r. 
 
 LIBRARY 
 
 OF THK 
 
 UNIVERSITY OF CALIFORNIA. 
 
 OR 
 
 Received 
 Accession No. frJtS-tf'ty Class No. 
 
OS 
 
 oc 
 
 LU 
 (/> 
 
 CQ 
 
 o 
 
 LJ 
 
 z 
 o 
 
 \i 
 
 
 
 < 
 
THREE ESSAYS 
 
 ON 
 
 AUSTRALIAN WEATHER 
 
 HON. RALPH ABERCROMBY, 
 
 FELLOW OF THE ROYAL METEOROLOGICAL SOCIETY, LONDON ; MEMBER 
 
 OF THE SCOTTISH METEOROLOGICAL SOCIETY ; AND AUTHOR OF 
 "PRINCIPLES OF FORECASTING BY MEANS OF WEATHER CHARTS." 
 
 ^gg? 
 
 FEEDEEICK W. WHITE, PEINTEE, 39 MAEKET STEEET WEST. 
 
. 
 
 UBTI7BRSIT7 
 
 [NTRODUCIION 
 
 THE following essays on "AUSTRALIAN WEATHER" have been 
 brought together and published in book form by the Hon. 
 RALPH ABERCROMBY, for convenient reference by students of 
 Meteorology, because he thinks that they are important contri- 
 butions to our knowledge of the Meteorology of a new field. In 
 passing it may be mentioned that the preparation of them has 
 been made possible by the very complete Meteorological records 
 of the Sydney Observatory, especially the continuous automatic 
 records of wind, rain, barometer, and weather charts introduced 
 by the present Director. These afford the means of reference to 
 wind and weather at any moment since 1863, and the barograph 
 since 1870, without which it would have been impossible to study 
 in the necessary detail the so-called " Southerly Burster," which 
 is perhaps the most remarkable of the " squall " winds which are 
 found in various parts of the earth. 
 
 The first essay, "Moving Anticyclones," is here reproduced 
 with the consent of the author. It establishes the progressive 
 easterly motion of weather over Australia. The great importance 
 and controlling influence of anticyclones upon the usual weather, 
 traces their progress over Australia, and the changes produced in 
 them by the contour of the country, gives records of the number 
 of such weather systems which pass over Australia every month 
 and year, and traces their motion in latitude with the seasons, 
 and shews that their average daily progress over the mainland is 
 
IV. 
 
 four hundred miles per day, and produces evidence which seems 
 to justify the conclusion that they travel rather faster over 
 the ocean between South Africa and Australia, that is, at the 
 rate of four hundred and fifty-eight miles per day. Twenty 
 diagrams of weather and anticyclone tracks are used in illustration. 
 
 The Prize Essay on " Southerly Bursters " was the outcome 
 of a valuable prize offered by The Hon. RALPH ABERCROMBY. 
 It follows the " burster " through all its phases, its connection 
 with moving anticyclones, with the state of the barometer, with 
 the heat of the interior, and with various other phases of weather. 
 Traces them historically, and gives reasons for thinking that 
 they are not so violent now as they used to be. It is illustrated 
 by four cloud pictures and eleven diagrams. 
 
 "Types of Australian Weather." These types forming the 
 third essay, were selected by Mr. ABERCROMBY, and he marked 
 out the way they should be treated, and engaged Mr. HUNT to 
 do the actual work necessary for their publication. These studies, 
 as well as those in the essay on "Moving Anticyclones" have 
 been rendered possible by the extreme care with which the 
 Sydney daily weather chart has been prepared, and passing phases 
 of weather submitted to special study since its commencement. 
 Forty of the weather maps on a reduced scale are used in illus- 
 trating the types. 
 
0* TH3 ~4^ 
 
 UiriVBRSITT 
 
 MOVING ANTICYCLONES IN THE SOUTHERN 
 HEMISPHERE. 
 
 BY H. C. RUSSELL, B.A., F.E.S., F. B. Met. Soc. 
 
 Sydney Observatory, New South Wales. 
 [With Twenty Diagrams.] 
 
 [From the Quarterly Journal of the Royal Meteorological Society, Vol. XIX. 
 No. 85. January, 1893.] 
 
 SINCE Western Australia established in 1887 a number of new 
 Meteorological Stations, it has been possible to make our Weather 
 Charts more complete and to trace the progress of meteorological 
 conditions more minutely : and the opportunity has not been lost. 
 Some of the results are of purely local value as aids in forecasting. 
 Others seem to have more general significance, and will, I think, 
 be of interest to the Fellows of this Society and to meteorologists 
 generally, because they have an important bearing upon accepted 
 theories in explanation of " weather" in latitude 20 to 50 South. 
 So much so that I think it will be necessary to modify those 
 theories. There can be no doubt that the great extent of ocean, 
 as compared with the land, in the latitudes named, affords to 
 atmospheric circulation a field in which it may approach what it 
 would be, if the earth were completely covered by water, and the 
 atmosphere therefore free from the disturbing influences of 
 unequal heating, surface land friction, and mountains. 
 
 The leading fact that our 1 investigations have brought to light 
 is that Australian weather south of 20 South latitude is the 
 product of a series of rapidly moving anticyclones, which follow 
 one another with remarkable regularity and are the great 
 controlling force in determining local weather. 
 
 1 In the investigations which lead up to the results detailed in this 
 paper I have been very ably assisted by Mr. H. A. Hunt, who prepares 
 the daily Weather Chart, and who has carried out many investigations 
 to the successful discovery of weather laws here. 
 
 A 
 
AUSTRALIAN WEATHER. 
 
 . 
 
 The fixed anticyclone over the Indian Ocean in these latitudes, 
 which is found in books of reference, must give place to a moving 
 series. It is not difficult, now that the true weather conditions 
 are known, to trace the development of the idea of a fixed 
 anticyclone over this part of the ocean. The moving anticyclones 
 are about five times as large in area as the low pressure A 
 between them, and they are also moving to the east at the rate 
 of about 400 miles per day. Now vessels crossing these at 
 random would necessarily find five times as much high pressure 
 as low pressure : and if travelling with the anticyclone would 
 keep in it for many days, perhaps all the way from the Cape to 
 Sydney, as the Havanah did. When all these barometer readings 
 came to be plotted on the chart the result would be a fixed anti- 
 cyclone, which we now know has no existence in fact. It is not 
 so easy to trace the history of the " overland " fixed anticyclone 
 of more modern writers, but it is equally imaginary. The high 
 pressures regularly move on over land as well as over ocean. 
 
 It will perhaps be more convenient if I state here, in as few 
 words as possible, some of the results obtained, and bring forward 
 subsequently in more detail the data upon which they are based. 
 These results may be briefly stated as follows : 
 
 1. Instead of fixed anticyclones we have a series of moving ones. 
 
 2. The average number of anticyclones passing over Australia 
 in a year is forty-two, and so far as the observations go, the 
 number varies but little. See Table I. (p. 5). 
 
 3. Anticyclones are more numerous in summer than in winter. 
 See Table I., and diagrams 9 to 20. 
 
 4. The latitude of anticyclone tracks varies with the season, 
 being in latitude 37 to 38 in summer, and 29 to 32 in winter. 
 See Table II. (p. 7). 
 
 5. Upon the average, an anticyclone travels across Australia 
 in seven days in summer, and in nine days in winter. Since 
 forty-two pass over each year, the average time of passage over 
 any place is 8 -7 days. 
 
MOVING ANTICYCLONES. O 
 
 6. The average daily rate of translation derivgd from all the 
 available records is, over Australia, four hundred miles; and over 
 the sea and land from Natal to Sydney, four hundred and fifty- 
 eight miles. (See page 8.) The rate of translation of moving 
 anticyclones varies from two hundred to nine hundred and fifty 
 miles over land. 
 
 7. The shape of the anticyclone over the comparatively flat 
 lands of Australia is an ellipse, with axes in ratio of two to one, 
 the longer axis being east and west. The shape, as well as the 
 direction of the major axis, are, as a rule, modified when the 
 anticyclone reaches the coast range, the result being a shortening 
 of the major axis, and a bending of the major axis to or towards 
 a position at right angles to what it had on the low lands, i.e., 
 making it north and south. See diagrams 1, 2, 3, 4, 5, 6, 7, 
 8 and 9. 
 
 8. The winds on the north side of the anticyclone are not so 
 strong as those on the south side ; but at the ends the winds 
 have greater velocity, and the two winds, northerly and southerly, 
 pass each other as if struggling to get through between two 
 obstacles, i.e. the preceding and following anticyclones. See 
 -diagrams 1 to 8. 
 
 9. The intensity of weather is in proportion to the difference 
 in pressure between the anticyclone and the A depression, but the 
 relation of the pressure varies frequently before the wind responds, 
 and it seems as if the pressure was controlled from above by the 
 more or less rapid descent of air, which feeds the anticyclone. 
 The centre of the descending current, assuming that to be the 
 point of greatest pressure, not unfrequently moves about 
 independently of the general motion, so that the centre at times 
 seems to retreat, without corresponding motion in the extreme 
 parts of the anticyclone. 
 
 10. When the A depression is deep it is usual for the South- 
 east Trade wind, blowing in the north of Australia, to be deflected 
 into a Northerly wind in the rear of an anticyclone, which gets 
 
4 AUSTRALIAN WEATHER. 
 
 heated as it bloys over Central Australia, and becomes the true 
 hot wind of the Southern Colonies. This also explains the sudden 
 shift from hot northerly to the cold southerly wind, blowing on 
 the following side of the A depression. 
 
 11. Cyclonic storms are very unusual, and do not appear more 
 than once in two or three months. These come from north-east 
 to east, or from the north-west coast across the Australian 
 continent to the sea at the Great Bight, thence they travel 
 eastward. They do not seem in any sense to be part of our 
 weather system, but to be offsets from tropical storms. 
 
 The depression between anticyclones is essentially a A depres- 
 sion, both in the shape of the isobars and in the sudden change of 
 wind which follows the passage of the lowest pressure. 
 
 It is, of course, impossible here to pass in review all the 1,400 
 Weather Charts which have contributed to these deductions ; and 
 I have, therefore, selected for reproduction here a set of eight 
 Weather Charts in which the passage of an average cyclone is 
 clearly depicted (Diagrams 1 to 8). The isobars have been repro- 
 duced on a convenient and small scale, and they show much better 
 than any description what the ordinary fine weather sequence of 
 events is. Attention may, however, be called to one or two points. 
 No. 1 diagram shows the incoming anticyclone well established on 
 the coast of Western Australia, with a passing A depression in 
 front of it extending over the southern parts of South Australia, 
 Victoria, and New South Wales. No. 2 shows the twenty-four 
 hours' forward motion and the closing up of the isobars as they 
 reach the east coast mountain range. Nos. 3 and 4 show the further 
 progress of the high pressure and the decided effect of New 
 Zealand mountains in intensifying the A depression. No. 5. 
 As on Sunday we get no telegrams, probable isobars have been 
 drawn. No. 6 shows the anticyclone over New South Wales, and 
 the preceding end of its major axis tilted northwards by the 
 mountains, while in Western Australia we see the first isobar of 
 the incoming high pressure, and over Perth the A depression 
 which divides the two. No. 7. The axis of the anticyclone is 
 
MOVING ANTICYCLONES. 
 
 5 
 
 now nearly north and south, and it has passed over the 
 mountains. No. 8. The anticyclone is over the sea between 
 Australia and New Zealand, and with its two axes nearly equal. 
 Such is the passage of an ordinary anticyclone. This one was 
 moving at the rate of four hundred and fifty miles per day, which 
 is somewhat above the average speed ; and by referring to the 
 diagrams it will be seen that its track is nearly straight, and from 
 west to east. The majority of such anticyclone tracks are bent 
 southwards in the middle, as if when the anticyclone reaches the west 
 coast range, which trends to south-south-east, it is deflected so as to 
 move to the east-south-east instead of east ; and that when it meets 
 the east coast range which trends to north-north-east the track is 
 again deflected by mountains and made to go east-north-east or 
 north-east. See illustrations of this in diagrams 11, 15, 17, 18, 
 19, and 20. 
 
 Another feature brought out in the diagrams is the occasional 
 stoppage of the anticyclone. For instance, in diagram 12, track 
 C, the centre was about the same place from April 25th to May 
 1st. In No. 15 the track is remarkable, the anticyclone lasting 
 tw T enty-five days, although it moved every day. Of its kind, this is 
 the most remarkable one we have on record. 
 
 Out of a total of forty-two anticyclones which passed over 
 Australia in 1891, six, or fifteen per cent., hesitated or actually 
 stopped in their forward motion. Table I. shows the number of 
 cyclones in each month since February 1888. 
 
 TABLE I. NUMBER OF ANTICYCLONES IN EACH MONTH. 
 
 Years. 
 
 January. 
 
 February. 
 
 March. 
 
 i 
 
 I 
 
 
 
 i-a 
 
 >* 
 
 i-s 
 
 1 
 1 
 
 | 
 
 ,Q 
 
 1 
 
 02 
 
 October. 
 
 1 
 
 1 
 
 1 
 
 1888 
 
 
 
 2 
 
 3 
 
 3 
 
 2 
 
 3 
 
 5 
 
 3 
 
 5 
 
 4 
 
 4 
 
 1889 .... 
 
 3 
 
 4 
 
 2 
 
 3 
 
 4 
 
 4 
 
 3 
 
 3 
 
 
 4 
 
 4 
 
 5 
 
 1890 
 
 5 
 
 3 
 
 4 
 
 3 
 
 2 
 
 2 
 
 2 
 
 3 
 
 2 
 
 5 
 
 5 
 
 6 
 
 1891 
 
 4 
 
 4 
 
 3 
 
 3 
 
 3 
 
 4 
 
 2 
 
 4 
 
 4 
 
 4 
 
 4 
 
 2 
 
 1892 
 
 3 
 
 4 
 
 3 
 
 4 
 
 3 
 
 4 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Totals 
 
 15 
 
 15 
 
 14 
 
 16 
 
 15 
 
 16 
 
 13 
 
 15 
 
 13 
 
 18 
 
 17 
 
 17 
 
 Mean 
 
 3-8 
 
 3-8 
 
 2-8 
 
 3-2 
 
 3-0 
 
 3-2 
 
 2-6 
 
 3-8 
 
 3-3 
 
 4-5 
 
 4-3 
 
 4-3 
 
6 AUSTRALIAN WEATHER. 
 
 It is difficult to understand how an anticyclone can, to all 
 intents and purposes, stand still for several days when the whole 
 surrounding atmosphere is moving forward : and we have not yet 
 made out the explanation. But facts, and some very significant 
 ones, are accumulating which indicate the probable explanation. 
 For instance, when such a stoppage occurs, the isobars in front 
 widen out, showing that the preceding system is moving forward ; 
 and the closing up of the isobars in the rear shows that the 
 following one is coming forward. We know also, as already 
 pointed out, that the source of pressure, the descending current in 
 the anticyclone, may, and does, vary in locality from day to day 
 with regard to the outlying isobars, coming down at one time on 
 the preceding side of the centre and at the next on the following 
 side. This gives an oscillating position to the apparent centre. 
 But it seems impossible that a mass of air, even that in one anti- 
 cyclone, without reference to its surroundings, measuring as it 
 does 2,000 miles by 1,000 miles, can be actually stopped in its 
 forward motion. This point is however, one of those still under 
 investigation, and we hope in another paper to make the explan- 
 ation complete. 
 
 In America it is well known that cyclones move ten to fifteen 
 per cent, faster than anticyclones, but there the low pressure 
 acquires a velocity of its own in addition to that which it has in 
 common with the general mass. The conditions here, as already 
 pointed out, are different, the low pressure not being independent, 
 but tied to the anticyclone, as effect is to cause. 
 
 Another point of considerable importance is the normal latitude 
 of the anticyclones in each month of the year. At times it is a 
 very uncertain matter, for there is as much difference in the 
 latitudes of tracks in the same month in some cases as there is in 
 the average tracks for each month of the year. Still, taking the 
 four and a half years of available records, there is a very obvious 
 monthly change of latitude, which is best indicated in tabular 
 form. 
 
MOVING ANTICYCLONES. 
 
 TABLE II. MONTHLY CHANGE or LATITUDE or THE ANTICYCLONES. 
 
 Months. 
 
 1888. 
 
 1889. 
 
 1890. 
 
 1891. 
 
 1892. 
 
 Mean 
 Latitude. 
 
 January 
 
 
 
 
 
 37 
 
 38 
 
 36 
 
 
 
 36 
 
 
 
 36f 
 
 February . . 
 
 
 38 
 
 41 
 
 36 
 
 38 
 
 381- 
 
 IVEarcb. 
 
 34 
 
 38 
 
 39 
 
 37 
 
 37 
 
 37 
 
 April 
 
 37 
 
 32 
 
 35 
 
 36 
 
 35 
 
 35 
 
 IVIav . 
 
 33 
 
 32 
 
 31 
 
 34 
 
 33 
 
 32?- 
 
 June 
 
 27 
 
 27 
 
 27 
 
 37* 
 
 29 
 
 27* 
 
 July. 
 
 27 
 
 30 
 
 30 
 
 29 
 
 33 
 
 29-* 
 
 August 
 
 28 
 
 31 
 
 30 
 
 30 
 
 
 29 
 
 September 
 
 32 
 
 32 
 
 30 
 
 30 
 
 
 30^ 
 
 October 
 
 33 
 
 32 
 
 29 
 
 33 
 
 
 311 
 
 November 
 
 38 
 
 37 
 
 32 
 
 36 
 
 
 354 
 
 December '. . 
 
 39 
 
 37 
 
 35 
 
 35 
 
 
 36^ 
 
 Not included in mean. 
 
 It must be borne in mind that the tracks are seldom straight, 
 and that at times they are very erractic. The above positions 
 for each month have been obtained from careful eye estimates of 
 the mean positions, and this must be borne in mind in using the 
 table. The tracks for 1891, diagrams 9 to 20, will illustrate what 
 is meant, and, will justify the course adopted in preference to 
 that of taking the measured latitude of a series of points in each 
 curve. The individual curves are too irregular for such a method. 
 
 It will be observed that the maximum of latitude is in. February, 
 a month after the hottest month ; and the minimum in June, 
 which is not our coldest month. Perhaps however, June should 
 be rejected from this attempt to determine the monthly latitude, 
 because in that month, for some reason not obvious, the tracks 
 are much more erratic than in any other month of the year, and 
 it is almost impossible to take a mean latitude for it ; and if it be 
 rejected the minimum would be between July and August, July 
 being our coldest month. 
 
 It is our experience that when an anticyclone track is far from 
 the mean, the weather is also far from the mean. For instance, 
 in June 1891 two tracks were down in latitude 37 instead of 27, 
 and we had, as a consequence, fourteen and a half inches of rain, 
 an excessive quantity, the mean being five and a quarter inches. 
 
8 AUSTRALIAN WEATHER. 
 
 I have already stated, that taking a large number of anticyclones 
 on their passage over Australia the daily translation eastward is 
 four hundred miles ; and it is a matter of considerable interest to 
 ascertain if they maintain the same velocity of translation over 
 the ocean. I have been so far unable to trace any connection 
 between the variations in the barometers at Buenos Ayres and 
 Sydney, and infer from this that the great mountain chain of the 
 Andes so breaks up the anticyclones that the curves are not alike. 
 The only other place in like latitude for which I have daily 
 barometer readings, and these for only one year, 1890, is Natal. 
 Fortunately there is an obvious similarity in the curves. 1 have 
 compared thece with Sydney by having the two plotted on the 
 same scale and taking the difference in the times at which marked 
 points of high and low barometer readings pass the two places. 
 There are 
 
 In January 5 cases with a mean of 1 3 days 
 February '5 16 
 
 March 5 15 
 
 April 3 18 
 
 May 4 12 
 
 June 3 ,, 14 
 
 July 4 16 
 
 August 5 17 
 
 Septembers ,, 14 
 
 October 4 14 
 
 Novembers 17 
 
 December 5 15 
 
 Total 51 cases Average No. of days 15'08 
 
 The greatest number of days for the translation of the waves 
 from Natal to Sydney is eighteen days for April, which makes 
 the daily velocity of translation three hundred and eithty-two 
 miles ; the least number of days in any month is twelve, in May, 
 which makes the velocity of translation five hundred and seventy- 
 three miles per day. The average rate for all the months is 
 fifteen days, which is equal to a velocity of translation of four 
 hundred and fifty-eight miles per day. 
 
a - * 
 fall 
 
 MOVING ANTICYC: 
 
 1^ 
 
 , 
 
 Taking individual anticyclones in Australia the slowest moved 
 at one hundred and twenty miles per day, and the quickest nine 
 hundred and fifty miles per day, the average being four hundred 
 miles per day. The result of this comparison surprised me by 
 the similarity of the two results. I do not, however, think that 
 very much value should be attached to a comparison of barometer 
 curves for one year only, still it agrees remarkably with that 
 obtained by four and three-quarter years' results for Australasia, 
 and the little difference is in the direction it ought to be, owing 
 to the obvious friction of the land surface, mountains, etc., as 
 compared with the smooth sea surface. 
 
