UC-NRLF *c 3fl aoo Digitized by the Internet Archive in 2007 with funding from IVIicrosoft Corporation http://www.archive.org/details/discussionofaustOOsolarich SOLAR riTYSICS COMMITTEE. - 1276^ - A DISCUSSION OF AUSTRALIAN METEOROLOGY Being a Study of the Pressure, Rainfall and River Changes, both Seasonal and from Year to Year, together with a Comparison of the Air Movements over Australia with those over South Africa and South America. BY WILLIAM J. S. LOCKYER, MA. (Cantab.), Ph.D. (Gottingen), F.R.A.S., Chief Assistant, Solar Physics Observatory. UNDER THK DIRECTION OF Sir NORMAN LOCKYER, K.C.B., LL.D., ScD., F.R.S. LICK OBSERVATORY LIBRARY UNIVERSITY OF CnUfOR^NIA j;\.l 13 1S58 LONDON: PRINTED FOR HIS MAJESTir'S STATIONERY OFFICE, By eyre and SPOTTISWOODE, Ltd., PRINTERS TO THE KING's MOST EXCELLEXT MAJESTY. And to be ])iircliase(l, either directly or through any Bookseller, from WYMAN Axi) SONS, Ltd., Fetter Lane, E.C. ; or OLIVER ANU BOYD, Tweeddale Court, EoixBURciii ; or E. PONSONBY, 116, Giiaftox Street, Dublin. 1909. [Price Five Shillings.] LIST OF PUBLICATIONS OF THE SOLAR PHYSICS COMMITTEE (FiH)M 1879 TO 1908). Yciir of Publication. Preliniinaiy Re])oif of the Committee on Soliir Physics - - . - . (1880) Report by the Committee on Soiiir Physics -..-■--- (1882) iSecoud Report by the Committee on Solar Pliysics .... . (1889) Positions and Areas of Sun Spots and 1>' acute, 1887 and 1888 .... (1889) 1889 .... . (1891) „ „ ,, „ „ 1871-1881 (inclusive) and 1890 - - (1892) , , 1891 . _ - . . (1894) 1892 ..... (1895) „ V V „ ., 1893 .... . (1896) „ „ „ „ „ 1894, 1895, 1896, and 1897 - - (1899) Spectra of Sun Spots, 1879-1897 (inclusive) . - . . . - (1900) Positions and Areas of Sun Spots and Facuhv, 1898 and 1899 .... (1901) Catalogue of 470 of the Hri '^ Fig. 3. mean tracks or belts of the centres of the winter and smiimer "highs'' are repre- sented by continuous and dotted lines respectively. The directions of the winter and summer " lows " are also shown, the former travelling from west to east, while the latter recurve from a south-easterly to a north-westerly course, and reach the north-east coast as " tropical north-easters," and the north-west coast as " tropical north-westers." While dealing with the subject of the general features of xVustralian meteorology, and having previously indicated that the mean annual pressure over the Avhole continent varies very considerably from one year to another, it is important to find oiit the chief characteristic features which have been noted during these exceptional years of high and low pressure. E 142. C 18 SOLAR PHYSICS COM:\riTTEE. For the purpose of this inquiry the annual vohimes of the " Meteorological " Observations made at the Adelaide Observatory and other Places in Soiith Australia " and the Northern Territory " have been utilised. An attempt was tirst made to take the years of exceptionally high and low pressure, and extract statements which would indicate the general conditions of the anticyclones or cyclones during these years. Such characteristic features as were sought for were not found recorded in the earlier volumes, the remarks being confined chiefly to more detailed barometric changes. The desired statements were, however, found in the report for the year 1891, a year of exceptionally high pressure. One may therefore assiime that the experi- ences recorded in this year may be considered as typical of all high-pressure years. The extracts from the Monthly Summaries are as follows, the most important points being printed in italics : — Page 13. February. " As regards the weather generally during the month, the extreme dryness was doubtless due to the high pressure which prevailed. The Colony seems to have been protected, as it were, from the influence of south coast disturbances hy a belt of high pressure covering the Bight and southern parts of the Colony, At times the barometer within this belt read as high as SO' 2 and 30 "3 inches, but even whmi they were only relatively high (as 30 "0 inches) it offered an impenetrahle barrier to the low-pressure energy in the south . . . ." Page 29. April. " All along the southern seaboard of the continent there has been this month an almost unbroken continuance of high-pressure conditions. Anticyclonic ' centres ' passed, from time to time, eastward fi-om the Great Bight to New South Wales or Tasmania . . ." " At the end of the mouth the maximum pressure was reached, a vast mound or mountain of pressure being steadily built up diiring the last few days reaching its maximum on the last day, lohen the lohole of Australia, Tasmania, the Southern Ocean, the Pacific Ocean nearly to 'New Zealand, and the Indian Ocean to the west of the continent (or some 50° of longitude and 35'^ of latitude) loere enclosed within the isobars of one vast high-pressure system." Page 37. May. " May was characterised by intense and prolonged Iiigh- pressure conditions. ..." " This month the dry area eiubraced the whole of the Colony. . . ." " This anticyclonic belt, or ridge of higli barometers over the southern coast line of the continent, lasted nearly all through May, and, as is usually the case, acted as an effectual barrier to low-pressure extensions from the Southern Ocean . . ." "The maximum point the pressure reached was 30*664 inches on the ist, when the vast high-pressure system, which steadily formed during the last few days of April, still covered the continent." A DISCUSSION OF AUSTIiALIAN ilETEOROLOGY. 19 Page 53. July. " Tlie character of the weather charts then completely altered — the hirge antic y clonic sijfttems, which were so marked a feature of the previous months, again covering the continent." Page 61. August. "The most marked feature of the month was again its dryness and exceptionally high, barometric conditions." " Looking through the daily weather charts for the month, the most noticeable feature is the large numher and extent of the anticyclonio systems which passed over the continent -at times the whole continent and the surroundhig oceans, even Including New Zealand, being enclosed within the Isobar of one vast high-pressure system — within the central area of which the barometer read as high as 30 '5 and 30 "6 inches. Those large systems have been a striking feature of the weather charts all through this xointer season, and special attention is called to the circumstance as associated with the exceptional character of the season, as regards the small, variable, and light rainfall, especially over the southern portions of the Colony." Page 69. September. " Continued dry weather was the chief feature of the month ..." " The pressure was very high through the latter part (or during the last ten days) of the month. . . ." Page 85. November. " November was dry with rather high barometric pressure . " . . . During the remainder of the month, a high-pressure system gradually formed over the Southern Ocean, extending over- the greater part of the continent . Page 94. Summary for the year. "As far as the southern districts are concerned, this year will rank amongst the driest years experienced . . ." " One striking feature of the year was the absence of general soaking rains." From the above exti-acts it will be gathered that the cliief characteristics which can be attributed to this high-pressure year are : — 1. Greater size of anticyclonio systems. 2. Fewer number of low-pressure areas. 3. Great deficiency of rainfall. By plotting the values for the daily means of pressure for Adelaide, day by day, for several of the high-pressure months referred to above, it is observed that the progressive movements of the anticyclones from West to East is still in evidence, though their motion seems to be slower. It does occur sometimes that C 2 20 SOLAR I'HYSICS CO.M.M ITTEE. the anticyclones move very slowlj- and l^econie even stationary ; and this fact was clearly stated by Russell* in the followi])g words : — " Another feature brought out in the diagrams is the occasional stoppage of the anticj^clones. . . . Out of a total of 42 anticyclones which passed over Australia in 1891, 6 or 15 per cent, hesitated or actually stopped in their forward motion." It is interesting to note that the year 1891 was one of excessively high pressure, so that such an occurrence of this hesitation or stoppage may be a feature of the anticyclones during these epochs. It Avill be shown, however, later on (page 29), that the number of anticyclones per year does not seem to alter juuch, and Russell also came to the same conclusion. One may conclude, then, that such years as 1877, 1881, 1885, 1888, and 1891, Avhicli were years of very excessive pressure, were caused by the anticyclones having altered their normal condition and become, on the average, rcqi much larger. During these epochs they presented very formidable barriers to the inroads of the low-pressure systems, both on their northern and southern boundaries, and thus prevented these rain-bearing systems from watering the country. High- pressure years should therefore be years of deficient rainfall over the icliole of Australia. If, now, the Adelaide Annual Meteorological Reports be examined for those years when Aiistralia was undergoing a great deficiency of pressure, it will l)e found that such large anticyclonic systems, as have been referred to above, are more the exception than the rule, and that the Colony is frequently swept by the low-pressure areas, as will be shown in a subsequent chapter (page 29). As low^-pressure areas aj-e closely associated with rainfall, low-pressure years should be wet years for the whole of the Colony. It is possible that the above statement, associating high-pressure years with years of deficient rainfall over the whole of Australia, may require to be slightly modified when more rainfall statistics over a greater area of Australia become available. Only a small amount of rainfall data can at present be utilised for the important and large State of Queensland, since they are as yet unpublished. With regard to the year (1891) in question, Mr. Hunt has informed me that — 1891 was not a year of marked deficiency of rainfall in the Eastern States, and although the absence of remarkably heavy daily falls in New South Wales was conspicuous in South Australia, yet, in Queensland, Ayr had 10 '19 inches on March 25th, Burketown 13 '58 inches on June 15th, Cairno 14 '08 inches on * Quart. Jour. U. Met. Soc, Vol. XIX., No. 85, January 1893, page 26. A DISCUSSION OF AUSTRALIAN .AIP:TF.01;0U0(;Y. '21 April -Jtli, JJouaiclsoii 11'2'J iiiclies on .lauuary 27th, Uloncurry lU'o3 Indies on .7ainiary 23r(I, Townsville lO-fil inches on :\hirch 2St]i, and ^lackay lO-^O inches in Marcli." In til is memoir tlie Queensland rainfall statistics which have heen employed extend only from 1895 to 1906, the only data which could be obtained from Australia, so that no discussion has l)een possible with respect to the previous years, including 1891. These when properly grouped together and compared with the pressure changes, as is done on page 48, indicate clearly that the low- pressure years are years of increased rainfall and vice versa in Queensland. Chapter TIL THE MEAN ANNUAL PRESSURE VARIATIONS. A DISCrsSlOX OK AliSTl.'Al.lAX :\IF;rE()I!()T/)( 1 Y. THE MEAN ANNUAL PRESSURE VARIATIONS. After the preliuiiiiaiy remarks, given in the last chapter, concerning the general atmospheric movements which take place over Australia, the curves representing the mean annual pressure variations can be better interpreted. In order to select stations well distributed over the land area, and which present satisfactory series of observations, the following places given in the accom- panying table have been chosen. The respective columns show the States and Districts, the names and latitudes of the stations used, and, linally, the number of vears over which the observatifuis here utilised extend : — state. South Austialia West Austnilia South Austiiiliii West Aiistiiiliii New Soiitli Wale? South Australia Victoria - District. Station. N. N. M. N.W. Central W. W. s.w. 8.W. Coast S. Port Darwin Wymlhani Derby Cossack Alice S])riu«;s Carnarvou Geralilton Perth Bunbury Sydney Adehiifle Melbourue - Lat. S. No. (if Veal's. ( 12 28 24 1.5 20 9 17 20 11 20 40 1 23 38 23-24 24 50 11 28 40 13 31 57 15 33 20 15 33 52 28 34 57 41 37 50 43 The curves representing the mean annual variations for these stations are shown in Fig. 4, the values from which they have been drawn being given in the Appendix (Table 3). The main feature of these curves, which is common to them all aud to which attention is first drawn, is that all the curves have one principal mininmm. For the higher latitudes this miniminn occurs in December, while, as the equator is approached, it is more distinctly pronoiuiced in Januaiy. The second prominent trait is the double maximum in April and July in the stations which have high latitudes, and the single maximum in July in the lower latitudes. It is worthy of notice to observe the gradual change from the double to the single maximum as the more equatorial regions are reached, and also the less rapid descent after July of the more southern stations. The form and changes of type of these curves seem to be simply and efhciently explained by the change of declination of the sun throughout the year. In the Australian region of the world, places which have the sun in the zenith at noon are dominated by a region of low pressure, a region in which the mousoonal "lows" occur. Bordering on this low-pressure region and on the southern side of it are found the anticyclones which pursue their west to east E U2. D 26 soT.Ait I'livsirs ^o^r:\r^TF.E. PRESSURE INCHES 299 12-28 5. 29-7 DERBV i7*»9'a 29-9 298 29^7 3(>(H ALICE SPRINGS^^^ 23'4os. 29-8 30-1 300-- GERALDTON 2g9- 28*46 3. 301 30-0 299 301 ADELAIDE ^^° 34*57 5. 299 BUNBURV 33'20S. NCHES Fig. 4. 299 2^^ WVNDHAM 297 \5'20'e: 29^9 ^^^ COSSACK 297 20*40'5. ^^^CARNARVON 29-8 24* 40" 5. 30-1 '^^ PERTH 29-9 31*57 s. SYDNEY 33'52S. 2^9 298 300 ^^^ MELBOURNE 29 8 ^7'^°'^- course. Any movement in latitude of the low-pressure region under the sun must, therefore, influence the positions, also in latitude, of the anticyclonic areas to the soTithward of it. A T)ISr[-SSI()X OK AI'S'l'ltATJAX MKTE< )|J()I/)(! Y In tlie course of a year the sim has its greatest northern declination on June 21 and its greatest southern declination on December 22. Let us consider for a moment this change of the sim's position in latitude with reference to the annual pressuiv changes of the lower latitude Australian stations. Taking the Australian midsummer condition first, namely, that when the sun has its greatest southern declination, Decem1)er 22, the low-pressure region under the sun will have its most southern position, and consec|uently the anticyclonic region bordering its southern side will likewise be furthest south from the equator. About December, therefore, the stations in North Australia should indicate their lowest seasonal pressure, as at this time they are nearest the centre of the low-pressure region and furthest from the centre of the anticyclonic area. As a matter of fact, the pressure curves indicate their lowest values in January, or a month late. The delay in the occurrence of the maximum is of the same length. If now the Australian mid-winter be considered, this epoch being when the sun has his greatest northern latitude, that is, June 21, the low-pressure ai"ea under the sun will be in the northern hemisphere. The path of the southern anticyclones will now lie in lower southern latitudes, i.e., nearer the equator. At this season of the year stations in North Australia wall be nearer the centre of the anticyclones and furthest away from the low-pressure area under the sun, so that in this month they should indicate their highest pressure during the year. The month of highest pressure for these stations is July, that is, a month late, or a similar lag to that of the lowest pressure. Turning attention now to the mean annual pressvire curves of Australian stations in higher latitudes, these show maxima during a year in Aj)ril and July and a principal minimum in December and January. It has been shown aliove that in December the sun has its greatest southern declination and that the low-pressure region under it reaches the northern part ol: Australia, while the centres of the antic\^clones pass along from w^est to east, skirting the southern shore of the continent. About this time the wdiole of Australia is undergoing a period of relative low barometer. As the sun begins to decrease his southern declination and the track of the centres of the anticyclones consequently moves to lower latitudes, all places in Australia experience a rapid rise in pressure. When the month of April is reached the mean position of the centi-e of the anticyclonic track is about latitude 28° S., and all places have up to then been increasing their mean monthly values of pressure. From April, places in lower latitudes than 28° S. will continue to show an increase of pressure because the centre of the anticyclonic area is still advancing towards them, but those in higher latitudes will display a decrease because they are now to the southwards of the track of the centres of the high- pressure areas which is retreating from them. The southern Australian stations D 2 28 S0T>A1I PHYSICS CO.MilLTTEE. will therefore indicate a maximum when the centre of the antic^yclonic track is over them. Up to the time of the Australian mid-winter on June 2), when the sun has his greatest northern declination, the centre of the anticyclonic belt moves still more towards the equator, raiding the mean monthlj' pressure of all the northern stations. During this period the anticyclones become much larger in dimensions so that the belt becomes very much more extensive, spreading out more both northwards and southwards, and influencing the pressure in the southern stations to such an extent that a secondary maximum is formed in the curves of the more southerly stations. This increase in size of the anticyclones during the winter months has been previously drawn attention to on page 13. The decrease in the nortliern declination of the smi and the consequent retreat of the centre of the anticyclonic belt to higher southern latitudes, should now cause the pressure at all the stations, of lower latitude than the centre of this belt, to fall. Those statioirs to the southward of the centre <^f the Ijelt should show a rise in their pressure curves as the centre of the belt approaches them, a maxinmm being reached in about September to October. At the same time as this rise should occur, the anticyclones are taking on their summer aspect and becoming smaller, so that it is quite possible that this expected rise about September to October may be neutralised. The curves themselves do not indicate this maximum in these months, but at the same time they do not show a rapid descent, but fall gradually and much less gradually than those of the northern stations. In fact, they seem to indicate that the suggestion made above holds good. So soon as the centre of the track of the high-pressure areas reaches the south coast the pressure curves of all the stations on this continent decrease together. In consequence of this change of latitude of the track of the anticyclones during the course of a year, it should be expected that, in the southern portion of Australia, the low-pressure wind systems, which travel approximately from west to east, skirting the southern area, should be more prominent in the winter or high-pressure months (April to September) than in the summer or low-pressure months (October to March). On the other hand, the anticyclones ought to be more conspicuous in the sunmier or low-pressure n)onths (October to March) than in the winter or high-pressure months (April to September). In order to test this, the individual daily barometric readings of the Adelaide Observatory barometer were carefully examined. These readings were taken from four volumes of tlie publications of the Meteorological Observations made at the Adelaide Observatory, and covered the period from April 1891 to March 18!)4. This period was chosen because the mean annual values of pressure for 1891, 1892, and 1893 were very high, average, and very low respectively, and it was A DISCUSSION OF AUSTRALIAN METEOROLOGY. 