Popular Science Monthly/Volume 43/July 1893/Recent Science II
By PRINCE KROPOTKIN.
DURING the last thirty years the data of meteorology have been accumulated with a very great rapidity, and the chief desideratum of the moment is, to construct with these data such a general theory of the circulation of the atmosphere as would embody the distribution of heat, pressure, moisture, and winds over the surface of the earth, and represent them as consequences of well-established mechanical laws. The old provisory hypothesis of atmospheric circulation, advocated by Hadley in 1735, and further elaborated by Dove in our century, can be held no more, and a new theory has become of absolute necessity.
We all have learned Dove's theory at school, even though we often found it difficult to understand. The air, greatly heated on or near the equator, rises in the same way as it rises in the summer over a sunburned plain. On reaching the higher strata of the atmosphere it flows toward the poles, but, owing to the speed of rotation which it has acquired in the lower latitudes, it is deflected — to consider the northern hemisphere only — to the right, and blows in the upper strata as a current from the southwest. To compensate this flow, air rushes on the earth's surface toward the equator, and as it also is deflected from its course by the same inertia of rotation, it appears in the tropics as a trade wind blowing from the northeast. However, the upper warm current does not flow all the way to the pole in the upper regions; it is gradually cooled down, and in about the thirtieth degree of latitude it begins to descend to the earth's surface, where it meets with the cold polar current. A struggle between the two winds ensues, and it lasts until they make a temporary peace by blowing side by side, or one above the other, the struggle giving origin to storms and to changes of wind which are fully analyzed in Dove's theory. A rope without end rolling over two pulleys, one of which lies horizontally near the equator, and the other stands upright in 392 THE POPULAR SCIENCE MONTHLY.
higher latitudes — such was the simplest expression of Dove's the- ory given in text-books.*
Under this provisory hypothesis meteorology made an im- mense progress, and some five-and-thirty years ago, Loverrier in France, and Fitzroy in England, ventured for the first time to foretell weather twenty-four hours in advance, or at least to send out warnings as to the coming storms. This bold step brought meteorologists face to face with a quite new problem. From the air pressure, the temperature, the moisture, and the winds ob- served at a certain hour of the day at various spots and tele- graphed to a central station, they had to infer the next probable state of weather. So, leaving aside the great problems of atmos- pheric circulation, they directed their attention to the changes of weather rather than to the causes of the changes. f For this pur- pose purely empirical laws were of great value. "When the me- teorologist saw on a weather chart a region of low atmospheric pressure, with winds blowing in spirals round and toward its center, he named it, by analogy with real cyclones, a " cyclonic disturbance or a " cyclone," giving the name of " anticyclone " to the region of high atmospheric pressure — and he studied the tracks of both disturbances in their advance across the oceans and the continents. He did not inquire for the moment into the causes of the disturbances ; he took them as facts, and, following Buys Ballot's law, he said that the wind will blow as a rule from the region of high barometic pressure (the anticyclone) to the region of low pressure (the cyclone), with a certain deflection to the right or to the left. Immense researches were made to study the routes followed by the centers of barometrical minima, and we now have splendid atlases showing the normal tracks of cyclones across the Atlantic Ocean, over Europe and the States, in Japan, in the In- dian Ocean, and so on, at various seasons of the year. J With these empirical data meteorologists attained such a perfection in their weather forecasts that in five cases out of six their previsions are now correct, while the coming gales are even foretold with a still greater accuracy.
��*E. E. Schmid, Lchrbucli der Meteorologie, Leipsic, 1860, p. 568.
f Sec AV. Bezold's short sketch of lucteoroloffical progress in Sitzungsberichte der Ber- liner Akademie der Wissenschaften, 1890, ii, 1295, sq.
X Besides the earlier works of Ley (Laws of the Winds prevailing in Western Europe, Part I, 1872) and Ktippen (Wissenschaftliche Ergebnisse aus der monatlichen Uebcrsichten des Wetters, IS'TS-'^S), we have now the splendid work of W. J. Van Bebber, which em- bodies the tracks of all cyclones in Europe for the last fifteen years (Die Zugstrassen der barometrischen Minima, fur 1875-'90), the researches of Blanford, S. E. Hill, and Elliot in the Indian Meteorological Meirioirs and Cyclone Memoirs, Part IV (published by the Mete- orological Department of India), the work of E. Knipping lor Japan, in Annual Meteoro- logical Report for 1890, Part II, Appendix, and several excellent works for Russia.
