Chapter XIX: Section C: shows a plain letter of modern cut known erroneously as (8)

Believing that such a service would be appreciated by printers and users of typography, the author made a study of the thousands of Roman type-faces on the market, with the purpose of selecting a half-dozen representative standard faces. It was felt that these type-faces should be legible and good-looking, and possess character; that their merits should be acknowledged by authorities, and that the types should be capable of wide serviceability on both book and job work.

EXAMPLE 464-B

Proportions of Roman capitals as found in the Trajan alphabet, and a
few additional letters
]

EXAMPLE 465

Evolution of Roman lower-case type-faces. (A) Pen-made Roman capitals.
(B) Development into Minuscules or lower-case thru rapid lettering.
(C) Black Letter or German Text developed from Roman Uncials. (D)
White Letter, the open, legible Caroline Minuscules, on which Jenson
based his Roman type-face of 1470. (E) A recent type-face closely
modeled on Jenson’s Roman types. (F) Joseph Moxon’s letters of 1676.
(G) Caslon’s type-face of 1722
]

The face first selected—and without hesitation—was Caslon Oldstyle as originally designed. Scotch Roman was the second selection, Cheltenham Oldstyle the third, Cloister Oldstyle the fourth, Bodoni Book the fifth, and French Oldstyle the sixth. (All shown in Example 467.)

Type-faces designed and cut for private use were not considered in making these selections, as it was believed best to adhere to type-faces that are procurable from most foundries and that are available for machine composition. It may be well to inject here a warning that most so-called Caslon Oldstyles are not as good as the one selected (Example 467-B); that Jenson Oldstyle is inferior to Cloister Oldstyle (Example 467-A) as a representative of the original Jenson type. However, good representatives of Scotch Roman (Example 467-D) are obtainable under the name of Wayside, of National Roman, etc.

EXAMPLE 466

Two standard type-faces that rate high in legibility, but that are
colorless in the mass and lacking in the pleasing irregularities of
form that characterized Roman type-faces before the nineteenth
century. The various qualities of legibility found in Modernized
Oldstyle have been converted to narrower letter shapes and more
“modern” form in Century Expanded
]

EXAMPLE 467

Six standard representative Roman type-faces, approved by authorities
for both beauty and legibility, and selected by the author from the
thousands of type-faces available for hand and machine composition
]

Selection of these faces was also made just as a person makes a wise selection of records for a Victrola—for permanency and investment. As standard records can be selected which will “wear” for a long time, so standard type-faces can be selected which will look well for many years. Good type-faces are like good music.

We will discuss these six faces.

* * * * *

_Cloister Oldstyle._—Use in American printing of a standard type-face based on the Roman letter cut by Nicholas Jenson about 1470 has been retarded by the fact that typefounders in this country some years ago used as a model Morris’s interpretation of the Jenson face instead of going direct to the original letter. Jenson’s face had been copied by Bruce Rogers, T. J. Cobden-Sanderson, Frederic W. Goudy and others, and the fonts used privately, but for general use it was not until 1914 that a creditable copy of the face was made available by the good judgment displayed by the American Type Founders Company in bringing out Cloister Oldstyle. This is probably the best rendition of the spirit of an old Roman type-face that a modern type foundry has made. If there is a fault, it is that the lower-case letters set a trifle too close.

EXAMPLE 468

Type-face used in Italy by Sweinheim and Pannartz in 1465. From De
Vinne
]

EXAMPLE 469

The Roman types of John and Wendelin of Spires, Venice, 1469. Greek
letters were to be written in the line now half blank. From the
original in Typographic Library and Museum, Jersey City, N. J.
]

EXAMPLE 470

The beautiful “White Letter” Roman type-face of Nicholas Jenson, 1470,
from “Eusebius,” the first book printed by him at Venice. The dot
over the “i,” small and slightly to the right, is found, also, in
the manuscript specimen here inserted. From the original in the
Typographic Library and Museum, Jersey City, N. J.
]

EXAMPLE 471

Section of manuscript of the fifteenth century, in the Charlemagne
“White Letter” on which Jenson may have modeled his Roman type-face.
From Humphrey
]

Jenson’s face was not the first Roman type made, as is sometimes supposed; others had previously been used (Example 469), but it was the first Roman type to meet the approval of those who preferred the Italian White Letter to the Gothic Black Letter.

EXAMPLE 472

Type-face used by Paul Manutius at Venice in 1566
]

Sweinheim and Pannartz were German printers who, in 1465, began to print near Rome. Until this time the black Text letter such as Gutenberg had printed with ten years before was the only one used in typography, but book buyers in Italy wanted Italian lettering. The German printers attempted to cut an Italian face, but in appearance they did not advance far from Gothic lettering (Example 468).

EXAMPLE 473

Cloister Oldstyle set in imitation of the Jenson title on page 14.
Printed and zinc-etched, as was the original
]

EXAMPLE 474

Type-face cut in 1693, now used by the National Printing Office,
Paris. Notice slight projection on lower-case “l,” a decorative
feature used by Gutenberg in his types
]

EXAMPLE 475

The first line is set in Cloister Oldstyle; second reproduced from
Elzevir page opposite; third set in Cadmus; fourth in Caslon
Oldstyle. These lines are zinc-etched
]

John and Wendelin of Spires, Germans, began to print at Venice in 1469 with a type-face that was more Italian in character, yet it had Gothic characteristics. Nicholas Jenson, a Frenchman, who began to print in 1470 at Venice, was more successful, and he gave to the world the fine Roman type-face that he used on his first book, shown, together with a sample of contemporary lettering of similar design, as Example 470. More facts about Jenson will be found on page 14.

