THE DESICCATION OF AFRICA. 185 



over the Union in proportion to the requirements df each 

 locality, a simple calculation shows that it would add an 

 additional 6 inches to the rainfall, bringing the average up to 17 

 inches, thus leaving an additional 3 inches to be met from 

 artificial watering. It is of course not only possible, but very 

 probable, that a certain proportion would be re-e\aporated from 

 the ground and re-precipitated, but this would take place in 

 decreasing amount as the distance from the source of supply 

 increased, and with each recurrence of precipitation and subse- 

 quent evaporation. 



In order to produce the desired result it is not sufficient 

 merely to get the moisture into the air, but the mechanism must 

 also be available, not only to transport the moist air to the desired 

 area, but also to re-precipitate it over that region. The main 

 cause of aridity in any part of the world is not that there is not 

 sufficient moisture in the air, but that nature has withheld the 

 means of cooling the atmosphere sufficiently over sudh regions to 

 cause condensation. It is evident that in order that this scheme 

 should benefit the arid and semi-arid portions of the Union, there 

 ought to be a fairly persistent northerly to north-easterly wind to 

 transport the moisture ; whilst over these dry regions the means 

 of dynamically cooling the air to such an extent as to produce 

 precipitation must be at hand to take advantage of the presence 

 of the moist currents. This might take the form of a denser 

 stratum of air, which, by under-running the warm, moist layer, 

 would cause it to rise and so cool by expansion to such an 

 extent as to cause it to part with its moisture ; or the atmosphere 

 over t'he dry regions must be so warmed as to cause convectional 

 (vertical) ascent, resulting in thunderstorms, etc. As the 

 attitude of the proposed scheme is approximately 3,500 feet, and 

 as the dry areas of the Union vary from 1,500 to 4,500 feet 

 (mostly about 4,000 feet) it is evident that but little rain could be 

 expected from orographic causes, that is produced by cooling 

 due to forced ascent over mountains or other high ground. 



Athough nothing is known of the meteorological conditions 

 prevailing over the site of this proposed scheme, it may be per- 

 missible to examine the weather conditions at Bulawayo as 

 published by the Rev. Father Goetz in his papers on " Some 

 Meteorological Features of Southern Rhodesia " and " The 

 Rainfall of Rhodesia." 



Bulawayo (20°9' South Lat.) is in practically the same 

 latitude as, but a little further east than, the Alakarikari, and so 

 lies well within the Tropics. It is not, therefore, surprising to 

 learn that Father Goetz asserts that a rainfall not accompanied 

 by electrical phenomena is an exception from October to April. 

 An examination of the tables of hourly distribution of rainfall 

 shows that 75 per cent, of the total occurs between i p.m. and 

 midnig'ht, the largest portion (42 per cent.) falling between i 

 and 6 p.m. 



It would therefore appear that thunderstorms are the most 



