Skip to main content

Full text of "The atmosphere : that which gives life to the earth and by which everything has its being : an ethereal sea that covers the whole world"

See other formats


Presented to the 
LIBRARY of the 



Robert Roseberi^'- 

Digitized by the Internet Archive 

in 2007 with funding from 

IVIicrosoft Corporation 




> , 


-4. Mar if pinx' 

EiM. CicSri chromolith' 



rHR ATMOSPHERE that which gives 






The following work is translated and abridged from M. Flammarion's 
L' Atmosphere^ Paris, 1872. That some curtailment of the text of the 
original work was requisite will be apparent when it is stated that the 
French Edition contains 824 large pages of closely printed matter, and 
is of more than twice the extent of the present volume. Not only was 
some compression necessary in order to bring the work within a rea- 
sonable compass, but, independently of this, one or two chapters, such 
as that on the Eespiration and Alimentation of Plants, appeared to have 
so remote a connection with the subject of the work — the Atmosphere 
— that their omission would in any case have been desirable. 

Every one who has any acquaintance with French popular works on 
Science is aware that very many exhibit a tendency to imaginative, or, 
to express my meaning colloquially, "fine" writing, which ill accords 
with the precision and accuracy that ought to be a characteristic of 
scientific information, even when expressed in language free from tech- 
nicalities. There is a good deal of this exalted kind of composition in 
M. Flammarion's book, which — even in the French not very agreeable 
to an English reader — becomes, when translated, intolerable. I have, 
therefore, omitted these rhapsodies very freely, though traces enough 
of them will be found here and there to betray the French origin of 
the work. 

I may add that the task of editing has not been a light one ; besides 
the necessity for compression and the consequent selection of the mat- 
ter to be included, I have been obliged to exercise some sort of censor- 
ship over the facts contained in the work. It is impossible for any one 
man to have a complete knowledge of so great a variety of subjects as 
are treated of by M. Flammarion, and the compiler of such a book must 
include many things taken from others, of the accuracy of which he is 
not fully competent to judge. In cases where a statement contained in 
the original work appeared to me clearly erroneous, I have corrected it, 


appended a note, or omitted it altogether ; and in cases where I have 
been doubtful of the accuracy of a passage, or have differed in opinion 
from the author, I have not considered myself j ustified in making an 
alteration, so long as there was no strong prima facie presumption that 
the original was incorrect. In spite of obvious blemishes, inseparable 
from a translation, and a certain want of continuity in a few places, 
which is due to the omission of portions of the book as originally 
written, I believe the volume will be found to be readable, popular, and 
accurate, and it covers ground not occupied by any one work in our 

The work treats on the form, dimensions, and movements of the 
earth, and of the influence exerted on meteorology by the physical 
conformation of our globe; of the figure, height, color, weight, and 
chemical components of the atmosphere ; of the meteorological phe- 
nomena induced by the action of light, and the optical appearances 
which objects present as seen through different atmospheric strata ; of 
the phenomena connected with heat, wind, clouds, rain, and electricity, 
including the subjects of the laws of climate: the contents are there- 
fore of deep importance to all classes of persons, especially to the ob- 
server of nature, the agriculturist, and the navigator. 

The whole is explained in a very popular manner, and as free as pos- 
sible from all technicalities ; the object having been to produce a work 
giving a broad outline of the causes which give rise to facts of every- 
day occurrence in the atmosphere, in such a form that any reader who 
wished to obtain a general view of such phenomena and their origin 
would be readily enabled to do so. The great number of subjects 
treated of will thus, to the majority of readers, who merely desire an 
insight into the general principles that produce phenomena which 
every one has seen or heard of, be found to be rather an advantage, as 
the whole range of atmospheric action is thus displayed in the same 
volume in moderate compass, without so much detail being anywhere 
given as to make the book other than interesting to even the most 
casual reader. 

The translation was made by Mr. C. B. Pitman. 

January., 1873. 


In ea vivimus, movemur et sumus. 

Of all the various subjects which invite a studious examination, it is 
impossible to select one possessing a more direct, a more permanent, or 
a more real interest than that which forms the subject of this work. 
The Atmosphere gives life to earth, ocean, lakes, rivers, streams, forests, 
plants, animals, and men ; in and by the Atmosphere every thing has 
its being. It is an ethereal sea reaching over the whole world; its 
waves wash the mountains and the valleys, and we live beneath it and 
are penetrated by it. It is the Atmosphere which makes its way as a 
life-giving fluid into our lungs, which gives an impulse to the frail ex- 
istence of the new-born babe, and receives the last gasp of the dying 
man upon his bed of pain. It is the Atmosphere which imparts verd- 
ure to the fertile fields, nourishing at once the tiny flower and the 
mighty tree; which stores up the solar rays in order to give us the 
benefit of them in the future. It is the Atmosphere which adorns with 
an azure vault the planet in which we move, and makes us an abode in 
the midst of which we act as if we were the sole tenants of the infinite 
— the masters of the universe. It is the Atmosphere which illuminates 
this vault with the soft glitter of twilight, with the waving splendors of 
the aurora borealis, with the quivering of the lightning and the multi- 
form phenomena of the heavens. At one moment it inundates us with 
light and warmth, at another it causes the rain to pour down in torrents 
upon the thirsty land. It is the channel by which the sweet perfumes 
descend from the hills, and the vehicle of the sound which permits 
human beings to communicate with each other, of the song of the birds, 
of the sighing of the wind among the trees, of the moaning of the waves. 
Without it, our planet would be inert and arid, silent and lifeless. By 
it the globe is peopled with inhabitants of every kind. Its indestructi- 
ble atoms incorporate themselves in the various living organisms; the 


particle which escapes with our breath takes refuge in a plant, and, 
after a long journey, returns to other human bodies ; that which we 
breathe, eat, and drink has already been inhaled, eaten, and drunk mill- 
ions of times: dead and living, we are all formed of the same sub- 
stances What study can possess a vaster or more direct interest 

than that of the vital fluid to which we owe the manner of our being 
and the maintenance of our life? 

The study of the Atmosphere, of its physical condition, of its move- 
ments, of its functions, and of the laws which regulate its phenomena, 
forms a special branch of human research. This science, which since 
the days of Aristotle has been designated Meteorology^ belongs in part 
to Astronomy, which shows the movements of our planet around the 
sun — movements to which we owe day and night, season, climates, solar 
action, or, in a word, the basis of the subject. On the other hand, it 
appertains to Natural Philosophy and Mechanics, which explain and 
measure the forces brought into play. As it exists in the present day, 
Meteorology is a new science, of recent establishment, scarcely as yet 
fixed in its elementary principles. 

We are assisting at its elaboration-, at its struggling into life. The 
present generation has seen the establishment of meteorological societies 
throughout the different nations of Europe, and of special observatories 
for the exclusive study of the problems relating to the Atmosphere. 
The analysis of climates, seasons, currents, and periodical phenomena 
is scarcely terminated. The examination of atmospheric disturbances, 
of tempestuous movements, and of storms, has been made, so to speak, 
before our own eyes. The science of the Atmosphere is the question 
of the day. We are just now, in regard to this study, in an analogous 
situation to that of modern Astronomy in the days of Kepler. Astron- 
omy was founded in the seventeenth century. Meteorology will be the 
work of the nineteenth. 

I have endeavored to collect in this work all that is at present posi- 
tively known about this important subject, to represent as completely 
as possible the actual state of our knowledge about the Atmosphere 
and its work— that is, about the air, the seasons, the climates, the winds, 
the clouds, the rain, the hurricanes, the storms, the lightning, the me- 
teors — in a word, the phenomena of time, and above all, the general 
upholding of terrestrial life. It is, in fact, a synthesis of the research 
effected during the last half century (especially during the latter portion 
of It) as to the great phenomena of terrestrial nature, and the forces 


which produce them. The great majority of us, inhabitants of the 
earth, no matter to what nation we belong, pass our lives without at- 
tempting to form an idea of our actual position, without asking ourselves 
what is the force which prepares for us our daily bread, ripens for us 
the grapes that give the wine, presides over the change in the seasons, 
and alternates the exhilarating blue sky with the rains and cold of in- 
hospitable winter. Yet, why should we live in such a state of igno- 
rance? I venture to hope that after perusing this work there will be 
no difficulty in understanding the life and movements of the globe. 
Every thing which takes place around us is interesting when, instead of 
remaining as one born blind, a man has learned to appreciate external 
things and to keep himself in intelligent communication with Nature. 

I could have wished to keep this work, destined as it is for the gen- 
eral public, free from scientific terms and figures which constitute its 
basis. I have done so as far as possible, but without in any point sacri- 
ficing accuracy and precision in respect to observed facts. It seems to 
me, too, that what is termed the public (that is, every one) has become 
somewhat scientific itself, since so many excellent works have popular- 
ized ideas previously reserved for. a small circle of the elect. 

Camille Flammarion. 

Paris : November, 1871. 





I. The Terrestrial Globe 17 

II. The Atmospheric Envelope 23 

III. The Height of the Atmosphere 28 

IV. Weight of the Terrestrial Atmosphere — The Barometer 

AND Atmospheric Pressure 38 

V. Chemical Components of the Air 57 

VI. Sound and the Voice 75 

VII. Aeronautical Ascents 85 



I, The Day 103 

II. Evening 113 

III. The Rainbow 121 

TV. Anthelia : Spectre-shadows upon Mountains — The Ulloa 

Circle — Circle seen from a Balloon 127 

V. Halos : Parhelia — Paraselenes — Circles surrounding 
and traversing the Sun — Coronas — Columns — Vari- 
ous Phenomena 137 

VI. The Mirage 149 




VII. SiiooTixG-STARS— Bolides— Aerolites— Stones falling from 

THE Sky 1^^ 

S'lII. The Zodiacal Light 1*^4 



I. Heat: the Thermometer — Quantity of Heat received — 

Temperature of the Sun — Temperature of Space 181 

11. Heat in the Atmosphere 1 90 

III. The Temperature of the Air : Its mean Condition — Daily 

AND Monthly Variations of the Temperature — Tem- 
perature OF EACH Summer, Winter, and Year at Paris 
AND AT Greenwich since the last Century — Daily and 
Monthly Variations of the Barometer 202 


V. Autumn — Winter : Winter Landscapes — Cold — Snow — Ice 
— Hoar-frost, Rime, etc. — Remarkable Winters — The 


VI. Climate : Distribution of TeisIperature over the Globe — 
Isothermal Lines — The Equator — The Tropics — The 
Temperate Regions — The Poles — The Climate of 
France 245 



I. The Wind and its Causes: General Circulation of the 
Atmosphere — The Regular and Periodical Winds — 
Trade-winds— The Monsoon — Breezes 269 

IL The Sea Currents: Meteorology of the Ocean — Maritime 

Routes — The Gulf Stream 284 



III. The Variable Winds — The Wind in our Climates — Mean 

Directions in Europe and in France — Relative Fre- 
quency OF different Winds — Rise op the Winds accord- 
ing TO the Times and Places — Monthly and Diurnal Va- 
riation IN Intensity 29V 

IV. Respecting certain Special Winds : The Bise — The Bora — 

The Gallego — The Mistral — The Harmattan — The Si- 
moom — The Khamseen — The Sirocco — The Solano 318 

V. The Power of the Air : The 'Hurricane — The Cyclone — 

The Tempest 327 

VI. Trombes, Whirlwinds, or Water-spouts 337 



I. The Water upon the Surface of the Earth and in the At- 
mosphere : The Earth — Volume and Weight of the Wa- 
ter throughout the Globe — Perpetual Circulation — 
Vapor of Water in the Atmosphere — Its Variations ac- 
cording TO THE Height, the Locality, and the Weather 
— The Hygrometer — Dew — White Frost 355 

II. The Clouds : What a Cloud is — The Manner of its Forma- 
tion — Mist — Observations taken from a Balloon and 
from Mountains — Different Kinds of Clouds — Their 
Shapes — Their Heights 363 

III. Rain: General Conditions of the Formation of Rain — Its 

Distribution over the Globe — Rain in Europe 381 

IV. Hail: Production of Hail — Course of Hailstorms — Vary- 

ing Distribution of Hailstorms in different Parts of 
the Country -r- Heaviest Hailstorms known — Nature, 
Size, and Shape of Hailstones — Periods of their Occur- 
rence 390 

V. Prodigies: Showers of Blood — of Earth — of Sulphur — 
OF Plants — of Frogs — of Fish— of various Kinds of An- 
imals 401 





I. Electricity upon the Earth and in the Atmosphere : Elec- 
tric Condition of the Terrestrial Globe — Discovery of 
Atmospheric Electricity — Experiments of Otto de Gub- 
RiCKE, Wall, ISTollet, Franklin, Romas, Richmann, Saus- 
suRE, ETC. — Electricity of the Soil, of the Clouds, of the 
Air — Formation of Thunder-Storms 423 

II. Lightning and Thunder -. 431 

III. The Saint Elmo Fires and the Jack-o'-Lanterns 441 

IV. AuROR-fi: Boreales 445 




1. Halo Frontispiece. 

2. Sunset at Sea To face 119 

3. TheKainbow " 121 

4. Lunar Eainbow seen at Compi^gne - " 126 

5. Sunrise from the Righi " 127 

6. African Mirage " 149 

7. Summer Landscape " 218 

8. Winter Landscape " 229 

9. The Storm " 423 

10. Aurora Borealis seen at Paris, May 13, 1869 " 445 


1. Mathematical Limit of the Shape of the Atmosphere 29 

2. Measure of the Height of the Atmosphere, according to the Length of Twilight 32 

3. Thickness of the Earth's Crust, of our Atmosphere, and of a higher Atmosphere 34 

4. Suction-Pump 39 

5. Suction and Forcing Pump 40 

6. Torricelli inventing the Barometer •. 41 

7. Barometer Tube full of Quicksilver 43 

8. The Tube in the Basin •. 43 

9. Otto de Guericke's Experiment 45 

10. The Magdeburg Hemispheres 46 

11. Atmospheric Pressure. Rupture of Equilibrium 47 

12. Atmospheric Pressure imder an inverted Glass 47 

13. Diagram showing the Decrease of atmospheric Pressure, according to Height 51 

14. Variation in the atmospheric Pressure at the Level of the Sea 52 

15. Lavoisier analyzing atmospheric Air 56 

16. Matrass or Glass Vessel 58 

17. The Apparatus for Analysis of Air 58 

18. Mercury-Eudiometer, for analyzing Air 59 

19. Apparatus for analyzing Air by the Method of Weight.. 60 

20. Apparatus for obtaining the Proportion of carbonic Acid in Air 61 

21. Apparatus for separating the Oxygen from the Nitrogen 62 

22. Vibrations of a Blade 75 

23. Vibration of a Cord 76 

24. Illustration of Hawksbee's Experiment 78 

25. Baroscope 86 

26. Soap-bubbles inflated with Hydrogen 88 

27. Distribution of Kinds of Birds according to Height of Flight 97 

28. Lunar Day Ill 

29. Atmospheric Refraction 114 

30. Simple Reflection of Rays in a Drop of Rain 121 




31. Foimiition of t!ic Rainbow 123 

32. Double Ueflection of Hays in ii Drop of Kain 124 

33. Theory of the two Arches of a Rainbow 124 

34. Triple Rainbow 125 

35. The S])ectre of the Brocken 129 

36. The Ulloa Circle .- 132 

37. Theory of the Halo 140 

38. Halo seen in Norway 142 

39. Corona formed around the Moon by Diffraction 147 

40. Explanation of the ordinary Mirage 152 

41. Mirage seen at Paris in 18G9 158 

42. Lateral Mirage seen on the Lake of Geneva 160 

43. La Fata Morgana 162 

44. Shooting-stars 165 

45. Fall of a Bolide in the Daytime 170 

46. The Caille Aerolite, weighing 12^ cwt 172 

47. The Pyrheliometer 183 

48. Relative Intensity of the calorific, luminous, and chemical Rays of the Sun 192 

49. Inequality of the Thickness of Air traversed by the Sun 196 

50. Regular Diurnal Oscillation of the Barometer 213 

51. Regular Monthly Oscillation of the Barometer 216 

52. Snow Crystals 231 

53. Winter.— The Seine fiill of floating Ice 235 

54. Comparative Temperatures of Rome, London, Paris, Vienna, St. Petersburg 251 

55. The last human Dwelling-places. Esquimaux of the Polar Regions 262 

56. Ice at the Pole 264 

57. Section of the Atmosphere, showing its general Circulation 272 

58. Average annual Prevalence of different Winds at London 305 

59. Average annual Prevalence of the different Winds at Brussels 305 

60. Monthly Intensity of the Winds 307 

61. Diurnal Intensity of the Winds 307 

62. The Simoom 323 

63. Whirlwind 346 

64. Sand Whirlwind 348 

65. Water-spout at Sea 350 

66. Intense Fog in one of the Islands of the Antipodes 368 

67. Intense Fog in the Spitzbergen Mountains 369 

68. Formation of a Thunder-cloud 377 

69. Above and below the Rain-cloud 380 

70. Diminution in the Rain-fall from the Tropics to the Poles 383 

71. Increase of Rain, according to the Undulations of the Soil 384 

72. Comparative Depths of Rain-fall 385 

73. Section of Hailstones, showing their ordinary interior Structure 398 

74. Section of a Hailstone, enlarged 399 

75. Different Forms of Hail 400 

76. Rain of Blood in Provence, July, 1G08 404 

77. Shower of Locusts 417 

78. Shower of Cock-chafers 418 

79. Experiments of Franklin and Romas 424 

80. Richmann, of St. Petersburg, struck by Lightning during an electrical Experiment. . 426 

81. Harvesters killed by Lightning 438 

82. Curious Freak of Lightning 440 

83. Saint Elmo Fire over the Spire of Notre-Dame, Paris 442 

84. An Aurora Boreal is over the Polar Sea 447 

85. Aurora Borealis observed at Bossekop (Spitzbergen), January 6, 1839 449 

86. Aurora Borealis observed at Bossekop (Spitzbergen), January 21, 1839 451 






Borne forward in space, in obedience to the mysterious laws of uni- 
versal gravity, our globe travels therein with a rapidity that our closest 
study can scarcely conceive. Let us imagine a sphere absolutely free, 
isolated on all sides, without any prop or stay, placed in the midst of 
space. If this sphere were alone in the immensity, it would remain 
thus suspended, motionless, without power to incline to this side or to 
that. Eternally fixed, it would constitute in itself the whole of crea- 
tion ; astronomy and physics, mechanics and biology, would all be in- 
cluded in its conception. But the earth is not the only world existing 
in space. Millions of celestial bodies have been formed, like itself, in 
the infinite heavens, and their co-existence establishes between them 
relations inherent in the very constitution of matter. The earth, in 
particular, belongs to a system of planets analogous to itself, having the 
same origin and the same destiny, situated at various distances around 
the same centre, and governed by the same motive power. Our planet- 
ary system is composed essentially of eight worlds, made to revolve in 
successive orbits, the exterior one of which is seven thousand million 
leagues in extent. The sun, a colossal star nearly a million and a half 
times larger than the earth, and 350,000 times as heavy, occupies the 
centre of these orbits ; or, to speak more accurately, a focus of one of 
the nearly circular ellipses which they describe. It is around this gi- 
gantic star that take place the revolutions of the planets, which are per- 
formed with an indescribable speed on account of the length of the cir- 
cumference to be traversed. Far from being motionless, as it appears 
to us, the globe which we inhabit revolves at an average distance of 
ninety-one and a half millions of miles from the sun, and over an orbit 



which does not measure less than 587 millions of miles. These are 
traversed in 365 days and six hours — that is to say, that we move 
through space with a speed of more than one and a half million of 
miles per day, or more than 66,000 miles an hour. The most rapid of 
express trains can scarcely accomplish more than twenty-five leagues 
an hour. Upon the invisible roads of the heavens the earth moves 
with a speed eleven hundred times greater. The difference is so enor- 
mous, that it is impossible to express it in this work by a geometrical 
tio-ure. If the distance traversed in an hour by a locomotive was rep- 
resented by one tenth of an inch, it would be necessary to trace a line 
more than nine feet long to indicate the comparative advance made by 
our planet during the same space of time. I will add, as a point of 
comparison, that the movement of the tortoise is about eleven hundred 
times less rapid than that of an express train. Consequently, were an 
express train to be sent in pursuit of the earth, it would be as a tortoise 
in pursuit of an express train. 

Situated as we are about the globe, infinitely small mollusks, made to 
adhere to its surface by its central attraction, and carried away with it, 
we are unable to appreciate this movement or form a direct idea con- 
cerning it. It is only by the observation of the corresponding change 
of position in the celestial perspectives, and by calculations based thereon, 
that we have been able — and this only during the last few centuries — 
to acquire a knowledge of its nature, its form, and its importance. From 
the deck of a ship, from a rail way -carriage, or the car of a bdlloon, we 
are alike unable to form an idea of the movement that is transferring 
us from one place to another, because we participate in it; and with- 
out some object of comparison not partaking of the motion, it is impos- 
sible for us to appreciate it. To form an idea of the rapidity of the 
earth's motion, we must imagine ourselves placed not upon the earth's 
surface but outside it, in space itself, not far from the course along which 
it hurries so impetuously. Then we should see far in the distance — to 
our left, I will suppose — a little star shining amidst the rest in the gloom 
of space. Then this little star would seem to grow larger, and to draw 
nearer to us. Soon there would be perceptible a disk like that of the 
moon, upon which we should also recognize spots formed by the op- 
tical difference between continents and seas, by the polar snows and the 
cloudy bands of the tropics. We should endeavor to distinguish upon 
this gradually swelling globe the principal geographical shapes visi- 
ble athwart the vapors and clouds of the atmosphere, when suddenly. 


Standing out against the sky and covering the immensity of its dome, 
the globe would meet our affrighted gaze, as if it were a giant emer- 
ging from the abysses of space. Then, rapidly, without giving us time 
to recognize it, the colossus would rush away to our right, quickly 
diminishing in size, and silently burying itself in the dark depths be- 
yond. So moves the globe we inhabit, and we are borne along by it 
like so many grains of dust adhering to the whirling surface of a can- 
non-ball projected into space. 

How great a difference there is between this truth and the ancient 
fallacy which represented the earth as the support of the firmament! 
During the reign of illusion — so old, and yet so difficult to dispel, even 
in our epoch, from certain minds — the earth was believed to form in it- 
self alone the living universe, and to represent the whole of nature. It 
was the centre and objective of all creation, while the rest of space was 
but a vast and silent solitude. There was a higher region in the uni 
verse — viz., the heavens, the empyreum ; a lower region — viz., the earth, 
hell. Mysticism had created the world for terrestrial humanity alone, 
as being the centre of Divine Will. In the present day we know that 
the heavens are but boundless space, and that the earth is in the heav- 
ens just as the other stars ; we contemplate in the firmament worlds sim- 
ilar to our own, and the starry night addresses itself to our minds with 
a new eloquence. The terrestrial globe, with its humanity, is no longer 
more than an atom cast into the infinite — one of the countless fly-wheels 
which, in tens of thousands, constitute the mysterious mechanism of the 
physical world. Our planetary system, despite its vastness, compared 
to the microscopical volume of this earth, is, sun and all, eclipsed in the 
presence of the extent and number of the stars, which are solar centres 
of systems distinct from ours. The astonished gaze encounters distant 
suns whose light takes hundreds and thousands of years to reach us, not- 
withstanding its wondrous speed of 186,000 miles a second ; farther still 
the eye may contemplate pale masses of stars which, seen nearer, would 
resemble our Milky Way, and would be found to be composed of mill- 
ions of suns and systems ; be3'-ond these, again, the eye and the mind 
still seek to discover more distant creations, but the sweep of our fa- 
tigued conceptions soon falls to a lower level, worn out and lost by 
this interminable flight into the regions of infinity. 

An invisible star, lost in the myriads of stars, the earth is borne along 
in the heavens by various movements far more numerous and peculiar 
than most people would be inclined to suppose. The most important 


one is that of revolution, which we have noticed above, and by virtue 
of which the earth moves round the sun at the rate of one and a half 
million of miles a day. A second movement, that of rotation, causes it 
to turn round its own axis in the course of every four-and-twenty hours. 
It may be at once seen, in examining this movement of the globe, that 
the different points of the terrestrial surface have a different speed, ac- 
cording to their distance from the axis of rotation. At the equator, 
where the speed is greatest, the terrestpial surface has to traverse 25,000 
miles in twenty-four hours ; that is, more than 1040 miles an hour, or 
about seventeen a minute. In the latitude of London, where the circle 
is perceptibly smaller, the speed is eleven miles a minute. At Rekia- 
witz, one of the towns almost in the heart of the polar region, the speed 
is seven and a half miles a minute; and finally, at the poles themselves, 
it is nil. A third movement, that which constitutes the precessi07i of the 
equinoxes, causes the terrestrial axis to accomplish a slow rotation, which 
occupies not less than 25,868 years, and in virtue of which all the stars 
of heaven annually seem to change their position, to return to the same 
point only at the close of this great secular cycle. A fourth movement 
gradually makes a change in the position of the perihelion, which makes 
the circuit of the orbit in 20,984 years, so that in this other cycle the 
seasons successively take the place the one of the other. A fifth move- 
ment causes the plane of the earth's orbit, which it describes around the 
sun, to oscillate, and diminishes the obliquity of the ecliptic at present, 
to increase it in the future. A sixth movement, due to the action of 
the moon, and called nutation, causes the pole of the equator to describe 
upon the celestial sphere a small ellipse in eighteen years and eight 
months. A seventh movement, caused by the attraction of the planets, 
and principally by the gigantic world of Jupiter and our neighbor Ye- 
nus, occasions perturbations, calculable beforehand, in the curve de- 
scribed by our planet around the sun, swelling or flattening it, according 
to the variations of distance. An eighth movement, more considerable 
and less exactly measured than the preceding ones, though its existence 
is incontestable, is the transport of the whole planetary system in space. 
The sun is thus not motionless, but traverses an immense orbital line, 
the direction of which is at present toward the constellation of Her- 
cules. The speed of this general movement is estimated at 487,000 
miles a day. The laws of motion would incline one to believe that the 
sun gravitates around a centre as yet unknown to us. If so, how vast 
must be the extent of the circumference of the ellipse which it describes, 


since for the last century it has followed, as far as we can judge, a per- 
fectly straight line ! 

These different movements, which cause the earth to travel in space, 
are ascertained with certainty, thanks to the vast number of the ob- 
servations of the stars made for more than 4000 years, and to the defi- 
nite nature of the modern principles of celestial mechanics. The knowl- 
edge of these constitutes the essential basis of the highest and most 
substantial of sciences. The earth is henceforth inscribed in the ranks 
of the stars, in spite of the evidence of the senses, in spite of secular 
illusions and errors, and, above all, in spite of human conceit, which 
had for a long time complacently formed a creation for man alone. 
Drawn here and there by these diverse movements — some of which, 
such as that of the perturbations^ are extremely complicated — the terres- 
trial globe travels onward, whirling along, balancing itself under the in- 
fluence of varied forces, rushing with an incomprehensible rapidity to- 
ward an unknown goal. Since the beginning of the world, the earth 
has not twice passed the same spot, and the place which we occupy at 
this very moment is rapidly sinking behind into our track never to re- 
turn. The very terrestrial surface, too, undergoes changes every centu- 
ry, every year, every day, and the conditions of life change throughout 
eternity as throughout space. After having thus examined the move- 
ment of the earth in space, we must join to it, in order to complete its 
astronomical aspect, the motion of the moon round the earth in twenty- 
nine days and a half. The moon is only -^-^ of the size, and -g-^ of the 
weight of the earth. Its action upon the ocean and the atmosphere is, 
nevertheless, comparable with that of the sun, and is even more impor- 
tant as regards the production of tides : it is as useful to know its move- 
ment about us as to know that of our planet about its primary. The 
revolution of the moon around the earth takes place really in twenty- 
seven days and eight hours, but during these twenty-seven days the 
earth has not been motionless, but, on the contrary, has advanced a cer- 
tain distance. The moon employs about two days more to complete its 
'revolution and to return to the same point in relation to the sun, which 
gives twenty-nine days and thirteen hours for the lunation or the cycle 
of phases. The revolution in twenty-seven days is called the sidereal 
revolution, because in that time the moon returns upon the celestial 
sphere to the same position in relation to the stars. We see that to re- 
turn to the same position in relation to the sun, and to accomplish its 
synodical revolution, our satellite must make more than a circle upon 


the cclcstinl sphere, and pass over in addition the distance which the 
earth has traveled during that time. If we suppose the earth motion- 
less, the movement of the moon round it may be nearly represented by 
a circle. In reality, it is a sinuous line, resulting from the combination 
of the two movements. 

Three stars thus command our attention in the general history of na- 
ture — the sun, the earth, and the moon. They are held up, isolated, 
in space in a manner dependent on their respective weights. The sun 
weio-hs two quadrillions of tons (two followed by twenty-four zeros). 
The sun is 355,000 times heavier than the earth, the latter eighty times 
more so than the moon. The sun holds the earth at arms-length, so to 
speak, ninety-one and a half millions of miles distant ; the earth holds 
the moon — also by the influence of its mass — at a distance of 237,000 

In gravitating around our luminary, the earth, constantly immersed 
in its rays, brings the different portions of its surface successively into 
its fertilizing emanations. Morning succeeds evening, and spring au- 
tumn. Night, like winter, is but the transition from one light to anoth- 
er. The solar heat keeps in continual work the mighty factory of the 
terrestrial atmosphere, forming the currents, the winds, the tempests, 
and the breezes ; preserving the water liquid and the air gaseous, rais- 
ing water from the inexhaustible wells of the ocean, producing the 
mists, the clouds, the rains, and the storms ; organizing, in a word, 
the permanent system of the vital circulation of the globe. 

It is this system of circulation, with the varied phenomena of the 
atmospheric world, which we are about to study in this work. The 
subject is vast and grand, for upon it depends all terrestrial life. In 
studying it we learn, therefore, the very organism of existence upon 
the planet we inhabit. 




Our globe, the motions of which we have been explaining, is encir- 
cled by a gaseous film which adheres to its entire spherical surface. 
This layer of fluid extends with uniform thickness all round the globe, 
covering it on every side. We have already compared the earth in the 
midst of space to a cannon-ball launched into the air; by imagining 
this cannon-ball surrounded by a thin ring of smoke not more than -^^ 
of an inch thick, we may form some idea of the position of the atmos- 
phere around the terrestrial globe. It is, indeed, from this position that 
the atmosphere derives its name ('Ar/uoc, vapor ; and ^^alpa, sphere), 
being, as it were, a second sphere of vapor concentric with the solid 
sphere of the globe itself. As a rule, sufficient importance is not at- 
tached to the functions of this atmospheric envelope. It is from it that 
we draw our being. Plants, animals, and men imbibe therefrom the 
first elements of their existence. The earth's organization is so ordered 
that the atmosphere is sovereign of all things, and that the savant can 
say of it as the theologian said of God: "In it we live and move, and 
have our being." 

The air is the first bond of society. Were the atmosphere to vanish 
into space, an eternal silence would be the lot of the terrestrial surface. 
We may not think of the fact with our forgetfulness of nature, but none 
the less the air is the great medium of sound, the liquid channel in 
which our words travel, the vehicle of language, of ideas, and of social 

It is also the first element of our bodily tissues. Breathing affords 
three-quarters of our nourishment ; the other quarter we obtain in the 
aliment, solid and fluid, in which oxygen, hydrogen, nitrogen, and car- 
bonic acid are the chief component parts. Further, the particles which 
are at the present moment incorporated in our organism will make 
their escape either in perspiration or in the process of breathing; and, 
after having sojourned for a certain time in the atmosphere, will be re- 
incorporated in some other organism, either of plant, animal, or man. 

With the unceasing metamorphoses in beings and in things, there is 


at the same time going on a continuous exchange between the products 
of nature and the moving flood of the atmosphere, by virtue of which 
the gases of the air take up their abode in the animal, the plant, or the 
stone, while the primitive elements, momentarily incorporated in an 
orf^anism, or in the terrestrial strata, effect their release and help to re- 
compose the aerial fluid. Each atom of air, therefore, passes from life 
to life, as it escapes from death after death ; being in turn wind, flood, 
earth animal, or flower, it is successively employed in the composition 
of a thousand different beings. The inexhaustible source whence ev- 
ery thing that lives draws breath, the air is, besides, an immense reser- 
voir into which every thing that dies pours its last breath ; under its 
action, vegetables and animals and various organisms are brought into 
existence, and then perish. Life and death are alike in the air which 
we breathe, and perpetually succeed the one to the other by the ex- 
change of gaseous particles ; thus the atom of oxygen which escapes 
from the ancient oak may make its way into the lungs of the infant in 
the cradle, and the last sigh of the dying man may go to nourish the 
brilliant petal of a flower. The breeze which caresses the blades of 
grass goes on its way until it becomes a tempest that uproots the forest- 
trees and strews the shore with shipwrecks ; and so, by an infinite con- 
centration of partial death, the atmosphere provides an unfailing sup- 
ply of aliment for the universal life spread over the surface of the 

It is this unceasing activity of the aerial envelope of gas which forms, 
nourishes, and sustains the vegetable carpet that extends over the sur- 
face of the dry land. From the meanest blade of grass to the colossal 
Baobab^ this rich and diversified covering draws all its sustenance from 
the air. 

And while it keeps up the vital circulation of the earth by incessant 
exchanges of which it is the vehicle, the atmosphere is also the aerial 
laboratory of that splendid world of colors which brightens the surface 
of our planet. It is owing to the reflection of the blue rays that the 
sky and the distant heights near the horizon assume their lovely azure 
tint, which varies according to the altitude of the spot and the abun- 
dance of the exhalations ; and to it also we owe the contrast of the 
clouds. It is in consequence of the refraction of the luminous rays, as 
they pass obliquely across the aerial strata, that the sun announces its 
approach every morning by the soft and pure melody of the glowing 
dawn, and makes its appearance before the astronomical hour at which 


it should rise ; it is owing to a similar phenomenon that, toward even- 
ing, it apparently slackens the speed of its descent beneath the horizon, 
and, when it has disappeared, leaves floating upon the western heights 
the fantastic fragments of its blazoned bed. Without the gaseous en- 
velope of our planet, we should never have that varied play of light, 
those changing harmonies of color, those gradual transformations of 
delicate shades which lighten up the world, from the gleaming bright- 
ness of the summer sun down to the shadows which cover, as with a 
veil, the forest depths. 

The study of the atmosphere embraces also the general conditions of 
terrestrial existence. The notion of life is so bound up in all our con- 
ceptions with that of the forces which we see ever at work in nature, 
that the myths of the early inhabitants of the world always attributed 
to these forces the generation of plants and animals, and imagined the 
epoch anterior to life as that of primitive chaos and struggle of the 
elements. " If we do not consider," says Humboldt, " the study of phys- 
ical phenomena so much as bearing on our material wants as in their 
general influence upon the intellectual progress of humanity, it will be 
found that the highest and most important result of our investigation 
will be the knowledge of the intercommunication of the forces of na- 
ture, and the certainty of their mutual dependence upon each other. 
It is the perception of these relations which enlarges the views and en- 
nobles our enjoyment of them. This enlargement of the view is the 
result of observation, of meditation, and of the spirit of the age in which 
all the directions of thought concentrate themselves. History teaches 
him who can travel back through the strata of preceding centuries to 
the farthest roots of knowledge how, for thousands of years, the human 
race has labored to grasp, through ever-recurring changes, the fixity of 
the laws of nature, and to gradually conquer a large portion of the 
physical world by the force of intelligence." 

The most important result of a rational examination of nature is, that 
it leads one to comprehend unity and harmony in this immense assem- 
bly of things and forces, to embrace with equal ardor what is due to the 
discoveries of past ages and to those of our own time, and to analyze 
the details of phenomena without succumbing beneath their weight. 
It is thus that it has been given to man to show himself worthy of 
his high destiny, by penetrating into the meaning of nature, unveiling 
its secrets, and mastering by thought the materials collected by obser- 


Wc may now coiitcmpLate our planet traveling in space, and keeping 
about it the aerial envelope which adheres to its surface. Our imagina- 
tion can easily comprehend the general shape of this gaseous sphere 
which encircles the solid globe, and which is comparatively thin and of 
slight bulk. 

The exterior surface of the atmosphere is therefore curved like that 
of the sea, for, like water, the external layer of air tends to a level, all 
points of which are at equal distances from the centre. To the eyes of 
novices, it seems difficult to reconcile the idea of the spherical surface of 
the ocean with what is commonly termed a level; the idea that the air 
has a horizontal level like water, and that, like an aerial ocean, this 
level is always tending to an equilibrium, seems at first sight somewhat 
obscure. Nevertheless, not only does the air possess to an unlimited 
degree all the properties of elasticity and mobility of a fluid seeking 
equilibrium, but, different in this respect from water and other liquids, 
it is extremely capable of compression and, consequently, susceptible of 
extreme expansion. These are facts which must always be kept in 
mind, for they will assist in the understanding of a great number of at- 
mospheric conditions explained in future chapters of this work. 

What, then, is the thickness of this gaseous stratum which envelops 
our globe? This is the point which we shall examine in the next 

To aiscertain the height to which the atmosphere extends, it would be 
necessary to calculate the density of the air at different elevations in the 
average state, leaving out of consideration accidental disturbances. This 
can be done when we know the temperature of the air, its pressure, and 
the tension of the vapor of water which it contains. It would, further, 
be necessary, in order to obtain an exact determination, to take account, 
first, of the gradual diminution in weight as the distance from the cen- 
tre of the earth is increased ; secondly, of the variation in the centrifu- 
gal force according to the latitude. These variations are, however, slight, 
and scarcely affect the calculation, in consequence of the coat of air being 
of such insignificant thickness as compared to the radius of the globe. 

The height of the atmosphere has its limits, which, as we shall see, are 
somewhat confined. If the air had no elasticity, its limit would be at a 
distance where the centrifugal force was in equilibrium with the weight; 
but as this condition does not exist, its elasticity must necessarily be coun- 
terbalanced by a force of some kind, and this force is the weight of the 
strata of air which are above the particular one we are considering. But 


the higher we ascend the more rarefied does the air become, and when 
the last strata are reached there is nothing to keep them down. Nev- 
ertheless, the atmosphere being limited, as we shall presently see, these 
strata can not be lost in space ; and it is probable that in consequence of 
their rarefaction and the great decline in their temperature, their phys- 
ical condition is so modified that the elastic force becomes nil. Laplace 
has pointed out this indispensable condition ; Poisson has specified it, by 
showing that the equilibrium would still be possible with a very consid- 
erable limiting density, provided that the fluid was not capable of ex- 
pansion ; and Biot, who has summed up these conditions, clearly indi- 
cates the state of these external inexpansible strata in his remark that 
they must be like " a liquid which does not evaporate." We will now 
examine the mechanical and physical conditions of this aerial envelope, 
estimate its exterior shape, and measure its height. 




As the earth travels in space with enormous swiftness, carrying along 
with it, adhering to its surface, the gaseous body that encircles it, it 
naturally follows that this latter does not extend indefinitely into space, 
but ceases to exist at a certain distance from the surface. How far 
can it extend ? Carried along by the rotation of the globe in its daily 
movement, we may conclude that at a certain height above the ground 
the movement of the atmosphere is so rapid that the centrifugal force 
which it acquires would hurl into space the outside particles of air, 
which would then cease to adhere to the surface and, for the same rea- 
son, to form part of the atmosphere. 

Certain inventors of methods of aerial navigation have vaguely im- 
agined that the atmosphere does not entirely turn round with the earth, 
so that, by rising to a certain height, we could see the globe moving 
around beneath our feet, and should only have to wait until the me- 
ridian, where we wished to alight, passed under the balloon, to find 
ourselves transported there by the rotation of the globe. Such an idea 
is, of course, absurd, as the atmosphere and all that it contains partake 
equally with the earth in the rotation of the latter. 

The centrifugal force increases as the square of the velocity, and at 
the equator its amount is -^ig- part of that of gravity, so that a body at 
the equator weighs less than the same body at either of the poles by 
•jj-gir of its weight. If, therefore, the earth rotated on its axis seventeen 
times as fast as it does, since seventeen times seventeen is equal to tw^o 
hundred and eighty-nine, a body at the equator would not have an}^ 
weight. A stone, for instance, detached from the ground by the action 
of the hand, would not fall down again ; we should become so feath- 
er-like, that, in dancing upon the surface, we should resemble aerial 
nymphs displaced by the wind. As the circumferences of circles vary 
as their radii, at seventeen times the distance from the surface to the 
centre of the earth — that is to say, at a height of about sixteen times 
the radius of the earth, or about 63,000 miles— if the other quantities 
involved remained unchanged, the atmosphere would cease to rotate 



with the earth; but, in point of fact, the weight does not remain un- 
changed, but diminishes as the distance from the centre of attraction is 

By combining this diminution with the increase of centrifugal force, 
we find that at a distance of about 6'61 times the radius of the earth 
from its centre, which corresponds to a height above its surface of about 
21,000 miles, the centrifugal force is equal to the weight, and conse- 
quently the aerial particles which might happen to be in these regions 
must of necessity escape. This is the distance at which a satellite 
would gravitate in exactly twenty-three hours fifty-six minutes, the 
time occupied by our planet in its rotation. It is, theoretically^ the "max- 
imum limit of the atmosphere, which, however, as a matter of fact, is 
far from extending to so great a height, as we shall see; but, mathe- 
matically, it might do so, and it is only at this enormous distance that 
the centrifugal force would be sufficiently great to prevent the atmos- 
phere from existing as such. 

Such is the extreme and maximum limit of the atmosphere; but it is 
at a far lower elevation that the air we breathe really ceases. Thus, at 
the height of 10,000 feet — the height of Mount ^tna — there is beneath 
the mountaineer nearly a third of the aerial mass; at 18,000 feet, which 
is less than that of the peaks of many mountains, the column of air 
which presses upon the soil has already lost half its weight, and conse- 
quently at this point the whole gaseous mass, which reaches far up into 
the sky, does not weigh more than the strata which are compressed into 
the region below. 

In consequence of the forces that act upon it, the shape of the atmos- 
phere is not absolutely spherical, 
but swollen out at the equator, 
where it is much higher than at 
the poles. The maximum limit 
of this figure, in the case where 
the flattening is greatest, has been 
given by Laplace. The diameter 
of the atmosphere at the equator 
is a third greater than at the 
poles.* It is the mathematical 
limit, beyond which the terrestrial 

Fig. 1.— Mathematical limit of the shape of the at- 

* [This is inaccurate. Laplace proved that the ratio of the least (the polar) diameter to the 
greatest (the equatorial) diameter could not be less than f (not f , as in the text). Fig. 1 is 


atmosphere can not pass. But it has not this exaggerated shape, 
though in reality it is perceptibly denser at the equator than at the 
poles. It may be remarked that it is probable that a detached train 
of the lighter gases remains constantly in the rear of the globe during 
its rapid revolution around the sun. It need scarcely be added that 
the shape of the atmosphere undergoes further change, owing to the 
atmospheric tides, which are due to the varying attraction of the sun 
and the moon. 

The decreasing weight of the atmospheric strata affords us the first 
means of calculating a minimum limit of the height of the atmosphere. 
Mechanics have given us the maximum limit, and it is in this instance 
to physics that we shall have recourse. 

Consider a vertical column of air, then the pressure at any point 
must be equal to the weight of air above ; or, in other words, any por- 
tion of the column measured from the ground supports all the rest of 
the column above; the lower strata of the atmosphere are, therefore, 
more pressed down (and consequently denser), because they have a 
greater weight resting on them. The barometer, which measures this 
pressure of the air, is higher at the foot than at the summit of a mount- 
ain ; and the relation which exists between the pressure and the height 
is so close, that the difference in level between two points may be de- 
duced from the difference in the heights of the barometrical columns 
simultaneously placed at these two stations. The smaller the pressure 
the more dilated is the air ; so that, at first sight, it would seem as if the 
atmosphere mus't extend to an immense distance. 

A celebrated natural philosopher, Mariotte, first determined the law 
of the compression of gases ; and the result of his researches shows that 
the quantity of air contained in the same volume — or, in other words, 
the density of the air — is proportionate to the pressure to which it is 
subjected. Until within the last few years this law was considered en- 
tirely accurate ; but recently it has appeared most difficult to conceive 
why the terrestrial atmosphere does not extend very far into space ; 
while other considerations indicate that it is necessarily limited, and 

therefore incorrectly drawn, as the protuberance should be considerably greater. It may be 
mentioned that one consequence deduced by Laplace from his result is, that the Zodiacal light 
can not be produced by reflection on the atmosphere of the sun, as the former always appears 
in the form of a thin lens, the ratio of the polar to the equatorial diameter being much less 
than |. Laplace's investigation is given in vol, ii., pp. 194-197 of the Mecanique Celeste 
(National Edition).— Ed.] 


ceases at a short distance above the ground. This apparent contra- 
diction was the result of a too extensive generalization of Mariotte's 
law, which is simply relative instead of rigorously definite; and Reg- 
nault has studied the differences which exist between the theoretical 
law and the facts of the case. 

Subsequently to these investigations, M. Liais has ascertained, by in- 
troducing very small portions of air into a large barometrical instru- 
ment made for the purpose, that the differences between the results of 
observation and the theory usually adopted are still greater. By di- 
minishing sufficiently the quantity of air, it has been possible to find a 
limit at which the particles, far from separating from each other, as 
would happen were the gases capable of indefinite dilatation, seem, on 
the contrary, to have a mutual tendency to adhesion similar to that of 
the molecules in a viscous liquid. The elasticity of the air, producing 
expansion, ceases, therefore, at a certain degree of dilatation, from which 
point this gas assumes the character of a liquid, but a liquid out of all 
comparison lighter than those with which we are acquainted. 

By means of this decrease in the density of the air in proportion to 
its height, Biot has, by an examination of the physical conditions of 
equilibrium and a complete discussion of the observations obtained at 
different degrees of altitude by Gay-Lussac, Humboldt, and Boussin- 
gault, demonstrated that the minimum height of the atmosphere is 
160,000 feet, or about thirty miles. At that height the air must be as 
rarefied as beneath the exhausted receiver of an air-pump ; that is to 
say, as rarefied as the air in the nearest approach to a vacuum that we 
can make. 

Thus the minimum height of the atmosphere is thirty miles, and the 
maximum 21,000. Hence we have two defined limits, but with a great 
distance between them. There are, however, other methods by which 
we can get nearer to the truth. Efforts have been made to measure the 
height of the atmosphere optically, by studying the length of the twi- 
light and the length of time during which the solar rays continue to 
reach the aerial regions when the luminary himself has sunk below the 

If the atmosphere were unlimited, the phenomenon of night would 
be entirely unknown to us; the light of the sun, reaching the strata of 
air which are sufficiently distant from the earth, would be continuously 
sent on to us by reflection from these strata. On the other hand, the 
absence of any aerial envelope would cause the night to begin exactly 



at sunset, and the light of day to burst upon us immediately the sun 
rose. As it is, every one knows that the twilight of evening and the 
morninfT dawn prolong the time during which we enjoy the solar light. 
It will be readily imagined that the observation of these phenomena at 
once suggested the idea of seeking to resolve, by their agency, the 
height to which the atmosphere extended. 

Suppose the earth to be represented by the circle, radius o A, and 
that its atmosphere is limited by the circumference F G H i c. It is 
evident that, when the sun has sunk beneath the horizon f a c b of the 

place A, it will only give 
light to a portion of the 
atmosphere. Thus, when 
the sun arrives at j, if 
we imagine a tangent 
cone to the earth, hav- 
ing the sun for its sum- 
mit, all those parts of 
the atmosphere situated 
below J G will be de- 
prived of light, and the 
part c I H G will alone 

■Measure of the height of the atmosphere, according to be Illuminated. Later 
the length of twilight. , , , 

on, when the sun reaches 
.j', the portion bounded by c i H will alone be subject to its light; 
later still, only from c to i ; and finally, when the sun gets to j'", upon 
the tangent line from c, the intersection of the plane of the horizon 
FACE and the limiting sphere of the atmosphere, the twilight ceases. 
From the moment, therefore, that the sun sets, we ought to see a sort 
of arc appear on the opposite side of the horizon, rising gradually until 
it reaches the zenith, and then slowly descend until it finally disap- 
pears. Such is the theory that the earliest astronomers conceived as 
to the phenomenon of twilight. In the optics of Alhasen (in the tenth 
century) we find that the angle of the sun's declivity for the close of the 
twilight or the break of dawn was taken as 18°, and this estimate is 
still adopted by modern astronomers as the average amount. 

In our climate it is difficult to distinguish with accuracy the limit of 
separation between that part of the atmosphere which is lighted by the 
sun and that which does not receive its rays directly. But Lacaille, in 
his voyage to the Cape of Good Hope, recognized all the phases which 



have been enumerated theoretically. He says: "Upon the 16th and 
17th of April, 1751, while at sea and in calm weather, the sky being ex- 
tremely clear and serene, at the point where I could distinguish Venus 
at the horizon as a star of the second magnitude, I saw the twilight ter- 
minated in the arc of a circle as regularly as possible. Having regula- 
ted my watch by the exact hour, according to sunset, I saw this arc lost 
in the horizon, and I calculated, by the hour at which I made this ob- 
servation, that the sun had descended below the horizon, on the 16th 
of April, 16° 88', and on the 17th, 17° 13'." 

Other observations have since been made, as we shall see further on. 

It is easy to understand that, once having ascertained the apparent 
daily circle described by the sun upon a certain date, and the position 
of the observer upon the earth, we can calculate, by the time that has 
elapsed between the hour of sunset and the moment of the crepuscular 
arc's disappearance, the angle traversed by the sun below the hori- 
zon. It will also be understood that, according to the time and place, 
there will be found a difference both in regard to twilight and dawn, 
since the variations in the relative position of the sun and the state 
of the air must necessarily influence the direction and quantity of the 
light which, &fter countless reflections and refractions, reaches the ob- 

We will study, in the second book, the optical effects of twilight; at 
present we are only concerned with the relation existing between its 
duration and the height of the atmosphere. 

Now, the time during which the sun, after sinking below the horizon 
of a particular spot, continues to give light directly to part of the at- 
mosphere visible from this place, depends upon the thickness of the 
aerial strata which envelop the earth. Let us suppose, for instance, 
that we pass a plane (Fig. 2) through the place A, the centre, o, of the 
earth and the centre of the sun ; this plane will cut the earth in the cir- 
cle o A. Let F A B be the intersection of the horizon of the spot A with 
this same plane ; from c draw the tangent c D to the earth ; all that 
part of the atmosphere visible at A will cease to be illuminated by the 
sun when, in its apparent diurnal movement, it has sunk below CD j'". 
Now we have seen that, from the duration of twilight, it was concluded 
that it came to an end when the angle BC j'" of descent below the ho- 
rizon was 18°. As the angle o A c is a right angle, and as o A is the 
radius of the earth, we know one side and the angles of the triangle 
o AC, and consequently are enabled to calculate the other parts, oc 




may therefore be regarded as known, and thence it results that we have 
the height, E c, of the atmosphere, for e c = o c— o E. 

Such is the method devised by Kepler for deducing the height of the 
atmosphere from the phenomena of twilight. The results which it has 
furnished agree with the preceding, and give our atmosphere a height 
of from thirty to thirty-seven miles * The average radius of the earth 
being 3908 miles, it will be seen that this height is but a little more 
than the 130th part of this radius ; that is to say, that if the earth were 
represented by a sphere about twenty-two feet in 
p4 diameter, the atmosphere would be like a coat of 
vapor adhering to the surface, with a thickness of 
about one inch. 

Figure 8 represents exactly this relation. It 
shows — firstly, the incandescent interior of the 
globe, which is a; secondly, the solid crust, 6, on 
which we live (it is but twelve leagues, or thirty 
miles, thick, as, in consequence of the increased tem- 
perature of one degree (Fahrenheit) for fifty feet, 
minerals fuse at this depth) ;f thirdly, the thickness 
of the aerial layer which we breathe, ^nd which is 
represented by c; and, fourthly, the probable height 
of a very light atmosphere, c/, over and above ours, 
of which we are about to treat. 

It may be further mentioned, in reference to the 

ITig. 3 — Section showinf; '' 

the relative thickness of measurement of the height of the atmosphere by 

the earth's crust, of our , , . f ^^■ ^ ^ • i i 

atmosphere, aud of a the duration of twilight, that Certain ODscrvcrs bavc 
higher atmosphere. obtained, as the rcsult of similar researches, an ele- 
vation much greater than that given above, affording a clear proof that 
the twelve leagues actually represents the minimum only. M. Liais has 
made a direct calculation of this height by observing the duration of 

* [It is to be noted that different methods give different heights for the atmosphere, but 
there is no discrepancy, as different things are meant. Thus, if experiments on twilight give 
forty miles as the height, this implies that the air above this elevation reflects no appreciable 
amount of light ; while, if we define the height to be to the point where the friction will not 
set light to a meteor, we have about seventy miles; but, of course, there is no reason why there 
should not be some air at much greater heights. — Ed.] 

t [This is the observed rate of decrease at the surface of the earth, but it is not true that 
the thickness of the crust must be as stated in the text. It follows, from several considera- 
tions of other kinds, that the thickness of the crust is in all probability not less than 600 
miles. — Ed.] 


twilight and of the crepuscular curve, which colors the skj with that 
lovely rose tint which is so remarkable, especially in southern countries. 
These observations have been made both on the Atlantic, during a 
voyage from France to Rio Janeiro, and in the bay upon the shores 
of which the last-named city stands. They give, as a minimum, 180 
miles, and, as a probable height, 204 miles. 

By observing, from the summit of the Faulhorn, the course of the 
crepuscular arcs, Bravais obtained a height of seventy-one and a half 
miles. The height, however, varies according to the temperature of the 
seasons, and remains always greatest at the equator. Another method, 
different from the preceding, consists in measuring the thickness of the 
penumbra which surrounds the earth's shadow on the moon during lu- 
nar eclipses, as well as the phenomena of refraction produced. This 
measurement gives from fifty to sixty miles as the thickness of the 
terrestrial atmosphere, the influence of which is felt under this special 

The observations which accord the atmosphere a height far greater 
than the theoretical thirty-eight miles have been for many years the 
object of special discussion. Quetelet, director of the Brussels Observa- 
tory, has, after much research on this head, arrived at the conclusion 
that it does indeed extend much higher than had been supposed, but 
that the upper strata are not quite of the same nature as nearer 
the earth. 

This addition is supposed to be due to an ethereal atmosphere, very 
rarefied and differing from the lower atmosphere in which we live. It 
is the region where are mostly seen the shooting stars, which afterward 
disappear when they reach the terrestrial atmosphere. 

The upper atmosphere* is still, the lower in continual motion. The 
special movements caused by the action of the winds and tempests are 
limited in their height by the effect of the seasons. Thus, as regards 
our climate, the agitated portion, in the vicinity of the earth, would not 
be more than from seven to ten miles high during the winter, while its 
height must be almost double in summer. All that part of the atmos- 
phere which is above the latter would only experience a very slight 
and scarcely sensible movement, arising from the movable basis upon 
which it reposes. 

The continual disturbances going on in the lower regions cause the 
air in the inferior atmosphere to be very much alike in its chemical 
* [The existence of such an atmosphere seems to me very uncertain. — Ed.] 


components. No difference has been discovered at the various eleva- 
tions which it is possible to attain for the purpose of collecting air and 
submitting it to analysis. 

In the upper atmosphere the phenomena, of which we are scarcely 
able to form an idea by judging them from the surface of our globe, 
take place. There, also, appear the shooting stars; descending from 
a still greater height, the aurora borealis, and those mighty luminous 
phenomena which we often witness without having the power to sub- 
mit them directly to the test of experiment. All these facts do not es- 
cape us altogether, especially as regards the aurora borealis and the 
mao-netic phenomena. If we can not determine the cause, we can at 
least feel the effect with suflScient force to be in a position to appre- 
ciate them. 

Sir John Herschel, De la Eive, and Hansteen seem to share upon this 
point the opinion of Qaetelet. We can quite admit that, above our at- 
mosphere ot oxygen, nitrogen, and vapor of water, there exists an at- 
mosphere excessively light, which may extend two hundred miles in 
height, and which is naturally composed of the very lightest gases. 

The terrestrial globe being about 8000 miles in diameter, this total 
thickness represents the fortieth of the globe's diameter. The simulta- 
neous existence of these two atmospheres is, therefore, the general con- 
clusion at which we will, momentarily at least, stop. 

As to the basis of the atmosphere, we may now inquire if it ceases 
at the surface of the ground, and does not descend into the interior of 
the globe itself. 

Pressing upon all bodies upon the surface of the earth, it tends to 
penetrate in all directions between the molecules of liquids as into the 
interstices of the rocks. It is to be found in water as in all vegetables 
and all organic structures ; the earth and the porous stones are impreg- 
nated with it, and that in proportion to the force with which it presses. 
It will be seen, therefore, that the air is not limited to the part which 
is, so to speak, a gaseous envelope, and that a sensible fraction of its 
constituent elements penetrates the waters of the ocean and the inter- 
stices of the ground. Certain savants have imagined that the air of 
which the atmosphere is composed is but the continuation of an inte- 
rior atmosphere ; but the rise in the temperature, due to the central 
heat, would prevent the condensation of gases, and must limit the pres- 
ence of air in the under strata. 

A rough estimate of the quantity of air which is thus introduced into 


tlie waters of the ocean may be formed by measuring the absorption of 
gases by various liquids. Under ordinary pressure, sea-water absorbs 
from two to three per cent, of its volume, only the proportion of oxygen 
is much greater than in the ordinary air. The result of the calculation 
is, that the quantity of air absorbed by the ocean is not above a three-- 
hundredth part of the atmosphere. 

We thus have a tolerably complete determination both as to the 
height and shape of this terrestrial atmosphere. 




While treating of the height of the atmosphere, we have already seen 
that the air is denser in the lower regions of the aerial ocean — that is to 
say, near the surface of the earth — than in the higher regions. The air, 
light and unsubstantial as it may appear to us to be, has consequently a 
positive weight. Each square foot of the earth's surface sustains a con- 
siderable pressure, the amount of which we shall presently attempt to 
estimate, corresponding to the height and density of the column of air 
above it. 

Our ancestors were not able to measure the atmospheric pressure; but 
we must not conclude from this that they were ignorant of the effects 
which it exercised, especially when the wind was violent. Yet this 
force, which every one felt without being able to measure, was not ren- 
dered determinate until the middle of the seventeenth century. 

In 1640, the Grand Duke of Tuscany having ordered the construc- 
tion of fountains upon the terrace of the palace, it was found impossible 
to make the water rise more than thirty-two feet. The duke wrote to 
Galileo in reference to this strange refusal of the water to obey the 
pumps. Torricelli, the pupil and friend of Galileo, gave the true ex- 
planation of the fact, and proved, as we shall see, that this column of 
water of thirty-two feet was in equilibrium with the weight of the at- 

The celebrated invention of Torricelli has sometimes been erroneous- 
ly attributed to Pascal. The French philosopher himself alludes to the 
mistake, and shows how much of the merit is due to him in the follow- 
ing terms: "The report of my experiments having been spread abroad 
in Paris, they have been confounded with those made in Italy ; and, 
thanks to this misunderstanding, some, according me an honor to which 
I can lay no claim, attributed the Italian experiment to me, while oth- 
ers unjustly deprived me of the credit of those to which I was really en- 
titled. To give to others and to myself the justice due to us, I published, 
in 1647, the experiments which I had made the year before in Norman- 



dy ; and tbat they might not be confounded with one made in Italy, I 
gave the latter separately and in italics, whereas mine were printed in 
Eoman letters. Not content with giving it these distinctive marks, I 
have stated in so many words that I am not the inventor of the barom- 
eter; that it was made in Italy four years previously, and was the cause 
of my making similar experiments," 

It was, then, the refusal of the water to rise more than thirty-two feet, 
in obedience to the pumps, which revealed to Torricelli the fact that the 
atmosphere had weight, and that its whole weight was balanced by a 
column of water thirty-two feet in height. Let us then examine for a 
moment the mechanism and action of the pump. 

Every one knows that these simple and old-fashioned contrivances 
serve to raise water either by suction or pressure, or by both combined. 
Hence their classification as suction-pumps ., fcnxing-pumps^ and suction 
and forcing pumps. Before Galileo's day, the ascension of water in the 
suction-pump was ascribed to the fact of nature abhorring a vacuum ; 
but it is, in reality, merely an effect of atmospheric pressure. 

Take a tube, at the lower extremity of which is a piston, and place 
this lower end in water. If the piston is 
drawn up, a vacuum is created below, and the 
atmospheric pressure, acting upon the surface 
of the liquid external to the pump, makes it 
rise in the tube and follow the movement of 
the piston. 

Herein lies the principle of the suction- 
pump, which is essentially composed of the 
body of the pump, in which a piston moves, 
communicating by a tube with a reservoir of 
water (see Fig. 4). At the point where the 
body of the pump and the suction-tube join is 
placed a valve, opening upward, and in the 
body of the piston there is an opening formed 
by a similar valve. 

For water to reach the body of the pump, 
the suction-valve must be less than thirty-two 
or thirty-three feet above the level of the wa- 
ter in the well, otherwise the water would 
cease to rise at a certain point in the tube. Fig. 4.-suction-pump. 
and tne motion of the piston would be unable to raise it any farther. 



In addition, to insure raising at each ascent of the piston a volume 
of water equal to the volume of the body of the pump, the spout must 
be placed at a less height than thirty-two feet above the reservoir. 

Thus the suction-pump will not raise water 
to a height of more than thirty-two feet ; but 
the water having once passed above the pis- 
ton, the height to which it can then be raised 
depends solely upon the force which drives 
the piston. 

The suction and force pump (see Fig. 5) 
raises water both by suction and pressure. 
At the base of the body of the pump, over the 
orifice of the suction-pipe, is, as before, a valve 
opening upward. Another valve, also open- 
ing upward, closes the aperture of the bent 
tube, which runs into a receptacle called the 
air-vessel.* Then from this reservoir there 
starts a pipe which serves to raise the water 
to the required height. Finally, the force- 
pump only acts mechanically, and does not 
utilize atmospheric pressure. It differs only 
from the other in that it has no suction-pipe, its body going right 
into the water which is to be drawn up. 

In reference to this elevation of the water only to a certain height, 
Torricelli, throwing aside, like his master, all idea of a hidden cause, 
shoived that the pressure of the air compels the water to mount iqj into the 
pipe from which the air is withdrawn, until the weight of water raised 
into the pipe is equivalent to that of the air which presses upon an 
equal section of the reservoir from which the water is being raised. By 
the aid of this principle he was led to invent the barometer. To exer- 
cise equal pressures, the liquid columns must be of heights inversely 
proportional to their density. Thus, a liquid twice as heavy as water 
would, with a column of sixteen feet, be in equilibrium with the atmos- 
phere ; and quicksilver, which is nearly thirteen and a half times as 
heavy as water, would be in equilibrium if the height of the column 
were diminished in this proportion — that is, to about twenty-nine inches. 

* [The air-vessel is not essential to the principle of the pump ; if it wei-e not used the supply 
of water would be intermittent, as in the common suction-pump, but the effect of the elasticity 
of the air in the air-vessel is to render the stream of water continuous. — Ed.] 

Fig. 5.— Suction and forcing pump. 

Fig. C. — Torricelli iuveutiug the Barometer. 



This conclusion is easily verified. Take a glass tube, three feet in 
length, and open only at one end ; fill it with quicksilver, and then, 

placing the finger on the open end (see 
Fig. 7),, put the lower portion of the 
tube into a basin filled with the same 
liquid, with the end closed by the fin- 

rig. T.— The tube full of quicksilver. 

Fig. 8 — Tlie tube in the basin. 

ger downward. Immediately the finger is removed, the quicksilver 
inside will descend several inches and then stop (see Fig. 8 ). The 
equilibrium is established, and the liquid column which remains sus- 
pended in the pipe is a true balance, for the weight of the column of 
mercury is exactly in equilibrium with the atmospheric pressure. 

Torricelli gave to this tube of quicksilver, thus placed vertically in a 
basin of quicksilver, the name of Barometer; that is to say, a contriv- 
ance to indicate the weight of the air, from the Greek (iapog, weight, 
and fxirpov, measure. Its invention by Torricelli dates from 1643. 
Three years later, Pascal repeated the experiment in France with a wa- 
ter-barometer, and even a wine-barometer. This was at Eouen. His 
tube was forty-nine feet long, and to avoid the difficulty, insurmounta- 


ble in that day, of exhausting the air in it directly, he had it sealed at 
one end, filled it with wine, and closed the other end with a cork. 
Then, by means of cords and pulleys, the tube was placed upright and 
the lower end put into a vessel full of water. As soon as the cork that 
kept it closed was removed, the whole liquid column in the tube fell, 
until its surface was about thirty-three feet above the level of the water 
in the vessel. The remaining sixteen feet above were destitute of air. 
Consequently, the liquid column itself formed an equilibrium to the at- 
mospheric pressure, and from this he drew the conclusion that a column 
of water (or of wine of the same density) thirty-two feet high weighs as 
much as a column of air on the same base. 

The surface of the earth is pressed upon as if it was covered with 
a body of water thirty-two or thirty-three^ feet deep, and we who live 
upon the bed of this ocean of air undergo the same pressure. 

If it is the pressure of the air which causes the elevation of the quick- 
silver or the water, as we ascend into the atmosphere, the weight of the 
column of quicksilver raised, and consequently the height of this col- 
umn, must gradually diminish in a manner dependent on the strata of 
air left beneath it. The experiment was made on the Puy-de-D6me, ac- 
cording to the instructions of Pascal, by his brother-in-law, Florin Pe- 
rier, upon, the 19th of September, 1648, and repeated by Pascal himself 
on the Tour St. Jacques at Paris. The results were decisive, and the 
barometer became an easy and accurate means of measuring the total 
weight of the atmosphere, and the variations in the pressure which it 
exerts at different times and places upon the surface of the globe. We 
thus see that it was between 1643 and 1648 that the atmospheric press- 
ure was demonstrated by the construction of the barometer and the ex- 
periments which its discoverers at once entered upon. 

By a coincidence not at all unusual in the history of science, while 
the indications of the barometer were being studied in Italy and 
Prance, experiments were being made in Holland to ascertain the pre- 
cise weight of the air, but by quite a different process. 

In 1650, Otto de Guericke, burgomaster' of Magdeburg, invented the 
air-pump, by which the air may be exhausted from any receptacle and 
a nearly absolute vacuum created. 

The ingenious inventor conceived in the same year the idea of weigh- 
ing a globe of glass, first leaving in it the air which it contained, and 
then weighing it again when the air had been removed by the air- 
pump. The globe, when emptied of air, was found to be less heavy 



by about one-third of a grain for every cubic inch of the globe's ca- 

Aristotle had long before suspected that air had weight, and to make 
sure of the fact, he weighed a leather bottle, first empty and afterward 
when inflated with air; for, he remarked, if the air has weight, the 
leather bottle will be heavier when weighed the second time than it 
was the first time. The experiment not confirming his supposition, he 
concluded that the air had no weight. Nevertheless, several of the 
ancient philosophers admitted the material nature of air as a fact. 
Thus the Epicureans compared the efiects of the wind with those of 
water in motion, and considered the elements of the air as invisible 
bodies. During the reign of the peripatetic philosophy, however, it 
was assumed that air was without weight, and there were but few 
philosophers who did not share this erroneous opinion. 

We have seen that, by repeating judiciously the experiment of Aris- 
totle, Otto de Guericke demonstrated the real weight of air. If Aris- 
totle's experiment led to a contrary result, it must be attributed to the 
change in the volume of the leather bottle during his two trials, for 
every body, when weighed in a fluid, loses in weight a quantity equal 
to the weight of the fluid displaced. 
The leather bottle made use of by Aris- 
totle would have shown an increase of 
weight if weighed in a vacuum. Let 
us suppose that about 1835 cubic inch- 
es of air were introduced into it by in- 
spiration ; its weight would have in- 
creased by about 550 grains, but at the 
same time the bottle would become in- 
flated, and its volume, being increased 
by 1835 cubic inches, would have dis- 
placed a volume of air of equal weight, 
so that its loss in weight would, be also 
550 grains, and the weight .of the air 
and bottle together would consequent- 
ly remain the same as before. But in 
the experiment of Otto de Guericke 
the globe was always of the same size, 

whether empty or full of air, and its Fig- 9— otto de Guericke's experiment. 

loss in weight through the displacement of the air being in each case 


the same, there was, of course, a difference, which proved that air had 
weight. Otto de Guericke, at the same time, conceived the idea of the 
Magdeburg Ilemispheres, so called from the town in which he invented 
them, and which consist of two hollow hemispheres of copper, with a 
diameter of from four to five inches. The hemispheres fit each other 
hermetically. One of them has attached to it a cock that screws on 
to the plate of an air-pump, and the other a ring which acts as a 
handle to move it backward or forward. As long as the two hem- 
ispheres, when in contact, contain air within them they can easily be 
separated, for there is equilibrium between the expansive force of the 
interior air and the outside pressure of the atmosphere, but when 
once a vacuum is formed by the exhaustion of the air, it requires a 
considerable effort to draw them apart. 

In one of these experiments, the learned burgomaster had each 
hemisphere pulled by four strong horses without succeeding in sep- 
arating them. The diameter was more than two feet, which gives a 
total of more than three and a quarter tons as the atmospheric pressure 
brought to bear in the way of resistance. 

The pressure of the atmosphere on a square inch is equivalent to the 

weight of a column of quicksilver with 
a volume of 29"92 cubic inches, viz., 
about fifteen pounds. 

It is easy and interesting to draw 
from this the conclusion that, as the 
superficies of an average human body 

Fig. 10._The Magdeburg Hemispheres. -^ g-^^^^^^^ ^^^^^,^ feet, WC may Cach 

of us be said to be subject to a pressure of about fifteen tons. 

That we are not crushed by this enormous pressure, is because it does 
not all press vertically down on us. As the air surrounds us on all 
sides, its pressure is transmitted over our body in all directions, and, 
in consequence, becomes neutralized. Air penetrates readily and with 
full pressure into the profoundest cavities of our organism; hence we 
have the same pressure inside and outside, and thus these weights be- 
come exactly balanced. This is easily proved by the experiment of 
bursting a bladder under the receiver of an air-pump. Take a cylin- 
drical glass vessel, hermetically closed at the upper end by a piece of 
gold-beater's skin, with the other end placed (see Fig. 11) on the plate 
of an air-pump ; as soon as the air begins to be exhausted from the 
vessel, the gold-beater's skin becomes depressed under the influence of 



the atmospheric pressure upon it from above, and soon bursts. The 
opposite result occurs if the pressure from out- 
side is lessened. If a bird is placed in the vac- 
uum of an air-pump, its body will be seen to 
swell, its blood to spurt out with violence, and 
in a short time it perishes, a victim to a kind of 
explosion the inverse of that just described. 

This fact is confirmed, as we shall see farther 
on, by the ascents that have been made to great 
elevations. Upon reaching the regions where 
the air is much rarefied, the limbs swell, and the 
blood has a tendency to force its way through the 
skin, in consequence of the want of equilibrium 
between its own tension and that of the external 

Fig. 11.— Atmospheric press 
ure; rupture of equilibrium. 


Any one can show the effect of atmospheric pressure by a very 
simple experiment. This consists in filling a glass 
quite with water and laying over the top a sheet of 
paper. It can then be turned over without spilling 
any of the liquid, a fact which must be attributed 
to the pressure which the atmosphere exercises 
upon the sheet of paper. 

It was stated above that, where a vacuum is 
created, the atmospheric pressure is about fifteen 
pounds to the square inch. It is this pressure 
which causes the limpet to adhere to the rock, ^^''Tnirt^tred 
when this mollusk has by contraction created a s'^^^- 
vacuum under its shell. The fly, excluding the air from between its 
feet and the ceiling, is enabled, apparently, to violate the laws of grav- 
ity. Cupping-glasses, when applied to the body, act on this same 
principle, and we can not take a step without observing some fact 
which is founded on the effects of atmospheric pressure. Such are 
the general facts and experiments which demonstrated that the air 
had weight, and gave birth to the instrument wherewith this weight 
was to be determined, viz., the barometer. It now remains to apply 
these ideas to the whole atmosphere, the extent of which we endeav- 
ored to "explain in the preceding chapter. 

* [I have neither experienced any of these symptoms myself, nor have I observed them in 
others. — Ed.] 


At the level of the sea the pressure, upon the average, sustains the 
barometrical column at a height of about 29-92 inches. 

Experiments frequently repeated by physical philosophers— and the 
accuracy of which has been verified — have proved that the weight of 
the air at 32° (Fahr.) of temperature, and under a pressure of 29-92 
inches of mercury, is to the weight of an equal volume of quicksilver 
in the proportion of unity to 10,509 — that is to say, that 10,509 cubic 
inches of air have the same weight as one cubic inch of mercury. If 
the density of the strata of air were everywhere the same, it would be 
easy to deduce from the above result not only the height of a given 
spot by the aid of the barometer reading there, but also the total height 
of the atmosphere. It is, indeed, evident that if a fall of an inch in the 
height of the barometer corresponded to a change of height of 10,509 
inches, a fall of 29*92 inches, which is the total height of the barome- 
ter, would correspond to 29*92 times 10,509 inches — that is, about five 
miles. Such would be the height of the atmosphere if its density re- 
mained the same from top to bottom, but we have seen that its lower 
strata are denser than the higher. It follows, therefore, that, to pro- 
cure a fall of an inch in the mercury of the barometer, it is necessary 
to traverse a greater distance above the level of the ground or the sea. 

Halley was the first to deduce a formula by which heights might be 
obtained by means of the barometer. 

We have seen in the previous chapters that, since the experiments 
of Mariotte, it has been recognized that air becomes compressed in pro- 
portion to the weight above, or to the pressure exerted upon it. 
Thence it is inferred that, in rising vertically in the atmosphere to suc- 
cessive elevations, increasing in arithmetical progression, the density 
of the corresponding strata of air would diminish in geometrical pro- 
gression. This would be accurate if the temperature were everywhere 
the same, and the difference in height would scarcely be any more com- 
plicated than if the density were constant. But the temperature of the 
air diminishes with increased height, so that the variation in density is 
not so simple, as the upper strata are more condensed by their lower 
temperatures than those below. 

The relation between temperature and height is rather complicated, 
as we shall see farther on ; and this, of course, renders more difficult 
the process of measuring heights by the barometer. At the same time, 
the atmospheric strata always contain a certain quantity of aqueous 
vapor, the weight of which must be added to that of the air. 


Furthermore, the weight of any body, and consequently that of a 
stratum of air, is proportionately less as the body in question is farther 
removed from the centre of the earth. And as the weight of bodies 
varies also according to the latitude on account of centrifugal force, it 
becomes evident that, for a single formula to be in general use for ob- 
servations made at different points of the globe, it is indispensable that 
it should include the latitude of the place of observation. 

Laplace has given, in the "Mecanique Celeste," the corrections ren- 
dered necessary by these different causes in measuring height, and has 
deduced from theory alone a formula the accuracy of which has been 
confirmed by numerous experiments. 

To determine the height of a mountain it is necessary that two per- 
sons take simultaneous observations of the readings of the barometer, 
one at its foot, the other at its summit. They must be careful, at the 
same time, to read the thermometers attached to the barometers, as well 
as others to determine the temperature of the surrounding air. Two 
observations will be sufficient ; but it is better to have several. 

A single observer can also ascertain the difference in level between 
two stations, not very distant the one from the other, with very fair ac- 
curacy, if he takes care to observe the thermometer and barometer at 
the lower stations, both when he leaves it and returns to it, and infers, 
from the difference, the reading at the lower when taking that at the 
higher station. 

When, by a long series of observations, the average readings of the 
barometer and thermometer at a given place have been determined, • 
they may be employed to calculate the absolute elevation of the place 
above the level of the sea by taking corresponding observations at 
the level of the ocean. Sufficient barometrical observations have al- 
ready been made at various elevations for us to be in a position to rep- 
resent this decrease of atmospheric pressure, with increase of elevation, 
no longer theoretically, but from direct observation. 

From a series of observations, made at very different elevations, the 
table on the following page has been formed. 

This satisfactory series of barometrical observations, which we are 
able to establish by means of numerous ascents, either in the balloon 
or up the mountain path, and by researches of several observers in 
inhabited regions far above the level of the sea, enables us also to en- 
deavor to represent, by a curve and a tint, this rapid decrease in the 
weight of the atmosphere. In Fig. 13, the horizontal line which forms 




Level of the Ocean 

Mean barometric reading at Greenwich Observatory 

Paris " 

" " Strasburg " 

" " Toulouse " 

Dijon (Perrey) 

Geneva Observatory (Plantainonr) 

Kodez ( Blondeau) 

Summit of Vesuvius (Palmieri) 

Guatemala (R. P. Canudas) 

Guanaxuato (Humboldt) 

The Monastery of the Great St. Bernard 

The Summit of the Faulhorn (Bravais) 

Town of Quito (Fouque') 

Summit of ^tna (Elie de Beaumont) 

In several aeronautical ascents (Flammarion) 

Summit of Mont Blanc (Ch. Martins) 

On the Chimborazo (Humboldt and Bonjiland) 

The summit of Ibi-Gamin (the highest mountain that has been 

climbed) (Schlagintweit) 

In an aeronautical ascent (Gay-Lussac) 

" " (Bixio and Barral) 

In several aeronautical ascents (Glaisher) 

In an aeronautical ascent (Glaisher) 

In the highest ascent (Glaisher) 

Height above 
the Sea. 










































the base represents the mean state of the barometer at the level of the 
sea (29 "92 inches). Each other horizontal line indicates the reading of 
the barometer corresponding to the elevation which is shown bj the 
vertical line. In this way, or by the aid of the tinted portion, it will 
be noticed that at 8200 feet the pressure is diminished by one-quarter, 
at 18,000 feet by ojae-half, and at 31,168 feet by three-quarters. 

The reading of the barometer diminishes, therefore, rapidly as we 
rise above the level of the sea. But even there it is not the same all 
over the globe's surface. It is lower at the equator than between the 
tropics ; at the equator it is about 29'84 inches ; it then increases up to 
the 33d degree of latitude, where it is 30'16 inches; then decreases un- 
til the 43d degree (30-00 inches), toward which point it becomes sta- 
tionary, and so remains up to the forty-eighth degree. Thence it con- 
tinues to decrease so far as sixty-four degrees, where it stands at 29"65 
inches. Lastly, it again increases from that point as far as the remotest 
latitudes— rat Spitzbergen (seventy-fifth degree), where the height of the 
barometer is 29*84 inches. Between the pressures at the thirty-third 
degree and the sixty-fourth degree of latitude, there is, therefore, a dif- 
ference of half an inch. I have laid down these results on a diagram, 
and traced the following curve (see Fig. 14, p. 52) : 

These variations in the atmospheric, pressure are probably caused by 



the trade- winds and upper currents of air, •which slightly raise the whole 
mass of the atmosphere. 

It is easy to conceive that the latitude may exercise some influence 
upon the pressure of the air, inasmuch as the conditions of temperature, 
pressure, and rotary movement vary with it. It is less easy to explain 

Fig. 13.— Diagram showing the decrease of atmospheric pressure, according to height. 

why the longitude should exercise any, but it seems, nevertheless, to do 
so. In the same latitude, the average pressure of the atmosphere is 
0'14 inch greater in the Atlantic than in the Pacific Ocean. 

The readings of the barometer are continually changing ; but, not- 
withstanding this, by a careful determination of the mean atmospheric 



pressure at many places, a map showing the lines of equal barometrical 
pressure (isobaric lines) can be drawn over the surface of our planet. 

The lines of equal pressure — or isobaric lines, as they are technically 
termed— are at first pretty equally distributed from K to S., running 
from W.S.W. to E.N.B. The isobaric line of 29*96 inches passes 
throuo-h the south of England and Holland; that of 30-02 inches near 
Tours and Nancy ; but the centre of France shows a very remarkable 

Latitade 5 10 15 20 25 30 35 *0 45 -50 55 60 65 70 lb 60 

Fig. 14.— Variation in the atmospheric pressure at the level of the sea, from the Equator to the North 


line of pressure, for the isobaric line of 80'04 inches crosses France 
diagonally, passing close to Strasburg, Chaumont, Dijon, Clermont, and 
Toulouse. On the other side, toward the S.E., the pressure diminishes, 
and attains a minimum not less remarkable in the Gulf of Genoa, where 
the pressure is about 29-98 inches. 

The curve of 3000 inches is formed, and its path pretty well known, 
in consequence of the numerous points at which observations have been 


made. The isobaric line of 30"08 inches, which passes close to Oran, 
and somewhat farther from Algiers, necessarily continues toward the 
west, nearly parallel with the above. A maximum of pressure in the 
Atlantic is in thirty -five degrees of north latitude; a minimum of 
pressure is met with at five degrees north of the equator ; a maximum 
at sixteen degrees south latitude, near St. Helena; and the lowest 
pressure existing in the world is to the south of Cape Horn, where it 
does not exceed 29 "33 inches. Upon the Asiatic continent the distribu- 
tion is quite different, and Siberia shows a maximum of about 30*24 
inches between Nertchinsk and Bernaoul. 

The chief difficulty in calculating altitudes is in reference to the 
mean level of the sea. Equilibrium upon the surface of the sea is not 
absolute; its level is affected by various causes, such as centrifugal 
force in the zone of the equator, the wind, barometrical pressure, and 
temperature. To these may be added the configuration of the sea- 
board, which gives a varying effect to the action of the winds and 
tides. It is well known that the sea rises quicker than it recedes, and 
when the gulfs are landlocked this effect is more decided. Along the 
coast the sea must rise higher than it does farther from shore. 

The level of the sea at Marseilles is 31 "5 inches lower than the aver- 
age level of the ocean upon the French coast. The Mediterranean must 
be an inclined plane, falling from the Straits of Gibraltar to the coast of 
Syria. The last level taken in Egypt, from the Mediterranean to the 
Red Sea, showed that the latter is higher than the Mediterranean. It is 
easy to comprehend that these seas, receiving much less water than 
evaporates from them, must have a tendency to become shallow, and 
that they are only kept up by the straits that unite them with the ocean. 

This first general description of the weight of the air and its pressure 
upon the spherical surface of the globe will answer our present purpose. 
It explains in some degree the statics ; and we shall soon reach the dy- 
namics. The atmosphere is unceasingly in motion, with its displace- 
ments, horizontal, vertical, and oblique. From this cause it results that 
the weight of air upon a given place, or the height of the barometer, is 
always changing. Solar heat gives rise to regular diurnal and monthly 
variations, the intensity of which differs according to the latitude. The 
change in th"e position of the great currents gives rise to extensive vari- 
ations upon a vast scale. Changes of weather are heralded by these 
fluctuations, which are bound up with the general pressure. 

Under the title of '•'■ Gomhien pese la masse entiere de tout Tair qui est an 


Monde^'' Pascal wrote, at the epoch when he devoted himself to his cele- 
brated experiments on atmospheric pressure, a small treatise as simple 
as it is curious, the first sketch of all that has since been written on this 
subject, and containing from the outset the absolute reply to the ques- 
tion which forms its title. "We learn," he says, "by these experi- 
ments that the air which is over the sea-level weighs as much as water 
to a height of thirty-two feet; but inasmuch as the air weighs less over 
more elevated places, and consequently does not press equally over all 
points of the earth alike, it is impossible to measure exactly what is the 
pressure upon all parts of the world by the same process, although an 
approximate measure, very nearly accurate, may be taken. Thus, for 
instance, it may be assumed that all the places of the earth have as much 
pressure upon them as if there was a depth of rather more than thirty- 
two feet of water over them ; and it is certain that this supposition is 
not half a foot in error. 

"Now we have seen that air which is above the mountains 3000 feet 
high is as heavy as water to a height of twenty-nine feet. Consequent- 
ly, all the air which extends from the level of the sea to the summit of 
the mountains weighs nearly the seventh part of the whole atmosphere. 

"We gather, too, from this, that if the whole sphere of the air was 
compressed against the earth by a force which, driving it downward, re- 
duced it to so small a space that it became of the density of water, it 
would then be only thirty-two feet high. The whole mass of air may 
be regarded as if it had been formerly a mass of water, thirty-two feet 
deep, which had become rarefied and very much dilated, and converted 
into the state which we call air; whereas it occupies, in truth, more 
space, thoagh it preserves exactly the same weight. 

"And as nothing would be simpler than to calculate what would be 
the weight in pounds of water surrounding the earth to a depth of thir- 
ty-two feet, we should find, by the same means, the weight of the entire 
mass of air. 

" Curiousity led me to make this calculation, and I found that the 
weight of this mass of water would be about nine trillions of pounds 
— that is, nine followed by eighteen ciphers represents the weight, in 
pounds, of air surrounding the earth." 

This weight is about ^^^^^oo part of the weight of the earth. 

If all this mass of air were agglomerated into a single ball, it would 
weigh as much as a ball of copper with a diameter of sixty-two miles. 
Thus the weight of the air is far from being insignificant. 

Fig. 15.— Lavoisier aoalyziug Atmospheric Air. 




It is to the great French chemist Lavoisier that science owes the dis- 
covery of the chemical components of the air. 

Let us go back to the researches of this laborious observer, and hear 
from his own lips the recapitulation of his interesting studies. 

Our atmosphere, he remarks, must be made up of all the substances 
capable of remaining in an aeriform state at the ordinary degree of tem- 
perature and atmospheric pressure which we experience. These fluids 
form a mass, almost homogeneous,* from the surface of the earth to the 
highest elevation which man has ever reached, and the density of which 
decreases with elevation. But it is possible that above our atmosphere 
there are several strata of very different fluids. 

What is the number, and what is the nature, of the elastic fluids 
which compose this lower stratum that we inhabit '/ 

After having established the fact that chemistry offers two methods 
essential for the study of bodies — that is to say, analysis and synthesis — 
Lavoisier describes as follows the celebrated experiment of the first 
analysis of air: 

" Taking a vessel, or long-necked tube, with a bell or globe at its ex- 
tremity, containing about thirty-six cubic inches (see Fig. 16, p. 58), I bent 
it (see Fig. 17, p. 58) so as to place it in the furnace while the extreme end 
of the neck was under a glass cover, which was placed in a basin of mer- 
cury. Into this vessel I poured four ounces of very pure mercury ; and 
then, by means of a siphon, I raised the mercury to about three-quarters 
the height of the glass cover, and marked the level by gumming on a 
strip of paper. I then lighted the fire in the furnace, and kept it up in- 
cessantly for twelve days, the mercury being just sufficiently heated to 
boil. At the expiration of the second day, small red particles formed 
upon the surface of the mercury, and increased in size and number for 
the next four or five days, when they became stationary. At the end 

* [Homogeneous must be understood to mean that the components of the atmosphere are 
found mixed in the same proportion at all heights. Its usual meaning is, of course, " of uni- 
form density." — Ed.] 



of the twelve days, seeing that the calcination of the mercury made no 
further progress, I let out the fire and set the vessels to cool. The vol- 
ume of air contained in the body and neck of the vessel before the op- 
eration was fifty cubic inches ; and this was reduced by evaporation to 
forty-two or forty-three. On the other hand, I found, upon carefully 
collecting the red particles out of the melted mercury, that their weight 
was about forty-five grains. The air which remained after this opera- 
tion, and which had lost a sixth of its volume by the calcination of the 
mercury, was no longer fit for respiration or combustion, as animals 
placed in it died at once, and a candle was extinguished as if it had been 
plunged in water. Taking the forty-five grains of red particles, and 
placing them in a small glass vessel, to which was adapted an apparatus 
for receiving the liquids and aeriform bodies which might become sepa- 

Fig. 16. — The glass vessel. 

Fig. 17.— The apparatus. 

rated, and having lighted the fire in the furnace, I observed that the 
more the red matter became heated, the deeper became its color. When 
the vessel approached incandescence, the red matter commenced to be- 
come smaller, and in a few minutes had quite disappeared ; and at the 
same time forty-one and a half grains of mercury became condensed in 
the small receiver, and from seven to eight cubic inches of an elastic 
fluid, better adapted than the air of the atmosphere to supply the respi- 
ration of animals and combustion, passed under the glass cover. From 
the consideration of this experiment, we see that the mercury, while it 
is being calcined, absorbs the only portion of the air fit for respiration, 
or, to speak more correctly, the base of this portion ; and the rest of the 
air which remains is unable to support combustion or undergo respira- 
tion. Atmospheric air is, therefore, composed of two elastic fluids of 
different, and even opposite, natures." 


The nature of air was thus clearly established by these experiments, 
which were made in 1777. Its real components were not, however, 
completely ascertained until the present century. The first exact analy- 
sis of air is scarcely fifty years old, and is due to Gay-Lussac and Hum- 
boldt, who analyzed it by the use of the eudiometer. 

. Fig. 18. — Meieiiry-Kiidiometer, for analyzing air. • 

When an equal mixture of air and pure hydrogen are set fire to in 
the eudiometer, all the oxygen disappears in the shape of water, which 
becomes condensed into dew, the volume of which is insensible, and 
there remains a mixture formed of nitrogen and the excess of hydro- 
gen employed. Now the hydrogen causes a volume of oxygen equal 
to half itself to disappear as water; whence it follows that the volume 
of oxygen contained in the measured air is equal to one-third of the 
volume that has disappeared. If the measures of the air, the hydro- 
gen, and the gases after explosion, are made at the same pressure and 
the same temperature, and if, in addition, the gases were saturated 
with humidity before explosion, the determination would require no 
correction. Such is the principle of the method. Gay-Lussac and 
Humboldt found that there was twenty-one per cent, of oxygen, and 
seventy-nine per cent, of nitrogen, in the air. This analysis has since 
been confirmed by nearly all chemists. There is another method by 
means of which the relative quantities of oxygen and nitrogen con- 
tained in the air of the atmosphere can be weighed — a process which 
gives results far more accurate than the measuring of the volumes (al- 



ways very small) of the gases employed in the other processes. The 
apparatus used is composed — first, of a tube which brings in the air 
from outside of the room where the operation is proceeding ; secondly, 
of a set of Liebig balls, L, containing a concentrated solution of caustic 
potash ; thirdly, of a tube, / in the shape of the letter U several times 
repeated, and filled with fragments of caustic potash ; fourthly, of a 
second set of balls, o, containing concentrated sulphuric acid ; fifthly, 
of a second tube, /, of the same shape as the one above mentioned, 
filled with pumice-stone steeped in concentrated sulphuric acid ; sixth- 
ly, of a straight tube, T, of hard glass. This tube is filled with copper 
filings, and laid upon a long iron furnace, so that it can be heated 

Fig. 19.— Apparatus for analyzing air by the method of weight. 

throughout its whole length, and is moreover furnished at its extremi- 
ties with two taps, r and /, which admit of its being emptied; seventh- 
ly, of a glass globe, B, holding from two to three gallons, and the neck 
of which is fitted with a tap, r. 

To perform the experiment, as complete a vacuum as possible is 
made in the tube T: the two taps are closed tight, and the tube, thus 
emptied of air, is weighed. The glass ball B, having been emptied of 
air, is also weighed. The various portions are then put together in 
the order described, and the tube T is made red-hot. Then the taps 
r r' of the tube T, and the tap r of the glass ball, are successively 
opened. The air, entering by the suction-tube to the right, traverses 
first of all the balls l and the tube/, where it parts with its carbonic 
acid ; then it passes into the second set of balls, o, and into the tube I, 



where the sulphuric acid removes all the vapor of water it contains. 
Separated from these, the air makes its way into the tube t, containing 
the red-hot copper, which retains the oxygen, and then passes into the 
empty glass ball in a state of pure nitrogen. The increase of weight 
in the tube clearly gives the weight of the oxygen which has been de- 
posited in the operation. The difference between the weight of the 
globe when empty and when full of nitrogen as clearly represents the 
weight of this gas. By means of this analysis, made with every con- 
ceivable precaution, MM. Dumas and Boussingault ascertained that 
one hundred parts of air contain — 

Oxygen, 23 in weight; 20"8 in volume. 
Nitrogen, 77 " 79-2 " 

The difference between the proportion of weight and that of volume 
is due to the fact that oxygen is rather heavier than nitrogen. 

These, therefore, are the two fundamental elements of the chemical 
constitution of air. But there 
exist other elements in far 
smaller quantities; such, for 
instance, as carbonic acid and 
aqueous vapor. Their quan- 
tity is determined by the ap- 
paratus described for finding 
the weight of the oxygen and 
nitrogen in the air. (See Fig. 
20.) An iron vessel is filled 
with water, and emptied by 
means of a tap inserted in the 
lower part. The water which 
runs out is gradually replaced 
by external air, which has to 
pass through the six curved tubes before it reaches the reservoir. The 
first two of these are filled with pumice-stone steeped in sulphuric 
acid, and the air, on its way through them, leaves behind the water 
which was mixed with it. The two middle tubes are filled with a 
concentrated solution of potash, which absorbs the carbonic acid. Of 
the last two tubes, containing pumice-stone steeped in sulphuric acid, 
the first is intended to extract the humidity which the potash has im- 
parted to the air, and the other to prevent the humidity from making 

Fig 20. — Apparatus for obtainiug the proportiou of car- 
bonic acid in air. 



its way back from the sucker into the tubes. By weighing, before 
and after the experiment, the series of analyzing tubes, we obtain the 
weight of the ivater and the weight of the carbonic acid contained in a 
volume of air equal to that of the reservoir. 

The atmosphere contains about looou ^^ its volume of carbonic acid. 
There is also a very simple process by which the oxygen and the 

nitrogen can be separated. Into a 
graduated tube, containing a certain 
volume of air, with its open end 
placed in a vessel containing water 
or mercury, is inserted a long stick 
of phosphorus. (Fig. 21.) At the 
expiration of six or seven hours, as 
a rule, the oxygen is absorbed, and 
the stick of phosphorus may be 
withdrawn, and the gas which re- 
mains — that is to say, the nitrogen 
— measured. The absorption is con- 
sidered to be complete (the appara- 
Fig. 21.— Apparatus for separating the oxygen tus being placed in the dark) when 

from the nitrogen. , , , . 

there ceases to be any glmimer upon 
the surface of the phosphorus. The rapid absorption of the oxygen 
by the phosphorus may be shown by heating the gas in a bell-glass 
into which a fragment of phosphorus has been introduced ; the phos- 
phorus is heated by an alcohol-lamp, and a portion of it volatilized; 
and when the flame has reached all the space occupied by the gas, the 
experiment is complete. Time is left for it to get cool ; the volume 
of nitrogen is transferred into a graduated tube and measured, the dif- 
ference from the original weight giving the quantity of oxygen. 

Oxygen and nitrogen are two permanent gases — that is to say, it has 
been found- impossible hitherto, either by compression or cold, to de- 
stroy their gaseous form. 

The first, oxygen, is the ordinary agent of combustion, whether of 
the kind which takes place in our fire-places or in our organisms. 
The second, nitrogen, exercises a moderating influence over the first. 

Carbonic acid, which exists in quantities varying according to time 
and place, but always very small in amount, has been liquefied under 
a strong pressure conjoined to intense cold ; it has even been solidified. 
In that state it has the appearance of light and very compressible snow, 


the contact of which with the skin produces a burning sensation, this 
excessive cold acting upon the epidermis in the same way as great 
heat.* In the small quantities in which it is found, carbonic acid pro- 
duces no ill effects ; in larger quantities it is hurtful to the breathing, 
and finally produces asphyxia. 

Emanations from the earth, the abundant sources of carbonic acid, 
are often met with in volcanic districts. When M. Boussingault ex- 
plored the craters at the equator, he was shown a locality where no 
animals could remain ; this was at Tunguravilla, not far from the vol- 
cano of Tunguragua. He thus describes his visit of 1851 : " Our horses 
soon gave us indications that we were approaching it ; they refused to 
obey the spur, and threw up their heads in a most disagreeable fashion. 
The ground was strewn with dead birds, among which was a magnifi- 
cent black-cock, that our guides at once picked up. Among the vic- 
tims were also several reptiles and a multitude of butterflies. The 
sport was good, and the game did not seem too high. An old In- 
dian, Quichua, who accompanied us, declared that, to procure a good 
sleep, there was nothing like making one's bed upon the Tungura- 

This deleterious emanation made itself manifest by the sterility of 
the ground for a circle of some hundred yards ; it was especially great 
at a point where there were many large trees lying dried up and half 
buried in the vegetable earth, which implies that these trees had flour- 
ished upon the spot where they have been lying since the eruption of 
the carbonic acid. This gas, like that which is also met with in similar 
circumstances in various regions of the globe, is carbonic acid more or 
less mixed with air, according to its distance above the soil. 

Carbonic acid exercises a directly deleterious effect upon the nerves 
and brain. Hence the anaesthetic effects which it may produce, and 
which all visitors to Pouzzoles, near Naples, may have seen at a grotto 
which has become famous from this cause. 

The keeper has a dog whose legs he ties together, to prevent his 
running away; he then places him in the middle of the grotto. The 
animal displays evident fear, struggles to escape, and soon appears to 
be dying. His master then takes him out into the open air, where he 
gradually recovers himself One of these dogs has been used for this 
purpose more than three years. It is all but proved now that the con- 

* [The snow-like flakes can be handled with impunity ; it is only when forcibly pressed 
against the skin that a blister is produced. — Ed.] 


vulsions of the pythonesses charged with expounding the decrees of 
the gods were produced by the priests with carbonic gas. 

This grotto is situated upon the slope of a very fertile hill, opposite, 
and not far from, Lake Agnano. The entrance is closed by a gate of 
which the keeper retains the key. It has the appearance and shape of 
a small cell the walls and vault of which have been rudely cut in the 
rock. It is about one yard wide, three deep, and one and a half high, 
and it is difficult to judge from its aspect whether it is the work of man 
or of nature. The ground in this cavern is very earthy, damp, black, 
and at times heated. It is, as it were, steeped in a whitish mist, in 
which can be distinguished small bubbles. This mist is composed of 
carbonic acid gas, which is colored by a small quantity of aqueous 
vapor. The stratum of gas is from ten to twenty-five inches high. It 
represents, therefore, an inclined plane the highest part of which cor- 
responds to the deepest portion of the grotto, and this is a physical 
consequence of the formation of the ground. The grotto being about 
on the same level as the opening leading into it, the gas finds its way 
out at the door, and flows like a rivulet along the hill-path. The 
stream may be traced for a long distance, and a candle dipped into it 
at a distance of more than six or seven feet from the grotto is extin- 
guished at once. A dog dies in the grotto in three minutes, a cat in 
four, a rabbit in seventy-five seconds. A man could not live more 
than ten minutes if he were to lie down upon this fatal ground. It is 
said that the Emperor Tiberius had two slaves chained up there, and 
that they perished at once; and that Peter of Toledo, Viceroy of Na- 
ples, shut up in the grotto two men condemned to death, whose end 
was as rapid. 

Two analyses of the air in this grotto, which had been collected at 
different times (see Ch. Ste. CI. Devi lie and F. Le Blanc), gave in vol- 
ume — 

Carbonic acid 67*1 73*6 

Oxygen 6-5 5-3 

Nitrogen 26-4 21-1 

100-0 100-0 

It is not necessary to travel so far for this predominance of carbonic 
acid. At Montrouge, near Paris, and in the neighborhood, there are 
large quarries, and even cellars, which are filled from time to time with 
this mephitic gas. 

Upon the borders of Lake Laacher, near the Rhine, and at Aigue- 


perse, in Auvergne, there are two sources of carbonic acid so abundant 
that the J give rise to accidents in the open country. The gaS rises out 
of small hollows in the ground, where the vegetation is very rich ; the 
insects and small animals, attracted by the richness of the verdure, seek 
shelter there, and are at once asphyxiated. Their bodies attract the 
birds, which also perish. 

In former times the accidents caused by this gas in caves, mines, and 
even in wells, gave rise to the most extravagant stories. Such locali- 
ties were said to be haunted by demons, gnomes, or genii, the guardians 
of subterranean treasures, whose glance alone caused death, as no trace 
of lesion or bruise was to be found on the unfortunate persons so sud- 
denly struck down. 

