has taken place since. In some cases, these tracts compose basins
apparently scarcely at all exposed to any local influence in the way of
percolation or infiltration of water towards or from neighboring
valleys. But in such situations, apart from accidental disturbances, the
ground is growing drier and drier from year to year, springs are still
disappearing, and rivulets still diminishing in their summer supply of
water. A probable explanation of this is to be found in the rapid
drainage of the surface of cleared ground, which prevents the
subterranean natural reservoirs, whether cavities or merely strata of
bibulous earth, from filling up. How long this process is to last before
an equilibrium is reached, none can say. It may be, for years; it may
be, for centuries.

Livingstone states facts which strongly favor the supposition that a
secular desiccation is still going on in central Africa, and there is
reason to suspect that a like change is taking place in California. When
the regions where the earth is growing drier were cleared of wood, or,
indeed, whether forests ever grew there, we are unable to say, but the
change appears to have been long in progress. A similar revolution
appears to have occurred in Arabia Petraea. In many of the wadis, and
particularly in the gorges between Wadi Feiran and Wadi Esh Sheikh,
there are water-worn banks showing that, at no very remote period, the
winter floods must have risen fifty feet in channels where the growth of
acacias and tamarisks and the testimony of the Arabs concur to prove
that they have not risen six feet within the memory or tradition of the
present inhabitants. Recent travellers have discovered traces of
extensive ancient cultivation, and of the former existence of large
towns in the Tih desert, in localities where all agriculture is now
impossible for want of water. Is this drought due to the destruction of
ancient forests or to some other cause?