152 PULSATIONS. 



with certainty. There was, perhaps, only one slow-run trace on which they were distinctly recognisable, 

 that of August 6, 1902. On that day pulsations were distinctly visible in both the D and H curves 

 during most of the time between 8 a.m. and 3.30 p.m. They were distinct enough at times, the day 

 being unusually free from large disturbances, to admit of an approximate determination of the period from 

 both the D and H curves. The results obtained showed no certain variation in the period, the mean 

 obtained being 1 9 minutes, i.e. there were about 32 complete pulsations in the hour. 



Several quick-run curves showed pulsations of short period. On June 16, 1902, pulsations were visible 

 in the D and H curves between 8.30 and 8.40 p.m., having a period of about 27 seconds, and again 

 with about the same period from 10.30 to 10.40 p.m. On August 15, 1902, pulsations appeared at 

 intervals in both the D and H curves from 11.30 a.m. to 2.10 p.m. For 8 minutes, from 11.40 to 

 11.48 a.m., they had quite a considerable size, the maximum value of the double amplitude being at least 

 8 mm. (i.e. 12') in D and 6 mm. (8 - 4y) in H. During most of the time the amplitude was no larger than 

 in Plate XII. The periods derived from the D and H curves were the same, about 7 seconds, and appeared 

 independent of whether the amplitude were large or small. This, of course, strongly supports the view 

 that the largeness of the amplitudes between 11.40 and 11.48 a.m. was due to natural as opposed to 

 instrumental causes. 



On August 16, 1902, exceedingly small but very regular pulsations with a period of 7 or 8 seconds 

 could be made out in the H curve from noon to 1 p.m. At one or two points the edge of the D curve 

 seemed also serrated, but the movements were too small to be certain of. 



Pulsations were also visible in the H curves of September 1, 1902, from 2.35 to 4.20 p.m., and on 

 those of September 16, 1902, from 10.5 to 11.0 p.m. On both occasions the period was about 8 seconds. 

 On September 1 the D curve also showed a trace of pulsations. On September 16 there seemed a faint 

 indication of minute pulsations in the H curve later in the day between 3.0 and 3.20 p.m. 



The above are the only cases in which pulsations were detected. As to whether they are more or less 

 numerous in the Antarctic than elsewhere it would be impossible to say without a very minute investi- 

 gation. It was really only when the maghetographs were quick run that pulsations were at all likely to 

 be detected. The number of hours' trace in the Antarctic from quick-run curves was approximately 130, 

 but this represents but a very small fraction of the time during which the magnetographs were in action. 



54. The curves in Plate XIII represent one of the quietest times, if not the very quietest time 

 experienced in the Antarctic. The time represented is from 11 p.m. on June 27, 1903, to 9 a.m. on June 28. 

 The slight movement that is apparent in the V curve is mainly due to temperature, there being a slight 

 rise of temperature towards the end of the period. There were only 2 or 3 days' curves which even 

 distantly approached those shown for quietness. Undisturbed as the D and H curves on June 27-28 are, 

 when considered relatively to other Antarctic curves, they display a practically continual succession of 

 small movements such as would not appear on an ordinary quiet day at Kew. The curves at Christchurch 

 corresponding to those in Plate XIII were also very quiet. At Kew the time was also distinctly quiet, 

 but not more so than on an average quiet day. 



In the evening of the same day there was a large disturbance in the Antarctic, which is described later. 



