Return to Index

The Byzantine
Gospel Lectionary

Introduction:

In his famour letter 96 to Trajan, ¹ the younger Pliny describes an early second-century Christian service. He tells of the congregation meeting before dawn and vowing to abstain from improper conduct. During the service the members chanted psalms in Christ’s honor; then, after a recesss, those who were baptized gathered for a meal.

Around the middle of the century, Justin Martyr mentions the use of the writings of the prophets and what he calls the “memoirs of the Apostles” (απομνημονευματα των αποστολων) as part of the service. ²

Such sources reveal a liturgy that, although doubtless primitive, was structured in a familiar way, in two distinct parts. The first centered on a reading, which might serve as the basis of a sermon. In the private house converted for use by the Christian community of Dura Europos, Syria, exists some indication of the spaces in which these early rites were conducted.

Later evidence casts an oblique light on the growth of the liturgy after the second century. The list of vessels and other objects confiscated during a raid on the North African church at Cirta Constantina in 303, ³ as well as church construction from Constantine to Justinian, speaks of cult practices requiring both a large, impressive stage ⁴ and substantial equipment in the form of patens, chalices, censors, and so on.

But of the materials needed for the service, books are perhaps the most important. Books are what the authorities were searching for in the small North African church when they uncovered cups, plates, and garments. The importance books held for the early Church is attested by Constantine the Great’s letter to Eusebius; in it he orders fifty well-made copies of the Holy Scripture to serve the needs of the various churches in the newly established capital. ⁵

In the centuries after the death of Constantine, the liturgy became increasingly complex and richly ceremonial. As church feasts were added, passages from the Old and New Testaments were selected and assigned as more or less fixed readings. The Gospels, though, played a role beyond that of merely being the source for one set of lessons (and lessons of such importance that they were entrusted only to a deacon, never to a reader or chanter).

By the early eighth century the Divine Liturgy began with the First (Little, Lesser) Entrance, which the Patriarch of Constantinople, Germanos I (715-30), described as a procession headed by the deacon carrying the Gospels, followed by the celebrant. To Germanos the entrance of the priest and deacon bearing the manuscript signified nothing less than the coming of the Son of God into this world. ⁶

A roughly contemporary Gospel book of the kind used in the service survives in the University Library at Basel. ⁷ The scribe who copied the text marked the beginnings and ends of the passages read in church, and in the margins he added the various phrases that served to introduce, as freestanding units, readings cut or pruned (pericope) from their narrative context.

It is not difficult to find medieval manuscripts of the Greek Gospel text that include the liturgical apparatus: a set of tables to allow the reader to identify the passage for the day, the marginal or interlinear arche and telos signs that told him precisely where to begin and end the lesson, and the standard incipits, such as “At that time…” or “The Lord said to his disciples…” ⁸

Such copies of the Gospels continued to be produced until the fall of Constantinpole. Occasionally, the owner of a Gospel book made without the lectionary apparatus would have a scribe prepare it for use in the service. The Vienna Gospels (Vind. Theol. gr. 154) was made around 1050, but its table of readings and the interlinear markings were added centuries later. ⁹

However long and widespread the liturgical use of the Gospel book was, there also appeared at some time in early Byzantine history the kind of book known as the Gospel lectionary, the kind discussed here in a magnificent cruciform example in the Pierpont Morgan Library, New York. The New York Lectionary, along with most of the others I will use for comparison, is an invention of the Middle Ages.

A Gospel passage is assigned to nearly every day on which a service might be held, and for important feasts readings are provided to cover morning and even hourly services. Owing to the nature of the calendar and the requirements of the church, the text was divided and subdivided into various parts. The basic division is twofold: the synaxarion and the menologion.

The synarion contains the movable readings beginning with Easter Sunday and ending with the celebration of Christ’s Passion. The first half is divided into four parts, which by the early eleventh century had come to be associated with the Evangelists: the John readings extend from Easter to Pentecost, those from Matthew from Pentecost to around mid-September and the Luke readings from mid-September to the beginning of Lent, and Mark supplies the Lenten pericopes.

The menologion, the second major part of the lectionary, contains the fixed celebrations arranged daily and by month, beginning with September. So complete is the medieval calendar that the text of four Gospels was exhausted before the end of the lectionary. Some readings would be given more than once in the course of the year, and cross-references occur with increasing frequency toward the end of the menologion.

