V \ NATIONAL ANTARCTIC EXPEDITION T^-t'- 1901-1904 i ¥ PHYSICAL OBSERVATIONS WITH DISCUSSIONS BY VARIOUS AUTHORS PREPARED UNDER THE SUPERINTENDENCE OF THE ROYAL SOCIETY LONDON: PUBLISHED BY THE ROYAL SOCIETY 1908 ^ (P \^ ~^ ^kL^ ^-J- \ iii 1 CONTENTS. Preface v I. Tidal Observations in the Antarctic Eegions, 1902-1903 1 II. Pendulum Observations 17 III. Earthquakes and other Earth Movements recorded in the Antarctic Kegions, 1902-1903 . 37 IV. Antarctic Observations of Aurora, 1902-1903 97 V. Antarctic Magnetic Observations, 1902-1904 127 Index 191 53^95 t iv ] PLATES. Earthquake Observations — Plates 1-G, fullmmng jj. 96. Plate 1. — Velocity Diagrams. „ 2. — -Chart of Earthquake Origins. ,, 3. — Isoseists of Earthquakes. Plates 4-6. — Reproductions of Earthquake Records. Aurora Observations — Plates 7-14. Plate 7. — Charting Diagram, facing p. 99. Plates 8-14. — Coloured Plates of Auroras, folloiuiiu/ p. 126. Magnetic Observations — Plates 15-19, following p. 156. Plate 15. — Views of Winter Quarters and Magnetic Houses. „ 16. — Panoramic Views of Winter Harbour. ,, 17. — Magnetic Declination Chart. ,, 18. — Chart of South Magnetic Pole. „ 19. — Inclination Chart. "Scotia" Magnetic Observations — Plates 20, 2\, follmvimg p. 192. Plate 20. — View of Observatory, Scotia Bay, Laurie Island. ,, 21. — Manner of taking Preliminary Observations, and Plan of Observatory Hut. MAPS. Map of Ross Island, McMurdo Sound, Sic, facing p. 129. Map of Laurie Island, at end of volume. PREFACE. The present volume contains a series of Reports on various portions of the Physical observations made during the voyage of the "Discovery" in the National Antarctic Expedition, together with other observations taken in difterent regions of the world in further illustration of the subjects under discussion. The Royal Society, having undertaken the supervision and publication of the Physical work of the Expedition, placed the preparation of these Reports in the most competent hands available. The Tidal work of the " Discovery," which was in charge of Lieutenant M. Barxe, R.N., has been here discussed by Sir George Darwin, to whom we are also indebted for the discussion of the Tidal observations of the " Scotia," which were entrusted to the Royal Society by Dr. W. S. Bruce, the leader of the Scottish Antarctic Expedition. The analysis of the " Scotia " observations has been furnished by Mr. Selby and Mr. Hunter, Tidal Assistants at the National Physical Laboratory. The other Physical work of the "Discovery" which is included in the present volume, consisting of Pendulum observations, Earthquake registers. Auroral journal and Magnetic observations, was in charge of Mr. L. C. Bernacchi during the Expedition, except the Magnetic work at sea, which was undertaken by Lieutenant Aemitage. Engineer Commander R. W. Skei.ton, besides his valuable contributions to the Photographic work, rendered much assistance in the Pendulum observations. All the officei's of the ship, both scientific and naval, lent their help in the recording of Auroral phenomena. The results of these various lines of physical research are presented in the following pages. The account of the Pendulum observations has been prepared by Mr. Bernacchi. The discussion of these results has been supplied by Dr. Chree, F.R.S., of the National Physical Laboratory, whose valuable assistance has been placed at the service of the Royal Society by the Director, Dr. Glaze- brook, F.R.S. To Dr. John Milne, F.R.S. , we are indebted for the interesting discussion of the results of the Seismological observations and the comparison of these observations with others made contemporaneously in different parts of the world. Mr. Bernacchi has re-written and arranged the Auroral journal, to which Dr. E. A. Wilson has contributed a series of striking drawings. The Report on the Magnetic observations has been supplied by several contributors. Mr. Bernacchi, who spent some months in the reduction of the Difterential Magnetic work, has written the introductory statement as to the conditions in which the observations were taken. Commander Chetwynd, R.N., of the Hydrographic Department of the Admiralty, has been so good as to supply the section on the reduction of the Absolute and Relative Magnetic observations. As planned by the Royal Society, simultaneous Magnetic observations were taken on pre-arranged term- days at certain observatories, as well as at the Winter Quarters of the "Discovery." The results as measured from the magnetograms have been given in Tables of Hourly Values compiled l\v Commander Chetwynd and by Dr. Chree. The rest of the Magnetic work of the " Discovery " is now in course of being worked out at the National Physical Laboratory and will form the subject of a separate volume. The Magnetic work of the "Scotia," which was entrusted to the Royal Society by Dr. Bruce, was placed in the hands of Dr. Chree, who has discussed it in the present volume, while Mr. Mossman, who took the original observations, has furnished the account of the conditions in which they were taken. Arch. Geikie, Sec. R. S. Royal Society, Burlington House, 2ndJuly, 1908. [. TIDAL OBSERVATIONS IN THE ANTARCTIC REGIONS, 1902-1903, ANTARCTIC TIDAL OBSERVATIONS. I. Observations of the " DiscoTery," by Sir Geoege Daewin, K.C.B,, F.R.S. II. Observations of the " Scotia," by F. J. Sblby, M.A., J. de Geaafp Huntee, B.A., and Sir Geoege Daewin, K.C.B., F.R.S. I. TIDAL OBSERVATIONS OF THE "DISCOVERY." BY SIR GEORGE DARWIN, K.C.B., F.R.S. The "Discovery" wintered in 1902 and in 1903 at the south-eastern extremity of Ross Island, on which Mount Erebus is situated, in south latitude 78° 49' and east longitude 16G° 20'. The station is near the west coast of a great bay in the Antarctic Continent, and the westerly coast line runs northward from the station for about 9° of latitude. To the eastward of the bay, however, the coast only attains a latitude of about 75° and follows approximately a small circle of latitude. Since the tide- wave comes from the east and travels to the west the station is not sheltered by the coast to the westward, and the continent to the eastward can do but little to impede the full sweep of the tide-wave in the Antarctic Ocean. It is true that Ross Island itself is partially to the east of the anchorage, but it is so small that its influence cannot be important. Of course the westward coast line must exercise an influence on the state of tidal oscillation, for regarding the tide-wave as a free wave coming in from the east, it is clear that it will run up to the end of the bay and then wheel round northward along the westerly coast. It would seem, then, that the situation is on the whole a good one for such observations. Of course their value woidd have been much increased if it had been possiljle to obtain other observations elsewhere. The following account by Lieutenant MiCHAEL Barne, R.N., explains the manner in which the tidal observations were made : — " On our arrival in the vicinity of our Winter Quarters on February 8, 1902, a good deal of the previous year's ice remained attached to the land. As there was no foreshore, and pieces of this ice were constantly moving out, it was impossible to erect a tide-pole. With the final departure of the old ice, the temperature fell, and young ice formed continually, only to be quickly broken up by the almost incessant easterly winds. " As this state of aftairs promised to last for a considerable time, an effort was made to obtain records of the tides. A stout graduated pole was erected alongside the ice foot in about 10 feet of water, the lower end being heavily weighted and the upper end securely guyed. Some intermittent observations were secured in this manner, but they are probably of little value, as the ice was continually forming round the pole, which was only with difficulty freed from it. Besides this, communication with the shore, and consequently approach to the tide-pole, was con- stantly interrupted. "On the ship being finally frozen in, a tide gauge of the following nature was erected (fig. 1). " A single length of pianoforte wire (sounding wire) was led through a block, secured to the head of a tripod. One end of this length was attached to eight 25-lb. sinkers, which were lowered to the bottom. Four 25-lb. sinkers were secured to the other end in such a manner as to allow of their free movement, between the ice and the block, as the ice, with the tripod, rose and fell with the tide. An indicator was clamped to the wire, and a suitable scale secured to the tripod. Fig. 1. " It was thought that the motive force supplied by the weight of four sinkers would be sufficient to draw the smooth surface of the wire through the ice as the water rose and fell, whilst, in case it should fail to do so, the weight of eight sinkers would not be sufficient to break the wire. B 2 Indicator 4 25lb. sinkers " As it was considered possible that the ice, owing to the proximity of the land, might not maintain a uniform position relative to the surface of the water, a small hole was occasionally opened close to one of the tripod legs, to which was attached a mark, indicating the height to which the water should rise. A few observations showed that no error from this cause was to be apprehended. " This gauge was placed about 200 yards from the ship, and two-hourly readings with but slight interruption were continued from April 12 to April 28. "Some sluggishness in its movement, which was eventually noticed, and its final breakdown, was possibly partly due to the thickening of the ice, but principally, I think, to the fact that too small a block was employed at the tripod head. The scale was, by accident, secured so that the readings increased upwards, consequently they have to be inverted.* " It was originally intended to place a tide gauge in the ship, owing to the far greater convenience of position, but it was thought that the position of the ship relatively to the water surface might alter and this might lead to errors. It was hoped that by placing one on the ice as well as one in the ship, check observations might be obtained to determine if this source of error existed. This was eventually accomplished on April 25, but by the time the ship gauge was erected the outside gauge had ceased to be entirely satisfactory, for the reasons given. The observations, however, show a close approximation of movement. Fig. 2. " The ship gauge was arranged as .shown in fig. 2. The supporting blocks were secured rigidly, and, until May 10, the wire was led directly through the ice. As the friction was gradually increasing, a suggestion made by Dr. Wilson was adopted on that date, and the wire was taken through a ttibe, filled with paraffin oil and closed at the top and bottom with a hard wooden plug through which the wire passed. A maximum and minimum arrangement, with balanced weights, was added, as shown in the * This oversight was rectified before May 12.— G. H. D. sketch.* Unfortunately, both on May 10 and May 12, in refitting the gauge, the indicator had to be fixed afresh and therefore the observations cannot be referred to a common zero. "A mark was phiced on the ship's side to ascertain any vertical movement of the ship relatively to the water surface, and a long plummet was secured in the engine room, to show any alteration in her inclination to the vertical. "On April 6, 1903, the tide gauge was re-erected and observations continued, but, owing to the large number of observers employed, the maximum and minimum arrangement was not fitted. " The height of the mark (b) on the ship's side above the water was ascertained about once a month in the same manner as during the winter of 1902, i.e., by digging a hole through the ice below the mark and measuring its height. On these occasions the difference between the heights of the leading blocks wa.s measured in the following manner. A wooden scale (c) marked in half inches was secured to 'the beam (a) in a vertical position, close to the outer block, with its zero mark on a level with the top of the sheave. A wooden instrument, shaped like the letter T, and having a lead weight attached to its lower end and a hole, A (see fig. 3) in the centre of its upper end, was hung freely on a nail in such a manner that its upper side was horizontal and on a level with the top of the sheave of the inner block. By bringing the eye on a level with the upper side of this T-piece, and noting the position on the scale at which the upper side, if produced, would cut it, the reading of the scale was obtained which gave the difference in height of the blocks. " By taking periodical measurements of the height of the mark, and the difference between the heights of the blocks, data were obtained by which readings could be corrected for alteration of the trim and the list of the ship respectively and reduced to a common zero, namely that on April 6, when the tide gauge was erected for the winter of 1903. "On September 21, 1903, the wire carried away close to the place where it was secured to the weight resting on the bottom. On examination the wire was found to be greatly eaten away, from the point of attachment to a height of several feet, presumably on account of an action between the cast-iron sinkers and the steel pianoforte wire." The series of hourly observations was occasionally interrupted by accidents, and the trim and list of the ship changed a little from time to time. Accordingly it is not possible to treat the observations as a continuous whole. The series was therefore broken into a succession of months, so chosen as to avoid periods of manifest irregularity or of accidental interruption, and each month was treated independently. The choice of the method of harmonic analysis to be employed seemed to lie between that explained in the " Achniralty Scientific Manual" and that devised for the use of the tidal Abacus. t The method of the * No use has been made of this arrangement. — G-. H. D. t 'Manual of Scientific Inquiry,' Article "Tides"; and "On an Apparatus for Facilitating tlic Reduction of Tidal Observations," 'Roy. Soc. Proc.,' vol. 52, p. 345. I take this opportunity of correcting a mistake in the Manual article, discovered by Mr. Selbt when reducing tlie " Scotia" tidal observations. At p. 63 — For the tide K.,. In the formula for tan y\i, m the denominator, for 3 -67 p, read 3 71 p, for a fortnight's observation, and 3 '84 p, for a month's observation. In the formula for Hs wherever 3 "67 occurs read 3 -71 for a fortnight, and 3 -84 for a month's observation. The formula H" = -— — Hg remains correct. 3 -67 For the tides 'K^ and P. In the formula for H' the 3 in the numerator (but not that in the denominator) should be replaced by 3 '007 for a fortnight's observation, or by 3 -027 for a month's observation. The formula 1, °A Fig. 3. Hn = - H' remains correct. f + V + 1^ + 6^ -88 for a fortnight, and For k' = Kp = f -I- V + <(> read k = for a month. The succeeding numerical example must be corrected accordmgly. Tlie only sensible change is that k place of 327°. + t" + V + (() + ]3''-29 Manual is considerably more laborious than the other, and it was highly desirable that the Abacus should be used if it could be trusted for a short series of observations. I therefore asked Mr. Wright, who carried out the reductions and was familiar with the use of the Abacus, to reduce the first month in duplicate by the two methods. Curves were drawn through ordinates representing the mean height of water at the 24 hours of mean hmar time, as derived in the two ways. Although the whole range of height in the 24 mean lunar hours was only about six inches, the two curves showed a substantial agreement. The same process was then applied with 0-time, when the range was found to be about 15 inches, and the agreement of the two curves was very close. The method of the Manual showed several sharp peaks or irregularities in the curves which were nearly smoothed out by the use of the Abacus. Such peaks would not aflfect the values of semi-diurnal or of diurnal components to a sensible amount, and as they are clearly accidental I concluded that the use of the Abacus was quite satisfactory, and accordingly that method was adopted throughout. In the use of harmonic analysis it is necessary that the month under discus.sion should differ a little in length according to the tide which is being evaluated. For finding the Mo-tide months of .30 days or of 29 days would be almost equally advantageous, but as 30 days gives us one more day of observation that period was adopted. Similarly 30 days is appropriate for the S2-tide." For a short period of observation it is necessary to regard this tide as compounded of the So and Ko-tides, and we must also suppose its range to vary with the sun's parallax. The separation of these two tides from one another depends on theoretical considerations, which appear to be well founded. Similarly, in a short series of observations the Ki and P-tides must be treated as fused together in a single tide, and they are separable by theoretical considerations only. For these two tides a month of 27 days is appropriate. Lastly the analysis for the 0-tide demands the use of a month of 28 days.* I determined, then, to separate the months in such a way that the shortest months (27 days) should follow one another as closely as possible, while the longer months should overlap .slightly. Whenever any event occurred whereby it seemed likely that the observations might be vitiated, the months were chosen so as to omit the time of possible or actual abnormality. It was clearly desirable that the largest possible number of independent or nearly independent months should be discussed. This consideration led in one case to an overlap of as much as six days ; thus the fifth month of 27 days ended on October 19, while the sixth month began on October 13. In the few cases where hourly obsei'vations were missing, the defects were made good by interpolation. Although the observations began in April, 1902, the first satisfactory continuous period began on May 12. It will be well to state the epochs for the succession of twelve months which it was possible to obtain, and to add a few comments on the observations. First month. This begins with C' May 12, 1902. The observations really begin at 2'', but extrapolated values were used for 0'* and 1''. Second month. This begins with C' June .5, 1902. On the afternoon of July 5 the wire attached to the sinker parted and the observations ceased. The apparatus was only reinstalled at .5 p.m." on July 23. Third month. This begins with 0>' July 24, 1902. Fourth month. This begins with 0'' August 23, 1902. Fifth month. This begins with 0'' September 23, 1902. The height for 6^ on October 20 was interpolated. On October 1 it was found that the ship had shifted so as to affect the readings by one inch. The date at which the shift had occurred was unknown, and, moreover, so small a change could not aft'ect the results sensibly. Sixth month. This begins with 0'' October 13, 1902. On November 9, the four hourly values, l'> to 4'' inclusive, were missing and were supplied by interpolation. As already remarked, this month considerably overlaps the one before it. This was necessary if a * This use of montlis of various lengths necessitates some small arithmetical changes in the method as explained in the paper on the Apparatus referred to aboTe. seventh month was to be secured before the observations ceased for the season, but the choice of the stage at which the overlapping should be made to occur was more or less arbitrary. Scivnlh month. This liegins with 0'' November 13, 1903. The observation for 22'' of December 9 is missing and was supplied by interpolation. On December 13 the wire parted and the series ended for the year. In the second winter, that of 1903, somewhat greater care seems to have been taken to note the small shifting of the ship. Eiffhth month. This begins with 0'' April 6, 1903. Between April 22 and May 3 the ship shifted so as to make the readings too high Ijy 3 inches, compared with the earlier ones. As an arbitrary correction I deducted 1 inch from all heights from 0'' April 24 to 0"^ April 27 ; from 1'' April 27 to Q^ April 30 I deducted 2 inches ; and for the rest of the month the full 3 inches. These arbitrary corrections were submitted to independent harmonic analysis, and it appeared that they afforded cori'cctions so minute as to leave the tidal constants virtually unchanged. Ninth month. This begins with 0'' May 9, 1903. The ship shifted considerably at some time about June 12, and as it is only possible to obtain one month before that date, thcie is an unutilized gap of a few days between this month and the one before it. Tenth month. This begins with 0'' June 15, 1903. On July 10 a sensible shift in the trim and height of the ship was discovered. This necessitates the addition of 4^ inches to all heights, 2i inches being due to angular movement and 2 inches to vertical movement. As an arbitrary correction I added 2 inches to all heights from O'' July 8 to 0'' Jidy 9 ; and afterwards I added the full 4|- inches. Eleventh month. This begins with O^ July 14, 1903. Twelfth month. This begins with 0'' August 14, 1903. After September 8 the observations were only taken every two hours, and for the remainder of the month the values at the odd hours were interpolated. The observations stop on September 20, but are not used in the reductions after September 13. No corrections have been applied for changes in the barometric pressure. As the application of such a correction would have been very laborious and, moreover, somewhat speculative, I have relied on the automatic elimination of the inequalities produced hy taking mean values. The following are the results of the twelve harmonic analyses, the heights being stated in inches : — N.B. — The values of Aq represent merely the changes in the position of the ship and have therefore no physical significance ; all the heights are stated in inches. The values of H and k are somewhat irregular from month to month, and it is therefore not permissible to adopt the mean values of H and k as representing the mean tide. I have therefore formed H cos k and H sin K for each month and have taken the mean of each as giving the mean values of H cos k and H sin k. It is easy to compute from these the proper mean vahies of H and k for each tide. The results are given in the following table : — Mean Values of Tidal Constants. Semidiurnal tides. Diurnal tides. Ms H = 1 -966 inches = -164 feet K = 9° -9 = 10° K, H = 9 -245 inches = -770 feet H = 14° -05 = 14° So H = 1 142 inches = 095 feet K = 272° -1 = 272° P H = 3 -082 inches = -257 feet K = 14° -05 = 14° K., H = -311 inches = -026 feet K = 272° -1 = 272° O H = 9 -264 inches = 772 feet K = 359° -5 = 0° The sum of the semi-ranges of the three diurnal tides is 21 • 6 inches and of the three semidiurnal tides is only 3'4 inches. This result corresponds with the fact that little trace of the semidiurnal tide is to be discovered from mere inspection of the tide curve. When tidal observations have been reduced it is always important to verify that the constants found do really represent the tidal oscillation, for, in computations of such complexity, it is always possible that some gross mistake of principle may have slipped in unnoticed. Such a verification is especially important in a case where the tides are found to be very abnormal, as here, and where the results from month to month are not closely consistent. I accordingly asked Mr. Glazebrook to run off curves for two periods with the Indian tide-predicter at the National Physical Laboratory. The constants used were the means for the tides evaluated. It is jirobable that a better result might be attained if a number of other tides, with constants assigned by theoretical considerations from analogy with the constants actually evaluated, had also been introduced, but I did not think it was worth while to do so. Evidence will be given hereafter to show that the smaller elliptic diurnal tides must exercise an apprecialile influence. The periods chosen for the compariison were about three weeks, beginning on May 12, 1902, and nearly the same time in November. It does not seem worth while to reproduce the whole of the observed and computed curves for these periods. The observed tide curve has frequently sharp irregularities, presumably produced by weather or by unperceived shifts of the ship, and the maxima are sometimes sharp peaks instead of flowing curves. However, on the whole, the computed and observed curves follow one another very well, at least throughout all those portions where the diurnal tide is pronounced. Where the diurnal inequality is nearly evanescent, and the semidiurnal tide becomes perceptible, the discordance is sometimes considerable, although, even in these cases, every rise and fall of the water is traceable in the computed curve. Such discordance was inevitable, for at this part of the curve all those tidal oscillations which have any importance have disappeared, and only those tides remain which are very small ; moreover, most of these tides are avowedly omitted from the computed curve. I give two figures. The first shows the two curves where the diurnal tide is large, viz., from 0'' to 24'^ November 18; it is a rather favourable example of the general agreement referred to above. The second figure, from 12'' May 29 to 12'' May 30, is selected because it exhibits by far the worst discordance which occurred in the six weeks under comparison. I conclude that the reductions are quite as good as could be expected from tide-curves which present as much irregularity as these do. It would not be possible to make a very good tide-table from the constants, but no one wants a tide-table for Eoss Island. We only need sufficient accuracy to obtain an insight into the nature of the Antarctic tides, and the constants are quite sufficient for that end. When the mean heights of water at the 24 hours of mean lunar time were plotted in curves for each month, it became obvious that a pure semidiurnal inequality did not represent the facts very closely, and that there remained also a sensible diurnal inequality. Such an inequality is given by the tide Mi, and if we neglect the minute portion of the tide Mi, which depends on the terms in the tide-generating potential, which vary as the fourth power of the moon's parallax, such an inequality is found to depend on the composition of two elliptic tides with speeds y-tr-ar and -y-cr + ar. The genesis of this compounded tide is explained in the Report to the British Association for 1883. Fig. 5. I accordingly thought it worth while to evaluate the Mi tide for each of the twelve months under reduction. The results come out sufficiently discordant to render it impossible to assign any definite value to the tide, yet there appears to be some sort of method in the phases. Thas the phases for the twelve months come out for 1902, 9°, -3°, -45°, 6°, -32°, 70°, 12°, and for 1903, 6°, -159°, -179° -42', -10°. Two of the phases, those for the 9th and 10th months, are very discordant, but for these months the C 10 amplitude of Mj is small ; it is also very small for the 6th month with phase 70°. The mean of all the other phases is such that k is pretty small, and this agrees with what is to be expected, because k for the tide is small. It thus appears probable that there has been a sensible disturbance from the Mj tide of the values of the mean heights of water as arranged in mean lunar time. It should be noted that the whole amplitude of oscillation is so small that it is really surprising that this eti'ect should be traceable at all. There is one feature in the results which is so singular that it is well to refer to it. If we look at the heights and phases of the Mo it will be observed that there is a progressive change both in amplitude and phase as the season of 1902 advances, and this change is repeated in 1903. Mere inspection does not convince one of the degree of regularity, and I have, therefore, prepared a figure which exhibits the march of H cos k and of H sin k. The values for each month may be taken to appertain to the middle of the month, and the points surrounded by rings in fig. 6 give the values for the season of 1902, while those marked with crosses give the values for 1903. The broken line shows conjectural curves which appear to satisfy the observations. The conjectural curves are such that (in inches) HcosK = 1-65 - 0-75 cos (i// + 2°), Hi = 0-23 + 0-53 cos (vi! + 79°), where >; is 360° per annum and t is expressed in months. There would thus be an annual inequality in II cos k and H sin k, and their mean values, viz., 1 •65 and 0'23 inches, would show that the mean lunar semidiurnal tide is expressed by H = 1| inches, k = 8°. The mean given previously as derived only from the observations was H = 2 inches, k = 10°. It will be noticed that the greatest retardation occurs about midsummer, and at the same season there is a considerable decrease of amplitude. It is almost impossible to believe that the thawing of the sea could decrease the amplitude of the tide, although it might possibly increase it. Dec I JanI Fig. 6. It would be strange if this result, depending as it does on 12 independent observations, should arise from mere chance. Yet there is no astronomical tide which can give an annual inequality in the lunar semidiurnal tide. I note that if the observations of 1903 were pushed backward one month the whole of the observations would fall into a more perfect curve. Hence, an inequality of 13 months would satisfy 11 the conditions more perfectly than one of 1'2 months. There is, theoretically, a minute tidal ineiiiiality of long period (Lai'I.ace's first species) with a period of U months due to the variation of latitude, liut it is difficult to see how any perturbation of the lunar semidiurnal tide could lie produced in this way. But if we have found a true physical phei\omenon, the same kind of effect ought proliably to be produced on all the other tides. Yet when the oliservations for the other tides are plotted out in the same way, the points appear to he arranged almost chaotically. It is true that some slight tendency may be perceived for an increase of amplitude towards midwinter, but the effect is too uncertain to justify reduction to numbers. A much longer series of observations would l)e needed to throw a clear light on the jioint raised, but the result is so curious that it would not have been right to pass it by in silence. Tidal observations were made at Ross Island (called Erebus Island on the memorandum) by Dr. Wir.sON from 2'' January 11, 1904, to 8'' January 13. The place of observation was some 40 or 50 miles to the northward of the winter station. As there seemed some reason to suspect a seasonal variability in the tides, it seemed worth while to compare with actuality a tide-curve computed with the constants derived fi-om the winter observations. A curve was therefore run off at the National Physical Laboratory for a few days beginning with 0'' January 11, 1904. Although the sites of the two sets of observations are not identical, comparison with actuality shows a satisfactory agreement. It is unfortunate that these observations were made just after the time when the diurnal inequality had vanished and was beginning to increase again; for at these times the agreement is liable to be imperfect between computed and observed curves. On these grounds no surprise need be felt on account of the fact that the semidiurnal tide is somewhat more clearly marked in the observed tide-curve than in the computed one, and that the whole range of the diiu-nal tide on January 11 was 3 inches greater, and on January 12 about 6 inches (out of 28 inches) greater than appears from the computed curve. The computed and observed times of high and low water agree closely with one another. We may, on the whole, accept these summer observations as proving that our tidal constants are substantially correct. The semidiurnal tides, although small, exhibit clearly another peculiarity; it is that (k of S^..) - {k of Mo) exhibits a seasonal change of roughly the same character in both years. In all cases " the age of the tide " is negative and its mean value is about - 4 days ; in other words, spring-tide occurs four days before or ten days after full and change of moon. If the phases of Mo and So differed by 180° we should have neaps at full and change, and springs at half moon. This case corresponds to " direct " lunar tide and " inverted " solar tide. In the actual case (k of Mo) - {k of So) = 370° - 272° = 98° ; thus the observations show a result a verj' little nearer to this condition than to the ordinary one where springs coincide with full and change of moon. The unusual relationship between the Mo and S2 tides is such as to make it worth while to examine what would be the condition of affairs in an ocean of uniform depth covering the whole planet. From the few soundings which have been made it would seem that the ocean may be about 600 fathoms in depth, although further north the depth appears to be considerably greater. I have therefore taken the formulae of Mr. Hough ('Phil. Trans.,' A, 191 (1878), pp. 177, 180) and evaluated the lunar and solar semidiurnal tides for an ocean of 7260 ft. in latitudes 60°, 65°, 70°, 75° with the following results : — Lunar Semidiurnal Tide. H of equilibrium tide 6 '052 eeutims. Factor of augmentation for dynamical tide . . I 1 '932 H of dynamical tide for ocean of 7260 ft. 1 11 '69 centims. "1 (direct tide) J j 4^ inches / C 2 i '324 centima. 1-196 6 -47 centims. \ 2J inches J 70°. 2 -832 centims. 1-098 3 '11 centims. \ li inches J 1 -022 centims. 0-755 1 "22 centims. "1 i inch / 12 Solar Semidiurnal Tide. Latitude 60°. 65°. 70°. 75°. H of equilibrium tide Factor of augmentation for dynamical tide . . H of dynamical tide for ocean of 7260 ft.! (inverted tide) J 2-816 centims. -6-441 — 18-14 eentims. "1 —7 inches J 2-012 centims. -4-390 - 8-83 centims. "1 -3i inches J 1-318 centims. 0-755 centim. -2-556 -1-003 - 3-37 centims. 1 ■ -0 751 centim. "1 — IJ inches / —J inch J We thus find that in these high latitudes the solar tide is more magnified than the lunar, and is inverted. Thus in latitude 60° the solar tide is much larger than the lunar and is inverted, whereas in latitude 70° they are nearly of equal magnitude and the inversion of the solar tide still continues. For an ocean of twice the depth both the tides are direct, and they are not so much magnified. Although the Antarctic Ocean runs all round the globe it is of course unjustifiable to apply these results directly to the oscillations of the actual ocean, but they serve to show that we have no reason to expect considerable semidiurnal tides so near to the pole, and also that the great discrepancy between the phases of Mo and So is not so surprising a fact as it might appear at first sight. It is useless to carry out a similar investigation for the diurnal tides, because the variations in the depth of ocean exercise so large an influence on the result. We know, in fact, that for an ocean of uniform depth the Ki-tide vanishes completely, and the 0-tide nearly vanishes. I find that the equilibrium 0-tide is 3i inches in latitude 60° and falls to 2 inches in latitude 75°. Thus the amplitudes of the diurnal tides observed by the " Discovery " are very much larger than the equilibrium values. The Scottish Antarctic Expedition passed the winter of 1903 in S. latitude 60° 44' and W. longitude 44° 39' at the South Orkney Islands; they were thus nearly opposite to the station of the "Discovery." Their station was well adapted for determining the general character of the tides in the Antarctic Ocean. The reduction of their observations was made by Mr. Selby at the National Physical Laboratory, and gave the following results : — M,. B,. K,. K,. P. 0. H 1 -522 ft. 172° -912 ft. 198° -249 ft. 198° -482 ft. 15° 0-161 ft. 15° 0-558 ft. 359° It will be noticed that these results are quite normal, save that the So-tide is rather large compared with M2, and there is a well-marked diurnal inequality. They acquire a special interest when considered in connection with the " Discovery's " results. We see that the semidiurnal tides are " inverted," but have little or no retardation, whereas the Mo of the " Discovery " is small, but " direct," also with little retardation. We are thus led to suspect that to the northward of the latitude of the South Orknej's, where the " Scotia " wintered, the semidiurnal tides are inverted with small retardation ; that somewhere between the South Orkneys and near to the Antarctic Continent there is a nodal line for the Mo-tide. There must be also a similar node for the Sj-tide, and we may, perhaps, suppose that the node of the S2-tide is nearer to Ross Island than that of the Mo-tide. When we turn to the diurnal tides we find an entirely diff"erent condition, for at both places the phases are virtually identical, and there seems a. prima facie case for maintaining that the phase of the diurnal tide throughout the whole Antarctic Ocean is approximately the same as in the equilibrium theory. I cannot venture to offer any theory in explanation of the greater magnitude of the diurnal tide at Eoss Island than at the South Orkneys. 13 II. TIDAL OBSERVATIONS OF THE "SCOTIA," 1902-1904. I. Analysis of the Observations. The anchorage at which the tidal observations here dealt with were taken was at the head of Scotia Bay, Laurie Island, South Orkneys, in latitude 60' 43' 42" S., longitude 44° 38' 33" W. The bay is about 3 miles deep and faces S.E. ; the depth of water at the anchorage was 10 fathoms, increasing to about 100 fathoms at the mouth of the bay. Here the " Scotia" was frozen in from March 25 to Novemlier 23, 1903. (See Map at end of volume.) The apparatus used consisted of a long wire fixed to the sea bottom by means of a heavy weight. The wire was carried over a davit by means of a pulley. At the extremity of the wire was another lighter weight, which rose and fell with the tide along a graduated wooden scale. The floe in which the "Scotia" was frozen moved with the tide, the height of which was thus shown by the position of the movable weight. The observations were made by the leader, officers, and staff of the Expedition under the direction of Captain Thomas Robertson. The period covered by the observations is from 10.30 a.m.. May 26, 1903, to midnight on October 16 of the same year. On September 4 the wire broke and had to be replaced, and at this point there may be a discontinuity of zero. The observations for the last month have accordingly been treated as a distinct series. The observations recorded give the height of the tide at every half-hour from May 26 to September 4, and at every hour from September 5 to October 16. Records of the barometric pressure and of the strength and direction of the wind are also available. On May 26 the height of the tide was noted every 5 minutes from 5.45 a.m. to 9 a.m. In analysing the observations hourly heights were used, commencing for the first three months from 1 a.m. on May 26, 1903 (13 h. May 25, local mean time). The harmonic components for which an analysis was carried out were M.2, 0, N2, S2, K2, Kj, P. For the first three components named the method of analysis was that used for the Indian Tides, and described in the 'B.A. Report' for 1883. The periods chosen were 86 M days, 89 days ami 78 N days, these periods being selected to minimise the effects of the S2, Ki, and Mo components respectively, and to be as large as the first series of observations permitted. For the components S2 and K2 and Ki and P the process employed was essentially that given by (Sir) G. H. Darwin in the 'R.S. Proceeding.?,' vol. LII (1893), p. 365, where a method is indicated for dealing with a short series of observations extending over a few months. In analysing for Ki and P, however, 27 days' observations in each month were employed instead of 30. [See Article by (Sir) G. H. Darwin in Appendix to 'B.A. Report' for 1886.] Each month's observations were separately analysed and the means formed as recommended on p. 367 of the paper referred to above. The most important feature of the method, in regard to the results of the analysis tabulated below, is that the ratios of the amplitudes of So and Ko, and of Ki and P respectively are assumed to have their theoretical values, and that the "lags," in the case of each pair of components having nearly equal speeds, are taken to be the same. That in this instance this theoretical assumption is in sufficiently close agreement with the facts may be regarded as established by the results of the analysis. The values of the tidal constants obtained from the analysis are given in the subjoined table, H denoting the semi-range or amplitude (in feet) and k the epoch or " lag " of a component : — 14 Ms. 0. Sj. Kj. K,. P. N. 1st montli < 2nd month -j r IT 3rd nioutli -j 4th month | ^ 1 , 1 -518 f 172' -3 1-537 172° -1 0-564 357° -7 " I 0-543 3°-l 0-897 200° -6 0-892 198° -4 0-953 197° -6 0-907 194° -7 0-244 200° -6 0-243 198° -4 0-260 197° -6 0-247 194° -7 16° 11° 13° 16° 446 8 487 3 473 8 522 5 16° 11° 13° 16° 149 ~1 8 162 1 3 [ 158 1 8 J 174 0-298 153° -5 MeanH Values K 1-522 172° -3 0-558 359° -0 0-912 197° -8 0-249 197" -8 0-482 14° -6 161 14° -6 0-298 153° -5 The agreement between the values of the constants in successive months is as close as can be expected and may be regarded as quite satisfactory. The zero readings (mean sea-level) for the successive months were found to be 4-37, 3 -61, 3-90, 3-00 feet respectively. In regard to the last value, it should be remembered that this belongs to the period following the breaking of the wire. It would .seem certain that there was an irregular change of zero due to shifting of the ship's position in the ice, consumption of coal, &c., and that from these values there is nothing to be inferred. As a check on the analysis the initial values for the various components obtained were calculated for h., June 1, 1903, and, with the aid of the Indian tide-predicting machine in the charge of the National Physical Laboratory, a curve was run off for the two months June and July. The observed heights for the five days beginning h., July 3, 1903 (astronomical time), have been plotted for comparison with the curve given by the machine. The two curves are shown in the diagram on p. 1-5 opposite. The curve obtained from the observations, where distinguishable from the machine curve, is indicated by a broken line. Crosses mark the observed points. The mean sea-level for the " observation " curve is that given by the mean of the five days' heights. The agreement over these five days is sufficiently close. It may be noted that where the two curves separate they could in each case be brought much nearer to coincidence by a change of zero, as distinct from the addition of another harmonic term, and that such change of zero, as already pointed out, is to be expected. The differences between the observed heights and the machine heights were also obtained for every hour on the following days: June 4, June 19, July 1, July 11, and July 24. The zero in each case is arbitrary, but if the analysis were accurate and complete the difference should be constant. The actual differences on these five days, however, would appear to indicate a variation having an approximate amplitude of 5 inches, and possibly of approximately diurnal period. This is not apparent in the comparison curves, and further investigation would be necessary before any conclusion could l)e arrived at in regard to it. With reference to the wind records, there is nothing calling for any special comment in coiniection with the tides. It is clear that the winds have not sensibly impaired the value of the observations, which appear to have been taken with great care, and have been found quite satisfactory for purpose of analysis. F. J. Selby, M.A., 1 Tidal Assistants at the National J. DE Graaff Hunter, B.A., f Physical Laboratory. December 20, 190G. 15 _5 S ^ 2 ■G -^ IS o .S 1-5 .? a "73 O 16 II. Discussion of the Preceding Eesults. The tides seem to be normal for a place in the Southern Ocean. The semi-diurnal tides are considerable, but the solar tide is unusually large compared with the lunar tide, the ratio being f, or 0-6, as compared with ■ 465 of the equilibrium theory. The semi-diurnal tides are almost exactly " inverted," so that low water occurs very nearly when the moon is on the meridian. The "age of the tide," or the mean interval from full and change of moon to springs, is (197°'8 - 172°'3) -^1°-016 hours, or 2.5 hours. This is a normal result, for the ages at Madras, Bombay, and Karachi are 29 hours, 32 hours, and 27J hours, respectively. The diurnal tides are well marked, as might be expected ; and it is interesting to note that they are " direct " and almost exactly in the phase indicated by the equilibrium theory. The age of the diurnal inequality may be defined as the mean interval which occurs after the moon has attained her maximum declination before the diurnal tide reaches its maximum. This is given by the excess of k for K] over k for 0, divided by twice the moon's mean motion. Thus, in the present case, the age is { 14'-6 - ( - T'O)} -^ 1°'098 hours, or 14^ hours. There does not seem to be any prevalent rule as to this "age" in India, for whereas at Madras the corresponding period is 14 hours, at Bomliay and Karachi this retardation is replaced by a small acceleration. These results are very valuable, as relating to the only ocean uninterrupted by land throughout the whole circumference of the globe, yet in themselves they do not seem to present any features of special interest. But they do acquire much importance when considered in connection with the very abnormal results obtained by the " Discovery," which I hope to discuss in the volumes to be devoted to the scientific work of that expedition. I wish to use the present opportunity of drawing attention to a mistake which was made in the article on Tides in the ' Admiralty Scientific Manual.' It was discovered by Mr. Selby when he came to apply the methods of that article to these reductions. The mistake occurs in the ' British Association Report ' for 1886, referred to above by Mr. Selby, and was carried on into the Manual. The principle of the method was given correctly in my paper in the ' Proceedings of the Eoyal Society ' for 1893, vol. 52, p. 365, but certain small changes are needed for applying the method to the case in point. I hope to correct the mistake in vol. 1 of my Collected Papers, which are to be puljlished by the Cambridge University Press, but it may suffice here merely to correct the errata in the Manual as follows : — At p. 63, — Far the tides Ko and S. In the formula for tan i/-, in the denominator, for 3 • 67 p, read 3 '71 p, for a fortnight's observation, and 3 '84 p^ for a month's observation. In the formula for H,,, wherever 3 '67 occurs, read 3-71 for a fortnight, and 3 '84 for a month's observation. The formula H" = — ^ H. remains coirect. 