philosophical discussion were active, when the facts were wholly inadequate and records hardly existed; to-day we have a multitude of facts and records, though even now they are insufficient, but philosophers are less disposed to regard their exposition as part of their ordinary duty than they were before the introduction of the weather-map. The place of forecasting in modern meteorology This most recent development leads us to remark upon another misconception in the public mind with regard to meteorology, and that is that meteorology means forecasting the weather and nothing else. It is a great misfortune for the subject. A curious sense of depression for some meteorologists arises from the question which their friends, without exception, ask on the mention of any new scientific advance in meteorology: "Will it be useful in forecasting?" It is hardly an exaggeration to say that meteorologists have a natural aversion from the iteration of the duty of forecasting. No student of physical science in any of the recognised institutes, however sure he might feel about the result of a scientific calculation, with the exception of such well-tried predictions as those of the Nautical Almanac, would dream of announcing it in the public press without waiting for its verification, if waiting were the only avenue to verification. This feeling of premature disclosure is deep-seated in the scientific student. When Admiral FitzRoy, overpowered by the glamour of a telegraphic synoptic map, published his anticipations of coming weather as forecasts in the newspapers, his colleagues of the Royal Society were shocked and gave expression to their feelings; so much so that, after FitzRoy's death in 1865, when the Royal Society took over the administration of his office, no forecasts were allowed. In 1879 the controlling body, with full scientific authority, had arrived at the conclusion that the primary problems were solved, and forecasts have been issued day by day since that date. But the problems were not really solved with the degree of precision necessary for unfailing accuracy. It is a fair question whether forecasts would not really have been better to-day if meteorologists had not been compelled to issue a series of 50,000 sets of forecasts, only more or less correct. They might have given their attention to more purely scientific aspects, with some assurance of useful results, because the scientific aspect is the aspect from which true knowledge is derived. But any such question is chose jugée from the first. The universal desire for information about future weather opens the main artery of communication between the science and the public and is the chief vindication of an appeal for public funds; and when once a scientific subject enjoys public money it is difficult to persuade anybody that it is not provided for in all particulars. Public money is not always an unmixed blessing; science sometimes prospers in what appears to the outsider as poverty or adversity. In the meantime forecasting is an invaluable touchstone, or test, of the reality of the scientific conclusions that are reached in the study of the atmosphere. If the general conclusions are correct, the inferences fairly drawn from them in particular cases must necessarily be correct too, and the comparison between inference and fact is an indispensable part of scientific reasoning which depends partly upon inductive principles and partly upon deductive calculations. It is from that point of view that we propose to treat the subject in this book. We shall survey the methods and results of observation, we shall set out the inductive principles which have been established by their aid, and illustrate the application of physical and dynamical reasoning, with the understanding that the final test of the conclusions is their agreement with the facts of observation. THE EARLIEST RECORDS Reverting after this digression to the main purpose of this chapter, we note that the civilisations from which our own culture is derived include those of Egypt, Crete, Babylonia, Phoenicia, Chaldaea, Palestine and Greece; also that for satisfactory evidence of the dawn of meteorology we must not look to those civilisations which were independent of weather. We naturally look to the documents that remain of the civilisations which took their sustenance from the sky for the earliest indications of the nature and progress of meteorological study. The indications which have survived are only fragmentary until we come to the time of Aristotle, who wrote the first formal treatise with the title Meteorologica; but such fragments as there are, are not without interest. There are still a considerable number of popular superstitions about the weather. In a lecture before the Royal Meteorological Society in 1908 on The Dawn of Meteorology1, Dr G. Hellmann of Berlin traced some of them back to very early times. For example, the widespread belief in the first twelve days of the year as prophetic of the weather of the ensuing twelve months is traceable in the old Indian or Vedic texts, and the signa tonitrui, the forecasts of weather and fertility deduced from the thunder heard in each of the twelve months, contained in mediaeval almanacs, though attributed to the Persian prophet Zoroaster, are in fact of Chaldaic or Babylonian