Desert Conditions in the Trias. 559d 
The above excerpts give one a good idea of what the desert salt 
lakes are like, and one or two facts common to all are worth noting. 
First, we have the clay bottom sometimes augmented in thickness by 
layers of sand which have drifted into the pool; then we note the 
constant association of salt and gypsum in the deposits. It is well 
known that concentrated brine precipitates the sulphate of lime out 
of solution. Further, we have no mention of carbonate of lime as 
one of the constituents left after evaporation, or, at least, if present 
it is not in sufficient quantity to call for special notice. Yet in the 
waters flowing from the hills towards the lakes this substance, as 
the bicarbonate of lime, must have been present, and probably 
exceeding the others in amount. The explanation is simple. The 
bicarbonate is chemically not a stable compound, and readily parts 
with the excess of carbonic acid. Hence, in passing over or through 
the soils the carbonate of lime will be precipitated and form the 
nodules and limestones described above, while the more chemically 
stable compounds will reach the lakes undiminished in quantity. 
So it happens, as we should expect, that carbonate of lime 
segregates on the land, while salt and gypsum are the chief substances 
left as residues in the desiccated pools. Dolomite in perfect rhombo- 
hedral crystals is occasionally found in association with the products 
of desert pools. ' 
Drifting sands blowing over the plains may fill up the hollows and 
leave no sign on the surface of the pools which formerly existed there. 
The lake deposits would still be there buried under the sand, and if 
a section could be made through them we should find beds composed 
mainly of sand with thin bands of clay following the limits of the lake 
and irregularly disposed at various horizons. On the margins the clay 
would show desiccation cracks and ripple-marks, the footprints of 
animals, and the remains of such forms of life as find a suitable habitat 
under such conditions. 
Mr. T. H. Holland,! referring to the deposits in the Rajputana 
Desert, states that silt beds occur, filling in hollows in the Archean 
surface. They have a general plano-convex lens shape, and are 
charged with salt, beds of gypsum, and concretionary nodules of 
carbonate of lime. 
In our Triassic rocks we get the exact counterpart of these filled-in 
desert lakes. Bands of clay occur in all the divisions, but they are 
more common in the higher beds. When the full extent of the clay 
bands can be determined, they are always half lenses in shape, with 
the convexity downwards. It is on the lower surfaces of the sand- 
stones, immediately resting on the clay, that we find casts of footprints, 
raindrops, ripple-marks, and desiccation cracks, and the clays often 
either contain pseudomorphs of rock-salt or deposits of this mineral 
and gypsum overlie the clay. The salt beds of Cheshire have these 
associations, and are now generally regarded as resulting from dried-up 
pools or lakes. Even when beds of pure salt are absent it is found 
that water from the Keuper Marls is almost invariably strongly 
impregnated with salt. It may be that where beds of pure salt are 
1 Brit. Assoc. Report, 1906, p. 575. 