Figure 31b, for example, contains one such reference. At the head of the pericope the scribe wrote two lines of gilt half-unical, the first of which gives the day and then an instruction to the reader:

“Saturday before the (Feast of the) Elevation (of the Holy Cross): see the seventh Saturday in Matthew.”

When preparing the service for that day, the deacon knew he would need to turn back to the synaxarion and find the Matthew readings, then select the passage for the seventh Saturday. The reproduction shows another feature of the Gospel lectionary used in the Middle Ages; above the lines of text are the neumes, the markings of Greek ekphonetic notation. Generally written in carmine ink, they were added to enable the reader to chant the pericopes.

Although the structure of the medieval lectionary can be easily summarized, the historical circumstances surrounding its invention remain unclear. Who, for instance, ordered and supervised the arrangement of the calendar, and by what means were the manuscripts disseminated? What can be said about the lectionary prior to the ninth century? One finds that even basic issues, such as what the book was called, also remain unclear.

The lectionary appears to have been known by a number of names. A scribe might refer to one he had written as the “book of readings” or a “collection of readings; ¹⁰ in the Typikon of Saint Sophia, which gives the pericopes and hymns used in the patriarchal service, the lectionary is called the megaleion, the “grat book” ¹¹

In fact, in trying to discover what the Byzantines normally called the lectionary, we encounter a linguistic confusion that seems to arise from their indifference to any need for a distinction between lectionary and Gospel book. Over centuries, both texts were identified by the same ancient name, evangelion, a name that stressed performance of the reading. ¹²

Yet there seems to come a time when it became either necessary or convenient to have separate designations, and the Byzantines then coined the term tetravangelion to denote the Gospel book; the added prefix necessarily shifts the emphasis away from the performance of the reading to a book divided into four parts. The two terms regularly appear in the lists and inventories that survive in some number from the eleventh century onward, ¹³ just when the Gospel lectionary with a complete calendar of readings became popular.

That the Byzantines would retain the ancient term for the new kind of book, or so often add lectionary tables to copies of the Gospels, should alert us to the petential difficulty of writing the history of the early lectionary. In addition, the centuries-long absence of clear language distinctions suggests that it is impossible to grasp the significance and purpose of the Gospels without at least having some working knowledge of the contemporary lectionary.

Despite the amount of evidence available to us in the form of about 2,200 manuscripts, ¹⁴ the origin and history of the Gospel lectionary remain in considerable doubt.

Regarding the origin of the lectionary, we cannot determine with certainly whether we should speak of a process of development or perhaps something closer to an invention.

According to a recent argument, ¹⁵ the Emperor Justinian (527-65) may have been instrumental in establishing the two-part structure of the service book; in the sixth century and after its development followed an organic process dictated by the growth of the liturgy, itself an expression of evolving doctrine acknowledged through corporate devotion.

Unfortunately, no sixth-century lectionaries survive to confirm the origin. In fact, no lectionary, counting manuscript fragments, can be securely dated before the eighth century. ¹⁶ It is not possible, in other words, to sift through a mass of surviving manuscripts, to place them in chronological order, and then, using these books as evidence, to trace the history of the lectionary and the Eastern church calendar. The early history of service books must be inferred from the lectionary and the Eastern church calendar.

The early history of service books must be inferred from sources whose actual dates range from the eighth to fourteenth century and whose contents are far more restricted than those of the most eleventh-and twelfth-century lectionaries. There exists a handful of manuscripts whose limited number of readings does not seem to satisfy what we could reasonably assume to be the church calendar as it was observed when the manuscripts were made. Some of the books with only a few readings may have been copied from much earlier manuscripts.

The element of doubt that can always exist as to whether a manuscript is a fresh compilation or a copy of what happened to be available allows the surviving lectionaries to be studied by content rather than by their actual dated. Thus a fourteenth-century manuscript can be said to reveal the seventh-century practices, whereas a tenth-century lectionary might reflect the liturgy of the ninth century. Although arranging the manuscripts according to content can create the appearance of an orderly evolution over time and can yield some truly important results, ¹⁷ one hesitates to consider this a wholly satisfactory way to study the surviving material because it demands that we ignore the provenance and chronology of the documents themselves.

Putting the problem into the starkest terms may be helpful: At what point did the number of readings become large enough to justify, in terms of opposing pressures, the production of a separate book? The cost of producing the lectionary for churches that had access to the readings in the form of the Gospel book or New Testament was undoubtedly a factor since the sixth to the ninth century in the East was often marked by economic uncertainty.

In the Gospel book or New Testament, the user had a text that not only encompassed all possible readings but was written in a continuous narrative form that could be used for study and teaching Outside the church service, the lectionary had no value.

