336 



Wheat Prices and Rainfall. 



[July, 



most of these occurred at dates which might be calculated by 

 carrying back some of the cycles found. 



The material actually used covers only the three centuries 

 1550 to 1850, the later years being entirely ignored in the inves- 

 tigation, for the reason indicated above, viz., that European 

 prices of wheat since then (or at least since 1870) depend far 

 more upon the harvests of the entire world than of the locality 

 where the price is paid, and thus — as cycles probably affect 

 different parts of the world differently — high prices 

 would not necessarily indicate scarcity in western 

 Europe. This precaution enables him to test the con- 

 tinuance of the phenomena since 1850, as well as their 

 utility for prophecy, by comparing a calculated curve with 

 actual results. He accordingly adds together the theoretical 

 effects of the eleven best established cycles in each year, and 

 constructs a " synthetic " (curve — as he calls it — of wheat 

 prices. Upon the assumption that, if we are looking for a single 

 factor which is uniformly adverse to a good harvest, we shall 

 get nearest to finding it in rainfall, the " synthetic " curve 

 between 1850 and 1920 is compared with the curve of rainfall 

 in western Europe during the same years, and the main peaks 

 in the two are seen to correspond very closely. 



Eeference may be made to the deduction drawn by Sir William 

 Beveridge some year or two ago as to the probability of heavy 

 rain and bad harvests in 1923. This was taken by many people 

 as a prophecy, but his later inquiries do not lead to the same 

 conclusion as to the general meteorological condition of the near 

 future, and such an interpretation is now to be regarded as with- 

 drawn. He fully believes that trustworthy prophecy of the 

 weather will, in due course, become possible, bat that it is not 

 yet possible on the facts as he has given them. Prophecy will 

 become possible, if at all, only after detailed investigation has 

 shown the nature, shape, relative importance, and, above all, 

 local variations of each cycle. At present little more can be 

 said than that such cycles exist and are noticeable as periodic 

 changes in the rainfall. As his examination shows, these cycles 

 do not necessarily persist indefinitely : many of them have per- 

 sisted for centuries, but others have died away, or their periods 

 have become modified. Such considerations render it dangerous 

 to forecast the weather of any given year; and. in fact, the 

 " synthetic " curve above referred to shows several discre- 

 pancies in certain years with the actual rainfall records, although 



