p QC L r George Da via son 1825-1911 STANDARD MEASURES OF UNITED STATES, GREAT BRITAIN, AND FRANCE. HISTORY AND ACTUAL COMPARISONS. WITH APPENDIX OX INTRODUCTION OF THE MJ&TKE. BY ARTHUR S. C. /WURTELE, ASS'T ENG., N. Y. Or^H? R. R. E. & P. N. SPON, NEW YORK; 44 MURRAY STREET. LONDON: 16 CHARING CROSS. 1882. STANDARD MEASURES OF UNITED STATES, GREAT BRITAIN, AND FRANCE. HISTORY AND ACTUAL COMPAEISONS. WITH APPENDIX ON INTRODUCTION OF THE M^TRE. BY ARTHUR S. C. WURTELE, ASS'T ENG., N. Y. C. & H. R. B. E. & F. N. SPON, NEW YORK; 44 MURRAY STREET. LONDON: 16 CHARING CROSS. 1882. COPYRIGHT, 1882, BY ARTHUR S. C. WURTELE. INTRODUCTION. DURING the preparation of this investigation of Stand- ard Measures a large number of authorities were ex- amined, including the following : Kelly's " Universal Cambist," Maunder's " Weights and Measures," " Encyclo- paedia Britannica," " Chambers' Encyclopaedia," Williams' " Geodesy," Hymer's works, " Smithsonian Reports," " Coast Survey Reports," Herschel's " Astronomy," etc. The only concise and clear statement I found was J. E. Hilgard's report to the Coast Survey on standards in 1876, which I was gratified to find coincides with my deductions. ARTHUR S. C. WURTELE. ALBANY, November 26, 1881. 20779 STANDARD MEASURES, A STANDARD measure of length at first sight appears to be very simple merely a bar of metal of any length, ac- cording to the unit of any country ; and comparisons of different standards do not seem to present any difficulty. But on looking further into the thing, we find that stand- ards are referred to some natural invariable length, and we are at once confronted with a mass of scientific reduc- tions giving different values to the same thing, according to successively improved means of observation. We find, also, that comparisons of one standard with another differ, as given by reductions carried to great apparent exactness. 1 Every author appears to assume the right of using his own judgment as to what reduction is to be considered the most exact, and the result is a very confusing differ- ence in apparently exact figures, with nothing to show how these differences arise. I have endeavored to indicate what may be the cause of this confusion by giving the figures of actually ob- served comparisons and reductions; in a manner, the roots of the figures used as statements of length. Sir Joseph Whitworth gives ^-QU- of an inch as the smallest length that can be measured with certainty, with an ultimate possibility of YOOOOOO of an inch ; but imper- ceptible variations of temperature affect these infmitesi- rnal lengths to such an extent that he believes the limit can only be reached at a standard temperature of 85 F. to avoid the effect of heat of the body. It appears to me that comparisons should be made of double yards and metres with the old French toise, as the limit of exactness would be thereby doubled. Another great defect in statements of relative values is the omission of necessary facts the material of which the bars or standards are made, the temperature at which comparison was made, and the standard temperatures used as to the final reduction, with the coefficient of expansion adopted. Again, bars of different metals appear in time to sensi- bly change their relative length. ENGLISH STANDARDS OF LENGTH. The first establishment of a uniform standard appears to have been made in 1101 by Henry I, who is said to have fixed the ulna (now the yard) at the length of his arm; but nothing definite was done till 1736, when the Royal Society took steps toward securing a general stand- ard, and in 1742 they had a standard yard made by Graham from a comparison of various yards and ells of Henry VII. and Elizabeth, that were kept in the Ex- chequer. Two copies of the Eoyal Society standard yard were made by Bird in 1758 for a committee of Parliament, one of which was marked "standard of 1758," and the other 1760. But no exact legal standard was yet established, as shown by comparisons in 1802 of the various standard measures in use which Pictet, of Geneva, made with an accurate scale by Troughton, using means exact to the ten thousandth part of an inch, with