10 



It was unfortunate that during a blizzard our instrument was broken. It should always be taken in 

 before a heavy wind, because the dome for the fan will always get badly choked with drift snow. 1 do 

 not know whether the speed which the fan reaches would in any way affect the readings. If so, this is 

 another matter for consideration. 



CAPE ARMITAGE THERMOMETER. 



This was a minimum thermometer, placed on a stand about 4 feet 6 inches above the snow-level and 

 well clear of all land (see map, p. 16, for position). It was read always once a day, weather permitting, and 

 very often twice and three times, as any Officer going out would make this an excuse for a walk, and would 

 read and reset the thermometer, taking the time he did so. The differences between the readings of this 

 thermometer and those of the minimum in the screen for the same interval were very marked. 



During the Winter months the thermometer was open to the sky, but during the Summer a shade was 

 placed to protect it from the sun's rays (fig. 4). Dr. KOETTUTZ was the originator of this series of 

 observations, and it is mainly due to him that the series remained as unbroken as it is. During the 

 second year Lieutenant MULOCK assisted Dr. KOETTLITZ in taking the observations, and generally recorded 

 two observations in the day. 



Fig. 4. Thermometers at Cape Armitage, looking North (showing the screen placed 

 over the thermometer during the Summer). 



DINES SMALL ANEMOMETER. 



This instrument was set up on the screen, the head being about 10 feet above the level, and the 

 connecting pipes about 5 feet long. During the heavy snow drifts which always accompanied a 

 southerly gale the open head of this instrument was invariably choked, and necessitated clearing every 

 two hours, before observations were taken. The highest and lowest readings were recorded during the 

 time the observer was looking at the instrument. Snow was found not only to collect in the head, but 

 also in the pipes, and the only way to clear them was to take them on board and thaw them out. A leak 

 was observed at the lower glass joint, and the Engineer soldered all round it and so more or less stopped 

 the leak. Pure alcohol was used, coloured by staining matter, and the evaporation of the alcohol 

 appeared to be rapid. The observations taken with the instrument seemed to agree more or less with 

 those of the Eobinson, and more especially the estimates on the Beaufort scale. 



