Home
  By Author [ A  B  C  D  E  F  G  H  I  J  K  L  M  N  O  P  Q  R  S  T  U  V  W  X  Y  Z |  Other Symbols ]
  By Title [ A  B  C  D  E  F  G  H  I  J  K  L  M  N  O  P  Q  R  S  T  U  V  W  X  Y  Z |  Other Symbols ]
  By Language
all Classics books content using ISYS

Download this book: [ ASCII | HTML | PDF ]

Look for this book on Amazon


We have new books nearly every day.
If you would like a news letter once a week or once a month
fill out this form and we will give you a summary of the books for that week or month by email.

Title: The History, Theory, and Practice of Illuminating - Condensed from 'The Art of Illuminating' by the same - illustrator and author
Author: Wyatt, M. Digby
Language: English
As this book started as an ASCII text book there are no pictures available.


*** Start of this LibraryBlog Digital Book "The History, Theory, and Practice of Illuminating - Condensed from 'The Art of Illuminating' by the same - illustrator and author" ***


Transcriber's note: Text enclosed by underscores is in italics (_italics_).


       *       *       *       *       *



[Illustration]



  THE HISTORY,
  THEORY, AND PRACTICE
  OF
  ILLUMINATING.

  SKETCHED BY
  M. DIGBY WYATT, V.P.R.I.B.A.
  ETC.

  With Illustrations by
  W. R. TYMMS.

  CONDENSED FROM

 "The Art of Illuminating,"
  BY THE
  SAME ILLUSTRATOR AND AUTHOR.

  _London_:
  DAY AND SON, _Lithographers to the Queen_,
  GATE STREET, LINCOLN'S INN FIELDS.



HISTORICAL MANUAL.

LIST OF PLATES.

  PLATE I.--From the fragments of the Bible of Charles the Bald, preserved
  in the British Museum, Harleian 7551. In most of the MSS. of the date of
  Charlemagne and his immediate successors, the ornamental forms are
  generally compounded of Anglo-Saxon and semi-classical details; thus,
  fig. 1 presents us with a lunette, or arch-filling, borrowed from some
  Latin type; while in figs. 3, 4, 5, the interlaced knots, and in figs. 2
  and 6 the "zoomorphic" terminations, are equally characteristic of Celtic
  art. This class of conventional design, although apparently complicated,
  is of comparatively easy execution, and on that account forms a suitable
  style for the young illuminator to try his "'prentice hand" upon.

  PLATE II. gives the outline of the preceding plate, and the beginner may
  make his first attempt at practical illumination in an endeavour to make
  it resemble Plate I. as closely as possible.

  PLATE III., from the same source as Plate I., gives, in figs. 1 and 4,
  two alphabets, and in fig. 2 one sentence, in the characters in which the
  Latin text of the original is written throughout the volume, with the
  usual form of initial letter; together with, in fig. 3, an ornament
  showing, on a largish scale, the principle upon which the most common
  interlacement of the Saxon school is usually worked out.

  It may be here noted that, considering it as likely to be more useful to
  the student, throughout these illuminations the characters, which in the
  originals express Latin, French, or barbarous English, have been arranged
  to exhibit Scripture texts of simple language, such as may be frequently
  desired for the embellishment of churches or schoolrooms.

  PLATE IV., from the British Museum, Reg. 1, c. vii. This manuscript
  consists of the Books of Joshua, Judges, and Ruth, in the Vulgate
  version, with St. Jerome's prologues. It is probably of German execution,
  and is attributed by Sir Frederick Madden to the middle of the 12th
  century. In its illustrations may be recognised a series of good
  specimens of Romanesque forms. In these the scroll may be observed as
  having almost entirely superseded the Carlovingian interlacement, while
  in the foliated ends of the leading stems (more particularly in fig. 1)
  the germination of Gothic is distinctly perceptible. The student will
  scarcely fail to observe how entirely dependent this style of
  illumination is upon the steadiness with which the pen is handled for all
  its charm of expression.

  PLATE V. gives the outline of the preceding plate, to be coloured as a
  lesson in shading with the brush.

  PLATE VI. provides an alphabet of capital letters, some initials, and a
  complete sentence, taken from the same MS. which has furnished materials
  for the two preceding plates.

  PLATE VII. contains fully-coloured examples from the British Museum, Reg.
  1 D, fully described at pages 47 and 48. In these we meet with the
  plenitude of English mediæval illumination,--its freedom of drawing, its
  vigour of colour, its exquisite delicacy, and the general facility of
  design its wayward lines attest. In the best work there is a playfulness
  not to be often found in the productions of the contemporary European
  schools. It needs but little ingenuity to expand such features as those
  which constitute figs. 1, 3, 6, 7, 8, from small book embellishments to
  large _motives_ of elegant mural decoration.

  PLATE VIII. is the corresponding outline to Plate VII.

  PLATE IX. supplies the student with models taken from the same MS. (Reg.
  1 D), of capital and initial letters, and ordinary text, suitable for
  combination with the rich ornaments of Plate VII.

  In this, as in all corresponding plates, the object aimed at is to
  provide forms of lettering, at once tolerably legible, well-proportioned,
  and adapted to harmonize, both historically and artistically, with the
  styles of pictorial illumination given in connection with them. There are
  few faults more common in modern work, or more offensive to the educated
  eye, than the association of styles of lettering and styles of
  ornamentation warring with each other in the proprieties of both time and
  form. Against such it is our earnest object to guard the student, both by
  precept and example.

  PLATE X. furnishes some specimens of the beautiful borderings which
  enrich the pages of that most precious relic of the 15th century, the
  "Bedford Missal," or, more properly speaking, "Book of Hours." This
  exquisite volume has been so fully dwelt upon at pages 55 and 56, that it
  is necessary only to refer the reader to the notice given thereat.

  In the examples collected on this plate, it is manifest that the
  pictorial has not been allowed to usurp the proper province of the
  conventional element of design, as it too frequently does in many works
  of the same period, particularly in Italy,--undoubtedly beautiful as much
  of the illumination of the 15th century was in that country.

  PLATE XI. provides an outline of Plate X., for fully colouring in
  facsimile.

  PLATE XII. shows the general style of the lettering, both capital and
  lower-case, the initials, and the borders which pervade this beautiful
  triumph of Flemish art. Of these, figs. 3, 4, 5, 6, 7, and 8 are
  especially adapted for mural decoration, on, of course, a greatly
  enlarged scale.

  M. D. W.

  37, TAVISTOCK PLACE, LONDON.
          _April, 1861._



TECHNICAL MANUAL.

LIST OF PLATES.

  PLATE I. is from the Speyer Passavant Charlemagne Bible, British Museum
  add. 10,546, described at page 39 of the "Historical Manual." The student
  cannot fail to observe how distinctly legible, and indeed how entirely
  Roman in character, the alphabet of capitals remained so long after the
  Augustan period as the ninth century after Christ. In the lower-case
  letters, in which the text is written, the legibility is evident; but the
  classical derivation through the uncial form of writing is not equally
  manifest. Desiring to guide the beginner in the class of exercises most
  likely to lead him on satisfactorily, we have in this technical manual in
  every case allowed the outline illustrations of each style of writing to
  precede the fully-coloured specimens of the corresponding ornament of the
  leading epochs of the art of illumination; enforcing thereby our
  conviction that the study and practice of that, which of old fell more
  directly within the province of the scribe than within that of the
  miniature painter, should invariably receive the student's first and
  chief attention--for, let it always be remembered that good writing looks
  well if enhanced by little, or the very simplest, ornament, while in
  illumination the effect of the best possible painting is irretrievably
  marred if the writing is irregular and badly formed or spaced.

  PLATE II., from the same precious volume, provides some simple but
  excellent conventional forms, suitable for execution both on a small and
  on an enlarged scale. By repeating either one of the three borders at the
  bottom of the plate, very pretty margins may be produced, suitable either
  for surrounding a page of writing or for enclosing a panel in mural
  decoration.

  PLATE III. is an outline for practising colouring upon. In using these
  outlines, it may be a profitable exercise occasionally to vary the
  colours from those given on the corresponding plate. A comparison of the
  effect produced by the original, and by its copy with variations, will
  tend to fix on the memory of the artist the exact degree of merit of the
  original and of the altered combinations of colour. The practical value
  of an educated eye, no less than of an educated mind, is dependent on the
  force and intelligence of the memory, and every exercise which can assist
  in fixing a fleeting image on the brain is no less efficacious in
  strengthening the one than it can be in developing the other.

  PLATE IV., from the Harleian MS. No. 2,804, gives one of our usual
  exercises upon the main structural features of all illumination--the
  alphabets, initial letters, and small borderings. These, in this case in
  the Romanesque style, have been taken from a very remarkable Bible
  formerly belonging to the church of St. Mary, near Worms. For further
  notice of this and similar volumes, see "Historical Manual," page 43. The
  main use to the student of this class of lesson is to give him steadiness
  of hand in the use of the pen; a word or two of counsel upon which may
  not be altogether unprofitable to him. Firstly, then, let him avoid the
  habit of allowing the pen to touch the paper before he has clearly made
  up his mind where it is to go and when it is to be taken off. An
  ill-directed line instantly reveals a listless mind, and a careful master
  can generally detect the exact points in his work at which the attention
  of a usually diligent pupil has been abstracted from it. Secondly, he
  should never express by half a dozen or more separate strokes forms which
  may be defined by a single continuous line. Thirdly, let him practise
  moving the pen or pencil, not up and down only, but in every direction,
  until equal facility is acquired in drawing spirals from left to right
  and from right to left. Fourthly, it is well to hold the pen or pencil
  nearly vertical, just touching, but scarcely ever pressing heavily on,
  the surface of the drawing. Fifthly, he should by no means aim at dash or
  spirit until he is quite sure that his lines are correct: nothing betrays
  the ill-educated artist more surely and readily to those who know better
  than a bold stroke where a delicate one would be more appropriate, or a
  dark touch in the wrong place. It is the ignorant only who are misled by
  an appearance of _bravura_, vigour, and facility.

  PLATE V., fully coloured from the same source as Plate IV., offers in
  figs. 2, 3, 4, 5, 7, and 8, some easy borders, well adapted for enriching
  string courses or filling in long upright panels or pilasters. The
  initial letters (figs. 1 and 6) are designed with great spirit, and the
  student may profitably amuse himself by endeavouring to invent other
  capital letters made up as these are of apocryphal animals writhing in
  convoluted scrollwork.

  PLATE VI. gives the outline for colouring in _fac simile_ of Plate V.

  PLATE VII., from a Latin Bible of fine English illumination early in the
  fourteenth century (British Museum, Reg. 15, D. 2), corresponds in the
  general character of both its technical and chronological peculiarities
  with those shown on Plates VII., VIII., and IX., of the "Historical
  Manual." The two MSS., however, from which the two series of plates have
  been taken, differ in some material respects, and it will be well for the
  student to practise the leading characteristics of each. One is of
  extraordinary delicacy, the other of great vigour of execution. The
  latter stamps the MS. from which the plate under consideration is taken.
  The student is invited to observe the grace and freedom with which the
  floral terminations of the principal initial (fig. 2) dash away,
  extending to both the top and bottom of the page, and not unfrequently in
  similar examples embracing, as it were, the text on two or more sides.
  (See "Historical Manual," page 47.) Great attention must now be bestowed
  upon the writing; so that neither the true Mediæval character may be
  destroyed, nor so exaggerated as to lose clearness and legibility: a
  little care and dexterity may preserve both.

  PLATE VIII. is intended to draw out all the capabilities of the
  illuminator. Raised, burnished, and engraved, or indented gold, are
  essential to a proper realization of a revival of such old work; and the
  student who would rival in his productions the sober richness of the
  brushes of the artist monks of the fourteenth century, must carefully
  study the combinations of colour given in the "Mappæ Clavicula." Figs. 2,
  5, and 6, offer examples of the tesselated burnished diaper grounds, and
  fillings in, which superseded to a great extent the flat burnished golden
  grounds of earlier dates. Such diapers are little less well adapted for
  walls or ceilings than they are for book decoration. It can be scarcely
  necessary to dwell upon what must be perfectly obvious, the great beauty
  of the initial letters (figs. 1, 3, and 4) given on this sheet.

  PLATE IX. is a careful outline of the above.

  PLATE X., from the Missal of Ferdinand and Isabella (British Museum, Add.
  1851), described at page 57 of the "Historical Manual," introduces us to
  the pictorial, or rather miniature style,--one, which can only be
  excelled in by those who are prepared to devote themselves to painting as
  no longer a decorative, but as essentially a "fine art." Far am I from
  saying that the highest possible art was not brought to bear upon much
  Mediæval illumination; all that I would convey is, that care and neatness
  may produce very respectable reproductions of ordinary ornamental work,
  such as was commonly used during the fourteenth, and early in the
  fifteenth centuries; but that they alone will be found quite inadequate
  to imitate successfully the highly modelled and fully shadowed foliage,
  landscape, architectural groups, and figure subjects, which incessantly
  recur in books illuminated at periods corresponding with the great
  Renaissance of art under the Van Eycks and Memlings of Flanders, the
  Durers of Germany, and the Peruginos, Pinturicchios, and Raffaelles, of
  Italy.

  PLATE XI., fully coloured from the same source as Plate X., can only be
  satisfactorily copied by the student, who may have learnt to shadow with
  the brush from either objects "in the round," or from really good copies,
  either by very great personal devotion and perseverance, or under an
  experienced master.

  PLATE XII. gives a careful outline of the preceding plate.

  M. D. W.

  37, TAVISTOCK PLACE, LONDON.
          _April, 1861._



[Illustration]



PART I.

WHAT ILLUMINATING WAS.

In the following pages an attempt has been made to concentrate into limited
dimensions that which has generally been treated very voluminously. Few
authors, who have tried both, will feel inclined to deny, that it is a much
more difficult task to compress a great subject into a little book
successfully, than it is to expand a little subject into a great book.
Where materials of the highest interest, historically, artistically, and
intellectually, abound, the danger is lest suppression and condensation may
not break the links essential to bind a perspicuous narrative together. I
must, therefore, on these grounds claim the indulgence of the reader, who
may, I trust, be induced by the very imperfections of my story, to recur to
the pages of those more copious and learned writers on the subjects, who
have bestowed upon its elucidation long lives of exemplary and pertinacious
industry.

Before, however, entering on my theme, it is my duty to point out to the
reader, that although, for popular convenience and simplicity, it has been
deemed expedient to divide the history of the Art of Illuminating from the
theory of its use and practice, I have considered that each of the
subdivided parts would be made more valuable by association, and by being
made mutually suggestive and illustrative--by being, in fact, cast as two
parts of one work, rather than as two separate works. I have not,
therefore, hesitated to refer in this, the "Historical Manual," to the
historical interest of plates contained in the technical, nor in the
"Technical Manual" to the technical interest of plates contained in the
historical. Much, indeed, of the matter contained in both should be
considered as common to the two. Thus the ancient technical processes are
no less historically interesting, than they may be likely, by a judicious
revival of such as may be worthy, to prove practically valuable in the
present day. Again, whatever proficiency a student may attain in the
manipulation of his or her drawing, gilding, or painting, it will be in
vain to hope to be enabled to produce a work of art which shall be
satisfactory to the educated eye and taste, until a very considerable
acquaintance has been made with the peculiarities of the various styles in
which our forefathers delighted. No originality can ever be permanently
agreeable which does not discard the precise conventional form of a period,
which is but a mode or transient fashion, in favour of the principles which
pervade all synchronous works of art, and which, transmute them as we may,
must ever remain permanent through all time. Historically we should
remember that miniature ornament of every period reflects on a diminished
scale, and frequently in a highly concentrated form, the leading spirit
which may have pervaded the greater revolutions of monumental art. Owing to
the license which the diminished scale afforded, the imagination of the
artist in these works was restricted by none of those material impediments
which, in the execution of the major monuments of art, protracted the
realization of the changing fashions of the day, frequently until long
after the period when the original impulse may have been communicated to
the art in which those variations were possibly but transient fluctuations.

Thus it is that in these relics of the past may frequently be traced
artistic impulses destined to find no other embodiment than the form in
which they are presented to us in the pages of a manuscript. The
copiousness, then, of such documentary illustrations of the invention of
remote periods is one of the most valuable features of the teaching they
should convey to us. No revival nowadays of any historical style by the
architect can be satisfactory which is not based upon a knowledge, not of
the purely architectural features of the period alone, but of the condition
and characteristics of all those decorative details which distinguished it
as a living reality from the effete and denuded relic which may now only
present itself for our information. Thus even the Saxon and Romanesque
styles of architecture may, through the architect's careful attention to
the decorative features exhibited to us in the pages of ancient illuminated
books, be revived, not in their rude and structural nudity, but as glowing
with those colours, and decorated with those forms, which we may observe as
peculiarly affected in the ornamental and pictorial embellishments of the
best artists of the days when those styles were the only ones popularly
adopted. And not only are the beautiful ornaments and decorative features
of illuminated manuscripts valuable as supplying us with correct
information as to the system of embellishment regarded by the best artists
of each period as harmonizing most perfectly with the structural styles
prevalent in their days; but in the measure of their permanent beauty they
are no less valuable to us as indications of what is excellent for all
time.

Thus, then, they may be used, either as enabling us to restore the most
brilliant features of the historic styles with an accuracy to be acquired
from no other sources of information, or they may be regarded as providing
us with materials for that more extended system of eclectic selection which
must afford the only basis of perfection and originality in any styles
which we may desire now or hereafter to originate; and the origination and
perfection of which we may desire to bequeath to succeeding generations, as
testimonies that, in the nineteenth century, there lived men as capable of
the creation of beauty as any whose happiest inventions are to be found in
the pages of these ancient and most precious volumes.

In opening this historical sketch, I need scarcely recall the facts, that
not only was that which we know as the earliest type of writing the most
pictorial, but that it was also embellished with colour from the most
remote ages. A glance at the pages of Rosselini or Lepsius will suffice to
convince us that the monumental hieroglyphics of the Egyptians were almost
invariably painted with the liveliest tints; and when similar hieroglyphics
were executed on a reduced scale, and in a more cursive form upon papyri,
or scrolls made from the leaves of the papyrus, the common flowering rush
of the Nile, illumination was also employed to make the leading pages more
attractive to the eye. Nor was such illumination peculiar to hieroglyphic
characters; it prevailed also, but not to the same extent, in the hieratic
and demotic modes of writing. Of such papyri notable specimens may be seen
in the British Museum; the most wonderful in existence, however, is the
remarkably interesting and graphic illustration of the funeral of a
Pharaoh, preserved in the Royal Museum at Turin.

Extraordinary dexterity was acquired in a conventional mode of expressing
complicated forms by a few rapid touches, and the life and spirit with
which familiar scenes are represented, and ornaments executed, in both the
early and late papyri, are truly remarkable. The precise extent to which
the Greeks and Romans were indebted to the Egyptians for the origination
and use of alphabetic symbols, the learned have not yet agreed upon. They
have, however, concurred in recognizing the fact that Egypt certainly
supplied the principal materials by means of which writing was ordinarily
practised. The primitive books of the ancients were no other than rolls
formed of papyri, prepared in the following manner:--Two leaves of the rush
were plastered together, usually with the mud of the Nile, in such a
fashion that the fibres of one leaf should cross the fibres of the other at
right angles; the ends of each being then cut off, a square leaf was
obtained, equally capable of resisting fracture when pulled or taken hold
of in any direction. In this form the papyri were exported in great
quantities. In order to form these single leaves into the "scapi," or rolls
of the ancients (the prototypes of the rotuli of the Middle Ages), about
twenty were glued together end to end. The writing was then executed in
parallel columns a few inches wide, running transversely to the length of
the scroll. To each end of the scrolls were attached round staves similar
to those we use for maps. To these staves, strings, known as "umbilici,"
were attached, to the ends of which bullæ or weights were fixed. The books
when rolled up, were bound up with these umbilici, and were generally kept
in cylindrical boxes or capsæ, a term from which the Mediæval "capsula," or
book-cover, was derived. The mode in which the students held the rolls in
order to read from them is well shown in a painting in the house of a
surgeon at Pompeii. One of the staves, with the papyrus rolled round it,
was held in each hand, at a distance apart equal to the width of one or
more of the transverse columns of writing. As soon as the eye was carried
down to the bottom of a column, one hand rolled up and the other unrolled
sufficient of the papyrus to bring a fresh column opposite to the reader's
eye, and so on until the whole was wound round one of the staves, when, of
course, the student had arrived at the end of his book. Eumenes, king of
Pergamus, being unable to procure the Egyptian papyrus, through the
jealousy of one of the Ptolemies, who occupied himself in forming a rival
library to the one which subsequently became so celebrated at Pergamus,
introduced the use of parchment properly "dressed" for taking ink and
pigments; and hence the derivation of the word "pergamena" as applied to
parchment or vellum; the former substance being the prepared skin of sheep,
and the latter of calves.[1]

The sheets of parchment were joined end to end, as the sheets of papyrus
had been, and when written upon, on one side only, and in narrow columns
across the breadth of the scroll, were rolled up round staves and bound
with strings, to which seals of wax were occasionally attached, in place of
the more common leaden bullæ.

The custom of dividing books into pages is said by Suetonius to have been
introduced by Julius Cæsar, whose letters to the Senate were so made up,
and after whose time the practice became usual for all documents either
addressed to, or issuing from, that body, or the emperors. As that form
subsequently crept into general use, the books were known as "codices;" and
hence the ordinary term as applied to manuscript volumes. All classes of
books, the reeds for writing in them, the inkstands, and the "capsæ" or
"scrinia," the boxes in which the "scapi" or rolls were kept, are minutely
portrayed in ancient wall-paintings and ivory diptychs. The inkstands are
generally shown as double, no doubt for containing both black and red ink,
with the latter of which certain portions of the text were written.[2]

Nearly two thousand actual rolls were discovered at Herculaneum, of course
in a highly-carbonized condition, and of them some hundreds have been
unrolled. None appear to have been embellished with illumination;[3] so
that for proof of the practice of the art in classical times, we are thrown
back upon the classical authors themselves. The allusions in their writings
to the employment of red and black ink are frequent. Martial, in his first
epistle, points out the bookseller's shop opposite the Julian Forum, in
which his works may be obtained "smoothed with pumice-stone and decorated
with purple." Seneca mentions books ornamented "cum imaginibus." Varro is
related by Pliny to have illustrated his works by likenesses of more than
seven hundred illustrious persons. Pliny again informs us that writers on
medicine gave representations in their treatises of the plants which they
described. Martial dwells on the editions of Virgil, with his portrait as a
frontispiece. The earliest recorded instance of the richer adornments of
golden lettering on purple or rose-stained vellum, is given by Julius
Capitolinus in his life of the Emperor Maximinus the younger. He therein
mentions that the mother of the emperor presented to him, on his return to
his tutor (early in the 3rd century), a copy of the works of Homer, written
in gold upon purple vellum. Whether derived from Egypt or the East, this
luxurious mode of embellishment appears to have been popular among the
later Greeks, a class of whose scribes were denominated "writers in gold."
From Greece it was, no doubt, transplanted to Rome, where, from about the
2nd century, it, at first slowly, and ultimately rapidly, acquired
popularity. St. Jerome, indeed, writing in the 4th century, in a well-known
passage in his preface to the Book of Job, exclaims:--"Habeant qui volunt
veteres libros vel in membranis purpureis auro argentoque descriptos vel
uncialibus, ut vulgo aiunt, literis, onera magis exarata quam codices;
dummodo mihi meisque permittant pauperes habere scedulas, et non tam
pulchros codices quam emendatos."[4]

This almost pathetic appeal of the great commentator was scarcely necessary
to assure us that such sumptuous volumes were executed for the rich alone,
since the value of the gold and vellum, irrespective of the labour
employed, must necessarily have taken them, as he indicates, altogether out
of the reach of the poor. Evidence indeed is not wanting, that many of the
Fathers of the Church laboured with their own hands to supply themselves
with writings, which no golden letters or purpled vellums could make more
valuable to them or their primitive followers: thus, Pamphilus, the martyr,
who suffered in the year 309, possessed, in his own handwriting,
twenty-five stitched books, containing the works of Origen. St. Ambrose,
St. Fulgentius, and others, themselves transcribed many volumes, precious
to themselves and most edifying to the faithful. Whatever ornaments or
pictures these volumes contained, no doubt reproduced the style of art
fostered, if not engendered, in the Catacombs.

Roman illuminated manuscripts would appear, therefore, to have been mainly
divisible into two classes; firstly, those in which the text, simply but
elegantly written in perfectly-formed, or rustic (that is, inclined)
capitals, mainly in black and sparingly in red ink, was illustrated by
pictures, usually square, inserted in simple frames, generally of a red
border only; and secondly, the richer kind, in which at first gold letters,
on white and stained vellum grounds, and subsequently black and coloured
letters and ornaments on gold grounds, were introduced. The first of these
appears to have been the most ancient style, and to have long remained
popular in the Western Empire, while the second, which, as Sir Frederick
Madden has observed, no doubt came originally to the Romans from the
Greeks, acquired its greatest perfection under the early emperors of the
East.

Of both styles there are still extant some invaluable specimens, which,
although not of the finest periods of art, may still be regarded as typical
of masterpieces which may have existed, and which fire or flood, Goth or
Vandal, may have destroyed. Before proceeding, however, to an enumeration
of any of these, it may be well to define certain terms which must be
employed to designate the peculiarities of character in which the different
texts were written, some slight knowledge of which is of great assistance
in arriving at a proximate knowledge of the dates at which they may have
been executed. Such a definition cannot be more succinctly given than in
the following passage, extracted from Mr. Noel Humphrey's interesting work
"On the Origin and Progress of the Art of Writing:"--[5]

"Nearly all the principal methods of ancient writing may be divided into
square capitals, rounded capitals, and cursive letters; the square capitals
being termed simply _capitals_, the rounded capitals _uncials_, and the
small letters, or such as had changed their form during the creation of a
running hand, _minuscule_. Capitals are, strictly speaking, such letters as
retain the earliest settled form of an alphabet; being generally of such
angular shapes as could conveniently be carved on wood or stone, or
engraved in metal, to be stamped on coins. The earliest Latin MSS. known
are written entirely in capitals, like inscriptions in metal or marble.

"The uncial letters, as they are termed, appear to have arisen as writing
on papyrus or vellum became common, when many of the straight lines of the
capitals, in that kind of writing, gradually acquired a _curved_ form, to
facilitate their more rapid execution. However this may be, from the 6th to
the 8th, or even 10th century, these uncials or partly-rounded capitals
prevail.

"The modern minuscule, differing from the ancient cursive character,
appears to have arisen in the following manner. During the 6th and 7th
centuries, a kind of transition style prevailed in Italy and some other
parts of Europe, the letters composing which have been termed
_semi-uncials_, which, in a further transition, became more like those of
the old Roman cursive. This manner, when definitively formed, became what
is now termed the minuscule manner; it began to prevail over uncials in a
certain class of MSS. about the 8th century, and towards the 10th its
general use was, with few exceptions, established. It is said to have been
occasionally used as early as the 5th century; but I am unable to cite an
authentic existing monument. The Psalter of Alfred the Great, written in
the 9th century, is in a small Roman cursive hand, which has induced Casley
to consider it the work of some Italian ecclesiastic."

To return from this digression on the character of ancient handwriting, to
the examples still extant of the two great sections into which the
manuscripts of classical ages may be divided, I would observe, that, first
in importance and interest of the first class may certainly be reckoned the
Vatican square Virgil with miniatures, which has been referred by many of
the best palæographers to the 3rd century. It is written throughout in
majuscule Roman capitals, which, although MM. Champollion and Sylvestre[6]
describe them as of an "elegant but careless form," appeared to me, when I
examined the volume minutely in 1846,[7] to exhibit great care and
regularity. The miniatures, many engravings from drawings traced from which
are given in D'Agincourt's "Histoire de l'Art par les Monuments,"[8] are
altogether classical, both in design and in the technical handling of the
colours, which are applied with a free brush, and apparently in the true
antique manner, _i.e._, with scarcely any previous or finishing outline.
These miniatures have also been engraved by Pietro Santo Bartoli, but not
with his usual accuracy of style. A complete set of coloured tracings made
by him are in the British Museum (Lansdowne Coll.), but they even are not
quite satisfactory. The Terence of the Vatican, which is without
miniatures, is in a somewhat similar writing, and belongs to about the same
period. The third in importance of the ancient Vatican manuscripts of this
class, is in the rustic instead of elegant capital lettering, and is
supposed to be of the 5th century; certainly not later. It is a Virgil,
decorated throughout with pictures executed in apparent imitation of the
square Virgil, but in a much more barbarous and lifeless style.[9] From an
entry of the 13th century contained in the volume,[10] and from our
knowledge of its having been long and at a remote period, preserved in
France, it would appear to have belonged to the Parisian monastery of St.
Denis, if not to the Saint himself.

So far as antiquity, irrespective of merit in point of illumination is
concerned, the most remarkable ancient Roman manuscript[11] existing
belongs to the curious class known as "Palimpsests," or books from which
the colouring matter of an original writing has been discharged, in order
to prepare the vellum for receiving an altogether different text, the
latter being generally written at right angles to the former.[12] This
precious document is the celebrated treatise "de Republicâ," by Cicero,
written in uncial characters, evidently in an Augustan period, and was
discovered by Cardinal Angelo Mai, under a copy of St. Augustine's
Commentary on the Psalms, made previous to the 10th century.

The Ambrosian Library at Milan contains a codex of Homer, of equal
antiquity with the Cicero, with fifty-eight pictures, much in the style of
the Vatican square Virgil. This important MS. has been commented upon by
the same distinguished antiquary.[13]

The Vienna Roman calendar, supposed to have been executed in the 4th
century, and embellished with eight allegorical figures of the months, is
both an early and very important specimen of Roman illumination, not only
on account of the elegance and dexterous execution of these figures, but
because it is the most ancient manuscript in which anything like ornament,
independent of pictured illustration of the author's text, is introduced.
Of little less note in the history of art, is the celebrated Dioscorides of
the same imperial library, the date of which is fixed by the fact of its
being enriched with a very graceful portrait of the Empress Juliana Anicia,
for whom it is known to have been written at the commencement of the 6th
century. Both Lambecius[14] and D'Agincourt give various facsimiles
(omitting colour) of the fine illustrations which decorate this remarkable
volume.

Another 5th century Virgil of remarkable purity in the text, although
without miniatures, is the well-known "Medicean" of the Laurentian Library
at Florence. The Paris Prudentius, in elegant rustic capitals of the 6th
century, is another fine codex of the same type. There are, in addition to
those already cited, various other early texts of the classics contained in
the different public libraries of Europe; and it is singular to remark,
that (so far as I have been able to ascertain) none of them are embellished
with those richer decorations, which appear to have been reserved after the
end of the 5th century, for the great text-books of the Christian, and more
particularly of the Eastern Church. Of these sacred volumes, that which is
generally supposed to be the oldest complete version of the Bible in
Greek,[15] is the Codex Alexandrinus of the British Museum, attributed, by
consent of all the best Palæographers, to the commencement of the 5th
century. It is without gold altogether, and has no other illumination than
the occasional contrast of red and black inks, and a line slightly
flourished, at the close of each book.[16] The next fragment of the
Scriptures, in point of probable date, is the once celebrated Cottonian
Genesis, or at least its ghost; for unfortunately a few charred and
shrunken fragments are all that have been saved from the disastrous fire
which destroyed so many of Sir Robert Cotton's precious volumes in 1731. In
its original state, as we know from several collations made previous to the
fire, it contained, on 165 pages, no less than 250 miniatures, each about
four inches square. Astle[17] has given a facsimile of a page, which, on
comparison with the existing shrivelled fragments, proves that in their
present state they are just about one half their original size. The
paintings are in all respects antique, and correspond in general character
with contemporary secular miniatures. Dr. Waagen[18] remarks that "only the
hatched gold upon the borders, the glories, and the lights on the crimson
mantle indicate the commencement of Byzantine art." The great rival to the
"Codex Cottonianus Geneseos" is the "Codex Vindobonensis Geneseos," which
consists of twenty-six leaves with eighty-eight miniatures. It forms one of
the four great lions of the Vienna Imperial Library. These two remarkable
versions of Genesis are supposed to be of nearly equal date, and correspond
as to the character of the truly antique miniatures very fairly; the fact,
however, of the text of the English version being in black ink with very
regularly-formed letters, while that of the Vienna one is, for the most
part, written in gold and silver, and in less evenly-distributed
characters, induces a fair presumption in favour of the greater antiquity
of the Cottonian fragments. In the more gorgeous details of the Vienna
Genesis, coupled with its square and unadorned classic pictures, we may
thus clearly recognize the transition from our first or Latin class of
ancient illumination, to our second or purely Byzantine style. We
especially designate this class as "Byzantine," because as art in
illumination, as in all other branches, declined in the seven-hilled city,
it rose in the seat of empire founded in the East by the first great
Christian emperor. It is true that ideal art degenerated almost
contemporaneously in the capitals of both empires; but in decorative art,
at least, there can be no question but that Byzantium gained, as Rome lost,
ground. The former no doubt drew fresh inspiration from her close
intercourse with the Persian and other nations of the East, while the
latter was content to produce little, and that little in slavish
reminiscence of the past. Italy no doubt fed the earliest monastic
libraries of Western Europe with the quantities of texts of ancient authors
we know them to have contained; but we may fairly assume those texts to
have been but rarely illustrated, since the original styles of illumination
produced in those countries to which the classic volumes travelled, would
unquestionably have betrayed an antique influence more strongly than they
did, had the means of deriving that influence been brought copiously within
their reach.

I proceed now to a slight notice of the second class of ancient codices,
that on which the ultimate splendour of the Byzantine school was founded.
Fortunately, time has spared to our days several brilliant specimens of the
richest of these quasi-classic manuscripts. Of such, the principal are, as
Sir Frederick Madden observes,[19] "the celebrated Codex Argenteus of
Ulphilas, written in silver and gold letters on a purple ground, about A.D.
360, which is, perhaps, the most ancient existing specimen of this
magnificent mode of caligraphy; after it, may be instanced the copy of
Genesis at Vienna," already mentioned, the Psalter of St. Germain des Prés,
at Paris, and the fragment of the New Testament in the Cottonian Library,
Titus C. xv., all executed in the 5th and 6th centuries.

