322 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1930 



North America, render De Geer's correlation of Hackensack clays 

 with Dutchess Junction and Connecticut Valley clays as premature 

 and misleading. It will be necessary for Brooks to revise his coeffi- 

 cients of correlation again when the teleconnections between Sweden 

 and eastern North America have been established. 



SUGGESTED CORRELATION OF SOLAR RADIATION, WEATHER, AND 

 VARVED CLAY VARIATIONS 



It has been stated that the character of the weather during the 

 melting period of the ice varied from year to year, and that these 

 fluctuations arc reflected in the varying thickness of the annual de- 

 posits (Bruckner, 1921; Osborn and Reeds, 1922; Sauramo, 1929). 

 After calling attention to wide range correlations or teleconnections 

 between Sweden, North America, Argentina, and northwestern 

 Himalayas, in which more than 80 per cent of the varves are said 

 to agree, De Geer (1927) states: 



This remarkable coincidence of such rapid variations at such considerable 

 distances, caused by simultaneous ice melting, seems not to be explicable in any 

 other way than by variations in the amount of heat from the sun. 



De Geer (1926) states that the rate of melting depends almost en- 

 tirely on the intensity of solar radiation. In those temperate regions 

 which were glaciated during the Pleistocene epoch he considers that 

 practically all of the solar radiation was spent in melting the ac- 

 cumulated snow and ice, while now, after the yearly snow and ice 

 is melted, the balance is available for warming the ground. During 

 the retreat of the ice he notes that the changes in solar radiatiojn 

 correspond fairly well with the changing amount of melting water. 

 Since this water found its way to the ice edge, he considers that the 

 amount of reassorted morainic material eroded and carried away 

 by the subglacial rivers varied very nearly at the same rate as the 

 amount of melting water. Furthermore, by getting accurate meas- 

 urements of the annual amount of the finest and most regularly de- 

 posited sediment from such waters (varved clays), he considers it 

 would be about the same as getting readings from a gigantic natural 

 self -registering thermograph. Brooks (1928) suggests solarigraph 

 for thermograph. 



A comparison of Figure 1 with Figures 2 to 13 shows that the 

 solar radiation variations, when plotted in the same manner, are 

 not different in kind from those of the varved clays. In fact, it is 

 surprising how close the solar fluctuations simulate those of the 

 varved clays. No direct comparison can be made between them, 

 since the years represented in Figures 1 and 2 to 13 are not con- 

 temporaneous. The close agreement in the form of the graphs, 

 however, leads one to offer a tentative correlation of the annual solar 

 radiation variations with those of the varved clays. Clayton has 



