GROWTH AND DECLINE OP CULTURE. 163 



both fact and inference are sound, every sucli argument is a 

 step gained, while if either be unsound, a distiiict statement of 

 fact and issue is the best means of getting them corrected, or, 

 if needful, discarded altogether. A principal object of the 

 present chapter, is to bring forward a variety of instances 

 drawn from sources where indirect evidence bearing on our 

 early history is to be sought. 



As examples of evidence from language, a few cases may be 

 given. The word calculation, indicating the primitive art of 

 reckoning by pebbles or calculi, has passed on with the growth 

 of science to designate the working of problems far beyond 

 the reach of the abacus. So, though the Mexicans, when they 

 were discovered, had a high numerical system and were good 

 reckoners, the word tetl, " stone," remained as an integral part 

 of one of their sets of numerals for counting animals and 

 things; centetl "one stone," ontetl "two stone," etetl "three 

 stone," etc., meaning nothing more than one, two, three. Nor 

 is Mexico the only country where this curious phenomenon 

 occurs. The Malays say for " one " not only sa, but also sa- 

 tvatu, that is literally " one stone," and the Javans say not only 

 .sa but sawiji, that is, " one corn, or seed," and in like manner 

 the Nias language calls one and two samhua and duvibua, that 

 is, apparently, "one fruit," "two fruits."^ 



Still more notable is the Aztec term for an eclipse. The 

 idea that the sun and moon are swallowed or bitten by dragons, 

 or great dogs, or other creatures, is not only very common in 

 the Old World, but it is even found in North and South 

 America, and Polynesia.^ But there is evidence that the 

 ancient Mexicans understood the real cause of eclipses. They 

 are represented in the picture-writings by a figure of the 

 moon's disc covering part of the sun's, and this symbol, Hum- 

 boldt remarks, " proves exact notions as to the cause of 

 eclipses ; it reminds us of the allegorical dance of the Mexican 



' Crawfurd, Gr. and Die. of Malay Language ; London, 1852, vol. i. pp. Ivi. 

 Iviii. Ixvii., and see ccxviii. 



^ Jacob G-rimni, ' Deutsche Mytliologie ; ' pp. 224-5, 663. Solioolcraft, part i. 

 p. 271. Dobrizhoffer, vol. ii. p. 84. Du Tertre, Hist. Gen. des Antilles, etc.; 

 Paris, 1667, vol. ii. p. 371. Turner, ' Polynesia,' p. 531. 



^ Humboldt, Vues, pi. 56. 



M 2 



