LIBRARY OF THE UNIVERSITY OF CALIFORNIA. Class THE ONE SHILLING. NET 'CLIMATE OF I I AUSTRALASIA IN REFERENCE TO ITS CONTROL BY THE SOUTHERN OCEAN BY PROFESSOR J. W. GREGORY, D.Sc., F.R.S. or T IVER MELBOURNE ; CHRISTCHURCH, WELLINGTON, DUNEDIN, AND LONDON : WHITOOMBE AND TOMBS LIMITED. Works by the same Author. THE AUSTRAL GEOGRAPHIES, second edition, revised to date, as used in Australian schools. Class II. and III., each 48 pp., stiff paper, 4d.; Class IV., 116 pp., cloth, 9d.; Class V., 112 pp., cloth, 9d.; Class VI., 144 pp., cloth, Is. "Easily first among the many Geographies that have been produced for the use of teachers and pupils in Australia,."— Education Gazette of Victoria, July, 1903. The series received special attention in the New South Wales Commissioners' Report on the Systems of Education in various countries, as the only series which has successfully attempted to apply to Australian conditions the new methods in Geography teaching. THE TEACHING OF GEOGRAPHY, second edition, revised and enlarged. Price Is Prescribed for the use of teachers by the New South Wales Department of Public Instruction under its new Syllabus. " We earnestly commend them to read Professor Gregory's thoughtful words."— N.S.W. Education Gazette. "Should be carefully studied by every teacher." — Australasian School- master. "All teachers of geography should provide themselves with this treatise."— Professor David, F.R.S. THE GEOGRAPHY OF VICTORIA, 290 pages, cloth, 3s. 6d. Prescribed for the Teachers' Examinations by the Victorian Education Department, and for the Senior Public Examinations of the University of Sydney. Though treating primarily of the Geography of Victoria, this book is incidentally one of the most reliable and up-to-date text-books of general Physical Geography. THE Glimafe of Australasia IN REFERENCE TO ITS CONTROL BY THE SOUTHERN OCEAN. Expanded from the Presidential Address (delivered as an Evening Lecture) to the Geographical Section of the Australasian Association for the Advancement of Science, Dunedin, nth January, 1904. BY PROFESSOR J. W. GREGORY, D.Sc., F.R.S. Author of the " Geography of Victoria" the " Austral Geographies" " The Teaching of Geography? etc. MELBOURNE ; CHRISTCHURCH, WELLINGTON, DUNEDIN AND LONDON : WHITCOMBE AND TOMBS LIMITED. 011444 BEJEBAL CONTENTS. PART I. INTRODUCTORY. PAGE. Geography and Technical Education. ... ... 7 PART II. THE SOUTHERN OCEAN. CHAP. I. Its Name, Range, and General Characters ... 14 2. The Temperature of its Waters 23 3. Its Currents and Drifts ... ... ... 25 4. Our Comparative Ignorance thereof ... ... 27 PART III. OCEANIC CONTROL OVER CLIMATE. 29 CHAP. I. Monsoonal Winds and Rainfall ... ... 31 2. Cyclonic and Anti -cyclonic Systems ... ... 35 3. Winds and Oceanic Circulation ... ... 36 4. The Effect of Ocean Currents on Climate ... 39 5. The Irregularities in Oceanic Circulation ... 40 PART IV. THE WEATHER CYCLE. 45 CHAP. i. Bruckner's Period ... .. ... ... 51 2. The Balance of Oceanic and Continental Weather Conditions ... ... ... ... 57 3. Necessary Irregularity in Length ... ... 64 4. The Meteorological Provinces of Australasia ... 66 5. Briickner's Period and the Sun ... ... 69 6. Historic Recognition of the Bruckner Period ... 71 7. Lockyer's Law of Famine Recurrence in India ... 72 PART V. THE SOUTHERN OCEAN AND OUR WEATHER. CHAP. I. The Influence of the Southern Ocean on the Indian Climate ... .. ... ... 78 2. Iceberg Irruptions into the Southern Ocean ... 81 3. The Interval between Disturbances in Oceanic Circulation and their effect on Weather ... 83 PART VI. CONCLUSION. CHAP. I. Wanted, a United Meteorological Service for Australia ... ... • 85 2. The Prospects of Long Period Weather Prediction 88 APPENDICES. 1. Soundings and Temperature Observations made by the Challenger^ Valdivia, and Gauss, in the Southern Ocean ... ... ... 92 2. Some Iceberg Records in the Southern Ocean, Oct. and Nov. 1903 .. ... ... 95 3. Extracts from one of the Annual Seasonal Fore- casts issued by the Indian Meteorological Department ... ... ... ... 96 LIST OF ILLUSTRATIONS. PAGE I — Map of Antarctica ... ... ... ... ... 15 2 — Southern, South Pacific, and South Atlantic Oceans at close of Challenger Expedition ... ... ... ... 17 3 — Temperatures in the Southern Ocean between 50° S. and the Antarctic Circle, Feb., 1874 ... ... ... 24 4 — Distribution of the soundings in part of the Southern Ocean shown by reference to an equal area of Australia .. 28 5 — General Diagram of Oceanic Circulation... ... ... 38 6 — Melbourne Rainfall, 1840-99 ... ... ... ... 47 7 — Variations in the volume of the Murray near Mildura ... 48 8— Floods of the R. Darling ... ... ... .. 49 9 — Variations in depth of water in Lake George ... ... 53 10 — Early and Late Harvests, 1501-1890 ... ... ... 56 H — Rainfall at Bourke, Wilcannia, Horsham, Deniliquin, and Bathurst, 1855-99 ... ... ... ... 59 12 — Departures from Average Rainfall on the Coast and Interior of New South Wales ... ... ... ... 60 13 — Rainfall Map of New South Wales for 1894 ... ... 61 14— „ „ ,, ,, 1899 ... 62 15 -Departures from Average Rainfall at Sydney and Adelaide.. 63 16— Climatic Provinces (after Supan) ... ... ... 67 17 — Long Period Variations of Sun Spots, Magnetic Phenomena, and Climate ... ... ... ... ... 70 18 — Relations of Rainfall Variation of India and Mauritius - according to the Lockyers' Principle .. ... 79 PREFACE. WHEN invited to deliver the Presidential Address to the Geographical Section of the Australasian Association for the * Advancement of Science as an evening lecture, I thought it advisable to choose a subject of more general interest than that which I had previously selected. Meteorology is the branch of geography in which, at present, the widest interest seems to be felt in Australasia, and in which well directed research promises the richest reward. I accordingly resolved to attempt to explain the basis of hopes for the possibility of long distance weather forecasts, and to suggest some lines of research, which must be undertaken before Australasia can secure the incalculable benefits of such foreknowledge. Success in this work largely depends on the existence of a permanent weather cycle, which is now well established, and has only so long escaped recognition^ because, owing to its dependence on changes in the sun, its length is variable, and it inevitably produces contrary effects in different geographical areas. This cycle does not offer us any simple direct evidence as to the nature of the forthcoming seaspns ; for it affects the weather of the continents mainly by causing changes in the circulation of the oceans. Now we know that ocean currents vary from year to year, we can understand how the actual irregularities in the seasons are fully consistent with the obvious oceanic control of the climate of the lands in the southern hemisphere. As these irregularities are caused rpy processes which take a year or more to run their course, we have the means of watching in one year the evolution of the weather we shall experience in the next. The problems of long distance weather forecasting may be solved by two methods, each applicable to special areas. 6 PREFACE In this address I have referred mainly to the influence of the oceanic circulation, as its effects may be traced directly in the land masses of the southern hemisphere. In the more complex conditions of the northern hemisphere the oceanic control may not be recognised so directly ; and there seasonal weather changes are necessarily considered in relation to the atmospheric pressure systems, which have been investigated with .such interesting results by the American meteorologists. Some meteorologists still regard the prospects of seasonal weather predictions as an unattainable vision. But I find it difficult to understand this view. The work is no longer a mere speculative possibility. The Indian Meteorological Department has issued seasonal forecasts for the last 15 years, and with brilliant success (see Appendix III), where the extracts illustrate the Seasonal Forecasts issued by the Indian Meteorological Department, and show their depen- dence on the variations in the currents of the Indian Ocean. It is quoted from the Report of the Administration of the Meteorological Department of the Government of India, 1898-1899, pp. 26-29. There has been only one failure, which happened during a year when the conditions of the sun were abnormal, and which could have been avoided had the Lockyers' principle been then available. Similar forecasts for Australasia will no doubt continue impossible until our meteorology has been advanced another stage. I have agreed to the separate publication of this address in the hope that it may do something to help the proposals for the establishment of a united meteorological service for Australasia, which should work in close co-operation with those of India and New Zealand. J. W. GREGORY. January, 1904. The Climate of Australasia. PART I.