2 6* 
Account of' the Poison Plants of 
caedron. When the angles (i . ), (z : z) of the dodecaedron will 
be known, without knowing, at the same time, its position with 
respect to the primitive, there will be two answers, for then the 
rhomboid the oblique diagonals of which correspond to db , %dr, 
may be derived in two different ways from the primitive, each 
will give a different equation, and each of these combined with 
Id. i) _ y— z 
X 
cos 
-, a set of values of the indices. The me- 
Cos § \i : z) x—y 
thod I have just explained to determine the indices of a dode- 
caedron resulting from an intermediary decrement, will be 
found very simple in practice, because logarithmic calculation 
may be used. 
The formulae contained in this and the preceding paper, are 
sufficient to find the indices of rhomboids and dodecaedrons, 
when some of their incidences are known. It remains now to 
explain in what manner their angles may be calculated when 
their indices are given, and which may, at a future time, be the 
subject of another communication. 
Art. VI.— Account of the Poison Plants of the Southern Parts 
f Brazil. Continued from p. 100. 
HI HE first historians of Brazil have spoken much of the art 
with which the Indians prepared their poisons. Piso says* they 
can at their pleasure infect the air and waters, —poison their ar- 
rows, — the clothes of their enemies, and even the fruits upon 
which they may have to feed. But, as Southey *j- sagaciously 
insinuates, it is very probable that such tales have been imagined;, 
to gratify the hatred of the oppressors against the oppressed ; 
and the latter, perhaps to make themselves be feared in their 
turn, may have sought to believe themselves the fables which 
were originally invented for the purpose of rendering them odi- 
ous. Piso sufficiently justifies this assumption, when he asserts 
that the Indians, while they made a mystery of their poisons, 
readily disclosed their antidotes. It is evident, that, if these men 
were interested in not divulging the fatal secrets which are attri- 
* Bras. 46. 
•f History of Brazil; vpk i, p. 237. 
