SEA ICE 183 



to whether the latter phenomenon actually takes place. Certainly plate-like crystals 

 developed considerably below the surface on a dredging line which had been left for 

 some weeks without being used. These tine hexagonal plates were in some cases as 

 much as 6 inches in diameter, and had formed vertically down the dredging line for 

 a distance of several yards. When dredging off the cracks extending westwards 

 from the Penguin Rookery at Cape Eoyds and southwards through Flagstaff Point 

 numbers of these plate-like crystals were brought up by the dredge, though it is 

 possible that they may have formed in super-cooled water upon the cordage of the 

 dredge coming in contact with it. It should be mentioned that in this case the crystals 

 were not attached to the netting of the dredge in the way that the ice-crystal plates 

 were attached to the dredging line just mentioned. It was observed that if the 

 weather was calm the felted ice crystals, which formed with temperatures falling 

 below 28° F., became firmly adherent to one another, and the ice continually in- 

 creased in thickness, although still plastic enough to undulate under the influence 

 of a slight swell. If calm weather continued we observed that fibrous sea ice began 

 to form, while the alternations of rise and fall of temperatures gave rise to a banded 

 structure in the first few layers of the ice, as when the temperature rose the ice 

 ceased to form, and when it fell the layer of ice thickened. As the periods of high 

 and low temperature vary in length, so do the bands in the ice vary in thickness, the 

 latter being from half an inch, or less, up to several inches. A similar banded appear- 

 ance has already been noted as occurring in the ice of some of the lakes, particularly 

 in Green Lake and Coast Lake. A similar reason to that just given has been 

 advanced by us to account for it in the latter instance also. During the formation 

 of the first few inches of sea ice the ice appears transparent, and seems black on 

 account of the deep water seen through it. As the ice becomes thicker it gradually 

 takes on a greenish tinge, and finally becomes opaque. Once the sea ice is thick 

 enough to prevent the under side being seriously affected by changes of air tempera- 

 ture a uniform vertical fibrous structure sets in, without the transverse lamination 

 already described. 



It is in this way that all the rest of the ice is formed. During the growth of 

 the fibrous ice salts still in solution are forced out in the form of brines, the mineral 

 matter of which becomes more and more concentrated as the temperature falls. 

 Such brines are partly forced back into the sea water below, partly forced up to 

 the surface of the sea ice, where they form ice flowers, and partly remain entangled 

 between the crystals. It is suggested that this movement of the brines together 

 with that of the air dissolved in sea water which has been expelled on freezing 

 helped to impart a remarkably fibrous structure to the ice. The brines extruded to 

 the surface of the ice obviously freeze as cryohydrates. Such cryohydrates were 

 also observed by us frozen at the tips of stalactites in caverns in the grounded bergs 

 between Cape Royds and Cape Barne. The icicles dependent from the roofs of 

 these caves were slightly sticky from concentrated brine at normal winter tempera- 