EXAMPLE 476

Cheltenham Oldstyle, with the capitals a size smaller, set in
imitation of the Plantin typography on page 16
]

EXAMPLE 477

Type-faces used by Daniel Elzevir at Amsterdam in 1675. From “The
American Printer” library
]

EXAMPLE 478

The Roman types of Fournier, the French founder, taken from his type
book of 1766
]

EXAMPLE 479

A capital alphabet drawn by Joseph Moxon (1676) to a scale of
forty-two small squares
]

EXAMPLE 481

Earliest known specimen sheet of printing types, issued by William
Caslon in 1734
]

As time went on, printers and typefounders made changes in the Jenson design, but these alterations were not numerous (the cross stroke in the lower-case _e_ was one, Example 495), and fairly close resemblance to the Jenson types is found in the Roman types of Paul Manutius, 1566 (Example 472), Daniel Elzevir, 1675 (Example 477), and Fournier, 1766 (Example 478).

EXAMPLE 480

Moxon’s lower-case alphabet, which he says was patterned after the
letters of a Dutch punch cutter
]

In selecting a page from Jenson’s book, “Eusebius,” to be photographed, the writer had great difficulty in finding one of sufficient evenness and clearness of print. It may be that Jenson, in his endeavor to enhance the “White Letter” effect of his types, used a minimum of ink, and, under these circumstances it was perhaps more difficult for him to maintain consistent page color than it was for Gutenberg to do so with his “Black Letter” pages. Bruce Rogers’s printing of his Jenson-like pages of typography would be a revelation to the old-time Venetian printer.

Jenson made his Roman letter in only one size, and that fairly large (about sixteen-point), so that the open design of the characters would admit white light from the paper background. When, some years later, he was required to print cheap devotional books, he cut and used a compact Black Letter Text type for the purpose.

There is now a movement away from the mechanical style of type-face, restricted in design by arbitrary craft rules, in the direction of the graceful, interesting yet legible faces represented by Cloister Oldstyle.

This type will be found in use in Examples 171, 240, 256, 266, 286, 316, 337, 340, 341, 345, 357, 359, 372, 383, 399 and 452.

The Italic that accompanies Cloister Oldstyle is based on the Italic of Aldus Manutius.

* * * * *

_Caslon Oldstyle._—This is a historic American type-face. It was used by that renowned American printer, Benjamin Franklin; it appears on eighteenth-century school books, and even our paper money was printed with it. It was the type-face used by John Dunlap, the Philadelphia printer, in setting up and printing the first published copies of the Declaration of Independence, a few hours after it had been passed by Congress, July 4, 1776. (See facsimile that appears as a frontispiece to this book). It was one of these copies that was read to the Continental Army.

EXAMPLES 479-A and 480-A

The Moxon alphabets inked in and reproduced in a smaller size
]

Today, nearly two hundred years after the first font was cut, Caslon Oldstyle has the approval of the good printers of America, who regard it, in its original form, as the best and most useful Roman type-face available, as it is one of the few old faces the punches and matrices of which have been preserved and placed at the disposal of all printers.

The Caslon foundry in England now casts the original letter as “Caslon Old-face.” In this country, all typefounders make some sort of Caslon, and it may be procured practically in its original form, with long descenders, kerned _f_ and _j_, old special characters, decorated Italic capitals and all, from the American Type Founders Company, which has named it “Caslon Oldstyle 471” (Example 517).

Other foundries are preparing to furnish the letter more in its original form, and composing-machine makers are doing likewise. The monotype company, after cutting several faces in imitation of imitations of Caslon’s type, finally copied the original fairly successfully, and this, as now obtainable, is known as Caslon Oldstyle No. 337. The linotype company is recutting its Caslons and is making a sincere effort to reproduce the original letter as closely as possible. Mechanical difficulties prevent an exact duplication in every detail of the Caslon type in machine composition, yet the companies that are attempting it show a spirit that augurs well for the future quality of printing in America.

Caslon Oldstyle is not at its best printed on a smooth, highly-calendered surface.

When properly composed and printed, it presents as a page a dark-gray tone, formed of a pleasant mixture of black ink and white paper; its letters show more of a contrast in thick and thin strokes than do the Roman types of Nicholas Jenson, and have more color and interest than the modernized old-styles of the nineteenth century. In the sizes above ten-point it measures up in legibility and beauty to the requirements of the best printers and other students of typography.

Caslon did not create an entirely new Roman letter, but with the elements that had already been provided by Jenson and the type designers of Holland he did, with good taste and skill, build a type-face that is distinctive, legible and beautiful. The illustration of the evolution of Roman lower-case in Example 465 furnishes opportunity for study of this point.

The story of the designing of Caslon Oldstyle has been often told, yet a few brief facts may be of interest here.

EXAMPLE 482

There were two slightly different faces to many of the Caslon fonts.
From the specimen pages in Luckombe’s “History of Printing,”
published in 1770
]

Typefounding and printing had deteriorated in England, and the product of the printing press was pathetically poor (the author was surprised, in examining important volumes printed in England about 1700, at the consistently poor work of typefounder, compositor and pressman), when, in 1720, William Caslon, an engraver on metal, was called upon to cut a font of Arabic characters, to be used in printing a New Testament for the poor Christians in Palestine and Arabia. At the same time, for use at the foot of the proofs of his Arabic types, he cut his own name in Roman, pica size. (This was at the time when Benjamin Franklin was working as a journeyman at Watt’s printing office.)

The pica size, however, was not cast immediately. So far as is known, the size of Roman first to be cast by Caslon was that called “English” (about fourteen-point), which he cut in 1722. Additional sizes followed, and at about the period of his death, in 1766, there were at least thirty fonts of Roman (pearl to five-line pica), nineteen fonts of Italic, ten fonts of Black Text, and various fonts of Greek, Hebrew and other alphabets, together with more than a half-hundred fonts of flowers and bands.

