LIBRARY OF THE UNIVERSITY OF CALIFORNIA. RECEIVED BY EXCHANGE Class Frontispiece. Enlargement of Localit of Winter Quarters SKETCH MAP TO snow APPROXIMATE POSITIONS OF SOUTH MAGNETIC POLE. A. " Discovery " Expedition. B. Lieut. SHACKLETON'S Expedition. C. " Southern Cross " Expedition. NATIONAL ANTARCTIC EXPEDITION 1901-1904 MAGNETIC OBSERVATIONS PREPARED UNDER THE SUPERINTENDENCE OF THE ROYAL SOCIETY F THE UNIVERSITY LONDON : PUBLISHED BY THE ROYAL SOCIETY 1909 - [ iii ] CONTENTS. Page Preface, by Sir A. GEIKIE, K.C.B., Pres. E.S vii Introductory Note, by L. C. BERNACCHI . 1 Historical Note, by Dr. C. CHREE, F.R.S 5 Explanation of Tables of Hourly Values 7 Tables of Hourly Values 8 Table of Days made use of for Diurnal Inequalities 71 DISCUSSION OF THE OBSERVATIONS, BY DR. C. CHREE, F.R.S. CHAPTER I. INSTRUMENTS AND RECORDS. Sections Page 1-5. Instruments 73 6, 7. Scale Values of Curves 75 8, 9. Temperature Coefficient of Vertical-Force Magnet . 76 CHAPTER II. BASE-LINE VALUES. ANNUAL INEQUALITY. SECULAR CHANGE. 10, 11. Declination and Horizontal-Force Base-Line Values 78 12, 13. Mean Monthly Values 80 14. ,, ,, Secular Change 81 15, 16. ,, ,, ,, Annual Inequality . 81 17-19. Vertical-Force Base-Line Values and Temperature Corrections 83 CHAPTER III. DIURNAL INEQUALITIES AND THEIR DISCUSSION. 20-22. Diurnal Inequalities of Declination, Horizontal Force and Vertical Force 87 23-24. Explanation of Diurnal-Inequality Tables and Curves 89 Diurnal-Inequality Tables XII-XXII 90 25 Non-Cyclic Changes 100 26-28. Discussion of Declination, Horizontal Force, and Vertical-Force Inequalities 101 29. Diurnal Inequality of Inclination 102 30. ,, South and West Components 103 Figures 1-13, illustrating Diurnal Inequalities 104 CHAPTER IV. DIURNAL INEQUALITIES. FOURIER COEFFICIENTS. 31, 32. Fourier a and b Coefficients 117 33-35. c and a. Coefficients 121 a 2 189805 CHAPTER V. Sections Page 36, 37. ANNUAL VARIATION. FOURIER COEFFICIENTS 129 CHAPTER VI. ABSOLUTE DAILY RANGES. DAILY MAXIMA AND MINIMA. 38-43. Absolute Daily Ranges of Declination, Horizontal Force and Vertical Force 131 44, 45. Incidence of Daily Maxima and Minima 138 CHAPTER VII. TERM-HOUR AND SPECIAL RECORDS. . 46, 47. Term-Hour Records from Mauritius 146 48. Christchurch 147 49-51. ,, Antarctic Winter Quarters 148 52. Antarctic Record during Solar Eclipse 151 53. Pulsations 151 54. Antarctic Record on exceptionally Quiet Day 152 CHAPTER VIII. RECORDS OF DISTURBANCES FROM CO-OPERATING STATIONS AND WINTER QUARTERS. 55-62. Comparative Records from Falmouth, Colaba, Mauritius, and Christchurch 153 63-76. Antarctic Records of Disturbances 158 77-84. Various Representative Types of Antarctic Disturbances 166 CHAPTER IX. MATHEMATICAL ANALYSIS OF DISTURBANCES, AND THEIR DISCUSSION. 85. General Discussion of Theories 170 86-88. Case of Sphere of Magnetic Material in a Uniform Field 170 89, 90. Definition and Nature of Disturbances 173 91-101. Analysis of Selected Disturbances recorded at the Co-operating Stations and at Winter Quarters 175 CHAPTER X. 102-108. Special Type of Disturbance 186 CHAPTER XI. 109-114. Comparison of Antarctic Magnetic Disturbances and Aurora 193 APPENDIX A. Christchurch Term-Day Observations 201 APPENDIX B. An Examination of Antarctic Disturbances from October, 1902, to March, 1903, which are simultaneous with Arctic Disturbances discussed by Prof. KR. BIRKELAND . . 246 ILLUSTRATIONS. Map of South Victoria Land (frontispiece). Sketch of magnetographs as mounted (p. 3). Figs. 1-13. Declination and other curves and diagrams (pp. 104-116). Plates (at end of volume) Plate I. Term Hours, quick run, at Mauritius. ,, II. slow run, at Mauritius. Ill . June 15, and July 1, 1902, at Christchurch. IV. Hour, June 15, 1902, at Winter Quarters. V. July 1, 1902, at Winter Quarters. VI. Hours, April 1 and 15, and June 1, 1902, at Winter Quarters. VII. July 15, and August 1 and 15, 1902, at Winter Quarters. VIII. September 1 and 15, and October 1 and 15, 1902, at Winter Quarters. IX. November 1 and 15, and December 1 and 15, 1902, at Winter Quarters. X. January 1 and 15, and February 1 and 15, 1903, at Winter Quarters. XI. Eecord during Solar Eclipse at Winter Quarters (September 21, 1903). XII." Pulsations " (February 2, 1903). XIII Record during especially Quiet Day (June 27 and 28, 1903). XIV. Disturbance of May 8, 1902, at Co-operating Stations. XV. August 20, and November 6 and 24, 1902, at Co-operating Stations. XVI. April 5 and 6, 1903, at Co-operating Stations. XVII. June 19, and July 26, 1903, at Christchurch. XVIIL August 22, 1903, at Christchurch. XIX. August 25 and 26, 1903, at Co-operating Stations. XX. December 13, 1903, at Falmouth. XXL December 13, 1903, at Colaba and Mauritius. XXII. May 8 and 9, 1902, at Winter Quarters. XXIII. August 21, 1902, at Winter Quarters. XXIV. November 7, 1902, at Winter Quarters. XXV. November 24, 1902, at Winter Quarters. XXVI. April 6, 1903, at Winter Quarters. XXVII. June 19, 1903, at Winter Quarters. XXVIII. June 28, 1903, at Winter Quarters. XXIX. June 29, 1903, at Winter Quarters. XXX. July 26, 1903, at Winter Quarters. XXXI. August 17, 1903, at Winter Quarters. XXXIL August 22 and 23, 1903, at Winter Quarters. XXXIII. August 26, 1903, at Winter Quarters. XXXIV. August 26, 1903, at Winter Quarters. XXXV. October 12 and 13, 1903, at Winter Quarters. XXXVI. December 13 and 14, 1903, at Winter Quarters. Plates XXXVlI to XLIII. Representatives of different Types of Disturbance at Winter Quarters. [ vii ] PREFACE. THE volume of "Physical Observations " of the National Antarctic Expedition, 1901-1904, under Captain R. F. SCOTT, E.N., which was published in the summer of 1908, included a report on one portion of the Magnetic work. This Report consisted mainly of a reduction of the absolute and relative observations by Commander L. W. P. CHETWYND, R.N. It contained, also, an account by Mr. BERNACCHI of the Observatory site in McMurdo Sound, of the geological features of the district, of the instruments employed, and of other matters, likewise Tables of the Hourly Values of Declination, Horizontal Force and Vertical Force on term-days at different observatories during 1902-1903, and a report by Mr. R. C. MOSSMAN and Dr. CHREE on the Magnetic Observations of the " Scotia," under Dr. W. S. BRUCE. The present volume, with its detailed Tables of Hourly Values and its exhaustive discussion of the Observations, completes the presentation of the Magnetic work of the " Discovery." The Royal Society, in arranging for this investigation, was fortunate, with the sanction and co-operation of Dr. GLAZEBROOK, F.R.S., Director of the National Physical Laboratory, in obtaining the invaluable services of Dr. CHARLES CHREE, F.R.S., of the Kew Observatory. His pre-eminent qualifications for the onerous task imposed upon him, and the unwearied zeal with which he has prosecuted it to the end, give to this portion of the Reports of the Expedition a special scientific interest and importance. Dr. CHREE has mentioned in his Historical Note (p. 5) the various Institutions and individuals who have contributed their services towards the preparation of the materials of this volume. To all of them the thanks of the Royal Society are due. Special allusion, however, should here be made to the assistance generously given by Mr. BERNACCHI, who was in charge of the Magnetographs of the Expedition. Not only has he taken a large share in the tabulation of the magnetic curves, but he has been always ready to help during the preparation of this Report. A reference is also called for to the great zeal with which, as Dr. CHREE has heartily acknowledged, the successive Directors of the Christchurch Observatory, New Zealand, Dr. COLERIDGE FARR and Mr. H. F. SKEY, entered into the co-operative scheme of observations suggested by the Royal Society, and to the interest and importance of the information which they were so good as to transmit. Magnetic Stations in the Southern Hemisphere are so few in number that it was particularly useful to obtain so ample a record of observations from Christchurch, which is the nearest observatory to the Antarctic Winter Quarters of the " Discovery." From the map, which forms the frontispiece, it will be seen how closely the positions agree which have been assigned by three successive recent expeditions to the South Magnetic Pole. The position found by the "Southern Cross" was about Lat. 72 40' S., Long. 152 30' E. That obtained from the observations of the "Discovery" was Lat. 72 51' S., Long. 156 25' E. ("Physical Observations" of National Antarctic Expedition, 1908, p. 156). Lieutenant SHACKLETON has been so good as to furnish the exact position found by him, which is Lat. 72 25' S., Long. 155 16' E. ARCH. GEIKIE. Royal Society, Burlington House, September,. 1909. HE UNIVERSITY OF INTRODUCTORY NOTE. BY L. C. BERNACCHI, F.R.G.S. THE self-recording Magnetographs supplied to the National Antarctic Expedition were of the delicate transportable type devised by the late Professor VON ESCHENHAGEN and made by the firm of 0. TOEPFER, of Potsdam, Germany. They were received at the National Physical Laboratory, Surrey, England, in August, 1901, after the departure of the " Discovery." The set consisted of three instruments, viz., a Declinometer, a Horizontal Force Magnetometer, and a Vertical-Force Magnetometer, with a self-recording photographic apparatus. No instructions were supplied by the makers, and as they were of a type absolutely strange in this country, some little difficulty was at first experienced in erecting and adjusting the instruments. Dr. HARKER and Mr. F. E. SMITH, of the National Physical Laboratory, succeeded in adjusting them satisfactorily, and had them recording for a few days before my departure from England by mail boat to join the " Discovery " in New Zealand. The Horizontal-Force instrument proved to be considerably more sensitive than the Kew form of differential magnetometers a difference which was due to the fineness of the quartz suspension fibre employed. Unfortunately, it was not discovered until too late that the boxes of spare quartz-fibres supplied contained even finer threads than that already in the instrument. On arriving at Melbourne, at the end of October, 1901, the instruments were taken to the Government Observatory and erected on a wood bench in a cellar. From November 1st to November 10th they worked in a satisfactory manner. They were subsequently conveyed to the "Discovery" at Lyttelton, New Zealand, carefully secured in my cabin, and were not again removed until the ship reached her Winter Quarters in McMurdo Sound in February of 1902. Winter Quarters. The Winter Quarters were situated in latitude 77 50' 50" S., longitude 166 44' 45'' E. of Greenwich, and to the south of a narrow peninsula extending in a south-west direction from the base of an island formed by Mounts Erebus and Terror. The " Discovery " remained frozen-up in her Winter Quarters from February, 1902, until February, 1904. Observatory Site. The spot selected for the Observatory, although the best available, was hardly an ideal one for magnetic observations. From a magnetic point of view, an observatory of this kind should be placed in a position undisturbed by the presence of magnetic rocks ; but it would be difficult to find such an undisturbed locality in Victoria Land, unless it were on the surface and near the seaward edge of one of the extensive ice-floes, far from the actual coast line, such as the Great Ice Barrier. Geological Formation. A description of the Geological Formation in the neighbourhood of the station will be found in " Physical Observations," pp. 129-130. Observations for Local Attraction. A comparison of the results of absolute observations made in the Magnetic Hut with some taken on the ice in McMurdo Sound at a considerable distance from land will be found in " Physical Observations," p. 134. The site selected for the houses was a low and fairly level piece of rocky ground close to the extremity of the peninsula, arid at a distance of about 300 yards from the ship (see Frontispiece). The peninsula is about 10 miles long by a mile broad, and has an average height of 600 to 700 feet, although the extremity where the Observatories were placed was only 30 feet above the mean sea level. Observation Houses. The Observation Houses were constructed of large asbestos slates, screwed on to the outside and inside of a wood framework. The larger of the two used for the Variation House was 11 feet 6 inches by 11 feet 6 inches, and 6 feet 8 inches high. Although, perhaps, small log houses would have been more suitable, they certainly would not have B 2 INTRODUCTORY NOTE. been so light, compact, and easily portable. The asbestos houses were fairly satisfactory, but had some disadvantages. Installation. The instruments were erected on a strong firm wood bench (see p. 3), 8 feet 5 inches in length and 1 foot 6 inches in breadth, and supported at one end by a drain pipe, 1 '6 feet in diameter, sunk into the frozen ground, and at the other end by a thick pillar of wood sunk in the same manner. The thickness of the wood slab forming the bench was 3 inches, and it was 2-6 feet above the floor of the house. The bench was carefully erected in the magnetic Meridian, the magnetometer and Declination magnet being used for finding the Meridian, and the instruments fixed upon it in the following order : Magnetic North (S.S.E. true) extremity the recording cylinder and clock apparatus, nearest the cylinder the declinometer, then the Horizontal-Force instrument, and at the magnetic South extremity the Vertical-Force instrument. The suspension fibre of the declinometer was the original one employed at Kew ; it was used throughout the two years and returned to England with the instrument. On examining the Horizontal-Force fibre originally supplied with the instrument it was found to be broken ; this unfortunately happened to be the stoutest fibre supplied, and it became necessary to replace it by a succession of fibres which gave the magnet an unnecessary and rather troublesome degree of sensitiveness. The magnets for the declinometer and Horizontal-Force instrument consist of well-hardened laminar pieces of watch-spring steel, 25 mm. in length and weighing about 1'5 gramme. A light aluminium frame supports the mirror and the magnet, and this is hung by means of a double hook on a small cross- piece attached to the bottom of the quartz-fibre suspension. The Vertical-Force instrument is a modification of the Lloyd balance, and as it was only completed shortly before leaving England, very little was known of its behaviour even at Potsdam. This instrument was a source of constant trouble. The needle was balanced for a dip of about 70" N. The magnetic dip at Winter Quarters being nearly 85 S., the pull on the S. end of the needle could not be overcome by the small weights and auxiliary magnets supplied for the purpose, and therefore additional weights had to be added to the N. end, which increased the temperature co- efficient of the balance. The chief feature of the recording apparatus is that all three elements, base lines, and a temperature curve are on the same photogram for the day. On disturbed days and they were frequent and in the Summer when the movements of the magnets are large the result was a considerable confusion of the curves. From May, 1902, until January, 1904, the declinometer was never interfered with, to the knowledge of the observer, nor its zero mirror altered. During the first year the Horizontal-Force instrument was two or three times found to be out of adjust- ment and altered, but remained untouched after April 1, 1903, while the Vertical-Force instrument was altered from time to time during both years. The method of determining sensitiveness was by deflecting; the suspended magnets with one of the unifilar collimating magnets at certain known distances, and then carefully finding the moment of the deflecting magnet by a set of absolute observations. Routine. The general routine was usually as follows: The Observatory was entered at between 11 a.m. and noon each day, the light-shutter of the magnetograph closed, and the time of doing so noted by means of a chronometer watch. The thermometer inserted in the Vertical-Force instrument was then read. After changing the paper on the recording cylinder, filling and trimming the lamps, the thermometer was again reid, the light-shutter dropped, and the time of doing so noted as before by means of the chronometer. The whole operation occupied about 30 minutes, and times of stopping, starting, temperatures, and error of watch on mean time were entered in a note book. Temperature of House. During the first year the walls only of the Variation House were banked with snow, and a large brass heating lamp was kept burning within, so as to maintain as uniform a temperature as possible. This lamp was frequently a source of danger and inconvenience of an aggravating nature, and required constant watching. Unless there was a draught underneath the lamp it emitted dense smoke which on calm days filled the room and extinguished the small oil lamp which provided the beam of INTRODUCTORY NOTE. I I B 2 4 INTRODUCTORY NOTE. light for the sensitive paper on the recording drum. This accounts for the loss of a few days' records during the first year. Nor was the lamp successful in keeping a uniform temperature and giving out a fair proportion of heat for the amount of oil burned. During the second year the lamp was dispensed with, and the house was entirely buried under snow, and although at times the temperature within was low, viz. : - 30 F., its diurnal variation was seldom large throughout the 24 hours. General. During the first year the curves are finer and sharper than during the second, on account of a more sensitive bromide paper being employed, and, consequently, a smaller light-slit. The magnetograms were usually developed once a week by means of ortol-soda developer, which has the advantage of being exceptionally clean to use and gives rich dark tones to the curves. Towards the end of the second year the supply of recording paper became very short, and from the end of September, 1903, had to be distributed equally over the subsequent months, amounting to about 4 days in each month. This is the only serious break in the two years' record. In all there are records for about 600 days. My thanks are due to the New Zealand Government for the courtesy in placing the Christchurch Observatory at our disposal, to Dr. COLERIDGE FARR and Mr. H. F. SKEY, of that Observatory, for their valuable assistance, to Mr. P. BARACCHI, Government Astronomer of Victoria, and to Dr. CHARLES CHREE, F.R.S., who has been so closely associated with the " Discovery " magnetic work from the beginning. V OF THE f UNIVERSITY ) Of HISTORICAL NOTE. HISTORICAL NOTE. BY DR. CHARLES CHREE, F.R.S. THE tabulation of the Magnetic Curves registered at the Winter Quarters of the National Antarctic Expedition between March, 1902, and January, 1904, was commenced in the Observatory Department of the National Physical Laboratory in December, 1904. The material consisted of photographic records of Magnetic Declination, Horizontal Force and Vertical Force for some 600 days. For the first six months the work was carried out by Mr. L. C. BERNACCHI, who had been in charge of the magnetographs of the Antarctic Expedition. He went carefully through the curves, and decided the exact times of starting and stopping registration on each day by reference to the daily notes he had made in the observation books and the register of watch- and chronometer-rates. He also tabulated, with the assistance of Mr. B. JOHNSON, a large portion of the records of Magnetic Declination. On Mr. BERNACCHI'S relinquishing the work it was entrusted to Mr. H. A. MAUDLING, who carried it on, with Mr. JOHNSON'S assistance, until he left for another post in the summer of 1906. He was succeeded by Mr. A. E. GENDLE, who continued the work until his appointment, early in 1 908, to Eskdalemuir Observatory. The completion of the work was then entrusted to Mr. B. FRANCIS, Librarian in the Observatory Department, who, with the assistance of Mr. F. LEVIN, brought it to a conclusion. In addition to the tabulating work, the assistants mentioned carried out a great deal of arithmetical work, in connection more especially with the tables of Diurnal Inequalities and the calculation of Fourier Coefficients. Whilst the repeated changes in the personnel tended to delay the work, and interfered, possibly, to some extent with its continuity, there was the compensating advantage that the checking of the measurements almost invariably fell to a new and unprejudiced observer. The fact that under these circumstances few serious mistakes were discovered encourages the belief that a high standard of accuracy was maintained. During 1908 a good deal of photographic work in connection with the reproduction of the Antarctic curves was done by Mr. W. J. BOXALL, a senior assistant in the Observatory Department, and a good many other curves were copied with the Schmidt tracer at Bushy House, by Mr. W. H. BROOKES, under the supervision of Mr. F. J. SELBY. My work has been much facilitated by the care exercised by all these gentlemen. To Mr. BERNACCHI I am particularly indebted for the trouble which he has taken throughout the whole course of the work in assisting in the removal of sources of uncertainty on which he alone could throw any direct light. Thanks are also due to Dr. COLERIDGE FARR and Mr. H. F. SKEY, successive Directors of Christchurch Observatory, New Zealand, to Mr. C. T. F. CLAXTON, Director of the Royal Alfred Observatory, Mauritius, to Mr. N. A. F. Moos, Director of the Colaba Observatory, Bombay, and to Mr. E. KITTO, Superintendent of Falmouth Observatory. In response to an appeal issued by the Royal Society's Antarctic Magnetic Committee, these gentlemen put at my disposal, in the most generous way, copies or originals of the records of a number of magnetic disturbances, synchronous with disturbances recorded at the Antarctic Winter Quarters. The extent to which the work has benefited by the co-operation of these gentlemen will be appreciated only after a study of Chapters VIII and IX. Valuable as was the contribution of copies of disturbed curves from Christchurch Observatory the nearest observatory in existence to Winter Quarters it represents but a small part of what has been done by Dr. FARR and Mr. SKEY. In his anxiety to utilise to the full the opportunities presented by the Antarctic Expedition, Dr. FARR extended the scheme of co-operation arranged with the German Antarctic Expedition and co-operating Observatories, to the extent of running the Christchurch magnetographs at high speed during the whole 2-i hours of the two monthly " term " days, so as to secure a very open time scale. Mr. SKEY had all these quick-run curves tabulated at 1-minute intervals and part at 20-second intervals, and the results of all these measurements were sent to this country, along with photographic copies of the 6 HISTORICAL NOTE. original curves. The measurements were received by the Hydrographic Department of the Admiralty who had undertaken the " term " day observations along with the absolute observations only a short time before the printing of the volume of " Physical Observations " was completed. Under these circumstances the Hydrographer and Captain CHETWYND agreed that the material had better be handed over to me, to be discussed in the present volume. The discussion will be found in Appendix A. In view of the absence of reference in the text to the recently published ' The Norwegian Aurora Polaris Expedition, 1902-1903,' vol. 1, of Professor KR. BIRKELAND, I should explain that Professor BIKKELAND'S volume did not appear until the present work had been practically completed. The disturbed curves had been selected and copies had been made and sent to the engraver, and all the mathematical calculations of Chapters IX and X had been carried out and the discussion written. The preparation of the present volume had already taken much longer time than was originally anticipated, and to have deferred the printing, pending an examination of Professor BIRKELAND'S large volume, did not appear advisable. The text has thus been left unaltered, and any conclusions or theoretical views which it may contain are absolutely independent of any similar or conflicting results which Professor BIRKELAND has reached. On studying Professor BIRKELAND'S volume, however, I found that almost all the disturbances which he had selected for discussion were represented in the Antarctic. The opportunity for comparing Arctic and Antarctic results appeared so unique that it was decided that a special Appendix, B, should be written dealing with the subject. It is hoped that the interest attaching to the results will be deemed sufficient justification for the three or four months' delay which the preparation of the Appendix has entailed in the appearance of the present volume. My part of the work has had to be carried out with due regard to the claims of official duties, which at times leave very little unoccupied leisure. A great many difficulties had to be dealt with, some of them calling for very delicate discrimination. Under these circumstances the ordinary tendency of humanity to err is pretty certain to have asserted itself, but, at all events, no pains have been spared to aim at that measure of accuracy which it is given to erring mortals to secure. Observatory Department of the National Physical Laboratory, Richmond, Surrey, July, 1909. EXPLANATION OF TABLES OF HOURLY VALUES. EXPLANATION OF TABLES OF HOURLY VALUES. Pages 8 to 70 contain tables giving the hourly values of Declination, Horizontal Force and Vertical Force at Winter Quarters (lat. 77 50' 50" S., long. 166 44' 45" E.). The way in which these values were arrived at is explained later (see Chap. III., 20). As a rule, each page contains data for a single month for one of the elements, but November and December, 1903, and January, 1904, contained so few days of registration that a single page sufficed for these three months. The time shown in these tables is local mean time, which was llh. 7m. fast on Greenwich. As is usual in magnetic tables, there appears at the top of the Declination tables a value in degrees common to all the hourly values, and at the top of the Horizontal- and Vertical-Force tables the commencing figures or figure of the values expressed in C.G.S. units. For instance, the March, 1902, Declination table is headed D = 151+ , and the value entered under 1 a.m. of the 2nd is 118''1. This means a Declination of 151 + 118' - 1, or 152 58' - l, at the hour stated. Owing to the large size of the daily range of Declination, it was impossible to avoid entries exceeding 60'. The last four columns give the absolutely highest and lowest values of the element shown on the day's trace, and the time or times of their occurrence. Thus on March 2, 1902, the absolutely largest or maximum reading on the Declination trace was 151 + 179'"6, i.e. 153 59''6, and its hour of occurrence was 6h. 7m. in the morning, local time. In the case of the Vertical Force, information as to maxima and minima is confined to days when the photographic record was complete except for the interval occupied in changing the papers. In general the cause of absence of hourly readings is indicated in the tables, the letters a, a, b, c, d, being employed for brevity with the following meanings : a no trace of the element, usually from no paper being on the drum ; a (in case of Vertical Force) no temperature trace ; b (special case of a) a gap between successive sheets which could not be filled in satisfactorily ; c trace existent, but either too faint or too confused to tabulate ; d instrument recording, but not working satisfactorily. When the magnetograph in question was working, but the trace was beyond the limits of registration, the fact is indicated by inserting the value answering to the extreme limit of registration, followed by a plus or a minus sign according to circumstances. For instance, in December, 1902, the Declination is given in the table as (150 + ) 326'"0 + at 6.5 a.m. on the 2nd, and as (150 + ) 30'"5 - at 8.10 a.m. on the 10th. This means that on the former occasion the Declination was beyond the limit (150 + ) 326' - of possible registration in the direction of high Declination, while on the latter occasion it was beyond the limit (150 + ) 30'-5 of possible registration in the direction of low Declination. This explanation applies both to maxima and minima and to hourly values. In the case of Declination and Vertical Force the trace was registered right up to either edge of the sheet. In the case, however, of Horizontal Force the light from the lamp began to be eclipsed whilst the trace was still considerably short of that edge of the sheet which answered to low values of the force, and the trace became fainter and fainter, eventually vanishing at a distance from the edge which varied sensibly with the brightness of the lamp and the development of the photographic paper. The most usual time to change papers was within an hour or two of noon, so that a day's trace appeared usually on two different sheets (sometimes on more, when there were quick runs). In the case of the Vertical Force there was the complication that the edge of the sheet had a constantly varying value unless temperature were absolutely steady. When Declination or Horizontal Force was highly disturbed, the trace might be repeatedly off and on in the course of an hour. The most appropriate entry in these cases was only decided after investigation of the special features. That absolute consistency prevailed in the treatment is hardly likely, but uniformity was aimed at, so far as possible (see Chap. III). HOURLY VALUES. E s s - s s - a SB e - a a - a - - S c ce) ft G, a 8 O. d i. ^ p, oj ft ee i r 1 3 a 1 OS . s t B a : - - s a : s s s s s a : s s : : : i fi ^co ^ Oi .-.aji-HOOooocooaooocoi-.t-cc.nco s a* 1 g eoiO ooo ooTOececooooiOoooecoooeoiQcooo Sg ^^j?SSoS^gg^s5sSS?S8 rH r- i-l < r S c-i - . 1 . - - . . . . I - 1 ^ | | , | . | S 1 c I | | | . Sa^S^^saBssl^SIS^^i^S ^ | i S 1 " rtrHrtrtrtrtrH-,rt M 3 9 3 - ***'' '< g S 5*N?3So5fflS2c3 . - - aB Sc? a ' aaaa 'S a ' 3 S p " T:!0 "i S ?>0 | 9 S;: s 5 ia o m o KJ aoiHO too CDOOO 01 ^osrtajoacegos sj caoscagca SS^-cg^^o^522^S* ^1 O O lOtDtDf-HhO O<-(iOOOO .ri -caflrfojcag ca -. d ce^cacj^^g^gj^rto^^o^^g^os rt rH < + O >A CD OtOiAOiOTO OOOtOOr^ JH II CO - a S-IS-^IIisl=2lSS2S- P ci V^T* ^ ^ ^ < ?7 H 9 ^9^??*^ Otp O p 00 p to p OrHptpp 1 a ..ls a I 8 a -| a 3gg|s T! g SOOO o Kli m O iO iO i-H O O i i O l- tOrHtOtO -- 1 i - - S - 1 1 1 - 1 1 1 . . o u i 1 1 1 tft lO D C0OOO.OO OOOCpCD - >- i i - - S i - s - i s * I - - - I Ills 8 O CD rH OrHtOU$OOO>O OkOtflCD - ieoooouiiUcROi O'WM '. T* 7^9.7'9..T' .99. ^^ pOO r .p r ,p^ r ,p,p 2 d p i-H 5CprHJSrHr-'plOOK5i-Hp 2 s? -C, < , a = t -, ,c 5r . t ,c,,^0> S - S 0*-. 8 gO s w 525" O O w p HOURLY VALUES. 2 ,S TT * 38 !t*ctttictxt*ttx:cj<9:Bsf3t: 04 H A * A ^-1 1-1 l-< ~1 i i SB tti*c*tt*SS*tl4 : S : : i : s : : S S S B ~~iS o Ss t^ ** us o u5~ ~ 7^" VH p "" p oo ' p tp p p iia o p o S t^ f^~ rH 5 O iO iS O O O O O S * 5 -5" ^HO^^O>ft^HOO^H00 1 ^'^'OOOl ~~i i ~ ~ ^ ~ ^ ^ ^ ~ ~ ^ ^ ^ ^ ~ j ^ ^ ^ ^ -r C5 iQ p CO A in i(5 S S Si . O hH H 1 a d, s a rf d . s s " a, S ::::::::: a S S a ;:;;::: :: a a a a : : : in t/1 S 55 p p ^ 00 M ^ 56 p S rt ^ S p ^ 00 W p M ^! O> S 5i~ t- ^ w n p us 5s p P^ P^ y tf3 "" H 55 w ^ w o ^ 5 ^ oo 85 S p ^a OiOSb-OOCOOJpO^- a jPOCOOS^|eOOOf-t-.aiOiCC^-iOiO>tO-^irt3 OOC 5rH^-t W '-"r->^-l >^HC^Cr-(CJ(MCNt--CNMN;5OiC ^H p S p 5 p S p 55 F^ Si 55 55 w F^ p i ^i ^i 55 p 55 55 f^ S~ p p w t^ ^ oo 55 55 S ^ S p S 55 55 W ST 06 yi F^ S t^ p 55 TH ^i ^ ^~ w 06 p y p 55 w 06 55 p p p 55 S p p S p w 55 ^i w S 55 55 S 55 oo c 2 12 HOURLY VALUES. el ill -. a A s I at & s 9 i; 9 9 s a a n x a s 3 . B| 31 s as 3 I S aaasBB-a.-s as ^ O 00 ^ tC M >A i" i - i" ; - . : i". SSSSSSS w us ep p p t- oo t-pODr->USp OOCOp^1ppOOF-i'VCp o us 01 p us p co COSOeOU3t-HiOi(SeviC^ >(.MO>i-(i-tOHO tomoooo -ipco.^wpusai--' t* O >-l o o to us o us <-> "~ : ' "' > 00 p US iO USUSpb-oOppsDUao c^^idC'SeouSi-HaQiOt 3 Cl'^'f-H 010.01 usrou37Husus---iM-*t- ^ 10 S p ^i S p ^ | C4 O1 IM sq S SS S g S HOURLY VALUES. 13 . a a ' a a s s a a a a a a . B a ' a ||1 SS3 S = 111 ^ a a . s a " a a s &. a a E d o n - .-< o * r-u5X>pU5^Si^p^ip^5U5 n 00 CO CD ^ (M O 05 oo ri o ^ p n 5ir t- Ss p p~ H S 10 o o I ;p"o&'p3tjcsptpijiWN;popp^pui.opi^-i 00 iO O O O w Q CO t- ift O ift CO rH iaiot-u50w^H.oo^oiowo6(boo".Hi.o oo oo to t TO to * ^> m ip^copr-i^oppip. ii-t 00 CO t- US O O SO F-iff^,>.o3enppppoo--ippppccp^iFH cb ra o ^t oo OlOiCOCOiOCO~-C*U3i-Hi I u^ >Q U3 p a s s a IV 4V il 14 HOURLY VALUES. el a a a ......... a a a " ft ft a ....SB' a a : : ....... d " ft a A a ^i < rH O O rH rH S 35 55 rH 55 O 55 f- 'S i-i O lprHrHtf5tf3lQt-p iH rH ~~55 56 oo p p "5 5 ^o Ss p 55 p ^ ^ p S5 S p p ^H p p ^ p " p S ^ ~^5 F^ cq p p 55 p p S Si ^H 55 p 06 A p 35 55 y< ^ 55 55 ^H 55 o 55 p _____ ^_ ^_ ___ .._ .._ .._ ^_ ^_ _ ^ __ _, ^ ^ ^_ _. ^ _ _ _ ^ ^ _ __ ._ __ rHrHrHi-lrHrHi-Hi-li-HrHrH^-lr-li-HrHrH rH rHt-H^i-lrHr-lt-IC^t-HrH O'OOppppprHptfSpU5U3rHrHprH COtpUSrHOrHQOO rH Di-HiOMOCQt-ClOOOSCNIr-lQrHfr-t OOCli-H ' ^MCqaOO^'JlO *N ~p ui 55 ^i 55 ^t p p p 55 p ^ S S 55 p TH p p 7* ?H p S~ J p 06 ao rji S p ^ 55 55 p p p 5 p S w ^H p p p 55 p 5s p w p 5i~ sssisiillissil2 "slsssssl a s ~55 55 f^ W TH ^i p S S S5 55 55 p S ^ M ^ p ffl p p p p S 55 ^ 55 p ifsiSf5^DiSOiO^SiOiOOU3iraAOiOiraiO *tto^ot-.^tOtO^D S sssiigsigsigsssgiii *!!!s!g *s i I PI 02 n c J Q HOUELY VALUES. 15 Sg 3 0) .5 ' ' a a, QO - O O ^i "* CO CO t-i -W S H ea a 5? 8 ;^ p in in oo i CO W W oo rt * <-i 2 rH rt W* rH 00 i-l OO CO O 3 R 8 IS o in in WWW 'O O ^O WWW i-t to O WWW W W W W W W 7" 9 s i W (M W W TH^ppOpppVp I o o in 74 si. ^ =; 1. I-t l.t I I o 1-1 00 O o- co to to I-H to 10 8 8 8 S3 x H it i Of-lf-iioOO>OO B v 3r 3 w ^ eg o< + r 9 p p r-l O O i i i i i 9 P p o if> o '-p 9 7 1 ?; M rH m I-H rH 1C O rH 8 3 S N M s s in CO ri w w 7 1 rH W rH V V 9 WWW w to o to I-t W r-t 9 * i S in in p p o o Si S S 2 8 16 HOURLY VALUES. Ill IF SS : 6 S S S S : ; j 6 g 5 - : : c S t 6 o3 ft J ?i 5^ = SS oiooood - O >J ~~CO O TO iS 00 i^ ^H io rt r^i 15 55 O ^r- 3 10 spcqeooeousspi^toopfr-py ooot-sc i tfS'^5pcO'HipMt-TOpcRep^ir-iMrtUi^rt^ "^O ri ^^5'~ " 00 C'iai'J'O^SDOOfNiCiOOCO co^i-iot-cieotfj co ec i -< Cq IH fr- D rl t- OOOrHOOCOSDfHCOr-OOOt- ji t-. tO C^ CO V 9 *O ^ O O t-i ^ i-H O rH 3SSSSS5SSS" IIS COCO5t^OOOOOiO o -i ec ooo po- i * oou3iom^oooiion i *rH<^ t ~7<7^ i -'W ppTp.;*pp i r 1 ^p7 < -V : ':P7 1 I HOUKLY VALUES. 17 II. G SC * *5 .s .S .s -c *j as e s a a a a a - E . e s a a a. a & i A 4 A S I is. Ill a a s. s : - a a Pk a ee (M O OS m S w a w Q =, ?i i' -^ ;; i - ;; r. ' v .; r-: -c. -. *sp<0cptf5sp7< 7 p7ieopirt^7Jciftwccqr-ippp>p^5t-M^w3^5oopp *^^HOOr- ( ua .p (p y p p o> <-H o us t-< r-l ** t O3 00 H O t- O O iO lO to" 00 IB fi ft ~9E fe rH Cl C^ !M T-H t-.t^o VS 9 5 p w p w O 00 i : "C* * lO OO io ?O ^ O TO io O ?) O O O O ff ^* i(5 O A 00 9 ^ >O M pOOU5Olppp^l^- OiaooOt CC'OiG 18 HOURLY VALUES. ll I a . B : s s - s - g s . . E s s s . s - sg.s. a o 8 8 o 2 .' ii ffqoi-wirft-.^HdiHt~o6o6t-OTO i *a6t^oi^Qdcie^t^t-5o 10 id t-^ oi w a II II 1 II OOiOUiiOrH^DrtQOCOOiOOOOO5CQifteOK5tOiftCOO3O OlOCOOOO "opoc^-*f'*oinCTif-c^0>0&OOOs^'CT.iOOCOart 9 * l~ * H g .--s's s a s s" SB ... 6^ s 1 In! oj n Adce a ee 5Oo ow 3'5"* iri * r-cqi-.itoo'N'^'riraeC'-igoQ ow ecso rHiftCOiOCOCoO'-('*^o>--o rt ec eo o> ^< _* 4. t- cp 01 cq ^f ^^cot---Dt-^:cowt-rtt-.ot-.^'-*'-*oo t~.ooiftot-4n * ***** OOWiO(MOt-t-.-*OOMOO>-*OCOa>00 * CO * O 5 t- OtAOrHO 1 1 1*4 ^**^-5DOOp>raOI-*t^'^l^'iD?Db-OOt-Cp B Cfi-OOCO- 1 ^' . _t^DK5C^5O5p {SSSSS^I's35S**92MJRK *Tct-ioic>oc% * ^wrfo^K o^H^oot-ooire^-i--i--NtDoora^Hi-- oous^cooooo ^owoo^ o -SSSSSS??2S5SJ835SSaSSBSS>32SSS ci -SSSKSSSSSSSSSS?l!S2S5 ?!SSSS3 S 5 S S S DOOO(OOO^-OOiJOC(OWOCO NNOC^OOfH O*400 00 -ggsss^sssssssssasB-sss^ssa-gssss OiOOOCOi-f^Hr-(?Or-eo Oit-OiiOiOO> O1C<1OOO*O t- -S tD S'' :il " :;etIb "S l ^S5 CJia>t ^ 1 ^ 00 cacoQcoiOooco d eacqgico^ao c^our-tf^o 5er-(M -^"rtflg-icici t-5jT-i3ico I OCOOO-*OsCOiCI>.OO 1 *^'iO/5 00 O O ** Cl 00 Sraoooooflooico i *cciraco'*c9ccirao> < ioiooocoi-i o*t^t-o> --Di~(oiOiSDCOc4ti^cqScooo^if-i 9 9 V Q o ocgfq ^O-*r^O^HOOOt-0000O>OOOOO Ol 00 M CC Ol -. ~) ^t + * .^^S^^^^S^^^OJ^^^DMiQiOt^OO - _i Ol t~ t- ^ tM _. -.-^ -O^O^< &3SoBr-COt-rt(Mt^ So5iMr-<' :aei oo-*^Q-= 5:rt ?. d ;HCl i rHOCO-*OOl^-10000^-(Ct-t-i-COOOit~O3tOt~r-(eC toOi-lOO W lO ft t- w ^OS^^Otg^Di-HNOlCOCOrH^it i-Hi-^t-.if5 _, _,r^^"tO-^QO _5D rttf^OOi "^(Mt-SS^cr. r^b-m^ot-rtt-eocNDiw * ^oi^too^D .gJ "rtcfiin COOtDCOOOeC^COWCO^ONOiftOO COOO1OO O MO-t - i oi COOCOeDCO5DaOrtf-C tflb-COt-O CO PH O -gS|SS3gSg5S!?SgSS-*S2S><9g3S ri "ISS58S3S2ssgggeococ3O3Ii- *5rHt- S'~ |1 I SP *-Jiiora>3 ^" oot-iMcnaooooo^oocnoi-HTnec ^H ocooooww s t- a. o -2gfig3ijgS|gig5 " 5 M SS|SSS "E3S32 CO laoiipcoiOFHCiMcocoiocn^DNta^tieoo coco^-*toiReDxi!OOio-. E8 ;oococ:ni7-*if5C^'5"5 oi -SgSgSS8SSBSSSS3SSS KSSSSS "S5353 ,_; p c^tooa-^i ^cjopOit^ioc^ooeo^ptoo^'Oiftoao^P'-H _c i it^b-t>.ftt^ M co i ^ ^ JS^S'^* 000 ' 000 ^' 00 ^'"^''*^"'" 00 *o>rt^-v * ca co5'3rt> i ( ^H rH rt a S o *'Stl!^2JS'* f 23IS ClOQ 5 OJa0l ~ l '^'^^S (B eotpOi'*'(M^) _. , t- -r ~i :) iovooi^^ot-iooocot-.-^''3<-**"oo w i-.ooifteDt^iS * w ^^i^^l $ '-IOeO'^llOtDt-OOCOr-OCO'^'H Q O 03 J Q HOURLY VALUES. 19 II III ss a a a a .... a a .a - - a a a a - a ... a d. a o. a & " x 'A a A 4 ' A r~ o j> i~r~ro o i o o ~ i a : a a F-taOp^W^!t-^ ^ ) ^ i i | ? (3 . . . P '..";".. I ' 4 a> '"a * ^< - " - as ' " ~ ~ o ~ t i ^ ( i ~ d ^1 O O r^ ^i 00 N S t~ O O 00 00 S pH O TO O "?O O O N S lO W rH E^ ^H ~ ~p rt co y S us rt ^T! io -tfi r-i o G 5 p 00 p p~ TO~ >6 "~O S 5 ^H y ^ P-I C9 00 t- ^i TO^ io ^iS O 00 S S5 SI O ffi ^ ^" >O ift O 'so S S S t^ W O rt' 00 t W ^< ^ ^ 00 CO S PH i5 Oi f^ S to cq ~ O> F^ O iS ^" N TO OS 00 O O QO BB ill O 135 i o O hH H J_ d I D 2 20 HOURLY VALUES. g -as .as ..ss -as : a* P 3 . , ofPn cQA -" (5 ft to ooi^m t o * o us co Q i-'ojtDiooogjia^oiooeo ^D ^iraiO^io-^*o-*- | irar-iWrHfowt-(cceo- i '-ODC^?J3*Cq^~ ; ^*re^*** 1 3 1 ' ^COd^lrtao^^J'OoaOI'QO K K 3 OB w p3 9 If v v 9 S V v av eo . -il *9 i 9 ~ C 9 S S S a 3 W B B 6 1 ri 1 j|a6S8*-8a-at5S8!i,ai9a*884a3siJ8M B gfl ira^t^^cdcoaietiooodtfi'-ait^cotoodt-^aDQdt^cot^oioiQO.t^tcoicocJ a jg ooNo^'O^'oou^odtDi'Soooo^'oiniDO'-fioooi/soi^Oincoooioc^ ! OO!?10Ct^ 1 ^C^Cfai-H^J'Ol'?O>OOC>CC?Dt C"li (COiClOOrHt-OOlOig>ClO ^^S5^oSS^^>ftiftif;5w.ft*ccrft-t-x>'-c)S.>.H;5Mcc?*S^ M -^ ODO*- (OOO!DC 1 5C4iO^ | l^'^ l ^C 1 ^^^ l O'~'O>tO'-O~-"-C 1 ^'^oOOiO5OiCO(Dt-- 5 ci -fti|SS&2iS.8!ig||8iJ3g8 8 ,J ~g3S8SSSSS383gSSgSSSgSggSS a^^^nOOt--aOCOO>'^'"-O' (MOaOO^iQ'-O'l'C^OVOOC^i (OlCOOSOJOCCOO d rib V ft* fl9 ft d fi tt S fi tt fi ^ Cl 9 t fl 3i tft it 9i tt 4 to 4 A 4 tt "coaSo3it--asSooooicowososi3ii:5CioiooQo; soic 3* oS tDCCi-HNOC1Cpr-(OSOi 1 *CO^-I^DeOCO?g^'HK5asO:pCO'*COlQ^-lD'?IOO ^OOCfeOOitDClTfioDQOQodoCOaiCOOJOOOiCOI^-OiC^iSX't: 5t--O> CO iQ-^'C OOt-CSO OTjoo o O'^"COCllift^ H OO^D'^''--li-H^-t^HOlO4OCpOO' l^pWt^-OOQOift -SooooaD*ooi3oo^*ao;i5coa>o>o>cOiCDb-t~o6oQoo*oost-oi50 id >Qt--eoa3cDooira2ft3iaioioia>*t-iaoooooSSo*cOQOoo + * -^'irapi-HOrHi5 1 fOC5l-b-aJ(-(CpC3C'OOt-r-l^p^OC^i-li-l-J(p5-rf ^*ot-*i-ao^35*ct-oooc3ioooia>i->GO*ooooAt-.cooo H ^D^iOOiCQCOaO-^f-tOlOsiOOOiOCOeOCCCO^CQ^aOt-l-ttiOi-I^DWifS ii Q CO QOmct--'X>QOtDCOrtCOODiOC l *-DCOOOlf:CO : 5t- 1 ^tHrH(M ^Sai^oit-'i>t--*oooo--HOo^oaicocSoot-.ooi-ooooi*o><* N *g8$S8$$S$3Bg$3558883g|$$38g J -I3sissi$38ss*;5gi888|2gss;s|a | N -SS522S83SSSSS^S^?;22c?S^g -^S^SSSSS 5 ^ CO'Qi COr-fOCOOOrOiOO"-D' ICO^HO*O^' l fS?IOO"-D e fl3ff'"t'*iO''5C^tOCO *SiMoeTjic* i *'iD ^OOOcomoOC ?|iCOO<-"t-?5( J-H-^-^IMNt^-^'iMOI'-'rHNe 1-1 00 C3coioao'f"f-Oi'-icirot^N i *ajtoo>racocooo'^' u ^-*^HOrtoob-Q ^rtiftdi:5oioocooowMNw^--ioio-i*"rHC*iioooasccocc!O'*ao^)Qoeit-t-iaoooo t-^ _QQOOoom^-i3|cocit-coa;'*cOQO^NiQ i *^'ococ5eo'^"ttJo>oi<-( ^H5o^i^oic^eoi>>o* iwoo5i--i'-tMC ; ir-(M^t jj Q OJ-^ClOX>COfOt--(rt!3)OSOC^tOCiU5t-t~OCa>e^-^Mr-(GD -.0 fHOi^D'-icccooco^-t-'Mifscoift^-io^eoot-Oi^'^'cocqoTit^'Virac* ~ 1 COeOi-lCC>-lOCOCO-fO'M'MrtiraCC'Wf-''-(^-lC*li-l'Mo5I'r9!J5S Ol * c ** cs &' PC <-HCiOOO i *CIi(5oo^cotDeoioto^'-'QOi-H'*c i ain'-'t-eot-i3'SESES'~' t:o '~ lclODI o^H('Ooo-tooooaio--'OOOi3io-<*ooO'-HOtHOf~( J CTiGOSO^OOmoOt *CCQ 1 ^''ffCOJQeOC^C ( lOnOltDO>(MOi^OlO*O) o 01 < 5oicocfto3jo0oaioooo j S * 1 S 1 9 rt l-< ft i-l rH MCCOi^OODiCCliO^*-^i l t--(t--lift'--.-H-.pQOQCi3esic>o^o^- (C *>oiast-Mt-io -ocoooos3ioso>oooas*Qo5ooO')5o5ooi-'OOOi-Ho^ j? -.wm^^ajt.eoo.o.-.Nre-wingb.jMm^fjgg^g^^ggg- Q CO o OJ O w Q HOURLY VALUES. 21 II S d i II - g 3 S = d H iftt-HCQCOMr-lO(53irocOOOIOCOr-li-l!Or-(iOfHM w oji io-"*5Ciracicoc'5coi i ioec oo co >o o ^ >o" rt TO rjr f^ ^ psy^ccoot-i-Hi-H'^ic^oi- tra o 3B 85 5 55 ^- *p . ^i E f- ^ =p 7* v> '.'-' ooou5rttoooT-imaio i ^ ( cr. o ^ a> ^c o -jft- co r-l-(i-Hi-' fMW ^Hl-H 25 co o "i?a toi-imDU5'- **i^WOO'^Oi005OlftW5CODr-iOQpCO^''--OOCfllOCOOi(M SHsfiaSsSsSSSSSaSSsSSiN PH CO -^i T(( b- 5D Tf< ^f it* U3 O CQ ^ O CM W 05 ' OS " O> " lO CD CO -o is" Ol ftt 9 p ip rt O TO ^H p O O> O p p p p O> ^ T" O * O ^H ^6 ^ CO 00 tfi COpOOO>r-OOOi-iec^H-^'O>ao ""coosio--*ifi oocof-teot^^c-i^ooTOc^ t--*oooo -*NTO -^lOr-to 8 3 S t ^>P3loeo330'*i?S'r>t-!Di>- w w "obt-oot-t- "oo^oti w oiooao icoc'iOirtosxi^'OOt-'V cqoooqeoo ooctko WCOCQCI 00 o,HCCt-cNaoiftaco t- o t- w ^H NUS-V H(DCOIO b^ -SSSS^2^S^SSg*-SSSSSSgS5S^S ^t-WQOOiOiO^'CftOMO * t t-l ^1 3 OrHO t-tOtOO s 4 ^QOt-t-^- .. - ., - -OOTC^-iiOTji -IM-.OCO -JDCDQOiO ^"iO^O^"Oii^'*"i>^O^t--^> *t t OO5D!D00 3l-- CCt-* 1 ^'CO 2 (M^t,-c^jt-rH"M^Hecc i ira cc as p in y CM i i- tntoocc .0 -SSSSaSSSS8Sg SSSSSSSS 51S8? coaiocDina>oi-<-u5aio OONCNOM I-HO-* CCKSOSC^ + 4 *8ggggfggss <| S8gSS*S-3gs|j 1 C^t-OOO-iCC>OOOC't*00 OiOOSSDiO Xff^t-OU5d(DW CO -!S88888i8BS!!*'J38ISaC*gCl:ilS8| QOCOOSCOO^H'O'-'CQCC^'* SOOiONiO iOiOOOW03-*OI^ * -ssssssassas sss8s: s5Kssgsse lO^Ht-CQXt-Oi-lOiOe^COOl OliftC-CO OCC MIQO p4 -SSSSSSiSSSSSS * SiSg S 8g 1 a 2 odoOJ^iraNoDooas^oiiO ^D > t o ^HO -KSS|3Sf:SSSS:'8SSS'''S| 3 OOOiQkntOr-Oi-Ht-t-t-.O K5OCD ^"O> c>*5DOiw^DOr-tiOiOooecwco rnoa^H o ^HOOO S ^CDOiHOfc^O^AOiACOrH ^O^ ''S ^"OX o-: nm:oot~ia-'pmoo^'C4C4Oi eooo^H^^ 1 cqccoi 00 t-i -SaKggS5$SSSSSSgSSisS g r = T -: ~i i- :: o b- ao r-( c looco^-iio tp CB o> 3 <6 i* r-(fH^HiHi-Hf-IO ^ in -SPS5SS5S;SSSSS'''gSSSg'SSS; ot-oDot-^pM5 T't'T* 1 ??*^~ > * B 6 S 8 R | C S S fc C R 1 S j S g * g j g S S S rt r-l 2 S A -SSSKSRS5SSSKS''"='gS8SSSS'i?8S i & rtN-*ot-a><3io-j322;^2( : ;23gggsg55^s5gjj|^ 1 HOURLY VALUES. 23 111 .Sr'-S as 3 I - B B a d S - 1 : B - n od . B s e g : S g B g A a A a I a . Ill K . S a* if a oo so -i w ia iO S irj i^ S ^ S S w ^i n F^ oo "5~ CO o oa I p s o ^r p F- 00 CQ rl O > (N O O COMOOCOWO^t- rH ^ OpCprHO ^p^COp ip CO p ai pHe^CN>At*OaD-*tipHOOCQ^Cp ooiftt coeoio^oocoeoeot-eoi-t o M j *