the following results at the temperature of 62 F. : Troughton Scale 36.00000 inches. Parliamentary Standard (1758, Bird) 36.00023 " Royal Society (1760, " ) 35.99955 " (Graham) 36.00130 " Exchequer " , 35.99330 " Tower " 36.00400 " Gen. Roy " (Trig. Survey) .. .36.00036 " Parliament finally undertook to reform the measures of England, and appointed a commission in 1818, under whose authority Capt. Kater compared the standard yards then in use with the following results, as referred to the Indian Survey standard : Col. Lambton Standard (Indian Survey) . .36.000000 inches. Bird's Standard (1760) 36.000059 " Sir Geo. Schuckburgh's Standard 36. 000642 " Ramsden's Bar. Ordnance Survey 36.003147 " Gen. Roy's Scale 36.001537 " Royal Society Standard 36.002007 " The commission reported in favor of adopting Bird's standard of 1760, as it differed so slightly from Sir George Schuckburglrs standard (w r hich had been used in deduc- ing the value of the French metre) that those values could be assumed as correct. They also established the length of the seconds pendulum at level of sea in London and in vacuo as 39.13929 inches. The seconds pendulum had been previously fixed by Wollaston and Play fair in 1814 as 39.13047 inches. On this report, an Act of Parliament in 1823 declared the only standard measure of length for the United King- dom to be the yard as given by the distance at 32 F. between two points in gold studs on the brass bar, made by Bird, and marked "Standard of 1760," and in the keeping of the Clerk of the House of Commons; also it referred this standard yard to the natural standard of a pendulum vibrating seconds of mean solar time at the level of the sea, in vacuo at London and temperature of 32 F., as in the proportion of 36 to 39.13929 ; so that a pen- dulum 36 inches long ought to make 90088.42 vibrations in 24 hours. The Royal Society had a copy of the legal standard made by Bailey in 1834 ; and in the same year the Par- liamentary standard was destroyed by fire at the burning of the Houses of Parliament, leaving the kingdom again without a legal standard. All attempts made by a commission consisting of Airy, Bailey, Herschel, Lubbock, and Sheepshanks, to restore the standard by means of the seconds pendulum failed in exactness, on account of the many conditions of a vibrating pendulum, and recourse was had to the Royal Society standard, which had been carefully compared by Captain Katerin 1818, and from this in 1838 Bailey and Sheepshanks made six bronze bars, one inch square, and 38 inches long, which in 1855 were legalized by Act of Parliament, and the English standard of length defined as follows : "That the straight line on distance between the centres of the transverse lines in the two gold plugs on the bronze bar deposited in the Exchequer shall be the genuine standard yard at the temperature of 62 Fahrenheit ; and if lost, it shall be replaced by means of its copies." The French metrical system was made legal permissively in 1864, at the length established by Captain Kater, re- ferred to in Act of Parliament of 1823, of 1 metre equal to 39.37079 inches, or 3.28089916 feet. These are the standards now in use in the United Kingdom. UNITED STATES. By the Constitution of the United States Congress is charged with fixing the standard of measures (Art. 1, sec. 8) ; but as no enactment has been made by Congress, the standard yard in England, which was legal previous to 1776 in the Colonies, is the standard yard of the United States, and does not differ with the English standard yard. Under resolution of Congress in 1 830, Mr. Hassler was employed to examine the standards in use. Considerable discrepancies were found, but the mean of all examined corresponded very nearly with the English standard, and in 1832 the recommendation of Mr. Hassler was adopted, and the standard yard defined as the distance between the 27th and 63d inch marks, at the temperature of 62 F., on the brass scale 82 inches long, being an exact copy of Sir George Schuckburgh's standard, made by Troughton, of London, for the Coast Survey, and deposited in the Office of Weights and Measures at Wash- ington. In 1836 