THE HEVEA RUBBER TREE IN THE AMAZON VALLEY 47 
gueiro as vermelha. There are also intergradations between the two 
types, so that frequently it is hard to say whether a tree belongs to 
the one or to the other. These two sorts may be separate varieties, 
and the fact that true vermelha trees are rare on eastern plantations, 
if indeed they occur there at all, seems to lend weight to this idea. 
Breeding tests alone will tell the truth about these relationships. 
The names "seringera boa (good rubber tree)" and "seringera 
verdadeira (true rubber tree) " are common in this region also. They 
do not refer to the type of the tree, but rather to the type of rubber 
produced. A tree which produces strong rubber will always be called 
by one or the other of these names, but they are not applied to a 
tree which produces borracha fraca (weak rubber). 
On the Madeira River and its tributaries and in the Acre Territory 
the names used for different types of trees are the same as on the lower 
Amazon, but the basis of classification appears to be somewhat differ- 
ent. The type known as seringa branca on the lower river, that is, 
the one with white inner bark, is not common here, though it is occa- 
sionally seen along the river banks, but the name is used here to 
designate trees which have a trunk with a whitish exterior. This 
appearance is due to a development of soft light-colored cork, and 
it often happens that it is developed on one side of the tree and not 
around the whole trunk. So a tree viewed from one direction might 
be called a branca tree, but from another direction it might be 
classed as something else. This type of cork development is found 
on plantation trees in Sumatra, though it is not common and does 
not give so light a color to the trunk. It is very certain that the 
so-called branca type of the upper rivers is not a true type at all. 
The vermelha or rosa trees are so named from the color of the 
inner bark, as on the lower rivers. The preta trees are distinguished 
by an inner bark which varies in color from dark red to a very deep 
purple. This type of bark is very different from anything which 
exists on the eastern plantations. The true preta tree has also a 
very thick bark, which is soft and contains a much smaller number 
of stone cells than the vermelha, the branca, or the plantation type 
of tree. This bark contains a large number of latex vessels and is 
very full of latex. 
There is no sharp line of demarcation between the vermelha tree 
and the preta, though it is easy to distinguish typical representatives 
of the two sorts. 
Ule (4-2) considered that the branca and preta (narrow-leaved) 
types are varieties of Hevea hrasiliensis. These he called, respectively, 
"variety latifolia" and " variety angustifolia" The types can not 
with certainty be thus separated, for when the leaves of numerous 
trees of each kind are examined they are found to be not distinguish- 
able. 
LEAF CHARACTERS 
So far as the writer was able to determine, it is not possible to 
separate any of these types from the others on the basis of leaf 
characters. Huber (9, 11) has probably gone as far as anyone can 
in distinguishing types by the use of leaf characters. In some cases 
his species based on such characters seem very imperfectly defined, 
and the leaf typical of Hevea hrasiliensis is very like that of some of 
these other species. Perhaps one ought not to speak of a form of 
leaf as typical of H. hrasiliensis, because of the great range in 
variability of leaves within this species. This is shown particularly 
