Page:Popular Science Monthly Volume 63.djvu/397

Rh The records employed are the printed impressions of the ten digits placed in a definite order upon a card, and the separate cards are placed on file by means of a classification wholly dependent upon the individual patterns.

Of those latter there are three types, the arch, the loop and the whorl, designated in descriptive formulæ by their initial letters. A, L and W (hence the name of 'the alw system' by which Galton has designated it). Of these the loop, which may turn to either the radial or the ulnar side of the hand, is for some purposes farther subdivided into radial and ulnar loops designated respectively as R and U. The patterns are definite in their nature; transitions between them are of rare occurrence and these are nearly always referable to one type or the other. This system of 'ALW,' with the occasional subdivisions of the R and U, forms what Galton designates the ' Primary Classification,' and the finger tip records of any number of individuals are arranged in accordance with a preconceived order. Although he has made several experiments in this, in his final method (1895) the sets are first arranged in four divisions (ARUW) in accordance with the type of pattern found upon the right index finger. This is followed by (small) letters designating the patterns upon the middle and ring fingers of the same hand, using l instead of r and u for all looped patterns. This will be seen to subdivide each of the first four divisions into 9 or will divide an entire set into 36, as follows:

If, now, the prints in each of these 36 divisions be farther subdivided in accordance with the same three fingers of the left hand, each subdivision consisting likewise of 36 compartments, the entire collection will be divided into 362 or 1,296. The thumb and little finger of the right hand which show 9 possible combinations, will subdivide each of the 1,296 compartments and, in like manner, the patterns of the thumb and little finger of the left hand will give another subdivision of 9, so that the number of possible compartments or subdivisions into which a set of prints may be arranged by means of this primary classification alone is 1,296 × 9 × 9 or 104,976; and since the various combinations with a few exceptions occur with about the same frequency, the number of separate prints in a collection of five