29 (.lesired to see at the saiue time whether tliose ditt'erent pressure conditions made much valuation in the nuinl)er oi' anticyclones or cyclones recorded. The method of incjuiry was to examine the mean dailj' readings of the barometer from day to (hiy, and count up for each month the number of passages of high or low pressure systems. Low-pressure systems were sought after lirst, and when completed the whole series of data was again gone through to pick out the high-pressure systems. A valuable lead was given in the published columns of remarks, as the weather was briefly described there. The following tables, A. and H., sum up the results obtained : — Table A. II igh-pressure Systems. Mont lis. 1891-2. 18S2-3. 1893-4. No. Total. No. Total. No. Total. r April May Ili-fii , June prcssiii-e ' July August L Sopt(Miil>er f Uctobor i X()veml)er Low J Decenilicr - pressure Jiinuary - Fel)rn!iry - - Murcli (3 2 6 6 3 6 () 3 () 8 7 28 J > 36 7 ! 6 o 7 8 6 9 i 8 7 5 '' i 1 . 39 41 6 o 3 6 6 3 6 6 8 4 6 5 ■ 29 3o 32 37 Totals — 64 — 1 80 ' 1 — 64 Table B. Low-pressure Systems. 1891-2. 1892-3. 1893-4. No. Total. No. Total. No. Total. High ^ pressure Low pressure ■ April May Juno July August , Septeinlx'r ' October NovoinUer DeceuiluM- - flauuary - February - . March (1 1 3 1 2 1 >■ 5 3 1 2 2 2 4 4 4 2 1 1 1 ► 15 J 9 2 4 4 3 4 6 3 2 2 2 1 > 23 9 14 7 Totals - — 8 — 24 — 32 30 SOLAR PHYSICS C0:\I!\1TTTEE. Now the first results wliicli follow from the above tables are, first, that the anticyclones in Table A. are more numerous in the low-pressure or summer months than they are in the high-pressure or winter months. In all three years examined the same result is obtained, the mean result he'mg 37 for the low- pressiire months and 32 for the high-pressure months. In the case of the cyclones (Table 15.) or low-pressure areas, they are more nxunerous in the high-pressure or winter months than they are in the sinnmer months, the means for the three years being as 14 to 7 respectively. The al)Ove results are, therefore, in accordance Avith what was expected. From Table B. one can gather also how long, on the average, it takes an anticyclone to pass over a station. It will be seen that during the three years investigated the mean number of anticyclones per year was 69, or nearly six passed over Adelaide during each month. This means that an anticyclone takes about five days to pass over a station. Mr. Russell, as previously mentioned (page 13), gave the average time of transit as 8' 7 days, so that the length of time here determined is considerably shorter than that found by him. There is little doubt that the method of inquiiy adopted by him, namely, deductions from a review of weather charts, and that employed here — which was to count the highest daily readings of the barometei- — may account for the greater number of anticyclones suggested by the latter method. A comparison of the Tables A. and B. indicates further very clearly that Australia is dominated by anticyclones, and that low-pressure systems take quite a secondary place. Chapter IV. THE MEAN ANNUAL RAINFALL VARIATIONS. A ])ISCL"SSI()N OF .\rS'l1!.\IJ.\N .\IETE0Ji()LO( lY. 33 THE MEAN ANNUAL RAINFALL VARIATIONS. The next step in the inquiry Avas to find out in which months o£ the year rain fell in different parts of the Commonwealth. In looking at a map of Anstralia on which isohyets are drawn, snch as that, for instance, Avhich appears on Plate 26 in Volume III. of Bartholomew's " Physical Atlas (1899), "■••■■ it will be seen that the greatest rainfall is at nearly all the coast districts, with the exception of the region adjoining the Great Australian Bight in the south and that in the north-west division. As the inner regions of the continent are approached, the rainfall becomes a rapidly diminishing quantitj^ until the Great Victorian Desert is reached, where the fall is under 10 inches in the year. Fig. 3, based on the above-mentioned map, shows at a glance the positions of the isohyets (here dotted) for differences of 5 and 10 inches of rainfall. The isohyet for 5 inches and under has been omitted, as it is not corro])orated by more recent data discussed by Mr. Hunt. Now the rainfall season in the course of a year is not the same for everj'' region of Anstralia. This fact can be well studied by comparing curves repre- senting the mean annual rainfall variations, and this procedvire has been adopted here. Excepting for stations near the coast and those lying on or near the great telegraph line, which passes through the heart of Austi'alia from south to north, rainfall statistics are not numerous. Sufficient data are, however, available to determine the main features of Australian seasonal rainfall. In order not to deal with curves representing the rainfall of single stations those of areas which exhibit similar variations have been employed in which the values at two, three, or more stations have been combined. Such curves are used as types for those regions. The following tables give the State, position, number of stations included, years of observation, and seasonal rainfall or annual " rain-beat," reckoning twelve months from the minimmn month. In these tables the districts are grouped according as their wettest months occur in the iniddle of the year. Table A. (winter in Australia), or at the end of the year. Table B. (summer in Australia). * Bulletin No. 2 (issued July 190W) of the Commonwealth Bureau of Meteorology, Meltionrne, by H. A. Hunt, contains a new Kainfall Map of the Commonwealth of Australia. E 142. E 34 SOLAR IMlVSrCS COMMITTEE. Table A. state. Position. Station. Years of Observation. Raiu-beat. ( w. ( Geraldton - - - 23 ) West Australia - - < (Sliark's Bay Hiimelin Pool U V December to Jannary. I District). Carnarvon 17 ( Pinjarrali - - - 23 ) ^1 1» " 1 S.W. (Perth District). Xewcastle ^ Bnnbury [Albany - 23 23 23 \ January to December. 1 S. (Coast). I Espcrance - - - j Israelite Bay ) Eyre ( Eucla 17 lo 16 > December to Jannary. ' Fowler's Bay Streaky Bay 23 ' ( S. 23 South Australia - I (Eyre's Penin- ■i Warrow 24 J> February to January. ( sula). Port Lincoln 35 '. Warratta Vale 24 „ „ -{ S. (Agric. District). [ 50 Stations Bendifio - - - All over 20 42 February to -Jannary. Victoria — Portland - j Cajjc Otway 42 40 - February to January. ( Wilson Promontory 27 Taule B. Yciirs of state. Position. Station. Observation. Kaiu-be.'it. 1 Cossack - - - 18 West Australia - N.W. 1 Boodarie - - - ( Condon 12 12 September to August. 91 »» N. i Derby \ Wyndliam I Port Darwin 14 13 31 August to July. South Australia N. Daly Wnter.s ( River Katherine 28 > August to July. 28 ) 1 Coen - . . 8 V August to July. Queensland - N. Cook town - - - 8 ( Port Douglas 8 i Brisbane - - - 14 ) »» " " S. Hocklianipton ( Mitchell I Hugbenden 14 8 8 August to Jidy. . Inland I Barcaldinc - - - ( Boulia - . - 8 Jtilv to June. 8 1 ■ 1 Alice Springs ) Charlotte Waters - 1 Hermaunsberg 27 South Australia Central 27 13 > August to July. ( Tempo Downs 13 ) f Anna Creek - - - 18 1 9» " ' ' ) Lake Eyre District. I Stuart's Creeks 24 <( MuUoorina - - - 1 Peechawnrina 19 19 ► July to June. I Cowarie - - - 18 , i Armidale - - - \ Cassilis - - - 31 ) New South Wales Inland 30 \ August to July. ( Teiiterfield 30 i »? ij 99 Coast ( Sydney ( Port Macquarie 38 30 [ August to July. A DISCUSSION OK AUSTKAIJAN AllOTKOltOl/H 1 Y. 35 The niiniall (iistricts iu the al)()ve tables are !--ho\vu lor the most part iu the accoinpanying map (Fig. 5) by means oi' areas surrounded by dotted lines. In the case of Queensland the districts represented are those later employed for determining the changes oi rainfall from one year to another. The mean annual variation rainfall curves in Queensland are, however, so similar with regard to the epochs of maxima and minima that the slight dift'erence between the dis- tricts may be neglected. AUSTRALIA ^^ HAME.i>VwPofu &ERALDTO\M J^EISTELRN \ •OAf^WATE-RSi" NORTHERN •/ ^:^t^\J TeRRITORY X \ ^<" SWJNVI l_L.E \\.\K r • n i •'^- ' ' .TAr^ai/ II. TrkiPcDo**N9. I ' . . \ \ • \ •CtARLCTTEW'ME.SS. ( • Adavale .IM.tivhel.. AnnaC* • ' *^ Stuart*Cx. / . i>RlSBANEL 3.^' •.SOUTH KcUA NewcastVe.' n PlNJARRrtH* \ Israelite ^k\ SeTpERANce ^v.B^N^ P''Lmccui»l . J IV'^DEi.ATofe • VICTORIA "- MAP SHOWING THE RAINFALL STATIONS AND DISTRICTS (INCLUDED IN DOTTED LINES) EMPLOYED. 1m (^. .-) Talile 4 (Appendix) gives the mean monthly values of rainfall which have been used in this inquiry. Curves of these seasonal rainfall beats are reproduced in Figs. 6 and 7 ; they are all drawn on the same scale, and ai-e nninbered according to the key map. The curves here shown, and they represent the seasonal rainfall for regions widely distributed over Anstralia, can be divided in the first instance into tAvo main classes, namelv, one in which the c-hief rainy months are in the Avinter E 2 30 SOT.Al! riTYSTrS COMMITTF.K, season (Fig. 6) and the other in which the sunnner months have the most rain (Fig. 7). If the map shown in Fig. 5 be consulted, it will be seen that these two classes or types of rainfall can be almost completely se]iarated l)y a line di-awii across Anstralia extending from the north-west coast about Onslow to the south- east coast about Sydney. INCHES _W£AySIRAUA_2 SHARKS BAV (3 stations) q U9 I. SOUTH COAST p (4 STATIONS') N?3. S"^" AUSTRALIA o A&R\C. DISTRICT. (50 STATIONS) q N?5. RAINFALL. CWINTER.) ^ ^ ^ -, lj,o o u - ■ • INCHES PERTH DISTRICT Q (4S>TAT\0NS>^ N"? Z. P S"^^" AUSTRALIA eVRESPLNlN^ Q (5 STATIONS) N?4-. 2 V1CT0R\K. (4 STATIONS) Fig. 6. Regions neighbouring on this line, such as New South Wales, Lake Eyre District, &c.> appear to some extent to be affected by both the rainy seasons, since the curves for these places exhil)it two maxima — one aliout January or P'ebrnary and anothei- about June. It will be noticed, further, that in both the sets of curves each is divided into two groups by a thick horizontal line. This was done as it was observed that in each main type into which these rainfall curves have here l)een grouped there were regions in which the minima of the curves sometimes indicated very little or no rain and sometimes quite a considerable amount in relation to the annual amount. A DISCUSSION OF xiUSTRALIAN METEOROLOGY. 37 RAIN FALL.C summer) ttCHES W^^ AUSTRALIA NORTH WEST (3 STATIONS) N97. .S^."AUSTRAUA NORTH'^TER^ {3STATI0NS1 NOPTH N?9. SOUTH (3 STMIONS) N?ll n S^■"AUSTR^L\A Y CCNTRKL ■ • (fV STATIONS) N" 13. NEW.SWALLS ^ INLAND 2 (3 STATION^ N? 15. NORTH C2 STATIONS) N?8. QUEENSLAND NORTH Q (3 STAT IONS) N?IO QUEENSLAMD INLAND (3STAT\0NS)N?ia. S^-"AU5TRAL\A LAKEEVRE^ (5 STAT IONS) N?I4. NEW.S. WALES 4 COAST (2 STATIONS) _2. N'Jie. Fk:. :\S SOLAi; PHYSK'S CO.M.MIT'l'EF. Tims, to take an example in Fig. 7, tlie curve representing tliree stations in the northern territory of South Australia records no rain for the months of June, July, or August, while in the series of curves at the l)ottom of tlie ligure the minima never reached zero in any month, and iu three of the cases the minima indi<;ate a fall of one or two inches. Somewhat similar remarks apply to the curves in Fig. 0. Since the curves are all drawn on the same scale, they serve also to show relatively the quantitative distribution of rainfall over the land as j)reviously indicated by isohyets in Fig. 3. Chapter V COMPAUISUN OF THE MEAN ANNUAL PRESSURE AND RAINFALL VARIATIONS. A DISCCSSlOX (3F AUS'l'IfAi,! AX .METE()I{()T.0C;A'. 41 THE RELATION I5E1AVEEN THE ANNUAL PRESSURE AND RAINFALL VARIATIONS. Having now dealt indeijeudently in some detail with the curves representing the mean annual variations of pressure and rainfall, it is important next to con- sider them in relation to each other. The annual harometric change, as pointed, out previously, has a maximum in -lidy and a uu'nimum in Janiuiry. The two main types of rainfall cui'ves have their maxima iu July and January respectively. PRESSURE AND RAINFALL PRESSURE P-^ DARWIN RAINFALL Queensland: SOUTH. (NET. coast) W^'^ AUSTRALIA: PERTH DISTRICT (sw coast) NEW. S.WALES, INLAND. (continuous uinc.) PRESSURE MELBOURNE INCHES. 29-95 Kk.. s. One is therefore confronted ^vith the fact that the single annual pressure change is responsible for two cpiite distinct rainfall pulses. During the high-pressui'e months, that is, irom April to September, the south and south-west regions of this continent receive their maximum water supply. E 142. 42 S()l,Ai; IMIVSICS CoALMl'ITKIv wliile diiriug the low-preysiire inontlis the uoilli and mH-tli-east rej^ious Jiave tlieir maximum rainfall. This relationship is best seen Ijv looking at Fig. cS, in Avhich the annual pressure curve is compared with the typical i-aini'all curves of these two regions. One can quite understand that there must l)e a region where both these pulses of rainfall should be felt in a year, and this region should be found in a l)elt the centre of which shoidd pass from about Onslow in the west to al)out Sydney in the east (see page 36). ' In the accompanying ligure the rainfall curve for New h'outli Wales is given as a type for this doxdjle annual pulse of rainfall. To illustrate how this curve appears to be built up, the portions about the points of maxinimn have been pro- longed by dotted lines to correspond with the principal minima of the other two type curves. It will be seen that by adding the ordinates of the two dotted curves together the continuous curve represents very fairly the resulting cui-ve. It will be noticed that the north-eastern type of rainfall affects the rainfall in New South Wales more than that which comes from the west and is represented by the Perth District curve. The main conclusion to be arrived at from the comparison of these pressure and rainfall curves is that care must be taken to associate the different rain pulses with the pressure changes for the same period. Thus the rain pulse of Queensland, which extends from August to the following -luly, nnist he compared with tlu; pressure for the same group of months, wliile the western rainfall should be compared with the pressure for the months January to Decendjer. Chapter VI. COMPAKMSOX OF THE PHESSUKE AND RAINFAEL CHANGES, AND THE FREQUENCY OF "SOUTHERLY BURSTERS" FROM YEAR, TO YEAli V 2 A DlSCrSSION OF AI'STU'AI.IAX .\1 K'l'F.o I M )!,()( ;Y. COMPARISOX (JF the pressure AND RAINFALL CHANGES EROxM YEAR TO YEAR. It has heeii shown above (page 9) that the mean animal pressure values for the different stations in Australia are not the same from year to year, but exhibit very similar variations of short duration, lasting for about o'8 years. 'J'lie question now arises, termine the mean annual rainfall for any area the raingauges must be well distributed over it, for the values recorded l)y any one gauge are subject to variations due to local conditions, such as contiguration of district (plain or mountainous), elevation above sea level, &c. \n any discussion of rainfall values it is therefore necessary to group together as many stations as possil)le, each group containing only those records which indicate a similar mean annual variation. Wet or dry years in Australia are so pronounced in the curves illustrating the variations of rainfall over Australia that it has not been necessary for the present inquiry, which, of course, can only be considered as a ffrst approximation, to group together for each area investigated a great number of rainfall stations. This is very fortiuiate because, considering the size of the comitry, raingauges ai"e verj' scattered, and, furthei', statistics are not very numerous and they are limited for the most part to the coast areas, and in the niajority of cases do not extend over a great number of years. Nevertheless the material that is available is sufficient, I think, to indicate, at any rate as a first approximation, that a close relationship does exist between the two meteorological elements now^ in question. * This statenipiit only applie.-; to the cases wlicre ehaiifjes occui-, tlie ]>erioils of wliicli cover several ycais. A consiileiable nunibcr of well-distributeJ barometers is necessary for the (hilly forecasting work in order lo know the |)ositii)ns nf the isobars. 46 soi.Ai; iMiYsics r():\i:\irriT;K. It luis been pointed out aljove (page i2) that wlien c(>iisi(l(>i-iii^- the rehitiun- sliip between rainfall and pressure, the group of niontlis which include a complete rain-beat, that is, the I'ainfall from one miiiiiimiu to the next, should be compared with the moan pressure for the same period. Thus, for instance, as the rainfall period on the north-east coast extends from August to the following July, the mean pressure for this group of months should only be considered. In the same way, the rain-beat on the south-west coast, which extends from Januarj' to December, should be compared with the mean pressure for this i:)eriod. Where a region has two rain-beats in twelve months, each should be treated separately and compared with the pressure v^alues for the corresponding periotls. Ivxaminaiion shows that it does not seem necessary always to compare the mean pressure for a irliolc 12 months with that of the rainfall lor the same period, but that the pressure for the middle six months of the year can be employed instead. This is due to the faet that the amplitudes of the mean pressure variations from year to j'ear of the middle six months predominate in the variations exhibited by the yearly means. Thus, instead of comparing the mean pressure for the twelve months Janiiury to December with the rainfall for that period, we can employ the pressure for the six months April to 8epteu:;ber. Coming now to the actual comparisons of the pressure and rainfall curves themselves, these are shown in the two sets of curves given on Plates 2 and o. The rainfall data from which the curves are drawn are given in the Appendix (Tables 5 and C). To (hnd with Plate 2 first, here are indicated the winter (April to Sept(>ml)er) conditions, when the anticyclones reach their most northern latitudes and enable the southern " lows " to give rain to the west, south, and south-eastern j)arts of Australia. As has been shown in a previous section (page 3G), the times of minimum and maximum rainfall during a year for these regions occur in January and June respectively. The rainfall curves here displayed are compared by plotting the total falls recorded during the wdiole interval of the 12 months, January to the following December, and joining them together consecutively. Each district is treated separately', and, going from west to east, they are as follows : — Western Australia, Shark's Bay - - - „ ,, Perth District - - - ,, ,, South (,'oast - - - South Australia, Evre's Peninsida _ . - „ „ Agricultural Districts Victoria -____. 