314 ANNUAL REPORT SMITHSONIAN INSTITUTION, 19 31 



The several sediments recording the year are perliaps most impor- 

 tant among the means of dating, but they are rather rare, occurring 

 in any frequency and extent only during widely separated ages of 

 the earth's history. First among these sediments comes the varved 

 glacial clay. " Varve " and " varved " are v7ords of Nordic origin 

 that have been applied to designate the " annual deposit " of a sedi- 

 ment, and "yearly laminated." The striking alternation of light 

 and dark layers in the glacial clay recalls annual rings in a cross 

 section of wood. This similarity caused a Swedish scientist, as early 

 as 1769, to assume that a pair of layers in the clay constituted the 

 deposit of one year. Later, the same conclusion was independently 

 reached by several geologists, next by an American in 1832. Vari- 

 ous evidences have been presented to prove that a pair of layers 

 actually represents the year (pi. Ih). 



The varved glacial clay consists of alternating layers of sand, silt, 

 or coarse clay, light in color, and of fine to extremely minute clay, 

 dark in color. The thickness of the coarse layers varies from a frac- 

 tion of an inch to several inches. The thickness of the fine-grained 

 layers is normally smaller and varies less. The varved clay was 

 formed in lakes and slightly brackish bays outside melting glaciers 

 and ice sheets. It was formed of mud brought directly from the 

 melting ice. Similar clay is now depositing in lakes fed by glacial 

 brooks, for instance, in Lake Louise in the Canadian Rockies. In 

 winter time the late-glacial lakes evidently iroze over. 



In summer the temperature of the suri'ace water may have ranged 

 from a little above freezing to about 35° F. It may not have at- 

 tained 39.2° F., for, owing to the fact that water is densest at this 

 temperature, the entire water mass would then have been of 39.2° 

 and the lake would have had complete circulation, enabling winds 

 to mix suspended mud in all water strata and even to stir up 

 the bottom deposits. If this had been the case, the varved clay, if 

 it could be formed, would show signs of erosion, which it does not 

 do except rarely, when deposited in very shallow water. Further- 

 more, if the surface water at a distance from the ice sheet had risen 

 to 39.2°, this surface water would have sunk in the main lake 

 body, since it was heavily loaded with mud. At a still greater dis- 

 tance from the ice border, where the water temperature was higher, 

 the mud would have sunk rather quickly, because of the lower vis- 

 cosity of the water. Transportation of large quantities of mud in 

 glacial lakes for more than 100 miles shows that the mud did not 

 sink quickly. Therefore, the surface water of the glacial lakes may 

 have ranged in temperature during summer from 32^ to about 35° ; 

 the bulk of the water may have been, both in summer and winter, 