In addition to the oxygen, nitrogen, and carbonic acid, the air con- 
tains a certain number of other substances, in smaller and very varying 

The most important is aqueous vapor, of which I have spoken above 
in describing the method of analysis for determining its presence. The 
air always contains a certain proportion of aqueous vapor in a state of 
solution, and invisible. When this water passes into the state termed 
vesicular^ it constitutes clouds or mists. The quantity of aqueous vapor 
varies with the seasons, the temperature, the altitude, the geographical 
position, etc. At the same temperature and under the same pressure 
the maximum quantity capable of being mixed with the air is invaria- 
ble. Ttie hygrometrical state of the air, for a given temperature, is but 
the relation between the quantity of moisture really existing in the air 
and the quantity which would exist if the air were saturated at the same 
temperature. The millions of cubic feet of va2Mr of water which, mix- 
ing with the air, form the clouds and the rain constitute the most im- 
portant element of the atmosphere in respect to the circulation of life. 
Therefore water will be in a subsequent chapter the object of special 
study. The quantity of heat necessary for the evaporation of the water 
from the earth's surface has been ascertained. The volume annually 
evaporated may be represented by the volume of water which falls from 
the atmosphere in that space of time ; and, in comparing the results of 
observations taken at different latitudes and in both hemispheres, we 
are led to estimate this volume as corresponding to a depth of fifty-four 
and a quarter inches over the whole earth. The amount of heat neces- 
sary to evaporate such a volume of water would sufiice, according to 
Daubree, to liquefy a thickness of ice of nearly thirty-three feet in depth 



enveloping the whole globe. From the calculations of Dalton, the at- 
mosphere contains about the 0-0142th part of its weight in water: the 
upper strata are nearly free from water. 

What other substances are there to be found in the atmosphere ? It 
unquestionably contains small quantities of ammonia, partially in a 
state of carbonate of ammonia; perhaps, too, partially in a state of ni- 
trate, or even nitrite, of ammonia. The origin of this substance must 
evidently be attributed principally to the decomposition of vegetable 
and animal matter ; and its presence in the air is of peculiar importance 
in regard to the phenomena of vegetation and the chemical statics of 
plants. Several chemists have attempted to determine its exact propor- 
tion, which does not seem to exceed a few millionths of the volume of 
the air. 

The quantity of ammonia found in different waters is (in weight) : 

Rain-water 00000008 

Fresh-water 0-0000002 

Spring-water 0-0000001 

From one to two grains of ammonia per cubic foot have been found 
in sea-water. This is, no doubt, a very trifling quantity; but when we 
reflect that the ocean covers more than three-quarters of the globe, and 
when we consider also its enormous mass, it may be fairly looked upon 
as a vast reservoir of ammoniacal salts, whence the atmosphere can 
make good the losses which it is continually undergoing. 

The streams, too, carry to the sea prodigious quantities of ammoniacal 
matter. I will give one instance. According to M. Desfontaines, the 
engineer, the Ehine at Lauterburg has, on the average, a flow of 39,000 
cubic feet of water a second; and from a careful examination of the 
amount of ammonia contained in the water, it results that the Ehine, in 
its passage by Lauterburg, carries down with it every twenty -four 
hours at least 22,500 lbs. of ammonia — that is, 13,000,000 lbs. a year. 
The atmosphere, incessantly undergoing change (although its constitu- 
tion remains unaltered) by the immense labor of human beings who, 
like so many chemical pairs of bellows, are in continual motion on the 
bed of the aerial ocean, is the theatre of accidental chemical modifica- 
tions which play their part in the general organization. We see rising 
from the ground aqueous vapor, effluvia of carbonic acid gas, nearly 
always unmixed with nitrogen, sulphureted hydrogen gas, sulphurous 
vapors ; less frequently we notice vapors of sulphuric or hydrochloric 
acid ; and, lastly, carbureted hydrogen gas, which has for thousands of 



years been in use among different nations for the purposes of producing 
warmth and light. 

Of all these gaseous emanations the most numerous and abundant are 
those of carbonic acid. In former ages, the greater heat of the globe 
and the large number of crevices that the igneous rocks had not yet, 
covered contributed considerably to these emissions. Large quantities 
of hot vapor and of this gas became mixed with the aerial fluid, and 
produced that exuberant vegetation of pit-coal and lignites which is 
nearly an inexhaustible source of physical strength for a nation. The 
enormous quantity of carbonic acid the combination of which with lime 
has produced the chalky rocks then rose out of the bosom of the earth 
under the predominant influence of volcanic forces. What the alkaline 
soils could not absorb spread itself into the air, whence the vegetable 
matter of the Old World drew continuous sustenance. Then, too, abun- 
dant emissions of sulphuric acid in vapor have led to the destruction of 
mollusks and fish, and to the formation of beds of gypsum. Humboldt 
adds, that the introduction of carbonate of ammonia into the air is prob- 
ably anterior to the appearance of organic life upon the globe's surface. 
Besides" the ammoniacal vapors, the atmosphere also contains many 
traces of nitrogen, and even nitric acid. Several observers have also 
demonstrated, especially in large towns, the presence of a small quanti- 
ty of hydrogen in some form, probably carbureted. M. Boussingault was 
the first to prove, by precise experiments, the presence of a hydrogenous 
gas or vapor equal, at the most, to a ^oioo P^-rt of the air in volume. 

Analysis has also brought to light a certain quantity of iodine. The 
entire, or nearly entire, absence of iodine in the air or water of certain 
mountainous countries has, according to M. Chatin, a close connection 
with the existence of goitre among the inhabitants of these countries. 
His conclusions have been received, as a rule, with incredulity by 
chemists. Yet, when we consider that rain-water collected in a plu- 
viometer contains various kinds of salts, which arise from the washing 
of the dust suspended in the atmosphere, and that chemists have often 
found evidence of the presence of iodine in rain-water, there can be no 
difficulty in admitting that the presence in the air of iodine, free or in 
combination, may be, if not a normal, at least an occasional occurrence. 
We now arrive at the last element ascertained by special investigations 
to be existent in the atmosphere, viz., ozone. 

Van Marum, about the year 1780, by means of powerful electric 
machines, excited a large number of sparks in a tube full of oxygen. 


about six or seven inches long. After passing about five hundred 
sparks into the tube, he found that the gas had acquired a very strong 
sraell which, to use his own words, " seemed clearly the smell of elec- 
tric matter." Every one, indeed, is aware that if lightning strikes any 
.object it leaves behind it what is commonly called a sulphurous smell. 
Van Marum also found that the gas acquired, after the experiment, the 
property of oxidizing mercury without heat. Nearly sixty years later, 
in 1839, M. Schoenbein, professor at Basle, informed the Academy of 
Sciences at Munich that, having decomposed some water, he had been 
struck by the smell of gas emitted. After a few researches he drew 
the conclusion that a new body was brought to light by his experi- 
ment, which he called ozone, from o^w (to emit an odor), A large 
number of contributions have been subsequently made to the subject 
by various savants. 

Ozone is interesting in a chemical point of view, both in its nature 
and its energetic affinities, for it oxidizes directly both silver and mer- 
cury, at least when these metals are moist. It also liberates iodine 
from potassic iodide, and forms, with the metal, an oxide which, doubt- 
less, contains far more oxygen than the potash. The hydracids" impart 
to it their hydrogen. The salts of magnesium become decomposed by 
its contact with the formation of peroxide. Chlorine, bromine, and 
iodine, pass, when moist, under the influence of ozone, into chloric, 
bromic, and iodic acid. 

This agent has an exciting effect upon the lungs, provokes coughing 
and suffocation, and presents all the characteristics of a poisonous sub- 

. Notwithstanding all the researches that have been made in reference 
to ozone, the knowledge of it is, from a physical and chemical point of 
view, very imperfect;, a fact easy to understand when I state that it 
is impossible, even with the most perfect methods, to transform more 
than T-^Vir of a mass . of oxygen into pure ozone. This maximum 
reached, action ceases. How can it be easy to study a body which is 
spread over at least 1300 times its own volume of another gas?* 

It has occurred to several experimentalists, such , as Schoenbein, 
Berigny, Pouriau, Boeckel, Houzeau, and Scoutetten, to join to the 
ordinary meteorological observations ozonometrical observations also! 

* [By a continuous electrical discharge, maintained for many hours, Andrews and Tait 
were enabled to transform into ozone one-twelfth of the volume of oxygen operated on. — 
Phil Trans., I860.— Ed.] 


M. Schoenbein, in his experiments,- boiled one part of potassic iodide, 
ten parts of starch, and two hundred of water, a preparation of "Jo- 
seph's paper " being afterward steeped in it. The latter is dried in a 
close room, and then cut up into small strips. This paper becomes 
blue bj contact with the ozone, for the iodine is set at liberty and re- 
acts upon the starch. The deepness of the tint, however, depends upon 
the quantity of oxygen whicK has been turned into ozone. A small 
strip is exposed each day for twelve hours, sheltered both from the 
sun's rays and the rain, and its tint is then compared with a scale of 
ten colors, varying from white to indigo. 

In 1851, MM. Marignac and De la Eive undertook several experi- 
mental researches as to ozone ; and their conclusion was, that this sub- 
stance must be simply oxygen in a particular condition of chemical ac- 
tivity, determined by electricity. Berzelius and Faraday gave their ad- 
hesion to this opinion of the Geneva savants; and MM. Fremy and Bec- 
querel demonstrated, by fresh experiments in 1852, its legitimacy. The 
works of Thomas Andrews, published in 1855, leave no doubt upon 
this head. Ozone, no matter from what source it is derived, is a unique 
and separate body, with identical properties and the same constitution ; 
it is not a composite body, but an allotropic condition of oxygen. This 
allotropic condition is due to the action of electricity upon the oxygen. 
This opinion, based upon the best experiments, has now been univer- 
sally accepted, and this constitution of ozone appears incontestable. 

Let us further add to all these divers substances the presence of oxy- 
genated water^ as indicated by M. Struve, director of the Pulkowa Ob- 
servatory. While engaged in a chemical analysis of the water in the 
River Kusa, M. Struve was struck with the presence of a certain quan- 
tity of nitrite of ammonia, which was only to be found after a fall of 
snow or of rain. Soon after the downfall had ceased, all trace of this 
substance had again disappeared; M. Struve therefore supposed that 
the nitrite of ammonia existed in the air, and that it had been brought 
away by the snow or the rain. He entered upon researches on the sub- 
ject, and in the course of them made the interesting discovery of the 
presence of oxygenated water in the atmosphere. From these research- 
es may be drawn the following conclusions : 1st. Oxygenated water is 
formed in the atmosphere like ozone and nitrite of ammonia, and be- 
comes separated from the air through the atmospheric deposits. 2d. 
Ozone, oxygenated water, and nitrite of ammonia, are always intimately 
connected. 3d. The alterations which the atmospheric air brings about 


in the starch-iodine papers are caused by the ozone and oxygenated 

One word more. In absorbing into our lungs the quantity of air due 
to us, we often unwittingly inhale whole hosts of microscopical animals 
which are in suspension in the atmospheric fluid, and even portions of 
antediluvian animals, mummies, and skeletons of past ages ! 

Paris is nearly entirely built with chalky microscopical skeletons and 
tortoise-shells. The shells of the foraminifera^ for instance, in a fossil 
state, by themselves form entire chains of lofty hills and immense beds 
of building-stone. The rough chalk in the neighborhood of Paris is in 
some places so full of these remains that a cubic inch in the Gentilly 
quarries contains at least 100,000 of them. When we pass close by a 
house that is being pulled down, or one in course of construction, and 
find ourselves enveloped in a cloud of dust that penetrates down our 
throats, we often, beyond a doubt, inhale hundreds of these tiny atoms. 

Each day and each hour we inhale and take into our chest legions of 
animal and vegetable life. There are the living microzoa, several spe- 
cies of which are the fish of our blood ; there are the vibriones, which 
attach themselves to our teeth like oyster-banks to rocks. Then, again, 
there is the dust of microscopical animalcules, so small that it takes 
75,000,000 to make a grain ; and, besides these, there are the grains of 
pollen which, germinating in our lungs, further the spread of parasite 
life, which is out of all comparison more developed than the normal life 
visible to our eyes. 

The winds and storms, by their violent agitation of the atmosphere; 
the ascending currents due to the inequalities of temperature ; the vol- 
canoes, by their incessant emission of gas, vapors, and ashes, so finely 
divided that they often fall at a prodigious distance, carry up and main- 
tain in the higher regions corpuscles drawn away from the surface of 
the ground, or forced out of the internal and, perhaps, still incandescent 
portion of the globe. In the phenomena connected with the organism 
of plants and animals, these substances, so slight and of such different 
origins, the vehicle of communication for which is the air, very proba- 
bly exercise a far more pronounced action than is generally believed. 
Their permanence is, too, placed beyond doubt by the mere evidence 
of the senses, when a ray of sun penetrates a darkened room. As M. 
Boussingault remarks, "The imagination may conceive very readily, 
though not without a certain disgust, what is contained in these morsels 
of dust which we are incessantly inhaling, and which have been aptly 


denominated the refuse of the atmosphere. They establish, in a certain 
sense, a contact between individuals far removed from each other ; and 
though their proportion, their nature, and, consequently, their effects, 
are so varied, it is not too much to attribute to them a part of the insa- 
lubrity which generally manifests itself in all great agglomerations of 
human beings." 

Eain carries away these morsels of dust, while it dissolves their solu- 
ble matter, among which are found ammoniacal salts, as they also dis- 
solve the vapor of carbonate of ammonia and the carbonic acid gas 
diffused in the air. There must, therefore, exist in a fall of rain, at its 
commencement, more soluble substances than at its close; and if the 
rain continues uninterruptedly in calm weather, after a certain interval 
there can only be very insignificant indications of the existence of the 

Miasmas, the propagators of epidemics, are superinduced by the 
aerial currents ; the cholera, the small-pox, the yellow fever, and the 
diseases which periodically attack a district, seem to have their princi- 
pal source of propagation in the atmosphere — the factory of death as it 
is of life. The rate of mortality, which was so heavy in Paris during 
the early part of 1870, in consequence of small-pox, pleurisy, and in- 
flammation of the lungs, was especially severe in the northern districts 
of the city, over which the southerly wind spread the miasmas of the 
whole town, and where there was scarcely any ozone. A knowledge 
of the conditions of public health will be furnished in part by a study 
of the relations of meteorology to the variations in the rate of mortal- 
ity, which is as continually oscillating under the slight breath of the 
wind as under the trifling alterations in barometrical pressure. 

The air which Gay-Lussac brought down with him from his aero- 
nautical voyage, and which was collected at a height of 23,000 feet, 
had the same composition as that which floats upon the earth's sur- 
face. The experiments of M. Boussingault in America, and those of 
M. Brunner in the Alps, lead to the same conclusions. This similarity 
in results arises from the fact that currents of air and continual varia- 
tions in density are unceasingly mixing up together the atmospheric 

Is it the same at a greater height ? It is scarcely probable, for the 
nitrogen and oxygen being in a state of mixture, and not chemically 
combined, the gases must be ranged according to their density, allow- 
ing, of course, for the law of expansion ; that is to say, there are, as it 


were, two distinct atmospheres, the least dense of which does not ex- 
tend so far as the other, so that the proportion of nitrogen, the density 
of which is 0-972, that of the air being 1, must increase the higher one 
rises in the atmosphere; while the oxygen, the density of which is 
1'057 (and which is the denser of the two), must be in a greater pro- 
portion near the surface. According to this hypothesis, the latter gas, 
at 23,000 feet, would constitute only -^^ of the volume of air ; but at 
present experiment has failed to note so great a difference, because 
this calculation supposes the air to be in a state of tranquillity, whereas 
at these heights it is, as a matter of fact, in a continuous state of agi- 

The composition of the air varies very little : when it rains, the con- 
densed water dissolves more oxygen than nitrogen ; in frost, the water 
leaves these two gases alone ; the water which evaporates returns then 
to the atmosphere. 

We may now ask ourselves, in terminating this study of the chem- 
ical composition of the air, if this constitution is variable over the ter- 
restrial globe. By virtue of one of the great natural harmonies which 
unite the animal and the vegetable kingdoms, while the animals act as 
combustion-machines, taking the oxygen from the air and throwing it 
back into the atmosphere in the state of carbonic acid, the vegetables 
play the reverse part, acting as reducing-machines. Under the influ- 
ence of the solar rays, the green portions of the plants react upon the 
carbonic acid, decompose it, concentrate the carbon, and restore the 
oxygen to the air. The atmosphere, vitiated by the animals, is puri- 
fied by the action of the vegetables. The chemical equilibrium of the 
air's components has thus a tendency to self-preservation by virtue of 
this inverse action brought to bear upon its constituent elements. 

Certain phenomena due to the decomposition of rocks through oxida- 
tion seemed, at first sight, calculated to modify in the long run the com- 
position of the air; but a series of inverse actions of reduction tends to 
restore, in the shape of carbonic acid, the oxygen that has disappeared. 
As Ebelmen has pointed out, in his memoir upon changes in rocks, the 
process of reactions in the mineral matter upon the globe's surface seems 
also calculated to establish a compensation which maintains the chem- 
ical composition of the atmosphere. 

The question is whether this compensation is complete. Supposing 
it does not take place — as, indeed, is possible — does the quantity of oxy- 
gen diminish? As Thenard has remarked, "Thia is a very important 


question, the solution of which can only be arrived at in the course of 
several centuries, because of the enormous volume of air by which our 
planet is surrounded." 

In their remarkable memoir upon the true constitution of the atmos- 
pheric air, MM. Dumas and Boussingault thus expressed themselves in 

"Some calculations, which, though not of absolute precision, never- 
theless are based upon sufficiently certain grounds, tend to prove how 
far an analysis should extend to reach the limit at which the varia- 
tions in oxygen would be sensibly manifest. The atmosphere is un- 
ceasingly agitated ; the currents, stirred up by heat, by winds, by elec- 
tric phenomena, are continually being mixed up and confusing to- 
gether the various strata. The whole mass would, therefore, have to 
be changed in order to admit of an analysis indicating the difference 
between one epoch and another. But this mass is enormous. If we 
could place the whole atmosphere into a balloon, and suspend it in one 
side of a pair of scales, it would be necessary to put on the other side 
138,000 cubes of copper (each a mile in length, breadth, and thickness) 
to balance it. Let us now suppose that each man consumes a little more 
than two pounds of oxygen a day, that there are a thousand millions 
of men upon the earth, and that, through the respiration of animals and 
the putrefaction of organic matter, this consumption attributed to man 
be quadrupled. Let us further suppose that the oxygen disengaged 
from plants is only the compensating agent of the causes of absorp- 
tion omitted in our calculation, which would assuredly be putting the 
chances of alteration of the air in the strongest light. Well, even on 
this overdrawn hypothesis, at the end of a century the whole human 
race, and three times its equivalent, would only have absorbed a quan- 
tity of oxygen equal to fourteen or fifteen of the cubic miles of copper. 

"Thus, to assert that, with their utmost efforts, the animals which 
people the face of the earth could in a century render the air ihej, 
breathe impure, to the extent of depriving it of the ^oVir part of the 
oxygen that nature has placed there, is to make a supposition far be- 
yond the reality." 

In habitations badly ventilated, the effects of the breathing of men 
or animals, and the phenomena of the combustion of coal or of com- 
bustible matters, may cause a sensible alteration in the state of the air. 
Thus, in barracks, hospital rooms, theatres, wells, mines, etc., chemical 
analysis, when it is accurate enough, indicates a different composition 


from that of the open air. Furthermore, in habitations even out of the 
influence of the presence of sick persons, the animal emanations which 
escape with the aqueous vapor in respiration and perspiration may ex- 
ercise an incontestable physiological influence, often more injurious 
than that caused by the production of carbonic acid or the disappear- 
ance of the oxygen in small quantities. 

It is especially when the air arrives at a state of saturation from the 
causes cited above that there is reason to consider it deleterious. There 
is an unanimity of opinion in the present day that, to avoid a disas- 
trous influence upon the organic economy, dwelling-houses, and espe- 
cially hospitals, should be so constructed as to give more than 20,000 
cubic feet of air per day to each individual. 






D' B J) '' 

Among the works of the atmosphere in terrestrial life, one of the 
most important is unquestionably that of serving as a vehicle for hu- 
man thought, and enveloping the world in a sphere of harmony and 
activity which could not exist without it. 
What is sound? 

It is a movement produced in the air, and transmitted therein by 
successive undulations. To be perceived by the ear, this vibratory 
movement must be neither too slow nor too rapid. When the air, 

agitated by sound, vibrates at the rate of 
sixty undulations a second, it emits the 
dullest sound which can reach the ear. 
When the vibrations are 40,000 per second, 
they convey the sharpest sound which the 
auditory nerves can perceive. 

To appreciate the nature of the sonorous 
movement, let us suppose that between the 
chaps in a vise, A (see Fig. 22), is fixed one 
of the extremities, c, of an elastic blade, 
c D ; that the upper end, D, is pulled back 
to d', and then let go. By virtue of its 
elasticity, the blade will return to its primi- 
tive position ; but in consequence of the 
speed it has acquired, it will pass it and go 
on to d", executing on both sides of c D a 
series of oscillations the amplitude of which 
will gradually decrease, and in a more or 
less short space of time altogether cease. 
The longer the elastic blade is, the slower will be the vibrations; 
while, in proportion as the blade is shortened, the vibratory movement 
will become more rapid, and at a certain point will be imperceptible to 
the eye. But when the organ of vision ceases to play a part, so to 
speak, that of the organ of hearing begins, and the ear can distinctly 

Fig. 22.— Vibrations of a blade. 

Fig. 23. — Vibra- 


catch a sound, the nature of which depends upon the physical condi- 
tions of the vibrating body. Another instance of the production of 
sound is furnished by the vibration of a piece of cord fas- 
tened at its extremities, A b, and pulled in the middle.* 
Its vibration is rendered perceptible by the fact of the 
cord presenting the shape of a bobbin. By reason of the 
persistent impression upon the retina, and the speed of 
the vibratory movement, the eye sees the cord in all its 
positions together, as it were, the time of a vibration 
being less than that of a luminous impression, which is 
the tenth of a second. Sound, therefore, is but an im- 
pression upon the organ of hearing, caused by the vibra- 
ting movement of a given body. But the existence of a 
vibratory body on the one hand, and of an ear on the 
other, is not enough to cause an impression: a relation 
must be established between that body and the organ of 
tionofacord. hearing, and this is effected by a ponderable medium, 
liquid or gaseous, constituted of more or less elastic matter. If we im- 
agine a body vibrating in a complete vacuum, or in the centre of a 
space entirely devoid of elasticity, the ear, at a certain distance off, 
would catch no sound. Sound, in the proper sense of the word, does 
not exist in such a case. 

We may in fact form, from what is mentioned above, the following 
definition of sound : 

Sound is an impression produced hy the vibrations of a hody transmitted 
to the organ of hearing hy the intervention of a ponderable and elastic me- 

At what rate is sound propagated ? 

The first exact measurements were made in 1738 by a commission 
of the Academy of Sciences, of which Lacaille and Cassini de Thury 
were members. Several pieces of ordnance were placed upon the 
heights of Montmartre (then outside the walls of Paris) and at Mon- 
tlhery (an elevated position in the department of the Seine-et-Oise, dis- 
tant about 16 miles from Paris), and it was arranged that from a given 
hour a gun should be fired at equal stated intervals. The persons en- 
gaged in the experiment counted the time that elapsed between the 
flash and the arrival of the report ; and this was found to be, on an 

* [This is, of course, the principle of all stringed instruments — the harp, violin, etc. It is 
difficult to hold the cord sufficiently tight by the hand to produce a note. — Ed.] 



average, 1 minute 24 seconds for a distance of about 95,000 feet, which 
is at the rate of 1037 feet per second. 

These experiments were repeated in 1822 by the Bureau des Longi- 
tudes — a section of the Academy of Sciences — the persons taking part 
in them being Arago, Gay-Lussac, Humboldt, Prony, Bouvard, and 
Mathieu. Yillejuif and Montlhery, distant from each other 61,000 
feet, were the places selected ; and it was found that at a temperature 
of 61° the velocity of transmission was 1047 feet a second. 

A great number of similar experiments have been made in different 
countries. Very recently, M. Eegnault investigated this subject, em- 
ploying all the resources of modern physics, and especially telegraphic 
signals, for registering instantaneously the discharge and the arrival of 
the sound. 

The velocity of sound varies with the density and the elasticity of 
the air, and therefore with its temperature. According to the most 
accurate measurements, the following table may be given in reference 
thereto : 






per Second. 


per Second. 


1056 feet. 


1122 feet. 


1070 " 


1132 " 


1079 " 


1142 " 


1089 " 


1152 " 


1096 " 


1161 " 


1102 " 


1171 " 


1112 " 


1181 " 

Sound is propagated in the air by successive undulations, which may 
be roughly compared to the circular waves which are produced on the 
surface of water around a point disturbed by the fall of a stone. But 
they are, in reality, very different phenomena. In the liquid waves, 
the molecules are alternately raised and lowered in regard to the gen- 
eral level, but undergo no change of density ; while this change is, on 
the contrary, a characteristic of the waves of sound. There is, however, 
one circumstance common to both these phenomena which is worth 
pointing out — and that is, that the wave causes no real progressive 
movement. Thus, when waves of water follow each other, if we notice 
any small floating object, it is seen to alternately rise and fall, but it re- 
mains in the same place upon the surface of the water. Similarly, in 
the waves of sound, the molecules of the air execute oscillatory move- 
ments in regard to the propagation of sound, but the centre of these 
movements remains unchanged. 


Scientific education should teach us to behold in nature the invisible 
as well as the visible — to depict to the eyes of the intellect what escapes 
the eyes of the body. We may, with a little application, form a true 
idea of a sound-wave ; we may mentally see the molecules of air first 
pressed the one against the other; then, immediately after, this con- 
densation brought away again by an opposite effect of dilatation or 
rarefaction. We thus learn that a wave of sound is composed of two 
parts : in one the air is condensed ; while in the other, on the contrary, 
it is rarefied. A condensation and a dilatation are then the essential 
constituents of a sound-wave. But, if the air is necessary to the propa- 
gation of sound, what happens when a sounding body, such as the bell 
of a clock, is placed in a space destitute of air ? The result is that no 
sound proceeds from the empty space; the hammer strikes the bell, 
but silently. Hawksbee demonstrated this fact in a memorable experi- 
ment in 1705, before the Eoyal Society of London. He placed a clock 
under the receiver of an air-pump, in such a way that the striking of 
the clapper would continue after the air had been exhausted. While 
the receiver was full of air, the sound was quite audible; but it was no 
longer so (or at least in a very slight degree) when a vacuum had been 
created. The appended illustration is that of a contrivance which en- 
ables us to repeat Hawksbee's experiment 
in an improved manner. Under the re- 
ceiver B, placed firmly on the plate of an 
air-pump, will be seen the works of a strik- 
ing clock, A. The hammeris kept back by 
a spring and ratchet, c. As much as possi- 
ble of the air is exhausted ; then, by means 
of a stem, ^, which passes out through the 
top of the receiver, without letting in the 
exterior air, the trigger c?, which holds back 
the hammer &, is pulled. The bell, a, vi- 
brates silently. But if we let the air into 
the recipient, we at once hear a sound, very 
feeble at first, but growing louder as the air 
becomes denser. At great heights in the at- 
mosphere, the intensity of sound is notably 
less. According to the calculations of Saus- 
sure, the detonation of a pistol upon the summit of Mont Blanc is about 
equal in intensity to that of a common cracker at the level of the sea. 

Fig. 24. 


Since it is proved that there is no sound in a vacuum, fearful catas- 
trophes might take place in the planetary regions without the slightest 
audible notice of them reaching the surface of the earth. 

The vibratory movement of the air has been represented as being a 
circular wave, which spreads out in all directions with equal velocity, 
and diminishes in intensity as it advances. Where does it cease? 
where is it extinguished ? We must regard this as taking place at the 
point in space where it is no longer sensible to the most delicate ear ; 
and we all know how much this limit varies with the organization and 
habits of different individuals. At the same time, there can be no 
doubt that the aerial wave continues to spread out after the most prac- 
ticed ear has ceased to be sensible of it. In the places where there is 
a numerous population, the incessant noise kept up in the air by so 
many thousands of people creates a characteristic difference between 
day and night ; the noises become confounded together, and are propa- 
gated in a confused mass. During the night there is nothing to lessen 
the intensity of sound, and the ear perceives in all their force the howl- 
ing of the tempest, the blast of the winds, the roaring of the waves, 
the shrill cry of the bird of prey or the wild b'east; and it is then that 
pusillanimous fears and superstitious terror take possession of the timid. 
Traveling in a balloon over the plains of Charente, the stream of a 
river seemed to make as much noise as that of a great cascade, and the 
croaking of the frogs were audible at the height of 3000 feet. Above 
two miles all noise ceases. I never encountered a silence more com- 
plete and solemn than in the heights of the atmosphere — in those chill- 
ing solitudes to which no terrestrial sound reaches. 

"Two conditions determine essentially," says Tyndall, " the velocity 
of the sound-wave, viz., the elasticity and density of the medium which 
it passes through." The elasticity of the air is measured by the press- 
ure which it supports, and to which it forms an equilibrium. We have 
seen that, at the level of the sea, this pressure is equal to that of a col- 
umn of quicksilver 29-92 inches high. Upon the summit of Mont 
Blanc the barometrical column scarcely exceeds half this height, and, 
therefore, at the highest point of this mountain, the elasticity of the air 
is only half what it is upon the sea-coast. 

If we could increase the elasticity of the air without at the same time 
augmenting its density, we should increase the velocity of sound. We 
should also effect that object if we could diminish the density without 
making any change in the elasticity. The air heated in a closed vessel, 


ill which it can not become dilated, has its elasticity increased by the 
warmth, while its density remains the same. Sound will, therefore, be 
propagated more rapidly through air thus heated than through the air 
at its normal temperature. In like manner, air which is free to dilate 
has its density diminished by heat,* while its elasticity remains the 
same, and consequently it will propagate sound more rapidly than cold 
air — this, indeed, takes place when our atmosphere is heated by the 
sun; the air becomes dilated and much lighter, volume for volume, 
while its pressure, or, in other words, its elasticity, remains the same. 
This is the explanation of the statement that the velocity of sound in 
air is 1090 feet a second at the temperature of melting ice. At a lower 
temperature the velocity is less, and at higher temperatures greater, 
with an average difference of about one foot for one degree (Fahr.). 
Under the same pressure — that is to say, with the same elasticity — the 
density of hydrogen is much less than that of the air, and, in conse- 
quence, the velocity of sound through hydrogen gas considerably ex- 
ceeds its velocity through air. The reverse is the case with carbonic 
gas, which is denser than air ; for under the same pressure sound trav- 
els less rapidly through this gas than through air. 

The fact that air, even when very rarefied, can transmit intense 
sounds, is proved by the explosion of meteors at a great height above 
the earth, though it is true that, for this to be the case, the initial cause 
of the atmospheric disturbance must be very violent. 

The movement of sound, like all others, is less in amount when it 
communicates from a light body to one more dense. The action of 
hydrogen on the voice is a phenomenon of this kind. The voice is 
formed by the injection of air from the lungs into the larynx ; in its 
passage through this organ the air is set vibrating by the vocal chords, 
which thus give rise to sound ; and if one wishes to speak when the 
lungs are full of hydrogen, the vocal chords still impress their move- 
ment on the hydrogen, which transmits it to the air outside. But this 
transmission of a light gas to one much denser causes a considerable 
diminution in the intensity of the sound. The efiect of this is yery re- 
markable, Tyndall demonstrated it to the Eoyal Institution in Lon- 
don. Having, by a great effort of inhalation, filled his lungs with 
hydrogen, he began to speak, and his voice, generally powerful, was 
hoarse and hollow ; there was no ring in it ; it seemed to issue from 
the depths of the grave. 

* [The air must be contained in a vessel so constructed that the elasticity (pressure) is kept 
the same (for instance, in a cylinder in which fits a piston of constant weight).— Ed.] 


The intensity of sound mainly depends upon the density of the air 
from which it proceeds, not on that of the air in which it is heard. 

The wave of sound, propagated in all directions from the point where 
the sound has been produced, diffuses itself in the mass of air in which 
the motion takes place, and consequently lessens the amount of move- 
ment at any point. Let us imagine around the centre of disturbance a 
spherical layer of air, with a radius of a yard ; another layer of the 
same thickness, with a radius of two yards, contains four times as much 
air; one with a radius of three yards contains nine times as much ; one 
with a radius of four yards, sixteen times as much ; and so on. The 
quantity of matter set in motion increases, therefore, as the square of 
the distance from the centre of disturbance ; the intensity of the sound 
diminishes in the same degree. This law is expressed by the statement 
that the intensity of sound varies inversely as the square of the dis- 
tance from the point of initial disturbance. The decrease in the sound 
in inverse ratio to the square of the distance would not occur if the 
sound-wave spread in such a way as to prevent its being diffused later- 
ally. By producing a sound in a tube the interior surface of which is 
perfectly smooth these conditions may be realized, and the wave thus 
confined reaches a great distance, with but a slight loss of intensity. 
In this way Biot, noting the transmission of sound through the conduit 
pipes that supply Paris with water, found that he could carry on a con- 
versation in a low tone at a distance of 3300 feet ; the faintest murmur 
of the voice was heard at this distance, and the firing of a pistol at one 
end of the pipes extinguished a candle placed at the other end. 

Echoes depend, in a great measure, upon the compressibility and 
elasticity of the air. The sound-wave, as has been stated, spreads in- 
definitely, and is finally lost in space ; but if it encounters a body capa- 
ble of opposing it, it undergoes a reflection like that of light when it 
falls upon a smooth surface. For an echo to be distinctly produced, 
there must be a distance of fifty-five feet at least — the tenth of a sec- 
ond in time — between the person speaking and the reflecting surface. 
When the former is nearer, the echo is replaced by a confused reso- 
nance, which, in some buildings, renders it impossible for a speaker to 
make himself heard. 

AVhether acute or grave, sounds have the same velocity* — that of 

* [This is proved by the fact that, if a band of music be heard at a distance, the sounds are 
not confused, the distinctness of the tune being unaffected by the distance, though the loudness 
is of course diminished. — Ed.] 



1115 feet a second in air of 61° (Falir.). At half this distance the echo 
<Mves back four syllables rapidly pronounced ; at a greater distance it 
will distinctly reflect a larger number of syllables and whole phrases. 
The echo in Woodstock Park repeats seventeen syllables in the day- 
time and sixty at night. Pliny tells us that a portico was built at 
Olympia which repeated sounds twenty times. The echo at the Cha- 
teau de Simonetti was said to repeat the same word forty times. The 
theory is the same for the multiplied echoes; they result from the re- 
llecting sLirfices against which the aerial wave is thrown back several 
times from the one to the other, like a ray of light between two parallel 
glass plates. Perceptible sounds are included between the limits of 
60,000 and 40,000 simple vibrations a second, except in the case of 
ears which are exceptionally sharp. The undulations of the ether, 
which produce light, are far more rapid.* Visible colors are the result 
of vibrations so rapid that between 400 and 800 billions take place in 
a second. 

Of perceptible sounds, the extreme limits of the human voice are the 
lowest, /«, of 87, and the highest, ut, of 4200 vibrations. 

Sound has four fundamental properties — duration, height, intensity, 
and timhre or quality. The first three are defined by the words used 
to express them. As to the timbre., it is the resonance peculiar to each 
instrument and to each voice which enables us to clearly distinguish 
the sounds of a violin from those of a clarionet or a flute, and to recog- 
nize a person by hearing him speak or sing.f 

The tiiiibre of sounds has long been an insoluble enigma to natural 
philosophers and physiologists. It is only within the last few years 
that the excellent experiments of Helmholtz have proved that it de- 
pends upon the number of harmonic sounds which are produced simul- 
taneously with the fundamental tone, and upon their relative intensity. 

The intensity of sounds generated upon the surface of the earth 
spreads upward far more readily than in any other direction, and is 
transmitted to great heights in the atmosphere. Citing some few in- 
stances from my aeronautical travels, I will, in the first place, mention 

* [It must be borne in mind tbat there is only a general analogy between light and sound. 
In the latter, the vibrations consist of condensations and rarefactions in the air (or other gas), 
which are longitudinal — i. e., take place in the direction in which the sound is proceeding: 
while, in light, the vibrations are transversal (i. e., perpendicular to the direction of the ray), 
and take place in an ether which is supposed to pervade all space. — Ed.] 

t It is, of course, more difficult to recognize a person by his song than by his speech. 


that a noise, immense, colossal, and indescribable, is ever to be heard 
at 1000 to 1500 feet above Paris. In rising from a relatively quiet 
garden — as from the Observatory — we are astonished to hear a chaos 
of sound and a thousand various noises. The following details will, 
however, illustrate more strikingly this ascent of sound: 

The whistle of a steam-engine may be heard at 10,000 feet; the noise 
of a train at 8200 ;* the barking of a dog at 6000 ; the report of a gun 
attains the same height ; the shouts of people sometimes are audible at 
5000 feet, as also the crowing of a cock or the tolling of a bell. At 
1500 feet the beating of a drum and the sound of a band are audible; 
at 3900 feet the rumble of vehicles upon the pavement; and at 3300 
feet the shout of a single individual. At this last height, during the 
silence of the night, the current of a stream at all rapid produces the 
same effect as the rush of a cascade ; and at 2950 feet the croaking of 
frogs is plaintively distinct. At 2620 feet the slight noises made by 
the cricket are heard very plainly. 

This does not hold good of sound when descending. While we hear 
distinctly the voice of a person speaking from 1600 feet underneath us, 
it is impossible to catch what is said at a height of more than 300 feet 
above us. 

The occasion upon which I was most struck by this astonishing 
transmission of sounds vertically upward was in an ascent that took 
place on June 23, 1867. Having been in the midst of the clouds for 
several minutes, we were surrounded by a white and opaque veil that 
concealed both the sky and the earth, when I noticed with surprise a 
singular increase of light taking place around us, and all at once the 
sounds of a band reached our ears. We could follow the piece of music 
as distinctly as if the band had been in the clouds, a few yards distant 
from us. We were then just above Antony, a village near Paris. Hav- 
ing mentioned the fact in a newspaper, I. was glad to receive, a few 
days afterward, a letter from the President of the Philharmonic Society 
in that place, informing me that his society had seen the balloon above 
them, and had purposely played a very soft piece, in the hope that 
they might be of service to us in our researches. 

In this case the balloon was about 2950 feet above the place. At 
3280, 3940, and even 4590 feet, the parts were still distinctly audible. 
Far from being an obstacle to the transmission of sound, the clouds in- 
creased its intensity, and made the band seem close to us. 

* [On June 26, 1803, 1 heard a rail-way train when at the height of 22,000 feet.— Ed.] 


When sound has ceased, there still continues in the air a movement 
which may cause to vibrate membranes placed to receive and to inter- 
pret these impressions. M. Eegnault has measured these silent waves ; 
he has determined the distance traversed both by the sonorous wave 
and the silent wave which continues after the former has ceased. In a 
o-as-pipe, twelve inches in diameter, a pistol, with a charge of fifteen 
grains of gunpowder, was heard at the other extremity, 6250 feet off; 
and when the pipe was closed with an iron plate, the echo of the report 
was perceptible to any one listening attentively. The limit of the so- 
norous wave was therefore, in this instance, 12,600 feet; that oi silent 
waves is much greater. 

Air, the vehicle of sound, is at the same time the vehicle of smells 
and of all the emanations that are exhaled from the terrestrial surface. 
But smells are due not only to the vibratory movement, like sound and 
light. Fourcroy was the first to establish the fact that they are in part 
caused by the volatilization of vegetables or other matter ; that smells 
are caused by actual molecules suspended in the air — material particles, 
very slender and volatilized in the atmosphere. But the matter seems 
to become almost intangible. 

Nothing can give a more faithful idea of the divisibility of matter 
than the diffusion of smells. Three-quarters of a grain of musk placed 
in a room develop a very strong smell in it for a considerable time, 
without the musk perceptibly losing weight, and the box containing 
the musk will retain the perfume almost indefinitely, Haller states 
that papers perfumed with a grain of ambergris were quite odoriferous 
at the expiration of forty years. I remember purchasing upon the 
quay in Paris, some twelve years ago, a pamphlet which had a pro- 
nounced odor of musk about it. It had, no doubt, been there many 
months, exposed to the sun, the wind, and the rain. Since that time it 
has remained upon a library shelf, where the air has full access to it, 
and having just opened its pages, I find it as fully scented as ever. 

Smells are transported by the air to great distances. A dog can rec- 
ognize his master's approach from a distance ; and it is asserted that at 
twenty-five miles from the coast of Ceylon the delicious perfume of its 
balmy forests is still borne upon the wind. These sweet perfumes, like 
the harmony and the activity of the terrestrial surface, we owe to the 




The air being a fluid possessing weight, analogous to water in regard 
to the principles of pressure,* but, as we have seen, very much lighter, 
an instant's reflection will suffice to show that, if a body lighter than 
air be placed in the atmosphere, it will rise just as a body lighter than 
water — such as wood or cork — will, if placed at the bottom, at once 
ascend to the surface, because of its less specific gravit3^ 

If the atmosphere formed a homogeneous ocean above the surface of 
the globe, equally dense throughout, and terminated, like the sea, by a 
defined surface, every body the density of which was less than the 
density of this aerial ocean would rise, when left to itself, by the as- 
censional force of a pressure dependent on the difference of densities, 
and would remain floating upon the upper surface of this atmosphere. 
This was the notion of several of the predecessors of Montgolfier ; 
among others, of the worthy Father Galien, in his fantastic scheme for 
aerial navigation, pubhshed in 1755. His famous ship was to contain 
" fifty-four times as much weight as Noah's ark," its dimensions were 
to be equal to those of the town of Avignon ; for the hypothesis of this 
excellent ecclesiastic was that this vast iron vessel would float in the 
atmosphere in virtue of the same principle as that by which a ship 
floats upon the ocean. But as the density of the atmospheric strata 
diminishes with elevation, all objects lighter than the lower strata 
mount merely to the region the density of which is such that the 
weight of the body is equal to the weight of the volume of fluid dis- 

Archimedes established for liquids a principle which we can apply 
with precision to the atmospheric fluid, enunciating it as follows: All 
bodies situated in the atmosphere lose a portion of their absolute 
weight, equal to the weight of the air which they displace. 

This actual loss of weight in the air is proved by means of a pair 
of scales specially constructed for the purpose, as the name indicates, 

* [It must be borne in mind that water is very slightly compressible indeed ; while air is 
an elastic fluid, capable of almost indefinite compression or expansion. — Ed.] 


Fit;;. 25.— The Baroscope. 


of seeing the weight— the baroscope. One extremity of the beam has 
attached to it a hollow copper sphere; the other end carries a small 

piece of lead, balancing in the air the 
copper sphere. If this apparatus is 
placed under the glass-receiver of an 
air-pump, as soon as a vacuum has 
been created the balance inclines to 
the side of the sphere, showing that in 
reality it weighs more than the mass 
of lead which was in equilibrium with 
it when in the aii;; or, in other words, 
that it has lost in the air a portion 
of its weight, because its volume was 
larger than that of the piece of lead. 
To verify, by means of the same apparatus, that this loss is just equal 
to the weight of the air displaced, the volume of the sphere must be 
measured, and if it holds say about a pint, or 84-6 cubic inches, the 
weight of this volume of air being 11-3 grains, the corresponding 
weight must be attached to the piece of lead, and the equilibrium will 
be re-established in the vacuum, but will be destroyed upon the re- 
introduction of air. 

Let us note, en jMssant, in reference to this subject, that when any 
object is weighed in scales it is never its exact weight which is ob- 
tained, but its apparent weight. To get at the actual weight, the ob- 
ject must be weighed in vacuum. This is a source of continual error 
which is rarely taken into consideration. But, on the other hand, it 
may be asked, what is the real weight of any particular body? and the 
reply must be, there is no such thing. It is a purely relative matter, 
resulting from the volume and density of the planet which we inhabit. 
A pound weight does not constitute an absolute quantity, notwith- 
standing appearances to the contrary. The proof of this is, that if a 
pound weight were transported to the surface of the sun it would weigh 
nearly twenty-eight pounds;* whereas it would weigh two pounds and 
three quarters, nearly, upon the surface of Jupiter, and only one-sixth 
of a pound at the moon ! And even without going so far as this, if we 
imagine our atmosphere graduallj^ becoming denser and denser, we 

* [The weighing must, of course, be made by means of a spring-balance, or other bahmce 
of the same kind. If a certain object balances a pound weight on the earth in a pair of scales, 
it would do so also anywhere else — on the sun, moon, etc. — Ed.] 


sliould, in that case, become lighter; or, again, if the earth revolved 
seventeen times faster than it does, the inhabitants of tropical countries 
would have no weight at all, and only weigh a few grains in the lati- 
tude of London or Paris. This may serve to confirm the doctrine of 
those English philosophers who, with Berkeley at their head, argued 
that the only real fact is, that there is nothing real in the world. 

But let us return to the weight of the air. A balloon is, in fact, 
merely a body lighter than the weight of the air which it displaces, and 
which consequently rises in search of its equilibrium into higher re- 
gions of less density, where it will only displace a volume of air equal 
to its own weight. It is clear that, far from being in opposition to the 
laws of gravity, the ascent of balloons is, on the contrary, a special con- 
firmation of them. 

Whatever may be the substance which is used for filling a globe of 
silk or other material, if the whole apparatus — the gas which fills the 
envelope, the car, the net to which it is attached, the aeronauts, etc. — 
weighs less than the air which it displaces, it constitutes by that very 
fact an aerostatical machine, and rises in the atmosphere. 

When Montgolfier launched, for the first time, a balloon into the air, 
his balloon was simply inflated with hot air. The density of air heated 
up to 122° (Fahr.) is 0-84, that of air at 32° being represented by 1. 
The density at 212°, the temperature of boiling-water, is 0-72, giving 
scarcely a difierence of one-third for the ascensional force. The den- 
sity of pure hydrogen is only 0-07 ; that is, one-fourteenth of that of air. 
The density of carbureted hydrogen is about 0-55 ; that is, tibout one- 
half the density of air. The latter of these two gases is generally used 
for filling balloons. 

By a happy coincidence not rare in the history of science, hydrogen 
gas was discovered alrnost simultaneously with the invention of bal- 
loons. In 1782, Cavallo exhibited before audiences, at his London 
lectures, soap-bubbles formed of hydrogen, which rose by their less 
specific gravity up to the ceiling of the hall. In the following year 
(June 5, 1783) Montgolfier launched the first aerostat. With a little 
study and energy, Cavallo might have deprived the Annonay manu- 
facturer of the immortality of his invention. 

A balloon inflated with hot air is still often called a Montgolfier bal- 
loon, after its inventor, A balloon inflated with gas is denominated 
a gas-balloon, and often, popularly, an air-balloon. Gas has been 
adopted almost exclusively since its first trial, which was made at 



Fig. 20.— Soap-bubbles inflated with liydrogen. 

Paris, on the 27th of August, 1783, by M. Charles, Member of the 
Academy of Sciences, and the Brothers Eobert. 

The first time that a car was suspended to a balloon was on the 19th 
of September, 1783, in presence of Louis XVI. and Marie Antoinette, 
at Versailles ; and the earliest passengers were a sheep, a cock, and a 
duck. The first real aerial voyage was accomplished on the 21st of 
October following, by Pilatre des Hosiers and the Marquis dArlandes, 
who rose, by means of a fire-balloon, from the Chateau de la Muette 
(the Bois de Boulogne), and made their descent at Montrouge (on the 
south side of Paris), after having crossed the capital. 

To say that one feels one's self being carried up by a balloon perhaps 
scarcely gives a correct idea of the situation. It is better to say, sees 
one's self carried up, for the voyager feels no kind of movement, and 
the earth seems to him to he descending. 

As personal impressions are unquestionably those the recital of 
which comes nearest to the reality, I will take the liberty of citing 
some. My first ascent took place on Ascension-Day (May 25) in 1867. 
Eugene Godard, the aeronaut, having verified the perfect equilibrium 
of the balloon, orders the four assistants to let slip through their hands, 
without losing hold of them, the ropes which secured the car, and thus 
we find ourselves a few yards above the ground. The sky is clear, the 


wind light, and the balloon, filled with hydrogen gas, becomes impa- 
tient and endeavors to rise. Then, taking a sack of ballast in his 
hand, Godard gives the word to " let go," throwing over a few pounds 
of sand, and the aerostat rises with majestic ease. 

The balloon rises in an oblique curve, caused by two component 
forces — its ascensional power on the one hand, and the velocity of the 
wind on the other. If, as is proper from all points of view, we take 
care to let the balloon have only a slight ascensional force, the most 
magnificent of panoramas is slowly developed before the charmed gaze. 
If we wish only to ascend to a height of 3000 to 4000 feet, the balloon 
is allowed to move horizontally as soon as it reaches an atmospheric 
stratum of this elevation, whose density is then equal to that of the 
balloon. For higher ascents, the balloon is lightened by throwing out 

The aeronaut, the meteorologist, or the astronomer who thus hovers 
in the air, is in a most enviable position for studying the atmosphere. 
Penetrating into the very midst of the clouds, traversing them to de- 
termine the light and heat which influence them, following the storm 
in its mysterious formation, studying the production of rain, snow, and 
the hail, transporting himself, in fact, into the very regions where these 
phenomena are occurring, it is there alone that the observer is reall}^ 
master of the globe. The savant may in vain spend years by his fire- 
side in forming hypotheses by the aid of books and apparatus ; but in 
this, as in most other things, the surest method of ascertaining what is 
going on, is " to go and see for one's self," as the old proverb has it. 
And, assuredly, no attempt can yield more fruitful results. 

I do not intend to revert to a subject which was largely and com- 
pletely dealt with in 1870 in a work specially devoted thereto. The 
purpose of this chapter is not to record my travels in the air; the scien- 
tific results flowing from them will be found embodied in the various 
explanations which compose the present book. It was merely necessa- 
ry to lay down the general theory of the ascent of a balloon in its rela- 
tions to the study of the atmosphere, and to give some idea of the effects 
of the hio-her regions. 

If aerial travels may be profitably applied to the study of the forces 
at work in the atmosphere, and of the laws which preside over its mul- 
tiform movements, they are also a special subject of interest for the ob- 
server, and open for him an exclusive vista of vast and useful contem- 
plation. Borne into the fields of the sky by the invisible breath of the 


winds, the solitary balloon rises above the earth, and the traveler views 
its surface as a map stretched out on a boundless plain seen with all the 
characteristics of its local topography. Capitals situated on the banks 
of rivers, the central cities of provinces, innumerable villages dissemi- 
nated over the country, and succeeding each other in hundreds like the 
little chateaux one used to see dotted down in old-fashioned maps, hill- 
sides brown with the vine, furrows golden with grain, verdant mead- 
ows, cragged mountains whose tops are covered with sombre forests, 
sparkling streams and sinuous rivers running to the distant ocean — all 
the charms, soft or stern, of landscape and perspective are slowly re- 
vealed to the delighted gaze of the aeronaut who, without feeling the 
slightest movement, hovers as in a dream until he again sets foot upon 
the earth that he has been contemplating from on high. A less power- 
ful impression, but of a similar kind, is derived from a mountain ascent. 

The purity of the upper air, and the variation in atmospheric press- 
ure, are physical elements which must be taken into account in order 
to explain the benefit of a sojourn at a moderate altitude. The peculiar 
action which may be exercised upon impressionable organizations by 
the contemplation of mountains, where nature has bestowed so liberally 
that mixture of the gracious and the terrible which tends to make u]» 
the picturesque, is undeniable. J. J. Rousseau says: "Every one must 
feel, though he may not observe it, that in the purer and more subtle 
air of the mountains he has a greater facility of breathing, more nimble- 
ness in the body, more serenity of mind; the pleasures are less ardent 
there, as the passions are more subdued. Meditation assumes a certain 
tranquil voluptuousness, which is not in the least sensuous or bitter. It 
seems that, as we rise above the abode of man, we leave all terrestrial 
and base sentiments behind, and as we approach the ethereal regions, 
the soul gains something of their inalterable purity. We become grave 
without being melancholy, placid without indolence, content to live and 
to think. I doubt whether any violent agitation, any hysterical affec- 
tion, could hold out against a lengthened sojourn there; and I am as- 
tonished that a bath of the healthy mountain air is not one of the great- 
est medical remedies." 

It is, however, proper to state that, beyond moderate altitudes, the 
human organism is susceptible of a deleterious influence, owing to the 
change in atmospheric pressure, the dryness of the air, and the cold. 

The physiological uneasiness and disturbances which are felt at great 
heights have long been ascertained facts. As early as the fifteenth cen- 



tury tbey were observed and described by Da Costa, under the name of 
mal de montagne. Later, all mountain explorers in the Alps, the Andes, 
and the Himalayas, as well as aeronauts, have noted these singular per- 
turbations of organism, and have published theories more or less plau- 
sible in explanation of them. The principal cause assigned since De 
Saussure has been merely the rareflxction of the air ; but by what series 
of actions and reactions does this rarefaction affect the human body? 
That was the point which needed elucidation. 

In 1804, Gay-Lussac and Biot rose as high as 13,000 feet in a balloon. 
Gay-Lussac's pulse went up from 62 to 80 a minute ; that of Biot from 
79 to 111. In the memorable ascent of July 17, 1862, Messrs. Glaisher 
and Coxwell attained the enormous elevation of 37,000 feet. Previous 
to the start, Glaisher's' pulse stood at 76 beats a minute, Mr. Cox.well's 
at 74. At 17,000 feet the pulse of the former was at 84 — of the latter, 
at 100; at 19,000 feet Glaisher's hands and lips were quite blue, but 
not his face ; at 21,000 feet he heard his heart beating, and his breath- 
ing was becoming oppressed; at 29,000 feet he became senseless, and 
only returned to himself when the balloon had come down again to the 
same level ; at 37,000 feet the aeronaut could no longer use his hands, 
and was obliged to pull the string of the valve with his teeth. A few 
minutes later he would have swooned away, and probably lost his life. 
The temperature of the air was at this time 12° below zero. In aero- 
stats, however, the explorer remains motionless, expending little or none 
of his strength, and he can therefore reach a greater elevation before 
feeling the disturbance which brings to a halt at a far lower level tHe 
traveler who ascends by the sole strength of his muscles the steep sides 
of a mountain. 

De Saussure, in his ascent of Mont Blanc on the 2d of August, 1787, 
has given an account of the uneasiness which his companions and him- 
self began to experience when a long distance from the summit. Thus, 
at 13,000 feet, upon the Petit-Plateau, where he passed the night, the 
hardy guides who accompanied him, to whom the few hours' previous 
marching was absolutely child's plaj'-, had only removed five or six 
spadefuls of snow in order to pitch the tent, when they were obliged 
to give in and take a rest, while several felt so indisposed that they 
were compelled to lie upon the snow to prevent themselves from faint- 
ing. "The next day," De Saussure tells us, "in mounting the last 
ridge which leads to the summit, I was obliged to halt for breath at 
every fifteen or sixteen paces, generally remaining upright and leaning 


on my stock; but on more than one occasion I Lad to lie down, as I 
felt an absolute need of repose. If I attempted to surmount the feel- 
ino-, ray legs refused to perform their functions; I had an initiatory 
feeling of faintness, and was dazzled in a way quite independent of the 
action of the light, for the double crape over my face entirely sheltered 
the eyes. As I saw with regret the time which I had intended for ex- 
periments upon the summit slipping away, I made several attempts 
to shorten these intervals of rest. I tried, for instance, a momentary 
stoppage every four or five paces, instead of going to the limit of m}^ 
strength, but to no purpose, as at the end of the fifteen or sixteen paces 
I was obliged to rest again for as long a time as if I had done them 
at a stretch ; indeed, the uneasy feeling was strongest about eight or 
ten seconds after a stoppage. The only thing which refreshed me and 
augmented my strength was the fresh wind from the north. When, in 
mounting, I had this in my face, and could swallow it down in gulps, I 
could take twenty-five or twenty-six paces without stopping." 

Bravais, Martins, and Le Pileur, in their celebrated expedition to 
Mont Blanc in 1844, experienced and investigated the same phenome- 
na upon the Grand Plateau. In clearing the tent, which was half filled 
with snow, the guides had continually to stop for breath. An internal 
uneasiness, according to Martins, made itself apparent in many diflerent 
ways. The appetite was gone. The strongest, biggest, and most har- 
dy of the guides fell upon the snow, and was nearly in a fit when the 
doctor, Le Pileur, felt his pulse. On nearing the summit, Bravais was 
anxious to see how far he could go without a rest ; at the thirty-second 
step he was obliged to stop short. 

All the indispositions felt by the savans of whom we have been 
speaking, and by many other travelers, at great elevations, have been 
classed in the following list: 

Breathing. — The breathing is accelerated, impeded, laborious; and 
there is a feeling of extreme dyspnoea at the least movement. 

Circulation. — The great majority of travelers have noticed palpita- 
tions, quickening of the pulse, beating of the carotids, a sensation of 
plenitude in the vessels, and sometimes the imminent approach of suf- 
focation and various kinds of hemorrhage. 

Innervation. — Very painful headache, a sometimes irresistible desire 
to sleep, dullness of the senses, loss of memory, and moral prostration. 

Digestion. — Thirst, strong desire for cooling drinks, dryness of the 
tongue, distaste for solid food, nausea, and eructations. 


Functions of Locomotion. — Pains more or less severe in the knees and 
legs; walking causing great fatigue and exhausting all strength. 

These disturbances are not regular, they do not all come on at once, 
and evidently depend a good deal upon the strength, the age, the hab- 
its, and the previous actions of the individual. They seem to have a 
greater effect upon Alpine climbers than in other mountainous regions. 
Thus, at the Great St. Bernard, the monastery of which has an altitude 
of only 8117 feet, most of the monks become asthmatic. They are 
compelled to descend frequently into the valley of the Ehone to regain 
their health, and at the end of ten or twelve years' service to quit the 
monastery for good, under penalty of becoming quite infirm ; and yet, 
in the Andes and Thibet, there are whole cities where people can en- 
joy as good health as anywhere else. Boussingault says, that " when 
one has seen the activity which goes on in towns like Bogota, Micui- 
pampa, Potosi, etc., which have a height of from 8500 feet to 13,000 
feet ; has witnessed the strength and agility of the toreadors in a bull- 
fight at Quito (which is 9541 feet); when one has seen young and deli- 
cate women dance for the whole night long in localities almost as lofty 
as Mont Blanc, where De Saussure had scarcely the strength to read his 
instruments, and where the vigorous mountaineers fainted ; when one 
remembers that a celebrated combat, that of Pichincha, took place at 
a height as great as that of Monte Rosa (15,000 feet), it will be admit- 
ted that man can become habituated to the rarefied air of the highest 

The same writer is also of opinion that in the vast fields of snow, 
the discomfort is increased by an emission of vitiated air under the 
action of the solar rays, and he bases this impression upon an experi- 
ment of De Saussure, who found the air near the surface of snow to 
contain less oxygen than that of the surrounding atmosphere. In cer- 
tain hollows and inclosed valleys of the higher part of Mont Blanc — 
in the Corridor^ for instance — people generall}^ feel so unwell in trav- 
ersing it, that the guides long thought that this part of the mountain 
was impregnated with some mephitic exhalation. Thus, even now, 
whenever the weather permits, people ascend by the Bosses ridge, 
where a purer air prevents the physiological disturbances from being 
so intense. 

Notwithstanding that one may become gradually accustomed to the 
attenuated air of high elevations, certain animals can not live there. 
Thus cats, taken up to the altitude of 13,000 feet, invariably succumb, 


after having been subject to singular attacks of tetanus, of gradually 
increasing intensity; and, after making tremendous leaps, succumb 
from fatigue, and die in convulsions. 

We will conclude these remarks by mentioning that the highest in- 
habited spot in the world is the Buddhist cloister of Hanle (Thibet), 
where twenty priests live at the enormous height of 16,500 feet. 
There are other cloisters built at a nearly equal height in the province 
of Guari Khorsum, upon the banks of the lakes Monsaraour and Ba- 
kous, and they are inhabited all the year round. In these equatorial 
regions one can live very easily for ten or twelve days at an altitude 
of 18,000 feet, but not for a longer time. The Brothers Schlagintweit, 
when they explored the glaciers of the Ibi-Gamin in Thibet, encamped 
and passed the night, with eight men of their expedition, from the 13th 
to the 23d of August, 1855, at these exceptional elevations, which are 
rarely visited by a human being. For ten days their encampments 
varied from 18,000 to 21,000 feet; that is to say, the greatest altitude 
at which a European ever passed the night. These three brothers 
succeeded, on August 19, 1856, in mounting to an elevation of 24,339 
feet — farther than man has ever yet reached. At first they suffered a 
good deal when they got to 17,000 feet; but, after a few days, they felt 
nothing but a passing uneasiness even at 19,000 feet. It is, however, 
probable that a prolonged stay at this altitude would have produced 
ill effects. 

Three or four years ago, Professor Tyndall, in order to take scientific 
observations, passed the whole night upon the summit of Mont Blanc, 
sheltered only by a small tent. The guides who accompanied him were 
so unwell that the next morning they were obliged to make their way 
downward as quickly as possible. 

A year or two ago, M. Lortel, who had several times ascended to 
14,000 feet upon Mont Blanc without discomfort, and who doubted 
whether another 1600 feet could superinduce the symptoms asserted, 
went to the summit to judge for himself. He writes: "I am now con- 
vinced, and am compelled to admit, de visu and rather at my expense, 
that there really do exist causes of disturbance at this height which af- 
fect a person who ascends so far, especially if he is in motion^ in this 
rarefied air. This is also the result of my personal observations ; and 
I have satisfied myself that it is much less hurtful to the organic func- 
tions to rise to great heights when sitting still in a car than by climbing 
over the snows." 



To complete our atmospheric panorama, it is interesting to see what 
are the highest points of the mountainous peaks upon which man is liv- 
ing, and what are the highest points of the mountain chains which raise 
into the rarefied atmosphere their silent and icy peaks. The highest 
spots of the earth which are inhabited are : 

The Buddhist cloister of Hanle (Thibet) 16,532 feet. 

Cloisters on the sides of the Himalaya 14,7G4 to 16,404 

The post-house of Apo (Peru) 14,377 

The post-house of Ancomarca (do.) ,.... 14,206 

The village of Tacora (do.) 13,691 

The town of Calamarca (Bolivia) 13,651 

The vineyard of Antisana (Republic of Ecuador) 13,455 

The town of Potosi (Bolivia), ancient pop. : 100,000 13,323 

The town ofPuno (Peru) 12,871 

The town of Oruro (Bolivia) 12,455 

The town of La Paz (do.) 12,225 

Quito, capital of the Ecuador Eepublic, is situated at an altitude of 
9541 feet ; La Plata, capital of Bolivia, at 9331 feet ; Safita Fe de Bo- 
gota, at 8730 feet. The highest inhabited spot in Europe is the Monas- 
tery of Mount St. Bernard, which is 8117 feet high. 

The highest passes of the Alps are — the pass of Mount Cervin, 11,188 
feet; the Great St. Bernard, 8110 feet; the Col de Seigne, 8074 feet; 
and the Furka, 8002 feet. The highest passes in the Pyrenees are — 
the Port d'Oo, 9843 feet; the Port Vi el d'Estaube, 8402 feet; and the 
Port de Pinede, 8202 feet. 

The highest mountains in the world are: 

Asia : The Gaurisankar, or Mount Everest (Himalaya). 

The Kanchinjinga (Sikkim, Himalaya) 

The Dhaulagiri (Nepaul, do. ) 

The Juwahir (Kemaon, do. ) 

Choomalari (Thibet, do. ) 

America : The Aconcagua (Chili) 

The Sahama (Peru) 

The Chimborazo (Republic of Ecuador) 

The Scrota (Bolivia) 

Africa: The Kilimanjaro 

Mount Woso (Ethiopia) 

Oceania: The Mownna-Roa, volcano (Sandwich Isles). 

Europe: Mont Blanc 

Monte Rosa 

The birds, of course, represent the population of the 
altitudes. In the Andes the condor, in the Alps the eagle 

29,003 feet. 














very highest 
and the vult- 


ure, hover above the topmost peaks. Fitted for the longest journeys, 
they are the greatest sailors in the atmospheric ocean, just as the petrels 
and the gigantic sea-swallows are the great sailors over the Atlantic. 
The choucas (a kind of jackdaw), with its intensely black plumage and 
yellow beak and red legs, does not rise so high into the atmosphere, 
but it is especially the bird of the highest peaks, of the region of snows 
and barren cones. It is met with at the summit of Monte Eosa and at 
the Col du Geant, at over 11,500 feet. 

There are also birds more graceful in form which live in the region 
of hoar-frost, and lend a little animation to those bleak and unchanging 
landscapes. The snow-chafiinch has so great a preference for this cold 
region that he rarely descends to the zone of the woods. The accenteur 
of the Alps also follows him to great elevations, preferring the stony 
and barren region which separates the zone of vegetation from that 
of perpetual snow, and both of these birds sometimes soar as high as 
11,000 to 15,000 feet in pursuit of insects. 

The engraviing (see Fig. 27) represents the principal kinds of birds 
according to the maximum height to which they fly. The earth has its 
birds, like the air. Certain kinds never use their wings but for a few 
moments, when it is impossible for them to move along the ground: 
for instance, all the gallinaceous kinds. The region of snow has its 
own kind, just as it has its characteristic sparrows. The ptarmigan, or 
snow-hen, is met with in Iceland as in Switzerland. It soars fjir above 
the everlasting hoar-frosts, and is so fond of the snow that at the ap- 
proach of summer it mounts farther in search of it, plunging into it 
with evident delight. A few lichens, grains brought up there by the 
air, suffice for its food. It looks for insects, with which it nourishes its 

The insects are, indeed, the only animals which are abundant in these 
bleak regions — a fresh analogy with the polar countries. It is also the 
class of coleoptera which predominate in the higher Alpine regions. 
They attain to 9800 feet on the southern slope, and to 7900 feet on the 
opposite side. Their wings are so short that they scarcely seem to have 
any ; one would imagine that nature had intended to protect them from 
the strong currents of air which would undoubtedlj^ carry them away 
if their wings had not been, so to speak, reefed. One does every now 
and then encounter other insects, neuroptera and butterflies, which the 
winds have taken up to these heights, and which are afterward lost 
amidst the snows. The seas of ice are covered with victims that have 


Fig. 27.— Distribution of kiuds of Birds according to heigtit of flight. 

Condor (has been seen as high as 9000 metres, or 29,600 feet) ; 2. Griffon ; 3. Vulture ; 4. Sarcoroniphus ; 5. Eagle ; 6. Urudu ; 7. Kite ; 8. Falcoj , 
9. Sparrow-hawlt ; 10. Fly-bird ; 11. Pigeon ; 12. Buzzard ; 13. Swallow ; 14. Heron ; 15. Crane ; 16. Duck and Swan (found In lakes at an altitude 
of 1800 metres, or 5900 feet) ; 17. Crow ; 18. Lark ; 19. Quail ; 20. Parrot ; 21. Partridges and Pheasants ; 22. Penguin. 




perished in this way. Nevertheless, there are certain kinds which ap- 
pear to travel freely as high as 13,000 or 16,800 feet. In my aerial 
voyages, I have met with butterflies at heights to which the birds of 
our latitudes do not ascend, and at more than 9800 feet above the 
ground. Dr. J. D. Hooker noticed some at Mount Momay, at an alti- 
tude of more than 17,700 feet. Such is the scale of animal life in these 
Alpine zones, where the fauna gradually becomes scarcer, finally giving 
way to solitude and desolation. Beyond the last stage of vegetation, 
beyond the extreme region attained by the insect and mammifers, all 
becomes silent and uninhabited ; yet the air is still full of microscopic 
animalcules, which the wind raises up like dust, and which are dissemi- 
nated to an unknown height. 




THE DAT. 103 



As the atmosphere is the organizer of life ; as all beings, animal and 
vegetable, are so constituted as to be able to breathe in its midst and 
construct, by means of its fluid molecules, the solid tissue of their or- 
ganisms, we must now turn our attention with admiration to the atmos- 
phere, as being still further the ornament of nature, and we shall see 
that we owe to it not only the picture, but also the frame. 

Whether the sky be clear or cloudy, it always seems to us to have 
the shape of an elliptic arch ; far from having the form of a circular 
arch, it always seems flattened and depressed above our heads, and 
gradually to become farther removed toward the horizon. Our ances- 
tors imagined that this blue vault was really what the eye would lead 
them to believe it to be ; but, as Voltaire remarks, this is about as rea- 
sonable as if a silk-worm took his web for the limits of the universe. 
The Greek astronomers represented it as formed of a solid crystal sub- 
stance ; and so recently as Copernicus, a large number of astronomers 
thought it was as solid as plate-glass. The Latin poets placed the di- 
vinities of Olympus and the stately mythological court upon this vault, 
above the planets and the fixed stars. Previous to the knowledge that 
the earth was moving in space, and that space is everywhere, theologi- 
ans had installed the Trinity in the empyrean, the angelic hierarchy, 
the saints, and all the heavenly host. ... A missionary of the Middle 
Ages even tells us that, in one of his voyages in search of the terres- 
trial paradise, he reached the horizon where the earth and the heavens 
met, and that he discovered a certain point where they were not joined 
together, and where, by stooping, he passed under the roof of the heav- 
ens. . . . And yet this vault has, in fact, no real existence ! I have 
myself risen higher in a balloon than the Greek Olympus was sup- 
posed to be situated, without being able to reach this limit, which, of 
course, recedes in proportion as one travels in pursuit of it — like the 
apples of Tantalus. 

What, then, is this blue, which certainly does exist, and which veils 
from us the stars during the day ? 


The vault which we behold is formed by the atmospheric strata 
which, in reflecting the light that emanates from the sun, interpose be- 
tween space and ourselves a sort of fluid veil, which varies in intensity 
and height with the density of the aerial zones. The illusion referred 
to above took a long time to dispel, and it was also a work of time to 
make it known that the shape and dimensions of the celestial vault 
change with the constitution of the atmosphere, with its state of trans- 
parency and its degree of illumination. One part of the celestial rays 
sent from the sun to our planet is absorbed by the air, the other part 
is reflected ; the air, nevertheless, does not act equally on all the col- 
ored rays of which white light is composed, but acts like a glass, al- 
lowing the rays toward the red end of the solar spectrum to pass more 
readily than those in the neighborhood of the blue end. This differ- 
ence is only perceptible when the light passes through a great thick- 
ness of air. De Saussure pointed out that the blue color of the sky 
was due to the reflection of light, and not to a hue peculiar to aerial par- 
ticles. "If the air were blue," he says, "the distant mountains, which 
are covered with snow, would appear blue also, which is not the case." 
An experiment made by Hassenfratz also proves that the blue ray is 
more reflected; in fact, the thicker the atmospheric stratum is which 
a ray traverses, the more do the blue rays disappear to give place to 
the red; and as, when the sun is near to the horizon, the ray has to 
traverse a greater thickness of air, the sun therefore appears red, purple, 
or yellow. The blue rays are also frequently absent in rainbows which 
make their appearance just before sunset. 

We shall see further on that it is the vapor of water accumulated in 
the air which plays the principal part in this reflection of the light, to 
which we owe the azure of the sky and the brightness of day. 

Very recently. Professor Tyndall reproduced the blue of the sky and 
the tint of the clouds in an experiment at the Eoyal Institution. Va- 
por of different substances, of nitrite of butylene, of benzone, and of 
carbonic sulphide, is introduced into a glass tube; a succession of elec- 
tric sparks is then passed through it, and the condensation and rarefac- 
tion of the vapor augmented ad libitum. As soon as the vapors em- 
ployed, no matter what their nature is, are sufficiently attenuated, the 
reflection of the light first manifests itself by the formation of a blue 
like that of the sky. There is, I will suppose, in the tube a half atmos- 
phere of air mixed with vapor, and another half atmosphere of air that 
has passed through hydrochlorfc acid. The proportion and density of 
the gas can, of course, be varied. 


THE DAT. 105 

The vaporish cloud, after having first assumed the blue tint, becomes 
more condensed and white, and as it thickens, becomes exactly like real 
cloud, presenting, as regards polarization, the same variation of phe- 

The atmospheric air is one of the most transparent bodies known. 
When it is not charged with mist or obscured by other bodies, we can 
see objects at an immense distance, and mountains only disappear from 
our view when they are below the horizon ; but, in spite of its slight 
power of absorption, the air is not completely transparent; its mole- 
cules absorb a portion of the light which they receive, permit the pas- 
sage of another part, and reflect the third; and hence it is that they 
give rise to what appears a vault, that they light up terrestrial objects 
which the sun does not reach directly, and effect an imperceptible tran- 
sition between day and night. 

It is easy to convince one's self of the decrease in the intensity of 
the solar light during its passage through the atmosphere by daily ob- 
servations. If an object situated near the horizon is watched for sev- 
eral days together, it will be seen that it is more visible at one time 
than at another. The distance at which its details fade out of sight is 
at one moment less than at another, as may be proved by direct meas- 
urement; the transparency of the air can be even expressed numeric- 
ally, as has been done by De Saussure through the instrumentality of 
the diaphanometer. The distance at which objects disappear does not 
depend upon the angle of vision alone, but also upon the manner of 
their illumination, and the contrast which their color offers to surround- 
ing objects. This explains why the stars, despite their small diameter, 
are so visible in the vault of heaven. It is the same with some terres- 
trial objects. It is difficult to distinguish a man, as he stands out in 
the fields, as against dark surfaces ; but he is very easily seen if he is 
placed upon an elevation so as to stand out against the clear sky. 
Hence the optical illusions so common in mountainous countries. 

While the chain of the Alps, seen from the plain at a great distance, 
is visible in its minute details, the spectator standing upon one of its 
peaks can distinguish hardly any thing in the plain. From the Faul- 
horn, for instance, it is easy to make out very distinctly the chain of 
the high Alps ; but every thing in the valley below is dim and con- 
fused. The summits of the Pilate, the Black Forest, and the Vosges 
are clearly defined at a great distance, whereas nothing can be distin- 
guished in the plain between the Alps and the Jura. Any one who 


has passed a few months amidst the lakes and mountains of Switzer- 
land must have noticed the same variations in the visibility of objects. 

To measure the intensity of the blue color, De Saussure invented the 
cyanometer, which is composed simply of a strip of paper divided into 
thirty rectangles, the first of which is of the deepest cobalt blue, while 
the last is nearly white, the intermediate colors offering every conceiv- 
able shade between dark blue and white. If it be found that the blue 
of one of these rectangles is identical with that of the sky, this identi- 
ty is then represented by a number corresponding to one of the rect- 
angles, and all that remains to be done is to arrange the scale of the 

Humboldt perfected this apparatus, and rendered it capable of giving 
very precise measurements of the blue tint. 

The mere contemplation of the heavens tells us that their color is not 
the same at every altitude, being generally deeper at the zenith, and 
gradually becoming lighter toward the horizon, where it is often nearly 
white. The contrast is rendered the more striking by the use of the 
cyanometer. Thus it will be found that sometimes the color corre- 
sponds to the number twenty-three in the neighborhood of the zenith, 
and to the number four near the horizon. But the color of the same 
part of the sky also changes pretty regularly during the day, as it be- 
comes darker from morning until noon, and lighter again from noon 
until evening. In our climates the deepest blue is when, after several 
days of rain, the wind drives away the clouds. 

The color of the sky is modified by the combination of three tints — 
the blue, which is reflected by the aerial particles ; the black of infinite 
space; and the white of the vesicles of mists and snow-flakes which 
float at the high elevations. If we rise sufficiently high in the atmos- 
phere, we leave a part of the vesicles of vapor below us. Thus the 
white rays reach the eye in a lesser proportion, and, the sky being cov- 
ered with fewer particles which reflect the light, its color becomes of a 
deeper blue. 

The nature of the ground also plays an important part in these effects 
of reflection and atmospheric transparency. 

In the regions where there are vast surfaces devoid of vegetation, as 
in a great part of Africa, the air is very dry, and loses part of its trans- 
parency, especially in consequence of the dust borne by the winds and 
the absence of heavy rain to cleanse the air. In the other parts of the 
intertropical zone, upon the Atlantic, on the American continent, in the 

THE DAT. 107 

South Sea Islands, and in certain regions of India, aqueous vapor, in a 
state of transparent gas, is abundantly mixed with the air ; and in place 
of the grayish hue which it possesses in our climates and in sandy des- 
erts, the sky presents a strongly-marked tint of azure blue, which spe- 
cially characterizes the regions about the zenith, and sometimes even 
the sky near the horizon. 

The limiting surface of the atmosphere being parallel to that of the 
earth, and the visible portion being that only which is above the plane 
of the horizon, it is clear that rays of light reaching the eye in different 
directions have traversed different thicknesses of air. If the sun were 
at the zenith, its rays would pass through the thinnest stratum of air ; 
the nearer Ihe sun approaches the horizon, the thicker becomes the mass 
of air which its rays have to pierce, and consequently the weaker its 
rays become. The light of the sun at its meridian passage is dazzling, 
whereas we can look at it with the naked eye when near the horizon ; 
and for the same reason the regions situated near the horizon seem 
always to be without stars. 

The color of the sky is thus explained by the reflection of light upon 
the molecules of the vapor of water which invisibly pervades the air. 

How are we now to explain the very perceptible shape of an ellip- 
tical vault which the sky presents, whether cloudy or entirely clear? 

This may be explained as a simple effect of perspective. 

I will suppose we have before us an avenue of poplars, all of the 
same height. Every one knows that this height will apparently de- 
crease with distance, and that the top of the trees at the extreme end of 
the avenue will appear to be at the height of our eyes. 

The trees' roots are upon a horizontal surface, because we ourselves 
are upon the ground. It is by the top line that the inclination toward 
the ground operates. If we were in the upper branches of the nearest 
tree, then it would he from below that the perspective inclination would 
operate. The same train of reasoning may be applied to the clouds. 
Starting from those which are vertically above our heads, they succes- 
sively decline in height according to their distances above the horizon. 

When we are above the clouds in a balloon, they no longer seem to 
sink toward the earth like a vault, but to extend like the plane surface 
of an immense ocean of snow. When but a few miles above them, they 
describe a curve in the contrary direction.* 

* [Having been led theoretically to expect such a phenomenon, I always, when some miles 
above the clouds, attentively looked for its appearance, and invariably without success. It 


With a clear sky, the surface of the earth, seen from a great height, 
is liollow underneath the car of a balloon, and gradually rises around up 
to the circular horizon. Far from appearing convex, as might be ex- 
pected if one imagined that at a great height in the atmosphere the 
spherical shape of the globe would be recognized, the surface of the 
o-round is hollowed out underneath us, rising till it reaches the horizon, 
which seems always to be on a level with the eye. 

This aspect of the earth, hollowed out like a basin, surprised me very 
much the first time I saw it from a balloon, for at the height which I 
had attained I had expected to see it convex. 

Thus the sinking of the apparent vault of the sky above our heads is 
due to an effect of perspective, as we can not estimate vertical heights 
in the same way as horizontal lengths. A tree forty-five feet high 
seems much longer on the ground than when standing. A tower three 
hundred feet high would appear far more if laid along the ground than 
when vertical. Being in the habit of walking along the ground, and 
not of soaring into the air, we appreciate lengths at their true estimate, 
whereas heights are beyond our powers of direct judgment. 

It results from the apparent shape of the celestial vault that the con- 
stellations seem to us much larger toward the horizon than at the ze- 
nith (as, for instance, the Great Bear when it skirts the horizon, and 
Orion when he rises), and that the sun and the moon appear to have 
larger disks at their rising and setting than at their culminating points. 
It further results that we are constantly in error in estimating the 
height of stars above the horizon. A star which is at 45° of altitude — 
that is, just half-way between the horizon and the zenith — seems to us 
much higher ; and when we point out a star as being at 45°, it m.ay 
happen that it is only at 80°.* 

Modern treatises on physics and meteorology have not gone into this 

is trae that, the dip of the horizon being very small, objects on the horizon practically ap- 
pear to be on the same level as the eye, while the ground underneath of course seems far be- 
low, so that, in this sense, the appearance of the earth is cup-shaped. But, in point of fact, 
if the day be clear, the distance of the horizon is so much greater than is that of the ground 
below, that the effect is no more noticeable than it is from the top of a hill. If the air be not 
clear, all traces of the appearance are of course absent. — Ed.] 

* [Most people imagine they are looking at the zenith when they are looking at a point 10° 
or 20° below it, and on this account their estimates of heights are too great. As regards the 
shape of the celestial sphere, it may be remarked that the distance to the horizon would appear 
greater than to the zenith, if it were only because of the intervening objects which occur in the 
former case; while, looking upward, there is nothing to aid the eye in its estimation. — Ed.] 


THE DAT. 109 

curious question of the aspect of the sky. I find it discussed in certain 
works of the seventeenth and eighteenth centuries, but rather from a 
philosophical point of view than in its purely geometrical aspect. Af- 
ter a long dispute between Mallebranche and Eegis Upon this point, 
Eobert Smith examined it in his "Optics" (1728), and concluded that 
the horizontal diameter of the celestial vault must seem to us six times 
as long as the vertical diameter. He is of opinion that this is due to 
the fact that "our view does not extend distinctly to the point at which 
the objects form an angle of the 8000th part of an inch in our eye, so 
that all objects seem to us to sink under the horizon at a distance of 
25,000 yards." 

The mathematician Euler, in his "Letters to a German Princess" 
(1762), devotes several chapters to an explanation of it, which may be 
stated in a few words. First, the light of the stars which are near the 
horizon is much weakened, because their rays have a greater distance 
to travel through our lower atmosphere than those which are at a great- 
er height; secondly, being less luminous, we deem them to be farther 
off, because we always take the objects which are most clear to be near- 
est to us (for instance, a conflagration at night seems much closer to us 
than it really is) ; thirdly, this apparent distance of the celestial objects 
which are near the horizon gives rise to the imaginary elliptic vault of 
the heavens. 

The logical arrangement of these last two points seems the inverse 
of the theory explained above, yet it may be seen that these two facts 
do not follow the one from the other, but are simultaneous in our ob- 
servation. Perspective is due to the distance and to the diminution in 
brightness, and it gives a clear explanation of the apparent shape pre- 
sented by the atmospheric strata, and the variation in size according to 
the elevation above the horizon. There is, so to speak, a double effect 
of geometrical and luminous perspective. 

We do not appreciate the beauty or the practical importance of the 
diffusion of light by the air, because it is always present to us. A so- 
journ of a few hours in our neighbor the moon would suffice to show 
us the enormous difference there is between an atmospheric day and 
one without air. 

As Biot remarked, in a very correct simile, the air is around the 
earth a sort of brilliant veil, which multiplies and disperses the sunlight 
by an infinity of repercussions. It is to it that we owe the light which 
we enjoy when the sun is below the horizon. After the latter has risen 


there is no spot so secluded, provided the air has access to it, which 
does not receive some light, although the sun's rays may not reach it 
directly. If the atmosphere did not exist, each point of the terrestrial 
surface would only receive the light reaching it directly from the sun. 

The strange effect of the absence of the atmosphere would be far 
more complete and striking if we had the power of transporting our- 
selves into our satellite. Let us compare the cheerful spectacle that 
the earth presents, partly covered with its humid and wavy mantle, 
and decked with flowers, to the aspect of the moon, with its stony or 
metallic surface, abounding with crevasses and vast mountainous des- 
erts, with its extinct volcanoes and peaks that seem like gigantic tombs, 
with its sky invariably black and shapeless, in which reign, day and 
night, stars without scintillation, the sun and the earth. There day- 
time is, so to speak, nothing but night lighted up by a rayless sun, 
No dawn in the morning, no twilight in the evening. The nights are 
pitch-dark. Those parts of the lunar hemisphere which are toward us 
are lighted by an earth-light, the first quarter of which coincides with 
sunset, the full earth with midnight, and the neio earth with sunrise.* 
In day-time the solar rays are lost against the jagged ridges, the sharp 
points of the rocks, or the steep sides of their abysses, designing here 
and there grotesque shapes against the angular contours, and only strik- 
ing the surfaces exposed to their action to become at once reflected and 
lose themselves in space — fantastic shadows standing out in the midst 
of a sepulchral world. Fig. 28 represents a landscape taken in the 
moon, in the centre of the mountainous region of Aristarchus. There 
is nothing but white and black. The rocks reflect passively the light 
of the sun ; the craters remain partially wrapped in shade ; fantastic 
steeples seem to stand out like phantoms in this glacial cemetery ; the 
absence of the atmosphere leaves the black space of the starry heaven 
perpetually hanging over this dismal region, to which, fortunately, the 
earth can offer no sort of analogy. 

* [The moon always turns the same fiice to the earth ; so that there is one-half of the 
moon's surface that has never been seen from the earth. The words one-half must not be 
taken quite literally, as, owing to a slight oscillatory motion of the moon, called libration, 
we sometimes see a little more round the comer, as it were, than at other times. Speaking 
generally, therefore, an inhabitant of the moon, if he saw the earth at all {i. e., was on the 
hemisphere turned toward us), would always see it in the same position in the sky (and in 
size about four times as large as the moon appears to us). The statement in the text is only 
true for a spectator placed at the middle point of the visible hemisphere of the moon ; the 
lunar day is of course about four weeks.— Ed.] 

Fig. 2S.— Luuar Day. 




Light, that imponderable agent which enables us to see objects, and 
which by its qualities illuminates the magnificent atmospheric world in 
which we live, gives rise to an ever-changing series of effects. The at- 
mosphere not only bathes the landscape with light by reflection, but 
also decomposes it by refraction, and gives additional variety to the 
beauties of the earth and sky. 

When a ray of light passes from one transparent medium to another, 
it undergoes a deviation caused by the difference of density of the two 
media.* In passing from air to water the ray is bent toward the ver- 
tical, because water is denser than air. It is the same with a ray which 
passes from a higher to a lower stratum of air, for, as we have seen, the 
lower strata are denser than those above. 

If a ray of common light be admitted through a small hole in a dark- 
ened room, and, after passing through a glass prism, be received on a 
screen, it will be seen that the ray of white light has been decomposed 
by refraction through the prism into seven colors — violet, indigo, 
brown, green, yellow, orange, red — which occupy different positions, in 
the above order, on the screen. The red rays, being the least bent from 
the direction of the original ray, are said to be least refrangible, and the 
violet rays, which form the other end of the spectrum, are said to be 
most refrangible. 

In refracting light the air produces two distinct effects. On the one 
hand, it causes a ray of light which has its origin beyond the earth's at- 

* [M. Flammarion here adds the sentence, "A stick plunged into water appears bent at the 
surface of the liquid, and the immersed portion appears more nearly vertical." As this illus- 
tration of the eiFect of refraction is given in many popular works, I think it worth while to 
point out its inac^curacy. A ray of light entering a densier fluid (the surface of which is hori- 
zontal) is bent nearer to the vertical ; but a stick is not a ray of light, and in no way resembles 
one. The immersed portion of the stick is seen by rays that have been refracted at the surface 
of the water ; and it easily follows, from the principles of optics, that the part under water ap- 
pears hQXii from (not toward) the vertical. This any one can verify for himself experimentally. 
The sentence quoted above is therefore not only erroneous in theory, but also incorrect in fact. 
The apparent bending of the stick is only indirectly due to refraction. — Ed.] 



mosphere to become bent as it approaches the earth, so that we see the 
sun, moon, planets, comets, and the stars, as if they were higher in the 
heavens than they really are. On the other hand, it causes a more or 
less considerable separation between the various rays that constitute 
white light, according to its state of transparency and density. 

The first effect mainly produces twilight ; the second gives that soft, 
undulating beauty which is seen in the serenity of the evening, 

Eefraction is greater or less, in proportion as the luminous ray trav- 
erses the atmosphere in a direction more or less inclined to the ver- 
tical, being greatest for horizontal and vanishing for vertical rays. As- 
tronomical observations would all be false with regard to the positions 
of objects if they were not corrected for the effect of refraction. Thus, 

Fig. 29.— Atmospheric refraction. 

for instance, the star a is seen at a'; the star B at b'; at the zenith 
alone stars are where they appear to be, there being no alteration in the 
direction of the ray of light due to refraction. To make these neces- 
sary corrections, tables have been constructed giving refractions, based 
upon the hypothesis of a uniform disposition of the different strata 
of air lying one above the other. The refracting power of the air is 
determined on the hypothesis that it contains only oxygen and nitro- 
gen ; but we have seen that it further contains from four to six parts in 
10,000 of carbonic acid, and an ever-varying quantity of the vapor of 
water. The refracting power of the vapor of water differs so little from 
that of air properly so called, that the correction depending on it need 
not, as a rule, be taken into the calculation. 

To calculate the amount of correction to be applied to any observa- 



tion, it is only necessary to note at the time the temperature of the air 
and the pressure of the atmosphere at the place of observation. 

To illustrate the effect of refraction, I have selected from a table of 
refractions a few numbers, at different zenith distances. They show to 
what extent objects are apparently raised by its influence : 




from the 


from tlie 




90 deg. 

33 min. 47 sec. 

1 74 deg. 

3 mill. 20 sec. 

89 " 

24 " 22 " 

72 " 

2 '• 57 " 

88 " 

18 " 23 " 

1 70 " 

2 " 38 " 

87 " 

14 " 28 " 

' 65 " 

2 " 4 " 

86 " 

11 " 48 " 

60 " 

■ I " 40 " 

85 " 

9 " 54 " 

: 55 " 

1 " 23 " 

84 " 

8 " 30 " 

50 " 

1 " 9 " 

83 " 

7 " 25 " 

45 " 

" 58 " 

82 " 

6 " 34 " 

40 " 

" 48 " 

81 " 

5 " .53 " 

30 " 

" 33 " 

80 " 

5 " 20 " 

20 " 

" 21 " 

78 " 

4 " 28 " 

10 " 

" 10 " 

76 " 

3 " 50 " 

, " 

" " 

From this table we see that an object situated just upon the horizon 
is raised by more than 33', or about t4tj of ^^^ distance from the hori- 
zon to the zenith. Neither the sun nor the moon is 33' in diameter. 
When, therefore, they appear to have just risen, they are still entirely 
below the horizon. In the same way, the sun does not appear to begin 
to set until after sunset has actually taken place. 

It follows from these considerations that the sun may be seen in the 
west and the moon in the east at the time of full moon, and even an 
eclipse of the moon may be visible while the sun is still above the hori- 
zon, although the earth is then exactly between the two luminaries, and 
the latter are both, astronomically speaking, below the horizon. This 
is due to refraction. This curious circumstance was noted during 
eclipses of the moon on June 16, 1666, and May 26, 1668. 

Owing: to the same cause, the sun and the moon seem to be flattened 
both at their rising and setting, the rays proceeding from the lower edge 
of the luminary being more refracted than those proceeding from the 
upper, so that the apparent vertical diameter is diminished, while the 
horizontal diameter remains, of course, unaltered. The length of the 
day is thus increased, and that of the night decreased. It is for this rea- 
son that at Paris the longest day of the year is sixteen hours seven min- 
utes, and the shortest eight hours eleven minutes, instead of being fif- 


teen hours fifty-eight minutes, and eight hours two minutes. We see 
that the length of the day at Paris at the time of the solstices is thus 
prolonged by nine minutes, and by seven minutes at the equinoxes. 
At the North Pole the sun seems to be in the horizon, not when it ar- 
rives at the spring equinox, nor when its angular distance from the 
North Pole is 90°, but when it is 90° 33'; it then remains visible until, 
having passed to the autumnal equinox, its polar distance has again be- 
come equal to 90° 33'. Care must always be taken to keep account of 
refraction in calculating the hours of sunrise and sunset. 

Twilight is that light which remains after the sun has set or which is 
seen before sunrise. The duration of twilight is, in many respects, a 
useful element to be acquainted with. It depends chiefly upon the an- 
gle to which the sun has descended below the horizon ; but it is modi- 
fled by several circumstances, the chief of which is the degree of clear- 
ness of the atmosphere. The direct light of the sun at the time of sun- 
set reaches to the west ; as the sun sinks, its boundary-line rises, and 
some little time afterward crosses the zenith, when civil twilight ends; 
the planets and large stars then become visible to the naked eye. The 
eastern half of the sky being thus first deprived of direct solar light, 
night begins there. Afterward, the boundary -line (the crepuscular 
curve) itself disappears in the west; then the astronomical twilight 
ceases and night has fully set in. Twilight begins or ends when the 
sun is at a certain distance below the horizon ; this distance is variable, 
depending upon the state of the atmosphere. It may be taken that civil 
twilight ends when the sun is about 8° below the horizon, and that as- 
tronomical twilight ends when the sun is about 18° below the horizon. 
The phenomena of twilight are hardly known in tropical climates; as 
soon as the sun has descended below the horizon, darkness sets in sud- 
denly. This was remarked by Bruce at Senegal, where, however, the 
air is so transparent that Yen us may sometimes be distinguished at mid- 
day, and in the interior of Africa night succeeds day almost immedi- 
ately after sunset. At Cumana, Humboldt tells us, twilight lasts but a 
very few minutes, although the atmosphere is higher under the tropics 
than in other regions. 

The following tables give the length of the civil and astronomical 

'twilight in France for the various seasons and for the fifteenth day of 

each month. By adding the duration of twilight to the hour of sunset, 

the time at which each of the twilights terminates is readily obtained, 

and subtracting it from the hour of sunrise, the times of their com- 



raencement are found. France, from the Pyrenees to Dunkirk, is with- 
in the 41st and 42d degrees of latitude. It will be seen that, even 
within these trifling limits, there is a perceptible difference." The short- 
est civil twilights take place on the 29th of September and the 15th of 
March, the longest on the 21st of June; the shortest astronomical twi- 
lights fall upon the 7th of October and the 6th of March, the longest on 
the 21st of June. North of 50° latitude, the astronomical twilight con- 
tinues all night for some time both before and after the summer solstice. 



Length of the Day. 

The longest : 
June 21. 

The shortest: 
December 21. 

42 degrees. 


46 " 


15 hrs. 13 min. 

15 " 28 " 

15 " 44 " 

16 " 2 " 

16 " 24 " 

9 hrs. 

8 " 47 min. 
8 " 30 " 
8 " 14 " 

7 " .55 " 












42 deg. 

34 min. 












44 deg. 

35 min. 
33 " 

32 " 

33 " 

36 " 
39 " 
38 " 

34 " 
32 " 
32 " 
34 " 
36 " 


36 min. 

34 " 

33 " 

34 " 

38 " 
41 " 

39 " 

36 " 
33 " 
33 " 

35 " 

37 " 

48 deg. 

38 min. 

35 " 
34 " 

36 " 
40 " 
44 " 
42 " 

37 " 

34 " 

35 " 
37 " 

39 " 


40 min. 















42 deg. 

44 deg. 

46 deg. 

48 deg. 

50 deg. 

n. M. 

n. M. 

H. M. 

11. M. 

H. M. 


1 31 

1 33 

1 36 

1 40 

1 45 


1 24 

1 26 

1 29 

1 32 

1 36 


1 24 

1 26 

1 29 

1 33 

1 37 


1 33 

1 35 

1 39 

1 44 

1 50 


1 46 

1 52 

2 1 

2 11 

2 26 


1 56 

1 48 

2 5 
1 54 

2 19 
2 4 

2 36 
2 14 

3 13 
2 31 



1 32 

1 37 

1 42 

1 47 

1 54 


1 24 

1 26 

1 30 

1 34 

1 38 


1 23 

1 25 

1 29 

1 33 

1 36 


1 30 

1 32 

1 35 

1 39 

1 43 


1 34 

1 36 

1 40 

1 45 

1 50 


In warm countries, the presence of humidity in the air not only gives 
to the sky its dark azure tint, but also has the effect of modifying the 
vital power of tlie solar rays. At the equator it adds to the thousand 
other wonders of nature an incomparably beautiful display of light both 
at sunrise and sunset. The sunset, in particular, affords a spectacle 
indescribably magnificent — a superiority over sunrise attributable to 
the presence of moisture in the air. This is more abundant in the 
evening, after the heat of the day, than in the morning, when it is par- 
tially condensed into dew by the effect of the cooler temperature of 

It is not in our climate that the fi:nest sunsets are seen. The celestial 
blue of distant mountains, the rose or violet tints which in turn tinge 
the nearer hills, and the warm tones of the soil, harmonize in a mar- 
velous manner, when the sun disappears below the horizon, with the 
gleaming gold of the west, the red or roseate tints that crown it in the 
sky, the dark azure of the zenith, and the more sombre and often, in 
contrast to the others, greenish hue which prevails in the east. In the 
equinoctial regions, these soft and delicate tints, joined to the varied as- 
pect of the earth's configuration and the richness of vegetation, produce 
more striking effects than with us. Sometimes light and roseate clouds, 
fringed with a coppery red, produce peculiar effects, similar to certain 
sunsets in our regions ; but whenever the sky is clear the shades differ 
entirely from those of the temperate zone, and present a special charac- 
ter. Sometimes, too, the indentations of mountains situated below the 
horizon, or invisible clouds intercepting a part of the solar rays which, 
after sunset, still reach the elevated regions of the atmosphere, give rise 
to the curious phenomenon of crepuscular rays. Then may be seen, 
starting from the point where the sun has disappeared, a series of rays, 
or rather of diverging "glories," which sometimes extend as far as 90°, 
and even in some instances are prolonged as far as the point opposite 
to the sun. "Upon the ocean," as M. Liais remarks, "when the sky, 
near the equator, is free from cloud in the visible part, and when the 
diverging rays mingle with the crepuscular arcs, the play of light as- 
sumes a form and brilliance which defy all description or pictorial illus- 
tration. How, indeed, is it possible to depict completely the rosy tints 
of the arc fringed by the crepuscular rays that border the segment 
which is still strongly lighted up from the west, the segment itself being 
tinged with a bright gold hue? How, above all, is it possible to de- 
scribe the tint of an inimitable blue, different from that of noonday, and 


occupying that portion of the sky which is included between the ordi- 
nary azure and the crepuscular arc? 

"To all this splendor of the western sky must be added the descrip- 
tion of its fires as reflected upon the surface of the waters agitated by 
the trade- wind, the dark blue color of the sea to the east, the white foam 
of the wave, which sharply defines upon this gloomy background the 
pale roseate arc of the eastern sky, and the sombre and greenish seg- 
ment of the horizon." 

What spectacle can be more sublime than a sunset at sea? We have 
attempted in the illustration to recall this beautiful spectacle. The col- 
ored clouds which float in this western sky are cirro-cumuli^ which will 
be described in the chapter upon the Clouds. 

The setting sun is nearly always accompanied by these cirro-cumuli 
clouds, which serve to display those aspects of the sky which are of so 
remarkable a beauty in the west. In consequence of the curvature of 
the earth, sea-clouds which are sometimes seen from Paris are more 
than two miles above the ocean, and are formed of ice and snow, even 
in the month of July. These are nearly the highest clouds, and pro- 
duce the varied forms of mountains, fishes, animals, and other fantastic 
shapes, which one may discern of an evening upon a bright and rich 
ground of every tint that light can give. 

To the preceding remarks may be added one of a more general and 
curious nature, in reference to the influence of the evening light in the 
construction of cities. Towns grow in a westward direction. Paris, 
the cradle of which was the Ih de la Cite^ has, in its successive aggran- 
dizements, constantly extended toward the west. Two thousand years 
ago, Paris was situated on the north-east slope of Mount St. Genevieve, 
where the arenas have recently been discovered. Under the Merovin- 
gians it commenced its descent toward the west, and has unceasingly 
progressed in that direction ever since. The wealthy classes have a 
pronounced tendency to emigrate westward, leaving the eastern districts 
for the laboring populations. This remark applies not only to Paris, 
but to most great cities — London, Vienna, Berlin, St. Petersburg, Turin, 
Li6ge, Toulouse, Montpellier, Caen, and even Pompeii. 