— INTRODUCTION. GEOGRAPHY AND TECHNICAL EDUCATION. THE great educational controversy of the century was whether the study of literature and philosophy was of higher educational value than reasoning that can be checked by experi- ment an'd observation. In that conflict the defenders of the existing educational system ranged themselves round the standards of classics and authority. The attack came from^ those who were inspired by the new-growing enthusiasm for science. The results of the battle were the admission of science into the old schools of learning, and its predominant influence in all the new, a revolution in the methods of primary education, and a remarkable revival in classical research. The victory of science is so complete, and in one respect so unexpected that, even in Oxford, compulsory Greek at the THE CLIMATE OF AUSTRALASIA first examination — that last relic of the claim that some knowledge of Greek is indispensable to culture — cannot last much longer ; and it need not in the seat of learning where the renewed activity in classical research has re-discovered for us the revenue laws of Ptolemy, and has revealed to us further sayings of Christ. The old controversy is closed. With the new century, the centre of educational conflict is changing, and there are signs of a new strife, which has some analogies with that of the last century. With the vast growth of the extent of science, no student can keep in touch with it all. Decade after decade the sciences are being further divided and subdivided, and the minor divisions of a subject to-day have a greater literature than the parent science had a century ago. Men engaged in scientific research are not expected to know more of other sciences than is necessary for the efficient pursuit of their •own. But our students still have to spread themselves over many subjects, although all of them are ever widening in their range, and ever increasing in the refinement and complexity of their methods. The time is approaching when a change in scientific education will be necessary to secure some limitation to its range. The problem that is now pressing upon us, is the selection of the subjects which are of the INTRODUCTION 9 most educational value. There are branches of science which appear to be travelling on blind roads, or have already been driven far ahead in useless isolation. There are some, which, so far as we can judge, are at present as purely academic as the classics. On the other hand, there are branches of science in which any progress is at once turned to practical account ; while there are industries whose work is hampered because science does not give them as much help as it would do, had important lines of research not lagged far behind the rest. Some of the modern advocacy of technical education is no doubt due to impatience with anything that is not immediately useful. To modify our educational system to satisfy that spirit, would injure education, would cramp scientific progress, and ultimately prove disas- trous to industry. The best feeling in the advocacy of technical science seems to me due to the recognition of the alternative, that scientific teaching must inevitably be restricted in its range, or must become so superficial and elementary in its standard, that it will lose much of its value as a medium of education. It is generally held that the essence of true education is method, not matter. As Mark Pattison has so well put it in his Oxford Studies, it is not knowledge, but a discipline that is required, not science, but scientific habit. In accordance with this principle, 1O THE CLIMATE OF AUSTRALASIA the advocates of technical education claim that, in the selection of branches of science that are to be taught, if two subjects are of equal educational value, preference should be given to that which is also of economic value. More- over, many teachers believe that the principles of a science can, in most cases, be best taught by the study of those branches of it which are of most practical service to man. To justify that belief, the teaching of the technical subjects must be improved, and, must give at all costs, a rigorous training in scientific method. Most branches of applied science afford excellent educational material, for their results are tested daily by experiment on a scale vastly larger than science could afford. Applied science can be so taught as to give an especially powerful stimulus to the imagina- tion, owing to its victories over space and time, its conversion of waste matter into useful products, and those triumphs over disease which have dispelled the delusion that there is any- thing in climate to prevent the successful occupation of any part of the world by white races. The claims of technical education may be resisted by the advocates of pure science, on the ground that, in education, it is better to concentrate the attention of the student solely on the methods of work, and not to distract him and excite his impatience by even glimpses of the goal. But teachers can be trusted to INTRODUCTION I I convince their students that the goal will never be reached except by thorough and honest work ; and surely any student will work all the more cheerfully if he know that college work is arm- ing him with ammunition for subsequent use, and is useful not only as an intellectual treadmill. The fear that academic science will be neglected if technical science be used for educa- tional work, seems to me somewhat exaggerated. The revival of classics since the loss of its educational monoply is a valuable lesson. Academic science is even less likely to be neglected : it will always have its attractions for intellectual hermits, who prefer to work on subjects where they will not be worried by the practical man of the world ; and it will always have its appeal to the fancy with its true fairy tales. As applied science must always be dependent on theoretical science, the greater the prosperity of applied science, the more urgent will be its demands on pure science for further information, and the more generously the world will be prepared to pay for pure scientific research. The victory of science in the last century has not led to the predicted overthrow of culture. Instead of its having been fatal to the study of classics, it has inspired a period of unrivalled brilliancy in classical research. So far from its having destroyed all respect for past history, and led to the neglect of books of authority, it has caused the re- writing of history on more instructive lines, and 12 THE CLIMATE OF AUSTRALASIA has brought the various scriptures into the category of popular literature. So far from being fatal to art, it has supplied the means for a modern renaissance. I venture to introduce these general educa- tional principles, because they have especial application to geography, the teaching of which has, of all familiar subjects, been the last to benefit by recent progress. What passes as commercial or economic geography is too often a mere catalogue of natural products and climatic conditions. Geography in British education has been too often treated merely as popular topography, and has suffered from its inclusion among the arts subjects, and by being pursued by the methods of the humanist studies. Geography must be studied as a science, with methods better adapted to its materials and problems, and with a more exact and adaptable terminology. It must also look below the surface, and seek deeper into the nature of things. In the more crowded, settled parts of Europe, geography may not be so useful a subject educationally as it is in a new country, such as Australia. In Europe, with every parish mapped, every lake known, every mountain range explored, unsolved geographical problems do not present themselves daily ; but here in Australasia, geography provides the most useful educational material that is' available to us. On all sides we are confronted by the INTRODUCTION 13 unknown, or what educationally is perhaps even better — the imperfectly known. Geo- graphy accordingly gives us the readiest means of exciting that passion for research which is the life blood of science, and of stimulating that habit of independent enquiry which is the essential aim of education. While in Australasia geography should be one of the chief subjects in elementary education, it should also receive more recognition as a good subject for the teaching of research. Geography on land has made excellent progress, owing to its practical value. The need for irrigation has caused the study of our rivers, which in some states, such as Victoria, have been investigated with unusual detail and accuracy. Interest in the rainfall has enlisted an army of volunteer observers, who collect rainfall statistics with admirable intelligence and patience. The cutting up of our vast territories has necessitated the preparation of lease plans, which form an admirable basis for fuller maps. But the oceans that surround Australia have been almost entirely neglected. What we know of them, except what our feeble eyes can see from the surface, we owe almost entirely to England and to Germany ; and an appeal that Australasia should herself take up the investigation of the adjacent oceans, can be fully justified by economic consideration alone. PART II.— THE SOUTHERN OCEAN. CHAPTER i. — NAME, RANGE, AND GENERAL CHARACTERS. When on 25th September, 1513, Nunezde Bal- boa and Francisco Pizzarro, "silent from a peakin Darien," discovered the great ocean that divides the western from the eastern world, the part they saw lay to the south of them. So they called it the Mar del Zur, the South Sea. That inappropriate name was replaced by Magellan's more attractive title of the Pacific Ocean, which was given eight years later ; and though the South Sea Islands is still a familiar expression, it was only in modern times that the name of Southern Ocean was revived for the unbroken belt of sea that encircles the southern hemis- phere. The title of the Southern Ocean to recognition is not yet universally admitted. A committee of the Royal Geographical Society,* which, in 1 845, considered the nomenclatureof the oceans, recommended their division into five— the Atlantic, Pacific, Indian, Arctic, and Ant- arctic. The Southern Ocean was divided among the Pacific, Indian, and Atlantic, along arbitrary *The report of this Committee was not published at the time, but its minutes have been issued in a note entitled " Nomenclature of the Oceans." Geographical journal. Vol. I., London, 1893, pp. 535-6. THE SOUTHERN OCEAN 15 rtieridians, while it was cut short to the south by the Antarctic Ocean. This classification of the oceans has been so widely accepted by the compilers of British atlases, that it is in common use. Modifications in this simple scheme are, however, necessary. Thus the Antarctic Ocean, Fig. I — -Map of Antarctica. which is defined as occupying the area within the Antarctic circle, must be abandoned ; for within that limit there is probably far more land than sea. Instead of the Antarctic area being occupied by ocean, it appears to consist of a great land mass, into which two seas — the Weddell Sea and the Ross Sea, project pole- ward from the oceans that lie to the north. A 1 6 THE CLIMATE OF AUSTRALASIA second important change in the popular British classification has been made by calling the gathering of waters south of the latitude of 40° S., the Southern Ocean. This innovation has been sanctioned by the English Admiralty, and is adopted on recent charts ; though the official Sailing Directories still divide the Southern Ocean between the South Atlantic, Indian, and South Pacific Oceans. Recent oceanographic work has supported the claim of the Southern Ocean to recognition as an in- dependent geographical unit, but in a some- what restricted sense. Its northern boundary is generally defined as the 4ind5|c.Tm8|S.