti Io: Baptistae prope pagum Vertemati, prioratum nuncupatam. Quae habet atrium ante pene dirutum et redolet antiquitate elegantis structures; intus habet tres naves, mediam et duas laterales omnes convenientes, in quibus lateralibus sunt sexdecim fornices cum infradicendis altaribus. (Ninguarda, ed. Monti, II, 18). 19 Prioratus S. Ioannis de Vercemate, qui est vnitus Prioratui de Cernobio, vbi debent esse iuxta diffinitionem anni 1367. cum Priore sex Monachi, & fit eleemosyna omnibus petentibus. (Marrier, 1746). 87 LOMBARD ARCHITECTURE In 1621 the Cluniac monks were supplanted by Frati Minimi of the order of S. Francesco da Paola. The last of the commendatary abbots died in 1788, and soon after the monastery passed into secular hands. It is at present used as a barn. III. The church consisted originally of a nave four bays long (Plate 18, Fig. 1), non-projecting transepts, a choir of a single bay flanked by side aisles, and three apses, of which the central one was preceded by a barrel-vaulted compartment; but the southern side aisle of the choir and its absidiole have been destroyed, and the transepts have been walled off from the nave. The nave is at present covered with groin vaults supported on a clumsy alternate system (Plate 18, Fig. 1), but both vaults and system are modern, and the nave was without doubt originally roofed in timber. The cloistered vault that covers the crossing is also modern and supplants the original Lombard cupola. The slightly domed groin vaults of the side aisles are original, however, and are supplied with loaded transverse arches but have no wall ribs. Original, too, are the barrel vaults of the transepts, and of the choir. The piers of the crossing are cruciform in section, but those of the nave were originally all cylindrical (Plate 18, Fig. 1). The side-aisle responds are rectangular in section and comprise one or three members. The north wall is reinforced externally by vigorous buttresses. The masonry consists of stone ashlar, on the whole fairly well laid, although the stones vary extremely in size and the courses are frequently broken. The church has unfortunately been covered internally and externally with intonaco, which makes it extremely difficult to study the structure of the walls. IV. The basilica is characterized by the restraint, one might almost say the absence, of decoration. The piers of the nave are crowned by cubic capitals of fully developed type. The aisle responds and the imposts of the choir and transepts (Plate 18, Fig. 1) are crowned by simple impost mouldings. Most of the bases are at present not visible, but I believe that they were all of a similar character and consisted of a simple bulging torus surmounted by a fillet. The unmoulded archivolts are in a single order. The exterior, though much modernized, retains in the facade traces of its ancient cornice of arched corbel-tables. In the central apse the arched corbel- tables are well preserved and are supported on shafts. The edifice contains notable frescos of different ages. V. The priory of Vertemate is an authentically dated monument of 1083-1095. 88 BARDOLINO, S. SEVERO BARDOLINO,1 S. SEVERO (Plate 19, Fig. 4) I. The church of S. Severo at Bardolino has been mentioned by Melani,2 and has been studied by Cipolla. Crosatti has made a careful search for documents relating to the history of the edifice. II. Since S. Severo is mentioned in a diploma of Berengario of 893,3 Crosatti is doubtless correct in assuming that our monument was the early parish church of Bardolino. Cipolla tried to interpret a monogram on one of the columns as signifying: 0[BIIT] M[ENSE] AVGO? MCIX, but I confess that to me this reading seems very imaginative and the meaning of the monogram in question quite enigmatical.4 The church is mentioned for the second time in a document of 1186.5 In 13496 and in 13987 bequests were made to the laborerio for the con struction of the church. In 1415 and 1416 other legacies were left for the reparation of the church. It is strange, however, that the existing edifice shows no signs of alterations executed at this time. I therefore suspect that the contemplated reconstruction of S. Severo was never executed and that a new church (S. Nicolo) was erected instead. Confirmation is lent to this hypothesis by the fact that as early as 1447 the new church of S. Nicolo had supplanted S. Severo as the parish church of Bardolino.8 In 1530 S. Severo was abandoned, but was subsequently reopened for worship.9 In 1574 the edifice menaced ruin.10 About this time the church became the chapel of the cemetery. In 1750 the old apse was replaced by a new choir.11 In the XIX century the church was quite abandoned again, and was used as a powder- magazine by the Austrians. In 1869 it was in a ruinous condition, and in 1872 the campanile was mutilated.12 To-day the edifice serves as a concert hall for the band locally known as the Societa Filarmonica of Bardolino. III. The edifice consists of a nave (Plate 19, Fig. 4) five bays long, two side aisles, a modern choir and a modern campanile. Test excavations have, however, revealed the fact that there was anciently a crypt. The building is at present entirely roofed in timber (Plate 19, Fig. 4), but it is probable that the side aisles were originally vaulted. The southern side-aisle wall has been entirely reconstructed in modern times, but the north side- aisle wall, which is original, is divided into bays by colonnettes engaged on i (Verona). 2315. 3 Crosatti, 130. 4 Crosatti, 130-131. 5 ibid., 131. 6 . . . relinquo laborerio ecclesie sancti severij de bardolino decern soldos . . . (Ibid., 132). 7 . . . pro fabricatione ecclesie s. seuerij de bardol. (Ibid., 132). s Crosatti, 134. 9 ibid., 134, 135. lo/biU, 137. " Ibid., 139. 12 Ibid., 141-142. 89 LOMBARD ARCHITECTURE pilaster strips. The colonnettes end in inconsequential capitals which support nothing, but the pilaster strips are continued along the wall surface as blind arches. These arches, executed in plaster, are evidently a modern makeshift to disguise the amortizements of the vault. The supports of the nave are cylindrical piers (Plate 19, Fig. 4), the bases of which are now buried. The ancient windows were very small, widely splayed and intended to serve without glass. The existing windows are of the Renaissance, but one or two of the ancient ones are still extant in the north wall, although walled up. The masonry consists of unhewn stones of rectangular form, laid in courses approximately horizontal, and separated by thick mortar-beds. The masonry of the piers, on the other hand, is formed of roughly hewn stones separated at intervals by bands of brick. This masonry is thus crudely polychromatic. The piers have a very decided entasis, or rather, inward slope. IV. The capitals of the responds of the north side aisle are of a high cubic variety, with chamfered edges. The capitals of the main piers (Plate 19, Fig. 4) are also cubic, but the proportions are low. The angular cushion does not recede, that is, the piers have the same diameter as their load, and the office of the capital is merely to form a transition from the cylindrical support to the rectangular archivolt. On the abaci of these capitals are carved, or rather crudely scratched, rosettes, clover leaves, interlaces, zigzags, a head, a mono gram, and similar motives. All these ornaments and the capitals themselves recall Stradella (Plate 210). The archivolts are in a single unmoulded order (Plate 19, Fig. 4). The exterior is decorated with arched corbel-tables resting on pilaster strips only at the angles. The expanse of the wall is unbroken even by buttresses. V. The capitals of the main arcade recall those of Stradella, which date from c. 1035. Those of the north side-aisle responds, however, are of a somewhat more advanced character. Also the character of the masonry and of the arched corbel-tables indicates that the edifice was constructed in the second, rather than in the first, half of the XI century. We may, therefore, assign it to c. 1050. BARDOLINO,1 S. ZENO (Plate 19, Fig. 1, 3) I. Crosatti has made a conscientious study of the local archives for documents referring to this church, which has also been described by Cipolla. II. According to Biancolini,2 the church of S. Zeno at Bardolino was given to the abbey of S. Zeno of Verona by Pepin by a diploma of 807. The i (Verona). 21, 44. 90 BARDOLINO, S. ZENO diploma itself is lost, and we have knowledge of it only from Biancolini's reference. That the notice is authentic, however, is confirmed by the fact that Bardolino is confirmed to S. Zeno in a diploma of 8473 and in others of 1014,* 1027, 1163 and 1186.5 In many other documents as well our church appears as a dependency of the Veronese monastery. In 1529 the church of S. Zeno at Bardolino was abandoned and in a ruinous condition.6 In 1541 it was desecrated. In 1697 it was reopened for worship, and a radical restoration undertaken.7 At the time of the Revolution the edifice was again desecrated, but was restored and consecrated anew in 1863.8 III. The edifice consists of a single-aisled nave with projecting transepts and a square apse. Over the crossing rises a lantern covered with an undomed groin vault. The rest of the edifice is spanned by barrel vaults. The walls are covered internally and externally with plaster and intonaco, so it is impossible to see the masonry except in a few scattered spots. To judge from these the church is built of rubble of many different qualities, and it is probable that the walls and vaults also have been many times made over and repaired in the various restorations to which the church has been subjected. It appears, however, that the original Carlovingian dispositions are still, in the main, preserved. IV. The nave walls are decorated internally with two blind arches on either side, supported on two free-standing columns and corbels in the west wall. In the angle of the apse and transepts are inserted two free-standing columns. The original capitals are of a thoroughly Carlovingian type. One is an imitation of the Ionic, with angle volutes (Plate 19, Fig. 3). It has a bead- moulding on the necking and eggs and darts on the echinus. The abacus is surmounted by a high stilt-block, decorated with a peculiar crocket-like ornament. Two capitals (Plate 19, Fig. 1) are of a Corinthianesque type, with a single row of thick, stiffly carved leaves, except under the volutes, where an extra leaf is inserted. The veins on the leaves are indicated by incised lines which, however, are scratched rather deeply. On the centre of each face is a rosette, and on the front of the stilt-block are two rosettes and a Greek cross. The volutes are rather crudely executed, and have a sagging curve, or else are continued to the lower row of leaves. On certain faces the stilt-blocks are ornamented with the same crocket-like motive as the Ionic capital. The fourth capital is pilfered Roman Ionic, with carved angle volutes, and seems to have been the model from which the first capital was copied. The bases s Crosatti, 166. ¦a.Historia di Verona, MS. of 1587-1597, No. 1968/Storia/90.5 of the Biblioteca Comunale of Verona, Libro Sesto, f. 13, sotto anno 1014. See also Biancolini, I, 48. s Crosatti, 166-167. 