an Act of Congress ordered standards to be sent to each Governor of a State, and the work was done under direction of Mr. Hassler. In 1856, two copies of the English standard yard, as restored after destruction of the original standard by fire in 1834, No. 11 of bronze, and No. 57 of Low Moor wrought iron, were presented to the United States by Airy. The United States Troughton standard bar being com- pared with No. 11 was found to be longer by 0.00085 inch, or in proportion of 1 to 1.0000237216, about l inches in a mile, according to Report of Secretary of Treasury in 1857. Later comparisons made by J. E. Hilgard, of the Coast Survey, at the British Standards Office, between No. 11 and the standard imperial yard, give No. 11 as 0.000088 inch shorter, or it would be of standard length at temperature of 62.25 F. We may infer that the Troughton standard is too long by 0.000762 inch, or would be standard length at tem- perature 59.77 F. instead of at 62 by making expansion reduction with Airy's coefficient for the bronze of the imperial standards, 0.000342 inch per yard for 1 F. The metre was made a legal standard permissively in 1866 ; the United States metre standard being one of the 12 iron metre bars made and verified for the French Gov- ernment in 1799 on the adoption of the metrical system, and brought to America by Mr. Hassler in 1800, the relative value being fixed by Act of Congress at 39.37 inches. The relative value of 39.36850154 United States inches, as obtained by Mr. Hassler, corrected to 62 F., was used by the Coast Survey till 1868, when it was found advisable to use the relative value of 39.3704 as deduced by Clarke. Since 1800 several standard metre bars were sent to the United States by the French Government, and on com- parison, there appearing to be a slight discrepancy, the original iron standard metre bar was sent to Dr. F. A. P. Barnard in Paris, and in 1867 it was compared with the French platinum standard, which is only used once in ten years to verify other standards. A difference was found by this comparison of only .00017 millimetre or Y6TTo~oo~ i ncn > which being only y^of an inch in a mile is inappreciable. FRANCE. The standard of length of the systenie ancient was the toise of 6 pieds, divided into 12 pouces of 12 lignes each. The origin of the toise is not known, but it was probably legally established by Philip Le Bel, about 1300, as he first appears to have taken steps toward a uniform sys- tern of measures in France. In the 13th century the toise is mentioned by Ch. Le Rains. In the 14th century Menongier writes that, in marching, the sight should strike the ground 4 toises in front. In the fifteenth century Pere- forest brings in the toise, and in the sixteenth century the Contume de Berry says. " We use in this country two toises ; one for carpenters of 5 pieds and a half, the other for ma- sons of 6 pieds." Picard used the toise in his measurement of an arc of meridian from Malvoisin to London in 1669. The meridians measured by the Academy in 1735 to settle the question of the figure of the earth were made by means of two standard toises, known as the " Toise du Nord," and the " Toise du Sud." The first, used by Maupertuis, Clairanlt, and Le Monnier, in Lapland, was destroyed by immersion in sea-water, when their ship was wrecked on the return voyage. The second, with which La Condamine, Bourgner, and Godin operated in Peru, was the original of the toise Cani- vet made in 1768, and of the standards used in determining the metre. The commencement of the move for a scientific standard of length in France which resulted in the m&tre was in 1790, when the revolutionary government proposed to England the formation of a commission of equal numbers from the English Royal Society and the French Academy, for the purpose of fixing the length of the seconds pendu- lum at latitude 45 as the basis of a new system of meas- ures. This proposal was not favorably received, and the Academy, at the request of government, appointed as a commission Borda, Lagrange, Laplace, Monge, and Condor- cet, to decide whether the seconds pendulum, the quarter of the equator, or the quarter of a meridian, should be used as the natural standard for the new system of measures. 