 The two methods of determining the velocity of anticyclones, 
 that is, over Australia alone, where it is four hundred miles per day, 
 and over the space from Natal to Sydney, where it is four hundred 
 and fifty-eight miles per day, seem to leave no doubt as to their 
 persistence. For if they can thus be followed one-third of the 
 circumference of the earth, i.e. from Natal to Sydney, it may 
 safely be assumed that they travel the other two-thirds of the way, 
 and that they keep up their general characteristics. What 
 influence that great obstacle in their path, the Andes of South 
 America, may have on them I am not at present in a position 
 to say, but I have no doubt, from what we see so clearly in the in- 
 fluence of our own comparatively small range of mountains along 
 the east coast of New South Wales, that it is a very material one. 
 
 If from Buenos Ayres we could get by cablegram the state of 
 the weather from day to day, we should be in a position to 
 forecast the coming weather for about a month in advance ; and 
 it may yet be that when our investigations, which are now in 
 progress, are completed we shall be able to forecast far longer 
 periods. If, for instance, we could ascertain the velocity of the 
 translation of the anticyclone round the other two-thirds of the 
 globe, as we have done for the one-third from Natal to Sydney, or 
 rather more than one-third because it extends to New Zealand, 
 then we could ultimately forecast the return, in say, seven weeks, 
 of weather passing over Sydney. Certainly the discovery of the 
 
10 AUSTRALIAN WEATHER. 
 
 daily translation of anticyclones in our latitude, over such a large 
 section of the circumference of the globe, holds out a reasonable 
 hope that they may be traced all round, and the proportion of 
 water surface points clearly to the fact that the conditions are 
 more favourable here than in any other part of the earth for 
 normal atmospheric circulation. I do not by this intend to convey 
 the idea that I think an anticyclone keeps its shape, size, form, 
 and pecularities for weeks together, because I see them changing 
 every day. But nevertheless there are obvious pecularities which 
 affect some anticyclones general characteristics I mean, such as 
 dryness or moisture, which, it may be; are attached to them 
 more persistently than the mere form of the isobars. And if so, 
 it will afford good data for long period forecasting. 
 
 If I have succeeded in showing the normal conditions of our 
 weather, to be that of an endless series of anticyclones passing over 
 to the eastward at the average rate of four hundred miles per day 
 and keeping within a very moderate range of latitude, it will be 
 obvious that these conditions hold out the prospect of our being 
 able to predict the weather for some weeks in advance ; because 
 an anticyclone with such a rate of motion in latitude 38 south 
 would pass round the earth in forty nine days as stated above. It 
 is true that in isolated anticyclones we find the rate of motion 
 vary considerably over a part of its track ; but I think it is a fair 
 assumption to make, that the average velocity of the anticyclones, 
 in a given latitude, is the rate of progress of the atmosphere at 
 that latitude, and that the apparent variations in rate are 
 simply local accelerations or retardations, which would naturally 
 affect masses of air in motion under several influences, that is, the 
 translation, the constant variations in pressure, temperature, etc. 
 Some of the conclusions arrived at from the examination of four 
 and a half years weather charts, may have to be modified slightly 
 when the investigation is continued over other years ; more 
 especially, when the number of observing stations make the data 
 more complete, but the general agreement of the results in the 
 years examined, seems to support the opinion that there will be 
 no material alterations in the results here given. 
 

U5 
 
16 AUSTRALIAN WEATHER. 
 
 AN ESSAY ON SOUTHERLY BURSTERS. 
 BY HENRY A. HUNT. 
 
 Second Meteorological Assistant, Sydney Observatory. 
 [With Four Photographs and Five Diagrams.] 
 
 [Awarded the Prize of 25 offered by the Hon. Ralph Abercromby, for the 
 best Essay on Southerly Bursters, 2 May, 1894.] 
 
 ORIGIN ON THE PRIZE. 
 
 IN December 1892, the Honorable Ralph Abercromby gave to 
 the Royal Society a sum of 100 to promote the study of Austra- 
 lasian Meteorology by offering prizes for essays, upon particular 
 phases of weather, and in rewards for special investigations, 
 suggesting that the subject of the first essay should be "The 
 Southerly Burster." The foundation of this Prize Fund was 
 announced at the general meeting of the Royal Society at the 
 December 1892 meeting. Subsequently the Council appointed a 
 Committee consisting of the Hon. R. Abercromby, Professor 
 Liversidge (Chemistry), Professor David (Geology and Physical 
 Geography), and H. C. Russell, Government Astronomer. 
 
 The Committee met to determine the conditions for the compe- 
 tition and advertised these freely in Melbourne and Sydney, and 
 by correspondence with various kindred Societies in Europe and 
 America, and a note appeared in Nature about it ; one of the 
 conditions was that the competition was open to all. It was 
 advertised in March 1893, and the last day for receiving essays 
 was 31st March, 1894. 
 
 The contents of this essay may be briefly summarized as follows : 
 It begins with a short note on bursters past and present, and 
 weather indicating their approach. Deals with the burster in 
 other colonies, shows that it is intensified in New South Wales 
 by geographical features. Is sometimes caused by monsoonal 
 depressions. Traces the changes in isobars with various kinds of 
 
SOUTHERLY BURSTERS. 17 
 
 bursters. Shows that duration and strength of bursters have 
 wide ranges. Gives a sketch of the Pampero. Traces different 
 kinds of bursters, their rate of progress along the sea coast and 
 relation to general weather conditions. Gives detailed description 
 of two bursters with diagrams of weather before and after, and 
 photographs of clouds, also full notes of cloud changes. Gives 
 diagram and short note about the most violent burster ever known 
 on the coast. Gives tabular particulars of all the bursters that 
 have taken place between September 1863 and March 1894. 
 Showing the number in each hour of the day in each month and 
 in each year, and the total number nine hundred and ninety-one 
 which have been recorded, the prevalence in each month in each 
 year with the greatest velocities of wind and the mean velocities, 
 etc , also a diagram showing the relation of the number of 
 bursters in each hour of the day to the barometer curve. 
 
 THE PRIZE ESSAY. 
 
 In the early days of Australian settlement, when the shores of 
 Port Jackson were occupied by a sparse population and the region 
 beyond was unknown wilderness and desolation, a great part of 
 the Haymarket was occupied by the brickfields from which 
 Brickfield Hill takes its name. 
 
 A BRICKFIELDER. 
 
 When a "Southerly Burster " struck the infant city its approach 
 was always heralded by a cloud of reddish dust from this locality, 
 and in consequence the phenomenon gained the local name of 
 " brickfielder." The brickfields have long since vanished and with 
 them the name to which they gave rise, but the wind continues to 
 raise clouds of dust as of old under its modern name of "Southerly 
 Burster." A consideration of the earliest reliable records, and a 
 comparison of them with these of later times, appears to prove 
 that the phenomenon itself, as well as its surroundings, has changed. 
 
 SIXTY MILES PER HOUR. 
 
 Even up to within ten or fifteen years ago the velocity of wind 
 was frequently as high as sixty miles per hour, and occasionally 
 B 
 
18 AUSTRALIAN WEATHER. 
 
 attained the tremendous force of eighty miles on one memorable 
 occasion it went far beyond this and registered the unprecedented 
 velocity of one hundred and fifty-three miles per hour in a gust. 
 
 ONLY FORTY MILES PER HOUR. 
 
 Now the southerly burster seldom exceeds fifty miles, and 
 generally ranges between twenty and forty miles per hour. 
 Whether this result arises from the fact that civilisation has raised 
 much brick and mortar to obstruct the atmospheric disturbance, 
 or whether it is that the absorption or radiation of heat is less 
 from the cultivated soil than from the hard, unbroken surface of 
 the pre-cultivation days, is a matter of conjecture only. 
 
 BURSTER ANTECEDENTS. 
 
 The climatic conditions preceding a southerly buster are, first, 
 a period of high temperature varying from three hours to three 
 or more days, accompanied in the early part of the summer, or 
 towards its close, by wind from the west or north-west, and 
 in the midsummer months, generally from the north-east. In 
 the early morning on the day of a "burst" the sky is white 
 and hazy of aspect. As the hour of the outbreak approaches 
 there begin to accumulate in the south, ball-shaped cirro-cumulus 
 or pilot clouds, and frequently, if electric disturbances are to 
 accompany the squalls, heavy cumuli thunder-clouds rise gradually 
 in the south-west. 
 
 THE CLOUD ROLL. 
 
 An hour or so before the squall, a heavy cumulus roll appears 
 low down on the southern horizon the interval between this 
 apparition and the beginning of the gusts depends entirely on the 
 velocity of the wind. Afar off this cloud roll appears most 
 frequently due south, but sometimes south-south-west, or even 
 south-west ; it is sharply defined, dark on the edges with lighter 
 shades towards the centre. The roll is from thirty to sixty miles 
 in length. Sometimes it appears singly ; on other occasions there 
 are a multitude of these formations heaped one above the other, 
 with light cirrus below. Generally, if the burst is of the first 
 
SOUTHERLY BURSTERS. 19 
 
 order, it is followed by an overcast sky composed of nimbus from 
 which patchy rain descends. (See Plates i to 4.) 
 
 As the cloud roll approaches it gradually loses its symmetrical 
 aspect, and careful observation reveals a light cirrus fringe rising 
 from underneath it in front, falling over the top, and trailing 
 behind the advancing cloud. Up to this time the wind has been 
 blowing with steadily increasing force from a northerly direction, 
 
 INTERVAL OF CALM 
 
 but at this stage it drops suddenly and a profound calm prevails, 
 broken only at intervals by a few fitful puffs. This state of things 
 lasts for a varying space \ if the southerly arrives during the heat 
 of the day, it endures but a very short time ; if at night the calm 
 is of longer duration. 
 
 THE BURST. 
 
 Meantime the cloud roll is seen rapidly approaching, clouds of 
 dust rise in the southern part of the city four miles away, and 
 gather volume as they come, until they almost hide the city as 
 viewed from Observatory Hill, while from immediately under the 
 roll light clouds rush forward with great velocity only to be thrown 
 back over the top of it as they reach the front, the wind vane on 
 the Time Ball tower flies to the south and the wind reaches us on 
 the ground a moment later, and in a few moments is blowing 
 with the force of a gale. 
 
 VERY LITTLE DEW. 
 
 It is noteworthy that the night preceding a burster, however 
 clear it may appear, seldom precipitates any measurable quantity 
 of dew. This affect is, no doubt, owing to the excessive dryness 
 of the northerly current which absorbs any moisture obtainable 
 by radiation from the earth, or by condensation from the upper 
 strata. 
 
 A BURSTER FOLLOWS A FOGGY MORNING. 
 
 But it is a curious circumstance that during a period of hot 
 weather, should a fog exist at daybreak a southerly change is 
 almost certain to follow within twenty-four hours. Such a con- 
 
20 AUSTRALIAN WEATHER. 
 
 dition is of rare occurrence and is always preceded by high pressures 
 without energy or grade, and consequently at a time when land 
 and sea breezes prevail. 
 
 THUNDER AND RAIN. 
 
 A burster rarely brings immediate rain, except when it is 
 accompanied by a thunderstorm. In the event of a three days 
 blow the downpour, if any, seldom takes place until the second day 
 and even then as is the case with all coastal ruins is heaviest on 
 the promontories, but little falls over the eastern slopes of the 
 mountains and it seldom reaches the highlands. In nearly all 
 cases the coastal region benefits by rain from the southerly half 
 of the following anticyclone. 
 
 CONDUCIVE TO DRYNESS. 
 
 As noted above, the burster is itself conducive to dry ness rather 
 than rain, it being caused, apparently, by the hot, dry conditions 
 prevailing on the plains. The rain by which it is accompanied is 
 caused entirely by electrical disturbances. 
 
 A GRADUAL VEERING OF WIND TO SOUTH WITHOUT A BURSTER. 
 
 This fact may be demonstrated by studying the progress of a 
 southerly change when it is not accompanied by the " burst." As 
 the advance isobars of the approaching anticyclone reach Central 
 Australia, the northern part remains stationary while the southern 
 expands eastward with rapidity. The barometers rise rapidly in 
 Tasmania and the region of high pressure, then extends itself 
 northwards east of this coast, and thus in part surrounds a pocket 
 of low pressure. This low pressure area is then forced north- 
 wards or towards the equator, but not to the eastward, and this 
 temporary stoppage of the usual easterly motion is the precursor 
 of gales, not only in the burster season but at all times in the 
 year. In the course of the further development of this system 
 the low pressure is ultimately forced to north-east off the coast 
 of New South Wales, and the isobars trend from north-west to 
 south-east about the thirty-second or thirty-third parallels of 
 latitude. 
 
SOUTHERLY BURSTERS. 21 
 
 STRAIGHTENING OF ISOBARS AND RAIN. 
 
 At the same time the rear isobars of this retreating high pressure 
 are becoming flatter or straighter, and causing a divergence of the 
 south-east trades which prevail almost constantly north of New 
 South Wales into a north-east wind. The result of this move- 
 ment is an inflow of humid air which penetrates into the southern 
 part of the continent as far as the South Australian border, and 
 sometimes still further west, that is, six to seven hundred miles. 
 
 SOUTHERLY WIND THE PRECIPITATING AGENT. 
 
 As the following high pressure approaches, the southerly or 
 surface precipating current in front of it travels eastward, and 
 meeting this humid atmosphere rain invariably falls inland, the 
 most copious downpour being along the western slopes of the 
 mountains. Ultimately this approaching anticyclone reaches the 
 eastern sea-board, thereby forcing the low pressure off and up the 
 coast, and by this time the wind which was southerly inland has 
 a tendency to veer to east following the isobars, and develops one 
 of the ordinary forms of easterly weather, which although some- 
 times dangerous to mariners by reason of its high winds and 
 stormy seas, is by no means so hazardous as the burster. The 
 wind rises gradually, anrl thus several hours warning is given 
 before the disturbance assumes a serious aspect. 
 
 A change of this character rarely makes itself felt beyond the 
 Queensland border. While the fury of the storm is spending 
 itself, the high pressure continues its forward motion south of the 
 storm, the low pressure fills up, and the storm is over. 
 
 The weather during the forty-eight hours preceding this change 
 may be, and generally is, locally hot and fine, but the develop- 
 ments in the twenty-four hours immediately antecedent to the 
 outbreak are exceedingly rapid. On the day of such a change the 
 weather reports, even from the inland districts, may tell of fine 
 weather, but by evening the disturbance is felt in the far west, 
 travelling with great speed towards the highlands, and during the 
 night, or by daybreak, it reaches the coast lands. 
 
22 AUSTRALIAN WEATHER. 
 
 LOCAL CONDITIONS. 
 
 The weather prevailing locally presents exactly the same features 
 as that which heralds the southerly burster, and unless the observer 
 is furnished with complete data, both from South Australia and 
 Queensland, it is practically impossible to differentiate the con- 
 ditions. 
 
 DIFFICULTY IN FORECASTING. 
 
 Given these data however, it is possible to foretell accurately a 
 disturbance of the kind above described. So far back as the date 
 of the very first of the series of daily weather charts upon which 
 this essay is mainly founded, the conditions which I have 
 endeavoured to describe have produced the same results without 
 exception. I have thus traced the changes by which the conditions 
 promising a " burster " i.e. an approaching anticyclone with its 
 low pressure A are modified locally so that we have no southerly 
 burster but in its place a south-east gale. There are other though 
 less imprtant factors, tending to rob the southerly wind of its 
 velocity. 
 
 RAINFALL IN THE WESTERN COLONIES. 
 
 The first and most important of these is rainfall, only in this 
 instance the precipitation takes place further to the westward. 
 The anticyclone appears on the coast of West Australia accom- 
 panied by severe gales and rain, which have, in a great measure, 
 wasted themselves before the disturbance reaches Bass Straits, 
 and on the coast of New South Wales only the last dying breath 
 of the atmospheric upheaval is felt. The barometric evolutions 
 may be related as follows : In the first place the high pressures 
 both to east and west of the \ depression, are in unseasonable 
 latitudes, the summer tracks lying between 36 and 37 south, and 
 never travelling much below that. At first the adjacent isobars 
 of both high pressures are very full and round ; the col between 
 is narrow, subsequently the conditions become intensified, the 
 isobars straighten, and the A becomes acute. 
 
SOUTHERLY BURSTERS. 23 
 
 EFFECT OF COASTAL FORMATION. 
 
 As they advance, the anticyclones gradually become separated 
 by a wider space and isobars of the depression open out and 
 become round instead of pointed, grow more obtuse in fact, and 
 were it not for the mountains of the east coast intensifying these 
 conditions the wind should reach us as westerly instead of 
 southerly. 
 
 BURSTERS IN SOUTHERN COLONIES. 
 
 While the barometric conditions just referred to are travelling 
 over West and South Australia they produce weather somewhat 
 similar to that experienced here under similar conditions, and 
 hence it is said that the burster is felt in South Australia and 
 Victoria as well as on the eastern coast, but although this is true 
 in a sense, the local conditions on the east coast intensify the 
 characteristics of the "burster " to such a degree that it is very 
 unlike the burster of the southern Colonies, and is in fact quite 
 different. 
 
 GEOGRAPHICAL CONDITIONS IN SOUTHERN COLONIES. 
 
 In South and Western Australia particularly, the atmospheric 
 waves pass over comparatively level country, and the fact that 
 two high pressures are sometimes so close together may be ascribed 
 to accident. But in New South Wales the geographical aspect is 
 widely different. This province is traversed along the east coast 
 by a mountain range, whose pinnacles have a mean altitude 
 of from three to four thousand feet, and this checks the forward 
 motion of the anticyclone until the moment of the mass of high 
 pressure carries it over, but while the power to do this is accumul- 
 ating there is formed in the hollow or vast basin, west of the 
 
 BURSTERS INTENSIFIED ON THE PLAINS. 
 
 highlands, what may be termed a local depression, the sun heating 
 the plains and causing an up draught from the soil, which makes 
 the A assume an intensified character. And it is retained in this 
 situation until the high pressure at its back gathers sufficient 
 force to send it also over the mountains, which it ultimately does 
 
24 AUSTRALIAN WEATHER. 
 
 suddenly, and the southerly wind relieved of the surface friction 
 of the land comes up over the coastal water with a violent rush 
 or " burst." 
 
 DIVERTING THE ANTI-CYCLONES TO SOUTH EAST. 
 
 Under the influence of this accumulated motive power which 
 has placed the burster on the coast, the anticyclone is occasionally 
 diverted from its easterly course to a south-east direction, and 
 instances are recorded where, so far as it is possible to trace its 
 movements it has disappeared travelling almost due south. 
 
 ANOTHER CONDITION MODIFYING BURSTERS. 
 
 There is yet another condition, or set of conditions, which not 
 only modifies the southerly burster but retards it to a considerable 
 extent. As a general rule when there is a A depression lying 
 with its centre approximately south of VVentworth, it may be 
 confidently anticipated that the southerly will reach the sea-board 
 after a period varying from twelve to eighteen hours. 
 
 EFFECT OF ELECTRICAL CONDITIONS. 
 
 But should the front of the \ develop more than the normal 
 electrical activity, it will inevitably be delayed for another 
 day, and in the meantime the barometric conditions advance but 
 slightly. A southerly of this type is almost invariably only one 
 of a series. The local weather features that herald its advent 
 are the same as those which foretell the coming of the ordinary 
 burster, except that the temperature on the day of thunderstorm 
 is higher than that of the day on which the burster reaches us. 
 The isobaric curves accompanying this particular form of dis- 
 turbance are notably disorganised and irregular to the east of the 
 A, while on the high pressure to the west they remain smooth 
 and even. 
 
 TYPES OF LOW PRESSURE. 
 
 Southerlies have been observed to result from three distinct 
 types of low pressure. The first of these is the familiar A depress- 
 ion, resulting in the true southerly burster. This is the one most 
 
.a 
 
 SOUTHERLY BURSTERS. 25 
 
 ' 
 
 commonly experienced, and as a rule, the sharper the A the more 
 sudden is the change. 
 
 THE MONSOONAL DEPRESSION. 
 
 The second variety is the tropical depression or tongue which 
 may be looked upon as an inverted \, which only occurs during 
 the monsoonal season, and even then only on rare occasions ; in 
 fact the tongue has never been observed to exist east of the moun- 
 tains, though west of the range it has been known to reach to the 
 Victorian border. This peculiar type of disturbance is accom- 
 panied by an overcast tropical sky, which almost invariably is 
 found preceding the southerly current. 
 
 BURSTERS WITH HIGH TEMPERATURE. 
 
