, y ad ey $m . Ay, ne 16 ois saet Bettas stnentT® iste tena | sa pid ry apaye 4 pd ’ ett ee a eb ee spelt . ee ‘ re leq iy at 8 a4 Hats >. ther interes 4 “ TeAS ad Pabel el ley he Oder pe Ci be bie | . PPP ete AE cine shh, 2 +4 = Ria tow at HHO te Genk Ame tga yoo Ae ace raed 7 % ib wes SUN POPE le len ple yee eee? beg Miernt yee. sg pet th : " ‘ te pane bey y ch} va a! A ; ah Pe nih ismarers “) ’ . avy age 4 ‘ ‘ x -s " Mu : ‘ ¥ it ity Epa bt ays . retain oh) ak oteie Wee onteatettt age att t tytiremiy be owe Poe Wh i Tencaeeaed wal . 4 Pag ‘ eats Hick < Aas bal ED e t Bai a Vie a - , yeas Mii, rin rai ier ‘ me sate 8 4 1 4b. oe ted Oke Tania! ; in v0 Wodatet y Pie itat Thy q Ry et 1yrd yy) bw pe 4 \ ae ins seal AN * eae og Sh : 4 ree ey rere ~ ys a Tart yt - o tha i Melgwitian ste st ne t . eiatinitel 4A Hrs 9 18a Fat aiagaisd 5; ieial ater oak SNe atta ee aiiaaaiety pe Leth ‘had } Vary aients nia oi a i we we gee 1 tide & LD Memb] e 4 oh t ste evir : *, % ora’ * puer “7 a ay ” is Vhaie tah fh wile te * vate PE + ea t M An er er ice A bu hace ify eve heh | a hit i by ie ian WF ois ae ’ oe Ht ea th oe ‘ ¢ be Uediliees Wnavgeretedet rl desbods jae dx Mi 3 Fe Mi ae i ttt. ’ Leg iad , ‘ sited Mf +3 ¢ ath Medd leg id ita. tdog Vat : a ts a Vey Hi iat i " . lo ee ve) rievew tb yy at ys sh } prt Pai nth aah Fi - “t 4 dow i Magee aed we ea ati wh ite + ea Le vite “ tee 4 rivae z Ane it i ab ATE 7 by Tech ated a ae ate ‘ sary yi / f 1% pre ‘ pe aet nt " * } ‘ 4 He st ’ t} ‘ ' nt A reli A ii a { ‘i ga HAN te le So ian ‘ & belt Ey Wd Ra Ae ast ti PHU ze aes oe ' jeg i, Ot wit tt nik HA tage yal i { 4 pd eed edety | is nH ao ft iti i Sia \ se aio i ; aa ia ava j rig 344" ‘ abate Mise } i PW ‘ai 5 ae ha HY phbi ta Ne v if ; abe) beep ie aY neh at if RE EH 4 Hab Gaieray Nagai pal : r DHMH ite) a NE Let gH Ea PETRY buy Geshe Uy Hh i nee ia Hh MS oe ; j feiss aah i! ve ae iin ati Be a ik Ati eft hi iv Hedy. iL rh, Tie at : vad me he it é ‘ Oia " a + Wirwet vo ‘ Hs iat j Hue i t at) sieht Bh hs i ea robe Hy L re ee i Le iG yes ae iat ae HY haath y aadinyasiay 14") a bh Mt { hat salted g ‘ id) 4 alt 5 Spe. «! a) ye oi Petre i i) ; 4\) Ht ihe fated yl : ire ah. sa pit EY +) p ; Hero one cma OE RN A i ae itu titeteyt ini ti Hien : a iar Riba cet Leste F Moat 148 9) of Rate as HN Nea Had as : Pe ' Ht i ETE ' Hshatgs shot 8 athehd ifady tate ; : t Nae Hh i iy Asli WMalence aAe Le olen ue dete ha Trp Na tet aa 1 wit ait 3! aia iil Mal en i t - as 7 Me Aca eda yea aa | 4 Y, ; We i aa Ey iy } ae Ag! ane ihn entcia) bv diet ; Blas ; aie yt aya fi aH Reieteen Re led i iv ate ‘ a a sf \ it ie ‘at ee it Aas AG : Nar) ' fa ets Depa tar at rh alii! A al atts Wan Ait as a Da ih iy Ni Starting Vhs tek H Ri Y iy aang ih at ph ys iste se Mina tty t iene ty us et) ; \ ah el canta ee HEAR ta ane } a | iret ada: Rant Ah Bats iH ia tb ire i wit aman ekec ‘ * A bh teaaee : me y hit Hate int in his 159 20 if ie i Wand ad GEN i unt aes i me 4 vt , i 1% OYE Wats Wats Hiss row ar + ag Si etait Ns } i tae Hanna ge et 4 i gt ; is vera 8 ht ab nant ‘ ’ uy AYE ns Gara auaianeahe Aaa a } Th shi in ‘ : ‘ Hy 4d eter Rel > \ Ba i in tea aa ah ae iy a! \ eights qa’ iat aie sth t Wii et " ry stash s Saad sea thay ty iaetarab rata! ‘A sale hs wets ts t os tah Pid “ti i¢ at ve at aban panned : , Hae i ahd Ae) hat ae wetet ihe by idian 4, fale SSE MEE Litt \ | \ one At ; iy bi mht Vest izle tet OEE 4 ar aac at ‘ te ty ; " ey * ' oe ith ts " 14 ' : { a Pe adele Nadi ly Fea Oda se Hi piaaeaNt n it ira , is . 8 ee ; ; 4 sit? . M Sane ty os tite . ‘ Ee 9 if ’ : ‘ 4 wad 44 eH : rs talhiat i fe ' ‘ ) Nath YL) eye Poy ohne hte Prt aan wits Sines Ke 7) ‘ 4 Rae AY nye iy th y ay : ' . ' +h Velen ae she PORTE Leet A aa age ie AMotttn REL yds : 4) 4 . ‘ , ta Qena Hs 1," Sueee 4) : ‘ : : V x ve ‘ ke on f Ans MBAR ‘ i tad 4 fete aot t a phan tse Paes atte ee ‘ ‘ ’ , fount . Hieber gh 4 ‘ » andre aw steel SONS I as KHTER oR sisi pla deen aga iateaesuiatie tt The person charging this material is re- sponsible for its return to the library from which it was withdrawn on or before the Latest Date stamped below. Theft, mutilation, and underlining of books are reasons for disciplinary action and may result in dismissal from the University. To renew call Telephone Center, 333-8400 UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN L161—O-1096 Ns wis Auth ig oe) tA fos | ere Ri 2 ny ao {2 ae : , TABLE OF SQUARES AND CULES. Cubes. )) NO. |Squar, 3}264127 51286 907|508I51 dlsaalls. (441/462/485}5041525}546)567/588\609\6s0\651 el 51598'621|6441667|690\715 TABLE OF THE MUTUAL PRODUCTS OF NUMBERS BELOW ONE HUNDRED. 26] 27|26]29 s0\\51 56 | 37 | 58 | 59 | 40 | 62 | 65 | 64 41 | 42 | a0 | 44145 46 | 47 | 48 49 | 50} 51 | 52] 55 —_—_—— 54, 55 is Bec 596] 407] 418] 429} 440] 451) 462) 473] 484] 495) 506] 517] ses| 550] ssol s6il 572 sss| 594] 60s! 616] 627 671 9241556\548)560 A52| 444) 456) 468) 480!) 492) 504) 516} 528) 540) 559) 564] 576] 588] 600! 612] 624) 656] 648| 660] 679] 684 752) 744) 756 998155115641577|500) 468] 481) 494) 507) 520!) 533) 546] 559) 572) 585] 598) 611] 624] 637] 650] 663] 676] 689| 702] 715| 728] 741 793) 806} 819) 56415781592) 406)490 504) 518} 552) 546) 560} 574) 588] 602) 616) 650) 644] 658| 672) 686] 700) 714] 728] 749] 756] 770] 784] 798 854| 868] 882 i 515901405 420\455|4.50) 540) 555} 570) 585) GOO} 615) 650] 645) 660) 675] 690} 705| 720] 735] 750] 765] 780] 795} 810] 825] 840] 855 915} 950) 945) § =H poalaoc yo to2|448}464 480) 576} 592| 608) 624] 640}) 656] 672] 688) 704] 720] 756) 752] 768| 784] 800} 816] 832| 848] 864] ss0l s96| 919 976| 992|1008|1024 ai alae 44214.5914'76]4951530\|5e7 612) 629} 646) 665} G80} G97} 714) 751) 748) 765] 782) 799] 816] 855} 850|| 867] 884] 901] 918] 935] 952] 969] 98611003/1090/1037]1054 596/41 4)4521450)468)486|504\5201540|5: 648] 666) 684) 702} 720) 758] 756) 774) 792) 810] 828! 846] 864] 882} 900] 918} 936] 954] 9721 990|100811026]1044]106211080]|1098}1116 915|5352)5511570)589) 665) 684) 705) 722) 741) 760} 779) 798) 817] 856] 855| 874] 895] 912] 951] 950] 969] 988|1007|1026]1045|1064|1083]110211197111 40]1159|1178 5405601 580\600|620 700} 720| 740} 760 780} 800} 820} 840} 860} 880] 900} 920} 940] 960} 980|1000]102011040]1060]10801100|1 12011140111 60]1180|1 200] 1220] 240|1260|1280]1500]1520]1540/1560]1580|1400}}1 420/1 440 per wee 755) 756) 777 770) 792 798} 819) 814) 856) 858 805) 828) 851] 874] 897 840]! 861] 882] 9035) 880]} 902] 924] 946 92011 945] S66 594161 616381660 682 600\624/6481679\606\790 [744 840} 864] 888} 912} 956] 960 95:4)1008 12007) 05611080}110411128)1 152/1176)1200}1224)1 248/127 2}1296)1520)1544)1568)1592|1 41611 440 1464/1485) 650167 51700)72517501775 875} 900} 925} 950) 975)1000}102.5)1050}1075)1 LOO}1 1 25}11.50}1175}1 200]1225}1250]}1275|1500|1525}155011575|1400|1425]14.50|1 4'75|1 500/1525]1 550|1 57 5|1600)1625/1650/1675]1700)1725 676/702'728'754I7s0|ls06 910} 956) 962} 98S)1014)1040/1066]1092)1118)1144)1170)1196|1229]12948112741500]1526]155211578)1404}1 430|1456|148911 50811 55411 560) 1586|1612|1658]1 66411 690}1 71 6|174211 7681794 729 945] 972} 999)1026}1055)1080})1107|1154)1161]1188}1 21 5124911 269)1996|1525/15.50/1377]1404|1451|1 458}1 485|151 21153911 566|159311620)1647|1674 924] 952] 98 928} 957} 986}1015}1044]1073/1102/1151/1160}}1189]1218]1247}1276113505 900/950} 960, 990}1020}1050]1080/1110}1140/1170}1200}1 250/1 260)1 2901 520/13.50}1580}1410f1 440|1470|1500}|1 550|1 560|1 590|1620}165011680|1 710117401177 (961) 992 1025)1054)1085}1116)1147/1178)1209}}240)127 1/1502)1533}1564}1595}1426|1457|1 488|1 519|1 5501 581|1612|1645|1674|1 705|1756|1767|1'798]1 82911 860]|1891]1922]1955 {1024 10.56)1088)1120)1152/1184)1216)1248)1280}}1 51 2)1544)1576)1 40811 440}1 47911 504|1556)1 5681 600)}1 65211 664|1696|1728]1 760|17991182411856|1888|1920|1952|1984|2016|2048 1089) 122)115511188}122112.54/1287|1520})1 5.55)13586/1419)14.52/1 485}1518)1551|1584|1617|1 650|1685|1716|1749}1782|1815|1848|1881|1914|1947|1980||2015|2046|2079|211 2 1156) 190|1224 12.58)1292/1526|1560))1594/1428)] 462}1 496]1 550)1 564|1598)1652)1666|1700|1 75411 768|1802|1836|187011904|193811972|200612040]|2074|2108)21 4212176 1225|1260,1295,1550}1565)1400}] 45.5/1470)1505)1540)1 575)1610}1645}1680|1715]1750]]1785|1820]1 855|1890|1925|1960|1995|205012065]2100]2135|21 70}2205]2240} 1296]1552)1568)1404)1440})1 476)1 5129/1 548}1 584]1620)1656|1692/1728|1 764) 1 800}]1 $56|1872)1908]194411980|2016|205212088]21 24121 60/|2196|225212268|2504 1569 1406)1445|1 480/71 517|1554)1 59116281 665)1702|1759}1776|1 815]1850}1887/1924|1961|1998]2035}2079|21 09]21 46121 85|2290/|2257|2294295112568 11444]1489/1 520/11 558/1596|1654|167211 7101 748|1 7861182411 86911 900]1938|1976|20141205212090 2128/2166]2204/2249}9980)|251 8}25.56)2594|2452 1521}1.560})1 599]1658/1677|1716]1755]1794|1853]1872|1911|1950)|1989]2028|2067]21 06/21 45|21 84|2295|29691950119540/12579|241 812457|2496) 1600}/1640}1680}1720)1760)1 800}1840}1 880}1920}1 960]2000)]2040/2080)2120]21 60!2200]22.40/2280125201236012400!24.4012480/252012560) 1681}1722}1763)1804/1845}1886|1927|1 968'2009}2050}|2091|21 52121 75}221 4/2955|2996|2557125781241 912460] 2501/254212585 1764/1806/1848/1890/1 952]1974]2016|2058]2100]21 42)21 84|2996|2268]2510]255912394|2456|2478]2520/|2562|2604|2646|2688 2 756/785 810, eo 954}1565|1592)1421/1450)}1 4791 508}1537|1566)1 595]1624/1655|1682!1711|174 1849}1892)1 955}1978]2021]2064/2107|2150)|21 93]2236)2279]252212565|2408|245112494/2557/9580]|2623|2666|2709|27522 1956}1980}2024|2068]21 12/21 56|2200]/2244}2288}2332}2576|2420)2464|2508]2552|2596|2640]2684|2728|27 72/2816 2025}2070]2115)21 60 2205|22501}2295}2540]2585}2450}247 5}2520|2565|261 0/2655|2700|2745|2790|2855|2880 2116]2162]2908 2254|2500)]12546/2592)2458/2484]< 2576|262212668/2714 2760) 2806]2852]2898/2944 129092256'2305]2550/]2397|2444|249112558|2585|265910679|27961277510890|2867|291 4|2961|5008 12504 2552|2400/2448!2496|254412592|2640)2688|2756)27841285912880]2928|2976|5024|507215 2500) es 2650/2700)27 50/2800]28.50!2900)2950|5000}3050|5100|51 50|5200 2601)2652 2703)27 54 2805|2856|2907|2958|5009 5060)51 11/51625 2704/2756 2808/2860/291 2}2964|3016'5068)5120)31 72/52241. '2809]2862!291 5/2968|3021|5074/5127/51 80/3235]5286 2916}2970/502415078/315215186|5240/152941554815409 5025/5080/5155}2 k 15300))33.5515.41 0/5465/5520) bisels199ls04 52491530 315420)15477|5554}559115648 5564/5422'5480)5558|5596|5654/071 2/5 15481|3540]5599|5658/571 719776) 3600]|5660|5720|5780|5840 5721|5782!5845|3904 3844/5906|5968 924] 945] 966] 987}1008}1029}1050}}1071|1092|1115}1154|1155|1176/1197|121 S]1 95911 260} 1281]1502|1325]1544)1565}1586/1407)1428)1449 968] 990|1012)1054|1056|1078|1100}1129}1144|1166|1188|1210}1 25911254|1976|129911520/1542|1564|1586]1 408)1430)1 452)147411496 959'1012}1055}1058}1 081}1104|1127|1150}}1175|1196|1219|1249}1 965)1988|1511|1554|1357|1590)1405]1426]1449]1472)1495 1008}1056|1064)1092}1 1201] 148/1176)1204)1252)] 260)1 288]1516|1544|1572}1400}]1428]1456/1484]1519}1 540)1 568|159611624]16591680]1708|1736|1764]1792/1820]1 848]1876|1904 1769]1798]1827|1856]1885 UO eS ee | eae ee 1800]|1850|1860|1890]1920]1950}1 980]2010/2040}2070!2100!|2150}21 60]21 90/2220/2250)2980/2510)2540 2570)2400)1245¢ 2046|2077]2108/2139]21 7011220112232122965/2294)2525|25 56/258 7/241 812449]2480)/251 1/2542)2575|2604126; 2568|24001245212464|2496|2528]2560]]259212624 126 56/2688)|2720|2752|27 '270612759]2772|2805|2838|287 56|2890|292412958]299215026|5060]|3094 151 28151 62|51 96|5250|52 40/297 5|5010|5045|508 3 1984/2015 go Rte 67 | 68 | 69 | 70 76|77| 78 80 || 81 ] 82] 835) 84 757| 748] 759) 770 847| 858 880]| 891} 902) 913) 924 804] 816] 828} 840) 924] 936] 94 871| 884} 897) 910) 958} 952} 966 955) 946] 95 972} 984] 996]1008]1020]1052}1044}1056 1]1144]1157|1170 511260)1274/1288 5115 50)|1565|1580}1 5951 4|1440]]1456/1472)1488 923] 936] 949] 962] 975] 988]1001/1014|1027/1040] 1055/1 06611079) 1092]1105|1118|115 980|' 994]1008]1022/1036]1050)1064]1078]1092}1106)1120))1154}]1 148]1162)1176]1190|1204]121 5] 990]100511020]1035/1050]1065}1080]1095}11 10]112.5/1140)11 55/11 70)1185)1200)|121 5/1 250}1 24.5)1 260)1275|1 290) 150. 1040|1056|1072}1088]1 10411 120]}1136]1152}1168]1184{1200}1 21 6]1252)1248)1264)1280)|1296)1312)1528)1544/1560)1576|139 1952|1960]]1 988]2016|2044]2079121 00)21 98/21 56/21 84/221 2/2240)}2968/2296/252.4/955: ted mae 2080]211 2)21 44/21 76/2208/2240)'2979)/2504/255' 21 4512178|221 112244] 2977251 0)12545|2576|2409|2442/2475|250812541|2574|2607|2640)/2675 2210|224412278)251 212346/2580]1241 4|2448/2482]2516/2550/2584/261 8)2652/2686]2720 2754/2 92275|251012545|2580|241 5|2450]]2485}252012555|2590|2625|2660|2695]2750/2765|2800)/2855)2 234019576|241212448|2484/2590)12556]2592/2628|2664)2700|2756/2772|2808]2844}2880 }2916 2s 12405]2442912479]251 6|2553|2590]|2627|2664/2701|2758]27 75] 281 2/2849/2886|2925}2960 2997 5 2.47012508|2546|2584)2622|2660]12698|2756|2774|281 2|2850)2888]2926| 2964/5002, 5040 5078)5 255512574|2615|2652|2691 |2750|12769]2808|2847|2886|2925}296 4|5003|5042!5081/5120)51 59 5198 3 2600|2640|268012720]2760}2800}]2840|2880}2920|2960}5000|3040|5080)51 20/51 60}5200))5240) 328015% 5D62/5 44415 ¢ 502: 2706|2747|2788]2829|2870}|291 1]295212993|5054|507 5/51 16)51 57/51 98/5259/5280)}55 2750|2772/281 412856|2898|2940]1298215024|5066|51 0851 50151 92)5254)5276|5518/5560)5 2967|3010]3053/3096|5139|018219225|5268|551 1155.54|5597|5440)15 4/3168}: 5 5 2838)2881/2924 2860|2904|2948/2992)5056/50801'5 2970/3501 5|3060}5105}51 50}! '2990/5056|5082)51 28)5174/5220)/55 5055|3102|5149)5196)5245/5290)/5; 5168/3521 6)5264/5512)5560)3 5608 5652 56|9500/5544|5588)5452|5476/5520)|5. 357 5154201546515510/3555|5600)15 ibe a) 5250|5500/3550|5400}5450/5.500}15550|5600|3650)37 21548415 556|5.588)5640)|5692)5 3965|4026/4087|4148/4209]4270)/4551|4592|4455}4.51 4/457 5}4656/4697|47 58/481 9/4880) 4050|4092)41 54/421 6/4278/4540]|4402!4464|4526|4588|4650)4712}4774/4856|4898/4960 4095|4158)4221/4284/4547/441 0)'4475|/4556|4599/4662/4725]47 88/4851 |4914 5022 EXPLANATION OF THIS ENLARGED MULTIPLICATION TABLE. B960}4052 The numbers as far as 100, except the first decad, which is omitted as unnecessary, are ranged horizontally in their order from left to right ; and their products by like numbers, from 11 to 99 inclusive, which descend along the right hand side, are placed in the corresponding vertical columns. If the numbers to be multiplied be equal, their product, which is in this case a square, will occupy the exterior diagonal; but if those numbers be unequal, the greater is to be sought in the row along the top of the Table, and then glancing down the column to where a line from the other number on the right hand crosses it, the product -is found. Thus, the sqtiare of 34 is 1156, which stands at the bottom of its column; and to multiply 78 by 62, cast the eye along the column ander 78 Aill opposite to 62’on the side, and the product 4836 occurs. Example of Common Example of the Squa- Example of Abbreviated swam leof Ets # Mulii lication. ring of'a Number. Decimal Multiplie 685214 7912 62599744 117716257604(545098 4489 504957115891(6752 559671 7912 -785598 1156 5 500765 994 7144 48356 6850\ 211625 15467 4194115891 2972 1896 - 4602 Saree 62351 405 4828 6241 7566 900 961 1612 1544 62599744; 54 686098 \ 67237604 135294061 2914 5072 being the 5286 oe eee 5] 1891 6528 square of the 5127 5880 490 mean diameter 52 9604 112 of our globe, 6076 2580 in English 59 245752262594 miles, 491657143 which must conse- quently denote, in English square miles, the fourth part of the surface of the earth. 4160)4224|4288145 52/441 6/4480]}4544|4608|4672/4756|4800}4864|4928|4992)50.56] 51 20)15184}5248/5519 4925|4290}45.55/4420/4485]4550/ 461 5|4680/4745}481 0)4875}4940)5005|5070 '44.89}4556]4625/4690)'47.57|4824|4891 |4958 4624/4692/4760)/4823/4896/4964|5052 5100}5168|5256|5504|5572! 5440/5508 a 4970) 5041 oll6075|61501629516300 76|5352|592816004/60801/6156|6252l6308\6594 592916006 |6083,6160 }6257|6514\6591 6468 6084 6162,6240)651816596 647416559 le241 6520) 639916478|6557\6656 6480'6560\664016720 6400 | {6561/6642'6725/6804 672416806/6888 A 6889/6972) 191411945/197212001|2030/12059|2088|21 17/2146]2175]220412955|2262)2291 2520 12549)2578)2407|2436)2- 2460}2490|2520|2550|2580|2 2666|2697 5576 5265|5350|5595|5460 4356/4422}4488/4554|4620]| 4686/4752) 431 8]4884]4950/501 6|5082]51 48]521 4] 5280)|5546}5412)547815544)5610 5025|5092)51 59}5226}5295}5560}}5427/5494/556115628|5695 1071}1088]1105]1 1 22/1159]1156)1173]1190)|1 207}1224)1241]1258)1275)1292)1 509) 1526 1545}1560)1577|159411 411|1428)1445|1462)147 1154(1159111 70111881 2061122411 24911 2601 1978|1296|151 4153211 550|1568]1586|1 404|1422/1 4401 458/1476|1494]151911 550)1548)156 5411 273}12991511/1530111549|1368|1587|1406|1425]1444}1 463]1452)1 501/1520)1 559)}1 558)1577/1596)1615)1654/165 1460|1480}1500)1 520)1 540)1560)1580)1600)1620]164C}1 660) 1680)1700}1 720/174 147011 491|1519]1533|1554|1575]1596|1617|1658|1659|1680)|1701|1722)1745|1764|1785|1806]182 1518|1540I11 562/1584|1606|1628}1650]1672}1694|1716|1 753)1760)|1 782) 180411 826) 1848}1870)1892|191 1.518}1541/1564]1.587|1610]}1655|1656|1679|1 70211 725)1748}1771|1 7941 81 7)1840)|1 865}1 886) 1 909]19352}1955}1978]200 29951255 5525 5676 629016564 6575|6450|6525, 6460/6 556|6612 6545|/6622/6699 6650|6708|6786 6800/6880|6960) 6885|6966|7047 6970)7052|7154 7055715: 7056 We may observe, that this table contains 14,052 digits; but had it com- menced at zero, it would have contained 16,638. In this aggregate, the several digits recur in very different proportions; of the series 0, l, 2, 5, 4, 5, 6, 7, 8 and 9, the corresponding numbers of recurrences are 2101, 1966, 2588, 1548, 1854, 1577, 1495, 1261, 1554 and 916. 415060|5096|9152)5 35108)314.5/5182)521 9/52 319215250|5268)5506 76|551 5|55.54|3595)5 5400|5440}5480!5520|5.560|5600]15640/568 3485|5.526|5567|5608|3649|5690||5751|57 72/581 5}58.5415895|5956|5977|4018 15.570)56 1 2}565415696|5758|5780]|5822|5864|5906|5948|5990) 40; 1 391561 2)5655|5698|574 6715/6794|6873)6 85 te he tee 90 y 91 7| 968 8/1252)12 51 S2O1G. 211408)14 9)1496|15 4 5 2 € 3)1672)1691|171 49|2376!2405|2430!|2457 5412499/2520)/2 81|2610}/2 4/2816 1/2904/2957 5440}5504|5568]56521569615760 6675/6750) 6764/6840 6855/6950 7051)7110) 712017200) 7128/7209|7290 7216]7298}7580 7504/7587|7470 7592)7476|7 560 7296/7482 7569 5)7480/7565/7650 |7568/7654|7740 7656/7743]7 8350) 7744;7852)7920 979} 990}1001 1068|1080]|1092)1104/1116) 15|1530)1547|1564/1581 6]1584)1602|1620) 1638)1656/1674)/1692 0|1760/1780}1800}}1 820]1840}1860}1880 7|1848]1 869]1890}/1911}1952)1953)1974 '2002!2024|2046|2068/2090)2112)2154/21 56/2178) 90 2093|2116|21359|21 62/2185|2208|2251 1512)1596]1 560}1 584|1 608}1 652}1656|1680}}1 704/1728}1752)1776]1 800)1 824/1848)1872 1896|1920}|1944|1968|1 992} 201 6/2040|2064|2088|21 1 2]21 56}2160}|21 84|2208)2252}2256)2280/2304|2528)25521937G 2 1750]]1'77.5|1800]1825]1 850|1875|1900]1925|1 950/197 5}2000)!2025}2050/207 5|21 00/21 25/21 50/21 7 5|2200/2225}2250)| 22 1820!11846|1872]1898]1924|1950}1976|2002|2028|2054)2080)|21 06/21 52)21 58/21 84)221 0/2956|2262)2988)251 4}2540 1701|1728}17.55|1782|1 809]1 856}1 863}1 890} 1917|1944]1971}1998]2025/2052/2079}2106 215521 60}12187|2214/2941/2968 4)1956/1958}1980; 1/2024|2047|2070; 5150 iS 355013567|5404)544 115478): 5420||3458)5496/5554|5572)5 71}5510)5549 5963|6050||6097 j 5780|5848|5916]5984|60526120]6188 4761]4850||4899|4968|5037|5106/5175|524415515|558215451|5520/15589|5658|5727|5796|5865|5054}6005/6072 5040|5110151 80|5250|532015590|54601555015600}15670|5740158101588015950|602016090161 60 (680}|5751|5822/5895|5964|605516106|6177 6248/6531 9|6590)|6461 '904|5976/6048/6120/61 6141/6210)/6279)6. 6250|6500)|6370 1183]1196]1209 1302) O}}1729}1748}1767|1 786) 2525|25.50|257 5124 2418/2444 4)2511)2! Z 2604|2652)2¢ - 2697|2726|2 2750|2760|2790)2820) 285028: 2821)28% 2912 5003 3069|35102)5155|51 31 290/355 }9276 3548)5584154 3815627 |566 972015760 ra 6006 7098 7189, 7280/7: 7371|7452;7555)761 7462)7544) 7626/7 7081779) \7644/7728/7819!7806}708 7735}7820) 7826/7912/7998 7917, 8008! 7921}8010}; 2885/29 1 4/2945|2976|35007|5038 4|2976|5008/3040|5072/5104 52|4074/4116 1/6784|5827|5870113915|5956|5999]4042!4085/4198|/417114914 696|5740|5784)5828)587 2/591 6|5960]|4004|4048}4092] 41 56/41 80/4224) 496814519 5}5690/5755|5780|5825|3870|291 5|3960|4005|4050]/4095}41 40)41 85|4230|497514520143651441 014, 1154.50]5496|554215588|365415680}5726157 721581 8158641591 015956|4002!4048/4094]41 40/141 86|4252)4278]4524| 45701441 6|446214508 5.525|5572|561 9}5666|5713}5760||5807}585415901)5948)5995|4042|4089141 56/41 85|4250/|4277|452414571/441 814465/451914559|4606 552|3600|564815696|5744|5792|5840}5888/5956/598 4/4052) 4080]41 28|4176|42941497914590|14568/441 61446414512 4560/4608/4656|4704 31 85/3234|5285/95521958115450113479|55281957'7|9626|967 515724157 75|5822|587 1|5920)|5969)401 8/406 7/41 16/41 65/421 4|4263]451 9145614410 [2401|24.50/}24991954819597|2646]2695/27441279512849/2891|2940/2989|5058|3087151 3613 4459)4508 |4557|4606)4655|4704 57 50!5800|5850|5900/5950|4000}/4050/4100)41 50}4200/4250/4500]4550)4400)4.4.50|4500))4550|4600'46.50}47001475014800 5417154681551 915570115621 19672/5725/9774|0825|5876|5927|5978]4029| 4080/41 51/41 824955) 498414555}4586|4457|448814 55914590 5796|5848}5900|5952/4004] 4056/41 08) 41 60/)4212)426 4/45 16)4568}4490|4472)4524)4576|4628|4680)|4752|4784|4856/4888|4040/499915044 5551|5604|3657|571015765|981 6|5869|5922|597 5|4028]4081|4154|4187|4240||4295/4546)459914459!4505/4558|4611/4664/471714770)4825|4876|4929| 4989] - 55641361 8|567215726|5'780|5854|5888|5942|5996|4050!41 04/41 58/421 214266/4520!/4574/4428/4489)4556145901464414698]47 521480614 860]491 4|4968150221507615 5.5'75|5635013685|57401579.519850115905|5960|401 5|4070/4125]41 80/425 5}4290}4545)4400) 4455 4510 4565]4620)467 5|4750|47 85/48 40/489 5/4950] 5005|5060|5115|51 7015 53601541 61547915598155841564015696|9752'58081586415920||3976|4052|4088|41 44|4200]49.56|451 2}4568/4424|4480}]4556/4592)4648) 470414760)481 6/4872}49281498415040115096|51 5215208)5264| 55 5'705|5762|581915876|5955|5990]|4047/41 04141 61|421 8427 5]4552|4589}4446|4.505]4560}]461 7/467 4/475114788]4845|4902/4959|501 6|50751/5150/151 87/5244|5501/5558) 54) 5828|5886|5944|4002/4060)/41 18]41 76|4254}4292/4550]4408/4466|4524|4582)4640}/4698/4756)48] 4|487914950]4988|5046151 04/51 6215290)527815556 g 5835|5894|5955/401 2}4071|41 501141 89|4248|4507|4566|4425]4484|4545] 4602/4661 |4720)]47 79/4858}4897| 49561501 51507415195151991595)1551011556915428): 3900|3960|4020/4080}41 40/4200]]4260]4520/4580]4440|4500) 4.560|4620]4680)4740}4800]}4860/4920/4980) 5040)5100}5160}522015280153401540011546015520 4941)5002|5065)51 24151 85|5246|5507|5568)5420|54901|5551156121: 5084151 46)5208|527015552|5594154.561551 8155801156421570. F 5105}5166/5229152991555515418|54811554415607|56701|573515796|585915929| soesleo46 6111)6174'5 556 5824|5888|595216016\608016144\62081627216. 5590|5655)572015785|5850}]591 5|5980|604.5|6110)6175 5742)5808/5874|5940) 5762)5829|5896 4641 |4692/47435)4794)/4845)4896 55|5088)5141 2/1164'1176!1188] 19 1261|1274|1987| 15 1358|1372/1586) 14 1455/147011485) 15 1552/1568|15841 1¢ 1649|1666|1683] 17 1746|1764|1 7881 18 1243/1862/1881| 19 1940|1960)19801 99 2037|2058]2079} 91 22541227] 9s 00)2425/2450/247 os 2522/254812574) og 2619}264612673| o7 2716|2744127791 5g 2815/2849|987)). 80}2910}2940)2970) ay 5069) 5) 5156|5168) 59 : 5267) 35 68|5' 645) 4947/4998|5 5890}5952!601 4|6076|61358 7695}77 76/7857 17 7954|S036)811 7555; Eh le ale 7885,79681805118134|5 21" 8064/81 49/8939 8281/8572 8465 8554!s64s\8736\ggo7 8464 8556 8648/8740 8839! gq04\c016 9108 8649 8742!8e55|soa8 8856/8950!5024 2 5}91 9216 {9021 9114/9909 \9118/921 2195, 9215/9353 0ilo4Q 9512/9408) 9504) 9409/9506/o6 6; 9604] 97, 80 |) 81 |] 82 | 85] 84] 85 | 86 oda bel ths 90 » 91 | 92] 93, 94] 95] 96 97 | 98 | 99 | sso] 891] 902} 915} 924] 955] 946] 957} 968} 979] 990}}1001|1012)1023}103411045]1056/1067|1078!1089 960|| 972} 984} 996/1008}1020}1052)1044)10.56)1068)1080)|1092)1104)1116/1128)1140/1152/1164/1176'1188 1040/]1055|1066]1079}1092)1105]1118)1151}1144)1157/1170)}1185}1196)1209/1 222)1935}1248]1261 127411987 \1120}}1154/1148/1162)1176)1190)1204)1218)1252)1246)1260)/1274)1288)1502)1516)13530)1544}1558/157211586 1200}}1215]1230}124.5|1 260/127 5]1290]13505}1520}133.5]13550}1565}1580}139.5}1410/1425/1440)/1455)147011485 1280}}1296]1512)1528/1544)1560)1576/1592)1 4081 424)1440/1456)1472)1488)1 504/1 520/1.536}1552)1568/1584 1560)]13577|1594/1 41 1|1428)1 445)1462)1479}1496/1 515}1530}1 547/1 564/1 581)1598}161 5|1652)1649) 1666/1685 '1440]|14.58]1476|1 4941151211 550}1 5481 566]1 584|1 602|1620]1658]1656|1674|1692}1710|1728|1746 1764/1782 15201 539]1 558}1577|1596]161 5}1634|1653)1672)/1691|1710}}1 729|1748}1767|1786)1 805}1824/1845/1862!1881 '1600)/1620}164C}1660)1680)1700}1'720}1740}1 760|1'780}1 800}}1 820}1840}1 860)1880}1900}1920)1940)1960]1 980). |1680||1'701]1722)1745|1764|1785|1806|1827|1848}1 869]1890}}1911}1952/1953)1974)1995}2016|2037|2058|2079] . |1760}]1'782/1804)1826/1848]1870)1892/1914/1956/1958/1980}2002/2024/2046/2068/2090|21 12/2154\21 56/2178 1840)|1863}1886]1 909}19352}19.55}1978|2001|202412047|2070)//2093)21 1 6|2159|21 62/21 85|/2208|2231|2254laa77 '1920)|1944|1968}1 992) 201 6|2040/2064|2088)21 1 2}21 36/21 60/21 84|2208/2252)225612280|2304|2328/2352103576 2000}|2025}2050/207 5|2100)21 25/21 50|2175}2200/222.5|22.50}1227 5|2500|2525|25 501237 5|2400)242.5/2450\24'75 _.2080)|2106}2152)21 58/21 841221 0/2256/2262/2288)23 1 4|2540112566|2592)24 1 8)2444/2470/2496/2522/254810574 “2160}12187|221 4/2241 |2968,22995|252912549|23576|24035/2430}|2457|2484/251 112558|2565|2592|2619/2646|2675 2240}|2268|2296|2524|25.52/258012408|2456|24641249912520112548]2576|260412652|2660|2688/2716|2744/2779 2520)|2549)2578/2407|2456|2465|2494/2525/2552/2581|26 1 0i2639|2668|2697|2726|27 55|2784|2815|2849\987) 2400)|2450!7460[2490) 2520/25 50/2 5801261 O26 401967197 901127 50127 EM2LTANIPS IMP LED/ IRAN 291 012940|9 970) \ 06 68 IT THE PHILOSOPHY or ARITHMETIC; EXHIBITING A PROGRESSIVE VIEW OF THE THEORY AND PRACTICE OF CALCULATION, WITH AN ENLARGED TABLE OF THE PRODUCTS OF NUMBERS UNDER ONE HUNDRED. BY JOHN LESLIE, F.R.S. E. PROFESSOR OF MATHEMATICS IN THE UNIVERSITY OF EDINBURGH. EDINBURGH: Printed by Abernethy & Walker, FOR ARCHIBALD CONSTABLE AND COMPANY, EDINBURGH ; AND LONGMAN, HURST, REES, ORME, AND BROWN LONDON. 1817. SMH TESORO HHS te pA: jaan. ier ’ As HJ > ky P f : An Oy pent} ¢} SR ORD Ta a fee ep ) aes iyhts Ny & ¥ panies / Va AW i be REP | Fvliyt x . ‘ We ee Cel A vs i rye aT ne ei Haein yy. ey ve ay . be r J ay ne mere PALIOTAD Ww #31 TOA ‘dam a * aie a 7 ke 9 ci dis ar SAITO Ti ag wir ip ER 1k ange | - RS. ES nA a atts i Peet eee Hi: 3 aie a at e St TR AION a Sepa * oa. hey ns ven voir: Herons $ ae ae “Riy) ee Uer ae 1 aonalire P rit Nay 2 aa P Me ee yi “ sepanulatehtiall a x J ‘ss hee 4 ae Te es SOR SECT i oc aaa vaduamtag, er 6 la Svknrens, sin gica Bek sy Seite, ara, ‘Pais ay Cee ey une oan ia ME a hy 5¢ é “DL, 2 on “« ; MATHEMALICS LIBRARY PREFACE. I wow discharge my promise, by the publication of a volume, in which Arithmetic is deduced from its principles, and treated as a branch of liberal _. education. The object proposed was not merely *-to teach the rules of calculation, but to train the iC young student to the invaluable habit of close ~ and patient investigation. I have therefore pre- ferred the analytical mode of advancing, and have pursued a route entirely different from that which is followed by the common treatises of arithmetic. In seeking to unfold the natural progress of dis- covery, I have traced the science of numbers, through the succession of ages, from its early germs, till it acquired the strength and expansion of full maturity. ‘This species of history, combin- BLO ing solid instruction with curious details, cannot fail to engage the attention of inquisitive readers. If the execution of this little work, which 1s the | result of very considerable research, should at all correspond with the importance of its design, it may supply a capital defect in our systems of pu- 349756 1V PREFACE. blic instruction. A great portion of the materials which compose it has already been given to the world, through the wide circulation of the Supple- ment to the Encyclopedia Britannica; but the distinct and improved form in which it now ap- pears is alone suited for general use. Some per- | sons will perhaps complain. that it takes a wider scope than might answer all the common objects of tuition. But there is yet no royal road to knowledge, and whatever is acquired without ef- fort becomes quickly effaced from the memory. Nothing indeed can be more fallacious than to ex- pect any solid or lasting advantage from the sub- stitution of a concise and mechanical procedure. The time required for the study of this treatise could scarcely be more beneficially employed, since it will not only rivet in the mind of the pupil the theory and enlarged practice of calculation, but invigorate, by proper exercise, his reasoning faculties, and consequently prepare him either for entering the labyrinths of business, or engaging in the higher pursuits of science. CoLLEeGE oF EDINBURGH, 15th October 1817. INTRODUCTION. ray I HE idea of number, though not the most easily ac- quired, remounts to the earliest epochs of society, and must be nearly coéval with the formation of language. The very savage, who draws from the practice of fishing or hunt- ing a precarious support for himself and family, is eager, on his return home, to count over the produce of his toil- some exertions. But the leader of a troop is obliged to carry farther his skill in numeration. He prepares to at- tack a rival tribe, by marshalling his followers ; and, af- ter the bloody conflict is over, he reckons up the slain, and marks his unhappy and devoted captives. If the numbers were small, they could easily be represented by very portable emblems, by round pebbles, by dwarf-shells, by fine nuts, by hard grains, by small beans, or by knots tied on a string. But to express the larger numbers, it became necessary, for the sake of distinctness, to place those little objects or counters in regular rows, which the eye could comprehend at a single glance; as, in the actual telling of money, it would soon have become customary to dispose the rude counters, in two, three, four, or more B 2 INTRODUCTION. ranks, according as circumstances might suggest. The at- tention of the reckoner would then be less distracted, rest- ing chiefly on the number of marks presented by each se- parate row. , Language insensibly moulds itself to our wants. But it was impossible to furnish a name for each particular num- ber: No invention could supply such a multitude of words as would be required, and no memory could ever retain them. The only practical mode of proceeding, was to have recourse, as on other occasions, to the powers of CLASSIFI- CATION. By conceiving the individuals of a mass to be dis- tributed into successive ranks and divisions, a few compo- nent terms might be made sufficient to express the whole. We may discern around us traces of the progress of nu- meration, through all its gradations. The earliest and simplest mode of reckoning was by pairs, arising naturally from the circumstance of both hands being employed in it, for the sake of expedition. It is now familiar among sportsmen, who use the names of brace and couple, words that signify pazring or yoking.—To count by ¢hrees was another step, though not practised to an equal extent. It has been preserved, however, by the same class of men, under the term /eash, meaning the strings by which three dogs and no more can be held at once in the hand. — The numbering by fours, has had a more extensive appli- cation: It was evidently suggested by the custom of taking, in the rapid tale of objects, a pair in each hand, Our fishermen, who generally count in this way, call every double pair of herrings, for instance, a throw or cast ; and the term warp, which, from its German origin, has exact- ly the same import, is employed to denote our, in various articles of trade. INTRODUCTION. g These simple arrangements would, on their first appli- cation, carry the power of reckoning but a very little way. To express larger numbers, it became necessary to renew the process of classification ; and the ordinary steps by which language ascends from particular to general objects, might point out the right path of proceeding. A collec- tion of individuals forms a species ; a cluster of species makes a genus ; a bundle of genera composes an order ; a group of orders constitutes a class; and an aggregation of classes may complete a kingdom. Such is the method indispensably required in framing the successive distribu- tion of the almost unbounded subjects of Natural History. In following out the classification of numbers, it seemed easy and natural, after the first step had been made, to re- peat the same procedure. If a heap of pebbles were dis- posed in certain rows, it would evidently facilitate their enumeration, to break down each of those rows into simi- Jar parcels, and thus carry forward the successive subdivi- sion till it stopped. ‘The heap, so analysed by a series of par- titions, might then be expressed with a very few low num- bers, capable of being distinctly retained. The particular system adopted for this decomposition would soon become clothed in terms borrowed from the vernacular idiom. Let us endeavour to trace the steps by which a child or a savage, prompted by native curiosity, would proceed in classing, for instance, twenty-three similar objects.— 1. He might be conceived to arrange them by succes- sive pairs. Selecting twenty-three of the smallest shells or grains he could find, he might dis- pose these in /wo rows, containing each eleven counters, and one over. Having thus reduced the number to eleven, he might sub- eo. © '@ Ste o 6 Oe 6 6 o oe @e @ie € @ 6 G8, @ 4 INTRODUCTION. divide this again, by representing only one of the rows, with shells ¢wice as large as before. He would consequently obtain two rows of *° ® ® © eeeeoe e@ Jive each, with an excess of one. Instead of these shells, were he to employ shells of a double size, it would be sufficient to denote one of the rows, @ @ or to dispose it into worows. ‘These rowscon- @®@ @ tain each only ¢wo counters, with one remaining. Again, by adopting counters of a double size, the last row C3) might be represented by one pair, each contain- @ ing only two marks exactly. These again could be deno- ted by a single counter, of fwzce the former di- S mensions.—Hence the number d/wenty-three, as decomposed by repeated pairing, would be denoted by one counter of the fifth order, one of the third, one @ of the second, and another of the first. ® ® 2. Again, suppose a person should attempt to represent the same number, byzrzple rows of shells or counters. He would first have seven bce of the smallest shells in each row, with ....... s @¢ an excess of two. Then, expressing one of the rows by shells of three times the size or value, ee it would be again resolved into ¢hree ats i: rows, each containing ¢wo counters, with an excess of one. But these rows might be repre- © sented by /wo counters of triple size ; e and here the decomposition stops. The number éwen- ty-three is thus expressed, on the sys- tem of ¢riplication, by two counters of @@ the third order, ove of the second, and ° two of the first. oN oi INTRODUCTION. 3. Lastly, conceive the number ¢wenty- ++ °° ° threeto be reckoned by double pairs or qua- *'* * °° druple rows. Each row wouldnow contain °° * * ° only five counters, with an excess of three. But a single row would express the same as all these four, if each counter in it were changed into ano- @ 1G :@. [6 2S ee * ther of quadruple size or value; and, conse- quently, this row might be again distributed @ © into four ranks, each consisting of @ single mark, with one left. Retain one of these ranks, and substi- tute a counter of quadruple effect, which will ex- @ press the whole amount. Hence twenty-three, analysed by the system of double pairs or warps, would @ be represented by ove counter of the third or- e der, one of the second, and three of the first °° It is easy to conceive how, in other cases, the process of decomposition could be carried forward. In the ruder periods of society, a gradation of counters, accommodated to such a process of numerical analysis, was supplied by grains, beans, pebbles, or shells of differ- ent sizes. ‘The series of such natural emblems, however, is very limited, and would soon confine the range of de- composition. To obtain a greater extent, it was neces- sary to proceed by a swifter analysis; to distribute the counters, for instance, successively into ten or twenty rows, and to make pebbles, shells, or other marks, having their size only doubled perhaps or tripled, to represent values increased ten or twenty fold. Beyond this stage in the progress of numeration, none of the various tribes disper- sed over the vast American Continent seem ever to have passed. In the Old World, it is probable that a long pause of improvement had ensued among the nations 6 INTRODUCTION. which were advanced to the same point in the arts of life. But the necessity, in such arithmetical notation, of em- ploying the natural objects to signify a great deal more than their relative size imports, would lead at last to a most important step in the ascent. Instead of distinguishing the different orders of counters by their magnitude, they might be made to derive an artificial value from their rank alone. It would be sufficient, for that purpose, to employ marks all of the same kind, only disposed on a graduating series of vertical bars or columns. The augmented value which these marks acquire in rising through the successive bars, would evidently be quite arbitrary, depending, in every case, on a key to be fixed by convention. ‘This point in the chain of discovery was attained by the Greeks at a very early period, and communicated to the Romans, who continued, during their whole career of empire, to practise a sort of tangible arithmetic, which they trans- mitted to their successors in modern Europe. Such humble expedients might suffice for all the compu- tations required in the ruder periods of society. But men were insensibly led to frame permanent symbols to denote numbers by that feeling, which unceasingly prompts them to seek the approbation and applause of their fellows. The aspiring leader of a successful band, or the petty legislator of a rising community, is anxious to preserve the memory of the exploits he performed, or the be- nefits he ‘conferred. He is not content with obtaining the applause of his contemporaries; this fleeting exist- ence is insufficient to fill his imagination; he looks anxiously beyond the grave, and sighs for the admi- ration of generations yet unborn. Hence the anxiety among all people to erect monuments of high achieve- INTRODUCTION. 7 ments or illustrious characters. In the early periods of society, a vast mound of earth, or a huge block of stone, was the only memorial of any great event. But, after the simpler arts came to be known, efforts were made to trans- mit to posterity the representations of the objects them- selves. Sculptures of the humblest kind occur on monu- mental stones in all parts of the world, sufficient to convey tolerably distinct images of the usual occupation and em- ployments of the personages so commemorated. The next step in the progress of ‘society was to reduce and abridge those rude sculptures, and thence form combina- tions of figuresapproaching to the hieroglyphical characters. At this epoch of improvement, the first attempts to repre- sent numerals would be made. Instead of repeating the same objects, it was an obvious contrivance to annex to the mere individual the simpler marks of such repetition. ‘Those marks would of necessity be suited to the nature of the materials on which they were inscribed, and the quality of the instruments employed to trace them. In the histori- cal representations, for instance, which the Mexicans and certain Tartar hordes painted on skins, a small coloured circle, as exhibiting the original counter, shell, or pebble, was repeated to denote numbers. But, on the Egyptian Obelisks, the lower numerals at least, are expressed by combined strokes. None but straight lines, indeed, are fitted for being carved on pillars of stone, or cut on the face of wooden posts. Even after the use of Hieroglyphics had been laid aside, and the artificial system of alphabetic characters adopted, the rectilineal forms were still prefer- red ; as evidently appears in the Greek and Roman capi- tals, which, being originally of the lapidary sort, are much older than the small or current letters. The Runic let: 8 INTRODUCTION. ters, in which the Northern Languages of Europe were at first engraved, consist almost entirely of simple strokes in- serted at different angles. The primary numeral traces may, therefore, be regard- ed as the commencement of a philosophical and universal character, drawn from nature itself, and alike intelligible to all ages and nations. ‘They are still preserved, with very little change, in the Roman notation ; consisting of simple strokes variously combined, which were imported perhaps before the adoption of the alphabet itself by the Grecian colonies that settled in Italy, and gave rise to the Latin name and commonwealth. The lower classifica- tions of numbers had gradually fallen into disuse, and gi- ven place to the more expeditious and convenient system of advancing by successive tens, which arose very naturally indeed from the practice common among rude people, of counting by their fingers on both hands. Assuming, therefore, a perpendicular line | to signify one, another such {| would express ¢wo, the junction of a third ||| three, and so repeatedly till the reckoner had reached ¢en. The first class was now completed, and to intimate this, he threw a dash across the stroke or common unit ; that is, he employed two decussating strokes X to denote ten. He next repeated this mark, to express twenty, thir- ty, and so forth, till he had finished the second class of numbers. Arrived at an hundred, he would signify it, by joining another dash to the mark for ten, or by merely connecting three strokes thus L_- Again, the same spirit of invention might lead him to repeat this character, in denoting two hundred, three hundred, and so forth, till the third class was completed. 4 thousand, which begins the fourth class on the ordinary mode of numeration, was INTRODUCTION. 9 therefore expressed by four combined strokes M); and this was the utmost length to which the Romans first pro- ceeded by direct notation. But the division of these marks afterwards furnished characters for the intermediate numbers, and thence great- ly shortened the repetition of the lower ones. ‘Thus, ha- ving parted in the middle the two decussating strokes \X denoting ¢en, either the under half A, or the upper half \/; was employed to signify five. Next, the mark [7 for an hundred, consisting of a triple stroke, was largely divided into [~ and |_, either of which represented ji/iy, Again, the four combined strokes (Mj, which originally formed the character for a thousand, came afterwards, in the progress of the arts, to assume a round shape {¥), fre- quently expressed thus C] >, by two disparted semicir- cles divided by a diameter: This last form, by abbrevia- tion on either side, gave two portions C| and |7) to re- present five hundred. The process appears easy, therefore, to devise an universal character for expressing numbers. But it was a very dif- ferent task, to reduce the exhibition of language in gene- ral to such concise philosophical principles. This attempt seems accordingly to have been early abandoned by all nations, except the Chinese. ‘The inestimable advantage of uniting together the whole human race, in spite of the di- versity of tongues, by the same permanent system of com- munication, was sacrificed for the simpler attainment, of re- presenting, by artificial signs, those elementary and fugi- tive sounds, into which the words of each particular dia- lect could be resolved. Hence the ALPHABET was invent- ed, which, notwithstanding its obvious (fects, must ever be regarded as the finest and happiest effort of genius. i0 INTRODUCTION. More letters were afterwards added in succession, as the analysis of the primary sounds became more complete; but the alphabet had very nearly attained its present form, at the period when the Roman commonwealth was extending its usurpation over Italy. About that epoch, a sort of re- action seems to have arisen between the artificial and the natural systems, and the numeral strokes were finally dis- placed by such alphabetic characters as most resembled them. The ancient Romans employed the letter I, to re- present the single stroke or mark for one ; they selected the letter V, since it resembles the upper half of the two de- cussating strokes, or symbol for five ; the letter X exactly depicted the doubled mark for ten ; again, the letter L was adopted, as resembling the divided symbol for, /fty ; while the entire symbol, or the tripled stroke, denoting an hun- dred, was exhibited by the hollow square [~, the origi- nal form of the letter C before it became rounded over. The quadrupled stroke for a thousand was distinctly repre- sented by the letter M, and its variety by the compound character clo, consisting of the letter I inclosed on both sides by C and by the same letter reversed; the latter portion of this again, or Io, being condensed into the let- ter D, expressed jive hundred. The Greeks, after having communicated to the founders of Rome the elements of the numeral characters, which are still preserved, again exercised their inventive genius in framing new systems of notation. Discarding the sims ple original strokes, they sought to draw materials of con- struction from their extended alphabet. ‘They had no fewer than three different modes of proceeding. 1. The letters of the alphabet, in,their natural succession, were employed by them to signify the smaller ordinal numbers. In this ‘i. — —— - INTRODUCTION. 11 way, for instance, the books of Homer’s Jdad and Odyssey are usually marked. 