The first-named of these contains, on about 160 leaves, a considerable
portion of the four gospels, and is now preserved in the Royal Library of
Upsal, in Sweden. It is the earliest version of any part of the sacred
writings in the Moesogothic or ancient Wallachian dialect.[20] The second
of Sir Frederick Madden's notabilities has been alluded to as of transition
character. The third, the Psalter of St. Germain des Prés, is ascribed by
M. Champollion Figeac, who has given a portion of it in coloured facsimile
in the "Moyen Age et la Renaissance"[21] to the 6th century. It is
unquestionably a beautiful specimen of gold writing on purple; but neither
in the size of the letters nor in the ample spacing of the lines, will it
bear comparison with the, no doubt, earlier example, the Cottonian, Titus
C. xv. Our greatest authority upon all matters connected with early
illuminated versions of the Holy Scriptures, Mr. Westwood, remarks, in
speaking of this last-named manuscript, that "Codices purpureo-argentei are
much rarer than those in golden writing, the latter material being used not
only on purple, but also on white vellum; whereas the silver letters would
not easily be legible except on a dark ground. The writing is in very large
and massive Greek uncials; the words denoting God, Father, Jesus, Lord,
Son, and Saviour, being, for dignity's sake, written in golden letters. The
colour of the stain has faded into a dingy reddish purple, and the silver
is greatly tarnished and turned black. This fragment is stated by Horne to
be one of the oldest (if not the most ancient) manuscripts of any part of
the New Testament that is extant, and is generally acknowledged to have
been executed at the end of the 4th, or, at the latest, at the beginning of
the 5th century."

The Vienna gold, silver, and purple Gospels, the lettering of which
corresponds closely with that last described, may be regarded as certainly
next in importance, and are of about equal antiquity. In none of these
relics of magnificence are we enabled to trace the Eastern or Persian
influence, which unquestionably imported a previously unknown originality
and character into the art of Byzantium during the reign of Justinian the
Great, A.D. 527 to 565. It is, no doubt, true, as Dr. Waagen remarks,[22]
that "the style of painting up to his time, both in conception, form, and
colour, was much the same as that which has been preserved to us in the
paintings at Pompeii; while the spirit of Christianity, operating upon the
artistic Greek nature, stimulated it anew to beautiful and original
inventions. In a few single instances this style of art was maintained
until the 10th century; but, generally speaking, a gradual degeneracy
ensued, which may be dated from Justinian's period. The proportions of the
figures gradually became exaggerated, elongated, the forms contracted with
excessive meagreness, the motives of the drapery grew paltry, appearing
either in narrow parallel folds stiffly drawn together, or so overladen
with barbaric pearls and jewels as to exclude all indication of form. The
flesh assumed a dark tone, the other colours became heavy, gaudy, and hard;
while in glories, hatchings, and grounds, gold was called into requisition.
In these qualities, united to a gloomy and ascetic character of heads,
consist the elements of the Byzantine school." But, on the other hand, it
is ever to be remembered that the mortification of the old flesh was but a
symptom of the more active life beneath it, sloughing off the Pagan
tradition, and gradually replacing it by that new and healthy Christian
vigour which, for many centuries, nourished and aided the northern and
western nations of Europe in their efforts to organize those national
styles of Christian Art which are commonly designated as Gothic.[23]

To return to Justinian, and his direct influence on the change of style
which took place during his reign, it may be noted as a curious fact, that
the year in which the great Church of Sta. Sophia was commenced was the
very year in which he concluded an eternal peace with Chosroes Nushirvan,
king of Persia. In one or two reigns antecedent to his, Greek artists had
been employed in Persia, and there had been a friendly communication
between the two countries. It may be therefore assumed, that when Justinian
proposed to build this structure in so short a time, he not only enlisted
the ability of those about him, but that he recalled those straying Greeks
who had gone to seek their fortunes in other countries. He most likely,
indeed, employed not only his own subjects, but foreigners; and in that way
probably a considerable portion of what no one can fail to recognize as
Oriental Art, was mixed with that known as Byzantine. Certain it is that in
many of the mosaic ornaments of Sta. Sophia a very marked Oriental
character is still to be traced.

On a close comparison of these mosaics[24] with the unique Eusebian Canons
on an entirely gold ground, two leaves of which, painted on both sides, are
preserved in the British Museum (Addit. No. 5111),[25] the student will
certainly, I think, be induced rather to agree with Sir Frederick Madden,
in ascribing them to the 6th century, than with Dr. Waagen, who considers
that they "can scarcely be older than the 9th century." To the practical
illuminator, these fragments are of far higher importance than all the
others to which we have as yet alluded, since, while of equal archæological
interest, they constitute the earliest specimens from which really
decorative illumination can be studied.[26]

Another illustration of the Eastern influence brought to bear upon
Christian manuscripts of the age of Justinian, is furnished by the
celebrated Syriac Gospels of the 6th century, written in the year 586
(one-and-twenty years after the emperor's death), by Rabula, a scribe in
the monastery of St. John, in Zagba, a city of Mesopotamia, and now
preserved in the Laurentian Library at Florence. Mr. Westwood regards this
as "so important a manuscript in respect to the history of the arts of
illumination and design in the East," and by reflection in the West, that
he is induced[27] to give an elaborate description of its embellishments,
from which the following is a short extract:--

"The first illumination represents Christ and the twelve apostles seated in
a circle, with three lamps burning beneath a wide arch supported by two
plain columns, with foliated capitals, and with two birds at the top. The
second illumination represents the Virgin and Child standing within a
double arch, the columns supporting which are tessellated, and the upper
arch with several rows of zigzags, and peacocks standing at the top. The
third represents Eusebius and Ammonius standing beneath a kind of tent-like
canopy, supported by three columns, with undulated ornament, two peacocks
with expanded tails standing at the top. The nineteen following plates are
occupied by the tables of the Eusebian Canons, arranged in columns, between
pillars supporting rounded arches, generally enclosed between larger and
more ornamented columns supporting a large rounded arch, on the outsides of
which are represented various groups of figures illustrating scriptural
texts, plants, and birds. In some of these, however, the smaller arches are
of the horseshoe character. The capitals are, for the most part, foliated;
but in one or two they are composed of human faces, and a few of birds'
heads. The arches, as well as the columns by which they are supported, are
ornamented with chevrons, lozenges, nebules, quatrefoils, zigzags, flowers,
fruit, birds, &c.; many of which singularly resemble those found in early
Anglo-Saxon manuscripts, especially in the columns supporting the Eusebian
Canons in the purple Latin Gospels of the British Museum (MS. Reg. I. E.
6). There is, however, none of the singular interlacing of the patterns so
characteristic of the Anglo-Saxon and Irish manuscripts."

I have dwelt thus in detail upon these Greek pictorial and decorative
features, because there can be no doubt that the exportation of books so
adorned, by the early missionaries, who carried Christianity and a degree
of civilization to the Northern and Western countries, supplied the
original types from which, however barbaric the imitations, the first
attempts were made to rival, in the extreme West, the arts and spiritual
graces of the East. On this plea, I hope I may be pardoned for dwelling yet
further upon some of the leading distinctions between the Byzantine and
Latin (that is, between the Eastern and Western) modes of working out
religious conceptions, which were, that in the Western or Latin mode
symbolism was universal; the art of the Catacombs was followed distinctly,
though frequently remotely, developing itself in mythical and sentimental
forms, and systems of parallelism between type and prototype. In the Greek
Church, the exposition of faith, through art, took a more tangible form.
Symbolism was avoided on all possible occasions, and the direct
representation of sacred themes led to a partial transfer to the
representation of the adoration due to the thing represented. Iconoclasm
was the reaction to this abuse. In the advanced periods of Greek art, this
realistic tendency led to a painful view of the nature of religion, more
particularly in connection with the martyrdom of saints, and the physical
sufferings of our Saviour and his followers, which are frequently
represented in the most positive and repellent forms.

Long, however, before Byzantine Art had time to deviate much from its
ancient traditions, and even while it maintained an easy supremacy over the
Western empire, the Lombard kingdom, and all the Visi- and Moeso-Gothic and
Frankish races, a formidable competitor for the leadership in the Art of
Illumination had sprung up in the extreme West, in the island homes of the
Celtic races.

It is not necessary now to prove, what historians have freely admitted,
that Ireland was certainly christianized for a long time previous to the
date of the mission of Augustine to England. The disputes which arose
between the followers of that saint and the Irish priests, so soon as they
clearly apprehended the nature of the supremacy claimed by the Church of
Rome, assure us of their early isolation in the Christian world. Even in
their, at first entire, and ultimately partial, rejection of the Vulgate
text of the Gospels, and their retention of the older versions, from which
no doubt their formulas of faith were derived, they steadily maintained
their Ecclesiastical freedom from the dogmatism of Rome. As their creed was
independent, so was their Art original; nothing resembling it can be traced
previous to it.

Before proceeding to examine the precise form assumed by this "original
art," it may be well to remind the student that, with the exception of a
few manuscripts decorated in the style of the Laurentian Syriac Gospels and
the British Museum golden fragments, the general character of the
decoration of all writings, previous to the origination of the Celtic style
in Ireland, had been limited to the use of different-coloured, golden, and
silver inks, on stained purple and white vellum grounds, to the occasional
enlargement of, and slight flourishing about, initial letters; to the
introduction of pictures, generally square, or oblong, enclosed in plain,
or slightly bordered, frames; and, occasionally, to the scattering about,
throughout the volumes, of a few lines and scrolls. Let us now see--in the
words of Mr. Westwood, who has done more than any previous writer had done
to vindicate the honour of the Irish school of caligraphy[28]--what
features of novelty it was mainly reserved for that school to originate.
"Its peculiarities,"[29] he states, "consist in the illumination of the
first page of each of the Sacred Books,--the letters of the first few
words, and more especially the initial, being represented of a very large
size, and highly ornamented in patterns of the most intricate design, with
marginal rows of red dots; the classical Acanthus being never represented.
The principles of these most elaborate ornaments are, however, but few in
number, and may be reduced to the four following:--1st. One or more narrow
ribbons, diagonally but symmetrically interlaced, forming an endless
variety of patterns. 2nd. One, two, or three slender spiral lines, coiling
one within another till they meet in the centre of the circle, their
opposite ends going off to other circles. 3rd. A vast variety of lacertine
animals and birds, hideously attenuated, and coiled one within another,
with their tails, tongues, and top-knots forming long narrow ribbons
irregularly interlaced. 4th. A series of diagonal lines, forming various
kinds of Chinese-like patterns. These ornaments are generally introduced
into small compartments, a number of which are arranged so as to form the
large initial letters and borders, or tessellated pages, with which the
finest manuscripts are decorated. The Irish missionaries brought their
national style of art with them from Iona to Lindisfarne in the 7th
century, as well as their fine, large, very characteristic style of
writing; and as these were adopted by their Anglo-Saxon converts, and as
most of the manuscripts which have been hitherto described are of
Anglo-Saxon origin, it has been the practice to give the name of
Anglo-Saxon to this style of art. Thus several of the finest facsimiles
given by Astle as Anglo-Saxon, are from Irish manuscripts; and thus
Sylvestre, who has copied them (without acknowledgment), has fallen into
the same error; whilst Wanley, Casley, and others, appear never to have had
a suspicion of the existence of an ancient school of art in Ireland."

The monks of Iona, under the great Irish saint and scribe Columba, or
Columbkill, and their Anglo-Saxon disciples at Lindisfarne, under his
friend St. Aidan, together with the Irish monks at Glastonbury, spread
Celtic ornament in England, from whence it had, to a great extent, retired
with the expulsion of the ancient British. St. Boniface, the principal
awakener of Germany to Christianity, carried with him his
singularly-ornamented book of Gospels, which is still preserved as a relic
at Fulda. Similar evidence of the transmission of the Art prevalent during
the early centuries of the Church in Ireland, to other lands, by means of
the missionaries who left her shores, is to be found in the books of St.
Kilian, the apostle of Franconia, still preserved at Wurtzburg; in those of
St. Gall, now in the public library of St. Gall, in the canton of
Switzerland which still bears his name; and in the very important series,
of which Muratori has given an interesting catalogue, connected with the
monastic institution founded by St. Columbanus, at Bobbio, in Italy, and
now principally in the Ambrosian Library at Milan. Many of these pious men
were themselves scribes, and their autograph copies of the Holy Gospels are
still in existence, with the name of the writers, in some cases,
identifying the volumes, and absolutely fixing their date. Thus we have the
Gospels of St. Columba, the Leabhar Dhimma, or Gospels of St. Dhimma
MacNathi, and the MacRegol Gospels in the Bodleian Library. All of these
are anterior to the 9th century, and are distinguished by an elaborate
style of ornament unlike any other European type. The extent of influence
exercised by these eminent men and the "Episcopi Vagantes," or
missionaries, is strongly insisted upon by M. Libri, unquestionably one of
the most eminent and correctly-informed bibliographers of the present day.
Speaking of the latitudinarianism of some among these Christian men, he
observes, "No doubt certain pious but narrow minds hoped to open the door
to ecclesiastical literature only; but the exclusion sometimes pronounced
against the classics was never general amongst writers who, even in their
rudeness, always showed themselves imitators of antiquity. Thus we find
that the celebrated manuscript of Livy, in the Imperial Library at Vienna,
belonged to Sutbert, an Irish monk, one of those wandering bishops who,
towards the close of the seventh century, had gone to preach Christianity,
and, as it would seem also, to teach Roman history in Belgium. One cannot
help remarking, that the most celebrated of these pious missionaries, St.
Columbanus, laid the foundations at Luxeuil in France, at St. Gall in
Switzerland, and at Bobbio in Italy, of three monasteries which afterwards
became famous for their admirable manuscripts, in many of which the
influence of the Irish and Anglo-Saxon schools can be recognized at a
glance. The library of St. Gall is too celebrated to require mention. The
Bobbio manuscripts are known everywhere by the discoveries which have been
made in the _palimpsests_ which once belonged to that collection. As for
the manuscripts of Luxeuil, they have been dispersed; but the specimens of
them which are to be found in the Libri collection, joined to what has been
published on the subject by Mabillon, O'Conor, and others, prove
unanswerably that in this abbey, as well as in that of Stavelot in Belgium,
and other ancient monasteries on the Continent, a school of writing and
_miniature_ had sprung up as remarkable for the beauty of its caligraphy,
as for the care applied to reproduce the forms of the Anglo-Irish
schools."[30]

In delicacy of handling, and minute but faultless execution, the whole
range of palæography offers nothing comparable to these early Irish
manuscripts, and those produced in the same style in England. When in
Dublin, some years ago, I had the opportunity of studying very carefully
the most marvellous of all--"The Book of Kells;" some of the ornaments of
which I attempted to copy, but broke down in despair. Of this very book,
Mr. Westwood examined the pages, as I did, for hours together, without ever
detecting a false line, or an irregular interlacement. In one space of
about a quarter of an inch superficial, he counted, with a
magnifying-glass, no less than one hundred and fifty-eight interlacements,
of a slender ribbon pattern, formed of white lines, edged by black ones,
upon a black ground. No wonder that tradition should allege that these
unerring lines should have been traced by angels.[31] However "angelic" the
ornaments may be, but little can be said in favour of the figure subjects
occasionally introduced. In some manuscripts, such as the Book of Kells, in
pose and motive it is generally obvious that some ancient model has been
held in view; but nothing can be more barbaric than the imitation; while in
the other specimens, such as the so-called autograph Gospels of St.
Columba, or Columbkill, who died A.D. 594, two years before the advent of
St. Augustine--the Book of St. Chads, or the Gospels of MacRegol--no such
evidence of imitation is to be met with, and the figures are altogether
abortive.

I was enabled some years ago, by the kindness of the Rev. J. H. Todd, the
learned librarian of Trinity College, Dublin, to compare the so-called
autograph Gospels of St. Columba, with the Book of Kells, which is
traditionally supposed to have belonged to that saint, and remained
strongly impressed with the superior antiquity of the former to the latter.
The one may have been his property, and the other illuminated in his honour
after his death, as was the case with the Gospels of St. Cuthbert. In none
of them, at any period, were shadows represented otherwise than by apparent
inlayings under the eyes and beside the nose; and yet, at the same time,
the ornaments were most intricate, and often very beautiful, both in form
and colour. The purple stain is frequently introduced, and is of excellent
quality; but gold appears, so far as I have been able to observe, only in
the Durham Book, and in that even most sparingly.[32] It is the most
celebrated production of the Anglo-Hibernian monastery of Lindisfarne,
founded by St. Aidan and the Irish monks of Iona, or Icolumkille, in the
year 634.

St. Cuthbert, who was made bishop of Lindisfarne in 685, was renowned as
well for his piety as for his learning; he died in 698, and, as a monument
to his memory, his successor, Bishop Eadfrith, caused to be written this
noble volume, generally called the Durham Book, and known also as St.
Cuthbert's Gospels, now in the British Museum. This manuscript, surpassed
in grandeur only by the Book of Kells, in the same style, was greatly
enriched by Æthelwald, bishop of Lindisfarne, who succeeded Eadfrith in
721, and caused St. Cuthbert's book to be richly illuminated by the hermit
Bilfrith, who prefixed an elaborate painting of an Evangelist to each of
the four Gospels, and also illuminated the capital letters at the
commencement of each book. The bishop caused the whole to be encased in a
splendid binding of gold, set with precious stones; and in 950, a priest
named Aldred rendered the book still more valuable by interlining it with a
Saxon version of the original manuscript, which is the Latin text of St.
Jerome.

Want of space alone prevents our following Simeon of Durham in his touching
narrative of the circumstances which attended the translation of this
volume, together with the body of the much-loved saint, to Durham
Cathedral, in which both were long and profoundly venerated. The peculiar
importance of this volume in the history of Illumination, consists in its
clearly establishing, by its coincidence with earlier examples, the class
of caligraphy practised by that primitive Church[33] and people, to whom
Gregory the Great despatched St. Augustine, at the end of the 6th century.
With the mission, which reached its destination, and effected the
conversion of Ethelbert and of many of his subjects, in the year 597,
Gregory forwarded certain sacred volumes, of which the following were long
preserved with the greatest veneration:--A Bible in two volumes; two
Psalters; two books of the Gospels; a book of Martyrology; apocryphal Lives
of the Apostles, and expositions of certain Epistles and Gospels.

The first--the Bible--which was beautifully written on purple and
rose-coloured leaves, with rubricated capitals, was certainly in existence
in the reign of James I. Mr. Westwood ("Palæographia Sacra," 1843-45) looks
upon the magnificent purple Latin Gospels of the British Museum (Royal
Library, 1 E 6) as "no other than the remains of the Gregorian Bible." In
this, with the utmost respect for his opinion, I cannot concur, since the
fragment exhibits far too many genuine Saxon features to have been possibly
executed in the Eastern or Western empires previous to the date of the
mission of St. Augustine. That it may have been produced in this country,
in imitation of the more classical original, by the immediate followers of
the saint, is, I consider, highly probable.

The second--the two Psalters--have disappeared. Several learned men have
indeed looked upon the British Museum Cottonian MSS. Vesp. A 1, as one of
these celebrated books, but, as I venture to think, erroneously; for it is
difficult to believe that ornaments, so entirely of the Anglo-Irish school
of Lindisfarne, as those which decorate this volume, could have been
executed at Rome during either the 6th or even the 7th century. Nothing is
more probable than that, out of the forty persons who are believed to have
constituted Augustine's mission, several should have been skilled, as most
ecclesiastics then were, in writing and in the embellishment of books; and
in any school, established by St. Augustine for the multiplication of those
precious volumes, without which ministrations and teachings in consonance
with Roman dogmas could not be carried on in the new churches and monastic
institutions founded among the converts, it is most likely that the native
scribes, on their conversion, should be employed to write and decorate the
holy texts, with every ornament excepting those of a pictorial nature. In
the execution of these, they could scarcely prove themselves as skilful as
the followers of St. Augustine would, from their retention of some
classical traditions, be likely to be. Thus, and thus only, as I believe,
can we account for the singular combination of semi-antique with Saxon
writing, and of Latin body-colour pictures, executed almost entirely with
the brush, and regularly shadowed (such as David with his Attendants, in
the frontispiece to the Vespasian A 1 Psalter), with ornaments of an
absolutely different character, such as the arch and pilaster which form
the framework for this very picture of King David. Another argument, which
weighs greatly in my mind against the probability of such a Psalter as
Vespasian A 1 being a prototype, is the fact, that the Utrecht and Harleian
Psalters, to both of which I shall have occasion again to allude, in their
pictorial illustrations, present us with evident copies, in outline, of
some classic coloured original; just, in fact, of such a manuscript of the
Psalms as the celebrated Vatican Roll[34] is of the book of Joshua. What
more likely than that one of the venerated Psalters brought from Rome
should have been such a manuscript, and should have been the very one
copied in the case of the Utrecht Psalter, in the "rustic capitals" of the
original, and in the later Harleian replica in the current Saxon uncial?

As respects the third class of Augustinian books--the Gospels--the case is
far different; for the accredited and traditional originals are, in every
respect, such as would be likely to have been produced at Rome or at
Constantinople, but most probably the former, during the pontificate of
Gregory the Great. Fragments of the most important of these Gospels are
preserved in the library of Corpus Christi College, Cambridge. They are
written in black ink generally, with occasional lines in red, in the
ancient manner. Two pages only of illuminations are left, though it is
evident that the volume once contained a large and complete series. The
most important of these represents St. Luke, clad in tunic and toga, seated
under just such a triumphal arch as is frequently to be met with in the
Roman Mosaics of the 5th and 6th centuries.[35] The second illuminated page
comprises a series of small square pictures, framed round with the simple
red line of the oldest Latin manuscripts.

The other Augustinian fragmentary Gospel is to be found among the Hatton
manuscripts in the Bodleian Library, Oxford: it is without any other
illumination than the contrast of red and black ink, and a few ornaments
about some of the initial letters. The evidence, upon which it may be
assumed that these volumes were either brought to this country by St.
Augustine, or formed some of the "codices multos,"[36] sent by Gregory the
Great to the mission on its establishment, rests not only upon the
antiquity and purely Latin character of the fragments, but on the fact that
both Gospels contain entries in Saxon of upwards of one thousand years old,
connecting them with the library of the Abbey of St. Augustine at
Canterbury; and, furthermore, they correspond with the description given by
a monk of that monastery, who, writing in the reign of Henry V., dwells
upon the "primitie librorum totius Ecclesie Anglicane" preserved in that
library, as the very Gospels in the version of St. Jerome, brought to
England by St. Augustine himself.

The Martyrology, the apocryphal Lives of the Apostles, and the Expositions
which completed the series, cannot be now identified.

To rapidly multiply copies of these text-books of the Church of Rome, was,
no doubt, one of the first and most important duties of the monks of
Canterbury; and from the traces we may detect in various manuscripts of the
Anglo-Saxon mode of writing and ornamenting writing, combined with
paintings such as the Anglo-Saxons were incompetent to execute for some
time after the close of the 6th century, we may safely infer that the monks
both worked themselves and largely employed the native scribes. Thus, as
Mr. Westwood observes in a recent article in the "Archæological Journal,"
"We have sufficient evidence that, soon after the settlement of the
followers of St. Augustine, there must have been established a
_scriptorium_, where some of the most beautiful manuscripts were written in
the purest uncial or rustic capitals, but decorated with initials in the
Anglo-Saxon or Irish style. Of such MSS. we can now record--

"1. The purple Gospels at Stockholm, written in very large uncials, but
with illuminated title-pages, with pure Anglo-Saxon ornaments, and grand
figures of the Evangelists in a mixed classical and Anglo-Saxon style.

"2. The Utrecht Gospels.

"3. The Gospels in the Cathedral Library, Durham; Astle's 'Origin and
Progress of Writing,' pl. 14, fig. B, p. 83.

"4. The Utrecht Psalter.

"5. The so-called Psalter of St. Augustine, MSS. Cotton., Vespasian, A 1;
Astle, pl. 9, fig. 2.

"6. The Bodleian MS. of the Rule of St. Benedict, Lord Hatton's MSS., No.
93; Astle, pl. 9, fig. 1, p. 82.

"Were it not for the initials, and other illuminations in the genuine
Anglo-Saxon style, not one of these MSS. could be supposed to have been
executed in England. They are, nevertheless, among the finest specimens of
early caligraphic art in existence."

One of the most important of this interesting class of manuscripts is,
unquestionably, that of the Psalms, now preserved in the public library at
Utrecht. It was formerly in the possession of Sir Robert Cotton, and should
be now with the rest of his library in the British Museum. The volume
contains, besides the Psalms, the "Pusillus eram," the Credo, and the
Canticles, with a few leaves from the Gospel of St. Matthew. It is written
upon vellum, and each psalm has a pen-and-ink illustration, in the same
style as those in the Harleian Psalter, No. 603, which was written in the
10th century; and similar also to those in the Cambridge Psalter of the
12th century. The writing in the Utrecht Psalter is executed in Roman
rustic capitals; it is arranged in three columns in each page; and the
elegance with which the letters are formed, would place the manuscripts
amongst those of the 6th or 7th century: but the illustrations before
mentioned, with the large uncial B, heightened with gold, in the Saxon
interlaced style, which commences the first psalm, would give it a later
date, certainly not earlier than the 7th or 8th century; and the
pen-and-ink drawings were probably executed a century later.

Mr. Westwood, to whose highly interesting "Archæological Notes of a Tour in
Denmark, Prussia, and Holland," published in the "Archæological Journal," I
am indebted for the above information, tells us that the date of the few
pages of the Gospel, mentioned as being bound up in this volume, is as
uncertain as that of the Psalter; the text being written in a style which
would place it amongst the works of the 6th or 7th century, whilst the word
"Liber," with which it commences, is written in large square Roman
capitals, in gold, with the remains of ornament similar to that in
Vespasian, A 1.

That which gives, however, its greatest value to the Utrecht Psalter, is
the remarkable freedom and cleverness of the pen-and-ink drawings with
which it is embellished. In them may be recognized, I believe, the earliest
trace of those peculiar fluttered draperies, elongated proportions, and
flourished touches, which became almost a distinct style in later
Anglo-Saxon illumination. So different is it, both from the Anglo-Hibernian
work, prevalent in England up to the advent of St. Augustine, and from the
contemporary imitation of the antique, practised by Byzantine, Latin,
Lombard, or Frankish illuminators, that the conclusion seems, as it were,
forced upon us, that it can have been originated in no other way than by
setting the already most skilful penman, but altogether ignorant artist, to
reproduce, as he best could, the freely-painted miniatures of the books,
sacred and profane, imported, as we know, in abundance, from Rome, during
the 7th and 8th centuries.

To so great an extent do antique types and features prevail in the earlier
specimens of this class of Anglo-Saxon volumes, that, until comparatively
recently, the catalogue of the Utrecht Library has designated the
illustrations of the Psalter now under notice, as evidently productions of
the reign of Valentinian;[37] while the outline subjects, in a similar
style, and of considerably later date, which are introduced in the British
Museum "Aratus," were attributed, by even Mr. Ottley's critical judgment,
to a somewhat similar period.

The Harleian Psalter (No. 603), to which allusion has been already made,
although written in later characters, is decorated with many pictures, all
but identical with those in the Utrecht manuscript, thereby demonstrating,
with comparative certainty, that both were taken from some popular
prototype, possibly one of the Augustinian Psalters already alluded to.[38]

The Bodleian Cædmon's, or pseudo-Cædmon's, "Metrical Paraphrase of the Book
of Genesis," written and illustrated in outline,[39] during the 10th or
11th century, and the Ælfric's Heptateuch of the British Museum,
"Cottonian, Claudius B iv.," of a somewhat later date, afford excellent
illustrations of the enduring popularity of this peculiar mode of
outline-drawing. The striking difference may, however, be noted between
these later and the earlier specimens in the same style, that whereas the
types of the latter are, with scarcely any exception, antique, those of the
former are comparatively original, and exhibit that strong inclination to
caricature, which has always formed one of the leading features of English
illumination.

While, in this class of Anglo-Saxon manuscripts, the influence of Latin art
may be traced on the original Hiberno-British school of scribes, a
corresponding change was effected, through the introduction into this
country of specimens of the more brilliant examples of Byzantine execution
or derivation. Thus, as Sir Frederick Madden observes,[40] "The taste for
gold and purple manuscripts seems only to have reached England at the close
of the 7th century, when Wilfrid, archbishop of York, enriched his church
with a copy of the Gospels thus adorned; and it is described by his
biographer, Eddius (who lived at that period or shortly after), as
'inauditum ante seculis nostris quoddam miraculum,'--almost a miracle, and
before that time unheard of in this part of the world. But in the 8th and
9th centuries the art of staining the vellum appears to have declined, and
the colour is no longer the same bright and beautiful purple, violet, or
rose-colour of the preceding centuries. It is rare also to meet with a
volume stained throughout; the artist contenting himself with colouring a
certain portion, such as the title, preface, or canon of the mass.
Manuscripts written in letters of gold, on white vellum, are chiefly
confined to the 8th, 9th, and 10th centuries. Of these, the Bible and Hours
of Charles the Bald, preserved in the Royal Library at Paris, and the
Gospels of the Harleian collection, No. 2788, are probably the finest
examples extant. In England, the art of writing in gold seems to have been
but imperfectly understood in early times, and the instances of it very
uncommon. Indeed, the only remarkable one that occurs of it is the charter
of King Edgar to the new minster at Winchester, in the year 966. This
volume is written throughout in gold."

Although but few books were thus gorgeously written, many were sumptuously
decorated; and, indeed, there exist no more brilliant volumes than some of
those produced by Anglo-Saxon scribes. Of these several have been
preserved; but if two or three only are noticed, it will be quite
sufficient to establish the leading characteristics of the school, which
appears to have been organized under Ethelwold, bishop of Winchester, at
New Minster, or Hyde Abbey, near Winchester, during the 10th century. The
names of several leading masters of that great nursery of illumination have
been handed down to us. Thus Ethric and Wulfric--monks--are recorded as
having been "painters;" but Godemann is spoken of as the greatest of all.
Fortunately, a magnificent specimen of his art is preserved in the
celebrated benedictional of St. Ethelwold, in the library of the Duke of
Devonshire, and engraved _in extenso_, with great care, in the
twenty-fourth volume of the "Archæologia."[41] This is one of the most
sumptuous manuscripts which has been executed in any age by any scribe, and
differs widely from the Anglo-Saxon MSS. previously described. The text is
generally enclosed within a rich framework, formed by wide and solid bars
of gold, about and over which twine and break elegantly-shaded masses of
conventional foliation. In the initial letters, and occasionally in the
ornament, the peculiarly Saxon interlacing and knotwork is retained; but in
most of the embellishments, a reaction can be traced from the Carlovingian
manuscripts themselves, originally acted upon, as will be hereafter seen,
by the Saxon school of caligraphy.[42] The figure subjects in this volume
are cramped in style and action, exhibit but little classical influence,
and possess, as a leading merit, only a singularly sustained brilliancy of
tint and even execution throughout.

Next to this great masterpiece, and from the same fountain-head, come the
following, several of which are exceedingly beautiful:--The two Rouen
Gospels; the Gospels of King Canute, in the British Museum, Reg. D 9; the
Cottonian Psalter, Tib. C vi.; the Hyde Abbey Book, lately in the Stowe
Library; and the Gospels of Trinity College, Cambridge. The ornaments in
all these volumes are painted in thick body-colours, and with a vehicle so
viscid in texture, that Dr. Dibdin[43] infers from its character, as
evidenced in the Benedictional, "the possibility or even probability of oil
being mixed up in the colours of the more ancient illuminations." In this
opinion I do not concur, as I believe the peculiar body and gloss of the
pigment to be produced by the use of white of egg.

If the character of Anglo-Saxon architecture and sculpture agreed with the
representations of both given in the Benedictional of Ethelwold--as I have
every reason to believe it did--it must have been massive and elaborate in
the highest degree; and there is no reason to suppose that a people who
were capable of drawing so well as they assuredly could, should have
limited their productions in the sister arts to the rude and clumsy, long
and short, and other similar work, which we are in the habit of supposing,
characterized all their principal productions.

I have dwelt in some detail upon Saxon illumination, for two reasons:
firstly, because it is a theme on which some national self-gratulation may
be justifiably entertained;[44] and, secondly, because it is one on which,
although much has been written, comparatively little light has as yet been
thrown. Before leaving it, however, some general observations should be
made upon the classes of books most in demand, and the means by which they
were multiplied in this country; and, indeed, with slight local
differences, on the great continent of Europe as well,--Byzantium, Ravenna,
Rome, Monte Cassino, Subiaco, Paris, Tours, Limoges, Arles, Soissons,
Blois, Aix-la-Chapelle, Cologne, Hildesheim, Worms, Treves, Glastonbury,
Canterbury, Winchester, York, Durham, Lindisfarne, Wearmouth, Jarrow,
Croyland, and Peterborough, being the great centres of production.

From the earliest period religious zeal was much shown in its offerings to
the Church, by laymen, more or less pious,--the least pious being, in fact,
sometimes the most liberal donors,--and very large sums were expended in
illuminating and ornamenting manuscripts for that purpose. Many of these
books were remarkable for the extreme beauty of the paintings and
ornamental letters enriched with gold and silver, which decorate them, as
well as for the execution of the writing, the most precious bindings
frequently adding greatly to their cost. Gospels, books of anthems, and
missals, were most frequently chosen for such gifts; but they were not
confined to sacred subjects, including occasionally the best writings of
Greece and Rome, which were eagerly sought after as models of eloquence,
and, still more, as often being supposed to contain prophecies of the
coming of Christ, and proofs of the truth of his doctrines.

The piety of individuals often led them to expend large sums in the
preparation of their offerings to the Church; the finest and best parchment
which could be procured being used for manuscripts. When black ink was used
in liturgical writings, the title-page and heads of the chapters were
written in _red ink_; whence comes the term Rubric. Green, blue, and yellow
inks were used, sometimes for words, but chiefly for ornamental capital
letters; the writers and miniature-painters exercising their own taste and
judgment in the decoration, and heightening its effect with gold and the
most expensive colours, such as azure and the purest cinnabar or vermilion.