3 stations 4 )j i i» 5 ,, 50 )) 4 A DlSCrsSloX OF .\CS'n;.\I.IAX mktkoiioi.ocv. 47 Curves representing the niiiiL'iiU cluuiges I'roiii year to year of all the above regions are reproduced in Plate 2 and are drawn on the same scale. The last curve on the plate is that obtained by picking out six stations having about the same mean annual fall, and which are well distributed and included in the above regions. Af tlie top ol the plate, the lirst curve illustrates the variation from year to year of the mean l^arometi'ic pressure for the 12 months January to December, and the second curve the barometric variation, also from year to year, of the means for the six months April to Septendjer. Botli of these curves have been inverted so that their highest points, years of deficient pressure, could be more easily identified with years of excess rainfall. If the rain fall curves l)e now compared uifei' sc and also with the inverted pressure curves it will be seen that they all indicate in a general way similar variations. It is \ery probable that the resemblances would be still more striking if it were possible to form rainfall cnrves based on a greater number of stations in each region, l)at, as indicated previously, the data are lacking. The al)ove comparison, considering it as a Hrst approximation only, indicates that, for these regions of Australia, a year with pressure much above the normal corresponds to a year of deticient rainfall, and vice versa. Turning now to Plate o, the sunnner (October to March) conditions are dealt with. It is during this time that the anticyclones pursue a more southern track and the mousoonal " lows " give rain to the northern, north-eastern, and eastern portions of Australia. In these regions the minimum and maximnm months of rainfall are August and January respectively, so that the total rainfall for the 12 months from August to the end of the following Jidy are plotted for each year. Each district into which this portion of Aiistralia has been divided has, as before, been treated separately, and all the curves have been plotted on the same scale. The regions dealt with are as follows : — West Australia, North Coast _ _ - South Australia, Northern Territory Queensland, North-east Coast _ _ _ South Australia, Northern Territory (Central) ,, ,, Lake Eyre District New South Wales, Inland . _ - - ,, ,, Coast - - - - - 2 stations - 3 )> - 3 >> - 3 )> - 55 - •J )> . 2 48 soT.AR rnysics co^^diittee. UnfoTtnnately, the record available for Qiieenslancl is very short. This State WHS subdivided into four areas, and the rainfall variation for each was determined. As they all indicated changes similar in kind but different in quantity (as was to l)e expected from coast and inland stations), it was considered sufficient to reproduce only one of the curves. As before, two inverted pi'essure curves stand at the head of the plate. The first represents the mean pressure changes, from year to year, Ibr the 12 months August to the following Jidj', while the second shows the mean pressure changes, also from year to year, of the middle six months of this period, namely, Xovendx'r to April. These curves have, as in the preceding case, Iteen inverted. The rainfall curves have been arranged in order from the top with regard to the general advance of the monsoonal low-pressure area over the northern part of the continent. Thus the northern regions first feel the effect of this low- pressure area, then in order the central and eastern districtts, and lastly the south-eastern region. Comparing now the rainfall curves with those of the inverted pressure curves given at the top of Plate o, it will be seen that there is a general tendency for the years of greatest i-ainfall to be coincident with those of lowest x'^i'essnre (highest points of pressure curves). The coincidences are more striking, perhaps, in the curves of the northern areas, for there the rainfall is solely dependent on the monsoonal " lows," while towards the more central and southern parts the effects of the sotillicm " loirs " are not aliogether absent. The coincidences are, however, sufficiently numerous, and they would pi'o!)- ably be more in number if the rainfall statistics represented the facts better, to indicate the close connection between the variations of the wintei- rainfall and the pressure for that period. It is unfortunate that the only available data for (Queensland extend over tlie short period of ten years. Nevertheless, for this period high-pressure years are years of reduced rainfall, and vice versa. Year. Au.?ust to .Inly. 1896-1897 1897-1898 1898-1899 1899-1900 1900-1901 1901-1902 1902-190:5 1903-1904 1904-190.) 1905-1906 I'lessiirf. Adelaide. ] Kiiinfall. QuceiislaMiljN.E I (3 Stations.) Iiiolios. 30- 109' 30 051 30 048 30-082* 30 070 .30-08!)* 30-OS9 30 053 30-068 Iiic-lies. 39-57* 65-88 52-76 31-73* 49-75 40-27' 69-38 62 04 44-84 34-30 Rainfall. Queenslaii(l. (Iti Stations.) Inches. 20-50* 33-44 24-43 18-44* 22-38 12-88* 29 ■ 68 33-80 23 07 26-37 A niscrssiox of atstkaijax .MF/i'Koitor/^cy, 40 Tlie above tabular statement deaionstrates this relationsliip very clearly, the high-pressnre years and years of least rainfall being indicated by an asterisk. In column 3 only one district is considered, consisting of three stations in the north-east part of the State. If all the districts be combined, giving a summation of 15 stations, including coast and inland stations, the figures which are shown in the fourth column indicate similar epochs for the minima. No pressure data are yet available for the epoch lOOo-lOOO. The aliove comparisons of tlie variations from year to year of the pressure and i-ainfall f(jr tli" winter and summer seasons, suggest what a step in advance would lie made if it were possible to predict the pressure even a few months ahead. TUK Fl!E(^LK\CY OF " Soi'TIlEllLY BuiJSTKIt.S." In the southern portion of Australia, chiefly in the region of Victoria and New South AVales, certain peculiar conditions sometimes occur which give rise to southerly -winds of great violence. These storms are known as " Southerly Bursters," as they have the peculiarity of commencing very suddenly ; in some years these bursters are much more prevalent than in others. It seemed of interest to find out whether their occurrence, like that of rainfall, was closely associated with the pressure changes, and whether high- pressure ov low-pressure years presented the more favourable conditions for their formation. The inquiry was very much facilitated, because most of the information required has already been published in an admirable essay on " Southerly Jiursters,"''- by Mr. Henry A. Hunt, now Commonwealth Meteorologist for Australia, who has brought together and discussed all the available information concerning these peculiar atmospheric disturbances. The period over which the observations he discussed covered, extended from September 1863 to March 1894, a)x)ut thirty years, so that we have here a valuable series of statistics of bursters recorded at Sydney to deal with. It is hoped that this list will be brought up to date.f Mr. Hunt differentiates between three different kinds of "bursters." The first, the true "southerly burster," is associated with the familiar inverted V depression : and the sharper the V, the more sudden the change. The second type occurs with tropical, or V, depressions, is associated with high temperatures and thimderstorms : this type is of seldom occurrence. The third and last variety results from a secondary. They develop cm the south coast of New South Wales through the formation of a " kink " in the outlying isobars of the i-etreating high pressure. • " Tliree Kssavs on Aiistraliau Weather," by Hon. Ralph Abercroniby, page 16. [Sydney, 1896.3 f Ml'. Hunt lia-i since informeil me tliat tlie list has been brouglit up t) date, but is not yet published. K 112. ' G 50 SOLA]; I'HVSICS C()M.\nTTKK, The statistics show that ]:)ursters are most frequent in the sunnner months. 'I'hiis the totals recorded at Sydney for the period aliove mentioned an^ given by Mr. limit as follows : — Month. Total Number o£ Bursters. Mean Pressure. Sydney. (185i)-1887.) Aiin;ust Septeinlii'r October Noveniher December January - February Marcb April - Mav o 62 139 166 182 170 132 90 41 4 Inches. •29 ■ 927 •sao •807 •742 •769 •801 •890 •9;;o •913 It will be seen that the month of December has the greatest number, while the months between May and August have only five or less each. When we compare these figures with the annual variation of pressure at Sydney, the values for which are included in the above tal)le, one is led at first to associate the great frequency of bursters with the low-pressure months. It must be jjorne in mind, however, that during the low-pressure months the paths of the anticyclones over Australia pursue a more southerly course, Imrring out to a considerable extent the southerly lows which attempt to wedge themselves in between them. Since the southerly Imrsters are for the most part met with in Victoria and New South Wales, their frequency is therefore closely associated with the presence of the anticyclones in these higher latitiides. The southerly positions of the anticyclones at this time of the year render the noi'thern portions of a V depression to the south of them very sharp in the region of Victoria and New South Wales. The higher the pressui^e in the anticyclones the sharper and more intense must the V flepression be, otherwise it could not be recorded f)y the Sytlney barometei^s. One is thus led to conclude that a " southerlj' burster " is really an attempt of an intense low-pressure system or part of it to wedge itself in between two anticyclones at a time when such a movement is very difficult. There are several other conditions which render this attempt successful, such as those which Mr. Hunt mentions, namely, the flattening of the front of the isoljars of the anticyclones against the mountains of New South Wales, the consequent deformation ol the V depressions, isobars, &c. In dealing with the variation in tlie frequency of l)ursters from year to year Mr. Hmit gives a very interesting tal)le, which shows that some years are very much more liable to their occurrence than others. The numbers recorded for each yea,r, reckoning the year from August to the following July, are as follows : — A DISCrSSK^N OF AI'STItAI.! AX .MF/rKOliOLOCI V. m Vfiir. 1864- 18fi.')- 1 S6(>- 1867- 18()S- 1 8'- Attention must, however, be drawn to the exceptional value of lioth the gauges for the j-ear 1880, as this has no apparent equivalent in the rainfall for Xew South Wales for the period August 1885 to July 1886, which period should correspond to that of the luver height. This point is referred to in a subscMinent paragraph. The Murrmj (J(ui> in order to make this memoir more complete. Importunately, three excellent series of barometric observations are available for this region, nanaely, those for Adelaide, Melbourne, and Sydney, commencing respectively in the years 1857, 1859, and 1858. The changes at Perth are also included here, as tliera is a set of observations commencing in 1870, which will represent tho variations which take place well to the west of Australia. The tirst step taken to render more apparent the changes of long duration involved in all these curves was to eliminate as far as possible the prominent short variations of abovit four years' din-ation. This was to a great extent accomplisheil by gronping the years in sets of four and emplojang the mean values of each of these groups. Thus the means for the years 1873 to 1876, 187-1 to 1877, 1875 to 1878, and so on, were determined, and curves were drawn through each of these points after they had been plotted. Each mean valne was actually plotted on the time scale at the end of the second year of the group of which it was the mean. Tims the mean for 1873 to 1876 was plotted at the end of 1874. The curves for the four stations are reproduced on the same scale in Plate 6, and their monthly, anmia), and four-year mean values are given in Table 2 in the Appendix. The prominent features of the curves are as follows : — About the year 1807 and 1868 three of the curves, the records of which go back to this date, indicate a prominent maximum. About the year 1878 three of the curves, namely, Melbourne, Sydney, and Perth, show a small subsidiary maximum, while that in the case of Adelaide is more pronounced. About 1887 all the curves indicate a large maxinmm, greater than any of those that preceded it. About 1890 another subsidiary maximum seems to be indicated, and it looks as if the curves are tending towards a prominent maximum about 1900. • Roy. Sop. Proc, A., Vol. 78, 1906, pajre 43. E U2. ()t) soT.Ai! PHYSICS c'o:\r:\nTTKE. If for a moment the curve for Adelaide at the epoch 1878 be left out of consideration it will be seen that the curves bear a close reseml)lanee to one another. An hypothetical curve euibodying the main features of these c'hanges has been drawai at the bottom of the set of curves. This is intended to indicate in one curve the general nature of the variations as regards their epochs of maxiuiuiu and. minimum. The epochs of the two principal maxima are seen to occur at about 18G8 and 1887, while three subsidiary maxima are suggested about the years 1859(?J, 1878, and 1897, but the epoch of the first of these is uncertain, as the curves do not extend over a sufficiently long period of time. Nevertheless it may be remarked that the interval l)etween the two chief maxima is 19 years, while those between the successive secondary maxima are about the same length. In the hypothetical curve those portions representing the fall fi-om and the rise to the principal maxima have been connected by a dotted line as if a subsidiary maximum did not exist, thus forming a principal (but really non-existent) minimum. From this it is suggested that the curve i"ises quicker than it falls, the interval being eight years for the former and eleven years for the latter. The Magnitude of the Pressure Change of Long Duration. In order to determine the amplitudes of pressure variations of long (hiration the difference between the readings of the several maxima and minima of the curves plotted were determined. For the three stations Adelaide, Melbourne, and Sydney the diiferences for each were 0'()54, 0"0I3, and 0'058 inches respectively, the means of them being 0'0o2 inches. The actual values from which the above figures were derived are stated in tabular form below, and it must be remembered that as the points on the curve are derived from means of four years, the years given refer to the two middle years of each four :— Adelaide. Melhourne. Sydney. Year. Inches. Mean. Minima. 1861-62 30-023 1870-71 33 035 Maxima. 30-029 1877-78 30 082 1886-87 30 093 1895-96 Difference 30-073 30-083 0-0.)4 Year. Inches. Mean. Minima. 1862-63 29914 1873-74 29-921 1893-94 29-915 Maxima. 29-917 1867-68 29-947 1886-87 Difference 29-973 29-960 — 0-013 \\-.u: Indies. Mean. Minima. 1 862-63 29-842 1874-75 29-829 1880-81 29-837 Maxima. 29-836 1 857-58 29 ■ 896 18S6-87 29-902 1 895-96 DiffcrcMce 29 - 883 29-894 — 0-058 The mean difference for the three stations, taken together, is ()'()52 inches. A DiSCrSSION OF Al'STUALIAN METEOROLOGY. 67 The Relation between the Pressure and Rainfall. It lias l)eeu pi-evioiisly shown (page 48) in the case of the pressure and rainfall rlianges of short dnration that in a vear of excess or deficient pressure the rainfall over the whole of A^lstralia was (leficient or in excess respectively. It was expected, therefore, that a similar relationship would hold good in the case of the changes extending over several years. For this purpose a search w^as made to secure the longest honiogeneons series of rainfall data. Unfortunately, such records are not very numerous, and it was only found possible to examine the data for Sydney, Brisbane, Adelaide, Yanko, Melbourne, Perth, and a summary of stations representing the agricultv;ral districts of South Australia. In order to eliminate the variation of short duration (four years) prominent in all the curves, four-year means were utilised as was done in the case of pressures. By this method the variation of long duration becomes more consi)icuous, although the curves are still not very smooth. The separate curves thus obtained are reproduced in Plate 7, and for purposes of comparison the four-year mean pressure curve (inverted) for Sydney and the inverted hypothetical barometric curve are also added. The annual and four-year mean rainfall values employed above are given in the Appendix (Table 9). Although the agreement between the rainfall curves is not as good as one would have desired, especially during the years 1840 to 1860, when the Sydney and Adelaide curves are the inverse of each other, yet they all show that nearly a similar rainfall change of long duration is in existence, and that it is associated (with the exception of the epoch above mentioned) with the pressure variation of about the same period. On the suggestion of ]\lr. Hunt, the rainfall data for Horsham were also dealt with, and the curve obtained (but not here shown) was found to indicate similar epochs of maxima and minima to the curves given in the plate. The data on which the curve was based will be found given in the Appendix (Table 9). So far as the data of pressure and rainfall permit, one is thus led to deduce tliat there is a variation of about 19 years period in both these meteorological elements. The long harometric saving of 19 years in Australia does not seem to have been pointed out before (Bruckner omitted the Australian area in his pressure investigation), ='■ but the existence of a 19-year period of rainfall change has often been mentioned. * " Kliraaschwankiuigeu scit 17U0," Eduard Bruckner, page 194. 1890. I 2 68 ROr^Al? PriYSlCS CO.M.MITTEE. In an article on the " Development of j\leteorology in Australia," by Mr. Andrew Noble, of the Sydney Observatory, it is stated : " Australian meteorology " is greatly indebted to the Rev. W. B. Clarke for his untiring efforts in its " behalf during those early j^ears, beginning with his observations at Paramatta " in the year 1839 and continuing long after the inauguration of the New South Wales service imder Government auspices in the year 1858. . . . The " lS)-year cj'cle theory, elaborated by Mr. Russell in more recent years, was '• advanced by Mr. Clarke in the ' Sydney Morning Herald ' of May 1, 184()."