Whence arises this tendency ? A fact so universal can not be due to 
accident. Is it the stream of the Seine which has taken Paris westward 
in its wake? Not so, for the Thames flows in a contrary direction, 
while London has none the less extended to the west like Paris. Twelve 
years ago. Doctor Junod {Comptes-Rendus of the Academy of Sciences 


ia 1858) offered, as an explanation of this fact, the statement that the 
east wind is that which raises in the greatest degree the barometrical 
column, while the west wind lowers it the most, and therefore inundates 
the eastern part of a town with deleterious gases, so that the latter has 
to put up not only with its own smoke and miasmas, but also with those 
coming from the western portion. It may, in fact, be admitted that 
people prefer going where fresh air is to be found, and in the direction 
from which the wind blows most frequently. 

But the wind is not the same in all countries. For my own part, I 
am more inclined to see in this fact an evidence of the attraction of 
light. And the suggestion is an extremely simple one. It may be re- 
marked that people, as a rule, take their promenade of an evening, and 
not of a morning, and always, or nearly always, in the direction of sun- 
set. This disposition has led to the formation of gardens, country 
houses, and places of public resort, and, little by little, the wealthy pop- 
ulation of a large city extends in this direction. 


^V - •'■n*-.^ -^s^ 





The general action of light in nature is always evident to our eyes ; 
its effects in the atmosphere are of very different kinds, and produce 
a thousand optical phenomena, always curious, often fantastic, but all 
capable of explanation in these days by physical laws. We shall de- 
vote the following chapters to the examination of the phenomena that 
are due to this agent, at once so powerful and so delicate. 

The most common of these phenomena is the rainbow, and the ex- 
planation of it will aid us in understanding the others. 

There are few persons who have not remarked in water foiling from 
a fountain or cascade the production of a miniature rainbow analogous 
to the majestic arch which crosses the sky. Whenever these small 
rainbows are seen, three circumstances will be observed in connection 
with them : first, that drops of water must be present ; secondly, that 
the sun must be shining ; and, thirdly, that the observer must be be- 
tween the sun and the water. 

These three conditions in regard to the production of the rainbow will 
explain the phenomenon in which the Jewish religion saw the presence 
of Jehovah, and the Greek mythology the auspicious influence of the 
goddess Iris. In order to see a rainbow as a result of the action of light, 
whether on artificial rain or on 
the drops of rain falling from 
the clouds in the atmosphere, 
the spectator's back must be to 
the sun. In this position, the 
solar rays which shine upon 
the drops of water are reflected 
and refracted as follows : Let us 
suppose a drop of water, a i i', 
in the atmosphere. A solar ray 
reaches this drop at i, and pass- 
es into it, being deflected from 

a straight line by refraction, Pig. SO.-Simple reflection of rays in a drop of rain. 


inasmuch as the ray passes into a medium of different density. Arriv- 
ing at A on the surface of the small sphere of liquid which constitutes 
the drop, it is reflected and returns in the direction of A i', being refract- 
ed on emergence into the direction l' M. 

This ray, so decomposed by refraction, presents all the colors ar- 
ranged in regular order, as each color possesses a different degree of re- 
frangibility. The inclination increases from red to violet; that is to 
say, that if the red ray from a particular drop reaches the eye, the other 
rays proceeding from the same drop can not reach it too; but a drop at 
a less elevation in the air can send a violet ray which will be visible at 
the same time. Thus, the observer sees, in the direction of these drops, 
a red hue above and a violet hue below. The intermediate drops simi- 
larly emit rays which, when seen by the eye, are of the colors included 
between red and violet, forming a solar spectrum, the colors of which, 
starting from the lowest arc, are violet, indigo, blue, green, yellow, orange, 

Let us now imagine a conical surface passing through the drop, and 
having for axis the straight line drawn from the eye of the observer to 
the sun. Every drop of water which is upon this surface of the cone 
produces the same effect, so that there is a mass of spectra forming a 
circular band, in which the simple colors succeed each other in the or- 
der indicated, the violet, a (see Fig. 83, p. 124), being inside, and the red, 
h, outside. 

The phenomenon continues as long as the drops of water go on fall- 
ing in the same region of space, the luminous appearance being inces- 
santly renewed by the falling of the drops, so that the arch appears per- 
manent while the rain lasts. 

Calculation has shown that the angle of the cone of the red rays is 
42° 20', and that of the violet rays 40° 30'. This is, therefore, the dis- 
tance from the arc to the centre or the point of the sky on which the 
shadow of the head of the spectator, P (see Fig. 33), would be cast. The 
diameter, h h' (see Fig. 33), of the whole arc subtends an angle of about 
84°, the width of the arc being 2°, or nearly four times the apparent 
diameter of the sun. 

The rainbow, therefore, demonstrates the existence of small spheres 
of liquid water, falling as rain in the midst of the atmosphere. The 
arch is more brilliant as their size increases. They must be much 
larger than those which form the clouds for the eye to be able to distin- 
guish the colors, and that is the reason why mists and clouds do not 



produce any rainbow. Knowing that the rainbow is caused by the re- 
fraction of the sun's rays through drops of rain as they fall, we may de- 
duce therefrom not only the size of this arch, but also the conditions 
without which it could not exist. If the sun were on the horizon, the 
shadow of the spectator's head would be cast there also, and as the axis 
of the cone would be horizontal, it follows that we should see a semi- 
circle of an apparent radius of 41°. As the sun rises, the axis of the 
cone is inclined, and the arch becomes smaller; and finally, when the 
sun reaches a height of 41°, the axis of the cone forms the same angle 
with the plane of the horizon, and the top of the arch just touches this 

Fig. 31.— Formation of the rainbow. 

latter plane. If the sun were still higher, the arch would be projected 
upon the ground. The phenomenon is rarely visible under this last con- 
dition. The secondary rainbow, of which I am about to speak, disap- 
pears when the sun reaches an altitude of 52°, for which reason a rain- 
bow can not be seen at noon in summer. The observer standing upon 
the earth can, therefore, never see more than half a circumference (viz., 
when the sun is on the horizon); and, as a rule, the arch is only 100° 
to 150° in length. When the earth does not stand in the way of the 
production of the lower part, more than a semi-circumference, and even 
a whole circumference, may be seen. This occurred to me once in a 



balloon ; and by a curious coincidence (the upper part being concealed), 

I saw a 7'ainbow upside down^ in 
which the violet color was inside. 
A second arch, in which the 
colors appear in an inverse order 
to those in the rainbow described 
above, is frequently remarked. 
This second arch is explained by 
a double reflection, s i A B i' M (see 
Fig. 32) and s'a'o, ^'h'o (see Fig. 

Fig. 32.-Double reflection of rays in a drop of rain, ggy j^ ^|^ jg ^^g^^ ^|^g deviations 

of the rays after they emerge from the liquid sphere are 51° for the red 
rays, and 54° for the violet rays. This secondary arch is always paler 
than the first. 

The zone comprised between the principal and the secondary arch is 
generally darker than the rest of the sky, and appears to me, after nu- 
merous observations, to be a region of absorption for the luminous rays. 

It is ascertained that a 
larger number of reflections 
may be produced, and that 
other arches, more and more 
pale in hue, may exist. But 
the diffused light prevents 
them from being seen. How- 
ever, a third has been seen, 
at 40° from the sun. By 
causing the solar rays to fall 
upon a jet of water in a dark 
place, as many as seventeen 
rainbows have been counted. 

It ma}^ happen that the 
sun is reflected toward a 
cloud by the surface of a 
piece of still water; and 
then this reflection will also 
give rise to a rainbow. It has been found that in this case the rain- 
bow must cut the arch formed by the direct rays at a height dependent 
upon that of the sun. If the two phenomena produce a secondary arch, 
the four curves intertwined form a very beautiful spectacle. A case in 

Fig. 33.— Theory of the two arches of a rainbow. 



which they were quite complete and perfectly distinct is cited by 
Monge. Halley observed three arches, one of which was formed by the 
rays reflected upon a river. This arch first intersected the exterior 
arch so as to divide it into three equal parts. When the sun sunk to- 
ward the horizon, the points of meeting were drawn close together. 
There soon was seen but one single arch, and as the colors were in in- 
verse order, pure white was formed by the superposition of the two se- 
ries. The sun, too, may produce, after being reflected upon a piece of 
water, a complete circle, the upper part of which being sometimes invis- 

Fig. 34 Triple rainbow. 

ible, gives rise to the singular phenomenon of a rainbow upside down. 
The Academicians dispatched to the polar regions to measure an arc of 
the meridian, observed upon the Ketima Mountain, on July 17, 1736, a 
triple rainbow analogous to that of which Halley speaks. In the lower 
bow the violet was underneath, the red outside as usual : this was the 
principal arch. The second, which was^parallel to it, was the secondary 
arch. In this the red was underneath and the violet at the top. The 
third arch, starting from the extremities of the first, ci'ossed the second. 


and had, like the principal one, the violet inside and the red outside. 
This is the phenomenon drawn in Fig. 34. 

Seeino-, then, that the rainbow is due to the refraction and reflection 
of the solar rays upon little drops of water falling in the air, it is easy 
to conceive that moonlight may cause an analogous appearance, though 
less intense; and this indeed is the case, though a lunar rainbow is not 
very common. The illustration represents a lunar rainbow which I 
had an opportunity of remarking one spring evening at Compi^gne. 

Many observers have remarked and described this nocturnal rain- 
bow. I gather from the writings of Americ Yespuce (1501) that he 
had several times observed " the iris at night." He considers that the 
red of the arch is due to fire, the green to the earth, the white to the 
air, and the blue to the water; and, he adds, "this sign will cease to ap- 
pear when the elements are used up, forty years before the end of the 

I notice in an ancient treatise on meteorology (that of P. Cotte) that, 
in addition to the ordinary rainbow, the secondary rainbow, the reflect- 
ed arches, and the lunar rainbow, there has been mentioned yet another 
optical effect, called the " marine rainbow," formed upon the surface of 
the sea, and composed of a large number of zones. It sometimes ap- 
pears upon wet meadows lying opposite to the sun. This fifth aspect is 
a kind of anthelion, which I will allude to in the next chapter. The 
name of "white rainbow" has also been given to the anthelical circle, 
which will also be considered in the same chapter. 

Lastly, there are sometimes seen colored bands below the violet of 
the ordinary rainbow, which appear to belong to an arch lying over the 
first. This arch then takes the name o{ supernumerary arch, and is due 
to very complex effects of interference of light, explainable on the un- 
dulatory theory. 

The first person who attempted to explain the phenomenon of the 
rainbow by the reflection of light upon the interior of the drops of rain 
was a German monk of the name of Theodoric ; the second an arch- 
bishop, A. De Dominis (1611). But the true theory was first given by 
Descartes, with the exception of the separation of colors, which was only 
determined by the discovery of Newton as to the unequal refrangibility 
of the rays of the solar spectrum. 

A. Marie pime' £i*HI- Oie^ ekromoUth^ 





Treatises on meteorology have not, up to the present day, classified 
with sufficient regularity the diverse optical phenomena of the air. 
Some of these phenomena have, however, been seen but rarely, and 
have not been sufficiently studied to admit of their classification. We 
have examined the common phenomenon of the rainbow, and we have 
seen that it is due to the refraction and reflection of light on drops of 
water, and that it is seen upon the opposite side .of the sky to the sun in 
day-time or the moon at night. We are now about to consider an order 
of phenomena which are of rarer occurrence, but which have this prop- 
erty in common with the rainbow, viz., that they take place also upon 
the side of the sky opposite to the sun. These difierent optical effects 
are classed together under the name of anthelia (from avQi, opposite to, 
and n\LOQ, the sun). The optical phenomena which occur on the same 
side as, or around the sun, such as halos, parhelia, etc., will form the 
subject of the next chapter. 

Before coming to the anthelia, properly so called, or to the colored 
rings which appear around a shadow, it is as well first to note the effects 
produced on the clouds and mists that are facing the sun when it rises 
or sets. 

Upon high mountains, the shadow of the mountain is often seen 
thrown either upon the surface of the lower mists or upon the neigh- 
boring mountains, and projected opposite to the sun almost horizon- 
tally. I once saw the shadow of the Righi very distinctly traced upon 
Mount Pilate, which is situated to the west of the Righi, on the other 
side of the Lake of Lucerne. This phenomenon occurs a few minutes 
after sunrise, and the triangular form of Righi is delineated in a shape 
very easy to recognize. 

The shadow of Mont Blanc is discerned more easily at sunset. MM. 
Bravais and Martins, in one of their scientific ascents, noticed it under 
specially favorable circumstances, the shadow being thrown upon the 
snow-covered mountains, and gradually rising in the atmosphere until it 

this the regulnr contC' 
lallj at their upper edge. 

: lines to converge the orie to I. 
• ''"!ow of Mont Blano- t1, it ■ 
■iws of our own - 

.'he na? 

olcome* ; 

, .: ex I. 

-,-ven l>v t 

•ly, 1797. ./. 

lecL oi I: 

. . . :-ing 'i',u:. 

apors which 


ing this, at 
'iced by tK 

Fig. 35.— The Spectre of the Brocken. 


spectre. He then called another person to him, and placing themselves 
in the very spot where the apparition was first seen, the pair kept their 
eyes fixed on the Achtermannshohe, but saw nothing. After a short 
interval, however, two colossal figures appeared, which repeated the 
gestures made by them, and then disappeared. 

Some few years ago, in the summer of 1862, a French artist, M. 
Stroobant, witnessed and carefully sketched this phenomenon, which is 
drawn in Fig. 35. He had slept at the inn of the Brocken, and rising 
at two in the morning, he repaired to the plateau upon the summit in 
the company of a guide. They reached the highest point just as the 
first glimmer of the rising sun enabled them to distinguish clearly ob- 
jects at a great distance. To use M. Stroobant's own words, "My 
guide, who had for some time appeared to be walking in search of some- 
thing, suddenly led me to an elevation whence I had the singular priv- 
ilege of contemplating for a few instants the magnificent effect of mi- 
rage, which is termed the Spectre of the Brocken. The appearance is 
most striking, A thick mist, which seemed to emerge from the clouds 
like an immense curtain, suddenly rose to the west of the mountain, a 
rainbow was formed, then certain indistinct shapes were delineated. 
First, the large tower of the inn was reproduced upon a gigantic scale; 
after that we saw our two selves in a more vague and less exact shape, 
and these shadows were in each instance surrounded by the colors of 
the rainbow, which served as a frame to this fairy picture. Some tour- 
ists who were staying at the inn had seen the sun rise from their vv'in- 
dows, but no one had witnessed the magnificent spectacle which had 
taken place on the other side of the mountain." 

Sometimes these spectres are surrounded by colored concentric arcs. 
Since the beginning of the present century, treatises on meteorology 
designate, under the name of the Ulloa circle^ the pale external arch 
which surrounds the phenomenon, and this same circle has sometimes 
been called the " white rainbow." But it is not formed at the same an- 
gular distance as the rainbow, and, although pale, it often envelops a 
series of interior colored arcs. 

Ulloa, being in company with six fellow - travelers upon the Pam- 
'bamarca at day-break one morning, observed that the summit of the 
mountain was entirely covered with thick clouds, and that the sun, when 
it rose, dissipated them, leaving only in their stead light vapors, which it 
was almost impossible to distinguish. Suddenly, in the opposite direc- 
tion to where the sun was rising, " each of the travelers beheld, at about 



seventy feet from where he was standing, his own image reflected in the 
air as in a mirror. The image was in the centre of three rainbows of 
different colors, and surrounded at a certain distance by a fourth bow 
with only one color. The inside color of each bow was carnation or 
red, the next shade was violet, the third yellow, the fourth straw color, 
the last green. All these bows were perpendicular to the horizon; 
they moved in the direction of, and followed, the image of the person 
whom they enveloped as with a glory." The most remarkable point 
was that, although the seven spectators were standing in a group, each 

Fig. 36.— The Ulloa circle. 

person only saw the phenomenon in regard to his own person, and was 
disposed to disbelieve that it was repeated in respect to his companions. 
The extent of the bows increased continually and in proportion to the 
height of the sun ; at the same time their colors faded away, the spec- 
tres became paler and more indistinct, and finally the phenomenon dis- 
appeared altogether. At the first appearance the shape of the bows was 
oval, but toward the end they became quite circular. The same appa- 
rition was observed in the polar regions by Scoresby, and described by 
him. He states that the phenomenon appears whenever there is mist 
and at the same time shining sun. In the polar seas, whenever a rather 


thick mist rises over the ocean, an observer, placed on the mast, sees 
one or several circles upon the mist. 

These circles are concentric, and their common centre is in the straight 
line joining the eye of the observer to the sun, and extended from the 
sun toward the mist. The number of circles varies from one to five; 
they are particularly numerous and well colored when the sun is very 
brilliant and the mist thick and low. On July 23, 1821, Scoresby saw 
four concentric circles around his head. The colors of the first and of 
the second were very well defined ; those of the third, only visible at 
intervals, were very faint, and the fourth only showed a slight greenish 

The meteorologist Kaemtz has often observed the same fact in the 
Alps. Whenever his shadow was projected upon a cloud, his head ap- 
peared surrounded by a luminous aureola. 

To what action of light is this phenomenon due? Bouguer is of 
opinion that it must be attributed to the passage of light through icy 
particles. Such, also, is the opinion of De Saussure, Scoresby, and oth- 
er meteorologists. 

In regard to the mountains, as we can not assure ourselves directly of 
the fact by entering into the clouds, we are reduced to conjecture. The 
aerostat traversing the clouds completely, and passing by the very point 
where the apparition is seen, aftbrds one an opportunity of ascertaining 
the state of the cloud. • This observation I have been able to make, and 
so to offer an explanation of the phenomenon.* 

As the balloon sails on, borne forward by the wind, its shadow trav- 
els either on the ground or on the clouds. This shadow is, as a rule, 
black, like all others; bat it frequently happens that it appears alone on 
the surface of the ground, and thus appears luminous. Examining this 
shadow by the aid of a telescope, I have noticed that it is often com- 
posed of a dark nucleus and a penumbra of the shape of an aureola. 
This aureola, frequently very large in proportion to the diameter of the 
central nucleus, eclipses it to the naked eye, so that the whole shadow 
appears like a nebulous circle projected in yellow upon the green 
ground of the woods and meadows. I have noticed, too, that this lu- 
minous shadow is generally all the more strongly maVked in proportion 
to the greater humidity of the surface of the ground. 

Seen upon the clouds, this shadow sometimes presents a curious as- 

* [The explanation of the phenomenon offered by M. Flammarion (viz., that it is due to dif- 
fraction) was generally recognized long previous to M. Flammarion's ascents. — Ed.] 


pect. I have often, when the balloon emerged from the clouds into the 
clear sky, suddenly perceived, at twenty or thirty yards' distance, a sec- 
ond balloon distinctly delineated, and apparently of a grayish color, 
against the white ground of the clouds. This phenomenon manifests 
itself at the moment when the sun re-appears. The smallest details of 
the car can be made out clearly, and our gestures are strikingly repro- 
duced by the shadow. 

On April 15, 1868, at about half- past three in the afternoon, we 
emerged from a stratum of clouds, when the shadow of the balloon was 
seen by us, surrounded by colored concentric circles, of which the car 
formed the centre. It was very plainly visible upon a yellowish white 
ground. A first circle of pale blue encompassed this ground and the 
car in a kind of ring. Around this ring was a second of a deeper yel- 
low, then a grayish red zone, and lastly, as the exterior circumference, 
a fourth circle, violet in hue, and imperceptibly toning down into the 
gray tint of the clouds. The slightest details were clearly discernible — 
net, ropes, and instruments. Every one of our gestures was instantane- 
ously reproduced by the aerial spectres. The anthelion remained upon 
the clouds sufficiently distinct, and for a sufficiently long time, to permit 
of my taking a sketch in my journal and studying the physical condi- 
tion of the clouds upon which it was produced.* I was able to deter- 
mine directly the circumstances of its production. Indeed, as this brill- 
iant phenomenon occurred in the midst of the very clouds which I was 
traversing, it was easy for me to ascertain that these clouds were not 
formed of frozen particles. The thermometer marked 2° above zero. 
The hygrometer marked a maximum of humidity experienced, namely, 
77 at 3770 feet, and the balloon was then at 4600 feet, where the hu- 
midity was only 73. It is therefore certain that this is a phenomenon 
of the diffraction of light simply produced by the vesicles of the mist. 

The name of diffraction is given to all the modifications which the lu- 
minous rays undergo when they come in contact with the surface of 
bodies. Light, under these circumstances, is subject to a sort of devia- 
tion, at the same time becoming decomposed, whence result those curi- 
ous appearances in the shadows of objects which were observed for the 
first time by Grimaldi and Newton. 

The most interesting phenomena of diffraction are those presetted by 

* A colored illustration of this remarkable phenomenon is given in the Voyages Aeriens, 
which was published by MM. Glaisher, De Fonvielle, and G. Tissandier, in conjunction with 
myself, part 2, p. 292. 


gratings^ as are technically denominated the systems of linear and very 
narrow openings situated parallel to one another and at very small in- 
tervals. A system of this kind may be realized by tracing with a dia- 
mond, for instance, on a pane of glass equidistant lines very close to- 
gether. As the light would be able to pass in the interstices between 
the strokes, whereas it would be stopped in the points corresponding to 
those where the glass was not smooth, there is, in reality, an effect pro- 
duced as if there were a series of openings very near to each other. A 
hundred strokes, about -^^ of an inch in length, may thus be drawn 
without difficulty. The light is then decomposed in spectra, each over- 
lapping the other. It is a phenomenon of this kind which is seen when 
we, look into the light with the eye half closed; the eyelashes, in this 
case, acting as a net-work or grating. These net-works may also be 
produced by reflection, and it is to this circumstance that are due the 
brilliant colors observed when a pencil of luminous rays is reflected on 
a metallic surface regularly striated. 

To the phenomena of gratings must be attributed, too, the colors, oft- 
en so brilliant, to be 'seen in mother-of-pearl. This substance is of a 
laminated structure; so much so, that in carving it the different folds 
are often cut in such a way as to form a regular net- work upon the sur- 
face. It is, again, to a phenomenon of this sort that are due the rain- 
bow hues seen in the feathers of certain birds, and sometimes in spiders' 
webs. The latter, although very fine, are not simple, for they are com- 
posed of a large number of pieces joined together by a viscous sub- 
stance, and thus constitute a kind of net-work. 

If the sun is near the horizon, and the shadow of the observer falls 
upon the grass, upon a field of corn, or other surface covered with dew, 
there is visible an aureola, the light of which is especially bright about 
the head, but which diminishes from below the middle of the body. 
This light is due to the reflection of light by the moist stubble and the 
drops of due. It is brighter about the head, because the blades that are 
near where the shadow of the head falls expose to it all that part of 
them which is lighted up, whereas those farther off expose not only the 
part which is lighted up, but other parts which are not, and this dimin- 
ishes the brightness in proportion as their distance from the head in- 
creases. The phenomenon is seen whenever there is simultaneously 
mist and sun. This fact is easily verified upon a mountain. As soon 
as the shadow of the mountaineer is projected upon a mist, his head 
gives rise to a shadow surrounded by a luminous aureola. 


The Illustrated London News of July 8, 1871, illustrates one of these 
apparitions, " The Fog Bow, seen from the Matterhorn," observed by E. 
Whymper in this celebrated region of the Alps. The observation was 
taken just after the catastrophe of July 14, 1865 ; and by a curious co- 
incidence, two immense white aerial crosses projected into the interior 
of the external arc. These two crosses were no doubt formed by the 
intersection of circles, the remaining parts of which were invisible. The 
apparition was of a grand and solemn character, further increased by 
the silence of the fathomless abyss into which the four ill-fated tourists 
had just been precipitated. 

Other optical appearances of an analogous kind are manifested under 
different conditions. Thus, for instance, if any one, turning his baclj to 
the sun, looks into water, he will perceive the shadow of his head, but 
always very much deformed. At the same time he will see starting 
from this shadow what seem to be luminous bodies, which dart their 
rays in all directions with inconceivable rapidity, and to a great dis- 
tance. These luminous appearances — these aureola rays — have, in ad- 
dition to the darting movement, a rapid rotatory Tnovement around the 

MALOIS. 137 



The description of optical phenomena now brings us to one of the 
most singular and complicated effects of the reflection of light in the at- 
mosphere. Under the name of halo {aXwg, area) is designated a brill- 
iant circle which, under certain atmospheric conditions, surrounds the 
sun at a distance of 22° or 46° ; while, under the name of parhelia, or 
mock suns {irapa, near, and iiXiog, sun), are designated luminous circu- 
lar spaces, generally of a red, yellow, or greenish color, which appear 
both to the right and to the left of the sun, at the same distance (viz., 
about 22°), bearing a sort of rough resemblance to the sun itself The 
same appearances may be seen about the moon ; and it is, indeed, easier 
to observe them, as the diminished brilliancy of the moon's light ren- 
ders an examination of the area around it less difficult. These lumi- 
nous spaces are called paraselenes {-rrapa, near, and (nXijvt], moon), or 
mock moons. The two cases only differ as to the intensity of the lu- 
minary from which they are derived — a difference similar to that which 
may be observed between ordinary solar and lunar rainbows. 

In addition to the halo and the two parhelia, a number of other cir- 
cles, arches, bands, or luminous spots, are sometimes seen upon the sky. 
These are more or less bright, and accompany the halo. 

It is well known that, when a triangular prism of glass is submitted 
to the action of the sun's rays, part of the light falling on it is reflected 
from the surface of the prism as upon a mirror, and another part pene- 
trates into the glass and leaves it in a direction different from that by 
which it entered, producing an image formed of different colors. It is 
upon this fact that Mariotte based the explanation of the phenomenon 
which we are about to consider. The origin of halos, in his opinion, is 
to be discovered in the crystals of ice in the shape of equilateral trian- 
gular prisms in the air. These prisms may be situated at all possible 
angles, and in all directions in the atmosphere, some among them being 
in such positions as to produce the absolute minimum of deviation of 
the rays of light which, entering by one of the three lateral surfaces of 


the prisms, traverse one of the other two on their way out of it. Mari- 
otte has shown that, at an angular distance from the sun equal to that 
of minimum deviation, which is 22°, a brilliant circle must be formed, 
and this is the ordinary halo. If from some cause or other all the 
prisms become vertical, the halo is replaced by two parhelia. The tan- 
gent arcs seen near the ordinary halo, the halo with a radius of 46° and 
the parhelion circle, have been explained by Young upon the hypothe- 
sis that, in certain cases, the prisms may be situated in such a way that 
their axes are all horizontal. 

Twenty years ago, Bravais devoted to the analysis of these phenom- 
ena a work which will be useful to us as a guide. The theory of these 
phenomena is somewhat complex, and demands a certain amount of 
attention in order to be intelligible. Voltaire confessed that he was 
obliged to read the same things twice over in order to comprehend 
them thoroughly ; and perhaps those of us who do not consider our- 
selves more acute than the sage of Ferney will do well to imitate him 
in this instance. 

When a halo appears upon the sky, light cirri clouds (of which we 
shall speak presently) are generally seen, and it is upon them that the 
phenomenon appears to be delineated. Sometimes, too, these cirri are 
collected into one single mass, so that the eye can not seize their shapes : 
a white vapor predominates in that part of the sky near to the sun ; and 
the blue tint of the atmosphere is replaced by a kind of light mist, the 
brilliancy of which is sometimes unbearable to the eye. But these light 
clouds of snow, placed high in the air, are so distant that it is difficult 
to decide upon their real nature. Hence we see how easily the mode in 
which the phenomenon is produced might for a long period have re- 
mained unknown ; and this is unquestionably one of the reasons why 
halos and parhelia were in early ages deemed marvelous phenomena, 
signs of celestial ire, presages of the death of princes, etc., etc. 

It is not enough for the clouds of the higher strata of the atmosphere 
to be formed of snowy particles for the phenomenon of the halo to be- 
come visible; the two following conditions are further necessary. The 
cloud must be of a certain degree of thickness ; for, if too thin, the halo 
would not occur; if too dense, the light would be intercepted. The 
crystallization of the water must also proceed slowly and not be disturb- 
ed by wind, as with a rapid, and therefore irregular, crystallization the 
points lose their transparency, the angles Qf the facets their consistency, 
and the surfaces by which the rays enter and leave, their smoothness. 

HALOS. 139 

The appearance of halos is less rare than might be supposed. It is cal- 
culated that in our latitudes the number of days on which this phenom- 
enon occurs, in the rudimentary state at least, are fifty a year, and in the 
north of Europe many more. 

The most simple form of crystals of ice, snow, or hoar-frost — viz., 
that seen in the earliest process of crystallization — is a right prism, hav- 
ing for its section a regular hexagon, and terminated by two bases per- 
pendicular to the lateral surfaces, which are rectangular. 

These simple forms are, however, rarely seen in a fall of snow, be- 
cause, before reaching the ground, lateral crystallization, due to the con- 
densation of vapor in the lower strata, makes an addition to the primi- 
tive nucleus. 

The hexagonal prism gives rise to all the spots or curves, the appear- 
ance of which has been placed beyond doubt by numerous observa- 

The halo, with all its aspects, is explained on the hypothesis of snow 
or ice-crystals falling slowly in a calm atmosphere. 

It is therefore due simply to the refraction of the solar rays upon 
crystals of ice. The different positions of the prisms of ice are the cause 
of the diversity of the appearances. The situation of these sharp-point- 
ed needles of ice in the atmosphere may be divided into three classes: 
1st, prisms placed at any angle; 2d, prisms axes of which are vertical; 
3d, prisms placed horizontally. 

In order to comprehend the production of the phenomena, let us, as 
in explaining the rainbow, take the first case and examine its effects. 
If a prism is turned round, the ray which emerges from it is seen to 
make a variable angle with that which enters it. But there is a certain 
position in which the entering and departing rays make the smallest 
angle possible with each other ; the prism then relative to the incident 
ray is said to be in its position of minimum deviation. ISTow, in this 
position, the prism may be turned a little one way or a little the other 
without causing any perceptible change in the direction of the refracted 

If a prism of this kind turns upon its own axis in the atmosphere, 
rays are incessantly emanating from it, which reach the eye and disap- 
pear immediately afterward; but, as has just been remarked, it is clear 
that the ray will catch the eye for the greatest length of time when its 
deviation is a minimum. If the number of these prisms is very great, 
we receive at the same time the rays refracted by a prism at the mo- 



ment at which the others disappear, so that the impression upon our 
eye is persistent, although the rays are not transmitted to it by the same 
crystals. A solar ray enters a triangular prism by the surface A (see 
Fig. 37), and undergoes a deviation. This ray is, of course, decom- 
posed. Let us suppose the violet portion, after emergence from the 
surface B, reaches the eye of the spectator placed at o. Another prism, 
c, nearer to the direction o s of the sun, will send red rays which have 
deviated the least, so that in fact the cone passing through a will be 
violet, the cone passing through c red, and the intermediate one colored 
with various intermediate colors of the spectrum. 


Fig. 37.— Thecji-y of Uie halo. 

Refraction of the solar rays will thus produce all round the sun, and 
at the same distance, a series of luminous impressions. The deviation 
is about 22°, but is not the same for all colors. Calculation, coinciding 
with observation, gives 21° 87' for the red, which is the least refrangi- 
ble color, 21° 48' for the yellow, 21° 57' for the green, 22° 10' for the 
blue, and 20° 40' for the violet. This circle of 22° radius which is thus 
formed around the sun and the moon is the ordinary halo which is seen 
most frequently. The red is inside ; then we have orange, yellow, green ; 
but these colors gradually become weaker, because they are influenced 
by the prisms, which are not in the position of minimum deviation. 

HALOS. 141 

The red remains most visible. The sun, however, is not, as we have 
assumed, a mere luminous point, but each part of its disk contributes 
to the production of this phenomenon; and this circumstance tends to 
blend still further the various colors, which are, in consequence, never 
very clearly defined, and the halo generally appears as a bright ring 
with a reddish tint on the inside, 2° to 8° in width, and inclosing a cir- 
cular area of which the sun occupies the centre. 

By a well-known optical effect, a spectator not previously instructed 
upon the point would be inclined to attribute an elliptic shape to the 
halo, considering it an oval with a longer vertical axis; but this illu- 
sion, which also takes place when an entire rainbow is seen, disappears 
before angular measurement. From a similar cause, the halo appears 
to get smaller as the sun rises, just as the moon loses, at a certain eleva- 
tion, the gigantic proportions that its disk presented soon after rising. 
In addition to the halo of 22° radius, a second is also frequently seen, 
the diameter of which is about twice as large as that of the preceding 

The latter is produced by the refraction of light across the dihedral 
angles of 90° that the sides of the prisms make with the bases, just as 
the angles of 60° produce the ordinary halo. Like the latter, it is com- 
posed of a succession of rings, the first of which (viz., the one nearest to 
the sun) is red. But, by a superposition of colors similar to that which 
occurs in the halo of 22°, there is scarcely discernible more than a ring, 
reddish upon its inside and yellowish in the middle, whereas the exter- 
nal part seems of a whitish hue, and gradually becomes lost in the gen- 
eral light of the atmosphere. The total width of this halo is rather 
large, embracing about the 3° between 45° and 48° distance from the 
sun, the white light that borders it included. 

These two circles are, therefore, formed by the reflection of light upon 
the prisms of ice placed at all angles in the air. Let us now consider 
what effects may be produced by prisms placed vertically. When the 
light is reflected across the dihedral angles of 60°, which the six sides 
of the prisms of ice falling vertically form between them, there are two 
parhelia produced, one to the right, the other to the left of the sun, and 
both situated at the same height as the latter. To rightly understand 
the reason of this phenomenon, the principle must first be enunciated 
that the light given by a group of prisms, all of which have their axes 
vertical, but which are situated in every conceivable position as to the 
direction of their sides, is similar to that which would be transmitted 



by a single prism turning rapidly on its own axis. It follows, in 
fact, that the prism, in the movement indicated above, passes in suc- 
cession through all the positions compatible with the verticality of its 

When the sun is on the horizon, the distance at which these appear- 
ances are formed is exactly the angle of minimum deviation, or, in oth- 
er words, the radius of the halo. If the halo and the parhelia are seen 
together, the latter appear to be situated just upon the circumference of 
the prism, and occupy in height a distance equal to the diameter of the 
sun. The various tints are clearer than in the halo ; the yellow is very 
distinct, and so is the green, but the blue is pale, and scarcely visible ; 
while the violet, overlapped by the other colors, is too indistinct to be 

Fig. 38. — Halo seen in Norway. 

seen. The phenomenon is completed by a tail of white light, some- 
times very indistinct, but occasionally attaining a length of from 10° to 
20° in the opposite direction to the sun, and parallel to the horizon. 
This light is due to those prisms, the positions of which are somewhat 
out of the line that corresponds to the minimum deviation. 

When the sun rises about the horizon, the luminous rays traverse the 
prisms, moving in oblique planes, and the smallest of the deviations 
produced during the rotation is greater than the absolute corresponding 
minimum, when the sun is at the horizon. This shows that the parhe- 
lia must emerge slowly from the circumference of the halo, in propor- 
tion as the latter rises in height; but on the other hand, as the halo is 
nearly 2° in width (including the white light that borders it), the par- 

HALOS. 148 

helia only become completely separated from it when the sun is at an 
elevation of 20° or 30°. 

Optical considerations show that the formation of parhelia becomes 
impossible when the sun has reached an elevation of 60°. 

Parhelia are sometimes very brilliant, and their brightness may then 
be in a certain measure compared to that of the sun itself, in which case 
it is quite conceivable that each parhelion may become in its turn the 
origin of two others, which are then the parhelia of parhelia, or seconda- 
ry parhelia. The effect caused by the refraction of light across angles 
of 90°, which produce the large halo, is still more remarkable. The so- 
lar rays enter obliquely at the upper base of the prisms, and, passing 
through it, emerge by one of the vertical surfaces. 

If we imagine, as we have already done in the case of the parhelia, 
that the prism on the upper base of which the rays are falling, turns 
rapidly upon its own axis, it may be proved by optics that the light 
emerging from it will be scattered in the form of a bright curve with its 
axis vertical, whence it is easy to conclude that the corresponding optic- 
al appearance upon the celestial sphere will be a luminous arc parallel 
to the horizon and situated at a great distance above the sun. 

The arc thus produced, which may be termed the upj^er tangent arc of 
the halo of 46°, or, more briefly, the circunizenithal ajx, deserves special 
notice, for it is unquestionably the most remarkable of all the appear- 
ances which may accompany the halo. The brightness of the tints, the 
distinctness of the colors, the precision with wiiich its edges, as well as 
its extreme limits, are shown upon the sky, give it the characteristic of 
a real rainbow. Of the respective rings composing it, the red is nearest 
to the sun, the violet fringes the concave part of the arc, and is on the 
opposite side ; the width of the various rings is about the same as in the 
rainbow, though rather less, owing to the illusion caused by the prox- 
imity of the zenith. When the halo of 46° is visible, the circumzenithal 
arc generally appears to touch it at its highest point, the red of the arc 
being then in contact with the red of the halo, the orange with the or- 
ange, and so on with the other colors; but very often the circumze- 
nithal arc is seen without the halo of 46°, just as the parhelia may ap- 
pear without the halo of 22°, although they owe their existence to the 
same kind of dihedral angles. 

From the observations that have been made of this arc, it appears 
that it never is to be seen when the height of the sun is less than 12° or 
more than 31°. 


It follows also from optical consideration that prisms, falling and 
turning upon their sides, can reflect the sun, forming upon the celestial 
sphere a luminous horizontal band, extending right round the horizon 
and passing through the exact centre of the sun. As reflection does 
not separate the colors which compose white light, this circle will ap- 
pear to be quite white, and its apparent width will be equal to the di- 
ameter of the sun. Such is the origin of the white circle called the ^ar- 
heliacal ring. It is upon its circumference that the ordinary parhelia 
always appear, as also the secondary parhelia situated at about 45° from 
the sun ; hence the name. 

Sometimes the solar rays experience two successive reflections upon 
the vertical surfaces of one of the prisms. There is then visible, at 
120° from the sun, a white image more or less diffuse, which has re- 
ceived the name of paranthelion. The horizontal bases of the ice-crys- 
tals reflect also the solar light, but in an upward direction, which pre- 
vents the spectator from perceiving it, unless he be upon the summit of 
a steep mountain, or in the car of a balloon, above the cloud containing 
the icy particles. It will be readily admitted that these conditions can 
be rarely fulfilled; but MM. Barral and Bixio were, fortunately, able to 
realize them on July 27th, 1850. The image of the sun thus reflected 
appeared almost as luminous as the sun itself Bravais suggested for 
this phenomenon, at once so remarkable and so rare, the name of pseu- 

Finally, the prisms of ice which are horizontal in the atmosphere give 
rise, by reflections and refractions analogous to the above, to tangent 
arcs which often appear on each side of the halo. 

The most complete halo that has yet been seen is that which Lowitz 
observed at St. Petersburg, on June 29, 1790, from 7 hours 30 minutes 
A.M. to 12 hours 80 minutes p.m. Since that time there have, of course, 
been a great number of halos observed ; but this is, perhaps, the most 
complete that has been recorded. MM. Bravais and Martins observed 
one at Piteo, in Sweden, on October 4, 1839, which was also very re- 
markable, but less complete than that seen by Lowitz. 

The examination which we have made of the general phenomenon 
of halos leads us to speak of other optical effects, the explanation of 
which is more or less akin to that of the above. 

The columns of white light, the crosses, and the different luminous 
aspects sometimes visible at sunrise and sunset, are due to the reflection 
of light upon the surfaces of crystals of ice situated high in the atmos- 

HALOS. 145 

phere. It is well known that if we look at the reflection of any light 
(such as the sun, the moon, or a street-lamp) on the surface of rather 
troubled water, the reflection extends vertically ; the motion of the wa- 
ter gives rise to a multitude of small surface-planes which are oscillating 
unceasingly about the horizontal, in all possible directions. This is the 
exact reproduction of what is going on in the region of the ice-cloud ; 
the small coruscating bases of the prisms, to which I have attributed 
above the reflection of the sun as seen from a balloon, are perpetually 
shifting their position. The reflection produced will therefore also be 
very elongated, and its upper part may, at sunrise or sunset, rise several 
degrees above the horizon. 

Such is the origin of those columns of white light which are some- 
times seen to form at the moment of sunset, and to increase in size as the 
sun gradually sinks lower. It is scarcely necessary to add that, when 
the sun has descended below the horizon, the reflection of the light 
takes place at the lower and not at the upper surfaces of the prisms. 

Previous to sunset, on April 22, 1847, four luminous columns, each 
about 15° in extent, were seen from Paris, presenting the appearance of 
a cross with the sun in the centre. After sunset one of these four col- 
umns (the uppermost of the four, of course) still remained visible for 
some little time. 

When the sun is near the horizon, part of a vertical circle may rise 
above that luminary in the shape of a column. On June 8, 1824, ap- 
pearances of this kind were seen in several parts of Germany. At 
Dohna, near Dresden, at eight in the evening, just as the sun was about 
to disappear behind the mountains, Lohrmann perceived a luminous 
band, perpendicular to the crepuscular arc, and similar to the tail of a 
comet. This column was 80° high and 1° in width. It is more unusual 
to see a band below the sun or the moon, and more unusual still to see 
also a horizontal arc passing the sun in such a way that it is situated in 
the middle of a cross. Roth saw very distinctly a phenomenon of this 
kind at Cassel, on January 2, 1586. Before the sun appeared, a lumi- 
nous vertical column, with a diameter equal to that of the sun, was visi- 
ble at the spot where the sun was about to rise, resembling a brilliant 
flame, except that its brightness was of uniform intensity throughout. 
Soon after there appeared a reflection of the sun, so brilliant that it was 
taken for the sun itself; and this parhelion had scarcely risen above the 
horizon when the sun rose immediately under it, followed by a column 
resembling that which had appeared above it. 



This latter, with its three suns, remained continuously vertical. The 
three suns were each exactly similar in appearance, but the true sun 
was the most brilliant. The phenomenon lasted about an hour. 

If the sun, instead of being on the horizon, is some few degrees above 
it, the luminous column which rises from the pseudohelion then situ- 
ated below the horizon, and consequently invisible, may reach to the 
centre of the sun, but can not extend perceptibly beyond it. We then 
have the appearance of a luminous ascending column, which seems to 
support the solar disk. Instances of this are afforded by the observa- 
tions taken by Parry at Melville Island on March 8, 1820 ; by Sturm 
on December 9, 1689 ; and by many others. 

The vertical gleams which, passing through the centre of the sun, ex- 
tend symmetrically above and below it, without having their base at 
the horizon, and which accompany the sun in his apparent course from 
east to west, seem due to the same cause. It is easy to see that they are 
caused by the rays twice reflected upon the horizontal bases of the ver- 
tical prisms, or, at all events, by some even number of successive reflec- 
tions. They are never seen but at heights less than 25° ; and are far 
more frequently seen about the moon than about the sun — a fact which 
is no doubt due to the greater brightness of the latter, which thus eclipses 
all the gleams near to it. The reverse is the case with the columns 
which are seen at sunset, because the sun then being below the horizon, 
the phenomenon is projected upon a partially lighted ground, and may 
thus be seen in all its brilliancy. 

The combination of the parheliacal circle with the vertical column 
passing through the centre of the sun, produces the phenomenon of the 
solar or lunar crosses which are often seen when the halo of 22° is not 
visible. Sometimes the arms of the cross may be nearly equal in length, 
and sometimes the horizontal are larger than the vertical. 

The vertical columns, and the lunar and solar crosses, are mostly seen 
in northern countries during the long winters which envelop those re- 
gions in snow and ice. 

To these optical effects must be added, finally, the coronas (see Fig. 
39) which appear around the sun and the moon when the air is not 
clear, and when small drops of vesicular vapor, or light clouds, are 
passing before their bodies. 

These colored rings, which are frequently seen round the moon, owe 
their origin not to refraction, but to diffraction ; they have the red out- 
side, and the violet inside, like the primary rainbow, and their colors 



are the converse of those of the two halos concentric with the sun and 
moon. The diameters of coronas of the same color are in the propor- 
tion of the natural numbers, 1, 2, 8, 4, etc., but the diameter of the first 
ring seems enlarged. This diameter, varying from 1° to 4°, depends 
upon that of the vesicles of water interposed between the luminary and 
the observer. Generally, the color of it is blue mixed with white for a 
certain distance round the luminary; then follows a red circle, and then 
other colored circles, as in Newton's rings. For the phenomenon to 

Fi>,'. By.— Coroua formed around the moon by diffraction. 

take place there must be a certain number of globules of the same char- 
acter, and, indeed, a far greater number of this diameter than of any 
other. If the diameters of the spherules of cloud were all different, the 
corona would not be produced. An exactly similar effect is observable 
when a luminous object is examined through a piece of glass that has 
been sprinkled with lycopodium powder, or, in a less marked degree, if 
the glass has merely been breathed upon before use. 

To these different effects, due to the refraction and reflection of light 
in the atmospheric strata, must. also be added the deformation of the 


sun at the horizon, which occasionally gives rise to most singular ap- 
pearances, in consequence of the want of homogeneousness in the lower 
strata, and the curious action of atmospheric refraction. 

With the progress of astronomy and physics, the decadence of astrol- 
ogy, and the expansion of inquiry, these optical phenomena lose their 
supernatural attributes. For the last century they have undergone a 
calm and impartial study and analysis ; while we see in this chapter that 
they may be explained upon theory, and savans merely recognize them 
as so many physical facts belonging to the vast domain of meteorology. 
The historian Josephus relates that at the beginning of the siege of Je- 
rusalem by the Komans, a.d. 70, the Jews foresaw their disaster "in ar- 
mies marching upon red clouds." Nearly analogous apparitions were 
visible at the commencement of the siege of Paris in September, 1870, 
to say nothing of the aurora borealis on the 24th of October ; but we 
now know that the phj^sical effects are purely natural, and are produced 
merely by the action of light in the atmosphere. 




THE inBAQE. 149 



Not only does the atmosphere produce remarkable phenomena in 
the aerial heights, but it gives play to its fancy even in the lower re- 
gions where we move, and the very surface of the ground and of the 
water is occasionally the field of strange metamorphoses due to the rays 
of light in the air. 

Under the name of mirage we designate those optical apparitions 
caused by a peculiar state of the densities of the atmospheric strata — a 
state which produces variations in the ordinary refractions which we 
considered in the previous chapter. 

In consequence of these variations distant objects appear either de- 
formed, transported to a certain distance, or inverted and reflected, ac- 
cording to the deviation which the abnormal density of the air causes 
in the luminous rays. 

The mirage is no new phenomenon. In Diodorus Siculus we read: 
"An extraordinary phenomenon occurs in Africa at certain periods, 
especially in calm weather; the air becomes filled with images of all 
sorts of animals, some motionless, others floating in the air: now they 
seem running away, now pursuing; they are all of enormous propor- 
tions, and this spectacle fills with terror and awe those who are not ac- 
customed to it. When these figures overtake the traveler whom they 
seem to be pursuing, they surround him with a cold and shivering feel- 
ing. Strangers not used to this extraordinary phenomenon are seized 
with fear; but the inhabitants, who are in the habit of seeing it, take 
no particular notice of it. 

" Certain physical philosophers attempt to explain the true causes of 
this phenomenon, which seems extraordinary and fabulous. They say 
that there is no wind, or scarcely any, in this country. The masses of 
condensed air produce in Libya what the clouds sometimes produce 
with us on rainy days, viz., images of all shapes rising on every side in 
the air. These strata of air, suspended by light breezes, become mixed 
with other strata, executing at the same time very rapid oscillatory 
movements; and when calm again sets in they descend toward the 


ground by their own weight, preserving the shapes that they had ac- 
cidentally assumed. If no cause occurs to disperse them, they spon- 
taneously attach themselves to the first animals which come near. 
Their movements do not appear to be the effect of volition, for it is im- 
possible for an inanimate being to progress or go backward. But it is 
the animated beings who, unwittingly, produce these voluntary move- 
ments, for, as they advance, they cause a violent recoil in the images 
which seem to fly before them. Similarly, those which recoil seem, by 
producing a void and a relaxation in the strata of the air, to be pursued 
by the aerial spectres. The persons running away are probably struck, 
when they stop or return to their former position, by the matter of these 
figures, which break against their bodies and produce, at the moment 
of the shock, the chilly sensation." 

We see that, before the epoch of Diodorus, the mirage had been ob- 
served; the philosophers of the period were nevertheless far from being 
in possession of the true scientific explanation, although it was then at- 
tributed to a change of density in the aerial strata. 

This same phenomenon (of which Quintus Curtius has also spoken) 
has long been remarked by the Arabs, and it is often discussed in the 
treatises of Oriental writers. Among other instances may be cited the 
Koran, which says that "the works of the incredulous are like the mi- 
rage {serah) of the plain ; the thirsty man takes it for water until he 
draws nigh to it, and then he discovers that it is nothing." 

In about the middle of the seventeenth century the mirage began to 
attract the special attention of physicists. The discovery of telescopes 
rendered possible a great number of observations, which were beyond 
the power of the naked eye; and the knowledge of the laws of the re- 
fraction of light, and of the variations in the density of the air caused 
by changes in its temperature, prepared the way for the theoretical ex- 
planation of these singular apparitions. It is not till 1783 that we find 
the first really scientific work treating of the mirage. This was from 
the pen of Professor Busch, who observed its effects on the Elbe, near 
Hamburg, and on the coasts of the Northern and Baltic seas. He often 
made use of a telescope, and this method of observation disclosed to him 
man^' details hitherto unknown. He saw upon several occasions a mir- 
ror of the waters and mock hank, beneath which figures upside down 
seemed to be delineated ; he saw ships suspended in the air, and bearing 
beneath their keels the reversed image of their masts and sails. On 
the 5th October, 1779, he saw, at the distance of two German miles from 


Bremen, the ordinary image of that town and a second image, very dis- 
tinct but upside down ; between him and the town there was a large 
and verdant common. The principal circumstances of the phenomenon 
are clearly indicated in his work, without, however, the theoretical ex- 
planation of them. 

It was during Bonaparte's expedition to Egypt that the true expla- 
nation of the phenomenon was first given. 

The soil of Lower Egypt forms a vast and perfectly horizontal plain, 
the uniformity of which is only broken by gentle eminences upon which 
are built the villages that are thus protected from the overflowings of 
the Nile. At morning and evening there is no change in the aspect of 
the country ; but when the sun has heated the surface of the soil, it 
seems, at a certain distance off, to be inundated; the villages look like 
islands in the middle of an immense lake, and below each village is to 
be seen its inverted reflection. To complete the illusion, the ground 
vanishes, and the vault of the firmament is apparently reflected in still 
water. It is easy to understand the cruel disappointment of the French 
army. Exhausted by fatigue, with a devouring thirst under the burn- 
ing sky, the men fancied they had reached a great pool of still water in 
which they saw reflected the shadow of the villages and the palm-trees ; 
but as they gradually approached, the limits of this seeming inundation 
retreated ; the imaginary lake, that appeared to surround the village, 
drew back, and finally melted away altogether, the same illusion being 
repeated in the case of the next village. The savans attached to the 
expedition who witnessed this phenomenon were not less surprised than 
the rest of the army ; but Monge succeeded in giving the explanation 
of it. 

The theory of the mirage, in order to be perfectly understood, de- 
mands very special attention. The phenomenon occurs when the lu- 
minous rays, through whose agency we see objects, are made (before 
they reach our eye) to undergo a deviation caused by differences of 
density in the strata of air they pass through. We have seen that 
when a luminous ray penetrates from a less dense into a more dense 
medium, it undergoes a deviation which bends it nearer to the line per- 
pendicular to the boundaries of the two surfaces; and when it passes 
from a more dense to a less dense medium, it suffers a deviation bend- 
ing it from the perpendicular. 

Further, the angle of refraction is greater than the angle of incidence, 
and at a given moment a certain ray will, after refraction, make an an- 



gle of 90° with the perpendicular to the surface. This is called the 
critical angle. 

Beyond this angle the rays are reflected, and do not enter the me- 
dium at all ; this is known in physics under the name of total reflection. 

An illustration of this fact may be obtained by filling a glass with 
water and holding it so as to see the surface of the liquid from under- 
neath; this surface acts like a mirror, and appears very bright. A 
spoon dipped into it is reflected. Another instance : a prism of glass 
properly placed at the opening of a dark room is capable of intercepting 
entirely the passage of light by this very fact of total reflection. In 
fact, when a luminous ray tends to emerge from a more reflecting me- 
dium into one that is less so, at an angle greater than the critical angle, 
the ray is entirely reflected. 

This being taken for granted, we may now afiirm that the mirage is 
a phenomenon of total reflection. 

By the action of the solar rays, when the atmosphere is calm, the 
strata of air which are in contact with the soil become very much heat- 
ed, and it may happen that for a short distance up their density may in- 
crease as they are farther from the ground. This is a purely accidental 
fact, which depends upon various circumstances peculiar to the place 
where it occurs; it does not extend very far, and consequently in nowise 
affects the general law of the decrease of density in proportion to the 

Fig. 40._Explanatlon of the ordinary mirage. 

elevation. In the event of these physical conditions happening, the fol- 
lowing may be the result: a luminous ray, starting from the point M 
(see Fig. 40), is successively refracted in a' d', as it is bent from the nor- 
mal ; at a given moment the direction will coincide with that of the 


stratum of air A, and this latter will serve as a mirror: the ray will fol- 
low, therefore, in an opposite direction a path, a d' a\ similar to that 
which it has already taken, and will reach the eye of the spectator, who 
will see in the lower direction, o M, the reflection of the palm-tree M, at 
the same time that he will see the object directly. It is, therefore, the 
stratum of air which, at a given moment, becomes a mirror, and con- 
sequently acts in the same way as a piece of reflecting water, which 
gives rise to the phenomenon. Such is the ordinary or inferior mirage. 

This lower and reflected deviation of the luminous rays does not al- 
ways attract attention so much as might be fancied. Many people will 
pass by it without remarking it, and, even when their attention is called 
to the fact, will declare that they perceive nothing extraordinary or 
worthy of notice. To clearly discern the mirage, a person must not 
only possess long and very accurate eyesight, but must also know how 
to observe details, and be accustomed to the view. To travelers, sailors, 
and meteorologists, this is a practice that has become familiar; but very 
frequently non-scientific eyes fail to distinguish these details. 

Yet, in some cases, and especially in certain regions of the globe, the 
mirage is so plainly evident that it arrests the most inattentive gaze. 
Such is at times the mirage upon the coasts of the Gulf of Messina; and 
such, it appears, is that seen upon the sandy plains of Arabia and 

The mirage is sometimes visible upon the surface of the sea, and of 
lakes and large streams ; sometimes upon the great dry and sandy 
plains, or upon high-roads or the sea-shore. 

Very frequently these misleading appearances, due to the action of 
solar rays, and to their prismatic reflection across the strata of air of 
unequal density, present purely imaginary shapes which one is inclined 
to consider as real, although their origin is as fortuitous as that of the 
appearances occasionally seen in the clouds. The same may be said of 
those unknown islands which rise up in mid-ocean before the astonished 
navigator, and which lead him astray toward imaginary lands. The 
Swedish sailors for a long time went in search of a magic island that 
seemed to rise between those of Aland and of Upland ; it turned out to 
be only a mirage. The towns which seem evolved by the wand of a 
fairy are sometimes but the reflection of distant towns; but more fre- 
quently there is nothing to explain, if not their nature, at least their 
origin. M. Grellois says, "During the summer of 1847, I was proceed- 
ing one very hot day on horseback, at a walking pace, between Ghelma 


and Bone, in company with a young friend wlio has since died. When 
we had arrived within about two leagues of Bone, toward one in the 
afternoon, we were suddenly brought to a halt at a turn in the road by 
the appearance of a marvelous picture unfolded before our eyes. To 
the east of Bone, upon a sandy stretch of ground which a few days be- 
fore we had seen arid and bare, there rose at this moment, upon a gently 
sloping hill running down to the sea, a vast and beautiful city, adorned 
with monuments, domes, and steeples. The illusion was so complete, 
that reason refused to admit that this was only a vision which held us 
entranced for nearly half an hour. Whence came this apparition? 
There was no resemblance to Bone, still less to La Calle or Ghelma, 
both distant twenty leagues at least. Are we to suppose it was the re- 
flected image of some large city on the Sicilian coast? That seems to 
me very improbable." 

The inferior mirage is sometimes affected by simple effects of refrac- 
tion, by a change or magnifying of the objects observed. Thus, for in- 
stance, in May of 1837, during the Algerian expedition, which preceded 
the treaty made with Abd-el-Kader, M. Bonnefont observed, among oth- 
ers, the curious mirage described below : 

"A flock of flamingoes, birds of prey which are very common in this 
province, were seen upon the south-east bank, about three miles and a 
half off. These birds, as they left the ground to fly to the surface of 
the lake, assumed such enormous dimensions as to give the idea of 
Arab horsemen defiling one after the other. The illusion was for a 
moment so complete, that General Bugeaud sent a Spahi forward as a 
scout. The latter crossed the lake in a straight line; but when he had 
reached a point where the undulations commenced, the horse's legs be- 
came so elongated, that both steed and rider seemed to be borne up by 
a fantastic horse several yards high, and disporting itself in the midst 
of the water that appeared to submerge it. All eyes were fixed on this 
curious phenomenon, until a thick cloud intercepting the sun's rays 
caused these optical illusions to disappear, and re-established objects in 
their natural shape. 

" Sometimes another effect, which became a source of amusement to 
the soldiers, was produced. If, while the sun was in the east and the 
wind blowing from an opposite direction, a small and buoyant object, 
susceptible of being floated along by the wind, was cast into the lake, 
it was curious to observe how it became larger as it got farther off, and, 
as soon as the wind had made it undulate, it suddenly took the shape of 


THE Mill AGE. 155 

a small boat, the movement of which, above the waves, was in propor- 
tion to the shaking it experienced from the wind. The objects that an- 
swered best for the experiment were thistle-heads, as they were most 
easily influenced, even by the lightest breeze, and rendered the illusion 
complete. At about half-past eight on the morning of June 18, with a 
temperature of 26° centigrade, while a somewhat strong breeze was 
blowing from the east, and a nebulous stratum was beginning to dissi- 
pate the heat, a certain number of these thistle-heads were launched 
upon the water, and no sooner had the wind driven them to the point 
where undulation commenced, than they presented the curious spectacle 
of a fleet in disorder. The vessels seemed to dash one against the other, 
and then, driven by the wind to a great distance, they disappeared as 
completely as if they had gone down." 

We now come to a second kind of mirage which is often seen, but the 
effects of which are less striking, and which has consequently been less 
studied, viz., the approach of objects situated beyond the horizon, and 
which are raised above it. In the ordinary mirage which we have just 
described, the density of the air increases with the height, the trajecto- 
ries being convex toward the earth, at least in their lower parts. In 
the case under consideration, the density decreases and the trajectories 
become very concave toward the ground. A luminous ray, at first hor- 
izontal, should, as it moves through the void, remain rectilinear; but 
the ordinary atmospheric refraction inflects this trajectory, imparting to 
it about a twelfth part of the terrestrial curvature. But if the condition 
of the strata is modified, and if, by the effect of an abnormal increase in 
the temperature, the density decreases with the height much more than 
is usual, the refracting effect of these strata may impart to these traject- 
ories a greater curvature, amounting to a quarter, a half, or even the 
whole of the curvature of a great circle of the earth. Indeed, some- 
times this action may cause it to exceed this latter limit. 

In these fresh conditions, the various trajectories passing through the 
eye and situated in the same vertical plane, instead of cutting each other 
two and two, as in the case of an ordinary mirage, generally diverge. 
Hence it results that we can not obtain two reflections of one object. 
If the depression of the apparent horizon is measured, it is found to be 
very much raised, sometimes to the level of the rational horizon; and 
objects, usually invisible by reason of their great distance and curva- 
ture of the earth, may become visible. The accidental position of these 
objects beyond the apparent contour of the visible horizon makes them 


appear to be much nearer than usual, wliile another circumstance fa- 
vors the illusion, viz., the transparency of the air during the occur- 
rence of the phenomenon. It is clear that, as no reversal of the ob- 
jects takes place, one would be less struck with this particular form of 
mirage than with that which corresponds to the cases previously de- 
scribed. Woltmann and Biot point out that when the atmosphere is 
in this particular condition the sea seems to be concave, at the same 
time the horizon is seen above the hulls of ships, distant shores take 
the shape of high cliffs, and very distant objects seem to rise in the air 
like clouds. 

An optical circumstance well worthy of attention is the following: at 
the same time that some objects are thus raised above others by which 
they are ordinarily hidden from view, or when they are apparently 
removed to this side of the apparent horizon, they seem to the eye to be 
very much nearer. Heim has described a case of this kind observed in 
the mountains of Thuringia, where he suddenly beheld three lofty peaks 
appear above an intermediate chain which generally concealed them 
from sight ; and these peaks appeared to be so clearly defined that he 
was able to distinguish, with an ordinary glass, tufts of grass that were 
distant four German miles. M. de Tessan saw a phenomenon of the 
same kind in the harbor of San Bias (California). 

A letter from Teneriffe, published in the Courrier des Sciences^ states 
that from the summit of this mountain, whence the view embraces a 
horizon of fifty leagues radius, a mirage rendered visible the Alleghany 
Mountains in North America, a thousand leagues distant. I scarcely 
can venture to credit this story. 

Having explained the two great classes of facts relating to the phe- 
nomena of mirages, one of which is due to the depression of the objects, 
and the other to their elevation, we now come to the consideration of 
another effect scarcely less curious, viz., the superior mirage. 

This presents three different aspects. Sometimes the reflection is seen 
inverted above the object, and, above the former, a second reflection, 
erect as the object; sometimes the first reflection alone is seen, the up- 
per one having disappeared ; and, thirdly, the upper reflection remains 
without any inverted reflection beneath it. 

Woltmann noticed the superior mirage on three different occasions ; 
objects appeared to be reflected in the sky ; in the air was seen the re- 
flection of the horizon of the waters, and below were suspended, upside 
down, the shores, houses, trees, hills, and windmills. Frequently a lay- 



er of air separated the objects turned upside down from those beneath, 
but usually the reflection and the object were in contact. 

Welterling made analogous observations upon the Svenska-Hogar, 
islands situated at the entrance of the harbor of Stockholm. He sajs: 
"Above each of the sand-banks a black spot rises and appears in the 
air; these spots then become elongated downward, and finally reach the 
sand-bank, which assumes the appearance of a column nine or ten times 
higher than it really is. Hence there results a mock horizon, to which 
all the objects are transported, all thus appearing in a straight line upon 
the same level, though their absolute height differs considerably." 

Crauz saw in Greenland the shores of the Kokernen Islands, raised in 
the shape of high cliifs, ancient towers, and ruined edifices, Brandes 
several times witnessed the superior mirage ; as a rule the reflection of 
objects were not seen very distinctly by him, for he adds that the upper 
or direct reflection was generally wanting, and he attributes this fact to 
the want of spherical shape in the homogeneous strata. He also re- 
marks that this is a very local phenomenon, being seen often upon the 
houses in the eastern part of Damgast, and at the same time being invis- 
ible upon those in the west part of the town. 

In December, 1869, between the hours of three and four in the morn- 
ing, a mirnge was seen in Paris, as represented in the opposite plate. 

These objects are occasionally delineated in the sky at a considerable 
height above the horizon. Some move very rapidly, and others are 
stationary, while they are sometimes tinged with colors. In proportion 
as the light augments, the shape becomes more airy, and they vanish 
entirely when the sun is shining with full brightness. Mirage may also 
be produced by two strata of air separated by a vertical plane. This 
notably occurs in the case of large walls with a southern aspect, when 
they are heated by the sun, and then the ordinary mirage is formed. 
It is in this case termed the lateral mirage. The wall in this instance 
acts in the same way as the soil when exposed to the solar rays, and a 
line perpendicular to the wall replaces the vertical line in the horizontal 
mirage. But, as the heated strata of air are easily renewed as they 
rise along the wall, the disturbing influence of the densities does not ex- 
tend very far. The eye must, therefore, be placed in front of the plane 
of the wall, and must view in a parallel direction any objects that may 
approach and recede. The persons who approach the doors in the wall, 
the images which cross in the sky the vertical parallel to that of the 
wall, are always seen inverted, as indicated in the theory of the ordinary 



mirao-e. Gruber seems to have been one of the earliest spectators of 
this phenomenon, Blackader has described a lateral mirage that he 
saw upon a wall at Leith. It was also observed by Gilbert. 

Let us add to the above the multiplied mirage which is seen when 
several reflections, all inverted, are superposed upon the object. Biot 
and Arago saw phenomena of this kind from the mountain Desserto de 
las Palmas, and observed at night, with the repeating circle, an illumi- 
nated reflector in the island of Ivyza. Besides the ordinary reflection, 
two, three, or even four false reflections, superposed in the same vertical 
line, have been seen. Scoresby observed, on July 18, 1822, a brig with 
three reflections superposed, all inverted, and in each of them the ves- 
sel was in contact with the reflection, also inverted, of the field of ice 
beyond which it was situated. 

The mirage does not always present such regular characteristics as 

we have indicated; sometimes 
the second reflection is seen 
above the original one ; some- 
times the two are seen beside 
each other; and, lastly, the re- 
flections sometimes are not in- 

Dr. Vince relates several re- 
markable observations. From 
Ramsgate, in fine weather, may 
be seen the tops of the four 
hio;hest towers of Dover Castle. 
The remainder of the edifice is concealed by a hill, which is about 
twelve miles from Ramsgate. On the 6th of August, 1866, Dr. Vince, 
looking toward Dover at seven in the evening, perceived, not only the 
four towers as usual, but the entire castle from roof to base, as dis- 
tinctly as if it had been transported to the hill near Ramsgate. 

In the polar regions, the action of refraction is seen under the most 
capricious and extraordinary conditions. Admiral Wrangell writes: 
" The extreme condensation of the air in winter, and the vapor diffused 
in the atmosphere in summer, give great power to refraction in the fro- 
zen sea. In these circumstances the mountains of ice often assume the 
most grotesque shapes ; sometimes, indeed, they seem to be detached 
from the icy surface which serves as their base, so as to appear to be 
suspended in the air." 

Fig. 42. — Lateral mirage seen on Lake Geneva. 


Very frequently Admiral Wrangell and his companions thought they 
perceived mountains of a bluish color, whose shapes were clearly de- 
fined, and between which they thought they could discern valleys and 
even rocks. But just as they were congratulating themselves on hav- 
ing discovered the long-sought land, the bluish mass, carried away by 
the wind, extended on each side, and finally embraced the whole hori- 
zon. Scoresby, who collected so much interesting information in these 
Greenland regions, has also pointed out that ice assumes at the horizon 
the most regular shapes, and even appears, at many points, suspended 
in the air. 

The most curious phenomenon was to see the reflection, inverted and 
very distinct, of a vessel below the horizon. He says : " We had al- 
ready observed similar apparitions, but this one was peculiar for the 
distinctness of the reflection, in spite of the great distance of the vessel. 
Its contour was so well defined that, in looking at it with a DoUand's 
glass, I could distinguish the details of the masts and the hull of the 
ship, which I recognized as that of my father. On comparing our 
books, we saw that we were 34 miles from each other, that is, 19^ miles 
from the horizon, and far beyond the limits of vision." 

Upon the shores of the Orinoco, Humboldt and Bonpland discovered 
that at noon the temperature of the sand was 127°, while at six yards 
above the ground the temperature of the air was only 104°. The hil- 
locks of San Juan and Ortez, the chain called the Galera, situated three 
or four leagues off, seemed suspended in the air; the palm-trees ap- 
peared to have no hold on the ground, and, in the midst of the savanna 
of Caraccas, these savans saw, at a distance of a mile and a half, a herd 
of oxen apparently in the air. They noticed no double reflection, Hum- 
boldt also remarked a herd of wild cattle, part of which seemed to be 
above the surface of the ground, while the remainder were standing 
upon the soil. 

Mirages are not exclusively phenomena of warm climates; as we 
have seen, they have been observed in the very heart of the polar seas. 

When, instead of occurring in plane and regular strata, refractions 
and reflections take place in the curved and irregular strata, a mirage is 
produced, the reflections of which are deformed in all directions, broken 
or repeated several times, and very far distant from one another. 

This is the case with the fantastic aerial vision, formerly attributed 
to a fairy — the Fata Morgana— yjhioh sometimes attracts crowds of peo- 
ple to the sea-shore at Naples and at Reggio, upon the Sicilian coast. 




The phenomenon generally occurs of a morning in very calm weather. 
For an extent of several leagues the sea upon the Sicilian coast assumes 
the appearance of a chain of sombre mountains, while the waters upon 
the Calabrian side remain quite unaffected. Above the latter is seen 
depicted a row of several thousands of pilasters, all of equal elevation, 
of equal distance apart, and of equal degrees of light and shade. In 
the twinkling of an eye these pilasters sometimes lose half their height, 
and appear to take the shape of arcades and vaults, like the Koman 

Fifr. 43.— L:i Fata MorL'ana. 

aqueducts. There is often, also, noticeable a long cornice upon their 
summits, and there are also seen countless castles, all exactly alike. 
These soon fade away, and give place to towers which in turn disap- 
pear, leaving nothing but a colonnade, then windows, and lastly pine- 
trees and cypresses, several times repeated. 

Similar fantastic apparitions were noticed with great surprise in the 
neighborhood of Edinburgh on the 16th and 17th of June, 1870, pre- 
vious to a severe thunder-storm. These are unquestionably among the 
most curious kinds of mirage that exist. 





None of my readers will have failed to have been struck with sur- 
prise, during the calm of a fine starry night, by the spectacle of a star 
gliding noiselessly through the celestial vault to extinction. Some, 
perhaps, of those who peruse these pages, may have enjoyed the rare 
privilege of beholding, not only a shootmg-star, but a more brilliant and 
sometimes very exciting phenomenon, viz., the rapid passage through 
space of a flaming bolide, scattering a gleaming light in all directions — 
a globe of fire, leaving a luminous track behind it, and sometimes burst- 
ing with an explosion like that of an enormous shell, and a report like 
that of a cannon. Some, perhaps, also, by a still more fortunate chance, 
have had an opportunity of picking up a fragment of an exploded bo- 
lide — a piece that has fallen from the sky — an aerolite or stone that has 
come down from the heights of the atmosphere. 

We here have three distinct facts, which nevertheless seem to be re- 
lated to each other in their origin. The progress made during the last 
few years in the special study of these meteors is a reason for consider- 
ing them separately, taking first the shooting-stars, then the bolides, and 
lastly the aerolites. 

The first point to consider in the study of shooting-stars is the meas- 
urement of the height at which they are seen. Two spectators, placed 
at a distance of some miles from each other, notice the passage of a 
shooting - star among the constellations ; its path is not exactly the 
same to both observers, owing to perspective. From the observation 
of these two paths the distance can be obtained. This method, as early 
as 1798, two German savans, Brandes and Benzemberg, had already 
made use of. From the latest researches upon this head made by Al- 
exander Herschel (grandson of the famous Sir William Jlerscnel), by 
Professor Newton, of New Haven, Conn., and by Father Secchi, Director 
of the Observatory at Rome, it has been concluded that the average 
height of a shooting-star is seventy-five miles when first seen, and fifty 
miles at the end of its visible journey. 


The velocity varies from seven to forty miles a second. 

Shooting-stars are not common to all nights of the year alike, for the 
resuH of observations shows that there are yearly, monthly, and daily 
periods of recurrence of certain sets of shooting-stars. Great showers 
of shooting-stars on particular nights have been remarked since the last 
century ; Brandes relates that, on December 6, 1798, during a carriage- 
drive to Bremen, he counted four hundred and eighty from the coach 
window; and he estimates that, at that rate, there must have been at 
least two thousand in the course of the night. 

During the night of the 11th to the 12th of November, 1799, Hum- 
boldt and Bonpland witnessed a perfect shower of shooting-stars at Cu- ' 
mana (South America). Bonpland states that there was no part of the 
sky equal in extent to three diameters of the moon that was not con- 
tinuously being filled with shooting-stars. The inhabitants of Cumana 
were terrified by this phenomenon, and the oldest of them remembered 
an analogous occurrence in 1766, accompanied by an earthquake. 

This shower of stars at the close of the last century had been nearly 
forgotten, when a fresh shower was seen in America on November 13, 
1833. Professor Olmsted, of New Haven, Conn., basing his calculations 
upon data which had been transmitted to him, regards the number of 
shooting-stars that appeared in certain districts on that occasion as over 
two hundred thousand. Olmsted was the first to point out that the 
great display of November must be periodical, and would be reproduced 
every year at the same epoch. A very considerable increase in the 
number of shooting-stars at that date has, in fact, been noticed, but not 
to the extent of the extraordinary phenomenon in America in 1833. 
The astronomer Olbers, writing on the same subject in 1837, says: "We 
shall, perhaps, have to wait until 1867 for the recurrence of the splendid 
phenomenon witnessed in 1799 and 1833." This bold prediction was 
completely realized just a twelvemonth earlier, in 1866. 

From a general discussion of the observations, it results that the num- 
ber of shooting-stars which ordinarily appear over the whole extent of 
the visible sky in the space of an hour is, on an average, from ten to 

Now, at the time of the maximum on November 12 and 13, this 
hourly number, which was equal to fifty in 1834, gradually fell annual- 
ly, until it was reduced to thirty in 1839, to twenty in 1844, to seven- 
teen in 1849 ; three or four years later the maximum had disappeared, 
and was replaced by a normal appearance of from ten to eleven an hour 



Matters remained in this condition until 1863, when a maximum of 
thirtj-seven an hour again occurred at the same epoch, rising to seventy- 
four an hour the next year, and thus acting as a precursor of the great 
phenomenon of 1866, when Olbers's prediction was fulfilled. Another 

Fig. 44.— Shooting-stars. 

maximum occurred on August 10, and was noticed by M. Quetelet so 
long ago as 1837. The maximum hourly number of shooting-stars was, 
on that night, fifty -nine. There was a progressive rise in the number 
to seventy-nine in 1841, to eighty-five in 1845, and to one hundred and 
ten in 1848, from which date it gradually decreased each year, standing 
at thirty-eight in 1859, since which time it has alternately risen and 
fallen, varying between the numbers thirty-seven and sixty-seven. 

Here we have a well-ascertained annual variation in these periodical 
showers. The researches of Coulvier-Gravier clearly establish the ex- 
istence of a montldy variation, the number of shooting-stars being great- 
er in autumn than in spring. There is, also, a daily variation. The 
hourly numbers, from six in the evening to six in the morning, are 
twice as great as for the corresponding hours in the day-time. 

Shooting-stars are seen in all parts of the sky ; but if the directions 
whence they seem to come are examined, it is found that the different 


parts of the horizon furnish different numbers. There is thus a varia- 
tion in this respect which is termed the azimuihal variation^ and which 
has been thoroughly studied by means of carefully registered observa- 
tions. Many more shooting-stars come from the east than from the 
west, but nearly equal numbers from the north and the south. 

At the periods of the maxima, toward the 12th and 13th of Novem- 
ber, and toward the 9th and 10th of August, the shooting-stars, instead 
of appearing in all the regions of space indifferently, nearly all come 
from given directions. Some (those of November) start from the con- 
stellation Leo; the others (August) emanate from the constellation Per- 
seus. What path in space is then taken by these periodical showers, 
the existence of which is ascertained ? 

It has been observed that the speed of the meteors is equal to that of 
comets descending toward the earth from the depths of space, and their 
orbit has been also assimilated to the orbits of the comets. Signor Schi- 
aparelli. Director of the Milan Observatory, sought to determine the ele- 
ments which characterize the shape and the position of the apparent 
parabola followed by the meteoric current of the 10th of August. He 
then compared these astronomical elements with those obtained b}^ cal- 
culating the orbits of the different comets. He was thus able to estab- 
lish a very unexpected similarity between the orbit that he had just dis- 
covered for the swarm of shooting-stars of the 10th of August and that 
of the great comet seen in 1862. 

Supposing that every one hundred and eight years these meteors 
have a frequency neither so sudden nor so short in duration as that of 
November meteors, but lasting twenty or thirty years, this period agrees 
with the duration of the revolution of the great comet of 1862, and may 
be, therefore, taken to represent that of the successive returns of the 
comet to its perihelion. 

M. Schiaparelli then set to work to discover the elements of the orbit 
of the November swarm of shooting-stars. Observation in this instance 
supplied him with further data ; the period of return for the great dis- 
plays of November, indicated by Olbers in 1837, had just been confirmed 
in 1866, and might be fixed at thirty-three years and a fraction.* 

* [Taking as data the observed directions, etc., of the November meteors, the researches of 
Professors Newton (New Haven, Conn.) and Adams have shown that their orbit must be an 
ellipse, the periodic time of which is about 33^ years, agreeing exactly with observation. A 
small discrepancy has also been satisfactorily explained as the result of the attraction of the 
larger planets, especially Jupiter. — Ed.] 


A swarm of shooting-stars, seen on the 10th of December, describes 
in space the same ellipse as the well-known Biela's comet, and the shoot- 
ing-stars seen on the 20th of April move along the orbit of the first 
comet of 1861. Such researches have thrown a great light upon the 
question of shooting-stars. The comet which traces in space the same 
path as the swarm of meteors must be considered as an integral part of 
it. It is, in fact, merely a local concentration of the matter of the swarm 
— a concentration so intense that the mass of matter it forms is visi- 
ble even at a great distance from the earth. According to this theory, 
shooting-stars are of the same nature as comets, consisting of small neb- 
ulous objects which move in space without being visible to us because 
of their smallness, and only becoming so when they penetrate into the 
atmosphere of the earth. Like comets, they seem to be gaseous. 

A current of these meteors which encounters the orbit of the earth 
at a certain point, and the different parts of which take several years to 
pass this point of meeting, must be crossed by the earth each year at 
the same epoch. Hence the periodical showers of shooting-stars which 
are reproduced from year to year, with varying intensity, according to 
the greater or less concentration of the nebulous matter in the various 
parts of the current which the earth successively reaches. 

Such are shooting-stars. Now we come to the Bolides. If shooting- 
stars are gaseous, there is an essential distinction between them and 
bolides, for the great majority of the latter are unquestionably solid. 
To give an idea of the meteoric phenomenon of the explosion of a bo- 
lide, I will cite, among the most recent falls, one that occurred by day 
and another that occurred at night, both in 1868. 

This is the account of the fall of a bolide by day, which took place in 
the arrondissement of Casale, in Piedmont, on the 29th of February.. 
It was half-past ten in the morning, but the sky was rather dark. Sud- 
denly a loud detonation was heard, similar to the discharge of a heavy 
piece of artillery, or, perhaps, rather to the explosion of a mine. This 
was followed, at an interval of two seconds, by another report resulting 
from two distinct detonations, which succeeded each other so closely 
that the second seemed to be the continuation or the prolongation of the 
first. These detonations were heard as far off as Alexandrie, a distance 
of twenty miles. The sound had not yet died away when there became 
visible, at a considerable height above the ground, a mass irregular in 
shape and enveloped in smoke, thus resembling a small cloud. It left 
behind a long train of smoke ; other spectators saw distinctly, and at a 


great height, not one but several spots like small clouds which disap- 
peared nearly instantaneously. Some men at work in the fields saw 
several blocks fall through the air, and heard the noise which they made 
as they struck the ground. Every one whom it was possible to ques- 
tion on the subject was unanimous in affirming that there were a large 
number of these blocks, and that they must have occasioned a regular 
shower of aerolites of all sizes. Laborers at work felling trees in a 
wood three-quarters of a mile from Villeneuve, on the high-road from 
Casale to Vercelli, saw something like a hailstorm of grains of sand after 
these detonations, and a somewhat large fragment struck the hat that 
one of them was wearing. The aerolites found upon the ground con- 
sisted of: 1st, a piece weighing 4^ lbs., which fell in a wheat-field 650 
yards to the south-east of Villeneuve, and penetrated sixteen inches 
into the ground ; 2d, a piece weighing 14| lbs., which fell in a newly- 
sown field to the north of Villeneuve, 7700 feet from the first, and en- 
tered the ground to a depth of 14^ inches ; 3d, the numerous frag- 
ments into which a third piece broke by falling upon the pavement in 
front of the inn of Molta dei Conti, at a distance of 10,335 feet from 
the first piece, and of 10,630 feet from the second. 

The recital of the nocturnal fall will help to complete the comprehen- 
sion of these singular occurrences. It took place in the arrondissement 
of Mauleon, in the Lower Pyrenees, on September 7, 1868, at half-past 
ten in the morning. 

The sky was suddenly illuminated by a meteor, which looked like a 
burning ball with a long train of fire in its track. It emitted a bright 
light of a pale greenish hue, and lasted for six or ten seconds. Its dis- 
appearance was preceded by an explosion, and by the simultaneous pro- 
jection of flaming fragments, while there remained for some time after 
a light and whitish cloud. This was followed by a continuous noise, 
like the distant rolling of thunder, then by three or four detonations of 
extreme violence, which were heard at points distant fifty miles from 
each other. Immediately after these detonations the inhabitants of 
Sanguis-Saint-Etienne heard a hissing noise like that made by red-hot 
iron when it is plunged into water, then a dull sound indicating the fall 
of a solid body to the ground. The mass had fallen at about thirty 
yards from the church of Sanguis, in the bed of a small stream, and was 
shattered into fragments, the largest of which was scarcely two inches 
long. The fall was witnessed by two men who were talking together, 
and who, terrified at the detonations and the hissing noise, had thrown 


themselves upon the ground just as the stone fell about twenty paces 
before them. The weight of the stone was estimated at from six to 
eight pounds. 

These two instances, which I select from an immense number, give 
a sufficient idea of these downfalls from the sky, which were formerly 
looked upon as fabulous. It is only in the last half century that the 
facts have been credited and scientifically confirmed. 

In contradistinction to the shooting-stars which become extinguished 
and lost in the upper regions, the bolides traverse all the atmospheric 
strata, and often reach the surface of the earth. This is the reason why 
the luminous phenomena that accompany them usually appear to us 
much more intense ; because, in fact, the regions in which they occur 
are much nearer to us. But when seen from afar, as is the case with 
those whose directions prevent them from reaching the lower strata of 
the atmosphere, bolides present the same appearance to our eyes as 
shooting-stars. When they do reach the lower air, an explosion, sim- 
ple or repeated, often takes place, followed in the majority of cases by 
a fall of fragments from the bolide that have become detached from 
the main mass by the effect of the explosion. Bolides, then, are solid 
bodies, like the fragments detached from them. The orbits described 
by these bolides, in their movement relative to the earth, have some- 
times been found to be ellipses of such limited dimensions, that one 
would be led to suppose that the former were nothing but satellites 
of the earth, only visible during their passage through the atmos- 
phere — a view adopted by Petit, of Toulouse. On the other hand, 
their orbits have sometimes been found to be hyperbolic arcs, nearly 
rectilinear, and traversed with great speed — a fixct tending to show 
that bolides possessing such rapid movement must come from the 
stellar regions. 

The aerolites are minerals that fall from the sky to the earth. They 
proceed from the explosion of a bolide. 

Sometimes they plunge deeply into the soil upon which they fall. 
Thus the island of Lanaia-Uawai possesses an aerolite six or seven 
yards in diameter, which has remained imbedded in the ground in de- 
spite of all the efforts made to raise it to the surface. This aerolite fell 
at the beginning of the century. (Very recently, on the 9th of March, 
1868, at 9-30 p.m., another bolide fell upon the same island.) 

These stones, if touched immediately after their fall, seem to be burn- 
ing hot ; but they cool very rapidly — a fact indicating that their higher 



temperature was altogether superficial, and did not extend to the inte- 
rior of their mass. 

As to the shape of these aerolites, it is neither that of a ball, more or 
less round, nor that of an object with a rounded surface; they rather 
resemble polyhedra, with rough, irregular sides and ridges. The plane 
parts of their surface have often hollows analogous to those produced 
by the pressure of a round body upon a pasty substance. They are, 
moreover, enveloped in a black crust, generally of a dark hue, but some- 
i>imes lustrous, as if covered with very thin varnish. 

Fig. 45.— Fall of a bolide in the day-time. 

The light displayed in the movements of the bolides is due entirely 
to the heat produced by the compression of the air. Let us examine 
in what way the phenomena of explosion, and the falls of the aerolites 
which often succeed it, are produced. 

The enormous compression of the air forced back by the bolide can 
not occur without this air reacting upon the anterior part of the surface 
of this body, and exercising a considerable pressure upon it. Attrib- 
uting to the bolide a speed of four and a half miles per second — by no 
means an exaggerated estimate — M. Haidinger calculates the resisting 


pressure which the bolide meets with from the air at more than twenty- 
two atmospheres. Such a pressure evidently tends to crush the body 
which is exposed to it ; and if this body, in its more or less irregular 
shape and constitution, offers portions of itself which are more opposed 
than the others to the action of this pressure, these portions may give 
way and become suddenly detached from the mass of the bolide. 

Broken off" and started in a direction contrary to that in which they 
were traveling a few moments before with the main mass of bolide, 
these fragments soon lose the speed with which they were endowed, and 
reach the terrestrial surface, still moving with very great velocity, but 
not with the rapidity of bodies falling to the earth from space. 

We are inclined to look upon the bolides as being somewhat similar 
in origin and being to the planets which circulate in such great num- 
bers around the sun, and as probably themselves forming part of our 
planetary system. Besides, the discovery recently made of a large 
number of planets of very small dimensions, induces us to believe th'at 
there exists a multitude of others still smaller which have escaped ob- 

In consequence of the great difficulties that were encountered in at- 
tributing to the bolides a purely terrestrial origin, it was long ago sug- 
gested that they might be stones hurled to the earth from the volcanoes 
of the moon. This idea was taken up and developed, in 1795, by Olbers, 
and in the early part of the present century by Laplace, Lagrange, Pois- 
son, and Biot; but serious objections of more than one kind soon ap- 
peared to render this theory untenable, and it was finally abandoned for 
that of Chladni, whose system consisted in regarding the bolides as 
bodies wandering freely in space, and penetrating every now and then 
the atmosphere of the earth. 

Whatever may be the part played by the bolides in the universe, the 
possibility afforded us of examining the fragments which they leave in 
their passage is very useful in regard to the information which we are 
enabled to extract from them as to the constitution and nature of bodies 
foreign to the globe which we inhabit. Thus great pains have been 
taken of late years to collect from all quarters stones that have fallen 
from the sky after the explosion of bolides ; and collections of this spe- 
cial kind of rock have been made, to which, in order to distinguish 
them from the terrestrial rocks, the special denomination of meteorites 
has been given. There are at various places beautiful and valuable 
collections of this kind ; among others, that in the Museum of Natural 



History in Paris, that in the British Museum, and that in the Mineral- 
ogical Museum at Vienna. The Paris collection, under the superintend- 
ence of M. Daubree, contains at present specimens of 240 meteorites, 
while all the known falls do not exceed 255. 

It is easy to understand that conflagrations may have been caused by 
the fall of aerolites, and that people may have been killed by them. 
Fourteen deaths have been ascertained to have taken place from this 
cause at various times. 

The largest stones known to have fallen are as follows : 
The aerolite that fell at Juvenas in the Ardeche, on June 15, 1821, 
weighed 212 lbs., exclusive of the fragments detached from it. 

Fig. 46.— The Caille aerolite, weighing Vi\ cwt. 

The aerolite found in Chili, between Eio-Juncal and Padernal, in the 
Upper Cordilleras of Atacama, weighed 240 lbs., and was in the shape 
of a cone, measuring nineteen inches in length and eight inches in di- 
ameter. The miners who brought it home upon their mules had taken 
it for a block of silver. It was in the Paris Exhibition of 1867. 

The meteoric stone of Murcia, which is in the Museum of Natural 
Sciences at Madrid, weighs 1\ cwt. 

The aerolite which fell in 1492, at Ensisheim, in the Upper Rhine, in 
the presence of Maximilian I., king of the Romans, weighs 2f cwt. ; it 


is imbedded five feet in the ground, and was long venerated by the 
Church as a miraculous object. 

The aerolite that fell on Christmas-day, 1869, at Mourzouk (latitude 
26° N., longitude 12° E. of Paris), in the midst of a group of terrifiea 
Arabs, must weigh much more, for it is nearly a yard in diameter, it 
is to be taken to Constantinople, but will, unfortunately, have to be pre- 
viously divided. 

None of these, however, approach the Caille aerolite, in the Maritime 
Alps, which was used as a seat at a church porch, and which is now in 
the Paris Museum. It weighs 12J cwt. (see Fig. 46). 

The aerolite that fell in 1810 at Santa-Kosa (New Granada) in the 
night of April 20, 21, weighs 14f cwt. When found, it was almost im- 
bedded in the ground by the force of the fall. 

Lastly, the most colossal of the known stones that have fallen from 
the sky is the aerolite brought back from the Mexico campaign, weigh- 
ing more than 15^ cwt. It had from time immemorial been lying at 
Charcas. Its shape is that of a truncated triangular pyramid, measuring 
a yard in height, and it is a fair specimen of the world that sent it to us. 
From several hundred analyses made by the most eminent chemists, it 
appears that the meteorites have added no single substance to the globe 
which it did not possess before. The elements up to this time discover- 
ed to be existent in them are twenty-two in number. 




To complete the panorama of the optical phenomena of the sky, we 
will now consider the nature of a nocturnal brightness which is seen in 
the heights of the atmosphere on certain clear nights. As in the case 
of shooting-stars and bolides, its origin is in the depth of space, and the 
explanation of it belongs to astronomy ; but, as it reveals itself in our 
sky, it deserves notice in these pages. 

After sunset in January, February, March, and April, and after sun- 
rise in November, the celestial vault sometimes displays a band of light 
inclined toward the horizon and in the plane of the zodiac; that is, in 
the apparent path that, by its annual change of position, the sun seems 
to trace out in the sky. This light was not remarked till comparatively 
recently, and the discovery of it is due to Childrey, who speaks of it in 
his " Natural History of England," published about 1659. The earliest 
scientific researches with regard to this phenomenon were not, howev- 
er, made until 1683 ; they are due to J. D. Cassini. When the zodiacal 
light first appears in the evening after sunset, it is interfered with 
near the horizon by the last traces of the twilight glimmer, and the 
union of these two lights presents the appearance of a cone. This 
oblique cone, at least in our climates, has its base upon the horizon and 
its summit at a certain height above. 

Toward the equator this brightness rapidly loses its conical aspect as 
the last traces of twilight disappear, and when night has fully set in a 
band of light may be distinguished right round the sky, and making 
the zodiac luminous, so to speak ; sometimes this band is visible unin- 
terruptedly from sunset to sunrise. The parts nearest to the sun exceed 
in brilliancy the intensity of the Milky Way ; the other parts are dim, 
and if they are visible at all in the intertropical zone, it is because of 
the great limpidity of the atmosphere in these regions. 

The zodiacal light, when it is distinctly seen, as in the intertropical 
zone, is one of the most beautiful of the celestial phenomena. Its color 
is pure white. Certain observers in Europe have sometimes thought 
that they could discern a reddish tint in it. This tint has no real exist- 



ence ; for, if it had, it would be most distinctly discerned at the tropics, 
as the color would become more perceptible when the intensity of the 
light was increased. The last traces of twilight have been mistaken for 
it. In the tropics (in the months of January and February for the 
Tropic of Cancer) it rises perpendicularly to the horizon ; then, when 
night has fully set in, there is seen rising in the west a beautiful white 
vertical column, the central axis of which equals and even exceeds in 
intensity the more brilliant parts of the Milky Way. Upon the edges 
of this column, the light gradually blends with the feeble glimmer of 
the sky. It differs in that respect from the Milky Way, the edges of 
which at certain points offer a noticeable contrast of light to the general 
darkness, as in the black hollow of the Southern Cross, called the coal- 

It is not visible in Europe during the summer. This is owing to its 
inclined position upon the southern horizon, which then grazes the part 
of the zodiac which is visible at night and during the twilights. In 
February its appearance is most complete. In warm countries, the 
shortness of twilights, and the elevation of the ecliptic, cause the phe- 
nomenon to be visible all the year round. There are, however, even in 
countries where this is the case, periodical maxima of beauty which de- 
pend upon the inclination of the plane of the zodiac to the horizon. 

The observations of Cassini and of Mairan, who sometimes saw the 
zodiacal light at more than 100° from the sun, had long since indicated 
that this beautiful phenomenon extends beyond the terrestrial orbit. 
Humboldt and Brorsen had also remarked a luminous thread uniting 
the east and west. 

Let us now consider what is the nature of this nebulosity which sur- 
rounds the sun. Several astronomers of the last century thought it was 
the atmosphere of that luminary, extending to an immense distance in 
the direction of its equator. From mathematical considerations, La- 
place has shown that this hypothesis is inadmissible, and that the solar 
atmosphere can not extend beyond the limit at which the centrifugal 
force due to rotation would be in equilibrium with the attraction of the 
sun. It can easily be shown that at a distance from the sun equal to 
thirty-six times its semi-diameter, the centrifugal force developed by its 
rotation equals the weight of the atmospheric particles at that distance. 
It is mathematically impossible that the solar atmosphere can extend 
beyond this limit. It is not half the distance from Mercury to the sun, 
and but a sixth part of the distance at which the earth gravitates, for we 


are situated at a distance of two hundred and fourteen times the semi- 
diameter of this gigantic luminary from its centre. Therefore the zodi- 
acal light, which extends beyond the terrestrial orbit, is not an atmos- 
phere of the sun. 

Physicists have ascertained that all reflected lights acquire the prop- 
erties peculiar to polarization, but that at the same time these proper- 
ties may be lost in the event of the reflection arising, not from a gas or 
a continuous surface, but from a series of distinct particles, as in the 
clouds, which are composed of globules of water. The zodiacal light 
not being polarized, it results either that this light is not reflected, and 
issues directly from matter luminous in itself, or, if it proceeds from the 
sun, that it is caused by the reflection of the light of that luminary from 
a multitude of corpuscles having no connection with each other, but obe- 
dient, like all matter, to the laws of universal gravitation. These bodies 
we must regard as circulating round the sun, and describing elliptical 
orbits like the planets or the comets. Now, if the zodiacal light pro- 
ceeded from matter luminous in itself, this substance would still reflect 
a certain quantity of the solar light, so that traces of polarization in the 
zodiacal light would be perceived if it was not composed of distinct 
corpuscles. Therefore, in any case, we may consider as proved that it 
is due to corpuscles with no connection between each other, and cir- 
culating in accordance to the laws of gravitation round the sun, from 
which they receive their light. Judging by the trifling intensity of 
the light which they shed, it is improbable that they further possess a 
proper light of their own. 

It is possible that the aerolites, to the number of milliards upon mil- 
liards, distributed throughout the whole planetary system, and chiefly 
in the general plane of movement — that is, in the plane of the ecliptic 
— the bolides, the shooting-stars, corpuscles, solid, liquid, and gaseous, 
form but one general kind of celestial fragmentary bodies, and that the 
zone in which they chiefly gravitate is manifested to us by the reflec- 
tion of the solar light, and constitutes the zodiacal light; and that, by 
falling against the sun, these corpuscles cause the spots on its disk, and 
help to keep up its immense heat. If this whirlwind of corpuscles 
does not circulate around the sun itself — a fact not proved — it circu- 
lates around the earth; and it is just possible that from afar it may 
look like the ring of Saturn. 

The appearance of the zodiacal light is somewhat rare in France ; it 
is scarcely ever seen distinctly more than once or twice a year, and then 


in February, It was seen in Paris very clearly on the 20th of Febru- 
ary, 1871, and lasted from 6'50 to 7-30. In the shape of a spindle, in 
which it is always seen, it measured 18° in width at its base, at the ho- 
rizon, and, rising obliquely along the zodiac, terminated in a point be- 
fore reaching the Pleiades. From the sun, which had set an hour and 
a half earlier, to the extremity of the spindle, it measured 86°; the part 
which was visible above the horizon measured 63°. 

The determination of its intensity was all the more easy, as the at- 
mosphere of Paris was scarcely lighted up at all, in consequence of 
there being no gas. Calm and motionless, this light was very different 
from the quivering gleam of the aurora borealis. This spindle was 
much more intense in the middle than at the edges, and at its base than 
at its apex. The tint, about half as brilliant again as the Milky Way, 
was rather more yellow. The smallest stars were visible through this 
veil ; while in the case of the aurora borealis in October, 1870, the brill- 
iancy of the stars in Ursa Major was eclipsed. 




HEAT. 181 


heat: the thermometer — quantity of heat received — TEMPER- 

We have, in the First Book, contemplated the earth as it is borne 
along in the midst of space by the force of universal gravitation, revolv- 
ing in an orbit distant 91^ millions of miles from the sun, which not 
only retains it, but also gives it beauty and life. From it we also de- 
rive heat, to the consideration of which we now proceed. Let us first 
see how heat, and its distribution over the surface of the globe, are to 
be estimated. 

To measure the variations of temperature, the thermometer {Bepfxog, 
heat; fxirpov, measure) is used, just as the barometer was invented, as 
we have seen above, for ascertaining the variations in atmospheric 
pressure. Without discussing at greater length the employment of the 
thermometer, or the various forms of the instrument, than we did the 
above contrivance, it is, nevertheless, interesting to go back to its dis- 
covery, which also dates from the middle of the seventeenth century. 

Our ancestors judged of temperature pretty much in the same way as 
we do in the present day, viz., by the principal effects resulting from it. 
Nowadays, science measures it more in detail and more uniformly by 
means of special instruments which permit of a comparison between the 
results obtained in different countries, or between those of one epoch 
and another. When the academicians of Florence established the fact 
that all bodies undergo a change in volume under the influence of heat, 
they laid the basis of thermometry. The instrument of which these sa- 
vans made use consisted of a sphere soldered to a narrow tube, and con- 
taining colored alcohol. When this apparatus is transferred from one 
place to another warmer place, the liquid becomes dilated and the level 
rises, thus showing the augmentation of the temperature. This appara- 
tus dates from 1660. In order that thermometers might be suitable for 
comparing with each other (that they might, that is to say, give the 
same indications under the same circumstances), the academicians of 
Florence had them all constructed, as nearly as was possible, upon one 
standard. A natural philosopher of Pavia, one Charles Renaldi, was 


the first to suggest, about 1694, the means, still in use, for obtaining 
thermometers suitable for making comparisons. The plan consists in 
placing the instrument successively in two calorific positions, invariable 
and easy of reproduction, viz., those corresponding to the melting of ice 
and the boiling of water. Between these limits of temperature any 
given body becomes dilated by the same fraction of its volume. As 
a rule, is marked at the point at which the liquid of the thermometer 
stands in melting ice, and 100 at the point where it remains stationary 
in the midst of boiling water. These two points being marked upon 
the stem, the interval between them is divided into one hundred equal 
parts. Newton, having conclusively demonstrated the fixity* of the 
point at which water boils, the means adopted by Eenaldi to render 
thermometers capable of comparison was adopted by all physical phi- 
losophers. This is the Centigrade thermometer, the most convenient, 
and the most in use.f Thirty years ago, Pouillet engaged in a series 
of ingenious and patient experiments, with a view of determining the 
quantity of heat transmitted to the earth by the sun, and the tempera- 
ture of space — that is to say, the two constituent elements of the tem- 
perature of the globe. 

The two contrivances made use of for the purpose were the pjrheli- 
ometer and the actinonieter. The latter, being only used for researches 
as to the temperature of the zenith, need not occupy our attention 

The pyrheliometer is in principle composed of a thin silvered box, a 
(see Fig. 47), four or five inches in diameter, and holding, perhaps, three 
or four ounces of water. Its surface, turned toward the sun, is black- 
ened. A thermometer is introduced into the box and embedded in the 
copper frame- work, b. The water in the box, at the same temperature 
as the surrounding air, is exposed for five minutes to the sun. In order 
to ascertain that the side of the box is quite perpendicular to the sun's 
rays, care is taken to see that its shadow falls exactly upon the lower 
disk, c, of the same diameter. By comparing its temperature with the 
temperature of the air previous and subsequent to its exposure, the 

[* That is to say, under the same atmospheric pressure. The boiling-point of water varies 
every day with the height of the barometer ; and, in fact, a method often used by explorers for 
determining the heiglit of the barometer (and therefore then- own elevation above the sea-level) 
is to find the temperature at which water boils at the place in question. — Ed.] 