|iTradcs.| faTmq |N.E.T00 3 ~ £ S- T H 1 1 ? H i M 1 1 1 f 7 i j it i r 1 1 < T smii x Is the year of maximum discharge ; x, x, mark periods n years apart. L is the year of lowest discharge ; L, L, 1 1 year periods from it. FIG. 7. — Variations in the volume of the River Murray, near Mildura. gauge in Australia — the River Murray. The river gaugings taken near Mildura show great variations in the volume of the Murray ; and the variations show no agreement with the 1 1 years period. The floods of the Darling (fig. 8) also show no support to the 1 1 year cycle. While Lockyer and Meldrum have advocated an 1 1 years cycle in connection with the sun- spots, other men have claimed that the weather varies in cycles of longer periods. Dr. Russell, of Sydney, to whom Australian meteorology is ols J 1 II I 1 1! 1 ) } \ \ 1 I i M 1 1 1 I I „ - - • — t i 1 1 M I 1 1 M 1 1 1 II) 1 1 » iTT >-*«-« o oo^o ^ in o*"^0 •*• 1-5^ -3- -* S? ^, 4; ->2 ^ -*^ . 5O THE CLIMATE OF AUSTRALASIA so greatly indebted, advocates a weather cycle of 19 years,* and Bruckner of Vienna holds that the cycle is about 34 or 35 years. Again Wolf holds that, in addition to the 1 1 years period in sun spot variation, there is also a long period variation, 55 years in length, which has claims to consideration, in reference to weather cycles. In the face of such divergent opinions as to the length of the supposed weather cycle, we may well ask, is there any weather period at all ? Ordinary weather statistics do not show any obvious periodicity. For instance, years of heavy rainfall at Melbourne appear to be simply capricious in their recurrence ; and Symons, to quote a European authority, denies that there is any regular oscillation in the British rainfall, or that the rainfall increases and decreases regularly like the swing of the pendulum. Such indisputable irregularities and the different lengths assigned to the supposed cycle by different authorities, appear, at first sight, to be convincing proofs that there is no weather cycle at all. If there were, it may be thought, the cycle should be as easily discovered, and as well known as the succession of the seasons. The view is accordingly prevalent that the wind bloweth where it listeth. Practical men follow the fatalist creed of Ecclesiastes, " He that observeth the wind shall not sow, and he that regardeth the clouds shall not reap ;" and *H. C. Russell " On the Periodicity of Good and Bad Seasons," Proc. Royal Soc. N.S. Wales, vol. xxx (1897) pp. 70-115. THE WEATHER CYCLE 5! many meteorologists still adopt an agnostic attitude in regard to the existence of any regular periodicity controlling the weather of the world. But the fact that inconsistent lengths are assigned to the weather cycle merely shows that there is no simple weather cycle of precise and unvarying length, which affects the whole world, in the same way, at the same time. But there is no reason to expect any such crude and simple arrangement. To search for such is to chase a Will o' the Wisp. The weather of our infinitely complex earth is not likely to be con- trolled by any governor so simple in its action. It does not follow that there is no system at all. If there be a system, we must expect one which is subtle, refined, and difficult to trace. CHAPTER I. — BRUCKNER'S WEATHER PERIOD. As the possibility of a weather cycle is the fundamental problem of meteorology, let us look at some of the evidence that has been collected to determine its existence. Owing to the complexity of our world and the delicacy of our atmosphere, minor variations are sure to be so numerous and apparently capricious, that a vast mass of material must be examined in order to secure satisfactory results, and to eliminate local irregularities. The most serious attempt 52 THE CLIMATE OF AUSTRALASIA to handle this question has been made by Bruckner of Vienna, whose great book — " Klimaschwankungen," the Variations of Cli- mate— published in 1891, first placed this subject on a satisfactory footing.1* Bruckner was led to his investigation by consideration of the changes that take place in the level of the Caspian Sea. It has long been known that the waters of the Caspian alternately advance and recede. The records go back to the tenth century, and they are complete since early in the eighteenth century. Bruckner found that there was a regular cycle in the rise and fall of the Caspian waters, and that the average dura- tion of a complete cycle was from 34 to 36 years. The dates of the maxima for the Caspian were about 1743, 1780, 1809, 1847, and 1879. As the Caspian is an inland sea, its level must depend upon the amount of water supplied by rainfall, and on the rate of its removal by evaporation. The Caspian acts as a great rain gauge for its vast drainage area. A high level of the Caspian, therefore, indicates that the preceding years have been abnormally wet, or that the temperature of the Caspian was low, so that evaporation was slow, or else that both factors worked together. Lakes which have *The best summaries of this work in English are in a paper by F. Waldo in the American Journal of Science, ser. 3, vol. xli., 1891, pp. 141-151. . There is a longer and admirable summary in Waldo's " Modern Meteorology, "London, 1893, pp. 406-421 ; and some account in the recent translation of Hann's Climatology. (London, 1903). The three books are available for reference in the Melbourne Public Library. THE WEATHER CYCLE 53 no rivers flowing out of them, all act as rain gauges for their drainage basins ; and Bruckner found that the levels of other inland lakes, with- out outlets, also varied in cycles of the same length, and reached their maximum at about ore* v» o -jt> *•• o» o- i A r i i i i r T r T r tn~i » i n » i M r \ i • < i i ' \l \1 . II • lo • 1J ^ n » it • it : a - ti - 10 - « - G . T - 3 . X _ 1 n - FIG. 9. — The variations in depth of the water in Lake George. (From Russell). the same dates as the Caspian. Bruckner therefore considered the rainfall of 32 1 localities, some of which have rainfall records for over two centuries. Thus the rainfall data of Paris go back to 1691. Bruckner's study of the rainfall convinced him that it also varies in a cycle of 54 THE CLIMATE OF AUSTRALASIA about 35 years ; and Hann's more recent, and still more detailed study of the rainfall of Padua since 1725, and of Milan since 1764, has con- firmed Bruckner's conclusion.! Bruckner's study of the rainfall, therefore, showed that the variation of the lakes was the result of the variation in the rainfall. Thus the ake levels were lowest in 1720, 1760, i8oc 1835, 1865; they were highest in 1740, 1780, 1820, 1850, and 1880. Bruckner found that there were dry periods in 1831-40, 1861-65, anc^ wet periods from 1846-55 and 1876-80. So the lakes were at their highest, at about the time of the wet periods. As rain is a product of distillation, variations in its amount probably indicate periods of warm and cold weather. Bruckner, therefore, en- quired whether there was any connection between the temperature records and these dry and wet periods. He found that from 1736- 1885 there has been the same 34-35 years cycle of variation in temperature. Severe winters and the freezing over of European rivers are recorded in history since early times ; records are available as far back as the year 800, but they are too incomplete to be of much use before the year 1000 ; and the records show a temperature variation of the same length as the rainfall and the lake levels. Thus, Bruckner finds that there were cold periods from 1591- 1600, 1611-35, ^646-65, 1691-1715, 1730-1750, fllann's Memoir is published in Sitzber. K. Akad. Wien., vol. CXI. II. a, 1902, pp. 1-120. THE WEATHER CYCLE 55 1766-75, 1806-20, 1836-55. Taking the cold periods, Bruckner finds that from the years 1 020 to n go the oscillation period was 34 years ; from 1 190 to 1370 it was 36 years ; from 137010 1545 it was 35 years ; 1545 to 1715 it was 34 years, and from 1715 to 1890 it was 35 years. The date at which the rivers of Northern Russia and Siberia are opened to trade by the melting of the ice in the spring, becomes on the whole, alternately earlier and later, and the length of this oscillation is consistent with the rest. The snouts of the Alpine glaciers advance down their valleys and then recede and advance again ; and though — owing to the local variation in snowfall— all the glaciers do not act simultaneously, the movement, as a whole, is also in a 34—35 years cycle. The beginning of the grape harvest in France, southern Germany and Switzerland is celebrated as one of the chief national festivals of the year. The harvest is ripe earlier when the summer is warm and dry, than when the season is cold and wet. And the dates of the first day of harvest are known far back into the Middle Ages. These dates show an advance and retreat in an oscillation of the same length as the other weather variations. The price of grain is a good index to the nature of the seasons ; and Bruckner claims that the price, looked at broadly, shows the same vicissitudes as the various meteorological factors, which control it. THE CLIMATE OF AUSTRALASIA I < jsr w» j C7 c o a W S ft 0 S £ ft 0.5 - - • — c» -Sf 1 or .