The fonts that have been preserved to us, and that are purchasable, do not include all. There were Nos. 1 and 2 faces in nonpareil, brevier, bourgeois, long-primer, small-pica, pica, english and double-pica (two-line small-pica). The No. 1 face in each case was a trifle larger than that of No. 2, and a third face, still larger, was made in bourgeois and long-primer.

These faces were all shown, in honor of the memory of William Caslon and as a tribute to the merit of the type, in Luckombe’s “History of Printing,” published by Adlard & Browne, London, in 1770 (Example 482).

It is probably the smaller and lighter No. 2 faces that have been preserved to us in the ten-point (long-primer) sizes and under. The lighter, more open eighteen-point (great-primer) Italic (Example 517), so different from the Italic of the other sizes, is as Caslon made it. In some respects it is better than the Italic of other sizes.

While the general appearance of all sizes of the Roman is the same (and of the Italic also, with the exception noted), there are slight differences in details of design in the various sizes. For this reason, when Caslon Oldstyle is compared with the original letters, the comparison should be with the same size in each instance.

De Vinne, perhaps influenced by the prejudices and practices of nineteenth-century typefounders, in one of his books points out as defects in the original Caslon types features that typefounders and composing-machine makers are today restoring to the face as used in America. The “defects” were supposed to be: “Too long a beak to the _f_ and _j_; unnecessary narrowness in the _s_ and _a_, and in some capitals; too great width of the _c_, _o_ and _v_.”

Typefounders and composing-machine makers have filled the type cases of the printers of this country with Caslons that have been improved (!), recut, modified, adjusted to “lining” systems, made bold, disfigured, sawed-off and ill-treated generally; and as they bring out good copies of the original letter, it is to be hoped that, so far as possible, sale of the defective and offending type-faces and matrices will be discontinued.

William Caslon died in 1766, and the business was continued by William Caslon the second, and then by William Caslon the third. Mrs. Henry Caslon conducted the business in 1796, when was brought out the modernized Caslon (Example 484-B) that was probably the model for the type we know as Scotch Roman. This modernization was a concession to the demand for types in the new style that had been designed by Giambattista Bodoni, an Italian typefounder.

The influence of Bodoni’s ideas in type design was such that Caslon Oldstyle was not included in the 1805 specimen book of the Caslon foundry, and did not again appear therein until 1860, a year after the face had been reintroduced to America by a Philadelphia foundry.

The story of the revival of Caslon Oldstyle in England has been told in these words:

In the year 1843, Mr. Whittingham, of the Chiswick Press, waited
on Mr. Caslon to ask his aid in carrying out the then new idea of
printing, in appropriate type, “The Diary of Lady Willoughby,” a
work of fiction, the period and diction of which were supposed to
be of the reign of Charles I. The original matrices of the first
William Caslon having been fortunately preserved, Mr. Caslon
undertook to supply a small font of great-primer. So well was Mr.
Whittingham satisfied with the result of his experiment that he
determined on printing other volumes in the same style, and
eventually he was supplied with a complete series of all the old
fonts.

America took part early in the revival of Caslon’s old-style types, for in 1858 matrices were brought to the United States by the Johnson Type Foundry, afterward MacKellar, Smiths & Jordan, now the Philadelphia branch of the American Type Founders Company.

The first size made in this country was brevier, which was announced in the “Typographic Advertiser,” the company’s house-organ, of January, 1859. Six months later thirteen sizes were shown, and in 1865 Caslon Oldstyle was made in sizes from pearl to six-line pica.

The decorative Italic types, known as Swash letters, were not until 1916 included in the American fonts, but these, redrawn by T. M. Cleland, are now procurable.

Selection of Caslon Oldstyle, in 1892, for the text and advertisements of _Vogue_, the use of the letter by Will Bradley at the Wayside Press, in 1896, and the general old-style revival, influenced by the work of William Morris, of the Kelmscott Press, in England, were causes that led to the present popularity of Caslon Oldstyle.

This Caslon style of type-face is one of the few standard letters that has practically unlimited usefulness. Properly treated, it looks well on any class of printing, from a business card to the text and advertising pages of periodicals. It may be seen in action in this book by referring to Examples 18, 34, 64, 65, 66, 67, 68, 81, 82, 86, 96, 119, 121, 126, 127, 132, 133, 134, 135, 138, 150, 151, 167, 168, 169, 172, 178, 180, 188, 190, 191, 195, 203, 215, 221, 222, 223, 224, 225, 228, 230, 231, 232, 236, 239, 245, 246, 247, 248, 249, 250, 260, 262, 268, 274, 282, 284, 287, 289, 290, 292, 294, 296, 297, 300, 374, 375, 382, 384, 385, 393, 394, 396, 426, 429, 430, 433, 434, 436, 438, 439, 440, 441, 448, 450, 451, 453, 455, 459, 460 and 461; the frontispiece; upper specimen, page 21; reproductions, page 23; insert opposite page 32; upper specimen, page 33.

EXAMPLE 483

On the left is a reduced reproduction (from Reed) of Baskerville’s
types used in a book of 1758, and on the right, for comparison, is
the same matter set in Baskerville Roman, twelve-point size,
introduced to America in 1917
]

EXAMPLE 484

The possible descent of Scotch Roman. (A) Baskerville Roman, based on
Baskerville’s letter of 1758. (B) Mrs. Caslon’s “modern” Roman of
1796. (C) Scotch Roman, as we know it today
]

_Bodoni Book._—When a representative of the Bodoni type family was selected for a place in the group of standard faces, Bodoni Book was chosen for its refinement and legibility. In its lower-case, especially when printed on other than gloss-surfaced paper, it seems to have more of the qualities of the best types of the Parma printer than does Bodoni, the first series brought out by the American Type Founders Company. Scientific tests have demonstrated that in the small sizes usually used for text purposes, type-faces with considerable contrast of hairlines and heavy strokes strain the eyes. The contrast found in Bodoni Book is moderate.

