298 PROCEEDINGS OP THE 



tendency in these particulars is to cut off the crest of 

 the destructive floods, depriving them of their power 

 to do harm. 



The presence or absence of forests undoubtedly has 

 a marked bearing, too, upon the quantity of the run- 

 off. This effect varies with a number of different 

 conditions, chief amongst which are the permeability 

 and porosity of the soil, the different habits in different 

 species of plant life in the matter of transpiration and 

 the differences in evaporation influences. The soil con- 

 ditions have already been touched upon. Retention of a 

 large part of the precipitation by the soil instead of its 

 being permitted to flow off rapidly may, and probably 

 must in many localities as, for example, in the arid re- 

 gions result in a decreased total run-off owing to 

 the probably greater increased "fly-off," as the sum 

 of the evaporation and the transpiration is sometimes 

 termed. This diminution in the total quantity is, 

 however, considerably more than offset by the advan- 

 tages incident to a regulation of the run-off and conse- 

 quent increase in the low water discharge. As our 

 old friend "Mike" once said: "It's better to have 

 a little liquid refreshment when you need it, than to 

 have a high old time twice in a year." 



It has been demonstrated that evaporation, greatest 

 of all from a water surface in the open, is nearly as 

 great from a wet earth surface similarly situated, and 

 that the evaporation from a tract surrounded by forests 

 is far less than it is from otherwise similar, but un- 

 protected areas, this being due principally to the char- 

 acteristics of the forest as a modifier of temperature 

 and as a wind-break and shield. 



This matter has been discussed at length by Mr. G. 

 W. Rafter in a number of valuable papers, in which 

 he shows beyond doubt that in humid regions at any 