« OT .si s 0 91 CA jio In -rt> u 5 .ti i .« cS •c* on rt .tf ta ,<\ 0 ot 'S -W g cfc rt 4 5 1 '5 0 XI 6 2 £ of 07 "*' A 3J The weather cycle thus estab- lished, is not invariable in length, and it does not affect the whole world similarly and sim- ultaneously. The figures above quoted show that the cycle has varied in different centuries from 34 to 35 to 36 years. The mean length is given as 34-8 +_ 7 years. Moreover, it must be remembered that the variations of any one of the meteorological elements is the result of the action of a complex of varying and opposing agencies. It is inevitable that there will be irregularities, which will be exaggerated in appearance by the artificial divisions of our annual weather records. The calendar year is not altogether satisfactory meteorologically ; the year from spring to spring or from autumn to autumn would give better results than the present system of dividing the southern summer and the north- ern winter between two years' records. We must be prepared for unexplained inconsistencies in the data, and when we con- sider the extreme sensitiveness of the weather to local causes, the surprising fact is that only THE WEATHER CYCLE 57 8 per cent, of Bruckner's material gave dis- cordant results. Bruckner found that the rainfall records of Europe and Asia agree precisely ; and are in general agreement with those of North America ; but the records of South America and Australia are much less regular in their agreement. CHAPTER II. — THE BALANCE OF OCEANIC AND CONTINENTAL WEATHER CONDITIONS. TKe most important of the apparent incon- sistencies were due to intelligible, geographical causes. Bruckner found, early in his investi- gation, that, while the level of those continental lakes which have no outlets rises and falls in agreement with the Caspian, yet many other lakes, though they vary in a cycle of the same length, are highest when the Caspian is lowest. Bruckner, therefore, described these lakes as being in areas of permanent ex- ception. Such exceptions, in this case, verily prove the rule. The Caspian is lowest after years when the rainfall in its drainage area is at its lowest. These are years when the great continental mass of Europe and Asia is under conditions of especially high atmospheric pres- sure. A prevalence of such anticyclonic conditions results in comparative drought. 58 THE CLIMATE OF AUSTRALASIA When anticyclonic conditions prevail on land, cyclonic conditions, with extra rainfall, are developed at sea. Thus, when a smaller percentage of rain falls on land, a larger percentage should fall on the sea. There are no records of the annual rainfall at fixed stations out at sea, and but too few on the small oceanic islarids ; but the coast stations share in the meteorological conditions preva- lent in the adjacent oceans. Hence, when the centre of the continents lie under their anti- cyclonic conditions, and are passing through dry years, the coastal regions are undergoing cyclonic conditions and periods of compara- tively heavy rainfall. Therefore, the Caspian and great inland lakes should be emptiest, when the coastal lakes are fullest. This fact is abundantly established by historical records, and it shews us that we cannot expect all the rainfall records or all the lake levels of a great continent to rise and fall together. The movement will be a great see-saw, the interior and coastal districts varying inversely. This important fact is well illustrated by the rainfall returns for Australia. For example, rainfall curves for localities in the western plains of New South Wales and at Horsham in north-western Victoria vary in common (fig. 1 1 ). If now we look at the rainfall curve for one of those localities and contrast it with that of Sydney, we find that the conditions are reversed (fig. 1 2). When the rainfall is heavy THE WEATHER CYCLE 59 1 6O THE CLIMATE OF AUSTRALASIA Sydney, it is light in the interior. This fact is illustrated for the whole of New South Wales by the instructive rainfall maps of that State, prepared by Dr. Russell. The shaded areas in the maps (figs. 13 and 14) are the districts where the rainfall was above the average ; and the plain areas are districts where the rainfall was below the average. In the FIG. 1 2. — Departures from average rainfall on coast and interior of New South Wales. year 1894, the whole of the inland or western district had a heavier rainfall than usual, while the coastlands had less than their average. The reverse arrangement is shown for the year 1899. The balance of the coastal and inland conditions is not always shewn with such diagrammatic simplicity, for in some years the coastal and continental rainfalls are both above the average, owing no doubt to an increase in THE WEATHER CYCLE 6l the amount of rain raised from the sea by evaporation. We cannot expect that rainfall will anywhere vary with mathematical regu- larity. Rainfall is one of the most complex of FIG. 13. — Rainfall map of New South Wales for 1894, shaded area showing where the rainfall was above the average. (After Russell). meteorological products, and is the resultant of a series of conflicting agencies. The supply of moisture available depends on the evaporation at sea, on the strength and direc- tion of the winds that carry it inland ; the amount of rain that falls at any locality depends on the temperature of the soil in the district, 62 THE CLIMATE OF AUSTRALASIA and also of that in the areas over which the moisture has had to pass ; and the temperature of the ground depends on the winds, on the cloudiness of the sky at night, on the tem- FIG. 14.— Rainfall map of New South Wales for 1899, shaded area showing where the rainfall was above the average. (After Russell). perature of the adjacent sea, on the amount of dust in the air, and on other factors. Whether the moisture will reach any given locality, and, if so, whether it will fall there, depends on such a complicated series of con- ditions, that it is idle to seek for any precise THE WEATHER CYCLE and invariable cycle of change. Even the records may be apparently falsified ; two sets of heavy Christ- mas storms may fall into one year's £ record, and miss 3 the alternate ye'ar < altogether ; even 1 the same storm i at ,the passage of •^ the year may fall ^ into different years 5 in different parts !s of the country. * The surprising | fact in rainfall | statistics ' is that | the cycle should ^ be as marked as g it is ; but we can- | not hope to recog- Q nise it unless we ^ separate the vari- " ous elements in £ the rainfall. For instance, we have already seen (fig 6) that the variations in the Melbourne 64 THE CLIMATE OF AUSTRALASIA rainfall appear to be very irregular. The rainfall conditions for Sydney and Adelaide are less so, although, as Russell has pointed out, they are reversed (fig. 15). Years of heavy rainfall at Sydney are years of light rainfall at Adelaide, because Sydney has entirely a coastal rainfall, and the Adelaide rainfall is mainly continental. This fact explains the apparent caprice of the Mel- bourne rainfall. Melbourne is intermediate between the conditions of Sydney and Adelaide. It receives some of the rains of the Australian coastlands, while it shares the inland rains which are the most powerful at Adelaide. If we could separate the Melbourne rainfall into its two distinct sources, we should probably find that each of them would be represented by regular curves ; but when combined the results are capricious and unintelligible. Looked at from this point of view, instead of the rainfall statistics of Adelaide and Sydney being fatal to a belief in a periodic system, they flash into agreement, and those of Mel- bourne are not inconsistent with the existence of the great cycle. CHAPTER III. — THE WEATHER CYCLE IR'REGULAR IN LENGTH. There is another essential precaution to be observed in search for a weather cycle. We THE WEATHER CYCLE 65 must not expect periods, due to sun spots or any form of solar change, to be ' always the same in length. The statement that the sun spot period is i i . 1 1 1 1 years in length, looks very precise ; but that period is only the average of nearly two centuries of records, and any one solar cycle may differ in length from the mean by as much as two years. * If, therefore, our weather cycles are connected with the sun spot period, they also will vary from 9 to 13 years. We cannot expect maximum to follow maximum with the punctuality of clock work. There is no sanctity to the number n, any more than the number 13 is unhallowed. The great variation in the length of the solar period shatters for ever any hope of our getting certain predictions as to the weather of future years from simple averages. Averages are useful, but they are easily misused. In trying to determine weather cycles we must not put our trust in the mere application of simple addition and division to haphazard collections of figures. " Professor Armstrong once had a German chemical student, who, considering his nationality, was abnormally lazy but normally systematic, and who used to determine the weight of his precipitates by asking all the students in the laboratory to guess the amount, and then taking the average. That student's * The precise length of the sun-spot cycle as given by Wolf, (1875) is ii-iin etc. + 2-030 years. 