Mean Height of the Barometer. The unequal and irregular atmospherical pressures between Port Jackson and Port Philip are apparent throughout their respective registers. The discrepancy of that pressure between Woolnorth and Circular Head are not less so, notwithstanding that these localities are situated on the same latitude of the northern littoral of the island, and separated by only twenty minutes of longitude.

Reduced, however, to winter and summer means, as seen above in Table IV., the height of the barometer in each station, discovers a rate which is constant and regular at the return of each season; and when the means of seasons are reduced to the last term of averages, as in Table V., it shows farther, that, within the extent of 12° of latitude, the locality of the maximum of atmospheric pressure, during summer and winter, is Woolnorth, from whence, to the north and south, the pressure decreases. In that decrease, Circular Head is almost isobarometrical with Port Philip, as Port Arthur is with Port Jackson.

Barometrical Oscillations. Of these, the monthly maximum and minimum coincide in data throughout New South Wales and Van Diemen's Land, with some remarkable exceptions, which will be noticed hereafter. The differences which they give in the means of seasons are, at each station, uniformly greater in winter than in summer; at both seasons, nevertheless, as shown by Table V., the amplitude of oscillation uniformly diminishes from Port Arthur, the extreme south, to Port Jackson, the extreme north, of the space embraced by the meteorological survey.

This decrease in the oscillation from the pole to the equator, which, according to the Australian observation, is, in every degree of latitude, equal in summer to 0287 of an inch, and in winter to 0223, is a fact so generally applicable to every part of the globe where the barometer has been recorded, that it

becomes, as Saussure has justly said, a phenomenon that must be accounted for by every hypothesis which pretends satisfactorily to explain the barometrical oscillations.

Before venturing on the analysis of the results which are derived from the registers of New South Wales and Van Diemen's Land, it is most important to review, in a few lines, the progress which has been made in Europe in the study of atmospheric pressures, a study which as yet, for lack of the needed observations, presents more problems than solutions.

Lamanon, the naturalist of the ill-fated expedition of La Peyrouse, was the first who detected a regular variation in the daily range of the barometer, embracing two maxima and two minima, in twenty-four hours; and Humboldt, whose name is associated with every branch of terrestrial physics, made it a subject of special inquiry. With him, indeed, may be said to have commenced, in all parts of the world, a series of the most elaborate observations, which have led to the discovery, that the differences of this variation are greater at the equator than they are in the higher latitudes; and that the hours at which the changes happen vary according to the altitude at which they are observed.

De Buch, again, extended the study of the diurnal variation of the barometer to that of its monthly oscillation; and came to the conclusion that, in the entropical regions, the height of the barometer decreases as the sun approaches to the zenith. Dove extended the inquiry still further, and proved, by accumulated observations, that its rise and fall are influenced by seasons.

These discoveries rendered the existence of some laws evident, and gave rise to most ingenious speculations as to the causes on which they may depend.

Professor Daniell, in his valuable meteorological

essay, confessing all the difficulties with which the question of the barometrical fluctuation is beset, ventured nevertheless an opinion upon its causes; which opinion, from the fact of its never losing sight of the decrease of the barometrical oscillation from the poles to the equator, deserves the greatest attention of meteorologists.

The hypothesis of Professor Daniell rests upon the assumption of an atmospheric circulation between the poles and the equator, in which circulation "the cold dense air of the former region flows in a lower current to the latter; while the elastic air of the latter is returned in an upper current to the former;" and farther, upon a change in the thermometrical state of the two currents, which must naturally effect an unequal distribution of the ponderable matter, and consequently an unequal distribution of atmospheric pressure, at all the intermediate stations between the equator and the poles; the changes are throughout traced and analysed, and identified with the theory, which, as far as the scanty numerical elements relating to the subject allow us to infer, is perfectly accordant with actual observation.

Kämtz, again, following up the march of the barometer and thermometer, in different localities, was struck with the fact, that whenever one of the instruments was observed to rise, the other showed a fall. The examination and comparison of the observations of pressure and temperature made at Bagdad, Buda, Cambridge (Massachusetts), Eyaford (Iceland), and St. Fé de Bagota, localities most different in climate, and most distant in position, did but confirm him in the opinion that the barometer and thermometer assume an inverse ratio in their respective oscillations. Furthermore, the experiment with Leslie's differential thermometer, and the consideration and analysis of the effects which a differ

ence of temperature between two distant localities would produce both on the thermometer and the barometer, led him to this remarkable conclusion, that "whenever the barometer falls in a country, the temperature of that country is higher than that of the countries which are in its vicinity; and, on the contrary, when the barometer rises, that country is colder than those by which it is surrounded. *

But Dove, who discovered a connection between the hygrometrical condition of the air and the barometer, referred the variation of the atmospheric pressure to it. His disquisition on the observations of Neuber in Andrade, detailed in the Poggendorff's Annalen der Phisick, and his final reduction of the tension of vapours to the hourly pressure of a dry atmosphere, bear, like his other labours, contained in Dove's Repertorium, the stamp of unusual research and unwearying zeal in the discovery of truth.

Two more hypotheses relating to this interesting subject deserve to be recorded here.

The first is that of Bouger, who deduced the variation of pressure from the calorific effects of solar

rays.

The last is the influence of winds upon pressure.

Both hypotheses stand the test of observations, and the last in particular is singularly adapted to extend our knowledge of the nature of barometrical oscillation in Terra Australis.

This influence of winds was detected by Lambert, as far back as 1771. He was the first who suggested the connecting the observation of each wind with that of the barometer. The suggestion was not lost upon the lovers of science. Burckhardt, Ramond, De Buch, Bueck, Dove, Eisenlohr, Kupffer, Schouw,

Vorlesungen über die Meteorologie. Lehrbuch der Meteorologie.

and, lastly, but not least, the enlightened meteorologist Kämtz, directed their attention to the subject, and the results of more than forty years of observations have been such, that, at the localities where they were made, we may infer, with the greatest approximation to the truth, either the height of the barometer from the existing wind, or the quarter the wind is in from the atmospheric pressure.

Other results, not less important, have been obtained. The observations made in London, Middleburg, and Hamburg, have shown uniformly that the barometer rises during the (N. + N. E. + N. W.) polar wind, and that it falls in the (S. + S. E. + S. W.) equatorial one.

This conquest in meteorology was achieved only by the extreme perseverance with which the obser vers accumulated the numerical elements, both as regards the wind and the barometer.

In the Australian colonies these elements are as yet too scanty, too imperfect, to yield a result similar to that obtained in Europe; but, trifling as they are, they nevertheless tend to show that the atmospheric pressure in New South Wales and Van Diemen's Land is influenced by winds, and in the same way as it is in Europe and North America.

In order, however, to detect and appreciate the dependence of one phenomenon upon the other, we must deal, not with the prevailing winds, as illustrated in Table II., but with the great circuit of atmospheric circulation which surrounds the Australian continent, and of which the winds of the five stations are mere off-shoots or eddies.

Taking, then, the diagram Pl. II. fig. 2. of the three winters, we see that the northerly segment of the exterior circulation, proceeding from east to west, had less influence upon the motion of the represented eddy; and, consequently, that the prevailing winter winds in