422 



IDENTIFICATION AND DESCRIPTION, PERSONAL. 



measurement of profiles seems to be the best 

 means of personal identification. Prof. Gallon 

 prefers for reference-lines B C (Fig. 3), touch- 

 ing the concavity above the nose and the con- 



FlG. 1. 



vexity of the chin, and a line parallel to this, 

 touching the tip of the nose. From these lines 

 various measures may be taken which are char- 

 acteristic of the individual. Instead of these 

 profile measurements, measures of the head 

 and limbs are generally employed in prisons 

 for purposes of identification, this idea originat- 

 ing in France with Alphonse Bertillon. But, 

 whatever the system, the practical 

 difficulty is to classify the sets of 

 measures that are so made, so that 

 it may be told at a glance whether 

 any given set of measurements 

 agrees with any or none of them 

 within specified limits, and for this 

 purpose Prof. Galton has devised 

 what he calls a mechanical selector. 

 It consists of a large number of strips of card 

 or metal, d, C a (Fig. 4), eight or nine inches 

 long, and having a common axis, A, passing 

 through all their smaller ends. A tilting-frame 

 T, turning on the same axis, has a front cross- 

 bar, F, on which the tips of the larger ends of 

 the cards rest when the machine is left alone, 

 the opposite end of the frame resting on the 

 base-board, S. When this end is raised, as in 

 the figure, the cards descend by their own 



FIG. 2. 



weight. Each card has notches cut in its 

 lower edge, whose distances from the axis rep- 

 resent the measurements in one of the sets. 

 The greater the number in the set the longer the 

 cards must be. In the figure the card has only 

 four notches. The given set of measures that 

 is to be compared with the sets already made 

 is represented by the positions of movable 

 wires, strung perpendicularly to the plane of 

 the figure. When the cards are released, by 

 raising the end of the tilting-frame, if the posi- 

 tions of all the notches in any card correspond 

 with those of the wires, that card will fall so 

 that a wire enters each notch ; but, otherwise, 

 the card will rest on one or more of the wires. 

 A glance thus enables the experimenter to de- 

 termine whether any sets of measurements 

 agree with the one to be tested, 

 and, if so, what sets so agree. In 

 the figure, the card C a so agrees, 

 and has therefore fallen lower 

 than Ci, which rests on the second 

 wire. By making the notches fit 

 the wires closely or loosely, the 

 limits within which the sets must 

 agree may be made small or large. 

 There is thus theoretically no limit 

 to the number of sets of measure- 

 ments that can be compared with 

 a given set by this machine, by a 

 single movement of the hand, and 

 in practice the only limit is the 

 necessity of making the machine 

 of convenient size. It seems a 

 valuable adjunct to the system of 

 personal identification in prisons. 

 Various markings on the hu- 

 man body remain unchanged for FIG. 3. 

 years, and so afford a basis for 

 identification, where the question is simply 

 whether two persons are or are not identical. 

 Those of them that admit of approximate 

 measurement by the method of least discerni- 

 ble differences, described above, can also be 

 used for the comparison of one person with a 

 thousand others. Among them are the mark- 

 ings on the iris (of which there are thousands 

 of varieties) the arrangement of the superficial 

 veins, the shape of the ear, and the furrows on 

 the hands and feet. The markings on the un- 

 der surface of .the finger-tips can be made more 

 plainly visible by rubbing on the finger a paste 

 of prepared chalk and water, which fills the 

 furrows. They may be made to leave a per- 



FIG. 4. 