�� � �RECENT SCIENCE. 393
However, the very progress achieved demonstrated the neces- sity of a more thorough knowledge of the too much neglected up- per currents of the atmosphere. In Dove's scheme, the upper equatorial current, after part of it had been sent back to the equa- tor, was entirely abandoned to itself, to make its way as best it could against the opposed polar winds ; but the existence of a strong, nearly permanent, and relatively warm upper wind blow- ing toward the east in our latitudes — which was only probable thirty years ago* — became more and more evident, especially since the movements of clouds began to be systematically studied and observatories were erected on high mountains ; and this wind remained unexplained in Dove's theory, while in Maury's scheme of atmospheric circulation, which is still in great vogue in our schools, there was eA^en substituted for it a current in an opposite direction, which does not exist, and which Maury himself could not account for.f An entire revision of the subject was thus ne- cessary, and this revision has been done by the American meteor- ologist Ferrel, in a series of elaborate works which are only now beginning to receive from meteorologists the attention they fully deserve.
Ferrel's theory is based upon considerations as to the laws of motion of liquids and gases of different densities. If the whole atmosphere were equally heated in all its parts, and at full rest, the air would be disposed in horizontal layers, of greater density at the bottom, and of decreasing density toward the top. Consid- ering some part only of the atmosphere, from j)ole to equator, and neglecting the curved surface of the earth, we should thus have something analogous to a trough filled with layers of different liquids. If one end of the trough were now warmed, and the other end were cooled, the layers would be horizontal no more.
- Observations in Siberia — namely, at the graphite works on Mount Alibert, at a height
of eight thousand feet (52° north latitude) — were especially conclusive. Alibert's observa- tions, buried in the Russian Trudy of the Siberian expedition, proved the existence of a nearly permanent west and west-northwest wind on the top of the peak, and they showed at the same time that the average yearly temperature on the top of the peak was by some fourteen to eighteen Fahrenheit degrees higher than it otherwise ought to be. When I visited the then abandcmed mine in 1864, and saw the peak dominating all surrounding mountains, and could judge of the force of the west wind from the immense works accom- plished to protect the road which was traced on the western side of the peak, I could not refrain from explaining the extraordinary great height of the snow-line in east Siberia by the existence of a relatively warm equatorial current blowing with a great foi-ce at a height of from eight to ten thousand feet in the latitude of 52° north. Later on the observations which I brought from the Voznesensk mine (60° north, altitude twenty-six hundred and twenty feet) induced my friend Ferd. Miiller, who calculated those observations, to conclude that in higher latitudes the same current descends still lower to the earth's surface, and still maintains some of its initial warmth.
f Sec James Thomson's paper On the Grand Currents of the Atmosphere, in Philosoph- ical Transactions, A. 1892, p. 071.
�� � 39+ THE POPULAR SCIENCE MONTHLY.
They "would be inclined, but in two different ways : the lower ones would be inclined toward the warm part, while in the upper layers the inclination would be the reverse. A full circuit of the lighter liquids flowing one way on the surface, and of heavier liquids flowing the other way on the bottom, would thus be established. The same would happen in our atmosphere with the lighter warm currents and the heavier cold currents if the earth had no rota- tion on its axis. But it rotates — the solid globe as well as its gaseous envelope — and this modifies the whole circulation. The air which flows from the equator to the poles maintains, not its velocity of rotation, as has been hitherto taught, but its energy of rotation, which means that it obeys the law of preservation of areas ; therefore, when it is transported from the equator to a higher latitude it is endowed (in the northern hemisphere) with a much greater easterly velocity than if it simply maintained its speed of rotation. On the other side, the air which is flowing from the higlier latitudes toward the equator also obeys the same law and acquires a westward velocity, but much smaller than the eastward velocity of the former; this is why the west winds have such a preponderance in our latitudes.* Moreover, in virtue of the centrifugal force, all masses of air moving in any direction — not only north or south, but also due west or east — are also de- flected to the right in the northern hemisphere, and to the left in the southern hemisphere, f Consequently the air flows in great spirals toward the poles, both in the upper strata of the atmosphere and on the earth's surface beyond the thirtieth degree of latitude ; while the return current blows at nearly right angles to the above spirals, in the middle strata as also on the earth's surface, in a zone comprised between the parallels 30° north and 30° south. J
Such are, very briefly stated, the leading features of the theory which Ferrol laboriously worked out during the last thirty years, submitting all its parts to the test of both observation and mathe- matical analysis. By the end of his life (he died in 1891) he em- bodied his theory in a well-written and suggestive popular work,
- Full tables giving the eastward (or westward) velocities for each latitude, under the two
different hypotheses, have been calculated for the Meteorologische Zeitung, 1890, pp. 399 and 420.