[t The thermometer used in England is Fahrenheit's. The temperature of melting ice is 
marked 32°, and that of boiling water (when the height of the barometer is 29*92 in.) 212°. 
thus 180 graduations on the Fahrenheit scale correspond to 100 on the Centigrade. — Ed.] 



quantity of heat received from the sun in a minute by each square inch 
of ground can be found and expressed in heat-units.* 

Making allowance for the atmospheric strata traversed by the solar 
rays, the experimentalist discovered that the pyrheliometer would be 
raised 12°-1 Fahr. if the atmosphere were capable of transmitting in its 
totality all the solar heat, without itself absorbing any, or if the appa- 
ratus could be placed at the limits of the atmosphere to receive at that 
point, without any loss, the heat transmitted to us by the sun. 

Fig. 47.— The Pyrheliometer. 

We can thus tell the quantity of heat which the sun spreads in the 
space of a minute over a square inch at the limit of the atmosphere, and 
which, would also be received at the surface of the ground, were it not 
that the air of the atmosphere absorbed some of the rays as they passed 
through it. 

From these data and the law in accordance with which transmitted 

[* The heat-unit generally adopted in English works is the quantity of heat necessary to 
raise one pound of ice-cold water one degree Centigrade, viz., to raise one pound of water from 

0° to 1° C— Ed.] 


heat diminishes in proportion as the obliquity increases, it is easy to 
calculate the proportion of incident heat which arrives each instant upon 
the lighted hemisphere of the globe, and the proportion absorbed in the 
corresponding half of the atmosphere. The calculation shows that 
when the atmosphere is, to all appearance, quite still, it is absorbing 
nearly one-half of the total quantity of heat which the sun emits to us, 
and that it is only the other half of this heat which reaches the ground. 
Since the sun, as has been calculated, transmits every minute to each 
square yard of the ground that it shines perpendicularly upon a de- 
gree of heat equal to about 85,200 heat-units, it is easy to conclude 
therefrom the total quantity of heat which the terrestrial globe and 
its atmosphere together receive in a year. The result is more than 
2,660,000,000,000,000,000,000 heat -units! This heat would raise, if 
such were possible, by 2315 degrees, a body of water three feet three 
inches deep, and enveloping the whole. By transforming this quantity 
of heat into a quantity of melted ice, the following result is arrived at : 
If the total quantity of heat which the earth receives from the sun in 
the course of the year was uniformly distributed over all parts of the 
globe, without any loss in melting ice, it would be sufficient to melt a 
coat of ice enveloping the whole globe to a depth of about one hundred 
feet. Such is the simplest way of expressing the total quantity of heat 
which the earth receives each year from the sun. 

It is this gigantic quantity of heat which sets in motion the mechan- 
ism of terrestrial action, which lets loose the tempests over the ocean, 
and, in a word, sustains the vast aerial life of this planet. The same 
fundamental data permit of our ascertaining the total amount of heat 
which is emitted from the sun in a given time. 

Let us consider this luminary as the centre of a vast sphere, the ra- 
dius of which is equal to the mean distance of the earth from it; then it 
is evident that over the surface of this sphere each square 3'ard receives 
every minute from the sun precisely as much heat as the square yard 
of the earth — that is to say, 35,200. Consequently, the total quantity of 
heat which it receives is equal to its entire surface, expressed in yards 
and multiplied by 35,200. 

The same thing may be expressed by stating that the terrestrial 
globe, with its 8000 miles of diameter, only intercepts, in this sphere of 
^1^ million miles of radius, 2 5 0^0 of the total heat that leaves that 
luminary, and that the heat emitted by the sun is 2,300,000,000 greater 
than that received by the earth. 

HEAT. 185 

Transforming into the quantity of melting ice, we obtain the follow- 
ing result : 

If the total amount of heat emitted by the sun were exclusively em- 
ployed in melting a coat of ice placed right around the body of the sun, 
it would be capable of melting in a minute a thickness of nearly 39f 
feet — that is, a thickness of more than 10|- miles in the twenty-four 
hours ! 

One part of this immense source of " energy " is employed in heating 
the terrestrial rind to a certain depth ; but as the soil and the atmos- 
phere radiate into space, and as the terrestrial globe does not seem to 
lose or gain in reference to the mean temperature, at least during long 
periods of years, all this part of the sun's radiation may be considered 
as maintaining the equilibrium of the temperature of our planet. An- 
other part is transformed into molecular movements, in chemical action 
and reactions, which are the source whence the life of animals and veg- 
etables derives unceasingly the wherewithal of their perpetuation and 
sustenance. Heat, which thus seems necessary for these beings, is but 
an emanation from our luminary. As Tyndall remarks, "It is thus 
we are, not merely in the poetical sense, but practically, children of the 

The American engineer, Ericsson, the inventor of the solar steam-en- 
gine, has calculated that the mechanical effect of the solar heat which 
falls upon the roofs at Philadelphia would keep in motion more than 
5000 steam-engines, each of 20-horse power. 

The work done in raising the temperature of a pound of water by 
one degree Fahrenheit is exactly as great as that required to raise a 
weight of one pound to the height of 772 feet. 

Solar heat is the source of the only natural works that man has yet 
been able to divert to his profit, and among them we must include 
water-courses and the winds. 

Moreover, the combustible matter of manufacture is derived from the 
same luminary ; as wood, it is carbon absorbed by the vegetables breath- 
ing in the air under the influence of the sun ; as coal, it is still carbon 
that has been in earlier ages fixed by the same influence in the large 
antediluvian forests. 

The sun's rays, after having traversed either the air, a pane of glass, 
or any transparent body, lose the faculty of retreating through the same 
transparent body to return toward celestial space. It is by a procedure 
founded upon this physical law that gardeners accelerate in spring the 


vegetation of delicate plants by covering them with a glass bell, admit- 
ting the solar rays, which have great difficulty in effecting their egress. 
If the gardener places two or three of these bell-glasses one upon an- 
other, he invariably burns up the plant underneath them, and even in 
the mild weather of March or April he is often obliged to raise one of 
the edges of the glass to prevent the plant from being injured by the 
sun at noon. By means of an apparatus composed of a box blackened 
inside, and of several pieces of glass laid one upon the other, Saussure 
was enabled to raise water to boiling-point; and Sir John Herschel, 
during his stay at the Cape of Good Hope, in the burning heat of the 
last days of December, was enabled to cook a piece of " boeuf a la mode" 
of very fair dimensions, by means of two blackened boxes placed one 
inside the other, and each provided with one single glass, with no other 
source of heat than the solar rays which were ingulfed without possi- 
bility of escape in this kind of trap. " There was," says M. Babinet, 
"sufficient to regale the whole of his family and their guests at this 
meal, prepared with a stove of such a novel kind." 

Herschel's box, closed only by two panes of glass, reached successive- 
ly 80, 100, and 120 degrees of heat. 

Although this oven appears so novel, it may almost be said to be 
taken from the Greeks. We find, indeed, that a century before the 
Christian era. Hero of Alexandria described in his "Pneumatics" a 
large number of ingenious contrivances devised by the ancients, and, no 
doubt, by the learned hierophants of Egypt. One of these, which seems 
to have been constructed by Hero, draws water from a reservoir by the 
sole effect of the dilatation and condensation of air under the influence 
of the sun alternately shining on and concealed from the apparatus. 

At the close of the sixteenth century, the Neapolitan savant, J. B. 
Porta, set forth in his "Natural Magic" the mechanical applications of 
solar heat. If, he says, a hollow copper globe is placed upon the sum- 
mit of a tower, and if from it there descends a pipe into a reservoir of 
water, by heating the globe above, either by means of fire or the sun's 
rays, the rarefied air escapes. Soon after, when the sun declines, the 
copper globe cools, the air becomes condensed, and the water rises up 
the pipe. 

The concentration of solar heat in a glass-covered inclosure is an ex- 
periment so easy that the observation of it must have followed very 
closely upon the invention of glass. Nevertheless, despite the different 
proofs of this fact and the applications of its principle to which I have 

HEAT. 187 

alluded, there is no complete scientific study of the phenomenon earlier 
than that of Saussure. Subsequent to his work and that of Herschel, 
the subject had been considered in various lights by different philoso- 
phers. This curious problem is just now in perhaps its most interest- 
ing phase, viz., that which gives on the one hand serious results, and 
on the other allows the imagination to guess at others in the future still 
more important. 

It is a natural question to ask. What is the temperature of the sun? 
To this we can give no satisfactory answer. Two estimates have been 
made by Secchi and Zollner, which, however, differ enormously, the 
former giving about 19,000,000° Fahr., while the latter only amounts to 
about 49,000° Fahr. 

To determine the temperature of the sun, an apparatus has been used 
which exposes the thermometer to its rays in an inclosed place, the 
temperature of which is previously ascertained. Eeading the indica- 
tion given by the mercurial column, the number is multiplied by the 
ratio of the surface of the celestial sphere to the apparent surface of the 
sun. As the solar disk has a mean diameter of 31' 3"'6, the ratio of 
the whole celestial sphere to this is 183,960. The apparatus in ques- 
tion is as follows: Two concentric cylinders soldered together form a 
kind of double caldron, the annular interval of which may be filled 
with water or oil at a given temperature. A thermometer passes by 
means of a small tube through the annular space and penetrates to the 
axis of the cylinder, where it receives the solar rays, which are intro- 
duced by means of a diaphragm, the orifice of which is scarcely larger 
than the ball of the thermometer. The interior cylinder and its ther- 
mometer are covered with lamp-black ; a second thermometer gives the 
temperature of the annular space and, consequently, that of the inclos- 
ure. The whole apparatus is mounted upon a stand having a parallac- 
tic movement, corresponding to the diurnal motion of the sun. 

The apparatus being exposed to its rays, the two thermometers are 
noticed, the difference of their temperature gradually increases, and at 
the end of a certain time becomes constant. The two temperatures are 
then marked and the difference calculated. 

One word must be added as to the interior heat of the earth, Mai- 
ran, Buffon, and Bailly estimated, so far as France is concerned, the 
heat which escapes from the interior of the earth at twenty-nine times 
as much in summer, and four hundred times as much in winter, as that 
which reaches us from the sun. Thus, the heat of the luminary which 


gives us light would, if this were true, form but a small fi-action of that 
of the globe. This idea was developed with great eloquence in the 
"fipoques de la Nature," but the ingenious romance to which it forms 
a basis is dispelled like a phantom before the stern evidence of math- 
ematical calculations. Fourier having discovered that the excess of the 
temperature of the terrestrial surface over that which results from the 
mere action of the solar rays has a necessary relation to the increase of 
the temperatures at different depths, succeeded in deducing from the 
amount of this increase, as found by experiment, a numerical determi- 
nation for the excess in question — that is, for the thermometrical effect 
which the central heat produces upon the surface. And, instead of the 
high figures given by Mairan, Bailly, and Buffon, he obtained as his re- 
sult only the thirtieth part of a degree ! 

The surface of the globe, which, at the beginning of the world, was 
probably incandescent, has cooled down, in the lapse of ages, so much 
as to retain scarcely a trace of its primitive temperature. Nevertheless, 
we know that the temperature increases as we descend into the interior 
of the earth at the rate of 1° to about 112 feet, on the average, and that 
the heat must be very great underneath volcanoes. Upon the surface 
(and the phenomena of the surface can alone alter or compromise the 
existence of human beings) all changes are limited to about the thir- 
tieth part of a degree. The terrible congelation of the globe, which 
Buffon fixed for the epoch when the interior heat should be entirely 
dissipated, is therefore a mere dream.* 

* [M. Flammarion concludes this chapter with a discussion of the temperature of space, and 
he states that the mechanical theory of heat shows that there is an absolute zero of temperature 
at — 459° Fahr. ( — 273° Cent.), so that no body can be colder than this; it being, in fact, the 
temperature of a body totally devoid of heat, and therefore the temperature of space. I should 
merely have contented myself with the omission of this portion of the chapter without remark, 
only it appears to me that the error reproduced by M. Flammarion is sufficiently wide-spread to 
make it worth while to call attention to the matter. In point of fact, we have no evidence for 
asserting that the temperature of space is — 459°. We know that gases, at ordinary tempera- 
ture, expand equally by heat, so that if a thermometer were made in which the fluid was air 
kept at a constant pressure, its reading would be the same as if any other gas were used, the 
pressure being the same. Consider, therefore, a thermometer composed of air contained in a 
long straight tube, so arranged that the pressure of the gas is kept constant whatever its vol- 
ume may be, and suppose the freezing and boiling points determined as usual, and the inter- 
vening space divided into 180 equal parts, as in Fahrenheit's scale, then it follows, assuming 
Boyle and Mariotte's law, that if the graduations were continued right down to the end of the 
tube, the last division would be marked veiy nearly — 459°, so that it is clear that no tempera- 
ture, however low, can correspond to — 459° of the thermometer (i. e. , the air thermometer can 
never read so low as — 459°), as in that case the air would have been compressed into nothing ; 

HEAT. 189 

but as it is clearly convenient to start from the end of the tube, this point can very well be 
taken as our zero, merely to measure from. There are other reasons, of a more strictly scien- 
tific character, derived from thermo-dynamical considerations, that also lead to approximately 
the same point as the absolute zero of temperature ; but they do not, in the very slightest de- 
gree, imply that this is the temperature of space ; in fact, such an assertion would be unintel- 
ligible, even if true, without much explanation. The lowest artificial temperature observed is 
— 220° Fahr. ( — 140° Cent.), obtained by Natterer, by exposing to evaporation a mixture of 
nitrous oxide and carbonic disulphide. — Ed.] 




It now becomes necessary to ascertain what part of the immense cal- 
orific radiation which is incessantly emanating from the sun is at work 
in the atmosphere. 

Meteorology is nothing but a great physical problem. We have to 
determine what are the laws which regulate the manner in which heat, 
barometrical pressure, vapor of water, and electricity, are distributed in 
our atmosphere, in relation to the movements which the solar heat en- 
genders in the solid, liquid, and gaseous superficial stratum of our 
globe. This problem, vast as it is, says Father Secchi, is in reality but 
an application of the best known laws of physics ; the difiiculties of 
solving it are owing rather to the large number of disturbing causes, 
and to the incalculable reactions of effects upon causes, than to any real 
deficiency in the general theory. Hence the necessity of numerous ex- 
perimental data in order to arrive at a complete solution. 

The atmosphere is in reality an immense machine, to the action of 
which is subordinated every thing upon our planet that has life. 
Though there are neither fly-wheels nor pistons in this machine, it none 
the less does the work of millions of horses — a work the aim and effect 
of which is the sustenance of life. 

All the movements of the atmosphere are the consequence of the 
property which gases possess of being expanded by heat. The varia- 
tions of volume, and, consequently, of density, are, at each instant, dis- 
turbing the equilibrium which would be tending to establish itself in 
the atmosphere. The air, heated in the equatorial zones, rises into the 
upper regions to fall again near the poles; there it becomes cool, re- 
turns to the equator, and recommences its circulating movement. The 
work thus performed in the atmosphere is enormous. To this property 
of gas must be added another, not less important — that of dissolving'^ 

[* The air and the vapor of water form, as it were, different atmospheres : that is to say, 
that the vapor atmosphere could still remain if all the air were removed. Water, placed im- 
der the air-pump, evaporates till the space under the receiver is filled mth aqueous vapor to an 
extent dependent on the temperature. — Ed.] 


the vapor of water which, as it rises in prodigious quantities about the 
equator, is thence distributed all over the earth in the shape of rain. 
Thus is effected another and scarcely less potent work — the distribution 
of rain over the surface of the globe. The running waters which set 
our machines in motion were originally raised into the air by this 
mighty agency ; from thence they pour down on the mountains in the 
shape of rain, run into the rivers, and so make their way again into the 
ocean from whence they started. 

The sun is the power that regulates all the movements of the 
planetary system ; not only their motions in their orbits, but also 
the physical or physiological phenomena which take place upon their 
surface. On the earth, in particular, the atmospheric movements and 
those of the waters, the development of vegetation, the production 
of the force which results from combustion and the nutrition of ani- 
mals — all these phenomena are due to the influence of the sun's heat 

What may seem to us still more perfectly organized is the way in 
which this calorific power is, so to speak, stored up in the vegetables ; 
not only in those which, still alive, serve for our use and nourishment, 
but also in those which, buried for ages in the bowels of the globe, at 
length emerge therefrom to warm us and supply our machines with the 
required motive power. Each plant is a veritable machine, in which 
are elaborated the extremely combustible substances which serve to 
furnish us, in the absence of the sun, with heat and light, or to produce, 
in providing us with nutriment, the force and vital warmth which we 
stand in need of. It is, therefore, on the sun, as Father Secchi again 
remarks, that depend entirely all the phenomena of nature, and our ex- 
istence itself. 

In the solar radiation what is at first so striking is the light which 
gives us day, and the heat which warms us ; but, besides these two or- 
ders of phenomena, there is a third of equal importance, viz., the chem- 
ical actions which accompany the two others. Thus three classes of ac- 
tion must be distinguished in the solar work — the luminous rays, the 
calorific rays, and the chemical rays. It is well known that, to analyze 
a sunbeam, it is passed through a triangular prism of glass, on emerging 
from which the ray is decomposed into a colored ribbon, as we have 
already seen in our study of the rainbow. But the visible spectrum is 
not the only component part of a sunbeam. The many-hued ribbon is 
continued at each end by an invisible ribbon. The waves — the length 


of which is included between -0000167 and -0000266 of an inch— are 
capable of causing our optical nerve to vibrate, and thus producing the 
sensation of light, the diversity of colors being dependent only upon the 
length of the waves, the longest of which belong to the red rays, and 
which gradually diminish toward the violet. To the left of the red ex- 
tremity of the spectrum, there are long and slow waves of heat. To 
the right of the violet end, there are short and rapid waves of chemical 
action. The eye sees neither the first nor the second of these, but they 
may be recognized by the use of suitable apparatus. In reality, how- 
ever, there exists in nature but one single series of waves, the lengths 
of which continually decrease from the extremity of the obscure calo- 
rific spectrum to the extremity of the invisible chemical spectrum. 
Between these two extremes there is but a very limited part which has 
the power of giving sensation to the optical nerve. 

Fig. 48 shows the relative extent and intensity of these different ac- 
tions, separated from each other as they are made manifest to us by the 

Fig. 48.— Relative Intensity of the calorific, luminous, and chemical rays of the sun. 

dispersive action of a prism. The band which forms the basis of this 
figure indicates the length of the solar spectrum. From a to h is the 
luminous part ; to the right, from H to P, is the invisible chemical part ; 
to the left, from A to s, is the calorific part, also invisible. The curves 
traced above show the relative intensities at each point of the spectrum. 
The intensity of the light is represented by the curve r' m' t', that of 
chemical action by m m" p, that of calorific radiations by r m T. It has 
been attempted to represent the three respective intensities by the three 
bands, 1 (light), 2 (heat), and 3 (chemical action). 

Thus we do not see all that goes on in nature. The luminous ravs 


are the only ones which we can behold ; but the calorific and chemical 
rays take effect without being visible to us. 

The illuminating power of the different rays consists in their greater 
or less capacity for giving an impulse to the optical nerve. It is prob- 
able that the faculty of perceiving luminous phenomena has not the 
same scale for every individual, and that it is much more extended in 
the case of certain animals than with man, both at the red and violet 
ends of the spectrum. Pure water possesses a very considerable absorb- 
ing power for thermal rays. The moisture contained in the eye differs 
very little from pure water, and it is this fact which very likely renders 
the organ of sight insensible to calorific rays. The extent of the lu- 
minous waves which are sensible to the eye ordinarily corresponds to 
what is called in acoustics an octave, so that man is only placed in re- 
lation with nature by a very small part of the solar rays. 

Gases possess the faculty of absorbing heat rays, and consequently 
our atmosphere absorbs a very considerable portion of the rays which 
are transmitted to us. The longest waves are those which are most 
easily absorbed ; thus a large number of the less refrangible rays which 
fall upon our atmosphere are stopped, and do not reach us at all. 

The absorption produced by the simple gases, oxygen and nitrogen, 
is extremely small ; but this does not hold good of the compound gases 
existent in our atmosphere, such as carbonic acid, vapor of water, am- 
monia, etc. Professor P. M. Garibaldi,* of Genoa, haa- proved by con- 
clusive experiments that, for a pressure of 29'92 inches, these gases have 
absorption-powers represented by the following figures : 

Atmospheric air 1 

Carbonic acid 92 

Ammonia 546 

Vapor of water 7937 

A quantity of vapor of water capable of producing a pressure of 0'4 
inch exercises an absorption a hundred times greater than that of at- 
mospheric air. Thus, a considerable portion of the dark heat rays pro- 
ceeding from the sun are intercepted by the vapor of water contained in 
the air, and are unable to reach the surface of the earth. 

The luminous rays have been separated from the heat rays by Pro- 
fessor Tyndall. To effect this, a pencil of solar rays was made to pass 

[* Professor Tyndall has made an elaborate and careful series of experiments on the ab- 
sorptive power of different gas3s. He concludes that on an average day the water present in 
the air absorbs about sixty times as much heat as the air itself. — Ed.] 



through a solution of iodine in carbonic disulphide. The rays beconne 
invisible without losing their calorific power, and if the vessel contain- 
ing the solution is made in the shape of a convex lens,* the temperature 
in the focus of the lens is increased to such an extent that combustible 
bodies may take fire there. Professor Tyndall on one occasion placed 
his eye at the focus, and the retina received no luminous influence. 
The calorific rays were, however, so powerful that a sheet of metal was 
immediately made red-hot at the very spot where the eye bad received 
no impression. The ratio of luminous radiations to obscure radiations 
is equal to -^-^ for incandescent platinum. For sunlight, the heat which 
accompanies the luminous portion of the spectrum is only i of that 
which is found in the obscure part. 

The action of the atmosphere is to raise the temperature of the earth, 
for it allows the calorific rays to reach the earth, and then prevents them 
from making their way back into space. The greater portion of heat 
rays sent out from the earth are no longer able to traverse the atmos- 
phere, while only a few of the sun's rays, which are of high tempera- 
ture, are stopped. 

Further, the nocturnal radiation is considerably diminished by the 
presence of the atmosphere, and in this way the cooling of the globe 
and of the plants which it nourishes becomes modified and diminished. 
The vapor of water acts very efficiently, and a moist stratum, only a 
few yards thick, arrests the nocturnal process of cooling as completely 
as the whole atmosphere. 

But the most striking fact in connection with this matter is the great 
absorption of heat which accompanies the transformation of water (or 
any other liquid) into vapor. The heat so absorbed is termed latent 
heat, from its not being spent in raising the temperature of the vapor. 
Water evaporates in large quantities, especially in the equatorial regions, 
and thus absorbs a large quantity of heat which remains latent. As 
much heat is necessary to vaporize one pound of water (at the tempera- 
ture of the boiling-point) as to increase by 1° (Centigrade) the heat of 
537 pounds of water. The vapor of water absorbs this enormous pro- 
portion oF heat, which it, however, restores in its entirety when it re- 
turns to the liquid state as rain. This heat is destined to be transport- 
ed to the most distant latitudes, and to establish in the atmospheric en- 

[* The solution must be inclosed in a lens of rock-salt, a substance which allows heat to 
pass through it without absorbing scarcely any ; it is therefore termed diathermanous, an ad- 
jective having the same reference to heat that transparent has to light. — Ed.] 



velop which surrounds the globe an equality of temperature which 
would not otherwise be produced. The quantity of heat which thus 
passes from the equator to the poles is beyond conception. 

Thus, for instance, numerous and rather exact observations have 
taught us that in the equatorial regions evaporation each year removes 
a body of water at least sixteen feet deep. Let us suppose that in the 
same regions there is an annual rain-fall of rather over six feet, there 
still remains a quantity of water represented by a depth of nearly ten 
feet, and which must pass, in the form of vapor, into the countries nearer 
to the poles. The surface over which the evaporation takes place may 
be estimated at seventy million geographical miles ; and, starting from 
this datum, it will be seen that the depth of ten feet represents a volume 
of water equal to twenty-five thousand billions of cubic feet (25 x 10'^). 
This enormous mass of heat passes incognito^ so to speak, from the equa- 
tor to the poles, transported by the action of the vapor, and this latter, 
as it becomes transformed into water and ice, sets free all the heat which 
it had absorbed, thus contributing to make milder the climate of these 
desolate regions. In this way the heat is distributed in the atmosphere, 
and thus are created clouds and rain, which will be explained below. 

The thickness of the strata of air traversed by the solar rays has a 
notable influence upon heat and light. The rays do not fall upon the 
earth perpendicularly, but obliquely, and the loss is greater the more 
they are inclined to the vertical. 

This diminution has been submitted to different calculations : the two 
formulae which seem to be most in harmony are those of Bouguer and 
Laplace. Making use of them, the following results are arrived at, as 
to the thickness of the strata of air for the different heights of the sun : 

Height above 


Thickness of the 

the Horizon. 


Strata of Air. 

90 deg. 



70 " 

20 " 


50 " 

40 " 


30 " 

60 " 


20 " 

70 " 


15 " 

75 " 


10 " 

80 " 


5 " 

85 " 


4 " 

86 " 


3 " 

87 " 


2 " 

88 " 


1 " 

89 " 



90 " 


Thus, if the thickness of the atmosphere traversed by a ray of the 



sun at the zenith be represented by 1, the thickness traversed by the 
sun's rays at the horizon is more than thirty-five times greater. This 
diflference is much larger than can be indicated in the annexed illustra- 
tion (Fig. 49). The first 
s'> result of this inequali- 

ty is, that the sunlight 
becomes feebler in pro- 
portion as the sun sinks 
toward the horizon. At 
the zenith and in the 
higher regions of the 
sky the sun is dazzling, 
and no human eye can withstand its blaze. At sunrise and at sunset we 
are able to fix our eyes upon its reddened disk without inconvenience. 
The smaller stars do not become visible till they reach a certain height, 
and we can only witness the rising and setting of those of the first mag-' 
nitude. According to the researches of Bouguer, if 10,000 be taken to 
represent the luminous intensity of the sun as it would be seen from a 
point external to the atmosphere, its intensity at the different altitudes 
above the horizon may be thus stated : 

At 50 degrees 8123 

Fig 49. — Inequality of the thickness of air traversed by the sun, ac- 
cording to its position above the horizon. 

' 30 
' 20 
' 10 
' 5 
' 4 
' 3 
' 2 
' 1 










That is to say, that, at sunrise and sunset, this luminary has only 
YTTi of its apparent brilliance when at the zenith, and two of its brill- 
iance when at its midday elevation over our horizon during the sum- 
mer solstice. These comparisons are made on the supposition that 
the sky is clear, and consequently vary with the more or less misty 
state of the atmosphere. Heat varies, like light, with its angle of inci- 
dence. The most accurate observations prove that the atmosphere ab- 
sorbs, of vertical rays, '28 of the heat which falls upon its surface, and 
the total absorption in the illuminated hemisphere is about equal to 



three-fifths of the incident heat ; the transmitted part at different heights 
being represented as follows : 



At the zenith 


' ' 70 degrees 

" 50 " 

" 30 " 

" 10 " 

" " 

As was remarked above, it is not the air itself — that is to say, the 
mixture formed of oxygen and nitrogen — which absorbs the most heat, 
but the vapor of water, which always exists in the air, but in very vary- 
ing proportions. 

The luminous rays pass almost in their entirety, and reach the ground ; 
the heat rays are, on the contrary, absorbed to a large extent. If, there- 
■fore, the atmosphere prevents a great part of the solar heat from reach- 
ing the surface of our globe, it makes up for it by retaining for us the 
part that we do receive. Without the atmosphere and the vapor of 
water contained in it, since the radiation of the soil goes on almost with- 
out obstacle toward the interplanetary space, the loss would be enor- 
mous, as indeed is the ease in the higher regions. No sooner has the 
sun set than a rapid coldness succeeds the intense heat of the sun's di- 
rect rays ; in a word, there is an enormous difference between the max- 
ima and minima of temperature, either daily or monthly. This oc- 
curs upon the lofty plateaux of Thibet, and explains the severity of 
the winters, and the decline of the isothermal lines in these regions. 
Tyndall says very truly : " The suppression, for a single summer's night, 
of the vapor of water contained in the atmosphere over England (and 
the proposition holds true for all the countries in similar latitudes) would 
be accompanied by the destruction of all the plants which are killed by 
frost. In the desert of Sahara, where the ground is fire and the wind a 
flame., the cold at night is often very difficult to support. In this hot 
country ice is seen to form in the course of the night." 

Moisture is not distributed in equal proportions at all elevations of 
the atmosphere. "We shall see, further on, that it decreases in amount 
beyond a certain height. Heat traverses air the more easily, the less 
moisture it contains. After the lower regions of the atmosphere have 
been passed, and (say) an altitude of 6000 feet attained, it is impossible 
to avoid noticing the very considerable increase in the heat of the sun 


relatively to the temperature of the surrounding air. This fact never 
struck me so strongly as in an aeronautical ascent on June 10, 1867, on 
which occasion I noted, at 7 a.m., at a height of 10,000 feet above the 
ground, that there was, for half an hour, a difference of 27° Fahr. be- 
tween the temperatures of my feet and head; or, to speak more accurate- 
ly, between the temperature of the interior of the car (shade), and that 
of the exterior (sun). The thermometer marked 46° Fahr. in the shade, 
and 73° Fahr. in the sun. While our feet felt the effects of this relative 
cold, a hot sun scorched our necks and faces, and those parts of the 
body directly exposed to the solar radiation. The effect of this heat is, 
of course, further augmented by the absence of the slightest current of 

In a subsequent ascent, I experienced at the same time the remarka- 
ble difference of 36° Fahr. between the temperature in the shade and 
that in the sun, at an altitude of 13,500 feet. 

The influence of altitude upon the intensity of the sun's calorific in- 
fluence at points nearly vertically above one another has recently been 
studied very carefully by M. Desains and a colleague at the Schweitzer- 
hofif, Lucerne, and at the Eighi-Culm Hotel, about 1500 yards above the 
lake. These experiments demonstrated that at the same hour, and un- 
der equal conditions, the solar radiation was more intense upon the 
summit of the Eighi than at Lucerne, but that it was less capable of 
being transmitted through water. It was found that the solar rays in 
their passage — at an angle of about seventy degrees with the vertical — 
through the stratum of air comprised between the level of the Eighi 
and that of Lucerne, underwent a loss of about seventeen per cent. 

This shows that the terrestrial temperatures depend not only on the 
quantity of heat received from the sun, but also and especially upon the 
ahsorhing power of the air in regard to the rays of heat. Such also, no 
doubt, is the case in the other planets; and the influence of the atmos- 
pheres is such that, in spite of its close proximity to the sun, it is possi- 
ble that Mercury may possess a much lower temperature than that of 
the earth, and the surface of Jupiter may present a climate far warmer 
than ours, in despite of its greater distance from the sun, 

Eecent spectroscopic experiments of M. Janssen render the existence 
of planetary atmospheres generally similar to our own probable, As- 

[* My experiences at high elevations were quite opposed to those stated in the text. At 
great heights I observed no difference in reading between a thermometer witli the sun shining 
full on its bulb, and another in which the bulb was carefully shaded.— Ed.] 


tronopical observations also long since pointed to the same conclusion 
with regard to some of the planets. 

After having appreciated the action of solar heat throughout the at- 
mosphere and upon the surface of the globe, to which, in fact, almost 
every movement that takes place there can be traced, we must now 
complete the account by 'noticing that the amount of this heat dimin- 
ishes as we ascend higher into the atmosphere. We have seen that the 
pressure of the air diminishes in proportion as we rise higher into it. 
The temperature is subject to an analogous decrease, which may be esti- 
mated, though not nearly so accurately as in the case of the diminution 
of the atmospheric pressure. Corresponding to the indications of the 
barometer, the following are those given by the thermometer: 

When an ascent in a balloon is made with the sky cloudy, the tem- 
perature generally declines until the clouds have been reached ; once 
above them, a rise of several degrees always takes place, but the tem- 
perature soon begins to fall again. With a clear sky, the initial tem- 
perature is, cceteris paribus, higher than in the preceding case by a quan- 
tity about equal to the rise observed after emerging from tjie clouds. 
The diminution of heat is never absolutely regular, as strata of hot air 
are nearly always encountered in the atmosphere, sometimes five or six 
in succession, at very great elevations. These alternations, and this va- 
riability of the sky, do not prevent the manifestation of one general 
fact — that of the decrease in the temperature with an increase of eleva- 

The following is the result of a series of observations upon this point 
which I have made in the course of my various ascents : 

The decrease in the temperature of the air, which plays so important 
a part in the formation of the clouds and in the elements of meteorolo- 
gy, is far from following a regular and fixed law. It varies according 
to the hours, the seasons, the state of the sky, the direction of the 
winds, the condition of the vapor of water, etc., etc. It is only after a 
great number of observations that it is possible to deduce any fixed 
rule, several secondary causes which must be first ascertained and elim- 
inated being always at work. 

From several observations, taken in very dissimilar conditions (which 
are, however, less unfavorable than those under which observations are 
taken on mountain sides), it follows that the decrease in the tempera- 
ture of the air differs, in the first instance, according to whether the sky 
is clear or cloudy, being more rapid in the first case than in the second. 

200 -^-ff^ ATMOBPHERJE. 

With a clear sky, the mean fall in the temperature has been fc^ind to 
be 7° Fahr. for the first 1600 feet from the surface of the ground; 13° 
at 3280 feet; 19° at 4900 feet; 23° at 6560 feet; 27° at 8200 feet; 31° 
at 9840 feet ; 34° at 12,500 feet— an average of 1° Fahr. per 340 feet. 

With the sky cloudy, the fall in the temperature is 5^° Fahr. for the 
first 1500 feet; 11° at 3000 feet; 16° at 4900 feet; 19° at 6560 feet; 
29° at 9840 feet; 32° at 12,500 feet— of average of 1° Fahr. per 350 

The temperature of the clouds is higher than that of the air immedi- 
ately above or beneath them. The decrease is more rapid near the sur- 
face of the ground, and more gradual at greater elevations. It is also 
more rapid in the evening than in the morning, and also in warmer 
than in colder weather. Regions hotter or colder than the mean tem- 
perature for their altitude are sometimes met with in the atmosphere, 
crossing it like aerial rivers. Notwithstanding these variations, the 
general law enunciated above is the expression of the true state of 
things. The difference between the indications of the thermometer in 
the shade and in the sun augments with elevation. Thus, the general 
result of these aerial ascents tends to show that the temperature de- 
creases about 1° Fahr. for an elevation of 345 feet. The result of the 
well-known and numerous aerostatical observations taken by Glaisher 
differs but little from the above. The ascents of mountains have fur- 
nished a certain number of important data, among which may be 
quoted the following : 

Humboldt found that the decrease, in a southern atmosphere, was 1° 
Fahr. to 344 feet in the mountains, and to 440 feet upon the table-lands. 
A series of places in Southern India gave 320 feet ; in the north of Hin- 
doostan, on the other hand, the decrease was 1° in 410, an amount ap- 
proaching to that noted by Humboldt upon the table-lands of America. 
Everywhere analogous differences of level are remarked ; in Western 
Siberia, 1° in 450 feet is the result arrived at; and this number is con- 
verted into 440, if the comparison includes the elevated regions of 
Northern India. In the United States the decrease is 1° to 400 feet. 
The configuration of the country seems to be the most important ele- 
ment in the calculation. If there is a gentle rise in the ground, or if 
the country is made up of successive gradients, the decrease in the tem- 
perature is much more gradual than upon the sides of steep mountains. 
In the first case, 1° may be taken to represent a difference in level of 
420 feet ; in the second, of 350 only. 


Schouw remarked in Italy, upon the southern slopes of the Alps, a 
decrease of 1° to 300 feet ; less on Mount Ventoipc, a steep and isolated 
mountain in Provence (lat. 44° 10' K, long. 2° 56', height 6270 feet 
above the level of the Mediterranean). Martins found, after nineteen 
observations, taken under dissimilar conditions, a decrease of 1° to 840 
feet in winter, and 230 feet in summer, or an average of 260 feet. The 
observations of Ramond, made between 43° and 44° of latitude, give an 
average of 1° to 265 feet. 




We have seen that the earth, by its annual revolution round the 
sun, and by its daily rotation upon its axis, produces a variation in the 
obliquity of the solar rays which find their way to it. By its annual 
revolution, they become more and more vertical during six months of 
the year — from December 21 to June 21 — and less and less so during the 
other six months. By its rotation the horizon each morning is brought 
into the presence of the sun, causing the heat-giving luminary to reign^ 
in the heights of the heavens during the day, and in appearance to sink 
again to the horizon on each evening. Thus, it is evident that, by these 
two movements of the earth, there are two general principles in regard 
to solar heat upon our planet; the one annual, the other diurnal. 

Let us consider the latter first. To determine it exactly, the ther- 
mometer must be consulted hourly, night and day, for several years to- 
gether, in order to distinguish and eliminate the effects due to the rota- 
tion of the earth from those due to the numerous other causes which in- 
fluence change of temperature. Few meteorologists have been willing 
to undertake so arduous a task. Ciminello of Padua made such obser- 
vations for nearly sixteen consecutive months. I say verij nearly^ be- 
cause the observations at midnight, and at the hours of one, two, and 
three in the morning, were replaced by two, taken during the same in- 
terval at different hours. He was the first to make hourly series of 
thermometrical observations. Since that time others have been made 
by Gatterer, a contemporary of his ; by the artillery officers at Leith ; 
by Neuber at Apenrade, in Denmark; by Lohrmann at Dresden; by 
Koller at Kremsmunster ; by Kaemtz at Halle; and at the Observa- 
tories of Milan, St. Petersburg, Munich, and Greenwich. Such observa- 
tions are now continuously recorded at the Roman Observatory, and 
some others by means of a self- registering apparatus. 


The result of these observations, and of many others which have been 
made every two or every three hours, shows that the hottest moment of 
the day is two in the afternoon, and the coldest about half an hour be- 
fore sunrise. These two limits vary but little from one month to an- 
other. The diflfei'ence of temperature between the hottest moment and 
the average coldest period of the twenty-four hours is about 14° at Paris. 
This amount, however, varies with the time of the year. 

The average yearly maximum temperature at the Paris Observatory 
is 58° at 2 p.m. ; the average minimum is 44°-8 at four in the morn- 
ing; and the average mean temperature of the year, as taken at 8"20 
A.M. and 8-20 p.m., is 51° -3. 

The interval of time, between the minimum in the morning and the 
maximum in the afternoon, is only ten hours; and the interval is four- 
teen hours, viz., from 2 P.M. to 4 a.m. between the time of maximum 
and the next minimum. The minimum of the diurnal variation, as 
a rule, takes place just before sunrise; in the early part of the year it 
is just before 6 a.m., and occurs earlier as the days lengthen. After 
the month of February it occurs at about 5 a.m., then at 4 a.m., af- 
terward oscillating between three and four in the morning during the 
longest days. In the beginning of August the minimum is again at 
4 a.m., returning to about 6 A.M., when the days are at their shortest. 
It is even somewhat later than this for a short period, but soon after- 
ward resumes the annual progress given above. 

The mean temperature of a day, in the mathematical acceptation of 
the term, represents the average of the temperatures corresponding to 
every instant of the day. If the duration of these instants be a minute, 
it would be necessary to divide the sum of the 1440 thermometrical ob- 
servations taken between two consecutive midnights by 1440 (the num- 
ber of minutes in twenty-four hours), and the quotient would give the 
required mean temperature. Again, by dividing by 365, the sum of 
the 365 mean temperatures of every day in the year, we should obtain 
the mean annual temperature. 

It would seem, from the preceding definition, that to obtain the mean 
temperatures accurately, observations at short intervals would be indis- 
pensable ; but the change of temperature under ordinary conditions is, 
fortunately, of such a nature that the half sum of the maximum and 
minimum temperatures (at 2 p.m. and sunrise) is found to differ but lit- 
tle from the mean of observation taken at every hour. So early as 
1818, Arago pointed out that the average temperature at 8*20 A.M. was 


nearly the same as the average temperature of the year. Numerous 
thermometrical observations taken under his direction were based upon 
the fact of the mean temperature of the day, occurring twice in the 
course of the day. But it has since been found that this method is de- 
fective; for from 8 to 9 A.M., and also from 8 to 9 p.m., the temperature 
often varies very rapidly. The averages were afterward formed by 
taking the temperatures at 4 a.m. and 10 a.m., and again at 4 p.m. and 
10 P.M., adding and dividing by four. The arithmetical mean of the 
observations taken at 6 a.m., 2 p.m., and 10 p.m., also gives about the 
same value, the difference being about two-tenths of a degree. Since 
meteorology has been more methodically followed, still greater accura- 
cy has been acquired, and it has been found that the twenty-four hour- 
ly observations may be replaced by eight tri-hourly observations, taken 
ut 1 A.M., 4 A.M., 7 a.m., and 10 a.m. ; and at 1 p.m., 4 p.m., 7 p.m., and 
10 P.M. 

Let us now consider the annual movement of the temperature. 

The various causes which influence the action of the sun's heat vary 
but little throughout the year in the regions near the equator, whether 
situated in the northern or the southern hemisphere, the tropical re- 
gions, as they are called, and which form the torrid zone. The day has 
about the same length all the year round ; the meridian height of the 
sun undergoes but little variation there; and the four seasons differ 
very little, in regard to temperature, the one from the other. For an 
entirely opposite cause, the seasons are very dissimilar both to the 
north and to the south of the equator in the regions where the length 
of the day varies very much in course of the year, or, to express the 
same thing in other words, where the meridian height of the sun at one 
solstice is very different from that of the other. 

We have considered the general condition of the seasons in our lati- 
tudes. Let us now consult the figures themselves. The table append- 
ed gives the mean temperature at the Paris Observatory. 

It shows that, whether the average maximum or the average mini- 
mum of each month be taken into account, or, indeed, if we merely 
take the mean temperatures alone, the heat follows an ascending scale 
from January to July, and a descending scale from July to December. 
The hottest month is that of July, which follows upon the summer sol- 
stice, and the coldest that of January, which comes after the winter sol- 
stice. The average of the minima is only once (in January) below 32° ; 
the coldest months are December, January, and February, constituting 



the real climatological winter ; spring is made up of the months of 
March, April, and May ; summer of the three hottest months, June, 
July, and August ; and the other three months, September, October, 
and November, form the true autumn. 












October ■.... 



Annual Temperatures. 










The above averages are those which Arago arrived at after forty-six 
years of observations (1806-1851). Since then, further observations 
have given a result still more in conformity with the secular mean tem- 
perature of Paris, representing, as it does, a longer series of years. 

The heat received from the sun by the earth varies inversely as the 
square of its distance from the sun ; and as the earth does not move in 
a circular orbit, there is, in addition to the monthly variation caused by 
the inclination of the solar rays, a variation due to the distance of the 
sun. In fact, we are farther from the sun during: the summer than we 
are in the winter; and the difference is considerable. The following 
are the deviations, taking as unit the mean solar distance, and regard- 
ing the heat as reciprocal to the square of the distance : 


Solar Heat. 

Mean distance 





In Perihelion (least distance) 

In Aphelion (greatest distance) 

Thus, before even reaching our atmosphere, the solar heat rays are 
subject to a variation of nearly one-fifteenth ; that is to say, that the 
solar heat during winter is, in respect to our globe, about one-fifteenth 
greater than during summer. 

This difference is sufficiently great to be taken into account. 


The diurnal and monthly variations of temperature increase as the 
distance from the equator increases. From the equator to 10° north 
latitude, the mean temperatures of the various months scarcely differ 
more than 4° to 6°. At 20° north latitude they vary from 10° to 12°. 
At 30° distance the regular mean monthly variation is found to reach 
22°. In Italy the regular curve at Palermo, in Sicily, extends from 
51° to 74°; and this range is moreover diminished by the contiguity of 
the sea. At Paris the mean curve varies from 35|-° (January) to QQ!^ 
(July), and the changes become much greater between the frosts of win- 
ter and the heat of summer. At Moscow the mean monthly range 
extends from 12° (January) to 75° (July) ; showing a difference of 
63° of mean temperature. Lastly, we may add to this scale of varia- 
tions that of Boothia Felix, a northern country of America, situated be- 
yond 72° of north latitude. There the range varies from —40° (72° 
below the freezing-point of water) in February to 41° in July; exhibit- 
ing a difference of 81° between the mean monthly temperatures of the 

The diurnal variation also gives rise to remarkable curves in its suc- 
cessive temperatures. The range of thermometrical oscillation is great- 
er in warm climates and inland countries than it is in colder lands and 
in the neighborhood of the sea. Apart from the equalizing influence 
of the sea, which remains about the same all the year round, the dis- 
tance from the equator acts in an opposite way upon the annual and 
the diurnal oscillations of the thermometer. While the first increases 
on account of the length of the nights in winter and of the days in sum- 
mer, the second decreases because in the southern countries the heat of 
the sun's rays is greater and the sky clearer during the night. Thus, 
for instance, at Padua the diurnal variation in July is about 16° ; at 
Paris it is on an average about 13° ; at Leith it is about 9°. 

These are the mean values. But if the changes of temperature in a 
given district be constantly recorded, it will be found that, apart from 
these regular mean variations caused by the sun, there are other varia- 
tions of a much larger amount which exercise the greatest influence 
upon the public health ; these are the diurnal variations that occur in 
the space of twenty-four hours. These differences are very interesting, 
especially if we notice the reading of a thermometer with its bulb 
placed in the full rays of the sun by day, and of another with its bulb 
exposed fully to the clear sky at night. There are also often very 
great differences between the maximum and the minimum temperatures 



of the air of the same day, especially in the months of May and June — 
differences which reach, in Paris, to as much as 45° to 55°. 

The following are some of the maxima observed at Montsouris, be- 
tween 1 and 4 p.m., with a thermometer with green bulb, exposed to 
the sun at a height of about four inches above grass, as also some of 
the minima taken from the same thermometer between one and four the 
following morning. I select those that exhibit the greatest differences : 
























- 52-3 








































May 11, 1870. 

" 16, " . 

" 17, " . 

" 18, " . 

" 19, " . 

" 20, " . 

" 21, " . 

" 25, " . 

" 27, " . 

" 30, " . 

June 8, " . 

" 12 " 

" 13; " : 

" 14, " . 

" 16, " . 

" 23, " . 

" 29, " . 

" 30, " . 

July 2, " . 

This shows how great at times are the diurnal variations in these lat- 
itudes. The mean temperature of a place is that found by adding up 
the annual mean temperatures and dividing their sum by the number 
of years during which the observations have been taken. This mode 
of operation is only applicable to a limited number of stations. It was 
necessary, therefore, to seek a method of obtaining, by means of experi- 
ments which could be readily made, approximate mean temperatures, 
with a fair approach to accuracy. We know that the surface of the soil 
undergoes daily variations of temperature, that lower down there is a 
stratum which experiences only small annual variations, and that at a 
greater depth still, at about seventy to eighty feet, there is a stratum 
with constant temperature which is found to be very nearly the same 
as the average of a long series of the daily temperatures of the atmos- 
phere made at the same place. By finding the temperature of this stra- 
tum at a sufficient depth, or, which comes to the same thing, by ascer- 
taining the constant temperature of springs or wells in a certain district, 
or even of tunnels, we may thus succeed in obtaining for the tempera- 



ture of each place a number differing but slightly from that which 
would be found by taking a long series of annual temperatures at that 
place. In the equinoctial regions, a thermometer simply sunk in the 
earth to the depth of thirteen inches in sheltered spots will continue to 
mark the same degree of temperature with a difference of 0° 2' or 0° 4' 
of a degree at most. For this purpose a hole is dug under the tents of 
the Indians or inside a shed, in a place where the ground is protected 
from the heat caused by the direct absorption of the solar rays, from 
nocturnal radiation, and from the infiltration of rain. 

By taking the temperature of springs as that of the highest stratum 
of constant temperature, there will be found a great similarity in re- 
spect to the zone comprised between 30° and 55° latitude, provided 
that the places are not more than 3000 feet above the level of the sea. 

In respect to latitudes above 55°, the difference between the tempera- 
ture of the air and of springs increases to a marked extent. 

Toward the peak of the Swiss Alps, above an elevation of from 4600 
to 4900 feet, as in the high latitudes, the springs are nearly 6° Fahr. 
warmer than the air. 

In Southern countries the temperatures of springs and of the ground 
are less than the mean temperatures of the air, as may be gathered from 
the accounts of Humboldt and Leopold von Buch. 

In our latitudes this temperature is equal to that of the soil near the 
surface, and is a little higher than the average of the particular place. 

It is worth our while to complete this general study of the meteorol- 
ogy of our climate by enumerating the mean temperatures at Paris and at 
Oreenivich since the heginning of the present century. They are furnished 
from the archives of the Observatories at Paris and at Greenwich. 




(Dec, Jan., Feb.) 

(June, July, August.) 






























































































(Dec, Jan., Feb.) 

(June, July, August.) 





































































































































































































































































































1 ^ 























































































































49 1 






































































































































' The preceding table shows that the coldest winter of the present 
century in Paris was that of 1830 ; the mildest, that of 1869 ; the cold- 
est summer, that of 1816; the hottest, that of 1812 ; the coldest year 
was 1829 ; and the hottest, 1834* 

This list gives the mean annual temperature of winter and summer, 
as ascertained at the Paris Observatory. We shall see farther on that 
there have been more severe frosts and greater heat in France than 
those given, but they have been observed at different places. 

I have already stated that, taking the mean temperatures of each day 
of the year at Paris, it would be seen that there is an increase in heat 
from the first week in January to the middle of July, with a continuous 
decrease from the latter date until the close of the year. The general 
phenomenon, however, exhibits certain discontinuities which can not 
be treated so simply. 

It is true that, generally speaking, it is the movement of the earth 
which gives rise to the grand phases of the temperature, and which 
produces in our climates, for instance, a minimum in January and a 
maximum in July. But the curve which unites these two extreme 
points is not regular. There are unmistakable departures from conti- 
nuity which seem subject to periodical returns. 

In its more general aspect, the question may be put in the following 
manner : 

What is, for a given locality, the mean departure which the tempera- 
ture of each day in the year exhibits in relation to the supposed regular 
march of these temperatures between the annual extremes? 

[* M. Flammarion has given this table for Paris only ; I have added the correspoTiding 
values for Greenwich, as taken from my paper in the Philosophical Transactions for the year 
1848, supplemented by subsequent results. I may remark that I have altered some values 
in M. Flammarion's table as seemed to be necessary by comparison with the tables in the 
"Annuaire" for 1872. 

This table shows that the coldest and warmest winters, the coldest summer, and the coldest 
year, were the same at Paris and at Greenwich, and that the warmest summer and the warm- 
est year at Greenwich was 18(58. 

It also shows that the most severe winter of all was at Paris, and that the winter tempera- 
ture of Paris is frequently lower than at Greenwich, although generally it is higher. 

The mean temperature of the winter at Paris from all the years is 38° '4, while that at 
Greenwich is 37° '1. 

The mean temperature of the summer months has in every case been warmer at Paris than 
at Greenwich. 

The mean of summer at Paris is 64°-7 ; at Greenwich, 60°'4. 

The mean temperature of every year is higher at Paris than at Greenwich : the mean at 
^Paris is 51°-3 ; that at Greenwich, 48°-9.— Ed.] 


Is this departure about the same for each year, or for a small group 
of years — or does it, on the contrary, vary from one year to another, or 
from one group of years to another, so as to present a certain periodical 
recurrence ? 

The questions, which are secondary to the first general question, are 
very numerous, inasmuch as the quantities of light which enter into the 
atmosphere, the electric state of the air, and its so-called ozonometrical 
properties, its hygrometrical condition, as also the variations in the at- 
mospheric pressure, the displacement of the air, or the winds and tem- 
pests — in a word, all the atmospheric phenomena are intimately bound 
up with the distribution of heat over the surface of the globe. 

Lastly, a very natural and important addition consists in the influ- 
ence of these thermometrical perturbations upon the health of men, ani- 
mals, and of plants. 

Four epochs in the year are remarkable for a fall in the temperature, 
and atmospheric perturbations caused thereby, viz., about the 12th of 
February, the 12th of May, the 12th of August, and the 12th of No- 

The periodical cold of the month of May is a popular tradition ; hor- 
ticulturists term St. Mamert, St. Pancras, and St. Servais, whose anniver- 
saries are on the 11th, 12th, and 13th of May, the three ice-saints. 

In February there are the same indications, but they are even more 
marked. The fall after the 7th of February is very sudden, and con- 
tinues to the 12th, which gives but a single minimum even in the mid- 
dle of the ice-saints of February. As February with us represents 
Northern climates, every thing will be extreme, the rise as well as the 
fall ; in August, on the other hand, which gives us an idea of the trop- 
ical climates, the changes are less sudden, and the slight movement cor- 
responding to that of the 10th to the 14th in May, or, in another form, 
of the August ice-saints, continues until the 16th. 

In November, as in August, the decline of the temperature is seen to 
be struggling against influences which tend to an abnormal return of 
heat; the points of inflection correspond precisely to those of the three 
other months, and one of the last of them produces, on the lith, the 
Martinmas summer. 

The careful examination of a large number of years shows that, at 
London and Berlin, as at Paris, there is a certain agreement between the 
four days of the same date, as exhibited in their mean temperatures. M. 
Ch. Sainte-Claire Deville ascertained that these curious periods are to be 


found in the most ancient of known meteorological documents; for in- 
stance, in the observations of the pupils of Galileo, and of the Academy 
of Cimento. These observations extend over fifteen j^ears (1655-1670). 
The minimum of the ice-saints is found to occur on the 12th, M^ith a 
remarkable regularit3\ 

It is certain for the last two centuries, in this part of Europe, that the 
periodic anomalies of the temperature, some of which were proverbial 
among our ancestors, have manifested themselves in the same manner 
stated above. 

Certain astronomers, Erman and Petit among the number, have at- 
tributed these frigorific phenomena to masses of asteroids which, in their 
orbit, sometimes come between the sun and the earth. 

The action of the sun produces, therefore, in the temperature of the 
air, variations according to the hours of the day and the month of the 
year. This same solar action produces a diurnal and a monthly varia- 
tion in the readings of the barometer, which, perhaps, had better be con- 
sidered here, as it is a consequence of the temperature. 

The atmospheric pressure increases and decreases twice each day with 
regularity, in a manner dependent on the sun's position. The reading 
of the barometer, which shows the w^eight of the atmosphere, gradu- 
ally increases from 4 to 10 A.M. This atmospheric tide is not due, like 
that of the sea, to the attraction of the moon and the sun, since it takes 
place every day at the same hour, and does not follow the course of the 
moon. It is due to the expansion produced by the solar heat, and to 
the increase in the vapor of water, also produced by this same heat. 

This barometrical variation is not great, for it never attains so much 
as one-tenth of an inch. It was about the year 1722 that the diurnal 
variations of the barometer were first ascertained by a Dutchman, whose 
name has not reached us. Since that epoch, several observers have en- 
deavored to determine their amounts and their periods for different 
parts of the earth. Humboldt proved, by a long series of observations, 
that, at the equator, the maximum of elevation corresponds with 9 a.m.; 
after that hour, the barometer reading decreases until 4 or 8'30 p.m., 
when it attains its minimum. It afterward increases again until 11 p.m., 
when it reaches a second maximum, and, lastly, decreases again until 4 
A.M. Thus there are each day two minima, at 4 a.m. and 4 p.m. ; and 
two maxima, at 9 a.m. and 11 p.m. The movements are so regular that 
a simple glance at the barometer suffices to ascertain the hour, especial- 
ly during the day, without any probability of being more than a quar- 




























— - 


ter of an hour in error. They are so permanent that neither tempest, 

nor '^torm, nor rain, 

fects it ; they main- 
tain themselves as 
steady in the warm 
regions of the coast 
of the New World 
as upon table -lands 
more than 13,000 
feet high, where the 
mean temperature 
falls to 441°. The 
amount of the os- 
cillations diminish- 
es as the latitude in- 
creases, in the same 
manner as the mean 
temperature of a 
place is, in general, 
higher the nearer it 
is to the equator. 

At the Antilles, it 
is found that there 
is a distinctly-mark- 
ed maximum for the 
diurnal oscillation 
along the northern 
coast of America, 
which is situated op- 
posite to the sea of 
the Antilles. The 
stations upon this 
coast -line give, on 
an average, an am- 
plitude of Oil inch- 
es; whereas at all 
v\„ ^ T.O 1 V 1 Mw r.v- T, . , . . ., the other stations 

iMg. 50.— Regular diurnal oscillation of the Barometer: I.Ascension Isl- 
and ; 2. Port d'Espagne ; 3. Acapulco ; 4. Cumaua; 5. Basse-Terra. the amOUnt is Small- 
































+ '^v r ""n. T 

__^-^^_lt ^ — Z 1- 



er, whether they are situated to the north or south of the littoral region 
in question. " 

The northern coasts of Venezuela and New Grenada are exactly those 
which the thermal equator follows, rising in this district to the twelfth 
degree of north latitude, whence it descends again toward the equator, 
on both sides of the continent. The place of the maxima oscillations 
of the barometer is, therefore, the same as that of the maxima tempera- 
tures, and the two phenomena follow a similar march in the intertropic- 
al American zone. This is, moreover, quite in accord with the causes 
which influence the distribution of temperature over the different hours 
of the day. 

Various observations have made it evident that the amplitude of the 
total oscillation diminishes with increased altitude. It may be stated 
as a general rule that this amplitude is dependent on the mean tempera- 
ture of the place, and that it decreases with it not only according to the 
vertical co-ordinate of the altitude, but according to the two co-ordinates 
of latitude and longitude. 

The diurnal oscillation of the barometer varies with the latitude as 
follows : 



Mean Height. 








12-3 S. 





10-31 N. 
5-6 S. 




4-36 N. 
















Rio de Janeiro 

22-54 S. 
19-26 N. 
















































St. Petersburg 





The last column of this table shows that at 60° of latitude the diurnal 
barometrical oscillation is very small. 

In our climates these hourly variations are so masked by accidental 



variations that to discover and measure them was a work requiring the 
greatest sagacity and precision. It is only by the average of many 
years' observations taken with care and at suitable hours that the hour- 
ly periods can be arrived at. 

The following table gives the diurnal and monthly atmospheric vari- 
ation due to the expansion of air by solar heat, as found from observa- 
tions at the Paris Observatory : 














Means of the Year. 

Meau Heights of the Barometer reduced to the 
Temperature of 0°. 

At 9 A.M. At Noon. At 3 p.m. At 9 p.m. 











This table shows that the morning maximum attains on an average 
29-772 inches, and the afternoon minimum 29-743 inches: the differ- 
ence is 0-029 inch. It, moreover, shows that there is not only a diur- 
nal variation of the barometer, but also a monthly variation. 

The atmospheric pressure decreases gradually from January to April, 
increases a little up to July, decreases a little until November, and then 
increases in December and January. This movement of the atmospher- 
ic pressure is almost the exact opposite of that of the temperature ; it is 
much more marked in the tropical regions, as may be seen by consult- 
ing the curves which M. Deville traced in the Antilles. The amplitude 
of the monthly oscillation is there on an average (29'81 — 29*69 = ) 0"12 
inch, between January and April, according to observations taken at 
noon. The nearer one approaches the tropics, the greater it is : corre- 
spondents at the Calcutta Institute inform me that 0^7 inch represent 
the amplitude between January and July, and at Benares 0*6 inch. 

The observations at Brussels show that the diurnal and monthly va- 
riations in our climates are distinct. By comparing them it is seen that 
the diurnal maxima of temperature are pretty constant during the year, 
occurring about 10 a.m. and 10 p.m. As to the minima, the interval 




between them is greater in summer than in winter; the two quantities 
also exhibit a greater deviation in the summer months. During the 

Jan. Tel). Mar. %. May. June. July, Aug. Sept. Oct. ?fov. Dec Jan. ^^^^^^^t dajS (Novem- 
ber, December, January), 
there are only eight 
hours between the mini- 
ma, which occur at 6 a.m. 
and 2 p.m., whereas dur- 
ing the other months the 
interval between them is 

The time at which 
the first minimum takes 
place varies more than 
two hours, being at 8"30 
A.M. in June, and 6'22 
A.M. in December. 

There is an equally 
great change in the time 
of the first maximum. 
The extreme limits take 
place at 10-50 a.m. in 
February, and at 8*40 
A. M. in June. Local 
causes exercise a certain 
influence upon the epochs 
of these extreme limits. 

The epoch of the sec- 
ond minimum varies still 
more, as it occurs at 2*15 
P.M. in January, and at 
5*30 P.M. in June, show- 
ing a difference of time 
of three and a quarter 
hours. The limits with- 
in which the baromet- 
rical epoch varies are, in 
the case of the first max- 
imum and the first minimum, about two hours. The interval of time 


























O yp^ V 

Fig. 51.— Regular monthly oscillation of the Barometer : 1. Cay- 
enne; 2. Guiana; 3. Trinidad; 4. Santa -Fe de Bogota; 5. 


which elapses between the first maximum and the second minimum de- 
serves especial attention, there being a separation of four hours only in 
January, which increases to eight in June, the latter being the double 
of the former. The results show that the total diurnal variation is 
made up by the combination of two waves : the one scarcely percepti- 
ble, which, in the space of twenty-four hours, has a maximum and a 
minimum of 0'009 inch only; the other much greater, with two maxi- 
ma and two minima of O'Ol inch. 

Such are the regular variations of the barometer, due to diurnal and 
annual action of solar heat. These are the least important variations. 
The atmosphere is unceasingly in movement by influences which ac- 
quire a greater intensity, although they have the same origin. The 
irregular variations are larger, and increase from the Equator to the 
Poles. While the extreme differences of the barometer do not exceed 
upon an average two-tenths of an inch in the equinoctial regions (ex- 
clusive of the cyclones, which will be alluded to hereafter), they reach 
to two and three inches in our latitudes. 

The greatest barometrical variations occur in winter, the smallest in 

At all times of the year the barometer reading is higher during the 
minima of temperature than during the maxima. 

It is especially during autumn and winter nights that the differences 
of temperature have the greatest effect on the reading of the barometer. 
In spring this influence is much less, and is to a great extent disguised 
by other causes. 





The first summer of the present century, or, to speak more exactly, 
according to chronology, the summer of the last year of the eighteenth 
century, was remarkable for its high temperature, and I might com- 
mence the series with it but for the fact that Europe experienced an 
exceptional degree of heat at a date which will remain famous, that of 

The summer of this year was memorable for the intense and unexam- 
pled heat, which occurred in July and August. According to Cassini 
IV., then director of the Observatory, the results for Paris were: 

Great heat (77° to 88°) 36 days. 

Very great heat (90° to 93°) 9 " 

Extraordinary heat (95° or higher) 6 " 

The highest temperatures occurred as follows: 

Valence, July 11 104 '0 degrees. 

Paris, July 8 lOl'l " 

Paris, August IG 99-1 " 

Chartres, August 8 100-4 " 

Chartres, August 16 100-6 " 

Verona, July and August 96-1 " 

London, July 16 89-1 " 

The great heat began in Paris about the 1st of July, and increased 
very rapidly. The sky remained continually clear and cloudless; the 
wind was in the north and generally very gentle, and the barometer re- 
mained very high. The hottest days were the 8th and 16th of July. 
On the 9th, a fearful thunder-storm raged at Senlis and the immediate 
neighborhood. Hailstones as large as eggs destroyed the crops ; a tre- 
mendous wind blew down more than 120 houses. This tempest was 
followed by very heavy rain, and the water, collecting in the fields, 
swept off cattle, furniture, women, and children. At Bougueval, in the 
Oise, a woman, after rescuing her nine children, was swept off by the 
flood. On the 10th of July^ to complete the destruction, there came a 
second hailstorm. 

The extreme heat of the month of July continued through part of 


xlugust. On the 7th of this latter month it was very great, very gen- 
eral, and most oppressive. The sky was still clear, and the wind, from 
the north-east, was so scorchino- that it seemed as if emitted from the 
mouth of a furnace. It came by whiffs, and was as severe in the shade 
as in the sun. This was experienced not only throughout Paris, but 
in the country districts as well. The suffocating heat paralyzed the 
breathing, and was felt far more severely on that day, with the ther- 
mometer at 86°'6 Fahr., than on the 8th of July, when it was 101°"2. 

The dryness of the ground became very great. The level of the wa- 
ters of the Seine fell to the lowest water-mark of 1719, at the end of 
August and in the middle of September. There fell in Paris but 10'9 
inches of rain in the year. In the country, trees and shrubs were gen- 
erally burned up; and fruits, includijig apples among them, showed signs 
of having been burned. There was a great scarcity of vegetables. The 
land, dried up, hardened and cracked : it was impenetrable both to the 
plow and the spade. Workmen engaged in sinking a well in a place 
exposed to the sun, found the soil dried up to a depth of more than five 
feet. By the 1st of September the trees had lost nearly all their leaves ; 
the dryness and the heat had caused the bark to crack and the branches 
to look dead. Very many of the trees did, in fact, die. 

In Burgundy the vintage began on the 23d of September; the wine 
was abundant, but of inferior quality, as the vines had been affected by 
a cold rain which fell in that district. The summer was dry and hot in 
the neighborhood of Toulouse, and the maize crop was a complete fail- 
ure. It will be remembered that 1793 was a year of great scarcity in 

1800. — The summer was marked by extreme heat, which extended 
over part of Europe. From the 6th of July to the 21st of August the 
temperature decreased but five times below 7-l°.2 ; there were, according 
to Bouvard, of 

Great heat 25 days. 

Very gi-eat heat 5 " 

Extraordinary heat 2 " 

The highest temperatures were as follows : 

Bordeaux, August 6 101 "8 degrees. 

Nantes, August 18 101 '8 

Montmorency, August 18 100'2 

Limoges 99 "5 

Paris, August 18 95-9 

London, August 2 88 "0 


Conf3agrations were very numerous in the early part of April. A 
whole village in the department of the Eure, the forest of Haguenau, 
and part of the Black Forest, became a prey to the flames. Myriads of 
grasshoppers alighted in the neighborhood of Strasbourg, In the night 
of July 20th the ancient monastery of the Augustins, in Paris, was 
struck by lightning. In the south there occurred numerous cases of 
sudden madness. 

1811. — The summer of 1811 was in many respects one of the most 
memorable known in Northern Europe. 

The following is the table of the maxima temperatures: 

Augsburg, July 30 99'5 degrees. 

Vienna, July 6 96-3 " 

Avignon, July 27 95-0 " 

Hamburg, July 19 94-6 " 

Naples, July 20 94-3 " 

Copenbagen, July 92 '8 " 

Liege 92-7 " 

Strasbourg 91'4 " 

St. Petersbin-g, June 27 88-0 " 

Paris, July 19 87-8 " 

In Burgundy the vintage began on the 14th of September. A hail- 
storm that occurred on the 11th of April spoiled two-thirds of the crop; 
but the summer was very favorable for vines, and the small crop yield- 
ed wine of an excellent quality, which was long famous as the comet 

1822. — The summer of 1822 was remarkable throughout France for 
high mean temperature. 

At Paris there were of 

Great heat 55 days. 

Very great heat 3 " 

The maxima of temperature were as follows : 

Malines, in July 101 "8 degrees. 

Joyeuse, .June 23 99-1 " 

Alais, June 14 and 23 97-7 " 

Lie'ge 95-0 " 

Maestricht, June 11 93-2 " 

Paris, June 10 , 92-8 " 

The drought was very great in France during the warm season : from 
the 21st of August to the 26th of September the Seine was nearly con- 
tinuously below the mark of zero at the Pont de la Tournelle. As early 
as the month of March there was a scarcity of water ; for cattle in the 


south of France water had to be brouglit from great distances upon the 
backs of mules, and the spring temperature in that country was as high 
as that generally experienced in August, The harvest was finished in 
Languedoc by the 23d of June, and though there was very little straw 
the ears were well filled. In Burgundy the sky was unusually clear. 
The vintage began on the 2d of September, but, according to the vine- 
dressers, it might have been begun on the 15th of August, and in the 
neighborhood of Vesoul the grapes were gathered on the 19th of Au- 
gust. There was an average quantity of wine, and the quality was ex- 
ceptionally good ; the grain crops were, as a rule, less abundant than 

1826. — A very hot and dry summer : thirty -six days of great heat in 
Paris, seven of very great, and two of extraordinary heat. The mean 
of the summer was very high, 69|-° Fahr. Crops were destroyed, and 
forests burned, in Sweden and Denmark. 

The highest temperatures observed were — 

Maestricht, August 2 •. lOl'S degrees. 

Spinal, July 1 97-7 " 

Paris, August 1 97*2 " 

Metz, August 3 97-0 " 

Strasbourg 93 "6 " 

1834:. — This year, though not remarkable for any very great heat, is 
noticeable for the very high mean temperature of the spring and sum- 
mer throughout France. Vegetation was very forward, and there fell, 
in many places, rain distributed in such a manner as to be most favor- 
able to the crops. In Paris there were forty-three days of great, and 
three days of very great heat. 

The mean average of the summer, 69°, is the highest of the century, 
next to 1826, 1812, and 1846. The drought was very great in August. 

The maxima temperatures of 1834 are thus distributed : 

Avignon, July 14 9;V0 degrees. 

Geneva, July 18 93-9 " 

Lie'ge 92-3 " 

Metz, July 12 91-4 " - 

Strasbourg 91 "0 " 

Paris, July 12 and 18 9()-7 " 

In the south, the temperature was lowered by plentiful rains, and was 
very mild. Burgundy was this year celebrated for the superior quali- 
ty of its wine, though the quantity ran very short. Such was also the 


case in the Bordeaux district. The harvest was almost universally 
good in France. 

1836. — The summer of this year was memorable for the stormy na- 
ture of the month of June and the early part of July, and for the num- 
ber of fatal accidents caused by the heat in the south of France. In 
Denmark, Russia, and Spain, the temperature also produced some re- 
markable effects. 

The drought in the month of August was intense ; the Seine fell about 
ten inches below the low-water mark of 1719. There was an average 
crop of wine in the south of France, the quality being fairly good. The 
vintage did not begin in Burgundy till the 6th of October. The corn 
harvest was bad. 

18i2. — The summer of this year was the hottest during the first half 
of this century, especially in Paris and the north. 

In Paris there were of 

Great heat 51 days. 

Very great 11 " 

Extraordinary 4 " 

The mean temperature of the season in Paris was 69° 4. 
The following is a list of the highest temperatures recorded : 

Paris, August 18 99"0 degrees. 

Agen, July 4 98-6 " 

Bordeaux, July 16 94-6 " 

Toulouse, July 17 93-9 " 

Many accidents caused by the heat were registered; wheels of sev- 
eral mail-carts took fire. At Badajoz, in Spain, three laborers died of 
the heat on the 28th of June, and a lady expired from its effects in a 
diligence. At Cordova several reapers fell down asphyxiated, and fre- 
quent cases of madness were attributed to the same cause. 

In Burgundy the vintage began on the 21st of September, the crop 
being abundant and the quality good. The grain crop was below the 

1846. — The temperature this summer was very remarkable, and there 

were periods of intense heat in France, Belgium, and England. In 

Paris there were of 

Great heat 48 daj's. 

Very great 9 '• 

Extraordinary 2 " 

The mean summer temperature was 69°"2. 


The maxima of this year are as follows : 

Toulouse, July 7 104 '0 degrees. 

Quimper, June 19 100'4 

Rouen, July 5 98 "2 

Paris, July 5 97-7 

Orange, July 13 97-7 

Angers, July 29 95-0 

Metz, August 1 1 94-(; 

Accidents occurred in Brittany. At the Pont-de-Croix fair several 
persons had fits, occasioned by the heat ; at Benzec, a little girl, left in 
the sun, was found dead a few minutes afterward. The temperature 
during June was also very high at Toulouse, Toulon, and Bordeaux. 
In the Landes, farmers obtained a second crop of rye. Near Niort, 
early in July, three laborers died while at work. 

The vintage in Burgundy began on the 14th of September, the qual- 
ity being exceptionally good, though there was only a half crop. The 
corn harvest, too, was much below the average. 

1849. — The heat was very great in the south of France, and the max- 
imum at Orange is the highest temperature in the shade yet recorded 
in France. 

The table gives the following figures: 

Orange, July 9 lOG'S degrees. 

Toulouse, June 23 99-7 " 

Bordeaux, July 7 94-3 " 

Gand..... 93-9 " 

Metz, Julys 92-5 " 

1852. — This was a remarkable summer in Russia, England, Holland, 
Belgium, and France. There were in Paris of 

Great heat 30 days. 

Very great (! " 

Extraordinary 1 " 

The summer mean in Paris was 67°. The mean of July was 724-°. 
There was an unusual continuance of great heat: July 9th, 88°'0; the 
10th, 92° -3 ; the 11th, 87°-8 ; the 12th, 90°-5 ; the 13th, 92°-8 ; the 14th, 
93°-6; the 15th, 93°-6; the 16th, 95°-2. 

The highest temperatures throughout Europe were— 

Constantinople, July 27 101 "3 degrees. 

Rouen, July ,5 97"0 " 

Versailles, July 1(5 96-3 " 


Orange, August 25 95-5 

Dunkerque, July 7 96-3 " 

Paris, July 16 95*2 " 

Verviers, July 18 95*2 " 

London, July 12 95*0 " 

At Amsterdam, a thermometer rose, on July the 12th, to 102°-2. At 
Alphen, near Leyden, two peasants, asphyxiated by the heat, were found 
dead in a field ; at Alkenaer an engine-driver became insane, after con- 
gestion of the brain produced by sun-stroke. In the centre of France 
the thermometer stood for more than 10 days at over 86°. Many do- 
mestic animals perished from the heat. At Thourotte, in Belgium, 
there fell a disastrous hailstorm on the 11th of August: many of the 
hailstones weighed from two to three ounces, and were from two to 
three inches in diameter. 

In France the harvest was mostly over by the middle of July, and 
was an average. On the other hand, the vintage did not begin till the 
early part of October, and the wine crop was small in most vineyards, 
and of inferior quality. 

1857. — This summer was hotter than usual in France, and the months' 
of July and August were nearly everywhere distinguished for extreme 
heat. The highest temperatures observed were — 

Montpellier, July 29 101*5 degrees. 

Orange, July 18 100-9 " 

Les Mesneux, August 4 98"6 " 

Toulouse, July 27 98*2 " 

Clermont, July 14 and 15, and August 3 98-2 " 

Blois, in August 97'7 , " 

Paris, August 4 97-2 " 

Metz 96-1 

There were three distinct streams of summer heat. The first, on the 
27th of June, passed over the highest and the most southerly stations 
in France, and reached, on the 28th, the northern frontier; the second 
extended over the north-west, from the 14th to the 16th of July ; the 
3d, and the most intense, moved slowly and in the same direction, trav- 
eled from south to north in the interval between July the 27th and 
August the 4th. The drought this summer was very great through- 
out nearly the whole of France; fortunately, in the middle of August, 
some rain fell. 

In Burgundy the vintage began on the 16th of September, and the 


The corn 

crop was passable as to quantity and good as to quality, 
crops were, generally speaking, up to the average. 

1858. — This summer was remarkable for great drought, and pro- 
longed rather than intense heat, in England, Belgium, the centre and a 
part of the south of France, and Algeria. In the north the heat was 
less than in 1857, but greater in the south. The maxima tempera- 
tures were — 

Montpellier, June 20 100'9 degrees. 

Orange, July 19 100-9 

Vendome, June 15 97'0 

Tours, June 96-8 

Clermont 96'4: 

Lille, June 15 95-9 

London, June 16 94*8 

Paris, Junes 89-6 

The drought was very great throughout nearly all France in toe 
spring and part of the summer, and was very inimical to the rearing of 
stock ; during June the sky was remarkably clear, but in July and Au- 
gust some rain fell, at least in the north, so that the meadows that had 
been scorched up, owing to a want of moisture dating from the year be- 
fore, partially recovered themselves. The harvest, which terminated on 
the 1st of July in the south, and the 1st of August in the north, was an 
average crop in respect to quantity, and a rather more than average one 
in respect to quality. The vintage, begun in Burgundy on the 18th of 
September, yielded a remarkable crop, both in respect to quantity and 

During recent years I must mention 1865 and 1868 as having been 
marked by a long series of hot days. The former, as is well known, 
was very favorable for the vintage. 

1865. — The mean monthly temperatures at the Paris Observatory 
were — 

January 38*5 degrees. 

February 36-1 

March 36-0 

April 60-4 

May 61-3 

June 64 "4 

July 67'8 degrees. 

August 63-9 

September 66*6 

October 54-0 

November 46 '4 

December 36'1 

The extreme heat in Paris was 91°-9 on the 6th of July. The aver- 
age of the three summer months was 65°*3. Adding to them Septem- 
ber, the average of the four months was 65° '5; an average that rarely 




lasts SO long. The mean of the year was 52°-5, being l°-2 above the 
averao-e. The month of January was relatively warm. In April, after 
the 4th, the weather was exceptionally fine, and the thermometer read- 
ino-s were very high: from the 8th the temperature was that of June. 
In May and June. the temperatures were above their normal points. 
July and August were cold. In September the temperature was higher 
than in August. October and November were warm. The highest 
temperatures were — 

Nimes, July 5 100'2 degrees. 

Nice, July 10 95-5 " 

Perpignan, July 4 95'4 " 

Aix, August 28 94-5 " 

Montpellier, July 26 93-2 " 

The mean monthly temperatures at the Paris Observatory 

were — 

January 32 "0 degrees, 

February 41-7 " 

March 44-6 " 

April 50-9 " 

May 64-2 " 

June 64-4 " 

July 70-2 degr 

August 65 "7 

September 63-7 

October 50-9 

November , 40"8 

December 47'5 

The maximum temperature in Paris w\as 93°"2 on the 22d of July. The 
average of the three summer months was 66°"9. This summer is nota- 
ble in the annals of meteorology for its thermometrical elevation, and 
its combination of circumstances favorable to the crops, both as to their 
quantity and quality. The averages of the temperatures of May, June, 
and July were very high in the south. Thus at Tours the average of 
May was 65°-l; that of June, 67°-6 ; that of July, 71°-2. The highest 
temperatures observed in France are appended : 

Nimes, July 20 106*5 degrees. 

Perpignan, July 25 990 

Draguignan, July 24 98*4 

Montauban, July 20 98-1 

Toulouse, July 19 95-0 

Montpellier, July 20 94-3 

Aix, July 20 93-2 

The temperature rose higher in 1859, without giving so high an aver- 
age. This latter was due less to the height of the diurnal maxima than 
to that of the nocturnal minima. In fact, notwithstanding the almost 
uninterrupted serenity of the nights, the cold caused by nocturnal radi- 
ation was at no time very remarkable. Nearly every morning before 



sunrise a slight fog, indicating a somewhat elevated hygrometrical con- 
dition, covered the soil, moistening the plants, and moditying the effects 
of the great heat during the day. The vapor of water prevents the 
radiation of the obscure heat; the air which was lying over our part 
of the country, and the somewhat elevated hygrometrical condition of 
which increased the transparency for the stellar light, nullified the ef- 
fects of nocturnal radiation, which is so potent even in the tropical re- 
gions when it has only to traverse an air devoid of moisture. 

This remarkable summer affected the temperature of the soil to 
the depth of more than a yard. During the summers of 1864, 1865, 
1866, and 1867, the heat at the depth of 39 inches was 57°-7, 58°-5, 
57°-2, and 57°-6. In 1868 it was 60°-6, nearly 61°-0. 

Such are the memorahle summers of the present century. The follow- 
ing are the highest temperatures of the air {in the shade, and to the 
north) observed in France since they have been truthfully ascertained 
by the thermometer. I have recorded all those which have reached or 
exceeded 37°, and those only, except in the case of Paris, where there 
are several readings. The towns are given in the order of latitude from 
N. to S. 





Les Mesneux 

Metz ., 

Montmorency . . . , 











Seurre (Cote-d'Or) , 

















the Sea. 





August 10, 


" 18, 

" 4, 

" 4, 

" 18, 

" " 26, 

" 14, 

" 19, 

" 5,6, 

July 16, 


" 10, 

August 18, 

July 31, 






















98 ^6 

























La Rochelle 

Saint-Jean d'Angely. 




Joyeuse (Ardeche). ... 









Beziers , 

Sor^ze , 





43-57 N. 












the Sea. 
























" 23 






(July 30, 


21, 1777 
, 5, 1836 

-25, 1800 

11, 1793 
6, 1800 

23, 1822 

4, 1842 

9, 1849 

14, 1802 

16, 1803 

20, 1868 

18, 1782 

20, 1806 

31, 1753 


29, 1857 


12, 1824 
4, 1838 

29, 1857 










99 1 



r 100-6 











The greatest heat in the shade and in the north is 106^° for France 
(Orange July 9th, 1849, and Nimes July 20th, 1868); 96° for Great 
Britain; 102° for Holland and Belgium; 99^° for Denmark, Sweden, 
and Norway ; 102° for Russia ; 103° for Germany ; 105° for Greece ; 
104° for Italy ; 102° for Spain and Portugal. In non-European coun- 
tries, the highest temperatures, as given by Arago, are as follows : 

Tunis 112-5 ( 

ManUla 113-5 

Nubia 115-2 

Am-Dize (Egypt) 116-1 

Esne' (Africa) 117-3 

Bagdad (Asia) 120-0 

Near Suez (French Expedition to Egypt) 1265 

Near Port Macquarie (Archipelago) 129-0 

Near Syene (Africa) 129-2 

Murzouk (Africa) 133-4 

These are the maxima of the temperatures of ^Ae air in the shade. 

In presence of such elevations of temperature, it may be asked to 
what point human organism can support it without incurring the dan- 
ger of sudden death. The mean temperature of the human body is 
about 96° (it is easily ascertained by placing the bulb of a thermometer 
under the tongue). That of birds is higher, and reaches 111° with cer- 
tain kinds. That of fish is lower, and about 87°. 





Take, in the first place, this winter landscape which is represented in 
the preceding page. It is the same as that which we saw, full of color 
and movement, on a fine summer's day. It is now transformed be- 
neath the gray and sombre sky of winter. The green foliage has dis- 
appeared from the trees, the meadow is covered with a pall of snow, the 
rivulet is frozen over, and the laborer's cottage seems as lifeless as Na- 
ture herself. With the progressive decline of temperature the ther- 
mometer has fallen to 32°, a remarkable point, at which water ceases to 
preserve its liquid condition and becomes solid. It then may assume 
various forms, becoming either massive in the shape of ice, light in the 
shape of hoar-frost, or falling slowly as snow-flakes. It is, as a rule, in 
this latter form that winter begins to manifest itself, for snow is pro- 
duced as soon as the temperature is at or about 32°. If this tempera- 
ture extends from the clouds to the surface of the earth, the water 
reaches the ground as snow. If snow in falling has only a thin stratum 
of air above 32° to traverse, and if it be abundant, it still reaches the 
ground and preserves its consistency. This occurs sometimes in sum- 

Snow, in covering the earth as a carpet, forms at once a covering and 
a screen ; a screen, because, possessing but little conducting power, it 
obstructs the passage of heat from the earth, and thus prevents the 
earth from becoming as cold as the air. Snow also adds its influence 
in favor of the fertilizing of the soil. Like rain and mists, it contains a 
considerable proportion of ammonia, which exists in a volatile state in 
the atmosphere, and which it conveys to the soil, afterward preventing 
it from becoming volatile again, as is the case after rain, especially after 
warm rain. 

In the origin — that is to say, in the frozen clouds high up in the at- 
mosphere — the snow appears to be formed of very slender fibres of ice. 
When the small drops of water, which form mists and ordinary clouds. 


become congealed, it is probable that these drops do not preserve their 
spheroidal shape, but that they fall an instant and take the shape of a 
filament which freezes concurrently with its physical transformation. 
By virtue of the laws of crystallization, these small filaments of ice, be- 
come cohesive at angles of 60°, and form the figures which, though so 
numerous, still appertain to the same geometrical order. 

Glaisher, in his ascent of June 26, 1868, encountered at 13,000 feet 
an immense cloud of snow, extending to a thickness of nearly one mile. 
It was a truly wonderful sight. This snow was composed entirely 
of small and perfectly -formed crystals, of an extreme delicacy. The 
points were visible, separate from each other, following two systems of 
crystallization ; for the angular intervals were some at 60°, and others 
at 90°. 

The construction of snow-flakes has long attracted the attention of 
observers. Kepler speaks of their structure with admiration, and other 
natural philosophers have endeavored to determine their cause; but it 
is only since the laws of crystallization in general have been ascertain- 
ed that it has been possible to throw any light upon this subject. 

In a circle, of all the polygons which can be inscribed, there is but 
one whose sides are equal to its radius; that is, the regular hexagon, oi" 
figure with six sides. This hexagonal figure is traced upon the flowers 
of the field, and we meet with it also in the crystallization of ice and 
snow, in the analysis of all the forms presented to our notice. The 
tendency of ice to take a crystalline shape is made evident by the fern- 
like leaves noticeable on window-panes during winter, when water be- 
comes congealed upon them. 

The examination of the figures of snow leads to impressions not less 
marked as to the existence of geometry, Number and Beauty, in the 
works of nature. It is not merely a few ice-flowers, such as the above, 
which have been remarked and designed in the slender snow-flakes, but 
there are many hundred different kinds, all constructed upon the same 
fundamental ans;le of 60°. 

The snow sometimes falls in such compact flakes that behind the first 
planes it forms a white, cloudy veil, which hides the landscape. These 
heavy falls of snow are mostly met with upon the lofty table-lands of 
Asia or the Andes, where the caravans have often to encounter them. 
The routes soon become concealed beneath the pall that covers them ; 
it becomes difficult to find one's way; and just as, in the rarer falls of 
snow in our countries, travelers wander over St. Bernard, and even 


)bab]e that tbeae drops do not preeervf 
they fall an instant and take the sba) 
currently with its physical transforn 
^-'^talli'/ation, these small filaments of i. . * 

.1 form the figures which, though so 
o tlie same geometrical order. 
i.i f June 26, 1863, encountered at lOj'",-.; (tet, 

■' : ; extending to a thickness of nearly one mile, 

onderful sight. Thi vas composed entirely 

i^eifectly-formed crystals, of au extreme delicacy. The 
', separat'^' ^''' -^ '■-^' -^ -^'^r, following two syst^^''"" 
r the a'. -re some at 60°, and 

u of snovv-ilake,' 
- L. i , .:. . J ,^r speaks of their £... ..^....v. . .. ,. .,. . 
licit.:.'" ^>hi]o«onher? have endeavored to determine 
j,s o; of crystallization in general have 

ed til. jeea pi^ssible to throw any light upon this subject 

"i f ;ill the polygons which can be inscribed, there 

e equal to its radius; that is, the regular hexa^ 
tigur Kagonal figure is traced upon the 1^ 

"' : ' ' . • Uization of i(V: una 

•^r jjOtice. The 
' the fern- 
Mceable on window-ptinf.- 

marked as to the existen. umber and Beauty, 

works of nature. . merely -flowers, such as the above, 

which h:;ve br-'^ ;.> u and desii.'^ '-■■< .h the slender snow-flakes, bur 

there -iTP' mnv "^ fiiffcront kinds, all constructed npor> I'tie same 


.> va such compaC' 
.u :..,x... , -udy veil, whi-^' 
' pnow fire mostly met wit 
r.e concealed beri. .ot/eri] ■: 

..;..... ''nl one's way; a:. ;_.,,_ :j rarer Ihl 
'ur cour velers wander over St. Bernard, and ev< 

Fig. 52.-Snow Crystals. 


crowd, are in no danger when standing upon ice eleven or twelve inch- 
es thick. 

In very severe Eussian winters the ice in the rivers is more than one 
yard thick ; but in France it has never exceeded more than about two 
feet. Its power of resistance is so great that, in 1740, a large palace of 
ice was constructed at St. Petersburg, 55^ feet long, 17 feet wide, and 
21 feet high, the weight of the top and of the higher parts of the edifice 
being readily supported by the foundations. In front of the building 
■were placed six guns in ice, with their carriages made of the same ma- 
terial. They were made to fire ball ; and each piece pierced, at a dis- 
tance of sixty yards, a plank two inches in thickness. The guns were 
not more than four inches thick ; they were loaded with a quarter of a 
pound of powder, and not one of them burst. The Neva supplied the 
materials for this singular edifice. 

I have said that water when congealed increases in volume. One 
consequence and one proof of this expansion is the bursting of the ves- 
sels containing it — a fact which occurs all the more readily when the 
process of freezing is rapid and the vessel narrow in the neck. 

I will complete this chapter by a notice of some of the hardest ivinters 
upon record — considering those as hard winters in which the cold has 
been of sufficient length and severity to freeze certain sections of large 
rivers, such as the Seine, the Saone, and the Ehine — to congeal wine, 
to destroy the tissues of certain trees, and to be followed by very grave 
consequences for both the vegetable as for the animal world. 

The following among the remarkable winters are the severest during 
the last hundred years. Let me, in the first place, mention that the 
hardest winters of past centuries were those of 1544, 1608, and 1709, in 
which latter year the thermometer at the Paris Observatory fell as low 
as — 9°*6 Fahr. The winter of 1776 next comes as an exceptionally 
cold one. The Tiber, the Ehine, the Seine, and even the Ehone, rapid 
as it is, were nearly entirely frozen over. 

After 1776, we come to the winter of 1788-1789, precursor of the 
Eevolution. This was one of the severest and longest winters that have 
ever prevailed in Europe. In Paris the cold commenced on the 25th 
of November, and lasted, with the exception of Christmas-day, when it 
did not freeze, for fifty consecutive days. The thaw began on the 13th 
of January, and the snow was found to be twenty-six inches deep. In 
the great canal at Versailles, in the ponds and in several streams, the 
ice was two feet thick. The water also froze in several very deep wells, 



and wine became congealed in cellars. The Seine began to freeze as 
early as November 26th (1788), and for several days its course was im- 
peded, the breaking up of the ice not taking place until the 20th of 
January. The lowest temperature observed at Paris was — 7°*2 Fahr., 
on the 31st of December. The frost was equally severe in other parts of 
France and throughout Europe. The Ehone was quite frozen over at 
Lyons, the Garonne at Toulouse; and at Marseilles the sides of the docks 


Fig. 53.— Wiuter.— The Seine full of floating ice. 

were covered with ice. Upon the shores of the Atlantic the sea was 
frozen to a distance of several leagues. The ice upon the- Rhine was so 
thick that loaded wagons were able to cross it. The Elbe was covered 
with ice, and also bore up heavy carts. The harbor at Ostend was 
frozen so hard that people could cross it on horseback ; the sea was 
congealed to a distance of four leagues from the exterior fortifications, 
and no vessel could approach the harbor. The Thames was frozen as 
low as Gravesend, and during the Christmas holidays and the early 

236 ^^^ ATMOSPHERE. 

part of January the stream in the neighborhood of London was covered 
with shops. 

The following are the lowest temperatures that were noted in differ- 
ent places : 

Bale (Suisse), December 18 — 35'5 degrees. 

Bremen (Germany), December 16 — 32'1 " 

St. Alban's, December 31 -28-8 " 

Warsaw (Poland), December 18 -26-5 " 

Dresden (Germany), December 17 — 25*8 " 

Eosberg (Norway), December 29 — 24'3 " 

St. Petersburg, December 12 —23-1 " 

Berlin (Prussia), December 28 -19-8 " 

Strasbourg, December 31 — 15'3 " 

Tours, December 31 -130 " 

Lons-le-Saulnier, December 31 — 11*2 " 

Troyes, December 31 -10-8 " 

Orle'ans, December 31 — 8"5 " 

Lyons, December 31 — 7'4 " 

Rouen, December 30 -7*2 " 

Paris, December 31 - 7-2 " 

Grenoble, December 31 -6.2 " 

Angouleme, December 31 — 1'7 " 

Marseilles, December 31 + 1'4 " 

The cold of this winter was very fatal to men and animals, and in- 
jurious to vegetables. In the Toulouse district the bread was nearly 
everywhere frozen, and it was impossible to cut it until it had been laid 
before the fire. Several travelers perished in the snow ; at Lemberg, 
in Gralicia, thirty-seven persons were found dead in three days toward 
the end of December. The birds that belong to the extreme north 
were seen in several parts of France. Fish were killed in nearly all 
the ponds by the great depth to which the ice penetrated. 

1794r-95. — This was a remarkably long and severe winter throughout 
Europe. In Paris there were forty-two consecutive days' frost; and 
January 25th (1795) was the coldest day ever known, the thermometer 
falling to — 10°'3, or 42°'3 below the freezing-point of water. In Lon- 
don the minimum temperature, 8°"1, occurred upon the same day ; and 
at midnight, on the banks of the Ehone, near Geneva, it was 6°'8. 
The Maine, the Scheldt, the Ehine, and the Seine were so frozen over, 
that carriages and army corps crossed them in several places. The 
Thames was frozen over in the beginning of January, near Whitehall, 
in spite of the height of the tide. Pichegru, then in the north of Hol- 
land, sent detachments of cavalry and infantry about the 20th of Jan- 


uary, with orders to the former to cross the Texel, and to capture the 
enemy's vessels caught at anchor by the frost. The French horsemen 
crossed the plains of ice at full gallop, approached the vessels, called on 
them to surrender, captured them without a struggle, and took the crews 

1798-99. — This was a very cold winter all over Europe. In Paris 
there were thirty-two consecutive days' frost, and the Seine was com- 
pletely frozen from the 29th of December to the 19th of January, from 
the Pont de la Tournelle to beyond the Pont Eoyal, but not sufficient- 
ly so to admit of its being crossed on foot. The lowest temperature 
remarked was +0°'3, or 31°-7 below the freezing-point of water in 
Fahrenheit's scale, on December 10th, 1798. An Alpine eagle was 
shot at Chaillot. The Meuse, the Elbe, and the Ehine were frozen 
more completely than the Seine. Carriages crossed the Meuse ; at the 
Hague and at Rotterdam fairs were held upon the stream. A regiment 
of dragoons, starting from Mayence, crossed the Rhine upon the ice in- 
stead of by the bridge at Cassel, which it had been found necessary to 

1812-13. — This winter will ever be remembered for the terrible dis- 
asters which attended the retreat of the French army through Russia, 
after the capture and conflagration of Moscow. The frost set in early 
all over Europe. The retreat of the army began on the 18th of No- 
vember; Napoleon left the capital of the Muscovite Empire on the 
19th, and the evacuation of the city was complete on the 23d. The 
army marched toward Smolensk, the snow falling without intermis- 
sion. The cold became very intense after the 7th of November, and 
on the 9th the thermometer marked 5°-0 (Fahr.). On the 17th the tem- 
perature fell to — 15°'2 (Fahr.) according to Larry, who had a thermom- 
eter suspended from his button-hole. The army corps commanded by 
Ney escaped from the Russian troops, by whom it was surrounded, ac- 
cording to Arago, by crossing the Dnieper, which was frozen over, on 
the night of the 18th-19th of November. The day before some Russian 
troops, with their artillery, had crossed the Dwina upon the ice. The 
cold diminished, and a thaw began on the 24th, but did not last ; so that 
from the 26th to the 29th, during the fatal passage of the Berezina, 
the water contained numerous blocks of ice without offering a passage 
at any part to the troops. The cold soon set in again with fresh in- 
tensity; the thermometer fell again to — 13°'0 (Fahr.) on the 30th of 
November; to —22° (Fahr.) on December 3d ; and to —35° on the 6th 


at Molodeczno, the day after Napoleon left Smorgoni, and published 
the bulletin (No. 29) which informed France of a part of the disasters 
incurred during this terrible campaign. 

1819-20. — This was also a very severe winter throughout Europe, 
although the extreme cold did not last so long. In Paris there were 
forty-seven days' frost, nineteen of which were consecutive, from the 
30th of December, 1818, to January 17th, 1819. The minimum tem- 
perature occurred on the 11th of January, viz., — 14°-3. The Seine was 
entirely frozen over from the 12th to the 19th of January. The Saone, 
the Ehone, the Rhine, the Danube, the Garonne, the Thames, the La- 
goons of Venice, and the Sound, were so far frozen that it was possible 
to walk upon the ice. The lowest temperatures observed in different 
towns are as follows : 

St. Petersburg, Januarj' 18 —25*6 degrees. 

Berlin, January 10 — 11"9 

Maestricht, January 10 — 2'7 

Strasbourg, January 15 — 1"8 

Commerc}^ (Meuse), January 12 — l"8 

Marseilles, January 12 + 0*5 

Metz, January 10 + 2'7 

Mons, January 11-15 + 3'9 

Paris, Januaiy 11 + G"3 

In France the intensity of the cold was heralded by the passage along 
the coast of the Pas de Calais of a great number of birds coming from 
the farthest regions of the north, by wild swans and ducks of variegated 
plumage. Several travelers perished of cold ; among others a farmer 
near Arras, a gamekeeper near Nogent (Haute Marne), a man and a 
woman in the Cote d'Or, two travelers at Breuil, on the Meuse, a woman 
and a child on the road from Etain to Verdun, six persons near Chateau 
Salins (Meurthe), and two little Savoyards on the road from Clermont 
to Chalons-sur-Saone. In the experiments made at the Metz School 
of Artillery on the 10th of January, to ascertain how iron resisted low 
temperatures, several soldiers had their hands or their ears frozen. 

1829-80.— This was the earliest and longest winter of the first part 
of the nineteenth century ; its duration was especially injurious to agri- 
culture in southern countries. The cold, without being extremely rig- 
orous, extended all over Europe; a great number of rivers were con- 
gealed, and the thaw was accompanied by disastrous inundations; many 
men and animals perished, and field labor was for a long time inter- 
rupted. The following are the principal temperatures observed: 


St. Petersburg, December 10 —26-5 degr 

Mulhouse, February 3 — 18'6 

Bale, February 3 -16-6 

Nancy, February 3 , — 15'3 

%inal, February 3 —14-1 

Aurillac, December 27 — 10'5 

Strasbourg, February 3 — 10"1 

Berlin, December 23 — 5'8 

Metz, January 31 — 4-9 

Pan, December 27 + O'S 

Paris, January 17 + 1"0 

In Switzerland the winter was severe in the great altitudes. At 
Freiburg there were one hundred and eighteen days' frost, sixty-nine 
of which were consecutive, and the minimum was — 1°-3 Fahr. In 
the plains, at Yverdun, among other places, the effects of radiation 
were felt very intensely : the thermometer fell in a few hours from 
+ 14° to —4°. The snow termed polar snow, the crystallization of 
which is very close, and which is peculiar to very low temperatures, 
also fell there. 

The length of time during which the Seine w^as frozen and its subse- 
quent thaw, excited public curiosity to the highest degree. The river 
remained frozen from December 28th to the 26th of January; that is, 
for twenty-nine days, on the first occasion. It was frozen over after- 
ward from the 5th to the 10th of February, making in all thirty-four 
days, or as long as was the case in 1763. It was frozen over at Havre 
from the 27th of December, and a fair was established upon the ice at 
Rouen on the 18th of January. On the 25th of January, after six days' 
thaw, the ice from Corbeil and Melun blocked up the bridge at Choisy, 
forming a wall sixteen and a half feet high. 