 The temperature is very high ; thunderstorms are prevalent ; 
 and the southerly itself is not, generally speaking, very strong. 
 Still it is, perhaps, more beneficial than the first description of 
 southerly, as it is attended by rain in the northern parts of 
 Australia, and west of the highlands in the northern parts of New 
 South Wales. 
 
 TROPICAL DEPRESSIONS. 
 
 Should the tropical tongue, when it retreats towards the equator 
 travel east, as occasionally happens, the lower part seems to become 
 detached and forms an active rain cyclone on the south-east coast 
 of Queensland. The floods which took place in that locality early 
 in 1892, as well as those on the northern coast of New South 
 Wales, may be ascribed in part to this cause. This opinion, 
 however, it is necessary to mention, rests upon few data and must 
 be taken subject to correction. 
 
 The apparent parting of the tropical tongue to which I have 
 alluded is by no means the invariable, or even the most common 
 form of development. The low pressure always exists in a fluctu- 
 ating state, north of this continent, the isobars sagging down 
 between the high pressures as they pass along, and drawing up 
 again as the high pressures pass under them. In this way they 
 act upon the low pressure to much the same effect as a succession 
 
26 AUSTRALIAN WEATHER. 
 
 of ocean waves might do, while passing under a long, pliable, 
 floating body set at right angles to them. 
 
 BURSTERS FROM SECONDARIES. 
 
 The third and last modification results from a secondary. 
 Of this particular variety very few have been recorded on the 
 weather charts. They develop so suddenly that unless they are 
 actually in a state of evolution at the hour when observations are 
 taken (9 a.m.), the meteorologist has no indication of their 
 proximity. They develop on the south coast of New South Wales 
 through the formation of a " kink " in the outlying isobars of the 
 retreating high pressure. Prior to such a development the baro- 
 metric conditions are neutral or dormant in other words, the 
 existing high and low pressures have little grade, and are only 
 relatively high and low. Southerlies caused in this way partake 
 of the well known characteristics of secondaries, wherever they 
 may occur that is, violent and of small compass, most severe in 
 the extreme south east, and seldom affecting the coastal districts 
 north of Cape St. George. The depression which follows the 
 retreating high pressure brings on the following day, a southerly 
 of the ordinary character, the strength of which is regulated 
 according to the intensification or otherwise which has taken place 
 during the interval in the adjacent high and low pressures. 
 
 BURSTERS NOT CONDUCIVE TO RAIN. 
 
 The next point to be considered is the relation of bursters to 
 rainfall. The burster, on the whole, must be regarded as unfavour- 
 able to rain, for though its advent may benefit the coastal areas 
 to the extent of a few showers it serves as an indication that the 
 country west of the ranges is at least in a temporarily dry con- 
 dition. A succession of bursters especially denotes intermittent 
 rain in the interior. As a proof of this, and also as an evidence of 
 the comparative powerlessness of merely local weather conditions 
 to directly cause, or even to intensify, the burster, it is only neces- 
 sary to point out its frequent occurrence under circumstances 
 apparently most unfavourable for its development. 
 
SOUTHERLY BURSTERS. 27 
 
 ARRIVES WHEN TEMPERATURES ARE LOW. 
 
 It often arrives during a period of low temperature at Sydney, 
 and also when rain has actually been falling on some occasions 
 when it has been falling heavily. This statement is, of course, 
 not intended to imply that the southerly may not be modified by 
 the existence of such conditions upon the immediate scene of its 
 action. Still, despite any modification which may take place, the 
 wind attains a by no means inconsiderable velocity, some storms 
 recording as high a rate as forty miles per hour. Moreover the 
 rain generally ceases and a period of fine weather follows, proving 
 
 RAINFALL FROM NORTHERLY WIND 
 
 conclusively that, in these particular instances, at all events, the 
 rainfall which takes place on the arrival of the burster is the pre- 
 cipitation from the moist northerly current, and not from the 
 southerly one. Bursters have also been recorded when the local 
 maximum thermometer has registered little over 74 in a mid- 
 summer month, and effecting a diurnal range of only four degrees. 
 At such times the temperatures in the interior are always high, 
 especially to the north-west of New South Wales and in south- 
 west Queensland. 
 
 BURSTERS DESIRABLE. 
 
 Though the individuality of the burster is, in itself, opposed to 
 rainfall, yet the frequent recurrence of this phenomenon is to be 
 desired to cool the atmosphere, and its frequency indicates that 
 the cyclonic systems in the north-east, and the anticyclonic systems 
 in the south have less than the average energy ; also in the early 
 part of the year it denotes the failure of the monsoonal rains. 
 
 SEVERE DROUGHT COINCIDENT WITH THEIR ABSENCE. 
 
 For instance, their absence was very marked during the severe 
 drought in the summer of 1875-76. In seasons when southerlies 
 are rare the loss of the interval of cool weather adds to the severity 
 of the summer. 
 
28 AUSTRALIAN WEATHER. 
 
 MONOTONOUS WEATHER. 
 
 During such droughty periods as that just referred to, the 
 barometrical changes over the whole of Australia are very slight. 
 The temperatures, however, are frequently very high inland, but 
 the diurnal ranges are not great. Locally, in Sydney, during the 
 mid-summer months, the maximum shade temperature has been 
 known to range between 80 and 86 without hot nights, for 
 twelve consecutive days, with persistent north-east winds. Experi- 
 ences of this nature are monotonous for the pastoralist and 
 agriculturist, and equally so for the meteorologist, presenting as 
 they do, no interesting features for observation, and no change of 
 moment, either in weather or temperature can be anticipated until 
 the barometer shows renewed activity. 
 
 CHANGES OP TEMPERATURE. 
 
 Generally the greatest diurnal range of the temperature re- 
 sulting from bursters occurs in October. The mean minimum 
 temperature for this month is 55' 1 so that when the maximum 
 reaches from 85 to 90 and a southerly takes place, the tem- 
 perature in some cases drops as much as 30 to 35 This range 
 is seldom reached in the hotter months, in fact it is only attained 
 when a maximum of about 100 is recorded an event of very 
 rare occurrence. 
 
 A BURSTER OF THE POPULAR TYPE. 
 
 An anticyclone of good energy, and one from which a popular 
 type of southerly burster lasting three days results, has a latitu- 
 dinal axis of approximately two thousand four hundred miles and 
 as it moves at the rate of four hundred miles per diem, 1 it follows 
 that if a vertical line is drawn in front of it, an observer stationed 
 in this line will record three days of southerly weather in the front 
 half of it and three days of northerly winds in the rear, due to the 
 normal circulation about an anticyclone. This is the actual 
 experience : provided no meteorological agencies affect or modify 
 its symmetrically oval form during its passage over the Australian 
 
 1 See Moving Anticyclones, p. 3. 
 
SOUTHERLY BURSTERS. 29 
 
 continent, steady breezes throughout its circulating areas are the 
 rule during the six days it occupies in passing over the compara- 
 tively low lands of Australia, as it travels from west to east. 
 
 ISOBARS FLATTENED BY MOUNTAINS. 
 
 But when the anticyclone reaches the mountains of New South 
 Wales, the symmetry is disturbed by a flattening of the front 
 isobars against the mountains, which concurrently show a tendency 
 to spread northwards, aided probably, by the inclination of the 
 mountain chain towards the north east. 
 
 INTENSIFICATION OF THE LOW PRESSURE BY MOUNTAINS. 
 
 While this change is taking place in the high pressure, the 
 isobars of the \ depression in front of it are undergoing a similar 
 process. But as this depression is a much smaller system, the 
 front isobars are being squeezed against the highlands, while those 
 at the rear are also being compressed by the advancing high 
 pressure. The result is an acceleration of force in the northerly 
 and southerly winds which enclose the \ depression. When the 
 squeezing process has amassed sufficient energy to overcome the 
 obstruction, the gradients of the high pressure, in their efforts to 
 regain their normal condition, expand with great rapidity on the 
 eastern slopes and send the \ depression at a great rate to the 
 south-east. 
 
 For the previous twenty-four hours the northerly currents have 
 been maintained by the \ depression, but when once the moun- 
 tains are passed the anti-cyclone takes up the running of the 
 southerly one. As a proof of this speaking particularly of the 
 central and northern parts of the coast the southerlies have 
 nearly always been found to circulate along the line of the sea- 
 board in the anticyclone isobars. After an anticyclone of this 
 character has entered by a three days' march into Australia, it is 
 very rarely indeed that its progress is checked ; but occasionally, 
 as the A depression still travels on, it seems to drag the southern 
 side of the high pressure after it. At the same time, while the 
 body of the high pressure remains over Australia, the one preced- 
 
30 AUSTRALIAN WEATHER. 
 
 ing continues on its way towards New Zealand. This results in an 
 attenuated state of atmosphere on the northern coast of New 
 South Wales and on the shores of Southern .Queensland. 
 
 TROPICAL CYCLONES. 
 
 If it so happens that one of the north-east cyclones is on a visit 
 to this region at the same time it is diverted from its customary 
 course the track of these storms being generally to the eastward 
 of our coast and avails itself of this partial void. Its irruption, 
 acting with the already existing A depression, seems to result in 
 that peculiar class of storm, of which the Dandenong gale is a 
 notable example. 
 
 Concerning the intensity of this particular storm a few facts 
 concerning another of a similar character may not be considered 
 out of place. A study of its leading characteristics may, in some 
 future time assist the meteorologist in recognising the early signs 
 and portents of another such calamity. 
 
 A RECENT GALE OP THE DANDENONG TYPE. 
 
 The second gale referred to occurred on the 23rd of September, 
 1892 sixteen years after the famous " Dandenong" storm. On 
 this occasion the wind attained a velocity of one hundred and 
 twenty miles an hour. The series of anticyclones in August and 
 those in the early days of September were of considerable energy, 
 and in each case the high pressure to the west, which, from its 
 unusually energetic state, doubtless aggravated the violence of 
 the storm, had completely disappeared on the following days. 
 All that remained were horizontal isobars of a relatively low 
 pressure in the southern ocean. Particulars of the "Dandenong " 
 gale may be found in Mr. Russell's paper on " Storms on the New 
 South Wales Coast," read before the Royal Society of New South 
 Wales on 7th August, 1878. (See following diagram, page 32.) 
 
 BAROMETERS FALL AFTER WIND VEERS TO SOUTH. 
 
 These are two noteworthy instances in which the barometers, 
 after a southerly, have been observed to fall instead of rise, with 
 
SOUTHERLY BURSTERS. 31 
 
 active high pressure in the rear at the commencement of the blow. 
 The gale of 1892 was not a burster, but a gradual change to south 
 which took place some thirty hours previously. In this instance 
 the A depression, which seemed to act in concert with the 
 tropical cyclone, appears to have backed from a situation midway 
 between Australia and New Zealand. 
 
 CONDITIONS WHICH REGULATE THE NUMBER OP BURSTERS. 
 
 Before proceeding further, it may be advisable to offer a sug- 
 gestion as to the causes which seem to the writer to justify the 
 two apparently anomalous statements made in an earlier part of 
 this essay : (1) that bursters are less frequent and also less violent 
 than usual, during those seasons when the interior of the country 
 is suffering from drought ; and (2) that they are also less frequent 
 and of less than their ordinary violence, when the same region is 
 visited by persistent rains with overcast skies. In other words, 
 seasons of drought and desolation and seasons of deluge are both 
 inimical to the existence of bursters. 
 
 The following theory is tentatively submitted, to account for 
 the greater prevalence of bursters during what may be called a 
 moderately dry season. This term is used here to indicate, not 
 so much a season in which the rainfall does not exceed the 
 average, as one in which the number of rainy days is a shade 
 below the average. It will be remembered that strong southerlies 
 have been said to result from energetic anticyclones ; also that 
 energetic anticyclones bring with them possibilities of great 
 extremes in our summer temperatures. But since in summer 
 their sphere of action is in high latitudes the weather of the 
 northern parts of Australia is controlled chiefly by tropical 
 depressions, and consequently great heat prevails there as a rule. 
 Now if, after a period of hot weather, rain were suddenly pre- 
 cipitated on this northern area, the consequence would be a great 
 uprush of air with the vapour produced by the rain falling on 
 heated ground, and a consequent inrush of air from southern 
 areas. 
 
Cf 
 
SOUTHERLY BURSTERS. 33 
 
 RECURRENCE AND DURATION OF BURSTERS. 
 
 Bursters are not only uncertain in their duration, but the periods 
 of their recurrence are extremely erratic. Two have been fre- 
 quently known to take place within twenty-four hours ; between 
 two others an interval of a month has been known to elapse. 
 The shortest one recorded extended over a period of three hours ; 
 the longest covered the space of ten days. These figures of course, 
 embrace all the variations of the wind during the continuance of 
 the storm. 
 
 SHORT-LIVED BURSTERS. 
 
 The short lived bursters are generally experienced during a 
 seasonal prevalence of north-east winds those of longer duration 
 are usually met with during a southerly prevalence. 
 
 PARTIAL TO COAST OF NEW SOUTH WALES. 
 
 Not only during the summer, but in all seasons of the year, the 
 southerly has a remarkable partiality for the sea-board of New 
 South Wales. 
 
 OVER STEP ISOBARS. 
 
 So much is this the case in the hot months that the approaching 
 currents from the south have been observed on the coast when the 
 isobars indicated northerly winds. 
 
 BLOW CONTRARY TO ISOBARS. 
 
 Sometimes, too, for several consecutive days, the wind has blown 
 with appreciable velocity from this quarter when the isobaric 
 lines from their relation to the centre of the anticyclone should 
 be producing northerly winds. This latter pecularity is most 
 noticeable in winter months and with necessarily small gradients. 
 
 THE AREA OVER WHICH THE BURSTER EXTENDS. 
 
 The land area over which the burster exercises its influence may 
 be generally described as including all the country east of the 
 mountains in New South Wales, from the extreme south-east 
 point of the colony to a little above Port Macquarie on the north 
 coast. It occasionally oversteps the northern limit, but when it 
 
34 AUSTRALIAN WEATHER. 
 
 does so it is usually the result of a cyclonic disturbance on the 
 north coast, such as existed, for example, in the " Dandenong " 
 burster, the effects of which were felt considerably north of 
 Brisbane. The northern boundary of the burster may be defined, 
 as limited by the south-east trade winds, which blow almost 
 incessantly north from about 30 S. latitude. Other areas of 
 actual experiences are given on page 40. 
 
 CHANGE OF TEMPERATURE. 
 
 Bursters always result in a diurnal drop in temperature, the 
 said drop ranging between 37*5 and 4' 2 with an average fall of 
 18'1. The greatest diminution of temperature takes place during 
 the first hour, and the fall is most sudden when the burst comes 
 at midday. The steepest drop on record was that registered at 
 1*50 p.m. on the 30th December, 1891. The maximum then read 
 97*5 ; five minutes after the change to south it read 80 f 3, being 
 a difference of 17*2 ; two minutes later there was a fall to 7 9 '7 
 or a further difference of 0'6. The thermometer remained at 
 this reading until after two o'clock, but at 2 '15 it descended to 
 75, being a drop of 4'7 more and a total drop of 22'5 in twenty- 
 five minutes. The diurnal range in this instance was 23 '8. 
 
 VELOCITY OF WIND IN THE BURSTEfe. 
 
 It occasionally, though rarely, happens that the greatest velocity 
 of a burster is reached at the change or during the first hour 
 thereafter. In most cases, however, th'e maximum force is attained 
 about twelve hours afterwards. As the majority of bursters occur 
 between the hours of seven and twelve p.m. (see Table I.), it there- 
 fore follows that the wind is usually strongest between seven and 
 twelve a.m. on the following day. 
 
 UNSEASONABLE WINDS. 
 
 In these instances where the greatest velocity was reached im-. 
 mediately on the arrival of the burster, the winds preceding were 
 generally of an unseasonable character that is to say north- 
 westerly in the summer and north-easterly in the spring and 
 
SOUTHERLY BURSTERS. 35 
 
 autumn months. Under these circumstances the burster, though 
 strong, was generally of brief duration. 
 
 THE PAMPERO. 
 
 The opinion has long been held that there is a close analogy 
 between the southerly burster of Australia and the pampero of 
 South America. The writer therefore sent a detailed description 
 of the burster to a friend formerly resident in Brazil, and requested 
 him to note the points of similarity between it and the pampero. 
 The following is a copy of his reply which is based, as will be seen, 
 partly on personal observation and partly on information gathered 
 by him from sources on which he places reliance : 
 
 " In answer to your letter of 14th of April, I am afraid I can 
 afford you very little information on the subject of the South 
 American pamperos from personal observation. They never 
 reached, on the coast, as far north as Rio de Janeiro which was 
 my usual place of residence, but I twice experienced their violence 
 when on my exploring expedition across Brazil in 1872-3. The 
 first time was in September, 1872, when we were camped on the 
 banks of the Paraguay River as far north as 20 south. On the 
 particular occasion we were awakened in the night by a roaring 
 sound which rapidly drew nearer and burst upon us overturning 
 tents and everything else that offered much resistance without 
 being stable. My hammock was slung to a branch of a big tree, 
 which was torn off, but did me no damage as I had turned out to 
 save the tents. It had all blown over by the morning. On the 
 second occasion, in October, 1873, I was surveying the Alto 
 Parana River, about 21 south, when my Indians noticed the 
 clouds gathering in south, and made at once for the lee of the 
 islands in their canoes. The river at this part is due north and 
 south for many miles, and from one to four miles wide, and as the 
 wind came from due south it beat up a great sea, the waves being 
 quite three feet from crest to bottom. I sailed up before it, and 
 was able to go up rapids that, on ordinary occasions could only 
 be negotiated by poling and ropes. It did not last long, and on 
 both these occasions was, I imagine, a more violent effort than 
 
36 AUSTRALIAN WEATHER. 
 
 usual of a pampero which found its way so far north owing to 
 the immense width and openness of the river Plate Valley, of 
 which river both the Paraguay and Parana are confluents. The 
 pamperos are mostly felt at the mouth of the River Plate, where 
 by their violence they often cause considerable damage to shipping. 
 If they are not recorded further south, it is because there are no 
 ports on the east coast nor inhabitants or towns where their 
 recurrence could be noted. 
 
 " In the southern States of Brazil, they are frequently felt, and 
 a friend of mine who has just arrived from Rio Grande du Sul, 
 the most southerly state of Brazil, tells me they occur mostly in 
 the winter months (my friend was asked to verify this statement 
 and his reply will be found further on. H.A.H.), from April to 
 November. He says they are preceded always by three days 
 rain, which, when over, is immediately followed by the pampero, 
 coming from what he called the south-south-west or nearly due 
 south. They last then for nearly three days and blow with great 
 violence and low temperature. My informant goes on to say, 
 from his own experience, that the pampero does not get so far 
 north on the coast, although, as I have told you, in the interior 
 it does seem to penetrate. There are high ranges of mountains 
 south of Rio de Janeiro which would, I have no doubt, have the 
 effect of breaking up and deflecting any part of the pampero 
 which might go there." 
 
 The friend's account is as follows : " Pamperos proper occur in 
 the winter, say from the end of May till October, and generally 
 last three days. Occasionally in the summer time we have, in 
 Rio Grande, smart breezes from the south-west after rain, but 
 they do not last long, and although they come from the same 
 quarter and cool the atmosphere wonderfully, are not called 
 pamperos. While the pampero is blowing the sky is beautifully 
 clear and cloudless. Occasionally in the summer in Rio Grande 
 when the weather is very sultry, it breaks by a squall suddenly 
 setting up from the west by south-west, and though the sky was 
 clear ten minutes before, you can see the cloud roll forming in 
 
SOUTHERLY BURSTERS. 37 
 
 the west as the cool breeze advances, and down the rain comes in 
 torrents, and in a few hours all, wind and rain, is over. Thunder 
 may or may not accompany the rain, which almost always precedes 
 (at least in Rio Grande) pamperos. I have seen pamperos as 
 strong after rain, when there was no thunder accompanying, as 
 when there was. Our thunder, too, is generally, in the summer 
 and rarely in winter. I must add that Rio Grande being further 
 north than the Plate, we do not get the pamperos with the full 
 force experienced in the Plate region. With us they are a steady 
 continuous blow for three days, varying little in force till the third 
 day, when they are felt to be grudually declining. They always 
 blow from the same quarter, the south-west, and are cold and dry." 
 
 The most interesting fact evidenced in this description of the 
 pampero is that it always follows rain. This would seem to imply 
 that the evaporation arising from the plains is one of the immediate 
 causes of its existence. If this is the case it lends support to the 
 theory, hereinbefore submitted, that the vapour arising after rain 
 from the vast heated areas of the Australian interior is at least 
 one of the agencies from which the burster has its origin. The 
 difference between the two sets of circumstances is that, in South 
 America, the cause and effect act on the same region, and there- 
 fore one follows the other with no appreciable interval, while in 
 Australia the seat of the effect is removed by something like one 
 thousand miles from that of the cause, and consequently the con- 
 nection is more difficult to trace. 
 
