350 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1917. 



vestiges of the ice sheet, but in Germany, where elevation was less and 

 the ice was more remote, the traces of the continuation of the dry 

 period into the period of higher temperatures are slight. Probably 

 also, in the course of dev elopment of modern storm tracks, North 

 Germany became more humid earlier than Scandinavia. 



The " Grenztorf " is separated from this earlier dry period by a 

 considerable interval, during which Sphagnuui peat accunudated. 

 This Sphagmim peat falls naturally into place in the period of the 

 Litorina depression, and the "Grenztorf" would then be associated 

 with the elevation which terminated that period. It3 occurrence 

 finds an obvious explanation in the high temperatures which pre- 

 vailed at that time over Scandinavia, allowing the cyclonic storms to 

 take a more northerly course, so that north Germany largely escaped 

 them. The " Grenztorf " then corresponds to the boreal period of 

 Sweden, a correlation which is confirmed by the fact that each of 

 them is associated with predominant foi-ests of oak. 



At the International Congress at Stockholm short accounts were 

 given also by Kupfer and Lindberg of investigations in the Baltic 

 Provinces of Russia and in Finland. Kupfer found that the cold 

 climate of the last glaciation was followed first by a subarctic cli- 

 mate and then by a dry climate resembling that of central Russia. 

 This, which evidently corresponds to the Ancylus elevation, gave 

 place during the Litorma subsidence to a damp, warm period wdiich 

 brought the climate and vegetation of the western European coasts 

 to the eastern Baltic. Kupfer makes no mention of a second dry 

 period between this and the present day, but such a dry period is 

 suggested by the reference made by G. I. Tanfiljew at the same con- 

 gress, to layers of tree stools in the peat, and to the gradual replace- 

 ment still continuing of forests by steppe in central Russia, though 

 Tanfiljew did not attribute these phenomena to climatic changes. 



The most important result of the work of H. Lindberg was the dis- 

 covery of a well-marked warm period in Finland, characterized by 

 the occurrence of a horizon with Trapa natans, Carex pseudoGype7nis, 

 and Ceratophyllum demerswn. Trapa is now entirely extinct in 

 Finland; the other tw^o are found in peat beds far to the north of 

 their present habitat. This warm period coincided wdth the latest 

 Ancylus period and the maximum extent of the Litorina Sea, for at 

 Sakkola in the Carelian Isthmus (and in the island of Aland in the 

 Gulf of Bothnia, as recorded by P. H. Olsson in 1900) the Trapa 

 bearing deposits are directly overlain by beaches of the Litorla Sea. 



Briefly summing up, we find that in postglacial times in northern 

 Europe. there have been two dry periods, one preceding and the 

 other following the Litorina subsidence, and that in the Fennoscan- 

 dian region these two dry periods were also warm. 



