Davis.] 
366 
[Jauuary 18 , 
north and south arms of Lugano are open to the valley of the Po, 
except for alluvial and diluvial valley-filling 1 of an undetermined 
depth; and it is worth noting that the cross-piece of this lake fol- 
lows the outcrop of a certain series of rocks, just as longitudinal 
river valleys do ; the southeastern arm of Como (known as Lago 
di Lecco) is determined by the same structural characters. In the 
same way, to the north of the Alps, the outlets of many lakes flow 
for long distances on quaternary deposits ; below Lakes Neuchatel, 
Bienne, Thun, Zurich, Hallwyl and its neighbors, and Constance, 2 
the valleys are so clogged with unstratified and stratified drift and 
alluvium that it is impossible to say how low the lakes would be 
drained were the obstructions well scoured out down to rock- 
bottom. To account for the deposits occurring below the lakes 
without filling their basins, it is not necessary to suppose any 
glacial erosion or postglacial subsidence, but simply that glacial 
ice occupied them to the exclusion of sediment, while the lower 
parts of the valleys were filled. 3 
A sufficient reason for giving so great importance to this origin 
of lakes, is that if the existence of so-called rock-basins can be 
largely disproved, one of the strongest arguments for the glacial- 
erosion origin of lakes is destroyed. The leading advocates of 
that theory lay much stress on the point that all other causes 
being excluded as insufficient, glacial erosion alone remains ; we 
question very strongly whether this obstructive cause (as well, 
indeed, as some others) has not been too much neglected. It 
would be an interesting task to examine Scotland and Scandinavia 
with this question in mind. 
Warped Valleys, Glacial Erosion Basins and Moraine Basins 
are very fertile in crossing with this species, and hybrids should 
be looked for under those headings. 
C. 5. Lava Barrier Basins. Lava-flows sometimes perform 
the same office as ice-streams in forming lake-barriers, but they are 
1 See maps of Swiss Geological Survey. 
2 L. Wiirtenberger states that the Rhine at Schaffhausen now runs in a new, post- 
glacial channel; Ueber die Entstehung des Schaffhauser Rheinfalls, Neues Jahrb. 
1871, 582-588. The old channel would assuredly drain Lake Constance lower than 
its present level. 
3 This has been suggested by Escher v.d. Linth and by Desor ; see Gebirgsbau; 
der Alpen, 136. But de Charpentier seems to be the earliest author of this idea, 
Annales der mines, vm, 1834, 228. 
