mn te 
as it is mainly based on studies in South Russia and Trans- 
caspia. But the distinction between steppe (grass-steppe) on 
the one hand and desert on the other, seems easier to apply 
than the one maintained by most authors, for instance 
WARMING and VAHL. Phytogeographical literature defines so 
many different kinds of steppe — Meadow-steppe, Grass-steppe, 
Vermuth-steppe, Salt-steppe, Sand-steppe, Bush-steppe and 
even such types as Orchard-steppe poor in grass, Tree-steppe, 
Steppe-forest &c. (ENGLER 1910) —- that it is hard to see 
what they have in common except that all are more or less 
xerophilous formations. 
What ScHIMPER, KrAssnow and TANFILJEW have termed 
steppe, is not an extreme xerophilous formation. The Sand- 
steppe (“die Sandpuszte”, Apamovic p. 320, WoEniG) belongs 
to the steppe type, but differs from the ‘“Sand-steppes’’ des- 
cribed from Transcaspia which are deserts. Steppes, according 
to the definition given here, do not occur in Transcaspia 
at all. 
CHAPTER 6 
Formation of the Salt-Deserts. 
Under this heading are classed localities where the plants 
grow on soil so saline that the salt crystallizes out as a layer 
covering the surface. The conditions which render this pos- 
sible are: 1) the presence of salts in solution, 2) facilities for 
the solution reaching the surface. It is essential for this that 
the underground water-table is not located at too great a 
depth. It is usual therefore to find salt-deserts in depressions. 
The constant evaporation from the surface causes the salts 
to crystallize out, and a fresh solution diffuses constantly 
from below. Where the soil is loess, this upward diffusion 
takes place rather easily on account of the capillary structure. 
“Ssor” is the name given by the natives to wet saline flats. 
They are often found among dunes and are said to be gener- 
ally flooded during winter and spring. A “Ssor’” is not a 
very pleasant sight (fig. 2). The ground is flat and white like 
