100 AMERICAN FERN JOURNAL 
However, the seurce of supply of these non-volatile 
substances for epiphytes must be rather limited, and 
it is presumably for this reason that all epiphytes 
(so far as known to the writer) are evergreen. For 
it is obviously out of the question for an air-plant of 
any size to get enough solid “food”? to make a complete 
new set of leaves every year, as most terrestrial herbs 
and all deciduous shrubs and trees do, and consequently 
such plants have leaves (or fronds or thalli in the case 
of some cryptogams) that last more than one year. 
Or the proposition might be reversed by saying that 
plants growing in fertile soils take up so much inorganic 
matter, which is deposited in leaves and other ex- 
ternal parts by the process of transpiration, that they 
have to shed their leaves periodically to get rid of it. 
It has been suggested that air-plants get some of 
their solid nutriment from dust, which is probably true; 
but in the heart of a vast trackless forest the quantity 
of dust that falls on any one herb in the course of a 
year must be quite infinitesimal, and there is no reason 
for believing that true epiphytes are less abundant 
in such places than near highways and habitations. 
Pollen of anemophilous trees is another possible source 
of food, just as available in a wilderness as elsewhere, 
but whether it contains iron and potassium or not the 
writer is not informed. 
A much more likely source of inorganic matter for 
epiphytes of the type here discussed is the bark on 
which they grow. The bark of most trees contains 
2 to 6% of mineral matter, and as it increases little 
in thickness with the growth of the tree it must be 
continually scaling off and decaying on the outside. 
One could with a little trouble make a rough estimate 
of the amount of mineral liberated in a year by a unit 
area of bark of a given species under normal conditions. 
The terrestrial herbs and deciduous shrubs that one 
