108 



THE CLIMATIC FACTOR AS ILLUSTRATED IN ARID AMERICA. 



to form a single ring in 22 sections, a double ring in 19, and to be clearly separate in 

 only 11 cases. They seemed to represent one year. Nos. 53 and 54 were found to form 

 a single ring in 6 cases, a double one in one, and clearly 2 in the remaining 24 cases for 

 which sections were available ; accordingly they were considered to represent two years. On 

 comparing the plotted curve of tree growth with the curve of rainfall, the two were found 

 to agree more closely if ring No. 6 were assumed to represent two years (1903 and 1904) 

 rather than one, but the real evidence strangely enough came from Flagstaff. The cross- 

 identification between the sections from Prescottand Flagstaff made it possible to identify, 

 unquestionably, most of the rings, both before and after 1903, and Flagstaff plainly showed 

 two rings in place of the doubtful ring or rings called No. 6 at Prescott. Hence this was 

 apportioned to the two years 1903 and 1904. Apparently, if a sufficient niunber of com- 

 parisons be made, and if the trees thus compared be distributed over widely different 

 localities, the yearly identification of rings may be made with almost perfect certainty. 



Year 



Fig. 10. — Annual Rainfall and Growth of Trees (Group V) at Prescott. 



Dotted line = Rainfall. Solid line = Growth. 



The final curves resulting from the process described above are given in figure 9. 

 The upper four curves represent the amount of growth year by year of each of the four 

 groups mentioned above. The lower curve shows the mean of all four groups. It will be 

 seen that on the whole these four groups from different localities, 10 miles or so apart, agree 

 quite closely. Nevertheless, as has already been said, the trees of the group nearest to 

 Prescott agree most closely with the rainfall at that place. Accordingly, their growth has 

 been plotted in figure 10, together with the rainfall at Prescott. On the whole there is 

 much agreement, as may be seen by comparing the crests and troughs of one with those 

 of the other. The most conspicuous discrepancy is in 1886, where the rainfall decreases 

 and the growth of the trees increases. In 1873 the growth seems to have responded to 

 the decrease in rainfall, but to a greatly diminished degree. The tree maximum of 1875, 

 one year behind the extreme maximum of 1874 in the rainfall, is entirely reasonable, since 

 the ground may become so saturated that the effects last until the following year. The 

 general falling off of the tree curve during the last twenty years will be discussed later; it 

 is due merely to the fact that the trees grow slowly in old age. On the whole, the curves 

 shown in both figures 9 and 10 support the idea not only of the similarity of the rings of 

 a given year in different trees, but of a proportional relation between annual rainfall and 

 annual growth. 



The conclusions regarding yearly identity drawn from the curves at Prescott are sup- 

 ported by those of Flagstaff. In addition to the 25 sections procured there in 1904, 7 

 others were procured in 1911. The pieces for examination were not cut horizontally as 

 hitherto, but were secured by making two slanting saw-cuts at right angles to one another 

 on the top of the stump, thus bringing away a triangular pyramid of wood, which included 

 the outer 50 to 100 rings. These cuttings were for the purpose of checking the growth 

 in the last half century but made no pretense of reaching the center of the trees, whose 

 average age was three or fom' hundred years. Figure 11 shows how well the 7 cuttings of 