2. The first letters of the words for numerals were adopted as abbreviated symbols. A sim- ple and ingenious device was used for augmenting the powers of those symbols: any letter inclosed by a line on each side, and another drawn over the top thus ||, being made to signify five thousand times more. 3. But a mighty stride was afterwards made in numerical notation by the Greeks, when they distributed the twenty-four letters of their alphabet into three classes, corresponding to units, tens, and hundreds. ‘To complete the symbols for all the nine digits, an additional appropriate character was intro- duced in each class. This beautiful system was vastly superior in clearness and simplicity to the combinations of strokes, retained by the Romans, and transmitted by them to the nations of modern Europe. It was even tolerably fitted as an instru- ment of calculation, to which the Roman numerals were totally inapplicable. The Greek notation proceeded directly as far as nine hundred and ninety-nine ; but, by subscribing an 76a or short dash under any character, its value was augmented a thousand fold; or by writing the initial letter of my- riad, the effect was increased still ten times more. With the help of punctuated letters, therefore, it reached to ten thousand, comprising four terms of our ordinary scale; and by means of the subscribed m, it was carried over another similar period, or fitted to express a hundred mil- lion. But the penetrating genius of Archimedes quick- ly discerned the powers, and unfolded the properties, of such progressions. He took the square of the limit of the common numeral system, or an hundred millions, being ten- lv INTRODUCTION. thousand times ten-thousand, for the index of a new scale of arrangement, which therefore advanced by strides eight times faster than the simple denary notation. ‘This com- prehensive series he proposed to carry as far as eight pe- riods, which would hence correspond to a number express- ed in our mode by siz/y-four digits. Apollonius, who, next to the Sicilian philosopher, was the most ingenious and inventive of all the ancient mathe- maticians, resumed that scheme of numeration which had been suffered to lie neglected, simplified the construction ofits scale, and reduced it to acommodious practice. For greater convenience, he preferred the simple myriad as the root of the system, which therefore proceeded by successive periods, corresponding to four of our digits. These periods were distinguished by breaks or blanks. The learned and profound astronomer Ptolemy modi- fied this system in its descending range, by applying it to the sexagesimal subdivisions of the lines inscribed in a cir- cle. He likewise advanced a most important step, by em- ploying a small or accentuated 0, to supply the place of any number wanting in the order of progression. The Arithmetic of the Greeks, thus successively mould- ed by the ingenuity of their great Geometers, had attain- ed a singular degree of perfection, and was capable, notwithstanding its cumbrous structure, of performing operations of very considerable difficulty and magnitude. But those masters of science, rich in their mental re- sources, overlooked the advantages resulting from a sim- pler mode of arrangement. They had only to ascend more slowly, and proceed by Zens instead of periods of myriads ; that is, to retain as numerals no more than the first set of their alphabetic characters, which were already INTRODUCTION. 13 ae | employed with a point or a short dash subscribed to denote thousands. ‘This might seem an easy step in the progress of invention, but the current of ideas had already flowed beyond it. Nor during the ebbing tide that preceded the fatal extinction of science among the Greeks, was any farther simplification effected, which would have shed 2 pale ray over the evening of that Philosophy which was again destined to emerge from the thickest darkness, and relume the world. For the knowledge of our system of | elementary numerals, which may be justly styled the Al- phabet of Arithmetic, we are indebted to a people extreme- ly inferior to those instructors of mankind, in genius, acuteness, and general energy of character. Whether the Hindis lighted on that happy contrivance themselves, or derived it from their communication with the natives of Up- per Asia, there is yet no sufficient evidence to decide. They seem, however, to have become acquainted with it nearly two thousand years ago, and to have thenceforth common- ly employed that mode of notation. T’rom the Hindis, again, their Arabian conquerors appear, about the ninth century of our zera, to have received an improvement at once so simple andimportant. ‘These industrious cultivaters of science afterwards imparted the valuable present to their countrymen the Moors, who still occupied the finest por- tion of Spain. From this centre, it was gradually com- municated over Europe. ‘The earliest traces of the nu- meral characters among the Christians may be referred to the end of the thirteenth century. They were at first in- troduced only into almanacs and astronomical tables; but their great convenience soon brought them into more gene- raluse. Originally those characters were somewhat differ- _ ently shaped from the present; they had attained, however, 14 INTRODUCTION. their actual forms not long after the invention of the no- ble art of printing, or about the commencement of the sixteenth century. ‘The digits were now adopted almost universally throughout Europe into the practice of Arith- metic. Yet vestiges still remained of the performing of nu- merical operations by means of counters and other palpa- ble emblems. Nor is this rude mode of computation quite so contemptible as would at first appear, since the Chi- nese continue to employ it with success in all their mer- cantile transactions. At any rate, it deserves attention, both for its connection with the history of science, and its im- portance in illustrating the properties of numbers, and explaining the grounds of calculation. We shall therefore view an1ITHMETIC under two very dis- tinct forms, that would require separate appellations. 1. Pal- pable Arithmetic, in which the numbers are exhibited by counters, or abbreviated representatives of the objects them- selves; and, 2. Zigurate Arithmetic, in which the numbers are denoted by help of certain symbols, or artificial charac- ters, disposed after a particular order. The progress of Arithmetic is analogous to that of writing, but it has fol- lowed the advances and transitions of this sublime art at a great distance. ‘The numeration by counters, balls or strokes, evidently resembles hieroglyphics or picture wri- ting; while the invention of the alphabet, so happily con- trived for the rapid transmission of thought, probably led the way to the subsequent discovery of the science of 27- gurate Arithmetic, founded nearly on similar principles. These capital divisions of Arithmetic we shall consider in succession, pointing out the application of each mode ta Numeration, and to the several depending operations of Addition, Subtraction, Multiplication, and Division. PALPABLE ARITHMETIC. NUMERATION. Suppose the objects to be reckoned were so numerous, that eighty-six counters might be required to represent them. Placed in a single row, these counters would only give the very confused idea of multitude. But, if counted by pairs, or divided into two rows of forty-three each, they would become a little more distinct. Were every counter now, in each row, to denote a pair, a single row of them would have the power of both. Let this row be reckoned again by pairs, and it will change into two higher rows, each consisting of twenty-one counters, with an excess of ove. But one of these ¢hird rows would be sufficient alone, if each counter in it were esteemed equal to a pair in the second rows, or equal to a duplicate pati’, or four in the frst row. Again, tell one of the third rows over by pairs, and the ¢wenty-one counters will be convert- ed into two fourth rows, containing ten each, and one over. In like manner, each counter of a fourth row, being con- _ ceived equal to a pair in the ¢hird row, or a triplicate pair, _ that is, eight on the first row ; a single row, including tex _ of those higher counters, would have the same effect. Now these ten would be reduced to a pair of rows of five counters each. Let each counter in the f/th row have the 16 PALPABLE [NUMERATION,, power of two in the fourth row, or of a quadruplicate pair, or sixteen in the first row ; and five such counters would be sufficient. But Jive would give two pairs of sixth rows, one of which might denote the whole, if each counter in it were held equal to wo in the preceding row, or to a quintuplicate pair, or thirty-two on the first row. Again, the last wo counters would be divided into a single pair of a higher order. This analysis appears tedious when so detailed, but it would proceed with great ease and rapidity in practice. The number eighty-six would, therefore, on the system of successive pairing, be expressed by one sextuplicate pair, one quadruplicate pair, one duplicate pair, and one single pair. The language appears very uncouth, merely from its novelty and inaptness to our idiom; but its elements are extremely clear and simple. If a few cognate words had been devised to express the several combinations of pairs, or the ascending scale of the powers of two, it would have removed every objection. This arrangement, whereby a number is analysed into certain elements by the operation of distributing it and its sections into successive pairs or duads, may be called the Binary Scale, of which two is the root or index. This scale, resting on so narrow a basis, expands slowly, and is therefore not very fit for expressing large numbers by words. But it is well adapted for the simplicity of emble-_ matic exhibition. Suppose marks or counters were placed in perpendicular rows or parallel bars, proceeding from the right hand to the left, such that a counter on any bar should be equivalent to ¢wo laid on the bar immediately — below it. Instead of putting the ezghty-six counters on the first bar, it would be the same thing, to place forty- NUMERATION. ] ARITHMETIC. 17 three on the second bar; the effect here again would be the same as to leave one, and drop twenty-one on the third bar. Counting these twenty-one also by pairs, we should place one on the third bar, and carry zen to the next. But ten, divided into pairs, would leave the fourth bar vacant, and throw jive to the next. The decomposition thus effected would appear as below; where only four counters and three blanks are sufficient to exhibit the num- ber etghty-siz. By this elementary arrangement a very distinct idea is conveyed: BINARY SCALE. The eye caneasilycatch the | eign picture, and the memory od ® & preserve it. A similar effect would be produced, though much less clearly, by the combination of strokes, like the Runic sculp= ture, dots being employed to indicate the blanks or va- cant terms. Seem ee Pa eh Sk Pee It is obvious, that this very simple scale would require only one set of marks. ‘These on the ascending bars sig- nify two, four, eight, sixteen, &c.; and the number now expressed is divided into sixty-four, sixteen, four and two. _ Some feeble traces of the Binary Notation are found in the early monuments of China. Fount, the first Emperor and the founder of that vast monarchy, is venerated in the East as a promoter of Geometry; and the inventor of a science, of which the knowledge has been since lost. The emblem of this occult science appears to consist of eight separate clusters of three parallel lines or ¢rigrams, drawn one above another after the Chinese manner of writing, and represented either‘entire or broken in the middle. Those varied trigrams were called Koud or suspended sym- bols, from the circumstance of their being exposed in the c is PALPABLE { NUMERATION.- public places. In the composition of such clusters we - may perceive the application of the Binary Scale, carried only to three ranks, or as far as the number eight. The entire lines signify one, two or four, according to their or- der, while the broken lines are void, and serve merely to indicate the rank of the others. The Ternary Scale of numeration, which reckons by suc- cessive threes or triads, advances with more speed. ‘Thus, suppose, as before, that ezghty-stx were to be exhibited on it. Counted by ¢hrees, or, in the sportsman’s phrase, by leashes, two counters would be left on the first bar, and dwenty-eight thrown to the next bar, or that of the simple triads. ‘These twenty-eight counters, being again told by threes, would leave one on the second bar, and carry nine to the third bar, or that of duplicate triads. The nine reckoned by ¢hree in succession, would now pass over the third and fourth bars, and throw one mark to the sixth bar, or that of quintuple triads. ‘The original number so decomposed, might therefore be de- nominated one quintuple triad, one | | single triad, andtwo. Itisdenoted @ @ @ a TERNARY SCALE. by four counters, as in the mode here | | annexed. ‘The number might like- | wise be readily expressed, though less perfectly, by combi- ned strokes and points. | miko, “phig' Mets Wha aoe Ae It is apparent that the Zernary Scale, though more power- fal than the Binary, requires two sets of marks or coun-. ters. .In the example now taken, each counter on the ascending bars represents three, nine, twenty-seven, ov eighty-one; and the number itself is consequently divided | into one e7ghty-one, one three and two units. NUMERATION. ] ARITHMETIC. 19 Let still the same number be arranged on the Quater- nary Scale, which proceeds by Fours or Tetrads. Lighty- _siz, told over by double pairs or warps, would leave two counters on the first bar, and carry twenty-one to the next. This twenty-one again, reckoned by warps or throws, would drop one counter on the second bar, and transfer jive counters to the third bar. The five being now count- ed, would leave one counter on this bar, and carry one to the next. The original number QUATERNARY would therefore be described as con- SCALE. taining one triplicate tetrads, one dupli- gy @ @ cate tetrads, one tetrad and two; and it | | 3 would be designated in this manner : -@6-3— Or, if the less satisfactory mode of strokes were em- ployed, one hundred and sixty-five would — — — — be thus exhibited : i The original number is thus analysed into once sicty- four, once sixteen, four and two.—Three sets of counters would evidently be required to fit this scale for its applica- tion. | The Quaternary Scale may be considered as a duplica- tion of the Binary, each bar of the former comprising two bars of the latter. It is alleged that the Guaranis and Lulos, two of the very lowest races of savages which inha- bit the boundless forests of South America, count only by fours; at least that they express the number jive by four and one, six by four and two, and so forth. We may likewise infer from a passage in Aristotle, that a certain tribe of _Thracians were accustomed to use the quaternary scale of numeration. If such was the historical fact, that simple ‘race must have not advanced beyond the early practice of reckoning successively by casts or warps. 20 PALPABLE [ NUMERATION. To: rise a step farther, let the same number be re-— presented on the Quinary Scale, which reckons by the se- ries of fives or pentads. Classed in this way, it would leave one on the first bar, and throw seventeen counters to the second. ‘Told over ae 2UINARY SCALE, gain, it would leave éwo counters on the # ®@ 6 second bar, and carry three to the third. deca | Lighty-six would therefore be denomi- i | nated three duplicate pentads, two single i | pentads, and one. It would thus be denoted: This number might also be exhibited on the same scale by a combination of —§ — strokes, | zy By this classification, the number eighty-six is divided into three twenty-/ives, two fives, and one.—The root or in~ — dex of the scale being five, it would require four sets of counters to adapt it for practice. The first bar of the Quénary Scale is actually used in this country among wholesale traders. In reckoning ar- — ticles delivered at the warehouse, the person who takes — charge of the Zale, having traced a long horizontal line, — continues to draw, alternately Ii i above and below it, a warp or Jour vertical strokes, each he of WTA TY which he crosses by an oblique score, and calls out tally! as often as the number jive is completed. The Quinary system has its foundation in nature, being evidently derived from the practice of counting over the fingers of one hand. It appears accordingly, at a certain stage of society, to have been adopted among different na- tions. Thus, the Omaguas and the Zamucas of South America reckon generally by fives, which they call hands. NUMERATION.] ARITHMETIC. 21 The Toupinambos, a most ferocious and warlike race that inhabit the wilds of Brazil, would seem, according to the relation of Lery, to use the same kind of numeration. To denominate six, seven and eight, those tribes only add to the word hand, the names for one, two, three, &c. The same mode, as we learn from Mungo Park, is practised by some African nations, particularly the Yolofs and Foulahs, who designate ten by two hands, fifteen by three hands, and so progressively. ‘The Quinary Numeration seems like- wise, at a former period, to have obtained in Persia; the word pentcha, which denotes five, being obviously derived from the radical term pendj, which signifies a hand. Suppose ezgity-six to be now disposed in the Senary Scale, which proceeds by successive séxes or heatads. Parted in- to*stz rows, that number would leave ¢wo counters on the first bar, and cast fourteen to the next; this fourteen, being reckoned by szxes, would drop ¢wo coun- guyapy scale. ters on the second bar, and transfer two e ® e to the third bar. The original number nom would hence be described as two dupli- 6 @ e cate hextads, two single hextads and two. It is likewise represented thus by — — strokes. The Senary arrangement has few advantages to recom- mend it; yet it seems at one period to have been adopted in China, at the mandate of a capricious tyrant, who, ha- ving conceived an astrological fancy for the number six, commanded its several combinations to be used in all con- cerns of business or learning throughout his vast Empire. When the index of the progressive scale is larger, it of- ten becomes inconvenient to place so many counters as £2 PALPABLE [NUMERATION. are wanted on the same bar. But this notation may be abridged in the case where the index is an even num- _ ber, by adopting a counter of greater dimensions to sig- nify the half of it. Thus, in the Octary Scale, which proceeds by succes- } sive eights or octads, the original number would leave szr_ counters on the first bar, and throw ¢en to the next; and — this ten being told over again by eights, would leave two counters on the second bar, and carry two tothe third. If, therefore, the large counter signify half the index, or four, — eighty-six would be thus denoted. sue ee i The number is denominated one dupii- ; cate octad, two single octads, and sia ; [1 @ and it has been decomposed into one siz- ® @ ty-four, two eights, and siz. % 8 f ; . i Suppose now that the Denary Scale of Notation were i employed. ‘Thesame number, reckoned by tens or decads, would leave szx counters on the first bar, and cast eight to the next bar. This arrangement fur- pgnany SCALE. nishes the denomination of eight decads, | and. six ; which is simpler than any of & the former appellations, and yet it sounds uncouth, owing merely to our want of $ familiarity with the terms. It would be marked in this way: @ . i We are thus conducted by successive advances to that system of numeration which has prevailed among all civi- _ ture of language. ‘This almost universal consent clearly _ bespeaks the influence of some common principle. Nor | is it hard to perceive, that the arrangement of numbers by — NUMERATION. ] ARITHMETIC, 23 reNns would naturally flow from the practice so familiar in the earlier periods of society,—that of counting by the fingers on both hands. The composition of the terms em- ployed in the more polished tongues of antiquity, how- ever, is not easily or clearly traced. But the origin of the names imposed on the radical numbers, appears conspi- cuously displayed in the nakedness of the savage dialects. The Muysea Indians, who formerly occupied the high plain of Bogota in the province of Grenada, were accus- tomed to reckon first as far as ten, which they called qua hicha or a foot, meaning no doubt the number of toes on both feet, with which they commonly went bare and ex- posed; and, beyond this number, they used terms equi- valent to foot one, foot two, &c., corresponding to twelve, thirteen, &c. Another tribe, likewise inhabitants of South America, the Sabiconos, call fen, or the root of the scale, tunca, and merely repeat the same word to signify an hundred and a thousand, the former being termed tunca- tunca, and the latter tunca-tunca-tunca. Etymology, guided by the spirit of philosophy, fur- nishes a sure instrument for disclosing the monuments of early conception, preserved, though disguised, in the structure of language. Our own dialect, as immediately derived from its Gothic stem, betrays a composition not less rude or expressive than the simple articulation of the Sabiconos. The words eleven, twelve, &c. anciently sig- nified merely one, two, &c. leave ; intimating no doubt that one and two are to be kept, while ten, the root of the scale, is set aside. TZ'wenty, thirty, &c. meant simply two, three, &c. drawings ; importing that so many Zens are to be taken from the heap. An hundred, in the Gothic and Saxon, in- timated that ten was to be ten times drawn or told. 24 PALPABLE [NUMERATION, It is remarkable that the Peruvian language was ace _ tually richer in the names for numerals than the polished dialects of ancient Rome or Greece. The Romans, we have seen, went not farther than mille, a thousand; and the Greeks made no distinctive word beyond zug, or ten thousand. But the inhabitants of Peru, under the Incas, following the Denary System, had the term huc, to denote ONE; chunca, TEN 3 pachac, ONE HUNDRED; Auaranca, A THOYSAND ; and hunu, a MILLION. ‘These words are eis — ther original, or have been formed, like our numerical terms, by the abbreviation of certain compound expres- sions, tribes who surrounded the infant colony of New England, used the Denary Scale of arrangement, and had a set of © distinct words to express the numbers as far as a thousand, and could even advance as high as one hundred thousand by help of combined terms, Thus, one they named nguit ; TEN, piuck ; AN HUNDRED, pdwsuck ; and a THOU- SAND, mittdnug, But these words are apparently com- pound, and would doubtless be found to throw much light on the subject of numeration, if we had any means of ana- lysing them. It is likewise asserted, that those savages employed grains of corn for numerical symbols, and were very expert in computation. The Laplanders, in their system of numeration, join very significantly the cardinal to the ordinal numbers, Thus, to express ELEVEN and TWELVE, they say auft nubbe lokkai, and gouft nubbe lokkai ; that is, one to the second It appears, from an early document, that the Indian _ -~ > ee OF Oe A ae TEN, and ¢wo zo the second TEN. In like manner, to sig- nify TWENTY-ONE and TWENTY-TWO, those rude people use the expressions auft gooalmad lokkat, and gouft gooal- NUMERATION. ] ARITHMETIC. 25 mad lokkai, meaning one added to the third TEN, and two added to the third ren. ‘This procedure affords a curious and very happy illustration of the principle of numeral ar- rangement, The Duodenary System of arrangement was introduced at a more advanced stage of society. It plainly drew its origin from the observation of the celestial phenomena, | there being twelve months or lunations commonly reckon- ed in a solar year. ‘The Romans likewise adopted that. index, to mark their subdivision of the unit of measure or of weight. | | ny. The mode of reckoning by ¢welves or dozens has been very generally employed in wholesale business, Nor is its application there confined to the first term of the progres- sion, but extends to the second or even the third term. Twelve dozen, or an hundred and forty-four, makes the long or great hundred of the Northern Nations, or. the Gross of traders. ‘Twelve times this again, or seventeen hundred and twenty-eight, forms the Double Gross. DUODENARY SCALE, Let the same number, ezghty-szx, as before, be reckoned on the Duodenary | Scale. It contains seven dozen and | @ two, and consequently will be repre- ’ © sented as here annexed. . © Next to the Denary Scale itself, the system of counting _ by progressive scores or twenties, derived from the same source, appears to have been the most prevalent. The _ savage who had reckoned the fingers on both hands, and then the toes on both feet, amounting to fwenty in all, 26 PALPABLE [NUMERATION. © might seem to have reached the utmost limit of natural calculation. The Guaranis, a very simple and inoffen- sive tribe, who live on the shores of the Maranon, are ac- cordingly said to proceed no farther in their direct nu- meration. When these people want to signify an hundred, they only place in a row five heaps of maize, each consist- ing of twenty grains. The Mexicans, however, being more advanced in society, were accustomed to employ the higher terms of the same progression, thus combining the Denary with the Binary scales. In the ancient hierogly- phic paintings of that unfortunate race, wnits, as far as a score, are exhibited by small balls ; and twenty is denoted by a figure, which some authors, and particularly Clavi- gero, have mistaken for a club, but which was really a small standard or flag. In the same curious monuments, twenty scores, or four hundred, is signified by a spreading open feather ; probably, because the grains of gold, lodged in the hollow of a quill, represented, in some places, mo- ney or the medium of exchange. ‘This symbol has, from the rudeness of the drawing, been taken at times for a pine-apple, an ear of maize, or even the head of a spear ; but its application to intimate the duplicate scores is cer- tain and invariable. A sack or bag was also painted by those ingenious people, to represent twenty times twenty scores, or eight thousand: It was of the same form as a purse called ziquipilli, and supposed to hold eight thou- sand grains of cacao. ‘The annexed figures, (Or | ; copied from Hum- q g 8000 400 ANTE boldt’s splendid work, entitled Vues des Cor- dilleres, exhibit the series of Mexican hieroglyphical nu- merals. To avoid the multiplication of the balls, and other 20 1 NUMERATION. ] ARITHMETIC. 27 symbols, those people sometimes divided the flag denoting a score by two cross lines, and coloured the ong half of it to signify ¢en, or covered three quarters of it with colour to mark fifteen. ‘This mode of abbreviating the signs was evidently capable of farther extension. Traces of numeration by scores or ¢wenties still exist in the old continent. The expression three score and ten, in our own language, is more venerable than seventy ; and the compound quatre-vingts et dix, is the ordinary mode in French for signifying nznety. ‘The inhabitants of the province of Biscay, and of Armorica, people descended from the ancient Celts, are said to reckon like the Mexi- cans, by the powers of ¢wenty, or the terms of progressive SCOres. The principle of numerical scales might likewise be con- veniently employed in the exhibition of uniform systems of weights and measures. But, for this effect, as many distinct sets of ascending objects would be required, as one less than the number of units contained in the index of each progression. Thus, on the Bznary System, one set of weights, rising by a successive duplication, is very ge- nerally used. If the Denary Scale were preferred, there would be wanted no fewer than nzne sets of separate weights. This number might indeed be reduced to four, by employing in combination, besides the single series, others having double, triple, and quadruple its value. Since, in the notation by numerical scales, the import of a counter depends on the position of the bar on which it stands, any alteration of the place of units must produce a proportional change on the value of the whole amount of an expression. ‘Thus, in the Binary classification, the 28 | PALPABLE [NUMERATION. shifting of the bar of units one place lower, would, in ef- fect, double all the preceding terms, and a second shift would double these again. In like manner, to carry the beginning of the scale a bar lower would, in the Denary system, convert the units into tens, the tens into hundreds, and the hundreds into thousands; thus augmenting the © whole expression tenfold. On the contrary, if the units | were moved to a bar higher, the amount of any expression — would, in the Binary scale, be reduced to one-half, and, in the Denary, to the tenth part of its former value. Hence, to multiply or divide by the index of any scale or © its powers, we have only to change the names of the bars, or to shift the place of the units. | The systems of progressive numeration are hence as well ‘adapted to represent a descending as an ascending series ; a property which greatly facilitates and simplifies the ex- hibition of fractions. Suppose, for example, it were sought to express thirteen-sixtenths on the Binary Scale. Since a counter depressed by four bars will signify only the szxteenth of its BINARY SCALE, original value, so thirteen, reckon- | fet i ing upwards from the low bar, will 6 w ® express the value required. | But the analysis of the fraction might ‘be performed otherwise. It is evident, that thirteen-sixteenths on the jirst bar, or the bar of units, are only equivalent to ¢wen- ty-six such parts, or one counter and ¢en-sixteenths placed on the descending bar. ‘This excess again corresponds to twenty, catried to the ¢hird bar, making one counter and four-siateenths. But these four-sixteenths, by successive duplication, pass over the fourth bar, and leave a whole counter on the fifth. The result is thus the same as NUMERATION. | ARITHMETIC. 23 before, and the fraction proposed appears to consist of one-half, one-fourth, and one-sixteenth. On the Quaternary Scale, this fraction would require only ¢wo bars, since sixteen is only the second power of the index four. But the second mode of decomposition is, on the whole, simpler. The ¢hirteen- QUATERNARY, sixteenths of a counter on the bar of units, SAF; are equivalent to fifty-two, or three coun- | ters, and four such parts on the next bar; and these four-sixteenths correspond to sixteen, which make a whole counter on | the third bar. To express the fraction on the Senary Scale, more bars are wanted. It is now equivalent, on the bar immediately after that of units, to seventy-eight-sixteenths, or four coun- ters and fourteen parts. ‘This excess corresponds to eighty- four, or five counters and four parts, on the third bar. These four parts again give to the fourth bar one counter and eight parts, which correspond to three counters on the fifth bar. hea SENARY SCALE, | © a inch © -¢ } @ eo @ thus: @ | ¢ HE LT ? Let the same fraction be expressed by . ete of successive decomposition on the Denary Scale. Ten times thirteen-sixteenths, or one hundred and. thirty parts on the 30 PALPABLE [ NUMERATION. bar immediately below the units, make ezght counters and two parts. These two parts DENARY SCALE. are equivalent to twenty parts, or J one bar and four parts on the l third bar. This excess, again, | ° 7 | je And, lastly, these remaining eight, | | ; parts are represented by five counters placed on the fifth bar. The fraction thirteen-sizteenths is thus analysed into eight-tenths, one hundredth, two thoeusandths, and five ten thousandths. There is not the same facility, however, in decomposing — all fractions, and reducing them to the terms of a descend- — ing scale. It often happens that the expressions for these will run through many bars, or even maintain a perpetual — circulation, without ever drawing to atermination. Sup- pose, for example, that three-sevenths were to be represent- ed on the Binary Scale: It would correspond to siz-se- wenths for the second bar; which is equivalent to twelve parts, or one counter and five parts on the third bar. Thesey — again, make one counter and ¢hree parts for the fourth bar. This bar must therefore leave out the same portion as the first bar; and consequently the same notation will be con- tinually repeated at the interval of three bars. gives forty parts, or ¢wo counters and eight parts to the fourth bar. | BINARY SCALE. i | me me ee ey ey eS OSS oi feileileliellelfell =. The same fraction would be thus expressed on the Se- — nary Scale : NUMERATION. ] ARITHMETIC. 31° SENARY SCALE. The three-sevenths of a counter are here equivalent to eighteen parts, or ¢wo counters and four parts on the se- cond bar; and these four parts correspond to twenty-four parts, or three counters and three parts on the third bar. There being thus a remainder of three-sevenths, it is ob- vious that the process of decomposition will be incessantly renewed on the alternate bars. If this fraction be reduced to the Denary Scale, the cir- culation will be still more complex. It will give for the Second bar thirty parts, or four counters and fwo parts; for the third bar ¢wenty parts, or #wo counters and six parts; for the fourth bar szxty parts, or eight counters and four parts ; for the fifth bar forty parts, or five counters and jive parts; for the sixth bar fifty parts, or seven coun- ters and one part; and for the seventh bar fen parts, or one counter and ¢hree parts, which being the same re- mainder as at first, will occasion the perpetual recurrence of the same series of counters. DENARY SCALE. 82 PALPABLE [NUMERATION, In a similar way, might compound fractions and num- bers of involved denominations be reduced to arithmetical scales. Thus, to express thirty-eight pounds sixteen shil- kings and tenpence halfpenny, on the Quaternary Scale. The integer thirty-eight being analysed, gives two coun-_ ters to the bar of units, one to QUATERNARY SCALE. the next higher bar, and two | | to the bar above this. Again, & r.) & @ @ : | @@ | if quadrupled and placed on | t @ ) the bar below that of units, ‘ t : | . @ the fractional parts ofa pound, @ makes three pounds seven shillings and six pence ; this exe cess, again, carried to the next lower bar, is augmented ta — oOo 3 $ oD f éwo pounds and five shillings; and these five shillings are 5 é equivalent to one pound on the third descending bar. But it seldom happens that the expression of such coms _ plex quantities terminate so soon, or indeed close without circulation. ‘To reduce a fraction to any descending scale, may therefore prove a tedious, and often impossible task. — But the converted expression approximates rapidly to its true value, and a very few terms will be sufficient for eve- ry practical use. The numerical scales are thus equally fitted by their constitution for ascending or descending,—whether for exhibiting huge multitudes or the minutest subdivision — of parts. But men were, in general, very slow to per- * ceive or to avail themselves of this most yaluable though — distinguishing property of such progressions. The Chi-— nese are the only people who have for ages been accus= ; tomed to employ the descending terms of the Denary Scale, — or to reckon by Decimal parts in all their commercial * * . ey transactions. The same uniform system directs the whole — e Ps ‘ , : NUMERATION. ] ARITHMETIC. 33 subdivision of their weights and measures ; an advantage of the highest importance, since it gives to the calculations of those ingenious traders the utmost degree of simplicity and readiness. When the index of a Numerical Scale is large, the notation may be conveniently abridged, by marking only what counters are wanted to complete any bar, or ren- der its expression equivalent to that of an additional counter placed on the bar immediately before it. ‘Thus, instead of eight counters on a particular bar, it would be the same thing, to join one to the preceding bar, and put ¢wo deficient or open counters in the Denary Scale, or four such counters in the malt Scale. For the sake of illustration, let Ib some of the former ex- i amples be resumed. The | 4 number eighty-siz, asre- @ @ pe 6 presented on the Octary | @ @ | @ Scale, may have the szx | | © counters on the bar of units changed, by substituting ¢wo deficient counters, and joining another counter to the bar immediately above it, as here exhibited. On the Denary I. Scale, the expres- sion for the same | Ce) T number will thus ‘appear succes- ! oY changed. | i. -@-6-6-0-@- = @--- —-—%oO-— The last form of notation, consequently, signifies one hundred, abating go fourteen. D $4 PALPABLE [ NUMERATION, ; Il. | Again, the number v eighty-six, as represent- & l ed on the Duodenary @ Scale, may bethus modi- | fied ; implying @ gross | I and two, diminished by Jive dozen. This method of employing open or deficient counters - is applicable likewise to the notation of fractions. Thus, — the fraction thirteen-sixteenths may be expressed two different ways on, és a the Quaternary Scale ; the second 66 | $08 form intimating it to be the same as e one and a sixteenth diminished by a | a | | Sourth. The same fraction may be represented in three distinct forms on the SENARY SCALE. : I II, eole lel i © ! The fraction three-sevenths, which, on the Denary Scale, formed a perpetual recurrence, may, in like manner, be | ' (ee DENARY SCALE, - NUMERATION. ] ARITHMETIC. 35 abbreviated. The first expression is converted into ano- ther more commodious one, by changing the counters that _ exceed five on any bar into deficient counters. There is al- _so the same circulation as before, at the interval of siz bars, From the application of the same principles, it is easy to reduce any number expressed by rows of counters, to its original heap. ‘Thus, in the annexed arrangement on the | Binary Scale, the counter _on the highest bar is equi- BINARY SCALE. valent to /wo on the next a | | | @ Q ps bar, to,four on the bar be-_! Errnaiay & low this; and so forth by a repeated duplication, till it leaves sivty-four in the bar of units. In like manner, the coun- ters on the third and second bars, carried downwards, give four and two to the bar of units; and consequently the aggregate amount of this decomposition is seventy: one. In the Ternary Scale, the corresponding expression is thus marked. But the ¢wo counters on the highest bar correspond to siz on TERNARY SCALE. ithe next, to eighteen on the following @ @ @ 6 bar, and to fifty-four on that of units. @ | @ | The single counter on the ¢hird bar is equivalent to nine on the bar of units ; the ¢wo lower coun- ‘ers are equivalent to szr on the same bar, which there- ‘ore holds two, six, nine, and fifty-four, amounting in all 0 seventy-one units. In the Quaternary Scale, the fourth QUATERNARY ind second counters correspond to i as ixty-four and four, on the bar of o inits; which, with the three coun- ers already occupying that bar, make Ip seventy-one. 36 PALPABLE [ NUMERATION. Lastly, in the Quinary Scale, the QUINARY #wo counters on the third bar would give SCALE, vel ten to the next, which joined to the four a4 © i counters placed on it, would furnish se- & wenty, to be annexed to the one occupy- @ ° . | ing the bar of units. © Numbers thus reduced to their primary elements, might be distributed again into other classes, and consequently | arranged on any scale. But in certain cases, they can be- transferred directly to a higher scale, without undergoing | such previous decomposition. This will happen, whenever | some power of the one index becomes equal to a power of the other, or when the result of their repeated multipli- cation actually coincides. Thus, resuming the expres- sion of the Binary Scale, if @ BINARY SCALE. tf , , heal mitted, and their values ; cast to the bars immediately below them, which are mark+ ed here with dots, the whole would be converted into. the Quaternary Scale. The counter on * Agel f 3 ; . the second. bar furnishes two to that of QUATERNARY SCALE. the alternate bars were O- units; the counter on the ¢hird bar re- mains untouched; the fk bar remains @ @ vacant, and forms the ¢Azrd inthe new hehionh arrangement; and the counter on the se- | venth bar continues unaltered. Again, by condensing three bars into one, OCTARY — the expression of the Binary Scale will be SCALE. changed into another on the Octary Scale. 1 Here the fourth bar transfers its vacuity to | the second, the seventh bar is converted in- @ toa third, while the counters which occu- pied the third and second bars increase the number of units to seven. _ NUMERATION. ] ARITHMETIC. 37 In like manner might any expression on the Ternary, be reduced to one on the Nonary, Scale, by condensing every pair of bars into a single bar. On the same prin- ciple, the exhibition of a number by the Quaternary Scale would be transferred to the Octary Scale, by reducing the import of three bars always to that of two bars, since four multiplied ¢hree times gives the same result as e¢ght mul- tiplied twice. We shall now investigate some general properties of ‘those Numerical Scales. Suppose, in the Binary Progres- sion, there was standing but a single counter on a high bar. It is obvious, that this counter might be removed, and ¢wo such placed on the inferior bar. But one of these | might likewise be removed, and ¢wo counters substituted for it on the bar next lower. The same process could be |pursued through any num- Su anniae Sas) ber of bars, the removed ‘oe 1 ; 7 ; counters beingalwaysmark- ©, @ @ @ 6 @ ed by a dot, and the one eee @ @e es ee which is finally rejected pla- ced on the outside of the last bar with two dots over it. It hence appears, that one placed on any bar of the Binary Scale, is equal in value to one joined to the sum of a series lof units running through all the inferior bars. Thus, in ithe example now produced, one counter occupying the sixth bar, and therefore indicating the number thirty-two, is equal in effect to one annexed to the sum of sixteen, eight, four, two, and one. Let similar modifications be introduced into the Ter- mary Scale. Suppose a single counter to stand by itself, al the rest of the bars being empty. It may be taken 38 PALPABLE [NUMERATION. away, and three counters TERNARY SCALE. substituted for it on the ' next inferior bar. But one ® e © Ve) @ 3 of these may be now with- & & @ © © drawn, and three others @: @: %- @: . placed on the following bar: Of this triplet, the under- most might again be removed, and three substituted for it on the next bar. The same process, it is evident, could be repeated, till the change reached the lowest bar, leaving out an excess of one, marked by two dots to signify its being transferred.— Hence the single counter on the Ter- nary Scale is, by successive mutations, converted into two rows of counters extending through all the inferior coun. ters, and leaving an excess of one. Thus, the number éwo hundred and forty-three, the value of a counter placed on the fifth bar of the Ternary Scale, is equal to one added to double the sum of ezghty-one, twenty-seven, nine, three, and one, the values of counters occupying all the inferior bars. +f In like manner, if a solitary counter in the Quaternary Scale be withdrawn, four coun- J ters may be substituted on the QUATERNARY SCALE. next bar. Remove the under- e. Lad most of these, andset four more |! : © on the succeeding bar. Take e ps away one of these again, and | put other four counters on the adjacent bar. Proceed in the same way, till the quatere nion reaches the last bar, and is reduced to a triplet, by | the exclusion of one counter.—By this analysis, therefore, the simple counter is resolved, with an excess of one, into three rows of counters, which run through the whole of the lower bars. In the present instance, the number /wo _ hundred and fifty-six, the import of a counter on the (fifth NUMERATION. ] ARITHMETIC. 39 bar of the Quaternary Scale, is equivalent to one joined to triple the sum of sixty-four, sixteen, four, and one, the va- lues of all the succeeding bars. It may seem scarcely necessary to pursue this investiga- tion farther, but we shall extend it likewise to the Qu7- nary Scale. A single counter, it is obvious, may now be removed, and five substituted for it on the next bar. The undermost of these, again, may be withdrawn, and five placed instead QUINARY SCALE. of it on the following bar. Oneof 3g 3 3 es ech these may then be taken away, and @® 6 © @ @ five substituted for it on the adja- | > ° ° ‘cent bar. The same procedure is e ® e repeated to the last bar, leaving four | @. @. @. rows, with an excess of one counter. In the present instance, a single counter on the fourth bar, ‘and corresponding, therefore, on this scale to the number six hundred and twenty-five, is equivalent to one added to four times the sum of one hundred and twenty- five, twenty-= frve, jive, and one, the values of all the inferior bars. We may hence conclude in general, that if one be tae ken from any power of the index of any numerical scale, the jremainder ‘will be equal to all the inferior powers repeated as often as the units in that index diminished by one. Wherefore if the counters, reckoned as mere units, be se- parated from any compound expression, the whole will be converted into as many trains of counters, occupying all the inferior bars, as correspond to the index of the scale di- minished by one. Suppose, for ‘example, the expression here ‘noted, which is disposed on five bars of the Ternary Scale, ‘and is equivalent to one hundred TERNARY SCALE. r33 4.0 PALPABLE [ NUMERATION, and seventy-eight. By decomposing aa ems ae cessive bar, and placing the ex- cluded counters close beside the place of units, it will be chan- ged into this regular but com- plex form, which consists of Zwo rows of two counters, leay- ing out ¢wo ; two rows of sin- gle counters, leaving out one; the ¢wo counters on the last bar but one redoubled, the final counter being excluded. ——- ——_ -6-® ——- ——- -@-0-6-@- —-@@ -@0@6@ -@-8-8-6- 0-@- ae 5 6:6: 9: If we omit altogether the six excluded' counters, an take only the single rows of the rest, the result must evi- © dently express the quotient of @ the remainder by two, or by the index of the scale diminished by one. Collecting these coun- y ters on the Ternary Scale into a | ° 1 o @ e @ 8 9 more compact form, the expression belowis obtained; which corresponds to eighty-six, or , ae the half of one hundred and © | | Hs @ seventy-two, the original num- | ‘eb: 9 fy ber, leaving out the siz counters employed in denoting its pind ape similar property, belonging to numerical scales, may be deduced from the combination of the deficient or open counters, Let us begin with the Binary System, Suppose a solitary counter to occupy the sixth bar. It may be removed, if ‘wo counters be placed in its stead on the next'bar. But, without changing their value, we may to this pair evidently join another, composed of a full and an open counter, which perfectly balance each othe NUMERATION. } ARITHMETIC, 41 Withdraw the open counter that stands undermost, and substitute for it #wo open counters on the fourth bar. To these again, add a balanced pair, consisting of an open AINARX SCALE, and a full counter. Take @. @ re) @ roy ° oO away the undermost coun- | @ 0 @@ re @ ter, set #wo similar coun- | @ Oo @ re) @ ters on the third bar, and | O. 6. O. @. O. to this pair annex a full and an open counter. By continuing this process, the ‘singie counter will be decomposed into three rows of coun- ters, alternately full and open, with an excluded counter, which is open when the number of the bars, as in the present case, is even, but full if that number be odd.