When we consider those lectionaries that seem to be out of date because their scribes copied only a small selection of readings, a number of explanations come to mind as ways to account for the apparent anachronism: the book that does not adequately represent the practices of its time because it contains fewer feasts than we know were then celebrated. But celebrated where, and by whom?

For much of its history the Byzantine Empire covered a vast territory. Within it existed varying liturgical practices, some local in nature and others arising from differences between monastic and cathedral observances. The role of chanting may have some bearing on the growth of the lectionary. ¹⁸

Differences in handwriting styles suggest that some of the early manuscripts and surviving fragments were produces over an area that stretched from Italy, through Constantinople, to Palestine. ¹⁹

Calendars may include – and illustration call attention to – celebrations of purely local significance; the handsome tenth-century lectionary at Saint Catherine’s, Sinai. gr. 204, contains a full-page miniature of only one saint, the obscure Peter of Monabata. ²⁰ It is also easy to lose sight of the fact that the Gospels were not the only source of readings in the Byzantine church.

As Justin Martyr states, and as liturgical and direct evidence confirms, ²¹ Old Testament writings provided lessons as well, though we may have reason to believe that some began to fall into disuse between the eighth and tenth centuries, at least in the cathedral services of Constantinople. ²²

The practical argument fails because it cannot account for why lectionary manuscripts begin to appear in the eighth century. It actually leads one to wonder why any were ever made. Since the owner of a Gospel book or New Testament could always add the apparatus and even update it when necessary. ²³ Such notations are found in the famous Codex Ephraemi Rescriptus ²⁴ and in the equally well-known Codex Besae. ²⁵ A ninth-century Gospel book in the Great Lavra, cod. B. 52, ²⁶ even has the ekphonetic notation to guide the reader in church. What we learn from considering the manuscript evidence is that the lectionary and Gospel book were not, in the stark way I have implied, somehow in competition.

Practical considerations having proved to be largely irrelevant, one begins to think of the lectionary as an unusual kind of book. The care with which some of the early examples were written and decorated indicates that this may be so; Gospel lectionaries were apparently considered to be especially sacred collections produced for the most solemn times of the year.

The Vatican lectionary gr. 351, which contains a mere sixteen lessons in addition to morning readings, must be counted amond the finest manuscripts to have been made during the tenth century. ²⁷ Its headpieces and initials bear a close resemblance to the ornament of two Gregory manuscripts written in miniscule: Paris. Coisl. 51 ²⁸ and Patmos cod. 40. ²⁹ As an important manuscript of tenth-century Constantinople, the lectionary Vat. gr. 351 cannot be considered a provincial copy of an out-of-date service manual. The same claim can be made for other Constantinoplitan lectionaries of the late ninth and early tenth centuries, ones like Paris. gr. 278 ³⁰ and the related Vat. gr. 1522. ³¹

One certainty is that the Gospels, regardless of their form, enjoyed a unique status among books. The evidence for this statement is mainly literary, but it can be supplemented by inferences made on the basis of book illumination.

Early Christian and Byzantine authors tell us that the Gospels were an object of particular veneration. Cyril of Alexandria writes that at the Council of Ephesus, held in 431, the Gospels were placed on a throne to signify Christ’s presence. ³²

In the eighth century at Saint Sophia, the deacon incensed the Gospel text on the altar before he removed it to begin the reading; ³³ and, as already noted, Germanos interpreted the opening procession into the church by the deacon, book, and celebrant as Christ’s coming into the world. When, in 787, the Orthodox needed an illustration of the kind of reverence due to icons, they chose the veneration paid to the cross and to the Gospels. ³⁴

It seems clear that the book from which the readings were taken was regarded with particular reverence over the fifth through the eighth centuries and, furthermore, that this reverence often arose from an identification of the manuscript with Christ.

As the dogma of the Incarnation came to be a major issue in eighth-and ninth-century theology the Gospels could provide the congregation with an objective analogue of impeccable orthodoxy, no matter what the beholder’s point of view.

In the passages cited, the vocabulary does not allow us to say with certainty whether a Gospel book or Gospel lectionary is the object understood, and it would not be appropriate to look for precise terminology in such writings; using the technical vocabulary lists and inventories would not only be unnecessary in most literary contexts, it indeed would be graceless. Furthermore, we do not know exactly when the different terms came into widespread use.

