rEKSI])KXTIAL A])])RESS— SKCTIOX C. 53 



toward the extinction of the ancient flora. But any change 

 in the flora eventually has an effect on the climate also, as I 

 hope to show further on. 



Climatic Cycles. 



That cyclical changes of climate occur everywhere is now 

 well established — the causes may not be clear or convincing, 

 the duration of the cycle may be irregular and unreliable, but 

 the effects are pronounced, though not always direct, or even 

 easily connected. 



The cosmic readjustments whi(di have produced in succes- 

 sion one or more whole-world glacial periods of enormous 

 duration, alternating with periods of more or less torrid 

 equatorial conditions, have rendered necessary gradual read- 

 justments of the flora and fauna capable of enduring each new 

 condition as it came into existence. 



But climatic cycles of much shorter duration, distinctly 

 affecting the flora and fauna, have also been established which 

 apparently have their origin in the more or less regular cycles 

 of sunspot production, the periods of maximum activity 

 extending over several years, and by interfei'ence with solar 

 radiation producing on earth conditions of less intense light, 

 with consequent effect upon barometric pressure, upon rain 

 production, and upon the vegetation and the fauna. 



The Senate Select Committee on Droughts, Rainfall and 

 Desiccation, 1914. reported, inter alia, re rainfall: "4. (c) 

 That while there is some evidence to support the theory of the 

 periods of maximum and minimum rainfall corresponding with 

 certain cycles, there are not sufficient data available to define 

 any such cycles." 



That there occur periods or seasons of heavy rainfall at 

 considerable but more or less irregular intervals, with much' 

 drier years or periods intei'vening, has long been recognised in 

 South Africa as well as elsewhere, and for many years these 

 have been correlated with sunspot phenomena. 



Close study in America and in Europe (see writings of 

 Douglas, Huntington, Clements, cited at the end of this paper) 

 place this beyond further doubt, especially as it is fully 

 supported by rings showing annual tree growth for 2,000 years 

 past, and I advise those who still doubt this relationship to 

 investigate the matter further before they commit themselves 

 to op])osition views. 



The relationship appears to be an indirect rather than a 

 direct one, apparently in the direction of a direct influence 

 on the migration of the banc factors, which in due course, but 

 not everywhere contemporaneously or to equal extent, affect 

 local climate, and naturally not always in tlie same direction, 

 since if extra pressure exists in one place it corresponds with 

 reduced pressure in another, or, in other M'ords, increased 

 rainfall in one locality corresponds with the absence of rainfall 

 somewhere else. 



The sunspots themselves appear with an irregular cycle, 

 which ranges from seven to seventeen vears between the 