* This reference is of great interest, since it indicates that this 19-year variation was evidently quite a pro;ninent feature of Australian weather before the o^^.seri-af/on-s- discussed in the present memoir were made. In "Notes on the Climate of New South Wales," 1870, Mr. H. C. Russell advocated a 19-year rainfall change, and in a later paper, which he published in 1876,1 he more strongly advocated this rainfall cycle. In a still later paper, which he published in 1896,:]: ?ilr. Russell collected information of a miscellaneous kind and extended his 19-year rainfall cycle both over a greater period of time and a wider area. In fact, Australia, India, Europe, Asia, Africa, North and South America, all tended to give him general ideas relating to droughts, which he marshalled, and from which he deduced that this cycle was occiirring over the whole earth, epoch for epoch, nearly simidtaneously. With this latter conclusion I cannot, however, agree, for reseaz'ch of late years has shown that while practically one half of the woiid is undergoing excess pressure the other half is experiencing diminishing pressm-e. In conseqiunice of this pressure variation the rainfall must and does follow suit, so that where the pressure is in excess the rainfall is in deficit, and vice versa. Thus, a drought cannot occur over the ichole world simidtaneously, although it may extend over a considerable portion of the glo1)e at one time, such as over East arul South x\frica, Arabia, India, East Indies, and Australia. A drought over this area would pj'oh- ably correspond simultaneously to an excess of rainfall over North-west Africa, North and South America, and Siberia, for in those regions the pressure at that time would be deficient. * "Monthly VVeiitliiT Review," Vol. 33, No. 11, November 1905, page 480. [Wiishinjrton, U.S.A., \Vfather Bureau.] t •' Journal of the Royal Society of New South Wales," Vol. 10, pajje 151, 1876. I "Journal of the Royal Society of New South Wales," Vol. 30, jjagc 70. 1896. ClTAFTEn IX. THE RELATION OF AUSTRALIAN PRESSURE CHANGES TO VARIATIONS IN OTHER PARTS OF THE WORLD. A DISCUSSION OF AIJSTRAIJAN METEOROLOGY. 71 TKE RELATION OF THE AUSTRALLVN PRESSURE CHANGES TO VARIATIONS IN OTHER PARTS OF THE WORLD. TUK \'.\I!1.\TI()NS OF SlIOIiT DuiiATION (aBOUT 4 YeAKS.) Ill llie preceding chapters it has been shown that the pressure over Austraha i'roni yeai' to year undergoes t^vo variations, one indicating a variation which takes about four years to complete a cycle, and the other a cycle which seems to be about 11) vears in length. 's^' It has also been sliowu tliat these two variations are not solely restricted to Australia, but that they are very intimately associated with changes that occur in other parts of the world. In dealing, therefore, with the meteorology of Australia it is fundamental that a careful watch shoidd be kept on the changes taking place elsewhere, and b}' so doing it is quite possible that in time to come the ke\' to long-period fore- casting, that is, the prediction of the nature of seasons, will l)e based on such broad views as are above indicated. It is proposed, therefore, in this chapter to indicate briefly the probable association of Australian pressure changes with those taking place in India, South America, and South Africa. Ip a previous part of this memcjir reference has been made to a world-wide barometric see-saw which has recently been discovered by the comparison of pressure clianges from year to year at a great number of stations well disti-ibuted over tlie earth's surface. So far as could be made out, the two centres of these inverse pressure types were situated, one in South America fabout Cordoba), and the other in India, with Bombay as the type. The presence of this see-saw iiidicateil distinctly that in some years there seemed to bo a cause in operation which produced over the South American region a higlier pressure than usual, while at the mmc time there was a deficiency of pressure over the Indian ai-ea, and v/Vt versa. This transference of pressure, from approximately west to east and east to west, was found to have a period of about o'S years in the mean. The result of the discussion of the pressure changes indicated that the South American type of variation was closely associateil with the changes going on in the United States, Central America, Nortli-west Africa, Siberia, and Honolulu. The Iiiflian type, on the other hand, was intimately associated with the Variations taking place in luirope, North-east and South Africa, Arabia, East Indies, and Australia. 72 sor.Ai; I'livsirs co.mmi'I"] 111 the study, therefore, of Australian meteorology the pressure changes of Batavia (East Indies), Boniba)*, and Cape Town should at least be kept well in view, and all of these should be carefully compared with the reverse pressure changes which occur simultaneously at Cordoba. To indicate the nature of these barometric variations in widely separated stations, attention may here ))e tlrawn to the changes which occur from year to year in the regions of South America, South Africa, India, I'^ast Indies, and Australia. Not only are these changes very prominent in the annual values of pressure which have been use\ate. The study of the inter-relationship of these pressure changes is therefore of extreme importance for Aixstralian meteorology, but a great amount of research work is still needed before the variations can be accurately interpreted. A ])rSCUSSI()N OF AUSHfATJAN METEOROLOGY. 73 Vakiations of Long Duration. While the above remarks have reference only to the pressure changes of short duration, such as that which covers about 3'8 years, the barometric variation of about 19 years duration must not be ignored ; in fact, it seems to be a very important factor in Australian weather changes. Reference has already (page 67) been made to these changes of long duration, both as regards pressure and rainfall, showing that the maxima of the curves are separated by an interval of about 19 years. In South America, changes covering a similar length of time, but with different epochs of maxima and minima, are also in operation, and these have recently formed a subject for a communication by me to the Koyal Society."' Unfortunately, the barometric data for South America are not very numerous, 1)11 1 utilising those that were available, four-year mean curves, formed in a similar way to those obtained for the Australian stations, were drawn. The stations employed were Cordoba, Goya, and San Juan (Argentine Republic), Santiago (Chili), and Ciirityba (Brazil). Although the curves extend over different periods of time, there is sufficient overlapping in all cases to connect up one series Avith another. The Cordoba curve undoubtedly indicated that a long barometric change was taking place, but the shortness of the period over which the observations extended, namely, from 1873 to 1904, rendered it unserviceable for the determination of its possible periodicity. A neighbouring station, Goya, corroborated in a general manner, so far as the observations extended, the Cordoba variation, with perhaps the exception of the first three points on the curve. To carry back the pressure changes to an earlier date, the observations at San Juan (Buenos Ayres) were emploj^ed ; the available data for this station extended from 1867 to 1889. Here the fall of pressure at Cordoba from 1875 to 1882 was well corroborated, followed by a subsidiary maximum similar to that at Goya in 1885. So far as these observations extend, there seemed to be two prominent maxima at about the epochs 1874 and 1893, which were followed by minima at about the years 1882 and 1901 respectively. The curve for Santiago, a station to the west of the Andes, indicated also very clearly these two principal maxima and the second of the two minima at the same epochs, but the minimum about the year 1882 occurred somewhat earlier. At a station in Brazil, Curityba, in which only a short series of observa- tions was available, this long variation is also in existence ; the second principal maximum, however, fell a little later than at the previously mentioned stations. Forming a hypothetical curve in exactly the same Avay for the South American region as was done in the case of Australia {sec Plate 6), the two » Koy. Soc. Pioc, A., Vol. 78. 1906. K 142. K 74 SOLAR PHYSICS C0M:MITTEE. principal maxima fall in the years 1874 and 1893, while a subsidiary maximum occurs somewhere between 1880 and 1885. Here agaia the interval between the two main maxima is 19 years. This curve is reproduced in Plate 9, and below it is given the Australian curve for comparison. In both of the hypothetical curves those portions representing the fall from and the rise to a principal maxinmni have been connected by a dotted line as if a subsidiary maximum did not exist. The object of doing this is to indicate that in the Australian area the rise to the principal maxima seems to be more abrupt than the fall from them, while in the South American area the opposite features seems to be the case. An unsymmetrical curve seemed in both cases to represent the main features better than one drawn symmetrically. In fact, in the Australian area there is suggested an 8-year rise and an 11-year fall, while in the South American region an 11-year rise and an 8-year fall is indicated. Particular attention is called to this unsymmetrical peculiarity of the curves, since a similar feature was found to be present in the curve representing the barometric variation of about four years duration,'" in operation in India and Cordoba. It was there stated that for Cordoba " the points of maxima of the " hypothetical curve at the top of the plate do not lie midway between the " minima on either side of them, but nearer the preceding minimum." The first striking fact which this comparison indicates is the remarkable similarity of the nature of the variation in the two cases. Both curves seem to have principal maxima occurring at intervals of about 19 years, while situated between these is another maximum of a subsidiary nature. The second point of importance is that the epochs of these maxima in these two areas are not coincident. Further, we are not here in the presence of a barometric see-saw, or opposite pressure variation, because the Australian maxima do not occur simultaneously with the South American minima; there seems 1o be a general time-difference of phase amounting to about six years, the epochs of the Australian high pressures preceding those of the South American region. In the case of the barometric variations of short duration existing between India and South America, the inversion of the latter curve corresponded exactly with the direct curve of the former. In order to make a similar comparison, the South American curve, representing the curve of long duration, has here also been inverted, and it will be observed that on Plate 9 the curves are not quite the inverse of each other. It is unfortunate that the length of time, covered by the observations discussed above, is not sufficient to determine whether these variations of long duration are periodic or not ; the curves, so far as they go, suggest that distinct changes of long duration are in operation. » Roy. Soc. Proc, A., Vol. 76, page 503, 1905, !iote. A ])18CUSSl(W OF AUSTRALIAN METEOROT,0(IY. 75 Whether the difference of phase between the South American and Australian pressure curves will in the future he a means of forecasting Australian Aveather cannot be stated, but the curves are so suggestive that if future observations give further indications of this apparent periodicity, an attempt might be made. The examination of the Australian and South American pressure observations for variations of long duration and the interesting conclusions derived, suggested an inquiry into the natiire of the changes occurring in South Africa and India. The variations of long duration occiirring in the Indian area have already been discussed by me iu a previous publication,'-' so only a brief reference will here be made to them. By forming four-year means of the Indian pressure values (see Appendix, Table 10) fairly smooth curves were obtained for several stations. These curves were seen to be exactly alike, indicating that the origin of the variation was general to all Indian stations. One of these curves, namely, that for Bombay, is reproduced in Plate 10 and shows the nature of the variation which was found. As a stepping stone to the Australian area the observations made at Batavia, Java, in the East Indies, were similarly treated {see Appendix, Table 10). Although the observations do not extend over such a long series of years, yet where the curve overlaps that of liombay it will be seen (Plate 10) that they present striking similarities, suggesting a conomon origin. Coming now to the Australian continent, curves for four stations having already been reproduced in Plate 6, I have previously pointed out that the variation exhibited by the Adelaide curve differed from those of Melbourne, Sydney, and Perth. This I did in the following wordsf : — "It will be seen in the first instance that the Adelaide curve resembles in a general way that of Bombay and that the maximum about the years 1877 and 1878 in India is almost equally pronounced in Adelaide. Attention is specially drawn to this particular maximimi, as it wiU be observed when examining the curves for Melbourne, Sydney, and Perth, that during these years it becomes of quite secondary importance." The above statement indicates, therefore, that while the Adelaide curve bears a family likeness to the other three stations on the same continent, yet it shows a stronger resemblance to the Indian and Batavian curves. In Plate 10 the curves for Adelaide and Sydney have been reproduced again in order that they can be directly compared with the curves representing the changes over the Indian and East Indian areas. The conclusion to l)e drawn from all the curves above referred to is that while they all indicate somewhat similar variations of long duration, yet there seems, on the whole, to be a difference between the variations over the Indian and Australian areas. • Hoy. Soc. Proc, A., Vol. 78, 1906. page 43. + Roy. Soc. Proc, A., Vol. 78, 1906, page 48. K -2 76 SOLAR PHYSICS COMMITTEE. It has been previously stated (page 72) that in discussing tlie pressure variation of abox;t four years duration, tlie changes indicated l)y the Cape Town barometer seemed to be intermediate between those shown by tlie South American and Australian barometers. An attempt was therefore made to see whether any variation of long duration was exhibited in the Cape barometric observations, and to compare it with those previously found for South America, Australia, and India. As a good series of barometric observations of Durban was at the same time available, this station was discussed as well. For both stations four-year mean values were determined, and curves drawn in the usual manner. Both of these are reproduced in Plate 10, from data given in the Appendix (Table 10). In examining the curves the first point that strikes one's attention is that the curves differ from one another in many respects. Thus the minimum about the year 1880 at Diu-ban is not repeated at Cape Tow^n. The high values for the years 1881 to 1883, which make the curve indicate a maximum at this epoch, do not correspond to the maximum in the Durban curve, w^hich occurs about the year 1888. When one comes to compare these curves with those of India, East Indies, and Australia, one is very inclined to throw doubt on the Cape observations aboiit the year 1880, because the Cape curve about that epoch alone difl'ers not only from that of Durban but from those of India and Australia, with which Durban agrees. If, therefore, the Cape curve for the interval 1879 to 1884 be omitted from the discussion, then it wnll be seen that the two South African curves bear a close resemblance to the Indian, East Indies, and Australian curves, but approximate more to the Adelaide variation than to that exhibited by Sydney. If the South African curves be compared with that of Cordoba, which represents the South American type of variation, their dissimilarity will be at once observed (Plate 10). One is thus led to the conclusion that, so far as the variation of long duration is concerned, South Africa must be associated directly with the Australian and Indian regions, and more especially Avith the latter, and not with South America. A study of Plate 10 will show the inter-relationship between all the barometric curves to which reference above has been made. Possible Okigin of A'aihations of Long Duration. In looking for the cause of these barometric changes which extend over several years, I have suggested* that possibly solar changes, as exhibited by the • Proe. Roy. Soc, A., Vol. 78, 1906, pan;e 55. A DISCUSSION OF AUSTRATJAN METEOROLOGY. 77 frequency or areas of sun-spots (the only indication of solar activity extending over a long period of time that exists) may be responsible for the Indian fluctuations. To indicate this relationship the sun-spot curve (inverted) is placed at the top of Plate 10. This curve represents the variation from year to year of the mean dail}^ areas of sun-spots deduced from both hemispheres of the sun. Perhaps different solar data handled in another manner naay indicate at some future date a closer relationship than is at present suggested. Although this existing relationship may be considered of too approximate a character to indicate clearly a cause and effect, there is undoubtedly a general similarity Ijetween the sun-spot variation curve and that representing barometric changes in India from 1844 to 1903, a period of 59 years. Years of average high pressure are j^ears of few sun-spots, and vice versa, but there is a marked exception to this about the epoch of sun-spot maximum in 1883, which maximum was much smaller in intensity than those of 1870 and 1860. If India be thus dominated by the solar changes, then the curves for Australia and South America become of secondary importance from the solar point of view, and may be considered as a modification of the Indian variation, due possibly to soiue terrestrial cause. How this modification is brought about I am not yet prepared to say, but I do not think we need be driven to explain the Australian or the South American barometric changes as depending either on lunar influence or a solar variation of about 19 years. The simihirity of the curves representing the pressure changes in India and the sun-spot curve is not pointed out here for the first time. In fact, so striking was the resemblance between curves representing these changes in years previous to 1880 that the attention of several meteorologists was draAvn to the close association of these two phenomena. Thus F. Chambers, writing in 1878,"* concluded that the curves "support " each other in showing a low pressure about the time of sun-spot maximum and " a high pressure at the time of sun-spot minimum." He further stated : — " The range of the variation of the year by mean pressure from the minimum of 1862 to the maximum of 1868 is 0*042 inch, and the mean range of the barometer from January to July is 0'291 inch, from which it appears that the variation of pressure produced by the absolute variations of the sun's heat are, in (comparison with the usual seasonal changes, by no means insignificant." J. A. Brown,t S. A. Hill,^ Sir John Eliot,§ H. F. Blauford,|| E. Douglas Archibald,^ and others, have all corroborated in a general manner this relationship * "Nature," Vol. 18, page 568. t " Nature," Vol. 19, page 7. 