THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 293 



Only on one occasion was force 10 exceeded, viz. ou 12th July 1903, when at 8 p.m. 

 a uorthei'ly gale of force 1 1 was recorded. 



It will be seen that the most frequent gales are recorded from the S.W., while at no 

 hour of observation during the two years was force 8 reached with winds from the 

 S.S.E., S.E., or E.N.E. Taking the values by seasons, we find that 28 per cent, of all 

 gales occurred in Spring, 29 per cent, in Summer, while the values for Autumn and 

 Winter were 22 per cent, and 21 per cent, respectively. The maximum percentage was 

 14 in November, and the least 3 in April and October. 



Miscellaneous Phenomkna. 



Fog. — On the average, fog occurs ou 36 days of the year, the total number of cases 

 recorded in the two years being 71. Of these cases Summer and Winter had 21 each, 

 Spring 17, and Autumn 12. An examination of the direction of the wind during the 

 hours that fog was recorded showed that in nearly all cases the direction was either 

 N., N.N.E., or N.E., and that the earlj^ morning was the foggiest part of the day. 



Thunderstorm!^. — Thunderstorms were recorded on 30th March and 27th April 1903 

 and 9th January 1904, while lightning without thunder was seen on 3rd April 1903 

 and 3rd February 1904. 



Lunar Halos were seen on six occasions in 1903 and 8 times in 1904, while Lunar 

 Coronas were observed 11 times in 1903 and 7 times in 1904. A mock moon was 

 recorded on 4th I\Inrch 1904. 



On the Relation of thk Wind Direction to Pressure, Temperature, 

 Wind Force, and Amount of Cloud. 



In order to ascertain the effect of wind direction upon the meteorological elements 

 the observations for the year 1903 have been tabulated according to the direction of 

 the wind prevailing at the time, the values thus treated being those of pressure, 

 temperature, wind force, and amount of cloud. The observations were first tabulated 

 to 16 points, N., N.N.W., N.W., and so on, and the values resolved to the eight 

 principal directions by halving N.N.W., — for example, between N. and N.W., W.N.W., 

 between W. and N.W., — the values in this way being resolved to the octagonal directions. 

 On looking at the monthly tables (not printed) it was seen that the almost complete 

 absence of S.E. and E. winds during the year under review prevented values being 

 given in several of the months, and even as regards the seasonal and annual means the 

 number of observations was too small to give a true mean for the winds referred to. 

 In order to obtain a closer approximation to the true mean prevailing with winds from 

 the S.E. and E. the following method was adopted. The data observed with S.S.E., 

 S.E., E.S.E., and K, E.N.E. winds for the years 1903 and 1904 were extracted, and 

 the values resolved to the two' directions, viz. S.E. and E., as described above. Then 

 the seasonal means for each of the years were taken and the mean of the two struck, 