 Winter northers of Texas are a somewhat similar experience. 
 The following is a short extract : "The northers prevail from 
 November to March, and commence with thermometer at 80 or 
 85. A calm ensues on the coast ; black clouds roll up from the 
 north ; the wind is heard several minutes before it is felt ; the 
 thermometers begin to fall ; the cold northers burst upon the 
 people bringing the thermometers down to 28 and sometimes even 
 to 25, men and cattle being killed from the severe cold." This is 
 the only description I have come across of similar changes in the 
 northern hemisphere. l 
 
 1 Physical Geography of the Sea by Maury, pp. 93-94. 
 
38 AUSTRALIAN WEATHER. 
 
 SIMILARITY OF WEATHER ON EAST AUSTRALIA AND SOUTH AMERICA. 
 
 Before finally leaving the subject of the Pampero, I may mention 
 that many old sailing ship captains recognise numerous points of 
 similarity between the southerly bursts upon the east coast of 
 South America and those upon the east coast of Australia, and 
 make the same preparations to meet either case. 
 
 ADMIRAL FITZROY AND OTHERS ON THE PAMPERO AND BRICKFIELDER. 
 
 Perhaps the earliest reference to the pampero appears in the 
 "Weather Book" of Admiral Fitzroy, pp. 150 and 151. In 
 chapter xxi., he gives thrilling accounts of two occasions when 
 his vessel was struck by it and nearly foundered. The same 
 author, in the same book, speaks of the brickfielder (southerly 
 burster) of New South Wales, page 263. A more exhaustive 
 account of the pampero in its strictly scientific aspect, appears in 
 " Weather," by the Hon. Ralph Abercromby, page 263, and a 
 description will be found in the Scottish Meteorological Journal, 
 Vol. v., No. lx., p. 335. 
 
 A DESCRIPTION OF A BURSTER WRITTEN EIGHTEEN YEARS SINCE. 
 
 The following description of a southerly burster is taken from 
 " The Climate of New South Wales," by Mr. H. C. Russell, B.A., 
 The passage in question contains the best description of it that I 
 have met with, and it is the more interesting as being one of the 
 first accounts published by any scientific observer of this interest- 
 ing phenomenon. It is as follows : " If in fine north-east hot 
 weather the barometer falls fast in the forenoon, a southerly wind 
 (burster) may be expected before night ; if the day is very hot 
 the change will come sooner ; and if the barometer is falling very 
 fast and clouds be seen in the west, a thunderstorm may be 
 expected in the afternoon. Sometimes the thunderstorm bursts 
 first, and the wind sets in from south afterwards ; if only the storm 
 comes it will probably be hot again next day. 
 
 "Southerly bursters are generally to be expected from November 
 to the end of February ; they are always attended with strong 
 electrical excitement, a stream of sparks being sometimes produced 
 
SOUTHERLY BURSTERS. 39 
 
 for an hour at the electrometer at the ends of the exploring wire. 
 The approach of the true burster is indicated by a peculiar roll of 
 clouds, which when once seen cannot be mistaken ; it is just above 
 the south horizon, and extends on either side of it 15 or 20, and 
 looks as if a thin sheet of cloud were being rolled up like a scroll 
 by the advancing wind. 
 
 "Cloudsof dust, which penetrate every where, announce the arrival 
 of the wind, scud flies by overhead with great rapidity, being 
 sometimes less than two thousand feet high ; rain may follow, but 
 if so, thunder and lightning come first. The velocity of wind is, 
 in most cases, greatest within the first two hours, and varies from 
 thirty to seventy miles per hour, but is usually from fifty to sixty 
 and the rate of progress along the coast about forty miles per hour. 
 The change of wind is sometimes very sudden ; it may be fresh 
 north-east, and in ten minutes a gale from south, hence vessels 
 not on the lookout are sometimes caught unprepared, and suffer 
 accordingly. When the wind is light these storms are often 
 carried to sea by the general easterly motion of the atmosphere, 
 and may be seen passing by, the peculiar clouds indicating unmis- 
 takably their position." 
 
 APPARENT DOUBLE BURSTERS. 
 
 The following is a detailed account of a double burster, or of 
 two distinct simultaneous bursts on different parts of the coast. 
 It will be convenient to commence with a table, showing the 
 hours at which the burst reached various points on the coast-line, 
 with relative rates of travel from place to place : 
 
 10th October, 1893. 
 
 Rate of Travel. 
 
 Stations. Distances. Hour. Interval. -..-., 
 
 Miles. 
 
 Eden 
 jYToruya 
 
 79 
 
 2 a.m. S 
 8'30 6 30 12-2 ft 
 
 Jervis Bay 
 Wollongong 
 Sydney 
 Newcastle 
 
 Port Macquarie... 
 Clarence 
 
 47 
 41 
 41 
 62 
 
 110 
 122 
 
 10-36 2 6 22-4 j> 
 11-45 1 9 32-8 'g'a ^ 
 1 p.m. 1 15 32-8 13 10 
 3-40 2 40 23-3 " J 
 Mean rate of travel 25 -3 a 
 2 p.m. g r^ 
 5-30 3 30 34-9 
 
 V '*-~^ " 
 y> 01 THX ^^\ 
 
 5I-FBRSIT7 
 
40 AUSTRALIAN WEATHER. 
 
 From this table it will be seen that the wind changed to south 
 at Eden, the southern point of the coast at 2 a.m., and made its 
 way gradually up the coast, reaching Sydney at 1 p.m. and New- 
 castle at 3 '40 p.m. in the mean time that is at 2 p.m. Port 
 Macquarie, one hundred and ten miles north of Newcastle, reports 
 the change at 2 p.m., or one hour and forty minutes before it got 
 to Newcastle, or to put it another way, when one burster reached 
 a point on the coast twenty miles north of Sydney, another made 
 itself felt at Port Macquarie, one hundred and ten miles farther 
 north, it appears then that there were two bursters separated by 
 one hundred and ninety miles and simultaneously making their 
 way along the coast. It is probable, however, that the southerly 
 burster is a line storm, and that the change of wind occurs along 
 a certain isobar, and if this be the true explanation it is entirely 
 in accordance with what has already (page 29) been shown as to 
 the trend of the isobars in certain cases. When the approaching 
 anticyclone flattens its isobars against our dividing range, and 
 then taking advantage of the easier western slope in the northern 
 districts, and the lower altitude of the mountains generally, to 
 the west of Port Macquarie, protrudes some of the isobars over 
 the mountains, so that they come to the eastward at that point 
 like a nose or easterly extension of the isobars, while the southern 
 half of the anti-cyclone is retarded. This feature was noticeable 
 upon the occasion in question, and, as is invariably the case under 
 such circumstances, thunderstorms were prevalent. These, I think, 
 may be safely accepted as the causes which give rise to the 
 phenomenon of two simultaneous, yet entirely independent, 
 bursters. 
 
 In speaking of the areas over which the burster exercises its 
 coastal influence, mention has been made of the fact that it is rarely 
 met with north of Port Macquarie. The present case was not, 
 in reality, any exception to the general rule ; the burster did not 
 reach Clarence Heads as such, and the change there was simply a 
 veering of the wind from north-east through west to south-east, 
 accompanied by a thunderstorm and hail. The downfall of hail 
 
SOUTHERLY BURSTERS. 41 
 
 was heavy and some of the stones were very large. This storm 
 lasted on the north coast until 7 p.m., and was also very severe 
 in the northern highlands. It was accompanied by a rainfall of 
 from one to three inches. 
 
 DESCRIPTION OF THE DOUBLE BURSTER. 
 
 The notes taken of this burster the second of the season are 
 as follows : During the 7th, 8th and 9th October the barometers 
 all over the continent showed little or no grade. The centres of 
 the high pressures were situated, one over the Tasman Sea, and 
 the other off the west coast of West Australia, with a shallow 
 trough of low pressure between. On the 10th the western high 
 pressure had intensified and had made much progress, compressing 
 the low pressure into a sharp y\, with its axis lying in a north- 
 west and south-east direction, or from Bourke to a little east of 
 Gabo. Strong north to north-west winds were experienced on 
 the western borders of Queensland and in New South Wales 
 north-east, while exceptionally hot weather was reported from the 
 Queensland inland stations. In Sydney, for three days previous 
 to the arrival of the burster, the sky was hazy and almost tropical 
 in its aspect. The barometer at Sydney fell three-tenths during 
 the twenty-four hours immediately preceding the burst, and rose 
 rather sharply after it had passed, the lowest point making half- 
 an-inch in twenty-four hours. The temperature was moderate 
 immediately before the change, the thermometer stood at 75, and 
 at 2 p.m., rather less than one hour after, it had fallen to 65. 
 A roll of cumulus cloud of a rather undefined character was first 
 seen at 12-30 p.m., with a line of ragged cirrus beneath. The 
 latter, as the storm advanced, rose in front of and obscured the 
 cumulus. Five minutes later, cirrus were moving horizontally, 
 vertically, and in every other direction from a point in the cloud 
 lying due south. The change of wind at Sydney came at 1 p.m., 
 and was attended by much dust, and some rain which was entirely 
 owing to electrical influences, it was generally light from the 
 western slopes to the coast. 
 
42 AUSTRALIAN WEATHER. 
 
 Passing from this record of the double burster, it is now neces- 
 sary to consider various facts relating to the time taken by 
 bursters in travelling along the coast. 
 
 RATE OP TRANSLATION OF BURSTERS. 
 
 .From the preceding remarks and the figures quoted in the table 
 page 39, as well as other figures appended to page 43, it will appear 
 that no definite relation can be traced between the rate of trans- 
 lation and velocity of the wind in bursters. As already demon- 
 strated, it is quite possible for two bursts to occur on different 
 parts of the coast at the same time, and it is also possible for a 
 
 BURST SIMULTANEOUS OVER A WIDE AREA. 
 
 burst to be felt at the same moment over an extensive area. (See 
 table 28th November, page 43. Since there is no visible con- 
 nection between the velocity of the wind and the ratio of transla- 
 tion of the burst itself, it may throw light on the matter if we 
 look for some explanation of the fact that the southerly change is 
 generally first experienced on the south coast of the Colony. 
 
 BURSTERS FELT FIRST SOUTH OF SYDNEY. 
 
 The most probable explanation is as follows : The high pressure 
 following the /\ depression in many cases moves faster over 
 Victoria than it does over this Colony, and thus forces the lower 
 part of the A to the east making the axis of it more or less 
 towards south-east and north-west. Were it not for this swinging 
 of the A depression, the wind, from its natural inclination to the 
 centre of a low pressure, would be south-west instead of due south. 
 
 CONDITIONS GOVERNING THE RATE OF TRANSLATION. 
 
 The southerly isobars of the A depression usually reach the 
 south coast first, hence it follows that the burster touches the 
 coastal stations in rotation, beginning at the south. The rate at 
 which it travels in its northerly course is decided, first by the 
 inclination of the axis of the A to the trend of the coast, and 
 secondly, by the then prevailing rate of the general atmospheric 
 motion to the eastward. 
 
SOUTHERLY BURSTERS. 
 
 43 
 
 DIFFICULTY IN FORECASTING TIME OF BURSTER'S ARRIVAL. 
 
 These few remarks will show how many and how serious are the 
 difficulties to be contended with in predicting the moment at 
 which a burst will arrive at any given point for up to the present 
 moment it has been found impossible to decide how far the axis of 
 the A may deviate from its vertical position before reaching the 
 coast, or whether the rate of atmospheric motion may dimmish, 
 maintain or increase. 
 
 TABULAE STATEMENT OF THE PEOGEESS OF SOUTHEELY 
 BUESTEES UP THE COAST. 
 
 13th November, 1893. 
 
 Distances. Interval. Translate. 
 
 Stations. 
 Moruya 
 
 Time. 
 Miles. 
 4'45 p.m. 
 
 ] 
 b. m. 
 
 late ^ h 
 
 Miles. 
 
 r. 
 Velocity of 
 
 Jervis Bay 
 Wollongong 
 Sydney .. 
 
 5-30 p.m. 47 
 7 p.m. 88 
 
 45 
 1 30 
 
 62-7 
 
 587 
 
 Wind at Syd- 
 ney 36 m. in 
 short gusts 
 
 Newcastle 
 
 Jervis^Bay 
 Sydney . , 
 
 11 a.m. 62 
 
 16th October, 
 2-30 p.m. 
 9-50 p.m. 88 
 
 1893. 
 7 20 
 
 12 
 
 42 miles. 
 
 Velocity of 
 wind at Syd- 
 
 Pt. Macquarie (Oct. 17) 
 Moruya 
 
 8-30 a.m. 172 
 
 28th November 
 8-10 a.m. 
 
 19 20 
 , 1893. 
 
 9 
 
 ney 20 miles. 
 
 Jervis Bay 
 Wollongong 
 Sydney 
 
 7-30 p.m. 47 
 7 p.m. 41 
 7 p.m 41 
 
 
 " 
 
 Velocity of 
 wind at Syd- 
 ney 48 miles 
 
 Newcastle 
 Port Macquarie 
 
 Moruya . 
 
 10 p.m. 62 
 2 a.m. 110 
 
 30th November, 
 2 -40 p.m. 
 
 3 
 4 
 
 , 1893. 
 
 207 
 23-3 
 
 
 Jervis Bay 
 Wollongong 
 Sydney 
 Newcastle 
 
 Moruya 
 Jervis Bay 
 
 Sydney 
 
 4'15 p.m. 47 
 6-15 p.m. 41 
 7'15 p.m. 41 
 9'30 p.m. 62 
 
 6th December, 
 1 p.m. 
 3-30 p.m. 79 
 6 p.m. 82 
 
 1 35 
 2 
 1 
 2 15 
 
 1893. 
 
 2 30 
 2 30 
 
 29-7 
 20-5 
 41-0 
 27'5 
 
 31-6 
 32-9 
 
 Velocity of 
 wind at Syd- 
 ney 26 miles. 
 
 Velocity of 
 wind at Syd- 
 
 Newcastle 
 Port Macquarie 
 
 Jervis Bay 
 Sydney 
 
 Moruya ... 
 
 10 p. UK 62 
 2-30 a.m. 110 
 
 12th December, 
 11 p.m. 
 4-30 a.m. 82 
 
 15th February, 
 9 a.m. 
 
 4 
 4 30 
 
 1893. 
 5 30 
 1894. 
 
 15-5 
 
 24-4 
 
 15 
 
 ney 40 miles. 
 
 Velocity of 
 wind at Syd- 
 ney 25 miles. 
 
 Velocity of 
 
 Cape George 1p.m. 79 40 
 
 Sydney 5'55 p.m. 82 4 55 
 
 19 8 wind at Syd- 
 167 ney 21 miles. 
 
44 AUSTRALIAN WEATHER. 
 
 CHARACTER OF WINDS PRECEDING A BURSTER. 
 
 When the wind preceding a burster blows from some point 
 between north-west and west, it is always drier and hotter than 
 when it proceeds from between north-east and east. The latter, 
 however, produces peculiarly uncomfortable and relaxing effects, 
 such as might be experienced by one who went into a steam- 
 bath. The north-westerly wind on the other hand is of a dry 
 and parching character, but apart from its occasionally irritating 
 effect upon the nostrils, and the stinging sensation which it some- 
 times causes in the eyes, it is endurable, and is enjoyed by many 
 persons and much preferable to the sweltering north-easter. 
 
 As would naturally be imagined, when the northerly wind is 
 strong the sontherly by which it is succeeded is strong likewise, 
 being effects arising from the same cause. This intimate rela- 
 tion between the force of the two currents is especially noticeable 
 should the burst occur while the northerly wind is in full force 
 in other words, should it occur in the daytime, for northerly winds, 
 and particularly those from the north-east, moderate considerably 
 by sunset. 
 
 RELATION OF VELOCITY OF NORTH AND SOUTH WINDS. 
 
 But while it may be said that, as an almost invariable rule, 
 strong southerlies follow strong northerlies, the mean velocity of 
 the former is generally less than that of the latter, by about ten 
 per cent. 
 
 WITH STRONG WINDS THE BAROMETER RISES SLOWLY. 
 
 And it may be noted, as a peculiar circumstance in a gale with 
 such antecedents, the barometer rises very slightly after the 
 first short and rapid rise, the accumulation of pressure in the 
 west, no matter how steep the gradient is, seems to be used up in 
 maintaining the gale on the coast, and along the coast the 
 barometers rise little or nothing. There are remarkable records 
 of cases in which steady high barometic readings have been the 
 precursors of prolonged southerly gales upon the coast, even the 
 diurnal fluctuations being totally ignored. On the other hand 
 
SOUTHERLY BURSTERS. 45 
 
 cases are on record in which at Sydney the barometer has fallen 
 0-5 in. in twenty-four hours, and in Tasmania as much as 0'8 in. 
 these differences expressing the ratio of the changes in the same 
 burster at the two places, due to latitude. 
 
 DESCRIPTIONS OF PARTICULAR BURSTERS. 
 
 The past twelve months have, unfortunately, been a most 
 unsuitable season for the study of the subject with which this 
 essay is concerned, and it has been especially unsuitable for the 
 observation of cloud movements. The bursters during the past 
 year have not only been few in number, forty per cent, below the 
 average, but also unusually mild in character, and hence it has 
 been impossible to find a burster of strongly marked character 
 for study, of the comparatively small number however, the best 
 have been chosen. 
 
 The first of these occurred on February 15, 1894, it has been 
 selected mainly because the clouds admitted of better notes than 
 others, and because I am able, through the courtesy of Mr. Russell 
 to submit photographs of it from the Observatory records. At 
 9 a.m. on the 14th an anticyclone was over Perth in West 
 Australia, another of rather more decided character was over 
 the Tasman. Sea, and between these there was a low pressure 
 wanting in intensity ; winds were light and without character, 
 and the weather generally hot and unpleasant, but fine, except 
 some passing showers on the coast of New South Wales. Fresh 
 northerly winds were blowing in Bass's Straits. Diagram No. 1 
 shows the instrumental conditions of this change from 7 p.m. of the 
 14th to 7 p.m. of the 16th. It will be observed that the barometer 
 began to fall steadily at 10 p.m. of the 14th, and reached the 
 lowest point at 2 p.m. of the 15th, two and a-half hours before 
 the burster arrived, at the time of lowest barometer the wind was 
 from east, and attained its greatest velocity for the day, seventeen 
 miles per hour, at 3 -30 p.m. Temperature was highest 79, 
 between 1 and 2 p.m. of the 15th, and the coming of the southerly 
 at a quarter to six p.m. made but little difference ; light rain 
 
46 
 
 AUSTRALIAN WEATHER. 
 
 began to fall two hours after mimimum temperature was reached, 
 and became gradually heavier from between 2 and 3 p.m. of 
 the 16th, at which time the wind veered a little to west of south 
 with a considerable increase of velocity. 
 
SOUTHERLY BURSTERS. 47 
 
 On the morning of the 15th February weather chart No. 1 gives 
 the barometic conditions over Australia and New Zealand, con- 
 ditions which had materially hardened since the 14th, winds 
 generally were fresher and conforming to the isobars, and weather 
 fine and warm in the eastern colonies. Up to noon there were 
 light cirro-cumulus clouds in the south-west, at 1 p.m. they had 
 become much more dense and somewhat thundery looking, and 
 seemed to be working round the horizon to the north, with one 
 remarkable mass of cumulus. The following diagram of this date 
 shows the instrumental changes during this southerly. 
 
