Apart from the way line width affects the application of justification, as mentioned earlier, another aspect of typography is to be considered when deciding upon line width, namely typesize. The more points per character, the less characters will fit in a line of a certain width. It is fundamental that a ratio of typesize to line width is chosen which results in a balanced appearance. When the properties of typesize and line width have to be applied to justified text, a badly chosen ratio would result in stretched words separated by large gaps.

In early research, line width was expressed in millimeters. Tinker and Paterson (1929), for example, used the Chapman-Cook Speed of Reading Test on 560 subjects and found 80 mm to be the optimal width for 10-point type in order to achieve the highest reading pace. Luckiesh and Moss (1941), who studied the effect of various line widths on readability by measuring rate of involuntary blinking of the eye, patterns of eye- movement and speed of reading, expressed their results of optimal line width in picas. The confusion that might result from using these different measures is avoided by expressing line width as the average number of words or characters that should fill one line. For the conventional typesizes (9 to 12 points) it has been concluded that, on the average, the optimal line width lies between 10 to 12 words per line. This amounts to a range of between 50 and 70 characters per line of text, also counting word spaces, to achieve the highest readability (Matis, 1996). It should be noted, however, that the aspect of typesize which is relevant for line width is character width, not height (Hartley, 1978).