2.1.1 Width and Relative Dimension of Character
The term character width is used to indicate the width of a letter. It is the distance between the most leftward part and the most rightward part of the letter, excluding any attached space. Expressing width of a character in points, however, is not very useful. Comparing different typefaces and typesizes is only possible when character width is expressed in its relation to character height. Whereas the term typesize refers to character height only , character width just as much determines the reader’s perception of how large a letter is. Consequently, the most useful measurement is the width -to- height ratio.

Width -to- height ratio has mainly been studied for applications other than books; such as signs, which are limited in space and where odd shapes and dimensions of letters are used to draw attention. Width seems not to be considered an issue with regard to books. After all, the commonly used ratio in many book-typefaces seems to meet the criterion of legibility quite well. For signs and related applications, however, conspicuity is often considered at least as important as legibility. Do note that this priority needs to be decided per situation; the factors that have to be considered in the design of a company-name displayed on top of a building differ from those required in the design of a brochure.

A width to height ratio of 3:5 (0.6) is recommended for most applications. This choice, supported by research, is based on the fact that five is the maximum number of elements in the height of letters and three is the maximum number of elements in width. As shown in figure 6, the letter ‘E’ has three strokes separated by two spaces protruding from its stem; this corresponds with a total of five elements (grid-boxes) in height. The letter ‘T’ has a stem and a topstroke that protrudes on two sides to result in three elements in width. In research on numerals, Soar (1955) concluded that a ratio of 7.5 : 10 (0.75) may result in even greater conspicuity, but this is not necessarily the case for every single letter or digit; an ‘8’ may be optimally presented with this ratio, whereas the optimal ‘7’ requires another ratio. Likewise, one would think that individual letters of the alphabet differ with regard to the optimal ratio as well. This assumption can be related to Heglin’s (1973) propositions in his design guide for military naval use. He argues that keeping basic geometric forms intact would result in higher legibility; this means using a circular ‘o’ and an equilateral triangle for a ‘v’ rather than a cylindrical ‘o’ and a ‘v’ with a very sharp angle between its two strokes. Special applications, such as translucent letters, require a ratio of 1:1 (Heglin, 1973).


Figure 6:
The Ratio of width to height and stroke-width to height (after Sanders and McCormick, 1993)
s = stroke widthw = character widthh = character height
in this case:
w : h = 3 : 5(0.6) s : h = 1 : 5(0.2)

Many typeface families have several fonts which vary in width. Univers, for instance, has several fonts which differ in width only (Univers 63, 65, 67), as illustrated in figure 7. This does not mean that the ‘i’ of the 63-font is wider than the ‘m’ of the 65-font; one font is just a condensed version of the other.


Figure 7:
The Univers Family, designed systematically by Adrian Frutiger in 1954. Reading from top to bottom, an increase in stroke width (boldness) is seen. From left to right, character width decreases. Even numbers represent italic versions, while uneven numbers represent regular family members. The red boxes contain those versions of Univers discussed in the text.




to preceding paragraph back to top to next paragraph