6 ibid., 171. 7 Ibid., 172, 173, 287. s ibid., 174. 91 LOMBARD ARCHITECTURE show considerable variation, but approach more or less closely to the Roman Doric type, with a single torus and a plinth. The exterior of the church is absolutely without ornament. V. The capital (Plate 19, Fig. 1) shows strong analogies with the capital of S. Satiro at Milan (Plate 132, Fig. 5), an authentic monument of 875, in the carving of the Greek cross and in the technique of the leaves and volutes. Moreover, the plan of S. Zeno at Bardolino is analogous to that of S. Satiro, in that it consists of a central vaulted area with four arms, also vaulted. This type of plan is exceedingly common in IX century churches and therefore in itself is not sufficient to fix with precision the date of our monument. When taken, however, in connection with the -capital, the analogy to S. Satiro at Milan is so striking that we may assign this edifice to the same date, c. 875. BARDONE,1 S. MARIA ASSUNTA I. The little church of Bardone, hidden away in the wilds of the Apennines, lies a half hour's walk from the highway. The sculptures have been described and in part illustrated by Venturi.2 II. According to a memorial composed by the late priest and now preserved in the sacristy, the church is mentioned in a document of 1004, and has always enjoyed the rank of a pieve. A vague tradition3 that the basilica was reconstructed by the countess Matilda appears to be confirmed by no trustworthy evidence. On one of the piers is a painted inscription of 1514, referring doubtless to frescos executed at that epoch. III. The edifice has been entirely rebuilt in the Renaissance period, and of the old Romanesque edifice there remain visible only two piers and numerous fragments of sculpture. IV. In a store-room to the north of the church, near the stairway which formerly led to the houses of the canons, there are several sculptured frag ments, — a St. Peter dressed in episcopal robes and holding the keys, a nice head, probably of a caryatid figure, and several capitals of the XIII century. The altar of S. Antonio, the second from the west on the north side of the church, contains in its principal face a carved Romanesque slab repre senting Christ in glory. In the centre is Christ with an inscribed halo and an aureole in the form of the figure eight, supported by the symbols of the four Evangelists. To the right of Christ stands a female figure with hands i Frazione di Lesignano Palma (Parma). 2 III, 134, Fig. 114, 116. 3 Molossi, 13. 92 BARDONE, S. MARIA ASSUNTA raised in adoration, doubtless the Virgin. About are seven angels flying or standing, holding scrolls, candles, censers. One hands a crown to Christ. Opposite this altar, in another, dedicated to the Angelo Custode, there is a companion relief, representing the Deposition. Christ's right hand has been loosened from the cross, and is held by the Virgin. Nicodemus puts his arms around Christ's waist to support the body, while Joseph of Arimathea, on a ladder, pulls the nail out of the left-hand palm with a pair of pincers. Behind the Virgin stand the two other Marys ; an angel flies above their heads. Beyond Nicodemus an angel with bared sword drives Adam and Eve, naked except for fig-leaves, from the Garden of Eden. The holy-water basin is supported by a female caryatid with finely folded drapery. Over the northern portal is a weakly composed lunette showing a seated Madonna with Child, and a kneeling, beardless saint with a book, probably St. John. At the west portal are two caryatids and two lions, which formerly supported the columns of the Lombard porch. The lions hold between their paws animals the species of which it is now impossible to determine. Over a gateway east of the church, leading to the canonica, are grotesque figures, representing a hunting scene. This arch, cusped, crocketed and finely moulded, is evidently of the XIV century, although it has been taken by Venturi to be contemporary with the other sculptures. The latter, with the exception of the lunette of the portal, a contemporary, but far inferior work, are all by the same hand, and that hand must have been that of a local sculptor, strongly under the influence of Benedetto. The style is coarse and crude, and possesses something of the vigour and roughness of the mountain region in which the church is placed. Numerous details prove the influence of Benedetto, and especially of the works of his earliest period: the angels flying in a horizontal position in the relief of Christ in glory, recall the similar figures in the Deposition of the cathedral of Parma (Plate 165, Fig. 4) and in the northern lunette of the baptistery (Plate 164, Fig. 1). The figure of the angel in the lower plane to the right of Christ is inclined in a straight, oblique line, precisely like that of the Church in the Parma Deposition. The very subject of the Deposition in the other altar front recalls the plaque of the Parma cathedral, and the details of the iconography are not only identical in the two compositions, but also unique in the field of Lombard plastic art. The cross in the Bardone plaque is covered with little knots precisely as in the Parma composition. The rosettes at Bardone around the edges of the relief are evidently merely crude copies of Benedetto's originals. Numerous other details, such as the treatment of the hair, the drapery and the composition, complete the proof that the Bardone sculptor closely imitated Benedetto. V. The fact that the sculptures at Bardone show very strongly the influence of the Deposition of Benedetto executed in 1178, and but faintly, 93 LOMBARD ARCHITECTURE if at all, that of the same sculptor's later works at the baptistery of Parma — all of which are later than 1196 — gives reason to believe that the Bardone works were executed at a time when the more mature works of Benedetto had not yet become generally known. The Bardone sculptures may, therefore, be assigned to the very last years of the XII century, or to c. 1200. BARZANO,1 S. SALVATORE (Plate 19, Fig. 2; Plate 20, Fig. 1, 2; Plate 21, Fig. 1, 2) I. The unpretending little church of S. Salvatore, commonly called Chiesetta della Canonica, at Barzano, has frequently been referred to by writers on Lombard antiquities. The indefatigable Mella2 illustrated the portal. Barelli3 studied the architecture, which has also been called to notice by Malvezzi,4 Monti5 and Melani. The little monograph of Mantovani contains, amid many inaccuracies, some notices of great value, and the same may be said of the historical compilation of Dozio. Of all the authors who have written on the church, however, not one has appreciated its archaeological significance. II. According to Mantovani6 the church is nothing less than the temple which Novelliano, according to an inscription, erected to all the pagan gods and goddesses. This Novelliano, he believes, lived at the end of the IV cen tury. About the year 700 (always according to our author), the pagan shrine was remade as a Christian church. A mere inspection of the building, however, suffices to show that there is in the present structure nothing Roman. That the church was built by the early Lombard kings, on the other hand, and more particularly by the famous queen, Teodolinda (who lived about 590), is a constant local tradition at Barzano.7 Such traditions are common in Lombardy, and especially in Brianza, and this would merit no faith were it not for the circumstance that in this case the tradition is in some degree confirmed by the style of the monument. The villa of Barzano is mentioned in a document of October, 1015,8 but nothing is said of the church. At any event, it is certain that the latter, after the XIII century, enjoyed the rank of pieve.0 In fact, the existing edifice was merely the baptistery of a basilica which has disappeared. This is clear from the existence of a baptismal font at least as old as the XIII century, still preserved in the centre of the nave. I conjecture that S. Salvatore was originally erected, not as a baptistery, but as a church; that at the end of the i (Como). 2 Elementi, T. V. s Not. Arch., 39. 4 6. 5 483. 67. 7 See Barelli; Malvezzi, 6, etc. 8 Dozio, 55. 9 Chronicon Mediolanense, ed. Cinquini, 17. 94 BARZANO, S. SALVATORE XII century, a new and more imposing structure was built for the use of the lately established canons, and that in the XIII century the old church was remade as a baptistery. An inscription, formerly near the church, but which has now disappeared, and is preserved only in a copy of Bombognini,10 states that the basilica was erected by Galdino Pirovano, archbishop of Milan.11 This inscription, however, offers some difficulty, because S. Galdino, archbishop of Milan from 1166 to 1176, did not belong to the Pirovano family as did his predecessor Oberto (1146-1166) and his successor Algisio (1176-1185). It therefore seems probable that Bombognini's copy is incorrect. There are grounds for believing that the church and castle belonged to the Pirovano family, and that the chapter, consisting of a prevosto and five canons, was established by the arch bishop Algisio Pirovano.12 I therefore conjecture that it was not S. Galdino but Algisio Pirovano who constructed the church c. 1180, and that the church he constructed was not the existing church of S. Salvatore, but the basilica served by a chapter of five canons, for which S. Salvatore was merely a baptistery. On the archivolt of the portal, beneath the gable, may still be seen the remains of an inscription painted in Gothic letters, all but a few of which are now illegible, but the first part of which, in Barelli's time, could still be read, and was thus transcribed by that author: Anno dominice incarnationis millesimo ducentesimo [?] trigesimo primo On the keystone is the incised inscription: Q[VI] FECIT HOC OPUS APELLATUR SERIN PETRVS Even in the time of Barelli the date was not easily decipherable, and that archaeologist expresses considerable doubt as to the exactness of his reading. Furthermore, the inscription seen by Barelli was not the original inscription, but a copy of it made in 1611, at the order of Federigo Borromeo, to replace the original inscription, which had become in part effaced.13 We have, there fore, a copy of a copy of the original inscription. The inexact manner in which mediaeval inscriptions were renovated in the XVII century is so notorious that it is needless to insist upon the fact that no faith can be placed in them. However, the main facts recorded in this inscription offer no difficulty. The style of the portal, as indeed that of the entire western bay of the nave, is io 168. n GALDINVS . PIROVANVS . ARCHIE . MEDIOL . B ASILICAM . HANC . CONSTRVXIT . 12 Mantovani, 26. is Litterae ostio ecclesiae inscriptae, quibus tempus constructionis ecclesiae demon- stratur, renoventur qua parte corrosae sunt, nihil immutata figura sive charactere. (Acts of Pastoral Visit of 1611, cit. Mantovani, 15). 