162 



THE CLIMATIC FACTOR AS ILLUSTRATED IN ARID AMERICA. 



Trees, like other living beings, possess much inertia. If they are subjected to favorable 

 conditions for a few years and make a good growth, they are in a position to keep on in 

 the same way unless seriously checked. Therefore, in estimating the effect which the 

 conditions of an individual season have upon their growth, the actual amount of growth 

 is less important than the change from preceding j-ears. Accordingly I have added to 

 figure 42 a lower line showing the differential growth of the trees at Dillonwood — that is, 

 the amount of increase or decrease of growth compared with the year immediately pre- 

 ceding. This curve shows a fair degree of agreement with the curves of precipitation, but 

 there are also disagreements, such as 1887 and 1903. Other factors besides the rainfall of 

 the season immediately preceding the year of growth evidently have a large effect upon the 

 trees. Moreover, our records of rainfall are derived from places 30 to 50 miles from where the 

 trees were located. Nevertheless, in cases like 1881, 1890, 1897, 1900, and 1907, there seems 

 reason to think that we can see the direct effect of the precipitation of the preceding winters. 

 Yet in spite of this, these Dillonwood curves might be used almost as well to show that 

 rainfall and tree growth do not agree as to show that they agree. They have been intro- 

 duced here purposelj' in order to show the difficulties, and in order to emphasize certain 

 facts which will be brought out later; namely, that the growth of the trees depends upon 

 the rainfall of several years, and not of one year, and that it is influenced bj' the season 

 at which precipitation falls quite as much as by the actual amount of precipitation. 



Dale 1850 



1860 



1870 



1880 



1890 



1910 Growth 

 in mm. 



5,00 



^■°0 Group I-B 

 14 trees 

 Group I- A 

 3.00 23 trees 



Group II -A 

 ^■^" 31 trees 

 Group II -B 

 IS trees 

 ) Group III 

 25 trees 



Fig. 43. — Annual Growth of 111 Sequoias at Hume. 



(See Table I, pp. 328-329.) 



It would be possible to go on and show more clearly the exact nature and degree of the 

 effects produced upon the Dillonwood trees by past rainfall and by variations in its 

 seasonal occurrence, but as only a small part of the measurements used in our final 

 long curve of growth (extending back over many centuries) came from Dillonwood and the 

 other regions near Portersville, it seems wiser to discuss the matter in relation to the trees 

 at Hume. At that place I pursued a method somewhat different, and on the whole distinctly 

 more reliable, than that followed at Dillonwood. Instead of cutting young trees, whose 

 youth makes them particularly liable to be affected by accidents, I selected over 100 vigor- 

 ous trees in early maturity; they were from 5 to 8 feet in diameter and had a probable age 

 of from 500 to 1,000 years. The Hume-Bennett Lumber Company, on whose land they 

 stood, kindly gave me permission to cut from them small sections 8 or 12 inches in length 

 and showing the rings of growth for a period varying from 30 to nearly 200 years. The 

 difference in the number of rings depends partly on the size of the sections, but it is far 

 more largely dependent on the rate of growth of the trees, for a slow-growing tree of course 

 shows many more rings per inch than does a fast-growing one. On the basis of the number 

 of rings the 111 sections have been divided into three groups, and the first two groups have 

 each been divided into two subgroups, one of which grew in wet, swampy places, and the 

 other upon drier slopes. 