The place of the liturgy in Byzantine life may be a factor in the development of the lectionary. In the sixth century, we are told, the people of Constantinople would father enthusiastically around the ambo to touch or kiss the silver-covered Gospels used by the deacon. ³⁵ The growing role of the liturgy in Byzantine piety can be documented in a variety of ways, beginning with the seventh-and eighth-century interpretative treatises of Maximos the Confessor ³⁶ and Germanos.

Popular enthusiasm for the Gospels and their role in the liturgy suggest that we consider the invention of the lectionary as part of the early Byzantine search for the holy within the mundane world. Unfolding over centuries, the sometimes painful debate tended to precise questions that directly or obliquely involved the Divine Liturgy: Did Christ appear in the elements? Did he appear in painted images? As late as the ninth century, the Iconoclast emperors could recall priests censured for mixing the paint from icons with the bread and wine of the service. ³⁷

The content of some early lectionaries suggests an assembly of passages made to encourage a parallel between Christ and the book. The tendency to relate the Four Gospels with Christ Incarnate took on particular significance in the seventh and eighth centuries. The parallel became unmistakable through the selection of events accepted as the most significant in Christ’s life. The decoration of the lectionary tends to support the conclusion that the manuscript was of unique significance.

The evidence from manuscripts earlier than the ninth century is limited but rewards careful sifting and interpretation. Some of the evidence reflects the Byzantine indifference, encountered in the early literary sources, to a distinction between the Gospel book and lectionary.

The Basel Gospels numbers among the most important illuminated manuscripts made prior to the late ninth century, perhaps because it was patterned on the contemporary lectionary. It contains painted headpieces that frame titles written in a style derived from that of monumental inscriptions. Beneath the frame, each Gospel begins with an initial filled with geometric decoration. The readings themselves are marked with elaborate uncial letters. ³⁸ This system of decoration continues in Byzantine art, but in the lectionary, not in the Gospel book. There may be reason to believe that ornamental letters made with birds, fish, and human hands were used in Greek manuscripts before the ninth century. One of the most common initial types, the epsilon, whose cross-stroke is a hand gesturing in speech or holding a pen (Figs. 42. 44) may be of pre-Iconoclastic origin. One such letter is found in a mid-seventh-century Latin sacramentary that shows Greek influence. ³⁹ When holding a pen, the epsilon could easily fall at the beginning of Luke’s Gospel (“Inasmuch as many have undertaken to compile a narrative…”) or be used in the context of authorship: the Evangelist writing the opening word of his Gospel (Fig. 3). The common lectionary incipit “The Lord said…” offers numerous instances, if not for the invention, then at least for the spread of the epsilon with gesturing hand. The design can be found in early lectionaries and throughout those made during the Middle Ages. ⁴⁰ Rich ornament appears to have become a mark of sacred texts in the eighth and early ninth century, during the Iconoclast Controversy. Iconoclast patrons possibly considered book ornament, like the decoration of church buildings with plants and animals, a suitable way to express their piety.

The evidence of surviving ninth-century manuscripts shows that decoration was considered especially appropriate to the lectionary, and it suggests a well-established base from which scribes and illuminators drew in making other kinds of texts in magnificent editions; the epsilon with gesturing hand appears in the Paris Gregory, gr. 510, made for the Emporor Basil I toward the end of the ninth century. ⁴¹

Against the history of the early lectionary, which can only be pieced together from a number of sources, all incomplete, stands that of the medieval lectionary. In studying the MSS of the 11th and 12th centuries, the principal difficulty arises from the daunting amount of source material. The underlying questions posed by the manuscripts are: Why does the lectionary containing a full calendar of readings appear at this time? and What mechanism for the creation? ...

---

Original Footnotes:

1. Pliny: Letters and Panegyricus, II, trans. B. Radice, Loeb Class. Lib. Camb.Mass. 1975, 288-89.

2. Apology I.67: Saint Justin: Apologies, ed. and trans. A. Wartelle, Paris 1987, 190-93.

3. We know this through quotations of the transcript of the bishop's trial in 320. C. Zwissa, S. Optati Milevitani Libri VII, Coprus scriptorum Ecclesiasticorum Latinorum, xxvi, Vienna, 1893, 187; on this matter see also Lepelley, Les cites de l'Afrique romaine au Bas-empire, Paris, 1979, 336-38.

4. T. Matthews, The Early Churches of Constantinople: Archetecture and Liturgy, Univ. Park, Pa., and London 1971, explores the relationship between the service and building forms.

