122 
•• 
we may conceive how, in languages rich in figu~ 
rative expressions, the words Uvoy three^ and 
seven, may be applied to the ideas of a yoke 
(jugiim ), of all 'powef'ful ( trimurti of the Hin- 
doos), of enchantment, and misfortune : but is it 
possible to admit, that, when man in an uncul- 
tivated state first feels the necessity of reckoning, 
he calls four a black thing ( muykica ) ; six, har- 
vest (ta) ; and twenty, a house (gue or gueta); 
because in the arrangement of a lunar alma- 
nack, from the recurrence of the ten terms of a 
periodical series, the term four precedes by one 
day the conjunctions of the Moon ; or because the 
harvest is reaped six months after the winter 
solstice ? In all languages a certain independ- 
ence is observed between the roots which desig- 
nate the numbers, and those which express 
other objects of the natural world ; and we must 
suppose, that, wherever this independence dis- 
appears, two systems of numeration exist, one 
of which is posterior to the other ; or that the 
etymological affinities, which were presumed to 
be discovered, are only apparent, because they 
rest on figurative significations. P. Lugo, who 
wrote in 1618 , informs us indeed, that the Muys- 
cas had two modes of denoting the number 
twenty ; and that they said either gueta, house, 
or quihicha ubckihica, foot ten ; but we shall 
enter no farther into discussions foreign to the 
object of this work. What we know with cer- 