The greater part of these works were intrusted to monks and their clerks,
who were exhorted, by the rules of their order, to learn writing, and to
persevere in the work of copying manuscripts, as being one most acceptable
to God; those who could not write being recommended to learn to bind books.
Alcuin entreats all to employ themselves in copying books, saying, "It is a
most meritorious work, more useful to the health than working in the
fields, which profits only a man's body, whilst the labour of a copyist
profits his soul."[45]

Home production could, however, by no means suffice to multiply books, and
especially religious books, with sufficient rapidity to satisfy the eager
demand for them. Long journeys appear to have been taken to foreign
countries, by learned ecclesiastics, for scarcely any other purpose than
the collection of manuscripts; while quantities were imported into England
from abroad. Thus Bede tells us, that Wilfrid, bishop of Wearmouth and
Jarrow, and Acca, Wilfrid's successor, collected many books abroad for
their libraries, at the end of the seventh century. Thus Theodore of Tarsus
brought back an extensive library of Grecian and Roman authors, on his
return to Canterbury, in 668, from a mission to Rome; and thus, as we are
told by Mr. Maitland,[46] when "Aldhelm, who became Bishop of Schireburn in
the year 705, went to Canterbury, to be consecrated by his old friend and
companion Berthwold (pariter literis studuerant, pariterque viam religionis
triverant), the archbishop kept him there many days, taking counsel with
him about the affairs of his diocese. Hearing of the arrival of the ships
at Dover during this time, he went there to inspect their unloading, and to
see if they had brought anything in his way (_si quid forte commodum
ecclesiastico usui attidissent nautæ qui e Gallico sinu in Angliam,
provecti librorum copiam apportassent_). Among many other books, he saw one
containing the whole of the Old and New Testaments, which he at length
bought: and William of Malmesbury, who wrote his life in the twelfth
century, tells us it was still preserved at that place."

How deeply must all lovers of illumination regret the infinite destruction
of books that has prevailed in all ages! Of all this "librorum copiam," how
few survive. Even in the days of Alfred the Great, the Danes had destroyed
the majority of them; for, as that great royal Bibliomaniac exclaims, in
his preface to the "Pastoral of Gregory,"--"I saw, before _all_ were
spoiled and burnt, how the churches throughout Britain were filled with
treasures and books."

I now leave our own country for a while, and return to the general
continent of Europe; having, I trust, satisfactorily established the
individuality of those three great styles of illumination, from the fusion
of which the Romanesque, and ultimately the Mediæval, system sprang,--viz.,
the Roman, or pictorial; the Greek, or golden; and the Hiberno-Saxon, or
intricate. The commencement of that fusion has been traced in the later
Anglo-Saxon work, and it now remains to observe the circumstances under
which a similar, and even more marked, amalgamation took place on the
continent, under the auspices of Charlemagne, the greatest patron of the
art who ever lived.

Much has been assumed by early Palæographers, and even some recent ones,
with respect to the influence exercised by the Lombard MSS. executed
between the establishment of the Lombard kingdom in the year 568, and its
absorption A.D. 774, in the empire of Charlemagne, on the class of
illumination introduced under his auspices; but the specimens which have
descended to these days exhibit such an entire decrepitude of style, as to
justify the belief that, with the exception of a peculiar broken-backed
letter, known as "Lombard brisé," the Lombards themselves contributed
little or nothing to the results which attended the efforts made by that
great sovereign to raise the art of book-decoration in his day to its
highest pitch. It was mainly by the aid, and through the direct
instrumentality, of the learned Anglo-Saxon, Alcuin, that Charlemagne
carried out his laudable design. This industrious ecclesiastic, who was
born in the year 735, received his education under Egbert and Elbert,
successive archbishops of the see of York,--having been appointed at an
early age "custodian" to the library collected by the former. On the death
of Elbert, he was sent to Rome to receive the pallium of investiture for
the new archbishop Enbalde. On his journey home, in 780, he passed through
Parma, where Charlemagne happened to be at the time. The consequence of
their meeting in that city was, that Alcuin received and accepted an
invitation to take up his residence at the court of the Frankish sovereign.
During four-and-twenty years, until his death, indeed, in 804, he retained
the affection and respect of his royal patron, and occupied himself in
incessant labour for the advancement of learning, and the multiplication of
pure texts of the Holy Scriptures and other good books. Several of Alcuin's
letters to Charlemagne are still extant, in which the supremacy of the
English schools and libraries is distinctly recognized, as well as the
direct influence exercised by them on Frankish literature, and, as in those
days literature and illumination were inseparable, on illumination also.
Thus, in one place he begs his master to give him "those exquisite books of
erudition which I had in my own country by the good and devout industry of
my master Egbert, the archbishop." Again, referring to the same "treasures
of wisdom," he proposes,--"If it shall please your wisdom, I will send some
of our boys, who may copy from thence whatever is necessary, and carry back
into France the flowers of Britain; that the garden may not be shut up in
York, but the fruits of it may be placed in the paradise of Tours."

One of the evidences of the eagerness with which this task of multiplying
the sources of learning was carried on, is to be found in the attempts made
to abridge and expedite labour. Thus, as M. Chassant[47] observes in his
useful little manual of abbreviations[48] used during the Middle Ages, the
texts of all documents of importance were comparatively free from
contractions from the period when Justinian the Great banished them, by an
imperial edict, from all legal instruments, until the accession of
Charlemagne, "during whose reign, either to save time or vellum, the
scribes revived the ancient Roman practice of using initials, and
frequently arbitrary signs, to represent whole words of frequent
recurrence."

It is, however, in the quality, rather than the quantity, of Carlovingian
MSS. that the reader is most likely to be interested; and I therefore
hasten to note two or three of the most imposing specimens. The earliest of
the grand class is believed to be the Evangelistiarium, long preserved in
the Abbey of St. Servin, at Toulouse, and ultimately presented to Napoleon
I., on the baptism of the King of Rome, in the name of the city. From
contemporary entries, it appears to have been completed, after eight years'
labour, in the year 781, by the scribe Godescalc. Of whatever nation
"Godescalc" may have been, the volume[49] exhibits far too many composite
features to justify the belief that any one individual, or even many
individuals of one nation, could have executed the whole. The paintings are
probably by an Italian hand, being executed freely with the brush, in
opaque colours, in the antique manner. Many of the golden borders are quite
Greek in style, while the initial letters, and others of the borders, are
thoroughly Hiberno-Saxon. A nearly similar dissection would apply to most
of the manuscripts executed for Charlemagne's descendants, to the third
generation. The volume contains 127 leaves, every leaf, not entirely filled
with illumination, being stained purple, with a white margin, and covered
with a text, written in golden initials, in two columns, separated by very
graceful and delicately-executed borders. Our plates, Technical Manual,
Nos. 1 and 2, taken from the great Charlemagne Bible of the British Museum,
give a good idea of the nature of the ornament usually employed in similar
MSS. to fill up such borders and to form and decorate initial letters. They
will serve to show also the common type of the Alphabets in use.

From Charlemagne's "Scriptorium," which was no doubt the head-quarters of
the best artists of all nations in his time, proceeded many other volumes
of scarcely less interest and magnificence. Among these, the most
noteworthy are, the Gospels of St. Medard de Soissons,[50] so called
because believed to have been presented by Charlemagne to that Abbey;[51]
the Vienna Psalter, written for Pope Hadrian; the Gospels preserved in the
library of the Arsenal at Paris, and formerly belonging to the Abbaye of
St. Martin des Champs;[52] the Gospels found upon the knees of the Emperor
on opening his tomb at Aix-la-Chapelle; the Harleian MS. No. 3788, known as
the "Codex Aureus";[53] and last, not least, the Bible, known as that of
San Calisto, preserved in the Benedictine monastery of that saint at Rome,
and formerly in the monastery of San Paolo fuori le Mura. The frontispiece
to this volume, which is no less than one foot four inches high, by one
foot one inch wide, represents a sitting emperor holding a globe, on which
are inscribed various letters, arranged in the peculiar form adopted by
Charlemagne in his signs manual.

The learned have disputed hotly whether this portrait is intended for that
of Charlemagne, or of Charles the Bald, his grandson. Whether this
manuscript, which, in all respects, except beauty in the figure-subjects, I
look upon as the finest I have ever seen, was executed in the days of the
former or latter monarch, is of no very great moment, as its leading
features would harmonize very well with accredited reliques of either. It
still contains no less than 339 pages, and is one blaze of illumination
from the first page to the last.[54] The large initial letters are quite
Saxon in form; the borders, of which there are endless and beautiful
varieties, are more strictly classic in character than is usual in Caroline
manuscripts; and the pictures are in an indeterminate style, between Greek,
Latin, and that original Frankish, which subsequently absorbed in Western
Europe all previous tradition, and grew into the peculiar type of French
12th century work--the progenitor of the pure Gothic of the 13th.

Ample materials happily exist for tracing the gradual development of this
Frankish element; at first through the works of the immediate descendants
of Charlemagne, and subsequently through various liturgical works,
collected from suppressed abbeys, and preserved for the most part in the
Imperial library at Paris. Of these, some of the most important are, the
Bible of Louis le Debonnaire, executed in the eighth year of his reign; the
Gospels of the same monarch; and the Sacramentaire de Metz,--all produced
for sons of Charlemagne. The first-named is of the barbaric style, on which
Alcuin and others improved; the second contains some very curious symbolic
initial letters; and the third, a good deal of originality, both in
ornaments and figures.

The principal volumes still preserved, once belonging to the grandsons of
Charlemagne, appear less original in several respects, than do those
executed for his sons. Thus, in the case of Louis le Debonnaire's eldest
son Lothaire, whose Gospels, written and decorated at the Abbey of St.
Martin, at Tours, exhibit a mixed Latin and Saxon style, with but little
specifically Frankish work,--and thus also in the person of Lothaire's
youngest brother, Charles the Bald, whose two celebrated Bibles, the one
known as the Bible of St. Denis, and the other as that presented to the
monarch by Count Vivien, abbot of the same monastery at which the Gospels
of Lothaire were executed,--illustrate a similar composite, but scarcely
original, style. The former manuscript is illuminated with intertwined
lacertine monsters, knotwork, single (but not the three-whorl) spirals, and
rows of red dots following many of the leading outlines, all of which may
be regarded as distinctive features of the Hiberno-Saxon school; while the
latter, with several of the above peculiarities freely introduced, combines
an unmistakable classicality, shown in the various figure-subjects, and
especially in the arcading which encloses the Eusebian Canons at the
commencement of the volume.

The British Museum is fortunate in possessing in the Harleian MS.--No.
7551--a curious collection of ancient Biblical fragments, and amongst these
are a few pages taken from a Bible executed for Charles the Bald. From
these Mr. Tymms has selected the elegant Alphabets, initial letters, and
ornaments which are to be found in plates 1, 2, and 3, of this manual. In
these the student will not fail to recognize what he may have already
observed in studying the specimens given from the Charlemagne Bible
(Technical manual, plates 1, 2, and 3), that while the form of the text and
the ornamental borderings are founded on antique models, the initial
letters scarcely ever fail to exhibit in their Celtic animals' heads and
interlaced strap-work the influence of Alcuin and the Saxon scribes.

We can feel but little surprise at the production of such works at the
Abbey of St. Martin, at Tours, for it was within the walls of that
"Paradise," as Alcuin calls it, that the Saxon sage gave all the latter
years of his life to the recension of the Holy Scriptures,[55] and to the
organization of a "scriptorium" worthy of his affectionate patron.

The impulse given to the Art of Illumination in that celebrated
establishment was speedily communicated to rival scriptoria in other
localities; thus from the abbeys of St. Martial, at Limoges, from Metz,
Mans, St. Majour in Provence, Rheims, St. Germain and St. Denis at Paris,
issued, from the age of Charlemagne to the 13th century, an almost
uninterrupted series of highly-illuminated volumes, many of which still
remain to attest the vigorous efforts by which the foreign elements were
gradually thrown aside in France, to make way for that expressive and
original outline style[56] which achieved its greatest power in the early
part of the 13th century. The throes and struggles by which this was
achieved, are singularly well shown by a page engraved in Count Bastard's
splendid work from the "Apocalypse of St. Sever," written during the first
half of the 11th century. The page presents a curious emblematical
frontispiece, the general form of which is perfectly Oriental; the border
ornaments are founded on Cufic inscriptions; the animals which decorate the
Arabian framework are classical; and the interlacing fretwork of several
portions of the design is purely Saxon.

Many Byzantine features were brought into French illumination through the
schools at St. Martial's and the other abbeys of Limoges, but it was at
Paris itself that the greatest changes and improvements were effected;
thus, at St. Germain and St. Denis were produced, during the first half and
middle of the 11th century, two volumes, still existing in the Imperial
Library of France, which distinctly show the budding of "Gothic." The St.
Germain "Mysteries of the Life of Christ" is illustrated by many original
and very spirited outline compositions, some of which are slightly
coloured; while the "Missal of St. Denis," of a few years later, displays
that peculiar grace and _naïveté_ in the action and expression of the
figures, together with that soft elegance in foliated ornament, which for
several centuries remained a dominant excellence in the best French
illuminations.

As classical tradition and Hiberno-Saxon intricacies died out in France to
make way for the true Mediæval styles, so did they, although somewhat more
slowly, in England, Germany, Spain, and the Netherlands. In Italy, a
degeneracy occurred, from which the revival at length, under Cimabue and
Giotto, was as rapid and brilliant as the previous collapse appears to have
been fatal.[57]

Alike from any such complete change, complete degeneracy, or ultimate
attainment of life and perfection, the genuine Greek style of the Byzantine
empire was exempted. That Oriental splendour of gold and colour by which so
early as the days of Justinian the Great, it sought to gloss over the
feebleness of its reminiscences of classical beauty, remained the unchanged
leading characteristic of its illuminations down to the final extinction of
the empire in 1453.

In such an essay as the present, it is quite impossible to convey any idea
of the minute, but extremely interesting varieties of type adopted in
Byzantine manuscripts; it must suffice to state, in general terms,--that
the dispersion of many of the most skilful Greek artists, by the
iconoclastic emperors (commencing with Leo the Isaurian, A.D. 726), gave a
great impetus to the arts of design in those countries in which they took
refuge, and no doubt contributed specially to the improvements effected
under Charlemagne,--that on the abandonment of such religious persecutions,
in the middle of the 9th century, a fresh start appears to have been
taken,[58]--and that from the date of that revival, which may be specially
noted under the reign of Basil the Macedonian, until about the year 1200,
many very noble and dignified pictures[59] were executed. From the
last-named era, until the taking of Constantinople by the Turks, although
the treatment of figure-subjects became more and more weak and mannered,
much beautiful ornament was painted upon gold grounds, and the influence
originally communicated to Arabian art from the Eastern Empire, was
reflected back upon its later productions from the contemporary schools of
Saracenic and Moorish decoration.[60] It is scarcely necessary to remark,
that in all these inflexions of style the Russian, Syrian, and Armenian
illuminators closely followed the example set them by the Byzantine scribes
and painters.

Returning from the East to the extreme west of Europe, it is worthy of note
how entirely the primitive Saxon styles, which wrought so important an
influence upon the rest of Europe, were lost in the country from which they
had been mainly promulgated. The successive social and political changes
wrought by the ascendancy of the Danes, and ultimately of the Normans, put
an almost total stop to Saxon illumination; and so complete was the
abandonment of the Saxon character, that Ingulphus, in describing the fire
which destroyed the noble library of his abbey at Croyland, in the year
1091, after dwelling on the splendour of the "chirographs written in the
Roman character, adorned with golden crosses and most beautiful paintings,"
and especially "the privileges of the kings of Mercia, the most ancient and
the best, in like manner beautifully executed with golden illuminations,
but written in the Saxon character," goes on to state: "All our documents
of this kind, greater and less, were about four hundred in number; and in
one moment of a most dismal night, they were destroyed and lost to us by
this lamentable misfortune. A few years before, I had taken from our
archives a good many chirographs, written in the Saxon character, because
we had duplicates, and in some cases triplicates, of them; and had given
them to our cantor, Master Fulmar, to be kept in the cloister, to help the
juniors to learn the Saxon character, because that letter had for a long
while been despised and neglected by reason of the Normans,[61] and was now
known only to a few of the more aged; that so the younger ones, being
instructed to read this character, might be more competent to use the
documents of their monastery against their adversaries in their old age."

The Normans, a warlike but unlettered race, did but little for the first
century after the Conquest, to restore the taste for learning which they
and the Danes had displaced. While English progress in illumination was
thus comparatively paralyzed, in France and Germany new styles,
corresponding with those known in architecture as Romanesque, rapidly
sprang into popularity. Of the leading decorative features of such styles,
as well as of the corresponding alphabets and initial letters, we have
endeavoured to give some elegant reductions in plates 4, 5, and 6 of this
manual, and in plates 4, 5, and 6 of its technical companion. The
illustrations in the former have been taken from the British Museum, "Reg.
1, C. VII.," a folio MS., of bold rather than beautiful execution, but
containing throughout many well-designed initial letters and ornaments. The
volume comprises the vulgate version of the books of Joshua, Judges, and
Ruth, and it is believed by Sir Frederick Madden, no doubt the most
competent judge in this country, to have been executed about the middle of
the 12th century. The materials for the plates in the Technical Manual,
Nos. 4, 5, and 6 have been gleaned from a manuscript of rather later date,
preserved in the Harleian Collection, Nos. 2803 and 2804. There can be
little doubt that the numerous ornaments which decorate this great Bible
were the work of German industry, for, independently of the evidences of
style offered by the writing and illumination, an entry in the volume
informs us that it once belonged to the church of the Blessed Virgin, in
one of the suburbs of Worms. All of the ornaments in this series of
illustrations show a manifest disposition on the part of their designers to
break away from the rigidity of pure convention into a class of foliation,
which, if not directly copied from nature, at least recalls the general
aspect of her germinating, growing, and, finally, luxuriant forms of
vegetation. The combination, with reminiscences of Carlovingian knotted
ends to the initial letters, of foliated ornament, during the 12th century,
may be frequently found developed, in Germany especially, into a fresh
luxuriant, and complete system. The complicated conventionality of foliage
shown in many Teutonic manuscripts, and greatly encouraged by the Emperor
Frederick Barbarossa, A.D. 1152 to 1190, was never entirely abandoned by
the Germans in their ornament; and at the end of the 13th and early part of
the 14th centuries, when France and England were successfully imitating
nature, they continued to cling to that peculiarly crabbed style of
crinkled foliation, which they reluctantly abandoned only in the 17th
century.

With the accession of the Plantagenets, in 1154, and especially through the
marriage of Henry II. with Eleanor of Guienne, French influence acquired a
marked predominance in English illumination; and for about one hundred
years from that date, the progress of style in England and France was
parallel and almost identical. Gradually, in each, the Romanesque features
disappeared, and by the middle of the 13th century, the fulness of mediæval
illumination, as reflecting the perfection of Gothic architecture, was
attained. The rapid growth of the Dominican and Franciscan orders during
the first half of the century, and their eagerness to dispel the drowsiness
into which the old well-to-do monastic establishments were fast slipping,
gave a new life to all arts, including, of course, that of the
transcription and illumination of the sources of learning, and in those
days, consequently, of power.

The present appears to be the most fitting place for a few notes, derived
chiefly from the "Consuetudines" of the regulars,[62] on the general
mediæval practice in relation to monastic libraries, of which England,
France, Germany, and Italy possessed many during the 13th, 14th, and 15th
centuries, rich, not only in sacred and patristic, but in profane
literature as well.

The libraries of such establishments were placed by the abbot under the
sole charge of the "armarian," an officer who was made responsible for the
preservation of the volumes under his care: he was expected frequently to
examine them, lest damp or insects should injure them; he was to cover them
with wooden covers to preserve them, and carefully to mend and restore any
damage which time or accident might cause; he was to make a note of any
book borrowed from the library, with the name of the borrower; but this
rule applied only to the less valuable portion of it, as the "great and
precious books" could only be lent by the permission of the abbot himself.
It was also the duty of the armarian to have all the books in his charge
marked with their correct titles, and to keep a perfect list of the whole.
Some of these catalogues are still in existence, and are curious and
interesting, as showing the state of literature in the Middle Ages, as well
as giving us the names of many authors whose works have never reached us.
In perusing these catalogues, it is impossible not to be struck by the
assiduous collection of classical authors, whose works sometimes equal, and
at others actually preponderate over, the books of scholastic divinity. It
was also the duty of the armarian, under the orders of his superior, to
provide the transcribers of manuscripts with the writings which they were
to copy, as well as with all the materials necessary for their labours; to
make bargains as to payment, and to superintend the works during their
progress. These books were not always destined for the library of the
monastery in which they were transcribed, but were often eagerly bought by
others, or by some generous layman, for the purpose of presenting to a
monastic library; and their sale, particularly at an early period, added
largely to the revenues of the establishment in which they were written or
illuminated.

The different branches of the transcribing trade were occasionally united
in the same person, but were more generally divided and practised
separately, and by secular as well as by religious copyists. Of the former,
there were at least three distinct branches--the illuminators, the notarii,
and the librarii antiquarii. The last-mentioned were employed chiefly in
restoring and repairing old and defaced manuscripts and their bindings. The
public scribes were employed chiefly by monks and lawyers, sometimes
working at their own houses; and at others, when any valuable work was to
be copied, in that of their employer, where they were lodged and boarded
during the time of their engagement.

A large room, as has been already stated, was in most monasteries set apart
for such labours, and here the general transcribers pursued their
avocation; but there were also, in addition, small rooms or cells, known
also as scriptoria, which were occupied by such monks as were considered,
from their piety and learning, to be entitled to the indulgence,[63] and
used by them for their private devotions, as well as for the purpose of
transcribing works for the use of the church or library. The scriptoria
were frequently enriched by donations and bequests from those who knew the
value of the works carried on in them, and large estates were often devoted
to their support. The tithes of Wythessy and Impitor, two shillings and
twopence,--and some land in Ely, with two parts of the tithes of the
lordship of Pampesward, were granted by Bishop Nigellus to the scriptorium
of the monastery of Ely, the charter of which still exists in the church
there. A Norman named Robert gave to the scriptorium of St. Alban's the
tithes of Redburn, and two parts of the tithes of Hatfield; and that of St.
Edmondesbury was endowed with two mills, by the same person.

During the whole of the 12th and 13th centuries the pen played a more
distinguished part than the brush in the art of illumination; since, not
only was the former almost exclusively employed in outlining both foliage
and figures, but the use of the latter was generally limited to filling up,
and heightening with timid shadowing, the various parts defined by the
former, and which were altogether dependent upon it for expression. In
fact, it appears as if the principal patterns in 13th century illumination
had been designed by stained-glass painters, the black outlines being
equivalent in artistic result to the lead lines which, in the best
specimens of grisaille and mosaic windows, keep the forms and colours
distinct and perfect. This firm dark outlining was retained in England
later than in France, and was combined in the former country with a more
solid and somewhat less gay tone of colour than ever prevailed in the
latter.

So late as the 15th century, this correspondence between stained glass work
and illumination still obtained; thus, as Mr. Scharf remarks, in a note to
his interesting paper on the King's College, Cambridge, windows, in the
Transactions of the Archæological Institute for 1855, "The forty windows of
the monastery of Horschau contained a series of subjects minutely
corresponding to those of the Biblia Pauperum," &c.

The initial letters which in Romanesque illumination had expanded into very
large proportions as a general rule,[64] diminished; but, in compensation,
effloresced, as it were, into floreated terminations, which were at last
not only carried down the side of the page, but even made to extend right
across both the top and bottom of it. During the reigns of the three first
Edwards in England, the tail, as it might be called, of the initial letter,
running down the side of the page, gradually widened, until at length it
grew into a band of ornament, occasionally panelled, and with small
subjects introduced into the panels. In such cases, the initial letter
occupying the angle formed by the side and top ornaments of the page,
became subsidiary to the bracket-shaped bordering, which, in earlier
examples, had been decidedly subsidiary to the initial letter.

As no one can doubt that the 14th century was the period during which
illumination attained its highest perfection, not only in point of artistic
spirit in design, but in the dexterous processes of execution as well, it
has been considered that it might prove useful to the English student to
supply him or her with as large a proportion as possible of illustration of
that which we may really regard in matter of illuminating as our national
style. Thus our plates in this manual, Nos. 7, 8, and 9, have been taken
from a Latin Bible (B. M. Reg. 1, D. 1), exquisitely written on Uterine
vellum, about the commencement of the 14th century, by an English scribe,
whose autograph at the end of the holy text declares that

  "Wills. devoniensis scripsit istum librum."

Well may the pious writer render thanks as he does, in a paragraph just
preceding the colophon, "to God, to Jesus Christ, to the Blessed Virgin
Mary, and to all Saints," on the completion of such a volume, in every
respect a model of what illuminated writing may be. It is somewhat
deficient in pictures, although in the prologue and in that part of the
Psalms in which David prophesies concerning our Saviour, specimens of the
artist's abilities on a more extended scale than usual may be met with. In
these, as in the initials and borders, manual dexterity is pushed to
perfection, and combined with that occasional feeling for beauty and
constant appreciation of humour, which form leading characteristics of that
English school of illumination, of which "William of Devon" must ever be
ranked among the worthiest. The expression of the little heads, and of the
hands and feet, which are unusually well drawn for the period, is
invariably given with the pen, scarcely any attempt being made at shading
with the brush. The high lights are touched on most delicately with pure
white; and deep blue, and burnished gold grounds looking like solid metal,
are universal throughout the volume.

Our plates, Technical Manual, Nos. 7, 8, 9, also from a Latin Bible in the
Royal collection (No. 15, D 2), are of nearly the same period and style,
but not quite so delicately wrought perhaps as the illuminations are which
we meet with in Reg. 1, D 1. The former offer, however, the least
exceptional aspect of English illumination of the Edwardian period--one in
which vigorous but rather heavy colouring and firm but rather loaded
outline dominate. In these specimens we at length see natural leafage of
the vine, maple, &c., introduced, but scarcely yet allowed to throw itself
about in Nature's wildly wilful way.

From the 12th century onwards, important illuminated manuscripts exist to
the present day in such profusion as to deter me from individualizing in
this necessarily brief essay. I shall rather dwell upon general
characteristics of style, and upon the influence of the leading patrons of
the art, in its palmiest days, in England, France, Germany, and the
Netherlands. In these countries the infinite activity of the mendicant
friars kept up a steady demand for manuscripts of all kinds: thus Richard
de Bury, bishop of Durham, the greatest bibliophile of his age, and the
tutor when prince, and friend while sovereign, of Edward III., relates,
that in all his book-hunting travels: "Whenever we happened to turn aside
to the towns and places where the aforesaid paupers had convents, we were
not slack in visiting their chests and other repositories of books; for
there, amidst the deepest poverty, we found the most exalted riches
treasured up; there, in their satchels and baskets, we discovered not only
the crumbs that fell from the masters table for the little dogs, but,
indeed, the show-bread without leaven,--the bread of angels, containing in
itself all that is delectable."

These mendicant friars were looked upon with great jealousy by the clergy,
who attributed to them the decrease in the number of students in the
universities. Fitz Ralph, archbishop of Armagh and chaplain to Richard de
Bury, accuses them of doing "grete damage to learning:" curiously enough,
his accusation, contained in an oration denouncing them, bears testimony to
their love of books and to their industry in collecting them. "For these
orders of beggers, for endeles wynnynges that thei geteth by beggyng of the
foreside pryvyleges of schriftes and sepultures and othere, thei beth now
so multiplyed in conventes and in persons. That many men tellith that in
general studies unnethe, is it founde to sillynge a pfitable book of ye
faculte of art, of dyvynyte, of lawe canon, of phisik, other of lawe civil,
but all bookes beth y bougt of freres so that en ech convent of freres is a
noble librarye and a grete, and so that ene sech frere that hath state in
schole, siche as thei beth nowe, hath an hughe librarye. And also y sent of
my sugettes to schole thre other foure persons, and hit is said me that
some of them beth come home agen for thei myst nought find to selle ovn
goode Bible; nother othere couenable books." Richard de Bury's example gave
a stimulus to those who succeeded him, both at Durham and elsewhere.

As the styles of architecture varied in England and France,--agreeing in
leading particulars, but each acquiring for itself a set of distinctive
characteristics,--so did the art of illumination. In the purely Gothic
work, such as prevailed from 1250 to 1400, extreme _finesse_ in execution,
tenderness of colour, gentleness of expression, piquancy of ornament, and
elegance of composition, may be regarded as almost invariable attributes of
French productions. In England, on the other hand, the style was not so
harmonious but more vigorous, the colouring was fuller and deeper, the
action of the figures more intense, the power of expression more
concentrated, and reaching occasionally in its energy almost to caricature,
the sense of humour always freely developed, and a more generally active
sentiment of life impressed upon design, not only in figure subjects, but
in ornament. In the latter, monkeys and other animals, dragons, and comic
incidents, are very frequently intermingled with graceful foliage and
heraldic embellishments. In fact it is to the credit of both countries
that, with so much that is excellent in common, they should still have
displayed such free and distinctive features as marked the works of each
respectively. About the year 1400, in both countries the mechanical
reproduction of the accredited types and leading incidents of Scripture and
of Catholic faith began to be abandoned; and, mainly from the necessity of
giving to the historical personages introduced in secular romances and
chronicles individual force and vigour, an attention to portraiture and a
transcription of characteristic traits of active life are freely developed.

Considering how few traces of the art of painting, as exhibited either in
panel pictures or in mural embellishments, remain to attest the condition
of the arts in England and France in the thirteenth, fourteenth, and
fifteenth centuries, it is impossible for the student of Gothic art to
overestimate the extreme interest which attaches to the chronological
series of specimens of the painter's art which may be examined in the great
metropolitan libraries of either country. It is very fortunate for our
reputation that we are enabled, through so large a series of volumes as
still exist, to trace such distinctive and national characteristics as
enable us to assert, without fear of error, that so far as graphic
dexterity is concerned, the English artificers were fully competent to
execute all the artistic productions which have as yet been found upon our
soil. That foreigners were freely employed there can be no doubt, but that
the works which were executed by them could not have been executed by
Englishmen, no one can with safety assert, who has traced with any
considerable care the gradual development of English art through a series
of English illuminated manuscripts.

That illumination was excessively popular in England during the 14th
century among the leading families, is proved by the numbers of coats of
arms emblazoned in many of the most remarkable English manuscripts. Thus in
the Salisbury Lectionary, in the Douce, in Queen Mary's, and in the
Braybrooke Psalters, appear the ancient coats of some of the best blood in
the country. A most interesting contemporary illustration of the precise
terms upon which these noble patrons employed the best illuminators of the
day has been furnished me by a kind and learned antiquarian friend,[65] in
the shape of an extract from the fabric rolls of "York Minster,"[66] of
which the following is a translation:--

"August 26th, 1346.--There appeared Robert Brekeling, scribe, and swore
that he would observe the contract made between him and Sir John Forbor,
viz., that he the said Robert would write one Psalter with the Kalender for
the work of the said Sir John for 5_s._ and 6_d._; and in the same Psalter,
in the same character, a _Placebo_ and a _Dirige_, with a Hymnal and
Collectary, for 4_s._ and 3_d._ And the said Robert will illuminate
('luminabit') all the Psalms with great gilded letters, laid in with
colours; and all the large letters of the Hymnal and Collectary will he
illuminate with gold and vermilion, except the great letters of double
feasts, which shall be as the large gilt letters are in the Psalter. And
all the letters at the commencement of the verses shall be illuminated with
good azure and vermilion; and all the letters at the beginning of the
_Nocturns_ shall be great uncial (unciales) letters, containing V. lines,
but the _Beatus Vir_ and _Dixit Dominus_ shall contain VI. or VII. lines;
and for the aforesaid illumination and for colours he [John] will give
5_s._ 6_d._, and for gold he will give 18_d._, and 2_s._ for a cloak and
fur trimming. Item one robe--one coverlet, one sheet, and one pillow."[67]

Under such contracts, and on much more extravagant terms, were no doubt
produced the finest of those "specimens of English miniature painting" of
the Edwardian period, which Dr. Waagen considers "excel those of all other
nations of the time, with the exception of the Italian, and are not
inferior even to these."[68]

There is probably no document in existence which better illustrates the
nature, cost, and classification of illuminated and other manuscripts
during the 14th and 15th centuries, than the catalogue of the library
founded by William of Wykeham, himself one of the greatest English patrons
of literature, at the College of St. Mary, near Winchester. This catalogue
has been printed _in extenso_ in the "Archæological Journal" (vol. xv. pp.
69 to 74), with notes by the Rev. W. H. Gunner. It is essentially a
_catalogue raisonné_, divided into the following classes, which give a good
idea of the staple commodities in mediæval and monastic libraries:--

"Ordinalia, Antiphonaria, Portiphoria, Legendæ, Collectaria, Graduales,
Manualia, Processionalia, Gradales, Pontificates et Epistolares, Libri
Theologiæ, Doctores super Bibliam, Libri Sententiarum, Doctores super
Sententias, Libri Historiales, Psalteria Glossata, Libri Augustini, Libri
Gregorii, Libri Morales Diversorum Doctorum [to which in many libraries
might, I fear, be added, Libri Immorales Diversorum Auctorum], Libri
Chronici, Libri Philosophiæ [strange to say, a total blank in the
Winchester Collection], Libri Juris Canonici, Decreta et Doctores super
Decreta, Decretales, Libri Sexti cum Doctoribus, Clementinæ, Summæ et alii
Tractatus Diversorum Doctorum Juris Canonici, Libri Juris Civilis, and
Libri Grammaticales."

Most of the volumes in this library were donations from both laity and
clergy, but mainly from the former. The price of every volume is given. The
founder himself presented one Missal valued at £20, and John Yve, "formerly
a fellow of this College, bequeathed a great Portiphoriam for laying before
the senior fellow standing on the right hand of the upper stall," valued at
an equal amount. The York contract, previously quoted, shows precisely how
much illumination could be obtained for much less than one pound; and we
may therefore form from it a tolerable idea of the magnificence of volumes
upon the production of which such large sums were expended. The student
will find this catalogue well repay his careful examination.

During the last half of the 14th century and the beginning of the 15th, the
art of illumination received a great impulse in France, from the
magnificent patronage bestowed upon it by Jean, Duc de Berri, brother of
Charles V. Of his unique library, which excited the envy of all the princes
of his time, and stimulated especially Philip the Bold, Duke of Burgundy,
and the great Duke of Bedford, to enter into competition with him, many
magnificent specimens still remain--such as his Psalter, his two
Prayer-books, and his copy of the "Merveilles du Monde."[69]

French illumination attained perfection in these works, and in some few
specimens of the more decidedly Renaissance period, such as the unsurpassed
"Hours of Anne of Brittany," executed about the year 1500: all of these are
models for the study of the illuminator of the 19th century, since in them
gaiety and charm of ornament will be found united to a style of
miniature-painting of real excellence in art. In the MSS. of the period of
Jean de Berri, we meet with the perfection of that lace-like foliation
known as the Ivy pattern--one that attained an extraordinary popularity in
France, England, and the Netherlands.

In the illuminations of both France and England, during the 14th and first
half of the 15th centuries, the application of raised and highly-burnished
gold became a leading feature, and reached its highest pitch of perfection.
When used, as it frequently was, as a ground for miniature subjects and
ornaments, it was frequently diapered in the most brilliant and delicate
manner. This diapered background gave way at length to an architectural,
and, ultimately, under the influence of the Italian school and that of the
Van Eycks, to a landscape one.

