350 Observations on the 
By the preceding generalities, it will be seen that almost all the 
groups of predatory birds are common to the old world and to the 
new. 
To recapitulate, we may state, that the only genera belonging ex- 
clusively to America, are, among the vultures, the Sarcoramphi 
and Cathartes, replacing in this quarter of the world the true vul- 
tures, which are peculiar to our own hemisphere ; among the Falco- 
nidse the small group of Caracaras, which have no analogues in the 
old world, and which differ from the vultures only in their forms, 
for their manners really connect them with^that genus. The harpy* 
and the rostramet are likewise birds exclusively American, without 
any analogues in Europe ; for none of our eagles are so strong as the 
harpy, and none of our birds so sociable as the rostrames. 
Let us now consider the birds of prey with respect to their geo- 
graphical distribution, or the number of species which the different 
localities of South America present to our notice. 
We shall divide the whole extent of country to which our obser- 
vations extend, into three zones ; and these three zones shall each 
be considered in three points of view, namely, in regard to their la- 
titude, their elevation above the level of the sea, and the nature of 
the countries which compose them. 
Our Jirst zone extends from the 11 of latitude to the 28, and 
in elevation from to 5000 French feet above the level of the sea ; 
our second, from 28 to 34 9 of latitude, and from 5000 to 11,000 feet 
in height ; our third, from 34 to 45* of latitude, and comprehends 
every elevation above 11,000 feet. 
The Jirst zone of latitude, (that which extends from the J 1 to 
the 28) besides its immense plains and forests, naturally compre- 
hends the most elevated tracks of the Andes ; it, therefore, possesses 
the greatest number of species ; for, independently of such as are 
peculiar to the warm and temperate zone, we there find the kinds 
which inhabit a lower latitude, the proportional decrease of heat, as 
we ascend the plateau of the Andes, soon rendering the temperature 
of this latitude equal to that of the most southern. Thus when we 
regard in the aggregate all the species comprised within these two 
parallels, without taking into account the zones of elevation, we find 
them amount to thirty- eight, while the total number of species ob- 
served does not exceed forty-one. According to this calculation there 
will only remain for the two other latitudinal zones, three species not 
found in the first, and peculiar to the southern plains. This propor- 
tion will at first appear enormous, but the explanations on which we 
are about to enter will make it seem less extraordinary. 
If we consider, in reference to this first zone, the species confined 
* Harpyia. f Cymindis, 
