ICEBERGS IN THE SOUTHERN OCEAN. 223 
concurrently southerly winds had asserted themselves. In Janu- 
ary 1897, twenty ships reported ice, and the charts shew strong 
southerly winds, and during February and March icebergs and 
southerly winds were concurrent, but from the end of March to 
the end of September 1897 strong north-west winds have been 
prevalent, and only one ship, in September, has reported ice. 
This is too short an experience to settle the question, but so far 
as the records go, we find that when there is a prevalence of north- 
west winds no ice is reported, and with southerly winds plenty of 
ice is reported. The fact that these icebergs are about 3000 mlies 
distant from Australasia where the winds were observed, must 
not be overlooked, and therefore the experience just given may 
shew an accidental relation between the position of the icebergs 
and the direction of winds. I do not however think so, because 
some years since, I investigated the winds between the Cape and 
Australia, and found that the atmosphere as a whole, was moving 
to the eastward at the rate of about five hundred miles per day, 
carrying storms and change of wind with it; so that a storm, in 
the iceberg area travels to Australia in six or seven days. The 
probability that the iceberg area has the same winds that we 
have in Australia, only a few days earlier, is very strong indeed. 
T have no doubt that winds of the same general character effect 
the two places, Australia and the iceberg area shewn on the map 
herewith. 
We must have a longer experience before it can be considered 
proven. Meantime it would materially aid the proof, if any 
vessels sighting icebergs on a fine day with strong northerly or 
southerly winds, would stop the engines and watch the berg care- 
fully for three or four hours to see if it does move with the wind. 
As soon as the motion with the wind is definitely determined by 
actual observation of the bergs, it will be possible by careful study 
of the winds in South Africa and Australia to forecast the positions 
of icebergs between Africa and Australia with some git of 
exactness, 
