Personal Identification and Description 201 



In other words, between a fifth and a quarter of the sets fall into groups 

 which are far too unwieldy for rapid index searching. It is clear that the 

 loops and whorls are the chief source of this trouble (see our pp. 149, 165 and 

 173) and Galton proceeds to break them up by what he terras a Secondary 

 Classification, or a system of adding subscripts to the letters of his primary 

 classification. The subscripts or suffixes as Galton calls them are very 

 numerous, although some can only be attached to certain patterns. For 

 example, what would have appeared in his old (his present primary) classi- 

 fication as 



oiviv, oil, vow, 11, 

 now becomes 



uw v w, ul~\l vy , w lvy w, ll v , 



where subscript y means that the core of the corresponding whorl is pear- or 

 racquet-shaped ; f denotes that there was a scar on the middle finger of the 

 left hand; l vy denotes a loop with invasion of ridges from the side and with 

 a racquet core ; w lvv means a whorl which might be mistaken for a loop, has 

 an invasion of ridges from the side and a racquet core, and l v denotes a loop 

 with a like invasion only. Thus 18 symbols are used to index the set. Galton 

 defines and discusses 28 letters and symbols which may be used as suffixes. 

 Obviously the above system of subscripts is one liable to error either in writing 

 or printing, and Galton, although he suggests its use, does not actually adopt 

 it in the Directory he publishes of 300 sets of prints of the 10 digits. Here 

 he gives the primary classification symbols on the left of his page, and then 

 on the right in 10 columns the suffixes to be attached to each of these 

 symbols. For example, the above formula appears as 



| Uivw \ull\9,5\\ — ,y, — | — , f , vy \ Ivy, — | — , v\, 



where the last 10 columns correspond to the digits in order of the primary 

 formula (9 = ww, 5 = 11, the thumb and little finger formulae of right and left 

 hands : see our p. 198). 



Besides the 28 symbols which are chiefly devoted to breaking up the 

 large loop and whorl groups, Galton introduces for the troublesome all- 

 loops group the counting of the ridges on the forefingers. This counting he 

 now does in a different manner from that of his earlier papers, and one which 

 seems less liable to misinterpretation. He first determines a better line for 

 counting the ridges on (see his pp. 78-80) than he had previously selected (see 

 our pp. 163 and 165). The following are his rules (see Fig. 34, p. 202): 



"The terminus from which the count begins is reckoned as 0; it proceeds thence up to, and 

 including, the other terminus. 



"The inner terminus lies at the top of the core of the loop, the outer terminus at the delta, 

 but it is necessary to define their positions more exactly, as follows : 



"Inner terminus. There are two cases : 



"(a) The core of the loop may consist of an uneven number of ridges, as in each of the two 

 figures, a 1 and a 2 ; then the top of the central ridge is the inner terminus*. 



* I think there is a risk of confusion here to which Galton does not refer. The ridge or 

 ridges within the "staple" may or may not meet the latter. In Figs, a 1 and a 3 the inner 

 ridges are made to meet the staple, and the inner terminus is not put at the top of the 

 pa in 26 



