INSTRUMENT MARKING 



81 



dence from some studies of typography in- 

 dicates that certain condensed types, es- 

 pecially if bold, may be less legible than 

 normal, the problem of instrument numeral 

 proportions would seem to be worthy of 

 further special study. 



Berger studied digit proportions in one 

 of his experiments (7) but only after he had 

 arbitrarily chosen a width to height pro- 

 portion of 42:80 for his standard numeral, 

 the digit 8. He then determined what pro- 

 portions each of the other digits should have 

 in order to be equal in legibility to the 8, 

 Except for the digit 1, all figures required 

 proportions similar to that of the 8, namely 



TABLE I 



Berger's Data Showing Optimum Width and 

 Spacing of Digits Which Are 80 mm. High 



about 1:2. The exact values of the figure 

 width recommended by Berger are presented 

 in Table I. It is interesting to note that 

 these 1:2 proportioned digits of Berger de- 

 part from the 2 : 3 proportioned digits in the 

 U.S. Army-Navy Aeronautical Design Stand- 

 ard (see Fig. 1) and that the departure is 

 in the direction suggested by the foregoing 

 quotation from Loucks. 



Digit Spacing. The data which Berger 

 collected on digit spacing (7) are also pre- 

 sented in Table I. The spacings given in the 

 table are additive for any two adjacent fig- 

 ures: e.g., in the number 10, a space of 11.1 

 mm. should be allowed to the right of the 1 

 and another 12.3 mm. to the left of the 0, 



making an over all space 23.4 mm. between 

 the two digits. These spacing values were 

 determined principally by finding how far 

 each digit had to be separated from zero 

 in order for the combination to be legible. 



In view of the repeated reference made 

 here to Berger's results, it is necessary to 

 remark that they represent only one ap- 

 proach to the digit legibility problem and 

 that Berger himself warns against adopting 

 his results unquestioningly for application 

 in the design of displays other than license 

 plates. Since all of his data were based upon 

 the reading of distant stimulus cards, com- 

 parable investigations to check his results 

 at normal reading distances are needed. 



'v\ 



FOOT LAMBERTS 



Fig. 3. Number-reading time as a function of 

 brightness. Each curve is for a different size of 

 number, being given in visual angle. Data from 

 Craik (19) for two subjects. 



Digit Size. The most pertinent data on 

 the subject of instrument numeral size are 

 those of Craik (19). Working with only 

 two subjects, Craik determined reading time 

 for numbers of four different sizes as a func- 

 tion of dial illumination. His number sizes 

 are given as angular size. The plot of aver- 

 age reading time vs. brightness in foot 1am- 

 berts is reproduced in Fig. 3. Craik would 

 suggest the use of this chart, or similar ones 

 based on more adequate data, for selecting 

 instrument numeral size once illumination 

 level and observation distance have been 

 fixed for a given operating situation. 



Instrument Labeling 



Two principles have been applied in sup- 

 port of the suggestion that instrument names 



