June 21, 1888] 



NA TURE 



is simply by sliding the rings to which they are attached 

 at either end, along the rod which passes through them. 

 It is easy to arrange a more delicate method of effecting 

 this if desired. Hitherto I have snipped out the notches 

 in the cards with a cutter made on the same principle 

 as that used by railway guards in marking the tickets 

 of travellers. The width of the notch is greater than 

 the width of the wire by an amount proportionate to the 

 allowance intended to be made for error of measurement, 

 and also for that due to mechanical misfit. There is 

 room for 500 cards or metal strips to be arranged in 

 sufficiently loose order within the width of 30 inches, 

 and a key of that effective length would test all these by 

 a single movement. It could also be applied in quick 

 succession to any number of other collections of 500 

 in each. 



Measurement of Profiles. — The sharp outline of a 

 photograph in profile admits of more easy and precise 

 measurement than the yielding outline of the face 

 itself. The measurable differences between the profiles 

 of different persons are small, but they are much more 

 numerous than might have been expected, and they are 

 more independent of one another than those of the 

 limbs. I suspect that measures of the profile may 

 be nearly as trustworthy as those of the limbs for ap- 

 proximate identification — that is, for excluding a very large 

 proportion of persons from the possibility of being mis- 

 taken for the one whose measurements are given. The 

 measurement of a profile enables us to use a me- 

 chanical selector for finding those in a large standard 

 collection to which they nearly correspond. From the 

 selection thus made the eye could easily make a further 

 selection of those that suited best in other respects. A 

 mechanical selector also enables us to quickly build up 

 a standard collection step by step, by telling us whether 

 or no each fresh set of measures falls within the limits 

 of any of those already collected. If it does, we know 

 that it is already provided for ; if not, a new card must be 

 added to the collection. There will be no fear of duplica- 

 tions, as every freshly-added standard will differ from all 

 its predecessors by more than the specified range of 

 permitted differences. After numerous trials of different 

 methods for comparing portraits successively by the eye, 

 I have found none so handy and generally efficient as a 

 double-image prism, which I largely used in my earlier 

 attempts in making composite portraits. As regards the 

 roost convenient measurements to be applied to a profile 

 or use with the selector, I am unable as yet to speak 

 lecidedly. If we are dealing merely with a black 

 silhouette, such as the shadow cast on a wall by a small 

 >r brilliant light, the best line from which to measure 

 ftems to be B c in Fig. 8 ; namely, that which touches 

 both the concavity of the notch between the brow and 

 nose, and the convexity of the chin. I have taken a con- 

 siderable number of measures from the line that touches 

 the brow and chin, but am now inclined to prefer the 

 former line. A sharp unit of measurement is given by 

 the distance between the above line and another drawn 

 parallel to it just touching the nose, as at N in the figure. 

 A small uncertainty in the direction of p c has but a very 

 trifling effect on this distance. By dividing the interval 

 >etween these parallel lines into four parts, and drawing 

 a line through the third of the divisions, parallel to B c, 

 we obtain the two important points of reference, M and R. 

 M is a particularly well-defined point, from which o is 

 determined by dropping a perpendicular from M upon B c. 

 seems the best of all points from which to measure. It is 

 xcellently placed for defining the shape and position of the 

 notch between the nose and the upper lip, which is perhaps 

 r .he most distinctive feature in the profile. O L can be deter- 

 mined with some precision ; o B and O C are but coarse 

 neasurements. In addition to these and other obvious 

 neasures, such as one or more to define the projection of 

 he lips, it would be well to measure the radius of the circle of 



curvature of the depression at B, also of that between the 

 nose and the lip, for they are both very variable and very 

 distinctive. So is the general slope of the base of the nose. 

 The difficulty lies not in selecting a few measures that will 

 go far towards negatively identifying a face, but in selecting 

 the best— namely, those that can be most precisely deter- 

 mined, are most independent of each other, most variable, 

 and most expressive of the general form of the profile. I 

 have tried many different sets, and found all to be more or 

 less efficient, but have not yet decided to my own satis- 

 faction which to adopt. 



A closer definition of a profile or other curve, can be 

 based upon the standard to which it is referred. Short 

 cross-lines may be drawn at critical positions between 

 the two outlines of the standard, and be each divided 

 into eight equal parts. The intersection of the cross- 

 lines with the outer border would always count as o, that 

 with the inner border as 8, and the intermediate divisions 

 would count from 1 to 7. As the cross-lines are very 

 short, a single numeral would thus define the position of 

 a point in any one of them, with perhaps as much pre- 

 cision as the naked eye could utilize. By employing as 



N M 



Fig. 8. 



many figures as there are cross-lines in the standard, each 

 successive figure for each successive cross-line, a corre- 

 sponding number of points in the profile would be accu- 

 rately fixed. Suppose a total of nine figures to be given, 

 together with a standard collection of under a thousand 

 doubly outlined portraits, each with six cross-lines. The 

 first three figures would specify the catalogue number of 

 the portrait to be referred to, -and the remaining six 

 figures would determine with much accuracy, six points 

 in the outline of the portrait that it is desired to describe. 



I have not succeeded in contriving an instrument that 

 shall directly compare a given profile with those in a 

 standard collection, and which shall at the same time act 

 with anything like the simplicity of the above, and with 

 the same quick decision in acceptance or rejection. 

 Still, I recognize some waste of opportunity in not 

 utilizing the power of varying the depths of the notches 

 in the cards, independently of their longitudinal position. 



I shall have next to speak of other data that may 

 serve for personal identification, and especially on the 

 marks left by blackened finger-tips upon paper. 



(To be continued.) 



SOAP-BUBBLES. 



SOAP-BUBBLES fill the same happy position as do 

 those charming books in which Lewis Carroll de- 

 scribes the adventures of Alice, in that they serve equally 

 to delight the young and to attract the old. Clerk- 

 Maxwell has mentioned the fact that on an Etruscan 

 vase in the Louvre are seen the figures of children 

 amusing themselves with bubbles, while to-day the same 

 subject is being forced on the attention of the world 