5. Incorporated verbatim into Eusebius's Life of Constantine, 4.36: 1. Heikel, Eusebius Werke, i, Die griechischen christlichen Schriftsteller der ersten drei Jahrhunderte, vii, Leipzig, 1902, 131-32.

6. N. Borgia, Il commentario liturgico di S. Germano patriarca Constantinopolitano, Studi liturgici, i, Grottaferrata, 1912, 21 (sec. 24); see also his comments, 26 (sec. 31) and 27 (sec. 32), that relate the text to Christ and his Incarnation.

7. Universitatsbibliothek, cod. AN.3.12: W. Hatch, The Principal Uncial Manuscripts of the New Testament, Chicago, 1939, pl xxxviii; G. Cavallo, Ricerche sulla maiuscola biblica, Florence, 1967, 107; C. Nordenfalk, Die spatantiken Buchstaben, Stockholm, 1970, 189-94. figs 52-56.

8. Eventually a set of six essentially standard incipits comes into use; the phrases are given in chapter III, note 10.

9. P. Buberl and H. Gerstinger, Die byzantinischen Handschriften, ii. Die Handschriften des X.XVIII. Jahrhunderts, Die illuminierten Handschriften und inkunabeln dir Nationalbibliothek in Wien, viii/4, Leipzig, 1938, 21.

10. That is, ευαγγελικη βιβλος (Lake, Dated Greek MSS, IV, MS 173), and εκλογαδη του ευαγγελιου (VI, MS 266).

11. J. Mateow, Le typikon de la Grande Eglise, II, Le cycle des fetes mobiles, Orientalia Christiana analecta, CLXVI, Rome 1963, 80, 82, 84.

12. G. Lampe, A Patristic Greek Lectionary, Oxford, 1968, s.v.

13. For the two terms used together, one turns to the inventories and donors' lists of the Middle Ages, to, for example, P. Gautier, "Le typikon du sebaste Gregoire Pakourianos," REB, XLII, 1984, 121, or C. Diehl, "Le tresor et la bibliotheque de Patmos au commencement du 13* siecle," Byzantinische Zeitschrift 1, 1892, 514. These sources also provide us with the Byzantine terms for the Saturday-Sunday, σαββατοκυριακον, and daily, καθημερινον , lectionary types.

14. The estimate of K. and B. Aland, The Text of the New Testament: An Introduction to the Critical Editions and to the THeory and Practice of Modern Criticism, trans. E. Rhodes, Grand Rapids, Mich., and Leiden, 1987, 160.

15. Y. Burns, "The Historical Events that Occasioned the Inception of the Gospel Lectionary", XVI, Internationaler Byzantinistenkongress, Akien, II/4, Vienna, 1982, 119-27.

16. Ibid., 125, identifies the single-leaf Paris. suppl. gr. 1155, fol. 19, as the earliest lectionary fragment to survive and dates it in the seventh century. As Burns points out, the small amount of text we have suggests a kind of transition between a Gospel book and an actual lectionary of the medieval type. The script of the manuscript shows Western traits and has been associated, as an 8th century work, with an Italian scriptorium: G. Cavallo, "Funzione e strutture della maiuscole greca tra i secoli VIII-XI", La paleographic grecque et byzantine, Paris, 1977, 106-7. For a leaf, see H. Omont, Facsimiles des plus anciens mannuscrits en onciale et en minscule de la Bibliotheque Nationale du IV* au XII* siecle, Paris 1892, pl. xx bis.

17. Burns (as in note 15.) 124-26. Her analysis of the Vat. Ottob. gr. 175 yields important information on the patriarch's use of the lectionary in Contantinopolitan services prior to the 10th century: see Y. Burns, "The Lectionary of the Patriarch of Constantinople", Studia Patristica, XV/I, ed. E. Livingstone, Texte und Untersuchungen zur Geschichte der altchristlichen Literatur, CXXVIII, Berlin, 1984, 515-20.

18. R. Devreesse, Introduction a l'etude des manuscrits grecs, Paris, 1954, 197-98, notes that texts were chanted from a early period; while generally true, the statement is too comprehensive. As late as the Middle Ages, one of the most important liturgical books, the psalter, was still not uniformly treated; see O. Strunk, "The Byzantine Office at Hagia Sophia", DOP ix-x, 1956, 192.

19. For Italy, see Paris. suppl. gr. 1155, fol. 19: references in note 16 above; for Palestine, see the Sinai. gr. 210 and gr. NE Meg. Perg. 12, dated 861/62; D. Harlfinger, D. Reinsch. J. Sonderkamp, Specimina Sinaitica: Die datierten griechischen Handschriften des Katharinen-Klosters auf dem Berg Sinai, Berlin, 1988, 13-14.