10 They settled on the last as best for the purpose, and resolved that the ten millionth of the meridian quadrant, or distance from equator to pole, measured at sea level, be taken for basis of the new system, and be called a metre. Delambre and Mechin were at once charged with re- measurement of the meridian surveyed in 1739 by La Caille and Cassini, from Dunkirk to Perpignan, and its ex- tension to Barcelona. Operations were commenced in 1792, and carried on with great accuracy to completion in 1799; Delambre working between Dunkirk and Paris, and Mechin between Paris and Barcelona. The distance measured from Dunkirk to Barcelona was 9 40' 24.24" of arc, or 1,075,059 metres, as reduced to the new standard. The " toise de Peru" was the standard used in the work at a temperature of 13 E. Two base-lines were measured with Borda's compensat- ing bars of brass and platinum ; one at Melun, near Paris, 6076 toises long, and the second at Perpignan, 6028 toises long, and though over 900,000 metres' apart, the calculated length differed by only 10 pouces. This meridian was afterward, in 1806, extended by Gen. Boy to Greenwich, on the north, and by Biot and Arago to Formentera, on the south. The results, as given by Laplace in centesimal degrees and metres, are as follows : Greenwich 57.19753 .0 metres. Pantheon, Paris 54.27431 292,719.3 " Formentera 42.96178 1,423,636.1 " The middle of the arc being 50.079655 Cent., or 45 4' 18.0822" Sexa., and the middle degree centesimal being veiy nearly 100,000 metres. 11 The determination of the final result of these geodetic measurements was referred to a committee of 20 members; 9 named by the French Government, and the others by the governments of Holland, Savoy, Denmark, Spain, Tuscany, and of the Cisalpine, Ligurian, and Swiss republics, on the invitation of France. This committee established the meridian quadrant at 5,130,740 toises ; making the metre 0.513074 of the toise, or 36.9413 pouces, or 443.296 lignes, and the toise 1.94903659 metres. Iron standard metre bars, 12 in number were made by Borda, also 2 of platinum and 4 standard toise bars. The 12 standard iron metre bars were sent to different countries, after being verified by the French Government, and on the 2d of November, 1801, the metrical systeme was legalized by France, and the standard unit of length declared to be the ten millionth part of a meridian quad- rant of the earth, as defined by the distance at a tem- perature of Centigrade (32 F.) between two points on a platinum bar in the keeping of the Academy of Science at Paris. This standard bar is used only once every ten years for exact comparisons, as stated by Dr. F. A. P. Barnard. About 1837 Bessel, by a combination of 11 measured arcs of meridian, deduced the quadrant of meridian as 5,131,179.81 toises instead of 5,130,740 toises, as fixed by law. This would make to quadrant 10,000,565.278 legal metres, or would increase the metre length from 443.296 lignes to 443.334 lignes, agreeing very nearly with result obtained by Airy in 1830, from a combination of 13 measured arcs. The following are the measured arcs used by Bessel and Airy ; the combinations being indicated by initial letters, A and B. 12 CO CD ^J" 1 Oi O rH CO Oi O 00 CO i> O O i> 1Q i> rH CO i> CO CO O t> 1O CO iO CO o o o rH O CO G^J O rH O * O O G^ CO CO iO *O G G> OQOOOOOO GOrHrHG>iCOGMGOG> CiGOt>TtCOT^t^ (M iO CO rH CO ^ 0) +. : - 3 ^ : ! r 8 ' 2 93 : i ^ rH f > C . Svanberg, Sweden . Maupertuis, Sweden.. .. RfvilTTO T?TK2C!ia % o P r I] ill > C Hi i & p . Schumacher, Denmark . . .Ganss, Hanover . Roy and Kater, England. a n a . Lacaille and Cassini, Frai . Delambre and Mechin, Fi . Boscovich, Rome . Lambton, India . ....... . Lambton and Everest, In . Lacondamine, Peru T,Q/^Qillo r^QT-vi C^-r\r\f\ TTr\T-i/: r a 1 5 r ! 5 o '- I r.l > T J a 13 The following different lengths of the metre have been obtained : As adopted by France, 1801 443.296 lignes. According to Delambre 443.264 " " Bessel 443.33394 . " Airy 443.32387 " " Clarke 443.36146 From Peru Meridian 443. 