1840-1841. — During this winter there were fifty-nine days' frost, 
twenty-seven of them consecutive in Paris. The cold began on the 5th 
of December, and lasted, with an intermissiop from the 1st to the 3d of 
January, until the 10th of that latter month. There was another frost 
from the 30th of January to the 10th of February. On the 3d of Feb- 
ruary the thermometer still marked 16°'2 Fahr. From the 16th of De- 
cember the Seine was full of blocks of ice, and one of the arches of the 
Pont Royal was obstructed. Upon the evening of the same day the 
current was stopped at the Pont d'Austerlitz, and was frozen from Pont 
Marie to Charenton. The next day it was frozen at the bridge of Notre- 
Dame, and on the 18th people crossed from Bercy to the railway sta- 


tion. In several places the blocks of ice forced together were as much 
as seven to eight feet thick. On the loth of December the ashes of 
Napoleon, brought back from St. Helena, entered Paris by the Arc de 
Triomphe. The thermometer, in places exposed to nocturnal radiation, 
had that day marked +6°*8 Fahr. An immense crowd, the National 
Guard of Paris and its suburbs, and numerous regiments, lined the 
Champs Elysees from the early morning until two in the afternoon. 
Every one suffered severely from the cold. Soldiers and workmen, 
hoping to obtain warmth by drinking brandy, were seized by the cold, 
and dropped down dead of congestion. Several persons perished, vic- 
tims of their curiosity : having climbed up into the trees to see the pro- 
cession, their extremities, benumbed by the cold, failed to support them, 
and they were killed by the fall. I append some of the temperatures 
noted during this winter. 

Mount St. Bernard, January 22 — 9'9 degrees. 

Geneva, January 10 +0*0 " 

Metz, December 17 +4*5 " 

Paris, December 17 +8-2 " 

Paris, January 8 +8*4 " 

1853-54. — This was a severe winter in the temperate regions of 
Europe. It lasted from November to March, and caused several rivers 
to be frozen over. The cold was intense in many places, yet it proved 
rather beneficial to agriculture than otherwise. 

The principal temperatures were as follows: 

Clermont, December 26 .' — 4*0 degi-ees. 

Chalons sur Mame, December 26 — 4'0 " 

Lille, December 26 -0-4 " • 

Kehl, December 26 + O'S " 

Metz, December 27 + O'S " 

Brussels, December 26 + 3'0 " 

Lyons, December 30 + 5-7 " 

Paris, December 30 + 6*8 

Bordeaux, December 30 +14"0 

The next winter was also severe, especially in Southern Russia, Den- 
mark, England, and France, and was of unusual length. The frosts 
commenced as early as October in the east of France, and lasted until 
the 28th of April. The Loire was blocked with ice on the 17th of 
January, and its course was arrested the next day. The Seine, though 
full of blocks of ice on the 19th of January, was not frozen over. The 
Rhone was impeded on the 20th, and the Saone on the same day. The 


Rhine was completely frozen over at Manheim on the 24th, and people 
crossed it on foot. The appended table gives the lowest temperatures : 

Vendome, January 20 — 0'4 deg! 

Clermont, January 21 + 1"4: 

Brussels, February 2 + 1'9 

Turin, January 24 + 2-3 

Metz, January 29 + 3-2 

Strasbourg, January 29 + 3'2 

Montpellier, January 21 + 3"2 

Lille, February 2 + 7-2 

Paris, January 21 +11*7 

Toulouse, January 20 +12-7 

The winter of 1857-1858 was the type of the average severity of a 
winter in the temperate zone. The Seine contained blocks of ice on the 
5th of January, and the small arm of the stream by the Cite was cover- 
ed with ice on the 6th. The Loire, the Cher, the Nievre, the Rhone, 
the Saone, and the Dordogne were stopped in several places. The 
Danube and the Russian ports in the Black Sea were frozen in January. 
The lowest temperatures were : 

Le Puy, January 25 + 6'1 degrees. 

Clermont, January 7 + 6"8 " 

Bourg, January 29 + D'S " 

Vendome, January 6 +12'2 " 

Lille, January 7 —14-0 " 

Paris, January 7 — 15'8 " 

The winter of 1864-1865 was more severe. The Seine was frozen 
over at Paris, and people crossed it by the Pont des Arts. The ex- 
treme temperatures were : 

Haparanda, February 7 — 28*1 degrees. 

St. Petersburg, February 9 —19-8 " 

Riga, February 4 — 14'4 " 

Berne, February 14 -f 5'0 " 

Dunkirk, February 15 +10-4 " 

Strasbourg, February 11 -fl2"2 " 

Lastly, the winter of 1870-1871 will also be classed among severe 
winters, because of the extreme cold in December and January (not- 
withstanding the mild weather of February), and also because of the 
fatal influence which the cold exercised upon the public health at the 
close of the war with Germany. The great equatorial current, which 
generally extends to Norway, stopped this year at Spain and Portugal, 



the prevailing wind being from the north. On the 5th of December 
there was a temperature of 21° Fahr.; and on the 8th, at Montpellier, 
the thermometer stood at 17° -6 Fahr. A second period of cold set in 
on the 22d of December, lasting until the 5th of January. In Paris the 
Seine was blocked with ice, and seemed likely to become frozen over. 
On the 2'4th there were 21°-6 of frost; and at Montpellier, on the 31st, 
28°"8. It is well known that many of the outposts around Paris, and 
several of the wounded who had been lying for fifteen hours upon the 
field, were found frozen to death. From the 9th to the 15th of January 
a third period of cold set in, the thermometer marking 4-17°"6 Fahr. at 
Paris, and 4-8°*6 Fahr. at Montpellier. The most curious fact was that 
the cold was greater in the south than in the north of France. At Brus- 
sels the minima were +ll°'l in December, and +8°'2 Fahr. in Janu- 
ary. There were forty days' frost at Montpellier, forty-two in Paris, 
and forty-seven at Brussels, during these two months. Finally, the 
winter average (December, January, and February) is 35°'2 in Paris, 
whereas the general average is 37°"9. In the north of Europe this was 
also a very hard winter, though the cold set in at a different time from 
what it did in France. There were 40° of fro^ at Copenhagen on the 
12th of February, or the temperature was — 7°"6 Fahr. By the docu- 
ments which M. Renou has furnished me with for France, I discover a 
minimum of — 9°'4 Fahr. at Perigueux, and of —13° Fahr. at Moulins! 
I find by the documents supplied me by Mr. Grlaisher, that he also con- 
siders the winter of 1870-1871 as appertaining to the class of winters 
memorable for their severity. 

For the Seine to freeze in Paris there must be a temperature about 
4-16° Fahr., lasting several days. We have seen above how this is 
brought about. Since the beginning of the century it has been entire- 
ly frozen over eleven times: January, 1803; December, 1812; January, 
1820, 1821, 1823, 1829, 1830, and 1838 ; in December, 1840 ; in Janu- 
ary, 1854; and in January, 1865. 

M. Renou has noticed that the severest winters seem to recur about 
every forty years: 1709, 1749 (less severe), 1789, 1830, 1870. 

The following are the lowest temperatures observed in France since 
they have been carefully noted by the thermometer. They are in- 
scribed, like the previous list of the highest temperatures, in geographic- 
al order from north to south. I have taken all those that have reach- 
ed 20° of frost, and only those, except in the case of Paris, where there 
are several means of comparison. 













































































3 13 





























Clermont (Oise) 

Les Mesneux 



Chalons-sur-Marne , 






















Lons-le-Saulnier. . . . 






Grande-Chartreuse . 










48-35 N. 











January 28, 1776 

December 30, 1788 

February 27, 1776 

January 28, 1776 

December 31, 1788 

January 29, 1776 

December 30, 1788 

" 26, 1853 

January 19, 1855 

" 31, 1830 

" 1795 

(December, 1788 

\ " 26, 1853 

" 27, 1853 

"January 25, 1795 

" 13, 1709 

December 31, 1788 

February 6, 1665 

<! January 22, 1716 

" 29, 1776 

December 30, 1783 

January 20, 1838 

" ' 17, 1830 

December, 1788 

30, 1788 

(February 1, 1776 

t " 3, 1830 

(December 31, 1788 

trebruary 3, 1830 

December 31, 1788 

" 1788 

" 31, 1788 

" 31, 1788 

February 3, 1830 

December 19, 1788 

" 18, 1788 

" 31, 1788 

(January, 1784 

(February 3, 1830 

December 31, 1788 

" 31, 1788 

February 1, 1776 

December, 1788 

January, 1789 

(December 31, 1788 

t " 14, 1846 

j " 31, 1788 

(January 16, 1838 

December, 1788 

(December 31, 1788 


22, 1870 
31, 1788 
31, 1788 
16, 1838 

December 30, 1788 
February, 1776 
December, 1870 
December 27, 1829 


- 5-1 

- 4-5 

- 5-1 

- 7-4 

- 8-5 

- 7-2 

- 4-0 

- 4-4 

- 4-9 

- 4-0 

- 5-1 

- 4-0 

- 7-2 


- 9-6 

- 7-2 

- 6-2 

- 3-5 

- 2-4 

- 2-2 

+ 1-0 

- 6-7 

- 7-2 

- 8-7 




- 7-4 

- 4-0 

- 9-4 





- 8-5 

- 8-3 


- 8-5 


- 4-0 


- 9-4 





- 4-0 

- 8-7 


- 5-1 


- 7-4 

- 4-0 


- 6-9 

- 9-4 


The greatest coW yet experienced has been —24° in France; —5° in 


England ; —12° in Holland and Belgium ; —67° in Denmark, Sweden, 
and Norway ; —46° in Eussia ; —32° in Germany ; zero in Italy ; —10° 
in Spain and Portugal. As to other countries, not European, more ob- 
servations must be taken before one can speak with any degree of cer- 
tainty upon the point. It is, nevertheless, certain that at Fort Eeliance, 
in British North America, there have been —70° of cold, and at Semi- 
palatinsk —76°. Quicksilver freezes at —39°. There are inhabited 
points of the globe where it remains congealed for several months of the 
year — on Melville Island, for instance. Captain Parry, moreover, as- 
serts that a person sufficiently wrapped up may safely expose himself 
to the open air in —60°, or 82° below freezing-point of water — that is, 
if there is no wind. In this latter event the skin is rapidly affected. 
Frozen mercury looks like lead, but it is not so hard, is more fragile, 
and less coherent. If touched it burns like hot iron. Small statu- 
ettes can be made with it which melt when the temperature is higher 
than -39°. 

Such are the greatest frosts that have been experienced. If they are 
compared with the extremes of heat noticed in the previous chapter 
(165° upon the surface of the soil of Africa), it will be seen that the ex- 
tremes of temperature upon the globe may attain a scale of nearly 240 




If two lines parallel to the equator be traced upon the globe, at the 
distance of 23° 28' in each hemisphere, they will mark two circles be- 
tween which the sun is seen to pass across the zenith at certain epochs 
of the year; these are the tropics. That of the northern hemisphere is 
known as the Tropic of Cancer, because, during the summer solstice, the 
sun passes at its zenith and is in the zodiacal sign of Cancer. That of 
the southern hemisphere is known as the Tropic of Capricorn, because 
the sun passes at its zenith during the winter solstice in the zodiacal 
sign of Capricorn. The zone included between these two circles is the 
hottest part of the earth, inasmuch as it comprises the places over which 
the sun rises to its greatest altitudes; it is termed the torrid or inter- 
tropical zone. 

If two other circles, distant 23° 28' from the pole, or at ^Q° 32' 
from the equator, be drawn upon this same terrestrial globe, they will 
mark the points below which the sun may remain for several days to- 
gether, and above which it remains at its least altitudes ; these are the 
polar circles. During one half of the year the sun rises spirally above 
them to the height of 23° 28', and during the other half descends below 
them to the same amount. Between these two zones is the temperate- 
zone, in respect to which the sun rises and sets each day, without ever 
reaching so high as the zenith, attaining an increasing elevation, and 
giving a greater length of day, so far as our hemisphere is concerned, 
from the solstice of December to the solstice of June, corresponding with 
which there is an inverse rate of progress in the other hemisphere. 

The two glacial zones form 82 thousandths of the surface of the 
earth; the two temperate zones represent 520 thousandths; and, finally, 
the torrid zone, composed of the two regions comprised between the 
tropics and the equator, is 398 thousandths of the whole surface of our 

The length of the longest and the shortest days, in the different lati- 



tudes of our hemisphere, from the equator to the polar circles, gives 
the followina: scale : 


Names of Places. 

Longest Day. 

Shortest Day. 

hrs. min. 

hrs. min. 



12 17 

11 43 

12 35 

11 25 

12 53 

11 7 

13 13 

10 47 

13 34 

10 26 

13 56 

10 4 

14 22 

9 38 

14 51 

9 9 

15 26 

8 34 

16 9 

7 51 

17 7 

6 63 

18 30 

5 30 

21 9 

2 51 






(Gondar, Madras) 

(St. Louis) 

(Mexico, Bombay) 




(Madrid, Naples) 

(Bordeaux, Turin) 

(Dieppe, Frankfort) 

(Edinburgh, Copenhagen).... 
(St. Petersburg, Christiania). 


(Polar Circle) 

It is of course the same in the southern hemisphere. Beyond the polar 
circles the length of the day varies from to 24 hours in that part of 
the year during which the sun rises or sets. The number of days dur- 
ing which the sun is constantly above or constantly below the horizon 
in different latitudes, from Q&° 32' to 90°, is given in the following table, 
the phenomena being just the reverse for the two glacial zones : 


The Sun does not set 
in the Northern Hemi- 
sphere nor rise in the 
Southern during 

The Sun does not rise 
in the Southern Hemi- 
sphere nor set in the 
Northern during 






















In this theory of the climates we suppose the sun to be reduced to a 
point at its centre ; we have, moreover, neglected to take into account 
the phenomena of 'starlight produced by the refraction of light. As the 
diameter of the sun is about 32', the latitude at which it would disap- 
pear altogether must be placed at 16' farther back. The refraction, too, 
raising it by 33' at the horizon, the absolute polar circles must be also 
placed that distance farther back. Lastly, night is not complete until 
the sun has descended to about 18° below the horizon ; and this cir- 
cumstance must also be taken into account, whence it results that near 
the poles day is hardly ever extinct, and night, in the absolute sense of 
the term, almost unknown. 



The seasons are exactly opposite in the two hemispheres, as we have 
already stated ; they are indeed neither more nor less than the intervals 
of time which the earth takes to traverse the four parts of its orbit com- 
prised between the equinoxes and the solstices. In consequence of the 
eccentricity of the earth's orbit, and by virtue of the law of superficies., 
the lengths of the seasons differ ; they are represented by the following 
figures, which show that the sun is, in the course of each year, about 
eight days longer in the northern hemisphere than in the southern 
hemisphere : 




Autumn (September 22 to December 21) 





Winter (December 21 to March 21) 

Sojourn of the sun in the southern hemisphere... 

Spring (March 21 to June 21) 

Summer (June 21 to September 22) 

Sojourn of the sun in the northern hemisphere... 










The sun being, in fact, the source of heat for the surface of the earth, 
it follows that the hottest countries are those over which it remains 
the longest, and upon which it darts its rays the most vertically, that 
is, the regions situated along the equator, and upon each side of it, as 
far as the tropics. Thus, these warm regions are known by the generic 
appellation of " the torrid zones." In proportion as one recedes from 
the equator and approaches the poles, it is seen that the sun attains a 
lesser elevation, and that for six months the nights are longer than the 
days; these are the temperate regions, where the seasons lend a far 
greater variety to the products of nature, but where the mean of the 
annual temperature gradually diminishes according to the diminution 
in the apparent height of the sun at noon. Lastly, when we pass be- 
yond ^Q° of latitude, the glacial polar region is reached over which the 
sun, even on the finest days, scarcely rises sufficiently high to melt the 
eternal ice subsisting in these regions. 

It is superfluous for me to mention that the south pole is cold like . 
the north pole, nothwithstanding the idea attaching to this direction. 
Some few poets still talk of traveling 

"Du pole brulant jusqu'au pole nord;" 

but such metaphors are no longer admissible. The equator is to the 
south of our position, and the winds blowing from there toward us are 
hot. The equator is to the north of the other hemisphere, and the 


winds which reach it from the equator are also hot, though they come 
from the north. In respect to meteorological direction, as in regard to 
the seasons, the inhabitants of Australia, the Cape of Good Hope, Cape 
Horn, Buenos Ayres, and Santiago feel and speak just contrary to what 
we do. 

Latitude — that is, the angle at which the solar rays reach the surface 
of the ground — being the great cause of the succession of climates from 
the equator to the poles, the diminution would be progressive and reg- 
ular if the earth was a globe, perfectly regular in shape, instead of being 
divided into earth and water, and broken by mountains, table-lands, 
and valleys. The quantity of heat, estimated at 1000, for instance, at 
the equator, would follow a constantly descending scale, marking 923 
at each of the tropics, 720 at the latitude of Paris, and 500 at the polar 
circle. But the earth is not a smooth and undisturbed sphere, and rev- 
olutions more or less harmonious are constantlv occurrinsr. 

"We shall see in this work that the atmosphere is in a perpetual state 
of circulation, and that there are general winds which periodically trav- 
erse different countries of the globe. These regular currents modify 
the normal distribution of climates. Thus the trade-winds, which es- 
tablish a double current between the equator and the poles, temper at 
once the cold of the high latitudes over which they pass and the heat 
of the tropical regions, heating the former and cooling the latter. 

A second cause is added to this by which the temperature along the 
same circles of latitude is varied. The globe is divided into oceans and 
continents. Water has a greater capacity for heat than land, whence 
it results that the sea is cooler than the land in summer, and warmer in 
winter. The winds which blow from the sea prevent the coast-line from 
being as cold as the country farther inland. As the south-west wind is 
that which blows oftenest, the western coasts of Spain, France, Scotland, 
and Norway are warmer than the inland country in the same latitudes. 
The great marine current known as the Gulf Stream adds still further 
to this modifying cause. 

Water becomes less readily heated upon the surface than earthy mat- 
ter, because the latter has a specific heat much below that of water ; so 
that the quantity of solar heat required to raise its temperature by 10°, 
for instance, is much less than that which would raise the temperature 
of a liquid surface the same number of degrees. 

It must, furthermore, be mentioned that the solar rays which become 
absorbed in a very thin layer of earth penetrate, at least in part, to a 


very considerable depth in the water; that at sea, especiallj, they do 
not become totally extinguished until they have reached a depth of 
more than three hundred yards ; so that the heat arising from absorp- 
tion, instead of being concentrated upon the surface, is spread over a 
great mass of water, and must be the less in proportion as this mass is 

Evaporation, which is, as we have seen, a very great cause of cold, is 
the greater according as this phenomenon occurs upon a larger scale. 
And, where the liquid mass continually furnishes the means of evapo- 
ration, there exists a cause of cold which does not exist at all, or in a 
much less degree, upon dry land. From these three causes (specific 
heat, diathermacy, and evaporation) it follows that water, and the at- 
mosphere which is in contact with it, must be less heated than conti- 
nental regions situated in the same latitude. In winter, on the other 
hand, it is warmer, and that for a reason which it is easy to compre- 

I have already stated that the superficial moleculae, rendered cold by 
their radiation toward the cold regions of space, are precipitated toward 
the bottom by reason of the excess of their specific weight ; consequent- 
ly the surface of the sea must preserve a higher temperature than that 
of the surface of continents, since in this case the superficial moleculae 
that have become cold do not plunge into the ground. 

These consequences, deduced from a minute examination of the ac- 
tion of the solar rays upon a liquid and a continental surface, are con- 
firmed by observation. 

Thus, at Bordeaux, the mean winter temperature is 42°-8 : whereas, 
in the same latitude, the temperature of the Atlantic never falls below 

In latitude 50° the ocean has never been known to be less than 48°"2. 

The mass of observations collected show that, in the northern hemi- 
sphere and in the temperate zone, the mean temperature of an island 
situated in the midst of the Atlantic would be higher than the mean 
temperature of a spot similarly situated upon the main-land — that the 
winter would be warmer and the summer cooler. This has been espe- 
cially remarked in the Island of Madeira. 

The sea serves to equalize the temperatures. Hence there is an im- 
portant difference between the climate of islands or coast-lands pecuhar 
to all continents that abound in gulfs and peninsulas, and the climate 
of the interior of a great and compact mass of dry land. In the interi- 


or of Asia, at Tobolsk, Barnaul-upon-Obi, and Irkoutsk, the summer is 
the same as at Berlin, Miinster, and Cherbourg ; but these summers are 
followed by winters when the temperature is as low as — 0°4 or —4°. 
During the summer months the thermometer will remain for weeks to- 
gether at 86° or 88°. These continental climates have been very appro- 
priately termed excessive by Buffon, and the inhabitants of countries in 
which they prevail seem to be condemned, like the spirits alluded to by 

"A sofFerir tormenti e caldi e gieli," 

The climate of Ireland, of Jersey and Guernsey, of the peninsula of 
Brittany, of the coasts of Normandy, and the south of England, coun- 
tries in which the winters are mild and the summers cool, contrasts 
very strikingly with the continental climate of the interior of Eastern 
Europe. In the north-east of Ireland (54°*66), in the same latitude as 
Konigsberg, the myrtle grows in the open ground just as it does in 
Portugal. The temperature of the month of August in Hungary is 
69°*8; in Dublin (upon the same isothermal line of 49°) it is 61 de- 
grees at most. The mean temperature of winter descends to 36°"3 at 
Buda. In Dublin, where the annual temperature is only 49°, that of 
the winter is, nevertheless, 7°'7 above the freezing-point, or nearly four 
degrees higher than at Milan, Pavia, Padua, and all Lombardy, where 
the mean heat of the year reaches 65°. In the Orkney Islands, at 
Stromness, a little to the south of Stockholm (there is not one degree 
difference in latitude), the mean winter temperature is 7°, or higher 
than that of London or Paris. Stranger still, the inland waters of the 
Faroe Islands never freeze, situated in 62° of north latitude, beneath 
the mild influences of the west wind and the sea. Upon the coast of 
Devonshire, one part of which has been termed the Montpellier of the 
North, because of the mildness of its climate, the Agave Mexicana has 
been known to flower when planted in the open air, and orange-trees 
trained upon a wall to bear fruit, though only scantily protected by a 
thin matting. There, as at Penzance, Gosport, Cherbourg, and the 
coast of Normandy, the mean temperature of winter is 42°, being but 
18°'5 below that of Montpellier and Florence. 

The mean annual temperature of London, as deduced by Glaisher 
from one hundred years' observations (1771-1870), is 48°"6. The mean 
summer temperature is 60° '2, and that of winter 88°. The winter, 
therefore, is warmer at London than at Paris, and the summer and the 
year cooler. Although Cherbourg is one degree of latitude north of 



Paris, its mean temperature is, notwithstanding, higher, being 62°-3, 
while that of Paris is only 51° -3. The difference between the winter 
climates of the two towns is much greater, since the winter mean is 
43°-7 at Cherbourg, and 37°-8 at Paris. Thus fig-trees, laurels, and 
myrtles, which would perish in the neighborhood of Paris, are found to 
flourish in the former place. The enormous fig-trees which grow at 
Eoscofi", in Brittany, are almost equal to those of Smyrna. 

Jan.. 1 Feb. | Mar ] h.VV. 1 Mav | Junc.l JtUv. | Auq, | St-pt. | Ocl, | Wov. | Dec. , 

Jf 31 \ 

28 1 











_ 1 






































\ .■•. 









/ ,v 








' / 

































, ,■ 



> . 
















Fig. 54.— Comparative temperatures of the European capitals of Rome, London, Paris, Vienna, 

St. Petersburg. 

These comparisons are sufficient evidence as to how the same mean 
annual temperature may be distributed in many different proportions 
over the various seasons, and how great an influence these diverse 
modes of distribution of heat may exercise in the course of the year 
upon vegetation, agriculture, the ripening of fruit, and the comfort of 

The same relations of climate, which are remarked as existing be- 
tween the peninsula of Brittany and the rest of France, the mass of 
which is more compact, and where the summers are hotter, and the 
winters colder, are reproduced to a certain extent as between Europe 
and the continent of Asia. Europe owes the mildness of its climate to 
its abundantly indented configuration, to the ocean which washes its 
western shores, to the sea which separates it from the polar regions, 
and, above all, to the existence and to the geographical situation of the 
African continent, the intertropical regions of which radiate excessively 


and cause the ascent of an immense current of hot air, whereas the re- 
gions situated to the south of Asia are, for the most part, oceanic. 

Europe would become colder if Africa were submerged — if the fabled 
Atlantides, emerging from the ocean, were to unite Europe to America 
— if the warm waters of the Gulf Stream did not flow into the northern 
seas, or if a new land, upheaved by volcanic agency, were to become in- 
serted between the Scandinavian Peninsula and Spitzbergen. In pro- 
portion as we advance from west to east, along the same latitude, in 
France, Germany, Poland, Eussia, and as far as the Ural Mountains, we 
find that the mean annual temperatures follow a uniformly descending 
scale. But as we penetrate inland, the form of the continent becomes 
more and more compact, its breadth increases, the influence of the 
sea diminishes, and that of the west wind becomes less perceptible. 
Therein lies the chief cause of the progressive decline in the tem- 

The mean temperature of the equator is 81°-5. Owing to the causes 
which I have specified, and to the absence of vegetation, that of inland 
Africa is 86 degrees with the thermometer placed in the shade and pro- 
tected from hot winds ; but there are points at which the action of the 
burning breeze, and the absence of clouds, combine in producing an in- 
tolerable degree of heat. Thus, in the interior of Abyssinia, and in the 
neighborhood of the Eed Sea, it is by no means rare to meet with a 
summer temperature of 118 to 122 degrees in the shade. That of the 
soil is higher still. In the afternoon the valleys of Abyssinia are regu- 
lar furnaces; M. d'Abbadie having observed the temperature of the soil 
at 160 degrees nearly, while Colonels Ferret and Galinier met with a 
temperature of 167 degrees. The air is stagnant in the midst of all this 
heat, and there is no refreshing breeze. The air in the depths of these 
ravines is often mephitic, and to repose therein after or before the rainy 
season is fatal. It is necessary at that period to travel by night, as 
plains have to be crossed which afford no place of shelter. 

"Sometimes in crossing these deserts," says M. d'Abbadie, "one is 
assailed by the Icarif^ a sort of aerial hurricane, a phantom of burning 
dust which appears upon the horizon, and seems to grow in size as it 
approaches. The wind which wafts it blows like a hurricane; men and 
animals are obhged to turn their backs, and are enveloped in a dry and 
black cloud, which covers them as with a hideous mantle. Fortunately 
this storm of fire lasts only a few minutes, and after it has passed the 
intense heat which is peculiar to these regions is felt as a relief. 


"At Other times one is overtaken by the simoom (the poison), a wind 
of flame which begins to blow without any premonitory sign. The 
camel is then seen to lay his head upon the ground, seeking coolness 
from it, though it is itself like a furnace. The hardiest of the natives 
are struck down in despair. The prostration is so great, that I was 
myself unable to lift a small thermometer placed within my reach, in 
order to ascertain, at all events, the temperature of this remarkable 
wind. Its duration was five minutes : it causes death when it continues 
for a quarter of an hour. 

" If one happens to meet with a small stream in these regions, it soon 
disappears, absorbed by the sand. These miniature oases, composed of 
a few trees and some grass, are very rare. 

" These same valleys are the theatre of a very extraordinary phenom- 
enon ; a sudden irruption of water, which at certain periods of the year 
causes inundations, to which those occurring in European countries are 
trifling. And, strange to relate, they take place during the summer. 

" One may be traveling with a fall sense of security, when a native, 
hearing a strange and distant noise, commences to shout at the top of 
his voice, 'The torrent!' and climbs as fast as possible to the nearest 
elevated point. In a few seconds the hollow of the valley is hidden by 
a deep body of water, which carries with it trees, rocks, and even wild 
animals. These torrents, formed in an instant, disappear in the course 
of the same day, and leave no trace of their passage, save debris and 
muddy deposits. 

"How is this strange phenomenon to be explained? The barrenness 
of the mountains accounts for these sudden down-pours. From the 
hollow of the ravine in which the traveler is journeying, he is unable 
to see the narrow clouds which suddenly dissolve into water, with an 
abundance unknown, save in tropical regions. There is very little soil, 
and still fewer roots of trees, to absorb this sudden rain, which conse- 
quently runs off at once, leaping from rock to rock, as down a roof, 
flowing from each minor valley into the principal ravine, and there 
forming a stream which, though short-lived, is of mighty dimensions." 

M. dAbbadie further relates how he was just too late on one occasion 
to witness, in all its grandeur, one of these sudden inundations. He 
found a native, who was regarding the wet ground with the air of one 
who had been stunned. "Peace be with you," said M. d'Abbadie; 
" what news? Where are your arms? Surely you can not be without 
your lance and shield?" "Peace be with you," replied the African, 


" the torrent has carried off my lance, my shield, my camel, and all my 
fortune, my wife and my children." 

It will thus be seen that various causes influence the climates of dif- 
ferent countries of the globe ; and it would involve great errors were 
we to take into account the distance from the equator only in calcula- 
ting the decrease of the temperature toward the poles. We have seen 
that the average temperature of the equator is 81°'5 ; the mean temper- 
ature of Paris is 51°-3 ; that of regions within the polar circle about 5°. 

To establish a correct table of the distribution of temperature over 
the surface of the earth, Humboldt marked upon a map all the points 
at which reliable thermometrical observations had been taken, noted 
the degrees recorded, and then traced lines passing respectively through 
all the places where the mean temperature was the same. These he 
termed isothermal lines (from '/(roe, equal, and Oepiuog, heat). During 
the fifty years since, observations have been multiplied and the maps 
made more perfect. 

We see in diagrams of isothermal lines, or lines of equal temperature 
running along the western shores of Europe, that the line of 50°, for in- 
stance, touches the fortieth degree of latitude south-west of New York, 
and reaches as far as 55° near England ; so that Dublin and London 
have nearly the same mean temperature as New York, although they 
are situated much farther north ; the same temperature then falls again 
toward the south, passing to Vienna, Astrakhan, and Pekin, and de- 
scending even below the fortieth parallel of latitude. The greatest heat 
line, called the thermic equator, is nearly entirely to the north of the 
equator, and its temperature varies, according to situation, from 81° to 
86°. Within the polar regions the mean temperature of different places 
decreases to as much as 1°, which has, as yet, scarcely been traced, in 
consequence of the difficulty of traveling in these inhospitable regions. 

Humboldt has pointed out that, notwithstanding these great differ- 
ences, the mean temperature decreases almost uniformly at the rate of 
nearly a degree of the thermometer to each degree of latitude. But as, 
on the other hand, the heat diminishes by 1° for an increase of height 
of about three hundred feet, it follows that an elevation of about one 
hundred yards produces the same effect upon the temperature of the 
year as an approach of 1° of latitude toward the north. Thus, the mean 
annual temperature of the monastery of Mount St. Bernard, situated at 
a height of 8173 feet, in latitude 45° 50', is the same as that of low 
ground in 75° 50' latitude. By studying the distribution of heat over 


the surface of the globe, and by tracing a system of isothermal lines, 
Humboldt demonstrated the causes which raise the temperature of a 
particular spot, and those which lower it. The augmenting causes are 
as follows : 

The proximity of the ocean on the west in the temperate zone. The 
configuration peculiar to continents which are cut up into numerous 
peninsulas. The Mediterranean, and the gulfs penetrating far inland. 
The direction, that is to say, the position of a country in respect to a 
sea free from ice, which extends beyond the polar circle, or in regard 
to a continent of considerable extent, situated upon the same meridian, 
at the equator, or at least in the interior of the tropical zone. The 
south-westerly direction of the prevailing winds in the case of the west- 
ern fringe of a continent situated in the temperate zone, the chains of 
mountains acting as a rampart and a protection against the winds which 
blow from colder countries. The scarcity of pieces of water, the sur- 
face of which is covered with ice during the spring, and up to the be- 
ginning of summer. The absence of forests on a dry and sandy soil, the 
constant serenity of the sky during the summer months, and, lastly, the 
near neighborhood of a maritime current, whose waters are warmer than 
those of the surrounding ocean. 

The decreasing causes are : the height above the level of the sea of a 
region which does not possess extensive table-land. The distance of 
the sea to the west and the south in our hemisphere. The compact 
shape of a continent, upon the coasts of which there are no bays ; a 
great extent of land toward the pole, and toward the regions of eternal 
frost, except in the case of there being between the land and this region 
a sea that is free of ice during the winter; a geographical position such 
that the tropical regions in the same longitude are covered by the sea : 
in other words, the absence of any tropical land upon the meridian of 
the country whose climate is being studied ; a chain of mountains which 
by its shape or direction prevents the access of warm winds, or indeed 
the presence of isolated peaks, because in both these cases currents of 
cold air make their way down the slopes ; forests of great extent, for 
these prevent the solar rays from acting upon the soil ; the leaves cause 
the evaporation of large quantities of water, by reason of their organic 
activity, and increase the superficies liable to be rendered cold by radia- 
tion. The forests act, therefore, in three ways : by their shade, by their 
evaporation, and by their radiation. The numerous pieces of water 
which, in the north, are regular receptacles of ice up to the middle of 


summer. A cloudy sky in summer, because it intercepts a portion of 
the sun's rays ; a very clear sky in winter, because it facilitates the ra- 
diation of the heat. 

To the general conditions of climates must be added the influence 
which local circumstances may have upon the state of the temperature. 
It is far more difficult than is generally supposed to ascertain exactly the 
temperature of a given spot upon the surface of the globe, and especially 
of an inhabited spot ; for ten thermometers, identically the same, and 
carefully compared, will not mark the same point at the same moment 
in ten different streets of the same town. The principal remark to be 
made in reference to this is, that in consequence of the radiation of 
dwelling-houses, and on account of the obstacles which an agglomera- 
tion of buildings puts in the way of free circulation of air, the tempera- 
ture of large towns is always less marked and higher than that of the 
country around them. Howard showed that the mean temperature of 
London exceeds by 2° that of the surrounding district.* The ther- 
mometers of the Paris Observatory are never so high as those in the 
heart of the city, but are higher than those placed in the open air in the 
field adjoining. Every one has noticed that it is cooler in summer and 
warmer in winter in the narrow streets of old Paris than upon the mod- 
ern squares and boulevards. There is frequently a difference of sev- 
eral degrees. Even in the open country, at the same altitude and in the 
same frontage, the temperature differs according to the distance from 
woods. These latter act upon the temperature of the air, which is low- 
er in than it is outside them. The mean maxima outside of woods are 
higher than inside. The mean temperature of summer is also higher in 
the former case than in the latter. These facts are clearly shown, ac- 
cording to MM. Becquerel, by the results of more than fourteen thou- 
sand observations made during the last few years. 

The hours of maxima and minima are not the same inside the trees 
(even when they stand alone) as in the open air. They vary according 
to the kind and the diameter of the tree. The variations of temperature 
among the leaves are about the same as those in the surrounding air; 
in the young branches they occur more slowly, and so on to the trunk, 
where they are very gradual. I am excluding from the question the 

[* In a paper published in the Philosophical Transactions, Part II., for 1850, I proved that 
those parts of London situated near the river Thames iire somewhat warmer upon the whole 
year than the country, but that those parts of London which are situated at some distance 
from the river do not enjoy higher temperatures than those due to their latitudes. — Ed.] 


special heat of the trees which results from the various reactions which 
take place in their tissues, and that which they derive from liquids ab- 
sorbed by the roots, because it is very slight as compared to that caused 
by solar or nocturnal radiation, as is proved by the maxima and mini- 
ma of temperature which correspond with the maxima and minima of 
the air, though occurring at different hours of the day. This special 
heat of trees plays an important part in winter, by preventing a decline 
in temperature which would be fatal to them. In a tree twenty to 
twenty-four inches in diameter, the maximum temperature occurs in 
summer about 10 or 11 p.m. ; in winter, toward 6 p.m. ; whereas in the 
air it is at 2 or 8 p.m., according to the season. From this difference be- 
tween the hours of maxima, it results, as experience has proved, that the 
temperature of the air may be lowered by some cause, such as the pas- 
sage of a cloud, a change in the direction of the wind, etc., and yet rise 
in the interior of trees, because of the heat acquired by the outer sur- 
face, which is transmitted slowly to the inner portion of the tree, owing 
to its non-conductibility. The abundance of forests and moisture tend 
to lower the temperature, while clearing away timber and causing 
dryness of atmosphere produce a contrary effect ; the difference in 
some cases for the mean temperature of the year being as much as four 
degrees nearly. 

The numerous observations taken by MM. Becquerel in the Loiret 
have been particularized by them as follows : 

1st, in summer, the mean temperatures of the air outside of woods 
are higher than they are inside. 

2d, in winter, the reverse is the case. 

8d, the difference between the mean annual temperature of the air at 
several miles from woodland and that inside a wood is about 3°. 

The mean temperatures of the air in summer being about 2\° higher 
outside than they are inside a wood, and the reverse being the case in 
winter, it follows that the woodland climate is not so extreme as that of 
the open plain ; it partakes, therefore, of the nature of a warm climate 
in respect to temperature. Local conditions modify more or less the 
general type of climates. The greatest local action is always exercised 
by unevenness of soil. The mountain chains divide the surface of the 
earth into large basins, into deep, hollow, or circular valleys. These 
valleys, often shut in, as between ramparts, individualize local climates 
(in Greece, for instance, and in part of Asia Minor), and place them in 
special conditions in reference to heat, moisture, the transparency of the 



air, and the frequency of winds and storms. After having studied the 
general condition of climates, and before coming to the poles in the 
course of this short geographical review, it is interesting to endeavor to 
form a correct idea of the extreme differences of temperature through- 
out the world. 

In no place of the globe, and in no season, has the thermometer at an 
elevation of two or three yards above the soil, and sheltered, reached 

In the open sea the temperature of the air has never exceeded 86°. 

The most extreme degree of cold ever recorded upon a thermometer 
suspended in the air is 72° below zero. 

The extreme difference in the temperatures of the atmospheric air is, 
therefore, 207°. 

Comparing together the most extreme temperatures recorded, Arago 
constructed the remarkable table appended. The places are given ac- 
cording to their decrease in latitude. 














Melville Island 

74-47 N. 




+ 60-1 
-f 70-0 

+ 72-5 




Port Felix 





+ 68-9 

- 4 





+ 83-7 







-1- 86-0 







-f 95-0 




St. Petersburg 



+ 90-0 







+ 86-0 







+ 99-5 







-f 95-0 






+ 96-8 
+ 94-1 









+ 95-0 






+ 102-7 







+ 95-0 

+ 5 











+ 95-0 
+ 99-5 

- 6 







+■ 96-1 
+ 92-3 


- 3 







+ 100-4 

— 7 






+ 100-6 

- 6 







+ 104-0 
-f 96-6 





Munich (1765 feet) 



+ 95-0 







+ 93-2 
-f- 96-3 


- 8 







-f 100-4 






+ 96-1 

- 4 





+ 99-5 



Lausanne (1732 feet) 



+ 95-0 

- 4 





+ 97-2 







St. Bernard (8172 feet). 
Gr.-Chartr. (6660 feet). 



Le Puy (2493 feet) 















Pulo-Penang Island 

Quito (9540 feet) 

St. Louis de Marana.... 
Isle of Bourbon 






















14 S. 



















+ 106 
+ 104 
+ 101 
+ 98 
+ 101 
+ 100 
+ 104 
+ 109 
+ 101 
+ 103 






- 6 


- 3 


+ 15 
+ 19 
+ 23 
+ 3 
+ 24 
+ 27 
+ 60 
+ 75 
+ 75 
+ 42 
+ 75 
+ 60 








Generally speaking, the differences between the highest and the low- 
est temperatures are less the farther one travels from the pole toward 
the equator. 

Let us now deal with the limits of climates, the extremity of the 
world, the icy regions of the poles. 

In the neighborhood of the Polar Circle the sea becomes frozen, and 
assumes a special character. This phenomenon seems to increase as the 
water gets less briny, and as the rotatory movement declines in rapidi- 
ty. Even in 50° of latitude pieces of floating ice are met with in the 
sea. These have become detached from some more northern region, 
and carried off by the currents which run from the poles to the equa- 
tor. At 55° it is by no means rare to find the sea-shore strewn with 
ice. At 60° the gulfs and the inland seas are often frozen all over. At 
70° the floating blocks of ice become very numerous and very large, 
forming sometimes regular islands as much as a half league in diame- 
ter. Finally, at 80°, there is found, as a rule, fixed ice — that is, ice 
which has become accumulated and bound together. 

These solitary regions offer a striking spectacle. 

The polar ices are tinted with the brightest hues, and seem like 
blocks of precious stones, forming vast plains and lofty mountains. 

The fields of ice are often composed of extensive plains, perfectly 
level, without either fissure, hollow, or elevation. Scoresby saw one 


of these floating-fields upon which a carriage might have been driven 
for thirty-five leagues without the slightest interruption. When these 
masses meet, the report of the shock is like a clap of thunder. 

The mountains of floating ice, as seen for the first time by the navi- 
o-ator who has made his way into the polar regions, present a striking 
spectacle. Dr. Hayes, in his voyage to the Arctic seas (1860), has con- 
veyed to his readers the first impression produced by the sight of them. 
He says : 

"We met our first iceberg the day before we reached the Polar 
Circle. Hearing the sea breaking furiously against the mass, as yet 
concealed by the mist, the helmsman was upon the point of crying 
out, ' Land ahead !' But almost immediately the formidable colossus 
emerged from the fog, bearing down upon us, terrible and threatening; 
we hastened to get out of its way. It formed an irregular pyramid, 
about three hundred feet wide and one hundred and fifty feet high ; its 
summit was half hid in the mist; but the latter suddenly lifting, ex- 
posed to our gaze a dazzling peak, around which were folded light va- 
pors. There was something very striking in the indifference of this gi- 
ant, which the waves caressed in vain, while it passed on its way, deaf 
to their charms. 

" In Davis Straits we had to pass many cruel hours ; and upon one 
occasion I thought that our last moment had arrived. We were run- 
ning against the wind, all sails bent and a heavy swell on, when the 
bows gave way, and all the sails fell on to the deck, nothing remaining 
save the chief sail, which was flapping violently against the mast; and 
it was only owing to a miracle of firmness on the part of the helmsman 
that we escaped complete shipwreck. 

" For most of us Greenland was still a kind of myth ; for some days 
we had been following the coast-line : beyond the appearance of Disco, 
the clouds and fog had kept it constantly hidden from our gaze. But 
suddenly it emerged from its mantle of mist, and stood out before us in 
all its splendor; its extensive valleys, its noble mountains, its abrupt 
and sombre rocks adding to its terrible desolation. 

" In proportion as the fog and mist rolled slowly over the surface of 
the blue waters, the mountains of ice succeeded each other and defiled 
before us like the fantastic palaces in a fairy tale. Forgetting that they 
would come spontaneously toward the region, they seemed to us to be 
attracted by an invisible hand into this enchanted land." 

The ice met with on the coasts of Spitzbergen and Greenland is, gen- 


erally, from twenty to twenty-five feet thick, often forming immense 
plains, the limits of which can not be seen from the topmast of a vessel : 
these are called the ice-fields. They may be estimated as having an ex- 
tent of three hundred to four hundred square leagues. An ice-field 
sometimes presents an entirely level surface ; at others, it is rough and 
uneven, with, at intervals, columns twenty or thirty feet high. These 
columns give it a very picturesque aspect, and which are sometimes of 
a topaz blue tint, sometimes covered with thick snow. 

The undulations of the water, the movement of the waves, or some 
other potent cause, break up a field of ice in a moment, and reduce it 
into fragments of 1000 or 2000 square feet. These fragments, becom- 
ing separated, come into collision and disperse ; but sometimes they are 
carried off by a rapid current. In that case, if they meet a current run- 
ning in the opposite direction, which is floating away large masses of 
ice from some other field, these mountains meet with a terrible shock. 

The icebergs, lifted up out of the water, fall the one on to the other, 
become covered with fragments more or less voluminous, and thus com- 
pose regular mountains, with ravines and indentations, which rise from 
thirty to fifty feet above the water. The part out of the water is, as a 
rule, in regard to the portion submerged as one to four; consequently 
the total height of these mountains is from 180 to 200 feet. Sometimes, 
too, icebergs 100 to 130 feet long, which are very heavy at their two 
extremities, sink so deep into the water that a vessel may pass over 
them. But in this case the crew is exposed to the most fearful risk, as 
the least shock, the least cause, may disturb the equilibrium which 
keeps the mass submerged, and if that cause occurs, the iceberg rises 
suddenly and hurls the vessel into the air or, at any rate, shivers it into 

In Baffin's Bay there are mountains of ice much higher than in the 
seas of Greenland, some having been found to measure 100 to 130 feet 
out of the water, which is equivalent to a total height of 660 feet. It is 
supposed that these fearful masses are formed upon the coasts where 
they shut in the valleys which abut upon the sea, and that they can be- 
come detached. In summer-time the waters flow from their summits 
and form immense cascades, which are sometimes overtaken by frost. 
This is a majestic spectacle, but it must be witnessed from a distance, as 
all of a sudden these columns suspended in the air will snap short and 
fall into the sea. 

Scoresby often saw ice form upon the open sea at twenty leagues from 



the shore. As soon as the first embryos of the crystals become percep- 
tible, the sea gets calm, just as if oil had been poured over its surface. 
These crystals soon attain three or four inches in size, and it is then 
that they begin to agglomerate if the cold continues, forming a sheet of 
ice which soon attains a thickness of from eight inches to a foot. 

In these countries the density of sea-water is 1-026; when still, it 
freezes at 28° 4. The water which has been concentrated by the frost 
may attain a density of 1*104, in which case it will only congeal at 14°; 
and it is well known that water saturated with salt will not solidify till 
the temperature is less than 5°. 

Fig. 55 The last human dwelling-places. Eequimanx of the Polar Regions. 

These desolate regions where mercury freezes in the open air, are 
nevertheless inhabited by the Esquimaux, who are the remotest inhabit- 
ants to the north, living as they do in the 79th degree of latitude. Dr. 
Kane visited, in 1853, two of their villages upon the Greenland coast of 
Smith's Straits, at 11° from the pole. These villages are called Etah and 
Peterovik, and the capital of the country is Upernavik, which was vis- 
ited in 1861 by Dr. Hayes, An idea of the place in which these people 
(from whom America is descended) dwell may be gathered from Fig. 
55. The huts are constructed upon landings with blocks of snow cut 
into the shape of domes. The entrance is by a circular and very small 

Fig. 56.— Ice at the Pole. 


opening, and light is admitted by means of a small window, in which a 
diaphanous piece of snow serves the purpose of a pane of glass. 

The point nearest to the pole as yet reached is six degrees and a 
quarter (lat. 83° 45'), which is only about 170 leagues from it. Parry 
and Sir James Eoss approached thus far in 1826. The ill-fated Frank- 
lin did not pass beyond 77°. Dr. Hayes navigated in the polar sea as 
far as 81° 40' in the month of May, 1861. 

Let us conclude this general view of climate by remarking that the 
last isothermal line, clearly established by observations, is that of 4-5°, 
which descends to the north of America, re-ascends to the north of Baf- 
fin's Bay, crosses the 80th degree of latitude, afterward extends to de- 
gree 70, and even to degree 65. This line forms two bends, in each of 
which there is recorded an increase of cold. It is not at the pole itself 
that the mean temperature is lowest, but on either side of it. There are 
thus what may be termed two poles of cold, one situated to the north 
of the Asiatic continent, not far from the archipelago known as New Si- 
beria, where the mean temperature is -|-l°-4; the other to the north of 
the American continent, in the western isles of the Polar Archipelago, 
and its temperature appears to be — 2°"2. It is probable that two anal- 
ogous poles of cold exist as well in the frozen Antarctic Ocean. As to 
the North Pole itself, the early calculations of Plana, the mathematician, 
of the geometer Lambert, and of the astronomer Halley, as well as those 
of my regretted friend Gustave Lambert, establish conclusively the fact 
that the cold is much less intense there. As to our pole (taking into 
account refraction), the sun rises in the beginning of March, mounts 
slowly, skimming the horizon, and follows a spiral line which takes a 
greater elevation each successive day. It does not again set until the 
end of September. On the 21st of June it attains its greatest elevation. 
The maximum of heat prevails in July and August. From these calcu- 
lations and the direct observations of navigators who have penetrated 
the nearest, it follows that the sea is not frozen at the North Pole itself. 






We now come to the study of the great currents of the atmosphere, 
which are themselves the incessant manifestation of the sun's action 
upon our planet. Without the wind the atmosphere would remain 
motionless about the globe ; heavy, cold, deadened, enveloping the earth 
in a regular pall, never agitated by a breath of air, a Receptacle for ev- 
ery kind of miasma — poisonous and deleterious. By its agency an im- 
mense circulation is established from one end of the world to the other, 
renewing all the strata, sweeping away unhealthy exhalations, substi- 
tuting for oppressive heat a refreshing coolness, or replacing the period 
of frost by the warmth of spring. 

What is windf In this section of our work, and in the succeeding 
one, which deals in clouds and rain, we take in hand the general data 
of meteorology ; for the currents of air on the one hand, and water on 
the other, cause the varying meteorological conditions of the seasons 
and of years. It is on this head particularly that we have an exact base 
for our knowledge, and that we are in a position to consider the general 
mechanism of this vast factory, which distributes benefits and disasters 
over the earth, and among the people which inhabit it. Meteorology 
will not be able to hold her own with her elder sister. Astronomy — that 
is, to be precise in respect to ascertained principles, and to enable science 
to announce the movements to the atmosphere, the winds, the rains, the 
droughts, and the tempests, as the latter announces the movement of the 
stars — until we are able to embrace, in one glance, the general circula- 
tion which is constantly going on all over the glcHbe, and which gives 
rise to divergencies which occur in different seasons and at different 

What is the wind? 

It is neither more nor less than a certain quantity of air set in motion 
by a change in the equilibrium of the atmosphere. The varying tempera- 
tures to which the different parts of the atmosphere are constantly ex- 

270 ^-^^ ATMOSPHERE. 

posed rarefy each of these parts in a different manner. When air is 
heated its weight diminishes, and it has a tendency to rise; whereas 
colder air becomes heavier, and flows to supply the place of the heated 
air, and, in its passage toward the re-establishment of an equilibrium, 
will cause a current of air which is termed wind^ and which will con- 
tinue till an equilibrium is restored. 

Let us suppose, for a moment, that the atmosphere is perfectly calm 
everywhere. A cloud passes over the sun, the air that is situated in 
a line with its passage is rendered cooler, undergoes condensation, and 
becomes denser ; this air seeks an equilibrium ; a primary movement 
will take place in the direction of the cloud, and here we have a current 
of fresh air, the tendency of which will be to occupy, as quickly as pos- 
sible, the place of the hotter and more dilated air which is next to it. 
Suppose that the sun, shining in a clear sky, remains motionless above 
our heads. The air situated immediately underneath will become heat- 
ed more rapidly than that which receives its rays obliquely. Becoming 
dilated, it will rise toward the less dense aerial regions, the air which is 
contiguous to it will force itself into its place, and thus another current 
of air is established. 

The great atmospheric currents, the winds, general and special, are 
nothing else than this unceasing pursuit toward an equilibrium which 
is perpetually being destroyed by the various influences of the sun. 
This will be seen by applying to the entire surface of the globe the in- 
stance cited above. In what way are two contiguous parts of the at- 
mosphere affected if they become heated in unequal proportions ? Near 
the equator, the sun, as its rays reach the earth in a perpendicular or 
nearly vertical direction, causes a temperature which is constantly high- 
er than at other points of the globe. It follows from this that two in- 
ferior currents must flow from the two hemispheres toward the equator. 

The air, which is very heated in the equatorial zone, rises in a mass 
toward the higher regions of the atmosphere. Having reached an ele- 
vation of several miles (but which we are unable to calculate exactly), 
the ascending mass»breaks into two, which pass away in the direction 
of the two poles. 

This ascensional movement thus produced gives rise to a rush of air 
from the two sides of the torrid zones, and two other masses, skimming 
the surface of the ground, make their way from the temperate regions 
toward this line. Thus we discover all over the earth a double aerial 


Let US first take the northern circuit. A current of air, starting from 
ihe tropical regions, proceeds toward the equator. Situated in the lower 
regions of the atmosphere, and upon the surface of the globe, this cur- 
rent comes directly beneath our observation, and it constitutes the trade- 
luinds of the northern hemisphere. When within a short distance of 
the equator, a distance which varies with the seasons, it suddenly rises, 
and, when it has reached a certain level, takes a directly horizontal 
march toward the pole, gradually descending toward the surface as its 
distance from the equator increases. Maury termed this kind of cur- 
rent the upper anti-trade-wind. 

If it stopped there the current would not be complete ; the trade- 
winds and the anti-trade-winds, connected with each other by the as- 
cending branch of the equatorial region, are not, as yet, united on the 
northern side. If the earth were motionless, and the whole of its sur- 
face received light at the same time ; if, moreover, its surface was uni- 
versally homogeneous, the meeting of the two hrizontal branches would, 
no doubt, take place toward the north, as it does toward the south, ex- 
cepting, of course, the reversal of the direction of the movement. The 
upper anti-trade-wind would incline toward the ground, so as to join 
the trade-wind, and the circulation of the atmosphere would be almost 
comprised within heights of an inconsiderable elevation. Let us re- 
mark, however, that as the first origin of the movement is at the equa- 
tor, the movement will be regular there, like the cause which produces 
it. The trade-winds and the anti-trade-winds will themselves partici- 
pate of this regularity in the neighborhood of the equinoctial line ; but 
the farther one recedes from this line the less directly will the motive 
force act. The descending mass will, therefore, be more difi'use, less 
compact, and less fixed in its quantity than the ascending mass. Its 
mean position will depend upon the mean activity of the equatorial 
draught, and upon the height to which the trade-winds reach. This 
height is itself dependent upon the law of the decrease in the tempera- 
ture, according to the altitude. It may vary with the seasons, and has 
probably not been the same in all ages of the world. 

The southern circuit is rather more extensive than the northern; it 
encroaches upon the northern hemisphere, upon the surface of the At- 
lantic, and in summer this encroachment is more marked than is the 
case in winter. 

Circulation, regular as it may be, can not take place in the midst of 
an atmosphere always in motion like ours, without reacting upon the 



part which is not directly comprehended in the movement. The de- 
crease of the temperature extends also toward the poles, and atmos- 
pheric movements are the forced consequences in these high latitudes. 
Two leading circumstances cause the aerial currents to travel out of 
the limits comprised in the above circuits, and give rise to two second- 
ary circuits (N' and S'); these are the rotation of the earth on its axis 
around the sun, and the division of land and water over the globe. 


Sonth Pole 

Tlorlh Pole 

Pig. 57.— Section of the atmosphere, showing its general circulation. 

The earth turns upon its axis in the direction of west to east. In 
virtue of this rotation, every point of it completes a revolution in the 
same period of twenty-four hours ; but in this interval of time all parts 
do not traverse the same distance or move at the same rate of speed. 
At the equator the speed is about 416 leagues an hour ; in the latitude 
of Paris it is 273 ; at degree 56 it is 231 — as at Edinburgh, for instance; 
at the poles it is nothing. 

The air whicb seems to us to be in repose at Paris is, in reality, mov- 
ing there at the rate of 273 leagues an hour. Let us imagine this air 
transported to the latitude of 56° without any change in its velocity ; it 
will continue to travel 273 leagues per hour. As each point in latitude 
56° travels at 231 leagues per hour, the air will gain upon the ground, 
in an easterly direction, at the rate of forty-two leagues an hour ! which 
would constitute a hurricane. The reverse would be the case if a mass 
of air, relatively still, in parallel 56°, were suddenly transported into 


parallel 49°. This air would appear to us to be traveling from east to 
west at the rate of forty-two leagues per hour. 

In reality, these passages of air from one parallel to another always 
take place gradually, and, during their transition, resisting causes of va- 
rious' kinds tend to equalize their velocity. The lessened differences 
none the less continue in operation, and, as the size of the parallels of 
latitude diminishes the more rapidly on approaching the poles, the ef- 
fects pointed out above become more and more pronounced as they oc- 
cur in higher latitudes. Many tempests are derived from this cause. 

The influence of the earth's rotation upon the direction of the trade- 
winds is as follows : 

Take, first, the trade-winds of the northern circuit. We have sup- 
posed that they move from north to south toward the equator. During 
this movement they pass gradually by the parallels, whose diameters, 
and consequently whose speed, progressively increase. If their abso- 
lute velocity does not diminish, they will apparently move toward the 
west, and their seeming direction will be from north-east to south-west, 
which is, in fact, somewhere about the direction of the trade-winds in 
the northern hemisphere. A like result follows in the case of the 
southern trade-winds, which also seem to retrograde toward the west; 
but as these winds travel from south to north toward the equator, their 
apparent direction will be from the south-east toward the north-west, 
which is, in fact, the general direction of the trade-winds in the south- 
ern hemisphere. 

When the ascending mass, having reached a certain height, divides 

into two horizontal masses, which form the upper or anti-trade-winds, 

flowing from the equator toward the poles, and, little by little, travel 

past parallels the speed of which is successively less and less, they soon 

take an easterly bend in these parallels, and their apparent direction is 

toward the north-east. When they have arrived at a certain distance 

from the neighborhood of the tropics, they descend toward the earth ; 

then is reproduced the phenomenon noticeable in the ascending mass ; 

the anti-trade-winds find their way with the velocity which they have 

acquired andtheir easterly tendency ; the inclination of their speed in a 

vertical direction renders this speed less apparent, and we meet with 

two new regions in these latitudes, called tropical calms. In moving 

from the equator toward the North Pole, we thus encounter: 1st. The 

region of equatorial calms ; 2d. The north-easterly trade-winds ; 3d, The 

tropical calms; and 4th, beyond these, winds varying from south-west 


274 y^-E" ATMOSPHERE. 

to north-west. The same series is met with in the southern hemi- 

In a word, we find that there are in each hemisphere two circuits 
which have as a common basis the ascending equatorial mass. The 
first, a direct circuit, is generally limited to the intertropical regions ; the 
second, a derived circuit, is, in reality, only a prolonged arm of the first, 
and extends from the tropics to a varying distance from the poles. 
These two circuits are distinguished from each other by essential char- 
acteristics ensuing from their different positions in the atmosphere. 

The direct circuit spreads upward. While the trade-winds skim the 
ground, the anti-trade-winds circulate in very lofty regions of the air. 
The distance which separates these two currents, joined to the regulari- 
ty of their movements, prevents them from encroaching upon each oth- 
er or influencing each other's progress. This does not hold good of the 
derived circuit. The prolonged arm of the anti-trade- winds there be- 
comes superficial. It sweeps along the ground ; and so, also, does the 
returning current. Both, therefore, are upon the same level, simply 
contiguous, and separated only by the action of the earth's rotation. 
There are points at which these currents come together ; and their dif- 
ferent qualities cause numerous, and sometimes disastrous, atmospheric 
disturbances. Their beds get shifted over the surface of the globe, and 
the succession of one after another in the same place produces sudden 
variations in the state of the sky. To avoid confusion, the branch of 
the upper anti-trade-winds which is prolonged into the derived circuit 
is termed the equatorial current, and the back current in the same cir- 
cuit is called the polar current. 

This general circulation of the atmosphere is influenced to a certain 
extent by the seasons. 

At the end of our summer the regions about the North Pole have for 
several months had days without any nights; the temperature there 
has become perceptibly milder and the air rarefied. To days without 
nights soon succeed nights without days, accompanied by excessive 
cold ; the air becomes contracted, and draws in a fresh supply to fill up 
the vacancy caused by this contraction. Each of these changes in our 
hemisphere corresponds with an exactly reverse change in the other 
hemisphere ; there is, therefore, a general translation each year of the 
atmosphere of the northern hemisphere into the southern, and vice 

The rush of air toward the North Pole during winter is brought 


about bj the equatorial currents, which then acquire a very large vol- 
ume. The perturbations increase there in the same proportions: it is 
the season of tempests. As the sun makes its way back to us, and our 
atmosphere becomes heated and dilates, the equatorial current slackens 
its speed and reaches lower latitudes. On the other hand, the polar 
currents become more active ; but as they are diffused over the surface 
of Asia and of Europe, their speed is rarely very great, and summer is 
the calm season in our hemisphere. The atmospheric disturbances at 
this season never extend very far, and their local gravity is due to elec- 
trical phenomena of a special nature : it is the season of thunder-storms. 

The equatorial currents take, at their polar extremities, a direction 
parallel to the equator, and march from west to east. Notwithstanding 
their variations, both in volume and intensity, it is easy to understand 
that they cause the atmosphere at the poles to make a continuous rota- 
tory movement in the same direction as the earth. 

For many centuries the trade- winds were an enigma, both to meteor- 
ologists and to navigators. Halley and Hadley first suggested the ex- 
planation which has been developed, and which contemporary research 
has modified in the course of the last century. 

Between the two trade -winds there are two zones; these are the 
zones of equatorial calms. These calm regions occupy very difierent 
positions at the close of winter to what they do at the end of summer ; 
they follow, in fact, but at a distance, the progress of the sun between 
the tropics. They never cross the equator upon the surface of the At- 
lantic. In February and March, months when they approach nearest to 
it, the north-easterly trade-winds stop at about 4° north latitude ; in Au- 
gust and September, the months in which they are farthest away from 
it, the same trade-winds stop at about 11°. When a vessel sailing in 
the Atlantic approaches the equator, the crew begin to feel anxious, for 
they know that the favorable wind which has brought them thus far 
will gradually fail, and finally disappear altogether. The waters extend 
around them like a vast sheet of ice, and the ship is, so to speak, nail- 
ed to the limpid crystal. The solar rays fall vertically upon the deck. 

The sun which, twice in the course of the year, pours down its rays 
perpendicularly upon these regions, never recedes far enough for any 
thing like coolness to ensue. The heated atmosphere becomes so light 
that it is continually ascending. There evaporates also from the Atlan- 
tic and the Pacific Oceans an immense quantity of water, which becomes 
diffused and mixed with the heated air, and ascends with it ; but as the 


air ascends to the lofty regions it gradually cools, sometimes very sud- 
denly, so that a great part of the water which had accompanied it is 
transformed into drops. These sudden changes produce passing tem- 
pests, which are frequent in the equinoctial regions. 

We have seen that, as the wind approaches the temperate zones, upon 
which it will descend and become converted into surface currents, the 
upper current encounters strata of air, the speed of which in regard to 
the diurnal movement is at a minimum. It follows that the return of 
the trade-winds gives rise in the temperate zones to a wind which blows 
from south-west in the northern hemisphere, and from north-west in the 
southern hemisphere. Thus, in France, the wind blows oftener from 
the south-west than from any other direction. At the time of the dis- 
cussions upon the real movement of the earth, the followers of Coperni- 
cus adduced the trade-winds as a proof of the diurnal rotatory move- 
ment, from west to east. This was quite an illusion on their part. Car- 
ried by the movement of the globe, the observer would, had such been 
the case, have quitted the air of the atmosphere, which would, under 
those circumstances, have seemed to give rise to a wind blowing in a 
contrary direction, viz., from east to west. But we have seen that it is' 
the combination of different rates of speed, on the one hand, the strata 
of air which are displaced by the differences of temperature in the vari- 
ous parts of the globe ; and on the other hand, the atmospheric strata 
which are brought under the influence of the diurnal movement, which, 
in reality, produce the trade- winds. The theory of the motion of the 
earth does not require this pretended meteorological proof. 

The existence of the upper counter-current has been ascertained di- 
rectly by Captain Basil Hall, who observed that in the region of the 
trade-winds very high clouds are continually sailing in an opposite di- 
rection to that followed by the wind beneath. The same traveler re- 
marked upon the summit of Teneriffe in August, 1829, a south-westerly 
wind; that is to say, a wind of a diametrically opposite direction to the 
trade- wind which was blowing upon the surface of the ground. When 
Humboldt ascended the same mountain in 1799, a very strong westerly 
wind was blowing upon the peak. 

Another proof of the existence of this same counter-current of the 
trade- winds may be deduced from the fact of dust emitted by the vol- 
cano in St. Vincent Island falling upon Barbados. 

During the evening of April 30, 1812, explosions resembling the dis- 
charge of heavy pieces of artillery were audible at Barbados ; the gar- 


rison of the Chateau St. Anne remained under arms all night. On the 
following morning the horizon of the sea, to the east, was clear and 
well-defined, but just above it was seen a black cloud which already 
covered the rest of the sky, and which, soon after, spread over that part 
where the light of day was beginning to break. The obscurity became 
so intense that persons sitting in a room were unable to distinguish the 
window, and in the open air trees and houses, and even a white hand- 
kerchief held up at a distance of six inches before the eyes, became in- 
visible. This phenomenon was caused by the fall of a large quantity 
of volcanic ashes, emitted by a volcano in the Island of St. Vincent. 
This new kind of rain, and the profound obscurity which accompanied 
it, did not entirely cease until nearly one o'clock. The trees, whose 
timber bends readily, bent beneath its weight, and the crash of the 
limbs of other trees as they snapped off short was in striking contrast 
to the complete calm of the atmosphere; the sugar-canes were pros- 
trated upon the ground, and the whole island was covered with a layer 
of greenish ashes to a depth of one inch. 

St. Vincent is fifty miles nearly due west of the Barbados, and the 
volcano there had shot this immense mass of ashes to the height at 
which the upper current was traveling — -a current which was itself suffi- 
ciently strong to transport the mass. 

Halley was the first to afl&rm the existence of the upper trade-winds 
as a consequence of the ordinary trade-winds. Though he advanced no 
direct proof of the fact, he assured himself of its truth by the almost in- 
stantaneous rotation of the wind in opposite directions, when the polar 
limits of the trade-winds are passed. In his opinion, as in that of all 
meteorologists of the present day, the equatorial south-west current 
which prevails in the mean latitudes of our hemisphere is, in reality, 
only a continuation of part of our upper trade-winds on their return 

The higher branch of the intertropical circuit is, at its equatorial ori- 
gin, at such a height that it has been impossible to ascertain its exist- 
ence with precision, even by climbing the loftiest peaks of the Cordil- 
leras in the neighborhood of the region of calms. But, as this branch 
gradually descends toward the surface of the globe, in proportion as it 
approaches the tropics, and as, moreover, its course lies through colder 
regions, some few clouds appear in the air which it carries in its train. 
These serve as so many proofs of the direction which it takes. 

The existence of trade-winds was ascertained during the first voyage 


made by Christopher Columbus. The regular winds, which impelled 
that adventurous navigator along the new route by which he expected 
to reach India, excited the fears of his associates, who doubted the pos- 
sibility of getting back to Europe. Had Columbus, after the discovery 
of the New World which he alighted upon when he imagined that he 
had reached India, not taken pains to avoid the trade- winds, by steering 
to the north before he turned westward, he would assuredly never have 
found his way back to Spain. With his vessels both ill provided with 
food and defective in construction, he and his crews would have perish- 
ed of hunger in the vast regions of the trade-winds. It is upon the 
struggle between these two currents, upon the point at which the upper 
current descends to the surface, and upon their reciprocal mingling, that 
depend the most important of atmospheric variations, the changes of 
temperature in the strata of air, the precipitation of aqueous vapor, and 
even, as Dove has shown, the varying shape and form which clouds 
take. The shape of the clouds, which lends so much charm to our 
landscape, indicates to us what is going on in the higher regions of the 
atmosphere. When the air is calm the clouds delineate upon the sky 
on a warm summer day " the projected shape " of the soil, the caloric of 
which radiates freely toward space. 

In the great ocean and the Atlantic the trade- winds extend nearly to 
the tropics ; but in the Indian Ocean the presence of land prevents the 
regular or the trade- winds from setting in; whereas in the southern 
hemisphere, at a certain distance from land, the south-east trade- winds 
prevail almost uninterruptedly. In the northern hemisphere of the In- 
dian Ocean there prevails a south-west wind, blowing toward the penin- 
sula of Hindoostan, to the north of India and China, from April till 
October; and from October to April the prevailing wind is, on the 
contrary, from the north-east. These are the monsoons of the Indian 
Ocean, This word is derived from the Malay moussin., which signifies 
season. Thus, during the summer of our hemisphere, when the sun 
has a north declination, it is the south-west monsoon which prevails; 
whereas in our winter, when the sun has a south declination, the mon- 
soon is the north-east. These winds penetrate into the interior of con- 
tinents, where they are influenced by the shape of the land. The 
mountain chains generally tend to attract the gaseous masses in their 
direction. The explanation of these periodical winds is this : In Jan- 
uary the temperature of South Africa is at its maximum^ that of Asia at 
its minimum. The northern portion of the Indian Ocean is hotter than 


the continent, but not so hot as the southern part of the same ocean at 
an equal latitude. We find, then, in each hemisphere, easterly winds 
blowing toward the hottest points. From October to April the south- 
east trade- winds prevail in the southern hemisphere ; the north-east 
trade-winds are blowing in the northern hemisphere, and are termed 
the north-east monsoon. Between the two is the region of calms. 
When the sun advances toward the north, the temperature of the conti- 
nent and that of the sea become more or less equalized ; thus, about 
the period of the spring equinox, there are no prevailing winds in 
the northern hemisphere, but varying winds, which alternate between 
dead calms and hurricanes ; whereas the south-east monsoon prevails 
throughout the year in the southern hemisphere. As the north decli- 
nation of the sun increases, the temperature of Asia rises above that of 
the sea ; whereas it declines below it in New Holland and South Africa. 
The relative positions of the two continents, the differences in the tem- 
perature which are most marked, and the rotatory movement of the 
earth, thus create a current from the south-west — a monsoon which pre- 
vails from April to October. Thus, whereas in the southern hemi- 
sphere the trade-winds from the south-east prevail throughout the year, 
the north-east monsoon in winter and that from the south-west in sum- 
mer are met with to the north of the equator. 

Thus are indicated in a brief manner the general directions of these 
winds. So far back as any records exist, they facilitated the communi- 
cations which were then so frequent between India and Egypt. Upon 
the decadence of that empire these relations ceased, and the tradition of 
these winds was lost ; for, otherwise, Nearchus would not have been so 
long on his voyage from the mouths of the Indus to the extreme end 
of the Persian Gulf 

In many places periodical winds are met with which alternate with 
the seasons, and which are influenced by the shape of the coast-line ; 
thus, for instance, in Brazil there is a north-east monsoon in spring and 
a south-west monsoon in autumn. The Mediterranean has its mon- 
soons, known to the ancients, who indicated their sense of dependence 
upon the winds by the term etesian winds (from troq, year or season). 
To the south of the Mediterranean basin the vast desert of Sahara ex- 
tends. Devoid of water, made up merely of sand or conglomerated 
pebbles, it becomes very heated under the influence of an almost vertic- 
al sun ; whereas the Mediterranean preserves its ordinary temperature. 
Thus, in summer, the air rises above the desert of Sahara with great ra- 


pidity, and floats off mostly toward the north, while underneath are 
northerly winds which extend as far as Greece and Italy. In North 
Africa, at Cairo and Alexandria, there are none but northerly winds. 
All navigators are aware that in summer the voyage from Europe to 
Africa is effected more rapidly than the return passage. Thus, if we 
compare the half-duration of passages to and fro between Toulon and 
Algiers, it will be found that the return passage is one-fourth longer in 
the case of sailing-vessels, and one-tenth in the case of steamers. This 
fact can not be attributed to the currents, which are very trifling. Be- 
sides, the north coasts of the islands of Majorca and Minorca — that of 
the latter in particular — are swept by this same wind, which causes a 
perceptible stunting of vegetation there. These winds prevail at Al- 
giers, Toulon, and Marseilles. In winter, on the contrary, when the 
sand radiates considerably, the air of the desert is colder than that of 
the sea, and in Egypt there is a very cold south wind, though not so 
strong as the summer winds. — Kaemtz and Martin. 

To these periodical winds, to the trade-winds and the monsoons, we 
may add the breezes caused upon sea-coasts by the difference between 
the heat of the land and of the water. This, in the early part of the 
chapter, was pointed out as produced by solar heat, like the trade- 

Periodical and diurnal displacement of air takes place in mountain- 
ous regions. These consist in a breeze which creeps along the side of 
the mountain at night, and in an ascending breeze during the day. 
These movements of air vary according to the shape and aspect of the 

Of all the causes which are assigned to the winds, one of the most 
powerful is, beyond doubt, the condensation of vapor in the atmos- 
phere. Sometimes one inch of water will fall, in the course of an hour, 
over a wide tract of country, especially in the equatorial regions. 
Now, suppose this tract to be but a hundred square leagues in extent. 
If the vapor necessary for the production of a depth of one inch over a 
hundred square leagues were in an elastic condition in the air, and had 
only 50° temperature, it would occupy a space a hundred thousand 
times greater than in its liquid state ; that is to say, it would occupy a 
space of a hundred square leagues by 8860 feet in height. Such, there- 
fore, would be the dimensions of a void resulting from this condensa- 
tion. In reality, the vapor is not in an elastic but in a vesicular state, 
although, from the very fact of its remaining suspended in the atmos- 


phere, it is probably of less density than if it were in a liquid state, and 
its condensation into drops of rain also occasions an immense void, the 
filling of which must necessarily give rise to great atmospheric disturb- 

The constant circulation going on in the atmosphere renders impossi- 
ble the entire consumption of any of the substances necessary to main- 
tain the life of organized matter, such as oxygen, aqueous vapors, etc. ; 
and it also prevents any dangerous accumulation of deleterious matter, 
such as carbonic acid. The existence of animated nature is intimately 
connected with this circulation. These simple features do not, at first 
sight, seem to apply to the apparently capricious play of the weather, 
nor to delineate it in its true aspect or type of versatility and change- 
ableness. The weather is not less variable, especially in our climates, 
as we shall presently see. We may divide the surface of the globe 
into two unequal parts — the regions of fixed and variable weather. 
The state of the air may be predicted to the limit to which the trade- 
winds extend, and that for several years to come. The mean zone (in- 
cluded between 2° and 4° N. and S. latitudes) is that where throughout 
the whole year great heat and calms alternate with nocturnal rain-falls 
and tempests. Next to them, both north and south, is another zone 
(4° to 10° latitude), where similar weather occurs only in summer or in 
winter, and trade-winds render the sky clear. There is a third zone 
(10° to 28° K. latitude) where, in winter as in summer, the trade-winds 
do not usher in the slightest moisture, where years pass without the 
soil being refreshed by the least drop of rain. 

Finally, another zone, both north and south (from 20° to 30° latitude), 
which forms the limit of fixed weather; there the trade-winds cause the 
summer to be without rain and the winter to be mild and rainy, though 
the rain is never continuous. The approximate indication of the lati- 
tudes refers to the northern hemisphere and the Atlantic Ocean, the 
sole region where reliable observations have been collected. 

We now have to consider a zone of 2-4° latitude, where the meeting 
between the polar and the equatorial currents occasions a variable cli- 
mate, which only seems to us capricious and uncertain because the cir- 
cumstances influencing the predominance of one of the two currents 
in a given locality are so complicated that we have been unable to de- 
duce from observations a law by which these modifications can be clas- 
sified. If we study the question we find, as I have said, that there are 
in reality but two winds in the atmosphere ; that which blows from the 


poles toward the equator, and that wliich makes its way back from the 
equator to the poles. Let us now take a place situated in the region of 
variable weather (the latitudes of Paris, Vienna, or London, for instance), 
and further, let us admit that this place is just in the direction of the 
polar current When the north wind blows there, the cold becomes 
accentuated, the sky gets clear, even if the wind, deviating slightly from 
its direction, turns toward the east. The polar air which it brings with 
it is, as Schleiden remarks, very dangerous for consumptive persons, by 
reason of its extreme dryness and the abundance of oxygen in it. The 
east wind blows until some other wind comes to take its place, and this 
can only be done by the equatorial current which arrives as a southerly 
wind. The immediate result produced by this meeting is to give birth 
to an intermediate direction, or to the south-east wind, the hot and hu- 
mid air of which, cooled by the polar current, is obliged to abandon a 
part of its water in the shape of clouds, snow, or rain. The equatorial 
current gradually gains the mastery, the weather clears up, becomes 
warmer, and maintains itself with a southerly wind, which imperceptibly 
veers to the west. There is only the polar current which can, in turn, 
take its place ; the fusion of these, passing to the north-west, produces 
abundant atmospheric precipitation. Then we have those cold and 
damp days which are so unpleasant to persons of a nervous tempera- 

Strange to say, this variable zone, which one would be inclined to re- 
gard as the most unfavorable for the development of the human race, 
embraces nearly all midland Asia, Europe, North America, and the 
north coast of Africa, and consequently comprises the scene upon which 
has been illustrated the history of humanity and of its intellectual de- 
velopment. Perhaps there is some secret connection between this phe- 
nomenon and the special development of the vegetable world in this re- 

This sketch of the distribution of weather over the surface of the 
globe is modified by many causes. The elevation of countries above 
the sea-level, plains and mountains, sandy deserts and forests, cause 
great disturbances in the action of these laws. 

Among the influences which modify weather, one of the most impor- 
tant is the manner in which the sea and the land are spread over the 
surface of the globe. The land, being exposed to the solar rays, is heat- 
ed more rapidly than the sea, and, after a certain interval, attains a 
higher temperature, which, moreover, cools again far more slowly. The 


first consequence is that tbe hottest zone, the region of calms, is not 
equally extensive both to the north and to the south of the equator ; 
but, on the contrarj^, occupies the largest space in the northern hemi- 

We have seen that heat and its unequal distribution in all direc- 
tions is the fundamental phenomenon around which the others, which 
are dependent upon it, group themselves. The moisture of the air has 
an intimate co-relation with this phenomenon, and the latter, together 
with the heat, are the causes of vegetable life. It is upon these two 
conditions that principally depends the distribution of plants over the 
globe. The animal world follows the plants, for the existence of herbiv- 
orous beings is directly connected with that of the carnivora. The first 
supreme principle, that which not only vivifies, but stirs up and regu- 
lates all, is the sun ; its rays are the pencils with which it traces light 
and shadow, the burning yellow of the arid sand, and the fresh green 
of meadows, and even the sketch of an ethnographical map for the hu- 
man race. 




We have seen that the distribution of solar heat over the globe cre- 
ates in the atmosphere a general regular circulation. In the next chap- 
ter we will prove that the irregular and variable winds are alike due to 
this heat, and that they are subject to laws of periodicity which science 
is engaged in studying. But, before having done with the great cur- 
rents of the atmosphere, it is necessary that we should form some idea 
of the great ocean currents, also dependent on the action of the very 
same heat which regulates all things here below. 

The sea is not motionless; neither its waters nor the atmosphere 
above them. A great general oscillation of the surface occurs twice a 
day, under the attracting influences of the moon and sun : these oscilla- 
tions are the tides, whose flux and reflux alternately cover and lay bare 
the shores of the ocean, and give to the coast that endless variety which 
never fails to charm us. This movement of the waters is due to an as- 
tronomical cause, and need not be gone into here. But the sea is ani- 
mated by another meteorological circulation, more complex and wider, 
which may almost be compared to the circulation of the blood in oui- 
veins; it is traversed by currents which, running from the equator to 
the poles, and vice versd^ thus forming a connecting link between the 
most distant seas, distribute heat to colder regions, exercise a cooling 
influence within the torrid zones, equalize the briny and chemical com- 
position of the ocean, and form, to a certain extent, the vital circuit of 
the globe ; like the sap which rises and falls in plants, like the blood 
which becomes regenerated at the heart after having carried its tribute 
to the farthest extremities of the organization. 

These ocean currents merit our special attention, and their study will 
embrace at once the currents of the atmosphere which accompany and 
complete them, constituting the meteorology of the ocean. Both have, 
especially for the last thirty years, been the subject of detailed research. 

Maritime travel differs ah initio from journeys by land, in the absence 
of any fixed route. For a long period, indeed, modern navigators nev- 


er suspected that there existed upon the surface of the ocean numerous 
highways, traced by the hand of nature. The constancy of the mon- 
soons, the periodical return of the marine breezes along the coasts of 
the Red Sea and in the Indian Ocean, are phenomena which our fore- 
fathers had ascertained and utilized. When the astronomer Hippalus 
discovered the physical fact of the return journey of the summer mon- 
soon, he made a discovery which the Arabian sailors had for centuries 
been acquainted with, and which they had taken advantage of to pre- 
serve the monopoly in the trade of Ceylon spices and perfumes, which 
they sold as the products of Arabia. The discovery of Hippalus caused 
a complete revolution in the system of maritime services among the 
Europeans who flourished at the commencement of the Christian era. 
The discoveries effected by the researches of Lieutenant Maury, of the 
Washington Observatory, during our own day, are analogous to the 
above, but on a much larger scale. On account of their great inter- 
course with other peoples, and the geographical position of their coun- 
try, which is bounded by two oceans, the Americans were more inter- 
ested than any other nation in the discovery of the shortest sea- routes. 
To effect this purpose, it was necessary to compare with each other the 
thousands of routes that had been followed by thousands of navigators. 
This herculean task rendered it possible to deal with the whole globe 
as Hippalus had dealt with the short distance between Egypt and Ta- 

The great navigators of early ages seemed to have struck out the 
only routes practicable, without its occurring to them to introduce the 
modifications which the comparative study of the data of experience 
might have led them to. But when the application of steam to the 
means of transport had proved the advantages of a rapid system of in- 
tercommunication, and the great value of time, attention naturally be-' 
came turned to the discussion of better routes, and the means of decid- 
ing as to how they could be arrived at, A steam vessel, taking no ac- 
count of the wind, can trace upon the sphere the shortest and the most 
direct line between its point of departure and its place of arrival ; but 
with the sailing vessel, subject as it is to aerial currents which consti- 
tute its sole means of progression, the line which is shortest in point of 
distance often becomes the longest in respect to the time occupied in 
traveling along it. To find the greatest possible sum of favorable winds, 
without deviating more than can be avoided from the straight line, is 
the surest way to accomplish a quick passage. The observations taken 


at the surface of the seas by navigators were long allowed to remain 
profitless for the purposes of science and navigation. Under Maury's 
auspices they led, in a few years, to a knowledge of the general circula- 
tion of the atmosphere and the seas. At the same time they have been 
instrumental in reducing by a fourth, a third, and even a half, in some 
instances, the length of long voyages, and in eSecting an immense sav- 
ing in the cost of transport. 

To awaken public interest by some practical result which would dem- 
onstrate the great importance of these new studies, he concentrated all 
his efforts upon one single route — that from the United States to Eio 
Janeiro. The data which he collected enabled him to ascertain a route 
far shorter and better than that followed by the great mass of naviga- 
tors. The ship Wright, Captain Jackson, from Baltimore, was the first 
to steer by Maury's course. Starting from Baltimore, on the 9th of 
February, 1848, this vessel crossed the equator in twenty -four days, 
while the time occupied had previously averaged forty-one days. 

This route from the United States to the equator is all the more im- 
portant because it is the road of all ships sailing from North America to 
the southern hemisphere, whether their ultimate destination be the Pa- 
cific, the Indian Ocean, or the Atlantic. From forty-one days this pas- 
sage was reduced to twenty-four days, afterward to twenty, and finally 
as low as eighteen. This is a gain of fifty per cent. 

The passage from the States to California took, on an average, rather 
more than 180 days; after Maury had brouglit his knowledge to bear 
upon the subject, it was at first shortened to 185 days, and since then to 
100 ; while one of the fleet of clippers trading there — the Flying Fish — 
cast anchor in the harbor of San Francisco on the 92d day after leaving 
New York. 

But the most remarkable instance is furnished by the voyage to Aus- 
tralia, From England to Sydney, a vessel sailing under the old sys- 
tem used to take at least 125 days, which was the usual average. The 
return journey being about the same, the total length of the voj^age 
amounted to 250 days. When Maury passed through England to at- 
tend the Congress at Brussels, he promised the British sailors and mer- 
chants that, as a recognition of the help they had afforded him, he 
would diminish by at least a month the voyage to Australia, and reduce 
the return passage to a still greater extent; or, in other words, lessen 
by a quarter the distance between England and its wealthy colony. A 
little later, when the notions with respect to this route were complete, 


Maury pointed out the immense advantage that would be derived by 
making the voyage to Australia a regular circumnavigation of the 
globe — that is, doubling the Cape of Good Hope on the outward voy- 
age, and Cape Horn on the return passage. The total length of these 
two voyages would, he said, occupy 130 days, or even less, in place 
of 250, as were taken before. This prediction has been fulfilled, 
and even exceeded, the saving of time being equivalent to fifty per 

Let us see what is the economy from a pecuniary point of view. 
The price of freight to Australia is about one shilling per ton per day. 
Taking the average tonnage of the vessels upon this line as being only 
500 (they are 700 in reality), and assuming a reduction of only thirty 
days in the passage, it will result that each ship will have realized in 
each passage a saving of 15,000 shillings. If we take Maury's calcula- 
tions, and put the number of vessels of all flags that ply annually be- 
tween the North Atlantic ports and Australia at 1800, there will be a 
clear gain of twenty-five millions of shillings at the expiration of a 

For English commerce alone, in the Indian Ocean, the annual econo- 
my is nearly £500,000. Taking all passages effected by ships of vari- 
ous nations, this discovery must effect an annual saving of four millions 
of pounds sterling. 

The greater the distance to be accomplished, the greater is the advan- 
tage in deviating from the straight line to seek a region where continu- 
ous breezes will impel the vessel at the greatest speed. Thus, generally 
speaking, if one is sailing from east to west, it is in the intertropical re- 
gion that the speed is greatest; whereas, in order to sail very rapidly 
from west to east, it is necessary to go beyond the tropics, either north 
or south. 

Each day's delay in the arrival of a merchant- vessel beyond the fixed 
date or the average of passages is not only a more or less considerable 
cause of annoyance to the passengers, whose health, and even life, may 
be depending upon their speedy arrival ; it is also a cause of loss to the 
shipper and the merchant. The expenses of a large vessel vary, as Ad- 
miral Fitzroy pointed out — including wages, provisions, material, a full 
cargo, and an average number of passengers — from £50 to £200 a day ; 
moreover, to these immediate expenses must be added the diminution 
in the annual earnings of the vessel which are consequent upon the 
forced delay in its next departure. The evils incident upon a long pas- 


sage are, therefore, complex in nature, affecting the interest of the ship- 
per and of the public at large. 

The progress realized by the •' Sailing Directions " in shipping indus- 
try is, consequently, equivalent to that effected by the adjunction of a 
new motive power. Thus a ship which, sailing in the ancient track, 
would have been at sea for one hundred days, now follows the new 
course, and reaches its destination in half the time, and is thus, so to 
speak, supplied with a traction-engine powerful enough to double its 
speed. These fortunate results have been universally accepted. In a 
conference held at Brussels in 1853, the United States, France, England, 
Eussia, Sweden, Norway, Denmark, Holland, Belgium, and Portugal 
agreed upon a uniform plan of meteorological observations at sea, and 
this plan was soon adopted by Prussia, Austria, Spain, Italy, and Bra- 
zil. Since then, all the trans-oceanic vessels belonging to these powers 
have become floating observatories, which register night and day all the 
incidents of navigation calculated to secure a complete knowledge of the 
movements of the atmosphere and the sea. 

Thanks to these researches and to the development in late years of 
meteorological observations, I am enabled to give, in the previous and 
following chapter, a general sketch of the distribution of winds over the 
surface of the globe. 

Let us now consider the circulation of water, also due to the influence 
of solar heat. 

It is well known that the seas are divided, first into three great oceans, 
viz., the Atlantic, which separates Europe and Africa from America; 
the Pacific, which covers half of the globe between the two Americas 
upon the one hand, and upon the other Eastern Asia and New Holland, 
with the Archipelago between ; and thirdly, the small ocean known as 
the Indian Oceayi^ which is almost entirely upon the south of the equa- 
tor between Africa, Asia, and New Holland. 

If the two great oceans be divided into two parts, that to the north 
and that to the south of the equator, and if the polar seas be taken into 
account, we shall have altogether seven divisions in which the move- 
ment of the hot or cold waters, their flow from the equator toward the 
poles, and their return to the point from which they started, can be 
studied. It is to this movement that are due, throughout the sea, cur- 
rents of hot and currents of cold water, the majestic and steady changes 
of which, and the more or less varying temperature of which, give rise 
to effects of a far more important nature in the economy of climates 



than might be supposed by those whose only knowledge of the globe is 
derived from ordinary maps. 

Let us analyze and weigh these important currents, taking as an ex- 
ample the circuit formed by the waters of the Atlantic to the north as 
being best known to us, and which is continually being traversed 
by vessels coming and going between Europe and North or Central 

In the equatorial regions, the waters of the ocean are impelled to- 
ward the west by an incessant movement which, in the Atlantic, car- 
ries them toward tropical America. This vast current, 30° in width, 
twenty of which are to the north and ten to the south, breaks against 
the shores of the New World. In accordance with the shape of Amer- 
ica, the eastern point of which is a long way below the equator, the 
greater part of these waters make their way toward the Gulf of Mexico^ 
the bends of which it follows, and finally makes its way out again by 
the extreme point of Florida, running along the coast of the United 
States from south to north. This gulf, situated in the torrid zone, is on 
all sides surrounded by lofty mountains, which shut in the solar rays as 
within a vast funnel, and store up therein the heat of a burning climate. 
It is from this focus that the equatorial current starts. It runs across 
the Straits of Florida, and produces an impetuous stream, nearly 1000 
feet deep and fourteen leagues wide, running at the speed of five miles 
an hour. Its waters, which are warm and very salt, are of the color of 
indigo blue, and differ from their greenish borders formed by the waves 
of the sea. This mass creates in its passage a great agitation, and thus 
follows its course without becoming confounded with the ocean. Shut 
in between two liquid walls, the waters of the Gulf Stream form a mov- 
ing vault which glides along the sea, carrying off to a great distance all 
objects which get drifted into it. " In the greatest droughts it never 
fails, in the greatest floods it never runs over. Nowhere in the world 
does there exist so majestic a current. It is more rapid than the Ama- 
zon, more impetuous than the Mississippi ; and the collective waters of 
these two streams would not equal the thousandth fraction of the vol- 
ume of water displaces." — Maury. 

By means of the thermometer the navigator can follow the great 
liquid vein. The instrument, plunged alternately into its edges and its 
mid-stream, shows a difference of 27° of temperature. 

Powerful and rapid the Grulf Stream runs northward, following the 
coast of tlie United States, as far as the Banks of Newfoundland. There 



it encounters the tremendous shock of a polar current, upon which are 
floating enormous icebergs, veritable mountains of ice, the force of 
which is such that one of them, weighing more than twenty billions of 
tons, carried the vessel commanded by Lieutenant de Haven more than 
three hundred leagues southward. The Gulf Stream, whose waters are 
lukewarm, dissolves the floating ice. The icebergs melt, and the earth, 
and even the fragments of rock which they contained, are swallowed up 
by the waters. 

Upon reaching the neighborhood of Europe, it sends a great part of 
its waters in the direction of the Polar Sea, along the coasts of Ireland, 
Scotland, and Norway ; the remainder turns off to the south, opposite 
the west coast of Spain, and regains the great tropical current off the 
centre of Africa. After having effected their junction with this cur- 
rent, of which they are, so to speak, the source, its waters make their 
way westward, to reach once more the coasts of Mexico and the United 
States, and to traverse, for the second time, the space which separates 
the United States from Europe, thus forming a continuous circuit from 
Africa to Mexico, returning to the point from which they started by the 
route given. The bottles which sailors throw into the sea, with a men- 
tion of the day and the spot where they were confided to the ocean, 
have shown us that this voyage of from 13,000 to 19,000 miles is ac- 
complished in three years and a half The winds have about the same 
direction as the waters, that is to say, that between the tropics the east- 
erly trade-winds prevail, driving the atmosphere from Africa to Ameri- 
ca, just as the tropical current conveys the water thither. Between the 
United States and Europe, just as this current causes the sea to flow 
eastward, so also do the counter-currents of the trade- winds blow toward 
Europe, whence it happens that the passage from the United States to 
Europe is effected more rapidly than the return journey, for in this lat- 
ter case the wind and the current are against the vessel. It is well 
known that when Christopher Columbus ventured to give himself up to 
the west winds, he got as low down as Africa to take advantage of the 
easterly winds which, according to his calculation, would lead him to 
China. As the late M. Babinet remarks, it is difiicult to understand 
how, at this epoch, when geographical knowledge was sufficiently ad- 
vanced to permit of the globe's dimensions and the distance from India 
and China being pretty accurately known, any one could have expected 
to reach the eastern coasts of China after a navigation equal to three 
or four times the distance beween the Old and the New World. If 


America had not been in existence, he would have perished a hundred 
times over before he could have reached China. 

Before passing to the other maritime circuits analogous to these of 
the North Atlantic, let us consider carefully the circumstances by which 
it is characterized. 

The tropical waters, in their journey from the coasts of Africa to 
those of America, pass beneath the rays of a zenithal sun, and are con- 
tinually being heated until they reach the Gulf of Mexico ; they then 
flow by way of the Straits of Bahama, where they form a rapid current 
of hot water, which re-ascends to the east of the United States, toward 
the Banks of Newfoundland. There the current turns eastward on its 
way to Europe, but still preserves the high temperature due to its trop- 
ical origin, and this is one of the most powerful agencies of nature for 
increasing the temperature of our globe — viz., the conveyance, by means 
of these waters, of the heat which the sun sheds between the tropics to- 
ward the northern regions. In proportion as this current advances, it 
parts its heat, which it distributes into the atmosphere and over the 
seas which it traverses ; then it returns, leaving Spain and the north of 
Africa to its left, to resume its place in the tropical current, and again 
to receive heat, which it will transfer as before to European latitudes. 

It is by means of the winds that the heat of the sea communicates it- 
self to the main-land. We shall see presently that in Europe the pre- 
vailing winds of the globe are westerly, inclining to south-west. It is 
seen at once that these currents of air, having a current of hot water for 
basis, will share its temperature, and pass over Europe and be much 
warmer than if the sea, deprived of this warm current, had only the 
same degree of warmth as is due to latitude. To demonstrate this as- 
sertion, we have only to compare the climates and temperatures of 
American cities with those of France and England which are in the 
same latitudes. 

None of the masses of water which move from place to place in the 
seas merit such close attention as that of the Gulf Stream ; none are of 
greater importance in regard to the commerce of nations, nor exercise a 
greater influence upon climate. It is to the Gulf Stream that the Bri- 
tannic isles, France, and neighboring countries, owe, in a great measure, 
their mild temperature, their agricultural wealth, and, moreover, a very 
large part of their material and moral strength. Its history is almost 
that of the whole of the North Atlantic, so great is the hydrological and 
climacteric influence of this current of the seas. 



Owing to the rotfitorj motion of the globe, and probably also to the 
general direction of the coasts, the current follows without intermission 
a north-easterly course, and comes in contact with none of the advanced 
points of the continent. Beyond New York and Cape Cod it bends far- 
ther eastward, and, ceasing to run parallel with the American coast, 
turns off into the mid- Atlantic, toward the shores of Western Europe. 
As Maury says: "If an enormous cannon could fire a ball from the 
Strait of Bahama to the North Pole, the projectile would follow almost 
exactly in the curve or course of the Gulf Stream, and, deviating grad- 
ually as it went, would reach Europe traveling eastward." 

From the 43d to the 47th degree of north latitude, in the neighbor- 
hood of the Banks of Newfoundland, the Gulf Stream, traveling from 
the south-west, encounters upon the surface of the sea the polar current. 
The line of demarkation between these oceanic streams is never abso- 
lutely the same, and varies with the seasons. In winter — that is, from 
September to March — the cold current drives the Gulf Stream toward 
the south; for, during this season, all the circulatory system of the At- 
lantic — winds, rain, and currents — veer toward the southern hemi- 
sphere, above which the sun is situated. In summer — that is, from 
March to September — the Gulf Stream regains the preponderance, and 
repels the polar current farther north. After having come in collision 
with the waters of the Gulf Stream, those of the arctic current cease, in 
a great measure, to flow upon the surface, and sink by reason of their 
being cold, and consequently heavy. It is easy to trace the direction 
of this counter-current, which is exactly opposite to that of the Gulf 
Stream, by the mountains of ice which the mild temperature of lower 
latitudes fails to melt, and which float in a south-easterly direction, un- 
til they meet the superficial current, which they cleave like the prow of 
a vessel. Farther south, it is only by sounding that- the existe-nce can 
be ascertained of this hidden current, the cold waters of which serve 
as a bed to the warm stream proceeding from, the Gulf of Mexico. It 
descends lower and lower until it reaches the Straits of the Bahama 
Islands, where the thermometer indicates it at a depth of 1300 feet. — 

We have the pendant of the Gulf Stream in the Pacific Ocean, in the 
shape of a warm current which follows the coasts of China and Japan, 
and which has long been known to Japanese geographers by the name 
of Kuro-Siwo (the Black Stream) — a name which originated, no doubt, 
in the dark hue of its waters. In the southern seas the currents are 


not so well known to us, and are, in fact, much less numerous. It is, 
moreover, probable that the marine streams are not isolated currents, 
but several portions of one net-work, distinct veins in a comprehensive 
system of circulation. 

The quantity of heat which the Gulf current carries northward forms 
a very considerable part of the caloric which is stored up in the waters 
of the torrid zone. The total heat of the current would suffice, if it 
were concentrated upon a single point, to melt mountains of iron and to 
form a stream of metal as voluminous as the Mississippi ; it would, fur- 
ther, suffice to raise from winter to summer temperature the whole col- 
umn of air which lies over France and Great Britain. 

Notwithstanding the march of the sun, it is, upon an average, as 
warm in Ireland at 52° N". latitude as it is in the United States at 38° 
N. latitude, or a place more than 1000 miles nearer the equator. 

The Gulf current, which carries the tropical heat to the temperate re- 
gions of Europe, often serves, too, as a highway for the hurricanes; 
hence the names of Weather-breeder and Storm-king, which have been 
given to the Gulf Stream. The movements of the atmospheric ocean 
and those of the ocean of waters are so completely parallel that we are 
tempted to view them as one and the same phenomenon both in the 
currents aerial and marine. Thus the Gulf Stream seems to be for the 
winds what it in reality is for the waters — the great intermediary be- 
tween the two worlds. It transmits to the seas of Northern Europe the 
saline matters of the Gulf of the Antilles; it carries with it the tropical 
heat for the benefit of the temperate regions ; it marks the route follow- 
ed by the torrents of electricity proceeding from the storms in the An- 
tilles. It is, in fact, the great serpent of the Scandinavian poets, which 
displays its immense ring along the ocean, and which, by the motion of 
its head, either causes a mild breeze to blow or emits the raging hurri- 
cane. While, in the North Atlantic, the equatorial current, which falls 
into the Gulf of Mexico, returns from whence it came, traversing high 
latitudes, another part of this currerit, much less voluminous, after hav- 
ing touched Cape St. Roch, which forms the eastern extremity of South- 
ern America, descends along the eastern coast of that same continent, 
and then, crossing the Atlantic from west to east, returns toward Lower 
Africa, running along its western shores and rejoining, by the south, 
the great tropical current, just as the Gulf Stream meets it northward. 
Down to the quantity even of water which it contains, this current 
bears a marked resemblance to the circuit which occupies the north of 


this ocean. The portion which runs off beyond the tropics, and which 
returns from west to east, from South America to South Africa, is also 
a current of hot water, like the Gulf Stream between the United States 
and Europe. The comparison of the masses of water which each of 
these circuits separately conveys shows how much better the north is 
provided with hot waters than the south. It is not too much to say 
that the north circuit forms a current five or six times more abundant 
than the south circuit. If we now consider the Pacific Ocean, there 
also we find tropical waters which flow on to the shores of New Hol- 
land, the Northern Archipelago, and Lower Asia. Most of these waters 
re-ascend northward in vast currents of lukewarm water which give 
to High California and to Oregon climates very similar to those of 

The ISToi'th and South Atlantic, the North and South Pacific, and the 
Indian Ocean, each contain a current, that of the former ocean being the 
most voluminous. The Arctic seas, north and south, also appear to be 
traversed by a current running eastward, round the Pole, — Bahinet. 

The circulation of the sea is completed by submarine currents. There 
must exist one of these, conveying the waters of the Mediterranean into 
the Atlantic. Its existence is, in a way, demonstrated by a calculation 
which shows that the quantity of salt water in the upper current of the 
Straits of Gibraltar is 2900 cubic miles per year, the quantity of soft 
water contributed by the rivers 240 cubic miles, and that which is lost 
by evaporation 480 cubic miles. Thus there would be an annual ex- 
cess of 2660 cubic miles, if the equilibrium were not re-established by 
a submarine current. This hypothesis seems to have received confir- 
mation by a very curious fact. 