 At 2 p.m. south-west clouds slightly advanced, at the same time 
 those low down obscured by haze or dust : up to 2 p.m. north-east, 
 east, and south-east horizons beautifully clear. 2'40 p.m., clouds 
 still working further round to due north, with a few light cirrus 
 forming and evaporating east of this point, and a general drift to 
 the east, then thunder cumulus worked up from south-west very 
 slowly, the advanced cirrus reaching Sydney at 3-30 p.m. ; no 
 trace of cloud observable with surface wind up to this time ; a 
 roll of cloud appeared- at 3*15 p.m. above south horizon (See 
 Plate i, taken at 3-43 p.m.), which seemed to be made up of 
 a band of stratus surmounted by cumulus 30 in length ; at 4 
 p.m. the roll merged into the general clouds, which were then 
 fringed overhead and stratified on the southern boundary, beyond 
 which the sky was clear for a space of about 10, at the same 
 moment an uneven roll of heavy cumulus began to rise above the 
 horizon ; as it lifted, the sky still further beyond was clear. Up to 
 this hour, 4 p.m., only one strata of cloud visible, and that moving 
 from south-west ; a very dense and extensive cloud of smoke to 
 south-west, which worked backwards and forwards between north- 
 west and south-east on the horizon. At 4'20 p.m. the eastern 
 point of the roll was immediately over Botany Bay, the western 
 limit extending indefinitely to the west. (See the details in Plate 2, 
 taken at this time.) Small shower at 5-15 p.m. The cloud roll 
 then seemed to melt away like the earlier one, and the upper 
 clouds still moving from south-west became very wild looking, 
 
O 
 
 CE 
 
 a 
 
 1 
 
SOUTHERLY BURSTERS. 49 
 
 and Plate 3 was taken at 5-50 p.m., showing only a trace of the 
 cloud roll and the disturbed looking upper clouds. Southerly 
 burster arrived at 3 -55 p.m., but depth of the current very shallow, 
 for the clouds were still maintaining their course from the south- 
 west. The cumulus observed earlier in the evening, at 6 p.m. 
 remained stationary until 7 -30 p.m., with general outline unchanged, 
 but with variations in facial aspects meanwhile rising to higher 
 altitudes. At 7*15 p.m. a fine cluster of festoon clouds developed 
 on a prominent outstanding cloud to the west, As the evening 
 became cooler a few light cirro-cumulus clouds came up from the 
 south at 7-35 p.m., the cumulus of the upper strata first dissolved 
 into cirrus when the influence of the southerly reached them, 
 and then quickly dispersed before the wind ; lightning visible 
 occasionally to the north until 8-30 p.m. 9 p.m., cirro-cumulus 
 and cumulus extent 4; 10 p.m., overcast nimbus, raining lightly 
 and intermittently until 3 a.m. on 16th, when it came down 
 heavily in Sydney ; to south of city it started to rain heavily 
 three quarters of an hour before, and at 2*45 a.m. a heavy clap of 
 thunder was heard with continued rumblings. The rain con- 
 tinued heavy for another hour, when it tailed off. 
 
 February 16th A heavy shower at 9 a.m., light showers 
 during rest of day. 
 
 The weather chart of the morning of the 16th February shews 
 that in the previous twenty-four hours the whole storm system 
 had moved about five hundred miles to the east (see weather 
 chart No. 2), the front isobars of the anticyclone overlapping the 
 coast line and the low pressure between Tasmania and the Bluff. 
 The isobars are numerically of the same value as on the 15th, but 
 they have spread out and lost energy, weather generally cooler, 
 with rain in Victoria and on western slopes of the main range in 
 New South Wales. 
 
 The second burster to which I shall refer, reached Sydney at 
 2-15 a.m., February 22nd, 1894, weather chart No. 3 and the 
 accompanying diagram of this period show a great deal more 
 

 "t 
 
 s- 
 
 ".'*>*/ "fit 
 
 11 
 
 1: 
 
 3 
 
 X 
 
 / 
 
 V 
 
 
 
 AL 
 
 I 
 
 
SOUTHERLY BURSTERS. 51 
 
 energy in weather conditions generally than there was on February 
 15th. The isobars enclosing the approaching anticyclone are 
 more numerous and of higher value. The wind circulation is 
 more regular and moderate to fresh in force generally. The 
 tropical low pressure is also a denned and active feature. The 
 A depression it will be observed on the 22nd was ill-formed, and 
 hence notwithstanding the anticyclonic feature favouring a good 
 blow, this burster was little better than that of the 15th February. 
 
 The weather antecedent to this southerly was comparatively 
 speaking cooler in central parts of Australia than on the coast, 
 there was also much thunder cloud, and scattered showers fell 
 in south-west Queensland and northern parts of South Australia 
 on the 21st, these points may in part account for the want of force 
 in this burster. The preceding north-east wind blew for the 
 greater part of three days with a force varying between five 
 and fifteen miles per hour, and at 1 a.m. on the 22nd immediately 
 before the burster it was practically calm. This burster began 
 with a velocity of seventeen miles per hour ; between 4 and 11 a.m. 
 the hourly force pulsated between eleven and fifteen miles, it then 
 again rose and twenty -seven miles per hour was reached at 3 a.m., 
 and at 3 -15 a.m. thirty-eight miles the greatest velocity ; it then 
 gradually fell away and veered through east to north early on the 
 26th. This may be taken as a modified type of the popular 
 burster. (See page 28). 
 
 Cloud notes on burster of 2 2nd February, 1894 Warm swelter- 
 ing day on 21st, excepting a small patch of cumulus visible at 3 
 p.m., not a cloud was seen up to midday 22nd. At daylight on 
 the 23rd cirro-cumulus to extent of -3 moving from the south, 
 upper strata very light mackerel with just a perceptible motion 
 from west-north-west; clouds gradually extending up to 11 a.m., 
 when it became completely overcast and a light misty rain began 
 to fall at 11-40 a.m. Thunder at 12-50 p.m.; thunderstorm 
 between 1 and 2 p.m., clouds breaking at 3 p.m. to south. An 
 upper strata of cirrus visible to the south-west, but too distant to 
 note motion ; cleared by evening. 
 
52 AUSTRALIAN WEATHER. 
 
 On November 13, 1893, came a burster of little intensity or 
 interest save from the very remarkable cloud roll, the finest that 
 has been seen in Sydney for some years. The weather conditions 
 at 9 a.m. of November 13 are well shown in weather chart No. 5. 
 An anticyclone rested over Western Australia, with its centre 
 about latitude 30 south, and longitude about seven hundred 
 miles west of Adelaide, while in front of it is a well marked A 
 depression, with its axis north-north-west and south-south-east, 
 and lying centrally over Victoria and the western districts of 
 New South Wales, about three hundred miles west of Sydney. 
 The whole system was moving very rapidly, and from the position 
 of the centre of the high pressure at 9 a.m. of the 14th (weather 
 chart No. 6) the forward motion in twenty-four hours is seen to 
 be fully eight hundred miles. About the A there was a well 
 defined circulation of fresh winds, north-west and south-west, 
 rain had fallen in the northern districts of South Australia and 
 Queensland, and during the forenoon of 13th there were passing 
 thunder squalls and half formed festoon clouds, which were in 
 rapid motion and had a mild disturbed appearance. All day the 
 upper strata of cloud was coming from north-west to west, during 
 the forenoon temperature in the shade rose to 82. At 3 p.m. 
 seven-tenths of the sky was obscured by cumulus, cirro-cumulus, 
 cirrus, and stratus. At 4 p.m. they were much the same, except 
 that in the north the cumulus clouds were bolder and looked like 
 a wall of rocks, snow-white in colour, with horizontal seams or 
 joints. At 5 p.m. thunder clouds seemed to be in all directions. 
 At 6 p.m. a southerly roll could be seen low down in south-west, 
 with much lightning playing about over it in the cumulo-stratus 
 and cumulus (Plate T.) 
 
 As the burster came on, the stratus on the horizon rose 
 gradually up and revealed its true character, at 6 -25 p.m. the 
 roll was about 4 in diameter, and had a lower fringe like a very 
 narrow roll of cumulus, above this a dark black roll of stratus 
 surmounted by a feathery fringe, turning up and trending to 
 south, while under the great roll at its northern end, could be 
 
SOUTHERLY BURSTERS. 53 
 
 
 seen a well defined shower with rain drifting to north at an angle 
 of 45. (See Plate 4 taken at 7 p.m.) Above this well defined roll 
 cloud another fainter and less defined one is clearly but faintly 
 shown in the negative. The whole storm could be seen to great 
 advantage from the Observatory, and the roll came along getting 
 more and more defined, showing all the features just described 
 from 6*25 to 7 p.m., including the shower and the duplicate roll 
 above. At 7 p.m. it looked very close to us, and could not have 
 been more than three or four miles away, for five minutes later 
 the squall with a velocity of forty-two miles per hour and the rain 
 were upon us. On the photograph taken at 7 p.m., (reproduced 
 as Plate 4) the altitude of the lower edge measures 5, and taking 
 the distance to be four miles, the actual height above the ground 
 comes out one thousand eight hundred and thirty feet, and it 
 spread over at least 100 of the horizon, seventy of which the 
 photograph includes. The sun had set at 6 '38 p.m., and night 
 was closing in fast, hastened by the dark masses of cloud which 
 almost covered the sky. 
 
 At 645 p.m., a long way off and due south, a thunder squall 
 was seen and no doubt marked the arrival of the burster there, 
 showing that the axis of the A was still inclined to south-south- 
 east as it was in the morning. The rate of motion of the axis of 
 the A eastward to Sydney from its barometrically defined position 
 at 9 a.m., three hundred miles west of Sydney, to the Observatory 
 by 7 p.m. i.e., in ten hours, is seven hundred miles in the twenty- 
 four hours, or thirty-three miles per hour. 
 
 In the original photograph delicate shading shows the rounded 
 cloud perfectly, with another delicate roll above it, but much of 
 the fine detail is lost in the reproduction. 
 
 DANDENONG GALE. 
 
 I have also added a diagram, (page 32) giving barometer and 
 anemometer conditions at Sydney during the famous Dandenong 
 gale, for comparison, the extensive and disastrous possibilities which 
 a burster may develop made this desirable. During this gale 
 
64 AUSTRALIAN WEATHER. 
 
 the wind attained, locally, to the abnormal velocity of one hundred 
 and fifty-three miles per hour in a gust, the rate of one hundred 
 and twelve miles for ten minutes, and fifty-seven miles per hour 
 for nine hours, while its influence was severely felt for hundreds 
 of miles in every direction. The most remarkable points in this 
 diagram are, first the comparatively steady barometer curve ; and 
 second, the pulsatory action of the wind velocity shown by the 
 mean hourly number of miles registered during the heaviest part 
 of the gale. 
 
 UNFAVOURABLE SEASONS FOR CLOUD OBSERVATIONS. 
 
 The opportunities for making cloud observations this summer 
 have been eminently unfavourable, but I have made the best 
 use of such chances as presented themselves, showing the conditions 
 which prevailed for several hours, both before and after the burst. 
 During the present season they have been preceded by clear skies 
 and by overcast skies, and by cirrus and stratus and cumulus, 
 separately and in combination, and to varying extents. 
 
 The skies obtaining with the north-westers were more often 
 than not, hazy or overcast, almost tropical of aspect, with hazy 
 stratus and occasional heavy thunder cumulus on the several 
 horizons, but with a special partiality to the south-west. 
 
 With the north-easters the skies have been clearer and the 
 smoke-cloud arising from the metropolis, which is seldom notice- 
 able during a north-wester owing to the thick condition of the 
 atmosphere, has generally been perceptible round the horizon in 
 stratified form. The background, or upper strata, when seen, has 
 been composed of fine cirrus, the surface wind occasionally bring- 
 ing in detached cumulus travelling very low. At times also, as 
 with the north-westers, detached cumulus have been visible to the 
 south-west. The lower strata always moved with the wind, while 
 the upper moved from the west, with an occasional tendency to 
 west-north-west or west-south-west, but this motion could only be 
 discerned in a few instances. In the majority of cases no move- 
 ment could be detected. Whether the easterly motion is a current 
 
SOUTHERLY BURSTERS. 55 
 
 or merely the general drift of the atmosphere cannot be decided 
 but as it was noticeable at night I should incline towards the 
 latter hypothesis. No atmospheric current, apart from these two, 
 was observed at any time, except when an agitation was remotely 
 visible overhead between the two, due to vertical uprising or 
 lowering of either one or the other. In these rare instances the 
 clouds moved in all directions. 
 
 There were fewer opportunities for observing the upper strata 
 after the burst than while the northerly winds prevailed, but 
 when the conditions permitted observation the cloud movements 
 generally tended from due west. I do not, however, feel inclined 
 to commit myself more definitely on the subject of cloud aspects 
 since the observations from which these statements are deduced 
 are for one season only, and that season an eminently unfavourable 
 one for the collection of useful data. 
 
 The question whether bursters arrive here by preference at any 
 particular hour, and whether such hour varies from year to year, 
 as well as the average number for each hour, the percentage of 
 the whole that came at each hour seem to me best answered by a 
 diagram and 
 
 TABLE I. 
 
 In Table No. 1, all the bursters that have been recorded at 
 Sydney from September 30, 1863 to March 31, 1894, are grouped 
 together so that they can be seen at a glance, all the bursters that 
 
 NUMBER IN EACH HOUR AND YEAR. l 
 
 are on record for each year, and for each hour of the day, also 
 the total number at each hour for the whole period, and the per- 
 centage of bursters that have taken place at each hour in figures 
 and in diagram. Each stroke in the table represents a burster, 
 where several bursters have occurred at the same hour there is a 
 group of strokes, and the eye catches at once those hours in each 
 year in which bursters have been most frequent, the greatest 
 
 1 See also at the foot of Diagram in., page 58. 
 
56 AUSTRALIAN WEATHER. 
 
 number for any hour in any year was between 9 and 10 p.m., 1891, 
 where ten were recorded. 
 
 VARY FROM YEAR TO YEAR. 
 
 The number for each year varies considerably, the greatest being 
 fifty-six in 1869 ; the least, sixteen, in 1890. Since 1888 there 
 has been a gradual falling off in the number, and this is coincident 
 with abundance of rain each year. 
 
 PREFERENCE FOR A PARTICULAR HOUR. 
 
 The Table No. I shows that from 11 a.m. to 1 p.m. bursters are 
 fewest, while they are most frequent from 6 p.m. to midnight, 
 the chances being slightly in favour of 7 to 8 p.m. 
 
 It also appears that the hour of maximum varies from year to 
 year, in 1891 it was from 9 to 10; 1888, 11 p.m. to midnight; 
 1875, 1 to 2 p.m. and so on. 
 
 The questions as to the number and strength of bursters in the 
 several seasons and months of the year I have treated in the same 
 manner. 
 
 TABLE II. 
 
 In Table No. 2, all the bursters on record have been grouped 
 in another way, which brings into evidence other characteristics. 
 The table has the years on the sides and months across the top, 
 and the information given under each month and for each year is 
 the maximum velocity of the wind, and if there are several bursters 
 in the month the average of their greatest velocities ; at the right 
 hand side the average of the greatest velocities for the year, and 
 the greatest velocity attained at any time during the year. It is 
 also shown that the average number in each year is thirty-two, 
 the greatest in any year fifty-six, and the least sixteen. The 
 average of all the greatest velocities is 4 2 '7 miles per hour, the 
 greatest one hundred and fifty-three miles per hour. (It must be 
 borne in mind that all the anemometer results at Sydney are 
 recorded on the assumption that the velocity of the wind is three 
 times that of the centres of the cup, this has been the practice 
 since 1862, and it is deemed better not to alter it until the exact 
 ratio has been decided and generally accepted.) 
 
SOUTHERLY BURSTERS. 57 
 
 It is necessary to state here that in the conditions under which 
 this prize essay was written, there is no express definition of the 
 minimum velocity of wind which shall constitute a burster, I have 
 therefore taken as such every sudden shift of wind to south or 
 south-west, from any point between west and north on the one 
 side and east and north on the other side, irrespective of the 
 velocity of the wind ; always provided however, that the velocity 
 was maintained subsequently for some hours with force. 
 
 It will be observed that the table brings to light the fact, that 
 the mean velocity of the wind is greater in spring than in autumn, 
 which may be accounted for by the fact that our September 
 equinox is windy and the March one wet. The discrepancy 
 between the seasonal total in the tables is accounted for by the 
 fact that the anemometer on a few occasions failed to record the 
 hours of arrival, in these cases the burster is omitted from Table I. 
 
 One other point seems important in respect to the arrival of 
 bursters, and that is the time between the lowest barometer and 
 the arrival of the burst. 
 
 DIAGRAM III. 
 
 Gives the fluctuations of the self recording barograph for twelve 
 hours before and after nineteen bursters of the summer 1893 to 
 1894. The curves have been drawn with their lowest points in 
 a vertical line, and the times at which the burst commenced on 
 each curve have been connected by lines, thus showing the rela- 
 tion between the barometic minimum and the time that the 
 burst of wind reached Sydney. It will at once be seen that the 
 bursters occur some hours after the barometers have commenced to 
 rise, and if this set of curves be accepted as the general rule, the 
 diagram shows that in any season the lowest readings take place 
 more remotely from the change of wind, early in the summer. It 
 is also patent that no particular height of barometer is peculiar to 
 the burster. A sudden rise does not indicate that the blow will 
 be a hard one, see the sharpest that of 17th September and of 
 30th November, while the velocity of forty-eight miles per hour, 
 
DIAGRAM SHOWING FLUCTUATIONS OF BAROMETER- 
 
 FORK HOURS BEFORE AND AFTER BURSTERS OFTHE SUMMER 1895 TV/89?. 
 TABLE K23. 
 
 arfaZ6^ reed/nys 
 
 f -1 
 
 
 
 
 
 
 
 
 ' 
 
 
 > 
 
 x ^. 
 
 
 
 
 
 
 
 
 
 
 
 ' 
 
 
 
 6 
 
 
 
 
 
 
 
 f .' 
 
 
 
 
 
 >s 
 
 
 
 
 
 
 
 / 
 
 --- 
 
 "" 
 
 /'- 
 
 
 X- 
 
 ^ 
 
 \ 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 i 
 
 
 
 _, 
 
 
 
 
 f 
 
 
 
 
 
 ,' 
 
 
 
 
 
 
 
 
 
 X N 
 
 
 -' 
 
 
 I 
 
 
 .- 
 
 
 
 
 \ 
 
 ) 
 
 
 
 'V 
 
 > 
 
 
 
 
 
 
 
 
 
 - 
 
 
 ^ 
 
 
 ' 
 
 
 ? 
 
 
 
 
 
 
 ? 
 
 
 
 
 
 
 \ 
 
 .,'- 
 
 
 
 
 s 
 
 <i 
 
 / 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
SOUTHERLY BURSTERS. 59 
 
 the heaviest of the season resulted from a very gradual rise and 
 with a range of only -1 in the whole twenty-four hours. The 
 curves at the foot of diagram 3 relate to one feature of Table I. 
 
 HOT SOUTHERLY WINDS. 
 
 Instances have been recorded of hot southerly winds which 
 lasted for some hours. These are no doubt, the winds blowing on 
 the western aspect of the tropical depression. The following is 
 an account of a hot southerly wind copied from notes made by 
 Mr. Russell, and kindly placed at my disposal : 
 
 " December 12th, 1883 -Yesterday was hot, to-day hotter still, 
 and by 9 a.m. the temperature had risen to 82 -6, wind from north 
 and hot. At noon the humidity fell to twenty-nine, and the 
 temperature had risen to 99 '6. At 3 p.m. the dry bulb stood at 
 95-6 and the wet at 71 1 ; from 9 a.m. to 6'30 p.m. the barometer 
 fell 0-350 and then began to rise rapidly, and by 8 p.m. had risen 
 0-215. At 6-30 p.m. as the barometer began to rise the wind 
 changed suddenly to south, and it felt more like a furnace blast 
 than a southerly burster. The wind continued as a hot wind for 
 more than an hour, but not strongly after the first violent gusts 
 were over ; by 9 p.m. a light warm air from south-west had set in 
 but the temperature was still 86-6. I do not remember a hot 
 southerly like this before. On the morning following we had a 
 cool wind from north-west." 
 
 In conclusion, it is needless to state that this investigation has 
 involved much labour and a vast amount of patient research, and 
 though I will be gratified if my efforts should in some degree, how- 
 ever small, contribute to a better knowledge of this phenomenon : 
 in any case, I will consider myself rewarded for the labour I have 
 given to is by the knowledge I have acquired while gathering 
 material for this essay. I am largely indebted to Mr. H. C. 
 Russell, who has kindly given me access to all the valuable records 
 of the Sydney Observatory, and furnised me with copies of 
 weather maps, photographs, etc., for which my warmest thanks 
 are hereby expressed. 
 
60 
 
 AUSTRALIAN WEATHER. 
 
 TABLE II. TABLE SHOWING NUMBER OF BURSTERS IN EACH 
 
 YKAR. 
 
 AUGUST. 
 
 SEPTEMBER. 
 
 OCTOBER. 
 
 NOVEMBER. 
 
 DECEMBER. 
 
 1 
 
 fc 
 
 Velocity. 
 
 1 
 
 1 
 
 Velocity. 
 
 | 
 
 Ra 
 Mile 
 6 
 3 
 5 
 
 3 
 
 7 
 4 
 7 
 5 
 2 
 5 
 6 
 6 
 5 
 3 
 7 
 4 
 7 
 3 
 5 
 5 
 7 
 8 
 4 
 3 
 4 
 1 
 2 
 2 
 6 
 4 
 
 Velocity. 
 
 1 
 
 a 
 K 
 
 Velocity. 
 
 I 
 
 
 
 Velocity. 
 
 d 
 
 9 
 
 I 
 1 
 
 M 
 
 
 1 
 
 a 
 
 M 
 
 w 
 
 hour 
 108 
 49 
 64 
 
 29 
 60 
 47 
 52 
 54 
 44 
 54 
 54 
 34 
 43 
 59 
 72 
 76 
 59 
 63 
 59 
 45 
 41 
 36 
 63 
 37 
 43 
 26 
 35 
 21 
 48 
 43 
 
 S 
 
 0> 
 
 1 
 
 1 
 3 
 
 B 
 
 Summer. 
 