— Thus, thirty-two, the value of the counter on the sixth bar, is equal to three times all the inferior alternating bars, or the excess of szxteen above eight, joined to the excess of four above two, together with one, that is, eleven ; abating, however, the excluded counter which is here open; but thrice eleven, omitting one, is thirty-two. Let a similar analysis be applied to the Ternary Scale. Change the solitary counter for three counters laid on the next bar, and to these join a balanced pair, consisting of a full and an open counter. Remove this open counter, jand substitute three open counters for it on the succeeding bar, and to the triplet annex an open and a full counter. ‘Take away the full counter, and place three such counters on the following bar. Repeat the procedure, till the first ‘bar comes to be occupied; and TERNARY SCALE. there will evidently emergeyfour) oP } vows of counters extending i © ° e 2 @ ‘through all the inferior bars, | 4 5 4 o and alternately full and open, |! > GC ° 5 veh an excluded counter of | QO. @. 6. 6. 42 PALPABLE [ NUMERATION, an opposite character to those which terminate the decom- position. In the present instance, where the single counter stood on the fifth bar, a counter of the ordinary kind is left out. Wherefore the number ezghty-one is equal to four times the amount of the excess of ¢wenty-seven above nine, and of that of three above ane, or twenty, together with the wnt excluded. | It may be deemed sufficient for grounding a general | conclusion, to repeat the same process on the Quaternary — Scale, Instead of the solitary counter, place four coun- ters on the next bar, and conjoin with these a balanced pair of counters, composed of a full andan open one. Re- | move the undermost of these, and substitute four open counters on the following bar. QUATERNARY SCALE. To these, again, add an open and a full counter, which will not affect the value of the ca- lumn. Pursue the operation till all the bars are occupied by counters, and one excluded. It is evident therefore, that,on the Quaternary Scale, the decom- position of a single counter pro- duces five rows of alternating counters through all the in- ferior bars, with an excluded counter of an opposite na- ture to that of the row which completes the analysis. In the example now given, the number two hundred and Sifty- siz, the value of the counter on the fifth bar, is equal to five times the amount of the excess of sixty-four above six- teen, and of four above one, or fifty-one, that is, two hun- dred and fifty-five, together with the unit left out. evident, could be transfer- &- ed but to change the coun- Thus, inthe Binary Scale, | | ® pression: Nothing is want- i ® NUMERATION. ] ARITHMETIC. 43 The conclusion may, therefore, be extended to any pro- gressive scale: If the value of unit on a separate bar be in= creased or diminished by one, according as the rank is even or odd, the remainder will be divisible by ane greater than the index of the scale, and the quotient will be equal to the amount of the values of unit, as alternating in excess and de- Sect through all the inferior bars. BINARY SCALE. The same property, it is red to any compound ex- oO: ters on each successive bar Oo: into alternating rows.— the expression which sig- nifies forty-three will be converted into. another @:0: composed of triple alter- -9-@-6—-@-6-6-9-6-0 nating rows, with an ex- clusion of ove counter in ‘excess, and ¢hreg in defect, or the balance of éwo open counters. If this arrangement be é ' é ' now divided by three, the result, %§ after restoring the deficient unit, will be as here exhibited. By collecting and condensing the counters on each bar, the whole will be reduced to an- other very simple form ; which @ | indicates fifteen, the third part of forty-five, or of the original num- @ ber, and ¢wo counters more. i 44 PALPABLE [NUMERATION: Let another example be selected from the notation of the Quaternary Scale. ‘The counters on these six bars will represent the number fwo thou- @ sand three hundred and fifty- @ | QUATERNARY SCALE. : : four. The excluded counters | | will range as here noted ; and, ve Sve consequently, there is on tte QOO@@GOGS® whole an excess of four counters @ , @ ¢ | which must be rejected.—The e ; : g ® | original number is, therefore, i ° | reduced to two thousand three | ® 6 : hundred and fifty, of which | the fifth part will be denoted | by repeating the counters alter- | nately full and open on the low- er bars. : This last expression being condensed and abridged, will ; stand as in the first form ; or with full counters only, as in the second. :f rae eetee. The amount of the division by vd is evidently four hums dred and seventy. We may hence conclude generally, that 7f a number ex pressed on any scale be diminished by the counters on the al- ternate bars beginning with that of units, and augmented by the counters on the even bars, the result will be divisible by one greater than the index, and the quotient will be ree presented by repeating the counters through all the descend= ing bars of alternate character. NUMERATION. ] ARITHMETIC. 45 In all these decompositions, we have stopped at the bar of units; but if we pursue the analysis through the de- scending bars, we shall discover trains of equivalent frac- tions which never terminate. ‘Thus, to begin with the Binary Scale: A counter on the bar of units may be taken away, and two counters _ placed instead of it on the BINARY SCALES | pair, again, one may be _ removed, and another pair following bar. Of this ay @e@ e ® 4 ® ®. | pe Ong _ substituted for it on the next lower bar. rr of eg | again, may be withdrawn, and two placed on the fol- lowing bar. The same operation of exchange, it is ob- _ vious, may be repeated for ever. Wherefore, the value ofa single counter is here the same as that of a single row of counters, extending indefinitely over the lower bars. | But the counter on the bar immediately below the place of units, indicates one-half, that on the next one-fourth, that on the following bar one-eighth, and so forth conti- _nually. Wherefore the sum of the fractions one-half, one- fourth, one-eighth, one-sixteenth, extended without limit, | must always approach to unit or one whole. Let a similar transformation be carried through the | Ternary Scale. Sup- —_—- pose a half counter to TERNARY SCALE, stand on the bar of 9-€ @ ®@ ® © e @ © wi. units: It may be re- i }. 6.6. 6.6 6 o 6-6. b moved, and three half Y counters, or a whole counter and half of one substituted on the next bar. Take away this half counter, and set three such, or a counter and a half, on the succeeding bar. Re- _ peat the same process continually, and the half counter on 46 PALPABLE [ NUMERATION, the bar of units will be converted into a single row of entire counters, extending without limitation through all the in- ferior bars. But these successive counters signify one-third, one-ninth, one-twenty-seventh, &c. Whertce the fraction one-half is equal to the sum of one-third; one-ninth, one- twenty-seventh, &c. continued without end. In the Quaternary Scale; let the third of a counter oc- cupy alone the bar of units. It may QUATERNARY SCAEE. be withdrawn, and f' , @ 6 « four such parts, or re Be ; a 09 Side a whole counter, and the third of one placed in its stead on the next bar. This third; again, may be removed, and a counter, with an- other third, substituted for it on the following bar. The same procedure being repeated, the third part of a coun- ter in the place of units will be changed into a row of en- tire counters running through all the inferior bars. It therefore follows, that the fraction one-third is equal to the sum of the infinite series one-fourth, one-sixteenth, one-_ sixty-fourth, &c. Again, let similar modifications be carried through the Quinary Scale. ‘Fhe fourth of a counter on the bar of units may be exchanged for five such parts, or one coun- ter and a quarter on the following bar ; and this quarter may now be removed, and five quarters, or one counter and a quarter set on the next bar. The process of de- composition may thus be continued perpetually, leaving instead of the fourth‘ QUINARY SCALE. of a counter, an un- | limited range of @. ® ® @ ce counters stretching | ‘ + $. o. @. @. | over the inferior — NUMERATION. ] ARITHMETIC, 47 bars. Consequently the fraction one-fourth is equal to the aggregate terms of the progression one-/i/ih, one twenty- Sifth, one hundred and twenty-jifih, one six hundred and _ twenty-fifth, continued without termination. _ From these very simple analyses, we may therefore con- clude in general, that the fraction of unit, which has for its denominator one less than the index of any numerical | scale, is equal to the sum of all the descending powers, or the value of a single row of counters, extending indefinite- ly through the inferior bars.—Thus, one-ninth is equal to a tenth, a hundredth, a thousandth, &e. or one-eleventh is equal to a twelfth part, a hundred and forty-fourth, a thou- sand seven hundred and twenty-eighth, &c. But the summation of a descending series, whose terms alternate, may with equal facility be discovered, by in- troducing the admixture of deficient counters.—-Thus, not to multiply examples, suppose on the Ternary Scale the fourth part of a counter occupied the bar of units. Remove this, and substitute three-quarters, or a whole counter abating one-quarter, on the next bar. In- stead of this deficient quarter again, place three such, or ome open counter, con- TERNARY SCALE. joined with the quarter of | ! | a full counter, on the suc- ‘t © ? 8 te) @ &e, ceeding bar. Repeat the | : 9. P.9s @.0. same procedure, and the quarter of a counter will be trans- formed into a single row of counters, alternately full and open, extending without limitation over the lower bars. Wherefore the fraction one-fourth is equal to the excess of the perpetual series one-third, one twenty-seventh, &c. above the similar series, one-ninth, one cighty-first, &e. 48 PALPABLE [ NUMERATION. We may hence infer generally, that the fraction of unit, divided by one greater than the index of any numerical scale, zs equal to the amount of all the descending powers taken alternatively as additive and subtractive. In all these transformations of fractions, arising from the index of the numerical scale, increased or diminished by one, the operation is repeated or alternated at each suc- cessive bar. But similar changes may be made on fractions derived from the same modifications of the powers of the index, which will regularly circulate along the bars at a corresponding interval. Thus, on the Bznary Scale, the fraction one-third, or the second power of the index two diminished by one, will form by decomposition an inter- mitting row, or a perpetual circulation, passing over the successive alternate bars. For one-third of a counter on | the bar of units is equivalent to two- BINARY SCALE. thirds on the follow- o. 6. @ o. 8 6. ° & ' i La owGe 8 te ¢. ing bar, which a- | four-thirds, or an entire counter and a. ¢hird, on the next gain are equal to bar. Pursuing the same analysis, a row of counters emer- _ ges on the alternate bars. In reality, if the intermediate bars, which here serve only for the transit of the pair of thirds, were left out altogether, the notation would pass into that of the Quaternary Scale, and obey the general rule. Again, on the same Binary Scale, the fraction one-se- venth, or the reciprocal of the third power of two, di- ininished by one, will be found to circulate at every third bar. Thus, one-seventh of a counter on the bar of units NUMERATION. } ARITHMETIC. 49 gives ¢wo such parts for the second bar, four for the third and eighth, or a whole counter and an excess of one-se- venth for the fourth counter; and if this kind of decom- position be carried forward, another counter will appear on the seventh bar, a third on the tenth bar ; and so forth in perpetual succession. BINARY SCALE. 4 ! fietitite titty? | TTETE : | : But the same conclusion might also be drawn from the general principle, if we consider that the Binary Progres- ston, by omitting always two consecutive bars, is convert- ed into the Octary Scale. It is not difficult to perceive, that every fraction is ca- pable of being either exactly represented on any given scale, or of being denoted by an expression which circu- lates after an interval of fewer bars than the denominator of the fraction contains units. In fact, the moment the same set of fractional counters comes to appear a second time, the whole expression must evidently recur in the same order. But all the possible variations or series of remainders must ever lie within the number itself, which constitutes the divisor. Thus, it was found that the ex- pression for any fraction having the denominator seven, circulates on the Binary Scale, at the interval of three bars. The same fraction represented on the Quaternary Scale has a like recurrence; but, on the Octary Scale, the ex- E 50 PALPABLE [NUMERATION. pression is renewed at every ¢wo bars, while it does not circulate till after passing over sz bars, in the Ternary, Quaternary, Quinary, and Denary Scales. Employing a similar decomposition, it will appear that a fraction, with eleven for its denominator, will, in the Quaternary and Quinary Scales, circulate on five bars, but will embrace no — fewer than ten bars, by its circulation in the Ternary, Quaternary, Senary, Octary, and Denary Scales. Having considered at some length the properties of nu-_ merical scales, and their various transformations, we have now to explain the ordinary operations performed on numbers themselves. ‘These operations are all reducible to two very simple changes,—the conjoining and the se-— parating of numbers. When two or more numerical ex-. pressions are conjoined, that is, condensed into a single ex-_ pression, or collected into one sum, the process is called Addition. But when one numerical expression is separa-_ ted or drawn out from another, leaving only a di iftsencl | or remainder, the process is called Subtraction. If the ad-_ dition should be employed merely in repeating the same — number, it admits of abbreviation, and is then termed I. Multtplication. On the contrary, if the subtraction be li- ‘i mited to the continued withdrawing of the same number from another, the process becomes capable of abridgment, and is termed Division. ADDITION. ] ARITHMETIC. 51 ADDITION. Tue whole operation consists in collecting and con- densing the separate expressions. Beginning with the lowest bar, the counters are gathered together, and if they exceed the index of the scale, this excess only is retained, and one counter annexed or carried to the next bar. But if the counters on any bar should contain the index more than once, the number of repetitions is transferred a place higher, while the remainder of the reckoning is left as it stood. A very few examples will render the mode of pro- ceeding quite clear. Let it be required to collect the ex- pressions here disposed on the Quaternary Scale, and cor- responding to four hundred and se- QUATERNARY venty-two, one hundred and seventy- SALE nine, and two hundred and thirty. | On the bar of units, five counters ® | occur, which leave one, and advance ‘one to represent four on the next bar. This second bar now holds four counters, which, therefore, leave it vacant, and furnish one to represent them on the third bar. ‘On this, again, seven counters are found, leaving three- consequently, and furnishing a counter to the fourth bar, which thus comes to contain ‘nine counters. One counter is, therefore, left, and two carried to the counter occupying the highest bar. ‘The sum of those three compound numbers hence corresponds to eight hundred and eighty-one, S- - &- -6-@ —e- wet 2: —@-& -6'——-8-@-6-@-3-————- -@-8-8-|— @-9/-8-0-€ -8-8-8--——_-- 52 PALPABLE Let the same numbers be trans- ferred to the Quinary Scale. On the bar of units, szz counters oc- cur, which leave an excess of one, and send another to represent five on the next bar. ‘This second bar has now likewise szz counters, and hence leaves one, and advan- ces one to the higher bar. The third bar, containing fen coun- ters, is left vacant, after giving two representatives to the fourth bar. This last bar now holds seven coun- ters, or retains /wo, and sends one to the highest bar. ‘The expres- sion for the sum corresponds, as before, to eight hundred and ecighty- one. Let those numbers be arranged on the Denary Scale ; they will be thus represented. In the bar of units, the four lower counters, with one of the upper, leave a counter, and furnish another to represent the two five on the next bar. This advanced counter, joined to the single counters on the second bar, make five, with an excess of three, while the two remaining fives give a counter to the high- er bar, making likewise jive coun- ters, with an excess of three. The counters on the several bars being now collected on the opposite page, g -0-© [ADDITION. QUINARY SCALE. @@0é Settee entheee tie eel DENARY SCALE. —@-@ ———_—_—_—. -§-—_- ——-|8-@-0-@-|__—8- 86-36-06 000. @-@-¢,__-e e-@ ———'-8-@66-@ o¢e@ at —_—__'e-@-@-6'—_ ee —@-|-——--@|—__——_ -@-— ADDITION. | ARITHMETIC, rive obviously the same result as be- ore. . The working of these examples would be generally simplified by he judicious application of deficient counters. Thus, in the first expres- ion of the Quaternary Scale, one ounter being on the highest bar, our counters may be taken up, or in open one left. (The same process nay be applied to the two following »xpressions; and farther, the last one nd additional counter being pla- ed on the second bar, four coun- ers may be taken up from the bar f units, leaving ¢wo open counters. vollecting, therefore, the counters m the several bars, and observing Cl an ~Y + 35 | QUATERNARY SCALL. i —O- ®———e-e- -0-0-0 o-——_——e-8- ; = -O—| 8-6-0 - (600/60 hanes | he opposite effects of the full and the open, there is a ba- ance of one counter on the first bar ; the four counters on he second bar leave it vacant, and throw one to the next, n which the partial balance gives an excess of three; the pposite counters on the fourth bar leave a balance of one ; ind three counters are still found on the highest bar. If open or deficient counters be idopted in the expressions on the Quinary Scale, the quantities will ye thus denoted. The four defi- sient counters on the bar of units ire equivalent to one full counter mn the same bar, and one taken from she next bar. In the third bar, the opposite counters are exactly balanced, and in the fourth bar there is an excess of ¢wo counters. nant SCALE. 6} ¢ ee eee €- -6-€\— 6 —-0- —0}-@-6|-——- — FF —-& --—— —e'—_'—-c0-0- -——@- -©-@- ——— 54 PALPABLE [ ADDITION, Let the numbers be expressed DENARY SCALE. by open counters on the Denary , Scale. The arrangement will stand & | thus: But, on the bar of units, set- ting aside the full and the deficient | counter, one full counter is left; on the next bar, the ¢hree full balance three of the open counters, leaving 0-60-00 ss, Pt Sone at | | another open ove ; on the third bars | & rd Bs the four full counters are equivalent | ® ; i to five such, with an open one. | @ | - e008 | © % The whole amount may be express- : | bis ed otherwise thus: there being eight & 9 ; counters on the second and on the ® @ @ . third bars, and one counter on the S @ : % first. cy ® ) It is more natural, however, and more like the sponta- g neous practice of uninstructed men, to proceed by succes- . sive steps, and only conjoin two numbers at the same time. _ Nor is it requisite, in this mode of proceeding, that the uf numbers to be severally added be actually expressed ; it” u may be sufficient, at each advance, to retain them mentally, ~ Thus, on the Quaternary Scale, the & first number .being represented, %747=RNARY SCALE. ; Fed three counters belonging to the se- @ @ ® | fi cond number are to be laid on the ! al if first bar, none joined to those of | | : | my: the second bar, ¢hree counters add- ) & ed to the one on the third bar, Jeaving it consequently vacant, and throwing one to the fourth bar. On this fourth bar, there now occur ADDITION. | ARITHMETIC. 55 four counters, which furnish one counter to the fifth bar, and leave the ¢wo annexed counters. Again, the third number gives /wo counters ere | ° to the bar of units, making up jive, or leaving one counter, and carrying another to the next bar, on which the trans- ferred counter, with one to be joined to it, make four, and consequently leaving a void, send a counter to the third bar. The third bar having an accession of ¢wo counters, now holds ¢hree, while the wo counters of the fourth bar, joined to other three, give five or leave a counter, and fur- nish one to augment those of the fifth bar to ¢hree, Let the same process be perform- ed on the Denary Scale. To the _#wo counters on the first bar, zine _ being joined, give eleven, or an ex- _ cess of one, and another carried to the next bar, which, by the acces- sion of seven more, leaves five coun- ters, and sends another counter to _ the third bar, where sz are now col- lected. ‘The third number does not affect the bar of units; it furnishes _ three counters, however, to the jive of the next bar, and ¢wo more to _ the six of the third bar. DENARY SCALE. —-@- — ee e-3-@- 9-@- e-8-6-e-— eee@ e666 ———- B-— 56 PALPABLE [ SUBTRACTION, SUBTRACTION Is that process by which a number is severed or drawn — out from another. The number so parted is called the Minuend ; the one detached from it the Subtrahend ; and — what is left after of the separation forms the Remainder or — Difference. If the counters representing the minuend ex-_ ceed on each bar those denoting the subtrahend, we need — only mark the several excesses. But if the minuend have fewer counters on any bar than the subtrahend, it will be necessary to carry the decomposition farther, by taking a — counter from the next higher bar, and throwing its value to the expression of the other, by joining as many coun- ) J ters as there are units in the index of the scale. Suppose — it were required to subtract nine hundred and forty-seven from thirteen hundred and fifty-two, as thus arranged on the Quaternary Scale. Were the object is, without dis-_ i turbing the order of the counters, to tell out from those of the minuend the expression of the subtrahend, and note the residue. As no counter QUATERNARY SCALE, appears on the first bar of the minuend, a counter is taken from the second, which, ha- ving the value four, gives the three counters of the subtra- hend, and an excess of one. On the second bar there is nothing to take away, and consequently the single coun- ter now left is placed below. , i ; ; -yalue of one that should be SUBTRACTION. | ARITHMETIC. BY To supply the vacancy of the third bar of the minuend, a counter is borrowed from the fourth, and its value represents the three counters of the subtrahend, with a surplus of one. The fourth bar is likewise augmented by four, by anticipating the counter of the next bar, and gives ‘an excess of two. By taking the highest counter again, a difference of one is left on the fifth bar. The whole re- mainder expresses four hundred and five ; and, as in the process of subtraction, the minuend was only split into two portions, so these combined again must form the ori- ginal number. Let the same operation be repeated on the Senary Scale. In this case the ¢#wo counters of SENARY SCALE. the minuend, in the bar of units, @ are, by help of one borrowed §! from the higher bar, augmented | to eight, which gives five for the subtrahend, and leaves an excess of three. On the next bar, the two counters now left furnish a counter to the subtrahend, and another to the remainder. ‘The counter on the third bar of the | minuend, increased by six, the 6 drawn from the fourth bar, de- posits zwo on the subtrahend, and delivers over jive to the remainder. ‘The counter borrowed from the fourth bar is supplied from the six counters corresponding to the expression of the highest bar ; so that four counters are dropped, and one furnished to the remainder, which has collectively the same value as before. 58 PALPABLE [SUBTRACTION, Lastly, suppose this subtrac- DENARY SCALE. tion were performed on the —@ 8-@ —_@-@— Denary Scale. 'The two coun- augmented to fwelve by the counter borrowed from that of tens, give seven to the minuend, with an excess of five. The five counters on the bar of tens, re- duced now to four, merely sup- ply the subtrahend, leaving the remainder vacant. On the bar of hundreds, the ¢hree counters increased by the value of the higher counter, furnish nzne to the subtrahend, and leave four for the remainder. | , These operations may sometimes be conveniently a-— ters on the bar of units being | 1 86-@-8-— bridged, by the judicious introduction of open counters ; the value of the last bar will not be altered. Thus, on the Senary Scale, if a full and an open counter be annex- — ed; consequently, while the open counter of the subtra~ _ hend is supplied, three full counters SENARY SCALE, are thrown to the remainder. On @ the second bar the three counters | give two to the subtrahend, and one to the remainder. The single coun- | 3 & @ ° ter on the third bar, being combin- @ @ a) ed with a full and an open counter, © o@ @ supply the ¢wo counters below, and @ | | e surrender this excess of an open | @ counter. Onthe fourth bar, the vacuity may be occupied fh SUBTRACTION. ] ARITHMETIC. 59 by ¢wo full and ¢wo open counters, and consequently the latter go to the remainder. On the Denary Scale, the pro- cess of subtraction is likewise short- ened. The ¢wo counters of the bar of units having three full and three open counters annexed, furnish the latter to the subtrahend, and give five full counters to the remainder. On the bar of tens, the counters of the minuend and of the subtra- hend are equal, and consequently leave a vacuity. But on the bar of hundreds, the three counters with a full and an open combined, sur- render the latter to the subtrahend, and deliver four full counters to the remainder. DENARY SCALE. lle. Tt 647.9 od gh O17 pe ap poate In every instance where a counter is borrowed from a higher bar, the effect would evidently be unaltered, if a counter were added on the same bar to the num- ber below. This modification of the process is what has been generally termed carrying. It is farther illus- trated, by operating with open coun- ters. ‘Thus, resuming the foregoing | example, which might be expressed in this way on the Octary Scale. Con- ceive three full and three open coun- ters were placed on the first bar, and the latter would evidently go to the -Yemainder. Again, on the second bar, two open, and two full counters, would ——- &- 06] © Oo e060 | | 6006 ® | 0@ [1 ape | @ 0 60 PALPABLE (SUBTRACTION. throw three full counters to the re- mainder. On the third bar, two of the open counters are left in excess ; and on the fourth bar, there is an ex- cess of a full counter.—The result may be changed, as here done, into a more commodious form. © It is obvious from the procedure now followed, that the effect_would be exactly the same, if the counters of the - * es oe ee ee ee et subtrahend, converted into others of an opposite charac- — ter, were all conjoined with those of the minuend. By — such a change, the operation of subtraction is readily transformed into one of addition. To illustrate this remark, we may take any of the former ex- amples. Suppose the expressions on the Quinary Scale to be assu- med, but the counters of the sub- trahend inverted. The various counters on the several bars be- ing now collected and balanced, would give the annexed result ; which, being modified again, forms the remainder of the original sub- traction, In all these operations, the procedure is alike, whether on the ascending or the descending bars. Hence frac= tions may be added or subtracted by help of counters, — precisely in the same way as integers themselves. It would be superfiuous, therefore, to produce any examples. muttipricaTion.] ARITHMETIC. 61 MULTIPLICATION Is nothing but a process of repeated Addition. When ‘the terms, however, to be multiplied are complex, and the index of the Numerical Scale is large, the operation will admit of being very considerably abridged. It has been already shown, that a number is virtually multiplied by the zndex of the scale, by advancing its expression one bar ; that it is multiplied by the second power of that in- dex by advancing it ¢wo bars; and so forth continually, according to the progressive powers. Again, if any term of the multiplier be great, it is preferable, instead of re- peating the counters of the multiplicand, to collect them mentally, and only to mark the result. The ready perform- ance of multiplication depends entirely on the right ap- plication of these two principles. A few examples will elucidate the process. Suppose it were required to multiply the number thirty- seven by twenty-one, that is, to add twenty-one times together the units contained in thirty-seven. First, let those numbers be disposed on the Binary Scale. The counter onthe unit bar of the multi- BINARY SCALE. plier, shows |! | Baa! thatthe whole : ° | of the multi- plicand is to be set down once, as it stands. The @ next counter, passing over + le|— —|—e-@-|-@--6- | | | | | — | @ 62 PALPABLE — [MUuLTIPLICATION. © the vacant bar, indicates by its position, that the whole of - the upper range of counters must be advanced ‘wo bars. The last counter intimates a similar advance to be made again. ‘These various counters are next collected into a single row, which would give by reduction seven hundred — and seventy-seven. Next, let the same numbers be arranged on the Ter- nary Scale. Here the TERNARY SCALE. counter placed on the se- | | cond bar of the multiplier =#=£§ ;—j——|—'—|— shows that the counters of the multiplicand are to : | be carried one bar forward. | The ¢wo counters on the third bar, intimate that @ the range of the multipli- © cand must be doubled, and @ pat | a -@- ee ‘ advanced two bars. The counters on the several | | bars being now collected and condensed, give this result, composed of ¢hree, eighteen, twenty-seven, and seven hun- dred and twenty-nine ; making in all seven hundred and seventy-seven. On the Quaternary Scale, QUATERNARY SCALE. ‘the process will be simpler, er ee and require fewer bars. The | three successive counters of the | multiplier show that the mul- | | eee tiplicand is first to be repeated -@- *: as it now stands, and then at the advance of one and two 413 BIR ey lett " -@- = ‘: ai MuULTIPLIcATION.] ARITHMETIC. bars. . These three successive multiplications give for their collective amount, thrice two hundred and fifty-six, twice four and one, or seven hundred and seventy-seven, as in the for- mer examples. | By the Quinary and Senary Scales, though fewer bars will be required, the operation is on the whole a little more complex. A single instance may be judged suf- ficient. Thus, the numbers to be multiplied will, on the Senary Scale, be represented by one counter on the first and the third bars, and by three counters on the first and se- cond bars. Consequently the range of the multiplicand must be repeat- ed thrice in the order in which it stands, and likewise by one bar in advance. The result is, therefore, equal to triple the sum of one, siz, thirty-six, and two hundred and six- feen, or to seven hundred and seventy= —— -@-@- ——§-@- - 2 - -@-@- a Te or | © ws fm me. es Ce ogee - [a ees 2 | SENARY SCALE. -8-8-8--8-e-8-—-—__, en ee ee -@-8-&- ee ee -@-@-@- BS Se oe --@-6|-8-6-$-__| © @-6- -8-8-¢-'_—_'9-0-e'9-@-¢6'¢- ‘ 64 PALPABLE _ [MULTIPLIcATION. Lastly, let the multiplication of thir- ty-seven by twenty-one be performed on the Denary Scale. The counter on the unit bar of the multiplier shows that the multiplicand is to be set down once in its place, and the ena two counters of the next bar intimate that it must likewise be redoubled, ‘and placed a bar in advance. But the seven counters of the multipli- cand, on being doubled, leave four on the bar of tens, and ‘send one to the higher bar, which, with the other three eounters of the multiplicand likewise doubled, now holds seven. These counters being collected, give seven hundred and seventy-seven. —@e@ The process of multiplication is often considerably sim ; plified by introducing open or deficient counters. Thus, resuming the example on the Ternary Scale, the expresd sion of the multiplier is | | @ i; . changed into a single row | of counters. The full | counter onthesecondbar | @ ' shows that the multipli- @ @ | cand is to be advanced @ | . by a whole bar, the open counter above it indicates : 7 —O e\é—@/6 o-— @|——_@ —oO ~ *+QO50. *% he Sul ARITHMETIC. 65 ) _— . a similar advance only | bar, and e7ght full ones moved to the _ the next higher bar. This multiplicand is therefore set, as the multiplier indi- : cates, down once in the same place, and then doubled and shifted a bar higher. _ | Siz open counters are hence, in the se- ‘ with inverted counters, nM | ° : ° | | and the highest counter =~ intimates that the first o eration is merely to be repeated. The several counters are then collected, and again modi- fied into the final result. | In like manner, the example on the Denary Scale is thus modified, the seven counters on the unit bar of the multipli- | cand being exchanged for three open counters, with a full counter thrown to & -0'-—-®--0-0-0 cond operation, placed on the second third. Collecting the several counters, the result is eight hundred, abating twenty-three ; that is, seven hundred and —0o-0|9-0-e-@ seoeee se 0-0-0-—|---—— —0-0 seventy-seven, as before. © ‘; § Let another example in Multiplication be taken, where @-3-6- the numbers concerned are rather larger, being seventy- eight and fifty-seven, Arranged on the Quaternary Scale, they will exhibit the form on the following page. Con- sequently the multiplicand is to be set down once in its place, then doubled and moved a bar higher, and - Next ¢ripled and advanced another bar. Inthe second opera- tion, therefore, the ¢wo right-hand counters of the multipli- 1% $$!’ ®t * “ 66 PALPABLE — oe ee cand would be changed to four on the second bar, which is hence left void; and an equivalent coun- ter joined. to the doubled three, or the six of the third bar. On this bar, the three of excess are. placed, and one carried } to the empty bars above : it. In the third opera- “tion, the two units of the meeryipltand being érz- pled; give. an@excess of fwo, with one thrown to the nine of the next bar. On this fourth bar, two counters of surplus are laid, and #wo more sent to occupy the fifth bar. All those counters being gathered together on their several bars, exhibit a re- sult which corresponds to ve four thousand four hundred and forty- Site _ Suppose the same eraple were transformed to the Quz- nary Scale, it is evident that here the multiplicand must be doubled, commencing, at the first, and again at the third, bar; and that it must be set . ie QUATERNARY SCALE. Oo QUINARY SCALE. # - MULTIPLICATION. ] “ARITHMETIC. down once, beginning atthese- cond bar. Thedoubling ofthe | three counters of the multipli- | cand on the first bar, gives a counter in excess, with another sites to thenext bar; and the same process converts the three coun- ® ters of the third bar into one @ counter on that bar, and ano- ther on the fourth bar. ‘The td amount of all these counters afford by reduction the same J product as before. On the Senary Scale, the mul- tiplicand must be successively ¢r7- pled, and then repeated two bars in advance. The triplication of its zwo highest counters will exact- ly fill up the corresponding bar; or, what is the same thing, that bar will be left void, and a single representative counter transferred to the next bar higher. The sub- sequent collecting of the counters requires no condensation, and the whole process in this case becomes extremely simple; the first bar | heing vacant, the second, third, ‘and fifth being occupied by three _ counters, and the fourth bar hold- ing only two. e-6—--e _ o> ~y ——-@- @-0-8--6- eee -@ cf igirtig ee CeeS wo -@- -@- ———@-@ -9-8 9-3 SENARY SCALE. | aa —-—@ } e-e-%/-——-@-e-¢ —_—- | a . & ' @-2-¢|—- @-6-6 —@¢ es “upper number, when repeated siz 68 " PALPABLE | [ MULTIPLICATION. ym ‘ Suppose the same Multiplica- tion were performed on the De- nary Scale. The eight counters on the first bar of the multiplicand being re eated seven times, leave siz, and send Jive counters to the higher bar. The seven counters again being repeated as often, furnish, with the addition of those _ five, an excess of four, and throw |» other Jive to the bar of hundreds. ~ In the second operation, the eight ———— = @—ee@. eet | =~ * 7 - a od -O;——_ counters repeated jive times, send, sy 0-00-26 without any excess, four counters to the third bar ; ; where the seven counters being repeated as often, give a farther excess of five, and transmit ¢hree counters to the bar of thousands. Adding together those counters, the six counters of the first bar, and the four of the second remain unaltered, while the two fives on the third bar, leav- ing the four surplus counters, furnish out another to the next bar. & 6-0-6-6— a . ferred to the Duodenary Scale. | Here, for the sake of convenience, | the multiplier and multiplicand are made to change places. But the | nine counters on the first bar of the | *! > “ | oh | MULTTPLICATTON.] ” ARITHMETIC. _ times, give four dozen and siv ; and _ the four counters repeated after, make two at 4 or leave the four "advanced counters, and send two to 1e fourth bar. The same opera- | er. Collecting now the several gross, six gross, ten dozen and six ; : vite is equivalent, therefore, to the amount of three thousand, four hun- eed and Sifty-sizx, eight hundred and | sixty-four, an hundred and twenty- six, making collectively Sour thou- sand faMieundeed and forty-six. tion is again renewed with the other | siz counters, and carried a bar high- — counters on the bars, the product i is, in mercantile language, two double v - . a » 69 t, : —_—, . Mas Conceive the same operations to be Cae ba by help of deficient counters. On id the Quinary Scale, the num- bers seventy-eight and fifty- seven, will stand thus. The former, or the multiplicand, is, therefore, first tae, then repeated a bar higher, and next doubled again and ad- vanced 1 o bars. In collect- ing the counters, four open ones appear on the first bar ; and consequently. a vacuity on the second bar ; eight open and one full counter occupy the third. bar, leaving a sur- plus of Zwo open cotinters, l } two open and ¢wo full ones & 6060 | ‘produce a mutual balance, | e+ : | $ ee SCALE. 4 mm t AY Pe 70 . PALPABLE — [MULTIPLICATION, _ with another open counter to be carried to the fourth bar ; on which a pair of full coun- _ters balance another pair which are open, and the éwo ti { remaining full counters are reduced to a single one, by | the open counter annexed to them; while, on the fifth j bar, the four open counters are abridged to three, by the | influence of the full counter immediately above them. | This amount, after reduction, gives the same result as be- i fore. DEWARY SCALE. On the Denary Scale, the num- Oe bers to be multiplied will assume this form. The three open coun- ters on the right hand of the mul- tiplier intimate that the counters of the multiplicand are to be ¢ripled, and their characters reversed; which gives six full counters for the first and second bars, and three open counters for the third. Again, the sig fall counters of the multiplier show that all the counters of the ' multiplicand are to be repeated six r times, and moved a whole bar in advance. But the two open counters repeated so often, give a surplus of | two, and transmit one to augment the product on the next bar, which acquires three open counters, and sends another to the fourth bar, where the six full counters is re- duced by this junction to Jive. The ‘result of the whole is, therefore, five-thousand and Sort: sia, abating six hundred. is eel Pe : ‘s “S i wf hh + % Ps d | MULTIPLICATION: | »~ARITHMETIC. é 71 A od " The application of open or deficient counters will be found useful, in varying and simplifying the process of | multiplication, even where smaller numbers are concerned. : Thus, to confine our views to the com- @ ? “mon Denary Scale, suppose it were re- quired to multiply eight by seven. ~The 1 ~¢ former, being two less than ten, may be denoted by one counter on the second bar, and. sa en counters on the first ; and the 7 latter, being ¢hree less than ten, is express- % Qo ed by one counter on the second, and three if @ open counters on the first bar. ‘These o- | jen counters show that the terms of the | upper number should be subtracted three @ oO | times; that is, repeated zhrice with an op- mm?) | | | eee) Se = SS Pe] ad posite character; which gives six full counters for the first bar, and three open ones for the second, Again, the full counter on the second bar indicates that those terms are to be repeated unchanged, but placed one bar in advance. The combination now made exhibits H the product. But, instead of the single S counter on the third bar, substitute two. ° si FIvEs on the second; the result will then be decomposed into five, abating three, | Q & and five abating Wo, reckoned as TENS, ‘oak | together with the counters on the first bar, | or wa considered as UNITS. | A Hence the explication of the method for multiplying any numbers under ten, by help of the fingers merely ; an arithmetical curiosity, probably communicated by the Arabians, but certainly ‘ng to the mathematicians of | “ a val ea 4 * | m9 72 d _, . Taruunicicatrogs Europe at the periotl of the revival of ence, though lately into the practice of elementary schools. In the preceding example, beginning at the left, and thence going to the right hand, eight fingers, (including the oe) are counts | off leaving two ipgers to close. Again, proceeding the reverse way, the num- ber seven leaves an excess of three fingers to be shut on the left hand, as in the a ! bove disposition of counters. Now, joining the projecting | fingers of both hands, or five abating two, and five abating three, we obtain Jive’ TENS, or fifty; while the prodia of the closed fingers or ¢wo times ¢hree, gives an ene of s7z UNITS; and consequently the combined result is Sifty-siz. 7 ea As another example of this mode, suppose j ‘it were de- sired to multiply nzne by siz. These numbers may be se- verally expressed by ten abating one, and by ten abating | four. But the four open counters of the multiplier signify ' that the upper counters are to be chan ged and then repeat- ed four times; while the full counter intimates that those - counters are, without alteration, to be advanced a whole. bar. Instead, however, of the single counter on the third bar, two fives may be combined with the open counters on. the second. But this procedure can be imitated by the play of the fingers. Counting nine fingers, beginning with the left hand, and passing to the right, we leave one finger close; and again reckoning six, by proceeding from the’ right to the left, we have four fingers remaining to clo If we now bring together the erect fingers of both alle we shall have, of Tens, five abating Jour, and five abating one, that is, jifty; together with the product of the shut fingers, or four UNITS told once. * eat - * « y € Ketel , re ARITHM CG. ¢. This RE sok trick cannot fail to appear striking to : young practitioners, and may prove really useful to them, by helping to hho: oughly and accurately in their memory the ordinary mltipiicallant table. DENARY SCALE, | | But the same principle might be ex- | | @ | Q tended farthe Suppose it were re; $ quired es multiply ninety-nine by st @ Oo | ninety- cil These numbers are | merely e hundred, abating respec- | | tively one and two. The two open 4 counters of the multiplier signify es -o Ta Te 4 , vat the counters of the multipli- es pl nd He to be doubled and rever- | = sed, while the single full counter in- timates that those must also be repeated two bars in ad- vance. The product i is, therefore, of hundreds one hundred fold, abating Hheree,—together with | ‘teo units,—that is, nine thousand seven hundred and two, In this example, the nmtibers may be conceived as ar- ranged ona new scale, which proceeds by hundreds. This index is hence diminished by three, their joint defects, to denote so many hundreds; while two, the product of those defects, exhibits the units to be added. But such, we es seen, is the very mode furnished by the fingers on iw. 7 for multiplying the mum bers under ten. a scale, their multiplication proceeds with equal facility as | that of integers ; it being only requisite to commence with the us of units, and to descend wip the lower bars. Thus, i sought to multiply five ¢ and a half by three and . 7. “4 a evidently, on the Bi= + When fractions are expressed on the same numerical _ i, ih A tiplications from the de- ~ he . ote ¢ * ¢ é * 4 " 3 ‘i hy ¢. y ae j yw £ 7 4, ~ PALPABLE — [MuLTIPLICATIO wh nary Scale, be thus ex- BINARY SCALE. pressed. If we start from the bar of units, “ all the counters of the multiplicand must be ~ | © | a) repeated in the same | | a position, and the mul-~ i ‘, | | scending bars would popes td | only be carried by equal ceraey radations to the right @ } band. To preserve re- ne So ? ge cularity, therefore, it will be more convenient to begin witl ; the lowest counter, which occupies the second descending, bar. Hence the whole train of the multiplicand is to be set down ¢wo bars lower. Then the process goes on as usual; that row of counters is repeated ¢wo bars in ad- vance, and again at the interval of three bars. The coun- | ters being now collected together, express seventeen and | seven-eights. Let the same mixed numbers be represented on the Quaternary Scale. Beginning at the right hand, the counter on the first de- scending bar shows that the whole of the multiplicand must be repeat- ed one bar lower; and the three counters on the bar of units inti- mate that it is to be érzpled in its actual position. ‘The result of the summation of the several bars is the same as before. (ent ‘ Pt * » ¢ 5 . Fr ‘ | - ’ a a * eerie Mok 4 _ ARITHMETIC. Lastly, suppose those quan- tities were transferred to the Denary Scale. The five coun- ters of the multiplier on the se- cond descending bar, show that the multiplicand is to be repeat- ed as often ¢wo bars lower; which gives on the descending bars five for the third bar, seven for the second, and ¢wo for the first. The next ¢wo counters ‘double the range, one bar in | advance ; and the three highest counters triple the whole, at the ‘advance of another bar. All those counters, again, being col- lected together, give for the pro- duct, seventeen, with eight hun- | dred and seventy five-thousand parts. lf DIVISION | | 8 © & sd | e026 _ Is the opposite process to Multiplication, and consists in _ finding how often the same number can be separated or | wn out from another. In the rudest way, therefore, | th s operation would be performed, by telling over a cer- | tain number of counters repeatedly from the same heap. But instead of a slow process of repeated subtraction, the ¥ 9 #3 , A # * 76 PALPABLE |.) [pwviston, _and the operation renewed, till nothing is left of the divis | »and note the excess, , . 