THE CORRECTION AND COMPARISON OF CURVES OF GROWTH. 



135 



distinct positive reasons for thinking that they are due to cUmatic pulsations. The first 

 of these has already been given, namely, the agreement which Professor Douglass finds 

 between annual variations in the rate of growth of the yellow pine and in the rainfall. 

 Subsequent to the publication of his original paper other investigators have come to the 

 same conclusion in respect to other trees; and in the next chapter I shall show how 

 far it is true in respect to the Sequoia washingtoniana of California. Another strong 

 reason for the belief that cUmatic pulsations are the cause of the fluctuations in the 

 curve is found in the agreement of curves for widely different areas; for instance, the 

 curves derived from 32 specimens of the bull pine of San Bernardino County, in southern 

 California, from 73 specimens of the red fir at Henry's Lake, Idaho, and from 26 

 specimens of the white oak in Fenton, Missouri, all agree in having a marked maximum 

 between 1720 and 1730 a. d., a minor maximum between 1770 and 1780, and a second 

 marked maximum between 1820 and 1830— or, in the case of the white oak, a decade 

 later. In other cases there is a similar agreement: for example, the 163 specimens of red 

 spruce from Piscataqua County, Maine, the 223 spruces from Nicholas County, West 

 Virginia, and the 29 Douglas firs from the Salmon Forest, Idaho, all have a pronounced 

 maximum between 1680 and 1690 a. d. Next comes a period of slow growth, lasting a 

 century or more, but broken by a shght maximum about half-way. Then again the 

 curves rise, and reach a maximum between 1810 and 1830. On the other hand, the white 

 oak of West Virginia, with its 728 trees, has its maxima and minima at almost opposite 

 times from those of the red spruce from Maine. In the same way the curves of the bull 

 pine of southern California and the Sequoia sempervirens from the northern part of the 

 State are aknost diametrically opposed to one another. This is not surprising, for the 

 records of the 30 years from 1878 to 1908 show that the main fluctuations of the rainfall 

 of these two regions are in general the reverse of one another, which is precisely what we 

 should infer from the trees. 



The evidence of these 17 curves of growth, so far as our main problem of climatic 

 changes is concerned, all points to one conclusion. The growth of trees in the United States 

 seems to be characterized by long and important cycles, having a periodicity of 100 or 200 

 years, more or less, and affecting all parts of the country. So widespread a phenomenon 

 can scarcely be due to anything but climate. If we were able to interpret each curve more 

 exactly in terms of rainfall, snowfall, drought, temperature and the like, we should probably 

 find the agreement much closer than is now the case. The very fact, however, that curves 

 from Maine and Idaho, or from southern California and Missouri, agree so closely, seems 

 to prove that cUmatic pulsations of relatively long period affect the whole country almost 

 simultaneously, and can be read in the trees. It must not be inferred, however, that the 

 effect is supposed to be the same in all place-. A change which causes drought in the north 

 may produce an excess of moisture in the south. 



The possibility of this is well illustrated in figure 32, which shows two curves for the 



Date 1600 



1700 



1800 



1900 



Growth in 

 inches 



0.200 



0.150 



Fig. 32.— Curves of Growth of Western Yellow Pine in New Mexico (dotted line) and Idaho (solid line). 



(See Table H, pp. 325-327.) 



