276 RESULTS OF THE METEOROLOGICAL OBSERVATIONS TAKEN DURING 



Tlie Ijfirometric readings are reduced to 32° and sea-level, and the wind direction is 

 true. The observations are not quite synchronous at the two positions, which differ 

 nearly 10° of longitude, but the purely local effect of the "Fohn" is well marked. 

 That the phenomenon is quite local and not associated in any way with the temperature 

 of surrounding areas, is shown by the circumstance that as soon as the wind changed to 

 NW. by N. the temperature invariably fell. A few notes made on the da}' above 

 mentioned indicate the restricted area over which the "Fohn" effect prevailed. At 

 5 P.M. Cape Bennett, twelve miles west of the station, and the horizon beyond, were quite 

 clear, but in the direction of Saddle Island, some seven miles due north, the view was 

 restricted to above five miles, there being much stratus fog and haze. This shows the 

 extent to which the dry " Fohn " wind extended in a north-west direction. At this time 

 the upper sky was three-fourths covered with beautiful cirrus and cirro-cumulus clouds 

 travelling very quickly from the north-west. 



An increase of tenqaerature with height was not uncommon under anti-cyclonic 

 conditions. Thus on 19th August 1904 the temperature at 11 a.m., at a height of 896 

 feet, was 20°'9, while at sea-level at the same time it was 10°"4. Many such inversions 

 of the normal conditions were observed, and it is very probable that during anti-cyclones 

 the large inversions of temperature observed in Europe would manifest themselves. 

 Unfortunately the kite experiments were not a success, and the ascent of lofty peaks 

 was out of the question owing to crevasses and hanging glaciers, so no data were 

 obtained from heights exceeding 1400 feet. 



Thermal and other Windroses. 



In the calculation of the thermal and other windroses, the whole series of hourly 

 observations made during the wintering of the " Scotia " in Scotia Bay has been utilised. 

 It was not considered expedient to make use of the observations taken from 1st November 

 1903 at Omond House, as the wind circulation there, under the conditions already 

 referred to, differs considerably from the general mean for the locality. In dealing 

 with the observations, the values were tabulated to sixteen directions, N., NNW., NW., 

 and so on ; another column being added for calms. The values given in the tables 