Toward the close of the seventeenth century a Dutch brig, pursued 
by a French corsair, the Phoenix., was overtaken between Tangier and 
Tarifa, and disabled by a single cannonade. Instead of sinking at once, 
the brig, which had a cargo of oil and alcohol, floated beneath the sur- 
face of the waters, and did not finally go to the bottom for two or three 
days, after having been carried twelve miles toward Tangier from the 
point at which she first disappeared. It was evidently carried this dis- 
tance by an under-current, in an opposite direction to that of the sur- 
face-current. This fact, in conjunction with some recent experiments, 
confirms the opinion which admits the existence of a current issuing 
from the Straits of Gibraltar. Lieutenant Maury also considered it 
certain that there is a submarine counter-current to the south of Cape 


Horn, which carries the overflow of the Atlantic into the Pacific. As 
a matter of fact, the Atlantic is continually being fed by very large 
rivers ; whereas the Pacific, into which debouches no important stream, 
must, on the other hand, lose an immense body of water, owing to the 
evaporation which takes place from its surface. 

Certain lower currents have been ascertained by weighting a piece 
of wood and plunging it into the water, keeping hold of it at the same 
time with a piece of string, so as to let it sink to any depth which may 
be desired. At the other end of the line is attached an empty barrel 
strong enough to support the apparatus, and then the whole is set free. 
The sailors who tried this experiment for the first time were astonished 
to find this little barrel traveling in an opposite direction to the wind 
and the sea at the rate of a knot or more per hour. The crew were 
even inclined to look upon it as a supernatural phenomenon. The 
speed of the barrel was evidently equal to the difference in speed be- 
tween the upper and the lower currents. 

In 1773, Captain Deslandes cast anchor in the Gulf of Guinea; a 
strong current running into this bay prevented him from going farther 
south. Deslandes then noticed that there was an under counter-cur- 
rent at the depth of eighty feet, and he adopted an ingenious plan for 
availing himself of it. A machine, with considerable surface, was let 
down to the depth of this submarine current. This was hurried along 
with so much force that it towed the vessel at the rate of one and a 
half miles per hour. 

In the Antilles seas a vessel is sometimes brought to a halt even in 
the middle of a current. 

In the Sound there has long been known to exist both an upper and 
an under current. 

The mean temperature of the surface of the sea differs but little 
from that of the air, so long as warm currents do not add their influ- 
ence. In the tropics it appears that the surface of the water is rather 
warmer than that of surrounding air. 

An examination of the temperature at the surface at various depths 
gives the following results : 

1st. In the tropics the temperature diminishes with depth. 

2d. In the polar seas it augments with increased depth. 

3d. In the temperate seas, included between 30° and 70° latitude, 
the temperature decreases in a smaller degree as the latitude gets high- 
er, and beyond degree 70 begins to increase. 


There exists, then, a zone the temperature of which is almost sta- 
tionary, from its surface down to a great depth. 

It is scarcel}'- possible to doubt that the currents caused by the dif- 
ference in pressure which strata of the same level are subject to, at the 
equator and toward the poles, contribute materially to this distribution 
of heat. It seems certain that there is, as a rule, a surface current 
which carries the warm waters of the tropics toward the polar seas, 
and an under-current which takes back from the poles to the equator 
the frigid water of the polar regions ; but the direction and intensity 
of these currents are modified by a number of causes, which depend 
upon the depth of the sea basins, their shape, and the influence of winds 
and tides. In very deep water there is a uniform temperature of 39°, 
which corresponds, as physical science has proved, to the maximum of 
the density of water. This temperature exists at the equator at a depth 
of 7200 feet. In the polar regions, where the water is colder upon the 
surface, this same temperature is met with at a depth of 4600 feet. The 
isothermal lines of 39° form the demarkation between the zones where 
the surface of the sea-water is colder, and those where it is hotter than 
the stratum which marks 39°. 

Lastly, the quantity of salt in the waters of the ocean differs accord- 
ing to the points of the globe, and is unquestionably an important ele- 
ment in the density, and, consequently, in the actual formation, of mari- 
time currents. 




Having observed the regular and periodical currents of the atmos- 
phere and the seas, let us now consider the irregular winds which blow 
over our climates. These latter are only apparently irregular, for in 
nature there is no such thing as chance, and each molecule of air that 
changes position is obeying laws as absolute as those which regulate the 
worlds of space. We will endeavor to throw some light upon the mul- 
titude of winds which succeed each other in our regions, and to ascer- 
tain the causes which set them in motion. 

Beyond the changing limits within which blow the trade-winds and 
the periodical breezes of the two hemispheres, the temperate zones are 
the seat of variable winds. Europe, for instance, is entirely subject to 
that regime, and the masses of air float off sometimes in one direction, 
sometimes in another. Now and then one kind of wind will prevail for 
weeks together; sometimes, on the other hand, the wind will blow from 
two or three different points of the compass in as many hours; some- 
times, again, the air remains calm, and there is not a breath of wind to 
agitate even the foliage of the poplar-tree. Thus the instrument used 
to indicate the direction of the winds in our climates, the weather-cock, 
has long been taken to signify inconstancy. 

Nevertheless, even inconstancy has a cause, and is often more appar- 
ent than real. The winds in our climates, which seem to us so capri- 
cious and variable, leave behind them a trace of the laws which they 

We have seen that the upper trade - winds, which travel from the 
equator to the pole, modify their primitive direction from north to south 
in our hemisphere, and veer gradually to the south-west as they reach 
higher latitudes. They lose at the same time both in velocity and heat, 
and gradually come nearer to the ground. About 30° latitude they are 
almost on a level with the surface. This south-west wind, in fact, pre-- 



vails throughout Europe. Thus, amidst the variety of winds, we al- 
ready fiud that there is one which is regular, since it is no other than 
the upper trade- wind which has descended thus far, and which occupies 
the largest place in the meteorology of our climates. 

We have seen that the great oceanic current, the Gulf Stream, reach- 
es the coasts of Europe from the south-west. The air circulates in the 
same direction, and increases still further the inflection of the upper 
trade-winds, or, to speak more correctly, it is the same equatorial aerial 
and maritime current turned off in a south-west direction by the rota- 
tion of the earth. 

To ascertain precisely the direction of the winds, it is necessary to 
keep an account of the time during which each wind has prevailed, tak- 
ing a supposititious total upon which the calculation is based. Thus, for 
instance, let us suppose that the south-west wind has been blowing for a 
little more than ninety days of the year ; it would be put down that it 
had prevailed for a quarter of the whole time. If the total 1000 be 
taken to signify this time, 250 would be placed to the account of the 
south-west wind (that is, if it had been blowing for ninety-one days and 
seven hours, which is exactly a quarter of a year). In the same way 
all directions indicated by the vane would be similarly put down, and 
thus we should obtain a comparative table giving the average result for 
a long series of years. 

This plan has been adopted in Europe for many years, and the fol- 
lowing table will show the result of the observations made. It indi- 
cates a decided preponderance of a south-westerly wind over the Euro- 
pean continent and even North America: 











Direction of 

the Wiud. 

Force of 
the Wind. 



















S. 88° W. 
S. 66° W. 
S. 76° W. 
S. 62° W. 
S. 50° W. 
S. 86°W. 







N. America. . 


It will be seen that the souih-ivest is the prevailing wind. By adding 
up the numbers set down, as they run horizontally, the total will be 
found to be 1000 ; thus in France the south-west wind blows yVV, or 


nearly a fifth of the whole time. The proportion is greater still in En- 
gland, By adding together the west and the south, it will be seen that 
the continuance of wind from this quarter amounts to nearly one-half 
of the prevailing winds ; -^^ in France, and ^^V i^i England. The 
careful observations taken at Brussels and in various parts of Belgium 
since 1830 show a like preponderance to exist there. The prevailing 
wind is, indeed, just S. 45° W. In Eussia there is a greater variety, 
owing to its distance from the ocean. 

Thus we are under the benign influence of the equatorial current. 
^But, if the return trade-winds reach so far, and even to the pole, the 
lower polar current, which conveys the cold air from north to south, and 
forms in the tropics the north-easterly trade-winds, must also have its 
influence upon our regions. It must pass us somewhere on its way 
from the pole to the equator; and if the air which travels from the 
equator to the pole did not return, there would cease to be any atmos- 
phere at all in the tropics. Now, let us study for a moment the pre- 
ceding table of the relative frequency of the winds. The maximum is 
to the south-west, as is shown by the figures underlined, whence the to- 
tals become smaller and gradually swell again, giving a second max- 
imum in the shape of a N.E. wind. That is our polar current. The 
N.E. wind forms the y\^ of the winds in France, ^^V of those in En- 
gland, and tW in Russia. 

There exist, therefore, in our hemisphere two general directions of 
winds. Now it is the equatorial, now the polar current which predom- 
inates. The first is warm and moist, the latter cold and dry. Each has 
an opposite influence upon the productions of the soil, and the state of 
the crops depends in a great measure upon the epoch and continuity of 
their prevalence. 

The S.W., W., and S. winds on the one hand, the N.E. and N. winds 
on the other, constitute the gQnev^ primitive winds to which our regions 
are subject. All the other directions of the wind are due to these two 
currents, and for the following reasons : 

If the two currents are blowing in proximity to each other, each oc- 
cupying a certain extent, as they are proceeding in opposite directions, 
there must exist, about the limit which separates them, whirlwinds and 
circular blasts engendered by the action of the two currents of air. 
These circular blasts will revolve from N.E. to S.W. at the tangent of 
the polar current, from S.W. to N.E. at the tangent of the equatorial 


As an instant's reflection will show, this is a simple horizontal move- 
ment like that of a grinding-stone. Each point in the circumference of 
this grinding-stone will have its own direction, since we are supposing 
that this mass revolves in its entirety. It would be, in fact, a zone of 
variable winds which would be liable to change its place under the in- 
fluence of the two great currents from which it springs, and which them- 
selves vary in position, extent, and intensity. Here we have one cause 
for the change of wind which is almost constant (since the two currents 
are always in existence), and which must be multiplied to a vast extent. 
There is a second and not less important cause. 

There is a constant difference of temperature in the various regions 
of the same country. In one place there is water, in another land; here 
deserts, there forests; at one point low and sultry plains, at another 
bleak table-lands. These differences of temperature modify our two 
currents on their passage through them. A cloudy sky is favorable to 
the progress of the one, and arrests the march of the other. Thus par- 
tial winds spring out, like lateral branches, from the trunks of the two 
great trees which are lying prostrate. 

A third cause must be superadded to the above — the protuberances 
upon the land. The general currents which pass over a chain of mount- 
ains do not blow with the same regularity that they do in the plains. 
In fact, the winds must be all the more unequal in their successive 
blasts in proportion as the surface over which they sweep is uneven. 
The same aerial surface which moves over the waters with the uniform- 
ity of a vast river, loses the regularity of its movement when it is inter- 
rupted in its course by the protuberances of the soil. At the foot of the 
Swiss mountains, and especially around Geneva, where the ground is 
very uneven, the alterations in the force of the wind are so great that 
the anemometer sometimes shows a variation in intensity from one to 
three. In the lofty ravines of the Alps it often happens, even in the 
midst of the fiercest storms, that the atmosphere is at intervals perfectly 
calm. Even in the countries which are not very hilly, and in the plains 
studded with houses and plantations, the wind does not blow with the 
regularity of the trade-winds at sea, but advances with a succession of 
blasts, each of which represents a victory of the atmospheric current 
over some obstacle upon the plain. 

At the level of the ground the wind is always intermittent, whereas 
in the heights of the air it almost always proceeds with the regular and 
majestic motion of a river. 


Thus laws regulate these minor changes as well as the general move- 
ment of circulation. We may now consider whether there is any law 
as to the succession of winds. 

Let us revert to the first cause of change dealt with above. As a 
rule, our hemisphere is divided into large oblique bands composed of 
masses of air running in an inverse direction, some toward the poles, 
others toward the equator. These bands shift their position around the 
globe, so that at one moment the polar wind, at another moment the 
tropical wind, will prevail in the same place; but there is always a 
compensating balance between these atmospheric currents, and the wind, 
which is neutralized or repelled in one part of the hemisphere, is soon 
felt at some other point. As long as the struggle between the two 
masses of air animated by opposing movements continues, the vicissi- 
tudes of the conflict and the general preponderance of one of the winds 
cause a temporary modification in the march of the air, and make the 
vane turn toward the different points of the horizon. It is from the en- 
counter of the two regular winds that chiefly arises the apparent irregu- 
larity of the whole atmospheric system. 

Although the struggle between the two aerial streams is continually 
going on at one point or another, nevertheless they are not of equal 
force, and one of them always obtains the mastery after a more or less 
prolonged period of resistance. This wind, which proves the superior 
in impulse, is the back current that has come down from a great eleva- 
tion, and reaches the level of the ground outside the zone of the trade- 
winds. The atmospheric currents coming from the equator naturally 
incline toward the east, whence it results that, in the northern hemi- 
sphere, the majority of the winds are from the west. 

Many centuries ago the savants ascertained that in the northern hemi- 
sphere the normal succession of the winds is from south-west to north- 
east by west and north, and from north-east to south-west by east and 
south. This is a rotatory movement analogous to that which the sun 
seems to describe in the sky when, after rising in the east, it travels 
westward, developing its vast curve around the zenith. Aristotle, in 
his "Meteorology," wrote, more than two thousand years ago, "When 
a wind ceases to give place to another, the direction of which is next in 
order to that of the former, the change always takes place with the sun." 
Since the time of the great Greek naturalist, several authors enumerated 
by Dove have re-affirmed this fact of the regular rotation of the winds, 
which was, indeed, known to sailors in the earliest ages. Dove was the 

302 ^^-^ ATMOSPHERE. 

first who collected the scattered proofs of this generally accredited the- 
ory, and transformed the primitive hypothesis into a scientific certainty. 
It no longer admits of any doubt that, in the northern hemisphere, the 
winds generally succeed each other in the following order : S.W., W., 
N.W., K, N.E., E., S.B., S., S.W. 

In the southern hemisphere the normal rotation of the aerial currents 
is exactly the opposite. Thus, as E. Eeclus remarks, the procession of 
the winds in each of the two hemispheres coincides with the apparent 
march of the sun, which, so far as Europe is concerned, describes its 
daily course to the south of the zenith, and, in Australia, passes to the 
north of it. Such is the regular order which Dove termed the law of 
gyration, but which is generally called after the name of its discoverer. 

I have noticed in my aerial travels a gyratory deviation, which shows 
that the wind can not extend in a straight line when it spreads over a 
great area, but inclines in the direction indicated by the above theory. 

Immersed in the atmospheric current which bears him along, the 
aeronaut is placed in the most favorable position imaginable, both for 
ascertaining the continuous direction of the current and for measuring 
its speed. Upon each occasion I took care to trace accurately on a map 
of France or Europe the aerial line taken by the balloon, which is done 
with extreme ease when the sky is clear, and which may always be ob- 
tained even with a cloudy sky, either by availing one's self of the mo- 
mentary breaks or by descending every now and then below the clouds. 

The balloon marks so accurately the direction and speed of the cur- 
rent, that the first sensation in navigating the air is that of being com- 
pletely at a stand-still. It is a peculiar and always surprising impres- 
sion experienced when, traveling along with the velocity of the wind, 
one feels neither the slightest breath of air nor the least movement, even 
when hurriedly carried off into space by the most violent tempest. I 
never felt but once any thing like a breeze. This was on the 15th of 
April, 1868, and then only for a few minutes. This I attribute to the 
fact that the balloon, which was traveling at the rate of thirty-four miles 
an hour, had reached a region where the air was shifting its position less 
rapidly. One capital fact is brought to light by the aerial lines which 
I have traced, and that is, that these routes all incline in the same di- 
rection, by virtue of a general gyratory deviation. 

The actual direction of a wind is the most easily observed of its char- 
acteristics. To ascertain it, we suppose the horizon to be divided into 
four equal sections by two diameters perpendicular to one another, one 


running from south to north, the other from east to west. The points 
at which the diameters intersect the horizon are called the four cardinal 
points. But they would not of themselves suffice, for it is necessary to 
have a number of intermediate directions. These are indicated by oth- 
er diameters, which divide the horizon into sixteen equal parts; and 
thus we obtain the indications of the wind at as many different direc- 
tions, called, starting from N. round by E., N.N.E. ; N.E. ; E.N.E. ; E. ; 
E.S.E.; S.E.; S.S.E. ; S. ; S.S.W. ; S.W.: W.S.W. ; W. ; W.N.W. ; 
N.W.; N.N.W.; K 

When the points of the compass are known, and objects are affected 
by the movement of the air, it is easy to ascertain the direction of the 
wind ; but often recourse is had to an instrument which is no doubt 
the oldest of those used in meteorology, viz., the weather-vane. This 
simple apparatus consists of a metal plate, generally of tin or zinc, cut 
into a figure of some kind, and turning upon a rod, to which is attach- 
ed a horizontal cross with the letters N., S., W., E., at its extremities. 
The weather-vane is placed upon the highest part of a building, and in 
by-gone days no house of moderate size was deemed complete without 
it. Exposed to the weather, it becomes corroded, and ceases to follow 
implicitly the impulsion of the winds. Sometimes the rod gets out 
of order, and the vane inclines to one side. Its indications are not 
worth consideration unless they are verified from time to time, and the 
vane is situated beyond the influence of obstacles which obstruct the 
free passage of the wind. It is not a rare occurrence for the atmos- 
phere to be influenced by several different currents, one superposed 
upon the other. In this case, the principal current — that which, so 
to speak, governs the weather — is generally placed at a considerable 
height, even when it is not the highest of all ; it is discovered by the 
motion of the clouds. This is the best and surest indication of the 
direction of the wind. 

As the mass or density of the air only varies within very restricted 
limits, the force of the wind depends almost entirelj^ upon its speed, 
and varies as the square of its velocity, or very nearly. The terms 
" force of the wind " and " velocity of the wind " are therefore almost 
identical. To measure the speed, an apparatus, called an anemometer^ 
is used. One of those most frequently in use is that by Dr. Eobinson, 
of Armagh. This instrument is composed of a vertical axis support- 
ing four horizontal radii of the same length, crossing at right angles, 
and at the extremities of which are four hollow half-spheres. 


A moment's reflection will suffice to make it clear that the wind is 
always pressing against two concave and two convex half-spheres. As 
it has more power over the former than over the latter, it causes a ro- 
tatory motion, and the number of revolutions which the half-spheres 
make is proportional to the velocity of the wind. The number three 
represents with approximate accuracy the relation which exists be- 
tween the horizontal movement of the air and the horizontal move- 
ment of the half-spheres. Thus, by measuring the circumference of 
the circle which the centre of one of the demi-spheres describes, and 
by multiplying half its length by three, we obtain the distance traveled 
by the wind for each revolution of the apparatus. 

The monthly averages of each wind referred to eight points of the 
compass, as found from sixty years' observations at the Observatory at 
Paris (1806-1866), are as follows: 


TheN 1039 

" N.W , 1084 

" W 1782 

" S.W 1935 

" S 1476 

" S.E 799 

" E 694 

" N.E 1191 

These numbers show the dominant winds to be S.W. and S. 

The monthly averages of the winds at London show a prevalence of 
south-westerly winds to an even more marked extent than in Paris. 
The result of observations taken for twenty consecutive years at the 
Grreenwich Observatory, which I have received from Mr. Glaisher, the 
director of the meteorological service there, gives the following averages 
of the relative frequency of each wind (see Fig. 58) : 

The N. y, 

ind blows on an average 

for 41 days 


48 " 


22 " 


20 " 


34 " 


104 " 


38 " 


24 " 


of complete 


34 " 




-Average annual prevalence of the different 
winds at London. 

The average direction of the winds at Brussels gives the same result 
(see Fig. 59), and we have already remarked the predominance of the 
equatorial current in the study 
of the general mass of observa- 
tions taken throughout Europe. 
It seems certain that the wind 
is propagated not only by im- 
pulsion, but by aspiration. This 
second mode deserves attention 
because it furnishes important 
data as to the cause of the 
movement. Franklin appears 
to have been the first to ob- 
serve this fact. He mentions in 
one of his letters that, when at- 
tempting to watch an eclipse of 
the moon at Philadelphia, he sw 
was prevented from doing so rig. 58.- 
by a hurricane from the north- 
east, which took place at about seven in the evening, and was fol- 
lowed, as is usually the case, by clouds which obscured the whole sky. 
He learned to his surprise, some time afterward, that at Boston, which 
n.n.m, », -KT,^ is about 400 miles to the north-east 

of Philadelphia, the storm had not 
commenced until 11 p.m., long after 
the first phases of the eclipse had 
been observed ; and, by a compari- 
son of the various accounts collect- 
ed in different colonies, Franklin 
remarked that, according as the 

Fig. 59.-Average annual prevalence of the differ- place WaS farther north, the later 
ent winds at Brussels. ^^g ^^^ ^^^^ ^^ ^^-^-^ ^^-g ^^^^-^^ 

easterly tempest occurred there, and that thus the wind was blowing in 
one direction and was advancing progressively in another. 

Since that time a great number of tempests have been remarked, 
which presented this peculiarity in respect to their direction. Never- 
theless, in nearly every case, the wind advances in the direction toward 
which it is blowing. 

The terrible storm from the south-west, which occurred on Novem- 



ber 29, 1836, passed over London at 10 a.m., the Hague at 1 p.m., Am- 
sterdam at 1'30 P.M., Emden at 4 P.M., Hamburg at 6 p.m., and Stettin 
at 9*30 p.m. It traveled, therefore, in the same direction as that in 
which it was blowing, and took ten hours to reach Stettin from London. 

The following is a general sketch of the prevailing distribution of 
wind over the surface of the globe : 

Suppose a ship to start from the Arctic Polar Circle for the equator, 
to cross it, and proceed onward to the Southern Arctic Circle, it will 
meet with the following succession of winds : 

1st. At the outset, it navigates in the region of south-westerly winds 
or of the northern anti-trade-winds, so called because they blow in an 
opposite direction to the trade-winds of their hemisphere. 

2d. After having crossed the parallel of latitude 50°, and until it 
reaches that of 35°, it encounters the zone of partially western winds, 
in which south-west predominates, and in which the north-easterly cur- 
rent also prevails over the other winds, 

3d. Between N. latitudes, 40° and 46° there is a region where the 
winds are very variable, and where there are calms. The winds blow, 
in the course of the year, in equal proportions from the four quarters 
during three months. 

4th. To the west winds, which have predominated thus far, succeeds 
the calm region of the Tropic of Cancer, then that of the trade-winds 
which conduct the vessel to the latitude of 10° north, where it reaches 
the zone of equatorial calm, which is only 5° in breadth. 

5th. From 5° north to 30° south the south-easterly trade-winds pre- 

6th, Then succeeds the calm zone of the Tropic of Capricorn, analo- 
gous to that of the Tropic of Cancer. 

7th. From S. latitude 35° to 40° there prevail, as a rule, westerly 
winds, which sometimes veer to N.W. and to S.W. 

8th, Lastly, the vessel reaches at S. latitude 40° the southerly anti- 
trade-winds, which have a north-westerly direction, and prevail, as far 
as observations in the direction of the Southern Pole have extended. 

If we now consider the intensity of the wind, we notice that its varia- 
tion, apparently so irregular, is dependent, like every thing else, upon 
the movements of the earth, in the seasons and in the days. Twenty 
years' comparisons made at Brussels show that the wind is less intense 
during the longest days than during the shortest, as in June the indi- 



cations of intensity are 0-832, and in December 1-227. The month 
of September, however, seems to be an exception, for it gives the 
minimum, averaging only 0-804; but this month is, in many re- 
spects, an exceptional one in our 

It is, moreover, remarkable that 
durinsr the six months when the 
sun is below the equator the force 
of the wind is above the average 
of the year ; whereas, on the con- 
trary, its force is generally below 

, -, . ^ n .^ Jan.. Teb.^Mar.Apr.MayJime.TnlyAurf. Sept.OctXov-.Deo. Jan. 

the average durmg each oi the -^ '' 

. Fig. 60.— Monthly intensity of the winds. 

other SIX months. 

The intensity of the wind varies, too, according to the time of day. 
The anemometer at the Brussels Observatory, which registers the wind 
every five minutes, shows that this diurnal variation in the inten- ■ 

sity of the winds extends from 
an average of 0-15 (midnight to 
4 A.M.) to 0-21 (10 A.M.), 0-26 
(noon), 0-29 (2 p.m.), 0-28 (4 
d<ta5(..2^ 0" ^ 8^i(?»Noon2^ 4^ G^ 8^ io^M,i,igl>! P.M.), and 0-23 (6 p.m.) This is 

Fig. 61.-Diurnal intensity of the winds. ghoWU by Fig. 61. 

Thus the wind is almost twice as strong at 2 p.m. as in the middle 
of the night. 

The time will arrive when the march of the variable winds in our 
climates will be ascertained, just as the general circulation of the trade- 
winds and the monsoons in the tropical regions has long been made 
known. The day will come, too, when observations of the upper winds 
will have revealed to the meteorologist the route which they follow, 
just as observations of the planets have discovered to the astronomer 
the orbits they describe. Then we shall be able to tell the daily and 
yearly direction of the atmospheric wave which passes over our heads. 

The currents, the laws of which we have been studying, play a great 
part in nature. They favor the growth of flowers by causing the 
branches of the plants to oscillate, and blowing the seeds a long dis- 
tance. They renovate the air in cities, and render northern climates 
milder by supplying them with heat from the south. Without wind 
rain would be unknown in the interior of continents, which would be 
transformed into arid deserts. Without wind the earth would be al- 


uiost uninhabitable, and whole districts would become centres of con- 
tagion — vast cemeteries, in fact. We have seen the deleterious effects 
of air when confined. Man acts as a deadly poison to man, as typhus 
fever and plagues clearly demonstrate. The winds alone can avert 
these calamities, by blowing away the emanations, by disseminating 
them in the regions of space, and substituting for vitiated air a fresh, 
salubrious atmosphere. Moreover, it is the same with air as with wa- 
ter; motion alone keeps it pure, whether because it has a principle of 
life unknown to us, or because animalculse, or vegetable and animal 
debris, becoming decomposed when at rest, spread their deleterious 
principles throughout a motionless atmosphere. 

The winds not only bring life upon their blast, they may also trans- 
mit death to countries where the yellow fever, the plague, or cholera 

A distance of twenty leagues does not protect Eome from the deadly 

'air which has blown over the Pontine marshes. In Paris the west 

wind blows for seventy days in a year ; place an Agro Romano in the 

Mayenne, the Sarthe, or Touraine, and the population of Paris would 

be decimated by intermittent fever.* 

It has been mentioned that in all latitudes similar to those of Europe, 
and even rather more southerly, the prevailing wind is west, which 
conveys to Europe the warm air of the Atlantic, and endows it with 
that unique climate which admits of the cultivation of barley and other 
cereals as far as the North Cape; whereas in Greenland, which is de- 
prived of these balmy breezes, it never thaws, although this latter coun- 
try is in about the same latitude as the north of Scotland. The city 
of Boston, in the United States, is in the same latitude as the olive- 
growing districts of Spain. Nevertheless, during the winter there, the 
small lakes in the neighborhood are sometimes frozen a yard deep. 
The five great American lakes (which are, in truth, inland seas) freeze 
over, and are traversed by temporary railroads. What a striking con- 
trast between the climate which produces this ice and that where the 
olive-oil and wine afford an easy subsistence to the indolent cultivators 
about Bordeaux and in Spain ! Yet the intelligent activity of the in- 
habitant of the United States has transformed even this ice into a prof- 

* There are at times strange variations in the Bills of Mortality which can be due to no 
other cause than the wind. Tluis, for instance, on July 26th, 1871, half of tlie inhabitants of 
Paris were attacked by a mild form of cholerine. There had been no other perturbation than 
a heavy gale of wind which raged all the previous night. 



itable crop, which is exported to India and the tropical regions, fetch- 
ing a higher price than that obtained for the olives of the Asturias. 

Toward the centre of France there exists the most exquisite climate 
of the whole world, so that if a locality be selected somewhere about 
the east of the meridian of Paris, it will possess a more favorable cli- 
mate than any other place in the same latitude. 

Ijet us now consider the influence which the wind has on climatol- 
ogy. The winds have a dominating influence upon the distribution 
of temperature, as they effect in different countries, according to their 
positions in respect to the four cardinal points, permanent modifica- 
tions in the climate which these countries would otherwise have. The 
regime of the winds leads to a regime of temperature which is indis- 
solubly connected with it. The currents of the atmosphere bring with 
them the temperature of countries whence they come. Every one may 
have noticed that the north wind is generally cold, and the south wind 
generally warm. But it would be commonplace to be satisfied with 
these vague indications, and the role of science is to analyze facts. Con- 
sequently, for many years past, the temperatures which the thermome- 
ter denotes for the directions of the wind have been carefully compared, 
and one of the first results was to show that in France the winds blow- 
ing from the south-east and the south cause an increase of 5° or 7° in 
the temperature over those which blow from the opposite direction. A 
comparison of the mean corresponding temperatures of the different 
winds throughout the various cities of Europe has made it evident that , 
the influence of the wind varies according to places, as may be seen by 
the appended table : 



























Zecken (Silesia) 

Arys (Prussia) 

Reikiawick (Iceland). . 

Thus the mean difference between the influence of the warm and of 
the cold winds reaches 7°"2 in Paris, and as much as 11°*5 in Iceland. 
There are often differences even more marked. 



The coldest wind is nearly always that which blows from a direction 
between north and east. The warmest wind is nearly always from 
S.S.W. The farther one passes inland the nearer it approaches to 
the west. 

The preceding fact is a confirmation of the meteorological truth that 
no phenomenon stands alone: all act and react upon each other. No 
sooner does the S.W. wind begin to blow than it takes effect upon the 
temperature, not only by its warmth, but by the vapor which it brings, 
and the condition of the sky which is the consequence. In winter, the 
moist west winds are remarkably warm, because they cover the sky with 
clouds, and thus prevent loss of heat by terrestrial radiation. 

The winds affect not only temperature, but also atmospheric pressure. 

When the north and north-easterly winds are blowing, the barometer 
rises ; it falls when the wind is from the S. or the S.W. 

The following is the result of a great many years' observations in the 
principal cities of Europe, and it shows very clearly the influence of the 
wind upon the reading of the barometer: 










St. Pe- 





































The general result of these researches is that the barometer rises 
highest with the wind between north and east — that is to say, when the 
current is coldest ; and that its minimum elevation is when the wind is 
anywhere between south and west, the points from which its current 
blows the warmest. Analogous conclusions have been obtained in oth- 
er countries. Thus, upon the eastern coasts of the United States and 
China, the barometer is generally highest when the wind is in the north- 
west — the coldest which prevails in those regions — and, as a rule, low- 
est when it is in the south-east, the temperature being at its maximum 
when the wind is in this direction. 

The fact of the readino^ of the barometer increasing with cold winds, 


and decreasing when the winds are warm, is one that has been made 
evident wherever observations upon the point have been taken. 

It may be generally stated, so far as our hemisphere is concerned, that 
the barometer reaches its maximum when the ivinds blow from the north and 
the interior of continents^ and its minimum when they come from the equator 
or the sea. 

In Europe the most rain-bringing winds are those between south and 
west, and the driest those between north and east: this is the reason 
why it rains oftener when the barometer is low than when it stands 

Just as the winds, according to the direction whence they come, in- 
fluence the temperature and the pressure of the air, the reading of the 
thermometer and the barometer, so do they affect humidity^ announcing, 
bringing on, and keeping off rain. Daily experience tells us that the 
air has not always the same degree of moisture irrespective of the direc- 
tion of the wind. When the farmer desires to harvest his hay or corn, 
when the laundress puts out her linen to dry, their task is accomplished 
far more rapidly with an easterly than with a westerly wind. Certain 
dyeing operations can only be attempted with the wind in the east. In- 
structive as these observations may be, they can not, however, provide 
us with rigorous and unchanging laws. 

The air always contains, in addition to the gases of which it is com- 
posed, a certain quantity of vapor of water, and this element plays a 
principal part in the absorption and distribution of heat over the sur- 
face of the globe. 

It would be to the highest degree important to be able to ascertain 
numerically the quantity of vapor which exists in the several regions 
of the globe. The life of plants and of animals, the nature of the land- 
scape, are dependent upon this element as well as upon temperature: 
the dryness and the humidity of the air have the greatest influence upon 
the development of disease. What we do know is that the air above 
all the seas is saturated with vapor of water. 

The farther inland, the drier the air becomes : nevertheless, after 
long-continued rain, it is at times saturated with moisture overland, be- 
cause soft water vaporizes more readily than salt water. But, generally 
speaking, the quantity of vapor of water contained in the air varies ac- 
cording to the country ; and there are regions — the deserts of Africa 
and Asia and the steppes of Siberia, for instance — where there is not 
the slightest evaporation from the soil, and where the air is dry in the 


extreme. The winds which come from the sea bring humid air ; those 
which blow from the land bring dry air. 

The quantity of vapor with which the air may be laden, varies, ac- 
cording to temperature, in the following proportions : 

At 14°, a cubic foot of air 

49°, " 


s saturated with water by the weight of one grain. 

" '■ " two grains. 

" " " three grains. 

" " " four grains. 

" " " five grains. 

" " " seven grains. 

" " " eleven grains. 

" " " fourteen grains. 

" " " twenty grains. 

At 212°, the air is capable of absorbing a quantity of vapor of water 
equal to its own volume ; the tension of the water becomes equal to 
that of the air ; it boils ; and the pressure of the vapor is equal to one 

Thus the hotter the air the more it can contain of water in a state of 
invisible vapor. Let us suppose a cubic foot of air to be saturated with 
vapor at 100° : it contains twenty grains. Now, if a current of cold air 
sets in and reduces it to 30°, as it can only now contain two grains, it is 
obliged to part with about eighteen grains of water. This condensation 
would lead to diurnal rains if cold currents were to encounter daily sat- 
urated masses of air. 

The quantity of vapor is at its minimum when the wind is blowing 
between N. and N.E. ; it increases when the wind is in the E., the S.E., 
and the S., and attains its maximum when the vane points to S. and 
S.W., diminishing again when the breeze is from the W. and the N.W. 
The cause of these differences is very simple. Before reaching us, the 
west winds pass over the Atlantic, and are loaded with vapor, whereas 
those which blow from the east come from the interior of Europe and 
Asia. These vapors resolve themselves into rain when the west winds 
reach France ; but this water is vaporized almost immediately, and the 
result is that these winds continue to be more charged with vapor than 
those which come from the east. The W.S.W. wind, blowing both 
from the sea and from warmer countries, is capable of containing a 
larger quantity of vapor of water than the west wind, which is colder. 
This is not the case in regard to relative humidity. 

Thus, although with a north wind the air may contain a much small- 



er proportion of vapor of water than when the wind is south, it is far 
more humid, because of its low temperature. The seasons again modify 
this general rule. The following is the influence of the wind for each 
season, complete saturation being represented by 1000 : 



















The contrast here shown between winter and summer is striking. 
Although in these two seasons the proportion of vapor is less with an 
easterly than with a westerly wind, nevertheless the low temperature 
of these winds in winter re-establishes the equilibrium, and in this sea- 
son the east wind is the most humid and the west the driest. In sum- 
mer it is just the contrary; it is when either of these winds begins to 
blow that the contrast is the most striking. If, for instance, in winter 
the westerly winds have prevailed for some time, the sky being clear, 
and there suddenly springs up an east or a north-east breeze, then the 
sky becomes cloudy, and the lower regions of the atmosphere become 
filled with mist. But if the wind continues to blow, then the sky be- 
comes clear again, although the air remains moist. If the reverse takes 
place — that is to say, if the sky is overcast, the wind being in the east, 
and if it suddenly veers round to the south — the sky becomes clear and 
the atmosphere dry, the reason being that the heated air dissolves the 
vapor of water and becomes further removed from the point of satura- 
tion. It is only when this wind has prevailed for several days and 
collected a large quantity of vapor that the atmosphere again becomes 

We will now consider the force and velocity of the wind. It is at 
times very gentle, and at others extremely powerful. No other ele- 
ment is so capricious and so changeable ; none so capable of soft ca- 
resses or of wild rage. The scale of its variations is so extensive that 
it is difficult to give a very exact account of its range, from the breeze 
which scarcely raises a ripple on the surface of a lake to the hurri- 
cane which uproots trees and throws down buildings. The following 



table will give an idea of the different degrees of velocity which it ac- 
quires : 


Velocity per 
Second near- 
ly in Feet. 


Velocity per 

Hour in Miles. 































Scarcely perceptible wind 

Perceptible wind 

Light breeze 

Moderate wind 

Good breeze 

Fresh wind (swelling sails) 

Wind that causes windmills to revolve 

Good sea-breeze 

Strong breeze 

Very fresh (reefing top-sails) 

Violent wind 


Hurricane that blows down buildings 

Maximum speed of a cyclone's rotation 

Maximum of the rotation and of translation as well. 

It is not known to what degree of speed masses of air borne off by 
cyclones may attain, for it is in the upper regions of the atmosphere, 
where there is but a feeble resistance to aerial currents, that the wind 
of the tempest must be most rapid. Therefore it is not enough to as- 
certain the rate of speed of the moleculee of air near the level of the 
ground in order to form an idea of the rapidity at which the atmos- 
pheric mass moves when hurried along by the tempest. I have re- 
marked in my aerial travels that the speed of air generally increases in 
proportion to the height.* The balloon which, during the siege of 

* [On March 31, 1863, the balloon left the Crystal Palace, Sydenham, at 4 hrs. 16 min. p.m., 
and fell at Barking, in Essex, a point fifteen miles from the place of ascent, at 6 hrs. 30 min. 
P.M. Leaving out of the calculation all motion of the balloon, excepting the distance between 
the places of ascent and descent, its hourly velocity was seven miles ; the horizontal movement 
of the air at Greenwich, as shown by Robinson's anemometer, was five miles per hour. 

On April 18, 1863, the balloon left the Crystal Palace at 1 hr. 16 min. p.m., and descended 
at Newhaven at 2 hrs. 46 min. The distance is about forty-five miles passed over in an hour 
and a half, or at the rate of thirty miles per hour. Eobinson's anemometer had registered 
less than two miles per hour. 

On June 26, 1863, the balloon left Wolverton at 1 hr. 2 min. p.m., and fell at Littleport at 
2 hrs. 28 min. p.m. The distance between these two places is sixty miles; the velocity was 
therefore forty-two miles per hour. The anemometer at Greenwich registered ten miles per 

On July 11, 1863, the balloon left the Crystal Palace at 4 hrs. 53 min. p.m., and fell at 
Goodwood at 8 hrs. 50 min. p.m., having traveled seventy miles, or at the rate of eighteen 
miles per hour. The anemometer at Greenwich registered less than two miles per hour. 

On July 21, 1863, the balloon left the Crystal Palace at 4 hrs. 52 min. p.m., and fell near 


Paris, traveled from that city to Christiania accomplished the distance 
(nearly 1000 miles) in fifteen hours, or at the rate of 66^ miles per 
hour; and this, although there was but little wind on the ground. The 
balloon sent up from Paris at the coronation of Napoleon, in 1804 (at 
11 P.M.), carried the news of the Pope's submission to the emperor di- 
rect to Eome, reaching that city at seven the next morning, having 
done the 800 miles at an average hourly speed of 100 miles ! These 
facts serve to give us an idea of the speed of the cyclone at a certain 
height above the ground, when even along the earth, which is covered 
with points of resistance to it, its rapidity is as much as 100 miles in 
the hour, and upon the ocean 150 to 170 miles. 

As to the pressure exercised by the aerial current which moves at so 
great a rate, it is indeed formidable. In a notice upon the construction 
of light-houses, Fresnel calculated that the highest wind-pressure was 
sixty pounds on a square foot, but it is very probable that in many 

Waltham Abbey, having traveled about twenty-five miles in fifty-three minutes, or at the rate 
of twenty-nine miles per hour. The horizontal movement of the air by Robinson's anemome- 
ter was at the rate of ten miles per houi-. 

On September 29, 1864, the balloon left Wolverhampton at 7 hrs. 43 min., and fell at Slea- 
ford, a point ninety-five miles from the place of ascent, at 10 hrs. 30 min. a.m. During this 
time the horizontal movement of the air was thirty-three miles, as registered at Wrottesley 

On October 9, 1864, the balloon left the Crystal Palace at 4 hrs. 29 min. p.m., and descend- 
ed at Pirton Grange, a point thirty-five miles from the place of ascent, at 6 hrs. 30 min. p.m. 
Robinson's anemometer during this time registered eight miles at the Royal Observatory, 
Greenwich, as the horizontal movement of the air. 

On January 12, 1865, the balloon left the Royal Arsenal, Woolwich, at 2 hrs. 8 min. p.m., 
and descended at Lakenheath, a point seventy miles from the place of ascent, at 4 hrs. 19 
min. p.m. At the Royal Observatory, by Robinson's anemometer, during this time the mo- 
tion of the air was six miles only. 

On April 6, 1865, the balloon left the Royal Arsenal, Woolwich, at 4 hrs. 8 min. p.m. Its 
correct path is not known, as it entered several different currents of air, the earth being invisi- 
ble, owing to the mist; it descended at Sevenoaks, in Kent, at 5 hrs. 17 min. p.m., a point 
fifteen miles from the place of ascent. Five miles was registered during this time by Robin- 
son's anemometer at the Royal Observatory, Greenwich. 

On June 13, 1865, the balloon left the Crystal Palace at 7 hrs. min. p.m., and descended 
at East Horndon, a point twenty miles from the place of ascent, at 8 hrs. 15 min. p.m. Rob- 
inson's anemometer during this time registered seventeen miles at the Royal Observatory, 
Greenwich. * 

On August 29, 1865, the balloon left the Crystal Palace at 4 hrs. 6 min. p.m., and descend- 
ed at Wey bridge at 5 hrs. 30 min. p.m., a point thirteen miles from the place of ascent. Dur- 
ing this time fifteen miles was registered by Robinson's anemometer at the Royal Observatory, 
Greenwich. — Ed.] 


cases this is exceeded. Leaving out of the question the effects of 
strong cyclones in the tropics, several cases have occurred in the tem- 
perate zones where the pressure exercised by the wind in a very limited 
space was much above the calculations of meteorologists. To cite only 
one instance, the tempest which occurred on the 27th February, 1860, 
and which blew from the west in the plains ofNarbonne, was so violent 
as to blow trains off the rails on the line between Salces and Eivesaltes. 
The pressure must have been at least eighty pounds to the square foot. 

It has been calculated that, approximately, the mechanical force of 
the wind is in proportion to the surface of the object exposed to it, and 
in direct ratio to the square of the velocity, and that for a velocity of a 
yard per second, for each square yard, the effect produced is about 
a quarter of a pound. With strong winds, the velocity of which is 
twenty yards per second, there is a pressure of ten pounds per square 
foot; when, as in hurricanes, the speed is forty yards, the pressure be- 
comes quadrupled. This renders it easy to understand how trees are 
uprooted and houses blown down. 

The extreme smallness of the molecules of air is often more than com- 
pensated by the rapidity of their motion, so that they are capable of 
producing effects which appear incredible, but which are in conformity 
with the laws of mechanics. 

To give a correct idea of these effects, I may anticipate the chapter 
upon Cyclones, and cite a few of the great disasters caused by certain 
hurricanes. At Guadaloupe, on July 25th, 1825, solidly - constructed 
houses were demolished, and a new building belonging to the State had 
one wing completely blown down. 

The wind had imparted such a rate of speed to the tiles, that many of 
them penetrated through thick doors. 

A piece of deal, thirty-nine inches long, ten inches wide, and nearly 
one inch thick, moved through the air so rapidly, that it went right 
through a palm-tree, eighteen inches in diameter. 

A piece of wood about eight inches wide, and four or five yards long, 
projected by the wind along a hard road, was driven a yard deep into 
the ground. 

A large iron railing, in front of the Governor's Palace, was shattered 
to pieces. 

Three twenty-four pounders were blown from one end of a battery to 
the other. 

In 1823, a hurricane, about half a mile in diameter, passed close by 


Calcutta, killed in the space of four hours 215 persons and wounded 
223, blew down 1239 fishermen's huts, and drove a piece of bamboo 
through a wall five feet in thickness : the blast of the air must have had 
a force equal to that of a six-pounder cannon. 

At St. Thomas, in 1837, the fortress which protects the entrance into 
the harbor was demolished as if by bombardment. Fragments of rock 
were projected from a depth of thirty to forty feet, and hurled on .the 
shore. In other places, strong houses, torn up from their foundations, 
were swept along the ground before the wind. On the banks of the 
Ganges, the Antilles coast, and at Charlestown, several vessels were car- 
ried from the sea some distance inland. In 1681, an Antigua vessel was 
carried out of the water to a point ten feet above the highest known 
tide. In 1825, the vessels which were in the harbor of Basseterre dis- 
appeared, and one of the captains, who had escaped, said that his ship 
was lifted by the hurricane out of the sea, and was, so to speak, " ship- 
wrecked in the air." A quantity of the debris from Guadaloupe was 
carried to Montserrat, over an arm of the sea fifty miles wide. In the 
tempest which blew across the English Channel on January 11th, 1866, 
stones weighing from four to six hundred pounds were hurled over the 
Breakwater at Cherbourg to a height of more than eight yards. Ad- 
miral Le Noury states that the sea dashed against the fort, which is 185 
feet above the level of the shore. 

The only difficulty in explaining these phenomena is to discover how 
the air can attain in the atmosphere so prodigious a velocity ; for, 
granting that velocity, the most extraordinary chemical action becomes 
the necessary consequence. It is gas in motion which drives the can- 
non-ball from the gun, and which hurls into the air vast masses of rock 
when a mine explodes. An oak olank, nearly an inch thick, may be 
pierced by a candle fired out of a gun ; the force of the projectile being 
only due in this case to its velocity. 




Having considered the theory and the action of the general winds 
(both those that are regular and irregular) which blow over the surface 
of the globe, we must now turn our attention to special winds which 
characterize certain countries, and to atmospheric movements which at 
times traverse oceans and continents with the rapidity of a bird of prey, 
and which seem to form an exception to the system of organized laws 
by which nature is regulated. Scientific analysis has shown that these 
phenomena are obedient, like every thing else in the universe, to defi- 
nite and fixed laws. 

In France, the temperate climate which we enjoy precludes the in- 
tense atmospheric phenomena which occur in less favored regions. 
Among the winds, properly so called, which differ slightly in their char- 
acter from most of the general winds, may first be cited the hise^ or 
north wind, which is very cold, and occasionally very violent. In the 
east of France it is much dreaded, for it comes nearly in a straight line 
from the North Sea ; and having traversed Holland and Belgium when 
those countries are covered with snow, it becomes even colder during 
its passage. At Istria and in Dalmatia the bise is known as the hora, 
and it is so strong that it sometimes blows over a horse and cart. In 
Spain, this same north wind — which is sometimes a north-east wind in 
that region — is designated the gallego. 

In the south of France, the cold and violent south-west wind which 
has passed over the snows of the Alps and the Pyrenees, and which is 
known as the mistral^ deserves particular notice. 

Its cause was long unknown. It was attributed to a sudden coldness 
of the wind that passed over the Alps and the Pyrenees. M. Marie- 
Davy, in several notes published in the Bulletin de T Ohservatoire in June, 
1864, proved that the cause of this wind is not local, and that the move- 
ments which give rise to it pass eastward like whirlwinds. Kaemtz, in 
a communication to the Institute in July, 1865, shows, by means of a 


list of barometrical pressures in France, Spain, and Italy, before, during, 
and after the passage of the mistral, that it is a regular tempest, coming 
from a great distance, and that it is not due to a sudden fall in the tem- 
perature of the wind while passing over the mountains. 

It is remarkable that, in proportion as meteorology advances, we learn 
not to look for the causes of most phenomena in the localities where 
they occur, but to general preponderating causes to which the local cir- 
cumstances are subordinate. 

Whenever the mistral blows, there is an excess of atmospheric press- 
ure to the west of the Gulf of Lyons. Whatever may be the origin of 
this pressure, it always is an accompaniment of the mistral. 

For the mistral to occur, no matter in what season, there must always 
be a combination of circumstances which are always identically alike. 
Whether there be fine or bad weather in the south-west of Europe, 
there is always an excess of pressure to the west of the Cevennes. 

The violence of the wind is due to the form of the Pyrenean isthmus. 
As soon as the general direction of the atmospheric movement veers 
slightly from west to north, the central plateau and the main body of 
the Alps cause an inclination of the current toward the Gulf of Lyons. 
This current, compressed between the Alps and the Pyrenees in the di- 
rection of length, and by the Cevennes in a vertical direction, consti- 
tutes a rapid upon the coast of Languedoc, and this is one of the causes 
of the excess of pressure upon the north-east slope of the Cevennes, and 
of the diminution in pressure upon the Mediterranean, where the wind 
maintains a velocity no longer commensurate with the width of the 
channel. Hence also arises the violence of the north wind in the valley 
of the Rhone between the spurs of the Alps and those of the central 

The mistral is the driest of the winds in these regions, because it has 
been rendered dry in its passage over the Cevennes. It is, indeed, plu- 
vious or moist upon the north-western slope of those mountains. The 
winds from the east or south regions bring rain with them, because they 
are sea-winds upon the coasts and upon the south-western slope of the 
Cevennes ; they are dry upon the opposite slope. 

The high temperature of the interior of Africa is the cause of the ex- 
traordinary winds which are met with on the coasts of Guinea and Bar- 
bary, in Egypt, Arabia, Syria, the steppes of Southern Russia, and even 
in Italy. These winds, the names of which are harmattan, simoom, and 
kharaseen, have unusual accompaniments, some details of which it may 


be interesting to mention. They are remarkably hot and dry, and are 
attended by whirlwinds of dust. 

The name harmaitan is given to a wind which blows three or four 
times each season from the interior of Africa toward the Atlantic, in 
the space comprised between Cape Verd (lat. 14° 44' N.) and Cape Lo- 
pez, on the African coast near the equator. The harmattan is generally 
noticed in the months of December, January, and February. Its direc- 
tion is from E.S.E. to N.N.E., and its ordinary duration is one or two 
days — sometimes five or six. This wind is only moderately strong. 
A peculiar kind of mist, so thick as to shut out all but a few red rays of 
the sun at noon, always rises when the harmattan begins to blow. The 
particles of which this mist is composed alighting upon turf, the leaves 
of trees, and the skin of the negroes, make every thing white. Their 
nature is not known, and all that has been ascertained respecting them 
is, that they are carried but a very little way out to sea : at a league 
from the shore, for instance, the mist is very slight, and at three leagues 
there is no trace of it, though the harmattan may be blowing with its 
full force. 

The extreme dryness of the harmattan is one of its most marked char- 
acteristics. If it continues any length of time, the branches of the or- 
ange and lemon trees become parched and begin to die ; the covers of 
books, not even excepting those which are wrapped up in linen and in- 
closed in a case, become bent as if they had been laid before a fire. The 
panels of doors and windows, the furniture of rooms, crack, and often 
snap. The effect of this wind upon the human body is not less pro- 
nounced. The eyes and the lips dry up and smart. If the continuance 
of the harmattan be for four or five consecutive days, the cuticle of the 
hands and the face begins to peel, and it is necessary to anoint the body 
with grease. 

All this would lead one to suppose that the harmattan must be very 
unhealthy ; but, so far from this being the case, it is the opposite. In- 
termittent fever, for instance, is radically cured by the first breath of 
the harmattan. Persons who have become weakened by the excessive 
blood-letting practiced in those countries at once recover their strength; 
remittent and epidemic fevers also disappear as if by enchantment. So 
salutary, in fact,, is the influence of this wind while it lasts, that it is 
said to be impossible to communicate infection even artificially, for it 
appears that vaccine virus will not act during its continuance. 

Its asserted poisonous properties are therefore pure invention, and 


may possibly have been circulated by the Arabs to deter travelers from 
penetrating into what they consider their kingdom. 

Kaemtz says: "At every epoch the Arab of the desert, poor and of 
nomad habits, has detested the inhabitant of towns who leads a steady 
and orderly life. Thus, when the merchant is compelled to cross the 

desert, the Bedouin exacts an enormous price for protecting him 

In the eyes of the inhabitant of a town the desert was the theatre of 
scenes horrible beyond description. All the marvelous stories of ad- 
venture found a ready audience, just as in our days the Turks have the 
most grotesque and unreal ideas concerning Europe. The dwellers in 
the desert took care not to destroy these fancies, but rather to confirm 
them, and the few merchants who knew the exact truth kept it to them- 
selves in order to maintain a monopoly of commerce. It is in this way 
that visionary ideas maintain their sway." 

The Arab writers teem with falsehoods concerning the desert, and the 
European travelers have outrivaled them. The Mussulman believes 
he is acting meritoriously when he deceives the unbelievers and keeps 
them away from the desert. L. Burckhardt, of Bale, was the first who 
supplied reliable information concerning the phenomena of the desert, 
and especially touching the winds which prevail there. He thus re- 
duced to their true value the fabulous stories of his predecessors, Beau- 
champ, Bruce, and Niebuhr. 

Burckhardt relates that this wind of the desert surprised him between 
Siout and Esne. He says, " When it rose I was alone, mounted upon 
my dromedary, and far away from any houses or trees. I endeavored 
to protect my face by covering it with a handkerchief. In the mean 
while, the dromedary, into whose eyes the sand was driving, became 
alarmed, and began to gallop, causing me to fall off. I remained flat 
upon the earth, for I could not see ten yards in front of me, and I cov- 
ered myself with my clothes as well as I could until the wind became 
less violent. I then went in search of the dromedary, which I found 
some distance off, lying with his head against a bush to protect it from 
the sand." Malcolm and Morier, who crossed the Persian deserts, Ker- 
Porter, who visited that which lies to the east of the Euphrates, agree 
with Burckhardt that when they were exposed to the simoom they felt 
no ill effects from it beyond the momentary inconvenience. 

" It is not only in the sandy deserts of Africa and Asia that the hot 
winds are to be dreaded, but in nearly all continental countries near 
the tropics. In India they are known as the 'devil winds.' They fre- 



quentlj occur during the dry season, and scatter terror and desolation 
through country and town. Without being absolutely poisonous, it 
may be admitted that winds whose speed is so formidable, laden witb 
grains of sand, and the temperature of which is as much as 104°, may 
exercise an unhealthy influence upon the regions through which they 
pass, and be especially dangerous for Europeans who do not know how 
to protect themselves." 

About the time of the equinox the tempests in the desert become ter- 
rible. Every one has heard of the burning wind, the simoom — a wind 
which, in Arabia, means poison. This formidable wind blows also in 
Egypt, where it is called khamseen (fifty), because it lasts that number 
of days, five-and-twenty before, and five-and-twenty after the spring 

The simoom is preceded by a black spot which rises in the horizon. 
This spot grows rapidly larger. A murky veil obscures the sky, gusts 
of sand darken the sun and dry up all verdure. As soon as it begins 
to blow, birds fly off affrighted ; the dromedary seeks a bush to pro- 
tect him from the sand ; the Arab covers his face, rubs his body with 
grease or wet mud, and lies on the ground or hides himself behind a 
tree until it is over. The simoom is the most dangerous of the acci- 
dents to which a caravan crossing the desert can be exposed, and to it 
is attributed the destruction of the 50,000 men sent by Cambyses to 
destroy the Temple of Jupiter Ammon. 

In 1805 the simoom buried in the sand a whole caravan, causing the 
death of 2000 men and 1800 camels. More than once French generals 
have feared that columns of troops which they have been obliged to 
send into the desert have been overtaken and destroyed by the simoom. 

The impalpable dust which the air carries along in thick clouds 
enters the nostrils, the eyes, the mouth, and the lungs, and causes as- 
phyxia. "When it does not absolutely kill, the rapid evaporation from 
the surface of the body dries up the skin, inflames the throat, makes 
the breathing rapid, and produces violent thirst. The terrible blast of 
the simoom dries up the sap of trees in its passage, and causes the 
water contained in the skins carried by the camels to evaporate. The 
caravan is then a prey to horrible thirst, which sets the blood on fire, 
and the route which they follow is strewed with the whitened bones of 
men and animals who perish for want of water. 

Thomas William Atkinson was a witness, in 1850, of the fierce hur- 
ricanes which swoop down upon the steppes of Mongolia. He says, 

'•;M I 


"A solemn silence prevails in these vast and arid plains, which are de- 
serted alike by men, quadrupeds, and birds. People talk of the soli- 
tude of forests. I have often ridden through their sombre alleys for 
days together: the soughing of the wind, the cracking of the branches, 
and the murmur of the leaves are to be heard there; sometimes, too, 
the crash of one of the giants of the forest as it falls to the ground 
wakes the distant echoes and startles from their lairs the tenants of the 
glades, making the birds utter a cry of terror. This is not solitude : 
trees and leaves have a language which man recognizes from a great 
distance; but in these arid deserts there is no sound to break the 
death-like silence which prevails. 

" The sand was raised into circular terraces, some from fifteen to 
twenty feet high, and they extended as far as the eye could reach into 
the desert. Seen from one of the knolls they presented the singular 
appearance of an immense necropolis, over which were dotted countless 

" While I was taking a sketch of this, I was witness of the formation 
of a hurricane above the level of the water. It was coming from the 
north direct upon us. The Cossacks and Tchuck-a-boi went to place 
their horses in security behind the reeds and bushes, leaving two of 
their companions with me. The tempest came on at a fearful rate, 
driving enormous waves into the air, and striking down all vegetation 
in its path. A long white wave came moving along the lake, and 
when it was within half a verst its roar became audible. The men 
begged me to move away, and we rejoined the rest of the troop behind 
the bushes. I had scarcely reached there when the hurricane burst 
forth, bending the bushes to the ground. When it reached the sand it 
began to revolve in a circle, lifting whole mounds of sand into the air, 
and causing others to spring up where there had been none previous to 
the storm. This tempest lasted a quarter of an hour, when the atmos- 
phere became as calm as it had been before. 

" It is very dangerous to be overtaken in the plain by one of these 
typhoons. I have since seen them swoop down from the mountains or 
rise from the hollow of some deep ravine, in the shape of a black and 
compact mass, with a diameter of as much as 1000 yards or more, and 
rushing along the steppe with the speed of a race-horse. All animals, 
whether tame or savage, take flight before it, for once enveloped in its 
sphere of action they must infallibly perish. The wild horses gallop 
off in terror before the storm which pursues them." 


In Europe we have the sirocco (Italy), and the solano (Spain), both 
of which have a very enervating effect upon those exposed to them. 

Brydone, who was at Palermo on July 8th, 1770, during a sirocco, 
writes, " I opened my door at eight in the morning without suspecting 
there was any change in the temperature, when all at once I felt a 
burning impression upon my face like the air from a hot oven. I 
closed my door, exclaiming to Fullarton that all the atmosphere was on 
fire." At this moment the thermometer, in the open air, marked 111°. 

An army surgeon who accompanied the French troops in a march 
between Oran and Tlemcen, in the desert, gives the following account 
of a sirocco : " It was toward the end of July, 1846. Several soldiers 
had succumbed to the heat. The sirocco assailed our little column. 
Under the influence of this dry, heavy, and enervating air, the breath- 
ing became difficult; the lips and the nostrils, cracked by the burning 
dust driven up by the wind, were both dry and painful, and the throat, 
as it were, became contracted. The face was burned by gusts of heat, 
sometimes followed by tremor, and a fainting away which resembled 
syncope. The perspiration ran off in streams, and the water, which 
was eagerly swallowed, did not quench the thirst, but increased the 
stomachic pains and the difficulty of breathing. To walk would have 
been impossible ; we felt half suffiDcated under cover of the tents, and 

in the open air the burning breeze caused a choking sensation 

But for the water, our column must have perished." 





The two great general currents which have been adverted to, the 
one moving from the equator to the poles, and the other from the poles 
to the equator, come into collision with each other in the equatorial 
zone. Various causes counterbalance the periodical action of the solar 
rays, and place obstacles in the way of the ordinary progress of air. 
The diversity in the temperature of continents and seas causes a varia- 
tion in the normal direction and intensity of the currents. The state 
of the sky in the tropics, according as it remains clear or cloudy for 
any length of time together, condenses the heat as in a focus of absorp- 
tion, or disseminates it over vast tracts of country. The undulations 
of the soil, the high chains of mountains and their temperature, the less 
lofty plateaux, and even the valleys themselves, cause in one place the 
storing up and the repose of large masses of air, in another their dis- 
tribution in different directions, while in other cases this same uneven- 
ness of the ground forces the currents back right and left, causing them 
to eddy like the waters of a stream, or to rush furiously past the ob- 
stacles in their way. The blasts of the air as they meet, either join 
forces or oppose each other, increasing or destroying their mutual 
power. It is in this way that strong winds, -hurricanes, and tempests 
arise. These atmospheric contentions, which sometimes attain gigantic 
proportions, create a great disturbance in the course of nature. They 
have been studied by sailors and meteorologists, and the principal laws 
which seem to regulate them have been ascertained. Eedfield and 
Reid, Professor Dove, of Berlin, and Admiral Fitzroy, have, after great 
labor, succeeded in forming a theory of the tempests which explains 
the most violent of the movements in the atmosphere, and their re- 
searches will be useful in considering this subject here. 

One of the chief observations made is, that hurricanes do not proceed 
in straight lines, but follow a curve, turning horizontally upon their 
own axes by a rapid rotatory movement. 

This characteristic movement of horizontal rotation has earned for 


these gigantic whirlwinds the name of cyclone (kvkXoc, circle). They 
are the general hurricanes, which are not local tempests resulting from 
the deviation of the wind, owing to the configuration of the soil or the 
meeting of several ordinary currents, but extend over several hundred 
square leagues, and travel a distance of many thousand miles. 

The cyclones are vast whirlwinds, of various size in diameter, in 
which the force of the wind increases from all the points of the circum- 
ference to the centre, where a calm prevails. In this centre, however, 
the sea remains rough. There is no cloud in this calm region ; the sun 
shines brightly, the stars appear, and fine weather seems to have return- 
ed, when in reality it is surrounded on all sides by a vast belt of fierce 

All around this central calm the rotatory movement is of the same 
force, and this force is the greatest. Consequently, on passing to this 
central region, a ship passes from a violent tempest into a complete 
calm, and, on crossing this calm space, passes on the opposite side into 
a violent tempest again. But in this latter case the hurricane blows in 
the opposite direction to what it did before entering the calm, since the 
movement of cyclones is circular. 

The first central zone, which constitutes in reality a hurricane, and 
during the passage of which occur all the disasters, is generally from 
100 to 120 leagues in diameter, whatever may be the extreme limits 
which the phenomenon reaches, for its power is not in proportion to its 

The rotatory speed of the hurricanes varies very much : it is that 
which constitutes chiefly the violence of the whirlwind, and which 
causes it to be, in regard to the places against which it blows and the 
vessels it assails, either a hurricane, a gust of wind, or a simple gale. In 
violent storms, it is estimated that the moleculas of air turn around its 
centre with a rotatory speed of sixty leagues an hour — a rapidity which 
explains the ravages and disasters produced by the passage of this ter- 
rible wind. 

The cyclone generally begins between the latitudes of 5° and 10°. 
It moves, in our hemisphere, in a north-westerly direction, continuing 
thus until it reaches a particular latitude, when it turns toward the 
north-east, and thus forms a parabola, the two branches of which di- 
verge farther and farther. 

The difference in the density of the different atmospheric strata 
which are encountered in its passage, the rotatory movement itself. 


must impart an oscillating movement to the cyclone, so that, instead 
of describing a regular parabola, the course of the cyclone is rather 
spiral, infolding itself around the parabola. Ships that happen to be 
in the midst of it are exposed to its oscillating action: hence those 
terrible gales which are succeeded by a more or less complete calm ; 
hence those dramatic situations in which the ship in distress sees the 
wind veer round to all the points of the compass in a short space of 

The sudden and dangerous variations of the wind, which were for- 
merly considered as essential to hurricanes, typhoons, tornadoes, etc., 
can not, and in fact do not, occur save in the immediate path of the 
centre of the cyclone. The cyclone contains in itself the germ of its 
own early destruction : in proportion as it advances it is approaching 
nearer to regions which are colder than those whence it started ; the 
vapor which it contains becomes condensed into torrential rain ; the 
electricity issues from it in large currents ; the equilibrium which ex- 
isted becomes destroyed ; and the centrifugal force, being no longer 
counterbalanced, permits it to extend to an immense size. It then loses 
in force what it gains in volume ; at its starting-point it does not meas- 
ure more than a few leagues, but when, having lost its equilibrium, it 
collapses — as generally occurs between the latitudes of 40° and 45° — it 
extends over hundreds of miles. 

The more rapid the escape of electricity the quicker will the meteor 
collapse: thus it sometimes happens that a cyclone terminates its course 
before reaching these high latitudes, and without describing the second 
branch of its parabola, which therefore remains incomplete. 

Between latitudes 5° and 10°, and longitudes 45° and 60°, when a 
cyclone is near its starting-point, it has been ascertained that the rate 
of revolution is inconsiderable, varying from one to five miles an hour, 
increasing as the hurricane advances westward. 

In latitudes 35° to 45°, and in longitudes from 50° to 30°, the rate 
of revolution varies from six to twelve miles an hour. In the higher 
latitudes it is greater, and has been known to be as much as twenty 
miles per hour. 

The greatest rate of revolution ever registered is that of a cyclone 
which reached the Banks of Newfoundland from the Antilles in August, 
1853, when the speed was thirty-one miles per hour. This velocity 
gradually increased to ninety; and without affecting the speed of rota- 
tion, which was sixty leagues an hour. Thus the wind is capable of 


traveling along the surface of the sea at a speed of seventy-five leagues 
an hour, perhaps more. 

The origin of cyclones, so far as can be judged by the comparisons 
that have been made, is probably due to the encounter of two currents 
of air moving in opposite directions. The place of meeting forms a 
neutral point, where the air receives a rotatory movement from the col- 
lision of the two currents. It is like an eddy in a stream, and a mo- 
ment's reflection will enable the reader to form an idea of it. 

These immense whirlwinds come into existence to the south as well 
as to the north of the equator. The astronomer Poey, Director of the 
Observatory at Havana, has ascertained, by a laborious research into 
the hurricanes which have raged in the West Indies since the discov- 
ery of America (1493) to the present day, that, out of 365 grand cy- 
clones, more than two-thirds have occurred between the months of Au- 
gust and October, that is, during the period when the heated shores of 
South America are beginning to attract toward them the colder and 
denser air of North America. In the Indian Ocean cyclones are most 
frequent when the change occurs in the direction of the monsoons and 
at the end of summer. In the list of hurricanes in the southern hemi- 
sphere, drawn up by Piddington and completed by Bridet, there is not 
a single mention of a hurricane in the months of July or August; 
more than three-fifths took place in the first three months of the year. 
It is at the epoch of change of seasons that the powerful masses of air, 
loaded with electricity, enter into a struggle for the mastery, and give 
rise, by their meeting, to those great eddies which develop themselves 
in a spiral shape over sea and land. At the same time, the whirlwind 
never extends very high into the aerial ocean. According to Bridet, 
the average height of hurricanes in the Indian Ocean is less than 10,000 
feet ; Eedfield puts it at no more than 6000. As a rule, the stratum of 
air which revolves in this way is not nearly so thick ; sometimes it is 
so shallow that the crew of a vessel which is exposed to a cyclone see 
above their heads a clear sky. Above the cyclone the storm-winds 
follow their regular course. 

The analysis of cyclones is especially due to Redfield. America is a 
country peculiarly well adapted for observing these phenomena, as the 
hurricanes which run along the shores of the United States pass, dur- 
ing their progress through the tropics, over the West India Islands, 
where their remarkable character has earned them the appellation of 
" West India hurricanes." 


In regard to the cyclones which occur in Central Europe, it is rarely 
possible to ascertain through what part of the tropics they have passed, 
and this is a sufficient proof that the wider the extent of our observa- 
tions, the less likely shall we be to form incorrect ideas of these phe- 
nomena of nature. 

The meteorologist Dove proved, in his work upon the "Law of Tem- 
pests" (Paris edition, p. 173), that a cyclone movement occurs whenever 
any obstacle stands in the way of the regular change in the direction 
of the wind (which is due to the rotation of the earth), and therefore 
prevents the regular rotation of the vane to some given point. He 

"The hurricanes in the West Indies generally commence at the in- 
ner limit of the zone of trade-winds, in the region of calms, where the 
air rises and becomes disseminated in the upper strata of the atmos- 
phere, and in a direction contrary to that of the trade-wind. This 
renders it probable that the primary cause of cyclones is the intrusion 
of a part of this upper current into that below. 

"Let us also imagine that the air which rises over Asia and Africa 
flows laterally into the upper strata of the atmosphere — a fact which is 
made evident by the sand which falls in the North Atlantic, and which 
rises to a great height, for on the Peak of Teneriffe the sun is some- 
times obscured by it. A similar current must have a tendency to op- 
pose the free passage of the upper anti-current of the trade-winds, and 
must force it back into the lower current, or the direct trade-wind. 
The point at which this intrusion takes place must advance with as 
great rapidity as the oblique upper current which causes it. The in- 
terposition of a current traveling from E. to W. with another traveling 
from S.W. to N.E. must necessarily create a rotatory movement in a 
direction the opposite to that followed by the hands of a watch. Ac- 
cording to that, the cyclone, which advances from S.W. toward N.E. 
in the lower trade-winds, represents the point of contact of two other 
currents which in their higher layers advance in directions perpendic- 
ular to each other. That is the origin of the rotatory movement, and 
the ulterior progress of the cyclone will necessarily be based upon the 
same principles. The cyclone being thus considered as the result of 
the meeting of currents at different points, one after the other, may 
therefore preserve its diameter unchanged for a considerable period, 
and it may even diminish in size, though it ordinarily increases. 

" It is, moreover, quite clear that, if the above explanation be cor- 


rect, a cyclone which turns in the same direction may originate by the 
interposition of some mechanical obstacle in the route of the current, 
as it travels toward the high latitudes of the north — an obstacle which 
compels this current to assume a more southerly course (that of a south 
wind) upon its eastern than upon its western edge, where it always re- 
mains nearly due west. This was what happened, to cite one instance, 
during a cyclone in the Bay of Bengal on the 3d, 4th, and 5th of 
June, 1839." 

The name of cyclone is therefore, in a certain measure, the geomet- 
rical designation of the more ancient term hurricane^ like the tornadoes 
which are seen on the coasts of Africa, like the typhoons of the Chinese 
seas; the great tempests that occur in these regions are of the same 
kind as the cyclones in the Atlantic. Dampier, the prince of navigators, 
describes the approach of the typhoon with that accuracy which ren- 
ders all his works so reliable. We read in his "Voyages" (vol. ii., p. 26): 

" The typhoons are a special kind of violent tempests which blow 
along the coast of Tonquin and the neighboring shores in the months 
of July, August, and September. They generally occur about the 
period of full moon, and are, as a rule, preceded by very fine weather, 
light breezes, and a clear sky. These light breezes are the ordinary 
trade-winds, which blow from the S.W. at this season, and which veer 
to about N. or N.E. Before the tempest begins, a thick cloud forms in 
the N.E. ; it is very black near the horizon, copper-colored toward the 
summit, and gradually lighter in color toward its outside edge, which 
is perfectly white. The aspect of this cloud is very strange, and it ap- 
pears sometimes twelve hours before the storm breaks. When it be- 
gins to move very rapidly, the wind breaks out at once, augmenting in 
force with great suddenness, and blowing with great violence for about 
twelve hours. It is often accompanied by thunder and lightning, and 
thick rain. When the wind begins to diminish, it drops very sudden- 
ly for about an hour, after which it recommences to blow from the 
S.W. for about the same period as it did from the N.E., rain falling 
as before." 

The course followed by the centre divides the hurricane into two 
equal parts, but into parts which differ from each other. In the one 
the movement of rotation and that of translation have the same direc- 
tion ; in the other, on the contrary, the direction of translation and 
that of rotation are different. It follows that at an equal distance from 
the centre there is much more wind in the first hemicycle than in the 


second ; hence the name of daiigerous hemicycle is given to the one, and 
that of manageable hemicycle to the other. 

In the northern hemisphere the cyclone turns from right to left — 
that is to say, that an observer placed in the centre of the whirlwind 
would see the wind pass before him from right to left. The dangerous 
hemicycle will be to his right if he follows the same route as the centre 
of the hurricane, and the manageable hemicycle to his left. 

In the southern hemicycle, on the contrary, the hurricane turns from 
left to right ; the dangerous hemicycle is to the left, and the manage- 
able hemicycle to the right of the line through which the centre passes 
if he follows the same direction as the hurricane. 

The direction of the wind observed at a given point of the cyclone is 
very near to a tangent drawn to the concentric circle upon the circum- 
ference of which it is placed. Consequently, it is always nearly per- 
pendicular to the radius drawn from this point to the centre of the con- 
centric circle or cyclone. Now, the law of gyration indicates that if 
one faces the wind the centre will be to the right in the northern hem- 
isphere, and to the left in the southern, but always at right angles to 
the direction of the wind. 

It is upon this latter fact, which numerous observations place beyond 
a doubt, that all the theories as to the means of avoiding the centre of a 
cyclone, by moving away from the line which it takes, are based. The 
nearer the centre, the more violent the wind, and the more sudden its 
variations. The sea is also roughest at the centre, being subject, at very 
short intervals, to violent gusts of wind from all directions — and this 
after having been under the influence of comparatively regular winds 
which have had time to cause a heavy swell, and to impart to the water 
a direction different from that of the wind. Hence arise the short and 
chopping waves which assail a vessel on all sides at once. 

It is easy, however, to avoid the place over which the centre of a cy- 
clone passes. 

Let us suppose that the centre of a cyclone is coming toward a vessel. 
It will pass over this vessel, or to the right or to its left. If it is about 
to pass over, its direction in respect to the ship will not change; but 
then the direction of the wind, which is always perpendicular, will not 
change either, and the crew will find the wind increase in violence with- 
out changing direction. 

If the centre pass to the right of the vessel, it will shift slightly to- 
ward the right. Its direction will vary from left to right; but that of 

334 ^^^ ATMOSPHERE. 

the wind, which is connected with the first, will vary in the same direc- 
tion — that is, from left to right. 

The exact opposite will take place if the centre pass to the left of the 

Thus, if the wind increases without changing its direction, the vessel 
will be upon the line along which the centre passes ; if the wind veers 
from left to right, it will be to the left of this line ; if the wind changes 
from right to left, it will be to the right of the same line. 

From these laws regulating cyclones, we may gather that the worst 
position in which a vessel can be is that which leads to the centre of the 
hurricane, and, to avoid this, all efforts should be directed. 

The premonitory signs of the cyclone are : Some days before the hur- 
ricane, both at sunrise and sunset, the clouds assume a reddish and or- 
ange hue, which becomes reflected upon the sea ; and it is this which 
renders them so brilliant and splendid, and which inspires with such 
sentiments of admiration those who do not dream of the imminent dan- 
ger which they foreshadow. 

As the cyclone approaches, this reddish tint gets deeper in color; 
then a black and deep band extends across the sky ; the edges of the 
cumulus are of a copper hue, imparting to the sea and the land an analo- 
gous glitter which makes the atmosphere look as if it were on fire ; the 
sea-birds fly rapidly inland to seek shelter from the fury of the tempest 
which they have an instinct is coming, thus hoping to escape the death 
which would overtake them at sea. 

But of all the premonitory signs of the tempest, the surest and the 
easiest to interpret is the movement of the mercury of the barometer. 

As the pressure of the air gradually diminishes from the circumfer- 
ence to the centre of the whirlwind, the approach of the phenomenon is 
always made evident by a fall of the barometer. This same symptom 
characterizes the tempests in our temperate regions, which are in reali- 
ty, so to speak, the continuations of the oceanic cyclone. 

The barometer begins to fall twelve, twenty - four, and even forty- 
eight hours before the arrival of the cyclone. 

An oppressive calm, accompanied by a suffocating air, prevails for 
four-and-twenty hours ; Nature seems to be collecting all her strength 
to accomplish the work of devastation. 

Whatever may be the course taken by the hurricane, the point near- 
est to its centre is known when the barometer reading ceases to de- 
crease. Then, for a space of two or three hours, the reading of the ba- 



rometer will rise and fall every half-hour, without making any decided 
movement up or down, this being a certain sign of proximity to the 
centre, that the heaviest blasts have been felt, and that the violence of 
the storm will gradually abate. 

The total decrease of the barometer is proportionately greater as the 
central rarefaction is more complete, and this rarefaction, chiefly caused 
by centrifugal force, augments in ratio to the increase of the rotatory 
movement, which causes hurricanes to be so violent. The barometer, 
therefore, declines in proportion as the violence of the wind becomes 
more intense, and the most disastrous hurricanes are those which influ- 
ence it to the greatest degree. 

The rarefaction of the atmosphere at the centre of cyclones is clearly 
proved by the following table, taken from the register of a barometer 
during the hurricane that passed over St. Thomas on the 2d of August, 
1837, when the central calm occurred at 8 P.M. : 

August 2, 6 
" 2 



.5 -45 








A.M 29-922 

p.M 29-764 

" 29-646 

" 29-489 ^ 

" 29-292 



28-268 3 August 3, 
Variation, 0-89 inch! 



August 2, 7-50 P.M 28-032") g 

" 8-20 " 28-032J o 

" 8-22 " 28-386 

" 8-38 " 28-583 

" 8-50 " 28-780 

" 9 " 28-938 

" 9-25 " 29-213 

" 9-50 " 29-410 

" 11 " 29-607J 

2 A.M 29-725 

9 " 29-922 


These large perturbations of the air are perhaps, next to great vol- 
canic eruptions, the most fearful phenomena that take place upon the 
globe, and, as Keclus remarks in his work upon the "Earth," we can not 
be astonished that, in Hindoo mythology, Eudra, the chief of winds and 
storms, should have become, under the name of Siva, the god of de- 
struction and death. For some days before the outbreak of a hurri- 
cane, Nature, desolate and gloomy, seems to foresee a disaster. The 
small white clouds which travel in the air with the anti-trade-winds are 
concealed by a yellowish vapor ; the stars are surrounded by halos with 
a vague iris, and by heavy banks of clouds which, about evening, are 
beautifully tinted with purple and gold. The air is suffocating, as if it 
issued from the mouth of a furnace. The cyclone, which is already re- 
volving in the upper regions, gradually descends. Jagged remnants of 


reddish or black clouds are borne furiously along by the tempest, which 
plunges rapidly through space, and the column of mercury descends in 
the barometer. Soon an obscure mass becomes visible in the stormy 
part of the sky, and, increasing in size, gradually covers the firmament 
with a veil of darkness and a blood-red glitter. This is the cyclone, 
which is swooping down upon the earth, and a terrible silence succeeds 
the moaning of the sea and of the skies. 

In the early part of the cyclone a strange, dull sound is sometimes 
heard, " with a noise like that of the wind in very old houses during 
winter nights." — Piddington. The gusts which rend the air during the 
time the cyclone continues are said to create a noise like that of the 
roaring of wild beasts, a tumult of countless voices, and cries of terror. 
At the points where the centre passes, a formidable sound like the dis- 
charge of artillery, an incessant rolling of thunder (the voice of the hur- 
ricane, as it in fact is), is heard above all others. 

The progress of the winds meets with a certain degree of resistance 
upon land, but the destruction caused is none the less terrible. Build- 
ings which lie in their path are overturned ; the waters of a stream are 
driven back toward their source; isolated trees are torn up by their 
roots; forests are bent down as if they formed but one compact mass, 
and their branches and leaves are scattered ; the grass is swept off the 
ground. In the track of the hurricane fly countless debris like the flot- 
sam carried along by a stream. Generally speaking, the action of elec- 
tricity is superadded to the violence of the air in motion, and helps to 
augment the ravages of the tempest : sometimes flashes of lightning are 
so rapid that they descend like a sheet of flame ; the clouds, and even 
drops of rain, emit light ; the electric tension is so great that, according 
to Reid, sparks have been seen to fly from the body of a negro. A 
whole forest in the Island of St. Vincent was killed without the trunk 
of a single tree being blown down. In Europe, too, upon the shores of 
Lake Constance, many trees were skinned of their bark, though they 
still remained upright in the ground. 

The most terrible cyclone of modern times is probably that which oc- 
curred on October 10, 1780, which has been specially called the Great 
Hurricane, and which seems to have embodied all the horrible scenes 
that attend a phenomenon of this kind. Starting from Barbados, where 
trees and houses were all blown down, it ingulfed an English fleet 
anchored before St. Lucie, and then ravaged the whole of that island, 
where six thousand persons were buried beneath the ruins. From 


thence it traveled to Martinique, overtook a French transport fleet, and 
sunk forty ships conveying four thousand soldiers. " The vessels disap- 
peared ;^'' such is the laconic language in which the governor reported 
this disaster. Farther north, St. Domingo, St. Vincent, St. Eustache, and 
Porto Eico were also devastated, and most of the vessels that were sail- 
ing in the track of the cyclone were lost, with all on board. Beyond 
Porto Eico the tempest turned north-east toward Bermuda, and though 
its violence gradually decreased, it nevertheless sunk several English 
vessels. This hurricane was quite as destructive inland. Nine thou- 
sand persons perished in Martinique, and a thousand at St. Pierre, where 
not a single house was left upstanding, for the sea rose to a height of 
twenty-five feet, and 150 houses that were built along the shores were 
ingulfed. At Port Eoyal, the cathedral, seven churches, and 1400 
houses were blown down ; 1600 sick and wounded were buried beneath 
the ruins of the hospital. At St. Eustache, seven vessels were dashed 
to pieces against the rocks; and of the nineteen which lifted their anchors 
and sailed to sea, only one returned. At St. Lucie the strongest build- 
ings were torn up from their foundations; a cannon was hurled to a 
distance of more than thirty yards, and men as well as animals were 
lifted off their feet and carried several yards. The sea rose so high that 
it destroyed the fort, and drove a vessel against the hospital with such 
force as to stave in the walls of that building. Of the 600 houses at 
Kingstown, in the Island of St. Vincent, fourteen alone remained intact, 
and the French frigate Junon was lost. 

In the Leeward Islands, the inhabitants of the Government Palace 
took refuge in the centre of the building during the height of the storm, 
thinking that the immense thickness of the walls (nearly a yard) and 
their circular shape would preserve them from the fury of the wind. 
At half-past eleven they were obliged to repair to the cellar, as the wind 
had penetrated everywhere and lifted off the roof. The water rising 
there to the height of more than a yard, they were driven into the bat- 
tery and protected themselves behind cannons, some of which were 
driven from their places by the force of the wind. The hurricane was 
so violent that, seconded by the sea, it carried a twelve-pounder a dis- 
tance of more than 400 feet. (This cannon was, it must be supposed, 
upon its carriage, which had wheels.) By the light of day the country 
looked as it does in midwinter; there was not a single leaf, or even a 
branch, remaining upon the trees. Human passions are quelled in pres- 
ence of such a war of the elements. When the Laurier and the Andro- 



mede were lost at Martinique, the Marquis de Bouille set at liberty the 
five-and-twenty English sailors who had survived the shipwreck, writ- 
ing to the Governor of St. Lucie that he was unwilling to retain prison- 
ers men who had fallen into his hands during a disaster to which every 
one was liable. 

The last memorable tempest is that of March 3, 1869, when the three- 
masted vessel La Lerida^ of Nantes, was lost off Le Havre, on her way 
from Haiti. On March 2, at 10 a.m., this vessel, which for two hours 
had been struggling against a fearful sea, approached the jetty, where a 
tremendous current, the force of which was further increased by the 
north-west wind, raised up an insuperable barrier. The vessel soon felt 
the first shock of the current which, two hours later, would have had 
little effect. Hitherto it had managed to sail with the wind blowing 
aft, and this manoeuvre, diminishing its speed, left it almost at the mercy 
of the hostile elements. A feeling of despair came over the spectators, 
most of whom were seamen. They saw that this movement had grave- 
ly compromised the chances of the vessel's escape. The captain tried 
another effort. He endeavored to luff, so as to run his ship into the 
mouth of the Seine, but this was attempted too late. One last chance 
remained — the two anchors were cast out, but they did not grip ! Still, 
for a moment there seemed room for hope ; the anchors had caught, but 
the heavy sea snapped the chains. It was all over in less time than it 
takes to write these lines: the Lerida, at the mercy of the waves, ran 
against the angle of the bastion, which stove in its poop and bulwarks. 
The only thing that remained was to endeavor to save the crew. For- 
tunately the ship was near enough to land to admit of ropes being 
thrown to them, and all were rescued save two, who, losing their pres- 
ence of mind, clung to a rope that was not strong enough to bear their 
weight. The captain, who had staid on the vessel last of all, had scarce- 
ly left when she went down. 

I may finally add that in the torrid zone, and in climates where tem- 
perature is high, hurricanes are numerous, and extremely violent; in 
our temperate climates they are at once rarer and less violent; and in 
the polar regions, the great atmospheric disturbances, which occur very 
frequently, are limited to winds which, though tempestuous, do not con- 
stitute a hurricane. 




Among the chief phenomena which disturb the apparently regular 
order and harmony of Nature, scattering terror and desolation in their 
paths, there is one remarkable for its peculiar and colossal form, for the 
forces which it seemingly obeys, for the unknown and apparently con- 
tradictory laws which it follows, and for the disasters which it causes. 
These disasters are themselves accompanied by such strange circum- 
stances, that their origin can not be confounded with the other phenom- 
ena which prove so fatal to man. This meteor, fortunately rare in this 
part of the world, is designated in France by the general term trombe. 

Previous to Peltier's explanation of this peculiar atmospheric phe- 
nomenon, it was imperfectly known. We are now able to describe 
with precision its nature and its character; and we know that a trombe 
is a column of air which generally turns rapidly upon its own axis, 
and which revolves comparatively slowly, for, as a rule, a person can 
keep up with it at a walking pace. This whirling column of air is 
both caused and set in motion by electricity. The sometimes violent 
wind which its movement produces, and which acts so disastrously, as 
we shall presently see, is not the result of atmospheric currents upon a 
large scale, as with the cyclones, but is confined to very limited dimen- 
sions. The trombes are often only a few yards in diameter, but their 
force is very great. They sweep the soil over which they pass, de- 
stroying trees and houses so completely that sometimes nothing re- 
mains upright in the track along which they have passed. This phe- 
nomenon generally has its origin as follows : 

By virtue of considerable electric tension, the lower surface of a 
stormy cloud descends toward the earth in the shape of a cylinder, or 
rather of a cone, like a great speaking-tube, the top of which is lost in 
the clouds, while the orifice is relatively close to the surface. This re- 
versed cone may be more or less developed, more or less different in 
shape, according to the special condition of the clouds or the locality. 
^That which is always present is a connecting link of vapor between the 
clouds and the earth. 


Beneath the cloudy column there is a great agitation upon the sea or 
upon the ground. Sailors compare it to a boiling process which would 
emit vapor and streams of liquid sheaves. Upon land the dust of the 
roads and light substances form an analogous kind of smoke. In a 
short space of time the lower whirlwind rises sufficiently high, atid the 
upper column descends low enough to admit of their joining and being- 
fused into one and the same column, which is thicker at its higher than 
at its lower part, and which is often transparent like a tube, within 
which vapor can be seen rising and falling. 

When the centre of waters raised over the sea is more compact, it 
appears like a pillar placed to sustain the descending column. There 
proceeds from this column a noise which varies considerably, from 
what seems like the hissing of a serpent to the noise of heavy wagons 
being driven over stony roads. This noise is much more pronounced 
on land than at sea. 

The germs of destruction seem to be embodied in this singular for- 
mation. The trombe advances slowly, to all appearances, blowing vio- 
lently, writhing convulsively, leaving its mark upon all the productions 
of nature and humanity, and reading into atoms all that oppose its ad- 
vance. The disasters caused by this formidable agent show that its 
pressure is sometimes as much as eighty to one hundred pounds to the 
square foot. Flocks of cattle, men, and even rivers, are lifted to an im- 
mense height. The roofs of houses are carried into the air; walls are 
leveled by the sudden violence of an irresistible pressure. To judge 
of the force of this strange phenomenon, let us consider some of its 
most remarkable effects. 

Take, for instance, two tronibes which were observed to the south of 
Paris, May 16th, 1806, from one to two P.M., and which are particu- 
larly good instances of these phenomena. Peltier copies them from 
Professor Debrnn. They may be termed the Paris tromhes. " The 
first began about one o'clock, and seemed to be at least twelve feet 
wide at its base near the cloud, like that of a cone turned upside down. 
It then became successively fifteen, twenty, and forty feet long. The 
lower it descended the more pointed became its conical form, for, 
when it first left the cloud, it formed a perfect cone. Gradually in- 
creasing in length and decreasing in breadth, it finally became no big- 
ger than a rnan's arm. 

"This whirlwind traveled very slowly toward the south, then west 
and south-west, and seemed to be, suspended over the last houses of 


the Faubourg St. Jacques, then above the plain of Montrouge and 
Montsouris. It was of a gray and white color like ordinary clouds, 
and stood out very clearly against the background of the darker 

" Wkat struck me most was that it formed a long tube, partially semi- 
transparent, gradually making several curves and inflections, something 
like a long flexible piece of gut, in which I saw vapors mounting with 
an undulating movement, like smoke which might ascend a stove-pipe 
in glass. The most curious fact was, that the ascent of the vapors was 
much more marked and active toward the lower part, which was then 
about 300 or 400 feet above the ground. 

"As the cloud which formed the head of the column advanced, the 
main mass described a curve and followed it, becoming elongated by 
1500 or 1600 fathoms, and remaining attached to it. But when the 
column became extremely long, and consequently very slight in vol- 
ume, and when it formed an angle of 20° or 25° with the horizon, then 
the main body of the column began to curl off (or become detached). 
This whirlwind, when its inflection was most pronounced, seemed to 
have its head over Chatillon and its tail over Arcueil ; but while the 
head of the column was moving forward, I remarked that the lower 
part seemed to be attracted by the valley of Arcueil, and that it had 
great difficulty in emerging from it. 

"It lasted for more than three-quarters of an hour, and went off to a 
point at last. Its upper part seemed to me to work its way back into 
the cloud whence it started, though, as it was then at a great distance 
to the S.S.W. of Paris, and very small in volume, I could not afl&rm 
this positively. 

"About twenty minutes after the formation of this whirlwind I saw 
a second, which did not, indeed, present so many marked peculiarities 
as the first, but which was far more majestic in appearance. It was 
produced by a cloud, not nearly so high in the air as that which caused 
the first, and it was visible above the Eue du Faubourg St. Jacques 
and the Observatory. It was of a grayish hue, and was traversed from 
top to bottom by a tube as luminous as the moon. I saw the vapors 
rising and falling in the lower part of it very distinctly. At short in- 
tervals the body of this whirlwind lengthened and shortened, and some- 
times rapidly. It passed before the first, and seemed not to be more 
than from 1600 to 2000 paces to the north ; but the first, just before 
it disappeared, traveled much more rapidly southward. It followed 


about the same direction as the first, and its lower part curved slightly 
toward the west. 

" There was a thunder-clap from a cloud not very far from the whirl- 
winds, especially from the second ; they did not seem to be in any way 
affected by it. We judged, from the loudness of the report, that the 
lightning had struck the ground. Drops of rain as large as a man's 
thumb fell at the point where I was standing, followed by hailstones as 
big as nuts. 

"The second whirlwind gradually made its way back to the cloud 
out of which it had proceeded, and by which it was rapidly re-absorbed. 
It had not lasted altogether more than five-and-twenty minutes." 

These whirlwinds were, as will have been gathered, harmless. They 
do not seem to have touched the ground ; but there is no doubt they 
would have proved more dangerous to any balloon which might have 
approached them. 

We now come to tromhes of another kind, the passage of which along 
the surface of the ground leaves unmistakable traces of their power. 

"At 1-30 P.M. of the 6th of July, 1822, in the plain of Assonval, six 
leagues distant from St. Omer and Boulogne, the clouds, coming from 
different points of the horizon, suddenly effected a junction, and cover- 
ed with one mass the whole sky. Directly afterward, a thick vapor, 
with the bluish hue of burning sulphur, was seen to descend from this 
cloud. It formed a reversed cone, the base of which touched the cloud. 
The lower part of the cone, which reached to the ground, soon formed 
an oblong mass of about thirty feet detached from the cloud, revolving 
very rapidly. 

"As it rose, it emitted a sound like that caused by the bursting of a 
large shell, leaving an indentation upon the ground about twenty-five 
to thirty feet in circumference, and to a depth of three or four feet in 
the middle. When at about a hundred yards from its point of depart- 
ure, and moving in an easterly direction, the whirlwind blew down a 
barn, and shook a solidly-constructed house with the force of an earth- 
quake. On its way, it rooted up a group of very large trees, which 
were found lying in many different directions, showing that the whirl- 
wind was revolving while advancing. Others had their topmost branch- 
es torn off, and some of these were found hanging to the tops of other 
trees sixty or seventy feet from the ground. 

"The whirlwind then went a distance of two leagues without touch- 
ing the soil, tearins; off larsre branches of trees which it scattered right 


and left : reaching the corner of a wood, it carried off the tops of some 
large oaks which were blown over the village of Yendome, situated at 
the foot of the hill to the east of the forest. 

" Globes of sulphurous vapor were from time to time emitted from 
the centre of this whirlwind, and the noise which it made was like that 
of a heavy carriage driven rapidly over paving-stones. Each time that 
a globe of fire or vapor was emitted there was an explosion like that 
of a gun, the wind, which was very violent, adding a wild shriek. Af- 
ter having torn up the soil and every thing which resisted it, the whirl- 
wind rose into the air and went on to a distance of a league and a half, 
where it recommenced its ravages. 

" Thence it reached the valleys of Witernestre and Lambre. In the 
first of these villages, composed of forty houses, only eight were left in- 
tact; and it was noticed that the gables and walls of the houses were 
blown in all directions — showing that the wind had blown from every 

" The disasters which it caused at Lambre were not less extensive. 
Several persons remarked the circular progress of the trombe, its sul- 
phurous hue, and the focus of flaming fire which issued with the sparks 
of bituminous vapor. The trees around the church were broken and 
uprooted, the cure's house carried away, and eighteen others, mostly of 
brick, were snapped off at their foundations, with the curious phenome- 
non of the walls falling outward." 

The following whirlwind was not less remarkable : At 3 P.M. on the 
26th of August, 1828, after some calm and warm weather, a whirlwind 
appeared at Rouvier (Eure et Loir). It was preceded by a black cloud 
from the S. W., followed by others of a yellowish hue, with intermittent 
thunder and hail. Apparently touching the cloud at its summit, and 
with its base on a level with the ground, it threw down every thing in 
its passage, hurling the soil and trees to a great distance. The whirl- 
wind was of a dark yellow color — due, no doubt, to the dust and other 
substances which it carried off. The leaves of the hedges and trees 
which were not blown off were dried as if by fire. In the hamlet of 
Marchefroid, where it continued only a minute, it destroyed fifty-three 
houses. The inhabitants heard no thunder, nor did much hail fall. A 
child of three years of age was killed. A deep wound was found in its 
neck, but it was impossible to tell what body had caused it. In the 
valley of St. Ouen, the meteor destroyed a range of trees extending 800 
feet, and then moved toward Mantes, extending over a width of from 


forty to fifty fathoms. Whole houses were swept away, and in the di- 
rection followed by the whirlwind branches of trees were found scat- 
tered on all sides. Trees were snapped off at a height of four, six, and 
ten feet from the ground in the valley — a fact which would lead one to 
suppose that the tempest there did not reach quite so low as the ground. 
In one instance the destruction was very regular. The four walls of a 
garden, built of solid stone, were blown down, each wall in a straight 
line, and as if the stones had been placed there for constructing a wall. 
The body of a three-horse wagon loaded with grain was blown off the 
carriage and carried on to the top of a building, the roof of which it 
stove in. Pieces of the wood-work of the wagon were found upon the 
other side of the building. The grain had disappeared, and the horses, 
though uninjured, had been entirely stripped of their harness. 

The following case is equally remarkable : On the 26th of August, 
1826, the neighborhood of Carcassonne was visited by an enormous 
column of fire which, sweeping along the surface of the soil, destroyed 
every thing that lay in its passage. A young man was carried off by 
it into the air and hurled head foremost against a rock. Fourteen 
sheep were taken off their legs and asphyxiated. This column of air 
and fire overturned walls, displaced enormous rocks, uprooted the largest 
trees, and did great damage to a very solidly-constructed country house. 
The air, wherever it passed, was impregnated with sulphur. 

Among the whirlwinds which have left traces of great destruction 
behind them must be cited that of Monville on the 19th of August, 
1845. The valley in question, which is so attractive a point in the 
railway journey between Eouen and Dieppe, was visited at about 1 
P.M., the weather being hot and oppressive, by a whirlwind of a very 
remarkable kind. The large mills existing at Monville were suddenly 
enveloped and blown down. The factory, in which hundreds of wom- 
en were at work, fell in, amidst a sudden discharge of electricity, and 
they were buried beneath its ruins. Some of them who escaped death 
were unable to understand what had happened, and believed that the 
end of the world had arrived. Men were hurled over hedges; others 
were cut to pieces by the machinery which was whirled about in the 
air; others, without being actually hurt, were so terrified that they died 
from the effects of the fright in the course of a few days. Whole rooms 
and walls were turned upside down, so as to be no longer recognizable. 
At other points the buildings were literally pulverized, and their site 
swept clean. Planks measuring a yard long, five inches wide, and near- 


]y half an inch thick, archives and papers, were carried to distances of 
fifteen to twenty-five miles, almost to Dieppe. Trees situated in the 
track of the storm were blown down and dried up. The extent of the 
ground thus devastated was as much as nine miles, increasing from 100 
yards in width near the Seine, at Canteleu, to 800 yards about Monville, 
and decreasing again to thirty yards at Cleres. The barometer fell sud- 
denly from 29-92 to 27-75 inches. 

This sudden dilatation of the air necessarily upset the equilibrium 
of the atmosphere in the immediate neighborhood. An inhabitant of 
Havre informed me that on the day this catastrophe occurred he saw a 
vessel which was three leagues ofi" the shore enveloped in a tempest, al- 
though the sea just outside Havre was relatively calm. 

The catastrophe of Monville is remembered in Normandy, just as 
a terrible shipwreck is handed down in the recollection of a sea-port 
town. Fortunately whirlwinds do not often assume such immense pro- 
portions, or do not occur at the spot where large masses of people are 
congregated. Several others, equally violent, perhaps, have not found any 
element of resistance in their path. That which occurred in the neigh- 
borhood of Treves, in 1829, was in the form of a chimney hanging from 
a cloud, and emitting jets of flame and vapor. It soon changed to the 
shape of a serpent, undulating above the land, and leaving a track from 
ten to eighteen paces broad, along a distance of 2000 yards, where even 
the grass, plants, and vegetables growing upon the ground were swept 
away. There was, however, no loss of life nor destruction of houses. 
That which devastated Chatenay, near Paris, in June, 1839, burned up 
the trees that lay within its circumference, and uprooted those which 
were upon its line of passage. The former, in fact, were found with 
the side which was exposed to the storm completely scorched and burn- 
ed, whereas the opposite side remained green and fresh. Thousands of 
large trees were blown down and lay all one way, like wheat-sheaves. 
An apple-tree was carried over 200 yards on to a group of oaks and 
elms. Houses were gutted inside without being blown down. Several 
roofs were carried off as if they were kites. An inside wall was cut 
into five nearly equal parts of eight yards each ; the first, the third, and 
the fifth were laid in one direction ; the second and the fourth in an ex- 
actly opposite direction. Several rows of slates had their fixings torn 
out, without being themselves displaced. In a whirlwind which raged 
over the village of Aubepierre, in the Haute-Marne, on the 30th of 
April, the slates on the roof of a wash-house were turned completely 


upside down, each rank being reversed as if by the hand of a work- 


In the sandy regions of the African and Asian deserts, the traveler 

Fitr. di. — Saud whirhviud. 

sometimes encounters gigantic whirlwinds of sand which rise from the 
earth to the clouds, twisting convulsively, and emitting a sound like the 
hissing of a serpent. This is the phenomenon represented in Fig. 64, 


and it is taken from the travels of T. W. Atkinson on the frontier-land 
of Russia and China. 