440 " The length of a pendulum vibrating 100,000 times in a mean solar day was determined in numerous careful ex- periments by Biot, Arago, and Mathieu, in metres of 443.296 lignes, as follows : Dunkirk 56.671at. Cent. . above sea. .0.7419076 metres. Paris 54.26 " ..65 " ..0.7418870 " " by Borda. 54.26 " ..0 " ..0.7416274 " Bordeau 49.82 " ..0 " ..0.7412615 " Formentera.. .42.96 " ..196 " ..0.7412061 Borda also determined the length of the seconds pendulum at Paris, in vacuo : First result 440.5595 lignes=0.9938267 metre. Second result " " =0.9938460 " As given by Ganot " " =0.9935 " In 1812 the systeme usuelle was established, of which the unit was one third of the metre, with the old name of pied, and duodecimally divided into pouces and lignes. This system continued in use till 1840, when it was abolished by law, and the names of pied, pouce, and ligne forbidden under penalties. So the metre, decimally di- vided, remains the only legal measure of length in France. COMPARISONS OF UNITED STATES AND ENGLISH STANDARDS. In 1832, under resolution of Congress, Mr. Hassler compared the different standard yards in America, with 14 * the following results, using the yard between the twenty seventh and sixty-third inches on the scale made of bronze by Troughton, of London, for the United States Coast Survey, as the reference, that being identical with Sir George Schuckburg's standard : Troughton Scale, mid. yard 36.0000000 inches. " " between platinum points. ..35.9989758 " Jones yard in State Department 35.9990285 " Iron yard in Engineer Department 35.9987760 " Brass yard, Albany, Sec. of State 36.0002465 " Gilbert yard, University of Virginia 35.9952318 " In 1856 the Troughton standard bronze scale was compared with the bronze standard yard No. 11, which was sent over by Airy as a copy of the English im- perial standard, as restored after destruction of the origi- nal standard by fire in 1834, and the United States standard was found to be longer by 0.00085 inch. Later comparisons by J. E. Hilgard, of the Coast Sur- vey, of the bronze standard No. 11 with the imperial standard yard, at the British Standards Office, gave No. 1 1 as 0.000088 shorter than the imperial standard. Hassler's reduction of the metre, as deduced by Beach at 62 F., 39.36850154, compared with the English reduc- tion of the metre, 39.37079 inches, gives an excess to the United States Standard of 0.002029 inch. The following reductions have been given for the United States yard in English inches : Report of Sec. of Treas., 1857 36.00087 = 1.00002416 Chambers' Encyclopaedia, 1872. . . .35.00087 xu_ s~ " " " ....36.0020892 =1.0000580334 ^/ Trautwine 36.0020894 = 1.000058038 ' \JVlathewson, U. S. surveyor 36.00208944 = 1.00005804 Hassler and Beach 36.002092 = 1.00005811 J. E. Hilgard, Coast Survey 36.00076 = 1.000021 15 To Mr. Hassler's reduction the name of United States inch has been applied ; but his reduction is not correct, as he used a rate of expansion for brass deduced by himself of 0.0003783 inch in one yard for 1 F., and later experi- ments show that the smaller rate of 0.000342, deduced by Airy, is more correct. By correcting Hassler's reduction with the later rate of expansion, J. E. Hilgard shows that the difference would be very small, or only 36.0002286 = 1.00000635, or about f- of an inch in a mile. In Coast Survey report for 1876, J. E. Hilgard calls at- tention to another difficulty in the matter of extreme accuracy, in the uncertainty with regard to the permanence in the length of a bar, and states that the bronze standard bar No. 11 and the Low Moor iron standard bar No. 57, presented to the United States by Great Britain, are found to have changed their relative length by 0.00025 inch in 25 years ; the bronze bar being now relatively shorter by that amount. This subject, he states, is undergoing further investigation. COMPAKISON OF UNITED STATES AND FRENCH STANDARDS. In 1817 Mr. Hassler examined the French standards in America, for the Coast survey, using the Troughton bronze standard scale, which is identical with Sir George Schuck- burg's standard, as the reference, with the following results, all being reduced to temperature of 32 