20. K. Weitzmann, Illustrated Manuscripts at St. Catherine's Monastery on Mount Sinai, Collegeville, Minn. 1973, 14, fig 14. The saint's name has various spellings: Monabata, Monobata, and Monabagou.

21. A. Rahlfs, Verzeichnis der griechischen Handschriften des Alten Testaments, Mitteilungen des Septuaginta-Unternehmens..., II, Berlin, 1914, xix-xx.

22. J. Mateos, La celebration de la Parole dans la liturgie byzantine, Orientalia Christiana analecta, CXCI, Rome 1971, 131.

23. The requisite notes might not always appear where we expect; C. Nordenfalk, "Canon Tables on Papyrus", DOP, xxxvi, 1982, 36-37, has called attention to what appear to be lectionary markings on a set of canon table fragments dated to around the second half of the 6th century.

24. Paris. gr. 9: see the leaf published by Cavallo (as above note 7), pl. 82 (with comments on the date, 88-93).

25. Cambridge, Univ. Lib. cod. Nn.2.41: ibid., 75 n. 4, pl. 60; Hatch (as in note 7), pl. xx, and Codex Bezae Cantabrigiensis Quattuor Evangelica et Actis...photypece repraesentatis Camb. 1899. J. Harris, The Annotators of the Codex Bezae, London 1901, discusses some of its later history.

26. B. Metzger, Manuscripts of the Greek Bible, NY, 1981, pl. 24.

27. R. Devreesse, Codices Vaticani graeci II, Codices 330-603,Vat.Cit. 1937, 32-33. Burns (as in note 15). 423; Cavallo (as in note 7), 118, dates thems to the early 11th century and compares its script to those of Bodl. Canon. gr. 92 and Paris. Coisl. 31; the scripts of all three are similar, but the decoration of Vat. gr. 351 cannot be compared withthat of the two early 11th cent. MSS with which Cavallo draws specific comparison.

28. Weitzmann, Byz. Buchmarlerei, II, fig. 54.

29. Ibid., 7 , figs. 30-32.

30. Ibid., 6-7 , figs. 28-29.

31. Ibid., 6 , figs. 25-27.

32. Apologeticus ad Theodosium ed. E. Schwartz, Acta Conciliorum Oecumenicorum, 1/1.3. Berlin, 1927, 83. One's mental image of Cyril's description of hte enthroned Gospels at the Council is inevitably shaped by the anachronistic depiction of the Council of Constantinople (381) in the late 9th cent. Paris.gr.510: Omont, Miniatures, pl.I. The view of J. Crehan, "Patristic Evidence for the INspiration of Councils", Studia Patristica, ix/e, ed. F Cross, Texte und Untersuchungen zur Geschichte der altchristlichen Literatur, xciv, Berlin , 1966, 210-15, that the Gospels represented the Holy Spirit, is without foundation; C Walter, L'iconographic des conciles dans la tradition byzantine Archives de l'Orient chretien, xiii, Paris, 1970, 147-48.

33. As inferred from the order in which Germanos discusses the service: Liturgy Commentary, sec. 30: ed. Borgia (as note 6 above), 25; see Mateos (as in note 22), 135-36.

34. J. Mansi, Sacrorum conciliorum nova et amplissima collectio, xiii, Florence, 1767, 377D-E.

35. As reported by Paul the Silentiary: P. Friedlander, Johannes von Gaza und Paulus Silentiarius, Kunstbeschreibungen justinianischer Zeit, Berlin and Leipzig, 1912, 263, lines 247-48.

36. R. Cantarella, S. Massimo Confessore, La Mistagogia ed altriscritti, Florence, 1931, 122-215.

37. Michael II and Theophilos writing to Louis the Pious: Monumenta Germaniae Historica: Leges, sec. III: Concilia, II/2, ed. A. Werminghoff, Hannover and Leipzig, 1908, 479.

38. Nordenfalk (as in note 7), fig. 53.

39. St. Gall; Stiftsbibliothek, cod. 908: Nordenfalk (as in note 7), 126-28, pl. 53c.

40. For a 9th or 10th cent. example, see Mount Athos, Great Lavra, cod. A>86, fol 114: Weitzmann, Byz. Buchmalerei, fig 321, (also fig. 308, the writing Evangelist type).

41. Ibid., text fig. 1.