3-67 For the tides Ki and P. In the formula for H' the 3 in the numerator (but not that in the denominator) should be replaced by 3-007 for a fortnight's observation, or by 3-027 for a month's observation. The formula Hp = ^H' remains correct. For k' = Kp = f ' + V + <^ read k' = K-p = f'-|-V' + <^ + 6°-88 for a fortnight, and k = k^ = f + V'-|-<^+ 13° -29 for a month. The succeeding numerical example must be corrected accordingly. The only sensible change is that k' = Kp = 334° in place of 327°. G. H. Darwin. December, 1906. II. PENDULUM OBSERVATIONS. PENDULUM OBSEKVATIONS. I. Results of the ObserTations, by L. C. Bebnaoc'HI, F.R.G.S II. Dispussion of the Results, by C. Chree, Sc.D., LL.D., F.R.S. 19 RESULTS OF THE PENDULUM OBSERVATIONS. L. C. BERNACCHI, F.R.G.S. INTRODUCTORY. Among the investigations that the " Discovery " Polar Expedition had placed upon its progi'amme were those for the determination of the force of gravity in a high southern latitude. As there could be no question of anj'thing but relative determinations, it was decided, after consulting Professor \'0N Helmert, to apply for the loan of the Stiickrath pendulum apparatus of the South Kensington Museum. Instrumental Equipment. The apparatus was lent to the Expedition by the authorities of the South Kensington Museum. The complete outfit comprised a set of three quarter-metre invariable pendulums with agate knife-edges, to swing on three separate agate planes, an air-tight case in which they were swung, a dummy or temperature pendulum, flash apparatus, air pump, dry cells, and various accessories. The stand was a heavy metal one, provided with levelling screws, and arrangements for starting, stopping, raising and lowering the pendulums from the outside. A heavy case fitted over the pendulums and rested on a smooth level rim of brass at the base of the stand, which was fitted with a similar rim so that the rims came in close contact, a thin layer of vaseline being previously spread between them. The knife-edges were made horizontal by means of small pendulums with levelling tubes in their heads. Two windows in the case permit the mirrors at the top of the pendulums and the thermometer being seen. The bulb of this thermometer was inserted in the stem of a dummy pendulum of the same size and metal as the swinging ones and held in the case near them. The flash apparatus is for the purpose of observing coincidences between a chronometer and the swinging pendulum. An electromagnet in circuit with a break-circuit chronometer moves a shutter at the end of each second, thus throwing a flash of light through a narrow slit. The image of this slit is seen in an observing telescope supplied with a reticle of wires. These are so adjusted that when a pendulum is at rest the image of the slit coincides with one of the wires. When the pendulum is moving, the apparent position of the flash depends on the position of the pendulum when the reflection occurs. The period being slightly greater than half a second, the pendulum falls behind the chronometer at each swing. If s is the number of seconds lietween two coincidences, or the coincidence interval, then the pendulum executes 2s -I vibrations in -s seconds, hence its period = -j^ — - seconds. The Reductions to Standard Conditions. The eflfects of changes of temperature and pressure were investigated at the National Physical Laboratory (Kew Observatory Department), and coefficients were deduced. The following are the corrections employed, applying to all three pendulums the mean results obtained : — Temperature correction to reduce to 0° C. -o-oooooie-t/, where t is the observed temperature in degrees Centigrade. D 2 20 Pressure correction = — ^ ' ; ■^' . where p is the iiressure in millimetres of merctirv at 0' C. and t 1+0-00367/ ' -^ i ■> the temperature inside the receiver. The " pressure correction " really depends on the density, and so on the temperature as well as the pressure of the air in the receiver ; that is how - — AAop^-i ^''' accounted for. There should be an allowance made for the fact that the pressure p is due in a small degree to aqueous vapour, whose density is only | of that of air under the same pressure. But to make this small allowance one requires to know the vapour pressure, and of this we had no note. At low temperatures, however, the vapour present is necessarily very small. Fiate correction = - 0' 00001 157EP, where R is the rate in seconds per day, P the period of the pendulum in seconds. The rate correction is negative if E is positive or chronometer gaining. Sidereal time was employed throughout. The correction for finite arc of oscillation is given in various forms. That of the United States Coast and Geodetic Survey' is equivalent to - 0-01357 (w + »')(/>-'»' )p logio («/"') where P is the period of the j)endulum. Also _ semi-arc in millimetres at start distance between scale and plane on which pendulum swings in millimetres ' , _ semi-arc in millimetres at end distance as above The value of one scale-division on the instrument was 3 millims., the distance l)otween scale and mirrors at Winter Quarters about 2858 millims. No experiments were made in the Antarctic for flexure coefficient. As it is a function of the pillar, no certain allowance seems feasible. The types of j^iHar used were, however, sufficiently similar to make it uidikely that the neglect of a flexure correction causes any serious error. The above formuliB assume 0° C. and millim. as the standard temperature and pressure, as it is easiest to reduce to these even when the mean temperature and pressure of the experiments are widely remote from these \'alues. The Melbourne Observations {Norcmljcr 10, 11, 1901). The pendulum apparatus \\-as erected in the cellar of the Melbourne Observatory. Swings were taken with two pendulums, Nos. 37 and 39. It was found impossible to stop the leakage of air into the receiver when exhausted, and accordingly the observations were made at atmospheric pressure. The rates of the chronometer 6711 employed in the experiments were determined by direct comparison with the sidereal clock in the transit room of the Observatory. Oliservations with the two pendulums were also made by Mr, P, Baracchi, Government Astronomer of Victoria. The Christchurch Obsei;vations, New Zealand. Sets of swings were taken at Christchurch Magnetic Ob.servatory in November, 1901, before the sailing of the "Discovery" for the Antarctic, and again on her return in May, 1904. The Oliservatory is in latitude 43° '31' 50" S., and longitude 172° 38' 9" E, ; it is situated on a lai-go alluvial plain (Canterbury Plain), and is 25 feet above the level of the sea. The observations were taken in the Absolute Magnetic House, a small wooil building exposed to changes of temperature ; therefore the temperature during the experiments had a large range, and changed vapidl)-. There was no way of obviating this difficulty, no cellar being available. 21 Ohsfi-nifion.-: of Xuiriiil^r 20 mnJ 27, 1901. These were taken with all three pendulums, Nos. ."ST, 38, 39, swung at atmosphoiie pressure, the pressure being recorded l)y a mercurial barometer lent by Dr. Evans of Canterbury College, N.Z., particulars of the errors of the instrument being supplied. The cistern of the barometer was placed on a level with the pendulums. A break-circuit sidereal chronometer. No. 6711, by KuLLBEUG, London, was used throughout, and its rate was determined by telegraphic signals from the Astronomical Observator}' at Wellington, N.Z. Eleven signals were sent at intervals of 30 seconds, commencing at 9.30 p.m. on the evenings of November 25, 26, 27. The results arc as follows : — November 26, chronometer lost, in 24 hours, 0'87 second sidereal. 27, „ „ „ 0-66 The arrangement of observing was as follows : — Twelve coincidences were observed, six to the right and si.x; to the left, then an interval of fifty coincidences was allowed to pass, and then twelve more coincidences were observed. During this process four readings of pressure, temperature, and arc were obtained. The penduhmi was then left swinging for two hours, and then the same observations repeated. All three penduhmis were swimg in this manner. On the following day the case was taken off, the pendulums reversed on the agate planes, and the whole process of observing repeated. The same thermometer. No. 753 (Centigrade), was used throughout the observations in noting the temperature of the pendulums, of which the corrections were determined at the National Physical Laboratory. Ohxcrvationf: of May 30, 190-1. These were conducted in the same house as those of 1901, and the method of observing was the same. The pendulums, however, were not swung at atmospheric pressure, but at a considerably reduced pressure. The manometer employed was an open U-tube kindly lent by Dr. Evans of Canterbury College. The barometer used in connection with the manometer was No. C 895, by Hicks, London, whose corrections were known and applied. The rates of the chronometer. No. 6711, were determined by telegraph signals from Wellington Observatory, and taken as - 1 ■ 20 seconds daily. The Antarctic Observations. Winter Harbour. Latitude 77" 50' 50" S. Longitude 166" -t-i' -45' E. Olmerralions of July 31 and August 1, 1902. The pendulum apparatus was set up in a wood hut on shore at Winter Harbour in a small room partitioned off from the rest, where the temperature was kept as uniform as possible by means of lamps. A brick pillar firmly cemented was sunk below the surface of the frozen earth to a depth of 1 foot, and rose 3 feet above the surface. The section of the pillar was 2 feet by 2 feet. Upon this the pendulum stand was placed, small holes being drilled into the surface of the Ijricks to take the legs. The agate planes were then carefully le\'elled l)y means of the two small levelling pendulums, and the pendulums were placed in the supporting V's, No. 39 at the back, No. 37 on the right, and No. 38 on the left, the two latter being at right angles to the first. The light from the mirrors of the two pendulums at the sides is thrown into the observing telescope by means of two prisms placed near the centre. 22 • The flash apparatus was erected on a small l)Ox filled with cement, and therefore very heavy, which, in turn, rested upon a larger box on the floor filled witli heavy materials. The distance between the scale and the mirrors was 2858 millims. PreSf^ure. — The pressure was observed by means of a large manometer open to the atmosphere and therefore subject to the variations of atmospheric pressure. The manometer was tightly secured to a flat board, and a long millimetre scale of wood screwed alongside the tube. It was then placed upright close to the pendulum case, and the level of the mercury in both ends of the tube was read off by means of a T-square four times during each set of fifty coincidences. A mercury station barometer placed close to the manometer was read off simultaneously. The correction to the mercury barometer, - • 002 inch throughout, was determined at the National Physical Laboratory. The small and very delicate manometer supplied with the apparatus, and which was most carefully packed by Messrs. Negretti & Zambra, London, and sent out to New Zealand by mail boat, was found Interior of OUservation Hut, showing — 1. Marine barometer. 2. Observing telescope and flasli apparatus on opening the case to be broken. The manometer employed was constructed by Engineer Commander R. W. Skelton, R.N., the chief engineer of the " Discovery." At Kew Observatory, in 1901, some trouble was experienced with the "air-tight" case. At Melbourne and Christchurch, in the same year, it was found quite impossible to reduce the pressure to anything like 60 millims., so that the pendulums had to be swung under atmospheric pressure. The whole weight of the extremely heavy stand and case is supported by only three comparatively slender screws, and the metal rim above the screws was " pressed up " and the contact between the rims was no longer perfect. Mr. Skelton, by means of a surfacing plate, actually found this to be the case. He therefore re-surfaced both rims, which process considerably mitigated the evil. Indeed, it appeared at first as if we should be troubled with only a very small leakage, for when the pressui-e within the case was exhausted in 23 15 minutes to 50 millims., in the subsequent '25 minutes tliere was not a leakage of half a millimetre. On returning to the instrument on the following morning (Julj- 30) it was found, however, that the leakage had amounted to 280 millims. in 11 hours 12 minutes. Rates in the Antarctic. The sidereal chronometer No. 6711 by Kui.LBERG was employed and connected up with the electro- magnet, the current being supplied by two Obaeh dry cells. A portable transit instrument was set up in the Absolute Magnetic House for the purpose of observing stars at the commencement and ending of the observations for determining the rate of the chronometer ; but the persistent bad weather made these observations impossiljle, so that the only rate availalile for the Inteiior uf (Jbservation Hut, showing — 1. Open U -tube manometer. 2. T-square for reading-otF same. .3. Pendulum a])paratus. 4. Brick pier. b. Exhausting pump. sets of pendulum observations taken on July :il and August 1 is that oljtained by comparison with the " Discovery's " marine chronometer on board. The rate of chronometer No. 6711 in England, Australia, and New Zealand was small and uniform, and the comparisons with the "Discovery" chronometers in Winter Harbour during 1902 and 1903 indicate that its rate still remained small and uniform. Soon after the "Discovery" had reached her winter quarters this chronometer was set going in the Physicist's cabin, and compared from time to time with the marine chronometers in charge of the Navigator, Lieut. A. B. Armitage. The average temperature within the cabin was 52° F., seldom varying b' on each side ; although the temperature in the observing hut was much lower, it was kept as uniform as possible, and the temperature of the chrcjiiometer frequently noted during the observations. 24 Observations of Fcbruari/ 1 to G, 1903. Observations were taken under exactly similar conditions to those taken in July and August, 1902. The pressure was about the same and the temperature fairly uniform throughout. The method of noting coincidences and the distance of the flash apparatus from the pendulums were the same as previously. The flash apparatus worked well. When completing the last pendulum swing the vacuum suddenly failed. On taking the ease off, small risings were found in the rims of the base near the supporting screws. The rim had to be re-surfaced, which process delayed the completion of the observations until February 6. Observations of September 5 and 6, 1903. The pendulum apparatus was set up in a similar manner and in exactly the same spot as in July- August, 1902, and February, 1903. The distance between the pendulums and scale was a little greater than before, viz., 2932 millims. Preparations were made to commence observing on the 4th. Observing was commenced at about 10 a.m., with everything in good adjustment and the temperature of the chamber + 1° C, but, unfortunately, soon after starting, the mirrors and prisms under the receiver liecame " fogged " and covered with moisture, and observing had to be discontinued. The thorough drying of the pendulum apparatus and readjustments occupied the rest of tlie daj', and observing had to he postponed until the 5th. An unsuccessful attempt was made during the evening of the 4th to take transit observations of stars for determining the rate of the chronometer. The portable instrument supplied to the Expedition was not of a very satisfactory character, the axis or pivots which rest in the V's being so worn that the telescope would not remain at any required altitude. The low temperature — nearly - 40' F. — so contracted the spirit in the striding level that the ends of the bubble could not be seen, and therefore no value for level error could be determined. The cold was also extremely trying to the observer, who was compelled to give up the attempt after about two hours. The subsequent nights were overcast. The rate of the chronometer was therefore determined, as on the previous occasions, by comparison with the " Discoverer's " chronometers. Complete sets of swings were taken on the two following days, September 5 and G, Engineer Commander Skelton rendering assistance throughout and observing independently. Local Geology. The Winter Quarters were situated near the end of a peninsula running out in a south-westerly direction from the base of the island formed by jNIounts Erebus and Terror. The peniu-sula is about ten miles long by a mile broad, and has an average height of 600 to 700 feet. The rocks of which it is composed are of practically three varieties : — 1. A yellow breccia, which occurs in three well-marked heights, the nearest of which is three miles distant from AVinter Quarters and 1400 feet high. This rock docs not appear to be developed to any great extent, but occurs as a volcanic pipe surrounded liy the basalt which forms the major part of the peninsula. 2. The trachyte found on Observation Hill, a hill three-quarters of a mile distant from the ship and 750 feet in height. This hill is conical in shape, the upper half being composed of a trachyte of specific gravity 2*2, and the lower half of a lava containing lapilli of a very varying composition and with a specific gravity of 2 '87 in one case, but on the south-east side of it there is a rock of greater specific gravity. 3. A black basalt, which is by far the most important rock both as regards its development and physical properties. It forms the hill called Harbour Height and reaches from Hut Point to the base of Castle Kock, if not beyond it. It forms three-quarters of the rock of the peninsula, and rises to an average height of 700 feet between the two points mentioned above and lies roughly perpendicular to the magnetic meridian. It has a specific gravity of 2 • 9, and under the microscope shows frequent plates of magnetite. Hut Point is entirely formed of it, and it is on this rock that the pendulum observations were made. 25 Further uticKl, Mount KroUus vises ;is :i full-l)()ilio(l cono with its 1);iso 12 miles and its suuunit 20 miles distant. The rocks found at its base have a specific gravity of approximately 2-9. The mountain lies north hy east of Winter Quarters, and is 12,000 feet high. Mount Terror, lying -tO miles east of the ship, is only 10,000 feet high, and is rnmposod of basic rocks of speeifip gravity 2-0, with small local intrusions of trachyte of a specific gravity of 2-4. The two are joined by a ridge some 8000 feet high and of similar rock to that which forms the nias.ses of Ereljus and Terror. There i.s no important land development, to the southwaid, there being only two islands undei' 3000 feet high and composed chiefly of l)asalt of sjieeifie gra\ily aliout 2-9. These are resjiectively 15 and 20 miles distant from the ship. Moinit Discovery lies south-west from the ship at a distance of 30 miles. It is also conical, with a height of 8000 feet and the diameter of the base some 10 miles. It appears to be chiefly composed of the basic rock so common in this locality. Turning to the west there is a totally different development. A great mountain chain running nearly due north and south lies at a xlistance of 40 miles from the ship and rises to heights of 12,000 and 13,000 feet, and is on an average 11,000 feet high. This chain is composed of granites, diabases, and quartzites. The granites form the core of the chain and rise to a height of 4.500 feet above sea-level. They vary in composition and have a specific gravity between 2 '6 and 2' 7. Above this occurs a dialsase up to a height of 8000 feet. This rock lies practically horizontally on the plutonic rocks (though interrupted by faults) and has a specific gravity of roughly 2 • 8, while above it and also horizontal a sandstone occurs which has a specific gravity not greater than 2 • 67. This practically completes the series of the Prince Albert Mountains. Near Hut Point the soundings showed McMurdo Sound to be comparatively shallow. The water quickly deepened from 2 fathoms at Hut Point to ISO fathoms a mile further out in the Sound to the west, while 10 miles away to the west-north-west the sounding was 100 fathoms. The deepest sounding was 400 fathoms at a point 2 miles south-east of Observation Hill, and other soundings showed that the water was much deeper to the south and the south-east than to the north and north-west of Winter Quarters. The Sound therefore may be taken to be 25 miles wide with an average depth of 200 fathoms. The ship anchored in Winter Harbour had 9 to 11 fathoms of water, while on the north side of Hut Point the water quickly deepened to 50 fathoms close in to the land. Hut Point itself is continued half a mile to the south-west below water in a shoal which gives soundings of from 2 to 25 and 40 fathoms. Most of the above geological information has been kindly supplied by Mr. H. T. Ferrar, Geologist to the Expedition. It may be possible from this geological information to investigate the effect of topo- graphical irregularities and determine a correction for the density of the local rocks. Concluding Remarks. In the following reductions the period has been determined for each separate set of 50 coincidences, and corrected independently for temperature, pressure, &c. In the reduction of the pressure observations the temperature of the mercury in the barometer has been assumed to be the same as that of the mercury in the U-tube, since they were quite close to one another. The difference in the height of the columns was then corrected for temperature. It must be borne in mind that the observations of July, August, and September were taken in what is practically the middle of the Antarctic winter, or, more correctly, during the coldest months of the year. The February observations are at the end of the comparatively warm summer, when the pack ice has moved far to the north, and after huge masses of ice have been shed from the glaciers and the great ice fields of the Antarctic lands, and in turn drifted away to the north. My thanks are due to Engineer Commander R. W. Skelton, R.N., who not only rendered most valuable services in assisting to set up the pendulum apparatus, &c., but soon made himself thoroughly competent in taking observations. Independent sets of swings were taken by him on every occasion at Winter Harbour. E 26 -spuoods III pd)ajj.ioa poujj •5083144 123 159 195 -5083067 085 103 054 -5083341 322 311 344 -5083156 165 159 155 -5083090 132 089 103 •5083380 288 339 319 Corrections (seventh deciui-al place). -SUOI) -aaj.ioj JO umg IN X -f O CD X ^ X 03 IN 01 -f ■* Tf + + + + + + 1 1 cr. 03 oS + + ■(- -f 03 C -fi -fl I- N -1 + + + + 03 CD ^ -I -H -H- 1 ^ o 'J'oq X X t- + + + 1 -a^ua -? -? -^^ -i" + + -<- + ^ ^ !? ^ '•?'-? ^' -? + + -f-i- + + ■^ + l^ 1^ 1^ 1^ + -f + -f + -f + -f 1> 1^ t^ 1^ •* -)■ -t -t + + + + -ajnssajj ^ -f 03 -t 10 03 1(1 03 1 1 I 1 lO --C' CO CO UO 05 lO C3 1 1 1 1 lO O O M -* 03 lO 03 1 1 1 1 - 58 -101 - 55 - 98 - 59 -100 - 57 - 96 lO 03 lO 03 1 1 1 1 -9jn)t!.i3dmax + 105 + 79 + 30 + 49 -1- 65 + HI + 53 + 1 -H199 •t-134 + 92 •1- 54 + 98 -1- 77 + 25 •f 53 -H CO C O CD CD ID + + + -fl90 -1-128 + 87 + 49 •o.iy 03 -Tfi ■■* 0^ 1 1 1 1 ^-*;dco CO-I-.OCO 7 1 1 1 1 1 1 1 X CO CO rt 1 1 1 1 O ^ CD N 7 1 1 1 ?5 rHCD N 1 1 1 1 -spuooas m pjjodjjooun poujj -5083052 095 135 195 -5083021 077 062 102 -5083143 232 228 338 •5083077 145 145 ]54 •5083051 126 055 154 -5083196 208 265 321 -saj}oum|nu u; ojnss^jd pajoj-uoj 56-0 104-8 55-3 105-2 61-2 107-7 59-2 104-3 50-1 100-1 55-4 103-0 «pMCO 1 xcopco -ji N IN O ' lO -N ^ X CDrHCOr-l CDr-HCOO 5-801 9- 09 5-901 0-93 1 "^3 ■i«"Ta ■*p-1it- x^-pp C3XXi— 1 I— lO'-^^O IMrHrtrH COWMl-l 17-7 11-1 24-3 15-1 27-0 17-7 17-8 11-8 30-1 19-8 23-4 15 -0 ■* p 00 -31 ■FRFI 31-5 21-6 21-0 13-6 34-8 22-5 26-4 18-0 19-8 12-9 27-0 18-1 30-7 18-9 19-2 12-7 33-3 21-9 26-2 16-2 8-91 3.95 0-5T 5- 61 -oan4«addiua4 paja^jjog "C. -2-27 -1-71 -0-64 -1-05 -1-41 -1-32 -1-15 -0-03 -4-29 -2-89 -1-99 -1-17 -2-12 -1-66 -0-.54 -1-14 -1-32 -1-36 -1-07 -fO-01 -4-10 -2-76 -1-88 -1-05 -spuoaas 30 -6010 •5861 -5709 •5495 3883- 8869- 8569- 9519- 08 CD t- C33 rH 30 -5923 •5680 -5678 •5679 30 -6021 •5748 ■6004 -5646 30 -5493 •5455 •5243 •5043 •uonisoj O ::« = fi c^p^ : fi =M = e =f^ :: P -M = ft =« = ■uminpuaj si I 1 ?; X 03 CO ! CO s S 1 g Bernacchi g Date and place. c 27 -50832M 238 242 280 IN O CO jH 0> M •5083373 380 390 407 -5083230 222 255 274 rH t^ r^ CO t^ 05 -5083360 386 383 376 r-i CO i rH t~i 05 CO •^ CO CO CO W N + 26 - 40 - 18 -140 O ^1^ + + + iH iH »-( rH + + + + + + + + ^ f-i r^ C^ t^ l'- + + + + + + + + + + + + + rH r-t + + CO CO O CO 10 03 CD m f-t MM 1 1 1 O t^ '* CO CO O lO 05 MM X X 05 IN O Oi to CD MM X rf CO Tjl X CO 1 1 1 CO O C35 I:^ CD rH lO 05 MM lO r-A • O 05 1 1 CO o 1 1 lO X O X 1 1 fS2 1 1 1> lO O U5 rH FH »1 lO + M 1 X -1 X CO CD lO 1 1 1 + + + 1 IN t> iH CO ii 1-1 oq lo + M 1 g3S 1 1 1 -H oq IN lO CO CO IN + + + 1 + + 532 + + X CO ^ CO rH T-{ + + X t> CO lO r-i 1-1 + + 05 CO 05 CO MM C5Tf(t> 1 1 1 O CO r-< -* 7 M 1 X CO X CO MM 05 oq t^ 1 1 1 C5 ■* O ^ M 7 1 IN O 7 1 CO IN ' 1 rH ^ 1 1 IN lO rH 1 1 •5083219 278 260 420 IM ■* O t^ X I> rt (M rj 05 LO O CO CO ?3 n -Ji >-< CO X cq oq N N -t CO CO CD X lO O rH oq N S S S 3 CO CO CO -T C^ O CD X CO OS o is g5i <-( N X rH i>IN O r-i 60-0 105-1 67-8 172^1 M p oq ^ S CO CO ip p -^i 1^ o ^ ^ CO (M CO O X X rH 00 10 O X IN CO rH I> X X rH N lO 05 t- 00 rH 00 CO O -* CO 05 I> C^ CD O OS 05 8S o t^ S2 00 p rH X CO OS is 28-2 17-1 29-2 16-6 Ip X 05 OS 05 lO oq i-H oq M t^ 00 C3 OJ> rH X CO r-l CO rH t- -fl P 05 CO «> O lO IN rH IN rH rH -^ 00 IN Jh M ^pcpp 05 X O X Oq rH CO rH CD ^ CO •'Jl T}l -* Oq rH rH p eq X 33-3 20-8 33-4 20-8 M IN 05 CO IN oq IN ip 05 m ^ 35 CD IN CO rH CO oq 05 IN iH 05 O 05 N X CO rH CO rH ip ■* IN m IN oq P ip rH «J CO O 10 -1 CO oq CO IN iO I;- CO W p X p p CO N CO eq CD M CO X CO CO n< -H O O O -H 1 + + + + + + lo -* oq CO CO t- ip o rH O O O 1 M + N ro ■* N O O O rH 1 + + + W rH 05 00 m jq O rH rt + + + rH O O O M 1 + 8g coco CO CO ■* ■* 1 1 OS t^ r* ^ CO oq 1 1 CO X p n CO CO 1 1 30 -5408 •5198 -5259 -4684 05 CO CO t-t- oq O 00 o: ir> »o -^ Tp -^ CO CO X 05 o CD X X o CO OS "* 05 »q CO O rH rH 1-^ 05 t^ lO ^ -Jl 'I' rH CO X S k' as 00 p o S OS ri< CO 'ji m r^ CO OS t^ ip »p o q srt s p, =« R 'f^ = fl =« = O =rt fi 'f^ - « = « = !A s rt = CO ^ 05 fe X CO i> X CO CO CO s 8 a g Hi M 03 a o 5 M t35 r* 28 -spuooas up po}oo«oo pouoj CO CO •5083118 123 157 144 •5083089 036 049 144 •5083300 348 260 286 ■5083135 179 115 145 •5083037 072 073 065 •5083278 314 320 290 "El 1 1 > •1 J ■snoi) -oajjoo JO uing + -f 01 rH ^ f-H rH rH + -^ 00 -# C5 CD 0; CO rH IN + -f 1 1 10 CD I- l^ J> + + + + CD CO CO CO + + + + CO CO CD CO l>I> Ot- 'H + + + CO CC CO CO- t^ l' t> J> + -(- + -f •ajiissajj CO 1 7 00 rH in 05 1 1 01 CD Q CO T? 00 ^ J> 1 1 1 1 CO IN CO 05 'J CD Hft- 1 1 1 1 t^ ^ CO rH ■^ -Ji -i< tXl 1 1 1 1 1 1 1 1 1 1 1 1 1 7 1 1 •ajntujadmaj^ ft r^ + + -* 05 on CD + + 1 1 + + 1 1 !>'? O CO CDrH.0^ + + 1 1 -* O CO CO I> 00 ^ (N + + 1 1 IN t- rH O IN t-T}i^o + + 1 1 + + 1 1 ■3.iy CD IN 1 1 ■* 7 1 lO I> CO 7 1 1 1 J> CO 00 ■# 1 1 1 1 O -* t> CO 7 1 1 1 O T? CD CO 7 1 1 1 .t>coi^eo Ci-'J'ccco 1 1 1 1 1 1 1 1 c •spuooas ui po^oajjooun poi-taj cn CD 00 p> CO 05 CD 05 lO CO W CO C5 05 IN § i -fico-*o rHi^ino •-I >n XI •* -frHCOl^ 00 00 s i s l^l> lO -H CS rH O CO 00 00 IN UO N OO 1 ip o cS I3 -sai}Dnirj|itn n: ajuseajd pa^joajjog oq 05 is p (N S 1 (T. ipeqrH Mt-eqc i-*iNsp-* COCDIO-M rHiNrHCB INlOOOrH ^05t}ioO IIOOIOOO >00^05 rH r~t pt-ipt- t- IN >p 05 1 -* rt p ■* COCOCOCO OOrHTflin >nOin05 lOrHU305 1 f-i i S •inm^ CO rii IM r-l 05 9 « C ip o CO rH N rH rH 05 CO gi 03^10 CO rH IN rH ip 00 ip ■* in xi in t^ IN rH 0.1 rH rH !E 00 -^ 05 X CO in IN rH N rH ■im)iui CD rH IN M CO ?5 0: X ip cpi> OD ^-cp 05 p CO M (N rH rH p P p C? IN t^ 00 CO IN Jq rH « P rH ip t- 00 05 05 (N rH IN rH rH t- OD ip IN O •^J> CO M IN rH 1 t-H •a.miutadnia^ paqoajjOQ Q V=9 ss OOCOIO COOrHCO O5U3Q0CB C0l:^0»ra J>i>a5rH [o-iPOOS gffSS SS882 >0 IN CO 05 N >n rH 05 ~ Tf -* 1 1 CO N 1 1 rH rH rH 1 1 + + (NrHOrH 1 CONrHO rHrHrHO NrHOrH 1 COMrHO ll++!||-4-+ II++ ll+-f'll+ + 1 1 -spuooas ni ^uAa3:)ni aauapiouiog 00 CO CO ip ip CO 05 I^ ta 10 oq CO 10 2 rH CD lO CO 00 IN rH 05 lo in lo •* CO CD lO CD >0 05 rH 00 CD in m U5 k' ' ' CO rH OOt^ 00 •* in 5^ CO S in in §' ' ' o •noi^isoj w = 1:^ = R =P? = O ::« - fi ;« :; fi :;« = P rP^ - P r« = •uminpuaj 00 CO CO fe S 05 ^- CO CO CO 05 t 1 i2! H ►J M 00 K -< « |5 O g M 03 1 a 1 1903 February 6 (continued) at Winter Quarters September 5 and 6 29 2S CO (M CO CO CD O X N W X ^ M l~ CO X t^ in 'f ^ 1 CO 0) CO ?3 p eq in 'i' S§ 'O -f X (N CD 05 X •* o S 1 W X M 'N o; X X X in in rj. .^ t- ?- W l-r m 05 03 05 OS 05 s a X X X X 00 o o O o o o o o >p ■P m '•" '•" :S S CD -H O 03 CO 5-1 M t^ o r^ o: o CO oj '; 9 9 Q CO M I~ =? ■# IN -M rH 05 CO in l» CO X t^ Ol 1~ CO n CD r^ CD -P CO oo rH a> rH M 00 CO CO CO -f in rH 05 05 05 05 05 + + t^ t> o in CO CO CO CO CO TO in m in in r-w i>i> t~ t^ i^ t~ t^t^ l^ I- + + + + + + + + + + + + + + + + + + + + + + + + + + «>t- ^ N CO CO O I^ CO 05 CO O O CD CO t^ (N 05 CO X 1^ ^ OS CO O CD 52 =2 COCO-fCO COIN^JCI m -? -* CO in o in o in S CO rJ 50 O 1 1 CD CO CDCDCDCO COCCaSCO 1 1 1 1 1 1 1 1 CD CD CO CO 1 1 1 1 1 7 1 7 1 1 1 7 1 7 1 7 1:- •* lO LO CO CO X CO O CD X 33 CO CO CO in -H in tP C35 lO i^ rH X I'- » CO-H t» rH O -1 .-1 in rt i> l^ ^ X X N rH CO CO X CD CO CO 05 X 1 1 00 00 1 1 X X CC CD 1 1 1 1 t^ X X 03 1 1 1 1 in CO CD t^ 1 1 1 1 in CD 1 1 CO CD 1 1 1 1 in -.c 1 1 CO CO 1 1 CO * 1 1 1 1 1 1 X r-l t- rH 1.0 rH in rH 1 1 1 1 1 1 1 1 1> rH m rH O ■* 1 1 1 1 7 1 05 « 1 1 03 CO 1 1 05 CO t 1 1 00 CO 1 1 05 CO 1 1 -* O CI CO 1 CO rH o in o •* X t^ I> r^ 05 CO OS OS rH^ CD 05 t- ca CO X t^ 05 X in O IN I^ CO 05 ■* Tjl t^ (M (N lO lO rH O rH rH O C^^ O Cvi O *l iH ■— 1 rH N a N rH rH rH rH , O o o o a> OJ 03 i o; 05 1 Oi 05 o o o 1 o o 1 o O in 'P '■" « V ■* CO p m rH^lf-HM rHOSCpCp CD 05 -ji in o eq 05 IN t , ■* ip m 9^ 01 p CO J>J> a>i> t> t~ t- t- t- t- 1^ l> t~ t- C~ W rH p 00 ^"P incpcorH riof? epx-*(N , int- N in 9 ■? ^f ri ?> epos 00 t> t-t- inci-jx oojoco -txot- rHX X t^ O l> 0> CO CD to t~. in CMIN INC<1 0)?J COrH IN rH CO rH Cq rH IN rH T' V

lO lO l> x-jicqo OC0005 cocooso^ (Nin t^ X 05 CO in cq M in to l^ r"P « in ip ■* CO o mmt-x mxi^gj cq ^ CO t- N p ■* CO t* h^ ■? ^ 1>1> i> J^ t^XCO'tfl 1 ini>XO CqcO^CO rHCO CO CO X O r^ CO CO CO X o l-l l-i r-i t~< ^ r^ r-i r-t ^ ^ r-i C^ I rH-HrHrH rHrH r-t r-t r-i rH + + + + + + + + 1 + + + + 1 + + + + + + + + + + + + + + + + COI> (M t~ o (M CO o t^ o o m m o CO r^ o co IN X 05 CD CVI i-i rH X CO CO CO m ^ O! 0>'-H'»'N inrHCOCO OlXrHCO C003 Hp X J> IN in X tX Tf 05 O) O CTl COXrHO 00305CO CO-fliMrH lOOJ •* CO in 05 ■* O CO IN Ip ip ip -^ roospp ogspp m 5 oa OS M -h IN |^l M IN (N IN W r^ t^ 1> t^ X X 1^ I^ X X X X X X oq ■M cq oq O) IN oq oj " IN (N P = fi = q =Ph : fl =M = O =« = n = fi = R ~ fi = R = R = t^ 03 t~ X 05 i> X 03 £: X 05 CO CO CO CO CO CO CO CO CO M „ M n H a ^ o t> o o s ■< -< g »; |2S ^ 14 M M n » QQ m m m ^ 1^ cq S a a h ig 53 C a 3 jSl rH CD -r o ^ .2 m o g-^ '"'•ss '"' -o ^ si; ao j O r i"S ' -§ |Zi ^ 30 II. DISCUSSION OF PENDULUM RESULTS C. CHREE, Sc.D., LL.D., F.R.S. (from the national physical laboratory). § 1. In drawing conclusions from the pendulum observations made during the British Antarctic Expedition of 1902-1904, due allowance must be made for the conditions under which the work was done. Those responsilile for the Expedition found themselves shortly before its departure without a physical observer. At the last moment, Mr. L. C. Bernacchi consented to fill the breach, and in the very short time that remained he did all that was possible to obtain familiarity with the instruments. He had fortunately had a good deal of previous experience in observing times of vibration in connection with magnetic observa- tions, and the observational results obtained by him and Mr. Skelton during the Expedition appear as consistent as could be expected under the conditions of observation. The apparatus had arrived at Kew some time before Mr. Bernacchi joined the Expedition, and the pendulums had been swung by Mr. E. G. CoNKTABLE, senior assistant in the Observatory Department, in order to obtain their periods. Even then some difficulty was experienced in getting the cylinder containing the pendulums to remain air-tight during the observations, which were taken at a pressure of about 60 millims. of mercury. Greater difficulty was experienced during Mr. Bernacchi's introduction to the instruments, but this was attributed to the fact that meantime the apparatus had been dismounted and had been somewhat hurriedly re-erected. The defect, however, as explained in Mr. Bernacchi's introduction, proved even more troublesome in the Antarctic. When observing, Mr. Bernacchi's usual practice was to set the pendulum swinging, and then shortly after take two sets of observations of the time answering to 50 coincidences, one set with the pendulum moving in the one direction, the other with it moving in the opposite direction. The pendulum was then left swinging unobserved for about an hour and a half, and thereafter two other similar sets of 50 coincidences were taken. The mean time of the two sets of observations differed by about two hours, and the leakage was such that in this interval the pressure inside the receiver rose on an average from about 60 millims. to 110 millims. The leakage was not conspicuously worse during anyone set of experiments than during the others. During each set of coincidences four readings were taken of the pressure, and the arithmetic mean of these was accepted as the pressure of the observation. The " pressure correction " is, within the limits of accuracy of its determination, a linear function of the pressure, and the rate of leak would normally be nearly uniform during the time occupied by a set of coincidences. Thus the fact that the cylinder was leaky will presumably have made little if any difference in the accuracy of the mean final values ; but it is probably in part accountable for the somewhat large discrepancies occasionally apparent between the results of the different sets of coincidence ol)servations with the same pendulum on the same day. Difficulty was also experienced in connection with the temjierature. At Winter Quarters, whilst the regular diurnal inequality of temperature was small, large sudden changes were not unusual. The room in which the pendulums were swung usually varied very perceptibly in temperature during the observations, and the temperature in different parts of the room {e.g., beside the pendulums and beside the barometer) sometimes differed rather largely. The change of temperature in progress during the observations was sometimes a rise, sometimes a fall. A change of 1° C. in the temperature of the pendulums means an alteration of 46 x 10"" second in their time of swing. Thus a very little error in the temperaturt assigned has an appreciable effect on the period. Here, again, there was probably little, if any, effect on the mean final values, but there was unquestionably in the temperature variations an active source of irregularity lietween the different individual results. 81 § 2. Some other sources of uueertaiiity remain to l)e mentioned. Under Antarctic conditions, with the instrumental outfit supplied, it did not prove possible to take astronomical observations of sufficient accuracy to determine chronometer rates from day to day, with the high precision desirable for pendulum observations. From time to time, at intervals varying from 8 days to 4 months, stellar and solar observa- tions were taken by Ijieutenant Armitage with a theodolite, and from these he deduced the error of the chronometer " A," which served as the standard to which the others were referred. These observations were carefully made, l)ut between the dates of two successive oliservations tlie rate of the standard A had to be assumed uniform. During the pendulum observations, Mr. Bernacciii compared the chronometer, Kullljerg 6711, used in the pendulum observations, with A, through the intermediary of a chronometer watch ; this was compared with A in the ship, and with 6711 in the observational hut. The rates accepted for 6711 thus depend on the accuracy of the comparisons with it and with A of the chronometer watch used as inter- mediary, on the steadiness of A, and ultimately on the accuracy of Lieutenant Arjiitage's observations. To reduce the uncertainties of the comparison with the intermediary watch, Mr. Bernacchi's usual practice in the later observations was to compare it with 6711 after an exact 24-hour interval by the watch, taking seven successive readings of the watch at 10-second intervals, and estimating the corresponding times on 6711 to O'l second. Different watches were employed during the different sets of observations. That employed in September, 1903, had the steadiest rate, and the uncertainties as to the rate of 6711 were then probably least. As to the accuracy of Lieutenant Armitage's observations, one can form an opinion only from the greater or less apparent regularity in the results. For some time after its arrival at AVinter Quarters, chronometer A seems to have gained slightly. It then began to lose, and continued to do so during the remainder of the time. The losing rates deduced from Lieutenant Armitage's observations varied as follows : — Eiite of A losing. . 0'60 seconds . . 0-90 „ . . 0-75 „ . . 1-35 „ • • 1-12 „ . ■ 1'17 „ . . \-l-2 „ The apparent irregularity in the rate about February, 1903, is suggestive of some uncertainty in the observation on February 4. The rates actually assumed as applicable to A during the pendulum observations at AVinter Quarters were : — July to August, 1902 -0-G seconds February, 1903 -M „ September „ , . - 1 • 1 ,, These assumed rates are hardly likely to be affected by any large errors. At the same time it is impossible to feel absolutely certain that an error as large as • 5 second may not have existed, especially in the result assumed for February, 1903. The rates finally deduced for the i)endulum Chronometer 6711 during the obseivations at Winter Quarters were : — July to August, 1902 -0-8 seconds February, 1903 -1-2 „ September ,, -1'3 ,, Chronometer 6711 when at Kew, both before and after the Expedition, possessed a very steady rate, so that the uniformity in the above results is at least in harmony with its general character. May 13, 1902, to August 7, 1902 August 7 „ October 26 „ October 26 ,, Februai'y 4, 1903 February 4, 1903, to Feliruary 21 ,, February 21 „ March 1 ,, March 1 ,, July 6 ,, July 6 „ October 8 „ 32 An error in the r.ate accejited for 6711 of 1 second per diem means an error of approximately 59 X 10^" second in the time of swing of the pendulums. § 3. A second source of uncertainty, already alluded to by Mr. Bernacchi, is the absence of observations for determining the so-called " flexure " correction, which arises from the absence of alisolute rigidity in the pendulum and its supports. Flexure experiments were made at Kew before the E.xpedition set sail, and after its return, but the results apply strictly only to the conditions existent at Kew. It has seemed on the whole best to apply no flexure correction to the results obtained during the Expedition, and to compare these results with those obtained at Kew, also uncorrected for "flexure." This is equivalent to the assumption that the " flexure " with the piers used at Winter Quarters, Christchurch, and Melljourne was the same in each case as that with the pier used at Kew. Judging by Mr. Beknacchi's descri^ition and the photograph, the pier used at Winter Quarters was fairly similar to that used at Kew, so it is probable that the plan adopted will lead to but little error. The " flexure " tends to lengthen the time of swing, and so, if uncorrected or underestimated, leads to too low a value for ;/. Whether the method adopted is equivalent to an overestimate or an underestimate it is, of course, impossible to say. On the Kew pier, on the average of experiments with three pendulums, the flexure correction to the period was 69 X 10~" second. The error in g corresponding to the total omission of a correction of this size is approximately 0"027 C.S."^, or about 1 part in 36,000. The error actually arising seems hardly likely to have exceeded a third of this, and may, of course, be absolutely nil. § 4. The last source of uncertainty to be mentioned is the fact that during their three years' absence on the Expedition the pendulums seem all to have altered slightly. The apparatus had really four pendulums, Nos. 36, 37, 38, 39, but with a view to possilile changes in the Antarctic it was considered advisable to retain one pendulum. No. 36, at Kew. The times of swing observed at Kew in 1901 and 1904 were as follows, all the ordinary corrections — temperature, pressure, arc, and clock-rate — having been applied : — Flexure correction omitted. I Flexure correction applied. No. 36. No. 37. No. 38. No. 39. No. 36. No. 37. No. 38. 1901 1904 ■5087654 652 •5087801 752 •5087745 719 •5088202 7925 ■5087586 583 •5087731 683 •5087675 650 Change [in 7th decimal place) -2 -49 -26 -277 -3 -48 -25 The apparent change in No. 36 does not exceed the probable error of the observations. Thus the presumption is that during the Expedition no appreciable change took place except in the pendulums themselves. That some change actually took place in Nos. 37, 38, 39 can hardly be doubted. There is independent evidence of the fact from the observations at Christchurch, presently to be discussed. The changes in Nos. 37 and 38 are, fortunately, not large, but that in No. 39 appears much more serious. Comparatively few observations were made with this pendulum at Kew before the Expedition set sail, and it is quite possible that the change in it is overestimated, Ijut in any case the results deri's-ed from it must be regarded as apprecialily more uncertain than those derived from Nos. 37 and 38. § 5. Table I., p. 26, gives particulars of all the observations taken during the course of the Expedition. The reductions, involving a large amount of laborious calculation, were made by Mr. Bernacchi. They were then done independently by Mr. E. G. Constable. The results obtained by Mr. Bernacchi were accepted as correct unless Mr. Constable's difl'ered by more than I in the seventh figure. In cases where larger differences existed I investigated the cause myself. I also revised the rates accepted for the chronometer during the observations. The usual set of observations gave four values for the period of each pendulum, two with the pendulum facing D (or direct), and two with it facing E (or reversed). But on February 6, 1903, at Winter Quarters, the R position alone was used, and at ISIelbourne, in 1901, the D position only was used. If one compares the residts for the D and R positions in Table I., when both 33 exist, one finds that on the ;tver.\i,'o the K periods exceeded the D in pendulums 37 and 38 by 17 x 10~^ and 15 x 10"' second respectively, whereas in No. 39 the D period exceeded the R by 9 x 10"". Individual ditferences, however, fiuetuate largely, and corresponding results, based on a greater number of observations at Kew, make the difference much less for No. 37, and both numerically less and of opposite sign for No. 38. It has thus been decided to neglect any difference that may possibly have existed between the D and R positions in dealing with the observations at Melbourne, and those on February 6 at Winter Quarters. Tables II., III., and IV. summari.se the results of Table I. T.VBLE II. — Results at Winter Quarters. Date. Observer. Periods of Pendulums. 37. 38. 39. 1902 July 31, August 1 BEKN.lCCni Skelto.v Means . . . ■5083155 159 ■5083077 103 •5083330 332 •5083157 •5083090 •5083331 1903 February 1, 2 Bernacchi Skelton Means . . , •5083251 245 ■5083202 183 •5083388 376 •5083248 ■5083192 •5083382 1903 February 6 Beenacchi Skelton Means . . . •5083287 249 ■50831G7 181 •50S3341 347 •5083268 •5083176 •5083344 1903 September 5, 6 Bernaccui Skelton Means . . . •5083135 143 •5083080 062 •5083299 301 •5083300 ■5083139 ■5083071 Tabi.e III. — Results at Winter Quarters. July 31, August 1, 1902 Febru-ary 1, 2, 1903 fi „ September 5, 6 ,, Means, allowing half weight to observations on February 6, 1903 „ from July 31, August 1, 1902, and September 5, 6, 1903 „ „ February 1, 2, 6, 1903 Periods of pendulums. 