2 number to be severed, or the Divisor, may be first multi | plied to approach the mass to be shared, or the dividend, — bY ; The remainder can again be treated in the same ce dend, or a difference less than the divisor itself. ‘Those multipliers, collected together, will express the quotzen or the number of subtractions required to exhaust the mas . A few examples shall be selected for illustration, Sup. pose two thousand three hundred and forty-six were to be | : | | divided by twenty-three. Let these numbers be arranged on the Ternary Scale, the dividend being the lowermost, t and space left for put- TERNARY SCALE. ce | ting the quotient im- | mediately under the divisor. Beginning at the left hand, it is easy to perceive that ed once in the first four bars; place one counter then for the quotient on the last of these bars; set the divisor directly be- neath the dividend, i the divisor is contain- ’ which is ‘wo counters on the sixth bar, and éwo counters on the fifth. Of this differ- ence, with the next ¢wo counters of the dividend brought down, three bars are less than the divisor, but four bars % | 9 gS a | BIVISION.] ti ARTI HMETLO 7% ; will evidently contain it ¢wice. Passing over one bar, th re, wo counters joined to the quotient are placed on’ the third bar of the range. The, divisor “is ‘then _doubled and set down. The remainder of this operation, with the zwo final counters annexed, is ; exactly the same the divisor, which must therefore be contained once ; jand here the operation terminates, leaving the first bar vacant. The quotient is hence by reduction equal to one | hundred and two. | Let "the same division be quatenNany SCALE. performed on the Quaternary. Seale. The divisor is evident- ly contained once on the three highest bars of the dividend : one counter of the quotient is, therefore, placed on the fourth bar, and the divisor itself set ; ay | diviso @ | ' down for subtraction. The remainder is denoted by three ] I p< © | @ ® © _ counters on the fifth bar, one | on the fourth, a and zwo brought from the dividend to the third bar. In this excess, the divi- sor is contained twice ; it is _ consequently doubled and sub- tracted. - But the surplus now, | with the two counters annex- | ed from the dividend, con- ® | tains the divisor once on the 8 ext bar. The remainder after he third subtraction is the same as the divisor doubled. | te je 2 -@- ee ~~ f 8-8- @-8- FERbe ora eng am 8 | Re Ao Gg CRA bee RLS Ee @-@ : “See A CE RE A RR! SRS ES CN EE TC: OTR TS f | | 4 Le, ee 78 PALPABLE Whence #ivo counters are pla- ced on the last bar of thedivisor, and the operation closes. The quotient is, therefore, equal to the amount of sixty-four, of twice sixteen, of four, and of éwo, or an hundred and two. Suppose the same opera- tion were performed on the Denary Scale. Here the di- visor being identical with the two highest bars of the divi- dend is, therefore, contained once ; but it is not contained at all in the four counters brought down from the next bar. That bar of the quo- tient is consequently left va- cant ; but, from the following bar, the siz counters are brought down, forming to- gether a number in which the divisor is obviously contained twice. But the divisor being doubled gives the very same number, and hence leaves no remainder. The collective quotient of the divisor is thus one hundred and two, as in the preceding examples. re th [ DIVISION, sug : te ee » ~ 66—-6-@ -@-66-6-@ seeees Se AT 7 ae ~ a “a, Res SR gama _[DIVISION, ARITHMETIC. Lastly, let the process be conducted ‘on the Duodenary Scale. To avoid prolixity, however, it will answer bet- ter in this case to employ the admix- ture of open counters. Here the di- visor may be assumed as once contain- ed in the two highest bars of the divi- dend. Conceive, therefore, a full counter to be placed on the fourth bar, and fwelve correspondent open ones; ‘making an excess of eight on the next bar. There consequently remain on that bar seven open counters after the first subtraction, to which the three full counters are brought down from the dividend on the second bar. In this deficient quantity, the divisor may be considered as contained four times, and the quotient is marked by so ma- ny open counters. But the counters of the divisor being multiplied by these four open ones, give four full counters to the second bar, and eight open ones tothethird. The remainder is hence a full counter on the third bar, and an open one on the second, to which ‘the six full counters are subjoined from ‘the first bar of the dividend. In this last surplus, the divisor is contained exactly stx times, the result of its mul- tiplication being six open counters on the first bar, and a full counter on the third, the second bar being passed over; but the open counter on the oe | © —__—@ '—0- ee SS ee ee a ‘@ ——0-0-0|—_0-0|__0--e-6-8-6-— —e- 3-6-2-e—-ee-e—'__|_¢ ee—_|0000—ee —O'—@ -2-—-6-— % at 80 PALPABLE Lorvistaat second bar of the remainder would in effect be taken away, by changing the six full counters on the first bat with as many open ones. | This mode of introducing deficient counters is often very convenient in practice, since it only requires, at am step, to know the nearest integral quotient, without re: garding whether this be less or greater than the quantity i sought. ‘s? # In all the preceding examples, the divisor is complete; } but it will often happen that a remainder is left, and con: sequently that the process may be continued on the de: scending bars, expressing an excess of a fractional Ee tient, which either terminates, or constantly recurs ag in a perpetual circle. Suppose, for example, it were pro: posed to divide one hundred and thirty by twenty-five These numbers would be thus arranged on the Under the third BINARY SCALES bar, the divisor is contained once, leaving a counter on theseventh, one on the sixth, and another on the se- condbar. Passing over the second bar, the divisor is contained once un- der the first ; but not again till after an interval of ¢wo bars, when there is left, as under thethird bar, four | « | DIVISION. ], ARITHMETIC. 81 / consecutive counters. At this point, therefore, a circula- ‘tion must take place, since the third bar below it cor- responds to that of units itself. ‘The same sequence will ‘be continually maintained: First an empty bar, then a counter, followed by two empty bars.—To indicate this -circle of renovation, the mark v for Arves, the first Sign | -of the Ecliptic, is adopted, as intimating the birth of the | revolving year; and, therefore, by extension, the recom- _mencement of a periodical cycle. Let the same process be transferred to the Quaternary Scale. The divisor is contain- QUATERNARY SCALE, ed once in the three highest | ‘bars of the dividend; and the subtraction being made, there remains one counteron | the third bar, and three coun- | ters on the second, to which are subjoined on the first bar — L. the ¢wo counters brought down ® | from the dividend. In this 2) | quantity, the divisor is con- @ 6 | @ tained once again; and one counter is left on the second, and another on the first bar. Now, passing over two bars, S : 4 times, with the same remain- 8-@0-¢6-6 the divisor is contained ¢hreeé rst der of two consecutive coun- ters. Wherefore the opera=_ °@ @ I tion is continually renewed, @ and three counters run through i & the whole train of subsequent @ S vars. «4 ; { 82 PALPABLE EDIVISION, Lastly, suppose this division | © ra were performed on the Quinary 7 Seale. The divisor occurs once in |} @| @ the first bar of the dividend, and © | € z once again, after an interval of two bars, in the remainder. _ In short, e | } the quotient here is precisely the | @ &: same as the dividend, only placed | two bars lower. The result is con- | @ | sequently, as before, five and one-/ifth. In these last examples, the integral part of the quotial may be considered as forming a distinct number; while the remainder of the division constitutes the numerator ol a fraction, of which the divisor is the denominator, the conversion or development of it along the range of in. ferior bars being effected in the way formerly explained in treating of Numeration. —_ EE -— We have thus explained, at some length, the modes of performing the four common Rules of Arithmetic, by means of counters. But the most complex calculations in which numbers are concerned being all reducible to such elemen- tary operations, it seems unnecessary to descend to the de- tails generally given, under the various heads of Propor ‘tion, Fellowship, Interest, or Exchange. In the application of Arithmetic, however, to Practical Geometry, the process of Extracting Roots becomes farther indispensable. — This will require some explanation. As the repeated multipli- cation of the same number by itself, forms its successive Powers; so, in reference to these again, the number which generates them is called their Root. To find any power of a number is hence an easy operation; but the con- ARITHMETIC. 83 verse of this problem, or the extracting of a root, that is, the discovering of the number from whose involution or repeated multiplication the given power had arisen, can be effected only by a sort of tentative procedure, which is in some cases attended with considerable difficulty. We shall, therefore, confine our views at present to finding che Square Root, or the evolving of the Root of the Se- cond Power. In order to investigate the method of proceeding, we have only to consider how the second power of a compound number is formed. Conceive this number to lonsist of two members, or to be represented on two con- secutive bars of any scale. The same number being ‘repeated as a multiplier, it is evident that their pro- duct, that is, the square or second power, must, fol- lowing the order of the multiplication, consist of the square of the first member, the product of it into the se- cond, and again this product by the second, together with the square of this member. Consequently, the square of the compound number is analysed into the square of its first member, and the product of twice that member join- ed to the second, by the same second member. Having substracted from the given number, therefore, the square of the first member of the root, the remainder is to be di- vided by twice that member augmented by the quotient tself, to find the second member. But if the division should not terminate, it is evident that the process of ex- haustion may be still continued; for, considering the num- vers exhibited on the two bars as condensed into one | group, the addition or correction on the following bar is discovered by the same mode of decomposing the residual portion. It is farther evident, from the practice of mul- tiplication in the forming of powers, that each bar of the Foot must correspond to a pair of bars on the square. 4 84 PALPABLE [EXTRACTION OF The whole procedure in the extraction of the square root will be readily understood from a few seleet examples. Suppose it were sought to discover the square root of eighteen hundred and forty-nine. 'This number expressed on the Ternary Scale will stand as below, occupying - partially four pair of bars, noted by so many asterisks. — Hence the root must likewise be contained on four bars. Beginning, therefore, with the ¢wo counters of the high- ; est pair, the nearest root is obviously one. A counter being hence placed on the fourth bar, its square is sub- tracted from those ¢wo, and the remainder conjoined with the counters brought down from the next pair of ‘bars give a single row of three counters. Let this be now divided by the double of the first member of the root, to discover the second or TERNARY SCALE. ‘ additional member; but before the division is com- pleted, annex the quo- tient itself to the divisor. Wherefore, above the counter of the root, and on the same bar, place two counters, which, be- ing contained once in the row of three counters to be @-@ 6-@-0-0- @\——_--8-® -e|-—-e|_—-@-0- ee HY — * * ;— e-@, decomposed, set on the next barone counter forthe root, and another for the : divisor. This compound @ @ is then multiplied by the & | é annexed counter of the e © @ root, and is consequently e set down two bars higher, or beginning with the fifth ¢ @ @ - i SQUARE ROOT. ] ARITHMETIC. | 85 bar. The subtraction being now performed, there are left two counters on the sixth, to which is subjoined the next period, consisting of wo counters on the third, and _ one on the second bar. For a third operation, then, the two highest bars of the root, viewed as consolidated into a single member, are doubled, which is done by repeating the counter on the third bar of the divisor, a point on the left side of it indicating this accession, as a point on the other side of a counter was employed to signify its being withdrawn. ‘This new divisor is contained twice in the remainder ; and consequently ¢wo counters are placed on the second bar, after the portion of the root already found, and after the divisor itself. The divisor being now mul- tiplied by those ¢wo counters of the root, and subtracted, leaves a single counter on the fifth bar, to which is annex- ed the ¢wo single counters of the last period. ‘To decom- pose this remainder, the counters on the three bars of the root, considered as forming one cluster, are doubled for a final divisor, which is effected by subscribing two doted counters below the other zwo on the second bar. The di- visor now occurs only once, and therefore a counter is an- nexed to it on the first bar, and likewise to the root. But there is at last no remainder; for the,four counters on the second bar of the divisor leave one and send one to the two counters of the next bar; which being then full, transmit a counter to the fourth bar, whence another counter is sent up to the fifth. This complex process of extraction is therefore completed, and the root sought is the Ternary Expression for the number forty-three. _ Let the number to be extracted be arranged on the Quaternary Scale. It will occupy three pairs of bars, and consequently its root must likewise stand on ¢hree bars. ; 36 PALPABLE | EXTRACTION OF Of the first period, consisting of a counter on the sixth bar, and ¢hree counters on the fifth, the nearest inferior square root is evidently wo; which being squared, gives four counters for the fifth bar, or one counter for the QUATERNARY ‘CALE, 6 4 : | : 6 | sixth. There are left after the subtraction, three coun- ters on the fifth bar, to which are subjoined, on the third bar, zhree counters brought down. In this remainder, the Jour counters of the third bar, or the first part of the root doubled, is contained twice. On the second bar, therefore, two counters are placed, both after the root and after the di- visor, which is multiplied by them. The product, express- ed by wo counters on the fifth bar, and one on the fourth, is next subtracted, leaving the fourth and third bars each oc- cupied by three counters. The last period, consisting of ?wo counters on the second, and one on the first bar, is now annexed, to complete the divi- dend. The two first bars of she root, considered as a dis- tinct group, are likewise dou- bled for the divisor, by repeat- ing with dots prefixed the two counters of the second bar. i OB —eesees -~————__—_—_—---@'-@-@-@- —@-@- SQUARE ROOT. ] ARITHMETIC. 87 This compound divisor is contained ¢hrice ; and, conse- quently, having three counters annexed to it, the whole is multiplied by the same three subjoined likewise to the root. The product gives one counter to the first bar, and sends two to the next, which hence acquires ¢wo counters, and transmits three to the third bar, where these three are dropped, and three more conveyed to the fourth bar. The operation consequently terminates, and the root thus obtained corresponds, as before, to forty-three. Not to multiply illustrations, let the same process be per- formed on the Quinary Scale, employing, however, open counters for the sake of simplicity. The original number of which the root is to be extracted here occupies partially three pair of bars. But the ¢hree counters of the highest pe- riod have two for their nearest superior root. Consequently, while #wo counters are placed on the third bar as the first portion of the root, four counters, being their square, are set under those three. The subtraction leaves an open counter on the fifth bar, to which is now subjoined ano- ther open counter brought down on the third bar. In this remainder the divisor, or doubled member of the root, is contained once ; and consequently an open counter is on the second bar annexed to that divisor, and likewise to the root. But this open counter being multiplied into the compound divisor, first changes the open counter above it into a full one, and then converts the four open counters preceding it into as many full counters. ‘The subtraction therefore leaves an open counter on the fourth bar, and two such on the third, to which are subjoined a single open coun- ter on the first bar. The two highest bars of the root, viewed as forming a distinct member of it, are then dow- 88 PALPABLE [EXTRACTION OF bled, for the divisor ; that is, the QUINARY SCALE. open counter on the second bar is again repeated with a dot placed before it. But this com- pound divisor being now con- tained twice in the remainder, two open counters on the first bar are annexed to the root, and again placed above it. ‘The pro-' ductofthe multiplication by those two lower open counters consists, therefore, of four full counters on the first and second bars, and eight open counters on the third bar, which leave three on that -@@-@-@- 9-0-0- bar, and deliver another open counter to the fourth, There is, however, no remainder; for an open counter joined to the two on the third bar of the minuend would be balanced by jive full ones on the second bar, leaving, consequently, another full one, which is equivalent to five such on the first bar where the open. counter reduces them to four, the very same as in the subtrahend. : Lastly, in performing this extraction on the Denary Scale, the notation at least will be somewhat abbreviated, by adopting open counters. ‘The given number now ranges Co | SQUARE ROOT.] — ARITHMETIC. 89 : four open ones on the third, being sub- : : on two pair of bars ; but of the highest | DENARY’SCALE. period, the nearest root is four, the square of which, or szxteen, denoted by two full counters on the fourth bar, and tracted, leave two open counters. The next period consisting of five full coun- ters on the second bar, and an open one on the first, is then subjoined to the ?wo full counters on the third. To ob- tain the corresponding divisor, the four counters of the root are doubled, and consequently are expressed, by placing #wo open counters on the same bar, and a full one on the third bar. The divisor being contained thrice, in this remain- der, three counters are accordingly pla- ~ ced both after that divisor and after the root itself. The product of the multi- plicand, though differently expressed, is obviously the same as the minuend. The operation ends here, and the square root sought is consequently forty-three. ~ -@e-0— 0-0-0-00-0—|.__—__+- © —__-—|—_@|-e-e-e-6|—-0-0- |____|_-9-|,--@-@-6le-@-e— —O-—' o-— -@ e- ——o ———-O-—-- @-0-0— It is easy to perceive, that if the process of decomposi- tion should not terminate, the extraction of the square root may, on the same principles, be pursued through all the descending bars, including a pair of those bars at each step. But to elucidate more clearly the mode of procedure when fractions are concerned, we shall take one or two ex- amples, which involve no integral part. Thus, suppose it were required to discover an approximation to the square root of the fraction one-third. Represented on the Ternary 99 PALPABLE [EXTRACTION OF Scale, this would evidently be denoted by a single counter on the first descending bar. Let every second bar, begin- ning with that of units, be marked by asterisks, to distin- guish the successive pairs in descent. The counter on the second bar, corresponding to ¢hree on the next bar, has consequently ove for its nearest root. ‘The square of this, again, throws one to the third bar, where, consequently, ¢wo counters are left TERNARY SCALE. after subtraction. The counter of the root is now doubled for the divisor; and, being contained twice in the remainder, two counters are sub- joined, both to the |@ 1 i 11 1) ) 1 Fast Me td the diviett Theda clo eee amr tae sor itself. Multiply- ing now that com- pound divisor by the two counters below it, four counters are thrown to the fourth descending bar, which e-o- -@-8- e ‘6-@ @-6_@ @ >) 4 —®,6-¢ oe ——@ oat 8 wee, | es drop one, therefore, and send another to the bar above, where the five counters leave ioe! | ¢ lst —e e-e- @-@- two, and convey one 6 | to the second bar. | On subtracting this S | | | product, only two ne re ee ee counters remain on the fourth bar; but the root, after having been doubled for a new divisor, is not souaRE RooT.] ARITHMETIC. 91 contained in excess, till after an interval of two periods, when it occurs once. A single counter, therefore, on the fourth bar, is subjoined both to the root and to the divisor, which is all repeated four bars lower, or condensed into the expression of single counters on the fourth, sixth and eighth bars. ‘The remainder of this third subtraction is considerable, being signified by fwo counters on the fifth bar, one on the sixth and ¢wo on the seventh and on the eighth bar. In this quantity, the corresponding divisor is contained fwice ; wherefore two counters on the fifth bar are annexed both to that divisor and to the root, and, the raultiplication being performed, leaves yet a small remain- der. The process of decomposition may, therefore, be continued indefinitely, though it has already approxima- ted within the five hundredth part of the truth. Another example may be deemed sufficient. Let the fraction one-third QUATERNARY SCALE. be transferred to the Quaternary Scale, and it will evidently be de-. noted by a train of single counters, beginning at the second, and run- ning down thro’ all the rest of the bars. Consequent- ly the nearest root of the first period is two, which, be- ing squared, give Jour to the third bar, or one to the second, The dif- 92 PALPABLE | ference, with the next period subjoined, exhibits three suc- - cessive counters. The root being now doubled for a divi- © sor, gives four to stand on the second bar, or one on the — first; which, being obviously contained once on the re- mainder, a counter is, on the third bar, annexed to the © root, and likewise to the divisor itself. This divisor: is, therefore, set down two bars lower, and, being sub- tracted, leaves a counter on the fourth bar, to which the next period is subjoined. Here the modified divisor — is very nearly contained once again, and leaves an open counter on the sixth bar, which admits of no division till — the two succeeding periods are joined to it. In this com- pound quantity the divisor is once found, and consequent- ly an open counter on the sixth bar is annexed both to the root and to the divisor itself. There is still a small remainder ; but the extraction has already been pushed so far as to approximate within the five thousandth part ef the true root. Such is the natural process of analysing numbers, and of variously combining and separating them; and such are likewise the simpler modes of abridging the la- bour of computation. From the copious illustrations which have been given, it appears that PaLpasLe AniTH- metic is capable, if skilfully conducted, of being applied, with considerable facility, to a wide range of combina- tions. All nations have, accordingly, at different periods, employed that symbolical method of Catcutation, which is indeed perpetuated in the term itself. The Egyp- # ARITHMETIC. 93 tians performed their computations merely by the help of pebbles ; and so did the Greeks for a lapse of ages. But while the latter, as Herodotus acquaints us, proceeded in such operations from left to right, always descending from the higher part of the number expressed, the for- mer used, both in writing and counting, to advance in the opposite direction, or from right to left, as still prac- tised very generally over the East. In the schools of an- cient Greece, the boys acquired the elements of knowledge by working on the 424X, a smooth board with a narrow rim; so named, evidently, from the combination of the three first letters of their Alphabet, and resembling the tablet, likewise called A, B, C, on which the children with us were accustomed to begin to learn the art of reading. The pupils, in those remote times, were instructed to calculate, by forming progressive rows of counters; which, accord- ing to the wealth or fancy of the individual, consisted of - small pebbles, of round bits of bone or ivory, or even of silver.coins. From the Greek and Latin word for a peb- ble, comes, in either language, the verb signifying to com- pute. ‘The same board, strewed with fine green sand, a colour soft and agreeable to the eye, served equally for teaching the rudiments of writing and the principles of Geometry. To their calculating board, the ancients make frequent allusions. It appears, that the practice of bestowing on pebbles an artificial value, according to the rank or place which they occupied, remounts higher than the age of Solon, the great reformer and legislator of the Athenian commonwealth. This sagacious observer and disinter- ested statesman, who was howeyer no admirer of regal 94 PALPABLE government, used to compare the passive ministers of Kings to the counters or pebbles of Arithmeticians, which, according to the place they hold, are sometimes most important, and at other times utterly insignificant. The Grecian orators, in speaking of balanced accounts, pice ture this termination, by saying that the pebbles were cleared away, and none left. It is evident, therefore, that — the ancients, in keeping their accounts, did not separately _ arrange the credits and the debts, but set down pebbles for the former, and took up others for the latter. As soon as the board became cleared, the opposite claims were exactly balanced.—It may be observed, that the common phrase to clear one’s scores or accounts, meaning to settle or ad-— just them, still preserved in the popular language of Europe, was certainly suggested by the same practice of reckoning with counters, which prevailed indeed until a_ comparatively late period. The Romans borrowed their Abacus fromthe Greeks, and never aspired higher in the pursuit ofnumerical science. To each pebble or counter required for that board, they gave the name of calculus, a diminutive formed from the word — signifying @ white stone; and applied the verb calculare, to express the operation of combining or separating such pebbles or counters. Hence innumerable allusions by the Latin authors. The use of the Abacus, called sometimes likewise the Mensa Pythagorica, formed an essential part of the education of every noble youth. A small box or coffer, called a Loculus, having compartments for holding the calculi or counters, was considered as a necessary ap- pendage. Instead of carrying a slate and satchel, as in modern times, the Roman boy was accustomed to trudge ARITHMETIC. 95 to school, loaded with those ruder implements—his arith- metical board, and his box of counters. In the progress of luxury, ¢alz, or dies made of ivory, were used instead of pebbles, and small silver coins came to supply the place of counters. Under the Emperors, every patrician living in a spacious mansion, and indul- ging in all the pomp and splendour of eastern princes, generally entertained, for various functions, a numerous train of foreign slaves or freedmen in his palace. Of these, the librarius or miniculator, was employed in teach- ing the children their letters; but the notarius registered expenses, the rationarius adjusted and settled accounts, and the ¢abularius or calculator, working with his counters and board, performed what computations might be re- quired. - To facilitate the working by counters, the construction of the Abacus was afterwards improved. Instead of the perpendicular lines or bars, the board had its surface di- vided by sets of parallel grooves, by stretched wires, or even by successive rows of holes. It was easy to move small counters in the grooves, to slide perforated heads along the wires, or to stick large knobs or round-headed nails in the different holes. ‘To diminish the number of marks required, every column was surmounted by a short- er one, wherein each counter had the same value as five of the ordinary kind, being half the index of the Denary Scale. The Abacus, instead of wood, was often, for the sake of convenience and durability, made of metal, fre- quently brass, and sometimes silver. ‘Two varieties of this instrument seem to have been used by the Romans. Both of them are delineated from antique monuments,—the first kind by Ursinus, and the second by Marcus Velserus. 96 PALPABLE In the former, the numbers are represented by flattish per. forated beads, . ranged on pa~ rallel wires ; and, in the lat- ter, they are sig- nified by small round counters moving in pa- rallel grooves, as represented in the figure an- nexed. ‘These instruments contain each seven capital die visions, expressing in regular order unzts, tens, hund) ‘eds, thousands, ten thousands, hundred thousands, and millions. For the sake of abbreviation, a,similar set of shorter grooves, following the same progression, but having jive times the relative value, are made to range immediately above them. With four beads on each of the long grooves” or wires, and a single bead on every corresponding short one, it is evident that any number could be expressed, as far as ten millions. ets In the Roman Abacus, the arrangement of the Denary Scale is uniformly followed; but there is, besides, a small ap- pendage tothe subdivisions, founded on the Duodenary Sys- tem. Immediately below the place of units, is added a bar, with its corresponding branch, both marked 9, being design- ed to signify ounces, or the twelfth parts of a pound. ive beads on the long wire, and one bead on theshort wire, equi- valent now to szz, would therefore denote eleven ounces. To express the simpler fractions of an ounce, three ver short bars are annexed behind the rest; a bead on the ARITHMETIC. 97 ene marked S or 3, the contraction for Semissis, denoting half-an-ounce ; a bead on the other, which is marked by the inverted 9, the contraction for Sicilicum, signifying the quarter of an ounce ; and a bead on the last very short | bar, marked , a contraction for the symbol % or Bine Sextula, intimating a duella or two-sixths, that is, the third part of an ounce.—'The second form of the abacus, here ‘patie differs in no esssential respect from the first, grooves merely supplying the place of parallel wires, and | coins, or ivory counters, being substituted for the perfora- ted beads. The Romans likewise applied the same word Abacus, | to signify an article of luxurious furniture, resembling in shape the arithmetical board, but often highly ornamental, and destined for a very different purpose,— the relaxation and the amusement of the opulent. It was used in a game apparently similar to that of chess, which displayed a live- ly image of the struggles and vicissitudes of war. The infamous and abandoned Nero took particular delight in this sort of play, and drove along the surface of the Aba- cus with a beautiful guadriga, or chariot of ivory. The Chinese have, from the remotest ages, used in all their computations an instrument similar in shape and con- struction to the Abacus of the Romans, but more complete and uniform. It is admirably fitted for representing the decimal system of measures, weights, and coins, which prevails throughout their vast Empire. The calculator could begin at any particular bar, and reckon, with the same facility either upwards or downwards, through the whole range, which includes ten bars. ‘This advantage of treating fractions exactly like integers, is of the utmost con- sequence in practice. Accordingly, those arithmetical ma- H 98 PALPABLE chines, of very different sizes, have been adopted by all ranks, from the man of letters to the humblest shopkeep- er, and are constantly used in all the bazars and booths of Canton and other cities, being handled, it is said, by — the native traders with a rapidity and address which quite — astonish the European factors. i The civil arts of Rome were isis umtaeited to other — nations by the tide of victory, and maintained through the — vigour and firmness of her imperial sway. But the sim- — pler and more useful improvements survived the wreck of empire, among the various people again restored by fore — tune to their barbarous independence. In all transactions — wherein money was concerned, it was found convenient f to follow the procedure of the Abacus, in representing numbers by counters placed in parallel rows. During the — middle ages, it became the usual practice over Europe for merchants, auditors of accounts, or judges appointed to decide in matters of revenue, to appear on a covered bank or bench, so called from an old Saxon or Franconian word signifying a seat. The term scaccarium, a Latinised— oriental word, from which was derived the French, and thence the English, name for the Exchequer, anciently indicated merely a chess-board, being formed from scaccum, denoting one of thé moveable pieces in that intricate game. The reason of this application of the term is sufficiently obvious. ia ‘The Court of Exchequer, which takes cognisance of all : questions of revenue, was introduced into England by the — Norman Conquest. Fitz-Nigel, in a dialogue on the sub- ject, written about the middle of the twelfth century, says. that the scaccarium was a quadrangular table about ten feet long and five feet broad, with a ledge or border about four inches high, to prevent any thing from rolling over; ARITHMETIC. 99 and was surrounded on all sides by seats for the judges, - the tellers, and other officers. It was covered every year, after the term of Easter, with fresh black cloth, divided _ by perpendicular white lines, or distinctures, at intervals of about a foot or a palm, and again parted by similar transverse lines. In reckoning accounts, they proceeded, he subjoins, according to the rules of arithmetic, using small coins for counters. ‘The lowest bar exhibited pence, the one above it shzllings, the next pounds ; and the high- er bars denoted successively tens, twenties, hundreds, thou- sands, and ten thousands of pounds; though, in those early times of penury and severe economy, it very seldom hap- pened that so large a sum as the last ever came to be rec- koned. ‘The first bar, therefore, advanced by dozens, the second and third by scores, and the rest of the stock of bars by the multiples of zen. The teller sat about the middle of the table; on his right hand, eleven pennies were heap- ed on the first bar, and a pile of nineteen shillings on the second; while a quantity of pounds was collected oppo- site to him, on the third bar. For the sake of expedition, he might employ a different mark, to represent half the value of any bar, a silver penny for ten shillings, and a gold penny for ten pounds. In early times, a chequered board, the emblem of calcu- lation, was hung out, to indicate an office for changing money. It was afterwards adopted as the sign of an inn or hostelry, where victuals were sold, or strangers lodged and entertained. We may perceive, particularly in the South, traces of that ancient practice existing even at pre- sent. | The use of the smaller abacus in assisting numerical computation was not unknown during the middle ages. In England, however, it appears to have scarcely entered 100 ~PALPABLE into actual practice, being mostly confined to those few individuals, who, in such a benighted period, passed for men of science and learning. The calculator was styled, in correct Latinity, abacista ; but, in the Italian dialect, abbachista, or abbachiere. A different appellation was afterwards introduced by the Arabians, who conquered Spain, and enriched that insulated country by commend- able industry, where they likewise introduced their mathe- matical science. Having adopted an improved species of numeration, to which they gave the barbarous name of algarismus, algorismus, or algorithmus, from their definite article a/, and the Greek word for number, this compound term was adopted by the Christians of the West, in their — admiration of superior skill, to signify calculation in ge- neral, long before the peculiar method of performing it had become known and practised among them. The term algarism was corrupted in English into augrim or awgrym, and applied even to the pebbles or counters used in ordi- nary calculation. The same word, algorithm, is now ap-— plied by mathematicians, to express any peculiar sort of | notation. The Abacus, with its symbolical furniture, had been adopt- ed merely as an instrument for expediting the process of computation. But it became likewise necessary to have recourse to some readier and simpler modes of expressing numbers. A very ancient practice, though quite arbi-— trary in its principle, consisted in employing the various disposition of the fingers and the hands, to signify the ) numerical series. On this narrow basis, a system was framed of very considerable extent. By a single inflexion of the fingers of the left hand, the ; Romans proceeded as far as ten ; and by combining ano- | ther inflexion with it, they could advance to an hundred. On the right hand, the same signs being augmented an — a ARITHMETIC. 101 hundred fold, carried them as far as a thousand, and ten thousand; and, by another extension, those signs variously referred to the head, the throat ;—or on ei- Oy: ther side, to the breast, the stomach, the waist, or the thigh, were a- gain multiplied an , hundred times, and Three ie i; consequently raised to the extreme limit of a million.— Thus, with the fingers of the left Frye hand, the first set of inflexions, as in the Six specimen here annex- ed, denoted the nine digits, and the se-= cond set of them res Eight presented the nine decads; but, perform- Nine ed on the right hand, the former Maacaition of the fingers indicated hundreds, and the latter was made to signify thousands. In this numerical play, the Romans especially had, from constant experience, acquired great dexterity and = Ten ys a Ra eee The >> Ou? yin VARSED Seven address. Many allusions to the practice are made in the writings of their poets and orators; and, without some _ knowledge of the principle adopted; many passages of the classics would-lose their whole force. This kind of pan- _ tomime even outlived the subversion of the Western Em- pire, and was particularly suited to the slothful habits of the religious orders who fattened on its ruins, and, relin- 108 PALPABLE ARITHMETIC. quishing every manly pursuit, recommended silence as a — virtue, or enjoined it as a duty. These signs were merely fugitive, and it became neces- _ sary to adopt other marks, of a permanent nature, for the — purpose of recording numbers. But of all the contrivan-— ces adopted with this view, the rudest undoubtedly is the _ method of registering by ¢allies, introduced into England along with the Court of Exchequer, as another badge of % ‘the Norman Conquest. These consist of straight well-_ seasoned sticks, of hazel or willow, so called from the French verb tailler, to cut, because they are squared at each end. The sum of money was marked on the side with notches, by the cutter of tallies, and likewise inscri- bed on both sides in Roman characters, by the writer of the tallies. The smallest notch signified a penny, a lar- ger one a shilling, and one still larger a pound; but — other notches, increasing successively in breadth, were made to denote ten, a hundred, and a thousand. ‘The stick was then cleft through the middle by the deputy-cham- berlains, with a knife and a mallet ; the one portion being — called the tally, or sometimes the scachia, stipes, or kan- cia; and the other portion named the counter-tally, or. folium. This strange custom might seem the practice of untu- tored Indians, and can be compared only to the rude simplicity of the ancient Romans, who kept their diary by means of Japilli or small pebbles, casting a white peb- ble into the urn on fortunate days, and dropping a black. one when matters looked unprosperous; and who sent, at _ the close of each revolving year, their Praetor Maximus — with great solemnity to drive a nail in the door of the — right side of the temple of Jupiter, and next to that of Minerva, the patron of learning, and the inventor of numbers. FIGURATE ARITHMETIC. 7 science of number received its capital improve- ment in the adoption of a ready and comprehensive system of characters, not only qualified to exhibit directly the widest range of objects, but also fitted as instruments for facilitating every process of calculation. This might ap- pear an easy step of advancement from the practice of the Abacus, since it was only requisite to devise the few sym- bols wanted in expressing the counters that could occupy any single bar of the common or denary scale. But the simplest and most valuable discoveries are seldom the first achieved. The transition made in the art of writing from the use. of emblematic figures, to that of arbitrary signs chosen to express merely the sounds of a conventional lan- guage, must have, on the whole, contributed to retard the progress of numeral notation. The system of characters among the Romans, especially after it had changed its pri- mitive forms into the letters most analogous, was so complex and unmanageable, as to reduce them to the necessity, in all cases, of employing the Abacus. The structure of the Greek numerals, though founded entirely on the distri- 104. FIGURATE bution of the alphabet, was more pliant and _ refined, Three sets of letters carried the direct notation to a thou. sand ; and, by punctuating the first series, its power was extended as far as en thousand, or four places of the de nary scale. To represent the largest numbers, the prin- ciple of position, after having been so long abandoned, was again resumed, and the same quaternion or period of progressive myriads was repeated in continual ascent. In this improved arrangement, the value of the characters depending on their place or rank, it became essential to mark or fill up the accidental vacuities in the scale of nue meration, and a small 0, insignificant by itself, was aceord- ingly adopted for that very purpose. t, The system of numerals, thus finally moulded by the Greek astronomers, though cumbrous and redundant in its structure, had therefore attained a high degree of per- ' fection, and was capable, with due labour and patience, of performing the most complex operations in Arithmetic. The extent of their alphabet was favourable to the first attempts at numeration ; since, with the help of only three intercalations, it furnished characters for the whole range below a thousand. This very circumstance, however, proved a bar to future improvements. The first series of letters was already distinguished from the two succeeding classes, in being employed with subscribed points to de- note thousands, and complete the quaternary period. It might hence appear as no violent, yet a most important, innovation, if only those letters denoting the nine digits had been retained, and the rest signifying tens and hun- dreds entirely dismissed. By such a change, the arith- metical notation of the Greeks would have reached its ut- ARITHMETIC. 105 most term of simplification, and have exactly resembled indeed our own. Had the genius of that people not suf- fered a fatal eclipse, they must have soon passed the few barriers which remained to obstruct their progress. Some writers, misled by very superficial views of the subject, have still ascribed the invention of the modern numeral characters to the Greeks, or even to the Romans. Both these people, for the sake of expedition, occasionally used contractions, especially in representing the fractions of weights and measures, which, to a credulous peruser of mutilated inscriptions, or ancient blurred manuscripts, might appear to resemble the forms of our ciphers. But this resemblance is merely casual, and very far indeed from indicating the adoption of a regular denary notation. The most contracted sort of the Roman writing was form- ed by the marks attributed to the freedman Tiro, and to Seneca the philosopher, while that of the Greeks was mixed with the symbols called Szgle ; both of which have exercised the patience and skill of our Antiquaries and Di- plomatists. In the latter species of characters, were kept the accounts of the revenues of the Empress Irene at Con- stantinople. The Arabians and Persians likewise employ- ed a concise but arbitrary mode of representing the larger numbers, by means of abbreviated words, a practice which still prevails in the East. The Chaldean astronomers, and their successors, in the Lower Empire, although ac- guainted with the simple and elegant system of the ancient Greek notation, yet preferred in certain cases a sort of tachygraphical marks, extremely abbreviated, but entirely conventional and arbitrary in their formation. These artificial characters were formed merely by a single broad 406 FIGURATE stroke, with a smaller branch guns inserted, as in the specimen below : Ae) yreep pe TAQVTD944 ‘mone 60 70 ; LEPKNER hhh 100 400 500 600 800 900 — TAidAdddad 1000 5000 4000 5000 6000 7000 8000 9000 One million was denoted, by merely doubling the chas racter for a thousand.—The principal stroke could have either a perpendicular or an horizontal position ; which gave occasion to two distinct ways of tracing those marks. Such, undoubtedly, were the characters which John Ba- singstoke, archdeacon of Leicester, according to the rela- tion of Matthew Paris or his continuator, brought into Eng- land, and communicated to some of his friends, as a pre- cious acquisition, shortly before his death in 1252. Had_ that eminent person, whom the thirst of knowledge in a_ dark age led to visit the East, and to study the Greek lan- guage amidst the ruins of Athens, been at all acquainted with the nature of the Arabic numerals, he must have per- ceived the comparative futility of every plan which aimed at mere abbreviation. It cannot be doubted, that we derived our knowledge of the numeral digits from the Arabians, who call the system _ Hindasi, and confess their having obtained this invaluable — acquisition from an extended intercourse with the East. But for want of precise historical evidence, much obscurity ARITHMETIC. 