It may be well now to advert to those styles of illumination which, through
the Flemings settled in this country, greatly affected English art; and
which, through the House of Burgundy, equally powerfully wrought upon the
French styles, not so much of ornament, as of miniature-painting. As M.
Hippolyte Fortoul[70] justly remarks, "The powerful school established at
Bruges by the Van Eycks, at the close of the 14th century, exercised an
immense influence on all the schools of Europe, not excepting those of
Italy;"--an influence which was, indeed, not altogether dissimilar from
that brought to bear upon mannerism in Art by the Pre-Raffaelitism of the
present day. The foundations of the Netherlandish school were sufficiently
remote, but may be satisfactorily traced through existing miniatures and
paintings. Herr Heinrich Otte, in his "Handbuch der Kirchlichen
Kunst-Archäologie" (p. 187), gives a chronological list of the principal
MSS. of Germanic production from the Carlovingian period to the
commencement of the 13th century. Up to that period the Byzantine manner
prevailed, mixed with a peculiar rudeness, such as may be recognised in the
works of the great saint and bishop, Bernward of Hildesheim, whom Fiorillo
and other writers look upon, with Willigis of Mainz, as the great animator
of German art in the 11th century.[71] The conversion of this latter
element into Gothic originality appears to have taken place during the 13th
century, and a fine manuscript in the British Museum (B. R 2, b. 11),
ascribed by Dr. Waagen to a period between 1240 and 1260, illustrates the
transition.[72]

With the commencement of the 14th century appear the "Lay of the
Minnesingers," one of the most peculiar of the Paris manuscripts, and
others cited by Dr. Kugler, which carry on the evidence of progressive
development until the power of expression obtained in painting by Meisters
Wilhelm and Stephen of Cologne, is reflected in the contemporary
miniatures.

Even did not the celebrated "Paris Breviary," and the British Museum
"Bedford Missal," or, more correctly speaking, "Book of Hours," both
executed in part by the three Van Eycks, Hubert, Jan, and Margaretha, for
the great Regent of France, exist, the style of the panel-pictures painted
by them would be quite sufficient to show that they must have been
illuminators before they became world-renowned _oil-painters_. Through
their conscientious study of nature, both in landscape and in portrait
subjects, a complete change was wrought in the miniatures of all
manuscripts produced after their influence had had time to penetrate into
the scriptoria and ateliers of the contemporary artist-scribes. Had not the
invention of printing rapidly supervened, there can be no doubt that even
more extraordinary results than followed the general appreciation of their
graces as illuminators would have been ensured.

It is not in publications such as this little manual that any attempts
could be successfully made to reproduce the pictorial results achieved by
such masters in such volumes; but an attempt may certainly be made to
convey some idea both of the general character of the handwriting and of
the ornamental adjuncts by which its effect, and that of the beautiful
little pictures framed in by its brilliant playfulness, was so greatly
heightened. In plates 10, 11, and 12 of this manual, Mr. Tymms has
collected from the "Bedford Book of Hours" much that the student will find
worthy of his careful attention. Well, indeed, may the enthusiastic Dr.
Dibdin soar off into the most transcendental raptures over a volume which,
tested even by the ignoble touchstone of a public sale in 1815, was not
knocked down to its eager purchaser, the then Marquis of Blandford, for a
less sum than £687. 15s. It has now happily found a final resting-place in
the British Museum (ranking as "add. 18,850"), of which it must always
remain as, probably, the greatest treasure, both from its historical
association and its intrinsic excellence and beauty--containing, as it
does, not less than fifty-nine whole-page miniatures, and about a thousand
smaller ones, enriched throughout with gilded lace-work, and ornaments of
the description of that shown in our plates, and commended to the student's
diligent observation.

The later manuscripts of the German and Netherlandish schools of
miniature-painting generally reflect the mixed cleverness and angularities
of such masters as Rogier van der Weyde the elder, Lucas van Leyden, Martin
Schongauer, &c.; where, however, the manner of Hemling prevailed, spiritual
beauty and refinement followed.

To dwell upon Spanish illuminated manuscripts would be comparatively
profitless to the practical student; for all the peculiarities and
excellencies they would appear to have at any time possessed, may be found
more perfectly developed at first in French, subsequently in Netherlandish,
and ultimately in Italian volumes.[73] In one most remarkable and indeed
historical volume, the actual alliance of Spanish writing and initial
illumination with Flemish subject-painting and Arabesque is clearly to be
recognized. The result of the union is certainly most happy, for few more
beautiful books exist than the exquisite missal which in a passage of
golden letters and honied words Francesco de Roias offers to Isabella "the
Catholic." This magnificent volume, from which our plates (Technical
Manual) Nos. 10, 11, and 12, have been taken, was purchased by the
authorities of the British Museum, in whose catalogue it figures as add.
18,851, of Messrs. W. and T. Boone in 1852. In this work the brush triumphs
over the pen, its decorations are essentially pictorial, and many of them
recall, if not the hand, at least the style, of Memling and Van Eyck.
Unlike volumes of earlier periods in which the illustrations are generally
the work of one hand throughout, in this elaborate volume, a division of
labour obtains. In this, as in many others of about the same period, not
only are the penman and the painter two individuals, but the latter
especially becomes half a dozen. This was, no doubt to a great extent,
occasioned by the almost universal substitution of lay for clerical
illuminators in the latter part of the 15th century, and by the production
at that date of illuminated books for dealers adopting the principles and
practice of that system of economic production which ultimately permitted
manufacture to almost universally supersede Art throughout Europe. It
remains now only to sketch, with a brevity altogether out of proportion to
the great interest of the subject, the progress of the art in Italy.

If the delineation of naïve and graceful romantic incident, combined with
elegant foliated ornament, reached perfection in the illuminations of the
French school; if blazoning on gilded grounds was carried to its most
gorgeous pitch in Oriental and Byzantine manuscripts; if intricate
interlacements and minute elaboration may be regarded as the special
characteristics of Hiberno-Saxon scribes; and if a noble tone of solid
colour, combined with great humour and intense energy of expression, marked
England's best productions,--it may be safely asserted, that it was
reserved for the Italians to introduce into the embellishment of
manuscripts those higher qualities of art, their peculiar aptitude for
which so long gave them a pre-eminence among contemporaneous schools.

I therefore proceed to trace the names and styles of some few of the most
celebrated among their illuminators; premising by a reminder to the student
of the miserably low pitch to which art had been reduced in Italy during
the 12th century. Even the most enthusiastic and patriotic writers agree in
the all but total dearth of native talent. Greeks were employed to
reproduce Byzantine mannerisms in pictures and mosaics, and to a slight
extent no doubt as scribes. Illumination was scarcely known or recognized
as an indigenous art; for Dante, even writing after the commencement of the
14th century, speaks of it as "quell' arte, che Alluminar è chiamata a
Parisi."[74]

Probably the earliest Italian manuscript showing signs of real art, is the
"Ordo Officiorum Senensis Ecclesiæ," preserved in the library of the
academy at Sienna, and illuminated with little subjects and friezes with
animals, by a certain Oderico, a canon of the cathedral, in the year 1213.

The Padre della Valle[75] expressly cautions the student against
confounding this Odericus with the Oderigi of Dante,[76] who died about the
year 1300. The latter was unquestionably an artist of some merit, for
Vasari[77] speaks of him as an "excellente miniatore," whose works for the
Papal library, although "in gran parte consumati dal tempo," he had himself
seen and admired. Some drawings by the hand of this "valente uomo," as he
is styled, Vasari speaks of possessing in his own collection.

Baldinucci makes out Oderigi to have been of the Florentine school on no
other grounds than because Vasari describes him as "molto amico di Giotto
in Roma;" and because Dante appears to have known him well. Lanzi,[78]
however, more correctly classes him with the Bolognese school, from his
teaching Franco Bolognese at Bologna, and on the strength of the direct
testimony of one of the earliest commentators on Dante--Benvenuto da Imola.
This same Franco worked much for Benedict IX., and far surpassed his
master.

Vasari especially commends the spirit with which he drew animals, and
mentions a drawing in his own possession of a lion tearing a tree as of
great merit. Thus Oderigi, the contemporary of Cimabue, and Franco, the
pupil of Oderigi and contemporary of Giotto, appear to have been to the Art
of Illumination what Cimabue and his pupil Giotto were to the Art of
Painting,--the pupil in both cases infinitely excelling the master. To them
succeeded, about the middle of the 14th century, a scarcely less celebrated
pair--Don Jacopo Fiorentino, and Don Silvestro, both monks in the
Camaldolese monastery, "degli Angeli," at Florence. The former, Baldinucci
tells us, "improving, with infinite study, every moment not devoted to his
monastic duties, acquired a style of writing greatly sought after for
choral books." The latter, who was rather an artist than a scribe, enriched
the productions of his friend with miniatures so beautiful, as to cause the
books thus jointly produced to excite, at a later period, the special
admiration of Lorenzo the Magnificent, and his son, the no less magnificent
Leo X.[79] So proud were their brother-monks of the skill of Frati Jacopo
and Silvestro, that after their death they preserved their two right hands
as honoured relics.

About a century later, the leading illuminators were Bartolomeo and
Gherardo,--the former abbot of San Clemente, at Arrezzo, and the latter a
Florentine painter and "miniatore," whom Vasari confounds with Attavante, a
painter, engraver, and mosaicist. Of all the Italian artists who adopted
the style of the illuminators, if they did not themselves illuminate, the
most celebrated certainly are Fra Angelico da Fiesole[80] and Gentile da
Fabriano. The majority of the works of both are little else than magnified
miniatures of the highest merit.

The school of Siennese illumination was scarcely less distinguished than
that of Florence. M. Rio dwells with enthusiasm on the books of the Kaleffi
and Leoni, still preserved in the Archivio delle Riformazioni, and
especially on those decorated by Nicolo di Sozzo, in 1334. The greatest
master of the school, Simone Memmi,[81] the intimate friend of Petrarch,
was himself an illuminator of extraordinary excellence, as may be seen by
the celebrated Virgil of the Ambrosian Library at Milan, which contains,
amongst other beautiful miniatures by his hand, the fine portrait of
Virgil, and a very remarkable allegorical figure of Poetry, quite equal in
artistic merit to any of the artist's larger and better-known works in
fresco or tempera.[82]

It is, however, in the library of the cathedral at Sienna, which retains
many of the magnificent choir-books executed by Fra Benedetto da Matera, a
Benedictine of Monte Cassino, and Fra Gabriele Mattei of Sienna, that the
greatest triumphs of the school are still to be recognized. This series of
volumes, although much reduced from its original extent by the abstractions
made by Cardinal Burgos, who carried off a vast quantity to Spain, is still
the finest belonging to any capitular establishment in Italy, and worthily
represents the grandeur of Italian illumination in "cinque cento" days.

The series of similar volumes next in importance to those of Sienna, is
attached to the choir of the church and monastic establishment of the
Benedictines at Perugia, known as "San de' Casinensi." Of these, nothing
more need be said than that they are worthy of the stalls of the same
choir, the design of which is attributed to Raffaelle, and the execution to
Stefano da Bergamo, and Fra Damiano, of the same town, the great
"intarsiatore."

Formerly, as M. Rio observes,[83] "Ferrara could boast of possessing a
series of miniatures, executed principally in the seclusion of its
convents, from the time of the Benedictine monk Serrati, who in 1240
ornamented the books of the choir with figures of a most noble
character,[84] till that of Fra Girolamo Fiorino, who, towards the
beginning of the 15th century, devoted himself to the same occupation in
the monastery of San Bartolomeo, and formed in his young disciple Cosmè a
successor who was destined to surpass his master, and to carry this branch
of art to a degree of perfection till then unknown. Even at the present day
we may see, in the twenty-three volumes presented by the Bishop Bartolomeo
delia Rovere to the cathedral, and in the twenty-eight enormous volumes
removed from the Certosa to the public library, how much reason the
Ferrarese have to be proud of the possession of such treasures, and to
place them by the side of the manuscripts of Tasso and Ariosto.

The "subjects generally treated by these mystical artists were marvellously
adapted to their special vocation: they were the life of the holy Virgin,
the principal festivals celebrated by the Church, or popular objects of
devotion; in short, all the dogmas which were susceptible of this mode of
representation, works of mercy, the different sacraments, the imposing
ceremonies of religion, and, in general, all that was most poetical in
liturgy or legend. In compositions of so exclusive a character, naturalism
could only be introduced in subordination to the religious element."[85]

While this was the case with the majority of illuminations executed under
the auspices of the Church, in those of a secular nature, undertaken for
the great princes and nobles, another set of characteristics prevailed. For
the Gonzagas, Sforzas, D'Estes, Medici, Strozzi, Visconti, and other great
families, the best artists were constantly employed in decorating both
written and printed volumes, in which portraiture is freely introduced, and
picturesque and historical subjects are represented with great vivacity and
attention to costume and local truth. Thus in the truly exquisite "Grant of
Lands," by Ludovico il Moro to his wife Beatrice D'Este, dated January
28th, 1494, and preserved in the British Museum, speaking portraits of both
Ludovico and Beatrice are introduced, with their arms and beautiful
arabesques.[86] Again, in the Hanrot "Sforziada," the first page contains
exquisite miniatures of three members of the princely family of the
Sforzas, by the hand of the all-accomplished Girolamo dai Libri.[87] This
artist, a truly celebrated Veronese and worthy fellow-townsman, with the
almost equally able Fra Liberale, whose work in the manner of Giovanni
Bellini excited the utmost envy on the part of the Siennese illuminators,
was himself the son of a miniature-painter, known as Francesco dai Libri,
and bequeathed the name and art of his father to his own son,--thus
maintaining the traditions of good design acquired in the great school of
Padua, under Andrea Mantegna[88] and Squarcione, during three generations
of illuminators. Girolamo was by far the most celebrated of the three. As a
painter, his works possess distinguished merit, and there still remain good
samples of his abilities in the churches of San Zeno and Sant' Anastasia,
at Verona. He also derives some credit from the transcendent merits of his
pupil Giulio Clovio. Vasari's description of the talents of Girolamo[89]
gives so lively a picture of the style which reached its highest vogue at
the end of the 15th, and during the first half of the 16th centuries in
Italy, that I am tempted to translate it. "Girolamo," he says, "executed
flowers so naturally and beautifully, and with so much care, as to appear
real to the beholder. In like manner he imitated little cameos and other
precious stones and jewels cut in intaglio, so that nothing like them, or
so minute, was ever seen. Among his smallest figures, such as he
represented on gems or cameos, some might be observed no larger than little
ants, and yet in all of them might be made out every limb and muscle, in a
manner which to be believed must needs be seen."

Mr. Ottley supposes that Giulio Clovio (born 1498, died 1578) worked
previous to his receipt of the instruction of Girolamo in a drier manner,
in which no evidence appears of that imitation of Michael Angelesque pose
in his figures, which in his subsequent production became so leading a
characteristic of his style. It is in his earlier manner that Giulio is
believed to have illuminated for Clement VII.[90] (1523-1534), while for
his successor, Paul III. (1534-1539), he worked abundantly, and gradually
acquired that which is best known as his later manner, in which he
continued to labour, according to Vasari, until 1578, at the great age of
eighty years. Mr. Ottley, however, recognizes his hand in MSS. which must
have been at least five years later--during the Pontificate of Gregory
XIII.[91]

It is obviously impossible, in such an essay as the present, to dwell in
detail upon the merits of so accomplished a master of his art. Fortunately
we possess in this metropolis two fine specimens of his skill, both
tolerably accessible--one in the Soane[92] and the other in the British
Museum.[93] A third, of great splendour, is in the possession of Mr.
Towneley, and a fourth, in the shape of an altar-card, attributed to him,
is to be found in the Kensington Museum; and several fragments, formerly in
Mr. Rogers's possession, have passed to Mr. Whitehead and to the British
Museum. All of these exhibit a refinement of execution, combined with a
brilliancy of colour and excellence of drawing, which has never been
surpassed by any illuminator. Vasari gives a complete list and description
of his principal works, and proves him to have been not less industrious
than able.

A contemporary of Giulio's, whose name has been overpowered by the greater
brilliancy of that of the Cellini of illumination, was a certain Apollonius
of Capranica, or, as he signs himself, "Apollonius de Bonfratellis de
Capranica, Capellæ et Sacristiæ Apostolicæ Miniator." Mr. Ottley most
justly states,[94] "that it is impossible to speak in too high terms of the
beauty of his borders, wherein he often introduces compartments with small
figures, representing subjects of the New Testament, which are touched with
infinite delicacy and spirit." His drawing, which is of a decidedly
Michael-Angelesque character, is of less merit when the nude is represented
on a larger scale. His harmony of colour is extraordinary, rather lower in
tone than Giulio Clovio's, but equally glowing, and more powerful. Some
beautiful specimens of his handicraft remain in the possession of Mr. T. M.
Whitehead. The late Mr. Rogers possessed many fragments, the most precious
of which have found their way into the National Collection. His work is
usually dated, and the dates appear to range from 1558 to 1572. Apollonius
having been official illuminator to the very institution from which Celotti
derived his richest spoils, it may readily be imagined that his collection
included an unprecedented series of beautiful examples of Buonfratelli's
style.

Long after the invention of printing, the Apostolic Chamber retained its
official illuminators; and among them one of the most noteworthy is
unquestionably the artist who signs his works, "Ant. Maria Antonotius
Auximas"--a native of Osimo, and a _protégé_ of the princely house of the
Barberini and its magnificent head, Urban VIII. (1623-1644). He was a pupil
of Pietro da Cortona, and an artist of great skill and refinement.[95]

For still more recent popes artists of great excellence continued to be
employed, including for Alexander VII. the celebrated Magdalena Corvina,
who worked from 1655 to 1657; and for Innocent XI (1676 to 1689) a German,
who signs his productions "Joann, frid-Heribach." As the popes retained
their illuminators for the decoration of precious documents, so did the
doges of Venice; and probably the most magnificent of all illumination,
executed after the general spread of printed books had checked, although
not extinguished, the art, may be found in the precious "Ducales," wrought
indeed by several of the greatest Venetian painters.[96]

I need scarcely remind the reader, that the earliest wood-cut and printed
books were made to imitate manuscripts so closely as to deceive the
inexperienced eye. "Artes moriendi," "Specula," "Bibliæ Pauperum," and
"Donatuses,"--the principal types of block books,[97]--represent
illuminated manuscripts in popular demand at the date of the introduction
into Europe of Xylographic Art. Spaces were frequently left, both in the
block books and in the earliest books printed with movable type, for the
illumination, by hand, of initial letters, so as to carry the illusion as
far as possible. This practice was abandoned as soon as the learned
discovered the means by which such wonderfully cheap apparent transcripts
of voluminous works could be brought into the market; and the old decorated
initial and ornamental letters were reproduced from type and wood blocks.

The Mainz Psalter of 1457, and other books printed by Fust and Schoeffer,
required only the addition of a little colour here and there to delude any
inexperienced eye into the belief that they were really hand-worked
throughout. Such imitations were but poor substitutes for the originals in
point of beauty, however excellent when regarded from a utilitarian point
of view.

Every country has more or less cause to mourn the senseless destruction of
many noble old volumes which the printing-press never has, and now, alas!
never can replace; but none more than England, in which cupidity and
intolerance destroyed recklessly. Thus, after the dissolution of monastic
establishments, persons were appointed to search out all missals, books of
legends, and such "superstitious books," and to destroy or sell them for
waste paper; reserving only their bindings, when, as was frequently the
case, they were ornamented with massive gold and silver, curiously chased,
and often further enriched with precious stones; and so industriously had
these men done their work, destroying all books in which they considered
popish tendencies to be shown by illumination, the use of red letters, or
of the Cross, or even by the--to them--mysterious diagrams of mathematical
problems--that when, some years after, Leland was appointed to examine the
monastic libraries, with a view to the preservation of what was valuable in
them, he found that those who had preceded him had left little to reward
his search.

Bale, himself an advocate for the dissolution of monasteries, says:

"Never had we bene offended for the losse of our lybraryes beyng so many in
nombre and in so desolate places for the moste parte, yf the chief
monuments and moste notable workes of our excellent wryters had bene
reserved, yf there had bene in every shyre of Englande but one solemyne
lybrary to the preservacyon of those noble workes, and preferrements of
good learnynges in our posteryte it had bene yet somewhat. But to destroye
all without consyderacion is and wyll be unto Englande for ever a most
horryble infamy amonge the grave senyours of other nations. A grete nombre
of them wych purchased of those superstycyose mansyons reserved of those
lybrarye bokes, some to serve their jaks, some to scoure theyr
candelstyckes, and some to rubbe theyr bootes; some they solde to the
grossers and sope sellers, and some they sent over see to the bokebynders,
not in small nombre, but at tymes whole shippes ful. I know a merchant man,
whyche shall at thys tyme be namelesse, that boughte the contents of two
noble lybraryes for xl shyllyngs pryce, a shame it is to be spoken. Thys
stuffe bathe he occupyed in the stide of greye paper for the space of more
than these ten years, and yet hathe store ynough for as manye years to
come. A prodyguous example is thys, and to be abhorred of all men who love
theyr natyon as they shoulde do."

Wherever the Reformation extended throughout Europe, a corresponding
destruction of ancient illuminated manuscripts took place, and in
localities where fanaticism failed to do its work of devastation,
indifference proved a consuming agent of almost equal energy; and, indeed,
there is no more forcible illustration of the untiring zeal and industry of
the illuminators of old, than the fact, that, after all that has been done
to stamp out the sparks still lingering in their embers, their works should
still glow with such shining lights in all the great public libraries of
Europe.

Despite all this ruthless destruction, and the universal extension of the
art of printing, ornamental penmanship has never been altogether
extinguished as a pictorial art. The Apostolic Chamber, as we have
remarked, retained, until quite recently, its official illuminator.

The luxuriant magnates of the court of the "Grande Monarque" still provided
employment for men like Jary and Prévost, while in England many heraldic
and genealogical MSS. of the 17th, and even 18th centuries, still exist to
prove that the Art was dormant rather than extinct. That it has a brilliant
future yet in store for it no one can hesitate to believe who is enabled to
recognize the power of design, and the capability to execute--either on
paper or vellum--with the brush or pen--by hand, or by calling in the aid
of the printer and lithographer for the rapid multiplication and
dissemination of beautiful specimens--manifested by Owen Jones, his pupil
Albert Warren, and many other able artists and amateurs, still gracing this
19th century of ours with works, many of which will doubtless survive to
our honour and credit so long as Arts and States may endure.

It is in a humble effort to assist in such a consummation that these little
manuals have been written, illustrated, and published.

M. DIGBY WYATT.

END OF PART I.



WHAT ILLUMINATING SHOULD BE IN THE PRESENT DAY.

Illumination, in whatever form practised, can never be properly regarded as
any other than one of the genera into which the art of Polychromatic
decoration may be subdivided. What was originally termed illumination, was
simply the application of minium or red lead, as a colour or ink, to
decorate, or draw marked attention to, any particular portion of a piece of
writing, the general text of which was in black ink. The term was retained
long after the original red lead was almost entirely superseded by the more
brilliant cinnabar, or vermilion. As ornaments of all kinds were gradually
superadded to the primitive distinctions, marked in manuscripts by the use
of different-coloured inks, the term acquired a wider significance, and,
from classical times to the present, has always been regarded as including
the practice of every description of ornamental or ornamented writing.

Because such embellishments were, during the early and Middle Ages, and, in
fact, until long after the invention of printing, almost invariably
executed on vellum, there is no reason whatever why illumination should be
applied to that material, or to paper, which has taken its place, only;
wood, metal, slate, stone, canvass, plaster, all may be made to receive it.
Again: because ancient illumination was almost entirely executed in
colours, in the use of which water and some glutinous medium were the only
"vehicles," there is no reason why modern illumination should not be worked
in oil, turpentine, encaustic, fresco, tempera, varnish, and by every
process in which decorative painting is ever wrought in these days. It is
in such an extension that the most valuable functions of the art are likely
to consist in all time to come. That utilitarian application which it,
originally and for so many centuries, found in the production of beautiful
books, copies of which could be elaborated by no other means than hand
labour, has been, to a great extent, superseded by chromolithography and
chromotypy. No doubt a wide field for useful, and even productive labour,
is still left to the practical illuminator on paper and vellum, in
designing and preparing exquisite originals for reproduction by those
processes, as well as in the rich and tasteful blazoning of pedigrees,
addresses, family records and memorials, and in the illustration for
presentation, or for private libraries, of transcripts from favourite
authors; but, at the same time, an equally elegant and useful application
of the art would be to enrich ceilings, walls, cornices, string-courses,
panels, labels round doors and windows, friezes, bands, chimney-pieces, and
stained and painted furniture in churches, school-rooms, dwellings, and
public buildings of all kinds, with beautiful and appropriate inscriptions,
of graceful form and harmonious colouring. Such illumination would form,
not only an agreeable, but an eminently useful decoration. How many texts
and sentences, worthy, in every sense, of being "written in letters of
gold," might not be thus brought prominently under the eyes of youth,
manhood, and old age, for hope, admonition, and comfort. No more skill,
energy, and taste are requisite for the production of this class of
illumination than are essential for satisfactory work upon vellum and
paper; and while in the one case the result of the labour may be made an
incessant enjoyment for many, in the other, it is seldom more than a
nine-days' wonder, shut up in a book or portfolio, and seen so seldom as
scarcely to repay the amateur for the expense and trouble involved in its
execution.

This, if I may be allowed the term, manifold application of forms,
primarily available for book decoration only, has not been lost sight of in
the selection and arrangement of the illustrations, both in this and in the
"Historical Manual." Mr. Tymms has, with excellent judgment, so arranged
them as to lead the student who may occupy himself in copying them, or
enlarging from them, gradually onwards from the comparatively easy to the
more difficult varieties of the art. Adopting, in all cases, the alphabet
of capital letters as a starting-point, the beginner will do well to learn
to write before attempting to learn to draw: he should copy the alphabet,
say on Plate I., fig. 1, on waste paper--common cartridge, or
paper-hanger's lining paper, will be best--many time; at first, in
fac-simile, then twice the size as printed, then four times, then eight
times, until he may be able to form letters as much as six inches high,
correctly. Having so far mastered the capitals, let him try in exactly the
same way to produce and reproduce the same text, or lower-case letters in
which the passages from Scripture, say Plate No. 1, fig. 2, have been
written. Let him then try a sentence not given in the plates, using for it
the capital and lower-case letters he has been learning how to form, and
let him work out his own sentence in as many different sizes as he has
previously tried Mr. Tymms's in. By the time he has drawn the enriched
initial letters of the same plate several times, he will find that his eye
and hand will have probably gained sufficient command to justify his
attempting to copy, in outline as before, on waste paper, and both on a
small and large scale, the ornaments given on Plate No. 3. In the intervals
between his outline studies, the young illuminator may occupy himself in
mastering the instructions given in this volume, so that he may have a
general idea of the theory of the different processes before be commences
an attempt to put them in practice. His first lesson in colouring should
then commence by his attempting to colour Plate No. 3, in fac-simile of
Plate No. 2.

It will be well to begin gilding and silvering with shell-gold and
aluminium, reserving for more advanced experiments the use of gold or
silver paper and leaf. The student may then with advantage copy in outline,
first of all the outline Plate No. 6, inking it in so as to produce, on
fine-grained drawing paper or card, a fac-simile of the printed plate. He
should next proceed to colour the printed plate to correspond with Plate
No. 5. Avoiding any defects he may have made in this operation, he may
colour his own outline as he had done the printed one; he will then find
himself able to copy both outline and colour on a small scale.

In his next set of lessons a much heavier demand will be made on his
capabilities.

To satisfactorily reproduce, either upon the same or upon an enlarged
scale, the compact black letter of Plate VII., and the solid brilliant
colours of Plate VIII., with the golden grounds, which should in this case
be highly burnished, will be found a much more difficult task than any yet
encountered. The student must not be discouraged by a little failure at
first. The technical operations of illumination are essentially
manipulative, and like the fingering of a musical instrument, must be
learnt by frequent and active exercise. The mere degree of skill requisite
to enable the artist to lay on a perfectly flat tint of very strong or of
very delicate colour, is only likely to be acquired after he may have
washed some fifty different tints, more or less cloudy and muddled. Few
hands will be found capable of tracing out a pure, firm, even outline, of
equal thickness and force in every part, with either pen, pencil, or brush,
which have not made many a score of ragged, feeble, or blotched attempts at
steadiness. To keep a number of lines perfectly upright, parallel, or
evenly spaced, demands an amount of dexterity which can only be gained by
laborious practice. The student must not therefore feel discouraged if at
first his hand may scarcely answer to the call made upon it. The failure of
to-day, with proper attention and perseverance, may become the germ of the
success of to-morrow; all that is essential is never on one day to repeat
the fault of its predecessor. Nothing will tend to give the beginner
greater confidence than the habit of working out the same forms and
processes upon various scales. Taking, for instance, such an initial letter
as the P, fig. 1, Plate X.; it would be an improving lesson to copy it in
pen-and-ink outline, exactly as it is shown, and then to copy it, say six
times the size given on the plate, thickening the lines, of course, in
proportion. Then let the student once again try to copy it in fac-simile,
and he will himself be probably surprised to find how much better and more
easily he will accomplish his task than he was enabled to do on his first
trial. A corresponding experiment, involving the application of gold and
brilliant colours, such as would be essential to a reproduction of figs. 1,
2, 3, 4, or 5, of Plate II., on various scales, will be found probably no
less useful and satisfactory.

A similar technical principle to that which has governed the selection and
order of the plates in this manual has also determined those in its
companion, the "Historical Manual."

Beginning with the simpler conventional styles of the Carlovingian school,
in Plates I., II., and III. (from the fragments of the Bible of Charles the
Bald, Harleian, 7,551), involving outline and flat tinting only, the
student may advance to the lightly-shaded pen-work and foliation of the
Romanesque style given in Plates IV., V., and VI., from the British Museum,
Reg. 1, C. VII. In the purely mediæval illumination of the 14th century,
Plates VII., VIII., and IX., from the British Museum (Reg. 1, D 1), the
tints become more solid; while the raised and embossed gold, the
complicated diapers, and more fully-shaded foliage, demand both
considerable mechanical dexterity, and some real artistic capability on the
part of the amateur, who would successfully revive the brilliant and
powerful execution of the master-scribes of the Edwardian age. Towards the
end of the 15th century, the miniatures of the illuminated books reflected
the general advance made all over Europe in the art of painting. Imitative
art rapidly superseded conventional, and although much ornament is freely
introduced in combination with small pictures, it is made to participate in
the general system of light and shade and arrangement of colour which
dominates over the more essentially-pictorial portions of the decoration.
Such a style of ornament is well shown in Plates X., XI., and XII., from
the Bedford (so called) missal, and in the three last plates of this
manual, from the beautiful Missal of Ferdinand and Isabella--British
Museum, add. 18,851.

Thus the student will find, that his own progress will tally with the
transitional changes of the art, from its infancy to its most artistic
phase, and that long before he may have learnt enough to enable him to
imitate successfully the miniature style of the 15th century, he may be in
a position to produce tolerably satisfactory reproductions of the early and
mediæval work.

Having thus suggested the most profitable mode in which the student can, I
believe, make use of the beautiful examples Mr. Tymms has prepared for his
assistance, I consider it well to proceed to offer to his notice such
counsel, as may, I trust, tend to induce him not to rest contented with
reproduction of old examples upon a small scale, but rather to extend the
sphere of his studies and operations into the origination of a fresh and
expanded system of decoration, based as a starting-point upon the labours
of the most zealous masters of the craft.

In the few remarks I am about to offer in respect to what the Art of
Illumination really should be now, I propose to treat briefly, but
specifically, of its application to each of the different substances on
which it may be most satisfactorily worked, in the following series:
vellum, paper, tracing-paper, canvass, plaster, stone, metal, wood. Dealing
with design only in this section of my essay, I propose, in the following
and concluding one, to adhere to the same order in noticing the best
processes by which amateurs may carry out the class of work I would
recommend to their notice.

To commence, therefore, with vellum: it is obvious that good copies of
ancient illuminated manuscripts can be made on this material only, for
there is a charm about the colour and texture of well-prepared calf-skin,
which no paper can be made to possess. For the same reason, and on account
of its extraordinary toughness and durability, it is especially suitable
for pedigrees, addresses, and other documents which it may be considered
desirable to preserve for future generations. To transcribe on vellum and
decorate the writings of ancient and modern authors so as to form unique
volumes, appears to me--nowadays, when God gives to every man and woman so
much good hard work to do, if they will but do it--little else than a waste
of human life. In days when few could read, and pictures drawn by hand were
the only means within the reach of the priesthood, of bringing home to the
minds of the ignorant populace the realities of Biblical history, and of
stimulating the eye of faith by exhibiting to the material eye pictures of
those sufferings and triumphs of saints and martyrs, on which the Church of
Rome during the Middle Ages mainly based its assertions of supremacy, it
was all very well to spend long lives of celibacy and monastic seclusion in
such labours; but the same justification can never be pleaded again. I am
quite ready to admit that the exceptional manufacture of these pretty
picture-books may be not only agreeable, but even useful: it is the abuse,
and not the occasional resort to the practice, I would venture to denounce.
For instance, a mother could scarcely do a thing more likely to benefit her
children, and to fix the lessons of love or piety she would desire to plant
in their memories, than to illuminate for them little volumes, which, from
their beauty or value, they might be inclined to treasure through life.
Interesting her children in her work as it grew under her hand, how many
precious associations in after-life might hang about these very books.
Again: for young people, the mere act of transcription, independent of the
amount of thought bestowed upon good words and pure thoughts, and the
selection of ornament to appropriately illustrate them, would tend to an
identification of the individual with the best and highest class of
sentiments.

All that has been said with respect to illumination on vellum applies, with
equal force, to illumination on paper. There has to be borne in mind,
however, the essential difference that exists between the relative
durability of the two substances. Elaboration is decidedly a great element
of beauty in illumination: and neatly-wrought elaboration cannot be
executed without care, patience, and a considerable sacrifice of time: why,
therefore, bestow that care, patience, and time upon a less permanent
material, when one only a trifle more costly, but infinitely more lasting,
is as easily procured? Work on paper, therefore, only as you would write
exercises or do sums upon a slate; learn and practise upon paper, but
reserve all more serious efforts for vellum only. No effect can be got upon
the former material, which cannot, with a little more dexterity, be
attained upon the latter.