37. •5083157 248 268 139 •5083090 192 176 071 ■5083331 382 344 301 •5083194 ■5083148 •5U83255 •50S3126 •5083080 •5083187 ■5083339 ■5083316 •5083369 34 Table IV.^Kesults ;it Melbourne and Christchurch. Place aud date. Periods of penduliims. 37. 38. S9. Melbourne, November 10, 11, 1901 •5091079 ■5091218 C'liristcburoli, November 26, 27, 1901 •5089600 •.lOsfl.^ias •5089876 738 May 30, 1904 580 477 •5089590 •5089508 •50&9807 § C. The period.s observed at Kew in 1901 and 1904 have been already given. Their mean values, uncorrected for flexure, are Pendulum Period . . .37 0-. 5087776 .'58 0^ 5087732 39 • .5088064 It has been judged best to employ these mean Kew values for comparison with the results obtained at Winter Quarters and Christchurch, but to employ only the 1901 results for compaiison with Melbourne. The observations made at Christchurch in 1904 gave lower values than those obtained in 1901, the ditt'erences in the seventh place of decimals being - 20 for No. 37, - 61 for No. 38, and - 138 for No. 39. The differences for Nos. 37 and 38 give a mean which is closely similar to the corresponding mean difference observed at Kew, thus suggesting that any change that took place in these two pendulums occurred at Winter Quarters, and so influenced the Kew and Christchurch observations alike, leaving the Melbourne observations unaffected. The comparative brevity of the interval between the oliservations made at Kew and Mel!)ourne in 1901 is an argument pointing in the same direction. § 7. If /i and t> denote the periods of a pendulum at two places where ;/i and [/-^ are the values of gravity, then, assuming the pendulums unchanged, and the conditions as to temperature, pressure, &c., the same at the two places, we have Uih- = f/Jr, or ;/.■ = ,'/! (/i/^.)'- Accepting 981 '200 (centimetre/second") as the value at Kew,* the values deduced ))y the above formula for Melbourne, Christchurch, and Winter Quarters are those given in Table V. under the heading " Oljserved values." In the probable mean the results from pendulum No. 39 have been allowed only half weight as compaicd to those from either 37 or 38. Table V. Place and date. Observed values. Theoretical sea-level value. 37. 38. 39. Probable mean. Melbourne Christeliureli Winter Quarters : — July, August, 1902, and "1 September, 1903. . J February, 1903 .... All observations combined . 979 936 980 -501 982 -987 982 -946 982 -970 980-515 982 -997 982 -955 982 -979 980 -037 980 -528 983 -034 983 -013 983 -023 979 -970 980 -512 983 -000 982 -963 982 -985 979-954 980 -463 982 -963 ' Roy, Soc. Proc.,' A, vol. 78, .19u6, [i. 245, .15 § 8. As to the v.ihios tn lie oxpectod 1)V tlioorv at those stations, the fni'iiniLi whiili at present has must claims to acceptance is that of von Hki.MKUT, viz. : — where ;/ = 978-000(l+0-005,31sin->./,){l -| +^| - 11^ +//} . . . . (1), (f) is the hititude (north or south), h the height above mean sea level, h' „ thickness of surface strata of low density, R ,, Earth's mean radius, A ,, ,, „ density (5 -6), S „ ,, ,, surface density (assumed 2 • 8), „ actual density of surface strata at the place, // an orographic correction, arising from mountain masses, i^'c. It is possible that at Winter Quarters, Mt. Erebus, Mt. Terror, ]\It. Discovery, and other mountain masses, and the proximity of McMurdo Sound with a considerable depth of water, might severally contriliute sensibly to the y term in (1), but without much more complete information than exists, no value derived from this could make any claims to accuracy. The observed rock densities at Winter Quarters would seem to indicate that the mean surface density did not differ much from 2 • 8. At sea level, at a station where the surface strata have a density of 2 • 8, and thei'e are no causes (such as high mountains or deep seas) in the neighl)Ourhood for an orographic correction, ;/ = 978-000 (1 + -00531 sin-^ <^), or, more conveniently, !l = 978-000 + 5-193 sin- oxes filled with eavy materials The instrument was so placed that the boom pointed S. — N. true, thus in an opposite direction to its former position. The period of the boom was made exactly 15 seconds. From January 29 until March 18, 1903, the seismograph was dismounted, the brick pillar being then employed for pendulum work, and was finally dismounted in December, 1903. L. C. Bernacchi. Amongst the various records brought home l)y the .ss. " Discovery " from the Antarctic Regions, a long series refer to the movements of a horizontal pendulum. This instrument, which is similar to a type adopted by the British Association and established at 38 widely separated stations in various parts of the world, was in charge of Mr. Louis Bernacchi. When we read Mr. Bernacchi's log we recognise the exceptional difficulties, meteorological and otherwise, under which he worked. This and the fact that a hurried departure only admitted of a few hours' instruction in the practical working of the instrument he had to use, entitle him to the sincerest congratulations on the results he has brought home. The huts, to which Mr. Bernacchi refers, were 30 to 50 feet above sea-level at a place in longitude 166° 44' 45" E. and latitude 77° 50' 50" S., about 15 miles distant from Mounts Erebus and Terror. The former of these volcanoes was always active. The records obtained refer to Changes in the Vertical, Tremors, Pulsations, and Earthquakes. In many instances these records, when taken by themselves, have little value, but when analysed in conjunction with registers obtained by similar and similarly installed apparatus at very distant stations they throw light upon hitherto unsuspected phenomena which take place within and on the surface of our world. In the following pages I give a register of the earthcpiakes recorded by the " Discovery." To this is appended a list of very large earthquakes which were not recorded by the " Discovery " seismograph, although at the time of their occurrence this instrument appears to have been in working order. Finally, I give a certain number of conclusions arrived at from an analysis of these various observations. The greater number of these are to be found in a paper on " Preliminary Notes on Observations made with a Horizontal Pendulum in the Antarctic Regions," see ' Proceedings of Royal Society,' Series A, vol. 76, May 29, 1905. J. Milne. 41 EARTHQUAKES RECORDED IN THE ANTARCTIC REGIONS. 77" 50' 50" S. Lat., 166" 44' 45" E. Long. 1 902- 1 903. D = The (listiince of a station from an origin. C and M give in minutes the time taken hy phases C and M to travel from an origin to a given station. The time used in the following registers is Greenwich Mean Civil Time : Midday =12 hours, Midnight = 24 or hours. C = Commencement. M = Maximum. D = Duration. A = Amijlitudc, or half of a complete swing. Pi refers to the commencement of the first phase. Po refers to the commencement of the second phase. P3 I'efers to the maxinuun motion. Towns printed in italics refer to instruments not of the Milne type. 1. March 14, 1902. C. M. D. A. Remarks. " Discovery " . . . . Bidston Mamburg h. m. 21 36-6 22 10-0 21 55-8 h. m. 22 50 li. m. 2.5 1 miUim, 0-5 Line tremulous. The difference in time between the "Discoveiy" records and M for Bidston suggests an origin to the S.W. of New Zealand. The Bidston records proLaWy respectively refer to Pj and P3, while that for Hamburg refers to Pi. We apparently have here the first illustration of an earthciuake being only recorded at its antipodes and not at intermediate stations. 2. March 25. C. M. D. A. Remarks. h. m. li. m. h. m. minims. " Diseovery " . . . . .5 10 7 — 13 0-3 } A thickening. Bidston . . . 4 1.5 1 4 18-7 10 — Edinburgli . . 4 14 5 4 24 19 0-2 Toronto . . . 3 42 3 47-5 36 1-5 Baltimore . . 3 41 3 46 1 5 2-3 Victoria, B.C. 3 58 2 4 G-3 43 0-7 San Fernando . 3 59 9 4 15 6 45 1-5 Christclmrch . 4 25 7 4 28-3 — 1 Hamburg . . 3 54 — 1 — Nicola iew . 4 14 4 17-0 38 — Taschkent . . 4 18 4 — — Tijlis . . . 4 6 3 4 2-7 — — Borpat . . . 4 4-0 — 28 42 The identificiition of the hour marks on the " Discovery " film is uncertain. If, however, we take the first reading as -th. 10'7m. the record is fairly in accord with those which follow and refers to a disturbance originating in or near to Central America. The area disturbed is similar to that given for No. 72. 3. March 28. C. M. J). A. Reniarlis. "DiscoTery" .... Cliristchurch .... Wellington .... h. m. 5 42-1 5 46-8 5 48-7 h. m. 5 47-2 5 47-4 5 51 -0 h. m. 13 22 millim. 0-5 0-6 1-0 A lenticular thickening. A possible approximate origin lies near to 150° E. Long., 50° S. Lat. 4. March 28. C. M. D. A. Eeinarks. h. m. h. m. li. m. millims. " Discovery " . . . . 6 30-9 6 31-9 10 5-0 Batavia . . 6 10-3 6 16 19 1-0 Kodaikanal 6 16-3 — 20 — Irkutsk . . 6 6-0 6 35-7 38 0-1 Tif!is . . 6 4-8 6 2.5 -1 — — Nieolaiem . — 7 19-0 — — Dorpat . fi 49 — — — Manila . . 6 3-1 6 5-7 15 ~ Slight tremors were felt at Zamboanga in Mindanao. The "Discovery" record suggests an origin not more than 5° distant. The entries for Manila, Batavia, and Irkutsk, however, suggest a distinct disturbance of very large extent which originated in the Southern Philippines at about 6h. Om. If this is the case, the second phase of motion, or Pj, would leach the Antarctic regions at Gh. 26m., or about the time the "Discovery" shock originated. Apparently, therefore, we may ha\'e a case of two shocks related to each other as a primary and a secondary. 5. March 28. C. M. D. A. Bemarks. " DiscOTery " . . . . h.m. 9 37-0 h. m. li. m. 10 millim. 0-5 A thickening. As entries corresponding to this do not appear in the registers from Christchurch, Wellington, Batavia, and comparatively near stations, one inference is that the origin was local. 43 6. March 28. C. M. D. A. Hemarks. li. ni. li, m h. m. millims. " Dis(Hiverv " . . . . 9 58 -3 10 6 -3 20 (1 (1 Christcluiroli 9 42 -0 10 3-0 55 1 AVellington 9 48 -7 10 3 24 1'5 Batavia . . 9 29-4 9 40-2 55 5 '2 Bonibav . . 9 3t-.-, 10 3 -3 40 0-5 Kodaikanal 9 40-.-, 9 59-5 40 0-5 Irkutsk . . 9 46-9 10 3-4 38 0'3 Manila . . 9 28 1 32 '4 8 ~~ It was also recorded at Nicolaiew, Tiflis, and Hamburg. At Ternati and Halmaheira (Celebes), shocks were felt at 8.35 and 8.4G (see ' Natuurkundig Tijdschrift v. Ned.-Indie,' Ixiii,, p. 194). M-C for Batavia and Manila indicate an origin about 132° E. Long, and 3° N. Lat. A similar origin is obtained from the differences in the value of M, given for these stations. The time of the origin deduced from the observations made at these two places would be 9.23 or 9.22. With an origin at 9.22 the times at which we should expect Pi, P^, and P3 to reach New Zealand and the " Discovery " would be as follows : — New Zealand Pi 9.32, Po 9.41, P3 9.57. "Discovery" Pi 9.36, P. 9.47, P3 10.15. The inference is that P^ and P3 were recorded in New Zealand and P3 only l)y the " Discovery." 7. March 28. C. M. D. A. Remarks. " Discovery " . . . . Wellington .... Christchurch .... Batavia Bombay Kodaikanal .... Cape Town .... Shide Kew Bidston Edinburgh San Fernando .... Toronto Baltimore Irkutsk Cordova Victoria, B.C Mauritius Bamburg Tiflis Dorpat Nicolaiew Manila h. m. 14 47 -1 14 34 -0 14 53 -8 14 49 1 14 54 -7 14 53 -4 14 53 -8 14 23 -8 15 3 -2 15 2 -5 15 4 -0 15 3 -5 15 7 -0 15 6-5 14 53 -8 15 3 -6 15 3-5 14 55 -1 14 58 -9 14 57 14 59 -0 14 47 -9 h. m. 15 26 -8 15 19 15 23 -3 14 59 -5 ? 15 '20 -0 15 40-4 15 20-7 15 58-8 15 50-2 15 16-5 16 7 -0 16 12-5 15 18-8 21 8 16 7-0 15 4 -2 15 4 1 15 7-9 15 25 -0 14 53 -7 h. m. 2 50 1 9 45 3 1 26 3 2 58 2 20 1 34 1 7 2 9 4 34 1 53 2 17 3 8 4 36 2 47 4 30 1 4 millims. 2 4-0 4-2 5-40 ? 1-5 0-01 1-2 0-7 0-6 0-6 2-5 0-4 0-5 2-7 0-5 2-5 10 This shock, which was one of a series, was felt strongly in the Celebes, The time given for a heavy shock at Ternate is 14h. 45ni. M - C for Manila, Batavia, and Irkutsk indicates an origin about 132° E. Long, and 3° N. Lat. To reach Ternate, 3° distant, would take 2- 5m. The time at the origin would therefore be 14h. 42 •5m, This time calculated from the Manila maximum is 14h. 42m., and from the Batavia maximum 14h. 41m. The time adopted is 14h. 42m. The region disturbed embrages the whole world. The following table G 2 44 give.s the distance, D, of various stations from the origin and the number of minutes occupied by C and M to travel from the origin to the station. The average velocity DjC may refer to Pi or P2, while DjM refers to P3. Distance. Minutes. Average arcual velocities in degrees per minute. D. C. M. P.. P^. P3. Manila . . Batavia . . Wellington Christchurch Irkutsk . . Kodaikanal Bombay Mauritius . " Discovery " Tijli', . . Nicolaiew . Dorpat . . Victoria, B.C. Bamburg . Cape Town Kew . . . Edinbui-gh Sliide . . Bidston . . San Fernando Toronto . . Baltimore . Cordova 18 26 51 52 52 60 72 82 86 95 97 97 108 108 111 111 112 112 125 130 131 150 6 7 11 11 11 12 13 5 15 17 21 17 12 21 20 21 25 24 21 11 17 37 41 36 38 22 44 26 43 85 22 58 76 34 38 68 85 30 3-0 4-7 4-7 4-7 5-0 5-5 16-4 5-7 5-7 4-6 6-3 9-0 5-3 5-0 5-6 5-9 5-2 5-4 7-1 3-2 3-3 4-9 3-2 2 '9 4 3 1-6 1-5 1-4 1-2 1-4 1-3 1-8 2-2 11 1-8 1-4 1-6 1-4 The times at which P], P^, and P3 would be expected to reach New Zealand and the "Discovery" would be as follows : — New Zealand Pi 14.51, Po 15.0, Pg 1.5.17. "Discovery" Pj 14.56, Po 15.7, P3 15.35. From this it appears that with a disturbance of greater intensity than No. 6, Pi was recognisable at Christchurch. 8. March 28. C. M. D. A. Eemarks. " Discovery " . . . . Batavia h. m. 17 57 -3 17 40 -7 h. m. 18 3-4 18 2-4 h. m. 30 millims. 1-5 1 -4 In all probability the first entry refers to a shock which had an Antarctic origin. The relations between the times of these two disturbances as recorded in Batavia and by the " Discovery " are somewhat similar to those for No. 4, and one may be the secondary of the other. 9. April 1. C. M. D. A. Remarks. " Discovery " . . . . h, m. 9 19-2 h. m. h. m. 8 millim. 0-3 At Christchurch a very slight disturbance was noted at 'J 12' 7. The inference is that the shock after reaching Christchurch had about 9° to travel liefore reaching the " Discovery," A possible origin would be about 10" S.W. of New Zealand. 45 10. Aprin. C. M. D. A. Remarka. "Disoovery" .... h. m. 2 41-2 h. ni. h. m. 8 millim. 0-5 At Christchurch a slight earthquake was noted at 2.31, with a maximum at 2.34. The interval of time between the records at the two places suggests an origin like that for No. 9, namely on the line of the submerged New Zealand ridge. 11. April 7. C. M. D. A. Bemarks. h m. h. m. h. m. millim. • 6 5 I "1 "Tlie character is that of & bead- " DiscoTei-y " . . . . 6 \ fi 17 6 26 3 \ 5 26 10 sliaped group of waves followed by two thickenings. ClirisWliiu'ch .... 5 39 5 52 2 1 4 1-0 Pertli 5 31-8 5 41 9 1 3 1 BataTia 5 25-7 5 45 40 0-7 Taschkent 5 39-9 6 5 4 — — Dorpat 5 51-3 6 10 — Origin very uncertain, possibly S. Indian Ocean, or distiict G. 12. Apil 9. C. M. D. A. Bemarks. h. m. \i. m. h. m. millim. " Discovery " . . . . 8 24-9 — 21 0-2 Small serrations. Christchurcli . . 8 8-7 — — 0-9 Bidston .... 8 15-0 — — Doubtful. Strassburg . 8 10-0 — — Tifiis .... 8 13-1 — — — Taschkent . . . 8 57-2 9 3-2 — — Dorpat .... 9 2-2 8 59-0 ~ Origin doubtful. Probably S. of New Zealand. 13. April 10. C. M. D. A. Bemarks. "Discovery" .... Christchurch .... Irkutsk Strassburg Taxchkent Dorpat h. m. 14 59 -5 14 43 -7 14 37 C 14 27 -0 14 29 1 14 57 -0 h. m. 14 51 14 52 -2 16 15-0 li. m. 16 29 1 ,50 millim. 0-2 0-2 Small serrations. Origin doubtful. District K ? April 1 1 to 15, film lost. Origin local. Origin local. 46 14. Api-n 17. C. M. D. A. Remarks. 1 " DiacoTery " . . . . li. m. 23 li. ra. h. m. — 10 millim. 0-5 15. April 17. C. M. D. A. Eemarks. " Discovery " . . . . Christchurcli .... h. m. 16 27 -2 17 22 h. m. 17 24 h. m. 13 8 millim. 0'5 0-1 Small serrations. 16. April 20. C. M. D. A. Eemarks. " Discovery " . . . . Christcliurch .... Wellington .... Perth h. m. 21 34-9 21 42 -0 21 44-5 21 49 1 h. m. i h. m. 21 45-2 1 10 21 45-0 — — 7 — j 46 millims. 2 1-3 0-2 (See facsimile of trace, Plate 4.) By the method of circles an origin is arrived at in 160° E. Long, and 65° S. Lat. 17. April 21. C. M. D. A. Remarks. h. m. h. m. h. m. milUme. " DiscoTery " . . . . 17 34 -4 17 54-7 1 28 1 Duration of P, = 4m., P = 10m. Christohurch 18 0-2 1 Perth . . 17 38 6 17 50-0 1 38 2 2 Batavia . 18 37 3 18 46-7 1 20 1 6 Kodaikanal 17 40 2 17 57-4 1 30 1 Bombay 17 44 7 18 3-0 1 6 1 7 Irkutsk . . 18 52 1 18 28 -1 1 23 7 Bidston. 18 37 3 18 48-2 24 3 Edinburgh . 18 44 5 18 57 -0 18 0-2 Strassburg . 17 45 3 — — (See facsimile of trace, Plate 4.) Mauritius . 17 37 17 39 — — Cape Town 17 42 2 17 50-5 16 0-3 Cdlcutta ! 17 55 I 1 24 — Nicolaiew . 17 53 18 23 ■() Taschkent . 17 49 3 18 14-8 Tidis . . 17 39 3 18 0-8 — Dorpat . . 17 55 18 17 -5 — — From the values C - M, and from those of M, an origin is indicated in district G, possibly in 70° E. Lon and 40° S. Lat. The area disturbed is a hemisphere embracing Europe, Asia, Africa, and Australia. 47 18. ,l/>ril 2.- C. M. D. A. Bemarka. " Discovery " . . . . Christchurcli .... Taschkent li. m. 11 23 1 11 44 12 19-8 h. ni. 11 46-5 12 0-5 12 37 -3 li. m. 43 32 iiiillim. -2 1 Sornitions. The last I'eeord probaljly refer.s to P3. Origin S. of New ZealiuKl. 19. Jjjril 25. C. M. D. A. Remarks. " Discovery " . . . . Cordova Bidston Victoria, B.C Mauritius SfrasslitrtjF Irkutsk Taschkent h. m. 23 44 22 52 23 38 23 46 23 53 23 15 24 25 23 26 5 9 4 I 6 h. m. 22 54-9 23 50 24 26 -3 h. m. 1 3 27 7 millims. 17-0 0-3 0-2 A slight tliickening. The disturbance probably originated in district D off the W. coast of South America at 22h. 50ra. The large waves would reach the Antipodean region at the times sjiecifiod for the last two stations, respectively 150° and 180° distant. 20. April 28. C. M. D. A. Remarks. " Discovery " . . . s Taschkent h. m. 3 9-6 5 43-2 13 48 -3 13 26 -8 h. ni. h. m. 5 5 5 20 millim. 0-2 0-2 0-2 I Sliglit ripples. Origin local. 21. Apil 28. C. M. D. A. Remarks. "Discovery" .... Christchurcli .... h. ra. 18 48 -3 18 15 -2 h. m. h. m. 6 millim. 0-2 0-4 Slight ripples. Origin near New Ze aland. 22. Jpril 28. C. M. D. A. Remarks. " Discovery " . . . . h. m. 19 .5-6 h. ni. h. m. 6 millim. 0-2 48 23. Jj^'ii 29. C. M. D. A. Eemarka. " Discovery " . . . . Kodaikanal .... h. m. 3 30 3 46-6 h. m. h. m. 6 millim. 0-2 Earthquake ? 24. Maij 1. C. M. D. A. Remarks. " Discovery " . . . . h, m. 17 57 -0 ll. HI. h. m. 30 millims. 1-5 Earthquake ? 25. May 2. " Discovery " Christchurch Calcutta Edinburgh . . . . Bidston Tokyo Manila < Shide Kew Baltimore Irkutsk Nicolaiew Taschkent Tiflis Dorpat Hamburg 11 49 and 12 10 11 55 11 51 11 46 11 52 11 52 11 32 11 4 11 37 11 55 11 58 12 27 11 36 11 45 11 40 11 42 11 51 11 42 11 43 h. m. 12 35-0 and 13 4-0 12 17-0 and 12 31 12 10-9 12 6-6 12 31 -5 12 32 -2 11 33 -4 11 5-6 11 38-6 12 31 -9 12 32 12 39 -4 11 49-2 12 12 11 54-9 12 19-9 12 18 -9 12 22-5 1 6 47 57 1 5 36 1 50 1 24 1 5 26 39 1 39 49 14 45 1 40 0-2 0-7 0-4 2-4 Sliglit tremors. _ 2 out of a series of 10 recorded on May 2. This earthquake originated off the N.E. coast of Japan, in about 144° E. Long, and 40° N. Lat. The time of origin would approximately be 4 minutes before the arrival of M at Tokyo, or at llh. 29- 4m. The following table gives the velocity for C which may refer to Pj or Po, and M, which refers to P3 in the form ^^S:?£i:- minutes Distance. Minutes. Average arcual velocities in degrees per minute. D. C. M. P.. P,. Pj- Irkutsk . . Calcutta . Bombay Tiflis . . Hamburg . Edinburgh . Bidston . . Christchurch Strassburg . Sliide . . Baltimore . '■■ Discovery " 29 50 62 70 79 80 81 82 83 84 91 116 7-0 16-9 22-3 13-0 13 -3 22-6 22-9 25-6 13 25-7 19 19-8 32-7 41-5 50-5 53 1 62-0 62-8 62-6 62-5 70-0 94-0 4-14 5-38 5-9 6-4 6 10 3 12 2-82 3-54 3-53 3-20 3-26 1-46 1-34 1-49 1-38 1-48 1-29 1-29 1-31 1 34 1-30 1-23 49 It will be observed that the entries for C chiefly refer to l\, which, with the exception of a slight increase in the equatorial regions, has a constant arcual velocity. For I'l there are five entries. If we express these velocities in knis. per second in the form ^ .°'^' , they become 7-5, 9-3, 10-1, 10-8, and 9-3. time The value for P3 as an arcual velocity is fairly constant, with a possible slight rise in value 62" to 70° distant from the origin (see Time Curve No. 25, Plate 1). 2G. Mai/ 2. C. M. D. A. Bemarks. li. 111. h. m. li. m. r 14 37 1 niillim. " Discovery " . . . . 13 .TOO U U-0 < and [ 14 46 — Sliglit tivinors. Calcutta 13 44-3 1-1 28 36 0-7 Earthcjuake ? Manila 13 .51-2 13 52 1 1 — 27. May 7. C. M. D. A. Remaris. " Discovery " . . . . Christchviroh .... h. m. 6 34-5 6 16-0 li. m. h. m. 10 millim. 0-2 1 Slight ripples. Origin near New Zealand. 28. May 7. C. M. D. A. Remarks. "Discovery" . . . . h. m. 10 26 -3 h. m. li. m. 3 millim. 0-2 Origin local. 29. May 8. C. M. D. A. Remarks, h. m. h. m. h. m. mill ms. " Discovery " . . . . 2 13 4 — 6 5 Bead-like line. Christchurch 2 55 ? 3 12 45 2 Wellington 2 57 7 — 14 7 Eatavia . 2 33 7 2 34 15 8 Irkutsk . 2 25 ,5 2 38-7 51 1 5 Tokyo . 2 20 9 2 24-6 1 10 4 Shide . 2 49 5 3 21-8 1 12 1 7 Kew . . 3 8 5 3 18-2 32 6 Bidston . 3 4 3 21-6 44 8 Edinburgh 3 3 16-5 47 6 Calcutta 2 36 1 2 46-7 38 2 Bombay 2 46 ] 2 56-5 23 10 Manila . 2 23 5 2 23-7 12 — Nicolaiew 2 41 5 3 3 — — Taschkent 2 34 6 2 59-8 — — Tiflis . 2 18 2 2 55-7 — — Dorpat . 2 40 4 2 57-4 — — Hamburg 2 30 9 3 4-7 — — Strassburg 2 31 9 1 15 •50 This earthquake was felt in Southern Japan, and its origin is given by Mi'. A. Imamura as 9° distant from Tokyo, off the south-eastern coast of Kiusiu. From the observations made in Kiusiu (see ' B.A. Ciicular,' No. 6, p. 270), the value M - C for Tokyo and the Manila record, I should place this origin further S.W. from Tokyo, and deduce 2h. 16m. as the time of origin. The following table gives the average arcual velocity of propagation : — Distance. Minutes. Average arcual velocities in degrees per minute. D. C. M. Pi. P.,. P3. Tokyo . . Irkutsk . . Calcutta Batavia . . Bombay Taschkent . Tijlis . . Dorpat . . Nicolaiew . Wellington Christcliurcli Hamburg . Strassbitrg Edinburgh Bidston . . Kew . . . Shide . . " Discovery " 12 28 37 40 50 51 66 72 74 70 77 80 84 85 87 87 88 110 5 9-5 20-0 17-5 30 18-6 2-2? 24-0 25-0 41 39 ? 15 16 44-0 44 52 33 3-0 8-5 22-7 30-7 18-0 40 43 49 43 47-0 56 48-0 60-0 65 62 65 30-0? 5-33 5-23 2-4 2-9 2-29 2-78 3-0 2-96 2-0 2-66 41 23 85 66 19 34 67 57 85 34 66 93 97 67 35 The values for P3 approximate to what we should expect from other analyses. Po was only observed at comparatively few places, while Pi, unless we accept the two low values of 5 • 3 kms. per second, has not been recorded. The earthquake is essentially one that only exhibits one type of wave motion, and this was propagated across Europe and Asia, and southwards beyond Australia. 30. May 10. "Discovery," 22h. 20m. A series of thickenings commencing on May 9 at 15h. 52ra. and ending May 11 at 19h. 34m. C. M. D. A. Remarks. Irkutsk Taschkent Tijlis Vorpat Batavia Perth Strassbitrg Hamburg h. m. 22 16-5 22 22-8 22 24-5 22 35 22 20-7 22 25 -1 22 23-1 22 24 -6 h. m. 22 36 -6 22 30 -9 22 23-0 22 42 -1 h. m. 36 1 8 15 25 37 37 miUim. 0-3 0-9 This is entered in 'B.A. Eegist«r' for May 11. 31. May 19. C. M. D. A. Remarks. " Discovery " . . . . Batavia San Fernando . . . h. ni. 12 23-9 12 31 11 50-8 h. m. 12 31 -7 h. m. 4 2 29 millim. 0-2 0-8 Two bead-like thickenings. Small serrations. 51 The occurrence of this earthquake, although it may have been local, is possibly connected with three heavy shocks which on the night of May 19-20 shook Amboina in the Moluccas. May 24. The clock ceased to drive the paper several times, with the result that the times of five thickenings cannot be obtained. 32. May 26. c. M. D. A. Remarks. " Discovery" .... Hamburg S/rasxhm-ff TiJIii h. m. 10 59 -6 11 40-8 11 37-3 10 53 -7 h. m. r 11 5-7 1 t 11 0-8 / 11 43-4 h. m. 37 20 12 1 39 millims. r 3-5 1 2-5 (See facsimile of trace, Plate 4.) The " Discovery " record refers to an earthquake with an origin about 15° distant from that station. The time of origin would, therefore, be at about 10.50. At 11.40 we should expect Po to have reached Europe, which is the time at which records were obtained there. 33 May 31. C. M. D. A. Remarks. " Discovery" .... Christchurch .... £odaikaiiaI .... li. m. 5 12-1 5 8-7 4 54-4 Ii. m. 5 18-2 5 20-0 4 54-4 h. m. 24 18 5 millim. 0-5 0-2 1-0 Origin S. or S.W. of New Zealand. 34. Mai/ 31. c. M. D. A. Remarks. " Discovery " . . . . Cliristchurcli .... Taschkeut Tiflis I)orpat h. m. 7 7-1 7 9-7 8 21-2 8 28 -9 ? 8 26 h. m. 7 17-2 / 7 12-2 1 \ 7 18-2 / 8 44-0 h. m. 32 32 millim. 0-7 r 0-5 t 0-6 M - C for the first two entries suggest an origin 25° distant from the " Discovery " and 22° distant from Christchurch, or 140° E. Long, and 55° S. Lat. The time of origin would be approximately 7h. 4m. The anticipated times of arrival of P3 at the last three stations, respectively 115°, 125°, and 145°, would be 8.14, 8.19, and 8.29. June 2, 4.30 to 23.52 the light was out. On June 8 for a portion of the day the clock stopped, while th& record for June 9 has been lost. 35. June 10. " Discovery " Christchurch h. m. 3 35 -9 3 37-2 h. ra. h. m. 3 39 1 18 3 46 21 iiiUim 0-5 0-3 Origin to S. or S.W. of New Zealand. June 10, llh., to June 11, 3h., the light was out. H 2 52 36. June 13. C. M. D. A. Kemaris. " Discovery " . . . . h. m. 9 22-6 h. m. h. m. 10 millim. 0-5 Origin local. June 13, 21h. Om., to 15, Ih. 26m., the light was out. 37. June 15. c. M. D. A. Remarks. h. m. h. m. h. m. millims. " DiscoTery " . . . . 12 47 2 — 20 0-2 Hippies. Wellington 12 26 b 12 31 -0 G 2-6 Batavia . . 12 50 7 13 8-0 30 0-8 Perth . 12 35 4 12 54-6 46 O-o Bitlston . 13 33 13 30 -6 22 0-3 TaSL-hkent 12 54 13 35-5 — — TiHis . 12 24 4 ? — 2 25 — Dorpat . 13 2 9 13 28-3 1 12 — Nicolaiew 13 17 13 370 40 — Samburg 12 34 8 — 1 26 — Strassburg 12 44 1 50 ~ The records from Wellington and Perth indicate an origin in the vicinity of 155° E. Long, and 50° S. Lat. The time of origin would be about 12.21. The following table gives approximate values for the arcual velocities of propagation in degrees per minute :— Distance. Minutes. Average arcual velocities in degrees per minute. D. C. M. P.- Po. Ps- " Dibcovery " . . . 25 26 ? 0-98 Batavia . . 60 39 47 — 1-54 1-28 Taschkent 120 33 74 — 3-63 1-62 Nicolaiew 135 56 76 — 2-41 1-77 Dorpat . 146 41 67 — 3-50 2-18 Rambtirg 156 11 — 14 1 — — Strassburg 157 23 — 6-82 — — Bidston . 164 72 76 2-27 2 15 The inference to be drawn from this table is that in all cases but two C refers to Pj. 38. June 17. " Discovery," 21 42-6, a slight thickening. Origin local. 39. June 17. C. M. D. A. Kemarks. " Discovery Christchurc Taschkent Tiflis . Dorpat . Samburg Strassburg h h. m. 23 49 24 17 24 34 23 53 24 11 23 3 23 3 9 2 7 3 5 7 h. m. 24 50 -0 24 28 -2 24 47 -9 24 8-7 24 35 h. m. 30 35 1 26 57 57 60 millim. 0-5 0-2 53 40. June 21. C. M. D. A. Remarks. li. m. h. m. h. m. rnillim. " PiscoTerv " . . . . 6 53-6 — 26 0-2 Ripples. Christchurch .... 6 51 -5 6 59 — 0-7 Pcrtli 7 lG-1 7 20-0 32 0-2 Tanchkent 7 28-1 7 48-5 — . Dorpat 7 6-7 — — — The records from Christchurch and Perth indicate an origin S.W. of New Zealand in 140° E. Long, and G5° S. Lat. The approximate time of origin would be 6h. 44m. The arcual velocity of propagation would be as follows : — • Distance. Minutes. Average arcual velocities in degrees per minute. D. C. M. P.. P,. Pa- Taschkent .... Dor-pat 120 145 44 22 62 6-60 2-72 1-93 June 22, 12.45 to 24.0, the light was out. The record for June 23 has been lost. 41. June, 26. "Discovery," 23h. 17' 7m., a slight thickening. Origin local. June 30, after 18h. the light was out. July 5, 17.30 to 24.0 the light was out. 42. July 6. C. M. D. A. Remarks. h. m. h. m. h. m. millinis. " Discovery" .... 13 8 6 — 2 50 10 Thickenings. Cliristchurcli .... 13 7 8 13 18 — 7-1 Wellington .... 13 5 2 13 15-2 2 14 2-8 Perth 13 14 8 13 50-2 2 14 2-8 Irkutsk 13 27 7 14 34-7 2 25 0-3 Batavia 13 15 13 50-8 1 40 2-0 Mauritius 13 30 0(abt.) — — Victoria, B.C. . . . 13 25 8 14 27 -3 1 51 0-4 Toronto 13 28 6 14 20-5 1 51 2 Kodaikanal .... 13 27 7 14 11-0 2 6 0-6 Baltimore 13 29 3 14 14 1 53 0-6 San Fernando. . . . 13 20 7 — 2 9 — Edinburgh 13 44 8 14 40-7 2 8 0-7 Bidston 13 38 7 14 28 -5 1 58 0-7 Kew 13 40 ri2 20 7 1 — 25 0-2 Shide ■l and 13 40 1 — — ~ TiHis 13 21 9 13 48-8 2 56 — Nicolaieio 13 40 14 0-0 1 20 — Dorpat 13 32 3 14 27 -9 1 45 — Strassburg 13 22-6 — 2 18 — Hamburg 13 231 3 7 The position of the origin appears to have been to the N.E. of New Zealand at a distance of 25° to 30°, but it cannot be determined with any accuracy. The time of origin would be about ISh. Om. 54 The following table of velocities indicates the character of motion which reached distant stations : — Distance. Minutes. Average aroual velocities in degrees per minute. D. C. M. Pi- P„. P3- " Discovery "... 60 9 5-55 Victoria, B.C 90 25 87 — 3-60 103 Kodaikanal . 95 27 71 — 3-52 1-33 Toronto 115 28 80 — 4-10 1-43 Baltimore . ]15 29 74 — 4-00 1-55 Tiflis. . . 132 22 48 6 00 2-74 — Borpat . . 142 32 87 — 4-43 1-63 Nicolaiew . 145 40 60 — 3-62 2-41 Strassburg . 158 22 — 7-18 — — Bidston . . 162 38 88 — 4-26 1-84 San Fernando 175 20 — 8-75 — — From this table it appears that, with the exception of three stations, the character of the movement recorded refers to Po and P3. July 7, Oh. 30m. to 19h. the light was out. 43. July 13. C. M. D. A. Eeniarks. li. m. h. m. h. m. millim. "Discoverj? " . . . . 12 9-6 — 7 0-5 Time uncertain. Bidston 12 9 -0 12 18 9 — Cordova 11 59 12 6 9 0-5 Trinidad 12 2 — 16 — Strasshitrg 12 2-0 — 37 ? — Hamburg 12 3 — 1 42 — This disturbance probably originated in the South Pacific to the E. of New Zealand, and spread in a north-easterly direction as far as the western side of Europe. The area disturbed is similar to that given for No. 123. July 29, Oh. 30m. to 30, Oh. 30m., the light was out. August 1, llh. 30m., to end of day the light was out. 44. August 2. C. M. D. A. Eemarks. h. m. li. m. h. m. niillims. " Discovery " Perth . . 14 45 14 29 3 14 43 -3 1 1 33 12 5 Irkutsk . 14 40 9 14 55 -3 55 4 Batavia . 14 28 6 14 41-5 1 5 2 5 Mauritius 14 45 1 15 12-4 49 09 Bidston . 14 42 15 27 -0 1 48 3 Edinburgh Strassburg Tascliltent 15 32 14 42 14 33 1 9 15 3G -5 15 21-3 17 1 55 0-3 Tiflis. . 14 39 4 15 1-9 — — Nicolaiew 14 48 15 17 59 — Dorpai . 14 46 -4 15 28 1 14 ~ 55 A possible origin is 150° E. Long, and 10" N. Lat. With this supposition, however, a very much smaller amplitude would bo expected for Perth, and the disturbance should have been noted at Tokyo, Victoria, Christchtu'ch, and Wellington, which was not the case. Another possil)Ie origin is that given for No. 4G. 45. August 7. C. M. D. A. Remarks. h. m. h. m. h. m. millims. " Discovery " . . . . 11 32-6 11 37-7 40 2 Perth 12 3-2 12 6-7 40 0-4 Irkutsk 12 22-6 r 11 50 -0 ? 1 12 25 -0 11 0-2 Batavia or L 11 48-1 11 51 -0 ? 42 25 Kodaikanal .... 12 0-7 12 11-1 19 10 Mauritius 12 4-2 12 15 21 0-4 Cape Town 12 29 - 12 36-0 9 0-2 Bidston 12 50-6 12 54-4 20 Small At Palembang, in Sumatra, where the shocks were severe, the time given is 11.49, and for Benkoelen 11.35. From these notes the inference is that the origin was nearer to the S.E. extremity of Sumatra, about 5° N.W. from Batavia. With the assumption that the largest waves travelled at a rate of 3 kms. per second, the time at the origin as derived from the records from Batavia would be 11.48, and from the Perth record 11 47-6, results which accord with the time noted at Palembang. For P3 to reach the "Discovery," 80° distant, would take 49 minutes. The Antarctic record for maximum motion, if it refers to the Sumatra earthquake, instead of reading 11.37, should read 12.37. As the "Discovery" observation appears to be correct, it is interesting to note that 16 minutes after the commencement of a somewhat severe earthquake in the Antarctic, a large earthquake originated near Sumatra, 80° distant. And 16 minutes is the time which preliminary tremors would take to traverse the path of that length. August 8, 6h. Om. to end of the day the light was out. 46. August 10. C. M. D. A. Remarks. " Discover; " . . . . Batavia Perth Bidston h. m. 13 48 -4 12 47 -7 12 47 -6 13 35 -2 h. m. 13 51 -5 12 59 -5 12 58-6 13 47 -0 h. m. 26 20 53 32 millims. 1-2 0-7 3-6 0-2 Time approximate. From the time observations the inference is that this earthquake, like No. 44, originated at a spot about 50° distant from Perth and Batavia. The amplitude records, however, suggest that the origin was much nearer to the former station than to the latter. It might also be at a spot nearly equally distant from Bidston and the "Discovery." The duration of the preliminary tremors at Batavia and Perth suggests an origin about 50° distant from these two places. A position roughly in accordance with these conditions would be about 50° E. Long, and 30° S. Lat. August IG, 2.30, to 18, 3.51, light cut ofl' by snow, which filled the slit. 56 47. August 16. C. M. T>. A. Remarks. " Discovery " . Perth .... h. m. 8 12-11 8 12 8 39-0 8 35 K 36 8 27-0 9 24 8 42-0 8 25-0 8 11-0 8 11-0 li. m. r 8 24 -4 \ I 8 28 -5 J 8 35 9 11 -0 10 1 9 12 8 45-0 8 49 8 25-0 8 22-0 h. m. 1 1 10 millims. 1 2 1 Seen on edge of paper. Three shocks here. Cape of Good Hope San Fernando. . Irkutsk .... Christchurcb . . "Wellington . . Local origin. 4:1b. August 18. C. M. D. A. Remarks. " Discovery " . . . . h. m. 18 39 -0 h. m. li. ni. — 5 millim. 0-2 48. August 21. C. M. D. A. Remarks. h. m. li. m. h. m. mQlims. "Discovery " . . . . 11 35-3 6 0-2 Perth . . 11 30 11 43 8 1 3 0-9 Irkutsk . . 11 32 11 41 7 59 0-5 Batavia . 11 20 -7 11 27 7 1 4-6 Kodaikanal 11 30-7 11 35 38 0-6 Mauritius . 11 40-9 11 41 5 9 — Calcutta 11 24-3 11 44 1 50 0-7 Shide . . 12 19 1 — Bidston . . n 39-7 11 48 2 1 5 0-6 Edinburgh . 12 10-0 12 23 33 0-5 Manila . . 11 17 -3 11 19-2 1 29 It was noted at many stations in Mindanao and at Minado in the Celebes at 11.21. The real origin was in the central region of Mindanao, where there were great upheavals of the ground and destruction of buildings, 122° 25' E. Long., 10° 56' N. Lat. Records obtained in Manila and Batavia each indicate 11.14 as the time of origin. The time taken to reach the "Discovery," 90° distant, was therefore 21 minutes. The time taken by preliminary tremors, the second phase of motion, and the large waves to traverse such a distance, would be approximately 16, 25, and 55 minutes. The suggestion, therefore, is that the " Discovery " record may refer to the first or second phase of the Mindanao disturbance, but this, however, fails to explain the absence of records at Wellington and Christchurcb. Another suggestion is that the "Discovery" record refers to a disturbance of local origin, resultant on tremors which originated in the Mindanao area. 57 iSb. August 24. C. M. D. A. Remarks. " Discovers " . . . . Bidston Sfrassbuiy .... Samburg h. m. 5 49 5 33 •() 5 53-8 5 33-3 h. m. 5 47-0 5 53-5 h. m. 11 8 50 37 niillim. 0-3 Slight. 48c. September 1. "Discovery," .about 14.30, a small earthquake. Time uncertain. Local origin. August 24, 17.30, to 25, 0.44, the light was out. August 28, 6.30, to 29, 0.30, the light was out. No records up to September 10. September 11, about 11.30, light out, and no records up to the 17th, 2h. 6-7m. September 19, 6.30, to 20, the light was out. On the afternoon of September 20, Mr. Bernacchi says that large volumes of smoke and a fire-like glow were seen emanating from the crater of Mount EreTius, which, as a rule, is very quiescent. 49. September 22. C. M. D. A. Kemarks. h. m. h. m. h. m. millims. " DiscoTery " . . . . 1 59 1 2 14 2 1 55 2 5 Christchurcli 1 57-3 2 ti-4 4 14 3 Wellington 1 42 r 2 8-2 -1 1 2 55 '8 / 2 16-9 4 20 10 Perth . . 1 57-7 2 32 9 7 Batavia . . 1 55 -3 2 6 3 41 11 Tokyo . . 1 51-3 1 57-3 2 20 20 Cordova 2 6-8 r 2 12-7 1 \ 2 43 -2 J 2 44 1 5 Irkutsk . . 1 55-0 2 8-7 3 13 9 Victoria 1 58 -5 2 10-5 2 48 7 1 Mauritius . 2 2 12-8 — 3 2 Bombay 1 57-7 2 14-8 2 1 2 7 Kodaikanal 1 57-5 2 22-8 1 59 1 Calcutta . 1 55-8 2 18-2 2 8 7 5 Paisley . . 2 5-2 2 20-6 2 7 2 5 (See facsimile of trace, Plate 4.) Shide . . 2 5-4 2 49-0 2 40 8 Kew . . . 2 5-3 2 48-6 2 54 8 5 Bidston . . 1 59-6 2 46-3 2 32 7 9 Edinburgh . 2 2-0 2 50-2 3 8 3 5 Toronto . . 2 6-4 2 17-4 2 29 4 4 San Fernando 2 5-2 2 54-2 3 16 10 Cape Town 2 6-3 2 55-6 2 40 1 9 Trinidad . 2 4-0 3 20-0 2 19 1-5 Cairo . . 2 4-0 2 17-0 1 44 10 Nicolaiew . 2 0-0 2 22-0 — — Taschkent . 1 59-2 2 26-6 — — Tiflis . . 2 0] 2 12-2 — — Dorpat . 1 59-6 2 13-7 — — Manila . . 1 54-4 1 57-4 — — Strassburg 2 1-9 — 3 40 — Samburg . 2 1-2 2 15-5 In Guam, 145° E. Long., 13° 36' N. Lat., there was a great destruction of buildings, while in the Island of Saypan buildings were also shattered. An origin in this region approximates to one dependent upon values for M - C at the four nearest stations. From the hours of maxima as recorded at Batavia, Irkutsk, Tokyo, and Manila, by the method of circles a district is arrived at, the centre of which is in 13° N. Lat. and 130° E. Long., about 6° W.N.W. of Guam. 58 As the maximum was recorded in Manila and in Tokyo at the same time the origin should be found on a line all points in which are equidistant from these two stations. For similar reasons it should be on a line equidistant from Batavia and a point about 480 kms. from Irkutsk, on a line drawn from that place in the direction of Guam. These two lines intersect in the region indicated. Guam and Saypan were, there- fore, about 300 miles distant from the origin, and that this might well be the case is testified by the destruction which took place. Had these islands been much nearer to the origin of a disturbance which was so definitely recorded at stations all over the globe, it is likely that the destruction in Guam would have been greatly intensified. The time of origin as deduced from the time of arrival of the maximum motion at Tokyo and Manila is 1.44, at Batavia 1.43, and at Irkutsk 1.40. The local times given for Guam are 1.35, 1.4-5 and 2.5. As a close approximation to time of origin we shall adopt 1.44. In the following table velocities are expressed in degrees per minute. The eighth column indicates the proportion of the wave path between the origin and the various stations which was sulxjceanic, while the last column gives the amplitudes recorded at these stations : — Distance. Minutes. On arc. On chord. On arc. Path. Amplilude. D. C. M. Pi- Po. P3- o mUlims. Manila .... 20 10 13 2 00 1-98 54 1 Tokyo .... 24 7 13 2 85 2 '84 54 1 20-0 Bata™. . . . 37 8 22 4 62 4-54 68 9 11-0 Irkutsk. . . . 4.5 11 24 4 09 3-99 87 r, 13-9 Calcutta . . , 46 11 33 4 18 4 07 39 5 7-5 Perth .... 51 13 32 3 92 3-80 59 7 9-7 Kodaikaual . . .'Jo 12 38 4 58 4-41 44 3 11-0 Wellington . . — — — - _ - _ 1 10-0 Bombay . . . 62 13 30 4 77 4-54 2 08 3 4-3 Cliristchurcli . . 62 12 21 5 16 4-91 2 95 1 14-3 Taschkent . . . 67 15 42 4 46 4-21 1 59 3 Victoria, B.C. . 80 13 25 6 15 5 66 3 20 1 7-1 Ti/lis .... 83 16 28 5 19 4-74 2 96 2 — Mauritius . . . 8.5 18 30 4 72 4-27 2 83 1 3-2 Dorpat .... 88 15 29 5 88 5-30 3 03 2 — " Discovery " . . 92 14 29 6 57 5-88 3 17 1 2-5 Nicolaiew . . . 92 16 38 5 75 5 15 2 42 2 — Cairo .... 95 19 32 5 00 4-44 2 97 0-2 10 Hamburg . . . 102 17 31 6 00 5-23 3 29 Strasslurg . . 103 18 — 5 72 4-98 _ — — Edinburgh . . 105 17 65 6 17 5 -34 1 61 0-2 3-5 Kew 105 20 14(?) 5 25 4-54 1 64 0-2 8-5 Bidston . . . 106 15 61 7 06 6 10 1 73 0-2 7-9 Paisley .... 106 20 36 5 30 4 -57 2 94 0-2 2-5 Shide .... 106 20 65 5 30 4-67 1 63 0-2 8-0 Toronto . . . 112 21 32 (?) 5 33 4-52 3 50(?) 0-7 4-4 San Fernando 120 21 69 5 71 4-72 1 74 0-2 10 Cape Town . . 120 21 71 .5 71 4-72 1 68 10 1-9 Trinidad . . . 146 20 95 7 30 5-47 1 53 0-9 i-5 Cordova . . . 160 21 59 7-61 5-37 2-71 10 1-5 An inspection of the above table shows that the values for Pi are such as might be expected, and also that the velocity of propagation on an arcual path is not constant. The average velocity along a path corresponding to a chord when the length exceeds 50° is more nearly constant. P3 shows a fairly constant arcual velocity to stations at less distances than 60° from the origin. To stations 60° to 90° distant the rate of propagation increases to values that suggest a rate of transmission for Po. Beyond this distance the rate decreases to values ajjproximating to those obtained at the nearer stations. In the middle regions it may be noted that the increased rate is derived not only from records of Milne pendulums, but from others adjusted to have different periods. It is likely, therefore, that the maxima recorded by different types of instruments refer to the same phase of motion, and we are not dealing with apparent maxima occasioned by coincidences of the period of the pendulum and that of its foundations. There does not appear to be any relationship between P3 and the nature of the path, nor is anything clearly shown with regard to decreasing amplitudes. (See Time Curve No. 49, Plate 1.) 59 50. September 23. " Discovery " , Christehurch . Wellington Perth . . . Trinidad Batatia . . Manila . . Tokyo . . Irkutsk . . Cordova . . Kodaikanal Bombay Calcutta Mauritius . Cape Town Victoria, B.C. Toronto. . San Fernando Shide . . Kew . . . Bidston . . Edinburgh . Paisley . . Cairo . . Nifoloiew . Taschkent . Tiflis . . Hamburg . SlraJsshurg h. m. 20 33 20 33 1 20 40 -0 20 2-i-O 19 40-5 20 38 -7 20 36 -4 20 34 -1 20 28 -0 20 39 20 39 -3 39 -.5 39-1 38-7 24-2 33-7 30-9 31-2 27-5 30-5 32-0 33 59-6 34 1 31-9 32-5 ?} D. h. m. 2 20 4 18 3 40 2 33 2 13 2 45 1 46 1 20 2 58 2 39 2 8 2 15 2 4 2 51 2 35 3 25 3 27 3 26 2 53 3 19 2 2 20 4 39 4 30 A. uiillims. 10 18 15 4-0 10 3 5 3 4-8 2-5 11 1-6 2 2-7 2 4 1 >20 >20 10 >17 >17 >17 22 -Oi 6-0 2-0 Bemarks. The origin was near Guatemala City, 15° N. Lat. and 90° W. Long. The shock was also severe in British Honduras. The values for M at Trinidad, Victoria and Toronto give times of origin 20.16, 20.19 and 20.17. The time adopted is 20.16. The following table is similar to that given for Eai'thquake No. 49. Velocities are expressed in degrees per minute : — Distance. Min utes. On arc. On chord. On arc. Path. Amplitude. D. C. M. Pi- P2. Ps- o miUims. Toronto . . . 29 8 18 3 62 3-50 61 20 Trinidad . 29 8 18 3-62 3-50 61 9 10 Victoria . 41 28 — 46 20 Bidston 73 11 47 6-63 6 18 55 1 17 Edinburgh 74 — 52 — — ■ 42 1 22 Paisley. . 74 14 49 5-28 4-92 51 1 6 San Fernando 74 17 4-35 4-05 - - 1 10 Shide . . 75 15 51 5 00 4-60 47 1 17 Kew. . . 76 15 54 5 06 4 •66 41 1 17 Strassburg 80 16 — 5 00 4-56 - - 1 — Hamburg . 80 15 27 5 -33 4-86 2 96 1 — Nicolaiew . 99 17 22 5-82 5 11 4 50 8 — Cairo . . 105 16 29 6-56 5 ■(52 3 62 9 2 TiflU . . 107 18 33 5-94 5 05 3 24 7 — Wellington 108 17 58 6 -.38 5-35 1 86 1 15 Tokyo . . 110 20 32 5-50 4-65 3 43 1 3 " Discovery" 110 17 31 6-47 5-47 3 55 1 10 Christchurch 112 — 61 — — 1 83 1 Irkutsk . . 118 18 78 6-55 5-44 1 51 7 4-8 Cape Town 119 22 66 5-43 4-45 1 80 6 4 1 Taschkent . 120 — 77 — — 1 55 6 — Manila. . 140 22 28 6-36 4-86 5 00 1 — Mauritius . 1.52 S3 38 6-60 4-82 4 00 8 2-0 Perth . . 155 24 37 6-45 4-66 4 18 7 4-0 Calcutta . 1.55 23 89 6-73 4-87 1 74 8 2-7 Batavia . 162 24 54 6-75 4 -70 3 00 1 3-5 Kodaikanal 165 23 90 7-17 4-91 1 83 8 11 Bonibiy . 165 23 46 7 17 4-91 3-58 0-8 2 I 2 60 Like the table for the preceding earthquake, the arcual values for Pi increase with the length of the wave path, while chordal values approximate to a constant value. From 99 to 110 the values for Ps are above the average. Where wave paths have been suboceanic, amplitudes recorded in Western Europe and in New Zealand have been high, which suggests that an ocean load has a less damping effect than a continental load. (vSee Time Curve No. 50, Plate 1.) September 26, 16.0, to October 3, 2.30, the clock was stopped. 51. October 5. C. M. D. A. Remarks. " Discovery " . . . . Perth ...... Strasshurg h. m. 15 2 -9 15 20 15 7-5 h. m. 15 27 -2 h. m. 10 40 40 millim. 0-2 0-4 Slight ripple. An Antarctic disturbance also noted near to its antipodes. 52. October 6. "Discovery" 8.30 to the 7th, 16.0. A rhythmic succession of 5 or 6 large and 9 or 12 small bead- shaped wave groups. Whether these are connected with a severe earthquake which originated in Ferghana, 40° N. Lat. and 72° E. Long., and which was recorded at the Cape of Good Hope, Perth, and many other stations, is doubtful. 53. October 12. C. M. D. A. Remarks. " Discovery " Bidston . . Hamburg . Strasshurg . Nicolaiew . Taschkent . Dorpat . . h. m. 8 40-0 8 24-4 7 59-5 8 9-2 8 39 8 36-3 8 10 li. m. time appr 8 32-1 8 9-5 9 3-6 8 35-3 h. m. oxiniate 19 51 millim. 0-3 A slight thickening. October 13. From 10.0 to 16.0 the trace was very faint. 54. October 14. C. M. D. A. Remarks. h. ni. h. m. h. m. millim. " Discovery " . . . . 16 44 -0 17 26 -0 2 30 0-5 Time uncertain. Bidston — 16 10-0 — — Batavia 17 43 -7 17 49 -3 21 1-0 Manila 17 37 -5 17 38-5 Taschkent 17 56 -3 18 12-2 — — The Manila record refers to an earthquake which was felt at Zamboanga and Jol6 Island. The " Discovery " record relates to an earthquake of local origin, but it is very doubtful whether the time of its occurrence is connected with the disturbance in the Philippines. October 18 to 22, the line is very faint. From 16.0 to 23.30 on the 21st it is invisible. 55. October ^%. C. D. A. Remarks. " Discovery " . . . . h. m. 3 59-4 h. m. 10 millim. 0-5 Of local origin. 61 56. A^mvmhcr 1. 0. M. D. A. Remarks. " Discovery " . . . . h. m. 11 5-9 h. m. 11 10-9 li. m. 18 millim. 0-5 Of near origin. On November 2 disturbances commenced at Tokyo and Batavia at 11.19 and 11. 48 respectively. 