along midrib, reclinate.’’ Richard Evans Schultes & Joao Murga Pires 
9000,—Same locality. January 30, 1948, Richard Evans Schultes & 
Francisco Lopez 9663, 9664.—Kstado do Amazonas, Rio Uaupés, Ipan- 
oré. February 2, 1948, Itchard Evans Schultes & Francisco Lopez 9666, 
9667, 9669. 
Micrandra Lopez appears to approach most closely 
to M. glabra, a tree of savannas or caatingas in British 
and Dutch Guiana. Whereas the latter is a corpulent, 
heavily buttressed tree, the former is a slender tree de- 
void of buttresses (except in the forma described below). 
The leaves of Micrandra glabra tend, in general, to be 
somewhat smaller than those of MW. Lopez, and they 
also seem to have had in life a waxy glaucescence on the 
upper surface, a character not present in M. Lopezi. 
The fruit of Micrandra Lopezti is much smaller (83 X 28 
mm.) than that of JZ. glabra (55-60 mm. X80 mm. ), and 
the capsule of the former is perfectly rounded at the apex, 
whereas that of the latter tends to be somewhat tapering. 
The staminate flowers of Micrandra Lopezii are very 
minute, only 1.8 mm. in diameter, much smaller than 
those of M. glabra; and the pistillate flowers, while of 
about the same size, have differentiating characters in 
the disk. Micrandra Lopezu has an extremely dense and 
contracted axillary inflorescence, whereas that of J/. 
glabra is lax with rigid pedicels and is subapical. 
Micrandra Lopez is an inhabitant of the caatinga- 
forest formation of the upper Rio Negro basin. Its simi- 
larity to Micrandra glabra emphasizes once more the 
phytogeographical relationship of the upper Rio Negro 
with the Venezuelan-Guianan land mass. We might 
justifiably infer, from its relationships and the isolated 
occurrence of Micrandra Lopezti near bases of ancient 
remnant mountains, that this species represents one of 
the most ancient elements of the genus. 
Be that as it may, there is no doubt that Micrandra 
[ 206 | 