As none of the other substances mentioned as those on which illumination
may be executed are available for making up into books, before proceeding
to a consideration of the special conditions under which the art may be
applied to them, I beg to offer the following recommendations with respect
to design, as suitable for book-illustration generally.[98]

Firstly:--Take care that your text be perfectly legible; for, however
cramped and confused the contents of many of those volumes we most admire
may now appear, it is to be remembered that they were all written in the
handwriting most easily read by the students of the periods in which they
were written. The old scribes never committed the solecism of which we are
too often guilty, of bestowing infinite pains on writing that which, when
written, not one in a hundred could, or can, decipher.

Secondly:--Fix the scale of your writing and ornament with reference to the
size of your page, and adhere to it throughout the volume. This rule, which
was rigidly observed in all the best periods of the art, is incessantly
disregarded in the present day; and to such an extent, that not only does
scale frequently differ, as we turn page after page, but the same page will
frequently exhibit scroll-work, derived from some great choral folio,
interwreathed with leafage borrowed from some pocket Missal or Book of
Hours.

Thirdly:--If you adopt any historical style or particular period as a basis
on which your text, miniatures, or ornamentation are to be constructed,
maintain its leading features consistently, so as to avoid letting your
work appear as though it had been begun in the 10th century, and only
completed in the 16th; or, as I have once or twice seen, _vice versâ_. For
however erratic changes of style may appear to be in Art, as they run one
another down along the course of time, it will be invariably found that
there exists a harmony between all contemporary features, which cannot be
successfully disregarded; and this it is which has ever rendered
eclecticism in art a problem,--not impossible, perhaps, to solve, but one
which, as yet at least, has never met with a satisfactory practical
solution.

Fourthly:--Sustain your energies evenly throughout your volume; for,
remember, your critics will estimate your powers, not by your best page,
but by a mean struck between your best and your worst. Book illumination is
generally looked upon as microscopic work, demanding the greatest
exactitude; and whatever merits any page may display, they will go for
little, if that page is disfigured by a crooked line, or a single leaf
insufficiently or incorrectly shadowed; and the greater the merit, the more
notable the drawback.

Fifthly:--Rigidly avoid contrasting natural with conventional foliage.
Adopt which you like, for by either beautiful effects may be produced; but
mix them, and the charm of both is gone. Natural foliage may be
successfully combined with any other varieties of conventional ornament,
excepting those based upon natural foliage.

Sixthly:--Take care that some at least of your dominant lines and borders
are kept parallel to the rectangular sides of your pages; for unless your
flowing and wayward ornaments are corrected by this soberer contrast, they
will, however beautiful in themselves, have a straggling and untidy
appearance in the volume. Where the lines of text are strongly marked, as
in black ink on a white ground, and the page is so far filled with text as
to leave but little space for ornament, this rule may be, to a great
extent, disregarded, for the lines of the text will themselves supply the
requisite contrast to the flowing forms; but where the page is nearly
filled with ornament, or when the text is faint only, as in gold lettering
on a white ground, it becomes imperative.

Seventhly:--Be decided, but temperate, in your contrasts of colour. It
would obviously exceed the limits of these notes to attempt in them to
enter upon the principles of the "harmony of colour;" they must be studied
from treatises specially devoted to the subject. Such study must, however,
be accompanied by constant experiment and practice; for it would be as
foolish to expect a man to be a good performer upon any instrument, because
he had learnt the theory of music, as it would be to suppose that he must
necessarily paint in harmonious colouring, because he studied the theory of
balance in combination. To the experienced eye and hand, functions become
intuitive, which, to the mere theorist, however profound, are toil and
weariness of spirit.

Such are a few of the rules, by attention to which the illuminators of old
achieved some of their happiest effects, and which can never be safely
disregarded by those who would emulate their efforts.

In taking up the class of substances on which illumination, as applied to
general decoration, may be best executed, we meet, firstly, with one
occupying a somewhat intermediate position,--viz., tracing-paper. I term
its position intermediate, because, it may be wrought upon in either oil or
water colour; and because, when so wrought upon, it may be either mounted
on paper or card, and so made to contribute to book or picture enrichment;
or attached to walls or other surfaces, brought forward in oil-colours, and
be so enlisted in a general system of mural illumination. How this may best
be done technically will be hereafter described; here I may notice only the
use which may be made of this convenient material, by many not sufficiently
advanced in design or drawing to be able to invent or even copy correctly
by free hand, and yet desirous of embellishing some particular surface with
decorative illumination. For instance, let it be desired to fill a
rectangular panel of any given dimension with an illuminated inscription.
Take a sheet of tracing-paper the exact size, double it up in both
directions, and the creases will give the vertical and horizontal
guidelines for keeping the writing square and even: then set out the number
of lines and spaces requisite for the inscription, fixing upon certain
initial letters or alphabets for reproduction on an enlarged scale, from
this work, or any other of a similar kind, and making the height of the
lines correspond therewith. Then lay the tracing over either the original
or the rough enlargement, and trace with pen, pencil, or brush, each letter
in succession, taking care to get each letter into its proper place, in
reference to the whole panel, to the letter last traced, and to the other
letters remaining to be traced. When this is completed, trace on whatever
ornaments may best fill up the open spaces and harmonize with the style of
lettering. When the tracing is completed, with a steady hand pick in all
the ground-tint, keeping it as even as possible; and heighten the letters
or ornaments in any way that may be requisite to make them correspond with
the models from which they may have been taken. By adopting this method of
working, with care and neatness of hand, very agreeable results may be
obtained, without its being indispensable for the illuminator to be a
skilful draughtsman. The tracing-paper may be ultimately attached to its
proper place, and finished off, as will be hereafter recommended; and, if
cleverly managed, it will be impossible to detect that that material has
ever been employed.

The special convenience of illuminating upon canvass is, that instead of
the operator having to work either from a ladder or scaffold, or on a
vertical or horizontal surface, he may do all that is necessary at an easel
or on a table on terra firma. His work when completed may be cut out of the
sheet of canvass on which it has been painted, and may be fastened to the
wall, ceiling, or piece of furniture for the decoration of which it may
have been intended. All that is essential, with respect to the designs
which may be wrought upon it, is, to take care that they are fitted for the
situations they may be ultimately intended to occupy. Thus it must be
obvious that it would be an entire waste of time to elaborate designs
destined to be fixed many yards from the eye, as minutely as those which
would be in immediate proximity to it. No branch of designing illuminated
or other ornament requires greater experience to succeed in than the
adjustment of the size of parts and patterns to the precise conditions of
light, distance, foreshortening, &c., under which they are most likely to
be viewed.

Illumination on plaster may be executed either in distemper, if the walls
or ceiling have been sized only, or in oil if they have been brought
forward in oil-colours. The former is the most rapid, but least durable
process. Hence decoration is usually applied in oil to walls which are
liable to be rubbed and brushed against, and in distemper, to ceilings,
which are, comparatively speaking, out of harm's way. Very pretty
decorations on plaster may be executed by combining hand-worked
illumination with diapered or other paper-hangings. Thus, for instance,
taking one side of a room, say about eleven feet high, to the under-side of
the plaster cornice, mark off about a foot in depth on the wall from the
bottom of the cornice, set out the width of the wall into three or more
panels, dividing the panels by upright pilasters of the same width as the
depth of the top border. At the height of about four feet from the ground
mark off the top edge of another horizontal band, which make also one foot
deep; continue on the lines of the pilasters to within six inches of the
top of the skirting, and draw in a horizontal border, six inches high,
running all round upon the top edge of the skirting: then paint, in a plain
colour, a margin, three or four inches wide, all round the panels formed by
the bands and pilasters, and let the paper-hanger fill in the panels with
any pretty diapered paper which may agree with the style and colour in
which you may desire to work your illumination. The side of your room will
then present two horizontal lines--one next the cornice, and one at about
dado-height, suitable for the reception of illuminated inscriptions. In
setting these out, care must be taken to bring a capital letter into a line
with the centre of each pilaster, so that a foliated ornament, descending
from the upper inscription, and ascending from the lower one, may meet and
intertwine on the pilasters, forming panelled compartments for the
introduction of subjects, if thought desirable.[99]

It is by no means necessary for the sides of these pilasters, or the
bounding lines of the bands containing inscriptions, to be kept straight;
they may be varied at pleasure, so long as they are kept symmetrical in
corresponding parts, and uniformly filled up with foliation emanating from,
or connected with, the illuminated letters. Agreeable results may be
produced by variations of such arrangements as the one suggested.
Frequently round doors, windows, fireplaces, &c., inscriptions may be
executed with very good effect, either on label-scrolls or simple borders,
and with greater or less brilliancy of colour, according to the
circumstances of the case. Often simplicity and quiet have greater charms
than glitter or brilliancy; thus black and red, on a light-coloured ground,
the most primitive combination in the history of writing, is always sure to
produce an agreeable impression: blue, crimson, or marone on gold, or _vice
versâ_, are no less safe: black, white, and gold, counterchanged, can
hardly go wrong.

Few amateurs will be likely to attempt illuminations upon plaster ceilings,
owing to the great difficulty they will experience in working overhead with
a steady hand. They will generally do wisely--to execute the principal
portions on paper, tracing-paper, or canvas,--to fasten them up, as will be
hereafter directed,--and to confine the decoration actually painted on the
ceiling, to a few panels, lines, or plain bands of colour, which may be
readily executed by any clever house-painter or grainer, even if altogether
ignorant of drawing and the art of design. The most beautiful illuminated
ceiling of mediæval times I believe to be that of the chapel in the
celebrated Jacques Coeur's house, at Bourges, in France. It is vaulted, and
each compartment contains inscribed labels held by floating angels. The
white draperies of the angels are relieved on a delicate blue ground only,
so that the stronger contrast of the black writing on the white labels
gives a marked predominance to the inscriptions; which, being arranged
symmetrically, produce in combination agreeable geometrical figures.

Most of the preceding remarks apply equally to stone; but in reference to
that material, there is one point to specially enforce,--namely, the
advisability of not covering the whole of the surface with paint. There is
about all stone a peculiar granulation, and in many varieties a slight
silicious sparkle, which it is always well to preserve as far as possible.
Illuminate, by all means, inscriptions, panels, friezes, &c., colour
occasionally the hollows of mouldings, and gild salient members
sufficiently to carry the colour about the monument, whether it may be a
font, a pulpit, a tomb, a reredos, a staircase, a screen, or a doorway, and
prevent the highly-illuminated portion from looking spotty and unsupported;
but by no means apply paint all over. It is not necessary to produce a good
effect; it destroys the surface and appearance of the stone, making it of
no more worth than if it were plaster, and it clogs up all the fine arrises
and angles of the moulded work or carving. Wherever stained glass is
inserted in stonework, the application of illumination, or at any rate of
coloured diaper-work of an analogous nature, is almost an imperative
necessity, in order to balance the appearance of chill and poverty given to
the stonework by its contrast with the brilliant translucent tints of the
painted glass. In illuminating stonework, it seldom answers to attempt to
apply decoration executed on paper or canvass; it should in all cases
(excepting when it is at a great distance from the eye) be done upon the
stone itself. The only exception is the one to which I shall allude in
speaking of metal.

Slate, although from its portability and non-liability to change its shape
under variations of temperature, a convenient material for filling panels,
and forming slabs for attachment to walls, is not to be recommended to the
amateur, owing to the difficulty he will experience in effecting a good and
safe adhesion between his pigments and the surface of the slate. In what is
called enamelled slate, an excellent attachment is secured by gradually and
repeatedly raising the slate to a high temperature; but the process would
be far too troublesome and expensive for practice by the great majority of
amateurs.

Metal in thin sheets is liable only to the objection from which slate is
free,--namely, that it is difficult to keep its surface from undulation in
changes of temperature. In all other respects, both zinc, copper, lead, and
iron, bind well with any oleaginous vehicle, and offer the great
convenience that they may be cut out to any desired shape, and attached to
any other kind of material by nails, screws, or even by strong cements,
such as marine glue. Zinc is, perhaps, the best of all, as it cuts more
readily than copper or iron, and keeps its shape better than lead; care
should, however, be always taken to hang it from such points as shall allow
it to freely contract and expand. If this is not attended to, its surface
will never remain flat. It is a material particularly well adapted for
cutting out into labels to surmount door and window arches, or to fill the
arcading of churches and chapels, and to be illuminated with texts or other
inscriptions. Very beautiful effects may be produced by combining
illumination with the polished brass-work which is now so admirably
manufactured by Messrs. Hardman, Hart, and others. Care should, however, be
taken not to overdo any objects of this nature. Let the main lines of
construction always remain unpainted, so that there may be no question as
to the substance in which the article is made, and restrict the application
of coloured ornament or lettering to panels, and, generally speaking, to
the least salient forms. Of course, where it can be afforded, enamelling
offers the most legitimate mode of illuminating metal-work; and ere long it
is to be hoped that the beautiful series of processes by means of which so
much durable beauty of colour was conferred on Mediæval metal-work may be
restored to their proper position in British Industry, and popularized as
they should, and, I believe, might readily be.

To woodwork, illumination may be made a most fitting embellishment; and the
application of a very little art will speedily be found to raise the
varnished deal cabinet or bookcase far above the majority of our standard
"institutions" in the way of heavy and expensive mahogany ones, in interest
at least, if not in money value. Almost every article of furniture may thus
be made, as it were, to speak and sympathize; for the return every
decorated object makes to the decorator is always in direct proportion to
the amount of life and thought he has put into his work. It is a common
saying, that, "what comes from the heart goes to the heart;" and in nothing
does it hold good more than in the production of works of art of all kinds,
including Illumination, which, through its special dealing with written
characters, has so direct an access to the intellect and affections.

In all appeals the decorative artist can make to the brain through the eye,
he has open to him two distinct channels of communication in making out the
scheme of his ornamentation,--the one by employing conventional forms,--and
the other by introducing representations of natural objects. In the former
he usually eschews light, shade, and accidental effects altogether; and in
the latter he aims at reproducing the aspect of the object he depicts as
nearly as possible as it appears to him. Both modes have found favour in
the eyes of the great illuminators of old, and by the best they have been
frequently and successfully blended. Under the "conventional" series maybe
classed all productions dependent on either an Oriental or Hiberno-Saxon
origin; among the "natural," the later, Netherlandish, Italian, and French
illuminations may be generally grouped; and, in a mixed style, the majority
of the book-decorations of the Mediæval period.

To be enabled to recognize intuitively how to blend or contrast, to adopt
or avoid, these different modes of treatment of ornament, is given to but
few, and is revealed to those few only, after years of study and of
practice. Rules may assist,[100] but can never suffice to communicate the
power; work of the most arduous kind, and persistent observation, can alone
bestow it. Still, with good models upon which to base his variations, and
goodwill, the amateur may do much, and will probably best succeed by
recurring incessantly to Nature, and combining direct, or nearly direct,
imitation of Nature with geometrical lines and masses of colour
symmetrically disposed. To aid his footsteps in this direction, I know no
more convenient councillor than Mr. Llewellyn Jewitt, whose historical
introduction to his brother's "Manual of Illuminated and Missal Painting"
contains some just remarks upon the subject.[101]

Having thus rapidly touched upon the series of materials upon which the Art
may be brought to bear, and the leading principles of design suitable under
different circumstances, I proceed to suggest the class of "legends," as
the mediæval decorators called them, likely to prove most fitting for
special situations. No doubt many more apt and piquant may suggest
themselves to some practical illuminators than the few I have culled (with
the assistance of one or two kind friends), principally from old English
writers; but to others, those I now present may not be without, at any
rate, a convenient suggestiveness. Something similar to the following I
would recommend for the embellishment of ceilings, friezes, string-courses,
or flat walls of the different apartments indicated. Of some I have given
four lines--one, say, for each side of a room; of others but a line, such
as might go over a door. Between the two are many suitable for panels or
irregular situations; and in one or two cases passages of many lines have
been chosen, fit for illumination on vellum or paper, and for framing to
hang up in the apartments specified, or to be inserted in panels or
furniture or on screens.


          FOR DRAWING-ROOMS.

 "For trouble in earth take no melancholy;
    Be rich in patience, if thou in goods be poor.
  Who lives merry, he lives mightily;
    Without gladness avails no treasùre."
                            WM. DUNBAR.

 "Since earthly joy abideth never,
  Work for the joy that lestis ever;
  For other joy is all in vain;
  All earthly joy returns in pain."
                            _Idem._

 "Who shuts his hand hath lost his gold;
  Who opens it, hath it twice told."
                            GEORGE HERBERT.

 "No bliss so great but cometh to an end;
  No hap so hard but may in time amend."
                            ROBERT SOUTHWELL.

 "Freedom all solace to man gives;
  He lives at ease, that freely lives."
                            JOHN BARBOUR.

 "That which is not good, is not delicious
  To a well-governed and wise appetite."
                            MILTON.


          FOR A STUDIO.

 "Order is Nature's beauty, and the way
  To order is by rules that Art hath found."
                            GWILLIM.


  FOR A FAMILY PORTRAIT-GALLERY OR HALL.

 "Boast not the titles of your ancestors,
  Brave youths: they're their possessions, none of yours.
  When your own virtues equall'd have their names,
 'Twill be but fair to lean upon their fames,
  For they are strong supporters; but till then
  The greatest are but growing gentlemen."
                            BEN JONSON.


  FOR BREAKFAST OR DINING ROOMS.

 "A good digestion turneth all to health."
                            WORDSWORTH.

 "If anything be set to a wrong taste,
 'Tis not the meat there, but the mouth's displeased.
  Remove but that sick palate, all is well."
                            BEN JONSON.

 "Nature's with little pleased, enough's a feast;
    A sober life but a small charge requires;
  But man, the author of his own unrest,
    The more he has, the more he still requires."

     "To bread or drink, to flesh or fish,
      Yet welcome is the best dish."
                            JOHN HEYWOOD.

 "It is the fair acceptance, Sir, creates
  The entertainment perfect, not the cates."
                            BEN JONSON, _Epigrams_, ci.

                       "No simple word
  That shall be utter'd at our mirthful board,
  Shall make us sad next morning."
                            _Ibid._

                       "To spur beyond
  Its wiser will the jaded appetite,
  Is this for pleasure? Learn a juster taste,
  And know that temperance is true luxury."
           ARMSTRONG, _Art of Preserving Health_, book ii.

 "What an excellent thing did God bestow on man,
  When He did give him a good stomach!"
                            BEAUMONT AND FLETCHER.

"The stomach is the mainspring of our system. If it be not sufficiently
wound up to warm the heart and support the circulation, we can neither

  _Think_ with precision,
  _Sleep_ with tranquillity,
  _Walk_ with vigour,
  Or sit down with comfort."
                            DR. KITCHENER.

"The destiny of Nations has often depended upon the digestion of a Prime
Minister."--DR. KITCHENER.

     "Is't a time to talk
      When we should be munching?"
  JUSTICE GREEDY, in MASSINGER'S _New Way to pay Old Debts_.

 "No roofs of gold o'er riotous tables shining,
  Whole days and sums devoured with endless dining."
          CRASHAW'S _Religious House_.

 "Now good digestion wait on appetite,
  And health on both."
                            SHAKSPERE.

 "When you doubt, abstain."
                            ZOROASTER.

 "Where there is no peace, there is no feast."
                            CLARENDON.

 "Not meat, but cheerfulness, makes the feast."
 "Who carves, is kind to two; who talks, to all."
                            GEORGE HERBERT.


          FOR KITCHENS.

 "A feast must be without a fault;
  And if 'tis not all right, 'tis nought."
          KING'S _Art of Cookery_.

 "Good-nature will some failings overlook,
  Forgive mischance, not errors of the cook."
                            _Ibid._


          FOR SUPPER-ROOMS.

 "Oppress not nature sinking down to rest
  With feasts too late, too solid, or too full."
          ARMSTRONG, _Art of Preserving Health_.

                                 "As men
  Do walk a mile, women should talk an hour
  After supper: 'tis their exercise."
          BEN JONSON, _Philaster_, act 2, sc. 4.


          FOR STILL-ROOMS.

 "The nature of flowers Dame Physic doth show;
  She teaches them all to be known to a few."
          TUSSER, _Five Hundred Points of Good Husbandry_.

 "The knowledge of stilling is one pretty feat,
  The waters be wholesome, the charges not great."
                            _Id. ibid._


          FOR A STORE-ROOM.

 "He that keeps nor crust nor crumb,
  Weary of all, he shall want some."
                            SHAKSPERE.


          FOR MUSIC-ROOMS.

 "Music removeth care, sadness ejects,
  Declineth anger, persuades clemency;
  Doth sweeten mirth, and heighten piety,
  And is to a body, often ill inclined,
  No less a sovereign cure than to the mind."
                            BEN JONSON.

 "Here will we sit, and let the sounds of music
  Creep in our ears."
          SHAKSPERE, _Merchant of Venice_.

 "Play on and give me surfeit."
                            _Ibid._


          FOR SMOKING-ROOMS.

 "Tobacco's a physician,
    Good both for sound and sickly;
     'Tis a hot perfume,
      That expels cold rheum,
    And makes it flow down quickly."
                            BARTEN HOLLIDAY.

 "Tobacco hic! Tobacco hic!
  If you are well, 'twill make you sick;
  Tobacco hic! Tobacco hic!
  Twill make you well, if you are sick."


          FOR DRINKING-ROOMS.

 "Backe and syde goo bare goo bare,
    Both hande and fote goo colde;
  But belly, God sende the gode ale inoughe,
    Whether hyt be newe or olde."
          BP. STILL, in _Gammer Gurton's Needle_.

 "The first draught serveth for health,
    The second for pleasure,
    The third for shame,
    The fourth for madness."

 "The greatness that would make us grave
    Is but an empty thing;
  What more than mirth would mortals have:
    The cheerful man's a king."
                            ISAAC BICKERSTAFF.


          FOR PUBLIC COFFEE-ROOMS.

 "If you your lips would keep from slips,
    Five things observe with care:
  Of whom you speak, to whom you speak,
    And how, and when, and where."

 "Every creature was decreed
    To aid each other's mutual need."
                            GAY.


          FOR BILLIARD-ROOMS.

 "The love of gaming is the worst of ills;
  With ceaseless storms the blacken'd soul it fills,
  Inveighs at Heaven, neglects the ties of blood,
  Destroys the power and will of doing good;
  Kills health, poisons honour, plunges in disgrace."
          YOUNG, _4th Satire_.

 "Play not for gain, but sport: who plays for more
  Than he can lose with pleasure, stakes his heart,
  Perhaps his wife too, and whom she hath bore."
          GEO. HERBERT, _The Church Porch_.


          FOR BEDROOMS.

 "Rise with the lark, and with the lark to bed;
  The breath of night's destructive to the hue
  Of every flower that blows.
         *       *       *   Oh, there is a charm
  Which morning has, that gives the brow of age
  A smack of youth, and makes the life of youth
  Shed perfume exquisite. Expect it not,
  Ye who till noon upon a down bed lie,
  Indulging feverous sleep."
                  HURDIS, _Village Curate_.

 "Watch and ward,
  And stand on your guard."
                            IZAAK WALTON.

 "Sleep is Nature's second course."


          UPON A LOOKING-GLASS.

 "Since as you know, you cannot see yourself
  So well as by reflection, I your glass
  Will modestly discover to yourself
  That of yourself which you yet know not of."
                            SHAKSPERE.


          FOR LADIES' BOUDOIRS.

 "_Birth_, _beauty_, _wealth_, are nothing worth alone.
  All these I would for good additions take:
  Tis the mind's beauty keeps the _others_ sweet."
          SIR THOMAS OVERBURY, _The Wife_.

 "'Tis beauty that doth oft make women proud;
 'Tis virtue that doth make them most admired;
 'Tis modesty that makes them seem divine."
                            SHAKSPERE.


          FOR A DRESSING-ROOM.

 "The apparel oft proclaims the man."


          FOR SCHOOLROOMS.

 "Extend generosity, it is profuseness;
  Confine economy, it is avarice;
  Unbridle courage, it is rashness;
  Indulge sensibility, it is weakness."

 "Catch Time by the forelock; he's bald behind."
 "Nothing is truly good that may be excell'd."
          _Motto of King Arthur's Table._

 "He may do what he will that will but do what he may."
                            ARTHUR WARWICK.

 "God dwelleth near about us,
    Ever within,
  Working the goodness,
    Consuming the sin."
          FULKE GREVILLE, LORD BROOKE, born 1554.


FOR LIBRARIES, STUDIES, AND BOOK-ROOMS.

"Reading furnishes the mind only with materials of knowledge: it is
thinking makes what we read ours. We are of the ruminating kind, and it is
not enough to cram ourselves with a great load of collections: unless we
chew them over again, they will not give us strength or
nourishment."--LOCKE.

"Crafty men contemn studies; simple men admire them; and wise men use
them."--BACON.

"Read not to contradict and refute, nor to believe and take for granted,
nor to find talk and discourse, but to weigh and consider."--_Idem._

"Some books are to be tasted, others to be swallowed, and some few to be
chewed and digested: that is, some books are to be read only in parts;
others to be read, but not curiously; and some few to be read wholly, and
with diligence and attention."--_Idem._

"In reading, we hold converse with the wise; in the business of life,
generally with the foolish."--BACON.

               "That place that does
  Contain my books, the best companions, is
  To me a glorious court, where hourly I
  Converse with the old sages and philosophers."
                            J. FLETCHER.

 "Bookes are a part of man's prerogative,
    In formal inke they thoughts and voyces hold,
  That we to them our solitude may give,
    And make time present travel that of old.
  Our life fame peceth longer at the end,
  And bookes it farther backward doe extend."
          SIR THOMAS OVERBURY, _The Wife_.

 "Books should for one of these four ends conduce,--
    For wisdom, piety, delight, or use."
                            SIR JOHN DENHAM.

 "Cease not to learne until thou cease to live;
    Think that day lost wherein thou draw'st no letter,
  Nor gain'st no lesson, that new grace may give
    To make thyself learneder, wiser, better."
          _Quadrains of Pibrac_, translated by JOSHUA SYLVESTER.

 "Who readeth much and never meditates,
    Is like a greedy eater of much food,
  Who so surcloyes his stomach with his cates,
    That commonly they do him little good."
                            _Ibid._

"Reading maketh a full man, conference a ready man, and writing an exact
man."--BACON'S _Essays--Of Studies_.

 "Calm let me live, and every care beguile,--
  Hold converse with the great of every time,
  The learn'd of ev'ry class, the good of ev'ry clime."
                  REV. SAMUEL BISHOP.

 "Of things that be strange,
    Who loveth to read,
  In these books let him range
    His fancy to feed."
                  RICHARD ROBINSON.


          FOR MUSEUMS OR LABORATORIES.

 "O mickle is the powerful grace that lies
  In herbs, plants, stones, and their true qualities:
  For nought's so vile that on the earth doth live,
  But to the earth some special good doth give."
                            SHAKSPERE.

 "Speak to the earth, and it shall teach thee."
                            SOLOMON.


          FOR A SURGICAL MUSEUM.

 "There is no theam more plentifull to scan,
  Than is the glorious, goodly frame of Man."
          JOSHUA SYLVESTER'S _Du Bartas_, 6th day.


          FOR JUSTICE-ROOMS.

 "'Tis not enough that thou do no man wrong,--
    Thou even in others must suppress the same,
  Righting the weake against th' unrighteous strong,
    Whether it touch his life, his goods, his name."
          _Quadrains of Pibrac_, trans. by JOSHUA SYLVESTER.

 "Upon the Law thy Judgments always ground,
    And not on man: For that's affection-less.
  But man in Passions strangely doth abound;
    Th' one all like God: Th' other too like to beasts."
                            _Id. cod._


          FOR CASINOS OR SUMMER-HOUSES.

   "Abusèd mortals, did you know
    Where joy, heart's ease, and comfort grow,
    You'd scorn proud towers,
    And seek them in these bowers;
  Where winds, perhaps, sometimes our woods may shake,
  But blustering care can never tempest make."
                            SIR HENRY WOTTON.

 "We trample grasse, and prize the flowers of May;
  Yet grasse is greene when flowers doe fade away."
                            ROBERT SOUTHWELL.

 "Blest who no false glare requiring,
  Nature's rural sweets admiring,
  Can, from grosser joys retiring,
  Seek the simple and serene."
                            ISAAC BICKERSTAFF.


          FOR A COUNTING-HOUSE.

            "Omnia Somnia."

 "Gae, silly worm, drudge, trudge, and travell,
          So thou maist gain
  Some honour or some golden gravell:
  But Death the while to fill his number,
          With sudden call
          Takes thee from all,
  To prove thy daies but dream and slumber."
          JOSHUA SYLVESTER, _Mottoes_.


          FOR OFFICES OR WORKSHOPS.

 "Have more than thou showest;
  Speak less than thou knowest;
  Lend more than thou owest;
  Learn more than thou trowest."
                            SHAKSPERE.

 "A spending hand that always poureth out,
    Had need to have a bringer-in as fast;
  And on the stone that still doth turn about
    There groweth no moss: these proverbs yet do last."
                            SIR T. WYATT.

 "How many might in time have wise been made,
    Before their time, had they not thought them so?
  What artist e'er was master of his trade
    Yer he began his prenticeship to know?

 "To some one act apply thy whole affection,
    And in the craft of others seldom mell;
  But in thine own strive to attain perfection,
    For 'tis no little honour to excell."
          _Quadrains of Pibrac_, translated by JOSHUA SYLVESTER.

 "If youth knew what age would crave,
  Youth would then both get and save."

 "Flee, flee, the idle brain;
    Flee, flee from doing nought;
  For never was there idle brain,
    But bred an idle thought."

"Get to live; then live to use it, else it is not true that thou hast
gotten."--G. HERBERT.

 "To him that is willing, ways are not wanting."


          FOR SHOPS.

 "Whoso trusteth ere he know,
  Doth hurt himself and please his foe."
                            SIR THOMAS WYATT.

 "Think much of a trifle,
    Though small it appear;
  Small sands make the mountain,
    And moments the year."


          FOR A BELL-TURRET.

 "We take no note of time
  But from its loss; to give it then a tongue
  Is wise in man."
          YOUNG'S _Night Thoughts_.


          FOR A BATHING-HOUSE.

 "Do not fear to put thy feet
  Naked in the river sweet;
  Think nor leach, or newt, or toad
  Will bite thy foot, where thou hast trod."
          BEAUMONT AND FLETCHER, _Faithful Shepherdess_.

With those still more admirable "legends" which may be selected from the
Bible I do not meddle. In it golden words of comfort and admonition lie
strewn so thickly, that error cannot be made by a selector. It may not be
amiss, also, for the illuminator to remember, that not unfrequently "a
verse may find him whom a sermon flies."

I cannot quit this portion of my theme without one word of summary, in the
way of advice, to the designer of illumination, on whatever material
applied. Briefly, then, let him eschew quaintness, and aim at beauty; let
him not shrink from beauty in old times because it was masked in
quaintness; but with a discriminating eye let him learn to winnow the chaff
from the wheat, and, scattering the one to the winds, let him garner up the
other in the storehouse of his memory, and for the sustenance of his
artistic life; and let him rest assured that the best designers, in all
ages, have been usually those who have gathered most widely and profoundly
from the failures, successes, and experiences of their predecessors.



PART II.

HOW ILLUMINATING MAY BE PRACTISED.

On analysis it will be found that this section of my Essay resolves itself
into three divisions, embracing respectively, 1stly, the ancient processes;
2ndly, the modern processes; and, 3rdly, the possible processes, not yet
introduced into common use. Of the last, I do not propose speaking in the
present work. Notices of the first of these might of course have been
presented in the historical manual; but, upon reflection, I considered it
would be most useful to the student to introduce them, in a collected form,
in this place; and for the following reasons:--1stly, In order that they
might not interrupt the thread of the narrative; and, 2ndly, because I
considered it desirable to put the ancient and modern processes in direct
contrast, so that the amateur might be the better enabled to reject what is
obsolete in the former, and to revive any which might appear to promise
greater technical excellence or facility than he might be enabled to obtain
through the employment of the latter.

I commence, therefore, with the _Ancient processes_.

Sir Charles Eastlake, who has profoundly studied the history and theory of
the subject, has justly remarked[102] the intimate relation which, in the
classical ages, existed between the physician and the painter,--the former
discovering, supplying, and frequently preparing, the materials used by the
latter. This ancient connection was not broken during those ages when
almost all knowledge and practice of either medicine or art were limited to
the walls of the cloister. The zealous fathers not only worked themselves
to the best of their ability, but delighted in training up their younger
brethren to perpetuate the credit and revenue derived from their skill,
knowledge, and labour, by the monasteries to which they were attached. "Nor
was it merely by oral instruction that technical secrets were communicated:
the traditional and practical knowledge of the monks was condensed in short
manuscript formulæ, sometimes on the subject of the arts alone, but oftener
mixed up with chemical and medicinal receipts. These collections, still
more heterogeneous in their contents as they received fresh additions from
other hands, were afterwards published by secular physicians, under the
title of 'Secreta.' The earliest of such manuals serve to show the nature
of the researches which were undertaken in the convents for the practical
benefit of the arts. Various motives might induce the monks to devote
themselves with zeal to such pursuits. It has been seen that their chemical
studies were analogous; that their knowledge of the materials fittest for
technical purposes, derived as it was from experiments which they had
abundant leisure to make, was likely to be of the best kind. Painting was
holy in their eyes; and, although the excellence of the work depended on
the artist, it was for them to ensure its durability. By a singular
combination of circumstances, the employers of the artist, the purchasers
of pictures (for such the fraternities were in the majority of cases), were
often the manufacturers of the painter's materials. Here, then, was another
plain and powerful reason for furnishing the best-prepared colours and
vehicles. The cost of the finer pigments was, in almost every case, charged
to the employer; but economy could be combined with excellence of quality,
when the manufacture was undertaken by the inmates of the convent."

All that is asserted in this passage with respect to painting, holds
equally good with regard to the materials requisite for the practice of the
Art of Illumination; and the same treatises which are illustrative of art
generally, almost invariably include specific instructions with regard to
the particular branch of it that I am now endeavouring to illustrate.

Fortunately, the series of these "Secreta" both commences from a remote
date, and is tolerably complete from that to a quite recent period.
Scattered allusions to the processes of art and industry may be met with in
the writings of several authors of the Alexandrian Neo-Platonic school in
the early ages of the Church, from whom the Byzantine Greeks, no doubt,
learnt much; but the most ancient collection on the subject is the treatise
of Heraclius, or Eraclius, "de Artibus Romanorum."[103] It would appear not
to have been written earlier than the 7th or later than the 10th
century,[104] its art being, as Mr. Robert Hendrie, the learned translator
and editor of the essay of Theophilus, of whom mention will presently be
made, observes, "of the school of Pliny, increased, it is true, by
Byzantine invention, but yet essentially Roman."[105] The next collection,
in point of age, is that published by Muratori,[106] and well known as the
"Lucca Manuscript," ascribed by Mabillon to the age of Charlemagne, and by
Muratori himself to a period certainly not later than the 10th century. Its
Latinity is barbarous, but I scarcely think I can do wrong in following the
translation of so careful a writer as Sharon Turner in the following
extracts, which treat of illumination, and give us a clear insight into the
practice of the school founded under the patronage of the great Frankish
emperor of the West.

