682 ME. E. J. GARWOOD ON THE [NoV. 1 899, 



The other tract lies immediately to the west of this, and occupies 

 the greater part of King James's Land ; they are separated one 

 from the other by Dickson Bay and Wijde Bay and by the depressed 

 area which connects these inlets (see Map, PL XLI). 



The expression 'inland ice' is a vague geographical term, useful 

 perhaps when applied generally to a country in which land-ice 

 predominates, but which has been neither mapped in detail nor 

 differentiated into separate glacier-systems. Por strict accuracy, 

 however, the terms ' ice-sheet ' and ' glacier ' convey a more definite 

 meaning. Though I am not aware that an ice -sheet has ever 

 been scientifically defined, most writers appear to use it in the 

 sense of a complete covering of ice, radiating from a watershed 

 consisting of snow and ice, and not confined or directed by visible 

 barriers of rock, while a gl'acier, though it may originate in an 

 ice-sheet, occupies a definite valley. As thus defined, a small 

 ice-sheet does exist on each of the above areas, the radiating- 

 point and chief gathering-ground being situated somewhat north- 

 west of the centres of each area, with supplementary radiating-points 

 to the north and east. If we accept this view, the rock-ridge of Mount 

 Chydenius, in the eastern tract, and the group of peaks including 

 the Three Crowns protruding above the ice in the centre of King 

 James's Land, may be regarded as groups of nunatakkr, though, in 

 the latter ease, the system is extremely complicated (see panorama, 

 PI. XLII). There can be no doubt, however, that, whether this appli- 

 cation of the term ' ice-sheet ' is accepted or not, the efi'ect produced 

 is practically the same as that which would result from the presence 

 of a much more extensive sheet. We have here a district from 

 which the ice-cap, which once buried it more deeply, is gradually 

 melting off: the melting has not, however, yet progressed sufficiently 

 to have reached the Alpine-glacier stage, and we still have un- 

 doubtedly the eftects of an ice-sheet with its valley-bound ground- 

 ice and freely moving surface-layers, with the result that the latter 

 travel in an independent direction, frequently at right angles to 

 the former. The district has, however, reached a stage in its history 

 when the original central surface radiating-point has been replaced by 

 several separate decentralized points by the lowering of the ice-surface 

 and the obstruction ofi'ered by a constantly increasing number of 

 nunatakkr.^ Here is in fact a condition of things almost identical 

 with that which obtained in our own country, on the land-ice 

 hypothesis, during the Glacial Epoch. As the cold conditions 

 passed away and the ice-sheet covering Britain melted off, the 

 radiating-point must have been replaced by decentralized points in a 

 similar manner ; and this fact is perhaps not sufficiently borne in 

 mind in accounting for such phenomena as the intercrossing of the 

 erratics in many of our own glaciated districts. 



^ A careful examination of the nunatakkr met with during the expedition 

 across these ice-sheets revealed nothing unexpected as regards their flora. 

 The plants all belong to the more universally distributed species collected on 

 the coast and in the ice-free valleys, such as Baxifraga ojp]^ositifolia, Papaver 

 nudicaule, Bryas octopetala, etc. 



