Sicily. In Sicily the time of sowing of wheat is very variable depending on the autumn rains, but sowing is possible from October to January, the time of earing varies from April 12 to May 5. Flowering takes place five to ten days after earing. The average date of harvest is about the first week in June but may vary from May 26 to July 5 according to the height above sea-level. More Northern Countries. In the South and East of England our own arrangement of the seasons would be approximately : March April spring sowing September root-harvest and tree-fruit harvest The seasons are still later further North; for the North of Scotland we might find an almost continuous crop of grass, and a harvest of oats, distributed over the months from September to November. The intermediate position between the Northern experience with regard to crops and seasons as compared with the Southern which we have been considering, is well expressed by the new kalendar devised by the National Convention of the first French Republic in 1793. Seasonal names were given to the 12 groups of 30 days each between September 22 and the next following September 16, as follows: The remaining five or six days were devoted to holiday celebrations. So in different parts of the world different climates have different associations of seasons, the details of which might form an introductory section of any comprehensive treatise on the application of meteorology to agriculture. Here we are only concerned with the subject as helping us to make out the times and seasons of ancient climates. THE PRESENT POSITION OF THE KALENDAR As time passes and statistics are compiled in a systematic manner the defects of our kalendar become more and more apparent, and this is the case especially with the compilations of meteorological data, which are now available for 50 years or more at many stations and for as long as 200 years in some. We are accustomed to group the data according to the unequal kalendar months. For their "Dekadenbericht" the Deutsche Seewarte divided the months into 10 days, 10 days and 8 days, 9 days, 10 days or 11 days. For a number of years the British Meteorological Council published the hourly values of the records at their observatories in the form of five-day means but their most effective contribution to meteorological statistics, the Weekly Weather Report, which runs from 1878 to the present time, is based on the year and the week. All the weeks are of equal length and the year consists occasionally of 53 weeks and its beginning ranges from December 29 to January 3. So far as meteorology is concerned the problem seems to compel us to make up our minds whether it is reasonable to continue to work with 12 months which are of unequal lengths in any case, even if the worship of the Emperor Augustus be now disregarded and all months made of 30 or 31 days. Five of the months are indivisible into two or three equal parts. They give us no practical clue to the dates of new and full moons, which are really pertinent as regards tides, so that it is really seeking trouble to call them months. On the other hand a year could be made up of 52 weeks with an additional or fifty-third spell of a single day or of two days in leap year. No doubt there are some discomforts or awkwardnesses; quarters of the year would be 13 weeks, not three months, and if groups of four weeks were used, the two solstices and two equinoxes would be unsymmetrically placed with regard to them. Thirteen weeks are awkward in harmonic analysis. The rational conclusion seems to be that the grouping of meteorological data for statistical purposes to form the basis of comparison with the statistical data of agriculture, trade, commerce and public health for each year should be arranged in the form most convenient for the effective representation of the material irrespective of the minor details of the kalendar. From this point of view the week is quite definitely the most suitable unit intermediate between the day and the year, both of which are fixed by causes beyond human control. Meteorological statistics grouped in months are of little use in reference to agriculture, trade, commerce or public health, even if the months be of equal length, for the simple reason that a month is too long a period; it is not generally possible to characterise effectively the weather of 28 days but a period of seven days is more naturally susceptible of definition and is indicated for the statistics of trade and commerce which in the present organisation of society and religion have inevitably a weekly period. Medicine seems to have special associations with the month, the use of the words lunacy and lunatic implies as much, yet the statistics of public health are already based generally on the week. The Meteorological Section of the International Union of Geodesy and Geophysics have given expression to these ideas in a resolution of a meeting at Madrid in 1924, in favour of the mean solar day (with its subdivisions), the week and the kalendar year as suitable units of time for meteorological statistics1. 