66 THE CLIMATE OF AUSTRALASIA method was as unsuitable for chemical research as is the use of blind averages in meteorology. The statements that you can prove anything by statistics, and that there is nothing so misleading as facts, is to some extent true of weather statistics of Australia. A careful selection of instances out of the Australian weather records would give any desired result. Owing to the oscillation between these coastal and inland conditions, it is obvious that the calculation of averages requires great judgment, in seeing that the stations included are equally distributed. Unfortunately Australian weather statistics, except for the coastal regions, do not date back for any considerable length of time ; so they are not particularly useful in the investigation of long period cycles. And the study of Australian meteorology cannot advance beyond a comparatively elementary stage, until the weather records of Australasia be divided amongst its natural meteorological provinces. CHAPTER IV.— THE METEOROLOGICAL PROVINCES OF AUSTRALASIA. Supan has divided Australia into four distinct climatic provinces. Most of the subtropical parts of northern Australia he includes in the 'Tndo- Australian Monsoon Province," which has THE WEATHER CYCLE 67 a tropical climate, with a very low yearly range of temperature. The eastern coastlands of Queensland, New South Wales, and the eastern part of Victoria, and the whole of Tasmania belong to the " East Australian Province." It has regular, plentiful rains, and a moderate annual range of temperature. The south- western corner of Western Australia forms another province, characterized by a subtropical i. African Tropical. 2. Kalahari. 3. Cape. 4. Indio- Australian Monsoon. 5. Inner Australia. 6. S.W. Australia. 7. East Australia. 8. Polynesian Tropical. 9. New Zealand. FIG. 16.— Map showing the Climatic Provinces. (After Supan.) climate, with fairly equable annual conditions. The rest of Australia, extending from the western coast to the head of the Australian Bight and Spencer Gulf, and bounded eastward by the main divide of eastern Australia, and northward by the I ndo- Australian Monsoonal Province, is the " Inner Australian Province." It is characterized by great extremes in temperatures, 68 THE CLIMATE OF AUSTRALASIA and very scanty rains, which fall at irregular intervals. The recognition of these meteoro- logical provinces is of primary importance, because it helps us to understand why the same meteorological variation affects some localities simultaneously, and others at different times. For the solar variation affects different regions by a different chain of circumstances. Thus Mauritius, being in a different meteorological province from Bombay, may feel the effects of the same variation in the great controlling agent at a different date. Moreover, there are interesting cases in which areas on the earth's surface, which are antipodal in position, have their meteorological variations reversed. Some parts of India lie under high pressure, while their antipodes in North America are under low pressure.* This opposition is to some extent a necessary con- sequence of the antipodal positions of land and water on the globe. Nearly every point on the land of the globe has water at its antipodes. In fact, only x/27 of the land surface of the world has land at its antipodes. As there is then a see-saw oscillation between the conditions of continents and oceans, it is only natural that antipodal localities, though they may show the same cycle, have their meteorological circum- stances reversed * Sir Norman and W. J. S. Lockyer, " On the Similarity of the Short-period Pressure Variation over large Areas," Proc. Royal Soc. , Vol. LXXL, 1903, pp. 134-5. THE WEATHER CYCLE 69 CHAPTER V. — BRUCKNER'S PERIOD AND THE SUN. The existence of the 35 years periods, which Bruckner has established with such a convincing array of evidence, is all the more remarkable because it is in agreement with a long period variation in the sun, which was unsuspected when Bruckner wrote his monograph. Bruck- ner was so convinced that his 35 years' cycle must be caused by changes in the sun, that he turned to solar records, confident that he would find in them some variation, of which the variation of the weather on earth was an echo. At the time of his work, it was believed that, in addition to the nx/9 years sun spot variation, there was a longer sun spot cycle, of which the length was 55 years. That cycle was claimed by Wolf. Bruckner searched the weather records for any trace of this 5,5 years period His elaborate data showed that the weather varies in a 34-5 years' cycle ; and as astronomers told him that the sun spot variation was 55 years, he logically accepted the conclusion that sun spots had no influence upon the weather. Bruckner therefore predicted that some other, and then unknown variation of the same period as his own, would be discovered in the sun. Bruckner's confidence has been justified ; and the main argument against any relation between sun spots and the weather has been removed. ;o THE CLIMATE OF AUSTRALASIA It is now known that there are three solar cycles. There is a short 3^ years' secondary period in sun spots, which is reflected by the Sunspot Curve. Magnetic Curve. Variations in Rainfall in Bruckner's Normal Districts. The rainfall of the whole earth. \7 . -\. FIG. 17.— Long Period Variations of Sun Spots, Magnetic Phenomena and Climate. (W. J. S. Lockyer.) THE WEATHER CYCLE 7 I records of atmospheric pressure in India and Argentina ;*. there is the well-known cycle of 1 1 r/9 years ; and there is a longer period variation, of which the length has been estab- lished by Dr. W. J. S. Lockyer at about 35 years.t CHAPTER VI. — HISTORIC RECOGNITION OF BRUCKNER'S PERIOD. It may be urged against the validity of Bruckner's weather cycle that, if it exist, it should have been recognised centuries ago, and long since placed beyond the range of contro- versy. But the meteorological cycle will be obvious to the casual observer only in areas of unusual meteorological simplicity, and where one meteorological factor is of great importance. Holland occupies such a position. It is part of the meteorological province of West Europe, and floods are of extreme importance to it. Bacon's essay on the " Vicissitudes of Things " announces the Dutch belief in a 35 years' cycle * Sir Norman and VV. J. S. Lockyer, " On some Phenomena which suggest a short period of Solar and Meteorological Changes," Proc. Royal Soc., Vol. LXX, 1902, pp. 500-504 ; " On the Similarity of the Short-period Pressure Variation over large Areas," Ibid., Vol. LXXL, 1903, pp. 134-135, 2P1- tW. J. S. Lockyer, "The Solar Activity, 1833-1900, Ibid., Vol. LXVIII., 1901, p. 294. 72 THE CLIMATE OF AUSTRALASIA in the following remarkable passage : — " There is a toy, which I have heard, and I would not have it given over, but waited upon a little. They say it is observed in the Low Countries (I know not in what part), that every five and thirty years the same kind and sute of years and weathers comes about again ; as great frosts, great wet, great droughts, warm winters, summers with little heat, and the like ; and they call it the prime ; it is a thing I do the rather mention, because, computing backwards, I have found some concurrence." Australia has also contributed to the evidence in favour of the Bruckner period. Mr. Charles Egeson, of Sydney, a clear-sighted meteorolo- gist, found by a study of some of the meteorological elements for Sydney for the month of April, that they varied in a cycle of from 33 to 34 years. Mr. Egeson's conclusions are of especial interest, as they were published* two years before the appearance of Bruckner's great work. CHAPTER VII. — THE LOCKYERS' LAW OF FAMINE RECURRENCE IN INDIA. The Australian weather records, however are not in a suitable form for satisfactory treatment, and I, at least, have not had time to * C. Egeson, " Weather System of Sunspot Causality," Sydney, 1889. THE WEATHER CYCLE 73 interpret them. The meteorology of India, which has had its weather more carefully studied than any other tropical country, affords more instructive lessons. It is naturally to India that meteorologists first turned for traces of the effects of the sun spot cycle. Some facts were found in agreement with this cycle, and some of them were so striking that even such cautious meteorologists as Blanford were impressed by them. But India, as a whole, gave no clear evidence of an IT years cycle, and the few agreements were dismissed as mere coincidences. Even when there were local agreements with the cycle, the action was often reversed. Thus, while the evidence of some stations suggested that abundant sun spots accompanied a high rainfall, other stations showed exactly the reverse, as sun spots were most numerous at the time of lowest rainfall. This contradictory evidence was discouraging ; and the first study of Indian rainfall records gave only enough agreement with the theory to tantalize, but not to convince. But during the last three years Sir Norman Lockyer, in conjunction with his son, Dr. W. J. S. Lockyer, has found a clue to the apparent irregularities of the Indian weather.^ Sir Norman Lockyer was one of the first to call attention to the possible connection of sun spots and the rainfall * Sir Norman and W. J. S. Lockyer, "On Solar Changes of Temperature and Variations in Rainfall in the Region surrounding the Indian Ocean,'' Proc. Royal Soc., Vol. LXVIL, 1901, pp. 409-430. 