in OCCCO^-lCMe4O4CC s s s QOiOtDtO^OOODCO iH f-l O ^*" iH fr- O C"3 OD t* O* O 00 Cfl CO TO lO00 1 *^*rHOrHO 1 *O ~S5 S ^~ V" ^ "^~^^ " ooc>coo>t-.i-.oi~i t-tr-QOiQOi-it-in rt o> ^ Oi co ^D co r-t^cpCRp-icpvusooro'^tfa Tj-oe>irtocNijr-i p o p t- ^s tncowNt-iMiacioj^t ?1 71 ^ ^i 2S8SSS -SSSS 00 b- O 00 O OO *O P tO * t- cDt^OlCOCCOOCD(C-l'l!-^C^MN o> w o ?li-H^H pecip^irao.'T.Hip^.j.j^.sQO^ ^OTO^^iO^^-i^ooooa; 9 t i i - Tji .-i CM OlfillACB'pl^r-lprHOOpp SN- 00 "oooiciaoaooiaico ~F^ 01 S O * f} 7" woor-r-ici^pWiO gCOOtl^'t--O'-DO)'-HCCt iO^JiiOt-.oO cOOOJOOt"Ot^OO>OOOI--iJb-OJp > st g i S 8 S S S o o SCT * to --< MAM (M W .-( N Ol 1-1 N 00 O -( " - TO i W t- 'O~ "cq 00 i^ " O> O O w Q TO^pTOopif: ira w5ajpaippcorau3Nu5poot-p-pipcc ^2^592232? 'sS^^Sa^^^r-SS^Kfe^SSS^St * ft O ' t* S F^ r^ M p c^i ^" r- pr-ioiu3oo(N^cp o^j;cii-tQOoocoinjp.--aoi r-3oot-lo-* idxi g j lOOiOt^QOCOOSDeOCO tCeOoOiOOOD OOCO -^S3SSS52SS SSS58SS ^S a' p a H< 3 rl S : : E B : : : E S B : : ,. S E S & ad A 4 B d o, a S!;2SS8Sq S3 5SSS|S S3 a>eQirfoc4(^oior-i .-4t-isci-4 oi wo> S + + + + + + OifSOSCOOOrHCOOiOip OOOOOCO OOO if5O^'jpC2C _> -i _. -* ., -. _. _. -.tt't^CO'^'O) -. ...Oit "^eoio^ioSBio-^Tf* * rt d * * rt * * * 3 IB Q 2 * * S- eooi^Oi itoaot-i c^oc?c^rH ^> oo 0> 00 a>CQ-t-C10>Or-4 tOF-IOOCDO tO>0 -ag5?ga8g : *'*9ii852ii<com^HOi'*i-i^"i3 r-.u3roo nr i <0 S < ,rf OD94t-b-rH>pO>tt ^-tOC4U9--I Op t-.OiNt-OlCTitfS.-HOO -. H> . : _ _ .> H> . _. i-< - OS W --^C 1 -i at f ^ -Tji-Tf^i-^cotNiOiMoo * * * * S * * * A C * * ^ooe-i + ^< r-lrtQOp 1 f 1 *)poeot^o oooot- os 1 * c4 ~83S38Q8S3*/5^o" oo rt O^tDlftt^OCCr^OlO iraiQU^rHOQ O -^8SSS|f:^SS* * * c8 =8 o O ^oot--oot-c pSSooS** '^*^** 5 ^^^ 5coSS5 '** c '5 c> -iI"lSiSiss s : i s - - - i Qooot-toasoot-^ooo COQOI**OS cq at pOOOCpCDtAOOrHCOtO lO -- Ol 3 O C>1 ~sSlSSlllSI ^T3Oir-Oi O -* co- U3tOC400O-4>r-iO1O^D t^ to D OO W O rf -SSS^SSSS^S*^** ***ca ce SSSSS* * * *2 t-l -1 f ODriOipCOgC-^'^-tOJ f-lOOQOkOC? O -SSSSSSSS5< C88JOJ <8 ceoIo!g5SSco* !aoSo,-,222S2SS2SS88S;NS ! 5N8S q PQ OQ ^" O HH H oJ eo odt-^cQ S 1 rc^ooo icooo IH o t- o -*" oo oo ift ~ggg gSB 35SS3SS if-! ximum 1 1 . r S : .- S S : 55 S S : : S^^g- a & * ftdad a"oS5 WO CO "" O> "-" ci5*jO' OO^H iOi(5 ift--H CQ rl o O "*oS5 lf5 QOO1O> CJI-aS rH^tCJOiOOCO ,-5-" ^t- i] i + + ++ ++ + -f + + lOO^D MOO WSOyrtpO QOSO 3 PHCD m i- C^pcqoo t- no i-oo iin -*ppao r* f H M 4| s>ic^ ^-" N^C 1 ) NW o CNI to x. z> p e- cp IQ p ^^^o8cart5ig5{8cart(a2 oSca ^w2 (8B!e85aa!Bjl!eea t2S ce{S * (a ol co oo : -r. to -^ oo t- p-t p ^2S' c 3*co5< teCD ^gjlJ*;i< ! ! SS eao! + cicop ^r N tpoooo prt 3 n Si n H ^ s55^ NW ii cp op co "^ '.' P '/ r ?' ? T 1 n M n Si C^N 01 01 n n n W ^*" 00 C* ifl i". 74 W rH ^l ~ T ^ S^ Sw w cqSe^i co8 is 9J + ^JJP"C -<- ^9 - + . fff aaa 9'? aaaa 'f as 'P^9 aaaaxaaa ^' saa "TC^ r ~ S *oS O^ ^^5 coco co e^coS Si ,0 ^*'>!='? aaa ? ! ? aaaa 'P aa ?-'P a <, a **x a <,? : ? a * ill si a i^8 si * -T='?''?= 8s P?'"2f;!eai.'aa n H tt iftira ^-i -^i (O o gSl SS S? SS8 S?, TO ,??? > ?^a^=t ? '^? > a l8 =!=iy"a= IsS ii g 1 i 1 as o co -* jo g -ill 1 | | 1 | aximum idmg ana time. S E - B - - - - & a a to O O iD >g m E CO 1(3 O 1(5 O O K5 O 71 5i 71 fl 71 71 71 ?J 5j S 2' -s s - - i s 1 - S -5 3 - - i 3 - g - d -s g - - s i 5 - oi -i i - - I I - I - ad i/5 -* O O OJ - v- ip '> 00 C-l ernoo d ~s a - - 3 g - s - * * rrTTTTJTfl OS OS 4 tf5 r- ID IO d rH ri -a 3 a a g j a | a s s II a cJ It2 SO rH O *e*iel" C - ~a'-aa^^aP 8 2 00 W5 p 10 a - a '- c P a a ? a '? a >O O O* t^. W CN (M CM d S + oi " i i * * 1 1 * i i ' a i i * a i ^ a i i - + oj '? ? a cs V 5 V~ 'P i - ~" J i * I I * I + O O O (O o io ad MS a i V ;p ? 7* 7 1 p P r f= is ' s i t! i co 3 rH ir; 10 p ip r -. p I d - I i - I s - I 1 rH ',? 1^ ?! Si ?5 cc ^ +j jfl i ci CO i/5 rH O OS -.O CO ~& _. ^ rf -.0 rt s CO CO * 00 - 2 3 S S * ^ o! H O - U5 CO O iO S S s S 1 & S "* * * i CO CO rH t~. iO iO - * -i ^ co ^ 00 > en C4 + ^ rH Oi + D O U5 5O co* + | ^ ^H p d o g t^ + 5 O Oi ip -D to N U7 8 I a s 1 j rH 1Q rH CO >" co ^" 5 CM W N HJ + rH p lO OS CO "* Irt fQ JH CN * -- O O rH " * 31 S t 5 5l S -H W co ^ CO N CO >C M 8 2 o I- O O iO - g s a s - p p p p -a S id o O CO rH t- s H d - | 8 g S ft Cl CO ** lO C Q limum ig d ft s - a (C ft CO tC. 00 10 d 3 ft f CM rH Ift <" rH a ta cs 3 B oo QO i(l i(7. CO CM CO -fl" 1C CO CO 3i .8 aximum 3 S i & a a s | CO d ft 2 o O rH S B A 5 o + + + + 00 >f5 * 1 J r M ej - i o 08 ?H OB rH i~ ^ iO g a =s a a a s a g A d CO ** o CO g tf 01 00 oB r a g oB --: ^ o j^ ca ce ce ce ce * S oB CO* i co p ^ S rt ee ea ce ca 3S a S oB - -1 t- SO '" a g CO ernooi ^ ^ o rH CO O 3 .t ee ee ce g I 5 ^ : i Jjj? j ca d ?. ca ee O iO s 4 - cfl 00 O u5 S? " 1 i .a ce ce ca ca .; cd ce 2 Si s ii B ^ .a ce ee ce ce ? ca ca CO + ^ w .t 00 S o I S 1 ^ 1 2 - ca 1 " - * I * - I I 4- us ?l a - oj 4- **' *c8 ce cS et) rt ee A ce ca id i d - ce o ca ce Ol CM * .a 1 a a a a cs a ee ce i CO .a 1 a a a a a a ce ce I 1 - .a ^ ce ca ce cs ea ee eS -f I I ^ - ca so o g g? ee ea I - ea ^ ^ ea ce ce ca ce S ^ ee ce s ^ - ce co g ca ee ca ca cs ca ee i m p oo e- o ce ea ca ce ce 2 55 ce a I- - ce co jo ea ee ce ce ce ee ca o - .a I | a a . a a ce ca s 1 d -a :P ? PS >, t- CO OS O < CM CO -J iO SO ,. R HOURLY VALUES. 29 5 I S II. 01 o O O O N) O w oq d d It = H=' dd E'S e i dd t Nift^ eooo U O O O U t ^ -c -c UOOO^2S JO iQ M9 -e ti *c ooooooo e -0 -s o oo^oouoo UO O> -. S S S -O *C -C t T3 o o o o o a 'e -e -c -a 8 12 10 m -e T- -c -e tcira c -e -d -e os cs 0) <8 OS ! 88 O O O O U ^ g 80OOOOC>2 SgSSS"$ iA -^ lA A A (O -V o! -C 'C "C 'C UOOOOSSS5 HOURLY VALUES. e 5 B s . s s s (i gj ft C8 cl ^ S B =>. r-l 1O S S 5 to co ^-i ^< co W O t~. CO ^H Cl ^H t- t- tD S 1 s s O 03 w j : ? O H I h- 1 M O a I I I I 999 CO O Os * 338 00 iQ CO t-4 I I S SiO 00 O t- to * t- t m tf3 S i to ^* S 8 M IS 111 __ . l i * - cq i s S 3 O I- + S i i s s o o 3 S 2 C A J9 A -* IQ l~ QO S 3 S 5- o e-i -H -* t- to S S S 3 ~ -> - 1 1 I i s I I I I m co m to 10 !o 10 I I I ! T? 3 S , t S HOUKLY VALUES. 31 II aS . s s s ' n >o >re ire 10 10 ire ire S ire 3 - O fH O =0 to '-O i S 2 S S 3 to ire n ire t- O iO 3 ri ri V- IP V CM CO t^ O - to 3 S ec> H ^( CO to ire to ?! 2 to to _. j i-H r-< O* 3 to to ire en oo Ire 5 OS CO t- to ire t-. i~- i- ire >re ire S S S S 8 SCI i I O O O to -X) 10 SOi CO ire ire Q I-- CO o in ire - - . . Op OO 00 ^H 55 o Ki to 83 -. CO -H S- 00 .re ire 10 ire 00 00 ire ire i i-l OO 00 ^H oo - i-- ire B' O A M OV oo -C a S S 3 S 8 S S ws in to 1-1 i~i is 3 a K .5 HOUKLY VALUES. = 1 3 . = t -. a O >!5 117 "f "* CC CC M X C 1 ! !C ^H s g s - . s - - s - s A a (4 ft - 3 oiaiec-i'i-'.-ii'.ptot- fcM.|B*4MB*r9iMv4 ei.-i.-oi'-SrH-S*aoq^wecrMiS?}'fl;i(5iS-^^;c5 i-3 * ^* o5 06 od t^si ** N l -5t-5^<^cioiddNc i i s . 01 o M P4 i?i ri S 3 i S H t- 00 to t- t* s s s s 222 t> --D S S S S 8 S S 8 g S 3 S o o s s 3 | S 00 C4 3 3 1 I 3 8 3 34 HOURLY VALUES. III S S 5 S : : S S : S - : S : : S - S S S 1 s f g 3 I : A A a A A it x> ci oo o -< M 4 S l. 65Pg "Sa5 S I >^ 1 * ;i ftO CC ,-< ,_, -H o'~ l 8V I t- t- IN O OJ I g ^ H O N a! tS O u a - Tjtcot^ ^HWMt-^- t- t- t- so w os i ---- ff-fI; ,II2S HOURLY VALUES. sfa S S 3 - 6 S a a ca a 3SSS585SISSS HOURLY VALUES. 37 ll. g : 03 2 3 ~~i I -BS " A a ft 53 10 03 i^ - 3 o ^ 3283 N <- ci -; S S = 8 =5 8 t- ?1 S 3 I i l to S t- O S9 & 1 S 3 S o os j^ m I I I I I i-H it" S 3 Si ^E 3 3 2 S V =O I I I I I ^ s ^" * ^* B fep B fl w5 so iC ^D ^ eo S*4< rH I-- tO tD 10 1-1 94 ^H I I I O 3 ~~i S s i s I - /_ .T r; S I I I S*^ 53 35 O sis HOURLY VALUES. ll A a ill + 5 lo + + + + n + ^S-S'^lS^SlQlOi^T + + + + + + + + + 2 *i -HjHlj- S i 3 3 ' i ' $ 8 l' ^'fl'CO 1 *^'*'*'* . i' i' i ^'^'* i ft & 8. 3 9. 1. 3 1 $ f 8 I S .i C-l o 05 132 * -y 10 ira PQ 10 SJ O II a < 9 41 - II3iii tt > fl -- tH^-tWM'ft-e-O'NOllg ^SQ'*'C-t)-^'Jo-ii< i 4' 9 w 9 Ma 3 Z Z v v. 4 R 9 / i~- z ' i - . m-#>4i-#&-;&-'*-v-i' H I *<# if *# ^ # o td ^^^^,^,^,^^f^ -"i-'^ JO S 1 HOURLY VALUES. 39 . a I 6 : S s 1 3 5 1 o ft " ^ i 3 i s 00 re' 5 i 2 S S S S S 3 1 | Si .^ 3 2 ' ft i S 3 9 C S'-S ID K S iii ii 1 1 1 l g 1 i i 1 i * * s S S S eS ^ S M 1 i i s 8 g - s -^ ^^^ -: A " : 6 : : : : : S : S E - S S . 5 g 5 1 ^' $ 11 o * I ' t % 'i i d a ft ft * ;-. a r* ^* o o 'r o a 1 2 'I ca s : 3 a jjj >--' '--' I IS 50 C> OO "7 f> " 1 " i^^^^oogg g i i | S. S 1 " 999 * M M " J 1 S 2 1 1 1 1 s I i 8iS8S88'S'a' 8 3JJ P = | - - | 88 1 CO o o M o I 2 fl a CCM ^r-t 3 9 - CC^' - B H P I 5M 9 9 C 11 SM ^ fi B COCO + ~+ +~ + re re " : T~+~ MCCM s e s re re re 9 B 9 H S S t s?s '2'S - + + + + + e 3 !l.'!i' - S S g g ccccrecc ^ i s s i u i i i 40 HOURLY VALUES. ill III s-.sa-.ss 3 - " A < , " ; A J 8 S . g.'B 'S -* * 8 . 3 8 .. S a . -9-.-S S 3J S S .d fc 9 I- ^ O ' If5 *f ^H 31 1- - 00 irt M eo ^ LT 5r so o Oi M W PH B O I S 01 S 5? rH I- CC a ' ec co cc 3 S 3 - CO CC B fl CC * CC C? S I jo 3 cc >- ec co o CO CO CO CO CO CO CC a . + + + + (-- us ti ec eo cc + + CO CO + + 5 U5 CO CO + Io S CO CC I I 7 Ol CO CO * CC CC K S iO Oi O O CC co eg cc* cc co i ift CO 1* S g >-H CO t- a i s I 1 - CC 2 2 CO CO CC O <3 -* co ct cS D f S c? I- t- CC CO t- <> cc ec cc CC O i-< O I- CO CO CO CO CO O> t* t~ ft t~ 4- t- O $ $ + ut t CO CO ^ 7 _; i i CO CO s> a ?i rf 00 CO O OJ r- t- "V 00 00 CO CJ F-I CO 8 S O) O 00 -* CO S CO W CO CO CO V to 10 if5 ---. Mft W Hit 'S * + + + OOl-^COOViOeDO OOl-t--t-^iiOOO i9 A ra A A 8& - A A i-- -VO5C>'* 1 *'*'*iO < A 9 A A O^O>^^25'^ l 3E* Si SK 3E SI uE 2E O jgO Si i * S S 3 s S c^ ^ '^' rH ' i-H cS rH i5 Cfl '~ l ^HC^I -. HOURLY VALUES. 43 ill s^^ E . S d a a a' o. E - E S -SB d a 1 d ad -.:::: S -SiSS32< c ,!3cog?5co'l ill iii a . a E 2, a. a a d a* dco d^'^d cS ^ d t~ a IN e ^ co Q> >ra ! oo c - - g S a d o ^4 FH ^ c> c-i *' o> .4 SSo^gg ! c-i ci co * o sa v* S !3 tO 8J flS rf _. iQ O "5 to to 4 tf S * * o I N HH I 6 d $ H ^ |Q ^4 US - O lO * CO W CO O t~ O t^ O t d 03 d -t _, ( C^ 3i i-i ^OOt-t-t-t-A ^-g^f; 46 HOURLY VALUES. Bl i*s 1 15 i] a .Baas d a oi (i d O ^< vl Oi ifi) < i >-: Jc d oirtoui^cioocid-. go"*. Aoioii'fda'? m ^'"S^ri '"<=> So S t- 00 S iii III 2 ::::: & co co ., t-> co co e- T) ' ' M co s . a . A A 00 iO * i/3 00 i a a s . ft S 7 a S O O -i -H Oi OO CD CO O t-i CO 00 s s s CO t- 00 CO t- 00 CO o H ^ O s ci OO 00 I- to * CO 00 B 8 B 00 00 CO 00 OO OO 2 S? S CO 00 CO S S3 Ol CC 66 00 OO 00 O O O O 1 S 8 00 * S S S ill tfS Ot O i-< I o ^ <* m & o> co oo oo oo 06 S oo s s 00 OO 3 S S C 00 00 OO co co co CO W ^ * 3 - 2: 8 2 CO 00 t- *- s s 3 * D CO H CO CO CO S CO * <3i Oi OO O 2 1C S3 S S _- i i CO 00 00 CJ '00 *" ITS CO * n -t 181 I g CO *- CO t- CO CO -n o ^-. 00 CO 00 I 8 I 9 CO uO 00 CO a s 1 9 CO CO CO CO CO OO 00 !N 10 "5 S S2 3! v) 8 S s s s - i . ._ ^^ co N CO CO 00 CO OO CO t- 00 00 00 00 g g s g 8 S S S S 8 S S S 8 2 t~ t- OO CO OO S) CO CO HOURLY VALUES. 47 !- .S B II. Ill CO o O5 PP S 02 O o S :S g Sd|g : a (i e . * s,?* g|?T cii * oo re'rt S E ' S - . HAs|=Ig 0. v*> a,l*> a g g a s i i i t i i i i i i it : : :B S - ' "H P- d co co d N ^ 8 e s? o, | s a e* + - t- to * o S 00 S 00 -O CO ** ^* O 1/3 jg B Eg E C E K a I I I a g a > * a g os ta (Sao CO t QO t- CO 00 00 CO 3 t S a 1 S S S S8 8 - S i i iIli a-aa ~i i i i i * | Si" 1 si ii ii i >r d ii ii i g g - . g g - g . I g s - . s g . g - p <3> 1 III gggg g 1 5 CO 1 1 1 ggaaSSJ a:3 a t- t- t^ t- t^ o" c - |2*gg 3 * t 8 *j i ggaa^g ag a * ID 3 d agaagg ag a 00 t- t^ 00 3 2 a d * a^aa;* !? a 3 * 5 W i cc - 1 " ' 1 g "1 5 w - S3 S S S 00 00 fr- OO ^ >- 3 i -1 00 OO t- t- 3 | 2 1 H 3 j 1 1 H q d 1 1 1 1 1 aggaag gg g | oi III H 3 00 ,i ,i ' tl H d feB 1 1 $ S" !3 ] orenoo d l ,i ' H 5 b .' ,' ii ' J 4 .'I ' ' ' to II II 1 aggaag ga g .i 1 .i J, ' ' -i II II 1 gj lO lO gj gj lO 1 ". _, 17 i-- t t- e-. t S d ii ii i I ^CflCO^UStO t00 Cft HOURLY VALUES. of H ^ g K a um and m g in m Mini eadi Maximum reading and time. i a I i i i i i o o o o o c5 c^ ^ o-i c*t | =^S = "" "* V "? fs ^l CO O CO CO W CD 1 1 1 1 t O 9 1 W i i i i O O O O W (N ?i N Iff 9 V M 1 1 -H o 9 >H M a 2 * * 2 ' d o- ; i i i s 1 i E 3 S 1 I i ** S ' : - g E d d 3 S 2 ^ 3 co d rt eo' r i + H 8 !3 8 S *" s a A i i a I i g a aaaaaag d S3 a aaaaaa co co e 0, 1 g g " I 00 1 32 S aaaaaa; CO CO C 1 ! ^ - 2^2 dd^ddrt CO CC CO CC 1 ri S5 ?2 8(812**^ CC CO * ao "* .- S de ' -' ^ ?J d deJdciddtf n ',2 d ddddddee CO oi d^ d Odddddo! CO CO d d ddddddci CO - 5^ ddddddrt CC CO I oi 1 ^ *2<-fc ddddddea cc oq * IO 1 ujjg 2 dddrtdrtoS ,, 1 - ^ r'- ddrtdddd CO 1 1 rt C- '^ ddddddd OI - - '-\ ddddddd - | 88 S 3 dddcsddd M CO 4 s d 1 | d^ Jg dddddd* C4 C4 Q t~ -ft OI Oi-HNCC^iOtt 50 HOURLY VALUES. l- is a .s-c -3 si S a S a 3a>c ala tO tO W^ O >O OO jP >O 9 ^Di ii-OCO ^< tD V CO CftCOrH S rt d assssssss -..ssssa,,??* -.- -ss?^-,,-;, a 11 ssassssss^sasaa^???-* --53? ---.,, 06 SS2S8SS22 -^sasass^sss-,, -.-. ss? ---., - aa jHOWCcg.noon*. S53Ss .ggKi:o S o> O>-HO. j ^ 83S2S882S- <8 5Sa8a8SS35- ll S-,- 1 ,?3?- rt< d rf 3^Sat8133^SSSSS3*** i ii < kSS*-ii -'" S" II H 2-.JS3S83SS-,, a S S 2 8 8 S 3 - a S3 - a - 3 3 3 --;,- a g g g g M ggOO^g 55 5 fflgoo rtrt rtr H^ rtrtrtrt rtrt rt rt 4 B - ie 2S8SSSS- a 822SS . 3 3 9 3 - - -. S ? - 3 a $5 rt . ^ d ^ H a - a 222SSSS2- - 8S228 a 333S-- el - (J -. ll S- a S a< , 00 i3fiS33- 338-38 33I'*'-*S i fc !SI '< >^3.'.giiigag88S*l*S i fc** a( 3 i fc* ^ S -0 a^-^sasssss-^asssa aSs^^s-i-aSs-,," , * o^goorom ggoogw *5^ rt aoooo ^cogoo^o ggg.g 533 o.ooo. rt _____ rtrt rtrt rtrtrt _, ^ Mr , 1 lOiO^POOOOTliO CO-HpOtO * 3- s s s 3 s -. * s a a s . 2 1 a n s n v s s s > -- -..*......, w >J HH W" O O H 2 52 HOUELY VALUES. s a a cj OiO a s e a & E s B B B S 3 A 2 A KS s s - s a a ' A 2 i P. S r-ioodooe4 2 a CO <> p l ~J3 l>.t^ r-^ (S CS rt 3 00 4 00 5Ot^^ i- i ^^' ' i ^CCCO ao co co co -r- OOQD t - T- -^ T- t-t-C - S M t- T- I- >- T. ^i 1 - ' 08 oj t- O> as t- O> O> 8 * ~ 8 S - S S-* 1 - S -iZ / C_- -- S -. S K SO O 61 ft M*S25Sli cO r- t- S d COOOTO^t ^ r li-rtrtSSBS OOCOCO COCO * 1 S 00 cfl ** i - t - --i "- O1COOQ HOUELY VALUES. 53 a . s s s - g B' s - - B' s s s g' limum !i d P. a: ft rt ft a a ft * ft o t-- r- oo oo V V V 9 JP V V M B W jtOOeD^D>'3 jiOiOiOiOiOiOirtiOOtO C3 t~-ev2i-Ht-.i-tT((C2eOOOOO COMMJ^COr-fQt- ^t^COOlOl'-CCOOt-OO i4 SSSSSS3S8S S SS5S-3SSS ..SSSSSSSSBS t^;jgr-r- (tr*< "*j-otooo ;; cc i - *ri -i< i < ao o ^i/soiot'-u'jt cori i 1 3 A s S 2 ' i S j S 9 'rt"^" 5 S fr ^^^^o^^^a.o^o,^,^^^..,, -c,,o Sa a g fl P 1-5 H t> 54 HOURLY VALUES. el s a . S **g Ssl B s s a os - ^ St^QiO wTOm a a p. a as m_ A s . e a s ft 4 A 4 e a s a A ft d OOt/3 >o ^H ** l-""*O ". i ~ i -. CCfM'fifSCO 3 . Ill E A n o ^ oi od 4 a S S t-i oi iM g S I i S i : i ft i 00 1O E S : : ft g 8 g * e4 5 eo f-I O^I i C-l o 05 w I i c d SOiOD 8 S o sssgsg-i. HOURLY VALUES. 55 11 . Ill as S d - S " S. : 6 : S . 133 s s ~ s s O O O t-i alsSISSslsSsSdsSSggssssg 999996 S.ov ^ I 56 HOURLY VALUES. ~ 02 3 o aj d t> ll. Ill 111 s p. B t S B S S B 8 . - a 0. a (i a P. o! S S S 8 8 8 S S C3SOM t-^ OOOOi-irH a s o, a e s a s a : ii & a d, a S o S S e, 8 SS _,-'- '** J in M t HOURLY VALUES. 57 . ssri . a 8 S oi o a E a 2 S3 O rH IH a s a . : : e a : s a A a A a A 8 2 o S 2 8 3 a 8 co O iH iC oi I-H CO O CO O O 3 II H - a a a : s . I-H O O 8 2 8 a. a a & i & S 8 a a a A a a. a rt & rt rl O> >-l O r-1 W O QO t-. s a . g Is J?i s A s rf iO i a a . ooi-wodoiooo^'O O ift cuci w w s Q g l'5'^ H t-O '*Ofl It n * ?- V -V If ! - "b D ' H lC 1 * rt d * I d I NOJlMtMCl'MC^oq I 2 60 HOURLY VALUES. a III B s a . a s s r of ti tf A rf A 10 IQ o o ire o >n *f U5 * t-l -* *< W rH OJ " Cfc ~ 1-* PH S fi abci - a If o >o o o o iO O O -" i-l i-Ieot-Iui.-Icc CO o 05 n O O rt C i i 1 1 iH iH Oi Ol 8 8 S S Ol Ol "* 1-1 O Ol Ol CO O 333 3 S S n 41 41 Si -a -o I PH a d v ri oi I-H 1-1 Oi I i 1 i O g 01 01 t- l>. ^H CO O CO SS R SS 3 Sg 5 ? W 01 C W I i i i H = -e S S s s s e M HOURLY VALUES. 63 1 a a . Sy)4 J c S s i .... A a - o. SSSSJS E - a 82 e - P. SS I! E s 3 A loB SS'-S ss i s s * f 3 SOO QO He 2 CQ o IH oaj SB a a *& OO r4 2 ooJ a - 2 coo s o 3 CO o 05 9 8 S Si S I oi a d O 3 3 8. 8 g 8 8 8 8 EH > I * * * 3 8 8 3 8 3 3 S 8 8 * *> i t 1 64 HOURLY VALUES. .S'c'S Sg B S a A S - S J " A ! S O. ca ooo sl 3 . CMS . S S ft at 2 & a) 91 OO Oi O O rH O O IO >c"i-i '~ N n H Si CO o 05 d o O w N M eq cq i i ft vi -H H Si 4V W M HOURLY VALUES. 65 5 s a. a 3 CO O S B - 6 B A a A rf O O O O tt o 4 O t-^ I-H 1 00 B 6 a a 3 3 B d. BBS : B ft 4 a tn 5$ to i-t o CO* Oi Cl M 11. ll a s - s 3 4 3 10 ro o S H 1-4 s e : O. rf s s s >-H I-H od a a s A S B a . . a i a ' = O O ^ o TO o w ^H O OO Ol rH Sep oo oo S *" '" w w o o d (S o s a S C*I r-l 35 o 3S o> 71 o C-S r* S S 00 l CC t 5 33 Cq M W o o o S A 41 n OS Oi Oi O 8 2 Ts" M CC A O> O S g g SB S 3 85 HOUELY VALUES. 67 | I. 1 S E : : : S S : : : S S S : S S B If A J oj & A B ' A ej SoSSSSS 333883888 o 8 - oi ei ctf d 00 td^Dooicii-Ioood oj o i tpt-osiraooo cotoo^T-neocot^t-- ^* oo c*3<2Qco^o .Q-*>^iot-a5eoag ^H rf i i i 1 ri s : t s s s s B - . s s a s s 1 | S~"- 3 L *s ft ea fk m A m ^ rapo OfMOio o tft in o n A A oi ^j oMraooeo<-ii-i-* 10 1-1 r4 cc oj d rn ci c I-H * ci o ei i- co TJI* ci & i-c*3OOi^-(;* MOO;*I-II-HO O T3 ri **i ai o o t w o o> ^ ^"^looioasoiOJ^Din o t ^ r-nc^ i oj 00 iCDOOOOQOt- (5 e8 QOGOOOt--OOt-l^l^( COr^OO^fO-^'--^^^- % fH . CO OS M t iO N O <3> iO ^OOOOiOlOJOl^t- O > i-^CS t oj 00 'COoocooot- os (gOOaooocot-t^ir-t-t^oo^o&t-^^^-rQt-t- ^"" d Sb-Ot~-if5C^OO) OU5QO>O>OOSOt-O> OtD tD O1W d OO lOOOOCOCOt- oJ C ' gjOOOOOOt-t-l OOt-(^COt-r0^^-fgt at St^-Qt-WJCgOOOiiQ OJOMOiOOlOQt O OJ^O t- OJi-n ria'SlaS^'^SeiSzsSS^^'O^'fl'tiS 00 t-Ot^if5C5QQO>OiOClt C<)ClOO31QtOt-t 3i^0 ^" t r4 (joo ioooocDooaot^ooaot-oooot-QOt-aot^oot^^-'^f^t~^t' t--O''-*raQOOQOu5<3sOOOd3lClO WtOO'C ^< fr-^H d OO >QOaOOOCOCOOOQOOOt-OOOOODl-0.t-OO^OOt-QOt-^^t-^tOt- d g Si-~oi-'*QOiocit-- i ^'c;oo^ooJoaaoj^'C > i^cot ^< t^n rt ao ' ao QO oo t- a> t^ *- oo t~- QO co ao r- 00 (~ i- t- oo t- oo t^ ^ ^ i- ^ ^) t- S 5 oo t or^-'^'ocioait looio-fot t-aiowoi'fl'oi'-D N t~^n a> ej 00 ioooooot-oot-.t-QOtooooooi^>-i -t^wt-t-t^^t^^-tot- oi O au a>oi-.'*Qaioait-i'aJOiOQe-t-aJt't^'*ooa3jaso-fe>it-c^ot-t-'r>--iiMO i fl' o eoo AS5SsSs^^&aSSSSsSSS^S*"'O'^'O'S '.' 11 Si^.i^ioiofNOOioitracat^t-oot-t-i-CTi'-H^D^i^cn t M H M M M . rtrt rt rt rt 10 t-irao^aiOio i~i~a>t~o5t~i~^HWtDt--i'tD t- rH rtrnrH rHrtr-tr-.r-lrt^HrH rH d ^.o.n^tao. g r-^o.^^t-t-o.-.gaj-jjt- ^ >. K rtrHrHrt """ H " rfrtrt rt rtrt r H rt 'e'o'O r , 81 t^LJOJOOl-X) t-l^t-Ol!3st-t^WiMai3t-^Ci r-( t- H t t-t^if5CflC*31^ Ol tt^OlO>Ott^"l'MOS^l- 1 ^' N *- Si- i- 10 -i* w t^ t- a> t^aso>a9a>t-t-9iioi^i~-o 01 ci 00 S^S^^S 01 ^ b-t-dO0t-t-tt)(Mair-t-D Tf - . OO OlOt-'^'WOiiSJ Q lrt'-OCCSt--t-tD^lOl t--^D ^" t . s SOi O t^- ^ N t t Q t-b-OiOiOltOit^if0> t-SD "* O gjoo^aoooooi ed*oj t B B B C 1- fe*B^)ES-iQiQk!>)Q E QO oai^^'Ot-*- o t t^aioost-cit-'ij'Oi oit- w o> 2SSG:^ ocno osooooot-oai 01 * (5 ^0 'cflo! 1 ^*"" t-CIt lt t^fr ^t~t-p^f^t-^- 2 SSSf^iSSS"* 1 * a>wowa>o>oiOTj(M oi< T* o> --OgjC 0000 5 gQO c jt--OOCO l^-t--t-OOt^COrQt t-^ t-iQ^ 1-1 00^l^O)WQ W OilQOO)OlOiOOTft Nt- f <-l ---O cS^ .0000000000 ce (ji^ooooxi ^r-oot-oo^oot-^^t-^t- O OJOt^l--NMOJ * aiOiOQQOiOiOi^Oi 2*^ "* ^ "-.S* ^ 'OOOOt^^jOO^t^OOOOOOQOt-l^-t t 30 (^ " ^~ t!- T" *"~ *H *~ d c S 00 ^SS^^S 13 * 3 -J3;ooioosai0i!0.o 2^ ^ -- 1 & * |. ' . 8 3, 9 9 3 S 3 S d A it R fl 8 $ 8 X ft X S 8 8 (5 CO o 05 VI g O K 2 68 HOUKLY VALUES. II. Is! s - s s s . s s a A 4 A * A 4 A !? S S n S S S Si S 8 S S S s l CO o 05 M I 02 W El a . 1*1 39 s s s a a a A 2 A rf 6 i I a d oa a ta a a c8sSSrtS2 < * < 2 sssssssss ra OiOiOUplO-^t-lCl 1 ^ S222S22SSSa=8! S222S3aSiadaS2S22S OO^'^J'lO 1 ^''^"t-.i I ^f CflOSOt W r* SS22222S22=8aQ ssssllsss3S5gsss I JH^2^(22;HS2^ cS8 '= i * o io3s*'S drt iHMSHSS rt ***t- t . SS2i!s**as!ososce*es5BBi* rtS *5 SRssSSsSSS*** ^^ = * = art rt^SBSS****^ S o ^^sss2sSS**'*a****BSSS (:a * (srt s s HOUKLY VALUES. 69 .= T ?.'< a a. S3 ' ll. .*! e a a 3 07 O M II I* H w ; fa SS d Cfl 2 rH Bt~> - S " 2 3* fi B * S--H roio 2l ft fe in ^H * r D ?i a S* Is r 1 S*-OC4 -f < ( * - -- . . S S ift-^H I -;* B ft fi! * 5 a -'S: spt-- 70 HOURLY VALUES. I w M o w Q Q CO O =5 w i ^ o g Minimum reading and time. E E a a I 1 Maximum reading and time. S - E => 5 1 2 n Ci O aximum 3 "a E E S ft 08 O, So eo C* t-i Ol * 1 i 5 1 4 2 S S S S a CO ft * * ^ ^ S * o ^ -* to S ^ H * d CO tO W tO 00* S 3 S S * - S S S S * 1 OI O CO CO ^ ^ ^ ^ rt CO < 4 <-> O CO O O r-H d d J O 00 CO O) rn (j M W eo* SCO O 00 ' rt ^ .-1 t- r- Oi S 2 S 2 * Q - e) * I 3 sill S sill S - S 1 S 1 oi a I a S S 2 8 a S S 2 S J 8 eq eo - 1-4 S d m ^ co t- 4 * S S S 5 ,0 CO 00 CO O* OJ iH O - s S s a I o" = 5 S S S W CO ^ Q JANUARY, 1904. oi d d b- II > Minimum reading and time. Maximum reading and time. 4 a M w rt rt 2 rt a < s S w td rt rt 2 rt o* rt cs g rt 0) rt rt rt 00 00 00 g a a g a t-I t- to oo re g g a g S - 1 i a I | lO i i i i i i - -* M B K K S re a * g | ' W I-H O 9 09 g - 00 g "Jg g g 00 fc 2 1 a a i Forenoon. a g a a a o rH g a a a a g 0> Ci ;O g a a a a 00 a a a a 2 t^ s a g a a a d c? 22 g a a a S lO CO a g a a a o 4 a g a a a eo a g a a a w a g a a a J a a a a 2 1 a d a g a a a 2 1 '- C. "1 i- --T '- DAYS USED FOR DIURNAL INEQUALITIES. 71 S 1-4 g ^ ^ d 1 S Q 1 ^ ? H I II | I o "g -1 F^ m I" w Q 02 P H i 1 if M ' i-H >-l 1-^ rt " N C* ' d Cfl '""* .-1 OO -f 00 % It it i H R R W> It o ss- * OO^,OO> ; t-(MtO fAp^ M ?4A; l 'M S S S o eo s c^ n S3 JH =5 R R SI 2 S S 1 8 8 * S R t f. N C*l R ft ! * f5 o O i-t (p CM co co n w H R R o R f R * * * *""" X St A ft fi sf 98 8 ft 10 It S S 8 N CO M C>) ?| 5> % to oo * o w ^. to ^ * * * * * * * * S ,! ^ tO O OOC*flCO [ aaaasassaSssS fc o , . s .. . M ' g KS ssa&aaaaaiasa fl C - gs aaaa*' * OSD-*OQOOOO>O>(MOC > JM^H a n H & & Oi-HCOi-ntO^Oi^OOV OtOi < b- lO O OO O OO OO o * * * s g * 000.& S2 ,S.^ vf-^ CO CO W N a Cft t t->aOCiCOCOOOtDOOtOtO - W O) " O tO ' * * * A HI & t_ * * * * * * to H . ooeoaoxncococoiec<>c*^o n 71 * 7lTli-_ l- '::,-.- NN CO ^H 1 < ^-1 1 Qt>nMt>Qn>aa>-Qa>Q M ^ A (*qWtOH>-aS S i g : : : : : . --y . J3 S = a -t s ,3 5 I 1 1 1 1 1 I 4 E V u O * < K Q o> S 72 DAYS USED FOE DIUKNAL INEQUALITIES. o il t O> 00 W i-H Cfl S TO P iOirac,oi-3-co^"O^'^ s s N N a - " 2 - to S 8 R 8 s k *S 8 * CO s g ! S 8 8 S S S 5 1 05 8 S 1 S 1 88 f W 5 H SI S M N cq S S S 1 3 t> & 1 1 & s . cq c*i ^ c5 c? S 2 w * * ft 12 M S ft * ^ IB B* -' ^-( O M If S 9i M M W M -^ oo w o> ffi lfi a a s s s ?! <= V & *- S 1 * * SSS 1 g 8 2 S3 2 S 2 a 8 S g i S 2 S S 2 ^ BS aa!sa*sa ^fc,S> 00 ^ ; ' ' ' * Si * W O* C 'O OO t- N oo i- Oi l- t- t- 5-1 t-t COOOOO'O^ 1 ^" t-iOO> o A * 4-4- M>4Ofe*t*V*OOOV to " oo w & 03 P .- 71 71 r T- ' >'. >'. O^O^^OO^^O S- > t^ oo n -)"*"*" i T: * * * * * * 4- i W s s ^ -< s 5 5 COoOoOiftiftOCOO Kn-.raJjnnrao, COCOCO^'^'^^DCOt COt-^tO^< "o s M t. nn n t-c^oc^5(cac^cos>Q >ost*Qat>pa [>Qttit > QWt>-QMt>^^^' I d> 1 5 4- .S . . . . 1 - 1 . . . . ^> i- i 1 I 1 ! I 1 : * 1 ill $ I rill ? < M O !5 O INSTRUMENTS AND RECORDS. 73 DISCUSSION OF THE OBSERVATIONS. BY DR. C. CHREE, F.R.S. CHAPTER I. INSTRUMENTS AND RECORDS. 1. The magnetographs used in the Antarctic were of the Eschenhagen pattern, constructed by 0. Toepfer und Sohn, of Potsdam. As full descriptions are readily accessible,* it is unnecessary to go into details here. The Declination, Horizontal Force and Vertical Force are recorded on a single drum. Answering to each element there is a separate base line, and there is further a record of temperature from a metallic thermometer inside the box of the Vertical-Force magnet. There are thus seven traces being simultaneously produced on each sheet. The photographic paper has a width of nearly 20 cms., and so long as there is little magnetic disturbance and only small variations of temperature, and the sensitiveness is similar to that customary in Europe, the difficulty of keeping the traces separate and all on the sheet is not very serious. In the Antarctic, however, the conditions were much less favourable than is usual. The magnetic elements possessed unusually large variations, both regular and irregular, whilst the changes of the external atmospheric conditions were such as to cause large fluctuations of temperature even inside the hut where the magnetographs were in action. The magnetographs arrived in this country from Germany at so late a date that no time remained for the observer or the staff at Kew to become really familiar with them, and it was unfortunately not discovered that the quartz-fibre suspensions for the Horizontal-Force magnet were all so fine as to give unduly high sensitiveness. The sensitiveness of the Declination magnetograph is determined by the greater or less distance of the magnet from the photographic paper. In the Antarctic instrument it was approximately l'-5 per mm. of ordinate. Though rather low for ordinary latitudes, this proved much too high a sensitiveness for the conditions experienced. The Vertical-Force magnet, as actually used, was by no means too sensitive ; in fact, during most of the time its sensitiveness might with advantage have been at least doubled. It possessed, however, a high temperature coefficient, and this, in consequence of the large temperature changes encountered, tended to make the Vertical-Force trace shift across the paper. The large variations of temperature also gave a wide range to the trace from the metallic thermometer, which possessed a high sensitiveness. The observer had to attempt to keep on the sheet three magnetic traces and a temperature trace, all four tending to shift their positions on the sheet, and two of them (the Declination and Horizontal-Force traces) subject to almost incessant fluctuations, which were not at all infrequently of the order of half the width of the sheet. Under these circumstances it is not surprising that traces not infrequently got off the sheet, and that those from different elements tended at times to become confused. Troubles of this kind were most serious near Midsummer, when the magnetic movements were largest. 2. The source of light was an oil lamp which had to be filled daily. Any movement of the lamp moved the positions of all seven spots of light on the sheet simultaneously. The Vertical-Force trace, the temperature trace and the base line common to these two traces were altered together if the Vertical-Force instrument was moved. These traces could also be shifted independently by means of screws. Any movement of the Declination or the Horizontal-Force instrument altered the corresponding base line, and also, to some extent, the trace of the corresponding magnetic element. Thus the number of ways in which traces might have their position changed on the sheet was very large, while the instruments were so light that a slight touch might cause movement. As a rule, mechanical disturbances can be distinguished from true magnetic changes. The latter are seldom, if ever, absolutely sudden, and when magnetic changes occur which appear sudden on the ordinary slow-run magnetograph trace they are seldom confined to one element ; they also, of course, have no effect on the base-line traces. Thus changes due to mechanical causes in the relative positions of the spots of light, though causing extra trouble to those * ' Terrestrial Magnetism,' vol. 5, 1900, p. 59 and Plate IV. L 74 INSTRUMENTS AND RECORDS. reducing the curves, are seldom a source of real uncertainty, provided they occur when the instrument is running. It is, unfortunately, otherwise with changes that take place when the instrument is not in action, At an ordinary station, it is true, supposing only a few minutes' trace lost whilst changing the sheets, an artificial shift of a trace can usually be detected at a glance, and at least a very approximate estimate be obtained of its amount. But in the Antarctic the conditions were not favourable for detecting artificial changes. The lamp had to be filled and trimmed, and there was not infrequently an interval of over 20 minutes between successive days' records. At times a trace was off the sheet when papers were changed. It will thus be readily understood that with elements perpetually altering an alteration of 1' per minute in the Declination was quite an ordinary one a very appreciable artificial shift in the trace of an element might occur during the changing time without the traces themselves suggesting it. In the case of the temperature trace and the Vertical Force, the observer on a good many occasions made a shift intentionally when changing the papers. He would, for instance, find the temperature trace off the sheet and bring it back. The extent of these changes was seldom much in doubt, because readings from a mercury thermometer afforded a check on the temperature record, while the Vertical Force was seldom much disturbed. 3. When a change was detected or suspected, the conditions were examined into by myself, and when the existence of a change was accepted, allowance was made for it after full consideration of the circumstances. In March, April, and September, 1902, and September, 1903, there were a good many interruptions in the record. Subsequent to September, 1903, only a few records were taken, as the photographic paper was nearly exhausted. At times, in very cold weather, the oil did not burn properly, and the trace became gradually fainter and finally invisible. On one or two occasions, during a blizzard, snow got into the magnetograph room and interrupted the record, and on one occasion a magnetograph suffered through the fall of some ice which had formed on the ceiling. The only satisfactory way of treating discontinuities in the base-line values that may occur on such occasions is by reference to the absolute observations. The primary object, of course, of absolute observations at a magnetic station furnished with a magnetograph is to determine base-line values. Under ordinary conditions, if one suspects a discontinuity, one simply compares the values given by the absolute observations for the base line before and after the date of the supposed discontinuity. In the Antarctic, unfortunately, with Declination changing half a degree or more during the course of an absolute observation, the results of a single observation are of no very high precision. To obtain information as reliable as that existing, for instance, at Kew where absolute observations are taken once a week it would have been necessary to observe at least once a day. As will be seen, however, on reference to Commander Chetwynd's discussion of the absolute observations, " Physical Observations," p. 133, the number of absolute observations available was really small. 4. After these remarks it is perhaps unnecessary to say that the uncertainties entering into the absolute values of the elements as given in the various tables are much greater than would be the case at an ordinary European station. They are probably greatest in the case of the Vertical Force, V. This is derived from the Horizontal Force, H, and the Inclination, I, by the formula V= Htanl. The absolute value of V corresponding to an observed Inclination I is deduced by multiplying tanl by the corresponding value of H, as obtained from measurement of the Horizontal Force curve at the time of the dip observation. If v be the ordinate of the V curve in centimetres at this same time, and s the scale value (i.e. the equivalent in C.G.S. measure of 1 cm. of curve ordinate), then the base-line value V is given by V = H tanl - vs. H was a quantity not far from -065, while I averaged about 84|. Thus an error of 1' in I means an error of some 200y (ly = O'OOOOl C.G.S.) in the value of H tanl. Accuracy to 0'-5 in the dip derived from observations with two needles is considered good under the most favourable conditions, at places where the dip is from 60 to 70. What it is reasonable to expect from a dip circle in the Antarctic it is impossible to say on our present knowledge. There can, however, be little doubt that even if we neglect the uncertainties in the values of H and the uncertainty in an absolute observation was probably nearer 1/500 than 1/1000 of H the uncertainty in the base-line values of the V curves was too large to admit of discontinuities of the order of lOOy being detected by reference solely to the base-line values. INSTRUMENTS AND RECORDS. 75 5. The Declination magnetograph remained at a fixed distance from the recording drum and its sus- pension was unaltered, so the scale value was accepted as constant. The light from the lamp, emerging from a narrow slit, traverses a slightly convex lens forming a window to the magnet box, is reflected from a plane mirror carried by the magnet, passes a second time through the lens window, and eventually falls on the photographic paper, after passing through a hemi-cylindrioal lens a little in front of the paper. The thickness of the hemi-cylindrical lens, approximately 0'5 cm., and the focal length of the lens window, about 120 cm., exercise a slight effect on the scale value; but for practical purposes it depends essentially on the distance of the plane mirror from the photographic paper. The scale value calculated from the optical conditions in the Antarctic was 1 mm. = l' - 43. This requires, however, a small correction for the torsion of the suspension, a very fine quartz fibre. This suspension returned safely to Europe, and independent observations by Mr. BERNACCHI and the Kew staff in which the torsion head was twisted through + 180 made the torsion correction very approximately 4J per cent. The scale value for the Antarctic curves was thus taken as 1 mm. = l''50. 