SURGES. 193 



it seems justifiable to conclude that the Antarctic pressure surges increase from Cape Adare 

 to Cape Evans to Framheim. 



Surges affecting the whole Antarctic. — Our next step is to examine the data for 1902, to 

 find whether the same surges affect the pressure at stations so widely separated as McMurdo 

 Sound, Snow Hill and the Gauss Winter Station. 



For this purpose plate III has been prepared showing the simultaneous observations at 

 the three stations in the same way as was used for plate I. 



Surges are very clearly marked on all three curves, and it is now necessary to see if 

 the same surges can be recognised on each. To facilitate the examination, the maxima and 

 minima of the large waves, i.e., of waves of a greater amplitude than -2", have been marked 

 as before on each trace, and we shall confine our attention chiefly to the surges thus marked. 

 We will follow the surges on the Hut Point curve and examine the other curves to see if 

 they show corresponding surges. 



The first surge at Hut Point extends from the middle of March to the middle of April, 

 with the minimum near the beginning of April. There can be no doubt that this surge is 

 present on the other curves, both of which have marked maxima and minima at approxi- 

 mately the same times. 



The next surge on the Hut Point curve extends from the middle of April to the middle 

 of May. This surge is shown quite clearly on the Gauss curve, although the maximum in the 

 middle of May is not marked with a + on account of the subsequent fall being less than 

 ■2". It appears at first sight that this surge is not shown on the Snow Hill curve, on which 

 there is a pronounced minimum at the time the maximum occurred at the other two stations. 

 But the disagreement is only apparent. A local deep depression apparently affected Snow 

 Hill just as the pressure commenced to rise due to the surge. There can be little doubt 

 that if the local depression had not existed the pressure changes would have been as shown 

 by the dotted line. 



The third surge at Hut Point, from the middle of May to the 9th of June, was less 

 intense and it failed to reduce the pressure at the other two stations by -2", hence the 

 minimum is marked only on the Hut Point curve ; but the curves clearly show that the 

 pressure was affected at all three stations to a greater or less extent, and the pressure rose 

 at all three stations to a maximum before the middle of June. 



The next maximum is shown on all three curves in the second half of July. The dip 

 between the two maxima obviously corresponds on all three curves, but the lowest points 

 are somewhat widely separated. This is obviously due to the wave being so shallow that the 

 lowest point is decided rather by the local inequalities superposed on the surge than by the 

 minimum of the surge itself. 



From this point until the minimum near the middle of October there are the same 

 number of marked maxima and minima on each curve, and there can be little doubt that 

 each sign refers to the same surge at each station. 



The rapid rise and fall on the Hut Point curve during the second half of October has 

 its counterpart on the Gauss curve where, however, it was so reduced in intensity that it 

 was less than -2" and therefore not marked. It is questionable whether the maximum 

 and minimum at the beginning of November on the Snow Hill curve are due to the same 

 surge. 



The curves come into line again at the maximum in the second half of November, and 

 run parallel to one another until the observations cease. 



S5 