95 LOMBARD ARCHITECTURE that of the second quarter of the XIII century. Moreover, we have seen that there is reason to believe that a chapter had been erected and the main basilica rebuilt in the last years of the XII century. It is therefore natural that the restoration of the baptistery should be next undertaken. Finally, the castle of Barzano was ruined, probably in 1222, an event which may well have necessitated a restoration of the ecclesiastical buildings.14 When S. Carlo Borromeo visited the church in 1583, the basilica had already been abandoned, since he prescribed that the baptismal font should be taken away from S. Salvatore.15 This order, however, appears never to have been carried out. In 1611 the archbishop Federigo Borromeo, in his pastoral visit, found the church abandoned and in ruin, collapso etiam tecto, and ordered that it should be restored in its original form. The sacristy and ossario (Plate 20, Fig. 2), at a lower level, were added at this epoch. Subsequently the ossario was converted into a second sacristy. In 1858 the church was restored. In the course of the works the break in the masonry between the newer western bay and the older eastern parts of the church was laid bare.16 III. The edifice consists of three distinct parts or bays (Plate 20, Fig. 1) ; the easternmost, a sort of square apse, is covered with a barrel vault; the next, half occupied by the nave, half by the choir, is surmounted, not by a cloistered vault, but by a true dome carried on arched squinches (Plate 19, Fig. 2). This dome is supported on arches in the wall, as in Byzantine architecture, the arches being filled in by thin screen walls, leaving deep reveals (Plate 19, Fig. 2). The westernmost bay has a timber roof (Plate 20, Fig. 1). In the middle of the nave is an octagonal immersion font, not sunk below the level of the floor but approached by a step and in turn depressed two steps (Plate 20, Fig. 2). Beneath the choir extends the crypt (Plate>l, Fig. 1), which is at present somewhat irregularly divided into a series of compartments covered with barrel vaults. The eastern of these compartments is covered by a transverse barrel vault extending the whole width of the church, with axis perpendicular to that of the edifice. The western part of the crypt is covered by two barrel vaults parallel to each other and to the church, but perpendicular to the axis of the barrel vault of the eastern part. Originally the crypt and the choir occupied only the eastern bay of the church, but they have been prolonged in modern times to occupy also half of the centre bay, and this portion of the crypt is not vaulted but covered with a wooden roof. The walls, constructed of coarse rubble, are enormously heavy, averaging about 1.75 metre in thickness. IV. The church is singularly destitute of ornament. It contains two ancient Roman pedestals, one of which is placed in the nave and serves as a holy-water basin, the other in the crypt. Beneath the intonaco with which 14 Mantovani, 18. is Ibid. io Mantovani. 96 BEDERO VALTRAVAGLIA, S. VITTORE the walls are covered are visible many traces of frescos. The font and the doorway are of the XIII century. V. It is evident that notwithstanding later restorations the body of the edifice is much older than the year 1231, when the portal and font were added. Indeed, the vast thickness of the walls, the complete absence of decoration, and the use of a dome make it certain that S. Salvatore is not only anterior to the year 1000, but one of the very earliest monuments extant in Lombardy. The structure of the building shows close relationship with the Byzantine style in the use of heavy relieving arches filled in by screen walls, in the spherical dome, and in the arched squinches. Moreover, the very plan of the edifice, departing widely from the basilican and circular types, recalls edifices of Greece or Constantinople, and their imitations of the V century in the Occident, such as the basilica of Fausta at Milan. There is extant in Lombardy no edifice erected after the VI century in which there is a dome carried on squinches and supported on arches which are closed by screen walls. This construction is thoroughly Byzantine. On the other hand, the crudeness of the masonry and the enormous thickness of the wall force us to recognize in the church at Barzano an edifice erected in the period of the greatest decadence in the arts, when the technique of construction had sunk to its lowest depths. The crudeness of this construction compared, for example, with the neat brickwork of the basilica of Fausta, prove that Barzano must be much later than the latter edifice. We have therefore in this church architectural forms familiar in edifices of the V century combined with the technique of construction that savours of the VII. These facts lead me to assign S. Salvatore of Barzano to the VI century, and more precisely to the end of that century. Since there is no other monument of this period extant, it would be exceedingly difficult to fix the date more exactly, were it not for the tradition that the church was erected by Queen Teodolinda. This tradition is not unworthy of belief when it is confirmed by the style of the architecture, and I therefore assign this edifice to c. 590. If this ascription be correct, S. Salvatore must be considered the only extant monument in Lombardy of the style of architecture used during the domination of the early Lombard kings. BEDERO VALTRAVAGLIA,1 S. VITTORE I. The basilica of S. Vittore at Bedero has been described by Barelli2 and by Monti.3 II. Nothing is known of the history of this church. Since the XVI century it has enjoyed the rank of pieve, this dignity having been transferred i (Como). 2 N0t. Arch. 3 483. 97 LOMBARD ARCHITECTURE from the neighbouring church of Domo, where the old baptistery still exists. It is probable that at about this time the edifice was baroccoized. In the latter part of the XIX century a restoration in the taste of the epoch was carried out. III. The edifice consisted originally of a nave six bays long, two side aisles and three apses, but the eastern and western bays of the side aisles have been walled off. The existing vaults are modern, and the edifice was originally covered with timber. The piers, which are rectangular, without bases, and with simple impost mouldings, are original, as is evident in the walled-off portions of the side aisles of the choir. The only really well preserved parts of the edifice are the side-aisle walls and the apses. The former are constructed of ashlar, consisting of fine large blocks which, however, are not brought to a smooth surface, but are laid in courses, frequently broken and often deviating from the horizontal. The mortar-beds, of moderate thickness, have been smudged over with modern plaster, and the ancient windows, widely splayed and with arcuated lintels, were intended to serve without glass. Two are still extant in the southern side-aisle wall, but have been walled up. The existing square windows are modern. The wall of the clearstory, unlike that of the side aisles, is built of smooth stones laid in courses roughly horizontal. Two of the old windows extant on the southern side are like those already described, but have true arches. Most of the windows, however, are great half lunettes of the Renaissance. IV. The interior of the church, restored in the worst possible taste, is of interest chiefly for the XV century frescos of the apse. The modern facade is a masterpiece of ugliness. The clearstory wall has at present no arched corbel-tables, and those of the side-aisle walls have been entirely remade in the recent restoration. Those of the apse, on the other hand, are original and supported on shafts. V. The apse cornice, analogous to that of the baptistery of Arsago, gives sufficient reason for assigning the Romanesque portions of the edifice to c. 1130. BELLAGIO,1 S. GIACOMO (Plate 22, Fig. 1, 2) I. The church of S. Giacomo of Bellagio, as is natural in the case of a striking monument of mediaeval art situated in a great tourist centre, has been frequently referred to by various authors. Barelli2 was the first to call atten tion to it, and in recent years have appeared, in addition to many notices of slight archaeological importance, monographs by Grandi and Perrone. i (Como). 2 Not. Arch., 21. 98 BELLAGIO, S. GIACOMO II. Of the history of the church in the mediaeval period nothing is known. The earliest notice in regard to it is a description of the XVI century made by the bishop Ninguarda,3 which, although it furnishes us with no data for the early history of the church, at least gives an accurate account of the condition of the building in 1593. "Visitata la chiesa di santo Iacomo nel borgo di belasio membro dell'arcipretato, et lontana piu d'un miglio. £ fatta in tre navi, ma non ci e volta alcuna, se non alle capelle maggiore et laterali in fronte. . . . Vi si ascende per andare alia capella magiore et altri doi altari otto gradi. . . . Ha due porte, una nella nave di mezzo nel frontispicio, et l'altra nella nave laterale dalla parte dell'epistola. . . . Ha campanile con due campane et orologio." The edifice did not long maintain its mediaeval forms as described by Ninguarda. In 1628 the duke Ercole Sfrondrato commenced the baroccoizing of the edifice by tearing down the two upper stories of the campanile which he planned to replace by a loftier structure in the barocco style. He also covered the lower part of the facade with intonaco, but did not live to complete the new campanile, which was finished by his grand-nephew, Giuseppe Valeriano. In 1657 S. Giacomo was raised to the rank of parrocchia prepositurale, a circumstance which seems to have provoked a complete restoration of the edifice in the style of the times. The wooden roof was replaced by a heavy vault. New barocco windows were opened — two circular ones in the sides of the choir, rectangular ones in the side wall, and a great lunette in the facade.4 The pavement of the church was raised, hiding the bases of the columns. Finally, the walls and columns were covered with a thick coating of intonaco.5 In 1690 the ancient apse was destroyed together with the ancient sacristy, and a new choir, much larger than the old one, erected.6 About the same time the facade was remade and a new central portal erected. In 1721 the chapels of the Vergine delle Grazie and of the Addolorata were added. These chapels were placed on ground which had formerly belonged to the cemetery. The disinterred bones were gathered together and placed in a chapel built to receive them alongside of the campanile. Throughout the XVIII and most of the XIX century, restorations continued to be executed in the church, always to the detriment of the mediaeval architecture.7 In 1884 the choir was repaved and various minor works executed.8 In 1902 the project of completely restoring the church in the mediaeval style began to be agitated; this restoration, carried out under the direction of the architect Luigi Perrone, was completed in 1907. The barocco intonaco was stripped from the walls, the barocco vaults destroyed, a new roof added, the ancient windows reopened, and the XVIII century ones closed. In addition the barocco choir was torn down and replaced by a new apse in the s A tti di Visita, ed. Monti, II, 115. 4 Grandi, 82, 88-89. 5 76^., 90. 6 Ibid. 7 ibid., 95-96. s ibid., 96. 