origin and date back to some thousands of years before the Christian era. From the astrological cuneiform library of Assurbanipal, now in the British Museum, we learn that predictions authorised with the forecaster's name were addressed to the king; for example: "When it thunders in the day of the moon's disappearance, the crops will prosper and the market will be steady. From Ašaridu.” The Babylonians had the wind-rose of eight rhumbs, and used already the names of the four cardinal points to denominate the intermediate directions; whereas it was till now generally supposed that we owe to Charles the Great, or perhaps to his learned monk Alcuin, who came from Yorkshire, this progress of the combination of the four principal winds to denote all others. That was indeed a great advance, for it is well-known that in the Greek and Roman periods each wind had its peculiar name, a practice still in use amongst the Italian mariners in the Mediterranean. 1 Q. J. Roy. Met. Soc. vol. XXXIV, 1908, p. 221, revised and extended in Met. Z. 1908, Bd. xxv, p. 482. The state of meteorology in the old Babylonian culture, namely, three to one thousand years B.C., shows quite another character than it did in those primeval times in which the weather proverbs originated. After having been formed into the beginnings of a learned profession by the priests, the atmospheric phenomena were brought by them into connexion with the constellations of the heavenly bodies, and a complete system of consequences and combinations was established, which gave rise to the Astro-Meteorology. It even formed an integral part of the Assyric-Babylonian religion. From the time of Meton, the fifth century B.C., the Greeks made regular meteorological observations, which were perhaps the first, as illustrated by the so-called parapegma, a kind of peg-almanac fixed on public columns for exhibiting the general data of the weather. In these parapegmata the observations of the wind prevail over all others, for they were of practical use in navigation. THE DIVERSITY OF CLIMATES Much more extensive information is available in the literature of Palestine and Greece, and to these we shall refer later, but here we note that the countries from which we seek the earliest information of the origins of meteorology are all related to a narrow strip of the earth's surface lying between the Atlantic Ocean and what we now call India, and including the shores of the Mediterranean Sea. That happens to be one of the most curious regions in the whole world from the point of view of weather and climate. It includes a number of arid deserts where no cultivation is possible, it includes also countries, Egypt, Babylonia, and India which, though arid, are fertile because they are irrigated by rivers, the Nile, the Euphrates, the Tigris, and the Indus. These are subject to very considerable rise and fall in the course of the year and provide in that way the seasonal variation which agriculture requires. On the other hand, the shores of the Mediterranean and the islands, although there is a very marked seasonal variation, are dependent, as we are, upon rainfall for the growth of crops. We cannot expect to comprehend any of the references to the meteorological conditions which are found in the ancient records without bearing in mind these essential differences of climate. We therefore devote a chapter to the climates of the Indo-Mediterranean Region, which is often referred to as "the world as known to the ancients." THE WORLD AS KNOWN TO THE ANCIENTS Fig. 1. A map marking the positions of stations for which climatic summaries are given on pp. 29-40, to illustrate the references to weather and climate in the works of Aristotle and other classical writers. CHAPTER II WEATHER AND CLIMATE IN THE "WORLD AS KNOWN TO THE ANCIENTS" Intensity may be indicated by attaching "exponents" o or 2 to the symbols, thus ✶ means light snow, heavy snow. No exponents other than o or 2 should be used with the international symbols. The use of b when the sky is one quarter clouded is authorised. Detailed instructions by the Director of the Meteorological Office for the use of b and c came into operation on January 1, 1926. The letter n is used in this book for number, to head columns which enumerate occurrences of observed phenomena in a specified period, week, month or year. THE World "known to the ancients," fig. 1, extends, as has been said, from the valley of the Indus on the east side to the Pillars of Hercules at the Strait of Gibraltar on the west, or perhaps beyond them to the Canary Islands; and from the tropic of Cancer on the south to Scythia or Southern Russia and Central Europe on the north. The tropic of Cancer runs across Africa and Arabia from the west coast, south of the Canary Islands, to the mouth of the Indus, crossing the Nile at the first cataract, the Red Sea between Mecca and Medina, and just missing the Persian Gulf. Aristotle considered that the region to the south beyond the tropic was not habitable because it was too hot and that the north, beyond the home of the Barbarians or Scythians, was equally uninhabitable because it was too cold. The northern limit he thought might be in the countries underneath the stars that never set. Taking the latitude of Athens as 38°, the stars that never go below the horizon would |