John Baptist Bodoni was born in Italy twenty years after William Caslon cut his first font of Roman letter. Bodoni’s subsequent career as a printer and typefounder doubtless resulted from his entrance, when eighteen years old, to the printing house of the Propaganda, in Rome. He began his important work in Parma in 1768.

The print of his books is remarkable for clearness, and in both type-cutting and printing he was skilful to a high degree. His types are not quite so regular in design as the recent Bodonis, and were given additional character by being impressed firmly in paper that did not have the enameled surface so much in use in our day.

His type-faces were a change from those previously used—different from the type-faces of Jenson, the Elzevirs and Caslon. The change was principally in the serifs, which were straight, thin strokes; and the general character of the letters was also distinctive. He used a few fonts in which the straight serifs were rounded by being filled in slightly at the corners, and these fonts probably suggested to Scotch, English and American founders the style known as “modern,” as compared with the “old-style” faces in use before Bodoni’s time.

The reader can compare for himself the resemblance of Bodoni Book to one of the text types of Bodoni as used in 1789. In order that the comparison could be fairly made, the Bodoni Book (Example 486) was printed on antique paper and etched on zinc, as the original Bodoni page had been (see page 26). This type is not only lighter, but other details are different from a 1790 type of Bodoni (Example 487-A), which may have served as a model for the present-day type-face called Bodoni. Compare the capital _R_s, and notice that the cross stroke on the lower-case _t_ in the 1783 alphabet is low and distinct, while on the 1789 page it is high and joined with the upper end of the vertical stroke. The latter treatment prevails on many of Bodoni’s types, as it does on practically all type-faces. Excepting for the straight serifs, many of Bodoni’s types have some general resemblance to other Roman types of the seventeenth and eighteenth centuries.

In designing the present-day Bodoni, the typefounders gave it much of the “finish” in design that was given type-faces by Scotch and American founders during the nineteenth century. Typefounders seldom succeed in faithfully reproducing old type-faces, for the reason that they “correct the faults” of the old faces, and in doing so lose much of the humanistic spirit of the original. Were it not that the punches of some of William Caslon’s types were preserved, we would not have the true rendering of this old-style that we now possess. Modernized Oldstyle represents the Caslon face corrected and improved as the typefounders of the nineteenth century thought it should be. Some American founders, however, are making excellent progress in reproducing the spirit of old type-faces, and it is possible that we may yet have a satisfactory copy of the famous “silver” types of the Elzevirs.

Bodoni types are shown in use on page 26, and in Examples 26, 36, 48, 175, 176, 194, 278, 310, 322, 330, 352, 355, 365, 387, 390, 412, 413, 414, 416, 428, 437 and 458.

* * * * *

_Scotch Roman_.—During the nineteenth century English, Scotch and American typefounders took Bodoni’s type-faces, redesigned them more along mechanical lines, filled in the corners of his straight serifs so as slightly to round them, and gave to the English-speaking world what are best known to printers as the modern Romans.

Looking over the specimen books of the type foundries of the last century—those of Bruce, MacKellar, Smiths & Jordan, and others—one marvels at the wonderful skill in the cutting of the hairlines and the general mechanical excellence of the modern Romans.

Because of their continued use on newspapers and in books of reference, these letters are in wide use today, altho there is a movement away from them.

The most generally used type-face on small city newspapers in America is the linotype company’s Modern Roman No. 2 (Example 489). In the hand-set days this broad, open-faced letter was so made for the reason that it would stereotype easier than would the narrower letters. It was adopted by the ready-print and plate houses, and when the newspapers installed linotypes they naturally selected the same face, so that the type matter thruout the publication would match. The larger city dailies give more attention to the details of the faces they use for body purposes. If No. 2 is selected, or the leaner No. 1, modifications are frequently made in some of the letters or figures. The No. 1 Modern Roman is used by the New York _Times_ (Example 406).

However, instead of the selection of a severe letter, such as the so-called Scotch-cut Modern Roman as a representative type-face in the group of six standard Roman letters, the more interesting Scotch Roman won the place. This type-face was first made in America by A. D. Farmer & Son, and the eight-point and ten-point sizes were used as body type by _The American Printer_ first in April, 1902. In September of that year there appeared in that publication showings of the eight-point, ten-point, eleven-point, twelve-point and fourteen-point sizes, with the statement that other sizes were in preparation. The name, “Scotch Roman,” probably comes from the fact that Miller & Richard, the Edinburgh typefounders, made a similar letter in sizes from eight-point to twelve-point. However, they call it “Old Roman.”

Scotch Roman is the link that connects the graceful old-style and the severe modern Roman. Compare it with the letter (Example 484-B) made at the foundry owned by Mrs. Henry Caslon in 1796, and notice the resemblance. (As a further study, there is added an alphabet of Baskerville Roman, recently introduced to America.) Scotch Roman is also procurable under the names of Wayside Roman and National Roman.

Scotch Roman has been used in Examples 38, 39, 77, 78, 85, 143, 152, 153, 154, 185, 186, 238, 271, 317, 400, 420, 423, 427, 428 and 444.

* * * * *

_French Oldstyle._—This is the title by which is known among printers a style of type-face made, under the name of “Elzevier,” by the Gustave Mayeur foundry of Paris in 1878. Its design, M. Mayeur tells us, was suggested by types used in a book printed in Leyden by the Elzevirs in 1634.

