6.1 8 



NATURE 



[October 17, 1907 



strument is another, and am inclined on the whole to 

 prefer the method of coordinates.' 



In my experiments I have chiefly used the sido-view 

 portraits by George Vance, R.A., of his distinguished 

 contemporaries, published in 1809 (2 vols., folio, 

 Lonifmans), which yielded si.vty-eig-ht pure profiles of 

 about one-third the natural size. I lexiconised these 

 in respect to the measures (entered to the nearest 

 cent) of the two coordinates of N and M respectively 

 (4 measures in all), and found, first, that no two of 

 ihe numerical formulae were the same, and, secondly, 

 that in two-thirds of them the smallest difference 

 between the most nearly resembling- pairs was 3 cents 

 in one or more of the four measures. This conspic- 

 uous, difference, equivalent to between i/6th and i/7th 

 of an inch in a portrait of the natural size, could never 

 be due to the inherent imperfection of the art of 

 measurement, but to some gross blunder. It follows 

 that the collection of sixty-eight portraits was lexicon- 

 ised with remarkable precision. The data were in- 

 sufficient to enable me to speak with much assurance 

 of the gain that would accrue from taking L and U 

 into additional account, but their correlations with C, 

 M, N, and F, seeming to be very small, the gain ought 

 10 be great. I am content to underrate this gain con- 

 siderably, and to allow only fifteen-fold for it. On- that 

 basis a collection of looo profiles from brow to chin 

 could be lexiconised and searched with great ease. In 

 667 cases each portrait would have a clearly distinc- 

 tive formula ; in the remaining 333 there would be 

 doubtful duplicates, and even triplicates, just as in 

 anv list of the names of 1000 British persons there 

 would be more than one Smith. 



In the report of a committee appointed by the 

 .Secretary of State in 18(54 'C. — 7263, price lorf.) to 

 inquire into the best means available for identifying 

 Iiabitual criminals, the following remark appears on 

 |i. 18 : — " An enormous amount of time is spent in 

 examining the books of photographs. It will be seen 

 from the figures furnished by Chief Inspector Neave 

 that on March i last twenty-one officers searched for 

 twenty-seven prisoners — the total time spent being 575 

 hours — and made seven identifications. This was an 

 average of more than two hours for each prisoner 

 sought for, and more than eight hours for each 

 identification." A similar search in a lexicon of por- 

 traits of the same size would occupy apparently fewer 

 minutes than the above occupied hours. 

 - I will go no further now into the results of my 

 experiments than to say that I have applied the above 

 method to portraits of persons of very different races, 

 and have thus far found it efficient in all of them. 

 Fr.'vncis Galton. 



WEIGHTS AND MEASURES REGULATIONS. 



THE new regulations, which came into force on 

 October i, apply only to weights, measures, and 

 weighing instruments used for the purposes of trade. 

 They are in some respects rather less stringent than 

 the preliminary draft issued bv the Board of Trade 

 in -August, ic)o6, a review of which appeared in these 

 columns last year. 



There are but few points of scientific interest in the 



^ In some cases brevity is very desirable, and may be obtained by regard- 

 log only the limits within which the variability of each link most commonly 

 occurs and by dividing the interval between those limits into 8 equal parts. 

 Then o would signify all measures helow the lower limit, and g .ill above 

 the upper one. ' The range of the rectangular coordinates to N and M 

 wiihin the limits above explained varies between 12 and, 20 cents, so the 

 value of each of the eight equal parts will vary front i A to 2^ cents according 

 to the courdit>ate in question. These " tKirts '- are more suitable for classi- 

 fication than cents, which are too small to be quite trustworthy. But I 

 will not go further here into this question than to add that the 8 rectangular 

 coordmates of ISI, N, I. and U can he described in this way by only 8 

 figures, and the conneaing outlines CL, LU, UM, MN, NF by 5 (or say 

 10) more, so that a portrait can be expressed (say be telegraphed) in a rude 

 but recognisal)le form by only 13 (or 18) figures. 



NO. I 98 I, VOL. 76] 



regulations. Specific instructions as to temperature 

 are now given for the first time to inspectors of 

 weights and measures. Measures of length are to be 

 verified by comparison with a local standard at or 

 near the normal temperature, which is 62° F. for im- 

 perial measures. For imperial measures of capacity' 

 the standard temperature of their water contents is 

 also 62° F. ; for metric measures of capacity it is 

 4° C. ; but metric measures graduated at, 15° C, or 

 60° F. may also be verified for chemical or pharma- 

 ceutical purposes, or for volumetric estimations. 

 Measures marked with the temperature at which thev 

 are graduated must be tested against measures 

 standardised at the same temperature. An apothe- 

 caries' measure, marked with equivalents in weight, 

 is permitted, provided that the words " of water " 

 are marked on it in addition to the denomination. 



Certain restrictions are placed upon the weighing 

 instruments to be used by dealers in precious metals 

 or precious stones, retail chemists or druggists, and 

 silk merchants. These traders are permitted to use 

 three kinds of weighing instruments, the first kind 

 being chemical and as>ay balances provided with means 

 for relieving all the bearings and knife edges, the 

 second being beam scalesof a lower order of sensitive- 

 ness' and accuracy, which must be marked " Class 

 B," and the third being instruments other than beam 

 scales which satisfy the requirements for Class B. 

 The first and third of these types of instruments are 

 not to be marked with a "class." 



The requirements of Nos. 69 and 78 of the regu- 

 lations, prohibiting adjusting contrivances which are 

 not permanently fixed to the weighing instrument, 

 will render some kinds of analytical and assay balances 

 ineligible for oflicial stamping in the future. Traders 

 who use such balances should be careful not to keep 

 them upon their trade premises if unstamped, otherwise 

 they may be liable to forfeiture if they come under 

 the notice of an inspector of weights and measures. 



Counter weighing machines of the " accelerating " 

 type are prohibited by the regulations. It is often 

 difficult to distinguish between a vibrating and an 

 accelerating instrument, especially when these are 

 sluggish or have been in use for some time. The re- 

 quirement prescribed, in No. 68 of the regulations 

 that the machine shall balance when unloaded will, 

 however, be sufficient in general to determine this 

 point. When the machine is unloaded, if either of the 

 pans be pressed down and then released, the beam 

 will be set in oscillation about its horizontal balancing 

 position, if the instrument is of the vibrating type ; 

 but if the instrument is accelerating, it will be found 

 that one or other of the pans when pressed clown will 

 remain down, and the beam, failing to oscillate, will' 

 rest out of balance. 



Spring balances are somewhat rigorously dealt with, 

 but, on the whole, the Board of Trade has taken a' 

 fairly lenient view with respect to these instruments 

 in permitting them to be used for ordinary trade, rather 

 than restricting their use to such purposes as the 

 weighing of postal parcels and passengers' luggage, 

 as is the case on the Continent. 



Weights, measures, and instruments at present 

 stamped and in use, but which do not comply with the 

 new regulations, may be continued in use for certain 

 prescribed periods, and, with the exception of a parti- 

 cular type of spring balance, may be re-stamped from 

 time to time. 



.\ppendix 2 cont.ains a useful list of all the de- 

 nominations of weights and measures which are at 

 present legal in this country. It will be seen from this 

 list that weights of i .grain, which are frequently em- 

 ployed bv chemists in coinpounding drugs, are not 

 legal for use in trade. 