99 LOMBARD ARCHITECTURE Lombard style. Fragments of the ancient ambo were removed from the campanile where they had been placed in the XVIII century, and with them was remade a new ambo. This ambo is supported by four capitals which had formerly been employed upside down as supports for the barocco baldacchino. It was a disappointment that no ancient frescos were discovered under the barocco intonaco, and the absence of such old paintings is scarcely compensated for by the modern frescos added in the restoration, although the latter, however unsatisfactory, are still not as bad as those that have been generally added to Lombard edifices in recent restorations. In 1908 the south wall was restored. It is at present planned to "restore" the campanile by replacing the barocco upper stories with a structure of pseudo-Lombard design. It is a great pity that the restorers should find it necessary to destroy the barocco campanile, a work of distinct architectural merit, and still more a pity that they should think of substituting for it a colourless structure which can only very dimly suggest the style of the XII century. III. The edifice consists of a nave four bays long, two side aisles, a highly raised choir of one bay flanked by side aisles, and three apses. The nave and side aisles are roofed in wood, but the choir and its side aisles are covered by groin vaults, and the three apses by half domes. There is no crypt, but since the church is placed on the side of a hill, the elevation of the choir corresponds to the natural configuration of the land, and the pavement of the nave slopes sharply towards the west. The nave at present has a clearstory of square windows, which, however, are not the original ones. The soffits of the choir vault have been remade, but their structure seems to be ancient. The vault of the nave is approximately square in plan, while those of the side aisles are distinctly oblong. All are slightly domed and are supplied with wall ribs. The piers separating nave and choir are cruciform, the groins and wall ribs of the choir vaults being carried on corbels. The other piers are all cylindrical. The choir vaults are reinforced externally by vigorous buttresses. IV. The capitals of the nave are all of cubic type, but are singular in that, instead of being formed of a single block of stone, they are constructed of small masonry like the piers themselves. They are unusually shallow, but supplied with a necking. The Attic bases without griffes are also worked in masonry. The piers of the choir have no capitals other than a simple impost moulding. The archivolts are of a single order unmoulded. The exterior of the church is severely plain. An arched corbel-table adorns the gable of the west facade, but the walls of the nave are without any decoration. Of the three apses the two lateral ones are ancient: that to the south is adorned with arched corbel-tables in two orders supported on pilaster strips with engaged shafts ; that to the north has plain corbel-tables ; both have a saw-tooth cornice. The central apse has been reconstructed on traces of the 100 BERCETO, S. REMIGIO (S. ABONDIO) original one which came to light during the restoration. The arched corbel- tables in two orders are grouped two and two and supported on pilaster strips. The windows of the apses, as indeed of all the church, were widely splayed and narrow, being evidently intended to serve without glass. The sculptures of the ambo (Plate 22, Fig. 1, 2) are among the most notable examples of the plastic art of the XII century extant in the diocese of Como. The symbols of the four Evangelists are sculptured in white marble, with a finesse which we look for in vain in other contemporary sculptures of this region. Yet the crude drawing betrays an inexperienced age, and these sculptures are not later than those of S. Fedele at Como. The four capitals of the present ambo are of a curious type which approaches the Corin thianesque, and may well be approximately contemporary with the sculptures. V. The masonry of the church, consisting of small but well laid blocks of Moltrasian stone, seems about contemporary with that of S. Abondio at Como, as do also the cubic capitals. We may therefore assign the monument to c. 1095. The sculptures of the ambo, however, not earlier than the portal of S. Fedele, must be somewhat later, or of c. 1115. BERCETO,1 S. REMIGIO (S. ABONDIO) (Plate 22, Fig. 3, 4) I. The town of Berceto lies in one of the wildest portions of the Appennini Parmigiani, near the summit of the Bardone pass which leads from Parma to Tuscany. In mediaeval times, this route used to be one of the most travelled between southern Italy and northern Europe. In modern times, however, the Bardone has been supplanted by easier or more direct passes, and the town of Berceto, with its interesting church, has remained almost unknown. For historical notices, the works of Affo and Molossi should be consulted. II. A monastery was founded in Berceto by Luitprando (712-743). The best source for this fact is the brief statement of Paolo Diacono.2 The notice of the chronicler is confirmed by the epitaph of Luitprando formerly preserved at S. Pietro in Ciel d'Oro at Pavia, which contains a distinct refer ence to the foundation of the monastery of Berceto,3 and by the legend of S. Moderanno narrated by Flodoard of Reims. According to the latter, in the time of Chilperic II (fl. c. 716), king of Neustria, S. Moderanno, bishop i (Parma). 2 In summa quoque Bardonis Alpe monasterium quod Bercetum dicitur aedificavit [Liuprandus]. (Pauli Diaconi, Hist. Long., VI, 58, ed. Waitz, 240). 3 See below, Vol. Ill, under Pavia, S. Pietro in Ciel d'Oro, Section II. 101 LOMBARD ARCHITECTURE of Rennes, set out on a pilgrimage to St. Peter's in Rome, but turned aside upon his way to visit the monastery of St.-Remi at Reims. There he obtained from the monks certain relics of the saint, which he carried away with him on his journey to Italy. While he was crossing the Bardone pass, he slept on a certain night by the wayside and hung the relics on the branch of a tree. In the morning, on resuming his journey, he forgot the relics, which he remembered only after he had proceeded some distance. He sent back a clerk to fetch them, but the messenger was unable to accomplish his mission, since, just as he was about to grasp the relics, they were miraculously lifted aloft. The bishop, when he heard of this miracle, returned and pitched his tent at the spot, nor was he able to lay hold of the relics until he had vowed to leave a part of them in the neighbouring monastery of S. Abondio at Berceto. Having done this, he resumed his journey. He was soon met by Luitprando, king of Italy, who, having heard of the miracle, immediately understood the virtue of the relics, and moved by love of St. Remi, gave to him the monastery of Berceto, with all its possessions. S. Moderanno returned from Rome and went again to the venerable tomb of St. Remi and laid at the feet of the saint the donation which he had received from the king. Then, having returned again to his own city, he appointed his successor, and, bidding farewell to his flock, returned to the monastery of Berceto, where he finished his days in the year 730. 4 The legend as told by Flodoard does not explicitly state 4 Temporibus Chilperici Francorum regis extitisse fertur Moderamnus Redonensis ecclesiae presul, vir nobili prosapia oriundus. Qui per licentiam predicti regis limina sancti Petri adire disponens, divertit in monasterium beati Remigii situm in suburbio Remensis urbis. Ubi liberaliter a fratribus eiusdem loci susceptus, impetravit a Bernehardo sacrorum custode reliquias de stola, cilicio atque sudario sancti Remigii. Quibus gratanter acceptis, iter inceptum laetus agens, dum permeat Italiam, in monte Bardonum quadam nocte metatum habens, memoratas in ilicis ramo suspendit reliquias. Cumque diluculo surgens iter coeptum arriperet inmemor horum, nutu, ceu creditur, divino haec ibidem remansere pignera. Procedente vero aliquanto longius episcopo, ubi relictarum memor fit reliquiarum, suum statim ad has recipiendum dirigit clericum nomine Vulfadum. Quo ad has perveniente, nullo valet eas ingenio contingere, dum mirabili signo, ut eas attingere vellet, elevarentur in sublime. Hoc prefatus episcopus audito miraculo regrediens, in eodem loco fixit tentorium; sed relicta pignera eadem nocte minime valuit recipere, donee facto mane in monasterio quod vocatur Bercetum, in honore sancti Abundii martiris inibi constructum, missam celebrans, predictorum partem munerum devoveret ibidem se relicturum. Sicque rapta sibi recipiens, impleto venerabiliter voto, coeptum repetit iter. Cui obvius factus Liutbrandus Italorum rex strenuus, qui hanc auditu iam compererat sacrorum virtutem, amore beati Remigii ductus, idem monasterium, Bercetum scilicet, cum omnibus adiacentiis omnique abbatia, mansos octingentos, ut tradunt, continenti, prefato presuli Moderamno delegavit eique in presentia fidelium suorum legali de more vestituram ex ea et cartam fecit. Remeans autem ab urbe Roma memoratus presul, accessit ad venerandum beati Remigii sepul- chrum, atque, sicut illi premissus rex hanc terram tradidit, ita nihilominus ille sancto Remigio eandem contulit. Sicque prospere in suum reversus episcopium, successorem sibi ordinari fecit, et valefaciens filiis suis, Bercetum monasterium repetiit et usque ad 102 BERCETO, S. REMIGIO (S. ABONDIO) that the monastery had been founded by Luitprando, but this is perhaps implied, since otherwise the monarch would not have had the power to place the abbey under the jurisdiction of St.-Remi of Reims. That the monastery was erected by Luitprando is confirmed finally by a diploma of Ugo, of 927. 6 According to the legend of S. Abondio published by the Bollandists, in the time of the emperors Lodovico II and Lotario I (849-855), Tiberio, abbot of Berceto, enlarged the church of the monastery and prepared a place under the principal altar for the body of S. Moderanno, which hitherto had been buried to the left of the altar. The saint, however, appeared to the abbot in a dream and warned him that the new tomb was destined for Abondio. Years afterwards the abbot chanced to go to a council in Pavia, where he learned that the body of S. Abondio rested at Foligno. The pious abbot immediately went thither and obtained by prayers the gift of the body, which he translated to Berceto in the year 850.6 This legend says nothing about the collapse of the mountain and the consequent removal of the monastery about this time, but implies, on the contrary, that the church remained always on the same spot, and was merely enlarged. The tradition of the collapse of the mountain is, however, constant among the historians of Parma.7 Affo8 has conjectured that in the time of Pope Benedict III (855-858) the church was officiated no longer by monks but by canons, and that this change of clergy was occasioned by the disaster of c. 850. Since a diploma of Ugo of 927 mentions that the church was officiated by canons, and yet calls it a monastery, no disproof of Affo's hypothesis is furnished by the fact that the church continued still to be known as an abbey, and as such was given to the bishop of Parma by Carlomanno in 879,° and that in 922 Rodolfo, king of Italy, confirmed to the bishop of Parma: abbatiam de Berceto, in honore obitus sui diem in loco illo moderate et honeste ut servus Dei conversatus vixit. (Flodoardi, Historia Remensis Ecclesiae, I, 20, ed. Heller et Waitz, M. G. H. Scrip., XIII, 443). Mabillon, Ann. Ben., II, 51, ascribes the first visit of the saint to Berceto to the year 718 and the death of the saint to 730. 