NOTES ON WOOD TYPE
posted: 12/29/15, type height section added 2/29/16
Sizes, leading, and spacing:Sizes: Wood type sizes are given in “lines”. “Line”, in wood type sizes, means “pica”. “12-line” wood type is twelve picas high. (Old catalogs may list sizes such as “12-line pica”. That was later shortened to “12-line”.) The size refers to the dimension from the base of a non-descending character to the top of a capital letter or ascender. (In wood type there is little or no shoulder above or below the printing surface of a full-height character.)
which you can cut out and stick onto a cabinet of wood type. They can be handier than pulling out a pica stick when you need to check sizes frequently.
There are two reasons why the standard block size is the smaller dimension required by most of the letters. At wood-type sizes, this greatly reduces the cost of type, since it allows more letters to be made out of a slab of end-grain wood. End-grain wood makes by far the best quailty wood type. But end-grain slabs are necessarily small, since they must be a cross-section of a trunk rather than a long board. No less importantly, smaller type greatly reduces the storage space required by wood type, which has always been a major issue for printers, since a complete font of large type, even if it has only a very few of each letter, can easily take up several cases—or a lot of shelf space, if it is stored standing up, as is common for the largest wood type.
This method of measuring type size, so unlike that used for metal type, confused people even back in the wood-type era. The 1888 specimen book of the Wm. Page Wood Type Co., one of the leading makers, admonishes printers not to send these long letters back under the impression that they do not belong to the font. (“The long or descending letters in the lower case are not kerned in wood type as they will split off. They are made solid, therefore Printers will please not send them back saying they do not belong to the font, as is often the case.”) Kerning would have allowed the descender to fit over a single continuous reglet in the line space. See the specimens on pages 1, 21, etc. of the Page catalogue.)
on this oddity of wood type sizes.
Another thing you may well run into is characters, typically short lower-case letters, that were cut short to allow another character, which was cut out on the side, to extend into the cut-away space at the top of the short letter. For example, a letter in a 12-line font might be cut down to 10 picas tall, but it will still belong with the 12-line font. When used without another letter filling the empty space, the space should be filled with reglet or lead to permit lockup without the line buckling or any types falling out.
Reglet was also probably the norm for word spaces and other in-line spacing. For display-sized type, think in terms of 4-to-em or 5-to-em spaces for word spaces; condensed type especially will need less word space. Thus, for fat or extended 12-line type, 3 picas of reglet (a fourth of an em) might be a good word space. For condensed type, 2 picas might well do—less than one fifth of an em. (Large type can take proportionately thinner word spacing than can text-size type.) You needn’t go out of your way to try to hit exactly one-fourth or one-fifth of an em. Consistency isn’t always as critical with wood type as it is with metal type in text sizes, and variations in physical spacing, making for good visual spacing, are more common and more necessary at large sizes.
In general, with wood type, spacing, leading, and other aspects of assembly are often much less cut-and-dried than with metal type. Improvisation is common.
For leading and spacing, use the pica-increment reglet—there’s a cabinet of it at MCBA with lengths of 8, 10, 11, 12, picas, etc., on up to about 50 picas or so. (Reglet in the other cabinets is in 5- or 10-pica increments up to 90 pica.) For smaller pieces, the spaces used with the larger sizes of metal type can be useful, as long as one side is one pica or multiple whole picas in width. Where cutting to size is needed, use 12-point lead slugs—don’t cut reglet, which is less easily and cheaply replaced.
Wood type height: higher than metal typeStandard height for U.S.-made wood type is 0.928", while metal type height is 0.918" in the U.S. and Britain (and other countries where practice was based on U.S./British usage). I don’t know why this is, but serious researchers may have more information, which I would be glad to know of. Standard metal type height is 0.928" in most of Europe (and other countries where practice was based on this usage). I suspect that it is not a coincidence that American wood type height is the same as European metal type height, though the industrial production of wood type was originated in the U.S., in the late 1820s.