57. November 4. C. M. D. A. Remarks. " Discovery " . . . . Perth Kodaikanal .... Bombay Calcutta Mauritius Cape Town .... Victoria Irkutsk Shide K.-W Bidston Nicolaiew Taschkent Tiflis Dorpat Hamburg Strassburg li. m. 12 44-0 12 5 -0 11 44-0 11 17-8 11 22-8 H 55-8 12 9 12 25 -6 11 37() 12 5-7 12 9-2 12 7-2 11 49-5 11 37 -9 11 47-4 11 .50-9 11 52 -3 11 53 h. m. 12 23 11 47 -0 11 4G'0 12 4 1 12 27 -8 11 48-1 12 12-7 12 20-0 12 18-7 12 3 11 48-4 12 4-4 12 8-3 12 12-3 li. m. 34 1 35 2 46 30 10 1 33 25 29 24 1 8 1 25 2 10 millinis, 0-3 4-5 3-8 0-6 0-5 0-1 C-3 0-5 0-5 0-5 A slight thickening. Sudden. Slight. Approximate position of origin, 85° E. Long., 35° N. Lat. This shock does not appear to have been recorded in New Zealand, Batavia, Tokyo, or Cordova. From the valnes of M at Kodaikanal and Bombay the time of origin would be 11.35. Large waves would arrive at the "Discovery," 125° distant, about 12.49. It would seem, therefore, that the "Discovery " record relates to P3. November 9 to 13 the instrument was dismounted. 58. November 15. c. M. D. A. Kemarks. h. m. h. m. h. m. niillirrs. r 10 41 2 11 21 -8 ■ " Discovery " . . ■ .'A or or !■ 1 20 1-5 [ 10 34 2 11 14-8 Perth 9 34 8 9 44-1 1 38 10-5 Batavia . . 9 28 1 9 41-2 1 3 1-6 Tokyo . . 9 32 5 9 37-0 30 1-4 Irkutsk . . 9 28 9 9 48 -5 1 12 0-2 Kodaikanal — 9 58-9 — — Mauritius . 9 39 5 10 10-8 1 11 0-5 Toronto . . 10 12 — 51 05 Victoria 9 58 7 — 59 0-1 Cape Town 10 6 10 23 30 0-3 Bidston . . 10 29 4 10 40 -0 22 0-1 Manila . . 1 9 24 1 9 40-0 1 7 — Nicolaiem . 9 50 10 25 1 — Taschkent . 9 40 9 59-2 — — Tiflis. . . 9 18 2 9 37-9 — — Dorpat . 9 48 — — — Strassburg . 9 41 9 — 2 30 — Hamburg . 9 38 -5 10 1-9 1 35 ~ It is interesting to note that this earthquake does not appear to have been recorded in New Zealand, whilst it was recorded to the N. and to the S. of the same. At Perth the movement was pronounced, from which it may be inferred that the origin was nearer to that station than to any other. In the 62 publications of the Earthquake Investigation Committee of Tokyo, No. 16, p. 99, Mr. A. Imamura analyses this earthquake on the assumption that its origin was probably near Manila, where it is said to have been registered at 9 21 -2. This, however, does not correspond with the entry in the 'Bulletin of the Philippine Weather Bureau,' November 1902, p. 281. The recorded amplitudes and the values for M for Perth, Batavia, and Tokyo indicate an origin near to the centre of a circle which would pass through these three stations, or 160° E. Long, and 0° N. or S. Lat. The values M - C f or these three stations suggest an origin near New Guinea, while the values of M together with that for Irkutsk place the origin about 1 30° E. Long. All that we can say about the origin is that it appears to be in the eastern portion of district F. 59. November 18. • C. M. D. A. Remarks. " Discovery" .... Strassburg Hamburg h. m. 2 12-3 2 2-4 h. 111. 2 12 -5 h. m. 10 16 millim. A slight thickening. Antipodean disturbance. 59&. November 19. C. M. D. A. Remarks. " DJBOOvery " . . . . h. m. 20 47 -0 h. m. h. m. millim. Slight, local. November 20. Instrument not working. 60. November 21. C. M. D. A. Remarks. h. m. h. m. h. m. millims. " Discovery " . . . . 7 24-3 — 1 35 0-2 A series of thickenings. Christchurch .... 7 23-6 — 1 25 0-8 Wellington .... Air tremors Perth 7 20-7 7 36-8 1 20 10 Batavia 7 7-5 7 29-2 1 20 8-0 Irkutsk 7 11-3 7 25 -0 1 26 4-3 Calcutta — 7 27-0 — — Trace faint. Mauritius 7 12-7 7 48-1 3 25 1-9 Bombay 7 11 -5 7 35 o 1 14 2-8 Tokyo 7 8-3 7 20-3 1 20 3-5 Origin in Formosa. Kodaikanal .... 7 10-0 7 33-3 51 0-7 Victoria 7 26-7 — 1 11 0-2 Toronto 7 32 -7 ? 8 18-5 52 1 Baltimore 7 30 — 2 — Cape Town .... 7 29-5 8 4-2 1 35 0-8 Cordova 7 29-0 — 48 — Shide 7 29 7 59 52 2 5 Kew 7 28-2 7 59-4 1 20 1 -8 Bidston 7 20-5 8 3-2 1 12 1 -2 7 23-5 7 56-0 1 30 11 Paisley 7 50 -0 ? — — — Obscured. San Fernando .... 7 55 -5 8 3-2 52 3 Manila 7 4-9 7 7-9 2 1 Felt at Santfl Domingo (Batanes at 7 6 -5. 1-) Nicolaieto 7 24-5 7 46-0 1 23 — Tasvhkent 7 12-9 7 40-7 — — Tiflis 7 13-6 7 44-9 1 45 — Dorpat 7 15-0 7 45 -0 1 45 — Hamburg 7 15-8 7 27 2 45 — Strassburg 7 16-1 ~ 3 In the publication of the Earthquake Investigation Committee of Tokyo, No, 16, p. 100, Mr. Imamura says that this earthquake originated off the southern coast of Formosa, and that in Manila the initial movement was registered at 7h. 1 -Sm. This hour, it will be observed, is not in accord with the one just given, wliich was obtained from 'The Bulletin of the Philippine Weather Bureau,' November, 1902, 63 pp. 280 and 281, Neither is it in accord with observations made at inctcorologiial stations in Foimosa, which were as follow : Taito, 7 12-0; Tainan, 7 4-2; Koshun, 7 6-7; Taichu, 7 2 '8; and Taihoku, 73-0. At the two latter places, which are in Northern Formosa, the motion was slight, whilst at the three former places, in the southern part of the island, the movement was strong, houses were shaken, and at Koshiui clocks were stopped. At Tainan there was vertical movement. The inference is that the origin was the S. of that city and its time would be at least 2 minutes earlier than that recorded in Tainan. From the times of arrival of maximum motion at Manila, Tokyo, and Irkutsk, an origin is arrived at a point about midway between the northern end of Luzon and the southern extremity of Formosa, or approximately at 120° E. Long, and 21° N. Lat. The values M -C for these three stations also indicate an origin in this district. From the times taken for largo waves to travel from such an origin to the above three places it would appear that the time at the origin would lie between 7h. 2m. and 7h. 4m. As the former of these is more nearly in accord with observations made in Formosa and at other stations than the latter, it is the one adopted and used in the construction of the following table : — Distance. Minutes. Average arcual velocity in degrees per minute. D. C. M. Pi. P.,. Pa- Manila Tokyo Batavia Calcutta .... Irkutsk Kodaikanal . . . Bombay .... Taschl-ent .... Perth TiUis Mauritius .... Doi'pat Nicolaiew .... Christehurch . . . Hamburg .... Strasibiirg. . . . Kew Edinburgh. . . . Sliide Bidston Paisley Victoria " Discovery "... San Fernando . . Cape Town . . . Toronto Baltimore .... Cordova .... 6 25 29 30 35 39 ■44 50 52 65 73 74 74 77 85 87 90 90 92 93 93 93 100 105 110 115 118 175 3 6 9 8 9 10 18 11 10 13 22 21 13 14 26 21 27 18 24 22 27 30 28 27 6 18 27 25 23 31 33 38 34 42 46 43 44 25 57 54 57 61 48? 53? 62 76 57 2 4 1 5-8 3-9 4-8 4-9 5-0 5-9 7-3 5 7 6-5 6 2 4-3 5 1 4-5 4 1 6-5 2-9 3-3 3-6 3-4 3-4 3-4 3-8 2-0? 3-8 4-3 3-0 1-0 1-4 1 1-2 1-5 1-2 1-3 1-3 1-5 1-5 1-6 17 17 1-5 1-6 1-6 1-5 1-9? 1-9 17 1 5 See Time Curve No. 60, Plate 1. Local origin. 62. December 23. 61. December 7. C. M. D. A. Bemarks. " Discovery " . . . . h. m. 8 52-2 h. m. h. m. 10 millim. 0-5 A thickening. C. M. D. A. Bemarks. " Discovery " . . . . h. m. 4 26-0 h. m. h. m. 11 millim. 0-5 Irregular thickenings. Local origin. 64 63. December 25. 0. M. D. A. Beniarks. h. m. h. m. h. m. millims. " Discovery" .... 5 32-7 f 5 42-9 5 45-9 I 49 2-0 Christchurch .... 5 32-7 1 1 1-7 Pertli 5 43-0 5 51 54 0-7 Origin 160° E. Long, and 60° S. Lat. On the continuation of the New Zealand fold. 64. December 25. C. M. D. A. Remarks. " Discovery " . . . . h. HI. 9 14-4 h. m. 9 22-5 li. m. 1 10 millinip. 7 The sudden commencement indicate.s a near origin. 65. Jannari/ 4 or 5, 1903. An earthquake with a duration of 50 minutes and an amplitude of 1 millim. was recorded, but the times are not given. It probably I'efers to a disturbance recorded at many stations round the world. In the Shide register it is numbered 668 (see ' Brit, Assoc. Circular,' No. 8). An appro.ximate position for the origin is 130° E. Long, and 30° N. Lat., from which the maximum motion would reach the " Discovery " on January 4 about 5.45 a.m. 66. Janvarij 14. C. M. D. A. Remarks. h. m. h. m. h. m. millims. " Discovery " . . . . 1 14-6 ~ 46 0-2 No central line, and the record only just visible on the edge the film. Time imcertain. 16 of Christcliurch . . . . 2 4 8 2 38 8 2 39 16 5 Wellington 2 4 8 2 38 S 2 39 16 5 Perth . . 2 8 8 3 55 8 2 52 3 1 Batavia . . 3 7 4 3 38 7 2 40 2 2 Calcutta 2 8 7, 2 .^8 8 2 32 11 Madras . . 2 11 4 2 58 2 1 45 2 Bombay 2 7 9 2 54 9 2 36 24 Mauritius . 2 18 8 / 2 24 1 3 21 I f 4 7 4 Tokyo . . 2 3 2 19 2 40 1 Irkutsk . . 2 7 4 3 15 9 2 2 3 Victoria 1 55 G 2 18 5 2 36 6 7 Toronto 1 54 4 2 11 2 2 34 >18 Cairo . . 2 12 2 15 2 30 2 Cape Town 2 7 4 2 52 3 3 7 o Timts approximate. Cordova 57 3 1 5 1 2 3 5 Sliide . . 1 59 1 2 36 7 2 3 Large waves also at 3.34. Kew . . . 1 59 3 2 37 4 3 4 >17 Bidston . . 1 54 fi 2 6 7 2 9 14 7 Recrudescence at 5.24. Edinburgh 2 5 2 37 2 51 11 Paisley . . 1 57 5 2 34 2 12 5 San Fernando 1 58 1 2 19 1 2 57 6 Trinidad . 1 57 2 17 1 39? 8 Manila . . 2 7 2 2 11 1 56 — Taschkenf . 2 3 2 3 48 fi 1 35 — TiflU . . 2 2 7 3 7 6 3 7 — Dorpat . . 2 1 4 2 11 -2 3 42 — Strassburg . 3 ? — — — Hamburg . 2 1-1 — — — Values for M and M - C indicate an origin about 100° W. Lc January 29 to March 18 the seismograph was dismounted. y. and 10° S. Lat. 65 G7. Marrh 20. C. M. D. A. Kemarks. "Discovery" . . . . li. m. 7 12-5 h. m. Ii. m. 5 millim. 0-2 Local origin. 68. March 20. C. M. D. A. Remarks. " Discovery " . . . . h. m. 10 35 -8 li. m. h. m. 5 millim. 0-2 This was followed by slight ripples with approximate periods of 60 seconds. The total duration is 65 minutes. At 10.31 an earthquake was recorded at Kodaikanal. The " Discovery " record probably refers to a shock of local origin. 69. March. 21. Between 10.30 and 12.30 the trace is slightly irregular, but whether this has any connection with an earthquake recorded at Bidston at 10.57, Tokyo 10.40, and Irkutsk at 10.45 is very doubtful. 70. March 22. From 14 to 16 hours a series of slight thickenings were recorded. It is possible that these refer to an earthquake which was noted at Shide, Bidston, Edinburgh, Paisley, Mauritius, Bombay, Calcutta, Tiflis, Irkutsk, Cordova and Cairo. The origin of the disturbance was to the N. of Bombay, 78° E. Long., 30° N. Lat. 71. March 26. C. M. D. A. Kemarks. h. m. h. m. h. m. millims. " Discovery " . . . . 9 5-3 — 53 0-2 Series of thickenings. Christcliurcli 8 .54 -7 9 6-2 42 2 Perth . . 8 .56 -0 9 7-9 31 0-8 Irkutsk . 8 58-8 9 2-8 21 — Bat a via . 8 55-7 9 6-5 30 0-5 Shicle . 9 42-2 9 59-6 25 0-2 Kew . . 9 41-0 — 15 0-2 Manila . 8 47 -5 8 45-8 35 — Taschkent 9 5-0 9 32-9 — — Dorpat , 9 8-9 9 45-7 ~ The time intervals between C and M at Christchurch, Perth, and Batavia, and the practical identity in the times at which C and M arrived at these three stations, suggest an origin in the eastern portion of the East Indies. The Manila and Batavia records suggest an origin to the N. of the Celebes. 66 72. March 29. C. M. D. ! A- Ketnarks. h. m. h. m. h. m. millims. ■'Discovery " . . . . Cordora 16 47 -7 16 30 16 32-5 50 1 20 0-5 4 A series of tliickenings decreasing in amplitude. Maurilius 16 42 4 17 26-2 1 42 4 Trinidad 16 56 17 8-0 35 1 Baltimore 16 49 — 7 3 Toronto 16 49 — 11 1 Victori;i, B.C. . . . 16 51 8 16 55 5 3 1 San Fernando . . . 16 47 17 29 1 11 1 5 Sliide 17 19 5 17 26 7 30 i ^ 5 Kew 17 19 5 17 28 24 4 Bidston 16 55 17 26 2 42 4 Edinburgh .... 17 20 17 27 5 21 2 Tiflis 16 53 9 10 59 7 — ' 0-4 Irkutsk 16 48 4 18 18 5 3 28 — Taschkent 16 44 4? 17 48 3 2 1 — Dorpat Strasshurg .... Hamburg 16 55 16 46 17 1 8 5 9 17 23 -9 1 15 i A well-defined seismogram obtained at Cordova indicates an origin about 12° distant from that place. The probability is that the same lies to the W. of Cordova off the West Coast of South America, or 75" W. Long, and 30° S. Lat. The approximate time at the origin would be about 9 minutes earlier than the time of arrival of M at Cordova, or 16.23. The area over which it was recorded may be represented by a radius of 100°. In New Zealand and at the Cape of Good Hope, which lie within this distance, however, it does not appear to have been noted. Distance. Minutes. Average arcual velocity in degrees per minute. D. C. M. Pi- P.,. P3. Cordova . . Trinidad . . " Discovery " . Baltimore . . Toronto . . Victoria, B.C. San Fernando Sliide . . . 12 45 65 70 72 90 90 102 102 103 104 105 108 120 130 145 178 7 33 24 20 26 28 24 56 56 32 57 23 38 32 30 21 25 9 45 33 66 63 65 63 64 60 36 85 55 1-7 4-6 4-3 6-9 7-1 2-7 2-7 2-7 3-2 3-7 3 2 3-7 3-6 3-2 13 1-3 2-8 13 1-8 1-8 16 1-9 2-8 2 1-7 Biilston . . . Edinburgh . Strasshurg Eamhurg . . Tiflis . . . Taschkent . . As already noted for Earthquake No. 50, the values for Pi increase with the length of the arc, while the values for P2 and P3 are fairly constant. 67 73. March 30. Christclnirch Perth . . TiBis . . . Batavia . . Bombay . . Calcutt a Mauritius . San Fernando Shide . . Irkutsk . . Kew ... Bidston . . Edinburgh . Manila . . Taachi-eiif . Dor pat . . S/rassburg . Hamburg . h. m. 3 31 -4 3 38 3 28 3 36 3 26 3 38 3 37 3 42 4 32 3 55 4 59 4 25 3 43 4 22 3 26 3 19 3 46 3 41 3 46 li. m. 3 40-5 4 1-1 3 43 -2 3 46 3 35-8 4 2-6 3 42-0 4 3-4 4 56 4 59-6 3 26-9 3 40 -9 4 26-3 D. h. m. 2 13 1 4 1 6 40 45 33 37 6 27 9 2 21 1 20 2 14 A series of thickenings and bead- formed groups of waves. Times approximate. The records from Batavia, Manila, and Irkutsk point to an origin near Ceram and 25° distant from Bata\'ia. At Amboina three heavy shocks were felt at 3h. 5m. There was also a heavy shock at 3h. 32m. The former was immediately followed by a disturbance in the sea. The time of origin deduced from the Batavian records is 3h. 19m., from the Manila maximum 3h. 15m., and from the Calcutta maximum 3h. 14m., which neither accord amongst themselves or with local observations. The disturbance did not reach Cape Town, or Victoria, B.C., but travelled westwards across Asia and Europe. 74. April 3. C. M. D. A. Bemarks. " Discovery " . . . . Batavia h. m. 4 24-3 4 45-7 h. m. 4 56 -6 h. m. 44 30 millim. 0-2 0-6 It is possible that these entries refer to distinct disturbances. 75. April 3. c. M. D. A. Remarks. " Discovery " . . . . Irkutsk Tiflis ...... Calcutta Mauritius Victoria, B.C Toronto Baltimore San Fernando. . . . Bidston Kew Shide Sfrasshurg Hamburg Taschkent h. m. 8 58-9 9 57-4 10 6-8 10 18 -7 8 5-8 9 40-6 9 51 9 58-5 10 14 -5 10 1-9 10 17 -0 10 17 -5 9 43 1 9 43-7 9 53-5 h. m. 10 5-5 10 28 -7 9 46-8 9 57-1 10 27 -5 10 12 10 21 -5 10 3 -3 h. m. 50 31 24 3 22 42 42 13 27 11 20 millims. 0-2 0-7 1-0 10 11 0-2 0-2 0-3 Minute irregular serrations. A possible origin is in District B, 160' W. Long, and 40° N. Lat. K 2 68 76. Jpril 10. C. M. D. A. Uemarks. " Discovery " . . . . Cordova h. m. 7 32-7 7 22-9 h. m. 7 44-0 7 24-0 h. m. 21 30 millim. 0-5 0-4 The commencement and end of the record from Cordova is not clear, otherwise it is well defined. It is possil:)le that the origin was off the West Coast of South America, in the same region as No. 72. 77. Jpril 10. C. M. D. A. Remarks. "Discovery" .... Wellington Perth Mauritius Shide Bidston Taschkent Tiflis Vorpat h. m. 15 12-2 15 16 15 19-9 15 40 ? ■ 16 4 "1 and \ 16 29 -6 J 16 34 -8 15 33 -5 15 32 -9 16 20 h. m. 15 18-4 15 20 15 23 16 39 -2 16 1-7 15 50 1 li. m. 1 10 13 34 18 17 millims. 4-2 1-5 1 1 1 Commencement and end liidden by tremors. (See facsimile of trace, Plate 5.) From the times of the arrival of M at the first three stations an origin is arrived at on the S.W. continiiatioa of the New Zealand axis, or approximately at 140° E. Long, and 55° S. Lat. The times of arrival of the larger waves at Shide and Bidston, roughly 160° distant from the origin, indicate they refer to the disturbance recorded by the " Discovery." The superficial area disturbed by this earthquake was apparently a band 50° in width running from its origin in a N.W. direction to its antipodes. It does not appear to have reached the continent of North and South America to the N.E. and E., Java, Manila, or Japan to the N., India to the N.N.W., or the Cape of Good Hope to the W.N.W. 78. April 12. C. M. D. A. BemarkB. " Discovery" .... Mauritius San Fernando. . . . Tijiis h. m. 23 10-8 23 2-7 23 44 -7 23 -4 ? h. m. 23 48-3 23 4-2 23 51 -7 23 21-6 h.m. 1 18 6 10 millim. 1-0 0-2 0-7 Pj lasts IG minutes. Assuming that these records refer to the same earthquake, the inference is that the area disturbed is similar to that given for No. 77. 79. Apiril 15. C. M. D. A. Kemarks. " Discovery " . . . . h. ni. 3 16-8 h. m. li. m. 24 millim. 0-2 Origin local. 69 80. April 15. C. M. D. A. Bemarks. " DiscoTei7 " . . . . Perth ]i. m. 12 1 -4 11 49 -3 li. m. 11 50-5 li. ni. 6 6 niillini. 0-5 0-3 The origin is probably on a line joining these stations 125° E. Long., 50° S. Lat. 81. Ajnil 25. c. M. D. A. Bemarks. " Discovery " . . . . Cordova CapeTowu . . . . Taschl-eiit h. m. 2 49-4 2 51-4 2 17 3 6-6 h. m. 2 14 3 25-5 h. m. 25 10 4 20 millim. 0-2 1 0-5 Probably lociil. The above records suggest an origin 60° distant from the first two stations, and about 6° S.S.W. from Cape Town, or 18° E. Long, and 40" S. Lat. 82. April 29. C. M. D. A. Bemarks. h. m. h. m. h. m. millims. " Discovery " . . . . 4 21-1 (■ 3 8-0 ] \ "'■ 1 4 30-3 2 15 1-0 Groups of bead- and gourd- waves. shaped Wellington .... 4 22-0 5 20 11-0 L 4 15 -2 J Christchurch .... 4 10-8 4 28-8 2 6 8-5 Perth 4 24-3 4 45-3 1 37 2-5 Tiflis 4 24-5 5 28-8 1 35 0-5 Cordova 4 33 1 4 43-2 12 0-5 Irkutsk 4 32 -6 — 53 — Batavia 4 15-2 5 5-5 1 30 2-1 Bombay 4 38-8 5 23-5 2 9 1-5 4 10 7 5 14-8 3 6 1-4 Cape Town .... 5 9-5 — ^ — Boom not free. Baltimore 4 41 -0 ? 5 30 1 30 1-2 Victoria 4 32-0 — 29 0-1 Boom partly anchored. Toronto 4 40-0 5 24 1 38 1-0 San Fernando . . . 4 44-3 5 54 1 48 2-0 Paisley — 5 48-0 — — Edinburgh 5 32 5 50-0 1 6 Bidston 5 0-0 5 52 7 1 3 1-0 Kew 5 33-5 5 48-5 2 2 1-6 Shide 5 26 -.3 5 53-1 2 1 -5 (See facsimile of trace, Pla te 5.) Manila 4 20 — 4 8 — Taschkent 4 31 4 5 23-6 — — Dorpat 4 28-4 5 27 -6 2 — Strassburg .... 4 29-6 — 1 40 — Hamburg 4 29 1 4 39-8 2 22 From the amplitude of movement and the time of arrival of the large waves at Wellington it is evident that the origin of the disturbance was nearer to that station than to any other. This is one of the largest earthquakes which has been recorded with its origin in district M. It disturbed the whole world. The time of origin based on the value of M - C for the nearer stations would be 4h. 5m., or half-an-hour after the occurrence of a disastrous earthquake in Caucasia, distant 150°. If we read this time of origin 4h. Om., this is the time at which the preliminary tremors from Caucasia, generated at 23h. 40m. on April 28, reached their antipodes, or the district in which the earthquake we are discussing took place. 70 83. April 29. C. M. D. A. Eemaris. " Discovery " . . . . Christchiireh .... Taschkent Tiflis li. m. 13 10-1 13 17 -0 13 39-0 14 52 -7 h. m. 14 41 -2 14 53 -2 li. til. 1 millims. 0-2 1-3 This shock originated at no great distance from Christchurch, and is in all likelihood an aftershock for No. 82. As confirmatory of this it appears that the times given for M at the two latter stations approximately accord with expectations. We have here another illustration of a disturbance only being recorded near to its origin and at its antipodes. 84. May 4. C. M. D. A. Bemavks. " Discovery " . . . . h. m. 10 48 -4 h. lu. h. 111. 6 niillim. 0-2 Local origin. May 11 to 18, the clock stopped. 85. May 19. 0. M. D. A. Remarks. " Discovery " . . . . Christchurch .... h. m. 7 52-7 8 14 h. in. h. m. 15 16 millim. 0-7 10 Origin between these two stations. 86. May 21. C. M. D. A. Remarks. " Discovery " . . . . h. m. 1 39-9 h. m. h. m. 4 millim. 0-2 Local origin. 87. May 21. c. M. D. A. Remarks. li. m. "Discovery". ... 17 57 9 h. m. h. in. millim. 5 0-2 Local origin. 71 88. May 23. " Discovery " . . . . Perth I Mauritius . . . . -^ Calcutta Kodaikanal . . . . Batavia -j Tokvo Irkutsk Tiflis Sliide Kew Edinburgh San Fernando . . . Manila Taschl-ent Dorpat Hamburg Strassburg h. m. 22 32 20 16 22 22 20 31 22 30 22 17 21 20 20 8 22 14 22 16 22 28 22 21 23 3 23 8 22 53 23 1 22 10 22 19 20 50 22 27 22 28 20 21 22 31 -6 22 50 22 28 21 25 20 10 22 20 22 19 22 31 22 58 23 19 23 9 23 14 22 44 23 3 h. m. 50 19 38 11 41 66 15 20 55 30 37 1 22 25 20 39 44 1 13 minims. 0-4 0-3 1-3 0-5 2-2 0-3 0-7 3-7 0-5 0-6 0-2 0-2 .Bemarks. Series of thickenings. This earthquake recorded by the " Discovery " apparently refers to the one recorded in Manila by the Vicintini seismograph. It originated in the island of Mindanao (see Earthquake No. 48), near to Darao and Caraga, at the head waters of the Eio Augusan (see 'Bulletin Philipp. Weather Bureau,' May, 1903, p. 115). Westwards across Europe and Asia the disturbance extended to Great Britain, 105° distant, but it does not appear to have reached Victoria, B.C., or Cape Town, respectively distant 95° and 105° along paths which would be suboceanic. The area covered closely resembles that for Nos. 73 or 88. The first entries for Perth, Mauritius, Batavia, and that for Kodaikanal refer to a distinct shock which maj- or may not be related to the one noted by the " Discovery." The Manila record indicates the time of origin at a distance of 14° as 22h. 7m. The entries for M at Batavia, Perth, and Calcutta respectively give 22h. 7m., 22h. 6m., and 22h. 5m. as the times for the origin. The following table is constructed on the assumption that the time of origin was 22h. 7m. : — Distance. Minutes. Average arcual velocity in degrees per minute. D. C. M. Pi- P,. P3- Manila . . Batavia . . Tokyo . . Calcutta Perth . . Kodaikanal Irkutsk . . Taschkent . Mauritius . Tiflis . . Dorpat . . Hamburg . Strassburg Kew. . . Edinburgh Shide . . San Fernando 9 22 31 36 40 43 46 61 70 76 86 100 101 105 105 106 115 4 7 9 10 14 13? 21 ? 12 23 14 20 21 46 56 54 13 13 21 24 19? 24? 37 43 51 56 61 62 72 67 2-2 3 1 3-4 3-6 3-3 5 5-4 5 5 2-4 2 9 2-2 3 2 3 2 1 1 1 1 1 1 7 7 7 7 9 6 6 5 7 7 9 7 (See Time Curve No. 88, Plate 1.) May 28 to 31, clock stopped. 72 89. May 31. C. M. D. A. Eemarks. h. xa. h. m. h. m. millims. " Discovery " . . . . 6 52-9 — 25 0-4 Wellington .... 6 32 6 39 20 3-5 Pertli 6 54-3 7 0-9 18 0-3 Shide 7 54-9 — 5 0-2 Irkutsk 6 40-8 7 15-8 1 50 After reaching Wellington the shock had about 35° to travel to reach Perth, and 31° to reach the " Discovery," which indicates an origin 15i!° E. Long, and 48° S. Lat., or on the continuation of the New Zealand axis. The time at the origin deduced from the value M - C at Wellington would be 6h. 24m. The time taken for large waves to reach Shide, 1G2° distant, would be Ih. 30m. The actual arrival and the anticipated time for the same are in accordance, and wc have another marked instance of antipodean reappearance. June 1 to 5, clock stopped. 90. June 8. c. M. D. A. Eemarks. h. m. h. ni. li. m. millims. "Discovery " . . . . 5 30 5 5 48-8 1 55 3-5 P, lasts 5m. 2(i». Perth . . 5 34 5 37 1 31 3-3 Wellington 5 31 5 49-0 1 8 1-5 Cliristcluircli 5 34 5 5 49-5 1 34 • 2 Batavia . 5 31 1? 5 41 -9 10 0-6 Mauritius . 5 45 9 5 56-4 46 0-6 Calcutta 5 58 o — 43 — Bombay Kodaikanal 5 58 5 47 4 2 6 12-3 6 10-7 34 32 10 0-3 (See facsimile of trace, Plate £ ■) Bidston . . 6 47 3 6 57 -9 16 0-3 Irkutsk . . 5 24 8 6 0-8 — — Taschkent . 5 54 — — — Tijlis . . 5 46 6 6 3-4 — — Vorpat . . 5 11 8 6 22-3 — — Hamburg . 5 49 6 5 57-4 1 3 — Strasshurg . 5 45 1 — 1 From the times given for M at the first four stations we should expect to find an origin at the centre of a circle which passed through Perth and touched a circle of 15° to 18° radius which passed through Wellington, Ghristchurch, and the " Discovery." This indicates a locality 120° E. Long, and 42° S. Lat. The probability, however, is that the same is to be found 10° farther eastwards on the line of the New Zealand axis. That the time of arrival for M at Bidston is 80 minutes later than Perth accords with this supposition. The area shaken closely resembles that for No. 88. 91. June 8. C. M. D. A. Eemarts. h. m. h. m. h. m. millim. " Discovery " . . . . 14 39 -2 r 12 44-4 l — 16 0-5 Time approximate. Shide •< and > 14 34 1 J — 6 0-2 Both shocks are similar. Kew 14 46 -2 14 56 -0 15 0-4 Antipodean disturbances which are nearly simultaneous. 73 92. Jimr. 9. C. M. D. A. Bemarks. h. m. h. m. li. in. millims. " Discovery " . . . . 11 22-3 — 1 20 0-2 A series of thickenings, approximate. Times Chiistchurch .... 11 16-6 11 18-7 22 3 Perth 11 35-4. 11 4.5-2 32 0-6 Batavia 11 37 -7 11 54-7 30 0-4 Mauritius 11 51 -6 12 91 35 0-4 Paisley — 12 42 -5 — — Shide 12 7 1 — 14 0-2 TiOis 11 7-6 11 19-8 _ 0-6 Strassburg 11 38 -7 — 50 — The difFerences in time between C and M at Christchurch, Perth, and Batavia indicate an origin on the S.W. continuation of the New Zealand axis, or approximately 150° E. Long, and 50° S. Lat. The area shaken practically resembles that for 77. 93. June 15. C. M. D. A. Kemarks. h. m. h. m. h. m. niillims. " Discovery " . . . . 22 30-8 22 40 1 30 0-5 Christchurch .... 22 18-1 22 22-9 — 2 1 Pertli 22 38 -9 22 43 -7 50 2-1 Irkutsk 22 33-6 23 0-4 2 — ■ Strassburg 22 29 -6 — — Although this disturbance was not recorded in Britain, it was noted in Strassburg. The area shaken is, therefore, similar to that given for No. 92. Also certain time intervals; for example, the interval between C and M at Christchurch approximates to those for No. 92. From resemblances like these it may be concluded that these disturbances had their origins in the same district. 9-1. June 17. C. M. D. A. Remarks. " Discovery " . . . . Cordova Mauritius Bidston Shide Taxchkent Dorpat Sfrassburg h. m 20 6 19 34 20 28 20 21 20 29 20 42 20 46 20 10 9 1 4 8 2 6 4 6 h. m. 19 36 1 20 37 ■^ 20 31 -5 20 55 -2 h. m. 10 8 9 19 uiillim. 0-2 0-5 0-3 0-2 The origin was evidently nearer to Cordova than to any other station, and at a distance of about 10° from that city measured westwards. The time taken for phase M to travel such a distance would be 6 minutes. The time at the origin would, therefore, be about 19h. 30m. With this assumption the times at which Pa would be expected to reach the "Discovery," Mauritius, and Britain would be 20h.l0m., 20h. 40m., 20h. 35m., which are not widely different from the observations. The area disturbed is somewhat similar to No. 72, which had an origin in the same region. 74 95. June 21. C. M. D. A. Remarks. h. m. h. m. li. 111. millini. " Discovery" .... 8 fi-5 — 10 0-2 Christchurch .... 7 43-9 — 33 0-3 Bidston 8 58-7. 9 2-9 21 0-2 Shide 9 0-6 — 45 0-2 TascAkent 8 20-3 8 20 — — There is also a record from Cordova at 7 16 '8, but as the duration is 8 hours, it suggests a tremor storm. If it is assumed that the shock originated a few minutes before 7.43, and at no great distance from Christchurch, the times of arrival for P3, as noted at the other stations, approximate to what would be expected. It is, however, difficult to understand why records were not obtained at stations lying between New Zealand and its antipodes. 96. June 25. C. M. D. A. Remarks. " Discorery " .... Bidston Sliide li. 111. 13 13 1 14 17 -3 r 14 18-2 < and [ 14 30 -5 J h. Di. 14 24 -0 h. m. 45 33 35 millini. 0-3 0-2 0-2 A disturbance was also noted at Strassburg at 13.24 by a Eebeur-Ehlert pendulum, and at 14.22 by a Milne pendulum. In Taschkent there was a record at 13 27-7.* The time interval for M between the "Discovery" and the English station lies between 71 and about 96 minutes. If the shock originated in the Antarctic region, we should expect the interval to have been about 90 minutes (sec Nos. 91 and 95). 97. June 27. C. M. D. A. Remarks. " DiscoTery " . . . . li. m. 4 35-6 h. m. h. m. 10 millim. 0-2 Local origin. 98. June 27. C. M. D. A. Remarks. " Discovery " . . . . li. m. 6 16-3 li. ni. h. 111. 25 iiiillim. 0-2 Local origin. • This is in tlic Strassburg register for June, but not in the 'Russian Bulletin,' April to June, 1903. For June 24 in the ' Bulletin ' there is a shock, recorded at Irknt.sk 13.6, Tasc-hkent 13 -5, Tiilis 13.8, which may correspond with a record for Mauritius 13.49 with a maximum at 14.4. 75 99. June 28. C. M. D. A. Remarks. " Discovcrv " . . . . h. m. 13 23 -9 h. m. li. ni. 5 millim. 0-2 Local origin. 100. July 2. C. M. D. A. EemarkB. li. m. li. 111. h. m. millims. " Discoverv " . . . . 21 32 4 — 55 0-5 Cliristchureli .... 21 21 6 21 30 5 13 1-4 Batavia 21 33 1 21 57 5 40 0-5 B.A. Begisler gives date as July 3. Mauritius 21 37 1 22 16 1 54 0-3 Bidst-ou 21 52 9 22 2 27 0-4 Sliide 21 40 2 22 2 5 1 15 0-2 M<,„ilu 21 24 4 21 25 1 47 — Tifis 21 36 1 21 54-7 — — S^rasslurf/ 21 34 5 — 1 30 — Hamburg 21 32-5 ~ 2 ~ The records from Christchurch, Batavia, and Mauritius indicate an origin 1 50° E. Long, and 50° S. Lat.j or on the submerged New Zealand axis. The area disturbed is similar to that for No. 73, &c. 101. July 3. C. M. D. A. Remarks. " DiscoTerj " . . . . Batavia h. m. 21 24-5 21 33 1 li. m. 21 57-3 li. m. 12 40 millim. 0-2 0-5 The seismic character of the "Discovery" record is doubtful. 102. July 8. C. M. D. A. Kemarks. " Discovei-; " . . . . BataTia h. m. 2 58 0(abt.) 2 58-0 li. m. 2 59 1 h. m. 25 7 millim. 0-5 0-6 A shock was felt at Bantam, Java, at aliout i\\. 56m. 103. July 9. C. M. D. A. Remarks. "DiscoTcry " .... Christchurch .... h. m. 4 40-6 4 52-2 h. m. 5 11 1 5 19-3 h. m. >0 50 millim. 0-7 Time of commencement and end uncertain. 1 Approximate origin, 150° E. Long, and 60° S. Lat. L 2 76 104. July 12. " Discovery " Christehurch Perth . . Irkutsk . . Tiflis . . . Mauritius . Cape Town . Baltimore . Victoria, B.C. Sun Feruando Bidston . . Kew . . . Shide . . Taschkent . Dorpat . . Strasshtirg . Hamburg . 28 34 13 36 40 37 23 42 G 35 6 15 6 11 5 48 5 21 4 29 5 38 5 39 h. 6 7-2 4 40 5 49 5 14 5 45 6 15 6 23 6 43 6 34 6 53 6 26 6 44 5 50 6 24 D. h. m. millims 1 12 0-5 24 2-5 50 2 1 1 36 0-4 — 0-4 1 4 0-2 51 05 1 0-6 1 1 10 28 0-3 6 0-3 1 55 7 1 20 — 2 20 — Tlie first entry probably refers to a maximum phase. The records from Christcliurch, Perth, and Irkutsk indicate an origin 150" E. Long, and 5° S. Lat., or E. of New Guinea. It does not appear to have been recorded at Batavia, Manila, and three Indian stations which are comparatively near or at Cordova in South America. With these exceptions it was noted all over the world. 105. July 12. C. M. D. A. Remarks. "Discovery" .... h. m. 13 27 6 h. ro. 13 35 -7 h. m. 18 millim. 0-5 Local origin. lOG. July 12. C. M. D. A. Remarks. h. m. h. m. h. m. millim. " Discovery " . . . . 4 0(abt.) — 23 0-2 Christehurch . 4 35 — 12 — Victoria, B.C. 4 8 4 12 13 0-4 Toronto . . 4 1 2 4 1 Edinburgh 4 31 — 4 0-2 Bidston . . 4 2] 4 28-3 20 0-3 Kew .... 4 30 5 — 4 0-2 Shide . . . 4 24 5 10 0-5 Strasshurg . . 4 3 — — Taschkent . . 4 19 6 4 45-3 — — Dorpat . . . 4 14-7 4 14-3 — — All these entries excepting that for the " Discovery " refer to July 28. 11 We have here a small disturbance which seems to have extended round the northern hemisphere N. of latitude 30", and to have also been recorded at Christchurch and possibly by the " Discovery " in the southern hemisphere. 107. Augud 11. C. M. D. A. Remarks. h. m. h. ni. h. m. millitns. " Discovery " . . . . 4 40 8 5 11 6 2 20 4 Irkutsk 4 46 5 5 10 7 50 3 Tiflis 4 37 1 4 41 7 2 4 2 7 Taschkent 4 32 1 4 51 6 — 1 Kodaikanal .... — 4 50 7 — 5 Bombay 4 48 4 51 9 33 1 4 Mauritius 4 53 2 4 54 3 — 5 Cape Town . . . . 4 52 9 5 9-8 40 0" 48 Victoria, B.C. . . . 4 56 — 48 2 Toronto 4 53 5 — 32 2 San Fernando . . . 4 39 2 4 43 8 1 7 4 A tremor at Lisbon. Cairo 4 38 5 4 42 52 2 5 Paisley 4 42 4 44 43? 3 Edinburgh .... 4 38 5 4 43 8 51 3 7 Bidstou 4 37 4 43 6 57 2 3 Kew 4 37 8 4 42 42 3 3 Shide 4 35 9? 4 44 1 40 2 Manila 4 45 5 4 46 3 24 — Dorpat 4 17 1 4 45 '9 — — Hamburg 4 36 3 — 2 28 — Strasshurg .... 4 36-8 1 20 This earthquake originated in the eastern part of the Mediterranean and was felt in Southern Italy, Cairo and Constantinople. At about 4.25 it was severe in the Island of Cerigo which was near the epicentre. In Athens there were heavy shocks at 4 31 • 7 and 4 33 ■ 0. The time of origin deduced from the first records in Athens would be 4h. 30m. ± 30s. The time of origin deduced from observations made at comparatively near stations is shown by the following table : — Distance. Time to travel. Time at origin. ' m. h. m. Tiais 18-0 11 4 30 7 Kew 20-0 12 4 30 Shide 21 13 4 31 1 22 14 4 29 8 Bidston 23-5 14-5 4 29 1 Edinburgli 24-0 15-0 4 28 8 Taschkent 33 20-5 Mean value 4 30-08 The following velocity table is constructed on the assumption that the shock originated near Cerigo at 4.30. (See Time Curve No. 107, Plate 1, and Map, Plate 2.) Distance. Minutes . Average arcual velocities in degrees per minute. D. C. M. P- P.. 1'3. Cairo Strasshurg .... Tiflis Hamburg .... Kew Shide Dorpat San Fernando . . Bidston Edinburgh .... Paisley Taschkent .... Bombay .... Kodaikanal . . . Irkutsk Mauritius .... Caj^e Town . . . Toronto Manila Victoria .... " DiscoTery "... 12 17 18 19 20 21 21 22 23 24 24 33 46 56 59 65 09 70 85 90 140 3-5 6-5 7-0 6 7-8 G-0? -13 9 7-0 8-0 12-0 ? 2-0 18-0 16 23 23-0 23-0 15-0 26 10-0 12 12 12 14 16 14 14 14 14 21 22 21 40 24 40 16 42 3-43 3 16 3-50? 3-28 3-0 16 -5 5 -66 2-5 2-57 2-56 2-44 2-00 ? 2-55 1-0 1-50 1-66 1-50 1-31 1-56 1-64 1-71 1-71 1-56 2-09 1- —' 1-47 2-70 2 66 3-68 2-82 3-00 3-04 5-31 1-72 14-0 3-45 3-33 If we omit the values for Pi for Taschkent and the " Discovery," as they are abnormally large, and that for Manila as being too small, the inference is that this phase of motion did not announce itself at stations more than 24° distant from its origin. Ps reached a distance of 90 or 140°, whilst P3 was lost at a distance of 70°. For area disturbed see Map (Plate 2). 108. August 11. C. M. D. A. Eeniarks. " Discovery " . . . . Eamhurg Strasshurg h. m. 10 50-1 11 5-3 11 1-5 h. m. h. m. 5 47 20 millim. 0-2 This is probably a shock which originated in region M some time before 10.50. If a phase Pi had been transmitted to Europe it would reach the German stations at about the time records were obtained. 109. Awjud 12. C. M. D. A. Remarks. " Disooverv " . . . . Perth . ■ h. m. 50-9 53-7 h. ni. 11-6 h. m. 20 18 niillim. 0-2 0-5 Records of a small disturbance were also obtained in Strasshurg at 0.48 and Hamburg at 0.47. The origin which would accord with the first two records would lie on the S.W. extension of the New Zealand ax if!. 110. August 12. "Discovery" 17 2'G and 17 39 -O. Each witli a duration of 5m. and an amplitude of 0'2 millim Local origin. 79 111. Amjusl 17. C. M. D. A. Remarks. " Discovery " . . . . CordoTii Irkutsk Taschkent Borpaf h. m. 18 21 -9 18 9-2 18 44 -2 18 36 -8 18 52 -0 h. m, IS 29 '0 18 10-2 18 45 19 31 1 h. m. 20 19 5 31 millim. 0-4 1 0-4 The first two records suggest an origin off the South American coast, W. of Cordova, to which Irkutsk and Taschkent are antipodean. 112. August 19. c. M. D. A. Remarks. h. m. Ii. m. h. m. niilliui. "Discovery" . . . . 9 21-2 9 44-5 1 50 0-5 Christehurch . 9 43 -fi 9 55-0 32 0-5 Calcutta . . 10 4-3 — 29 — Capo Town 8 58-6 9 19-7 38 1 San Fernando 9 48-8 y 57 -4 31 0-4 Bidston . . . — 10 41.1 -0 — Slude . . . 10 1 — 25 0'2 Taschkent . . 9 25-1 10 5 -4 — — Irkutsk . . . 9 23-5 9 42 2 12 Strassiurff . . 9 22-9 — 1 15 — The position of the origin is very uncertain. A possible origin would be 20° to 40° S.E. from Town. 113. August 21. C. M. D. A. Remarks. "Discovery" .... h. m. 14 16 - li. m. 5 millim. 0-2 The absence of other records and the sudden commencement indicate a local origin. 114. August 25. C. M. D. A. Remarks. " Discovery " . . . . Taschkent . . h. m. 12 16 -7 12 35 -7 h. m. 12 33 -8 12 42 -3 b. m. 30 milliui. 0-2 Times of beginning and uneerlain. end are It is probable that the " Discovery " record refers to a disturbance of Antarctic origin. 80 115. August 26. C. M. D. A. Remarks. h. m. h. m. h. m. millira. " Piscoverv " . . . . 13 14-2 13 33-5 30 0-5 Perth 13 17 -0 13 28 28 0-2 Bidaton 13 58-2 14 5-3 18 0-2 Taschlcent 13 30 13 45 -9 — — The origin probably lies in the S.W. extension of New Zealand axis, nearly equidistant from the "Discovery" and Perth. The interval between C at Perth and C at Bidston suggests that the record at the latter station refers to Po. As there are no records from intermediate stations, the Bidston entries suggest an antipodean reinforce- ment or resurgence. 116. August 29. C. M. D. A. Remarks. h. m. li. m. h. m. millims. " DisfiOTei-y " . . . . le 57 -5 — 5 0-2 CordoTa . . 15 20-9 15 24-4 17 1-5 Irkutsk . . 15 35 -5 16 16-0 4 5 — Taschkenl . 15 59 C 16 43-7 11 Bidston . . 15 59-5 16 6 3 18 0-3 Sirasshui'ff . 15 37 -7 — — — Hamhurg . 15 38 — 1 8 — The Cordova record refers to a disturbance that was felt at that city and in Tinogusta at 14h. 43m. The great difference in time between the times given for these localities suggests that they may refer to distinct disturbances. From the value M-C for Cordova the inference is that the disturbance originated at a spot about 1-5° distant from that city, and from this it follows that the time of origin would be about 15h. 16m. The records for Strassburg and Hamburg may refer to Pi or Po, but most likely the latter. The value for M at Bidston indicates a time at which P3 might be recorded. The record for Taschkent is suggestive of an antipodean focal effect. The disturbed area is not unlike those for Nos. 94 and 111. Instrument dismounted August 31 until September 22. 117. September 23. C. M. D. A. Remarks. h. m. h. m. h. m. millim. " Discovery " . . . . 26 1 27-1 25 0-8 Christehurcli 17 21-7 32 0-8 San Fernando 1 37 3 1 49-3 21 0-6 Paisley . . 1 52 7 1 53-3 6 0-5 Edinburgh . 1 53 5 1 54-5 5 0-4 Kew . . . Sliide . . 1 50 1 49 8 4 1 50-4 6 7 0-4 0-5 (See facsimile of trace, Plate 5.) Tiflis . . 1 23 1 1 22-7 — 0-5 Strassburg 1 46 3 — 35 — Hamburg . 1 46 8 1 55 -7 — — Irkutsk . . 1 28 6 1 37-9 1 — Taschkent . 44 6 1 28-9 — — Dorpai . . 1 27-4 1 34-3 — — 81 At 1.45 a shock was felt in Algiers. There were two heavy shocks in Blidah, near Algiers, anil also about this time there were two in the Canaries. The time records indicate that the origin was about 12" distant from Christchurch, and after the shock reached this station it had yet 10° to travel before arriving at the " Discovery." The probability, then, is that it lies appro.vimately in 1G0° E. Long, and 52° S. Lat. The time at the origin would be about 0.14. Large waves or Pg would reach Western Europe at 1.44 to 1.45, indicating that the C and M records for European stations both refer to this particular phase. That shocks should have been felt in Northern Africa at the time when these waves ani\'ed is worthy (jf note as also is the absence of records at stations lying between the origin and Europe. We have here a case not only of antipodean convergence, but possibly also of wave convergence resulting in the relief of seismic strain. The area over which it was recorded extends as a Ijand from New Zealand in a N.W. direction over Western Asia and Europe, which band might possilily be continued round the world. 118. September's. C. M. D. A. Eemarks. 1 h. 111. " Discovery " . . . . 7 27 '7 Christchurch .... 7 19 "l h. m. 7 33-7 7 23-2 h. ni. iiiillim. 30 5 31 0-5 Origin like No. 117. 119. Scptemhcr 2Q. C. M. D. A. Eeinarks. " Discovery " . . . . Christ^-hurch .... Taschkent h. m. 5 53-8 6 15-0 6 32-8 h. m. 5 56-9 6 17-8 h. m. 25 24 millims. 4-5 1-3 (Se > facsimile of trace, Plate 6.) The value M-C at the first two stations indicates an origin about 14" distant from each. The two values for M and the great diflerence between the two amplitudes, however, indicate that the origin was nearer to the "Discovery" than to Christchurch. A possible and extremely likely origin lies on the continuation of the New Zealand axis or in Long. 145° E. and 52° S. Lat. With a time of origin at 5.50 we should expect Po to reach Taschkent at 6.24, and P3 at 7.2. The record at that place may therefore refer to a local shock, and is not connected with the " Discovery " observations. 120. Octobe7- 4. C. M. D. A. Remarks. h. m. h. m. h. m. millim. "Discovery" .... 5 6-2 — 20 0-2 Time approximate. Christchurch .... 5 2-4 5 5-7 27 1-0 Baltimore 5 55 6 4-5 14 0-3 Bidston 6 7-2 6 12-7 23 0-4 Kew 6 19-2 — 11 0-2 Shide 6 0-3 6 4-4 45 0-5 1 The origin was apparently about 14° distant from Christchurch and nearer to that station than to any other. The time of origin would be about 4.57, and if the position of this origin is on the S.W. continuation of the New Zealand axis we should expect the time of arrival of phase Pg at British stations to be about 6.27, which, when we consider the durations given for these stations, might well be the case. M 82 Baltimore and the British stations, it may be remarked, are al)out equidistant from the antipodes of the supposed origin. (See No. 117.) It may be noted that this disturbance was not recorded at Hamburg, Strassburg, or Kussian stations, at all of which there are instruments of much greater sensibility than the Milne type. 121. October 8. C. M. D. A. Remarks. " Diseorery " . . . . CordoTa li. m. ]1 17-4 11 12-2 h. 111. 11 19 o h. m. 38 27 millim. 1-0 0-5 (See facsii lile of trace, Plate 6.) Origin about 15° distant from the " Discovery," probably in the direction of Cordova. 122. October 14. C. M. D. A. Remarks. " DiscoTery " . . . . Christeliurcli .... Pertli Irkutsk Tokyo Batavia Calcutta Shide Manila Taschkeiit Tiflix Dorpat Strasslurg Hamburg h. m. 3 40-1 .3 32 -2 3 20-0 3 32-3 3 32-7 3 9-7 3 35-4 4 4-5 3 18-0 3 35 3 41-8 3 33-4 3 35-4 ll. Ul. 4 8-6 3 52 -8 3 29 3 46-7 3 33 -4 3 33-5 3 18-5 3 49-3 3 42-9 4 5-8 h. m. 50 37 25 25 8 45 29 15 27 24 1 5 millim. 0-4 0-8 0-4 0-8 0-2 By the method of circles, with Manila as zero, an origin is indicated in or near the Carolines, 1.30° E. Long, and 10° N. Lat. The determination is, however, vague. It may be noted that the disturbance does not appear to have passed beneath the Pacific Ocean to stations in North America, whilst it has been transmitted to a great distance across the continents of Asia and Europe. Whether this indicates an oceanic damping eft'oct, or results from the direction of the initial impulse, is open to conjecture. (See map.) 123. October 14. C. M. D. A. Remarks. " Discovery " . . . . San Fernando . . . h. m. 7 26-9 7 24-1 h. m. 7 35 -2 7 49-4 ll. m. 25 45 millim. 0-5 0-8 The two values for C .suggest an origin somewhat nearer to San Fernando than to the " Discovery." The interval M - C for San Fernando indicates an origin some 60° distant from that station. An origin may, theiefore, be sought in the western part of the Southern Atlantic in about 35° W. Long, and 20° S. Lat. This shock could not have been recorded at Cordova, because at the time of its occurrence a tremor storm was in progress at that station. It is not likely to have disturbed the instruments in Cape Town, as shocks spreading in this direction have not so far made themselves sensible at that place (see Nos. 72, 94, and 116, &c.). The suggested origin, although it does not accord with the value M - C for the " Discovery," is ciuite in accordance with the amplitude and duration noted at that station. 83 124. October 20. C. M. D. A. Bomarks. " Discovery " . . . . Perth Irkutsk Batavia San Fernando Bidston Shide T-mrhheiit nflln Dorpat S/mssbiirg h. m. 2 58-5 3 6-8 3 15 -7 3 9 4 19-4 4 12-5 4 1-5 3 18 3 29-5 3 10 h. ni. 3 22-9 3 21-7 3 17-2 3 25 4 38-7 4 22 4 10-9 3 53-3 3 34-7 4 2 '8 h. m. 1 10 42 5 50 37 38 50 2 13 mitiinis. 0-5 1-8 0-6 10 0-2 0-5 The area disturlsed resembles that for number 122, but the position of the origin beyond being in the eastern seas is very uncertain. It docs not appear to have reached North America, but, avoiding India, it has travelled westwards across Asia and Europe. 125. October 21. C. M. D. A. Remarks. h. ni. li. ni. h. m. millinis. *' Discovery " . , . . 10 15 10 35 2 1 40 3-5 C refers to P2. Perth 10 14 1 10 21 1 1 7 3 1 Tiflis 10 13 7 10 47 8 47 1-9 Tiffis 10 8 9? 10 28 9 — Cordova — 10 21 9 1 19 0-1 Irkutsk 10 24 4 11 7 8 1 24 — m-ufsk 10 15 10 37 — — Kodaikanal .... 10 14 1 10 20 7 51 1 9 Local shock. Bombay 10 1 3 10 27 5 1 7 1 2 Calcutta 10 19 1 10 45 5 54 1 Cape Town .... 10 3 10 9-7 59 3 Victoria, B.C. . . . 10 42 — 27 1 Toronto 10 40 5 — 54 1 San Fernando . . . 10 19 6 10 52 -6 2 13 4 Paisley 10 30 — 55 1 5 Edinburgh 10 24 5 10 30 5 24 3 Bidston 10 24 10 57 7 1 25 1 4 (See facsiniiU' of trace, Dale 6.) Kew 10 23 5 11 48? 8 Shide 10 20 9 10 .54 4 1 15 2 Azores 10 27 2 10 54 7 1 4 0-5 Cairo 10 16 10 .37 1 20 2-2 Taschleent 8 41 4 10 33 -5 — — Sirassburg .... 10 10 5 — 2 — Samburg 10 11 -8 10 24 -9 Approximate origin -50° to 60° E. Long, and 30° S. Lat., S.E. of Madagascar, is represented by the surface of the globe. 126. October 24. The area disturbed c. M. D. A. Eemarks. " Discovery " . . . . Cape Town Bidston Shide Taschkent Tiflis Dorpat Sirassburg h. ni. 1 22 1 1 10 2 2-8 2 3-7 1 36 1 56-7 2 8 1 30 h. m. 1 34-2 1 15-7 2 7 1 3 0-2 h. in. 40 19 16 5 34 60 miUim. Oo 07 0-2 0-2 Possible origin, 35° E. Long, and 55° S. Lat., but this is very uncertain, M 2 84 127. Odober 29. C. M. D. A. Remarks. li. m. h. ni. h. m. millims. " Discovery " . . . . 14 35 -9 ri4 42-6] J 14 51 114 57 f 15 47 1} 1] 2 20 2-7 Irkutsk -j and V 14 .33 -7 J ll4 49 1 32 0-9 Batavia 14 30 -3 14 56 3 1 40 1-6 Kodaikanal .... 14 44 -1 15 16 1 1 5 0-5 Cajje Town .... — 15 32 5 36 Baltimore 14 19 -0 15 44 5 1 27 0-6 Victoria 14 43 -5 15 41 1 11 0-2 Toronto 15 25 -2 16 1 3 San Fernando .... 12 17-0 16 2 4 22 11 Edinburgh 15 51 -5 16 3 52 0-4 Bidston 15 1 -3 15 48 6 1 20 0-8 Kew 15 2 15 51 5 1 11 0-5 Shide 15 2 -5 15 49 2 1 35 0-5 Azores 14 41-1 15 1 1 56 0-2 Tasch/tent 14 46 -8 14 59 3 — 0-5 Tiflis 14 41 -9 15 40 -8 — 0-5 Strassburg 14 39 -5 — 2 30 — Hamburg 14 39 -8 14 59-1 ■ ~ In the register for Christchurch, Earthquake No. 158, an earthquake is entered with C for large waves at 14 29-9, M U 36-5, and with A = 11-5 millims., for October 20. It appears likely that this is a misprint for October 29. If this correction is accepted, an origin is arrived at near to 140° E. Long, and 55° S. Lat., in the extension of the New Zealand axis. The time of origin would be about 14.21, and from this origin the large waves would reach Batavia, India, the Cape, and England close upon the times they were noted to arrive. The area disturbed is represented by the surface of the globe. 