107 still hangs over the whole subject. If the exuberant fan- cy of the Greeks led them far beyond the denary nota- _ tion, it seems probable, that the feebler genius of the _ Hindus might just reach that desirable point, without di- verging into an excursive flight. ‘Though now familiar _ with that system, they are still mvieqnibinited with the — use of its descending decimal scale; and their manage- ment of fractions, accordingly, is said ‘die eae judges | to be tedious and embarrass- | ed. We give here the San- , 59 oR ng op 8? | eet digits, in pba are call-- 9 > : ° y ef Ob 896 ed the Devanagari charac- 125465678 910 ter, and place over them the numeral elements, consisting oF a succession of nine simple strokes, variously combined, from which they may, with great probability, be supposed to have been formed ; al- though some Arabian authors, who treat of astrological signs, allege that the Indian numerals were derived simply from the quartering of a circle. “The resemblance of those natural marks to the derivative, appears certainly very strik- ing. From the latter, the common Hindu ie here sub joined, and the vulgar Ben- galee, are evidently moulded, q ecg © y €9 CY 90 with onlyslight alterationsof ' 7 % 4 Spr ONeh Raw ata form. The Birman figures are of the same ol igin, but have a thin, wirey body, being generally written on the palmyra-leaf with the point of a needle. It appears, from a careful inspection of the manuscripts preserved in the different public libraries of Europe, that the Arabians were not acquainted with the use of the de-. nary numerals, or at least had not generally employed or adopted them, before the beginning of the thirteenth cen- 108 FIGURATE tury of the Christian era. They cultivated the mathema- tical sciences with ardour, but seldom aspired at original efs forts, and generally contented themselves with copying their Grecian masters. The Cufic letters, resembling the Syriac, were introduced a little before the time of Mahomet ; but the characters used at present, and called Neskhi, was invented about three hundred years afterwards. This alphabet the Arabians employed to express numbers, ex- actly in the same way‘as the Greeks. The letters, in their succession, were sometimes applied to signify the lower of the ordina] numbers; but more generally they were distin- — guished into three classes, each composed of nine charaec- ters, corresponding to units, tens, and hundreds. Though, | like most of the Oriental nations, the Arabians write from right to left, yet they followed implicitly the Greek mode of ranging the numerals and performing their calcula- tions. For several ages, they hired Christian scribes to keep their accounts; and Walid, Khalif of Syria, pro-— hibited any other marks for numbers to be used than those — of the Greeks. With the same deference, they received the other lessons of their great masters, and very seldom | hazarded any improvement, unless where industry and ) patient observation led them incidentally to extend Men- — suration, and to rectify and enlarge the basis of Astronomy. The Arabians might have received their information of © the Hindu mode of notation through the medium of the : Persians, who spoke a dialect of their language, had embra= ced the same religion, and were, like thems inflamed by the love of science and the spirit of conquest. The Ara- ps $4 VAQ te 123 4 i 6 7 bic numerals, here annexed, 4 8 9 10 though derived apparently from the inversion of the San=_ ARITHMETIC, _ 109 scrit digits, resemble more strikingly the Persic, which are now current over India, and there esteemed the fashion- able characters. But the Persians themselves, though no longer sovereigns of Hindostan, yet discover their superiority over the feeble Gentoos, since they gene- rally filled the offices of the revenue, and enjoyed the reputation of being the most expert calculators in the East. It should be observed, however, that, according to Gladwin, these accountants have introduced a peculiar contracted mode of registering very large sums, partly by the numeral characters, and partly by means of symbols formed with abbreviated words, somewhat analogous to the Chaldaic marks. The Indian origin of the denary numerals is farther confirmed, by the testimony of Maximus Planudes, a monk of Constantinople, who wrote, about the middle of the fourteenth century, a book on Logistics, or Practical Arith- metic, entitled, “ Zhe great Indian Mode of Calculating.” ‘In the introduction, he explains concisely the use of the -eharacters in notation. But Planudes appears not to have received his information either directly from India, or ‘through the channel of the Persians, the nearest neigh- bours on the eastern confines of the Greek Empire. It is most probable that he was made acquainted with those ‘numerals by his intercourse with Europe, having twice visited, on a sort of embassy, the Republic of Venice; for, of two manuscripts preserved in the library of St Mark, the one has the characters of the Arabians, and the other has that variety which was first current in Europe, while neither of them shows the original characters used in Hin- dostan. ’. It is a more important inquiry to ascertain the period ~ 110 FIGURATE when our present numerals were first spread over Europe, As it certainly preceded the invention of printing, the dif- ficulty of coming to a clear decision is much increased, by the necessity of examining old and often doubtful ma- nuscripts. Some authors would date the introduction of those ciphers as early as the beginning of the eleventh century, while others, with far greater appearance of reas son, are disposed to place it two hundred and fifty years later. . | i While the thickest darkness brooded over the Christian world, the Arabians, reposing after their brilliant con- quests, cultivated with assiduity the learning and science of Greece. If they contributed little from their native stores, they yet preserved and fanned the holy fire. Nor did they affect any sort of concealment, but freely com- municated to their visitors that precious knowledge which they had so zealously drawn from different quarters. Some of the more aspiring youths in England and France, disgusted with the wretched trifling of the Schools, resort- ed for information to Spain; and having the courage to subdue the rooted abhorrence entertained in that age against the infidels, took lessons in philosophy from the enlightened Moors. ‘The great objects of study were the Algarithm and the Almagest, terms derived from the Greek with the Arabic definite article prefixed to them; the for- mer including the rules of Arithmetic, and the latter com- | prehending the principles of Astronomy. Among the earliest of those who performed such a pil+ grimage, was the famous Gerbert, born of obscure parents — at Aurillac, in Auvergne, but promoted by his talents, from the condition of a monk, successively to the Bishoprick of Rheims and of Ravenna, and finally elevated to the Papal. Chair, which he filled during the last. four years of the ARITHMETIC. 11k tenth century, under the name of Sylvester II. This ar- dent genius studied Arithmetic, Geometry, and Astronomy among the Saracens ; and on his return to France, char- ed with various knowledge, he was esteemed a prodigy of learning by his contemporaries; nor did the ignorance ‘and malice of his clerical brethren fail to represent him as a magician, leagued with the Powers of Darkness. Ger- _ bert wrote largely on Arithmetic and Geometry, and gave rules for shortening the operations with the Abacus. In some manuscripts, the numbers are expressed in ciphers; but they had evidently crept, in through the licence of transcribers, and it would be most unwarrantable thence to conclude, as many writers have done, that Gerbert had actually the merit of introducing those characters into Europe. The same remark will extend likewise to our celebrated countrymen, Roger Bacon, and- John of Hali- fax, or Holywood, and therefore styled Sacro-Bosco in the rude Latinity of that age, who flourished indeed about three centuries later, but must have derived their information, — though perhaps not directly, from the same source. Bacon wrote on the reformation of the kalendar, yet he has given no proofs of his acquaintance with the denary nota- tion. Sacro-Bosco composed a treatise on the Sphere, which was long held as a standard work in the Schools, In the latter copies of that book, numeral characters had been sometimes inserted. The Digital Arithmetic, conjoined with the higher art of Algebra, seems to have been first brought into Europe by the zeal of Leonardo Bonacci, of Pisa, a wealthy mer- chant who traded to the coast of Africa and the various ports of the Levant. Commercial speculations haying tempted him frequently to visit those countries, he was 112 FIGURATE induced, by the love of knowledge, to study thoroughly the science of calculation among the Arabians. On his return to Italy in 1202, this meritorious person com- posed an Arithmetical Treatise, which he greatly en- larged in 1228. But typography had not yet lent its magic aid to the multiplication of thought, nor do the Tuscans, though long reputed the best calculators in Italy, and consequently in Europe, and to whom we owe the method of Book-keeping, appear to have derived their skill from an acquaintance with the writings of Bonacci. His manuscript had lain more than two centuries neglect- ed, till Lucas Paccioli, or de Burgo, instructed chiefly by its perusal, published, successively between the years 1470 and 1494, the earliest and most extensive printed treatise on Arithmetic and Algebra. The term cifer, or more correctly sipher, as appro- priated to the digital characters, is an Arabic word in- troduced by the Saracens into Spain, signifying to enume- rate. There is little doubt that the Arabic figures were first used by astronomers, and afterwards circulated in the almanacs over Europe. ‘The learned Gerard Vossius places this epoch about the year 1250; but the judicious and most laborious Du Cange thinks that ciphers were unknown before the fourteenth century; and Father Mabil- lon, whose diplomatic researches are immense, assures us, that he very rarely found them in the dates of any writings prior to the year 1400. Kircher, with some air of pro- bability, seeks to refer the introduction of our numerals to the astronomical tables which, after vast labour and ex- pense, were published by the famous Alphonso, King of Castile, in 1252, and again more correctly four years afterwards. But it is suspected, on very good grounds, ARITHMETIC. 118 that, in the original work, the numbers were expressed by Roman or Saxon letters. In the Arch@ologia, there is given a short account of an almanac preserved in the library of Bene’t College, Cambridge, containing a table of eclipses for the cycle be- tween 1330 to 1348. It has prefixed to it a very bricf ex- plication of the use of numerals, and the principles of the denary notation; from which may be seen how imper- fectly the practice of those ciphers was yet understood, The figures are of the oldest form, but differ not materially from the present, except that the four has a looped shape, and the five and seven are turned about to the left and to the right. The one, two, three, and four, are likewise, perhaps for elucidation, represented by so many dots, thus, . .. .. 3:5 while five, six, seven, and eight, are signified by a semicircle or inverted 9 with the addition of corresponding dots—9g 9: 9:9 -. Nine is de- noted by 0; ¢en by the same character, with a dash drawn across it; and twenty, thirty, or forty, by this last symbol repeated. As a farther evidence of the inaccurate conceptions which prevailed respecting the use of the digits in the fourteenth century, we may refer to the mixture of Saxon and Arabic numerals copied from some French manu- scripts by Mabillon. The or- 6 o% dinary series is thus express- ) a afi a olga be ed, the Saxon y being here employed to denote ten, and repeatedly combined with the common digits; nay, rrr and yyyi are immediately followed by 302 and 303, which were therefore intended to signify thirty-two and thirty- three, the force of the cipher not being yet rightly under- stood.—It. should be observed, that the Greek Episémon I 114 FIGURATE or Fau, for the number siz, had come to be represented: by a character similar to G. One of the oldest authentic dates expressed in numeral characters is thatof the year 1375, which seems to have been written by the hand of the famous Petrarch on a copy of St Augustin, that had belonged to that distinguished Poet and Philosopher. The use of those characters was only beginning to spread in Europe, and still confined to men of learning. A little tract in the German language, entitled, De Algorismo, and bearing the date 1390, explains, with great brevity, the digital notation and the elementary rules of Arithmetic. Atthe end ofa short Missal, similar directions are given in verse, which, from the form of the writing, may be judged to belong to the same period. Here annexed is a correct fac simile of the digits themselves, with the B-9:B-N:O-G.25-2-]- 0, 9, 8, 7, 6, 5, 4,5, % 1 modern figures placed below ~~ = | ; them; but what is very remarkable, the characters in both manuscripts range from right to left, the order which the Arabians would naturally follow. It was not very easy to comprehend at first the precise force of the cipher, which, insignificant by itself, only serves to determine the rank and value of the other di- gits. A sort of mystery, which has imprinted its trace on language, seemed to hang over the practice of numera- tion, for we still speak of deciphering, and of writing in cipher, in allusion to some dark or concealed art. After the digits had come to supply the place of the Roman numerals, a very considerable time probably elapsed be- fore they were generally adopted in calculation. The modern practice of arithmetic remained unknown in Eng- land, till about the middle of the sjxteenth century; and ARITHMETIC. - 115 the lower orders, imitating the clerks of a former age, were still accustomed to reckon by the help of their awerym stones. In Shakespear’s comedy of the Winter’s Tale, written at the commencement of the seventeenth century, the clown, staggered at a very simple multiplication, ex- claims that he must try it with counters. It cannot be doubted, that the kalendars composed in France or Germany, and sent to the different religious houses, were the means of dispersing the knowledge of Arabic numerals over Europe. In the library of the University of Edinburgh, there is a very curious almanac, presented to it, with a number of other valuable tracts, by the celebrated Drummond of Hawthornden, beauti- fully written on vellum, with most of the figures pencilled in vermillion. It had been calculated particularly for the year 1482, but contains the succession of lunar phases for three cycles, 1475, 1494, and 1513, with the visible eclip- ses of the sun and moon from 1482 to 1530 inclusive. The date of this precious manuscript, which had once - belonged to St Mary’s Ab- bey at Cupar in Angus, is ) ost VAY fe hence easily determined, and 1732G6A8 910: the numerals now exhibited 12 34 56 n 891% were exactly copied from it. For the He fe comparison, V2 34 i 6A8 90 the corresponding Arabic characters are placed above them, and again below them is given a specimen of the current forms which the digits acquired in England, about the middle of the sixteenth century; the last row show- ing the old figures used by Caxton, when he printed the Mirrour of the World, in 1480. The College accounts in the English Universities were generally kept in the Roman numerals, till the early part 116 FIGURATE of the sixteenth century ; nor, in the parish registers of the South, were the Arabic characters adopted before the year 1600. ‘The oldest date to be met with in Scotland is 1490, which occurs in the rent-roll of the Diocese of St Andrew’s; the change from Roman to Arabic numerals, with a corresponding alteration in the form of the writing, — appearing near the end of the volume. a Having endeavoured to trace the origin and introduc- tion of our numeral characters, it only remains now to explain the operations of Figurate Arithmetic. But the rules which guide the practice of numbers are easily de- duced from the principles already unfolded in treating of Palpable Arithmetic. ‘The same theory may likewise sug- gest other methods of varying and abridging the common operations. I shall follow the same order, selecting as few examples as may be wanted for illustration. The Denary Scale, being the one generally received, will claim the chief attention ; but its results shall be compared with | those of some other scales, particularly the Duodenary, — which is partially adopted in commerce, and possesses cer- tain peculiar advantages. For the expressions of this scale, however, it becomes necessary to devise two additional characters to denote ten and eleven. Not to seek far after such objects, I have contented myself with condensing the ordinary forms into ( and 0, which are perhaps sufficiently. distinct, while they shadow out the figures represented by them. The great recommendation of the Duodenary scale, con- sists in its fitness to denote fractional parts. Its index has indeed no fewer than four factors—2, 3, 4, and 6; while ven is divisible only by 2 and 5. Several attempts, accord- ARITHMETIC. 117 ingly, have, at different times, been made to carry this scale into actual practice. It is a curious fact, that the famous Charles XII. of Sweden, whose views, though often dis- turbed by the wildness of heroism, were on the whole be- neficent, seriously deliberated on a scheme of introducing this system of numeration into his dominions, a very short time before his death, while lying in the trenches, during the depth of winter, before the towering Norwegian for- tress of Frederickshall. In the Denary Scale, for the sake of expedition, it is often convenient to class the digits into members or periods of three places, ascending by the powers of a thousand. ‘This improvement is due to the Italians, who likewise drew from their elegant dialect the appropriate names for such periods: units, thousands, millions, bill- ions, trillions, quadrillions, quintillions, sextillions, sep- tillions, octillions, and nonillions. — Thus, the number 5,482,584,875,284,800, which expresses the square feet in the surface of our globe, is read, 5 quadrillions, 482 trillions, 584% billions, 878 millions, 284 thousands, and 800 units. In many cases, it would facilitate calculation, to have figures corresponding to open counters. To transform the ordinary characters, therefore, into deficient digits, I have caused modify their shape thus: eae eS GeO an alteration sufficient to distinguish, without entirely alter- ing, or disguising, them. With such reclined figures, it will be easy to represent numbers, by their defects as well as their excesses. ‘This answers most conveniently in expressing the digits from 6 to 9 inclusive. ‘Thus 38 may be denoted by 49., meaning 40 with 2 abated; for the same reason, $29 may be written 19°.3\, signifying that 1030 is to be diminished by 201, 118 FIGURATE [NUMERATION. NUMERATION Being only the mode of classing numbers by successive braces, leashes, warps, &c. is performed, by dividing them continually by the root of the scale of arrangement. In the reduction, consequently, to the Binary, Ternary, Quaternary, &c. system, the corresponding divisor is ¢wo, | three, four, &c. As an illustration, let it be required to 1 exhibit the number 430685 on a variety of scales. The decomposition will be thus effected : BINARY, TERNARY. QUATERNARY. 2)430685 3)4.30685 4.430685 143561)2 4:7853|2 1595 1/0 5317/0 1772\1 : 590/2 196|2 65)1 2112 7|0 QUINARY. Z|\l SENARY. 5)430685 SEPTENARY. 6)430685 861370 7)430685 ~ 717805 172272 61526/3 11963)\2 34452 8789/3 1993|5 6890 1255/4 3321 137 179 55/2 27\2 25/4 QL ih 3! 1/3 110 OCTARY. NONARY. UNDENARY. DUODENARY. $)430685 9) 480685 11)430685 12)430685 538355 47853)\8 3915312 35890/5 6729/3 5317/0 35591 29900 84:1|1 5907 39316 2499 105|1 6515 2914 209 13]1 712 Qi7 1! NUMERATION.] ARITHMETIC. 119 Hence the number 430685 will be thus represented on different scales : Binary, ...... }101001001001011101 Octary, os. 1511135 Ternary, .... 210212210022 Nonary, ..... 725708 Quaternary, 1221021131 Denary, .... 430685 Quinary, w+. 102240220 Undenary,.. 274642 DENATY»: seocsees 13121525 Duodenary, 189205 Septenary, ... 344.2433 The notation may be readily transferred to any higher scale, which has for its index some power of that of the lower. Let the given number be distinguished into pe- riods consisting of two, three, or four places ; the amount of these may be condensed into a single figure, and the corresponding index then is the second, third, or fourth power of the primary index. ‘Thus, conceive the ex- pression 1,10, 10, 01, 00, 10,01, 01,11, 01, of the Binary Scale, to articulate at every alternate place from the right hand, the value of each period, or 1 for 01, 3 for 11, and 2 for 10, will transform the whole into this Quaternary arrangement, 1221021131. If the same expression be distinguished into triplets, 1, 101, 001,001, 001, 011, 101, and each of these afterwards compressed into a single figure, assuming 5 for 101, and 3 for 011, it will be chan- ged into the Octary notation 1511135.—In like manner, if the representation of the same number on the Ternary Scale be broken thus, 21,02, 12, 21, 22, 00, 22, at every alternate place, and the values of those periods adopted, 8 being substituted for 22, 7 for 21, and 5 for 12, the whole will be converted into the expression 725708 of the Nonary Scale. To transfer, in general, any numeral expression from one scale to another, the most obvious way is to decompound it, and then dispose it again on the new scale. Suppose it were 120 FIGURATE RUauaRAmaM. required to convert the Ternary expression 210112 into — the Quinary Scale: Beginning at the left hand, and con- stantly tripling the terms, shifting them every time a place lower, the successive collected results are 2, 7, 64, 193, and 581. This final number, if decomposed again in the ordinary mode by a series of pentads, would give 4311. But this conversion might be performed directly with more expedition, by dividing the original expression and the quotients successively which arise, by the index of the new scale, as exhibited in the same notation. Thus, let it be required to transform the Quzn- 8)102240220 ary expression 102240220 to the Octary ir Scale. It is only necessary to observe, 11331 that the operation is carried on through Le the Quinary Scale, or by a series of pen- 23) tads. In the first division, 102 makes 1)5 twenty-seven, which contains the index three times with a remainder of 3 ; then 80 is equivalent to seventeen, which contains it only ¢wice with an excess of 1; and 14 or nine contains it only once, with an excess likewise of 1. In this manner, the division is pursued continually, till the de- composition becomes completed. The result is, there- fore, 1511135, the same as formerly stated. As another example, suppose it were sought to convert the Nonary expression 725708 into a Sena ‘y 6)725708 one. The constant divisor is now 6, but the. 118415 5 operation is performed on the Nonary Scale, the values of the digits advancing always by mines. Tere 6 is contained once in 7, and once again in 12 or eleven, but eight times in 55 or fifty. The rest of the operation pro- ceeds in like manner. The equivalent Senary expression is hence 13121525, t) NUMERATION. } ARITHMETIC, 12h Lastly, to convert a Septenary into an Undenary ex- pression, the process would be thus carried on. ‘The in- dex eleven being denoted on the Undenary 14) 3442433 Scale by 14, this number forms the divi- 222102) 2 : a2 13243] 4: sor; but 14 is contained in 34 twice, —Sal6 with a remainder of 3 on the Septenary Sil 4 Scale ; and the next dividends are 34, ° “210 $2, 14, &c. The remainder of the final division is 10, that is seven, on the scale of operation. Whence the cor- responding Undenary expression is 274642. From the principle before investigated in Palpable Arithmetic, we can easily find the remainder of the di- yision of the original number by another number, which is one less than the index of any particular scale. Thus, to begin with the Ternary Scale, if the amount of all the figures be divided by ‘wo; or, what is the same thing, if every ¢wo be rejected, and the other figures successively added, retaining only, at each step, the excess above ‘wo; at the end of this operation, there will be left one. Hence it might be concluded that 430685 is an odd number; a property indicated also by the character of its last digit.—In the Quaternary Scale, by adding the figures together, and constantly throw- ing out the threes, there remain two; which shows that the division of the original number by ¢Aree would leave two.—In the Quinary Scale, the several figures being col- lected, omitting the fours as they arise, give one, for the remainder of a division of 430685 by, four.—In the Senary Scale, omitting all the fives, there is no remainder ; a proof that the given number is divisible by jive.—-In the Septe- hary Scale, collecting the figures and rejecting the szes, 122 FIGURATE [NUMERATION. there is an excess of five, intimating five as the remainder of the division by siz.—In the Octary Scale, rejecting the sevens, there are left three, being the excess of the division by seven.—In the Nonary Scale, omitting the ezghts, the surplus is five, corresponding to the remainder of the di- vision by eight.—In the Denary Scale, by adding the figures, and rejecting the zines as fast as they arise, there is still an excess of ezght, being the remainder of the divi- sion of 430685 by nine.—In the Undenary Scale, by cast- ing out the ¢ens, there is left five, or the last digit of the original number, and consequently the remainder of its division by ten.—Finally, in the Duodenary Scale, by se- parating the elevens from the collected figures, there re- main ¢wo, the last figure in the preceding scale, or the re- sidue of the division by eleven. In general, the terminating figure of the expression on any scale, is the same as the remainder of the division by its index on the next higher scale. The casting out of the divisor may be shortened likewise, if it be contained in any power of the index, by taking only the corresponding terms. Thus, not to go farther than the Denary Scale, #eo may be cast out merely from the division of the final digit, or 5. Four may be separated by dividing the two last figures, or 85; and the remainder of the division by eight is detected, from the examination of the three terminating digits 685. But the quotients of such divisions are also directly dis- covered, from what was explained under Palpable Arith- metic. Not to multiply examples, let us begin with the Senary Scale. If the several figures 13121525 be repeated on all the lower places, their summation will appear as here annexed, The excesses corresponding to the differ- NUMERATION, ] ARITHMETIC. ent rows amount to 20, which, divided by 5, leaves no remainder, but gives 4 to the next column. The sum of this again, with the 4 carried to it, is 19, which contains 6 three times, with an excess of 1. This 3 is conveyed tothe next column, and the same operation is repeated. The 123 1111111)1 333333)3 11111) 2299'2 111]1 555 ol iret 15024400 result of the whole, or 1502441, being now transferred from the Senary to the Denary Scale, gives 86137, for the quotient of 430685 by five. The expression of the Octary Scale, treated in the same way, will stand thus. Here the excesses taken together make up 17, giving, after the division by 7, asurplus of 3, and 2 to be car- ried to the first column. The figures on all these columns, being collected and reckoned by ezghts, amount to 170126, which, converted again to the Denary 11111 {1 55555)}5 1111 1111 11)1 3/3 5 170126)3 Scale, is 61526, or the quotient of the original number by seven, with a remainder of 3. On the Denary Scale itself, the same de- composition is here pursued. The several excluded figures now amount to 26, making 2 nines, and a surplus of 8. The columns themselves being summed up, give 47853, for the quotient by zzne, with a remainder 8. Lastly, suppose the expression of the Duodenary Scale were thus analysed: The excesses amount to 35, or 3 elevens and 2. The columns added in succession give 10709, which, converted into the ordinary scale, makes 39153, with a surplus of 2, for the quotient of 430685 by eleven. 4444/4. 3333/3 66/6 8/8 5 4785318. 11111)1 8888/8 999)9 22/2 0 5 1090912 124 FIGURATE [ NUMERATION. But the extension which was made of the same prin- ciple to open counters, may be applied, to discover the quotient of any number divided by the index of the next higher scale. Without dwelling on this subject, we shall take the three last examples to illustrate the mode of operation. Beginning, therefore, at the left hand, each figure must be repeated through all the 4\.\1\11 succeeding places, alternately as an excess and = 59555) a defect, and the balance of their addition, or Aa 143436, taken as the result. ‘This number is M\{1 otherwise expressed 135356, which corresponds Bh; to 47854, the quotient of the original number, ya5qa6/7 with the accession of the excluded 1, by mzve. Again, the analysis will proceed thus in the 4)44);4/, ordinary scale. The extended digits give a Pa i surplus of 2, and the figures on the several co- 8/8 lumns amount to 41153, or 39153, being the mae 4153/2 quotient of 430685 by eleven, with an excess of 2. Lastly, in the Duodenary Scale, the de- composition will be more rapid. ‘The surplus figures here give a defect of 5, and the rest in the several columns make up 17200, which is equivalent to 33130, ‘the quotient of the original number, with the addition of 5 by eleven. It is of more consequence, however, in such divisions, to discover the remainder than the quotient. If the pro- cess be confined to the Denary Scale, the number eleven will appear to have properties analogous to those of nine. But to find the remainder of the division of any number by eleven, or, in the vulgar phrase, to cast out the elevens, NUMERATION.] ARITHMETIC. 125 will require attention to the alternate character of the ci- phers, fluctuating in succession from excess to defect, The easiest mode is, Beginning at the right hand, to mark the alternate figures ; and, from the amount of these, aug- mented by eleven, if necessary, take that of the rest, and the difference is the remainder sought. ‘Thus, resuming the original number 43/06’85’, the sum of the accented figures is 14, and that of the rest only 12; wherefore, if divided by eleven, it would leave an excess of 2. Again, taking the number 3/17/0625’, the marked figures amount to 21, and the others only make 3, leaving for the division by eleven a deficiency of 18 or 7, that is, an excess of 4. It hence follows, that, as a number is divisible by ne, when the amount of its figures is any multiple of xine ; so a number is divisible by eleven, when the sums of the alternate figures are either equal, or differ by eleven or its multiples. This proposition leads to some curious results, but I shall notice only the more striking and simple. It is an obvious consequence, that the difference between any -number and its reverse is always divisible by xzne : ‘Thus, the number 4306385 being reversed into 586034, gives the difference 155349, which may be divided without any re- mainder, by 9. ‘The reason is plain, since this number and its reverse are expressed by identical figures, they are both multiples of 9 with the same excess, and consequently their difference must only be some multiple of 9.—Again, the difference between a number and its reverse is like- wise divisible by eleven, if it has odd places of figures. Thus, the difference between 3170625 and its reverse, or 2090088, is divisible by eleven ; for the sums of the alter- nate figures, 19 and 8, differ by 11. But the swm of a num- ber and its reverse is divisible by e/even, when it consists of 126 FIGURATE { NUMERATION. even figures. ‘Thus, the original number 430685, having 586034 for its reverse, their sum is 1/01’67/19’; which is evidently diyisible by eleven, since the accented figures amount to 18, while the rest make only 7, or 11 less. It is not difficult to perceive the reason of these proper- ties of eleven. When the number consists of odd figures, — they preserve the same character of abundant or defective in its reverse, and consequently the subtraction of the op- posite numbers will destroy whatever inequality there had before existed; but when the number proposed consists of even figures, the abundant and defective, by reverting their order, mutually change places, and hence the addition of the number and its reverse will extinguish any original inequality between these, counterbalancing any surplus of the one set by an equal deficiency in the other. The number seven is likewise distinguished by its pro- perties on the Denary Scale, though they are not quite so remarkable as the relations of nine and eleven. All the remainders of any division by 7 must evidently be in- cluded in 1, 2, 3, 4, 5, or 6. But if wits having ciphers annexed in succession be divided by 7, the quotient is found to be 142857, with the corresponding remainders — 1, 3, 2, 6, 4, 5, which comprehend all the possible va- rieties. Wherefore, since the last of these remainders is just as it was at first, the series of divisions will again be renewed ; and consequently, in the expression of the quo- tient, th same digits must perpetually recur, and in the same order. Nay, if the first period of that quotient, or the number 142857, be multiplied by 2, 3, 4, 5, or 6, the products, 285714, 428571, 571428, 714285, or 857142, are still denoted by the same digits, and in the same NuMERATION.] ARITHMETIC. 127 order, only commencing from different points. ‘The rea- son of this very curious property is, that the remain- ders 1, 3, 2, 6, 4, 5, precede the digits, 4, 2, 8, 5, and 7, of the quotient; or, in other words, those remainders with ciphers annexed, when divided by 7, give quo- tients commencing with such digits, and afterwards run- ning through all the series of changes. Now, it is evi- dent, that these new quotients must be the same as the first one multiplied by the several remainders. Another remarkable property of the quotients by 7, which may be of some utility in Practical Arithmetic, is derived from similar principles. Since the remainders of the division of 1, 10, 100, 1000, &c. were respec- tively 1, 3, 2, 6, 4, and 5, it follows that the remainder of the division of any of these digits, with annexed ciphers by 7, will be found by pursuing the same concatenated order, 1, 3, 2, 6, 4, 53 1, 3, 2, 6, 4, 5, &c. and reckon- ing downwards from the given digit to the last place, or that of units. Thus, the remainder of the division of 400 by 7, is 1, because 1 stands in the series ‘wo places lower than 4, which answered to hundreds; for the same reason, 5 is the remainder of the division of 2000, since it occurs three places lower than 2. Ifa larger digit than 6, such as 8 or 9, with its train of ciphers, be divided by 7, the remain- der will evidently be the same as what corresponds to the excess of 1 or 2. It will be hence easy to discover the re- mainder of the division of any compound number by 7. Suppose, for example, the former number 430685 were proposed: Beginning at the right hand, and conceiving it to be composed of 5, 80, 600, &c. the corresponding remainders are 5, 3, 5, 5 and 6, making up 24, or a ge- neral excess of 3. 128 FIGURATE " [NUMERATION. Ii Palpable Arithmetic, the primary bar was always marked, as the key to ascertain the import of the rest. The same thing is done in the digital notation, by setting _ Below | a full point immediately after the place of units. ‘that point, the digits must diminish exactly as they in- | crease in standing above it. any numerical scale are hence fitted to denote the re- motest subdivision of parts with equal facility, as the as- cending ones are capable of expressing the largest possible number. This property, though ceived, constitutes one of the greatest advantages of such: scales. Hence to transfer the lower denominations from one scale to another, it is only required to multiply continually | the given expression by the index of the new scale, and conceive the product each time to be set a place lower. ut slow in being.per-. Thus, to change the fractional digits 15243134 .15243134, denoting the diameter of a reid. in ; : : ; : 1.13501024 eircle which has unit for its circumfer- 4, ence, from the Senary to the Quaternary 1.032041 44 Scale. 'This repeated multiplication is per- pee abide formed after the Senary notation; the Mad ‘*i ) products of the single digits being con- 1.2535 0424 stantly divided by szx, while the remain- tite Bra’ der is set down, and the quotient car- vauiaeng ried to the next place. The correspond- 3.5025 1504 ing Quaternary notation is therefore, in 4 round numbers, .11011331, the same as eet) what arises from the decimal expression -3810938062. 1.30445344 The descending scales now selected for explaining the method of transformation are seldom ever used. But the — The descending terms of | NUMERATION. ] ARITHMETIC. 199 descending terms of the Denary Scale, or those of the Decimal Subdivision, have at length been very generally adopted into practice, at least in all the calculations con- The duo- decimal system of partition, from its convenience in Men- nected with mathematical and physical science. suration, is employed to a certain extent among traders. It may be proper, therefore, to exemplify the re- — .785398 ciprocal transformation of decimals and duode- snaG cimals. In this view, let it be required to ~ = 19 change the expression, .785398, which denotes 5.097312 the ratio of the circle to its circumscribing Hk ‘ , ; ~ 1.167744 square, into duodecimals. The operation is 12 performed by multiplying the given fraction 2.012928 by 12, and pointing off the same number of di- ____!2 , F 0.155136 gits; the process being constantly repeated 12 with the several partial excesses. Hence the [.861632 equivalent expression in the duodecimal nota- 12 0.339584 tion is .9512010; of which the four first digits, however, may be judged sufficient in practice. Again, the cube root of 2 expressed in duodecimals is 1.315188. To reduce this té decimals, it must 1.315188 undergo a repeated multiplication by ¢en or 0. ae It is scarcely necessary to observe, that the nota- ~~ O tion being now duodenary, the whole process is 5.000428 to be carried through that scale, each digital Hise sty product which arises in succession being con- “situa tinually reckoned by fwelves: Thus, ten times 9.263228 8, or 80, is siz dozen and 8 over; and the _ 9% next product 80 with the 6 carried, is seven seta dozen and2 over. In like manner is the opera- 1.026628 tion performed with all the other figures. The result of the conversion is hence, in decimals, 1.259921. K 130 FIGURATE [ NUMERATION. Both these last descending scales, however simple and commodious in practice, are comparatively of recent adoption. The natives of India, who have so long been acquainted with the Denary Notation, are still ignorant of its application to fractions. Below the place of units, they change the rate of progression, and descend merely by a continued bisection, assuming successively the half, the fourth, the eighth, and the sixteenth ; and beyond this last partition they seldom advance. Nor did the Moors, during the short period in which they cultivated science after receiving the digital system from the East, ever at tain to the knowledge of Decimal Arithmetic. All pro- gress was fatally stopped by their cruel expulsion from Spain; but, that the principle of the Denary System should, in its native bed, have lain absolutely unproduc- tive through the course of many centuries, is a circum- stance which strongly marks the want of invention among the Hindus. It is curious to remark, that the use of decimal frac- tions in calculation was long preceded by the complex train of Sexagesimals. ‘This rapid progression had been introduced into the Alexandrian school by the famous Ptolemy, who had the merit of digesting the results of astronomical observations into a body of regular science. It descended by the powers of s¢xty; but though quite arti- ficial and seemingly arbitrary in its structure, it was easily engrafted on the Greek system of numeration. The astro- nomers of Alexandria and of Constantinople continued to employ the Seragesimal Notation, in which they were after- wards imitated by their successors among the Arabians and Persians. The system itself had no doubt its rise in the sub- divisions suggested by the celestial phenomena. ‘The par- NUMERATION. ] ARITHMETIC. 131 tition of the circumference of the circle into 360 equal de- grees was originally founded on the supposed length of the year, which, expressed in round numbers, consists of twelve months, each composed of thirty days. The ra- dius, approaching to the sixth part of the circumference, would contain nearly 60 of those degrees; and after its ratio to the circumference was more accurately determined, the radius still continued to be distinguished into the same number of divisions, which likewise bore the same name. As calculation became more refined, each of these 60 divisions of the radius was, following the uniform progres- sion, again subdivided into 60 equal portions, called mi- nutes or primes; and, by repeating the process of sexage- simal subdivision, seconds and thirds were successively formed. ‘The degrees were considered as integers, and the minutes, or primes, the seconds, and thirds, &c. distin- guished by intervening blanks, and sometimes the addition of corresponding dashes. As an illustration of the mode of converting decimals into sexagesimals, let it be required to express the side of an inscribed decagon, or the chord of 369, in seragesimal parts of the radius. Here the decimal .61803398428 . 60 618 8428 mS RNAS e bre. quantity, .6180339 ‘ , denating the Pa COODECOREE greater segment of this line considered as 60 ‘unit, and divided into extreme and mean #.9223434:08 ° ° ° "es R é 60 ‘ Katio, is multiplied hy 60, or; ate is SF a0R0LES equivalent, by 6, with one decimal abrid- 60 ged each time, the same process being 20.4362688 ; , 60 repeated po all the successive fractions. GATES The result is consequently 37° 4’ 55” 20” 60 26 10" 34%; which exactly corresponds, 10.56768 as far as the seconds, with what Ptolemy pe ae . 182 FIGURATE [NUMERATION. has assigned. | Hestops there, and the accuracy now pur- sued for the sake of exemplification, is indeed superflu- ous, approaching within a frillionth part of the truth. Another example will show, by the converse procedure, that the Greek astronomer knew more accurately, than has generally been supposed, the length of the circum- ference of the circle. He calculates the chord of one de- gree, which cannot differ sensibly from the arc itself, to be 1° 2/50”. Consequently, if the radius were considered as unit, the arc of 60 degrees would be represented sexagesi- mally by 1.2’ 50”; and therefore the triple of this, or 3.8’ 30”, must express the-circumference of a 4 gray circle whose diameter is 1. To reduce that ——_10 quantity to decimals, the fractional parts are ait 00 multiplied repeatedly by 10, and the successive rid products divided by 60. Hence the decimal 10 expression for the circumference of a circle is 1.40 10 3.1416, a very useful and celebrated approxi- ers mation. It was the practice of sexagesimals, at a late period, that led by gradual steps to the formation of Decimal Arith- metic. The great restorer of mathematical science in Europe, George Purbach, or Beurbach, of Vienna, a man of original and extensive genius, who died at an early age in 1462, in the table of sines which he ap- pears to have computed to every minute of the qua- drant, instead of distinguishing the radius into 216,000 seconds, or dividing it three times in succession by 60, made it to consist first of 600 instead of 60 equal por- tions, and then parted each of these into 100 primes, and each prime again into 100 seconds, thus blending in effect the sexagesimal with the decimal or centesimal nota- > * ba NUMERATION]. ARITHMETIC. 133 tion. His disciple and successor, John Muller, commonly styled Regiomontanus, from Koningsberg the place of his birth, extended those tables to seven places of figures, ma- king the radius to consist of 6,000,000 parts. After some hesitation, he finally abandoned that radical division, and having in 1464 enlarged the radius to ¢en million of parts, he recalculated the sines, to which he likewise joined, for the first time, a table of tangents. But this laborious and important work lay, many years after the author’s ~ death, in manuscript, and did not appear before the pub- lic until 1541, when it was printed under the direction of Schoner at Nuremberg. It is obvious that those mixed sexagesimal expressions would be reduced to common decimals, by multiplying by 10 and dividing the product by 6. Thus, 1552914 in Miuller’s first table, the sine of a Kar- | 10 _ daga, or the arc of 15°, so called, it would 6)15529140 i 2588i90 seem, from the Arabic verb Karatha, to divide, is 1552914, Wherefore the result of this reduc- tion, with the decimal point prefixed, is .2588190, which perfectly agrees with our modern tables. But the final step towards the use of decimals, which consisted in estimating the radius as unzt successively de- composed, was slowly attained, and a long period still elap- sed before mathematicians were trained to the new practice. It is curious to observe here the very gradual progress of improvement. Ptolemy had distinguished the sexage- simal subdivision into primes, seconds, thirds, &c. by corresponding accents placed over the successive parts. Michael Stifelius of Eslingen, the scholar and follower of Luther, having remarked the relations of arithmetical and 194: FIGURATE [NUMERATION. geometrical progressions, in his Arithmetica Integra, print- ed at Nuremberg in 1545, a work of great merit, noted the exponents of powers, both ascending and descending, by the digits 1, 2, 3,4, &c. Guided by analogy, he likewise appropriated the zero to indicate unit, or the commencement of every series. He therefore expressed integers and their_ sexagesimals by these characters, 0, 1, 2, 3, 4, &c. placed over them; and in representing astronomical quantities or physical subdivisions, he com- bined the exponents with certain contractions or modified accents; thus, 1, 2, 3, rm &c. Bombelli, in his Algebra, printed at Bologna in 1572, adopted this improvement. Simon Stevinus of Bruges, mathematician to the first Prince of Orange, and a man of great originality of conception, in his Arithmetic, composed in 1583, in the Flemish dia- lect, and published two years afterwards in French, em- ployed the marks @, @, ©, ©, @, @, @®, @; or the digits circumscribed by a circle, to signify the extended se- ries of powers. ‘This notation he applied principally to the Denary Scale, setting out from the place of units, and marking the tens, hundreds, thousands, &c. upwards, and again downwards, the tenths, the hundredths, the thousandths, &c. by continual decimation. The units themselves he indicated by ©, implying only the com- mencement of numeration. ‘Thus, according to him, @Q©O@OO@ & 3 1 4 1 6 would denote 3 units, 1 tenth, 4 hun- dredths, 1 thousandth, and 6 ten thousandths. He first explained the use of such decimals, and strongly urged their preference to vulgar fractions. Still, however, this notation, though sufficiently clear, might seem rather overloaded. NUMERATION. } ARITHMETIC. 