6. The scale values of the Horizontal- and Vertical-Force magnetographs were determined from time to time in the Antarctic by means of one or other of the collimator magnets 25A and 25D supplied with the unifilar magnetometer. In the case of the Horizontal-Force magnetograph, the magnet was placed horizontally in front of the recording drum in the magnetic Meridian. A paper being on the drum, the position of the spot of light was recorded photographically with the magnet as originally laid down and when turned end for end. The whole operation was repeated at least once on each occasion. The double deflection, i.e., the distance between the two deflected positions of the spot of light, answers to a change of very approximately 4m/r 3 in the Horizontal Force, r being the distance between the centres of the deflecting and the deflected magnets (usually at least 150 cms.), and m the moment of the magnet at the time. The change in m was small and slow, so that a sufficiently approximate value was derivable from the ordinary absolute observations. The procedure in the case of the Vertical Force was practically the same except that the collimator magnet was held vertically over the Vertical-Force magnet. The scale values thus found and the values actually employed are recorded in Table I. TABLE I. Scale Values. Value of 1 cm. of ordinate in terms of ly (ly = '00001 C.G.S.). Date. Observed Values. Values Accepted. Horizontal Force. Vertical Force. Horizontnl Force. Vertical Force. 1902 March 19 April 15 14-0 14-0 14-4 6-3 67 6-8 9-6 7-5 9-8 11 10-5 171 160 165 145 119 32 72 83 162 154 158 178 159 , . _ f From March to u | August 19 H f From August 21 to 't November 12 Q ft f From November 14 \ to January 17 H f ! From January 19 ' \ to March 31 -.(' From April 1 to 1 end ,,,- fFrom March to 11)5 < TLr ni \ May 31. 145 in June. 135 in July. 125 in August. , 1Q /From September 1 a 1 to 12. j From September 22 1 to November 12. ~ 9 I From November 14 \ to December SI. J From Januarv 1 to ' \ March 31.' , - f From April 1 to * 1 end. May 1 September 5 9 November 6 1903 March 11 . . . . . May 19 July 1 ,,27 1904 January 17 L 2 76 INSTRUMENTS AND RECOEDS. 7. On December 31, 1902, an important change was made in the Vertical-Force magnetograph. That instrument is fitted with three auxiliary magnets. Two are thin bars, almost wires, one on either side of the Vertical-Force magnet, with their centres near its level. The third is a short, much thicker bar, in a brass piece which screws on to the base of the Vertical-Force box, at some distance under the Vertical-Force magnet. These auxiliary magnets are intended apparently to assist in reducing the temperature coefficient. The use of auxiliary magnets is one of which the wisdom is open to much doubt. The results are of a somewhat complicated character, and any unrecognised change of moment or position in an auxiliary magnet may cause error and confusion. The auxiliary magnets in the present case may be described as of " soft " iron, but still they possess very appreciable " permanent " moments. In December, 1 902, the temperature changes in the Antarctic were large, and Mr. BERNACCHI, noticing a large temperature effect on the Vertical-Force trace, made a serious attempt to reduce it. After trying several less heroic remedies, he finally removed the short magnet from its place below the magnet box and placed it above, where it remained until the instruments were finally dismounted. When the tabulations were commenced at Kew, it was feared that nothing could be made of the Vertical- Force records. There were occasions when the instrument was obviously out of action, while on many occasions the Vertical-Force trace appeared so extraordinarily quiet compared to the others as to raise suspicions. Eventually I decided to have two months' trace measured, selecting December, 1902, and February, 1903, as months when the Vertical-Force changes seemed specially large and the information as to the scale values most complete. Even a superficial inspection of the sheets at times of large temperature change showed that the influence of temperature on the Vertical-Force trace must be considerable. But having had the 2-hour readings of Antarctic temperature through my hands, I knew that the regular diurnal variation of atmospheric temperature had a range of at most 3 or 4 F., and I thence inferred wrongly, as it proved that inside the Magnetic Hut the regular diurnal inequality of temperature would be so small even at Midsummer that its effect on the Vertical-Force diurnal inequality might be neglected without risk of serious error. I realised, of course, that this neglect might prejudice seriously individual curve readings, but for my immediate object that did not matter. Accordingly, diurnal inequalities were got out for December, 1902, and February, 1903, from the Vertical-Force curves uncorrected for temperature. The ranges appeared surprisingly large, and the inequality in the one month was inverted as compared to the other. This pointed to something being wrong. On inspecting the curves it was apparent that rise of temperature was associated with movement of the Vertical-Force trace down the sheet in 1902, but movement up the sheet in 1903. The explanation that may appear most natural viz., that there had occurred an actual change in sign in the temperature coefficient postulated conduct so contrary to my previous experience of magnetographs, that after reflection I inclined to the view that the Vertical-Force magnet must somehow or other have got turned end for end between December, 1902, and February, 1903. Minute inspection of the curves fixed the date of this supposed occurrence as December 31, 1902. Investigation of the written records then showed that on this afternoon Mr. BERNACCHI had made the alteration mentioned above in the position of the short magnet, and as the alteration entailed a rebalancing of the magnet whose ends are closely alike, he agreed with me in regarding the occurrence as at least a possible one. Experiment showed that the magnet worked equally well whether its N. end were east or west. The natural way of settling the question was to refer to the information on the sheets of scale-value determinations as to the position of the deflecting magnet, whether N. pole up or S. pole up. Information on this point was given, however, only on some of the sheets, and the different sheets for 1903 contradicted one another. When making the deflection experiment, the observer had to hold the magnet close up to the ceiling in the dark, and had to rely on his memory as to how he held it. Naturally it never occurred to him that circumstances might introduce an uncertainty as to the sign even of the Vertical-Force change. The difficulty, of course, would not have arisen if temperature coefficients had been determined in the Antarctic before and after changes of the instrument, but no determinations had been found practicable. 8. Though I did not at first think it possible that differences so large as those presented by the diurnal inequalities found for December, 1902, and February, 1903, could arise from temperature alone, inspection of the curves led me to suspect that the regular diurnal inequality of temperature in the Magnetic Hut INSTRUMENTS AND RECORDS. 77 during these months must have been much larger than I had supposed, and I decideil that some means must be devised for eliminating its effects. The absolute sign of the Vertical-Force changes during 1903 was, as already explained, in doubt, but there was no ambiguity either in 1902 or 1903 as to the direction in which rise of temperature deflected the Vertical-Force trace. It was thus apparent that if one could determine the change of Vertical-Force ordinate answering to a rise of 1 in temperature, one could from the trace of the metallic thermometer eliminate the effects of temperature, whatever their sign might prove to be. Thanks to the large irregular changes of the Antarctic temperature, the task proved simpler than anticipated. The method adopted was one which I have found to work successfully in several cases. The principle is that if one can get a number of instances in which there is a large change of temperature in the course of 24 hours, a coefficient calculated by assigning the 24-hour apparent change of Vertical Force to temperature alone will not be much in error. There are, of course, individual occasions when the true values of Vertical Force at the same hour on two successive days differ considerably ; but still, if one is dealing with 20 or 30 days on which no specially large magnetic disturbance has occurred, the effects of natural magnetic changes will in most instances be very nearly eliminated. In most months there were fairly copious temperature data from the trace given by the thermometer inside the Vertical-Force box, standardised by reference to the readings of the mercury thermometer. Numerical values having been obtained for the temperature coefficient, hourly measurements were made of the temperature trace, and corrections were thus obtained to the diurnal inequalities already calculated for December, 1902, and February, 1903. Considerably to my surprise, the result was not merely a large reduction of the range in each case, but a complete inversion of the inequality for February. 9. A difference of sign between the true inequalities of Vertical Force for December and February appearing highly improbable, I came to the conclusion that the hypothesis that the magnet had been changed end for end on December 31, 1902, must be wrong, and that there must in reality have been a change of sign in the temperature coefficient. To obtain further light on the subject, two direct determinations were made at Kew of the temperature coefficient, the soft-iron bar being on one occasion in the position it occupied during 1902, on the other occasion in the position it occupied during 1903. The coefficients obtained differed notably in size, but they agreed in sign. At first sight this was rather staggering, but surmising the true explanation of the phenomenon, I had a further determination made under conditions as similar as possible to those at Winter Quarters. With the aid of a number of bar magnets it proved possible to produce at the position of the Vertical-Force magnet a fairly uniform vertical field of about - '72, the natural field at Kew being about 4- - 44, and now, much to my relief, the expected difference in sign appeared. With the soft-iron bar below the Vertical-Force magnet, as in 1902, the trace went down the sheet as temperature was raised ; with the soft iron on the top of the magnet box the reverse happened. The experiments were made with both rising and falling temperatures, the readings being taken by Mr. T. W. BAKER and Mr. G. W. WALKER quite independently of me, and the results appeared quite decisive in favour of the view that the Vertical-Force magnet was not altered in position in December, 1902 movement up the sheet meaning increase of force throughout but that the temperature coefficient was negative in 1902, and positive in 1903. This view was thus finally accepted. If any additional evidence in its favour is thought necessary, it will be found in the general consistency of the results for the diurnal inequalities in Table XVIII. In several individual months the corrections from temperature to the mean diurnal inequality exert but a trifling effect on its nature, and it is out of the question that the diurnal inequalities in corresponding months of 1902 and 1903 should be the antitheses of one another. 78 BASE-LINE VALUES. CHAPTEE II. BASE-LINE VALUES, ANNUAL INEQUALITY, AND SECULAR CHANGE. 10. Table II gives particulars of the values given by the absolute observations for the base line of the Declination curves and the mean values for individual months as calculated and used. In the calculations regard was sometimes paid only to the observations of the particular month, as in May, 1902 and 1903; but in most cases the observations in adjacent months were also considered. Thus the calculated value for June, 1902, is the arithmetic mean of the observed results on May 26 and June 30. In no case was anything taken into account except the results of the absolute observations. The base line was assumed to have a constant value for each month. In passing from one month to the next the change in the assumed value of the base line introduces a discontinuity, the amount of which is shown in the last column of Table II. If the base-line value accepted for the second of two consecutive months is higher than that accepted for the preceding month, then the first midnight of the second month appears with a correspondingly higher Declination than the last midnight of the preceding month. The TABLE II. Values of the Declination Base Line. Pate. Observed values. Monthly mean values. Discontinuity. Calculated. Accepted. 1902 May 13 .. / 151 30 -3 150 3 -9 149 43 -3 149 17 '4 149 4 -5 148 59 7 148 51 '3 148 39 -6 148 35 '3 148 33 '8 149 16 '2 149 35 -1 149 40 -5 150 20 -0 150 35-2 150 36 -1 150 22 -6 148 59 -1 148 52 '8 148 40 -0 148 30 -1 148 51 -8 148 45 '6 148 47-8 148 41-6 147 52 -5 o / 150 47 -1 149 53 '6 149 7 "2 148 55 '5 148 51 '3 148 57 '6 148 59 7 149 257 150 '2 150 35 '6 150 22 '6 149 40 '8 148 59 -1 148 52 '8 148 46 '4 148 36 7 148 42 -5 148 48-2 148 47 '8 148 447 148 17 -0 / 150 47* 149 54 149 7 148 56 148 51 148 58 149 149 26 150 150 36 150 23 149 41 148 59 148 63 148 46 148 37 148 43 148 48 148 48 148 45 148 17 / -53 -47 -11 - 5 + 7 + 2 + 26 + 34 + 36 -13 -42 -42 - 6 - 7 - 9 + 6 + 5 - 3 -28 26 ... June 30 July 5 . . . , 21 . ,,22 August October 21 ... November 4 12 27 1903 January 7 30 . . . . February 10 15 March 10 May 19 June 28 July 31 September 29 30 November 2 1904 17 . * Applied also during March and April, 1902. two hours are of course the same, and the apparent difference between the Declinations assigned to them is wholly fictitious. The existence of such apparent discontinuities is of course undesirable. What they really imply in the present case it is impossible to say. They are contributed to, no doubt, by observational errors ; but even in the Antarctic it seems unlikely that errors of more than 5' or 6' would arise in absolute observations BASE-LINE VALUES. 79 taken by a practised observer like Mr. BKRNACCHI. On the other hand, it is difficult to imagine what natural cause could lead to alterations of 50' or even of 30' per month in the base-line value of a Declination magnetograph, the alteration continuing for several months in the same direction. The base line, it need hardly be mentioned, is due to light reflected from a mirror, which is supposed to be fixed, on to photographic paper on a drum which is also fixed. In the Kew magnetograph it is doubtful whether the total alteration in the base line between 1890 and 1900 attained to as much as 1'. 11. Table III gives the values deduced for the base line of the Horizontal-Force curves from individual observations, the values thence calculated for individual months, and the values actually used. The last column gives the excess of the base-line value for each month over that for the previous month. As already explained in the case of the Declination, the differences between the base-line values accepted for successive months appear as discontinuities modified in April, 1903, by a change of scale value between the values assigned respectively to the first midnight of a month and the last midnight of the previous month. As with the Declination, it is difficult to account for the large apparent variations in the base-line value. Owing to the limited number of absolute observations, observational errors doubtless come in ; but they can hardly account for any large fraction of the larger discontinuities. The collimator magnets used for the absolute observations were old, and there was but little change in their magnetic moments, especially in that of 25A, the magnet chiefly employed. If the absolute observations had been faulty a circumstance improbable in view of Mr. BERNACCHI'S experience this would have shown itself through irregularity in the values given by the individual observations for the moment of 25A. The accordance, however, in the values for the moment is satisfactory and suggests that the probable error in individual absolute observations of Horizontal Force was at most from lOy to 20y. Moreover, errors in the absolute determinations would naturally vary irregularly in sign, while the apparent monthly changes in the values of the base line in Table III appear on the whole systematic. TABLE III. Values of the Horizontal-Force Base Line. Bate. Observed values. Monthly mean values. Discontinuity. (Unit ly.) Calculated. Accepted. 1902 April 17 06371 06228 06314 06375 06414 06608 06626 06515 06504 06247 06207 off sheet 06214 06523 06619 06768 06810 06244 06371 06271 06345 06394 06511 06608 06626 06510 06298 06207 06210 06214 06368 06523 06600 06656 06731 06778 06800 06697 06470 06244 06370 06270 06345 06395 06510f 06610 06625 06510 06300 06205 06210 06215 06370 06525 06600 06655 06730 06780 06800 06695 06470 06245 -100 + 75 + 50 + 115 + 100 + 15 -115 -210 - 95 + 5 + 5 + 155 + 155 + 75 + 55 + 75 + 50 + 20 -105- -225 -225 May 12 ... 26 . . June 30 July 23 August .... September 5 October 21 November 12 12 December 27 1903 January 7 30 . February March 10 April May 19 June 28 July August 31 September ... . November 2 . December 1904 January 17 * Applied also during March. May base-line value used after Noon on April 22. t After August 20, September base-Hue value applied. 80 MONTHLY MEANS. 12. Table IV shows two sets of mean values of Declination for individual months derived from the hourly readings, accepting the base-line values given in Table II. The first set of data are derived from all the days of complete registration ; the second set are derived from a smaller number of days, selected as the least disturbed of the month. The differences between these two sets of values given in the last column would be unaffected by any alteration in the monthly values accepted for the base lines. The smallness of these differences seems to justify the conclusion that they are not seriously affected by the in- cidence of magnetic disturbances or by uncertainties in the measurements of individual days' curves. Whilst the differences are small, the all-days' mean appears in excess in so large a majority of cases as strongly to suggest that the phenomenon is not a purely accidental one. Differences between mean values derived from all and from quiet days have been observed elsewhere, though not of so large a size. TABLE IV. Mean Monthly Values of Declination (from the Curves), employing the Base-Line Values accepted in Table II. Month. Monthly means from Excess of all-dajs" mean. All daya. Quieter days. 1902 Mav . o / 153 39 -6 153 2 -7 152 37 -0 152 37 '9 152 40 -0 152 54 -2 152 49 -2 152 46-1 152 45 "2 153 0-9 152 46-1 152 19 -4 151 56 -5 152 10 -6 152 17 '8 152 30 '2 152 48 '7 o / 163 41 -1 153 1 -1 152 36-7 152 37 7 152 37 '5 152 49-5 152 48 '9 152 42 '9 152 47 '9 152 59 -7 152 45 '2 152 21 '5 151 56-9 162 7 '2 152 16 '1 152 22 -1 152 45 '3 / -1-5 + 1-6 + 0-3 + 0-2 + 2-5 + 4-7 + 0-3 + 3'2 -2-7 + 1-2 + 0-9 -2-1 -0-4 + 3'4 + 1-7 + 8-1 + 3'4 V " June July September October 1903 May . . . , * July . Mean excess of all-days' over quieter-days' mean + 1-46 13. Table V gives the mean monthly values of the Horizontal Force as derived from the hourly values in days of complete registration, accepting the base-line values given in Table III. TABLE V. Mean Monthly Values of Horizontal Force (from Curves), employing the Accepted Base-Line Values in Table III. Month. 1902. 1903. 06344 06334 March 06335 06558 06534 May . 06503 06673 " *} June 06584 06735 July . 06622 06780 06721 06838 06718 06873 October 06749 November 06627 = 06437 SECULAE CHANGE. 81 14. A comparison of the mean values of an element iov corresponding months of consecutive years enables an estimate to be formed of the rate of secular change. Table VI shows the results thus found for D and H, accepting the monthly mean values given in Tables IV and V for the five months May to September. The data subsequent to September, 1903, were too few to give representative results, while during March and April, 1902, matters were still somewhat in a preliminary stage. TABLE VI. Secular Change. Mean value from month of 1903 Mean value from same month of 1902. Month. Declination. Horizonal Force. May -1 43-1 -0 52 -1 -0 19-2 -0 7-7 + 87 + -00170 + -00151 + -00158 + -00117 + -00155 June .... . ... July . August September ... . -0 34-7 + -00150 The results for the secular change in H show a rather unexpected consistency. The only suspicious feature in the figures, as figures, is the extraordinarily large size of the apparent annual change, representing as it does nearly 2| per cent, of the absolute value of the Horizontal Force. If the mean is a true measure of the secular change, the natural inference is that the south magnetic Pole is receding from Winter Quarters i.e., is moving northwards at a rapid rate. The Declination figures are no less remarkable, but appear much less consistent. Starting with an apparent decrease of 103' -1 in the year ending with May, 1903, we finish with an apparent increase of 8' '7 in the year ending with September, 1903. Such a phenomenon seems hardly credible, and one cannot but suspect some instrumental source of error. In the case of the Declination, as already stated, there is no apparent reason why the base line should change, and it is conceivable that some seasonal change may have influenced the absolute observations, especially in view of the apparent inconsistencies in the azimuth readings obtained for the distant mark ("Physical Observations," p. 139). I have thus thought it worth while to ascertain what results would be obtained for the secular change of Declination if the base-line value were assumed to be invariable. On this hypothesis the results given in Table VI are replaced by those in the following Table VII. TABLE VII. Secular Change of Declination. Mean Talue from month of 1903 Mean value from same month of 1902. May. June. July. August. September. Mean. + 4' -8 + 8' -8 + l'-7 + 11' -2 + 16' -6 + 8' -6 The results given in Table VII are the antithesis of those in Table VI, and are more consistent amongst themselves. Whilst Table VI suggests that the south magnetic Pole is moving towards the west, Table VII suggests that it is moving towards the east.* 15. In even the best European stations two years is too short a period to give results of a really repre- sentative character for the annual inequality of the magnetic elements, i.e. the variation that remains in the * The results obtained for the secular change by Commander CHETWTND, R.N., from the absolute observations alone, disregarding diurnal variation, were for Declination 26' "4, for Horizontal Force +130y ("Physical Observations," pp. 137, 140). 82 ANNUAL INEQUALITY. mean monthly values after an allowance has been made for the effects of secular change, assumed to proceed at a uniform rate throughout the year. This is due, at least in the case of the Declination, to the fact that the regular annual inequality in temperate Europe if not absolutely nil is exceedingly small. Table VIII gives the results obtained for the annual inequality at Winter Quarters. In obtaining the Horizontal-Force results, the mean monthly values were taken from Table V, and the secular change accepted was the mean given in Table VI. The Declination results under both (i) and (ii) are based on all the days of registration. The results under (i) accept the mean monthly values given in Table IV, and the mean value given in Table VI for the secular change, i.e., they answer to a variable base line as given by the absolute observation. The results under (ii) assume the base-line value to be invariable, and the true value of the secular change to be the mean given in Table VII. TABLE VIII. Annual Inequality. Month. Declination. Horizontal Force. (Unit 1-y.) (i.) (ii.) January / + 2-5 + 21 -1 + 9-2 -14-6 - 0-4 - 8-9 -15-3 - 5-8 -l- 7'4 + 2-9 + 0-8 + 0-6 -28-5 -49-5 -52-0 -37-4 -16-6 + 0-8 + 17-8 + 33-7 + 42-3 + 44-8 + 37-0 + 7-2 -237 -260 -271 + 3 + 32 + 91 + 120 + 186 + 189 + 205 + 71 -132 March April May . "-3 July . August October November The range given for the annual inequality by any set of figures in Table VIII is simply enormous compared to anything that exists in ordinary latitudes. The two sets of figures for the Declination are far from similar. Those under (i) look at first sight the less improbable, as giving the smaller range. It should, however, be remembered that at Winter Quarters 1' of arc in Declination answered to only about 1 - 9y in force, so that the range under (ii) when converted into force is only about 184y, or 40 per cent, of the apparent range in Horizontal Force. Thus if a large range in an annual inequality is regarded as too improbable a result to be accepted, the argument is not so strong against the Declination results under (ii) as against the Horizontal-Force results. 16. If we take a Midsummer mean from the months November, 1902, to February, 1903, and a Midwinter mean from the months May to August of both years combined, both the epochs concerned centre at January 1, 1903, so that the results are free from the uncertainty as to the real value of the secular change. The results thus obtained are as follows : Season. Declination on hypothesis of Horizontal Force. Variable base line. Filed base line. o r 152 50 '3 152 36 '5 / 06436 06682 Excess of Midsummer . . + 13 -8 -17-4 - -00246 VERTICAL FORCE BASE-LINE VALUES. 83 In view of what has been already said and of the difference between the Declination results under (i) and (ii) in Table VIII, it is, perhaps, unwise to say more than that it is desirable that the attention of the observers of the next Antarctic Expedition should be called to the importance of making a careful study of the annual inequality. If an inequality with a range of the order suggested by Table VIII should be established, it would be a most important result, strongly suggestive of an annual oscillation in the position of the S. Magnetic pole. The question may be asked why a second set of results answering to those in Table VII and to those under (ii) in Table VIII has not been given for the Horizontal Force. The reason is simply that the Horizontal-Force base line inevitably changes with time as the moment of the suspended magnet alters, and there is the further reason that a discontinuity arose more than once through breakage of the suspension, or similar disturbing cause. A diminution in the moment at the rate of 1 per cent, per annum would have had at Winter Quarters the same apparent effect as a secular change of 65y per annum, and it is by no means improbable that the change of moment amounted to several per cent, as the magnet was exposed to large and numerous changes of temperature. 17. Table IX shows the Vertical-Force base-line values observed (i.e. derived by combining observed values of dip with the corresponding values given by the Horizontal-Force curves) and those actually employed. As already explained, the probable error in the value of V (Vertical Force) derived from a single observation of dip is very large. This made it advisable to derive base-line values from the observations of a number of months combined, when this appeared feasible, allowing for apparent curve discontinuities. Thus the base-line values up to September 12, 1902, were determined by combining the absolute observations of dip made in April, May, June, July and September. The base-line values from October 1 to November 12 depend on the absolute observation of October alone. The base-line values for the latter part of November, 1902, and up to the end of January, 1903, depend on the observation of December. The base-line value for February, 1903, is from the absolute observation of that month. From March to October, 1903, the base-line values depend on the observations of March, May, June, August and September combined, allowing for apparent curve discontinuities. The base-line values for November, 1903, and January, 1904, depend on the absolute observations of these respective months, and the base-line value for December, 1903, was interpolated. It is obvious from what has been already stated that the Vertical-Force base lines for individual months are affected by uncertainties which would render any deductions as to secular change or annual inequality of very problematical value. TABLE IX. Vertical-Force Base-Line Values. 1902. 1903. Month. Date of observation. Observed value. Accepted value. Date of observation. Observed value. Accepted value. 7348 10 73407 7341 March 7349 10 72508 7301 April 17 72720 7349 7414 May 12 '73506 7331 19 73933 7386 30 73229 7302 28 73574 7368 July 23 72430 7290 '7357 72S3 31 73740 '7336 September .... 8 21 73548 '73111 7273 73 11 30 73711 7355 7352 *Noyember .... December 28 27 73278 | '73142 7311 7321 7314 2 73348 7335 7306 Jan. 17, 1904 72770 7277 * Of the two values given for November, 1902, the fivpt was applied up to and including tlie 12th, the second for the reet of the month. M 2 84 TEMPERATUEE CORRECTIONS. 18. Table X contains data relating to the temperature correction to the Vertical Force. They were obtained by intercomparing the temperatures recorded by a mercury thermometer adjacent to the Vertical- Force instrument, at the times when the sheets were put on and taken off, with the readings of the temperature trace and Vertical-Force curve at the beginning and end of each sheet. Sometimes, of course, the natural values of V at the beginning and end of a day's trace differ considerably owing to magnetic disturbance. Again, at times when temperature was changing rapidly it is improbable that the temperature of the mercury thermometer, the metallic thermometer and the Vertical-Force magnet were really identical. Uncertainties also arose from discontinuities in the temperature trace and Vertical-Force curve. Comparing the changes during the 24 hours in the temperature trace with the corresponding changes in the readings of the mercury thermometer, one obtained the number of millimetres in the temperature trace answering to 1 C. The mean results thus obtained from the curves of individual months form the first column of Table X. Taking everything into account, I concluded that from March, 1902, to the end of February, 1903, there was no evidence of any real change in the scale of the metallic thermometer. Accordingly the mean value 16 mm. per 1 C. was accepted for these twelve months. Subsequent to February, 1903, for some unknown reason, the temperature scale was much less open and at the same time more variable. This continued until nearly the end of the year, when the scale again reverted to nearly its original value. During the last three months a common value of 14 - 8 mm. per 1C. was accepted, but from March to September, 1903, use was made of the individual monthly values actually found. TABLE X. Vertical Force. Temperature Correction. Month. Temperature scale, equivalent of 1 C. Correction to apparent V per 1 C. rise of temperature. (Unit 17.) Correction answering to 1 mm. change of ordinate of temperature curve. In millimetres of V-trace. In force. (Unit 1 7 .) 1902 mm. 14-7 15-8 15-7 16-3 16-8 14-4 16-5 16-6 16-2 19-8 15-7 16-4 16-4 5'9 4-1 3-8 3-6 3-3 3-5 3-1 1 14-8 J + 41 43 40 36 30 40 34 22 20 35 + 35 -21 17 17 22 18 14 22 21 15 -22 0-154 0-165 0-152 0-155 0-139 0-200 0-178 0-426 0-500 0-30 0-30 0-16 0-125 0-355 0-327 0-291 0-240 0-414 0-371 0-298 0-361 0-094 2-5 2-7 2-5 2-2 1-9 2-5 2-1 1-4 1-6 2-2 2-2 1-3 1 -0 2-9 53 4-7 3-9 6-7 6-0 4-8 5-8 1-5 May July . August September (to 12th) .... October November (to 12th) .... (after 14th) . . . 1903 February May . T J July . August September October December 1904 Comparing the daily changes in the temperature trace with the corresponding changes in the Vertical- Force ordinate, one got for each month assuming the effects of magnetic disturbances to neutralise one another the apparent change of Vertical-Force ordinate in millimetres answering to a change of 1 mm. in temperature trace ; while comparing the daily changes in the readings of the mercury thermometer with the corresponding changes in the Vertical-Force ordinate, one got the apparent change of Vertical-Force TEMPERATURE CORRECTIONS. 85 ordinate in millimetres for a change of 1C. From a mathematical standpoint the two sets of results just mentioned are not independent, the one being doducible from the other by means of the relationships given in the first column of Table X. In reality, however, the two sets of results were to some extent independent, because there were in all months days sometimes a good many days in which information was lacking as to the mercury temperature, the temperature trace or the Vertical-Force curve, and the loss was sometimes of one element, sometimes of another. The third column of Table X is in most months based on both sets of results (i.e. on Vertical-Force and mercury-temperature changes as well as on Vertical-Force and temperature-trace changes), allowing most weight to the direct Vertical-Force and temperature-trace comparisons. The second and fourth columns in Table X are calculated from the third column, employing the accepted scale values of the Vertical-Force curve and the temperature trace. Any uncertainty in the scale value of the Vertical-Force curve thus affects these two columns, but it does not and this is important have any influence on the accuracy of the temperature correction actually applied to the readings of the Vertical-Force curve. The ordinates of the Vertical-Force curve and temperature trace were read in millimetres, and for each millimetre change of ordinate of the temperature trace a corresponding correction in millimetres was applied to the Vertical-Force ordinate. There is not the slightest doubt that the application of the temperature corrections immensely improved the accuracy of the Vertical-Force results in fact, in some of the Midsummer months the results if uncorrected for temperature would have been practically useless but considering the many changes of scale value and other sources of uncertainty it would be too much to hope for complete success. 19. The uncertainties remaining after the application of the temperature correction increase, of course, with the probable error in the calculated value of the temperature coefficient, which there is no satisfactory means of estimating, but they are equally dependent on the range of temperature of the Vertical-Force magnet. So far as the regular diurnal inequality of V is concerned, it is only the regular diurnal change of temperature that counts. It is the range of this regular diurnal change that appears in the first column of Table XI. It is derived from the temperature trace on those days which were actually used in deducing the diurnal inequalities of V in Table XVIII. TABLE XI. Month. Eange, C. Hours of Equivalent of temperature range. (Unit l^.) Kange of corrected V-inequality. (Unit I?.) Highest temperature. Lowest temperature. 1902. 2-38 1-10 1 '40 0-53 47 0-46 1-37 1-31 4-16 6-33 4-31 0-40 0-37 0-35 0-24 0-46 0-39 0-36 0-25 4 p.m. 2 6 7 a.m. 6 midnight 7 p.m. 6 3 p.m. 3 p.m. 2 p.m. 2 and 3 p.m. 2 p.m. noon 11 a.m. and 3 p.m. 1 p.m. 1, 5 and 6 p.m. midnight ,, noon 10 p.m. midnight noon 11 a.m. 6 4 a.m. midnight 1 a.m. 9 p.m. 7 1 a.m. 7 and 8 p.m. 2 a.m. and 6 p.m. 7, 8, 11 a.m. and noon 97 48 66 19 14 18 46 29 145 135 75 9 7 5 5 10 6 16 8 6 37 46 19 22 15 22 26 27 46 51 80 52 52 35 25 19 33 39 40 44 66 130 May July 1903. April May July September November December 86 TEMPERATURE CORRECTIONS. The second and third columns give the hours at which the highest and lowest mean temperatures occurred. The fourth column shows the equivalent of the temperature range in the first column, in terms of the correction to V; for comparison with this, the fifth column gives the range of the corrected V diurnal inequality. To illustrate the table, take the case of December, 1902. The Vertical- Force magnet had its highest temperature at 3 p.m., its lowest at 4 a.m., the difference between the two being 4" 16 C. The correction to the apparent value of V necessary to eliminate the effects of temperature was greater at 3 p.m. than at 4 a.m. to the extent of 145y, or about 2 '8 times the total range obtained for V after applying the correction. In the Midwinter months the temperature range was small, and the hours of highest and lowest temperature varied much from month to month. Thus, the V-inequality results for the Midwinter months, and especially that for the whole season, should be but little influenced by temperature uncertainties. Subsequent to March, 1903, the regular diurnal variations of temperature in the Magnetic Hut were small. It is thus satisfactory that the inequalities finally deduced from the 1902 and 1903 curves agree in type. DIURNAL INEQUALITIES. 87 CHAPTEE III. DIURNAL INEQUALITIES. 20. The questions of secular change and annual inequality have been treated out of their natural order because they appeared so closely connected with instrumental questions that it was desirable to discuss them in that connection. Before passing to the diurnal inequalities, it will be convenient to deal with some points relating to the measurement of the curves and the construction of the tables of observational data (pp. 8-70).