Typeface anatomy describes the graphic elements that make up letters in a typeface.[1][2]
Typefaces are born from the struggle between rules and results. Squeezing a square about 1% helps it look more like a square; to appear the same height as a square, a circle must be measurably taller. The two strokes in an X aren't the same thickness, nor are their parallel edges actually parallel; the vertical stems of a lowercase alphabet are thinner than those of its capitals; the ascender on a d isn't the same length as the descender on a p, and so on. For the rational mind, type design can be a maddening game of drawing things differently in order to make them appear the same.
The strokes are the components of a letterform.[4] Strokes may be straight, as in k l v w x z, or curved, as in c o s. If straight, they may be horizontal, vertical, or diagonal; if curved, open or closed. Typographers also speak of an instroke, where one starts writing the letter, as at the top of a c f, and an outstroke, where the pen leaves off, as at the bottom of c e j k t y.[5]
A main vertical stroke is a stem. The letter m has three, the left, middle, and right stems. The central stroke of an s is known as the spine.[6] When the stroke is part of a lowercase[4] and rises above the height of an x (the x height), it is known as an ascender.[7] Letters with ascenders are b d f h k l. A stroke which drops below the baseline is a descender.[7] Letters with descenders are g j p q y.
An arching stroke is a shoulder as in the top of an R or sometimes just an arch, as in h n m.[4] A closed curved stroke is a bowl in b d o p q D O P Q; B has two bowls. A bowl with a flat end as in D P is a lobe.[8] A trailing outstroke, as in j y J Q R is a tail. The inferior diagonal stroke in K is a leg.[9] The bottom of the two-story g is a loop; the very short stroke at the top is the ear.[10] The letters i j each have a dot or tittle.[10]
A short horizontal stroke, as in the center of e f and the middle stroke of E F, is a bar. Strokes that connect, as in A and H, or cross other strokes, as in t, are also known as crossbars.[9] A longer horizontal stroke at the top or bottom, as in E T, is an arm.[4] The junction of two strokes intersecting above as in A M X x is an apex and the joining of two strokes intersecting below as in V W v w is a vertex.[10]
The font shown in the example is stressed; this means that strokes have varying widths. In this example, the stroke at the top of the "g" is thinner at the top and bottom than on the sides – a vertical stress. Fonts without any variation in the stroke width are known as monoline fonts.
The terminal (end) of an instroke or outstroke is often a serif or a stroke ending. A seriffed terminal may be described as a wedge, bulbous, teardrop, slab, etc., depending on the design of the type. Typefaces may be classified by their look, of which the weight and serif style – whether serif or sans-serif – are key features.[9] Some designs also have spurs, which are smaller than serifs and appear on angles rather than at a terminal, as on e or G.
Areas of negative space (white space) formed by straight or curved strokes are called counters. Closed counters are found in a b d e g o p q A B D O P Q R, and open counters in a c e f h m n r s t u. The closed counter in e is also named an eye.[9] The term storey refers to stacked counters: When the letter a appears with two counters it is referred to as double-storey a, and when the letter g appears with two closed counters it is known as double-storey g.[11]
Angles of white space, as in W w, are corners (w has three corners); the term is not used for angles of strokes. The small corner formed by a serif, whether curved or angular, is called the serif bracket.
Inter-letter space can be reduced with kerning. A kern is the part of a letter that intrudes into the "box" of an adjacent glyph.
A subtle change in proportion impacts weight, perception, measure, and legibility. The letterform height compared to its stroke width modifies the aspect ratio; a slight change in weight sometimes helps to create emphasis. The disparity between thick and thin strokes, known as stress, alters optical perception. As an example, the first sans serif typefaces used strokes of constant thickness, but subsequent technological advances permit drawing thinner strokes. Condensed type occupies less space than expanded type, so that a whole page of text can be reduced to half a page. The capline and x-height ratio improve or decrease word legibility.[4]
During the late metal type period, many fonts (particularly in American typefounding) were issued to "common line".[12] This meant that they were made to standardised proportions, so that fonts of different typefaces could be mixed with no difficulty. This made it possible to mix typefaces from completely different genres such as sans-serifs and serifs and have the cap height, baseline and linespacing match perfectly, something not possible with most digital fonts.[12] It even allowed mixing of different sizes of type with a consistent baseline. It however had the disadvantage of often forcing typefaces to be issued with cropped descenders compared to historical typefaces, to allow tight linespacing. A "script line" or "art line" was used for more delicate fonts with long descenders. Titling capitals, meanwhile, were issued taking up the whole space of the metal type area, with no room for descenders.[13][14]
individual parts such as the spine of the S