TBANSACTIOXS OF SECTION E. 277 



It will be observed that the Nvater-temperatures fi-om which 

 these observations are made are mostly taken from about the 

 latitude of 55° S., and also it maybe noted that no ice has been 

 seen in the course of the last seven years by observers to the 

 eastward of 117° W. long. Now, the principal ice-drift from 

 the Antarctic Circle is undoubtedly to the eastward of Cape 

 Horn, where by far the greater quantity of ice is always 

 encountered by sailing vessels ; and, indeed, it may perhaps be 

 assumed that Humboldt's current expends itself principally in 

 the South Atlantic, and that this portion of it existing between 

 long. 150° and 120° W. is but an of!"shoot. Furtheruiore, 

 although this current (Humboldt's) is by no means a sluggish 

 one, it is not considered by Maury to be infested with ice. He 

 ■ concludes that the launching-place of Antarctic ice is too far to 

 the eastward to trouble the route much between New Zealand 

 . and Cape Horn, for he says it is rare to sec an iceberg in Hum- 

 boldt's current so far from the Pole as the parallel of 55° S. 



Mr. Towson, some fifty years ago, from the best data then 

 at his disposal, calculated that the drift of an iceberg was, 

 under ordinary conditions, towards an east-north-east direc- 

 tion, and its rate of progression was about twelve miles per day. 

 This, of course, is a difficult matter to define, depending, as it 

 must necessarily do, upon current principally ; for when the 

 bulk of an iceberg below water is considered, with the small 

 proportion of it shown above and exposed to the wind — ;about 

 one-eighth — it will be at once seen that all drift to the nortli- 

 ward must be accounted for by current, though its progress to 

 the eastward may perhaps be accelerated by the prevailing 

 westerly winds of the vicinity. 



Assuming, therefore, that the foregoing premisses are cor- 

 rect, we have the fact that this ice-bearing current, no matter 

 -what it may be called, is setting to the northward of east, that 

 icebergs are fairly confined W'ithin its limits, and that the 

 further north it reaches the more the ice is scattered, until it is 

 finally dissolved in a low latitude. That it finally recurves to 

 the north-west may perhaps be demonstrated, and that it is 

 clear that the great mass of ice taken up to the eastward of 

 Cape Horn is taken there by an agency other than the current 

 now under discussion ; for ice, having once attained the latitude 

 of 55° S. in the middle of the South Pacific, can never be carried 

 round Cape Horn. 



There have been many theories started, in connection with 

 ice-navigation, as to the value of the thermometer as a means 

 for its detection, or indicating its immediate vicinity ; and these 

 theories, like many others, are some of them capable of inter- 

 Ijretation either way. For example, the question may be asked, 

 Will the temperature of the w^ater denote the presence of ice '? 

 The answer may be unhesitatingly " No." An iceberg may be 