- After examining the curves, I came to the conclusion that if the value assigned to a particular element at a definite hour were derived as usual from a single measurement of the curve ordinate at the exact hour, the irregularities arising from " accidental " disturbances would probably conceal the regular diurnal inequality altogether, or at least in great measure. To meet this difficulty, it is customary at some places to smooth curves by drawing a pencil trace, giving the general trend, as distinguished from " accidental " irregularities. An alternative plan, which has been sometimes suggested, is to determine mean ordinates for successive hourly intervals of time by planimeter measurements. Owing to the excessively disturbed character of the Antarctic Declination and Horizontal-Force curves, neither of these alternatives appeared feasible. Eventually I decided not to smooth the curves in any way, but to take as the ordinate at any hour the arithmetic mean of three ordinates, one exactly at the hour, the others at 20 minutes before and after. This necessitated the measurement of the curves at 20-minute intervals throughout the day. A suitable glass scale was constructed by Mr. FOSTER, one of the senior assistants in the Observatory Department. The ordinates on the scale were divided at millimetre intervals and, when reading, 1 mm. was usually aimed at, though this degree of accuracy is not claimed in the results. Readings were in all, or nearly all, cases repeated, usually by a different observer, and in the event of serious discrepancy a third measurement was made. Owing to the time occupied in changing papers, it was a frequent occurrence for one of the 20-minute readings to be lost, and, at some seasons, intervals exceeding 20 minutes were not unusual. The way of dealing with these gaps varied according to circumstances. Supposing, for instance, the reading at an exact hour missing, the value assigned to that hour was usually the mean from the readings at 20 minutes before and after ; but if a gap commenced or ended within a few minutes of an exact hour, the readings taken 20 minutes before and after the hour might be combined with an interpolated value, intended to represent the missing reading. The precise procedure to be adopted was determined by myself with the curves before me. In the case more particularly of the Horizontal Force, the limits of registration were exceeded rather frequently, especially in Summer. If this happened at, say, 20 minutes before the hour, the curve coming on the sheet at, say, 10 minutes to the hour and continuing on for some time, the value given in the table of hourly values was the mean from three readings which corresponded to the exact hour and to 10 minutes before and after. If, however, a limit of registration were exceeded for an appreciable time at the exact hour, the entry in the tables normally indicated an excess of the limit, precisely as it would have done if the limit had been exceeded not merely at the hour but at 20 minutes before and after as well. 21. In the case of the Declination, measurements were always taken to 0-1 mm. answering to 0'-15 but decimals are discarded in the hourly values in the tables as not really warranted by the accuracy attainable. In the case, however, of the diurnal inequalities (pp. 90-99), the data being means from a number of readings are given to the nearest 0' 1. In the Horizontal-Force tables hourly values are given to ly, and diurnal-inequality data (pp. 94 and 95) to O'ly. This degree of accuracy was fairly warranted, so far as mere uncertainties of reading are concerned, owing to the very open scale. It is, however, I freely admit, open to criticism on the ground that no temperature correction has been applied. What the exact degree of uncertainty on this ground may be it is difficult to say. None of the suspensions used in the Antarctic survived, and no inference seemed capable of being drawn by application of the method employed in the case of the Vertical Force. All that inspection showed was that the temperature coefficient was small, its sign even not being disclosed. Even if it had been large, the continually disturbed state of the Horizontal Force would have rendered any high accuracy in its determination impossible. We know that the rigidity of the quartz suspension 88 DIURNAL INEQUALITIES. would rise with increase of temperature, while the magnetic moment of the magnet would naturally diminish ; thus, presumably, the correction required was positive when temperature was above its mean. At Kew, quartz suspensions used in the Watson type of magnetograph have had temperature coefficients of from 3y to 6y per 1 C. But in the Antarctic the Horizontal Force was only about a third of that at Kew, so that correspondingly less torsion was put into the suspension, and the presumption, accordingly, is that the temperature coefficient was considerably lower than those found at Kew. The absence of a temperature correction may appreciably influence the diurnal inequality in two or three of the Midsummer months. At other seasons its influence is hardly likely to be of any importance, except in the case of individual hourly readings on days of large temperature change, In the Vertical-Force tables the hourly values are given only to the nearest lOy, but the absolute maximum and minimum are given to the nearest ly. The reasons for this are as follows : Owing to dislocations in both the Vertical-Force and the temperature traces there were at times somewhat large uncertainties as to the relative values of the base line at different parts of the same month. It was thus felt that the retention of five significant figures in hourly values represented far more than the accuracy attainable. There was also the consideration that the scale was very contracted in most months, and the further consideration that readings to the nearest lOy sufficed to give diurnal-inequality data, going to ly. When the Vertical-Force curves were first tabulated it seemed unlikely that anything beyond diurnal inequalities would be attempted. In fact, it was not until the absolute ranges in Declination and Horizontal Force had been analysed that it was decided to attempt to obtain corresponding data for the Vertical Force. It was obvious, however, that in most cases accuracy to nearer than lOy was obtainable in the ranges because the uncertainties attending absolute values of the base line largely disappear when considering differences of readings taken on the same day and accordingly it was decided to retain five significant figures in the values of the daily maximum and minimum. It was often exceedingly difficult to say which of several very nearly equal ordinates (allowance being made for temperature) represented the true maximum or minimum. The times of occurrence were not measured very exactly. 22. Before dealing with the data obtained for the diurnal inequalities reference must be made to one source of uncertainty to which attention has already been drawn by Commander CHETWYND (" Physical Observations," p. 134). Winter Quarters was a station at which there was appreciable local magnetic disturbance. Observations made on the ice in McMurdo Sound, about If miles from the Magnetic Huts (see Frontispiece), gave for the Horizontal Force a value of -0433 C.G.S., or only about two-thirds of that found at Winter Quarters. The differences in the other elements were much less. The Declination at the Ice Station was about 5 less and the Inclination 1| greater than at Winter Quarters. As the Ice Station was in fairly deep water, the presumption is that these differences represent local disturbance at Winter Quarters. To the question what the effect of such local disturbances is on diurnal inequalities one cannot give a positive answer. If we suppose diurnal inequalities to be due to overhead electric currents, as Dr. SCHUSTER'S mathematical calculations indicate, and as is most generally supposed, and if no sensible diurnal variation is caused by solar radiation or similar cause in the local disturbing field, then the presence of the local disturbance will influence only the Declination diurnal inequality, the amplitude of which will vary inversely as the value of the Horizontal Force. If this view is correct, then the only effect of the local disturbance at Winter Quarters would be to reduce the amplitude of all Declination changes in the ratio approximately of 2:3. It must be admitted that there is little, if any, positive evidence of the correctness of the view. There is some evidence, on the contrary, that in highly disturbed regions the effect on the diurnal inequality is much more complicated. But the local disturbance at Winter Quarters was not really large. An increase of 50 per cent, in the Horizontal Force is not, of Course, small from one point of view, but an increase of 02 C.G.S. does not mean anything very much out of the way at a place where the total force exceeded 7 C.G.S. If the source of the disturbance were basaltic rock close to the surface as seems most probable from the description of the station there would not appear to be much reason to fear anything more than a reduction in the Declination range. Under such circumstances, no doubt, direct heating by the sun would have some effect at Midsummer, but it could hardly be large. Even at the surface the range of the regular diurnal inequality of temperature, when largest, was only about 4 F. DIURNAL INEQUALITIES. 89 23. Tables XII to XXII (pp. 90 to 99) give the results obtained for the diurnal inequalities. In the case of the Declination two sets of results are given, the one derived from all days available, the other from the quieter only of these days. Table XII gives the all-days' Declination results for individual months. Table XIII gives the corresponding results for the twelve months of the year, taking a mean from 1902 and 1903 in cases when data from both years were available ; it also gives diurnal inequalities for the year as a whole and for three seasons, Midwinter (May to July), Equinox (March, April, September, and October), and Midsummer (November to January). Tables XIV and XV give corresponding results from the quieter days. For November, 1902, an additional inequality is added in Table XIV which excludes only the five most highly disturbed days, and so is intermediate between the inequalities based on " all " and on "quieter" days. Midwinter was limited to three months, so as to include only days throughout the whole of which the sun was below the horizon ; in Midsummer, on the other hand, the sun never set. The Horizontal Force and Vertical-Force data were treated similarly to the Declination, except that no quieter days' inequalities were formed. Tables XVI and XVII relate to the Horizontal Force, Tables XVIII and XIX to the Vertical Force. In the case of the Inclination, inequalities have to be calculated from the corresponding Horizontal -Force and Vertical-Force inequalities. It appeared sufficient to give a single table, Table XX, containing results for the twelve months of the year, and the three seasons, employing data from both 1902 and 1903 when available. Tables XXI and XXII give the diurnal inequalities of the components of the Horizontal Force, respectively in and perpendicular to the astronomical Meridian ; results are given only for the three seasons and the year. These were calculated from the corresponding inequalities AD and AH in Declination and Horizontal Force by means of the formulse AS = cos is the supplement of the easterly Declination, and A denotes the departure at any given hour from the mean value of the day. The mean values accepted for < and for H, and the force equivalent to 1' in Declination, were as follows : Midwinter. Equinox. Midsummer. Year. (j> 27 23' 27 12' 27 13' 27 17' H . . . . 06650 06606 06469 '06573 Equivalent of 1' in D . . 1-937 1-927 1-887 1-917 The working equations thence actually deduced, AH, AS, and AW, being measured in terms of ly as unit, and AD in terms of 1', were as follows : Midwinter. Equinox. Midsummer. Year. AS = AW = . . . . -89 (AH + AD) -0-46 AH + 1-71 AD -89 AH + -88 AD -0 -46 AH + 1 -71 AD -89 AH + -86 AD -0-46 AH + 1-67 AD -89 AH + -88 AD -0 46 AH + 1-70 AD 24. In all the diurnal inequality tables the algebraically largest and least of the hourly values are in heavy type. The tables also contain the ranges as derived from the largest and least of the hourly readings, and the sum of the 24-hourly differences from the mean for the day. In calculating the diurnal inequalities use was made in general only of days in which the record was complete from midnight to midnight. In some months, where the number of days of complete record was small, hourly values were interpolated when a gap of only an hour or two occurred in the record during a relatively quiet day. In some cases where material was scanty, use was made of periods of 24 successive hours which commenced at hours other than local midnight. This happened, for instance, with the Vertical Force in November and December, 1903. In the latter month the record extended from 2 p.m. on the 12th to 1 p.m. on the 16th, and by making the day start at 2 p.m. four complete days' record were obtained. Particulars of the days employed for the various inequalities are given in the table on pp. 71 and 72. N 90 DIUKNAL INEQUALITIES. V+|i| te*afi'* yp^p.pppy fill' ti i a a-irl i i i s i i i i i 3 1 *I|MIM1S1 I I U I I I I I j i i i i + i i + + i i i i i i i i i i ^-Mao3>tO-*CO n?pt-t-n>? rt ^ec-*. t-Ot0500U3rHino s ^lOOO-*lO-*-*t^)OC^ COCiOirHC^r-(Tj<^0} 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 OOrt-fflNO> oi ~Ot>OOt- - -f o co o i- 00 ^i-HOOO^Hi-llot--^OO SS2SSS2S 1 1 1 1 1 1 1 1 1 1 1 l l l l l l i i | 1 1 1 1 1 t 1 1 1 1 c^cooier-irHNWc-3 1 1 1 1 1 1 1 1 1 5 - 7 7 7 i 7 i i i i i T i 7 7 7 7 i 7 i 7 7 i i 7 i 7 i i i i i i 7 7 7 7 7 i IB.4rttCrHOinoOtO.13 wooo.nwt-a> ^ 7 7 i i i i 7 i i i i i i i 7 7 7 7 i ri '? T T ? 7 S T ? ? ? 77 af 7 f t 7 ? .|2SSSSSI SSSSSSSSi i i + + + + + i i + ^SSw2SNS OrtUJrtt-OO X> <0 TOi-cq^oeicccorH 2 >-iee-mw ;?!++???? *rtt-0tDCO tOCOOVONt-000.0 2 ^r-lt-(lOt-.OOQOQONaiO c^ift^io^cqroooo + + -). + + + + + + -l- + + + + + + + + + s -ooS^^goSmSSS Hiii Ilii eooooit-tDoo-^w d! r-l M Bl 3 C4C9C4i-o-ot-a>n -*pHrJ.ODiONOOCT- CO ~?? + + +??T?f ssassssss g ^ -^rtSiiiSSSI SSSSSi!2SS s + + + + + + + + + + + + 4- + + + + + + <*aM4b.>.nfi -OOOOCO3irat~W5i-HO) eo cc ~- ii ^- < i c't + + + + + + + + + + + + + + + + + + + u~> ~ISi2SSSSi| SSSS^SSSi + + + + + + + + + + + + + + + + + + + t-OOr S *3aK?3tSii! " >. 1 1 1 ! 1 1 ! ! I IB a J DIURNAL INEQUALITIES. 91 oj.. d *tO-f'N.-<0t-.-l-Ji-.ji >- * 5 ESg I*J, 1 1 s C>J O JO 1 1 1 1 1 1 1 1 1 ++ 1 M ^H O -^ 1 1 1 1 CX|tOOOO-'*eO<0iracOt0' ift 7 i i 7 ^r-li-iGOOOiiOCO-*ONCO ^H rt (D o> *5jCOiOCNO-irHTftOO<>ift S S 3 55 1 1 t 1 1 1 1 1 1 1 1 1 1 1 1 I o r- _-.'---- QO : to >-. .-. Ci lO TO O 00 "S t oo iow> iO eo i-t t- go O> Tf GO lO 1 1 1 1 1 1 1 1 1 1 1 1 1 t 1 1 cDoo^eoroaa^r-iooot 01 W CO | t^ ^OOt-iWt- "J Tf(O tt>.O Ol-O crroSi-in-i^-.fHr-.wiNc^co 1 1 1 1 1 1 t 1 1 1 1 1 Si 3 8 SS 1 1 1 1 '^wc-.^(ot-eo i t i i i i i t i i 7 i rH CO * tO 7 i 7 7 oioiomoo<-ioiMooo 00 01 00 0> w *,OiHO>iOi-iiHN^wnt~-eo + III1++1IIII CO O to 1 + 1 1 t-tHi-MiNfr-'*OS'^(3^i ^'ODt 1 "0&CO(5C l 5t- N W -J * ^T*5ooiot-ie-fl*ecTj<'rf at n v-t eo O4 - '- I~ "^ co N ~ z i- m o ' t- nip>f9AA?nilliB + + + + + + + + + -f + + 8 S 8 8 + + + + -*-"^oi^wcoeiioi-ciW O t-- Oi C^ CO ^i?S'^!l* oo '~ IOit " ooo N neCC^Ir-lrHt-l r-l .-I r- CO + + + + + + + + + + + + j-< 0> CJ O4 fH i-4 CO + + + + iOt-Ha>?4kO03c&>-io<-H&cn e* w o t- i t- -^^^ss^^s^sss + + + + + + + + + + + + S S 2 S + + + + | cot-cq-^ 03 m * - -.OeOOlTOOit~o>OOiOFHC^ Ci~t-tOTj CO O> CO + + + + emaoiftrHCOoo.-(ir3OSOO- > | & fe S s i i 1 1 fr i ^ 1 1 i 1 1 ^sl^l^^^l^ll s 8 a J .S g I I ij 1 o 3 c8 N 2 92 DIURNAL INEQUALITIES. (M W ^ mean. -.0 , s 8 a Oi rH i? Oi Ol * i CO t- e5 3 I-H CO g s * CO 11 00 00 iO SiQ Oi 00 CO t- O rH CO Ol 04 01 04 cr- oc Oi 01 01 p oo rH CO p op 04 iO Ol O 01 rH 04 CO CO 04 C? f 5 Ol us oi 00 1 00 C4 1 + rH + 1 CO CO t- 1 + I-H tO + 1 r-. CO fc- 7 i rH O 1 1 rH O O 1 1 1 rH CO rH CD + 1 00 e H -, CI 1 o o O4 rH 1 1 O rH + 1 00 t* 1 + CO C-l 04 o i 7 CO t- 7 i CO rH 1 1 O4 Ji CO 1 1 1 o to 1 1 CO 1 d ^ iO 1 CO 04 1 1 ? :' IN CO 1 1 9 ^ 1 1 CO CO CD CO i 7 CO CO 7 i it CO 1 1 -* rH CO 1 1 1 Ol OO i 7 1 01 1 US CO 1 1 o p 1 1 1 1 C? rH i 7 So 01 1 1 Oi CO CO t- 1 1 t- iO O i i 7 W U3 o t- i 7 Ol 1 iO CD i- CO -t CO CO o 01 to CD iO CO -* <# '-D to o 00 - to 1 t- CO 1 1 CO (0 1 1 Oi rH i 7 rJ SI 1 1 CO 1 1 V 7 7 to 01 7 i 7 7 i 7 us til I-H t- to CO CO o e rH 04 Ol CO 10 rH 0) t- 01 -J - t- iO CO to u rH 00 CO I-H l'-. iO -f r-l * O 2 S X 1 1 I 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 rH p b- CO tO i i i iO tp 1 1 CO 1 03 - 00 1 Oi 71 CO r-. 1 1 p p 04 # 1 1 00 CO 1 1 1 "l .0 CO 7 i 04 yf rH 00 7 i CO rH 01 1 1 1 01 cV. 1 1 1 CO 01 rH 9 7 1 tO CO U3 00 *< 01 O4 US co us r-. CD. * I-H O4 CO Ol H + ? CO ^ Ol US CO to iQ to co 00 us O t- rH CO y ~ 7- iO 0) to Ol iO CO fr- O CO CO o **' - 01 CO O + + W 01 00 * 10 01 <* CO CO 04 iO us r- CO 04 ^f CO CO iO O CO TP Ol O) 00 t- CO p CO 01 4- + C~ 9 7* Ol 00 CO O iO >o -* ' O) CO O) O CO CD ^ t- CO i-H OJ rH t. ^, cj 1 rH + 1 rH rH w 3 CO rH 00 i > rl 1 S S E I S tt) 6* o I 5 O" 1 > HH >2 7 7 i i i i i i i i i i i i i i i i i i 7 i i 7 " i 7 i i i i i i i i i "7771 i i i 7 7 7 i -r . CJ -- ;, tf> V t-< "0 > - '^ -j , . I - . - / 10 ' OO i." CO ' ?l i i 7 7 i i i i i i i i 7 i 7 i i i i i i i i i 777i i i i i 7 7 i i 7i7iiiiii7ii i i i i i i + i i i i i + + i i+ + +i i i i i + 7-T- + + + + + + + + + " + + + + + + + + + + + u-s :. ' : ' o : : :: *p / T. cp ~. i :. :. -* i?} as ^ a~- '> ?l ?4 PH i-t i-i w cc 'Sl^S* 5003 ' 3 22rH^ I ++ I + + + + + + + + Hi= :^ ;!!lil 8 8 1 e 1 3 O" o Q 1 i(5 00 1 1 1 1 CO CO 9 + 94 DIURNAL INEQUALITIES. i*|id S^ 53 $ I MS p 9 CO 5 CO h- CO CO y 5-J OO CO i 00 i 00 00 O "* oo ire M O4 S CO t^ 11 d 00 ^ 00 iO rH ,0 CO 00 to t- OS OS 04 04 b- 10 0> i R 8 " ?, Sfl OS O4 CM % o s * CO & % S S s s at p rH 00 *> o> 1- rH H CO o 04 n ** * tO 9 CO 04 1 e 7 1 1 * s 1 1 7 7 1 1 C-l rH 7 i 00 t- 1 1 7 i eo os M 1 1 d t- 7 q 1 CO O 00 to \ 1 o r- 00 O 1 1 r* 9 O CM 1 I p 7 to 1 O4 rH 7 7 9 *- 7 i CO CM OS O i 7 CD CO OS t- 1 1 o 00 7 04 1 oo to 1 i rH CO 00 00 1 t t~ OS 7 7 00 o 1 s 1 t CD OS 1 t b- O 1 1 CO 00 tO CD 1 1 * 4 to * OS rH iO t- t- 00 OS iTj os * * r- CO <# oi VD 1 1 CO iO 1 1 1 t 1 I N o O T- CD Ir* 1 1 CO iO t 1 '~r CO i 7 -f c* CO O rH t- iO t- * 04 OS rH 00 CM -? o to o 00 1 o + CM rH 1 1 o * 1 1 CO O4 1 1 CM 2 1 + 04 04 1 1 CM 1 o * 1 1 _ # to -* rH OS to OS 00 rH 00 OS OS 00 CO OS US - 7 " tO rH rH rH o eo OJ 00 iO * Ol rH O4 CM CM O + 1 5 -XI f OS t- tO CO f + OS t 00 rH 00 t- as R s s O t- OS - op OS. O> t- O) OS CO V iO 04 CO GO t* iO tD CO 00 CM S3 CO S3 04 co O ^< CO CO ' CM CM + rH CO CO CO 2 S R ro 00 10 00 sc- CO CM CO ^ O rH b- CO CO oo p p CD o tD CM O4 Os CD tp + + + + + + + + + + + + + + O -* to CM -<4l e> - CO 0> CM OS p CO rH CO CD 04* OS f- CO 0* 2 CD OS eS O CO 00 CO I- S 8- O CO OS O4 CO O CM to ire as CM 00 t- t- rH os w o p ^f CO 9 7< CO CO O OS iN n 1 CD 00 t- ri rH CO tO CO rH CO CM tp * CO O tO o o tO to o t- t- t- CO 04 f. HI 7 rH tO 00 tO CO rH 1 CM a fr- to t- rH * * OS t- o 1 n 1 O4 Ov rH 1 1 CO OS 9 T* 1 + 1 O CO 1 00 R 1 CO rH CO r-t I + CO rH CO 04 tO -S" CM '- + 1 * tp O rH 00 1 < 1 OO CO 1 1 CO CO CO 04 t 1 CO CO ** 00 1 1 CO 1 fen CO 7 '.- "f -V 1 1 p co ^t O + 1 U5 OS * CO 1 i CO rH i-H CO 1 t iO 9* iO 00 rH iO OS 0) ^ to f- eo co rH tO CO O 8 CO f 1 <* CO 1 1 tO * t- CO i 7 CO CM 1 i CO 00 I 1 CM 10 1 1 O> CD 1 1 CO fr- 1 1 OS a rH CO OS tO 04 f _ H CO 00 * CO 1 a to 1 1 tO CO 1 1 t- as 1 1 CO 1 S3 i as * 1 1 1 1 00 T oo n 0> to OS p O ^ CO tO 04 04 CO CO CM CO ri 50 9 t~ -ni * 2 O> gj re i - ! OO tD 1 1 CO rt 1 1 i 7 CO 7 7 7 7 CO 00 1 f 7 i rH 04 7 7 h t. IH S3 I . I fr P ? g 1 CM t- CM lO = &$&i if : B 5 g i S s 2 g l | s 1814 V I i ~ i i g s 1 1 n n io d to co eo cs CM u) COONMt-t-Nt-^t-rtr-l w -OrHi-(-.oe*) .-1 Oi ^H CO 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 !OC1^0>.-.00OOOt-00 to i -^ ic - ONOWCT. OOOOOOCftONO i 7 7 7 i i i i !_!__!_! CR CO t- 1 1 1 1 CUt-INt-OOONt-COI-OSOO fc- t- M + 1 1 1 1 1 1 1 1 1 1 1 l t- 00 CO 1 1 i 1 ooaiai-aoiiittt-iQt-t- O 00 '> CO OS + iiiiiii7ii + 111 + GO 2 S S S S S i I I I S^ p CO CO 1 1 + ooaoa>x>o^>n O CN O r- I oo^^rtra-*^NOoccN Ki eo CM 1 oi-HNoootooo-woo^aito * n n 4 <0 f?57+7++++5? + + " ? 00 CO T W 00 -^ - rf C>: ^. --0 CM 05 to 0! A SSSSS^^^^SSs? s s s s CO 0> * i-HODcc>w-'eNeoo-i | oa>cji QO CO IQ -; ; i : .-, 10 so eo co --, .- r. 1 fli A n v* pv M M P* m H n 2 S S 8 ri \\\ 11 11 11 III CO CXI C* ^* t- tH ^H rH P> ' - I ' -. -- " ^1 SO CO t- p W iO S 2 S S ci ^occgiojj^i,;.^^;. c* at e* + + 7- 7- t-rcooc^raaon"i~racoia )< CO W rH rot-j-low^^raooooiM t- .0 to K50iniOinO5tDOOlO B 00 tO O S N-.ora*-*-.o./5co rt + + + 1 =,- 1 1 + + + 1 + + + 1 1 .p o eo * Ct O rt t-l i + . 7 ^ SO 1^ i i i 7 I oycpyMj.cc.oNNO-, p t- o, o _ N tH CO 1 1 1 1 1 1 1 1 1 1 I 1 7 i i i id GO CO -r -; 1- ~ TJH ' TI .- .-H 80= ::--( r. co -r ' r- co iH i ^H P-) ^H CQ 1 1 1 1 1 1 1 1 1 1 1 1 Cft 00 -.0 -1 1 1 1 1 SSS;!S;IiSiSS S * S SI 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 rtoocoocooasa>eo Oi r-H eo :777i7i i i77i CO C i-H CO 1 1 1 1 * llll^llttlll S S 8 1 1 I 1 -' 7777 i 7 i i7777 CO lO 7 i 7 7 lllilllllill i I i i 1 1 1 1 f* I-H '5 > I "3 1 I 9G DIUENAL INEQUALITIES. SJ.|j S' g" i : d : g US 8 S3 s s ass S? * lA t- t- c? c^I 2 ^ S :p w M s s r-t ** > CO S. jl | I CO CO OS Si S ^ s a 5 S s s S S S 238 3 3 C O SO CO 0) OS * :o 10 0> s s CO CO =5 3 o oa CP M 2 S + + + + + + + + + + + + + + + + + + + + + + a A c-1 o S SI + + + + + + + + + + 1 + + + + + + + + + + + 00 oa eo D t- 01 rH t- IA IA O CR CO A 00 00 --C t- Ol CO + + + 1 + + + + + + + + + + + t-^ CO + N n 7 + >O IA + + O t- 7- + + ^ l^ 7- + CO tA 1 + lA CC >A + 7 + !O O lA + 7 + CO 7 + CO 7 7 3 d * + 2 o 7 + ^H !D + + 00 >0 CO + + + 00 * + + CO 1 * r* N 1 + + 00 *0 * + + + CO O + 7 o w 7 + u N 1 IA IA + + CO ^ 1 + t- C<1 Cq + 1 1 lA CO + + CO O> 7 i CO W A 7 + 1 CO t~ * + + + CO CO 1 1 CO + 7 * to 1 OS & CO 1 1 N d 7 t- * 1 + * IH 1 1 00 CD 1 7 00 2 7 7 a s i i 8 a S i l l CA C^ =O i 7 7 a 2 1 1 S3 S 1 1 - 3 1 s 1 1 Oi t- 1 1 f -l CO 1 1 1 S3 S 1 1 5 8 i i S S 3 i l l 00 r* 00 i 7 i s 3 1 1 3 S 1 1 cq 01 7 8 i i a i - i i US CO 777 r, n i i S S i l 8O> -l ^ 1 1 1 00 CO O i 7 i 2 2 7 7 s 1 1 ~ o 7 t- 00 7 i at oo 1 1 i-H O CO 7 i 7 S S 1 1 S 2 i i g S 1 1 1 * 2 S l l l - 71 7 7 S i i o i s s i i ^ t- 1 1 e ^< co 777 a s i i S * i i co cq w 7 1 1 o -* -i i 7 7 01 CO 1 1 "= S 1 1 of 7 i t~ 1 1 ^ 7 7 i s s 1 1 t - 7 + CO CO f-l i 7 i CD JJ 1 1 1 ^ * [ + 2 it l i 00 *- CO + 1 + lA O ic 1 1 w cc 1 1 O> N CO 1 1 + CO CO 1 1 O CO 7 + N t- + 1 + t- O + 7 S + + N (- 02 + + t- r-t + + N rH Oi + + + os ec + 7 S3 S + + CO 01 lA 77 + O CO A + + 7 <0 1A 7 7 2 S + + S h b s b s. 1 T: >? a < a i t -3 ^ ^ 4J M -g. 5 i r I >! w O NovemlK Decembe 9 Z. | >-9 fe III g i if \ 3 3 _= ^s f> < Septemb October 5 .S ! I > O - K n g "5 a fe DIUENAL INEQUALITIES. 97 las Irom the mean. ljf3i!!!>i 8 1 5 i ! s 3 = s = = 3 s 3 a = s 00 (O CO a + 7- + + + + + + + + + + St- * 00 r-l fH o.oi-^coo>-< r -,a>!00. W 00 * "* + + + + + + + + + + + + + + + + : ? ? ? ; ; 5 ; 5 ? ? ? ]f + 7 7 o>.notf>corto>t-w I-H O O 1 + + + + + + + + + + + + + + + 00 T 5 ; ? ; :? ? ? 3 * ? O> eo ^ I>I i +++++++++++ + + + + i 3 rf i ++++++I++ eo o w eo d 7 i i + + + i i + ^H 01 eo eo i + i i .- iii ii + i i i i i i i i CO 7 ? 7 i i i i i7777 n . 10 . A jo 7 i 7 7 Ol i i i i i i i i i i i i 2 * 3 8 1 1 1 1 - i i i i i i i i i i i i 8 * 53 S3 I 1 I 1 cj goo^.gcooo.o.o.-gg 8 * 1 1 1 1 1 1 1 1 1 1 1 1 lit) fj CO r-< W l-< ' 9 P| M 3 M 1 1 1 1 1 1 1 1 1 1 1 1 t-Oll-rH rH rH CO 1 1 1 1 i i i i i i 7 7 7 i 7 7 co oo w d 7 i 7 i oi i-H . <-1 1H ^ W *H + + 1 1 1 1 1 1 1 II D 00 O fr- 1 1 1 1 00 ?;; 7 7 7 7 7 7 7 7 7 O to * r- 111 + | - + + +I 1 1 1 1 1 + 1 + * rH CO 1 1 + i i id + + + 1 1 1 1 1 1 1 o eo o us 1 + - + + +I l + l I++I + eo r-( eo o + i + + 4 5 5 ? : : ; : ; ? 5 oo I-H m co c, ? 5 2 + :;:;::;; o eo oo o ci ?? t :.:;::? 5 ?? rH * rH S ' i". to :: -z. i- T r- ^ ri r- * St- * o> rH rH t - l^rtS^Sj-^2 > | S 8 B s .9 cr 1 CO 8 d - ra 98 DIURNAL INEQUALITIES. 63 I CQ *,^ D S a ss 1 S tNOtDCOaOWtO^OCOtOCO ^l-eSC&C&cpoC'OpOiO**- "cO^COoOiooooo.-Hf-ii^.-l O> Oi O f I-H .p o rt CO Oi C* O 1 ^OOt-tDr-iCOCl^Dt-O^in ^cif-D i- co >* o CH m * o -# S S S 5 H r-( rt ^ M e Scoocomooaokaoi-iiraco ^ _ y- <> f CO 1.7 ', ? ? P a + + + + + + + + + + + + + + + . + t-c9tf3**|-^-fl^^Tpio5"i ^ t3 g g " oooooooooooo + + + + + + + + + + + + 0000 + + + + b--Oi-<*Ha4t Cvjt-OCI CO'V'-^^'JtCOWTjtiOtDiTl OOOOOOOOOOOO !+ + + + + + + + + + + t- O t- i-l CO ^ -^ N o o o o + + + + Sco 05 cv ?i . i :t :: T- i - oc^eocQcoeoP5i* i - i .-; M 7. r - OOOOr-lOoOOWPii-lO s s a S3 OOOOOOOOOOOO 1 + + + + + + + + + + 1 0000 + + + 1 . ?3S2SSSg2S 3 2 S S 1 t oooooooooooo 1 1 1 t 1 1 1 1 + 1 1 1 o o o o 1 1 1 1 I ci i- r; < o o <> z> M co ^tt35rt^rt(Nrfo-^ioo6 s s s a 5 rHOOOOOOOOOOO 1 1 1 1 1 1 1 1 1 1 1 1 0000 1 1 1 1 B^*oqDC5oeot--Dt-ai ot-ift5rtocoe3>eort S S 5 S ^^OOOOOOOOOO<-H 1 - 1 1 1 1 1 1 1 1 1 1 1 0000 1 1 1 1 gt-3,--li-H 1 1 1 1 1 t 1 1 1 1 1 1 O O O ft 1 1 1 1 n OO-^OOCOv-IArH^IOCOiO ^cSi^w^-t-cpowSccip TT t- >o * # Oi r oo ^ 1 1 1 1 1 I 1 1 1 t 1 1 1 1 1 I ^-'*tioo-}'-'-i-a)oooo ^iJ"000*t-t-( lOOOCO U t- -t iH * o 35 us rHtHOOOOOOOO'Hi-4 ! 1 t 1 1 1 1 I 1 1 1 1 i-l O O i-t 1 1 1 1 j '. -" M c-t Ci i - : t i - -.: ' - -; ^rH^HO>O>^Ot-U3lpt-QOC^lS f-^HOOOOOOOO-^^- 1 1 1 1 1 1 f 1 1 1 1 1 StO *O * o oo rt 0000 1 1 1 1 c$ ^Ji^o^JiOiOCl' (t ^H ^^ ** b- % ( A [ { 7 p 7* ^* l" %* T* 7 1 oooooooooooo 1 1 1 1 1 1 1 1 1 1 1 1 CO O> O ift ? T* 1 V" ?* o o o o [III *cq^-fccoi-raoco^i i *co ^^Ip3|||^pi|fl^ldfl(- 5 53 S 51 """* oooooooooooo + 1 1 1 1 1 1 ! I I 1 1 o o o o 1 1 1 1 O 1 *OOOOQiOOt~ l ^ l ^^1' 1 * 9nfif9fi)C;4|t|5|i|K9 CO 00 W CO i^ ^-l CO O oooooooooooo + 1 1 1 1 1 1 1 1 1 1 1 O 1 1 ! 4- Sojgjwwgswoii-iiooej rHOrHOOOOi-tOt-lrt s s s + +I1I1IIII+ + o o o o + 11 + Scoooat-Htaoit^i-iiAaico CO C3 5O 00 oooooooooooo + + + 1 + + + + + + + + o o o o + + + + S-D-r"iJ*~ '-i~i-C^C*l c5-*CC4(Nr-lC4CO'-Of-l ci to w M cc oo a OOOOOOOOOOO--I + + + + + + + + + + + + 0000 + + + + 1 2^?:?5S^?8^^^^5; SD # N CO * .-HOOOOOOOOOOi-" + + + + + + + + + + + + O O O o + + + + 1- ri I 1 /. -* o ir> - ^ n -r- eo wsoitoirtikOf^co^^tcr- 12 S S S ^tOOOOOOOOOO^H + + + + + + + + + + + + t-l + + + + ** i': ^* -r : i i - -. o > ^ <* O> COOiOCO? (OU30OU3 -'OOOOOOOOOO'-' + + + + + + + + + + + + CO CO 00 (O t* 9 04 O O 4-1 + + + + mi-(iOinookOoooQo^)aoc>i *joieoiO^ieccoooq'-< t- t- CO t- TJI m 1-1 rHOOOOOOOOOO-H O O O o + + + + t--*coeoairagooi5po eot^b>-Ticctfic>3tOAA 8 3 S S -IOOOOOOOOOOO + + + + + + + + + + + + o + + + + t^.co-qit-iftMOi-HCoco'^'Oi ^e->tf>p-*f8c5ipfi5ioaiy> oooooooooooo + + + + + + + + + + + + i-l O O O so -^ to oo o o o o + + + + e t : : : : i 1 s:ii Ililllllllll 1 , I 'S 3 S 1 a s, a S H S 4- 9 fk D p s a c7

a a CO* O> s ! CO CO XO 00 a CO b i - OS 7* a 1 CO* Cl o Cjk O CO c; j r \ P*-"" \ o Nov6irii)c c . / \ \ / +36 / \ \ \ t ***" -IO' -20' / \ / +20 / r *- J \ -30' +10' ,/" / \ tt* 1 / \ \ 7 +30' / -20' y \ / -^^ t -^^ > \ J s +2O / ^x, \ \ ^ ftS** _ -30' + 10' / \ -40' / \ / \ ^ \ / / +40' . -^, \ r -IO' +30 / \ \ s, / -20' +20' y ^ \ ^-*. / -30' ) '^^ / \ +10' / . o / ,_ J \ o February / \ s^ -10' +30' \ / / / -20' +20' y I x" v \ \ ""^^ _ ^ / / -30' +10' / \ ^ / / \ - h ^ / 1 d, &> \ \ / t /- -10' ^ ~^ / 20' iMIDT 2 4 6 8 IO NOON 2 4 6 8 10 i: Fig. 1. Decimation. DIURNAL INEQUALITIES. 105 t-ZCf +20' +10' All days \ \ O 2 4 6 8 10 NOON 246 Fig. 2. Declination. P o April o May o June o July August September 1O 12"ID'T 1 00 DIUENAL INEQUALITIES. +10' +10' +IO' Qi eter \ \ \ \ o October o November -20' o December o January -10' -20' o February 20' o March o April o May June Juty O August 10' o September 2 4 6 8 10 NOON 2 4 6 8 10 I2MIDT Fig. 3. Declination. DIUKXAL INEQUALITIES. 107 +30 /\ +30 \ +30 +2O no \ \/ \ +20 \ \/ 1 / \ \ O Z \ \ \ \ \ \ \ X \ \ \ \ \ o October o November o December o January o February o March 6 & IO NOON 2 4 6 8 10 I2MID'T 4. Horizontal Force. (Unit ly.) P 2 108 DIUKNAL INEQUALITIES. +20 +2O +2O 6 8 IO NOON 2 o April o May o June o July o August September 8 10 12 MID'T Fig. 5. Horizontal Force. (Unit ly.) DIURNAL INEQUALITIES. 109 no o + 2O + 4O *2O + 2O \ \ \ \ \ \ \ I -20 -20 -30 o February 30 o M&rch 4 6 8 10 NOON 2 4 6 & TO I2MID'T Fig. 6. Vertical Force. (Unit ly.) 110 DIURNAL INEQUALITIES. tic- \ 02^ 6 8 1O NOOK 2468 Fig. 7. Vertical Force. (Unit ly.) o April -20 -30 o Mav o June July o August o September -20 IZMID'T DIURNAL INEQUALITIES. in +o'-5 + 0-5 +0-5 o to'5 O +0-5 o +0-5' \ \ \ \ \ \ \ \ \ \ \ \ \ i \ \ \ 1 V o October o November -o December o January o February o March o April o May O June o July o August o September * 4 6 8 IO NOON 2 4 6 8 10 IZMID'T Fig. 8. Inclination. 112 DIURNAL INEQUALITIES. + IO *- r - ^ *~-^ -^ - * -^, ^^, ^ --^ ^^ . < ^~ -10 no x ^ ^ ^ y ^*~ ^ / \ ^ o Equinox . / \ \ / 2O / \ x ^ ^ ^ X -IO' + 10 / / \ -20' / \ +IO X ^ X X x x x* ^~ ~\ \ \ \ \ \ \ X x -10' \ ^~- ^ ^ - Qt let ,er d qn \ x. ^^, * i , -^ X x x -10' - -zo' tio' o ^~ x ^ ^> , ^ ^N, N x, - o Midwinter ^ ^^ ' -^, s \ \ x x S ^ \ x -10' + 10' / X N ^ ~. \*^ ' - /> x \ -20 " ^ s ^ +30' + 20' / / / \ \ \ V. x / ' - -IO' -zo' + 10' X \ - / \ - X \ +20' + 10' x X" ' X ^ ^s \ \ \ \ x , ' ~ -10' -zo' X x \ --. - 30' , I2MID'T Fig. 10. Horizontal Force. (Unit ly.) 114 DIURNAL INEQUALITIES. \ \ \ \ \ \ N L_L o Midwinter o Ecjumox o Midsummer o Year 2O 2 4 6 6 IO NOON 2 4 6 8 K> I2MiD'T Fig. 11. Vertical Force. (Unit ly.) I JIURN AL INEQUALITIES. 115 +o'-6 +0-5 \ \ \ \ \ \ \ \ \ \ \ \ \ \ o Midwinter -05 o Ec[umox o Midsummer o-5 -1-5 o Year -0-5 0216 io NOON 24 6 8 10 IZMID'T Fig. 1:?. Inclination. Q -2 116 DIURNAL INEQUALITIES. Midwinter Mid aunt m.er> Fig. 13. Vector Diagrams DIURNAL INEQUALITIES. FOURIER COEFFICIENTS. 117 CHAPTER IV. DIURNAL INEQUALITIES. FOURIER COEFFICIENTS. 31. From the diurnal inequalities, calculations were made of the Fourier coefficients answering to the " waves " whose periods are 24, 12, 8 and 6 hours. The analysis of the diurnal inequality may be supposed to proceed according to either of the two equivalent series cos It + Jo sin It + ... fi sin (t + ai) + c., sin (21 + a 2 ) + . . . Here t is time, counted from local midnight, one hour being taken as equivalent to 15. The constants with suffix 1 refer to the 24-hour term, those with suffix 2 to the 1 2-hour term, and so on. The a and b constants are calculated directly from the inequality tables. The mean ai, for instance, for a particular season of the year is the arithmetic mean of the values of 0,1 for the months composing that season. The c (amplitude) and a (phase angle) constants are calculated from the corresponding a and b constants by means of the formulae a = tan" 1 (aft), c = a/sin a. = i/cos a. The c (or a) derivable from a seasonal diurnal inequality is not, as a rule, the arithmetic mean of the c's (or a's) of the individual months which form the season. 32. Tables XXIV to XXXII are devoted to the a and b coefficients. These were in all cases really calculated to at least one figure further than is retained. It is, however, hardly necessary to remark that even as thus curtailed they cannot be regarded as physical facts freed from observational uncertainties. This reservation ought especially to be borne in mind in the case of the coefficients with suffixes 3 and 4, which relate to the 8-hour and 6-hour waves. The differences between the values obtained for successive months probably owe at least as much to the existence of " accidental " disturbances as to any real difference between the magnetic conditions characteristic of successive months of an average year. In the case of the Declination, Horizontal Force and Vertical Force the values of a and b are recorded for the individual months of the two years, as well as for the months of a representative year in which common months of 1902 and 1903 are combined. Also two sets of values are given for Declination. Of these the first set, comprising Tables XXIV and XXV, relate to the all-days' inequality data of Tables XII and XIII ; while the second set, comprising Tables XXVI and XXVII, relate to the quieter-days' inequality data of Tables XIV and XV. For Inclination only one table is given. When the sign to be attributed to the numerical value of a constant is the same for each month and season it is indicated only at the top and bottom of the column. TABLE XXIV. Declination (All Days). Fourier Coefficients. (Unit 1'.) (7,. |. . a,. t t . 1902 April 3 -32 + 8-90 + -84 3 '96 + -19 + 1 -95 '30 '81 1 '78 5 '46 + '85 '30 + '03 + -06 + '50 '09 1 -35 3 '09 + '84 0-49 -f '28 -06 -18 4-0 '25 1 -61 4 '97 + 1 '84 "18 -t-0 '58 '34 + '12 '09 1 -21 5 '43 + '32 '78 4 '41 -fO'74 + '30 '20 1 '58 12 -49 -fO'58 2 -72 + '38 0'33 0'26 + '04 2 -62 13 '77 + 3 '57 1 '04 + '07 '05 + '01 + "47 2 "90 19 -01 + 1 '94 8'45 f 3 '17 + 1 '71 '91 + 1 '90 December 6-42 28 -10 1 -87 7 '29 + 2 '30 + 'It 2 '01 + '48 1903 14 -38 24 -<)8 + 2 '50 6 '88 + 2 -17 + 1 "18 I '36 3 '29 10 '58 21 -81 2 '10 '06 + 0" 50 1 '33 1 '05 March 4 "02 15 '37 + 1 '22 4 -57 + 1 '25 1 '00 '13 + '06 April . 4 -25 11 '51 -f 2 '95 4 "22 + '68 + '61 '71 1 '38 3 '31 -0-38 tO-25 -0-28 1 -86 7 -23 + 0-84 -0-24 + 0-58 -1-0-24 + 0-44 -0-17 4-20 14-50 + 1 -15 -0-79 -0-52 + 0-51 + 0-07 -1 -62 2 -62 1 3 -77 + 3 -57 - 1 04 + 0-07 -0-05 + -01 + 0-47 2 -90 19-01 + 1 -94 -8-45 + 3-17 + 1-71 -0-91 + 1-90 6-42 28-10 -1-87 -7 '29 + 2-30 + 0-11 -2-01 + 0-48 Year 4-80 14-44 + 1 -08 -3-02 + 0-90 + 0-33 -0-44 -0 36 Midwinter 2-29 3-66 6-12 13-4G + 1 -19 + 1 -96 -0-26 -2-62 + 0-45 + 0-31 -0-17 + 0-19 + -17 0-14 -o-oi -0'55 - 7-90 + 24 -03 + 0-86 -7-54 + 2 -55 + 1-00 -1 '43 -0-31 TABLE XXVIII. Horizontal Force. Fourier Coefficients. (Unit ly.) "l- |. a,. b,. "3- *3- 4- 4. 1902 April 12-20 7-16 + 1 -55 + 4 -13 1 '08 + -84 o -10 + '80 May 10-33 10-34 -0'30 3 -38 + 1 -35 + 0-42 0-65 + -97 8'99 5 -83 0-21 3 -53 + 0'38 P OL + -46 + -87 July . 6'84 4 '78 + -14 3 '04 + 0-02 0-37 '29 + 0">4 August 9-23 5 '75 + 0'41 2 -16 + -28 + -59 + 0-43 -12 12-85 6 -26 + 1 -81 1 '98 0-17 + '94 + '05 + -30 October 12 "55 7 -42 -t-0'66 3 '37 + -06 + -67 + "60 '53 November December 17-08 16-78 14-92 23-67 + 1-07 + 6-22 G-88 3 "59 -0 04 + -"4 -2-23 + -78 + 0-29 0-67 -1-46 + 1 '55 . 1903 January 10-61 27-48 -1 '75 1 "76 + 1 -62 -2 -f>6 + 2 -66 + 1 -48 13-24 12 -83 + -3 l ) 2 -91 + 1 -10 0-77 + -81 + 1 '25 March 12-58 10-22 -f 1 -19 1 -81 0'16 + 1 -32 + -08 '43 April 14-30 7-44 + 0'35 -19 + 1 -40 + -13 + 0-67 + 1 -20 Mav 10'03 5 -64 + '77 1 '89 + 0-85 f '93 + -57 + '27 ' > June .... 13-59 9 -13 + 1 -20 1 -59 0'68 + -63 -14 0-71 July 9 '53 8 -27 + '11 3 '11 + 1 -79 + 1 '22 + '13 + -34 8 '55 5 -18 1 '38 2 -10 0-98 + '55 1 '14 -01 September 11-06 5'64 + 0-46 + 4 -9G + 1 -81 -67 o -92 74 120 DIUENAL INEQUALITIES. FOURIEK COEFFICIENTS. TABLE XXIX. Horizontal Force. Fourier Coefficients. (Unit ly.) ,. *,- a?, is. 3- * 4- i 4 . 10-61 -27 -4S I -75 + 1 -76 + 1 '62 2 '66 + 2 '66 + 1 '48 February 13-24 12-83 + 0-39 2 -91 + 1 -10 0'77 + '81 + 1 '25 March . .... 12-58 10-22 + 1 -19 1 -81 -16 + 1 -32 + -08 '43 April 13-25 7-30 + '95 2'lfi + 0-1(3 + 0-48 + 0-28 + 1-00 May . 10-18 7-99 + 0-23 2-63 + 1 -10 + '68 '04 + 0'62 ^ . June 11 -29 7-48 + 0-49 2 -56 0-15 + -31 + '16 + -08 July 8-18 6-53 + -13 3 -07 + '90 + '43 -08 + -29 August September .... 8-89 11 -96 5-43 5 -95 -0-49 + 1 '14 2-13 3 "47 -0-35 + '81 + -57 + '14 -0-35 '4t -0-06 '22 12-55 7 -42 + -66 3 -37 + '06 + -67 + '60 '53 17-08 ]4-92 + 1 "07 6 -88 '04 2 -23 + 0'20 1 '46 December .... 16-78 23-67 H-G-22 3-59 + 0-74 + 078 -0-67 + 1-55 Ye.ir . . . 12 -21 11 -42 + -84 3 -03 + -49 '03 + '25 + '30 Midwinter .... I'Vuin Midsummer .... 9-88 12-58 -14-82 7-33 772 -22-02 + 0-29 + 0-98 + 1-85 2-75 2-71 + 4-08 + 0-62 + 0-22 + 0-77 + 0-47 + 0-65 -1-37 + 0-01 + -13 + 0-76 + 0-33 -0-05 + 0-52 TABLE XXX. Vertical Force. Fourier Coefficients. (Unit ly.) 0,. ,. 2 4, Of h a 4 . *4- 1902 + 13 -18 + -85 4-25 5 -19 + 2 '30 '30 1 '12 '76 April 16-78 8 '65 1 -38 '00 + 3 -66 + '13 '40 '90 5-62 4-00 -25 + 2 '74 + '44 0-52 '38 + '65 June 8 -05 2-20 + '05 0-92 + 0'09 '26 1 -61 '04 Julv . 5 -97 4 -05 -69 -16 + -fil + "23 + '07 + '55 August . 7 -00 8 '70 0'85 + 1 -72 + '27 '41 1 '07 o -42 September 11 -79 1 -96 2 -46 + 1 -22 -68 + 0-07 -40 + 1 '36 October 12-44 + 0-99 3 -22 2 -13 + 1 -15 + 0-29 + '27 + 0-56 November 22 '09 5 -47 2 -15 + 2 '84 '37 + 1 '72 '22 "45 December .... 19-63 + 3 '20 3 -49 1 '58 + 2 '73 + 2 -62 1 '98 1 '40 1903 January 20-11 + 18 -89 11 '98 3 '11 + 7 '34 + 1 "78 3 "35 + 3 '29 February 17 -60 + 8 -02 3 -34 6 '23 -11 + 5 '65 1 -13 '06 March ... 23 -00 + 7 -71 4 '41 4 '63 + 2 '30 + '29 '97 + 1 '70 April 14-93 4-62 1 -86 "55 '76 + 1 '13 1 "74 -09 8-38 + 1 -20 "45 4 "18 + 1 '19 + 1 "29 + 4 63 + -14 T J June 8-80 2 -12 1 '96 + "38 1 '05 + 1 '65 + 1 '07 + 0' 59 July . 12-68 7 -59 2 -15 1 '83 1 '73 '42 + "39 1-31 16 -08 3 '78 1 -18 1 -7O . n .RQ + 2 "30 l -fift + *15 September ... 16-41 + '49 1 '52 3 '84 + 1 "22 + 1 '78 + "15 O -74. October 15 -28 + 1'25 + 1 -33 5 '52 + 2 '21 + 2 "00 2 -47 '13 November 21 '95 + 2-81 8 '70 8 '02 + 4 "07 + 2 '99 '65 + 1 "71 + 50 "72 + 9 '87 11 '40 + 8 '38 + 1 '25 '69 2 -78 DIURNAL INEQUALITIES. FOUEIER COEFFICIENTS. 121 TABLE XXXI. Vertical Force. Fourier Coefficients. (Unit ly.) i- *i- 117 -0 7-7 9'3 7 53-5 3 '6 88-5 87-5 86-3 108-6 112-4 110-6 0-48 0-44 0-45 0-25 0-20 0-21 0-51 0-46 0-47 Sum of 24 differences .... Vertical Force Ranges 19-8-y 128-7 8-1 7'3 7 32-9 1-7 86-9 86-9 88-7 48-6 48-1 56-0 0-50 0-50 0-49 0-18 0-13 o-io 0-37 0-26 0-21 Sum of 24 differences Ci Inclination Kanges / 0-92 6-03 0-40 / 0-41 2-41 0-17 87-6 86-4 86-0 106-2 109-1 108-0 0-48 0-44 0-46 0-22 0-18 0-19 0-45 0-40 0-42 Sum of 24 differences ... . Annual means from D, H, V and I above .... at Kew .... 89-6 272 75-0 263 0-46 0-54 0-16 0-12 0-34 0-22 37. Table XLV shows approximately the dates when the 12-month and 6-month waves attain a maximum. The 12-month wave has of course only one maximum in the course of a year, separated by six months from the minimum. The 6-month wave has two maxima, 6 months apart, with minima equidistant between them. The date recorded for this wave in Table XLV refers to that one of the two maxima which falls nearest to January 1. TABLE XLV. Annual Variation. Dates of Maximum. Date of occurrence of maximum. Annual term. Semi-annual term. Declination (all days) Ranges December 23 26 27 January 21 February 5 3 Sum of 24 differences ... Declination (quieter days) January 4 ' 1 2 January 1 1 ' 21 20 Sum of 24 differences Horizontal Force Ranges January 2 ,, ' 3 4 December 22 20 21 Sum of 24 differences Vertical Force Ranges January 4 4 2 January 22 ,. ' 22 18 Sum of 24 differences Ci . Inclination January 3 , ' 4 5 December 23 22 22 Sum of 24 differences ABSOLUTE DAILY RANGES. 