GROWTH AND DECLINE OF CULTURE. 155 



member a commission at market by, she makes a rudimentary 

 quipu. Darius made one wben lie took a thong and tied sixty 

 knots in it, and gave it to the cliiefs of tlie lonians, tkat they 

 might untie a knot each day, till, if the knots were all undone, 

 and he had not returned, they might go back to their own 

 land.^ Such was the string on which Le Boo tied a knot for 

 each ship he met on his voyage, to keep in mind its name and 

 country, and that one on which his father, Abba Thulle, tied 

 first thirty knots, and then six more, to remember that Cap- 

 tain Wilson was to come back in thirty moons, or at least in 

 six beyond.^ 



This is so simple a device that it may, for all we know, have 

 been invented again and again, and its appearance in several 

 countries does not prove it to have been transmitted from one 

 country to another. It has been found in Asia,^ in Africa,* in 

 Mexico, among the North American Indians f but its greatest 

 development was in South America.^ The word quipu, that 

 is, "knot,^^ belongs to the language of Peru, and quipus served 

 there as the regular means of record and communication for a 

 highly-organized society. Von Tschudi describes them as con- 

 sisting of a thick main cord, with thinner cords tied on to it at 

 certain distances, in which the knots are tied. The length of 

 the quipus varies much, the main trunk being often many ells 

 long, sometimes only a single foot, the branches seldom more 

 than two feet, and usually much less. He has dug up a quipu, 

 he says, towards eight pounds in weight, a portion of which is 

 represented in the woodcut from which the accompanying (Fig. 

 15) is taken. The cords are often of various colours, each with 

 its own proper meaning; red for soldiers, yellow for gold, 

 white for silver, green for corn, and so on. This knot-writing 

 was especially suited for reckonings and statistical tables; a 



' HerocL, iv. 98. See Plin., x. 34. 



2 Keate, ' Pelew Islands ; ' London, 1788, pp. 367, 392. 



3 Erman (B. Tr.) ; London, 1848, toI. i. p. 4J2. 



* Goguet, vol. i. pp. 161, 212. Klemm, C. G., vol. i. p. 3. Bastian, vol. i. 

 p. 412. 



5 Charlevoix, vol. vi. p. 151. Long's Exp., vol. i. p. 235 (a passage which sug- 

 gests a reason for Lucina being the patroness of child-birth). Talbot, Disc, of 

 Lederer, p. 4. * Humboldt and Bonplaud, vol. iii. p. 20. 