135 It was not distinctly or immediately perceived, that the mark ©) might, by its position alone, ascertain the rest, The ultimate simplification, therefore, consisted in omit- ting altogether those various marks, and placing a full point or a comma, to represent the key, after the units and before the range of decimals. This capital step, which indeed leaves nothing more to be done in practice, we owe most probably to the great Napier, who, in his Rabdologia, printed at Edinburgh in 1617, while he quotes Stevinus with applause, incidentally proposes the final improvement. He seems not to trust exclusive- ly, however, to punctuation, and while he marks the ter- mination of the units by a comma, he also cautiously notes the successive decimals by superscribing repeated accents. But the noble invention of logarithms, deriving its birth in like manner from the efforts made to abridge the opera- tions of spherical trigonometry, gave a decided prepon- derance to decimals, which those artificial numbers, as next remodelled in the hands of Briggs, adopted into their actual composition. To reduce vulgar fractions to any scale, we have only to multiply the numerator by the root of that scale, and divide by the denominator; and to repeat this process, if requisite, on the ‘successive remainders, till the quotients either terminate absolutely, or glide into a circulation. Sup= pose it were sought to represent on the Senary, Octary, and Denary Scales, the fraction 34% or 3,5, which Peter Metius, a distinguished Dutch mathematician, and near relation of Adrian Metius of Alkmaer, about the close of the sixteenth century, assigned for the approxi- mative ratio of the circumference to the diameter of a nexed to it, by the denominator. 7 ' 4 ‘ f 136 FIGURATE [NUMERATION. citeles «i Themusiag vhs OCTARY. DENARY. merator 16 must, 16 16 16 . therefore, be mul- 6 8 10, tiplied by the in- MS)25i0 113)128(1 113)160(1 d Re ead 113 113 ‘ ex of each scale, 576 5 Pte 7 ¢ and the product 565 98: 10 Hi divided by the ik 1Z0(1 470(4 denominator 113; Med Bee aes enone 5 66(0 7 oe the remainder to 6 8 10 be treated in the 396(3 56(0 180(1 same way in re- sod a) 113 y 57 745(3 67 peated succes- 6 339 io = sion. ‘The ex- 342(3 709 670(5 x pression of 354 ies 8 565° @ 3 872(7 105 on the rina ‘Y 6 9} & taoa Scale is, there- 18(0 $1 1050(9 fore, 3.0503301 ; 1 ’ ata 1017 on the Octary 11 Hs Madly a 0 Scale, 3.110376; my eR oN 5303 and on the De- 339 = nary Scale, 3.141593, This last conversion would evi- _:113(16.000000).141593 dently take a more compact,form, pao ne: 4:70 h as here exhibited. The several 452 3 steps of the process are virtually 180 : the same as before. Hence the 113 foundation of the ordinary pro- iy , cess for changing a vulgar into 1050 : é a decimal fraction, by dividing 1017 ; the numerator, with ciphers an- pa : 4 | os ot -NuMERATION.] ARITHMETIC. 187 Mixed fractions are reduced to any scale 15 3% nearly in the same way. Suppose it were a sought to express decimally fi/teen shillings and 10. threepence three farthings. Multiply this sum 6.11 3 by 10, and the fractional part of the product — again repeatedly; the integral results being 10 deferred to the lower places, must evidentlyex- 65 — press the same value. The decimal corre- x sponding to the mixed fraction is hence “jg 765625. 5.0 The same result might have been obtained otherwise, by ascending progressively from the lowest term. Thus, three farthings, expressed in decimal parts of a penny, are -75; and dividing 3.75 by 12, the quotient .3125 denotes the value, with threepence annexed, in decimal parts of a shilling; and finally, having prefixed the fifteen shillings, and divided the compound 15.3125 by 20, the quotient, expressing the fraction of a pound, is the same as before. But examples of this kind are better adapted 15 33 for the Duodenary Scale. The multiplication aE here being performed by successive ¢welves, the 12 Duodecimal expression for the same sum isthere- 2. 5 0 12 ou As another example of the conversion of mixed frac- fore .923, consisting only of three figures. tions, let it be required to find the decimal of a Ton cor- responding to thirteen hun- 13 cwt.2q. 71b 28)7.00(.25 dred weight, two quarters, a a and seven pounds. ‘The ay ay 1 0 operation is performed in 7-16 1 0 eas two ways, which give the 73 4,)2.2500 same result, .678125. The 10 5625 ascending process, which hae O has 28, 4 and 20 for suc- os 20)13.562500 cessive divisors, is perhaps 10 ‘678125 the easier. 5. 0 138 FIGURATE [ADDITION. Having explained so fully the principles of Numera- tion, we now proceed to treat of the common operations in Figurate Arithmetic. ADDITION. From the principle of numerical notation, it follows that Addition is performed by collecting the digits of each bar or rank. Each class, whether it be units, hundreds or thousands, is treated in the same way. In adding two figures, it is only requisite to count forwards from one of them, as many steps as are signified by the other. Sup- pose 5 were to be joined to 8; reckoning onwards, we pass through 9, 10, 11, 12, to 13. ‘This simple process may be more conveniently performed by counting over the fingers. But, for a learn- Denary Scale. Saal 2; 3, 4 er, it is a preferable mode to EEK “| 3 -ano + frame a Table of Addition, 4) 56 7 8 oliolll| 2 — | | | | | which he may easily commit 6} 7} 8| 9111/12) 3 tomemory. ‘The construction iw] “| 9)10 11|12 13 4 and use of such a table are so LOM 21314) 5 very simple, as hardly to re- = ate quire any explanation. The ieli7 S one number occupies the hori- ~ {18} 9 zontal row at the top, and the other, which is not great~ er, the vertical row at the side. ‘Thus, below the column of 7, and opposite to the horizontal range of 6, stands 13, the sum of these numbers. Such tables are found in the more ancient treatises of arithmetic; but they have been most injudiciously omitted in the latter systems of education. Let it be sought to add these four numbers, 3709, 8540, 2618, and 706. Having set them in their ranks, the most ADDITION. ] ARITHMETIC. 139 natural way would be to write down the 3709 3709 8540 8540 sum of each porticein: The first column 9618 2618 on the right hand gives 23, the next 5, the 706 706 third 25, and the last column 13. Or the 13 23 ; : ; awe 5 operation might be carried on from the left 5 95 hand to the right, the several sums being 23 13 now 13, 25, 5, and 23. ‘These numbers, 15573 15573 collected by a second summation, give for the final result 15573. The former mode of working is generally preferred in Europe, but the latter is the method still practised by the Persians and the Hindus, who $17 )01]9 likewise, for the sake of greater : : , . distinctness, are accustomed to 7101/6 separate the different ranks of aa 3 ans 3 figures by perpendicular lines, as 2 POI cath 416 {51713 in the form here annexed. Of 13, the sum of the digits in the first left hand column, 3 is placed at the bottom, and 1 advanced to a higher cell in the line below; in the next column, 5 is set down, and 2 placed a step forward in the lower line; and the same pro- cess is carried through all the rest. But it is the fashion of the East, to preserve only the result of the summation, the other auxiliary rows of figures being always rubbed out. For this reason, they term the lower horizontal line, which in this example is 1202, the Khuti Mahi, or obliterating line. ‘To such a practice, the usual mode of operating in those countries is indeed well suited. The Hindus at present, as Dr John Taylor of Bombay acquaints us, perform their numerical computations on a board about a foot long and eight inches broad. A white ground being formed by a sort of pipe-clay, the board is then covered with fine sand, or with gulal, that is, flour dyed of a purple colour. ‘The figures or letters are traced by a wooden style, which, displacing the sand or 140 FIGURATE T ADDITION. purpled flour, leaves the white ground exposed. By pass- ing the finger gently over the surface of the powder, those forms are easily effaced, and the board is again fitted for receiving new impressions. It is customary for them to rub out each successive step, even of a short process, so as to leave on the board only the general result of the ope- ration. ‘The process of Addition from right to left, which is on the whole better adapted to our habits, would be rather shortened, by writing under each column the units of the sum, and below it the tens in a smaller cha- racter, which are to be joined to the figures Q2avo : ‘ 3709 i the next column in adding them. By a 8540 little practice, however, this precaution is ren- 2618 dered unnecessary, and the small subscribed ir: a“ ; figures are retained mentally, and carriedto 120 2 the successive higher columns. Above the several units, tens, hundreds, &c. are likewise placed here the marks contrived by Stevinus, to ascertain the respective ranks of the digits.. This operation is somewhat easier, if Q@O@O® 451% performed with deficient figures. Thus, Vos eH the numbers may be changed into others 3 M4 2 2 with the defects interspersed. In this i : . _ 2 mode, there being a sort of counter- orl 5 § 7 3 balance, it will seldom be required to carry any thing to the higher columns. Suppose those numberswere all transferred 2191 2191 to the -Duodenary Scale, they willstand thus: 4928 4038 : 1622 1622 The summation being carried on both from 400 400 right to left, and from left to right. In the pF OGG latter mode, the first column gives ¢wenty- se Ae one, that is, 1 dozen and 9. The rest 19.17 of the operation proceeds in the same way, 9019 9019 ADDITION. ] ARITHMETIC. ~ i4} If the condensed process be followed; 2191 221 hs A046 BV Ay the form will be a little different, and more 1622 20.9.9 distinct. ‘The summation is also short- 400 5\0. ened somewhat, by introducing deficient 9919 902 , ra) 1¢ figures; nor is there, in that case, any E049 occasion for carrying, since the accumulated excess of the digits is partly counterbalanced by the intermixed defects. b:. Duode nary Scale. To facilitate the work- 1} 2 314 5 a] 7 3 » 9 ing on the Duodenary 2 3 4 5 6 a8 Dice 7 Scale, it would be expe- ‘ ee = 5 ao ordi 9 dient to construct an —Harska OTB IST addition. Table. . By |. -oraiain | Of 01011 1215.14) 5 101112131415) 6 121374 15,16) 7 of pence, and carry the 1415|1617 8 excess to the place of 16 I7118) $ 9 shillings. 18:19 means of this, we may at once sum up a row It may serve as a check on the accuracy of Addition in the Denary Scale, to cast out the nines. Thus, in the exainple here adopted, the first number 3709, adding the digits 3 and 7, and passing over the 0 and 9, gives an ex- cess of 1; the next number 8540, treated in that way, leaves 8; and the second and third numbers give respec- tively 8 and 4. But these excesses again, or 1, 8, 8, 4, on casting out nie, leave a final surplus of 3; which is the same as what arises from the separation of nzne in the sum 15573, a presumption, though not an absolute proof in- deed, of the correctness of the operation. In the Duodenary Scale, the corresponding trial is made by casting out eleven. The several excesses are 142 FIGURATE [ ADDITION. 2,4, 0, and 2, making 8; the very same as results from the analysis of the sum 9019. The Hindus are still totally unacquainted with these curious properties. ‘They prove addition, by cutting off the uppermost line, and afterwards joining it to the amount of the rest, as frequently practised in Kurope. But a better mode of checking any error, and correct- ing wrong associations of numbers, is to begin at the top, and to sum the rows of digits downwards. It would greatly facilitate commercial transactions, if all subdivisions were carried downwards on the same scale, the decimal system being the best adapted to the prevail- ing mode of numeration. But as this obvious improve- ment is not likely to be soon embraced, an example or two shall be likewise given of the addition of mixed fractions. First in Apothecaries’ Weights, the pound consisting of 12 ounces, the ounce of 8 drams, the dram of 3 scruples, and the scruple of 20 grains. Here 4%, z 3 Der. the first column, amounting to 38 93°. BiG. alan : OB 11 Sie Care grains, leaves 18, and sends 1 to the 34 0 7 O15 scruples, which now make 5, or 2 16 8 0 1 O scruples and 1 dram. In the same 195 2 1 218 way, the rest of the process goes forward. The next example may be drawn from measures of length, where 12 inches make a foot, 3 feet a yard, 52 yards a pole, 40 poles a furlong, and 8 furlongs a mile, The only difficulty here consists in M. F. P. Y.F. In. adding the yards, which, with 2 car- 31°28 tle og ried, amount to 10; but 10 con- id r oy : : - tains 5} once and leaves 42, and 2601102 3 consequently 4 is set down and 1 117011483 6 foot 6 inches joined to the preceding columns. SUBTRACTION. } ARITHMETIC. 143 SUBTRACTION. ‘Turs operation, having for its object to find the differ- ence between two numbers, is precisely the reverse of ad- dition. The same auxiliary table may hence answer for both. Thus, if 7 joined to 6 makes 13, it is equally clear, that 7 taken away from 13 must leave 6. For the sake of distinction, the greater of the two numbers is called the Minuend, and the other the Subtrahend. The method of proceeding will be most clearly percei- ved from the inspection of an example. Let it be re- quired to take 428053 from 702632. ‘The process may be conducted, as in addition, either from right to left, or from left to right. In the former way, 3 cannot be taken from 2, but the effect will evidently be the same if ten were added to both the minuend and the sub- trahend. Ten may therefore be joined to the 2, while one, as equivalent to it, is thrown to 5, which occupies the place higher. This addition of the 10 is called borrowing, and the countervailing addition of 1 in the next bar is called carrying. Take 3 then from 2, with the junction of 10 borrowed, or 12, and there remains 9, which is set down and the 1 subscribed to the next higher bar. Again, 5 isto be taken from 3, which can be done only by annexing 10, for which 1 is sub- pelbis tracted on the advanced bar. Consequently 8 384689 is written down ; but no carrying is required in 11011 : ; ‘ 274579 the next bar which leaves 6. It is now required 144 FIGURATE [ SUBTRACTION. to subtract 8 from 2 or 12, and therefore to set down 4 and advance 1 for the 10 borrowed. Next 2 is taken from 0 or 10, leaving 8, and | to be subscribed under the highest bar. Lastly 4 is taken from 7, and 3 written down.—-The result would evidently be the same, if the operation had proceeded by an inverted order from left to right, as commonly practised by the Hindus. The subscribed figures 1, 1, 0, 1, 1, which had been joined to the minuend to promote the subtraction, must now be taken from the digits 3, 8, 4, 6, 8, 9, which marked the excess, in order to give the true remainder 274579. But instead of writing down those carried fi- ~o9g99 sures, they may with more convenience be ap- 428053 plied mentally. In this éase they are at each 2/4579 step either taken away from the minuend or joined to the subtrahend before the subtraction is made. Let the same example be worked with deficient digits. In substracting the lower number, it is only —_4793),39 required to change the character of its digits, 459.158 and then add them. The operation, there- marc 3 fore, needs no farther explanation. Next, suppose those numbers were converted into the Duodenary Scale. The subtraction will be per- 99974 formed thus: It is only to be observed, that 187871 when there is occasion for borrowing, twelve 112097 is joined to the digit of the minuend, and one is carried or annexed to the higher digit of the subtrahend. Subtraction is proved by adding the remainder to the sub- trahend, which should give the minuend. The mode of casting out mzmes on the Denary Scale, or elevens on the Duodenary, might likewise be employed. Thus, with the — former, the excesses are 2 and 4, leaving 7 for the excess ~ ie *. MuLTIPLicaTION.] ARITHMETIC. 145 of the remainder; and in the latter, the excess of the minuend is 7, the subtrahend being divisible by 11. Since the terms of a descending scale are treated in the same way, it would, perhaps, be superfluous to take ex- amples of the subtraction of decimal fractions. Of mixed quantities, a single exam- 4] 4° 7 36" 94)” ple in sexagesimals may be sufficient; 89 10 38 33 the remainder of this subtraction being 24 56 57 51 here the difference between the chords of 144° and 96°, and therefore equal to the chord of 24°. MULTIPLICATION. Tuts operation, it was observed, is nothing but a re-« peated addition. ‘The object which it seeks, is to collect, for a Product, the Multiplicand, as often as there are units contained in the Multiplier. Such a process, however, would have proved intolerably tedious, if the principle of numeri- cal arrangement had not come to lend its aid. Suppose it were required to double the number 748, this would be performed by adding it twice; andinthe 748 same way it might be ¢ripled, as here exhi- Ge Oe bited. But if the multiplier be a com- 9944 7496 posite number, the operation is shortened, by an inter- mediate procedure. ‘Thus, if 748 were to be Sas , : 14.96 multiplied by 6, it may be done, either by j49g go44, adding the double of it three times, or the ri- 1496 2244 ple of it twice: Again, the addition of this BASS \: 4ADG result five times, would give the product of the original L 146 FIGURATE [ MULTIPLICATION, number by 30. This product, it is obvious, could likewise be obtained by adding the ¢riple of 748 ten times, and consequently by subjoining a zero. In general, since any digit in the Denary Scale is augmented ten fold at each move to a higher place, its product into the multiplicand must give a similar increase. Let 748 be multiplied by the complex number 632. The multiplier now consists of 2 units, 3 tens, and 6 hun- dreds ; wherefore the compound product will be disco- vered, by resuming the separate products of those digits, and disposing them in their order. Here one zero 1486 is subjoined to the product of tens, two zeros to 99440 that of hundreds, and so forth. These zeros, how- 448800 ever, may be omitted, since the value of each figure 472736 in the lower rows is determined by the position of those in the first or uppermost one. The three rows being col- lected together, give 472736 for the result. But the operation is abridged, by performing mentally this repetition or summation of each digit in the multipli- cand. ‘Thus, resuming the last example: 748 748 8 repeated twice makes 16, which is set 632 652 down; 4, repeated as often on the next a ak. advanced bar, gives 8; and 7 repeated 14 48 likewise ¢wice, and shifted a place higher, 3 +m produces 14. In this way, the other rows 1 24: of separate products are successively _*% 14 ‘ 24 8 formed. The operation may be pursued 49 16 either from right to left, or from left to 472736 472736 right. Such was nearly the procedure of the ancient Greeks. Every step in the multiplication of complex numbers, re- presented by alphabetic symbols, appeared as detached — MULTIPLICATION. ] ARITHMETIC. 147 and separate members of the product. The operations of arithmetic, among that ingenious people, advanced like writing, from left to right, each part of the multiplier being combined in suceession with the several parts of the multiplicand. ‘The products were distinctly noted, or, for sake of compactness, grouped and conveniently dis- persed, to be collected afterwards into one general a- mount. Pappus of Alexandria, in his valuable Mathe- matical Collections, has preserved a set of rules which Apollonius had framed, for facilitating arithmetical opera- tions. ‘These are, in the cautious spirit of the ancient Geo- metry, branched out into no fewer than twenty-seven pro- positions, though all comprised in the principle stated by Archimedes, That the product of two integers of different ranks, will occupy a rank corresponding to the sum of the com- ponent orders. Suppose, in the last example, that 30 came to be multiplied into 700 ; ‘Take the lower corresponding cha- racters 3 and 7, which were called radicals, the one de- pressed /en times, and the other an hundred times; and multiply their product 21 successively by the zen, and by the hundred, or at once by a thousand, and the result is 21000. These methods of multiplying numbers would become in many cases excessively tedious and perplexed. ‘The modern practice of consolidating the figures at once on each bar before they are written down, by carrying, as in the process of addition, is much simpler, and 748 decidedly preferable. Instead of noting the 1 632 of the first product 16, it is joined immediately gue to the next product 8, and the sum 9 is written 4488 down. In the second row again, thrice 8 makes 472736 24, or leaves 4, and sends 2 to the product 12. The other figures are obtained in the same way. 148 FIGURATE — [MULTIPLICATION. . Stevinus followed in some measure the principles of Archimedes and the rules 448 laid down by Pappus,and markedthe pro- 6 32 gression ‘of zens by his series of encircled i ‘ : 4 6 exponents. Inthe highest product, for in- 4 49 8 stance, the 7 multiplied by the 6,and both 4727 3 6 advanced to a second place from that of BOOOOOO units, and therefore by ¢wo steps, and again by wo more, would give 42 for the fourth place, and consequently pro- mote the 4 itself to the fi/th place. This now is the ordinary form of Compound Multipl- cation, and it seems scarcely to admit of any material im- provement. But, to shorten the repeated summation of digits, it is expedient to° MULTIPLICATION TABLE. Denary Scale. Te atari construct a table, which must be engraved in the memory of the arithme- tician. It was anciently styled the Pinax, or Men- sa Pythagorica, from the name of the Philosopher who first taught the use of it to the Greeks. By those ingenious people, it was likewise called the Logistic, or Calculating Abacus. It is readily formed by repeated additions, but, though now so very common, I have annexed it here. The mecha- nical method of multiplying digits on the hands, which has been already explained, may serve as an useful auxiliary, in fixing the recollection of the series of products. It may be observed, that the numbers 1], 4, 9, 16, 25, 36, 49, 64, and 81, which occupy the diagonal, are the second powers or squares of the successive digits.—From MULTIPLICATION.] ARITHMETIC. 149 the inspection of the table, we gather that 1 is the ter- minating figure in the ¢Aree products, 1, 21, and 81; that 2 terminates the siz products 2, 12, 12, 32, 42, and 72; that 3 occurs as the terminating figure in only the two products 3 and 63; that 4 terminates the four pro- ducts, 4, 14, 24, and 543; that 5 terminates likewise the jive products, 5, 15, 25, 35, and 45; that 6 is the ter- minating figure in the five products, 6, 16, 16, 36, and 56; that 7 terminates only the two products 7 and 27; that 8 terminates the jive products, 8, 18, 28, and 48; and that 9 occurs only ¢wice as the terminating figure, in 9 and 9. It hence follows, that, out of ¢hzrty-four chances, there are szx that any composite number should end in 2; five chances that it should end in 5, 6, or 8; Jour chances that it should end in 43 three chances that it should end in 1; ¢wo chances that it should end in 3 or 7; and two chances likewise that the terminating figure should be 9. ‘These very different proportions in the re- currence of the several digits at the end of a number, may be remarked in the large tables of products. It likewise appears, that the bulk of the prime numbers must termi- nate with 9, 3, or 7, and the rest with 1. Notwithstanding the simplicity and obvious advantage of the Multiplication Table, it yet forms no part of the elementary education of the Hindus; asingular fact, which might seem to contradict the received opinion, that Pytha- goras brought the knowledge of it, with other higher ac- quisitions, from the East. ‘The boys in India are, no doubt, obliged to supply the want of this important help, by the tedious process of repeated additions, till practice at last renders them familiar with the products of the ordinary digits. In like manner, our young scholars are now left to grope their way without a guide through Addition, till the 150 FIGURATE [ MULTIPLICATION. — experience of many trials makes them acquainted with all the binary combinations of the lower numbers. It may be instructive, to compare the operation of an example of compound multiplication in the ordinary way, with another performed by deficient figures. In this instance, the working is 4819 5O.2\ evidently easier with the de- 378 49.9. ficient figures, since lower di- 38552 10452 ; . , 33733 L04542 gits are adopted in the multi- 14457 2028), plication. But it must be ob- 1821582 RS” o9 served, that a deficient figure or 193169.2 reverses the character of all the and 1821582 digits which it multiplies, The restoration of the ordi- nary figures is better understood, if, as here, it be made by successive steps. Another example, the same 4209.5 as what afterwards occurs, ae WARS. may be likewise produced. 295880 aca By a little practice, the work- 73970 159.60 147940 ing with deficient figures were10 Be would evidently become ea- 237739580 3>83.66580 sier, and more expeditious or 237739580 than the common way. Deficient figures applied to the multiplication of the di- gits themselves, furnish nearly the same practical rule as in Palpable Arithmetic. Thus, to multiply 9 by 6, the operation may be performed, by substituting for them 11 and 1, or 10 diminished by 1 and by 4. The result is consequently 154, or 54. But it appears ‘a obvious, that the last figure 4 is the product of the 354 deficient units, \ ands, and that the two first figures 1 15 or 5, denoting tens, are merely 10 less than the 154 sum of 1\ and 1),,or of 9and6. This accordingly if MULTIPLICATION.] ARITHMETIC. 151 is the rule given by the Arabian and Persian authors, It was farther simplified, however, in the earlier treatises on arithmetic published in Europe. Orontius Finzeus, of Briancon, Professor of Mathematics at Paris, who wrote his little tract in 1525, directs a crosstobe 9g l drawn, on the one side of which the digits ae 9 and 6 are placed, and opposite tothem, 6 4. on the other side, their defects from 10, or 4 4 1 and 4, are set down. ‘Then 4 is multiplied into 1, and the product 4 noted under them, while, following the oblique arms of the cross, 4 is taken from 9, or 1 from 6, to leave 5 for the place of tens. But it is evident, that to subtract 4, or the excess of 10 above 6, from 9, is precisely the same thing as to add 6 and 9 together, and then take away 10, as in the previous rule. The other variation of the process, by taking 1 from 6, must likewise give a similar result. As another example, a little more difficult, suppose 7 were to be multiplied by 6. These 1% digits are equivalent to 1% and 1s, so ah ; r : that their product is 162 or 42. ~ , : With the cross, the 4 multiplied into 19 et ie 3 gives 12, which leaves 2 units, and 162 3 advances 1 to the rank of tens; and 42 4: 2 6 diminished by 3, or 7 diminished by 4, supply 3 ad- ditional tens, making up 42 as before. The cross might also be accommodated to the multi- plication of numbers exceeding ten. Thus, j3 3 the product of 13 by 12 is found by setting oe opposite to them 3 and 2, their excesses 12 2 above 10; then multiplying the 3 by 2 for ~ 15 6 units, and adding cross-wise the 2 to 13, or the $ to 12 152 FIGURATE [ MULTIPLICATION. for tens. The latter part of the operation is evidently the same thing, as adding the numbers together, and deduct- ing ten from their amount; a rule which the Persians also employ in this case. Let this method of li afc be ive to the squares of the 2.7 digits from 9 o4 BY es aye toGinclusive. 9 208 Baty 3 6 4s Bac hy Cliente eke aoe 6 Theoperation is performed with great facility, the deficient figure being squared for units, and subtracted from its comespandlings digit to express the tens. It hence appears, that a square must have the same ter- minating figure, if the root end in | or 9, in 2 or 8, in 3 or 7, in 4 or 6, It likewise follows, that all square numbers terminate in these five digits, 1, 4, 5, 6,9, which lie equally distant on each side of the middle one 5, from which they differ by 1 and 4. When a number ends in 1, its square raot must end in 1 or 9 ; whenit ends in 4, the root ends in Z or 8; when it ends in 5, the root will also end in 5; when it ends in 6, the root will end in 4 or 6 5 and when it ends in 9, the root will end in 3 or 7. Un- less, therefore, in the case of 5, there are always two cor- responding terminations of the root, making together the number Zen, The Arabian and Persian, and likewise the earlier European writers on Arithmetic, enumerate several dif- ferent ways of performing Multiplication. Of these, it may be proper to select the most remarkable varieties. 1. The rudest mode was that of Cross-Multiplication, in which the distinct products of all the digits of the mul- liplier and of the multiplicand in every direction, are se- wv sf aad MULTIPLICATION.] ARITHMETIC. 158 parately set down. No figures are carried in this process, which is yet very complex and embarrassed. It was the method also that the Greeks were obliged to follow, for want of a simple denary notation, and has been already illustrated in the first example. 2. The commonest way was perhaps the Diced or Tessela- ded Multiplication. ‘This differs $69 8.4 nothing indeed from the ordi- a - 2 ¥ ar nary method, except that the ats{sirtor digits of the products, as they Tl4al7lolayol successively arise, are deposited aoe Ona in cells. The onlyadvantageof 2377395 80 such a form of operation, consists in the preserving better the regular ' alineation.of the figures. 3. Another mode of Multiplication was called the Qua- drilateral or Square. In this arrange- ment, it was not requisite to advance the products according to the place of the corresponding figure in the mul- tiplier. The same effect was pro- duced by an oblique summation. Be- 7% 9773 9 ginning on the right hand at the top, the 0 is marked at the corner of its cell; the 8 and 0 give 8, for a place lower; 8, 7, 0, throw 5 to the step below this, with 1 to be carried to the next addition, in the same slanting direction. 4. The last form of Multzplication which deserves any particular notice, is what was styled by Lucas de Burgo, the Latticed, or Rhomboidal. In this operation, the products of all the digits are severally dispersed in lozenges, or in square cells divided by diagonal lines, a form of procedure 154 FIGURATE [ MULTIPLICATION. by which the fatigue of carrying to the higher places is entirely spared. ‘The method, however, admits of some variation. 1. The multi- g dilip nel & oe! g plier and multiplicand may be written along the top, and down the left hand side of an oblong, which is sub- 5 divided into square cells, these again being parted by diagonals running obliquely SESE Se eee downwards from right to left. ‘The multiplication begins at the left corner above, and the successive products are inscribed in the cellular triangles of each horizontal zone. The summation is then performed along the diagonal lines. This figurate process was followed by the Hindus and Persians, among whom it obtained the technical name of Shabakh. 2. Another variation of the general mode consisted in writing the multiplier along the top, and the multiplicand down’the left hand of a divided quadrangle, the products beginning with the units, and proceeding along the horizontal columns from right to left; the sum- mation then sets out from the right corner, and runs up slanting to the left. —This mode of operating was peculiar to the Arabians and Persians, and by them communicated to the Hindus, who occasionally use it. On the next page is an example borrowed from the judicious travels of Sir John Chardin. Suppose it were sought to multiply the number 36985 by 6428. The Persian Arithmeticians, having drawn a rhomboid, would, beginning at the top, write these numbers downwards along the upper sides, and then divide the figure into equilateral triangles, by com- bining oblique with horizontal lines. mMuutipLicaTion.] ARITHMETIC. 155 — Now, the multiplication is carried along the rows on the left side of the rhomboid : 6 into 3 gives 18, which is dis- posed in the uppermost triangle and the one below it; 6 into 6 gives 36, which is deposited in the two next trian- gles; and the same process is continued through the se- ries. Again, 4 times 3 makes 12, which is placed in the two uppermost triangles of the next row. The rest of the operation of filling the triangles is easily understood. But to collect the products, the figures in each horizontal row, beginning at the bottom, are added up, and the tens car- ried to the one immediately above it. ‘Thus, the zero at the point of the rhomboid remains unchanged; in the row above this, 4, 4, 0 make 8; in the next row, 2, 6, 6, 1, 0 make 15, and 5 being set down, the 1 is carried to 156 FIGURATE [MULTIPLICATION. the higher row, 8, 7, 8, 1, 2, 2,0, making 29, of which 9 is set down, and 2 carried to the row above it. In this way, the summation is quickly performed, giving 237739580 for the complete product. It will be admitted, however, that such artificial helps may prove useful in laborious and protracted multiplica- tions, by sparing the exercise of memory, and preventing the attention from being overstrained. Of this description are the Rods or Bones, which we owe to the early studies of the great Napier, whose life, devoted to the improvement of the science of calculation, was crowned by the invention of logarithms, the noblest conquest ever achieved by man. These rods were small squared pieces of ivory or bone, box or silver, about three inches long, and only three-tenths of an inch in breadth and thickness. On their four sides, were engraved the successive columns of the common multiplication table, the cells being parted by diagonal lines running obliquely downward from right to left.— But instead of rods ar a surfaces, cee SEES SE SESE INNIS | \e CON | SESSA PAINS aaa BSESSIN A eS muitiptication.] ARITHMETIC. 157 mere slips of ivory could likewise be used, the one of them containing the row of digits being termed the indew. rod. ~ Let us resume the formerexample. The rods, inscribed with the several figures of the multiplicand, 36985, are selected and set in the same order, with the index placed before them ; then, opposite to the several figures of the multiplier, 6428, on the index, but going backwards, the num- bers in each horizon- tal column are taken, the pair of digits in each rhombus, or : double triangle, being always added; and finally, these rows of the products corresponding to each digit of the multiplier, being transcribed and properly disposed, are collected into one sum. ‘Thus, opposite to 8, the last digit of the multiplier, and proceeding from the right along the horizontal column, there ae occur these figures: first 0; then4 and 4, 147940 |& or 8; 6 and 2, or 8; 7 and 8, or 15; and 221910 1 carried to 4 and 4, make 9; and lastly, Ree the 2. The other rows are easily formed in the same way.—It is obvious, that if the horizontal columns op- 158 FIGURATE [ MULTIPLICATION, — posite to 8, 2, 4 and 6, were supposed to be detached and combined into an oblique group, the similarity to the Persian mode would be very striking. But, without formally adopting either the figurate rods or the rhomboidal cells, it will sometimes be convenient, in very long multiplications, to form, by successive addi~ tions, an extemporaneous tablet of the digital products of the multiplicand. ‘The application of this help is easily conceived. A very neat method of trying the accuracy of any pro- cess of multiplication, consists in casting out the 2imes.— Since any number must always be composed of repeated nines with some remainder, every multiplier and multipli- cand are only certain multiples of nzne, with correspond- ing excesses. Wherefore, their compound product will contain some involved multiples of nine, with the product -of those excesses. Or, conceive the numbers to be mul- tiplied, were exhibited on the Nonary Scale ; the first bar, or that of units, would evidently receive the product of the remainders of their division by nine. ‘This reasoning is quite general, and must consequently apply likewise to the casting out of eleven or of seven. But the remainder of the division of any number by zine or eleven is readily found in the way before explained; and it was likewise shown how the sevens could be cast out. Yo illustrate the application of the principle, let the first example be resumed. It was found, that 748 multi- plied by 632, give 472736. But the nznes, being cast out of the multiplier and multiplicand, leave 1 and 2, and out of the product, 2; which affords a strong presumption, though not an absolute proof, that the operation had been MULTIPLICATION. ] ARITHMETIC. 159 correctly performed. In the next example, the multiplier 378 is divisible by 9, and so is likewise the product 1821582. In the third example, 9 being cast out of 36985 and 6428, leaves 4 and 2; but on casting 9 also out of the product 237739580, the remainder is 8, or the product of 4 and 2, as it ought to be. Again, the multiplicand 748 in the first example being divisible by eleven, the product 472736 is likewise divisible, for the alternate figures 6, 7, and 7 make 20, and the other alternate figures 3, 2, and 4, give 9 or 11 less. In the se- cond example, if eleven be cast out of 4819, there will remain 1, and out of 378 there will remain 4; but the product 1821582 likewise leaves 4 on being divided by eleven. In the last example, eleven being cast out of 36985 and 6428, gives the remainders 3 and 4, which being mul- tiplied, make 12 or an excess of 1: But eleven, on being thrown out of the product 237739580, leaves alsol. In like manner, if seven be cast out of 748 and 632, the ex- cesses are 6 and 2, which give 12 or 5, for the remainder of the division of the product 472736 by 7. In the next example, the multiplier 378 is divisible by 7, and so is the product 1821582. Lastly, if seven be cast out of $6985 and 6428, the excesses are 4 and 2, which give 8 or 1 for the remainder of the division of 237739580 by 7. The last of these methods of proving multiplication is sometimes rather tedious, though greatly simplified by at- tending to the order of the recurring series, 1, 3, 2, 6, 4, 5. The casting out of the elevens may be conjoined with that of casting out the nzues, to strengthen the assurance of the accuracy of an operation. But the latter mode is the one most generally adopted. This elegant numerical pro- perty was known to the Arabian writers on arithmetic, 160 FIGURATE — [mvtriIprication. who styled it the Tarazu or Balance. Yet the Hindus are still unacquainted with it, and have no other way of pro- ving multiplication but by reversing the process itself, and converting it into division. It is evident, from the nature of notation, that, in the descending scale, the products corresponding to each — figure of the multiplier, instead of being advanced, should be shifted backwards. Hence the common rule for the multiplication of decimal fractions—to cut off as many de- cimals as are found in both factors. But, since the re- ~ mote decimals are of trifling import; a very com- 4 59¢ modious abbreviation is, to begin the process 1.618 at the place of units, and reject the very low 4-256 ae 2.542)" terms. Passing to the 6 of the multiplier, the 4.2 last figure 6 of the multiplicand is struck off ; but, 34) as it would have given 36, the nearest whole num- 6.854 ber 4, expressing the tens, is carried to the product of 6 into 3, making 22. The same thing is repeated at each multiplication. In far- MuLrieLicATION TABLE. ther illus- Duodenary Scale. tration of J[1/2|3} 4] 5] 6] 7] 8] 9] Of Of1 the prow #16] 8] O10] 12] 14 | 16 | 18} 10/2 sleeper 9 | 10 [13 | 16 | 19 [ 20 | 23 | 96 [2913 oe [774 [8%| a0 | 2a] 5if-a0 eaif B8if mnultipli- | 21 | 26 | 20 [34 [39 [42 [47] 5 cation, we [ 30 | 36 | 40 | 46 | 50 | 56 | 6 shall per- «| 41 | 48 | 53 | 50] 65 pure form the Rees te [§ same ex- er eto amples on | Sous - “MULTIPLICATION.] ARITHMETIC. 161 the Duodenary Scale. It will be convenient, however, though not quite essential, to construct previously a table of products. Let it be required then to multiply 2957 3965 the number 4819 by 378. Transform- arise aS: ed into the Duodenary Scale, they will 47634 sane? stand thus: The same operationislike- 5692 985% wise here performed by deficient fi- 730106 890106 730 106 gures. Suppose the large numbers 19401 2 59.1 3878 4)48)5 36985 pad 6428 were now con- TapaANIG RR, verted into the Duodenary Scale, 105907 13938 their multiplication, in common 123288 “issh, d in deficient fi ould 54263 71884 and in deficient figures, would pro- — Go75993g 758K TSH ceed in this manner. 67750938 The mode of proof which in the Denary Scale employ- ed nine and eleven, will evidently be eleven and thirteen on this Denary Scale. ‘Thus, in the former example, eleven being cast out of 2957 leaves 1; and out of 276 it leaves 4; but the product 730106 divided by eleven, gives like- wise a remainder of 4. If ¢hzrteen be thrown out of those factors, it leaves 9 and 1; and out of the product, it also leaves 9. In the latter example, eleven divides 19401 and 3878, with remainders of 3 and 4, but it likewise divides 67750938, leaving 1, the same in effect as #welve, the pro- duct of 3 and 4. Suppose ¢hzrteen to be cast out of the factors, the remainders are 9 and 6, which being multi- _ plied, give fifty-four, or an excess of 2; but the product 730106 likewise gives the same excess. But the Duodenary Scale is of greater consequence, when viewed in the descending progression, since the subdivi- M 162 FIGURATE [ MULTIPLICATION. sion by twelves has, toa certain extent, obtained currency in the denomination of money, and of weights and mea- sures. A few examples will explain the ma- 6.923 nagement of these fractions. Suppose it were sought to multiply L.6:15: 34, by 53, the opera- 230.90 tion with Duodecimals will be performed thus: 250.603 The product 250.693, converted again into the Denary Scale, gives L.358:11: 6%. This result is 6765625 more easily brought out than by Decimals, 53, as in the operation here annexed,—If the — 20.296875 Pound Sterling had been divided into 12 lead aaa 's) Ne Nie | ies, 358.575125 shillings, as the shilling is into 12 pence, the hs application of Duodecimals to accounts would have been extremely convenient. Duodecimals are best adapted, however, to practical mensuration, where feet, inches and their sub- 2.73 ordinate parts, are brought into play. Thus,if 2.73 it were required to find the solid contents of a 5.26 ‘ . 1.629 log of timber, 2 feet 74 inches square, and 27 799 feet 54 inches long. ‘The successive multipli- 6.9469 i oul fi : : 23.56 cations would be performed in this way, all the {169160 figures below the fourth place being excluded, 18.4183 as of little significance. The result is 156 cubic aris! feet, with about 2% twelfth parts, most inaccu- Foacne rately called, in this case, cubic inches. But these fractions will likewise readily apply to the mensuration of round timber ; for the relation of the cir- cle to its circumscribing square would be expressed in duodecimals by the number .9512. Suppose, for exam- ple, a cylinder 4 feet 2% in diameter, and 41 feet 10% long. The operation is thus performed: Four places of duo- decimals only are retained, though in actual practice two -MULTIPLICATION.] ARITHMETIC. 163 places may generally be reckoned 4.29 more than sufficient. The area of 4.99 the circular base hence exceeds by _—_14.00 12.0723 ; : 856 41.06 a minute fraction 12 square feet, ! 3209 4824900 but the solid contents of the cylin- 15.0769 12.0723 der amounts extremely nearly to. ‘9512 sh hee 2 ‘ 3 . 11.4081 — bs 4043, or about seven hundred cubic "7552 404.7050 feet.—A similar mode of proceeding 160 could easily be extended to the 30 12.0603 mensuration of cones and of spheres. To facilitate the operations with sexagesimals, it seem- ed indispensable to have a more extensive multiplication table, that should include the mutual products of all the numbers from one to sixty. Such a table was actually constructed early in the seventeenth century by Philip Lansberg, a Dutch clergyman, who resided at Middle- burg. It was nearly about the same time printed likewise in the Arithmetic of Adrian Metius; and has been since ex- hibited under various forms by Dr Wallis and others. The table of sexagesimal products, however, is now compara- tively of small utility, since the practice of those fractions has at last fallen into almost total disuse. It might be conveniently superseded by a table of products, carried as far as one hundred, which, if rightly managed, would vastly abridge and expedite the most laborious arithme- tical operations. The daily habit of using such a table could not fail to imprint a great part of it on the memory of the diligent scholar,—an acquisition of immense value, both in the pursuits of science and of commerce. To every person, it would give facility and correctness in their calculations. In this view, therefore, I have, with very considerable trouble and expense, framed an extended 164 FIGURATE [ MULTIPLICATION. table of products to accompany this volume. Its con- struction will be easily understood, and to render it more compact, I have omitted the rows of the products from 1 to 10, which are easily supplied. Along the upper horizontal line is placed the multiplicand, and, on the vertical one towards the right hand, the multiplier; and opposite to them, the product occurs in a descending co- Jumn. Suppose a former example were resumed: To multiply 36985 by 6428. These numbers are to be dis- tinguished into periods of two figures each. 36985 Beginning then at the left hand, 28 times _ 6428 85 gives 2380, which is set down; 28 times Brn 69 gives 1932, which is also set down two 84 mt 5440 figures in advance; and 28 times 3 gives 84 44.16 for the advanced place. Again, the products 192 by 64 are 5440, 4416, and 192, which are 237739580 successively advanced two places. Let it be required to multiply the decimal fraction .8134732 by .9135455, the former of which is double the sine, and the latter the cosine, of the arc of 24°. «Here the multiplication begins at the left hand, and all the figures of the product beyond the se- .g 1 3 47 3 9 venth place are excluded. But 91 9 1354 5 5 multiplied into 81, 34, 73, and 2, gives 73 é ; iby | : respectively 7371, 3094, 664 and 2, 66 4% which, at each step, are moved two _ 4% : , 2.8 Oe Bp places backwards. Again, 35 multi- 11 plied into 81, 34, 73, gives 2835, 119, 3 and 3, which are all drawn back two a: 5 places. ‘Then, 45 being multiplied by 4 81 and 34, gives 364 and2. And, -7 #3 14 4 8 lastly, 5 into 81 gives 4. The several distinct products MULTIPLICATION.] ARITHMETIC. 