5 See text cited below, p. 104. o Temporibus invictissimorum Imperatorum Ludovici et Lotharii . . . hie [Tyberius abbas monasterii Berceti, quod est situm in cacumine montis, cui nomen est Bardo] cum sui coenobii ecclesiam, justa quod necessitas commissae sibi congregationis exigebat, aliquantulum in longum porrexisset, quae prius erat modica, vel vix eapiens fratrum collectam; placuit, ut sub altari ejusdem basilicas, pararet congruum locum quo poneretur corpus S. Moderanni, quod istic ad laevam altaris jacet humatum. (Acta Translationis S. Abundii, ed. Boll., Acta Sanct., Julii I, 40). 7 See Affd, I, 163, who quotes Angeli to the following effect: Hebbe questo Castello suo cominciamento dalle ruine di Berce monastero fabbricato nell'alpe di Bardone da Luitprando Re de' Longobardi et donato dopo a San Moderanno, il quale cominciando a ruinare per le mosse de' monti fu trasportato in questo luogo. s Ibid. The basis for this conj ecture is a sentence in the diploma of Ugo cited below. 9 Affd, I, 294; Molossi, 17. 103 LOMBARD ARCHITECTURE Sancti Remigii constructam in comitatu Parmensi.10 In 913 the bishop Elbunco of Parma left a sum of money to restore the apse.11 Affo has published an important document of King Ugo, of 927, in which the monarch laments that the canons of Berceto, of the monastery of S. Remigio founded by King Luitprando, were reduced to great poverty, and suffered from lack of food and clothing, and to remedy this condition, he grants them certain lands.12 Three years later the same Ugo confirmed to the bishop of Parma: Abbaciam scilicet de Bercetum in honore Sancti Remigii extructam in integrum etc.13 In 1007 the bishop of Parma, Sigifredo II, granted part of the oblations of Berceto to the canons of his cathedral church,14 and the oblations of the altars of S. Remigio and S. Moderanno at Berceto were ceded by Ugo, bishop of Parma, to the same canons, in 1035. From this time the importance of the chapter of Berceto appears to have steadily declined. In 1313 the church was pillaged,15 a misfortune from which the canons never entirely recovered. In the XIX century S. Remigio became the parish church of Berceto. In 1845 the monument received a heavy blow in the shape of a quasi-archaeological restoration, which completely denatured the ancient architecture.16 At this epoch were added the chapels constructed, in part, of old materials. The northern transept and the upper part of the northern side-aisle walls were completely rebuilt. The existing facade seems to date almost entirely from this period, but here again old materials were employed in the new construction. The old clearstory disappeared behind the new side-aisle vaults. III. The edifice consists of a nave four bays long, two side aisles, modern chapels, projecting transepts, a tower rising over the crossing, and three apses, of which the two minor ones are semicircular, the central one square externally, but irregularly oblong internally. The nave is at present roofed in wood. The side aisles have undomed groin vaults, which, however, appear to have been added in 1845 and replace the original wooden roofs. It is probable, however, that the undomed groin vaults of the transepts are original, and the highly domed rib vault of the crossing (Plate 22, Fig. 4) is loMuratori, A. I. M. A., ed. A., XIV, 721. n See text cited below under Borgo S. Donnino, p. 171. 12 In nomine Domini Dei aeterni. Hugo gratia Dei Rex . . . Adelbertus vener abilis Episcopus nostri per omnia fidelissimus . . . retulit nobis inter caetera qualiter canonici de Bercedo monasterio Sancti Remigii, quod Luitprandus Rex a fundamentis asdificavit, subjecitque eum, ut sub sacri Palatii tutela esset, murmurarent, atque non haberent ad ciborum seu vestimentorum necessitatem, qualiter in ipso sancto loco deservire possent . . . ut imperium nostrum inviolabile perseveret, nee non auctoritatem Apostolicae Sedis, quam venerabilis Papa Benedictus de eisdem rebus illis fecit Data anno Dominicae Incarnationis 927. 13. Kal. Martii indictione 15. anno vero Domini Hugonis gloriosissimi Regis primo. (Aff6, I, 335). is Muratori, A. I. M. A., ed. A., VI, 320. 14 Affd, I, 383. 15 Molossi, 17. 16 This restoration is recorded by an inscription in the facade. 104 BERCETO, S. REMIGIO (S. ABONDIO) certainly ancient. The diagonals, rectangular in section, are pointed in elevation. There are no wall ribs; the wall arches are approximately semi circular in elevation, except one, which is very slightly pointed. The diagonals, heavily loaded at the groins, are carried on corbels. The doming of this vault is most exaggerated. The arches of the crossing are pointed, as are those of the main arcade. The other arches of the church are all semicircular. IV. The capitals of the church are of Romanesque style, but they were so much restored and made over in 1845 as to be quite misleading. The arched portal of the facade is in seven orders, and is adorned with zigzag and spiral shafts and with a sculptured lunette and architrave (Plate 22, Fig. 3). The influence of Benedetto is evident in the capitals, which have a continuous straight abacus, as in the cathedral of Borgo S. Donnino (Plate 27, Fig. 3). Particularly interesting is the subject of the architrave sculptures (Plate 22, Fig. 3) which appear to be a sort of parody of the Dance of David as represented by Benedetto in the Parma baptistery (Plate 163, Fig. 3). In the centre an animal (which I take to be an ass) strums a harp which is of the triangular form, symbolical of the Trinity. To the left are four animals, apparently symbolical of the four Evangelists, although for the bull and the angel are substituted curiously grotesque forms. All these animals are dancing to the music of the ass, and there is also a group of three human figures on the right-hand side of the lintel. These consist of a man holding a staff or a sword in his hand and a woman, dancing hand-in-hand. Between them is a short third figure, perhaps that of a child. The lintel is completed with the figure of a grotesque animal, a knight riding a stallion and a centaur shooting a bow. Above the centaur is a rosette, clearly showing the influence of Benedetto. The sacrilegious character of this relief, in which, apparently, Christ is depicted as an ass, is most amazing, and can only be explained by the hypothesis that some rebellious and cynical sculptor of this wild mountain region imposed upon the ignorance of his patrons to place over the portal of the church a shocking blasphemy. Equally irreligious is the relief of the Crucifixion above. Christ is depicted fastened on the cross by four nails. His head, with inscribed halo, is held upright. At the ends of the arms of the cross are sculptured two figures, probably Mary and John. This entire motive is a new one in Romanesque sculpture, and undoubtedly shows the influence of painting. In the XIII century painted wooden crucifixes were common, in which the figure of Christ was shown precisely as in these sculptures, fastened to the cross by four nails, and with painted scenes at the ends of the arms of the cross. Above the cross on either side are shown two angels flying in a horizontal position, crude imitations of the style of Benedetto. To the right of Christ are shown Mary and Joseph of Arimathea, both holding their left 105 BERGAMO, S. MARIA MAGGIORE hands to their faces in anguish. The darkest hour of the Dark Ages never produced in Italy works more crude, less life-like and more grotesque than these, and one feels that the sculptor took a certain malicious joy in making the figures as ridiculous as possible. Behind Joseph of Arimathea, squeezed into the curving angle of the lunette, is the figure of St. Remi dressed in archiepiscopal robes, with a mitre and crosier. To the left of Christ is a figure holding an enormous wine jar in which he catches the blood that flows from the wound inflicted by the centurion's spear. This wound, contrary to tradition, is placed in Christ's left side, and the large size of the wine jar in which the blood is being caught suggests again the satiric, irreverent feeling for which these reliefs are notable. At the extreme edge of the lunette are four soldiers with helmets and swords, all treated in a grotesque manner. V. The sculptures of Berceto, since they show imitation, or, rather, burlesquing, of the sculptures of Benedetto, can hardly be earlier than c. 1220. They are somewhat less crude than those of Bardone (c. 1200), and must be somewhat later, since they show the influence, not only of the earliest manner of Benedetto, but of his more mature works in the baptistery of Parma and in the cathedral of Borgo S. Donnino. There is nothing to indicate that the rib vault of the crossing is earlier. As far as can be judged from the present mutilated condition of the edifice, the building was a homogeneous structure until marred by Renaissance alterations and modern restorations. The existence of a rib vault of purely Lombard type at such a late date is indeed extraordinary, but must be explained by the fact that Berceto is placed in a remote mountain district where artistic forms developed somewhat tardily. Moreover, the vault of Berceto was placed beneath a tower where, as I have shown,17 the rib vault offers peculiar constructive advantages. The date of 1220 is, moreover, in perfect accord with the fact that pointed arches are freely used in the edifice. BERGAMO, S. MARIA MAGGIORE (Plate 22, Fig. 5, 6, 7; Plate 23, Fig. 1, 2, 4) I. Although situated in an important city, and one frequently visited by tourists, the notable fragments of XII century architecture preserved in S. Maria Maggiore at Bergamo have been strangely neglected by archaeolo gists. The first to describe the church were the brothers Sacchi,1 who wrote in the early part of the XIX century. They were followed about the middle of the century by Osten,2 who published a large engraving of the apse. In 1880 17 Construction of Lombard and Gothic Vaults, 12. 138. 2 Plate XXXVI. 106 BERGAMO, S. MARIA MAGGIORE appeared the monograph of Fornoni, a sumptuous publication, lavishly illustrated with large-scale drawings. Unfortunately in this work the citations of historical authorities are not always exact, and there are numerous errors in the measurements and restorations. In the Museo Civico at Bergamo are numerous old drawings of the church, but none which throw real light upon its history. The historians of Bergamo have all treated at length of the historical problems which the church presents. Among them should be consulted especially Pellegrini, who wrote in 1553, Calvi, who wrote in 1676, the classical work of Lupi, and the modern publications of Ronchetti (1807), and Locatelli (1879). Finally should be mentioned the little study in the compilation of Strafforello.3 II. It is usually stated that the church of S. Maria Maggiore was founded in 1137, but such cannot be the case, since it is mentioned in a document of 774. 