1878. + "Nature," Vol. 19, page 432. 1878. § "Iiuliaii Meteorological Reports," page 170. 1877. !| "Nature," Vol. 21, page 479. 1880. t " Iiuliaii Meteorological Memoirn," Vol. 9, page 543. 1897. 78 SOT..VR PHYSICS COMMITTEE. between iiressiire change and sun-spot variation ; Douglas Archibald nsed data which extended up to the year 1893, and his deductions, which I think are the most recent, were : — " The mean anomalies present all the characteristics of a true period, rising to a maximum of 0"0132 inch about the epoch of minimum sun-spot, and, with an exception in the sixth j'ear, falling to a mininnnn of 0"0100 inch coineideutly with that of maxinmm sun-spot, the former barometric epoch slightly preceding, antl the latter slightly following, the corresponding solar epoch as is usual in all other sun-spot comparisons. ..." " Still the figures from the other years, and the repetition in each cycle, show that there is a cyclical tendency to high pressiTre at the time of few spots and low pressure at the time of many spots .... the amplitude of the variation amoimts to 0*02 inch . . . ." By utilising in this inquiry observations made up to the most recent date possible, 1905, it will be seen (Plate 10) that the sun-spot maximum of 1893 corre- sponded with an epoch of mean low pressure about that epoch ; while up to 1901, a year of about sun-spot minimum, the pressure had steadily risen. It is thus evident that the same relationship is still in operation, only the amplitude is much smaller than was the case in the earlier years of observation. Further Discussion on other Views that have been expressed. Before leaving this subject of the variations of long duration of the barometer in the above regions and their possible origin, reference may be made to some recent work by Colonel A. E. Rawson, C.B., R.E., in which he has discussed the South African barometric changes. In a paper entitled " The Anticy clonic Belt of the Southern Hemisphere," ■■••" his inquiry leads him to the conclusion that there are changes of long duration indi- cated by the South African barometers. The changes, according to him, display a nineteen-year period, or, as he states, the anticyclonic belt has a movement in latitude and "performs its double oscillation in a period of 19 years." A certain ambiguity arises with the use of the term " double oscillation," for it is iincertain whether one or two complete oscillations are meant. Since, however, he stated that the period in question was of 19 years duration, and accompanied his remarks with only the values of the Cape pressures for the years 1845, 18G5, and 1884, pointing out that these years (about 19 years apart) exhibited exceptionally high pressures (and therefore, presumably, the intervening years were not so high), one is led to infer that his " double oscillation " means a " complete oscillation " and that the Cape yearly values exhibit maxima about •every 19 years. • Quart. Jour. Roy. Met. See, Vol. XXXIV., No. 147, July 1908, page 165. A DISCUSSION OF AUSTRALIAN METEOROLOGY. 79 This deduction as to a ''complete oscillation" seems Iiirtlier corroborated by another table which he gives showing the cyclical changes of the belt's latitude and years of reaching its extreme and mean positions. This table is as follows : — Anticyclonic Belt. Ut. Years. Interval. Extreme nortlierly position - 24i° S. 1855, 1874, 1893 - Years. 19 Menu position - - - - 29f S. 1850, 18.59, 1869, 1878, 1888, 1898 9-5 Extreme soutiiorly (wsitioii - 34° S. 1845, 1865, 1884, 1903 19 According to the above table the pressures at the Cape should have exceptionally high values about the years 1845, 1865, 1884, and 1903, and very low values about 1855, 1874, and 1893, because the Cape would be more tinder the influence of the anticyclonic belt in the former series of years than in the latter. During the intervening years, as given in the tables, the Cape pressure should have a mean value. In a preceding part of this memoir I have already (page 76) referred to the variation of long duration of the pressures in Soiith Africa, using the obser- vations made at Cape Town and Durban, and I pointed out that the changes found indicated a closer resemblance to those occurring in the Indian area than to those taking place in Australia as represented by the Sydney, Melbourne, and Perth curves. The Indian curves, it will be remembered, resembled closely the inverted sun-spot curve (Plate 10), or had a variation of about 11 years, while the three Australian curves above mentioned suggested a 19-years variation. This result as regards the Cape pressures is not, therefore, in accordance with that found by Colonel Rawson, who suggested a 19-years variation for South Africa. In dealing with a variation of long duration extending over many years it is important to eliminate as far as possible, in the first instance, changes the duration of which complete a cycle in three or four years. To this end, therefore, four-year mean values are preferable to the annual mean values. A ctirve representing the variation of the mean annual values of Cape Town from 1860 to 1904 is given in Plate 8,* and it will be seen how difficult it is to differentiate between the various peaks of the curves. If, however, one restricts oneself for a moment to mean annual values, it is found that between the years 1845 and 1865, for which years the pressures were 30 '057 and 30 "035 inches respectively, and Avhich Colonel Rawson considers as years of exceptionally high pressures, the pressures for the years 1853, 1854, * In this ])l!ite tlie Cape curve was not plotted for yeai-s previous to 1860, as it was unneeessary for tlie object of which the plate was made. so SOl.AH I'UVSICS COMMITTEE. 1855, aud 1856 were 30 "047, 30-049, 30-052, and 30-043 inches, all higher values than that recorded for 18<)5. Again, lie looks upon tlie pressures for the years 1865 and 1884 as abnormally higli, vahies 30-035 aud 30-049 inches respectively, but disregards such intervening years of high pressure as 1874 with 30-044 inches and 1876 with 30-052 inches. If, however, we consider the four-year mean values aud examine the curves made np from theni, the sequence of changes can be clearly discerned. Such curves for Cape Town and Durban are reproduced in Plate 10. I have previoiisly referred to the disagreement between these curves (page 76) for the years 1880 to 1884. The most prominent maxima and minima of these two four-j-ear mean curves can now be easily picked out, and leaving out of account the Cape curve after the year 1879, the epochs and values thus derived are shown in the following table : — Table showing Years and Values of the most prominent Maxima and Minima in THE Four-year Mean Curves. Years of Maxima. Capo Town. Durban. Interval in Years. 1842-1845 - - - - 1853-1856 - - . . 1866-1869 - - - - 1874-1877 1887-1890 - - - - 1898-1901 - . - . Years of Minima. 1849-1852 - - - - 1859-1862 ... - 1870-1873 - - - - 1878-1881 1892-1895 - - - - 30-043 30-048 30-038 30-040 30-016 30-014 30016 30-108 30-! 13 30 107 Mean 30 084 30-088 Mean 11 12 8 13 11 11 10 11 8 14 _ 10-7 The two chief facts to be derived from this table are as follows : — First, whether the intervals between the consecutive maxima or minima of the curves be considered, the mean interval between their occurrence is about 11 years. Second, the years of maximum pressure congregate about the years of sun- spot minima epochs, while the years of minima pressure occur about the times of the epochs of sun-spot maxima. These results clearly show that, so far as the data that have here been employed are concerned, the variation of pressure in South Africa has not such a very definite 19-year variation, but one covering a little more than half this period. A DISCUSSION OF AUSTRATJAN METEOROLOGY. 81 Whatever may be the hnigth ot the variation that is really in existence, Cokjnel Rawson suggests that the origin of the Ijarometric changes is the movement in latitude of the anticyclonic belt. Towards the latter portion of his paper (page 182) he refers to the work of Mr. Russell regarding Australia. He says : " On turning to Mr. Russell's con- " elusions regarding that of Australia there seems to be a cyclical variation of " the belt there alscj, which corresponds very closely with that which is given " in Tal)le XIV. He was led to advance the theory that thert^ is a general " periodic occurrence of seasons every 11) years, and that during ever^' cycle of " 19 years there are two periods of good seasons and two periods of bad seasons, " due to dnmghts, besides minor fluctuations." The Table XIV. referred to in the above quotation is that given here on page 70. From the above quotation Colonel Rawson suggests that the barometric change froui year to year in Australia is also due to the variation in the position of the anticyclonic belt, but, so fur as I am aware, Russell himself did not definitely state this as a cause. Colonel Rawson puts forward the idea, however, that " Mr. Russell's theory " probably rose out ol the existence of the belt's double oscillation over Australia." Leaving Australia for a moment, and dealing with Natal, Colonel Rawson refers to the important statejnent made by Mr. E. Nevill, the (lovernment Astro- nomer of Natal, which appears in his annual report for the year 1893-1894 : — " Natal is on the border of a great southern anticyclonic belt, and it would appear that it is the position and condition of this belt which mainly regulates the climate of Natal. As the position of this belt probably is cyclic in character, Avith a period approaching nine years or a multiple thereof, a knowledge of its position and condition for the preceding years will afford the means of judging of its probable character for the coming year, and thus for predicting the nature " of the coming season." Mr. Nevill, it will be seen, throws out the view that the position of the anticyclonic belt about Natal is only probably cyclic in character, but he does not state that the variation has a 19-year period, but a period "approaching nine " years or a multiple thereof." If the reader Avill glance at Plate 10 in this memoir he will see that the curve there shown, indicating the variation from year to year of the four-year mean pressure values at Durban, render three inaxima prominent, namely, at the epochs 1876, 1889, a]]d 1900. The intervals between these maxima are 13 and 11 years respectively, or 12 years in the mean. The shorter period given by Mr. NeviU seems therefore to come closer to the interval here found than the multiple of nine years of which he speaks. K 142. L 82 SOLAR PHYSICS COMMITTEE. Mr. Nevill not only speaks about the position of the anticyelonic belt, but also of the condition of it. It is this latter factor which may prove to have a more important bearing on these pressure variations of long duration, than tlieir position in latitude. With the object of attempting to throw soaie liglit on this question the data for Adelaide have been examined. Reference may lirst be made to tlie tables on page 29, in whicli the numbers of anticyclones and low-pressure areas were given for the three years 1891, 1892, and 1893, j'ears of high, average, and low pressures respectively. Although only three years were examined, their mean annual pressure values were so different that they represented good epochs for showing any great changes, if such occurred. The figures there suggest that the year whicli has the lowest mean pressure, namely, 1893, is also the year in which the greatest number of l()w-]n-essure areas is recorded. Thus for the six months (April to September) the numbers for the years are 5 for 1891, 15 for 1892, and 23 for 1893. In the case of the high-pressure year 1891 there does ncit seem to be a greater nundjer of anticyclones recorded, the nund)er for the six months (October to March) being 3G, 41, and 35 for 1891-1892, 1892-1893, and 1893-1894 respectively. Russell also found* that the average numlier of anticyclones passing over Austi-alia, so far as the ol)servations he discussed went, varied l)ut little. It seemed at iirst quite possible, however, that the nundjer of low-pressure areas may, in low-pressure years, be increased by assmning that the mean track of the anticyclones is a little nearer the equator at those epochs. The low- pressure areas would thus have a slightly more northerly course, and would therefore be more felt and therefore more prominently recorded by the South Australian barometers, which would then not be so much under the influence of the anticyclones. There does not, however, seem to be sufficient evidence at present to ^varrant such an assumption, for the data available do not indicate that years of deficient pressure in South Australia correspond to yeai's when the mean track of the anticyelonic belt occupies a more northerly position. Russell some years ago determined the mean monthly and annual latituihis of the anticyclones, so it is possible to compare these values directly with the pressure values for Adelaide. Quart. Jour. Roy. Met. Soc, Vol. XIX., No. 85, January 1893, page 24. A DISCrSSION OK AI'STKAIJAN METEOROLOGY. 83 111 the acconipaiiyijii;- luhli- these data are 2)laeed in vertical colunins :- 1889 1890 1891 1892 1893 1894 1 89.5 1896 1897 1898 1 899 U)()() Year. Iialitii(k'< of 33-6 S. 33 1 34- 1 34-4 35-9 34-3 34-4 34 o 33 -T 34-9 3.5-8 34 7 Barometric I'resKiiiv. As, 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 roimd 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 Aveeks, of weather passing over Sjaluey. Certainly the discovery of the daily translation of anticyclones in our latitude, over such a large section of the circund'erence of the glo])e, 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 * Three Essays on Aiistniliini Weathei-. "Moving Anticy(;lon<'s in the Soufhi-rn Homispiiere," by H. C. Russell. 1893. pajre 9. 88 SOLAR PHYSICS COMMITTEPI ill any other part ol tlie earth for normal atuiosphenc circulation. 1 tlo not l.)y this intend to convey the idea that 1 think an anticyclone keeps its shape, size, forn), and pecnliarities for weeks together, becanse I see them changing every day. But nevertheless there are obvious peculiarities 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." In order, if possible, to throw further light on this question by employing more recent data, the barometric conditions which prevail during a year in South Africa and South America have been examined. Dealing witli South Africa in the first instance, the inquiry is facilitated by the recent publication of an important paper on "The Barometer in Soutli Africa." by R. T. A. Lines.* Reading this paper with the knowledge of Australian barometric changes, it is found that practically the same kind of conditions prevail, remembering at the same time that the most extreme southern portion of South Africa is in latitude 34° 50' S., while that of Austraha is about 39" S. Just as Australian weather is controlled by a sequence of anticyclones passing from west to east, and that of its southern portion by inverted Y-shaped depressions carried along between them, so that of South Africa is dominated by similar conditions. A comparison made between the seasons at each oE these countries brings out their similarity very closely. Taking the winter conditions (April to September) first, the anticyclones take a course nearer the equator in both tliese regions. In both countries high pressure prevails, while the southern " lows " in l)otli cases skirt the southern coasts. The effect of these lows is felt more in Australia than in South Africa, because the former country extends further to the south and l)ecomes more enveloped in them. During this season both countries receive their rainfall on the west, south- west, and southern districts, brought bj' these southern lows. During the summer months (October to March) the auticyclonic track in both countries occupies higher latitudes, and the pressure is reduced in each case. The southern lows are now kept well south, and in neither country can they affect the laud areas. This southern position of the auticyclonic track is favourable in Australia for the monsoonal " lows " to recurve from their south-east course and striking the country as if they came from the north-east, t(j pursue afterwards a path lying in a south-easterly direction. These monsoonal " lows " * Report of the South African Association for the Advancement of Science. 1906. A Disrrssio.v of .\rsTi;.\i,i.