f I'errel seems not to have been aware that the game had been demonstrated by R. Lenz for rivers (about the year 1870), in a discussion of Baer's law, applied to the Amu Kiver, in the M6moires of the St. Petersburg Academy.
\ William Ferrel, A Popular Treatise on Winds, comprising the General Motion of the Atmosphere, Monsoons, Cyclones, Tornadoes, Waterspouts, Hailstorms, etc. New York : Wiley, 1889. See also analysis of it by W. M. Davis (in Science, xv, p. 142 ; translated in Meteorologische Zeitung, 1800; Literaturbericht, p. 41), who gave the best diagram of cir- culation according to Ferrel's theory, and by H. F. Blanford in Nature, xli, 124. A full bibliography of Ferrel's works was given after his death in the American Meteorological Journal, October, 1891.
�� � RECENT SCIENCE. 395
which. fully deserves being widely kno"vni. All taken, his views so well agree with the facts relative to the movements of the at- mosphere, and they give such a sound method for further investi- gation, that they are sure to become for some years to come the leading theory of meteorology. They already have given a strong impulse to theoretical research, and have created a whole litera- ture in Austria and Germany.*
Another theory of the general circulation of the atmosphere which is also awakening a good deal of interest among physical geographers was propounded in 1886 by Werner Siemens, and further developed by him in 1890. f Siemens did not consider that air might flow down the density surfaces, as supposed by Ferrel and Helmholtz, and admitted by many meteorologists, and he maintained that the source of the energy required for all dis- turbances of equilibrium in the atmosphere must be looked for in the unequal heating of its different strata by the sun, and in the unequal loss of heat through radiation in space. From these considerations he inferred the existence of an ascending current in the equatorial belt, an upper warm current, and a cold polar current. As to the eastward and westward directions of these currents, he made the very just remark that the energy of rota- tion of the whole atmosphere must remain constant and un- changed, even though masses of air move from one latitude to
- Roth has already abandoned the mathematica] objections he had raised against Fer-
rel's theory in the Wochenschrift fiir Astronomie, 1888. The objections raised by Teis- serene du Bort and Supan against the " density surfaces " have been answered by Prof. Davis in Science, and are not shared by the most prominent meteorologists. And the mathematical analysis of Prof. Waldo, Sprung (the author of the well-known Treatise of Meteorology), M. MoUer, and Pemter has further confirmed the accuracy of the theory. So also Hildebrandsson's observations of upper clouds (Annuaire de la Societe meteoro- logique de France, xxxix, 338), Teisserenc du Bort's high-level isobar?, and Guaran de Trommelin's researches relative to coast winds. The transport of the Krakatoa dust and Abercromby's observations of clouds having rendered the existence of an upper east cur- rent very probable on the equator, Pernter has mathematically deduced from Ferrel's the- or)- the existence of such a current in a belt 4' 45' wide on both sides of the equator, and he therefore has withdrawn the restrictions he had previously made in a lecture (published in Nature, 1892, xlv, 593) in favor of Siemens's views. It must be added that the idea of three superposed currents blowing in spirals may have been suggested to Ferrel by a com- munication of James Thomson to the British Association in 1857. Such was, at least, the claim raised and developed at some length by the Glasgow professor before the Royal Society in a Bakerian lecture, now published in the Transactions (A. 1892, pp. 653-685). Though Thomson's paper was never published, and only given in a very short abstract with- out a diagram (the diagram in the Transactions is now pubhshed for the first time), the few lines in which his theory was stated (British Association Reports, Dublin, 1857, pp. 38, 39) contained the idea clearly expressed. It is certainly a matter of great regret that James Thomson has not returned to this subject.
