54 
BULLETIN 1422, U. S. DEPARTMENT OF AGBICULTUEE 
Table 14.- 
— Classification of the bark of Hevea trees in an estrada at 
10, Cobija, Bolivia 
Kilometer 
Type 
Number 
of trees 
Type 
Number 
of trees 
Preta 
60 
14 
12 
8 
Tan to vermelha--. _ . 
5 
Tan to preta 
15 
Tan . 
Vermelha to preta. ._ _ 
4 
Total 
118 
Eight trees were found in this lot which had a noticeable develop- 
ment of the white soft cork on the outer surface which causes a tree 
to be called a branca in this region. Some of these when examined 
proved to be preta and some vermelha, and a few had a tan-colored 
inner bark. This shows very well that the branca of the Madeira 
and Acre regions is not a true type. 
The classification in Table 14 was made in the field, and it is possible 
that a comparison of collected bark samples might have given slightly 
different results. The preta trees constitute the greater number in 
this lot, whereas the vermelhas are relatively few. These two types 
are distinct. The other classes represent slight but noticeable varia- 
tions. It may be that these merely represent variates in the preta 
and vermelha populations. From the evidence of this estrada it 
might be considered that all the trees belong to one population, but 
the fact that the preta type is not found at all on the lower Amazon 
and that the vermelha type does occur there argues against this. 
The writer agrees with Cramer (6) that the testimony of the serin- 
gueiro is untrustworthy, at least in the case of individual trees; but 
it appears likely that he is right in considering that the preta and the 
vermelha trees represent distinct types. The branca trees of the 
lower Amazon are almost certainly distinct from the other two types. 
The branca trees of the up-river country apparently are not distinct 
from the preta and the vermelha types. 
It is possible that the differences between the up-river types and 
the lower Amazon branca are due to differences in climate and site- 
Most trees of the branca group are growing under unfavorable con- 
ditions as compared with the others. It does not seem likely that 
such factors play a part in producing the differences between the preta 
and the vermelha types, since these are found growing in identical 
situations. The differences between the preta and the vermelha types 
are much less than those between either of these and the branca type 
of the lower Amazon. The possibility must be admitted that the 
former types do not represent varieties at all, but merely variations 
in one population. The thing that most militates against the proba- 
bility of this view is the fact that vermelha trees are found on the lower 
Amazon, whereas true preta trees are not. 
RELATIVE VALUE OF DIFFERENT TYPES 
The relative value of the various types of trees is difficult, almost 
impossible, to determine, because of the lack of data from trees of 
the different sorts grown under comparable conditions. The trees of 
the lower Amazon mostly are growing under unfavorable conditions 
and so can not be compared directly with the others. Up-river rub- 
ber has long been considered superior to that of the lower Amazon 