 1863 - 64 
 1865 
 1866 
 1867 
 1868 
 1869 
 1870 
 1871 
 1872 
 1873 
 1874 
 1875 
 1876 
 1877 
 1878 
 1879 
 1880 
 1881 
 1882 
 1883 
 1884 
 1885 
 1886 
 1887 
 1888 
 1889 
 1890 
 1891 
 1892 
 1893 
 1894 
 
 1 
 1 
 
 20*0 
 18*0 
 
 20 
 18 
 
 Ra 
 Mile 
 2 
 2 
 2 
 6 
 3 
 5 
 2 
 1 
 1 
 2 
 
 te in 
 s per 
 58-0 
 11-0 
 26-5 
 25-7 
 30-0 
 17-4 
 29-0 
 39-0 
 37-0 
 40-0 
 
 hour 
 70 
 12 
 32 
 40 
 47 
 27 
 36 
 39 
 37 
 42 
 
 te in 
 sper 
 54-8 
 32-3 
 43-8 
 
 23-3 
 43-0 
 32-5 
 33-8 
 35-6 
 40-0 
 36-8 
 40-8 
 28-3 
 34-8 
 46-3 
 45-9 
 39-2 
 38-3 
 47-0 
 34-2 
 34-4 
 35-4 
 26-9 
 32-5 
 35-3 
 34-0 
 26-0 
 270 
 21-0 
 29-7 
 33-5 
 
 Ra 
 
 Mile 
 
 7 
 5 
 10 
 6 
 10 
 11 
 4 
 6 
 5 
 1 
 4 
 5 
 6 
 6 
 8 
 9 
 5 
 6 
 5 
 4 
 4 
 5 
 7 
 3 
 4 
 3 
 3 
 2 
 3 
 2 
 7 
 
 te in 
 sper 
 29-4 
 25-6 
 38-9 
 28-7 
 29-4 
 37-0 
 44-3 
 40-0 
 30-2 
 47-0 
 347 
 29-4 
 26-7 
 30-8 
 400 
 30-5 
 35-4 
 40-0 
 42-0 
 34-3 
 22-5 
 34-0 
 33-0 
 24-3 
 34-5 
 35-0 
 19-7 
 31-0 
 35-7 
 33-0 
 31-4 
 
 hour 
 41 
 50 
 74 
 42 
 43 
 72 
 70 
 69 
 40 
 47 
 44 
 41 
 37 
 55 
 53 
 45 
 63 
 68 
 54 
 53 
 27 
 48 
 47 
 29 
 48 
 46 
 24 
 48 
 41 
 42 
 48 
 
 Rn 
 Mile 
 3 
 4 
 7 
 4 
 4 
 8 
 8 
 6 
 6 
 4 
 6 
 8 
 4 
 7 
 9 
 9 
 8 
 8 
 5 
 7 
 7 
 5 
 4 
 4 
 5 
 6 
 8 
 4 
 5 
 6 
 3 
 
 te in 
 sper 
 32-3 
 24-3 
 28-3 
 38-0 
 36-3 
 37-8 
 30-5 
 45-0 
 42-5 
 29-3 
 37-3 
 32-9 
 31-0 
 32-9 
 401 
 31-0 
 37-8 
 61-1 
 29-6 
 32-1 
 30-6 
 38-2 
 38-0 
 30-3 
 28-0 
 29-0 
 2P-1 
 22-5 
 28-8 
 32-3 
 29-0 
 
 hour 
 49 
 36 
 48 
 5<> 
 69 
 73 
 53 
 63 
 64 
 45 
 61 
 54 
 50 
 64 
 64 
 52 
 01 
 82 
 48 
 43 
 54 
 50 
 41 
 37 
 50 
 43 
 42 
 30 
 35 
 58 
 40 
 
 2 
 
 33-5 
 
 37 
 
 1 
 3 
 6 
 2 
 4 
 
 20-0 
 24-3 
 47-2 
 29-0 
 29-3 
 
 20 
 30 
 153 
 32 
 42 
 
 1 
 
 16-0 
 
 16 
 
 2 
 2 
 4 
 1 
 5 
 1 
 
 28-5 
 36-5 
 43-8 
 32-0 
 28-8 
 30-0 
 
 33 
 41 
 
 68 
 32 
 48 
 30 
 
 ... 
 
 ... 
 
 ... 
 
 1 
 1 
 
 1 
 1 
 
 43-0 
 18-0 
 21-0 
 21-0 
 
 43 
 
 18 
 21 
 21 
 
 
 
 
 
 
 
 
 
 
 1 
 
 30-0 
 
 30 
 
 Total 
 
 5 
 
 ... 
 
 ... 
 
 62 
 
 
 
 
 139 
 
 
 
 
 
 166 
 16-8 
 
 
 
 
 
 182 
 18-3 
 
 
 
 
 
 Percent 
 
 5 
 
 ... 
 
 
 6-3 
 
 
 14-0 
 
 Mean 
 
 2 
 
 21-4 
 
 
 2-0 
 
 30-3 
 
 
 4-5 
 
 35-5 
 
 ... 
 
 5-4 
 
 33-5 
 
 ... 
 
 5-9 
 
 33-7 
 
 ... 
 
 Extreme 
 
 2 
 
 33-5 
 
 37 
 
 6 
 
 58-0 
 
 153 
 
 8 
 
 54-8 
 1864 
 
 108 
 1864 
 
 11 
 
 47-0 
 1873 
 
 74 
 1866 
 
 9 
 
 45-0 
 1871 
 
 73 
 
 1869 
 
 Year 
 
 
 18761876 
 
 ... 
 
 1864 
 
 1876 
 
 NOTE. In making out this Table the greatest velocity of each southerly has 
 if there are several bursters in the month, the average of their greatest veloci- 
 
SOUTHERLY BURSTERS. 
 
 61 
 
 [ONTH WITH MEANS AND EXTEEMES OF GEEATEST VELOCITIES. 
 
 LKUART. 
 
 FEBRUARY. 
 
 MARCH. 
 
 APRIL. 
 
 MAY. 
 
 1 
 1 
 
 
 * 
 
 I 
 
 Ra 
 Mile 
 31 
 36 
 37 
 37 
 38 
 56 
 29 
 34 
 34 
 22 
 30 
 38 
 35 
 37 
 32 
 47 
 28 
 37 
 34 
 30 
 35 
 43 
 32 
 20 
 32 
 27 | 
 16 
 23 
 21 
 21 
 19 
 
 991 
 
 Velocity. 
 
 h 
 
 1 
 
 1864 
 1865 
 1866 
 11867 
 186S 
 H869 
 1870 
 1871 
 1872 
 11873 
 11874 
 11875 
 :1876 
 !l877 
 1878 
 1879 
 1880 
 1881 
 1882 
 1883 
 1884 
 1885 
 1886' 
 1887! 
 1888 
 1889 
 1890 
 1891: 
 1692 
 1893 
 1894 
 
 Velocity. 
 
 1 
 
 p 
 Bq 
 
 Velocity. 
 
 1 
 1 
 
 Velocity. 
 
 | 
 
 fc 
 
 Velocity. 
 
 1 
 
 3 
 fc 
 
 Ra 
 
 Mile 
 
 Velocity. 
 
 1 
 
 1 
 H 
 
 d 
 
 1 
 
 te in 
 sper 
 22-0 
 29-5 
 29-8 
 48-4 
 35-4 
 34-4 
 39-8 
 35-3 
 34-0 
 35-7 
 34-3 
 38-1 
 32-3 
 26-0 
 35-3 
 39-0 
 29-0 
 48-0 
 37-6 
 34-0 
 39-3 
 29-3 
 25-5 
 30-5 
 29-8 
 27-8 
 37-3 
 26-3 
 40-5 
 33-0 
 
 0> 
 
 1 
 
 hour 
 48 
 45 
 46 
 69 
 54 
 63 
 49 
 52 
 50 
 45 
 47 
 52 
 44 
 38 
 44 
 52 
 36 
 71 
 67 
 50 
 54 
 40 
 30 
 32 
 46 
 35 
 40 
 40 
 46 
 57 
 
 a 
 
 te in 
 sper 
 30-5 
 22-4 
 40-5 
 26-7 
 51-0 
 24-0 
 41-6 
 38-8 
 33-4 
 29-5 
 28-8 
 38-0 
 32-4 
 26-0 
 39-0 
 31-3 
 36-0 
 34'6 
 30'8 
 28-7 
 29-9 
 30-1 
 35-8 
 41-0 
 31-6 
 31-0 
 
 0> 
 
 1 
 
 hour 
 31 
 36 
 46 
 44 
 77 
 33 
 55 
 51 
 40 
 34 
 44 
 64 
 39 
 34 
 41 
 36 
 36 
 54 
 66 
 37 
 46 
 50 
 46 
 41 
 45 
 34 
 
 1 
 
 te in 
 sper 
 34-0 
 24-5 
 50-0 
 16-0 
 33-3 
 34-3 
 
 9 
 
 1 
 
 1 
 
 hour 
 34 
 38 
 54 
 18 
 63 
 61 
 
 1 
 
 te in 
 sper 
 
 Q> 
 1 
 
 1 
 
 tein 
 
 sper 
 
 <c 
 
 I 
 
 hour 
 
 Ra 
 
 Mile 
 4 
 
 7 
 3 
 6 
 3 
 5 
 5 
 4 
 7 
 2 
 5 
 3 
 5 
 4 
 2 
 4 
 1 
 5 
 5 
 3 
 7 
 7 
 5 
 1 
 5 
 5 
 
 7 
 
 5 
 3 
 
 4 
 
 Ra 
 Mile 
 2 
 4 
 2 
 4 
 4 
 8 
 
 Ra 
 Mile 
 
 3 
 2 
 5 
 3 
 2 
 
 hour 
 
 ce in 
 sper 
 37-3 
 24-6 
 35-0 
 25-3 
 31-7 
 29-6 
 33-6 
 37'1 
 34-0 
 34-4 
 32-4 
 31-4 
 30-3 
 35-1 
 35-6 
 33-7 
 34-6 
 42-7 
 34-8 
 31-4 
 31-9 
 285 
 32 6 
 31-5 
 3L-6 
 28-4 
 26-6 
 25 6 
 27-8 
 30-3 
 30-2 
 
 hour 
 108 
 50 
 74 
 69 
 77 
 73 
 70 
 69 
 64 
 47 
 61 
 64 
 50 
 153 
 64 
 72 
 76 
 82 
 67 
 68 
 54 
 50 
 47 
 63 
 50 
 46 
 42 
 48 
 46 
 58 
 48 
 
 27-3 
 
 22-5 
 26-0 
 14-7 
 24-0 
 
 40 
 23 
 26 
 16 
 36 
 
 3 
 1 
 
 14-5 
 43-0 
 
 15 
 43 
 
 3 
 1 
 5 
 1 
 5 
 3 
 1 
 4 
 3 
 3 
 3 
 4 
 1 
 3 
 7 
 3 
 4 
 4 
 3 
 
 19-0 
 40-0 
 20-4 
 35-0 
 29-0 
 35-0 
 39-0 
 19-7 
 34-7 
 30-0 
 27-3 
 34-7 
 28-0 
 33-7 
 28-7 
 39-0 
 27-8 
 27-0 
 24-0 
 
 25 
 
 40 
 27 
 35 
 36 
 43 
 39 
 24 
 47 
 32 
 30 
 45 
 28 
 43 
 41 
 47 
 34 
 30 
 33 
 
 3 
 3 
 
 3 
 1 
 2 
 1 
 
 40-0 
 19-3 
 
 56 
 24 
 
 20-0 
 23-0 
 29-0 
 44-0 
 
 29 
 23 
 35 
 
 44 
 
 ... 
 
 
 ... 
 
 4 
 
 1 
 2 
 
 27-5 
 
 64-0 
 20-0 
 
 46 
 
 64 
 20 
 
 
 ... 
 
 
 
 
 
 
 1 
 
 "2 
 2 
 
 20-0 
 
 340 
 23-5 
 
 20 
 
 42 
 31 
 
 
 
 
 
 
 
 
 
 
 24-1 
 
 22-0 
 24'0 
 27-0 
 
 42 
 26 
 35 
 30 
 
 1 
 3 
 1 
 
 27-0 
 25-7 
 30-0 
 
 27 
 27 
 30 
 
 
 
 
 
 1 
 
 21-0 
 
 21 
 
 ... 
 
 ... 
 
 
 
 
 
 
 
 
 
 
 
 41 
 
 
 
 
 
 
 
 
 
 
 132 
 
 90 
 
 
 
 4 
 
 
 
 33-9 
 
 
 
 13-2 
 4-3 
 
 31-7 
 
 
 
 9-1 
 3-2 
 
 30-2 
 
 
 
 4-1 
 1-3 
 
 27'7 
 
 
 
 4 
 
 o-i 
 
 28-8 
 
 
 
 100 
 32 
 
 31-9 
 
 
 
 
 
 48-4 
 1867 
 
 71 
 
 1881 
 
 7 
 
 51-0 
 
 
 
 1868 
 
 77 
 1868 
 
 8 
 
 50-0 
 
 63 
 
 1868 
 
 5 
 
 64-0 
 1882 
 
 56 
 
 3 
 
 43 
 
 43 
 
 56 
 
 42-71 153 
 
 
 
 1866 
 
 
 1871 
 
 
 1871 
 
 1871 
 
 1869|l88l|l876 
 
 ... 
 
 en found, by ascertaining the shortest time in which four miles of wind was recorded; 
 js is shown under the head " Mean." 
 
OCOt^aOC5O< 1 O5 
 
 30 00 00 00 30 Ci O5 C. 
 
 I 
 
 <N 
 I 
 
 ii 
 
 H 
 I 
 g 
 
 I 
 
 C5 
 
 CO 
 
 ^ ^ 
 
 | ^DCDCDCOCOi^X>J^t^I^I>l>-l>-t>t^onQOOOXGOCOOOGCOOOOOiCiOiOi 
 
 ccooQOooooooaoaoooaDooooaoooooooooooooooaDooooaoooaDoooooooO 
 
TYPES. 63 
 
 TYPES OF AUSTRALIAN WEATHER. 
 By HENRY A. HUNT, 
 
 Second Meteorological Assistant, Sydney Observatory. 
 [With Forty Diagrams.] 
 
 IN continuation of the valuable work on Australian Meteorology 
 which the Hon. Ralph Abercromby initiated several years since, 
 by offering a money prize for the best essay on Southerly Bursters ; 
 he has recently selected the phases of Australian Weather which 
 are treated in the following twenty studies of " Types of Austra- 
 lian Weather." Many of these appear to be peculiar to Australia, 
 and at the same time connected with Equatorial and other weather. 
 That they throw much new light upon the source of the greater 
 part of Australian rain, and show how these rain storms develop 
 out of ordinary weather conditions is certain ; at the same time 
 they form an important contribution to the study of weather in 
 the Southern Hemisphere generally. The work has been done 
 by Mr. H. A. Hunt, at Mr. Abercromby's expense, and Mr. 
 H. C. Russell, has edited the work. 
 
 GENERAL REMARKS. 
 
 Asa general rule, weather is set fine when anticyclones move 
 rapidly, and in a straight line across Australia, i.e., at a rate 
 exceeding five hundred miles per day. And weather is unsettled 
 when they move slowly, and not in a straight line, i.e., in a zigzag 
 line, especially if they show no appreciable forward motion for a 
 day or two. When anticyclones move in low latitudes the con- 
 ditions favour dry weather, in high latitudes, wet weather, 
 especially if they rest for a time south of South Australia. 
 
 All the examples of weather phases which follow have been 
 selected from the Sydney Weather Charts, and illustrate each 
 type. The originals were carefully traced, and then reduced from 
 22" x 17" to the size used in this essay by means of the camera. 
 
64 AUSTRALIAN WEATHER. 
 
 Ordinary symbols have been used, except the circle half filled 
 for thunderstorms, and the straight line shading ; parallel lines 
 indicate the area of rainfall under one inch, crossed shading over 
 one inch. 
 
 TYPE I. MOVING ANTICYCLONES. 
 
 One of the best marked features of Australian weather is the 
 steady easterly progression of all the types, and the governing 
 type, that in fact about which all the other types seem to congre- 
 gate, is the anticyclone ; it has therefore been placed first in the 
 series, with three charts to show the progress made by a quick 
 moving one in forty-eight hours. The average daily progress of 
 anticyclones is four hundred miles per day, but the speed at times 
 rises to one thousand miles. 1 
 
 Investigation so far leaves no room to doubt that in these lati- 
 tudes a series of anticyclones surround the globe ; the latitude of 
 the average one varies with the season, being farther south in 
 summer than in winter. The normal circulation about an anti- 
 cyclone brings southerly winds in front of them, and northerly 
 winds in the rear, hence our cold and our hot winds. 
 
 Chart No. 1 shows the position, on 15th August, 1893, of the 
 eastern half of an incoming anticyclone; it rests over Western 
 Australia, while the departing one is seen over the Tasman Sea; 
 between these is seen the usual \ depression, which is of average 
 intensity, and a dormant tropical low pressure to the north. In 
 Chart No. 2 the anticyclone has moved nearly nine hundred miles 
 in the twenty-four hours, the centre being located near Fowler's 
 Bay, north of the Australian Bight ; the antarctic \ depression 
 is well across the Tasman Sea, while the tropical or monsoonal 
 isobar depicted in the previous chart has apparently merged into 
 the high pressure system, a curious and not unusual kink being 
 formed to the north-east of it, following the contour of the Gulf 
 of Carpentaria. On Chart No. 3 the anticyclone is shown to 
 have moved a further seven hundred and fifty miles, or a total of 
 
 1 Eussell Quarterly Journal K. M. S., Vol. xix., No. 85. 
 
TYPES. 
 
 65 
 
 TYPES 
 AUSTRALIAN WEATHER 
 
 Maying Anticyclones I to 
 
 TYPES 
 
 OF 
 
 AUSTRALIAN WEATHER 
 N2 
 
 one thousand six hundred and fifty miles in forty-eight hours; 
 the monsoonal dip is still noticeable south of the Gulf, while 
 the antarctic \ depression has its centre to the east of New 
 Zealand. 
 
66 
 
 AUSTRALIAN WEATHER. 
 
 TYPES 
 
 OF 
 
 AUSTRALIAN WEATHER 
 
 N3 
 
 The passage of this anticyclone was an unusually rapid one, 
 and is presented as portraying with a minimum number of charts 
 the easterly motion of anticyclones over Australia. 
 
 TYPE II. MONSOONAL RAIN STORM. 
 
 This type is undoubtedly the chief rain agent in the Australian 
 Continent. Monsoonal depressions or tongues may occur at any 
 time of the year, but particularly between the months of September 
 and April, and most frequently during January, February and 
 March. The readings of barometers in the depression seldom fall 
 very low, the grade from the surrounding areas to the centre of 
 the tongue ranging from one to three-tenths of an inch generally; 
 the depression may intensify, that is the tongues between high 
 pressures may protrude further south anywhere during their passage 
 across Australia, but show a preference to do so after they have 
 crossed central Australia, a fact which suggests that the heated 
 interior has at least some influence in their development. 
 
 When and wherever the tongue is well denned, rain certainly 
 follows in its track, and thunderstorms as a wide spread and 
 simultaneous feature are never experienced without it. 
 
TYPES. 
 
 67 
 
 On April 18fch and 19th, 1894 (Chart Nos. 4 and 5) occurred one 
 of the finest monsoonal rain storms on record, the area affected 
 being very extensive, embracing the whole of the eastern colonies, 
 Tasmania, and the greater part of South Australia. Many inches 
 up to five and six of rain were recorded on the north-east 
 coast of Queensland, and over the eastern half of the rain area 
 the country benefitted to the extent of over two inches generally 
 before the storm was over. 
 
 TYPES 
 
 X AUSTRALIAN WEATHER 
 
 N-t 
 
 4l Monsoonal Rain 4 5 
 
 On this Chart No. 4 the monsoonal isobars, after passing the 
 most southern part of the tongue, are shown sweeping round a 
 high pressure of considerable energy, situated over the Great 
 Bight. Following these isobars to the eastward, we find them 
 recurving over another high pressure of greater energy in the 
 Tasman Sea. On Chart No. 5, both anticyclones will be noticed to 
 have worked northward, and while so doing they have lost some- 
 what in pressure, and the low pressure tongue has extended further 
 south and broadened ; at the tip, two cyclonic rain centres will be 
 observed to have formed. 
 
 The chart antecedent to 18th April presented the general 
 characteristics of Chart No. 1, except that the antarctic \ depress- 
 
68 
 
 AUSTRALIAN WEATHER. 
 
 ion was a considerably less active feature, but on the other hand 
 the monsoonal dip was somewhat more pronounced, and possessed 
 two instead of one isobar. Following the 19th, the chart of 20th 
 revealed no sign of the monsoonal tongue, and the continent was 
 covered with a high pressure of very slight energy. 
 