74 THE CLIMATE OF AUSTRALASIA variations ; and his 30 years' pursuit of the rainfall cycle has at length been rewarded by the most fruitful and suggestive results yet obtained. The Lockyers' explanation of the apparent caprice of the Indian rainfall is based on a study of the spectra of all the sun spots that have appeared since 1876. Their work has enabled them to use sun spots as chemical thermometers, showing the variations in the temperature of the sun. The Lockyers have studied the spectrum of every sun spot, since 1876, which was sufficiently large for spectro- scopic measurement. The results show that the sun spots indicate changes in the heat of the sun, and are most numerous at periods of intense solar activity, when the sun is hottest. The Lockyers have found that at one time the majority of the lines in the sun spot spectrum are referable to elements which are known to exist upon the earth ; they therefore are called the " known" lines. At other times the majority of the lines in a sun spot spectrum do not occur in the spectra of elements known upon the earth's surface. When these unknown lines are predominant in the spectrum, the lines are wider apart, indicating that the temperature of the sun at that period is hotter than when the lines are close together. The increased heat, the Lockyers suggest, breaks up some of the elements in the sun, and so the spectrum of such material is different from that given by it THE WEATHER CYCLE 75 under any conditions yet produced on the earth's surface. The proportion of the known and unknown elements in the sun spot spectrum, of course, vary inversely. When the proportion of the known element is greatest, there is the lowest proportion of the unknown lines. When the numbers of the known and unknown lines are equal, the sun's temperature is at its mean. Accordingly the study of the sun spot spectra determines whether the heat of the sun is above or below the average. After the Lockyers had established this point, their next step was to study the records of the rainfall in Mauritius and India, to see if these variations in the solar temperature had any recognisable effect on the rainfall. When the sun spots are most numerous, their spectra show the greatest predominance of unknown lines, and the solar temperature is therefore at its hottest. It is found that at this period the south-western monsoon gives India its maximum of rainfall, while India, as a whole, being continental, is under cyclonic conditions. At the same time, Mauritius, being an oceanic island, is naturally under anticyclonic conditions, and has a minimum of rainfall. The maximum rainfall at Mauritius occurs when the sun spots and solar temperature are at their minimum, and the island is under cyclonic conditions. The rainfall curve of Mauritius is comparatively simple, and shows an T i years cycle ; but the 76 THE CLIMATE OF AUSTRALASIA rise and fall is inverted when compared with parts of India. But India, as a whole, gives no simple regular period. It has been pointed out by Archibald and Hill, that, while the south- western monsoons of India yield their maximum rains at the time of sun spot maximum, northern India receives most of its rain at the time of the sun spot minimum. This difference between northern and southern India at first seemed fatal to the belief in sun spot control. It is due to the fact that the Indian rainfall is a •compound product, which has to be resolved into its elements before the influence of the cycle can be seen. The Lockyers' work has now shown that, while at Mauritius the rainfall curve is comparatively simple, there being one maximum and one minimum in the course of the LI years, the Indian curve for the rainfall from the south-western monsoon is more complex. In India there are two maxima during the course of one sun spot cycle. In addition to its proper maximum of rainfall, corresponding to the Mauritius minimum, it has a secondary maximum at the same time as that at Mauritius ; for the conditions that deter- mine a year of high rainfall at Mauritius extend their influence to India, and thus produce there a secondary maximum. It is not surprising, therefore, that, if we lump the whole Indian rainfall together, we fail to find any simple agreement with the 1 1 years period. THE WEATHER CYCLE 77 The Indian rainfall, having two maxima during the course of the one sun spot cycle, has also two minima, which occur one on either side of that secondary maximum which corresponds to the minimum of Mauritius. Now the Indian famines have no 1 1 years period, and thus have hitherto appeared independent of sun spot influence, The Lockyers point out that all the Indian famines have occurred at the time of the minimum beside the secondary maximum. The relations of the Indian and Mauritius rainfalls from the south-west monsoon, and the position of the famine years are diagrammatically shewn in fig. 18. The agreement between the Lockyers' results and the distribution of famines in India, is so remarkable that it is claimed to be " clear from the above table that if as much had been known in 1836 as we know now, the probability of famines at all the subsequent dates indicated in the above tables, might have been foreseen." This great meteorological advance promises great assistance in the future administration of India. Sir John Eliot, the Meteorological Reporter to the Indian Govern ment, has expressed his opinion^ that the results 4 'accord closely with all the known facts of the large abnormal features of the temperature, pressure, and rainfall in India during the last 25 years, and hence that the indications already arrived at will be of great service in forecasting future droughts in India." * Lockyer, op. cit., p. 411. PART V.— THE SOUTHERN OCEAN AND AUSTRALASIAN WEATHER. CHAPTER I. — THE INFLUENCE OF THE SOUTHERN OCEAN ON THE INDIAN CLIMATE. AUSTRALASIA is more directly interested in a second outcome of the Lockyers' comparison of the rainfalls of India and Mauritius. Important meteorological events in India and Mauritius sometimes occur together, and at other times they occur a year apart. Thus the heavy rain- fall which occurs when the sun is hottest, occurs simultaneously in India and Mauritius ; but that maximum of rain which occurs at the minimum of solar intensity affects Mauritius a year earlier than it affects India. This fact the Lockyers explain by the probable assumption that the maximum of solar temperature affects the continental areas of Africa, Asia, and South America, directly, and thus India, the Cape, and Cordova in Argentina are influenced together. But the secondary maximum pro- duced at the period of solar cold, affects India and Mauritius indirectly, through changes which AUSTRALASIAN WEATHER 79 take place in the Southern Ocean. Such changes do not affect Mauritius in the same year as they influence India. To quote the Lockyers' own words : — " The delay of about a year in the effect of the Mauritius pulse being felt in Ceylon and India, is exactly what would Mauritius Rain- fall Hypothetical simple Rain- fall for India Compound Rain- fall for India FIG. 18. — Diagram of the relations of the Rainfall Variation of India and Mauritius, according to the Lockyer principle. F Famine Points. be expected if the rain at sun spot minimum comes from the south, as has been surmised. The fact that the pulses at Mauritius, Ceylon, and India in 1882 occur simultaneously is very strong evidence in favour of an origin in the equatorial region itself for the Indian rain at 8o THE CLIMATE OF AUSTRALASIA sun spot maximum. The pulse at maximum in the Indian south-west monsoon may depend to a large extent upon the action of the excess of solar heat on the equatorial waters to the south of India, and not on an abnormal effect on the south-east trade." The partial dependence of the Indian rainfall upon changes south of the Indian Ocean, even in the Southern Ocean, has been frequently considered. The Lockyers quote the following extract from Eliot's report of 1896, in which the highest official meteorologist in India expresses his belief in the connection between the Indian weather and the circulation of the Southern Ocean. " Mr. Eliot long ago conjectured that the rain- fall of India was profoundly modified by events taking place from time to time in the Southern Ocean. In his ' Annual Summary ' for [896 he wrote as follows ; 'It has apparently been established in the discussion that the variations of the rainfall in India during the past six years are parallel with, and in part, at least, due to variations in the gradients, and the strength of the winds in the south-east trade regions of the Indian Ocean. The discussion has indicated that there are variations from year to year in the strength of the atmospheric circulation obtaining over the large area of Southern Asia and the Indian Ocean, and that these variations are an important and large factor in determining the periodic variations in the rainfall of the AUSTRALASIAN WEATHER 8 1 whole area dependent on that circulation, and more especially in India. It has also been indicated that these variations which accompany, and are probably the result in part of abnormal temperature (and hence pressure) conditions in the Indian Ocean and Indian monsoon area, may be in part due to conditions in the Ant- arctic Ocean, which also determine the compara- tive prevalence or absence of icebergs in the northern portions of the Antarctic Ocean.' ' N. and W. J. S. Lockyer, op. cit., pp. 422-423. CHAPTER II.- ICEBERG IRRUPTIONS IN THE SOUTHERN OCEAN. That great periodic changes do take place in the Southern Ocean is shown by the distribution of the Antarctic ice. In some years the ice- bergs drift into comparatively low latitudes, and are a serious danger to shipping.* In 1892 one group of icebergs floated along the eastern coast of New Zealand, almost as far as Cook Strait. Three years later, one iceberg in the South Atlantic, in April, 1894, actually got within sight of the tropics (it was seen in latitude *The movements of the ice are best recorded in Towson's paper published by the Hoard of Trade, in 1858, of which I have not been able to find a copy in Victoria. Dinklage, Ann Hydrogr., Berlin, 1893, also 1894, 1897 and 1898. H. C. Russell, Icebergs in the Southern Ocean, Proc. Royal Soc. N.S. Wales, vol. xxix., 1895, PP- 286-315 pi. xii. Part II., ibid vol. xxxi., 1897, pp. 221-251. 82 THE CLIMATE OF AUSTRALASIA 26° 3C/ S.) In other years the ice is kept far to the south. These Antarctic icebergs migrate northward at intervals.