128. Odohcr 30. Discovery ChristchuiTli . Perth . . . Batavia . Kodaikanal Bombay . . Cape Town Azores . . . Baltimore . . Victoria Toronto . . . Paisley . . . Edinburgh . . Bidston . . . Kew . . . . Shide . . . Tiflis. . . . Irkutsk . Irkutsk . . . Taschkent . . Hamburg h. ni. 3 47-0 53-4 13-8 44-1 51-6 45-0 25-2 3-0 21-5 41 5 27 39-6 22 14-7 19-4 56-1 16-9 19-4 14 15-5 h. m. 4 22 4 30 4 10 to 4 20 4 28 4 43 4 55 5 10 — 5 14 5 8 5 13 5 43 6 6 5 28 5 32 5 21 5 35 4 45 5 21 4 39 — h. m. 2 15 >1 1 20 1 20 40 35 40 30 1 6 40 1 29 1 18 1 3 1 32 42 50 2 15 21 millims. 2-0 >6 Pi lasts 24 •4m. P.> lasts 11 ■2m. (See facsimile of (race, Plale 6.) The records for M and A at the first four stations suggest an origin in the same district as No. 127. The values M - C for the first three entries, however, are far greater than vvould lie expected. The disturbance spread all over the world. 85 129. October 50. C. M. D. A. Bemarkfi. h. ni. h. m. Ii. m. milliras. " Piscovcry " . . . . 15 12 -(j 1.5 25-7 1 5 0-5 Clirist<;hureh .... 12 0± 15 7-3 1-3 Perth 15 14-6 15 26 1 30 0-7 Bidston 16 24-2 16 28 -8 13 Shide 16 20 -7 5 0-2 TaschJceitt 15 32-5 15 57 -4 The quantities M- C for the "Discovery" and Perth records, and the dift'erence between M for each of these two stations and that for Christchtirch, each point to an origin in the southern extension of the New Zealand axis, or 150° E. Long, and 50° S. Lat. The time of origin would bo about 14h. 57ni. Large waves would be expected to reach England, 162° distant, at 16.27, from which it may )ie inferred that the entries for Shide and Bidston refer to this phase of motion. It may be noted that there are no records from stations in Germany and Russia. We therefore have here another illustration of antipodean convergence. 130. N'uwmher 1. C. M. D. A. Remarks. h. m. h. m. h. m. millim. " Discovery" .... f 18 8-8 "1 1 18 11-9 J 18 38 I 18 41 J 1 5 0-2 Chrislcliurch .... 18 2-5 18 7-7 34 10 Strassburg 18 13 — 1 40 — The entries for Christchurch suggest an origin at a distance of 15° to the S.W. of New Zealand. With a time of origin at 17.57, the anticipated arrival of P] at Strassburg, 155° distant, would be 18h. 17m. 131. Nmemher 10. C. M. D. A. Remarks. " Discovery " . . . . Christchurch .... Irkutsk 1 Tiflis Bombay Baltimore Victoria Toronto San Fernando .... Bidston Kew Shide Taschkent Manila Strassburg Samhurg h. m. 17 24-2 17 23-0 17 59-9 17 23 -6 17 36 -9 ? 18 2-8 18 14-5? 17 44-8 18 13 -0 18 26 -7 18 41 -6 18 47-7 18 40-2 17 41 -2 18 10-9 17 33 -5 17 44-8 h. m. 17 37-4 17 29 -2 18 9-7 18 47 17 38 18 20-3 18 32 18 33 -5 18 34-0 18 52 17 52 -7 18 11 -0 _ h. m. 1 22 57 28 ? 36 35 1 36 43 >0 25 28 35 millims. 0-5 3-5 1 0-3 0-4 0-4 0-3 0-3 0-6 0-9 0-2 0-3 The values M -C for the first two stations suggest an origin like that for 129 and 130. The time at this origin would be 17h. 17m., from which it would be anticipated that the times of arrival in England would for Ps be 18.47, at Victoria, B.C., 18.37, at Bombay 18.11, and at Manila 18,2, which are not widely different from what was observed. With the exception of an area which would include Cape Town, Mauritius, Kodaikanal, Calcutta, Batavia, and Perth, and a second area represented by South America, this earthcjuake was recorded all over the world. 132. Noveinlier 1.5. C. M. B. A. Remarks. " Discovery " .... h. m. 17 (abt.) h. m. li. m. 30 millim. 0-5 Origin local. Local origin. 133. Novemher 22. C. M. 1 P. A. Eeinarks. h. ni. h. m. h. m. " Diseovei-y " . . . . 14110 14 28 49 millim. Oo 134. November 22. C. M. D. A. Remarks. " Discovery " . . . . h. 111. h. m. 21 17 li. m. I millim. 4 1 0-2 Local origin. November 23 to 25, no record. 135. Novemlicr 26. "Discovery," alioiit 13.20 or 13.26 there is a slight thickening of the trace which may correspond to a disturbance which was recorded at nearly all stations, the only marked exceptions being Christchurch, Perth, and Cordova (see No. 789 in the Shide register). Origin, 95" E. Long., 35" N. Lat. 87 II. EARTHQUAKES NOT RECORDED BY THE " DISCOVERY." In the following list dates are given for a few very large earthquakes. These disturbances, although the " Discovery " seismograph appears to have been in good working order, were not recorded by the same. The stations at which the records were obtained are indicated by their initial letters, a key to which will be found on p. 96. The time of commencement is given for the first-mentioned station only. This is noted in Greenwich Mean Civil Time (24 or Oh. equals midnight). Details respecting the observations made at the stations mentioned are contained in British Association Seisraological Circulars Nos. G, 7, 8, and 9. The only earthquakes considered are those of which the origins are at least appro.ximately known. 1902. March 17, Bi. 12..3, S., V., T., Bal., I., St., H., Central Medco. „ 20, Bi. 2.28, S., K., SF., C, Ba., Bal., I., St., H., Caucasia. „ 22, Bi. 22..54, S., K., E., SF., T., V., CT., C, B., Ko., Tr., Bal., Ch., I., St., H., U'exf of rirhria or North Mexico ? „ 24, Bi. 18.28, S., K., K, SF., T., V., Tr., Bal., I, St., H., North Mexico or IVed Indies. April 19, Bi. 2.35, S., K., E., SF., T., V., CT., C, B., Ko., Ba., P., Bal., Ch., W., I., Cor., To., Guutemah. May 2.5, S. 17.28, Bi., E., SF., V., CT., C, B., Ba., Bal., I., St., H., W. Asia ? June 11, Bi. 6.30, S., K., E., SF., T., V., CT., C, B., Ko., Ba., Bal., L, To., East of Japan. „ 16, Bi. 2.16, K., B., Ko., I., St., North-JVest India. July 5, Bi. U..59, S., K., E., SF., T., V., CT., Ko., Cai., I., St., East of Greece. „ 9, Bi. 4.2, S., K., E., SF., CT., C, B., Ko., Ba., Cai., I., St., Bunder Jbbas. August 3, Bi. 17.2, S., K., E., V., B., Ko., Ba., I., St., Central Asia ? 7, To. 9.22, I., St., Bi., North-East .lapan. „ 22, Bi. 2.56, S., E., SF., T., V., CT., B., Ko., Ba., P., Bal., I., W., Ch., Cor., To., Ka.yaria. 22, S. 15.56, Bi., E., C, Ko., Tr., I., St., Kasijaria. 23, Bi. 13.24, E., C, Ko., I., Kasgaria. 24, Bi, 2.12, S., E., C, Ko., Ba., I., Kasgaria. September 20, Bi. 6.44, S., K., E., V., C. B., Ko., P., I., St., North-West India. 24, Ch. 5.31. I., K., Bi., E., T., V., St., West Indies. November 17, V. 19.57, T., S., K., Bi., K, Cai., I., St., near Victoria, B.C. December 12, V. 23.14, T., S., K., E., SF., Tr., Bal., I., Cor., W., St., West of Suutheru California. 13, C. 17.7, B., Ko., Ba., P., Cai., V., T., S., K., E., SF., I., To., St., Central Asia. 16, B. 5.12, C, Ko., Ba., Cai., S., K., Bi., E., SF., T., V., I., To., St., North-West India. 1903. January 5, To. 22.4, Ti., Ba., B., C, K., Bi., E., Pa., SF., Ta., near Formosa. 17, V. 16.11, S., K., Bi., E., Pa., Bal, Tr., CT., I., Cor., W., Ch., P., Ti., Ta., West Indies. „ 19, P. 12.44, I., Ti., Ba., Ta., S., K., Bi., Pa., North of New Guinea. 24, V. 5.33, T., Bal., S., K., Bi., E., SF., I., Cor., P., Ti., West of Mexico. „ 24, V. 15.4.5, T., Bal, S., K., Bi., E., Pa., SF., B., Ko., Ba., Cor., Ch., Ta., West of Mexico. March 28, I. 8.3, Ti., C, S., K., Bi., E., Pa., Ferghana. April 28, B. 23.52, Ko., I., Ti., S., K., Bi,, E., P., SF., Cai., St., near Tiflis. May 28, Ti. 3.58, Mau., S., K., Bi., E., St., near TiJlU. „ 29, Bi. 9.8, S., K., E., Pa., SF., I., Ti., St., lonisches Meer. 1903. June 7, Ba. 8.49, I., CT., Mau., C, S., K., Bi., E., Pa., SR, T., St., Ho., 95° E., 30° N. „ 10, Ch. 16.49, W., P., I., Ba., CT., Mau., C, S., K., Bi., SF., T., V., Ho., St., East of Philippines. July 27, T. 10.46, V., Bal, S., K., Bi., E., Pa., SF., Az., North-West Atlantic. „ 28, Ch. 4.35, T., V., S., K., Bi., E., St., Mid Atkmtic. August 6, Bi. 3.59, S., K, SF., I., Ti., St., Caucasia. October 10, C. 17.3, B., To., I., Ti., S., K., Bi., E., SF., St., Coast of Japan? 23, I. 2.40, Ti., B., S., K., Bi., E., Pa., SF., Cai., St., 90° E., ^5° N. November 17, Ma. 20.18, Ba., I., S., K., Bi., Ho., Philippines. An inspection of the above list shows that 37 large earthquakes, which without exception originated in the northern hemisphere, did not transmit motion sufficiently far south to be recorded by the " Discovery." Inasmuch as many of these were recorded over areas represented by the three northern continents, the fact that they failed to reach the Antarctic regions can hardly be attributed to want of intensity in originated impulses. The more probable explanation for the lacuna in the " Discovery " register is that the unrecorded earthquakes represent initial efforts or blows which were not delivered in a southerly direction. Isoseists which have been drawn for earthquakes originating between New Zealand and the " Discovery " find an explanation for their form by a supposition of this description (see pp. 91 and 92) and observations on certain recent earthquakes give strength to this idea. For example, the Californian earthquake of April 18, 1906, which originated from a fault parallel to the coast of that country, gave pronounced seismograms in countries lying to the east and west of the same. With the Jamaica earth- quake of January 14, 1907, where the originating line or lines of fracture were apparently east and west, the opposite took place. In Toronto a fairly marked record was obtained while a corresponding record in Europe was small. 89 III. CONCLUSIONS.* (a) Changes in the Vertical. Changes in the position of the outer end of the peiiduhim, which is an ahmiiniura boom three feet in length, have been measured on the seismographic films at intervals of four hours, and in certain instances everj' 30 minutes. These films are strips of bromide paper each 2 inches in width and 35 feet in length. They moved Ijeneath the end of the boom at a rate of 60 millims. per hour. The total length of film brought home by Mr. Bernacchi is about 3000 feet. One millimetre deflection of the photographic trace of the outer end of the boom is approximately equivalent to a tilt of 0'5". The measurement of the displacement of these traces was undertaken by my assistants, Mr. Shinobu Hirota, and Mr. Howard Burgess, of Newport, and it is in consequence of their assistance that the analj'ses of these records have reached their present stage. The results are at present in two forms — as a manuscript register and as a series of curves drawn on squared paper. They are in charge of the Royal Society. Before the analyses of these can be completed they must be supplemented with corresponding records from barographs and thermographs. The times of total darkness, continuous light, sunrise and sunset have already been entered on the squared paper. Also, as Mr. Bernacchi remarks, tidal fluctuations, ice movements, changes in volcanic activity may also hold some relation to the wanderings of the pendiUum. It is, therefore, desirable that information relating to these phenomena should be obtained. An examination of the curves indicates that there have been many comparatively large and rapid deflections of the pendulum, particularly after its removal from the magnetic observatory to the living hut. For example, subsequent to the removal, tiltings of 10" have taken place in 20 hours. Displacements of this magnitude suggest a yielding of the foundations or parts of the brick column on which the instrument was installed. My own experience is that in England it takes about 12 months for a masonry pier to become stable. A pier made with a glazed earthenware drain-pipe has only its foundation to settle, and becomes stable more quickly. There are other deviations which may be seasonal, whilst others have accompanied marked barometric fluctuations. At certain periods there have also been changes in position of the boom, indicating tilts of 0-5" to I'O" which have approximately a diurnal periodicity. In " Discovery " local time the western excursion of the pendulum was most frequently completed about 11 p.m., whilst it was usually farthest east about 3 p.m., and this took place whether there was sim or no sun. To explain these changes, possible distortions produced by sun heat on the earth's surface have been suggested. That an accumulation of a water load in a valley apparently causes its two sides to approach each other, whilst a body of men approaching an observatory will cause a pendulum inside the same to swing towards the advancing load, have strengthened the suggestions that changes of level observed at a station might be influenced by differences in evaporation or of vegetable transpiration on opposite sides of such a building. These suggestions, although they do not directly bear upon work carried out in the Antarctic regions, have received attention.! Another suggestion which I venture to make, and it is one which, for many reasons, I think deserves * This section is reprinted, with alterations, from ' Proceedings of tlie Eoyal Society,' series A, vol. 76, 1905. t See ' British Association Reports,' 1895, pp. 115-139, and 1896, pp. 212-218. N 90 consideration, is that the observed movements are not necessarily due to tilting, but are due to electrical attractions or repvdsions. Factors to be taken into account when discussing this possibility are to be found in ' Proceedings of the Royal Society,' vol. A 76, 1905, p. 286. (b) Tremors and Pulsations. As shown in the films brought home by the " Discovery," tremors usually commence as intermittent slight thickenings. The thickenings recur at shorter and shorter intervals until there is a thickened line. This may have a width of 0-2 millim. The period of the movements they represent is probably near to that of the pendulum, or 15 seconds. The duration of a storm usually lies between 6 and 20 hours. These thickenings may develop into serrations when we see that the period has been that of the pendulum. Regular movements with amplitudes of about 0-5 millim., and periods of 60 or 120 seconds, are evidently forced vibrations, and are referred to as pulsations. These various movements have been tabulated as a register, and also entered on squared paper, with the curves showing changes in the vertical. They have been placed in the charge of the Royal Society. (c) Earthquakes. Between March 14, 1902, and December 31, 1903, although there were many days when the instrument was not working, 136 earthquakes were recorded. As none of these were felt by the staff of the " Discovery," it may be assumed that none of them originated within 50 miles of the station on Ross Island. A certain number were recorded all over the world, whilst many were noted at very distant observatories. These latter must have originated at distances greater than 500 miles. The measurements of the various seismograms are contained in the accompanying register, which, as far as possible, also contains corresponding information from 43 other stations, 38 of which have seismographs similar to that used by the " Discovery." The results of analyses point to the following conclusions : — 1. Distrih'utwns of Origins. (See Plate 2.) Out of the 136 records, 73 refer to disturbances which originated in a sub-oceanic region lying between New Zealand and the " Discovery." A certain number of these were only recorded by the " Discovery," and the exact location of their origin is very doubtful; others were recorded at Christchurch and Wellington, others again reached Perth, while some travelled as far as their antipodes. On the maps published annually by the British Association to indicate the positions of origin of large earthquakes, 12 districts are shown. These are named by the letters of the alphabet from A to L. Districts J, I, L are not of great importance. The extremely active locality, the existence of which has been made known by the work of the " Discovery," I propose to call District M. The high frequency in the relief of seismic strain in the latter region indicates pronounced brady-seismical movement, an inference which is quite consistent \\ath the existence of the active Erebus and many other recent volcanic peaks. It also suggests that New Zealand may be continued towards the south-west as a sub-oceanic ridge, accelerations in the changes of which are announced by sudden yieldings along its base. The islands of Auckland, Macquarie, and others, may indicate the existence of such a ridge, but I am not aware that there are any soundings to confirm the suggestion. Sixteen records refer to shocks which originated near Japan — the Philippines and the Celebes. Five had their centres in the Himalayan region, and six off the West Coast of South America. (See Map, Plate 2.) 2. Seasonal Frequency of Antarctic Earthquakes. The relative frequency of disturbances with an Antarctic origin in different seasons and months for the years 1902 and 1903 is shown in the following table. The numerals in the body of the table are the index numbers of earthquakes in the " Discovery " register : — 91 Jan. Feb. Mar. April. May. June. July. Aug. Sept. Oct. Not. Dec. c 9 28 43 108 rii7 1118 51 56 61 .i' bi) 1 10 31 33 100 110 54 130 62 .S f3 12 32 35 103 113 119 55 131 r63 164 ^ 1^ J 4 1.5 34 36 105 114 120 132 1 » " 5 16 r67 1.68 40 115 127 ri33 1.134 8 18 41 ri28 1X29 "a o a 27 r20 121 84 90 • ■a" ■p. 28 85 92 g 95 i ^ 77 r86 187 97 be o a 78 98 1 .a 79 89 99 'S m 80 OJ 82 Number o earthqualses .... 7 13 11 11 4 5 3 7 6 4 " distinct seismic effects — 4 12 9 11 4 5 2 6 5 3 Earthquakes which are l:)racketed occurred within a few hours of each other, and, therefore, may possibly refer to the same relief of seismic strain. In the lower line of totals each of the groups has been regarded as a single disturbance. Whichever line we take, it seems that the greatest frequency has been in April, May and June, or the first part of the winter months. The seasonal distri])ution of Antarctic earthquakes is, therefore, similar to the distribution noticed in many other countries. Dr. Omori, however, has shown that earthquakes with a sub-oceanic origin off the coast of Japan have their greatest frequency in the summer, during which season a higher average sea level more than counterbalances a diminution of load on the sea bed, due to a lower barometric pressure. The seasonal difference in load amounts to 18'3 millims. of merciuy. Whether similar conditions prevail in the Antarctic regions remains to be determined. 3. Oft the Form of Areas Dishirhed hy Large Eartliquakes. (See Plate 3.) For local earthquakes, such, for example, as are from time to time noted in Great Britain, we are prepared to see isoseists occasionally in the form of circles, but more frequently in the form of ellipses. The major axis of any one of these ellipses is usually parallel to the strike of a fault, the sudden yielding on the face of which gave rise to the shaking. If the movement originates at no great depth, the epifocal area where motion is most pronounced has been shown by Dr. Charles Davidson to lie on the side of the fault towards which it hades. With very large earthquakes, which are not sufficiently strong to be recorded over the whole surface of the world, but which may reach stations near to their antipodes, the idea of elliptical isoseists requires modification. For example, earthquakes originating in District M to the S.W. of New Zealand have been recorded to the S.E. by the " Discovery," and along a band at least 20° in width, extending in a N.W. direction as far as Britain. They may or may not be recorded in India, whilst at comparatively near places like Batavia, Manila, and Japan, lying northwards from the origin, they have been seldom noted. Also it may be added that they have not been noted at Cape Town, or at Cordova in Argentina, each about 80° distant, nor anywhere on the American continents. It would appear, therefore, that recordable earth- quake motion originating in District M may be propagated as a band running in a N.W. direction as far as its antipodes. When more stations have been established in South America, it may be found that the motion proceeds to great distances in two directions round the world. This, however, is doubtful. Earthquakes originating off the West Coast of South America have been recorded by the "Discovery" to the S.W., but the greatest length of recognisable wave-path is found towards the N.E., in which direction they have been recorded in Western Europe and also near to their antipodes in Siberia. They have not been recorded at stations we should expect them to affect were they propagated with equal intensity in an opposite direction round the world. Disturbances with origins in Japan, the Philippines, and the East Indies have been recorded as far S, N 2 92 as the " Discovery " and westwards across Asia and Europe, whilst they do not appear to have reached nearer stations in North America. On the westward route it may be noticed that the path would be sub-continental, whilst in going eastwards it would be sub-oceanic. The loudness of the sound made by a gun depends in part upon the direction in which the gun is trained with regard to the observer. In a somewhat similar manner, if we hold the blade of a spade in water and then suddenly move it, the largest waves are forced in the direction of the primary impulse. If these analogies may be used to explain why earthquakes from District 1\I are propagated more vigorously in a N.W. direction rather than in any other, one inference is that th* fault or faults from which these disturbances spring strike in a N.E. and S.W. direction, that is, they are parallel to the New Zealand axis, and they hade towards the direction of the longest path along which movement is recorded. Similar inferences may be made with regard to the origins of movements in other districts. (See Map, Plate 3.) 4. Velocity Determinations. In a few instances, when accurate data have been obtainable, calculations have been made of the speeds with which earthquake motions have been ti'ansmitted in various directions round and through the world. Speeds along paths which are continental have been compared with those which are sub-oceanic. For example, for earthquakes with origins off the coast of Eastern Asia, the rate at which waves have been transmitted across Asia and Europe may be compared with the rate at which the same travelled beneath the Pacific Ocean to New Zealand and the "Discovery." The material at my disposal does not show that there is any certain difference in speeds. Certain tables relating to speed strengthen the suggestion that, for particular phases of earthquake motion, velocity is not constant. The large waves, or P3, apparently increase in speed in quadrantal regions. Other tables relating to rate of propagation are only of value as indications of the character of motion which has reached distant stations. To this I refer in the next section. A knowledge of the time taken by earthquake waves to travel from one seismic region to another occasionally leads to the conclusion that one earthquake may be regarded as the cause of a second disturbance. Illustrations of earthquakes having originated in a district at the times when teleseismic movement reached that district are to be found in earthquakes numbered 4, 8, 45, 48, and 117. 5. The Surviving Phase of Earthquake Motion. With exceptionally large earthquakes we may obtain at very distant stations seismograms which exhibit all three phases of earthquake motion. More fi'equently, however, at such stations the record is a mere thickening of the photographic trace, a small fraction of a millimetre in amplitude, and with a duration of 3 or 4 minutes. Near to its origin the maximum motion of the same earthquake may have been pronounced, while its total duration may have extended over at least 1 hour. The test which has been used to determine the phase of motion to which the surviving tremors represented by a thickening are to be referred has been determinations of the speed with which they have been transmitted from their origin to the station at which they were observed. In a few instances the times of origin and the positions of epifocal districts have been obtained with a fair amount of accuracy, and the results relating to earthquake speeds may be regarded as relialile determinations of the same. This, however, is not the case with the majority of velocity tables which have been compiled, the reason being that they have been dependent upon data relating to times of origin and positions of centres which in all probaliility may in certain instances deviate by 5° in distance and 5 minutes in time from the truth. Notwithstanding this, as the velocities of P], P2, P3 for long arcs are respectively about 12, 6, and 3 kms. per second, although the velocities deduced for surviving phases may want in accuracy, they seem to be sufficient to suggest the type of wave to which they belong. The type determined appears to be P3, which at stations comparatively neai' to the origin is announced as an undulation of the earth's surface.* ♦ For list of shocks showing these survivals, see ' Antipodean Eeourrenoes,' p. 292. 93 6. On a Suspected Qumlmntal Acceleration in Earthquake Speed. The earthquakes here referred to are those which have Ijeen recorded at stations situated at distances of at least 90° from their origins. In well-defined seisniograms these disturbances show three ])hases of motion. The preliminary tremors, or Pi, reach stations 60° to 180° distant from origins with average ehordal velocities increasing from 11 to 12 kms. per second. These may be compressional waves. Following these, but with larger amplitudes, we find a second phase, P.. These, which are regarded as distortional waves, have over paths from 30° to 160° in length average arcual velocities increasing from 4-2 to 6 •4 kms. per second. Lastly, there is the maximum motion, or P3, which has an approximately constant arcual velocity of 3 kms. per second. For the commencement of this phase, which is apparently recorded as an undulating movement of the surface of the earth, and may therefore be regarded as being partially gravitational in character,* the velocity becomes 3'3 kms. per second. With regard to P3, this, however, is a general statement. Within 10° of an origin, the value for P3 appears to be less than 3 kms. per second, whilst in the quadrantal region it may exceed 4 kms. per second. There are also indications of variation in velocity in the antipodean regions. The values for Pj also appear to be increased in the quadrantal region. These velocity changes were first discussed in a British Association Eeport for 1900, p. 64 et seq., but the data then at hand were not sufficient to sustain any definite conclusion. The observations made by the "Discovery," taken in conjunction with oliservations referring to the same earthquakes made at other stations, have added to the material illustrating the phenomena here considered, and it is for this reason that I again call attention to the same The speed acceleration, particularly for P3, is shown in the eleven time curves (Plate 1), six of which refer to the " Discovery " register. The flattening in these curves indicates an increased speed. This usually commences at a distance of from 40° to 70° from an origin. Something analogous to these movements recorded on the surface of the earth is seen in Whewei.l's Oceanic Cotidal Chart.f In the narrowest part of the Atlantic, between Africa and South America, the lines representing the hourly change in the position of the tidal crest are crowded together. As these travel northwards into the broader, and in places somewhat deeper, water, they are more widely separated. In other words, the tidal wave travels more quickly in the broader and deeper portions of ocean than in the narrower portions, where it is retarded. Although the chart may not be " perfectly trustworthy," | it at least suggests that a seismic wave of the type P3 may be less constrained, and therefore travel more quickly in its quadrantal than in its polar region. This comparison is only intended to illustrate a form of progress, and not to suggest that the factors governing the variations in speed of the tidal and seismic waves are altogether identical. Further, the seismic wave at its antipodes shows an apparent increase in its velocity, which is the reverse of that which would be expected by a tidal wave when approaching the head of an oceanic inlet. It might be assumed that the earthquake wave passes beneath a crust and over a nucleus, into which it merges. The upper portion of such a wave would be more retarded than its lower portion. It may also be imagined that the more swiftly moving lower portion on the first 90° of its path fails to give a surface indication of its existence because its external boundaries are widening. In the quadrantal region the periphery of the boundaries is fairly constant, and it is here that we find apparent acceleration in its speed. Still farther on its journey excessive contraction of the boundaries results in retardation of the waves. 7. Antipodean Re-appearances. For some years past I have noticed that earthquakes which had their origin in the vicinity of New Zealand, and were recorded in that country, have also been recorded in Britain, particularly at Bidstone, but had not necessarily been recorded at intermediate stations. The "Discovery" records, taken in * The influence of gravitation has been discussed by Bbomwich, in 'Proc. Lend. Math. Soc' t See ' The Tides,' by Gr. H. Dabwin, p. 172. t Ibid., p. 173. 94 conjunction with those from Christchurch, Wellington, and Perth, have confirmed this observation, and we have now a number of instances where the movement from an epifocal area has travelled round and through the world, to re-appear as a recordable quantity at its antipodes. It is not affirmed that in the region between an epicentral district and its pole seismic movement did not reach the surface of the earth, but oidy that even with instruments very nuich more sensitive than the Milne type motion has not been detected. The phenomena under consideration might also be described as antipodean resurgences, convergences, focal effects or amtrecmips, each of which, however, might l>e olijectod to as implying an explanation for this antipolar relationship. In the preceding registers we find the following 19 illustrations of possible re-appearanccs, viz. : — Numbers 1, 32, 34, 51, 53, 59, 83, 89, 91, 93, 95, 96, 108, 111, 115, 117, 120, 129, and 130. Out of these it seems that with earthquakes Nos. 1, 34, 83, 89, 95, 90, 117, 120, and 129, the surviving phase has been Pg. At Hamburg, Strassburg, and other stations wheie there are pendulums with a shorter period and a higher multiplication than those of the Milne type, Pi has occasionally been recorded, e.g., this is the case with Nos. 1, 93, 111, and 130. In other instances the polar responses have been nearly simultaneous, a conclusion, however, which for many reasons may be more apparent than real. The interpolar transit of a wave of the P3 type may be compared with that of a deep-sea wave down a rapidly widening and then up a similar but rapidly narrowing estuary. The dimensions of these estuaries are assumed to bo large. When half-way on its journey the height of the wave and its energy per unit area would be less than at its commencement or its terminus. It might, therefore, traverse the central area and not be noticed, but because of subsequent convergence it might become recognisalilo at points still farther from its origin. With very large earthquakes the movements were recorded all over the globe, and from experiments now in progress at Pribram, in Bohemia, the seismograms obtained at a depth of 1150 metres, although they show a diminished amplitude, differ but little from those relating to the same disturbances recorded on the surface. The earth((uakes we have to consider are of this type, but less in magnitude. Let us imagine one of these smaller efforts to start over an epifocal cap subtending 10° at the centre, and that this expands as a ring 5° in width until it reaches the quadiantal region. The area of the cap or ring in the two positions will be approximately as 1 to 11, and if we neglect loss due to friction and assume constant energy, the intensity will be diminished in like ratio. With such conditions it seems conceivable that a disturbance might be missed in the quadrantal region and recorded at its antipodes. The distance to which motion would invade the superficial region between the focus and the quadrantal region would depend upon the intensity of the disturbance at its origin. The reappearance of Pi, which is probably a condensational wave, may bo accounted for by assuming that reffections are focussed in an antipodean region. 8. SrisiiKiiirniiis, Pakalioiis, Maffnr/orjraiii.f, a ml. /he Valuf of g. It is now w(!ll known that at certain observatories magnetic needles are frequently disturbed b^' unfelt earthquake motion. To throw liglit upon the conscipiont in'cgularities which from time to time are shown in the niagnetograms at particular stations, horizontal i)endulums have been established. The iccdi'ds given by the latter instruments are due to mechanical movements, but whether the corresponding perturbations shown in the magnetograms are due to a similar cause is by no means certain. At one station teleseismic movement may disturl) surrounding and subjacent magnetic materials, with the result that needles at that station may respond to magnetic effects, which would not be the case at stations where the neighbouring materials which had been equally disturlied were non-magnetic. At Ross Island the ]>asalts are distinctly magnetic, while Mount Erebus and other recent cones indicate that physical and chemical characters, and also the arrangement of magnetic materials, have sufffered change. The varying activity of Erebus suggests that these hypogenic processes have not yet ceased, and with seismic distui-bances it seems probable that large bodies of magnetic magmas and rocks are, at least temporarily, disturbed and altered. We might, therefore, anticipate that the larger seismograms obtained 95 by the " Discovery " would bo .iccompanied by coiresponding perturbations in the magiietograms. That a slight relationship of this description exists has already been noticed by Mr. Bernacchi, hut now that the register of the " Discovery" has been extended this may be more clearly established. When making this enquiry, large earthquakes which for various reasons were not recorded by the "Discovery" should not be overlooked. A list of these is given (p. 87). Also that the time at which disturbances of magnetic needles might be expected would probably correspond with the arrival of phase P3 must be kept in mind. To strengthen the assumption that " pulsations " are actual movements of the earth's surface, it would be of interest to compare the times when these were frccjuent with the periods when magnetic needles were unsteady or showed oscillatory movements. The fact that the magnetic rocks on Ross Island have a high density is one reason which wouW lead us to expect a marked difference between the observed and calculated values for (j. 96 KEY TO ABBEEVIATIONS USED IN SECTION III. AND IN VELOCITY DIAGRAMS. Tlie small figures on curve, fig. 606, indicate the number of observations made to obtain the points to which they are attached. P] = commencement of first phase. Po = conimoncement of second phase. P3 = maximum of third phase or hirge waves. Az. = Azores. B. = Bombay. Ba. = Batavia. Bal. = Baltimore. Bi. = Bidston, near Liverpool. C. = Calcutta. Cai. = Cairo. Ch. = Christchurch, New Zealand. Cor, = Cordova, Argentina. CT. = Cape Town. D. = Dorpat. Dis. = Discovery. E. = Edinlmrgh. H. = Hamburg. Ho. = Honolulu. I. = Irkutsk. K. = Kew. Ko. = Kodaikanal. M. = Madras. Ma. = Manila. Mau. = Mauritius. Me. = Mexico. N. = Nicolaiew. P. = Perth, W. Australia. Pa. = Paisley. S. = Shide, Isle of Wight. St. = Strassburg. St.H. = St. Helena. SF. = San Fernando, Spain. T. = Toronto. Ta. = Taschkent. Ti = Tiflis. To. = Tokio. Tr. = Trinidad. V. = Victoria, B.C. W. = Wellington, New Zealand, Natioiuxl Antarclic K.qH'ililioii, 1901-l'.)0t Mijvures I'Inlf 1 {/■AuiliqanLr Ohsn-tulimis). All ME. //>P/I/V A/^es /"' 60 f\ y^a T,/ ?n z/ X y ^o° w 60° eo° loo' i2o° iw m' " 20" w' 60' JVl/AfUT£R 100" l?0° wo" 160" MiNDINflQ N'dd y M^ y > Ta / p, B^^T" 20 40 60' 100, lEO 14-0 160' C£/f/GO MS ao' 40 60' 20' OO" 60' 80' 100" 120" \W 150' fia Oi OH/IM •/L is''- V?606 '1902 y s ,r y 7 ~~iS/^ y sy y ^ ^ ^ ^ — '-^~ 100 120 WO 160 flL ISH/I N?3i7 80 /» Ba- ^^ /•^/ 6C 40 sjt ^ B r 20 ,.- 20° 40" 60° 60° 100° 120° 140° 160° /II ISKfl /VJJI Afati Bci ^^ jj. „ niy^ Time curves of eartliquakes. CERAi r N?3 ?7 > ^ ^CT 6 7 I -^ 10' 40' 60° aO° 100° 120° I40" 160° ■S Nos. 25, 49, 50, GO, 88, and 107 refer to tlie "Discovery" register. Nos. GOG, 333, 337, 338, and 347 refer to tlie Shide register. 20° 40° 60° 80° 100° 120° Xutioiial AnUmik ExpciJiium, 1001-1904. Plate 2 {Earthqnakc Obscrvatims). National Antarctic Ex.pcdition, 1901-1'JOt. riatc 3 (Earl/iqiiakc Observations). NaliM'it Aiihurlir H.f/ifJifin,,. IDOl-lOOk l'h(/r i {Ein-/li'- ' , ' . ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^I^^^^^^^^^^^^^^^^^^^^^^^^^l Nl>. 16. Aj',-;/ iO, 1902 ^■J0 /j^ f(Vl .^.^. t Jtf No. 17. J/);-i7 21, 19(12. ■■ ■ w y Nt.. 32. J/((y 26, 1902. / ih\»/. I- f^ la 5. Jo /kn , No. 49. September 22, 1902. l-3o /hn This sheet reads from rii^ht to left. Xational Antardk Esjuditioii, 1901-1904. rinlr .") (Eiirlhi[ii(ilr Ohxnralions). -**i_. > » m M M I Xo. 77. Ai.ril 10, 1903. ■■Tt»^t*tMesi^**«Mw-«o««««B No. 82. Ajiril 29, 1903. No. 90. Jtme 8, 1903. No. 117. Septemhe,- 23, 1903. No. 118. September 23, 1903 This sheet reads from left to riaht. Aalioiial .liitaniic KxpnUlioii, 1 1101-11104. I'lidc G {lutrlliqwih' Obfermtiouf). No. 119. September 2G, 1903. No. 121. Oetoher 8, 1903. No. 127. October 29, 1903. No. 128. October ZO, 1903. This sheet reads fiom left to right. IV. ANTARCTIC OBSERVATIONS OF AURORA 1902-1903. 98 OBSERVATIONS OF AURORA. Introduction, by L. C. Bernacchi. Journal of Observations. ANTARCTIC EXPEDITION I90I-4. PLATE 7 (aurora OBSERVATIONS.) Note . Centre of Circle is position of the observer and the skip in Winter Quarters. ■'Aurora Chart"April 8-9,1903. West. Newman Jith 99 INTRODUCTION BY L. C. BERNACCHI. A RECORD of the aurora visible during the two winters 1902 and 1903 was kept. The observations were generally made by the officer who was on meteorological duty for the night — a duty in which all the members of the " Discovery's " wardroom participated. Ordinarily, the observations consisted of noting the time, position of the aurora, its altitude and amplitude, intensity, form, movement and duration. These observatiotis were entered in a special journal kept for the purpose, and a rough " chart " of the surrounding hills was supplied each night for drawing in the position with regard to the magnetic meridian (see Plate 7). Whenever the display was fairly extensive, the physicist was called and special observations were taken, such as measurements of its intensity, width of bands, altitudes, and times of special movements. On the whole the displays, although very frequent, were extremely poor, and were generally in the following forms : — 1. Faint lights, with no defined forms. 2. Luminous patches, which frequently presented the appearance of clouds. 3. Incomplete arcs, or segments of arcs, of which the brilliancy was not uniform nor the border regular. From these arcs rays would frequently shoot up intermittently. 4. Rays, or vertical shafts, separated from each other at a greater or less distance, frequently described as streamers. 5. In one or two exceptional cases irregular bands, formed of rays or vertical shafts, pressed close together and forming " draped aurorse." The faint lights and luminous patches were of the most varied dimensions, sometimes very small, and at other times occupying almost the whole of the eastern (geographical) sky ; their brilliancy was rarely much more intense than that of stars of the 4th magnitude, or even the Milky Way. They formed, as it were, a white veil over the sky, through which stars of small magnitude were plainly visible. A clearly defined arc, formed of a homogeneous luminous mass touching the horizon at both extremities, was rarely seen. From the middle of the moon's first quarter to the middle of its last quarter the aurorse were generally quite invisible. Spectroscopic observations of the aurorae were not successful, due, apparently, to the weak intensity of the light. On some occasions the characteristic yellow line near D was seen by means of a direct-vision spectroscope, but, although plates were exposed on many nights in the prismatic camera, the times of exposure varying from a few minutes to twenty-four hours or more, not the slightest trace of the spectrum could be discovered on developing the plates. The spectrum plates (Cadet) appeared to be in fairly good condition, good photographs of the spectrum of krypton gas and of the sun and atmosphere having been obtained on them. The observations of atmospheric electricity taken during the displays reveal no special eftect referable to the aurora. An examination of the journal shows that the largest number of auroras occur during the mid-winter months, June and July, and that there is some indication of May being relatively a quiet aurora month in both years, but this may be purely an accident. The small number observed in March and September is, of course, due to the large amount of daylight. A daily variation of the aurora at Winter Harbour is probably shown, the maximum occurring at about 2h. a.m., which is also near the time of the mean maximum altitude of the display. The time of maximum intensity appears to depend upon the latitude, O 2 100 it being later as we go towards the pole. Thus, at Cape Adare (Lat. 71° S.) it is about 9h. p.m., and on the "Belgica" Expedition (Lat. 71° S.) also about 9h. p.m. The aurora with us usually appeared first at about 4h. p.m., low down on the horizon, and gradually moved up towards the zenith, reaching a maximum altitude at about 4h. a.m. (See the frequency diagram, p. 126.) There are many points of interest, such as a diurnal period, a monthly period due to the moon's phases — the magnetic direction of aurora at dift'erent hours of the day — simultaneous appearance of aurorje with those at northern stations and with sudden outbreaks of solar spots, the relation with terrestrial magnetism and meteorological phenomena, &c. With regard to the direction of aurora at Winter Harbour, it is interesting to note that displays were almost exclusively confined to the geographical eastern sky, which was also the direction from which the prevailing winds blew. Auroras were seldom seen in the geographical west. Arcs and segments of arcs at right angles to the magnetic meridian (N.) were frequently recorded. In the following observations all the times given are local mean time. All auroral directions are astronomical, unless clearly stated otherwise or where the word " magnetic " indicates that the direction is magnetic. All wind directions are astronomical. Where altitudes of beams, shafts, rays, &c., are given, the altitude of the lowest part closest to the horizon is meant, unless expressed otherwise. Temperatures are in Fahrenheit degrees. The very excellent drawings reproduced in Plates 8 to 1-1 were made on the spot by Dr. Edward A. Wilson. 101 JOURNAL OF OBSERVATIONS. r Latitude 77° 51' S. ,,. „ rL,atitucle 77 51 S. WINTER Quarters -^ ^ l Longitude 166° 45' E. Magnetic Declination 152° E. 1902. April 1. — At 111. ,1.111. oljserved faint curtain of aurora in N. true; altitude about i' to 4°; weak intensity and slow movement. At Ih. 15m. a.m. aurora glow in N. true. Beams of aurora visible at intervals in N. until about 3h. 30m. a.m. At about 3h. 2m. a.m. faint beam in N.N.E., occasionally stretching up towards zenith. Temperature of the air, - 8° F. No wind, clear sky, bright moonlight. Little Ci.-s. cloud. April 2. — Time Oh. 10m. a.m. Beam of aurora seen in S.S.W., about 30° above the horizon, very faint and slow in movement. Bright moonlight. 2h. a.m. Faint beam in N., about 5° above horizon, slow movement. Temperature, - 12° F. Ci.-s. cloud, 3. Bright moonlight. April 3.— Time 4h. a.m. Beam of aurora in S.W., very faint, slow movement. Altitude from 20° to 70°. Temperature of air, - 10" F. April 6. — Aurora (very faint) observed at about 2h. 20m. a.m. At from 2h. 12m. a.m. to 2h. 22m. a.m. faint aurora arc stretching from S. to N.E. true, apex nearly E. true, altitude 15°; occasional beams moving vertically, comparatively slow movement. Intensity very low, but little stronger than the " Milky Way " ; certainly too weak for spectroscopic observations. Temperature of air, - 10" F. Clear sky, no clouds. Very faint mist over sky, giving stars a " watery " appearance. Aurora had entirely disappeared at 3h. a.m. April 7. — At 9h. 10m. p.m., faint aurora glow in E., altitude 20°, only visible for few minutes, it being soon obscured by clouds. Temperature, -11" F. Blowing a gale from >S.E. April 9.