COMPARISON OF THEORY WITH OBSERVATIONS. 237 



(1) Pressure difference, Cape Ecans-Framkeim,—lt is interesting to notice that the pressure 

 difference remains positive with all kinds of air motion. In other words the pressure on the 

 average is higher at Cape Evans than Framheim in all winds. This result is obvious from 

 the lowest curve on the plates which is practically never below the zero line. The difference, 

 however, is least with high northerly winds and greatest with high southerly winds. This 

 is in a<Jreement with the whole theory which makes the high southerly winds increase djiia- 

 mically any pressure difference existing between the west and east of the Earner. 



(2) Pressure difference, Cape Evans-Cape Adare.— With, northerly winds the pressiu-e at 

 Cape Adare is higher than at Cape Evans, while with southerly winds the reverse is the 

 case. This requires no explanation, for the theoretical diagrams show a high pressure area 

 near Cape Adare during northerly winds and a low pressure area over the Ross Sea during 

 southerly winds. 



(3) Change of pressure difference, Cape Evans-Framheim. — This gives numerically the obvious 

 relationship between the wind and the pressure difference shown on the lowest curves of the 

 plates. Of all the relationships this is the most consistent. High northerly winds are associated 

 with the falling parts of the curve and high southerly winds with the rising parts, hence 

 the change in the pressure difference is negative in the former case and positive in the latter. 

 The chantJe is regular, from — -038" in 4 hours with high northerly winds through calms 

 to +-021" with high southerly winds. 



(4) & (5) Clumge of pressure at Cape Evans and Framheim. — It is interesting to notice 

 that the change in pressure at the two stations is of opposite sign during all winds at Cape 

 Evans. That is, during northerly winds the pressure falls at Cape Evans and rises at 

 Framheim, while during southerly winds it rises at Cape Evans and falls at Framheim. This 

 does not appear to agree with the theoretical curves on figure 656, according to which the 

 blizzards occur while the pressure is rising at both Cape Evans and Framheim, and the 

 northerly winds while it is falling at both stations. Nor does it seem consistent with the 

 fact that there is very little difference in phase between the waves at Cape Evans and Framheim. 

 For the pressure to change in opposite directions at the two stations would necessitate 

 a difference of phase of about 180°, this is obviously not the case as can be seen 

 from ths pres.sure curves on the plates. It has however been pointed out that the 

 change in the wind direction may occur at very different parts of the actual pressure curves 

 according to the shape, size and rate of travel of the various waves. The one thing, however, 

 which does not change is the dynamical pressure difference which always raises the pressure at 

 Cape Evans and depresses it at Framheim. Thus it is this effect which comes to light in the 

 mean of the whole series of observations. 



(6), (7) & (8) Departure of pressure from mean of the month at Cape Evans, Framheim 

 and Cape Adare. — We have here a very unexpected result. With the exception of a small 

 excess at Cape Evans during winds of 11-30 miles an hour from the south, the pressure is 

 below normal at all three stations during both northerly and southerly winds and above 

 normal only during calms. It will also be noticed that the higher the wnds both north and 

 south the greater the defect of the pressure. 



This is a result which would never be suspected from the curves. No indication can be 

 seen that the high winds, both northerly and southerly, are associated more with the troughs 

 of the waves than with the crests,^ nor do the crests as a rule appear to be associated 

 with calms. The only explanation I can offer is that the winds and calms are distributed 

 uniformly over the waves, but when the average pressure about which the waves oscillate 

 is low the air motion is intensified both from the northerly and southerly directions. 



Pressure distributio)i and temperature. — When discussing the mean pressure at Cape Evans 

 and Framheim, (page 173) it seemed a paradox that the station with the lower mean 