131 CHAPTER VI. ABSOLUTE DAILY RANGES. DAILY MAXIMA AND MINIMA. 38. By the absolute range of an element is meant the excess of the absolutely largest over the absolutely smallest value met with during the 24 hours. The term absolute is added to indicate that the quantity considered is not the range of the regular diurnal inequality, nor the range derived from mere hourly readings. The absolute ranges are at once derivable from the daily maxima and minima given in the tables of tabulated values, but for convenience of reference they have been collected and presented in Tables XLVI, XL VII, and XL VIII. It is of course impossible ever to say with certainty, in the case of an absolutely isolated station like Winter Quarters, what may have been taking place during the time of changing papers. There is always the possibility that, in the course of a few minutes during which no trace was being recorded, an element may suddenly have changed and then reverted to near its primitive value. But even in the Antarctic, though large sudden changes were not of very rare occurrence, it was very unusual for them to occur singly, and one could usually feel fairly confident that neither the daily maximum nor the daily minimum had occurred whilst there was no paper on the drum. When, however, the interval between successive sheets was considerable, as occasionally happened, or when the trace was highly disturbed about the time of changing, there might be considerable doubt as to whether a maximum or minimum might not have occurred in Declination or Horizontal Force. In Tables XLVI and XLVII, relating respectively to D and H, ranges are usually given whether the record for the day was complete or not. There are, however, omissions on a few days when the traces were confused or indistinctly visible. Figures inside [ ] brackets relate to days when the record was incomplete, but when the general appearance of the curve seemed to warrant the belief that both the maximum and the minimum for the day were actually recorded. Figures inside ( ) brackets are for days of incomplete record, when appearances suggested that either the maximum or the minimum, if not both, was unrecorded. The combination > + means that the trace went beyond the limit of registration in the direction of element increasing, while > - means that the trace exceeded the limit in the direction of element diminishing. When both limits were exceeded, the combination > is employed. The sign * denotes that the trace was too confused to decipher, while denotes that no trace, or only a few hours' trace, existed. Figures preceded by > are certainly, and those in ( ) brackets probably, under-estimates of the true range. In the case of D, and still more in that of H, the number of days when the record was incomplete was so considerable, and the cause was so frequently due to the limits of registration being exceeded, especially in Summer, that a very imperfect idea of the average amplitude would have been derived if days of incomplete record had been omitted. This consideration did not, however, apply to V, as loss of record of this element very seldom arose merely from the daily amplitude being large. The almost invariable cause was defective action in the magnetograph, or loss of temperature trace, and so ignorance of the temperature correction. There was thus no reason to regard the ranges derived from days of complete record as below the average size. It was thus decided to give maxima and minima and absolute ranges only for the days of complete record. These ranges appear in Table XLVIII, and means are given for the separate months. In February, October, November, and December, 1903, however, the number of days from which the means are derived are so small that the figures possess but slight significance. S 2 132 ABSOLUTE KANGES. DECLINATION. 1 ' -^ + + + Ii i i i ?* ? *? i ^~ i 10 =0 10 ^" 1 1 1 1 a 2 s 1 si - A A A i Ii i 1 1 1 1 1 1 1 1 1 1 _ A A 1 1 II 1 1 1 1 1 1 1 ? ? 1 1 1 1 II II eg I 11 1 1 1 1 1 1 1 1 1 1 1 (\ A A 1 1 O~ *< O | O iO I iO | II ^ OO ^-l 0* 1 Ui C - i aaM^"|aocOt*CM| | | ^ A A AAAAA I 1 1 1 1 1 1 1 1 1 1 1 1 1 1 U 3 qo,0 + + ,-, + 4 -s -H CM^iCO||OQOCM|| ^iiO * co SSmt-t-ratD | 1 | A A ~ 1 III s 1 1 fe 1 1 " A A A 3 r-< : ' A + -H + -H -, g n3n V *" * S^coS^S5S* * * * S >IOODiOt~iiOiOT)lOiiOt CMcOOOi-^C M il r-l ,-HCMCM-i-Hr (CMrHr-t ^ A A A 5 S 8 ora-^^o-oc.^^ocorat- 1 J 0? Anr A ~ r-< ^ T-l W T-( ^^T-*^-I" A I & * ~. *~. ~ ^ " 1 - aunf -ggSnSS;raSSwSS ?SSSSS5SSSSSgSSf '. S * A / \ I lat-^osiot-iioooit-woao II lOOOOWCfiCOC^^r^'wi . f- iuM --CO!OjjHCOOroGOg^tOlOO 1 I g. R * s e e. S S. E S & f 3 O O A 1 1 judy sSSSSSSoSSSSSSS? : S A A ^^ 1 1 i H O CO qOH "3SoSSS2S3S2|" ISgsgSSSScSSSSSSSr i iO CM A A A '^JBnjqoj wo-a,5om5'oo.o' ^6oi>6llO^l'OOo45i (OOOO7O5( 1 ,-, I^ICMCM^'' ltD-^if5Ol--COCO'3SD A A A "* - \ A ""A II II ~* *-* *-* i ^* i 1 1 O CM AJBniref "!- MS3N8m33S "SSogSroSSSrag 1 5 J CO CM A A A A "A A - A + -H -H + -, + .1 , ^TOONOONCOOlOOI^CIOOOOC >* CM n jaquisoaa ASSESS SoSSgSBS: 3SSM33'-;g3oSSc ' S S A A A A - A A " i^'e,^io rt oE'S'S5-c + + -H + + jsqraaAojj ~SSSS33Soc;S3Sf SSSSgSSSScSSSSSS? s A A - ~ AAAAA inw^u^i-ou-jooo-eoioioirac + + + + + ^ " + + ci~ "OCMOSTH-ffl^'OCMl CMr-lOSI s s A A A A A A A A A / v A A - -ElIISiSIISISI= '"' A ^ A ^ n CM CO O O CM 00 5 1 iO^ O CO * OS O C 1 + CO tO | -vmSnv -SSSKSSSSSSa-- 3 3 CM iO C" CM i-H s s *. itsistilssiifi ^ T** f^ 3 CM CM CMiO'-l'O^OOOOOkOio'Io'OiOO DOOr-((Dt- CO AH r ^ r ^ r ^^D^Ht -^liO O^ *-^ iO -^ CO CN f SOI CO os c-5 1 1 | , g, tJjdy -W*OOOT*CMr-i I !fO)--^OOC S OPH(NS J - J . l Ot-OQao 1 a I r- ^ I 134 ABSOLUTE EANGES. VERTICAL FORCE. MaquraewQ IlllllllllllgSgllllllllllllllll ST J9qui3AOJJ 1 g 1 1 1 S 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 II 1 1 1 1 1 &t to Jq 1 1 1 1 IC-1 1 l^oocoi-^eol 1 1 I^W-^coio-^TfcOTrnl ll | 1 | laJ m a- -.C|n f ooc^d-*aoooi-'cc'*~^-^D<3i | :e ao ci I-H i- r i |COIO^HI~IO 1- t- I- Tfi Oi iS (M 00 d - 00 Cl * K5 1 CO ire -fi Oi N I- 1 1 V i>4 t n M rl CH i-H ciri .r-l o aunj* CldODCOaDiOI--| 1 71 Ol OS "i* ^ 1 ICOI^CI It- 1 I^OOO^^O U50 ^2 tD |iC-^fOl IS 1 IS-VOiCOCO iO aj A-* K S^oot-c^eoMavOiOci^oi i i i i i 115 ^ -n i | i i i i ioor^ lOiocooiooi-HTC^iO 1 ^ 1 ! 1 LI I Irtco-*! i 1 1 1 1 ISio ^ ]udv '~\ ~. 'i. ~ ~: ] 1 i - 1 |O>ni 1 I 1 1 1 1 1 t CO 1 l^f! 1 II-H <# 85 > 35 cK I |M--.| lool 1 1 1 1 1 1 l-^-eoI 1 ao 1 1 13 Ci qDaBK IS l!lgS||'ggSg8l] SS 1 1 So ^ 5 5 S Xjutuqaj I 1 1 I 1 1 S 1 1 1 1 1 1 I 1 t 1 1 1 1 t $ 1 1 1 g f 1 01 A'jBnuBj- Oilll CO 1 i < j 1 -r iO CO iO O) I 1 1 "'Illllllll s s i s l| gsssas l ' IS jaqtusDSa 31gg2|S|8|S5l ISSIgl 1 1 1 1 S 1 1 1 1 1 1 g J3qiU3AOJJ SI S S 8 S S 1 1 | g 1 1 SSSSSS Illll s aaqcx)DO 1 i I i i |^^-'*^-.|oi^co<-ic>)C3| ojoiioi^i i 1 1 1 1 1 l4n^pnlc6*5'W4>!fc*Di$'*iHi<5l ^-^loaol 1 ~ < 3l t -'- ( OOO t-l-OOiOCQ 1 1 I SB fi ' 55 ^- H 1 Oi (M SS 3 iO -r lo*Oi-t--irao>iO-Vt' i iinp C^lr-iOOOlCO 1 1 M iTtieO.-! 1 ItMtO 1 1 1 iQ OS C4 tA iQ Q |tOOO*W -*w < I ^ ?J ^H oi 1(5 ?1 1 ICPNO1 1 1(5 00 i-4 ICSCO^^HOOTfltD^fO; to OiD I I 5 , '-O^HiMeO'l'iO^^OCiOlO'-i Q I S s '-2 E-< I w ABSOLUTE RANGES. 135 39. Table XLIX gives an analysis of the results obtained by grouping the ranges of Declination according to their amplitude. The first group gives the number of days in which the range did not exceed 30', the second the number of days in which the range, while in excess of 30', did not exceed 1, and so on, the seventh and last group giving the number of days in which the range exceeded 3. Days were arranged in two principal classes, the first including all days when the record was complete, or when loss of trace was due solely to one or both limits of registration being exceeded, the second including all days when there was imperfection of record through photographic failure, absence of sheet, or similar cause. Each of these two classes was sub-divided into two sub-classes, according as the trace did or did not keep within the limits of registration. The days are arranged under three seasons, Midwinter (May to July), Midsummer (November to January), Equinox (March, April, September, and October). The results for the remaining months, August, 1902 and 1903, and February, 1903, are combined. TABLE XLIX. Declination Ranges. Days when no incompleteness or failure of record, except through the trace going beyond the limits of registration. Days when trace kept within limits of registration. Days when trace went beyond one or both limits of registration. Total of Days. Range. Total of Days. Range. O'to 30'. 30' to 60'. 60' to 90' to 90'. 120'. laX to 150'. 150' to ISC'. Over ISC'. O'to so/. 30' to 60'. 60' to 90'. 90' to 120'. 120' to 1W. 150' to 180'. Over 180'. Midwinter .... 138 130 62 67 21 3 7 63 29 1 18 26 39 4 10 10 38 11 12 8 15 10 14 12 2 21 2 8 4 15 4 6 27 23 8 10 1 a 3 2 2 4 14 21 5 Midsummer .... February and Augusts Total .... 397 31 101 79 71 47 37 31 64 11 9 44 Midwinter .... Days when record incomplete from some cause other than trace going beyond limits of registration. Days when trace kept within limits of registration. Days when trace went beyond one or both limits of registration. 33 88 14 8 14 6 2 10 1 6 9 2 1 4 8 1 1 5 1 1 1 4 2 1 2 1 2 12 12 2 3 3 1 1 3 4 1 2 7 1 Midsummer .... February and Augusts Total .... 94 22 20 17 14 8 8 5 26 3 5 7 1 10 It will be noticed that out of a total of 111 days in Midsummer there was only one in which the range did not exceed 1, while at least 44 had a range in excess of 3. Taking the whole period, we find that out of 461 days for which the record was complete, except for the limits of registration being exceeded, 329, or 71 per cent., had a range over 1; 250, or 54 per cent., had a range over 1|; 179, or 39 per cent., a range over 2, and 75, or 16 per cent., a range over 3. Of the 581 days, complete and incomplete, included in the table, 407, or 70 per cent., had a range over 1; 218, or 38 per cent., a range over 2; 90, or 15| per cent., a range over 3; while 24, or fully 4 per cent., had a range over 4. On seven days the trace exceeded the limits of registration on both sides of the sheet, whose complete width represented from 4 50' to 4 55'. 40. Results for H corresponding to those for D, just discussed, appear in Table L. It contains two principal classes, each with two sub-classes analogous to those in Table XLIX. The ranges are again dealt with in seven groups, the first containing days in which the range did not exceed 25y, the second days in which the range exceeded 25y but did not exceed 50y, and so on, the last group containing days when the range exceeded 150y. Owing to the sensitiveness of the Horizontal-Force magnetograph, the limits of registration were exceeded in about one day out of two. In Midsummer the limits were exceeded 136 ABSOLUTE KANGES. in eight days out of eleven, so that our information at this season is unfortunately very imperfect. Even during Midwinter the range exceeded 25y on 90 per cent, of the days. TABLE L. Horizontal Force Eanges. (Unit ly.) Days when no incompleteness or failure of record, except through the trace going beyond the limits of registration. Days when trace kept within limits of registration. Days when trace went beyond one or both limits of registration. Total of Days. to 28. Hang 75 to 100. e. 100 to 125. Over 150. Total of Days. Range. 25 to 50. 50 to 75. 125 to 150. Oto 25. 25 to 60. 50 to 75. 75 to 100. 100 to 125. 125 to 150. Over 150. Midwinter .... 116 59 24 34 a i 45 19 15 29 21 3 13 18 13 4 5 10 4 6 4 1 11 2 1 29 94 58 43 17 8 4 3 25 7 16 8 29 12 13 10 7 13 2 8 16 18 8 Midsummer .... February and Augusts Total 233 79 66 40 20 16 3 224 29 51 62 32 50 Midwinter .... Days when record incomplete from some cause other thau trace going beyond limits of registration. Days when trace kept within limits of registration. Days when trace went beyond one or both limits of registration. 25 17 6 2 10 1 : : 13 5 : 1 7 1 1 2 1 1 1 1 7 29 21 5 1 1 1 5 2 2 7 7 1 2 2 4 2 3 4 2 1 11 4 Midsummer .... February and Augusts Total 50 11 10 9 3 3 1 62 1 2 9 15 10 9 16 Of the 457 .days for which the record was complete, except for the trace going beyond the limits of registration, 274, or 60 per cent., had a range over 75y, and 183, or 40 per cent., a range over lOOy. Of the 569 days, complete or incomplete, included in the table, 340, or 60 per cent., had a range over 75y, and 225, or 40 per cent., a range over lOOy. If we compare the D and H results we find that the percentage of days showing a range of over lOOy in H is closely similar to the percentage showing a range of over 2 in D. As 1' in D range answers to about 2y in H, the natural conclusion which one would also draw from the diurnal inequalities is that diurnal variations of force were considerably larger in the direction perpendicular to the magnetic Meridian than in the magnetic Meridian itself. The conclusion is almost certainly correct. At the same time, it should be noticed that even when the H magnetograph was least sensitive it was impossible to record a range much over 200y, while the possible limit for D was equivalent to nearly 600y throughout. Thus any deduction based on Tables XLIX and L as to the relative size of the average daily ranges in D and H is practically certain to underestimate the range in H. 41. Table LI gives some data for V analogous to those in the last two tables. Days are arranged in six groups, the first including cases where the range did not exceed 50y, the second cases in which it TABLE LI. Vertical-Force Ranges. (Unit ly.) Total of Ear ige. days. to 50. 50 to 100. 100 to 150. 150 to 200. 200 to 250. Over 250. 123 47 53 13 7 1 2 101 20 49 22 7 3 47 3 10 18 6 4 6 February and Augusts . . . 46 10 24 9 1 Total . 317 80 136 62 21 10 8 ABSOLUTE RANGES. 137 exceeded 50y but did not exceed lOOy, the last cases in which it exceeded 250y. Of the 317 days included, only 5 had a range as small as 25y, while 101, or 32 per cent., had a range exceeding lOOy. The latter figure is rather smaller than the corresponding percentage in the case of H. 42. In addition to the results for individual days, Table XLVIII gives means for individual months, rarying from 234y for January, 1903, to 50y for July, 1902. The monthly means, however, fluctuate somewhat irregularly, and can hardly claim to closely represent average conditions. The January mean, for example, appears abnormally large. This is partly accounted for by the exceptionally large ranges on January 11 and 12. It is by no means impossible that some of the irregularities may be due to errors in the scale values, or in the temperature corrections. The former source of error is most to be feared in the months of July, August, and September, 1902, the latter in the Midsummer months of 1902-3, when the temperature range was especially large. If we combine corresponding months from the two years, allowing equal weight to each day, we obtain the following somewhat more regular mean monthly values : TABLE LII. Mean Absolute Ranges of V. (Unit ly.) January . . . 234 February. . . 97 March ... 100 April .... 91 May .... 66 June .... 74 July .... 77 August ... 83 September . . 79 October ... 81 November . . 117 December . . 132 Year .... 91 Midwinter . . 73 Equinox ... 88 Midsummer. . 148 43. Table LIII gives the largest and smallest absolute ranges recorded in each individual month. TABLE LIII. Absolute Ranges. Declination. Horizontal Force. (Unit 17.) Vertical Force. (Unit ly.) Smallest recorded. Smallest recorded. Largest Largest Largest Smallest recorded. On day when record complete. On any day. recorded. On day when record complete. On any day. recorded. recorded. 1902 o / o / o / March . . >2 46 -5 '45-8 15-0 >196 66 162 44 April . . . >3 33 -0 22-5 15 -7 >216 36 36 190 50 May . . . 3 33-0 12-8 6-8 >224 30 11 Ifi8 28 June . . . 2 22-0 11-1 11 -1 >160 19 17 166 39 July . . . >3 23 -0 11 -0 8-6 159 23 17 165 15 August . . >3 30-7 20-3 16-0 >153 25 22 174 36 September . . 2 53-2 29 29-0 >112 29 29 92 39 October . . . >4 47 "I 50-7 50-7 >111 40 40 155 31 November . . >4 54-0 587 58 7 >160 52 >35 303 35 December . . >4 55 '5 1 28-8 1 28-8 >161 93 85 233 60 1903 January . . 42-7 '1 22-2 1 22-2 >161 84 53 415 94 February . . >4 54-0 15-7 157 >115 58 58 101 92 March . . >3 24 -0 39-6 39-6 >114 32 32 232 25 April . . . >3 47 -0 47-1 29 '4 >144 30 30 142 47 May . . . 3 43-5 24-0 187 >150 29 18 183 30 June . . . >4 30-4 17-1 15-7 >154 14 14 209 24 July . . . >4 0-0 23-7 23 -3 j >151 36 36 258 27 August >4 51 -0 39-7 397 >155 32 32 249 31 September . . 3 58-8 1 12-3 1 12-3 >155 51 51 161 33 October . . . >4 52 '5 >2 0-7 54-9 >157 >60 245 72 November . . >4 7-5 >3 25 -5 >2 67 >153 >23 164 161 December . . >4 54 -0 3 39-0 1 21-0 >1GO >75 318 122 1904 January . . . >3 50-2 1 41-3 >162 43 Even' in the case of D there are 16 months in which the largest range of the month is underestimated owing to the limit of registration being exceeded. In the case of H every month, except July, 1902, T 138 ABSOLUTE RANGES. suffered in this way. The information as to the smallest ranges does not suffer from any such uncertainty, but subsequent to September, 1903, the number of days of observation was too small to give results of much value. The smallest recorded ranges on days when the record was complete were forD, 11'- in July, 1902; H, 14y June, 1903; V, 15y July, 1902. Perhaps the most natural way of comparing the different months or seasons of the year as to their relative liability to disturbance is to consider the size of the ratio (mean absolute range)/(range of diurnal inequality). This criterion cannot be very readily applied in the present instance, except to the Vertical Force. For it the above ratio has the following values : Midwinter 4 1, Equinox 2 5, Midsummer 2 6. The ratio is thus decidedly largest at Midwinter. The same phenomenon has been observed at Kew in the case of the Declination. The practical equality of the ratios for Equinox and Midsummer suggests that in the Antarctic the former is not a season of specially large or frequent disturbances. If instead of taking for our criterion of disturbance the ratio of the mean absolute to the inequality range we were to take the amplitude o the mean absolute range, or the frequency of occurrence of specially large ranges, we should come to the conclusion that the equinoctial months were much less disturbed than those at Midsummer. For instance, in the case of D, taking all days, complete and incomplete, the absolute range was in excess of 3 on only 1 1 per cent, of equinoctial days as against 40 per cent, of Midsummer days. 44. Tables LIV to LIX show the frequency of occurrence of the daily maxima and minima in D, H and V at different hours of the day. For instance, from Table LIV we see that, taking the 24 days of April, 1902, for which data exist, the maximum in D occurred eight times between 8 and 9 a.m., six times between 9 and 10 a.m., three times in each of the hours 10-11 a.m. and 11 a.m. to noon, and once in each of the hours 5-6 a.m., 7-8 a.m., noon to 1 p.m., and 11-12 p,m. Maxima and minima were assigned for D and H even on days of incomplete record. In some cases the imperfection was of such a nature e.g. trace beyond a limit of registration during part of one hour as to introduce no uncer- tainty into the hour of occurrence of either maximum or minimum ; in a greater number of cases there was uncertainty as to one only of the two quantities. The number of days in which uncertainty affected the maximum and the minimum often differed widely. Thus in October, 1902, the hour of minimum in H could be assigned on 26 days, but the hour of maximum only on 12, In the case of V, only days of complete record were included, so that maxima and minima were equal in number. The occurrence of | will be noticed in some of the entries for D and H, especially the latter. This may mean equality in the extreme ordinates during two different hours, but it usually signifies that the trace was beyond a limit of registration during part of each of two hours. Days were omitted in which either limit of registration was exceeded during parts of more than two hours. If irregular disturbances did not exist, the daily maxima and minima would synchronise with the maxima and minima in the diurnal inequalities. But in the highly disturbed conditions of the Antarctic one would not have been entitled to assume a priori that the absolute maxima, for instance, would have shown even a preference for the hour of the inequality maximum. Comparing, however, Tables LIV and XIII, we see that the daily maxima for D cluster thickly round the hour 9 a.m., when the maximum presents itself in the mean diurnal inequality for the year. This is true of all the seasons, especially Midsummer. It is truly remarkable that at this season, when disturbances were so large, out of the 79 days for which data exist not one gave an absolute maximum during the 12 hours ending with 2 a.m. If we define the " concentration " of the frequency as the percentage which the occurrences during the three consecutive hours which, combined, give the greatest number of occurrences bear to the total number, we find in the case of the D maxima for the concentration 41 in Midwinter, 60 in Equinox, and 62 in Midsummer. The time of greatest frequency seems a trifle later in Equinox than in the other two seasons, but the difference is minute. DECLINATION MAXIMA. OCCURRENCES. 139 3 m I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I - - eo o i- o o o o o o o o H O O O O O O O o o o o H H O O o o o o CO O O -* 00 <* W o I > i i j CO rH CO fH r-C i-H i-H I I ~ I " I I i i i i i I I I 8 S S S ST -. .-. t-t i-t r-< .0 C Ql >O r-t i-i n o o (M O O . 8 BBS V till i-> < tn O ''I'll ' ' ' 1 I I 1 1 i 3 5 -S S s > c Ji 1 ** "o. S > ' II f 1 1 s i i ;! T 2 140 DECLINATION MINIMA. OCCURRENCES. g 5 S a a r I a 'S O o Q j < j | 1 1 j 1 1 1 1 1 " 1 1 I 1 1 1 1 1 1 CO O 00 ri i-l >O 19 CO t- -31 1 1 1 1 1 1 I 1 - - C9 M O 4V 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Illl O r-t O O CC fc b L. * 1 I 1 ..... fc . t, t. *'''' i i i 1 1 HOEIZONTAL FORCE MAXIMA. OCCURRENCES. 141 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 -w <. s pj g g o I O o o o o CO W O O t^ CO CO .H * w n O * ff\ o a o 1 1 1 -. rt 0000 1 1 1 1 1 1 1 1 rt CO O O CO 1 1 1

t-- ^H" ^H t- o> t 3- N i- CM CO CO -! 8 a CO ri aoto^Wt iicto^'ff'i^ieoo* io" iQ c*j oo r-l Cl > I- ,M CM CO s 3 00 s M a a a s- - - -a a - s a <-l 1 O i- 9 a 3 01 a . ft a s . .a ---:"* a a a a " >re ^H w t- O 00 s * rt s r ' -- 1 ^ H * s 00 ^-lOiCNWOsOi'f'-^l W i-t i iCONi (1/500 * CO t- H a a - t- WOOFMCM-Wt-lt-COlOr-ceqC^OlTTlCOCfllO kC CO f-H Cl 00 s , O-JOC.^-^^CC^^OCC^O^^^ - - CO 00 a t^ ^ ^ * " S3 ~ S s * a re --' ""a----a-- iO 00 CO OD - iO - 8 5J O?DCO>-l^)-*l7-tDWai^tD^)COCM^-iCM t- 01 -* CO a S pj ^NOi-ti "iO^"t WOOCOtfJCOCROi^-" - fll s - S d O^J-^COCRO^tO^-teDl-MCOOOC^CMtO - s -a - 9 a S oi ----*3*--si-----a CM t- cq o * - 8 CO CO^H^OO'ftO^'H'tDCMt-OOCMOf-Ht- (N i-i C-J (N -< T-I to CD O CO * * t- t- t-.tj i ^'ao'Cot~Nt^ -, ^ a a ~ a e " ^tOivraODt-WMiCt-WC^COt^l^Nr-IO 2 9 a s 01 -< rH -! rt ^-t ^H r- M . t-ioocOTfiirtc;^-ico^Hoooio i w^Oi in CC ^" O V b- TO j OOOtDCOO>m^OOoOO iCOt 1tOtD-^ CM 1 ^H b- -H C*l CM a a S s a 1 o a f 8 . fr e ,f * 4 SJ fe fc gg-ft-jl^gisja.xi ^ ^ " S 1 s 1 , o B s ~ o c 1 * February : = | _g SP O 4H O I J - 240 CHRISTCHURCH TERM-DAY OBSERVATIONS. ** v ;,;, 2*. , S 1 2 S S g tn us S CO * CO ift S" 5" TC I" I - rt ec PH ^1 O Ol 00 t~ CO CO Ol t- ^H CO O to o> o o t- OJ eo I-H M weoowcooioit^oioi O N CC eoeotooicooi-HCOco CC ?1 iO OV i oi^-^oooo^co^-ifMi- PH M o cq .-i t- ift ^yinc^omirfiocowcoeotoiKojNire^OJ CN rt i i fH ^" i-l <-" " -! OJOit-l'rH ^H(?1WC^ eoaoio-^coo^cooejcoeocoiftoocooo is^Hoogi-^c'ioocoi-iiraooo^coi-tD g| iQ 40 **! A tw fH fQ *^ ^ P"l 4 1 4 tt t* m j M fc 01 ^< IH CNcS^H I-H" t I i < S " ^ i I i I R 8 9 J 3 N Si s ^ ss rtN oc,oc<, rt c<,=, NN oc, 3NU53 c, ^ , - H - ri rt oooo=,o ra -c,oc, M c, ror H N M oo-o C-* w a rtM oc,o rt o Nrt o^.o=, ra c,o NN - * . t-^ .^o^^^.c^^.e.o.oooooo- oo. N o ? N oooo N ^- rt o ra oo rtrfrt o oo N TO o iC o t oe, TO ooc,oo. Nro c, N o rt - Nro o o O O . ^ ea s ^-^O^O^^^^OOO-.^OOC.OO rt0rt H - CJ aS ro ,o,oo, ret ,oo N -Ho rt o=, 0=00,^0 - i a, ON o N c, rtN c, rt ooo rt oo-c, oo^oo to ...^.o-o^oo^^^c^.^^o- .,., ^^ id ^*" V ''^ rf .,- rt -,0-0--^. W . M . M H . i-H w rt ooc, m oc, rtrtro c, NU5 ooc, rtM rtNNNM H " be s B 3 a S r.&.a ^ - 1 S S S S E? E 1 8i:o.-5-8:3"j.2 ^ . B B . 1 : * *i > I - I. 1 s 'l ! * : a " * 1 1 * 1 CQ rfl : bO ci e MH O 1 2 i 242 CHRISTCHURCH TERM-DAT OBSERVATIONS. < ** + ooQcoct~*-ooot-i + s s s S 8 >> . g= Ol^^^l^rf'^'COOJ^-r-HO' 1 O CO t O ^H ^H "" OO a 8 C*l 'o * w 3J a 'V a"- g' g ft S S 2 n ?i CO ri R a s ~- - - * > - * g s a a a M ? 2 a a w CO S3 a . ag ^.,.., aw g...jj 9a o 5T oo o 9 01 0^.ONN ! D, Ujn t-OOC<,o na . w oo as. *^ -. 8 " ^ *"" ^ 1-1 *"" S - 1 ^ rH IN - " - 1 S8 a tfj o^^ n ,o N _o N c,=, rortN;;5 , rt * O R CO * M . . M - M . . fl '. M . . , H . 71 V O * n M N . s =, too > roro ooc, t -oo, t , CO CO .-< - s H rt -< s - 1 " * o ! . 2 c,oc, rt , t0 c, t ,c, NU , rt CO ol SN ., S C,, ro oO,^,^C, ra a - * to - 1 t-i t-H rt Ci -j; t- 00 CO * to a9s-----..-..- afl a PO CO ^^ n CO * - s 2 s - - * s a s 2 - - s 2 s a s 2 ^U to o fT 01 (M -=o,o a > !0 occ..c, n ,o 230 ,o 3 - - 9 (N ri .. ^ .... i. a . '^ I - Ol lA W . ci - SS!Sa53S3 S8^SS2 SI " S a S . jc 9 h i 8 .&(<& s s ss lft*IB *** - i* &'< * * : B - : S 1 I ( Kj 11 V V V ^J "tt O > W o QJ a> ^ Q 03 t > CO J S a " 3 ' -3 1 , s | s : d o Q d I W H CHRISTCHURCH TERM-DAY OBSERVATIONS. 243 !.&>& jlj OOOICTOJ > i + a ' 5 ^> . Ill - 5 *o 233 = JBSa3SS2SSSSSSS5 E g a. s 2 S it ,., .. S *..,.g S .oS,.. B .. o t- " to ^ s 1-1 iH ' * sj . ,o^c, rora . N e )re o >2-'- s ie a - T* " ^ " " " a s s & i- rt H a* o^^coco^^o^c.ooc^^oco^^ccoo CO CO 10 -* <* * s * * -" 04 ~* *^ * flj M , SN ^ NK o re c, N c, nroN!0 ^ w CO iT - 4H 9 N _ rt 1-1 o^c,^ ON c,ooc, ro N ,c,c,c, . m - " CO 1-1 cd OOC, Nror H nt , U: , rt o, N N CO^C, *0 * CO =, - C, roNrt ,H t -C,C, N NN ^ W =,^ Sra , o w "CO ^ "* & ^ " s s - SI rt 4 ^.co^^^co^^oc^cccca^-o^go *" CO - O) -* s ^ "" " - 1 bC .5 a b 1 g B fr fi * - ^ t 1 fe fc fe ^ (Nh"& ::i 5"'''= 3 "?_^ J2 i J=> 1 a =9 a ? ^ v i a) a o s In s ' 1-1 rt ^ February ] I ^ $, ! g .a t~t o a o W ^ > w 2 I 2 244 CHRISTCHUKCH TEKM-DAY OBSERVATIONS. |b| e , ID 1 3 i , s w ft ! p 8 00 8 e 9 s 4 s p o CJ Si B ij g Oi ** a OJ 2 p et CO Ok OB tp 40 to - 04 (C o 00 rt to m CO r Ol o s H 9 o US H o Ol p 00 OP CO 00 8 a S 9 t- op Hft s OJ oo s 2 CO cn H> 00 s 00 ? 3 CD OO o at i p IO 00 0t 2 CO % S 2 N CO to Oft o t- 2 CO CO p to O) o to tD P p o tt s 00 Q p CO CO o s lO 9* to O| OJ CO p J 2 O) CO V 5 3 9 i CR Ofe p 00 9 *" 00 2 00 CO t Oi CO Ol 00 00 Ol rt 2 QO 00 09 o J 00 9 t CR s d CO 00 t- 9 o s op H 9 CO CO o *- CD p p t- a p a 2 CO H rf - p CO fc s - a I s CO CO* OO fl* n o CD op P M a CO CO H - 00 p p Q *r 00 n i p s 8 s * OJ S CO 0> ^ 08 CO 00 ;: a 5" * p 9 p " t- -4. IP s ip QO - Gft .2 p s J 1- co 2 e SO E ri 1ft CO * K* 00 i m et 9 oo O O) N s 01 00 H CO 00 to CO T" - o OB M p o OJ ^ p ft S IP S Hour ending . . . DECLINATION. lar Midwinter Midsummer HORIZONTAL FORCE. Midwinter Equinox VERTICAL FORCE. i H a o3 6 8 t_ 4 J CO p 1(3 CO 1 + W CO ri p * o CO p o Set a o o 1 CO tO o o 1 p p rH rH d p * o o 1 * o> US -T o o 1 rH O 1 O> rH CO CO 1 s p OS o a s m >o II - o o 1 I 7 rH rH 1 2 Ol * o o o 1 N p o o 1 IQ T 1 * 7 s 9 - ? 1 ? 9 1 CD t 0* rH CD 1 p p "" 1 rH CO * o o o CO W N o o 1 9 o o o g S o o CO" - o p o rH I(S O a s O * (M o o 1 S - o CO rH CO T* 9 O s s o o 1 5 g o o 2 p - o o o CD US T* 1 9 o o 8 S O 01 CO - o ? s s 1 t* Ol p 1 d - o t- o CM o o o s s O O ? 9 o o d * o o 1 US CO CO rH 1 S o o 1 s ? o o 1 o - o p o 1 CO rH 1 1 o o 1 rf - o -* 1 g p e=> Os N CO W o o 1 f SI ! 00 - o 9 o 1 o o 1 o o 1 US CO o o 1 - p - o us 1 g a 1 00 to 1 ? p 1 CO p - o o 1 ^J< CO o o 1 O Oi 1-, p O 1 o to rH rH us 5 1 9 P o e 1 o o 1 O) CO rH ^ 4 CO " * CD o 1 US O US " 1 o o 1 o o 1 m " V tfii CO t- CO US p t b- us o o fi CO ? ? 9 ? ? s s + + + rJ t-. s rH rH O OS rH O CM M ? 1 V, PS M M M M Hour ending \ 1 V ' H Midwinter . j I i V s I s a I -2 a C ^ U O d 8 to 8 It- O> US ?7 * CO * * CO -, N CO CN 0) CD o T t- \ US CO US O C- CO rH '/..CO 0> I 1 1 S t- \ p O ' ?l N rH CO O CO CO CJ CM + i ** 7 t us f 40 1 7 i T 00 p o 1 o o us CD o> o ' + ? 7 CO CD f ** It} CO O (H ** + ' + + (-.1 O CO us S S 2 J ' CO V CO CO * rH CO CO 9) 1 rH O C-1 O CO rH + + 1 US 00 o rH US *f t O ^f ? + ' ? s CO o W rH p W O> * ^0 CO CO + 1 + CO OS CO o 1 W N i-H rH US O CO rH CO rH CO O + 1 1 (N s ft 4 rH rH US US 00 US 1 + + 3 CO o ^ CO rH '*<=> + + 1 o CO CO 9* o 1 CO O CO (N C4 O 1 1 -t- CO CO oo rH O p US W -jf" 00 9 CO p 1 ff4 O US CN CO O t 1 1 N ecj CO O> US tO t rH t- 1 C9 O CO O CO US 1 1 I 4 US o 1 O CO O> O '> i-t MO US O * rH + 1 1 < 00 US ft M CD CO US O US O * 7* ?* + ? ' + CO CO t-. 9 ws s : s s s s ci t- M) ': -^ rH CO p OS CO O O CO O> t- CO p O1- IA O ^1 rH "' CD 4V 1O * TH N O O : M M M i-H H K" Hour ending " v 1 i s 1 s H 1 a g i 1 I 246 COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. APPENDIX B. An Examination of Antarctic disturbances from October, 1902, to March, 1903, which are simultaneous with Arctic disturbances discussed by Prof. KR. BIRKELAND. 1. AFTER the discussion of the Antarctic magnetic data on pages 73 to 200 had been completed, there appeared a large volume by Prof. KR. BIRKELAND* which contains a great mass of information respecting magnetic disturbances in the Arctic, from October, 1902, to March, 1903, which were contemporaneous with the observations discussed in this volume. Prof. BIRKELAND had the following four Arctic stations, all provided with self-recording magnetographs : - Station, Latitude N. Longitude. o / 77 41 o / 14 50 E 73 17 53 57 E 69 56 22 58 E 66 15 22 30 W Guided by the records from these stations and from Potsdam, Prof. BIRKELAND made a list of disturbances and issued a circular to magnetic observatories requesting copies of the records obtained on the days specified on his list. He thus became possessed of records of disturbances at 25 stations, including the four mentioned above and the following : In Europe Bossekop, Pawlowsk, Stonyhurst, Wilhelmshaven, Potsdam, Kew, Val Joyeux, Munich, Pola, San Fernando ; in Asia Tiflis, Zi-ka-wei, Dehra Dun, Bombay, Batavia ; in N. America Sitka, Toronto, Baldwin, Cheltenham ; in the Pacific Honolulu and Christchurch. The names in each group are in order of latitude from north to south. BIRKELAND reproduces the disturbed curves in 21 plates, each dealing with a disturbed period varying in length from 2 to 20 hours. Some of the observatories supplied no Vertical-Force curves, and only a few supplied material for all the magnetic storms. Still, the plates represent what is probably the most extensive series of contemporaneous magnetic data that has yet been published. In addition to the plates, the volume contains over 160 charts representing the results which BIRKELAND has deduced for the disturbing forces at the different stations, at different stages of the 21 storm periods. These charts are based on elaborate measurements of the magnetic curves and represent a large amount of work. Besides discussing the charts, BIRKELAND deals with experiments which he has made with a miniature Earth, or terrella, magnetised and exposed to kathode-ray discharges in a high vacuum. The experiments are intended to serve as an auxiliary to the elucidation of the causes that produce magnetic disturbances. The discussion of observations, experiments and theory occupies more than 300 large quarto pages. 2. As already mentioned, the discussion originally contemplated of Antarctic magnetic storms had been entirely completed when Prof. BIRKELAND'S volume came into my hands in February, 1909, and I was exceedingly reluctant to re-open the subject, considering the long time that had already elapsed since the work was begun. It was accordingly decided that no change should be made in what had been already written, and pages 73 to 200 are thus absolutely unaffected by any results or views in BIRKELAND'S! volume. When Prof. BIRKELAND'S conclusions and mine harmonize, the harmony thus merits increased consideration ; when they differ, the difference at least owes nothing to prejudice. * ' The Norwegian Aurora Polaris Expedition, 1902-1903,' Volume I. t The reference in Chapter IX is to Prof. BIBKELAND'S earlier work, ' Expedition NorTegienne de 1899-1900.' COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. 247 After having read Prof. BIRKELAND'S volume, I came, somewhat reluctantly I must confess, to the conclusion that those responsible for the production of the present volume would be open to criticism if the opportunity afforded for contrasting Arctic and Antarctic records were not utilised. This appendix has accordingly been written with the object of supplying information as to what was happening in the Antarctic during the disturbed times selected by Prof. BIRKELAND. Its primary object is to inform, not to criticise, and if it contains anything that savours of criticism, this is mainly in order to explain why Prof. BIRKELAND'S exact procedure has not been followed. 3. The method adopted by BIRKELAND in dealing with magnetic disturbances is fundamentally different from that adopted in Chapters IX and X. Practically following SABINE, he defines disturbance at any given hour as the difference from the normal for that hour ; and what his tables give, and his charts illustrate, are the values of disturbances so defined at different stages of each disturbed period. He believes that the cause of the disturbance is an electrical discharge, or a series of electrical discharges, in the Earth's atmosphere, such as he has succeeded in producing in the vicinity of his terrella, and his ultimate object apparently is to calculate the position and intensity of these discharges. One great difficulty, as I remarked in 90, Chapter IX, is the fixing of a iwrnml value. In doing this BIRKELAND seems to have been materially helped by the fact that at some observatories, including those provided with Kew-pattern magnetographs, it is usual to have two days' curves on the same photographic sheet. If the one day's curve happens to be undisturbed, its form greatly assists the eye in deciding as to the nature of the disturbance in the other. This is an advantage which I have often had occasion to appreciate myself. The Arctic stations, however, and some of the others had magnetographs of the Eschenhagen pattern, like those of the " Discovery," and there must have been considerable difficulty at times, as BIRKELAND himself allows, in deciding what the departure from the normal really was. This difficulty had probably a considerable indirect influence on BIRKELAND'S choice of disturbance periods. Those he has selected are largely represented at Kew and other non-polar stations by "bays" of comparatively short duration. What is meant by a "bay" will be readily grasped by reference to the accompanying figure. The continuous line ABCDEFGH represents an imaginary magnetic curve having two bays, one, BCD, occurring at a time when the regular diurnal variation is slow, the other, EFG, occurring at a time when it is rapid. The ordinates CM, FN drawn on to the base line represent the excess above the constant base-line value of the values of the magnetic element answering to the times M and N. The broken lines BC'D, EF'G are intended to represent the imaginary undisturbed curve, and the intercepts CC', FF' on the ordinates represent from BIRKELAND'S point of view the disturbances. In such a case as that represented by the figure the method appears simple, especially when the disturbance occurs at a time when the diurnal change is slow. In practice, however, there is usually a difficulty in deciding where the " bay " begins or ends, and the relative position of its two extremities is usually not quite what one would anticipate from the trend of the curve prior to its commencement. This latter difficulty naturally increases the longer the duration of the bay, and the more rapid the regular diurnal inequality changes at the hour. As a general rule, during really active magnetic disturbances, whilst bays of a kind are not infrequent, the curve adjacent to them is itself disturbed and sinuous, and affords very imperfect guidance as to where the normal curve would come. On days of real disturbance, one is usually obliged to have recourse to the 248 COMPARISON OF ARCTIC AND ANTAECTIC DISTURBANCES. curves of adjacent days, if quiet, or to a regular diurnal variation derived from a number of days. The latter alternative is, of course, the more satisfactory theoretically ; but there is this difficulty, which I have dealt with elsewhere, that the regular diurnal inequality which one gets depends on the nature of the days from which it is derived. For instance, taking means from 11-year results at Kew, the departure from the mean Declination for the day at 2 p.m. on a representative January day is : + 2' -21 on the Astronomer Royal's selected quiet days, + 2' 66 on ordinary undisturbed days, + 4' 88 on the average highly disturbed day. For simplicity, we may suppose the secular change and sun-spot influence non-existent, though in reality these are complications which have to be reckoned with. Let us suppose that on a certain disturbed day in January the Declination at 2 p.m. departs from the mean for the month by + 3' 66. Is the "disturbance" + l'-45, -fl'-OO, or -r-22 1 ! The +4' -88 departure from the daily mean on the representative disturbed day in January, be it noted, represents a regular diurnal inequality, or at all events something which we have no present means of distinguishing from a regular inequality. Its excess over the ordinary day value may, of course, indicate a tendency for a particular phase of disturbance to occur at a particular hour of the day, but it may mean that the causes operative in producing the regular diurnal inequality are for some reason e.g., increased conductivity in the upper atmosphere more effective on disturbed days than others. This source of uncertainty is equally present when the normal curve is derived, as BIRKELAND seems to have derived it, by reference to a day or days adjacent to the disturbed one. The adjacent days may be of the very quietest type, or may themselves be considerably disturbed. There is a further complication in that disturbance has sometimes a tendency to be associated with a temporary alteration in the value of magnetic elements, which disappears gradually like after-strain in a metal. Horizontal Force, for instance, is sometimes very considerably depressed for days after a really large disturbance. Thus it may make all the difference in the world to one's decision as to whether at a particular hour the element is above or below the normal value, if one happens to take for comparison the day after a disturbance rather than the day before. In one instance I observe that Prof. BIRKELAND noticed the occurrence of this precise source of uncertainty. In the case of the Antarctic curves, the uncertainties which exist elsewhere are mostly much enhanced. We have seen in Chapter III, Tables XII and XIV, that disturbance exerts an unusually marked influence on the amplitude of the regular diurnal inequality. The Antarctic D and H curves were always sensibly disturbed, and the disturbance was usually sufficient to obscure, if not to totally conceal from the eye, the trend of the ni ii 11 Equatorial Elementary polar Compound Elementary polar Compound Equatorial Compound Elementary polar Compound Elementary polar it ii / 6-5 16-2 4-4 8-6 6-9 4-8 18-2 14-5 4-4 3-6 3-8 5-2 2-0 4-7 19-3 8-3 18-8 6 *5 11-2 14-7 11-7 7 13 70 24 45 31 34 98 57 35 17 8 25 10 28' 49 43 55 41 25 39 26 11-12 23-24 27-28 28-29 29-30 ,, 31 to November 1 ... November 23-24 14-15 24-25 26-27 28 1903 January 26 26-27 February 8 . . ., 8 10-11 15 March 22-23 ... 