4 The campanile is again referred to in 928. 5 From these documents it results that the church must have been not founded, but reconstructed, in 1137. The principal source for this fact is an inscription on the archivolt of the southern portal. This inscription was originally painted in Gothic letters, but at some unknown time the painted inscription was replaced by an incised inscription, doubtless because the original was becoming so effaced as to be illegible. When I first visited the church in 1910, some traces of the ancient letters still existed, but unfortunately not enough to make it possible to control the reading made by the person who cut the incised inscription, and when I returned to the monument three years later, the painted letters had almost entirely disappeared. The inscription in substance records that on the upper lintel of the church of S. Maria Maggiore of Bergamo, there existed an inscription to the effect that the church was founded in the year 1137, in the time of Pope Innocent II, and King Lothair, and Roger, bishop of Bergamo. The name of the master-builder, Maestro Fredi, is also given: * . IN . XPI . NOMINE . AMEN . IN . LIMINE . SVPERIORI . ECCLESIE . BEATE . MARIE . VIRGINIS . CIVI . TATIS . PERGAMI . CONTINEBA . TVR . QVCCL . DICTA . ECCLESIA . FONDATA . FVIT . ANNO . DOMI- NICE . INCARNATIONIS . MILLESIMO . CENTESIMO| TREGESIMO . SEPTIMO . SVB . DOMINO . PAPPA . INNOCENTIO . SECONDO SVB . EPISCOPO . ROGERIO . REGNANTE . REGE . LOTERIO . PER . MAGIS- TRVM . FREDVM. This inscription is not without its difficulties. The monstrous word qvccl is written for quod, as I am fortunately able to attest from traces of the 3 Bergamo, 34. 4 See text cited below under S. Giulia di Bonate, p. 162. 5 Hist. Pat. Mon., XIII, 897. 107 LOMBARD ARCHITECTURE painted inscription visible in 1910. But what are we to say of Roger, bishop of Bergamo, when it is known that in 1137 Gregorio was bishop? And what of the King Lothair, when Lothair II was at this time really emperor? Are these mistakes of the master-builder of the XIV century, who misread the original XII century inscription which he undertook to preserve, or are they the errors of the modern restorer, who replaced the painted letters by an incised inscription? There can be little doubt that the former is the case. In the historians of Bergamo we have preserved a series of transcriptions of the inscription, which goes back to the XVII century, and hence must contain some copies made before the restoration of the inscription. In all these transcriptions, despite numerous copyists' errors, it is evident that the original inscription was always essentially the same.6 In addition to the inscription there were apparently documents still extant in the XVI century regarding the construction of the church in 1137 that are now lost. The existence of these is recorded by Pellegrini, who, however, probably in carelessness, ascribes to the year 1137, not the founda tion, but the consecration of the church.7 Giovanni Filippo assigns the foundation of the church to the year 1135, and says it was begun by the 6 The earliest of these copies is that of Celestino, who wrote in 1617 (II, pt. 2, p. 297) : M . CCC . LX . magister Iohannes f. q. Domni Iohannis de Capellio fecit hoc opus in Christi nomine, amen. In limine superiori Ecclesiae B. Mariae Virginis Ciuitatis Pergami, quae olim dicta Ecclesia fundata fuit anno Dominicae Incarnationis millesimo trigesimo septimo sub Domno Papa Innocentio secundo, sub Episcopo Rogerio, regnante rege Lothario per magistrum Fredum. Calvi (III, 296) read the inscription as follows: MCCCLD [sic] Mag. Ioannes f. q. D. Ioannis de Campellio fecit hoc opus in Christi nomine amen. In limine superiori Ecclesice S. Mariw Virg. Ciuitatis Pergami continebatur, quod olim dicta Ecclesia fundata fuit anno Dominicce Incarn. 10S7. Sub Domino Papa Innocentio II. Sub Episcopo Rogerio regnante Rege Lhotario per Magistrum Fredum; but appears to have taken it not from the original but from Celestino's copy. Ronchetti's version (III, 66), published in 1807, is as follows: In Christi nomine amen. In Limine superiori Beatae Mariae Virginis Civitatis Pergomi continebatur quod dicta Ecclesia fundata fuit anno Dominice Incarnationis millesimo centesimo III gesimo septimo sub dom. Papa Innocentio II. sub Episcopo Rogerio Regnante Rege Lothario per magistrum Fredum. In 1880, Fornoni published the inscription with new variations: In limine superiori ecclesice B. Marias Virginis civitatis Pergomi continebatur quod dicta ecclesia fundata fuit anno dominical incoro- nationis MCXXXVII sub domino Papa Innocentio II sub episcopo Rogerio regnante rege Lothario per magistrum Fredum. His version appears to have been taken verbatim from that of Locatelli (III, 195), with which it is identical, even to the italics used, except that in Locatelli we read eclesia for ecclesia, incarnationis for the monstrous incoronationis, Rugerio for Rogerio, and MAGISTRUM FREDUM. The little founded conjecture of Osten that magistrum Fredum is a copyist's error for magistrum Alfredum is hardly worth discussion, since Fredi, or Fredo, is a common Italian name, whereas the Saxon, Alfred, is almost unknown south of the Alps. 7 [Gregorius Episcopus] anno sequenti videlicet 1137. templum maius in vrbis nostras medio ad honorem beatissimae virginis Mariae dedicauit, & anno. 1144. die 19. Iunii occubuit. . . . Haec ex annalibus, & diurnalibus praedictae Abbatias s. Sepulchri, 108 BERGAMO, S. MARIA MAGGIORE citizens ex voto because of a great famine and plague.8 The same tradition appears in an inscription of the XVIII century, now over the choir9 and is also related by Bartolomeo Farina.10 It is, moreover, confirmed by a number of documents of the ancient archives now preserved in the Biblioteca Municipale at Bergamo. These documents, when I had access to them, were only in part classified, but from those which I examined it was evident that S. Maria Maggiore has always stood in a peculiar relationship to the commune of Bergamo,11 and was not, as has been asserted, co-cathedral with S. Vincenzo,12 notwithstanding the fact that in 1340 a baptistery was erected in the church. S. Maria Maggiore, as the most conspicuous church of Bergamo, has served as a burial place for illustrious citizens, such as Colleoni, Donizetti, Mayr, etc. In an unpublished document of the archives, dated 1195, is cited an earlier document of 1170, which is dated "under the porch" (sub porticu) of S. Maria Maggiore. This proves that in 1170 the construction of the church, if not completed, was at least far advanced, since the portal had been erected.13 & ex chronicis domini Bartholomaei de ossa Bergomatis parte 5, lib. 16, cap. 47. (Pellegrini, 7). 8 Templum maius quod misericordiae dicitur earn ob re[m] hac nostra in urbe Bergomo in honorem beatissima; uirginis Mariae in medio urbis fere iuxta cathedralem basilica indicibili prope: tu impesa: tu artis elegantia: ex lapidibus quadratis & sectis a conciuibus nostris ob misericordia[m] ipius dei genitricis Mariae hoc aestuati tp[or]e implorada hoc anno [1135] ceptu est. Atq[ue] inde p[er] tempora eximia pulchritudine cum capanaria pulcherrima atqfue] sublimi cosumatu fuit. Eo naq[ue] tempore Bergomates nri fundato [ut ita dixerim] misericordiae templo cu multi fame et peste laborarent: et misericordiae loca plurima ad elemosinas clamet ap[er]te paup[er]ibus errogandas in urbe & extra instituere: quae usqfue] in prassens tanta cu religione & pietate in unu collectae excreuit. Vt in tota italia eidem non inueniatur consimilis. (Giovanni Filippo sotto anno 1135). » DEO OPT' MAX' MARLE VIRG' MATRI CIVITAS EXSTR' AN' MCXXXVII io Refert Forestus, in supplemento, anno MCXXXIII, Bergomi & in Lombardia calores tam intensos fuisse, ut fruges exsiccarent, indeque annonae caritas inaudita prorsus profecta, eamque pestis immanis insecuta fuerit, ob quas calamitates, ad intercessionem B. V. Civitas confugiens, in honorem ejus fabricandum curavit sump- tuosissimum illud Templum S. Marice Majoris, urbis Cathedrale. (Bartholomaeus Farina, De Bergomi Origine et Fatis, ed. Graevius et Burmannus, Thesaurus Antiqui- tatum et Historiarum Italiae, IX, pt. 7, p. 10). n In documents of 1170 and 1195, cited below, ministri are seen to be in charge of the church of S. Maria Maggiore, and these are not canons of the cathedral. 12 In Lombard cities there were frequently two cathedrals, one of which was used for service during the summer, the other during the winter. is Vna die que e in mse aug. . . . Arnaldus d[e] corterezze [et] lafracus d 109 LOMBARD ARCHITECTURE Ronchetti, who wrote in 1807, saw in the archives of the cathedral chapter documents pertaining to a lawsuit of 1187, in which witnesses testified that about this time the church of S. Maria Maggiore was enlarged and embellished.14 These appear to have been known also by Fornoni, who wrote in 1880, since in speaking of them he adds certain details which could not have been derived from Ronchetti.15 In an ancient calendar16 of the church of Bergamo published by Finazzi, there is recorded a consecration of the altar of S. Maria Vetera in 1185. Since the indiction corresponds in this notice with the year, it is to be preferred gastaldio ministri eccl[esi]e see marie. . . . factu e hoc. anno dnice incarnatiois. millo centesimo septuagimo. Indjoe sexta [sic] Die decimo intnte mense decembr. Jn ciuitate bgami sub porticu see marie. maioris. Presentib; infra scriptis testib' Ottacius fili' [quon]da pet' de pappa. q[ui] p.[ro]fessus est lege uive longobar. Fecit datu uendit' noe. ad p[er]petatem. Jn manib; lanf de sea maria [et] petri redulfi [et] magri zaboni missorfum] eccie see marie maioris vice [et] noe illi' eccie. . . . Factum est hoc anno dni millo cento nonago q[ui]nto. indjone triadecima. Similar phrases are contained in another parchment in the same writing, dated 1195: Scd'o die intate mense febr. Jn civit' b[er]gi sub porticu see marie maioris. Presentia infra scriptor[um] testiu. Maginfredus filius [quon]da albti de lalio q[ui] p[ro]fessus e lege vive longobar. Fecit datum venditionis noe ad p[er]petatem. Jn manib; lanf d' sea maria petri redulfi [et] magistri zaboni ministor[um] eccie see marie maioris noe [et] uice illi' eccie. Nominatim de quada petia tre vidate. iuris sue qua habe visus e foris civit b[er]gi no m[ul]tu longe . . . pdictis lanf petro. [et] magistro zabono. noe et vice pfate eccie [et] eofrum] succes- sorib; . . . lanf d' sea maria [et] petri [et] mgist' zaboni . . . Factum est hoc anno dni millo cento nonago q[ui]nto. indjone triadecima. (Pergamene in Biblioteca Civica of Bergamo, Collezione Congregazione di Carita). 