165 being now collected, the amount is .7431448, which con- sequently expresses the sine of the double arc, or 48°. This table may likewise be accommodated to the opera- tions of sexagesimals, for nothing is wanted but to con- vert the products in seconds, thirds, fourths, &c. into the numbers of a higher dimension by an easy division of 60. Thus, resuming the last example, the cosine of the arc of Z4° is expressed sexagesimally by 54° 48’ 45” 50’’, and the double of its sine is 48° 48’ 30” 12’”; making the lat- ter, therefore, the multiplier, because the part 45 appears repeated, the operation will proceed thus: Beginning at the left hand, the product of 48° into 54° is 2592’, or 43° 12’, which is set down; the pro- 54° 487 487" 50m duct of 48° into 48’ is 2304”, or .48 48 30 12 38 24”, which is set down; the pro- -43 12 388 24 duct of 48° into 45” is 2160’, or 36”, 36 40 which is also set down; and the pro- 43 12 duct of 48° into 50” is 2400, or 40”, ao S which is set down after the 36”. “OT A 1] Again, the products of 48’ into the ——— several terms of the multiplicand, are FW eee the same as before, only removed all one place lower; consequently 37’” is substituted for 36’” 40%", as the near- est integral value. Next, 30” multiplied into the terms of the multiplicand evidently reduces them to one half, and throws them two places lower. Lastly, 12’” multi- plied into 55°, the nearest value in whole numbers of the multiplicand, in effect divides it by five, and throws the quotient 11 three places lower. ‘The amount of the whole is hence 44° $5’ 19” 16”; and therefore the double of this sine, or 89° 10’ 38” 32”’, must express the chord of 166 FIGURATE — [Mu.riptication. 96°. Ptolemy, going no farther than seconds, makes it g9° 10°39”, The Arabians performed the multiplication of sexagesi- mals by help of square cells, parted downwards from left to right by diagonal lines. The multiplicand being placed along the top of the quadrangle, the multiplier ascended on the right side, and the operation of multiplying-them proceeded from right to left, as customary in the writing of those Orientals. In short, this process was exactly the reverse of the ordinary mode followed in the multiplica- tion of numbers on the Denary Scale, which they had a- dopted, probably without any change or modification, from the Hindus. DIVISION. | TH1s process, being merely the reverse of Multiplica- tion, consists in subtracting one number repeatedly from another. The former is called the Divisor, the latter the Dividend, while the answer, signifying how often the sub- traction needs to be made, is termed the Quotient. The © principle of numerical arrangement suggests the means of abridging this operation. Suppose it 1554 74: were sought to divide 1554 by 37: 1 371, 37 : , 118 37 Let 37 be subtracted in succession [J], 37 ~~ I, 34 from 155, which, standing one place 81 or higher than the units, corresponds to HI. St tens; the several subtractions are jy. 37 marked by I, I, II, and IV, which 7 belong to the place of tens, and from remainder 7 with 4 DIVISION. ] ARITHMETIC. 167 annexed to it, the divisor 37 is again subtracted twice:— Whence the quotient is 42, or the number of times that $7 is contained in 1554, or must be subtracted before it exhausts this dividend. But such an operation is evidently circuitous. The most obvious improvement is to frame, as in compound multiplication, a small tablet of the digital products of the divisor, and to subtract always the nearest less number from the successive terms of the dividend and the re- mainder. Let it be required to , 423)22028148(52076 divide 22028148 by 423. The 92 846 2115 tablet of products is formed by 1269 878 . id > 41692 846 the successive addition of the di- 59115 “S018 visor 423 and its multiples; of 6/2538 2961 these; the number opposite to 5 deed 9 Bes comes nearest to the first four 9/3807 a aaee terms of the dividend; and the remainder 87, with the next figure annexed to it, is ap- proached the nearest by 846, the next remainder 32 with the annexed 1 is less than the divisor, and therefore, a zero is put in the quotient to preserve the place, and the following figure 4 is joined. The rest of the operation is easily conceived. This method, however, is more tedious than needful, unless the quotient should consist of several figures. In other cases, a little practice will show how to choose the proper multiples of the divisor. The dots placed under the figures of the dividend as fast as they are taken down, or annexed for a new division, point out the ranks of the divisor. Deficient figures may likewise be sometimes in- troduced with advantage. An example will explain this: Suppose 1797848 were to be divided by 472. With 168 FIGURATE [ DIVISION. he defici = the deficient 8 475)1797848(3809 5>2)1809.9.5(40.11 gures, the ope- 1416°°° 199.8 ++ ration is some- 3818 1109. what easier, tho’ 3776 | A00%5 . 4248 425 it was here unne- 4IAS 542 cessary to make 0 530. any alteration site on the dividend itself. In the course of working, it will often happen that the product of the divisor, after being written down, will appear greater, instead of less, than the part of the dividend from which this is to be taken; but without substituting a lower product, the oversight would be rectified by a defi- cient digit at the next step. ‘Thus, instead of the first two figures 38 of the quotient, we obtain 4°, which is equi- valent. The Arabians and Persians perform division like mul- tiplication by a figurate process, in which every step is distinctly set down. ..A sufficient number of equidistant vertical lines being drawn, another horizontal line near the top of the board is made to intersect them, and, im- mediately under it, is placed the dividend, the divisor be- ing set down at such a distance below as may allow space for the operation being repeated at each step on a lower bar. Having found how often the first figure of the divi- sor is contained in the corresponding part of the dividend, the quotient is placed above the horizontal line, opposite to the termination of the divisor, and now multiplied into each of those digits in succession, and the products subtracted from the dividend. ‘The divisor is then shifted upwards a step farther back, and the process recommences again. g DIVISION. | ARITHMETIC. 169 An example will show this A complex mode of proceed- \7)7 791718 )418 1 ing. Suppose, as before, bi jp Zoe! that 1797848 were to be 9 divided by 472. Hight | |97 vertical bars being drawn, outs the figures of the dividend i are inserted across thetop, Eas and those of the divisor at the bottom, the 4 being set opposite to 17, which it divides. ‘The quotient $ is then placed in the A, same column with 2, the Pg termination of the divisor, ‘ and multiplied into each yd Bi a ficure in succession. ‘The 4/7 | 2 products 12,21,and6are 417 {9 separately subtracted, lea- ving 381, &c. for anew dividend. The divisor 472 is now repeated a step backwards, and the operation renewed, the next digit of the quotient being 8, and its successive products 32, 56, and 16.—This method of performing di- vision, though unnecessarily tedious, requires no effort of memory. It is also sometimes a little varied. The mode of compound division, as now practised a- mong the Hindus, appears still more involved and labo- rious, only the figures of the dividend and its remainders are obliterated as fast as the operation proceeds. Re- suming the former example, the divisor 472 is placed under a corresponding portion of the dividend 1797848, in which it is contained 3 times; then thrice 2 is 6, ae a . 170 FIGURATE [ DIVISION. which, taken from 7, leaves 1, to be writ- 3 ten above it; thrice 7 is 21, which, taken 4 from 79, leaves 58 above it; and thrice 4 BA AG | is 12, which, taken from the 15, leaves 3. 1797848(3809 The remainder for the next division is, ° “gee d therefore, 3818, which contains the divisor, 472 repeated one place farther back, 8 times. 472 Now, 8 times 2 is 16, which, taken from 18, leaves 0 and 2 to be placed above and below those figures ; again, 8 times 7 is 56, which, taken from 80, leaves 24; and 8 times 4 is 32, which extinguishes that number. © The re- mainder for a new division is only 4248, which, passing over one place, contains the next divisor, when shifted to the end, 9 times; but 9 times 2 is 18, which, taken from 48, leaves 30; 9 times 7 is 63, which cancels the 3, and takes 6 from the 42, leaving only 36; and lastly, 9 times 4 is 36, which is therefore extinguished on the board. All the figures being successively obliterated, the result only of the operation is retained. Almost the same crowded and intricate mode of per- forming division had early prevailed in Europe, and even maintained its ground till about the beginning of the last century, since Dr Wallis constantly used it. This form of proceeding was by the Italians, according to Lucas de Burgo, styled the galley, either from the swiftness of its operation, as he thinks, or rather from the taper- ing shape which the group of digits acquires in the course of the work. In farther illustration of the pro- cess, I shall select an example from the numerous cal- culations by Regimontanus, in his tract relative to the quadrature of the circle, written as early as 1464, but not published until 1532. The question here proposed is te a DIVISION. ] ARITHMETIC. 171 divide 18190735 by 4153 the divisor being 71 repeated at every step backwards, and stand- 3134 i

- by 2 2 4 a . edd wath hod f ol- 7, it will be changed into 35, Or stay Wherefore, c 7 lecting these fragments into which the original fraction 4: was broken, it will assume this compound form, 59 3+ ; meaning that re joined to 3 is to divide 2, and the quo- tient then added to 5 for a divisor of 4. If the extension of the lines which separate the numerators from their com- plex denominators be omitted, the fraction will merely 4: run in a diagonal direction ; thus, b+ 2 $3+6 Fractions of this kind have, from the nature of their composition, been called Continued Fractions. They were first proposed about the year 1670 by Lord Brounker, President of the Royal Society, and improved by Dr Wallis, but afterwards cultivated chiefly on the Con- tinent. Few speculations have proved finer in theory, or led to a more useful or prolific disclosure of the relations and properties of numbers. The different parts of a continued fraction may, by an inverted process, be recombined and consolidated. ‘Thus, the fraction = 42 “846 7 is again restored, by beginning at the ARITHMETIC. 189 ." . 4 . ° . ° ° ] 4 extreme portion 35° which being multiplied by 7, gives Pr Pa . 4 and then the complex fraction —; BT the multiplication of its numerator and denominator by ; 1s finally converted by : 108 ; ‘ t 27 into cere But if the terms of a continued fraction be taken in succession, other subordinate fractions will arise, which approach at each step to the true value. ‘Thus, 4: ; assuming only the two first terms pies it forms the frac- ita (: : 12 108 tion j53 now the consecutive fractions B” iz. and 140" compose evidently an approximating series. | It is more convenient, however, to derive these second- ary fractions by a direct succession, than to combine them by help of a retrograde procedure. Nothing in fact is wanted, but to reconsider the several multiplications that took place. ‘The fractions as they are formed therefore 4.3.7 +-4.6 ca ate ee Pe Ports SUNG Sansa gts OOO Tey To 7S The first ’ 4. . ¢ fraction — has therefore both its numerator and denomi- nator multiplied by 3, the denominator of the next frac tion, and 2, the denominator of this fraction, added to the product of the denominator. Both numerator and de- nominator of this new fraction are next multiplied by 7, the denominator of the extreme fraction ; and the products of the numerator and denominator of the preceding frac- tion multiplied by the numerator 6 are added. The sym- metry of the process will appear more distinctly, if the ficti- ; meee hie ; tious fraction —- be placed at the beginning of the series: 190 FIGURATE Thus, 2, 4. #3492 9. 12, og 127446 108 we feels 17° ava-+5.6° | flee Hence, if the numerators of a continued fraction be all of them units, the recomposition is effected by multiplying with the successive denominators, and merely adding the numerator and denominator of the preceding fraction. This simpler kind of continued fractions, being the most common and convenient, the mode of transforming them deserves a separate investigation. Suppose it were sought to decompose the fraction 287 ry Divide both numerator and denominator by 287, rr ° I - : and the complex fraction rae arises, In like manner, eee = by its numerator 131, and it is changed into a } 25 the fraction 5; PaEES Again, reduce the fraction rr by di- viding its terms kph 25, and it assumes the complex form ee Wherefore, introducing those substitutions, the original eek a exhibit these Cane phases : ? 6 1 —z-- Lastly, the fraction 7B. is converted into ray 28 I I. ale Il.— tH “T33 2 3 ire aa IV. S19 De ie 2+35y? 241 G+s5 * pedo adt baw s bh $+]... .|, Or more simply $+) 2+! thus, without the 2-1 6 reals: prolonged lines, a 6+ 6+1 ~ * ARITHMETIC, igh It is obvious, thet this process 287)992(3 of decomposition is conducted 861 precisely in the same way as the —:131)287(2 operation for finding a common “91 (5 measure. ~The successive quo- 125 tients, 3, 2, 5, 4, and 6, here be- _ -6)25(4 come the denominators of the sls | 1)6(6 continued fraction. To restore a continued fraction again, the most obvious mode is, beginning at its extreme term, to re-ascend by successive combinations. ‘Thus, resuming the same ex- 1 6 1 ample, = is first changed into ¢ _ then >; «Rae into oz i “ is condensed int 1a d last] 131°? next gas ondensed in 07873 and lastly, gratis . 287 transformed into 595 992° By commencing this process of consolidation from any intermediate step, a series of subordinate and approxima- tive fractions will be likewise obtained. Thus, ar gives a» 2 1 and of the three terms, S41 241 5 Be 5 1 11 ron first 7 is equal to 77 and 3, equal to 36? and then ; caus gor , ] condensed into 5g) again, the four terms, cy 241 5+1 | a age 1 21 give first 5 or 5 then exe ya make [- 3 and lastly, » Sah is 192 FIGURATE equal a Thus, the following series of approximative ; | 1 2 11 46 287 fractions has been formed, > 3? 7? 3g’ 159" ggg? Commen- 87 992° If the first of these fractions have its numerator and deno- 1 cen ¥ cing with= » and terminating in minator multiplied by 7, and the second, in like manner, by e . 7 3, theywill be changed into the equivalent fractions 31 and 6 ; I a1? whence the value decreases in the second by a But : ; 2 11 let the next adjacent pair of fractions, 7 and aud have their numerators and denominators multiplied by 38 and ; 76 a7 7, and they will become 366 and 3663 8° that the value increases at the third fraction by the difference _ In the same way, it i 46 159 1 28 diminution of ——, and the original fraction —— re- 7 6042 992 . ; 1 : ceives an augmentation of 757728 The series of subor- dinate fractions thus alternately oscillate above and below the true value, to which, however, they rapidly ap- proach. Hence the original fraction might likewise be expressed, by combining those alternating differences: I 1 ] Pay a eo FTE The gradual approxi- mation is, therefore, clearly marked, since these successive fractions have only one for their numerator. . ARITHMETIC. 198 It would be more commodious to discover those subor- dinate fractions by a direct procedure. Let the continued ] fraction — eo Ab 6 be resumed. The progressive my 1 l Lae 2 steps are thus exhibited: I. 3" IT. 3 Sym a rgay 1 Si : , 9}; instead of 2 in the preceding expression, sub- pfs 1.2.5+1 F eg ' stitute 23, and the result is pS ak abe ee eet pai 3.2441 ~ 3.2.5-45.1-+43 UII. 2541 11 7648 38 7% ee 241 52; instead of 5 in the last 1 expression substitute 53, and there comes out a= 5143 2.544442 11.442 46 754ES447 — 38447 — 159) and finally, 1 V. S41 241 ‘6+1 4is Begg 4< for 4 in the preceding, and ALL2 11.4.649.6411 seat = = 36.4.64+7.64+38 — it becomes changed into (46.6411 287, 159.6438 992 The order of succession is hence easily perceived. If 9 194 FIGURATE ‘ s). 90 ) the simulated fraction 7 be placed first, the approxima- - tive fractions will be formed, by constantly multiplying the series of denominators, and adding the preceding terms : Q-;4 @ 112 46 4287 Thus, [> 5° 7? 38° 159? 992" It may likewise be shown, that the alternate products of the numerators and denominators of these fractions differ incessantly in excess and defect by unit: Thus, 3.2-+-1= 7.1, 7.11—1=38.2, 38.46--1=159.11, and 159.287—1 =992.46. For resuming the former analysis, 7 =3.2+1; 7.11L=7.2.5+7 atid 38.2=7.5.243.2; 38.46238.11.44 38.2, and 159.11=38.4.114-7.11 ; and finally, 159.287 = 159.46.64-159.11, and 992.46 159.6.46438.46. This mode of decomposition, though employed here in a parti- cular example, is evidently quite general. These approximative fractions hence will terminate al- ways in the true value. Any fraction may therefore be reduced to its lowest term directly from the series of quo- tients evolved in finding the common measure of its nu- : , el merator and denominator. ‘Thus, the fractidn 175 pave the quotients 2, 3, 1, ¢ and consequently is converted into the continued fraction 2+1 ‘ 34-1 i ee 2; of aa the ap- 15) 4 proximating valuies' are 729° 7? 9 and 5 ? the last be- a4 a ing the reduced fraction. Again, the fraction 3765 fur- nished these quotients, 3, 1, 2, 4, and 5, It may there- ? \ hoursgive86400”. Where- ARITHMETIC. 195 fore be changed into the continued fraction = But this again will produce the approximative fractions, = = ’ ) fas a and oe the last being only the origi- nal fraction reduced to its lowest terms. The secondary fractions derived from the successive composition of the members of a continued fraction, it has been shown, differ from each other by alternating varia- tions, which have always unit ds their numerator. Such derivative fractions will, therefore, not only advance ra- pidly to the true expression, but mtist constantly approach the nearest possible, or exhibit the approximate value in the lowest terms. This property is of essential consequence in all operations with numbers, and furnishes many useful practical results. A few examples will justify the remark. Let it be required to express approximately the frac- tional portion of 24 hours, 2098) 8640(4 by which the solar year exceeds 365 days. This “25520087 excess, or 5 hours 48 mi- 217) 268 (1 nutes and 50 seconds, be- 217 : : ~$1)217(4 ing reduced to seconds, 7 904 makes 20930”, while 24 ~18)51(4 52 — 20930 2093 86400" ” 8640 The successive quotients are 4, 7, 1, 4, and 4, without fore the fraction » is to be decomposed. 196 FIGURATE pushing the last division with rigour. There results, conse- : pt uently, th t fi — quently, the continued fraction 41 7+1 i+1 ath 45 from which are derived the approximative fractions L vn Tinga ® SOs. 64 4° 99° 33° 161 8° 677" Some of these fractions are remarkable. The first frac- - 4 iby oy oa : . tion a indicates the insertion of one day every four years, being the correction of the Kalendar by the Bissextile or j 8 Leap Year introduced by Julius Cesar. The fraction 3-- indicates a more correct intercalation of 8 days in 33 years; a method proposed about six centuries ago by the Persian Astronomers, who, after the lapse of seven ordinary leap years, always deferred the eight return of the period one year - 8 d longer than usual.—If this fraction 3> had its numera- | 1 tor and denominator multiplied by 12, and those of = added to the products, another fraction 2h will be form- ed, of very nearly the same value. This last represents the intercalation established by Pope Gregory in 1582, and afterwards successively adopted by all the Christian powers except Russia, which affects to maintain the inde- pendence of the Greek Church. It implies the insertion of 97 days in the space of 400 years ; which is performed by combining the Julian system with an omission of three ARITHMETIC. 197 intercalary days in four centuries ; that is, the last year of each century, which falls to be a leap year, is not consi- dered as such, unless the number of the century itself is divisible by four.—But the Persian mode of correcting the kalendar was evidently simpler and more elegant, since in the space of 33 years it restored the coincidence which we now require the course of 400 years to effect. As another example, let it be required to express the English foot by the French metre, or unit of linear mea- sures. The metre being 39.371 inches, gives, on a divj- sion by 12, the continued fraction. Wherefore the approxi- l S41 ‘ 3-4 . . 1 mating fractions are 5? SMEAR 8D } 10, 19, 29 — 89° I+41 $2. 89 655 341 39, — 9 LAN '=-~——* eo 105 292 2149 141 Hence the foot is to the 241 Te 1 1 metre nearly as 3 to 10, and still more accurately as 32 to 105. As a farther illustration, it was found, in the extraction of roots, that the reciprocal of .7937 must express the side of the double cube, to discover which, without the power- ful aid of the denary arithmetic, exercised the utmost in- genuity of the ancient Greeks. If that decimal in relation to unit were converted into a continued fraction, the quo- tients forming the series of denominators would be I, 3, 1, 5, 1, 1, 4, 1, 1, 2, 1, 1 and 2. Wherefore the successive ans 0 SA es OF 60 227 approximations are 7? 7? a? BS? 99? 34° 3? 286° 277 504 | 349 635 i implest and ore’ aaa &c. Of these fractions, one of the simp 198 FIGURATE most remarkable is =; ; the cube of 63 being 250047, while = ; that of 50 is 12500, very nearly the half. The expression 04 , 636! however, approximates still nearer, the cube of 635 being 256047875, and that of 504 being 128024064. Continued fractions will afford elegant approximations to the square root of any number. Thus, the square root of 10 is 3, together with the remainder 1, divided by 6 and this quotient itself. The fractional part, therefore, changes, by repeated substitution, into x : 6, then into 1 6+1 641 6, and thus passes into a continued fraction, which has likewise no termination. ‘The square root of 10 is therefore 3-41 | 641 641 6, &c. Ifthe fractional part be suc- cessively combined, it will form this range of approximating fractions, of which the numerators and denominators evi- 1 6 ,87.,998 dently constitute a recurring series, 2 aan &c. The last of these converted into decimals gives 3.162277 for the square root of 10, being true to every place of figures. It is what the Arabians and Hindus some- times most inaccurately assumed for the circumference of a circle whose diameter is 1, and which Joseph Scaliger afterwards pompously announced as a perfect quadrature. ee 7 ~~ ARITHMETIC. 199 The square root of 11 must be 3, with the remainder 2 divided by 6, and the quotient itself. Wherefore, by re- peated substitutions, the root will be expanded into this continued fraction, 3-++-2 6+2 6+2 6, &c., which shoots on- wards without intermission. It may be reduced, however, to a simpler form, the expression for 4 i 6 is evidently the same as es | 6, the terms only being divided by 2. Whence the square root of 11 is 3--1_ 6, &c., the circu- lation occurring at every second place.—In like manner, the square root of 12 is 3-++3_ 6+43 6; &c., which may be con- verted into 3+1 241 6, &c., circulating also at every second place. But the terms of a continued fraction, which expresses the square root of any number, are not always capable of being so easily reduced. Thus, the square root of 7 must be 2+3 443 | 4, &c.; where the numerator, not being con- tained in the denominator of the repeating fraction, this 200 . FIGURATE will admit of no direct change. It may still be transform- ed, however, by a circuitous procedure. ‘The fractional ; 3 expression t43 . 4, &ec. gives, by reduction, this series of ; : E 12... 67 - 264. 1227 approximative fractions, a? 5° BB” 409° 1900! But the last of these, converted into a simple continued fraction, furnishes the denominators 1, !, 1, 4, 1, 1, 4, &c., from which the circulation is easily inferred. Wherefore the square root of 7 is 2-4 By tae i+1 ees 4, &c.3 and conse- quently the fractional part is thus expressed by successive ‘ 3 1 1 2 9 ri 20 "Sly J44 approximation arate 3 3° aan 77’ 31° 38” 599? 175 319 494 2295 271° 494 767° 5560 &° There is another method of decompounding fractions, which likewise deserves particular notice. An example ee, Gene ‘ 6 Se will explain this procedure. Resuming the fraction 992. 861 992—-131 it may be written 3.992” or 3999 » OF more simply 1 131 131 37 F000" : Again, the latter part 955 of the compound f P aare 917 992—75 raction, is transformed into G55, or —= 99g» or mere- 75 1D) eh ores tae 7 999° : But the fraction 509 1» in like manner, ARITHMETIC. 201 be 975 992—17 1 17 changed into [3G992 OF Wa.999 9 °° 13 13999 * : 17 986 992-6 1 6 OW, 999 = 58.992 = 58.992 58 58.992 “Bain, 6 G90%. <1 99% 228-71 2g 992 165.992 165.992 165 165.992" And, lastly, 2 1 : 992406 Wherefore, collecting these elements and re- F 287 ascending, the original fraction 992 is converted by sub- sth ie Vi lee ; +: ad ! d stitution into the alternating series, ee 3.7 ty. l 1 1 5.7.13.58 * 3.7.13.58.165 5.7.13.58.165.406 This se- ries, however, conveys rapidly, as will appear from the ac=,.».)%s.». aint 8tr ] 1 ! tual multiplication of the factors; thus, Soi tare 1 1 1 : s CY MCAT TA iggheenae: It is obvious that the fac- tors which compose the series representing the fraction als are only the quotients of the continued division of the de- rai nominator 992, by the nume- 131)992(7 rator 287, and the successive 917 remainders. The operation is 75)992( 13 here given at length, the same i jen 58 dividend being constantly re- ie ce neces peated. As another example, suppose it were required to con- vert the decimal expression 3.14159265 for the circum- ference of a circle whose diameter is 1, into a converging series: Dividing 1 by the part .14159265, and by the re- mainders which appear in succession, these quotients will 202 FIGURATE result; 7,113,4739, and 47051, &c. Whence the circum- ference will be denoted by 3+ = — ae ta ] 7.113.4739.47051 The first term of the series indicates the Archimedean proportion, and the conjunction of the next term gives that of Metius.—-This ingenious mode of transforming fractions was the invention of the celebrated Lambert, a + &c. mathematician and philosopher distinguished by his strong and original powers of thinking, who left his native spot in the verge of the Swiss Cantons, and accepted the invi- tation of the great Frederic to settle at Berlin, where he died in 1777. The other properties and combinations of fractions will be more easily discussed. 1. To convert fractions into others which are equivalent, but have a common denominator, it is only requisite to multiply the numerator and denominator of each fraction into the continued product of all the rest of the denomi- Leon's 4 nators. Thus, the fractions, 33? S and 7 are changed Sus 1:3.6.7 2.2. 5179 312.327 ce alae 105 140 2.3.5.7 3.2.51? 6.2.3.7 OF G3 53” B10? 210? sao i nd = The reason is obvious from the constitution of fractions. | 2. To add or subtract fractions, nothing is wanted but to reduce them to the same denomination, and the addi- tion or subtraction of their numerators will give as the re- sult, new numerators to their common denominator. Thus, S772 63 70 d VO a at and = — are first reduced to — ‘ and ang fo 105° 105 105° _ 0 ARITHMETIC. 203 . aa: Say way 3 their sum is —— or 1,43. Again, — subtracted fi Pe is ma a re oe Ah salar 4 15 14 pa 1 Bove 35° 55 88 3. To multiply fractions. It is evident that the frac- tional multiplier implies that the multiplicand is to be re- peated as often as the former has units in its numeration, and then subdivided as often into as many parts as there are units in its denominator. Consequently, the numera- tor of the multiplicand should be multiplied by the nume- rator of the multiplier, and divided by its denominator ; or, instead of this last operation, the result must be the same, if the denominator of the multiplicand be multiplied by the denominator of the multiplier. ‘Thus, the fraction Pager te Gk s ke : - multiplied into at signifies that — is to be repeated twice, and the amount to be then divided into three equal portions ; but =being doubled, makes a and this divid- ed by 3 gives = Instead, however, of dividing the 12 by 3, the effect is the same to triple the denominator 7, for A 12 7 ; : . Aide Li - and 5 are evidently equivalent fractions. ‘The division of the numerator being seldom practicable in whole num- bers, the corresponding multiplication of the denominator is the general and preferable method. Hence, successive _ fractions are multiplied by multiplying all the numerators for a new numerator, and all the denominators for a new > denominator. 3 5 saan ie 3 Thus, the continued product of —s i, and —» is ——, 4° 6 ( i68 5 » ° bd ¥s or 55: Again, the product of the successive multiplica- 204 FIGURATE Saar. La 8 . 616, f : tion of 5+ -G? and =» is 35 or 20. This result might likewise in this case be found, though more la- 5 % boriously, bythe method of Cross Multiplication. 2 + 10 1 : 11 7 The operation would proceed thus: > and ~ 13% 5 being the same as 54 and 23, the 2 is multiplied “3 | 1 ¢ into 5, and next into >; then = is multiplied 125 2 3 Tits i) 202. ; aL: ; 5 into 5 and into gi The sum is 125, which ava multiplied again by = , or by 1 and then by =; gives the total result as before. 4. ‘he most obvious method of performing the division of fractions is to reduce them to a common denomination, and to state the snp tke of their transformed numerators. Thus, to aie ~ by > z3 they are first changed into the : ‘ 21 20 equivalent fractions 5g and sy 39? and now 5 must express 28 21 the quotient. The same sat would be obtained if the divisor were converted into its reciprocal, and then mul- tiplied ; for the numerator and denominator being inter- changed, must exactly reverse the nature of the operation. 3 4: 5 Thus, the reciprocal of [ or 3 being multiplied into = 20 , gives D7» as derived before. The general rule, conse- quently, for the division of fractions, is to multiply cross- wise the numerator of the divisor into the denominator of the dividend for a new denominator, and the denominator of the divisor into the numerator of the dividend for a new numerator. ARITHMETIC. 205 Those fractions which are the reciprocals of integers, or have one for their numerator, exhibit, when transfer- red to the descending bars of any scale, either the simple repetition of unit, or a succession of digits expanding in . 1 uniform progression. It was already shown, that 9? con- verted into a decimal expression, gives first 1, with a re- mainder of 1; and must therefore evolve in its sub- sequent expansion always the same train of units. If 1 ; | the fraction —- be represented on this scale, the first term will be 1, with 2 of a remainder; consequent- ly, each succeeding term is constantly doubled, forming : ] the series .1, .02, .004, .008, .00016, &c. Again, — treated in this manner, gives first 1, with an excess of 3; and hence all the subsequent terms are successively tri- pled, composing the progression .1, .03, .009, .0027, 00081, &c. In like manner, since the quotient of 10 by : 1 6 is 1, with the remainder 4, the fraction =e converted into decimals, will form the series .1, .04, .0016, .00064, 000256, &c. proceeding by constant quadruplication. It may, therefore, be generally inferred, that the recipro- cal of any number is exhibited on a descending scale by a series of digits consisting of unit, followed by the excess of the index and its successive powers. The descending progression, which represents the reci- procal of any number, though it never terminates, will some- times converge to a definite result. Thus, the expansion psht of the fraction gor the sim of the terms .1,2,4,8 16 32 64 128 &c. 206 FIGURATE being .12499968, receives always another 9 at each step of the progress, and consequently approximates to .125, the true decimal value. In other cases, the summation of the figures forms either a perpetual repetition or a periodic ; 1 se ; circulation. Thus, rae being expanded, gives .1,4 1 which makes .1663936, but when pursued farther, ap- proaches constantly nearer to the complete expression : HW, ene : 7 166666, &c. Again, the fraction = Is expanded into the decimal terms .1,3,9 243 19119 729, &c., which, being collected to- gether, indicate the period .142857, 142857, 142857, &c. of incessant circulation. If the number whose reciprocal is expressed on a de> scending scale be only less than some power of the index, the same progression of digits will emerge at corresponding i intervals. Thus the decimal of 08 is .01020408, the | ¥ terms doubling at every second place. Hence > which 14 A is the same as 55, will be expressed by continually dou- bling 14, and descending two places. Accordingly, .14,28,56 112 7 224 | y > diame 4.48, &c. taken collectively, give .142857, in per- ARITHMETIC. 207 _ petual circulation. It follows likewise, that os being the double of this fraction, is expressed by the same series of figures, only commencing two places lower ; thus, 285714, A: 285714, 285714, &c. In the samemanner z will yet be de- noted by the same descending progress, beginning at the next binary period, or two digits still lower ; thus, 571428, 3 571428, 571428, &c. But the fraction 7 being equal to 42 98? must, from the principle already stated, be composed by the summation of the decimal .42, redoubled at every second place, in this manner, .42,84 168 336 672, which makes up 428571, 428571, 428571, &c. the same series again : 6 only repeating from the second term. Hence 7» or the double of the last, will be represented by this identical series, only beginning two places lower, or at the fourth 5 term; thus .857142, 857142, 857142, &c. Lastly, wor - being expressed by 70 and its successive reduplications, or merely by 7 redoubled at every second place, will be represented by 714285, 714285, 714285, &c. It thus ap- pears, that the several multiples of 142857, as j,9g,% far as the product by six inclusive, exhibit al- 285714 ways the same train of digits, and in the same ie order of succession. Seven timesthe same num- 714285 ber gives 999999. 857142 208 FIGURATE This property of the number 142857, which becomes _ always renascent in its multiples, may be deemed a sin- - gular arithmetical curiosity. It depends on the peculiar é z : sy all circulation of the decimal expression for the fraction , the digits of which are not disturbed by the lower multiplica- tions, because the denominator is less than 10. When the reciprocal of any integer is expounded by a © decimal circulation, the figures forming the period must evidently amount to less than the number itself, since the moment any remainder of the division is repeated, a cycle will commence. Any number will hence divide an integer, with a definite succession of ciphers annexed, or this quantity diminished by unit; and conversely, any periodic decimal may be converted into a vulgar fraction, of which the denominator consists of a certain repetition of zines. ‘These observations are easily extended to other scales of Numeration. Enough, however, has perhaps been already stated in explication of the Theory of ter- minating, repeating, and circulating Decimals, | NOTES AND ILLUSTRATIONS. | Note I.—Pages 1—6. Pusrosorusrs, misled by the hasty and careless reports of travellers, have generally much underrated the attainments of savage tribes in the art of numeration. From the mere scantiness of the terms which a rude people employs to signify numbers, it would at least be rash to infer the narrow range of their application. The language even of the most polish- ed nations, when traced to its radical form, is yet found to betray uncommon poverty in numerical expression. In the ancient Gothic, Tachund, or simply Hund, denoted only Ten; and the word Hundred, composed by annexing Red or Ret, the participle of the verb Reitan, to reckon or place in rows, intimated consequently that its root was to be ten times told. Inthe translation of the Gospels, which Bishop Ulphilas made for the use of the Goths in the fourth century, Fourcera, One Hundred, is expressed by Tachund Tachund, or the name for Ten merely redoubled. The word Teon had also become synonymous with Hund, and in the Anglo-Saxon version, compiled three centuries after that period, One Hundred is called Hund Teontig, or ten ten times drawn, The name for P 210 NOTES AND ILLUSTRATIONS. the next number on the Denary Scale, or One Thousand, is nothing but an abbreviation of Dzuis-Hund, or twice-ten, meaning that ¢en was to be counted over in a double succes- sion, making first one hundred, and then one thousand. The ancient Scandinavians were fond of reckoning by dozens, and they sometimes combined to a certain extent the denary with the duodenary scale. The Chinese likewise employ both those scales in expressing their cycle of szxty years, which consists of ten roots and twelve branches, the year of the cyele being signified by the remainders in counting it by 10 and by 12. In like manner, ecclesiastical historians usually marked the dates of events by the numbers of the Solar and Lunar cy- cles, or reckoned by 28 and 19, which return again in the same order after a period of 532 years, called the Cycle of In- diction. — Bishop Tonstall, in his Arithmetic, aptly compares the ex- tension given to numbers, by help of the denary system of classification, to the growth of reeds, which, though slender, are enabled to shoot to such a great height, from the joints inter- posed along their stalks. Quemadmodum in arundinibus inter- nodia paribus distincta spatiis, procerttatem ipsam producunt calamorum 3 sic in majoribus numeris alia aliis aggregatagdena, welué numerorum internodia, crescentem magnitudinem "hon Note 1i.—Pages 7—9. The notation of numbers by combined strokes, which the Romans had received from their Grecian progenitors, was evidently founded in nature, and may be regarded as one of the earliest samples of a philosophical language. It is not sur- prising, therefore, that other nations, without supposing any communication, should have advanced by the same road. That the Roman system of notation was originally formed by successively combining the simple strokes, derives strong confirmation from the analogous practice of other people. It appears, from obvious inspection, that the Egyptians and the Chinese must have followed nearly the same mode. The in- ” a NOTES AND ILLUSTRATIONS. 911 scriptions on the ancient obelisks present a few numerals which are easily distinguished. Thus, the single stroke denotes one, ‘galt im gilks sotaatd POE 1 5 10 100 1000 5000 10,000 the St George’s cross ten, and the half of it five, and the same cross doubled an hundred, a zero a thousand, and this zero combined with the marks for 5 and 10, signifies five thousand and ten thousand. The substitution of capital letters for the combined strokes which they chanced most to resemble, gave uniformity indeed to the system of writing, but fatally prevented any farther im- provements in numeral notation, The only simplification which the Romans appear to have introduced, was to avoid the profuse repetition of letters, by reckoning in some cases backwards. Thus, they denoted Four by IV, Nine by IX, Forty by XL, and Ninety by XC, &c.; which signified jive and ten, abating one from each,—and forty and an hundred, dimi- nished each by ten. The series of Roman numerals is thus ex- nib: I II IIf. EVii ee WV igity VI. VII. VIII. IX, § pe oe oe 4. - 6. ie 8. 9. eS eS ET To ek ee XE, eC 10, 20; ao. 40. 50. 60. 70% 80. 90. Ch u0Co. COC CECE D..ODG~ DCC DGCGAs.cM. 100. §e 200; 300. 400. 500. 600. 700. 800. 900. Dan ddescMor 210+ doadedocloo.ooar. ecclodo: 500. 1000. 5000. 10,000, 50,000. 100,000, In illustration, it may be observed, that Cicero has this ex- pression in his fifth oration against Verres: C19 CIO CIO IDC. meaning 3600. The Romans, however, sometimes contracted or modified the forms of their numerals. This was done for the sake of expedition, chiefly in the carving of inscriptions on stones; and the abbreviated letters then used were called lapidary characters. The annexed specimen shows the princi- pal varieties : 212 NOTES AND ILLUSTRATIONS. Pes cal GR. *% x So oo M ch ma Le ee eee Ae HIM ay cb A 5000 10,000 The marks for any number could also be augmented in power one thousand times, either by inclosing them with two hooks or C’s, or by drawing a line over them. Thus, CXQ, or xX; denoted 10,000; and CLVIgM in Pliny means 156,000,000. Some- times a smaller letter was placed above another to signify their product ; thus M would express 50,000. Or the mul- tiplier was written like an exponent at the upper corner; thus IlI¢ was only another mode of signifying three hundred. In expressing very large numbers, points were sometimes inter- posed—a practice which, had it become more general, would have effected a material improvement. Thus, Pliny denotes 1,620,829 by these divided characters, XVI. XX. DCCC XXIX. But the Romans appear, in the latter ages of their Empire, to have likewise employed the small letters of the alphabet, in imitation of the numeral system of the Greeks. The letters a, b, c, d, e, f, g, h, and i, represented the nine digits, 1, 2, 3, 4, 5, 6,7, 8, and 9; the next series, k, 1, m, n, 0, p, q, r, and s, expressed 10, 20, 30, 40, 50, G0, 70, 80, and 90; and the remaining letters t, u, x, y, and z, denoted 100, 200, 300, 400, and 500. To exhibit the rest of the centuries, it was requisite to borrow capitals or other characters, and 600, 700, 800, and 900, were accordingly represented by I, V, hi andhu. But this mode of notation never obtained any degree of currency, being mostly confined to those foreign adventurers from Greece, Egypt or Chaldxa, who, by pretending to skill in ju- dicial astrology, were enabled to prey on the credulity of the wealthy Romans. In modern Europe the Roman numerals were supplied by Saxon characters. Thus, in the accounts of the Scottish ,. NOTES AND ILLUSTRATIONS. 213 Exchequer for the year 1331, the sum of L.6896 : 5: 5, stated as paid to the King of England, is thus marked, c xx vj®. viij. iiij .xvj.tj.v.S.v.d. It may be obser- ved, that, in Scotland, the contraction Gin for M,or one thou- sand, is still used, in the dates of charters and other legal in- struments. Note III.—Page 10. The Chinese had, from the earliest times, constructed a system of numerals, similar in many respects to what the Romans probably derived from their Pelasgic ancestors. It is only to be observed, that the Chinese mode of writing is the reverse of ours ; and that, beginning at the top of the leaf, they descend in parallel columns ‘to the bottom, pro- ceeding, however, from right to left, as practised by most of the Oriental nations.—Instead of the vertical lines used by the Romans, we therefore meet with horizontal ones in the Chinese notation. Thus, one is represented by a horizontal stroke, with a sort of barbed termination; two by a pair of such strokes; and three by as many parallel strokes ; the mark for fvur has four strokes, with a sort of flourish; three horizontal strokes, with two vertical ones, form the mark for foes and the other symbols exhibit the successive strokes ab- breviated, as far as nine. Ten is figured by a horizontal stroke, crossed with a vertical score, to show that the first rank of the Binary Scale was completed ; an hundred is signified by two vertical scores, connected by three short horizontal lines; a thousand is represented by a sort of double cross ; and the other ranks, ascending to an hundred millions, have the same marks successively compounded. The annexed figures are very exactly copied from the impressions given in Dr Marshman’s Elements of the Chinese Grammar, a work printed with me- tallic types, instead of the ordinary wooden blocks, at the Bap- tist Missionary Press at Serampore in 1814. 214 NOTES AND ILLUSTRATIONS. 1— Yuh. 10 ot Shih. eee ° ‘ 100 Fy Pith. oD as , ioe i 1000 ate Ts’hyen. PU se . Ne 10,000 a Wan. 6 5 Nevo. sian, Bit ¥ 100,000 APY Ee. +f ie z ak: 6 Lye. Ms a 1,000,000 Ik Chao. uP, ce] ¥ De TS hih. fo) 8 WAN Puh. fa OS ° 100,000,000 Ls 7 9 wh Kyéu. rare DP, hye (*] 10,000,000 JR, King. The numbers eleven, twelve, &c. are represented by putting the several marks for one, two, &c. the excesses above ten, im- mediately below its symbol. But, to denote twenty, thirty, &c. the marks of the multiples two, three, &c. are placed above the symbol for ten. This distinction is pursued through all the other cases. Thus, the.marks for two, three, &c. placed over the symbols of an hundred or of a thousand, signify so many hundreds or thousands.