— At about 2h. 25m. a.m., M.T. (3h. 20m. p.m., G.M.T., on 8th), observed aurora in N. and N. by E. true, in the form of three streamers radiating from N. (Plate 8). Very little movement discernible in curtains, the vertical beams remaining stationary for a considerable length of time, viz., 2 minutes, and only varying in intensity. The intensity was equal to a star of between the 2nd and 3rd magnitude. Temperature, - 3° F. Wind S.E., 4. No clouds. At about 2h. 44m. a.m. the two smaller streamers had almost faded away, but the principal streamer had become more intense and stretched across zenith forming a very grand glowing arc, containing faint tints of red, but with scarcely any appreciable movement in it. No vertical beams at all. Phenomenon had entirely disappeared at 2h. 55m. a.m. April 10. — From midnight April 9 to Oh. 20ni. a.m. aurora arc visible, extending from S. true to E.N.E. Apex of arc nearly due S.E., magnetic meridian, altitude 10°. Intensity star of between 3rd and 2ud magnitude. Slow movement ; constituted of vertical beams which remained stationary for some time. Temperature of air, - 1" F. Clear sky. Only dark band of cloud in S. beneath arc. No wind. 102 1902. April 11. — Very faint aurora beams visible in S. and S.E. at between 2h. a.m. and 4h. a.m. Altitude 20° Temperature, + 6° F. Clear sky. April 13. — At about midnight, M.T., aurora glow in E., about 5° above horizon. No beams visible; intensity very weak and varying. Wind S.E., 4-5. Cloud 3, Ci.-s. April 15. — At noon, M.T., faint curtain of aurora stretching from magnetic E. to W., 10" above horizon at extremity, and 50° at W. Centre of curtain within a few degrees of zenith. Movement rather rapid. Intensity weak. Temperature, - 12° F. Wind S.E. true. Clear, but slight mist partly obscuring stars. From above date luitil May 5 no sign of aurora was seen. During the latter part of April, the blight moonlight possibly made it impossible to see, and during the first 5 days of May a heavy storm blew from S. and S.W., and the drifting snow which accompanied it totally obscured the sky. Ice in McMurdo Sound was driven out, and the open water advanced to within a few hundred yards of the " Discovery." May 6. — An aurora was observed at 4h. a.m., M.T. The display extended from S. magnetic, round through W. to N. magnetic. The most brilliant portion was in S. 15° W. magnetic and 20° in altitude. Faint red was here visible at the base of the beams. A denser part of this totally eclipsed stars of the 4th magnitude. The following are the altitudes : S. W. (magnetic) 40°, W. (magnetic) 30°, N.W. magnetic 60°, N. magnetic 20", E.N.E. magnetic 10°. Temperature at the time was + 17" F. No clouds. Wind S.E., 2. May 7.— At 8h. 7m. p.m. faint aurora glow from N.W. by N. to W. 70° N. (magnetic). Streamers hidden behind the hills, one in W. 10° N. magnetic visible 2" above summit of ridge. Temperature, - 6° F. No clouds. Wind N.E., 4. Almost entirely disappeared at 8h. 45m. May 9. — 1. At 6h. a.m. faint streamers in W. and S.W. by W. magnetic, and faint-glowing aurora cloud in S.S.W. magnetic. Temperature, - 16° F. Wind E. by N., 2. Clouds 4, Ci.-s. in W. May 9. — 2. At 9h. 7m. p.m., aurora arc at right angles to magnetic meridian position, as indicated on chart. Altitude of centre, 4°; amplitude, 60°. Altitudes 9° and 14° to 15°. Aurora glow in N.W. by W., from which beams from time to time emanated, light white with a tint of yellow. From 9h. 17m. p.m. to 9h. 22m. a very well-defined arc, perpendicular to meridian. Alti- tude 9". Extremities as before, base sharp and dark beneath. Streamers above arc 10° to 15°, extending from N.W. to E.N.E. magnetic. Intensity variable, very slow movement, more of a glowing type. Too weak for spectroscopic observations. At 9h. .32m. p.m. arc was from 12° to 14° in altitude, extending from N. 20° W. to N.E. Ijy N. (magnetic). Intensity varying greatly. At one time the distance l)etween arc and streamers and beams was only about 1°, and a dark space lay between. At 9h. 42m. p.m. aurora arc had entirely disappeared, and only glow behind hill in N. 20° W. (magnetic) remained, and from which flashes of light would occasionally move up to about 15°. When arc faded away its altitude was 14°. At times during the display a doubtful appearance of the characteristic aurora line near D could be seen. Temperature, - 28" F. No wind, no clouds. May 10. — At 8h. jj.m., M.T., aurora glow appeared in N.W. magnetic, from which streamers would occasionally emanate. Amplitude of glow, W. 7" N. to N.W. l)y N. (magnetic). Altitude of streamers not more than 5°. At 8h. 47m. p.m., bright patch in form of rough arc in N.W. magnetic. Display had almost entirely disappeared at 9h. p.m., only an afterglow remained. Tempe- rature, - 6° F. No wind, no clouds. 103 1902. May 11. — 2h. ii.m. Very faint aurora, forming an incomplete arc. Total altitude, 15°. No streamers. 6h. a.m. An incomplete arc, height about 8°, moderately bright and composed of short, closely set streamers, in patches, very fluctuating. In 15 minutes no trace visible. May 13. — 6h. a.m. After a very clear and still night in which no aurora appeared there were two very faint lieanis rising some 10° from the western horizon (magnetic). May 1-1. — 6h. a.m. Air very still, faint streamers from S. to N.W. magnetic, altitude up to 40°, also small circle of aurora light near zenith. No colours. May 31. — Two patches of aurora glow just over hill in S.W. magnetic, very faint. Time 6h. a.m. Temperature, + 13° -8 F. Wind E.S.E., 2-3. June 1. — At 8h. a.m., faint aurora beams in W.N.W. magnetic. Altitude of beams or streamers between 10° and 20°. Temperature, - 2° • 9 F. Clouds St. and Ci.-s. Wind N.E., movement rapid, visiljle for very short time. June 2. — At 6h. a.m., two faint streamers visible for short time in S.W. magnetic, altitude 15° to 40°, rising from behind hill. Temperature, - 3° F. No clouds. Wind E. by S., 2. June 5. — 2h. 15m. a.m. Very faint, but rather extensive, aurora. A glow above the hills from E. to S. (true) in a low arc on the horizon. Temperature, -16° -5. Wind E. by N., 4. There were also a few faint beams in the S., altitude between 20° and 30°. Also an indistinct curtain, or ribbon of rays, very faint, and stretching across the heavens towards the S.E., and within a few degrees of the zenith. 4h. a.m. A faint arc, 20° in height, from N.N.E. to W (magnetic). - 15° F. Midnight. Faint aurora in N.N.E. to W.N.W. (magnetic). Altitude 20°. June 6. — At 8h. a.m. Aurora curtain from N.E. (magnetic) to S. 5° (magnetic) through W. (magnetic). The curtain rose rapidly from 10° to 40° in altitude. There was very little perceptible movement among the rays, but great and sudden variations in the intensity of the light. Phenomenon lasted about 20 minutes and then entirely disappeared. Temperature of the air, - 17' F. No clouds. Wind E. by N., 3. At 8h. p.m., faint aurora glow in N.W. (magnetic). - 9° F. Clear sky. Wind N.W. The glow moved round very gradually to W.S.W. (magnetic), taking nearly two hours to reach that point; no streamers visible, altitude no more than 5°. The temperature fell to - 28° F. in early part of afternoon, but rose again very rapidly with wind. June 7. — At Oh. 45m. a.m. observed fairly strong aurora in form of rough, broken-up arc, stretching from N.E. (magnetic) to S. 20° W. (magnetic), and passing through zenith. Altitude 40° in N.E., and down to summit of hills and behind them in S. 20° W. The phenomenon can scarcely be said to have taken the form of an arc, the light was too broken up and was more of the nature of luminous clouds, difficult at times to distinguish from the Milky Way, and varying greatly in intensity. The average width of these "clouds" was from 1° to 3°. No streamers whatever were visible, and very little movement was perceptible. The only movement appeared to be from S.W. to N.E. (magnetic). The glow at times became fairly intense, but never even faintly red. Stars of 3rd magnitude were frequently entirely eclipsed. The interval in time between maximum glow and fading away of any particular patch was a matter of a few seconds, viz., from 20 sec. to 50 sec. No clouds. 4h. 2m. and 4h. 7m. a.m. were the times of the brightest displays of a colourless aurora, stretching from S.W.S. to N.E. (magnetic), with a brighter patch in the S.W.S. (magnetic) and a fainter one in the N.E. (magnetic). The intervening part showed only a glow which occasionally faded away. The only streamers seen were in the S.W. by S. (magnetic), and these were 104 1902. brightest when a patch of light from below appeared. Length of streamers 2°, altitude of display 20°. Temperature, - 19° F. No clouds, light airs from N.E. true. At 4h. 35m. a.m., dull glow in S.W. by S. (magnetic), fading away towards N.W. (magnetic). 6h. a.m. Patch with faint streamers between N.E. and N.N.E. magnetic. 8h. a.m. Parts of three arcs, apparently concentric. The outermost, starting on the horizon at E.N.E. (magnetic), rose gradually to 30° at N.N.E. (magnetic). Within this rose a smaller portion at N.E. (magnetic), extending to N.N.E. (magnetic), where, like the last, it was discontinued. At N.N.E. (magnetic) rose the third, which just cleared Observation Hill, and then, rising to 20° or 30° over the hills at W. 5° N. (magnetic), dropped to the horizon again S. or a trifle E. of S. (magnetic). The arches were formed entirely of sheafs of vertical rays. The rays were very brilliant at times in the N.N.E. and N.W. (magnetic) of the arc. The rays were throughout 10° or 20° in length, except a few persistently longer and narrower and bright rays in S. (magnetic). Duration some 15 minutes at least. Movement was not flickering, but from N.E. to N. (magnetic) and W. (magnetic) by gradual transportation of the whole arc, or i-ather fading at N.E. (magnetic) and simultaneously appearing in N. and W. (magnetic). No wind, Ijut strong wind sprung up soon afterwards. lOh. p.m. Faint aurora clouds near zenith. June 8. — -At 4h. a.m., faint auroral streamers from S. 5° W. to W.N.W. (magnetic). Altitude of display from 15° to 45°. Length of streamers 2° to 20°. Patch of auroral luminosity in S.W. by W. (magnetic). The display was evidently shining through a thin mist. June 10. — An extensive, though faint, aurora glow, extending from E. to S.E., observed at Oh. 5m. a.m. Altitude 12°. No streamers. Temperature, - 22° F. Wind, 1-2. No sign of aurora at 2h. Om. a.m. At 4h. Om. a.m. very extensive faint aurora glow over hills N.E. to S.E., with one streamer due N., bright in comparison, especially just above the hills, reaching an altitude of 20°. At 6h. a.m. the conditions were exactly the same, except that the single streamer was at N. by E. (all bearings true). Temperature at 4 a.m., - 24° • 5 F. ; at 6h. Om. a.m., - 23° F. Light wind, clear sky. At 7h. 45m. to 8 a.m., vertical beams arranged more or less closely together to form an arc, which extended from near N.E. (magnetic) to S. magnetic, across N. and W. Altitude at each extremity the visible horizon ; at its highest over N.W. magnetic, about 20°, or a little more, to 30° at the summit of the streamers. Some of the W.N.W. streamers were 30° in length, but were very faint. Some of the N.E. streamers were very bright and glowing, the glow lasting for a few seconds. The S.W. and S. streamers were all faint, long, and very narrow, sometimes from 20° to 30° in length from the horizon. All the S. streamers appeared to be inclined to the right of the vertical, all the N.E. to the left. The end of the arc at the N.E. was indefinitely reduplicated and the ends of each piece turned upward rather from the horizon. The dark section of sky below the arc was well marked. - 21° -3 F. High E. airs (true). No clouds. Apparent movement S. to E. true. No corona. June 11. — 6h. a.m. Faint arc of streamers, highest above Observation Hill, discontinued eastwards just past that point. Greatest altitude 15°, E. extremity near Mount Discovery. Streamers appeared to be issuing from a dark cloud-like space below them and extended for 3° or 4° towards zenith. The streamers had a wavy, or flickering motion, but were always faint, the most pronounced being to the W. of Observation Hill (bearings true). 8h. a.m. Auroral arc S. to E. true, formed of faint streamers. Altitude 20°. Tempe- rature, - 21° F. No wind, clear sky. llh. a.m. Faint aurora streamers, 20° in length, stretching across zenith from N. by W. S. by E. magnetic. - 28° F. Wind E.S.E. No clcuds, clear sky. 105 1902. June 13. — Ahout llh. 45m. Aurora glow suddenly appeared in S. 25° W. (magnetic) and stretched up in a nari'ow winding ribbon to about 20° above the hills. The light was fairly intense, but died out almost entirely in 15 minutes. Examined the light with spectroscope, but could not distinguish even the characteristic line near D. Temperature, - 2G° F. Wind E. Ijy N., 2. No clouds. 2h. a.m. General dift'used light from N.N.E. to S.W. (true), bright but rapidly fluctuating streamers in N.N.E. from ridge top to zenith, others shorter and fainter in E. and S.E. Very little that was defined in the S. and W. 4h. a.m. Two companion streamers in N. (true), about 45° in altitude. A faint arc of streamers in N.E to (true) S.W., strongest to the N., but faint and fluctuating in intensity. A pale difTused light in the S. 6h. a.m. Only a very faint streamer in N.E. true, and more diffused streamers over Observation Hill. 8h. a.m. About a third of a semicircular arc of vertical rays rising from the horizon at N.E. (magnetic) and attaining an altitude of from 30° to 40° in N. (magnetic), or a little W. of N. (magnetic). Intensity faint, length of beams up to 10°. Movement practically nil. At the same time an irregularly vertical streamer of light was to be seen in E.N.E. magnetic, fading out at about 50° or 60° from the horizon where it originated. -31° F. Wind E., 4. The display was higher than the faint arcs usually seen at 8 a.m. June 14. — 2h. a.m. Faint auroral glow, about 11° in altitude, S.W. by W. (magnetic). 4h. a.m. Faint auroral arc, with curtains extending from S.W. by W. (magnetic) to N.N.W. (magnetic) and rising to an altitude of 45°. One bright streamer shot up to an altitude of 75° from S.W. by W. (magnetic). 4h. 30m. a.m. One definite streamer shooting up from S. 20° W. (magnetic) to altitude of 75°. This gradually faded away and was not to be seen at 4h. 40m. 5h. a.m. Long single streamer again visible in S. 20° W. magnetic, rising to 75°. 6h. a.m. Disconnected patches of curtain, extending from N.N.W. (magnetic) to E.N.E. (magnetic) at an altitude of 30°. Between these patches long beams arose to an altitude of 70°. Phenomenon faint. 8h. a.m. A disconnected aurora arc from N.E. by N. to S.W. (magnetic), and from 20° to 30° in altitude at centre, N. extremity 10° to 15° in altitude, and S. extremity about 5°. Length of individual streamers not more than 8°, on an average about 3°. Fluctuations in the intensity of the light, but very little apparent movement. There was, however, a very gradual movement up towards the zenith and advanced to within about 3° of it at noon. At 10 minutes after noon the phenomenon had entirely disappeared. Brightest patches were very carefully examined with spectroscope (direct vision), but no lines could be distin- guished, the light evidently being too weak. Colour of phenomenon, nebulous whiteness, at times faintly green. Temperature at 8h. a.m., - 32° F. ; at about lOh. a.m. or lOh. 30m. a.m., - 43° F. ; at noon, - 32° F. Barometer low, but steady. Light N.E. airs, clear sky. June 15. — 6h. a.m. Very faint rays in N.N.E. magnetic, altitude above horizon 10° to 12°, rays 1° to 2° in length. Temperature, - 24° • 8 F. Wind N., 2. No clouds. 8h. a.m. Arc of vertical streamers, starting from horizon at N.E. magnetic, rising to 20° and 30°, its highest at N.N.W. magnetic, and falling to S.W. and S. magnetic. Streamers bright, massed together and luminous, with a greenish tinge at N.N.E. magnetic, and at this end rather short, say 10°, whereas towards S.W. and S. magnetic they were very faint, narrow, and long, say 30° in length. -36°'8F. Calm, clear sky. Intensity faint, except for a few seconds in N.N.E. magnetic, where it was greenish and close, otherwise all white or faint straw colour. Movement imperceptible. Convergence of streamers on each side of centre of arc towards W. and N. 20° W. magnetic. From June 15 to June 30 bright moonlight or overcast skies prevented any aurora being seen. P 106 1902. June 30. — At 9h. 27m. p.m., faint aurora arc from 01).servation Hill to Crater Hill, altitude 15° to 18°. It had completely disappeared at 9h. .32m. p.m. At lOh. p.m. glow over Crater Hill. July 1. — 2h. a.m. More or less permanent glow E.N.E. to E.S.E. (true), waxing and waning rapidly in intensity. Streamers occasionally E.S.E. to S.S.E. Maximum intensity in various directions at various times and in any one direction only lasting for the briefest space (4 or 5 seconds). Temperature, - 16' F. Quite calm, clear sky. Intensity at greatest = 3rd magnitude stars. Curious shadow efl'ect under streamers, extending N. by E. to S.S.E. (true). Distinct double curtain in E.S.E., altitude 40°. Upper curtain fainter than lower and occasional streamers above it, extending to altitude 40°. Streamers to the E. faint, altitude 20° to 30°. Maximum intensity equal to star of 2nd magnitude. - 13° F. Wind N. by E, 1. Clear sky. 6h. a.m. Faint glow N. to N.E., faint streamer rising 20° in that direction. - 22" F. Calm and bright. 8h. a.m. Arc formed of vertical rays, extending from N.E. magnetic through N. to N.W. (magnetic) and ending in a faint glow along the hill tops at W. (magnetic). Maximum height of arc 20°. Length of beams 10° to 15°. Intensity faint. No pronounced beams. - 22° F. Calm and no wind. Fifteen minutes previous to this there were no beams between W. and N.N.W. (magnetic), but a faint arc of luminosity and three detached masses of auroral cloud and a few faint beams in N. (magnetic). July 2. — 2h. a.m. A faint sign of auroral curtain E.N.E. to E. by S. (true), altitude 45°, lasting for a few seconds. July 4. — 2h. a.m. Some slight patches above Harbour Hill and here and there towards zenith. Light streamers also over hills to the N. of Harbour Hill. 4h. a.m. Two bands of streamers of fair brightness, of greenish-yellow light, having a sharply defined lower edge, extended in the form of fractions of arcs from just above Crater Hill towards the S., and rising in that direction as well as from a little to N. of Harbour Hill and extending in same direction. From these bands streamers radiated towards, and almost extended to zenith, but continually varying. These portions of arcs then changed their form and assumed a more zigzag, or serpentine form, and continually changed their shape, glowing and waning in the same varied manner. At 4h. 10m. a.m. a bright strand of sti-eamers shot up over the hills further to the N., extending to within 40° of zenith. This also glowed and waned rapidly and was of a greenish- yellow line. At 4h. 20m. a.m. some streamers were also seen in the S., very faint, while a confused film of streamers and patches of auroral light continued to play in the E. and N.E. At 4h. 25m. a.m., part of double arc stretching from S.W. (magnetic) to about W. (magnetic). Very faint ; not much brighter than the " Milky Way." Space between the two arcs 3° ; larger space at W. extremity, where arcs were about 20° above horizon. In S.W. (magnetic) arcs rose up from above hills. One or two isolated streamers to S. of S.W. (magnetic). Phenomenon very transient. No movement, excepting that of light glowing up suddenly and then dying out again in a few seconds. Carefully examined light with Metz direct-vision spectroscope, but could see no sign of lines. At 4h. 35m. a.m., low arc of light suddenly formed from S.W. to N. 25° W. (magnetic) and became comparatively intense, as bright as any yet observed. Altitude in N. (magnetic) 15° and rising from behind hill in S.W. (magnetic). No sign of streamers. Colour yellowish green. Examination with the spectroscope revealed the characteristic yellow line in the greenish-yellow part of the spectrum quite plainly, but very faint, and not sufficiently well defined to measure its position, Temperature, - 9° F, Light air from E. by N. true, clear sky. 107 1902. The bright glow had quite disiippe.ircd at 4h. 37ni. a.m., and at 4h. 39m. a.m. a .similar arc in position, amplitude and altitude had formed. 6h. a.m. Arched band of streamers radiating towards zenith, extending irregularly and interruptedly from over Mount Discovery, Observation Hill (3° above it). Crater Hill, and over Harbour Hill to N.N.E. magnetic. The streamers were most brilliant and strongest in the S. magnetic. A little W. of Observation Hill, where they seemed to eclipse stars of 4th magnitude, these streamers appeared, with slight variations, to be fairly constant. 8h. a.m. The right half of an arc formed of vertical lieams. Started from the horizon at E.N.E. and N.E. (magnetic) and rising to 15° or 20° over Observation Hill at N.N.W. magnetic, was lost at about 20° over N.W. magnetic, where there were the only beams of any length (viz., about 10°). Intensity faint. Light N.E. airs. - 6° F. No clouds, very clear. 7h. 20m. a.m. Nothing of the arc described al)ove was visible, but there were two patches of faint vertical streamers over N.W. (magnetic), which started at an altitude of 40° and rose to 50° and 60°. Beyond these there was no glow or arc. Meteorological conditions same as at 8h. a.m. Time, noon. Very faint auroral beam across zenith from E. to W. (magnetic), faint on account of twilight. - 9° F. Wind E., 2. No clouds. July 5. — (Plates 9 and 10.) At Oh. 30m. a.m., broad and rather brilliant double arc, extending from S.S.E. to S. (true), altitude 10° to 70°, with a distending curtain above, altitude 45°, varying quickly in brilliancy. Auroral glare in the E. and S.E. true. From Oh. 52m. a.m. to Ih. 7m. a.m., observed aurora as follows ; — Dull, luminous arc from N.N.E. to W. (magnetic), altitude (apex over Observation Hill) 18°, almost stationary and little variation in the intensity of the light. A great mass of irregular aurora clouds, comparatively bright, stretching up from S.W. by S. (magnetic), across zenith, forming light luminous patches here and there in its course. All stars plainly visible through the clouds. With Metz direct-vision spectroscope the line in yellow-green part of spectrum visible, but very faintly, and impossible to measure its position. Large prismatic camera set up, plate exposed, and instrument directed to brightest patches. [On subsequent development, nothing appeared on the plate, although it was exposed for at least 8 hours. Many plates have already been exposed, but without the slightest result. The spectrum plates used (Cadet's) are apparently in good condition, and good photographs have been procured of all the lines of krypton gas and the pure spectrum from end to end of a bright-biuning oil lamp.] Temperature, - 10° F. Light N.E. airs, no clouds. At Ih. 7m. a.m. nearly whole of eastern sky (W. magnetic) was lit up with faint aurora clouds which crossed zenith and stopped 5° W. of zenith. Fairly brilliant display from Ih. 7m. a.m. to Ih. 57ni. a.m. (true), principally in S. (magnetic) in form of two radiating streamers. At Ih. 47m. a.m. to Ih. 52m. a.m. very beautiful folding curtain rising from hill in S. (magnetic) to an altitude of 30° in S. 15° E. (magnetic). With spectroscope characteristic line near D visible, but too faint to measure. The aurora clouds seemed to drift before the wind and ultimately reached to within 45° of W. horizon (true). From Ih. 15m. to 2h. Om. a.m. the auroral display in N. (true) was very fine and at times of a greenish tint, with portions of arcs radiating from N. to W.N.W. (true), maximum altitude 45°, and to N.E., maximum altitude 45°, and one from N.E. to S.E. (true), about 30°. The arc from N. to W.N.W. quickly changed from a glow to patches of streamers, which again changed to a curtain, and then to a series of curtains travelling up towards the zenith and W. As these faded, a spiral curtain brightened up from the same N. point, reaching about 30° in altitude, with very bright base, the curtain being of a green-yellowish tint, with very faint pink hue at the base. This spiral gradually transformed into a series of streamers, and subsequently diffused into a glow. At the same time there was an arc of streamers in the S.E. true, and a P 2 108 1902. glow in the E. (true), and very faint streamers from nearly all parts of the heavens, excepting the W. true. 4h. a.m. Very faint streamers in N., N.E., S., and N.W. true, reaching up to about 50° altitude, and disappearing about altitude 20°. July 6. — 2h. a.m. Faint aurora for a few moments. A semi-arc of streamers S.E. to E. (true), crest of arc S. true, where the longest and brightest streamer was 20° to 40° in altitude. 4h. a.m. Faint streamers in the E, and S. true, strongest in S., 20° to 40" in altitude. 6h. a.m. Luminous aurora clouds in S. (true) in a somewhat serpentine form, rather bright just for a few moments, a few very faint streamers in the E. true. 8h. a.m. Single aurora beam in N.W. (magnetic), altitude 20°. Midnight. Faint arc, extending from Observation Hill to Bluff, altitude at brightest part 10°. July 7. — 2h. a.m. A few faint beams shooting up from S.W. by W. (magnetic) to the zenith ; intermittent and lasting a few seconds only in each case. 5h. a.m. Faint auroral curtains, extending from Observation Hill to S. true, altitude generally 30°. Similar phenomenon at 6h. a.m. 8h. a.m. A faint semi-arc of more or less diffused vertical beams, rising from N.N.W. magnetic to a height of 30° over W.N.W. (magnetic), where it was lost. Outside this was a fainter half-arc, rising from N.E. (magnetic) and ending at an altitude of about 20° over N.N.W. magnetic. Intensity faint. Wind nil. Temperature, - 24° F. No clouds. 6h. p.m. Bright auroral glow from E. to S.E. true, at times forming bright arc, but generally diffused glow, altitude 7°. - 14° -5. E, 1-2. 8h. p.m. Faint arc of streamers from E.N.E. to S.E. true, altitude 10°. Light E. airs. -11° F. Midnight. An arc of luminosity, no beams apparent, stretching from a height of 10° at E.N.E. magnetic to 20° at N.N.W. magnetic. Vertical breadth of arc from 6° to 8°. Intensity very faint. - 19° F. Calm, no clouds. No marked darkness beneath and in the arc. July 8. — Time 6h. a.m. E.xtensive aurora of vertical streamers, arranged chiefly in pairs, some very broad and very long, length from 40° to 60", and extending up to the zenith at N.W. (magnetic), otherwise arranged in an arc, quickly shifting upward and being replaced by another from below, so that in the N.E. (magnetic) there were sometimes the extremities of as many as four arcs in view at once. Geneial intensity faint. Brightest streamers in E.N.E. (magnetic) very low down and short, to W. magnetic high and long, S.W. magnetic long, but base touching hills. There were no true curtains, and instead of a dark appearance beneath the arc there was more inclination to a faint luminosity, which may have been thin mist. Calm. Temperature, - 12° '5 F. No clouds. July 10. — 8h. a.m. Irregular arcs formed of vertical streamers, from horizon in N.E. magnetic to W., where altitude was from 10° to 20°. In N.W. by N. magnetic one of these arcs was moving fairly rapidly to the right. The upper part of the curtain was from 30° to 40° above the horizon when it had swung round to face the observer. Intensity variable, bright at times, movement of curtain visible. Temperature, - 8° F. Calm, clear sky, no clouds. At 4h. 27m. p.m. fairly bright aurora glow, just showing upon hills from W. 5° N. to N.W. magnetic. Only lasted a few minutes. No clouds. Wind E., 2. - 4° F. July 12. — At 6h. a.m. faint curtain from N.N.E. (magnetic) through 10° of amplitude, altitude 10° to 15°, light fading in upper part, two or three vertical shafts about N. Others N. 40° W. to N. 70° W. (magnetic), altitude 30° to 45°, light fading in altitude. No part of phenomenon exceeding 4th magnitude star in intensity. Wind E.N.E., 4-5. At about 8h. 20m. a.m. observed disconnected curtain of aurora from N.N.E. to N.W. 109 1902. (magnetic). Apex nearly over 01)serviitioii Hill, and altitude 1.5°; movement slow assing a little K. of zenith to N. Patches in E. 4h. a.m. Streamers from N.E. to S. 6h. a.m. Irregular streamers from N.E. to S. and in S.M'. J\u>e 30. — 8h. p.m. Bright streamers from N.E. to S.E., altitude 10°. July 1. — Sh. p.m. Bright aurora glow in E. and S.E. behind hills. 8h. 26m. p.m. Short curtain, extending from W. 20° N. to W. magnetic. Bright as star of 2nd magnitude. Movement left to right, rather rapid. Breadth of curtain 5° to 6°, altitude 15°. Sky covered with thin mist. This disappeared in a few minutes, leaving only a diffused light, which remained some time after. Exposed spectrum plates. lOh. p.m. Very faint streamers between W.N.W. and N.N.E. (magnetic), altitude 8° to 15°. Temperature, - 20° F. Misty all round to an altitude of 80°. Calm. Midnight. Bright are between "W.N.W. 8° to N.AY. 15° (magnetic), be, 2. Clouds W. to N.W. true. July 2. — Ih. a.m. Four faint bands between N. 20° W. and N.N.E. (magnetic), altitiides 8°, 10°, 12°, and 15°. Bright confused patch S.W., altitude 12° to 20°. 2h. a.m. Faint auroral bands of light between S.W. by W. and N. 20° W. magnetic, and faint streamers S.AV. magnetic, 10° to 15° in altitude. 6h. a.m. Bright patch between S.W. and N.N.E., altitude 10° to 20°. The above auroree, July 1 and 2 (lOh. p.m. to 6h. a.m.), varied considerably in intensity and form. The intensity never exceeded a star of the 2nd magnitude, but was generally much fainter. 8h. a.m. Bright auroral arcs and streamers S. to N.W. true, and two beams over White Island (S. by E.) resembling clouds, the whole being very unsteady. Visible for 30 seconds E. to N.W. arcs, altitudes 20° to 30°, the brightest vertical ray reaching to about 40°. Weather bright. Wind E.S.E., 1-2. Temperature, - 18° F. July 3. — 6h. a.m. Eapidly changing, but generally faint aurora, extending N.E. to S.S.E. Streamers from N.E. to E.S.E., extending to an altitude of 20°. An incomplete curtain E.S.E. to S.S.E. between the altitudes of 7 and 18°. Temperatm-e, - 23°-5 F. Weather b. Wind E. by N., 6-7. lOh. a.m. Faint arc from N.N.E to S.S.E., altitude 10°. Streamer in N. 4h. p.m. Bright arc from S. to N.E. by E. Part of arc, low down on horizon, seen from Cape Armitage, altitude 5°, E. to S.S.E. 6h. p.m. Faint auroral arc in E.S.E. July 4. — 2h. a.m. Faint streamers in N., altitude 40°, three in number and quite detached. 4h. a.m. Very faint patch in N., altitude 20°. Very faint band in E., altitude 10°, parallel to hills. Above this, at an altitude of 15°, very faint streamers occasionally manifested themselves. 6h. a.m. Very faint streamers in S., altitude 10°. Midnight. Faint arc S.E. to E. July 5. — 8h. a.m. Faint arc from N.E. to S.E., altitude 10°. lOh. a.m. Faint arc from N. to S.E. July 6. — Arc from S.S.E. through zenith to N.N.E. (No time stated.) July 12. — 4h. p.m. Extensive faint aurora iu N. extending to 20° and 30°. 120 1903. July 15. — 4h. a.m. Faint aurora in E., altitude 15°. Auroral sti-eamers were seen low down on horizon from E. by S. to S.E., at about 4h. 5m. p.m., and lasted some time. Fairly bright. None of the streamers attained a greater height than 4°. July IG.— 4h. p.m. Faint arc from E. by S. to N.W. by W., altitude 20°. Midnight. Exten.sive, though very faint, auroral glow N. to S.E., reaching to 40° in altitude, fading away completely and reappearing again at intervals of about 30 seconds. Temperature, - 32° F. Weather b. Calm. July 18. — 4h. p.m. Diffused aurora light in S.E. and .streamers to altitude 8°. July 19.— 4h. a.m. Faint arc N. to E., altitude 15° and 7°. Wind N.E., 1-2. No clouds. 6h. a.m. Arcs, N. to S.E., seen at intervals of 20 to 40 seconds. Temperature, + 3° F. b. Calm. An especially brilliant aurora suddenly appeared a few minutes after 4h. p.m., in the shape of a curtain, or segment of an arc, extending from W. 20° N. to N.E. magnetic. There was more movement, both vertical and horizontal, than has yet been observed. The vertical movement of the whole display en masse was fairly rapid from S. (or N. magnetic) towards the zenith, and the horizontal motion of the huge shafts of light at one time too rapid for the eye to follow. The intensity of the light rapidly changed, frequently showing a green hue, and occasionally a faint pink. Directed spectroscope towards the light, but it was too evanescent and shifting to see anything. Altitude at first was about 10° at the extremities E. and W., and 20° in centre, but this gradually rose to 50° and 60° in the centre. The brightest display was at about 4h. 10m. to 4h. 15m. p.m. ; had almost entirely disappeared at 4h. 25m., and at 4h. 35m. there were very faint, slightly luminous patches here and there. During this special display a bright auroral glow showed up above the hills almost at right angles to the curtain. The display originated quite suddenly in the direction of Mount Discovery (E.N.E. magnetic) and flashed across the sky towards Observation Hill in a few seconds. Temperature of air, -3° -2 F. Wind E.N.E., 2-4. Clouds nil. The temperature during the last two days has been abnormally high, the maximum yesterday being +12° F. This seems to indicate a warm current from some direction, perhaps from the Eoss Sea. Brisk N. and N.W. winds prevailed yesterday. A deep red glow from the sun below the horizon appeared this morning and remained in the northern sky until about 2h. 30m. p.m. This glow extended to quite 20° in altitude. This characteristic glow appears for two or three days some weeks before the return of the sun and some weeks after it leaves. The display, which from the ship showed just above the hills, was seen by an observer in the Sound and took the form of two segments of arcs, extending from N.E. to S.S.W., with streamers radiating from the upper edges towards the zenith. July 20. — 8h. p.m. Faint auroral glow behind hills to N.E. Midt. Band of auroral light crossing zenith E. and W., 10° to 15° in length. Faint luminosity low in sky, S. to S.S.E. Temperature, - 17° F. b. Wind E. by N., 1. July 21. — 4h. p.m. Auroral display in E. July 22. — 2h. a.m. Extensive, complicated, and rapidly changing display from S. by W. to S.E. true, rising to zenith, and occasionally beyond. At first appearing as a number of parallel segments of arcs, from horizon to 70° in altitude, though as many as seven could be distinguished, some were irregular and dimly outlined ; numbers 2 and 3 from zenith were especially bright, uniform, and exact in parallelism. Later this phase changed to one in which the same portion of the sky was covered with numerous irregular patches of light. At 2h. 15m. a.m. tha arcs above and below 121 1903. re-formed. There seemed a movement in parts of the display to the N.N.E., but some patches, carefully watched, remained quite stationary. One arc distinctly passed from S. to N. of zenith, and, when S., its upper edge had the usual vertically fibrous fringe. Immediately overhead there was a mere band of light, Init to the N. the vertical fil>res again sprang up. The greatest brilliancy was about equal to a star of the 2nd magnitude. Temperature, - .30° F. b. Calm. July '2X — 2h. a.m. Faint auroral streamers converging N. to E., altitude .50". Temperature, + 2° F. Wind E., 1-2. July 24. — 2h. a.m. Ditt'used aurora in S.E. and S. and slight trace in S.E. at 4h. a.m. Midnight. Very faint nariow auroral liand, extending from N.E. to S.W. through zenith. No movement visible. Calm. b. Temperature, - 8° F. July 25. — 4h. a.m. Very faint and disconnected arc of rays, extending from N. through E. to S.W., altitude between 30° and 40°. No movement visible. Wind N., 1. Temperature, +9°-2 F. Gh. a.m. Very faint and disconnected arc of rays, e.xtending from N. to S.E., altitude 20° to 30°. This shortly changed into a narrow arc of ditt'used light, with extremely faint rays at the N. end. From 4h. to 6m. a.m. there was a continuous aurora of the above nature, sometimes extending to the S., but always faint and of the same altitude. 8h. a.m. Very faint auroral streamers S.E. and E. Midnight. Faint auroral streamer in S.E. July 26. — 2h. a.m. Faint auroral streamer in S.E. 4h. a.m. Low arc S. to S.E., altitude of apex 20". Occasional bright rays, .showing above land, from S.E. to N.E., extending to an altitude of .50". Auroral curtain in N., altitude 30° to 70°. 6h. a.m. Low arc S. to S.S.E., streamers, or rays, through E. to N. 8h. p.m. DitTused e.xtensive aurora from E. to S.W., very faint. lOh. p.m. Aurora light just showing up above hills from N.N.W. (magnetic) to S. 15° W. magnetic. Midnight. Aurora just above hills, from S. to S.W. magnetic, altitude about 5°. July 27. — 2h. a.m. Isolated patches of diffused aurora, from N. to S.E., true, altitude 10° to 30°. 4h. a.m. Streamers, or rays, from N. to S.E., true; various heights, mean 40°. The aurora was more or less visible all night, and confined principally to N.E. true, average altitude 20°. Died entirely away at about 5h. a.m., when a very thin mist of Ci.-s. cloud covered the sky, partially obscuring the light of the stars. Temperature, - 24" F. to - 31° F. Calm. b. At about 9h. 45m. p.m. an uiuisual form of aurora appeared. A band of light extended from due S. to due N., passing round through E. Breadth of band 5° and averaged 12° in altitude. Intensity fairly strong in N., where a greenish tint predominated, and occasionally a reddish hue manifested itself. The display reached its maximum brilliancy at about 9h. 50m. p.m., and had almost entirely disappeared at lOh. 10m. p.m. A few arrow-like beams were visible here and there just above the band. Midnight. A few faint streamers in E.S.E. true, and then a segment of a low arc appeared in gap. It became bright and faded away in the space of about three minutes. July 28. — 2h. a.m. Faint aurora diffused over the E., forming a narrow curtain and a few streamers scattered irregularly. 3h. a.m. Fine display of aurora, involving the whole heavens from N.W. l)y E. to S. true. Nothing ever visible in S.W. Three fine compact curtains in the E., one above the other, height R 122 1903. of uppermost approximately 60°. Three more curtains, more difl'iised, but liright and much folded, extended from N."\V. to zenith, where there were two large bright luminous clouds. The rest of the area was filled with more or less isolated streamers, small or fragmentary curtains or clouds. All was constantly changing Ijoth in shape, position, and brightness. The three N.W. cxn-tains broadened and faded, then becoming more brilliant and folded ; the easternmost disappeared, the central remained more or less, and the westernmost travelled over to the westward and disappeared at an altitude of 60°, or thereabouts. Beyond this nothing was ever seen. The eastern curtains disappeared (these had the form of ares) and were replaced by scattered fragments of curtains and a large faintly luminous cloud (sufficient to throw up a large part of the outline of Mount Erebus). At one time a single curtain appeared to have foiu- folds, each fold Ijeing very biight and lasted a few minutes. To the S.E. was an occasional curtain and numerous isolated streamers, one of these 40° in altitude, very bright. 4h, a.m. Only the remains of the above display visible in the form of two ciu tains: several faintly luminous clouds and streamers, principallj- to the N. and N.AV. true. 6h. a.m. Faint streamers to N.E. true, fainter to the S.E., and an auroral cloud over White Island. Temperature, - 18° to 30° F. Calm. b. 8h. a.m. Double auroral arc, verj' bright, l>ut not lasting longer than one minute. Streamers to an altitude of 30°. Direction E. to S.S.E. true. Temperature, - 19° F. Wind E., 2-3. No clouds. -Ih. p.m. Diffused aurora low down to S.S.E. Sh. p.m. AiU'ora glow, with a few faint streamers in S.E., altitude 10°. iSh. 45m. p.m. Faint auroral bands in g.ip, and one over Oliservation Hill. Directed prismatic camera towar to PLATE II. a. =3 ANTARCTIC EXPEDITION 1901-4. PLATE 12 (aurora OBSERVATIONS.) Weit, Newman llth. Corona, April Stb 1903. 2^ A M. ANTARCTIC EXPEDITION 1901-4. PLATE 13 (aurora OBSERVATIONS J Corona. May 3U' 1903.4*^ PM. West, Newman lith. PLATE 14. V. ANTARCTIC MAGNETIC OBSERVATIONS 1902-1904. 128 ANTARCTIC MAGNETIC OBSERVATIONS, 1902-1904. I. Introductory Statement, by L. C. Beknacchi. II. Reduction of the Absolute and Relative ObserTafions, by Commander L. W. P. Chettvtnd, R.N. III. Hourly Values of Declination, Horizontal Force, and Vertical Force on Term Days, 1902-1903, at various Observatories. IV. Magnetic Observations of the " Scotia," 1902-1904. Aatarctic Expedition 1901-4 Magiietisiri Scale of Mil c <^ . I I - I i ao 4,0 so 60 uhe parts izoLoiorejil hrowrrt tncU^caie^ foc/c-outcfops throjjjgh the. ic^^-shjeet. . ( For ttue. njodiire of the rocks see. the, lrij~g& m/zp acconzpanyvnxj the. vohune on Geology) Heights are given in feet . 129 I. INTRODUCTORY STATEMENT. BY L. C. BERNACCHI, F.R.G.S. JFinter Qmrters. — The Winter Quarters of the ship " Discovery," of the National Antarctic Expedition, were situated in latitude 77° 50' 50" south, longitude 166° 44' 45" east 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. AVith the exception of the first and last months the magnetic observations extended over the whole period. Indruments. — Besides the Fox and Lloyd-Creak instruments for the determination of Inclination and Total Force at sea, the " Discovery " was supplied with Unifilar Magnetometers and Dover Circles for the determination of absolute Declination, Horizontal Force, and Inclination on shore, and with a set of Eschenhagen variometers, or self-recording instruments, for obtaining a contiruious photographic record of the changes in Declination, Horizontal Force, and Vertical Force. The Magnetic 01:)servatory at Christchurch, New Zealand, was made the pi-imary base station of the Expedition in the Southern Hemisphere ; there the constants for the instruments were determined before sailing in 1901, and again on returning in 1904. Our thanks are due to the New Zealand Government for their courtesy in placing the Ol^ervatory at our disposal, and to Dr. Coleridge Fakr, D.Sc, and Mr. H. F. Skey, B.Sc, of the Observatory, for their valuable assistance. As the magnetic programme of the Expedition was an important one, preparations were at once made to commence observational work and to complete the construction of the Magnetic Houses for March 1. Ohservatonj Site. — The spot selected foi' the Observatory, although the Ijest 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 as little as possible disturbed by the presence of magnetic rocks ; but it would be difficult, if not impossible, in the whole length of Victoria Land to find .such an undisturl)ed locality, 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. The site selected for the houses was a low and fairly level piece of rocky ground close to the extremity of the peninsula, and at a distance of about 300 yards from the ship. (See Plates 15, 16.) The peninsula {vide Map) is about 10 miles long by a mile broad, and has iui a\'erage height of 600 to 700 feet, although the extremity where the Observatories were placed was only 30 feet above mean sea level. The rocks of which it is composed are practically of three varieties. Geologiail Fonnution. — (1) A yellow breccia which occurs in three well-marked heights, the nearest of which was 3 miles distant from the Observatory and 1400 feet high. This rock did not appear to be developed to any great extent, but occurred as a volcanic pipe surrounded by the basalt which forms the major part of the jjeninsula. (2) The trachyte found on Observation Hill, a hill three quarters of a mile distant ivom the 01:)servatory, and 750 feet in height. This hill was conical in shape, the upper half being composed of a trachyte of a specific gravity 2 • 244, and the lower half of a lava containing lapilli of a very varying composition, and with a specific gravity in one case of 2 "87. (3) A black Ijasalt which is by far the most important rock both as regards its development and S 130 physical properties. It forms the ridge called HarLour Heights, and reaches from Hut Point (the extremity of the peninsula) to the base of a conspicuous rock some 6 miles along the peninsula named Castle Rock. It forms three quarters of the rock of the peninsula, and rises to an average height of 700 feet between the two points mentioned above, and lies neaily perpendicular to the magnetic meridian. It has a specific gravity of 2 • 929, and under the microscope shows frequent plates of magnetite. Hut Point is entirely formed of it, and it was over this rock that the magnetic observations were made. Mount Erebus, 25 miles distant, rises as a full-bodied cone, with its base 12 miles distant. The specific gravity of the external rock may be taken as 2 • 9. The mountain lies north by east of the Observatory, and is nearly 13,000 feet high. Mount Terror, lying nearly 40 miles north-east of the ship, is nearly 11,000 feet high, and is composed of basic rocks of specific gravity 2-97. The two are joined l)y a ridge probably 8000 feet high, and of a similar rock to that which forms the masses of Erebus and Terroi'. There is no important land development to the southward, there being only the two islands under 3000 feet high and composed chiefly of basalt of specific gravity 2 ■ 9. These are respectively 20 and 25 miles distant from the Observatory. Mount Discovery lies south-west at a distance of '>0 miles. It is also conical, with a height of nearly 10,000 feet, and the diameter of the ba.se some 10 miles. It appears to be chiefly composed of the same basic rock so common in this locality. Turning to the west there is a totally different development of rocks. A great mountain chain, running nearly due north and south, lies at a distance of 70 miles from the ship, and rises to heights of 12,000 and 15,000 feet, and is on an average about 11,000 feet high. This chain is composed of granites, diabases, and quartzites. The granites form the core of the chain and rise to a height of 4500 feet above sea level. They vary in composition and have a specific gravity between 2-6 and 2-7. Above this occurs a diabase up to a height of 8000 feet. This rock lies practically horizontally on the plutonic rocks (though interrupted by faults) and has a specific gravity of roughly 2-8, while above it, and also horizontal, a sandstone occurs which has a specific gravity not greater than 2 " 67. It must be borne in mind that the above directions of the various land masses are triie, and that, as the Declination amounted to about 152° east, the magnetic directions are entirely different, the north-seeking end of the magnet pointing within 30° of the geographical south. I am indebted to Mr. H. T. Ferrar, M.A., for the whole of the above geological information. Obscrvatiun Houses. — The Observation Houses were constructed of large asbestos slates, screwed on to the outside and inside of a wooden framework. The laiger of the two, used for the Variation House, was 11-6 feet by 11 -G feet and 6-8 feet high. The Absolute House was slightly smallei-. Although, perhaps, small log houses would have been more suitalile, they certainly would not have been so light, comjjact, and easily portable. The asbestos houses were fairly satisfactory, but had some grave disadvantages. By the end of February, 1902, the erection of the Variation House (A) was completed, and the variometers set np and working. The Absolute House (B) was completed later, and placed 25 yards to the north of (A). For the absolute instruments a brick pillar was built up through the floor of B, 3 feet 6 inches above it, and 2 feet by 1 foot 6 inches square. The door of the house faced nearly due west (true), and narrow openings with sliding doors were made across the roof and down the north and south walls in, as near as possible, the geographical meridian, for the purpose of using a transit instrument or theodolite. Adnudh Mark. — An azimuth peg was erected a little to the south of west, 30 yards from the observational pillar. The jieg was an iron one, driven into the frozen ground to a considerable depth, and only 1 foot showing above the surface. At the top was a circular hole, across which a wire was stretched, and a light was placed )>ehind when bearings were taken from the Absolute House in the dark winter months. This mark lemained throughout the two years, and was at no time disturbed. The azimuth of this fixed mark for Declination observations was determined by a number of sun asiiniutUs in the spring ami summer of 1902-1903 and 1903-1904. 