In 1889, Farmer, Little & Co. of New York procured drives of five sizes from France and began making the letter in America, naming it “Cadmus.” The type foundry of Phelps, Dalton & Co. of Boston had as early as 1884 made a similar and rather pleasing letter in capitals only, which it called French Oldstyle.

Theodore L. De Vinne was the first to use the Mayeur French Oldstyle in America. It was a favorite with Walter Gilliss, who adopted it in remodeling the catalogs and other printed matter used by the Metropolitan Museum of Art, New York.

EXAMPLE 485

A study of the French Oldstyle design of Roman capitals
]

The capitals of French Oldstyle are especially pleasing, and when this letter is employed they should be used in display with frequency. The Mayeur letter would be even more attractive if the proportions of its capitals were nearer those of early Roman lettering (Example 464-B), as are some other French Oldstyles.

French Oldstyle is tall and formed with liberal space between the strokes of the lower-case letters. It should always be leaded, as its open character and large lower-case demand that it shall not be crowded. Like Cheltenham, it is best suited for pages that are narrow and long, such as the 12mos issued in the seventeenth century by the Elzevirs.

The Mayeur French Oldstyle has a slight resemblance to the type-faces found in the Elzevir seventeenth-century books (Example 475), but is a rather free rendering of those types. The capitals of the Elzevirs are more in the proportions of Jenson’s (see Cloister Oldstyle, Example 467-A), and the small letters (_o_) of the lower-case are not so large in comparison with the tall letters (_l_) as they are in French Oldstyle. The Mayeur letter also contains the shortened descenders and short-kerned _f_ of the nineteenth-century typefounders. In fact, both ascending and descending strokes have been shortened.

Examples 45, 183, 184, 217, 410, 411, and the Jacobi specimen on page 33 show French Oldstyle in use.

* * * * *

_Cheltenham Oldstyle._—This extremely popular type-face was designed in America and here developed into the most numerous family known to typefounding—at least thirty series. A pamphlet issued in 1905 by the Mergenthaler Linotype Company tells the story of its origin. Ingalls Kimball, director of the Cheltenham Press, New York, was commissioned to suggest and supervise the making of a letter that would have both beauty and legibility. Cheltenham Oldstyle resulted, the drawings of which were made by Bertram G. Goodhue.

It was realized by the designers of this type-face that space between lines increases legibility, and experiments seem to show that the upper half of lower-case letters are easier to read than the lower half. For this reason the lower-case was then designed with long ascending strokes and short descending strokes. While the theory regarding the superior legibility of the upper half was announced as a new one, it had been put forth about 1882 by Dr. Émile Javal, a Parisian oculist. (See the demonstration in Example 508-G.)

The first announcement by the American Type Founders Company of Cheltenham Oldstyle in _The American Printer_, January, 1903, showed eleven sizes—eight-point to forty-eight-point. Claim was made that it was a “compact and legible book type,” and that it would become the most popular job type.

Because of its lean formation and close set it has been frequently used for narrow booklets.

A Roman type-face entirely new in design would be a freak and unreadable in the mass. As a completed design, Cheltenham Oldstyle was different from others in use in 1903, yet it is made up of features of Roman type design in vogue between the designing of Jenson’s letter and Caslon’s types. Study Examples 470, 472, 476, 477 and 478. Compare with the Portuguese capitals, especially _G_, _P_, _W_, in Example 485.

Cheltenham Oldstyle has strength, readability, and a certain amount of beauty. Its capitals are awkwardly large in comparison with its lower-case. It is shown in use with smaller capitals in Example 476.

Because of the smallness of most of the lower-case and the close set of the letters, very little space is needed between words; yet, as used by printers generally, much of the legibility the type-face possesses is lost by wide spacing between words, especially on machines.

Because the eye, as a rule, gives less attention to the lower part of a line of reading matter, the descenders were cut off almost entirely in the designing of Cheltenham. It would be as reasonable, because the eye usually sees only the upper part of a man, to amputate his legs. This amputation of the legs of type-faces by modern founders has maimed many other letters, notably Caslon Oldstyle as we most frequently see it.

The serif, which is supposed to distinguish modern types from old-style, is very small in Cheltenham Oldstyle, and is more like the serif found in modern types. The general character of Cheltenham, however, is extremely old-style.

This letter has been used in the composition of Examples 35, 40, 47, 106, 122, 125, 209, 267, 273 and 350.

DEVELOPMENT OF THE ROMAN TYPE-FACE

In the beginning, Roman letters were in capitals only (Example 464-A), and, carved on stone, were at their best about the beginning of the Christian era. The present so-called Roman, Italic and Text types are descended from them.

EXAMPLE 486

Resetting in Bodoni Book of the 1789 Bodoni specimen on page 26. Like
the original, this specimen was printed and then zinc-etched
]

Use of the pen and brush in lettering and the tendency to letter rapidly finally gave us the lower-case type of today. The evolution is pictured in Example 465. “A” shows the Roman capital alphabet as made with the pen (compare with the chiseled letters of Example 464-A). In “B” we have the result (known as uncials) of hurried writing of the Roman capitals, and one can see how the lower-case letters were forming. At about this point in the development, the minuscules, or lower-case letters, took two separate growths—one in the direction of the “Black Letter” or German Text types of today, and the other toward the “White Letter” of the Caroline Minuscules, which, as Roman lower-case, are now used almost exclusively in America, England, France and other countries in North and South America and Europe, the notable exception being Germany.