The 0.928" wood type standard apparently goes back a long way in the U.S. (Again, others may have more information.) I have examined a font (Nesbitt Gothic Condensed Roman) that probably dates from between 1837 and 1851, and almost certainly from before 1863. I measured several letters at 0.927" high. Printing with the font showed that many of the rest, with a few exceptions, were substantially the same height, in spite of many surface irregularities. The presumed 0.001" discrepancy from 0.928" represents an impressively small amount of shrinkage and wear, considering that the font had obviously seen much heavy use, and was in use for a long time.
The difference in height, of course, presents a problem if wood type and (U.S. or British) metal type are to be used together. I don’t have much experience with this (yet), but one solution is to put backing under the metal type. (For easier handling, and to reduce the chance of the backing slipping out from under the metal type, the backing should probably extend under the surrounding furniture as well, as long as it can’t get under the wood type.) Dura-Lar (similar to Mylar) is dimensionally stable, and 0.005" sheets are available from art materials suppliers. I have used two such sheets to make up the 0.01" difference. I believe 0.01" sheets are available also, as well as sheets of thicknesses down to 0.001" that could be useful for finer makeready.
Font completeness:Wood type fonts found in shops and collections are often incomplete. Or they may be scant, with only one or two of many letters. Wood type is big, so storage is a limitation on font completeness. Often, small printers would only order the letters needed for a particular job. Only large firms that specialized in publicity printing were likely to invest money and space in numerous complete fonts with enough types for a variety of jobs.
One consequence of this scantiness of fonts was the rather frequent improvisation of extra letters. It’s not unusual to see a piece of wood type with a second letter carved into the bottom. In the wood type at MCBA, a number of fonts have additional letters made by cutting the letter shape out of a piece of leather, which is nailed, smooth side out, to a wood block. Some of these improvised letters are quite skillfully done, and probably indetectable in a printed job. (But on one improvised leather-faced letter at MCBA, someone forgot to reverse the image, and produced a type that is right-reading and will print wrong-reading.)
Note also that it was normal to use some characters interchangeably: zero and O, sometimes 1 (one) and l.c. l (ell). And some characters could be used interchangeably by flipping the character: d/p, b/q, and 6/9. In a pinch, you can even do this with n/u (requiring extra line space and spacing out with reglet, as for descenders, since the top of the n will extend below the baseline).
Many wood type fonts, especially the larger sizes, were not made with lower-case letters or figures.
If necessary, especially for punctuation, you could substitute characters from a similar font for a character lacking the the font you are using. But keep track of where you got each substitute character from, and put it back there when you’re done. You should be working from written copy, or at least writing down what you compose as you go, to allow for writing such notes, since improvisation is often needed in working with wood type.
(In the MCBA composing room, there may eventually be a case or three of miscellaneous characters, and perhaps one of punctuation, to use for substitution or other improvisation.)
Typeface names and designs:Wood type designs are generally either linear, bold, and simple, or more-or-less prominently ornamented, or pictorial, or downright bizarre. The particular subtle curves and modeling that make text faces readable and appealing in mass serve no purpose at large sizes. They are not often seen in large display faces of the wood-type era. Where they do appear at those sizes, they are apt to look odd. Finely-modeled display faces were designed in later periods, but the modeling is of a different kind.
The identification and nomenclature of wood type is not nearly as exact a matter as that of metal type. A face from one manufacturer was often closely similar to that of another manufacturer, and there was also a lot of design piracy going on, so one maker’s fonts may have been made directly from another’s, with a few letters varied for the sake of appearances. The name of the maker was not marked on most individual types, though it was often stamped on one of the sides of the capital A. (That is the only place I have looked for a stamp to record in the “Stamp” section of the inventory listings for the wood type at MCBA. There seems to be no regularity as to which side is stamped.)
Due to the size of the type (and a wish to avoid piracy of designs), specimens showing full alphabets are almost never available for comparison—much less full alphabets in all the sizes made, though letterforms might vary in detail from one size to another. (American Wood Type gives a number of full-alphabet specimens that were composed specially for that book.)
However, exact identification can be less important with wood type than with metal type. And there was a more-or-less consistent nomenclature used for the many common type designs of which versions were made by more than one manufacturer. Thus, for example, Hamilton’s French Clarendon Condensed was likely to be very, very closely similar to Page’s French Clarendon Condensed, so that it might be possible to use them interchangeably in a single job, and mixed in the same case. It would usually be impossible (and if possible, rarely worth the trouble) to identify the specific maker of an individual piece of wood type. Discrepancies between different cuts of the same design might not be noticeable when only a few letters appeared in the piece, and if noticeable, might not be faulted in a 19th-century publicity piece, or a 20th-century grocery-store flyer. Fine printing could be done with wood type, but that was not the rule in the wood-type era.