—The character for ten thousand, called wan, appears to have been the highest known at an early pe- riod of the Chinese history, since, in the popular Janguage at present, it is equivalent to all. But the Greeks themselves advanced no farther. In China, wan, wan, signifies ten thou- | sand times ten thousand, or an hundred millions; though there is also a distinct character for this very large number. In the Eastern strain of hyperbole, the phrase wan, wan, far out-doing a thousand years, the measure of Spanish loyalty, is the usual shout of Long Live the Emperor! The Chinese character of the cross for ten, and the — NOTES AND ILLUSTRATIONS, 215 chad for a million, though not of the greatest antiquity, is yet as old as the time of Confucius. The characters for ten, and for an hundred, millions, are not found in their oldest books, but occur in the Imperial Dictionary. Such is the very complete but intricate system of Chinese numerals. It has been constantly used, from the remotest times, in all the historical, moral and philosophical composi- tions of that singular people. The ordinary symbols for words, or rather things, are, in their writings, gene- ally blended with skill among those characters. But the Chinese merchants and traders have trans- formed this system of notation into another, which is more concise, and better adapted for the details of business. The changes made on the elementary characters, it will be seen, are not very material. The one, two and three are represented by perpendicular strokes ; the symbols for four and five are altered: six is denoted by a short score above an horizontal stroke, as if to signify that five, the half of the index of the scale, had been counted over ; seven and eight are expressed by the addition of one and two horizontal lines; and the mark for nine is composed of that for six, or per- haps at first a variety of five, joined to that of four. Jivcy lJ 2 | | 21 l twelve, 20 To represent eleven, twelve ry : of &c. in this mode, a single stroke is placed on the left I-90 several additions of one, two, &c. annexed on the right. > -f- 40 oa om From twenty to an hundred, the signs of the multiples are 4 -f 50 4 —— 51 prefixed to the mark for ten. - fs \ 216 NOTES AND ILLUSTRATIONS. The same method is pursued through the hundreds, the marks of the several multiples being always placed on the left hand before the con- tracted symbol of pith, i ion or anhundred. The ad- ditions are made on the " # ue W right, with a small ci- e — pher or circle (0), called fi Fi gooe % os ling, when necessary, to a) separate the place of yg ‘700 ass | 739 units. The distinction between two hundred and three, and five hundred and thirty, deserves to be parti- cularly remarked. A similar process extends to the notation of thousands ; but, for ten thousand, the character wan is abbreviated. As a specimen of their combination, we se- > rican mult EEX o lect the following com- 6 0 4 Hy plex expression, which A+ PF A denotes 543,475,003. The same number would be thus represented in the regular system of Chinese no- tation: Where the first Ef” if yy = : column on the right AY i. hand presents the marks =~ tt ) | ENE. #B for fifty and four, with the interjacent charac- —- ter wan, or ten thau- Ba BB py sand ; the next column to the left has the several marks for a thousand, three, and an hundred; the middle column exhibits the symbols of forty and of seven; the adjacent column repeats the character wan, or ten thousand, and then presents those for five and a thou- sand ; and the last column has the symbol ding, or the rest, which fills up the blank, with the mark for three. The last expression seems abundantly complicated, and yet NOTES AND ILLUSTRATIONS, OTT a it is unquestionably simpler and clearer than the correspond- ing notation with Roman numerals. From such an intricate example, the imperfection of the Roman system will appear the more striking. The abbreviated process of the Chinese traders was proba- bly suggested by some communication with India, where the admirable system’ of denary notation has, from remote ages, been understood and practised. The adoption of a small cipher to fill the void spaces, was a most material improve- ment on the very complex character /ing, used formerly for the same purpose. About the close of the seventeenth century, the Jesuit mis- sionaries Bouvet, Gerbillon, and others, then residing at the Court of Pekin, and able mathematicians, appear to have still farther improved the numeral symbols of the Chinese traders, and reduced the whole system to a degree of simplicity and elegance of form scarcely inferior to that of our modern ¢i- phers. With these abbreviated characters they printed, at the imperial press, Vlacq’s Tables of Logarithms, extending to ten places of decimals, in a beautiful volume, of which a copy was presented by Father Gaubil on his return to Europe, about the year 1750, tothe Royal Society of London. No more than nine characters, it will be seen, ? i are wanted, the upright cross ++ for ten Be ae being a mere redundancy. The marks ~~ Uo for one, two, and three, consist of parallel —- 12 —ll strokes as before; an oblique cross x 4 XK li denotes four ;. and a sort of bisected ten ? i i signifies five. This symbol again, being 14 —-X contracted into the angular mark --, 6 L.. lo] and combined with one, two, or three Lo strokes drawn below it, represents siz, 7 — 235 i|=u1| seven, or eight ; and still more abridged ye p60 4 and annexed to the sign of four, it de- 3 notes nine. The distinction of units, ae 538 ulzS tens, hundreds, &c. is indicated by giving the strokes alter- nately an horizontal and vertical position ; while the blanks er 218 NOTES AND ILLUSTRATIONS. vacant bars are expressed by placing small zeros.—The very important collection of logarithmic tables just mentioned, was printed by the command of Kang-shi, the second Emperor of the present dynasty, a man of enlarged views, who govern- ed China with dignity and wisdom during a long course of years. This enlightened Prince was much devoted to the learn- ing of Europe, and is reported to have been so fond of cal- culation, as to have those tables abridged and printed in a smaller character, which precious volume he carried constant- ly fastened to his girdle. Note 1V.—Page 11. The oriental nations appear generally to have represented the numbers as far as one thousand, by dividing their alphabet into three distinct classes. But the Hebrew, the rudest and poorest of all written languages, having only twenty-two let- ters, could advance no farther than 400; and to exhibit 500, 600, 700, 800, and 900, it had recourse to the clumsy expe- dient of addition, by joining 400 and 100, 400 and 200, 400 and 300, 400 and 400, and 400 with 400 and 100. The Ara- bic alphabet, containing twenty-eight letters, supplied fully the three classes. It is very remarkable, that, when these let- ters are employed to signify numbers, they are written, in the customary way, from right to left ; but in adopting the pecu- liar numeral character appropriately styled Indian, the order is inverted, or proceeds from the left to the right. The Ara- bians and Persians have also another set of symbols called the Diwani to express numbers, consisting merely of disguised and contracted words. This system extends as far as 400,000, and is much used in the East for keeping of accounts. Note V.—Page 1], 12. The Greeks, to represent numbers, distinguished their al- phabet into three classes, in each of which they inserted a supplementary character. The first class exhibited the nine units, called secvedimas OF tadixas; the next class denoted the ° ete ees ee oe NOTES AND ILLUSTRATIONS. 919 series of lens, thence named d:xadixas; and the third class ex- pressed the successive hundreds, which were termed ‘vc/lovDiseces- To complete those classes, the mark ¢, called episémon, was introduced among the units after é to denote sx, and the koppa and sanpi, represented by S, 4, or %, terminated respectively the range of tens and hundreds, or expressed ninety and nine hundred. The notation of numbers, as far as one thousand, was therefore effected in this way, the artificial word cae, which contains the letters that commence each series, serving to aid the recollection of their order. a . eh Y: b. és Ss. ‘4 Z é Ts wie it a Se 4, 5. aS 8 9 fot) Bepupide U6 fy ad E: 0. alvin hy be SOW RO £0b nl SOM, 60s. | 7Oay ot, 8000. ! Oth Cen oxi itt Hetil D. pore ewhiabs io! alts way, 100. 200. 300. 400. 500. 600. 700. 800. 900. But the same letters, by having an 76¢a subscribed, were aug- mented one thousand times. ‘Thus, &, 2, Y &c. denoted 1000, cee 2000, 3000, &c. or the ysrsadinae; ¢, x, A, &C. expressed 10,000, : / / / 20,000, 30,000, &c. or the dexadas ab sce Ounces ; and 2, 0, T, &C. re- TO Pili} presented 100,000, 200,000, 300,000, &c. The values of the se- veral characters could likewise be augmented ten thousandtimes, by placing under them the initial letter Mm of the word Mute. Thus, c signified one million. Another mode of representing large numbers was to superscribe repeated dots. Thus « ex- pressed fen thousand, or the commencement of the series of myriads, Or ugar] acdexocs amree; and a denoted one hundred mil- lions, beginning the pevgsovladinas dvrras, or the successive squares of the former set. As an example, the papiner 3,280, 196,529 would have been written by the Greeks, a. 2. n 1 Oo. Qe x. and read pugiadinay Ouray AG. mueladinay mr Aly nie, ok camiosAsoy Wivlaexoric enors SVG. 7 220 NOTES AND ILLUSTRATIONS ? Such is the beautiful system of Greek numerals, so vastly superior in clearness and simplicity to the Roman combina- tion of strokes. It was even tolerably fitted as an instrument of calculation. Hence the Greeks early laid aside the use of the abacus ; while the Romans, who never showed any taste for science, were, from the total inaptitude of their numerical symbols, obliged at all times to practise the same laborious manipulation. | Allusions to the Grecian mode of notation occur frequently in the ancient classics. Hence the point of the following epi- gram : ‘EZ pcs pe0x,0o4s ixcevalolas. os O& peel aevlas Vecepepeacs daxveeveer ZHOL Avyscs Peclois. The meaning of the passage is, that the space from sun- rise to noon, or the seventh hour, which was consequently de- noted by the letter zé#a, being consumed in labour, the rest of the day may be fairly devoted to relaxation. Note VI.—Page 97. In the early ages of the Republic, the private Romans were accustomed to register their time, by casting every day a /apillus, or little pebble into an urn. If the day closed happily, a white pebble was chosen; but if they deemed it unfortunate, they selected a black one. Hunc, Macrine, diem numera meliore lapiillo, Qui tibi labentes apponit candidus annos. Pers. Sat. Il. 1, 2. The Abacus, or Tabula Logistica, with its furniture, is fre- quently mentioned in the Classics. Quo pueri magnis e centurionibus orti, Levo suspensi loculos tabulamque lacerto. Hor. Sat. I. vi. 73. For the purpose of elementary education, this table or board was strewed with sand. Nec qui abaco numeros, et secto in pulvere metas Scit risisse vafer. Pers, Sat. I. 152. The sand used was, according to Martianus Capella, of a sea-green colour : . oo oe NOTES AND ILLUSTRATIONS. Za nw = Sic abacum perstare jubet, sic tegmine glauco Pandere pulyereum formarum ductibus equor. Lib. vil. De Arithmetica. The Abacus appears to have continued in familiar use among the modern nations of Europe, even till a recent pe- riod. The counters or pebbles were, from the corruption of the word algorithm, called in England augrim, cr awgrym, stones. Thus, in Chaucer’s lively description of the chamber of Clerk Nicholas : His almageste and bokes grete and smale, His astrelabre, longing for his art, His augrim stones, layen faire apart On shelves couched at his beddes head. Miller's Tale, v. 22-+24. Note VII.—Page 97. It is observed in the text, that the Computing Tadle, or Swan- Pan of the Chinesé, nearly resembles the Roman Aba- cus. It consists of a small oblong board surrounded bya high ledge, and parted downwards near the left side by a similar ledge. It is then divided horizontally | by ten smooth and slender reds of bam- boo, on which are strung two small balls of ivory or bone in the narrow compartment, and five such balls in the wider compartment; each of the latter on the several bars denoting one, and each of the former, for the sake of expedition, expressing five. The pro- gressive bars, descending after the Chinese manner of writing, have their values increased tenfold at each step. The arrangement here figured will hence signify, reckoning downwards, 5,804,712,063.. The Swan-Pan ad- vances the length of ten billions, and therefore a thousand times farther than the Roman Abacus, But the capital im- i Ten STR I Oo € Nii! : iH ee 292 NOTES AND ILLUSTRATIONS. provement which the Chinese had made, was, by commencing the units from any particular bar, to represent the decimal subdivision on the same instrument. Yet this most useful ex- tension of the denary scale, however obvious it may now ap- pear, was unknown in Europe before the time of Stevinus. Note VIII.—Page 97. The famous Dr Saunderson, professor of mathematics in the University of Cambridge, who had the misfortune to be totally deprived of sight by small-pox when only twelve months old, contrived a very ingenious kind of Abacus, well adapted to his forlorn case, and requiring comparatively few implements for its operations. It consisted of a smooth thin board, rather more than a foot square, divided by equidistant vertical and horizontal lines about the tenth part of an inch apart, and ha- ving their intersections perforated with holes. Conceiving the surface to be crossed by belts composed of double lines, and consequently distinguished into a multitude of quartered squares, he assigned a digit to each of these clusters, by planting a Jarge headed pin in the central hole, and com- bining with it another small headed one, stuck into the various holes round the margin. The large pin placed 9 2° 35 alone in the centre of the compound square ex- »—¢—» pressed merely the 0, or zero; and the small pin 8 ti 4 being substituted, signified 1. The rest of the pro- fede cedure was quiteregular. Onreplacingthelarge 7 6 5 pin, a small one stuck in the hole directly above it denoted 2; in the hole on the right corner, 3; in the hole below this, 4; in the next corner, 5 ;—thus, following the exterior circuit, with 7 and 8, till 9 is signified by placing the variable pin in the upper and left corner. In other respects, the board had precisely the same con- struction as the Abacus. The position of each central pin de- termined the value of its digit in reference to units, tens, hun- dreds, thousands, &c. But the better to distinguish those belts or double bars, the spaces were marked by notches along the outer edge of the board. A large stock of pins with their points cut off, was kept ready for use in two separate boxes. a ee NOTES AND ILLUSTRATIONS. 923 for calculation will be clearly understood from the inspection of the an- nexed diagram, in which the numbers 7052,4608, and 5190, are denoted, running from left to right, and reckoning from the top. Dr Saunderson is said to have acquired, fi om assiduous prac- tice, great facility and quickness in performing arithmetical operations. Note IX.—Page 100, 101. The mode of signifying numbers by different inflexions of the fingers, is of very great antiquity. The symbols for one and for seven are nearly the same, only the little finger is ex- tended a joint farther in the latter. Zhree and eight are dis- tinguished likewise, by the different extension of the two last fingers of the hand. Similar differences may be perceived in the expressions of other numbers. Those digital symbols were quite familiar to the ancients. According to Pliny, the image of Janus, or the Sun, was for- merly moulded with the fingers bent, to signify the 365 days of the solar year. Janus geminus a Numa rege dicatus, qui pacis bellique argumento colitur, digitis ita figuratis, ut trecen- torum sexaginta quingue dierum nota per significationem anni temporis, et evt se Deum indicaret. Hist. Nat. Lib, xxxiv. 7. In allusion to the circumstance that hundreds came to be expressed on the right hand, many passages occur in the Classics. In digitos rem redire, in digitos mittere, in digitis i) LE es expressions which occur in Cicero. Juvenal thus describes the aged Nestor : Rex Pylius (magno si quicquam credis Homero) Exemplum vite fuit a cornice secunde. Felix nimirum, qui tot per secula mortem Distulit, atque suos jam dexéra computat annos. Sat, x, 246 2o4 NOTES AND ILLUSTRATIONS. Some commentators would even explain, from the same practice of numeration, the allegorical description of Wisdom in the Proverbs of Solomon : Length of days is in her right hand, and in her left hand riches and honour. Proy, iii. 16. The Chinese have also contrived a very neat and simple kind of digital signs for denoting numbers, greatly superior, both in precision and extent, to the method practised by the Romans. Since every finger has three joints, let the thumb- nail of the other hand touch those joints in succession, pass- ing up the one side of the finger, down the middle, and again up the other side, and it will give mine different marks, appli- cable to the Denary Scale of arrange- ment. On the little finger, those marks signify units, on the next fin- ger tens, on the mid-finger hundreds, on the index thousands, and on the thumb hundred thousands. With the combined positions of the joints of the one hand, therefore, it was easy to advance by signs as far as a mil- lion. ‘To illustrate more fully this ingenious practice, I have here co- pied, from a Chinese elementary treatise of education, the figure of a hand, noted at the several joints of each finger, by characters along the inside, corresponding to 1, 2, and 3; down the middle, by those answering to 4, 5, and 6; and again up the outside, by characters expressing 7, 8, and 9. The length of the projecting nails betokens it to be the hand of one of the literary or higher classes in society. The merchants of China are accustomed, it is said, to conclude bargains with each other by help of those signs; and often, prompted by sel- fish or fraudulent views, conceal the pantomime from the know- ledge of bystanders, by only seeming to seize the hand with a hearty grasp. — < ‘ee NOTES AND ILLUSTRATIONS, 225 Note X.—Pages 106—110. I am sorry, after all tlre pains I have taken, at not being able to come to a more definite conclusion relative te the origin of our present numeral characters. There is strong ground to suspect that the Hindus obtained the knowledge of those figures from Upper Asia or perhaps Tartary. The humble attainments of this people are not entitled to claim any very high antiquity. The expedition of Alexander the Great into the East, opened a channel of learned intercourse ; but, according to the testi- mony of his Generals, Megasthenes and Nearchus, the people of India were at that period entirely ignorant of letters, nor had they acquired any skill in arithmetic about the time of Arrian and Philostratus. The Sanscrit alphabet is asserted by Anquetil to have been formerly distributed into three classes, which, as in other languages, were employed to de- note the successive ranks of units, tens, and hundreds; and Croze contends, that the followers of Zoroaster on the Malabar coast continued long afterwards to represent numbers by means of letters. This application of the alphabet has no doubt, for some ages, given place among the Hindus to the simpler and more perfect system of digital characters. At what epocli such a mighty change was effected, it would be difficult to conjecture. The most ancient monument of the Sanscrit nu- merals at present believed to exist, is a Royal Grant of Tand engraved on a large folding plate of copper discovered in the ruins of Mongueer, which has been described in the first volume of the. Asiatic Researches by the learned Dr Wilkins ; and its date, the 33d of Sombot, referred by him to the 23d year before the Christian era. This curious document may be genuine; but the Brahmins of India, like their brethren the monks of Europe, are well known to have been strongly ad- dicted to the pious fraud of forging charters and other deeds favourable to the interests of their order. The oldest treatise on arithmetic possessed by the Hindus, the Lilavati, remounts no higher than the eleventh century of our era. This famous composition, to which the vanity and ignorance of that people claim a divine original, is but a very Q 226 NOTES AND ILLUSTRATIONS. poor performance, containing merely a few scanty precepts couched in obscure memorial verses. The examples annex- ed to those rules, often written probably by later hands in the margin, are generally trifling and ill-chosen. Indeed, the Lilavati exhibits nothing that deserves the slightest no- tice, except the additions made by its Persian commentators. The Hindus had not the sagacity to perceive the various advantages to be derived from the denary notation. They re- mained entirely ignorant of the use of decimal fractions, with which their acute neighbours, the Chinese, have been fami- liarly acquainted from the remotest ages. Their numerical operations are unnecessarily complicated, following closely the procedure which the application of an alphabet had obliged the Greeks to employ. It would seem that, while the Hin- dus communicated their numerals to the Arabians, they were glad, in return, to accept the mode of calculation adapted to a very different and inferior system of notation. Note XI.—Page 111. I strongly suspect that the person styled Leonard of Pisa or Fibonacci, lived two hundred years later than the period as- signed in the text. Fabricius places him at the beginning of the fifteenth century. But Cossali, in a ponderous work, en- titled, Origine, trasporio in Italia, é primi progresst in essa dell? Algebra, and printed at Parma in 1797, maintains his antiqui- ty in a triumphant tone, and discourses with more than Italian prolixity on his merits. Still, however, the claims urged for Fibonacci to such an early date, appear to rest on very slen- der authority. Targioni Tozzetti, about sixty years ago, dis- covered, in the celebrated Magliabecchi library, a manuscript with this inscription: Incipet Liber Abbact, compositus a Leonar- do, filio Bonaccz, Pisano, in anno 1202. Some time afterwards, Zaccaria found, in the Ambrosian library, another enlarged manuscript by the same author, treating of mensuration or geo- desia, and bearing date 1220.. Now, we are left merely to guess at the age of these copies, nor can the dates attached to them be considered as fixing any thing but the opinion of the = NOTES AND ILLUSTRATIONS. 927 transcribers. Besides, in the older forms of the digits, the character of 4 very nearly resembled that of 2. I am inclin- ed, therefore, to believe, that both those tracts of the son of Bonacci were only the same wark, and ought to be referred to the year 1420. This view of the matter would reconcile the various discordant facts. All the early treatises of Arithmetic in Europe, being formed after Arabian models, had common- ly subjoined to them some portions of Algebra and of Practical Geometry. But is it not in the highest degree improbable that, if Leonard had once taught his townsmen of Pisa the use of the denary numerals, an art so useful and so simple could have been ever lost? No certain traces of the digital arith- metic are found among the Christians, before the close of the fourteenth century; and from this epoch, the knowledge of it, diffused by commercial intercourse, appears to have been soon conveyed from Italy to France, Germany and England. Trea- tises on the subject of calculation were now composed in dif- ferent parts of Europe, and the noble art of printing came most opportunely to multiply their circulation and perpetuate their influence. Note XII.—Page 127: Pythagoras brought from the East a passion for the mystical _ properties of numbers, under the veil of which he probably con- cealed some of his secret or esoteric doctrines. He regarded Numbers as of divine origin, the fountain of existence, and the model and archetype of all things. He divided them into 4 variety of different classes, to each of which were assigned dis- tinct properties. They were prime or composite, perfect or im- perfect, redundant,or deficient, plane or solid; they were tri- angular, square, cubic, or pyramidal. Even numbers were held by that visionary philosopher as feminine, and allied to earth ; but the odd numbers were considered by him as endued with masculine virtue, and partaking of the celestial nature. Unit, or monad, was held as the most eminently sacred, as the parent of scientific numbers. Two, or the duad, was 998 NOTES AND ILLUSTRATIONS. viewed as the associate of the monad, and the mother of the elements, and the recipient of all things material; and three, or the triad, was regarded as perfect, being the first of the masculine numbers, comprehending the beginning, middle, and end, and hence fitted to regulate by its combinations the repetition of prayers and libations. It was the source of love and symphony, the fountain of energy and intelligence, the director of music, geometry, and astronomy. As the monad represented the Divinity, or the Creative Power, so the duad was the image of Matter; and the ériad, resulting from their mutual conjunction, became the emblem of Ideal Forms. But four, or the tetrad, was the number which Pythagoras affected to venerate the most. It is a square, and contains within itself all the musical proportions, and exhibits by sum- mation all the digits as far as ten, the root of the universal scale of numeration ; it marks the seasons, the elements, and the successive ages of man; and it likewise represents the car- dinal virtues, and the opposite vices. The ancient division of mathematical science into Arithmetic, Geometry, Astronomy, and Music, was four-fold, and the course was therefore term- ed a tetractys, or guaternion. Hence Dr Barrow would ex- plain the oath familiar to the disciples of Pythagoras: “I swear by him who communicated the Tetractys.”” Five, or the pentad, being composed of the first male and female numbers, was styled the number of the world. Re- peated any how by an odd multiple, it always re-appeared ; and it marked the animal senses, and the zones of the globe. Six, or the hexad, being composed of its several factors, was reckoned perfect and analogical. It was likewise valued, as indicating the sides of the cube, and as entering into the com- position of other important numbers. It was deemed harmo- nious, kind and nuptial. The third power of 6, or 216, was conceived to indicate the number of years that constitute the period of metempsychosis. Seven, or the heptad, formed from the junction of the triad with the tetrad, has been celebrated in every age. Being un- productive, it was dedicated to the virgin Minerva, though Pe NOTES AND ILLUSTRATIONS. 239 possessed of a masculine character. It marked the series of the lunar phases, the number of the planets, and seemed to modify and pervade all nature. It was called the horn of Amalthza, and reckoned the guardian and director of the universe. Eight, or the octad, being the first cube that occurred, was dedicated to Cybelé, the mother of the Gods, whose image in the remotest times was only a cubical block of stone. From its evenly composition, it was termed Justice, and made to sig- nify the highest or inerratic sphere. Nine, or the ennead, was esteemed as the square of the triad. It denotes the number of the Muses, and, being the last of the series of digits, and terminating the tones of music, it was in- scribed to Mars. Sometimes it received the appellation of Horizon, because, like the spreading ocean, it seemed to flow round the other numbers within the Decad: For the same reason, it was also called Terpsichore, enlivening the produc- tive principles in the circle of the dance. Ten, or the decad, from the important office which it per- forms in numeration, was, however, the most celebrated for its properties. Having completed the cycle, and begun a new series of numbers, it was aptly styled apocatastasic or periodic, and therefore dedicated to the double-faced Janus. It had likewise the epithet of Atlas, the unwearied supporter of the world. The cube of the triad, or the number twenty-seven, express~ ing the time of the moon’s periodic revolution, was supposed to signify the power of the lunar circle. The quaternion of celestial numbers, one, three, five, and seven, joined to that of the terrestrial numbers, ‘wo, four, six, and eight, compose the number thirty-six, the square of the first perfect number s7zr, and the symbol of the universe, distinguished by wonderful properties. But it would be endless to recount all the visions of the Pythagorean school; nor should we stop to notice such fan- cies, if, by a perpetual descent, the dreams of ancient Philoso- phers had not, in the actual state of society, still tinctured our language, and mingled themselves with the various institutions 230 NOTES AND ILLUSTRATIONS. of civillife. The mystical properties of numbers, originally nur- sed in the sombre imagination of the- Egyptians, were eager- ly embraced by the Jewish Cabbalistic writers, and afterwards implicitly adopted by the Fathers of the Christian Church. But those fancies maintained an ascendancy in public belief until a very late period, nor were the Reformers themselves exempt from their influence. Luther, whose vigorous mind was yet deep- ly tinctured with the credulity of his age, was accustomed to venerate certain numbers with a species of idolatry. Peter Bungus, canon of Bergamot, published, in 1585, athick quar- to, De mysticis numerorum significationibus, chiefly with a view to explain some passages in the Old and New Testament. The famous number of the Beast, 666, which has so often tortured the ingenuity of the expounders of the Apocalypse, is regarded by some Divines as of Egyptian descent, the archetype of the three monads, and combining the genial and siderial powers ; being indeed only the sum of all the terms of the magic square of 6, the first of the perfect numbers, and dedicated to the Sun. But we still see the predilection for Luther’s favourite num- ber, seven, strongly marked in the customary term of appren- ticeships, in the period acquired for obtaining academical de- grees, and in the legal age of majority. Note XJII.—Page 128. The Chinese appear, from the remotest epochs of their empire, to have entertained the same admiration of the mys- tical properties of numbers that Pythagoras imported from the East. Distinguishing numbers into even and odd, they con- sidered the former as terrestrial, and partaking of the feminine principle Yang ; while they regarded the latter as of celestial extraction, and endued with the masculine principle Y. The even numbers were represented by small black circles, and the odd ones by similar white circles, variously disposed and connected by straight lines. The sum of the five even num- bers, /wo, four, six, eight, and ten, being thirty, was called the number of the Earth; but the sum of the five odd numbers one, three, five, seven, and nine, or twenty-five, being the square NOTES AND ILLUSTRATIONS. 231 of five, was styled the number of Heaven. The nine digits were likewise | grouped in two eo OPP 2 ot different ways, termed the Lo- chou, and the Ho-tou. The former expres- a sion signifies the Book of the river Lo, or what the Great Yu saw “> delineated on the back of the mysterious tortoise which rose out of that river: It is here represented. Nine was reckoned the head, and one the tail of the tortoise ; three,and seven were considered as its left and right shoulders, and four, and two, eight and six, were viewed as the fore and the hind feet. The number five, which represented the heart, was also the emblem of Heaven. It need scarcely be obser- ved, that this group of numbers is nothing but the common magic-sqguare of the nine digits, each row of which makes up Sifteen. As the Zo-chou had the figure of a square, so the Ho-tou had that of a cross. It is what the Emperor Fou-hi observed on the body of the O-0-0-0-0-0-0 e-6 _horse-dragon, which he saw b > & a> oO @-6-0-@-6-© .e) O-O spring out of the river Ho. The central number was ten, which, it is re- marked by the . commentators, terminates all the operations on numbers. @- 232 NOTES AND ILLUSTRATIONS. Note XIV.—Page 132. From the data given by Ptolemy, the ratio of the diameter to the circumference of a circle may be derived somewhat differently. Since the length of the arc of one degree, con- sidered as equal to its chord, is 1.2.50” in sexagesimal parts of the radius ; the circumference of a circle, whose diameter is unit, will be denoted by 3.8’4, the triple of that measure. Wherefore, the diameter is to the circumference of a circle, as 1 to 33, or as 60 to 1884, that is, as 120 to 377. If, from these numbers again, the corresponding terms 7 and 22 of the Ar- chimedian approximation be respectively subtracted, there will remain the ratio of 113 to 355, which is the expression assign- ed by Metius. It has been very generally supposed, that no nearer approxi- mation to the ratio of the circumference to the diameter of a circle, than that of Archimedes, was known to the Mathema- ticians of Europe before the time of Vieta. This opinion, however, is unquestionably erroneous. Purbach, the great restorer of science, who flourished more than one hundred years before that period, expressly says, in his introduction to a compend of Ptolemy’s Almagest, which he bad been able to study only through the medium of a corrupt Latin version from the Arabic, that the Greek astronomers made the cir- cumference of a circle to consist of 377 degrees, of which the diameter contains 150 (120), having deduced this esti- mate from the side of the inscribed decagon, which amounts to 27 (37) degrees and 4 minutes nearly. But no such pas- sage occurs in Ptolemy’s preliminary book on chords. It is true, that from the side of the inscribed decagon, or 37°4/55", he computed the chord of 24°, or the difference between 36° and 60°, and afterwards derived the chords of the successive bisections of this arc, till he found that of 14° to be 1°34/15", which gives, consequently, in the same pro- portion, 1°2’50" for the chord of 1°. The Arabian commen- tators must, therefore, have spontaneously drawn the inference already stated, that the diameter of a circle, being divided into 120 parts, will include $77. They had also given a diffe- NOTES AND ILLUSTRATIONS. 232 rent form to the same conclusion ; for Purbach subjoins, that some authors make the diameter to be to the circumference in the ratio of 20000 to 62832, the same evidently as 1 to the decimal expression 3.1416. He further adds, that the Indians allege the ratio to be the same as that of 1 to the square root of 10, if such a number were capable of extraction. This differs widely, however, from the truth, though the famous philologer Joseph Scaliger afterwards proposed it, with most presumptive dogmatism, as an absolute quadrature of the circle. It may be remarked, that the sines near the beginning of the quadrant, calculated to seven places of figures by Purbach’s disciple and successor, Regiomontanus, would afford a nearer approximation. Thus, the sine of half a degree expressed sexa- gesimally and centesimally, is .62359, which, being multiplied by 6, and marked decimally, gives 3.14154, for the circumfe- rence of a circle whose diameter is I. Note XV.—Page 174. It should be remarked, that the ancient Greeks distin- guished the Theory from the Practice of Arithmetic, by se- parate names. The term Arithmetic itself was restricted by them to the science which treats of the nature and general properties of numbers ; while the appellation Logistic was ap- propriated to the collection of rules framed to direct and fa- cilitate the common operations of calculation. The ancient systems of Arithmetic, accordingly, from the books of Euclid to the numerical treatise of Boéthius, are merely speculative, and often abound with fanciful analogies. The Logistic or Practical Arithmetic of the Greeks, though unavoidably encumbered by the triplicate form of their nume- ral notation, had, by successive improvements, attained to a remarkable degree of perfection. A few select examples will explain the mode of operation. Suppose the number 862 were to be multiplied by 523, the process would have been thus con- ducted by the Greek Arithmeticians, the several steps being, for the sake of clearness, marked likewise in modern numerals, 234 NOTES AND ILLUSTRATIONS. Here, in the first range, ¢ multiplied into #, being the same as 40, the product of § and 5 augmented ten thousand w & 6 * © 962 times, is consequently de- Puny 523 noted by x or &; mul- wb 7 ee rt aS tiplied into % gives the as 2 aL same result as 30 or 5 1G.'s Hats times 6 increased a thou- “Oo wb 1B es sand fold, and therefore ? 4. i 3 v in te expressed by yor %;and 180 ? multiplied into £, evi- Say 6 dently makes a thousand bs 2 than beds ele weeny eae ore. In the second range » multiplied into # gives the same product as 8 repeated twice, and then augmented a thousand times, or denoted by 4 53 * ! multiplied to Z is equivalent to 6 repeated twice, and after- wards increased an hundred fold, or expressed by # ¢; and x multiplied by 6 gives 40, the value of « In the third range, y multiplied into # produces 2400, which is denoted by 8 v; 4 y multiplied into makes 180 or ex; and, lastly, y multiplied into @ gives s, the symbol of G. Collecting now the scattered members into one sum, the result of the multiplication of » 22 by Quy is we wx ¢, or 450, 826. But the notation of the Greeks was not at all adapted to the descending scale. ‘They had no decimals, and to express vul- gar fractions they took two different methods. 1. If the nu- merator happened to be unit, the denominator was indicated by an accent. Thus, 0’ signified as and Sahoo: but one-half, being of more frequent occurrence, was ‘denoted by a particu- lar character, varying in its form, C,.<, (’, or KX. 2. In other cases, it was the practice of the Greeks to write the de- nominator, as we do an exponent, a little above the numera- tor, and towards the right hand: Thus, Br intimated ,%, and 8x TOPRX rove — NOTES AND ILLUSTRATIONS. 235 In illustration of the management of fractions, I shall take an example which is rather complicated, from the commen- tary which Eutocius of Ascalon wrote, about the third cen- tury of our zra, on the Tract of Archimedes concerning the quadrature of the circle. Let it be required to find the square of ohn fie, or 1838.2,. & g RY OR. MM M *%Bds v 0 ee. 7 BO% o Mm !? 7 SUE TMG M or, 7 A 4 M arRm be arn be a) a 4 Bi % yp P) gié jp 2d st & 0 S ale ®W by eee % y r) gm xO. si% Ss gi& ToL Pm “ov oO Cie oregpne aov pe ACpxe 18382. 1838.9, PEE ieee pee 818.2, ss “ee 8 @ 64... 64... 6 54,5 Ble cccve Poh s alee & 33812514), ae Tis a: 8x47 rar or, 3,381,2523,7- This complex process needs no explication: It is only to be observed, that to multiply the several integers by the fraction -2_, is the same thing as to multiply them first by 9, and then divide the product by 11. 236 NOTES AND ILLUSTRATIONS. It may be proper, likewise, to give an example of the mul- tiplication of sexagesimals. For this purpose, I shall borrow a question proposed by Theon, to find the square of the side of a regular decagon inscribed in a circle, or the chord of 36° which, according to Ptolemy’s computation, measured in sexa- gesimal parts of the radius 37° 4/55”. The multiplication is thus effected : 0G, ; VE 37° ul BBY a 37 4 55 ard eum gai 1369’ 1487 2035” en IS Ox 148 16 220iv Bre ox 2035 220 yne 3025v arog OO 60 & HE 1375’ 47 14 iow Q5v The square now found is that of the greater segment of the radius divided into extreme and mean ratio, and consequently the same as the rectangle under the radius and the smaller segment, or 22° 55/ 5", But the product of 60° into 22° 55/ 5” is 1375’, differing only by a very minute defect from the ac- tual result. From these operations, it is not difficult to conccive how the Greeks would proceed in other cases, such as Division and the Extraction of the Square Root. Note XVI.—Page 175. From the Greek word {Qos, a pebble, came the verb YnPiZev, to calculate, and likewise the substantive noun YyPoPogia, de- noting calculation. Corresponding to dYngeos, the Romans had calculus, a little chalk stone, or counter. Hence, also, various phrases used in the Classics: Hic calculus accedat—ponere cal- culum—calculum detrahere—decedere calculum—plurium calculis wincitur—calculos movere. The following sentences are extract- ed from Cicero: ‘* Quare nunc saltem ad illos calculous rever- tamur, quos tum abjecimus; ut non solum gloriosis consiliis utamur, sed etiam paulo salubrioribus,”’—- Hoc quidem est NOTES AND ILLUSTRATIONS. 237 nimis exigue et exiliter ad calculos vocare amicitiam, ut par sit ratio datorum et acceptorum.”’ The verb calculare, and the substantive noun calculatio, formed by derivation long afterwards, are considered as bar- barous Latinity, though they have been retained in most of the modern languages. Note XVII.—Page 184. When ciphers were first introduced into Europe, it was deem- ed necessary to prefix a short abstract of their nature and ap- plication. ‘These brief notices are often met with attached to old vellum almanacs, or inserted in the blank leaves of mis- sals, and frequently intermixed among famous prophecies, most direful prodigies, and infallible remedies for scalds and burns. In such strange company, the denary characters copied in page 126 were found, but followed by a neat explication of their use. After the present numerals had been generally adopted, it was the practice throughout Europe, to reduce the rules of Arithmetic, like these of the Latin Grammar, to memorial ver- ses. A small tract composed on that plan, in the reign of Edward VI. by Buckley of Litchfield, a fellow of the Univer- sity of Cambridge, appears at one period to have gained posses- sion of the schools and colleges of England. It bore this title,— ARITHMETICA MEMORATIVA, sive COMPENDIARIA ARITH- METICH TRACTATIO, non solum tyronibus, sed etiam veteranis, et bene exercitatis in ea arte viris, memoria juvandae gratia, ad- modum necessaria: Authore Gulielno Bucleo, Cantabrigiensi. I shall here extract a few specimens. DE NUMERATIONE. ————— Numerorum signa decem sunt, Quorum significant aliquid per se omnia, praeter Postremum, nihili quae dicitur esse figura. .Circulus hace alias, alias quoque cyphra vocatur, Que supplere locum nata est non significare. Hi characteres prima si sede locentur, Significant se simpliciter, positique secunda, 238 NOTES AND ILLUSTRATIONS. Significant decies se, quod si tertius illis Obtigerit locus, ad centum se porrigit usque Summa, locus quartus solus tibi millia fundit, Et quartum quintus decies complectitur, huneque Tantundem sextus superat quid multa sequens cum Quisque locus, soleat decies augere priorem. Ratio numeros tum scribendi, tum exprimendé. Scripturis numerum a dextris fac incipias, hinc In levam tendens, donec conscripseris omnes. Post signa minimis loca quarternaria punctis. Punctaque quot fuerint, totidem tibi millia monstrant. A leva vero numerorum expressio fiat. DE MULTIPLICATIONE, Est numerum in numerum diducere, multiplicare Et cujus ductu numerus producitur, in se Contineat toties numerum qui multiplicatum, Multiplicans quoties in se complectitur unum. Scribatur primo numerus, qui multiplicari Debeat, et recte sub eodem multiplicantem Ponito, ducatur solito mox linea more. Et numerum primum seriei multiplicantis, Multiplica in cunctos seriei multiplicande, Inferius scribens quicquid producitur, atque Si plures fuerint numeri tibi multiplicantes, Omnes in numerum deducito multiplicandum, Semper subscribens quicquid producitur, idque Recte sub numero scribatur multiplicante. Et quia quot fuerint numeri tibi multiplicantes, Productos totidem numeros quoque adessse necesse est : Idcirco hos omnes conjunge per additionem. Subscriptus numerus, productus jure vocetur, « Nam, quam querebas, solet hic producere summam, Examen. Divide productum, numerum per multiplicantem. Si nihil errasti, prodibit multiplicandus. DE DIVISIONE- Ostendit numeri quasvis divisio partes. Ponatur numerus, suprema parte secandus, Lineolasque duas ille supponito rectas, Divisor sub iis ponatur parte sinistra. Deinde vide quoties diyisor contineatur NOTES AND ILLUSTRATIONS. In supraposito numero quotiensque lovetur In spatio, mox et divisorem per eundem Multiplica, totumque quod hinc provenerit, aufer Supremo ex numero, supra ponendo relictum, Transfigens numerum de quo subtractio facta est. Si plures numeros contingat adesse secandos, Divisor dextram versus tibi promoveatur. Unam per seriem, Rursus quoque querere oporiet Divisor quoties in eo, qui dividitur sit, Et quotientem intra spatium deponere ut ante. Sic reliqua adsolvis prorsus, queecunque supersunt. Nec labor hic quicquam distat, variatve priori. Sin, qui dividitur, fuerit minor inferiori, Supremo intacto, divisor progrediatur, Et medio in spatio ponatur cyphra, modoque, Hoc facies, donec summam diviseris omnem, _ Modus scribendi residuum. Si quid restiterit postquam divisio facta est, Id supra scribi divisorem solet omne. Inter et hos numeros est linea parva trahenda, Que fractum numerum, non integrum notet esse. Rei totius brevis comprehensio. Divide, multiplica, subduc, transferque secantem. Examen, Per divisorem, quotientem multiplicabis. Producto reliquum, si quod fuit, adde, priorque Exhibit numerus, nisi te deceperit error. DE EXTRACTIONE RADICIS QUADRATE. Quadratz est numerum semel in se multiplicare. Querere radicem, numerum est exquirere, qui in se Ductus, propositam poterit producere summam. Cujus radicem numeri vis querere, scribe. Descripti alternas punctis signato figuras, Lineolasque duas illi supponito rectas. Et quia presenti similis diviso parti est, A puncto versus levam incipies operari, Querendo sub eo digitum, qui multiplicatus In se, vel totum, vel magnam tollere partem Signati puncto numeri possit digitusque Sub puncto in medio spatio scribatur, et inde In se ducatur, productum tolle supremo Ex numero, reliquum scribens, ut quando secares 39 240 NOTES AND ILLUSTRATIONS. Dupletur Quotiens, producti prima figura, Si binz fuerint, versus dextram statuatur Sub numero, punctum cui non supereminet ullum, Et reliqui numeri ponantur parte sinistra. Sic novus emergit Divisor, qui quoties si¢ In supra posito numero, qu : Quotientem Inventum in spatio sub suinctpone sequenti. Hunc primum in se, mox divisorem per eundem Multiplica, producta duo summam simul unam Efficiant, numero quz subducenda, supremo est, Et reliquum solito debes ascribere more. Dupletur rursus quicquid tibi linea duplex Suggerit, et duplum divisor erit nonus, huncque Divide per numerum suprema parte relictum, Czteraque expedias quadrando, multiplicando, Hinc subducendo, supra ponendo relictum. Quod facies donec numeros percurreris omnes. — Si semel in reliquo duplum non possit haberi, Pone cyphram in spatio, divisoremque novato. : Examen. Quadra radicem, quadrato junge relictum, Si modo quid fuerit, numerus si prodeat idem Cum primo, recte est, si non opus est iterandum. Modos colligendi minutias ex residuo. Duplo radicis numerus superadditur unus, Producto numerum mox supra scribe relictum, Lineola adjecta numeros quz separet ambos. It deserves to be mentioned, that the great Napier himself did not disdain to give, in his Rabdologia, a short and neat set of memorial verses adapted to the use of his Rods. Note XVII.—Page 208. The formation of circulating decimals affords a fine illustra- tion of that secret concatenation which binds the succession of physical events, and determines the various lengthened Cycles of the returning seasons—a principle which the ancient Stoics, and some other Philosophers, have boldly extended to the moral world : Alter erit tam Tiphys, et altera que vehat Argo Delectos heroas: erunt etiam altera bella, Atque iterum ad Trojam magnus mittetur Achilles. FHE END. ‘eo res oh a Pt : i, ‘ ; : ot UNIVERSITY OF ILLINOIS-URBANA 513.01L56P C001 THE PHILOSOPHY OF ARITHMETIC EDINBURGH FUL 3 0112 017100428 Lea sen : ' fines { , i i vse petubetetebed wh itte eh ‘ I vite ju ihe fat i be ' a oa : / s ’ rei ity hes i f f h on Chis v ; " reeds aut ' ip ots it bes Heeb ‘ ‘ ! ’ ver aly fae Rr >s) y ¥ i : ibe (we het ’ ‘ f * he) ’ Giseniteatn 4 tlpaly Hipet Aubert i ' , PORTE Ae AT ' f ' és ' \ . i pibye j ; ee neres al apraryt Vib yee fy ’ fet : pied ts sol y \ ‘ } ti . i hikes ‘ ‘ ' , ‘ f ety Pires? ‘ . ‘ . bt ‘ ol teh , t ; neve ai , a8 f ; ’ ' . ‘ . ‘ Ye ’ Aehahras *1 ’ 411 ' : phat atin ne GSthet " f ’ ' yf t ‘ bh . ' hele f , hereh Heed enee a kira eel \ 7 Bere REL vb tellebehi et Ft ' 4 fe be hey eben t tf hed { ’ ae : : ‘ ver te + A fet ‘ ' iF be he ‘ : f ‘ ’ HeCorbel spit pueniee vt te } t ' i tg ; oe 4 ye ; reine pet eprint mitt Vit siya i }} Hhothetadenat ‘ 1 beat ' ‘ ' ' ei + td bel : . ‘ } : , 4 7 - ni ia . , ’ wt i : ' 4 . Heese ‘ . § : , : bab } , ' " ' ' r ; Pras