^ One well known ice irruption occurred between November 1854, and March 1855, and another from 1892-95. Mr. Baracchi has recently stated (Age, 9th Feb. 1904) that' " there is no year on record since 1855 in which we have had so much rain during the first two months as we have already had up to date." The recurrence of the 1855 weather with similar and exceptional ice con- ditions south of Australia is significant. Icebergs this year have travelled unusually far to the north and to the east. The northward voyage of these icebergs must reduce the temperature of the sea through which they flow, and chill the winds which pass them. But they are them- selves perhaps more important than their effects. They are symptomatic of less conspicuous, but more powerful agencies. The long life of these icebergs may indicate the absence of the warmer waters in which, under ordinary conditions, they would soon melt ; and probably the extremely cold, changeable weather of the past two months, (November and December 1903), is a consequence of the proximity of this ice. And Mr. F. W. Hales tells me that it is a matter of experience in Southern New Zealand, that years when abundant Antarctic icebergs *Suggestions have been made that the ice came northward owing to the great quantities set free from the Antarctic area by the effect of earthquakes or earthquake waves. AUSTRALASIAN WEATHER 83 are reported by ships between New Zealand and the Cape, " we, (in Southern New Zealand), invariably got a showery spring and summer with plentiful rains in the winter follow- ing." CHAPTER III. — THE INTERVAL BETWEEN DIS- TURBANCES IN OCEANIC CIRCULATION, AND THEIR EFFECT ON THE WEATHER. Unfortunately we cannot trace the full influence of these ice-masses on the Australasian climate, as we know so little of the temperature of the sea water in the Southern Ocean. Analogy with other oceans however, suggests that the lowering of the temperature of the sea water, which is in part the cause and in part an effect of the northern position of the icebergs, must influence the climate of Australia and New Zealand. One striking case in the North Atlantic, knowledge of which we owe mainly to the work of Mr. H. N. Dickson, the lecturer on Oceanography at Oxford, shows how an unusual condition of weather in north-western Europe was due to an unusual position of the North Atlantic anticyclone, 18 months before.* Dickson's comparison of the North Atlantic in 1896 and 1897, shows that there were several differences in the circulation, which were due to *For a non-technical statement of this case see the Argus, iQth Sept. 1903. 84 THE CLIMATE OF AUSTRALASIA the fact that the North Atlantic high pressure area extended unusually far to the north, and north-east in the first half of 1896. The result was the early driving of the tropical waters far to the north, and an early spring, followed by a cold spell due to the southerly movement of the Arctic ice ; then followed a hot summer after the ice had melted. The Arctic ice broke up very early that year, and going to Spitzbergen in the spring of that year, we had to encounter a phenomenally heavy pack in the Norwegian Sea. But by the end of summer that had all gone ; the Arctic Ocean north of Europe was comparatively ice free ; and northern Europe had a warmer summer than usual. In 1897, the reduced Arctic ice broke up •early and came south, giving England a cold spell in spring ; after which, as the North Atlantic was free of ice, there was a phenome- nally hot, dry summer. That English drought was the result of the position of the Atlantic anticyclone, 12 or 18 months before. PART VI.— CONCLUSION CHAPTER I. — WANTED, A UNITED METEORO- LOGICAL SERVICE FOR AUSTRALIA. WE have seen how changes in the Southern Ocean may affect the Australian climate ; that it is clearly recognised that movements in the Southern Ocean determine some of the most important events in the Indian weather ; and that it is claimed that the earlier discovery of the Lockyers' law would have enabled all the Indian famines of the last half century, to have been accurately foretold. Meteorology in Australia is far behind that of India, and will require to make up much ground before Aus- tralia is in the same position of vantage as India. The continent which gave meteorology that powerful agent of research, the Har- greave's kite, is the only continent which has not employed it for meteorological work, Professor Schuster, in a recent address to the British Association deplored the conservatism of meteorological work. He declared that meteorologists were enslaved to continuity, that the brilliant progress in the science during the past few years has been achieved, not by 86 THE CLIMATE OF AUSTRALASIA meteorologists at their Observatories, but in spite of them, by experimental work along fresh lines. If Schuster could make such complaints in Europe, one wonders in what terms he would express his opinions on the condition of meteorology in some other places. If Australian meteorology is behind that of India, it is not the fault of our meteorologists, who have done wonderfully well, with the means at their disposal. Their means have been inadequate. Federal Australia wants a meteorological service, which will adopt the same methods of observation, and the same system of publication for the whole of Australia and for New Zealand ; and which will be so organized, that the data collected will be fully and promptly used. We want a united meteo- rological service, working on a uniform plan and publishing uniform records ; and that service should have a sufficient staff and suffi- cient money to undertake experiments outside the ordinary routine of Observatory work. Such a service, to be efficient, must be as elastic, and free from red tape rules as a Govern- ment Department can be. Its officers must carry on their work, animated by a love of scientific research and not as a matter of business routine. The British Government largely relies on the Royal Society for advice in dealing with its scientific work. And in the case of the estab- lishment of a Federal Meteorological Service, the Australian Government might well follow CONCLUSION 87 the British precedent, and consult the Austra- lasian Association for the Advancement of Science, in reference to the organization of that work. The benefits such a meteorological service might confer on Australasia are incalculable. Proposals have been made to introduce into Central Australia a sheet of salt water, which, though large enough to be somewhat costly, would be small in comparison to the vast waterless plains it is proposed to benefit. But in the summer, when rain would do the most good, the country is often already covered with a vast sea of water. Day after day, in the summer of 1901-2 the districts around Lake Eyre lay under a heavy pall of morose, grey cloud. The fall of one tithe of that sea of moisture would have broken the long spell of drought, which had laid that country waste. The clouds at times descended as if endea- vouring to reach the earth ; but the ground was too warm and they were repelled again to the sky. More than once we had a few drops of rain, which showed that the clouds were so near the precipitating point that the slightest impulse would have upset the balance and brought down heavy rain. How high those clouds were above us, how thick they were, how much their temperature was above the precipitating point, we could not tell. No one knows. As I watched those clouds drifting steadily overhead, I used to long for a 88 THE CLIMATE OF AUSTRALASIA meteorological kite to sound that great sea of moisture ; and I dreamt of the time when kites would spray those clouds with liquid air, and discharge their now wasted contents on to the wasted plains below. CHAPTER II. — PROSPECTS OF LONG PERIOD WEATHER FORECASTS. Few investments promise Australia a higher return than meteorological research ; but to be successful that research must be conducted patiently, and on well considered lines ; it must sound the ocean of air that floats above us, and must watch, by the collection of water samples, the influential changes in the circulation of the seas around our shores. These studies are necessary for the solar changes, which act upon our weather, act indirectly, by variations in the temperature and currents of the oceans. The variations in solar radiation are probably insufficient to cause any appreciable direct change in the temperature of the surface waters. The changes in the sun probably act on the far more sensitive atmosphere, and the winds disturb the oceanic circulation and change the surface temperature of a wide expanse of ocean, and thus affect the temperature and rainfall of the adjacent lands. CONCLUSION 89 It is this fact which, though it renders the problem of long distance weather prediction very complex, gives us the best hope of ulti- mate success. If the changes that take place in the sun affected our weather at once, we should always be liable to disturbances that are extra-terrestrial in origin, and which at present we cannot reliably foretell. But the main effect of solar variations upon our weather is an indirect effect, and it works by a series of changes which may take years to run their course. The weather we shall have a year or two ahead is already determined ; it is ad- vancing upon us as silently and irresistibly as fate. Man may never be able to mould it an iota ; but he can watch its progress and be forewarned of its effects. Many men distrust the possibility of weather predictions for any sufficient distance ahead to be of any service to agriculturists ; but I have more faith in the future progress of science. We can see a little ahead already. We cannot see farther because of our ignorance ; and more knowledge would increase the penetration of our vision. Fortunately for Australasia, our meteorological conditions are far simpler than those of Europe, so that we may expect much greater certainty in weather predictions. I see no impossibility in future Australian meteorologists foretelling correctly a year ahead, the general nature of the approaching seasons. But such insight will never come to us until we have done our 90 THE