University of Virginia Library

Use of the Tables

Bibliographers may approach the following tables in either of two ways: (1) from their twenty-line measurements which may range from fifteen

through 1088 millimeters, or (2) from the name of a specific type size. In the latter case, the table conveniently provides the equivalent common foreign name for the type already known.

With Millimeter Measurements

Bibliographers in the former case will be undertaking original measurements directly from books and should heed the following advice. Twenty-line measurements should only be based on lines set solid. Unfortunately, it is not always possible to know whether there is leading between the lines. A convenient test for the presence of leading is to measure the interline spacing between a descender in the top line and an ascender in the next line down. A measurement between .5 and 1.0 millimeter usually suggests that the text is set solid.

Comparisons with other work may be complicated by the fact that bibliographers have not taken their measurements in a consistent manner. Some variation occurs because of ink squash—perhaps as much as one millimeter on the top and bottom of the sample. Thus, the use of a 6-10X power glass is recommended to determine the border of the type impression (Barker Interview). Additional variation can enter when one measures paper that is cockled or that has stretched or shrunk. In measurements taken for the Catalogue of Books Printed in the XVth Century Now in the British Museum (1908), for instance, A. W. Pollard notes up to a two per cent deviation ("Introduction," p. xx). Likewise, Philip Gaskell reports that "Paper shrinkage, which was more pronounced across the chain-lines than along them, generally reduced its dimensions by about 1 per cent, and occasionally by as much as 2½ per cent" (New Introduction to Bibliography [1972; repr. with corr. 1974], p. 13). These differences can be minimized by taking measurements on several different leaves.[4]

More importantly, Proctor measured "from the top of the short letters in line 1 to the bottom of the short letters in line 20, no account being taken of the tails on the longer letters."[5] As a consequence, his calculations are off by a fraction of an inch, as are measurements from the top of an ascender (e.g., h or l) in the first line to the bottom of a descender (e.g., g or y) in the twentieth line. The most accurate way is to begin any place in line one and measure to the same spot in line twenty-one. Any twenty-line measurement can be checked by measuring the type-page height and multiplying this figure by twenty and then dividing the product by the number of lines. This result should closely approximate the twenty-line measurement, although because the text type page measurement is usually taken to be the height from the text's highest ascender to its lowest descender, the result of the formula will not be precisely accurate.

With the Bruce System

Second, the measurement in inches will help bibliographers examining more recent books, especially American ones. Calculations in inches have been carried to four places to the right of the decimal point. Such precise

information will serve those dealing with books from the nineteenth century which have been set in type measured by the so-called Bruce System of Geometrical Progression, which is calculated to at least four decimal points (De Vinne, pp. 147-149). Although a rather curious system today, it was the rational approach of George Bruce's "based on the theory that bodies of type should increase by arithmetical progression—that small pica should be made as much larger than long primer as bourgeois was larger than brevier" (Updike, p. 33). His system, however, was not widely adopted outside of his own foundry.

With Point Systems

In other cases the bibliographer may only know the point size. Note, however, that typefoundries in different countries use different measurement systems for body sizes of type. In Anglo-American countries, the Pica Point System, originating in 1878, has been widely adopted since England led the way in 1898.[6] One Pica point is equal to .013888 of an inch. Earlier on the continent Simon P. Fournier (generally known as Fournier le jeune) advanced his "point" system in 1737 (see his Tables des Proportions qu'il faut observer entre les caracteres or Manuel Typographique, 1764); Updike reprints the 1742 table between pages 28 and 29 in volume one of his Printing Types. Many consider Fournier's system, in which a point equals .013728 of an inch, "the first successful attempt at a mathematical systematization of type-bodies."[7] The other system advocated by Françoise Ambroise Didot is widely employed on the Continent today. Based on the pied du roi—12 French or 12.7892 Anglo-American inches—one Didot point is equal to .01483 of an inch.