F. Original Iron Metre, 1799 39.381022708 inches. Lenoir Iron Metre, Coast Survey 39.37972015 " Brass " " 39.380247972 " " Eng. Dept 39.38052739 Canivet Iron Toise, 1768 76.74334472 Lenoir " " 76.74192710 16 In 1814 Trough ton had compared with his own scale in London two of the above. Lenoir Iron Metre, C. S 39.3802506 inches.. " Brass " " 39.3803333 In 1832, under resolution of Congress, Hassler again compared the French standards in the United States, using as before the Trough ton scale, and reducing all to tem- perature of 32 F. as follows : Original Iron Metre, 1799 39.3808643 inches. Lenoir " " C. S 39.3799120 Brass Metre C. S 39.380447 Eng. Dept 39.3801714 " in 1829 39.3807095 Fortin " State Dept 39.3796084 Treas. " 39.3795983 Iron Metre " " 39.3807827 Gilbert " Univ. of Virg 39.365408 " Platinum Metre 39.3803278 " (Nicollet) 39.380511 Canivet Iron Toise, 1768 76. 74290511 " Lenoir " 1799 76.74047599 " From the mean of his comparisons between the United States brass Troughton standard yard and the authentic French standard metres used by the Coast Survey, Hassler, in 1832, deduced the value of the metre at 39.3809172 inches, at 32 F., and by correction for expansion to United States standard temperature of 62 F., he made the metre at 32 equal to 39.36850154 inches at 62 F. The British imperial standard and the United States Troughton standard differ by only 0.000762 inch, which applied to the English reduction of 39.37079, would give 39.36996 as the relative value according to Troughton standard. 17 The difference between these reductions is probably to be attributed to the use of different rates of expansion, in correcting for standard temperatures, which vary con- siderably, according to high authority as follows for brass atlF. Whitworth, 1876. . . . 0.00000956 = 0.00034416 in. per yard. Borda, 1799 0.000009913 = 0.00035687 Smeaton, 1750. ..... 0.000010417 = 0.00037501 Hassler 0.000010508 = 0.0003783 " Eamsden, 1760 0.000010516 = 0.0003786 Faraday, 1830 0.00001059 = 0.00038124 And for the bronze of which the British imperial stand- ards are made : Airy and Sheepshanks. . . 0.0000095 = 0.000342 in. per yard. Fizeau 0.00000975-= 0.000351 " The correction at Ramsden's rate is nearly identical with Hassler's, and gives 39.3684933 ; at Whitworth's rate it would give 39.36962, very nearly the same as deduced from the difference between the British Imperial standard and the United States Troughton standard. The results of Sir Joseph Whitworth were obtained by use of all late improvements for scientific precision, and they must be accepted as most reliable. It would appear preferable to give comparisons at the same temperature in connection with the corrected result, so that international comparisons of scientific measurements may not be vitiated by accidental variations. COMPARISON OF ENGLISH AND FRENCH STANDARDS. When the metre standard was established in France' 1799, it was compared with Sir George Schuckburg's 18 standard yard by Captain Kater. The quadrant of 10,000,000 metres, or 5,130,740 toises, was determined to be 32,808,992 English feet, giving the metre equal to 3.2808992 English feet, or 39.37079 inches, and the toise equal to 6.3945925921 English feet. In 1814 Wollaston and Playfair, by comparison with the platinum metre standard at 55 F., deduced the metre as equal to 39.3828 English inches/ During the geodetic operations of General Roy in 1802, who used 60 F. as standard temperature, Pictet's compari- sons, using means capable of measuring the 10,000th part of an inch, gave the metre standard, which is used at 32 F. as standard temperature, at 39.3828 English inches ; this corrected for temperature by Dr. Young, gave 3 .371 English inches at 62 F.; which result was confirmed by Bird, Mas- kelyne and Laudale. In 1823, by Act of Parliament on report of committee, the metre is fixed as 39.37079 English inches. In 1800 the Royal Society, by comparison with two toise standards sent by Lalande to Maskelyne, deduced the metre as 39.3702 English inches. Later comparisons by Clarke in the Ordnance Survey Office at Southampton, in 1866, give the metre as 39.37043 inches. The French Academy of Sciences by comparison with Sir George Schuckburg's standard at temperature of 32 F., deduced the metre as 39.3824 English inches, which reduced to standard temperature of 62 F., would be 39.3711, or slightly in excess of the value deduced by Dr. Young from Pictet's comparisons. The legal value in England is one metre equal to 39.37079, and the latest reduction is 39.37043 inches by Clarke in 1866, which is probably the most exact re- duction. 