\N me'it.ouoi.ogy. 8er of ship's logs. He found that the cyclonic disturbances followed paths which are south of the 4.3rd parallel during the winter months, and south of the 46th parallel during the sunnner months. This change of latitude of the cyclonic track thus harmonises with the move- ments already referred to, when they pass near the continent of Australia. To make the most favourable passages from the Cape to Australia it is necessary to take a course on which the greatest amount of westerly wind can be experienced. Such a course will be that which skirts the northern portion of the cyclones. This, according to Commander Hepworth, is to the northward of the 42jid parallel in the winter months, and somewhat to the northward of the 40th parallel in the summer months. This change in the seasonal position of the tracks of these southern depressions is, with little doubt, regulated by the sequence of anticyclonic areas which lie adjacent to the northern parts of the depressions. Westerly winds will therefore be found just as nmch on the southern borders of the anticyclones as on the northern portion of the cyclones, although, perhaps, they may be stronger in the latter case than in the former. A l)ISCrSS[ON OF AT^STRALTAN ilETEOROLOGY, 91 I'oimuuiicler llepworth gives many instances of the easterly progress of those southerly depressions, which seem to clearly show that in this southern part of the ocean we have to deal with a series of separateLij moving cyclonic xchirU, as is experienced on land, and not a simple extensive air movement from west to east. One example will illustrate the nature and movement of such depressions : — " These systems of low atmospheric pressure frequently travel to the eastward, for tlays, at so slow a rate of speed that a steamer running on the left-hand side of their centres will often keep up with them for himdreds of miles, sometimes overtake them, and not unireqiiently leave them astern. In July 1887, the S.S. Port Pirie ran on the left front of such a system for upwards of 1,200 nautical miles, or nearly 1,400 statute miles, and would in all probability have remained imder its influence for hundreds of miles more had she continued her onward progress ; but having been stopped and hove-to under canvas in order to effect some repair to machinery, the trough of the depression within three hours passed the ship. . . . The Port Pirie had been making about 300 miles per day whih^ in company with this system." In this case the velocity of the depression is quite in accordance with the experience of South Australia. Coming now to the South American region, it is found that practically the same sequence of changes occurs during a year as has been shoAvn to be the case in South Africa and Australia. As the extreme southern portion of South America is in about latitude 50° S., the belt lying between latitudes 20° S. to 40° S. need only be considered here for comparison with the other two countries. The data Avhich have been employed have, in the case of the Argentine Republic, been taken from the valuable memoir on " Climate of the Argentine Republic," l)y Mr. Walter G. Davis, director of the Argentine Meteorological (jffice, and, in the case of Chili, from various numbers of the " Meteorologische Zeitschi'ift." Dealing with the pressure conditions first, it is very clearly shown that the country lying between latitiides 25° S. and 35° S. is continually being swept by a succession of anticyclones, while to the soiith, about latitudes 40° to 50° S., a series of cyclones transit the country. Although it is difficult to determine the mean latitudes of the summer and winter nnticyclones without analysing a great nund)er of daily weather charts one is able to see that the tracks of the anticyclones and their attendant cyclones exhibit a march in latitude according to the season of the year. In the winter months they travel nearer to the equator than they do during the summer months, as was seen to be the case with those in South Africa and Australia. Most interesting is a study of the mean annual pressure variation- curves of several stations in the Argentine Republic arranged in order of latitude. Here M 2 02 80T.au rfTYSICS C():\i:\HTTEE. is found, just as was experienced in Australia, that the curves have single maxima in the year at stations in low latitudes, while in higher latitudes the single maximum is converted into a double one. For Australia the latitude where this change appeared to commence was about 24° ; in the case of South America the latitude is about 32°. (See page 26.) Turning our attention now to the rainfall of the two I'egions, the Argentine Republic and Chili, we here find two quite distinct seasonal rainfalls. In the Argentine Republic north of latitude 38° S. the rainy season comprises the summer months, October to March, the maximum fall occurring in January. In C^hili, in the same latitudes, the rainy season is confined to the winter months, the maximum fall taking place in June. Further south, about latitude 40° S., regions are met with which receive rain at both seasons of the year. It is important to note that when the rainfall occurs on the west coast the path of the anticyclones is nearer the equator and affords the southern cyclones an opporttinity of reaching lower latitudes and watering the country. \i\ the other season of the year, when the east coast receives its annual rainfall, the anticyclones are in higher latitudes, and thus allow the monsoonal lows from the South x'Vtlantic ocean to reach higher latitudes. Whether these lows originate in the south-east trades, recurve and reach the Argentine Republic from a north- easterly direction, and eventually pass away in a south-easterly direction, is a matter for investigation, but it seems very probable that such may be the case. The comparison of the general weather conditions prevailing in these different countries, lying on about the same parallel of latitude, indicates that great similarity exists between them. If, therefore, each of these countries has series of anticyclones passing across them in about the same latitudes, and from west to east, it seems quite reasonable to suppose that valuable meteorological information for any one of these countries could be secured by inquiring about the conditions which prevailed in the country lying more towards the west. It has been previously indicated that when the pressure over Australia in any one year is in excess, that over South America is deficient, and vicx versa. If, therefore, the anticyclones do encircde the earth in the parallel of latitude just referred to, they must suffer some modification as they travel alternately through these regions of relatively high and low pressure. The following summary brings together the main feature regarding the similarity of the conditions prevailing in the three regions, namelj^ Aiistralia, South Africa, and Soiith America : — 1. A successive series of anticyclones travelling from west to east cross the countries between latitudes 20° and 35° S. 2. A series of successive " lows," bordering on the south side of the anticyclones, pass also eastward. A DISCrsSKlN OF Al'STIfAMAN METEOROLOGY. d?> .'3. Tlie monsooiial " lows " recurve Iroiia the soutli-oast trades and enter the countries from the north-east or north directions. 4. A seasonal movement in latitude of the anticyclonic tracks takes place, and this controls the positions of the cyclonic tracks. 5. The rainfall on all the east coasts occurs in the same season of the year, namely, October to March. 0. The rainfall on all the west coasts takes place during the winter months, April to September. 7. Each country has a region which receives rainfall during both of these seasons during a year. In the accompanying figure (Fig. 10) an attempt has been made to illustrate at a glance in diagram form the positions of the anticyclonic tracks and the _v ■' -_ OCT- mar; Fig. 10. regions of the southern and monsoonal low-pressure disturbances, Avith the accompanying regional rainfall for the two seasons, winter and summer. The small arrows in the isobars indicate the general movement of the air in the systems, while the large arrows show the direction in which they travel. The tendency of the low-pressure systems is to travel with the anticyclones from west to east, but, at the same time, they are trying to wedge themselves in between them and pass through to the opposite side. Both the southern and monsoonal " lows " endeavour to do this, and often succeed. A glance at Fig. 10 will show the relation between the two wind systems, indicating the advisability of the shipping routes being restricted to the north of the cyclones or to the south of the anticyclones, thus experiencing in both cases Avesterly winds. It will be seen, further, that if a more southerly route were chosen, ships would eitlier have to pass through the centres of these depressions and experience heavy weather or traverse their southerly portions and meet easterly or opposing winds. !M SOLAR PHYSICS COM:\nTTEE. The cunchision that all tlie three countries, luunely, South America, South Africa, and Australia, between ahont latitnrles 20° S. to o5° S., experience series of anticyclones travelling from west to east, i-enders it very difficult to believe that these systems do not circulate roinid the earth in those latitudes. During the winter months, that is, when the anticyclones are nearer the equator, these systems are also distinctly felt at Alauritius (latitude 20° 5' S.). So clearly are they pronounced in the barometric records there that one can state with certainty that they affect Mauritius aljout three days after they have been recorded at Durban. This additional evidence, and more is being collected, still further endorses the view that they ci-oss the soiithern oceans as individual systems, as Russell siiggested. APPENDIX. X A DiscrssioN OF .M'S'ri;.\T,i.\x .Mi-;rF,oi;()T.or;y. !>7 Table 1. PRESSURE.— YEARLY. (Jan la in k. December.) Vear. ^ •oit viwlii. W ins. + 29 Daly aters. York. 1 Carnarvon. j Albany. Kucla. Deuiliquin. Goulburn. J^^^^^ ' 29 . . .- 1 ii;s, + 30 ins. 4- 1 29 :ns. + 30 ins + 30 ins. + 30 ins. + 30 ins. + ' 30 ins. -(- isdi - - . ' - - - - - ' _ ■112 „ 18(iS - ■ - 1 - - _ — 1 1 _. ■125 _ 1 _ 186'.) - - - - _ _ ■176 _ _ 187(1 - - - ' - - - - i _ '' _ •118 29 937 _ ),S71 - " - - - - ■127 1 30 (169 _ 1S7L' - - - - - - •133 i 069 _ l,s7:i - - - ' - - - _ - - •132 075 _ 1871 - - - - - _ _ ! _ ■085 ' 072 _ 187.-, - - - - - - - •030 044 „ 1871; - - - - - - •064 077 _ 1877 - - . _ _ •163 017 _ 1878 - 827 — - - 1 i _ ■047 013 _ 1879 - - - 810 - • 028 ■034 039 . 1880 - - - ', 838 - _ 046 050 •070 i 026 061 1881 - - - 1 ! 90.5 ~ - 088 090 ■126 ' 086 096 1882 ■ - - : 822 914 •016 _ 031 034 ■085 040 031 1883 - 819 j 933 ■042 ' 041 050 •094 053 054 1884 844 9.52 _ 014 056 •124 •fKJS TI«T 188", - 8(i2 980 ■065 ■980 093 090 •146 106 115 1880 - ■ - 813 8!K) - •940 071 061 •119 064 (I(i4 1887 . - - |i 829 897 ■049 •962 073 056 ■100 1 052 066 1888 - ■ ■ \ 850 931 ■065 •987 096 089 ■141 093 104 1889 - 831 891 ■010 •964 026 027 ■107 062 045 1890 - - - 'l 813 873 •001 •9.51 034 018 ■084 039 OOrt 1891 - - - ' 862 905 ■068 ■998 108 - ■153 1 064 079 1892 ■ - - ! 826 866 ■036 •960 056 045 ■108 , 035 , 029 189S - - - '■ 826 861 ■001 •9S2 010 , 010 •071 024 004 1894 - 831 880 ■0.54 ■940 076 065 •1.52 069 043 189.5 - 861 893 ■057 _ 0.58 ! 061 •144 072 051 189(i - 894 914 ■049 ■973 058 066 •138 070 062 1897 - 878 901 •056 ■962 056 065 •126 066 i 070 1898 - - - 847 863 ■016 _ 021 017 •094 06O 026 1899 - 899 908 ■048 ■984 056 051 •106 092 070 1900 - - - 903 1 915 ■056 ^002 062 086 ■066 074 052 1901 - 898 899 ■054 1004 078 (1.58 ■107 102 074 1902 922 919 •071 1019 : 077 060 ■109 109 085 19o:! - 883 878 •026 ■966 038 017 - _ 040 1904 - ■ ■ 887 896 ■042 •997 039 0(1 _ _ 063 r.)o,-, - - !l 1 920 910 ■070 1-024 i ( 077 i 067 " 1 072 E 142. ^" 98 SOJ.Wl IMIVSKS ('()^[:\riTTEE. Table 2. PRESSURE.— PERTH. Lat. S. 31° oT, Ldxc. E. llo" 4o'. Ellvatiox hi- C'istkiin 197 i't. Aftek 1884, Meav of 9 a.m. axd 3 i\M. N()(.).\ ivKAI)I\(iS. •\r„„„ 29 inc hes + Mean tor Year. — J Ian. I 'eb. Ma-ch. April. May. June. Julv. Aug. Sept. Oct. Nov. Dec. Trar. Four Years. 187G 818 931 •971 1135 1138 1-151 1-229 1-1.50 1-099 I-02it -928 •996 1-048 1877 931 931 103 1 1.59 1-073 1-327 140 1-191 232 103 1 072 1-(H)5 106 1H78 939 9(i0 995 056 1-133 1-071 972 1-090 006 997 1 005 -907 Oil u.-u 044 030 032 031 038 039 056 ((63 072 065 048 052 044 040 051 048 050 063 053 047 044 043 058 056 059 nn' t 187!) 9(t4 924 052 276 ■969 1-022 14! 1135 091 137 984 -897 044 1880 814 877 934 028 1'047 1-067 179 1-069 104 075 990 -972 013 1881 913 971 019 098 1-087 1-136 176 1-210 140 053 915 •897 051 1882 wr.K ! 936 964 930 1-109 1-135 127 1-014 101 038 1 010 -903 1 019 1883 948 9.51 023 0.50 -995 -971 200 1-126 207 094 971 -9.52 041 1884 9.-,l 959 962 054 1144 1-060 211 1-074 094 077 958 -953 041 188.-. 974 960 030 110 1046 1-186 090 1-078 , 134 110 1 048 -922 1 057 188(: - 1 99t> !i28 061 170 1-150 1 ^.-.a 134 1-048 044 146 1 091 1-012 086 1887 9r)(> 960 048 129 1-138 1-129 127 l-0<)7 118 118 1 018 1-004 068 1888 908 994 I 075 123 1-111 1-095 216 1-176 118 112 1 012 -993 078 188!» - 1 008 974 084 050 1-036 -969 173 1-120 036 044 937 -912 029 189(1 904 917 040 084 1043 1-024 162 1-098 020 915 1 065 -941 OlS 1891 - ' 928 983 036 132 1-065 1-106 258 1-216 150 074 1 0(;6 •982 083 1892 93.5 976 926 162 1 - 136 1 - 196 094 1-008 100 084 1 005 -954 . 048 1893 890 879 045 976 1-027 1-121 038 1 - 194 006 010 1 022 -918 010 1894 934 1 017 oor. 135 1-160 1-114 194 1-096 096 070 1 006 -924 063 189.^ - i 927 982 102 157 1 - 206 1-088 138 1-035 097 080 1 053 •973 070 189(! 872 954 956 138 1-171 1-0.58 102 1-164 178 100 984 1013 • 058 1897 939 990 029 106 1-1.54 1-028 180 1-183 087 080 1 014 -956 062 1898 910 860 964 200 1122 1-076 120 1-008 085 966 1 007 •931 1 021 1899 944 944 964 041 1-168 1-062 130 1-186 142 968 1 042 •982 1 048 1 900 980 975 088 054 1-168 ■928 090 1-046 172 043 986 1-007 045 1901 953 915 018 U71 1-108 1-103 204 1-101 136 102 1 046 -945 058 1902 884 ; 990 040 120 1-104 1-186 188 1-209 088 116 1 082 -982 082 1903 - 1 000 1 000 035 060 1-156 1-111 155 1-092 009 966 970 -909 1 041 1904 938 966 983 086 1-169 1-0.59 112 1-156 088 031 1 093 -966 054 1 190.5 994 1 036 100 1104 - 932 1-181 1-157 1-196 1-082 118 1-050 -984 1-078 ) 1906 918 950 024 A |)|S('1-SS[(A' OF AUS'I'KAIJAX MKTKOK'OIXH ; V. 09 Taislk 2 — CA))d'niued. PRESSURE.— .MELBOURNE. LvT. S. .i? .')0', LoN(i. E. 111°.W. Klkv.vtion of Cistekn 91 -JJ ir. Reuuckd th A'lV. 29 inches + Mean fur ' ,] a:i. F 1 cb. XI ivch. April. Miiy. •111 no. .J Illy. Aug. Sept. Oct. Nov. Deo. Y ear. Four Yeari'. i8r)9 838 1 833 943 1-014 •893 1-066 1143 •934 •969 860 953 •847 941 18(!(l 73:1 933 915 •952 roo8 -917 I -195 1168 •929 958 835 -844 952 1861 833 79! 956 1-008 •981 -832 -965 P078 ■!i42 837 928 -749 900 V - 928 1862 825 851 917 •979 -923 1-012 •832 P060 -825 1 047 888 -839 918 1 -916 1863 - 1 816 849 921 l-09ii •951 1-062 •868 •971 •978 652 , 780 •783 896 ' -914 186+ - : K26 895 1 ((71 ■966 1-129 1-046 •905 ■921 -907 903 909 •848 •944 - 923 I86r, - ; 889 863, 924 1004 ■ 952 1106 1-006 ro.-)2 •831 959 883 •764 •936 -932 -938 1866 85.-) 896 1 0311 1-069 •976 1-123 1-003 1014 ■S73 797 878 •903 -954 1867 - S92 855 1 024 1-012 1-107 1 -033 -917 r099 ■743 672 91(1 ■752 -918 ■ 946 1868 837 937 957 1-096 1-2.56 ■932 1-068 -933 ■ 936 917 -967 •833 -977 -947 -941 -943 ■929 -931 -931 1869 7.50 1 878 987 1-081 -955 1-081 1181 1-0.50 1-076 806 -779 ■737 -938 1870 - 1 801 ! 911 1 050 -967 1-039 -916 ro.53 ■ 793 -925 955 -872 -876 •930 1871 - i 810 800 995 1-059 1 - 005 ■960 •886 •930 -9.58 -96S 854 -842 -925 1872 - 1 806 963 925 1-021 1-00 1 -808 •902 roi8 1-064 •919 -s(;7 -784 •923 1873 - S 942 885 983 -956 •980 1-045 1108 •948 -861 886 -919 -819 •944 1871 8M4 910 965 1-0.50 •959 -995 1-069 •893 -790 I 999 -834 -815 •930 -921 -923 •935 •929 •937 •936 •930 -929 •929 •933 -940 -955 •961 •973 -9.59 -951 -961 •944 •929 •929 •915 •921 •937 -934 •939 •932 . -932 1875 788 899 961 •964 •909 -866 1128 -881 •971 •818 - 709 -743 •886 1876 781 ; 877 921 •923 1 • 102 1-051 1-122 1-036 -920 •826 -741 ■866 -931 1877 835 908 1 033 1-070 •844 1-212 1-188 1-075 I -071 1 007 -858 •821 -993 1878 953 946 955 •978 1^068 -909 -910 ■912 -804 808 •853 -769 •905 1879 867 839 946 1^092 •891 1-0.54 1-035 1-020 •911 875 -736 •766 •919 1880 857 925 889 1-019 •887 •996 1-069 •879 •941 849 •9.52 -871 -928 1881 833 939 1 003 1-089 1-019 -939 1-180 1-075 •962 948 812 -794 -966 1882 777 959 881 -908 -934 -968 •942 -960 -M57 865 987 -787 902 1883 891) 745 977 1-041 -932 •991 1-021 •962 •929 900 875 -759 919 1884 783 899 1 030 1-075 1-066 -971 1-166 -902 -921 891 946 -6.84 944 1885 888 841 925 1-148 1-074 1-048 1-150 -907 ■997 1 050 978 -949 996 1886 885 838 1 022 1-007 1-038 1-25.1 1-136 ■ 783 -961 777 919 -889 959 1887 808 856 926 1-125 1-070 -891 -918 1-125 ■834 882 952 -945 944 1888 841 907 972 1-177 1-099 -999 -942 1-048 1^082 1 046 911 -884 992 1889 S62 887 1 015 1-042 1-08-1 -7.58 1-158 1-020 ■937 935 844 -777 943 1890 910 870 1 005 1-105 1112 -899 •993 -919 ■882 663 861 •868 924 1891 809 921 1 025 1154 1-233 -980 1^027 1-065 1-029 894 945 ■753 985 1892 86(i 933 910 1-007 1-036 1-018 1 -064 •903 -856 889 840 -781 923 1893 779 840 1 013 -870 -883 ■993 •926 1039 •794 810 833 •805 882 1894 - ' 784 926 920 1023 1-027 -965 •886 -911 -960 898 899 •916 925 I89."i 866 882 1 058 1-017 1-103 •999 •905 -865 •824 915 984 •722 928 1896 817 980 910 •910 1053 -976 ■881 1006 1-027 \ 958 984 •893 948 1897 799 894 956 •992 1-029 1-149 1104 •960 •910 785 868 -911 946 1898 8-tl 806 940 P0I9 1-087 1-033 ■940 1119 •885 758 694 -841 913 1K99 706 918 900 -981 1 022 •962 1170 1111 1026 953 778 -874 950 1900 852 929 928 •969 1-113 -886 •967 •739 1-009 864 • 921 -845 918 1901 811 949 912 ■981 1106 •939 1-105 r(K>8 •877 879 975 -837 948 1 1902 - ' - - - - - - - - - - - - - 1903 ~ - - - - - - - - - - - - 1901 ' 1 N 2 100 SOT-AIJ I'MYSrCS ('()M:\1ITTEF,. Table 2 — contlmicd. PRESSURE.— SYDNEY. L.iT. S. 33° 52', Lox(4. E. 152" 11'. Elkvaxiox ok Cistkkn 155 it. Reducki) to '.VI F. Vcar 29 iuohes + Jan. I Feb. , Mar. April, i May. ; June. July. [ Aujr. \ Supt. ; Oct. Kov, ] Dec. 1858 - ■775 ■919 •970 1014 ■829 1022 ■917 ■764 ■745 1859 - - ■815 ■ 755 •851 ■933 ■888 •971 1015 ■971 ■875 ISCiO - •682 •830 ■8.H9 ■875 ■944 ■823 1032 M05 ■941 I8(;i - - ■765 ■698 •886 ■914 ■896 ■856 ■801 ■991 ■ 952 1862 - - ' ■763 ■770 ■862 ■870 ■8.57! ■921 ■813 ■ 952 ■ 783 1 803 - - ■760 •714 •879 1009 ■ 930 ■971 ■898 ■865 ■ 968 lSli4 - •737 •801 •967 ■89(i 1 ■ 059 ■910 ■818 ■817 -838 18(i5 - • ■782 •784 •844 ■962 ■894 1043 ■917 I ■001 ■780 18r)<> - •■773 •834 ■998 1001 ■957 roi9 ■912 ■999 ■812 i8<;7 - - ■790 ■776 ■933 ■ 885 ■975 1^008 ■928 \-Q\i ■698 i8t;s - ■762 ■838 ■899 1^026 1175 ■900 roo7 •900 ■925 18()9 - - ■661 •805 ■916 ■978 ■824 i-ooo r088 roos r0O2 1870- •597 •820 \ ■890 ■873 ■941 ■865 ■ 973 ■928 1-029 1871 - - •787 •713 ■821 ■882 ■770 ■764 ■879 ■965 ■973 1872 - •851 ■863 ■807 ■858 ■897 ■ 756 ■799 •883 1017 1873 - - ■857 ■801 ■874 •844 ■946 928 ■967 ■884 ■ 796 1871 - ■786 ' ■7(il ■872 ■925 ■ 823 ■ 1 02 941 ■801 ■694 1875 - - ■ 706 ■794 1 •857 ■861 ■812 ■825 1-008 ■834 •907 187(i - ■71H ■782 •848 ■882 ■999 ■957 ■945 ■9.56 ■82s 1877 - - ■717 •850 ■961 ■948 ■724 1 117 1075 1^009 -963 1878 - ■853 ■870 ■886 ■921 ■967 ■802 ■8-26 ■857 -772 1879 - - ■769 ■751 •868 1016 ■707 ■927 •938 ■917 •775 1880 - ■776 •8.52 ■793 •940 ■779 ■!M)3 •906 •854 •858 1881 - - ■753 ■84-i ■910 1010 ■975 ■839 1082 ■968 •884 1882 - ■703 •899 •802 •805 ■843 ■856 ■892 •879 -835 1883 - - ■806 •660 •948 •9!59 ■8.52 ■996 ■952 •930 -866 1884- ■697 •838 ■961 ■991 ■997 •941 1 ■ 051 ■874 - e9(> 1885 - - ■807 ■ 755 ■8-22 rOiJ2 ■999 •938 r037 ■906 •965 188f> - ■ 839 ■726 •903 ■906 ■956 1155 1062 ■769 •911 1887 - - ■746 ■ 795 •883 1^047 ■963 ■786 ■888 1018 •794 1888 - ■769 ■843 •857 1116 1019 •990 ■898 •967 1^02ll! 1889 - - ■802 ■803 •9.53 1012 ■982 • 702 1^070 ■975 •885 1890 - ■871 ■ 796 •925 1^003 1^034 ■817 ■894 ■887 -845 1891 - - ■766 ■824 ■966 1027 1148 ■805 ■924 ■968 -949 1892 - ■786 •856 ■810 ■897 •995 ■961 •995 ■827 -784 1893 - - •690 •786 ■967 • 760 •886 ■904 ■880 1-010 •791 1894 - •760 ■868 •889 •972 ■960 ■897 ■;-;36 ■892 •925 1895 - - •799 ■ 853 ■997 ■994 1037 ■951 ■877 ■878 -786 1890 - ■770 ■870 ■861 ■905 ■968 ■832 ■822 ■926 ■ 965 1897 - - ■728 ■813 ■884 ■890 ■9.-)3 1143 ■977 ■905 -859 1898 - ■794 ■776 ■886 ■960 ■961 ■991 ■912 1^0.59 •865 1899 - - ■604 ■942 ■905 ■886 ■948 ■878 1061 ■986 1 - 002 1900 - •836 ■874 ■887 ■903 ■985 •835 •871 ■746 -918 1901 - - ■ 730 ■919 ■866 ■908 1069 •884 JV004 ■939 -905 19(t2 - ■698 ; ^779 ] ^870 ■989 ro58 1^027 roo4 1-065 •846 1903 - - ■835 •788 ■862 •870 •998 ■941 ■894 1024 •784 1904 - ■805 •693 ■912 1069 roi7 ■897 ! ^999 ■978 ■880 1905 - - ■793 •838 ■921 1^042 ■892 ■992 ! ^924 ■985 ■770 1906 - ■812 •879 ■913 ■949 1^033 ■965 ■843 1^019 ■871 1907 - - ■741 •869 ■848 •861 1083 1^011 ■966 ■829 ■851 ■801 ■961 • 643 ■i-02 •857 •732 •6,58 ■851 ■ 752 •927 ■915 ■920 ■829 ■941 ■732 •721 ■891 ■788 ■820 ■778 ■813 ■801 •856 •791 ■986 ■707 ■809 ■931 ■902 ■644 -841 -839 -780 -917 ■1-90 ■911 ■7.54 •750 ■894 ■818 ■868 ■S62 •948 •864 •748 •811 •850 •831 ■770 ■862 ■805 •678 ■808 ■ 802 ■ 745 ■801 ■871 ■017 ■771 •808 ■844 ■779 ■ 752 ■63! ■ 6-'S0 •764 •761 ■633 ■ 7.59 ■716 ■930 ■842 ■867 ■886 ■862 ■871 ■831 ■782 ■ 760 •867 ■793 ■777 ■832 ■897 ■ 9.50 •823 •679 •732 •857 •906 ■873 ■880 -806 ■810 ■714 ■7,30 •803 ■783 ■686 ■726 ■728 ■ 7.53 ■623 ■818 ■653 ■732 ■690 ■772 ■844 ■722 ■ 755 ■720 ■662 ■785 ■732 ■673 ■ 668 ■771 692 ■703 ■677 ■.592 ■887 ■831 ■910 ■806 ■ 705 ■7-<2 ■680 ■702 ■718 ■876 ■6-13 •815 •873 ■790 ■814 ■796 ■768 ■727 •730 •788 •783 •839 Mean for Year. April- Oct.