14 Ci6 che dunque rileviamo da giurati testimonianze dell'anno 1187. esistenti nell'archivio capitolare si e, che circa questo tempo detta chiesa di S. Maria fu ampliata, e resa piu bella, il che per6 non si esegul, che nel corso di molti anni non essendo per anco nel detto anno 1187. terminata; che dopo essersi cominciato tale rifacimento continuava il clero di S. Vincenzo ad officiarvi nelle feste della Beata Vergiue [sic], e nella quaresima vi si cantava dal medesimo la messa dopo nona; che la collazione de' suoi beneficj facevasi per libera elezione del capitolo; che il vescovo col clero ne] Sabbato santo vi celebrava l'officio, e andava processionalmente a benedire il fonte, e ad amministrare il battesimo, trovanosi [sic] sino da' tempi antichi in quella chiesa il fonte battesimale, ehe era l'unico in tutto l'ampio circondario della citt& e de' borghi, e continud ad esservi sino al secolo decimo settimo, nel quale fu trasportato in S. Vincenzo. 15 Un documento poi che si rinvenne nell'archivio del Capitolo giustifica anche il resto della lapide; poiche in esso certo Lanfranco Mazocchi nell'anno 1187 con dichiarazione giurata attesta che le fabbriche, le quali si innalzavano in quella localita, vennero completamente distrutte per ampliare e rendere piu bella una Chiesa dedicata a Maria, che ivi sorgeva. (9). i° V K. [Decembr.] Consecratum est altare see Marie veteris, M C octuagesimo quinto, Ind. tertia. (Finazzi, 415). 110 BERGAMO, S. MARIA MAGGIORE to another similar notice in another calendar17 published by the same authority in which the consecration is referred to 1184. In view of the other documents just noted, which prove that shortly before 1187 the church was in construc tion, I have no hesitation in referring these texts to S. Maria Maggiore, although I know of no other case in which the church is called vetus or antica. It seems clear, therefore, that in 1185 the construction, if not finished, was at least sufficiently far advanced to admit of the consecration of the altar. Two inedited documents of 1211 and 1235 respectively make it evident that in the early part of the XIII century the church of S. Maria Maggiore was still officiated by a clergy entirely separate from that of the cathedral, but yet subordinate to the latter.18 According to Pellegrini, the church was consecrated in 1273. This historian, who wrote in the middle of the XVI century, assures us that he derived this information from documents of the archives,19 but the notice is unknown to all the other historians of Bergamo, and is unconfirmed by any documents which have come down to our day. Moreover, in the church itself it is possible to find no evidence of any additions or embellishments made at this time, and it is almost inconceivable that changes sufficiently radical to necessitate a consecration could have been made without leaving some trace in the monument as we have it to-day.20* I am, therefore, inclined to believe that the date 1273 of Pellegrini, owing to a misprint, or other mistake, is erroneous, and that the consecration really was celebrated after the completion of the works which were in progress in 1187. In 1351 the northern portal was added to the church according to an i?V. K. [Decembris]. Consecratum est Altare See Marie veteris millesimo centesimo octuagesimo quarto. Indict, prima. (Finazzi, 407). is Die ttio intrante sept. Jn civit poi [= Bergami], sub porticu cupticelle ecclie maioris see marie. . . . Feceru' datu [et] car dati noie vendiciois ad p[er]petate [et] iure p[er]petatio. dno petro redulfi. [et] dno mar. [et] iohi frib [et] concis [et] ministris illi' ecclie. vice ac noie illi' ecclie. d' qda dnr ficto. q. habebat. . . . Factu est hoc ano dni millo ducentoe vndecimo, indict joe qartadecima. . . . (Pergamene in Biblioteca Civica, Bergamo, Collezione Congregazione di Carit&). Die mcurii nono intnte novembri [the chapter of S. Vincenzo authorizes] dno Giufredo archipsbro d'pagano psbro ecclie maioris see marie ciuite p[er]gi. comutandi [et] in causam comutationis dandi noie ipi' ecclie see marie [et] p[er] ea omes tras [et] possessiones ipi' ecclie see marie etc. . . . Factum est hoc anno dni mille ducent'o trig'o nono (Ibid.). 19 Hie [Suardus episcopus] ecclesiam sanctae Mariae Matris Domini dedicauit, anno Christi. 1273. & episcopatus eius primo, & concessit omnibus visitantibus ea in ilia die dedicationis videlicet. 35 martii dies quadraginta. Haec ex dicto memoriali, & antiquis scripturis ipsius monasterii, & ex nostro Kalendario. (Pellegrini, 9). 20 Unless, indeed, the changes were confined to the clearstory and vaults of the nave, which have disappeared, or to the addition of frescos which have been covered with intonaco. Ill LOMBARD ARCHITECTURE inscription originally painted and now preserved in an incised copy.21 The portal was finished probably in 135522 and in 1360 the southern porch was erected by the same master-builder, according to the existing incised inscrip tion,23 or by his son, according to old copies which probably reproduce more faithfully the original painted inscription.24 The northern portal of the choir was added in 1367, as is evident from the fact that the date M.CCC.LX.VII. is incised on the lintel. According to Locatelli25 the sculptures of God and the Annunciation were completed in 1403, and were the work of Antonio de Alemania. In 1449 the care of the church was ceded by the city to the Misericordia,26 which subsequently became the Congregazione di Carita, to which corporation it still belongs. Four years later the building was exempted from the jurisdiction of the bishop.27 The church must have been restored in the XV century since in the gallery of the eastern absidiole of the southern transept there are numerous indications of alterations executed at this epoch, among others a capital of the early Renaissance style. A description of the church, written in 1516, by Antonio Michele, gives a picture of the edifice as it was in the early, part of the XVI century, shortly after the addition of the Colleoni chapel, but before the building had been baroccoized.28 In 1647 the church was much damaged by lightning,29 an 21 * M . CCC . LI . MAGISTER . IOHANES . DE. CAMPLEONO CIVIS . PGAMI . FECIT . HOC. OPVS. 22 Merzario, I, 138. 23 -KM. CCC . LX . MAGISTER . IOHAN ES . FILIVS C[= cuiusdam or quondam?] . DNI . VGI . DE . CAMPILIO . FECIT . HOC OPVS 24 See above, p. 108. 25 III, 205. 20 Calvi, I, 291, and II, 343. "Ibid., I, 292. 28 Contra vero D. Mariae aedes neque tam [as S. Vicenzo] vetustae dicationis, utpote ducentesimo circiter ab hinc anno asdificari inchoata, neque sacerdotii dignitate par, ob operantium tamen sanctitatem & frequentiam ita a populo visitur, ut nulla sit in urbe aedes celebrior. Ejus longitudo secus areolam quam dixi ab ortu ad occasum patet, forma extra quadrata est, intus christiano ritu Crucis figuram praesefert. Arae 112 BERGAMO, S. MARIA MAGGIORE event which probably provoked a restoration in the barocco style, for in 1651 Calvi makes a reference to the preparations in progress for covering the interior with stucco.30 In 1655 a new misfortune befell the edifice. The lights hung in the campanile to celebrate the accession to the pontifical throne of Alexander VII, set fire to the church, and the conflagration was checked only after all the lead with which the cupola was covered had been destroyed.31 In 1660 the baptistery was removed from the church.32 From the description of Calvi, written in 1676, it is evident that at this date33 the baroccoization of the church had been almost completed. In 1771 the wooden doors of the southern portal were added.34 The monument has been happy in escaping a modern restoration, and still preserves its barocco interior, with all the gilt and restless over-decoration. Veritable oases in the desert of stucco angels and barocco curves are the lovely Renaissance tapestries and intarsia choir-stalls. maximae (quae ad ortum spectat) fornicatum tribunal circumagitur, cui duo alia minora tribunalia unum k dextra, alterum a, laeva adhasrent, pro his tribunalibus transversa ambulatio patet, per cujus utrumque cornu in aedem est ingressus: reliquum asdes quod ad occasum extenditur duabus cellis [= side aisles], una hinc, altera illinc, mediaque ambulatione [— nave] constat, e cellis ilia quae est a meridie k tergo parietem in hemicycli formam ductum habet, k fronte (qua transversam ambulationem spectat) tota patet: altera vero quae est a septentrione in sacellum est versa in quod temere non datur ingressus, utpote ara & Bartholomcei Colleonis monumento religiosum. (M. Antonii Michaelis, Agri et urbis Bergomatis descriptio, anno MDXVI, ed. Graevius et Burmannus, Thesaurus Antiquitatum et Historiarum Italiae, IX, pt. 7, p. 31). 29 1647. Piombd di mattina a buon hora la saetta in Santa Maria Maggiore, entrd nel campanile, passd per l'organo, trascorse gfaltari, cagiando moltissimi danni. (Calvi, II, 539). 30 1651. In S. Maria Maggiore cauandosi verso la parte Orientale la terra, a fine di piantar li ponti per la stuccatura, si faceua in quella parte del tempio. . . . (Ibid., II, 420). si 1655. Solennizandosi con ogni dimostrazione d'allegrezza l'assontione al Ponti ficate d'Alessandro Papa VII dalla moltitudine di luminari posti nella sommita del Campanile di S. Maria Maggiore acceso il fuoco consumd in puoco tempo, & distrusse il piombo tutto [sic] di cui era la cupola ricoperta. Danno, che non si potfe riparare se non con la spesa di mille scudi per rifarla di nuovo. Diar. Par. (Calvi, II, 91). 32 Genaio XV, 1660. A fine di render piu spatiosa, e vaga la Chiesa di S. Maria Maggiore souerchiamete dal recinto dell'antico Battisterio in essa riposto occupata, diede in questo giorno la Citta nel pieno maggior Consiglio a Presidente della Miseri- cordia licenza, di poterlo far demolire, come poi fu esseguito nel venturo Febraio. Ex. lib. Cons. Ciuit. 1660. (Calvi, I, 93). 33 £ disposta la nobil Chiesa a forma di Croce, con sublime non meno che vaghissima cupola nel mezzo tutta a stucco, & oro di pieno rilieuo, come fe anco il rimanente della Chiesa, leuata vna parte verso il Vescouato, non ancor terminata. (Ibid., 290). . . . Tutta la Chiesa si fe posta a stucco l'anno 1670. (Ibid., 355). 3i They bear the inscription: ANNO DOMINI MDCCLXXI. 