The water-spouts that occur upon the water differ only from the 
whirlwinds of the air in respect to their situation. In place of dust, 
leaves, and other solid substances drawn up by the whirling column, 
they are composed of water, generally in the form of very condensed 
vapor, but sometimes in a liquid state, which becomes mixed with the 
air of the water-spout. Peltier cites many instances, extending over ev- 
ery degree of latitude. I can find no case in which they have been 
proved to have swallowed up a vessel. Generally the base of the col- 
umn is severed by discharging a cannon into it. Upon one occasion, 
however (in the Ionian Sea, on October 29, 1832), it appears that a ship 
was caught in a water-spout and tossed up and down, to the great alarm 
of the passengers, who were situated "like a person at the bottom of a 
well who is looking up into the air." 

The cloud which is attracted may descend near enough to the ground 
to raise up masses of water as well as floating substances; the heaviest 
of these will become detached from the mass one by one by reason of 
their specific weight, but the smaller bodies may be carried a great dis- 
tance, and may fall all together. This is one cause of showers of frogs 
and fish. 








The globe to which the force of attraction attaches us is nearly 7958 
miles in diameter, and is therefore about 25,000 miles in circumference. 
It is a sphere, the cubic volume of which is about 264,000,000,000 
cubic miles. If it consisted entirely of water, it would weigh about 
1080 trillions of tons, inasmuch as water weighs about 62|- pounds 
per cubic foot. But, as the earth is more than five times (544) as 
heavy as water, the weight of the globe is about 5880 trillions of 
tons. The atmosphere which envelops our planet weighs, as I have 
said, scarcely the millionth part of the weight of the entire earth (the 
T.TT^.-rro- part). Water occupies a not less important part in the ter- 
restrial system than air. The mean depth of the oceans is about two 
and a half miles, taking into account the uneven nature of their beds, 
the level of which, owing to the shores, table-lands, valleys, and mount- 
ains, varies from a few yards to six miles. 

Embodied in one mass, the water of the sea (taking its average depth 
at two and a half miles) would form a sphere OOO' miles in diameter. 
Spread over the whole spherical surface of the globe — supposing the 
surface to be perfectly even — it would cover it to the depth of nearly 
two miles. The density of sea- water is rather above that of soft water: 
its entire mass would weigh less than the 4 0^0 u P^i^^ of the weight of the 

The maximum depth of the ocean is about six miles, and the part of 
the atmosphere in which we can breathe is of about the same extent. 
It is within this limited zone of twelve miles that all the phenomena of 
life take place, from the submarine forests and strange animals which 
inhabit the lowest depths, to the plants which vegetate upon the surface 
where man has his being, to the various kinds of animals which live in 


the open skj, to the condor which soars above the limit of perpetual 
snow. This zone of life is very limited, when compared to the size of the 
earth, which itself appears so small in relation to the planetary system. 

To form an idea of the immense difierence, we have only to examine 
an equatorial section of the globe. Even if the sinuosities are increased 
fifty-fold, the terrestrial rind is almost a complete circle. The conti- 
nents and islands are but the summits of table-lands and mountains, the 
lower parts of which are submerged. 

This water covers nearly three-fourths of the earth, in the state which 
corresponds to the mean temperature of its surface — that is to say, in a 
liquid state. Its currents constitute, as we have seen, the grand arterial 
circulation of the planet. Not content with thus prevailing in its ordi- 
nary state, it reigns in a solid state, in the silent regions of the poles and 
upon the ice-bound sides of inaccessible mountains ; and, in a gaseous 
state, it reigns with more absolute sovereignty in the atmosphere, the 
life of which it regulates, and in which it in turn promulgates abun- 
dance and dearth, the gladness of fine weather, and the gloom caused by 
sombre skies. 

This water is not motionless, either in the depth of the oceanic basin, 
in the solid ice, or in the atmosphere. Thanks to the always active 
power of the sun, to the aerial currents, the water rises vertically from 
the bed of the sea to its surface, becomes vaporized at all temperatures, 
ascends in the shape of invisible vapor through the ocean of the air, be- 
comes condensed into clouds, travels across continents, falls again in the 
shape of rain, filters through the surface of the soil, passes along the 
strata of impermeable clay, springs up as a source or fountain-head, de- 
scends by the streamlet into the river, and falls from the river back into 
the sea again. 

Every source, every streamlet, every river, every stream, has its ori- 
gin in rain. Even the mineral waters are produced by the same cause, 
and their heat is merely due to the profound depths from which these 
meteoric waters have been brought up ; and they, moreover, continue 
to ascend through the interstices of the rocks, afterward returning to 
the level of their primitive reservoir, like a siphon. The sun, as it 
evaporates the sea-water, leaves behind it the salt, which is not volatile. 
That is why rain-water is soft, and that of a running stream also. The 
salt never leaves the sea, and its quantity is such that it would cover 
the whole surface of the globe to a depth of ten yards. 

Just as the blue color of the sky is due to the vapor of water, so also 


is the color of water itself, taken in a mass, blue ; its shades vary from 
that hue to green, according to the action of light. The vapor of water 
mixed with the air is of the highest importance in the distribution of 
temperatures ; both its formation, as well as its movement from place to 
place, represent a formidable force which is permanently in action. The 
air can contain more vapor of water in proportion as it is heated. A 
given diminution of temperature brings it to its saturating limit, with- 
out in any way adding to the quantity of vapor which it contains. To 
ascertain the quantity of vapor of water mixed with the air at a given 
moment, a thermometer, for instance, suspended in the air might be 
made gradually colder until it indicated the limit of saturation — that is, 
until its bulb was covered with condensed vapor, or dew. By ascer- 
taining what quantity of vapor of water corresponds to this thermo- 
metrical degree of saturation, we should learn the real quantity suspend- 
ed in the air at the moment of the experiment. 

The instruments for measuring the moisture of the air have received 
the name of hygrometer {vypog, moist, fxtrpov, measure). That in most 
general use is formed of two thermometers exactly alike,* and placed side 
by side. The bulb of one is enveloped by a piece of muslin, which is 
kept constantly moist. The moistened thermometer has a lower tem- 
perature in consequence of evaporation proceeding from the moist mus- 
lin. The difference of reading between the two thermometers is, there- 
fore, dependent upon the more or less moisture in the air. The hy- 
grometrical state of the atmosphere is not the same from the top to the 
bottom, like the proportions of oxygen and nitrogen. As a rule, it in- 
creases, beginning from the surface of the ground up to a certain height, 
where there exists a zone of maximum moisture; above that point it 
decreases. I will not venture to trace a diagram of this variation in the 
moisture according to the height, as I have done in regard to the de- 
crease of atmospheric pressure and of the temperature, for the observa- 
tions which I have made upon this head are wanting in number and in 
precision. Those taken by Glaisher are much more precise, and have 
been made by different hygrometers. They show that, generally speak- 
ing, the moisture increases from the surface of the soil to a height of 
3500 feet, and that after that it diminishes, there being, however, spaces 
which represent moist strata of air of varying thickness. 

The observations taken upon mountains confirm the increase which 

* See " Glaisher's Hygrometrical Tables " for description and use of dry and wet bulb ther- 
mometers. — Eu. 



was first noticed to be in proportion to elevation. Kaemtz has ascer- 
tained the degree of humidity to be a mean of 84 upon Mount Righi 
as against 74 at Ziirich beneath it. Bravais and Martins registered 76 
upon the summit of the Faulhorn, when at Milan there was only 63. 
At heights exceeding 3300 feet the moisture decreases, in spite of the 
special augmentations due here and there to currents which lie one over 
the other. 

Upon the surface of the ground, the relative moisture of the air va- 
ries according to the time of day, in inverse ratio to the temperature. 
The warmer the air, the drier it will be ; the colder it is, the more read- 
ily will it be saturated with moisture. In our temperate regions, the 
hygrometrical state of the air augments, with little fluctuation, toward 
sunrise during the minimum of temperature ; afterward falls, until about 
2 P.M., at the maximum of heat; and rises again toward evening and at 
night. Twenty years of daily repeated observations (1843-1863), taken 
at Brussels by the aid of the Saussure hygrometer and dry and wet 
bulb thermometers, have furnished M. Quetelet with the information 
that the mean degree of humidity at noon is as follows : 

January 87 

February 84 

March 73 

ril 66 





July 67 

August 68 

September 74 

October 80 

November 85 

December 89 

Where complete saturation is represented by 100. 

We see that the maximum of relative humidity occurs in December, 
and the minimum in June. This invisible atmospheric moisture, the 
presence of which is only revealed by aid of delicate instruments, con- 
fers upon the landscape all the variety with which it is endowed — the 
emerald green of the Irish pastures, the blue sky of the Mediterranean, 
the splendor of tropical vegetation — and it becomes visible in the shape 
of dew as soon as a diminution of temperature brings it to the point 
of saturation. If it is the air itself which becomes colder, it is made 
opaque by the passage of the vapor in a liquid state, and hence arises 
fog. If it be a solid body which is thus rendered cold, the moisture 
becomes condensed upon its surface, and the result is dew. 

Bew does not come down from the sky, as is still taught in the 
French primary schools. Its production is in no degree assimilated to 
that of rain. It is formed at the spot where it is seen. If small por- 


tioiis of grass, cotton, or other fibrous substance be exposed to the sky 
on a fine night, it is found that, after a certain time, their temperatures 
are fifteen, eighteen, and even twenty degrees below that of the cir- 
cumambient atmosphere. 

In places where the sunlight does not penetrate, and whence a large 
extent of sky can be seen, this difierence between the temperature of 
the grass, cotton, wool, etc., and the atmosphere is noticeable between 3 
and 4 p.m. — that is to say, as soon as the temperature diminishes; in 
the morning it continues for several hours after sunrise. 

The observations of Wells, continued by Arago,* have proved that 
on a clear night the grass of a meadow may be ten to twenty degrees 
colder than the air ; if the weather becomes cloudy, the grass at once 
increases several degrees in temperature, without any increase in that 
of the atmosphere. 

This diminution of heat is due to nocturnal radiation. When there 
is nothing to prevent the heat of a body from becoming dispersed, it 
gradually becomes irradiated and lost. The transparent air does not 
suffice to prevent this loss of heat. But a cloud, a wooden screen, a 
sheet of paper, a little smoke even, will answer the purpose. Without 
obstacles of some kind, the substance becomes colder according to its 
power of radiation, which is itself dependent upon the nature of the 
substance (it is, for instance, very great in the case of glass, and very 
trifling with metals) ; and when the temperature of the body thus ex- 
posed has reached that of the point of saturation, the atmospheric mois- 
ture is deposited upon it, taking at first the shape of spheroidal drops ; 
then, when these drops are sufficiently weighty and near together, they 
extend like a shallow pool of water over the surface of the substance. 

Dew is never abundant except when the nights are calm and bright. 
A little dew may be seen when the nights are cloudy, if there be no 
wind, or even with wind if the weather is bright; but there is never 
any sign of it when there is wind, and the sky is cloudy as well. The 
circumstances which favor an abundant deposition of dew more gener- 
ally occur in spring-time and in autumn — the latter especially — than in 
summer. It must be remembered, too, in addition to the above fact, 
that the differences between the temperature of day and night are never 
greater than they are in spring and autumn. 

The phenomenon of the deposit of dew upon a dense and smooth 

* [And by the Editor. See "Phil. Trans.," part ii., 1847, for paper on "Radiation at 
Night from the Earth, and several Substances placed on or near it. "] 


substance — upon a sheet of glass, for instance — resembles that seen 
when a pane of glass is exposed to a current of vapor of water warmer 
than itself: first, a light and uniform layer of moisture dims the sur- 
face; then are formed irregular and flat drops, which run together af- 
ter they have acquired a certain volume, and flow in all directions. 

This may be seen whenever any substance which has been rendered 
cold by exposure to a low temperature is taken into a warm room ; the 
substance at once becomes covered with moisture. In the same way 
glass placed in a room where a large number of persons are dining is 
at once dimmed by the thick stratum of dew which the invisible vapor 
mixed with the surrounding air deposits. The glasses of a pair of 
spectacles which have been exposed to cold air will often be found ob- 
scured in the same way. 

If, during a frost, the windows of a room in which a large company 
has been dining are suddenly thrown open, a cloud forms instantane- 
ously in the path of the cold air, and the ceiling is made damp by a 
long stain of condensed vapor. 

Dew is a phenomenon of importance, not only because of the absolute 
quantity which any one point of the globe receives, but because of the 
extent of ground over which it may be deposited. 'It is mostly in 
tropical regions that its effects upon vegetation are the most marked 
and the most favorable. When the air, nearly saturated with vapor at 
the temperature of 86°, contains more than thirteen grains of water to 
the cubic foot, the water falls abundantly during the declining tempera- 
ture of night ; it makes the leaves drip, and in the morning grass is 
as wet as if there had been heavy rain. The dew is known to deposit 
in greater or in lesser quantities, but it has not been found possible to 
measure it, because it does not fall like rain ; its appearance depends 
upon the radiating power of the body which it moistens, for it is only 
deposited upon substances which are colder than the surrounding air, 
and in increased quantity according as the difference of temperature is 
greater. Plowed land, fallow, forests, rocks, and sand vary much in re- 
spect to the dew which deposits upon them ; and, more than that, the 
leaves of all plants do not possess an equal radiating power, and the 
intensity of their diminution of temperature, influencing the deposit of 
dew which ensues upon it, is dependent upon their distance from the 
ground, their color, the smoothness or the ruggedness of their epidermis. 
The dew alights upon the leaves of mangel-wurzel, while the tops of 
potatoes in an adjoining field will hardly be moist. 


M. Boussingault has endeavored to measure these quantities of dew. 
After certain nights, when the dew had fiillen abundantly, he used to 
repair to the meadows on the banks of the Saiier before sunrise ; there, 
by aid of a sponge, he soaked up the water over forty-three square feet 
of glass, and this he placed in a bottle and weighed. In some instances 
it was found to exceed two pounds in weight. 

Dew and mist contain about the same proportions of ammonia and 
nitric acid ; both, moreover, have a great analogy to rain when it be- 
gins to fall — when it is, so to speak, in process of washing the air. It 
is, in fact, in the first part of a shower of rain after a season of long 
drought, that there is present the greatest amount of carbonic acid, car- 
bonate and nitrate of ammonia, organic matters, and dust of every kind. 
If a close examination be made of the substances which air contains in 
infinitesimally small quantities, it is in the mist, the dew, the first drops 
of rain, the first flakes of snow and hail, that we must look for them. 

White frost, which is so fatal to vegetation in spring, and which has 
given such a bad reputation to the harvest-moon, is, in reality, the dew 
frozen by the same cause as that which led to its formation — nocturnal 

In 1871, A. Wilson, having followed the movements of a thermom- 
eter during a winter night when the weather alternated constantly be- 
tween clear and foggy, found that it always rose about a degree at the 
same moment that the atmosphere clouded over, and fell to the point 
at which it had previously stood when the mist cleared off. His son, 
Patrick Wilson, asserts that the instantaneous effect of a thermometer 
hung up in the open air is to cause an elevation of 5°. The researches 
of Pictet, undertaken in 1777, and published in 1792, coincide nearly 
with the above. 

It is a curious circumstance, which was discovered by Pictet, that, 
when the nights are still and clear, the temperature of the air, instead 
of diminishing the higher from the ground, shows, on the contrary, a 
progressive rate of increase, at least up to a certain height. A ther- 
mometer nine feet above the soil marked throughout the night 4|-° 
Fahr. less than an exactly similar instrument which was attached to 
the summit of a pole fifty feet high. About two hours after sunrise, 
and two hours before sunset, the two instruments were exactly the 
same. Toward noon the thermometer nearest the ground was often 
4^° higher than the other. When the sky was covered with clouds, 
the two instruments corresponded exactly, both by day and night. 


These observations have been confirmed. Wells having placed at 
the four corners of a square four small pegs, which stood perpendicu- 
larly four inches above the surface of a meadow, spread over them 
horizontally a fine cambric handkerchief, and during five nights com- 
pared the temperature of the small square of grass covered by it with 
the surrounding portion which remained fully exposed to the air. The 
turf that was protected from radiation by the handkerchief was at times 
11° warmer than the other. While the latter was completely frozen, 
the temperature of the turf protected from the air was several degrees 
above 32°. With the sky very cloudy, a screen of cambric, matting, 
or any other substance, produces scarcely any effect. 

Mr. Glaisher finds, after three years' consecutive observations at 
Greenwich, that the temperature of the air twenty-two feet above the 
ground is higher than that at four feet at every hour of the night and 
day during the months of November, December, January, and Febru- 
ary; that it is higher at night and in the evening in May, June, and 
July ; and that it is also higher during night-time and in the afternoon 
in March, April, August, September, and October. At an elevation 
of fifty feet the temperature is also higher during the night throughout 
the whole year. With the sky cloudy, the temperature remains the 

In June of 1871 the attention of the Academy of Sciences was di- 
rected to the subject of late frosts by M. Ste, Claire-Deville and M. Elie 
de Beaumont. The immediate instance in hand was the frost which 
occurred on the 18th of May (Ascension-day) and extended to the 
vines and their crops around Paris and in the centre of France. As I 
myself had seen a vine which had been frozen in the Haute-Marne, I 
showed by a few comparisons that this disastrous frost extended over 
quite one-half of France at the same moment. It would certainly be 
most desirable to find some means for protecting crops during the criti- 
cal period which follows the blossoming, as many severe losses would 
thus be prevented. 




The invisible vapor of water spread throughout the atmosphere, the 
distribution and variations in which I have just pointed out, becomes 
visible when a dechne in the temperature or an addition of moisture 
brings it to the point of saturation. Suppose, for instance, that a cer- 
tain quantity of air at eighty-six degrees contains 478 grains of vapor 
of water, this air will be quite transparent. If by some cause or other 
this air descends to seventy-seven degrees, or receives an accession of 
moisture, it will become opaque. A diminution of nine degrees of 
heat will cause 108 grains of vapor of water to be condensed and to 
become visible. That is what a cloud really is : vapor of water which 
the air, being saturated, is no longer able to absorb, and which becomes 
separated from it by passing into the state of small vesicles. 

This passage from the gaseous to the liquid state takes place indif- 
ferently at all elevations. When it occurs at the level of the soil, it is 
termed mist. But there is no essential difference between a cloud and 
mist. Traveling through the clouds in a balloon, meeting no resist- 
ance, the air is simply more or less opaque, more or less cold, more or 
less damp, just as is the case upon the surface of the ground, according 
to the diversity of the mists. This is also the case with the clouds 
when one is enveloped in them upon the summits of a mountain. 

Though there is no essential difference between mists and clouds, there 
is, however, one in fact, viz., that a mist is vapor of water passing from 
the visible to the invisible state ; whereas a cloud is a grouping of visi- 
ble vapors in some given shape. The first is motionless^ the second is 
endowed with movement. Let us consider the mist first. 

Seen through a glass, mist is composed of small and opaque bodies. 
A closer study shows that these small bodies are composed of water, 
obeying the laws of universal gravitation. The molecules of water are 
grouped together in the form of spherules. Are these spherules full 


or hollow ? Such is the question upon which meteorologists are 
divided. The opinion already given by Halley that these spherules 
are hollow, and that the water is but an envelope, seems the best 

Take a cupful of some dark-colored liquid, such as coffee or China 
ink, dissolved in water; warm it, and place it in the sun's rays: if the 
air be still, the vapor will ascend and soon disappear; if looked at 
through the glass, it will be seen that globules are rising. The small- 
est run rapidly over the surface of the magnifying-glass, the others fall 
back on to the liquid mass. Saussure adds that the small vesicles 
which rise difi'er so much from those which fall back that it is impossi- 
ble to doubt that the first are hollow. 

The way in which they act when exposed to the light is also favor- 
able to this supposition, for they do not scintillate like the full drops 
when they are exposed to a bright light. Every one must have re- 
marked that soap-bubbles are generally very brilliant in color. The 
same must also have been noticed with bubbles from other viscous sub- 
stances, and it is the easier to observe them because they continue a 
longer time. These colors rise from the division of the incident rays 
into two parts. Some of the rays are reflected by the outside surface ; 
others penetrate through, and are reflected by, the inner surface. The 
envelope of the sphere must be thin, to admit of this taking place. 
Kratzenstein having examined in the sun and through a magnifying- 
glass the vesicles which ascend out of hot water, observed upon their 
surface colored rings like those of soap-bubbles ; and not only was he 
convinced that their structure is analogous to that of soap-bubbles, 
Ijut he was further successful in calculating the thickness of their en- 

De Saussure and Kratzenstein attempted to measure by aid of the 
microscope the diameter of the vesicles which compose the vapor of 
water. But it is difficult to arrive at any positive result, for it is the 
vesicles rising from mist, and not those from hot water, which it is nec- 
essary to measure. Fortunately, some of the optical phenomena which 
occur, when the sun shines through clouds, furnish us with a means of 
arriving at this result. 

Kaemtz has taken a great number of measurements in Central Ger- 
many and Switzerland; he has ascertained that upon an average the 
diameter of the vesicles of mist is about '00087 of an inch, and that it 
varies in the different seasons as follows : 












October. . . 



It will be seen that there is an almost regular progression from win- 
ter till summer ; the anomalies arise from the small number of obser- 
vations that have been taken. Thus in winter, when the air is very 
moist, the diameter of the vesicles is twice as great as in summer, when 
the air is dry ; but this diameter also varies in the course of a single 
month. It attains its Tninimum when the weather is very fine; it in- 
creases when there are signs of rain ; and before the fall it varies con- 
siderably in the same cloud, which probably contains a large number 
of drops of water mixed with vesicular vapor. 

Autumn, like spring, is the season of abundant dew. The cooling 
process to which the ground is subject, when the nights are clear, and 
the moisture of the air nearer precipitation than in summer, causes the 
atmospheric water to be deposited upon terrestrial objects which have 
diminished in temperature, just as in a crowded room the moisture of 
the heated air affects the glass brought in from outside. The steam of 
hot dishes, the breath of the persons present, the combustion of the 
lights, make the air of the dining-room hot and moist, and cause water 
to trickle down the vases containing ice. In autumn, the nocturnal 
coldness of the ground often communicates itself to the stratum of air 
immediately above, and hence arise the low fogs which are soon dissi- 
pated by the sun's rays. If the ground be uneven, the cold air of fogs 
descends into the valleys, and seems, to any one standing upon an emi- 
nence, a white sea perfectly level. As a child, I have often watched 
before sunrise, from the ramparts ofLangres, the ocean of grayish va- 
pors that extend through the valley of the Marne, and the waves of 
which reached to within a few feet of where I was standing. The 
height of the ramparts at Langres is near 1500 feet above the level of 
the sea. In winter the view sometimes extends at sunrise so far be- 
yond the mist in the plain, that the white outline of Mont Blanc is dis- 
cernible with the naked eye. 

To witness a spectacle of this kind at its best, it is necessary to be 


upon the top of a lofty mountain, whence the view embraces a vast 
horizon, and at sunrise after a day when the clouds have obscured the 
sky of the country below. The clouds, disturbed in a thousand ways 
by the rays of the sun and the light winds which are the natural con- 
sequence, are not very level during the day-time. But at night the 
equilibrium and the level are restored, and a sea of aerial vapors ex- 
tends far as the eye can reach beneath the feet of the observer. The 
elevated summits of the isolated mountains around him break here and 
there through the nebulous ocean, above which soars from time to time 
an eagle in quest of its prey. Standing in the valley, in the midst of 
the mist, the sun's rays, as they play through the foliage, delineate brill- 
iant beams of light, the ensemble of which forms what is called a gloi-y 
not more than a few yards above the head of the spectator. This glo- 
ry, which emanates from the tree immersed in the fog, recalls to mind 
Moses's burning bush. 

Sometimes only the surface of rivers is covered with fog, because 
water emits vapor which becomes condensed in the air which lies over 
them, and which becomes cold after sunset. The air takes almost in- 
stantaneously the temperature of the bodies to which it is in contigu- 
ity. During a calm and clear night, the portion of the atmosphere 
which lies over water will be warmer than that above dry land. 

In calm weather, where water is abundant, the lower strata of the 
atmosphere become laden with the extreme amount of moisture com- 
patible with their temperatures. I have already stated that the mois- 
ture which the air contains when it is saturated is of a fixed quantity, 
which varies according to its temperature. If saturated air becomes 
cold by contact with a solid body, it deposits upon the surface of that 
body a portion of its moisture ; but when the cooling process takes 
place in the very midst of the gaseous mass, the moisture that is set 
free passes off in small floating vesicles, which affect its transparency ; 
it is these vesicles which constitute clouds and mists. 

Let us suppose that some circumstance — a small declivity of the soil, 
for instance, a slight puff of wind — causes a fusion to take place at 
night between the air that lies over a river or sea, and that which is 
above the land : the latter, which is colder, diminishes the temperature 
of the former; the former also loses a part of the humidity which it 
contained, and which did not at first cause any alteration in its diapha- 
nous condition. But as this moisture gradually resolves itself into a 
state of vesicular vapor, the air becomes thick ; when the number of 



floating vesicles becomes very large, a heavy fog comes on. The dis- 
tribution of fog throughout the year corresponds with that of humidi- 
ty and temperature. Fogs are much more numerous in winter than 
in summer. The Brussels Observatory, which has recorded them with 
great care, gives the following as the number of days on which there 
have been fogs for the last thirty years (1833-1863) : 

January 259 

Februaiy 1G8 

March 138 

April 62 

May 71 

June 42 

July 28 

August 76 

September 159 

October 228 

November 276 

December , 315 

Total 1822 

Under certain circumstances the fog is very thick, and is bounded by 
a plane surface like a sheet of water, rising slowly in the still air, and 
enveloping all surrounding objects with a cold and damp embrace. 
M. Eaynal, whose vessel was wrecked off Auckland Island in 1864, 
was witness of a curious instance of fog, which he relates in this way : 
Having, on the 9th of August, climbed one of the mountains in the 
island, he was making his way down again with one of his compan- 
ions, following a narrow path between two precipices. " I was una- 
ble," he says, " to move a step, for we could not see where to put our 
feet. We passed at least an hour in this way, absolutely motionless, 
and holding each other by the hand, while the cold began to benumb 
our limbs. Fortunately a breeze sprang up, and dividing the fog into 
two parts, gradually carried it away." 

But it is in the frozen latitudes that the fogs are thickest. At Spitz- 
bergen, says M. Martins, the mists are almost continuous, and so thick 
that it is impossible to make out objects which are a few paces off. 
These damp, cold, and piercing mists often wet as much as rain. Thun- 
der-storms are unknown in these regions, even during summer. To- 
ward autumn the fogs increase, rain changes into snow. Fig. 67, illus- 
trative of an incident during the scientific voyage to which I have re- 
ferred, gives an idea of these immense and perpetual fogs. 

In countries where the soil is damp and hot, and the air damp and 
cold, thick and frequently recurring fogs must be expected; this is the 
case in England, the shores of which are surrounded by seas with a high 
temperature. It is the same with the polar seas and Newfoundland, 



where the Grulf Stream, which comes from the south, has a higher tem- 
perature than that of the air. 

Fig. 60.— luleuse fog in one of the islauds of the Antipodes. 

In London the fogs are at times dense. Every year the journals re- 
cord that it has been found necessary to light gas in the middle of the 
day, both in the streets and houses. Very heavy fogs* also occur in 

* There are at times dry fogs. They have no connection with the hygrometrical states I 
am now discussing. They are generally due to the smoke of burning prairies, and may ex- 



Paris and Amsterdam, the sk}^, at a short distance from these cities, be- 
ing at the same moment perfectly clear. 

Pig. 67.— Intense fo^ in the Spitzbergen Mountains. 

Thick fogs emit, too, a noxious odor when they become impregnated 

tend over a vast distance. The smoke of the heath in Holland sometimes reaches as far as 
Austria, hundreds of leagues off. The smoke of volcanoes also extends very far, that from 
Honolulu having been seen in 1868 at a distance of 200 miles from the mouth of the volcano. 
In 1865 the smoke from a great fire at Limoges covered the sky seventy-five miles off. The 
most intense dry fog known is one that occurred in 1783. 



with the different exhalations which may find their way into the lower 
strata of the atmosphere. Ammonia may often be discovered. It is 
not rare to find it accompanied by a smell of peat in Belgium and the 
north of France. During the cold and damp fogs of the month of Oc- 
tober, 1871, in Paris, the smell of petroleum was several times per- 

If a chain of mountains be looked at from a distance, it will often be 
seen that a cloud hangs over each peak, but that the intervals between 
them are clear. This state of things may last for hours and even days, 
but this absence of motion is only apparent, for there is frequently a 
strong wind blowing over these summits, which condenses the vapor as 
it ascends the flanks of the mountains. As soon as it disappears from 
the summits, the wind also vanishes. In Alpine passes, the formation, 
the movements, and the disappearance of clouds form a spectacle of 
very varied beauty. 

The clouds which ascend the mountain side of a day-time, by virtue 
of the diurnal ascending currents, often dissolve when they reach the 
summits under the influence of an upper wind, which is comparatively 
dry and warm. It is of an evening especially that this is the most no- 
ticeable, and the phenomenon generally occurs upon the ridges and 
summits of the passes which lead to them. The fog then seems to make 
its way in the direction from- which the wind is blowing, yet, notwith- 
standing the surface by which it is bounded, remains stationary. 

Very often, sombre clouds, passing rapidly over the St. Gothard Hos- 
pice, are precipitated in vast masses into the deep gorge of Lake Tre- 
mola. It might be fancied that all Lombardy would be obscured by a 
thick fog, but before it has issued from Lake Tremola, the warm ascend- 
ing currents dissolve it. 

Let us now consider the clouds in themselves, their formation, and 
the manner in which they are suspended in space. 

We saw, in the previous chapter, that the moisture of the air increases 
up to a certain height until it reaches a zone of maximum humidity^ the 
elevation of which varies according to the seasons and hours, and above 
which the air is drier and drier. This zone was seen by De Saussure 
in his Alpine travels, and by Commander Eozet both in the Alps and 
the Pyrenees. It is a blue, transparent vapor, which it is difficult to 
distinguish when one is immersed in it, but the upper surface of which 
is easily made out when situated beyond it. This surface is always 
horizontal, like that of the sea. From a great height upon some peak 


of the Alps or the Pyrenees, the topmost limit of this atmosphere of 
vapor is clearly delineated on the horizon by a bluish line, like that 
which bounds the horizon of the sea. Its height varies according to 
the season and hour ; it has been found to vary between 3500 and 13,000 
feet. Its temperature never falls below 32°. 

It is upon this surface of the atmosphere of vapor that clouds are 
formed, and on which they seem to repose. On the 15th of July, 1867, 
I rose to a height of 5000 or 6000 feet before sunrise, and for once I 
was present at the formation of clouds in the workshop of Nature. It 
was above the Rhine plain, between Cologne and Aix-la-Chapelle. The 
atmosphere had remained pure, when small white flakes began to ap- 
pear in the zone of maximum moisture. These gradually ran together, 
became grouped in large numbers, and dissolved with as much rapidity 
as they had formed. The small white clouds, agglomerated together, 
formed cumuli. This formation of clouds was proceeding several hun- 
dred yards below us. As the sun rose, the moisture on the balloon 
evaporated, and we gradually ascended to a height of 7900 feet. It 
was the same with the clouds, which indeed rose rather more rapidly 
than the balloon, and finally surrounded and surmounted it. 

The clouds are generally carried along by the wind, following its 
course and being relatively motionless in the current with which they 
float. The measurement of their speed gives, indeed, the measurement 
of the velocity of the upper wind. But this rule is not without excep- 
tions. There are, however, clouds which do not progress^ even when they 
are traversed by a more or less powerful wind, which it would be 
thought must take them along with it. 

When traveling in company with M. Eugene Godard in a balloon, 
while we were over the forest of Villers-Cotterets, I was much surprised 
to see for more than twenty minutes a small cloud which might have 
been about 200 yards in length and 150 in breadth, suspended motion- 
less about eighty yards above the trees. As we approached, we noticed 
five or six smaller, which were disseminated and also motionless, not- 
withstanding the air was moving at the rate of eight yards per second, 
and we were curious to ascertain what invisible anchor retained these 
small clouds. When we were above them, we found that the principal 
was suspended over a piece of water, and that the others were over the 
course of a stream, from which arose a current of humid air, the invisible 
moisture of which, reaching its saturating point, became visible in its 
passage through the cool wind that prevailed above the wood. 


Kaemtz witnessed an analogous occurrence near Wiesbaden after 
heavy rain. He says, "The clouds dividing, tlie sun burst forth, and I 
saw a column of mist that continued to ascend from the same point. I 
hastened thither, and found a newly-mown meadow surrounded by pas- 
ture-lands, the high grass of which, being less heated than the bare sur- 
face of the mown meadow, gave rise to a less active evaporation." In 
Switzerland the phenomenon occurs on a smaller scale. While it is 
fine upon the Faulhorn, the Swiss lakes are often covered with fogs of 
very different densities. The same meteorologist has observed that the 
fogs over lakes Zug, Ziirich, and Neuchatel were very thick, while 
those which rested over lakes Than and Brienz were merely light 
vapor. This phenomenon has occurred too often to be attributed to 
chance. Lake Zug is rather deep, and its tributaries do not descend 
directly from the regions of perpetual snow. Its temperature must be 
higher than that of Lake Brienz, into which the Aar empties itself im- 
mediately after having descended from the Grimsel glaciers. With the 
temperature the same, the first would become more readily involved in 
fog than the second. 

I must now explain the causes which lead to the suspension of clouds 
in the atmosphere. 

When a ck)ud is dissolved into rain, and pours down thousands of 
gallons of water, the question may well be asked how it is possible for 
such a weight of water to have remained suspended. The cause lies in 
its extreme divisibility. Left to themselves, the vesicles would fall. 
Calculation shows that it would take them more than half an hour to 
fall a little more than one mile in the atmosphere — that is to say, that 
the rapidity of their descent is about one yard per second ; it is often 
less. But during the day the air is constantly traversed by warm as- 
cending currents, which rise with a speed of several yards per second. 
Thus the clouds can not descend during day-time unless the circum- 
stances be exceptional. It is not necessary to suppose that their vesi- 
cles are filled with dilated and lighter air, as if they were so many small 
balloons. Nevertheless, as Fresnel has remarked, the solar heat ab- 
sorbed by the cloud must contribute to its remaining suspended. At 
night the clouds are nearer to the ground. But we have seen that the 
conditions, under which the vapor of water becomes visible, depend 
upon the temperature and the degree of saturation. It follows that the 
lower surface of the clouds dissolves as they descend into a warmer air, 
and frequen.tly, too, the upper surface dissolves when exposed to the 


action of the sun ; so that, as a matter of fact, they are constantly chan- 
ging in thickness, shape, and even substance. The clouds being but 
water in a special state, seem to us motionless even when the particles 
which compose them are incessantly descending from their upper to 
their lower surface, below which they become dissolved. They rest, 
moreover, upon the zone of invisible vapor which I have already spoken 
of. The horizontal march of the currents represents a somewhat con- 
siderable effort to maintain the clouds at the same elevation, even when 
all the aqueous particles are full. 

Having dealt with the formation of clouds, and their position in the 
air, let us consider their varied and characteristic shapes. 

The forms of the clouds are of infinite diversity, from the thick fog 
wh.ich bathes the surface of the soil to the luminous detached filaments 
which hover in the heights of the atmosphere. A methodical nomen- 
clature of clouds, to enable observers to record with precision observa- 
tions of their various forms, became a necessity. Howard first gave 
names to the principal types in order to have a means of recognizing 
each, and his classification has been generally adopted, so much so that 
his figures have become, so to speak, classic. His description alone I 
shall use as a basis for my remarks on this subject. 

In our climates the clouds are, in most cases, rather oval in shape; 
they seem to be piled one upon another, and their clearly-defined edges 
trace curves upon the azure of the sky. This class of clouds have re- 
ceived the name of cumulus., and it is in summer that their shape is the 
most marked. Sailors call them hales of cotton. They rise and aug- 
ment in size during the morning; reach their greatest elevation when 
the temperature is highest; from which time they descend, and ulti- 
mately disappear, when they are not numerous. Their thickness varies 
from 1300 feet to 1700 feet; their height from 1500 feet to 10,000 feet. 

Sometimes these half-spheres become heaped one upon the other, 
and form those large, accumulated clouds near the horizon which, seen 
from a distance, resemble mountains covered with snow. These are 
the clouds which lend themselves most readily to the play of the im- 
agination, for their lightness and the extreme variability of their shape 
give rise to incessant metamorphoses. It is not difficult to see in them 
the forms of men, animals, dragons, trees, and mountains. Ossian has 
utilized them for some of his finest imageries. The popular legends 
of mountainous regions are filled with strange events, in which these 
clouds play an important part. 


This frequently-occurring shape is coincident with the warm wind 
from the S. and S.W. — that is to sav, with the equatorial current. 
When this moist current prevails for some time, cumuli become more 
numerous, more dense, and spread in beds over the sky. This second 
form is seen almost as often in our variable climates as the first, and it 
is characteristic of winter as the latter is of summer, the principal dif- 
ference being that condensation, or rain, takes place more rapidly when 
the sky is in this state than it does during the summer phase. This 
kind of cloud is termed cumido-stratus. The fleecy clouds, the dappled 
sky, represent it in well-known aspects. 

When the clouds, instead of being detached, form one vast sheet ex- 
tending to the horizon, the term stratus is given. 

When a cloud is about to dissolve in rain, it acquires a greater 
density, becomes more sombre, and, except in the case of hail or par- 
tial storms, extends over a vast space. The water which is discharged 
from it would fall vertically if the atmosphere were calm, and the 
drops of water heavy enough ; but two causes, of which one at least is 
always in existence — the wind, and the lightness of the rain-drops — 
cause the water which falls from the cloud to follow an oblique course, 
generally preceded by the cloud, which the wind drives at a greater 
rate of speed. The special state of the cloud resolving itself into rain 
is termed nimhus. 

All these clouds are formed of aqueous vesicles, more or less consid- 
erable in size, and more or less compact. But the clouds do not only 
reside in the strata, the temperature of which is above 32°; they also 
float in the regions where the temperature is below the freezing-point. 
In this state the vesicular water becomes congealed into minute fila- 
ments of ice, and the clouds formed in this way are clouds of ice or 
snow, which have already served to explain such optical phenomena 
as halos, parhelia, etc. These clouds of ice are those which reach the 
loftiest regions. No matter the height to which the balloon may rise, 
these clouds always appear so far above that they seem no nearer than 
when viewed from the earth ; whereas it is a work of scarcely any time 
to travel through cumuli and the other forms of clouds which I have 
mentioned. Mr. Glaisher found that at 37,000 feet above the soil of 
England, he was still far below them. 

They are composed of loose filaments, the ensemble of which is 
sometimes like the sweep of a broom, sometimes like a bunch of feath- 
ers, sometimes like a mass of hair, or a light and irregular piece of net- 


work. Their mean height is from twenty to twenty-three thousand 

By reason of their very constitution, they remain in the ethereal 
regions of eternal snow. But, as I have said, the zone of 32° varies in 
height according to climates and season, whence it follows that these 
clouds may make their appearance in the lower regions of the atmos- 
phere in the frosty latitudes of the polar regions, and even in our lati- 
tudes during a severe frost. 

These clouds are designated cirrus. With a little practice it is easy 
to recognize them, and what is most striking in them is that they are 
nearly always divided into long and narrow strips, quite straight, and 
white in color, which correspond with the upper currents that direct, 
mold, or dissolve them. 

Sometimes their whitish hue gets bedimmed, their strice interlace 
each other, and they become denser because the upper air is moist. In 
this case they look like carded cotton, and this change generally fore- 
tells rain. When in this state of excessive density, they are called 

Sometimes, too, they become transformed into light transparent 
clouds of vesicular vapor — so transparent that the stars and the spots 
on the moon can be seen through them. These are clouds which give 
rise to the coronce ; when they are in receipt of abundant light, they 
seem to be well rounded and fleecy; when the sky is covered with 
them, it is said to be dappled ; their mean height is from ten to thirteen 
thousand feet; they are termed cirro-cumulus. The cumulus and the 
cirro-cumulus are those which impart the most beautiful hues to sun- 
set ; their transparency and their distant reflection refracting and color- 
ing its rays. The beautiful sunsets seen in Paris are partially due to 
the fact that these clouds, situated above Havre for the horizon of 
Paris, give us a softened reflection of the luminous effects that are pro- 
duced by the sea. 

Such are the principal shapes which clouds take, and which are due 
to the difference in their constitution and their elevation. These va- 
rieties do not constitute, in reality, more than two great categories — 
the cumulus, formed of liquid vesicles, and the cirrus, formed of frozen 

M. A. Poey gives the following "scientific and popular classification'" 
of the various shapes of clouds : 


1st Type.-CiRKUs. Curly cloud | ^^^^^^ ^^^^^^^ ^^.^^^^ ^6,000 to 40,000 feet. 

f Cirro-stratus. Sneaky cloud } 
Cirro-cumulus. Dappled cloud ) ^, , , „ • v,^ ,o nr,« * «.. ^„rv <• . 

\ Snow clouds. Height, 13,000 to 26,000 feet. 
Pallio-cirrus. Cloud in strata i 

2d Type.— CniatJLUS. Mountainous cloud "J 

( Pallio-cumulus. Eain cloud } ^'"" '^l""'^^' vesicular or of vapor of water. 

^^''"'''""''■iFraHo-cumulus. Wind cloud j ^^^'''^g'^ ^^^'S'^^' ^200 feet. 

Among the clouds composed of liquid vesicles, we must now con- 
sider the peculiar and characteristic shapes corresponding to the pro- 
duction of aqueous meteors, of which they are either the cause or the 

My colleague, J. Silbermann, Vice-president of the Meteorological 
Society, has spent thirty years in studying and making designs of these 
specially typical shapes. Out of the large number which he has stereo- 
typed and collected in a kind of meteorological museum, I will cite the 

Every one is acquainted with the shape of the clouds which usher in 
a lengthy period of rain ; the sky is covered with an immense leaden 
sheet, and the rain falls continuously from horizontal strata slightly 
undulated, which are scarcely distinguishable from the sombre mass in 
its entirety. For days and nights together the sky continues covered 
with this opaque slieet, the thickness of which is sometimes many thou- 
sand yards, there being successive strata by which the light of the 
autumn sun is entirely absorbed. These are clouds of continental rain, 
which extend over vast tracts of country, and the contour of which it 
is impossible to make out. 

The clouds ofj^cirticd rain resemble them so far that they are length- 
ened into horizontal strata; but in this case their shape, less extended, 
is more definite, as it stands out against the background of the sky, 
which is no longer darkened by the immensity of the strata that lie one 
over the other, but is partially covered with cumuli that have different f 

densities in different places. The rain issues from the sides of the 
clouds; it is delineated upon the pale perspective of the sky in oblique 
streaks of gray, the general tone of which varies with the motion of 
the wind. These clouds do not always dissolve entirely; certain parts 
seem, after they have discharged a great quantity of rain, to dry up 
and fall back into the centre of the cloud, as if attracted by the molec- 
ular affinity which gives to clouds their var3nng contour. 

The. hail-squall is different; it does not spread out in a large hori- 
zontal sheet, but forms a definite mass, which often stands out by itself 


in the blue sky. The sun reaches to its edges and sets off its white sur- 
face against the rest of the sky; there issues from its open sides a cold 
rain, hail, and rime, which a March wind blows into our faces. 

The clouds which produce hailhsiwe the singular aspect of an adhe- 
sion of moleculie, as if attraction tended to unite them in condensed 
masses of a globular form, and their shape has a strange resemblance to 
that of a cauliflower. This peculiar adhesion has also been noticed in 
thunder-clouds; the lower plane of this species of cloud is horizontal, 
and from this kind of table-like base rise projections, the shape of which 

Fig. 6S. — Fonuatiou of a thunder-cloud. 

may be compared with enormous balls of wool more or less carded, and 
connected the one with the other. These are typical instances which 
accentuate rather than attenuate the average appearance of clouds. The 
color, the white or the sombre hue of the clouds, can scarcely be taken 
as characteristic, for they are dependent upon their position in respect 
to the sun, and in regard to the situation of the observer. 

If we see a cloud at a great distance, and are standing between it and 
the sun, it will seem to us to be white. If, on the contrary, we notice it 



as it passes over our heads, we see the lower surface which the light 
does not reach, and then it appears black. 

The snow clouds have not this definite shape. They generally are of 
an immense thickness in the atmosphere, and of slight density. The 
light sifted athwart their vast extent gives them a yellowish tint, whence 
the flakes descend and cover the earth. 

Fijr. 09.— Abuvc and below the raiu cloud. 

HAix. 381 


rain: general conditions of the formation of rain — ITS DIS- 

Having treated of the distribution of moisture in the atmosphere, the 
manner in which the clouds are formed and remain suspended in space, 
their division into two distinct kinds, and the action of temperature 
upon the vapor of water, we shall have no difficulty in discovering how 
the formation of rain takes place. 

Rain is the precipitation of the aqueous vapor which constitutes the 
clouds. For this vapor to become precipitate — that is, to form drops, 
the weight of which causes them to descend and to produce rain — the 
molecular state of the cloud must be modified by some external cause. 
This modification may be effected by the influence of upper clouds — 
clouds of ice. Under certain circumstances, the least decline of tem- 
perature sets them in motion and destroys them. Such is the case with 
saturated cumuli; the least diminution of temperature precipitates them 
in the form of rain. 

The ordinary condition of the production of rain consists, therefore, 
in the existence of two layers of clouds, one above the other, and it is 
the higher which causes the precipitation of the one below it. This is 
an observation which an}^ one may verify for himself; and in the course 
of many years' observation of the sky when rain is about to fall, I have 
never found this condition wanting. 

Monck Mason remarked, in his aeronautical voyages, that when rain 
falls, the sky being at the time totally covered with clouds, there is al- 
ways a similar range of clouds situated at a certain height above, and 
that when, on the contrary, though it does not rain, the sky presents 
the same appearance below, bright sunshine prevails in the space im- 
mediately above. Saussure had already noted the same fact in his Al- 
pine explorations. Hatton had noticed that when two masses of air, 
saturated or nearly saturated, but of unequal temperature, meet, there is 
a precipitation of aqueous vapor. Peltier observed in regard to another 
point, that a thunder-storm is always composed of two banks of clouds 
which are of opposite electricity. Rozet arrived at the conclusion that 


thunder-storms and rain both result from the encounter between the 
cirrus and the cumulus, between the frozen and the vesicular vapor. 
Kaemtz and Martins adopt the same theory. M. Renou further adds 
that water may fall without being frozen at temperatures as low as 27°, 
36°, or 45° below the freezing-point of water, in the state of extreme di- 
visibility which constitutes fogs and mists, and that rain and frost are 
due to the admixture of fi-ozen cirrus with the still liquid cumulus be- 
neath the varying influence of temperature. 

Such is the general manner in which rain is formed. It sometimes, 
however, falls when the sky is clear. On August 9, 1837, at 9 p.m., 
Wartmann of Geneva noticed that during the space of two minutes 
large drops of warm rain fell from the sky, then studded with stars. 
The edges of the horizon were covered with broken patches of black 

On the 31st of May, 1838, at 7 p.m., M. Wartmann again remarked an 
analogous phenomenon, which this time lasted for six minutes. The 
warm drops, which were at first very large and thick, gradually de- 
creased in size. On the 11th of May, 181'1, at 10 a.m. and 3 p.m., he no- 
ticed the same occurrence, and during the time the air being quite 

The transit of masses of clouds is an important factor in their disso- 
lution, and in the abundance and the distribution of rain. This has 
been already pointed out when we were considering how the various 
directions of the wind corresponded with the amount of rain that fell. 
The south-west wind, which prevails in our country, brings the greatest 
amount of rain, because it is accompanied by the cloudy strata formed 
over the ocean, these strata of humidity being, moreover, sometimes in- 

Thus we can form an idea of the immense evaporation which daily 
takes place from the surface of the ocean, and see in it the origin of 
clouds and rain. The trade- winds, which blow over the surface of the 
sea in the tropics, carry this vapor of water as far as the regions of equa- 
torial calm, where they rise into the higher and colder part of the at- 
mosphere, and from thence pass to the temperate countries laden with 
moisture. As they rise through the atmosphere in the equatorial re- 
gions, a portion of vapor is condensed ; and as this occurs every day, 
there is a constant zone of clouds and rain. It is what English sailors 
term the cloiid-ring^ and French sailors the Pot au Noir. 

The oceanic clouds from the south and the south-west distribute the 



water which they contain according to their course, height, and temper- 
ature ; the more or less thick, and more or less cold, strata of clouds 
which weigh down upon them varying with the accidental winds which 
may affect them, and influenced by the undulations of the ground which 
alter their course. All other conditions being unchanged, the propor- 
tion of rain decreases from the equator to the poles, since, on the one 
hand, evaporation takes place almost entirely in the warm latitudes; 
and, on the other, the quantity of vapor which the air is capable of dis- 
solving augments rapidly as the temperature* increases. Thus, for in- 
stance, there is an annual rain-fall of more than six and a half feet at 
Guiana and Panama, while it is only seven and three-quarter inches at 


United States Tropical Zone South of the Alps Noith of the Alps 

; ScandinaV 

Fig. 70. — Diminution in tlie raiu-fall from the tropics to the poles. 

There is also a second law in regard to the proportion of rain, viz., 
that it diminishes in amount according to the distance from the sea, 
measured in the direction of the prevailing winds. It is easy to under- 
stand that clouds, being unable to reform in the interior of continents, 
yield less rain in proportion as they pass farther from the ocean. The 
evaporation that proceeds from rivers, lakes, pools, and moist plains, 
does indeed give rise to clouds ; but this is a very insignificant source 
of rain compared to that of the ocean. There falls forty-nine inches 
nearly at Bayonne ; forty-seven inches at Gibraltar; fifty-one inches at 
Nantes ; only sixteen and a half inches at Frankfort ; seventeen and 
three-quarter inches at St. Petersburg and Vienna. In Siberia the rain- 
fall is but seven and three-quarter inches, and less still farther east. 
At Algiers there is a mean of seven and three-quarter inches, and at 
Oran and Mostaganem of less than fouf inches. Farther south, the 
quantity of rain diminishes rapidly ; and at Biskra, on the borders of 
the desert, there falls two-tenths of an inch in the course of the year. 

Numerous observations enable us to establish a third law. The un- 
dulating nature of the ground causes a variation in the two distributing 
elements which we have just been considering. If a mass of air, satu- 
rated with moisture, encounters a mountain chain, it will be partially 
stopped by this protuberance of the soil. But the check is not a long 



one. The currents of air which ascend the slopes of mountains will 
elevate them at the same time ; they will become colder at the rate of 
one degree to 200, 250, or 330 feet ; according to the season and tem- 
perature, they will consequently undergo a progressive condensation, so 
that when they reach the summit they will be able to pass above it; a 
great part of the water they contained will already have fallen, and the 
remainder will descend upon the summit of the mountain. The lessen- 
ed speed of the air also deprives them of their water, much in the same 
way as the diminishing rapidity of a stream facilitates the fall of the 
deposits which it keeps suspended. There falls, therefore, more rain in 
a mountainous than in a level region ; there is also more rain upon the 
slope that faces the sea-wind than upon the opposite. Thus, clouds 
which, as they pass over Lisbon, give but an annual rain-fall of twenty- 

Altitudes 2^00 


Fitf. Tl.— Increase of ruin, accA)rdii)g to the uudukUiuns of ihc 

seven and a half inches, are soon arrested by the cold-tipped mount- 
ains of Portugal and Spain, there being a rain-fall of 118 inches at Coim- 
bra. The clouds which pass at the zenith of Paris yield nineteen and 
three-quarter inches of rain in a year. As the altitude augments, so 
does the rain. Thus, taking merely the basin of the Seine, we have 
three and a quarter feet of rain-water upon the plateau of Langres, 
and six feet nearly at the higher point of Morvan, in the Niovre. At 
Geneva, at the foot of the Alps, the annual quantity of rain is thirty- 
two and a half inches, and at the Great St. Bernard ridge it is six and 
a half feet in the year. 

There are regions in which these conditions are so complete, that the 
rain stops as if attracted there permanently. Thus the Great Himalaya 
chain stops the clouds which come from the immense evaporation of 
the Indian Ocean. x\t Cherra-Poejen, upon the Garrows Mountains, at 
a height of 4500 feet, and to the south of the Brahmapootra Valley, 
the quantity of rain which the clouds pour down is forty-eight and a 
half feet. In these mountainous regions near the tropics, the maximum 
rain-fall is probably to be found ; they are also the great reservoirs of 
the Asiatic rivers. In these same lower slopes of the Himalayas, upon 

EAIK 365 

the eastern side of the Ghauts, an average annual rain-fall of twenty- 
five feet nearly has been recorded, after observations extending over a 
period of fourteen years. A downfall lasting only four hours has been 
known to cover the ground to a depth of thirty inches — more than falls 
at Paris in a whole year. It is certain that in no other part of the tor- 
rid zone is the precipitation of the rain so much facilitated by attendant 
circumstances. The Antilles are not wide enough to prevent the winds 
and clouds from veering obliquely to the right or to the left ; but not- 
withstanding, certain districts there receive thirty-two and three-quar- 
ter feet in the course of the twelvemonth. In the Gulf of Mexico the 
summer rains also give a depth of more than thirteen feet at Vera 
Cruz. Farther from the tropical regions we only noticed these remark- 
able maxima of rain upon the mountain chains which, being in the way 

feet ft 

•u' s 

>i9S -» 




Cherra-Poejen.* Mahabuleshwar. Vera Cruz. Bergen. Names. Paris. Alexandria. 

Fig. 72.— Comparative depths of rain-fall. 

of the general current, bring it to a stop. Such, for instance, is the ef- 
fect produced by the Scandinavian Alps that separate Sweden and 
Norway, for its western slope receives much more rain than the eastern 
side, there being an annual rain-fall of eight and three-quarter feet at 
Bergen, which exceeds that of any other town in Europe. Moreover, 
several points are again specially favored in respect to their frontage to 
the south-west current ; as Nantes, for instance, where there is a mean 
annual rain-fall of four and a quarter feet. 

Collecting and comparing the observations that have been made at a 
great number of places in different parts of the globe, it has been found 
possible to register the three predominating causes which we have re- 

* [At Cherra-Poejen the fall of rain in April is 22 inches ; in May, 62 inches ; in June, 
19.5 inches ; in July, 121 inches ; in August, 104 Indies ; in September, 75 inches ; and in 
October, 29 inches ; making a total fall in seven months of 608 inches. No rain falls either 
in November or December, and less than five inches in the months of January, February, 
and March. See my "Report on the Meteorology of India," in relation to the health of the 
troops stationed there, 1863. — Ed.] 



viewed, to lay down upon a diagram the depths of the rain-fall that 
have been observed, and to make a map exhibiting the comparative 
depth of rain all over the globe. The heaviest rain takes place to the 
north of the equator in the Atlantic, in the Pacific, and to the east 
of America. In these regions, the maximum falls exceed six and a 
half feet in depth ; in Asia, in the islands of Borneo, Sumatra, and 
Java, along the Himalaya and Ghauts Mountains ; in Africa, along the 
table-lands of the eastern coast; in the Atlantic, between Guinea and 
Guiana ; in South America, upon the Andes in Chili, at Cape Horn, 
and upon the summit, above Peru, which, by contrast, is a country 
where no rain falls. Lastly, the mountain chain which runs eastward 
along the borders of North America, from fifty to sixty degrees longi- 
tude, yields an annual maximum of more than six and a half feet. 

The rainless regions extend along the desert of Sahara, Egypt, Ara- 
bia, and Persia, reaching as far as Mongolia, and even to Siberia, with 
the exception of the region of Central Asia, upon which the monsoons 
and the winter rains yield some little moisture. 

If we consider Europe in particular, we find, relatively, abundant 
rain, ranging from three and a quarter to six and a half feet, in the 
marine zones of Portugal, Brittany, Ireland, and Sweden. The propor- 
tion of rain gradually diminishes toward the east, with the zones of 
condensation produced by the undulating nature of the soil. There 
are certain points where rain is very rare, as in Greece, for instance. 
The climate of Attica is dry, and the sky is generally clear, the air 
having always been considered the purest in Greece. As an instance 
of this, I may mention that M. Lusieri exposed a piece of paper to the 
air all night, and that he was able to write upon it the next morning. 
To this remarkable dryness of the air has been attributed the excellent 
state of preservation of the Athenian monuments. 

The northern hemisphere receives more rain than the southern by 
about one-fourth. This excess of rain is especially due to the northern 
equatorial zone of rains and monsoons. Nevertheless, there is much 
more dry land in the former than in the latter, and evaporation pro- 
ceeds on a much larger scale in the southern hemisphere, which is near- 
ly all sea. Thus, our clouds, our rain, our rivers, and our streams are 
chiefly fed by the ocean in the hemisphere of our antipodes. 

As the distribution of rain has for its origin both the variations of 
temperature and the prevailing winds, it can be easily seen that in dif 
ferent countries it is more or less abundant according to the time of year. 

RAIN. 887 

The countries in ■wbich there is what is termed a rainy season are 
those situated in the tropics, where the sun, which twice a year passes 
perpendicularly over them, occasions at those epochs an excessive heat, 
which must, of course, be succeeded both by a great rarefaction of the 
strata next to the ground ; as these latter, becoming too light to bear 
the weight of the upper strata, rise, and afterward by the diminution 
of temperature and fall of rain, which always follow, no matter what 
may have been the producing cause. It is impossible to form an idea 
of the mass of w^ater which, during the rainy season, falls into the 
basins of the Amazon and the Orinoco, After these streams and their 
tributaries have overflowed their banks to a height of several feet, a 
tract of country as large as Europe becomes a fresh-water sea, the out- 
flow of which into the ocean destroys the salt for some distance from 
the shore, and in comparison with which the North American lakes are 
mere mill-ponds. The scientific study of this great display of physical 
forces, in which Nature, whose action is irresistible, commands the at- 
tention of us whose existence is menaced, is making rapid progress, and 
none are better qualified to throw light upon the subject than the in- 
habitants themselves, whose life depends upon their being familiar with 
the vicissitudes of the seasons. 

Thus, in the United States, upon the Atlantic, from the twenty- 
fourth and as far as the fortieth degree of latitude, in Spain, in the 
south of France, in Italy, Greece, Turkey, Asia, China, Japan, and in 
the Pacific, in the same latitudes, nearly all the rain falls in winter, ex- 
cepting the region of periodical monsoons and in certain southern coun- 
tries, where, during the summer months, no cloud appears in the sky. 
It is the same between the twenty-fifth and fortieth degrees of south 
latitude, at Buenos Ayres, the Cape, and at Melbourne. 

Over a zone extending from twelve to fifteen degrees of south lati- 
tude, over nearly all the globe, it is in summer that most rain falls. 

Over a zone extending from forty to sixty degrees of north latitude, 
and which reaches as far as seventy-five degrees, beyond Iceland and 
Sweden, and within a limited zone in Asia, rain falls at all times of the 

Nevertheless, even in our variable regions, there are well-defined 
proportions for each particular season. Thus, taking France in partic- 
ular, we find that it may be divided into two parts. The western re- 
gion has the maximum of rain in summer, and the minimum in winter. 
Such is also the case in England, while in Germany it is the reverse. 



under even more marked conditions. The same holds good with re- 
gard to Eussia. 

We have said that there is an annual rain-fall of seven and a quarter 
feet at Bergen, in Norway, This town is, in this respect, a remarkable 
exception in the meteorology of the globe. It is, in all Europe, the 
town where there is the most rain. It is situated in the centre of a 
deep bay, exposed to westerly winds, which are stopped by the mount- 
ains, so that the rain is, to use Kaemtz's expression, mechanically 
pressed out. 

The following table gives the rain-falls throughout Europe, and is 
the result of many years' observations : 


Names of Places. 


















. » 









51 6 









50 5 









59 52 









48 13 

St. Petersburg 








59 56 


















52 34 









48 50 









59 21 









38 8 








55 41 








60 27 









48 46 









43 36 









49 7 









47 19 









55 57 









.50 51 









49 26 









51 3 









53 23 










46 12 









43 36 










45 24 
53 29 























43 47 
45 4 

45 28 

46 31 












46 5 

The quantity of rain at Breslau, Prague, Upsal, Vienna, and St. 
Petersburg, shows how little falls in these places, as the mean is less 
than 15f inches. 

RAIN. 389 

In tbe Netherlands, Belgium, France, Germany, and Poland, the 
average is 19|, 23|-, and 27|^ inches. It is easy to see that there is a 
diminution as one recedes from the sea inland. Thus, in the Belgian 
cities, there is more than 27-| inches of rain, while in the same latitudes 
at the German towns and those nearer to Asia the quantity is much 
smaller. Upon the other hand, it is evident that the two most rainy 
seasons are summer and autumn, no matter what the distance of the 
locality from the sea. England is very peculiarly situated in this re- 
spect, as, being surrounded by the sea, she receives more rain than her 
latitude would lead one a priori to expect. 





When several strata of black and grayish clouds are flying through 
the atmosphere, and when the thunder-storm has burst forth, millions 
of pounds of hailstones are launched from the clouds as if precipitated 
from the opened cataracts of a vast reservoir. For several minutes the 
hail drives through space, pelting trees and gardens ; it then ceases as 
the wind blows it off in some other direction, and the close and sultry 
temperature which had preceded it gives place to the fresh odor of 
refreshed plants, light returns, the rainbow appears, and the blue sky 
emerges from the banks of clouds. What is the force which produces 
in the clouds these lumps of ice (often very large), what bears them up 
in space, and then launches them upon the earth ? While studying the 
production of rain, we saw that it does not, as a rule, occur except when 
there are two or more strata of clouds one over the other. Such is also 
the case with the formation of hail, though there is a difference in the 
respective physical conditions of the clouds. 

Hail occurs during a thunder-storm, when the temperature is very 
high upon the surface of the ground, but decreases rapidly with ele- 
vation. This rapid decrease is the principal element in the formation 
of hail, and it has been known to be as much as one degree in a little 
more than 100 feet. What then takes place in the region of clouds? 
Those above, from 10,000 to 20,000 and 25,000 feet high, contain, the 
highest of them, ice at —22° or at —40° Fahr. ; the lowest of them, 
vesicular water at +14° and at —4°. The lower clouds contain vesic- 
ular water above 32°. As a rule, these clouds travel in different direc- 
tions, and hail is formed when there is a collision and admixture of 
winds that are opposed to currents and clouds the temperatures of 
which are different. The vapor, which then resolves into rain, freezes 
instantaneously in so cold a temperature. Carried off by the wind, and 
even exposed to the influence of opposite electricities of the diverse 

HAIL. 391 

strata of cloud, these frozen drops do not fall at once, notwithstanding 
their weight, and they have time to become enlarged .by the addition 
of a considerable a^uantity of water which they collect during their pas- 
sage through the air. 

The extreme cold that prevails in the clouds below the region of 
perpetual snow is due, in a great measure, to evaporation, which has 
itself a double cause — the action of the sun and of electricity — it hav- 
ing been remarked that after every electric discharge the rain or hail 
falls in great quantities, and the reaction produces a dilatation which 
gives rise to rapid evaporation. 

The formation of hailstones is always a very speedy process. Volta 
was of opinion that the upper cloud was formed by the condensation 
of vapor from the lower strata, and that it contained positive electricit}', 
while the latter retained negative electricity. Just as pith-balls placed 
between two copper plates laden with opposite electricity are seen to 
bob up and down under the influences of this double attraction, in the 
same way he thought that hail was formed by a like movement of the 
corpuscles of ice or snow, becoming successively enlarged by condensed 
vapors. This theory is not now considered admissible, and it is, indeed, 
far simpler to suppose that hail is formed like rain, but amidst an at- 
mospheric cold which freezes the globules of water at the very moment 
of their formation. 

It appears that this formation, or the shock of hailstones that are 
borne along by the wind, sometimes produces a noise audible upon the 
surface of the ground. Aristotle and Lucretius, of ancient writers, re- 
cord this fact; and the meteorologists Kalm and Tessier assert that 
they heard it, the former in France on July 13, 1788, the latter at Mos- 
cow on the 30th of April, 1744. Peltier states that at Ham the ap- 
proach of a hailstorm was preceded by a sound like that of a cavalry 
squadron at full gallop. In 1871, M. Pessot, corresponding member of 
the Montsouris Observatory, reported from Doulevant-le-Chateau (Haute- 
Marne) a hailstorm which was preceded by this same phenomenon. 

The surfaces of hail-clouds show here and there immense irregular 
protuberances. Seen from underneath, they are generally dark in col- 
or, because of their opaqueness, which the solar light is scarcely able to 
traverse. Arago pointed out that they seem to be thick, and to be dis- 
tinguishable from other storm-clouds by their ashen hue. Their edges 
are indented ; but they very soon are lost in the general mass of the 
nimbi which discharge rain. 


To what height do they soar? From what elevation do hailstones 
fall ? Saussure noticed a hailstorm upon the Col du Geant at a height 
of 11,246 feet, Balmat upon the summit of Mont Blanc itself, and Pac- 
card discovered hailstones beneath the snow which forms its peak. 
Hail often falls upon the high slopes of the Alps. Thus the phenome- 
non of hail occurs at all elevations. But when the height at which it 
commences to fall is very great, the hailstones melt during their pas- 
sage through the thousands of feet of air above the temperature of 32° 
which cover the surface of the globe. In the case of our hailstorms, 
on the contrary, the clouds which emit them are at a less height, and 
seem to be between 5000 and 6500 feet above the ground. Below 
them extend the storm and the rain clouds, at a height of about 3300 
feet, or even lower. The clouds which discharge hail are never very 
large. Borne along by the wind, they cover a narrow strip of land, 
which is often only three-fifths of a mile in breadth, and rarely more 
than ten miles long ; but the length is sometimes as much as 500 miles. 

One of the most curious and remarkable hailstorms in the annals of 
meteorology is that of July 13, 1788. It was divided into two bands : 
that on the left, or the western one, began at Touraine, near Loches, at 
6'30 A.M. ; passed over Chartres at 7'30 a.m. ; over Rambouillet at 8 
A.M. ; Pontoise, 8'30 A.M. ; Clermont (Oise) at 9 A.M. ; Douai, 11 A.M. ; 
whence it entered Belgium, passing over Courtrai at 12"30 p.m. ; and 
finally dying out beyond Flashing at 1*30 p.m. The total length was 
420 miles, and it extended over a width of ten miles. 

The right, or eastern, branch began at Orleans at 7"30 a.m., passing 
over Arthenay and Andonville, reached the Faubourg St. Antoine in 
Paris at 8-30 a.m., Crepy-en-Valois at 9-30 a.m., Cateau-Cambresis at 11 
A.M., and Utrecht at 2'30 p.m., the length being near 500 miles, and the 
width only five miles. There was a mean interval of twelve miles of 
ground between the two bands, and rain fell in this space. The pas- 
sage of the hailstorm was preceded on each line by a profound dark- 
ness. The speed of the storm was thirty-two miles per hour on both 
lines, the hail not falling for more than seven or eight minutes in the 
same place, but with so much violence that the crops were cut to pieces. 
This is the greatest hailstorm known. No less than 1039 communes in 
France suffered from its ravages ; the destruction of property was found 
to amount to no less than £1,000,000. The hailstones were not all of 
the same shape; some were round, others long and pointed; and some 
were found to weigh 3900 grains, or more than half a pound. 

HAIL. 393 

It is seldom that the same hailstorm extends over such a length of 
country, and in so regular a line. It is probable that the clouds which 
produced this hail were more than half a mile high. Generally they 
are at a less height than this, and are influenced by the undulations of 
the soil. Certain storms, without having extended over so much 
ground,' are remarkable for their abundant quantity. On May 9, 1865, 
for instance, a storm began at 8'30 a.m. over Bordeaux and proceeded 
in a N.N.E. direction, passing over Perigueux at 10 a.m., Limoges at 
noon, Bourges at 2 p.m., Orleans at 5"30 p.m., Paris at 7*45 p.m., Laon at 
11 P.M., and collapsing a little after midnight in Belgium and the North 
Sea. Its mean breadth was from fifteen to twenty leagues. The hail 
fell only in certain places: to the left of Perigueux, over the arron- 
dissement of Limoges, to the right of Chateauroux, to the south-east of 
Paris, from Corbeil to Lagny, and in the arrondissements of Soissons 
and Saint Quentin. At this latter point it was of a formidable charac- 
ter. The crystal mass which fell from the sky upon the Catelet mead- 
ows formed a bed a mile and a quarter long and 2000 feet broad, esti- 
mated to amount altogether to 21,000,000 of cubic feet. The hailstones 
did not disappear for more than four days afterward. These hailstones 
sometimes destroy all the crops, as, for instance, that which occurred in 
the neighborhood of Angouleme on August 3, 1813. The day had 
been fine, and the wind was due north until 3 p.m., when it suddenly 
veered right round ; the sky gradually became covered with clouds, 
which, collecting one on the top of the other, offered a terrible specta- 
cle. The wind, which from noon until 5 P.M. had been rather violent, 
suddenly dropped. Thunder was heard in the distance, and gradually 
became louder ; the sky, at last, became totally obscured, and at 6 P.M. 
there was a tremendous fall of hail, the stones being as large as eggs. 
Several persons were severely wounded, and a child was killed near 
Barbezieux. The next day the ground looked as it might do in mid- 
winter: the hailstones had accumulated in the hollows and the roads to 
a height of thirty to forty inches; trees were entirely stripped of their 
leaves; vines were cut into pieces, the crops crushed, the cattle, sheep 
and pigs especially, were severely injured. The whole neighborhood 
was deprived of game, and some few young wolves were found dead. 
The effects of the storm were still visible in 1818, the vines, in particu- 
lar, not having recovered their productive powers. 

The storm which burst over Chaumont, in the Haute Marne, on July 
17, 1852, spread over a district nearly sixty miles long by five miles 


broad; wheat, vines, and nearly every tree, were destroyed by bail- 
stones of abnormal dimensions. The same hurricane swept violently 
over the department of the Aisne, uprooting trees, blowing down cot- 
tages, and killing several persons. In a few seconds all trace of the 
crops had disappeared from the fields. 

On July 17, 1868, at about 8 p.m., a heavy hailstorm devastated the 
neighborhood of Rheims; the stones were as large as Barcelona nuts, 
and the downfall lasted three parts of an hour. In some of the hollows, 
where the ground was sandy, there were remarked impressions like 
those which might be made by a cannon-ball. These cavities, into 
which the hailstones were first driven, constitute regular physical im- 
pressions of the hail, which seemed, in regard to the construction placed 
by geologists upon similar marks, to possess a special importance. 

Disastrous hailstorms are, fortunately, rare in our climates, though 
they do, from time to time, remind us of their existence. A heavy 
storm began at Brussels on June 18, 1839, about 7 P.M. ; thick clouds 
drove from the S.S.W., while at the same time the vane indicated a 
lower current from the N.W. Until 7"30 there was a continuous roll- 
ing sound, during which the flashes of lightning succeeded each other 
with astonishing rapidity. Soon after, a large cloud of very ashen 
hue, and the direction of which was from W.N.W. to S.E., veiled the 
city in the most complete obscurity, and burst in a shower of hail 
which did immense damage. Most of the hailstones were from a half 
to three-quarters of an inch in size, some of them as much as one inch. 
In shape some were spherical, but the greater number were more or less 
flat. The depth of water that fell during the storm was an inch and a 
half The temperatui'e rose as high as 92° Fahr. — the maximum re- 
corded at Brussels ; the barometer reading was 29"70 inches at 4 p.m. 

Hailstorms have a tendency to follow the direction of the valleys and 
the rivers when the clouds are not high ; for, as is shown by the cases 
cited above, the storms then become regular currents, which come from 
the Atlantic and, following the ordinary course of the currents which 
reach us, continue their progress from the south-westerly regions to- 
ward those of the north-east. But in all partial secondary storms (which 
are the most frequent, and are generally confined to a limited area) there 
is an evident deviation from the valleys. It seems, too, that they keep 
away from forests. Since meteorological facts have been registered by 
the French Ecoles Kormales, there has been plenty of evidence collected 
as to the influence of the ground in regard to the distribution of storms 

HAIL. 395 

and of hail. One district may be visited by hailstorms every year, an- 
other not once in ten years. It has even been found possible to com- 
pose statistical maps showing the damage done by the hail in each de- 
partment, by aid of the documents appertaining to insurance companies. 
These maps are scarcely reliable from a meteorological point of view, 
as they are based on pecuniary losses ; and the same quantity of hail 
would cause ten times as much damage were it to fall over a tobacco 
plantation of the Lower Ehine, as it would if it were to rage over an 
uncultivated or even a wooded district. It is true that the intrinsic 
quantity of hail differs in neighboring countries, according to their geo- 
logical, orographical, and climatological situation. 

Hailstorms are those in which the development of electrity attains the 
largest proportions. The thick clouds in which the meteor becomes 
elaborated are laden with a large quantity of the electrical fluid, part of 
which becomes exhausted within themselves or in reciprocal discharges 
with neighboring clouds. 

The thunder is then not merely a report following the flasb; it is a 
continuous rolling sound, during which it is not unusual for no light- 
ning to be perceptible, either because the flashes are of very small di- 
mensions or because they take place entirely within the interior of the 
clouds. Thus, on the 4th of September, 1871, 1 noticed, in the hail- 
storm which took place in Paris at 3'36 p.m., that when the hail had 
passed over the district in which the Observatory is situated, and when 
it was over Menilmontant, there was a continuous rolling of thunder, 
unaccomjxw.ied hy lightning., which lasted six minutes, and recommenced 
again after several short intervals. On the 7tli of May, 1865, a violent 
storm burst over the department of the Aisne, causing damages amount- 
ing to several million francs. Above the strata of clouds there was 
visible a thick cumulus, of a livid white hue, from which there was a 
continuous flashing of lightning; the rolling of the thunder was unin- 
terrupted, though not very loud ; there was an unintermittent crepita- 
tion of the lightning, and the explosions seemed to be confined to the 
interior of the largest cloud. When the cloud had slowly ascended the 
heights of Roussay, upon the apex of the basins of the Somme and the 
Scheldt, it swept down with tremendous rapidity into the valley of this 
latter stream, pelting Vend'huile, Catelet, and Beaurevoir, with so many 
hailstones that they lay five yards deep upon the ground. The}'' were 
still visible five days after, and, at some places, formed such a solid mass 
that they acted as a dike to keep back the water. When it was at- 


tempted to sweep them away they slipped along like fields of ice ! M. 
Quetelet remarked, during a severe storm that occurred at Brussels on 
June 18, 1839, a continuous rolling of thunder, during which time the 
flashes of lightning succeeded each other with marvelous rapidity. Soon 
after, a thick ashen cloud plunged the whole city into profound dark- 
ness, and burst in a heavy fall of hail. 

It is interesting to ascertain what is the greatest dimension which a 
hailstone can attain. I am able to give some very curious comparisons 
on this subject from a number of well-authenticated documents. 

After the great hailstorm of July 13, 1788, alluded to above, the geol- 
ogist Tessier cut pieces of ice which seemed to him to be of the con- 
sistency of hail, into the shape and size of pigeons', hens', and turkeys' 
eggs, in order that meteorologists might be enabled to calculate approx- 
imately the weight of hailstones according to their size. The first 
weighed 169 grains; the second, 254; and the third, 1065 grains. 

The most ordinary size of a hailstone is that of a small nut: some, 
indeed, are not larger than a good-sized pea. In ordinary storms, the 
stones weigh from 46 -to 120 grains. 

The three weights above often occur in the annals of meteorology. 
There is nothing absolutely abnormal in a fall of hailstones weighing 
from a quarter of an ounce to two and a quarter ounces. 

Some extraordinary facts are the following, which are, however, per- 
fectly authenticated and certified by well-known savants : In a disastrous 
hailstorm near the Khine, a hailstone was picked up by Yoget at Heins- 
berg weighing 1400 grains. At Randerath they weighed twice as much. 
During a storm that occurred at Morbihan, and which lasted three-quar- 
ters of an hour, on June 21, 1846, the hailstones were of all dimensions, 
from the size of a nut to that of a turkey's Qgg. One was eight and 
three-quarter inches in circumference, Muncke weighed some hailstones 
in Hainault that exceeded three and three-quarter ounces in weight. 
Halley relates that some hailstones were picked up on April 29, 1697, 
in Flintshire, the weight of which exceeded four ounces; and on May 
4, in the same year, Taylor found that the circumference of some that 
fell in Staffordshire was eleven and three-quarter inches. 

Volney tells us how, during the storm of July 13, 1788, he was stay- 
ing at Pontchartrain, ten miles from Versailles. The sun's rays were 
almost unbearable ; the air still and suffocating ; the sky was cloudless, 
and claps of thunder were from time to time audible. Toward 7*15 
P.M. a cloud appeared in the south-west, followed by a very sharp wind. 

HAIL. 397 

" A few minutes afterward the cloud filled the horizon and sped toward 
our zenith, accompanied by a wind which had become quite cool ; hail 
began to fall obliquely at an angle of 45°, the stones being as large as 
pieces of plaster thrown down from the top of a house. I could scarce- 
ly believe my eyes ; several of the stones were as large as a man's fist, 
and some of these were but pieces that had been broken off stones still 
larger. When I ventured to put out my hand beyond the door of the 
house where I had taken refuge, I picked up one and found that it 
weighed more than five ounces. It was very irregular in shape, there 
being three protuberances, thick as the thumb and nearly as long, 
which projected from the main body of the stone!" 

Volta states that, during the night of April 19-20, 1787, among the 
enormous hailstones which fell in Como and the neighborhood, there 
was one which weighed nearly nine ounces. Parent, member of the 
Academy of Sciences, relates that hailstones as big as a man's fist, and 
weighing from nine and a half ounces to twelve and three-quarter 
ounces, fell in Le Perche on May 15, 1703. Montignot and Tressan 
picked up some at Toul on July 11, 1753, which had the shape of an 
irregular polyhedron, with a diameter of three inches. 

During a hailstorm at Constantinople on October 5, 1831, there fell 
stones weighing more than one pound, and larger than a man's fist. 
Analogous stones are said to have been picked up in May, 1821, at 
Palestrina (Italy). 

The following are, however, even more remarkable instances: On 
June 15, 1829, there was a hailstorm at Cazorta, in Spain, which crushed 
in houses ; some of the blocks of ice weighed four and a half pounds. 
For hailstones to attain such proportions, several must have become 
agglomerated together, either when they reached the ground or during 
their descent. This is, in fact, in accordance with experience. And 
this explanation is, therefore, specially applicable to the following cases, 
if, indeed, they be authentic: During the latter part of October, 1844, 
durins: a terrible hurricane which devastated the south of France, there 
fell hailstones weighing eleven pounds ; the town of Cette, in particular, 
was severely damaged ; men were struck to the ground as if they had 
been stoned, partition walls were blown down, and vessels sunk. 

It seems that there was a very singular hailstorm on May 8, 1802, a 
piece of ice having been picked up which measured more than three feet 
both in length and in width, with a thickness of two and a quarter feet. 
Dr. Foissac, who cites this fact, does not consider it to be an exaggera- 



tion ; and he adds, "M. Hue, a Catholic missionary in Tartary, relates 
that hailstones of a remarkable size often fall in Mongolia, and that some 
of them have been found to weigh twelve pounds. Daring a heavy 
storm in 1843 the noise as of a terrible wind was heard in the air, and 
soon after there fell in a field not far from our house a piece of ice larger 
than a millstone. It was broken up with a hatchet; and though the 
weather was very warm, it took three days to melt completely." 

Fig. 73.— Section of liailstones, showing their ordinary interior stnicture. 

If this be true, there is nothing improbable in the chronicle dating 
from Charlemagne, which relates that there fell hailstones fifteen feet 
wide by six long and eleven thick, nor in that of Tippoo Sahib, which 
speaks of a hailstone as big as an elephant. 

The shape of hailstones differs very much. They are, as a rule, 
round, spherical, more or less irregular, like peas, grapes, or nuts. Sev- 
eral are 'more elongated, like a grain of wheat, cornelian cherries, or 
olives. When very large, they are formed by the juxtaposition of 
crystallized particles. On July 4, 1819, during a nocturnal storm which 
spread over a large portion of Western France, Delcros picked up sev- 
eral entire spherical hailstones, in which was visible a first spherical 
nucleus of a somewhat opaque, whitish hue, offering the traces of con- 
centric strata. Around this nucleus was an envelope of compact ice, 
radiated from the centre to the circumference, and terminating upon 
the exterior with twelve large pyramids, between which were inter- 
calated smaller pyramids. The whole /ormed a spherical mass nearly 
three and a half inches in diameter. 

Some hailstones picked up on September 12, 1863, in a road to the 
south-west of Tiflis, drawinsrs of which were exhibited to the Academv 
of Sciences at St. Petersburg, were ellipsoidal in shape, and their sur- 



Fig. 74.— Section of a hailstone, enlarged. 

face was covered with a large number of small prominences. The poly- 
hedric tissue, examined through 
a glass, had the aspect of a series 
of six -fronted pyramids; and a 
section of the interior revealed 
the existence of a hexagonal net- 
work of meshes, which is repre- V\. 
sented in Fig. T-i. - 

On July 29, 1871, at 6 p.m., the 4- 
sun shining brightly, and there 
being hardly any clouds, a sound 
was heard at Auxerre, like that 
of a heavy luggage-train. A few 
flashes of lightning preceded the 
fall of the hail, which came down 
unaccompanied by any tempest or atmospheric disturbance. The hail- 
stones preserved their shapes when they reached the ground, which are 
represented in the four corners of Fig. 74, after the designs of M. 
Daudin. The two stones in the centre are those to which I alluded in 
connection with the Academy of St. Petersburg, and the remainder have 
been added as illustrative of the smaller and more usual size of hail- 
stones. During the same storm M. Parent remarked at Montargis that 
there was a heavy fall of hail at 6*45 p.m., the pieces of ice being from 
one to two inches in length, oval in shape, and transparent as crystal. 

During the storm of May 22, 1870, in Paris, M. Trecul, of the Insti- 
tute, noticed that several of the hailstones were conical, or rather pyri- 
form — that is, larger at the base than at the top, some of them being 
about three-quarters of an inch long by half an inch wide. One of 
them, carefully examined, presented characteristics worthy of notice. 
The third part of it, at the top (the narrowest portion of the hailstone), 
was opaque and white; while the lower, or the broadest, part was per- 
fectly translucid, like the purest ice. In addition, this hailstone, when 
looked at from its broadest end — that is, when the narrowest diameter 
was placed crosswise in respect to the visual axis — presented the shape 
of an obtuse-angled rhombus; and from the sides there started oblique 
facets which converged and died away toward the obtuse summit of the 

As to the epochs of hailstorms, it is generally known that they occur 
in summer and in the afternoon — that is, when the meteorological con- 
ditions mentioned above happen together — viz., great heat upon the 



Fiyr. 75. — Diflferent forms of hail. 

surface of the ground, which diminishes rapidly with increase of eleva- 
tion, and which is accompanied by a considerable evaporation from the 
clouds under the action of the sun. As, however, the mere collision of 
a very cold upper wind with a very warm wind at the same altitude may 
produce hail, it occasionally falls in winter and at night; but this is of 
rare occurrence. 

Meteorologists often class together hoar-frost and hail, and hence as- 
sert that these aqueous meteors occur oftener in winter and spring than 
in summer and autumn. But hoar-frost differs from hail, not only 
from being divided into so much smaller particles, but in its mode of 
formation, for it does not spring from the bosom of the clouds, nor does 
it necessitate great atmospheric movements. It is merely frozen rain, 
or a rou.srh-ofrained and dense snow. 




Apart from the ordinary showers, more or less heavy, of rain, snow, 
or hail, which we have been considering above, the history of meteors 
is supplemented by certain extraordinary showers which have often in- 
spired the ignorant a»d credulous with terror, who have seen in them 
direct manifestations of God's anger. 

I do not refer to stones falling from the sky, the aerolites, which 
Greek philosophers looked upon as fragments detached from the celes- 
tial vault, but which are, as we have seen, cosmical corpuscles circu- 
lating in space. Nor will we deal with the showers of stones, bricks, 
planks, and earthenware, which are caused by whirlwinds. But we will 
just glance at certain phenomena which we have not yet taken notice 
of. We will begin by the Showers of Blood. 

Homer relates how a shower of blood fell upon the heroes of Greece, 
as a presage of death for many of their number, Obsequens cites the 
following : After the capture of Fidenes, in the year 14 of the Romish 
era, drops of blood fell from the sk}^, to the great surprise of all men. 
In 538 a heavy shower of blood fell over the Aventine Hill and at 
Aricia. In 570 and 572 it rained blood for two days upon the Squares 
ofYulcan and Concordia; in 585 during one day. In 587 this prodi- 
gy occurred in several districts of the Campagna, upon the territory of 
Preeneste ; in 626 at Ceres, in 648 at Rome, in 650 at Duna, in 652 in 
the neighborhood of the Anio. There was a shower of blood when 
Tatius was murdered. Plutarch speaks of showers of blood after great 
battles — in the Cimbric war, for instance, after the massacre of so many 
thousand Cimbri upon the plains of Marseilles. He admits that the 
bloody vapors distilled from the corpses and diluted in the clouds 
would lend to these their crimson inge. The following are the show- 
ers of blood which, principally by aid of the researches made by M. 
Grellois, I have succeeded in collecting as having occurred since the 
commencement of the Christian era down to the close of the last cen- 
tury. In the first instance, Gregory of Tours relates that in the year 



582 A.D. "a shower of blood fell over the district about Paris. Manv 
persons had their clothes stained with it, and cast them ofi" in terror." 
An analogous shower is said to have taken place at Constantinople in 
652. In 654 the sky seemed on fire in Gaul, blood descending from 
the clouds in large quantities. In 787 Fritsch mentions a shower of 
blood in Hungary, followed by the plague. Others were witnessed at 
Brixen in 869, and at Bagdad in 929. In 1117 there occurred strange 
phenomena, showers of blood, and subterraneous noises, which scat- 
tered terror throughout Lombardy during the struggle for freedom 
there, and a meeting of Bishops took place at Milan to consider their 
origin. The same phenomenon was remarked at Brescia for three 
days and three nights before the death of the Pope, Adrian II. In 
1141: there were several showers of blood in Germany; in 1163 at La 
Eochelle. In 1181, during the month of March, -there was a constant 
rain of blood for three days in France and Germany : a luminous cross 
was visible in the skies. Toward the end of 1543 blood fell at the cas- 
tle of Sassemburg, near Barendorf, in Westphalia ; in 1580 at Louvain. 
In 1571 there fell near Einden, during the night, so much blood that 
over a space of five or six miles the grass and clothes exposed had 
assumed a dark purple hue. Many persons preserved some of it in 
vessels. It was attempted, but unsuccessfully, to show that this prodigy 
was due to the rising into the air of the vapor from the blood of oxen 
that had been killed. No other explanation was found more deserving 
of credit among natural causes. These phenomena were also noticed 
at Strasbourg in 1623, at Tournay in 1638, and at Brussels in 1640. 

We learn from the records of the Academy of Sciences that on 
March 17, 1669, at 4 a.m., there fell in several parts of the town of 
Chatillon-sur-Seine a kind of rain or reddish liquor, thick, viscous, and 
putrid, which resembled a shower of blood. Large drops were seen 
imprinted against walls, and one wall was even splashed all over on 
both sides, " which would lead one to believe that thjs rain was com- 
posed of stagnant and muddy waters, carried into the air by a hurricane 
out of some neighboring marshes." There was a shower of blood at 
Venice in 1689. 

In 1744 there fell a red rain in the Faubourg of St. Peter d'Arena, 
at Genoa, which, on account of the war then going on in the territory 
of the Republic, terrified the inhabitants very much ; but it was subse- 
quently ascertained that this tint was due to some red earth which a 
strong wind had carried into the air from a neighboring mountain. 


History speaks of showers of blood at Cleves in 1763, in Picardy in 
1765, and in Italy in 1803. Eain of a red color has been observed oft- 
en enough in our own day to prevent there being any doubt as to the 
reality of the phenomenon, and the only mistake of our forefathers was 
in assigning it a supernatural origin. Bede was of opinion that a rain 
thicker and warmer than usual might become blood-red, and so deceive 
the uninstructed. Kaswini, El Hazen, and other savans of the Middle 
Ages, relate that about the middle of the ninth century there fell a red 
powder and a matter resembling coagulated blood. These philosophers 
were thus on the road to a reasonable explanation ; they saw in it only 
a resemblance which might be correct, and not a reality which is re- 
pugnant to the simplest logic. " What the vulgar call a shower of 
blood," says G. Schott, " is generally a mere fall of vapors tinted with 
vermilion or red chalk. But when blood actually does fall, which it 
would be difficult to deny takes place, it is a miracle due to the will of 
God." Eustathius, the commentator of Homer, says that in Armenia 
tbe clouds discharge showers of blood because this country contains the 
Cinabrian mines, the dust of which, mixed with water, colors the drops 
of rain. 

Conrad Lycosthenes, in his " Book upon Prodigies," represents the 
showers of blood and the showers of crosses in the shape of childish 
figures, which give us an idea of the simple-mindedness prevalent in 
those days. 

In the early part of July, 1608, one of these pretended showers of 
blood fell in the outskirts of Aix (Provence), and this shower extended 
to the distance of half a league from the town. Some priests, either 
being themselves deceived or wishing to work upon the credulity of 
the people, at once attributed it to diabolic influence. Fortunately, a 
person of education, M. de Peiresc, examined very minutely into this 
apparent prodigy, studying in particular some drops that fell upon the 
wall of the cemetery attached to the principal church in Aix. He soon 
discovered that they were in reality the excrements of some butterflies 
which had been noticed in large numbers during the early part of July. 
There were no spots of the kind in the centre of the town, where the 
butterflies had not made their appearance, and, moreover, none were 
noticed upon the higher parts of the houses, aboye the level to which 
they flew. Besides, the presence of these drops in places protected 
from the air rendered it impossible that they could have their origin 
in the atmosphere. He at once pointed this out to those who regarded 



the occurrence as miraculous; but, in despite of the proofs which he 
adduced, the inhabitants persisted in attributing these drops to a su- 
pernatural cause. 

Fig. 76. — llaiu of bluoj iu Provence, July, 1608. 

Eeaumur gives the butterfly known as "the great turtle" as being 
the most capable of depositing these drops. " There are thousands of 
others," he says, " which turn into chrysalises toward the end of May 
or the beginning of June. When this transformation is about to take 
place, they leave the trees and often take refuge upon walls, entering 
houses, hanging on to the arch of a door-way or a plank. If the but- 
terflies which emerge from them at the end of June or the beginning 
of July flew iu masses together, they would' be numerous enough to 
form small clouds, and consequently to cover the stones in certain 
places with spots of a blood-red color, and thus to make the timid be- 
lieve that they were spectators of a supernatural occurrence." Gen- 


erally speaking, showers of blood are not only red spots produced by 
certain insects, but regular showers, colored by the dust which the wind 
carries into the air. This general origin was not ascertained until the 
present century. On March 14, 1813, one of these strange red showers 
fell in the kingdom of Naples and the Two Calabrias. Sementina ex- 
amined and analyzed it, rendering the following account to the l^^aples 
Academy of Sciences: "An east wind had been blowing for two days, 
when the inhabitants of Gerace noticed a dense cloud moving toward 
the sea. At 2 p.m. the sea became calm, but the cloud already covered 
the neighboring mountains and began to intercept the light of the sun. 
Its color, originally a pale red, soon became deep as fire. The town 
was then plunged into such profound darkness that, about 4 p.m., it 
was necessary to light candles in the houses. The inhabitants, alarmed 
by the obscurity and the color of the cloud, rushed in crowds to the 
cathedral to pray. The obscurity increased, and the whole sky seemed 
red as fire; thunder began to growl; and the sea, though six miles dis- 
tant, added to the general alarm by the roar of its waves. There then 
began to fall large drops of reddish rain, which many persons took for 
blood, and others for fire. At last, as night advanced, the air became 
clear, the thunder and lightning ceased, and the inhabitants regained 
their self-possession." 

With the exception of there being no popular alarm, the same phe- 
nomenon of a shower of reddish dust occurred not only in the Two 
Calabrias, but also at the opposite extremity of the Abruzzes. This 
dust was of a yellowish hue, like cinnamon, and had a slight earthy 
taste ; it was unctuous to the touch, and, seen through a glass, con- 
tained small and hard bodies resembling pyroxene. Heat at first 
embrowned it, then made it black, and finally gave it a reddish tint. 
After the action of the heat, this dust displayed, even to the naked eye, 
an immense number of small and brilliant points, which were of yellow 
mica. Its specific gravity, when deprived of hard substances, was 2 '07 : 
it was composed of silica, 33-0; aluminium, 15*5; lime, 11-5; chrome, 
I'O ; iron, 14-5 ; and carbonic acid, 9-0. 

Whence came this dust? This it was found impossible to ascertain 
at that time. It was not until 1846 that a general examination of these 
rains was made, and their origin found by following them up into 
space. On May 16 in that year an earthy rain fouled all the water at 
Syam (Jura). In the autumn of the same year there was a similar fall, 
accompanied by lightning, diluvian rain, very disastrous hurricanes^ 


etc., which occurred alternately, or nearly so, over a large circular tract 
of country, in such a way as to be only explicable by some great dis- 
turbance in the system of the trade-winds. The cyclones also swept 
over the Atlantic; amidst fearful squalls, whirlwinds, and hailstorms, 
vessels were dismasted and their decks swept clean. Then also oc- 
curred severe tempests in France, Italy, and at Constantinople ; while, 
farther eastward, the typhoons spent their fury in the China seas. The 
winds were sufficiently intense to detach a stratum of land in districts 
where the surface of the ground was sandy or of some other soft sub- 
stance. This earth, carried into the air, was, of course, certain to be 
deposited somewhere. This took place in the south of France, between 
Puy and Mont Cenis, in the direction of the prevailing wind, and cross- 
wise from Bourg to Drome. The quantity of earth precipitated varied, 
however, according to the locality ; at Lyons, in fact, it was scarcely 
apparent, though it occurred in the shape of a reddish slime which was 
popularly converted into a shoiver of blood. But at Meximieux a bat- 
talion of soldiers marching toward the Swiss frontier were covered with 
the mud, and their uniforms impregnated with it. The Chateau de 
Chamagnieu was bespattered in such a way that it could scarcely be 
recognized, and there was such a thick laj^er at Yalence that the inhab- 
itants were compelled to clean water-shoots and gutters. Fournet gives 
a calculation which shows that in the department of the Drome the 
clouds must have taken up from and again discharged upon the ground 
the enormous weight of 720 tons, which represent 180 four-horse wag- 
on-loads. Ehrenberg, who analyzed samples of this earth, found in 
them seventy-three organic formations, some of which were peculiar to 
Southern America. This earth must, therefore, have come from the 
New World. The interval of time between their leaving America, Oc- 
tober 13, and their arrival in France, October 17, was about four days, 
which gives a speed of eighteen and three-quarter yards per second. 

Subsequent to that date we have had a remarkable fall of colored 
rain in the neighborhood of Chambery, on March 31, 1847. It was 
imbued with a milky matter, which seemed like thin clay suspended in 
the air. The clothes of persons exposed to this rain were bespattered 
with whitish spots. Information from Savoy and tlie Great St. Ber- 
nard came to hand soon after this, stating that there had been a fall of 
earthy red snow, coming from the south-west, and covering the ground 
to the depth of several inches. 

This coloring of the snow by the dust must not be confounded with 


a hue •which, it often derives from a small insect which lives in it — 
uredo nivalis — a kind of microscopic infusorj often extraordinarily nu- 
merous in the Alps and the Polar regions. 

At the period of the red rain in 1847 cited above, the falls of snow 
extended over a large portion of France — at Orleans, at Paris, in the 
Vosges, and La Bresse ; and there were hurricanes at Havana, Bahama, 
the Azores, Newfoundland, the Sorlingues, Portugal, and Spain. There 
were numerous atmospheric whirlwinds in the north and the west, at 
Le Havre, Paris, and at Grignan, no less than twenty-four storks fall- 
ing dead at this place. At Nantua, a whirlwind, which carried a sen- 
try-box ten feet into the air, covered the streets with debris of tiles, 
chimneys, and windows. The numbers given by Fournet show a very 
rapid and marked depression of the barometer on March 81, followed 
by a still greater decrease on April 2, 

There was also a very remarkable shower of earth on March 27, 
1862. The residue, when moist, was, like that of 1846, so far red in 
hue as to revive the popular belief about a shower of blood ; when dry, 
the earth was fine and yellowish. Ehrenberg discovered in it forty- 
four organic forms, among which were those microscopical galionelles, a 
cubic inch of which may contain 466,000. 

The shower which fell at Beauvais in May, 1863, from 5 to 11 A.M., 
was also very remarkable, the spots which it left upon clothes being as 
marked as in the preceding cases. 

About 3 A.M. on the morning of May 1, a violent thunder-storm 
broke over Perpignan, and afterward a reddish dust was noticed in 
several parts of the town, which, it was subsequently ascertained, must 
have fallen during the storm. The same storm extended to the level 
district in the department of the Eastern Pyrenees ; but here the phe- 
nomenon witnessed was a fall of red snow, and the appearance of these 
flakes alarmed the inhabitants. The occurrence was also noticed on 
many coast-towns of the Mediterranean. There was discovered in 
them a dust of marshy and ferruginous clay, mixed up with fine sand, 
which, as it passed through the atmosphere, deprived it of a portion of 
the organic matters in suspension there. In this way these rains serve 
a fertilizing purpose, being in fact showers of manure. Each heavy gust 
of wind raises clouds of dust, as may especially be remarked when, ani- 
mated by a gyratory movement, it possesses a certain force of aspira- 
tion which enables it to form those small whirlwinds of dust which 
may be seen upon the high-roads. 



The whole extent of the vast zone of deserts which reaches over the 
intertropical and the subtropical countries of the Old as of the New 
World contains earthy elements, which the wind drives to an immense 
distance. Europe, like Asia, Africa, and America, furnishes the wind 
with a supply of this kind. 

We have already pointed out the powers of whirlwinds. To cite 
but that of 1780: it developed its force near Carcassonne, upon the 
banks of the Aude, raised high into the air immense quantities of sand, 
unroofed eighty houses, and blew in all directions stacks of wheat 
standing in fields. Large ash-trees were uprooted, and their biggest 
branches carried to a distance of forty yards. Such a power amply ex- 
plains the fact of earth and sand being taken so much farther. The 
shower of blood which fell at Sienna on December 28-31, 1860, ana- 
lyzed by D. Campani, seemed to be of organic origin. 

One of the latest showers of blood recorded is that which occurred 
on March 10, 1869. On this day the sirocco was blowing at Naples, 
and its squalls were accompanied by that nebulosity which is peculiar 
to it, and which resembles a slight mist ; the barometer had fallen con- 
siderably ; the weather was very warm, and from time to time there 
fell sharp but short showers, either of very fine rain or in large drops ; 
each drop of this rain left a muddy spot behind it. 

These spots, when examined carefully, had a marked yellowish brown 
tint, and resembled spots left by water containing iron. A sheet of 
white paper, first damped and then exposed to the wind, was soon cov- 
ered with a number of small and reddish grains, nearly spherical in 
shape, the diameter of which varied from 0'004 inch to 0-0004 inch. 
There can be no doubt, considering the direction of the wind at the 
time, that these grains of sand came direct from the desert of Sahara. 

