88 



CLIMATIC CYCLES AND TREE GROWTH 



that it is not to be regarded as a random effect whose variations will never 

 be explained. The writer made studies of it in 1913, calling it the "2-year 

 zigzag." This has been referred to in volume I, page 106, and in Astro- 

 physical Journal, vol. XLI, pages 180 to 186. 



This short cycle, or zigzag, is of the highest importance in cross-identifica- 

 tion and in mean sensitivity, and it has influenced the development of methods 



1800 I 

 2 



20 30 40 1850 60 70 80 90 1900 10 20 30 



Fig. 36 — Two-year reversal test, to show similarity in Arizona area in short cycles. 

 FAWR is Pueblo area winter rainfall; LRI (Lynch s Rainfall Indices) applies to south- 

 ern California. Note: For example, the general rise toward 1900 and the minor crest 

 at that date. 



of analysis because of its great amplitude. Analyses were tried extensively 

 upon unsmoothed curves, but resulting cycle lengths came too often in exact 

 multiples of one year (the unit used) and this was recognized as an error. 

 Resort was made to a form of smoothing which would least distort the general 

 march of the data, and that was the Hanned curve, as explained above. 



Next it is recognized that mean sensitivity, or the percentage variation 

 from each ring thickness to the next, is very largely a measure of the amount 



