184 THE DESICCATION OF AFRICA. 



become heated much more slowly than the land during the day, 

 and would part with its heat equally slowly during the night, 

 so that as a general rule the water would be colder than the 

 adjoining land by day and warmer during the night. As a result 

 there would be on-shore breezes during the day, and on-lake 

 breezes at night, or, at least, a modification of the general cir- 

 culation due to the pressure distribution, in consequence of the 

 tendency to the production of such winds. 



A second result would be an increase in the humidity of the 

 atmosphere due to the evaporation from this large water-surface. 



A third effect would be a tendency to the reduction of the 

 extremes of daily temperature, tlie maximum being slightly 

 lowered, the minima being raised more particularly, owing to hot 

 winds being cooled and cold winds warmed in their passage over 

 the water. These last two effects would be more apparent on 

 the leeward than on the windward side. A somewhat similar 

 result would be locally apparent in the seasonal variation. A 

 similar effect is met with along the eastern shore of Lake 

 Michigan (20,000 square miles in area), where the amelioration 

 of the climate due to the presence oif tthis large body of water is 

 such that peaches, grapes, etc., can be grown in a narrow strip 

 known as the " Fruit Belt." Whilst there can be but little doubt 

 that the impounding of a large sheet of water such as this would 

 produce the modifications of local climate just mentioned, it is a 

 very different matter when we come to consider the modifica- 

 tions of actual precipitation that are likely to ensue. It then 

 becomes necessary to ask the questions. Is the proposed scheme 

 capable of producing the effects claimed for it? And if so, is it 

 likely that the results claimed for the scheme would be actually 

 produced? The mere proximity of extensive sheets of water is 

 of itself insufficient to^ ensure any area having a large or even 

 moderate rainfall, otherwise there would not be such a state 

 of affairs as prevails at Walvisch Bay, on the West Coast, where 

 precipitation amounts tO' something like one-third of an inch in 

 the year. If for the sake oif argument it be allowed that a rain- 

 fall of 20 inches per annum is the minimum on which agricultural 

 operations can be carried on without artificial watering {i.e., 

 without the assistance of irrigation), it is found that in the Union 

 of South Africa alone there are approximately 240,000 square 

 miles having less than this amoimt.* Of these, something like 

 128,000 square miles have less than 10 inches; so that, 

 on a rough estimate, an additional amount of 9 inches 

 per annum would on the average require to be dis- 

 tributed over these 240,000 square miles to bring the annual 

 rainfall up to the minimum assumed necessary for agricultural 

 purposes. Even granting that the evaporation from this water- 

 surface amounts to 100 inches per annum, and further assuming 

 that the whole of this amount were transported and distributed 



*Schunke Hollway states that the desert region of South Africa 

 measures 700,000 square miles. 