 TYPES 
 AUSTRALIAN WEATHER 
 
 JLPSIL 19_..ie.< 
 
 TYPE III. DEVELOPMENT OF A CYCLONIC STORM IN LOW LATITUDES 
 FROM A MONSOONAL DEPRESSION. 
 
 In Type No. 3 we have the development of a cyclonic storm 
 out of a monsoonal depression. The seasonal peculiarity of this 
 phase of the tropical low pressure is similar to that in Type No. 2. 
 The cyclone seems to develop when the southern extension of the 
 monsoon is out of proportion to its width, and it becomes so 
 narrow at one part of it that the opposing winds which circulate 
 round it interfere, set up the cyclonic circulation, and it then 
 progresses eastward as a rain storm. (See Charts 6 and 7.) 
 
 These storms frequently develop in South-east Queensland, 
 and they are generally most severe there ; the quantity of rain 
 which sometimes comes with them is very remarkable, as in the 
 case of the phenomenal flood in Brisbane in 1893, which was the 
 result of oue of these storms. 
 
TYPES. 
 
 69 
 
 The example selected and shown in Charts 6 and 7 took place 
 on September 28th and 29th, 1892 ; on the 28th (Chart 6), it 
 appears that one of these storms had developed in the previous 
 twenty-four hours, at the southern end of a narrow tropical or 
 monsoonal tongue of low pressure. An anticyclone of good energy 
 lies to the east of it, with its isobars extending well to the north 
 and contracting the width of the tongue in the north ; another 
 anticyclone lies to the west, and this also seems to be pressing on 
 the narrow monsoonal tongue and helping to contract its diameter, 
 at the same time the energy in both high pressures is adding force 
 to the circulation of the wind, and so aiding in the development 
 of the cyclonic circulation. A modified \ depression exists to the 
 south-east of the Australian Bight, and there is another over New 
 Zealand. 
 
 TYPES 
 
 AUSTRALIAN WEATHER 
 ^ N6 
 
 On September 29th, (Chart 7), the small cyclone has extended 
 its area and energy on its way to the coast, but its motion of 
 rotation has been rapid, and this probably accounts in a measure 
 for the comparatively small area over which rain has fallen, 
 although, in this instance, over an inch of rain fell in the central 
 
70 
 
 AUSTRALIAN WEATHER. 
 
 and eastern parts of New South Wales. The winds circulating 
 about the western isobars of this storm are rather stronger than 
 usual in such cases. In the twenty-four hours both anticyclones 
 have lost a considerable portion of their energy, while that of the 
 depression in New Zealand is about the same as on the previous 
 day. 
 
 TYPES 
 
 OF 
 
 AUSTRALIAN WEATHER 
 N7 
 
 TYPE IV. DEVELOPMENT OF A CYCLONIC STORM IN HIGH LATITUDES 
 
 FROM A MONSOONAL DEPRESSION. 
 
 These are somewhat similar to Type 3, but the rainfall is 
 usually not so heavy, and the wind much more violent. 
 
 Chart No. S shows the development of one of these cyclones on 
 April 15th, 1889; in this case the monsoonal depression had 
 extended across Australia into the Australian Bight. The 
 development of one of these storms is heralded by the strong 
 easterly gales on the south-east coast of South Australia and 
 south coast of Victoria. The energy of wind circulation increases 
 over South Australia, and as the whole system moves bodily east- 
 ward from the Australian Bight to the mainland, the circulation is 
 seen to be that of a fully developed cyclone of small area, with a 
 
TYPES. 
 
 71 
 
 diameter three to five hundred miles, formed out of the southern 
 part of the extensive monsoonal depression. 
 
 In Chart 8 a high pressure is shown over the south-western 
 part of Australia, and evidently encroaching on the moonsoonal 
 about central Australia. 
 
 TYPES 
 
 OF 
 
 AUSTRALIAN WEATHER 
 N8 
 
 Cyclone in high lalitiLdes from 
 ** Monsoonal depression 8 to 10 
 
 TYPES 
 
 AUSTRALIA)! WEATHER 
 Ne 
 
 APRIL 16. i*a 
 
1'2 AUSTRALIAN WEATHER. 
 
 In Chart 9, the western anticyclone has extended northwards 
 and has seriously contracted the diameter of the monsoona! low 
 pressure, thereby facilitating or helping to cut off and start the 
 cyclonic storm in the south ; which has now developed a much 
 steeper barometric grade arid an energetic wind circulation, and 
 is moving eastward without extending the rain area. 
 
 Chart No. 10 shows comparatively little motion in the cyclone, 
 but it is considerably distorted, especially to the west, where the 
 anticyclone is compressing the isobars by its easterly progress and 
 in its endeavours to maintain its rate has overridden the cyclone 
 in the north. As will be seen by a glance at these three charts, 
 the rain resulting from this storm was most extensive and bene- 
 iicial, and the winds under its influences were strong. 
 
 TYPES 
 
 or 
 
 AUSTRALIAN WEATHER 
 NIO 
 
 APRIL 17 ! 
 
 TYPE V. CONDITIONS FAVORABLE FOR THUNDERSTORMS. 
 
 Upon comparison of charts setting forth this type with those 
 of cyclonic thunderstorms, they will be found very similar ; the 
 main difference being the absence in this set of the cyclonic area 
 at the end of the monsoonal tongue. The chief feature indicating 
 
TYPES. 
 
 73 
 
 thunderstorms is the narrowness of the col 1 and somewhat con- 
 gested state of isobars in the hisjh pressures west and east of it, 
 resulting in opposing winds. Those of the tropical tongues are 
 hot and charged with moisture, while those of the A depressions 
 are strong and dry. This type is met with during the monsoonal 
 season. The rains resulting from this feature are not generally 
 heavy, and though thunderstorms may be experienced over exten- 
 sive areas of Queensland, New South Wales, and Northern 
 Territory, a number of them may occur without any rain falling. 
 Chart No. 11, January 17th, 1893, shows an extended and 
 narrow tongue lying between two relatively high pressures, a \ 
 depression similarly situated also exists in the south, this col or 
 area of low pressure separating these two depressions is very 
 small, and opposing currents of wind are noted there as blowing 
 within close limits. Thunderstorms were occurring or had 
 occurred in Northern Territory generally, over a great part of 
 Queensland and in central parts of New South Wales, the rains 
 upon this occasion, as is often the case, were not heavy, though 
 they fell over an extensive and generally unfavoured area. 
 
 TYPES 
 
 or 
 
 AUSTRALIAN WEATHER 
 Nn 
 
 toM for thunderstorms 
 ,a // & 12 
 
 For explanation of this term see p. 80. 
 
74 
 
 AUSTRALIAN WEATHER. 
 
 On Chart No. 12, the following day, though a few storms were 
 recorded, shows a great diminution in number. The col has 
 widened and the monsoonal tongue has lost to some extent its 
 thundery characteristics, having widened at the end. The 
 accompanying barometric systems show no motion since the 
 previous day, but the high pressures have intensified. 
 
 TYPES 
 
 OP 
 
 AUSTRALIAN WEATHER 
 
 NI2 
 JAM VARY 18. *< 
 
 TYPE VI. CYCLONIC THUNDERSTORMS. 
 
 This also, like the preceding one, is allied to the tropical low 
 pressures, but in this case a defined cyclonic circulation develops 
 in the lower extension of the tongue without the usual intensifi- 
 cation of grades. From this source the thunderstorms radiate in 
 easterly and southerly directions, and at times, as in the instance 
 presented, a vast area is affected. 
 
 Chart No. 13, December 12th, 1893. As on the previous set 
 the monsoonal tongue lies over much the same country, though 
 with its axis more east and west; the high pressure to the west is 
 very small, and the systems are somewhat fragmentary. 
 
TYPES. 
 
 75 
 
 TYPES 
 
 AUSTRALIAN WEATHER 
 NI3 
 
 DECEMBER I2asos 
 
 LOW 
 
 Cjclonjc thunderstorms 
 /3 &./4 
 
 TYPES 
 
 OF 
 
 AUSTRALIAN WEATHER 
 
 On the following day, Chart No. 14, a cyclonic .circulation has 
 formed in the end of the tongue, while the A depression over 1 
 Sydney on the 12th has moved eastward and nearly filled up, and 
 the high pressures have followed and intensified. The thunder- 
 
76 AUSTRALIAN WEATHER. 
 
 storms and rains were not so general as on the thunderstorm 
 type, but curiously, were recorded in an almost parallel area of 
 country, nearly three hundred miles wide by twelve hundred 
 long. Though great barometric changes have taken place, the 
 pressures remain about the same as on Chart No. 13. 
 
 TYPE VII. A RAIN STORM WITH VERTICAL AND NEARLY STRAIGHT 
 
 ISOBARS. 
 
 This type is one of the best defined and reliable of the series 
 for forecasting purposes, because with them, good general rains 
 almost invariably come. They are the rear isobars of a departing 
 anticyclone, and the wind circulation from north and north-east 
 brings into the interior winds laden with tropical moisture to 
 meet in the west southerly winds laden with antarctic cold, 
 and therefore precipitating power. The sequence of rain is 
 rendered even more certain if rain be recorded in the north-east 
 before the isobars straighten, or in the \ depression to the south. 
 
 The actual height of the barometers is not material, but the 
 greater the number of isobars in a given area, the more extensive 
 will be the rainfall ; the rain usually lasts three days. The rain 
 begins to fall north- west of New South Wales, spreads southwards, 
 then eastward, and finally northwards, crossing the mountains 
 near the Queensland boundary. A fine example of this type 
 occurred on 15th, 16th, and 17th October, 1894. (Charts 16, 17.) 
 
 On the 15th, Chart 16, a departing anticyclone rests over 
 Tasman Sea, and its rear isobars are shown running north and 
 south over central Australia ; another anticyclone is shown over 
 Western Australia and a \ depression east of the Australian 
 Bight ; a trough of low pressure rests over Central Australia from 
 north to south. On this day the only indications of the pending 
 rain were found in the cloudy skies generally over South Australia, 
 western parts of New South Wales and Queensland, and a small 
 area of rain in South Australia. 
 
 In Chart 16, the straight isobars of Chart 15 have entirely 
 disappeared, but the rain has come over Central and South 
 
TYPES. 
 
 77 
 
 TYPES 
 
 AUSTRALIAN WEATHER 
 
 NI5 
 OCTOBERS ie*- 
 
 s Vertical isobars 
 
 15 
 
 Australia. Excepting the Gulf country and its central area, all 
 Queensland, all New South Wales, Victoria and Tasmania had 
 rain. (See shading on Chart 16). For other instances of straight 
 isobars see Charts 5, 27, and 28. 
 
 TYPES 
 
 AUSTRALIAN WEATHER 
 
 N6 
 ocroezf? ic ,89* 
 
 1*1 
 
78 
 
 AUSTRALIAN WEATHER. 
 
 TYPE VIII. CYCLONES FROM NORTH-WEST. 
 
 From time to time fully developed cyclonic storms appear on 
 the north-west and west coasts of Australia, and in the Australian 
 Bight, but the absence of observing stations in the unoccupied 
 country which lies between the overland telegraph line and the 
 west coast of Australia, makes it impossible to trace them over 
 that part of the continent, but cyclones are well known on the 
 northern coast of Western Australia, and their formation in the 
 tropics equally well known. There can therefore be no doubt 
 that when we find a cyclone on the western coast of Australia 
 or in the Australian Bight, that it is one which has come from 
 the north-west, and is in fact recurving to the east and south-east 
 as they do on the east coast. 
 
 The one selected for Type 8 was picked up on the west coast of 
 Australia in Latitude 28 on July 4th, 1892, (See Chart No. 17). 
 The winds were light, but the rain heavy along the coast; an inert 
 anticyclone rested over South Australia, Victoria, and New 
 South Wales, where it was moving to the east and making room 
 for the approaching storm. 
 
 TYPES 
 
 AUSTRALIAN WEATHER 
 NI7 
 
 1. iea 
 
 Cyclones from North West 
 J7to/9 
 
TYPES. 
 
 79 
 
 Chart 18, July 5th, shows that the whole system has moved 
 rapidly, that the cyclone now with an elongated centre lying 
 north-west to south-east now rests over South Australia, strong 
 winds are rapidly developing, and the barometric grade about the 
 
 TYPES 
 
 AUSTRALIAN WEATHER 
 KI8 
 
 AUSTRALIAN WEATHER 
 Nie 
 
 ..; Att 6y.fi.. 
 
80 AUSTRALIAN WEATHER. 
 
 centre is much steeper, and rain has fallen over the south coast 
 generally and extended northwards almost to Central Australia. 
 
 On July 6th at 9 a.m. the weather chart presented the features 
 shown in Chart 19. The cyclone has intensified all round, and 
 has moved rapidly to the east, its centre is just entering 
 Bass' Straits ; very heavy gales from south-west are blowing 
 in the rear of the centre ; heavy rain is falling over Victoria and 
 extends over the greater part of New South Wales. All the winds 
 controlled by this storm were very heavy, and during the 6th 
 July, as the storm passed through Bass' Straits, extremely heavy 
 weather was experienced there. On July 7th it had filled up. 
 
 TYPE IX. TORNADOES. 
 
 These occur during the summer months, and are most frequent 
 in the western plains ; they are developed in hot weather and in 
 the low pressure known as a " Col " between two high pressures, 
 when there is not enough grade to control the winds and the heat- 
 ing power of the sun is great ; if to these there is added the pre- 
 sence of moisture from recent showers, we have all the conditions 
 for the formation of a tornado. The force of wind is often suf- 
 ficient to break off growing trees, two and even three feet in 
 diameter, and the reason there is so little damage to life and 
 property is not the want of power, but the sparse population and 
 the very small number of towns. 
 
 Chart 20, March 20th, 1894. An extensive anticyclone lies 
 south of Australia, giving way in its central parts to an extensive 
 monsoonal dip. The isobars are generally uniform and of even 
 gradients, though a suspicious interval exists to the west of Now 
 South Wales between the 30 -0 and 29*9 curves. This is un- 
 doubtedly the area in which the secondary developed. Light rains 
 were recorded in New South Wales and Victoria, but in Central 
 Australia temperatures were high. 
 
 Chart 21, March 21st. A marvellous change has taken place. 
 The area of high pressure on the previous day over the Tasinan 
 Sea has lost two-tenths in pressure. The moiisoonal dip so pro- 
 
TYPES. 
 
 TYPES 
 
 OF 
 
 AUSTRALIAN WEATHER 
 
 N20 
 
 TYPES 
 STRALIAN WEATHER 
 
 nounced then has retreated, and is now only represented by one 
 isobar, and where the western portion of the anticyclone existed 
 on the 20th, an antarctic depression has protruded itself, and lastly 
 an anticyclone is entering Western Australia. Thus a most 
 favourable and extensive col area exists between these four systems. 
 F 
 
82 AUSTRALIAN WEATHER. 
 
 for the generation of these storms. The winds are generally 
 blowing any way, and excepting those of the eastern high 
 pressure have apparently no circulating power. 
 
 These were the conditions at 9 a.m., but as the day advanced 
 the col area advanced with its arid heat, and this acting upon the 
 precipitated moisture of the previous two days, resulted in the 
 tornado which we are about to describe, and which occured at 
 Bourke on the morning of March 21st, being one of the most 
 terrific ever witnessed in that district. It struck the town at 10 
 o'clock, but could be seen approaching for some time from a 
 north-westerly direction. It only lasted six minutes, but during 
 that period thirty-nine points of rain fell, and several hundred 
 pounds' worth of damage was done to houses. Chimneys, veran- 
 dahs, trees, etc., suffered and general consternation prevailed. 
 Many narrow escapes occurred, as the cyclone came across the 
 common from the direction of Fort Bourke. The theatre was 
 unroofed and a quantity of beams and iron was deposited in an 
 adjoining yard. Pleasure boats on the river were sunk, and a 
 steam-boat was considerably damaged by the falling of a large 
 gum tree. 
 
 TYPE X. CYCLONES FROM NORTH-EAST. 
 
 A comparatively small number of these storms reach the coast 
 of Australia, and owing to the almost complete absence of observ- 
 ing stations, New Caledonia excepted, and the small number of 
 vessels passing their tracks, it is usually impossible to trace their 
 course before they reach Australia, but there seems to be no reason 
 to doubt that they are more or less spent tropical cyclones, which 
 reach Australia in the act of recurving. The majority reach the 
 coast of Queensland between latitudes 20 and 26 ; some farther 
 north and south ; only one has been traced from the coast inland, 
 and then recurving there to south-east. It reached the coast in 
 the neighbourhood of Brisbane in January 1893, passed inland 
 over the mountains, gradually curving to south past Mudgee and 
 Dubbo, thence curving easterly it left the mainland about latitude 
 35 S. Its course was marked by violent cyclonic wind and rain. 
 
TYPES. 
 
 83 
 
 Picton reported 4-76" rain. It has not been possible to present 
 this as an example, because one of the days was a Sunday when 
 very few observations are recorded. 
 
 As a rule these north-east cyclones recurve at the coast line. 
 Charts 22 and 23 show the weather conditions on March 10th 
 and llth, 1891. 
 
 On March 10th there was a sudden fall of pressure on the coast 
 of Queensland, about latitude 24 (Chart 22), and an accession 
 
 TYPES 
 AUSTRALIAN WEATHER 
 
 Cyclones from North East 
 22 to 24 
 
 of wind and sea which clearly heralded the coming storm. This 
 was intensified on March llth, (Chart 23) by a further fall of 
 three-tenths of an inch, and three more isobars of the cyclone were 
 marked on the coast south to south-east ; gales prevailed in its 
 neighbourhood with very heavy rains. 
 
 On March 12th, (Chart 24) the anticyclone had retreated to 
 New Zealand, and the cyclone travelled to the south along the 
 coast intensifying as it came. The barometer curve at Brisbane 
 was exactly of the cyclone type and dropped to 29 '5. In the 
 southern part of Queensland and northern of New South Wales 
 heavy southerly to easterly gales were experienced. On the 13th 
 the storm had disappeared to the eastward. 
 
84 
 
 AUSTRALIAN WEATHER. 
 
 TYPES 
 
 OF 
 
 AUSTRALIAN WEATHER 
 
 H 
 
 TYPES 
 AUSTRALIAN WEATHER & 
 
 TYPE XI. SOUTH-EAST GALES. 
 
 The south-east gale here referred to is peculiar to the east coast 
 of Australia, and it has been responsible for tl.e most memorable 
 
TYPES. 
 
 85 
 
 wrecks on this coast. For the most part these gales appear to be 
 partially spent cyclones, which come in from north-east or east, 
 and travel down the coast until they begin to recurve to the 
 eastward. 
 
 The warning of their coming is usually very short ; it consists 
 of a sudden increase in the sea on some northern part of the coast 
 with wind from east to south, and falling barometers, while the 
 high pressure over Victoria and South Australia becomes inten- 
 sified and progresses into the Tasman Sea. The south-east circula- 
 tion about this anticyclone increases in force with the increasing 
 barometric grade, and also by the wind circulation about the 
 cyclone, and the effect of the two causes acting together is to 
 produce a most serious gale. Rarely, these storms originate in 
 a monsooiial depression somewhere over South Australia, which 
 travelling eastward intensifies on the east coast. Heavy rain is 
 a marked feature of these storms, but it is confined to the coast, 
 and rarely if ever extends inland. 
 
 The storm selected to illustrate this type was a very severe one, 
 and began on September 23rd, 1892, about 6 p.m. The barometric 
 conditions antecedent to it are shown in Chart 25 ; the main 
 
 TYPES 
 
 AUSTRALIAN WEATHER 
 N25 
 
86 
 
 AUSTRALIAN WEATHER. 
 
 features being the bend in the high pressure isobars, and the 
 dormant low pressure off the coast of Queensland. 
 
 At 9 a.m. on this day there was nothing in the local weather 
 conditions which would lead one to anticipate the gale that 
 eventuated, the winds being generally light, and at Sydney only 
 a light breeze was blowing from the South-south-west ; but at 3 
 p.m. an unusual fall took place in the barometers to the north-east 
 of New South Wales, and the winds there were freshening gener- 
 ally. On preparing a chart at this hour we found the depression 
 was intensifying and had a cyclonic tendency ; at 6 p.m. in 
 Sydney the wind, which had been blowing from the south-east 
 and gradually increasing in velocity, reached the force of a gale, 
 a thick driving rain began to fall, which continued with little inter- 
 mission until daybreak next day, when over 2" were registered, 
 and an extensive area around the metropolis benefitted to the 
 extent of an inch and upwards ; the barometer at this hour also 
 began to fall rapidly and steadily, until at 5 a.m. on the 24th it 
 read 29*203, and the wind had reached in one squall, lasting only 
 a few seconds, the extraordinary rate of one hundred and twenty 
 miles per hour, the mean rate of the gale being thirty-two miles 
 per hour. 
 
