rKKSIJ)ENTIAL ADDllKSS SKCTIOX C. 55 



The exact correspoudence of certain increases and 

 decreases of rainfall (four rises and four falls) shown by 

 Arctowsivi (1915) during the period 1900 to 1910' for certain 

 stations in Peru, Mauritius, Madagascar and South Africa 

 (Bulawayo) (see Clements, p. ooO and fig. 29). indicate that 

 the SoutLern Hemisphere has some independent general factor 

 producing simultaneous coincidence in certain localities which 

 is worth further investigation, as also is the distribution of 

 local variations arising therefrom. 



In 1888 Mr. D. E. Hutchins ])ublished '; Cycles of 

 Drought and Good Seasons in South Africa"; in 1890 he 

 brought the same subject before the Royal Meteorological 

 Societj', London, and he further dealt with it in the Cape 

 Agricultural Journal, 1897, xi, p. 701,* and 1898, xii, pp. 138, 

 211 and 26T. In all these he claims that Soutli Africa is 

 subject to three rainfall cycles, of different period, working- 

 simultaneously, viz. : — 



(1) " Storm cj'cle, bringing the heaviest rain to western 



winter rainfalls, but usually only wind to eastern 

 stations. Period nine and ten years alternating " 

 ( = Ilusseirs nineteen-vear cvcle, Cape Arjric. Jour., 

 xii, 272). 



(2) " Meldrum's cycle, liringing the heaviest rain to 



eastern summer rainfalls, 1nit usually little rain to 

 western stations." This lie previously' called 

 " Mitigation cycle " ; period, 12'5 years. 



(3) Sunspot cycle; period, ll'll years. — "The least 



important, or at any late the least powerful 

 and punctual, of tlie three South African weather 

 cy(des." 



These three cycles so rarely fall due at one time that it is 

 126 years since they were so near together as happened in 

 191G-1917, which years Hutchins predicted thirty years ago 

 would be a time of unecjualled eastern rainfall, in which, at 

 most localities, his prediction has come wonderfully correct, 

 as also have others of his ]U'edictions as to rainfall and also 

 as to drought. 



The net result of all this, however, is that if Hutchins' 

 theory is on a sound basis (which I neither affirm nor contra- 

 dict), and apart from any deferred rains which may fall up 

 to 1920, the eastern rainfall may be expected to be less yearly 

 for some years from 1918, and not much again u]) till 1927- 

 1930, but that the south-west districts should get good rains 

 about 1926, Avhich may extend as storms to the south-east 

 coast also, and that towards the close of the dry period above 

 mentioned intense desiccation is to be expected throTighout 

 Soutli Africa, except on the coast belt and sub-coastal 

 mountain ranges, since the rainfall of seasons 1917-1918 and 

 of 1920. which kee]i tlie average fairly high, mostly went as 



* Reprinted as an Agricultural Department Bulletin, with correc- 

 tions, for, nnfortmiately. p. 709 in the A(jric. ■Journ. has^ many evident 

 printer's misplacements as to due dates. 