113 LOMBARD ARCHITECTURE III. Beneath the accumulated additions of later centuries, the church still preserves almost intact the original Lombard structure, and is, indeed, notwithstanding the tasteless modern decorations, one of the very finest XII century edifices of all Italy. It consists of a nave two bays long, two side aisles, projecting transepts with eastern absidioles (the southern transept has also a western absidiole) (Plate 23, Fig. 2), a choir of a single bay, flanked by side aisles, and an apse. The bays of the nave are slightly oblong, those of the transepts very oblong. It is probable that the nave was originally spanned bj' a single transverse arch, supporting a wooden roof, but in the present condition of the church it is impossible to be certain, for the nave walls have been rebuilt in their upper portions and in their lower portions covered with intonaco. The system of the choir was evidently alternate, a single bay of the nave embracing two of the side aisles. The existing groin vault of the choir, I believe, is not original, but was reconstructed in the XVI century. The octagonal cloistered vault over the crossing was probably originally carried on arched squinches, but this part of the edifice also has been entirely made over, so that it is impossible to be certain of the original dispositions. The upper portions of the north and south transept have similarly been entirely rebuilt, and, like the nave, are now supplied with a roof carried on modern transverse arches. The side-aisle vaults, which are not domed, have probably been rebuilt or at least denatured. Over the side aisles of the nave and choir there existed in the Lombard basilica a high gallery, which is still extant, although walled off. These galleries are among the best preserved parts of the Lombard edifice (Plate 23, Fig. 1). They are covered by groin vaults, of which the transverse and wall arches disappear towards the springing. The original vaults are highly domed. The diagonal ribs of the vaults of the gallery of the choir must be barocco additions, and the vaults of the north gallery of the choir, which are not domed, have evidently been entirely made over. In the choir the galleries opened upon the nave by means of bifora, but in the main body of the edifice by grouped triforia (Plate 23, Fig. 1), of which the central arch was pointed. At present the galleries and nave are covered by a roof of continuous slope, and although the upper part of the edifice has obviously been remade, it is probable that there never was a clearstory. The lower story of the campanile cuts across the southern absidiole, and therefore must be later than it. An inspection of the masonry is sufficient to show that there are in the church two distinct eras of construction. To the later belong the upper part of the southern transept, the entire west wall of the southern transept, with its absidiole (Plate 23, Fig. 2), the south wall of the nave (Plate 23, Fig. 2), the north wall of the nave, and the west facade. Tliese portions are con structed of rough ashlar, with small stones of irregular shape and horizontal courses frequently broken. This masonry contrasts strongly with the careful 114 BERGAMO, S. MARIA MAGGIORE ashlar of the rest of the edifice (Plate 23, Fig. 4). Strangely enough, however, the rougher ashlar is of a later epoch than the smoother. This will be demon strated below in studying the ornament of the two portions of the edifice. It is, however, shown also by a careful study of the masonry itself, for the inside of the nave gallery (Plate 23, Fig. 1), belonging to the second epoch of construction, is finished with ashlar even finer than that of the first epoch. It is evident that the work of constructing the church was suspended for a considerable time and that, when it was resumed, for reasons of economy, inferior masonry was emplojred. The existing cupola is a curious mixture of old and new fragments. The masonry is covered with plaster and the Gothic capitals are modern. It appears to preserve very little of its original character. IV. The capitals of the church form an interesting study, and confirm the fact already indicated by the masonry that the church belongs to two distinct eras of construction. In the gallery of the choir the capitals are foliated, with broad, flat leaves, or are of the wreathed type familiar at Fontanella al Monte. In the exterior galleries of the eastern absidioles is a remarkable series of capitals showing a great variety of design. Some are of Corinthianesque type, with feathery acanthus leaves (Plate 22, Fig. 6) ; several have well sculptured eagles (Plate 22, Fig. 7) ; some are adorned with grotesques and one has sculptures representing the four archangels blowing trumpets. These capitals are very similar to those of S. Fedele at Como (Plate 63, Fig. 1, 8) although evidently somewhat more advanced in style. To the second era of construction belong the capitals of the nave gallery, of Corinthian type, with broad, uncarved leaves (Plate 23, Fig. 1) and those of the western absidiole of the southern transept of a high bell type somewhat Gothic in character (Plate 23, Fig. 2). The apses and absidioles are decorated with practicable galleries (Plate 22, Fig. 5; Plate 23, Fig. 2). In the portions of the edifice belonging to the first era of construction, beneath the galleries are placed a row of blind arches finely moulded and supported on shafts engaged on pilaster strips (Plate 22, Fig. 5). These arches recall those of the cathedral of Parma. In those belonging to the second era of construction (Plate 23, Fig. 2), there are simply arched corbel-tables supported on pilaster strips. In the southern absidiole is an elaborate cornice and a string-course adorned with a guilloche in which are sprinkled grotesques. The north absidiole is elaborately ornamented with moulded ornaments in both string-course and cornice (Plate 22, Fig. 5), and the same motives are repeated in the central apse. The mouldings are numerous and fine. The triforium, both in the choir and in the nave, is lighted by bifora, with broad-leaved capitals (Plate 23, Fig. 2). In the eastern absidioles of the southern transept, there is a sculptured capital, unfortunately broken. On one side is represented the sacrifice of 115 LOMBARD ARCHITECTURE Isaac by Abraham, with the angel and ram (Gen., xxii, 11, 13), on the other the angel appearing to Abraham (Gen., xxii, 15). Notwithstanding a certain crudity in the details of the faces, feet, etc., these figures are very fine. The technique is excellent, as witnessed by the deep undercutting, the composition superb and splendidly architectural. By the same hand appears to be the southern portal (Plate 23, Fig. 4), which is part of the original basilica, and antedates the Lombard porch which was merely added on at a subsequent epoch. This portal is in many orders, very elaborately shafted and moulded, and ornamented with spirals, interlaces, bead-mouldings, etc. In more than one detail it recalls the portal at Borgo S. Donnino (Plate 27, Fig. 3). The capitals of the western jamb are adorned with feathery s finely carved acanthus leaves. On the eastern jamb the capitals are replaced by a frieze with figure reliefs. To the left is shown the Visitation, with the singular detail that two handmaidens on either side hold the cloaks of Mary and Elizabeth. Elizabeth's handmaiden is apparently naked to the waist. The sculptor has succeeded in elegantly expressing by means of the flying draperies the motion of Mary and her handmaiden, who appear fairly to rush towards Elizabeth. There follows the apparition of the angel to Joseph. Joseph holds a cane in his right hand, the angel grasps his left arm with a firmness that suggests coercion rather than moral suasion. The last scene represents the Presentation. The Virgin offers the Christ-Child to Simeon, who stands behind an altar. Back of Simeon is seen a handmaiden bearing two turtle doves. The draperies in these sculptures are treated in broad masses, a little differently from those of the capital of the absidiole. On the other hand numerous details, such as the curious zigzags at the bottom of the garments, the enormous hands and the crudity in the treatment of the faces and details of the anatomy, combined with the architectural feeling and excellent composition common to the two works, leave no doubt that they are by the same hand. The southern porch is, from many points of view, a most interesting monument (Plate 23, Fig. 4). As is evident from the inscription already cited, it was added to the church in 1360. A close study of the structure, however, gives reason to believe that the attic was built subsequently, and, as the style would seem to indicate, in the early part of the XV century. Now, we have already seen that the statues of Christ and the Annunciation over this southern portal were added in 1403; it is therefore altogether probable that the attic is of the same epoch. This attic is a most important monument of mediaeval archaeology because of the reliefs with which it is sculptured. Those on the south and west faces representing Christ, the twelve apostles and five saints, among whom figure SS. Lawrence and Anthony, need concern us little. Not so, however, those on the east face,' which show four genre scenes representing the life and activities of mediaeval builders. In the great dearth of documents referring to the building trades in the Middle Ages, these reliefs, which have never before been published, nor, I 116 BERGAMO, S. MARIA MAGGIORE believe, even noticed, assume the greatest importance. In the first relief is shown a man seated at a desk and drawing with a pair of compasses. He has a curious skull cap which covers his ears and neck, and wears a sort of jerkin, apparently of leather, whose sleeves are fastened with buttons or thongs on the lower edge. Behind him is what looks like a pine-apple, but which is probably merely an ornament of the bench on which he is seated. This figure is probably intended to represent the head master-builder. His bench is more elaborate than that of the others, all of whom face him, and he is distinguished from them by his head-dress. Below is the inscription [SCVJLTORETVS, exceedingly difficult to interpret, but which is probably a latinization of the diminutive of the Italian word, scultore. In the second relief is shown a builder engaged in working a capital, like the similar figures In the Porta dei Principi of Modena cathedral (Plate 142, Fig. 4) and in the bronze doors of S. Zeno at Verona (Plate 234, Fig. 1). He has a head-dress of cloth caught back of his ear and allowed to flow behind, leaving the neck and ear uncovered. He appears to have a leathern jerkin like the head master-builder. He is seated on a stool and is working at a capital, which is held inverted, perhaps on a stand before him. Below is the inscription ARISTATIVS. This word, so far as I have been able to discover, is not found elsewhere in mediaeval Latin, but is doubtless a technical term to indicate one of the grades in the profession of builder. The root must be identical with that of aristato, also a word of obscure meaning, but which is believed to have reference to a wooden structure erected on tombs.35 The third relief shows a builder, clothed precisely like the second, but holding a chisel in his left hand and a hammer with a stone head in his right, and engaged in chiseling the neck of a capital after it has been placed in position. Taken in connection with the second relief, this shows that the mediaeval builders blocked out roughly their capitals before placing them in position, but finished them afterwards. Below is the inscription PIS[C]HOMASTIVS, another word which does not occur elsewhere in mediaeval Latin, but which it is natural to connect with the Italian verbs piszare and pizzicare. The fourth scene represents a builder with a curious Oriental cap, gathered in a tuft at the top and with a flat band below. He is engaged in hollowing out with a stone hammer the inside of an object the precise nature of which it is impossible to determine, but which looks as though it might be a capital in its first stage of manufacture. Below is the inscription GRECHVS, which gives rise to many conjectures upon the extent of Byzantine influence over the mediaeval building trades. From these reliefs it may be adduced that c. 1400 the head master-builder at Bergamo was known as a scultoretus, who occupied himself chiefly with making drawings, and that he had under him three grades of assistants, known respectively as grechus, pischomastius and 35 See Du Cange. Cf. the Greek word 8.pi