152 



SCIEXCE. 



[Vol. VII., No. 158 



tain ranges, many wet strips and spots where no 

 definite records are kept, and which therefore 

 cannot at present be shown on the map. Multipli- 

 cation of records, as well as general local knowl- 

 edge, will, no doubt, reveal many such places. 



South Africa may be described as a central 

 tableland, rising in successive terraces from the 

 seacoast. The country has been subject to great 

 erosion from water; and the mountains with 

 which the country is studded, and the dee]) val- 

 leys, locally termed 'kloofs,' with which it is in- 

 tersected, are principally due to this cause. In 

 some cases the hills are covered with forests, but 

 generally they are now denuded of such covering. 

 Where the streams take then rise in an area de- 



winter half of the year from April to September, 

 than in the summer half from October to March ; 

 May, June, and July being the wettest months, 

 and the north-west the general rain-bearing wind. 

 On the eastern side, however, the fall is greater in 

 the summer half, March and February being the 

 wettest months, and the south-east the rain-bear- 

 ing wind. On the south-east coast the dividing- 

 line between these two different characteristics 

 appears to run out to sea very nearly at Port 

 Alfred, where the fall in both halves of the year 

 appears to be very nearly identical, that in the 

 summer half being rather the greater of the 

 two. 



Mr. Tripp resided and kept records for some 



Map of 



SOUTH AFRICA 



sJxowing the 

 MEAN ANNUAL RAINFALL 



"William B . Tripp . 

 TlLInst CE. P.RJWet-Soc. 



ENGLISH MILES 



_SB_ 



JUL 



nuded of forest, their volumes are highly variable, 

 their beds being dry for, in some cases, three hun- 

 dred days in the year ; their only existence being, 

 in fact, as torrents after sudden thunder-storms. 

 When, however, it fortunately happens that the 

 streams take their rise in an area still largely 

 clothed with forests, they are frequently perennial. 



The climate over such a large tract of country 

 as South Africa varies, of course, considerably. 

 The rainfall varies from 2 inches to 50 or 60 

 inches doubtless, and perhaps more in some of the 

 mountain districts : in the north-west corner of 

 Natal it is considerably over 30 inches. Dividing 

 the country according to the half-yearly distribu- 

 tinn of rainfall, we find that on the western and 

 southern sides tin- fall is generally greater in t lie 



years at King William's Town, in the eastern 

 province, where, on a total of 283 days, from 

 June, 1880, to May, 1883, 70 inches of rain were 

 registered, 30.2.") inches of which (recorded on 242 

 days) were made up of falls under 0.50 of an inch, 

 19.88 inches (on 28 days) from 0.50 of an inch to 1 

 inch, and 19.87 inches (on 13 days) of falls of f rom 

 1 inch or upwards, in twenty-four hours. The 

 heaviest fall on any one day was 2.04 inches, and 

 the next heaviest fall was 3.11 inches, in forty- 

 eight hours. 



Although this record proves that the rains are 

 moderate and tolerably distributed, and do not all 

 occur in sudden storms, yet the author has heard 

 accounts of sudden falls there of a very different 

 character, and most disastrous in their results. 



