44 CLIMATIC CYCLES AND TREE-GROWTH 



from erratic estimations, on account of which ordinary eye-smoothing 

 may be criticized. The graphic Hann thus formed is the basis of the 

 cycle plot whose process of formation will be continued below. 



Emphasis point — In the Hanning process just described the 

 emphasis is laid on the middle point of the three. This has been used 

 in so large a part of this curve-production that it is here given pref- 

 erence. But there is some question about its use when conservation 

 is considered, for it intimates a reversed or negative conservation in 

 the last year of the three (see p. 101). If rainfall is retroactive, that 

 is, if it affects rings already formed, the tree records ought to show 

 some anticipation of abrupt changes in the rainfall. On the other hand, 

 placing the emphasis on the last of the three years used amounts to 

 admitting a conservation of moisture from the two preceding years. 

 On the whole, it is felt that middle-point emphasis has given more 

 satisfactory curves than emphasis on the final year. 



Cutting-line — The cycle plot has the maxima of the curve cut out 

 so that light may pass through. The curve produced by the graphic 

 Hann forms the upper side of this area to be cut, but the position of 

 the base of the cut area has proved very important in the successful 

 use of the analyzing instrument and therefore I have always had the 

 curves at that stage returned to me to have the base or "cutting-line" 

 marked. In any analysis the variations of the curve are the important 

 features; hence, if the cutting-line is placed along the X-axis or the true 

 base of the curve, the variations are reduced to very small percentages 

 of the total light coming through and can not be seen. Even when 

 the cutting-line is placed at the lower minima, the light is so abundant 

 that it is very hard to get the variations visually or photographically. 

 After extensive trials of every sort of height for this line, I have come 

 to the general plan of sacrificing about one-third of the vertical height 

 at the bottom of the minima and marking a long, sweeping line nearly 

 straight, but not entirely so, as that brings the best display of varia- 

 tions within the range of the instrument and has not been found to 

 affect the results. 



The range of the instrument as now in use is confined to periods 

 between 6 and 32 years (see "Recent changes," below). The cutting- 

 line, therefore, to show these best, may be curved so as to cut out or 

 reduce longer periods. They, however, are taken care of by plotting 

 at a reduced scale. This has been done extensively with long sequence 

 of rings extending 500 years or more. 



Cutting the plots — The final work on the cycle plots is cutting out 

 the maxima, which, of course, is a simple matter usually done with a 

 razor blade. 



