Instrumental weather observations did not become the rule, however, until the end of the seventeenth century, when the use of thermometers, hygrometers, barometers, and rain gauges began in Italy and spread rapidly to other countries. The origin of each of these instruments is commonly ascribed to a particular inventor--the thermometer to Galileo, the barometer to Torricelli, etc.--but the truth is that the idea of the instrument was, in each case, a slow growth, to which many minds contributed. Thus a form of thermoscope--a device for showing but not for measuring the expansion and contraction of air with changes of temperature--was described by Philo of Byzantium in the third century B. C. Galileo supplied such an instrument with a scale, but without fixed points, thus converting it into a crude thermometer, but it was not until half a century later that the Grand Duke Ferdinand II of Tuscany introduced the idea of filling the thermometer with alcohol, in place of air, and sealing it so that it was not affected by changes in barometric pressure. The thermometric scale now used in English-speaking countries, which bears the name of Fahrenheit, appears to have been devised by the Danish astronomer Ole Römer, from whom Fahrenheit borrowed it. In short, any _brief_ account of the invention of the principal meteorological instruments necessarily ignores the just claims of many inventors; to say nothing of the fact that what is written on the subject to-day is likely to be refuted to-morrow by the discovery of some forgotten book or manuscript. We are on safer ground in saying that the plan of _measuring_ the weather, instead of merely observing it, became general early in the eighteenth century; and that about the middle of the nineteenth century the further improvement was introduced of making meteorological instruments trace their own records, so that the human observer was, to a great extent, dispensed with. Self-registering instruments are now the rule at important meteorological observatories and stations, though they do not, even yet, record all the elements of weather, and at a host of minor stations none of them have yet replaced the eye of the observer. Now let us see what things go to make up the weather, and how these things are observed by the modern meteorologist. The pressure of the atmosphere, if not exactly a part of the weather, is so intimately associated with it that we cannot exclude it from our list of weather phenomena. Atmospheric pressure is measured with the _barometer_, and the importance of this instrument as a key to weather changes is fully recognized--and indeed overrated--by the layman, who sometimes calls it the “weather glass.” [Illustration: MERCURIAL BAROMETER (Fortin type)]