172 AMUND HELLAND ON THE FJORDS, LAKES,
raiiies and terraces in front of their lower ends. These, in South- eastern Norway, lie in four parallel rows, following the limits of the ice-masses indicated by the moraines.
Many lakes occur in Norway without moraines at the lower end. The theory, however, does not require that all lakes should have moraines in front of them ; for the motion of ice-masses over an undulating surface formed lakes in every place where their power was for some reason or other increased. The above-mentioned instances are those where the phenomena are clearest.
The Fjords, Fjord-valleys, Lakes, and Sea-banks of Greenland and Norway. — It is well known that certain districts in northern and southern latitudes abound in fjords and lakes ; and, further, that this abundance is not absolutely dependent on these districts being situated in high latitudes, but that fjords and lakes occur in all countries once covered by glaciers.
Thus there are many lakes in Switzerland, though it is south of the general zone of lakes and fjords in Europe. This intimate connexion between fjords, lakes, and the old glaciation, first ex- plained by Professor Dana and Professor Ramsay, led me to a closer examination of the fjords and lakes in Norway and Greenland, some of the results of which are given above. The fjords are usually regarded as valleys, or as parts of valleys, filled by the sea ; so that if the land rose, the fjord would be converted into a valley.
This view is not quite correct. All strongly marked fjords are not only valleys filled by the sea, but also lake-basins. This could be demonstrated by many instances from Norway ; but I will here confine myself to one. The Hardanger Fjord, in the inner part, gives soundings up to 800 metres ; the depth of the sea at its mouth does not exceed 350 metres. Hence, if the land rose this much, the fjord would become a lake 450 metres deep. If Norway and the bottom of the German Ocean were elevated together, its fjords on the whole would become long and deep lakes. That the same would happen with the Greenland fjords is shown further on. Thus the fjords and the lakes are formed in the same way.
We will endeavour to follow the theory of the glacial formation of fjords and lakes to its consequences, constantly using the ob- servations to verify the deductions. If, during a certain geological period, large portions of the land were covered with glaciers, and the fjords were formed by these, and could only be formed by these, then fjords must be confined to such portions of the land as were once covered by glaciers. Professor Dana has shown that this is the case ; and Professor Ramsay has further shown that there is an intimate connexion between the occurrence of lakes and old glacia- tion. The fjords, if formed by glaciers, must not be confined to certain rocks, but must traverse all deposits older than the Glacial epoch. Even a rapid journey along a fjord, for example the Sogne and its branches, or a glance at the geological maps of countries with fjords will prove that fjords and lakes are, on the whole, independent of the rocks in which they occur. The inner branches of the Sogne Fjord, as the Lyster Fjord, are surrounded by clay-slates and