CLIMATE OF AUSTRALASIA part, and studied the hydrography of the Southern Ocean, with the same methods which have yielded such profitable results in the North Atlantic, and to India. In meteorology each continent must work out its own salvation. Europe may help us with methods, but we must apply them our- selves, to our own waters, before we can share in the rewards. Patiently and excellently meteorologists all over Australia are recording the daily changes of our weather ; but far out in the Southern Ocean, the fundamental pro- cesses that are determining the nature of the seasons a year or two years ahead are passing unnoticed, and unknown. Australia has spent vast sums in irrigation works that have failed through lack of water, and undertakes the duty of recording the present weather, but, for the sake of some ^300 or ^500 a year, we are leaving unstudied the causes that produce and control it. We spend money freely in defence against hypothetical enemies who may never come ; while we stint the Intelligence Department, whose duty it is to watch the one foe whose onslaught is certain and merciless. What could add more to the commercial prosperity of Australasia than warnings to our vast agricultural and pastoral interests, whether next year our wide fields will be paralysed by drought or washed by a soil destroying deluge. The apparent fickle- ness and severity of our climatic changes CONCLUSION QI introduce as large an element of gambling into our farming as there is in many reckless mining ventures. The dragon of uncertainty that now preys on our agriculturists could be slain, if we had foreknowledge of approaching seasons of fair weather and of foul. That know- ledge is available, if we seek it properly. It is true that science holds out no hope of any quack formula that will tell us future weather without trouble or expense ; and it directly warns us that the forces that cause the major variations in our weather are too majestic for us ever to control. As, however, they work by processes that take years to run their course, we can watch the abnormal weather epochs in their slow and mighty progress, and science may herald their approach. Like the seer of old, science assures us, " cast thy bread upon the waters, for thou shalt find it after many days," though only if we devote those days to earnest and patient search. APPENDIX I. TEMPERATURES IN THE SOUTHERN OCEAN. 1. — The following are the records at some of the chief stations at which observations were made by "The Challenger." "Challenger" Expedition. Summary of Results : pt. 1., 1895 ; pp. 495-6, 505, 517-8, 523, 525-6. No. of Station Lat. S. Long E. Depths in fathoms. Temp'res. No. of Station Lat. S. Long E. Depths in fathoms. Temp'res. 153 65° 42' 79° 49' Surface 29-5° 157 53° 55' 108° 35' 400 . 32-6° ^n \29'0° 500 32-8° ou i , ,, i /29-0° V29'0° Bottom \32-0° j 32-2° 1UU J29-00 158 50° r 123° 4' Surface 45-0° 900 130-5° 50 44-5° £\J\J J30-50 100 43-4 Qftft, \32-0° 150 42 -3° oUU J32-00 200 41-3° ,-CAA ^32-8° 300 39-5° ouu /32-8° 400 38-2° Bottom \33'0° J33-00 500 600 37-0° 36-3° 157 53° 55' 108° 35' Surface 37-2° 700 36-0° 10 36.8° 800 36 -0D 20 36-8° 900 36-0° 30 36-6° 1000 36-0° 40 36 -6C 1100 36-0° 50 36-6° 1200 357' HO 36-6° 1300 35-3° 70 33-0° 1400 35-0° 80 32-5° 1500 34-7° 90 32-5° Bottom 33-5° 1 00 \32'7° 159 47° 25' 130° 22' Surface 51-5° 1 UU J32-80 50 50-0° 200 33-0° 100 49-0° 300 30-3° 150 48-1° THE CLIMATE OF AUSTRALASIA 93 No. of Station Lat. S. Long. E. Depths in fathoms. Temp'res. No. of Station Lat. S. Long. E. Depths in fathoms. Temp'res 159 47° 25' 130° 22' 200 47-4° 160 42° 42' 134° 10' Surface 55-0° 300 47'2° 50 51-8° 400 46-9° 100 48-5° 500 44-8° 150 47-9° 600 41-8° 200 47.70 700 39-4° 250 47-5° 800 38-3° 300 47'2° 900 37-9° 350 46-5° 1000 37-7° 400 44-9° 1100 37-5° 450 43-3° 1200 37-3° 500 41-6° 1300 37-1° 600 38-5° 1400 36-9° 700 37-0° 1500 36-7° 800 36-8° Bottom 34-5° 900 36-6° 1000 36-4° 1 Bottom 33-9° 2. LIST OF SOUNDINGS AND TEMPERATURE OBSERVA- TIONS IN THE SOUTHERN OCEAN HY THE "VALDIVIA." "The German Deep-Sea Expedition in Antarctic Waters." Geoy. Journ. Vol. xiii., 1899, p. 647. Lat. S. Long. E. Depth fathoms Bottom temp'ture. Lat. S. Long. E. Depth fathoms Bottom temp'ture 36° 23' 17° 38' 2280 33-2° F 56° 29' 7°25A 2758 37° 31' 17° 2' 2708 32-7° F 56° 16' 10° 53' 3018 31-2° F 40° 31' 15° 7' 1417 35-2° F 56° 30' 14° 29V 2784 • — 41° 5' 14° 52' 2860 33-2° F 55° 26' 18° 2' 2236 31-6° F 42° 18' 14° 1' 2512 32-7° F 54° 54' 22° 13X 2207 (32-8° F) 43° 52' 13° 6' 2962 32-70 F 54° 46' 26° 40' 2517 31 -5° F 46° 2' 11° 35' 2618 32-7° F 55° 27' 28° 59' 3025 31 -4° F 49° 8' 8° 41' 2415 32-7° F 56° 44' 32° 6' 3010 31'4°F 50° 57' 7° 40' 1960 58° 5' 35° 54' 3134* 31 -4° F 53° 31' 6° 14' 1240 33-0° F 59° 16' 40° 14' 2980 31 '5° F 54° 22' 4° 37' 1891 32-0° F 58° 53' 43° 1' 2965 — 54° -29' 3° 43' 310 59° r 47° 38' 3009 32-2° F 54 J 30' 3° 31' 240 33-8° F 60° 11' 49° 48' 3042 31 -6° F 53° 49' 3° 57' 1011 32-7° F 62° 27' 53° 22' 2829 31 '6° F 53 52' 4° 6' 1270 32-4° F 64° 9' 53° 12' 2540 31 -4° F 54U 29' 3° 30' 250 34-0° F 63° 17' 57° 51' 2534 31-2° F 55° 21' 5° 16' 1684 31'5°F 63J 32' 58° 40' 1504t — Deepest sounding in Antarctic area. t No bottom ; sounding incomplete. 94 TEMPERATURES IN THE SOUTHERN OCEAN Lat. S. Long. E. Depths fathoms. Bottom temp'tures. 61° 45' 61° 16' 1940 31 -8° F 58° 55' 64° 49' 2527 31 -6° F 56° 19' 66° 48' 1306 34-0° F 54° 33' 67° 52' 2690* 52° 48' 69° 13' 2145 32-5° F 51° 50' 69° 48' 1102 35-0° F 43° 45' 75° 34' 1878 34-5° F 41° 6' 76° 24' 1801 34 -5° F 38° 41' 77° 36' 86 55-0° F 38° 40' 77° 39' 367 49-8° F 37° 45' 77° 34' 800 37-8° F 37° 47' 77° 34' 271 51-0°F 3.— LIST OF SOUNDINGS MADE AND DEEP-SEA TEMPERATURES OBSERVED BY THE " GAUSS" BETWEEN CAPETOWN AND KERGUELEN. Lat. S. Long. E. Depth in fathoms. Bottom temp'ture F. 37° 41' 20° 47' 2570 34-2° 39° 42' 24° 49' 1676 36-0° 40° 0' 27° 13' 1706 36-0° 42° 30' 33° 45' 2780 36-9° ? 43° 4' 36° 22' 1980 34-4° 44° 18' 41° 48' 1342 35-8° 44° 41' 43° 54' 995 36-3° 46C 18' 48° 51' 1020 35-8° 46° 48' 50° 31' 1277 35-4° 46° 40' 51° 23' 149 — 47° 12' 58° 8' 2674 32-4° 47° 46' 61° 23' 2512 32-6° 47° 51' 66° 32' 210 35-4° No bottom ; sounding incomplete. CO g Q W P3 > O O %* & from Newspa which are ver Chief Office tain's repo ere marked Cap Sri £55 26 j" ^^ S £ So &fi s f II 6 s s ll 1 888 »:?. .? GOt— t^t^l>. G ,§ H3 ",0 J? 1 1 | 1 •^ _z •*"C !§ 1 1 1 ll CO 11 «j W «3 C C 1 I ! S I * COCO CO CO CO cr^ PQ APPENDIX III. " The forecast of the probable character of the rainfall during the south-west monsoon of 1898 was issued in the same form as in the preceding seven years in the Gazette of India of the 4th of June, 1898, under the heading of 'Memorandum on the Snowfall in the Mining Districts bordering Northern India, and the abnormal features of the weather in India during the past year, with a forecast of the probable character of the south-west monsoon rains of 1898.' " The following gives a statement of the forecast as published in the Memorandum: — "The consideration of the cyclical period through which India has passed indicates that the year 1897 was probably the last year of what may be termed the ' negative phase.' A period of fairly normal conditions is now probable for some time in the Indian monsoon and south-east trades regions. . . " Taking all the facts into consideration, it is probable the monsoon currents will be at least of normal strength, but that there may be, in consequence of the abnormal snowfall (moderate however) in the Punjab Himalayas during the month of May, slight delay in the establishment of the Bombay current. " Judging only from the conditions in India itself and the known conditions in the Indian Seas, it is, on the whole, very probable that the monsoon currents will be of normal strength, and probably they will be somewhat stronger than usual. The Bombay current is more likely to be above its normal strength than the Bay current ; bnt the probabilities for this are small, not exceeding 5 to 2. "Assuming that the currents will be of normal strength, the comparison with previous years (more especially 1883, 1888, 1890, and 1892) indicates that it is probable they will set in about the normal time on both the Bengal and Bombay coasts. The Seychelles observations are also in favour of this inference. It is possible that from the action stated above, the Bombay current may be slightly retarded, and there is a slight probability that it may be not so strong as usual in June. . . . " The conditions in the Indian seas and the Indian Ocean are, so far as can be ascertained, satisfactory and favourable, and indicate that the conditions in the south-east trades regions are at least normal, and that the air movement in that area is somewhat stronger than usual. "Conditions are favourable to the prevalence of monsoon currents of at least normal strength in the Bay of Bengal. The rains will probably commence about the normal date in Bengal. " Conditions are also, on the whole, favourable to the prevalence of monsoon currents of at least normal strength in the Arabian Sea. The abnormal snowfall in the Punjab Himalayas may slightly retard the establishment of the monsoon on the Bombay coast, and cause it to be slightly warmer than the normal in June. The influence of the snowfall will very probably be slight, and, so far as can be judged, the monsoon ought to set in on the Bombay coast before the 7th of June. " It should be carefully noted that the preceding probabilities are obtained on the assumption that the currents will be nor: in strength or slightly stronger than usual, and that . require to be considerably modified if the monsoon "' cu should be much stronger or much warmer than usual.1' ° GENERAL LIBRARY UNIVERSITY OF CALIFORNIA— BERKELEY RETURN TO DESK FROM WHICH BORROWED This book is due on the last date stamped below, or on the date to which renewed. Renewed books are subject to immediate recall. 14Hov'55VL| LD 21-100m-l,'54(1887sl6)476 QC99Z