475 



it is problematical to what extent the head of a nail driven into ice may be considered as a permanent 

 level relatively to the ice. Most certainly with the prevalence of sunshine, and a temperature not too far 

 below the freezing point, it would not bo permanent, for any small black object which does not float on 

 water sinks rapidly into ice under the influence of sunshine, even when the temperature is below the 

 freezing point, the ice thawing underneath under the influence of radiation, but in the actual Antarctic 

 winter this cause could not come into play. 



The general evidence seems to show that the air was very dry. It must not be forgotten that for some 

 unexplained cause the temperature at the " Discovery " was higher than at other places in the immediate 

 neighbourhood ; and greater dry ness generally accompanies greater heat, but, notwithstanding this, it 

 may, I think, be safely asserted that a low relative humidity prevailed generally in the district, and this 

 conclusion is supported by the fact of the intense solar radiation shown by the temperatures of the black 

 bulb in vacua. In the free atmosphere the humidity is an unfailing test of the vertical circulation, a low 

 humidity indicating a descending current and a high humidity an ascending one. Indeed, were it not 

 for the formation of rain and the accompanying extraction of water vapour the laws of diffusion of gases 

 must lead to the nearly complete saturation of the whole atmosphere. The humidity of the lower strata 

 must be partly dependent on the dryness or wetness of the earth's surface, but not entirely, for it is 

 notorious that the sea coast of England in the winter is dryer than the inland districts. Also the great 

 deserts of the earth are situated in the anticyclonic belts, and are the consequence rather than the 

 cause of the dry air that lies over them. It seems likely, therefore, that the Antarctic region, or at 

 least that part which lies in the neighbourhood of Boss' Bay, is an anticyclonic region, and that there is a 

 steady settling down of air. If this point can be established for the Antarctic regions as a whole, it is of 

 great importance in regard to the question of the general circulation of the atmosphere. 



There is one other matter of great interest in connection with these observations, and that is the 

 formation of the Great Barrier. The temperature observations show that the loss of ice or snow from the 

 surface by thawing is a negligible one, and the question is simply whether the snowfall exceeds the 

 evaporation, or vice versd. If the snowfall be the greater the Barrier should be increasing in thickness. 

 Unfortunately, precise observations on precipitation are impossible in such a climate, and in any case two 

 years would not suffice to give a reliable average either of the snowfall or the evaporation, even if we had 

 precise and exact observations of each. Hence nothing more than a guess can be made, but the 

 evaporation, as judged by the winter observations, seems sufficiently large to more than balance the 

 precipitation, and thus lead to a gradual decrease in the amount of ice and snow. 



3 P 2 



