1G6 DISCUSSION OF THE DESICCATION QUESTION. 



where the precipitation is comparatively small in amount. Great differences 

 are found to occur in the rain-fall of forest-covered areas ; so that it seems 

 hardly possible to say that any quantity of moisture is too great to allow of 

 the growth of trees, provided that the excess has an opportunity to run off 

 the surface, and does not stand upon it so as to form swamps or morasses. 

 At all events, it is a fact that forests are abundantly developed in i-egions 

 where the rain-fall exceeds 100 inches, and sometimes in those where it 

 much exceeds this amount. It is also true that there are dense forests in 

 regions where the total precipitation (in I'ain and melted snow) does not 

 much exceed twenty inches. Where the amount falls below this last-named 

 figure, forests do not thrive ; but the grasses usually do so, and often in the 

 greatest vigor and abundance. An inspection of a rain-chart of the earth, 

 and a comparison of the position of the rainless and drier areas with that of 

 the belts or tracts destitute of trees, will be sufficient to show at once that, 

 in a general way, regions where the rain-fall is deficient, or falls below twenty 

 or twenty-five inches, are those where trees are least developed ; and also 

 that a vigorous growth of grasses may be found where the precipitation is 

 considerably below twenty inches. 



It will perhaps surprise the reader to be told, as he may be with truth, 

 that certainly more than a quarter, and probably more than a third, of the 

 land surface of the earth belongs to the region in which the grasses and 

 carices, or a shrubby vegetation, constitute the natural growth, while trees 

 are almost entirely absent. Asia is the continent on which the amount of 

 treeless area is proportionally largest; but there are more than two millions 

 of square miles which may properly be classed in this division in South 

 America, and more than half that number on our own division of the 

 continent. 



Absolute deserts — that is to say, regions where no vegetation of any kind 

 covers the surface — are of very much more limited occurrence than are the 

 steppes above described. Even in localities where the cold is greatest, and 

 also in those where the heat is most intense and the atmosphere least moist, 

 some kind of vegetable life may continue to exist, provided the surface be 

 not a movable one. Thus in the Sahara, which is usually accejDted as a 

 typical desert region, there are large areas which are covered by a shrubby 

 vegetation, sparsely distributed, it is true, but not altogether absent. The 

 really desert regions are those over which movable sands form a heavy cov- 

 ering, continually shifting their position as urged forward by the driving 



