284 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1917. 



and also contains a fauna of large diluvial mammals (Bos primi- 

 genius, Cervus elaphus, Equus caballus, Elephas primigenius, Rhino- 

 ceros tichorhinus, Ovibos moschatus) which is very similar to the 

 fauna of the Rixdorf horizon in the Berlin region, and appears to be 

 of the same age. The lower part of this terrace is thus evidently 

 interglacial, but the upper part contains a molluscan fauna which, 

 according to Menzel, is of Arctic type, and thus represents the suc- 

 ceeding glacial period, and Grupe finds that this terrace interdigi- 

 tates to the north with moraine formations of the second ice sheet 

 at Hameln. The third ice sheet did not extend into the region of 

 the upper Weser Valley dealt with by Grupe, but as Stoller found 

 that in the lower Weser Valley the lower terrace (up to 5 meters) 

 was deposited during the glaciation by this ice sheet, and derived 

 part of its materials from its moraines, Grupe considers that the 

 lower terrace on the Weser at Hameln also corresponds to the third 

 glacial period, though a peat layer at its base shows that the forma- 

 tion of this terrace also commenced during the preceding interglacial 

 period. 



Siegert's conclusions are quite different. He was unable to con- 

 firm the existence of a high terrace 70 meters thick, but found instead 

 a much thinner terrace with northern material, overlain from Hameln 

 downstream by a thick series of banded clays, marl sands, ground 

 moraine and end moraines of the second glaciation. Older than the 

 high terrace are remains of the glacial formations of the first glacia- 

 tion of the district, and still older, higher Weser terraces of local 

 materials only. The Weser "high terrace'' thus dates from the first 

 interglacial period, and Grupe was in error in stating that it inter- 

 digitates with the moraines of the first glacial period. 



The lower part of the middle terrace, with its temperate fauna and 

 flora, thus corresponds, not with the first interglacial, but with the 

 second, and the upper part, with the arctic mollusea, with the third 

 glacial period. The low terrace, Siegert therefore places in the post- 

 glacial period. 



In deciding between these conflicting age determinations, it seems 

 safe to place the mammal deposit in the lower part of the middle 

 terrace on the same horizon as the Rixdorf mammal bed of Berlin. 

 i. e., in the second interglacial, which confirms Siegert's view of the 

 age of the high and middle terraces. On the other hand, there is no 

 reason to reject Stoller's conclusion that the low terrace of the Weser 

 west of Luneburg also belongs to the third glacial though to the con- 

 cluding stages of it. This refers both middle and low terraces to 

 the third glacial, separated by a period of erosion and improved 

 climate corresponding to the Baltic interstadial, to be described later. 

 This seems legitimate, for a climate allowing the growth of small 