- I Kmir Sept. I Mar. 1 Yiv.rs. ■883 ■878 ■873 ■842 ■840 ■K37 ■8.50 -857 ■8,S3 ■843 ■907 ■895 ■866 ■844 ■851 •8.55 •827 •803 •833 ■896 ■831 ■816 ■831 ■874 ■829 ■862 ■875 ■920 ■888 ■878 ■921 ■881 ■8.55 ■897 ■8,54 ■829 ■885 ■878 ■885 ■883 •852 ■888 ■860 ■902 ■900 ■880 ■892 ■875 ■887 •898 ■942 ■953 •902 ■866 ■940 ■,S90 ■933 ■9,50 ■918 ■989 ■983 ■935 ■872 ■868 ■894 ■848 ■875 ■919 ■973 ■857 ■880i -873! ■ 960 i ■8,52 ■926 •9,58 •983 •965 •921 r002 ■938 ■913 ■970 ■910 ■872 ■914 ■920 ■903 ■955 ■9,58 ■960 ■876 •951 •998 -918 •973 •934 •947 •828 •802 • 792 •791 -807 - 759 • 795 •815 • 799 •768 ■806 ■794 ■799 ■ 848 •836 •797 • 795 •728 ■777 ■ 833 • 7(!8 ■757 ■ 802 ■ 770 ■ 808 ■812 ■772 •871 ■804 •843 ■854 •830 ■790 ■ 806 ■796 ■799 ■879 ■829 ■8,55 ■818 ■778 ■840 ■831 ■ 825 ■825 ■828 ■ 835 ■824 ■804 • -877 - S59 -S4S ■S+2 -S15 •S5ii - SCi.') ■ NSI'i -89i; ■ SS( I - 875 ■854 -814 -846 -s;i4 ■ S2',l ■S(ii -Stl ■814 ■843 ■838 ■837 ■ 849 •8611 ■871 ■886 ■890 ■902 •892 •884 •889 •872 ■859 ■866 ■861 ■ 869 ■ 883 •874 •877 •871 ■875 ■888 ■886 , ^894 ■887 V884 A DISCI'SSinX OF ArSTi;,\l,l.\X MKTK'VROT.OCY Tarle 2 — rout i Intnl. riJESSURE.— ADELAIDE. Lat. S. 34 56', LoN(i. K. llW Mean- Ska J^kvk .'{.v. Kl.KVATIOX Ol- ClSTEIiX . Mean ok 9.0 a.>i. and ;i.() 40 KT. Keducei) to I'.M. OliSEUVATlOXS. ■.i-2' V. 29 incln'8 + Year. I«.i7 18.-,8 IK.V.t ISfld iwn 1W>2 1 »(VA l,Sfi4 i>%r. 1,S<>(> l**l>7 IKCS 1 SKI l.S7(l 1871 1872 187a 1874 187;") 1871! 1877 1878 187!l 188(» 1881 1882 1883 I8s4 188,-) 188(1 1887 1888 188!» 18!)ll 18!tl 18112 1 8y:i 18114 18!C) l8!)fi 18!t7 1898 18!(9 IIMH) 11101 1!MI2 111(13 1!I(I4 litO.J 1 <)()(; Mean for Jan. Feb. Mar. Apr. | May. ] June, j July. Aug Sept. ■998 •883 ■94.5 •825 •936 ■909 ■912 •937i ■990; ■936 ■979 •9.52 •862 ■889 ■916 ■849 1 •971 ■9.-)l 1 ■ 890 ■911 ■9.58 1()32 1 ■9.59 ■942 •910 1 •911 1 roo2 •9()0 •996 ■974 ■893 •931 1 •984 I ■931 •932 1 ■995 1 •930 ■914 1 ■935 ■942; ■966: •946 •9lH) •972 1 ■949 •876: 1(127 ■947, •943 1 ■946 •874 ■972 •0(»2 I •050 •967 P044 •(t26 •994 •912 1-012 •933 I ■oo,s •958 •99(1 •970 M31 • 998 1((42 •972 1118 ■940 1(»96 • 999 1()42 •995 r(l76 ■942 1 -098 ■850 IIOO ■022 1-025 ■962 1^091 ■030 1-072 • 939 1-09I •992 1-017 ■997 i-132 ■042 1 •090 •9.52 P(.I78 ■981 1-011 ■052 1-1.57 ■039 •995 ■896 1-09I ■990 1-095 ■978 1-072 ■988 1-138 ■942 1-050 ■008 1-134 ■004 1109 ■963 1-083 ■026 1112 ■0.32 ■996 ■970 1 ■ 1 10 ■058 100(1 ■991 1186 •994 1^03o •990 1-110 •8r>o M48 ■986 1-013 ■033 1056 ■990 1-058 ■986 1-099 ■991 1-00(1 •898 1104 ■023 1-128 ■944 1-060 1-196 I -185 M82 1^075 1-08(1 I •126 1-1.53 1-0.56 1102 1168 1101 4-201 1-165 1-047 1-144 1-138 1-091 1162 1-076 1-095 1-213 1-110 1-196 1-123 1-222 1-046 1-168 1-184 1-238 r 113 {•■>m 1^256 1-144 1-210 1-250 1-201 •984 1-162 1-126 1-0.58 1142 1-189 1 -09(1 1-121 1-1.54 1-204 1-062 1-186 1156 •B45 •993 1 •072 1 •197i'l •107 - 034 i 1 -021 ; 1 ■230 1 •145 I •073J1 •163 1 ■3311 1 ■075; 1 100 1 ■074 1 -119 ■0.59 1 091 1 (166 2-22 i 1 996 ; 1 220 i 1 (157 073 1 36 097 (127 185 187 186 194 204 174 174 298 151 994 192 272 214 180 218 I 176 1 260 1 238 1 250 1 242 1 184 1 070 1 -966 1 ■1G2 1 ■191 1 •016 1 -930 1 -142 1 ■1.50 1 173 1 ■269 1 236 1 (158 1 (102 1 207 1 080 1 076 1 924 1 171 1 066 1 970 1 203 1 1 396 1 1 123 1 189:1' 186, 1' 129 il 1.58! I' 136 07(1 231 :-i61 051 1.57 938 009 132 182 094 114 172 134 202 102 114 013 110 189 131 090 174 -0.54 -175 •271 •201 -(J50 -013 ■063 ■076 ■157 '1(»6 (156 ■187 301 140 077 055 255 193 269 276 304 046 172 243 343 145 I 212 330 261 180 119 118 274 104 195 191 lol 098 068 046 188 128 290 120 257 233 078 230 158 1 1 • 192 : 1 1-003 I 1 1 - 102 1 1-271 1 1-163 1 1-191 1-098 1 1-082 1 1-178 M83 1 1 ■ 230 1 131 1 1193 1 ■893 1 1^066 1 M87 1 1-088 1 I (125 1-064 I 1197 1 1-235 1 1111 113^i 1 1-065 1 1-21(1 1 1-087 1 1-116 1 1-0(19 1 1-056 1 -886 1 1-212 1 1-176 1 1-134 1 1-058 1 1-218 1 1-032 1-214 1-078 1 1-032 1-190 1 1119 I 1-200 1 1-238 1 -900 1 1-133 1 1-302 1 1-132, 11711 P167 1 -118 ■001 -095 ■045 ■083 ■948 ■103 ■027 -991 ■035 ■883 ■ 024 ■200 ■0119 ■071 ■171 ■033 ■ 936 ■141 ■095 ■261 ■966 ■064 ■113 ■107 ■063 ■110 ■071 •1.50, ■058 •025 •196 •066 ■020 ■1.50 ■980 ■9(>5 ■130 ■994 ■186 ■068 ■ 050 ■1(11 ■1.50 •026 ■ 055 ■952 ■113 ■062 Oct. \-0(H 1122 1-008 P048 ■949 1108 •809 1031 1 ■ 103 -922 -844 1-024 -912 1-015 1-079 1-043 1-021 1-090 •980 1-000 M69 •968 P007 roo6 M05 •995 P064 1-0.56 1-146 - 972 1051 l^l&l 1-035 •844 1^032 •976 •961 1-004 1-069 1086 •9.56 •902 1-082 1039 1-039 1-020 10221 •9851 1035 Nov. Dec. \, Tear. Apr.- Oct.- Sept. Mar. 1-020 -963 1-043 -951 1-030 -968 •907 ro2i •989 r(MI9 1()63 1^074 •883 1-024 -941 ■932 1(1(14 ■988 •914 ■918 1 1^0(12 1 P031 ■896 ro27 1 •970 P(»85 •990 I 1^040 1^ 124,1 ro36 r 024,1 roil ■920' 1 ■ 034 ro8o -960 -9801 I -060! 1-116 1-054I1 roi4j -901 -922 1 1-026; 1-064 1-016 ■976 1-034 1-025 ■942 ■906 •940 •917 ■870 ■917 905 ■987 ■914 ■980 ■931 -988 ■ 854 ■9lil ■920 ■ 924 -940 ■938 ■938 ■005 ■000 ■934 ■913 -027 ■932 ■942 ■907 ■849 ■012 -979 -022 ■972 ■916 ■998 ■902 ■952 ■960 •995 •884 •010 ■981 ■928 ■005 •975 ■943 •900 •918 •9.59 -990 1062 1(138 1^072 I 043 roil 1025 P"06 r060 1073 1()62 1029 1()80 1062 I112I 1(126 11132 1()62 1045 P028 ro78 1144 1^05(1 P 0.521 roiiei 1107 P047 1 • 060 1-075 1121 1^075 1-066 1112 1-058 1-036 1-111 1-0.54 1022 ro67 1070 1079 1076 1()38 ro84 1055 1(180 1(194 1(144 1(175 P078 I 145 POSfi I ■ 1 52 1-134 I 071 1-076 1-098 1107 1 1140 1 1-133 1082 1146 1 1 ■ 190 1055 1 ■ 085 1-099 1116 1 1(179 r098 1181 1 1-234 1 1-096 1-136 1134 1 1192: 1 -09911 1-128:1 1-153 1-187 1 1136 1135 1 1-184 1 11221 1-096: 1-207 1 1-123' ro.59 1129 1 1111 1 1-13811 11.50! 1148 I-I8O' 1 1-092 1 1153 1205 1100 M()2 1131 1 -994 •991 ■972 ■963 ■9.50 ■976 943 ■Oil ■005 ■988 ■ 972 ■003 ■935 -978 ■972 ■987 ■013 -989 -9.58 ■002 ■066 ■987 •9.58 ■030 •992 ■002 ■001 ■998 064 ■979 ■028 (144 ■975 991 '006 9S3 979 028 (107 036 966 942 012 '006 (101 992 977 995 000 Aug.- July. 11)65 1 0.55 1043 1023 r022 1 (132 1(»33 P071 1()65 1 -044 1 (155 1(»77 1031 roi7 1-017 1-071 1(159 P023 r063 1 ■ 101 M16! 1-051 1-048 1099 1O60 ro.59 084 Nov.- Four Apr. Years. 1 ro87 1-121 1-032 1-095 1-0971 1-045; 1075 ro94' 1-007 11151 MI85 1 -043 1109 1(151 1 -048 1-082 1-070 1-086 1-069 1-053 1-068 1014 1()01 ■984 ■9fi9 ■9711 ■984 •984 1(123 101(1 P018 I (131 I •027 •9.52 ■999 •982 •995 1()36 ■987 ■978 1037 1(1.56 ro25 • 978 1^066 •982 1031 1-031 1029 r063 1-018 1062 ro37 1004 r059 1 ■ 034 -984 1012 1-049 1 005 1045 100.5 ■974 l-(>\x 1-025 1029 -999 1-005 1041 y 1 ■ 055 1042 1039 1-023 1(125 104I 1O50 1 056 1-061 1058 1()48 l(t47 1035 1-035 1041 1-042 10.53 1073 1 07(1 1 ■((82 \-OT.I [•070 ro(i8 1-070 1072 1-076 1083 1-084 1093 1-078 l-0()8 1079 1065 I 0.56 1063 10.53 1059 1073 I Odd 1(169 1-063 1-064 1078 I-(I68 P074 /I •074 102 SOT.AK PHYSICS CO:\nnTTEF,. Table 3. PRESSUHE.— MEAN MONTHLY VALUES. Mean Yeare. Jan. Feb. Mar. April. May. June. July. Aug. Sept. Oct. Xov. Dec. for Year. Ins. Ins. Ins. Ins. Ins. Ins. Ins. Ins. rns. Ins. Ins. Ini. Ins. I'fiit Harwin ISTK-l'JOl 29-738 •7.50 789 840 888 •!;2i 942 -930 905 870 820 - 769 29 847 Wyiiilliam - 1881t-189!) 29-728 •762 804 892 957 •988 30 017 30-000 29 931 878 823 -780 29 880 Derbv 1888-1899 29-750 •775 828 909 990 30-007 30 032 30-016 29 970 912 860 -824 29 906 Cossack 1890-1809 29-709 •744 S02 932 30 008 .30-029 30 (KiS 30-048 29 994 923 8.54 -807 29 910 Alice Springs 1879-1901 29-816 •844 993 30 104 185 •218 248 -206 153 30 OOl 29 917 -851 30 0-12 Carnarvon - 1887-1901 29-789 •809 88*; 972 30 032 -043 OM) - 096 O60 3o 010 29 92(1 - 858 29 965 Geraldton - 188fi-1899 29-890 -898 979 30 078 110 -103 150 ■127 103 30 069 29 999 •940 30 037 Perth - 188r, 1899 29-935 •954 30 024 114 116 •100 144 -112 094 058 ' 30 024 29-961 30 0.-)3 Bunbury - 1885-1899 29-960 •977 .10 038 30 111 107 •075 115 ■079 076 043 I 30 034 29-978 30 050 Sydney 18.-.9 1887 29- 709 •8DI 890 935 913 ■931 954 ■!I27 882 830 807 -742 29 865 Adelairiii - - - , - 1-80 1 • 35 1-97 277 3-51 4^04 360 370 325 2-95 2^20 193 33 ■(19 i W't'ftt Au4rii!iii : Xorth-Wesf - North ... - 4^07 7-57 187 (i-88 289 420 123 •83 ■76 ■85 •76 ■40 ■31 ■27 •23 •06 •10 •33 ■18 L70 ■73 4 37 13 ■OS 27-44 Siiiitli Aiixt rill ill : X.irth ]2-2f> l(i'35 849 2-25 ■32 •06 ■06 •04 ■45 208 464 813 39-23 QiieoiLiliiiiil : North - - - ■ South .... Inlaiiil - . . - 12-82 7-18 4 -IS 1445 (•■14 3^05 1045 5 ■.58 2 75 1013 291 2 32 2 33 259 149 ■99 2^42 1^12 ■7(! 183 ■39 ■68 1^4(; ■45 ■71 186 ■64 1^26 2-19 ■71 4^28 2^81 1-28 461 404 123 63 .50 40 92 1947 Simth Austniliii : . Central Lake Eyre District L7S •93 1-39 •70 ■89 •54 •86 ■71 •89 •62 •45 •65 ■19 •19 •20 ■44 ■27 ■31 •48 •33 •69 •54 ■83 ■52 893 1 5-18 .Veir Siut/i Il'^to.- !iilaiul(N.p:.)- •Joast - - . - 379 4-90 3 33 2-69 5^66 174 5 39 189 6^00 2-36 5-49 201 4 ■.54 1^69 3^40 i 224 372 2^41 317 2-98 4 03 3 34 4 62 30-45 .57^34 104 S(n,Ai; PllVSICS COMMITTEE. Table 5. RAINFALL.— TOTAL VALUES. (Janiakv to Di-cKMiiKu.) Year. West Australia, West Australia, Perth District. We.st Australia, South Australia, Eyre's Peuiusula. South Australia, Vii-lciria. South-West .\iistralia. Shark's Bay. South Coast. Past )ral. Summary Inches. Incheii. Inches. Inc!ies. Inches. Indies. Inches. 18711 • - - - ' - - - - 27-74 - - 1871 . - - - - - - - 23-55 - 1872 - - - - - - " - 24-42 33-66 1873- - - - - 20-87 27-81 - 1874 ... - - ~ _ 20-92 31-28 _ 187.5 _ - . 26-60 33-71 _ 1876 - - - - - - 15-74 32-06 1877 ~ 23-71 - - 21-92 28-30 - 1878 - . - - 34-37 - 13-96 19-23 34-37 21-01 187!) - 25-61 - 14-66 20-54 29-91 22-34 18811 - - - . _ 24-57 _ 15-36 21-14 32 - 70 20-76 1881 -- 22-50 - 11-85 18-13 29 ■ 36 16-17 1882 - • - - - 32-87 _ 17-00 16-ti9 30-79 22 ■ 54 1883 - 31-20 - 19-48 24-07 31-09 23-19 188+ - - - - - 28-78 - 15-00 21-70 31-88 19-19 I88r> - 30-23 14-73 17-25 17-(i4 31-13 20-99 188() - - . . 9 45 30-13 15-50 12-68 17-36 34 - 90 19-75 1887 11 11 33-83 16-76 18-12 23-39 34-83 23-55 1888 .... 8 69 26-96 14-09 12-81 15-13 2«-20 1817 188!* 18 53 35-99 15-09 21-95 30-22 36 - 52 26-04 1890 - . - - 18 02 42-19 16-32 22-41 26-65 33-93 29-17 1891 r, 13 IS 25-33 13-81 13-14 16-13 28-16 18-00 1892 - - - - 30 28-40 19-09 17-72 21-14 34-67 22 - 36 1893 IG 37 36-42 16-92 20-30 24-67 33-98 26-26 1894 - - - . 6 25 22-85 11-92 14-36 21-40 36-31 16-15 189r. 12 86 33-31 9-67 16- 13 19-71 33-44 21-61 1890 - - - - (> 42 28 - 48 13-51 15-80 16-92 29-37 20- 19 1897 11 52 27-03 9-95 11-05 15-91 28-61 16-45 1898 - . - - - 13 33 33-73 11-74 13-16 21 -.59 29-32 20-01 1899 11 37 31-78 12-25 12-94 17-20 34-98 21-13 190(1 - - - - 14 72 37-66 19-47 16-11 20-63 30-77 25-03 1901 y 64 28-07 15-23 17-58 18-26 2 7 -.58 22-06 1902 .... 10 12 26-52 18-55 13-25 16-85 _ 20-79 1903 14 72 37-47 15-44 16-32 23-81 _ 24-51 1904 - . - . 12 70 36-02 20-00 14-81 18-46 - - lyori 13 11 35-91 12-83 10-81 21-92 A J)lSCrSSION OF AUSTRALIAN METEOROLOGY. I(»5 Tadle G. RAINFALL.— TOTAL VALUES. (August to Julv.i West, Australia, North. South Australia, North. Queensland, Nortli-oast. South Australia, Central. South Australia, Lake Eyre Distri<'t. New South Wales. Inland. New South Wales, Coast. Inches. Inches. Inches. Inches. Inches. Inches. Inches. 1S7()-IS71 . _ - _ _ _ ^ _ 1S71 1S72 _ - _ ^ _ 21-79 37 65 1K71'-1.S7S - - - - - - 33-77 76 80 1H78-1S71 - - 43- (-.3 - _ 27-71 71 74 IS74-187.-. - 49-04 - _ _ 28-60 59 68 1875-lS7t> - - 45-12 _ _ _ 30 -GO 61- 98 18711 1S77 - 49-58 _ lfi-9/ _ .. 45 12 1877 1S78 - - 38-86 - 8-78 - 23-82 57 42 1878-187'.* - 52-97 - 15-57 - 40-63 55 16 1879-18SO . - 41-03 - 10-58 - 31-52 57 42 1880-1881 - - 31-2(1 - 3-50 _ 21-49 45 58 1881 1882 - 40-20 - 8-19 _ 23-61 48 82 1882-188;) - - 38 - 40 _ 4-24 3-26 28-47 58 03 1883 1884 - 3.) -.53 _ 5-ti2 3-50 22-35 48 26 188t-188.-. - - - - 42-22 - 11-33 7 -.53 23-48 44 00 188.) lS8i; - 39 -.30 - 5-51 3-11 23-31 45 86 188r>-1887 - - 4(>-79 - 14-73 7-77 41-15 71 01 1887-1888 29 ^.H 41 -5(; - ti-57 4 -OS 22-77 53 04 1888-188a - 2(>-83 39 -.57 - 13-42 8-68 28-48 56 32 188'.t- 18911 31-52 47-88 - 8-98 7-30 43-12 96 93 1890-181)1 - 190t 42-01 - 10-34 4-98 35-98 60 12 1891-1892 17(i2 24 - 84 - 5-70 2-14 36-98 67 97 1892-18:t:! - 25- 15 39-13 - 7-92 4-34 52-98 68 07 1893 1894 19-92 42-82 - 12-19 4-51 33-75 47 00 1894-1895 - 27-24 .50-24 _ 15-05 9-11 24-48 57 04 189.'.-189fi 34 - 1>8 51-72 - 7-10 362 26-24 43 38 189ri 1897 - 15-92 39- u; 39-57 5-37 4- 19 25-32 43 72 1897 1898 29-45 00-48 (i5-88 8-50 4 -60 30-58 55 65 1898-1899 - 1 49-24 54-20 52-76 C-17 4-01 22-18 50 33 1899-1900 18-05 28 - 2(i 31-73 6 -.56 2-20 30-17 88 41 1900 1901 - ! lti-12 44-92 49-75 6-Hl 2-61 17-46 45 26 1901-1902 23 - 38 35-22 40-27 2-72 - 18-78 31-86 1902-19OH - 38-38 33 ■ 22 09 -38 13-00 - - - 1 90S- 1904 3(i-51 (;o-oi 02-04 11-61 - - - 1901-190.-, - l(!-2 li 4 17 s 1 Wilcaiinia 188(l-I!t02 fi 9 9 2 13 7 11 5 8 (i 7 9 10 i; 13 2 11 11 111 7 9 7 2 111 Moama 1879-1902 5 3 1 10 2 9 4 4 9 6 l.j 8 li; 111 19 8 18 i; 1.-, 2 1 9 1! 10 2 Balranal.l 188(1-1902 4 2 3 2 3 2 4 3 4 10 8 9 9 8 11 ti IL' i 111 7 7 i; 1! 9 Kustoii 1880-1902 10 1 C 4 4 6 .5 10 4 8 7 13 in; 1 17 9 19 11 17 9 11 3 11 11 Weiitwdi-th - 1880-1902 7 8 ■) 9 5 n 1 ;> l> 7 4 10 i-j (i 13 II 14 9 14 11 7 1 '.I 4 A DISCUSSION OF ATSTKALIAN METEOROLOGY. 107 T.viii.K (S. RIVER CiAUOES.— MEAN READINGS.- War. January to Deciimber. Vuai-. April to March. Bourkc. Wilcanni.i. Pooucanc. Moama. Balranalil. East on. Wentvvoith. Ft. in. Ft. ill. Ft. iu. Ft. in. Ft. in. Ft. in. Ft. in. 1879 - - - - 1879 18811 9 8 8 5 9 3 13 10 18811 9 7 13 4 1880-1881 - 13 6 8 1 10 10 12 1 1881 - 9 2 1 i 1881-1882 7 10 3 8 8 10 li 8 1882 - 3 5 4 5 r> 18S2-I883 - 8 11 4 9 7 3 7 5 18S3 - 4 4 4 8 .*> 8 1883- 1884 10 11 4 9 8 7 7 7 188-1 9 1 6 1 10 1884-188.5 - 5 8 3 6 5 9 4 10 188.") - 8 2 2 2 7 1885-1886 7 10 2 6 5 9 4 11 188(i Ifi 3 Iti 1 11 1 1886-1887 - 7 10 5 8 6 3 il 5 1887 - 14 11 19 9 21 3 1887-1888 17 5 12 2 15 Iti 1 1888 3 2 r, 1 6 9 1888-1889 - 8 10 4 5 6 8 7 3 1889 - 8 7 111 10 11 "t 1889 1890 16 3 9 10 15 4 14 189(1 - 23 10 2(> 9 23 II 1890-1891 - 16 6 12 1 19 8t 18 2 1S91 - 18 3 23 8 23 9 1891- 1S92 10 1 10 5 15 8 14 1892 - 11 1(1 15 13 8 1892-18il3 . 11 5 8 9 14 11 11 .5 1893 - 19 7 23 11 2.> 2 1893-18111 14 7 11 17 10 14 9 1894 - 14 3 18 .-, 19 10 I.s94-18!l5 . 16 8 13 11 21 1 16 7 189n - 3 II ti 2 f> 11 1895-1896 9 3 5 2 12 7 8 .5 1896 - 4 7 8 1) 8 4 1896-1897 - 7 3 i 11 11 4 7 2 1897 - 9 1 7 1) i (1 1897-1898 6 3 3 3 9 8 6 3 1898 7 4 7 (1 ."i 1898-1899 - 8 3 3 11 1 6 2 1899 - 4 (i 4 3 9 1899-190(1 7 6 4 2 10 3 .5 li 1900 - 7 2 7 3 i; 11 19(10-1901 . 10 10 8 4 14 8 9 19(il - 2 8 3 10 3 ii 1901-1902 8 3 4 11 10 11 5 ( 19ii2 6 Oi '» 1 * hi 0(ins<'i]iuiire of alltr.Mlinn in lifiglit nf two trau.urs. tlio corrci^pnniiiiif.' ('i|]vi.'K in I'lute t Tlic gaujre at Pooncariu was lowered 2 feet at the beginning of 1889. J The gauge at Euston wan lowered 4 feet 8 inolies ficini tlie beK'nninf; of 1890. an- left broken. 2 108 soT.Ait PiiYSTCS co:\r:\iriTEE. Tap.le 9. RAINFALL. Yearly Totals and Four-yearly Means. Year. 1840 1841 1842 1843 1844 184.5 1846 1847 1848 1849 1850 18.51 18.52 1853 1854 1855 1856 18.57 1858 1859 1860 1861 1862 IKOH 1861 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 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 Syilney, New Soulli Wales. One Year. Four Years. 58 -.52 76-31 48-82 62 - 78 70-67 62-03 43-83 42-80 59-17 21-48 44-88 35-14 43 - 78 46-11 29-28 52 - 85 43-31 50-95 59-60] 42-06 82-81 58-36 23-98 47-08 69-12 36-29 30-81 59-68 43 05 48-19 64-22 52-27 37-12 73-40 63-60 46-25 45-69 ,59-66 49-77 63-19 29-51 41-09 42-28 46-92 44-04 39-91 39-43 160-16 23-01 57-16 81-42 .55-30 69 •2S 49 ■ 90 38-22 31-86 42-40 42 - 52 43- 17 .55-90 66-54 40-15 43-07 .38-62 45-93 35-03 31-89 Bi-isbano, Qiuieiisland. On"! I Four Year. Years. •61-61 64-61 } 61-07 59-83! 54-83! 51-95 41-821 42 0^ 40-16 ,36-32 42-48 38-58 43-00 42-90 44-10 51-68 48-98 58-85 60-71 51-80 .53-06! 49-61 ! 41-12 47-32 50-481 48-96'! 46-93 .53-78 51-93i 50-45,1 56-751 ,56-60'[ 55-09: ,57-231 53-80 50-34, .54-58 50 - 53 I 45-89 44-02 39-95 43-58 43-29 42-5/ 45-88 40-63 44-94 ,55-44 54-22 65 ■ 78 63-97 53-17 47-31 40-59 38-75 39-99 46-00 ,52-03 51-44 51-41 47-10 41-94 40-66 37-87 29-32 49-31 28-82 51-23 63-21 89-19 31-43 42 ,59 Ailolaidc. One Year. 43-00 35 00 .54-63 69-44 28-27 08 ■ 82 47-00 24-11 51-18 61-04 35-98 54 • 39 79-06 45 ■ 45 49-22 62-02 38-71 67-03 53-42 30-28 .56-33 67-30 49- 12 29-39 42 - 02 32-22 43-49 20 ■ 85 .53-66 81-54 3! -08 49-30 73-02 41-08 04-97 88-20 44-02 59 • 1 1 44-98 42-53 60-08 37-35 34-41 38-49 16-17 49-28 33-24 36-76 42-85 • 50 -.52 46-83 55-29 53-38; 42-05 47-78 45-841 43 -OS; 50-65 57-62' .53-72} 57-03! 58-941 48-8.5 54-24 55-29 47-36 50-76! 5(1-531 4711 38-44i 36-93 36-29: 39 • 05; 51-38 48-78 54-41 59-25 49-28 57-26 66-98 59-70 64-09 59-09 47-60 51-07 46-23 43-59 42-58 31-61 34-59 34-30 33-86 40-53 24- 23 17- 96 20- 32 17- 19 16- 88 18- 83 20- 88 27- 61 19- 75 25- 44 19- 56 30- 86 27- 44 27- 09 15- 34 23- 15 24 93 ''O 15 20 25 14 46 18 5(J 24 04 21 85 23 67 19 75 15 50 20 11 19 05 19 99 14 74 23 84 23 25 22 06 21 00 17 23 ■29 21 !13 43 24 95 22 08 20 69 22 48 18 02 15 70 20 76 118 74 !l5 89 !l4 42 25 70 14 55 30 87 25 •78 14 01 21 -53 21 ■49 20 ■78 21 28 15 -42 20 -75 18 -84 21 -68 118 -01 10 -02 '25 -47 |20 -51 22 -28 20 -51 Four Years. ■19-92 18-09 18-31 19-95 22 - 55 23-27! 24-92 23 09 23-88 25-82 26-24 25- 18 23-26 22-63 20-89 22-12 19-95' 18-34 19-31 19-71 22-01 22-33 20-19, 19-76 18-60 18-66' 18-47 19-40 20-45 21-12:: 22 ■691; 21-03! 22-52! 20-22 21-21 22-42; 20-29, 22 - 551 20-82 19-22 20-74 19-81 19-27 18-95 18-09 17-64 21-38 24-23 21-30 23-05 20-70 19-45 21-27 19-68 18- 16 18-16 17-55 19- 1: 19-82 18-04 20-29 19- 9.5 21-02 23-01 Soiitli Australia, Agricultural Di-tricts. One Y'ea'-. 27-83 21-76 27-98 18-83 16 08 21-80 23-31 20-70 16-85 27-74 23-55 24-42 211-87 20-92 20-00 15-74 21-92 19-23 20-54 21-14 18-13 16-69 24-07 21-70 17-64 17-36 23-39 15-13 30-22 26-05 10-13 2114 24 ■ 67 23-40 19-71 10-92 15-92 21-59 17-20 20-03 18-26 10-85 23-81 18-40 21-92 Four Y^ears. '24-10 21 31 21 -341 2O-I7I 20-04. 20-68' 22-15 22-21 23-14 24-14 22-44 23-20 21- Ol 21-29 20-87 19-30 20-71 19-76 19- 12 20-01 20-15 20-02 20-19 20-02 18-38 21-52 23-85 22-03 23-54 22-15 21-34 22-23 21-17 18-99 18 -.54 17-91 18-83 19-42 18-23 19-89 19-34 20-20 Yauko, New Soutli Wales. One Year. 22-80 24-28 6-11 21-49 17-70 12-53 17-78 36-00 19-^5 20-24 10-42 9-44 23-87 12-20 13-08 21-37 15-45 14-79 16-44 16 01 12-71 10-78 15-45 14-11 30-86 9-51 24 -.59 17-01 22 - 53 16-26 14-84 22-91 15-14 14-18 11-22 10-13 11 ,50 17-66 11-50 8-88 Four Y^eais. ili'lbouriu', Vijtoria, One Year. Four Yeai-s. .18-67 17-41 14-47 17-39 21-02 21-39 23-32 22-98 16-34 17-49, 15-48; 14-80 17-78: 15-67: 16-32 17-01 18-17 14-99 13-99 13-74 13 --26 17-80 17-48 19-77 20-64 18-50 211-25 17-81 19-13 17-29 10-77 15-80 1207 1 1 - 70 12-03 12-70 12-43 22-57 HO- 18 31-10 21-54 28-20 23-93 30-53 30-18 33-15 44-25 20-98 28-21 29 - 75 28-90 211-01 21-82 25-38 29-16 22-08 36-42 27-40 15-94 22-41 25-79 27 { 5S ■17 IS 24 33 30 32-52 25-60 28-10 32-87 2404 24-10 25-36 19-28 28-48 24-08 22-40 23-71 25-85 26-94 24-00 32-39 19-42 27-14 24-20 20-73 24 - 09 26-80 22-00 17-04 25 IS 25-85 15-01 28-87 28-09 27-45 23-08 28-43 29-72 25-64 22 29 >20-3O 27-78 20-22 20-00: 28-22 29-45 34 -.53 33-64 28- 22 26- 62 25 53 25 59 24 01 28 20 28 70 25 40 25 54 22 88 20 00 22 76 25 60 20 711 30 26 30 51 29 10 29 77 27 65 27 28 26 59 23 19 24 30 24 30 23 56 24 67 24 01 24 72 25 12 27 29 25 Oil 25 74 25 79 24 37 24 m' 24 60 25 20 22 78 22 90 22 ■06 20 91 23 •87 24 •00 25 00 20 •87 26 -70 27 •17 20 -72 26 •52 Perlli. Western Australia. One Four Year. 1 Years. 