to be deposited on the spot, a fact which causes very fine 

 deposits, especially as also the other formerly above-mentioned 

 factors will be of no great importance here. The samples which 

 were found in Carlsberg Fjord, Fleming and Hurry Inlets 

 belong, as has already been shown, to the very finest, while, on 

 the contrary , a sample obtained from S с о r e s b y Sund is 

 of much coarser consistency. 



Generally speaking , it will be very difficult to determine 

 what proportion of the ingredients of a sample have been 

 deposited by transport direct from land. We have here 

 two things to go by, namely the size of the grains, and 

 the mineralogicai constitution. With regard to the size of 

 grains, it has already been explained what the size will be in 

 deposits which have been formed exclusively in this way. The 

 other factors which contribute to the formation of the deposits 

 will in most cases produce a very mixed deposit with finer 

 and coarser ingredients intermingled. In almost every case 

 we can, therefore, only be sure that the material conveyed from 

 land will form a certain part of the finer ingredients of the 

 deposit; how great a part, it is often impossible to determine. 

 Ice-transport, for instance, may deposit large quantities of clay 

 at the bottom of the sea, which cannot in any way be disting- 

 uished from other clay; we have therefore hardly any facts to 

 go by except that there must probably be a certain proportion 

 between the finer and the coarser particles in the material conveyed 

 by the ice, this proportion being tolerably persistent every- 

 where. If in a sample very few rock-fragments are found 

 we may suppose that only an insignificant part of the clay in 

 the same sample can have been transporled by ice. But the 

 circumstances are rendered still more complicated by the fact 

 that part of the material may have originated from the bottom 

 of the sea, and this part may consist of all sizes of grains, 

 and we have no means vvhertby we can determine the quan- 

 tities of each of these bei'oreliaiul. Onlv after a close examina- 