The first I select refers to the preparation of the calf-skin.

  "Put it under lime and let it lie for three days; then stretch it, scrape
  it well on both sides, and dry it; then stain it with the colours you
  wish."

The second directs how skins may be gilt.

  "Take the red skin and carefully pumice and temper it in tepid water, and
  pour the water on it till it runs off limpid; stretch it afterwards, and
  smooth it diligently with clean wood. When it is dry, take the whites of
  eggs and smear it therewith thoroughly; when it is dry, sponge it with
  water, press it, dry it again, and polish it; then rub it with a clean
  skin, and polish it again and gild it."

Such gilding was effected with gold _leaf_, beaten out between small sheets
of "Greek parchment, which is made from linen cloth" (_i.e._ paper),
enclosed in vellum. White of egg was used as the mordant for fixing on the
gold.

The following two passages instruct the student in preparing gold for
writing:--

  "File gold very finely, put it in a mortar, and add the sharpest vinegar;
  rub it till it becomes black, and then pour it out. Put to it some salt
  or nitre, and so it will dissolve. So you may write with it; and thus all
  the metals may be dissolved.

  "Take thin plates of gold and silver, rub them in a mortar with Greek
  salt or nitre till it disappears. Pour on water and repeat it; then add
  salt, and so wash it. When the gold remains even, add a moderate portion
  of the flowers of copper and bullock's gall; rub them together, and write
  and burnish the letters."

The next and last, alludes to the amalgam, which appears to have been for
many centuries a favourite method of applying gold to parchment and other
surfaces.

  "Melt some lead, and frequently immerse it in cold water. Melt gold, and
  pour that into the same water, and it will become brittle. Then rub the
  gold filings carefully with quicksilver, and purge it carefully while it
  is liquid. Before you write, dip the pen in liquid alum, which is best
  purified by salt and vinegar."

In these instructions the student may distinctly recognize the processes
adopted in the production of those gilt texts on stained vellum grounds
which were so highly prized in the Carlovingian age.

In the writings of an ecclesiastic, probably nearly contemporary with the
Norman conquest, the monk Rugerus, or "Theophilus," we arrive at a really
perfect picture[107] of the arts of the 11th century. The first of the
three books into which his "Schedule of different Arts" is divided, is
dedicated entirely to painting. It contains forty chapters, of which thirty
refer to the preparation and application of pigments generally, both for
oil, tempera, and fresco painting, and ten to the various processes
connected with illumination. Of these, the following are the most
important:--

  CHAPTER XXX.

  OF GRINDING GOLD FOR BOOKS, AND OF CASTING THE MILL.

  When you have traced out figures or letters in books, take pure gold and
  file it very finely in a clean cup or small basin, and wash it with a
  pencil in the shell of a tortoise, or a shell which is taken out of the
  water. Have then a mill with its pestle, both cast from metal of copper
  and tin mixed together, so that three parts may be of pure copper, and
  the fourth of pure tin, free from lead. With this composition the mill is
  cast in the form of a small mortar, and its pestle round about an iron in
  the form of a knot, so that the iron may protrude of the thickness of a
  finger, and in length a little more than half a foot, the third part of
  which iron is fixed in wood carefully turned, in length about one yard,
  and pierced very straightly; in the lower part of which, however, of the
  length of four fingers from the end, must be a revolving wheel, either of
  wood or of lead, and in the middle of the upper part is fixed a leather
  strap, by which it can be pulled, and, in revolving, be drawn back. Then
  this mill is placed in a hollow, upon a bench fitted for it, between two
  small wooden pillars firmly fixed into the same bench, upon which another
  piece of wood is to be inserted, which can be taken out and replaced, in
  the middle of which, at the lower part, is a hole in which the pestle of
  the mill will revolve. These things thus disposed, the gold, carefully
  cleansed, is put into the mill, a little water added, and the pestle
  placed, and the upper piece of wood fitted, the strap is drawn and is
  permitted to revolve, again pulled, and again it revolves, and this must
  so be done for two or three hours. Then the upper wood is taken off, and
  the pestle washed in the same water with a pencil. Afterwards the mill is
  taken up, and the gold, with the water, is stirred to the bottom with the
  pencil, and is left a little, until the grosser part subsides; the water
  is presently poured into a very clean basin, and whatever gold comes away
  with the water is ground. Replacing the water and the pestle, and wood
  above being placed, again it is milled in the same way as before, until
  it altogether comes away with the water. In the like manner are ground
  silver, brass, and copper. But gold is ground most carefully, and must be
  lightly milled; and you must often inspect it, because it is softer than
  the other metals, that it may not adhere to the mill or the pestle, and
  become heaped together. If through negligence this should happen, that
  which is conglomerate is scraped together and taken out, and what is left
  is milled until finished. Which being done, pouring out the upper water
  with the impurities from the basin, wash the gold carefully in a clean
  shell; then pouring the water from it, agitate it with the pencil, and
  when you have had it in your hand for one hour, pour it into another
  shell, and keep that very fine part which has come away with the waters.
  Then again, water being placed with it, warm it and stir it over the
  fire, and, as before, pour away the fine particles with the water, and
  you may act thus until you shall have purified it entirely. After this
  wash with water the same refined part, and in the same manner a second
  and a third time, and whatever gold you gather mix with the former. In
  the same way you will wash silver, brass, and copper. Afterwards take the
  bladder of a fish which is called _huso_ (sturgeon), and washing it three
  times in tepid water, leave it to soften a night, and on the morrow warm
  it on the fire, so that it does not boil up until you prove with your
  finger if it adhere, and when it does adhere strongly, the glue is good.

  CHAPTER XXXI.

  HOW GOLD AND SILVER ARE LAID IN BOOKS.

  Afterwards take pure minium (red lead), and add to it a third part of
  cinnabar (vermilion), grinding it upon a stone with water. Which being
  carefully ground, beat up the clear of the white of an egg, in summer
  with water, in winter without water; and when it is clear, put the minium
  into a horn and pour the clear upon it, and stir it a little with a piece
  of wood put into it, and with a pencil fill up all places with it upon
  which you wish to lay gold. Then place a little pot with glue over the
  fire, and when it is liquefied, pour it into the shell of gold and wash
  it with it. When you have poured which into another shell, in which the
  purifying is kept, again pour in warm glue, and holding it in the palm of
  your left hand, stir it carefully with the pencil, and lay it on where
  you wish, thick or thin, so, however, that there be little glue, because,
  should it exceed, it blackens the gold and does not receive a polish; but
  after it has dried, polish it with a tooth or bloodstone carefully filed
  and polished, upon a smooth and shining horn tablet. But should it
  happen, through negligence of the glue not being well cooked, that the
  gold pulverizes in rubbing, or rises on account of too great thickness,
  have near you some old clear of egg, beat up without water, and directly
  with a pencil paint slightly and quickly over the gold; when it is dry
  again rub it with the tooth or stone. Lay in this manner silver, brass,
  and copper in their place and polish them.

The raised gold was not always produced by the mixture of red lead and
white of egg recommended by Theophilus. It was, especially in Italy,
frequently made of a composition of "gesso," or plaster, and in the 15th
century was often punctured all over by way of ornament. It may be
occasionally met with stamped over in patterns, with intaglio punches. This
"gesso raising," though very brilliant, possessed little tenacity, and in
many examples it has scaled off, while the more ancient "raising"
prescribed by Theophilus has adhered perfectly.

  CHAPTER XXXII.

  HOW A PICTURE IS ORNAMENTED IN BOOKS WITH TIN AND SAFFRON.

  But if you have neither of these (gold, silver, brass, or copper), and
  yet wish to decorate your work in some manner, take tin pure and finely
  scraped, mill it and wash it like gold, and apply it with the same glue,
  upon letters or other places which you wish to ornament with gold or
  silver; and when you have polished it with a tooth, take saffron, with
  which silk is coloured, moistening it with clear of egg without water,
  and when it has stood a night, on the following day cover with a pencil
  the places which you wish to gild, the rest holding the place of silver.
  Then make fine traits round the letters and leaves, and flourishes from
  minium, with a pen, also the stuffs of dresses and other ornaments.

  CHAPTER XXXIII.

  OF EVERY SORT OF GLUE FOR A PICTURE OF GOLD.

  If you have not a bladder (of the sturgeon), cut up thick parchment or
  vellum in the same manner,--wash and cook it. Prepare also the skin of an
  eel carefully scraped, cut up and washed in the same manner. Prepare thus
  also the bones of the head of the wolf-fish washed and dried, carefully
  washed in water three times. To whichever of these you have prepared, add
  a third part of very transparent gum, simmer it a little, and you can
  keep it as long as you wish.

  CHAPTER XXXIV.

  HOW COLOURS ARE TEMPERED FOR BOOKS.

  These things thus accomplished, make a mixture of the clearest gum and
  water as above, and temper all colours except green, and ceruse, and
  minium, and carmine. Salt green is worth nothing for books. You will
  temper Spanish green with pure wine, and if you wish to make shadows, add
  a little sap of iris, or cabbage, or leek. You will temper minium, and
  ceruse, and carmine, with clear of egg. Compose all preparations of
  colours for a book as above, if you want them for painting figures. All
  colours are laid on twice in books, at first very thinly, then more
  thickly; but twice for letters.

The next extract I give is of great interest in the technical history of
illumination, on three accounts: firstly, because it guides the student to
recognize in madder the purple stain and colour, so highly prized in the
early periods of the art; secondly, because it shows him the manner in
which fugitive vegetable tints were protected from the decomposing
influence of the atmosphere by an albuminous varnish; and thirdly, because
it illustrates the ordinary modern processes of under painting, and glazing
with transparent colour. The "folium" of the Greek illuminators was
procured from plants growing abundantly near Athens, while that of the
Hiberno-Saxon scribes was obtained from the "norma" or "gorma" of the
Celts. Mr. Hendrie, in his learned notes to Theophilus, has traced
successive recipes for the preparation of "folium," in which the identity
of the base giving the colouring matter is clearly established. It is
curious that the collections of "Secreta" should give as the only countries
supplying the materials for making "folium," those two in which the use of
the bright purple stain ascends to the very earliest of their decorated
manuscripts. The following is the description given by Theophilus:--

  CHAPTER XXXV.

  OF THE KINDS AND THE TEMPERING OF FOLIUM.

  There are three kinds of folium, one red, another purple, a third blue,
  which you will thus temper. Take ashes, and sift them through a cloth,
  and, sprinkling them with cold water, make rolls of them in form of
  loaves, and placing them in the fire, leave them until they quite glow.
  After they have first burnt for a very long time, and have afterwards
  cooled, place a portion of them in a vessel of clay, pouring urine upon
  them and stirring with wood. When it has deposed in a clear manner, pour
  it upon the red folium, and, grinding it slightly upon a stone, add to it
  a fourth part of quick lime, and when it shall be ground and sufficiently
  moistened, strain it through a cloth, and paint with a pencil where you
  wish, thinly; afterwards more thickly. And if you wish to imitate a robe
  in a page of a book, with purple folium; with the same tempering, without
  the mixture of lime, paint first with a pen, in the same page, flourishes
  or circles, and in them birds or beasts, or leaves; and when it is dry,
  paint red folium over all, thinly, then more thickly, and a third time if
  necessary; and afterwards paint over it some old clear of egg. Paint over
  also with glaire of egg, draperies, and all things which you have painted
  with folium and carmine. You can likewise preserve the burned ashes which
  remain for a long time, dry.

I conclude the series of receipts extracted from Theophilus[108] by one not
further bearing upon the Art of Illumination, than as proving the nature of
the ink which has generally retained its colour so wonderfully in the
ancient manuscripts.

  CHAPTER XL.

  OF INK.

  To make ink, cut for yourself wood of the thorn-trees in April or May,
  before they produce flowers or leaves, and collecting them in small
  bundles, allow them to lie in the shade for two, three, or four weeks,
  until they are somewhat dry. Then have wooden mallets, with which you
  beat these thorns upon another piece of hardwood, until you peel off the
  bark everywhere, put which immediately into a barrelful of water. When
  you have filled two, or three, or four, or five barrels with bark and
  water, allow them so to stand for eight days, until the waters imbibe all
  the sap of the bark. Afterwards put this water into a very clean pan, or
  into a cauldron, and fire being placed under it, boil it; from time to
  time, also, throw into the pan some of this bark, so that whatever sap
  may remain in it may be boiled out. When you have cooked it a little,
  throw it out, and again put in more; which done, boil down the remaining
  water unto a third part, and then pouring it out of this pan, put it into
  one smaller, and cook it until it grows black and begins to thicken, add
  one third part of pure wine, and putting it into two or three new pots,
  cook it until you see a sort of skin show itself on the surface; then
  taking these pots from the fire, place them in the sun until the black
  ink purifies itself from the red dregs. Afterwards take small bags of
  parchment carefully sewn, and bladders, and pouring in the pure ink,
  suspend them in the sun until all is quite dry; and when dry, take from
  it as much as you wish, and temper it with wine over the fire, and,
  adding a little vitriol, write. But, if it should happen through
  negligence that your ink be not black enough, take a fragment of the
  thickness of a finger, and putting it into the fire, allow it to glow,
  and throw it directly into the ink.

The next collection of "Secreta," in point of importance and probably
antiquity, is the "Mappæ Clavicula," or "little key to drawing," a
manuscript treatise on the preparation of pigments, and on various
processes of the decorative arts practised during the Middle Ages, in the
possession of Sir Thomas Phillipps, of Middle Hill.[109] The proprietor of
the volume, Mr. Hendrie, Sir Charles Eastlake, and (last, not least) Mr.
Albert Way, agree in considering it highly probable that it may be an
English collection, probably of about the reign of Henry II. Like the
"Schedula" of Theophilus, it presents a very miscellaneous series of
recipes, and tends to prove, what is very generally believed by the
learned, that the "Masters of Arts" of old were frequently skilled, not in
special departments of production, such as the modern division-of-labour
system has created, but in multifarious avocations, such as we should not
now readily recognize as likely to be practised by any single individual.

These collections remarkably illustrate the class of knowledge likely to
have been possessed by such apparently versatile geniuses as St. Dunstan,
St. Eloi, Bernward of Hildesheim, Tutilo the monk of St. Gall, and many
others. The author of the "Mappæ Clavicula," in a few lines of poetical
introduction to his teachings, defines the first necessity for painters to
be, a knowledge of the manufacture of colours, then a command over the
various modes of mixing them, then dexterity in using and heightening them
in different kinds of work; and, ultimately, he commends to their attention
a variety of information for the advancement of art generally, derived from
the writings of many learned men.--"Sicut liber iste docebit." Thus under
two hundred and nine heads, but with some tautology, he proceeds to treat,
as Sir Thomas Phillipps observes, not only of the composition of colours,
but "of a variety of other subjects, in a concise and simple manner, and
generally very intelligibly; as for instance, architecture, mensuration of
altitudes, the art of war, &c." Among the recipes, in addition to those
referring to pigments, are many relating to illuminating. The following,
for instance, is curious as defining clearly what were the best and most
important tints for illumination:--

  _Of different Colours._

  "These colours are clear and full-bodied for vellum:--Azorium (azure),
  Vermiculum (vermilion), Sanguis Draconis (dragon's blood), Carum (yellow
  ochre), Minium (red lead), Folium (madder purple), Auripigmentum
  (orpiment), Viride Græcum (acetate of copper), Gravetum Indicum (indigo),
  Brunum (brown), Crocus (yellow), Minium Rubeum vel Album (red or white
  lead), Nigrum Optimum ex carbone vitis (the best black made from
  carbonized vine twigs); all these colours are mixed with white of egg."

The mixture of colours appears to have been reduced to a perfect system,
each hue having others specially adapted and used, for heightening and
lowering the pure tint; thus the author gives directions which are likely
to be scarcely less useful to the illuminator of the present day than they
were to those of old.

  _Of Mixtures._

  "If, therefore, you should desire to know the natures and mixtures of
  these [the above given] colours, and which are antagonistic to each
  other, lend your ear diligently.

  "Mix azure with white lead, lower with indigo, heighten with white lead.
  Pure vermilion you may lower with brown or with dragon's blood, and
  heighten with orpiment. Mix vermilion with white lead, and make the
  colour which is called _Rosa_, lower it with vermilion, heighten it with
  white lead. Item, you may make a colour with dragon's blood and orpiment,
  which you may lower with brown, and heighten with orpiment. Yellow ochre
  you may lower with brown, and heighten with red lead (query, with white).
  Item, you may make Rosam[110] of yellow ochre and white lead, deepen with
  yellow ochre, heighten with white lead. Reddish purple (folium) may be
  lowered with brown, and heightened with white lead. Item, mix folium with
  white lead, lower with folium, and heighten with white lead. Orpiment may
  be lowered with vermilion, but cannot be heightened, because it stains
  all other colours."

  _Of Tempering._

  "Greek green you will temper with acid, deepen with black, and heighten
  with white, made from stag's horn (ivory black). Mix green with white
  lead, deepen with pure green, and heighten with white lead. Greenish
  blue, deepen with green, heighten with white lead. Yellow, deepen with
  vermilion, heighten with white lead. Indigo, deepen with black, heighten
  with azure. Item, mix indigo with white, deepen with azure, heighten with
  white lead. Brown, deepen with black, heighten with red lead. Item, make
  of brown and white lead a drab (Rosam), lower with brown, heighten with
  white lead. Item, mix yellow with white lead, lower with yellow, heighten
  with white lead. Lower red lead with brown, heighten with white lead.
  Item, red lead with brown, deepen with black, heighten with red lead.
  Item, you may make flesh-colour of red lead and white, lower with
  vermilion, heighten with white lead."

  _Which Colours are Antagonistic._

  "If you wish to know in what manner colours are antagonistic, this is it.
  Orpiment (sulphuret of arsenic) does not agree with purple (folio), nor
  with green (acetate of copper), nor with red lead, nor white lead. Green
  does not agree with purple.[111]

  "If you wish to make grounds, make a fine rose-colour of vermilion and
  white. Item, make a ground of purple mixed with chalk. Item, make a
  ground of green mixed with vinegar. Item, make a ground of the same
  green, and when it shall have become dry, cover it with size ('caule').

  "If you wish to write in gold, take powder of gold and moisten it with
  size, made from the very same parchment on which you have to write; and
  with the gold and size near to the fire; and, when the writing shall be
  dry, burnish with a very smooth stone, or with the tooth of a wild boar.
  Item, if then you should wish to make a robe or a picture, you may apply
  gold to the parchment, as I have above directed, and shade with ink or
  with indigo, and heighten with orpiment."

The above are the principal passages in the "Mappæ Clavicula," which supply
deficiencies in most other books of Secreta; and I have translated them at
length, both on account of the accuracy with which I have found the
directions followed in ancient illuminated manuscripts, and because I
believed that a knowledge of this ancient scale of colours might greatly
facilitate accurate copying from old examples. I need scarcely say, that as
the art of painting improved in Italy and the Netherlands, the
illuminator's palette became enriched with several new and very brilliant
colours;--such as the ultramarines and carmines (exceedingly scarce in
early manuscripts) which make the books produced at Rome and in Northern
Italy, during the 16th and 17th centuries, glow with a vivacity never
previously attained. Every improvement made in one country was, however,
speedily, communicated through these very art-treatises to other countries,
and thus we find lakes and carmines freely used in England during the 15th
century.[112] Ultramarine, indeed, forms the special subject of an essay by
a Norman, comprised among the Le Bègue MSS. (already referred to), under
the following title, which proves its novelty in Western Europe, at the
beginning of the 15th century:--

"Anno 1411, Johannes de ... [illegible] Normannus de _Azurro novo_, lapidis
lazulli ultramarini."

The next collection of Secreta in importance, and probably in date to the
"Mappæ Clavicula," is that of a Frenchman, Peter de St. Audemar. "With this
treatise," observes Sir Charles Eastlake, "may be classed a similar one in
the British Museum, written in the 14th century," but treating of a
somewhat earlier practice in art. The identity of the colours for, and
practice of, painters on wall and panel, and illuminators on vellum, is
proved by the instructions to both being almost invariably given in the
same books. Thus, the volume last mentioned commences--"Incipit tractatus
de coloribus Illuminatorum _seu_ Pictorum"--as though there existed no
practical distinction between the two. Another manuscript, of later date,
also in the Le Bègue collection, exhibits, in its title even, a curious
picture of the industry with which the Art of Illumination was studied in
the principal countries of Europe,--introducing the student to a scribe,
actually keeping a school at Milan. Thus, "Liber Johannis Acherius, A.D.
1398. Ut accessit a Jacobo Cona, Flamingo pictore:--Capitula de coloribus
ad illuminandum libros ab eodem Archerio sive Alcherio, ut accessit ab
Antonio de compendio illuminatore librorum in Parisiis et a Magistro
Alberto Pozotto perfectissimo in omnibus modis scribendi, Mediolani scholas
tenente."

Here we have, in a few lines, evidence of the concurrence of no less,
probably, than four distinct nationalities to make up one set of
instructions. However illuminated manuscripts may differ in style from each
other, according to the countries in which they may have been produced, the
technical processes, from the commencement of the 15th century, scarcely
differed at all, probably through the general spread of these "handbooks of
the Middle Ages."

From the 14th century onwards, the treatises, or rather probably composite
transcripts from earlier treatises, multiply greatly; so far, however, as I
have been able to make out from the able analysis made by Sir C. Eastlake,
Mr. Hendrie, and Mrs. Merrifield, of many, they contain little more
information than is conveyed in the extracts already given. Some curious
details, however, may be gathered as to the London practice in the 15th
century, which may interest the reader. A manuscript, written in German, as
is believed at that date, is preserved in the public library at Strasburg,
which distinctly proves that the colours for illuminating were commonly
preserved by steeping small pieces of linen in the tinted extracts,
sometimes mixed with alkaline solutions. The process is minutely described
in this MS.; the dyes so prepared are there called "tüchlein varwen,"
literally "clothlet colours." The following passage from another
compendium, a Venetian MS., gives the result in few words:--"When the
aforesaid pieces of cloth are dry, put them in a book of cotton paper, and
keep the book under your pillow, that it may take no damp; and when you
wish to use the colours, cut off a small portion [of the cloth], and place
it in a shell with a little water, the evening before. In the morning the
tint will be ready, the colour being extracted from the linen." This
practice is alluded to by Cennini, when he says:--"You can shade with
colours, and by means of small pieces of cloth, according to the process of
the illuminators."

The German compiler, speaking of the preparation of a blue colour in this
mode, says, "If you wish to make a beautiful clothlet blue colour according
to the London practice," &c. After describing the method of preparing it,
he adds:--"These [pieces of cloth] may be preserved fresh and brilliant,
without any change in their tints, for twenty years; and this colour, in
Paris and in London, is called [blue] for missals, and here in this country
clothlet blue; it is a beautiful and valuable colour."

"The place denominated _Lampten_, mentioned together with _Paris_, can be
no other than London."[113]

As pursuing the subject of ancient processes further than I have now done,
would scarcely he profitable to the student, I proceed to the second
division of this part of my subject, and accordingly take up _the modern
processes_. In offering the following details on this subject, however, to
the amateur's attention, I would not for one moment let it be supposed that
a knowledge of them alone will be sufficient to make him an efficient
illuminator. Fortunately many very excellent artists have of late devoted
themselves to giving instruction in the practical manipulation of the art,
and amateurs cannot do better than place themselves at once in
communication with masters, whose addresses may be obtained at the shops of
the principal artists' colourmen. There will still be, no doubt, in
different parts of the country, many desirous of illuminating, and yet
unable to obtain the benefit of seeing a practised hand work before them,
or even to pick up information as to the _modus operandi_. To such, at
least, the following observations may prove useful.[114]

The two great sections into which all the processes by which illumination
of any kind may be executed, divide themselves, are--1st, those in which
water and glutinous substances soluble in water form the vehicles for
applying the pigments, and causing them to adhere to the surfaces on which
they may be applied; and 2ndly, those in which oil or spirit, and resins,
or other substances which combine readily with such fluids, are made to
perform corresponding functions. The pigments, reduced to an impalpable
powder, are the same in both classes of processes, which are commonly known
as watercolour-painting and oil-painting. That which was of old the
artist's greatest stumbling-block--the manufacture and preparation of his
pigments--need now no longer occasion him the slightest embarrassment; for
every colour with which his palette could be enriched is to be bought,
ready prepared, of the principal artists' colourmen. In like manner every
other essential for his use is now freely at his command; and all that is
required on his part is knowledge how to employ the materials which others
most dexterously and carefully place at his disposal.

In commencing the collection of that information which I am now
endeavouring to communicate, I felt it my duty to enter into correspondence
with all those manufacturers whose products I had at different times
personally tested; and I accordingly addressed myself to the following,
whose materials, with insignificant exceptions, I have invariably found
satisfactory, both in nature and quality.

  R. ACKERMAN, 191, Regent-street, W.
  L. BARBE, 60, Quadrant, Regent-street, W.
  J. BARNARD, 339, Oxford-street, W.
  Messrs. BRODIE & MIDDLETON, 79, Long-acre, W.C.
  H. MILLER, 56, Long-acre, W.C.
  J. NEWMAN, 24, Soho-square, W.
  Messrs. REEVES & SONS, 113, Cheapside, E.C.
  Messrs. ROBERSON, 99, Long-acre, W.C.
  Messrs. ROWNEY & CO., 51, Rathbone-place, W.
  Messrs. SHERBORNE & TILLYER, 321, Oxford-street, W.
  Messrs. WINSOR & NEWTON, 38, Rathbone-place, W.

From each of the above-mentioned firms I have obtained valuable
information, and from several, excellent samples of their products. I am
glad, therefore, to take the present opportunity of expressing my
obligations to them. From Messrs. Winsor & Newton, especially, I have
received the kindest and most intelligent co-operation; and I am happy to
be the channel of making public the results of a series of experiments, on
the combinations of colours and the use of various materials for
illuminating purposes, suggested by me, and made with great tact and
judgment by Mr. W. H. Winsor. Messrs. Winsor & Newton and Mr. Barnard have,
up to the present time, done most to smooth away the difficulties which
beset the illuminator. Messrs. Newman, Messrs. Rowney & Co., Messrs. Reeves
& Sons, and Mr. Barbe, have also recently contributed valuable improvements
or special adaptations.[115]

The colours best suited for illuminating I believe to be as follows:--

  B Lemon Yellow             }
  A Gamboge                  }    Yellow.
  A Cadmium Yellow           }
  D Mars Yellow              }

  B Rose Madder              }
  A Crimson Lake             }
  C Carmine                  }    Red.
  C Orange Vermilion         }
  A Vermilion                }

  A Cobalt                   }
  A French Blue              }    Blue.
  D Smalt                    }

  D Mars Orange              }    Orange.
  B Burnt Sienna             }

  C Burnt Carmine            }    Purple.
  D Indian Purple            }

  A Emerald Green            }    Green.
  C Green Oxide of Chromium  }

  B Vandyke Brown                 Brown.

  A Lampblack                     Black.

  A Chinese White                 White.

These colours are selected from the list of water-colours made at the
present day (upwards of eighty), and will, I think, be found to be all that
can well be required for illuminating. The whole number is by no means
indispensable, and I have therefore marked by different letters of the
alphabet,--1st, A, those without which it would be useless to commence
work; 2ndly, B, those which should first be added; 3rdly, C, those which
are required for very great brilliancy in certain effects; and, 4thly, D,
those which may be regarded as luxuries in the art. The C are really
important; the D are much less so. Messrs. Winsor & Newton have arranged
them into four different lists, which are placed in boxes (complete with
colours and materials for working in water-colours), of the respective
retail values of £1. 1_s._, £1. 11_s._ 6_d._, £2. 2_s._, and £3. 3_s._
Boxes corresponding with, or slightly varying from these, in selection of
colours and materials, may be obtained from other artists' colourmen.

I now proceed to notice these colours _seriatim_, in reference to their
tints, both when used alone and when mixed with other colours.


  YELLOWS.

  _Lemon Yellow._--A vivid high-toned yellow, semi-opaque, is extremely
  telling upon gold. Mixed with cadmium yellow it furnishes a range of
  brilliant warm yellows. It mixes well with gamboge, orange vermilion,
  cobalt, emerald green, and oxide of chromium, and with any of these
  produces clean and useful tints.

  _Gamboge._--A bright transparent yellow of light tone; works freely, and
  is very useful for glazing purposes. In combination with lemon yellow it
  affords a range of clean tints. When mixed with a little Mars yellow it
  produces a clear, warm, transparent tone of colour.

  _Cadmium Yellow._--A rich glowing yellow, powerful in tint, and
  semi-transparent. This is a most effective colour for illuminating. When
  judiciously toned with white, it furnishes a series of useful shades.
  Mixed with lemon yellow it produces a range of clean vivid tints. It does
  not, however, make good greens--they are dingy. Mixed with carmine, or
  glazed with it, it gives a series of strong luminous shades.

  _Mars Yellow._--A semi-transparent warm yellow, of slightly russet tone,
  but clean and bright in tint. Useful where a quiet yellow is required;
  mixes well with gamboge; does not make good greens.


  REDS.

  _Rose Madder._--A light transparent pink colour of extremely pure tone.
  It is delicate in tint, but very effective, on account of its purity.
  Mixed with cobalt, it affords clean, warm, and cold purples. The addition
  of a little carmine materially heightens the tone of this colour, though
  at the same time it somewhat impairs its purity.

  _Crimson Lake._--A rich crimson colour, clean and transparent; washes and
  mixes well. More generally useful than carmine, though wanting the
  intense depth and brilliancy of the latter colour.

  _Carmine._--A deep-toned luminous crimson, much stronger than crimson
  lake; is clean and transparent. The brilliancy of this powerful colour
  can be increased, by using it over a ground of gamboge.

  _Orange Vermilion._--A high-toned opaque red, of pure and brilliant hue,
  standing in relation to ordinary vermilion as carmine to crimson lake. It
  is extremely effective, and answers admirably where vivid opaque red is
  required; it works, washes, and mixes well. Its admixture with cadmium
  results in a fine range of warm luminous tints. When mixed with lemon
  yellow, it furnishes a series of extremely clean and pure tints; when
  toned with white, the shades are clear and effective. This is a most
  useful colour.

  _Vermilion._--A dense, deep-toned red, powerful in colour, and opaque. It
  is not so pure in tone as orange vermilion, and is of most service when
  used alone; it can, however, be thinned with white and with yellows.


  BLUES.

  _Cobalt Blue._--A light-toned blue, clean and pure in tint, and
  semi-transparent. This is the lightest blue used in illuminating, and by
  the addition of white can he "paled" to any extent, the tints keeping
  clear and good. Mixed with lemon yellow, it makes a clean useful green.
  Its admixture with gamboge is not so satisfactory, and the green produced
  by its combination with Mars yellow is dirty and useless. With rose
  madder it produces middling, warm and cold purples (_i.e._, marones, and
  lilacs or violets); with crimson lake, strong and effective ones; with
  carmine, ditto. A series of quiet neutral tints can be produced by its
  admixture with orange vermilion. The tints in question are clean and
  good, and might occasionally be useful.

  _French Blue._--A deep rich blue, nearly transparent; is the best
  substitute for genuine ultramarine. The greens it makes with lemon
  yellow, gamboge, cadmium, and Mars yellow, are not very effective or
  useful. The violets and marones it forms with rose madder are granulous
  and unsatisfactory; with carmine they are somewhat better; but those
  formed with crimson lake are very good.

  _Smalt._--A brilliant, full-toned blue; deep in tone, and nearly
  transparent; luminous and very effective when used alone. It is
  granulous, and does not wash or mix well. The greens it makes are not
  particularly useful.


  ORANGES.

  _Mars Orange._--A brilliant orange of very pure tone, transparent and
  lighter in colour than burnt sienna; and is not so coarse or staring. An
  effective and useful colour.

  _Burnt Sienna._--A deep rich orange, transparent and effective; works
  well and mixes freely.


  PURPLES.

  _Indian Purple._--A rich deep-toned violet, or cold purple colour; most
  effective when used alone. Can be lighted with French blue or cobalt, and
  the tints will be found useful.

  _Burnt Carmine._--A rich deep-toned marone or warm purple colour;
  transparent and brilliant; luminous and effective when used alone; mixed
  with orange vermilion, it produces a strong rich colour, and a quiet
  fleshy one when mixed with cadmium yellow.


  GREENS.

  _Emerald Green._--An extremely vivid and high-toned green, opaque. No
  combination of blue and yellow will match this colour, which is
  indispensable in illuminating. It can be "paled" with white, and the
  tints thus produced are pure and clean. The tints afforded by its
  admixture with lemon-yellow are also clear and effective.

  _Green Oxide of Chromium._--A very rich deep green, opaque, but
  effective. The tone of this green renders it extremely useful in
  illuminating; mixed with emerald green, it furnishes a series of rich
  semi-transparent tints. Mixed with lemon-yellow, it gives quiet, useful
  shades of green; and when this combination is brightened with emerald
  green, the shades are luminous and effective.


  BROWN.

  _Vandyke Brown._--A deep, rich, transparent, brown, luminous and clear in
  tint; works, washes, and mixes well. The best of all the browns for
  illuminating.


  BLACK.

  _Lampblack._--The most dense and deep of all the blacks, free from any
  shade of brown or grey.


  WHITE.

  _Chinese White._--A preparation of oxide of zinc, permanent, and the
  white best adapted for illuminating. It is not only useful _per se_, but
  is indispensable for toning or reducing other colours.

In making the list of the colours just described, I have assumed as a _sine
quâ non_ that the colours used in illuminating should be permanent. All
those enumerated are so (in water-colours), with the exception of carmine
and crimson lake; and these, though theoretically not permanent, are yet
found in practice to be _very_ lasting, especially when not too much
exposed to the light. It is a curious fact, that crimson lake, though a
weaker colour than carmine, is yet more permanent, in consequence of its
different base, and that it will better stand exposure to light.

I here take the opportunity of warning amateurs, allured by their evident
brilliancy, against the use, in illumination, of the following five
colours, viz.--pure scarlet, red lead, chrome yellow, deep chrome, and
orange chrome. None of these are permanent; the first-named being fugitive,
and the others in time turning black; but this is the less to be regretted,
as there are permanent colours answering equally well for illumination. Of
course, these are less fugitive in books, which are generally protected
from the action of light and air, than they would be in pictures.

