Desert Conditions in the Trias. 909 
gradually getting smaller to leeward. The same sands may be worked 
up again and again by currents of differing velocities, and hence we 
should not expect to find the layers of even-sized grains to persist 
over very great distances. Again, the grains are not all composed of 
materials having the same relative density. Although quartz pre- 
dominates, we find mica, felspar, magnetite, zircon, and other 
minerals, and each will be affected according to its linear dimensions 
and specific gravity. 
If we consider two particles of different sizes and having the same 
terminal velocities when falling under gravity we find the resistance 
varies as the square of the linear dimensions, and the weight jointly as 
the cube of the linear dimensions and the specific gravity. This is 
only a very simple and incomplete way of stating a very complicated 
problem, but it is possible to find in terms of their relative densities 
and dimensions the sizes of the different kinds of grains which will 
come to rest together. 
We have already seen that where temporary streams flow down 
from high grounds fans of gravel are produced where they debouch 
into the plain, and along the beds of watercourses there exist layers of 
gravel, composed of water-worn stones. The distance these stones 
have travelled from their place of origin depends on the character and 
volume of the stream which brought them. It may be reckoned in 
miles or scores of miles. We pictured above (p. 511) the character of 
the deposits formed by the successive excavation and filling up of 
river valleys, and gave instances where sands with bands of pebbles 
intercalated at various horizons had resulted from this action. What 
is the effect of winds blowing over a deposit of this nature? The 
sands are carried by the air currents and form dunes, but the larger 
stones are too heavy to be moved and form a loose pavement on the 
desert floor. 
According to Mr. H. T. Ferrar, of the Geological Survey of Egypt, 
these pebble layers may be so thin that the tread of a camel may 
break through and send up puffs of the sand from below at every 
step. At other times they are more than 100 feet thick. These may 
result either from excessive rainfall and strong currents of water or 
from the successive accumulation of materials brought down by 
streams through many wet seasons, and the removal of the finer 
material during many dry seasons. In any case, we have in deserts 
great spreads of gravel brought down by rivers and reassorted by 
wind action. 
(3) Rounding and Etching.—Sands caught up by the wind and 
hurled against each other have a greater velocity and hence a 
greater impact than those carried along by moving water. In river 
and sea sands rounding does take place, but it does not approach 
the perfection seen in sands which have been subject to pro- 
longed movement in air. The constant battering of grains against 
each other not only results in rounding, but in the production of 
excessively fine splinters of sand. It is this sand dust which is 
carried to great distances by winds and tends to accumulate on the 
lee sides of desert regions. Particles too heavy to be lifted bodily in 
the air are rolled along the ground. If it-happens that the length of 