M. Breton, of Grenoble, noticed that this residue was exactly analo- 
gous to that which was picked up at Valence in September, 1846, after 
the red rain spoken of above. As was imagined, this sand came from 
Sahara. It appears from another account that Algeria was the theatre 
of a very violent hurricane on March 3, 1869. 

French soldiers were overtaken by the wind, near El-Outaia, in the 
midst of a sea of sand. It took them four hours to travel six and three- 
quarter miles, " During the seventeen years that I have been in Al- 
geria," says an eye-witness, " I have never seen such a whirlwind. Our 
little column was compelled to stop and to take precautions against be- 
ing killed. At the second halt we turned our backs to the squall, and 


for an hour and a half we could see neither the sun nor the sky, al- 
though just before there had been scarcely any clouds. For more than 
a quarter of an hour together we could not see a distance of two or 
three yards in front of us." 

The red rain which fell at Naples had undoubtedly been brought 
from the desert of Sahara, itself exposed to a tempest which in fact ex- 
tended over all Europe, the Mediterranean, and Africa. 

These phenomena are intimately connected with the great movements 
of the atmosphere, as M. Tarry has judiciously pointed out. 

Ten days after the red rain mentioned above, on the 20th of March, 
a violent tempest, coming from England, swept over the north coast of 
France. There was a very marked centre of atmospheric depression 
(28.90 inches) at Boulogne on the 20th ; by the next day it had reach- 
ed Lesina, upon the Adriatic. For several days a violent north-west 
wind raged over France, and afterward over Italy. On the 22d the cy- 
clone had reached Africa, where it raised into the air the sands of Sa- 
hara ; a retrograde movement then took place ; a fresh decrease of the 
barometer reading occurred in the south of Europe, where the pressure 
had risen after the passage of the cyclone. On the 24th the barometer 
fell to 29"13 inches at Palermo, and 29"2J. inches at Rome: the wind 
grew very violent; the instrument of Father Secchi, in the latter city, 
indicating a speed of 640 miles in the twenty-four hours — the greatest 
of the year. 

The atmosphere in Sicily was noticed, on the 23d, to be laden with 
thick clouds and a yellowish dust, which lent to the sky an unusual ap- 
pearance. Rain falling, each drop left a yellow residuum, which it 
needed two or three filterings to remove. This substance, analyzed by 
Professor Silvestre, at Catania, contained the following elements: clay, 
chalky sand, peroxide of hydrate of iron, nitrogenized sodium, silica, 
and organic matter. 

The same phenomenon was remarked at Subiaco, near Rome, and at 
Lesina, in Illyria. Thus the prodigies spoken of by Livy are now reg- 
istered at the Paris Observatory. 

The last remarkable red rain was that of February 18, 1870. On 
February 7 a great barometrical depression occurred in England; the 
barometer marked 29.33 inches at Penzance ; on the 9th it had reached 
the Mediterranean ; on the 10th, Sicily, where the barometer reading 
was lower than at Rome. This foil of the barometer was accompanied 
by a violent tempest ; at Rome there was a violent north wind for three 


days — the 8th, the 9th, and the 10th. It superinduced a severe frost in 
France and Italy, snow falling in Rome on the nights of the 8th and the 
9th. On the 11th and 12th the weather was calmer, and the barometer 
reading increased again, the cyclone raging over the desert of Sahara. 
The retrograde movement alluded to above soon made itself manifest. 
On the 12th the barometer fell to 2945 inches in the south of Spain; a 
violent wind from the south blew over Spain and Italy on the 13th and 
14th ; and from Africa the cyclone, accompanied by the hurricane, again 
made its way back to Europe, with the sand swept up from Sahara. As 
a matter of fact, at 2 p.m. on the 13th of February, a reddish sand was 
remarked in the rain that fell at Subiaco, near Rome, by M.Alvarez; 
at Tivoli, by Father Ciampri ; and at Mondragone, by Father Lavaggi. 
In the night of the 13th to the 14th there fell at Genoa an earthy and 
reddish substance; and at Moncalieri, Father Denza, Director of the 
Observatory, picked up some red snow which contained the same kind 
of sand. 

This recital of the showers of blood shows us — 1st, that they are a 
reality ; 2d, that they are mostly due to dust taken up by the wind into 
very distant regions ; 3d, that they are not so infrequent as they appear 
to be. Thus there are no less than twenty-one occasions upon which 
they have been known to occur during the present century in Europe 
and Algeria, as the following table will show : 

1803. Februaiy Italy. 

1813. February Calabria. 

1814. October Oneglia, between Nice and Genoa. 

1819. September Studein, MoraAna. 

1821. May Giessen. 

1839. April Philippeville, Algeria. 

1841. February Genoa, Parma, Canigon. 

1842. March Greece. 

1846. May Syam, Chambe'ry. 

1846. October Dauphine', Savoy, Vivarais. 

1847. March Chambery. 

1852. March Lyons. 

1854. May Horbourg, near Colmar. 

1860. 31stDecember Sienna. 

1862. March Beaunan, near Lyons. 

1863. March Rhodes. 

1863. April Between Lyons and Aragon. 

1868. 26th April Toulouse. 

1869. 10th March Naples. 

1869. 23d March Sicily. 

1870. 13th February Rome. 


It will be noticed that these remarkable showers mostly take place 
in the spring and the autumn, at the epoch of the equinoctial gales. 
We have seen that they may be due to the traces left by certain kinds 
of butterfly. A third cause must also be noticed — viz., volcanoes, the 
ashes of which are sometimes conveyed by the winds to an immense 
distance. Several cases in proof of this might be adduced, 
/ We now come to another series of remarkable showers spoken of in 
ancient legends, exaggerated and interpreted in different ways, and the 
true explanations of which it is not always easy to give. 

Showers of milk are often spoken of as having taken place. Thus 
Obsequens relates that upon the territory of Veies there fell a shower 
of milk and oil in 629. The absence of all definite information upon 
facts of this kind prevents one from doing more than hazard a few con- 
jectures borrowed from volcanic eruptions or the carrying into the air 
of white or chalky earth by some hurricane. In 620 streams of milk 
are said to have flowed into the Eoman lake. In 6-13 milk is reported 
to have flowed for three days in some place not mentioned; numerous 
victims were immolated when this prodigy took place. These so-called 
streams of milk are a common phenomenon in some countries; the 
washing of the rain over a white soil suffices to cause this illusion, 
which, however, the most cursory analysis would dispel. 

Dion Cassius speaks of a rain that looked like milk, and which, fall- 
ing on coins or copper vessels, made them retain the appearance of sil- 
ver for three days. If this fact be true, it is clear that it must have 
arisen from a downfall of sublimated mercury which had become con- 
densed, and consequently had fallen to the ground. But in what way 
this sublimation and condensation was brought about, it is first neces- 
sary to ascertain, before believing in the occurrence of this prodigy. 

Glycas also speaks of a shower of mercury, which might be the same 
as the above, though it is stated to have taken place during the reign of 

We may compare with these showers a phenomenon which has been 
observed too often to permit of its reality being questioned. I allude 
to the appearance of crosses upon men's clothes, a few instances of which 
I append : 

In 764 the misbehavior of the monks of St. Martin drew down the 
anger of God. Blood fell from the heavens on to the earth, and crosses 
appeared upon men's garments. — Gregory of Tours. 

Fritsch speaks of the same phenomenon as occurring in 783. In 


1094 crosses fell from heaven on to the garments of priests, for the pur- 
pose, no doubt, of warning them of their impiety, says Gr. Schott. In 
1534 there fell in Sweden a shower which left the mark of a red cross 
upon men's garments. Cardan explains this phenomenon by the state- 
ment that red dust was diluted in the rain-water, and that the crosses 
were formed by the drops falling in the woof of the cloth. Fromond 
and Schott do not accept this explanation, because, according to them, 
these crosses were formed not only upon certain parts of the garment, 
but all over it, and that when drops of blood fall upon a piece of cloth 
they never take this shape. The pious of that date considered it to be 
a direct intervention of the Deity. But this is not all. It is related 
that in 1501 crosses fell in Germany and Belgium, not only upon the 
garments, even when inclosed in boxes., and especially upon the garments 
of women, but that they left a mark upon the skin, and upon bread. 
This prodigy lasted three years, recurring during Passion-week and 
Easter ; no doubt, adds the chronicler, to inspire the respect too often 
forgotten to the blood and cross of the Lord. 

John of Horn, Prince of Liege, told the Emperor Maximilian I. of a 
young woman of that town, twenty-two years of age, whose garments 
were perpetually covered with blood-red crosses, although she continu- 
ally changed her clothes. 

It must, at the same time, be mentioned that many instances are cited 
in which nutritious substances have descended in a shower. Thus in 
1824 and 1828 there was so abundant a shower of this kind in one of 
the districts of Persia that it covered the ground to the depth of five or 
six inches. It was a kind of lichen, of a sort already known ; cattle 
and sheep devoured it greedily, and some bread was even made from it. 

We may also class with the preceding the descent of a soft substance 
which Muschenbroeck states to have occurred in Ireland in 1675. This 
was a glutinous and fat substance, which softened when held in the 
hand, and emitted an unpleasant smell when exposed to the action of 

On the 10th of March, 1695, at about 7 p.m., a heavy storm burst 
over Chatillon-sur-Seine : the front part of the cloud appeared inflamed, 
the air to be on fire, and the spectators who saw it believed that the 
neighboring villages were being burned, as sparks of flame fell to the 
ground in all directions. This shower lasted a quarter of an hour, and 
extended over a large tract of country, where it caused no conflagration ; 
immediately after the storm there was a heavy fall of large snow-flakes. 


In 828 there fell from the sky a number of grains like those of wheat, 
but much smaller. 

This fact may easily be credited, as also the following, which is told 
by Johnston : There fell for the space of two hours, over a tract of 
country two miles in extent, in Carinthia, a shower of wheat with which 
bread was afterward made. 

We may also accept the statement of Cassiodorus, that there fell in 
371 a shower of rain, in the country of the Atrebates, in which there 
was a plentiful admixture of wool. 

The showers of sulphur, which are often spoken of, are, as a rule, 
nothing more than the pollen of certain plants, pine and nut trees in 
particular, which may be carried by the wind to an immense distance. 
Without going so far back as the storm of sulphur which destroyed 
Sodom and Gomorrah, there are certain storms of the kind which ap- 
pear well authenticated. Olaus Wormius states that on May 16, 1646, 
there fell a heavy shower at Copenhagen which inundated the whole 
city, and contained a dust exactly like sulphur, both in regard to color 
and smell. Simon Paulli states that on May 19, 1665, there raged in 
Norway a fearful tempest, with a dust so like sulphur that, when thrown 
into the fire, it produced the same smell, and that, when mixed with 
spirits of turpentine, it produced a liquor the odor of which was just like 
that of balm of sulphur. The close neighborhood of the Iceland vol- 
canoes is sufficient to explain this occurrence. Phenomena of the same 
kind are not infrequent in Naples. Sigesbek, in the " Breslau Memoirs," 
speaks of a shower of sulphur which fell in Brunswick, and which was 
a regular mineral sulphur. This fiict can not be accepted without further 
proof: as to the showers of pollen, flowers, leaves, etc., they are well 

At Autreche (Indre et Loire), at 12-10 p.m. on April 9, 1869, the air 
was very still, and the sky cloudless. M. Jallois relates that one of his 
correspondents remarked a shower of dry oak-leaves falling from the 
higher regions of the atmosphere. Being gifted with excellent sight, 
he saw them appear like bright specks upon the azure of the sky, at a 
very great height, and fall about him, after having descended almost 
vertically, with a trifling inclination eastward. This continued for ten 
minutes ; but the shower of leaves had probably commenced previous- 
ly. There was at least one to each square yard upon a piece of water 
close by. 

This phenomenon seems to have resulted from a great squall which 


occurred on April 3 ; the oak leaves carried up by a hurricane into the 
higher regions of the atmosphere were kept there by the wind for six 
days, and fell again when the weather became calm. 

This shower of leaves reminds me of a shower of oranges. On July 
8, 1833, a water-spout, which took place at Pausilippus, near Naples, 
burst upon the shore and swept off two large baskets of oranges; a 
few minutes afterward they descended to the ground at some distance. 

After the vegetable showers we come to a series even more remark- 
able, and perfectly well authenticated. I refer to the showers of live 

In the chapter on water-spouts we have seen that fish are sometimes 
taken up in this way out of a pond. Peltier relates that frogs once fell 
upon his head from a water-spout. This was at Ham, in 1835, and the 
fact was duly certified. I may cite another, still more recent. 

In the morning of January 30, 1869, toward 4*30 a.m., after a vio- 
lent gust of wind, there began a fall of snow which lasted until day- 
light, at Arache, in Upper Savoy ; and in the morning a large number 
of live larvas were found in the snow. They could not have been 
hatched in the neighborhood, for, during the days preceding, the tem- 
perature had been very low. On January 24 the thermometer had 
marked 60*8°, and upon the following days a temperature of 41° at 
7 A.M. They seemed to be mostly the Trogosita inaun'tanica, which is 
common in the forests of Southern France. There were also found 
among them a few caterpillars of a small butterfly belonging to the 
noduelian tribe, probably the Stibia stagnicola. This caterpillar reaches 
its full size in the course of February, and is indigenous to the centre 
and the south of France. 

This shower of insects at Arache, at an altitude of from 1000 to 
1200 yards, can only be explained by a violent wind which must have 
brought them from some locality in the south of France. 

M. Tissot, the village school-master, who observed this phenomenon, 
adds, that in the course of November, 1854, the wind being very vio- 
lent, thousands of insects, most of them alive, alighted upon a planta- 
tion near Turin-; some of them were larvre, and others had attained 
their full growth, while all belonged to an order of hemiptera which 
are nowhere seen except in the island of Sardinia. Ancient authors 
have related several instances of falls of insects. 

Phanias, cited by Porta, states that there fell a shower of fish for 
three days in the Chersonesus. 


In Athens, Philarcus asserts that he saw large quantities of fish and 
frogs fall from the sky in many different places. Heraclides Lembus, 
in Book XXI. of his " Histories," says that God sent showers of frogs 
upon Poenia and Dardania in such large quantities that the houses and 
roads were covered with them. They were found mixed up in the 
food, and were consumed with it. The water was filled with, them ; it 
was impossible to walk without treading upon them. The decomposi- 
tion of their bodies produced such an odor that it was found necessary 
to quit the country. 

Varro declares that all the inhabitants of a certain town in Gaul 
were driven from their houses on account of the countless frogs which 
fell from the sky. 

Scaliger states that the town of Mirabel, in Aquitania, was filled with 
half-formed frogs which fell from the sky. Johnston relates that, in 
the island of Auckland (Friesland), "in which there are no frogs," a 
number fell in a shower of rain. Olaus Magnus also states that frogs, 
worms, and fish fall from the clouds in the north oftener than in the 
south, "on account of the viscosity of the clouds and the heat which 
they derive from the sulphurous principle !" 

Fromond relates that, while standing with several friends at one of 
the gates of Tournai, in 1625, a shower of rain suddenly fell, and pro- 
duced so many frogs, all of the same size and color, that the ground 
was covered with them. 

Porta says that he often saw, between Naples and Pouzzoles, a quan- 
tity of frogs suddenly emerge from the dust upon which a heavy show- 
er of rain had just fallen. This peculiarity, he adds, is well known to 
many inhabitants of these two towns. 

These sudden appearances of frogs and toads are generally due to 
the fact that these animals mostly issue from the mud after a thunder- 
storm, and are in the habit of crossing frequented routes. It is excess- 
ively rare for whirlwinds to carry up into the air either fishes or frogs. 

The showers of locusts are due to flying masses of orthoptera, the 
nomad cricket in particular. These insects are a scourge to agricul- 
ture. They are brought by the wind ; and when they alight, they 
transform a fertile region into a desert. Seen from a distance, their 
countless swarms present the appearance of thunder -clouds. These 
dark masses hide the sun. As far and as high as the eye can reach, 
the sky is black and the ground covered with them. The sound of 
their million wings is like the noise of a cataract. As they reach the 


o-round, tliey break the branches of the trees. In a few hours all signs 
of vegetation have disappeared over an extent of several leagues. The 
wheat is gnawed to its roots, the trees are stripped of their leaves. Ev- 
ery thing is destroyed, sawn, cut to pieces, and devoured. When noth- 
ino- is left, the terrible swarm rises, as if at a given signal, and flies off, 
leaving famine and desolation behind it. 

It often happens that, after having consumed every thing, they die 
of starvation before depositing their eggs. Their bodies, heated by the 
sun, soon become putrefied, and emit exhalations which breed terrible 
epidemics in the district. 

In 1690 locusts arrived in Poland and Lithuania from three different 
points, and in three distinct masses. The Abbe de Ussans, who saw 
them, says, "At certain places where they had died in large quantities 
they lay four feet deep. Those which were still alive, and which had 
settled upon the trees, made the boughs bend beneath their weight." 

In 1749 locusts arrested the march of Charles XII. 's army when it 
was retreating through Bessarabia after the defeat of Pultowa. The 
king thought that it was a hailstorm which was thus swooping down 
upon his army. The arrival of the locusts was announced by a hissing 
sound like that which precedes a tempest, and the rustling of their 
flight drowned the sound of the waves of the Black Sea. All the 
country in their track was laid bare. 

In the south of France locusts sometimes multiply at such a prodig- 
ious rate that they soon produce enough eggs to fill several barrels. 
They have at times caused terrible damages ; notably so in the years 
1805, 1820, 1822, 1824, 1825, 1832, and 1834. 

Mezeray states that in January, 1613, during the reign of Louis XIII., 
locusts invaded the district round Aries. In seven or eight hours all 
the wheat and forage were devoured to the very roots over 20,000 acres 
of ground. They then crossed the Ehone and visited Tarascon and 
Beaucaire, where they consumed the garden produce and the lucerne. 
They went from thence to Aramou, Monfrin, Yalebregues, etc., where 
most of them were, fortunately, destroyed by starlings and other insect- 
eating birds which had been attracted thither by the prospect of such 
a banquet. The consuls of Aries and Marseilles had their eggs picked 
up. It cost the former town 25,000 and the latter 20,000 francs, and 
3000 cwt. of eggs were thrown into the Rhone. Counting 1,750,000 
eggs to the cwt., 5,250,000,000 of locusts, as they would afterward have 
become, must have been destroyed. 



Fig. 77.— Shower of locusts. 

In 1825, in the territory of Saintes - Maries, not far from Aigues- 
Mortes, upon the shores of the Mediterranean, 1518 wheat-sacks were 
filled with dead locusts, the weight of which was nearly sixty-nine 
tons; at Aries there were picked up 165 sackfuls, or between six and 
seven tons. 

Locusts are always to be met with in Algeria, in the provinces of 
Oran, Bone, Algiers, and Bougie ; but they are not so numerous as to 
produce those terrible invasions which change a fertile country into a 
desert There are locust years in Algeria, just as in France there are 
years when beetles, caterpillars, etc., are especially abundant. These 
scourges are, fortunately, very rare. The most disastrous took place in 
1845 and 1866. 

Eegular showers of beetles have also been known to descend like a 
thick cloud and cover the fields and the highways. 




As with the locusts, they swarm from one province into another. 
Masses of these coleoptera, which are not transported by a whirlwind, 
but which are generally driven by the wind, emigrate from a district 
after they have devoured every thing in it. 

To give an idea of the prodigious numbers in which cock-chafers 
sometimes make their appearance, I will quote some few historical 

In 1574 these insects so abounded in England that they stopped 
several mills on the Severn. 

In 1688 they formed so dense a cloud in Galway that the sky v/as 
darkened to the distance of a league, and the peasants had a difficulty 
in finding their way about. They destroyed all vegetation, so that the 
country around had the look of winter. Their voracious jaws made a 
noise like that caused by the sawing of a thick piece of timber; and 

Fi_'. T>.— Shower (if cnck-cliafers. 


in the evening the flapping of their wings resembled the distant rolling 
of a drum. The unhappy Irish were compelled to cook and eat them, 
for want of other food. In 1804: vast clouds of cock-chafers, precipi- 
tated by a violent wind into the Lake of Ziirich, formed a thick mass 
upon the shore, where their bodies were heaped up, the putrid exhala- 
tions from which poisoned the atmosphere. On May 18, 1832, at 9 p.m., 
a legion of beetles encountered a diligence upon the route from Gour- 
nay to Gisors (as it was leaving Talmoutiers) with so much violence 
that the horses, blinded and frightened, were compelled to return. 

Such is the series of showers of blood, earth, vegetables, and animals, 
which the history of meteorology has registered. We will stop here. 
Just as in the preceding chapter we saw that there were writers who 
spoke of hailstones as big as elephants, so too, in this case, there has 
been considerable exaggeration. Fabulous as may be the force which 
the wind sometimes acquires, we may assign to the domain of romance 
the story told by Avicenne, that prince of Arab doctors, as to his hav- 
ing seen the body of a calf fall from the skies. Nevertheless, Xavier 
de Maistre declares that a young girl was carried off by a whirlwind in 
1820; but it is not said to what height. Cabeus, in the seventeenth 
century, declared that a violent wind had blown away a woman who 
was washing linen in the lake. In regard to large animals, the most 
exaggerated story is the one which is also the oldest — viz., as to the 
Nemsean Lion falling from the moon on to the Peloponnesus. ... It 
is true that stones to the weight of hundreds of pounds sometimes fall 
from the sky, as we saw in regard to aerolites. But hitherto the other 
worlds have sent us nothing more valuable than stones. The animals, 
fish, insects, grains, and leaves, which fall from the sky come originally 
from the earth, not from any of the planets. 






We now come to the most marvelous and singular agent that exists, 
the study of which will complete and close the immense panorama de- 
veloped in this work — viz., electricity, thunder-storms, and lightning. 
The study of them is by no means devoid of complications ; but our 
close attention will be amply repaid by the wonderful spectacles which 
it will reveal. Following our general plan, we will see how it is dis- 
tributed over the earth and in the atmosphere. It is, however, first 
necessary to obtain an idea of its history, which is somewhat remark- 

Otto de Guericke, burgomaster of Magdeburg, the celebrated inventor 
of the pneumatic machine, first discovered (about 1650) some signs of 
electric light. Dr. Wall, at about the same epoch, by applying friction 
along 'a cylinder of amber, saw a bright spark emitted, and heard a 
sharp noise ; and, curiously enough, this first electric spark produced 
by the hand of man was at once compared to the lightning's flash. 
This light and this sound, says Dr. Wall, in his " Memoirs " (see " Phil. 
Trans."), seem to represent, in a certain measure, the lightning and the 
thunder. The analogy was striking, and needed only an effort of the 
imagination to be understood ; but to demonstrate its truth, to discover 
in so insignificant a phenomenon the causes and the laws of the great- 
est phenomena in nature, required a series of proofs which could only 
be expected from a great genius. Nevertheless, many physical philos- 
ophers endeavored to obtain them by comparisons of a more or less in- 
genious kind: some remarked that the spark is zigzag, like lightning: 
others opined that thunder in the hands of nature is the same as elec- 
tricity in the hands of man. "I confess," said Abbe Nollet, "that I 
should look upon this idea with great complacency if it could be well 



sustained ; and thers are many specious reasons by which it might be." 
Still, this was nothing more than a train of reasoning v/hich could not 
be conclusive, inasmuch as in physics experiment alone is absolutely 
decisive. While Europe and the whole of the Old World were thus 
reasoning, America was conducting experiments in special reference to 
the subject of lightning. Franklin succeeded in bringing electricity 
down from the sky, in order to investigate it by direct examination. 
After having made several discoveries in respect to electricity, especial- 
ly in regard to the Leyden jar and the attractive power of fine points, 
Franklin went in search of electricity into the very midst of the clouds. 
He had concluded, as the result of certain experiments, that a stem of 
pointed metal, placed at a great height, upon the summit of a building. 

Experiments of Frankliu and Komas. 

formed a receptacle for the electricity of thunder-clouds. He was await- 
ing, with no little impatience, the construction of a steeple then being 
built at Philadelphia ; but unwilling to remain so long in doubt, he 
had recourse to a more expeditious and not less certain method for as- 
certaining what he desired to know. As all that was necessary was to 
raise a substance of some kind into the region of the thunder — that is 
to say, high enough into the air — he thought that an ordinary kite 
would serve his purpose as well as any steeple. He accordingly ar- 
ranged two pieces of stick, laid crosswise and covered with a silk hand- 
kerchief, which he took into the fields upon the occasion of the first 
thunder-storm. Fearing the ridicule which failure would entail, he was 
accompanied only by his son. The kite remained some time in the air 


without any perceptible effect being produced ; but at last the fibres of 
the rope were somewhat agitated. Encouraged by this, Franklin placed 
his finger upon the end of the rope, a motion which immediately led to 
the appearance of a bright spark, which was soon followed by several 
others. Thus, for the first time, the genius of man succeeded in play- 
ing with the lightning and discovering the secret of its existence. 

Franklin's experiment took place in June, 1752, and was shortly af- 
terward repeated in every civilized country with the same success. A 
French magistrate, De Eomas, assessor to the Nerac Tribunal, profiting 
by the ideas of Franklin, which had been made public in France, con- 
ceived the idea of using the kite with raised bars; and in the month 
of June, 1753, before the result of Franklin's experiments was known, 
he had obtained very strong electric signs, because he had prudently 
attached a metal wire to the cord along its whole length, which meas- 
ured 850 feet, A little later, in 1757, Romas repeated these experi- 
ments during a thunder-storm ; and this time he elicited sparks of an 
enormous size. He says, " Imagine tongues of fire nine or ten feet in 
length, and an inch thick, which made as loud a report as a pistol. In 
less than an hour I had obtained at least thirty sparks of these dimen- 
sions, to say nothing of a thousand others of seven feet or less." A 
great number of persons, including several ladies, were present at these 

As ma}'- be imagined, these experiments were not unattended with 
danger. Romas was on one occasion knocked down by an excessively 
heavy discharge, but, fortunately, escaped severe injury. Richmann, a 
member of the St. Petersburg Academy of Sciences, was not so fortu- 
nate, as one of his experiments cost him his life. He had erected an 
iron rod, which conducted the atmospheric electricity from the roof of 
the house to his study, so that he could measure its intensity every 
day. On the 6th of August, 1753, in the midst of a violent storm, 
and while standing at some distance from the rod, in order to avoid 
the large sparks, he incautiously approached too near the conductor. 
A globe of bluish fire struck him on the forehead, and killed him on 
the spot. 

For the last hundred years the study of electricity has been pursued 
both by experiments made in the laboratory and in the atmosphere — 
with what splendid results it is needless to relate. The electric tele- 
graph, which enables us to carry on a whispered conversation with our 
neighbors across the ocean, and the process which effects a faithful re- 



Fig. SO.— Kichmann, of St. Petersburg, struck by b'ghtning during au electrical experimeut. 

production of the chefs-d'oeuvre of statuary and engraving, are but two 
of the most important applications of the first. The experiments upon 
the electricity of the atmosphere, devoted to more complex and potent 
phenomena, have enabled us to acquire a more exact notion concerning 
the conditions of this electricity and its various manifestations. 

Electricity is a power the inner nature of which, like that of heat, 
light, and attraction, remains unknown to us. This power produces 


certain effects ; and it is the study of these effects which constitutes the 
science. To explain them, it is admitted — first, that electricity is a sub- 
tle fluid, capable of becoming amassed, condensed, and rarefied ; of dis- 
charging itself from one body into another ; of traversing immense dis- 
tances more rapidly even than light, which itself travels at the rate 
of about 185,000 miles per second ; secondly, that this fluid has two 
modes of existence — two modes of manifesting itself — which are dis- 
tinguished, the one from the other, by the terms positive and negative. 
These distinctions do not exist in nature, and are only perceptible to 
human sense by relative variations in intensity. Be this as it may, it 
has been ascertained that ojyiwsite electricities attract^ whereas like electrici- 
ties repel each other. The union of equal quantities of fluids of an op- 
posite denomination forms a neutral^ or natural, fluid, which, it is be- 
lieved, exists in inexhaustible quantities throughout all bodies. Under 
many influences, among which must be cited that of friction, the neu- 
tral fluid becomes decomposed into one or the other of these two ele- 
ments. The terrestrial globe and the atmosphere are two vast reser- 
voirs of electricity, between which there is a constant exchange by 
decomposition and reconstitution, which plays a complementary part 
to the action of heat and moisture in the life of plants and of animals. 

The general result of the researches into the conditions of electricity 
upon the surface of the globe and in the atmosphere is, that in a nor- 
mal condition the globe is charged with negative and the atmosphere 
with positive electricity. At the surface of the soil, where continual ex- 
changes are taking place, electricity is in a neutral state, as also in the 
lower stratum of air, which is in contact with the surface, upon the sea 
as well as upon land. Positive electricity increases in the atmosphere 
in proportion to height. 

The large amount of evaporation which takes place from the surface 
of the sea in the regions of the equator loads the clouds with positive 
electricity, and these, carried by the upper currents, travel toward the 
polar regions, and charge the atmosphere there with an accumulation 
of this electricity. Its influence causes in the soil of the polar regions 
an opposite condensation of negative electricity. The aurorse boreales 
are, in chief, caused by these two conflicting tensions ; it is a silent but 
visible reconstitution of the natural fluid by the two opposite tensions 
of the atmosphere and the soil. Thus the appearance of an aurora bo- 
realis is accompanied by electric currents, which circulate upon the soil 
at a distance sufficiently great to permit of the movements of the mag- 

428 ^^^ ATMOSPHERE. 

netic needle, indicating, at the Paris Observatory, for instance, an au- 
rora which may be visible in Sweden or Norway. 

Clouds are generally charged with positive electricity ; nevertheless, 
negative clouds are sometimes met with. It is not unusual to see upon 
the summit of a mountain clouds adhering to it, as if they were attract- 
ed thither, making a halt there, and then following the general move- 
ments of the wind. It often happens that in this case the clouds lose 
their positive electricity by their contact with the mountain, and as- 
sume the negative electricity of the latter, which, far from serving to at- 
tract, has, on the contrary, a tendency to repel and drive them away. 
On the other hand, a stratum of clouds situated between the ground, 
negative, and an upper stratum, positive, is almost neutral ; its positive 
electricity becomes accumulated upon its inside surface, and the first 
drops of rain cause it to disappear altogether. 

The electricity of the atmosphere is subject, like heat and atmospheric 
pressure, to a double annual and diurnal oscillation, and to accidental 
oscillations greater than those which are fixed and regular. The maxi- 
mum occurs from 6 to 7 A.M. in summer, and from 10 A.M. to noon in 
winter; the minimum is between 5 and 6 P.M. in summer, and between 
2 and 3 p.m. in winter. A second maximum is also noticeable at sun- 
set, followed by a diminution during the night until sunrise. This os- 
cillation is connected with that of the hygrometrical condition of the 
air. In the annual variation the maximum occurs in January and the 
minimum in July : it is due to the great atmospheric circulation. Win- 
ter is the period when the equatorial currents are most active in our 
hemisphere, and it is then that the aurorse boreales are most numerous. 

As the positive or negative conditions of electricity, as determined by 
apparatus constructed for measuring their intensities, are but a compari- 
son more or less between two different charges^ it follows that, when an 
electric cloud passes over our heads and dissolves itself into rain, the air 
may manifest negative electricity both before and after the rain, and 
even during its fall, according to the intensity of the cliarge contained 
in the cloud. M. Quetelet demonstrates this state of things in the fol- 
lowing manner : 

Let A, B, C, D, E, be five positions on the earth in a straight line, 
which we suppose to be neutral. The stratum of air above and parallel 
with the positions on the earth — A' B' C D' E' — is in a state of positive 
electricity in the absence of clouds, and to an equal extent throughout. 

The stratum A'' B'' Q" D" E", still higher and parallel, is also in a 


State of positive and more intense electricity. There comes suddenly a 
cloud at the three central positions B' C D', in a state of positive elec- 
tricity, greater than that of the circumambient air. It follows that, rel- 
atively to it, the air which is around will display negative electricity. 

To an observer situated at A, the electricity above the earth will be 
positive. As the cloud approaches, these indications will become gradu- 
ally less until they vanish altogether, and even become negative on the 
passage of the cloud. But the rain will bring back positive electricity. 
A corresponding variation will be manifest when the rain stops, and the 
cloud moves off. At D, the indications will be negative ; at E, they 
will again become positive. 

We saw in the conflicts of the great atmospheric currents in the trop- 
ical regions, where the node of the circuit accomplished from the equator 
to the poles takes place, that the evaporation of the seas, caused by 
solar heat in these foci of condensation, the variation of atmospheric 
pressure, etc., engender cyclones, hurricanes, and tempests, the whirling 
march of which reaches as far as our temperate latitudes. These vio- 
lent movements develop electricity in immense proportions, and it is 
rarely that these phenomena are not accompanied by thunder-storms, 
lightning, and thunder. The formation of the clouds upon sea and 
land, the fogs which occur in our regions, the course of the clouds along 
our valleys and mountains, all emit varying quantities of electricity. 
There is a storm when this electricity of the clouds, instead of effecting 
a mutual and tranquil interchange, collects at certain points, and, becom- 
ing condensed, saturates them, so to speak, and finally bursts, afterward 
uniting itself to the negative electricity which has been instantaneously 
amassed, either upon the ground or in other clouds. 

The great storms reach us from the Atlantic. They arise from the 
cyclones, and the clouds which convey them are generally more than 
3000 or 5000 feet high, traveling from S.W. to N.E., without being ap- 
parently affected by the undulations of the ground in France. The sec- 
ondary storms, which are formed in France itself, are conveyed by 
clouds of a less elevation than the above, and which sometimes just 
skim the ground, being, in fact, influenced by it, scarcely reaching over 
the mountains, following the valleys, amidst which they distribute in 
large quantities lightning and hailstorms. 

The formation of storms is preceded by a slow but steady decline in 
the reading of the barometer. The calm of the air and a stifling heat, 
due to the absence of evaporation from the surface of our bodies, are 


specially characteristic circumstances. The variations in the electric 
condition of the soil and the atmosphere, added to the above, have a 
great effect upon our organic system. A peculiar nervous feeling, with 
no visible cause for it, takes possession of many persons, in spite of all 
their efforts to shake it off. It is under these circumstances that one is 
especially enabled to see how intimate is the connection between man's 
physical and moral condition. 




When electricity is discharged from a cloud by which it is overload- 
ed, and is precipitated either into another cloud or to the ground with 
opposite electricity, electric light is produced — a rapid spark such as we 
•display on a small scale in our experiments in physics. This spark 
traverses in an instant the distance, whatever it may be, which separates 
the two electrized points. It has been ascertained that it does not last 
Tn oo o of a second. It is this electric spark which constitutes lightning; 
it is by it that lightning is made manifest during a storm. 

As a general rule, these flashes appear in the shape of a sudden dif- 
fused light which illuminates the clouds, the sky, and the earth, and is 
followed by a darkness which seems more intense than it was before by 
the force of contrast. Whether in this case the exchange of electricity 
between the clouds takes place simultaneously over a large surface which 
is lighted up and which dies away instantaneously, or whether there be 
a spark as in lightning concealed by the clouds, in either event one only 
sees — which is of the most frequent occurrence — a sudden diffused light, 
upon which are momentarily displayed the more or less marked con- 
tours of the clouds. 

These diffused lightnings are the most frequent. Hundreds of flashes 
are seen during a stormy day, or rather night, to one flash of linear 
lightning. The latter is, however, characteristic lightning. It is but 
a strong electric spark, a small ball of fire which darts from an over- 
charged cloud to the earth, or from one cloud to another, or which even 
rises from the earth to the clouds; the rapidity of its progress produces 
the effect of a narrow and luminous line. It is rare that it darts in a 
straight line, in spite of the axiom as to " the nearest road ;" whether 
because of the varied distribution of moisture in the air, which causes it 
to be a more or less better conductor, or because of the varying excess 
of electricity in different parts of the soil and of the clouds, the lightning 
is nearly always zigzag. The subtle fluid shows, by the way in which 
it traverses our dwelling-places, that it leaps suddenly from one point to 
another as if by caprice, but being evidently obedient to the laws of the 


distribution and conductibility of electricity. Generally speaking, linear 
lightning darts in obtuse-angled zigzags, or else is curled like a snake. 
Sometimes it splits into two or more branches. Nicholson and the 
Ahh6 Eichard observed forked flashes. Occasionally, though more 
rarely, it splits into three branches; Arago cites several instances of 
this, especially in the volcanic thunder-storms; Kaemtz noticed it 
once. At times, too, the flashes have four or five ramifications, or, it 
mav be, the branches which issue from the original flash become rami- 
fied into several small lateral branches. M. Liais observed and sketched 
flashes with five branches. 

The flashes are not always of a shining white hue, but have at times* 
a yellow, red, blue, and even a violet or purple tint; this color depends 
upon the quantity of electricity which traverses the air, upon the densi- 
ty of the latter, upon its moisture, and upon the substances suspended 
in it. The violet flashes generally indicate that the cloud from which 
they are emitted is at a great height, and the air which they travel 
through an air so rarefied as to call to mind that of the Geissler tubes. 

It is rarely that a correct idea as to the length of lightning-flashes is 
formed. While we produce with the greatest difficulty in our labora- 
tories an electric spark of a few inches. Nature shoots forth sparks as 
much as ten miles long. F. Petit measured at Toulouse some flashes 
which were ten and a half miles long— the extreme length with which 
I am acquainted. Arago found that a series of flashes which he meas- 
ured were seven or eight miles in length. 

In reply to the question as to the height of thunder-clouds, it is evi- 
dent that they are of difierent elevations. De I'lsle measured one on 
June 6, 1712, which was 26,250 feet above Paris; Chappe, on July 13, 
1761, remarked one that was situated 10,400 feet over Tobolsk ; and 
Kaemtz noticed another 10,200 feet above Halle. These observations 
give a decreasing series of elevations which gradually decline until 
they almost reach the ground. Haidinger measured thunder-cloi'ds 
which were only 230 feet above Gratz, on June 15, 1826, while upon 
another occasion he remarked some only ninety-two feet above the 
ground at Admont. This refers to a level country. In the mountains, 
Saussure observed some of these clouds over Mont Blanc; Bouguer 
and La Condamine, over the Pichincha, at 16,000 feet; Pamond, upon 
Mont Perdu, at 11,100 feet, and upon the Peak du Midi at 9630 feet, 
and indeed at all heights. They are generally from 2950 to 3280 feet 
high over the sea. 


Whether the flash takes place horizontally between two groups of 
clouds, or obliquely either between clouds of different strata or between 
the clouds and the ground, it is generally several iriiles long. It is this 
length which is the primary cause of the rolling of thunder. Thunder 
is, in reality, but the sound of the electric spark effecting an exchange 
of electricity, a neutralization, between two points more or less distant 
from each other. 

The noise of the thunder may be due to several different causes. 
The spark itself, as it traverses in an instant the atmospheric air, forces 
back the molecules upon its passage, and produces a momentary void 
into which the circumambient air at once rushes, and so on for a cer- 
tain distance. Pouillet met this rather natural explanation by the ob- 
jection that if the sound of thunder was produced in this way, the pas- 
sage of a cannon-ball would produce an analogous noise. The objec- 
tion is not well founded, for the cannon-ball is but a tortoise in compar- 
ison to the dart of the lightning. In the second place, the sound of 
thunder may be due to the fact that clouds become dilated under the 
influence of the electric tension which swells them in a certain measure, 
lengthens them, and stretches them with so much force at certain points 
that, if a spark causes the cloud to discharge, the outer air, being no 
longer retained by the expansive force of the electric fluid in equilibri- 
um with it, rushes from all directions toward the clouds. To this may 
be attributed the cause of thunder, and of the fall of rain which follows. 
The electric conditions of the various clouds which compose a storm 
being dependent the one upon the other, the discharge of one must lead 
to that of several others more or less distant. In the one case as in the 
other, the sound is always caused by the expansion of the air at the 
spot where the more or less partial void has just been made, as hap- 
pens with fire-arms, the bursting of a bladder, etc. When situated at 
the point where the lightning terminates — where the thunder-bolt falls, 
according to the vulgar expression — this noise is never very long, and 
is exactly like the report of a cannon, a fowling-piece, or a pistol, ac- 
cording to the intensity. But one of the special characteristics of thun- 
der consists in the rolling, as its name imitates it in every language— 
thunder, tonneire, tonitruum, hronte, donner. 

It is frequently asked to what this rolling, often very prolonged, can 
be due. There are several causes for it. The first is due to the length 
of the flash, and to the difference in the speed of sound and of light. 
Let us imagine, for instance, a horizontal flash 35,000 feet long and 



3000 feet high. An observer, placed beneath one extremity of the 
flash, will see this flash in its full length for an instant; the sound will 
be formed at the same moment along the whole line of the flash ; but 
the sound-waves will reach his ears at different times. That which 
starts from the nearest point will arrive in three seconds, as sound trav- 
els at the rate of about 1100 feet per second. That which was formed 
at the same instant at a point 6000 feet distant takes twice the time 
to arrive. That which pi'oceeds from a point at 13,200 feet will take 
twelve seconds. The sound formed at a distance of 35,000 feet would 
take thirtj-three seconds to travel ; thus the rolling will continue half 
a minute, gradually becoming fainter, until it dies away altogether. 

If, as most frequently happens, the observer is not situated exactly at 
one of the extremities of the flash, but at a certain point in its line of 
passage, he hears first the report, which gradually grows louder, and 
then diminishes. In this case, the sound starting from a point situated 
over his head, and at a height of 1000 yards, reaches him in three sec- 
onds; but the sounds formed on either side at equal distances arrive 
at the same time during several seconds, and sound ceases in less than 
thirty-two seconds. 

To this cause of the prolonged rolling must be added the numerous 
discharges which often take place very rapidly among thunder-clouds 
— the zigzags and ramifications of the lightning, caused by the hygro- 
metrical diversity of the various strata of air — the echoes repeated by 
mountains, the soil, waters, and the clouds themselves, to which must 
further be added the interference produced by the encounter of differ- 
ent systems of sound-waves. 

The duration of the rolling of thunder varies very much, as every 
one may have remarked. The greatest length recorded for a single 
flash is forty-five seconds, by De I'Isle, at Paris, on June 17, 1712. 
Upon the same day he remarked another, which lasted forty-one sec- 
onds; and on July 8, in the same year, one of thirty-nine seconds. The 
intervals, included between the commencement of the thunder and the 
different phases of intensity in its rolling, were as follows upon this 
last occasion (July 8) : 

At seconds, flash ; 

At 11 seconds, slight thunder; 

At 12 seconds, it bursts ; 

At 32 seconds, the explosions cease ; 

At 50 seconds, the sound dies gently away. 


The intensity of thunder varies to an enormous extent. In certain 
cases it has been compared to the report of a hundred pieces of artillery 
discharged at the same time. In other instances the report is no louder 
than that of a pistol, followed by a rolling sound more or less dull. At 
times the explosions remind one of the tearing of a piece of silk, at oth- 
ers of the noise made by a cart loaded with bars of iron sent loose 
down a steep paved street. 

The longest interval ever remarked between the flash and the report 
was seventy-two seconds. This was at Paris, and the same interval 
was also noticed to elapse by the astronomer De I'lsle on April 30, 
1712. In these two cases the cloud must have been six leagues off. 
Next to these exceptional cases, the longest interval was forty-nine 
seconds, which represents ten miles' distance. Direct researches have 
shown that a storm is never heard at a greater distance than thirteen 
miles, rarely at more than seven to ten ; the flashes are visible, but the 
sound does not travel so far. The fact is the more curious as cannon 
are heard at a much greater distance, as much as twenty-five miles ; and 
when very large, they may be heard at double that distance. 

Continued cannonading, as during a siege or a pitched battle, has 
been heard at a distance of thirty leagues. During the winter of 1870, 
the Krupp guns, exhibited in Paris in 1867, were heard at Dieppe, a 
distance of eighty-four miles, during the bombardment of Paris. The 
cannonade of March 30, 1814, was heard at Casson, a village between 
Lisieux and Caen, at a distance of forty-four leagues from Paris. Ara- 
go relates that the firing at Waterloo was audible at Creil, 120 miles 
distant. Thus the thunder manufactured by man reaches much farther 
than the thunder produced by nature. If thunder is not audible at 
more than six leagues, it follows that, if thunder is heard with the sky 
clear, the report must be produced by clouds below the visible horizon, 
as we can not see beyond six leagues. A person of five feet five inch- 
es in height is able to see, when the horizon is clear, an object placed 
upon the ground at a distance of 13,000 feet. If the object in question 
is eighty feet high in the air, it may be seen at five and a half leagues. 
If it is 1600 feet high, as in the case of an isolated mountain, it will be 
visible at a distance of fifty miles. If the object be 3300 feet high, as 
cumidus clouds are, as a rule, in our climates, it can be seen at a dis- 
tance of seventy miles. 

For a thunder-clap which takes place when the sky is clear, to be 
produced by a cloud, we must consequently suppose the cloud to be 


less than 100 feet above the ground — a state of things never witnessed. 
Thus electricity may be emitted from certain regions of the air, from 
invisible clouds, and may produce flashes and thunder-claps during fine 
clear weather. Observation has proved this to be a fact, but one of 
very rare occurrence. 

To these statements bearing upon the general action of thunder and 
lio^htning, I may add that, notwithstanding the extreme rapidity of the 
flash, it has been found possible to measure its duration, which does not 
exceed ttt auo of a second. To effect this, a round piece of card-board, 
divided from the centre to the circumference into black and white sec- 
tions, is made use of This circle is made to turn like a wheel, with a 
speed equal to that of the wind. It is well known that luminous im- 
pressions remain for one-tenth of a second upon the retina. Thus, if a 
hot coal is turned round, and if the revolution is made in one-tenth of a 
second, and as each successive position of the coal remains impressed 
for the same length of time upon the retina, a continuous circle becomes 
visible. If the circular piece of card-board, with its black and white 
stripes, is made to revolve, the sectors cease to be visible, and we can 
only see a grayish circle, if each stripe passes before our eyes in less 
than the tenth of a second. But it is possible to make the card-board 
revolve more than a hundred times in a second. This being the case, 
if the card-board circle is exposed to a continuous light, we shall be un- 
able to distinguish the lines, inasmuch as they come before our eye 
much more rapidly than the impression which they produce remains. 
But if the circle is made to revolve in a dark place, and an instantane- 
ous flash of light suddenly falls upon it, and as suddenly disappears, the 
impression produced upon our eye by each of the sectors will last less 
than one-tenth of a second, it will be almost instantaneous, and the circle 
loill seem to us to be motionless. By giving this apparatus a fixed rate of 
rotation, it has been ascertained that a flash lasts but ^^ J^^ of a second. 

Light, traveling a distance of 185,000 miles i'n a second, takes but an 
instant, too short to be reckoned, to come from the spot, never more 
than a few miles off, at which the flashes are produced. Thus we see 
the flash at the very moment at which it occurs. But sound travels, as 
we have seen above, less rapidly — at the rate of 1100 feet a second. It 
follows that the thunder-clap, which takes place at the same time as the 
flash, will only be audible to us ten seconds afterward if we are 11,000 
feet away from the storm ; and anyone can, therefore, calculate how far 
off the storm is by the interval between the flash and the thunder. 

Fi''. 81.— Harvesters killed by lightning 


^ second interval correspomls to 550 feet. 

1 " " '' 1,100 " 

2 " " " 2,200 " 

3 " " " 3,300 " 

4 " " " 4,400 " 

5 " " " 5,500 " 

6 " " " 6,G00 " 

7 " " " 7,700 " 

8 " " " 8,800 " 

9 " " " 9,900 " 

10 " " " 11,000 " 

11 " " " 12,100 " 

12 " " " 13,200 " 

There are about twelve beatings of the pulse to a league. AVhen 
the flash extends over a length of several miles, the spot struck by the 
thunder may be very distant, although the report is heard immediately 
after the flash, because it is the sound which starts from the nearest ex- 
tremity of the flash which is heard first. For instance, in a' storm on 
the 27th of June, 1866, M. Hirn remarked that the report followed im- 
mediately upon the flash, although this same flash had struck down 
two persons beneath a tree three miles distant. 

Many are the marvelous freaks and jests played by electricity, some- 
times ending in tragedy. Among the most remarkable is that of strik- 
ing a person dead, and leaving him in the exact position occupied at 
the moment the shock was given, just as if he were still alive, and yet 
so thoroughly consumed as to be nothing but a mass of cinders. Thus 
we are told that at Vic-sur- Aisne, France, in 1838, three soldiers 
sought refuge from a violent thunder-storm under a linden -tree. 
Some peasants, seeing them stand motionless long after the storm had 
passed, and receiving no response to a pleasant salutation, touched 
them on the shoulder. The bodies instantly crumbled to fine ashes! 
Yet the moment before there was no evidence that the lio-htnino; had 
touched them. Their clothing was not torn, and their fiices wore a 
natural appearance. The following remarkable circumstance was wit- 
nessed by Pastor Butler: On the 27th of July, 1691, ten harvesters 
took refuge under a hedge on the approach of a thunder-storm. The 
lightning struck and killed four of them, who remained as if suddenly 
petrified. One of them was just putting a bit of tobacco in his mouth, 
another was fondling a little dog on his knee with one hand and feed- 
ing him with the other. M. Cardan relates that eight harvesters, tak- 



ing their noonday repast under a maple-tree during a thunder-storm, 
were killed by one stroke of lightning. When approached by their 
companions, after the storm had cleared away, they seemed to be still 
at their repast. One was raising a glass to drink, another was in the 
act of taking a bit of bread, a third was reaching out his hand to a 
plate. There they sat as if petrified, in the exact position in whicl 
death surprised them. 

Fig. S2.— Curious freak of lightning. 

On the 10th of September, 1845, during a violent thunder-storm, a 
house in the village of Salagnac, France, was struck by lightning. A 
large ball of fire descended the chimney, and rolled across the floor of a 
room in which sat a child and three women. No one was hurt. It 
then rolled out through the centre of the kitchen, passing close to the 
feet of a young peasant, and disappeared through a crevice in the wall. 
Its erratic course ended in the pig-sty, the harmless occupant of which 
it despitefully slew, without setting on fire the straw on which the crea- 
ture lay. 




The Saint Elmo fires are a slow manifestation of electricity, a quiet 
and steady outflow (like that of the hydrogen in a gas-burner), which 
radiates gently over the topmost points of lightning conductors, of 
buildings and vessels, during thunder weather, when the terrestrial elec- 
tric tension is strongly attracted by that of the clouds. 

The Saint Elmo fires are generally seen as a light resting on the 
masts of ships. The following are some of the most recent observa- 
tions made : 

On December 23, 1869, in latitude 46° 53' north, and longitude 9° 
55' west, the barometer reading 29"61 inches, thermometer 49° 1', the 
log of the packet Imperatrice- Eugenie records the occurrence of very 
violent squalls. Sharp and numerous flashes of lightning were visible 
in all parts of the horizon, without being followed by a single clap of 
thunder. During the night these squalls were accompanied by heavy 
hailstorms, and, when they passed over the vessel, they presented the 
phenomenon known under the name of the Saint Elmo fire. 

Luminous tufts, blue in color and about a foot and a half high, ap- 
peared above the tips of the conductors upon each mast. The masts 
and the rigging looked phosphorescent, and the tips of the waves also 
seemed decked with tufts, but less showy than those that appeared 
above the masts. These glimmerings were visible whenever the squall 
reached the vessel. Very brilliant when the wind was blowing with its 
full force, they became less bright as it fell, and disappeared when it 
dropped altogether. Only those parts of the masts and the rigging 
which were exposed to the direct action of the squall presented this 
luminous appearance. They looked as if they had been rubbed with 
phosphorus. The phenomenon did not take place upon the parts which 
were at all sheltered from the wind, nor did it come down lower than 
the top-yards, about ninety feet above the level of the sea. The phe- 
nomenon repeated itself several times during the night, but only when 
the squalls were accompanied by hail. The Saint Elmo fires are also 
seen over steeples. The following is one of the most recent instances: 



On March 2, 1869, these flames appeared over the church at Sainte- 
Catherine-de-Fierbois, in the canton of Sainte-Maure and the arrondisse- 
ment of Cbinon ; no thunder was audible during the storm, and the 
steeple disarmed the thunder-clouds. A correspondent of the French 
Scientific Association wrote as follows : " Toward the end of the tem- 
pest, when the wind had somewhat abated and the rain wms not so 
heavy, several persons remarked a crown of fire around the cross that 
surmounted the steeple of the' church, about 130 feet high. One of the 
eye-witnesses saw it for at least five minutes (he did not perceive it 

^ ^^^__ begin) ; the light was so bright 

::^i that the steeple and cross were 

"^ as plain to the eye as in full 

4'- daylight; the light finally died 

'i away like that of a burned- 

i out candle, without the least 

change of position." 

Luminous tufts of electrici- 
ty have often been seen above 
the spire of Notre-Dame dur- 
ing certain violent thunder- 
storms of a summer evening. 

The Saint Elmo fires are oc- 
casionally seen playing over 
man himself, over his clothes, 
or any object that he has in 
his hand. 

Julius Caesar relates how in 
the month of February, about 
the second watch of the night, 
a thick cloud suddenly arose, 

Fig. 83.-Saiut Elmo fire over the spire of Notre-Dame, folloWcd by a shoWCr of StOUeS ; 

'"'^' and that during the same night 

the pike-heads of the fifth legion seemed to be on fire. 

According to Procopius, a similar phenomenon was seen over the 
pikes and lances of Belisarius's army in the war with the Vandals. 

Livy states that the pikes of some soldiers in Sicil}^, and a whip 
which a horseman in Sardinia had in his hand, seemed as if on fire. 
Even the coats of mail were luminous and bright with numerous flames 
of fire. 


When in 1769, in the- midst of a violent storm, bright tufts ap- 
peared over the cross upon the steeple at Hohen-Gebrachim, two per- 
sons, who had come to put out the conflagration, as thej thought, 
were at once surprised and terrified to see their heads covered with 
fire and light. 

On May 8, 1831, after sunset, the whole atmosphere was on fire, pre- 
saging a violent storm. At the extremity of the flag-staff at Algiers 
there appeared a white light in the shape of a brush which lasted for 
half an hour. Some artillery and engineer officers were walking upon 
the terrace -of Fort Bab-Azoun, and each noticed, to his surprise, that 
the heads of his companions w^ere tipped by small luminous tufts. 
When they raised their hands, brushes of light formed at the tips of 
their fingers. 

In some instances the Saint Elmo fires have been noticed in the shape 
of flames ; at other times a man's whole body has been seen radiant 
with light. Peytier and Hossard were frequently enveloped, in the 
Pyrenees, in centres of storms, which seemed so formidable as seen from 
the plains below, that the spectators believed they must have perished 
in them. On several occasions their hair and the tassels of their caps 
stood upright and emitted a bright light, accompanied by a loud hiss- 
ing noise. 

Letestu, in 1786, remained for three hours of the night in his balloon 
during a storm; he heard a deafening noise; the car was filled with 
snow and hail, and the gilding upon his flag emitted scintillations. 

The discharge of electricity from the soil into the atmosphere is some- 
times accompanied by remarkable phenomena — by a kind of electric 
hum upon the summits of mountains. 

These various phenomena are due solely to disengagements of elec- 
tricity. We must not confound with the Saint Elmo fires gleams of 
light which resemble them very much, viz., the ignes-fatui. These lat- 
ter are not caused by electricity. 

The ignes-fatui, or will-o'-the-wisp, is a wandering and shadowy fire, 
produced by the emanations of phosphureted hydrogen gas, which rises 
out of places where vegetable and animal substances are in process of 
decomposition, such as cemeteries, manure-heaps, or marshes, and which 
become spontaneously inflamed when combined with the oxygen of 
the air. 

These vacillating lights have appealed to the superstitious feelings of 
the people. The frightened imagination has often looked upon them as 


wandering spirits, and they have often terrified those who have seen 
them gliding between the graves of a church-yard during the silence 
of night. 

They are sometimes emitted suddenly when an old burying-vault is 
opened ; and as in former days lighted lamps were placed in the graves, 
the credulous believed they were inextinguishable. 






We now come to the most curious and the grandest of the various 
manifestations of electricity in the atmosphere. As we have seen, the 
globe is one vast reservoir for this subtle fluid, which exists in all the 
worlds appertaining to our system, and of which the radiating focus is 
in the sun itself. Like attraction, light, and heat, electricity is a general 
power in nature. Its palpitations sustain the life of the universe, and 
even upon our planet currents of it are in constant circulation from the 
equator to the poles, and from the poles to the equator. The delicate 
magnetic needle and the sea-compass indicate this perpetual circulation 
as moving northward. The magnetic needle oscillates and becomes 
agitated when these disturbances become violent and there are great 
changes in its position. The lightning which falls upon a ship often 
exercises an ineffaceable influence upon the compass; and while the 
pilot assumes that the needle is still pointing north, he runs the risk of 
being driven on to the rocks of some unknown shore. If a bright au- 
rora borealis is shining over Stockholm or Reikjavik, the compass in the 
Paris Observatory, hundreds of leagues off, is affected by it, and seems 
as if it were asking the editor of the Bulletin International to see what 
was the matter. 

The aurora borealis is one of the grand results of atmospheric elec- 
tricity. Instead of a furious and violent storm limited to a few leagues, 
it is a gentle and gradual recomposition of the negative fluid of the 
earth with the positive fluid of the atmosphere, taking place in the 
aerial heights, in the upper hydrogenous atmosphere. This disengage- 
ment of electricity in a vast sheet is only visible at night, and assumes 
every imaginable kind of shape, according to the way in which it takes 
place, and to the perspective caused by the distance of the observer. At 
one time the eye may scarcely have time to catch its rapid undulations, 
alternately rose-colored and white in hue, as they dart across the sky. 
Now it takes the shape of a cloth of gold and purple, which seems to 
fall from the celestial heights ; now it is a fiery dew, accompanied by a 
strange, rustling sound, or it may appear in the form of sheaves of 


flame, darting from the north to the various points of the compass. It 
is principally in the neighborhood of the polar circles, where thunder- 
storms are rare, that these manifestations of terrestrial electricity are seen 
to the fullest advantage. Michelet, who describes so graphically the 
great phenomena of nature, speaks of the aurora borealis in this way : 

"The pole seems a kingdom of death. But, in reality, general life is 
triumphant there. The two spirits of the globe (magnetic and electric) 
make their nightly rejoicing in this desert." 

"The aerial currents, and the currents of the sea, are their vehicles. 
The two torrents of heated waters which, from Java and Cuba, travel 
northward, where they cool and freeze, and then return refreshed to the 
centre whence they started, both assist in keeping up the magnetic and 
electric correspondence between the equator and the pole. Their storms 
are dependent upon each other. In summer, when the melted ice from 
the poles and the northern currents make their cooling influence felt, 
the magnetic element seems to extend in the direction of the centi-nl 
electricity ; hence the violent storms, especially those near to this 

Spitzbergen is a very favorable region for witnessing an aurora bo- 
realis. In a voyage undertaken in 1839, M. Ch. Martins observed and 
analyzed a large number, which he describes thus (see "Le Tour du 
Monde,"1865, vol. ii., p. 10): 

"At times they are simple diffused gleams or luminous patches; at 
others quivering rays of pure white which run across the sky, starting 
from the horizon as if an invisible pencil were being drawn over the 
celestial vault; at times it stops in its course, the incomplete rays do 
not reach the zenith, but the aurora continues at some other point; a 
bouquet of rays darts forth, spreads out into a fan, then becomes pale 
and dies out. At other times long golden draperies float above the 
head of the spectator, and take a thousand folds and undulations, as if 
agitated by the wind. They appear to be but at a slight elevation in 
the atmosphere, and it seems strange that the rustling of the folds, as 
they double back on to each other, is not audible. G-enerally a lumi- 
nous bow is seen in the north ; a black segment separates them from 
the horizon, its dark color forming a contrast with the pure white or 
bright red of the bow which darts forth the rays, extends, becomes di- 
vided, and soon pi'esents the appearance of a luminous fan, which fills 
the northern sky, mounts nearly to the zenith, where the rays, uniting, 
form a crown, which, in its turn, darts forth luminous jets in all direc- 




tions. The sky then looks like a cupola of fire: the blue, the green, 
the yellow, the red, and the white vibrate in the palpitating rays of 
the aurora. But this brilliant spectacle lasts only a few minutes; the 
crown first ceases to emit luminous jets, and then gradually dies out; 
a diffuse light fills the sky; here and there a few luminous patches, re- 
sembling light clouds, open and close with an incredible rapidity, like 
a heart that is beating fast. They soon get pale in their turn ; "every 

Fig. 84.— An aurora borealis over the Polar Sea. 

thing fades away and becomes confused ; the aurora seems to be in its 
death-throes; the stars, which its light had obscured, shine with a re- 
newed brightness; and the long polar night, sombre and profound, 
again assumes its sway over the icy solitudes of earth and ocean." In 
presence of such phenomena, the poet and the artist are compelled to 
confess their littleness — the savant alone does not despair. After hav- 
ing admired the spectacle, he studies, analyzes, compares, and discusses 


it ; he succeeds in proving that these auroree are due to electric radia- 
tions from the poles of the earth, which is a colossal magnet, the north- 
ern pole of which is situated to the north of North America, not far 
from the pole of our hemisphere, while its southern pole is in the sea to 
the south of Australia, near Victoria." 

A few instances will suffice to prove the electro-magnetic nature of 
the atirora borealis. At Spitzbergen, a magnetic needle suspended 
horizontally by an untwisted piece of silk-thread is turned toward the 
west. As soon as the aurora begins, the person observing this needle 
remarks that, instead of being sensibly motionless, it is agitated, pass- 
ing to and fro from right to left, and from left to right. In proportion 
as the aurora becomes more brilliant, the agitation of the needle in- 
creases, and the observer is able to judge of the intensity of the aurora 
by the motions of the needle without leaving his study. Lastly, when 
the corona is formed in the sky, its centre will be found exactly in the 
direction to which a magnetic needle, hanging freely, points. The 
aurorce boreales are therefore intimately connected with the magnetic 
phenomena of the terrestrial globe. 

What a strange world is that of the poles! Nearly every night 
there is a more or less brilliant display of these auroral lights; from 
the middle of January, there is an hour's twilight at noon ; the aurora, 
announcing the return of the sun, becomes grander as it mounts toward 
the zenith. Lastly, on the 16th of February, a segment of the solar 
disk, resembling a luminous point, shines brightly for an instant, and 
as rapidly disappears; but every day at no6n the segment increases, 
until the whole orb rises above the sea : it is the end of the long win- 
ter night; after that, day and night follow each other for sixty-five 
days, until the 21st of April, when begins day-time, lasting four months, 
during which period the sun revolves above the horizon, gradually be- 
coming lower, and finally disappearing. 

In North America, to the east of Beh ring's Straits, there is a large 
tract of territory little known to Frenchmen — Alaska — which is trav- 
ersed by the Arctic Circle. It formed part of Kussian America a few 
years ago, and was 45,000 square leagues in extent. It was purchased 
by the United States in October, 1867. In a curious account of a voy- 
age which Frederick Whymper made there, in 1865 (see "Le Tour du 
Monde," 1869, vol. ii., p. 247), there is recorded the observation of that 
very rare phenomenon, viz., an aurora borealis in the shape of a ribbon, 
extending in undulatincr folds in the heiarhts of the air. 








Fig. S5.— Aurora borealis observed at Bossekop (Spitzbergen) January 6, 1839. 

To use the traveler's own words, " It was on the 27th of December, 
as we were about to retire for the night, that we were informed that 
an aurora borealis was visible in the west. We at once climbed the 
roof of the highest building in the fort, in order to contemplate this 
splendid phenomenon. It was not in the form of an arch, as often is 
the case, but the light was serpentine-shaped and undulating, the form 
and the color varying every instant, being at one moment of a pale and 
soft tint like moonbeams, while at the next long bands of blue, rose, 
and violet stood out upon the silvery background. The scintillations 
extended from the lower extremity upward, and their brightness be- 
came fused with that of the stars, the brilliancy of which was visible 
through the spiral vapor." 

29 . 


from New York to Siberia, and from both hemispheres, at the Cape 
of Good Hope, in Australia, Salvador, Philadelphia, and Edinburgh! 
This was the first time that the eye verified what theory had advanced, 
viz., that aurorae boreales and the southern aurorae occur at the same 
time in the two hemispheres under the influence of the same current. 
The extremities of the globe are brought into intimate relation with 
each other by the fluid which circulates incessantly in the air and upon 
the soil. At certain solemn moments, magnetism augments in intensi- 
ty, and seems to reanimate the life of our planet. 

The production of aurorse boreales is, in Humboldt's opinion, one of 
the most striking proofs of the faculty which our planet possesses of 
emitting light. He says, " It results from the phenomenon of aurorae, 
that the earth is endowed with the property of emitting a light distinct 
from that of the sun. The intensity of this light is rather greater than 
that of the moon in its first quarter. It is at times (January 7, 1831) 
strong enough to admit of one's reading printed characters without 
difficulty. This light of the earth, the emission of which toward the 
poles is almost continuous, reminds us of the light of Venus, the part of 
which, not lighted by the sun, often glimmers with a dim phosphores- 
cent light. Other planets may also possess a light evolved out of their 
own substance. There are other instances in our atmosphere of this pro- 
duction of terrestrial light, such as the celebrated fogs of 1783 and 1831, 
which emitted a perceptible light during the night. Such, too, are those 
large clouds which are brilliant with a steady and motionless light; and 
such, too, as Arago has truly remarked, is that diffuse light which guides 
our steps during the nights of spring and autumn, when the clouds in- 
tercept all celestial light, and snow does not cover the ground." 

I may further remark that aurorae boreales are more or less period- 
ical. They were very numerous in Belgium and Western Europe dur- 
ing the last half of the eighteenth century. They were very rare in 
the seventeenth, and very frequent in the sixteenth century. This sec- 
ular periodicity seems to be of about a century and a half. There is 
a monthly variation, more accurately ascertained. They are most fre- 
quent about the time of the equinoxes, and seem to be seven times 
more numerous in March and October than in June. 

Such are the last and the grandest of the phenomena which we have 
to contemplate in this gallery of the works of the Atmosphere. 




Published by HARPER & BROTHERS, New York. 

C^~ For a full List of Books suitable for Libraries, see Hakpkr & Bkotukrb' Teadh- 
LiST and Catalogue, which may be had gratuitously on application to tkt 
Publishers personally, or by letter enclosing Six Cents. 

(ES"Harpeu & BuoTHEKB will Send any of the following works by mail, postage pre- 
paid, to any part of the United States, on receipt of the price. 

FLAMMARION'S ATMOSPHERE. The Atmosphere. Translated from the French 
of Camii.le Plammabion. Edited by James Glaisuek, F.R.S., Superintendent 
of the Magnetical and Meteorological Department of the Royal Observatory at 
Greenwich. With 10 Chromo-Lithographs and 86 Woodcuts. 8vo, Cloth, $6 00. 

HUDSON'S HISTORY OF JOURNALISM. Journalism in the United States, from 
1G90 to 1872. By Fbkdkkiok Hudson. Crown 8vo, Cloth, $5 00. 

PIKE'S SUB-TROPICAL RAMBLES. Sub-Tropical Rambles in the Land of the 
Aphanapteryx. By Nicholas Pike, U. S. Consul, Port Louis, Mauritius. Pro- 
fusely Illustrated from the Author's own Sketches; contaiDiug also Maps and 
Valuable Meteorological Charts. 8vo, Cloth, $3 50. 

TYERMAN'S OXFORD METHODISTS. The Oxford Methodists : Memoirs of the 
Rev. Messrs. Clayton, lugham,Ganibold,Hervey, and Broughton, with Biograph- 
ical Notices of others. By the Rev. L. Tyeum an. Author of " Life and Times of 
the Rev. John Wesley," &c. Crown Svo, Cloth, $2 50. 

TRISTRAM'S THE LAND OF MOAB. The Result of Travels and Discoveries on 
the Bast Side of the Dead Sea and the Jordan. By H. B. Tristram, M. A., LL.D., 
F.R.S., Master of the Greatham Hospital, and Honorary Canon of Durham. With 
New Map and Illustrations. Crown Svo, Cloth, $2 50. 

SANTO DOMINGO, Past and Present: with a Glance at Hayti. By Samuel Hazard. 
Maps and Illustrations. Crown Svo, Cloth, $3 50. 

LIFE OF ALFRED COOKMAN. The Life of the Rev. Alfred Cookman ; with 
some Account of his Father, the Rev. George Grimston Cookman. By Henry 
B. RiDGAWAY, D.D. With an Introduction by Bishop Foster, LL.D. Portrait 
on Steel. 12mo, Cloth, $2 00. 

HERVEY'S christian RHETORIC. A System of Christian Rhetoric, for the 
Use of Preachers and Other Speakers. By George Winpred Hervey, M.A., 
Author of " Rhetoric of Conversation," &c. Svo, Cloth, S3 50. 

CASTELAR'S old ROME AND NEW ITALY. Old Rome and New Italy. By 
Emilio Castelar. Translated by Mrs. Arthur Arnold. 12mo, Cloth, $1 75. 

THE TREATY OF WASHINGTON: Its Negotiation, Execution, and the Discus- 
sions Relating Thereto. By Caleb Cushins. Crown Svo, Cloth, $2 00. 

PRIME'S I GO A-FISHING. I Go a-Fishing. ByW.C.PEiMB. Crown Svo, Cloth, 
$2 50. 

HALLOCK'S FISHING TOURIST. The Fishing Tourist: Angler's Guide and Ref- 
erence Book. By Charles Hallook, Secretary of the " Blooming-Grove Park 
Association." Illustrations. Crown Svo, Cloth, $2 00. 

SCOTT'S AMERICAN FISHING. Fishing in American Waters. By Genio C. 
SooTT. With 170 Illustrations. Crown Svo, Cloth, $3 50. 

Si'KNCER F. Baird, of the Smithsonian Institution, with the Assistance of Emi- 
nent Men of Science. r2mo, over 700 pp.. Cloth, $2 00. (Uniform with the An- 
nual Re/:ord of Science and Industry for 1871. 12mo, Cloth, $2 00.) 

By John W. Forney. 12mo, Cloth, $2 00. 

Hundred Recipes for Economical and Healthful Cooking; also, many Directions 
for securing Health and Happiness. Approved by Physicians of all Classes. 
Illustrations. 12mo, Cloth, $1 50. 

FARM BALLADS. By Will Caeleton. Handsomelv Illustrated. Square 8yo, 
Ornamental Cloth, $2 00; Gilt Edges, $2 50. 

2 Harper &> Brothers' Valuable and Interesting Works. 

POETS OF THE NINETEENTH CENTURY. The Poets of the Nineteenth Cen- 
tury. Selected and Edited by the Kev. Kobert Aris Wili.mott. With English 
and Ameiicau Additions, airanged by Evkrt A. Uuyckinok, Editor of "Cyclo- 
piEdia of American Literature." Comprising Selections from the Greatest Au- 
thors of the Age. Superbly Illustrated with 141 Engravings from Designs by 
the most Eminent Artists. In elegant small 4to form, printed on Superfine 
Tinted Paper, richly bound in extra Cloth, Beveled, Gilt Edges, $5 00 ; Half Calf, 
$5 50; Full Turkey Morocco, $9 00. 

With an Introduction by the Rev. P. Soiiaff, D.D. CIS pp.. Crown Svo, Cloth, 
$3 00. 

This work embraces in one volume : 
By J. B. LiGHTKooT, I).D., Canon of St. Paul's, and Huleean Professor of 
Divinity, Cambridge. Second Edition, Revised. 196 pp. 
Connection with some Recent Proi)osals for its Revision. By Richard 
Chenevix Trencu, D.D., Archbishop of Dublin. 194 pp. 
OP THE NEW TESTAMENT. By J. C. Ei.lioott, D.D., Bishop of Glou- 
cester and Bristol. 178 pp. 

NORDHOFF'S CALIFORNIA. California: For Health, Pleasure, and Residence 
A Book for Travelers and Settlers. Illustrated. Svo, Paper, $2 00; Cloth, $2 50. 

MOTLEY'S DUTCH REPUBLIC. The Rise of the Dutch Republic. By John Lo- 
THBOP Motley, LL.D., D.C.L. With a Portrait of William of Orange. 3 vols., 
Svo, Cloth, $10 50. 

MOTLEY'S UNITED NETHERLAND'S. History of the United Netherlands : frora 
the Death of William the Silent to the Twelve Years' Truce— 1609. With a full 
View of the English-Dutch Struggle against Spain, and of the Origin and De- 
structitm of the Spanish Armada. By John Lotiirop Moti.ev, LL.D., D.C.L. 
Portraits. 4 vols., Svo, Cloth, $14 00. 

NAPOLEON'S LIFE OF C^SAR The History of Julius Csesar. By His late Im- 
perial Majesty Napolkon III. Two Volumes ready. Library Edition, Svo, Cloth, 
$3 50 per vol. 
Maps to Vols. I. and II. sold separately. Price $1 50 each, net. 

HAYDN'S DICTIONARY OF DATES, relating to all Ages and Nations. For Uni- 
versal Reference, Edited by Benjamin Vincent, Assistant Secretary and Keeper 
of the Library of the Roval "Institution of Great Britain ; and Revised for the Use 
of American Readers. Svo, Cloth, $5 00 ; Sheep, $C 00. 

MACGREGOR'S ROB ROY ON THE JORDAN. The Rob Boy on the Jordan, 
Nile, Red Sea, and Gennesareth, &c. A Canoe Cruise in Palestine and Egypt, 
and the Waters of Damascus. By J. Maogrkgob, M.A. With Maps and Illus- 
trations. Crown Svo, Cloth, $2 50. 

WALLACE'S MALAY ARCHIPELAGO. The Malay Archipelago: the Land of the 
Orang-Utan and the Bird of Paradise. A Narrative of Travel, ]S.')4-1862. With 
Studies of Man and Nature. By Aefked Rursel Wai.t.aoe. With Ten Maps 
and Fifty-one Elegant lUustration.s. Crown Svo, Cloth, $2 50. 

WHYMPER'S ALASKA. Travel and Adventure in the Territory of Alaska, for- 
merly Russian America— now Ceded to the United States— and in various other 
parts of the North Pacific. By Frehebiok Wuymper. With Map and Illustra- 
tions. Crown Svo, Cloth, $2 50. 

ORTON'S ANDES AND THE AMAZON. The Andes and the Amazon ; or. Across 
the Continent of South America. By James Orton, M.A., Professor of Natural 
History in Vassar College, Poughkeepsie, N. Y., and Corresponding Member of 
the Academy of Natural Sciences, Philadelphia. With a New Map of Equatorial 
America and numerous Illustrations. Crown Svo, Cloth, $2 00. 

WINCHELL'S SKETCHES OF CREATION. Sketches of Creation : a Popular 
View of some of the Grand Conclusions of the Sciences in reference to the His- 
tory of Matter and of Life. Together with a Statement of the Intimations of 
Science respecting the Primordial C(mdition and the Ultimate Destiny of the 
Earth and the Solar System. By Alexander Winohell, LL.D., Professor of 
Geology, Zoology, and Botany in the University of Michigan, and Director of the 
State Geological" Survey. With Illustrations. 12mo, Cloth, $2 00. 

WHITE'S MASSACRE OF ST. BARTHOLOMEW. The Massacre of St. Bartholo- 
mew : Preceded by a History of the Religious Wars in the Reign of Charles IX. 
By Henry White, M.A. With Illustrations. Svo, Cloth, $1 75. 


Harper 6^ Brothers' Valuable atid Interesting Works. 3 

Revolution ; or, Illustrations, by Pen aud Pencil, of the History, Biography, 
Scenery, Relics, aud Traditions of the War for ludependeuce. By Benson J. 
LoBsiNO. 'i vols., Svo, Cloth, $14 00; Sheep, $15 00; Half Calf, $18 00; Full 
Turkey Morocco, $2^ 00. 

LOSSING'S FIELD-BOOK OF THE WAR OF 1812. Pictorial Field-Book of thf 
War of 1812 ; or. Illustrations, by Pen aud Pencil, of the History, Biography, 
Scenery, Relics, aud Traditions of the Last War for American Independence. By 
Benson J. LossiNo. With several hundred Engravings on Wood, by Lossing and 
Barritt, chiefly from Original Sketches by the Author. 1088 pages, Svo, Cloth, 
$T 00 ; Sheep, $8 50 ; Half Calf, $10 00. 

ALFORD'S GREEK TESTAMENT. The Greek Testament : with a critically revised 
Text ; a Digest of Various Readings ; Marginal References to Verbal and Idio- 
matic Usage ; Prolegomena ; and a Critical and Exegetical Commentary. For 
the Use of Theological Students and Ministers. By Henry Alfoed, D.D., Dean 
of Canterbury. Vol. I., containing the Four Gospels. 944 pages, Svo, Cloth, 
$6 00 ; Sheep, $G 50. 

ABBOTT'S FREDERICK THE GREAT. The History of Frederick the Second, 
called Frederick the Great. By Joun S. C. Auisott. Elegantly Illustrated. Svo, 
Cloth, $5 00. 

tion of 1789, as viewed in the Light of Republican Institutions. By John S. C. Aii- 
BOTT. With 100 Engravings. Svo, Cloth, $5 00. 

ABBOTT'S NAPOLEON BONAPARTE. The History of Napoleon Bonaparte. By 
JouN S. C. AiusoTT. With Maps, Woodcuts, and Portraits on Steel. 2 vols., 
Svo, Cloth, $10 00. 

ABBOTT'S NAPOLEON AT ST. HELENA ; or. Interesting Anecdotes and Remark- 
able Conversations of the Emperor during the Five and a Half Years of his 
Captivity. Collected from the Memorials of Las Casas, O'Meara, Montholon, 
Antommarchi, and others. By John S. C. Abbott. With Illustrations. Svo, 
Cloth, $5 00. 

ADDISON'S COMPLETE WORKS. The Works of Joseph Addison, embracing the 
whole of the " Spectator." Complete iu 3 vols., Svo, Cloth, $6 00. 

ALCOCK'S JAPAN. The Capital of the Tycoon : a Narrative of a Three Years' 
Residence in Japan. By Sir Rutuekford Aloook, K.C.B., Her Majesty's Envoy 
Extraordinarv and Minister Plenipotentiary iu Japan. With Maps and Engravings. 
2 vols., 12mo,"Cloth, $3 50. 

ALISON'S HISTORY OF EUROPE. First Series : From the Commencement of 
the French Revolution, in 1789, to the Restoration of the Bourbons, in 1815. [In 
addition to the Notes on Chapter LXXVI., which correct the errors of the 
original work concerning the United States, a copious Analytical Index has been 
appended to this American edition.] Seoonp Series : From the Fall of Napoleon, 
in 1S15, to the Accession of Louis Napoleon, in 1852. S vols., Svo, Cloth, $16 00. 

BALDWIN'S PRE-HISTORIC NATIONS. Pre-Historic Nations ; or, Inquiries con- 
cerning some of the Great Peoples and Civilizations of Antiquity, and their 
Probable Relation to a still Older Civilization of the Ethiopians or Cushites of 
Arabia. By JotiN D. Baldwin, Member of the American Oriental Society, 
12mo, Cloth, $1 75. 

BARTH'S NORTH AND CENTRAL AFRICA. Travels and Discoveries in North 
and Central Africa : being a Journal of an Expedition undertaken under the 
Auspices of H. B. M.'s Government, in the Years 1849-1855. By Henry Barth, 
Ph.D., D.O.L. Illustrated. 3 vols., Svo, Cloth, $12 00. 

HENRY WARD BEECHER'S SERMONS. Sermons by Henry Ward Beeoher, 
PIvmouth Church, Brooklyn. Selected from Published and Unpublished Dis- 
courses, and Revised by their Author. With Steel Portrait. Complete in 2 vols., 
Svo, Cloth, $5 00. 

LYMAN BEECHER'S AUTOBIOGRAPHY, &c. Autobiography, Correspondence, 
ifec, of Lyman Beecher, D.D. Edited by his Son, Charles Beeoher. With Three 
Steel Portraits, and Engravings on Wood. In 2 Vols., 12mo, Cloth, $5 00. 

BOSWELL'S JOHNSON. The Life of Samuel Johnson, LL.D. Including a Journey 
to the Hebrides. By James Bobwell, Esq. A New Edition, with numerous 
Additions and Notes. By John Wilson Croker, LL.D., F.R.S. Portrait of 
Boswell. 2 vols., Svo, Cloth, $4 Oa 

4 Harper o^ Brothers'' Valuable and Interesting Works. 

DRAPER'S CIVIL WAR. History of the American Civil War. By John W. Dea- 
PEB, M.D., LL.D., Professor of Clieniistry and Physiology in the -University of 
New York. In Three Vols. 8vo, Cloth, $3 50 per vol. 

Intellectual Development of Europe. By John W. Drapf.b, M.D., LL.D., Profess- 
or of Chemistry and Physiology in the University of New York. 8vo, Cloth, $5 00 

DRAPER'S AMERICAN CIVIL POLICY. Thoughts on the Future Civil Policy of 
America. By John W. Drapbe, M.D., LL.D., Professor of Chemistry and Physiol- 
ogy in the University of New York. Crown Svo, Cloth, $2 50. 

DU CHAILLU'S AFRICA. Explorations and Adventures in Equatorial Africa with 
Accounts of the Manners and Customs of the People, and of the Chase of the Go- 
rilla, the Crocodile, Leopard, Elephant, Hippopotamus, and other Animals. By 
Paul K. Du Chaillu. Numerous Illustrations. Svo, Cloth, $5 00. 

BELLOWS'S OLD WORLD. The Old World in its New Face : Impressions of Eu- 
rope in 1S67-1S68. By Henky W. Bellows. 2 vols., 12mo, Cloth, $3 50. 

BROD HEAD'S HISTORY OF NEW YORK. History of the State of New York. 
By John Romeyn Bboduead. 1609-1691. 2 vols. Svo, Cloth, $3 00 per voL 

BROUGHAM'S AUTOBIOGRAPHY. Life and Times of Henry, Lobb Beohghaji. 
Written by Himself. In Three Volumes. 12mo, Cloth, $2 00 per vol. 

BULWER'S PROSE WORKS. Miscellaneous Prose Works of Edward Bulwer. 
Lord Lytton. 2 vols., 12mo, Cloth, $H 50. 

BULWER'S HORACE. The Odes and Epodes of Horace. A Metrical Translation 
into Enr?lish. With Introduction and Commentaries. By Lop.d Lytton. With 
Latin Text from the Editions of Orelli, Macleane, and Yonge. 12mo, Cloth, $1 75. 

BULWER'S KING ARTHUR. A Poem. By Eabl Lvtton. New Edition. 12mo, 
Cloth, $1 75. 

BURNS'S LIFE AND WORKS. The Life and Works of Robert Burns. Edited 
by Robebt Chambers. 4 vols., 12mo, Cloth, $6 00. 

REINDEER, DOGS, AND SNOW-SHOES. A Journal of Siberian Travel and Ex- 
plorations made in the Years 1865-'67. By Richard J. Bush, late of the Russo- 
American Telegraph Expedition. Illustrated. Crown Svo, Cloth, $3 00. 

C'ARLYLE'S FREDERICK THE GREAT. History of Friedrich II., called Frederick 
the Great. By Thomas Caklyle. Portraits, Maps, Plans. &c. 6 vols., 12mo, 
Cloth, $12 00. 

CARLYLE'S FRENCH REVOLUTION. History of the French Revolution. Newly 
Revised by the Author, with Index, »fec. 2 vols., 12mo, Cloth, $3 50. 

CARLYLE'S OLIVER CROMWELL. Letters and Speeches of Oliver Cromwell. 
With Elucidations and Connecting Narrative. 2 vols., 12mo, Cloth, $3 50. 

CHALMERS'S POSTHUMOUS WORKS. The Posthumous Works of Dr. Chalmers. 
Edited by his Son-in-Law, Rev. William Hanna, LL.D. Complete in 9 vols., 
12mo, Cloth, $13 50. 

COLERIDGE'S COMPLETE WORKS. The Complete Works of Samuel Taylor 
Coleridge. With an Introductory Essay upon his Philosophical and Theological 
Opinions. Edited by Professor Sheud. Complete in Seven Vols. With a fine 
Portrait. Small Svo, Cloth, $10 50. 

DOOLITTLE'S CHINA. Social Life of the Chinese : with some Account of their Re- 
ligious, Governmental, Educational, and Business Customs and Opinions. Witli 
special but not exclusive Reference to Fuhchau. By Rev. Jdstus Doolittle, 
Fourteen Years Member of the Fuhchau Mission of the American Board. Illus- 
trated with more than 150 characteristic Engravings on Wood. 2 vols., 12mo, 
Cloth, $5 00. 

GIBBON'S ROME. History of the Decline and Fall of the Roman Empire. By Ed- 
ward Giubon. With Notes by Kev. H. H. Milman and M. Guizot. A new cheap 
Edition. To which is added a complete Index of the whole Work, and a Portrait 
. of the Author. 6 vols., 12mo, Cloth, $9 00. 

and the Army in Germany and France, with a Diary of Siege Life at Versailles. 
By Brevet Major-General W. B. Hazen, U.S.A., Colflnel Sixth Infantry. Crown 
Svo, Cloth, %■'- .'^.0. 

Harper &* Brothers' Valuable and Interesting Works. 5 

— J 


The following Volumes are now ready. Portraits. 12mo, Cloth, $1 50 each. 


&c. — CioEEO ON Oratory and Orators. — Tacitus (2 vols.). — Terence.— 
Soi'uooLES. — JirvENAL. — Xenophon. — Homee's Iliai). — Homee'b Odyssey. — 
Herodotus. — Demostuenes. — Tuucydides. — .^Escuylus.— EnEiriDES (2 vols.). 
— LiVY (2 vols.). 

DAVIS'S CARTHAGE. Carthage and her Remains : heing an Account of the Exca 
vations and Researches on the Site of the Phoenician Metropolis in Africa and other 
adjacent Places. Conducted under the Auspices of Her Majesty's Government. 
Bv Dr. Davis, F.R.G.S. Profusely Illustrated with Maps, Woodcuts, Chromo- 
Lithographs, &c. 8vo, Cloth, $4 00. 

EDGEWORTH'S (Miss) NOVELS. With Engravings. 10 vols., 12mo, Cloth, $15 Ott 

GROTE'S HISTORY OP GREECE. 12 vols., 12mo, Cloth, $18 00. 

HELPS'S SPANISH CONQUEST. The Spanish Conquest in America, and its Rela- 
tion to the History of Slavery and to the Government of Colonies. By Aethuk 
Helps. 4 vols., 12mo, Cloth, $6 00. 

HALE'S (Mrs.) WOMAN'S RECORD. Woman's Record ; or. Biographical Sketches 
of all Distinguished Women, from the Creation to the Present Time. Arranged 
in Pour Eras, with Selections from Female Writers of each Era. By Mrs. Saeau 
JosEPiiA Hale. Illustrated with more than 200 Portraits. 8vo, Cloth, $5 00. 

HALL'S ARCTIC RESEARCHES. Arctic Researches and Life among the Esqui 
maux : being the Narrative of an Expedition in Search of Sir John Franklin, in 
the Years ISOO, 1861, and 1862. By Cuaeles Francis Hall. With Maps and 100 
Illustrations. The Illustrations are from Original Drawings by Charles Parsons, 
Henry L. Stephens, Solomon Eytinge, W. S. L. Jewett, and Granville Perkins, 
after Sketches by Captain Hall. 8vo, Cloth, $5 00. 

Henry VII. to the Death of George II. 8vo, Cloth, $2 00. 

HALLAM'S LITERATURE. Introduction to the Literature of Europe -during the 
Fifteenth, Sixteenth, and Seventeenth Centuries. By Heney Hallaai. 2 vols., 
Svo, Cloth, $4 00. 

HALLAM'S MIDDLE AGES. State of Europe during the Middle Ages. By Heney 
Hallam. 8vo, Cloth, $2 00. 

First Settlement of the Country to the Adoption of the Federal Constitution. 
Second Seeies: Prom the Adoption of the Federal Constitution to the End of 
the Sixteenth Congress. 6 vols., Svo, Cloth, $18 00. 

HUME'S HISTORY OP ENGLAND. History of England, from the Invasion of Ju- 
lius CiBsar to the Abdication of James IL, 1688. By David Hume. A new Edi- 
tion, with the Author's last Corrections and Improvements. To which is Prefix- 
ed a short Account of his Life, written by Himself. With a Portrait of the Au- 
thor. 6 vols., 12mo, Cloth, $9 00. 

JAY'S WORKS. Complete Works of Rev. William Jay : comprising his Sermons, 
Family Discourses, Morning and Evening Exercises for every Day in the Year, 
Family Prayers, &c. Author's enlarged Edition, revised. 3 vols., Svo, Cloth, 
$6 00. 

JEFFERSON'S DOMESTIC LIFE. The Domestic Life of Thomas Jefferson : com- 
piled from Family Letters and Reminiscences by his Great-Granddaughter, 
Saeau N. Randolph. With Illustrations. Crown Svo, Illuminated Cloth, Bev- 
eled Edges, $2 50. 

JOHNSON'S COMPLETE WORKS. The Works of Samuel Johnson, LL.D. With 
an Essay on his Life and Genius, by Aethub MuPvPhy, Esq. Portrait of Johnson. 

2 vols., Svo, Cloth, $4 00. 

KINGLAKE'S CRIMEAN WAR. The Invasion of the Crimea, and an Account of 
its Progress down to the Death of Lord Raglan. By Alexander William King- 
lake. "With Maps and Plans. Two Vols, ready. i2mo, Cloth, $2 00 per vol. 

KINGSLEY'S WEST INDIES. At Last: A Christmas in the West Indies. By 
Charles Kingsley. Illustrated. 12mo, Cloth, $1 50. 

6 Harper &' Brothers^ Valuable and Interesiifig Works. 

KRUMMACHER'S DAVID, KING OF ISRAEL. David, the King of Israel: a Por- 
trait drawn from Bible History and the Book of Psalms. By Predkrick William 
Kkcmmaoiikr, D.D., Author of "Elijah the Tishbite," &c. Translated under the 
express Sanction of the Author by the Rev. M. G. Easton, M.A. With a Letter 
from Dr. Krummacher to his American Readers, and a Portrait. 12mo, Cloth, 
$1 75. 

LAMB'S COMPLETE WORKS. The Works of Charles Lamb. Comprising his Let- 
ters, Poems, Essays of Elia, Essays upon Shakspeare, Hogarth, &c., and a Sketch 
of his Life, with the Pinal Memorials, by T. Noon Talfourd. Portrait. 2 vols., 
12mo, Cloth, $3 00. 

LIVINGSTONE'S SOUTH APRICA. Missionary Travels and Researches in South 
Africa; including a Sketch of Sixteen Years' "Residence in the Interior of Africa, 
and a Journey from the Cape of Good Hope to Loando on the West Coast ; thence 
across the Continent, down the River Zambesi, to the Eastern Ocean. By Davii> 
Livingstone, LL.D., D.C.L. With Portrait, Maps by Arrowsmith, and numerous 
Illustrations. Svo, Cloth, $4 50. 

LIVINGSTONES' ZAMBESL Narrative of an Expedition to the Zambesi and its 
Tributaries, and of the Discovei-y of the Lakes Shirwa and Nyassa. 1858-1864. 
By David and Cuarles Livingstone. With Map and Illustrations. Svo, Cloth, 
$5 00. 

M'CLINTOCK & STRONG'S CYCLOPEDIA. Cyclopaedia of Biblical, Theological, 
and Ecclesiastical Literature. Prepared by the Rev. Joun M'Clintock, D.D., 
and James Strong, S.T.D. 5 vols, now ready. Roj'al Svo. Price per vol., Cloth, 
$5 00 ; Sheep, $6 00; Half Morocco, $S 00. 

MARCY'S ARMY LIFE ON THE BORDER. Thirty Years of Army Life on the 
Border. Comprising Descriptions of the Indian Nomads of the Plains ; Explo- 
rations of New Territory; a Trip across the Rocky Mountains in the Winter; 
Descriptions of the Habits of Different Animals found in the West, and the Meth- 
ods of Hunting them ; with Incidents in the Life of Different Frontier Men, &c., 
&c. By Brevet Brigadier-General R. B. Marcv, U.S.A., Author of " The Prairie 
Traveller." With numerous Illustrations. Svo, Cloth, Beveled Edges, $3 00. 

MACAULAY'S HISTORY OF ENGLAND. The History of England from the Ac- 
cession of James II. By Thomas BAmNGTON Maoaulay. With an Original Por- 
trait of the Author. 5 vols., Svo, Cloth, $10 00 ; 12mo, Cloth, $T 50. 

MOSHEIM'S ECCLESIASTICAL HISTORY, Ancient and l^odern ; in which the 
Rise, Progress, and Variation of Church Power are considered in their Connec- 
tion withlhe State of Learning and Philosophy, and the Political History of Eu- 
rope during that Period. Translated, with Notes, &c., by A. Maolaine, D.D, 
A new Edition, continued to 1826, by C. Coote, LL.D. 2 vols., Svo, Cloth, $i 00. 

NEVIUS'S CHINA. China and the Chinese : a General Description of the Country 
and its Inhabitants ; its Civilization and Form of Government ; its Religious and 
Social Institutions ; its Intercourse with other Nations ; and its Present Condition 
and Prospects. By the Rev. John L. Nevifs, Ten Years a Missionary in China. 
With a Map and Illustrations. 12mo, Cloth, $1 75. 

THE DESERT OF THE EXODUS. Journeys on Foot in the Wilderness of the 
Forty Years' Wanderings ; undertaken in connection with the Ordnance Survey 
of Sinai and the Palestine Exploration Fund. By E. H. Palmer, M.A., Lord 
Almoner's Professor of Arabic, and Fellow of St. John's College, Cambridge. 
With Maps and numerous Illustrations from Photographs and Drawings taken 
ou the spot by the Sinai Survey Expedition and C. F. Tyrwhitt Drake. Crown 
Svo, Cloth, $3' 00. 

OLIPHANT'S CHINA AND JAPAN. Narrative of the Earl of Elgin's Mission to 
China and Japan, in the Years 1857, '58, '59. By Laurence OLiniANT, Private 
Secretary to Lord Elgin. Illustrations. Svo, Cloth, $3 50. 

OLIPHANT'S (Mrs.') LIFE OF EDWARD IRVING. The Life of Edward Irving, 
Minister of the National Scotch Church, London. Illustrated by his Journals and 
Correspondence. By Mr.=. Olipuant. Portrait. Svo, Cloth, $3 .50. 

tory, from the Earliest Times to the Fall of the Western Empire. Comprising 
the" History of Chaldsea, Assyria, Media, Babylonia, Lydia, Phoenicia, Syria, Ju- 
dsea, Egypt, Carthage, Persia, Greece, Macedonia, Parthia, and Rome. By 
Gf.orge''Rawi,inson, M.A., Camden Professor of Ancient History in the Univer- 
sity of Oxford. 12mo, Cloth, $2 50. 

Harper d>» Brothers' Valuable and Interesting Works. 7 

RECLUS'S THE EARTH. The Earth : a Descriptive History of the Phenomena and 
Life of the Globe. By Elisee Reolus. Translated by the late B. B. Woodward, 
and Edited by Henry Woodward. With 234 Maps and Illustrations, and 23 Page 
Maps printed in Colors. 8vo, Cloth, $5 00. 

RECLUS'S OCEAN. The Ocean, Atmosphere, and Life. Being the Second Series 
of a Descriptive History of the Life of the Globe. By finsijE Reclub. Profusely 
Illustrated with 250 Maps or Figures, and 27 Maps printed in Colors. 8vo, Cloth, 
$6 00. ' 

SHAKSPEARE. The Dramatic Works of William Shakspeare, with the Corrections 
and Illustrations of Dr. Johnson, G. Steevens, and others. Revised by Isaac 
Reed. Engravings. 6 vols., Royal 12mo, Cloth, $9 00. 

SMILES'S LIFE OF THE STEPHENSONS. The Life of George Stephenson, and 
of his Son, Robert Stephenson ; comprising, also, a History of the Invention and 
Introduction of the Railway Locomotive. By Samuel Smiles, Author of "Self- 
Help," &c. With Steel Portraits and numerous Illustrations. 8vo, Cloth, $3 00. 

SMILES'S HISTORY OP THE HUGUENOTS. The Huguenots : their Settlements, 
Churches, and Industries in England and Ireland. By Samuel Smiles. With an 
Appendix relating to the Huguenots in America. Crown Svo, Cloth, $1 75. 

SPEKE'S AFRICA. Journal of the Discovery of the Source of the Nile. By Cap- 
tain John Hannino Speke, Captain H. M. Indian Army, Fellow and Gold Med- 
alist of the Royal Geographical Society, Hon. Corresponding Member and Gold 
Medalist of the French Geographical Society, &c. With Maps and Portraits and 

• numerous Illustrations, chiefly from Drawings by Captain Gkant. Svo, Cloth, 
uniform with Livingstone, Banh, Burton, &c., $4 00. 

STRICKJvAND'S (Miss) QUEENS OF SCOTLAND. Lives of the Queens of Scot- 
land and English Princesses connected with the Regal Succession of Great Brit- 
ain. By Agnes Strickland. 8 vols., 12mo, Cloth, $12 00. 


France. Engravings. 12mo, Cloth, $2 00. 

Gibbon. Engravings. 12mo, Cloth, $2 00. 

Greece. Engravings. 12mo, Cloth, $2 00. 

Hume. Engravings. 12mo, Cloth, $2 00. 

Rome. By Liddell. Engravings. 12rao, Cloth, $2 00. 

Old Testament History. Engravings. 12mo, Cloth, $2 00. 

New Testament History. Engravings. 12mo, Cloth, $2 00. 

Strickland's Queens of England. Abridged. Engravings. 12mo, Cloth, $2 00. 

Ancient History of the East. 12mo, Cloth, $2 00. 

Hallam's Middle Ages. 1 'mo, Cloth. %-L 00. 

Hallnm's Constitutional History of England. 12mo, Cloth, $2 00. 

Lyell's Elements of Geology. 12mo, Cloth, $2 00. 

TENNYSON'S COMPLETE POEMS. The Complete Poems of Alfred Tennyson, 
Poet Laureate. With numerous Illustrations by Eminent Artists, and Three 
Characteristic Portraits. Svo, Paper, 75 cents ; Cloth, $1 25. 

THOMSON'S LAND AND THE BOOK. The Land and the Book ; or. Biblical Illus- 
trations drawn from the Manners and Customs, the Scenes and the Scenery of 
the Holy Land. By W. M. Thomson, D.D., Twenty-tive Years a Missionary of the 
A.B.C.F.M. in Syria and Palestine. With two elaborate Maps of Palestine, an ac- 
curate Plan of Jerusalem, and several hundred Engravings, representing the Scen- 
ery, Topography, and Productions of the Holy Land, and the Costumes, Manners, 
and Habits of the People. 2 large 12mo vols.. Cloth, $5 00. 

TYERMAN'S WESLEY. The Life and Times of the Rev. John Wesley, M.A., Found- 
er of the Methodists. By the Rev. Luke Tyerman, Author of " The Life of Rev. 
Samuel Wesley." Portraits. 3 vols.. Crown Svo, Cloth, $7 50. 

VAMBERY'S CENTRAL ASIA. Travels in Central Asia. Being the Account of a 
Journey from Teheran across the Turkoman Desert, on the Eastern Shore of the 
Caspian, to Khiva, Bokhara, and Samarcand, performed in the Year 1863. By 
Aeminius Vamueky, Member of the Hungarian Academy of Pesth, by whom he 
was sent on this Scientitic Mission. With Map and Woodcuts. Svo, Cloth, $4 60. 

WOOD'S HOMES WITHOUT HANDS. Homes Without Hands: being a Descrip- 
tion of the Habitations of Animals, classed according to their Principle of Con- 
struction. By J. G. Wood, M.A., F.L.S. With about 140 Illustrations. 8vOj 
Cloth, Beveled Edges, $4 50. 

7 07 

"^^^ i O Wf 



t !QC