The preceding remarks on pigments apply, with no difference worth noting,
to colours prepared either for oil or for water-colour; which may therefore
be laid on, by varying the vehicle for their proper application, to the
surfaces of any of those materials which have been specified in the Second
Part of this Essay, as available for different kinds of illumination. I now
proceed to notice the special processes requisite in each case, commencing
with those which may be best employed for vellum. This substance consists
of calf-skin, carefully cleansed and scraped, and repeatedly washed in
diluted sulphuric acid. The surface is rubbed down with fine pumice-stone
to a smooth face, and in that condition it is fit for working upon. It is
sold, prepared for use, at all the principal shops. If it has not been
previously strained, or if many tints are likely to be floated over the
surface, it will be well to strain it down upon a strainer or board before
attempting to draw upon it. This may be done by damping the vellum, and
then either gluing or nailing its edges down. When dry, it will be found to
lie perfectly flat and smooth. It may be well, then, to wash it over with a
dilute preparation of ox gall, to overcome any possible greasiness, and
prepare it to receive colour freely. Mr. Barnard, and, I believe, other
artists' colourmen, supply vellum mounted in block-books, similar to those
made up of drawing-paper for sketching on; and by providing himself with
one of those, the amateur may avoid the trouble of having to mount his own
vellum.

As it is by no means easy to remove pencil-marks from vellum (and indeed it
is never wise to attempt it, for the black-lead unites with the animal fat,
which can never be entirely got out of the material, and rubs under the
action of Indian-rubber or bread into a greasy smudge), it is always well
to set out the design in the first instance upon drawing-paper. The best
mode for good work is to complete the outline on drawing-paper, and then to
trace it carefully with a hard pencil on a piece of tracing-paper, about
one inch larger each way than the entire surface of the vellum; then cut
out, the exact size of the vellum, a piece of tracing or tissue paper,
rubbed evenly over with powdered red chalk.[116] Lay the tracing down
(pencilled side upwards) in its right place upon the vellum, and fasten
down one edge with pins, gum, or mouth-glue. Then slip the transfer-paper,
with the chalked side downwards, between the vellum and the tracing until
it exactly covers the former--touching the back of the transfer-paper with
two or three drops of gum on its margin. Then lay the tracing over, and
fasten down another of its edges. The gum drops will prevent the
transfer-paper slipping away from the tracing-paper, when the drawing-board
or strainer is placed upon a sloping desk or easel. Taking care to keep a
piece of stout card or pasteboard under the hand, go over all the lines of
the tracing with a blunted etching-point, or very hard pencil cut sharp.
This having been done, on removing both the tracing and the transfer-paper,
it will be found that a clear red outline has been conveyed to the surface
of the vellum. At this stage of the work, as nothing dirties more readily
than this material, it will be well to fasten over the surface a clean
sheet of paper with a flap cut in it, by raising up and folding back
portions of which, the artist may get to the part of the surface upon which
he may desire to work without exposing any of the rest. As the effect of
the writing on the page gives as it were the key-note for the general
effect of the illuminated ornaments, it will be well to complete the former
before proceeding to the latter.[117]

If the lines of the writing fixed upon are fine and delicate, they will
look best, and work most freely with Indian ink; but if they are bold and
solid, involving some extent of black surface, they will present a better
appearance if wrought in lampblack; the principal difference between the
two being that Indian ink is finer, and, if good, always retains a slight
gloss, while lampblack gives a fuller tint, and dries off quite mat, or
with a dead surface, corresponding with that of most other body-colour
tints used in illuminating. Great care must be taken to keep the writing
evenly spaced, upright, and perfectly neat, as it is almost impossible to
erase without spoiling the vellum, and as no beauty of ornament will redeem
an untidy text. If a portion of the writing is to be in red, it should be
in pure vermilion; and if in gold, it should be highly burnished, as will
be hereafter directed. The writing being satisfactorily completed, the
artist may proceed to lay in his ground tints, generally mixing them with
more or less white to give them body and solidity. Colours prepared with
water are best adapted for illumination on vellum; and those known as
_moist_ colours are to be preferred for this work, as they give out a
greater volume of colour, and possess more tenacity or power of adhering to
the surface of the material on which they are used than the dry colours. Of
moist colours there are two descriptions, viz., solid and liquid; and of
these I give the preference to the former, as some colours, such as lemon
yellow and smalt, will not keep well in tubes; added to which, there is
waste in using them in this form where only small quantities are required,
as the colour cannot be replaced in the tube when once squeezed out. The
tube colours possess, however, the valuable property of being always clean
when a bit of pure colour is required. The solid moist colours are apt to
get dirtied in rapid working, and occasionally mislead an eye which is not
quick at detecting a lowered tint. Mr. Barbe's body-colours, which are of
very good quality, are prepared in powder, combined with a glutinous
substance, on moistening which with water, the tints are fit for
application. Messrs. Winsor & Newton's body-colours are also very
excellent. Flatness of tint is best secured by using the first colour well
mixed with body, and put on boldly; this forms the brightest tint; then
shade with pure transparent colour, and finish off with the high lights.

Very useful models, both on a small scale for book illumination, and on an
enlarged one for wall decoration, are now prepared by several of the
artists' colour-men, for teaching amateurs the different modes of shading,
&c. They consist of outline plates partially coloured by hand. The beginner
will find it a very useful exercise to complete a few of these before
trying his hand upon more original works upon vellum. The greatest care
must be taken to have every implement perfectly clean. Experience alone can
teach the artist the value of what are called glazing or transparent
colours, such as the lakes, carmine, madders, gamboge, &c. Some tints may
be used either as glazing colours, or as body-tints, according to their
preparation, and according to the degree of thickness with which they are
applied. As a general principle, all shades should be painted in
transparent colour, all lights in opaque. Reflected lights may often be
best given by scumbling thin body-colour over transparent shade. In order
to prepare the tints for these operations, it may be well to use a little
of Newman's or Miller's preparations with them. The less tints are
retouched after the first application, the more clear and brilliant they
are likely to remain. Above all things never let the paint-brush go near
the mouth, and never attempt to correct or retouch a tint while it is in
process of drying, as doing so will infallibly make it look streaky and
muddy. In all these processes of manipulation, however, practice, good
example, and good tuition, must teach what the minutest directions would
fail to satisfactorily convey. The principal colours having been applied,
the next difficulty will be to heighten them with gold and silver. Any
large surfaces of gilding it will be well to apply previously to commencing
colouring, and as much as possible intended for burnishing.

The principal metallic preparations used in illumination may be enumerated
as follows:--gold leaf, gold paper, shell gold, saucer gold, gold paint,
silver leaf, shell silver, and shell aluminium. Of these, the leaves,
paper, and paint, are of English, and the shells and saucers of French
manufacture. Occasionally gold and silver powder and German-metal leaf are
employed, though too rarely to make them important enough to claim general
notice.

The first-mentioned preparation of gold--gold-leaf--is the pure metal
beaten into very thin leaves, generally 3-1/8 inches, 3¼ inches, or 3-3/8
inches square; but for illuminating purposes it should be still
smaller--say 2½ inches square, as it is easier to handle than a larger
size. For the same reason it is better to have the leaf double as thick as
it is usually beaten. Gold leaf is sold in "books," each of which contains
twenty-five gold "leaves," and for ordinary and general purposes, it is by
far the best and most useful metallic preparation; but the difficulty of
handling and laying it on deters amateurs from employing it, and it is
difficult in writing to furnish a practical description of the _modus
operandi_. The following is the usual mode:--

"Carefully open the book of gold, and if in so doing you disturb the leaf,
gently blow it down flat again. If a whole leaf be required, take a rounded
'tip,' and quietly so place it on the leaf that the top of the tip be close
to the edge of the leaf. In so doing, the sides of the tip will be brought
down upon the side edges of the leaf, which then can be securely taken up
and placed where required. If a small piece of gold leaf only be wanted,
cautiously take up a leaf from the book by passing a 'gilder's knife'
underneath, and place it on a 'gilders cushion;'[118] lay it flat with the
knife, with which then cut the piece of the size required. If when you have
laid gold leaf down with the tip it be wrinkly, blow it down flat."

The "gilder's tip" spoken of in the above extract is a very thin camel-hair
brush, and for unskilled hands a semicircular tip is to be preferred to one
of the ordinary form; as with it a leaf of gold may be firmly laid hold of,
balanced, adjusted, and placed, without needing any particular knack. For
long narrow pieces of gold, the ordinary gilder's tip is probably the best.

Gold paper consists of leaves of gold placed upon thin paper, a sheet of
which, measuring about 19 inches by 12¾ inches, requires one book of gold.
The mat or dead gold is most frequently used in illumination; but, when
required, the bright or burnished gold can be procured. Gold paper is
usually plain at the back, and when used, is required to be gummed on to
the work; but it is far better to have it prepared on the back with a
mixture of clear glue, sugar, &c., which can be laid on evenly and thickly,
and yet is very strong. Paper thus prepared needs only to have a wet flat
camel-hair brush passed over the back; it can then be laid down, and will
adhere very firmly. In laying down gold paper, it is well to place a piece
of white glazed paper on its face, then firmly to pass over it the edge of
a flat rule or burnisher, in order to press down all inequalities and
render the surface perfectly smooth.

Shell gold is gold powder mixed up and placed in mussel-shells for use. It
is removed from the shell by the application of water, like moist colours,
and is adapted for small work and fine lines, in which latter case a quill
or reed pen will be found useful. When the work is dry, the gold can be
brightened with a burnisher. Saucer gold only differs from shell gold in
being placed in china saucers instead of shells.

Gold paint is a preparation of bronze in imitation of gold, and is usually
sold in two bottles, one of powder and the other of liquid; which two
ingredients, when mixed together, form the "paint," the use of which I do
not recommend, as in course of time it turns black. The same objection
unfortunately applies more or less, also, to the preparations of silver,
which, however, are still occasionally used in illumination.

Silver leaf is made in the same manner as gold leaf, and the remarks made
in reference to that are generally applicable to silver leaf.

Shell silver is not really silver, but an amalgam of tin and mercury
prepared and placed in mussel-shells, and used with water in the same way
as gold shells.

Shell aluminium is a preparation of aluminium placed in mussel-shells for
use, and is warranted to keep its colour without tarnishing. _If_ this be
the case, it will form a valuable addition to the list of materials for
illumination, as it will be the only white metal known that can be depended
upon for not tarnishing. The preparation is at present a comparatively new
one, but bids fair to be very serviceable.

Water-mat gold size is a preparation for laying down gold leaf, _i.e._,
causing it to adhere to a given surface. The mode of using it is as
follows:--Take a small brush saturated with water, and thoroughly charge it
with the size. With the brush so charged, trace out the required form or
pattern, and upon this lay the gold leaf, pressing it lightly down with
cotton-wool. When all is dry, gently rub off the superfluous gold with
cotton-wool.

"Burnish gold size" is a preparation for laying down the gold leaf that is
intended afterwards to be burnished (_i.e._, polished with a tooth or agate
burnisher). That prepared by Messrs. Winsor & Newton may be used as
follows:--Place the bottle in warm water to dissolve its contents, which,
however, must not be allowed to get hot, but merely be made liquid. Stir up
the preparation with a hogs-hair brush, which then thoroughly charge with
the mixture; with it trace out the pattern required to be burnished, then
let the work dry. When quite dry, let the surface of the pattern be wetted
with clean cold water, and on it (while damp) place the gold leaf. Let all
get perfectly dry, and then burnish as required. When a very bright surface
is wanted, two coats of the size should be used; the second being put on
after the first is dry.

The "raising preparation" made by the same firm, is adapted for raising the
surface of the work, so as to obtain relief, and is particularly required
for imitating rich MSS. of the 14th and 15th centuries. It is used as
follows:--Place the bottle in hot water, and when its contents are
dissolved, stir it well up with a small hogs-hair brush, then fully charge
it, draw out the form intended to be raised, and deposit the "raising" on
the surface. If the height thus attained be not sufficient, wait till the
preparation is dry, and go over it again, and so on until you gain the
height you require, when it must be allowed to become quite hard; then go
over it with the water-mat gold-size, and while this is wet put on the
gold; press gently down with cotton-wool, and when dry brush off the
superfluous gold with cotton-wool; when putting on the "raising," take care
to keep the surface level, unless it may be required to be hollowed or
indented.

Mr. Barnard has also prepared a gold size and raising preparation, adapted
for laying gold on vellum or paper, which answers well both for mat and
burnish gilding. The mode of using it is as follows:--Wash a little of the
gold size off with a brush dipped in water, using it thinly for the flat
parts of your design, and in greater body for that portion of the drawing
which you wish to appear raised; after allowing it to remain for a few
minutes, till nearly dry, apply the gold, and press it down with a piece of
cotton-wool. It must now remain untouched for about an hour, when the
superfluous gold may be removed by means of the wool, and in case of
defect, the gold size and gold must be again applied. Preparations of a
somewhat similar nature are sold by Messrs. Rowney, Newman, and other
artists' colourmen.[119]

Very pretty effects may be obtained by partial burnishing of the gold in
patterns, and dotting it over with the point of the sharp burnisher in
indentations, arranged in geometrical forms. The best manuscripts of the
Edwardian period were often highly wrought after this fashion.

When finished, it is scarcely necessary to recommend that the vellum sheet
should be either put carefully away until enough of others corresponding
with it are done to make up a volume, or should be glazed so as to protect
its surface. One dirty or greasy finger laid upon it, and the effect of
much beautiful work, which may have taken weeks to elaborate, is fatally
marred.

All the above instructions apply as well for working on paper or cardboard
as on vellum. The amateur who has once succeeded on vellum, is not likely
to take again to the humbler practice of working on the less noble
materials, which, however, will always be exceedingly useful for practising
and sketching upon. I have occasionally seen printed volumes gracefully
illustrated by hand with borders, and with elegant inventions, in the form
of head and tail pieces, insertions, &c., applicable to the subject of the
volume. Many of the works of old English authors are peculiarly suited for
this class of embellishment. How beautiful might not a Walton's "Angler" or
a Bunyan's "Pilgrim's Progress" be made if appropriately enriched in this
style.

Tracing-paper, and the facilities it offers to those little gifted with
talents for drawing, I have already noticed. It remains, however, to
observe, that it possesses an additional practical convenience in being
ready for taking colour, either with oil, water, or varnish, as vehicles,
without the previous application of any special preparation. Hence it may
be fastened up when completed, either by pasting as ordinary paper, by
gluing, if for attachment to wood, or by paying over the back with boiled
oil and copal varnish, or with white lead ground in oil with some litharge,
and then pressing down until it may be made to lie perfectly flat and
adhere to any surface previously painted in oil-colour. Being very thin,
its edges will scarcely show at all, even if applied to the middle of a
flat panel; but, to make sure, it is always well to run a line with a full
brush of thick colour, either in oil or distemper, over the edge, extending
for one half of its width upon the tracing-paper, and for the other half
upon the surface to which it may have been applied.

Of the remaining materials on which illumination for the decoration, not of
books but of apartments, may be readily executed, canvas, stone, metal, and
wood, are generally wrought upon by the ordinary processes of oil-painting;
while plaster, especially in the form of ceilings, is more frequently
treated by means of distemper-painting. I propose, therefore, to give,
firstly, some general directions as to the setting out work, &c.,
applicable to both methods; secondly, a notice of the processes generally
required for oil-colour illumination; thirdly, a brief description of the
mode of working in distemper; and fourthly, to wind up with some
instructions as to the application of varnish which may be employed to
heighten and preserve illumination executed by either of the above methods.

The operation of setting out lines upon walls or other surfaces is by no
means easy. It involves care and judgment, a quick eye, and a very steady
hand. It is the indispensable preliminary before ornamental work or
illumination can be executed, as it can alone correctly give the forms of
panels, borders, &c., for which cartoons may have to be prepared. Lines may
be either drawn with pencil or prepared charcoal, or chalk, or else struck
by means of a chalked string. For lines which are vertical, a weight called
a plumb-bob must be attached to one end of the string. The best shape for
this is that of half an egg, as the flat side will then lie close to the
wall. Two persons are required in setting out these lines,--one working
above and the other below. The one at the top marks the points at the
distance each line is required to be from others. The string being chalked
either black or white,--according as the line has to show upon a light or
dark ground,--he holds it to one of the points, and lets fall the weighted
end, which, when quite steady, the person who is below strains tight, and
raising the string between his finger and thumb in the middle, lets it fall
back sharply on the wall. The result, if carefully executed, is a perfectly
straight and vertical line. The horizontal lines require to be drawn with a
straight-edge or ruler, and may be either set out at a true right angle to
the vertical lines geometrically by the intersection of arcs of circles, or
by a large square, or may be defined, irrespectively of mathematical
correctness, by measuring up or down from a ceiling or floor line. The
distances apart are as before measured out, but in long lines must be
marked as many times as the length of the straight-edge may require. This
being set at each end to the points marked, the line is drawn along it.
Circles and curved lines may be struck from their proper centres with large
wooden compasses, one leg carrying a pencil. Drawing lines with the brush
requires great practice. A straight-edge is placed upon the chalk lines,
with the edge next the line slightly raised, and the brush, well filled
with colour, drawn along it, just touching the wall, the pressure being
never increased, and the brush refilled whenever it is near failing; but
great care must be taken that it be not too full, as in that case it will
be apt to blotch the line, or drop the colour upon the lower portions of
the wall. Drawing lines in colour overhead upon a ceiling is even more
difficult, and is beyond the capabilities of most amateurs.

The patterns of ornament are executed either by means of stencils cut in
oiled paper, according to the method which will be next described, or else
by pounces, which are the full-sized drawings pricked along all the lines
with a needle upon a flat cushion; powdered charcoal, tied up in a cotton
bag, is then dabbed upon the paper which has been set up on the wall, or
else the back is rubbed over with drawing-charcoal and brushed well with a
flat brush, like a stove brush. In both cases the result is that the dust
passes on to the walls through the pricked holes, and forms are thus
sufficiently indicated to the painter.

Stencilling is a process by which colour is applied through interstices cut
in a prepared paper, by dabbing with a brush. The design to be stencilled
is drawn upon paper which has been soaked with linseed oil and well dried.
The pattern is then cut out with a sharp knife upon a sheet of glass, care
being taken to leave such connections as will keep the stencil together.
The next tint is then to be laid on in the same manner, and so on till the
darkest tint is done, each tint being allowed to dry before a second is
applied.

I do not purpose dwelling in detail on the preparation, or "bringing
forward," as it is called, of surfaces to receive oil-colour; since, for
such mechanical work, it will be always well to employ a good
house-painter. I may observe, however, that the first operation, where the
surface is absorbent, is to stop the suction, either by a plentiful
application of boiled oil alone, boiled oil and red lead, or size. Several
successive coats of paint should then be applied, and in order to obtain
smoothness, the surface of each should be well rubbed down. The last coat
should be mixed with turpentine, and no oil, in order to kill the gloss,
or, as it is termed, to "flat" the surface. For most decoration and
illumination, the work should be brought forward in white, as, by shining
partially through most of the pigments ultimately applied, it will greatly
add to their brilliancy. Zinc white will stand much better than white lead.
Messrs. Roberson, of Long Acre, prepare an excellent wax medium, which
dries with a perfectly dead encaustic surface, and answers admirably for
mural-painting of all kinds. I caused it to be employed for all the
decoration executed under my direction at the Sydenham Crystal Palace.
Miller's glass medium will also be found very useful to artists and
amateurs. In laying on all ground tints, great care should be taken to keep
them flat; and the less, as a general rule, tints are mixed, worked over
and over, and messed about, the brighter they will be. The principal
colours having dried, the setting out of the lettering, &c., may be
proceeded with; the following directions being duly attended to.


_The Setting-out of Letters._[120]

In regard to the proportion of Roman capital letters, it may be taken as a
general rule, that the whole of the letters, with the exception of S, J, I,
F, M, and N, are formed in squares. The top and bottom of the letters
should project the width of the thick line. The letters I and J are formed
in a vertical parallelogram, half the width of the square; the letters M
and N in a horizontal parallelogram, one third larger than the square. The
letters A, B, E, F, H, X, and Y, are either divided, or have projections
from the middle. This rule may be varied, and the division placed nearer
the top than the base of the square. Capitals in the same word should have
a space equal to half a square between them; at the beginning of a word, a
whole square, and between the divisions of a sentence two squares should be
left.

This is the general rule for the proportions of the letters; but they may
be made longer or wider, as may be deemed expedient.

The small letters are half the size of the capitals; the long lines of the
letters b, d, f, h, k, and l, are the same height as the capitals; the
tails of j, p, q, and y, descending in like proportion. The letter s is
founded on the form of two circles at a tangent to each other. These rules
are applicable to sloping as well as to upright letters. In _italic_
letters it is usual to make the capitals three times the height of the
smaller letters, and the long strokes of the small letters nearly equal to
the capitals.

The letters having been duly set out, and painted on the walls, the amateur
must next either himself encounter, or employ some experienced hand to
overcome, the technical difficulties of successfully gilding those portions
of his work he may desire to remain in gold. The following directions may
assist him; but he is not likely to succeed until practice shall have given
him considerable dexterity and confidence:--


_Gilding for Walls, &c._

The implements with which the gilder should provide himself are not
numerous, nor are they expensive, as they consist merely of a cushion of
particular form, a knife for cutting the gold-leaf, a tip for transferring
it, and a cotton ball or pad for pressing it down; these and a few brushes
are all the requisites, with the addition of an agate burnisher when
burnish gilding is desired.

The cushion is a species of palette made of wood, about 9 inches by 6
inches, having on the upper surface a covering of leather stuffed with
wool, and on the under side a loose band, through which the thumb being
passed, the cushion is kept firmly resting on the left hand. To prevent the
gold flying off (for, being extremely light, this very readily takes
place), a margin of parchment is fixed on the edge of the cushion, rising
about three inches, and enclosing it on three sides. The knife very much
resembles a palette-knife, the blade is about four inches long and half an
inch wide, perfectly straight, and cutting on one edge only.

The "tip" is the brush with which the gold-leaf is applied. It is formed by
placing a line of badger-hair between two thin pieces of cardboard, and is
generally about three inches wide. The "dabber" is merely a pinch of
cotton-wool, lightly tied up in a piece of very soft rag, or, what is
better, the thin silk called Persian. It is often used without covering,
but is then very apt to take up the uncovered gold-size, and so to soil the
leaf already laid down. Camel-hair brushes are useful for intricate parts,
and for cleaning off the superfluous gold a long-haired brush, called a
"softener," is requisite.

There should be also at hand a small stone and muller (these are also made
in glass, which is cleaner) for grinding up the oil and gold-size.

The operator, having stocked himself with the above tools, may now proceed
to lay the gold-leaf upon the work he desires to gild. There are two
methods of doing this, known in the trade as "Oil-gilding" and
"Water-gilding;" and so called from the composition of the size which
serves as a vehicle for making the gold-leaf adhere to the work.

  The following is the usual process in oil-gilding:--This method costs
  less and will wear much better than water-gilding, which will be
  presently described; but has not its delicate appearance and finish, nor
  can it be burnished or brightened up. Though the oil gold-size can always
  be purchased of good quality, it may be well to describe the fat oil of
  which it is principally composed.

  Linseed oil, in any quantity, is exposed during the summer in the open
  air, but as much away from dust as possible, for about two months, during
  which time it must be often stirred, and it will become as thick as
  treacle. It is a good practice to pour into the pot a quantity of water,
  so that the oil may be lifted from the bottom of it, as all the
  impurities of the oil sink into the water, and do not again mix when it
  is stirred. When of the consistency above mentioned, the oil is separated
  from the water, and being put into a bottle, is subjected to heat till it
  becomes fluid again, when all remaining impurities will sink, and the
  oil, being carefully poured off from the sediment, forms what is termed
  "fat oil." The gilder commences by priming the work, should it not have
  been painted, using for the purpose a small portion of yellow ochre and
  vermilion, mixed with drying oil. When this is quite dry, a coat of the
  oil gold size, compounded with the fat oil just described, japanner's
  gold-size, and yellow ochre, is laid on, and when this is perfectly dry,
  a second should be given, or even a third. A superior finish is produced
  by going over the work, before using the size, with Dutch rushes or
  fish-skin, which gives a finer surface to it. After the last coat of size
  is applied, the work must be left for about a day, to set, taking care to
  keep it from dust; and the proper state for receiving the gold-leaf is
  known by touching the size with the finger, when it should be just
  "tacky," that is adhesive, without leaving the ground on which it has
  been laid.

  The gilder then, taking on his left hand his cushion, transfers to it the
  gold-leaves from the books in which they are purchased. This is not very
  easy to a beginner, as the gold cannot be touched except by the knife.
  Gilders manage it by breathing under the leaf in the direction it is
  desired to send it, and flatten it on the cushion by the same gentle
  blowing or breathing. It is now cut to the required shape, and applied to
  the sized surfaces by means of the tip, which, if drawn across the hair
  or face each time it is used, will slightly adhere to the gold. The whole
  leaves are sometimes transferred from the books to the work at once; and
  when there is much flat space, it facilitates the process. As the leaves
  are laid on the size, they are pressed gently down with the cotton ball,
  or in sunken parts with camel-hair brushes; and when perfectly dry, the
  loose leaf is removed by gently brushing over the work with the softener,
  when if there should be found any places ungilt, such spots are touched
  with japanners' gold-size, and the leaf applied as before. The process of
  oil-gilding is then complete.

  Water or burnish-gilding differs from the former in the use of parchment
  instead of oil size, and has received its name from being moistened with
  water in rendering the size adhesive, and also from its fitness for
  burnishing. Its superior beauty, however, is balanced by its being less
  durable than oil-gilding, and, unlike the latter, unfit to be exposed to
  damp air; it is therefore only used for indoor work or ornamentation. The
  parchment size is made by boiling down slips of parchment or cuttings of
  glovers' leather, till a strong jelly be formed, the proportions being
  one pound of cuttings to six quarts of water, which must be boiled till
  it shrinks to two quarts. While hot, the liquid should be strained
  through flannel; and when cold, the jelly required will be fit for use.

  The work to be gilded will require several coats of composition: the
  first, or priming coat, is made of size thinned with water, and a little
  whiting; with this the work is brushed over, using a thicker mixture when
  there are defects which need to be stopped. Successive coats are then
  laid on to the number of seven or eight, and the last, being moistened
  with water, is worked over and smoothed on the plain parts with Dutch
  rushes. After this is completed, a coating is laid on, composed of bol
  ammoniac 1 pound, black lead 2 ounces, ground up on the stone with 2
  ounces of olive oil. This is one out of many receipts; all, however, are
  diluted for use with parchment size warmed up with two-thirds water, and
  forming what is called water gold-size. Two coats of this should be laid
  on; the part about to be burnished should then be again rubbed with a
  soft cloth till quite even, and care taken that each coat be perfectly
  dry before the subsequent one be laid on. The work is now moistened in
  successive portions with a camel-hair brush and water, and while moist
  covered with gold-leaf in precisely the same manner as described in the
  directions for oil-gilding, great caution being observed in order to
  avoid wetting the leaf already laid down, as a discoloration would be the
  result. The work is now left for about four-and-twenty hours, when the
  parts which are to be burnished may be tried in two or three places. Care
  should be taken not to let the work get too dry, as in that case it would
  require more burnishing, and yet not give a good result. This state is
  known by its polishing slowly, and if it be too wet it will peel off; but
  should the places where the trials are made all polish quickly and
  evenly, the work may then be finished; for which purpose agates cut in
  proper forms and set into handles, are sold at the artists'
  colour-shops.[121]

The gilding satisfactorily accomplished, the artist or amateur has only to
add the finishing tints and touches to his work, and then either to leave
it alone, or to varnish it in accordance with the directions which will be
given presently. If the work has been executed on canvass, it will remain
only to apply it to the surface for which it may have been destined. This
may be done by painting that surface with thick white lead, in two or three
coats, and by also similarly painting the back of the canvass. The latter
being then pressed evenly down upon the former, while the white lead upon
both is still tacky, and, left for a few days, will be found to have
attached itself with the greatest tenacity. Scrolls and panels cut out of
zinc sheets may be painted upon just as though they were cut out of
canvass, and may be fixed in their places by nails or screws. In
illuminating on wood, pretty effects may be obtained by varnishing
partially with transparent colours, such as the lakes, umber, Prussian
blue, burnt sienna, &c., so as to allow the grain of the wood to show
through,--restricting the use of opaque colour and gilding to a few
brilliant points.

Distempering is a method of colouring walls and ceilings in which powder
colour, ground up in water, and mixed with sufficient size to fix the
colour, is used instead of paint made with oil. The most simple employment
of distemper is in whitening ceilings, but it is also very much used in
theatrical decoration and scene-painting; and rooms are sometimes so
ornamented, the process being much less expensive than oil-painting. The
foundation of all the colours is whiting, which, having been set to soak in
water and break up of itself, is (when the top water is poured off) in a
fit state for use; common double size is then added, with as much of the
colour as will make the desired tint; but as this, when dry, will be many
shades lighter than it appears when wet, trials should be made on paper,
and dried by the fire till the colour required be attained. A gentle heat
is required for melting the size. Old walls are prepared for distemper by
being scraped and cleaned, and a coat of "clearcole" given to them. This is
merely thin size and water with a little whiting: it serves to wash and
smooth the walls and stop suction. Should there be any cracks or holes, a
thick paste of size-water and whiting is laid in them with a palette-knife,
and, when dry, smoothed down with pumice-stone, and another coat of
clearcole given, when the wall is in a proper state to receive the ground
tint; for new walls one coat of clearcole is sufficient. If it is intended
to lay on lines of various colours, the wall is, previous to the laying on
of the ground tint, set out as previously described; and the appropriate
colours put on in succession, according to the design to be followed. All
the colours required should be ground up, and kept ready prepared in
galley-pots well covered over, so as to be at hand at once. The colour
should be of the consistency of thick cream, and should run from the brush
on being raised from the pot in one thread; if it run in several, it is too
thin. If too thick, add more size and water; if too thin, more whiting. The
pots used are the common red paint-pots.


VARNISHING.

  Varnish is a solution of resin in oil or spirits of wine.[122]

  Surfaces which are to be varnished should be of the greatest smoothness
  and polish which it is possible to attain. Dark colours are best
  calculated for varnishing; the lighter colours, such as sky-blue,
  apple-green, rose-colour, delicate yellow, &c., will not bear varnishing
  so well, and in spite of the greatest care are liable to get dirty.

  The best preparation for stopping suction in absorbent surfaces, and so
  rendering them fit to take varnish, is made of isinglass or parchment
  size; for the darker colours it may be made of common clear glue. Four or
  five coats will be necessary for the brighter colours; two or three will
  be sufficient for the darker ones. Great care must be taken not to wash
  up water or distemper colours in laying on the first coat, nor to lay on
  a second coat before the first is perfectly dry; nor must the varnishing
  be proceeded with before the last coat of size is thoroughly dry. Varnish
  may be applied on surfaces brought forward in oil without any special
  preparation, provided the oil has become thoroughly dry and hard.

  This process serves both to enhance and preserve the beauty of the
  colours, and in some degree to counteract the destructive influence of
  the atmosphere and of insects.

  Varnishes suitable for the work in hand, such as clear copal spirit
  varnish, oil copal varnish, white hard varnish, &c., may be procured from
  any one who supplies drawing materials. The varnishing itself requires
  some little care. It should be performed in a place perfectly free from
  dust, in a bold manner with large brushes, steadily, rapidly, and
  uniformly, not returning too frequently to the same spot, more especially
  when using spirit varnish, which loses its fluidity much sooner than oil
  varnish. Whichever varnish is used, it should be very thin: if spirit
  varnish, the room must be of a moderate temperature; for if too cold, the
  varnishing is apt to be rough, white, and unequal; if too hot, it is
  liable to have air-bladders, and to crumble and spoil. Oil varnishing may
  be done in a room of warmer temperature. A second coat of varnish must on
  no account be laid on before the first coat is quite dry. If the work is
  to be polished, the spirit varnish must be applied from five to eight
  times, oil varnish three or four; but if the work is not to be polished,
  then four coats of the former and two of the latter will generally be
  found sufficient.

  When thoroughly dry, the face of the varnish may be polished with
  pumice-stone, tripoly, water, and sweet oil. If it be an oil varnish,
  procure some of the finest pulverized pumice-stone, and mix it with water
  to about the consistence of cream; with a piece of linen rag dipped in
  this mixture rub the work till all inequalities disappear, and the
  surface is as smooth as glass; then dry it with a cloth, and polish once
  more with tripoly and sweet oil; then dry it with a piece of soft linen,
  rub it with starch reduced to a fine powder, and finish with a clean soft
  linen cloth, until the varnish assumes a dazzling appearance. If it is a
  spirit varnish, omit the pumice-stone, and begin with the tripoly and
  water; after this use the tripoly and sweet oil, and finish as before
  described for the oil varnish.

  The difference is so striking between the polished and unpolished
  surfaces, as to amply repay the additional trouble required in the
  polishing. The polishing powders must be kept in thoroughly clean
  vessels, a single grain of sand being sufficient to spoil the polish.

  M. DIGBY WYATT.

  37, TAVISTOCK PLACE, W.C.
          _April, 1861._



FOOTNOTES.

  [1] M. Gabriel Peignot, in his "Essai sur l'Histoire da Parchemin et du
      Vélin," Paris, 1812, and in his paper on the same subject in "Le
      Moyen Age et la Renaissance," vol. ii. Paris, 1849, produces evidence
      of the use of parchment for writing upon anterior to the age of
      Eumenes; and consequently limits his interpretation of Pliny's words,
      "Varro membranas Pergami tradidit repertas," to an assertion of the
      discovery of improved processes by which parchment was rendered more
      available for writing upon than it had been previous to the accession
      of Eumenes.

  [2] A good representation of a scrinium and scapi, from a painting in the
      "Casa Falkener," described in the "Museum of Classical Antiquities,"
      vol. ii. p. 54, is given in one of the cubicula of the Pompeian Court
      at Sydenham.

  [3] See Gell's "Pompeiana" Appendix; and the "Memoir of the Canonico
      Iorio."

  [4] "Let those who will have old books written in gold and silver on
      purple parchment, or, as they are commonly called, in uncial
      letters,--rather ponderous loads than books,--so long as they permit
      me and mine to have poor copies, and rather correct than beautiful
      books."

  [5] P. 113. Ingram, Cooke, & Co., London, 1853.

  [6] "Universal Palæography." London, Bohn.

