INTERPRETATION OF THE CURVE OF THE SEQUOIA. 



167 



of this has been noticed, even in years where the precipitation amounts to two or three 

 times the average. In the past, it seems safe to say, the relation of precipitation and growth 

 must have been essentially the same as at present. Therefore we seem warranted in 

 concluding that in our long curve of growth, extending back 3,000 years, and given in 

 figures 38 and 50, high places indicate abundant moisture and low places indicate drought. 

 How greatly the rainfall of the past exceeded that of the present we can not yet ascertain 



1890 1900 1910 



1850 



1860 



1870 



1880 



iMms. 



Measured 

 growth of 

 bequoias 



4.00 



3.50 



3.00 



Growth of 

 Sequoias 

 calculated 

 from rainfall as 

 described in text 



Rainfall at 

 Fresno 10 

 San Francisco 



Fig. 48. — Tree Growth in California Calculated from Rainfall, by A. E. Douglass. 

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



positively. In the modern sequoias the growth during the group of 15 favorable years 

 exceeded that during the 14 unfavorable years by 0.43 mm., or 12.3 per cent of the mean. 

 The rainfall during the periods of 5 years preceding the favorable years exceeded that 

 during the similar periods preceding the unfavorable years by 0.64 inch, or 6 per cent 

 of the mean, while if a 4-year period is taken instead of 5 the excess is 1.22 inches, or 14.3 

 per cent. From this it would appear that the thickness of the rings of growth is closely 

 proportional to the rainfall. By this I do not mean to be understood as making any 

 exact or positive statement, but merely as indicating the order of magnitude of the relative 

 changes of rainfall and growth. Increasing the rainfall by 10 per cent might increase the 

 thickness of the rings by 5 per cent or 20 per cent, but it is quite certain that it would 

 not increase the thickness by 50 per cent, nor would its effect be so small as 1 per cent. 



Before attempting an analysis of the changes of climate indicated by the long curve 

 of the sequoia, let us attempt to gain some light on the nature of the monthly distribution 

 of the rainfall and the seasonal variations in storminess during favorable as compared 

 with unfavorable years. Because of the projection of the effects of past years into those 

 that follow, that is, because of the conservation factor, it is not easy to ascertain exactly 

 how much influence is to be attributed to the seasonal distribution of the rain of a single 

 year. Yet we have seen that this is a highly important factor. If the precipitation all came 

 in the form of rain in the fall, or if it all fell as snow after the ground was frozen and then 

 was rapidly melted by heavy rains in the early spring, the effect upon trees would be quite 

 different from that which would result from a uniform distribution throughout the fall, 

 winter, and spring, or from heavy precipitation from February to May. 



Four ways of testing the matter suggest themselves: (A) First we may pay attention 

 only to the amount of moisture and may compare years of exceptionally heavy and 

 exceptionally hght precipitation. (B) Next we may pay attention only to the growth 