28 ■ 73 20-49 39-71 41-34 31-79 24-78 .35 05 39 -.58 31-90 33-44 2-i-90 37 - 52 27 - 83 39-90 46-73 30-33 31-23 40-12 23-72 33-01 31-50 27-25 32 04 31-90 30-61 36-75 27^00 35 - 09 31-35 34-01 ■32-57 33-33 34-41 33-40 32-97 33-02 35-18 33-49 32^95 31-92 33 -.55 38-01 36-21 37-00 37-10 31-35 32-02 32-09 28-87 30-95 30 09 31-96 34-30 33-09 34-03 33 ■4( 32-93 Hirsliani," Victoria. One Year. Four Y'ears. 23-22 21-91 20-SO 11; 03 M-cr, 10-07 2014 11-71 23-01 14 04 1001 14-89 18-42 11 09 10-97 27-42 25-6I) 19-45 20-68 19-50 18-48 15-27 12-39 15-97 13-94 13-78 12-90 14-55 21 -.S3 15-41 15-02 19-70 19-14 11-17 26-64 21-27 15-91 18-98 22-35 21-43 14-69 12-12 13-39 18-73 15 -.SO 18-07 14-69 10-18 22-00 1 8 - 30 17-1)4 10-87 15-80 18-03 17-52 15-00 15-80 14-50 13-61 13-84 10-97 18-77 20-86 2i-29 21-31 19-53 18-48 10-41 15-53 14-39 14 02 14-16 13-81 15-78 lfi-19 10-70 18-00 17-33 10-27 19-18 19 -.55 18-75 20-80 19-63 19-67 19-36 15-15 15-41 14-73 15-01 16-50 16-82 14-08 1855-1808 inclusive at Waljner, 3 niilos away. 18!)9-1873 inchnive at Longernong 1874 and onwards at Horsham. II miles away. A DISCUSSION OF AUSTRALIAN ^[ETEOROLOGY. 109 Table ]0. riiESSURE. . I Year. 1841 1842 1843 1844 1845 184f) 1847 1848 1849 ISnO 18.-,1 1852 1853 1854 1855 185(! 1857 1 S58 185!) 18(10 I8()l 18(i2 18(13 18(14 18(15 18(16 18(17 1868 18(l!t 187(1 1871 1872 1 873 1874 1875 187(1 1877 1878 1S7!I 1880 18SI 18S2 1 8S3 1884 1885 188(1 1887 18S8 188!t 18!((l 18111 18'.I2 18'J3 1S94 1 8i)5 1 wk; 1 8!»7 1 S:i8 1 8!l!l HlOO P.lOl l!l(12 I'.KC! I!l(i4 111(15 llHKl 111(17 Coriliibii, 720 mm. Year. April- Sept. 74 Four Years. 4-65 (101 6-51 5-56 5-08 5-96 5-73 5-24 5-30 5-83 5-79 5-12 5 47 (1-56 5- 2(1 4-79 6-08 C-44 5-40 7-20 6- 09 I (1-39 i 5-69 5-48 68 51 I 04 21 79 47 78 70 97 90 42 38 42 37 27 14 09 98 8fi 88 95 12 19 19 33 32 31 54 70 73 75 64 54 34 13 02 80 69 95 96 0(1 34 45 Cape Tiiwn, 29 ins. + Year. April- Sept. Four Years. 036 046 033 057 030 036 ■004 ■027 ■009 ■023 038 •047 049 0.52 043 039 ■039 024 •023 •016 •993 0:U •030 035 036 •038 043 •034 013 •024 013 013 (144 032 052 0.32 •029 ■ 029 •048 •050 •042 ■02(1 ■049 ■028 •022 •043 ■029 ■(,35 029 •030 018 •019 024 •017 •037 •045 •02(1 ■029 ■022 ■039 •023 •04(1 ■037 115 089 111 112 127 093 089 067 (183 061 (188 (t95 123 123 121 115 105 (196 ■086 080 080 ■035 110 (196 lOl 112 102 103 108 ■098 ■087 ■078 (172 ■134 ■108 ■119 ■088 ■096 (191 125 ■139 •119 ■080 ■123 ■079 ■076 • 109 ■073 106 ■093 ■('95 ■087 084 102 092 ■098 ■124 ■im ■095 •oso •106 •060 ■102 ■102 Diirliai], 30 ins. + M-(143 I 1(143 1-041 1(133 1-(I26 1019 1016 1-024 1 1-029 1-039 1 -047 1-048 1-046 1-043 1-036 1-031 1-025 1-014 1-016 1-017 1-022 1-033 1-035 1-038 1-038 1-032 1-028 1-021 101(1 1-023 1(126 1-035 1-O40 1-030 1 -035 1(134 1-039 1-042 1-041 1-042 1-036 1-031 1-035 1(131 1 ■ 0H2 1-034 1-031 l-(»2f 1024 1-023 1(119 1-024 1-031 1-031 1-034 1-030 1-029 1-028 1-032 1-030 Year. April- Se])t. Four Years. -073 -102 -110 •110 -109 -089 -081 -(18.-i -084 -106 -lOI -no -101 -102 -133 -097 -110 -105 ■113 -0.59 -091 -108 -094 -107 -095 •094 -121 -0112 -112 ■082 -080 -094 -093 H5 179 175 191 182 148 103 1.59 159 173 J57 198 101 183 219 148 195 181 184 137 100 190 170 178 181 199 224 170 210 130 105 170 148 • -0119 -108 -104 -097 -090 -084 -088 •093 -100 -104 - 103 -111 -108 -112 -113 -108 -098 •092 -093 •088 ■100 -101 -097 -104 -100 •107 •104 •093 -093 -089 Hatuvia. 75(1 mm. + Year. April- Sept. Four Years. Bombay, 29 in?. + 8-84 8-85 9-20 8-99 8-14 8-49 8-20 8-40 8-57 8 8 9 8 43 65 77 68 17 79 87 -57 90 ■01 21 8-45 8-50 9-15 8-79 8-64 9-10 8-30 8-52 8-71 8-72 11-08 8-80 8-30 8-95 8-97 8-92 9-12 8-71 8-74 9-27 8 - 82 8-57 -35 ■94 -28 72 -34 -49 -48 8-67 8-68 9-75 8-68 8-30 8-81 8-97 8-50 8-96 9-04 9-10 8-36 8-72 9-01 8-65 8-«9 9-31 8-01 8-50 8 '70 8-82 9-14 8-94 8 -.55 9-05 8-90 8-8(1 9- 13 8-07 8-85 9-20 8 97 8 80 8 71 8 45 8 32 8 43 8 41 8 53 8 85 8 88 8 82 8 85 8- 63 8 60 8 78 8 84 8 92 8 89 8 79 8 83 8 72 8 77 8 93 8 74 8 67 8 09 8 58 8 70 8 83 8 72 8 78 8 75 8 78 8 99 8 93 8 87 8 96 Year. I April- Sept. Four Years. 792 800 793 803 791 8(K) 809 799 819 801 803 807 807 799 792 778 787 827 806 817 819 831 809 792 800 790 80(1 805 812 819 848 801 805 822 825 811 810 830 828 812 821 83(1 815 810 830 787 812 805 817 818 808 795 830 824 820 818 807 818 833 815 809 094 710 690 718 692 707 705 699 730 700 705 711 717 710 695 692 685 739 708 726 710 747 714 708 719 701 713 703 717 722 768 708 713 I 732 735 ; 716 719 729 ; 734 719 ] 738 ! 740 ' 714 i 720 742 093 720 708 1 730 i 720 ' 711 710 748 734 730 720 706 725 742 710 720 797 797 797 801 800 807 807 805 8(17 804 804 801 793 789 796 799 809 817 818 819 813 808 798 797 800 803 811 821 820 81S 819 813 810 818 820 821 821 823 824 821 821 823 8111 810 808 805 813 812 809 813 814 817 823 817 816 819 818 819 A I) D E N D U :\[. A DlSCrSSION OF AUSTRALIAN METEOROLOCJY. 113 AD])ENDUM. FuRTHKR Discission ox the Latitudes of AusTitALUX Anticyclones. While the nieaii animal (January to December) vahies for the latitudes of the anticyclones, as given liy Russell, have a variation of 2'S degrees between the two extreme values, it was considered more advisable to form the monthly values into two groups according to the seasons. This procedure was adopted as it seemed possible that changes might be occurring in the seasons which might neutralise each other, and therefore not be apparent in the mean for the year. This inquiry has led one to conclude that a law is possibly in operation which, if corroborated, nuiy prove of considerable importance. If, for instance, the year be divided into the two seasons, April to September (winter), Avhen the anticyclones in tlieir annual movement have a low mean latitude, and October to March (summer), when the anticyclones are in higher latitudes, it is found that the mean latitudes of the centres of the tracks are for each season respectively as follows : — Table 1. Mean Latitudes of Anticyclone Centres. 1888-1889 1889-1890 1890-1891 1891-1892 1892-1893 1893-1894 1894-189.) 1895-1896 1896-1897 1897-1898 1898-1899 1899-1900 1900-1901 Mean April-September (Winter). October-March (Summer). 30 40 30 40 30 30 32 40* maximiiiii 31 32 30 32 31 30 30 10 10* inaxiniiim 30 37 10 37 20 34 10* 35 50 39 20 40 10 niiiiimum. 36 37 36 10* minimum. 20 32 30 33 10 36 32 32 Uittereiice 10* maximum 40 4 4= 41' 36 20 34 50* minimum. 36 20 36 45 On examining the second and third colunms of figures, it will be noticed that the maxima in colunui 2 seem to be closely associated with the minima in colunui 3. Thus the maxima for the winter months in 1891, 1895, and 1899 follow closely after the minima of the preceding summer months in 1890-1891, 1894-1895, and 1898-1899. The values referred to have been marked with an asterisk. Similarly, the minima in colunm 2 for the winter months of 1890, 1894, and 189C are closely associated with the maxima in the third column for the preceding summer months, 1889-1890, 1893-1894, and 1895-1896. The only outstanding values which do not conform to the above-mentioned rule are those for April-September in 1892 and 1893. Summarising the results brought together in the above table, it is seen that at the epochs 1890-1891, 1894-1895, and 1898-1899 there is a great tendency of 142 114 SOLAR PHYSICS COMillTTEE. tlie range of anticyclones in latitude to be at a niinimuni, while, on the other hand, at the epochs 1889-1890, 1893-1894, and 1895-189(; the range attains a maxininm. There is thus a variation in range of latitude of about four years duration. This variation is shown graphically in the accompanying illustration (Fig. llj. AUSTRAL lA 30-i*-i UAT. ^0 -J LATITUPCS OF ANTICYCLONIC TRACKS FOR.— — >32'4'>APR.-SEPT .' J (WINTER.) ->36*^5loCT-MAR. J (SUMMER.) 1 1 1 r 18880 3 18900 I 1 1 1 9 !9Q00 YEARS. KlG. 11. The npper large arrow represents the direction and mean tracdv of the anti- cyclones for the winter months for the period 1888-1900, while the lower arrow represents the direction and ]iiean track for the summer jxionths for the same period. These values are 32° 4' and 36° 4.5' S. latitxide respectively, as seen from Table 1. The coast line of Australia is also shown to indicate tlie relation ni these mean tracks to the continent. With a horizontal scale of years given at the bottom the individual values for the latitudes for each of the seasons for each year, as given in colunuis 2 and 3 of Table 1, are plotted. For each season a belt can be drawn to include most of the variations in the different years from the mean value, and each of those belts has been shaded to render them more apparent. It will be noticed that wlien the points are above the mean line in the winter months they are for the most part below the mean line for the summer months, and vice versa. This shows that when the tracks of the anticyclones in the Avinter seasons are in lower latitudes than usual, those for the summer seasons are in higher latitudes, and vice versa. Thus there is an inclination for the annual swing of the anticyclonic track to be small in some years and large in others. This diagram shows further that there is much less evidence of the anticyclonic belt as a whole being in some years moi-e north or more south. If such a move- ment were in operation, then, when the points were above the winter mean line, the corresponding points should be above the summer mean line. That this is not the case can be seen at a glance, especially for the last four years, where the values for the winter seasons are below and those for the summer seasons above the mean lines. A DIRrrSSK^N OF AI'STU-ALIAN MKTKOIJOI.Ot IV. lir, The (letectiou of this cluinge of nuise of the paths of the anticyclones, showing that ill some years the range between summer and winter was large and in other years small, led one to inquire in what Avay the variation was associated with pressure differences between the two seasons. A comparison was therefore made between the range anomalies and those of pressuj-e in the following way : — In Table 1 the means of ench of the second and third columns were formed (namely, 32° 4' and 36° 45') and their difference (4° 41') determined. This difference is considered as the mean change of latitude due to the two seasons. To eliminate this the differences between columns 3 and 2 have been deter- mined for each epoch, and the mean value (4° 41') sui^tracted from each, thus resulting in a column of anomalies with positive and negative signs, the positive signs denoting large ranges and the negative small ranges of latitude change. Further, the winter group of months (April to September) for each year was con- sidered in relation to the preceding and following summer months (October to March) separately, and in this way two columns of variations from the mean formed, one for the twelve months April to March and the other for October to September. Kxactly the same procedure was adopted with the monthly pressure values at Adelaide. The winter and summer mean pressures were determined for each year and the means for the period 1888 to 1900 formed. The differences between the winter and sinnmer values were then separately found, and the differences from the mean for the wdiole period determined. In this way two sets of range anomalies were obtained, one representing the changes for the conseciitive group of 12 months April to March and the other the changes for the months Octol)er to September. The " plus " signs indicate large and the " minus " signs small ranges of pressures. These values are placed in columns 3 and 5 in Table 2, while the figures representing the anomalies of the range in latitude of the anticyclones are given in colunius 2 and 4 of the same table. The values thus obtained are as follows: — Table 2. Eange in Adelaide, Bange in Adclai. PRESSUF?E ^^^fi-oi.c. ADELMDL I S AOST ) APL-SEP ADELMDL RAINFALL (JAN.-DLC.1. W. AUSTRALIA. SHARKS BAY W. AUSTIN ALIA. PERTH DISTRICT W. AUSTRALIA. SOUTH. S.AUSTRALIA. 5 AUSTRALIA. PA5T0R/M-. VICTORIA AUSTRALIA 50UTH WE5T. [SUMMARX] 18700 5 18800 6 18$ PL.\TE 2. ^0-0 5 19000 6 -tn.na.-r— ^ ^ t- A A ^ ^0% ;^^ ^ / ^ S A A .06 X t 5^" N ^^ t it^ ^-^ ^ K-t^ ^ \ t % ^J\^ V ^ 3012 h/ ^ 2 ^ T 3007-r^ V A 3007 ^ ^^ Q -II "^J \ h h -, 4 ^ U^ 5 Jl ' ^ t^ t\ J J I 5^^ L^ t ^ •15- t- t :\JL V <^^ t -X^t ^^^J_t U^ VJL ^r^ _5 .'^ L,'^ 31 ^i V V 30.19- Xi: 2 V V I© ^ d f,,'^ ^J X ^ ^ IS ^ V t t: 4 lO ^ ^ / \ J- A /«>_- II ^-i itlJLS ^ N/ u^^ " /'^^ f^ r^, t ^=^ 7 ^5 V S / 3iv ^ ^ ■^R ; 4 ^ -^ OO ^ -\ -^ S ^^^ ■^4. -4- /- is 3 t ^ J. S L IE ^ Ji ^2 ^ 30 t fS ^J V^ lj r. r^ t_ iu t ^ V ti ^LAl In 2.R ^ I -T ^ V u ^ ^ 7 ^r^j il "^^ E 4: ^ /i 17 ^ ^7 Zv^ C^^^ 1- ^ " ^^"^^x ^^ 3 ^^^ ^ t IT -v^ ^ y^ 12 I 13-^ I 1 A^^zt^ lA A . r ^ A ^ Ml _^ ®" 7\a S^ ^t\^^ >^ A^ 14. =^'^ L ^^^2-. V 5^1 ^^ ^^- I+- =^5 ^ ¥j ^^ s, ^^ ^ ,0 JV . f t ^ o/i \ ^ /\ , f A ^^ A -•- ^' Xjj Tv Alt ^"5 P ^ 90 ^^-L/^^^::^ ' ^ Q:^ t-Z V A A t^ ^ t ^ ^^ tJ 7 V ^v- '\^^4 ^ ,6 V ^ !f-^ IT . ^2 ^ ^ Tyl /"\ 1 V -pu^V h -I- ^^~ ^ ^^K I A -,^A-t \ t \ t^ ^n \-t ^ ^sV^-^^ n ia^ \ ^ A ^° b V ^ V $y ^-^^^ V 7 A ^ n y1 f -\^ A A ^ -r ^^" ^ ^^ A '^ i74 -y \^^ OA ^ \ L i^^%J \J t' i^^ ^^ ^ ZO N— r- V'-*' V/ V- t ^r.^ iR ^ 5 ^ 15-1 1— j- 1 — 1 — 5ee page 46» I E 142. Q I87|00 5 le^OO 5 ^ I8S PLATE 3. )00 5 19000 6 1 1 pf:essure-op -V- -K- CAUSJUL) ' A- — iX- r- 4a 3004- 4 2a C 4^r- ^ ADLLAIDL^^ A P jl ^ I t^C f^V S. AUSTRALIA OO ^^' L r-\ CZ\ J.A 1 \l \7l 5 7^^ INVE.RTLD30O8- \ -) A-y 4 T ^t ^OJ ^^^v'^ 1. , A„, \ 9Q-Q7 -.N -A -/ / \_X— li^/- I ^ I^ (Noy-/\PU 2957 5 f f ^f '^^ V- t^X. ADLLAIDL,.^. ^^\ y T^r7V T '^ ^ ^S JZ^ ^v S AUSTRALIA ^0^' >7 J ?^1J^^^ ^^ / t i \7\j L-^y^< INVLRTED 3. ._ ^ S^ 3:/ \l\y \^^ ^L^f-X ' RAIMFAI r f ^ ^ "^ jl ^ rAI INiALL- . ^ t\ ^^V ^ (AUG-JUL) <^"T s, jX t 1- X W AIICiTRAI lA 9fl ' \ L\_r t X W /VUO 1 r\AI-lr\ /ID X -,\2 ^zi ^J~ t NUr\ 1 11 1 A ^1^^ i\ rv ti 10 CO ... ,L_ ^ t I- '^ 41^ -. C7Z. V- ^^\i I 11 <^ Allf^TRAI lA Afl. /\ \ 1 j\ J ^ 2 Ai i .^ ij 0. /\UO 1 nnLIA "rT- ' W I ■ ■ ■- ~ -y-\ -j f ^ L A_,x fir IMUr\ln 1 \ »> .^ "^ ' • A ^ 44^t Jb >^ A^ 4Z ^' Afi 4/ ' iS -. OD ^ (I \ Ol/F F MSI AMPl'^A 4a V W U C. C IM 1_ / \l M LC M r J^^ J A 'cn 4 C 4 J ou 4 A ti i no t A-j^t V 'f/L . - .. 16 \ s. t ^i *~ X :! C AII'^TRAI lA _S_i J ^ m^ ^X 4t .V ^ 0. Auo 1 r\/\Lin y \. \, / \ y > -^^ z ^^ ^ ^ i S^ CLNTP^AU ^/\ / ' ' ^7 ^^ --vZ in _, -^r — • S.AUSTRALIA'" \ ^y^ y \ 1 LAKE. LYRE. 9 —7~7 ^ 'n^ — ' \— — ""-^ 1 L 0\J il NEW SOUTH 42 ^ 4 WALE-S \ V -4 ^ t ¥»r\l_l_o -^ +- —J ^"^ , INLAND 3^ _\ . 1 \ p f A ^ ^ 26- 2^ t X ^\ t -T X^^J^A ^^ t r ^y^^^r V "^ '^ /\ -r Afv I i^^ 00 X " ' 7A 4 t 1 ^ 4 t NLW SOUTH 7o_ V L_ , hA WALES ^ \ ^V I ^-w 44 Z 4 44 COAST 6Z H p,r^ j-|-|^ 4 4 5^ 4 A^^-^^ n't i t^ \\- ^t *^ I Jj ^ l^ t 7X -4 .46-1 ^ i^ ^ t rA 7 TD 1 ' ^^-' ^r V -" t X ± x^ See page 47. Q2 PLATE 4. tNC HpS RAINFALU QUEENSLAND. SOUTH RAINFALL N.S.WALES. INLAND. NE HEIGHT OF DARLING AT 80URKE AT WILCANNIA 6- 5- 4- 3- 2- FEET 10- 5- 10- INCi Ptb RMNFALL N S WALES 1NL/\ND. S.E. RAlNFALU VICTORIA. HEIGHT OfHURRA\ atMOAMA 2- 4- "li 15- 10- hei&htofMURRUMBIDGEE 10- at BALRANALD 5- 10- hei&htof MURRAV AT EUSTON heisht at junction of MURRAY andDARUNG AtWENIVvJORTH 5- 10- JFMAMJJASONDJrMAMJ JASONDJFMAMJJASOND n^ ^ /:> U ^, L - \ \ x; ^\ ^v \ - -^ 1 A 1 A^ 4 T 1 -4^ it / t t t t t V V t -,^ t 4 t ^ ^^ ^^i^ V^^^ ^^^ ^-v-^^ ^- t ' - 5^ ^^ ^^^^ ^^ ^ ^ ^^y ^ "^"^ ^y ^ ^^^^^ ^ ^•^^y \ ^J'J^ s^ C ^^yS^ \ \ \ - -/"^^ /'"v^ ^'^^ ^^x ^^ ^'^X t ^ J^ ^^iv ^ X V "^ ^C^^ Z^ 5 ^'^ ^^ -^ /^^ ^^ ^;^^ z>^ ^ 4^^^^^%:^^ ^ ^^ - ^ S4 ^ ^^ 4 \ ^ ^^ / s /^^ 7 V-/ \^ / V-/ "^>^ / "^^ ^>r "i _, ■■ ^ ^ ^ --. >-v ^_, i:^'*";^^ ^!!!^ ^ , ^~» ^'^ ^ — \_Z ^=' — K^T^ ^^ =^w.-^ ^^^ =^^ ^'^^-c- /" -/ ^V V ^V ^^ ^V ^ ^^ z^ ^^ z ^^ V ^ J- ^^ y^ s_ -^J '^J ^ ^ V^^ - v"^ "^ ^^^ <^ \ ^ ^ V ^ ,^ V -T ^ -4 -5 ^^ i ^ V ^ t ^ ^ 7 V t S L t A ^ t -^4 K ^ 4 1 > ., X- -*- -/ -^ - ^^y ^^ \ ^2^ x^ N ^^J-^"^ ^'^ % C^^\\ ^ ^'^\K C ^"^^ J. ^ iv ^ /L ^0^ ^ ^ ^^ -^^^^"^z ^^^^^y ^ ^^^ — ^ / — r "T — '^ ^ r^J^ ""s '^a A r<> - ^^7 ^^"^v ^--^ .-^^^ ^-^ ^-^ "^^ /^ \ /^ ^ ^'^ \ ^ ^ ^ ^^ ^y ^^ ^y ^^ y - T T See page 58. RAINFALL (AUG-JUL.) N.S. WALEIS 1 N LAN H LIGHT OF (JAN DEC) 1870 ■0 5 188 00 5 18900 5 I900C PJ t .A 5 T K 5. 1 INS 50-7- A-O \ -iX ^:v -lA tU ^0 /\ A / ^- .r-i ( k, / ^ /\ / : v: OU FttT - on / / \ y \ / ' V r- - ^^ ^z\2X: / ^ - — / / A ^r ^ ZU / s s^^^- L. '■ R DARLING' I0-- n y \ / y t AT \. / \ _ 3 ^^i^^\- BOURKL V "— N_ ./ Vs *. ^. r darling wilcannia r. darling ; pooncarilI (APR-MAR) R MU^RRAY] u 20-- 1 f\ , \ >cV^ c A / y V I \ / - ^5 90-1- \, ' >'N u\^ '~ 5s, ..^ !^ _ I0-- 1 r ^_ J A / ^t V S \ 1 \ / j^ \ \, iH J > ? ~~ 5i. ^< ^ V '■>_ _/ A cru xV-, =^ \ s 10 1 n k_ /\ / \. -^ \ MOAMA . / \ y J ' V T ^ ^"^ AJ RMURRUMBIDG AT BALRANALD R MURRAY AT LUSTON . R. MURRAY WENTWORTH (junction) RAINFALL iJAN-DEC) VICTORIA. 1 u EE|0-r 5-- . 20-- ,a-: 1 n V z' \ / f -v!u^ -Z^' >^ /- /\ /^. ^'^I^ "^-^ / '\ 1 "^y A \^ / 7 t i_ ^ 7^^ v- ^> s/ _^\ ^ ^^ j\ - zv / ^ -Z A y\ J i^ VS- y s. /\ / ^ ^^-z: L 1 u^ 9(1 N ^ S. / V zu 1 <; V_ f r \ y\ XV-A 15 ^ 1 N /\ ' s ^/ X \ / ' \ / J rN* \ > ^ -/N- \ y \, 9n -> -^\- /\ / N. ^ A \ r A zu 1 c \ V r '^- ,/ \ 1 r- -> y'^-' \ \ / ^/ \ \, : s ^''^ 10 7 V y^ F See page 59. PLATE 6. 7" — • "'- r / - r- -4- ^^ "*^- O- ^- ^^ K % <" t i ^^ o "^^ -^^^ -^ -^^^ § - L_ I \ JT 2 ^ ^ ^ H^ / -=^ ^ ^ % ^ ^ J ' " < J. <^ -^ ^>, ^ ^^ V ^ ^^ ^ ^^ J^"^ ^^ ^^ / 2 ol ^^ ^^ ^ -^ «i=' o ^^ ^-^ ^^'' ^z ^^ : a> <- ^^ ^-' ^'^ ^-^ s_ ^^^ ^^ ^ ^^ ^^^ ,^^%^ -^^ ^^ ?* - ^^ ^^ -,^ ^^ ^7" - -^ ^=- ""^-^ ^ ^K >o ^ ^^ S^ "^ -- "^ -^ ^V ^"^^ "":^^ s; ^=-^ '- o- \ ^ ^ ^^ S ^s= S4i ^ ■-^ 4 ^ '-^ - O r^ t J ^2 \ 2._. ^'' ^^ ^^' ^^ ~ ^ X r^ -< !- % ^ \ ^^ ^ . ^^ 1 \ ^^ ^ _ % ^ '' - ■o ^^^ ^ 2 ," I I ^^ ^"^ ^^ ^ir Z ^^ o ^^ ^^ s ^^^"^ -=-"" ^^ « z! 7 .^ -' ^^z^ ^^ r u '1 ^^ s^ ^ ^^ ^--e- W, "^^ X ^^ ~" "^ ^ - ^^ X^^ ■^^ ""^v V ^s ^ k- V ^ >^ o ^^Z^"^ / ^ o - >- ^ =- ^ _/ _ ____ _— — — — or 09 LiJ CC z o ro in z ^.^O^o»^lOC^^^,^Oloto*■lOO — cmW ooQo>oocpcoooooooo oo) (hffi 0)<5' 06666 ro fJ N cj CM to r) III oe: y tt. LJ a UJ < o > UJ z Q >■ I H UJ Q. < a: H CO < ,SVr \\wn^ ()."). K If-". R PLATE 7. lO 2 < _i O uJ OC — I <: ^ -' := OC ■^ < O O z 2 CQ 1- jg See page 67. E 142 s 18600 1870 18800 18900 PLATE 19000 5 19 PRESSURE JAN -DEC 5 CORDOBA *5 40 CAPETOWN so 'N& * ADELAIDE „„ BATAVIA rso Mr< * BOMBAV 29tMS-fr B2- CORDOBA '"' INVERTED 4 5' APR- SEP es. CORDOBA *°' 720 MM+ CAPETOWN 30 INft-t- ADELAIDE 30 INS + BATAVIA 750 MM4 BOMBAV _^ 19 INS + T*" INVERTED 8. 00 See page 72. S 2 I ^LATE 9. \ ^^ / '. f ' ' ^ ^ \ ^ / \, '*'■ ■^ / \ cp v ■■ >

i /5 ' " ~ s :)■ 1 '1 I. . J '■ L y > •v 9! / ' i L c j X < ^ / c L k _ J ^■f ■> V -- 1 / s J^ - / /' i -• L p ^ F > \ » N p O " r nr > / EL y X > , £ n o OO - « • i << - - - 1 ' s < ._ oO a*^ >^ 'J r. ' ( 1 ^s ^g ,=^ C \ /' Tl ft C b r. s / \ / J L r t \ y \ r . . r -■ ■ ^■ ■- - •- > •■ e I , • - - ... .1 - ■ > r - r . L ^ \ - K* •i^ / o \ ^1 / "' o o ^ s ^ ■ - / -i ^ \ s / /• / s r - - -■ -J ' . . - - - -■ - - ,/ s / 3 3 1 c c < N 3 3 C c J 3 3 § o < a. u < X 1- < < - 1- J" • z > 3 i < 5' 3 1 se pa ?e 7 4. PLATE 10. o o o § [00 - Oo 00 00 "^ mo- CO? 2>< <-'uJ u B DC U 3 K CO > 2 CO QC Q-l > + < J. < K) CvlOD N m in- o 0) CD N «0 m