  [7] Through the kindness of the late Mr. Dennistoun, of Dennistoun, and
      Cardinal Acton, who obtained the requisite facilities for me.

  [8] Tome v. pl. lxv.; tome iii. p. 29.

  [9] D'Agincourt's famous mistake in attributing these miniatures to the
      12th or 13th century, and Ottley's ascription of those in the Saxon
      "Aratus" of the 9th century to the 2nd or 3rd, are among those slips
      of the learned which prove that even great men are fallible.

 [10] "Iste liber est beati Dionysii."

 [11] The palimpsest Homer of the British Museum, discovered by Mr.
      Cureton, is of equal importance in Grecian palæography.

 [12] In the case of the "de Republicâ," they are written in the same
      direction. See facsimiles in Sylvestre and Ferdinand Seré.

 [13] "Iliadis Fragmenta antiquissima cum Picturis," ed. Angelo Maio.

 [14] Petri Lambecii "Commentaria de Bibliotheca Vindobonensi," vol. ii.

 [15] The Bible formerly belonging to Theodore Beza, now at Cambridge, and
      one in the Vatican, are rival claimants to this honour.

 [16] It was given to Charles I. of England, by Cyrillus Lucaris, Patriarch
      of Constantinople.

 [17] In his "Origin and Progress of Writing."

 [18] "Treasures of Art in Great Britain," vol. i. p. 97.

 [19] Text to "Shaw's Illuminated Ornaments," page 4.

 [20] For a full description, with references to numerous commentators, see
      Westwood's "Palæographia Sacra Pictoria," cap. 49.

 [21] Tome ii. article "Manuscrits," fig. 15.

 [22] "Treasures of Art in Great Britain," vol. i. p. 96.

 [23] Dr. Kugler ("Handbuch der Kunstgeschichte," p.401), in speaking of
      Anglo-Saxon illuminated manuscripts, observes, "dass wir diese
      Arbeiten als ein der ersten Zeugnisse des germanischen Kunstgeistes
      in seiner Selbständigkeit, und zugleich als das Vorspiel oder als den
      ersten Beginn des romanischen Kunststyles, zu betrachten haben."

 [24] As represented in the plates to Salzenberg's fine work,
      "Alt-Christliche Baudenkmale von Constantinopel, vom V. bis XII.
      Jahrhundert." Folio, Berlin, 1854.

 [25] The whole are given in Shaw's "Illuminated Ornaments," plates 1, 2,
      3, and 4.

 [26] It is on this account that we have refrained from giving any
      specimens of manuscripts anterior to the 6th century.

 [27] "Palæographia Sacra Pictoria," cap. Syriac MSS.

 [28] O'Conor and others were of course earlier in the field.

 [29] "Palæographia Sacra Pictoria," Book of Kells, page 1.

 [30] Catalogue of the Libri collection of MSS., Introduction by M. Libri,
      pages xiv. and xxvi. London, 1859.

 [31] Giraldus Cambrensis, speaking probably of this very book, says, "Sin
      autem ad perspicacius intuentum oculorum aciem invitaveris, et longe
      penitus ad artis arcana transpenetraveris, tam delicatas et subtiles,
      tam actas et arctas, tam nodosas et vinculatim colligatas, tamque
      recentibus adhuc coloribus illustratas, notare poteris intricaturas,
      ut vere hæc omnia angelica potius quam humana diligentia jam
      asseveraveris esse composita."

 [32] It is more abundantly used in Vesp. A 1, which, as we shall have
      occasion to notice hereafter, is in a very mixed style.

 [33] Bede expressly says, that at Augustine's synod, held at the
      commencement of the 7th century, the bishops and learned men
      attending it, "after a long disputation, refused to comply with the
      entreaties, exhortations, or rebukes of the saint and his companions,
      but preferred their own traditions before all the churches in the
      world, which in Christ agree among themselves."

 [34] D'Agincourt, "Painting," plates xxviii. xxix. xxx.

 [35] This precious volume and its illustrations were first figured and
      described by Mr. Westwood.

 [36] "Life of Gregory the Great," by Johannes Diaconus, lib. ii. cap. 37.

 [37] The words are, "quæ omnia illustrantur Romano habitu, figuris, et
      antiquitate. Imperatoris Valentiniani tempora videntur attingere."
      This mistake of the old librarian has been corrected with much care
      and learning by the Baron van Tiellandt.--See his "Naspeuringen
      nopens zekeren Codex Psalmorum in de Utrechtsche Boekerij berustende,
      door W. H. J. Baron van Westreeinen van Tiellandt."

 [38] The MS. department of the British Museum possesses some tracings from
      the Utrecht Psalter, and on confronting them with the Harleian 603,
      it requires a sharp eye to detect the slight differences existing
      between several of the illustrations to each of the volumes. In the
      Harleian volume, all the subjects have not been filled in; some are
      left out altogether, spaces being reserved for them in the text, and
      others are faintly traced with a leaden or silver point, preparatory
      to inking in: very few artists of the present day could block in the
      general forms in so peculiar a style with greater freedom or more
      complete conveyance of expression, by similarly slight indications.

 [39] The whole of the illuminations are given in the twenty-fourth volume
      of the "Archæologia." The manuscript stands in the Bodleian
      Catalogue, "Junius, No. II."

 [40] Introduction to Shaw's "Illuminated Ornaments," pages 4 and 5.

 [41] The following inscription, written in letters of gold on the reverse
      of the fourth leaf and the bottom of the recto of the fifth,
      identifies both the artist and the patron under whose auspices the
      volume was executed, between the years 970 and 984, the term of
      Ethelwold's occupation of the see of Winchester:--

       "Presentem Biblum jussit perscribere Presul
        Wintoniæ Dñs que[m] fecerat esse Patronum
        Magnus _Æthelwoldus_       *       *       *
               *       *       *       *       *       *
        Atque Patri magno jussit qui scribere librum hunc
        Omnes cernentes biblum hunc semper rogitent hoc
        Post meta carnis valeam celis in herere
        Obnixe hoc rogitat Scriptor supplex _Godemann_."

 [42] If the celebrated coronation book of the Anglo-Saxon kings should
      turn out to have been written and illuminated in this country, it
      would afford a striking illustration of this reaction. The general
      opinion, however, appears to be, among the learned, that it may have
      been given to Athelstan by Otho of Germany, who married his sister,
      and by Matilda, Otho's mother. The arguments in favour of, and
      against, the Anglo-Saxon origin of the volume would be too long to
      discuss in this place. The writing is mainly Carlovingian.

 [43] "Bib. Dec." vol. i. p. cxxii.

 [44] It is to be regretted that the propriety of those just and learned
      remarks of Muratori, in which he exhibited himself as one of the
      earliest foreign scholars inclined to do justice to the ancient Irish
      and British schools,--"Neque enim silenda laus Britanniæ, Scotiæ, et
      Hiberniæ, quæ studio liberalium artium eo tempore antecellebant
      reliquis occidentalibus regnis; et cura præsertim monachorum, qui
      literarum gloriam, alibi aut languentem aut depressam, in iis
      regionibus impigrè suscitarent atque tuebantur" (Murat. "Antiq.
      Ital." diss. 43),--should have been impugned by the Rev. Mr.
      Berington in his "Literary History of the Middle Ages," pages 180,
      181.

 [45] These pious monks, until probably some time after the Norman
      conquest, generally worked together in an apartment capable of
      containing many persons, and in which many persons did, in fact, work
      together at the transcription of books. The first of these points is
      implied in a curious document, which is one of the very few specimens
      extant of French Visi-Gothic MS. in uncial characters, of the 8th
      century. It is a short but beautiful form of consecration or
      benediction, barbarously entitled "Orationem in Scripturio," and is
      to the following effect: "Vouchsafe, O Lord, to bless _this
      Scriptorium of thy servants and all that dwell therein_; that
      whatsoever sacred writing shall be here read or written by them, they
      may receive with understanding, and bring the same to good effect,
      through our Lord," &c.--See Merryweather's "Bibliomania in the Middle
      Ages."

 [46] "Dark Ages," second edition, p. 193.

 [47] Librarian of the town of Evreux.

 [48] Cornemillot, Evreux, 1846.

 [49] Du Sommerard, in "Les Arts du Moyen Age," has given copies of all the
      illuminations, and Mr. Westwood a page of specimens.

 [50] Count Bastard gives no less than six grand facsimiles from this
      volume, which is one of the greatest lions of the Bibliothèque
      Impériale at Paris.

 [51] One of the most curious illuminations in the book, the celebrated
      "fontaine mystique" of the church, is altogether antique in style and
      execution.

 [52] The colouring in this MS. is very elegant, being mainly restricted to
      gold, purple, white, and a little very brilliant vermilion;--the
      forms are principally Saxon.

 [53] Described at length by Dr. Waagen, "Treasures of Art in Great
      Britain," pages 104-106.

 [54] Many illustrations, but unfortunately without colour, are given by
      D'Agincourt, "Pittura," plates 40 to 45 inclusive.

 [55] The folio Vulgate (B. M. Addl. MSS. No. 10546) purchased by the
      British Museum authorities from M. Speyer Passavant, of Basle, in
      1836, for £750, was considered by its late possessor to have been the
      original transcript "diligently emended" by Alcuin himself, for
      presentation to Charlemagne on his coronation as Emperor of Rome, in
      the year 800. It is a very fine and interesting volume, but has been
      referred, by more recent authorities, to the reign of Charles the
      Bald. Mr. Westwood, however, considers that "it appears to have
      better claims than any of the several Caroline Bibles now in
      existence, to be considered as the volume so presented." Its chief
      rival is the great Bible of the Fathers of Sta. Maria, in Vallicella,
      at Rome. Sir Frederick Madden has entered into a minute analysis of
      the claims of the Speyer Passavant volume, in a series of most
      learned articles in the "Gentleman's Magazine" for 1836. See also
      Westwood's "Palæographia Sacra," and the pamphlet, by its late
      possessor, J. H. de Speyer Passavant, "Description de la Bible écrite
      par Alchuine, &c." Par. 1829, pp. 112.

 [56] It is singular, considering how generally Hiberno-Saxon ornament was
      adopted by continental illuminators, that the peculiar Saxon
      _fluttering_ outline never obtained a footing.

 [57] The learned and most eloquent author of the "Poésie Chrétienne," M.
      Rio (from whom it was my privilege, while yet a youthful student, to
      receive many a valuable lesson), in noting this "total eclipse,"
      remarks that "two rolls of parchment, one of which is preserved in
      the library of the Barbarini Palace, the other in the sacristy of the
      Cathedral of Pisa, are ornamented with miniatures which may serve to
      give us an idea of the state into which the arts of design had fallen
      in Italy in the 11th century. Those which were executed rather later,
      in the manuscript of a poem on the Countess Matilda (written by a
      certain 'Donizo,' in 1125), which is preserved in the Vatican,
      display no trace either of chiaroscuro or of correct imitation of
      form.

      "The Romano-Christian school ceased from this time to exist, after
      having fulfilled the whole of its mission, which had been to form the
      connecting link between the primitive inspirations of Christian art
      and the new schools which were destined to reap the harvest of this
      rich inheritance, and turn it to good account.

      "As for the Germano-Christian school, it may be compared to a
      vigorous shoot severed from a dying trunk, to revive and flourish in
      a better soil."

 [58] The "Menologion" of the Vatican, a magnificent volume, containing no
      less than 430 miniatures of remarkable interest and excellence, is
      the standing illustration of this assertion. The work was engraved
      and published at Urbino, in three folio volumes, in 1727, under the
      auspices of three pontiffs, Clement XI., Innocent XIII., and Benedict
      XIII.

 [59] It would be difficult to find in the production of the best Roman age
      anything nobler than several of the compositions in the Paris
      "Psalter," with commentaries (Imperial Library, Gr. No. 139), a Greek
      manuscript of the 10th century. One of the finest of the figures
      contained in it, that of "Night," I caused to be enlarged, and
      painted on the exterior of the Byzantine Court at Sydenham, as giving
      a more favourable impression of Greek art than any other pictorial
      representation I could meet with. A replica of this subject occurs in
      the Vatican "Prophecies of Isaiah." The two may be compared from the
      works of D'Agincourt and Seré. Most noteworthy also among the best of
      this class of Byzantine manuscripts, are the Paris "Commentaries of
      Gregory Nazianzen," the British Museum Psalter (Egerton, No. 1.139)
      of early 12th century work, and the Bodleian "Codex Ebnerianus."

 [60] Of this ornamental style the most remarkable specimens are the
      Vatican "Acts of the Apostles," and a beautiful volume in the library
      of the Duke of Hamilton. From the former, I have given some
      facsimiles in "The Geometrical Mosaics of the Middle Ages" (plate
      20), in order to show the similarity of design between the gold
      ground mosaics of the Greeks and early Italians, and the
      embellishments of the illuminated manuscripts of the former.

 [61] Ingulphus was at that very time indebted directly to the Conqueror,
      his early patron, for his abbacy.

 [62] See Martene Const. Canon. Reg. in "de Ant. Eccl. Ritibus," tom. iii.,
      for full details.

 [63] This indulgence was, after all, not very luxurious, for, as Mr.
      Maitland remarks ("Dark Ages," 2nd edition, p. 406), "Many a scribe
      has, I dare say, felt what Lewis, a monk of Wessobrun, in Bavaria,
      records as his own experience during his sedentary and protracted
      labours. In an inscription appended to a copy of Jerome's Commentary
      on Daniel, among other grounds on which he claims the sympathy and
      the prayers of the readers, he says,--

       "'Dum scripsit friguit, et quod cum lumine solis
        Scribere non potuit, perfecit lumine noctis.'"

      For whilst he wrote he froze, and that which by daylight he could not
      bring to perfection, he worked at again by the aid of the moonlight.

 [64] In Italy the propensity for large letters was never relinquished.

 [65] W. H. Blaauw, Esq.

 [66] Edited by James Raine, Jun., for the Surtees Society. 8vo. Durham,
      1859.

 [67] The same series of rolls contain many very interesting entries; as,
      for instance,--

      "1393 A.D. Soluti--de 4_l._ 6_s._ 8_d._ sol. hoc anno fratri Willelmo
      Ellerker pro scriptura duorum gradalium pro choro. de 40_s._ solutis
      domino Ricardo de Styrton pro eluminacione dictorum duorum
      gradalium--de 22_s._ 7½_d._ solutis dicto Willelmo pro pergameno
      empto per ipsum Willelmum.

      "A.D. 1395. Roberto Bukebinder pro ligatura unius magni gradalis pro
      choro ex convencione facta 10_s._ Eidem pro IIII. pellibus pergameni
      pro eadem custodiendo 20_d._ Eidem pro I. pelle cervi pro coopertura
      dicti libri 3_s._ 2_d._ Fratri Willelmo Ellerker pro pergameno 4_s._
      Domino Ricardo de Styrton in plenam solucionem _alumpnyng_ tryum
      gradalium, 40_s._ de 3_s._ 4_d._ solutis domino Johanni Brignale pro
      VIII. pellibus pergameni emptis pro magno gradali predicto."

      "Domino Ricardo de Styrton pro alumpnacione magni gradalis novi in
      choro, 20_s._

      "A.D. 1402. In expensis in _alumpnacione_ magni gradalis in choro per
      dominum Ricardum de Stretton, 20_s._"

      Throughout these accounts, and others too lengthy to note, it will be
      noticed that the value of the parchment, gold, colours, and current
      expenses, falls not very far short of the total cost of the labour of
      the illuminator.

 [68] "Treasures of Art in Great Britain," vol. i. p. 160. The same
      distinguished critic, who has made a special study of the illuminated
      MSS. of Europe, and especially of the French (see his "Kunstwerken
      und Kunstlern in Paris"), in describing some of the pictures in Queen
      Mary's Psalter (unquestionably English), observes (p. 166), "Upon the
      whole, I am acquainted with no miniatures, either Netherlandish,
      German, or French, of this time" (the 14th century) "which can
      compare in artistic value with the pictures executed by the best hand
      in this manuscript."

 [69] It is to be regretted that Count Bastard failed to complete more than
      thirty-two plates of the splendid work he announced under the title
      of "Librairie de Jean de France, Duc de Berri, frère de Charles V.,
      publié en son entier pour la première fois." Paris, 1834. Fol. max.
      &c.

 [70] "De l'Art en Allemagne," tome ii. page 153. Paris, 1842.

 [71] See casts from his bronze doors and columns in the Crystal Palace,
      and his Three Gospels in the treasury of the Cathedral at Hildesheim.
      In Dr. F. H. Müller's "Beiträge zur teutschen Kunst und
      Geschichtskunde," very careful engravings of the plastic art of
      Bernward and Willigis may be compared with facsimiles of contemporary
      German illumination.

 [72] The steps of the transition are also well indicated, and illustrated
      by reference to special MSS. in Kugler's "Kunstgeschichte," in his
      article on the "Nord., vornehml. Deutsche Malerei der Roman.
      Periode."

 [73] The subject is one that I am unable to find has been treated with any
      great ability. The reader may, however, be referred to the following
      old Spanish works on the subject:--Andres Merino de Jesu-Cristo,
      "Escuela Palæographica, ó de leer Letras universas, antiguas y
      modernas, desde la entrada de los Godos en España" (Madrid, 1780, in
      fol. fig.);--Estev. de Terreros, "Palæographia Española, que contiene
      todos los modos conocidos, que ha habido de escribir en España, desde
      su principio y fundación" (Madrid, Ibarra, 1758, in 4to. fig.); and
      Rodriguez-Christ., "Bibliotheca Universal de la Polygraphia Española"
      (Madrid, 1738, fol. fig.).

 [74] That art which is called "illumination" in Paris.

 [75] "Lettere Sanese," tom. i. p. 278.

 [76] The well-known passage in which Dante alludes to Oderigi occurs in
      the eleventh canto of the "Paradiso," and is as follows:--

       "Oh, dissi lui, non se' tu Oderisi,
        L' onor d' Agubbio, e l' onor di quell' arte
        Che alluminar è chiamata a Parisi?
        Frate, diss' egli, più ridon le carte
        Che pennelegia Franco Bolognese:
        L' onor è tutto or suo, e mio in parte.
        Ben non sarei stato si cortese
        Mentre ch' io vissi per lo gran disio
        Dell' excellentia, ove mio cor intese.
        Di tal superbia qui si paga il fio."

 [77] Vita di Giotto.

 [78] "Storia Pittorica," vol xi. p. 13, ed. Pisa, 1815; and vol. v. pp. 8,
      9, 10.

 [79] Lanzi speaks of these choral books as "De' più considerabili che
      abbia l'Italia."

 [80] The Kensington Museum possesses two splendid leaves from a great
      "Chorale," which contain miniatures completely in the manner of Fra
      Angelico.

 [81] The Duke of Hamilton possesses some beautiful MSS. illuminated by, or
      in the manner of Memmi. Mr. Layard is the fortunate owner of one leaf
      of surpassing grandeur and elevation of style.

 [82] The style, if not the hand, of Taddeo Bartolo, another of the great
      early masters of the Siennese school, may be distinctly traced in
      several existing miniatures.

 [83] "Poetry of Christian Art," p. 140.

 [84] "Ornò i libri corali di figure nobillissime."--Cittadella, "Catalogo
      dei Pittori e Scultori Ferraresi," vol. i. pp. 1-27.

 [85] Rio. It must be a matter of delight to all lovers of true art that
      that most useful society, the "Arundel," has been of late turning its
      attention to the production, by means of chromo-lithography, of some
      of the finest examples extant of Italian quattro and cinque-cento
      illumination.

 [86] A small volume, which passed from the hands of the late Mr.
      Dennistoun into the collection of Lord Ashburnham, contains a series
      of arabesques and miniatures of the most interesting character,
      recalling in different pages, and in the highest perfection, the
      varied styles of Pietro Perugino, Pinturicchio, Lo Spagna, and
      others. The Duke of Hamilton's library is extraordinarily rich in
      Italian MSS.; his Grace's Dante with outline illustrations being of
      great importance.

 [87] _See_ Mr. Shaw's truly beautiful reproduction, in that gentleman's
      "Illuminated Ornaments," &c., of a portion of Arabesque border from
      this volume, containing a medallion portrait, Plate XXXV. A very
      beautiful Sforza MS. has lately been transferred from the possession
      of Mr. Henry Farrer to that of the Marquis D'Azeglio.

 [88] That Andrea exercised a great influence upon miniature-painting may
      be recognized in the works of Girolamo: a grand leaf from a folio, on
      which is painted a seated allegorical figure of "Rome," in the
      possession of Mr. T. Whitehead, is so noble in every way, and so
      entirely in Andrea's manner, that it seems almost impossible to doubt
      its being by his hand. It may, however, possibly have been executed
      by his contemporary in the Mantuan school, "Giovanni dei Russi," who
      in 1455 illuminated the great Bible of the house of Este, for Borso,
      Duke of Modena.

 [89] "Vita di Fra Giocondo e di Liberale, e d'altri Veronesi."

 [90] The Celotti sale, which took place at Christie's on the 26th of May,
      1825, and which included by far the most important collection of
      Italian illuminations ever brought to the hammer, contained no less
      than nineteen beautiful specimens extracted from the choral books of
      that pope.

 [91] _See_ Baglioni, "Vite dei Pittori ed Architetti fioriti in Roma, dal
      1572 sino al 1642,"--Vita di Giulio Clovio.

 [92] Facsimiles of the exquisite pages of this volume are given in Mr.
      Noel Humphrey's work; they are perfect triumphs of chromolithographic
      skill, and their production by Mr. Owen Jones formed what Germans may
      hereafter call a "standpunkt" in the history of that art, of which
      this volume presents no unfavourable sample.

 [93] Grenville Collection.

 [94] In his catalogue of the sale of the Celotti collection.

 [95] The Kensington Museum possesses a beautiful specimen by this artist,
      formerly in Mr. Ottley's collection. Two others of equal excellence
      are treasured among other gems of art, by Mr. Ram, of Ramsfort,
      Ireland. They all came from Celotti.

 [96] Mr. Whitehead's small but choice collection of specimens includes one
      quite worthy of the hand of Tintoretto.

 [97] Mr. S. Leigh Sotheby, in his admirable "Principia typographica," Dr.
      Dibdin in his "Bibliotheca Spenceriana," and the Baron de Heinecken
      in his "Idée générale d'une Collection complète d'Estampes, &c.,"
      give the best literary and graphic illustrations of the block books
      of the middle ages.

 [98] Our good fortune in possessing at the present time, and in common
      use, a remarkably clear and easily intelligible set of alphabets, was
      thus admirably noted in an article in the _Times_ newspaper of
      December 28th, 1859:--

      "Happily for us, the written symbols employed by the Romans, which
      are now the chief medium of expression for all the languages of
      Europe, America, Australia, and the greater part of civilized Africa,
      reflect exactly the rough and stalwart energy which made Rome to
      Europe what we are to the world. They have bestowed on us an alphabet
      as practically effective, and as suited to the capabilities of human
      vision, as any that could have been devised. This alphabet of ours is
      like an Englishman's dress--plain and manageable; not very
      artistically arranged, it may be, nor remarkable for copiousness or
      flow of outline, but sufficiently elastic and capable of extension.
      Its symbols have certainly no graceful curves like the picturesque
      Persian; but, better than all flourishes, each letter has plain,
      unmistakable features of its own. The vowels, which are to the rest
      of the alphabet what the breath, or rather life itself, is to the
      body, are assigned their legitimate position, and are formed to be
      written continuously with the consonants. Lastly, though scanty in
      itself, it is abundantly equipped with capital letters, stops,
      italics, and every appliance for securing rapid legibility, so that
      the eye can take in the subject of a page at a glance. Oriental
      alphabets are the very reverse of all this. They are complex,
      cumbersome, unmanageable." Much the same might have been said of many
      of the mediæval ones.

 [99] For excellent examples, see plates Technical Manual, Nos. 7 and 8;
      and Historical Manual, Nos. 7 and 9.

[100] The best are contained in the writings of De Quincy, Owen Jones,
      Winkellman, Pugin, and Sir Charles Eastlake.

[101] _See_ especially pages 24 to 28 inclusive, from which I transcribe a
      few elegant and suggestive passages:--

      "The student should keep," says Mr. Jewitt, "both in form and colour,
      as near to Nature as possible. No fantastic design can be so elegant
      as one copied and studied from Nature. What, for instance, can be
      more beautiful or more appropriate for intertwining with rich
      scroll-work than the convolvulus, the maurandia, the woodbine, the
      tropeolum, or the passion-flower? These painted upon a rich
      groundwork of diapered gold, or upon one of the beautiful grounds of
      the 15th century, composed of gold and blue or green, in fine waved
      or winding lines, crossing each other in every conceivable direction,
      form truly elegant studies, for almost all varieties of
      ornamentation. Whenever birds, insects, &c., are introduced, they
      should, as a general rule, be drawn true to nature; but they may,
      nevertheless, be turned and twisted into almost any position or
      shape. For instance, a lizard, with its beautiful emerald-green back,
      its yellow underparts, and rich brown mottlings, might be introduced
      with its long tail wrapped and twisted round the stem of a plant, and
      its little head, with brilliant eyes, shown just peeping out from
      under one of the beautiful flowers. The ladybird, with its bright red
      wings, covered with small black spots, might also be well introduced,
      creeping upon a leaf or stem. Hairy caterpillars, ants, beetles,
      snails, glow-worms, and even spiders, form also beautiful additions
      to a design, and may be introduced in almost any form or shape.
      Butterflies and moths, in their endless and beautiful variety, with
      their wings of every conceivable colour and shade, and of the most
      exquisite forms, are truly amongst the most beautiful and appropriate
      objects which the student can have for his mind to dwell upon. But
      not only these,--for occasionally a squirrel might be introduced
      perched upon the scroll-work; a cat, a goat, a dog, a monkey peeping
      out from behind a leaf; or, indeed, any animal, if artistically and
      naturally treated, may be introduced with really good effect.
      Flowers, fruits, shells, corn, &c., all add their beauties to a
      design; and, indeed, there is nothing in nature, no, not one object,
      but which may well be introduced into ornamental designing, and may
      be so translated and poeticised as to become appropriate to any
      subject."

[102] "Materials for a History of Oil-painting," by Charles Lock Eastlake:
      London, 1847.

[103] The most copious text of Heraclius is contained in the Le Bègue
      collection of writers on art, brought together by Master John Le
      Bègue, of Paris, in the 15th century.

[104] Sir Charles Eastlake does not place Heraclius so early as Raspe and
      Mr. Hendrie do. I incline to agree with the last-named critics.

[105] The text of Heraclius is given not from the Le Bègue manuscript, but
      from one less perfect, formerly at Cambridge, but now in the British
      Museum, Egerton 840 A, in Raspe's work--"A Critical Essay on
      Oil-painting." London, 1781.

[106] Muratori, "Antiq. Ital. Medii Ævi," p. 269.

[107] The title he himself gives to his work illustrates its comprehensive
      character--"Theophili qui et Rugerus, Presbyteri et Monachi Libri
      III. de diversis Artibus, seu diversarum Artium Schedula."
      Translations, with excellent critical comments, have been made by the
      Count de l'Escalopier into French, and by Mr. Robert Hendrie into
      English. In the extracts here given I have followed the accurate text
      of the last-named gentleman.

[108] I cannot take leave of this good old monk, the influence exercised by
      whose writings during the whole of the Middle Ages is proved by the
      numerous transcripts of them executed at different periods, still
      preserved in most of the chief European libraries, without giving him
      credit for a pure and liberal philanthropy worthy of imitation in all
      ages. Nothing can be more dignified and noble than the words in which
      he concludes the introduction to his work. After reciting the various
      arts he has endeavoured to illustrate, and the sufferings and labour
      through which the knowledge he desires to convey to others had been
      acquired by himself, he winds up by saying:--

      "When you shall have re-read this often, and have committed it to
      your tenacious memory, you shall thus recompense me for this care of
      instruction, that, as often as you shall successfully have made use
      of my work, you pray for me for the pity of omnipotent God, who knows
      that I have written these things which are here arranged, neither
      through love of human approbation, nor through desire of temporal
      reward, nor have I stolen anything precious or rare through envious
      jealousy, nor have I kept back anything reserved for myself alone;
      but, in augmentation of the honour and glory of His name, I have
      consulted the progress and hastened to aid the necessities of many
      men."

[109] It will be found given in extenso in the 32nd vol. of "The
      Archæologia," pp. 183-244, with an elaborate letter from its
      possessor.

[110] There is some confusion about this word, for it is used to denote
      mixtures which would produce real rose-colour, light warm yellow, and
      a perfect drab.

[111] That is, the mineral green with the vegetable madder.

[112] A beautiful example may be found in Dan Lydgate's legends of St.
      Edmund and St. Fremund, MS. Harleian, 2278.

[113] "Materials for a History of Oil-painting," by Charles Lock Eastlake
      (Lond. 1847), pp. 127, 128.

[114] Mr. Edwin Jewitt's little "Manual of Illuminated and Missal
      Painting," Mr. Randle Harrison's, Mr. Albert Warren's, and Mr. Henry
      M. Lucien's, published by Messrs. Barnard, of Oxford-street; Mr. J.
      W. Bradley's, and Mr. T. G. Goodwin's, published by Messrs. Winsor &
      Newton, of Rathbone Place; and Mr. Noel Humphrey's hand-book on the
      same subject, have no doubt proved useful to many, and helped to
      produce the quantity of good illumination now executed.

[115] For illumination in water-colour on paper, cardboard, or vellum,
      Messrs. Winsor & Newton, Rowney, Barnard, Newman, and others, fit up
      boxes with special selections of all requisite materials; including
      all that can be wanted for the application and burnishing of gold and
      other metals. Messrs. Miller's "Glass Mediums, Nos. 1 and 2," and
      Newman's "Preparation for sizing albumenized papers," are exceedingly
      useful for mixing with illuminating colours; giving great hardness
      and body to them, and preventing them from "washing up," in working
      over with glazing and other tints. I have found Mr. Barbe's powder
      body-colours give remarkably solid tints, with great freedom in
      working.

[116] This had better be bought ready prepared, since some experience is
      requisite in so applying the red chalk as to prevent its depositing
      under the weight of the hand, and yet coming off sufficiently in the
      line traced by the point.

[117] The experienced illuminator will generally do his writing before he
      gets in the outline of his ornament, and he will frequently dispense
      with the transferring process altogether; but it would be by no means
      safe for a beginner to do so.

[118] Both the cushion and tip will be described in detail under the head
      of Oil-gilding.

[119] The amateur may of course prepare mordants of different degrees of
      tenacity and body for his own use, by the employment, and various
      combinations, of leather and parchment size, isinglass, red lead, gum
      arabic, sugar, honey, glycerine, borax, plaster of Paris, bol
      ammoniac, glaire, and similar substances; but his time will be more
      profitably spent in improving himself in design than it could be
      (nowadays) in experimenting on the "materia technica" of art.

[120] This information is principally derived from Nathaniel Whittock's
      "Decorative Painter's and Glazier's Guide." It gives the usual
      practice of "Writers to the trade," but must, of course, be modified
      according to the specialities of any of the historical styles
      adopted.

[121] Japanners' gilding is a branch of oil-gilding, the size or ground
      being made with 1 pound of linseed oil, to which, while boiling, is
      added gradually 4 ounces of gum animi in powder, the whole being
      stirred until the gum is completely dissolved, and kept boiling till
      the mixture is of a thick consistence, in which state it should be
      strained through a thick flannel, and stored in a wide-mouthed
      stoppered bottle. Vermilion is ground up with the size before it is
      applied, to render it opaque; and if it does not leave the brush
      freely, it should be thinned with oil of turpentine.

      The gold powder may be either real gold, or what is called Dutch
      metal, or imitation gold. Gold powder is produced by grinding the
      leaf gold with pure honey on the stone till it is perfectly reduced
      to powder, and afterwards dissolving the mixture in water till the
      honey is completely removed, and for this several waters are
      necessary; the water is then poured off, and the powder dried. If
      this gold be mixed up with weak gum-water and spread upon
      cockle-shells, it is then called shell gold, which is used in
      drawings only.

      The Dutch gold powder is made by reducing the Dutch leaf gold by
      exactly the same process; and if well protected by varnishing, its
      appearance is little inferior to the genuine metal. There is another
      method of procuring gold powder, which is by precipitating grain gold
      into powder by means of aqua regia, which is made by dissolving four
      parts of pure spirit of nitre and one part of sal ammoniac in powder.
      This process was (as has been already stated) well known to the
      mediæval illuminators. In 4 ounces of this compound, ½ an ounce of
      grain gold is dissolved under the action of a slight heat; a solution
      of green vitriol, consisting of copperas 1 dram, water 1 ounce, being
      gradually added. When the precipitation has ceased, the gold powder
      must be carefully washed and dried, and will be found to be more
      brilliant than that made from leaf gold. The use of japanners'
      gold-size is very similar to oil-gilding, and is equally simple. If
      the material to be gilded is brought to a smooth and clean face, the
      size may be laid on at once without other preparation; using great
      care, however, not to touch any part but what you wish to gild, as
      the gold will adhere wherever there is size. Priming with a mixture
      of chalk and size is sometimes used for a first coat, but not by the
      best japanners, as the work is liable to chip off; no material should
      therefore be japanned which cannot be made smooth. For hard or
      close-grained wood, metal, leather, or paper, one or two coats of
      varnish will answer all requirements; very great care being observed
      that each coat of varnish be perfectly dry and hard before it is
      again touched. It is a good practice to allow the work to stand a day
      or two between the applications; then the japanners' gold-size may be
      added, and touching with the finger as before described will indicate
      the proper state for applying the gold, whether in leaf or powder.
      Either may be employed; but in the case of colours being intermixed
      and subsequently varnished, the powder is usually adopted; it is
      easily laid on by means of a camel-hair brush, the work being set
      aside to get thoroughly dry, when the superfluous metal is removed
      with a soft brush. In case more size should have been prepared than
      is needed, the remainder, if water be poured over it, will keep for
      future use.

[122] The superiority of the Chinese and Japanese varnishing is chiefly
      owing to the excellence of a particular species of resin found in
      China and Japan. The varnishes made with oil are longer drying than
      those made with spirits of wine, but are of greater durability. The
      spirits of wine should be highly rectified: if oil is used, it should
      be linseed. It is safer to purchase the varnish ready prepared than
      to attempt the making of it, as the solution of resin, particularly
      in oil, is somewhat dangerous.





*** End of this LibraryBlog Digital Book "The History, Theory, and Practice of Illuminating - Condensed from 'The Art of Illuminating' by the same - illustrator and author" ***

Copyright 2023 LibraryBlog. All rights reserved.



Home