30-31 6. The principal object in mentioning the ranges at Kew is to bring out an important point which I hardly think BIRKELAND himself quite realised, and which I am confident will not be realised by readers of his volume who are not themselves experts in Terrestrial Magnetism. The mean value of the absolute daily range in D at Kew derived from all days of the eleven years 1890 to 1900 was 13' '57. This value, it will be observed, was exceeded on only six occasions in the table. During the eleven years the absolute daily range at Kew exceeded 20' a value not once attained during BIRKELAND'S disturbances on no less than 12 per cent, of the total number of days. It must not, of course, be forgotten that the average length of the period covered by one of BIRKELAND'S plates is slightly under 10 hours, and that the majority of the periods do not include the hours at which the daily maximum and minimum most frequently occur. Still, taking everything into account, the fact remains that the great majority of the days selected by BIRKELAND were not what are ordinarily called disturbed days. In the Arctic, it is true, there were movements which would rank at Kew or any other non-polar station as magnetic storms, but there is not a single one of the occasions on which the phenomena at Kew would ordinarily be dignified with that name. On perhaps three occasions, October 3 1 , to November 1 , 1902, November 23-24, 1902, and February 8, 1903, one would have little hesitation in describing the day as disturbed, but on the other hand there is quite a considerable proportion of the days which one would be likely to describe as quiet. It is not merely that the movements on BIRKELAND'S selected days were small, but that they were few in number, and in many cases represented slow changes. In the case of an ordinary magnetic storm at Kew, not only would the range be much larger than in any of the days selected by BIRKELAND a range of 30' in D and 300y in H represents what may be called a second class storm but the large movements would be much more numerous and some of them much more rapid. If we take a really first-class storm, like that recorded at Kew on October 31, 1903, it represents an altogether different order of conditions. Not merely is the range five or six times larger in D, more in H, but there are dozens of rapid oscillations, altogether without parallel in the most disturbed cases selected by BIRKELAND. The expenditure of energy during a first-class storm may, for all we know to the contrary, be 10, 100, or even 1000 times greater than that during BIRKELAND'S most disturbed day, and we cannot even say with certainty that the ultimate source of the energy, or the way in which it is expended, is the same in the two cases. What I have called a first-class storm is apparently experienced as a large storm over all the world or at least over a very large part of it and is invariably, or almost invariably, COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. 251 accompanied by aurora visible even in the south of England, whereas the disturbances treated by BIRKELAND attained the development of a first- or a second-class storm if anywhere only in a portion of the polar regions. In the stations in temperate latitudes they were mostly of the size one meets with every other day, and if there were any auroral discharge accompanying them, it was not visible outside the region where auroral frequency is high. 7. Returning to the table particularising BiRKELAND's plates, we have now to consider the significance attached to the descriptive terms appliqd to the disturbances. Explaining the term " equatorial," on p. 62, BiRKELAND says : "... it is not unusual to find perturbations that are best developed and most powerful at the Equator. It has even been found that these perturbations in the regions about the Equator act principally upon the horizontal intensity . . . Such perturbations we ... call equatorial ... Of these there are ... two kinds . . . , such as produce an increase in the horizontal intensity, and such as produce a diminution . . . The first . . . we have called positive .. . . , the second . . . negative equatorial perturbations." Of polar elementary storms BIRKELAND says, pp. 84-85 : (1) " They are comparatively strong at the poles (meaning the north polar regions). The simul- taneously perturbing forces, even as far north as the 60th parallel, have already sunk to about a tenth of their strength in the auroral zone. (2) " They are of short duration, frequently lasting not more than 2 or 3 hours. (3) " The conditions before and after are comparatively quiet. (4) " The oscillations at the (north) polar stations, especially the more southern ones, run a simple course. At the poles, they are often characterised by a simple increase to a maximum, and decrease to zero. We may sometimes, even at the northern stations, have to some extent an undulating form, answering to a slow turning of the perturbing force." Of compound perturbations no general definition seems to be given. Judging by individual cases they are a combination of phenomena, " equatorial perturbations " predominating at one stage, and " polar elementary storms " at another. BIRKELAND'S discussion of " cyck-median " storms on p. 144 is somewhat lacking in clearness. After expressing his belief that " electric cyclones, wandering over the Earth's surface," according to a suggestion of Dr. AD. SCHMIDT, are a very rare phenomenon at least in large storms, he adds : " It appears, however, that there is a class of perturbations that are due to current-systems which appear in lower latitudes at a height above the Earth that is small in proportion to the Earth's dimensions ... In the whole of our material, we have not found more than one considerable perturbation that in its entirety must be due to systems that come near to the Earth in lower latitudes." It is this single occurrence (October 6, 1902) that is characterised as " cyclo-median." Judging by a remark on p. 150, the term was intended to signify that the disturbance was "as great in medium as in high latitudes," and that the electrical currents to which it was due were " vortical in form." 8. It is difficult to discriminate between Prof. BIRKELAND'S observations and his theories, as the two are so interwoven in his pages. Thus some reference to his theoretical views may tend to clearness. He believes that the "equatorial" perturbations are due to electric currents encircling the Earth near the plane of the magnetic Equator, at a distance above the Earth which is similar to, possibly greater than, the Earth's radius. A current thus situated in a plane perpendicular to the Earth's magnetic axis would naturally give a force which in the equatorial regions would be roughly in the magnetic Meridian thus affecting H almost exclusively and the intensity would be greater in the Equator. Unless the height of the current were large, the disturbance would fall off rapidly as we departed from the Equator, and in even low latitudes there would be a large vertical component. There does not seem to be in the volume any close comparison of the amplitude of the " equatorial " perturbations experienced on the same occasion at different stations ; but the numerical data as to the disturbances during individual equatorial perturbations do not show such predominance in the equatorial regions as the definition leads one to expect. So far as I can judge, a considerable number of the movements discussed in our Chapter IX, including 2 K 2 252 COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. the " sudden commencements," would be classified by BIRKELAND as " equatorial " perturbations. In their case, however, as will be remembered, the amplitude at Colaba and Mauritius was usually less than that at Kew, and much less than in the synchronous movements seen in the Antarctic. BIRKELAND belie ves an "elementary polar"* storm to be due to what he calls "precipitation" in a comparatively limited Arctic area. By this he apparently means an influx (and efflux) of charged ions. In some calculations in the volume this is treated as equivalent to an electric current, approaching and receding from the Earth in lines which, if produced, would intersect at the Earth's centre. The current is regarded as stopping short of the Earth, then travelling in a straight line to a point equidistant from the surface, and finally receding. In some special cases on p. 103 which seem to be supposed to represent probable actual conditions the height of the connecting (so-called " horizontal ") portion is put at 200 or 300 kms., its length being taken as 1600 kms. and upwards. In the case of the " elementary polar " storms the disturbance at any given instant is large over only a very limited polar area. In temperate latitudes the disturbance is small and diminishes rapidly as the distance from the area of " precipitation " increases. This area usually keeps shifting its position, so that the disturbance travels across the polar regions. If Prof. BIKKELAND had seen what has been called in Chapter X the " special type of disturbance " he would not unlikely have called it a south polar elementary storm. It seems, however, to present a much greater definiteness of type than BIRKELAND'S " polar " storms, and its duration is usually much less than the two or three hours which BIRKELAND speaks of. 9. In his volume BIRKELAND does not take the disturbances in chronological order when discussing them, but treats first the " equatorial," secondly the " elementary polar," thirdly the " cyclo-median," and finally the " compound." One finds, however, that there are few of the occasions on which BIRKELAND failed to detect at one stage or another the presence of both " polar " and " equatorial " disturbances. I have thus thought it simplest to follow the chronological order, as BIRKELAND himself has done in the case of the 21 plates at the end of his volume. In what follows, the references are to BIRKELAND'S plates unless the contrary is explicitly stated, and the reader is strongly advised to have these plates before him while consulting the details given here. 10. October 6, 1902 (hours 13|-15, Plate I). Of this "cyclo-median" disturbance, BIRKELAND, p. 150, says: " Its chief characteristics are that it is as great in medium as in high latitudes," also " the effect over the district Wilhelmshaven, San Fernando, Stonyhurst, Pola is of about the same magnitude." The difference from the " equatorial " disturbances, which otherwise it closely resembles, is that there appears hardly any movement in Asia or the Tropics. BIRKELAND infers that it must be due to electrical currents at a height " small in proportion to the Earth's dimensions." The phenomenon, as presented at Kew and the other stations where it was best developed, had for its most prominent feature a sudden change in D, commencing at about 14h. 14m. At Kew, in the course of 5 or 6 minutes, the Declination needle moved about 4' 5 to the west and then returned much more slowly towards its normal position, taking nearly 30 minutes to reach it. H fell as D increased, but the total fall was only about 6y ; the return to the normal position was slow as in the case of D. The curves were exceedingly quiet for some hours before and after this movement. While the amplitude of the movement was very trifling, the isolation of the D movement and its nature are certainly unusual. Whether there was or was not a corresponding movement in the Antarctic is open to doubt. In the Antarctic, as elsewhere, the day as a whole was exceptionally quiet ; but, as was invariably the case in the Antarctic, both the D and the H curves show numerous small oscillations. The V curve was certainly not more disturbed between 14h. 10m. arid 14h. 50m. than it was earlier in the day, and decidedly less disturbed than it was a few hours later. The same appears true of the D curve. There was, however it may be a merely chance coincidence a distinct bay on the H curve, whose inception was at least very nearly simultaneous with the commencement of the disturbance in Europe. Between 14h. 10m. and 14h. 21m. H fell 9y, two-thirds of the fall taking place between 14h. 13m. and 14h. 18m. H remained below its previous value until 15h. 23m., the return movement being much the slower. * The terms " polar elementary " and " elementary polar " are applied indifferently. COMPARISON OF AECTIC AND ANTARCTIC DISTURBANCES. 253 This case is a good example of the uncertainties attending the application of BIRKELAND'S method to the Antarctic curves, even under the most favourable conditions. From inspection of the adjacent portions of trace one would conclude that D and V were normal, but H distinctly below the normal value from 14h. 13m. to 15h. Om. If, however, we compare the values of the elements at 14h. 53m. (otherwise 2 a.m., L.T., on October 7) with the means for the same hour from the four nearest days, this is what we find : D. H. V. O / October 7 (2 a.m., L.T.) . . . Adjacent days (2 a.m., L.T.) . . 153 3 -6 152 54-9 06743 06739 7258 7261 These figures point to exactly the opposite conclusion to that suggested by the form of the curves themselves. 11. October 11-12, 1902 (hours 12-2, Plate II). This " compound " perturbation is regarded by BIRKELAND, p. 251, as divisible into three so-called "sections," (i) from llh. to 17h. 20m., and (ii) from 17h. 20m. to 18h. 30m. on October 11, (iii) from the last-mentioned hour to Oh. 30m. on the 12th. The disturbance in section (i) is regarded as " mainly a positive equatorial perturbation," accompanied, however, from 12h. 25m. to 13h. 15m. by a " considerable polar perturbation." " The farther we go," BIRKELAND says, p. 252, "from the (N.) polar regions, the less perceptible does this brief polar perturbation become .... At Zi-ka-wei and Dehra Dun it is distinctly noticed, at Batavia it is almost imperceptible. At Christchurch, on the other hand, there is a rather violent perturbation .... (which) cannot have been produced by the same system .... for the effect of the latter is imperceptible (?) even at Honolulu and Batavia. The explanation of this seems to be that simultaneously with the descent (of ions or corpuscles) in the north, a similar phenomenon appears near the South Pole, and it is the effects of the latter that we observe at Christchurch." This quotation has been given at length because it constitutes one of the very few references which I have observed to the possible existence of disturbance centres near the S-pole. The "explanation" was presumably purely a hypothesis on BIRKELAND'S part, as he had no records from south of Christ- church. Section (ii), 17h. 20m. to 18h. 30m., was characterised by "violent storms in the Arctic," especially at Matotchkin Schar, but " the effect of the equatorial storm is still perceptible." Section (iii), 18h. 30m. to Oh. 30m., "is characterised by a long polar storm," in the course of which, however, there appeared three short " intermediate " polar storms, the first with maximum about 18h. 34m., the second lasting from 20h: 45m. to 21h. 20m., the third from 23h. 10m. to Oh. 25m. At Kew and all the other non-polar stations whose curves appear in Plate II there was a fairly sudden commencing movement, which does not seem to be mentioned by BIRKELAND. The H movement at Kew commenced about 12h. 18m., a fall of 2y and a rise of lOy occurring in about 6 minutes. A peak, representing a movement of about 1' to the west, appeared also in the Kew D curve at about 12h. 24m., i.e. simultaneously with the maximum in H. Plate II shows this commencing movement distinctly at Colaba, Batavia, and Christchurch. It forms in fact the commencement of the movement at Christchurch, which BIRKELAND suggests may be due to currents near the S-pole. At Kew and the other non-polar stations the disturbance consists mainly of bays in the H and D curves, with two or three small but fairly sharp peaks, the greatest departures from the normal appearing between 17h. and 22h. Christchurch differs a little from the other non-polar stations in that there is a bay on the H curve between 12h. 18m. and 14h. 10m., which commences less suddenly, but is larger than those encountered elsewhere ; the later movements at Christchurch, on the other hand, are exceptionally small. The division into three sections seems somewhat arbitrary, especially the line of demarcation drawn at 18h. 30m. between Sections (ii) and (iii). This comes in the middle of a very rapid rise 254 COMPAEISON OF AKCTIC AND ANTAKCTIC DISTURBANCES. in H at Matotchkin Schar, which according to Plate II extended from about 18h. 10m. to 18h. 35m., and which was immediately followed by a movement in the opposite direction, of so closely similar a character as to suggest its being an essential part of the same phenomenon. The significance of this will appear presently. The Antarctic curves were fairly quiet, according to the Antarctic standard, until about 17h., but the H curve shows a prominent bay from about 12h. 13m. to 13h. 57m., the element being depressed and the minimum coming at about 13h. 18m. The fall and rise were each about 60y. During part of the time covered by the H bay there was also a bay on the D curve, a rise of 58' being followed by a fall of 45', and the maximum coming at about 12h. 53m. As 1' in D represents a force of about l'9y, the D movement was really the larger as well as the more rapid, but owing to the high sensitiveness of the H magnetograph the H movement appeals more to the eye. During the occurrence of the bays on the D and H curves the V trace shows numerous small oscillations of an irregular character, bearing no obvious relationship to the D and H changes, and not so suggestive of the " special type of disturbance " as the H trace is. These D and H movements are synchronous with the " polar " storm of Section (i). Also, whilst there is no conspicuously rapid initial movement, the time of commencement is at least very approximately the same as that of the small sudden movement seen at Kew and elsewhere. After 17h. the Antarctic curves certainly deserve to be called disturbed. The H trace showed the following changes, superposed on which were the usual short-period smaller oscillations : 17h. llm. to 17h. 16m. fall 28y, 17h. 16m. 17h. 20m. rise 23y, I7h. 20m. 17h. 31m. fall 40y, 17h. 31m. 17h. 42m. rise 45y, I7h. 42m. 17h. 53rn. fall 41y. D showed a number of small oscillations, but none conspicuously large ; it increased, however, 90' between 16h. 53m. and 18h. 3m., going off the sheet at 18h. 3m. for a few minutes. During this time V rose and fell only about lOy, though there were numerous small oscillations. The above movements seem to be associated. They are synchronous with the earlier part of BIRKELAND'S Section (ii), but also with the end of his Section (i). They are followed by a larger H movement. Commencing to rise at about 18h. 21m., H, after increasing 74y, got off the sheet at 18h. 27m. Re- appearing at 18h. 31m., in the next 29 minutes it fell 108y, going beyond the limit of registration on the negative side. D commenced to fall at 18h. 21m., when H began to rise, and in the course of 26 minutes fell 61' and rose 70'. During this time V oscillations, though numerous, were small. The form of the Antarctic curves suggests that the phenomena from 18h. 21m. to 19h. Om. were associated together, and the most natural inference seems to be that they form part of the disturbance which BIRKELAND regards as the first " intermediate " storm of his Section (iii). This was the time, it may be added, of one of the most prominent movements seen at BIRKELAND'S co-operating stations. The next movements in the Antarctic worth mentioning were fairly synchronous with BIRKELAND'S second "intermediate" storm (20h. 45m. to 21h. 20m.) They are somewhat imperfectly shown, owing to lack of trace. Between 20h. 40m. and 21h. 3m. H fell 45y and rose 50y (possibly more, as the trace is very faint and part may be invisible). The D trace, which had gradually got off the sheet on the positive side, suddenly re-appeared at 20h. 51m. and fell 58' in 6 minutes. After a minor oscillation it began to rise rapidly at 21h. 6m., rising 55' before it again went off the sheet at 21h. 15m. During this time there were some very rapid movements in V; the total range between 20h. 46m. and 21h. 3m. was 22y. There was no Antarctic trace from 23h. 20m. to 23h. 45m. on the llth. Between 23h. 48m. and Oh. 31m. on the 12th a bay appeared in the V trace, the maximum depression being about 12y. There is a synchronous bay in the D curve, a fall of 43' being followed by a rise of 33'. The H trace was off the sheet practically all the time. The above D and V movements occur simultaneously with BIRKELAND'S third " intermediate " storm. They are by no means of an outstanding character and do not appeal much to the eye. They are followed, however, by a relatively quiet time. COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. 255 12. October 23-24, 1902 (hours 17-5, Plate III). This is described, p. 76, as "a positive equatorial perturbation. It commences suddenly at 19h. lira., simultaneously all over the Earth." It is added, however, "About 1 hours later, a polar storm . . . characteristic, simple, and well defined, appears around the Norwegian (i.e. Arctic) stations . . . especially distinct at Matotchkin Schar." The date was not in BIRKELAND'S list, so he obtained curves from only a few stations, including, however, Bombay and Dehra Dun. The sudden commencement is shown clearly at all the stations, including the Arctic ones. At Toronto, and at Axeloen in the D curve, the commencing movement appears distinctly double, the principal move- ment being preceded by a smaller movement in the opposite direction. At Kew there is only a suggestion of an initial fall in H, but nothing certain prior to a sudden rise, amounting to 17y in the course of 3 or 4 minutes. After the summit was reached at 19h. 14m., there was a gradual return to an undisturbed value at about 19h. 35m. Simultaneously with the change in H there was a very small change in D, westerly Declination rising and falling about 0' 7. In the Antarctic the curves had been unusually quiet for some hours when there suddenly began, at 19h. 10m., an exceedingly rapid rise in H. The movement is too rapid to be distinctly shown, but the trace seems to have gone off the sheet, remained off for 2 minutes, and returned to near its original position in 5 or 6 minutes. How much the oscillation exceeded that shown, + 40y, then - 35y, it is of course impossible to say. After slackening for a minute or two about 19h. 16m., H continued moving in the same direction as before, to a peak at about 19h. 22m. The fall in H since the curve came on the sheet was 77y. Simultaneously with the commencing movement in H there was an oscillation in D, a fall and rise each of 15' taking place in about 4 minutes ; this was followed by a second smaller oscillation. The V trace, which had been exceedingly quiet, showed also a marked commencing movement, consist- ing of a rise of 6y in 3 or 4 minutes to a sharp peak at about 19h. 14m., followed by a fall of 26y, occupying about 8 minutes. Halfway during the fall there was a nearly stationary position during about 2 minutes. The natural conclusion unquestionably is that these large sudden movements in the Antarctic correspond to the smaller commencing movements which appeared simultaneously elsewhere. The principal disturbance in the Arctic occurs between 21h. and 23h., the maximum coming about 22h. 20m., but the Axeloen curves appear considerably disturbed until 3 a.m. on the 24th. The non- polar curves in Plate III show only very trifling disturbances, the largest, between 21h. and 23h. on the 23rd, being only of the same order as the sudden commencement at 19h. llm. In the Antarctic the conditions remained highly disturbed from 19h. 10m. until the traces got on the clamp at about 23h. 6m. The D and H traces show incessant large oscillations. The largest movements recorded in H were a rise of 84y between 19h. 56m. and 20h. 13m., and a rise and fall each of 67y between 22h. 51m. and 23h. Om. The curve came on the sheet at 22h. 51m. and went off 9 minutes later. Coming on again immediately, H rose 44y in a few minutes, the trace coming on to the clamp and so being lost. In D there was a rise of 94' between 22h. 18m. and 22h. 25m., followed by a fall of 100' during the next 10 minutes. Between 22h. 51m. and 23h., synchronous with the large changes in H, there was a fall of 54' and a rise of 87', the latter immediately followed by a rapid fall of 135', the trace coming on to the clamp before the movement was completed. Amongst the larger V movements were a fall of 26y and a rise of 22y between 19h. 54m. and 20h. 26m., a fall of 20y between 22h. 26m. and 22h. 33m., and a rise of 51y between 22h. 33m. and 22h. 53m. Between 22h. 53m. and 23h. 6m., when the trace got on the clamp, V fell 24y, rose 37y, and fell 27y. The larger D and V movements, it will be noticed, occurred during the time when the Arctic stations were most disturbed; but the disturbance in the Antarctic was continuously large between 19h. 10m. and 23h. The conditions in the Antarctic had become distinctly quieter by Oh. llm. on the 24th, when the next sheet was put on, but might fairly be described as disturbed until 3h. 30m. There was a small bay from Oh. 17m. to Oh. 53m., the changes in the three elements being at least approximately simultaneous. D rose 25' and fell 51', just going off the sheet on the positive side; while V rose 16y and fell 13y. The H trace was off the sheet on the positive side for 10 minutes and the rise, 256 COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. 21y, and fall, 14y, actually shown were probably a good deal exceeded. After Oh. 53m. the oscillatory movements in the V curve were much reduced. The D trace, however, showed two moderate bays. During the first from Oh. 53m. to Ih. 16m. D rose 33' and fell 69'. During the second from Ih. 16m. to 3h. 23m. D rose 42' and fell 70'. Between Ih. 39m. and 2h. 17m. H fell and rose 39y. The trace went off the sheet at 2h. 17m. and did not reappear until 3h. 27m. It went off and came on steeply, so there may have been a considerable movement in the interval. 13. October 27-28, 1902 (hours 14-1, Plate IV). This "compound" disturbance is divided by BIRKELAND, p. 209, into two sections. The first section, from 14h. to 20h. 30m., is regarded as composed of a long storm, largest on the whole in the Equator, during which there is an " intermediate " storm, most powerful in the Arctic, especially at AxelSen and Matotchkin Schar, which lasted from about 15h. 30m. to 16h. 45m. At Kew, which was fairly representative of non-Arctic Europe, the most prominent phenomenon of Section (i) was a bay in the D curve from about 15h. 30m. to 16h. 55m., the element being depressed. The greatest departure from the normal value was about 7' and occurred about 16h. 20m. The corre- sponding H movement was a fall of 22y from 15h. 15m. to 16h. 5m., interrupted by two small recoveries, and a rise of 19y from 16h. 5m. to 16h. 30m., followed by a smaller fall. BIRKELAND'S Section (ii), from 21h. 40m. to about midnight, consisted of a "polar" storm, largest at Axeloen. A table on p. 212 gives particulars of its beginning and end and also as to the time of occurrence and the value P! of the largest disturbance in the horizontal plane. The commencement is about 21h. 40m. and the hour of maximum about 22h. 50m. at most stations; the end varies from 23h. 20m. on the 27th to Oh. 20m. on the 28th. P! varies from 265y at Axeloen to 4y at Dehra Dun, the value at Kew, 29y, being slightly above the average for non- Arctic European stations. The general nature of the disturbance during Section (ii) outside the Arctic is fairly represented by the phenomena observed at Kew. D there was distinctly depressed from 21h. 40m. on the 27th until about Oh. 10m. on the 28th. The most rapid change was a fall of 3' between 22h. 20m. and 22h. 45m. H, on the other hand, was distinctly above the normal value from 21h. 40m. to 23h. 20m. The maximum occurred about 22h. 54m., and the most rapid change was a fall of 20y between that hour and 23h. 34m. The Antarctic curves during the time covered by Plate IV show a moderate amount of disturbance, but nothing, perhaps, that would naturally attract attention. During the "intermediate" storm of BIKKELAND'S Section (i) there was a bay on the D curve between 15h. 29m. and 16h. 53m., the element rising 47' and falling 33'. The V trace showed numerous oscillations, but the largest only 3y or 4y in amplitude. The H trace was beyond the limit of registration in the negative direction from 15h. 43m. to 16h. 53m., and may of course have been considerably disturbed. The largest movements recorded during Section (i) took place later, between 18h. 53m. and 19h. 53m. During this hour D rose and fell 66', H rose 99y, while V fell 14y and rose 35y. The D trace was off the sheet on the positive side for some time after 21h. and there was no trace from 22h. 16m. to 22h. 36m. There is thus rather a lack of information as to what was happening during BIRKELAND'S Section (ii). After 22h. 36m., when registration was resumed, there was no really striking D movement. There was, however, a small bay between 23h. 16m. and 23h. 56m., the value rising 18' and falling 30'. The H trace, which had just got off the sheet on the positive side at 22h. 46m., came on again at 22h. 47m., and between that hour and 23h. 33m. H fell 53y. After 23h. 33m. H rose gradually with minor oscillations until Oh. 53m. on the 28th, when the trace again went off the sheet on the positive side and remained off for two hours. V fell 32y between 22h. 36m. and 23h. 17m., and rose 40y between 23h. 17m. and 23h. 43m. After a slight halt it continued to rise, but much more slowly, until Oh. 17m. on the 28th. Between Oh. 17m. and Oh. 45m. it fell 19y. Thereafter the V trace was relatively, quiet for some hours. 14. October 28-29, 1902 (hours 14-1, Plate V). This was another " compound " storm, which resembled that of the previous day in containing two " inter- mediate " elementary polar storms. The interval between these was, however, much less than on the previous day, and there was, according to BIRKELAND, p. 222, in lower latitudes, " no trace on the 28th of the long storm that occurred on the 27th, and was especially powerful at the Equator." According to BIRKELAND'S COMPAEISON OF ARCTIC AND ANTARCTIC DISTURBANCES. 257 table, p. 222, the first " intermediate " storm commenced about 18h. 10m. and ended about 19h. 30m. at most European stations, the maximum coming about 18h. 45m. The value of PI the maximum horizontal disturbing force varied from 248y at Axelo'en to 3y at Toronto, the value at Kew, 16y, being slightly under the average for non-polar European stations. The time of commencement of the second "intermediate" storm is given as about 21h. 30m. at most stations, but somewhat earlier in the Arctic. The end is given as usually somewhat after 23h. The maximum is said to have occurred about 22h. 10m., the value of Pj varying from 266y at Axeloen to 2 5y at Batavia. The value given for Kew, 16'5y, is again slightly below the mean for non- Arctic Europe. The disturbances outside the Arctic were really very trifling. The most notable change at Kew was a rise of 20y in H between 21h. 30m. and 21h. 48m. In the Antarctic there was no loss of V-trace during the time covered by Plate V, except from 22h. 43m. to 23h. 13m., when there was no sheet on the drum, and the H trace was off the sheet only for a short time before the end. The D trace, however, was off the sheet a good deal between 20h. and 22h. 43m. The largest D movements recorded took place between 19h. 10m. and 19h. 58m., and so synchronise with or overlap the latter part of BIRKELAND'S first "intermediate" storm. During this time D rose 41', fell 28', rose 45'+ (going off the sheet), fell 52' + , and rose 50'. Some rather notable oscillations also occurred in H. The element rose 37y and fell 53y between 18h. 38m. and 19h. Om., the turning point (a maximum) being at 18h. 50m. and so practically simultaneous with the maximum in BIRKELAND'S first " intermediate " storm. Another considerable oscillation took place between 19h. 40m. and 20h. 10m., H first rising 55y and then falling 52y. The V trace showed numerous small oscillations. There was one rather sharp oscillation between 20h. 5m. and 20h. 16m., a fall of 19y being followed by a rise of 18y. The intervening minimum occurred at 20h. 9m. Between this hour and 21h. 20m. there was a total rise of 36y in V, which was followed during the next 20 minutes by a fall of 27y. The D trace went off the sheet rather steeply at 21h. 45m., and came on rather steeply at 22h. 25m., so there may have been a considerable oscillation in this element during the time of BIRKELAND'S second " intermediate " storm. The H and V traces, however, after 21h. 40m. were quieter than they had been for some hours previously. Thus whilst there was decidedly more than the average amount of disturbance in the Antarctic during BIRKELAND'S first "intermediate" storm, it is at least doubtful whether the same was true of the second " intermediate " storm. 15. October 29-30, 1902 (hours 16-4, Plate VI). This "compound" perturbation is said, p. 161, to consist of an "equatorial" perturbation which commenced suddenly on the 29th at 16h. 52m., and whose most active phase in the southern stations appeared at about Ih. 30m. on the 30th and of " polar " storms. Whether BIRKELAND supposed the same equatorial perturbation to last continuously all the time is not clear. As to the nature of the coincidence of the equatorial and polar disturbances, p. 161 says: "The positive equatorial perturbations observed by us are always accompanied by polar storms. As a rule, the polar storms do not begin until a little while after the equatorial ; but on this occasion they begin almost simultaneously. . . ." In discussing his Chart I, which includes results for hours 18h. 52 -5m. and 20h. 30m. on the 29th, BIRKELAND concludes, p. 164, that " it is the polar systems that give the field its character," and he puts the " centre " of the polar system near Matotchkin Schar. In discussing Chart II for Ih. Om. on the 30th, he regards the field as " now mainly conditioned by the equatorial perturbation." During the major part of the polar storm BIRKELAND had records from only two polar stations, Axeloen and Matotchkin Schar. The largest movements shown at either occur between 18h. and 21h. At the non-polar stations there was a distinct sudden commencement not clearly apparent at the polar stations whose time of occurrence BIRKELAND puts at 16h. 52m. The original Kew H trace shows a small fall, about ly, between 16h. 52m. and 16h. 54m., followed by a rise of about 4y during the next 4 minutes. The most noteworthy movement at Kew and the other non-polar stations took place between 1 and. 2 L 258 COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. 