1 Journal of the Royal Statistical Society, 1925. THE QUARTERS OF THE MAY YEAR The question may perhaps be raised whether the division into as many as twelve intervals is really necessary for a general climatic summary and whether a group into four quarters with the possibility of more detailed expression for weeks or fortnights would not do instead. By way of examining this question we may note that with our present practice of using a solstitial year, that is to say a year beginning with the winter solstice, or nearly so, each group of three months, January to March, April to June, July to September and October to December, covers the whole interval between a solstice and an equinox; at those two extremes we have the most stationary condition and the most variable condition respectively, so each quarter includes half a stationary period and half a period of change; if, however, we could revert to the prehistoric practice of the May year, and group our figures for each of the 13 weeks, May 6 to August 8, August 8 to November 8, November 8 to February 4, and February 4 to May 6, we should get the data for the steady solstitial periods separated from those of the variable equinoctial periods. Counting in weeks the periods would be: Spring equinoctial period 13 weeks (6th to 18th) Feb. 4-May 5. The grouping of the days of the year in the manner thus indicated isolates the periods of thirteen weeks of greatest and least income of solar energy between two equal periods of rapid transition. Regarded from the point of view of solar influence the difference between the two periods thus isolated is remarkable. In illustration we give the results of observation of solar radiation at South Kensington, a metropolitan station, and at Rothamsted, a rural station, in the year 1924. The figures represent the average daily income of thermal energy from sun and sky expressed in kilowatt-hours received on a square dekametre of horizontal surface, about 1000 square feet. Since the electric energy which we buy is commonly counted in kilowatt-hours and would be cheap at id. per unit, we can assign a cash figure to the income of solar energy at that price. Daily average income of solar energy per square dekametre during the We have already noticed in the Scottish quarter days and certain English customs the survival of this interesting division of the year. It seems to be deeply ingrained in English rural life, for in a volume of addresses by the late Dr James Gow, Headmaster of Westminster School, we read that "on most of the commons of England the Lord of the Manor has the right of pasture and the right of the hay from Candlemas (Feb. 2) to Lammas (Aug. 1) and the tenants have the right of common from August to Candlemas." We give a tabular representation of the climate of Babylon on this plan. It contains only one-third of the number of data included in the monthly table of p. 30 but it gives a fair idea of the nature of the climate. CHALDAEA (IRAQ). Lat. 32° 30′ N. Long. 44° 20′ E. Alt. about 30 m. BABYLON. Seasonal Period 1911- 5 years, 1907-1912 PRECIPITATION Number of days extremes mm mm mm uy mm mm 287 321 288 313 295 May n 8 O n July CHAPTER IV POETS AND HISTORIANS: THE APPLICATIONS OF METEOROLOGY TO AGRICULTURE AND NAVIGATION. HERODOTUS THUS, having summarised the experiences which are characteristic of the Mediterranean climates at the present time and having indicated the natural associations of the cycle of the seasons with the information to be gathered from the face of the sky, we may now revert to the co-ordination of the experiences of weather to which reference has been made in the first chapter and the different ways in which the co-ordination was expressed by the poets on the one hand and by the philosophers on the other. It has been suggested that the experiences of weather led the ancients to regard the forces of nature as being under personal control and that they were in that way the origin of a religion, mythological or theological, which might regard an endeavour to trace the sequence of weather to natural causes, the scientific inquiry of the philosopher, as being an improper or even impious curiosity about an almighty power. But due submission to the divine will was not regarded as inconsistent with such anticipation of the future providence as a careful collation of past experience would indicate; so we find, together with exhortations to an implicit trust in providence, the beginnings of a weather-lore, or signs of coming weather and seasons, many of which are observed even to this day, and in some cases the form which the ancients gave still survives. This attitude is expressed by the Greek and Hebrew poets. We shall quote a number of extracts to illustrate that statement. As already indicated in this connexion we have nothing to say about Egypt because the Egyptians were dependent only on the rising of the Nile, and raised the structure of their mythology upon the sun and his risings and settings with which the changes in the Nile were so closely associated as to need no assistance from the signs of rain or storm1. The moon was probably worshipped before the sun, being connected with the animal gods, the baboon and the ibis of Tehuti, the god of lunar measure, of wise 1 Sir Norman Lockyer, The Dawn of Astronomy, Cassell and Co., Ltd., London, Paris and Melbourne, 1894, chap. XXIII. |