EXAMPLE 487

Comparison of one of Bodoni’s alphabets with a new type-face. Both are
rough, having been zinc-etched
]

When Gutenberg began to print he used a black Text type designed after the bold lettering of the German manuscript books, and when Germans went to Italy to print they had difficulty in rendering into types the character of Italian lettering. Example 468 shows the efforts in this direction by Sweinheim and Pannartz in 1465, and in Example 469, types by John and Wendelin of Spires, an improvement is noticed.

It remained, however, for the Frenchman, Jenson, to design and cut a letter that is the alpha of Roman type-faces (Example 470). Jenson was fortunate in the selection of a model, and it is interesting to include in the corner for comparison a section of an Italian manuscript of the same century. Development of the Roman letter into the beautiful and legible form known as “Caroline Minuscules” was due to the encouragement of Charlemagne. Why it is called “White Letter” in contradistinction to “Black Letter” may be seen by comparing the light-gray tone of the Jenson page (Example 470) with the missal specimen opposite page 14. It may be well to explain that the seeming disfigurement of the capitals in the Jenson page is due to the custom that book decorators had of placing a stroke of vermilion over the capitals. In the reproduction these strokes came out black. Both the Spires and the Jenson specimens were reproduced from the original volumes in the Typographic Library and Museum at Jersey City.

Example 465 also shows, for comparison with the style of lettering known as the Caroline Minuscules, a type-face based on Jenson’s letter of 1470, following which are Moxon’s and Caslon’s lower-case alphabets.

The faithfulness with which Jenson’s type-face was followed in the designing of Cloister Oldstyle may be seen by comparing the Cloister alphabet (Example 467-A) with the Jenson letters as they are found in Example 470. A very interesting comparison can also be made of Jenson’s capitals, on page 14, and Cloister Oldstyle, in Example 473. If a comparison of present-day type-faces with very old faces is to have any value, the present-day type-faces should be printed on antique-finished paper dampened if possible. And if the old face is reproduced by zinc etching from an old book the other type-face should also be zinc-etched. Printing and zinc etching were resorted to in the case of this example and in other instances in this chapter.

In some works reproductions of type-faces are valueless and misleading, for the reason that they have been retouched by an artist, redrawn, or cut in wood.

EXAMPLE 488

Modern Romans of the nineteenth century in three tones
]

The influence of the Jenson style of Roman was seen for some years, but in 1566 (Example 472) we find a change of form. The type is more condensed, and instead of a diagonal stroke on the lower-case e there is a horizontal stroke close to the top of the letter. This style, as used by Paul Manutius in 1566, seems to have been maintained with little alteration in the work of Daniel Elzevir a hundred years later (Example 477) and two hundred years later in the types of Fournier, the French founder (Example 478). This style of letter was also used by Plantin, the Antwerp printer, in 1569, and for the interest there may be in it the reproduction of a bit of his work (see page 16) should be compared with Example 476. This last-mentioned example has been set in Cheltenham Oldstyle. However, ten-point capitals have been used with twelve-point lower-case. It may be that the appearance of most reading matter would be improved if the capitals were a trifle lower in hight than the ascending strokes of the lower-case letters _b_, _d_, _f_, _h_, _k_, _l_.

EXAMPLE 489

Modern Roman as it is used on many American newspapers. Six-point,
seven-point and eight-point Linotype Roman No. 2. A readable but not
a handsome type-face
]

The National Printing Office at Paris is using on some of its productions a type-face designed in 1693 by Grandjean. While it is greatly like other letters of that period, it has peculiarities, one of which is a slight dot on the left center of the lower-case _l_, a feature also present in the Black Letter of Gutenberg, and in the types of Fust and Schœffer (see inserts opposite pages 7 and 12).

We have a key to the formation of the Roman types of the seventeenth century in the alphabets drawn by Joseph Moxon, an English typefounder, and published in 1676 (Examples 479 and 480). These are here reproduced from Moxon’s “Mechanick Exercises,” as reprinted, under the supervision of Theodore Low De Vinne, by the New York Typothetæ in 1896. Moxon says these letters are copied from the letters of Christopher van Dijk, a punch cutter of Holland. As will be seen by the scale in the upper left corner of Example 479, he attempted to show how the shapes of letters were “compounded of geometric figures, and mostly made by rule and compass.” Each letter was to be plotted on a framework of small squares, forty-two squares in hight and of a proportionate width. Passing up this idea as impractical, De Vinne said:

It is admitted that the characters are rudely drawn, and many have
faults of disproportion; but it must not be forgotten that they
were designed to meet the most important requirement of a reader—
to be read, and read easily. Here are the broad hair line, the
stubby serif on the lower-case and the bracketed serif on the
capitals, the thick stem, the strong and low crown on letters like
_m_ and _n_, with other peculiarities now commended in old-style
faces and often erroneously regarded as the original devices of
the first Caslon.

The writer was curious to see how these letters of Moxon’s made up into type, and for this purpose had the outlined alphabets inked in and reproduced in smaller size (Examples 479-A and 480-A).

When William Caslon, in 1720, cut his first Roman letter he may have had Moxon’s Roman alphabet before him. There are some resemblances, but that he did not slavishly follow the Moxon letters may be seen by comparing both type-faces (Examples 465-F and 465-G).

We could be more familiar with the types of Baskerville than we are, but now that the typefounders are reproducing them (Example 483) we shall come to a realization that they rival the letters of Caslon in beauty and readability. John Baskerville, the English typefounder and printer, spent six years and thirty thousand dollars in designing and cutting his first font of type. In 1758 he had cut eight fonts of Roman, but other printers would not buy his types, preferring those of Caslon. Two years later he attempted to sell his types and retire from the printing business. After his death, in 1775, his widow sold all his type and type-making material and they were removed to Kehl, near Strassburg. There is a legend that the types finally reached France and were melted into bullets for the French Revolution. It is a pity that Baskerville’s original punches are not available, like those of Caslon, so that printers of today could make use of the splendid type-faces that were unappreciated in his lifetime. However, Baskerville Roman is a satisfactory reproduction of the types of John Baskerville.