In the process of inventorying the wood type at MCBA, I have used the typeface names assigned by the manufacturer of the font, where that can be determined from stamps and from comparison with specimens in available specimen books. Otherwise, I have used the name that would be associated with that font in the standard nomenclature. Even in the wood-type era, more than this was often not possible or necessary. To someone experienced with wood type, and with the standard nomenclature, this name would be sufficient to express what the font looked like and enable it to be picked out from other, similar fonts.
The nomenclature for the common wood type designs was more or less consistent, but it was also profuse and arbitrary. The different families of wood type were distinguished by variations in every conceivable aspect of the letterform, notably weight, expansion, compression, the size and shape of serifs, the degree of modeling or contrast in the strokes, and various points, knobs, curlicues, and doodads attached to parts of the letter. Each such variation, or set of variations, could define a new family. (“Family”, here, is used in a somewhat looser sense than that in which it applies to the more coherent typeface families of the mid-twentieth century and later; these latter were, at least in origin, the productions of a single foundry, and would not show even the sort of minor variations that might be found between wood faces of the same family made by different manufacturers. Modern typeface families would also not show the extremely wide range of modifications, especially the shaded and ornamented modifications, that could be found in the wood types of the same family made by a range of manufacturers.)
The major families of wood type were referred to with terms like “Antique”, “Aldine”, “Clarendon”, or “Gothic”, and especially with an endless range of geographical terms like “Italian”, “Grecian”, “Egyptian”, etc. These names rarely had even the slightest connection with the look of the face, and could be in flagrant contradiction with common associations of the name.
Minor families tended to get names that were even more arbitrary, such as “Belgian” and “Madisonian” (the latter based on a supposed connection with the handwriting of Dolly Madison). I’ve never seen “Guamanian” or “Martian”, but it would be unwise to bet against them. There were wood type faces referred to as Bolivian, Athabascan, and Alaskan. There was also a Kurilian, which I presume refers to the Kuril Islands, some of the least-known, most remote and god-forsaken real estate north of Antarctica.)
Within each family, there were fairly consistent and intelligible names for the range of modifications, mainly Condensed, XX Condensed, XXX Condensed, Expanded, and names for various types of shading, outlining, 3-D effects, or ornamentation. This wide range of variations for each face was needed for posters and other typical uses of wood types. Condensed and expanded modifications allowed fitting single words or phrases on a line, without breaks, while preserving size distinctions called for by emphasis, sense, and the purpose of the announcement. Ornament modifications could be used to highlight the most important words.
From the 1870s, this system of nomenclature began to break down. In this period, the demand for novelty became greater with the rise of advertising (a thoroughly sleazy business in those days, before it learned that the cost of keeping up appearances could be added to the clients’ bills), and new technology enabled more people to to go into the wood type business, often with shoddy designs or faces pirated from the leading makers. These new firms often gave non-standard names to their faces, to reinforce claims to novelty and conceal the fact that the faces were pirated or derivative. For the same reasons, minor variations were treated and named as new families. The nomenclature became far more confusing and arbitrary than it had been, and many wood type makers gave up the names and just used numbers for their faces—which proved even less useful as references. (American Wood Type, p. 147.) But the unwieldiness was, in large part, the necessary concomitant of the profusion of designs demanded by the market for display type. No better nomenclature was ever arrived at, or is available today. (One might wish that the family names had not been so arbitrary. Not all of them were totally arbitrary when they were first used, though their rationales, often flimsy and obscure, were soon forgotten.) Kelly, in American Wood Type: 1828–1900, uses the old standard nomenclature of the early wood type era, augmented by explanation where needed. On page 229, he maintains “the feasibility of placing most nineteeth century wood type designs in one of the three primary families”. If he gave a list of those three, I missed it, but Antique and Gothic are probably on it. I wouldn’t be inclined to pare the list down quite that far.
