NO. 10 SOLAR RADIATION AND WEATHER STUDIES ABBOT 7 1 



tion that while a tree is young, with its roots shallow and but little 

 extended, the water supply on which growth so largely depends would 

 respond more directly to periodic changes in precipitation than when 

 the tree becomes very old, with a widely extended root system, possibly 

 tapping never failing sources of water supply at a considerable dis- 

 tance from its trunk. 



With this view in mind, I have for the most part restricted this in- 

 vestigation of 23-year periodicity, and the illustrative curves to which 

 I shall refer, so as merely to present periodic changes in the widths of 

 tree rings from about 20 selected localities from which wide rings 

 at the top of a Douglass table led down in a century or two to much 

 narrower rings. In these cases it seemed most probable that his 

 measurements had to do with young trees. 



Figures 29 and 30 give the results of these investigations. It appears 

 that in all of these cases, tabulations extending over 115 years indi- 

 cated changes of tree-ring width during 23-year cycles ranging be- 

 tween 40 and 120 percent, and with such definiteness of gradation, 

 from low to high and return, as seems in harmony with the idea of 

 periodicities of 23 years in the water supply on which the tree growth 

 depended. 



In another investigation of this subject I have kept each 23-year 

 cycle by itself, but have combined the results from five localities in 

 southern California and Nevada. In that way I have determined indi- 

 vidually the march of four successive 23-year cycles from 1829 to 

 1920 as represented by the average thickness of the rings of about 40 

 trees from five separate localities. Figure 30 shows these results. 

 Not only is a 23-year cycle apparent, but many details are reproduced 

 with such moderate alterations of phase and amplitude as to give 

 reasonable certainty of the veridity of these minor features in all four 

 cycles. As remarked above, the amplitudes of these features which 

 compose the cycles tend to diminish as the trees grow older. 



23. A Test of the 23-YEAR Cycle in Pleistocene Varves 



In a paper by C. A. Reeds," he gives many pages of illustrations 

 representing the march of the thickness of glacial varves near the 

 Connecticut and Hackensack Rivers. Independent measurements by 

 Antevs and Reeds are shown. Continuous series represent the present 

 thicknesses of these varves resulting, it is believed, from annual 

 weather-reactions extending in unbroken sequence for nearly 1,000 

 vears. 



"Ann. Rep. Smithsonian Inst. 1930, pp. 295-326, 1931. 