Like Caslon Oldstyle, however, it should be printed on other than gloss-surfaced paper (Example 463), as then, especially in the smaller sizes, the good qualities of the letter are brought out and it is made more legible.

When Mrs. Henry Caslon, in 1796, “Bodoni-ized” the types of William Caslon the first, she was probably influenced as much by the types of Baskerville as by those of Bodoni. This is borne out by the comparisons in Example 484. In “A” is shown an alphabet of Baskerville Roman as now made in America, and in “B” Mrs. Caslon’s Modern Roman. The third type-face (C) is Scotch Roman, one of the six selected representative type-faces (Example 467-D).

John Baptist Bodoni made some very good type-faces, but it is not difficult to lay at his door the responsibility for throwing typography out of gear for almost a century. Thru his influence and the desire of Scotch typefounders to improve on his work, the nineteenth century was inflicted with those stiff “Modern Romans” (Example 488) that in all their inhuman “perfection” yet abide with us and are still liked by Victorian persons of staid tastes. Many of our school children are being brought up on them, and in lighter form (Example 489) they are being fed to millions of Americans thru the homely typography of most newspapers. This double influence works for a lowering of typographic taste among printers and buyers of printing that for many years will keep the general standard of printing below what it should be.

To return to Bodoni. This Italian typefounder’s types, as a rule, were not so faultlessly finished as are the recent types that bear the name. There may be seen in Example 487 the alphabet of a type-face cut by Bodoni in 1790, together with an alphabet recently cut in America. The Bodoni type of the American foundry is a clever copy of the original, yet there is in the copy much of the modernization of the Scotch founders of the nineteenth century. Bodoni Book is a better-looking and more readable type-face.

EXAMPLE 491

(A) Optical changes caused by adding Vs at end of strokes. (B)
Appearance of vertical stroke altered by adding small horizontal
strokes, or serifs, and by joining both with curved line. (C) Cross
strokes or serifs added to E
]

EXAMPLE 495

(A) The diagonal stroke of the first Roman type-faces. (B) The lower
cross stroke or serif extended to right on the Roman type-face of
Jenson
]

The Scotch typefounders also attempted to improve the types of William Caslon, and the result of their labors is known as Modernized Oldstyle (Example 466-A). This kind of type was first made about 1852 for Miller & Richard of Edinburgh, Scotland. With Caslon Oldstyle unknown, Modernized Oldstyle would be a rather satisfactory letter, but with Caslon Oldstyle procurable there seems to be no need of the Scotch letter. There was brought out in 1884 by the MacKellar, Smiths & Jordan Company a decorative Modernized Oldstyle under the name of Ronaldson Oldstyle.

CHARACTERISTICS OF ROMAN TYPE-FACES

_The Serif._—The serif is that portion of a type-face indicated by the small black sections in Example 492. It has a powerful influence in determining the character and style of a face of type.

As an experiment, we will make three vertical lines exactly the same length (Example 491-A), leaving one as it was first drawn, but on the other two adding V-shaped marks, because of which the character of the line is changed and the length seemingly affected.

However, the serif, when properly used, has a decorative quality and is to type what the capital is to the column and the cornice to the building in architecture.

EXAMPLE 493

By altering the serifs, “old-style” type is changed into “modern.”
From Bullen
]

After further experimenting with three vertical lines that we will now identify as capital _I_s (Example 491-b), there can be no doubt that the serif is necessary as a finishing stroke. The serifless _I_ suggests a man in the public streets without hat or shoes. The crudest manner of adding a serif is to place a small horizontal stroke at the head and foot of the long vertical line, but the sharp angles thus formed can be softened and one stroke joined to another by a curve, with the corner filled in, as is shown. These two illustrations of the serif are merely for the purpose of introducing the novice to the subject. The fact that there are so many different faces of type is due somewhat to the manipulation of the serif—changes of length, thickness, direction and curve.

EXAMPLE 492

The black sections indicate the serifs, and show the differences in
serif construction on several well-known type-faces. From Bullen
]

It is by the serif that in most cases old-style types are distinguished from the so-called modern Romans. An interesting demonstration of this was made by Henry L. Bullen with Century Expanded, a modern Roman (Example 494). The first line shows the letter unchanged. As the serifs in most old-style faces project diagonally, he, in the second line, altered the modern serifs to conform to this idea. The third line shows Century Oldstyle as it actually appears in type.

There is, of course, more to a true old-style face than mere change of serifs, and in the actual cutting of the types other minor alterations were made, as will be seen by study of the third line.

EXAMPLE 494

By altering the serifs “modern” type is changed into “old-style.” From
Bullen
]

EXAMPLE 496

In most Roman alphabets all vertical strokes are thick, excepting in
M, N, U. Horizontal strokes are thin. Diagonal strokes running down
from left to right are thick and from right to left are thin,
excepting Z, z.
]

This process was reversed by Mr. Bullen, and in Example 493 certain serifs of Caslon Oldstyle are in the letters _i_, _n_, _k_, _p_, _h_ made horizontal, and in the letters _E_, _L_, _s_ made vertical. Modernized in this manner, Caslon Oldstyle resembles Scotch Roman (Example 484-C).

* * * * *

_Thick and Thin Strokes._—The inbred good taste and instinct for beauty possessed by the Roman calligrapher which directed him to add the serif to his lettering also impelled him to vary the width of the strokes (Example 464-A).