Examples of the common nomenclature can be handily found on pages 86–91 of American Wood Type, of which there are two copies in the MCBA library.
Here are details on a few of the common family names, and a few of the less obvious modification names:
Aldine: Very like the Clarendons, but they seem to have thicker, more uniform strokes, broader brackets at the serifs, and to be generally less well-drawn and coherent, especially perhaps in the lighter weights. [[A development of the later 1800s], when the demand for novelty pushed type makers to make sloppy variants on existing designs and give them new names.] AWT 131.
Antique: Bold or extra-bold letterforms distinguished by slab serifs that are about the same thickness as the main strokes; the serifs are unbracketed (i.e., the serif and the main stroke meet at right angles, with no curve, angle, or other device merging serif and stroke).
Clarendon: Like Antique, but with bracketed serifs, with a broad, even curve where the serif joins the main stroke. (Nicolette Gray, Nineteenth Century Ornamented Types, p. 51.
Doric: Seems to be characterized by shallow ripples and points on the outsides of the letter, with bifurcated flares at the bottom ends of the vertical strokes. There is perhaps no clear distinction between the Dorics and the more restrained varieties of Tuscan.
Egyptian: Originally, both Egyptian and Antique were used for faces with unbracketed slab serifs. (Cf. Gray, 28.) Later, Antique came to refer mainly to fat faces, and Egyptian to faces with thinner strokes and serifs, more like those of text faces. (In the 20th century, descendants of the Egyptians were occasionally used as short-text faces.) (AWT, p. 130, is a discussion of an anomalous issue.)
French: A term referring to a particular modification that could be applied to faces of more than one family. It is used for letterforms with serifs that are abnormally thick in proportion to the main strokes, often giving the marked optical effect of two thick bars at top and bottom of each line of text. Most often associated with Clarendons and Antiques; there was also a French Octagon. Strangely enough, the design actually seems to have come from France. Faces with extreme contrast between hugely thick serifs and very thin main strokes were called “Belgian”.
Gothic: In wood type, “Gothic” has the familiar sense of “sans-serif”; in wood type, gothics are generally bold, unless they are highly condensed. (“Gothic”, with reference to wood type, has nothing to do with the earlier sense of “gothic” meaning blackletter.)
Grecian: Monoline slab serifs with beveled corners instead of curves.
Ionic: Perhaps not originally distinguished from Clarendon (Gray 51); later on, “Ionic” came to refer mainly to lighter faces which could be, and eventually were, used as text faces. In fact, they evolved so far that they came to be used as “legibility” faces in applications such as newspapers, where legibility—in inexpensive, high-speed printing on relatively coarse paper—was at a premium.
Italian: Sort of reverse-contrast letterforms, that are often bold where letters are normally thin, and thin where they are normally bold. Hard to look at, and understandably not numerous, but impossible not to notice where they occur.
Latin: Simple pointed wedges instead of serifs. As Gray, p. 61, notes, Latin “seems to have no normal width”.
Old Style: This name was usually given to wood-type version of the traditional text faces.
Ornamented: A term referring to modifications that could be applied to faces of more than one family. Denotes, roughly, any ornamentation except: 1) the relatively simple variations in the outline of the letter that typify the main families; 2) shading, outlining, or tooling; or 3) 3-D effects.
Roman: In wood type, refers specifically to letters typified by narrow, unbracketed serifs, and narrow thin strokes, contrasting with very thick main strokes. Short strokes might end in a fat wedge or flare. Essentially, they are fat Bodonis, and their metal and digital successors are often treated and named as bold modifications of Bodoni.
Tuscan: The family is distinguished by a wide and arbitrary range of variations and doodads on the serifs or stroke-ends. These might be applied to letterforms of one of the other families, resulting in names like “Antique Tuscan” and “Gothic Tuscan”. Tuscan could be thought of as a modification (like Condensed, or Outline), but it seems that the Tuscans tend to be thought of as an exceptionally large and messy family or superfamily. Gray gives a chronological study of the varieties of Tuscans in her Appendix I.
All content © 2016 by Ken Dezhnev. All Rights Reserved.
The Winged Hat logo is a trademark of Ken Dezhnev.
top of page