19 DIFFERENT REDUCTIONS OF THE FRENCH TOISE INTO ENGLISH FEET. Captain Kater, 1799 6.3945925921 feet. Hassler, 1832 6.3951409 Chambers' Encyclopaedia 6.39456 Mathematics 6.394662 Wallace 6.39462 Nystrom 6.39625 Alexander 6.39435 Dana.. 6.3946 The following table of reductions as used shows clearly how great a confusion exists in the matter of comparisons : MfiTRE IN INCHES. Phcenixville Hand-book 39.368 inches. Hassler 39.36850154 " .39.370788 " 39.3809172 " Trautwine 39.368505 39.37079 Silliman 39.368505 39.37079 Chambers' Encyclopedia 39.36850535 " " 39.3707904 " Act of United States Congress, 1866 39.37 " Smithsonian Report 39.37 " Youmans 39.37 Davies '..3937 Homan's Encyclopaedia 39.37008 Weale ". 39.3702 Ordnance Survey (England, 1866) 39.37043 Clerk Maxwell 39.37043 Capt. Clarke 39.3704316 " J. M. Rankine (1870) 39.3704316 " (1866) 39.3707904 " Alexander (weights and measures) 39.37068 Ganot 39.370788 Vose . ..39.370788 20 Act of British Parliament, 1823 39.37079 inches. Encyclopaedia Britannica 39.37079 Hymer 39.37079 Davies and Peck 39.37079 J. W. Clarke 39.37079 Dana 39.37079 " Whittaker 39.37079 Sommerville 39.3707904 " Chambers' Mathematics 39.3707904 " Gwilt's Encyclopaedia 39.3707904 " Gillespie 39.3707904 " Capt. Kater 39.3708 Appleton's Encyclopaedia 39.37079 Van Nostrand 39.3708 D'Aubuisson 39.3708 Johnson (draftsman) 39.3708 Encyclopaedia Americana 39.371 Jameson's Dictionary 39.371 Herbert's Encyclopaedia 39.371 Popular " 39.371 Molesworth 39.371 Dr. Young (1802) 39.371 Wallace (engineer) 39.371 Nystrom 39.38091 Hencke 39.3809172 " Act of Canadian Parliament, 1873. . .... .39.3819 Paris Academy 39.3824 " LENGTH OF THE SECONDS PENDULUM AS GIVEN BY DIFFERENT WRITERS. NEW YOKK. Hencke 39.1012 inches. Bartlet 39.11256 " Nystrom 39.1017 Ganot 39.1012 Byrne 39.10153 " Wallace 39.10153 " LONDON. Hencke 39.13908 " Gillespie 39.13929 " Chambers' Encyclopaedia 39.13929 " Williams' Geodesy 39.13929 " 21 LONDON. Act of Parliament, 1823 39.13929 inches. Wallace (engineer) 39.1393 " Chambers' Mathematics 39.1393 Hymer Astronomy 39.13734 " Bartlet 39.13908 " Vose 39.1393 Sommerville 39.1393 Nystrom 39.1393 Davies and Peck 39.13908 " Ganot 39.1398 Wollaston (1814) 39.13047 " Galbraith 39.139 Byrne 39.1393 Capt. Kater 39.13829 " PARIS. Hencke 39.12843 " Ganot 39.1285 Galbraith 39.128 Byrne 39.12843 " Wallace. ..39.12843 " APPENDIX. HAVING shown in the preceding pages that in the point of view of scientific accuracy the yard, metre, and toise standards are on a common level, and that in the matter of comparisons there is no extreme accuracy, I will now refer to the proposed change of our standard from the yard to the metre. Theoretically the metre is the 10,000,000th part of the earth's quadrant, and the yard the -g^-j^VsT^ P ar ^ ^ a seconds pendulum at London. Practically, neither the metre nor yard could be recovered with exactness from their natural basis. The legal French metre differs from the latest reduction enough to give an excess of over three miles to the circumference of the earth. In fact, the metre- and yard are only the lengths of bars of metal kept in certain offices, from which copies are made. Decimally considered, it is as easy to divide one as the other into tenths, hundredths, etc., and the yard standard is often so divided. As to nomenclature, the metrical system is over- loaded with Greek arid Latin prefixes, which are in no way so easy and convenient in expression as the short, sharp Anglo-Saxon words yard, foot, inch. In all sciences Latin and Greek names are given for easier purposes of classification ; but the different peoples invariably keep their own household names for daily purposes, leaving prefix and affix to