f Ueber die Erhaltung der Kraft im Luftmeere, in Sitzungsberichte der Berliner Aka- demie der Wissenschaften, March, 1886, p. 261 ; Ueber das allgemeine Windsystem der Bide, in same publication, 1890, ii, p. 629. 396 THE POPULAR SCIENCE MONTHLY.
another. The velocitj'- of rotation of the atmosphere in tropical hititudes must therefore lag behind the rotation of the earth, and it must outstrip it in higher latitudes, mathematical calculation proving that the thirty-fifth parallel is, in both hemispheres, the line of division between the two. The general system of air cir- culation deduced from these principles is very similar in its re- sults to the system of Ferrel ; but the interest and importance of Siemens's views lie elsewhere. His memoirs were an appeal and an attempt to apply the principles of thermodynamics to the aerial currents, and they have opened the way for a series of im- portant researches, which, however, are not yet sufficiently ad- vanced to be discussed in these pages.
And, finally, a third new point of view has been introduced into the same discussions by Helmholtz. Sitting one day by the seaside, and observing how wind blows on the surface of the sea by sudden gushes, how it originates waves, and how they grow when wind blows with an increasing force, Helmholtz came to consider what would happen with two air currents blowing one above the other in different directions. A S5^stem of air waves, he concluded, must arise in this case, in the same way as they are formed on the sea. The upper current, if it is inclined toward the earth's surface (as is often the case), must originate in the lower current immense aerial waves rolling at a great speed. We do not generally see them, but when the lower current is so much saturated with moisture that clouds are formed in it, we do see a system of wavelike parallel clouds, which often extend over wide parts of the sky. To calculate the sizes. of the waves in different cases is extremely difficult, if not impossible; but by taking some simpler cases Helmholtz and Oberbeck showed that when the waves on the sea attain lengths of from sixteen to thirty- three feet, the air waves must attain lengths of from ten to twenty miles, and a proportional depth. Such waves would make the wind blow on the earth's surface in rhythmical gushes, which we all know, and they also would more thoroughly mix together the superposed strata, dissij)ating the energy stored in strong cur- rents. These views are so correct that they undoubtedly will throw some new light, as they already begin to do, upon the theory of cyclones.*
At the same time, Bezold is now endeavoring to reconstruct meteorology from the point of view of thermodynamics ; \ and the well-known Austrian meteorologist, J. Hann, whose work is
- H. Helmholtz, Ziir Theorie von Wind und Wetter, and Die Energie der Wogen und
des Windes, in the Sitzung.sberichte of the Berlin Academy, 1889, ii, and 1890, ii. Ober- beck's calculations of the waves are given in the Meteorologische Zeitimg, 1890, p. 81.
f Zur Thermodynamik der Atmospliiire, in Sitzungsberichte of the Berlin Academy of Sciences, 1888, p. 485; same year, p. 1189; 1890, p. 355; and 1892, p. 279.
�� � I
��RECENT SCIENCE. 397
exciting just now a great deal of interest, has openly broken with the old theory as regards the origin of cyclones and anticyclones.* From observations made for several years in succession on the top of the Sonnblick — a peak twelve thousand feet high, of the Tyrolese Alps — as well as from observations made on several high-level stations, he has concluded that a cyclone can not be due to a local heating of the earth's surface and to an ascending cur- rent of warm air provoked by this cause, just as an anticyclone can not be due to a local cooling of the earth's surface, and to a consequent condensation of the air. Contrary to the previsions of the meteorologists, the ascending column of air within a cy- clone, up to a height of some ten thousand feet, is not warmer than the surrounding air ; it is cooler within the cyclone, and its upward motion thus can not be due to its temperature. So also in an anticyclone the descending current of air is tcarmer than it is under normal conditions, and its downward motion must be due to some other cause than an increase of density resulting from a lowering of its temperature. The decrease of pressure in the one case, and its increase in the other, thus can not be caused by dif- ferences of heating or cooling of the lower strata ; and both cy- clones and anticyclones must be considered as parts of the general circulation of the atmosphere, such as it was conceived by Ferrel.f
Such a deep modification of the current views, though sup- ported to a great extent by weighty evidence, will obviously not be accepted without opposition; but it is already making its way, and certainly will exercise a deep influence on the further devel- opment of meteorology.