2 a.m. on the 30th, the maximum displacement being that already referred to as occurring about Ih. 30m. The disturbance at this hour appears to have been a good deal larger at Dehra Dun, Batavia, and Christ- church than at the non-polar European stations, BIRKELAND'S estimate of the horizontal disturbing force being 43y at Batavia and 40y at Christchurch, as against 14y at Kew. In the Antarctic there was an outstandingly rapid rise of H from about 16h. 55m. to 16h. 59m., followed by an equally rapid and larger fall. The trace was very faint near the time of the turning-point, and got beyond the range of registration at 17h. 5m., so that all one can be sure of is that between 16h. 55m. and 17h. 5m. H rose at least 27y, and fell at least 37y. A slow rise in H commenced about 16h. 53m., but this was checked for a few seconds at 16h. 55m., and the movement did not attain its highest rapidity until perhaps 16h. 56m. This H movement occurred at a time when the trace had been rather quieter than usual for an hour or more, and there can be little doubt that it represents the sudden commencement seen at BIRKELAND'S non-polar stations. In the Antarctic, synchronous apparently with the commencing movement in H, there was a sharp oscillation in V, a rise of lOy and fall of 6y taking place in about 6 minutes. D, which had been rising on the whole fairly steadily with minor oscillations, began a more decisive though not conspicuously rapid rise about 16h. 56m. During the next 27 minutes it rose 73' and the trace then got off the sheet. It came on the sheet 3 minutes later, but, after being on for about 18 minutes, got off once more, and thereafter was seen only at rare and short intervals during the remainder of the time covered by Plate VI. The H trace remained beyond the limits of registration until about 19h. 18m., and was again lost sight of about 19h. 45m. The light in the Antarctic magnetographs evidently became very faint towards the end of the sheet, as even the base lines are but faintly indicated after 21h. The V trace, which suffered less from weak illumination than the D and H traces, had become invisible by this hour. It is thus possible that faintness of light may have been partly responsible for the non-appearance of the D and H traces after 20h. The persistence, however, of active disturbances until the time when the V trace became invisible may be safely inferred from the following list of observed changes in V : From 17h. 6m. to 17h. 48m. rise 57y, 17h. 48m. 17h. 54m. fall 36y, 17h. 54m. 18h. 5m. rise 33y, 18h. 57m. 19h. 8m. rise 38y, 19h. 8m. 19h. 30m. fall 58y, 19h. 30m. 19h. 45m. rise 33y, 19h. 45m. 20h. 1m. fall 36y, 20h. 1m. 20h. 8m. rise 30y, 20h. 8m. 20h. 23m. fall 35y. The major part of these V disturbances synchronise with BIRKELAND'S " polar " storm, but some precede it. Shortly after the last movement recorded above the trace became invisible. After the next sheet was put on at 23h. 20m., distinctly disturbed conditions existed until after 3 a.m. on the 30th. The changes in V were especially noteworthy. Between Oh. Om. and Ih. 4m. on the 30th there was a rise of 92y, between Ih. 4m. and Ih. 46m. a fall of 74y, and between Ih. 46m. and 2h. 46m. a rise of 93y. H rose 52y between Oh. Om. and Oh. 40m., the trace then going off the sheet on the positive side. After being off for 9 minutes it came on, but 20 minutes later it went off again on the positive side. Coming on once more at Ih. 36m., it showed a fall of 59y between Ih. 36m. and 2h. 17m., and a rise of 59y between 2h. 17m. and 2h. 35m. The trace went off the sheet at the latter hour, and except for a short appearance of about 10 minutes re-appeared no more until after 8h. During the above changes in V and H the chief movements in D were a fall of 129' between Ih. 10m. and Ih. 43m., and a rise interrupted for 25 minutes by minor oscillations of 87' between Ih. 43m. and 2h. 45m. After 3h. 30m. the D and V traces were specially quiet during the next 5 hours. The H trace being off the sheet, one cannot be certain that it was equally quiet. It will doubtless have been noticed that the largest movements recorded in the Antarctic occurred COMPAKISON OF ARCTIC AND ANTARCTIC DISTURBANCES. 259 during the time when BIRKELAND'S "polar" storm was largest, and during the time of his largest "equatorial" disturbance. There is, however, no sign of intermission in the disturbance, though the evidence is not complete owing to failure of the trace. Later in the 30th, it may be added, after the time covered by Plate VI, there was a very prominent bay in the Antarctic D curve, extending from about 8h. 20m. to 9h. 33m. It included a fall of 142' and rise of 115'. This disturbance was at least an approach to the "special type," V oscillating about a mean position during the fall of D, and rising about 30y during the rise of D. There was a synchronous bay on the H curve, but details are lacking, as the trace was off the sheet most of the time. 16. October 31 to November 1, 1902 (hours 6-2, Plate VII). Of this "compound" storm BIRKELAND says, p. 230, "It appears at the poles with tremendous violence, although perhaps its strength is even more unusual at the equatorial stations. Considering its long duration and its universal distribution, we may say that it is the greatest storm that has been observed by us." He regards the disturbance as consisting of a long storm lasting from about 9h. on the 31st to 3h. on Nov. 1, with two, if not more, " intermediate " storms. Referring to his first eight charts, which answer to times from 9h. Om. to 12h. 30m. on October 31, he says, p. 232, that the equatorial stations show " powerful perturbing forces directed southwards," the forces at Dehra Dun and Batavia being almost double those in central and southern Europe. During this time BIRKELAND'S Arctic stations showed no very large disturbances, but Sitka was highly disturbed. Charts IX, X, and XI, for 13h. 30m., 13h. 42m., and 14h. Om., represent the conditions during the "first powerful intermediate storm," whose maximum is put at 13h. 42m. This includes the time of largest movements at the polar and equatorial stations. There are also movements at all the non-polar European stations, but these are on the whole smaller than the movements later in the day. After 14h. Om. conditions were everywhere less disturbed for some hours. But from 17h. 45m. to Ih. Om. on Nov. 1 there were further large disturbances in the Arctic and the European stations, which are dealt with in BIRKELAND'S Charts XII to XIX. The second " intermediate " storm is regarded as extending from 23h. 12m. to Oh. 42m. on Nov. 1, with maximum about 23h. 45m., and after its conclusion the conditions became much quieter. As usual, Kew seems to be fairly representative of non-polar European stations. It is very difficult there to assign even an approximate time for the commencement of the disturbance. One has to go back to 20h. on the 30th to get a time really free from the small undulatory movements which represent the disturbance up to noon on the 31st. The end of the disturbance between 3 and 4 a.m. on Nov. 1 is more definite. The Kew D curve shows two slow wave-like movements in immediate sequence, extending from 7h. 30m. to lOh. Om. on the 31st, the rise and fall in each being from 1' to 2'. From llh. 40m. to 14h. 30m. there was another group of movements of a more irregular character, which included a fall of 4' and rise of 3' between 13h. 20m. and 13h. 50m. This corresponds to BIRKELAND'S first "inter- mediate " storm. From 17h. on the 31st to 2h. on Nov. 1 there was considerably more disturbance at Kew than earlier. Between 17h. Om. and 17h. 48m. D fell and rose 3', reaching a sharp peak at the latter hour. After 17h. 48m. D continued to fall generally, with minor oscillations, until 22h. 10m., the fall in this time amounting to 13'. D then rose 4' - 3 in two steps to a rounded peak at 23h. 10m. Between this hour and Oh. 45m. on Nov. 1 it fell 6' and rose 7' ; the turning-point, which was the minimum during the disturbance, was at about 23h. 42m. In the Kew H curve the most rapid changes were a fall of 26y and rise of 23y between 13h. 15m. and 13h. 42m. corresponding to BIRKELAND'S first "intermediate" storm and a fall of 25y between 17h. 45m. and 17h. 53m. There was a comparatively quiet time from 14h. 40m. to I7h. 45m. After the latter hour there was no cessation of disturbance until about 2h. on Nov. 1. In the Antarctic a highly disturbed state of matters existed from about 8h. 50m. to 14h. Om. on the 31st. The phenomena resembled four disturbances of the " special type," following one after the other without any interlude ; but D and V were not quite in phase, and most of the turning-points on the H trace were beyond the limits of registration, so one can only see that this element was approximately in phase with V. 2 L 2 260 COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. We may regard the V movements during this time as composed of four " waves," whose times and amplitudes were as follows (+ denotes a rise, - a fall) : Wave From To Change (in 7). 1 h. m. r ? t 8 51 h. m. 8 51 9 23 ? + 26 2 / 9 23 \10 13 10 13 10 59 -16 + 43 3 flO 59 til GO 11 50 12 30 -29 + 21 4 /12 30 \13 33 13 33 14 1 -34 + 49 Each " wave " except the third left V enhanced, so that the final value exceeded the original by 60y. There were four " waves " in D roughly corresponding to those in V. During the first D fell 75' and rose 72'. second,, 102' 133'. third 81' 195'. fourth 173' 80'. H fell 108y between 9h. 32m. and lOh. 23m., rose 56y between lOh. 23m. and llh. 5m., and fell 57y between llh. 5m. and llh. 30m. The trace went off the sheet on the negative side at llh. 30m., remaining off until 12h. 46m. It came on again, but only for a few minutes, showing a double peak at 12h. 50m., and was thereafter no more seen until 20h. 53m. The last of the four waves above mentioned in the Antarctic synchronises with BIRKELAND'S first " intermediate " polar storm. But it seems impossible to draw any line of demarcation, such as BIRKELAND draws, between it and what precedes. The four waves are of the same type and follow in immediate sequence, and it is difficult to believe that the first three can be due to " equatorial " perturbation if the last is due to " polar." The storm in the Antarctic probably existed for some hours before 8h. 51m., as the D trace contains two bays similar to, though smaller than, the four we have described. The H trace, however, was off the sheet after Ih. 20m., and the V magnetograph not working until nearly 8h. 51m., so our information is very limited before this hour. After 14h. 1m. there was probably an absence of large rapid movements in the Antarctic until nearly 21h. The H trace was off the sheet, and the D trace also after 17h. 40m., but the V trace remained on until the sheet was taken off at 22h. 7m., and its course was unbroken by any oscillations at all comparable with the four we have described. There was, however, a persistent fall in V, as if the cumulative effect of the four waves were gradually disappearing. The total fall between 14h. 1m. and 20h. 41m. amounted to fully 70y. After 20h. 50m. there were some more wave-like movements in the V trace, though not so large as the earlier ones. The two principal consisted, the first of a fall of 21y and rise of 12y between 20h. 50m. and 21h. 28m., the second of a fall of 30y and rise of 16y between 21h. 28m. and 22h. 9m. The H trace, after having been beyond the limits of registration on the negative side for nearly 8 hours, came on the sheet at 20h. 58m., and H rose 89y between that hour and 21h. 30m. Rising further, with minor oscillations, it got off the sheet on the positive side at 22h. 2m., having risen llOy since 20h. 58m. There was no record from 22h. 7m. to 23h. 46m., which covers half the time of BIRKELAND'S second " intermediate " storm. When the next sheet was put on at 23h. 46m., the H trace was off the sheet on the positive side, and it remained off until 4h. 30m. on November 1. The D and V traces on the second sheet show, however, no signs of special disturbance. The largest movements in the V trace are two bays, the first, from Ih. 33m. to 2h. 45m., having a rise and fall of only 16y, the second, from 2h. 45m. to 3h. 56m., showing a rise of 43y and fall of 27y. COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. 261 17. November 23-24, 1902 (hours 15-7, Plate VIII). This " compound " storm, divided by BIRKELAND into three sections, is said, p. 272, to be the most powerful of a series which developed daily in the Arctic from the 19th to the 26th of November. Section (i), from 15h. 20m. to about 16h. Om., is described, p. 273, as a "typical positive equatorial storm," " strongest at the equatorial stations in the South of Asia." BIRKELAND speaks of a sudden rise occurring in H at 15h. 30m., except at the American stations, where the rise is slow. The time is not, I think, intended to be exact. For at Kew H begins to rise at about 15h. 22m., rising lOy to a rounded maximum at about 15h. 32m., and then diminishing about 67 during the next 7 minutes. There was a corresponding small movement in D, whose maximum was, however, a few minutes later. These trifling movements which can, however, be recognised at Axeloen, Kaafjord, and all the European and Southern stations constitute the largest disturbance of Section (i). During Section (ii), from 16h. to about 22h., the disturbing forces are generally small, but from 17h. 30m. to 18h. 20m. the disturbance is somewhat greater, " especially in southern latitudes." In the Arctic BIRKELAND sees indications of a " polar " storm of minor intensity. During this time the most noteworthy feature at the non-polar European stations is a bay on the D curve, extending from about 17h. Om. to 18h. 30m. at Kew, where the element fell and rose about 3'. The most prominent H changes at Kew were a fall of 12y between I7h. 23m. and 17h. 35m., and a fall of lly between 19h. 52m. and 20h. 3m. BIRKELAND'S discussion of his Section (iii), 22h. to 7h., appears involved. He attempted apparently to recognise a series of " elementary polar " storms in it, but was unable to disentangle them. Speaking generally, the Arctic curves show numerous rapid oscillatory movements going on pretty continuously all the time. At the non-polar stations the most disturbed time was from 22h. on the 23rd to 3h. on the 24th. The movements during this time were really of some size. At Kew, for instance, D fell 10' between 22h. 5m. and 22h. 42m., rose 5' between Oh. 30m. and Oh. 50m., fell 6' -8 between Oh. 50m. and Ih. 20m., rose 10' -4 between Ih. 20m. and 2h. 10m., fell 9' between 2h. 10m. and 2h. 43m., and rose 7' between 2h. 43m. and 3h. 12m. H fell 37y in three steps between 21h. 48m. and 23h. 1m., rose 35y in two steps between 23h. 1m. and 23h. 42m., fell 29y in two steps between 23h. 42m. on the 23rd and Ih. 35m. on the 24th, rose 42y between 2h. 2m. and 2h. 24m., and fell 30y between 2h. 48m. and 3h. 52m. The conditions existing at the southern stations during the early morning of the 24th will be best realised by consulting the Christchurch H curve given in Plate XV of the present volume. In the Antarctic there were some very rapid oscillatory movements synchronous with the commencing movements seen at Kew and elsewhere which BIRKELAND characterised as " equatorial." The D move- ments were so rapid that the turning-points are not very clearly shown, and the following measurements may be slightly under-estimates. Commencing suddenly, during a fairly quiet time, we have between 15h. 21m. and 15h. 37m. a rise of 81', a fall of 78', a second rise of 60' and a second fall of 85', with minor oscillations during the larger movements. There were presumably corresponding movements in H, as the curve which had been off the sheet for some time on the negative side came on for a couple of minutes about loh. 24m., forming a very sharp peak (maximum). During the second oscillatory movement in D there was a marked depression and recovery in V, of amplitude 44y. After this commencement the Antarctic curves remained highly disturbed until lOh. on the 24th. They are reproduced from Oh. 18m. to lOh. 31m. (or 11.25 a.m. to 9.38 p.m., L.T.) in Plate XXV of the present volume. The most conspicuous movements prior to the time covered by this plate were as follows : D after falling 226', with numerous oscillations, between 15h. 28m. and 16h. 25m., rose again, oscillating vigorously, about 197' between 16h. 25m. and 18h. 8m., and then fell 127' between 18h. 8m. and 18h. 58m. A specially rapid movement about 19h. 19m. is referred to presently in conjunction with a corresponding H change. Between 19h. 19m. and 21h. 3m. D rose, oscillating largely, 208', and the trace went off the sheet on the positive side for a few minutes at the latter hour. After coming on the sheet, the D trace in the course of about half an hour fell 97', rose 72', fell 75', and rose 100', going off the sheet again about 21h. 40m. Its reappearance a few minutes later is not very clearly shown, and at 22h. 8m. it disappeared, to be seen no more before the sheet was removed at 23h. 13m. 262 COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. The movements just recorded extend over BIRKELAND'S Sections (i) and (ii) and show no interlude. Until 21h. the H trace was mostly off the sheet on the negative side, appearing at intervals for a few minutes at a time. The most striking movement seen during this time answered to a peak at about 19h. 19m. The trace was visible for about 8 minutes, during which H rose and fell 687. This happened synchronously with an exceedingly rapid oscillation in D, that element in the course of 1 1 minutes falling 97' and rising 88'. Later the H trace was more in evidence. Between 20h. 53m. and 21h. 8m. H rose 66y and fell 49y with minor oscillations. Between 21h. 8m. and 21h. 20m. it rose 140y, the trace then going off the sheet on the positive side and remaining off for 20 minutes. On reappearing at 21h. 40m. the H trace fell 136y in 7 minutes, then rose 54y and fell 70y in the course of the next 11 minutes, and went off the sheet on the negative side about 21h. 58m. It was not again seen, except for a few minutes, until about 22h. 37m., when it appeared and rose 103y during the next 10 minutes. There were numerous minor oscillations all this time in the V trace, the most striking being a fall of 65y and rise of 50y between 19h. 17m. and 19h. 28m., a time during which there were also rapid oscillations in D and H. But perhaps the most outstanding feature is the continuous downward tendency of the curve from 16h. 15m. to 21h. 40m., the total fall during this time being about 300y. After some small oscillations V then commenced to rise rapidly, the rise between 21h. 57m. and 22h. 42m. representing 182y. From 23h. 13m. on the 23rd to Oh. 19m. on the 24th there was no paper on the drum, and though substantial alterations took place in the values of the elements during this interval it may have been quieter than the times before or after. But there is no direct evidence of any marked lull from the commencement, about 15h. 20m. on the 23rd, until lOh. on the 24th. No subdivision into sections nor difference in source is at all suggested by the Antarctic D and H curves. The case of V is rather different, as that element, on the whole, rose between 15h. 35m. and 16h. 15m., and again between 21h. 40m. and 22h. 42m., while it fell during the intervening time. In the Antarctic there can be no question that the disturbance of October 31 to November 1 was smaller than that of November 23-24. The range of the elements appears greater at Kew on the former occasion, but that is due to the fact that the period was longer and included the ordinary hours of the daily maximum and minimum. The appearance of the Kew curves would indicate that the later disturbance was the more intense, and this is really borne out by BIRKELAND'S figures on pp. 240 and 280 for the amplitude of the disturbing force. 18. December 9, 1902 (hours 5-18, Plate IX). This is entered amongst the " equatorial " storms. When discussing it on p. 70 BIRKELAND describes the disturbance as illustrating at its commencement " all the properties that characterise the positive equatorial perturbations. It commences quite suddenly, simultaneously all over the Earth, at 5h. 40' 6m." This movement is seen in the Arctic as well as the non-polar regions. It is, however, by no means large at most stations. At Kew, for instance, the initial rise in H was only about 5y and it was not very rapid. BIRKELAND regards "equatorial" conditions as persisting until nearly 15h. During this time there are no large movements even in the Arctic. "Between 15h. and 18h., the character of the perturbation conditions is essentially changed. It is this feature that we continually find repeated, namely, that when the equatorial storm has lasted for some hours, polar systems appear." p. 70. During the time stated there was a disturbance of some size at all the Arctic stations, and even at the non-polar stations there were appreciable bays on both the H and D curves. Thus at Kew the H trace shows a bay from 16h. 25m. to 17h. 40m., the greatest depression representing about 15y; corresponding to this was a hump on the D curve, the maximum representing a rise of about 2' - 5. In the Antarctic there were of course numerous movements larger than any at Kew, but there is no very decided trace of disturbance until towards the end of the period covered by Plate IX. On this occasion it is very doubtful whether there is anything in the Antarctic curves corresponding to the sudden movement seen elsewhere about 5h. 41m. There is, it is true, apparently at this exact time a trifling but sharp peak in the V trace which previously was very quiet representing a very rapid rise and fall of COMPARISON OF AECTIC AND ANTARCTIC DISTURBANCES. 263 ly or 2y, and this is followed by a rise of about 7y in the course of the next 6 or 7 minutes. There were peaks on the D curve at 5h. 41m. and 5h. 45m. Between these two times D rose T, and during the next 9 minutes it fell 14'. There was also a peak on the H trace at about 5h. 41m., the next turning- point being about 5h. 51m. ; in the interval H rose 20y. These movements in the D and H curves are, however, not conspicuously different either in size or rapidity from a good many others, and the apparent coincidence in time may be accidental. One would hardly describe the Antarctic curves as disturbed until after 16h. From then until 23h. the disturbance was continuous and very considerable. During the time covered by Plate IX the largest movements recorded in D were a fall of 156' between 16h. 44m. and 17h. 13m., followed by a rise of 213' between 17h. 13m. and 17h. 33m. The II trace was mostly off the sheet after 15h. V, however, showed gome considerable movements, rising 79y between 17h. 10m. and 17h. 35m., falling 62y between 17h. 37m. and 17h. 46m., then rising 38y between 17h. 46m. and 17h. 57m., ajid falling 80y between 17h. 57m. and 18h. 17m. This last movement, however, extends beyond BIRKELAND'S period. The Antarctic movements after 16h. 47m. are shown in the right-hand figure of our Plate XL. In it 4 a.m., December 10, answers to 16h. 53m., G.M.T., on the 9th. This figure shows the Antarctic disturbance for more than an hour subsequent to the time covered by BIRKELAND'S Plate IX. The movements which it shows, though large, are if anything inferior in size to some recorded between hours 20 and 22, G.M.T., i.e. about two hours later. The disturbed conditions continued until nearly 23h. One might thus be inclined to infer that the Antarctic storm, while so far synchronous with that observed in the Arctic and elsewhere after 16h., continued long after the disturbance elsewhere had ceased. The D and H Kew curves show, however, between 19h. and 22h., some movements which, though less than those between 16h. and 18h., are larger than those occurring between 5h. and 7h. It would be of interest to know what was happening in the Arctic between ISh. and 22h. 19. December 14-15, 1902 (hours 23-5, Chart X). This " elementary polar " storm is described by BIRKELAND, p. 87, as appearing " upon an otherwise very calm day .... without any preceding equatorial perturbation," and as consisting of " a great storm in the north, about Dyrafjord and AxelSen .... accompanied by a perturbation, small indeed, but well defined, .... observed in Northern America and Europe." The effect is described as "just perceptible " at Dehra Dun, but not visible at Batavia. At Dyrafjord, where the movement was largest, the storm is said to have lasted from Oh. 10m. to 3h. 15m., the maximum value of the disturbing force, 386y, being met with about Ih. 8m. At Axeloen, where the maximum disturbing force was about half that at Dyrafjord, the times were somewhat later, the maximum not appearing until Ih. 46m. In temperate Europe the disturbance is said to begin "rather suddenly at Oh. 45m." and to last about 3 hours. BIRKELAND adds, " This perturbation .... has its origin in the northern regions. Its sphere of action .... is concentrated about the neighbourhood of Dyrafjord and Axeloen. The shortness of its duration, as also the comparatively calm character of the curves .... seems to indicate that this is a polar elementary storm of the most typical nature ; it appears to be produced by a coherent impulse, which increases to a certain size, and then again decreases to .... At the same time, as the perturbation does not make its appearance at all places simultaneously, the perturbing cause must be supposed to move with a somewhat continuous motion," p. 87. This remark has been quoted at length, because in several respects it is so suggestive of the Antarctic " special type " of disturbance, the principal difference being that the value of V in the Antarctic usually remained elevated for some time after the apparent end of the disturbance in D and H. In Europe, as BIRKELAND says, the disturbance was small outside the Arctic. At Kew, for instance, H rose about lOy between Oh. 45m. and Ih. 5m. and then fell very gradually until about 2h. 40m., the total fall being about 15y. D rose about 2'- 9 between Oh. 45m. and Ih. 5m. and then fell about 3' 5 to a badly defined minimum about 2h. 5m. In the Antarctic there were some rather striking movements about three hours before the earliest time on Plate X, and one would put the commencement of the disturbances there at about 18h. 30m. on the 14th. There were, however, very sudden movements commencing about 23h. 5m. in both D and H, consisting of a rise and fall occupying in all some six minutes. H rose 22y and fell 34y. The D oscillation 264 COMPAKISON OF ARCTIC AND ANTARCTIC DISTURBANCES. was apparently larger, but owing to its rapidity the trace is too faint to follow to the turning-point. Simultaneously there was a very sudden rise of 20y in V, which was preceded and followed by slower movements in the opposite direction. On examining Plate X one observes a small sudden movement at least approximately synchronous with these Antarctic movements at Matotchkin Schar, Dyrafjord, and the American stations, but it cannot be identified with certainty in the Kew curves. There was no record obtained in the Antarctic from 23h. 56m. on the 14th until Oh. 19m. on the 15th. Alter the latter hour, however, there was a deep bay on the D curve, the element rising 99' between Oh. 33m. and Ih. lira., and falling 106' between Ih. llm. and 2h. 12m. H rose 48y between Oh. 53m. and Ih. 23m. ; the trace then went off the sheet on the positive side, remaining off until 2h. 33m. After coming on for a few minutes, it was again off until 3h. V rose 65y between Oh. 47m. and Ih. 23m., and then fell 36y between Ih. 23m. and 2h. 8m. These movements, it will be observed, occur about the time of the principal movements in BIRKELAND'S Arctic stations. On their conclusion the Antarctic curves were relatively quiet during the next 12 hours. 20. December 24-25, 1902 (hours 23-5, Plate XI). This disturbance is included amongst the " compound." In temperate Europe BIRKELAND says, p. 165, " the conditions are slightly disturbed from 23h. on the 24th to 5h. on the 25th. There are especially distinct perturbations about midnight, and from 2h. 30m. to 4h." In Toronto and at Baldwin and Cheltenham, U.S., the perturbation is practically confined, BIRKELAND says, to the short interval 3h. 14m. to 3h. 57m. on the 25th, with maximum at 3h. 21m. At Dehra Dun, Batavia, and Christchurch 3h. to 4h. is also decidedly the most disturbed time. The Arctic stations are as much disturbed between 23h. on the 24th and Oh. 15m. on the 25th as they were later, while temperate European stations are also sensibly disturbed at the earlier hour. BIRKELAND concludes, p. 165, that in Europe, as a whole, the conditions " are in the main connected with the polar storms at the Norwegian (i.e. Arctic) stations." The disturbance at the non-polar European and Asiatic stations was really very trifling. At Kew the only changes in D at all conspicuous were a rise of l'-2 between 23h. 5m. and 23h. 18m., followed by a fall of 1''6 between 23h. 24m. and 23h. 35m. on the 24th, and a rise of 2'- 6 between 3h. 14m. and 3h. 21m., followed by a fall of 2' ending about 4h. 5m. on the 25th. In H there was a rise of 9y between 3h. 34m. and 3h. 55m. on the 25th. The other changes hardly catch the eye. In the Antarctic it was rather quieter than usual for 2 or 3 hours prior to 23h. on the 24th. There then ensued a decidedly more disturbed time, extending from about 23h. 15m. on the 24th to Oh. 13m. on the 25th. Between 23h. 15m. and 23h. 54m. D rose 28', fell 84' and rose 57'. H, during this timo, had a total range of only 45y, but there was rather a prominent double oscillation composed of a rise of 38y in 4 minutes, a fall of 45y in 6 minutes, a rise of 42y in 9 minutes, and a fall of 24y in 8 minutes. Between 23h. 19m. and 23h. 56m. V fell 40y and rose 53y. Owing to the trace coming on the clamp, there was no Antarctic record from Oh. 18m. until the new sheet was put on at Oh. 55m. on the 25th. Conditions were distinctly quiet for over an hour after this. The H trace was off the sheet on the positive side from 2h. 23m. to 4h. 48m., so there may have been high values in that element. Between 2h. 15m. and 4h. 19m. D fell gradually 130', but a very appreciable fraction of this must be ascribed to the regular diurnal variation. The fall was interrupted as usual by a good many minor oscillations, the largest retrograde movement being a rise of 26' between 3h. 12m. and 3h. 21m. The V trace showed no large oscillations, but there were a number of minor oscillations between 3h. 20m. and 4h. 35m. The Antarctic movements are synchronous with and larger than those seen at the American and non-polar European stations. The reader should, however, be warned that one would not naturally regard any portion of the time covered by Plate XI as more than usually disturbed in the Antarctic, with the exception of the last hour of the 24th. At the same time, the end of Plate XI answers to the early afternoon in the Antarctic, and the diurnal changes at Midsummer were then so rapid that irregular disturbances appeal less to the eye than at most hours of the day. 21. December 26-27, 1902 (hours 18-2, Plate XII). This is included amongst the "polar elementary" storms. From the discussion on p. 137 we learn that it comprised two distinct " elementary" storms, the first especially powerful at Matotchkin Schar, having COMPAKISON OF AKCTIC AND ANTARCTIC DISTURBANCES. 265 a maximum at about 20h. 45m. to 21h., the second especially powerful at Dyrafjord, being most developed between 22h. 30m. and 24h. There is also a somewhat vague reference to a " more lengthy perturbation " as covering the time of the two polar storms, and to the possibility of " cyclo-median " perturbations being felt in lower latitudes. At the time of the first " polar elementary " storm the most prominent feature at Kew was a bay on the D curve. The element fell 4' '2 between 20h. 36m. and 20h. 46m., and then rose gradually to about its original value at about 21h. 45m., the rate of recovery slackening after 21h. 10m. H rose about 7y between 20h. 45m. and 20h. 57m., having been falling slowly for some time previously. At the time of the second " polar elementary " storm the most prominent feature at Kew was a bay on the H curve, the value of the element rising and falling about 14y between 23h. 8m. and 23h. 48m. In D there was a rise of about l' - 5 between 23h. Om. and 23h. 15m., followed by an equal fall and a further small rise. After midnight on the 26th one would ordinarily describe the curves as very quiet. The conditions at Kew appear fairly representative of the European non-polar stations. The Antarctic curves were unmistakably disturbed after 19h. on the 26th until the sheet was removed at 23h. 31m. The disturbance commenced apparently with a very rapid rise in D, the value increasing 141' in 6 minutes from 18h. 59m. to 19h. 5m. After some small rapid oscillations D continued to rise until 19h. 15m., the total rise since 18h. 59m. being 158'. This was followed by a fall of 196' between 19h. 15m. and 19h. 53m., and a second rise of 224' between 19h. 53m. and 20h. 55m., the trace then going off the sheet on the positive side for a few minutes. H commenced to fall at 18h. 59m., when D began to rise, and fell 79y in about 9 minutes, shortly thereafter getting beyond the limits of registration on the negative side. The largest change shown in H was between 20h. 27m. and 21h. 53m., when the element rose 160y in several steps interrupted by minor oscillations. The V trace during this time showed numerous oscillations, including the following ( + denotes a rise, - a fall) : From h. m. h. in. 7- 18 59 to 19 4 + 22, 19 4 19 16 -68, 19 16 19 24 + 40, 19 24 19 31 -65, 19 31 19 36 + 36, 19 36 19 46 -29, 20 7 20 33 + 47, 20 33 20 48 -33, 20 57 21 13 -49. After 21h. 45m. there were further considerable movements in the Antarctic, the most notable being on two occasions when the traces from the three elements show deep bays, occurring synchronously or very nearly so. On the first occasion, lasting from about 21h. 55m. to 22h. 30m., D fell 136' and rose 130', H 89y 69y, V 120y 112y. The turning-points (minima) occurred within a few minutes of one another. On the second occasion the oscillation was apparently not quite completed when the paper was changed. So far as recorded, it lasted from about 22h. 56m. to 23h. 31m. During it D rose 33' and fell 39', H 52y 66y, V 23y 37y. The second movement is comparatively small, and is mentioned chiefly because the three elements appear approximately in phase and the time synchronises with that of BIRKELAND'S second " polar elementary " storm. 2 M 266 COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. For some time after 23h. 50m. on the 26th, when the next sheet was put on, there were some fair movements, notably some rapid nearly synchronous oscillations in the traces of the three elements between Oh. 20m. and Ih. Om. on the 27th. After this the conditions became distinctly quieter. The temperature trace was lacking or imperfectly visible during most of the time, and the V changes recorded for times prior to 21h. 30m. are not corrected for temperature. The oscillations were, however, so rapid that the uncertainty thus arising is small. 22. December 28, 1902 (hours 3-8, Plate XIII). This date was not given in Prof. BIRKELAND'S circular, and he had few data for it from non-polar regions except from America. He has classified the storm, p. 169, as "compound," though the phenomena at Dyrafjord, where it was largest, suggested a " polar elementary " storm, whose centre was originally somewhere to the south of Greenland, and which moved to the westward. The classification seems to have been partly determined by the fact that the time of the chief disturbance from 4h. 40m. to 6h. was an unusual one for the occurrence of " elementary polar " storms. The curves reproduced from the European co-operating stations include only D and H from San Fernando and D from Stonyhurst. The Kew D curves between 4h. 30m. and 6h. 30m. showed two wave- like movements; in each a rise of about l'*5 was followed by a more gradual fall, the crests coming at about 4h. 50m. and 6h. Om. The H curve also showed two waves, H falling about 6y between 4h. 30m. and 4h. 50m., and rising 7y between 4h. 50m. and 5h. 15m., then falling 5y between 5h. 15m. and 5h. 30m., and rising very slowly about 5y to a maximum near 6h. 20m. There was a somewhat larger disturbance at Kew about midnight on the 27th, but it precedes the time covered by Plate XIII. The movements lasted from about 21h. on the 27th to Oh. 30m. on the 28th. In the Antarctic the most notable phenomenon was a deep bay on the D trace, of the same character as appeared during the " special type" of disturbance. The element began to fall suddenly about 5h. 13m. and in 13 minutes fell 90', going off the sheet on the negative side and remaining off for 11 minutes. During the next 12 minutes it was off and on the sheet once or twice, and then continued to rise with minor oscillations. The rise was not so rapid as the fall, and owing to the minor oscillations it is difficult to assign a definite time for the conclusion. D had, however, returned to its original value by about 6h. 20m. H was off the sheet on the positive side from 3h. 35m. to 6h. 25m., and so far as that element was concerned the disturbance might have been of the " special type." The V trace, however, was not of that type. There was a rapid fall of 18y between 5h. 14m. and 5h. 19m., followed by a rise of 29y between 5h. 19m. and 5h. 37m. But during the rise there were numerous small oscillations which continued until 6h. 20m. The Antarctic curves, it may be added, were somewhat highly disturbed on the 27th, from 16h. 15m. until 23h. 55m., when the sheet was taken off. 23. January 26, 1903 (hours 7-15, Plate XIV). During this time the Antarctic magnetographs were not in action. 24. January 26-27, 1903 (hours 18-7, Plate XV). This " compound " storm appears in Europe outside the Arctic as " a long perturbation . . . lasting from about 18h. Om. on the 26th ... to 7h. Om. on the 27th . . . We have . . . three . . . sharply defined intermediate storms," p. 287. These " intermediate " storms are said to coincide, on the whole, in time with three storms recorded in the Arctic, especially powerful at Axeloen. The last, however, of these three " intermediate " storms attained its maximum about Oh. 35m. on the 27th, and the Antarctic records do not commence until Ih. 5m. By this time there was comparatively little disturbance except in the Arctic, and even there it was much reduced. In the Antarctic after Ih. 5m. there was a fall in progress in D until 4h. 8m., when the curve got off the sheet for a few minutes. A fall is what we should expect in the ordinary course of events, but the fall shown between the hours mentioned, 198', is notably in excess of the average. The fall in D was interrupted as usual by minor oscillations. The most conspicuous of the retrograde movements, one of 57', took place between 2h. 23m. and 2h. 46m. The D trace was off the sheet between 5h. 25m. and 6h. Om., and the form of the curve when going off and coming on is not inconsistent with the existence of a bay of considerable depth. COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. 