131 Thi' AhuuJiitc Ohi^erration.'f. — As soon as possible the .absolute values of Declination, Horizontal Force, and Inclination were determined, the instruments employed throughout the two years being the same, viz., Unifilar >ragnetonieter No. 25 by Em.iott Bugs., and Inclinometer No. 27 by J. Dovek. All the magnets were adjusted in their stirrups for the latitude, and foi' the Horizontal Force observation Magnet 2^)\ was generally employed. The results of the observations show how constant the moment of this magnet remaineil. The silk suspension threads were the same throughout, and were never once broken or changed. The method of observation was the same as that employed at Kew and <.)ther observatories, the only dirt'erenee being that, instead of distances 30 centims. and 40 centims., in the deflection experiment 42 centims. and 56 centims. were used, owing to the small size of the force. Magnetically disturbed days, especially in the summer, were very frequent. It was only on a few days in each month that good absolute observations were possible. It was not always easy to select quiet days. Frequently attempts at absolute observations had to be abandoned on account of too great disturbance, and, in the winter, sometimes on account of a blizzard, which made intercourse l)etween the ship aiid the shore, and observing in the small exposed alisolute house, almost impossilile. Ohervations for Local Attradkm. — In order to determine the influence of the rocks at the Observatory on the absolute values, a large tent was erected on the unbroken sea-ice in jNIcMurdo Sound in November of 1903, 1-7 miles from the nearest shore line and over a spot where the depth of water was 200 fathoms (1200 feet). Soundings showed the water to deepen quickly from 2 fathoms at Hut Point to 180 fathoms a mile further out in the Sound to the west, while 10 miles away to the west-north-west the sounding was 100 fathoms. The deepest sounding was 400 fathoms, at a point 2 miles south-east of Observation Hill, and other soundings showed that the water was much deeper to the south and to the south-east than to the north and north-west of Winter Quarters. The Sound, therefore, may be taken to be 40 miles wide, with an average depth of 200 fathoms. The ship when anchored in Winter Harbour had 9 to 11 fathoms of water, while on the north side of Hut Point the water quickly deepened to 50 fathoms close in to the land. Hut Point itself is continued half a mile to the south-west below water, in a shoal which gives soundings of from 2 to 25 and 40 fathoms. Three sets of observations were taken, viz., on November 4, 6, and 8. The results dift'er considerably from those taken on shore, and indicate a larger dip and a smaller value for the Horizontal Force, whilst the Declination seems less easterly. These observations were the standard sets taken as being most undisturbed, and used as the base to which all the observations on board the " Discovery " were reduced. The establishment of the Absolute House out on the sea-ice, although perhaps possible during the second year, would have been attended by considerable difficulty and some risk, especially as it would have been fully exposed to the heavy winter storms, and the surface of the ice, being hard and smooth, offered very little holding ground. During the first year practically nothing was known of the ice conditions in the Sound ; indeed, up to quite late in the year the ice within a few hundred yards of the ship was continually breaking up and drifting away. Observations Across Barrier. — The tent on the ice was also made the base station for the observations taken on the Ice Barrier sledge journey of November 10 to December 10, 1903, with Inclinometer No. 27 by J. Dover, having two reversible Inclination needles and two Total-Force needles. The farthest point reached was about 155 geographical miles south-east of Mount Erebus. The geographical positions of the " camps " were determined, whenever possiljle, by means of sextant observations of the sun in an artificial mercury horizon. With the exception of those taken on November 28, all the magnetic observations were taken in the evenings between 7 p.m. and 9 p.m., after the day's march. On two occasions only was it possible to observe in the open air. In most cases the strong cold wind with drifting snow prevented open-air observations, and they were then taken in the small, low sleeping tent, while the other two occupants waited outside. These observations ought to give some indication of the rate of ch.-uigc over an apparently undisturbed area in these latitudes. s 2 132 Other Absolute Ohservaiions. — The only other observations on land that require special mention are a set of Dips and Total Force taken at Cape Adare in January, 1902. The observations were taken over exactly the same spot as those of 1899, and the values obtained show little sign of secular change. During February of 1904 the " Discovery " endeavoured to penetrate into Wood Bay for the purpose of getting magnetic observations on shore, or on fast-ice removed from the shore at the bottom of the bay and as close to the magnetic pole as possible ; but the attempt had to be abandoned on account of the bay being packed with heavy close ice. During the year 1900 a set of magnetic observations were taken on shore in Wood Bay by the Southern Cross Expedition, and gave an inclination of 88° 2' south, but as the volcanic character of the rocks there is much similar to that at Winter Quarters, it is possible that this value is too small, and that Wood Bay is closer to the magnetic pole than this would indicate. A sledge journey from Wood Bay in the direction of the magnetic pole may be attended l)y considerable difficulties, on account of the lofty mountain ranges that may have to be crossed ; but at Lady Newnes Bay, about a degree further north, the mountains are comparatively low, and entirely snow-clad. A journey to the magnetic pole from here might be successful, especially during the summer months of December, Jaiuiary, and February, when the temperatures are such as permit the handling of magnetic instruments without undue inconvenience. L. C. Bernacchi. 133 II. REDUCTION OF THE ABSOLUTE AND RELATIVE MAGNETIC OBSERVATIONS. COMMANDER L. W. P. CHETWYND, R.N. Table of Contents. Page 1 I- — lutrotluction. Comparison of results at Absolute Hut ami at Ie>' Station, McMurdo Sound 133 II. — Horizontal Force results at Absolute Hut I35 III. — Inclination results at Absolute Hut J35 n^. — DfcUnation results at Absolute Hut 138 V. — Horizontal Force, Inclination and Declination results at Ice Station, McMurdo Sound . . 141 VI. — Inclination, Total-Force and Declination results during south eastern sledge journev . . 141 VII. — Results at Cape Crozier, Cape Adare, and Falkland Islands 142 VIII. — Inclination, Total Force and Declination results during western sledge journey 143 IX. — Inclination results from observations at sea, January and February, 1902 144 X. — Inclination results from observations at sea, February and March, 1904 147 XI. — Declination results from observations at sea 150 XII. — Declination results from observations on shore during sledge journeys 154 XIII. — Determination of the position of the South Magnetic Pole I55 XIV. — Comparison of results with those given by Sabine isg SECTION I. The observations considered in this paper comprise : — Observations made at Winter Quarters, by Mr. L. C. Bernacchi. Observations made during south-eastern sledge journey, by Mr. L. C. Bernacchi. Observations made at Cape Adare, Cape Crozier, and Falkland Islands, by j\lr. L. C. Bernacchi. Observations made during western sledge journey, by Lieutenant A. B. Armitage, E.N.R. Observations for Inclination, made on l)oaid the " Discovery " at sea, by Lieutenant Armit.4GE and Mr. Bernacchi. Observations for Declination, made on board the " Discovery " at sea, by Lieutenant Armitage. Observations for Declination, made on shore, by Captain E. F. ScOTT, R.N., and other officers of the Expedition. Mr. L. C. Bernacchi and Lieutenant A. B. Armitage, R.N.R., by whom the observations were principally made, had had previous experience in magnetic observation. The instruments with which the Expedition was furnished were lent by the Admiralty and comprised the following : — Two Unifilar Magnetometers, Nos. 25 and 36, by Elliott Bros. Two Inclination Circles, Nos. 26 and 27, by Dover. Two Lloyd-Creak Circles, Nos. 143 and 149, by Dover, Two Fox Circles, Nos. 28 and 29, by Dover. One Set of Eschenhagen Magnetographs. 134 The "Discovery" was also fitted with the requisite compasses for navigational and observational work on board, and compasses for sledge woi'k. Observations made at Kew, before the Expedition sailed and also on its return, showed that the instruments were in good accord with the Observatory standards and had maintained their condition satisfactorily. Observations made at Christ Church Observatory also confirmed this. All the available information has been through my hands, and the reduction of the observations has been made by me and checked. The geographical positions were supplied by Lieutenant G. Mulock, R.N., one of the officers of the Expedition. The absolute observations were mainly made in a hut (hereinafter referred to as the Absolute Hut) set up near the vessel's Winter Quarters, its position being in latitude 77° 50' 50" S., longitude 166° 44' 45" E. In November, 1903, a temporary station was erected on the ice in McMurdo Sound, at a distance of 1 '7 geographical miles from the nearest visible land, and about the same distance from the " Discovery." The depth of water under the ice at this station was 200 fathoms, its position lieing in latitude 77° 51' 1" S., longitude 166° 36' 42" E. A comparison of the resvilts of observations made at the Ice Station with those made in the Absolute Hut shows that at the latter position the magnetic conditions were largely aflTected by local attraction. The Horizontal Force was approximately 50 per cent, greater at the hut than at the Ice Station, the Inclination 2° less. The Declination at the hut was about 5° greater (more easterly) than at the Ice Station. Comparison of Results at Absolute Hut with those at the Ice Station. Inclination, S. Horizontal Force. At the Ice Statio.v. Mean of results of observations made November 4, 6, and 8, 1903 . . 86 23 -3 -0433 At the Absolute Hut. Mean of the values measured from absolute observation diagrams for November 4, 6, and 8, 1903 84 34 0686 Declination. At the Ice Station, January .30 1904 147 45 52 E. 153 9 50 E. 1 All relative observations were referred to values obtained by observation at the Ice Station. Absolute observations for Horizontal Force, Inclination, and Declination were made in the Absolute Hut by Mr. Bernacchi, at intervals averaging a month to six weeks, between the dates April 17, 1902, and January 17, 1904. The results of these observations have in each case been plotted, and curves drawn, from which the approximate change during the year can be inferred, but these results are aft'ected by uncertainties as to the diurnal variation. In view of the fact that continuous photographic records of the changes in Horizontal and Vertical Force and in the Declination were obtained by means of the Eschenhagen magnetographs, and are being measured and considered by the staff of the Observatory Department of the National Physical Laboratory, no attempt has been made to derive values of the diurnal variations from the absolute observations themselves. The results of the absolute observations afford the means of standardising the values indicated by the photographic curves. 135 SECTION II. Absolute Horizontal Force RESur,T.s. The iiistniment used throughout was Unifilar No. 25. Magnet No. 25D was used for the observations of dates Ai)ril 17, May 12, May 26, and June 30, 1902. In all subsequent observations Magnet 25A was used. To obtriin the value of the correction for P (see Admiralty ' Manual of SeientiQc Inquiry '), the mean of the values derived from the whole serie.s of observations with each magnet was taken. For Magnet 25D the result gave P = ; for Magnet 25A the result was P -= +0'25. The logarithms of the correction factors due to this latter value for the distances used in the deflection observations, viz., magnet at 42 ceiitims. and 56 centims., are respectively 1-99994 and 1-99996, and these \alues have Ijccn used in the reduction of the observations. Table I.^Horizontal Force Results. (See figs. 1 and 2, p. 137.) SECTION III. Absolute Inclination Results. The standard adopted for observations of Inclination is the value as obtained with Circle No. 27, corrected for instrumental difference as determined at Christchiirch Observatory. Comparative observations were made with the instrument at Kew before the Expedition sailed, and again on its return, and the results show that the instrument had maintained its condition satisfactorily. The conditions in the Antarctic, however, differ considerably from those at Kew, and it is believed that observations made at Christchurch afford more reliable data for instrumental differences than those at Kew. Observations at Christchurch Observatory in December, 1901, and in April, 1904, showed the following differences from the Observatory standard : — 136 Corrections to be Applied to Observations with Circle No. 27 to Reduce them to Christchurch Observatory Standard. Date. Needle No. 1. Needle No. 2. 1901 December 1904 April Mean -0-93 -0-4i -1-36 -2 65 -0-68 -1-95 In other words, the south Inclination shown by Circle No. 27 was in excess of Christchurch Observatory standard by 0'-68 for Needle No. 1 and r-95 for Needle No. 2. These values for instrumental differences are confirmed by the results of the observations at the Absolute Hut, Winter Quarters. In ten cases out of thirteen the result of observation with Needle No. 2 is in excess of that with Needle No. 1. The mean value of the differences between the needles in all thirteen cases gives the result that observations with Needle No. 2 are in excess of those with Needle No. 1 by 1''2, which is in fair accord with the difference as determined from the comparative observations, viz., T'S. The observations at the Ice Station and during the sledge journeys are not so consistent, but in view of the better conditions for observing, which in all probability prevailed at the Absolute Hut, it seems reasonable to attach greater importance to results obtained at this station than to those at out-stations under varying conditions. Table II. — Absolute Inclination Results. (See fig. 3, opposite.) Date. Needle No. 1. Needle No. 2. Mean. Remarks. Time. Inclination, S. Time. Inclination, S. Time. Inclination, 8. 1902 b. m. o / li. m. o / li. m. a , April 17 — — 3 35 p.m. 84 39 -2 — — Needle No. 3 :— 3h.22m.p.m.,84°45'-5. Mavl2 3 35 p.m. 84 49 -6 3 53 „ 84 40 -4 3 44 p.ui. 84 49 -5 Juno 30 49 „ 84 45 -6 1 „ 64 43- 1 54 „ 84 44-3 July 23 *3 7 „ 84 41 -7 *2 53 „ 84 38 -8 3 „ 84 40 -2 I'iuies are only re- corded for com- mencing No. 2 and ending No. 1. September 5 32 „ 84 40 -1 49 „ 84 4IJ -6 40 „ 84 40 -3 October 21 1 36 „ 84 39 -7 1 50 „ 84 41 -4 1 43 „ 84 40 -5 NoTember 28 9 42 ,, 84 46 -1 — — — Needle No. 5 :— 9li. 45m. p.m., 84"47' 6. December 27 — — 50 p.m. 84 52 -6 — — Needle No. 3:— Hi. 3iu. p.m., 84°51' -7. 1903 Februarj 10 1 17 p.m. 85 2 1 2 p.m. 84 58 -1 1 9 p.m. 84 59 -1 March 10 30 „ 81 56 2 43 „ 84 57 -1 36 „ 84 56 -6 May 19 .56 „ 84 43 4 1 8 „ 84 44-0 1 2 „ 84 43 -7 .Tune 28 1 27 „ 84 37 8 1 38 „ 84 39 -4 1 32 „ 84 38 -6 August 31 1 „ 8t 34 5 1 24 „ 84 31-7 1 16 „ 84 34 -6 September 30 31 „ 84 30 4 46 „ 84 31 -4 3S „ 84 30 -9 November 2 28 „ 84 32 -2 49 „ 84 32 -4 38 „ 84 32 3 1904 January 17 6 p.m. 84 51 -3 14 „ 84 51 -8 10 p.m. 84 51 -5 * In tlic case of the observations of July 23, 1902, the mean time of observation for each needle has been obtained liy applying the mean of the differences in time between other observations to the mean lime" of these two observations. No comparisons are available to reduce for instrumental differences the results obtained with Needles Nos. 3 and 5. 137 •07 C.g-.5 Figs. 1 and 2. Absolute Horizontal Force Diagrams. 1902 •o6 IcfiSi I902 Jan •o rc.g.s ■o6 IcR-S I303 Jan. Feb. Mar. Dtc. Mean, April— December, 1902 -066:3 1903 0675 Difference, 1902-1903 -0013 increase. Mean for the year 1903 -0665 These Talues are obtained from measurements of the ordinatea of figs. 1 and 2, at the middle of each month. Fig. 3. Absolute Inclination Diagram. 85 84' Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec Values for Year 1902 indicated thus: 1303 ■• " : Mean, May— December, 1902 84 43 5 „ „ „ 1903 84 36 -8 Difference, 1902-1903 .... 67 decrease. Mean for the year 1903 84 42 7 These Talues are obtained from measurements of the ordinates of fig. 3 at the middle of each month. T 138 SECTION IV. Absolute Declination Eesults. The observation for Declination consists of two operations : — 1. The determination of the direction of the magnetic axis of the suspended magnet relative to the zero axis of the instrument. 2. The determination of the azimuth of a fixed mark, to which the direction of the zero axis of the instrument is referred. Shortly, Part 2 is known as the determination of the azimuth of the mark. Observations to determine the azimuth of the mark were made on ten occasions, during the years 1902 and 1903. The results of the observations, in some cases, show considerable inconsistency, and in order to form an opinion as to the best value, the change in the sun's bearing, corresponding to the interval of time between the two successive transits of each observation, was calculated independently from the observations themselves. The observed change of bearing, as derived from the difference of the readings of the unifilar circle, represents ; — The true change of bearing + twice the amount of instrumental error due to the axis of the mirror bearings not being at right angles to the line of the telescope (or to the plane of the mirror not being parallel to the axis of its bearings). Thus, by comparing the calculated and observed change of beai'ing, we have the means of determining the error due to the above causes. In the following table are given the calculated change of Ijearing, the change of bearing observed, and the amount of instrumental error derived from each observation. Table III. Date. Time interval. Calculated change of bearing. Observed cliange of bearing. Instrumental error. 1902 December 3 9 18 26 1903 January 7 18 „ 27 October 20 21 1904 January 11 min. sees. 3 3-5 2 53 2 40-5 3 33-2 3 35-7 3 5 2 46 3 45-7 4 24-5 4 4-4 44 30 41 39 39 11 51 43 52 59 45 33 40 34 55 11 1 5 33 1 13 44 50 41 40 39 20 51 10 53 10 45 20 40 40 1 43 50 17 1 50 40 - 10 - 0-5 - 4-5 + 16 -5 - 5-5 + 6-5 - 3 -24 19 -5 -24 16 -5 -25 13-5 In the observation of date October 21, 1903, Mr. Bernacchi apparently observed with the mirror in the reversed position first ; there is no objection to this, but it accounts for the oljserved change of bearing being only 17'. It is evident from the instrumental errors deduced above that the instrument was in fair adjustment up to and including the observation of date January 27, 1903; equally is it evident that subsecjuent to that date the instrument was considerably out of adjustment so far as the mirror is concerned. The error of observation due to this cause is eliminated by taking the mean of the readings observed with the mirror erect and reversed, but the evidence of the mirror being out of adjustment, together with the inconsistency of the results of the observations taken on two consecutive days, October 20 and 21, 1903, seems to suggest the probability that the instrument was out of adjustment in other respects. 139 The results of the three observations recorded subsequently to January 27, 1903, are inconsistent with each other and individually with previous observations, and cannot therefore be considered as affording values as reliable as the previous observations, of which there appears no reason to doubt the reliability. The results of all the observations are given in the following Table IV : — Table IV. Date. Azimuth of the mark. 1902 121 16 52 121 48 7 122 4,0 23 121 22 12 121 18 02 121 16 41 121 17 25 121 10 30 121 36 10 120 52 06 9 IS 2(3 1903 January 7 18 27 October 20 . . „ 21 1904. In the recorded times of transit of the sun's limbs, there is an inconsistency in the observations of dates January 27 and October 20, 1903, amounting in the former case to 7 seconds and in the latter to 5 seconds. These may possibly have been caused Isy a change of atmospheric conditions causing refraction, or by an error in noting the times. In either case the result of the ob.servation is not affected to any appreciable extent. No explanation can be found for the apparent inconsistency of the results of the observations dated December 9 and 18, 1902, but they appear to be quite outstanding, and it is considered advisable to discard them. Omitting the results which appear to be doubtful, viz., those of dates December 9 and 18, 1902, October 20 and 21, 1903, and January 11, 1904, there remain the following : — Date. Azimutli of the mark. 1902 121 16 52 121 22 12 121 18 02 121 16 41 121 17 25 26 1903 18 „ 27 Mean 121 18 14 This mean value has been used in reducing the observations for absolute Declination. The omission of the doubtful observations confines the accepted results to a short period of two months, December, 1902, and January, 1903, and it has therefore been necessary to assume that the azimuth of the mark did not alter during the two years in which the absolute observations were made. In this connection it is of interest to note that the mean of the results of the observations of dates October 20 and 21, 1903, gives the value 121° 23' 20", which differs by only 5' 6" from the accepted value. The mean of the results of all observations gives the value 121° 28' 26". In reducing the absolute Declination observations, no correction for tor.sion has been made, the necessary data not having been recorded. T 2 140 Table V. — Absolute Declination Results. (See figs. 4 and 5 below.) Date. Time. Declination, E. 1902 h. m. / May 13 „ 26 5 10 p.m. 4 1 „ 153 59 152 53 June 30 1 56 „ 1.53 3 July 5 „ 21 58 „ 1 5 „ 152 39 152 36 „ 21 4 26 „ 152 29 „ 22 30 „ 152 23 „ 22 3 22 „ 152 36 September 5 October 21 5 36 „ 38 „ 152 22 152 14 21 3 34 „ 152 9 31 4 48 „ 1.52 34 December 5 24 „ 152 13 5 3 51 „ 152 12 27 4 1 „ 152 .50 1903 January 7 30 3 8 „ 4 8 „ 153 3 153 4 February 10 15 3 22 „ 5 10 „ 152 44 152 55 March 10 4 21 „ 152 27 May 19 June 28 3 35 „ 4 6 „ 151 59 152 3 August 21 „ 21 1 6 „ 4 10 „ 152 24 152 20 „ 31 3 49 „ 152 21 September 29 30 4 11 „ 3 37 „ 152 2 152 22 November 2 3 33 „ 152 1904 January 12 17 5 7 „ 3 27 „ 152 16 153 8 Figs. 4 and 5. Absolute Declination Diagrams. 1902 I55f - — ^ © ^10> .^^ y N 1 1 1 1 1 May June July Aug. StPT. Oct. 1903 Feb. 1903 1903 Jan. Feb. Mar. May June July Aug. Sept. Oct. Nov. Dec. Jan. FeB.1904 Mean for 8 months, May to December, 1902 152 33 -9 E. 1903 152 7-5 E. Difference 26 4 decrease. Mean for year 1903 152 16 4 E. These values have been obtained by measuring the ordinates of the curves in figa. 4 and 5, at the middle of each month. 141 SECTION V. Horizontal Force, Inclination, and Declination Results at the Ice Station, McMuRDo Sound. Table VI. — Horizontal Force Eesults. Date. Time. X, Horizontal Force, c.g.s. m, moment of coUimatoF magnet at temperature 0° C. 1903 November 4 6 8 h. m. 4 7 p.m. 4 22 „ 4 48 „ -04327 04303 04358 943-6 942-3 942-2 Table VII. — Inclination Results. Date. Needle No. 1. Needle No. 2. Mean. I 1 Time. Inclination (S). Time. IncUnation (S). Time. Inclination (S). 1903 November 4 . . . 6. . . 8. . . 1904 January 30 . . . h. m. 44 p.m. 11 58 a.m. 45 p.m. 5 5 „ 86 24 1 86 24-5 86 25 -1 86 22 -6 h. m. 57 p.m. 11 „ 55 „ 5 18 „ 86 23 -3 86 21 -0 86 22 -6 86 23 -2 h. m. 50 p.m. 4 „ 50 „ 5 11 „ 86 23 -7 86 22 -7 86 23 -8 86 22 -9 Declination Result. Date. Time. Declination. 1904 January 30 h. m. 3 57 p.m. 147 45 52 E. SECTION VI. Inclination, Total Force, and Declination Results during South-Eastern Sledge Journey, November, 1903. Observations for Inclination and Total Force were made by Mr. Bernacchi with Dip Circle No. 27. The Ice Station, McMurdo Sound, was used as a base station, and observations for Inclination and to determine the Constant A, for reducing Total Force observations, were made here on November 8 as follows -. — 1903 November 8 h. m. 12 p.m. 47 „ 1 29 „ 4 48 „ Inclination, with Needles 1 and 2, polea direct. Obsei-vation, -with Needles 3 and 4, to determine A. Inclination, with Needles 1 and 2, poles reversed. Observation for absolute Horizontal force, with Unifilar No. 25. The results of these observations give the value of Constant A = log-i 1-55706, which has been used in reducing the Total Force observations. The observations for Inclination have been corrected for instrumental difierences (see Section III.). 142 In all the Total Force observations, in that part made with Needle No. 4 weighted, for one particular position of instrument and needle, viz., when the face of the instrument was west and face of the needle towards the face of the instrument, the recorded readings are nearly constant with value 63|°, whereas the readings for the three other positions of instrument and needle vary from 68° to 61°, but are in each observation consistent. The maintenance, ai different stations, of a constant reading for the one position seems to indicate that the direction of the needle was governed by some mechanical cause ; the reading obtained in this position has, therefore, been discarded and the mean of the three other combinations adopted. The needle appears to be well balanced, and the error introduced by the omission is inconsiderable. Table VIII. — Inclination and Total Force Results. Needle 2. Mean. Lati- Longi- Total Date. tude, tude, Force, S. E. Time; Inclination, S. Time. Inclination, S. Time. Inclination, S. e.g.8. 1903 c , o , h. m. o , h. m. o , h. m. , November 14 78 13 168 30 — 85 58-6 — 85 58-9 9 17 p.m. 85 58 -7 6993 17 78 33 170 22 9 2 p.m. 85 45-8 9 4 p.m. 85 42 -4 9 3 „ 85 44 -1 6933 20 78 47 172 16 8 28 „ 85 26-4 8 35 „ 85 26 -4 8 31 „ 85 26 -4 6923 23 79 2 173 8 8 26 „ 85 4-8 8 37 „ 85 1-0 8 31 „ 85 2-9 6945 26 79 17 174 55 8 38 „ 84 57 -7 8 49 „ 84 57 -4 8 43 „ 84 57 -5 6905 28 79 32 176 1 5 5 J, 84 50 -4 4 59 „ 84 45 -6 5 2 „ 84 48 -0 6896 December 8 78 18 169 33 8 13 „ 86 0-0 9 29 „ 85 58 -7 8 51 „ 85 59 -3 0-7280 Table IX. — Declination Results. Fmm Ohservatiom taken h// Lieutenant C. RoYDS, JLN., with a Prismatic Compass. Date. Latitude, S. Longitude, E. Declination, E. 1903 / // o , „ o , November 13 ... . 78 4 15 167 49 25 149 55 l-l 78 11 4 168 14 30 146 1 15 78 16 6 108 46 15 143 22 18 78 31 54 170 25 15 144 58 20 78 51 29 172 8 15 141 47 22 79 5 20 173 16 5 139 1 25 79 20 10 174 55 10 137 53 28 79 32 42 176 2 15 137 34 December 2 79 1 41 172 31 15 139 47 7 78 24 2 169 32 148 24 SECTION VII. Results at Cape Crozier, Cape Adare, and Falkland Islands. Frmn Observations bi/ Mr. Bernaccui with Circle No. 27. Date. Time. Inclination, S. Total Force, c.g.s. Cape Crozier Cape Adare Cape Adare 1902 January 23 9 1904 February 24 h. m. 48 p.m. 6 48 „ 7 „ 85 44 -1 86 34-2 86 34-7 -6891 -6806 -7055 No explanation can be found for the large difierence in the Total Force results at Cape Adare. 143 Results at Falkland Islands. From Observations hy Mr. Bernacchi with Unifilar No. 25. Date. Time. Inclination, S. Absolute Horizontal Force, C.g.9. 1904 July 13 „ U „ 19 „ 19 li. m. 10 59 a.m. 10 1 „ 10 p.m. 10 4 a.m. 46 9-2 46 6-8 -26937 -26972 SECTION VIII. Inclination, Total Force, and Declination Results during Western Sledge Journey, December, 1902, and January, 1903. From Observations hy Lieutenant Armitage. Table X. — Inclination and Total Force Results with Circle No. 27. Date. Latitude, S. Longitude, E. Time. Inclination, S. Time. Total Force, c.g.s. 1902 December 1 . . 10 . . 1903 January 4 . . 11 . . 77 46 77 52 77 47 77 52 164 49 162 35 158 43 162 7 h. m. 8 37 a.m. 12 p.m. 6 42 „ 4 50 „ 86 31 -9 86 19 -0 86 23 -1 86 15 -4 h. m. 7 p.m. 6 40 „ -6855 -6849 Table XL — Declination Results. From Observations with a Prismatic Compass. Date. Latitude, S. Longitude, E. Time. Declination, E. 1902 » , o , h. m. o , December 1 .... . 77 46 163 32 7 27 p.m. 155 44 15 77 53 161 17 6 8 „ 158 22 21 77 52 160 42 6 4 „ 164 34 26 77 48 159 38 6 22 „ 163 55 27 77 46 159 13 1 50 „ 165 25 27 77 42 159 3 6 28 „ 164 2 1903 January 4 77 47 157 26 5 50 „ 170 50 5 77 52 156 59 4 22 „ 172 23 144 SECTION IX. Inclination Results. From Observalions by Lieutenant Armitage and Mr. BernaCCHI, on board tlie " Discovery " at Sea between dates Jamtary 23 and February 8, 1902. The " Discovery " was swung for deviation of compasses — January 22, in latitude 77° 24' S., longitude 169° 5' E. ; and February 8, „ 77° 51' S., „ 165° 30' E. The resulting deviations are in close agreement and indicate that the magnetic condition of the vessel was the same on both occasions, and it may therefore be assumed that this condition did not alter between the above dates. Adopting the notation of the ' Admiralty Manual of Deviations of the Compass,' the mean value of the coeflBcients of deviation for the compass in the observatory cabin are : — A. B. C. D. E. -0°25' -13° 55' -3° 49' +1°24' +0°02' On January 23, in latitude 77° 25' S., longitude 169° 30' E., observations for Inclination were made on board, with the ship's head in four nearly equidistant directions, approximating to north, east, south and west. The results of these observations are the only data, obtained in the Antarctic, available for reducing the observations made in these regions. Observations for X = — '-— — — — — , that constant ratio which has such an important Absolute Horizontal iorce bearing on all observations made on board, were obtained at Spithead, August, 1901 giving result 0-975 Lyttleton, June, 1904 „ 0-976 Falkland Islands, July, 1904 „ 0-970 Spithead, September, 1904 „ 0-973 Mean 0-973 From the results of observations for Inclination made by Mr. Bernacchi and Lieutenant Armitage during their sledge journeys, and from those made by Mr. Bernacchi at the Ice Station, McMurdo Sound, a preliminary chart of lines of equal Inclination was drawn, from which the absolute Inclination at the position of swinging, January 23, was estimated to be 86° 25' S. The values observed on board were, with ship's head, (1) N. 4 E. (magnetic), 87 41 Inclination S. (2) S. 81i E. ( „ ), 86 21 (3) S. 3 W. ( „ ), 85 37 (4) S. 82 W. ( „ ), 87 12 The formulfe adopted for determining the errors of Inclination due to the direction of the ship's head are s cos f + N = (cos f + sin B) sec ^' tan d' for all observations made with ship's head between NE and NW, SE „ SW, and s cos f +N = {(1-2 sin D) sin (-sin C} cosec ^ tan 6^ 145 for all observations made with ship's head between N.E. and S.E., N.W. „ S.W. In these formulte f represents the direction of ship's head as shown by the compass. ^ represents the magnetic direction of the ship's head. 6' represents the uncorrected Inclination as observed. B, C, and D represent the coefficients of deviation of the compass. N represents the natural tangent of the Inclination value after correction for errors due to the direction of ship's head. s represents the maximum effect on the Inclination, which would be caused by the induced magnetism in a horizontal soft-iron rod in the fore and aft direction, one end of which is immediately below the dip circle (rod g of the ' Admiralty Manual of Deviations '). By means of these formulae, values of s cos f + N, for positions (1), (2), (3), and (4), were calculated to be (1) tan-i 18-365 = 86 53 Inclination S. (2) tan-i 15-821 = 86 23 (3) tan-i 16-198 = 86 28 (4) tan-i 17-793 = 86 47 Mean = 86° 38' Observations at Spithead in August, 1901, showed that s was zero, and therefore the above values represent N (the natural tangent of the Inclination, corrected for errors due to the direction of ship's head). The values of the Inclination so reduced should agree for all directions of the ship's head ; the differences of the above results from the mean are, however, considerable, viz. : — (1) +15 (2) -15 (3) -10 (4) +9 These differences are consistent with an error which has a maximum when the ship's head is N.W. or S.E., varying as the cosine of the azimuth of the ship's head measured from these points. Such an error might be caused by an elongated mass of horizontal soft. iron situated at an angle of 45° to the keel of the ship, and having one end directly under the position of the dip circle. The error is similar to that which is indicated in the formulte, by s cos (, but having a maximum value when the ship's head is N.W. or S.E. instead of N. or S. The differences as found were plotted and a curve drawn, from which the correction for any direction of ship's head could be measured, and these corrections were applied as a constant for the particular direction of the ship's head during each observation. The corrections were applied to the observed readings after the correction for instrumental differences. From the value of Inclination so deduced, the value of N was calculated by means of the formulae N = (cos f+sin B) sec (' tan 6' and N = {(1-2 sin D) sin f - sin C) cosec ^ tan 6'. To ascertain the correct values of the coefficients of compass deviation for different Inclinations and thus determine the necessary corrections to compass readings, it was assumed that over the area under consideration the Total Force might be treated as constant. This assumption is supported by the Total Force results during Mr. Bernacchi's S.E. sledge journey and U 146 by the results of Lieutenant Armitagk's observations during his western journey, also by consideration of the values given on Sabine's charts. Assuming the Total Force to be • 6876, as determined from Mr. Bernacchi's observations at the Ice Station, McMurdo Sound, values of Horizontal Force were calculated for each degree of Inclination between 82° and 87° S. The same assumption, viz., that the Total Force did not vary, was also employed in determining the correction to Inclination observations due to the vertical force of the ship. This correction at the position of swinging, January 23, 1902, was ascertained by comparison of the Inclination value estimated from the preliminary chart with the value of N taken as the mean of the results of the four observations near N., E., S., and W., the true vertical component and the vertical component on board being also calculated. Now if 6 be the absolute Inclination, On „ Inclination on board as indicated by N, H „ absolute Horizontal Force, Hi ,, mean Horizontal Component on board, V ,, absolute Vertical Force, Vi ,, mean Vertical Component on board. then from which but V Vi tan 9 = — and tan 6^ =■ ^ , H Hi tan 9 = tan 9^ x ^rr^ ; ViH Hi = AH, therefore tan 9 = tan 9n x =-r- . Over the area under consideration the ratio V/Vi will not vary to any appreciable extent and may be considered constant. A. is constant, therefore AV/Vj is constant. Its value as deternuned from observations of January 23 is 0-9393 = log-i 1-9728, which has been applied as a constant to the values of N as determined from each observation. The corrections for instrumental differences were determined from observations made at the Ice Station, McMurdo Sound, with Circle No. 27, and the two instriunents used for the observations considered in this section, viz.. Fox Circle No. 29, used by Mr. Bernacchi ; and Lloyd-Cieak Circle No. 143, used by Lieutenant Armitage. The results showed that the readings by these instruments were in excess of the standardised value derived from observations with Circle No. 27 by the following amounts ; — Excess over Standard. Fox Circle No. 29 2' -7 Needle No. 1. Needle No. 2. Lloyd-Creak Circle No. 143 3' -2 5' -7 These differences have been applied to the observations for Inclination made with the instruments quoted. To .sum up : The following corrections have been applied to the observed readings in the order given below : — 1. Correction for instrumental differences. 2. ,, ,, unsymmetrical iron. 3. ,, due to direction of ship's head. 4. ,, ,, vertical force of ship. The resulting values of Inclination are given in Tables XII. and XIII. 147 Table XII. — Inclination Results. From Ohservatimif: by Mr. Bernacchi uith Fox Circle No. 29. Date. Latitude, S. Longitude, E. Inclination, S. 1902 o / o , o , January 24 78 15 176 30 85 5 25 78 28 183 26 84 1 26 78 27 188 31 83 27 28 78 24, 191 31 83 50 29 78 2-i 197 24 84 5 31 70 15 207 40 82 36 February 2 77 2 202 6 83 46 5 78 12 190 12 84 35 6 77 52 180 43 85 48 7 77 10 169 40 86 35 8 77 23 164 37 86 5 Table XIII. — Inclination Eesults. From Observations by Lieutenant Armitage vdth Lloyd-Creak Circle No. 143. Date. Latitude, S. Longitude, E. Inclination, S. 1902 January 24 25 27 28 29 February 1 2 7 78 26 78 30 78 6 78 30 78 25 76 31 77 36 77 9 178 24 186 10 187 6 193 29 199 19 207 37 201 33 167 13 84 33 83 56 84 36 83 34 82 48 82 38 83 7 86 83 It is satisfactory corroboration of the method of reduction that the re.sult of the observation made by- Mr. Bernacchi, January 31, when the .ship's head was S. 71° W., is in close agreement with the result of the observation by Lieutenant Armitage, February 1, in nearly the same latitude and longitude, when the ship's head was S. 55° E. SECTION X. Inclination Results. From Observations made by Air. BERNACCHI ami Lieutenant Armitage on board the " Discovery " at Sea, in the Antarctic, February and March, 1904. The instruments used were : — By Mr. Bernacchi, Fox Circle No. 29. „ Lieutenant Armitage, Lloyd-Creak Circle No. 143. The instrumental differences were ascertained by comparison of instruments at the Ice Station, McMurdo Sound (see Section IX.). The "Discovery" was swung for deviation of the compass, February 21, 1904, in latitude 74° 26' S., longitude 165° 30' E. The resulting coefficients of the deviation were as follows, the notation being that of the ' Admiralty Manual of Deviation of the Compass ' : — A. B. C. D. E. - 2° 48' - 3° or The remarkable difference between these results and those obtained in the year 1902 (see Section IX.), in approximately the same locality, indicates that the magnetic conditions at the compass position in the u 2 148 observing cabin had, in the interval, undergone very great change ; possibly this change was due to alterations in the stowage of stores in the ship. The difference is so large that it has been considered advisable to reduce the observations made in 1904, by means of data obtained in the same year only. In addition to the swing for compass error already mentioned, the vessel was also swung at Auckland Islands, Lyttleton, Falkland Islands, and Spithead. Relative observations for horizontal and vertical force on board were made at Lyttleton, Falkland Islands, and Spithead. The results are given in the following Table XIV., the notation being that of the ' Admiralty Manual of Deviations of the Compass,' and of the ' Antarctic Manual ' : — Table XIV. Place. Date. X. ii. ff- s. Ai. A. B. C. D. E. 1904 » , o , o , o , Wood Bay . . . February 21 — — — — — -2 48 -3 1 Auckland Islands . Marcli 29 — — — — — -0 14 + 39 -0 15 + 1 14 -0 30 Lyttleton .... June 7 0-976 1-013 -0-014 -0-013 0-996 -0 '10 + 19 -0 22 + 1 43 -0 13 Falkland Islands . July 19 0-970 0-961 -0-011 -0-011 0-993 -1 18 + 1 29 + 22 + 1 + 3 Spithead .... September 10 0-973 -985 + 0-0016 0-990 -0 58 + 1 27 + 18 + 1 7 -0 2 The results of the observations at Lyttleton, on investigation, .show certain inconsistencies, and are on some points doubtful ; they have therefore not been used in the following reductions, except in the case of A, the mean values of the constant coefficients used being 0-973, ■0-0055, Ai = +0-991, D = +1° 7'. From the data in Table XIV. (omitting the Lyttleton results), by separate combinations, the values of the parameters P, Q, c, and / (' Admiralty Manual of Deviations of the Compass '), were calculated, with the following results : — Combination. P. Q. c. /• Spithead and Wood Bay „ Auckland Islands Falkland Islands Means + 0-0027 + -0029 + -0038 + -0001 + -0008 + 0006 + -0040 + 0-0031 + 0-0014 + 0-0016 + 0002 -0-0005 + 0-0031 + 0-0005 + 0-0028 + 0-0004 From these mean values, curves were drawn showing the values of for all values of c tan e f tan 6 P A ' ' X ' AH' -^ XH' 6 (the Inclination) and H (the Horizontal Force), the values of H being calculated on the assumption that over the area under consideration the Total Force did not vary (see Section IX.), and the value of 6 was taken from N, the natural tangent of the Inclination as calculated from each observation. By means of the data obtained from the curves and employing the formulae B = - ( c tan 6 + — ) (' Admiralty Manual of Deviations '), and G = i(/tan e+ ^), i)' 149 the values of the coefficients B and C were obtained, and the deviation of the compass calculated by means of the formula (' Admiralty Manual of Deviations ') Deviation = B sin {'+0 cos i+D sin 2^, assuming the coefficients A and E to be zero. To obtain the values of V, the vertical force on board due to permanent magnetism, and d, the coefficient of the vertical force due to induction in soft iron, of the formula V AiH + d tan = A [' Antarctic Manual '), the following procedure was adopted :- If Z be the absolute vertical force, (a) l+d+ -= = fj. ('Admiralty Manual of Deviations'). AtSpithead . . . Z = +0-438 and /x = 0-985, „ Falkland . . . Z= -0-280 „ /x = 0-961. Using these values in the above formula («), the resulting values of V and d are V = +0-0037, d = -0-0218. Then for each inclination observation the correction due to the vertical force of the ship (A) was calculated by means of the formula A = -^ + d tan 9. AiH The process of reduction of the observations for Inclination was as follows : — The deviation of the compass having been calculated and the observed readings of the Inclination corrected for instrumental differences, the value of N was calculated for each observation by means of the formulae s cos f+N = (cos f+sin B) sec (' tan 6' and s cos f + N = ((1-2 sin D) sin f - sin C} cosec f tan 0' (see Section IX.). The correction due to the vertical force of the ship was calculated from the formula A = -Xf + d tan d AiH and applied to the values of N. The resulting values of absolute Inclination are given in the following table : — Table XV. — Inclination Results. From Observaiinns on board the " Discovert/ " at Sea, by Mr. Bernacchi and Lieutenant Armitage. Date. Latitude, S. Longitude, E. luclmation, S. Observer. 1904 February 20 23 27 21 23 24 27 March 1 2 4 75 34 73 13 70 4 164 42 170 2 172 25 Off Cape Washington. 3' E. of Coulnian Island. A few miles N. of Possession Islands. 70 20 67 34 67 14 67 22 171 10 169 23 164 10 155 21 87 24 86 58 86 1 87 45 86 54 86 20 86 4 84 8 85 33 86 24 I Lieutenant Aemitaob j with L.C. Circle No. 143. I Mr. Beekacchi with Fox I Circle No. 29. 150 SECTION XI. Declination Kesults. From Observations by Lieutenant Armitage on board the "Discovery" at Sea. Table XVI. Date. Latitude. Longitude. Declination . Remarks. 1901 o , o , c. , August 10 . . . 44 45 N. 9 30 W. 17 27 W. Single observation. „ 11 . 