Abandoning now the domain of theoretical investigation, I must mention a work — also a life's work — which may safely be placed side by side with the best achievements in theory. I mean the beautiful charts of Mr. Buchan, representing the distribution of pressure, temperature, and winds over the surface of the globe, embodied in the last volume of the Challenger Expedition Re- ports. When Mr. Buchan published twenty-three years ago his first maps of monthly isobars and prevailing winds, they were quite a revelation, even though the data upon which they were based were very incomplete at that time.t But better data have
��* Das Luftdruckmaximum vom November 1889, in Denkschrift der Wiener Akademie dor Wissenschaften, 1890, Bd. Ivii, p. 401. Bemerkungen Uber die Temperatur der Cyclonen uud Antic} clonen, in Meteorologische Zeitscbrift, 1890, p. 328.
f See the discusision of this subject between Hazen and J. Hann in Science, 1890, xv, 382-384, and Meteorologische Zeitsehrift, 1890, p. 328.
X To trace the isobars, or lines of equal atmospheric pressure, reduced to the sea-level, the real altitude of each meteorological observatory must be known from direct geometrical levelings; but in 18G9 the altitude of not one single station in Siberia, central Asia, or even the Urals was known. A leveling across Sibeiia, as far as Lake Baikal, has been
�� � 398 THE POPULAR SCIENCE MONTHLY.
been collected since, and in the hands of Mr. Buchan they have undergone such a careful and able analysis that the Challenger Reports charts may be taken as the best reliable representation of the winds, the temperatures, and the pressure in the lowest strata of the atmosphere, as well as the surest basis for further generali- zations.* The theories which have been mentioned in the preced- ing pages give the grand lines of atmospheric circulation; on Buchan's maps we see how the grand lines are modified in the lowest strata by the distribution of land and sea, and the unequal heating or cooling of continents and oceans. The leading features indicated by theory are still maintained, and they become even still more apparent if we consult isobars traced for a certain height, like those of Teisserenc de Bort ; but the immense plateaus of East Asia and North America act in winter as colossal refriger- ators, where cold and heavy air accumulates, to flow down in all directions toward the lowlands. We see also how in July the air is heated in the lower lands of northwest India, in the corner be- tween the Afghanistan and the Thibet plateau, how pressure is lowered there by the ascending current, and how winds blow toward this region of lowered pressure. We see more than that : on looking on the maps it strikes the eye how the moisture or the dryness of the climate is dependent upon the distribution of press- ure, and how the dry anticyclonic winds make barren deserts of parts of North and South America, of Africa, and central Asia, and how they will continue to dry the lakes and the rivers of these regions and occasion total failures of crops so long as that distribution of pressure lasts on the globe, and man has not yet learned to eschew its effects by getting water from the depths of the earth. The life of the globe during the present period is written on these splendid charts. — Nineteenth Century.
��M. TnoRADDSEN, in the narrative of his travels in Iceland, observes a peculiar feature of the oases at the foot of Mount Hecla. These oases are subject to con- stant displacement by the violent sandstorms which are common. On the wind- ward side all vegetation is gradually destroyed, while on the other side grass takes root, and in a wonderfully sliort time the level and sterile surfaces are converted into good pasture lands.
made since, Mr. Buchan's iriobars having been one of our best arguments to press the neces- sity of the leveling. But Mr. Buchan may not be aware that the leveling beyond the ninetieth degree of longitude is now considered by Russian geodesists as utterly unreliable; it is supposed to contain some substantial error, so that a new leveling between Krasno- yarsk and Lake Baikal is insisted upon. The incertitude in the isobars on an immense space in northeast Asia resulting from this cause may attain as much as one or, perhaps, even three tenths of an inch.
- An excellent rhume of the whole work and its results in a popular form has been
published by Buchan himself in the Proceedings of the Geographical Society, March, 1891.