298 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. 
All of these islands extending along the north coast of Venezuela 
not only resemble each other very much but they are also like the 
coast of the mainland. There are to be fovmd much the same species 
constituting the seashore flora, the flora of the lagoon, of the wild 
cactus-covered hills, and of the few fertile coconut valleys. This is 
well illustrated by almost any part of the north coast. Carupano is 
in a long narrow valley with arid hills on each side. Cumana is on a 
sandy plain at the foot of the hills. Guanta is in a small valley with 
the appearance of a perpetual drought on every side. La Guaira 
is on a hillside by the edge of the sea and the hill is a bro^^^l and sun- 
baked exposure although it is broken here and there by green valleys 
and by a green mountain rising above. A short way inland but 
still in the coastal region between Caracas and Valencia and about 
the Lake of Valencia trees are scarce or lacking, the mountains are 
brown and clothed only in small shrubs or in dry grass, and in the 
valley are scorching sandy plains with here and there the shade of a 
tree. 
These islands are similar to the coastal land as naturally they 
should be, having been in early times a part of the coast and yet there 
is a vast country behind the coast to which they are not at all like. 
The mountain region of the Andes, anwhere from one thousand to 
four thousand meters high, the grassy plains of the Orinoco, and the 
fores'ts to the south present features vastly different in every respect. 
Unfortunately our knowledge of their flora is very limited. Many 
plants were described as new from Humboldt's travels, but since that 
time there have been few collections and fewer plants described. It 
is known that there are many plants which are common to the rest of 
the tropics. Altogether our information is one-sided as tending to 
show the cosmopolitan rather than the characteristic plants. I have 
compiled a list of all the published names of Venezuelan plants which 
comprises some three thousand names. That some of these are 
names which may not be in good standing today cannot be denied, 
but I have at least made reasonably sure that they represent nearly 
three thousand different species. 
Out of the six hvmdred and thirty-four ^Nlargaritan plants two 
hundred and ninety-five have not been published as occurring any- 
where else in Venezuela. Inasmuch as many of these are cosmopolitan 
plants it shows not the peculiarity of the Margaritan flora but the small 
amount of work that has been done on the mainland. 
