All stages from young ravines, with conditions extremely xerophytic up to 
mature ravines with mesophytic vegetation are found. 
The first stage of one of these ravines is merely a shallow groove down 
the side of the bluff, floored by the bare bedrock and partially choked by 
rock fragments. The debris accumulates as a cone at the foot. In this 
stage, a ravine is extremely xerophytic. Few if any plants grow within 
them, though a few appear upon the talus at the foot. The plants which do 
appear are the same as the pioneer plants on a cliff face. 
The position of such a ravine seems to be determined almost entirely 
by the surface drainage outside the gorge. Wherever the topography 
of the surface outside the gorge causes the flood water to be dis- 
charged down the bluff, a ravine will be formed. Cleavage planes, which 
commonly determine the location of ravines in more massive rock, are 
absent. Seepage lines which often determine ravines in clay bluffs prob- 
ably have little effect here, for they often occur on cliff faces where no 
tendency to ravine formation is evident. The rapidity with which a ravine 
will grow is of course dependent upon the water supply. 
Older ravines are generally not regular in gradient, but become very 
precipitous in some parts, on account of the occurrence of occasional harder 
layers of limestone. This produces vertical faces from a few inches to six 
or eight feet in height, and these are usually wet with seepage or run-off 
from above. These ravines are usually quite deep and well shaded. Ver- 
tical faces of the kind described are commonly covered with Cladophora 
and Vaucheria. Where the water is contaminated by sewage, Oscillatoria 
replaces these. Mosses grow luxuriantly in these situations, but no liver- 
worts grow anywhere in these ravines, with the single exception of Porella, 
which is common in mesophytie situations throughout the region. The 
absence of liverworts is difficult to account for, in view of the hydro- 
mesophytie conditions which prevail in such situations. Fimbriaria and 
Pegatella were found in abundance on damp rock shelves near Thistle- 
thuaite’s Falls, north of Richmond. Aneura and Blasia were found on 
clay in the same region. Marchantia occurs on rocks in similar localities 
near this point. 
Why mosses should be so abundant and liverworts entirely absent in 
these ravines is difficult to explain. The finding of several genera of liver- . 
worts on similar rock shows that their absence is not due to the chemical 
nature of the rock. The older parts of the Aneura found on clay were stiff 
