DAILY VARIATION— SEPTEMBER TO MARCH. 



59 



Comparison of the Daily Temperature Variation during September to March at McMurdo Sound 



and at other Stations in High Latitudes. 



So far as I am aware there is only one other set of temperature observations in high 

 latitudes so complete as these at McMurdo Sound with which to make comparison. The 

 observations made during the drift of the Fram in a mean latitude of between 82° and 

 83° N. give the daily temperature variation for each month from a homogeneous series of 

 observations extending over three years. For comparison the curves given in Professor Mohn's 

 masterly discussion of the meteorology of that expedition have been added to figure 20 as 

 they show marked and important differences from those for McMurdo Sound. For purposes 

 of comparison the northern observations have been plotted against those of the corresponding 

 months in the south ; thus January in the north is plotted against July in the south and 

 so on. 



The first striking difierence is in the time of year at which the daily variation has its 

 greatest amplitude. In McMurdo Sound the amplitude increased slowly from the return of 

 the sun on August 24:th until January (neglecting for the moment the decrease in December), 

 while in the north the amplitude increased more rapidly and the greatest amplitude was 

 reached as early as the second month after the return of the sun. From this month, April 

 (corresponding to October in the south), the amplitude in the north decreased rapidly to a 

 very small value at midsummer, after which there was a temporary rise during the last 

 two months, August and September, before the sun set for the winter. 



The contrast in this respect shown between the McMurdo Sound and Fram data is found 

 to be present when other stations are investigated. There appear to be two types of polar 

 climate. In one, of which the Fram conditions are typical, the daily amplitude of tem- 

 perature has its maximum in one of the spring months, while in the other the maximum 

 daily amplitude is found in a summer month. The following two tables give examples of the 

 two types : — 



Table 23. 



' Fram ' Type. 



*TUo Gauss diily a.npl.tulo for Mirch is g;veu as 3-S2^C. =ti-9°F. in D.-ut. Su^pjlar Expedition, Meteorc 

 logy, Vol. 1, part 1, page 53. This, however, is much too larg;. During the first .sixteen days of March the 

 teinparature observations ware taken on b:)ard the Gauss, and during twelve of the remaining days in a Stevenson 

 screen on the ice n3ar the ship. CilcuUtiug the amplitude for the 16 days on which the observations were made 

 on the ship and for tha 12 on the ice wa obtain the following: Ship 4-71=C., Ice 2 COX'. As the ship's 

 observations obviously give too large an amplitude and twelve days are not sufficient to give a reliable value, 

 I have not included thi.s month in my discussion. 