 TYPES 
 AUSTRALIAN WEATHER 
 
 N26 
 &*. SEPTEMBER 
 
TYPES. 87 
 
 At 9 a.m. on the 24th, (Chart 26) the cyclone was still in an 
 active state, but it had passed to the south of Sydney and was 
 receding from the coast; the barometers were rising rapidly on shore, 
 and before noon a light northerly wind was blowing. On the 25th 
 September, the day following the gale, the weather everywhere 
 was generally fine, while all thab remained of the energetic high 
 and low pressure systems were parallel isobars lying over the 
 southern areas of Australia with very shallow gradients. This 
 gale was very destructive and did much damage to property in 
 Sydney; in some instances houses were unroofed, and the wind 
 and sea on the coast were very heavy. 
 
 TYPE XII. DEVELOPMENT OF A CYCLONE FROM A A DEPRESSION. 
 
 The distinction between a \ depression and a monsoonal low 
 pressure is not by any means well defined, and it is possible that 
 this should be taken as a variation of Type 3; there are however, 
 marked differences, not only in the shape of the isobars, but also 
 in the wind ; and the most decided distinction is perhaps the 
 easterly wind circulating round the southern part of the monsoonal 
 low pressure and the northerly and southerly winds about the \ ; 
 but in some cases, as in the one selected, the wind circulation is 
 mixed, northerly, southerly and easterly winds being present, and 
 these from their want of energy tend to throw the forecaster off 
 his guard. The season is, however, some guide, as these storms 
 are most frequent from September to April. 
 
 Their sphere of influence is very extensive, as may be noted in 
 Chart 28, which shows rain over half Australia as the apparent 
 result of this storm. Most of these storms take a direct easterly 
 course over New South Wales and Victoria or through Bass' 
 Straits. At times they move to north-east, the Polar winds being 
 more energetic, and this feature intensifies all the storm and rain 
 conditions. The winds in all these storms are violent, and in 
 some very destructive. 
 
 The one selected for illustration appeared first on 27th May, 
 1893, (Chart 27). At 9 a.m. on that day, a dormant and irregular 
 
88 
 
 AUSTRALIAN WEATHER. 
 
 A depression existed over the Australian Bight, the winds were 
 moderate northerly and southerly, as usual in such conditions, 
 there were a few light easterly winds. Isobars were close over 
 eastern Victoria, but there was nothing which seemed to indicate 
 the violent cyclone that developed during that day over southern 
 parts of South Australia and south-west of New South Wales. 
 (See Charts 27 and 28.) This storm formed in the rear of a very 
 substantial anticyclone, then over the Tasman Sea, and it should 
 be noted that it did not act as a secondary and travel round the 
 southern and eastern parts of the high pressure, but it moved 
 towards the northern side of it and against its circulation, thus 
 proving its own Polar impulse and energy, and giving rise to very 
 strong gales and steep barometric grades with great fall of 
 temperatures ; these conditions produced extremely heavy and 
 wide spread rains, not only within the storm isobars, but over the 
 whole of the eastern half of Australia. We have no means of 
 tracing the rain to the west of the overland telegraph line, because 
 there are no observing stations there. 
 
 May 28th was unfortunately a Sunday, and we have no obser- 
 vations for that day, but on the 29th the cyclone is seen in full 
 
 T YPES 
 
 AUSTRALIAN WEATHER 
 N27 
 
 HIC 
 
 Cyclone from A depression 
 27 6* 28 
 
 29-7 
 
TYPES. 
 
 89 
 
 TYPES 
 
 OP 
 
 AUSTRALIAN WEATHER 
 
 . N28 
 MAY S3 xe 0a 
 
 29-9 50O 50 I 
 
 strength covering the southern colonies of South Australia and 
 Victoria, and the greater part of New South Wales ; its isobars 
 are unusually symmetrical and its rain influence the most exten- 
 sive we have on rpcord. During the 29th and 30th rain continued 
 to fall over considerable areas, although on the 30th the depression 
 filled up and the storm was displaced by the high pressure coming 
 on from the west. 
 
 TYPE XIII. WESTERLY WINDS. 
 
 The Winter anticyclone (See Chart 36) is much nore extensive 
 than the Summer one, and its grades are steeper and circulation 
 stronger, while its latitude is further north, often up to 30 S.; 
 hence the circulation on its southern side, unlike the Summer one, 
 affects the mainland of Australia and gives us our westerly winds. 
 And just as the trade wind intensifies its northern circulation by 
 adding force thereto, so the brave west winds of the southern 
 ocean follow the general move of the weather systems northwards, 
 and thus add force to the westerly circulation of the anticyclone, 
 and its greater dimensions increase the size of the \ depression, 
 so much that it no longer has the sharply defined change from 
 
90 
 
 AUSTRALIAN WEATHER. 
 
 notherly to southerly winds, but from north-west to south-west 
 winds ; all these conditions combine to strengthen the westerly 
 circulation, until at times they seem to be like the brave west 
 winds of the roaring forties. 
 
 On the east coast they extend further north than they do inland, 
 and at times include Brisbane within their influence. When 
 very strong the anticyclone is elongated and the southern isobars 
 are flattened. (See Chart 30). They are very cold and dry, 
 having all the raw feeling of the easterly winds of England, and 
 have been known to last for several weeks without intermission ; 
 they are the most persistent winds during our winter. In heavy 
 westerly winds the cold is very severe in Bass' Straits, between 
 the south coast of Victoria and Tasmania, rain, hail, and sleet being 
 very common. It is almost needless to say that on land they 
 bring very little rain, and they rapidly dry up the soil. The 
 severe drought of 1895 was largely due to the persistent westerly 
 winds which rapidly dried up the occasional rains; this feature 
 becomes intensified if instead of coming from due west they blow 
 from north-west. 
 
 TYPES 
 
 AUSTRALIAN WEATHER 
 
 N29 
 SEPTEMBER Z. ie 
 
 Westerly Gale 
 29 & 30. 
 
TYPES. 
 
 91 
 
 Turning now to the charts of the westerly gale selected for 
 illustration, it will be seen in Chart 29, that this storm began on 
 September 3rd, 1895; on that day an elongated anticyclone lay 
 over Western Australia, a flattened and extensive \ over New 
 South Wales and Tasman Sea, and the winds generally displayed 
 great energy, as might be expected from the close isobars, and 
 unusually low barometers over Tasmania ; light rain was falling 
 on the coasts of South Australia and Victoria. 
 
 On September 4th, Chart 30, the anticyclone is more elongated 
 and the \ flattened until its isobars are nearly horizontal, and 
 heavy westerly gales swept all the south-eastern part of Australia 
 and all Tasmania. On the 3rd the wind at Sydney at noon for a 
 short time reached a velocity of seventy-eight miles per hour. On 
 the 4th the wind was less gusty, but its average velocity was quite 
 as strong as it was on the 3rd. 
 
 TYPES 
 
 AUSTRALIAN WEATHER 
 N30 
 
 TYPE XIV. SOUTHERLY BURSTERS. 
 
 The southerly burster is a well known feature or type of Austra- 
 lian weather, so well marked in character indeed that it requires 
 no special training in meteorology to recognise it ; its character- 
 
92 AUSTRALIAN WEATHER. 
 
 istics are so obtrusive that they cannot be overlooked, and they 
 are welcomed as the most pleasant relief after the high tempera- 
 tures and oppressive northerly winds which precede them. They 
 come in the late spring, all summer, and part of autumn, but as 
 a rule, in order to get a strong southerly an antecedent excessively 
 hot day or days must be experienced. The duration of a southerly 
 burster may be anything from two hours to ten days, but it is not 
 to be understood that the term burster is applied to the whole 
 period or duration of the southerly wind. What is called a 
 burster is the squall or sudden and violent change of wind direc- 
 tion, and the violent rush or " burst " which marks the advent of 
 this wind. We need not go into all the characteristics; these will 
 be found in the Abercromby Prize Essay on this subject. 1 It is 
 there explained that the south wind comes in front of an approach- 
 ing anticyclone, and that it is felt from West to Eastern Australia, 
 but it is only on the eastern coast, where, aided by the smaller 
 friction of the ocean and the shelter which the mountains afford 
 from other winds, that the southerly becomes more vigorous and 
 rushes northwards in a squall, which happens so suddenly and 
 with such force, that at times ships drag their anchors in Sydney 
 harbour. 
 
 It is not definitely made out yet that these storms are ever 
 "line storms " in the sense that the change of wind comes as the 
 dip in the isobars passes over each place in succession, but there 
 are many facts which suggest that such is the fact in some instances. 
 Our present purpose is to describe a " burster" as a type of 
 Australian weather. The essential feature of it is a sharp 
 A ; in Chart 31 such a depression is shewn existing over Victoria 
 and Tasmania, with its axis lying from north-north-west to south- 
 south-east. An anticyclone of good energy for this time of the 
 year exists to the west, and hot northerly winds occupy northern 
 Australia ; these are the elements for the good burster that 
 followed. As a general rule, 2 the position and character of two 
 
 l Journal Royal Society, N. S. Wales, Vol. XXYIII. 
 2 See Moving Anticyclones, p. 4. 
 
TYPES. 
 
 93 
 
 TYPES 
 
 AUSTRALIAN WEATHER 
 N3i 
 
 such systems as shown would bring the burster to the coast of 
 New South Wales within twenty-four hours; in this case it took 
 thirteen hours. 
 
 November 16th, light to fresh north-east winds v/ere blowing 
 on the coast at 9 a.m., while in the front of the high pressure 
 strong south-west winds were bio wing. In Chart 32, the southerly 
 
 TYPES. 
 
 OP 
 
 AUSTRALIAN WEATHER . 
 N32 
 
AUSTRALIAN WEATHER. 
 
 is shown in full force in front of the anticyclone, which by the 
 way has lost none of its energy since the previous day ; the 
 depression is well off the coast and on its way to New Zealand. 
 This burster reached a velocity of forty-nine miles per hour, and 
 lasted thirty-five hours. 
 
 TYPE XV. THE BLACK NORTH-EASTER. 
 
 This is a somewhat uncommon but nevertheless well known 
 type of weather on the coast about Sydney. Its characteristics 
 are a very strong and persistent north-east gale, continuing day 
 and night for two or more days ; it has been known to last five 
 days and nights, and it ends with the advent of a southerly 
 burster. Its cause is found in an extensive col, the rear of one 
 anticyclone being at rest over this coast, while another lies over the 
 Australian Bight. If the grade is rather steep and the system 
 at rest for several days, then the north-east wind persists with 
 force proportionated to the grade, until the whole system moves 
 forward ; the southerly winds in the front of the approaching high 
 pressure then displace the north-easter and the storm is over. 
 
 There have been no good examples of this type since weather 
 charts have been printed here. Chart 33 shows the necessary 
 forms of isobars, but the grade is not steep enough for a gale. 
 
 TYPES 
 AUSTRALIAN WEATHER 
 
TYPES. 
 
 95 
 
 TYPE XVI. WIND BLOWING CONTRARY TO ISOBARS. 
 
 In this type the wind blows with considerable force in a direc- 
 tion directly opposed to that which the isobars would lead us to 
 expect. For instance in No. 34, it will be seen that an extensive 
 high pressure rests over the east coast, and the isobars are com- 
 paratively close together. The normal circulation with these 
 isobars would be fresh north to north-west winds, when, as a 
 matter of 'fact, strong southerly winds were blowing as far as 
 Sydney with a velocity of thirty miles per hour. Such conditions 
 are rather troublesome in forecasting ; fortunately they do not 
 come often, and the fact is not confined to southerly winds. The 
 general direction of the coast line is northerly bearing east a little, 
 a range of mountains from two to four thousand feet high 
 runs nearly paralled to it, and this local formation has a very 
 important effect on the circulation of the wind ; as in Chart 34 
 it seems to have more effect than the isobars, and probably the 
 grade was rapidly intensifying and had not been long enough in 
 existence to fully control the winds. It seems probable, so far as 
 this type has been studied, that we should find that when the 
 wind blows contrary to isobars, it does so because of some impulse 
 given to it before the new grade had time to control the circulation. 
 
 
96 
 
 AUSTRALIAN WEATHER. 
 
 TYPE XVII. - SUMMER ANTICYCLONES. 
 
 One of the most marked features of weather in Australia is the 
 regular easterly motion of it in all seasons of the year, added to 
 this the anticyclones the controlling element in our weather 
 change with the sun, so that the latitude they follow in their 
 easterly progress is further south in summer than in winter. 1 The 
 summer latitude is about 40 south. They are less extensive in 
 summer than in winter, and do not so completely control our 
 weather as they do in winter, for their southern position leaves 
 room for the southerly extension of monsoonal low pressures, 
 which make a great deal of our weather ; but the anticyclone 
 makes the change from hot northerly to cool southerly winds, the 
 bursters of the east coast. 
 
 TYPES 
 AUSTRALIAN WEATHER 
 
 N35 
 JANUARY 1.6... 
 
 TYPE XVIII. WINTER ANTICYCLONES. 
 
 During the winter months the antiyclones are much larger than 
 they are in summer, and their latitude about 30 S.; very com- 
 monly their area is equal to Australia, and their control of the 
 weather more complete than it is in summer. Fine weather 
 
 1 See Moving Anticyclones, page 2. 
 
TYPES. 
 
 97 
 
 marks their centres, and the rains come chiefly from the low 
 pressures between each pair of anticyclones, and the strong 
 westerly gales are but part of the circulation about these high 
 pressures. (See Charts 29 and 30.) 
 
 The one selected for illustration occupied the whole of Australia 
 on June 4th, 1895, Chart 36. Its form was remarkably sym- 
 metrical, and in the central area the barometers read 30 '6, which 
 is somewhat unusual, hence the circulation is active, and in 
 northern Australia where the trade wind adds force, it is very 
 strong. Under the central influence of these great anticyclones 
 the whole of Australia enjoys fine weather. 
 
 TYPES 
 AUSTRALIAN WEATHER 
 
 N36 
 
 TYPE XIX. SQUARE HEADED A DEPRESSION. 
 
 This type is a variation of the usual \ depression, but is 
 sufficiently characteristic to be placed by itself as a type or rather 
 a sub-type. Its isobaric peculiarity, as may be seen by refer- 
 ence to Charts 37 and 38, is that there is a flat top or square 
 head to the isobars of the A > the usual form of which is a sharp 
 and regular curve, and the marked feature of this is that under 
 it the weather is remarkably squally and charged with thunder 
 
98 
 
 AUSTRALIAN WEATHER. 
 
 and hail storms, and deluges of rain. This type is not peculiar to 
 any season^of the year, and the country they affect is usually 
 south of Lat. 30. The particular square headed \ depression 
 under discussion occurred in July, 1891. Its effects were wide- 
 spread and its life persistent, and its peculiarity in a more or less 
 distinctive form was maintained during its passage over six degrees 
 of longitude, and its effects most severe in Victoria. 
 
 In Chart 37, July 8th, 1891, the square headed A is shown 
 when over South Australia. The gradients about it excepting 
 those on its eastern side were moderate; winds were from fresh 
 to strong. Following it is a high pressure of decided energy, and 
 preceding it to the east an anticyclone of little or no character ; 
 on the previous day rain fell over greater part of the southern 
 seaboard and in Tasmania. 
 
 TYPES 
 AUSTRALIAN WEATHER 
 
 N. . 57 
 
 r striae a.. , 
 
 Square-Headed / depression 
 37 & 30 
 
 Chart 38, July 9th, the depression has become more extensive, 
 but has lost energy. The relative powers of the anticyclone 
 remains the same, winds were generally lighter, but the rain 
 covered a much wider area than on the previous day. As this 
 storm movecl to the east its characteristic weather was maintained. 
 
TYPES. 
 
 99 
 
 TYPES 
 AUSTRALIAN WEATHER 
 
 TYPE XX. THE ADVENT OP AN ANTARCTIC STORM. 
 
 This is a type of weather that does not often visit Australia 
 but its severity makes it noteworthy for the winds and weather 
 which come with it are very destructive, and the cold severe, it 
 might almost be called a southern blizzard ; yet the warning is 
 short and often difficult to read, for it comes from the Antarctic 
 where we have no out stations. 
 
 The storm under consideration began to affect the south coast 
 of Australia, making the winds fresh to strong on June 21st, 
 1892, (Chart 39), yet the season (winter) and the general conditions 
 pointed to westerly winds, and said nothing definite of the storm 
 which was telegraphed on the morning of June 22nd, (Chart 40). 
 Barometers were then seen to be four-tenths lower in Tasmania 
 than they were on the 2 1st, and the wind had increased to a 
 furious gale along the south coast of South Australia and Victoria, 
 and the isobars indicate a very steep grade commensurate with 
 the wind. On shore also the wind rose in places to hurricane 
 force ; at Ballarat, in western Victoria, buildings were unroofed 
 and trees blown down. So severe was it there that the storm is 
 recorded as the Ballarat storm. 
 
100 
 
 AUSTRALIAN WEATHER. 
 
 TYPES 
 
 OP 
 
 AUSTRALIAN WEATHER. 
 
 N.-39 
 JUNE 21 ie9* 
 
 Aduenf of an Antarctic Sfortn 
 59 A. 40 
 
 In South Australia and New South Wales the wind did a great 
 deal of damage without reaching the intensity it had in Victoria. 
 The isobars show a retreat of the anticyclone and a tilting of the 
 major axis, and a very remarkable increase in the number of 
 isobars 011 the south coasts of South Australia, Victoria, and 
 
 TYPES 
 
 AUSTRALIAN WEATHER 
 N._40 
 
 (20 
 
 '50 
 
 LOW 
 
TYPES. 101 
 
 Tasmania. It looks as if the low pressure had retreated and its 
 western parts forced their way north. What probably did take 
 place was that a storm centre south of the Australian Bight and 
 indicated by the northerly winds in Chart 39, had in the interval 
 surged northwards on to the west coast of Victoria, bringing with 
 it all its antarctic energy and severe cold. This view is sup- 
 ported by the fact that there was in the twenty-four hours but 
 little change in the New Zealand isobars, and further by the 
 upward tilting of the eastern part of the anticyclone caused by 
 the northing of the antarctic stown, and lastly by the blizzard- 
 like cold which was so marked a feature of this storm. 
 
 LIST OP TYPES OF AUSTRALIAN WEATHER. 
 
 I. Moving Anticyclones, Charts 1, 2, 3. 
 II. Monsoonal Rain Storm, Charts 4, 5. 
 III. Development of a Cyclonic Storm in Low Latitudes 
 
 from a Monsoonal Depression, Charts 6, 7. 
 IV. Development of a Cyclonic Storm in High Latitudes 
 
 from a Monsoonal Depression, Charts 8, 9, 10. 
 V. Conditions favourable for Thunderstorms, Charts 11, 12. 
 VI. Cyclonic Thunderstorms, Charts 13, 14. 
 VII. Vertical and nearly straight Isobars, Charts 15, 16. 
 VIII. Cyclones from North- West, Charts 17, 18, 19. 
 IX. Cyclones from North-East, Charts 20, 21, 22. 
 X. Tornadoes, Charts 23, 24. 
 XL South-East Gales, Charts 25, 26. 
 XII. Development of Cyclones from a A Depression, Charts 
 
 27, 28. 
 
 XIII. Westerly Winds, Charts 29, 30. 
 XIV. Southerly Bursters, Charts 31, 32. 
 
 XV. Black North-Easter, Chart 33. 
 XVI. Winds Blowing Against Isobars, Chart 34. 
 XVII. Summer Anticyclone, Chart 35. 
 XVIII. Winter Anticyclone, Chart 36. 
 
 XIX. Square Headed A Depression, Charts 37, 38. 
 XX. Advent of an Antarctic Storm, Charts 39, 40. 
 
F. W. WHITE, PRINTER, 39 MARKET STREET. 
 
OF 
 
 WVBRSIT7, 
 
Abercromby "Southerly Bursters." 
 
Abercromby "Southerly Bursters." 
 
 Plate II. 
 
JHI7ERSIT7 
 
Abercromby " Southerly Bursters." 
 
 Plate III. 
 

Abercromby " Southerly Bursters." 
 
 Plate IV. 
 
UNIVERSITY OF CALIFORNIA LIBRARY, 
 BERKELEY 
 
 THIS BOOK IS DUE ON THE LAST DATE 
 STAMPED BELOW 
 
 Books not returned on time are subject to a fine of 
 50c per volume after the third day overdue, increasing 
 to $1.00 per volume after the sixth day. Books not in 
 demand may be renewed if application is made before 
 expiration of loan period. 
 
 OCT 6 
 
 lOm-4,'23 
 
Three e.ssa; 
 II an weatn 
 
 !87 
 
 ) * 
 
 /s on Austra 
 
 sr. 
 
 A7A3 
 
 UNIVERSITY OF CALIFORNIA LIBRARY 
 
 -