2 67 On the H trace there were some fairly large oscillations, but only partly shown. H rose 56y between Ih. 13m. and Ih. 26m.; the trace was then off the sheet for 21 minutes on the positive side. Between Ih. 47m. and 2h. 31m. H fell and rose 53y, the trace again going off the sheet. Between 4h. 16m. and 5h. 13m. H rose 105y. The V trace was, on the whole, quiet. It contained, however, a small bay from about Ih. 13m. to Ih. 48m., V rising and falling about 19y. 25. February 8, 1903 (hours 8-12, Plate XVI; and hours 13-24, Plate XVII). Before discussing this disturbance BiRKEi/AND mentions, p. 187, that conditions in the Arctic had been on the whole very quiet since the end of November, 1902, until February 7, 1903. On that day, he says, a fairly powerful storm was experienced in the Arctic from 21h. 5m. on the 7th until about Ih. on the 28th. There was, I may add, also marked disturbance in the Antarctic at the same time, but it commenced earlier, about 19h. 30m. on the 7th. The disturbance of February 8 is classed amongst the " compound." It is regarded as divisible into three sections, the first of which covers the time to which Plate XVI refers. As this section is allowed a separate plate, and there is a gap of an hour between the times represented by the two plates, I shall treat the two parts separately. During his Section (i) BIRKELAND remarks, p. 187, that " The perturbation is particularly powerful at Sitka, and is (there) especially violent from 9h. to 9h. 35m." It continued to be considerable at Sitka until llh., but ran a rather irregular course there and at the Arctic stations. BIRKELAND adds, p. 188, " The simple conditions found between San Fernando in the west and Zi-ka-wei in the east, and between Kew in the north and Batavia in the south, form a strong contrast . . . the perturbation is throughout chiefly in H. It is well denned, and, as far as we can determine, commences everywhere simultaneously at about 8h. 35m. ... It terminates simultaneously at about lOh. 50m." The time of maximum at the co-operating stations outside North America is given as from lOh. Om. to lOh. 10m. BIRKELAND, p. 189, considers that the phenomena at the American stations suggest "a polar elementary storm, at first not very far north-east of Sitka," but this could not, he says, account for the phenomena "over the district between Kew and Batavia," which suggest a "negative equatorial storm," i.e. a disturbance due " to a current round the Earth from east to west ... at a distance from the Earth of at least a magnitude equal to the radius of the Earth." If I rightly follow him, the view he finally inclines to, p. 189, is "that at first the perturbation partakes most of the nature of a cyclo-median storm, and subsequently changes into a more purely polar one." At the non-polar European stations the phenomena were similar to those at Kew. There H fell gradually about 40y between 8h. 35m. and lOh. Om., and then rose 30y between lOh. Om. and lOh. 50m. For some reason which I do not understand, BIRKELAND, p. 188, regards this as the end of the storm at Kew, and elsewhere, during his Section (i). But as a matter of fact the Kew H curve began to fall again slightly immediately after lOh. 50m., and between llh. 20m. and llh. 50m. it rose 17y, falling 14y during the next 30 minutes. The trace was quieter for a short time after 12h. 20m. The D trace at Kew showed a rise of from 2' to 2' -5 above the normal between 9h. 30m. and lOh. 40m. D was, if anything, falling at lOh. 50m., the normal change at that hour being a rise. Decidedly the most conspicuous movement was a rise of 4' -8 between llh. 35m. and llh. 55m. There was a fall of 3' 5 between 12h. Om. and 12h. 40m. In the Antarctic the working of the V magnet appears doubtful. The H trace was off the sheet on the positive side, except during a few minutes, until 8h. 56m. Between 9h. 38m. and lOh. 39m. the trace shows a prominent to-and-fro movement a fall followed by a rise strongly suggestive of the special type of disturbance. The full extent of the movement is not shown, as the trace was beyond the limits of registration on the negative side for some 15 minutes, but the amplitude exceeded 68y. The turning- point, a minimum, must have occurred within a few minutes of lOh. Om. The D trace was off the sheet at this time on the negative side, having been off since about 8h. 10m., and did not appear until lOh. 53m. It went off and came on steeply, suggesting a deep bay or bays. Between lOh. 53m. and llh. 33m. D rose 82'. Between llh. 33m. and 12h. 25m. it executed a to-and- fro movement, somewhat suggestive of the special type of disturbance, a fall exceeding 82' being followed by 4 rise of over 105'. The trace was off the sheet only for a minute or two, the turning-point coming at 2 M 2 OF THE UNIVERSITY 268 COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. about lib. 50m. The H trace shows a bay synchronous with that on the D trace, H rising as D fell and conversely, but the H movements were relatively small, the rise being 17y, the fall 25y. It will be noticed that the large bay in the H curve occurs during the time of the first movements at Kew and elsewhere, while the bay in the D curve synchronises closely with the disturbance recorded at Kew near noon. BIRKELAND, p. 190, says that his Section (ii) extends from 14h. to 18h., but Plate XVII begins at 13h., and the disturbance is said to commence at more than one place at 13h. 45m. The conclusion drawn, p. 191, is that the perturbations represent " a series of short, principally polar impulses with somewhat changing centre." The disturbances in the Arctic are not very striking, but there is a considerable movement at Sitka, Kaafjord, and Matotchkin Schar with a maximum about 14h. 40m. This is also about the time when the departure from the normal is largest at the non-polar stations. At Kew H fell 17y between 14h. 28m. and 14h. 44m., and then returned gradually to the original value at about 15h. 50m. BIRKELAND'S Section (iii), 18h. to 23h., had for its principal feature in the Arctic a "violent storm . . . at all the . . . stations simultaneously, most powerful at Axeloen and Matotchkin Schar," p. 192. The time of commencement of this powerful storm is said to vary from 18h. 33m. at Dyrafjord to 19h. 7m. at Axeloen, the time of ending being about 22h. 30m. The intensity appeared to be greatest between 19h. 15m. and 20h. 15m. At the non-polar stations the disturbance was mainly from 19h. 5m. to 20h. 30m., the maximum being placed by BIRKELAND at 19h. 18m. At Kew the range between 19h. and 20h. was 48y in H and 14' -5 in D ; the most prominent movements were a rise of 44y in H from 19h. 18m. to 19h. 30m., and a fall of 14' in D from 19h. 5m. to 19h. 24m. In the Antarctic it does not seem possible to draw any line corresponding to BIRKELAND'S division into Sections (ii) and (iii). There was a comparatively quiet time for about half an hour prior to 13h., but thereafter there was a constant succession of large movements until after the time included in Plate XVII. The largest of several considerable oscillations in D during BIRKELAND'S Section (ii) consisted of a fall and rise each about 84' between 13h. 33m. and 14h. 28m. During BIRKELAND'S Section (iii) there were larger, but not more rapid movements. Between 18h. 47m. and 20h. 40m. there was a sort of bay in the D trace, a rise of 189' being followed by a fall of 151'. The decreasing movement which began about 20h. 5m. was interrupted by a rise of 78' between 20h. 41m. and 21h. 13m. When resumed, it continued until 22h. 21m., the value of D at that hour being 261' below that at 20h. 5m. Between 22h. 37m. and 23h. 34m. D rose 110'. The H trace was off the sheet on the negative side, most of the time from 14h. 10m. until 21h. 20m, Its longest appearance was from 16h. 28m. to 17h. 42m., when it rose and fell 687, with numerous minor oscillations. The chief turning-point (a maximum) was at 16h. 53m. After 21h. 20m. the H trace remained on the sheet until the end of the 8th. There was rather a rapid rise of 52y between 21h. 20m, and 21h. 33m., followed by a slower motion in the same direction interrupted by numerous oscillations. The total range in H between 21h. 20m. and 24h. Om. was lOly. 26. February 10-11, 1903 (hours 20-3, Plate XVIII). This is classed amongst the "polar elementary" storms. Of it BIRKELAND says, p. 106, "This magnetic disturbance is brief, and commences without any previous equatorial perturbation on an otherwise very quiet day. First a small disturbance appears rather suddenly at about 21h. 6m most powerful at the northern stations .... but is also perceptible in (temperate) Europe and North America. After about 30 minutes, the conditions are once more almost normal .... The powerful perturbation .... does not commence until 23h After about an hour and three-quarters the etorm is over .... At 2h. 30m. on the llth February another short, slight perturbation appears . . . ." On p. 107 BIRKELAND gives a table containing his estimate of the times of beginning and ending and of the maximum for the principal storm, and also the value of PI, the maximum disturbing force in the horizontal plane. At the non-polar stations the beginning is put at about 23h. Om., the maximum at abovit 23h. 20m. The end is given for the non-polar European stations as Ih. Om. on the llth. Jn the Arctic the times of beginning and ending are more variable, but do not depart much from those COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. 269 already given ; the maximum appears about half an hour later than elsewhere. PI is said to vary from 373y at Matotchkin Schar to less than 5y at Batavia. It is given as 35 -8y at Kew, a value similar to that assigned to the other non-polar European stations. Commenting on the fact that the value of P! at Christchurch, 12y, is larger than the values at Honolulu, Zi-ka-wei, and Batavia, BlRKELAND says, p. 108, " This may be explained by the fact that the perturbation in the Arctic regions is often accompanied by simultaneous perturbations in the Antarctic regions, and it is the effect of these latter that is noticed in Christchurch. Our material does not, however, allow of certain conclusions being drawn in this matter." As BIRKELAND does not seem to have had any records from south of Christchurch, the above is presumably pure surmise. At Kew the short commencing movement was represented by a fall of 5y in H between times which I make 21h. 8m. and 21h. 16m., and by a bay lasting from 21-h. 8m. to 21h. 35m. in the D curve, the element falling l'-9 and rising l'-0. The principal disturbance at Kew was represented in II by a rise of 29y from 23h. 4m. to 23h. 19m., a fall of 41y from 23h. 19m. on the 10th to Oh. 16m. on the llth, and a rise of 13y from Oh. 16m. to Oh. 45m. In the D curve there was a bay from about 23h. Om. on the 10th to Oh. 40m. on the llth, the element being depressed. The maximum depression occurred about 23h. 40m. and amounted to 5' -2. In the Antarctic, after being comparatively undisturbed for over 12 hours, the traces became distinctly disturbed about 18h. 36m. on the 10th, or 2 hours before the commencing disturbance noted by BIRKELAND, and the disturbance continued without appreciable intermission until 23h. Om., when the lamp went out. When registration was resumed at Oh. llm. on the llth the disturbance appeared much reduced. The period for which trace is lacking includes unfortunately the major part of BIRKELAND'S principal storm. The commencing movement in the Antarctic was a rapid fall of 72y in H between 18h. 36m. and 18h. 53m. The most striking movement, however, prior to Plate XVIII, was an oscillation which appeared simultaneously in D, H, and V. The turning-point, a maximum, occurred at about 19h. 46m. In the course of about 14 minutes D rose 70' and fell 54', H rose and fell 39y, while V rose 25y and fell 33y. There is, however, no distinct movement at this time in the Kew curves. During the time covered by Plate XVIII the largest Antarctic movements were as follows, + denoting a rise and - a fall in the element : D. H. V. h. m. h. m. / h. m. li. m. 7 h. in. h. m. 7 From 20 47 to 20 56 + 93 From 20 86 to 20 47 -56 From 19 41! to 19 53 - 33 20 56 21 26 -112 20 -)i; 21 6 -59 19 f>:t 20 29 + 51 21 26 22 13 + 93 21 15 21 43 + 94 20 -'!> 20 49 - 30 ,, 22 13 22 38 - 72 22 10 22 36 -77 20 40 20 55 + 33 ,, o 11 ,,11 + 67 20 66 21 5 - 45 21 28 21 35 + 36 22 8 22 36 -100 The H trace was off the sheet on the negative side for a few minutes after 21h. 6m., and got off on the positive side at 2h. 28m., remaining off for over 4| hours. After the new sheet was put on at Oh. llm. on the llth, the D and V traces appeared no more disturbed than usual. Owing to the lack of trace our information is mainly confined to the fact that the disturbances near the commencement of Plate XVIII, which BIRKELAND ascribes to equatorial perturbations, were synchronous with large oscillations in the Antarctic, which formed part of a series which commenced about 18h. 36m., and which persisted without intermission until at least 23h. Om. 27. February 15, 1903 (hours 13-20, Plate XIX). This is classified as a "compound" storm on the following grounds, p. 174, ... "It must thus be assumed that these are in the main polar perturbations ; but the conditions are not simple, indicating, as 270 COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. they do, both in the Arctic regions and in lower latitudes, that there are a number of systems acting to some extent simultaneously." According to BlRKELAND, one noteworthy peculiarity is that the disturbance at most stations commenced nearly 2 hours later, while ending nearly half an hour earlier in D than in H. In Europe, outside the Arctic, the H disturbance is given as lasting from about 14h. 15m. to 18h. 15m., and the D disturbance as lasting from about 16h. 10m. to 17h. 45m. The time of the maximum disturbance is given at most stations as near 16h. 40m. The maximum value of the disturbing force in the horizontal plane varied, according to BlRKELAND, from 392y at Axeloen to 10 '6y at Honolulu, the value at Kew, 38y, being similar to that at most non-polar European stations. There were no Christchurch results. In Europe generally, outside the Arctic, the disturbance consisted of a slight depression in H from about 14h. 15m. to 18h. 15m. the maximum- depression at Kew below the normal being about 20y and occurring about 16h. 30m. and of a more conspicuous depression in D. At Kew there was a fall of 6' -5 in D between 16h. 10m. and 16h. 45m., followed by a rise of 4' - 8 between 16h. 45m. and 17h. 20m. After a stoppage of nearly 20 minutes there was a further rise of about 2' going on until nearly 18h. 30m. This final rise, however, is so slow that opinions might well differ as to its duration and significance. In the Antarctic the curves would hardly be described as disturbed before 16h. Between 13h. 41m. and 14h. 53m., however, D rose 66', while H fell 64y between 13h. 17m. and 14h. 53m. There were two successive bays on the D curve, the first from 16h. 3m. to 17h. 10m. with a rise of 51' and fall of 31', the second from 17h. 10m. to 18h. 10m. with a fall of 52' and rise of 57'. The H trace, which had been off the sheet on the negative side for about 45 minutes, came on the sheet at 17h. 23m., and H rose 63y by 17h. 50m. Following this were a number of oscillations of some size, H rising on the whole until 19h. 33m., the total rise since the trace came on the sheet at 17h. 23m. being about 105y. During this time the V trace shows two fairly prominent bays, the first lasting from 15h. 53m. to 17h. 5m., the second from 17h. 5m. to 18h. 13m. The first bay represents a rise and fall of about 25y, the second a rise and fall of about 48y. Up to 18h. 40m. there were no very rapid movements. Between 18h. 40m. and 18h. 55m., however, there were simultaneous rapid oscillations in D and H ; D rose 52' and fell 48', while H fell 62y and rose 75y. During part of this time, from 18h. 43m. to 18h. 48m., there was a smart rise of 25y in V. This was followed by a larger but slower fall of 32y from 18h. 48m. to 19h. 3m. Conspicuous movements in D and H were just concluding at 20h. Om., the hour answering to the end of Plate XIX. D had risen 42' in the course of the previous 11 minutes, whilst H had fallen 64y since 19h. 46m. These movements seem, however, to be related rather to what follows than to what precedes them. For after a small retrograde movement H continued to fall to a very sharp peak at 20h. 3m., the total fall since 19h. 46m. amounting to Sly. The peak at 20h. 3m. answers to a depression (minimum) in V, rather suggesting that the preceding fall answered to the first phase of a disturbance of the " special type." If this view is correct, the second phase is represented by a rise of 78y in H and of 23y in V, which took place between 20h. 3m. and 20h. 16m. This was immediately followed, between 20h. 16m. and 20h. 38m., by another double movement, also suggestive of the "special type" of disturbance. The H change in this second oscillation was a fall of 72y and rise of 56y, the V change a fall of 36y and rise of 43y. This movement was followed by yet a third sharp oscillation in H and V, terminating about 20h. 54m. The range of H, however, in the third oscillation was only 32y, the fall and rise being equal, while the V movements were under 20y. During these three oscillations there were also oscillatory movements of some size in D, the total range in D between 20h. 3m. and 20h. 54m. amounting to 50'. The oscillations in D were, however, much interrupted by minor oscillations, and it is difficult to decide on their exact relationship to the H and V movements. One would put the commencement of the quiet time in the Antarctic at the end of the third oscillation, i.e. at 20h. 54m., and one would unquestionably regard the disturbances from 19h. 46m. to 20h. 54m. as forming part of a common system. 28. March 22-23, 1903 (hours 12-1, Plate XX). This is classed amongst the "elementary polar" storms, but BIRKELAND, p. 127, explains that "The COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. 271 perturbation ... is in reality . . . composed of two principal phenomena, an equatorial perturbation and a short, well-defined, comparatively powerful elementary polar storm." Of the " equatorial " perturbation he writes: "This . . . begins quite suddenly, at 12h. 58m., with an oscillation that is noticed simultaneously all over the world. In the equatorial regions, this sharp deflection . . . appears principally in H. About the auroral zone the curve oscillates, and the perturbation is noticeable both in I) and H." The time at which this commencing disturbance reached a maximum is given as 13h. 4m. BIRKELAXD concludes, apparently, that the small enhancement of H caused by the commencing disturbance, or at least a fraction of it, persists in low latitudes right through the subsequent polar storm until midnight. At Dyrafjord and Axeloen, however, he notes "a peculiar circumstance," p. 128, viz., that the commencing movement is distinctly oscillatory. The H curve, in fact, at the two stations shows a fall, a rise and then a second fall, and possibly a second, but much smaller, rise. According to BIRKELAND'S table, p. 130, the time of commencement of the "polar" storm varied from 20h. 30m. to 21h. 30m., the time at non-polar European stations being about 21h. 10m. At most stations the termination is put at about 23h. 45m. At most non-polar European stations the maximum occurred about 22h. 10m. ; in North America it was about half an hour later. The maximum value of the horizontal disturbing force is given as varying from about 370y at Axeloen to 12'4y at Batavia, the value at Christchurch being too small to measure satisfactorily. The value at Kew, 44y, is about a mean for non-polar Europe. My examination of the original Kew curves makes the sudden commencement 4 or 5 minutes earlier than the hour, 12h. 58m., given by BlRKELAND. But the movement appears distinctly oscillatory in both H and D, a small fall, lasting 2 or 3 minutes, preceding the rise. In the case of H the fall seems about ly, the rise about 13y ; while in D the fall is I'-O, the rise 2'- 4. There are a number of small movements at Kew in both H and D throughout the remainder of the 22nd, but much the most conspicuous phenomenon is a bay on the D curve from about 21h. 6m. to 23h. 50m., the element being depressed in value. The largest value of the depression is about 8'- 8 and it occurred about 22h. 5m. Looking at the curves in Plate XX, it is easy to recognise in most, if not all, a second movement which somewhat resembles the commencing movement both in size and character, and which occurred about an hour and fifty minutes later. At Kew this second movement occupied some 14 minutes. During it H fell 4y and rose lOy, while D fell 0' - 8 and rose l'-0, the turning-point in each case coming at about 14h. 46m. The reason for mentioning this is because on looking at the Antarctic curves one's eye is caught by two oscillatory D movements larger than their neighbours, occurring during an otherwise somewhat unusually quiet time. So far as I can judge, these movements correspond in time to the commencing movement and the second movement just referred to. The following are the results of the measurement made of the two groups of movement ( + denotes a rise, - a fall) : D. H. V. h. m. h. m. / h. m. h. m. 7 h. m. h. m. 7 { From 12 54 to 12 57 + 20 From 12 55 to 13 5 From 12 56 to 12 59 + 4 Group I. < 12 57 13 -42 13 13 4 + 5 12 59 13 4 - 9 I 13 6 13 13 + 15 \ 14 42 14 46 + 38 14 46 14 52 + 16 14 45 14 49 + 10 Group II. < 14 46 14 54 -47 14 52 14 57 -17 14 49 14 56 -17 [ 14 54 15 5 + 34 In each case it is a little doubtful whether the final rise in D forms part of the same system as the preceding rise and fall. The H movements do not stand out very conspicuously, and it is mainly their approximate coincidence in time with the others that has led to their enumeration. The V movements, though small, do stand out. Unless we assume an accidental coincidence of a truly remarkable kind, the conclusion seems inevitable that the above two groups of movements are duo to the same cause as the movements seen at Kew and elsewhere at the same times. If this be so, then the fact that the Antarctic movements represent 272 COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. disturbing forces much larger than those shown by the curves in equatorial and temperate latitudes is a fact which requires a good deal of explanation on the hypothesis of equatmial electric currents. From 15h. 5m. to 18h. 15m. the Antarctic curves, like those elsewhere, were, on the whole, distinctly quiet. From 18h. 15m. to 18h. 25m. there was a more prominent oscillation in H, a fall of 14y being followed by a rise of 16y, while V rose 16y between 18h. 22m. and 18h. 26m. Between 19h. 13m. and 19h. 22m. H fell 20y and rose 24y, while between 19h. 32m. and 19h. 35m. it rose 16y and fell 17y, the latter an exceptionally rapid though not large oscillation. After 20h. 15m. conditions became more disturbed in all the elements. The oscillations, though not really large, were unusually rapid between 20h. 15m. and 22h. 15m. The D and H traces got rather mixed up and are difficult to distinguish. Amongst the movements in the V trace after 20h. were a fall of 16y between 20h. 14m. and 20h. 19m., a rise of 24y 20h. 19m. 20h. 24m., and a fall of 39y 20h. 24m. 20h. 31m. After this there were numerous small but rapid oscillations, during whose incidence V rose to a maximum at about 21h. 13m. There then ensued a fall, interrupted by minor oscillations, which continued until about 22h. 40m., in the course of which V fell 56y. This was followed by a rise of 97y in two steps between 22h. 40m. and 23h. 25m., the rise being conspicuously most rapid during the first 20 minutes. From 22h. 20m. to 23h. 3m., simultaneously with the most rapid part of the change in V, there was a prominent bay on the H curve, the element falling 61y and rising 36y. After 23h. the D and H traces were somewhat uncertain owing to operations connected with the clearing away of snow near the Magnet Hut, and the record was suspended altogether from 23h. 37m. to Ih. 5m. on the 23rd. It will have been noticed that the large rise in V and the prominent bay on the H curve took place during BiRKELAND's " elementary polar " storm. So far as mere amplitude is concerned, the large rise in V after 22h. 40m. is, perhaps, the only phenomenon which merits the title of disturbance as judged by the Antarctic standard. The number, however, of minor oscillations in V and the rapidity of the oscillations in D and H, especially H, are certainly a little outstanding. 29. March 30-31, 1903 (hours 19-3, Plate XXI). This is classed amongst the "elementary polar" storms. BIRKELAND explains, however, pp. 115-117, that this "elementary" storm was preceded by an "equatorial" perturbation. He adds, p. 116: "As early as 19h., those little, sudden, very variable perturbations are noticed, which occur simultaneously all over the Earth. . . The ("equatorial") perturbation appears to be over at about 23h. 12m. . . ." The "polar" perturbation is said to be recognisable at Kaafjord about 23h., being earlier there than elsewhere. It did not commence at Dyrafjord until about Oh. 24m. At non-polar stations the commence- ment was, in general, about 23h. 50m., the end about 2h. 10m. on the 31st. The hour of occurrence of the maximum is givsn as Oh. 30m. for temperate Europe, but Oh. 58m. for Dyrafjord. The maximum value of the horizontal component of the disturbing force varied from 546y at Dyrafjord to 10 - 5y at Batavia. The value for Kew, 41 -5y, is about a mean for non- Arctic Europe. The earliest decided movement shown in Plate XXI at non-Arctic stations is a small oscillation apparent in almost all the H traces and in some of the D traces. At Kew the double movement lasts from about 19h. 23m. to 19h. 31m., the turning-point coming about 19h. 27m. There was a rise of 7y and then a fall of 6y in H, while D fell and then rose l' - 3. While examining the original Kew curves I noticed a somewhat conspicuous movement of similar size and character about an hour and fifty minutes earlier. This consisted in the case of H of a rise of 6y between 17h. 32m. and 17h. 36m., followed by a fall of 9y between 17h. 36m. and 17h. 40m. The reason for mentioning this will appear presently. During the whole time covered by Plate XXI the most conspicuous phenomenon at Kew (and the same is true generally of the other non-polar stations) took place between Oh. 10m. and Ih. 10m. of the 31st, i.e. during the "elementary polar" storm. D, which had fallen l'-8 between 23h. 45m. and Oh. 10m. on COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. 273 the 31st, rose 9'-6 between Oh. 10m. and Oh. 32m., and then fell ll'-5 between Oh. 32m. and Ih. lorn. Between Ih. lorn, and Ih. 40m. there was a rise of 2', and a further very gradual rise continued until nearly 4 a.m. H fell lOy between 23h. 45m. and Oh. 6m. on the 31st, rose 25-y between Oh. 15m. and Oh. 42m., and fell 25y between Oh. 42m. and Ih. 27m. The only other movements on the Kew curves likely to attract attention occurred shortly after 21h., i.e. 2| hours before the commencement of the "elementary polar" storm. Between 21h. 1m. and 21h. 35m. H rose 5y and fell 12y, then rose 19y in two steps and fell 67. The D trace shows only a very small movement about 21h. 3m., consisting of a fall of 0' 8, followed by an equal rise. The Antarctic curves had been rather quiet for some hours prior to 1 9h. on the 30th, but they exhibit a somewhat prominent oscillation occurring apparently simultaneously in the D, H, and V curves, all showing a sharp peak at 17h. 35m. The oscillation lasts from about 17h. 31m. to 17h. 39m. In D there is a rise and fall, each 12', in H a rise of 16y and fall of 12y, in V a rise and fall, each 7y. The V movement, though small, is clearly shown, the curve being unusually quiet for some time before and after. The fall in D and H only paused for a minute at 17h. 39m. and then proceeded for some minutes, but at a slower rate. The oscillation is synchronous with the earlier movement at Kew, which occurred prior to the time considered by BlRKELAND. The next movement in the Antarctic which catches the eye occurs between 19h. 25m. and 19h. 31m., with turning-point at about 19h. 28m. in all the elements. The D movement is small and not very distinct. It consists apparently of a fall of about 2' and a rise of 10'. H falls and rises 18y, while V falls and rises 6y. These movements synchronise with the second of the two movements at Kew, i.e. with BIRKELAND'S " equatorial " perturbation. One is rather reminded of the occurrence of the two oscillatory movements on March 22 ; the interval between the two movements is nearly the same on the two occasions. The movements of March 30 do not appeal so much to the eye as those of March 22. The fact that the disturbances in the Antarctic are again larger than the corresponding ones at stations nearer the Equator will doubtless have been noted. Though not seriously disturbed, the Antarctic curves all show oscillations of increased amplitude between 19h. 25m. and 21h. 40m. The largest movements during this time consisted of a bay, apparently simultaneous in the three traces, from about 21h. 6m. to 21h. 26m. During it D fell 35' and rose 52', H rose and fell 26y, while V fell 25y and rose 38y. This movement, it should be noticed, occurred at a time when the curves at Kew and the other non-polar stations showed increased activity of disturbance. From 21h. 40m. to 22h. 25m. there was a relatively quiet time in the Antarctic. There then ensued from 22h. 25m. to 23h. 30m. an interval during which the V trace showed a number of rather sharp oscillations, which were accompanied by more or less simultaneous oscillations in D and H, the H oscillations being of enhanced size. Several of these oscillations rather suggest the " special type " of disturbance, but their period is short and the rises in V do not exceed the falls. The two following cases gave the largest changes in V : From 22h. 38m. to 22h. 53m. From 23h. 5m. to 23h. 18m. Fall . . ~'V Rise .... D. H. V. Fall .... Bise .... D. H. V. 24 15 7 29 21 y 25 14 20 20 7 35 22 7 33 17 The oscillations in V continued for some hours, but with diminished size. In the case of D and H, however, movements of increased size and duration took place. The chief movements in D were a fall of 99' between 23h. 30m. on the 30th and Oh. 20m. on the 31st, a rise of 120' between Oh. 20m. and Oh. 56m., and a fall of 111' between Oh. 56m. and Ih. 57m. These movements were accompanied by shorter oscillations of comparatively small size. During this time there was a large bay on the H curve, but its nature is imperfectly shown, as the trace was off the sheet on the positive side from Oh. 16m. until Ih. 7m. Between 23h. 40m., when the bay may 2 N 274 COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. he said to commence, and Oh. 16m., when the trace went off the sheet, H rose 54y ; between Ih. 7m. and Ih. 42m. it fell 42y. Between Ih. 42m. and 2h. 12m. H rose 42y, again going off the sheet, but only for about a minute. After 2h. 10m. the Antarctic curves were quiet, according to the Antarctic standard. The large D and H movements observed in the Antarctic after 23h. 30m. on the 30th synchronise with BIRKELAND'S " elementary polar " storm. It is impossible to decide from the form of the curve alone at what instant D was really most remote from the normal, but this would seem to have been at the turning-point at Oh. 56m., when the maximum for the day occurred. This answers practically to noon, L.T., thus rendering possible a comparison with mean results for that hour from adjacent days. Taking a mean from the five previous days we deduce an excess on the 31st of 110', answering to about 200y in force. The simultaneous value of H was also substantially in excess of the normal, but how much it is impossible to say, as the trace was off the sheet and very steep both when going off and coming on. Probably the maximum value of the horizontal component of the disturbing force in the Antarctic, measured after BIRKELAND'S method, was fairly similar to that at Axeloen, 280y, or, say, half that at his most disturbed station, Dyrafjord. The disturbance of March 30-31, regarded from an Antarctic standpoint, was, of course, by no means a large one. 30. The intercomparison of the Antarctic curves with BIRKELAND'S records points to the following conclusions : 1. At the times of the small sudden movements which BIRKELAND assigns to "equatorial perturbations" there seem to be almost always (no certain exception has been noted) corresponding movements in the Antarctic. The Antarctic movements are larger, usually much larger, than those recorded at the equatorial or non-polar stations, and are usually, if not always, oscillatory in type. 2. At the times of BIRKELAND'S " elementary polar " storms in the Arctic, the conditions in the Antarctic are generally more than usually disturbed. The times of the largest movements in the Antarctic usually occur not far from the times of maximum disturbance in the Arctic. Not infrequently, however, the disturbances in the Antarctic continue without any marked subsidence of intensity throughout times which BIRKELAND regards as including two or more elementary polar storms. 3. In the Antarctic, the sudden commencing movements which BIRKELAND assigns to equatorial perturbations are sometimes immediately followed by disturbances which are not separated by any markedly quiet interlude from subsequent disturbances which synchronise with Arctic disturbances which BIRKELAND believes of the " elementary polar " type. 4. The " elementary polar " disturbances of BIRKELAND seem practically all confined to the hours 13 to 2, G.M.T. (or a.m. to 1 p.m., Antarctic L.T.), and so occur at a time when the Antarctic movements are usually of a rapidly oscillatory character, and avoid the hours when the rounded wave-like movements of the " special type " are usually found. Further, BIRKELAND'S results are limited to the Antarctic Summer, i.e. to the season at which few if any disturbances of the " special type " were recorded. It would clearly be of great interest to know what was the nature of the phenomena in the Arctic during the times of occurrence of the Antarctic " special type " of disturbance ; also whether the Arctic stations show a marked diurnal variation in the type of disturbance corresponding to that seen in the Antarctic. One of the results of the comparison has been to make me realise even more clearly than before the desirability of much reduced sensitiveness in magnetographs intended for use in polar regions. It is clear that most fundamentally important results might be hoped for if simultaneous complete records were obtainable from a series of stations in the Arctic and Antarctic regions so situated that the effects of day and night could be adequately brought out. This comparison should extend over a complete year, so as to bring out seasonal effects. Plate 1. s o ' o . - . M h K E M -' 1 1 < 1 p t-' cr t- J ?M ' en S <-< t- CO oC t --. U < ) 00 I S t"*l J 1 tO c J c^ c ** ' ) * * " ^ 5 O < o r* ^ r^ ") cO 4 N 1 b ^ 6 oo D co O tO N N O m >H r> in c n r-< u N <-i t- O cO f M N t ^ f ? ' s D *^ HH j, i I N N O CO N ^4 s N a. H 1-3 S t-i O tO J N 5 r, 4 j O O iO o f^ C^J n If H 2 /^ iTJ O IT) M t^i O CO ^ N i 3 g I bO Plate II. o K-> o o n | 2? $ cC O eo 2? & M O oO 3 o o ol & c o o -c p, u-i O >n < \ cO o 2? f-i O JO 9 N - o I s c8 3 _O I 3 O w a Pfafo ///. 3 a a CD S a P a > 10 f crG.M.T. at Christchurc S 1 I 1 O ra ff> a M cS - >o H <- >^ ^ o w 1 1 B P s U-,... U~ U,- n 2 o 1O - Plate IF. IT) 00 V0 > o w O w o p 00 oO i o oO CO Plate V. >J i * :> ^< i i i 5 5 * 4 h I 1 s - "> ? J s ^H i t C 1 Li t- 4 vC ? 9 c >i tj ( ? t 3 i < 8 ? ? ^ 5 < I -H H W* p i o ? o 1 T C f O* k 1 1 c ^ O ^ > & H o g R H K 1 I B 1 3- H j * W t >1 s i- t( s > s s 1 " > O O T 3 -H 1 o \- i o * f T 1 T c- u \ \ 1 vi I y O 5 y 2 X ri t 2 ^ ? ^ 5 Plate VI. O 0} H a) 4 en o O" I O co -^ ts 5 I o "-) H H H Plate VII. > I ' I a a 1 O >o H > 1- H 1- , xn a in H 1 H H 1 H ? ?- N "S O ~ ^ to ^ I M 3 1 o M > a N la p Vi CO o ft IT) -]3= w 10 Plate FIJI. o >o .to V) IT) pj o Plate IX. ride X. CO o 01 M ,fl 3 O" O rH in -s I >> 3 3 O Plate XL Plate'-XIl. Plate XIII. oo 1 02 I I a 2 , Plate XV. 1 o 5 fi o M tf> 3 to 3 0) O 1. o c o M CD CO O o z cu * >O CO to -s ffl j; a.' CO in c f> en a Plate XVI. April 5-6, 1903. I754' J Ch.ristch.urch Declination (East) ChristchurcK Horizontal Force. Mauritius Horizontal Force. Col aba Horizontal Force. Fal mouth Horizontal Force. Fal mouth Declination (West) IIRM. 12 IA.M. 2345 Disturbance of April 5 and 6, 1903, at Co-operating Stations. Plate XVII. o 01 C O jj ? o O c O o cO CO in in PlatelXIX. M- z o: 00- oO- 0) o O 3 f-( h 6 z: < ^- o 00 Plate XX. 8 I Q s o a 3 o .8 8 (5 J .2 ft Plate XXI. Plate XXII. Plate XXI II. I 1 1 Plate XXIV. Plate XXV. Plate XXl't. Plate XXVII. Plate XXVIU. Plate XXIX. Plate XXX. Plate XXXI. I4937 T 150 7' Disturbance of August 22 am Plate, XXXII. -06930 1903, at Winter Quarters. Plate XXXIII. Plate XXX IF. Plate XXXV. a- en .8 3 o O 8 S Plate XXXVI. Plate XXXVII. May 22, 1903 . August 31, 1902. August 15, 1903. \r I ^^^^^^^r H D 8RM. Ho Ha D52'; H53y; gp.M. Di39'-, H.7IX; V74V- IA.M. H Q ; Vi62y. Plate XXXV11I. 3P.M. June 4, 1903 D i5i' H 45 y H Oct 4-5, 1902 . H .HO D 42' Oct. 11-12, I9O2. H Plate XXXIX. li D September 6-7, 1302. IE H September 18-19, 1902. H IAJ1 IIP.M 12 1A.M. 2 B 56' H 441 H soy , V^^vSL \ CAL' Plate XL. o CD <8 I CO cO M O en C<1 m Cxi o en f 4) b N 3 Plate XL1. February 8, 1903. H February 10, 1903. H SA.M. H fl March 6, 1903. H H V 37X H ioiy V o r-l t~ M I Plate XLII. O E N N 00 o T i o l ~ iC - >* >. o Q ; > > I i < IT)' 01 ^ 3^ -- !> O * ff) Plate KL1IL CO i-i 1 u >o o o en <0 v O) I