EXAMPLE 497

(A) No serifs or stroke contrast. (B) Serifs added. (C) With serifs
and thick and thin strokes
]

EXAMPLE 498

Vertical thick strokes were made with pen held straight
]

EXAMPLE 499

Diagonal thick strokes were made with pen held at a slant
]

The most uninteresting kind of lettering is that which is not only serifless but without contrast in the width of its strokes—lettering that is known to the printer as Block, or Gothic (Example 497-A). Dignity and something of good looks are given to it by adding serifs (Example 497-b), yet it is not eligible for admission to the select society of pure Roman type-faces until the various strokes show, properly proportioned, a difference in thickness (Example 497-C). The difference should be at least as two to one. It is about that in Cloister Oldstyle, but in Baskerville Roman (Example 496) it is as six to one. When the difference in thick and thin lines is exaggerated, as was done in the nineteenth century, the result is a caricature and type that is almost impossible to read, even in large sizes.

EXAMPLE 500

Heavy strokes in the lower left and upper right part of curves
]

EXAMPLE 501

Heavy strokes in vertical positions
]

Generally the distribution of thick and thin lines in Roman letters, as shown in Example 496, is as follows: All vertical strokes are thick (excepting in _M_, _N_, _U_). Horizontal strokes are thin. Diagonal strokes running down from left to right are thick, and from right to left are thin (excepting _Z_, _z_).

When this rule is reversed, as was actually done by a nineteenth-century typefounder in a weird attempt at novelty, there is begotten a monstrosity. Good typography requires that basic principles be adhered to.

In the recent recutting of the Roman type-face of Nicholas Jenson, the founders went all the way back to his type for their models, and did not content themselves, as they did near the end of the last century, with copying Morris’s type or the faces of other foundries. As Jenson based the design of his types closely on Italian manuscripts, we have in such a face as Cloister Oldstyle many of the characteristics of the pen-made letters, one of which is the position of the thick strokes on round letters. The thick strokes on the _O_, as an instance, instead of being vertical, are diagonal. This is due to the manner of holding the pen when lettering. When held to write vertically and horizontally, the pen makes lines as in Example 498; when held at a slant, the pen makes thick lines diagonally as in Example 499. The Caroline Minuscules, upon which the lower-case types of Jenson were based, were written with a slanted pen.

In Example 500 are to be seen two well-known type-faces on which the slanted-pen thick strokes are used. In Example 501 are two faces that could be made by writing vertically. It was Mr. Goudy who reintroduced into type forms a characteristic of Jenson’s Roman—that of having the lower serifs extend more to the right than to the left, as shown by Example 495-B.

* * * * *

_Ascenders and Descenders._—Not a little of the beauty of Roman type-faces lies in the ascending and descending strokes, and one reason why our types are becoming more beautiful is that we are going thru a period of restoration—the long-missing ascenders and descenders are being given back to us, after an absence of many years.

What should be the length of these strokes? Edward Johnston, in “Writing and Illuminating and Lettering,” illustrates this point (Example 503), and says: “The lines in massed writing are kept as close together as is compatible with legibility. The usual distance apart of the writing lines is about three times the hight of the letter _o_. The descending strokes of the upper line must clear the ascending strokes of the lower line.”

Bodoni, the Italian typefounder, had a system of measurement for the proportions of his lower-case letters: “Divide the body of the type into seven parts and let two at the top and two at the bottom be for the ascenders and descenders and the three in the middle for the other letters.”

The ascenders and descenders, as found in Cloister Oldstyle and in the original Caslon Oldstyle, approximate the proportions given by Bodoni.

Half the letters of the lower-case alphabet—thirteen—are of the hight of the _o_, and we will call them the small letters. The letters with ascending strokes number eight, with descending strokes five (Example 502). Because the descending letters are only five in number, they are most frequently picked out for mutilation. This sawing off of a portion of the descending strokes cripples the letters just as the sawing off of a portion of a man’s legs cripples his body. Not only is typography made imperfect by missing descenders, but, generally, the practice has worked toward the degeneration of type print, making it less easy to read, as the lines set too close together. When the descenders are of proper length they maintain a strip of blank space between lines, but when they are cut off the printer must add leads to recover the space—which he seldom does.

The printer and user of typography have been using false logic in favoring type-faces that have descenders cut off. They think they are getting more for their money when they obtain a normal twelve-point type-face on an eleven-point body, but as the one-point linear space must be restored by leading, there is nothing gained. Scientists, in the study of eye hygiene, have ascertained that a certain amount of space between lines is necessary to save eyestrain, and the descenders are essential in furnishing some of this needed space.

EXAMPLE 502

Of the twenty-six letters of the alphabet, eight have ascending
strokes, five have descending strokes, and thirteen have neither
]

The descending curves of the lower-case _g_ are a thing of beauty when allowed freedom (Example 504-A), but on the shortened plan they resemble a man cramped with rheumatism, in need of crutches.

* * * * *

_Proportion of Letters._—One of the elements in the beauty of the ancient Roman stone-carved alphabet is the proportion of its letters. Goudy, drawing lettering and designing type-faces modeled after the inscription capitals of the Romans, has had a large influence in reviving good proportion in type. Cloister Oldstyle (Example 467-A), designed by L. M. Benton after Jenson’s letter, is an excellent study of the old proportions introduced in present-day type-faces. However, there is no type that has the beauty of proportion to be found in the Roman capitals (Example 464-A) reproduced from an inscription on the Trajan column at Rome. The author has endeavored, in plain lines, to trace these capitals and show their proportions on a background of squares. It will be seen that the width of the letters _B_, _E_, _F_, _L_, _P_, _S_ is less than half their hight, and about half the width of _O_, _M_, _Q_, _C_, _G_ and _D_. It is this contrast in width as well as in shape that makes the alphabet as a whole so pleasing.