specialists, probably with advantage to both parties. The units used for different purposes are entirely dis- 23 tinct from the base of any system, and though always re- ferable to such base, are not practically so referred. It therefore seems useless to burden the people with long scientific names in the ordinary transactions of daily life. For long distances the units in the yard and metrical systems are respectively the mile and the kilometre. The mile has a definite meaning in our minds, being associated, from the days of youth, with the measured dis- tances in race-courses, speed in walking, railway and steamer travel, length of surveyed lots the same being in use among about 100,000,000 people. For mechanical structures, the units are respectively the foot and the metre. The foot is used instead of the yard, as being the most convenient in practice, and is fixed in the minds of the people by constant association with length of foot-rules, size of buildings, doors, windows, etc., all of which are always before us. For commercial purposes the units are respectively the yard and the metre. The yard is associated with length of yard-sticks, distance between brass nails on counters, so many finger-lengths by ladies. Probably three fourths of the business of the world is conducted on the yard standard. For machine and shop work the English unit is the inch and fractions, and countries having the metrical standard have universally adopted the millimetre. The inch is well fixed in the minds of all mechanics by constant use, and the ease with which the fractions are had by halving only renders the system very convenient. As more figures must be used to indicate a size by millimetres than by inches and fractions, it appears that the metrical system cannot shorten the work of arithmeti- cal computation in shop work, and is therefore of no ad- vantage to the mechanic or draftsman, but rather the reverse. This is the opinion of Coleman Sellers, the dis- tinguished Philadelphia engineer and manufacturer, who, after a trial of the millimetre in his shops for some years, returned to the use of the inch, and writes in Engineering News : " The loss from the use of a small unit requiring many figures to express what is needed, takes away from the other advantages of the system when considered from a labor-saving point of view." In France itself the metrical system is not wholly deci- mal in actual practice, as we find the following measures in use in addition to the decimal divisions : double deca- metre, de mi-decametre, double metre, demi-metre, and double decimetre. The metrical system has been adopted in the following countries: France and colonies, Holland and colonies, Belgium, Spain and colonies, Portugal, Italy, Germany, Greece, Roumania, British India, Mexico, New Granada, Ecuador, Peru, Brazil, Uruguay, Argentine Confederacy, Chili, Venezuela ; and partially in Wurternburg, Bavaria, Baden, Hesse, Switzerland, Denmark, Austria, and Turkey. In the past centuries all the work and records of Eng- lish-speaking peoples now numbering about 100,000,000, and increasing and progressing faster than all other nationalities, as well as being closely connected by de- scent and business have been done and recorded under the yard standard, and any change now would inevitably render necessary continual reductions, to the great detri- ment and inconvenience of the mass of our people, and with little or no practical benefit, except perhaps to a small class of scientific and pseudo-scientific men, who can and do amuse themselves with the fancied uniformity of the metre. All our numerous text-books and tables, mechanical and scientific, would be rendered entirely useless by the change, and this is a serious final consideration. UNIVERSITY OF CALIFORNIA LIBRARY BERKELEY Return to desk from which borrowed. This book is DUE on the last date stamped below. -Ifcr^ *'n ., r. 20 10 REC'D LD DEC 2 01957 10 U 14 1988 OANIStK ' - ->' rtEC'D LD APR 2 9-65 -11 A tAMt 819741 ttt. MOrflll LD 21-100m-9,'47(A5702sl6)476 GAYLAMOUNT PAMPHLET BINDER Manufactured by eAYLORD BROS. Inc. Syracuse, N. V Stockton, T U.C. BERKELEY LIBRARIES CDDS37DSSfl