41 41 „ 11 22 „ 17 45 , 11 . 40 45 „ 12 2 „ 19 15 , 12 . 38 16 „ 13 39 „ 19 , 13 . 36 49 „ 15 „ 18 , 14 . 34 15 „ 16 „ 18 32 , 17 . 31 34 „ 17 13 „ 17 30 , 20 . 23 16 „ 18 59 „ 16 51 , 23 . 15 19 „ 20 22 „ 18 8 , 28 . 5 27 „ 21 40 „ 18 8 , 29 . 4 13 „ 20 46 „ 20 27 , 30 . 2 16 „ 20 50 „ 19 30 , 31 . 10 „ 22 22 „ 19 32 September 1 . 12 S. 24 21 „ 17 39 2 . 2 1 „ 25 47 „ 17 59 2 . 2 52 „ 26 18 „ 18 30 3 . 3 32 „ 27 45 „ 18 3 4 . 5 17 „ 28 45 „ 18 5 . 7 4 „ 30 7 „ 17 25 5 . 7 41 „ 30 36 „ 17 35 )) » 6 . 8 54 „ 31 20 „ 16 40 7 . 10 12 „ 32 34 „ 15 49 8 . 12 7 „ 33 29 „ 15 47 8 . 12 49 „ 33 43 „ 15 10 9 . 13 41 „ 34 28 „ 15 43 10 . 15 23 „ 33 28 „ 17 44 10 . 16 5 „ 32 56 „ IS 49 11 . 17 21 „ 32 16 „ 19 6 12 . 18 57 „ 31 12 „ 16 15 12 . 19 36 „ 30 13 „ 16 51 13 . 20 33 „ 29 20 „ 17 27 Mean of swings to port and starboard. 14 . 21 20 „ 28 15 „ 17 58 Single observation. 15 . 23 25 „ 27 20 „ 17 55 15 . 24 10 „ 27 2 „ 19 2 16 . 25 36 „ 25 15 „ 20 16 . 26 24 „ 24 50 „ 20 23 17 . 28 16 „ 24 12 „ 20 26 17 . 28 44 „ 23 33 „ 20 30 18 . 30 22 „ 21 55 „ 20 51 19 . 32 52 „ 20 „ 20 53 20 . 33 41 „ 17 57 „ 22 23 21 . 34 30 „ 15 45 „ 23 42 ; >, -. 1 22 . 35 2 „ 13 59 „ 23 20 „ „ 23 . 35 45 „ 11 27 E. 24 50 „ „ 26 . 37 43 „ 2 28 „ 29 14 )) )» 27 . 38 15 „ 6 „ 29 18 „ „ 28 . 37 50 „ 3 34 „ 30 23 » >> 29 . 37 37 „ 5 45 „ 30 J) )) 30 . 37 16 „ 8 56 „ 30 32 » )> October 1 . 1 36 5 „ 12 25 „ 30 26 )) „ 2 . 1 35 18 „ 14 57 „ 29 17 J) " 3 . 35 7 „ 15 36 „ 30 „ „ 16 . 36 55 „ 22 „ 2S 12 „ „ 17 . 38 47 „ 25 28 „ 28 12 „ „ 17 . 39 26 „ 26 48 „ 28 4 1) ■" 18 . 39 42 „ 29 3 „ 28 56 » " 19 . 40 3 „ 32 25 „ 30 2 II " 19 . 40 37 „ 32 39 „ 28 19 „ „ 19 . 40 53 „ 32 52 „ 28 50 „ „ 20 . 42 5 „ 33 22 „ 29 2 1. " . 22 . 44 59 „ 40 15 „ 28 45 „ „ 22 . 45 1 „ 41 48 „ 29 29 !) ■" 23 . 45 „ 43 „ 30 46 „ „ > 24 . 45 4 „ 47 27 „ 29 34 ,. ,> — -- -, 24 . 44 54 „ - -47 46 „ 29 36 - ^ - - 151 Table XVI. (continued). Date. Latitude. Longitude. Declination. Bemarkd. 1901 o , , o / October 25 . . . 45 28 S. 51 E. 28 37 W. Single obeervatiou. „ 27 . 46 18 „ 59 46 „ 31 20 ^^ „ 27 . 46 26 „ 60 11 „ 32 8 „ 28 . 47 21 „ 64 36 „ 31 24 28 . 47 22 „ 64 52 „ 31 58 „ 29 . 46 50 „ 69 33 „ 33 25 „ 31 . 47 50 „ 80 36 „ 35 55 November 1 . 46 56 „ 83 20 „ 34 54 2 . 46 55 „ 88 37 „ 34 30 2 . 46 57 „ 90 16 „ 33 40 3 . 46 36 „ 93 „ 32 15 4 . 47 51 „ 96 35 „ 31 50 5 . 48 28 „ 99 16 „ 32 38 5 . 49 4 „ 101 20 „ 31 24 6 . 49 57 „ 103 35 „ 33 45 7 . 51 5 „ 107 55 „ 30 22 7 . 51 17 „ 110 29 „ 29 36 8 . 51 43 „ 111 50 „ 27 4 8 . 51 45 „ 111 59 „ 27 16 10 . 50 51 „ 123 32 „ 12 46 10 . 50 51 „ 123 54 „ 10 35 10 . 51 1 „ 124 3 „ 13 5 11 . 51 20 „ 125 52 „ 8 5 12 . 52 16 „ 131 36 „ 3 20 13 . 54 10 „ 132 29 „ 1 32 13 . 55 14 „ 132 48 „ 3 48 14 . 57 27 „ 134 40 „ 6 4 15 . 59 1 „ 137 28 „ 26 15 . 59 50 „ 138 23 ., 4 18 . 61 25 „ 142 37 „ 6 5 18 . 60 34 „ 144 25 „ 10 8 19 . 59 45 „ 147 28 „ 12 19 . 58 43 „ 148 41 „ 14 46 20 . 57 41 „ 151 20 „ 14 37 Mean of swings to port and starboard. 21 . 56 52 „ 155 25 „ 18 34 Single obeerration. 21 . 56 8 „ 156 20 „ 17 56 23 . 54 8 „ 159 55 „ 16 27 23 . 53 21 „ 161 11 „ 18 40 24 . 52 29 „ 163 J 8 „ 19 1 24 . 51 41 „ 164 23 „ 20 1 26 . 48 30 „ 170 22 „ 19 40 December 24 . 46 17 „ 170 41 „ 18 35 25 . 48 54 „ 170 12 „ 19 42 25 . 49 50 „ 170 10 „ 20 26 . 52 29 „ 170 20 „ 20 50 27 . 53 24 „ 170 43 „ 22 42 27 . 54 8 „ 170 55 „ 25 40 28 . 54 43 „ 171 17 „ 20 34 28 . 55 20 „ 170 59 „ 19 46 29 . 57 31 „ 170 35 „ 25 6 30 . 58 42 „ 170 47 „ 26 10 31 . 60 52 „ 173 9 „ 28 1902 January 6 . . . 68 13 „ 175 25 „ 40 40 7 . 68 43 „ 175 37 „ 43 13 8 . 70 5 „ 173 42 „ 50 37 > 8 . 70 8 „ 173 36 „ 51 22 8 . 70 3 „ 173 5 „ 50 37 ^ 8 . 70 10 „ 172 49 „ 54 22 ^^ !) 10 . 71 28 „ 171 15 „ 64 40 "^ Mean of swings to port and starboard. 11 . 72 3 „ 171 59 „ 66 31 ,, Single observation. 12 . 72 41 „ 172 28 „ 74 2 ^^ j> 12 . 72 44 „ 172 17 „ 74 34 j^ li 13 . 73 33 „ 170 54 „ 87 30 ^ 13 . 73 23 „ 170 37 „ 90 46 ^ ;> 14 . 73 19 „ 169 52 „ 75 8 14 . 73 19 „ 169 46 „ 76 43 _^ 14 . 73 19 „ 170 „ 79 ^^ !I 15 . 73 12 „ 170 14 „ 82 11 ^^ » " 15 . 73 18 „ 170 1 „ 76 13 j_ 15 . 73 20 „ 169 53 „ 83 3 ^ Mean of ten single observations in different positions between Coul- man lelan d and mainland. | 152 Table XVI. (continued). Date. Latitude. Longitude. Declination. Bemarks. 1902 o , o , o , January 16 . . . 73 21 S. 169 21 E. 84 23 E. Single observation. 16 . 73 25 „ 170 51 „ 87 52 „ „ 16 . 73 37 „ 170 46 „ 87 14 J) J) 17 . 74 5 „ 169 35 „ 100 24 jj » 17 . 74 21 „ 168 31 „ 111 8 ») )> 18 . 74 32 „ 166 8 „ 118 30 Mean of swings to port and starboard. 18 . 74 48 „ 164 50 „ 123 27 Single observation. 18 . 74 56 „ 164 25 „ 123 32 >> 19 . 75 22 „ 164 54 „ 129 52 ,, 19 . 75 29 „ 165 4 „ 132 32 „ 19 . 75 54 „ 165 12 „ 137 19 „ 19 . 1 76 4 „ 165 9 „ 139 34 ,j 20 . 1 76 35 „ 163 41 „ 152 26 „ 20 . 76 39 „ 163 48 „ 150 13 21 . 77 8 „ 164 9 „ 151 38 jj 21 . 77 11 „ 1(J4 16 „ 151 50 J, 22 . 77 „ 166 47 „ 147 44 „ 22 . 77 10 „ 167 5 „ 143 12 J, 22 . 77 17 „ 167 43 „ 143 36 J, 22 . 77 24 „ 169 „ 139 41 Mean of swings to port and starboard. 24 . 78 12 „ 176 8 „ 130 37 Single observation. 25 . 78 27 „ 182 29 „ 123 41 » n 25 . 78 28 „ 183 11 „ 121 31 1 11 25 . 78 34 „ 185 21 „ 118 18 1 11 26 . 78 36 „ 186 10 „ 117 28 1 11 26 . 78 28 „ 189 2 „ 111 55 > 11 26 . 78 28 „ 188 26 „ 113 47 1 11 „ 27 . 78 28 „ 185 45 „ 116 31 1 11 „ 27 . 78 24 „ 185 50 „ 116 5 I 11 28 . 78 23 „ 190 54 „ 108 55 1 11 28 . 78 24 „ 191 36 „ 107 44 1 11 29 . 78 19 „ 196 30 „ 100 56 1 11 29 . 78 10 „ 199 23 „ 96 56 1 11 29 . 78 1 „ 200 20 „ 94 27 1 11 31 . 76 1 „ 207 33 „ 78 10 Mean of two single observations. February 2 . 76 58 „ 202 58 „ 85 33 Single observation. 2 . 77 1 „ 201 55 „ 87 38 1) >i 2 . 77 35 „ 201 33 „ 91 48 , , 2 . 77 42 „ 201 28 „ 93 6 J 3 . 78 7 „ 197 28 „ 97 11 , 3 . 78 11 „ 197 5 „ 99 32 , 5 . 78 12 „ 190 18 „ 105 40 , 7 . 77 10 „ 169 42 „ 136 28 , 8 . 77 48 „ 165 44 „ 151 31 , 8 . 77 49 „ 165 30 „ 151 50 Mean of swings to port and starboard. 8 . 77 51 „ 165 16 „ 154 35 Single observation. 1904 February 19 . . . 77 36 E, 166 15 „ 150 20 „ .. 20 . 75 31 „ 165 2 „ 133 18 „ „ 21 . 74 25 „ 165 16 „ 118 9 » »> 21 . 74 25 „ 165 16 „ 118 35 ?) »> 21 . 74 26 „ 165 30 „ 126 13 » )) 21 . 74 26 „ 165 30 „ 126 8 Mean of swings to port and starboard. 23 . 74 4 „ 169 31 „ 102 50 Single observation. 23 . 73 53 „ 169 49 „ 99 2 11 i> 24 . 72 10 „ 171 39 „ 69 47 11 11 24 . 72 6 „ 171 37 „ 72 27 11 11 25 . 71 2 „ 169 32 „ 61 56 11 11 26 . 70 20 „ 169 17 „ 52 31 11 11 26 . 70 20 „ 168 52 „ 56 14 11 11 29 . 67 46 „ 174 26 „ 41 17 11 11 29 . 67 39 „ 174 28 „ 40 15 11 11 29 . 67 24 „ 172 25 „ 44 36 11 11 March 1 . 67 31 „ 169 51 „ 40 40 11 11 2 . 67 22 „ 165 „ 40 6 1. 11 2 . 67 15 „ 163 44 „ 36 32 11 11 4 . 67 27 „ 155 41 „ 25 30 1. 11 7 . 62 5 „ 156 56 „ 23 28 8 . 60 47 „ 160 57 „ 21 54 11 11 9 . 59 56 „ 163 20 „ 22 15 .1 11 11 . 56 40 „ 163 55 „ 22 48 11 11 12 . 55 28 „ 164 33 „ 22 52 , 13 . 53 51 „ 165 22 „ 21 53 1 153 Table XVI. (continued). Date. Latitude. 1 Longitude. Declination. Remarks. 1904 . , o , o , March 14 . . 51 29 E. 164 17 E. 18 15 E. Single observation. „ 26 . . 50 33 „ 166 11 „ 19 30 ,^ Swing inside the harbour. 29 . . 50 33 „ 166 21 „ 21 5 J, Mean of swings to port and starboard. „ 30 . . 47 45 „ 168 1 „ 18 8 Single observation. Juue 10 . . 44 55 „ 177 56 „ 17 19 „ 10 . . 47 3 „ 178 9 W. 19 17 " » )i „ 11 . . 48 23 „ 175 32 „ 19 20 " )) )) „ 13 . . 51 „ 168 25 „ 20 8 " » 11 „ 14 . . 52 3 „ 165 13 „ 19 57 " " „ 16 . . 53 40 S. 156 29 „ 21 51 " " " „ 17 . . 53 52 „ 152 28 „ 22 18 " " " „ 19 . . 54 17 „ 144 4 „ 23 5 " " " „ 24 . . 58 50 „ 125 8 „ 29 58 " U )» „ 24 . . 58 56 „ 123 26 „ 31 31 " " " „ 25 . . 59 11 „ 121 4 „ 31 17 " " ., 27 . . 59 32 „ 112 30 „ 32 " )> )> „ 30 . . 57 31 „ 99 42 „ 28 42 " " July 1 . . 55 48 „ 95 52 „ 29 21 " " " „ 2 . . 54 17 „ 92 29 „ 27 53 „ 3 . . 53 42 „ 88 33 „ 26 49 „ 4 . . 52 58 „ 83 28 „ 24 53 " " „ 5 . . 52 27 „ 79 9 „ 23 8 " " " „ 10 . . 52 31 „ 65 53 „ 16 34 „ 22 . . 48 45 „ 52 3 „ 5 32 „ 23 . . 46 38 „ 48 57 „ 2 9 „ 24 . . 45 „ 45 57 „ 1 28 W. " " „ 25 . . 43 14 „ 43 32 „ 2 11 „ 25 . . 42 24 „ 42 41 „ 4 37 " " " „ 26 . . 41 2 „ 41 23 „ 5 5 „ 27 . . 39 JO „ 38 14 „ 9 8 „ 27 . . 38 43 „ 37 46 „ 9 24 " " „ 28 . . 37 22 „ 35 25 „ 11 47 " „ 30 . . 32 11 „ 30 27 „ 16 52 „ 30 . . 31 59 „ 30 21 „ 17 11 „ 28 . . 36 45 „ 34 39 „ 11 16 „ 29 . . 34 21 „ 31 50 „ 14 25 „ 30 . . 31 24 „ 29 42 „ 17 28 „ 31 . . 29 46 „ 28 28 „ 18 30 „ 31 . . 28 59 „ 27 54 „ 18 Auf list 1 . . 27 34 „ 26 58 „ 19 36 1 ■ ■ 26 47 „ 26 25 „ 19 39 2 . . 25 25 „ 25 48 „ 21 22 2 . . 24 28 „ 25 26 „ 20 30 3 . . 22 54 „ 25 32 „ 20 47 3 . . 21 58 „ 25 36 „ 19 37 J 4 . . 20 47 „ 26 16 „ 20 51 4 . . 19 54 „ 26 16 „ 19 39 5 . . 18 43 „ 26 12 „ 20 54 6 . . 16 6 „ 26 26 „ 21 1 6 . . 15 30 „ 26 26 „ 20 4 7 . . 14 34 „ 26 48 „ 21 20 , 7 . . 13 51 „ 26 55 „ 19 57 8 . . . : 12 54 „ 27 9 „ 20 55 9 . . 10 32 „ 27 23 „ 20 17 9 . . 9 24 „ 27 20 „ 19 19 10 . . . 7 36 „ 27 18 „ 20 6 11 . . . 4 54 „ 27 8 „ 20 15 12 . . . 2 37 „ 27 31 „ 19 56 12 . . . 1 22 „ 27 33 „ 18 35 1 13 . . . 40 „ 27 41 „ 19 58 jj 13 . . . ION. 27 32 „ 18 38 J 14 . . . 2 48 „ 27 56 „ 19 16 j^ 15 . . . 5 45 „ 27 30 „ 19 23 1, 1> 16 . . . 7 53 „ 27 1 „ 19 14 J 17 . . . 10 22 „ 27 4 „ 19 33 ,^ J 18 . . . 11 57 „ 26 56 „ 19 38 19 . . . 14 18 „ 27 35 „ 19 39 20 . . . 16 22 „ 27 7 „ 19 45 ], 21 . . . 18 5 „ 27 „ 20 jj I> » 21 . . . 18 52 „ 27 10 „ 19 7 J _, 22 . . . 19 45 „ 27 43 „ 19 46 » I) 22 . . . 20 21 „ 28 „ 20 42 , 23 . . . 21 33 „ 28 42 ., 20 1 . 24 . . . 23 48 „ 30 12 „ 20 13 „ » 1. 154 Table XVI. (continued). Date. Latitude. Longitude. Declination. Bemarks. 1904 o , o / o , August 25 . . 26 10 N. 31 45 VV. 20 52 W. Single obserTation. , 26 28 25 „ 32 10 „ 20 54 , 26 29 20 „ 31 51 „ 19 42 , 27 30 28 „ 30 45 „ 20 55 , 27 31 2 „ 30 19 „ 20 32 , 28 31 35 „ 30 10 „ 21 6 , 28 32 3 „ 29 49 „ 20 18 29 32 47 „ 29 18 „ 21 17 30 34 29 „ 27 59 „ 21 49 31 36 40 „ 26 27 „ 21 53 September 2 37 50 „ 25 6 „ 22 7 3 39 4 „ 24 25 „ 21 54 3 40 8 „ 23 45 „ 21 52 4 41 36 „ 22 44 „ 22 49 5 43 52 „ 20 26 „ 22 41 5 44 33 „ 18 32 „ 20 35 6 45 23 „ 16 20 „ 21 43 7 46 55 „ 13 9 „ 20 5 8 48 38 „ 9 „ 19 36 SECTION XII. Declination Results. From Observations made by Captain E. F. Scott, it.iV., ivith a Prismatic Compass. Western Sledge Journey. Table XVII. Date. Latitude, S. Longitude, E. Declination. 1903 o / „ / // o , October 14, a.m. . . . 77 41 30 164 38 45 158 9 E. „ ,, p.m. 77 38 50 163 53 20 157 13 „ 16, a.m. 77 44 50 163 17 30 160 7 >, 17, „ . 77 51 20 162 32 40 103 4 NoTember 2, „ . 77 47 10 161 3 30 104 11 15, „ . 77 43 159 2 15 172 27 W. 19, „ ■ 77 43 1.50 10 178 45 20, „ . 77 43 155 29 15 177 19 22, „ . 77 42 49 154 31 30 174 42 23, „ . 77 42 49 153 54 25 170 38 24, p.m. 77 45 7 152 23 10 171 26, „ . 77 47 49 1.50 .57 20 105 19 „ 27, „ . 77 48 32 149 43 35 103 3 29, „ . 77 54 47 147 52 30 159 44 30, „ . 77 58 39 147 10 157 50 IJecembcr 7, a.m. 77 56 9 153 25 15 170 18 „ 7, p.m. 77 57 16 154 9 30 174 37 10, „ . 77 50 53 156 50 5 178 38 E. „ 15, a.m. 77 45 160 10 15 169 1 155 Table XVIII.— Declination Results. From Observations hy Lieutenants RI. Barne and G. MuLOCK, 11. N., with a Prismatic Compass. South-South- J Test Sledge Journey. Date. Latitude, S. Longitude, E. Declination. 1903 o , „ o , „ , October 15 78 14 30 168 12 30 149 57 E. 16 78 22 40 168 22 150 3 „ 20 78 45 35 167 51 36 1.52 11 19 78 40 12 168 4 30 150 58 „ 21 78 55 10 167 38 153 44 29 79 7 20 166 26 165 7 November 2 79 26 25 165 9 167 7 79 39 25 163 56 164 49 9 79 43 10 163 41 30 165 26 13 79 49 163 13 160 33 14 79 54 10 162 47 168 50 20 80 7 161 4 25 172 30 24 79 54 35 161 49 30 169 56 25 79 47 40 161 17 170 30 30 79 19 50 164 16 162 43 December 2 79 9 25 165 32 159 54 11 78 22 25 168 7 149 54 Table XIX. — Declination Resnlts. From Observations by Captain Scott, B.N., ivith a Prismatic Compass. Southern Sledge Journey. Date. 1902 December 15 22 23 25 26 28 28 30 1903 January 3 6 7 10 11 14 20 21 25 26 Latitude, S. 80 27 81 22 36 81 32 10 81 47 23 81 51 28 82 5 38 82 11 37 82 11 37 81 56 81 32 15 81 25 24 81 2 45 80 40 15 80 25 17 79 47 30 79 33 50 79 10 25 79 Longitude, E. 161 42 30 162 13 30 162 42 10 163 5 40 163 19 38 163 41 45 163 52 163 52 163 22 30 162 41 45 162 22 7 161 43 21 161 39 161 48 32 lf;3 40 164 8 30 165 12 15 166 34 Declination. Remarks. 17°1 22 E. 171 25 171 43 170 36 171 37 170 32 170 28 p.m. observation. 169 54 a.m. „ 170 59 171 26 172 28 173 50 171 22 171 14 165 22 167 24 161 56 157 41 All Declination results were plotted, and lines of equal Declination drawn therefrom, as shown on Chart No. 1 (Plate 17). SECTION XIII. Determination of the Position of the South Magnetic Pole. In order that the determination of the position of the pole, liy means of the Declination results, might be entirely independent of the determination by means of the Inclination results, I handed the Declination results to my assistant, Commander F. Creagh-Osborne, R.N., with instructions to plot their positions and X 2 156 to extend the direction of the magnetic meridian, as indicated by the observations, towards the South Magnetic Pole. The result of his work is shown on Chart No. 2, where it will be seen that these lines of direction intersect within a space triangular in form (Plate 18). The radius of the circle inscribed in the triangle measures about 38 geographical miles, and the centre of the circle indicates the probable position of the pole, and is in latitude 72° 50' S., longitude 156° 20' E. Determination of the Position of the Magnetic Pole by Means of the Inclination Results. All the Inclination results were plotted on a chart, and lines of equal Inclination drawn (Chart No. 3), from a consideration of which the prol>able position of the pole is indicated to be in latitude 72° 52' S., longitude 156° 30' E. The agreement between this position and that determined by the Declination results is remarkable, and may be considered as corroboration of the results. The mean of the two positions, viz., latitude 72° 51' S., longitude 156° 25' E., is in all probability a close indication of the centre of the polar area (Plate 19). SECTION XIV. Comparison of Results with those given by General Sabine. Total Force. In Sabine's chart of lines of Total Force the highest value for which a line is drawn is that of 15 B.U. (= 0'6916 c.g.s.). This line passes outside all the positions at which observations were made during the 1902-1904 expedition. The observation position nearest to Sabine's line is that in latitude 79° 32' S., longitude 176° 1' E., during the south-eastern sledge journey by Mr. Bernacchl The result of this observation gives the value of Total Force = • 6896 c.g.s. The change of intensity thus indicated is very small. , ' Declination. To compare the Declination as given on Sabine's chart with present results, the values were obtained from the respective charts at the positions given in the following table : — Table XX. Position. Declination. Increase in 62 years. 1841. 1903. Ice Station, McMurdo Sound 124 ' 40 20 102 87 30 69 65 E. 147 45 62 30 122 30 95 79 109 E. 23 45 22 10 20 30 7 30 10 44 Latitude 78°, longitude 180° 78° „ 200° 76° „ 205° 74° „ 165° The change indicated at McMurdo Sound, viz., 23° 45' in 62 years, gives a mean aimual increase of 23'. The yearly difference as derived from the absolute oliservation diagrams (Section IV. for 1902-1903) shows a decrease of 26'. Knliomd Aulairlir E.qmlitwn, 1901-1904. Plat,' 1.-) {MiN/ndir (Jhsmvtions). Hut Point iiiiil Winter Quaiters. Tlie cross x shows position of magnetic houses Natimal Anlarclir Bipedilum, 1901-1004. nHf 10 (Mar,ne(ic ObfrnxUions). iXatioiial Antarrlic E.qmlitian, 1001-1904. Plate 17 {Magnetic Ohservalwns). West Lons: I40° I50° I60° I70° I8 0° I70' 160°. I5Q° I 40' C'hait I. Lines of equal magnetic declination. (See p. 155.) From observations made by the Officers of the National Antarctic Expedition, 1902-1904. By Commander L. W. P. Chetwynd, Royal Navy. National Antarctic Expedition, 1901-1904. Vlalc 18 {Magnetic Obaermtions). lao 170 160 I20°Easb Long : Cliart II. (See p. 156.) Nalioiml Aniarclic E.rpnlilion, 1901-1901. rial,- 19 {Ma{/ncli,- Obsaruiions). WeatLongiMO" 1 50" 160" 170° I60" 170" 160" 150" MO" 130° East Lonp;; 120° 7?^ ^°i>^ ,°6-? Chart III. Lines of equal magnetic iuoiination. (See p. 156.) From observations made by Mr. L. C. Bernacchi and Lieutenant A. B. Armitaoe, E.N.R., during the National Antarctic Ex2)e(litiou, 1902-1904. By Conmiander L. W. P. Chetwynd, Royal Navy. The Jiff ures plotted on this chart represent values of Co-inclination, 157 Comparison of InrMnation Results. The lines of equal Inclination on Sabine's chart were completed, where necessary, to conform as nearly as possible with those already drawn, and the values of Inclination obtained by inspection of the respective charts are as given in the following table : — Table XXI. Latitude, S. Longitude 170°, E. Longitude 175°, E. Longitude 180°, E. 1841. 1903. DJfEerence. -0 10 -0 55 -I 03 1841. 1903. Difference. 1841. 1903. Difference. 7°3 75 77 86 55 87 40 87 25 86 45 86 45 86 22 S6 15 87 5 87 5 85 50 85 55 85 43 -0 25 -1 10 -1 22 85 45 86 35 86 40 85 5 85 10 85 10 -0 40 -1 25 -1 30 -0 43 -0 59 -1 12 Mean for the whole area, -0' 58' in 62 years. Mean annual change, —1' nearly. The yearly diflference as derived from the absolute Inclination diagrams, 1902-1903 (Section II.) is - 6' -7. The position of the pole as now determined differs from Sabine's position by approximately 200 geographical miles, being 52' of latitude to the northward and 9°-l of longitude to the eastward of Sabine's, indicating a presumptive change of po.sition to the eastward. L. W. P. Chetwynd. 159 III. HOURLY VALUES OF DECLINATION, HORIZONTAL FORCE AND VERTICAL FORCE On Term Days during 1902-1903, in connection with the National Antarctic Exijedition, 1902-1904, comprising Results at tlie following Observatories : — Greenwich, Kew, Falmouth, Pola, Bombay and Mauritius, and also at the "Discovery's" Winter Quarters. -The Eoyal Society arranged for simultaneous magnetic observations to be made at certain ol)servatories on pre-arranged term days during the National Antarctic Expedition, 1902-1904, simultaneous observations being also made at the " Discovery's " Winter Quarters. The term days arranged for were the 1st and 15th of each month, each term day commencing at Oh., G.M.T., and ending at micbiight. The resulting data received, viz., the hourly values of the Declination, Horizontal Force, and, in some cases, of the Vertical Force, as measured from the magnetograms of the day, have been tabulated and compiled by Commander L. W. P. Chetwynd, R.N. The values derived from the "Discovery's" magnetograms have been measured and tabulated by the Stafl' of the Observatory Department of the National Physical Laboratory, under superintendence of Dr. G. Chkee, F.R.S. The Observatories from which results have been received are as follows : — * Greenwich Declination. Horizontal Force. Vertical Force. Kew ,, „ Falmouth ,, „ Pola „ „ Bombay „ „ „ Mauritius „ „ " Discovery's " Winter Quarters ... „ „ The maximum and minimum hourly values on each day are printed in heavy type ; where such maximum or minimum was recorded at more than one hour on each day, all such maximum and mininiuni \alues are so printed. In the tabulated values for Mauritius Observatory the data in brackets have been obtained by interpolation, and where one hourly value is lacking the mean, and where necessary the range, is marked with the letter "a" (denoting approximate). In the tabulated values for the " Discovery's " Winter Quarters, on several days one houi'ly value was lost, the hour coming when the sheets were changed ; on other occasions a good many hours' data are lacking, owing to no sheet being on the drum, or through the trace being too faint. When only one hour was lacking a mean has Ijeen calculated, to which the letter " a " is attached (to signify approximate). When several hours were lacking no mean is given. When the trace was off the sheet the value answering to the edge of the sheet is given with the sign > or < before it, according as the trace was off' in the one direction or in the other. When the trace was off the sheet for only one hour a mean is formed with > or < in front. AVhcn the trace was off for several hours no mean or range is given, except in cases where the general drift of the curve justified confidence that the maximum and minimum were both included. * As these pages were going to press, the magnetic returns from Christchurch, Now Zealand, were received. Tliese have been tabulated by Dr. Chbee and will be found on pp. 177, 178, 179. 160 |s-&i o CO . O. 05 o OS 0, ,o ., O « OS ^ w^^S " i^c „ -* o to to 00 O <-< rt o to ^ O to Cl ,- OS „ ,- ->• .r. 00 to 00 •X f Og « '-■' '■'• '.M ..•4 '•' 14 -' ■' -' '■'' '■'■ '■' IJ ■' " ■' ^J CO « _ 35 05 05 N o o 05 -x> 00 o 05 CO « to 00 o CO ,- 05 00 OS O o rH S ^ ' ' s S s S3 s ss S3 S g ^ s g g g s g S g 53 g g S g 53 ^ s ^ .T, m 05 rt N C 5<, CJ> o 1^ o CI „ 05 o rt to o 00 7= CO OS us a> rH n ~ 1 ' s s; ?> S3 s 53 S3 g S3 g s g s ?i s s 53 s 53 s g S 2 S g 53 ^ <31 OC 05 ^ n ^ 1i 05 o to OC 00 « « ^ to 05 CO ,4 o rH o us us 05 2 ~ ' ' s S SI s s S3 S3 g g g s S g g s 53 g g s s g 2 o s g g OV O! OS to « to t- ^ 00 05 o> to CO CO 05 •a CO 13 CO t^ OS 01 to t- » £ 0-- c^ ^M IM 1.-. CM .-. IM ^M IJ 14 " 14 14 CO (M » to m 05 o O cq to ,- 00 t- cc o ,- M as o CO CO oo ■r rH OS U5 US -a ~ 1 ' S SI ^ s S3 S S g s 53 s S 13 g s S3 53 g g g s S s s g s oo IN (T. 05 ■* ^ n OS o 9 00 U3 CO _ e> XI ■■ 1 ' s S s; S3 ^ s 3 S s g g ^ g s 53 8 g 53 g g o 2 S g g t^ •£ ra ■o 00 a> 00 « U5 t- 00 00 O CO rt 00 OO OS to o »fi cc 00 OO to to CO ^ "■ ' ' s S5 S! s; s S3 S g g g g 3 s g g s g g g g 53 2 s g g g a. ^ ^ .n to to rt t- o o> o OS 00 •a Oi 00 rt 00 o to r- ° 40 o U5 o s: - 1 1 S s S S s S -* 05 ^ S! S3 g s g g g SI SI S3 g 53 g g S g .o IN o o ,. o <0 00 CO 05 00 00 CO OC 05 .o CO * S g g g g ,, „ CO w „ o t- CO 00 rt CI o rt OS OS 05 00 Ift 00 •a OS 9 rH UO OS CO — ^1 1 I- I, CC S g ■^ B * t- ^ ■n fn g CI g g to CO o CO CO ^ 53 g PH !M Cfl w w ^ m o .o o OJ o rt to rt o 05 o Ol « CO to rt o O CO lO CO 00 o to CO ^ ■4* u? ■<• S M n f~* ° g s; g " g g « g CO g 00 g g 2 53 •-1 W CM Ol w OV rH .o O X 05 ^ - 1 1 S ?; S S3 s 2 g s s t- 2 2 s 2 g 2 s s § s g 2 2 s 2 2 „ .n m m ^ o t- OJ 00 « IX 05 c^ O to OS CO Tf o oo o « OS rH to rH CD c^s t^ to o (B CO 05 I- « CO CO to xa o « « rH to c^ to XI ~ 1 ' S 3 3 s S s s s S s s s S s 53 2 s g 53 53 53 2 s 53 53 8 m m .o to ^ U5 c^ to rt 00 (- t- U5 o 00 to o 00 ^ C5 >I3 CO US rH OS .o N (N (N !N (N OM '^ " '^ ■•* L4 ' ' ' ' '■' u "■' "■' ' ' 14 '' 0-. -, m Ol ^ 00 rt 05 CO „ t- 00 rt ^ « OS cq CO CO OC ? >» U3 tH CO U5 ja ~ 1 1 S S ■^ s s s g § g a s 8 g g g 2 s s s g 53 2 s 53 53 53 CT. o O! o N 05 05 05 o . CO ^ cq O rt rt la OS 00 to 00 05 CO U3

? 162 P K i i 163 SlC^I „ ^ c . Zi -^ TO ov .-o w^g i?c 31 ,, 00 -» o TJ t. 3, o „ „ » to -J- 00 M CO „ •£' 00 CO CM ^ ,_ „ iC ^ 3 Qg •J' ^ ■J- V ■' * ■* ■'' •J' •J' ■V 1< u: o a> ro -J" a 00 o » CM »■ 3 -!■ ■* 3 3 •V ^ ^ . n n TO ® _ » „ a> 00 to r- 03 CM o> 00 ^ a> „ „ CM "* ?• ■£> TO TO Tjt W rt s ^ ■fl" ■* ■<*■ '^ •d- TH ■* 'W ■* ■* ■>j< ■* ■V ^ -* ■W -)- "^ -0. -1- * ■* ■* N rt Ol o •n 00 « o 00 TO ■re 01 ^ CO ^ rt = TO CM TO ■^ TO TO TO U5 ca o ~ 5 ?? S •V ^ 3 "V ■q- ■* 3 ■^ 3! 3 "* 3 3! 3 5 3 3 3 ^ 3 s 5 ^ ^5 31 0-. ^ m 0-. lO CM ^ O rt TO 00 rt d = rt a. TO « TO <« « 00 ^ 3 1: 3 ^ ->• ■» ^ "* " " ■>*< " -K ^ ** ■» tj- » " " ■* -t ^ -« (, rt ^ o => o O » rt •0- ^ „ •^ at -. = rt 01 TO ■^ ■^ C4 ^ CO ■* rt 3 3 w fl" 1- ^ ■w ^ ■* ■* ■* 'rf V ■w ■* ■fl" wt ^ y -)' TK 13 oo CO 00 rt .o ■£> 00 ^ Cft o> w CM ^ ^ rt to TO CM ^ U3 ■^ CM ^ O) ^ ■9- '^ -s- ■w ■* ■^ ■* ■* ■* TJ» -s- ■w •V -a- V ■w V y< ■W V y -0. -^ _ -X ^ CM o « ■* 00 !0 ® oa rt ■V la 01 CM CM „ TO ■* CM •^ TO « 01 a -S- ■* ■* ■* ■* T)< Til ■* ■* ■* TJi ■^ Tj< ■* ■* ■* ■* ■* ■* •J« ■J" Mt y ■g- _ „ cc 00 >- rt o o X CM CO .0 •* M 00 ^ M TO TO (N N rt => ^ a to » ■* •D TO ■^ CO (N uj 1" •* Tf '* ■* V ■* y- •J' ** w M' •J> y <» a> n (N „ o w O 00 ^ ^ «> ^ r^ « J' gi -JI ^ o m Cl » „ a, Ol o ■«f US cr. t. ^ 01 « ^ to 00 to .0 ,^ 4 t^ TO ^ M M CO $ 3 Cv» -J- ■9" ■* -v •J- ■9- ^ ^ ■^ I- ^ -J- ■J- -J" -JI OS f. TO TO a a-. o a> O o> t^ n to ■* A ^ « « TO TO « ■X m t^ « 01 TT II ~s S 5 s 5 S $ s 9 ^ >o $ S; s to 3 -* 5 3 '-S ^ S; 3 :S ** ^ 3 c 1- 0-. UJ IM c« o « ej c^ TO Oi t- ^ .0 » t- ^ ,0 TO « C4 9 3 ■^ s^ -*■ ■V ^ "* «s ■^ ^ MS ■* "* ■* ^ "^ "* ■^ ■* „ .n TO UJ TO Ti. n o >o rt •i) O) ■* a, ^ Oi „ t^ ^ TO •-0 -i< „ TO ^ CO ,- 5 '^ -f -a- ■* ^ -fl' ■* 'S" ■* ■^ ** •* ^ I- •I- -J' -J" -J- -JI -JI ^Ji "V „ O CM CM Tf .n o 0! CM W 0> (_ eo to TO 00 ^ e^ CM TO TO Tf. x> OS 3 " ■* Tji ^ ^ -a- 1I< "J. ■* -* -w ■ji y" •W a> to e. TO e. o o o> ^ O A «> to ^ » to TO o> TO «*< to TO t- ■«»• u:. M TO s 5 $ S 3 3 s ^ 3 9 5 ■>(« 3 3 ^ 3 3 3 ^3 3 ^ 3 3 3 3 ■* S 3 o> 00 ej TO e. TO TO OS CO U5 ■* « TO w rt ^ ^ et t^ TO 10 ^ TO TO TO o» t- TO § - 5 5 9 S 3 s ?! SI 9 3 3 3 ■* 3 3 3 3 3 3 3 "J" 3 3 3 3 * 3 3 t^ 00 „ ^ _> TO N ^ « m 01 TO e. ^ •H s ^ 3 3 ■>* 3 ■w 3 3 3 3 3 3 3 ^ 5 3 3 3 ■* ■w ^ •a* ■* o> !0 t- rt » O O a. 00 m CM ■* CM UJ rt t- C» to rt TO Oi .- ■f CM 00 « y S ~ '^ 5 s •x> 3 ■* 3 3 5 '* S ■W 3 3 3 3 ■111 3 5! 3 ^ ^ 3 ^ 3 5 ^ 3 -, ^ TO ^ « ^ to ■» 00 CM .0 CM ■* 'P 00 rf fX rt M" TO .0 OS t- «o b- TO ^ - S S s s 3 to ■^ **" ->* ■^ 3 3 3 3 3 3 3 3 5! ■^ 3 «*« 3 3 3 -* ■* M« 4 9 •i o « i z s ^ 3 3 3 TO -* ■* Z 3 3 s "? t, 3 3 3 3 3 3 3 00 oo 01 ^ CM -X CM o » ■x> 00 ^ TO t- rt TO b- 3 ^ 3 3 5 •* 3 3 5 3 ■V ■* ■* 3 ^ 3 3 S 3 3 3 ^ t^ m cr. ;, CM o c» ^ o o U3 „ ^ ^ „ TO Ol ^ OS to TO t- TO " " " " ■* ^ " " -» " " " ^ ■>* tf t> -J- "fl" "' " '' ■* " "* *: » TO .c CM S° ^^ 5 » 3 3 3 -«< 3 3 3 5 3 ■* 3 m 3 3 3 s ^ ;s 3 3 3 ?? 3 3 3 1 TO Q - s " S " S " 2 " S ^ S ^ S ^ 2 " 2 '^ !li " 2 " 2 2 " 2 h9 J3 : < c "-3 >> 1-3 3 = g. = 1 1 & i 1 : Y 2 164 HI |1| t>. S S S 2 3 3 - ^ s S s s ?3 s s s 3 ?! 8 s 2 2 s 2 S S s a d II a S 1 1 i 1 ^ ^ s tC S S S % ^ 1 g s 1 S s ;s 1 % i 1 1 £ 1 s S i § 5 S 5 g 5 s g -* § i i i s s 1 s % S « S g Tjl s i 1 1 1 J3 ■1 S -5 S Jo 1 £ i g -J- t- i § to 3 % t~ § Til g t- ^ Tf •* i g i § t^ 1 i 1 1 5 1 i 1 g » 1 •J 1 s % b- ■* s ■* £ t- ^ 1 Tj< s s g § s i S i 1 g S ^ % 3 § s s 1 g "* % g ■"S- 1 £ 1 1 o ■W o s t- i g £ 1 s s s s "* -* ? s 1 S ■a- 1 i 1 1 i Tf "* to £ i g £ g § i g £ 1 g 1 5 1 K 1 5 i » s i 1 g s 1 i % 5 i i 1 i 1 1 I s 1 J3 S K 5 g £ ^ ■Tj" rj. rj. S 5 § s s 1 1 1 s S s T? ^ s t- i Tjl 1 £ g g 1 s' 1 sill S S2 g 1 e i i s 1 "a- g tH Tf g ■^ i Tj< g Tj« g g § £ ■* S 1 i 1 1 ■-J" £ i § i i s 1 ■>!< 1:- Tj« Tj< ■* s g 1 s i s e 4 1 1 1 1 1 ? 1 •»!" 1 S 1 i 1 1 ■^ -# 1 t- Tjl tH ■* g 1 i 1 £ 1 J3 p ills 1 1 "* ■^ i s 1 •j< 1 i S 1 ^ s 5 ■W 1 1 1 S 1 O Tf ja i s s s s tK -V -* ■* o o 5 S ■q" 1 s i s 1 K 1 i S s 1 ® 3 % S 1 % S 5 § as s s s 1 1 5 S S % 5 1 o S OS 1 b- £ s S % 1 1 1 o % 1 i 1 jr" 1 s i 3 s 1 i 1 1 g s s i § % o i 5 i Tj" t' s ? i i i 1 i S i i S 1 i o S 1 b- § i 1 S 1 1 i i i i i S i 1 s ^' ills S 1 g t^ i f 1 i 1 % i 1 i s 1 t^ i r- s TP s i o S S £ s '-3> Tj. £ 1 5 s 5 s 1 K 1 -1 i 1 5 ■w s % 1 £ a. s i ■a % 5 § § § S S e o ■* 1 § s s t- g 1 1 s o 5 i S S g 1 § s i s E s :« a S S S S 1 ?: £ ■* t^ § "* t- 1 S 5 o S i i i 1 S i s i s C- S 1 5 "* •"!• 1 "^ 1 s S § s 1 i OS 3 Ol 1 ■s. Tjl g g g ^ s 5 § 1 1 e- s 1 s 5 g s ■^ i 5^ t- ^ £ s i 1 1 1 i -^ o S5 Tjl g i s s 1 1 1 § s 1 5 s s i ^ 1 ■* 5 % i 3 1 -* 5 1 c. g TJ" g 1 -J- i ro g 3 S £ TC 'Ji ^ Tf 1 s 3 5 S S 1 s i S ■^ s 5' £ i 3 s s i •D ■* t- g 5 2 OS O N t- ^ 5 ? S S ? 1 1 § g Til s n 1 1 ^ CB S i s i S g £ 1 P' 3 O O lO « s; !; § s O 00 O % g ■^ S i s s s 1 1 § i 3 Ol 1 g r- g £ 1 Date. >> 1^ = 1=1 < s & g £ •-3 ■i >> •-3 s «! s •S s s 1 2 1 s 2 3 2 165 o _ >.Sl N 05 Oi 1^ o -* •fi is? * ^ •-C ^ CQ ■« ^ TO .J m I^ ^ 00 'JO Oi ,_ tTi tn TO ^ -4* 05 K-i i'c" » „ ra Oi to « O M rt O „ t- CO •* TO TO 00 » t.- a> ^ « TO TO „ « „ n OE ?! ?3 SS §3 s; s Si Si S! S S3 a 2 S3 S3 53 2 S3 a 2 S a 53 a a S s a t* t- 00 00 ^ 00 ^ « o. ^ O to - - « TO ^ 00 r.. o « w •* w Ol .- Ol - ;^ S ■^ ^ Si S! 'clj S3 S s s W S3 2 S3 S3 53 2 ci S3 s a s 2 a 2 a . 00 90 to C^ CO 00 d c< „ n t- _ Ol o .o C o to .15 ? ? w TO Ol o TO 01 o» ^ ^ 1.1 U'l ^"i -t (N u1 ^ 1.^ '^ . U5 » ^ a, •* c» ^ m « .n 00 o Ol « ■« n 1^ 3J M o> to to in X, 00 Ol rt o> 8 a ^ '■'* 1.1 LH ^•4 '-* '^ ^ ^ •^ '^ ^ m f_ ^ m -* »- 00 ►. N 00 to ^ o « C o ^ 00 TO Ol ».- CM o> TO 00 01 -r o> cr. ■ S s s a s; §5 S a S3 S a a 2 S3 §3 53 2 S3 a tr. to a n s 2 8 to OS t- oo CO O 00 00 TO 05 Ol 05 o 05 a -■^ ^ ■"• '"* ■ ' ^ t. 00 •a iti „ Ol r^ O 00 00 oo ^ to 35 TO OO 05 m Ol « € ~ 8 £^ §3 s S S3 S S3 S3 S a a 2 S S3 Oa 2 S3 a a s S3 8 2 cr. 53 s 8 00 ^ 00 a> m cc o ai 00 « 00 o CO to TO H* O o « CO o O TO 01 «s CO 01 » S ~ S ^ S3 S3 SI S3 i? S3 S3 s a OS <* Si a a a 53 2 2 S3 a a a S3 2 S 00 * "* « •r ro <3S « M to 00 „ M p TO rt ^ rt ^ CO OS lO •* w 00 ""*« o 00 ■s ■" a s ^ s; s; s a S3 S3 53 S3 a s s S3 53 SI S3 a s 53 S a a a o S3 --0 00 ■* rt -* ■* « « t- TO 3J o a> TO TO M. 3! t- o TO in to o o ^ „ o> 05 " ~ s s 'S" s s s s s S S3 S n a s a ■^ S3 a SI 13 2 53 a S3 53 a 8 n ^ 00 N oo 35 00 33 a> „ „ o « o ^ e. * 3) 35 c^ o .n TO IN TO 41 •a n 01 + So c. ~ s s s S3 •-D s s ^ ■* S3 s; ^ r-4 lO n a a a s S 2 S3 SI S3 a s; 53 la a> o -c CO ^ n n t. m rt ^ « c^ a rt 00 t- (M rt O '<«< to T ■»• to 05 Ol T ~a s s s s oe 53 S3 S S3 s s a s a s s ss s ss s 53 S S3 s a 8 a II c -» ■J) o * CO < ■* to w » -* Q ll ~ s s s S3 s s S S3 S3 8 kO s 2 S ss a s 83 a a a Si S S s a 2 53 .j5 ^ ,, » CO TO p r „ N « a, r, ..- TO o » « TO cr. TO -x> -f TO C<5 .- Ol rt S - s s; 5! S S3 -*■ S! « ■* s ^ S3 s SI 2 S3 a a 2 a a « 8 a 2 8 -il „ „ „ ^ m cr. rH U5 ^ ^ „ „ „ „ ^ 00 -» c^ ^ 35 in t- 35 Ol 00 to t- 3 ~ ?s s s S S S! S 8 S a 53 2 S 53 a a S SI 53 2 2 a s S a s 3 2 OS -^ lO r* rt r^ at rt ■■£> rt 00 o ^ o ■* •X. c» 00 o « IN o 00 to rt ,- 01 TO ~ s s s S n s s s ^ s S S >o Oi a 2 2 S3 a CD w 2 a 2 2 8 ^ 2 U5 ^ O oo m o ^ _ O) .o OO _ a-. ^ e. O o> ■o to t' O la 00 t. -* t- ■;< rH « "" s s ^ s S3 s S! ?. s g £ s ifj Ol s 2 s S 2 2 S 2 8 2 S 8 2 J, m a> » ■* tH O ^ » CD ^ rf 00 ■* ^ rt U3 o t- .o to a. oo i^ t' o Ol ~8 s Si s S S ^ s 8 OO s S 2 tr. s 2 S 0-- S S s a 2 2 a to 2 O Oi t^ T» tji TO ■^ CO oo ^ ^ 00 00 •^ 5, ^ o 00 Ol to . ».. ^ Cl I- b- 00 ■7 in s ~ s s s s S3 Si §3 S s a a 2 in 00 a 2 to 53 2 2 S 2 S3 2 2 a to 2 a -o m <»1 ^ ^ rt « <- „ rt IN rt rt -K TO IN IS o r.1 11 « Ol t.. 01 00 to ^ g - ?J !?) 3 s s; s S E5 a a s s s a 53 2 s 53 a; <31 00 a S3 a. a a s 8 ^ ns 00 m ^ o b- t- ^ IN ^ TO « 00 N c c m 00 IN to 3> ■* Ol ■* o; ^ t- ^ 53 §3 s S3 s; S3 Si S S3 53 E3 2 2 a S3 a ^ S 2 05 tr- 2 S a a a 2 a t- ^ m -^ V ^ 00 Ol to ■H CO TO rt U3 t~ "5 _ ,, 0> ^- m t- ^ tt> TO Ol TO TO s ~ g S Si S s S3 S3 S S c3 a a 2 53 Si a s c3 2 2 cr. 53 2 a 53 8 o 'O ~in~ ■* O M O t- ~»r — ~~ « « "!» t- ^ 05 c ^ TO .n ■* » -, r- Ol O - s S s s SI S3 s; S3 s 53 2 a 2 S Si a 5 s 2 a 3 2 S m a 8 2 S^ U3 --0 ^ o » 00 ^ in to o> w t- ^ „ s> 0-. ® o w . O o "oT „ o TO OS TO ■s> tr. 35 ^ ^ lO to rH m 35 o TO « s° ~S S! SI Si s; Si S3 S3 Si Si s 53 2 Si a a s S3 ^ a » t;j 8 2 2 a Ji. 2 ; s ■ C >■, S j;;- . " s . S s "" 1 " J. ¥ " 1 " , " c •-3 1^ a < a - ►^ = a. O >! •-5 & 166 S s 2 2 Ji s S " S s O s s S3 S s S s a s? s S - 2 ^ 3 S C- m d d II W Qg S 5 g g e K S S o t- S g g S g g g g ■* B s s s g g 1 S s s g g 5 g 5 g H t- 5 g t- g g s s t- g g g ^ ■* f S S S g ^' s £ i g g S S g t- g g K t- S g g g s g g g g g g re g g g 1 s i g g g t- s s t- i e 1 g s g e « t- s o g g 3 t- i g g .a ■a: g i S g 5 S g g s s g g s i g ? g g S g g g ■* s i g s 5- S g g g g Oi g g H g g S g s g g g g g g g g •* •* g g g g jd g s g P g g 1 g g 1 s g g i g 3 1 g g g g g s g 1 g s t- g 4 a s S g s S S g S t- s t- g s s S t- s t- R t^ s g 1 s t- g ^ s S g s 5 t^ g § t^ g g i g » g u g g g g S 1 g s g g g 4 P 2 g g g K g g g i s g g g g g g 1 S g g ■w g g g g t- S 3 s B g H g n S g g S g g S g g g s g g 1 s -£■ g g g g S g S s g i »^ g t^ g t- s g g «>• g ^- t- o i g g g B O t- o g g H g 5 1 2 S 1 S s g g S g e g g g S S s g CC' g S s ,? t^ i g g 1 g 1 g g s § B B g g o s g g g g "* g g S s g "* 1 g g g 5 g 1 g g g t^ g t- g g g g § g s t- t- S. l>- g s o g i 1 s g s 3 1 g g g g s B g g n § g g g g s g s g g ■^ -a" g 00 i2 g g re g J3 t^ g B g g g H g g g g g g g g g g p cc g g g g H 1 g g -X 2 g 3 g g g s n i t- g s g rP g g s g 1 g g g g g g g g g S g S g g Z s •x> s r- g g !>. g g s g g g g s ■1 g g ■* -J" 3 g OD p g g g § g g s I s t- s g l>- g s g g o g g S g g 5 g S g g •■c ^ s g S g g t- g g t- o t- s l~ t- g t- g g g g g g g s g '£■ •£■ J3 g 1 1 i l~ g g g g g s g ? g g g o 2 g g g g g g g g g ja s g 1 S g 3 § s g 5 s o g 3 g g s g g r- r- g g g s 2 g 1 J3 1^ S s g 1 g E g g Ti« s c3 s e g g s S g t^ t- g t- r- to g g g J3 t- g t- g g o S § 1 5 s g g g g g to s 1 g g s -t g 1 g ^ £ 2j= s s g B g H u g » t~ g ■v 2 g q; g § s t- s 2 g g g 1 i- g S n 1 s a •-9 2 2 j2 2 p. < 2 a 2 •-9 2 _>. 2 3 3 < 2 1 1 2 1 1 2 1 2 -a q 2 i 2 2 167 " a '^S o s-^il "■ f - ^ c^ TO l- - "- •" ^ * a, >- •^ » a '" 00 '- w CC ^ us TO 01 C^ ■•° - *>a ira .n .n •I- ^1 ^ -f -p ^ ■n in .p PI PI ^ ,, rn m -. ^ ^ PI •p -K ^ ^ &I ^S :o s s s « s 2 2 2 2 2 3 3 3 3 2 2 2 2 2 2 2 2 3 S V ^ A N o « O) TO _ 00 ® cq ■ij t- o oo o « 00 ■* „ ■»n ^ ,H in 1. TO o TO « to to rt OS TO us - ~ " o> V; s s 2 S 2 2 2 2 2 •« 2 3 3 2 2 2 2 2 2 2 2 2 3 3 3 -, <«• - rt TO cq b- a> « 00 n TO a 13 a> n Oi t- 00 « TO n TO OS o to « TO o ~ s CR s •X) s 2 ^ 2 S 2 2 2 3 •^ 2 ■<* 2 2 2 2 2 a 2 2 2 3 3 3 Ol o = o d rt t~ oo 00 ■* rt t- TO r^ et o» ,-1 ^ ^ t^ 00 O ^ a> TO ^ TO ~ E: s s •s s S 3 2 2 2 ■* 2 •^ 2 3 3 2 ^ 2 2 2 2 2 2 f-t 3 3 3 rt o> c rt TO rt 4 t^ ■o « TO rt N o ^ TO t^ „ o to to ' ^ to t- s s 2 « 2 2 2 2 2 3 2 3 2 ■* ■* 2 2 2 2 2 2 2 2 3 3 ^ » t o eg _ _ '■D 00 t- 4 2 2 2 n* 2 2 2 2 2 ^ ■^ 2 2 2 2 2 2 2 2 2 3 rt 00 U5 C^ a> a. 00 ,- CR >a •n ■« -« TO 04 o o -X TO oi us OI M -H TO TO OI « s ~ s s t- » 2 S 2 2 2 2 'ii 2 2 2 2 2 2 3 3 3 2 2 >-i 2 2 2 2 CD o< « ra TO !R TO cq -JS ^ ^ lO ■* :o ■-0 (- o 00 f U3 o OS OS _ TO 1- ^ OI Ol - t- s 2 3 S 2 !; 2 S 2 2 2 2 2 2 Tl. -* 3 3 3 2 2 2 2 2 2 2 3 ■* 00 rt m ^ «< o> o „ .ra ■S) -^ 00 ■n CM TO _ -, -t. o 00 1- 00 TO us to us -X, u ^ t- s S S a S S '£! - 2 ■^ rD 2 2 2 2 2 2 3 3 2 2 2 2 2 2 2 o .o 00 « M o rt Ol O TO :o w „ .o N us oo t- .o ■* 00 rt a o us « TO us us + II ^ Ci s S s 2 g s 2 2 - 2 s 00 2 s 2 S " I:; t^ 2 2 2 3 3 3 us US a M o n ^ » CI » m o * 01 OS a> .(3 « t^ t^ m us TO t- « 00 TO OS us OI OS n & ~g s S s 2 S « S 2 s 2 s s 2 CO S 00 2 00 S S S 2 s 2 S 3 3 ^ ^ oi iH ^ c 4 ■* 00 00 us t- t* -a" CO t~ * TO t- ■H 5 ^ cr- 00 00 ^ oo 00 '-O Ol 00 CO t~ 01 t^ 09 00 w lO ^ CO •^ 'K <«< Ol TO ^ ^ ^ ^ W4 ' ' ' ' ' ' ' ' ' •^ ■^ '^ ■^ '"' f-4 •-* ■^ ""^ '"' ■^ ^ N ro f. o 00 -. ^ to 00 00 ^ o 00 us OS a ^ ei rt -. ^ CO s M3 S s S 2 S 3 2 ^ 2 3 a " O 2 2 2 o 2 2 2 3 2 ■H 0> o o Ol M w oo 00 5it ,- eo « IM M f -. 00 9- TO o o CD rt Tic us to « US ^ 01 £ ' s 2 » 5 s TO »-i 3 s 3 TO a ^ " © s •H 2 1-1 OS 2 2 2 2 2 3 o o ^ m ^ 3i TO -J, ^ 00 » TK in M et n rt t- to « TO cq to oo TO to « rt Tit ,1 £ ~ := ^ S 2 S 2 2 2 5 3 2 s •H o ^J o ^ ^ ^ ^ 2 2 2 2 2 2 n 3 ^ „, „, ^ m m ^ •n -f .O CD eg ^ n ^ a. 04 U3 TO 00 00 TO tT ,- TO „ OS ■* g ~ E; s s S ^ 2 2 2 3 3 s s " o 2 := 2 2 2 2 2 2 2 2 2 2 3 3 O) w -o ,„ to oo ^ ,_ -* TO U3 TO ^ » to -. •« o t- to to Ol n. PI « b- £ - s fr- s a 2 3 3 ■* 3 3 2 2 2 3 2 2 2 2 2 2 2 2 2 2 2 2 3 m m ■<)« cq TO oo •> c ■^ -* ^ M o Oi o to to OS TO OI 00 us j3 ' S ^ t- '-0 s •x> 2 2 3 2 -* 3 ■* 2 2 2 2 2 2 2 2 2 2 2 2 2 ^ 3 n m .n ^ (XI ^ ic ■a lo 00 TO TO TO 00 t- • ■-0 U3 tK •a Ti. o oo lO to TO •q" OI 00 != ^ ?> « 9 OS ^ ~ s 2 t- 2 s to 2 2 2 2 2 2 ■* 3 3 2 2 2 3 2 S 3 2 3 2 2 _^ " 4i ^ ^ W M -f ^ ^ -^ t. ^ ^ Tf Tf ^ o lO 00 M O) « »- rH to " Tf- -"t M 2 -a ~ s 2 ^ •-0 s 2 2 2 2 2 2 2 3 3 :^ -J' 2 2 - 2 t^ ^ 2 OI 2 _5_ 3 " • • Q ^ 2 >1 s " 2 " 2 - " " " ^ " " " "^ " h " 3 c 1 £ 1 < g ^ < a in 1 o : S Q 1 f^ 168 Range from hourly readings. Unit ly. s; s S 3 S S! 2 ^ JS ?! s 3 a s s s S § S; S5 s 3 s o s s ^ S 6 d II Is S i S5 ro S S (M C^ C) N t § i 1 o i S i i i g i § S S i 8 i 3 f 8 -t 4 i a 8 i S i i s i i 1 i i § S o s; i 1 i O s; 1 i s 8 i i i i A i i 1 s s 1 i i S i o i g i S S a 8 i § 1 s i 8 i i i I i i i i S S i i g i 1 -D 1 i 1 i 1 i 1 1 i i 8 i i 1 i ^ i 3! s a i i S i i § i i i § 1 1 1 i § 1 i i i i s S i i i i 4> S S i i i 1 i i § i i 1 i H i i i 1 i 1 i i s i i i ■* s s A i S i S i i i § i % i g § i o i i i i i 1 i s s 1 1 s i i i i i i i i i i % i i i s s i i S5 i g s % i 8 i i i 8 § g i i S i i i i § g i S S i 1 g g i i i i Ol 8 i i i i i 1 A s § i 1 § i i s a SI § i S o Si i i i g i i S 1 i a i i 8 A 1 g i i i i g i i i § 5 i s § i i 8 8 1 ■* 1 8 S ^ 1^ s; 1 S a i i ills i i i i S i 1 a S 1 i i § § s i 8 i s s 8 i 5 i i i i s § i s 1 ^ i K i % i ^ 1 i i i i i 8 i g i i S 1 1 1 1 i i i i 1 i s i i i % § i i i s s i i i i 1 g ^ 8 ^' s i i S i i i g i g i § 1 t^ i i i i % i 1 s i o 1 S i i O s i s i § i i % ■* 1 S i S 1 i i S " i i § % 8 8 s i S 8 g* s i i i i s i i 1 i 1 i i i 5 2 S a s i 8 s i 00 i 8 x: s 1 i i i i i § li ?. s I s § i s i i s S 1 i s 1 i i I i jq g § i i i i i S i 1 i i i i s s S i 1 i 1 i i i i i I s % i § i i i i S i i 1 S i i i i s 1 § i i i i o s; 8 1 1 I ^ i i i i i i i 1 i 1 s o 1 s i I i 1 1 i 1 i i i 8 i i i 1 A i g S § i 1 i i -X) §5 i 1 i 8 i i i i § i i g i i i 1 i S i i 1 § s i i s g i 1 i 1 s 8 i i s i i 1 1 i i ^ 1 1 8 H A i i ills i i i i a i s i § 1 i i i i § 8 i i s ^ 8 s; A i i ills § i 8 s § i i i i o S! i § i 1 i i i i i i 8 i i i 1 i 1 i 1 i i i i i i i i i IS Si s i i i 1 8 i i i 8 Si Date. 1 ' 1-3 1 =1 = p. s g s 1-^ s "a s ■< " to s 1 o s 1 s g Q s 1 s s 169 Range from hourly readings. Unit iy. - s t- « 2 S3 S s; s S S 3 S3 ?! s s s » S 2 s 2 s s a .. s 2 6 6 II OS i s s 3 3 t- g 1 S 3 n S g S S g g o s (^ £ s 3 i 1 i f. O g ■a i i 3 i i 3 i i g g 70 s S t^ s -1> g g g t- t- •J 3 £ g i £ £ J3 i a i 3 3 £ t^ 1 i 3 £ g t^ 1-- t- s K t- g 1 t' b- -*« £ s 00 £ £ o i i o 3 3 t- ai £ 3 £ £ o t- s e to 40 S g S 1 i i £ ^ i i o 3 3 £ i 1 i 3 t- t- i i s g I- g t- g 3 1 i i i O s S i 2 s 3 S g g 1 s ■* £ 3 K g i 1 P g s g g g -* i i i i 1 J i ;:; i 3 3 1 g § e 3 £ 1 t-- t^ g g g g S g g g S i o i 1 .a s o o 3 s 1 1 £ i 3 £ O t- 1 t- g t- t- i t' g 3 i i 1 § i g § i i 3 3 S i. £ e 3 1 s t- s g g t^ g o t^ g g CO i 1 1 § i 43 S o i 3 1 i g £ o 3 g t^ b- S g It- g g e i^ t- g 3 i i 1 i 1 ja i i o 3 3 i g £ t- 1 £ i § 1 t- g g o t- t- OS 3 i 1 s i i g i 1 i S 3 s c- g t- s £ t- s t- l:^ g s t^ »- S g b- g 1> 00 i ^ s 1 1 5 i 1 i s i s s 1 ■p i S s s rs g s g r: g g t- g 2 i 1 i £ b- § a i 1 i 00 o o t^ S i s g g ■* i i g 1 g 1 3 b- b- i § i i b^ i J3 i i i 3 i g g E 1 i t- g g § g 1 t- S 1 1 g b' i s i i b- £ o i i i O 1 E g 1 g i g b- s g g g ?I g g g g ■^ i ? 1 i g 1 s i i i o X 3 b- 1 g 1 s t:^ Tf t^ g g g g g 1 g i g £ i i " jd s s i 3 i 6i g 1 -x> 3 t- t- £ s. t- g -*< fi s t b- g i i Ol 1 Ol 1 s s s s 3 3 £ S o £ 3 £ p g i g g (^ s b- g 1 i £ i £ i § i S s 3 s s I 1 g S £ £ s g g 1 g P g 'p 1 1 i 1 § £ i i i 3 3 3 * S 1 £ g 3 g S s s 1 i ■* t^ t^ t- S g 1 i 1 i b- i T? S3 S 3 3 3 8 s £ 1 3 £ i g t- i g ?1 g o 7** g g 1 i § i 1 § ^ i 2 3 3 3 Ci 1 b- S 3 £ s 1 1 s g 1 t- o s t' g 1 i i i g i ^' i S 3 3 3 o; i £ s 3 a s g i s g g K g ■3 g S s s i i £ i 2 i s 1 3 3 1 s £ i 3 £ s i 1 g g S t^ s It- g £ s £ i i S i i s S Z 7 3 3 g R 3 s g g g s g s t- g 1 ^ g i O i i CC 1 1 i 3 s 3 S 1 " ■< u^ S 1-3 s •-3 s < s -■ £ 1 o S 1 s s S •-3 s fii 2 170 to o ^ o in 115 5 2? 3 S 5 :; ?5 S S S S 3 ?! ?! S S 3 3 3 S S S a ?! ?! 6 d II ti 1 1 1 i 5 is 3 i 3 3 § i § 1 § § 3 3 5 3 1 3 S 1 3 3^3 13 1 1 i i 1 § e 3 1 i i 3 S i 8 § a s s s ■^ -j< -^ -j( 1 3 3 3 3 3 3 i xi 1 1 i i 5 i? S S 3 S -v ■«< ^ ^ ■* 3 i i i g § b. <£> t^ to •* •* -w -.)« CO » -^t* t- 00 ^ ^ 5: 5J ^ 3 3 i o 1 1 i s § g s S i § i § 3 S i § J^ S 2 2 '.n 5 5! 5 -w -* Tji 'fl- -j« 2 c3 S ^ ■* ■* g 1 1 § 1 s ^ w ^ w g 3 3 3 S 1 S 2 2 2 2 § S3 5 3 ^ s ^ s r* ^ J S 1 1 i i S §55^3 i 3 3 3 3 3 2 M CO XI i i s 3 i 3 3 3 1 1 S 3 § % % ^ ^4 % 3 3 3 3 3 3 •J" 1- 'J' ^ 3 3 3 3 3 3 3 3 fj 1 1 s 1 1 i § S i i i 3 3 3 3 3 3 § 1 3 3 3 3 3 3 3 3 3 o* 1 1 i i i 3 3 3 3 3 3 ^ ^ § 5 1 3 3 3 § a is g ■-s< ■q. ^ ja 113 3 5 a § 1 s s 13 3 3 3 3 S 3 1 S 8 S3 s a S ^ ■* ■* TJ. rj- 3 § 3 ja 1 1 § i 1 S S3 a S3 s ■V ■* "fj" ■* ■* 3 3 3 3 1 § ^555 3 3 3 3 3 § 3 3 d 1 1 i i 3 ^ ^ ?S ^ 2 •^ -!« -^ tC ■* § 3 3 3 § § » C-. w; »- 1 3 1 3 3 3 3 3 J3 1 1 3 S ^ § i 5 3 ?! 3 3 3 3 § 3 S 1 5 £ 1 3 1 3 3 3 3 3 1 1 1 1 1 1 1 1 1 s i 3 3 I i § 3 3 3 3 (N r-* 01 W 3 i 3 3 1 1 s? s s 3 3 S S S S S I 3 § 1 3 3 3 5 i 3 3 3 § 3 3 i O 1 1 3 3 3 §1 S5 S S 3 T^ Tjt ^ ^ ^ S 3 3 5 S 3 3 3 3 3 13 3 3 3 CM « « ■^ ■* -# § 1 1 S ^ S o in ^ ^ ^ § 3 S S g 3 ^ ^ ^ ^ ^ ^ 13 3 3 1 i 3 3 3 ^ ■* Tf i 1115 5 s i 3 i i 3 5 1 1 3 S S 3 3 i S 3 3 3 3 i 3 1 5 1 1 g 5 i 1 S S R S S i 3 S 3 3 S S s S ^ ^ T ^ 3 3 3 3 3 3 3 i -a 1 1 1 1 s i i i i 1 ID -^ CO S Tji ° 3 3 i 1 13 111 i i 3 s 1 1 1 1 1 5 1111 i 1 3 3 3 3 3 3 3 3 § s i 3 3 3 i § x: 1 1 S f 1 S E3 g iS S -* -J- ■* ^ ^ 3 5 S S S S ^ ■«" ^ ^ TJ. ^ 3 3 3 3 i i 1 3 3 3 3 3 j= 1 1 i § 3 s s a § s PD CC Ol ■•* CC W 3 1 3 3 s s s s s c3 3 3 3 3 3 3 3 d 1 1 3 S i g f 2 § S 3 3 3 3 3 § ■^ ■<*■■* -* i 3 3 3 3 3 i i 5 1 1 S 3 ^ iS 3 2 S 3 T -J- ^ -^ «* 3 i 5 3 3 § ■* •* 5 •* i 3 3 3 S 3 1 i I 1 S ?5 :; 1 1 rj" -^ -f 3 S § 1 S 3 i S i 3 3 a § 2 S i 3 3 3 i 333 1 1 : : : : '■'-■■% : : = 4 b E < ' 'S. ' ^ ' ^ ' < 1 1 ^ 1 ' s - 1 a Q ^ z 2 172 1 d P 1 1 S . ' s s? ' o s g g ^ 5 R -p ?§ s s " ^ ^ ^ ^ s a 2 ^d ^ O) rt C30 00 j^ __, j^ „ OO CO oo J^ OS ,, ^ r>i ifi ^- 00 Q (M ^ oo (M ^ o ^ X3 I 1 s if! fin s on E s OD ffi &' 00 00 ss s 00 00 00 00 s S SB ^ ^ ^ o> ro Oi n « ^ (3 50 in 00 <_, Q ^ 00 CI CO ■^ 00 00 in t^ 00 '-' ^ ^ N 00 pj j^ ^ 00 Q o Q ^ CJ ■^ OJ Ol 00 Oi a. «> * U. a. JU u. a> UJ UJ UJ tn m ^ ^ ^ m Ol ^ -n in „ rM .n ^ ^ 1 ! S s s « «> " s » «, s «> CC =o iC OV -£i ^ -X) •-D 00 CT> .— d CO CC lO ^ ^ 1 1 '^ :-p? « ^ 00 s 00 00 00 or s !W s ;?; « 00 s oo 00 00 j_^ tH in 00 oo CT> Oi 00 CO -£1 o ^ Q ^ 00 cr XI 1 1 s § s " s ^ a CO " ffi ffi " " s S CO » CO « CO CO Oi ■>* ^ P3 j^ Oi ^^ CM CO in Ol in t- t- ^ 1 1 § S s s «= a 00 «= CO CO CO «, ^ CO CO 0= S §8 ^ , •p -n ^ rr, -+" ^, i-n ^ -f m in t- ^ ^ ^ «i J3 1 1 s " s ^ S a CO CO ^ 00 s^ & CO S =o CO CO CO , ^ in >n -n ^ m T -v ^ Ol to in f. -n in ns n-1 o I- ^ 1 1 s O « " » ^ CO CO CO ^ ^ OD OJ ■-D Id (_, ^ 00 ^ OV in II & t~ OJ OJ u. OJ JJ UU UU UJ lA. UJ UJ ^ UJ "^ UJ ^ cc Q ^ J^ 00 ^ oo CO C) 00 00 ^ in CO 00 ^ 1 1 s " s g « a CO CO CO " s a « s s s 00 Sg 38 F^ OS in Ol OS ^ eo ^ Ca OS 00 OS ^ ^ in 00 -H ■-* ^ ^ ^ N CO CO « <=» 1 1 t' ^ 00 CO cu CO UJ OJ UJ UJ Oi u. CM in C) ^ m ■^ iQ .n 00 oo g s 3 H s » s - ^ CO CO CC CO CO g s ^ CO S " « ^ i eo o 1 i CO CO a. u> '' OJ OJ UJ «j w UJ UJ UJ UJ » CO ^ __, ^ J_, eg ^ €0 <» ;_, (M o in ja 1 1 3 s s g s s s « g s CO s S s " 4; CO as __, Ol N 00 p-i ■* (N ■rti ^ Q ^ CO 00 Q ^ O) c» u> t^ t- CO m 1 1 i t- s 3 s S s s S3 " CO ^ CO s^ ^ CO CO - CO a CO s s , ^ ro J^ nt in m fO r- ^ M ni in nn ,, W OJ « 1 1 t- 00 " O) CD UJ Oi u> UJ U) UJ UJ UJ « 00 ^ to « 1^ o> m Ol -*< ^ ^ -i> OS UJ UJ UJ UJ ^ 2j3 1 1 ° g s o 2 o s s S ?; s s s S5 s s ^ ? s s 5 o s •s. S i g° CD ■■ ■" t " e ^ " ^ •^ ^ '^ ^ " t. ' ' ■"■ . ■ ' ' ' ' "^ February i a -. < a C 3 •-3 J3 e i o O J2 i a •ts February 173 o H I— I o t-H 174 PS s S S P P i: g P g P e g P s s S g P 175 -1 1 1 1 1 l|SI I§SIS2?!S3I3SSS3SS|§ + II Q q| -1 1 1 1 1 1 131 lil|°SSS3|22gsll. g2| -1 1 1 lsBsSSSsS»2S§|S2Sg2|g|SgS J3 ~l 1 1 IS2SSsS3SSSSSSS|§3Rg|SS$S 1 ,i 1 1 issl3slSi2s"-Si"iS3"IBS"Ss -1 1 1 ii2sSs2saii^l2"i3ISSsS^§i i ~ 1 1 1 1 3 s s 5 s s g s 1 2 s s 1 s S S S 1 g » S s S S ja" -1 1 1 lS2sSB2S5|»S|Sg||3|S5|£gg J3 -1 1 1 issl3Sss2i^^2" = sSs"i"*"Si 5 -1 1 1 l3Ss2lsSSl»"-|g§g2gS2SS§2e :;" -1 1 1 1 i3Sl2SSsi^°2^»3iSais°sS3 4 -1 1 1 1 iS332S2S"""2""5SSss§ssss s 1 1 r ,^£2^2I'2'-^^'=^"-^'^'~'*^"''-'*^'*"5^''^=^'»"^'-" c i -1 1 1 1 iSSsSs53^^-"-2^^-Si23l*"»S2 -1 1 1 1 ISlSS2SS"»S§Sg2SgSSSgSSS ji -1 1 1 1 1 iSSS2^S^»"2»»2sS"S°S"S5 o -| 1 ] 1 I joa-r-* loi — aoOioooOixCK.Sxp^-^t^»-^?)S A ca -1 1 1 1 1 lO-^CCIOOfNOOaiaDOiaiOl QoSQOOt--J5SoiO J= .i 1 1 1 1 |Sg| iggesggssgsKSS^sss^ J3 -.1 1 1 1 1 |Sg| |SSSS£SSgSS5SS::g£3g2 J3 .| 1 1 1 1 ,33 1 |S3SS£gSgSSS3SSSSS5 x: tr~ >a Mdcc-j-oDt^-xicoeouri'^NOioooioJ-^tD ja .j j j j 1 l^^l |2gaioSS»SS^c^9w2o^Sg ^ -| 1 1 1 1 I^Sl |S^83ci£SSoMS^2Soi:raS2 X ~ 1 1 1 1 1 1 S s 1 1 S S « § » 2 » » S 3 — S S S — S Z3 ,| 1 1 1 1 1 |l| |3| liSSSSSSSSSSSSSS -1 1 1 1 1 1SS2I |!!°-2°22"-SS2°§l"2^l e January 1, 1902 15, „ February 1, ,, 16, „ March 1, ,, 15. ,, April 1, ,, 15, ,, May 1, ,, 15, ,, June 1, ,, 15, „ July 1, „ 16, „ August 1, ,, 15. .. September 1. ,, 15, ,, October 1, ,, 16, „ November 1, ,, 15, ,, December 1, ,, 15, „ January 1, 1903 15, „ February 1, ,, 15, „ 176 »,a ^ ^cT c 1 *"■ ;x «*^g 'g A A A A A A A ^C ^ ^ ^ .^ ^ ^ ■^1 •x> lO ^ «i ^ OS A V A ^ M rr, m ^ n-i ,, ^ in ^ -It ^ ^ rl 00 a A ^, -n ^ in in m ro -f ni in -I* ^ ^ ^ -1 in CO r-1 ^ ^ n ^ UJ lij "•' a. o. ^■' ^~ "^ "'■' 'M m ri ^ rh ■^ ni .n ni ^ ,, n- „ -f P-. 00 '-' ^ Of. rr. ■n in ^ ^ rr, ^, -f nn rr, ,^ ro ^ ov ^ ■* a. 1 1 1 1 t- ^ :o u. "*' ■^ ■"" '^' i- " "" V " -*■ '■ ^ ^ ^ ^ ^ ^ ■^ o ^ .n CI a-, o s g CD 1 1 1 1 X> ■^ UJ u^ a. *"■ ^ -' UJ ^j ^ LJ , ^ CI ni m *f ^ ^ f ro ^ I- .-. '- ' ' ' '"^ O. ■^ iij lO a> "^ '^ "^ fj CO >.'J CO CO -iJ in ^ ^ in ^ in ^ -tt _, ^ m -P r- a ^ ^ t- ■■" 1 1 1 1 « "•' OJ ^u ^ ^ ^' LiJ _ -H ^ m ^ rH m in m 3; lO in in s •n cr. OV s s " -i. a. 'W "' ^ OJ ^ u. 4J ■^ 1 " ^^ ^ „ ^ M ^ ■n ^ ^ ,, CO lO CO ^ ' «j *" U> ^ ^' "^ "^ ^ t'J CO CM IP ^ "1 , r-i ^ in -f ^ -n CO -N ra ■x> n. in ^ ^ »- " " ■"' "■' Ml '* ^^ m in rri ^, in ^ ^ rri m ^ r-. rn ^ ■n on j^ in -n in ^ in i-n m ^_ ^ Ift 00 -" "* UJ '■^ a. ^ ^ ^' tu '■^ ^ r- ^ ^ ^ ■^ m ^ m ^ ni ni m m CO t- .Q II K Z rr -, ^ ^ n^ m m _ ni ^ en m ov ^ I- w ID OS (N is " o. "•' "•' to ■^ t- ^~ X. UJ ^ " UJ rn ^ ni ,n t-. ^ M in ^ m Oi rr^ c-i a> in