INTRODUCTION 9 



of the tree, as we estimate them one after the other from center to outside 

 of the tree. This kind of series is called "complacent." Other ring records 

 contain strong increments from year to year, such as evident in the curves 

 shown in figure 1. These are called "sensitive" records. Examples of these 

 two types are given in Plate 6 which shows a very complacent ring series in 

 BE 133 and a sensitive series in MV 23. Apparent correlation could be very 

 high between complacent records and yet such records usually contain no 

 identification characters and often are valueless. Hence, another character 

 is needed; this has been expressed in previous years as "mean sensitivity." 

 Mean sensitivity is the average percentage increment from year to year 

 without regard to sign. A tree of high sensitivity is more datable and more 

 likely to make a contribution to our studies than a tree with a complacent 

 series of rings. This sensitivity, of course, is only determined where there is 

 no perceptible injury. 



Cross-Identity — Cross-identity which is established by a process called 

 "cross-dating or cross-identification," depends on the sensitive type of ring 

 sequence. It is the foundation stone on which long ring records have been 

 based. In modern trees it involves the determination of the exact year in 

 which each ring grew. 1 There are three practical steps in this process. First, 

 one counts in from the bark of trees whose outermost ring has an obvious 

 date — for example, the outermost ring in a tree just cut down — and thus 

 finds the exact dates for various thin or otherwise well-characterized rings. 

 Second, a careful note is made of the time spacing of the specialized rings, 

 which may now be regarded as a recognizable group, and search is made for 

 an identical group in another tree of unknown cutting date ; when found the 

 group rings in the latter tree become known as to exact date. Third, in the 

 second tree, a count is now made either to the outside to find when it was cut, 

 or toward the center to ascertain the exact dates of earlier groups of specialized 

 rings. All this is easily practised in a logging camp where many stumps are 

 available. On such stumps near Flagstaff anyone can always count from 

 the outside to some distinctly small ring or group of rings and find exactly 

 the same count from tree to tree except in the rare cases where some severe 

 drouth year has occurred that caused the failure of ring formation in a few 

 trees. The identification may then be carried to stumps cut years previous 

 to the dating and the date of their cutting ascertained for checking against 

 forest records. In this way confidence in the method is easily acquired by 



1 The word correlation has been suggested in place of the words here used, identifica- 

 tion and dating. Such a change would abandon the most important fundamental of 

 tree-ring work, which is the establishment of identity of the growth-date of the indi- 

 vidual ring. It is not resemblance or relation or correlation in date that we endeavor to 

 fix, but the actual identity of date, and all the other points of interest and value flow 

 from that. The significance of cross-identification and correlation is clearer in the 

 statement that the cross-identification of a ring depends on correlations between ring 

 distribution within groups and other characters in two or more different sets of rings. 



A definite ring sequence is usually first proved by showing the common identity in 

 date of rings within a number of trees without thought of the actual dates; hence the 

 word cross-identity or cross-dating to convey the idea that the identity must be carried 

 across from tree to tree (and not taken merely within a tree). If this cross-identity can 

 be established, then the student is well on the road to actual dating. 



