148 . ANTARCTIC TERM-HOUR RECORDS. 



scale of ordinates is as in the originals, 1 mm. representing l'-125 in Declination, 4'53y in Horizontal 

 Force, and 3 - 18y in Vertical Force. The slow-run Christchurch time scale is practically the same as at 

 Mauritius and Kew, but the quick-run scale was more open. This was reduced by the tracer so as to be 

 approximately the same as in the Mauritius curves. 



On July 1, trace was lacking for about a minute near the beginning of the term hour. 

 Of the curves themselves it is perhaps sufficient to say that it requires a very acute eye to detect their 

 difference from straight lines. 



49. In the Antarctic, quick-run curves were obtained for the whole of the term hour on June 15, 1902 

 (9-10 a.m., G.M.T.), for the greater part of the term hour on July 1, 1902 (10-11 a.m., G.M.T.), and for 

 a short portion of the term hour on July 15, 1902. 



Plate IV shows the records obtained on June 15. They were copied from the originals by Mr. BROOKES 

 with the Schmidt curve tracer, leaving the scale of ordinates unaltered, but reducing the time scale by 

 about 25 per cent., so as to make it approximately the same as in Plates II and III. D D , H H , V V 

 are the base lines for the Declination curve DD, the Horizontal-Force curve HH, and the temperature 

 curve TT respectively. The Vertical-Force magnet was out of action during the hour in question. 

 Vertical lines are drawn answering to 10-minute intervals, the times from the commencement of the hour 

 being marked on the Declination base line. The absolute values answering to the base lines and the 

 scales for the Declination and Horizontal-Force curves are shown on both margins. Fine horizontal lines 

 are drawn to aid the eye in estimating the extent of the movements, and arrows assist in indicating the 

 direction in which the elements increase. 



The alteration in the elements is sufficient to make the open time scale of real value. At the same time 

 it should be noticed that the Horizontal-Force magnet was unusually sensitive, so that 1 mm. of ordinate 

 answers to only 1 - 4y. In the case of the Declination the scale is less open, 1 mm. answering to l' - 5 in 

 force about 2 9y and the change during the hour in Declination is considerably more important than that 

 in Horizontal Force. 



Plate V shows the records obtained in the Antarctic during the term hour of July 1. They were 

 treated in the same way as those for June 15. In this case there is a record VV from the Vertical-Force 

 magnet. 



There are two gaps in the curves. The first, occurring from about 3 to 4^ minutes after the commence- 

 ment of the hour, was due to the light being intercepted by the clamp attaching the paper to the drum. 

 The second, occurring from 26 to 30 minutes after the commencement of the hour, represents the interval 

 occupied in changing papers. The Declination and Horizontal-Force scales have the same value as for 

 June 15. The changes in these elements are somewhat larger than on the previous term day. The 

 departure of the V trace from a straight line is not apparent to the eye. The changes in this element 

 were really less than for the other two, but the difference is exaggerated owing to the fact that the 

 Horizontal-Force scale is almost ten times as open as that of the Vertical Force, for which 1 mm. EE 13'5y. 

 50. Plates VI to X show the Antarctic records during the remainder of the term hours. April 1 was the 

 earliest term day for which a satisfactory record was obtained. The curves appear in chronological order, 

 omitting, of course, the two days already considered, and likewise the two May term-day hours for which 

 no record existed. The date is shown at the foot, and marks answering to the beginning and end of the 

 term hour appear on all the base lines. The hours (G.M.T.) are recorded on one only of the base lines. As 

 before, the base lines are distinguished by the suffix 0, and the letters D, H, V, and T indicate Declination, 

 Horizontal Force, Vertical Force, and temperature. The absolute values answering to the base lines and 

 the scale values are shown on the left margin, and arrows assist in indicating the directions in which 

 the elements increase (Vertical Force up the sheet, Declination and Horizontal Force down the sheet). 

 To show better the general trend of the curve, the portions of curve reproduced represent in most cases 

 about two hours, including half-an-hour before and half-an-hour after the true term hour. There are, 

 however, exceptions to this on January 15, 1903, when the original record stopped before the end of the 

 term hour, and on July 15, 1902. On the last-mentioned date the trace is restricted to the exact term 

 hour 11 a.m. to noon, G.M.T. It consists of three portions. Starting at 11 a.m., we have a quick run 

 covering about 1 minutes, then a gap corresponding to a minute, due to the clamp intercepting the light, 



